: i ’ . Vee mae eT me > Le 9.0 te & ims he ho S, OT, ws Pate wer

hy ' 5 at Ag ae Be aia? i ta Se Bde aes 7 Ls Me . ry a Pere he rales mAs

E ~ 3 ‘ as 2 are ae . rare a ie ry 4 " eh m9, 0% ele ax SOAS Ss

rf . sre - > ‘ . e § ss > ;, eel] a . ey mY . . wt 5 =

‘ 3 ‘ ‘ bon =A 4 :. ‘ . fh tot Wath on . A oh yy 52) 2 oe = y) ‘ ¢ Th tt ne . Seed: oat

‘ : ' . : a i . “ , Sony #4 Fe eT ar ea ee A “ the mie Oe ode bee, :

3 a , » oe * ? x ble Anca

Z % ‘ 4 2a a a 3 FA De Re Ke Sere - ‘ * ans ~ 4G

| oo . . u > * 4 : ' i nd Lol elm Ae * . 3

; “ . Wi Par Wes, K ao ‘ ‘aid . 4 nae a ay bk ae tata ear

j . 1g ‘ , : 3 aR ty, res Py 3 Sone the the Det ein ta a

r S . 4 ‘ . ’ ; . ’ 9 a - . + ys ‘ nat i ta Pe 2}.

Z i ate 4 ‘ 1 Bes ? $ , See er “a Lturerg DO Ne te ie Fae te Pm 3

i : ‘ , ' 7 Mpa Ste te ‘es 5 ‘

5a, . ‘ 3 6 + t ie ey v . tat Pa 4 A 4 Aen ae ’

gael . a . ~ a ig a areas . > : ‘i

ney 5 Bott bp hess t if . ; 3 : ‘ F1N32 Sollsthate Sp apy

roe : . Rk 9 4 h ‘ od > ‘ 7 Te oe ; Sareptay

n = 2 ALA es y A t 7 4 soy er . hy “4 “i ry

ure: : ¢ Adet. . CaS ete, te eee Pal i he he DEE roars Weis ee the We ste

% te ’ : 4 fale a » Amy Ce ? a “ ks a # Neyepntey op i ~ 7 27a in.

haa . r ‘ P : t ‘ Fee AS : Eel) tokehte we sede a8 Ba Faihat A ass el A

. . - . ' ery » uy ) : » 7 u ae ee .

t . . H PA tase. =o * ’ 4 ma) . ' alge of

ss. 3 3 oa eho a ae oer iy sb ke hy : cue Tea diab coh tc he te AY shite tabidtachrae

i- a . ; 2 ‘ ca . ‘ : A : y ata aa 3 Bath dy edad Rie &

5 : . of . - ‘ . ats } oe - teat 7 Fe . y mit te te" ats .

‘ - ; Ms ys aS ‘ y bo tea! i ops’) hit , 3 % “2 Rey ea

Da ‘ : - : 5 pie? 5 7 5 ‘ “Wh ee Ss Riot »

2» Tesaes ey 3 : INP te a Pagad TNA LER HA i CPA el Sees came A Meare WON tia

: = 7 ae : . on BUN ee So AW CL bes

2 2 Fa . 4 - “ . ‘ Py 4 tu . i my A . >. 7 hit F aR » a

4 p 3 aes ‘ & , SNR: fe sake ee Et 4 he's ih ala of Pita eee

. . rie ari) P te 4 ys 4 tae : EO B43 ree eG

. 3 4 , . 5 ag 4 pn 3 hh Se a , 2 Romie : t , bhacet on 4 ‘a a Mm AaHte wiiece mse

g ‘ ne . : Wa ‘ , a , cd , NaF sre : * “ony

nm - 4 Ly fide on Ste st. rat 2 . Re 5 + hyn 2 AD AW Ay Pa Ae

ay . EP i —_ re eh Bates i, tes RE ERR hoe) 1. eee meh arte eS

4 Pa asl y : ‘ # aa ie bed te Pasa o

é ae Det ; : : be ee 5 ot ek reas

; ; : oof ae ‘ee ahs palit TRE Ash ade on bie “ue. Of : aca

t ; 4 We etl ‘¢ Re EUs Oleic, nt 5 Pans sted ‘ PRY 24,

: ABS 2h + Puan ay he dea) Wows he q d ? “ ae al as ftese aie

py . 45 : 2 R fse¥ au pare =s on ales

. . 4 5 ; ‘ ‘y - ee : aD fumkrats mt. Mure a

at 5 f SRN en i ye , elt AS MNS “aad, oleic as ba Deen: Se

i ; 5 . 2 ¥% 7 AA le ag 4 £ ae - 7 a Fake 2 ie

q ‘ ’ : OP iy pais : > os Axa

A = ona iis Saha EPR RED H, adr 2 SAWN ee “oh AES

A 55 Mas A ais a eee $ i ta £, . Seats fey s ihe Bieta ys

A . ew 3 A BEDE t ee 4 afy r ah

aa ; cu een ae a eet : % ay) 7 atid Wanner y

? ‘ Abe vey : Fatih ‘ ete Ne BEN « RSAC Meath ae Lee

; 7 4 AERA aah eae : 5 ’ t pre ak Sey Gaon

Sa 2 ’ y ees Saeed A Gh y = i? + ~ y On ends wegen BY ot

. aha a2 zs Y Phe e BOA We Sek a eae

53 ‘ ad A FS Se Ce

: '? JV eS fae ° oan ’ ey fe ; : es AR

oh € te. PORTA A cnaenenisca) Megs lath arias ta . "e 2 Be iG oO

2 . 4 pe Reh ome hs hak a 5 : 3 2

% F ; 4 3 yin ’ me stan hi

, F L V3 he Mya ee ae ee Papo

Me 6 ates! A ee ST . oo:

2e A; * : . , ? he oats a Ty ei

s4 ree t te 2 ie ans : tS ade ‘. PAINS 4

: aS eG a he ke DNS Rat OSG Nae Pies a a eM

; u one i td eer) yf [aS Ea

? ; J ‘be 7 Pho tay ha? ‘ 4 H

, 4 , , ‘oo 4 md s ‘

o> 3 ae Arete yt Bie Waray ; ian

: A en ae Gos aural: Ely ete 49 Mi g-4 2a Se

‘ rs ‘ i aa a i yigee ep:

: A ; HAAN, 7 ae e nie a

Bh gh ag heer BS rg ay ORES Bp

. 7 ;. PR dh 3 2 , WO 44 wrt 1.5 f :

a) i : . on) “4 ; sae

h . $ ‘ a aM Gly J Mee MAG vy aa

g ah . b : he ae at bo 45. 88 E. a4

Mee) ae MS ra betty a Fad 34)

3 hea \. Nagi" oF eo. Pog .

an ' i . % ; . , ae, eT © 2"

. od Sf Py . s Gy : : at zen

. . 5 Bh : “ists Vata Et AP LE

ee | ' 8 . cre PT 2G Papheks

3 7 = ' . i :. ob Fe Py | 2

. ‘eas bo Ta 7 “ ye ‘ ay ? istee 5

3 a cs

v4 , PQ 3 : ts i [iad

‘ ; Say CoP ESA (5, ite Se . Lig Sas pal

ae, ¥ ‘ » eG Lt YR Re tain

ea? . 3 F igsiceg Pe

, >: A : r F hoa?

ria R na nt, WARM iy ny)

P Phy ae | Tl Ps E ” 3 (Bi Seat . ,

i ; Fic a a 3 q ’ F

; ‘ : : 7 eh ee ty g . i Fae par bar ee i

aX Pe 3 3 rAd : rS9 ered Tari eT Fe ai |

2 pers roe Shia, 0 te ‘ Cie Ag ibaa)

. c ae & 7 fu? a ee SWOaL ; § -.* 7 a poAee

Fag % 4 Bs sy ‘ one ‘ 4 ie

y 1 4 ’ ‘ i Oy i

: j , wt ; j 4 be dezy ‘ iJ 2 ei | .

. ti ‘ LY Wag? . ec ia a a i & ¥y'¢ ba

4 i g + . , . eh} is .

7% aie . ‘ 45 ,

: ;

4 ' 1 Bee , are t ty i”, eos r

Onur ae 3 i. See, . Gay ‘ ‘ Poeriara

1 oS 4 ec

‘ I mo ' ag i dettt ee) by

, , ; Ly aay Mi ‘7 7%) *

r So ioe) WY i 49 ‘

y ' ? G Fike fey = et Rahs

4 Pa ‘ 7 o ' ‘

A . ’ ry hy CT . Cad

a 4s ay

; : ay i bi Sle ‘ rata iunsek Shee

; os he teins res ' hintoa : : Saar ak tee es ot Ta

; ‘ Ore Fs : : ‘ Ls tp Wi deeS a. ces « s

4 5 VB eigre en Te

- tee dPSLB OI BCU vainerts oy ater noes

: Fi ' rman : , “3 Nee ited PRUE So

6 ; paca ‘ j Si oe ie ae

\ ; a | SERN ES ee

, t cht aeiy z

; Ja : r . i Lewd 2 98 VEE ngs

/ 4 i PME Care iy .

F . ‘oy gl Marrtgs Berge

i mW ge EA ed by J ES Sy

, OL LN fete eos ENS YY

a4 ‘ ; ; j : : rs WEES Be Fee eg ye EES RY Hse

t 7 6 yea see FOR eS gee

} ; PE 8. Oke bese gE

t : SB RD ee pa qe

‘ : fe SED eee gy ‘

Ce oF ey

. Pr Poe

: . ‘ ou

7 , 4 A bed

F of oe ae) SS

‘ ‘ t oY

' ‘ : ‘ ay

’ aver : +

yey oa

Sie © CLbijp 4 \% YM, CV SY iia BD NS, Cy CY

Ss = “jl O A Sa NS 5 Sa: Ros 5

2 Zz 4 re Taal odd Z

STITUTION NOILNLILSNI_NVINOSHLIWS _ $4 | yvyaiT LIBRAR TES SMITHSONIAN INST

_ 4 ws ~

= >) = 7 E We

= ss E a as e

Z Oo z oD 2 o z

1yVvusgit INSTITUTION NOILNLILSNI NVINOSHLINS S3/uvygit

< = : = < = =<

Wy ae = . 1 eS oon Z

tr 3. x SS aa £ Oo x a

* a= Z NES = 2 E

3 = oS Ss = > =

_NVINOSHLINS S3!1Y¥VYEIT LIBRARIES SMITHSONIAN INSTITUTION

“) =a : > dp) > ep)

<x c = < % a <

oc = = 25 ~i a

~ S fs = re) =

a = 2 a 2 =

1YVUSIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNS NVYINOSHLINS S3Iuvudli

Saiuvualy

INSTITUTION

N“

S3i1uyvuagiy

STITUTION NOILALILSNI NVINOSHLINS S31YVYSIT

NVINOSHLINS S@!IYVYYSII~ LIBRARIES SMITHSONIAN

NVINOSHLIWS

SMITHSONIAN

NVINOSHLIWS

SMITHSONIAN

NVINOSHLIW!

I'YVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI

=o w” > “” = >

a : 3 z : a

= a S a = oc =

S fs 7 S| e 3S = S

z oF A = zs z ey ed S

STITUTION _ NOHLALILSNI NVINOSHLINS SJI18VEat LIBRARIES SMITHSONIAN INSTITUT Ham ;

— °° =

5 ro \ 5 a 5 2 =

' a ? =

: 2 WHY E : S ‘3 e

= —_ NS \\ = ees a pes =

2 ee - z - z

== — = w aa

IMVUGIT_LIBRARIES SMITHSONIAN INSTITUTION NOILMLILSNI_NVINOSHLINS S31NVuaI7

= = =a, ” z 2) ‘

Wy z = ae 2 = =

Y x = 4

Ty FA Oo SS S mB i es aN = 3 = =

bhi? XS; 2 SNE 77? :

eo = NEY 2 Eee 2k = z

sh Ae “een : z

STITUTION NOILALILSNI_ NVINOSHLINS S3I¥YVYUSIT LIBRARIES SMITHSONIAN INSTITUTION

git LIBRARIES _ SMITHSONIAN _

” = ~ ” =_ —

» & NS Ww z us z

oe Cc Ye AS. < = cg roa

4 © SSN ce fond Cc

= re) NS - = mM. =

Bi z Ass a = = 2

MtYVUSiT LIBRAR ae SMIFHSONIAR INSTITUTION NOILNLILSNI NVINOSHLINS S31¥vudit

JB ie oa 2

AaiSYM Oo <MSON, oO oo No

Sat = 7 4? SM 2 6G = Me & ED

BNI _NVINOSHLIINS (satadVadii LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSE

i iS = zm = oa

a a = = a

ow = ER oO 3 = wo

Ps) Yous. 5 a = 5 as

> \ WW = > = = sca

= We : 7 5 ;

Ay tt, Y = Q zZ 2 y BS

ES SMITHSONIAN INSTITUTION NOILALILSNI_NVINOSHLINS S31uyvugi7_LIBRARIES SMITHSC

NVINOSHLINS S31NVYaI7

NVINOSHLINS

SMITHSONIAN

NVINOSHLINS

SMITHSONIAN

NVINOSHLIWS

BNI NVINOSHLINS S3IYVYEIT LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI_ NVINOSE

af ee co uh

é : ; w é “ Gi

a o = [es Le = ao a

var Ss fon e fea ce fad = Pa a

2 WY a = 3 = : = WSN

a ae a

ES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYUGIT LIBRARIES SMITHS(

z f z Sree c z

= = = SS < O 2

Sires : < ce = SS = a 5 tf

= ey: > = > MG 2 > es hy)

— ‘typ = 0] es > \ ‘SN E ea] — “yf

- “iy ae bo as re Bs - ii

SG « SS Se = 3 “We

me ep) = cae an ¢ p) *, <— wo ==

SNi NVINOSHLINS LIBRARIES SMITHSONIAN INSTITUTION NOILOALILSNI NVINOS

z wo = we ” z eS: 2) =

= 2 pes Sa = a ae = <

: : Z : EQN : %

E zy B z 2 NR 2;

mE Zz ts 2 = =o =

eS SMITHSONIAN INSTITUTION NOILNLILSNI_ NVINOSHLIWS S3iyvyusiT_LIBRARI Be oaliitet

a = ld = i = ud .

— 4 + ed + — 4 — ots

e : “ 2 2 3 = Pg

aa cS Pe = ac & a Vig

a “Gs 3 : : : en

- z= = = ae = 24

SNI_ NVINOSHLINS S3Ii4¥VYUdSIT LIBRARIES SMITHSONIAN _ INSTITUTION NOILNLILSNI_NVINOSH

a ~ = EF S : ay = | 278

o > — oo — uo] = w

> ea 5 2 : 2 - z

2 WE : hae : :

Pee |!) S OD z o z a

”

re)

ES SMITHSONIAN INSTITUTION NOILALILSNI_NVINOSHLINS S3tYVeGIT LIBRARIES

— < . 2 iy < = a

cat Zz =| Yt fy, = al S

ra ee Ne fit D un A NX

2 To WAN oO UY fof? ee fe) ss .

Zz = NN es Z E

BNI_NVINOSHLINS S3IYVYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_ NVINOS

ae ih ” > ” “ ” Zz |

5 3 = : = : a N

yar oc S a & a 4 AW

aa Mm _- m. — fea) = »\

4 — fe) ce ra) st re) s

ES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYSIT LIBRARIES SMITHS

Fa at = 5 “ Zz =e =

ac «<@ ASE Oo <asv al — “x q F re) ans > ree) StHSON > ~

Eel fa [RD = GAD © SNS = GAT 2 LAMY E OHH

boy

NH ANNALS OF THE ANNALE VAN DIE

SOUTH AFRICAN MUSEUM — SUID-AFRIKAANSE MUSEUM

VOLUME 97 BAND 97

vee 7 ; \

/ \

/ 1@

tae,

. ~ eg

ANNALS OF THE SOUTH AFRICAN MUSEUM

ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

VOLUME 97 BAND

(AS

TARAS

S BBWBWe

7 €

OY,

DQ <>

gy 9 NOVI ae

THE TRUSTEES OF THE DIE TRUSTEES VAN DIE

SOUTH AFRICAN MUSEUM SUID-AFRIKAANSE MUSEUM

CAPE TOWN KAAPSTAD

1985-1988

SET, PRINTED AND BOUND IN THE REPUBLIC OF SOUTH AFRICA

BY THE RUSTICA PRESS (PTY) LTD, NDABENI, CAPE

D2109

LIST OF CONTENTS

Cooke, H. B. S. see HENDEY, Q. B.

HAARHOFF, P. J.

A new fossil stork (Aves, Ciconiidae) from the Late Tertiary of Langebaanweg,

SouthyAtnicas (Published wanwanyl988s)mon- ane sac eee ieee

HENDEY, Q. B. & Cooke, H. B. S.

Kolpochoerus paiceae (Mammalia, Suidae) from Skurwerug, near Saldanha, South

Africa, and its palaeoenvironmental implications. (Published December 1985.)

HILxer, N.

The South African Museum’s Meiring Naude cruises. Part 16. Brachiopoda from the

ISS OPS crmsesm (kublishedwunedOSGy)) meas eee nae ane ee se

Hu.tey, P. A.

A taxonomic review of the lanternfish genus Triphoturus Frazer-Brunner, 1949

(Myctophidae, Osteichthyes). (Published May 1986.)...................2005.

Hu. ey, P. A.

Lanternfishes of the southern Benguela region. Part 1. Faunal complexity and distri-

budone (BublishedtAucustsl986s) tan) ssc. cn ohn ado woh oaths daeeca sence water

KENSLEY, B.

The fossil occurrence in southern Africa of the South American intertidal mollusc

Concholepas concholepas. (Published October 1985.) .................22-05-

KENSLEY, B. & PETHER, J.

Late Tertiary and Early Quaternary fossil Mollusca of the Hondeklip area, Cape

Provinces south Atricas (PublishedyAugust 1986;)25.5--..5-22422-5:05098-20-

Lyons, J. see Myers, A. A.

Myers, A. A. & Lyons, J.

A re-evaluation of the South African species of Lemboides Stebbing and Lembos

Bate (Amphipoda, Aoridae) described by K. H. Barnard (1916). (Published

Jeramerny DOT a his's wi aia, ate ey ais GON RIC eR ne Or re a Sea ed oer

Otson, S. L.

Early Pliocene ibises (Aves, Plataleidae) from south-western Cape Province, South

Attica. (Published: December 985!) sa. oie oe ese fe ees cheeses Soest

PETHER, J. see KENSLEY, B.

PICKFORD, M.

Miocene Suidae from Arrisdrift, South West Africa-Namibia. (Published October

IIGRSSZ'S)) 56 Go. B table Utes eat cic co ee aA a Oa

VERSEVELDT, J. & WILLIAMS, G. C.

A redescription of the soft coral Alcyonium valdiviae Kikenthal, 1906, with the

description of a new species of Litophyton Forskal, from southern Africa

(Octocorallia, Alcyonacea). (Published July 1988.).................00000005.

WILLIAMS, G. C. see VERSEVELDT, J.

WItson, M. L.

Notes on the nomenclature of the Khoisan. (Published August 1986.) ..............

Volume 97 is complete in 12 parts.

Page

u3)ii

Te:

DOs,

141

267

283

3115)

y -—«-—so IE: 97 PART 1 OCTOBER 1985 ISSN 0303-2515

agai. & 0 vl

‘OF THE SOUTH AFRICAN

" MUSEUM -

CAPE ‘TOWN

INSTRUCTIONS TO AUTHORS

1. MATERIAL should be original and not published elsewhere, in whole or in part.

2. LAYOUT should be as follows:

(a) Centred masthead to consist of ;

Title: informative but concise, without abbreviations and not including the names of new genera or species

Author’s(s’) name(s)

Address(es) of author(s) (institution where work was carried out)

Number of illustrations (figures, enumerated maps and tables, in this order)

(b) Abstract of not more than 200 words, intelligible to the reader without reference to the text

(d) Introduction

(e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents

(f) Summary, if paper is lengthy

(g) Acknowledgements

(h) References

(i) Abbreviations, where these are numerous.

3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with

3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for

illustrations should be typed separately, their positions indicated in the text. All pages should be num-

bered consecutively.

Major headings of the paper are centred capitals; first subheadings are shouldered small capitals;

second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further

subdivisions should be avoided, as also enumeration (never roman numerals) of headings and

abbreviations.

Footnotes should be avoided unless they are short and essential.

Only generic and specific names should be underlined to indicate italics; all other marking up

should be left to editor and publisher.

4. ILLUSTRATIONS should be reducible to a size not exceeding 12x 18cm (19 cm including

caption); the reduction or enlargement required should be indicated (and preferably uniform); orig-

inals larger than 35 X 47 cm should not be submitted; photographs should be rectangular in shape and

final size. A metric scale should appear with all illustrations, otherwise magnification or reduction

should be given in the caption; if the latter, then the final reduction or enlargement should be taken

into consideration.

All illustrations, whether line drawings or photographs, should be termed figures (plates are not

printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items

of composite figures should be designated by capital letters; lettering of figures is not set in type and

should be in lower-case letters. If Letraset is used authors are requested to use Helvetica-style letter-

ing, if possible.

The number of the figure should be lightly marked in pencil on the back of each illustration.

5. REFERENCES cited in text and synonymies should all be included in the list at the end of the

paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable):

(a) Author’s name and year of publication given in text, e.g.:

‘Smith (1969) describes. . .’

‘Smith (1969: 36, fig. 16) describes. . .’

‘As described (Smith 1969a, 1969b; Jones 1971)’

‘As described (Haughton & Broom 1927)...”

‘As described (Haughton et al. 1927)...”

Note: no comma separating name and year

pagination indicated by colon, not p.

names of joint authors connected by ampersand

et al. in text for more than two joint authors, but names of all authors given in list of references.

(b) Full references at the end of the paper, arranged alphabetically by names, chronologically within

each name, with suffixes a, b, etc., to the year for more than one paper by the same author in

that year, e.g. Smith (1969a, 1969b) and not Smith (1969, 1969a).

For books give title in italics, edition, volume number, place of publication, publisher.

For journal article give title of article, title of journal in italics (according to the World list of scientific periodicals. 4th ed.

London: Butterworths, 1963), series in parentheses, vclume number, part number in parentheses, pagination (first and

last pages of article).

Examples (note capitalization and punctuation)

BuLLouGH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan.

FISCHER, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140.

FiscHER, P. H., Duvat, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie

expérimentale et générale 74 (33): 627-634.

KouN, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and

Magazine of Natural History (13) 2 (17): 309-320.

KOHN, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of

the Bingham Oceanographic Collection, Yale University 17 (4): 1-51.

TuieLe, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthro-

pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika ausgefiihrt in den Jahren

1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270.

(continued inside back cover)

ANNALS OF THE SOUTH AFRICAN MUSEUM

ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

Volume 97 Band

October 1985 Oktober

Part 1 Deel

THE FOSSIL OCCURRENCE

IN SOUTHERN AFRICA OF THE

SOUTH AMERICAN INTERTIDAL MOLLUSC

CONCHOLEPAS CONCHOLEPAS

BRIAN KENSLEY

Cape Town Kaapstad

The ANNALS OF THE SOUTH AFRICAN MUSEUM

are issued in parts at irregular intervals as material

becomes available

Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000

Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

word uitgegee in dele op ongereelde tye na gelang van die

beskikbaarheid van stof

Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000

OUT OF PRINT/UIT DRUK

1 AGES), 523), ED, 245, & tot), SOS, 5, 7D,

GG, toa), TA), 8, O02, ?), 10 GS),

GED, 5, ion), WED), IC), 24), M7, BGS), 225), 63, 26), 45)

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 075 3

Printed in South Africa by In Suid-Afrika gedruk deur

The Rustica Press, Pty., Ltd., Die Rustica-pers, Edms., Bpk.,

Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap

THE FOSSIL OCCURRENCE IN SOUTHERN AFRICA OF

THE SOUTH AMERICAN INTERTIDAL MOLLUSC

CONCHOLEPAS CONCHOLEPAS

BRIAN KENSLEY

National Museum of Natural History,

Smithsonian Institution, Washington, D.C.

(With 2 figures)

[MS accepted 16 April 1985]

ABSTRACT

The occurrence of the thaidid gastropod genus Concholepas is recorded from presumed Late

Pleistocene coastal deposits in southern South West Africa-Namibia. The material is

indistinguishable from C. concholepas, a species known from the Pliocene to Recent on the west

coast of South America. The living species characteristically occurs in cold-temperate waters

from the intertidal to depths of 40 m. It is suggested that the southern African fossils represent a

short-lived pioneer population, established by larvae drifting from South America. Other

organisms having a similar disjunct distribution are discussed.

CONTENTS

PAGE

MN GHOCMUIC WM OMpEnpey eres wart AM ees ae. Phe Oe oe Bau F ecaur de ee be dys 1

SVSLCMIANICMISCUISSIONG A eae nee Nee ge ce Soe ees ee bas Nese 2

DISCUSSIONM EINE ee ee oan eel eats ONS ee ci a bbe ae s 2

PNCKMOME GEC INEM(SHe Pt he eats Ap en Sas nae on bee 6

INCICTICN CC CREne ErINre Tn tev tie ere ey eth. Dead lec me Sickah 2 ane es 6

INTRODUCTION

A small collection of fossil mollusc shells from South West Africa—Namibia

was submitted to the South African Museum for identification in 1983. The fossils

came from Area U of the Consolidated Diamond Mines diamond area number 1,

about 21 km north of the Orange River mouth.

The collection contained only three species, two of which, the bivalve

Choromytilus meridionalis, and the whelk Nucella squamosa, have been recorded

from the Pleistocene deposits on the west coast (see Barnard 1962; Tankard

1975). The third species is a Concholepas, which genus has never been recorded

either as fossil or alive from Africa. On searching the Cenozoic invertebrate

collections of the South African Museum, two further very worn specimens, also

from South West Africa—Namibia, were found. This report describes and records

this material and speculates on its history.

Ann. S. Afr. Mus. 97 (1), 1985: 1-7, 2 figs.

2: ANNALS OF THE SOUTH AFRICAN MUSEUM

SYSTEMATIC DISCUSSION

Family Thaididae

Genus Concholepas Lamarck

Concholepas concholepas (Bruguiere, 1789)

Jens I, 2

Material

SAM-—PQ2407, PN 33, 1 specimen, 107,6 x 75,9 mm (with adherent coarse

sand and gravel); PN 34, 1 specimen, 98,3 x 72,3 mm (very worn, smooth); South

West Africa, no further collection data.

SAM-PQ2408, 6 specimens, 114,4 x 90,0 mm (with attached barnacle

remains), 113,6 X 81,1 mm (with attached barnacle remains), 102,2 x 78,7 mm,

? X 88,7 mm (anterior body whorl damaged), 96,0 x 73,3 mm, 48,7 x 35,2 mm

(very worn); Diamond Area no. | of Consolidated Diamond Mines (Pty) Ltd.,

U11 megatranch assemblage, about 21 km north of Orange River mouth.

Description

Shell thick (up to 14 mm at outer lip of largest specimen). Entire shell of

1,5 whorls. Spire submerged, not extending beyond margin of aperture. Earliest

visible sculpture cancellate, but with spiral lines stronger than axial lines; 16 spiral

lines visible to siphonal ridge, weak and strong lines alternating. Body whorl

expanded, aperture flared, oval, with strong siphonal groove present at anterior

body whorl, ending in marginal notch; latter with two strong rounded marginal

teeth projecting slightly beyond margin in unworn specimens. Outer lip margin

faintly crenulate, becoming smooth in columellar region. Body whorl sculpture:

38-40 spiral ridges from apical region to siphonal ridge, 5—6 ridges below

siphonal ridge, but rapidly becoming obsolete. Siphonal ridge widening distally,

strongly and evenly rounded. Axial sculpture consisting of irregularly spaced

growth lines, becoming wavy as margin is approached; none becoming lamellate.

Two specimens have barnacle skeletons and bases in the area of the spire,

outer body whorl, as well as submarginally. The apertural margin appears

aragonitic, the lining of the body whorl white and lamellar-calcareous. Two of the

smaller specimens show red-brown iron-like staining.

DISCUSSION

The genus Concholepas contains eight recognized species or subspecies, both

fossil and living. Information on geological age and distribution is well

summarized in Stuardo (1979, tables 3, 4).

A brief comparative survey of these forms will help to characterize the

present fossil material.

Concholepas drezi Vokes, 1972: Chipola Formation, Florida (late Lower

Miocene); species with somewhat extended spire.

A FOSSIL CONCHOLEPAS FROM SOUTHERN AFRICA 3)

Fig. 1. Concholepas concholepas, SAM-PQ2407, inner and outer

view of two specimens from South West Africa—Namibia (no further

collection data available). Scale = 10 mm.

Concholepas antiquata Tate, 1894: Eocene of Mornington, Port Philip Bay,

and Muddy Creek, Hamilton, Australia; species with strong terminal spire, very

strong radial sculpture; aperture outline truncate due to posterior angulation.

Concholepas deshayesi Rambur, 1862: Middle Miocene of Touraine, France;

species with spire of 3,25—3,5 whorls; siphonal groove ending in faint bulge on

apertural margin; margin smooth, axial sculpture (growth lines) low and non-

lamellar.

Concholepas kieneri Hupe, 1854: Mio—Pliocene of Chile; species with spire

extending beyond aperture. (Moricke (1896) considered this species closely

related to, if not a direct ancestor of, C. concholepas.)

ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 2. Concholepas concholepas. A-F. SAM-PQ2408. Specimens from

Diamond Area no. 1, 21 km north of Orange River mouth. G. USNM BTS,

Recent specimen from Valparaiso, Chile. Scale = 10 mm.

A FOSSIL CONCHOLEPAS FROM SOUTHERN AFRICA 5

Concholepas pehuensis (Marwick, 1926): Upper Miocene of North Taranaki,

New Zealand; species subdiscoidal, with aperture wider than long. (Originally

described as a Lippistes.)

Concholepas nodosa Moricke, 1896: Tertiary of Coquimbo, Chile; species

with terminal spire, few strongly nodose spiral bands on body whorl.

Concholepas concholepas (Bruguiére, 1789), and subspecies C. c. fernan-

dezianus Stuardo, 1979 (confined to the Juan Fernandez Archipelago): only living

representative of genus; occurring in southern Peru and along the entire coast of

Chile. With the local name ‘loco’, the species is of economic importance as a

food-source. Schwabe (1959) records the area of optimum distribution for this

species as being between Valparaiso and Corral on the Chilean coast (roughly

between 32° and 40°S). The species has been recorded as a fossil from the

Pleistocene of Chile and Peru (Herm 1969).

Extensive studies on the morphology and variation of C. concholepas from

South America have been done, e.g. Schwabe (1959), Lozada et al. (1976), and

Stuardo (1979). While there would appear to be some variation in the length/

width ratio of the shell aperture, most specimens fall within a cluster, as

illustrated by Stuardo’s figure 3. If the dimensions for the six complete southern

African specimens are superimposed on Stuardo’s figure 3, these too, fall within

this cluster.

While axial sculpture is variable in the living Concholepas, with some

specimens having almost lamellate ridges, in others this is less marked. Specimens

from relatively circumscribed localities tend to have similar sculpture, although

this is to some degree also related to ecological conditions (Herm 1969: 136).

No differences either in sculpture or proportions or morphology of the shell

can be discerned between the southern African fossils and living South American

specimens (of which many have been examined). There is thus no basis for

taxonomically separating the southern African specimens from C. concholepas.

In South America, the species lives on rocky substrates from the intertidal to

a depth of 40m, usually in association with barnacles, mytilid bivalves,

bryozoans, and serpulid polychaetes (Guisado & Castilla 1983). The prey is

mainly barnacles and mytilids. The sea-temperature range of the species is in the

region of 10-14 °C (Gallardo 1979), i.e. a temperature range similar to that

occurring from the Late Pleistocene to the present on the west coast of South

Africa and southern South West Africa—Namibia.

Several unanswered questions remain with regard to this record. It is unlikely

that the southern African record represents a relict population of a Late

Cretaceous range that included the west and south-west coasts of the separating

masses of South America and Africa. There are no fossil records of

C. concholepas earlier than the Pliocene.

What is perhaps more likely, is that the southern African fossils represent a

chance pioneer population, established in the Pleistocene, long after the South

Atlantic had opened up. With a pelagic life of more than two months (Gallardo

1979), larvae could be carried by the West Wind Drift from southern South

6 ANNALS OF THE SOUTH AFRICAN MUSEUM

America to the west coast of southern Africa, eventually to settle on the rocky

intertidal, there to prey on barnacles and bivalves. A breeding population could

have been established, but which later died out either because of localized sea-

level or sea-temperature fluctuations. A similar larval-dispersal argument has

been used to explain the present-day southern oceanic distribution of the xanthid

crab Pilumnoides perlatus (Kensley 1981).

A similar west coast South America—west coast southern Africa distribution

has been noted for three living mytilid bivalves, viz. Aulacomya ater, Choromyti-

lis meridionalis (? = C. chorus of South America), and Semimytilus algosus, and

for the brachiopod Discinisca tenuis (Kensley & Penrith 1970). Of these species,

A. ater and C. meridionalis have both been recorded from the Pleistocene

(Barnard 1962), the latter occurring with the Concholepas fossils recorded here.

ACKNOWLEDGEMENTS

I am grateful to Mr K. R. Hazell and Mr B. Hawthorne of the Geology

Departments of Consolidated Diamond Mines (Pty) Ltd. and De Beers

Consolidated Mines Ltd. respectively, for making the material described here

available, and for collection data. Dr Louis DiSalvo of Coquimbo, Chile,

provided information on literature; Dr Q. B. Hendey of the South African

Museum, Cape Town, provided information and hospitality during my visits to

that institution; Miss E. Pretorius took the photographs used here; to all of these,

my sincere thanks. The manuscript was read by Dr R. Houbrick (Smithsonian

Institution) and Dr Q. B. Hendey (South African Museum); I am grateful for

their comments and criticisms.

REFERENCES

BARNARD, K. H. 1962. Revised list of South African Late Tertiary and Pleistocene marine

Mollusca. Transactions of the Royal Society of South Africa 36 (4): 179-196.

BRuGUIERE, J. G. 1789. Encyclopédie méthodique. Histoire naturelle des vers. Tome Premier.

Paris: Panckoucke.

GALLARDO, C. 1979. El ciclo vital del Muricidae Concholepas concholepas y consideraciones

sobre sus primeras fases de vida en el bentos. Biologia Pesquera, Santiago de Chile 12:

79-89.

Guisabo, C. & CastTILLA, J. C. 1983. Aspects of the ecology and growth of an intertidal juvenile

population of Concholepas concholepas (Mollusca: Gastropoda: Muricidae) at Las Cruces,

Chile. Marine Biology 78 (1): 99-103.

HerM. D. 1969. Marines Pliozan und Pleistozan in Nord- und Mittel-Chile unter besonderer

Berucksichtigung der Entwicklung der Mollusken-Faunen. Zitteliana 2: 1-159.

Hupe, L. H. 1854. Fauna Chilena. Moluscos. In: Gay, C. Historia fisica y politica de Chile.

Zoologia 8: 1-500.

KENSLEY, B. 1981. On the zoogeography of southern African decapod crustacea, with a

distributional checklist of the species. Smithsonian Contributions to Zoology 338: 1-64.

KENSLEY, B. & PenritH, M.-L. 1970. New records of Mytilidae from the northern South West

African coast. Annals of the South African Museum 57 (2): 15-24.

LozapA, E., Lopez, M. T. & DEsQuEyroux, R. 1976. Aspectos ecologicos de poblaciones

Chilenas de loco Concholepas concholepas (Bruguiére, 1789) (Mollusca, Gasteropoda,

Muricidae). Biologia Pesquera, Santiago de Chile 8: 5-29.

A FOSSIL CONCHOLEPAS FROM SOUTHERN AFRICA ii

Marwick, J. 1926. New Tertiary Mollusca from North Taranaki. Transactions of the New

Zealand Institute 56: 317-331.

MorickE, W. 1896. Versteinerungen der Tertiarformation von Chile. Jn: MOrickeE, W. &

STEINMANN, G. Die Tertiarbildungen des nordlichen Chile und ihre Fauna. Neues Jahrbuch

fiir Mineralogie, Geologie und Paldontologie (B) 10: 548-612.

RAMBUR, P. 1862. Diagnose d’un Concholepas fossile des Faluns de la Touraine. Journal de

Conchyliologie (3) 2: 86.

ScHwABE, G. H. 1959. Biometrische Daten zur Schale von Concholepas concholepas

(Bruguiére) (Moll. Muricidae) an der chilenischen Kiste und ihr 6kologischer Indikator-

wert. Internationales Revue der Gesamten Hydrobiologie 44: 449—462.

STuARDO, J. 1979. Sobre la clasificacion, distribucion y variacion de Concholepas concholepas

(Bruguiére, 1789): Un estudio de taxonomia beta. Biologia Pesquera, Santiago de Chile 12:

5-38.

TANKARD, A. J. 1975. Thermally anomalous Late Pleistocene molluscs from the south-western

Cape Province, South Africa. Annals of the South African Museum 69 (2): 17-45.

TaTeE, R. 1894. Unrecorded genera of the older Tertiary fauna of Australia including diagnoses

of some new genera and species. Journal of the Royal Society of New South Wales 27:

167-197.

VoKEs, E. H. 1972. Notes on the fauna of the Chipola Formation VII. On the occurrence of the

genus Concholepas (Gastropoda: Thaididae), with the description of a new species. Tulane

Studies in Geology and Paleontology 10: 31-33.

6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu-

larly Articles 22 and 51).

Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed

by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov.,

etc.

An author’s name when cited must follow the name of the taxon without intervening punctuation

and not be abbreviated; if the year is added, a comma must separate author’s name and year. The

author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is trans-

ferred from its original genus. The name of a subsequent user of a scientific name must be separated

from the scientific name by a colon.

Synonymy arrangement should be according to chronology of names, i.e. all published scientific

names by which the species previously has been designated are listed in chronological order, with all

references to that name following in chronological order, e.g.:

Family Nuculanidae

Nuculana (Lembulus) bicuspidata (Gould, 1845)

Figs 14-15A

Nucula (Leda) bicuspidata Gould, 1845: 37.

Leda plicifera A. Adams, 1856: 50.

Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b).

Nucula largillierti Philippi, 1861: 87.

Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9.

Note punctuation in the above example:

comma separates author’s name and year

semicolon separates more than one reference by the same author

full stop separates references by different authors

figures of plates are enclosed in parentheses to distinguish them from text-figures

dash, not comma, separates consecutive numbers.

Synonymy arrangement according to chronology of bibliographic references, whereby the year is

placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable.

In describing new species, one specimen must be designated as the holotype; other specimens

mentioned in the original description are to be designated paratypes; additional material not regarded

as paratypes should be listed separately. The complete data (registration number, depository, descrip-

tion of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.:

Holotype

SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach, Port Eliza-

beth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973.

Note standard form of writing South African Museum registration numbers and date.

7. SPECIAL HOUSE RULES

Capital initial letters

(a) The Figures, Maps and Tables of the paper when referred to in the text

e.g. ‘.. . the Figure depicting C. namacolus ...’: ‘. . . in C. namacolus (Fig. 10)...’

(b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by

initials or full names

e.g. Du Toit but A.L. du Toit; Von Huene but F. von Huene

e.g. Therocephalia, but therocephalian

Punctuation should be loose, omitting all not strictly necessary

Reference to the author should preferably be expressed in the third person

Roman numerals should be converted to arabic, except when forming part of the title of a book or

article, such as

‘Revision of the Crustacea. Part VIII. The Amphipoda.’

Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial

capital letter, provided the same generic name is used consecutively. The generic name should

not be abbreviated at the beginning of a sentence or paragraph.

Name of new genus or species is not to be included in the title; it should be included in the abstract,

counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts.

BRIAN KENSEEY

THE FOSSIL OCCURRERS

IN SOUTHERN AFRICA OF THE

SOUTH AMERICAN INTERTIDAL MOLLUSC

CONCHOLEPAS CONCHOLEPAS

E97 PART 2 DECEMBER 1985 ISSN 0303-2515

ANNALS

INSTRUCTIONS TO AUTHORS

1. MATERIAL should be original and not published elsewhere, in whole or in part.

2. LAYOUT should be as follows:

(a) Centred masthead to consist of

Title: informative but concise, without abbreviations and not including the names of new genera or species

Author’s(s’) name(s)

Address(es) of author(s) (institution where work was carried out)

Number of illustrations (figures, enumerated maps and tables, in this order)

(b) Abstract of not more than 200 words, intelligible to the reader without reference to the text

(d) Introduction

(e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents

(f) Summary, if paper is lengthy

(g) Acknowledgements

(h) References

(i) Abbreviations, where these are numerous.

3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with

3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for

illustrations should be typed separately, their positions indicated in the text. All pages should be num-

bered consecutively.

Major headings of the paper are centred capitals; first subheadings are shouldered small capitals;

second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further

subdivisions should be avoided, as also enumeration (never roman numerals) of headings and

abbreviations.

Footnotes should be avoided unless they are short and essential.

Only generic and specific names should be underlined to indicate italics; all other marking up

should be left to editor and publisher.

4. ILLUSTRATIONS should be reducible to a size not exceeding 12 x 18cm (19 cm including

caption); the reduction or enlargement required should be indicated (and preferably uniform); orig-

inals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and

final size. A metric scale should appear with all illustrations, otherwise magnification or reduction

should be given in the caption; if the latter, then the final reduction or enlargement should be taken

into consideration.

All illustrations, whether line drawings or photographs, should be termed figures (plates are not

printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items

of composite figures should be designated by capital letters; lettering of figures is not set in type and

should be in lower-case letters. If Letraset is used authors are requested to use Helvetica-style letter-

ing, if possible.

The number of the figure should be lightly marked in pencil on the back of each illustration.

5. REFERENCES cited in text and synonymies should all be included in the list at the end of the

paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable):

(a) Author’s name and year of publication given in text, e.g.:

‘Smith (1969) describes... .’

‘Smith (1969: 36, fig. 16) describes... .’

“As described (Smith 1969a, 1969b; Jones 1971)’

“As described (Haughton & Broom 1927) .. .’

‘As described (Haughton et al. 1927)...’

Note: no comma separating name and year

Pagination indicated by colon, not p.

names of joint authors connected by ampersand

et al. in text for more than two joint authors, but names of all authors given in list of references.

(b) Full references at the end of the paper, arranged alphabetically by names, chronologically within

each name, with suffixes a, b, etc., to the year for more than one paper by the same author in

that year, e.g. Smith (1969a, 1969b) and not Smith (1969, 1969a).

For books give title in italics, edition, volume number, place of publication, publisher.

For journal article give title of article, title of journal in italics (according to the World list of scientific periodicals. 4th ed.

London: Butterworths, 1963), series in parentheses, volume number, part number in parentheses, pagination (first and

last pages of article).

Examples (note capitalization and punctuation)

BuLLouGH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan.

FiscHER, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140.

FiscHER, P. H., DuvaL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie

expérimentale et générale 74 (33): 627-634.

Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and

Magazine of Natural History (13) 2 (17): 309-320.

Koun, A. J. 19606. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of

the Bingham Oceanographic Collection, Yale University 17 (4): 1-51.

TurELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: ScHuLTzE, L. Zoologische und anthro-

pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika ausgefiihrt in den Jahren

1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270.

(continued inside back cover)

ANNALS OF THE SOUTH AFRICAN MUSEUM

ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

Volume 97 Band

December 1985 Desember

Part 2 Deel

KOLPOCHOERUS PAICEAE

(MAMMALIA, SUIDAE) FROM SKURWERUG,

NEAR SALDANHA, SOUTH AFRICA, AND ITS

PALAEOENVIRONMENTAL IMPLICATIONS

Q. B. HENDEY

H. B. S. COOKE

Cape Town Kaapstad

The ANNALS OF THE SOUTH AFRICAN MUSEUM

are issued in parts at irregular intervals as material

becomes available

Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000

Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

word uitgegee in dele op ongereelde tye na gelang van die

beskikbaarheid van stof

Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000

OUT OF PRINT/UIT DRUK

i, AGES), 523), SCD, 2-5, 8, tnd), 5S0-S, 5, 7-9),

Gy tt), TO), 8, CC®, 7), 1003),

INGE, 5, 7, (tos), WIE), 15S), DAO), V7, SB), BAG), 38, 35), 45(0)

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 078 8

Printed in South Africa by In Suid-Afrika gedruk deur

ihe WKustica Press. Pty etd Die Rustica-pers, Edms., Bpk.,

Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap

KOLPOCHOERUS PAICEAE (MAMMALIA, SUIDAE)

FROM SKURWERUG, NEAR SALDANHA, SOUTH AFRICA,

AND ITS PALAEOENVIRONMENTAL IMPLICATIONS

Q. B. HENDEY

South African Museum, Cape Town

H. B. S. Cooke

2133-154th Street, White Rock B.C., Canada V4A 4585

(With 17 figures and 4 tables)

[MS accepted I May 1985]

ABSTRACT

A Kolpochoerus skull from a consolidated coastal dune at Skurwerug near Saldanha Bay is

identified as an early Pleistocene representative of the southern African K. paiceae lineage. It is

more primitive than the middle Pleistocene K. paiceae from the nearby Elandsfontein fossil site,

and is in a comparable evolutionary state to 1,0 to 2,0 Ma K. limnetes from East Africa. Its

suggested age is in accord with that of the high sea-level of the eustatic cycle Q2, during which the

Skurwerug dune was apparently formed under circumstances similar to those that have

influenced the formation of the nearby late Pleistocene—Holocene Spreeuwal dune plume. The

Skurwerug K. paiceae suggests the presence of woodland or forest patches in the Saldanha region

during the early Pleistocene.

CONTENTS

PAGE

MNPROGI CHO Meee er ey ie le ih are a ER 2a 9

Sy Stematicsandi descriptions ss senses aaa a eee oe aa 10

thesBlandstonteimmiratenale sa sense sees os soe. Sl

iihterZululandspecieninn ns). aes dent ees we ene oe 39

DISCUSSION rR er ier Lee EC Buys ii wn annul 8 40

Alacocnvinonmentalumplicatlons ane eeea nae ee ec 43

INCKMOWIEASEMEMUS Heres 2) oe ac heen dt sine oles 53)

INCKENCICC Spine earn See sn A rae adc eh Sas 8 Sa D3)

INTRODUCTION

During 1979, the South African Museum acquired a small collection of fossil

mammals from a large excavation that was being undertaken in a consolidated

calcareous dune at Skurwerug, near Saldanha in the south-western Cape Province

(Fig. 1). Access to the excavation was restricted for security reasons and it was

not possible to determine the nature of the fossil occurrence, or to undertake a

systematic investigation of the deposits from which the material was derived.

Ann. S. Afr. Mus. 97 (2), 1985: 9-56, 17 figs, 4 tables.

10 ANNALS OF THE SOUTH AFRICAN MUSEUM

Such occurrences are not uncommon in the region and the Skurwerug specimens

would have been of little interest and significance but for the fact that the

collection included a particularly fine skull of an extinct pig. This specimen was

found by Mr Albert W. Abraham and donated to the South African Museum by

Mr P. Hutton. Other fossils from the Skurwerug excavation, which are all in a

fragmentary state, were not necessarily associated with the pig skull and they will

not be considered in detail here. The mammals include Equus cf. capensis,

Connochaetes sp., Gazella sp. and Raphicerus sp.

As soon as the specimen was received it was recognized that the general

morphology of the skull and cheek teeth served to place the specimen in the genus

Kolpochoerus (formerly Mesochoerus) and the third molars were comparable in

size with material from East Africa assigned to K. limnetes (Cooke 1976; Harris &

White 1979). The marked reduction in the premolars indicated affinity with

K. paiceae from South Africa but the third molars are smaller than in the large

sample from Elandsfontein, suggesting that the Skurwerug specimen may be an

early form of this lineage. Despite the lack of a detailed account of this specimen,

it has come to be accepted as indicative of an early Pleistocene age for the coastal

dune from which it was derived (Hendey 1981a, 1983a; Rogers 1982, 1983). It is

the purpose of this paper to identify and describe the specimen and to consider its

implications in respect of the geological succession and palaeogeography of the

Saldanha region.

SYSTEMATICS AND DESCRIPTION

ORDER ARTIODACTYLA

Family Suidae

Genus Kolpochoerus van Hoepen & van Hoepen, 1932

Type-species: Kolpochoerus sinuosus van Hoepen & van Hoepen, 1932

(= Kolpochoerus paiceae (Broom, 1931)—see Cooke (1978)).

Discussion

Material assigned to Kolpochoerus has been recovered from many late

Pliocene to middle Pleistocene localities in Africa, and it has been reviewed in

recent publications by Cooke (1978), Cooke & Wilkinson (1978), White & Harris

(1977), Harris & White (1979), and Harris (1983). The latter authors recognized

only two valid species, whereas Cooke distinguished five. Many pig genera and

species recorded prior to 1977 have been synonymized with Kolpochoerus and

with the few species recognized by those authors cited above. The earlier

nomenclatural proliferation, and the subsequent difference of opinion on the

number of valid species, was due at least in part to the variation in dental

characters exhibited by Kolpochoerus populations over its three-million-year time

range so that recorded material represents temporal and, apparently, also

regional variants. |

KOLPOCHOERUS PAICEAE FROM SKURWERUG. SOUTH AFRICA 11

St Helena

Bay

,Langehaanwe

Skurwerug

Saldanha

Bay AN *Elandsfontein

Langebaan

Lagoon

Atlantic Bok Baai

Ocean Duinefontein

Cape Town 6

Swartklip 345

False

0 40 Bay

Cape Hangklip

Fig. 1 The location of Skurwerug.

12 ANNALS OF THE SOUTH AFRICAN MUSEUM

The only two species recognized by Harris & White (1979) are K. limnetes

and K. majus. Of these, K. limnetes is considered to be represented by a

continuum from an early form with small third molars to an ‘advanced’ form with

elongate third molars. However, Cooke (1976, 1978) regards the advanced form

as a separate species, K. olduvaiensis, but it would be inappropriate to debate this

matter here as it is the taxonomy and not the morphology that is in question.

Accordingly, to avoid unnecessary confusion in the present account, ‘typical’

K. limnetes refers to the material ascribed to this species both by Cooke and by

Harris & White, while ‘advanced’ Kolpochoerus (without a species designation)

denotes the advanced K. limnetes of Harris & White or the K. olduvaiensis of

Cooke. Kolpochoerus ‘majus’ is contemporary with the ‘advanced’ Kolpochoerus

but the third molars are in the size range of the ‘typical’ K. limnetes; they differ in

being relatively more hypsodont, have more crenulated enamel and the crowns of

the teeth have a distinctive lateral bulge just above the enamel line. Cooke &

Wilkinson (1978) consider that the name ‘majus’ is a nomen vanum but they

recognize the reality of this entity.

One of the differences of opinion referred to above concerns the recognition

by Cooke (1976, 1978) of a South African lineage that is represented by the

species K. paiceae. This species is recorded from middle Pleistocene occurrences

at Elandsfontein, Cornelia, the Vaal River gravels (Cooke & Wilkinson 1978),

and in Zululand (McCarthy & Orr 1978). It has hitherto been known from only

fragmentary specimens, with the largest assemblage coming from Elandsfontein,

which is 20 km south-east of Skurwerug (Fig. 1). This material does not show the

distinctive characteristics of K. ‘majus’ but there has hitherto been only slender

evidence for distinguishing it from the ‘advanced’ Kolpochoerus of East Africa.

Harris & White (1979) believe that until there is an unequivocal separation of

K. paiceae and K. limnetes on dental and cranial evidence, K. limnetes should

stand as the senior synonym of a single species.

It cannot be expected that the discovery of the remains of a single individual,

no matter how complete and well preserved, would settle the status of K. paiceae,

but the Skurwerug specimen does lend support to Cooke’s belief in a South

African Kolpochoerus lineage. This specimen is evidently a more primitive form

than that recorded from Elandsfontein, but it is very similar in some respects to

specimens in the Elandsfontein assemblage, and to the type material from the

Vaal River gravels; it is accordingly identified with K. paiceae.

In the final analysis, the specific identity of the Skurwerug Kolpochoerus is

not critical since its significance lies chiefly in indicating a somewhat earlier

presence of this taxon at the southern extremity of Africa than has been known

hitherto, with corresponding implications in respect of the relative dating of the

deposits from which it was derived.

Kolpochoerus paiceae has been known hitherto only by its dentition and by

parts of the mandible so the skull characters of SAM—PQ2166 must be considered

in relation to the skulls from East Africa. The general morphology of the East

African material has been discussed in several papers (e.g. Cooke & Wilkinson

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 13

1978; Harris & White 1979), with the latter publication including illustrations of

several specimens. These are KNM—ER 409 (pl. 10, centre, pl. 12, top), KNM-—

ER 212 (pl. 12, left), KNM—-ER772 (pl. 10, top), KNM-—ER 788 (pls 8, 9) and

KNM-ER 1085 (pl. 10, bottom), all from the Koobi Fora succession. The first

two are reasonably representative of the male skull, though they are larger than

the few cranial specimens from the Shungura Formation, still undescribed (Cooke

& Coppens in press). KNM—ER778 and 772 came from the upper part of the

Koobi Fora succession (‘Metridiochoerus compactus’ zone) and represent the

large ‘advanced’ Kolpochoerus that Cooke & Wilkinson (1978) regard as a

distinct species, Kolpochoerus olduvaiensis. Female cranial remains are rather

rare and the only female of K. limnetes so far illustrated is the incomplete

specimen KNM-ER 1085, apart from a rather crushed skull of this species from

Bed I, Olduvai (KNM-—OLD FLK NNI 1235), formerly the type of Ectopota-

mochoerus dubius (Leakey, 1965, pl. 25). An almost intact undescribed large

cranium from Peninj, Lake Natron, has been regarded by Cooke & Wilkinson

(1978) as the female of Kolpochoerus olduvaiensis. Sexual dimorphism is strong,

with the female having lesser development of the zygomatic arches and smaller

canine flanges than in the male. The degree of difference is comparable with, or

greater than, that seen in the forest hog Hylochoerus meinertzhageni, which is

apparently the closest living relative of Kolpochoerus although it has special

features of its own (Cooke 1976; White & Harris 1977).

Kolpochoerus paiceae (Broom, 1931)

Notochoerus paiceae Broom, 1931: 167, fig.1

Kolpochoerus sinuosus van Hoepen & van Hoepen, 1932: 59, figs 72, 75-77. Cooke, 1974: 73,

fig. 5.

Notochoerus capensis Shaw, 1939: 85, fig. 8 (partim).

Mesochoerus paiceae Shaw & Cooke, 1941: 293, pl. 54 (1-2). Cooke, 1949a: 35, figs 18, 19;

1949b: 44, pls 22, 23; 1976: 253, fig. 1. Keen & Singer, 1956: 352, pl. 33E. Leakey, 1958: 13.

Cooke & Maglio, 1972: 310.

Metridiochoerus andrewsi Arambourg, 1943: 473 (partim); 1947: 352, fig. 42 (partim).

Tapinochoerus meadowsi Cooke, 1949a: 31, fig. 15 (partim).

Mesochoerus lategani Singer & Keen, 1955: 170, fig. 1. Keen & Singer, 1956: 350, pl. 33A—D.

Cooke & Maglio, 1972: 310.

Kolpochoerus paiceae Cooke, 1978: 460. Cooke & Wilkinson, 1978: 462.

Mesochoerus limnetes Harris & White, 1979: 37 (partim).

Diagnosis

A species of Kolpochoerus comparable in size with the living forest hog and

similar in skull size and general morphology to the ‘typical’ K. limnetes of East

Africa, from which it differs in the following: relative reduction in the length of

the snout and symphysis; relative elevation of the occipital condyles above the

palatal plane, accompanied by reduction in height of the occiput; slight elevation

of the orbits; mandibular corpus very robust; anterior border of symphysis very

gently curved, resembling that of Hylochoerus; three pairs of lower incisors but

upper reduced to two pairs; three reduced upper premolars and only two !ower

premolars. Third molars comparable in size and structure to those of ‘advanced’

14 ANNALS OF THE SOUTH AFRICAN MUSEUM

K. limnetes (sensu Harris & White 1979) or K. olduvaiensis (sensu Cooke &

Wilkinson 1978). Male upper canines shorter and more strongly curved than in

normal K. limnetes; female canines small but with normal open roots.

Holotype

Mandible fragment with right third molar and stump of second molar, from

Vaal River gravels, Windsorton, Cape Province. McGregor Museum, Kimberley,

MMK 4088.

New material

SAM-—PQ2166: the incomplete skull and mandible of an adult, lacking the

anterior parts of the nasals, some of the premaxillae and maxillae, all of the right

and parts of the left zygomatic arches, left I°, right P’, left ascending ramus and

left Is.

Locality and horizon

Skurwerug, a consolidated dune on the farm Osfontein, 2 km inland from the

eastern shore of Saldanha Bay in the south-western Cape Province. This dune is

included in the Langebaan Limestone Member of the Bredasdorp Formation, as

it is defined by Rogers (1983).

Description

The Skurwerug skull is that of an adult, probably female, in which all the

permanent teeth are erupted, with only the most posterior pillars of the M? talon

being unworn. In size it is comparable with that of the forest hog and smaller than

the male specimens from the Shungura Formation, which, in turn, are smaller

than those from Koobi Fora.* Principal measurements are given in Table 1* in

comparison with the earliest of the Omo male crania, L193—109 from Shungura

Member C8, the smallest (KNM-—ER 212) and the largest and most advanced

(KNM-ER 788) of the male crania from the Koobi Fora Formation. Also

included are applicable measurements for the partial female cranium

KNM-ER 1085 from Koobi Fora and for the large undescribed female cranium

from West Natron (Peninj),* numbered A67—384 in the collections of the

National Museum of Tanzania, Dar-es-Salaam. It should be noted that these

specimens cover a substantial time range.

Cranial morphology (Table 1; Figs 2, 3)

In profile the Skurwerug cranium shows the marked angulation between the

rostral region and the frontals that is characteristic of Kolpochoerus. The back of

the braincase is not unduly elevated but the occipital condyles are much higher

above the palatal plane than is the case with ‘typical’ K. limnetes and this

*Personal observations by HBSC, as also are all measurements cited for East Africa; there

may be minor differences from measurements given by Harris & White (1979) or by Harris (1983)

but they are insignificant.

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 103)

elevation is accompanied by reduction in the height of the occiput itself. This is a

unique feature not seen in other specimens. The braincase itself is very slightly

concave between the orbits with the orbital rim rising just above the level of the

frontals, whereas in the ‘typical’ form they are normally below it. The cranial

vault is very gently convex and broad above the temporal fossae and the temporal

ridges are not strongly marked. However, the doming is less than in the ‘typical’

K. limnetes but the vault is not depressed as it is in the males of the advanced

form. The morphology is very similar to the condition seen in the Peninj cranium,

which suggests that there may be a sexual difference comparable to that in the

living forest hog. The supra-orbital foramina are in line with, or even slightly in

front of, the anterior margins of the orbits, as in the ‘typical’ form, whereas in the

advanced form these foramina are situated a little farther back.

Although the right zygomatic arch is lost and the left one is damaged, it is

clear that there was no great lateral expansion as in the males of K. limnetes, in

which the anterior margin sweeps outwards sharply until it is more or less

perpendicular to the sagittal axis. In PQ2166 the form of the zygoma is like that of

the Koobi Fora female ER 1085 and not as abruptly protuberant as in the Peninj

female; indeed it is rather reminiscent of the zygoma of Sus, although more

expanded laterally. The ventral margin must have been well above the occlusal

plane, so there was little of the lateral ‘droop’ that is seen in the early males of

K. limnetes.

The snout is damaged and the tips of the nasals are missing but the nasal on

the right side is preserved to the level of the canines. It is strongly arched in

transverse section along its preserved length. The nasals were widest above the

back of the canine flange and there is a distinct, but not strong, lateral projection

of the nasal-maxillary junction over the infra-orbital region. In males of early

K. limnetes the lateral projection is marked and is often accompanied by the

development of rugose areas on the nasal and adjoining maxillary similar to those

found in males of Potamochoerus porcus. These rugose areas have not been noted

in the later K. limnetes, although the overhang persists until it disappears

effectively in the advanced form. There is no sign of any rugosity in the

Skurwerug specimen, as is to be expected in a female.

The right canine flange is lacking and the left one is damaged but sufficient

remains to show that it was weakly developed in comparison with the strong

Potamochoerus-like bosses seen in ‘typical’ male K. limnetes. The flange in

PQ2166 curves gently out from the maxilla well in front of the infra-orbital

foramen and is intermediate in form between those of female Potamochoerus and

female Hylochoerus. There is a weak lateral crest about 20 mm in length that

arises some 25 mm from the root of the flange at the maxilla but the flange is not

at all inflated. No trace exists of any dorsal crest and it was probably absent. The

canine is moderately small and emerges more or less parallel to the palatal

surface. In the female cranium ER 1085 the canine flange is also small but has a

marked lateral crest and there is a more prominent gutter between the front of the

flange and the wall of the maxilla. In the Koobi Fora female the canine itself was

16 ANNALS OF THE SOUTH AFRICAN MUSEUM

{A it i

whe

ee A

Fig. 2. Dorsal, lateral and ventral views of the Kolpochoerus paiceae skull from Skurwerug

(SAM-PQ2166).

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 7)

TABLE 1

Dimensions (in mm) of the Skurwerug cranium, SAM—PQ2166,

and some other Kolpochoerus crania.

Locality Skurwerug Peninj Koobi Fora Omo

Sex g 2 2 3 3 3

Number PQ2166 A67-384 ER 1085 ER 212 ER 788 L193-109

Estimated age (Ma) 2 c. 0,9-1,3 ve We 1S DES

Vertex length 425 SS — 505e 602 465e

Basilar length Bie 465 — 436e 483 c. 400e

Palatal length 263e 341 €, ZNVE SWE 354 c. 280e

Bizygomatic breadth 260 318e — 278 364 e, AVE

Crest breadth 112 & ISVS — Ze 145 120e

Parietal constriction 46 62e — 65,5e 48,5 56e

Frontal breadth 134 143e — 133 159 We

Ocular breadth 104 WAS) 90 108,5 110 105

Maxillary breadth at M? 87,5 104,5 82 83 101,5 96

Post-canine breadth eC, SI 74 58e 79 84 76,5

Flange breadth 113e 123e 108e 140 180 150e

Precanine breadth Gals Vil — 73e OLS) —

Height of occiput 131 212e — 183,5 203 Wee

c. = approximate; e = estimated

*The horizons for ER 1085 and ER 212 are not certain but the age is probably about 2,0 Ma

(+ 0,5 Ma).

also directed less laterally and slightly downwards relative to the palatal plane. In

the Peninj cranium the flanges are damaged but were essentially similar except

that they are distinctly tapered towards the canine root area. At the canine

alveolar margin in ER1085 the opening is about 17 mm wide and 15 mm high

compared with 23 mm and 19 mm in the Skurwerug cranium; in the Peninj

cranium approximate estimates are 28 mm and 23 mm respectively. In all three

cases the canine had a strong dorsal groove.

The palate in ‘typical’ K. limnetes resembles that of the forest hog in general

but is proportionally a little narrower. The tooth rows are usually slightly curved

so that the palate is widest between the second molars and narrowest at the third

molars; the anterior premolars may curve gently outwards beneath the expanded

canine flanges. In PQ2166 the palate broadens anteriorly from M’, much as in

Hylochoerus. The palatine foramina, which normally lie level with the anterior

pillars of M°, are slightly farther back in the Skurwerug cranium. There are two

specimens of the Elandsfontein K. paiceae in which the palatine foramen is

preserved. In one (SAM-PQ-ESS) it is positioned as in PQ2166, but in the other

(SAM-—PQ-ES20) it is even more posteriorly situated, being opposite the

posterior pillars of the M® trigon (Singer & Keen 1955, pls 23A, 24A). In the

female cranium from Koobi Fora, KNM-—ER 1085, and also in the Peninj female

cranium, the palatine foramina are level with the front of M’, so this is not merely

a sexual difference. In the palate of PQ2166, the U-shaped palatonarial border is

situated only a short distance behind M’, which is apparently a primitive

ANNALS OF THE SOUTH AFRICAN MUSEUM

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA

‘({uruag) afeway (sisuamanpjo =) SaJUU] SN4ABOYIOd]OY ,paouvapy, °q ‘(99IZOd-WVS) avaa1nd sndgoyrodjoy ‘dD ‘(88L YA-WNM)

‘d “(60h UA-WNM) Saou snssoysodjoy jeardsL, vy “s[[nys snsaoysodjoy “¢ ‘314

(SisuaIvAnpjO =) Sajauiuy sndaoyrodjoy ,pooueapy,

20 ANNALS OF THE SOUTH AFRICAN MUSEUM

characteristic, since according to Harris & White (1979) this distance is increased

in advanced specimens. In the Koobi Fora female the palatonarial border is very

much in the same position as in the Skurwerug cranium whereas in the Peninj

cranium it extends much farther back.

The premaxiila is usually badly damaged but it is partially preserved in a few

specimens of ‘typical’ K. /imnetes, in the advanced Koobi Fora KNM-ER788,

KNM-ER772 and in the Peninj cranium. It is subtriangular in shape with a

rounded apex and the three incisors on each side lie almost in a straight line one

behind the other as in Sus or Potamochoerus. I? is small but is normally retained

into old age, although in KNM—-ER/772 it was shed during life. In PQ2166 the

premaxilla is shorter and broader with a more arcuate border, only two premolars

are present on each side and I’ is more laterally situated in relation to I'. There is

no trace of IP or of any scar. The I' and I’ are essentially similar in morphology to

the corresponding teeth of Potamochoerus porcus and to teeth attributed to

K. limnetes.

Upper canines

The upper canines of ‘typical’ male K. limnetes somewhat resemble those of

Hylochoerus but are shorter and relatively stouter, curving laterally at the tips

with less of a backward or upward sweep than in the forest hog. There is a broad

wear facet cutting across the front of the tooth. In structure the canines are also

like those of Hylochoerus, with a strong dorsal longitudinal groove and shallower

grooves on the anterior and posterior surfaces. There is a robust ventral band of

ribbed enamel, a narrow inset band of smooth enamel on the anterior edge of the

tusk and another on the posterior face, about 1 cm above the rear ridge of the

ventral band. In Hylochoerus the diameter of the tusk is greater in the vertical

direction whereas in Kolpochoerus limnetes the dimensions are about equal or the

tusk is dorso-ventrally flattened. In the few specimens attributed to females, the

canines are smaller, emerge with a downward and lateral component and taper at

the base to form closed or nearly closed roots. The teeth are less curved than in

males and the tip is cut by a small wear facet. In unworn teeth the tip of the

female canine is covered with thick enamel but it thins posteriorly, extending

farthest on the ventral side whereas on the dorsal side a V-shaped area of dentine

extends towards the tip, coinciding with a dorsal groove in the root area. There is

a good sample of both male and female canines from Olduvai Bed I in which the

female canines measure about 9-10 cm in length from tip to root and the maxi-

mum and minimum diameters of the cross-section are 19-22 mm by 12-16 mm;

the male canines are 20—25 cm in length, have open root areas, and cross-section

diameters in the range 29-45 mm with the vertical measurement sometimes

greater than the antero-posterior diameter. Male canines from the Shungura

Formation up to Member G are about the same size, as also are those of ‘typical’

K. limnetes from Koobi Fora. In advanced Kolpochoerus the canines are longer

(c. 30-35 cm) and stouter (50 mm by 39 mm in ER788).

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA wi

The canines of the Skurwerug cranium project 5 cm from the alveolar

margin and the root area may occupy about another 4-5 cm. The curvature is

fairly sharp and they are directed forwards and laterally almost in the palatal

plane. At the alveolar margin the antero-posterior diameter is close to 24 mm

and the dorso-ventral diameter 19-20 mm. There is a strong dorsal groove

and weaker anterior and posterior grooves, as well as a strong ventral band of

very weakly grooved enamel. There is nothing to suggest that the roots are

closed. The tips are cut by a moderately broad wear facet. The size and form of

these canines is thus a little more ‘masculine’ than are the Olduvai teeth, yet both

in size and in structure they fall short of being typically male. This may be taken

as indicating a degree of sexual dimorphism less marked than in ‘typical’ K.

limnetes or even in the ‘advanced’ Kolpochoerus represented by the Peninj

female. On the other hand, the convergence in structure is not quite as great as in

female Hylochoerus.

TABLE 2

Dental dimensions of Kolpochoerus paiceae specimen, SAM—PQ2166 from Skurwerug.

Upper Lower

Left Right Left Right

Length premolar series S17 33e AB) 5) AS)

Length molar series 91,5 94,2 99,9 99e

Length cheek teeth series 23} 126e 124,3 124,5e

Canine—antero-posterior diameter 24,5 DoS 21,6 22,4

—dorso-ventral diameter 18,9 AQ) II 18,1 ike3.3)

—length from alveolus 51 52) 67e 67

Diastema C—P?/P3 47 — SIS S558

P2—Length (L) 8,2 lost absent absent

—Breadth (B) 5,0 — — =

—Height (H) Syn — — —

P3—L ils) 5) ipl 11,0

—B 11,0 10,3 7,0 6,9

ll 6,3* 7,4¢ eis 8,0*

P4—L OF 1O3 IS) 72 14,6

—B 13.3 13,0 Lil 11,0

—H 5g" V3" 20" 8,5*

Mi—L 17,0s 17,4 I5.8) @ IOUS

—B 16,3s IS @, Bs, 14408

—H d Dead 3,0d 3)510|

M2—L 23,0 Mp) M31 MB)

—B DO 19.9 17,8 17,8

=H 20,0* ey Saou 8,07

M3—L 50,0 ell 60,2 60,0e

—B 25,0 24,1 US) 2 DOT

=H 25,0 ZS AO Ze) 22,0

—Trigon(id) length 28,7 29,8 US) BS) J)

c. = approximate; e = estimated; s = stump; d = mainly dentine

* =incomplete (worn)

22 ANNALS OF THE SOUTH AFRICAN MUSEUM

Upper cheek teeth (Table 2; Figs 2, 4, 5)

The upper dentition of the Skurwerug cranium is remarkable not only in

having the incisors reduced in number (to two pairs) but also in having the

premolars reduced both in size and in number (to three upper and two lower

pairs). The Peninj cranium has the upper premolars reduced to only two pairs,

but they are relatively enlarged rather than reduced. PQ2166 shows no sign of P’,

nor of any scar to indicate that it was ever present. In K. limnetes P' was present in

the rare female specimen KNM-—ER 1085 (although only the socket remains) and

in the Olduvai Bed I palate FLK W 626, so its retention may well have been a

normal feature in the ‘typical’ female. Although it is lacking in many of the male

specimens, it is certainly present in some cases. At Olduvai it occurs in a crushed

male skull FLK NN I 177, lying 0,5 cm in front of P’, as in the female skull

mentioned above; at Koobi Fora it was present in the male skull KNM—ER 212,

lying in contact with P?; and in the Shungura Formation was present in the male

cranium L193—109 from Member C8, lying 2,5 cm in front of P?.

Fig. 4. Occlusal view of the left upper cheek teeth of the Kolpochoerus paiceae skull from

Skurwerug (SAM-—PQ2166). Natural size.

The P? of PQ2166, which is preserved on the left side only, has the exposed

parts of its two roots fused. The normal P? has three cusps, one behind the other,

with the central cusp the largest. In PQ2166 there are only two cusps, well fused

and worn so that the dentine areas almost merge. The anterior cusp is longer than

the posterior one and tapers anteriorly, giving it a triangular outline in occlusal

view. It somewhat resembles the P! of other specimens although the true P' is

single-rooted.

The right P? is intact, that on the left slightly damaged, and both are well

worn. The tooth is triangular in occlusal view, with one root anteriorly and two

roots posteriorly. There is a large main cusp (paracone) on the labial side,

another postero-lingually (hypocone), and a very small cusp posteriorly. An

apparently small distinct cusp anteriorly has been linked to the central cusp

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA pie)

through wear. The rather Sus-like arrangement is normal for Kolpochoerus

limnetes but the tooth is considerably smaller than in any of the other material

referred to this species (Fig. 5).

The P* has a more complex crown, with an almost molariform morphology. It

apparently has three roots, and the tooth is broader posteriorly. The crown

morphology is somewhat obscured by wear and is most clearly seen in the right P*.

There are two pairs of more or less equally sized cusps anteriorly and posteriorly,

and a small accessory cusp situated medially on the lingual side. There is a

posterior cingulum. The enamel of the principal cusps is slightly crenulated. The

teeth of ‘typical’ K. limnetes are normally dominated by the enlarged paracone

and protocone with a lesser metacone; cingular cusps are commonly developed

and add to the complexity of the pattern but the essentially quadritubercular

pattern seen in the Skurwerug dentition has not been noted in the East African

material. The P* is substantially smaller than in any of the East African specimens

(Fig. 5).

Both left and right M' of PQ2166 are heavily worn and details of their cusp

morphology are no longer discernible. However, it is clear that this more or less

rectangular tooth has paired roots anteriorly and posteriorly that supported

paired cusps in an arrangement that is typical of the M!' of Kolpochoerus and

other pigs.

The M? of PQ2166 also has a typically Kolpochoerus morphology. This tooth

has paired roots anteriorly and posteriorly, supporting paired cusps that are

separated by a smaller median cusp, and flanked anteriorly and posteriorly by

well-developed cingula that are fused with small median cusps. In size both M!

and M? lie near the bottom of the range of East African K. limnetes material

(Fig. 5).

In PQ2166, the M? have the pillar-like structure that is characteristic of

Kolpochoerus. The trigon is morphologically similar to M’, and consists of two

sets of paired cusps separated by a smaller median cusp, an anterior cingulum that

is fused to a small median cusp, and another small median cusp posteriorly. The

trigon cusps are crenulated. The talon is shorter than the trigon and its cusps are

less symmetrically arranged. The most anterior of the talon cusps is a median one

that abuts the posterior median cusp of the trigon. It is flanked by a smaller buccal

cusp and a more posteriorly situated and larger lingual cusp. This arrangement is

repeated in a second row of slightly smaller cusps. Finally, there is a single

postero-buccal cusp on the left M’. There is no crenulation of the talon cusps. The

M? has a heavier cement cover than M7. In size the teeth are appreciably smaller

than all of those from Elandsfontein and fall well within the range for the sample

from Olduvai Bed I (Fig. 5), or that from the Shungura Formation, Member G.

Mandible

The general form and characteristics of the mandible of PQ2166 conform to

those seen in ‘typical’ K. limnetes but there are differences in the shape and

proportions of the symphysis and in the reduction of the premolars in the

BREADTH (mm)

ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 5.

0-9 O

ice ar. %, O

sb aa a e O

Do + /

Re / O @e oO

Va tA. + Boo p O *

+ a +h fo i an, fo. fo O O

hes ¥ # soe Cornelia

\ + eas

OA SKURWERUG

Limits of observed range ELANDSFONTEINS @ Undescribed

for Omo Member G O Described

O PENINJ

60 65

45 50 55

LENGTH (mm)

+ OLDUVAI Bed!

Length/breadth measurements of upper cheek teeth of Kolpochoerus paiceae from

Skurwerug, Elandsfontein and Cornelia, K. limnetes from Olduvai Bed I, and an advanced

Kolpochoerus from Peninj.

D5)

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA

9}eWINSa = 9 ‘ojeultxo1dde = 9

ia ee eee Se ee eee

9°66 90T POl IED 8°28 9721 — Oll 2 666 SOLIOS IRJOI\]

SCP S‘Or Iv 6G SE +67 CD O% ‘97 Solas Ie[OWsIg

8 69 18 96 79 16 aCL 379 9¢ d-D eusaiseiq

IS GS D OF SS ey 7 39¢ SS br FW 18 ssouyory) Mer

S‘r9 6) D 19 rae) LS SOL 99 2 agg S‘9 W/'d 18 148104 mer

Sg 5°16 £6 LOL EL 76 98 6 SEL UONSINSUOD

ran 671 rol S6CL 76 O€T ZOL S0TT —- 8S‘ €6 Yipesiq osuey ourued

ral Srl SZL al SII 6€1 901 2 901 ‘2 OI sisdyduiAs Jo yy3uaT

ae = p81 v6l ae a= = 3 orl So[Apuod Jo JYSIOP]

= = Sse L9p = OIp = = ZSE y}3ug] [PIO],

ee ee ee ee en ee ee

CC 07 é Le ST 3°09 ‘9 8°09 ‘2 r (VIN) asp pawwunsy

S-p9T €-LZ INN 374 AI SW

ONO OWO field 966 wa 0001 6L0LTLWWA SL99IA 8260TH 991ZOd 4aquinyy

Ott@) e104 Iqo0y TRANPIO ulojUOjspue[q sniominys A]NDIOT

Pa a a a ION ee ee ad en

‘So[qIpueU snia0yIOdjoy 13940 sUIOS puke

‘991TOd-WVS ‘21qIpuew Sniominys oY} JO SUOISUDTIG,

¢ FTav i,

26 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 6. Lateral and dorsal views of the Kolpochoerus paiceae mandible from

Skurwerug (SAM-—PQ2166).

Skurwerug specimen. Kolpochoerus shares with Hylochoerus a stout mandible

with great width across the canines, only modest constriction behind them, and

inflation of the corpus lateral to Ps, narrowing again fairly abruptly opposite the

middle of M3. In Hylochoerus the mandible of the female is a little less robust

than in the male but the differences are slight. In the forest hog the breadth across

the canines is about the same as the length of the symphysis—relatively slightly

narrower in females—whereas in Kolpochoerus limnetes the symphysis is longer

than the breadth across the canines. In the forest hog the back of the symphysis

lies well in front of the anterior premolar and the gap between it and the back of

the canine is about equal to the sum of the lengths of the three molars; in ‘typical’

K. limnetes the symphysis begins about level with the anterior premolar and the

gap is about as long as the third molar, or a little longer. In the forest hog the

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 27

Fig. 7. Kolpochoerus mandibles. A. ‘Typical’ Kolpochoerus limnetes (Omo L64-5 and incisors

of L36-27). B. ‘Advanced’ Kolpochoerus limnetes (= olduvaiensis) (KNM-ER 1314).

C. Kolpochoerus paiceae (SAM-—PQ2166). D. Kolpochoerus paiceae (SAM-PQ-E16675 and

E20928).

28 ANNALS OF THE SOUTH AFRICAN MUSEUM

anterior border of the symphysis is only slightly curved whereas in K. limnetes it is

more nearly semicircular and thus extends farther in front of the canines. The

junction of the two rami lies slightly below the lower surface of the corpus in

K. limnetes and the symphysis rises gently to the incisor border, almost as a

straight line in profile, although there is a very weak shelf supporting the

procumbent incisors, much as in Hylochoerus.

Compared with ‘typical’ Kolpochoerus limnetes, the Skurwerug mandible has

a shorter symphysis, only very slightly longer than the breadth across the canines,

but its origin is still about level with the front of the premolars. The distance from

the front of the premolars to the back of the canine is also reduced, being very

slightly less than the length of the M; alone. The anterior border of the symphysis

is less arcuate, although not as flattened as in Hylochoerus. The profile of the

symphysis is rounded, very much as in Phacochoerus and, like the latter, the

incisors are set in a shelf-like projection. All three pairs of incisors are retained,

with I3 not reduced (as it is in Hylochoerus) and the structure of the incisors is

basically similar to that in Sus, but perhaps closer to that of Hylochoerus. The

ascending ramus is a little different from the few K. limnetes jaws in which it is

preserved, for in these specimens the anterior edge is more upright than in

PQ2166 and antero-posteriorly narrower, especially in the ‘advanced’ form (e.g.

see Harris & White 1979, pl. 11—KNM-ER 1314). In PQ2166, also, the rear

edge of the ascending ramus is farther back from the end of the cheek teeth,

giving a ‘stretched-out’ appearance to the ascending ramus. The height to the

condyle is also less than in ‘typical’ K. limnetes and considerably less than in the

advanced form. Some measurements covering a substantial time range are given

in Table 3.

The lower dentition (Table 2; Figs 2, 8, 9)

The lower canines of PQ2166 emerge from the symphysis almost perpendicu-

lar to the axis of the mandible and rise at a low angle, which is typical of

Kolpochoerus limnetes. They extend 6 to 7 cm from the alveolar border and are

very similar to the canines from Olduvai Bed I that are attributed to females,

although perhaps a little more robust. The cross-section is subtriangular with a

somewhat rounded anterior keel. The posterior facet is 18 mm broad at the

alveolar margin, the antero-dorsal facet 21 mm broad and the antero-ventral one

22 mm. The two larger faces are covered with thin enamel that is weakly striated

longitudinally. There is a strong wear facet 4—5 cm long on the back of the tooth.

There are no anterior wear facets indicative of extensive digging.

The lower premolars of PQ2166 are reduced in size and number compared

with the ‘typical’ East African K. limnetes, in which P2 is almost invariably present

although in the ‘advanced’ form it may be shed in the mature adult. Only in one

‘advanced’ specimen from Koobi Fora, KNM-ER.6, is P2 apparently absent and

P3 is slightly reduced in size, but the Ps is normal.

The P3 of PQ2166 is a simple double-rooted tooth that is morphologically a

stouter version of the upper P?*. The anterior root is stouter and longer than the

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 29

Fig. 8. Occlusal view of the right lower cheek teeth of the Kolpochoerus paiceae mandible from

Skurwerug (SAM-—PQ2166). Natural size.

posterior one, and the anteriorly tapering part of the crown which it supports is

correspondingly longer, but narrower, than the posterior section. A single

elongated dentine island has been exposed by wear. In K. limnetes there is a stout

main central cone linked to a well-developed posterior cingulum cone and to a

weak anterior cingulum cusp. In PQ2166 the anterior cingulum cusp is lacking.

The Ps of PQ2166 is a double-rooted tooth that is appreciably larger than P3.

The Ps is well worn in both halves of the mandible, and the occlusal surfaces are

taken up by figure-of-eight exposures of dentine, flanked anteriorly by a

prominent cingulum. The characteristic offset of the double central cusp is

apparent and the pattern is normal for Kolpochoerus limnetes, though the crown

tapers a little anteriorly as compared with the usual rather rectangular crown. The

reduction in size of P3 and Ps in comparison with the sample from Olduvai Bed I is

clear from Figure 9.

Both left and right Mi of PQ2166 are well worn, the only enamel remaining

being on the lingual surfaces of these teeth. They evidently had paired cusps

anteriorly and posteriorly, the latter being slightly the broader.

The left and right Mp are also heavily worn, but sufficient enamel remains on

the occlusal surface to show that the cusps consist of anterior and posterior pairs,

separated by the two median cusps, and flanked by anterior and posterior cingula.

The M: and M are at the low end of the size range for the Olduvai Bed I sample

but are otherwise unremarkable (Fig. 9).

In the Mz of PQ2166, the trigonid and talonid are of similar size and, except

for the anterior cingulum of the trigonid, they are nearly replicas of one another.

Except for its slightly larger size and absence of a posterior cingulum, the M3

trigonid is similar to Mo, with paired cusps anteriorly and posteriorly separated by

two median cusps, with two other median cusps anteriorly and posteriorly. This

paired cross-shaped cluster of four cusps is also the pattern of the talonid,

although there is an additional small cusp developed postero-lingually on the

BREADTH (mm)

Bie Ai team)

BREADTH (mm)

30 ANNALS OF THE SOUTH AFRICAN MUSEUM

Absent in Skurwerug

and Elandsfontein

35, M3 A SKURWERUG-

Undescribed

O Described

+ OLDUVA! Bed!

ELANOSFONTEIN|

30 EB) VAAGINEOME=.

“Yr VAAL TYPE

aaa, ee BS

D5 Be } @

oe H @

a i] O

one ios +, ' e o sb eo” O

ae i + + ys @ e oO O O

20 / a “

ew aie ea

Limits of observed range

for Omo Member G

45 50 5) 60 65 70 ifs

LENGTH (mm)

Fig. 9. Length/breadth measurements of lower cheek teeth of Kolpochoerus paiceae from

Skurwerug, Elandsfontein and the Vaal River gravels, K. limnetes from Olduvai Bed I, and an

advanced Kolpochoerus from Peninj.

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA ai

most posterior median cusp. The presence of four pairs of laterals is a moderately

advanced characteristic and the Ms3 lies at or just above the upper limit for the

Olduvai Bed I sample (Fig. 9). The crown height is not abnormal for teeth at this

stage in the lineage.

THE ELANDSFONTEIN MATERIAL

Figs 7D, 10-14

Singer & Keen (1955) described a small collection of suid remains from the

farm Elandsfontein, the locality from which the ‘Saldanha’ hominid skull had

been recovered (Drennan 1953); the site is often referred to in the literature as

‘Hopefield’. The material consisted of jaw fragments and teeth apparently

representing the remains of 12 individuals. Comparisons were made with the then

known material of Mesochoerus olduvaiensis and a new species was created,

M. lategani, with the type consisting of seven fragments comprising an incomplete

set of upper and lower cheek teeth of a single animal. Keen & Singer (1956)

added descriptions of four more third molars referred to M. lategani and also a

mandibular fragment with RM3 (ES28), which they referred to M. paiceae; they

also removed the previously described M3 (ES21) and referred it to M. paiceae. A

left mandibular fragment with an unworn Mi, Ds, and part of D3 was referred to

Mesochoerus but not named specifically. The 1956 collection also included the

upper second and third molars (ES26) of a much larger suid referred to what was

then called Tapinochoerus meadowsi but Harris & White (1979) refer it to

Metridiochoerus compactus (= Stylochoerus compactus of Cooke & Wilkinson

1978). Three additional incomplete third molars (E9171, 11765 and 12040) have

since been recovered and serve to confirm this identification. The age of the bulk

of the Elandsfontein fauna is assessed as approximately coeval with the upper

part of the Olduvai succession (Bed III or IV), but it also includes younger

elements, and some material may even be older (Hendey 1974).

Subsequent to these accounts, additional material referable to Kolpochoerus

has been recovered, mostly pieces of maxilla or mandible with two or three cheek

teeth, and a few isolated molars and canines. The most complete specimens are

two partial maxillae of one individual with RP?-M?’ and LP*—M? (E16550A, B)

and two rather damaged mandibles (E16675 and E20928), which include the

symphysial areas. Measurements on all the teeth are plotted in Figures 5 and 9,

with the described and undescribed material distinguished by open and solid

circles respectively. Although few of the premolars are preserved intact, the roots

or sockets are present on several specimens and demonstrate the same reduction

in numbers and size as are displayed by the Skurwerug skull.

No upper incisors are known in the Elandsfontein collection but there are

three upper canines—a pair E16297 (Fig. 10) and a poorly preserved specimen

E11808. These specimens are appreciably stouter than the upper canines of the

Skurwerug cranium; they are considerably shorter than normal males of

K. limnetes but they probably do belong to male animals as they are too big to be

32 ANNALS OF THE SOUTH AFRICAN MUSEUM

cee ee He ee i a tt a ee

110 120 130 140 150 160 170 180 A 200. 210° 2

Fig. 10. Ventral and dorsal views of left and right upper canines respectively of Kolpochoerus

paiceae from Elandsfontein (SAM-PQ-E16297).

female. Nevertheless, they are very different from the strongly flared

hylochoerine to phacochoerine ones of ‘advanced’ Kolpochoerus (e.g. KNM-—

ER 788 in Harris & White 1979, pl. 9) and, as the Elandsfontein deposit is almost

certainly coeval with very advanced Kolpochoerus in Olduvai Beds III and IV,

this is regarded as a significant feature of K. paiceae.

The P? is known only by the posterior root in the maxilla of E16550A

(Fig. 11). The entire tooth is estimated to have been about 13 mm long and 11 mm

wide, or very slightly larger than in the Skurwerug cranium. The P* is also present

in this specimen and its mate in E16550B. They are also slightly larger than in

PQ2166 but are structurally similar in having a molariform aspect. There is an

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 333)

ee EN ee aad ase IM ee a Ne

70 80 90 100 110 120 130 140 150 160 170 180 790 = 200

Fig. 11. Lingual and occlusal views of the right upper cheek teeth of Kolpochoerus paiceae from

Elandsfontein (SAM—POQ-E16550).

anterior cingulum made up of four small cusps, paired cusps anteriorly and

posteriorly that are separated by a small median cusp, and a reduced posterior

cingulum made up of two small cusps. The process of molarization is not carried

quite as far as in Hylochoerus but the trend is similar. The only other P* is in a

maxilla fragment (ES7) but in this specimen the configuration is more normal for

Kolpochoerus with two buccal cusps, one lingual cusp, and anterior and posterior

cingula that are made up of several smaller cusps (Singer & Keen 1955, pl. 23D).

The enamel of the principal cusps is more crenulated than in the P* of PQ2166.

The upper first molar is preserved in E16550A (Fig. 11) but is broken away in

its mate. Although the fundamental pillar structure is normal, the tooth is worn to

the point at which large areas of dentine are exposed, together with ribbons and

rings of enamel; the M? in this specimen is still incompletely erupted with only the

34 ANNALS OF THE SOUTH AFRICAN MUSEUM

front pillars in wear, so the extent of the wear on M'is a little surprising. M! is also

present in ES7 in which the wear is a little less advanced and the structure is

normal (Singer & Keen 1955, pl. 23D).

There are five examples of M7’, in various stages of wear, and they are

essentially similar to the teeth of ‘typical’ K. limnetes. The enamel of the principal

cusps is more crenulated than in PQ2166 and there is a heavier covering of

cement, both of which Harris & White (1979) regard as progressive characteris-

tics. Another specimen, E2647/8, has the M' and M? worn at an unusual angle,

apparently due to abnormality in eruption.

There are 17 examples of M’, some of them incomplete but 12 are

measurable and represent at least 10 individuals. Length/breadth plots are given

in Figure 5. The teeth are all larger than the M? of the Skurwerug cranium, carry a

heavier coating of cement, and are relatively higher crowned. Most of the

specimens have three pairs of lateral pillars with a terminal complex that may

include an incipient fourth pair of laterals. One of the smaller specimens, ES23

(Keen & Singer 1956, pl. 33A) is unusual in having a total of only four pillars on

the talonid, namely a small median, the third pair of laterals, and a stout terminal

pillar. The smallest tooth, ES24, has the same arrangement of pillars as in the

Skurwerug cranium, although the crown is 6 mm longer. The third molars in the

pair of maxillae, ES16550, are similar in size to ES23 and ES24 yet have four fully

developed pairs of laterals and a pair of terminal pillars of smaller size (Fig. 11).

The unworn third pair of laterals have a height of 36 mm. The largest teeth are a

pair, ES14 and ES15, in which the fourth pair of lateral pillars is quite well

developed, plus a small double terminal pillar (Singer & Keen 1955, pl. 21A).

ES15 reaches a crown height of 42 mm on the unworn second lateral pillars. In

contrast, ES27 is equally large yet has only three pairs of laterals and a terminal

complex of smaller pillars (Keen & Singer 1956, pl. 33D). There is thus a good

deal of variation within the Elandsfontein sample, more so than in samples from

limited time zones in East Africa, perhaps because the former is temporally

heterogeneous.

Although there are a number of fragments of the lower jaw, there are only

two specimens that demonstrate the morphology. E16675 has the entire

symphysis preserved, with some damage to the incisor area, but retaining parts of

both canines (Fig. 12). The right ramus is lacking but the left ramus is preserved

as far back as the contact with the front of M3. Unfortunately the cheek teeth

were lost post mortem but the root sockets allow some inferences to be made. P2

was Clearly absent, the roots of P3 are very small and P3 and Ps together occupied a

total of 25 mm, while the roots of M: and M2 have a combined length of about

49 mm. The symphysis is slightly longer than in the Skurwerug mandible, and the

breadths across the canine flanges and across the post-canine constriction are also

greater in E16675 (Table 3). The back of the symphysis lies a little in front of the

P3. The cross-section of the canines is about the same size as in PQ2166 but they

appear to have been considerably longer. It seems likely that the jaw is either that

of a young male or, more probably, of a female more robust than PQ2166. The

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 33

«34 S27 tan 1A ser *3r ran 4¢ 7 a4

ev BU aes a 1e: BU A. SU Yu é ei0

Fig. 12. Dorsal view of Kolpochoerus paiceae mandible from Elandsfontein

(SAM-PQ-E16675).

second mandible, E20928, was shattered and the relative positions of the two

horizontal rami are slightly distorted in the reconstructed specimen, but it is clear

that it was very similar to E16675 (Fig. 13). The canines are appreciably larger

and it is virtually certain that this was a male animal. In the two specimens the

symphysis is identical in length and a reliable estimate of the breadth across the

canines in E20928 is slightly more than in E16675. The back of the symphysis is in

the same relative position and both have a curved profile to the symphysis with a

projecting shelf for the incisors. The gentle curve of the anterior border of the

symphysis is the same but E20928 retains Lh, Ll, Rl: and RI. Parts of both

canines are preserved. In this specimen, LMs3 is intact, but only the trigonid of

RMs3 remains. The rest of the tooth-bearing parts of the corpora are broken, but

LP. and LM2, and RPs, RMi, and RM: have been restored to the specimen. The

diastema in PQ2166 is shorter than those of the two Elandsfontein specimens, of

which E20928 is the shorter, apparently because of its relatively large canine. The

reconstruction of E20928 shows that the sum of the length of P3 to M2 was about

the same as that in E16675 and in PQ2166. As the horizontal rami and the Mi: and

Mb are larger in the Elandsfontein specimens than in PQ2166, the implication is

that the reduction of the premolars was even greater.

36 ANNALS OF THE SOUTH AFRICAN MUSEUM

3 9 sO oC 70 80 SO 100 (0 Oo 0 0 0 0 70 eo | 190 200 210 220 230 240 250 260

Fig. 13. Lateral and dorsal views of Kolpochoerus paiceae mandible from Elandsfontein

(SAM-PQ-E20928).

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA Sy

The incisors are known only from the mandible E20928. They are essentially

like those of the Skurwerug mandible although the RI: is a little wider.

The lower canines occur in the two symphyses, E16675 and 20928, although

they are broken above the alveolar margin. In E16675 they are very similar in

diameter to those in the Skurwerug mandible but they were clearly longer as the

whole of the worn part of the tip is broken away and the remainder is more than

6 cm in length. In cross-section, they are almost heart-shaped or U-shaped, with a

broad groove on the posterior face; this shape has been noted in female canines

from Bed I, Olduvai. The roots are open but the pulp cavity is small, and the

enamel is thin. In E20928 the cross-section is subtriangular with a faint keel

caused by a shallow groove on the lateral (lingual) face. The lateral face has a

length of 27 mm, the medial face 25 mm and the posterior 20,5 mm, as compared

with 22, 21 and 18 mm respectively in the Skurwerug mandible and 21, 21 and

16 mm in E16675. There is no posterior groove and the roots are open in the

normal fashion. The left canine of E20928 has been reconstructed and it extends

about 10 cm from the alveolar margin, with a root length of at least 8 cm (Fig. 13).

It is therefore appreciably longer than the canines of PQ2166, but much shorter

than the canines of advanced K. limnetes males from East Africa. There are also

two other specimens of the lower canine, E4019 and E7949. E4019 is part of a

symphysis with 9 cm of the root area preserved; it is essentially similar to E20928.

E7949 has lost the root area but the tip is preserved intact and shows a small

anterior wear facet; the length is 14 cm as preserved and the cross-section

resembles E20928.

In the only three specimens in which the relevant area is preserved, there is

no sign of the possible presence of P2 or of any scar to suggest its existence. Small

roots for P3 are seen in the symphysis E16675 and there is a P3 socket in one of the

specimens (ES17) described and figured by Singer & Keen (1955, pl. 22E, F).

Judging from the alveolus, P3 was even more reduced than that of PQ2166. The

posterior root is very compressed, and its alveolus merges with that of Ps. In

addition to an apparent absolute size reduction in the ES17 P3 compared with that

of PQ2166, there is an even greater relative size difference, given the fact that

ES17 represents a larger individual, with a molar row length of 123 mm against

the 100 mm in PQ2166. Much the same applies in the case of E20928 since,

although neither the P3 nor its alveoli are preserved in this specimen, it is clear

from the reconstruction that this tooth was as reduced as that of ES17 (Fig. 13).

The P, in the mandible fragment ES17 is slightly broken but the length is

estimated as 15,5 mm and the breadth is 11,7 mm, fractionally larger than in

PQ2166. There is a strong broad main pillar and a small posterior cusp, as in the

Skurwerug jaw. The enamel is somewhat rugose. The best-preserved specimens

in the Elandsfontein assemblage are an unequally worn pair associated with the

mandible, E20928 (Fig. 13). The more worn right Ps is morphologically very

similar to those of PQ2166 but is slightly smaller. This is a further indication of the

relatively greater reduction of the premolars in the Elandsfontein K. paiceae,

since the mandible of E20928 is considerably more robust than that of PQ2166

38 ANNALS OF THE SOUTH AFRICAN MUSEUM

(Table 3) and its M3 is appreciably larger. The Ps of E20928 is also distinct in

having markedly rugose enamel, which is said by Harris & White (1979) to be an

advanced character in East African K. limnetes.

There are five examples of Mi. These teeth in E20928 and ES17 are in

advanced wear (Fig. 13; Singer & Keen 1955, pl. 22E, F). In the latter the

anterior and posterior pairs of pillars respectively are united into enamel-rimmed

areas of dentine within which are small residual lakes of enamel; it is larger than

the corresponding tooth in PQ2166. The LM: is present and unworn in the

juvenile jaw fragment ES22 described and figured by Keen & Singer (1956,

pl. 34) and in the original ‘type’ specimen ES5/6 (Singer & Keen 1955, pls 20, 21).

M: is also present in two undescribed juveniles E5294 and E11859. The last

named is also very similar in size to the teeth in PQ2166 but the other two are

somewhat larger (Fig. 9). The morphology is normal.

PREP TL PLATT EEA PTVET EPEAT EE Pa

i Be rity i WITT |

| | | | |

0 10 120 130 140 180. 160 170. 180. 790. 205 ae

Fig. 14. Occlusal views of the smallest and largest lower third molars of Kolpochoerus paiceae

from Elandsfontein (SAM—PQ-E3032 and E11680).

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 39

Mz is present in ES17, ES3 and ES6 (Singer & Keen 1955, pls 20, 21, 22) and

also, though damaged, in the two fragments of a mandible E11680 and in three

other specimens, E20928 (Fig. 13), E20867 and E16910. ES3 is close in size to

PQ2166 while the others are somewhat larger (Fig. 9). The morphology is normal

for Kolpochoerus.

The lower M3 is well represented in the Elandsfontein assemblage by

14 complete specimens belonging to 12 individuals as well as by several incom-

plete specimens. Half of these were described by Singer & Keen (1955) or by

Keen & Singer (1956). The crown normally consists of four well-developed pairs

of lateral pillars with a fifth pair weakly developed at or near the back of the

crown. The two largest teeth, E11680A, B (Fig. 14), however, have only four

paired laterals and two small terminal pillars that can hardly be termed a fifth

pair. Most closely comparable with PQ2166 is the specimen ES28 (Keen & Singer

1956, pl. 33E). It was the only one to be selected by Keen & Singer (1956) for

differentiation from ‘Mesochoerus lategani’ and placement in Broom’s (1931)

species ‘M.’ paiceae. ES28 is indeed very similar to the M3 of the type and of the

‘neotype’ from the Vaal River gravels (Shaw & Cooke 1941), although a little

smaller. Both it and the smallest of the Elandsfontein third molars, E3032

(Fig. 14), are a little larger than the corresponding teeth in PQ2166 but are alike

in pillar structure. Since the older fossils from Elandsfontein could represent

more than one time interval, E3032 and ES28 may belong to an earlier temporal

variant of Kolpochoerus at this locality. However, in view of the variation in

detail in the number and arrangement of the minor elements at the back of the

talonid that is normal in Kolpochoerus samples, the Elandsfontein sample can be

regarded as a single species, so that ‘M. lategani’ 1s a junior synonym of

K. paiceae. The tallest unworn pillar (in E12822/11445) is 36 mm and it is the

same in ES6 and ES16 (Singer & Keen 1955, pl. 20B). Thus the crown height of

both upper and lower molars is generally greater than in most specimens from the

upper parts of the Shungura and the Koobi Fora formations.

THE ZULULAND SPECIMEN

lene, ILS)

During 1976, M. J. McCarthy of the University of Natal collected some

fossils, including an incomplete Kolpochoerus tooth, from the Port Durnford

Formation at Gabhagabha on the Zululand coast (McCarthy & Orr 1978).

This Kolpochoerus specimen (SAM-—PQ2174) is a left Ms in an early state of

wear that lacks most of the anterior half of the trigonid (Fig. 15). The remaining

part of the trigonid is similar in all observable respects to corresponding parts of

the Mz of PQ2166. The talonid also resembles that of PQ2166, as well as the

smaller of the specimens from Elandsfontein. It is distinct in having a double

median cusp anteriorly, a feature not observed in any other K. paiceae specimen.

This is probably an individual or population characteristic that has no taxonomic

significance. The talonid of PQ2174 lacks the small postero-lingual cusp present in

40 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 15. Lateral and occlusal view of the Kolpochoerus paiceae lower third molar from the Port

Durnford Formation, Zululand (SAM—PQ2174). Natural size.

PQ2166, a feature that accounts for the marginally greater talonid length in the

latter (c. 29 mm as against 27 mm in PQ2174). The talonids of the smaller

Elandsfontein specimens, and of the K. paiceae ‘neotype’ from the Vaal River

gravels, are distinguished from that of PQ2174 principally by their slightly greater

crown heights, which suggests that the Zululand specimen is in a more primitive

evolutionary state. This specimen may therefore be like the one from Skurwerug

in representing a somewhat earlier form of K. paiceae. Although the bases for

comparison are limited, there is nothing that would preclude the Zululand and

Skurwerug K. paiceae from being broadly contemporaneous.

DISCUSSION

The close similarity in size and morphology between the type M3 of

Kolpochoerus paiceae and the M3 of the ‘neotype’, both of which came from the

Vaal River gravels, leaves littlhe doubt about their mutual identity. In the

‘neotype’ both P3 and Ps are smaller than in any of the East African material and

very close in size to the Skurwerug specimen. The ‘neotype’ has a small pit in

front of the P3 that was interpreted as a root impression for P2, but as the M3 in

this specimen is not yet erupted, it is very possible that the anterior premolar

would be shed in the adult. The M3 in the Elandsfontein material covers a size

range that includes both the type and the ‘neotype’ and this, coupled with the

unusual reduction in the premolars, provides a sound basis for considering that all

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 41

represent the same species. The Skurwerug and Zululand specimens are distinct

in having the Ms slightly smaller than those in the other samples, but they are also

identified with K. paiceae.

The Skurwerug skull is a good deal smaller than the ‘advanced’ female

cranium from Peninj, although there are general resemblances, as may be

expected in females of Kolpochoerus. In the Peninj cranium the occipital

condyles are elevated well above the palatal plane but this is achieved by

elevation of the whole occiput and braincase, as well as by increase of the height

of the occiput itself as compared with ‘typical’ K. limnetes. In the Skurwerug

cranium the condyles are also higher above the palatal plane than in ‘typical’

K. limnetes, but the occiput itself is shorter and the net result is that the braincase

and orbits are not as elevated, although they may be higher than in the ‘typical’

form. Unfortunately, the braincase and occiput are not known in the best female

cranium, KNM-—ER1085 from Koobi Fora, but in that specimen the rim of the

orbit is lower than in the Skurwerug cranium. The reduction in size of the occiput

in the Skurwerug cranium is unexpected and it is also notable that the

paramastoid processes are not very robust. Hylochoerus also has elevated

condyles and a short occiput but a different occipital morphology. Another point

of difference between the Skurwerug cranium and that of the Koobi Fora female

KNM-ER 1085 is that, far from the premolar reduction seen in PQ2166, the

Koobi Fora female not only had normal-sized premolars but actually retained the

P! 18 mm in front of P?. The Peninj female on the other hand had lost P? but P?

and P* were larger than normal. The Peninj female had kept the full complement

of upper incisors as compared with the reduced complement in PQ2166. Thus the

Skurwerug cranium shows a number of important features of difference from

both the ‘typical’ and the ‘advanced’ Kolpochoerus of East Africa.

The two partial mandibles from Elandsfontein, E16675 and E20928,

probably represent female and male individuals respectively, and they make it

possible to reconstruct the mandible of ‘typical’ K. paiceae with some assurance.

By comparison with mandibles of males of ‘typical’ K. limnetes from the Shungura

Formation or the Koobi Fora Formation there are several striking features of

difference. The corpus is unusually robust and expanded lateral to the second and

third molars so that the overall width across the two horizontal rami is much

greater than the width at the canine alveoli. The symphysis is unusually short,

thus adding to the massive appearance of the jaw. The anterior border of the

symphysis is flattened rather than arcuate, emphasizing the relative shortening

and broadening of the jaw. The distance from the back of the canine to the front

of the anterior premolar is about normal for K. limnetes, so the shortening is

largely in the anterior part of the symphysis. The canines are stout but are more

laterally directed and shorter than in ‘typical’ K. limnetes. While these features,

apart from the premolar reduction, might be less striking in specimens from the

lower part of the Shungura Formation (i.e. more than 2 Ma), they are very

remarkable in the context of an age equivalent to the upper part of the Olduvai

Beds (i.e. less than 1 Ma). The evolutionary trend in the ‘advanced’ Kol-

42 ANNALS OF THE SOUTH AFRICAN MUSEUM

pochoerus from such levels is towards elongation of the jaw, especially of the

symphysis, so that the distance from canine to anterior premolar is increased, as

well as by forward arching of the incisor border. There is also a trend towards

increased curvature and elongation of the canines. There is thus good reason to

maintain the separateness of K. paiceae.

The Skurwerug mandible is less robust than the Elandsfontein specimens but

the width across the two horizontal rami is substantially greater than the width

across the canines. The symphysis is short and the incisive border flattened, as in

the Elandsfontein material, and the somewhat phacochoerine profile of the

symphysis is similar. Considered together with the premolar reduction, these

features leave little doubt that the jaw belongs to the same species as the

Elandsfontein material.

The lesser robustness of the Skurwerug mandible is consistent with the

specimen representing a female, but the difference is greater than might be

expected in a contemporary population, an opinion that would be confirmed if

E16675 does indeed belong to a ‘typical’ K. paiceae female. It seems likely that

the Skurwerug specimen represents an earlier, somewhat less specialized stock.

This is borne out by the smaller size of the third molars, for in suids there is

normally very little size difference in the cheek teeth of the two sexes. The upper

third molars are very slightly larger than those in the Koobi Fora female KNM-—

ER 1085 which, however, have only three well-developed pairs of laterals

whereas PQ2166 has an incipient fourth pair. The third molars from Olduvai

Bed I and from the Shungura Member G likewise have three pairs of laterals in

the uppers and four in the lowers, although additional posterior small pillars can

occur. In size the third molars of the Skurwerug specimen are closely comparable

with those from Member G of the Shungura Formation and with the sample from

Bed I at Olduvai, both with ages close to 2 Ma. It would seem likely that

K. paiceae diverged from a stock essentially at the 2—2,5 Ma level of develop-

ment, strengthening rather than lengthening the talon (talonid) of the third

molars while concurrently reducing the premolars and the upper incisors,

molarizing the P*, shortening the symphysis, and increasing the robustness of the

mandibular corpus, raising the level of the occipital condyles and reducing the

height of the occiput itself. It seems reasonable to suggest that the Skurwerug

skull retains more K. limnetes-like morphology than would be apparent in the

later Elandsfontein sample, although the Skurwerug specimen is already clearly

differentiated from its parent stock. It is therefore likely to date back between

1 and 2 Ma, and it seems desirable to allow about 0,5 million years for the

continued specialization represented by the material from Elandsfontein. The

bulk of the fossils from Elandsfontein have an age estimated as about 0,8 Ma

(+ 0,2 Ma) on the basis of correlation with Olduvai Beds III/IV.

Although not necessarily associated with PQ2166, the other fossils from

Skurwerug are not inconsistent with this age estimate. The Gazella sp. is of some

significance in suggesting a middle Pleistocene or older date, since this genus has

not been recorded in younger faunas from the south-western Cape Province.

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 43

PALAEOENVIRONMENTAL IMPLICATIONS

Apart from its significance in terms of the history of the genus in South

Africa, the Skurwerug Kolpochoerus specimen is remarkable in several other

respects. Firstly, Suidae are a rare element in the Pleistocene faunas of the south-

western Cape, and they were not present in this region at all during historic times.

Even the large and diverse fossil assemblage from Elandsfontein includes

relatively few suid specimens, with no more than a dozen Kolpochoerus paiceae

and even fewer Stylochoerus compactus individuals having been recorded on the

basis of very incomplete material. PQ2166 therefore stands out amongst the

approximately 30 fragmentary fossils that make up the mammalian assemblage

from Skurwerug. However, judging from other fossil occurrences in the vicinity

of Saldanha Bay, it is likely that immense numbers of specimens were preserved

in the Skurwerug dune, and it was only the scale and nature of the excavation and

the restricted access to the site that prevented a more representative sample being

collected. Thus, it is possible that the Skurwerug K. paiceae was as common an

element in the fauna of its time as its counterpart from Elandsfontein.

The recovery of PQ2166 was clearly fortuitous, and was almost certainly due

to the relatively large size and completeness of the specimen, which no doubt

made it an obvious curiosity to the persons involved in its discovery and

subsequent donation to the South African Museum. Indeed, it is the condition of

the specimen that is perhaps its most remarkable feature. It is unquestionably the

most complete suid specimen yet found in the Pleistocene deposits of the south-

western Cape, and one of the best-preserved skulls of Kolpochoerus known

anywhere. Many of the parts that are missing were lost after it had been

excavated, and apparently only the left ascending ramus of the mandible was not

preserved at all. The adjacent part of the left mandibular corpus shows signs of

weathering on its lateral surface, suggesting that this area and the missing

ascending ramus remained exposed after the rest of the skull had been rapidly

buried and so preserved intact. Rapid burial was evidently the exception rather

than the rule amongst recorded Pleistocene mammal fossils of the region. A high

proportion of these specimens show signs of post-mortem damage caused by

weathering and the activities of predators, scavengers and bone-gnawing animals

such as the porcupine.

Since there is no record of the relationship between PQ2166 and the deposits

in which it was incorporated, the taphonomy of this specimen cannot be

determined with certainty. However, since the deposits represent a coastal dune,

it is likely that the rapid burial of the specimen was effected by wind-blown sand.

An implication of this is that the Skurwerug dune was still active at that time, and

that PQ2166 therefore gives a direct indication of the age of the dune (i.e. early

Pleistocene). It is, of course, also possible that the specimen became incorporated

in the deposits after the dune had been stabilized through consolidation and a

covering of vegetation. Solution cavities and small caves or burrows excavated by

animals are features of consolidated calcareous dunes, and PQ2166 may have

been deposited in such a feature. This could have resulted from the death of the

44 ANNALS OF THE SOUTH AFRICAN MUSEUM

animal that had sheltered in such a cave or burrow, or as a result of the activities

of another cave occupant. The latter alternative is unlikely, since the specimen

shows no sign of the damage that characterizes the activities of man, hyaenas and

porcupines, the only habitual cave-inhabiting bone-collectors that might have

been involved in this instance. Even if PQ2166 does post-date the consolidation of

the Skurwerug dune, it need not necessarily be much younger. The consolidation

of local late Cenozoic calcareous dunes was often, or invariably, a relatively rapid

process, so an early Pleistocene date for the one at Skurwerug could still apply. In

spite of all these uncertainties, it is clear that the Skurwerug dune cannot post-

date the Kolpochoerus specimen.

An early Pleistocene age for this dune is in accord with a recently published

view on the relationship between late Cenozoic deposits and sea-level changes in

the Saldanha region (Hendey 1981a, 1981b, 1983a). According to this interpreta-

tion of available evidence, the succession has not been affected by local tectonism

and it reflects the global record of sea-level changes. That part of the record which

is relevant here began during the early Pliocene when the sea rose to about 100 m

above its present level, creating a channel linking Saldanha Bay to

St. Helena Bay, with a large island forming its western margin and the mainland

following a series of granite hills to the east (Hendey 1981), fig. 5). This event

represents the eustatic cycle TP1 of Vail et al. (1977), and is documented by the

Varswater Formation, particularly the exposures in ‘E’ Quarry near Langebaan-

weg, where a rich fossil assemblage provides evidence of its age. During the

middle and late Pliocene, sea-level fell and during this regressive phase there was

a stillstand at about 50 m above present sea-level (cycle TP2). During a late

Pliocene high-stand that was at about 20 m above present sea-level (cycle TP3 or

Q1*), a 12 m submarine platform was developed at wave-base away from the

shoreline in Saldanha Bay. This feature reflects the erosion that occurs in the

upper 10-15 m of the marine environment (Flemming 1976). The platform is still

visible inland from the north-eastern shore of Saldanha Bay, and is underlain by

marine deposits that represent the Uyekraal Shelly Sand Member of the

Bredasdorp Formation (Rogers 1983). The fluviatile deposits at Baard’s Quarry

near Langebaanweg are correlated with the 20 m high sea-level. The 12 m

platform has evidently remained above sea-level since the regression that

followed its formation. The Skurwerug dune is situated on the southern margin of

this platform, which it overlies and clearly post-dates (Fig. 16). On the basis of

this interpretation of local sea-level history during the late Tertiary, the Skurwe-

rug dune must be of late Pliocene or younger age. The possible correlation of this

dune with past sea-level changes may, however, be taken still further.

The actual elevation of the interface between the Skurwerug dune and the

12 m platform is not recorded, but judging from 1:10 000 orthophoto maps of the

*Q1 is a Quaternary cycle according to the definition of this period by Vail et al. (1977) and Beard

et al. (1982), but it falls into the Pliocene if the Plio-Pleistocene boundary is taken at about 1,6 Ma

(see Tauxe et al. 1983), the date which is accepted here. For the purposes of this study the early—

middle Pleistocene boundary is taken at 700 ka and the middle-late Pleistocene boundary at

130 ka (see Butzer 1974).

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA

(Avg eyuepres—VBIce ® ALIce ‘APg CUSIAH IS—O8IZE Y CLIZE dew Aoang

auljaioys Ld] —— — | spuim

? . uoseas Alp

Huiplendid

AV4

VHNVGIVS

Hawueeye ue

[PdISOJOSH UROLITW YINOS WO pajdepy) “}X9} dy} UI 0} PaIoJor somnjeay pue santeoo] Surmoys UoIseI eyULpyesS sy) Jo led Jo dewi-yo}0y4§ “oT ‘SIY

>>>>>>>> >>> > >> OP

>>>>>>>>F

>>>>>>>>>>> >>> >> > >> r>O Pr >>> Or >>

>>>>>>>>>>>>>>>>> >>> >> >>> >> >>

>>>>>>>>>>> >> >>> >> > >>> > r>r>Or rr rm

>>>>>>>> >>>

>>>

>>>>>>>>>>> >>>

>> >>>>>>> >> > > >>

AKAAAAAAAA

AKAKAKAAKAA

46 ANNALS OF THE SOUTH AFRICAN MUSEUM

area it must be in the order of 10 to 12 m. Between this dune and the Spreeuwal

dune complex that fringes the present coastline is another platform, this one

having an elevation of 8 to 9 m. It is here interpreted as a wave-cut platform that

was developed in the intertidal zone during the high sea-level that existed when

the Skurwerug dune itself fringed the coastline. Since this dune was situated on

the coast and was probably no more than a few metres above sea-level, and since

the 8 to 9 m platform has a veneer of aeolian sands, the actual elevation of the

high sea-level concerned must have been at or a little below 8 m. There is in fact

abundant evidence on the southern African west coast of a past high sea-level at

about 7 to 8 m above present sea-level (e.g. Krige 1927; Mabbutt 1957; Hallam

1964; Carrington & Kensley 1969; Tankard 1976; Davies 1980). On the basis of

the record of the Kolpochoerus specimen in the Skurwerug dune, this high sea-

level is now dated as early Pleistocene.

Since this dating conflicts with most previously published opinions on the age

of the 7-8 m high sea-level, the situation needs to be reviewed. Perhaps the most

widely held opinion is that this high sea-level dates from the last (or Eemian)

interglacial. For example, Davies (1980: 165) has stated that the Eem I (= oxygen

isotope substage Se) high sea-level at about 125 ka ‘is widely accepted as around

+8 m over much of the world’, and he cites local examples suggesting that this

was also the case in at least some parts of southern Africa. However, in instances

where evidence of age is available, that which indicates an Eemian age is

equivocal, while that which indicates a pre-Eemian age is not. The evidence for

selected areas or localities around the southern African coast is examined

below.

Apart from the Skurwerug evidence, perhaps the most secure indication that

the 7-8 m high sea-level on the southern African coast predates the last

interglacial comes from the Port Durnford Formation on the Zululand coast

(Hobday & Orme 1974). The mammalian fauna from near the base of this

formation is clearly pre-Eemian, and is generally regarded as being of middle

Pleistocene age (McCarthy & Orr 1978). The Port Durnford Formation

Kolpochoerus tooth described above could be broadly contemporaneous with the

Skurwerug K. paiceae, and the formation itself may therefore be earlier than has

hitherto been supposed. There is in fact nothing in the Port Durnford fauna that

precludes this possibility. The dating of this fauna had been based largely on an

elephant, which has been identified as either Loxodonta atlantica zulu (Maglio

1973) or Elephas zulu (Beden 1983). According to Maglio (1973) it is essentially

similar to the middle Pleistocene elephant from Elandsfontein. Elsewhere in

Africa L. atlantica dates back beyond 2 Ma, so if the identification with this

species is correct, then the Zululand elephant could be of pre-middle Pleistocene

age. However, it is more likely that the material represents a species of Elephas,

as indicated by Beden (1983). Since it is clearly less advanced than other middle

Pleistocene representatives of this genus in southern Africa (e.g. E. iolensis from

the Vaal River gravels and Victoria Falls), an early Pleistocene age for this

elephant, and for the Port Durnford Formation, is probable.

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 47

The decision by Hobday & Orme (1974) to assign an Eemian date to the Port

Durnford Formation appears to have been based largely on the artefact

associations of this formation, and the mistaken belief that the Acheulian and

Sangoan industries persisted well into the late Pleistocene. According to Hobday

& Orme (1974), an Acheulian artefact was recovered from this formation, and if

this was indeed the case, then it would be further evidence of a pre-Eemian age

for the formation, since the Acheulian is no younger than about 150-200 ka

anywhere in Africa (Volman 1984). In addition, a so-called ‘Sangoan’ handaxe

was found on the surface of this formation, and although the age of the ‘Natal

Sangoan’ or “Tugela Industry’ (Davies 1980) is not known, it is likely to be pre-

Eemian. While there may be doubts about the nature and implications of the

archeological associations of the Port Durnford Formation, the faunal evidence is

unequivocal in indicating a middle Pleistocene or earlier date for this formation.

Judged on both sea-level evidence and the Kolpochoerus occurrences, it is very

likely that this formation is an east-coast temporal equivalent of the Skurwerug

dune.

At the same latitude as the Port Durnford Formation, but on the opposite

side of the subcontinent near Oranjemund in South West Africa are deposits

associated with a 7-8 m high sea-level that are termed the ‘C Beach’ or ‘Main

Terrace’ (Hallam 1964: 701). The C Beach, like the Port Durnford Formation,

has recently been dated as an Eemian interglacial feature (Anonymous 1982).

This dating is unacceptable since Acheulian artefacts have been found overlying

the C Beach (Davies 1980; Corvinus 1983). Corvinus (1983) concluded that the

archeological evidence indicates an age for this beach of between 400 and 700 ka

although this dating is clearly tentative given the nature of the artefact association

with the beach and the uncertainties surrounding the chronology of the Acheulian

in Africa. Perhaps all that can be safely concluded from this evidence is that the

C Beach is middle Pleistocene or older.

There are several records in the south-western and southern Cape Province

of Middle Stone Age (MSA) artefacts associated with the 7-8 m high sea-level.

Such occurrences have been used to conclude an Eemian interglacial or even

younger age for this high sea-level. For example, this applies in the case of the

Open-air sites at Melkbos (Duinefontein) (Hendey 1968) and Bok Baai (Mabbutt

et al. 1955), and the cave sites of Die Kelders 1 (Tankard & Schweitzer 1974) and

Klasies River Mouth 1 (Butzer 1978). Vertebrate fossils and MSA artefacts were

found at Duinefontein (see Fig. 1) in deposits that post-date a 6-8 m high sea-

level (Hendey 1968). Although originally assigned to the last glacial period, it is

now clear that these materials are much older and pre-date the Eemian inter-

glacial, with the fauna indicating a ‘later Middle Pleistocene’ age (Klein 1976: 16).

Early Stone Age (Acheulian) artefacts are also known from the area (Hendey

1974). The situation at Duinefontein is similar to that at Bok Baai, 12 km further

north, where deposits overlying a 6-8 m beach contain both MSA and Acheulian

artefacts (Mabbutt et al. 1955). In both these instances the 7-8 m high sea-level

can be no younger than middle Pleistocene. The same applies at Cape Hangklip

48 ANNALS OF THE SOUTH AFRICAN MUSEUM

(Fig. 1), where Acheulian artefacts ‘extend to within 8 m of present sea-level’

(Mabbutt 1957). The caves, Die Kelders 1 and Klasies River Mouth 1, were cut

by a high sea-level of about 7—8 m and partly because the earliest recorded human

habitation of these caves was during the MSA, this high sea-level was dated as

Eemian. However, it was also admitted that in both these instances the caves

could have been created before this interglacial, and then scoured by the highest

of the Eemian transgressions (Tankard 1975; Butzer 1978).

The Eemian sea-level peak was reached during isotope substage Se and was

probably at about 4-5 m above the present level. This is the elevation of the

B Beach at Oranjemund, which is associated with MSA artefacts (Corvinus

1983). Boulder ramparts and other features in the Saldanha Bay—Langebaan

Lagoon area that indicate a high sea-level of up to 4,5 m were described by Parker

(1968). Flemming (1980) gave a slightly lower estimate (3—4 m) for this high sea-

level on the basis of wave erosion features and beach deposits on the Langebaan

compound barrier, and suggested a mid-Holocene age for this event. It is here

regarded as Eemian in age, although there is a beach ridge on this barrier 1 km

east of Kreefbaai with a minimum elevation of 1,5 m (Flemming 1980) that

evidently does represent a Holocene high sea-level. There is evidence that

indicates that the age of events involved in the development of the Langebaan

barrier were generally underestimated in earlier studies (e.g. Parker 1968; Birch

1976; Flemming 1977, 1980). The intertidal-flats deposits near Churchhaven

contain vertebrate fossils that are middle Pleistocene or older. These fossils

include undescribed elephant teeth that are in the collections of the South African

Museum. They represent two extinct species. One is unidentified, but it may

belong to a species that is of pre-middle Pleistocene age. The second is Elephas

iolensis, an elephant that in southern Africa is known only from middle

Pleistocene contexts. The deposits from which these specimens were derived

overlie the core of the compound barrier, which Flemming (1977, fig. 22; 1980,

fig. 51) incorrectly dated as ‘late Pleistocene’. Clearly, the barrier can be no

younger than middle Pleistocene, and it may in fact be even older. Consequently,

the circa 4 m high sea-level that is reflected by deposits and features on the barrier

need not be as young as Flemming and others have supposed.

The 7-8 m high sea-level has been dated as early Pleistocene on at least one

previous occasion. This was done by Tankard (1976) on the basis of the record in

the Saldanha region. Unfortunately, he based this dating on an incorrect

correlation of the 7-8 m high sea-level in this region with the “45-50 m

transgression complex’ in Namaqualand, and on the mistaken assumption that the

dating of this complex as ‘Early Pleistocene’ by Carrington & Kensley (1969) was

correct. The former error arose from his unsubstantiated belief that the south-

western Cape coast has been down-warped relative to that in Namaqualand, an

opinion that was discounted by Hendey (1981a). A major objection to Tankard’s

(1976) correlation is that the 7-8 m high sea-level in the south-western Cape is

associated with a cold-water molluscan fauna, whereas the fauna associated with

the 45-50 m complex in Namaqualand includes warm-water elements. Tankard

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 49

explained this inconsistency by postulating lower sea temperatures in the

Saldanha area. In fact, sea temperatures in the relatively shallow and enclosed

Saldanha Bay are higher than those on the open Namaqualand coast, and there is

no reason to suppose that the situation differed during the early Pleistocene.

Although the basis for Tankard’s dating of the 7-8 m high sea-level in the

Saldanha region is unacceptable, it is perhaps significant that he recognized it to

be of pre-Eemian age.

Another succession in the south-western Cape that has been correlated with

the 7-8 m high sea-level is that at Swartklip on the False Bay coast (see Fig. 1).

Barwis & Tankard (1983) have assigned an Eemian age to the Swartklip

succession largely by inference. The only dateable element in this succession is a

vertebrate fossil occurrence, which is probably of early last glacial age (Klein

1975). Since the fossils were accumulated in a small cave that was developed after

consolidation of the deposits, they clearly post-date the sea-level events that led

to the development of the Swartklip succession. The dating by Barwis & Tankard

(1983) depended upon their acceptance of a late Pleistocene date for similar

successions elsewhere on the South African coast, and their belief that these

successions are contemporaneous with that at Swartklip. The other successions

include the Port Durnford Formation, which was shown above to be of pre-late

Pleistocene age. Presumably, its correlation with the Swartklip succession still

applies, in which case the latter is also likely to be a temporal equivalent of the

Skurwerug dune. A pre-Eemian date for the Swartklip succession also follows if it

is accepted that the Spreeuwal and other dune plumes of the Cape west coast had

their origins during the Eemian interglacial (see below). These plumes, which in

part represent the Witsand Member of the Bredasdorp Formation, are younger

than the consolidated deposits of the Swartklip succession and the Skurwerug

dune, both of which are included in the Langebaan Limestone Member of this

formation (Rogers 1983).

Two important points emerge from the preceding discussion. Firstly, since

the late Cenozoic shorelines of the Saldanha region are unaffected by local

tectonism, and since the 7—8 m shoreline is virtually ubiquitous on the southern

African coast, it follows that this shoreline reflects a eustatic sea-level event.

Secondly, this event represents a high sea-level stand that followed the one that

peaked at about 20 m. The latter high sea-level has previously been dated as late

Pliocene, with the opinion that it is more likely to date from the eustatic cycle QI

than cycle TP3 (Hendey 1981b). If this is indeed the case, then the 7-8 m high

sea-level is most likely to date from the subsequent cycle, Q2. According to Beard

et al. (1982, fig. 1), the Q2 high sea-level stand occurred between 0,80 and

1,3 Ma, dates that are not inconsistent with the 1-2 Ma age estimate suggested for

the Skurwerug Kolpochoerus on the basis of its evolutionary development. This

fossil therefore dates from the latter part of the early Pleistocene, according to the |

subdivisions of this epoch accepted here (see footnote p. 44). The sea-level

changes indicated by features and deposits in the Saldanha region are illustrated

in Figure 17, and are summarized in Table 4.

ANNALS OF THE SOUTH AFRICAN MUSEUM

(7861 (12 Ja pleag Joie sajoAo Areurayend ‘//6] ‘Jv 1a [le Joqye sapoAo [aasy-eas Arensay, “Jowseq ueegosuey—7J ‘ounp

SNIOMINAS—G ‘sysodap syHerang Arend s,preeg—4jO ‘syisodap jerse1101 doyshuy—Ly ‘uoneuio, 1o}emsIe A ‘Alrengd .q,-WjAAq “p 30 ‘G1 361

Aapuay{ wor pojdepy) ‘worse: eyuepyes oy} ut syIsodap pue soinyeoy Aq poyeorpur sa8ueyo [oao]-eas jo uonejussoidor oneWUeIseIq “/] ‘34

syoody | JN3901S141d | IN39011d

$3,9A9 jaNa]-eaS

|8Aa|-eas

anoge | 1

09 LV

S118

001

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA Sl

TABLE 4

Pliocene to Holocene high sea-levels in the Saldanha region.

Height of

Feature/deposit Location sea above Age Eustatic

present level cycle

Beach ridge 1 km east of 1,5 m Holocene Q8

Kreefbaai

Beach deposits and Langebaan 3-4 m late (Oh)

wave erosion features barrier Pleistocene

(Eemian, 5e)

125 ka

Coastal dune and Skurwerug 7-8 m early Q2

wave-cut platform Pleistocene

0,8-1,3 Ma

Submarine platform Inland from G, AY ma late Q1

Skurwerug Pliocene

Fluviatile deposits Baard’s Quarry, 1,75—2,05 Ma

Langebaanweg

Coastal dune Anyskop, c. 50 m middle TP2

Langebaanweg Pliocene

Marine deposits Groot 3,8-4,2 Ma

Springfontein

Erosion surfaces Varswater,

Witteklip, etc.

Submarine platform Elandsfontein c. 100 m early TOP Al

(on Varswater Fm) Pliocene

Paralic deposits ‘E’ Quarry, 4,2—5,2 Ma

(Varswater Fm) Langebaanweg

References: Pliocene—Hendey (1981a, 1981b); Late Pliocene/Holocene—this paper.

The Skurwerug Kolpochoerus and dune have other palaeoenvironmental

implications. The dune itself is similar in some respects to the nearby Spreeuwal

dune complex on the present shoreline of Saldanha Bay. The latter has been

described by De la Cruz (1978) and it is similar to the series of ‘dune plumes’ on

the Cape west coast discussed by Tankard & Rogers (1978), Rodrigues (1978) and

Rogers (1982). It is a relatively small example of a dune plume and is situated

adjacent to the beach on the north-eastern shore of Saldanha Bay. The Spreeuwal

plume is not entirely unvegetated as indicated by Hendey (1983a, fig. 3), but

consists of a complex of vegetated parabolic dunes, with ‘hairpin’ extensions

inland that have a south to north orientation determined by the prevailing dry-

season (summer) winds (Fig. 16). There are, however, some patches of

unvegetated barchanoid dunes superimposed on the vegetated dunes. The largest

52 ANNALS OF THE SOUTH AFRICAN MUSEUM

of the ‘hairpin’ dunes is the most easterly, and it has a maximum height of about

40 m and a length of a little over 2 km. The relatively small size of this plume is

evidently due to the restricted sediment supply in Saldanha Bay, which is now

unconnected with the only large river in the area (i.e. the Berg River, which

discharges into St. Helena Bay). The largest of the west-coast plumes are situated

north of river mouths, where fluviatile sediments provide replenishment for the

dunes (Tankard & Rogers 1978).

At least three phases of dune activity are evident in the dune plumes of the

Cape west coast. The youngest dunes are active today and are clearly influenced

by existing sea-level and climatic conditions. The dating of the two earlier phases

of dune activity is by inference only. The partly vegetated dunes have been

correlated with the Flandrian transgression (Rodrigues 1978), and regarded as the

local equivalent of the ‘Flandrian Episode I’ dunes on the Californian coast

(Cooper 1967), an area that is environmentally similar to the Cape west coast.

The oldest dunes, which are completely vegetated, may be the local equivalent of

the ‘pre-Flandrian’ dunes of California, and they may well date back to the

highest of the sea-level stands during the Eemian interglacial (i.e. during oxygen

isotope substage Se).

The Skurwerug dune represents the vestiges of a dune plume that was

apparently developed under similar circumstances during the 7—8 m high sea-level

of the early Pleistocene. Judging from the remnants of this dune, its easterly parts

at least must have been very like the easterly ‘hairpin’ dunes of the Spreeuwal

plume in terms of orientation, height and length. This suggests that the climatic

conditions that prevailed during the period of development of the Skurwerug

dune (i.e. eustatic cycle Q2) must have been similar to those later in the

Quaternary that influenced the formation of the Spreeuwal plume (i.e. cycles Q5

and Q8). Given the well-documented cyclicity of climates during the Quaternary,

this is not unexpected, but the Skurwerug dune is significant in having provided

the first direct evidence of the nature of early Pleistocene climate in the south-

western Cape.

The similarity in the sizes of the Spreeuwal and Skurwerug dunes is also

significant in suggesting that the sediment supply in Saldanha Bay was as restricted

during the early Pleistocene as it was during the late Pleistocene and Holocene. It

follows that the Berg River must already have been following its present course by

the early Pleistocene, and that the last time it could have discharged into Saldanha

Bay was during the 20 m stillstand of the late Pliocene (Hendey 1981a). The

exclusively marine nature of the sediments that underlie the 12 m platform inland

of Saldanha Bay (Rogers 1983) is a further indication that the Berg River had

diverted to a more northerly course by the end of the Pliocene.

It has already been indicated elsewhere that the Skurwerug Kolpochoerus,

and the other mammals recorded from this locality, provide evidence that the

local vegetation in their lifetimes differed from that of the present (Hendey

1983b). Although the habitat preference of Kolpochoerus cannot be precisely

defined, this genus is likely to have inhabited fairly densely vegetated areas, since

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 53

its dentition is more like those of the living forest hog and bushpig than that of the

open-country warthog. The present strandveld and fynbos vegetation communi-

ties of the Saldanha region are clearly unsuitable habitats for Suidae, and it is

likely that at least some forested or wooded patches still existed in this region

during the early Pleistocene. Furthermore, even though they may not have been

contemporaries of the Kolpochoerus, the grazers represented in the Skurwerug

deposits (i.e. Equus, Connochaetes, Gazella) support the view that grasses were

more prominent in the Pleistocene vegetation communities of the south-western

Cape than is the case today (Hendey 19835). The present vegetation communities

of the Cape west coast are clearly not a prerequisite for the formation of the

coastal dune plumes.

To sum up, the discovery of the Skurwerug Kolpochoerus paiceae skull was a

fortunate event for a variety of reasons. Not only has this fine specimen shed

further light on the history of Kolpochoerus in South Africa, it has also resulted in

further support for the correlation of the west-coast late Cenozoic succession with

global sea-level changes. In addition, by indicating the likely age of the

Skurwerug dune, it has led to the conclusion that local climatic conditions during

the early Pleistocene were at least in some respects similar to those later in the

Quaternary. Finally, it adds to the growing body of evidence that Pleistocene

vegetation communities of the south-western Cape coastal plain were generally

very different from those of the present.

ACKNOWLEDGMENTS

We are indebted to the collector (A. W. Abraham) and donor (P. Hutton),

whose efforts led to the preservation of the Skurwerug Kolpochoerus paiceae

skull in the South African Museum, when so much other fossil material from this

locality was irretrievably lost due to the circumstances of the 1979 excavation.

We also thank Prof. R. V. Dingle and Dr J. Rogers (University of Cape

Town), Dr J. M. Harris (Los Angeles County Museum of Natural History) and

Dr T. D. White (University of California, Berkeley) for helpful comments on the

manuscript of this paper; Dr T. P. Volman (Cornell University) for information

on the archaeological associations of past high sea-levels; and Mr V. Branco,

Mr C. Hunter and Miss E. Pretorius (South African Museum) for preparing most

of the text-figures.

The research of one of us (Q. B. H.) is supported financially by the South

African Council for Scientific and Industrial Research, and this assistance is

gratefully acknowledged. The other (H. B. S. C.) is indebted to the Wenner-Gren

Foundation for Anthropological Research for its support in related projects,

including study of the Elandsfontein material.

REFERENCES

Anonymous. 1982. Itinerary for visit by the Winter Field School of the Geological Society of

South Africa from 10 October to 11 October 1982. Oranjemund: CDM (Pty) Ltd Geological

Department, unpublished document.

54 ANNALS OF THE SOUTH AFRICAN MUSEUM

ARAMBOURG, C. 1943. Observations sur les Suides fossiles du Pleistocene d’Afrique. Bulletin

of the Muséum d’Histoire Naturelle, Paris (2) 15: 471-476.

ARAMBOURG, C. 1947. Contribution a l’étude géologique et paléontologique du bassin du lac

Rodolphe et de la basse vallée de ’'OQmo. Deuxieme partie: Paléontologie. Mission

scientifique de !Omo, 1932-1933, 1, Géologie-Anthropologie: 232-562.

Barwis, J. H. & TANKARD, A. J. 1983. Pleistocene shoreline deposition and sea-level history at

Swartklip, South Africa. Journal of Sedimentary Petrology 53 (4): 1281-1294.

BEARD, J. H., SANGREE, J. B. & SmitH, L. A. 1982. Quaternary chronology, paleoclimate,

depositional sequences, and eustatic cycles. Bulletin of the American Association of

Petroleum Geologists 66 (2): 158-169.

BepDEN, M. 1983. Family Elephantidae. Jn: Harris, J. M. ed. Koobi Fora Research Project.

Volume 2. The fossil ungulates: Proboscidea, Perissodactyla, and Suidae: 40-129. Oxford:

Clarendon Press.

Bircu, G. F. 1976. Surficial sediments of Saldanha Bay and Langebaan Lagoon. Transactions

of the Geological Society of South Africa 79: 293-300.

Broom, R. 1931. A new extinct giant pig from the diamond gravels of Windsorton, South

Africa. Records of the Albany Museum 4: 167-168.

Butzer, K. W. 1974. Geological and ecological perspectives on the Middle Pleistocene.

Quaternary Research 4 (2): 136-148.

Butzer, K. W. 1978. Sediment stratigraphy of the Middle Stone Age sequence at Klasies River

Mouth, Tsitsikama coast, South Africa. South African Archaeological Bulletin 33 (128):

141-157.

CaRRINGTON, A. J. & KENSLEY, B. F. 1969. Pleistocene molluscs from the Namaqualand coast.

Annals of the South African Museum 52 (9): 189-223.

Cooke, H. B. S. 1949a. The fossil Suina of South Africa. Transactions of the Royal Society of

South Africa 32 (1): 1-44.

CookE, H. B.S. 1949b. Fossil mammals of the Vaal River deposits. Memoirs of the Geological

Survey of South Africa 35 (3): 1-109.

Cooke, H. B.S. 1974. The fossil mammals of Cornelia, O.F.S., South Africa. Memoirs of the

National Museum, Bloemfontein. 9: 63-84.

Cooke, H. B. S. 1976. Suidae from Plio-Pleistocene strata of the Rudolf Basin. Jn: CopPENns,

Y. et al. eds. Earliest man and environments in the Lake Rudolf Basin: 251-263. Chicago:

University of Chicago Press.

Cooke, H. B. S. 1978. Suid evolution and correlation of African hominid localities: An

alternative taxonomy. Science 201 (4354): 460-463.

Cooke, H. B. S. & Copprens, Y. In press. Suidae of the Omo Group deposits. Jn: COPPENS, Y.

& HoweELt, F. C. eds. International Omo Research Expedition. Volume 2. Paris: CNRS.

Cooke, H. B.S. & MaGtiio, V. 1972. Plio-Pleistocene stratigraphy in East Africa in relation to

proboscidean and suid evolution. Jn: BisHop, W. W. & MILLER, J. A. eds. Calibration of

hominoid evolution: 309-329. Edinburgh: Scottish Academic Press.

Cooke, H. B. S. & Wirkinson, A. F. 1978. Suidae and Tayassuidae. In: MAGLIO, V. J. &

CookeE, H. B. S. eds. Evolution of African mammals: 435-482. Cambridge, Mass.: Harvard

University Press.

Cooper, W. S. 1967. Coastal dunes of California. Memoirs. Geological Society of America

104: 1-131.

Corvinus, G. 1983. The raised beaches of the west coast of South West Africa/Namibia: an

interpretation of their archaeological and palaeontological data. Munich: Verlag C. H. Beck

(AVA-Forschungen 5).

Cruz, M. A. DE LA. 1978. Marine geophysical and geological investigations in Saldanha Bay.

Bulletin of the Joint Geological Survey/University of Cape Town Marine Geoscience Group 9:

1-115.

Davies, O. 1980. Last interglacial shorelines in the south Cape. Palaeontologia Africana 23:

153-171.

DRENNAN, M. R. 1953. A preliminary note on the Saldanha skull. South African Journal of

Science 50 (1): 7-11.

FLEMMING, B. W. 1976. Rocky Bank—evidence for a relict wave-cut platform. Annals of the

South African Museum 71: 33-48.

FLEMMING, B. W. 1977. Langebaan Lagoon: a mixed carbonate-siliciclastic tidal environment

in a semi-arid region. Sedimentary Geology 18: 61-95.

KOLPOCHOERUS PAICEAE FROM SKURWERUG, SOUTH AFRICA 39

FLEMMING, B. W. 1980. Guide to the Cainozoic deposits of the Saldanha Bay/Langebaan

Lagoon area. Council for Scientific and Industrial Research Report SEA 8002. (NRIO,

Stellenbosch. )

Hatiam, C. D. 1964. The geology of the coastal diamond deposits of southern Africa. Jn:

HaucuTon, S. H. ed. The geology of some ore deposits in South Africa 2: 671-728.

Johannesburg: Geological Society of South Africa.

Harris, J. M. 1983. Family Suidae. Jn: Harris, J. M. ed. Koobi Fora Research Project.

Volume 2. The fossil ungulates: Proboscidea, Perissodactyla, and Suidae: 215-302. Oxford:

Clarendon Press.

Harris, J. M. & Wuire, T. D. 1979. Evolution of the Plio-Pleistocene African Suidae.

Transactions of the American Philosophical Society 69 (2): 1-128.

HENDEY, QO. B. 1968. The Melkbos site: an Upper Pleistocene fossil occurrence in the south-

western Cape Province. Annals of the South African Museum 52 (4): 89-119.

HENDEY, Q. B. 1974. The late Cenozoic Carnivora of the south-western Cape Province.

Annals of the South African Museum 63: 1-369.

HENDEY, Q. B. 198la. Palaeoecology of the late Tertiary fossil occurrences in ‘E’ Quarry,

Langebaanweg, South Africa, and a reinterpretation of their geological context. Annals of

the South African Museum 84 (1): 1-104.

HENDEY, Q. B. 1981b. Geological succession at Langebaanweg, Cape Province, and global

events of the late Tertiary. South African Journal of Science 77 (1): 33-38.

HENDeEY, Q. B. 1983a. Cenozoic geology and palaeogeography of the fynbos region. In:

DEaAcon, H. J., HENDEY, Q. B. & Lamsrecuts, J. J. N. eds. Fynbos palaeoecology: a

preliminary synthesis: 35-60. Pretoria: South African National Scientific Programmes

Report No. 75.

HENpDeEY, Q. B. 1983b. Palaeontology and palaeoecology of the fynbos region: an introduction.

In: Deacon, H. J., HENDEY, Q. B. & LAmsrecuts, J. J. N. eds. Fynbos palaeoecology: a

preliminary synthesis: 87-99. Pretoria: South African National Scientific Programmes

Report No. 75.

Hospay, D. K. & ORME, A. R. 1974. The Port Durnford Formation: a major Pleistocene

barrier-lagoon complex along the Zululand coast. Transactions of the Geological Society of

South Africa 77: 141-149.

Hoepen, E. C. N. vAN & Hoepen, H. E. vAN. 1932. Vrystaatse wilde varke. Paleontologiese

Navorsings van die Nasionale Museum, Bloemfontein 2 (4): 39-62.

KEEN, E. N. & SINGER, R. 1956. Further fossil Suidae from Hopefield. Annals of the South

African Museum 42 (4): 350-360.

KLEIN, R. G. 1975. Palaeoanthropological implications of the non-archaeological bone

assemblage from Swartklip 1, south-western Cape Province, South Africa. Quaternary

Research 5 (2): 275-288.

Krein, R. G. 1976. A preliminary report on the ‘Middle Stone Age’ open-air site of

Duinefontein 2 (Melkbosstrand, south-western Cape Province, South Africa). South

African Archaeological Bulletin 31 (121 & 122): 12-20.

Krice, A. V. 1927. An examination of the Tertiary and Quaternary changes of sea-level in

South Africa. Annals of the University of Stellenbosch (A) 5 (1): 1-81.

Leakey, L. S. B. 1965. Olduvai Gorge 1951-1961. Volume 1. Fauna and background.

Cambridge: Cambridge University Press.

Massutt, J. A. 1957. Some Quaternary events in the winter rainfall area of the Cape Province.

In: Crark, J. D. ed. Third Pan-African Congress on Prehistory. Livingstone 1955: 6-13.

London: Chatto & Windus.

Massutt, J. A., RUDNER, I., RUDNER, J. & SINGER, R. 1955. Geomorphology, archaeology

and anthropology from Bok Baai, Darling District, Cape Province. South African

Archaeological Bulletin 10 (39): 85-93.

Macuio, V. J. 1973. Origin and evolution of the Elephantidae. Transactions of the American

Philosophical Society 63 (3): 1-149.

McCartny, M. J. & Orr, W. N. 1978. A rhinoceros tooth from basal strata of the Port

Durnford Formation, Zululand, South Africa: the relocation of Anderson’s mammalian

fossil site. Annals of the Natal Museum 23 (2): 485-496.

PARKER, R. J. 1968. Eustatic shore lines of Saldanha Bay. University of Cape Town:

Unpublished B.Sc. (Hons) thesis. —

56 ANNALS OF THE SOUTH AFRICAN MUSEUM

Ropricues, M. 1978. The Melkbos-Mamre dune complex. University of Cape Town:

Unpublished Geography III project. |

Rocers, J. 1982. Lithostratigraphy of Cenozoic sediments between Cape Town and Eland’s

Bay. Palaeoecology of Africa 15: 121-137.

Rocers, J. 1983. Lithostratigraphy of Cenozoic sediments on the coastal plain between Cape

Town and Saldanha. Technical Report of the Joint Geological Survey/University of the Cape

Town Marine Geoscience Unit 14: 87-103.

SHAW, J. C. M. 1939. Growth changes and variations in warthog third molars and their

palaeontological importance. Transactions of the Royal Society of South Africa 27 (1):

51-94.

SHAW, J. C. M. & Cooke, H. B. S. 1941. New fossil pig remains from the Vaal River gravels.

Transactions of the Royal Society of South Africa. 28 (4): 293-299.

SINGER, R. & KEENE. N. 1955. Fossil Suiformes from Hopefield. Annals of the South African

Museum 42 (12): 169-179.

TANKARD, A. J. 1975. The late Cenozoic history and palaeoenvironments of the coastal margin

of the south-western Cape Province, South Africa. Rhodes University: Unpublished Ph.D.

thesis.

TANKARD, A. J. 1976. Pleistocene history and coastal morphology of the Ysterfontein—Elands

Bay area, Cape Province. Annals of the South African Museum 69 (5): 73-119.

TANKARD, A. J. & RoGers, J. 1978. Late Cenozoic palaeoenvironments on the west coast of

southern Africa. Journal of Biogeography 5: 319-337.

TANKARD, A. J. & SCHWEITZER, F. R. 1974. The geology of Die Kelders Cave and environs: a

palaeoenvironmental study. South African Journal of Science 70 (12): 365-369.

TauxeE, L., OppykE, N. D., PASINI, G. & Ermi, C. 1983. Age of the Plio-Pleistocene boundary

in the Vrica section, southern Italy. Nature, London 304 (5922): 125-129.

VaIL, P. R., MircHum, R. M. & THomson, S. 1977. Seismic stratigraphy and global changes in

sea level. Memoirs of the American Association of Petroleum Geologists 26: 83-97.

VoLMAN, T. P. 1984. Early prehistory of southern Africa. Jn: KLEIN, R. G. ed. Southern

African prehistory and palaeoenvironments: 169-220. Rotterdam: Balkema.

Waite, T. D. & Harris, J. M. 1977. Suid evolution and correlation of African hominid

localities. Science 198 (4312): 13-21.

6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu-

larly Articles 22 and 51).

Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed

by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov.,

etc:

An author’s name when cited must follow the name of the taxon without intervening punctuation

and not be abbreviated; if the year is added, a comma must separate author’s name and year. The

author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is trans-

ferred from its original genus. The name of a subsequent user of a scientific name must be separated

from the scientific name by a colon.

Synonymy arrangement should be according to chronology of names, i.e. ali published scientific

names by which the species previously has been designated are listed in chronological order, with all

references to that name following in chronological order, e.g.:

Family Nuculanidae

Nuculana (Lembulus) bicuspidata (Gould, 1845)

Figs 14-15A

Nucula (Leda) bicuspidata Gould, 1845: 37.

Leda plicifera A. Adams, 1856: 50.

Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b).

Nucula largillierti Philippi, 1861: 87.

Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9.

Note punctuation in the above example:

comma separates author’s name and year

semicolon separates more than one reference by the same author

full stop separates references by different authors

figures of plates are enclosed in parentheses to distinguish them from text-figures

dash, not comma, separates consecutive numbers.

Synonymy arrangement according to chronology of bibliographic references, whereby the year is

placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable.

In describing new species, one specimen must be designated as the holotype; other specimens

mentioned in the original description are to be designated paratypes; additional material not regarded

as paratypes should be listed separately. The complete data (registration number, depository, descrip-

tion of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.:

Holotype

SAM-—A 13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach, Port Eliza-

beth (33°51'S 25°39’E), collected by A. Smith, 15 January 1973.

Note standard form of writing South African Museum registration numbers and date.

7. SPECIAL HOUSE RULES

Capital initial letters

(a) The Figures, Maps and Tables of the paper when referred to in the text

e.g. ‘. .. the Figure depicting C. namacolus .. .’: ‘. . . in C. namacolus (Fig. 10)...’

(b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by

initials or full names

e.g. Du Toit but A.L. du Toit; Von Huene but F. von Huene

e.g. Therocephalia, but therocephalian

Punctuation should be loose, omitting all not strictly necessary

Reference to the author should preferably be expressed in the third person

Roman numerals should be converted to arabic, except when forming part of the title of a book or

article, such as

‘Revision of the Crustacea. Part VIII. The Amphipoda.’

Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial

capital letter, provided the same generic name is used consecutively. The generic name should

not be abbreviated at the beginning of a sentence or paragraph.

Name of new genus or species is not to be included in the title; it should be included in the abstract,

counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts.

Q. B. HENDEY

H. B. S. COOKE

KOLPOCHOERUS PAICEAE

(MAMMALIA, SUIDAE) FROM SKURWERUG,

NEAR SALDANHA, SOUTH AFRICA, AND ITS

PALAEOENVIRONMENTAL IMPLICATIONS

ME 97 PART 3 DECEMBER 1985 ISSN 0303-2515

Q86l n f74

F THE OUTH AFRICAN

0 : MUSEUM

CAPE ‘TOWN

INSTRUCTIONS TO AUTHORS

1. MATERIAL should be original and not published elsewhere, in whole or in part.

2. LAYOUT should be as follows:

(a) Centred masthead to consist of

Title: informative but concise, without abbreviations and not including the names of new genera or species

Author’s(s’) name(s)

Address(es) of author(s) (institution where work was carried out)

Number of illustrations (figures, enumerated maps and tables, in this order)

(b) Abstract of not more than 200 words, intelligible to the reader without reference to the text

(d) Introduction

(e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents

(f) Summary, if paper is lengthy

(g) Acknowledgements

(h) References

(i) Abbreviations, where these are numerous.

3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with

3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for

illustrations should be typed separately, their positions indicated in the text. All pages should be num-

bered consecutively.

Major headings of the paper are centred capitals; first subheadings are shouldered small capitals;

second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further

subdivisions should be avoided, as also enumeration (never roman numerals) of headings and

abbreviations.

Footnotes should be avoided unless they are short and essential.

Only generic and specific names should be underlined to indicate italics; all other marking up

should be left to editor and publisher.

4. ILLUSTRATIONS should be reducible to a size not exceeding 12x 18cm (19 cm including

caption); the reduction or enlargement required should be indicated (and preferably uniform); orig-

inals larger than 35 X 47 cm should not be submitted; photographs should be rectangular in shape and

final size. A metric scale should appear with all illustrations, otherwise magnification or reduction

should be given in the caption; if the latter, then the final reduction or enlargement should be taken

into consideration.

All illustrations, whether line drawings or photographs, should be termed figures (plates are not

printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items

of composite figures should be designated by capital letters; lettering of figures is not set in type and

should be in lower-case letters. If Letraset is used authors are requested to use Helvetica-style letter-

ing, if possible.

The number of the figure should be lightly marked in pencil on the back of each illustration.

5. REFERENCES cited in text and synonymies should all be included in the list at the end of the

paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable):

(a) Author’s name and year of publication given in text, e.g.:

‘Smith (1969) describes. . .’

‘Smith (1969: 36, fig. 16) describes... .’

‘As described (Smith 1969a, 1969b; Jones 1971)’

‘As described (Haughton & Broom 1927) .. .’

‘As described (Haughton et al. 1927) .. .’

Note: no comma separating name and year

Pagination indicated by colon, not p.

names of joint authors connected by ampersand

et al. in text for more than two joint authors, but names of all authors given in list of references.

(b) Full references at the end of the paper, arranged alphabetically by names, chronologically within

each name, with suffixes a, b, etc., to the year for more than one paper by the same author in

that year, e.g. Smith (1969a, 1969b) and not Smith (1969, 1969a).

For books give title in italics, edition, volume number, place of publication, publisher.

For journal article give title of article, title of journal in italics (according to the World list of scientific periodicals. 4th ed.

London: Butterworths, 1963), series in parentheses, volume number, part number in parentheses, pagination (first and

last pages of article).

Examples (note capitalization and punctuation)

BuLLouGu, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan.

FiscHER, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140.

FIscHER, P. H., DuvaL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie

expérimentale et générale 74 (33): 627-634.

Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and

Magazine of Natural History (13) 2 (17): 309-320.

Koun, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of

the Bingham Oceanographic Collection, Yale University 17 (4): 1-51.

THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: ScHULTZE, L. Zoologische und anthro-

pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Stid-Afrika ausgefiihrt in den Jahren

1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270.

(continued inside back cover)

ANNALS OF THE SOUTH AFRICAN MUSEUM

ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

Volume 97 Band

December 1985 Desember

Part 3 Deel

EARLY PLIOCENE IBISES

(AVES, PLATALEIDAE)

FROM SOUTH-WESTERN CAPE PROVINCE,

SOUTH AFRICA

STORRS L. OLSON

Cape Town Kaapstad

The ANNALS OF THE SOUTH AFRICAN MUSEUM

are issued in parts at irregular intervals as material

becomes available

Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000

Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

word uitgegee in dele op ongereelde tye na gelang van die

beskikbaarheid van stof

Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000

OUT OF PRINT/UIT DRUK

il, AGES, 5-8), HUD, 4-5, 8, tact), SS, 5, 7D),

Gl, Gos), TOD), 8, COX, 1), OCS,

NGCD, 5, Ti; Gao.) ID), WES), VO), Wi, SUCES), 226), $3, SEO), SQ)

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 079 6

Printed in South Africa by In Suid-Afrika gedruk deur

The: Rustica Press; Rive, etd Die Rustica-pers, Edms., Bpk.,

Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap

EARLY PLIOCENE IBISES (AVES, PLATALEIDAE)

FROM SOUTH-WESTERN CAPE PROVINCE, SOUTH AFRICA

Storrs L. OLSON

National Museum of Natural History, Smithsonian Institution,

Washington, D.C.*

(With 5 figures and 2 tables)

[MS accepted 15 July 1985]

ABSTRACT

Two species of ibises (Plataleidae) are recognized in the early Pliocene avifauna from

Langebaanweg, south-western Cape Province, South Africa. One of these, Geronticus

apelex sp. nov., is described from a partial associated skeleton and a few other referred

specimens. Geronticus apelex was smaller than extant members of the genus and was more similar

in morphology to the Palaearctic species G. eremita than to the South African species G. calvus.

A single end of a tibiotarsus is referred to the genus Threskiornis and is indistinguishable from the

living species T. aethiopicus. These reports constitute the first Tertiary record for either genus.

The living species of Geronticus characteristically inhabit dry, open areas and are not

paludicolous. The South African species, G. calvus, is dependent upon short vegetation

maintained by fires and grazing by ungulates for optimum foraging habitat. These ecological

requirements would have been met in the early Pliocene at Langebaanweg, when the

environment was becoming increasingly arid and where periodic fires and abundant ungulates are

known to have been present.

CONTENTS

PAGE

INGO GUT Ct @ Meee mE eee We He ee ONL a Le, i)

SWSUSTDAUNES Sb Se. lets Wo oe Grore GRR Ree EN anne re 58

DISCUSSION MEN eee ree HE utente! Seat k Seto sa 68

INCKMOWIEUEEINENIS Anas Bes abut oe ties eee tee ee 68

IRCHEICEMNOSS © a 68 SS id So wake aee aan Oe us iuee A ee a en oe 69

INTRODUCTION

In the extensive collections of fossil birds from the early Pliocene Varswater

Formation at Langebaanweg (Rich 1980; Hendey 1981) are remains of two

species of ibises (Plataleidae). The present study describes these specimens and

completes the analysis of all the fossils recovered thus far from Langebaanweg

that are referable to the traditional order Ciconiiformes, which, however, is an

unnatural, polyphyletic assemblage (Olson 1979, 1984, 1985). Other ciconiiform

families known from Langebaanweg are the Scopidae (hamerkops), with one

species, Scopus xenopus Olson (1984), and Ciconiidae (storks), with a single new

* Research Associate, Percy FitzPatrick Institute of African Ornithology, University of Cape

Town, South Africa.

Ann. S. Afr. Mus. 97 (3), 1985: 57-69, 5 figs, 2 tables.

58 ANNALS OF THE SOUTH AFRICAN MUSEUM

species of Ciconia (P. Haarhoff MS). Herons (Ardeidae), flamingos (Phoenico-

pteridae), and shoe-bills (Balaenicipitidae) are absent in these deposits, despite

the fact that flamingos and herons are common in the Cape region today.

The fossils treated here are housed in the collections of the Department of

Cenozoic Palaeontology at the South African Museum, Cape Town; all fossil

specimen numbers are prefixed by the acronym SAM-PQ, which has been

omitted below. Other museum acronyms are explained in the acknowledgements.

Comparative material examined

Skeletal material of most species of ibises is scarce. The majority of the

more critical species needed for this study were represented largely or entirely

by a few unsexed captive birds. The specimens examined were as follows:

Geronticus eremita IRSB 2057, IRSB 7459, IRSB 41072, MVZ 155199;

G. calvus TM 33348, TM 33434, TM 60076; Pseudibis papillosa IRSB 64593;

Bostrychia hagedash UMMZ 158551, UMMZ 214345, USNM 322594;

B. carunculata’ AMNH 3894; Lophotibis cristata AMNH 3788; Threskiornis

aethiopicus USNM 558412, USNM 558413, USNM 558415, USNM 558416;

T. melanocephalus USNM 347314; T. spinicollis USNM 347785; Nipponia

nippon USNM 16687; Mesimbrinibis cayennensis USNM 345762, USNM 345763;

Theristicus caudatus USNM 227359, USNM 345764; T. branickii USNM 10022;

Harpiprion caerulescens USNM 227358; Cercibis oxycerca DMNH 58261;

Phimosus infuscatus UMMZ 158609, UMMZ 158610, UMMZ 218530; Eudoci-

mus albus USNM 19787, USNM 500882; Plegadis falcinellus USNM 430823,

USNM 502162, USNM 502696, USNM 502698; P. chihi USNM 498686; P. ridg-

wayt USNM 502127; Platalea alba USNM 558417.

SYSTEMATICS

Family Plataleidae

Any consideration of the systematics of ibises is hampered by the lack of a

modern generic revision of the family grounded in internal anatomy. In

examining the skeletons of most of the extant genera of Plataleidae, it became

apparent that osteological differences between genera are not particularly

profound. Nor are the more distinctive characters distributed in a manner that

consistently allows genera to be clustered meaningfully. Preliminary observations

nevertheless suggest that departures from the generic classification of Steinbacher

(1979) will eventually prove necessary. Although only a single pathological

captive specimen of Pseudibis papillosa was available for examination, no

trenchant characters by which Pseudibis could be separated from Geronticus were

detected. No skeleton of the giant ibis Thaumatibis gigantea was available for this

study; however, it would be premature to follow Holyoak (1970) in including this

distinctive species in the genus Pseudibis without anatomical confirmation.

EARLY PLIOCENE IBISES 59

Genus Geronticus Wagler, 1832

The fossil species described below is referable to the genus Geronticus by the

combination of (1) narrow, tapering premaxilla and mandibular symphysis,

lacking an expanded tip; (2) fairly shallow, curved mandibular rami; (3) retro-

articular processes of the mandible not strongly developed; (4) rostrum deep,

particularly the ventral bars; (5) narrow, elongate cranium; (6) short, robust

tarsometatarsi; and (7) proportionately very long wing elements.

This combination of characters is not met with in any other genus, with the

possible exception of Pseudibis. Among African genera of ibises, Threskiornis

and Plegadis have the bill markedly expanded at the tip and have much longer

tarsometatarsi, particularly in Plegadis. Geronticus and Bostrychia are similar in

sharing short tarsometatarsi and a narrow bill tip, but in Bostrychia the

retroarticular processes of the mandible are much better developed than in the

fossil species.

Geronticus apelex sp. nov.

Figs 1-5

Material

Holotype. 20692, partial associated skeleton including the following

elements: 37 mm section of premaxilla including the anterior portion of the

internal (ventral) narial opening; much of the cranium (partially reconstructed)

including frontals, occiput, and auditory regions; portions of the mandible

including 30 mm of the tip, the proximal third of the right and the left dentary,

and the right articular region; half of a pterygoid; atlas; axis; right forelimb

including most of the humerus (partially reconstructed), ulna lacking part of the

area around the brachial depression, radius, radiale, and carpometacarpus

lacking part of the minor metacarpal; right hindlimb including femur, tibiotarsus

lacking distal third, and tarsometatarsus; and various small fragments, mostly

cranial. The specimen is from the Quartzose Sand Member (QSM) of the

Varswater Formation at Langebaanweg, Cape Province, South Africa.

Paratypes. L13052W2, left coracoid (QSM); L28174M, scapular two-thirds

of right coracoid (QSM); L4236D, scapular half of left coracoid (Pelletal

Phosphorite Member (PPM), bed 3aS); L20755K, left scapula lacking posterior

fourth (QSM). Including the holotype, the minimum number of individuals is

two.

Measurements of holotype (in mm)

Cranium: width at point of articulation of nasal bars, 16,8; width through

auditory region, 23,3; estimated length from point of articulation of nasal bar to

posterior margin of occiput, 44. Mandible: length as reconstructed, 150; width at

a point 20 mm caudad from tip, 3,2; antero-posterior diameter of articular

surface, 7,0; medio-lateral diameter of articular surface, 8,0; least depth through

articulation, 6,7. Atlas: depth, 7,4. Axis: width, 11,3. Humerus: length, 111,4;

60 ANNALS OF THE SOUTH AFRICAN MUSEUM

rape

Fig. 1. Right lateral view of skulls and mandibles of Geronticus. A. G. apelex sp. nov.,

holotype, SAM—PQ-L20692. B. G. eremita, MVZ 155199. C. G. calvus, TM 33348. The

occipital crest in G. eremita may be better developed in presumably older individuals but never

approaches the condition in G. calvus. Scale is in mm.

EARLY PLIOCENE IBISES 61

Fig. 2. Dorsal view of mandibles (top row) and crania (bottom row) of Geronticus.

A. G. apelex sp. nov., holotype, SAM—PQ-L20692. B. G. eremita, MVZ 155199.

C. G. calvus, TM 33348. The cranium of G. apelex is partly reconstructed and is lacking

portions along the lateral margins that would make it slightly wider. Scale is in mm.

62 ANNALS OF THE SOUTH AFRICAN MUSEUM

shaft width at narrowest point, 8,4; shaft width at proximal limit of brachial

impression, 12,4; greatest diameter of brachial impression, 9,0; distal width,

approximately 18,5; length of dorsal condyle, 7,6. Ulna: length, 122,5; proximal

width and depth, 13,3 x 9,0; width and depth of shaft at midpoint, 5,5 x 6,1;

distal width and depth, 7,8 9,7. Radius: length, 116,4; greatest proximal

diameter, 6,7; least and greatest diameter of shaft at midpoint, 3,2 x 4,6; greatest

distal diameter, 8,9. Carpometacarpus: length, 60,8; depth through alular

metacarpal, 13,4; proximal width through trochleae, 6,0; width and depth of

major metacarpal at midpoint, 5,0 x 3,9; greatest distal diameter, 9,2. Femur:

length, 57,4; proximal width, 12,5; depth of head, 5,2; width and depth of shaft at

midpoint, 5,5 x 4,9; distal width, 12,7; depth through medial condyle, 8,8; depth

through lateral condyle, 10,3. Tibiotarsus: proximal width through articulation,

10,1; length of fibular crest, 17,1; width and depth of shaft at approximate

midpoint, 5,2 x 4,3. Tarsometatarsus: length, 64,6; proximal width, 11,6; depth

through hypotarsus, 10,9; width and depth of shaft at midpoint, 4,9 x 3,8; distal

width, 11,7; width and depth of middle trochlea, 4,5 <5,9.

TABLE 1

Length measurements (mm) of major elements of Geronticus apelex sp. nov. compared with

extant species of Geronticus and Pseudibis (means in parentheses).

G. apelex G. eremita G. calvus P. papillosa

(n= 1) (n= 4) (n= 3) (n= 1)

Coracoid 40,4 45,4-47,2 (46,1) 45,7-47,0 (46,1) —

Humerus 111,4 126,3-130,0 (128,6) 117,6-124,1 (119,9) —

Ulna 12S) 139,6-145,3 (142,5) 128,0-137,5 (131,8) 138,9

Carpometacarpus 60,8 71,0-72,9 (71,9) 64,4-69,8 (66,8) DS

Femur 57,4 64,8-66,7 (65,8) 65,5-69,9 (67,2) 64,7

Tarsometatarsus 64,6 72,9-74,6 (73,7) Mili W22) 81,5

Measurements of paratypes (in mm)

Coracoid, L13052W2: length from head to medial angle of sternal articula-

tion, 40,4; length and width of glenoid facet, 10,3 x 5,8. Coracoid, L28174M:

length and width of glenoid facet, 10,5 x 6,3. Coracoid, L24236D: length and

width of glenoid facet, 10,8 x 6,2. Scapula, L20755K: greatest diameter of

anterior end, 9,7.

Diagnosis

Smaller than either of the extant species of Geronticus (or Pseudibis)

(Table 1). Cranium rather narrow and elongate, lacking any of the occipital

expansion into a crest as seen in the two living forms (greatly exaggerated in

G. calvus, in which the top of the cranium is monstrously inflated both laterally

EARLY PLIOCENE IBISES 63

Fig. 3. Wing elements of Geronticus apelex sp. nov., holotype, SAM—PQ-L20692 (on left in

each pair) and G. eremita, MVZ 155199. A.Humeri. 8B. Ulnae. CC. Carpometacarpi.

Scale is in mm.

and posteriorly). Mandibular symphysis narrow as in G. eremita, not broader and

more flattened as in G. calvus. Compared to modern species of Geronticus, the

procoracoid foramen of the coracoid is smaller, the brachial depression of the

humerus is smaller, not extending as far proximally, with the brachial depression

of the ulna being correspondingly small; the trochanter of the femur in lateral

view is more proximally extended into a point, appearing narrower and less

truncate. The wing elements are proportionately longer than in G. calvus, being

more similar to those of G. eremita.

64 ANNALS OF THE SOUTH AFRICAN MUSEUM

Pad

MULT ITTY} TTT ETT

Fig. 4. Hindlimb elements of Geronticus apelex sp. nov., holotype, SAM—PQ-L20692 (on left

in each pair) and G. eremita, MVZ 155199. A. Femora. B. Tibiotarsi. C. Tarsometatarsi.

Scale is in mm.

As mentioned above, Pseudibis is very similar to, and possibly congeneric

with, Geronticus. In the one poor skeleton of P. papillosa examined, the distal

foramen of the tarsometatarsus was more proximally situated than in G. apelex or

in any of the modern specimens of Geronticus.

Distribution

Early Pliocene Varswater Formation (Quartzose Sand Member and Pelletal

Phosphorite Member bed 3aN) at Langebaanweg, south-western Cape Province,

South Africa.

EARLY PLIOCENE IBISES 65

Etymology

Greek, a-, without, and pelex, helmet, in reference to the lack of an

expanded bony occipital crest such as found in modern species of Geronticus. The

name is a feminine noun in apposition.

Remarks

The specimens of Geronticus apelex provide the only Tertiary record for the

genus.

The paratypical coracoids and scapula agree with a species the size of the

holotype of G. apelex. The referred coracoids differ from Plegadis in having the

sternal articulation convex rather than concave, from Threskiornis in not having

the area of muscular attachment on the dorso-sternal surface markedly

excavated, and from Platalea in lacking the distinct, wide, flat clavicular

articulation. The coracoids are distinctive in having the procoracoid process

expanded posteriorly and medially, giving the shaft a very robust appearance. In

this respect the coracoids also differ from Plegadis and Threskiornis and more

closely resemble Geronticus and Bostrychia. However, there appears to be

considerable intra- and interspecific variation in the development of the

procoracoid process in Geronticus and Bostrychia.

The wing elements seem disproportionately long in the holotype of

Geronticus apelex and this is confirmed by computing the ratio of the length of the

long bones to femur length (Table 2). Dividing by the length of the most

TABLE 2

Ratios of lengths of some long bones of African ibises divided by femur length. All ratios for

Geronticus apelex are computed from the holotype. Except for one male of G. eremita, all the

specimens of Geronticus and Bostrychia were unsexed. In Threskiornis and Plegadis, measure-

ments are for two males and two females each; in Plegadis there is considerable sexual

dimorphism in the size of the tarsometatarsus, which is proportionately much longer in males.

Carpometa- Tarsometa-

Humerus Ulna carpus tarsus

Geronticus apelex sp. nov. 1,94 DAS) 1,06 iz

(= 1)

Geronticus eremita 1,93-1,98 2,14—2,22 1,07-1,11 1,10-1,14

(o=4)

Geronticus calvus 1,77-1,80 1,93-1,98 0,97-1,01 1,05-1,10

(n= 3)

Bostrychia hagedash 1,78-1,82 2,05—2,10 0,99-1,02 1,11-1,24

G3)

Bostrychia carunculata 179 2,05 0,98 1,04

(n= 1)

Threskiornis aethiopicus 1,66-1,69 1,91-1,95 0,94-0,97 1,33-1,41

(n= 4)

Plegadis falcinellus 1,59-1,68 1,69-1,91 0,95-0,99 1,52-1,78

(n=4)

66 ANNALS OF THE SOUTH AFRICAN MUSEUM

complete paratypical coracoid yielded practically identical results. By either

standard, G. apelex has a relatively longer humerus, ulna, and carpometacarpus

than any of the species compared except G. eremita. The relative shortness of the

tarsometatarsus in Geronticus, as compared to Threskiornis and Plegadis, is also

demonstrated.

The greater similarity in the proportions of the wing of G. apelex to those of

G. eremita, aS opposed to G. calvus, may reflect migratory or nomadic

propensities in the fossil species. G. eremita migrates well out of its breeding area

in the autumn and early winter, whereas G. calvus is more sedentary. In lacking

the great occipital expansion of G. calvus, the fossil species is likewise more

similar to G. eremita, in which the occipital crest is much less developed. Within

Geronticus, the lack of an expanded occipital crest would almost certainly be

primitive, as such a crest occurs nowhere else in the family. The narrower and less

flattened mandible of G. apelex is also more similar to that of G. eremita than

G. calvus.

The ranges of G. eremita and G. calvus are now widely separated, with the

former having occurred historically in Europe, the Middle East, and in northern

Africa, although it is now reduced to two breeding populations, one in Turkey

and the other in Morocco (Smith 1970). Geronticus calvus is restricted to South

Africa, being found in mountainous areas from southern Transvaal to north-

eastern Cape Province, having formerly extended to the south-western Cape

(Siegfried 1966). The two species of Geronticus are now commonly regarded as

forming a ‘superspecies’ (e.g. Snow 1978). The superspecies concept has become

Fig. 5. Left coracoids of Geronticus (A-B) and distal ends of left tibiotarsi of Threskiornis

(C-D). A. G. apelex, sp. nov., paratype, SAM-—PQ-L13052W2. B. G. calvus, TM 33434.

C. T. aff. aethiopicus, SAM-—PQ-L28479G. _D. T. aethiopicus, USNM 558413.

A and B, natural size; C and D, twice natural size.

PAKDYS PIO GCENESIBISES 67

very fashionable in recent years but through constant abuse and misapplication

(e.g. American Ornithologists’ Union 1983) has lost whatever utility it may once

have had. In the case of Geronticus, it is doubtful that the superspecies category is

appropriate. The differences between the two living species, at least in skull

morphology, are considerable and it is possible that their ancestors may have

existed sympatrically, particularly if one of them were smaller in size, as is

G. apelex. The great differences in cranial adornment may perhaps have evolved

as specific isolating mechanisms at a time when the two species were formerly in

contact. Taking this view, G. apelex could be regarded as a smaller ancestral form

of G. eremita, to which it is most similar.

With the available evidence, however, there is no satisfactory basis for

deciding whether G. apelex represents (a) an extinct lineage that was contempor-

ary with the ancestors of G. eremita and G. calvus and that left no descendants,

(b) the primitive direct ancestor of one or the other of the two living species, or

both forms subsequently. The last seems the least likely, however, in view of the

geologically young age of the Langebaanweg deposits and of the fact that modern

species lineages were already clearly established in other groups of birds in the

same deposits.

Genus Threskiornis Gray, 1842

Threskiornis aff. aethiopicus (Latham, 1790)

JF, XC

Material

Distal end of left tibiotarsus L28479G.

Distribution

Early Pliocene Varswater Formation (QSM) at Langebaanweg, south-

western Cape Province, South Africa.

Measurements

The distal width of this specimen is 13,1 mm, which is very near the mean for

this measurement in the living species Threskiornis aethiopicus (12,1-14,4 mm,

mean 13,2 mm, n= 8).

Remarks

This specimen is larger than in modern African species of ibises except

T. aethiopicus, and differs further from Geronticus and Bostrychia in the much

more prominent tubercle above the intercondylar fossa, this being almost lacking

in Geronticus and only slightly better developed in Bostrychia. It is much too

large and robust for Plegadis. Compared to Platalea it is relatively stouter, the

intercondylar fossa is shallower and wider, and the distal end in distal view is not

68 ANNALS OF THE SOUTH AFRICAN MUSEUM

nearly as elongated antero-posteriorly. This specimen cannot be distinguished

from T. aethiopicus. It provides the only Tertiary record for the genus

Threskiornis.

DISCUSSION

Although ibises are usually considered to be closely associated with marshes

and fresh water, the modern relatives of the two species found in the

Langebaanweg deposits are not restricted to such habitats. In fact, Geronticus

apelex probably indicates quite different conditions, because both the modern

species of Geronticus characteristically inhabit open, arid areas or grasslands.

Both are also gregarious and nest in regions with rocky escarpments. Such

outcrops occur in the Piketberg massif, some 50 km from Langebaanweg.

Because G. apelex may have been migratory or nomadic, it need not have bred

close to the site of deposition.

The living species Geronticus eremita is found ‘most often on dry wadi beds,

rocky slopes or semi-desert’ (Cramp 1977: 344). Geronticus calvus frequents open

grasslands and forages extensively in burnt areas, resorting to heavily grazed

grasslands and pastures outside of the burning season (Manry 1981, 1985a,

1985b). Before the onset of human disturbance, Manry (1981) considered that

suitable habitat for G. calvus would have been maintained by lightning-caused

fires and grazing by native ungulates, particularly the black wildebeest,

Connochaetes gnou.

The habitat requirements of Geronticus in general, and of G. calvus in

particular, fit very well with the palaeoecological conditions inferred at the time

of deposition of the Varswater Formation at Langebaanweg, when the environ-

ment was becoming drier and cooler and the vegetation was giving way to open

grassland and fynbos (Hendey 1981). There is considerable evidence not only for

the presence of numerous ungulates but for the existence of lightning-caused fires

as well (Hendey 1981). Thus, the ecological conditions at Langebaanweg in the

early Pliocene would seem to have been ideal for a species of Geronticus.

Threskiornis aethiopicus is usually associated with water, particularly areas of

fresh-water marsh, which may be a prerequisite for breeding. However, in the

Cape region today it is also found along the coast and is commonly seen foraging

in open, arid pasturelands far from any body of water (McLachlan & Liversidge

1978), for which reason little palaeoecological significance attaches to the

presence of a single individual of Threskiornis in the Langebaanweg deposits.

ACKNOWLEDGEMENTS

I am particularly indebted to Q. Brett Hendey and Philippa Haarhoff of the

South African Museum, Cape Town, for making the fossil material available, and

to the Percy FitzPatrick Institute of African Ornithology, University of Cape

Town, for providing the initial impetus for my study of the Langebaanweg

EARLY PLIOCENE IBISES 69

avifauna. Comparative material came in large measure from the collections of the

National Museum of Natural History, Smithsonian Institution (USNM), but with

the most crucial specimens being supplied by other institutions, in which

connection I am grateful to Alan C. Kemp, Transvaal Museum (TM); Robert

W. Storer, University of Michigan Museum of Zoology (UMMZ); Ned K. John-

son, Museum of Vertebrate Zoology, Berkeley, Calif. (MVZ); David Niles,

Delaware Museum of Natural History (DMNH); and George Barrowclough,

American Museum of Natural History (AMNH). D. Scott Wood arranged for me

to examine specimens on loan to him from the Institut Royal des Sciences

Naturelles de Belgique (IRSB). The photography is by Victor E. Krantz.

Frederick V. Grady cleaned and reconstructed part of the holotype of G. apelex.

I am grateful to Graham Avery, Richard K. Brooke, Timothy Crowe, Philippa

Haarhoff, Q. Brett Hendey, James Hill, David W. Steadman, and D. Scott

Wood for commenting on the manuscript.

REFERENCES

AMERICAN ORNITHOLOGISTS’ UNION. 1983. Check-list of North American birds. 6th ed.

[ Washington, D.C.:] American Ornithologists’ Union.

Cramp, S., ed. 1977. Handbook of the birds of Europe, the Middle East and North Africa.

Vol. 1. Oxford: Oxford University Press.

HENDEY, QO. B. 1981. Palaeoecology of the late Tertiary fossil occurrences in ‘E’ Quarry,

Langebaanweg, South Africa, and a reinterpretation of their geological context. Annals of

the South African Museum 84 (1): 1-104.

Horyoak, D. 1970. Comments on the classification of the Old World ibises. Bulletin of the

British Ornithologists’ Club 90 (3): 67-73.

LatHaM, J. 1790. Index Ornithologicus. Vol. 2. London: Leigh & Sotheby.

Manry, D. E. 1981. Habitat use by foraging bald ibises Geronticus calvus in western Natal.

South African Journal of Wildlife Research 12 (3): 85-93.

Manry, D. E. 1985a. Birds of fire. Natural History 94 (1): 38-45.

Manry, D. E. 1985b. Reproductive performance of the bald ibis Geronticus calvus in relation

to rainfall and grass-burning. Ibis 127 (2): 159-173.

McLacu1ian, G. R. & LiveRSIDGE, R. 1978. Roberts Birds of South Africa. 4th ed. Cape

Town: Trustees of the John Voelcker Bird Book Fund.

Otson, S. L. 1979. Multiple origins of the Ciconiiformes. Proceedings 1978 Conference

Colonial Waterbird Group: 165-170.

Orson, S. L. 1984. A hamerkop from the Early Pliocene of South Africa (Aves: Scopidae).

Proceedings of the Biological Society of Washington 97 (4): 736-740.

Otson, S. L. 1985. The fossil record of birds. Jn: FARNER, D., KING, J. & PARKES, K. C. eds.

Avian biology. Vol. 8. New York: Academic Press.

Rico, P. V. 1980. Preliminary report on the fossil avian remains from late Tertiary sediments at

Langebaanweg (Cape Province), South Africa. South African Journal of Science 76 (4):

166-170.

SIEGFRIED, W. R. 1966. The present and past distribution of the bald ibis in the Province of the

Cape of Good Hope. Ostrich 37 (4): 216-218.

SmitH, K. D. 1970. The waldrapp Geronticus eremita (L.). Bulletin of the British Ornithol-

ogists’ Club 90 (1): 18-24.

Snow, D. W., ed. 1978. An atlas of speciation in African non-passerine birds. London: British

Museum (Natural History).

STEINBACHER, J. 1979. Family Threskiornithidae. In: Mayr, E. & CorrreLt, G. W. eds.

Check-list of birds of the world. 2nd ed. 1: 253-268. Cambridge, Mass.: Museum of

Comparative Zoology.

6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu-

larly Articles 22 and 51).

Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed

by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov.,

etc.

An author’s name when cited must follow the name of the taxon without intervening punctuation

and not be abbreviated; if the year is added, a comma must separate author’s name and year. The

author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is trans-

ferred from its original genus. The name of a subsequent user of a scientific name must be separated

from the scientific name by a colon.

Synonymy arrangement should be according to chronology of names, i.e. all published scientific

names by which the species previously has been designated are listed in chronological order, with all

references to that name following in chronological order, e.g.:

Family Nuculanidae

Nuculana (Lembulus) bicuspidata (Gould, 1845)

Figs 14-15A

Nucula (Leda) bicuspidata Gould, 1845: 37.

Leda plicifera A. Adams, 1856: 50.

Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b).

Nucula largillierti Philippi, 1861: 87.

Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9.

Note punctuation in the above example:

comma separates author’s name and year

semicolon separates more than one reference by the same author

full stop separates references by different authors

figures of plates are enclosed in parentheses to distinguish them from text-figures

dash, not comma, separates consecutive numbers.

Synonymy arrangement according to chronology of bibliographic references, whereby the year is

placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable.

In describing new species, one specimen must be designated as the holotype; other specimens

mentioned in the original description are to be designated paratypes; additional material not regarded

as paratypes should be listed separately. The complete data (registration number, depository, descrip-

tion of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.:

Holotype

SAM-A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach, Port Eliza-

beth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973.

Note standard form of writing South African Museum registration numbers and date.

7. SPECIAL HOUSE RULES

Capital initial letters

(a) The Figures, Maps and Tables of the paper when referred to in the text

e.g. .. . the Figure depicting C. namacolus ...: *. . . in C. namacolus (Fig. 10)...’

(b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by

initials or full names

e.g. DuToit but A.L. du Toit; Von Huene but F. von Huene

e.g. Therocephalia, but therocephalian

Punctuation should be loose, omitting all not strictly necessary

Reference to the author should preferably be expressed in the third person

Roman numerals should be converted to arabic, except when forming part of the title of a book or

article, such as

‘Revision of the Crustacea. Part VIII. The Amphipoda.’

Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial

capital letter, provided the same generic name is used consecutively. The generic name should

not be abbreviated at the beginning of a sentence or paragraph.

Name of new genus or species is not to be included in the title; it should be included in the abstract,

counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts.

STORRS L. OLSON

EARLY PLIOCENE IBISES

(AVES, PLATALEIDAE)

FROM SOUTH-WESTERN CAPE PROVINCE,

SOUTH AFRICA

_ QH

S67X 97 PART 4 MAY 1986 ISSN 0303-2515

OUTH AFRICAN

OF THE S MUSEUM

CAPE ‘TOWN

INSTRUCTIONS TO AUTHORS

1. MATERIAL should be original and not published elsewhere, in whole or in part.

2. LAYOUT should be as follows:

(a) Centred masthead to consist of

Title: informative but concise, without abbreviations and not including the names of new genera or species

Author’s(s’) name(s)

Address(es) of author(s) (institution where work was carried out)

Number of illustrations (figures, enumerated maps and tables, in this order)

(b) Abstract of not more than 200 words, intelligible to the reader without reference to the text

(d) Introduction

(e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents

(f) Summary, if paper is lengthy

(g) Acknowledgements

(h) References

(i) Abbreviations, where these are numerous.

3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with

3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for

illustrations should be typed separately, their positions indicated in the text. All pages should be num-

bered consecutively.

Major headings of the paper are centred capitals; first subheadings are shouldered small capitals;

second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further

subdivisions should be avoided, as also enumeration (never roman numerals) of headings and

abbreviations.

Footnotes should be avoided unless they are short and essential.

Only generic and specific names should be underlined to indicate italics; all other marking up

should be left to editor and publisher.

4. ILLUSTRATIONS should be reducible to a size not exceeding 12 x 18cm (19 cm including

caption); the reduction or enlargement required should be indicated (and preferably uniform); orig-

inals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and

final size. A metric scale should appear with all illustrations, otherwise magnification or reduction

should be given in the caption; if the latter, then the final reduction or enlargement should be taken

into consideration.

All illustrations, whether line drawings or photographs, should be termed figures (plates are not

printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items

of composite figures should be designated by capital letters; lettering of figures is not set in type and

should be in lower-case letters. If Letraset is used authors are requested to use Helvetica-style letter-

ing, if possible.

The number of the figure should be lightly marked in pencil on the back of each illustration.

5. REFERENCES cited in text and synonymies should all be included in the list at the end of the

paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable):

(a) Author’s name and year of publication given in text, e.g.:

‘Smith (1969) describes. . .”

‘Smith (1969: 36, fig. 16) describes...”

‘As described (Smith 1969a, 19695; Jones 1971)’

‘As described (Haughton & Broom 1927)...’

‘As described (Haughton et al. 1927) .. .’

Note: no comma separating name and year

pagination indicated by colon, not p.

names of joint authors connected by ampersand

et al. in text for more than two joint authors, but names of all authors given in list of references.

(b) Full references at the end of the paper, arranged alphabetically by names, chronologically within

each name, with suffixes a, b, etc., to the year for more than one paper by the same author in

that year, e.g. Smith (1969a, 1969b) and not Smith (1969, 1969a).

For books give title in italics, edition, volume number, place of publication, publisher.

For journal article give title of article, title of journal in italics (according to the World list of scientific periodicals. 4th ed.

London: Butterworths, 1963), series in parentheses, volume number, part number in parentheses, pagination (first and

last pages of article).

Examples (note capitalization and punctuation)

BuLLouGuH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan.

FiscHER, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140.

FISCHER, P. H., DuvaL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie

expérimentale et générale 74 (33): 627-634.

Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and

Magazine of Natural History (13) 2 (17): 309-320.

Koun, A. J. 19606. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of

the Bingham Oceanographic Collection, Yale University 17 (4): 1-51.

THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: ScHULTzZE, L. Zoologische und anthro-

pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika ausgefiihrt in den Jahren

1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270.

(continued inside back cover)

ANNALS OF THE SOUTH AFRICAN MUSEUM

ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

Volume 97 Band

May 1986 Mei

Part 4 Deel

A TAXONOMIC REVIEW OF THE

LANTERNFISH GENUS TRIPHOTURUS

FRASER-BRUNNER, 1949

(MYCTOPHIDAE, OSTEICHTHYES)

P. ALEXANDER HULLEY

Cape Town Kaapstad

The ANNALS OF THE SOUTH AFRICAN MUSEUM

are issued in parts at irregular intervals as material

becomes available

Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000

Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

word uitgegee in dele op ongereelde tye na gelang van die

beskikbaarheid van stof

Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000

OUT OF PRINT/UIT DRUK

i, NOES, SB), ALD, ES, 8 espa.) SUES, 5, FD).

6(1, t.-p.i.), 771-4), 8, 9(1-2, 7), 10(1-3),

GED, S, Tee), GD), 15@-S), 2), 27, SIGS), 226), 23, 36), 250)

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 081 8

Printed in South Africa by In Suid-Afrika gedruk

deur

ithe RKusticasPtess,, Btven lide Die Rustica-pers, Edms., Bpk.,

Court Road, Wynberg, Cape Courtweg, Wynberg,

Kaap

A TAXONOMIC REVIEW OF THE LANTERNFISH GENUS

TRIPHOTURUS FRASER-BRUNNER, 1949

(MYCTOPHIDAE, OSTEICHTHYES)

P. ALEXANDER HULLEY

Department of Marine Biology, South African Museum, Cape Town

(With 8 figures and 3 tables)

[MS accepted 1 November 1985]

ABSTRACT

The genus Triphoturus includes two known species: Triphoturus nigrescens (Brauer, 1904)

and Triphoturus mexicanus (Gilbert, 1890). For the latter species, a southern population and a

northern population may be distinguished on the basis of gill-raker counts. Phenotypic variation

in serial meristic data may be environmentally induced in the northern population, accounting for

differences between specimens from the California Current region and from the Gulf of

California. The two species are described, together with comments on their distributions,

including the first records of Triphoturus nigrescens in the Atlantic Ocean; here this species

possesses an Agulhas Subpattern of distribution. A lectotype (ZMB 17617) is designated for

Myctophum (Lampanyctus) nigrescens Brauer, 1904.

CONTENTS

PAGE

EN GRO GUC ENO Mees ee eas ere RT a ide eg ot 71

IVIG HOG Serer ee re eee eee eet ue ee elm TZ

SEES i Steer ese eee ry samen Bete rey ee eh Sect Re act 73

DESC USS UO TN ie ee SR ea Rs 76

SyStennaliC aCCOUME yam yee eee A ts eee be oh a 84

NCKMOWIEASEMENUS Oa cde ce de benno sae os 91

INELEREM CSS Hy eee ert eee ee rite tes eran a Buse 92

INTRODUCTION

The subgenus Triphoturus was created by Fraser-Brunner (1949) to include

five species: Lampanyctus microchir Gilbert, 1913; Myctophum (Lampanyctus)

micropterum Brauer, 1906 (with Myctophum oculeum Garman, 1899, being

considered a junior synonym); Myctophum (Lampanyctus) nigrescens Brauer,

1904; Serpa turneri Fowler, 1934; and Myctophum mexicanum Gilbert, 1890. On

the basis of osteological and photophore evidence, Paxton (1972) raised the status

to that of a genus, phylogenetically closely related to Lampanyctus Bonaparte,

1840, but distinguished from it by the possession of five VO photophores and by

the presence of a small dorsal process on the opercular head of the hyomandibu-

lar. Later, in his list of nominal genera and species, Paxton (1979) included two

Ann. S. Afr. Mus. 97 (4), 1986: 71-95, 8 figs, 3 tables.

We. ANNALS OF THE SOUTH AFRICAN MUSEUM

definite species, Myctophum mexicanum Gilbert (with the questionable junior

synonym Myctophum oculeum) and Myctophum (Lampanyctus) nigrescens (with

the questionable junior synonym Lampanyctus microchir), and one possible

species (Myctophum (Lampanyctus) micropterum), within the genus Triphoturus.

However, the taxonomic status of the species of the genus 1s still unresolved

(Hulley 1984). The type-series of Myctophum (Lampanyctus) micropterum

Brauer, 1904, comprises two species (Hulley 1981): Lampanyctus isaacsi Wisner,

1974 (ZMB 17614, 17615—Gulf of Guinea) and Triphoturus micropterus

(Brauer, 1906) (ZMB 17616—east of Seychelles), with the latter specimen

designated as the lectotype for the purposes of stability. Further, no differences

between this lectotype and the descriptions of Triphoturus microchir (Gilbert)

given by Gilbert (1913) and Nafpaktitis & Nafpaktitis (1969) were observed, so

that Triphoturus microchir was synonymized with Triphoturus micropterus

(Hulley 1981). On the other hand, Wisner (1976) synonymized Triphoturus

microchir (Gilbert) with Triphoturus nigrescens (Brauer, 1904), as there appeared

to be no differences warranting the retention of Gilbert’s species.

Brewer (1973) and Wisner (1976) suggested, not only on the basis of

numbers of vertebrae and gill rakers but also on the distinctiveness of the larvae

(Moser & Ahlstrom 1970; Ahlstrom 1971), that Triphoturus mexicanus is a

species-complex consisting of two species: a ‘southern’ species referable to

Triphoturus oculeus (Garman), and a ‘northern’ species referable to Triphoturus

mexicanus (Gilbert). Further, Wisner (1976) stated that the latter species consists

of two probable, but as yet unnamed, subspecies—one occurring in the Gulf of

California from 30°N to 18°N, and the other in oceanic waters of the California

Current between 38°N and 20°N, with a southern extension to 13°N in the region

of 127°W. A similar distinction was drawn by Robison (1972) on the basis of

higher relative abundance values in the Gulf of California. For the 7riphoturus

mexicanus-complex Imsand (1982, figs 22, 23, 25) differentiated: (1) a Gulf of

California population from a California Current population (both termed

Triphoturus mexicanus) on the criterion of occurring in areas south and east of

23°N 111°W; and (2) a southern population (termed Triphoturus oculeus) from

the two northern populations, on the criterion of occurring south of 16°N. She

stated, however, that these populations are allopatric, with very few areas where

individuals of the two populations can be captured in the same net tow (Imsand

1982, fig. 26).

The purpose of the present paper is to clarify the species composition of the

genus Triphoturus by the examination of the types and other material.

METHODS

The material examined is listed separately under each species in the

Systematic Account. The species are listed alphabetically and type localities are

given in parentheses in the synonymies. Methods for taking measurements and

counts follow Nafpaktitis (1973). A total of 13 morphometric measurements,

REVIEW OF THE LANTERNFISH GENUS TRIPHOTURUS 1

each to the nearest 0,1 mm, were made on each specimen with needle-point

callipers and using a binocular dissecting microscope where necessary. These

include: SL—standard length; HL—head length; HD—head depth; BD—body

depth; UJ—upper jaw length; ED—eye diameter; CPL—caudal peduncle

length; CPD—caudal peduncle depth; PreD—predorsal length;

PreAd—preadipose length; PreP—prepectoral length; PreV—preventral

length; PreA—preanal length. A total of six meristic counts were made on each

specimen, including: D—dorsal rays; A—anal rays; AOa—AOa photophores

(left and right); AOp—AOp photophores (left and right); [AOt = AOa + AOp];

GRu— gill rakers on upper limb (left); GRI—gill rakers on lower limb (left);

[GRt =GRu+1+GRI].

Photophore groupings are in accordance with Paxton (1972) and their

abbreviations are given by Hulley (1981). The following additional abbreviations

are used in reference to material examined:

CAS — California Academy of Sciences, San Francisco, USA

LACM — Los Angeles County Museum of Natural History, Los Angeles, USA

MCZ — Museum of Comparative Zoology, Harvard, USA

SAM — South African Museum, Cape Town, RSA

SU — Stanford University Natural History Museum, Stanford, USA

(specimens now housed in CAS)

USNM — National Museum of Natural History, Washington DC, USA

ZMB — Zoologisches Museum, Berlin, GDR

Statistical analyses and scatter plots were performed on an Apple Ile computer

using STATPRO (Wadsworth Electronic Publishing Company) and standard

reference texts (Snedecor & Cochran 1967; Sokal & Rohlf 1969; Zar 1974).

Computer programmes for analysing morphometric data to allow for size-free

comparisons of shape among specimens (Humphries et al. 1981; Johnson & Feltes

1984) were not available to the author. Meristic data were plotted according to

the method of Hubbs & Hubbs (1953).

RESULTS

Where possible, morphometric data (in mm) and meristic data for the type

specimens are given in Table 1. These specimens and the additional material have

been divided into five groups for the purposes of analysis, with the number of

specimens given in parentheses:

Group A. California Current region: meristics—CAS 24312 (30),

LACM 39201-1 (48), SAM—24926 (3); morphometrics—CAS 24312 (30).

Group B. Gulf of California: meristics and morphometrics—SU 68 (2),

SU 46808 (30).

Group C. Middle American Trench: meristics—LACM 31125-27 (27).

Group D. Central and eastern South Pacific: meristics—LACM 33603-18

(3), LACM 33676-7 (4), LACM 33696-11 (4), MCZ 28500 (3), MCZ 35185 (1),

ANNALS OF THE SOUTH AFRICAN MUSEUM

TABLE 1

Triphoturus. Measurements (in mm) and counts of examined type specimens.

Species

M. mexicanum

M. oculeum

L. microchir

M. micropterum

M. nigrescens

Status

lectotype

paralectotypes

syntypes

syntype

holotype

syntypet

syntype*

syntype

syntype*

No. of

speci-

mens

Catalogue

No.

SE Hi ob BD -Us

USNM 76343 42,8

SU 68

MCZ 28500

MCZ 35185

MCZ 35187

USNM 74468

ZMB 17614

ZMB 17615

ZMB 17616

ZMB 17617

ZMB 22379

3/455)

50,4

17720

74,7

21,0

18,2

+—Lampanyctus isaacsi Wisner, 1974 (fide Hulley 1981)

*—-lectotype of M. micropterum—fide Hulley (1981)

* specimen dried out or damaged

CRE

CPD PreD PreAd PreP PreV PreA

Dy AW A@©ayAOpGRu

— 216 —- — —

12 4 + 6 4

MO eye DO MOA S.7/

AM 2554 39,9) ibaa 220,

ig As —- — 4

Isyee 330 DO tO.7/ 4h

4,1 24,4 38,6 16,1 22,8

I 4 6 4

NOs TORI OAT ilies 5.0

4.9 282 456 — —

13 —- — — 3

Oe) yA ye OA ILD

MAW) Sayslk) AL SA 3) 2

— 15 —- — 3

—_— — — 3

1554 355 90) Sa At

AD 2788, 440 iG 7ee2Sen

i 1S 3

IO 74 79 G3 29

336) ) 2158) 34. 2 elsn@

12 116 5 5 4

5.1), 332. 2351 GO ORS

Li 93) 13.47 0)

(14) (17) 4 6 3

Ub TO = ths és) 3.8

Od) 83,9 spd) Aili 2.5

US UY 6 7 5

—- — 6 7 5

SiG 5 6 3

== 9,015" 5,08" TOs mes

is) WG) epi Os% 113)0

14 = 4 6 3

GRI

8,6

REVIEW OF THE LANTERNFISH GENUS TRIPHOTURUS 75

TABLE 2

Triphoturus. Morphometrics: standard length (SL) in millimetres; all other distances

expressed as percentages of SL.

n SL HL HD BD UJ ED CPL CPD PreD PreAd PreP PreV PreA

Group A 30 39,5-62,0 28-35 14-18 15-19 19-23 6-8 20-24 7-10 47-53 74-88 28-36 40-47 56-66

Group B 30 22,7-59,7 30-33 17-19 18-20 20-23 6-9 19-22 8-10 47-51 74-81 30-34 40-46 57-63

Group D 11 28,6—69,0 25-32 14-18 16-19 17-22 6-8 19-23 7-9 46-53 75-81 26-33 39-46 57-63

Group E 20 13,1—33,4 27-31 13-17 12-17 18-23 5-7 19-25 5-7 50-55 76-81 28-33 42-45 55-61

MCZ 35187 (1), MCZ 56962 (5); morphometrics—LACM 33603-18 (2),

PNG VIS S600 Sin) MEZz 23500772). MCZ 351185 (1), "MCZ 351187" (1),

MCZ 56962 (5).

Group E. South African region: meristics (22) and morphometrics (26); for

details see Table 3.

Morphometric data (expressed as the range of the measurement as a

percentage of SL) for these groups are given in Table 2. Scatter plots of relevant

morphometrics are given in Figures 1-3, and meristic data are presented in

Figure 4.

TABLE 3

Triphoturus nigrescens. Collection data on SAM specimens.

SAM Station Depth Size

Cae no. a Date Position (m) Gear No. (asim) Remarks

27309 IK6 23.04.1961 west of Cape Town 200-0 IKMT 1* — meristics only

27310 A 2967 28.03.1964 37°45’S 18°00’E 600-0 IKMT 1 3)

27311 A1896 11.07.1962 34°12’S 28°24’E 1000-0 N200B 1 33,4

27312 A2961 22.03.1964 40°12’S 14°41’E 700-0 IKMT 1 DRA

28091 SM125 10.05.1977 30°32,2’S 30°57,5'E 415-0 RMT 1 28,2

A292, SIMI IW OS IOV SOUS S SAS WNe oe IRON al Sills)

23093 e SMS 170551977 SOx IAS, 31225278 750-0 RMT 3° — meristics only

23094 eS Mali as .05 1977, 30n05:59"S) 3il257-07E 750=0 RME 1 Bor)

28308 SM190 01.06.1978 34°06,3’S 27°08,3’E 658-0 RMT 3 23,3-27,9

D350 4 SME Se 928, 0531978) 33°25.2'S 2754.7 E 683-0 RM 1 30,8

28346 SM 167 27.05.1978 33°10,5’S 28°17,5’E 1091-0 RMT 1 24,95

29040 SM 104D 24.05.1976 28°25,6'S 32°44,5'E 200-0 Bongo 1 AO

29062 SM 70D = 20.05.1976 27°23,9’S 33°02,9’E 200-0 Bongo 1 16,2

29070 SM 143D 15.05.1977 31°14,7'S 30°14,7’E 212-0 Bongo 2 17,9-28,6

29146 SM 80 21.05.1976 27°39,0'S 33°00,0’E 359-0 Bongo 1 1m

2OB 6S SIMEOZS S051 9762 7210:2'S. 33210 30 45-0 Bongo 1 26,1

29791 2013 09.08.1982 33°40,0’S 14°41,8’E 50 RMT-2 2 20,3-29,2

DIOS On 2023 10.08.1982 33°36,08’S 15°45,25’E 75 RMT-2 1 VG)

29979 2028 11.08.1982 33°40’S 16°45’E is) RMT-2 1 26,9

30016 2070 16.08.1982 29°27'S = 14°14’E 25 RMT-2 1 19,8

*—_damaged

76 ANNALS OF THE SOUTH AFRICAN MUSEUM

DISCUSSION

As pointed out by Johnson & Feltes (1984), the interpretation of serial

meristic data is difficult, since variation can be the result of ecophenotypic effects

(Barlow 1961; Fowler 1970; Johnson & Barnett 1975). Myctophid taxonomy,

however, is based on the nature and orientation of various established

photophore groupings. These are, in the main, species-specific, although there

may be some minor variation in individual photophore position, e.g. the position

of SAO, in relation to VO,—VO;3 in Lampanyctus ater. In a few cases

(Protomyctophum normani-complex and Symbolophorus boops-complex) photo-

phore grouping and position may be identical, so that the structure of the

secondary sexual characters (the supra- and infracaudal glands) has been

employed in distinguishing the species, particularly in the subfamily Myctophinae

(Hulley 1981). However, in the tribe Lampanyctini (subfamily Lampanyctinae),

the lengths of the supra- and infracaudal glands, rather than differences in

structural form, are of diagnostic value, e.g. Ceratoscopelus spp., Lampadena

spp., Lampanyctus spp., Taaningichthys spp. Variations in the total number of

gill rakers on the first arch and variation in the counts of serial meristic characters

(fin rays, AO photophores) are of lesser importance, although some exceptions

do occur, e.g. in the genera Notoscopelus and Gymnoscopelus. The number of gill

rakers on the lower limb of the first arch may exhibit clinal variation, e.g.

Hygophum taaningi, while variation in the number on the upper limb of the first

arch may be indicative of population structure (Diaphus dumerilii, Hygophum

hygomii, Lampanyctus alatus) or may be species-specific, when supported by

differences in photophore grouping and/or caudal gland structure (Hulley 1981).

In the Triphoturus nigrescens species-group, comprising specimens from the

South African region (Group E) and the types of Lampanyctus microchir,

Myctophum micropterum and M. nigrescens (Figs 6—7), the VLO is on or before

the vertical through the ventral base; PVO, is positioned slightly in front of, on,

or behind the vertical through PVO,; and only the VO, is elevated and anteriorly

displaced to before VO,. While VO3 may be slightly raised in some specimens, it

is well below the line passing through VO, and SAQ,. Further, PO; is level with

PO, and PO, and Pol, is well in front of the adipose origin. Some variation in

individual photophore position does occur, namely PO,; PVO,; PVO,; PLO;

SAO, (see Description), but this variation is of no taxonomic significance, being

encountered even amongst specimens from the same haul. In all specimens of this

species-group, the length of the infracaudal gland is less than one-half the length

of the caudal peduncle.

In the 7. mexicanus species-group, comprising specimens from Groups A, B,

C, and D, and including the types of Myctophum mexicanum and M. oculeum

(Fig. 5), the VLO is well behind the vertical through the pelvic base; PVO, is well

before the vertical through PVO,; and both VO, and VO; are highly elevated,

with VO; touching the line passing through VO, and SAO,—the VO, is more

anteriorly displaced (on vertical through outer ventral base) in the T. mexicanus

species-group than it is in the T. nigrescens species-group (behind vertical

REVIEW OF THE LANTERNFISH GENUS TRIPHOTURUS Ti

mm 8

O 18 35 53 70

Fig. 1. Triphoturus. Upper: head length (HL) plotted against standard length (SL).

Lower: head depth (HD) plotted against standard length (SL). Symbols: specimens from

Group A, California Current region—cross; Group B, Gulf of California region—circle;

Group D, Central and eastern South Pacific region—triangle; Group E, South African

region—square. All measurements in mm.

78 ANNALS OF THE SOUTH AFRICAN MUSEUM

mms

O 18 35 93 70

Fig. 2. Triphoturus. Upper: body depth (BD) plotted against standard length (SL).

Lower: caudal peduncle length (CPL) plotted against standard length (SL). Symbols as in

Figure 1. All measurements in mm.

REVIEW OF THE LANTERNFISH GENUS TRIPHOTURUS 79

Fig. 3. Triphoturus. Upper: caudal peduncle depth (CPD) plotted against standard length

(SL). Lower: predorsal length (PRED) plotted against standard length (SL). Symbols as in

Figure 1. All measurements in mm.

80 ANNALS OF THE SOUTH AFRICAN MUSEUM

through outer ventral base) (Figs 5, 7). The PO; is raised out of the series and

Pol, is typically on or behind the vertical through the adipose origin. Atypically

(one specimen only), Pol, may be situated slightly in advance of the vertical

through the adipose origin (Fig. 5A). Variation in individual photophore position

also includes the PO,, PVO,, SAOQ,, SAO,, and Prcz, photophores (see

Description). Such intraspecific variation is well documented in the closely

related genus Lampanyctus (Nafpaktitis et al. 1977; Hulley 1981) and has no

specific taxonomic value. Infracaudal gland structure in specimens of the

T. mexicanus species-group is similar and the infracaudal gland always extends

more than 70 per cent of CPL. It is therefore concluded that photophore pattern

and infracaudal gland structure and length would substantiate the identification of

only two species.

Anal and GRI values (and consequently GRt) would support this conclusion

(Fig. 4), as would differences in maximum size. Specimens in the 7. nigrescens

species-group attain a maximum SL of about 40 mm (Nafpaktitis & Nafpaktitis

1969; Wisner 1976; Parin et al. 1977; Hulley 1984, in press); the value of 74 mm

for M. micropterum given by Weber & de Beaufort (1913) is erroneous, since it is

based on the length given by Brauer (1906) for ZMB 17614, now known to be

Lampanyctus isaacsi (Hulley 1981). Specimens in the T. mexicanus species-group

attain a maximum SL of about 70 mm in both the eastern North and eastern

Central—South Pacific (Bolin 1939; Beebe & Van der Pyl 1944; Bussing 1965;

Berry & Perkins 1966; Craddock & Mead 1970; Parin et al. 1973; Wisner 1976;

Childress et al. 1980; Neighbours & Nafpaktitis 1982; Imsand 1982). Holton

(1969) reported a maximum (?total) length of 100 mm for specimens from off

California. The employment of maximum size as a diagnostic should, however, be

treated with extreme caution, since it is known that certain myctophids (e.g.

Diaphus brachycephalus) exhibit dwarfing, i.e. smaller maximum size and smaller

size at sexual maturity, in regions of low primary productivity within their

distributional range (Hulley 1981).

The 7. mexicanus species-group is said to comprise two species and a

possible subspecies, based on vertebral and gill-raker counts. However, the

general tendency is for teleost populations living in warm waters to have less

vertebrae than closely related populations living in cool waters, and that these

differences are due, in the main, to environmental temperatures prevailing during

early development (Fowler 1970; Johnson & Feltes 1984). Modal values and

ranges in vertebral counts for specimens from the Gulf of California (32; 30-33)

are less than those for specimens from the California Current region (34; 32-36)

(Wisner 1976, table 27) and may well be associated with differences in surface

water temperatures in the two areas during the spring—summer spawning peak

(Ahlstrom 1972; Imsand 1974, 1982; Moser et al. 1974). Concomitant with these

differences in vertebral count, are differences in serial meristics. Specimens from

the Gulf (Group B) have a significantly lower number of dorsal (t —5,12 df 111)

and anal (t —5,80 df 111) fin rays, and AOa (t —9,45 df 216) and AOp (t —6,63

df 214) photophores than specimens from the California Current region

REVIEW OF THE LANTERNFISH GENUS TRIPHOTURUS 81

(Group A) (Fig. 4). Group B specimens also vary from Group A specimens in

possessing higher GRI, and hence GRt counts (t 6,95 df 111; t 6,63 df 111,

respectively). The low weighting to the taxonomic significance of GRI variation

has already been noted above. Therefore the recognition of a separate subspecies

for the Gulf of California specimens does not appear to be warranted. This is

supported by the fact that values for D, A, AOa and AOp counts of specimens

from the Middle American Trench (Group C), just to the south of the mouth of

the Gulf, appear to be somewhat intermediate between those of Groups A and B

(Fig. 4). Both GRu and GRI counts in these Group C specimens approximate

those for California Current region (Group A) specimens.

Specimens from the eastern Central and South Pacific (Group D: represent-

ing 7. oculeus) may be distinguished meristically from specimens from Middle

American Trench (Group C) only by their lower number of gill rakers (GRu:

t —14,29 df 46; GRI: t —8,37 df 46; GRt: t —11,56 df 46). Dorsal, anal and AO

counts are not significantly different, and mirror equivalent modal values and

ranges in vertebral count (Wisner 1976, table 27). The identity of T. oculeus is

therefore based solely on gill-raker count and on pigmentation of the larvae

(Moser & Ahlstrom 1970; Ahlstrom 1971). However, the taxonomic value of

these gill-raker differences at the species level is questionable, especially in view

of the lack of diagnostic photophore characters. As pointed out above, variations

in gill-raker count (both GRu and GRI values) may not be species-specific,

especially when they are not corroborated by differences in photophore grouping

and/or supra- and infracaudal gland structure. Gill-raker variation is then only

interpreted in terms of population differences. This is applicable to both

geographically continuous populations (Diaphus dumerilii) and to geographically

separated populations (Hygophum hygomii) (Hulley 1981), as is probably the

case in Group D specimens. This is supported by the fact that specimens from off

Panama and Costa Rica were found to have GR counts both of 3 + 1 + 10 (9-11)

(LACM 33603-18 (3), LACM 33676-7 (4), LACM 33696-11 (4), SU 65711 (2))

and of 4+1+10-11 (SU 46829 (2)). Furthermore, one of the syntypes of

Myctophum oculeum (MCZ 35187) has GR4+1+10 (left side) and

GR 3+1+10 (right side).

On the other hand, larval characters (including pigmentation) apparently

show a considerable degree of variation within a myctophid species: two larval

forms of Hygophum proximum are found in the northern Indian Ocean

(Pertseva-Ostroumova 1974), while only one adult species, exhibiting clinal

variation, has been recognized (Nafpaktitis & Nafpaktitis 1969); three larval

forms for the Hygophum macrochir-complex in the eastern Central Atlantic have

been found (Moser & Ahlstrom 1974), but only two adult species are recognized

(Nafpaktitis et al. 1977; Hulley 1981). Two hypotheses about larval form are

therefore possible: either, that one species may have two or more larval forms; or

that adults of the second species have not been captured or recognized. In view of

the extensive sampling programmes and taxonomic investigations, particularly in

the eastern Central Atlantic and the Arabian Sea, the latter hypothesis seems to

oO ~t

tence

LOO)

WY)

a iG

Sito yD

C= 00} QO (a

ANNALS OF THE SOUTH AFRICAN MUSEUM

Ke Apel

Saale Me cae Ue!

AOt

159

BS)

I @, Ofer

REVIEW OF THE LANTERNFISH GENUS TRIPHOTURUS

‘ydess repnonsed oy} jon1}suos 0} pasn sjuNos Jo Joquinu =u ‘gq dnoiIn—q ‘gq dnoin—q

‘— dnoign—p9 ‘g dnoin—g ‘vy dnoiIgn—y ‘(€s6T) sqqnH 7 sqqny Jo poyjow ou) 0} BUIPIOINe pojjojd sonsuspy ‘snanjoyduy “p34

lie

LC ee ee

1Y9

Ge — o>

—

Ig SS

> - 22. 2... .. =. 2

U QC 61 8| Ll QI S| 7\ El Cl Ll

<f (a0) Coy (any) ali)

ea Ore

84 ANNALS OF THE SOUTH AFRICAN MUSEUM

be unlikely for Hygophum macrochir and H. proximum. Larval diagnostics for

Triphoturus in the eastern Central Pacific would appear to be unresolved as yet.

Ahlstrom (1971: 31) stated that ‘at least two species (my italics) of Triphoturus

were taken in the EASTROPAC area’ (13°N-20°S, 85°W-126°W), a region

potentially occupied at least by T. nigrescens, T. mexicanus and T. oculeus, if the

latter species is valid (fide Parin et al. 1973, fig. 18; Wisner 1976, fig. 156; and

above data for SU 46829). On the evidence presented by meristics and larval

characters, Group D specimens are therefore not considered specifically distinct,

and at best can only be recognized as a ‘southern’ population of 7. mexicanus.

Morphometric values (Table 2) would not substantiate the differentiation of

more than two species. There is little difference in the scatterplots for the various

groups (Figs 1-3), although in general, specimens of Group E (corresponding to

T. nigrescens) appear to be more slender, with the origin of the dorsal fin located

somewhat more posteriorly than in Groups A, B, C, and D (corresponding to

T. mexicanus) (Figs 2-3). Analysis of the data reveals that there is a significant

(F = 6,86 df 1, 87) difference in the case of the slopes of the regressions of

CPD vs SL when Group E data are compared to combined data for Groups A,

BaGyandeD:

In summary then, the genus 7riphoturus is considered to comprise only two

species, 7. nigrescens and T. mexicanus, the latter with a ‘northern’ and a

‘southern’ population. Specimens from the Gulf of California represent

ecophenotypic variants of the ‘northern’ population.

SYSTEMATIC ACCOUNT

Genus Triphoturus Fraser-Brunner, 1949

Type-species. Myctophum (Lampanyctus) micropterum Brauer, 1906

(= Myctophum (Lampanyctus) nigrescens Brauer, 1904) (by original designa-

tion).

Mouth large, jaws extending well behind vertical through posterior margin of

orbit, with maxillary abruptly enlarged posteriorly. Base of anal fin longer than

base of dorsal fin. Pectoral fins small. Dn absent, Vn present. Five PO, with PO,

elevated and anteriorly displaced to above PO3. VLO at or a little above lateral

line. Five VO, with VO, elevated and anteriorly displaced to in front of the

vertical through VO,, and with VO; either level, raised or elevated. SAO series

markedly angulate. AO series divided into AOa and AOp, both level. Two Pol.

Three Prc in an oblique line. Series of overlapping, luminous scales supracaudally

and infracaudally in both sexes, immediately in front of procurrent caudal rays.

Two species.

KEY TO SPECIES

la. VO; highly elevated, touching line through VO, and SAO,; PO; raised out

of series; VLO well behind vertical through outer ventral base;

GR. 3=5 1: 9=14 totais S20 ae ee ere T. mexicanus

REVIEW OF THE LANTERNFISH GENUS TRIPHOTURUS 85

1b. VO; level or raised, but always well below line joining VO, and SAO,; PO;

level with PO, and PO;; VLO on or before vertical through outer ventral

baseHGR eel 7-8. total 12 aia te ee oes ce ks T. nigrescens

Triphoturus mexicanus (Gilbert, 1890)

Fig. 5

Myctophum mexicanum Gilbert, 1890: 51 (lectotype USNM 76343—Gulf of California; para-

lectotypes SU 68 (3)—25°59’45"N 111°03'30"W; Bohlke 1953).

Scopelus mexicanus Litken, 1892: 266.

Nannobrachium mexicanum Goode & Bean, 1895: 512.

Myctophum oculeum Garman, 1899: 260, pl. LVI, fig. 2 (19 syntypes: MCZ 28500, 34945,

34946, 35162, 35182, 35185, 35187, 35189, USNM 120422—eastern Pacific; Paxton 1979).

Myctophum (Lampanyctus) mexicanum Brauer, 1904: 396.

Myctophum (Lampanyctus) oculeum Brauer, 1906: 167.

Lampanyctus mexicanus: Parr, 1928: 84 (key); 1931: 25 (key), 30, fig. 12. Bolin, 1939: 135,

fig. 21. Beebe & van der Pyl, 1944: 84, fig. 17.

Lampanyctus oculeus: Parr, 1928: 85 (key).

Lampanyctus (Triphoturus) mexicanus: Fraser-Brunner, 1949: 1084, fig. (key). Berry & Perkins,

1966: 660, fig. 22C.

Triphoturus mexicanus: Bussing, 1965: 203. Bekker, 1967: 179. Paxton, 1967: 424, figs 11, 14, 16.

Ahlstrom, 1969: 41; 1971: 31; 1972: x, 64-124. Craddock & Mead, 1970: 3.31. Ebeling

et al., 1970: 4, figs 1-4. Robison, 1972: 448, figs 3-4. Paxton, 1972: 6. Brewer, 1973: 23,

fig. 1OB. McNulty & Nafpaktitis, 1976: 579, pls 1-8. Childress et al., 1980: 28, figs 1, 4.

Pieper & Bargo, 1980: 935, figs 1-4. Neighbors & Nafpaktitis, 1982: 208, figs 1, 2. Barnett,

1983: 248. Loeb et al., 1983a: 134; 1983b: 155, figs 1, 2, 7.

Triphoturus oculeus: Ahlstrom, 1971: 31.

Material

Types. Myctophum mexicanum: lectotype, USNM 76343; paralectotypes,

SU 68 (2). Myctophum oculeum: syntypes, MCZ 28500 (3), 35185 (1), 35187 (1).

Other. CAS 24312 (30), Santa Catalina Islands; CAS 47824 (5), off Baja

California; LACM 31125-27 (27), Middle American Trench (22°21'N

108°12’W—22°25'N_ 108°29'W); LACM 33603-18 (3), 08°57'30"N 88°05'00"W;

LACM 33696-11 (4), 05°28’00"N 82°10'00"W; LACM 33676-7 (4), 05°29'30"N

82°33'00"W; LACM 39201-1 (48), San Pedro Basin (010°T from Long Point

Light, 6,0 miles); SAM-—24926 (3), 33°40’N—33°45'’N, 118°27’W-118°18'W;

SAM-30897 (5), 33°19’S 73°39'W; SU 46808 (30), Gulf of Mexico; SU 46829 (2),

Panama; SU 65711 (2), Panama.

Description

Meristics and measurements are given in Tables 1—2 and Figure 4.

Origin of dorsal fin well behind vertical through ventral base, usually nearer

to tip of snout than to end of lateral line; origin of anal fin under middle of dorsal

base or slightly more posterior; origin of adipose fin about on vertical through last

anal ray. Pectoral fins small, reaching to about POx,; ventral fins extending slightly

posterior to VO,.

Dn absent; Vn small, at anteroventral margin of orbit. Op, at about level of

upper, expanded end of maxillary. Five PO, with PO,—PO, interspace greatest,

86 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 5. Triphoturus mexicanus. Upper: southern population (MCZ 56962, 33°19’S 73°39'W,

SL 68,7 mm). Middle: northern population (SU 46808, Gulf of California, SL 49,5 mm). Lower:

northern population (LACM 39201-—1, San Pedro Basin, California, SL 47,6 mm).

REVIEW OF THE LANTERNFISH GENUS TRIPHOTURUS 87

with PO; raised, and with PO, elevated and anteriorly displaced to directly on or

slightly behind vertical through PO; and at level of upper pectoral base or higher.

PVO, above PO,—PO, interspace, closer to PO, than to PO, and below level of

Op,; PVO, well behind vertical through PVO, and below level of upper end of

pectoral base. PLO well in advance of vertical through PVO,, about one

photophore diameter below lateral line. VLO behind vertical through ventral

base, on vertical through VO,, and touching dorsal edge of lateral line. Five VO,

with VO, highly elevated and anteriorly displaced to vertical through outer

ventral base; with VO; highly elevated, touching line joining VO, and SAQ,.

SAO series markedly angulate; with SAO, midway between VO, and VO, or

closer to VO; and slightly above level of VO,; with SAO, above or slightly

behind anal origin and at or above level of SAO,; and with SAO, behind vertical

through anal origin, touching dorsal edge of lateral line. AOa level, with

AOa'—AOa?’ interspace greatest; AOp level, all behind base of last anal ray. Two

Pol, with Pol, behind last AOa, and with Pol, on or behind vertical through

adipose origin, touching dorsal edge of lateral line. Three Prc, in straight

ascending line, with Prcz nearer to Prc, and on or touching line through centres of

Pre, and Prez, and with Prez above level of lateral line.

Supracaudal gland consisting of 3—4 overlapping, luminous scales; infra-

caudal gland elongate (greater than 70% CPL), with 4—7 overlapping, luminous

scales.

Maximum length about 70 mm.

Distribution

Vertical distribution. California Current region: day 200-700 m (maximum

400-500 m), night 50-200 m (maximum 100-200 m). Gulf of California: day

200-700 m (maximum 400-500 m), night 25-550 m (maximum 200-300 m)

(Imsand 1982).

The geographic distribution of the species is given in Figure 8.

Triphoturus mexicanus apparently possesses a disjunct distribution in the

eastern Pacific Ocean. It is known from throughout the Gulf of California, where

it is the dominant midwater fish species between 25°N and 29°N. In the California

Current region the species occurs between 38°N and about 20°N, but offshore

may extend southwards to 13°N at 127°W. Off the coasts of Central and South

America it has been taken from 13°N to about 35°S. Towards its southern limit

the species has only been recorded east of 76°W, but off the coast of Peru it may

extend westwards to about 95°W.

Triphoturus nigrescens (Brauer, 1904)

Figs 6, 7

Myctophum (Lampanyctus) nigrescens Brauer, 1904: 403 (syntypes: ZMB 17617—03°24'06"S

58°38'01"E; ZMB 22379—02°38'09"S 63°37'09"E); 1906: 241, fig. 158. .

Myctophum (Lampanyctus) micropterum Brauer, 1906: 239, fig. 157 (partim) (lectotype:

ZMB 17616: 02°43’08"S 61°12'06"E; Hulley 1981: 205).

88 ANNALS OF THE SOUTH AFRICAN MUSEUM

Lampanyctus microchir Gilbert, 1913: 101 (holotype: USNM 74468—Suruga Bay, Japan). Parr,

1928: 85 (key).

?Myctophum micropterum: Weber & de Beaufort, 1913: 154, fig. 59.

non Myctophum (Lampanyctus) micropterum: Pappenheim, 1914: 197 (= Lampanyctus isaacsi

Wisner, 1974).

Lampanyctus nigrescens: Parr, 1928: 84 (key).

Lampanyctus micropterus: Parr, 1928: 85 (key).

Lampanyctus (Triphoturus) microchir Fraser-Brunner, 1949: 1083, fig. (key). Berry & Perkins,

1966: 660, fig. 22C.

Lampanyctus (Triphoturus) micropterus Fraser-Brunner, 1949: 1083, fig. (key).

Lampanyctus (Triphoturus) nigrescens Fraser-Brunner, 1949: 1083, fig. (key). Berry & Perkins,

1966: 660, fig. 22C.

Triphoturus microchir: Bekker, 1967: 179. Nafpaktitis & Nafpaktitis, 1969: 55, figs 62, 70.

Legand et al., 1972: 306, fig. 27. Kotthaus, 1972: 29, fig. 284. Clarke, 1973: 406, fig. 12;

1980: 625, figs 1-4. Hartmann & Clarke, 1975: 636. Parin et al., 1973: 114, fig. 19; 1977:

IPD), 109., ZI.

Triphoturus nigrescens: Paxton, 1972: 6. Wisner, 1976: 165, fig. 155. Loeb, 1979a: 178; 1979b:

789, fig. 10; 1980: 192. Barnett, 1983: 284; 1984: 201. Hulley, 1984: 90, fig. 19. Paxton et al.,

in press.

Material

Types. Lampanyctus microchir: holotype, USNM 74468. Myctophum (Lam-

panyctus) micropterum: lectotype, ZMB 17616. Myctophum (Lampanyctus)

nigrescens: syntypes, ZMB 17617, 22379.

Other. See Table 3.

Description

Meristics and measurements are given in Tables 1—2 and Figure 4.

Origin of dorsal fin well behind vertical through ventral base, nearer to end

of lateral line than to tip of snout; origin of anal fin under middle of dorsal base or

slightly more anterior; origin of adipose fin on vertical through base of last anal

ray. Pectoral fins small, reaching to about PO,; ventral fins extending slightly

posterior to VOx.

Dn absent; Vn small, at anteroventral margin of orbit. Op, at about level of

upper, expanded end of maxillary. Five PO, with PO,—PO, interspace greatest

and with PO, elevated and anteriorly displaced to directly on, anterior to, or

behind vertical through PO; and at level of upper pectoral base or higher. PVO,

above PO,—PO, interspace, closer to PO, than to PO, and at about level of Op,;

PVO, on or only slightly behind vertical through PVO, and below level of upper

pectoral base. PLO well in advance of vertical through upper pectoral base, at or

less than one photophore diameter below lateral line. VLO slightly anterior to

vertical through ventral base, touching dorsal edge of lateral line. Five VO, with

VO, highly elevated and anteriorly displaced to before vertical through VO;

VO; level or somewhat raised, never touching line joining VO, and SAO,;. SAO

series markedly angulate, with SAO, above VO3;—VO, interspace or above VO,

and at about level of ventral margin of orbit, with SAO, above anal origin and at

level of SAO,, and with SAO, behind vertical through anal origin, touching

dorsal edge of lateral line. AOa level, with AOa'—AQa? interspace greatest;

AOp level, all behind base of last anal ray. Two Pol, with Pol, behind last AOa

REVIEW OF THE LANTERNFISH GENUS TRIPHOTURUS 89

Fig. 6. Triphoturus nigrescens. Syntype (ZMB 17617, ‘Valdivia’ Station 231, 03°24’06"S 58°38’01"E,

SL 29,3 mm), now designated as lectotype.

Fig. 7. Triphoturus nigrescens (SAM—28094, 30°05,5’S 31°57,0'E, SL 33,6 mm).

and with Pol, in advance of vertical through adipose origin, touching dorsal edge

of lateral line. Three Prc, in straight ascending line, with Prc, nearer Pre, and

touching line through centres of Prec, and Prce3, and Prc3 above level of lateral

line.

Supracaudal gland consisting of four overlapping, luminous scales; infracau-

dal gland short (less than 50% CPL), with five overlapping, luminous scales.

Maximum length 40 mm.

Remarks

Due to the poor state of preservation of one (ZMB 22379) of the two

syntypes (Table 1), specimen ZMB 17617 is here designated the lectotype of

Myctophum (Lampanyctus) nigrescens Brauer, 1904. Comparison of this lec-

totype with the holotype of Lampanyctus microchir Gilbert (USNM 74468) and

the lectotype of Myctophum (Lampanyctus) micropterum Brauer (ZMB 17616)

indicates that only a single species is involved. GR counts are similar (3 + 1+ 8,

ANNALS OF THE SOUTH AFRICAN MUSEUM

“SuaISaIsIU SninjoYyduUT Pue SnuvIIXaW sninjoyduy Jo UONNGINSIq °g “314

o0L 00S Se ol ol

o0€

00S o0L 006

oll o0 El

00S o0LI o0LI o0S| o0El

=i L =SE =

Sie

—-

uoypjndog

SPS

sNUDIIxEW sninjoyduy

suaosaibiu snunjoyduy

wiayjnos

uoypjndog

WBYON

O€

REVIEW OF THE LANTERNFISH GENUS TRIPHOTURUS 91

total 12), PO3 is level with PO, and PO;, and VLO is anterior to the vertical

through the outer ventral base in all three specimens. Further, in the three

specimens the infracaudal gland is shorter than 40 per cent of CPL. Some minor

differences in individual photophore position were observed: VO; is level with

the rest of the series in L. microchir and M. nigrescens, but slightly raised in

M. micropterum; SAO, is directly above VO, in M. micropterum, but slightly in

advance of the vertical through that photophore in L. microchir and M. nigres-

cens; and Prcy is on the line joining Prc, and Prc3 in M. nigrescens, touches the

line in L. microchir, and is slightly below the line in L. micropterus. Such

variation is mirrored in the South African population. Due to damage, D and A

counts were not possible for L. microchir but the values given by Gilbert (1913)

(14 and 17 respectively) fall within the range for Triphoturus nigrescens

(Fig. 4).

Distribution

Vertical distribution. Off Hawaii: adults day 400-900 m (maximum

650-700 m and 900-1 000 m); adults night 200-300 m and 600-1 000 m; juveniles

(less than 20 mm SL) showing little diel migration above 400 m (Imsand 1982).

The geographic distribution of the species is given in Figure 8.

Triphoturus nigrescens is an Oceanic, mesopelagic, warm-water species. In

the Pacific Ocean it is distributed between about 35°N and 38°S, and extends

eastwards in a tongue (0°—08°N) to the coasts of Ecuador, Colombia and Panama.

It is absent from upwelled waters of the California and Peru currents. In the

western North Pacific it reaches at least to 35°N in the Kuroshio Current. The

species is known from South-east Asian Seas and from between 22°S and 38°30’S

off the east coast of Australia and in the Tasman Sea (Paxton et al. in press). It is

distributed in the Indian Ocean between 08°N and 15°S and off the west

Australian coast (22°S), and has been reported as far south as 34° in the Agulhas

Current. Recent data from the Sea Fisheries Research Institute’s phyllosoma

sampling cruises off the west coast of South Africa has revealed the presence of

the species in the pockets of warm Agulhas Water, that round the Cape of Good

Hope and occur in the eastern South Atlantic. These specimens represent the first

records of the genus and the species in the Atlantic Ocean and are representative

of a Broadly Tropical Pattern (Agulhas Subpattern) of distribution (Hulley

1981).

ACKNOWLEDGEMENTS

My thanks are due to Mr S. X. Kannemeyer for assistance during the course

of this project and to the following persons for supplying me with specimens and/

or data: Drs Daniel M. Cohen, William N. Eschmeyer, Robert H. Gibbs,

Christine Karrer, John R. Paxton, Stuart G. Poss, and Melanie L. Striassny, and

to Karsten E. Hartel. The work was carried out under a grant from CSIR (FRD).

92 ANNALS OF THE SOUTH AFRICAN MUSEUM

REFERENCES

AHLSTROM, E. H. 1969. Mesopelagic and bathypelagic fishes in the California Current region.

Reports of the California Co-operative Oceanic Fisheries Investigations 13: 39-44.

AHLSTROM, E. H. 1971. Kinds and abundance of fish larvae in the eastern tropical Pacific, based

on collections made on EASTROPACI. Fishery Bulletin of the National Oceanic and

Atmospheric Administration 69 (1): 2-77.

AHLstTROM, E. H. 1972. Distributional atlas of fish larvae in the California Current region: six

common mesopelagic fishes—Vinciguerria lucetia, Triphoturus mexicanus, Stenobrachius

leucopsarus, Leuroglossus stilbius, Bathylagus wesethi and Bathylagus ochotensis, 1955

through 1960. Atlas of the California Co-operative Oceanic Fisheries Investigations 17: i—xv,

1-306.

Bartow, G. W. 1961. Causes and significance of morphological variation in fishes. Systematic

Zoology 10: 105-117.

BarRnETT, M. A. 1983. Species structure and temporal stability of mesopelagic fish assemblages

in the central gyres of the North and South Pacific Ocean. Marine Biology 74 (3): 245-256.

Barnetr, M. A. 1984. Mesopelagic fish zoogeography in the central tropical and subtropical

Pacific Ocean: species composition and structure at representative locations in three

ecosystems. Marine Biology 82 (2): 199-208.

BEEBE, W. & VAN DER Py, M. 1944. Eastern Pacific expeditions of the New York Zoological

Society. XXXIII. Pacific Myctophidae. Fishes. Zoologica, New York 29 (2): 59-95.

BEKKER, V. E. 1967. Luminous anchovies (fam. Myctophidae). Tichii Okean 7. Biologi Tichigi

Okeana. 3. Moskow: USSR Academy of Sciences. (In Russian.)

Berry, F. H. & PERKINS, H. C. 1966. Survey of pelagic fishes of the California Current area.

Fishery Bulletin of the U.S. Fish and Wildlife Service 65 (3): 625-682.

BOHLKE, J. 1953. A catalogue of the type specimens of recent fishes in the Natural History

Museum of Stanford University. Stanford Ichthyological Bulletin 5: 1-168.

Bo.in, R. L. 1939. A review of the myctophid fishes of the Pacific coast of the United States and

of Lower California. Stanford Ichthyological Bulletin 1 (4): 89-156.

BonaPARTE, C. L. P. 1840. Iconografia della fauna Italica per le quattro classi degli animali

vertebrati. 3. Pesci. Roma: Salviucci.

Brauer, A. 1904. Die Gattung Myctophum. Zoologischer Anzeiger 28 (10): 377-404.

BrAvueER, A. 1906. Die Tiefsee-Fische. I. Systematischer Teil. Wissenschaftliche Ergebnisse der

Deutschen Tiefsee-Expedition auf dem Dampfer ‘Valdivia’ 1898-1899 15: 1—420.

Brewer, G. D. 1973. Midwater fishes from the Gulf of California and the adjacent eastern

tropical Pacific. Contributions in Science. Los Angeles County Museum of Natural History

242: 1-47.

BussinGc, W. A. 1965. Studies of the midwater fishes of the Peru-Chile Trench. Biology of the

Antarctic Seas I]. Antarctic Research Series 5: 185-227.

CuiLpress, J. J., TayLor, S. M., Camtiret, G. M. & Price, M. H. 1980. Patterns of growth,

energy utilization and reproduction in some meso- and bathypelagic fishes off southern

California. Marine Biology 61 (1): 27-40.

CLARKE, T. A. 1973. Some aspects of the ecology of lanternfishes (Myctophidae) in the Pacific

Ocean near Hawaii. Fishery Bulletin of the National Oceanic and Atmospheric Administra-

tion 71 (2): 401-433.

CLARKE, T. A. 1980. Diets of fourteen species of vertically migrating mesopelagic fishes in

Hawaiian waters. Fishery Bulletin of the National Oceanic and Atmospheric Administration

78 (3): 619-640.

Crappock, J. E. & MEAD, G. W. 1970. Midwater fishes from the eastern South Pacific Ocean.

Scientific results of the Southeast Pacific Expedition. Scientific Results of the Southeast

Pacific Expedition. Anton Bruun Report 3: 3.3-3.46.

EBELING, A. W., IBARA, R. M., LAVENBERG, R. J. & RouLF, F. J. 1970. Ecological groups of

deep-sea animals off southern California. Bulletin of the Los Angeles County Museum of

Natural History (Science) 6: 1-43.

Fow Ler, H. W. 1934. Descriptions of new fishes obtained in 1907 to 1910, chiefly in the

Philippine Islands and adjacent seas. Proceedings of the Academy of Natural Sciences of

Philadelphia 85: 233-367.

REVIEW OF THE LANTERNFISH GENUS TRIPHOTURUS 93

Fow er, J. A. 1970. Control of vertebral number in teleosts—an embryological problem.

Quarterly Reviews of Biology 45: 148-167.

FRASER-BRUNNER, A. 1949. A classification of the fishes of the family Myctophidae. Proceedings

of the Zoological Society of London 118 (4): 1019-1106.

GARMAN, S. 1899. Reports on an expedition off the west coasts of Mexico, Central and South

America, and off the Galapagos Islands in charge of Alexander Agassiz, by the U.S. Fish

Commission Steamer ‘Albatross’ during 1891, Lieut. Commander Z. L. Tanner, U.S.N.

commanding. XXVI. The fishes. Memoirs of the Museum of Comparative Zoology at

Harvard College 24: 1-431.

GILBERT, C. H. 1890. A preliminary report on the fishes collected by the steamer ‘Albatross’ on

the Pacific coast of North America during the year 1889, with descriptions of twelve new

genera and ninety-two new species. Proceedings of the U.S. National Museum 13 (797):

49-126.

GiwBerT, C. H. 1913. The lantern-fishes of Japan. Memoirs of the Carnegie Museum 6 (2):

67-107.

GoopE, G. B. & BEAN, T. H. 1895. Oceanic ichthyology, a treatise on the deep-sea and pelagic

fishes of the world, based chiefly upon the collections made by the steamers ‘Blake’,

‘Albatross’, and ‘Fish Hawk’ in the northwestern Atlantic, with an atlas containing 417

figures. Memoirs of the Museum of Comparative Zoology at Harvard College 1: 1-553.

HARTMANN, A. R. & CLARKE, T. A. 1975. The distribution of myctophid fishes across the

central equatorial Pacific. Fishery Bulletin of the National Oceanic and Atmospheric

Administration 73 (3): 633-641.

Hotton, A. A. 1969. Feeding behaviour of a vertically migrating lanternfish. Pacific Science 23

(3): 325-331.

Husss, C. L. & Husss, C. 1953. An improved graphical analysis and comparison of series of

samples. Systematic Zoology 2 (2): 49-57.

HUuLiey, P. A. 1981. Results of the research cruises of FRV “Walther Herwig’ to South America.

LVIII. Family Myctophidae (Osteichthyes, Myctophiformes). Archiv fiir Fischereiwissen-

schaft 31 (1): 1-300.

Hu.tey, P. A. 1984. The South African Museum’s Meiring Naude cruises. Part 14. Family

Myctophidae (Osteichthyes, Myctophiformes). Annals of the South African Museum 93 (2):

53-96.

Huey, P. A. In press. Family Myctophidae. Jn: SmirH, M. M. & Heemstra, P. C. eds.

Smith’s sea fishes.

Humenrigs, J. M., BooKSTEIN, F. L., CHERNOFF, B., SmiTH, G. R., ELDER, R. L. & Poss, S. G.

1981. Multivariate discrimination by shape in relation to size. Systematic Zoology 30 (3):

291-308.

IMsAND, S. 1974. Population dynamics of two lanternfish: Triphoturus mexicanus of the

California Current and T. nigrescens of the North Pacific Central Gyre. Biennial Report for

the two years ending 30 June 1974, Institute of Marine Sciences, University of California:

37-38.

IMSAND, S. 1982. Abundance in space and time of lanternfishes of the genus Triphoturus

(Myctophidae) in the Pacific Ocean. Ph.D. thesis: University of California (San Diego).

JoHnson, R. K. & Barnett, M. A. 1975. An inverse correlation between meristic characters

and food supply in midwater fishes; evidence and possible explanations. Fishery Bulletin of

the National Oceanic and Atmospheric Administration 73 (2): 284-298.

JOHNSON, R. K. & Fettes, R. M. 1984. A new species of Vinciguerria (Salmoniformes:

Photichthyidae) from the Red Sea and the Gulf of Aqaba, with comments on the

depauperacy of the Red Sea mesopelagic fish fauna. Fieldiana: Zoology (N.S.) 22: 1-35.

Korruaus, A. 1972. Fische des Indischen Ozeans. Ergebnisse der ichthyologischen Unter-

suchungen wahrend der Expedition des Forschungsschiffes ‘Meteor’ in den Indischen

Ozean, Oktober 1964 bis Mai 1965. A. Systematischer Teil, IX. Iniomi (Nachtrag: Fam.

Myctophidae). Meteor Forschungsergebnisse (D) 12: 12-35.

LeGAND, M., Bourret, P., FouRMANOIR, P., PERRIN, R., GUEREDRAT, J. A., MICHEL, A.,

REPELIN, R. & Rocer, C. 1972. Relations trophiques et distributions verticales en milieu

pélagique dans l’Océan Pacifique intertropical. Cahiers de I’ Office Recherche scientifique et

technique Outre-Mer (Série Océanographie) 10 (4): 304-393.

94 ANNALS OF THE SOUTH AFRICAN MUSEUM

Logs, V. J. 1979a. Larval fishes in the zooplankton community of the North Pacific Central

Gyre. Marine Biology 53 (2): 173-191.

Logs, V. J. 19796. Vertical distribution and development of larval fishes in the North Pacific

Central Gyre during summer. Fishery Bulletin of the National Oceanic and Atmospheric

Administration 77 (4): 777-793.

Loes, V. J. 1980. Pattern of spatial and species abundance within the larval fish assemblage of

the North Pacific Central Gyre during late summer. Marine Biology 60 (2/3): 189-200.

Logs, V. J., Smiru, P. E. & Moser, H. G. 1983a. Geographical and seasonal patterns of larval

fish species structure in the California Current area, 1975. Reports of the California Co-

operative Oceanic Fisheries Investigations 24: 132-151.

Logs, V. J., SmitH, P. E. & Moser, H. G. 19835. Recurrent groups of larval fish species in the

California Current area. Reports of the California Co-operative Oceanic Fisheries

Investigations 24: 152-164.

LUrken, C. F. 1892. Spolia Atlantica. Scopelini Musei Zoologici Universitatis Hauiensis.

Kongelige Danske Videnskabernes Selskabs Skrifter (6) 7: 221-297.

McNutty, J. A. & Narpaxtitis, B. G. 1976. The structure and development of the pineal

complex in the lanternfish Triphoturus mexicanus (family Myctophidae). Journal of

Morphology 150 (2): 579-605.

Moser, H. G. & Autstrom, E. H. 1970. Development of lanternfishes (family Myctophidae) in

the California Current. Part I. Species with narrow-eyed larvae. Bulletin of the Los Angeles

County Museum of Natural History (Science) 7: 1-145.

Moser, H. G. & Autstrom, E. H. 1974. Role of larval stages in systematic investigations of

marine teleosts: the Myctophidae, a case study. Fishery Bulletin of the National Oceanic and

Atmospheric Administration 72 (2): 391-413.

Moser, H. G., AuLSTROM, E. H., KrAmer, D. & STEVENS, E. G. 1974. Distribution and

abundance of fish eggs and larvae in the Gulf of California. Reports of the California Co-

operative Oceanic Fisheries Investigations 17: 112-128.

Narpaxtitis, B. G. 1973. A review of the lanternfishes (family Myctophidae) described by

A. Vedel Taning. Dana Report 83: 1-46.

NAFPAKTITIS, B. G., BAcKUS, R. H., CrAppock, J. E., HAEDRICH, R. L., Ropison, B. H. &

KARNELLA, C. 1977. Fishes of the western North Atlantic. Family Myctophidae. Memoirs

of the Sears Foundation for Marine Research 1 (7): 13-265.

NAFPAKTITIS, B. G. & NArFpaAktTitis, M. 1969. Lanternfishes (family Myctophidae) collected

during cruises 3 and 6 of the R/V Anton Bruun in the Indian Ocean. Bulletin of the Los

Angeles County Museum of Natural History (Science) 5: 1-79.

NeicHBors, M. A. & Narpakxtitis, B. G. 1982. Lipid compositions, water contents,

swimbladder morphologies and buoyancies of nineteen species of midwater fishes (18

myctophids and 1 neoscopelid). Marine Biology 66 (3): 207-215.

PAPPENHEIM, P. 1914. Die Fische der deutschen Stidpolar-Expedition 1901-1903. II. Die

Tiefseefische. Deutsche Stidpolar-Expedition, 1901-1903 15 (Zoologie 7): 163-200.

ParRIN, N. V., BEKKER, V. E., BoRODULINA, O. D. & TcHuvasov, V. M. 1973. Deep-sea

pelagic fishes of the south-eastern Pacific Ocean. Trudy Instituta okeanologii. Akademiya

nauk SSSR 94: 71-172. (In Russian.)

PaRIN, N. V., BEKKER, V. E., BoORODULINA, O. D., KarmovskayAa, E. S., FEDORYAKO, B. I.,

SHCHERBACHEV, J. N., POKHILSKAYA, G. N. & TcHuvasov, V. M. 1977. Midwater fishes in

the western tropical Pacific Ocean and the seas of the Indo-Australian Archipelago. Trudy

Instituta okeanologii. Akademiya nauk SSSR 107: 68-199. (In Russian.)

Parr, A. E. 1928. Deepsea fishes of the order Iniomi from the waters around the Bahama and

Bermuda islands, with annotated keys to the Sudidae, Myctophidae, Scopelarchidae,

Evermannellidae, Omosudidae, Cetomimidae and Rondeletiidae of the world. Bulletin of

the Bingham Oceanographic Collection 3 (3): 1-193.

Parr, A. E. 1931. Deepsea fishes from off the western coast of North and Central America.

Bulletin of the Bingham Oceanographic Collection 2 (4): 1-53.

PAXTON, J. R. 1967. A distributional analysis for the lanternfishes (family Myctophidae) of the

San Pedro Basin, California. Copeia 1967 (2): 422-440.

PAxTon, J. R. 1972. Osteology and relationships of the lanternfishes (family Myctophidae).

Science Bulletin. Natural History Museum, Los Angeles County 13: 1-81.

REVIEW OF THE LANTERNFISH GENUS TRIPHOTURUS 95

PaxToN, J. R. 1979. Nominal genera and species of lanternfishes (family Myctophidae).

Contributions in Science. Los Angeles County Museum 322: 1-28.

PERTSEVA-OstroumMovA, T. A. 1974. New data on lantern-fish larvae (Myctophidae, Pisces)

with oval eyes from the Indian and Pacific oceans. Trudy Instituta okeanologii. Akademiya

nauk SSSR 96: 77-142. (In Russian. )

PIEPER, R. E. & Barco, B. G. 1980. Acoustic measurements of a migrating layer of the Mexican

lampfish, Triphoturus mexicanus, at 102 kiloherz. Fishery Bulletin of the National Oceanic

and Atmospheric Administration 77 (4): 935-942.

Rosison, B. H. 1972. Distribution of midwater fishes of the Gulf of California. Copeia 1972 (3):

448-461.

SNEDECOR, G. W. & COcHRAN, W. 1967. Statistical methods. 6th ed. Ames: Iowa State

University Press.

SOKAL, R. R. & Routr, F. J. 1969. Biometry. The principles and practice of statistics in biological

research. San Francisco: Freeman & Co.

WEBER, M. & DE Beaufort, L. F. 1913. The fishes of the Indo-Australian Archipelago. II.

Malacopterygii, Myctophoidea, Ostariophysi: I. Siluroidea. Leiden: Brill.

WISNER, R. L. 1974. Descriptions of five new species of myctophid fishes from the Pacific, Indian

and Atlantic oceans. Occasional Papers of the California Academy of Sciences 110: 1-37.

WISNER, R. L. 1976. The taxonomy and distribution of lanternfishes (family Myctophidae) of the

eastern Pacific Ocean. Washington: U.S. Government Printing Office.

ZAR, J. H. 1974. Biostatistical analysis. Englewood Cliffs: Prentice-Hall.

pean aes

Ses Fs 2a

= ; < : -

- ¥ = = = 7 4 7 =< = a

a = :

= ae : 2 te a

<— = 7 i es =<

~ = 7

i 1 PS °

ep ana

{

6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu-

larly Articles 22 and 51).

Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed

by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov.,

etc.

An author’s name when cited must follow the name of the taxon without intervening punctuation

and not be abbreviated; if the year is added, a comma must separate author’s name and year. The

author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is trans-

ferred from its original genus. The name of a subsequent user of a scientific name must be separated

from the scientific name by a colon.

Synonymy arrangement should be according to chronology of names, i.e. all published scientific

names by which the species previously has been designated are listed in chronological order, with all

references to that name following in chronological order, e.g.:

Family Nuculanidae

Nuculana (Lembulus) bicuspidata (Gould, 1845)

Figs 14-15A

Nucula (Leda) bicuspidata Gould, 1845: 37.

Leda plicifera A. Adams, 1856: 50.

Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b).

Nucula largillierti Philippi, 1861: 87.

Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9.

Note punctuation in the above example:

comma separates author’s name and year

semicolon separates more than one reference by the same author

full stop separates references by different authors

figures of plates are enclosed in parentheses to distinguish them from text-figures

dash, not comma, separates consecutive numbers.

Synonymy arrangement according to chronology of bibliographic references, whereby the year is

placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable.

In describing new species, one specimen must be designated as the holotype; other specimens

mentioned in the original description are to be designated paratypes; additional material not regarded

as paratypes should be listed separately. The complete data (registration number, depository, descrip-

tion of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.:

Holotype

SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach, Port Eliza-

beth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973.

Note standard form of writing South African Museum registration numbers and date.

7. SPECIAL HOUSE RULES

Capital initial letters

(a) The Figures, Maps and Tables of the paper when referred to in the text

e.g. ‘. .. the Figure depicting C. namacolus .. .’: ‘. . «in C. namacolus (Fig. 10). . .’

(b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by

initials or full names

e.g. DuToit but A.L.du Toit; Von Huene but F. von Huene

e.g. Therocephalia, but therocephalian

Punctuation should be loose, omitting all not strictly necessary

Reference to the author should preferably be expressed in the third person

Roman numerals should be converted to arabic, except when forming part of the title of a book or

article, such as

‘Revision of the Crustacea. Part VIII. The Amphipoda.’ Lee

Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial

capital letter, provided the same generic name is used consecutively. The generic name should

not be abbreviated at the beginning of a sentence or paragraph.

Name of new genus or species is not to be included in the title; it should be included in the abstract,

counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts.

P. ALEXANDER HULLEY

A TAXONOMIC REVIEW OF THE

LANTERNFISH GENUS TRIPHOTURUS

FRASER-BRUNNER, 1949

(MYCTOPHIDAE, OSTEICHTHYES)

= 97 PART 5 JUNE 1986 ISSN 0303-2515

ae VEE

HS!

‘SON NAN in

5 ~~

imrmactly

Lt6 2aRitc

a ine peed

ee ad

ANNALS

OF THE SOUTH AFRICAN

MUSEUM

CAPE TOWN

INSTRUCTIONS TO AUTHORS

1. MATERIAL should be original and not published elsewhere, in whole or in part.

2. LAYOUT should be as follows:

(a) Centred masthead to consist of

Title: informative but concise, without abbreviations and not including the names of new genera or species

Author’s(s’) name(s)

Address(es) of author(s) (institution where work was carried out)

Number of illustrations (figures, enumerated maps and tables, in this order)

(b) Abstract of not more than 200 words, intelligible to the reader without reference to the text

(d) Introduction

(e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents

(f) Summary, if paper is lengthy

(g) Acknowledgements

(h) References

(i) Abbreviations, where these are numerous.

3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with

3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for

illustrations should be typed separately, their positions indicated in the text. All pages should be num-

bered consecutively.

Major headings of the paper are centred capitals; first subheadings are shouldered small capitals;

second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further

subdivisions should be avoided, as also enumeration (never roman numerals) of headings and

abbreviations.

Footnotes should be avoided unless they are short and essential.

Only generic and specific names should be underlined to indicate italics; all other marking up

should be left to editor and publisher.

4. ILLUSTRATIONS should be reducible to a size not exceeding 12 x 18 cm (19 cm including

caption); the reduction or enlargement required should be indicated (and preferably uniform); orig-

inals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and

final size. A metric scale should appear with all illustrations, otherwise magnification or reduction

should be given in the caption; if the latter, then the final reduction or enlargement should be taken

into consideration.

All illustrations, whether line drawings or photographs, should be termed figures (plates are not

printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items

of composite figures should be designated by capital letters; lettering of figures is not set in type and

should be in lower-case letters. If Letraset is used authors are requested to use Helvetica-style letter-

ing, if possible.

The number of the figure should be lightly marked in pencil on the back of each illustration.

5. REFERENCES cited in text and synonymies should all be included in the list at the end of the

paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable):

(a) Author’s name and year of publication given in text, e.g.:

‘Smith (1969) describes . . .’

‘Smith (1969: 36, fig. 16) describes. . .”

‘As described (Smith 1969a, 1969b; Jones 1971)’

‘As described (Haughton & Broom 1927)...’

‘As described (Haughton et al. 1927)...”

Note: no comma separating name and year

Pagination indicated by colon, not p.

names of joint authors connected by ampersand

et al. in text for more than two joint authors, but names of all authors given in list of references.

(b) Full references at the end of the paper, arranged alphabetically by names, chronologically within

each name, with suffixes a, b, etc., to the year for more than one paper by the same author in

that year, e.g. Smith (1969a, 1969b) and not Smith (1969, 1969a).

For books give title in italics, edition, volume number, place of publication, publisher.

For journal article give title of article, title of journal in italics (according to the World list of scientific periodicals. 4th ed.

London: Butterworths, 1963), series in parentheses, velume number, part number in parentheses, pagination (first and

last pages of article).

Examples (note capitalization and punctuation)

BuLLouGH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan.

FiscHER, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140.

FiscHER, P. H., DuvAL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie

expérimentale et générale 74 (33): 627-634.

Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and

Magazine of Natural History (13) 2 (17): 309-320.

Koun, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of

the Bingham Oceanographic Collection, Yale University 17 (4): 1-51.

THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: ScHULTZE, L. Zoologische und anthro-

pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Stid-Afrika ausgeftihrt in den Jahren

1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270.

(continued inside back cover)

ANNALS OF THE SOUTH AFRICAN MUSEUM

ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

Volume 97 Band

June 1986 Junie

Part 5 Deel

THE SOUTH AFRICAN MUSEUM’S

MEIRING NAUDE CRUISES

PART 16

BRACHIOPODA FROM THE

1975-1979 CRUISES

NORTON HILLER

Cape Town Kaapstad

The ANNALS OF THE SOUTH AFRICAN MUSEUM

are issued in parts at irregular intervals as material

becomes available

Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000

Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

word uitgegee in dele op ongereelde tye na gelang van die

beskikbaarheid van stof

Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000

OUT OF PRINT/UIT DRUK

i, DOS, SS), HU, 4-5, 8) eaas.)y, SES, 5, 7-9),

Gy eas), CD), HOOD) Dy 1009),

GED, Sh eros), IMGD), ISS), DD), D7, SNES), BAG), 38, SEO), 45)

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 082 6

In Suid-Afrika gedruk deur

Die Rustica-pers, Edms., Bpk.,

Courtweg, Wynberg, Kaap

Printed in South Africa by

The Rustica Press; Piva. Wtidy

Court Road, Wynberg, Cape

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES

PART 16

BRACHIOPODA FROM THE 1975-1979 CRUISES

NorTON HILLER

Department of Geology, Rhodes University, Grahamstown, South Africa

(With 19 figures and 1 table)

[MS accepted 3 December 1985]

ABSTRACT

Sixteen species of articulate brachiopods and a single inarticulate brachiopod are recorded

from off the east coast of South Africa. Of the seventeen species, Grammetaria africana,

Notozyga gracilis and Megerlia acrura are new, and five others are recorded for the first time from

South African waters.

CONTENTS

PAGE

MMNiOGUCHONe rr eerie une that Ae ees tk ha 97

Pe VIOMSPRESCATEM karen eteieiel ake Or A ga deine cs Ae ee v7

J EISICOLEG DELOS, Soa oa cera ORR an ee ae a 99

Brachopoddistmbuttonerayess 2. + sane Gan oer es se 4 - 100

Geopraphic distributions. 4-446 on een ee. 100

Batrhyiiletuc disthibutonese a en ear cee. . 101

SVStEMMALC ACEO UMUC eens sere Ps aes besa Ss oe 102

PNCKMOWICOSEMENISH 12g ayecin eel | Slat ak ek toe eaeee es 138

ING@IIEIRETIGSS dein hereto sed tee UES Neen tea ne ROC Ga 139

INTRODUCTION

The present paper deals with a number of brachiopod species that were

collected during the Meiring Naude cruises in the years from 1975 to 1979

inclusive. The positions of the various sampling stations from which brachiopods

were recovered are shown in Figure 1; further data on these stations may be

obtained from Louw (1977, 1980). A few specimens that have been separated

from coral material of various origins have also been included. Station data for

these specimens have been given separately along with the descriptions of the

species concerned.

PREVIOUS RESEARCH

Very little research work has been done on the South African brachiopod

fauna; the earliest records date from the late eighteenth and early nineteenth

centuries and deal with the commonly occurring members of the Kraussinidae.

From 1850 onwards a handful of species belonging to other families have been

added to the list and described in a number of publications, including the reports

oT)

Ann. S. Afr. Mus. 97 (5), 1986: 97-140, 19 figs, 1 table.

98 ANNALS OF THE SOUTH AFRICAN MUSEUM

Lake if 28

St Lucia

SOUTH

‘103

AFRICA

DURBAN

INDIAN

OCEAN

200km

Fig. 1. Map showing locations of SM stations from which brachiopods were recovered during the

Meiring Naude cruises 1975-1979.

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 99

of two major expeditions—the ‘Challenger’ Expedition (Davidson 1880) and the

‘Valdivia’ Expedition (Blochmann 1906).

The most comprehensive paper, and the only one dealing specifically with

South African brachiopods, was that of Jackson (1952), which includes an

account of all the earlier work. Jackson recorded fifteen species belonging to nine

genera from a number of stations extending from Saldanha Bay on the west coast

round to Delagoa Bay (now Maputo Bay, Mozambique) on the east. Of the forms

recorded by Jackson, five—Crania, Agulhasia, Terebratulina, Kraussina and

Megerlina—are recorded again in this paper; two of the species he described

from western Cape waters have not, as yet, been found east of Cape Agulhas, viz.

Megathiris capensis and Kraussina crassicostata.

Since the publication of Jackson’s work only three papers by Cooper (1973b,

1973c, 1981) have added to our knowledge of the brachiopods from South African

waters. The papers have mostly confirmed previous records but to the list of

species they have added Platidia anomioides and Xenobrochus africanus. Species

such as Chlidonophora chuni and Eucalathis fasciculata, described by Cooper

from off Madagascar, can now be added to the South African list.

LIST OF SPECIES

Station Complete Pedicle Brachial

No. specimens** valves valves

Family Craniidae

(CHRO SOS oes sy SLE Ae a SM 239 it (i) — —

Family Frieleiidae

*Grammetaria africana sp.nov. ........... SM 232 Gs) — —

Family Dyscoliidae

*Dyscolia cf. johannisdavisi (Alcock) ...... SM 174 4 (1) 2 1

Xenobrochus africanus (Cooper) .......... SM 103 13 (13) 1 —

SM 129 1 (1) — —

SM 131 — 3 1

SM 162 — 1 —

Xenobrochus agulhasensis (Helmcke) ...... SM 232 if — ==

XCHODLOCHUSDY SP eee ee nee ete ens SM 131 1 = =

XCHODFOCMUSIUS ID Wi Sid toda gin eek ele 8 SM 226 — 1 1

SM 232 — 6 3

Family Cancellothyrididae

NK CONOMI OSD s va ee tre eng a PRG Se ae i SM 131 8 1 3

SM 180 1 — =

Family Chlidonophoridae

*Chlidonophora chuni Blochmann......... SM 103 2) — ==

SM 246 31%) = ==

DNOlOZ) 2a eracilis Sp_MOVe aes, 2 56. > SM 131 i == =

NOLO DEUS Diem diet Rs Se ah ee Ak ws SM 131 1 == =

| Bucalathisyasciculata\ Cooper + .50.-.-: 4: - SM 60 4 (4) — =

SM 103 1 = =

SM 246 I @D) Z 2

ASUINGSI@AAVIGSOND KANG -e an... seae es: SM 131

* New records for South African waters

** Number of live specimens shown in brackets

100 ANNALS OF THE SOUTH AFRICAN MUSEUM

Station Complete Pedicle Brachial

No. specimens** valves valves

Family Kraussinidae

Kraussinarubra (Pallas). 35 ese eas SM 129 ~- 1 —

SM 131 1 1 —

SM 185 i (a) — —

* Mecerliaacrund Sp nOVaRnn ae ere SM 255 il (1) — —

SM 239 2 (2) —

Megerlina pisum (Lamarck) .............. SM 163 1 — —

SM 179 1 (1) — —

SM 180 1 — —

SM 185 TQ) 2 2

Family Phaneroporidae

*Leptothyrella cf. ignota (Muir-Wood) ..... SM 129 1 — —

* New records for South African waters

** Number of live specimens shown in brackets

BRACHIOPOD DISTRIBUTION

As reported by Cooper (19736, 1973c), Recent brachiopods are commonly

thought of as rare animals and while it is true they are greatly outnumbered by

molluscs in the world’s oceans, they are more widely distributed and in greater

variety than previously suspected. In the last 20 years research cruises by a

number of vessels from different countries have added considerably to our

understanding of modern-day brachiopod distribution. They are known from all

parts of the world and in some places form the major elements in the invertebrate

macrofauna (Cooper 1973c: 1).

On a local scale, the cruises of the Meiring Naude have added substantially to

the knowledge of the brachiopod fauna off the South African coast. In the cruises

from 1975 to 1979 the vessel has collected 17 species belonging to 13 genera, 8 of

the species recorded for the first time from South African waters.

GEOGRAPHIC DISTRIBUTION

The brachiopod fauna from off the Natal and eastern Cape coast, as is to be

expected, shows greatest affinity with other Indian Ocean faunas. Some species

are known only from South African waters: Agulhasia davidsoni, Kraussina

rubra, Xenobrochus africanus and X. agulhasensis, while others extend their

range from other parts of the Indian Ocean: Dyscolia johannisdavisi from around

the Maldive Islands, Chlidonophora chuni from the Maldive Islands and south of

Madagascar, Eucalathis fasciculata from south of Madagascar, and Leptothyrella

ignota from off Zanzibar and the Gulf of Aden—assuming the specimens of

Dyscolia and Leptothyrella described herein are indeed conspecific with the forms

mentioned. Two new species represent totally unexpected additions to the fauna:

Grammetaria africana is the first record of the genus outside the Philippines and

Notozyga gracilis is a new form of a genus previously only recorded from the

Caribbean.

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 101

BATHYMETRIC DISTRIBUTION

It is generally believed that the greatest diversity of brachiopod species is to

be found on the continental shelf areas, generally taken by geologists to extend

down to the 200 m line. Zezina (1970) states that the overwhelming majority of

brachiopods live at depths down to 500 m. If the figures of 200 m and 500 m are

taken to divide the ocean into shallow, intermediate and deep zones, then it can

be seen from Table 1 that by far the greatest number of specimens collected by

the Meiring Naude came from waters more than 500 m deep, with only Crania sp.

and the members of the Kraussinidae being essentially shallow-water forms.

According to previous descriptions, this is to be expected for species such as

Dyscolia cf. johannisdavisi, Chlidonophora chuni, Eucalathis fasciculata and

Leptothyrella cf. ignota, again assuming that the South African specimens of

Dyscolia and Leptothyrella are conspecific with the previously described forms.

However, previous records of Agulhasia davidsoni, Xenobrochus africanus and

X. agulhasensis suggest they are forms that live at shallow to intermediate depths.

In the case of A. davidsoni the state of the shells suggests they have been moved

from their living site and they may well have been carried by currents from

shallower water. Xenobrochus africanus has previously been recorded only from

Durban Bay at a depth of 366 m and since some of the specimens described

herein were found live, these new records simply extend the range of the species

into deeper water.

The single specimen of X. agulhasensis was recovered dead so it may well

have been transported from shallower water, but it must be remembered that the

TABLE 1

Depth ranges of the brachiopod species collected during the Meiring Naude cruises 1975-1979.

Shallow Intermediate Deep

0-200 m 200-500 m >500 m

Species (8 stations) (4 stations) (35 stations)

CHOTA S 0s Boe 6 SO a eee Xx

Grammetania africana... 2. ).... 02.2...

Dyscolia cf. johannisdavisi.............

MCHODTOCIUSIAIFICGNUS). =. a5 sass so:

Xenobrochus agulhasensis .............

XCHODLOCHUSISD Mn 0 balck a ou cen is nt

PCH ODLOCHUSISDOD Aa ae Sie crene aici ns 6 ee

ROD RETLTUTES) Yes eno eee ee x

Ghiidonophora chunt 2.65.52 .--4.-.-

IN OBDAYTE ECIGIIIS ees esi ee

INI OHOPASKTIIS Os ithe ined eee

FSUCGIAINISIIASCICUIGIA 26. 0.1 een cn.

VAGUIMGSUA AGVIGSOND as ae ee eae. a:

KG AUSSIRGMUOT Gna Ne ON. Plt ey xX

IMC RENE ACKUTG cso. eek Sok aes x

IMC CTIITIGI DISTT Bele At cis cele jaja ca eis x

Leptothyrellacttgnota...2:.0......---: xX

a a a as a

102 ANNALS OF THE SOUTH AFRICAN MUSEUM

continental shelf off the east coast of South Africa is very narrow with a steep

continental slope and this may well have a bearing on the bathymetric distribution

of the brachiopods. Certainly the previous records of X. agulhasensis are from

shallow and intermediate waters off the Cape of Good Hope, and on the Agulhas

Bank where the continental shelf is much broader.

SYSTEMATIC ACCOUNT

Most of the genera recorded here belong to the superfamilies Terebratulacea

and Cancellothyridacea of the suborder Terebratulidina. The classification

schemes employed herein for these two superfamilies are those proposed by

Cooper (1973a) for the Cancellothyridacea and Cooper (1983) for the Terebratu-

lacea. Otherwise the classification is that used by Williams et al. (1965) in the

Treatise on invertebrate paleontology.

Class INARTICULATA Huxley, 1869

Order ACROTRETIDA Kuhn, 1949

Suborder CRANIIDINA Waagen, 1885

Superfamily CRANIACEA Menke, 1828

Family Craniidae Menke, 1828

Genus Crania Retzius, 1781

Crania sp.

Jey, Z

Material

A single live specimen (SAM-—A25445) found attached to Megerlia acrura

sp. nov. from SM 239 at a depth of 90 m.

Description

Small elongately oval shell with conical profile; apex low, about one-third of

valve length from posterior margin. Posterior slope gently convex; anterior slope

Fig. 2. Crania sp., SAM—A25445, SM 239. A. Ventral valve exterior, ventral view. B. Dorsal

valve exterior, dorsal view. Both x8. The specimen is damaged.

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 103

gently convex near apex, becoming concave towards margin. Shell substance very

thin; ornamented by concentric growth lines.

Ventral valve concave from attachment to convex surface and pitted where it

covered tubercules on surface of pedicle valve of Megerlia acrura.

Details of interiors of both valves obscure except for pustules on inside of

ventral valve corresponding to pits on outer surface.

Dimensions (mm)

Length Width

SAM-A25445 c. 4 CxS

Discussion

Previous descriptions of Crania, such as that by Thomson (1927: 135), show

it to be a genus that displays considerable variation among its assigned species,

many of which are inadequately defined, a point made by Cooper (1973c: 19) in

his description of C. patagonica Dall. The nearest species of Crania to that

described here is C. roseoradiata Jackson, which is recorded from off the Cape

west coast, although Turton (1932: 260) records a shell that may doubtfully be

assigned to Crania from Port Alfred on the east coast. The present specimen was

damaged during recovery and the shell substance is so thin (although calcified)

that it offers no details of the internal structures that may be used in comparisons

with other species. Certainly, the specimen lacks the radiating rose-coloured

streaks that are said to distinguish Jackson’s species, although colour is not a good

criterion on which to define species. More and better-preserved specimens are

required before this one can be named.

Class ARTICULATA Huxley, 1869

Order RHYNCHONELLIDA Kuhn, 1949

Superfamily RHYNCHONELLACEA Gray, 1848

Family Frieleiidae Cooper, 1959

Genus Grammetaria Cooper, 1959

Grammetaria africana sp. nov.

Fig. 3

Diagnosis

Triangular Grammetaria with straight anterior margin parallel to hinge line;

ventral umbo sharply pointed, nearly straight; radial ornament only faintly

developed.

Material

Holotype. SAM—A25446 in the South African Museum, Cape Town. From

SM 232 (32°14,9'S 20°10,4’E) at a depth of 560-620 m, 25 June 1979. The sole

104 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 3. Grammetaria africana sp. nov., SAM—A25446, holotype, SM 232. A-B. Internal and

external views of pedicle valve. C. Close-up of ventral beak, dorsal view, showing auriculate

deltidial plates. D-—E. Ventral and lateral views of cardinalia. A-B <3. C—E X8. The brachial

valve is damaged. |

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 105

specimen was live when collected, being attached by a short slender pedicle to a

small pebble.

Description

Elongate triangular outline with maximum width close to anterior margin.

Valves convex with pedicle valve slightly deeper than brachial. Anterior

commissure rectimarginate. Pedicle valve with evenly convex lateral profile;

anterior profile gently convex in central portion between almost flat, steeply

sloping flanks. Beak small, pointed, nearly straight. Foramen small, elongately

oval, hypothyridid; deltidial plates auriculate, conjunct. Ornament of faintly

developed capillae and concentric growth rings.

Pedicle valve interior with small corrugated teeth, supported by strong

vertical dental plates. Pedicle collar well developed, free anteriorly.

Brachial valve interior with corrugated sockets bounded by strong socket

ridges. Crura short, curved, of spinulifer type, triangular in cross-section but

becoming flattened, blade-like distally. Outer hinge plates very narrow; inner

hinge plates fused medianly to plug of shell material that occupies space bounded

by median septum and crural bases. Median ridge thick, supporting proximal

ends of crural bases.

Dimensions (mm)

Length Width

SAM-A25446 Holotype 10,9 Do

Discussion

Grammetaria is characterized by its rectimarginate anterior commissure and

its auriculate and conjunct deltidial plates. Possession of these features allows the

present specimen to be referred immediately to that rare genus.

The only other species thus far assigned to the genus is G. bartschi (Dall),

which is represented by only two specimens recovered from Philippine waters

(Cooper 1959: 58) and a fragmentary specimen from off Bali (Zezina 1981: 12).

They show some differences with that described here and are therefore taken to

belong to a separate species. The anterior margin of the South African shell is not

curved like that of G. bartschi but is remarkably straight and parallel to the hinge

axis. The beak is much more pointed and nearly straight rather than suberect, and

the radial ornament is much less strongly developed.

The exact relationships between these two species cannot be properly

assessed on so few specimens. When more material becomes available it might be

possible to take into account the full range of variation but until that time it is felt

that sufficient differences exist between the South African and Philippine

specimens for them to be regarded as separate species.

Etymology

The specific name alludes to the fact that this is the first record of the genus

from African waters.

106 ANNALS OF THE SOUTH AFRICAN MUSEUM

Order TEREBRATULIDA Waagen, 1883

Suborder TEREBRATULIDINA Waagen, 1883

Superfamily TEREBRATULACEA Gray, 1840

Family Dyscoliidae Fischer & Oehlert, 1891

Subfamily Dyscoliinae Fischer & Oehlert, 1891

Genus Dyscolia Fischer & Oehlert, 1890

Dyscolia cf. johannisdavisi (Alcock, 1894)

Figs 4-5

Terebratula johannisdavisi Alcock, 1894: 139. Blochmann, 1908: 638.

Terebratula wyvillei Davidson: Thomson, 1927: 201 (in part).

Dyscolia johannisdavisi (Alcock): Helmcke, 1940: 261, figs 22, 25b. Muir-Wood, 1959: 300, pl. 1

(figs 1, 3, 4).

Material

Four pairs of conjoined valves, one live at time of collection, plus a brachial

and two pedicle valves (SAM-—A25447 to A25453) all from SM 174 at a depth of

760 m.

Description

Large, roundedly triangular to elongately oval shells; biconvex with

maximum width at midvalve or anterior to midvalve. Lateral and anterior

margins of largest specimens strongly incurved (flanged); anterior commissure

rectimarginate. Beak short, truncated, suberect to erect; foramen large,

subcircular, permesothyridid to epithyridid, labiate. Symphytium concave,

almost completely hidden by dorsal beak. Shell substance thick in large

specimens; surface marked by concentric growth lines and very faint radial

capillae.

Pedicle valve fairly evenly convex in lateral profile; anterior profile variably

convex, often strongly domed with steep flanks. Brachial valve moderately to

strongly convex in lateral profile; gently domed in anterior profile. Greatest

convexity in umbonal region with posterolateral corners slightly flattened.

Ventral interior with stout teeth not supported by dental plates and short

elevated pedicle collar. Muscle scar subrectangular; adductor scar occupies about

one-third of total width of scar. Dorsal interior with strong high socket ridges

bounding wide shallow sockets. Fulcral plates thick and extended laterally into

broad shelves. Cardinal process transversely elliptical, about one-tenth as wide as

valve, often roughened and pitted posteriorly; extends over posterior ends of

socket ridges. Outer hinge plates narrow, concave, indistinct. Loop preserved

only in two specimens. In the large specimen the crura are short rounded

extensions of the socket ridges and merge, at about midloop, with descending

lamellae without development of crural processes. In the smaller specimen the

crura are curved and flattened and united with descending lamellae anterior of

midloop with development of short, blunt crural processes. In both specimens,

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 107

Fig. 4. Dyscolia cf. johannisdavisi (Alcock, 1894), SM 174. A. SAM-—A25447: dorsal view of

conjoined valves. B-~D. SAM-—A25448: dorsal, lateral and anterior views of conjoined valves.

E-G. SAM~A25449: dorsal, lateral and anterior views of conjoined valves with pedicle and

serpulid worm tubes. H-J. SAM-—A25453: anterior, dorsal and lateral views of conjoined valves.

All X1.

108 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 5. Dyscolia cf. johannisdavisi (Alcock, 1894), SM 174. A-C. SAM-A25449: ventral,

anterior and lateral views of loop. D-F. SAM-—A25453: ventral, anterior and lateral views of

loop. G. SAM-—A25453: interior view of pedicle valve. H. SAM-—A25447: interior view of

pedicle valve showing muscle scars. A-F X2. G-—-H x1.

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES

109

descending lamellae and transverse bands are broad with the latter bearing low

median folds that are extended anteriorly as blunt points. Muscle field

subquadrate.

Dimensions (mm)

SAM-A25447

A25448

A25449

A25450

A25451

A25452

A25453

Muscle field dimensions (mm)

SAM-A25447

A25448

A25450

A25452

A25453

Length

41,8

41,3

41,1

34,9

35,8

Zoo

Brachial

valve

Width Thickness Length

350)

3/5

32,8

28,2

Soll

Ase

Pedicle valve

Length

16,7

2057

15,4

10,6

Loop dimensions (mm)

SAM-A25449

A25453

Discussion

Length

10,5

6,4

Width

9,0

ILE

8,1

Width

4.9

26,6 39,6

27,8 38,8

24,9 33)

= 33,4

11,0 28,6

Brachial valve

Length Width

Dye) 25)

Ll WL)

fle fle

Apical

angle

50%

84°

She

62°

The small sample of Dyscolia described here from a single locality shows

considerable morphological variation, which makes its assignment to one of the

known species of Dyscolia difficult. The specimens show differences and

similarities with most of the described species, e.g. the large size and triangular

outline are reminiscent of D. wyvillei (Davidson) and D. johannisdavisi

(Alcock), while the subcircular outline of the smallest specimen is more like those

of D.? radiata Cooper and D. ewingi Cooper. However, D. ewingi has now been

removed from Dyscolia to become the type-species of the new genus Gonio-

brochus (Cooper 1983: 261). The broad transverse band of the present specimens

differs from those of D. wyvillei and D. johannisdavisi, which are said to be thin

110 ANNALS OF THE SOUTH AFRICAN MUSEUM

and delicate (Cooper 1973c: 19; 1983: 254), but it is like that of the fossil

D. guiscardiana (Seguenza) from the Pliocene of Sicily. Dyscolia johannisdavisi

and D.? radiata are both recorded from the Indian Ocean, but the latter differs in

the form of its loop.

The author has examined the holotype of D. wyvillei and a brachial valve of

D. johannisdavisi in the British Museum (Natural History). These two forms are

closely similar and Muir-Wood (1959: 301) reports that although some authors

would wish to synonymize them (e.g. Thomson 1927: 201), there are differences

that are probably sufficient to separate the species. As pointed out by Muir-Wood

(1959), D. wyvillei has a longer, more narrowly tapering ventral umbo, a longer,

better-exposed symphytium and more strongly developed radial ornament.

Dyscolia johannisdavisi has a more marked marginal flange, a short concave

symphytium that is usually hidden by the dorsal umbo and only faintly developed

radial ornament. Indeed, in Alcock’s original description (1894: 139) no mention

is made of any radial ornament. Helmcke (1940: 267) regards the hinge plates of

the two forms to be different; those of D. wyvillei are large, wide, almost square,

while those of D. johannisdavisi are smail, narrow and barely developed. The

holotype of D. wyvillei has a row of denticles around the interior margin of both

valves. No such feature has been described from D. johannisdavisi nor are any

denticles visible on any of the South African specimens.

Of the two species, the South African shells described here are undoubtedly

closer to D. johannisdavisi, the main difference being in the form of the loop.

Cooper (1983) attaches considerable taxonomic importance to the form of the

loop, using it to distinguish between species and genera of terebratulaceans.

However, in view of the amount of variation in other characters of the shell

displayed by the Meiring Naude sample, it is felt that similar variation may be

seen in the form of the loop, and erection of a new taxon is unwarranted at this

stage.

Subfamily Aenigmathyridinae Cooper, 1983

Genus Xenobrochus Cooper, 1981

The classification and identification of terebratulacean brachiopods depends,

to a large extent, on the form of the loop and its relative proportions. Cooper

(1983) has established a number of new families, subfamilies, genera and species

based largely on this character of the shell. However, placing such taxonomic

importance on a single feature raises several problems, especially when what

appear to be minor differences are used to distinguish between genera. Most

species of Recent Terebratulacea, particularly those from fairly deep water, are

established on only a few specimens at most. Thus, the limits of loop variation in a

single population or species are still virtually unknown, although for some species

such as Liothyrella neozelanica and, as has been shown, Dyscolia cf. johannis-

davisi it is obvious that considerable variation may be present.

Further problems are encountered with the identification of juveniles of

short-looped genera when only the adult loop has been described, and specimens

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES HOU

in which the loop has not been completely preserved are especially difficult to

classify. Two such forms are present in the Meiring Naude material. However,

other features of the shells, such as their small to medium size, lack of radial

ornament, rectimarginate anterior commissures, narrow crura and poorly defined

outer hinge plates, suggest they belong to the subfamily Aenigmathyridinae

Cooper, 1983. Recent members of this subfamily include Abyssothyris, Acrobele-

sia and Xenobrochus, and of these Xenobrochus probably best accommodates the

specimens described below and it is to this genus that they are tentatively

assigned.

Xenobrochus africanus (Cooper, 1973)

Fig. 6

Gryphus africanus Cooper 1973b: 8, pl. 4 (figs 31-38).

Xenobrochus africanus (Cooper) Cooper, 1981: 20, pl. 4 (figs 30-35).

Material

Thirteen complete specimens and a single pedicle valve (SAM—A25454)

from SM 103 at a depth of 680 m; one complete specimen (SAM—A25455) from

SM 129 at a depth of 850m; a single brachial and three pedicle valves

(SAM—A25456) from SM 131 at a depth of 780 m; a damaged pedicle valve

(SAM-—A25457) from SM 162 at a depth of 630m. Most of the complete

specimens were live at the time of collection, some attached to small pebbles by

short slender pedicles.

Description

Small biconvex shells with elongately oval outline. Anterior commissure

rectimarginate; beak small, suberect; foramen mesothyridid; deltidial plates

conjunct forming completely visible symphytium. Shell surface smooth except for

concentric growth lines.

Pedicle valve quite strongly and evenly convex in lateral profile; anterior

profile strongly convex with evenly convex middle portion flanked by short steep

sides. Brachial valve subcircular with evenly and gently convex lateral profile;

anterior profile strongly convex with narrow median portion bounded by flatter

sides.

Ventral interior with strong teeth not supported by dental plates or thickened

valve wall; pedicle collar short, excavate anteriorly. Details of muscle scars not

visible. Dorsal interior with high socket ridges bounding deep sockets; outer

hinge plates narrow, concave and tapering anteriorly to join thin crural bases.

Crura narrow, bearing scoop-like anterior part of loop without development of

descending lamellae. Crural processes low and blunt and attached directly to

narrow transverse band, which is convex dorsally and anteriorly.

112 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 6. Xenobrochus africanus (Cooper, 1973). A-D. SAM-—A25454, SM 103: dorsal, lateral,

ventral and anterior views of conjoined valves with pedicle. E. SAM—A25454, SM 103: interior

view of pedicle valve. F. SAM-—A25455, SM 129: dorsal view of conjoined valves.

G. SAM-A25454, SM 103: interior view of brachial valve. H. SAM—A25454, SM 103: ventral

view of loop. A-E x3. F-—G x6. H x8.

Dimensions (mm)

Length Width Thickness

SAM-A25454 3,8 355 1,8

6,6 pl 3,4

A25455 6,1 Se 3,0

A25456 6,6 J3) ——

A25457 — S52 —

Loop dimensions (mm)

Length Width |

SAM-A25454 iS 0,7 |

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 113

Discussion

This genus is characterized by its small size, rectimarginate anterior

commissure and a loop that has a transverse band convex toward the anterior

(Cooper 1983: 275). Of the species assigned to the genus, X. indianensis

(Cooper), X. australis Cooper, X. anomalus Cooper and X. africanus (Cooper)

bear the closest resemblance to the present specimens. Of these, X. africanus is

recorded from the same area as these specimens but from shallower depths

(Cooper 19736: 3) and is so similar in morphology that the two forms must be

regarded as being conspecific. The other species can be distinguished principally

by their larger size but X. indianensis has a more incurved beak with partly

concealed symphytium, fairly large cardinal process and narrow median fold on

the transverse band of the loop. Xenobrochus australis is narrower and has an

angular transverse band; X. anomalus is distinguished by its tubular pedicle

collar.

Xenobrochus agulhasensis (Helmcke, 1938)

Riga

Terebratula vitrea var. minor Philippi: Davidson, 1880: 29, pl. 2 (figs 5-6).

Liothyrina (Gryphus) sp. Blochmann, 1908: 613, pl. 39 (fig. 31).

Liothyrina agulhasensis Helmcke, 1938: 243; 1940: 258, fig. 21.

Material

A single complete dead specimen (SAM-—A25458) from SM 232, at a depth

of 560-620 m.

Description

Elongately oval, biconvex shells with rectimarginate anterior commissure;

beak small, suberect with submesothyridid subcircular foramen. Deltidial plates

conjunct forming short symphytium. Shell surface smooth except for concentric

growth lines.

Pedicle valve quite strongly convex in both profiles. Brachial valve

subcircular, over nine-tenths as long as pedicle valve, with gently convex

profiles.

Ventral interior with small teeth not supported by dental plates; pedicle

collar very short, excavate. Details of muscle scars not visible. Dorsal interior

with quite low straight socket ridges bounding fairly wide sockets. Cardinal

process transversely elliptical, covering posterior ends of socket ridges. Outer

hinge plates narrow, triangular and tapering anteriorly to merge with crural

bases. Crura strong, rounded in section, bearing anterior part of loop, which is

convex anterodorsally. Crural processes small blunt points in midloop position

and attached directly to broad transverse band, which bears faint ventrally

directed median fold.

114 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 7. Xenobrochus agulhasensis (Helmcke, 1938), SAM—A25458, SM 232. A. Interior view of

brachial valve. B. Close-up of ventral beak, dorsal view. CC. Interior view of pedicle valve.

D-E. Ventral and lateral views of loop. A x3. B, D-E x8. C x2.

Dimensions (mm)

Length Width Thickness

SAM-A25458 ey 29 6,7

Loop dimensions (mm)

Length Width

SAM-A25458 Bei 2

Discussion

This specimen bears a close resemblance to Liothyrina agulhasensis Helmcke

from the Agulhas Bank. The species was first described by Davidson (1880) after

two specimens were dredged off the Cape of Good Hope during the Challenger

Expedition. Davidson identified the specimens as Terebratula vitrea var. minor

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 115

Philippi, which is synonymous with Liothyrella affinis Calcara. The author has

examined one of Davidson’s specimens in the British Museum (Natural History)

and found it to be very similar to that described here, particular in the form of the

loop.

Further specimens from the Agulhas Bank were recovered by the Valdivia

Expedition in 1898. Blochmann (1906, 1908) compared these to Davidson’s

specimens and concluded that both were separable from L. affinis on the grounds

that their loops were different. He assigned the specimens to Liothyrina

(Gryphus) but left the species unnamed. The loop figured by Blochmann (1908,

pl. 39 (fig. 31)) is very similar to that of the specimen described here.

Helmcke (1940: 258) revised Blochmann’s descriptions and again empha-

sized the differences between the loops of L. agulhasensis and L. affinis, that of

the former being more rounded while the latter had an angular transition between

the transverse band and the sides of the loop. Helmcke described the transverse

band of L. agulhasensis as being like a weakly fallen arch (? i.e. concave

ventrally).

The present author follows Cooper (1981: 20, 1983: 275) in assigning the

species to Xenobrochus because the transverse band of the loop is convex dorsally

and anteriorly, a distinguishing characteristic of the genus.

Jackson (1952: 17) mentions the possibility that the ‘Challenger’ and

‘Valdivia’ specimens might belong to his Gryphus capensis but in his description

of the loop of this species he states that the transverse band is narrow and arched

ventrally. This would immediately separate G. capensis from any of the

specimens mentioned here and indeed exclude it from Xeuobrochus. Also the

thread-like median septum separating adductor scars in the brachial valve of

G. capensis is a feature not seen in any species of Xenobrochus.

It is interesting to note that Cooper (1983: 275) has now questionably

assigned Liothyrella affinis to Xenobrochus.

Xenobrochus? sp. 1

Fig. 8

Material

One complete dead specimen (SAM-A25459) from SM 131 at a depth of

780 m.

Description

Small, elongately oval biconvex shell with maximum width about midvalve;

anterior commissure rectimarginate. Beak moderately long, narrowly rounded,

suberect to erect; foramen large, mesothyridid. Deltidial plates disjunct or

conjunct; symphytium short, visible. Shell surface smooth but for concentric

growth lines.

Pedicle valve deeper and more convex than brachial valve, evenly convex in

lateral profile, strongly convex in anterior profile, especially in median portion.

116 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 8. Xenobrochus? sp. 1, SAM—A25459, SM 131. A-D. Dorsal, ventral, lateral and anterior

views of conjoined valves. E. Interior view of pedicle valve. F. Interior view of brachial valve.

G. Lateral view of brachial valve showing crus. A—D X3. E-G x6.

Brachial valve moderately convex in lateral profile, more strongly convex in

anterior profile.

Ventral interior with strong teeth; pedicle collar excavate anteriorly. Details

of muscle scars not discernible. Dorsal interior with broad, transversely elliptical

cardinal process extending across posterior ends of socket ridges; socket ridges

high posteriorly but lower at their anterior ends, bounding narrow sockets.

Outer hinge plates flattish, tapering anteriorly and merging with crural bases.

Crura thin, subparallel; crural processes very small. Remainder of loop not

preserved.

Dimensions (mm)

Length Width Thickness

SAM-A25459 6,3 4,3 44

Discussion

This specimen bears a strong resemblance to Xenobrochus anomalus Cooper

from the waters around Marion Island. However, that species is characterized by

its extravagantly developed tubular pedicle collar and its crural processes are

situated at the anterior limit of the outer hinge plates (Cooper 1981: 20). While

the present specimen has a well-defined pedicle collar, it could hardly be

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES ER

described as extravagant and the crural processes are located anterior of the outer

hinge plates.

Xenobrochus? sp. 2

Fig. 9

Material

Six pedicle and three brachial valves (SAM-—A25460) from SM 232 at a depth

of 560-620 m and a pedicle and brachial valve (SAM-—A25461) from SM 226 at a

depth of 710-775 m.

Description

Small to medium elongately oval shells with maximum width at or slightly

anterior of midvalve; anterior commissure rectimarginate. Beak moderately long,

quite narrowly rounded, suberect to erect; foramen quite large, mesothyridid;

symphytium short, visible. Shell surface smooth except for concentric growth

lines.

Pedicle valve deeper and more convex than brachial valve; moderately

convex in lateral profile with maximum convexity in umbonal region; strongly

convex in anterior profile. Brachial valve gently and evenly convex in both

profiles.

Ventral interior with small elongate teeth; pedicle collar short, excavate

anteriorly. Muscle scars not visible. Dorsal interior with transversely elliptical

cardinal process; high socket ridges bounding fairly wide sockets; outer hinge

plates small, merging with crural bases. Crura thin; crural processes small, blunt.

Remainder of loop not preserved. Lightly impressed elongate rectangular muscle

scars, extending for about one-third of the valve length, barely discernible

beneath loop.

Dimensions (mm)

Length Width

SAM-A25460 ~—:114,7 ici

Sai 10,5

19 2.8)

IES SN

SAM-A25461 te 7 8,4

Discussion

These specimens resemble Xenobrochus indianensis (Cooper) from the

north-western part of the Indian Ocean, although without a complete loop further

comparison is impossible.

118 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 9. Xenobrochus? sp. 2. A-E. SAM-—A25461, SM 226. A. Ventral view of pedicle valve.

B. Interior view of pedicle valve. C. Close-up of ventral beak. D. Interior view of brachial valve.

E. Close-up of cardinalia. F-J. SAM—A25460, SM 232. F. Ventral view of pedicle valve.

G. Interior view of pedicle valve. H. Close-up of ventral beak. I. Dorsal view of

brachial valve. J. Close-up of cardinalia. A, F, G,I x2. B, D X3. C,E,H,J X8.

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 119

Superfamily CANCELLOTHYRIDACEA Thomson, 1926

Family Cancellothyrididae Thomson, 1926

Subfamily Cancellothyridinae Thomson, 1926

Genus Terebratulina d’Orbigny, 1847

Terebratulina sp.

Fig. 10

Material

A total of 13 immature specimens: 8 complete specimens, 3 brachial valves

and a pedicle valve (SAM-—A25462) from SM 131 at a depth of 780 m, and a

single complete specimen (SAM-—A25463) from SM 180 at a depth of 80 m. All

were dead at time of recovery.

Description

Very small biconvex shells with elongately oval to roundedly triangular

outline; maximum width in anterior third of shell. Beak nearly straight with large

submesothyridid foramen; deltidial plates very small. Anterior commissure

rectimarginate. Ornament consisting of rounded costae and costellae and strong

concentric growth lines. Many of the specimens have strongly beaded appearance

caused by interference of radial and concentric ornaments. Posterolateral

extremities with concentric ornament only.

Pedicle valve moderately convex in lateral profile; gently convex in anterior

profile but with steep posterolateral slopes. Brachial valve gently convex in both

profiles, but more arched umbonally. Posterolateral corners flattened.

Ventral interior with small hook-like teeth; pedicle collar short, excavate.

Other details obscure. Dorsal interior with high, strong, widely divergent socket

ridges fused posteriorly to weakly developed cardinal process; sockets short and

deep. Crura short, stout, rounded in cross-section. Remainder of loop not

preserved in any of these specimens.

Dimensions (mm)

Length Width Thickness

SAM-—A25462 3), Dd) 1,8

3 De} 1,5

2,6 Zell 12

2.3 1,9 Iho

SAM-A25463 4,2 3,5) 2.0)

Discussion

Terebratulina is very widespread in the world’s oceans and is represented in

South African waters by T. abyssicola (Adams & Reeve) and T. meridionalis

120 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 10. Terebratulina sp. A-B. SAM-—A25463, SM 180: dorsal and ventral views of conjoined

valves. C—P. SAM—A25462, SM 131. C. Dorsal view of conjoined valves. D. Dorsal view of

conjoined valves. E-—F. Dorsal and interior views of brachial valve. G—H. Ventral and interior

views of pedicle valve. I-J. Dorsal and interior v-ews of brachial valve. K-—L. Dorsal

and interior views of brachial valve. M-—N. Lateral and anterior views of conjoined valves.

O-P. Lateral and anterior views of conjoined valves. All x8.

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 121

Jackson, as well as unnamed species recorded by Jackson (1952) and Cooper

(1973b). The present specimens cannot be assigned to either of the above-named

species because of their much stronger radial ornament. The beaded appearance

of the ribs displayed by most of the present specimens is typical of many immature

Terebratulina (Cooper 1978: 7) so until associated adult shells become available a

more precise identification cannot be attempted. More than one species may be

present in this sample.

Family Chlidonophoridae Muir-Wood, 1959

Subfamily Chlidonophorinae Muir-Wood, 1959

Genus Chlidonophora Dall, 1903

Chlidonophora chuni Blochmann, 1903

Fig. 11

Terebratula sp.: Alcock, 1894: 139.

Terebratulina sp.: Chun, 1900: 404, 405, 2 figs.

Chlidonophora chuni Blochmann in Chun, 1903: 435, 436, 2 figs. Blochmann, 1906: 695.

Thomson, 1927: 182. Helmcke, 1940: 239, fig. 6. Muir-Wood, 1959: 296, pl. 4 (figs 5-7).

Cooper, 1973b: 13, pl. 8 (figs 17-26).

Material

Two complete specimens (SAM—A25464), dead at time of recovery, from

SM 103 at a depth of 680 m; three complete specimens (SAM-—A25465), one

dead and two live at time of recovery, from SM 246 at a depth of 1 640-1 660 m.

Description

Small subcircular biconvex shells with maximum width at about midvalve.

Hinge line nearly straight, less than one-half as wide as valve. Anterior

commissure broadly and very gently uniplicate. Beak short; interarea small;

foramen hypothyridid; deltidial plates disjunct but one specimen with pedicle

shows delthyrium closed anterior to pedicle by ?symphytium. Pedicle quite short,

slender and frayed at end. Ornamentation of rounded costae and costellae

crossed by strongly developed concentric growth lamellae to produce a reticulate

pattern.

Pedicle valve gently convex in lateral profile; maximum convexity in

umbonal region becoming flatter anteriorly. Anterior profile quite strongly

arched with median flattening due to development of broad shallow sulcus

Originating near umbo. Brachial valve evenly convex in lateral profile; anterior

profile arched by incipiently developed fold.

Pedicle valve interior with strong teeth but no dental plates; other details

obscure. Brachial valve interior with short socket ridges fused posteriorly to

transverse cardinal process; cardinalia project posterior to hinge line. Crura

short, stout, converge anteromedianly, crural processes sharply pointed with

points directed anteroventrally. Narrow transverse band with ventrally directed

median fold.

122 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 11. Chlidonophora chuni Blochman, 1903, SAM-—A25465, SM 246. A-B. Dorsal and

lateral views of conjoined valves with pedicle. C. Anterior view of conjoined valves.

D-E. Ventral and interior views of pedicle valve. F—H. Ventral, lateral and anterior views of

brachial valve interior showing loop. All x6.

Dimensions (mm)

Length Width Thickness

SAM-A25464 4.7 41 Dod

4.4 41 1.9)

SAM-A25465 4.7 4.5 al

41 Sy) lig)

4.5 4.2 Joa

Loop dimensions (mm)

Length Width

SAM-A25465 eo 0,9

Discussion

Two species of Chlidonophora are known; C. incerta (Davidson) from the

Atlantic Ocean can be distinguished from C. chuni Blochmann from the Indian

Ocean by the form of its pedicle, its loop and its wider hinge line. The pedicle of

C. chuni is long and slender and frays some distance from the umbo whereas that

of C. incerta frays immediately on emergence from the foramen to give a

radiating effect. The loop of C. chuni is longer and more pointed than that of

C. incerta, which is short and rounded. The Meiring Naude specimens have the

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 123

narrow hinge line and longer pointed loop of C. chuni and although the pedicle is

not long it does not fray until a short distance from the umbo. They must

therefore be placed in that species.

Genus Notozyga Cooper, 1977

Notozyga gracilis sp. nov.

Fig. 12

Diagnosis

Notozyga with subdued radial ornament and delicate, slender loop.

Material

Holotype. SAM—A25466 in the South African Museum, Cape Town. From

SM 131 (30°43,2’S 30°40,8’E) at a depth of 780 m, 11 May 1977. One complete,

dead specimen.

Description

Small biconvex shell with rounded subpentagonal outline; posterolateral

angles obtuse. Hinge line straight, about two-thirds as wide as valve; maximum

width slightly anterior of midvalve. Anterior commissure rectimarginate.

Interarea quite high, about one-third as long as valve; foramen quite large,

triangular, flanked by narrow deltidial plates. Well-defined triangular palintropes

between deltidial plates and beak ridges. Ornament of very subdued rounded

costae and costellae and very faint concentric growth lines.

Fig. 12. Notozyga gracilis sp. nov., SAM—A25466, holotype, SM 131. A-—D. Dorsal, lateral,

anterior and ventral views of conjoined valves. E. Interior view of pedicle valve. F—H. Ventral,

anterior and lateral views of brachial valve interior showing loop. All x8.

124 ANNALS OF THE SOUTH AFRICAN MUSEUM

Pedicle valve lateral profile strongly convex near umbo, becoming less so

anteriorly; anterior profile strongly convex. Brachial valve quite strongly convex

in both profiles; posterolateral corners flattened.

Ventral interior with short, excavate pedicle collar; teeth strong, semicircular

in shape, without dental plates. Dorsal interior with fairly high, widely divergent

socket ridges, fused posteriorly and bounding narrow, deep sockets. Crura short,

strong, projecting anteromedianly from below anterior ends of socket ridges.

Crural processes moderately high, bluntly pointed. Descending lamellae very

thin, curved, uniting medianly without development of true transverse band.

Both valves with flattened rim around anterior and lateral margins with well-

developed eminences and embayments inside this.

Dimensions (mm)

Length Width Thickness

SAM-A25466 Holotype Das Bolt 2

Loop dimensions (mm)

Length Width

SAM-A25466 Holotype 0,8 0,7

Discussion

The genus Notozyga is much like Eucalathis in appearance but can be

distinguished principally by its well-defined ventral interarea and the rounded, as

opposed to pointed, form of its loop, which extends only slightly beyond the

crural processes (Cooper 1977: 105). The specimen described here possesses

these distinctive features and is thus placed in that genus but it differs from the

only previously described species, N. lowenstami Cooper, in its subdued

ornament and finer, more delicate loop.

Eucalathis macrorhynchus Foster from the Pacific—Antarctic ridge is a form

with a subdued radial ornament and well-defined interarea and in these respects

resembles N. gracilis. The loops of these two species are also similar but in his

description Foster (1974: 81) points out that E. macrorhynchus shows consider-

able variation in the form of its loop from rounded Notozyga-type to a more

pointed shape typical of Eucalathis.

Another species worthy of investigation is Eucalathis trigona (Jeffreys),

which Cooper (19735: 13) reports as having a loop more rounded anteriorly than

is usual for the genus, and he states that Dall (1920: 324) questions the generic

affinities of E. trigona. This might also prove to belong to Notozyga.

Etymology

From the Latin gracilis meaning slender, referring to the form of the

loop.

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 125

Notozyga sp.

Fics

Material

A single complete, dead specimen (SAM-—A25467) from SM 131 at a depth

of 780 m.

Description

Small biconvex shell, subpentagonal to elongately oval in outline; maximum

width at about two-thirds valve length. Anterior commissure rectimarginate.

Hinge line nearly straight, about seven-tenths as wide as valve. Interarea quite

high, over one-third as long as valve; foramen large, triangular, flanked by

narrow deltidial plates. Fairly broad, triangular palintropes between deltidial

plates and beak ridges. Ornament consists of strong rounded costae, which

become broader anteriorly and increase by branching. Posterolateral corners of

brachial valve devoid of radial ornament. Concentric ornament of faint growth

lines.

Fig. 13. Notozyga sp., SAM—A25467, SM 131. A-B. Ventral and interior views of pedicle

valve. C—D. Dorsal and interior views of brachial valve. All <8.

Pedicle valve gently convex in lateral profile; quite strongly convex in

anterior profile. Brachial valve evenly convex in both profiles.

Ventral interior with short, elevated pedicle collar and small teeth. Other

details not seen. Dorsal interior with short, almost colinear socket ridges

bounding small sockets. Stout crura, slightly flattened, projecting anteromedianly

from valve wall in front of socket ridges. Remainder of loop not preserved.

Dimensions (mm)

Length Width

SAM-A25467 2,4 159

Discussion

This little specimen is placed in Notozyga on the basis of its well-defined

ventral interarea, but it differs from N. gracilis sp. nov. in its more elongate

outline and much stronger ribbing. However, more material with complete

brachidia is required before another new species can be erected.

126 ANNALS OF THE SOUTH AFRICAN MUSEUM

Subfamily Eucalathinae Muir-Wood, 1965

Genus Eucalathis Fischer & Oehlert, 1890

Eucalathis fasciculata Cooper, 1973

Fig. 14

Eucalathis fasciculata Cooper, 19736: 12, pl. 1 (figs 26-28).

Material

Four complete specimens (SAM-—A25468), live, from SM 60 at a depth of

800-810 m; one complete specimen (SAM-—A25469), dead, from SM 103 at a

depth of 680 m; seven complete, live specimens plus two brachial and two pedicle

valves (SAM—A25470) from SM 246 at a depth of 1 640-1 660 m.

Description

Small, broadly triangular biconvex shells with narrow hinge and rounded

anterior margin; posterolateral angles obtuse; maximum width, about twice hinge

width, situated at around two-thirds valve length. Anterior commissure faintly

uniplicate. Interarea very narrow; foramen wide, triangular; deltidial plates not

always present but several specimens show development of symphytium. Pedicle

relatively long, slender and frayed at end. Ornament consists of faint concentric

growth lines and about 10 subangular primary costae, each of which gives rise to

one or two costella to produce coarse fascicostellate ornament. Posterior portions

of shell nearly smooth although faint traces of ribs may be seen.

Pedicle valve gently convex in lateral profile; anterior profile quite strongly

convex in umbonal region but anterior of midvalve a weakly developed sulcus

flattens profile. Brachial valve gently convex in lateral profile; inconspicuous fold

originates near midvalve.

Pedicle valve interior with small teeth; pedicle collar developed; other details

obscure. Brachial valve interior with stout socket ridges united posteriorly with

wide, weakly developed cardinal process. Crura strong with blunt crural

processes; loop short, triangular, with broad descending lamellae, which

converge anteromedianly, and joined by narrow transverse band with small

ventrally directed median fold.

Dimensions (mm)

Length Width Thickness

SAM-—A25468 35 3,0 =

aril 2,8 i>

A25469 238 259 2

A25470 ree) 2,4 1,4

2,4 Pipi 3

Loop dimensions (mm)

Length Width

SAM-A25468 ied OF,

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 127

Fig. 14. Eucalathis fasciculata Cooper, 1973. A-D. SAM-A25470, SM 246: dorsal, lateral,

anterior and ventral views of conjoined valves. E. SAM—A25469, SM 103: interior view of

pedicle valve. F—H. SAM—A25470, SM 246: ventral, anterior and lateral views of brachial valve

interior showing loop. I-J. SAM—A25468, SM 60. I. Group of three specimens with pedicles

attached to foraminifera. J. Complete specimen showing long frayed pedicle.

A-H, J x8. I x6.

Discussions

These specimens can immediately be referred to Eucalathis fasciculata

Cooper from south of Madagascar. It is distinguished from the other known

species of the genus by its small size and distinctive ornament. This is the only

species in which the posterior part of the shell is more or less smooth. The other

Indian Ocean species, all from south of Madagascar, are E. costellata Cooper,

E.. rotundata Cooper, and an unnamed species recorded by Cooper (1981: 19).

All of these forms have a strongly developed radial ornament.

128 ANNALS OF THE SOUTH AFRICAN MUSEUM

Subfamily Agulhasiinae Muir-Wood, 1965

Genus Agulhasia King, 1871

Agulhasia davidsoni King, 1871

Fig. 15

Agulhasia davidsoni King, 1871: 111, pl. 11 (figs 1-7). Thomson, 1927: 182, fig. 52. Helmcke,

1940: 242, fig. 7. Jackson, 1952: 9. Cooper, 1973b: 14, pl. 4 (figs 1-14); 1973c: 15, pl. 8

(figs 18-24).

Terebratulina (Agulhasia) davidsoni King: Davidson, 1886: 36, pl. 7 (figs 1-5).

Material

Nine complete specimens (SAM-—A25471) from SM 131 at a depth of 780 m.

All appear to have been dead and sediment-filled at time of collection. Some

show signs of abrasion suggesting transportation from living site.

Description

Small biconvex shells with broadly triangular outline; maximum width near

anterior margin. Anterior commissure gently uniplicate. Ventral beak greatly

elongate, about one-third as long as shell, narrowly pointed; delthyrium bounded

by pair of disjunct deltidial plates and almost closed by long, triangular, concave

apical plate, which restricts foramen to a small opening at anterior end of beak.

Radial ornament consists of rounded costae and costellae; most specimens had

6 (pedicle valve) or 7 (brachial valve) primary costae extending from umbo,

secondary ribs arise mostly by intercalation between primaries from about 1 mm

growth stage to give total of 11 or 12 mbs by 2 mm growth stage. Largest

specimens with 18-22 ribs at anterior margin. Radial ornament absent from

lateral margins of both valves. Concentric ornament of closely spaced growth

lines.

Pedicle valve elongately triangular; lateral profile evenly convex; anterior

profile with median flattening and development of broad shallow sulcus in later

growth stages. Brachial valve elongately subpentagonal; lateral profile with

maximum convexity near umbo; anterior profile arched along midline but true

fold not developed.

Fig. 15. Agulhasia davidsoni King, 1871, SAM-—A25471, SM 131. A-D. Dorsal, lateral,

anterior and ventral views of conjoined valves. All x8.

THE SOUTH AFRICAN MUSEUM’S MEJRING NAUDE CRUISES 129

Details of internal characteristics of valves not available from present

specimens, but see Cooper (1973b: 14).

Dimensions (mm)

Length Width

SAM-—A25471 4,5 2,9

4,0 DAG

3,4 D3

28) eS

2,4 1,7

1,6 1,0

DD 1,4

2,9 1,9

1,8 ti

Discussion

Agulhasia davidsoni King is characterized by its small size, broadly triangular

shape and especially by its elongated ventral beak. It is unlikely to be mistaken

for any other species. It is known only from South African waters.

Suborder TEREBRATELLIDINA Muir-Wood, 1955

Superfamily TEREBRATELLACEA King, 1850

Family Kraussinidae Dall, 1870

Genus Kraussina Davidson, 1859

Kraussina rubra (Pallas, 1766)

Fig. 16

Anomia rubra Pallas, 1766: 182, pl. 14 (figs 2-11).

Terebratula capensis Kister (non Adams & Reeve), 1848: 32, pl. 3 (figs 15, 17).

Terebratula capensis Krauss (non Adams & Reeve), 1848b: 32, pl. 2 (fig. 10).

Terebratula (Kraussia) rubra (Pallas): Reeve, 1861: 9, fig. 37.

Kraussina rubra (Pallas): Davidson, 1887: 119, pl. 20 (figs 19-23). Jackson, 1952: 22, pl. 3

(figs 1-2). Cooper, 1973c: 23, fig. 5, pl. 9 (figs 10-22).

Material

A broken pedicle valve (SAM-—A25472) from SM 129 at a depth of 850 m; a

complete specimen and a pedicle valve (SAM-A25473) from SM 131 at a depth

of 780 m; and a complete specimen (SAM—A25474) collected live from SM 185 at

a depth of 90 m. Also included in this description are two specimens separated

from coral material collected during earlier research expeditions: a complete

130 ANNALS OF THE SOUTH AFRICAN MUSEUM

immature specimen (SAM—A25484) with coral SAM—H1366 dredged 28 Decem-

ber 1898 by S.S. Pieter Faure at station PF 796 (33°4'S 27°57'E) at a depth of

59m; and a complete specimen (SAM-—A25483) collected live at Somerset

Strand, 23 October 1897, with coral SAM—H1406.

Description

Biconvex shells with width usually slightly less than length; outline variable

but usually elongately oval. Hinge line nearly straight, about seven-tenths as wide

as valve. Anterior commissure gently sulcate. Beak suberect, irregularly

truncated with large incomplete submesothyridid foramen bounded by small

triangular deltidial plates. Narrow triangular palintropes between deltidial plates

and beak ridges. Ornament consists of concentric growth lines and strong

rounded costae, which increase by branching and intercalation; 18—22 ribs on

pedicle valve at 5 mm growth stage.

Fig. 16. Kraussina rubra (Pallas, 1766). A-E. SAM—A25483, Somerset Strand. A-B. Ventral

and interior views of pedicle valve. C—D. Dorsal and interior views of brachial valve. E. Lateral

view of brachidium. F—-I. SAM—A25474, SM 185: dorsal, ventral, lateral and anterior views of

conjoined valves with pedicle. A-E x2. F-I x4.

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 131

Pedicle valve gently convex in lateral profile but broadly carinate in anterior

profile. Brachial valve gently convex in lateral profile, anterior profile with broad

shallow sulcus extending from near umbo, separating gently convex flanks.

Pedicle valve interior with small teeth without dental plates; pedicle collar

short, sessile but may be slightly excavate anteriorly. Other details obscure.

Brachial valve interior with low, widely divergent socket ridges bounding narrow

sockets. Cardinal process transversely elliptical, between posterior ends of socket

ridges. Notothyrial platform consists of a pair of suboval thickenings between

socket ridges and posterior end of median septum; in middle of each, rounded

depressed attachment scars of pedicle muscles are situated. Low median septum

extends from in front of notothyrial platform and supports at its distal end a pair

of diverging lamellae, extending ventrally. Ventral end of each lamella drawn out

into short anteriorly directed prong; some specimens also bear posterior prongs

and some have short spine on top of each lamella. Valve floors of small specimens

show radial rows of coarse tubercules but these are not seen in adult shells.

Occasional small submarginal spines present in some specimens.

Dimensions (mm)

Length Width

SAM-—A25472 Ve Cn

A25473 3,6 30)

4,1 oe

A25474 Se 13

A25483 WZ 10,9

A25484 ed h3

Discussion

Kraussina is a fairly common genus in Cape coastal waters and is represented

by the species K. crassicostata Jackson, K. cognata (Sowerby) and K. rubra

(Pallas), which is perhaps the most common of all brachiopod species in this part

of the world. Kraussina crassicostata can immediately be separated from the

others by its smaller size, more convex valves and stronger, coarser ornament.

Kraussina cognata is poorly known and more material is required for its

relationship to K. rubra to be fully assessed. It is recorded from western Cape

waters whereas K. rubra, while known from the western Cape coast, is more

commonly found on the east coast. From previous descriptions, e.g. Jackson

(1952: 24), it seems that the principal difference between K. cognata and

K. rubra is one of colour; the former is reported to be pale yellow whereas the

latter is noted for its red and white markings although the author has seen pure

white shells in a sample of otherwise coloured specimens. Colour alone is a most

unsatisfactory criterion on which to separate species.

Of the present specimens, the very small shells are white but one has reddish

markings around its anterior margin and the specimen separated from coral

SAM-H1406 is yellowish. This last specimen may prove to be K. cognata but for

the present it is included with the others in K. rubra, which has such a variable

132 ANNALS OF THE SOUTH AFRICAN MUSEUM

shape that growth ratios are virtually useless in separating species. The species are

rather poorly defined and in need of revision so that the full range of morphological

variation can be taken into account when distinguishing between them.

Genus Megerlia King, 1850

Megerlia acrura sp. nov.

Fig. 17

Diagnosis

Megerlia similar in most respects to M. gigantea (Deshayes) but without the

development of crura.

Material

Holotype. SAM—A25475 in the South African Museum, Cape Town. From

SM 239 (32°14,8’S 29°00,8’E) at a depth of 90 m, 25 June 1979.

Paratypes. SAM-—A25476 in the South African Museum, Cape Town. From

SM 239 (32°14,8’S 29°00,8’E) at a depth of 90 m, 25 June 1979. SAM—A25477 in

the South African Museum, Cape Town. From SM 255 (31°37,8'S 29°40,8'E) at a

depthioml2>iim-s2seuinemO7 2:

All the specimens were live at time of collection; that from SM 255 is a

juvenile.

Description

Small subcircular to roundedly subrectangular shells with width slightly

greater than length. Anterior commissure broadly and very gently sulcate. Beak

broadly rounded; foramen hypothyridid, bounded by small, narrow, raised

deltidial plates. Interarea flat to gently curved.

Ornament on pedicle valve consists of concentric growth lines, faintly devel-

oped costellae, tubercules and small bristle-like spines; brachial valve ornament of

faint ribs and growth lines but no spines or tubercules. Pedicle short and thick.

Pedicle valve strongly and evenly convex in lateral profile; anterior profile

more arched medianly with flatter, sloping flanks. Brachial valve varying from

almost flat to gently convex especially in umbonal region; anterior profile evenly

convex.

Pedicle valve interior with short, slightly excavate pedicle collar and strong

teeth. Large subcircular pedicle muscle scars situated on either side of median

line about one-third valve length from posterior margin. Brachial valve interior

with high thickened socket ridges bounding relatively narrow sockets; no cardinal

process. Between socket ridges, thickened triangular platform passes anteriorly

into low median ridge, which extends to about midvalve and supports at its

anterior end a pair of anteroventrally directed plates, which form an open V-

shape when viewed from anterior. At about two-thirds of their height, posterior

margins of plates give rise to fine ribbon-like lamellae, which unite postero-

medianly to form complete ring. From about one-third of their height, anterior

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 133

Fig. 17. Megerlia acrura sp. nov. A-D. SAM-—A25475, holotype, SM 239: ventral, dorsal,

lateral and anterior views of conjoined valves with pedicle.

E-F. SAM-A25476, paratype, SM 239. E. Interior view of pedicle valve. F. Lateral view of

brachidium. G—-H. SAM-—A25475, holotype, SM 239: ventral and posterior views of brachial

valve interior showing brachidium. I-L. SAM-—A25477, paratype (juvenile), SM 255.

I. Interior view of pedicle valve. J-L. Ventral, lateral and posterior views of brachial valve

interior showing precampagiform stage brachidium. A-F x3. G-H x4. I-L x8.

134 ANNALS OF THE SOUTH AFRICAN MUSEUM

margins of plates give rise to another pair of ribbon-like lamellae, which extend

and converge posteriorly but do not meet. In none of the specimens is there any

sign of crural development. Diductor muscles attached to thickened notothyrial

platform between socket ridges; adductors located on valve floor anterior to this,

on either side of median ridge. Outside muscle field, valve floor covered in

coarse, radially disposed tubercules, which increase in size towards valve margin.

Dimensions (mm)

Length Width Thickness

SAM-A25475 Holotype 9,9 10,9 —

SAM-A25476 Paratype 8,1 9,8 4,0

SAM-—A25477 Paratype SEZ 3,4 il

Discussion

In most respects these specimens are very similar to Megerlia gigantea

(Deshayes), including the strong re-entrant at the dorsal beak suggesting an

amphithyridid foramen. The principal difference with this, and indeed other

species in the genus, is that none of the specimens described here shows any sign

of the development of crura. The juvenile specimen shows an early stage (pre-

campagiform) in loop development with a small cone, open anteriorly, supported

on a short pillar, similar to the situation described by Cooper (1981: 28) in

specimens of similar size. However, this specimen shows no sign of incipient crus

development as shown by Cooper’s specimens. Cooper (1981, pl. 6 (figs 9, 16,

22)) also figures adult specimens of similar size to the large Meiring Naude

specimens, in which the crura are fully developed and extend from in front of the

socket ridges to join the rest of the brachidium. No such development is seen in

the present specimens, which are thus distinguished by their lack of crura.

Etymology

The specific name refers to the lack of crura.

Genus Megerlina Deslongchamps, 1884

Megerlina pisum (Lamarck, 1819)

Fig. 18

Terebratula pisum Lamarck, 1819: 245.

Terebratula natalensis Krauss, 1844 (plates), pl. 2b (figs 4-7); 1848a (text): 36.

Material

One dead specimen (SAM—A25478) from SM 163 at a depth of 90 m; one

live specimen (SAM-—A25479) from SM 179 at a depth of 80m; one dead

specimen (SAM—A25480) from SM 180 at a depth of 80 m; and seven live and

four dead specimens plus a few fragments (SAM—A25481) from SM 185 at a

depth of 90 m. Also included in this description are two specimens separated

from coral material collected during an earlier research expedition: a complete

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 135

live specimen (SAM—A25485) with coral SAM—H1365 dredged 29 August 1901

by S.S. Pieter Faure at station PF 13601, 8km W by N of Great Fish Point

Lighthouse at a depth of c. 41 m; a complete live specimen (SAM-—A25486) with

coral SAM-—H1383 dredged 25 September 1901 by S.S. Pieter Faure at station

PF 13959, 4,8 km N by E % E of Bird Island Lighthouse at a depth of c. 66 m.

Description

Subpentagonal to subquadrate shells with length and width about equal;

maximum width about midvalve. Pedicle valve slightly deeper than brachial

valve. Anterior commissure sulcate. Hinge line nearly straight, about three-fifths

as wide as valve. Beak suberect with large incomplete submesothyridid foramen

bounded by small triangular deltidial plates. Narrow triangular palintropes

between deltidial plates and beak ridges. Ornamentation of rounded costellae,

which increase by branching and intercalation; 24-33, most commonly 27, ribs at

5 mm growth stage on brachial valve.

Pedicle valve gently convex in lateral profile; anterior profile broadly

carinate. Median fold narrow, extending from near umbo becoming broader and

more prominent anteriorly; lateral slopes flat to very gently convex. Brachial

valve over four-fifths as long as pedicle valve, gently convex in both profiles.

Narrow shallow median sulcus extends from close to umbo, becoming broader

and deeper anteriorly.

Pedicle valve interior with small teeth; pedicle collar very short, slightly

excavate anteriorly. Radially disposed tubercules cover anterior part of valve

floor; occasional spine present just inside anterior margin. Brachial valve interior

with strong socket ridges and fulcral plates bounding narrow deep sockets.

Notothyrial platform defined by buttresses joining anterior ends of socket ridges

to medium septum; cardinal process elliptical, weakly developed. Medium

septum extends from in front of cardinal process to just anterior of midvalve. At

its anterior end it supports a brachidium consisting of a pair of anteroventrally

directed diverging lamellae. Each lamella bears, at its distal end, a small prong

that projects posteromedianly, and midway down the outer surface of each

lamella is an accessory ledge (or process). Valve floor covered by radial rows of

coarse tubercules. Muscle scars lightly impressed on notothyrial platform and

valve floor immediately anterior to platform.

Dimensions (mm)

Length Width Thickness

SAM-A25478 8,3 8:3 Sip)

A25479 6,9 6,4 Perl

A25480 4,7 4.4 1,8

A25481 6,4 6,3 =

a 4.9 te)

A25485 6:1 Gee Vie)

A25486 4.5 4,3 1,6

136 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 18. Megerlia pisum (Lamarck, 1819). A-—D. SAM-—A25478, SM 163: dorsal, ventral, lateral

and anterior views of conjoined valves with serpulid worm tubes.

E-H. SAM-A25481, SM 185. E. Interior view of pedicle valve. F—H. Ventral, posterior and

lateral views of brachial valve interior showing brachidium. A-D x4. E-H x6.

Discussion

No less than four species of Megerlina have been recorded from the same

area off the eastern Cape coast from which the present specimens were

recovered. Previous descriptions of these species, M. pisum (Lamarck),

M. natalensis (Krauss), M. capensis (Adams & Reeve) and M. striata Jackson,

have been based on very few specimens and it is unlikely that the full extent of the

variation within the genus has been properly assessed, with the result that the

relationships between these four species are unknown.

The earlier descriptions distinguish between species on characters such as

colour, strength of ribbing, depth of sulcus and size and shape of the punctae. The

Meiring Naude specimens are all whitish in colour but pink or red around the

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES I E57/

margins. In this they are similar to the first three species named above. Their

strength of ribbing suggests affinities with both M. striata and M. capensis but no

figures of rib numbers are available for comparison. Megerlina pisum and

M. natalensis have less pronounced ribbing and may be smooth umbonally.

The depth of sulcus depends on the size of the shell, so comparisons should

only be made between specimens of approximately equal length; again examples

of other species are needed for comparison. In shape and density, the punctae of

the Meiring Naude specimens show closest similarity with the figures given by

Jackson (1952: 31) for M. pisum although the range overlaps with the ranges for

other species.

Wright (1972: 5) showed the range of variation that was possible within a

single species depending on its habitat especially when strength of ribbing and

colour were considered. It may be that the species of Megerlina recorded from off

the eastern Cape are simply variants of one species, M. pisum, so until such time

as re-assessment of these species is carried out, it is felt that the Meiring Naude

Specimens are best identified as M. pisum.

Family Phaneroporidae Zezina, 1981

Genus Leptothyrella Muir-Wood, 1965

Leptothyrella cf. ignota (Muir-Wood, 1959)

ig, IY)

Leptothyris ignota Muit-Wood, 1959: 308, pl. 4 (fig. 9), pl. 5 (figs 12-14).

Material

A single slightly damaged specimen (SAM-—A25482) from SM 129 at a depth

of 850 m. The specimen was dead at time of collection and shell was partially

filled with sediment.

Description

Small shell with elongately oval pedicle valve and almost circular brachial

valve; both valves gently and evenly convex, anterior commissure rectimarginate.

Ventral beak quite narrow, nearly straight; foramen hypothyridid; delthyrium

open with no, or extremely narrow, deltidial plates. Shell surface smooth except

for concentric growth lines. Punctae coarse.

Pedicle valve interior with strong teeth without dental plates; pedicle collar

broad, sessile, extending almost full length of delthyrium; broad, low, median

ridge extends to about two-thirds valve length. Brachial valve interior with strong

high socket ridges bounding fairly deep sockets; no cardinal process. Rounded,

narrow crura extend from anterior ends of socket ridges towards median septum,

anterior ends becoming flatter. High, pillar-like median septum originates in front

of notothyrial cavity and extends to about midvalve. No ring or hood present.

138 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 19. Leptothyrella cf. ignota (Muir-Wood, 1959), SAM—A25482, SM 129. A. Dorsal view of

conjoined valves. B. Interior view of pedicle valve. C—D. Ventral and lateral views of brachial

valve interior showing brachidium. All x8.

Dimensions (mm)

Length Width

SAM-A25482 DRS 1,8

Discussion

The inwardly directed crura, high pillar-like septum, wide foramen with

incipiently developed deltidial plates of this little specimen all point to an early

growth stage of a terebratellacean. However, the poorly known Leptothyrella

ignota (Muir-Wood) from off Zanzibar and the Gulf of Aden is remarkably

similar, although Muir-Wood’s specimens, which the author has examined, are

larger at 5 mm length. Both forms are coarsely punctate and have the same

overall shell outline with gently convex valves and rectimarginate anterior

commissure. Both have an open delthyrium flanked by extremely narrow deltidial

plates and floored by a pedicle collar that extends almost the full length of the

delthyrium. The brachial valve of L. ignota has a high, plate-like median septum,

which does not continue posteriorly into the notothyrial cavity; the crura are

slender and curved; the cardinal process is minute.

Any differences between L. ignota and the shell described here can be

accounted for by the larger size and presumably more adult nature of L. ignota,

by which stage the tiny cardinal process has developed as have the points of

attachment of the descending branches to the median septum. Otherwise both

forms are so similar that they must be regarded as being very closely related.

ACKNOWLEDGEMENTS

I wish to thank the Director of the South African Museum for allowing me

the opportunity to study the Meiring Naude brachiopods. Thanks are also due to

Professor A. D. Wright for permitting the author to use the facilities in the

Department of Geology, Queen’s University, Belfast, where much of the present

study was completed. Mr S. Watters of the same department took the

photographs and his contribution is gratefully acknowledged. I am indebted to Dr

C. H. C. Brunton of the British Museum (Natural History) for allowing me

access to the collections in his charge. Dr Daphne Lee read an earlier version of

the manuscript and her comments were greatly appreciated.

THE SOUTH AFRICAN MUSEUM’S MEIRING NAUDE CRUISES 139

REFERENCES

ADAMS, A. & REEVE, L. A. 1850. Mollusca. In: Zoology of the voyage of H.M.S. Samarang;

under the command of Captain Sir Edward Belcher . . . during the years 1843-1846: 1-88.

London.

Atcock, A. 1894. A new brachiopod. Jn: Natural history notes from the Royal Indian Marine

Survey Steamer /nvestigator, series 2, no. 13. Journal of the Asiatic Society of Bengal 63 (2):

139-140.

BLOCHMANN, F. 1906. Neue Brachiopoden der Valdivia und Gauss Expedition. Zoologische

Anzeiger 30: 690-702.

BLOCHMANN, F. 1908. Sur Systematik und geographischen Verbreitung der Brachiopoden.

Zeitschrift fiir Wissenschaftliche Zoologie 90: 596-644.

Cuun, C. 1900-1903. Brachiopoda. Jn: Aus den Tiefen des Weltmeeres 1, 2: 435-436. Jena.

Cooper, G. A. 1959. Genera of Tertiary and Recent rhynchonelloid brachiopods. Smith-

sonian Miscellaneous Collections 139 (5): 1-90.

Cooper, G. A. 1973a. Fossil and Recent Cancellothyridacea (Brachiopoda). Science Reports

of the Tohoku University (Sendai, Japan), (2) Geology, 6 (Hatai Memorial Volume):

371-390.

Cooper, G. A. 1973b. New Brachiopoda from the Indian Ocean. Smithsonian Contributions

to Paleobiology 16: 1-43.

Cooper, G. A. 1973c. Vema’s Brachiopoda (Recent). Smithsonian Contributions to

Paleobiology 17: 1-51.

Cooper, G. A. 1977. Brachiopods from the Caribbean Sea and adjacent waters. Studies in

Tropical Oceanography 14: 1-212.

Cooper, G. A. 1978. Tertiary and Quaternary brachiopods from the southwest Pacific.

Smithsonian Contributions to Paleobiology 38: 1-23.

Cooper, G. A. 1981. Brachiopoda from the southern Indian Ocean (Recent). Smithsonian

Contributions to Paleobiology 43: 1-93.

Cooper, G. A. 1983. The Terebratulacea (Brachiopoda), Triassic to Recent: A study of the

brachidia (loops). Smithsonian Contributions to Paleobiology 50: 1-445.

Da.L, W. H. 1920. Annotated list of the Recent Brachiopoda in the collection of the United

States National Museum, with descriptions of thirty-three new forms. Proceedings of the

United States National Museum 57 (2314): 261-377.

Davipson, T. 1880. Report on the Brachiopoda dredged by H.M.S. Challenger during the

years 1873-1876. Report of Scientific Results of the Challenger Expedition (Zoology) 1:

1—67.

Davipson, T. 1886-1888. A monograph of Recent Brachiopoda. Transactions of the Linnaean

Society of London (2) 4, Zoology (parts I-III): 1-248.

Foster, M. W. 1974. Recent Antarctic and Subantarctic brachiopods. American Geophysical

Union. Antarctic Research Series 21: 1-189.

HELMcKE, J. G. 1938. Die Brachiopoden des zoologischen Museums zu Berlin. Sitzungs-

berichte der Gesellschaft naturforschender Freunde zu Berlin ftir 1938: 221-268.

HELMcKE, J. G. 1940. Die Brachiopoden der deutschen Tiefsee-Expedition. Wissenschaftliche

Ergebnisse der Deutschen Tiefsee-Expedition auf dem Dampfer Valdivia, 1898—1899 24 (3):

217-316.

Jackson, J. W. 1952. A revision of some South African Brachiopoda; with descriptions of new

species. Annals of the South African Museum 41 (1): 1-40.

Kine, W. 1871. On Agulhasia davidsonii, a new palliobranchiate genus and species. Annals

and Magazine of Natural History (4) 7: 109-112.

Krauss, F. 1844-18484. Brachiopoda. In: Ktster, H. C. Conchylin-Cabinet von Martini und

Chemnitz I. 1844 (plates); 1848 (text).

Krauss, F. 1848b. Die Stidafrikanischen Mollusken: 1-140. Stuttgart.

Kuster, H. C. 1844-1848. Conchylin-Cabinet von Martini und Chemnitz 1. 1844 (plates); 1848

(ext):

Lamarck, J. P. B. A. DE M. pe. 1819. Histoire naturelle des animaux sans vertébres 6 (1):

1-343. Paris.

Louw, E. 1977. The South African Museum’s Meiring Naude cruises. Part 1. Station data 1975,

1976. Annals of the South African Museum 72 (8): 147-159.

140 ANNALS OF THE SOUTH AFRICAN MUSEUM

Louw, E. 1980. The South African Museum’s Meiring Naude cruises. Part 10. Station data

1977, 1978, 1979. Annals of the South African Museum 81 (5): 187-205.

Muir-Woop, H. M. 1959. Report on the Brachiopoda of the John Murray Expedition. John

Murray Expedition 1933-1934, Scientific Reports 10 (6): 238-317.

Patias, P. S. 1766. Miscellanea Zoologica, quibus imprimis et obscurae Animalium species

descriuntus: 182.

REEVE, L. A. 1860-1861. Terebratula. In: Conchologia Iconica, or Figures and descriptions of

molluscous animals 13. London.

THoMson, J. A. 1927. Brachiopod morphology and genera (Recent and Tertiary). New

Zealand Board of Science and Art, Manual 7: 1-338.

Turton, W. H. 1932. The marine shells of Port Alfred, South Africa. London: Oxford

University Press.

WiLuiaMs, A. et al. 1965. Brachiopoda. In: Moore, R. C. ed. Treatise on invertebrate

paleontology, part H (2 volumes). Lawrence, Kansas: University of Kansas Press.

Wricut, A. D. 1972. The relevance of zoological variation studies to the generic identification

of fossil brachiopods. Lethaia 5: 1-13.

ZEZINA, O. N. 1970. Brachiopod distribution in the Recent ocean with reference to problems of

zoogeographic zoning. Paleontologicheskiy Zhurnal (Academiya Nauk SSSR Paleontol-

ogicheskiy Institut) 2: 1-21.

ZEZINA, O. N. 1981. Recent deep-sea Brachiopoda from the western Pacific. Galathea Report

15: 7-20.

”

i }

a} }

1A)

}

iin

Up)

fal

6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu-

larly Articles 22 and 51).

Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed

by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov.,

ete.

An author’s name when cited must follow the name of the taxon without intervening punctuation

and not be abbreviated; if the year is added, a comma must separate author’s name and year. The

author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is trans-

ferred from its original genus. The name of a subsequent user of a scientific name must be separated

from the scientific name by a colon.

Synonymy arrangement should be according to chronology of names, i.e. all published scientific

names by which the species previously has been designated are listed in chronological order, with all

references to that name following in chronological order, e.g.:

Family Nuculanidae

Nuculana (Lembulus) bicuspidata (Gould, 1845)

Figs 14-15A

Nucula (Leda) bicuspidata Gould, 1845: 37.

Leda plicifera A. Adams, 1856: 50.

Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b).

Nucula largillierti Philippi, 1861: 87.

Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9.

Note punctuation in the above example:

comma separates author’s name and year

semicolon separates more than one reference by the same author

full stop separates references by different authors

figures of plates are enclosed in parentheses to distinguish them from text-figures

dash, not comma, separates consecutive numbers.

Synonymy arrangement according to chronology of bibliographic references, whereby the year is

placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable.

In describing new species, one specimen must be designated as the holotype; other specimens

mentioned in the original description are to be designated paratypes; additional material not regarded

as paratypes should be listed separately. The complete data (registration number, depository, descrip-

tion of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.:

Holotype

SAM-A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach, Port Eliza-

beth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973.

Note standard form of writing South African Museum registration numbers and date.

7. SPECIAL HOUSE RULES

Capital initial letters

(a) The Figures, Maps and Tables of the paper when referred to in the text

e.g. ‘. .. the Figure depicting C. namacolus ...’: ‘. . . in C. namacolus (Fig. 10)...’

(b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by

initials or full names

e.g. Du Toit but A.L. du Toit; Von Huene but F. von Huene

e.g. Therocephalia, but therocephalian

Punctuation should be loose, omitting all not strictly necessary

Reference to the author should preferably be expressed in the third person

Roman numerals should be converted to arabic, except when forming part of the title of a book or

article, such as

‘Revision of the Crustacea. Part VIII. The Amphipoda.’ i °!

Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial

capital letter, provided the same generic name is used consecutively. The generic name should

not be abbreviated at the beginning of a sentence or paragraph.

Name of new genus or species is not to be included in the title; it should be included in the abstract,

counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts.

NORTON HILLER

THE SOUTH AFRICAN MUSEUM’S

MEIRING NAUDE CRUISES

PART 16

BRACHIOPODA FROM THE

1975-1979 CRUISES

S67X : 97 PART 6 AUGUST 1986 ISSN 0303-2515

OF THE SOUTH AFRICAN

~ MUSEUM >

CAPE ‘TOWN

INSTRUCTIONS TO AUTHORS

1. MATERIAL should be original and not published elsewhere, in whole or in part.

2. LAYOUT should be as follows:

(a) Centred masthead to consist of

Title: informative but concise, without abbreviations and not including the names of new genera or species

Author’s(s’) name(s)

Address(es) of author(s) (institution where work was carried out)

Number of illustrations (figures, enumerated maps and tables, in this order)

(b) Abstract of not more than 200 words, intelligible to the reader without reference to the text

(d) Introduction

(e) Sudject-matter of the paper, divided into sections to correspond with those given in table of contents

(f) Summary, if paper is lengthy

(g) Acknowledgements

(h) References

(i) Abbreviations, where these are numerous.

3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with

3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for

illustrations should be typed separately, their positions indicated in the text. All pages should be num-

bered consecutively.

Major headings of the paper are centred capitals; first subheadings are shouldered small capitals;

second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further

subdivisions should be avoided, as also enumeration (never roman numerals) of headings and

abbreviations.

Footnotes should be avoided unless they are short and essential.

Only generic and specific names should be underlined to indicate italics; all other marking up

should be left to editor and publisher.

4. ILLUSTRATIONS should be reducible to a size not exceeding 12 x 18 cm (19 cm including

caption); the reduction or enlargement required should be indicated (and preferably uniform); orig-

inals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and

final size. A metric scale should appear with all illustrations, otherwise magnification or reduction

should be given in the caption; if the latter, then the final reduction or enlargement should be taken

into consideration.

All illustrations, whether line drawings or photographs, should be termed figures (plates are not

printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items

of composite figures should be designated by capital letters; lettering of figures is not set in type and

should be in lower-case letters. If Letraset is used authors are requested to use Helvetica-style letter-

ing, if possible.

The number of the figure should be lightly marked in pencil on the back of each illustration.

5. REFERENCES cited in text and synonymies should all be included in the list at the end of the

paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable):

(a) Author’s name and year of publication given in text, e.g.:

‘Smith (1969) describes. . .’

‘Smith (1969: 36, fig. 16) describes .. .’

‘As described (Smith 1969a, 1969b; Jones 1971)’

‘As described (Haughton & Broom 1927)...”

‘As described (Haughton et al. 1927)...”

Note: no comma separating name and year

Pagination indicated by colon, not p.

names of joint authors connected by ampersand

et al. in text for more than two joint authors, but names of all authors given in list of references.

(b) Full references at the end of the paper, arranged alphabetically by names, chronologically within

each name, with suffixes a, b, etc., to the year for more than one paper by the same author in

that year, e.g. Smith (1969a, 1969b) and not Smith (1969, 1969a).

For books give title in italics, edition, volume number, place of publication, publisher.

For journal article give title of article, title of journal in italics (according to the World list of scientific periodicals. 4th ed.

London: Butterworths, 1963), series in parentheses, vclume number, part number in parentheses, pagination (first and

last pages of article).

Examples (note capitalization and punctuation)

BuLLouGu, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan.

FiscHER, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100— 140.

FiscHer, P. H., Duvat, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie

expérimentale et générale 74 (33): 627-634.

Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and

Magazine of Natural History (13) 2 (17): 309-320.

Koun, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of

the Bingham Oceanographic Collection, Yale University 17 (4): 1-51.

THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthro-

pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Stid-Afrika ausgefiihrt in den Jahren

1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270.

(continued inside back cover)

ANNALS OF THE SOUTH AFRICAN MUSEUM

ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

Volume 97 Band

August 1986 Augustus

Part & Deel

LATE TERTIARY AND EARLY

QUATERNARY FOSSIL MOLLUSCA

OF THE HONDEKLIP AREA,

CAPE PROVINCE, SOUTH AFRICA

BRIAN KENSLEY

JOHN PETHER

Cape Town Kaapstad

The ANNALS OF THE SOUTH AFRICAN MUSEUM

are issued in parts at irregular intervals as material

becomes available

Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000

Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

word uitgegee in dele op ongereelde tye na gelang van die

beskikbaarheid van stof

Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000

OUT OF PRINT/UIT DRUK

iL, WOES, 5-8), SGD, 45, 8, tan.) SUS, 5, FO),

GGL tap.) TOD), 8, CID, 1), 1003),

11(1-2, 5, 7, t.-p.i.), 14(1-2), 15(4-5), 24(2), 27, 31(1-3), 32(5), 33, 36(2), 45(1)

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 083 4

Printed in South Africa by In Suid-Afrika gedruk deur

The Rustica Press, Pty., Ltd., Die Rustica-pers, Edms., Bpk.,

Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap

LATE TERTIARY AND EARLY QUATERNARY FOSSIL MOLLUSCA

OF THE HONDEKLIP AREA, CAPE PROVINCE, SOUTH AFRICA

BRIAN KENSLEY

National Museum of Natural History,

Smithsonian Institution, Washington, D.C.

JOHN PETHER

Department of Cenozoic Palaeontology,

South African Museum, Cape Town

(With 42 figures and 1 table)

[MS accepted 10 February 1986]

ABSTRACT

A brief review of the Quaternary mollusc research of the west coast of South Africa is

provided. Molluscan fossils were obtained from two coastal regressive sedimentary complexes.

The older complex records a regression from c. +50 m, and is called the 50 m Complex.

Stratigraphic context in terms of global sea-level curves and the presence of Equus in this

complex suggests a Late Pliocene age. The subsequent complex records a regression from

c. +30 m, is called the 30 m Complex, and is considered Early Pleistocene in age.

The following 21 species and one subspecies are described as new: Patella hendeyi, Patella

hoffmani, Clanculus lutosus, Gibbula zonata patula, Bolma anoropha, Crepidula deprima,

Argobuccinum casus, Epitonium lycocephalum, Thais arenae, Ocenebra petrocyon, Trophon

carringtoni, Burnupena rogersi, Burnupena aestus, Fasciolaria dinglei, Melapium hawthornei,

Pseudoliva lutulenta, Drillia tempestae, Terebra canisaxi, Glycymeris fulleri, Isognomon

gariesensis, Cardita unica, Dosinia sicarisinus. A number of previously described mollusc species

are discussed or redescribed. The faunal composition and biogeography of the Hondeklip

assemblage are discussed, and the assemblage is compared with other west-coast fossil

assemblages. Temperature tolerances of some extant forms present indicate that a warmer water

regime relative to modern conditions existed along the Namaqualand coast during the Late

Pliocene and Early Pleistocene. A temperature contrast across the Plio—Pleistocene boundary is

not clear-cut due to the faunal similarity of the 50 m and the 30 m Complexes. Significantly,

about 50 per cent of the fauna is extinct. Decreased diversity and the appearance of Choromytilus

meridionalis in the 30 m Complex may indicate cooling in the Early Pleistocene. Subsequently,

the restriction of warmer waters to the north led to the establishment of the modern west-coast

fauna. Lack of a broader regional database precludes more precise conclusions at this stage.

CONTENTS

PAGE

MatROGNC HOMER MeL erase Sora: ie heen e cee dels A acoo aetee 142

Gcolocicalise ttinpiy rr trc Menon wlo ste ge ens 2 ee el oe 142

The Hondeklip and Avontuur A deposits .............. 144

The Koingnaas and Swartlintjiesrivier localities......... 148

SWSUSMEMIS COIS OOS g.56500edensandoesdoosepeoge cour 149

DISCUSSION ee eee ie ais ae ee Nfs 5 5. Suse 5, Uke os 210

SAUNA GOMIOONWON soo gsoouce onnseoposesosoo lu ouUE 210

ASOSCOPKAP MC allIMIGIES Hee ae Ase ee ee DAW

BAUMAN COM PAtSOMS Es senha Hate ors oa eee = 219

Goncluichinegne maths payee ee ae acme Sani cate tee atin Noli 221

/ANOSNONUECUSTNEDIS 506 ob eae oe oon Semioe Ooo eee Deo dab amen 223

INORG GING Simm wan wn REA mM in Oye Sinus ha hati a eo ete! 223

141

Ann. S. Afr. Mus. 97 (6), 1986: 141-225, 42 figs, 1 table.

142 ANNALS OF THE SOUTH AFRICAN MUSEUM

INTRODUCTION

The fossil mollusc occurrences on the west coast of South Africa have

received scattered attention during the past 50 years, with reports becoming more

numerous in the last decade. Haughton (1932) presented an overview of the

west-coast deposits and described several of the fossil molluscs. Barnard (1962)

listed all the known Late Tertiary and Pleistocene molluscs from South Africa.

New species and records for the west coast have since been added by Carrington

& Kensley (1969), Kilburn & Tankard (1975), and Kensley (1972, 1977).

Diamond-mining activities on the Namaqualand coast have exposed coastal

marine deposits in several areas. Fossil molluscs from three localities, Koingnaas,

Swartlintjiesrivier, and Strandfontein, were the subject of the paper by

Carrington & Kensley (1969). Material recorded in the present paper also comes

from Koingnaas, plus two additional properties, Avontuur A, and Hondeklip

(Fig. 1A, B). Exposures of several fossil-bearing beds were bulk-sampled, while

selective sampling was carried out when appropriate.

The bulk of the material recorded in the accompanying list (Table 1) is

housed in the South African Museum, as are holotypes and paratypes, which bear

SAM-POQ catalogue numbers. Where possible, paratypic material has also been

deposited in the National Museum of Natural History, Smithsonian Institution,

Washington, D.C. and bears USNM catalogue numbers.

GEOLOGICAL SETTING

This paper is a sequel to that of Carrington & Kensley (1969) in which several

new species obtained from similar deposits were described and a summarized

general account of the Cenozoic coastal Namaqualand stratigraphy was presen-

ted. Important observations recorded in that work were the recognition of two

distinct littoral marine complexes, the 45-50 m and the 17-21 m transgression

complexes, and their zone fossils Donax haughtoni Carrington & Kensley, and

Donax rogersi Haughton, respectively. The complex names indicate perceived

transgressive altimetric maxima. The older 45-50 m complex was considered

typified by fine, green quartzose sands and the associated thin-shelled D. haugh-

toni, while the younger 17—21 m complex was characterized by brown-stained,

coarse, quartzo-feldspathic sands, high-energy bedforms, and the robust

D. rogersi. An evolutionary relationship was inferred between the two Donax

species, and linked to the changed environment implicit in the contrasted

lithologies of the two complexes. Additionally, species obtained from the

45-50 m complex suggested a fauna of warm-water affinity.

Subsequent fieldwork on the properties of Hondeklip and Avontuur A has

furnished more information on the depositional environments, ages, and sea-level

history of the 45-50 m and 17—21 m complexes. This work will be presented in

detail in another article (Pether in prep.); thus only a condensed account is given

here. Modifications of the succession presented by Carrington & Kensley (1969)

are shown in the following summarized stratigraphy.

FOSSIL MOLLUSCA OF HONDEKLIP 143

SWA/NAMIBIA

klipbaai

SOUTH

AFRICA

xX Main sampling localities

<10

30 Marine

50 packages

Hondeklipbaai\ 32.2 Qos SEIN Bedrock

Seeepe ee) gie “§ : > 30 masl.

ATLANTIC ie NY :{ 20-30 masl.

OCEAN oe ae een 10-20 masl.

Iam = ——=— 50m iI D

Fig. 1. A. Locality map. B. Hondeklip and surrounding properties with main sampling areas

indicated. C. Bedrock contour map of Hondeklip and a portion of Avontuur A in metres above

sea-level (masl). Large numerals (90, 50, 30) indicate the overlying deposits of the marine

complexes; < 10 refers to the low-elevation, Mid-Pleistocene to Holocene raised beaches.

Palaeoshorelines at transgressive maxima of the 30 m and 50 m Complexes are shown.

144 ANNALS OF THE SOUTH AFRICAN MUSEUM

Succession Age

*D metreWBeachirn s t Ae ce es teeter eater ae ate ree Holocene

*S5 ‘metre Beach siecce wanes ote ee ee Late Pleistocene

8 metre Beach Van sky lees unease eerie Middle Pleistocene

30 ‘metre Complex tse ee ne ane a akon Early Pleistocene

Regressive complex (incl. 29-34 m Beach and 17—21 m Complex)

SO'metne Complex sy. fey eee ele as een ee ears ae Late Pliocene

Regressive complex (= 45-50 m Complex)

90: metre Complex.) seas ages ce ee ee eee een Early Pliocene

Regressive complex? (= 75-90 m Complex)

(Includes lower shoreface to offshore deposits preserved in bedrock

depressions underlying the 50 m Complex at lower elevations. )

Kaolimizediiuviatile depositses saya e earner Latest Oligocene

to Early Miocene

BeGrock av S x.gucd sre ai ttaet ats alga ac ace (ae alae pa Bae GH ne a Archaean

THE HONDEKLIP AND AVONTUUR A DEPOSITS

Bedrock

The bedrock consists of Archaean quartzo-feldspathic gneiss, and bedrock

topography (Fig. 1C) has profoundly influenced the local character of the

overlying deposits. A major feature in the study area is a broad, coast-parallel,

channel-like structure in the bedrock, which has its western flank defined by a

bedrock ridge. Emergence of the bedrock ridge during regression resulted in a

barrier along the seaward flank of the channel, protecting the channel area from

open-coast conditions.

Kaolinized fluviatile deposits

This, the oldest sedimentary entity encountered, is locally preserved in a

bedrock depression. It is the remains of a fluvial arkosic infill deposited in the

bedrock channel. The underlying bedrock and this fluvial infill are kaolinized and

associated with it are silcrete slabs and boulders that are indurated portions of the

fluvial deposits that have been exhumed by extensive erosion.

The bedrock is considered to have been incised during the major Oligocene

regression that has been identified in the offshore record (Siesser & Dingle 1981).

Subsequently, infilling Late Oligocene to Earliest Miocene fluvial deposits were

kaolinized during the Early Miocene under tropical conditions; within the

weathering profile titanium-rich silcrete developed. Sea-level fluctuations and

concomitant marine and fluvial erosion during Middle Miocene to Late Pliocene

times exhumed and modified the bedrock channel. No Miocene marine deposits

are preserved.

* After Hendey & Volman (1986)

FOSSIL MOLLUSCA OF HONDEKLIP 145

The 90 m Complex deposits

This high elevation complex, the 75-90 m Complex of Carrington & Kensley

(1969), is present between 50—90 m above sea-level (masl), but no exposures at

these elevations exist in the study area. However, underlying the younger 50 m

Complex in a local bedrock depression, is a remnant deposit that may have been

deposited, wholly or in part, during the same sea-level cycle. This is a partly

indurated, mouldic, coquinoid, muddy conglomerate, which exhibits extensive

(though not laterally continuous), authigenic phosphorite (microsphorite) deposi-

tion. A ubiquitous, similar, phosphorite gravel content in overlying 50m

Complex gravels indicates the erosion of this previously more extensive lithology.

Fragmentary, worn, phosphate-mineralized, marine and terrestrial verte-

brate fossils are associated with this bed. These have a Mio-Pliocene aspect

(Q. B. Hendey pers. comm.). The mouldic coquinoid fauna has both intertidal

and offshore aspects. The bed is considered to have been storm-deposited in a

lower shoreface to offshore environment during regression, in an embayed

situation. Extensive microsphorite precipitation driven by upwelling took place

during calmer periods. This bed has previously been referred to as ‘lower E stage’

(e.g. Tankard 19755, 1975c).

The 50 m Complex deposits

This complex is present mainly to the east of the bedrock ridge (Fig. 1C).

Exposures in open-coast situations reveal a regressive vertical facies sequence of

lower-shoreface, upper-shoreface, foreshore, and aeolian environments. The

basal gravel is seen to be a transgressive veneer, which has been redeposited and

supplemented during regression in a lower-shoreface context. The bedrock ridge

promoted the development of back-barrier environments in its landward lee

during regressive emergence.

Evidence of a minor transgressive sea-level fluctuation is preserved in this

complex and consists of a laterally persistent surface produced in back-barrier

deposits, an anomalously thick foreshore deposit, a vertical aspect to normally

lateral open-coast barrier facies deposition, and a tidal inlet-dominated barrier

overlying back-barrier deposits.

The 30 m Complex deposits

The 50 m Complex was eroded by a subsequent transgression and overlying

the transgressive disconformity is the 30 m Complex. This is a seaward-thickening

wedge, which, beneath foreshore deposits extending from the transgressive

maximum near +30 m, progressively incorporates upper-shoreface, and then

lower-shoreface deposits. This complex is preserved mainly to the west (seaward)

of the bedrock ridge (Fig. 1C). Sea-level altitude relative to bedrock topography

suggests that generally during 30m Complex times the effect of bedrock

topography was not as marked as during the previous sea-level cycle. In the study

146 ANNALS OF THE SOUTH AFRICAN MUSEUM

area there is no evidence of back-barrier environments during 30 m Complex

times.

The 30 m Complex subsumes the 17—21 m Complex and the 29-34 m Beach

of Carrington & Kensley (1969). No evidence of a stillstand at c. 20 masl is

present in Hondeklip exposures. Instead, 30 m Complex lower-shoreface gravels

extend seawards from that elevation. However, the possibility that stillstands

occurred subsequently during the regression cannot be dismissed at this stage.

The complex is present almost up to the modern coastline, where it is overlapped

by younger palaeoshorelines at elevations of less than 10 masl.

Age of the marine complexes

Onshore and offshore evidence of Tertiary sea-levels around southern Africa

has been synthesized in a curve by Siesser & Dingle (1981) (Fig. 2A); this curve

exhibits a general correspondence with the trends of global sea-level produced by

Vail & Hardenbol (1979). Sea-level history and age inferred for the deposits at

Langebaanweg (Fig. 2B) is consistent with the global sea-level record (Hendey

1981a, 1981b) and, pointing out the feasibility of altimetric correlation along the

southern-African west coast, Hendey proposed a preliminary correlation scheme

between Langebaanweg and west-coast localities involving the high-elevation

(> 10 masl) deposits.

‘The correlation scheme was related to the sea-level curve of Vail &

Hardenbol (1979). Beard etal. (1982) provided greater resolution for the

Quaternary portion of the ‘Vail curve’ (sea-level cycles Q1 to Q8 (Fig. 2C)).

Although methodological aspects and the chronology of the curve are controver-

sial, it is accepted here as presented and considered to be of provisional utility.

For example, Vail & Hardenbol (1979) and Beard et al. (1982) placed the Plio—

Pleistocene boundary at 2,8 Ma; in contrast, in terms of the most recent ICS

decision (Aguirre & Pasini 1985), the boundary at the Vrica section is ~1,64 Ma.

Regardless of this controversy, it is the sea-level curves, as gross indicators, that

are deemed relevant, and interestingly they do show a general correspondence

with our present (though scant) knowledge of west-coast deposits. In this paper

the Plio—Pleistocene boundary is taken at ~1,6 Ma; thus sea-level cycle Ql

(Fig. 2C) is considered Late Pliocene.

An important feature of Hendey’s (1981a, 1981b) correlation scheme was the

correlation of the 90 m Complex in Namaqualand with the Early Pliocene

Varswater Formation at Langebaanweg; the latter, in turn, related to the

transgressive sea-level period TP1 of Vail & Hardenbol (1979) (Fig. 2B). An

Early Pliocene age for the 90 m Complex is consistent with observations at

Hondeklip. The likelihood of the complex being a regressive package in

Namaqualand suggests correlation with sea-level cycle TP2.

The presence of a species of Equus in the 50 m Complex suggests that it

cannot be older than the 1,9 Ma mammalian dispersal event in Africa (Lindsay

et al. 1980). Similar to the age inferred by Hendey (1981a) for the Baards Quarry

Fluviatile Deposits in the south-western Cape which contain Equus, the 50 m

FOSSIL MOLLUSCA OF HONDEKLIP 147

20 15 10

MIOCENE PLIOCENE PLEISTO.

Surface

deposits

Years BP x10°

3 -28

PEO EN E QUATERNARY

Fig. 2. A. Gross Tertiary sea-level trends after Siesser & Dingle (1981). B. Langebaanweg

deposits (south-western Cape) related to global sea-level trends of Vail & Hardenbol (1979)

(after Hendey 1981b). GM = Gravel Member, QSM = Quartzose Sand Member, PPM = Pelletal

Phosphorite Member, AM=Anyskop Marine Deposits—all of the Varswater Formation.

AT = Anyskop terrestrial deposits, BQF=Baard’s Quarry fluviatile deposits. C. Gross

Quaternary sea-levels of Beard et al. (1982).

148 ANNALS OF THE SOUTH AFRICAN MUSEUM

Complex is considered Late Pliocene in age and is correlated with the regressive

portion of sea-level cycle Q1 of Vail & Hardenbol (1979) and Beard et al. (1982).

The 30 m Complex must reflect a subsequent high sea-level, and it is thus

correlated with cycle Q2 in the Early Pleistocene. A Late Pliocene and Early

Pleistocene age for the 50 m and 30 m Complexes, respectively, is probably

consistent with the warmer-water faunal elements present, relative to the colder-

water assemblages of the younger, low-elevation complexes and of the modern

coast.

The age of these low-elevation (< 10 masl) palaeo-shorelines has recently

been assessed (Hendey & Volman 1986). Fossil evidence associated with the 8 m

Beach at Saldanha suggests a late Early Pleistocene age to Hendey & Cooke

(1985) and it was correlated with sea-level cycle Q2. Since this report prefers to

relate the 30 m Complex to cycle Q2, the 8 m Beach is here considered to reflect

cycle Q3. This younger (Middle Pleistocene) age is inconsistent with the age

constraints inferred from the fossil evidence at Saldanha, an issue yet to be

resolved. The 5 m and 2m shorelines are considered to be Late Pleistocene

(Eemian) and Holocene respectively (Hendey & Volman 1986). Flemming (1977)

has submitted evidence that the Flandrian transgression reached 3 masl and

profoundly influenced the development of the Langebaan lagoon. Recently a

radiocarbon date of 3800 BP has been obtained at 2,5 masl at Verlorevlei

(J. Parkington pers. comm.).

THE KOINGNAAS AND SWARTLINTIJIESRIVIER LOCALITIES

Some basal exposures of the deposits on De Beers properties neighbouring

Hondeklip (Fig. 1B) were examined and seen to contain molluscs not well

represented at Hondeklip. In the case of the 50 m Complex outcrops (Koingnaas

KN and Swartlintjies SL), this is due to the occurrence of thicker, muddy, distal

lower shoreface deposits in bedrock depressions at these localities. At Hondeklip

the deposits of this environment are thinner and more cryptic, due to the

embayed palaeocoastal geomorphology formed by the bedrock topography, in

contrast to the more open-coast, palaeocoastal situation considered applicable to

the De Beers properties. This facies overlies bedrock or local 90 m Complex

remnants, and is overlain by gravelly, proximal lower shoreface deposits of the

50 m Complex. It is characterized by a molluscan fauna with both intertidal and

sublittoral components, and in these respects the similarity with the 90m

Complex remnants is noteworthy.

An outcrop of the 30 m Complex on Koingnaas also yielded species not

encountered at Hondeklip (Koingnaas KL). At this exposure a massive sand,

containing thin gravel beds and a mud lens ~20 cm thick, overlies bedrock at

10 masl. It attains its greatest thickness (~2 m) in a local bedrock depression.

This unit is overlain by the 30 m Complex upper shoreface facies. It is thus

interpreted as 30 m Complex proximal lower shoreface deposits. The open-coast

forms Isognomon sp., Mactra sp., and Turritella carinifera dominate the

assemblage.

FOSSIL MOLLUSCA OF HONDEKLIP 149

SYSTEMATIC DESCRIPTIONS

In the following section, in addition to the previously undescribed material,

discussion is given to some previously recorded taxa, where relevant.

As is frequently the case in fossil molluscan systematics, generic allocation

cannot always be done with certainty, due to the lack of essential diagnostic

information (e.g. radular and opercular structure). This uncertainty in generic

placement was constantly in the forefront of the authors’ thoughts when writing

descriptions; nevertheless it was decided not to indicate this uncertainty with

quotation marks or any other such device.

Unless otherwise stated, all material mentioned in this work was collected by

the authors.

Class GASTROPODA

Family Haliotidae

Haliotis saldanhae Kensley, 1972

Haliotis saldanhae Kensley, 1972: 176, fig. 2.

Material

SAM-POQ-AV229, several fragments, Avontuur A, Trench 3.

SAM-—PQ-KN370, numerous fragments, including seven triangular frag-

ments, each the upper columella portion of an individual shell, Koingnaas KN-1.

Previous records

Langebaanweg, Pliocene.

Remarks

Several large abalone specimens have been seen in situ at Avontuur, but the

shell material is usually so fragile and friable as to make recovery almost

impossible. Several fragments of shell, however, have been recovered, bearing

the distinctive rounded spiral ridges and broadly rounded oblique axial ridges

characteristic of H. saldanhae.

A single specimen (SAM-PQ-AV506), c. 60 x 140 mm, was recovered,

which retains its overall shape and proportions. Unfortunately none of the

exterior shell structure remains, making positive identification impossible. The

size of this specimen places it well beyond the maximum for H. spadicea

Donovan, 1808, but within the size range of living H. midae Linnaeus, 1758, of

the west coast.

Family Fissurellidae

Amblychilepas scutellum (Gmelin, 1791)

Amblychilepas scutella: Barnard, 1962: 191; 1963: 286, figs 21b, 22d-f.

Amblychilepas scutellum: Kilburn & Rippey, 1982: 35, pl. 2 (fig. 11), pl. 6 (fig. 1a, b).

150 ANNALS OF THE SOUTH AFRICAN MUSEUM

Material

SAM-—PQ-HB319, 4 specimens; TL 22,5 mm, 23,3 mm, 29,0 mm, 1 dam-

aged, Hondeklip Zone 12. SAM—PQ-HB74, 2 specimens, Hondeklip, 50 m

Complex.

Previous records

Living: Angola to Natal.

Fossil: Saldanha, Little Brak River, Sedgefield, Durban, Inhambane.

Remarks

The living subspecies, A. s. scutellum of the west coast, is characterized by

the possession of ‘moderately to extremely raised ends’ (Kilburn & Rippey 1982:

35). The present material consists of six specimens not at all saddle-shaped, i.e.

sitting flat on a horizontal surface, and one specimen with very slightly raised

ends.

Fissurellidea aperta (Sowerby, 1825)

Pupillaea aperta Sowerby, Barnard, 1962: 191.

Fissurellidea aperta (Sowerby) Barnard, 1963: 288, fig. 21e. Kensley, 1973: 32, fig. 46. Kilburn &

Rippey, 1982: 36, pl. 3 (fig. 3), pl. 6 (fig. 13).

Material

SAM-—PQ-KN507, 1 specimen, 22,8 x 12,8 mm, 4 fragments, Koingnaas

KN-1.

Previous records

Living: Namibia to Transkei, shallow infratidal.

Fossil: Pleistocene, Algoa Bay.

Tugali barnardi (Tomlin, 1932)

Tugalia barnardi: Barnard, 1963: 300, figs 21d, 22a—c.

Material

SAM-—POQ-KN392, 2 specimens, 4,0 2,8 mm, 8,1 x 4,6 mm, Koingnaas

KN-1.

Previous records

Living: off Cape Point, 360 m; False Bay, 18 m; St. Francis Bay, 12 m.

Dead: Cape Point to Cape Morgan, intertidal to 360 m.

Remarks

The fossil specimens have been compared with material in the South African

Museum, and agree very closely with specimens from St. Francis Bay

(SAM-A9299). The external ridge sculpture is as figured by Barnard (1963: 291,

fig. 22b).

FOSSIL MOLLUSCA OF HONDEKLIP tel

Family Patellidae

Patella hendeyi sp. nov.

Figs 3, 4

Patella sp. (‘large’) Haughton, 1932: 30.

Material

Holotype. SAM—PQ-AV454, 126 x 107 mm, Avontuur A, 50 m Complex.

Paratypes. SAM—PQ-AV455—457, 7 specimens, 68 x 57 mm, 74 x 61 mm,

90 x 76 mm, 9477 mm, ? X 109mm, 142 x 133 mm, 146 x 137 mm, Avon-

tuur A, 50 m Complex. SAM—POQ-HB458, 148 x 130 mm, Hondeklip Z4A,

50 m Complex (with attached barnacle shells). USNM 400979, 2 specimens,

122 x 112 mm, 138 X 122 mm, Avontuur A, 50 m Complex.

Fig. 3. Patella hendeyi. Holotype in dorsal view.

Scale = 30 mm.

ANNALS OF THE SOUTH AFRICAN MUSEUM

ht corner, remainder of specimens paratypes.

Holotype in upper rig

Scale = 30 mm.

Patella hendey

FOSSIL MOLLUSCA OF HONDEKLIP 153

Non-type material. SAM-PQ-AV459, 4 damaged and incomplete speci-

mens, Avontuur A. SAM-—9921, 2 specimens, 10691 mm, 98 x 85 mm,

Graauweduinen, Vanrhynsdorp District coast, Cape Province, coll. H. Harger.

SAM-—9932, 136 x 124 mm, Graauweduinen, Vanrhynsdorp District coast, coll.

H. S. Harger.

Description

Shell heavy, relatively low, with subcircular outline, only slightly longer than

wide, with apex anterior to midline, apex in all specimens eroded. Sculpture

consisting of fine radial lines of subequal strength (about 300 on circumference of

holotype, 200-300 in paratypes), becoming divided and/or distorted close to

circumference. No juvenile scars on any specimens.

Remarks

The specimens numbered SAM-—9921 and 9932 are those referred to by

Haughton (1932: 30) as ‘Patella sp. (large)’.

Of present-day patellids living on the southern African coast, in terms of

size, only P. tabularis Krauss, 1848, reaches the size of the largest specimen

considered here. This species, however, characteristically has very large acute

radiating ribs alternating with weaker ribs, and an irregular shell margin.

Patella safiana Lamarck, 1819, of West Africa has more regularly spaced but

stronger ribs.

Patella argenvillei Krauss, 1848, has a narrower and higher shell.

Only very large shells of P. granularis Linnaeus, 1758, approach the present

material in shape and sculpture. Prof. G. Branch (U.C.T.) has donated a

specimen of P. granularis (91,0 x 74,9 mm), that in shape closely resembles some

of the smaller of the present fossils, and has faint radial sculpture as in the largest

fossil. The possibility exists that P. hendeyi represents either a precursor of

P. granularis or an earlier and much larger form of the same species.

Patella (Ancistromesus) fuenzalidai Herm, 1969, from the Pliocene of central

Chile, bears a striking resemblance to the present species. The holotype and

paratype of this species, at 209 mm and 188 mm length, are even larger than

specimens of P. hendeyi, but have a very similar outline and general proportions

and also the very fine radiating ribs seen in the Namaqualand species.

Etymology

The species is named for Dr QO. B. Hendey of the South African Museum.

Patella hoffmani sp. nov.

Figs 5, 6

Material

Holotype. SAM-PQ-AV508, 56,7 x 51,0 31,0 mm_ (length x width x

height), Avontuur A.

Paratypes. SAM-PQ-AV402, 9 specimens, 23,3 x 18,9 x 8,3 mm;

34.3) 6 ATs XS MOL woven 34.8 x 31,4 x 13,8 mm; ADO xX 38.5% 15.8 “mnt:

154 ANNALS OF THE SOUTH AFRICAN MUSEUM

44.5 x 41,0 x 20,0 mm, 47,7 X 43,6 X 22,4 mm; 47,3 x 42,9 x 20,0 mm;

46,7 < 41,8 X 25,4 mm; 54,2 x 48,2 x 30,8 mm; Avontuur A. SAM—POQ-HB188,

48,4 x 41,2 x 27,1 mm, Hondeklip, 50 m Complex. SAM —PQ-HB191, 2 speci-

mens, 48,5 xX 41,9 x 23,0 mm; 51,6 x 45,8 x 26,7 mm; Hondeklip, 50 m Com-

plex. SAM-—POQ-HB142, 3 specimens, 39,7 x 32,0 18,7 mm; 46,9 x 38,8 x

21,7 mm; 47,7 X 40,2 x 25,6 mm; Hondeklip, 50 m Complex. SAM-—K4755,

49,5 x 46,0 x 24,1 mm; SwartlintjiesSL2, coll. A.J. Tankard. SAM-—

PQ-HBS509, 3specimens, 43,8 x 40,432,3mm; 49,4 x 45,0 x 29,8 mm;

51,2 x 44,1 x 34,5 mm; Hondeklip Zone 4C, 50 m Complex. USNM 400980,

6 specimens, ?x37,2X18,4 mm; 42,2 <38)5:x 17,0 mm; 4444539"

18,6 mm; 45,8 x 40,6 X 27,5 mm; 52,3 X 47,5 x 31,1 mm; 54,3 x 49,0 x 25,3 mm;

Avontuur A, 50 m Complex.

Fig. 5. Patella hoffmani. A-C. Lateral, dorsal and ventral views of holotype.

Scale =20 mm. D. Specimen showing juvenile scars. Scale = 10 mm.

FOSSIL MOLLUSCA OF HONDEKLIP 155

Non-type material. SAM—PQ-HB249, 4 specimens, Hondeklip, B Block,

50 m Complex. SAM—PQ-—AV402, 12 specimens, Avontuur A, 50 m Complex.

SAM-PQ-AV510, 4 specimens, Avontuur A, 50 m Complex. SAM-—K4754,

3 specimens, Koingnaas, coll. A. J. Carrington. SAM-—K4756, 2 specimens,

Somnaas SM4, 50m, coll. A. J. Tankard. SAM—PQ-HB461, 27 specimens,

Hondeklip, Zone 4C, 50 m Complex.

Description

Shell generally high-conical; length/width ratio range 80-93 per cent; sides

very faintly convex; apex at or close to midpoint of antero-posterior line.

Sculpture consisting of radiating flattened ribs of equal strength; under

magnification, ribs seen to be built up of numerous concave growth lines,

Fig. 6. Patella hoffmani. Range of specimens in dorsal view, some showing colour-rays.

Scale = 30 mm.

156 ANNALS OF THE SOUTH AFRICAN MUSEUM

somewhat flattened, separated by shallow sharp groove. Number of ribs on

circumference of large specimens ranging from 105 to 124. Shell margin finely and

evenly serrate. Large shells often with several oval juvenile scars, up to nine per

shell; scars usually 4-10 mm greatest diameter. Some shells bearing dark-brown

radiating bands separated by narrow pale bands; dark rays more distinct closer to

shell apex; near margin, dark bands coalescing to form continuous dark-brown

band.

Remarks

The almost circular circumference and tall conical shape of Patella hoffmani

bears little resemblance either to any of the living or fossil southern or West

African patellids. The sculpture of equa! flattened radiating ribs, however, is

almost identical to that of Patella argenvillei.

The subcircular and high-conical shape of P. hoffmani is reminiscent of

Nacella clypeater (Lesson, 1831) from Chile. This species, however, possesses

fewer radiating ribs which, unlike the flattened ribs of P. hoffmani, are low and

rounded.

Etymology

The species is named for Mr F. Hoffman, of Transhex Pty Ltd diamond

company.

Family Trochidae

Clanculus lutosus sp. nov.

Fig. 7

Material

Holotype. SAM—PQ-KNS11, 16,3 x 18,3 mm, outer lip damaged, Koing-

naas KN-1.

Paratypes. SAM—PQ-KNS12, 5 specimens, Koingnaas KN-1.

Non-type material. SAM—PQ-—KNS513, 12 damaged specimens, Koingnaas

KN-1.

Description

Shell top-shaped, of at least 4 postnatal whorls. Whorl profile straight to

barely convex. Sculpture of spiral lines bearing bead-like, close-set tubercles;

second whorl with 4—6 spiral lirae; third whorl with 10-12 lirae; body whorl with

10 lirae above shoulder, 12—14 on base. Shoulder narrowly rounded. Umbilicus

broadly open, smooth, with single low spiral ridge. Columella with bipartite

tubercle on lower portion. Outer lip slightly thickened, but lacking internal

plicae.

Remarks

Clanculus murrayi has been described from the Quaternary of the Honde-

klipbaai area (Carrington & Kensley 1969). This species, while having beaded

FOSSIL MOLLUSCA OF HONDEKLIP 157

Fig. 7. Clanculus lutosus. Holotype. Scale = 10 mm.

spiral lirae as in C. lutosus, is characterized by a stepped profile, with two strong

lirae on the whorl periphery.

Clanculus atricatena Tomlin, 1921, known from Transkei to Zululand,

possesses a grooved lower tubercle on the columella, giving it a bipartite

appearance. The spiral lirae of this specids, however, are more finely beaded,

while the outer lip possesses a strong posterior tubercle on the inner surface.

Clanculus kraussi Philippi, 1846, from West Africa, possesses finely beaded

spiral lirae, but lacks the bipartite columella tubercle of C. lutosus. Two other

West African species, however, show a stronger resemblance to the present

material. Clanculus santamariae Gofas, 1984, from Benguela, Angola, has spiral

lirae with fewer but larger beads, a crenuiate umbilical opening, and a non-bifid

columella tooth. Clanculus pseudocorallinus Gofas, 1984, also from Benguela,

Angola, possesses sculpture very similar to that of C. /utosus, but also has a

crenulate umbilical opening, and a non-bifid columella tooth. Clanculus

corallinus (Gmelin, 1790) of the Mediterranean has fewer spiral lirae with larger

beads, but does have a bifid columella tooth, and a non-crenulate umbilical

opening as in C. lutosus.

Etymology

The specific name is derived from the Latin ‘lutosus’, muddy, and refers to

the sediments from which the material was collected.

158 ANNALS OF THE SOUTH AFRICAN MUSEUM

Gibbula zonata patula subsp. nov.

Fig. 8

Material

Holotype. SAM—PQ-HB514, 5,0 x 6,9 mm, body whorl damaged, Honde-

klip Zone 3.

Paratypes. SAM—PQ-HB117, 4 specimens, 4,9 5,6 mm, 4,3 x5,5 mm,

6,1xX7,2mm, 6,3x?mm, Hondeklip Zone 3. SAM-—PQ-HB345, 6,8 x

6,5 mm, Hondeklip Zone 12.

Description

Shell broader than high, somewhat globular, of 4 whorls. Profile of whorls

evenly rounded. Sculpture of rounded spiral bands; penultimate whorl with

6 bands; body whorl sculpture variable, with 8—10 bands on upper whorl; some

specimens with intermediate finer lirae; ventral whorl with 8-12 lirae. Umbilicus

broad, open in all specimens. Raised rounded bands in some specimens having

darker colour than rest of shell.

Fig. 8. Gibbula zonata patula. Holotype in upper left corner, remainder of specimens

paratypes. Scale =3 mm.

FOSSIL MOLLUSCA OF HONDEKLIP 159

Remarks

The present material is very similar to Gibbula zonata (Wood, 1828), known

from the intertidal and shallow infratidal of South West Africa—Namibia to False

Bay, Cape. The variable spiral sculpture of the present material agrees well with

the living species. The umbilicus is usually closed in adults, but occasionally

remains open in the living species, while all the fossil specimens have open

umbilici. As the fossils come from a distinctly warm-water assemblage, and

G. zonata is characteristically a cold-water species, it seems essential to regard

the fossil as a chrono-subspecies, or as a precursor species of G. zonata.

Etymology

The specific name derives from the Latin ‘patulus’, standing open, and refers

to the open umbilicus of the species.

Family Turbinidae

Bolma anoropha sp. nov.

Fig. 9

Material

Holotype. SAM—PQ-KN516, operculum, length 23,2 mm, Koingnaas

KN-1.

Paratypes. SAM—PQ-KNS517, 6 opercula, lengths 17,6, 20,0, 25,0, 26,0,

26,1 mm, one damaged, Koingnaas KN-1. SAM-—PQ-SL518, operculum, length

12,9 mm, Swartlintjies SL—20.

Description

Operculum narrowly elongate-oval, narrower at apical end; internal surface

gently convex, with numerous growth-lines. Apex on outer ventral margin.

External surface smooth, with rounded ridge in dorsal half. Operculum thickest

posterodorsally.

Remarks

None of the living southern African turbinids possesses an operculum

resembling the present material. Bolma andersoni (Smith, 1902), known from

Transkei to Zululand, possesses a more broadly oval operculum, but has similar

features, i.e. a smooth external surface, posterodorsally thickset, and with the

apex on the outer ventral margin.

Unfortunately, no turbinid shell that could be associated with these opercula

has been found in the deposits.

Etymology

The specific name, from the Greek ‘anorophos’, without a roof, alludes to

the fact that no shell has been found associated with the present opercula.

ANNALS OF THE SOUTH AFRICAN MUSEUM

160

‘WW OT = aJ899 ‘sodAje1ed suoumoods jo Ioputeulal ‘1auI09 yo] Joddn ut adAjojoH

‘pydosouv vuyjog “6 ‘314

FOSSIL MOLLUSCA OF HONDEKLIP 161

Family Turritellidae

Turritella declivis Adams & Reeve, 1850

Turritella declivis: Barnard, 1963: 167, fig. 33c. Kensley, 1973: 74, fig. 243.

Material

SAM-PQ-KN350, numerous fragments, up to 47 mm in length, maximum

diameter 13 mm, Koingnaas KN-1.

Previous records

Living: False Bay, 51 m; off Cape St. Francis, 12-112 m.

Dead: False Bay to Agulhas Bank, East London, 12-248 m.

No fossil record.

Remarks

Most of the present material is unworn, and shows the straight, even profile

of typical 7. declivis. A few fragments show slight development of a cingulum

above the suture, but none show the hollowing of the whorls seen in the

T. excavata form (see Barnard 1963: 168).

Family Crepidulidae

Calyptraea kilburni nom. nov.

Fig. 10

Calyptraea aurita striata Carrington & Kensley, 1969: 201, fig. 2c, pl. 22. Kilburn, 1980: 194.

Material |

SAM-K1433 (holotype of C. aurita striata), 8,1 x 19,6 mm, Swartlintjies,

45-50 m transgression. SAM—PQ-HB148, 2 specimens, 9,8 x 27,2 mm, one

damaged, Hondeklip HB-4, 50m Complex. USNM 400981, 2 specimens,

Fig. 10. Calyptraea kilburni. Left, dorsal view; centre and right, ventral view.

Scale = 10 mm.

162 ANNALS OF THE SOUTH AFRICAN MUSEUM

9,0 x 23,3 mm, 11,4 x 30,4 mm, numerous fragments, Hondeklip, 50 m Com-

plex.

Remarks

Kilburn (1980) correctly pointed out that the name ‘striata’, used as a

subspecific epithet by Carrington & Kensley (1969), was thrice preoccupied.

Further, C. aurita (Reeve) from Chile differed markedly from the living South

African species now named C. barnardi Kilburn, 1980. The west-coast fossil

species, while having fine radiating striae similar to the Chilean C. aurita, differs

in the structure of the septum. A new name is thus required for the Namaqualand

fossil species.

Etymology

The species is named for Dr R. N. Kilburn of the Natal Museum, foremost

southern African malacologist.

Crepidula deprima sp. nov.

Fig. 11

Material

Holotype. SAM—PQ-HBS19, 22,6 x 9,2 mm, Hondeklip Zone 4A.

Paratypes. SAM—PQ-HB520, 2 specimens, 12,3 ¥ 5,5 mm, 16,9 x 7,2 mm,

2 fragments, Hondeklip Zone 12. SAM-—PQ-HB125, 1 specimen, 14,2 x

7,4mm, Hondeklip HB-4. SAM—PQ-HB521, 3 specimens, 16,6 x 8,1 mm,

16,0x6,3 mm, 15,1X7,9mm, 2 fragments, Hondeklip Zone 4A.

USNM 400982, 1 specimen, 16,1 x 8,0 mm, 2 fragments, Avontuur A.

Description

Shell elongate-oval, strongly dorso-ventrally depressed, sides roughly paral-

lel, anteriorly evenly rounded, dorsum convex in long axis, dorsal surface with

irregular growth lines. Apex terminal on rounded posterior end. Ventral ‘shelf’

between posterior margin and start of septum. Free edge of septum evenly

concave.

Remarks

Crepidula deprima bears little resemblance in its general shape and

proportions to any of the southern or West African crepidulids. Crepidula plana

Say, 1822, of the east coast of the U.S.A. is a similarly depressed species with a

concave septum, but is much larger (up to 40 mm in length), with a narrowly

rounded posterior end and a truncate anterior end.

Etymology

The specific name is derived from the Latin ‘deprimo’, depressed, and refers

to the dorso-ventrally depressed condition of the present species.

FOSSIL MOLLUSCA OF HONDEKLIP 163

Fig. 11. Crepidula deprima. A. Holotype in dorsal and ventral view. B. Paratypes.

Scales = 10 mm.

Family Naticidae

Natica cf. andansoni (Blainville, 1824)

Fig. 12

Natica adansoni (Blainville), Nicklés, 1950: 80, fig. 117.

Material

SAM-PO-KN522, 8 specimens, ?X5,7mm, 6,3x6,4mm, 15,8 x

123 imme iop2emm 18.2 x2? mm, 20/9 x 17/8 mm, 20,4 x 19,0 mm,

21,5 X ? mm, Koingnaas KL south face.

Previous records

Living: Madeira; Cape Verde Islands; Morocco to Angola.

164

ANNALS OF THE

SOUTH AFRICAN MUSEUM

Fig. 12. Natica cf. adansoni. Scale = 10 mm.

FOSSIL MOLLUSCA OF HONDEKLIP 165

Description

Shell slightly wider than long; spire low. Umbilicus with slight ridge at outer

margin; columella callus forming convex lobe or bulge in umbilicus. Single

specimen with broad band of brown pigment stretching from above umbilicus on

to base and anterior end of outer lip.

Remarks

As all the present specimens are to some degree abraded, a definite

identification is difficult. The form of the umbilicus, however, most closely

resembles that of N. adansoni. The presence of a band of pigment around the

umbilicus and base, typical of the living species, further suggests this identifica-

tion.

Sinum concavum (Lamarck, 1822)

Fig. 13

Sigaretus concavus Lamarck, Nicklés, 1950: 81, fig. 121. Paes da Franca, 1960: 16.

Fig. 13. Sinum concavum. Scale = 5 mm.

166 ANNALS OF THE SOUTH AFRICAN MUSEUM

Material

SAM-—PQ-HBS15, longest diameter 16,8 mm, widest diameter 12,9 mm,

Hondeklip B Block.

Previous records

Senegal to Mocamedes, Angola.

Description

Shell thin, fragile. Protoconch indistinct; 4 postnatal whorls. Spire low,

barely protruding above surface of body whorl. Sculpture consisting of low

irregular flattened spiral lirae of varying widths, becoming obscure in columella

area. Lirae +50 on body whorl, difficult to count as wider lirae tend to divide

obscurely into several finer bands. Twenty-five lirae on penultimate whorl;

obscure on earlier whorls.

Family Cymatiidae

Argobuccinum casus sp. nov.

Figs 14, 15

Material

Holotype. SAM—PQ-AV523, 89,9 x 50,9 mm, apex and base damaged;

Avontuur A, 50 m Complex.

Paratypes. SAM-—PQ-AV405, 3 specimens, 66,2 x 37,8 mm, 50,7 X

28,6mm, 43,8 x 27,8 mm, Avontuur A, 50m Complex. SAM—PQ-HB339,

4 specimens, 62,7 X 37,4 mm, 40,3 x 19,4 mm, 32,5 x 18,2 mm, 31,2 x 19,4 mm,

Hondeklip Zone 12. SAM—PQ-AV524, 75,5 x 48,0 mm, 8 smaller specimens

48,8 X 21,9 mm to 29,0 x 15,8 mm, Avontuur A, 50 m Complex. USNM 400983,

2 specimens, 48,7 29,0mm, 33,3X17,9mm, Hondeklip Zone 12.

USNM 400984, 58,9 x 32,9 mm, Avontuur A, 50 m Complex. USNM 400985,

4 specimens, 27,8 X 15,1 mm to 42,0 X 22,8 mm, Avontuur A, 50 m Complex.

Non-type material. SAM—PQ-—AV525, 22 specimens, Avontuur A, 50 m

Complex. SAM—PQ-HB526, 15 specimens, Hondeklip Zone 4A.

Description

Shell high-spired, very slightly dorso-ventrally compressed, with at least

5 postnatal whorls bearing a total of about 6 varices, latter not continuous from

whorl to whorl. Profile of whorls evenly convex. Shell somewhat variable with

regard to degree of elongation; few squatter specimens resembling Argobuccinum

pustulosum (Lightfoot, 1786) in general proportions. Siphonal canal less than half

length of aperture, slightly flexed to left. Outer lip with 11 or 12 ridges, becoming

paired in older specimens. Columella smooth. Postnatal sculpture consisting of

strong flattened spiral ridges or ribbons with narrow intervening grooves, 7 or 8

on earlier whorls, 13 to 15 on body whorl with finer alternating ridges appearing.

FOSSIL MOLLUSCA OF HONDEKLIP 167

Fig. 14. Argobuccinum casus. A. Range of paratypes. Scale=20 mm. B. Holotype.

Scale = 10 mm.

Spiral ribbons becoming worn on body whorl, and seen to consist of 2 high ridges

connected and covered by outer shell layer. Spiral ridges running on to dorsum of

siphonal canal. Very faint rounded axial ridges seen in two specimens, 7 ridges

between varices, becoming obsolete on body whorl.

Remarks

Of the cymatiids currently living on the west coast, A. casus most closely

resembles Argobuccinum pustulosum (Lightfoot). This latter is said to occur in

two forms or subspecies (see Kilburn & Rippey 1982: 75), with A. pustulosum

proditor (Frauenfeld, 1865) being the west-coast form. Argobuccinum casus

generally has a narrower shell even than this west-coast form, lacks the rounded

tubercles on the spiral bands, has rounded tubercles rather than elongate ridges

on the inner surface of the outer lip, has a more clearly defined siphonal canal,

and lacks the short marginal digitiform projections of the lower outer lip. Even on

the few specimens showing faint axial ribs, where these intersect the spiral bands,

no rounded tubercles are formed as in the living species.

The present species was also compared with Pleistocene material of

A. pustulosum (Lightfoot) from the Saldanha Bay area, and again, all the

differences listed above are apparent.

168 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 15. Argobuccinum casus. Range of paratypes. Scale = 30 mm.

Cymatium parthenopeum (von Salis, 1793), living on the coasts of West and

south-east Africa, and known as a fossil from the Pliocene of Italy, possesses

fewer spiral bands, but stronger axial sculpture than A. casus.

Etymology

The specific name, the Latin for ‘adventure’, is derived from the farm-name

Avontuur, a locality for the present species as well as several others described in

this work.

FOSSIL MOLLUSCA OF HONDEKLIP 169

Family Epitoniidae

Epitonium (Gyroscala) lycocephalum sp. nov.

Fig. 16

Material

Holotype. SAM—PQ—-HB527, 14,5 x 6,6 mm, Protoconch damaged, Honde-

klip Zone 4A, 50 m Complex.

Paratypes. SAM—PQ-HB528, 2 specimens, 13,0 x 5,0 mm (apex missing),

12,9 x 4,6 mm, Hondeklip Zone 12, 50 m Complex.

Non-type material. SAM—PQ-—HB529, 15 fragments, Hondeklip A Block,

30 m Complex.

Fig. 16. Epitonium lycocephalum. A. Paratype. Scale = 3 mm.

B. Holotype. Scale = 5 mm.

170 ANNALS OF THE SOUTH AFRICAN MUSEUM

Description

Protoconch unknown; at least 6 postnatal whorls. First whorl with 12-13

well-defined rounded lamellae; second to fifth whorls each with 15 lamellae; body

whorl with 14 lamellae. At least 20 very fine spiral lines visible between lamellae,

becoming obscure near upper and lower sutures, and difficult to count. Strong

basal lira on body whorl. Suture crenulate. Outer lip varicoid.

Remarks

The present material resembles the living east-coast species E. coronatum

(Lamarck, 1816) in the possession of a basal lira, but is more slender. The

common Epitonium of the west coast, E. kraussi (Nyst), lacks both spiral lirae

and a basal lira. Epitonium lamellosum (Lamarck, 1822) of West and South

Africa is a somewhat broader shell, with fewer lamellae (9-10 on body whorl),

and lacks distinct spiral sculpture.

Etymology

The specific name, from the Latinized Greek meaning ‘wolf head’, alludes to

a low hill or koppie, Wolfkop, in the vicinity of Hondeklip Bay.

Family Thaididae

Nucella praecingulata (Haughton, 1932)

Figs 17, 18

Thais praecingulata Haughton, 1932: 48, pl. 5 (figs 6-10). Barnard, 1958: 219; 1962: 182.

Material

Syntypes. SAM-—9730, 3 specimens, Reuning’s Claim, Alexander Bay, Cape

Province, coll. J. B. Cilliers. SAM-—10598, 1 internal mould, The Point,

Namaqualand coast, Cape Province, oyster horizon, coll. E. Reuning.

SAM-9731, 4 specimens, Alexander Bay; SAM-—9742, 1 specimen; SAM-9743,

1 specimen; SAM-9745, 1 specimen; SAM-9746, 1 specimen; SAM-—9749,

1 specimen; SAM-—9750, 1 specimen; SAM-—9752, 1 specimen; SAM-—9753,

1 specimen; SAM-—9754, 1 specimen; SAM-—9934, 7 specimens; all previous

specimens from Graauweduinen, Vanrhynsdorp District coast, Cape Province,

unless otherwise stated, coll. S. Haughton.

Non-type_ material. SAM-—PQ-AV398, 18 specimens, 18-58 mm,

Avontuur A, 50m Complex. SAM—PQ-HB344, 28 specimens, 16-57 mm,

Hondeklip Zone 12. SAM—POQ-HB334, 5 specimens, 81-110 mm, Hondeklip

Zone 12. SAM—PQ-KN460, 2 specimens, 103 mm, 110 mm, Koingnaas, 50 m

Complex.

Description

Shell thaidid-like, variable, with aperture longer than spire. Protoconch of

1-13 whorls, apparently smooth (but all specimens apically somewhat eroded);

4+ postnatal whorls, generally with well-marked shoulder. Spire in relation to

body whorl variable, squat in some specimens and half length of aperture, to

FOSSIL MOLLUSCA OF HONDEKLIP ri

more slender and three-fourths length of aperture. Sculpture variable, squatter

specimens generally with stronger sculpture; early whorls with 4 strong spiral

bands and numerous axial lamellae between bands and upper and lower suture

lines, giving cancellate appearance. Spiral bands increasing on to body whorl,

with intermediaries appearing in some specimens; up to 20 major spiral bands,

running on to base. Cancellation usually lost on body whorl. Specimens having

weaker spiral sculpture usually with shoulder marked by single somewhat

stronger band; sculpture in these specimens often becoming obsolete on body

whorl. Columella smooth, concave in lower half. Base rounded, fasciole strong.

Inner surface of outer lip generally plicate, folds becoming obsolete in larger

specimens.

Remarks

Barnard (1962: 182) noted that Nucella praecingulata resembles both Nucella

cingulata (Linnaeus, 1758) and N. squamosa (Lamarck, 1816), but reaches a

much greater size than either of these living species.

Haughton (1932) quotes a note from Tomlin, in which the latter suggests the

name ‘praecingulata’, and also observes the similarity to N. cingulata and

N. squamosa. Haughton, while not presenting a formal description, noted the

variability of the species, did not designate a holotype, but saw all the material

listed above under ‘Syntypes’ (Fig. 17).

In spite of being highly variable, large specimens of N. praecingulata

(approx. 100 mm) are easily identified, being the only whelk of such dimensions

in the west-coast deposits. Smaller specimens, in the size range of N. cingulata

and N. squamosa, however, are more difficult to assign.

In the majority of smaller specimens of N. praecingulata the body whorl and

at least the two earlier whorls have a faint but distinct shoulder on the upper part

of the whorl. The strength of the spiral bands varies. In those specimens in which

the spiral bands are few and strong, the resemblance to N. cingulata is

unmistakable, even to the possession of five axial lines between the spiral bands,

leading to the squamose appearance sometimes seen in N. cingulata. In these

specimens, the whorls tend to have an evenly rounded profile. In those specimens

in which the spiral bands are finer and more numerous, the axial lines lead to a

cancellate appearance very similar to N. squamosa (see Fig. 19). In these, the

slightly angular whorl profile is more apparent, but even here, evenly rounded

whorls are encountered.

It is suggested that with the change to a colder regime on the west coast

during the Pleistocene, Nucella praecingulata gave rise to two forms, Nucella

cingulata and N. squamosa, both being intertidal to shallow infratidal forms,

neither attaining the dimensions of the ancestor, and with N. cingulata being

confined to the colder waters of the west coast to False Bay, while N. squamosa,

perhaps with greater temperature tolerance, extends to the Transkei coast.

Nucella praecingulata has not been recorded concurrently with N. cingulata

or N. squamosa, a further. indication of the form’s possible ancestral role.

ANNALS OF THE SOUTH AFRICAN MUSEUM

173

FOSSIL MOLLUSCA OF HONDEKLIP

‘WU OF = a[B9S

‘sadAyuds

d ‘WV ‘vojnsuisavid vyaonN *L{ “314

174 ANNALS OF THE SOUTH AFRICAN MUSEUM

| “WI OF = aTROS

usuisads jJa] JOMO] UO s]JaYS I9}skO d}ON ‘sudUIDads dipyopuoyH “gq “Wu Y¢ = aJeOg ‘suoUTIDads IMNJUOAYV “VY vinsunavdd vIJ2NN ‘y ‘Biq

iv,

©)

ap;

fy,

(O}

iZ2)

ANNALS OF THE SOUTH AFRICAN MUSEUM

176

"WU (¢ = BJeIS “Avg dS|[eq pue e[NsUIUDg Oded WoO susUIDSds yUDD0yY «“vsowNnbs DYaINN

“6. ‘SI

FOSSIL MOLLUSCA OF HONDEKLIP t7

Thais arenae sp. nov.

Fig. 20

Material

Holotype. SAM—PQ—KN530, almost complete specimen, length 52,5 mm,

greatest width 27,3 mm, Koingnaas KL south face.

Non-type material. SAM—PQ-—KN307, 3 fragments (body whorls, somewhat

worn), Koingnaas KL south face.

Description

Shell with at least 4 postnatal whorls; aperture longer than spire. (The

holotype, an almost complete specimen, lacks the tip of the spire.) Whorls with

distinct shoulder in upper half. Columella smooth; short anterior canal flexed

slightly to left. Second extant whorl with 14 axial ribs, spiral sculpture worn away.

Third whorl with 11-12 axial ribs, 2 strong spiral keels, upper forming shoulder;

axial ribs rounded at intersections with strong keels; fine spiral sculpture over

entire whorl, 11-12 lirae between suture and shoulder, 9-10 below shoulder.

Fig. 20. Thais arenae. Holotype. Scale = 10 mm.

178 ANNALS OF THE SOUTH AFRICAN MUSEUM

Body whorl with 10 axial ribs, becoming obsolete on outer lip; 5 strong keel-like

spiral bands; 12-13 fine lirae between suture and shoulder; 10-12 fine lirae

between each strong spiral band; 4 fine lirae between lowest band and base.

Remarks

The present species bears some resemblance to Thais capensis (Petit, 1852),

especially in the fine spiral lirae, but the latter is a relatively squatter species

having somewhat stronger tubercles, especially on the body whorl. Thais

haemastoma (Linnaeus, 1767), especially in the form occurring at present at the

southern end of its West African range, has a much squatter shell bearing variable

rounded tubercles on the shoulder, rather than the rounded ribs of T. arenae.

Lathyrus armatus A. Adams (see Nicklés 1950: 106) from the Azores, the

Canary Islands, and Senegal, has a similar overall shape and sculpture. Although

the holotype of 7. arenae shows some wear on the axial ridges and tubercles,

these are still more rounded than in the West African species. Also, the anterior

canal is more defined in the latter species.

Etymology

The specific name, from the Latin, of a sandy place, refers to the coarse

sands of the type-locality at Koingnaas KL south face.

Family Muricidae

Ocenebra bonaccorsii (Carrington & Kensley, 1969)

Tritonalia bonaccorsii Carrington & Kensley, 1969: 196, pl. 19.

Material

SAM-PQ-HB531, 39,1 x 22,0 mm, Hondeklip A Block, 30 m Complex.

SAM-—POQ-HB532, 2specimens, 8,4x4,8mm, 8,1x*4,0mm, Hondeklip

Zone 12, 50 m Complex. SAM—POQ-HB152, 22,1 x 10,8 mm (protoconch miss-

ing), Hondeklip Zone 3.

Remarks

The PQ—HBS531 specimen above is both larger and better preserved than the

holotype. The description of the species can thus be supplemented:

Protoconch of 2—23 whorls; 43 postnatal whorls. First two postnatal whorls

bicarinate; third whorl with upper carina stronger than lower; postnatal whorl 2

with 4 fine spiral lirae between suture and upper carina, 2-3 fine lirae between

upper and lower carina; 2-3 lirae between lower carina and suture. Postnatal

whorls 3 and 4 with 8-9 lirae between upper suture and upper carina; 4—5 lirae

between carinae; 4 lirae between lower carina and lower suture. Body whorl with

13-15 lirae between upper suture and strongest spiral ridge; 8 strong spiral

ridges, 9 fine lirae between uppermost strong ridge and next, number of fine lirae

decreasing anteriorly to two or three between stronger ridges. Inner surface of

outer lip with 8 rounded ridge-teeth.

FOSSIL MOLLUSCA OF HONDEKLIP 179

Ocenebra petrocyon sp. nov.

Figs 21

Material

Holotype. SAM—PQ—-HB533, 9,9 x 4,8 mm, Hondeklip, 30 m Complex.

Paratypes. SAM—PQ-HB534, 10,6 5,7 mm, 8,6 x 4,6 mm; Hondeklip,

30 m Complex. SAM—PQ-HB535, 5 damaged specimens, 20 juvenile speci-

mens; Hondeklip A Block, 30 m Complex.

Fig. 21. Ocenebra petrocyon. WHolotype at left, remainder of specimens paratypes.

Scale = 3 mm.

Description

Protoconch of 123 smooth whorls; 4 postnatal whorls. Aperture subequal to,

or slightly longer than, spire. Sculpture consisting of rounded axial ribs and spiral

lirae. Axial ribs: 11—12 on first whorl; 11-12 on second whorl; 11 on third whorl;

11 on body whorl. Spiral lirae: 4 on first whorl; 5 on second whorl; 5 on third

whorl; 5 (with 1 or 2 very fine intermediates) on fourth whorl; 18-20 on body

whorl! (outer lip), including several finer intermediates; lirae running on to

anterior canal. Lirae somewhat broadened at intersection with axial ribs, but

never becoming bead-like. Narrow, smooth columella callus present. Inner

surface of outer lip with 8—10 faint spiral ribs.

Remarks

Ocenebra petrocyon most closely resembles O. purpuroides (Reeve, 1845)

(with which it was collected), especially in general shape and body-whorl

sculpture. The new species differs from O. purpuroides in that the upper two or

three whorls are not cancellate; in having more spiral lirae per whorl; in having

spiral lirae evenly spaced between the upper and lower suture lines (OQ. pur-

180 ANNALS OF THE SOUTH AFRICAN MUSEUM

puroides has a distinct shoulder and no lirae between shoulder and the upper

suture line); and in having the axial ribs more rounded (almost carinate in

O. purpuroides).

Etymology

The specific name is the Latinized form of the Greek for ‘dog stone’, and

alludes to the farm Hondeklip.

Trophon carringtoni sp. nov.

Fig. 22

Latiaxis sp. Carrington & Kensley, 1969: 195, pl. 19.

Material

Holotype. SAM—PQ-AV536, 42,5 x 22,5 mm, Avontuur A, 50 m Complex.

Paratypes. SAM-—PQ-AV537, 2 specimens, 37,1 X 22,4 mm, 40,8 x

I toe | B

Fig. 22. Trophon carringtoni. A. Holotype. Scale=10 mm. B. Range of paratypes.

Scale = 30 mm.

FOSSIL MOLLUSCA OF HONDEKLIP 181

24,5 mm, Avontuur A, 50 m Complex. SAM-K1447, 62,5 x 37,0 mm, outer lip

missing, Strandfontein, 45-50m Complex, coll. A. J. Carrington.

SAM-PO-AV538, 8 specimens (all with eroded apices), 58,0 x 32,0 mm,

53,0 x 28,0 mm, 42,5 23,5 mm, 40,2 x24,0mm, 40,0 x 23,5 mm, 38,2 x

22,0 mm, 34,8 x 20,0 mm, 20,0 13,0 mm, Avontuur A, 50m Complex.

USNM 400986, 3 specimens (all with eroded apices), 42,0 x 24,8 mm,

43,0 x 23,3 mm, 38,0 x 25,9 mm, Avontuur A, 50 m Complex.

Description

Shell of 4-5 postnatal whorls; aperture longer than spire. Whorls with single

prominent keeled shoulder. Early whorls with irregular axial ridges, giving

appearance of cancellate sculpture below keel, forming more or less well defined

squamae where keel intersected. In later whorls, 6—8 axial ridges only faintly

indicated above keel, stronger below keel, especially on body whorl, forming

faint swelling at intersection with keel. Columella smooth, with callus developed.

Inner surface of outer lip with 5 tubercles below carinal notch. Short anterior

canal flexed to left.

Remarks

In the original mention of this species (Carrington & Kensley 1969: 195) the

resemblance to Latiaxis tortilis (=L. nakamigawai io Kilburn, 1974) was

mentioned. This comparison no longer stands, as the present species is more

slender, does not have a strongly flexed anterior canal, and appears to lack spiral

sculpture, other than the carina.

The general shape and sculpture are reminiscent of members of the

Trophoninae, but resemble nothing recorded in the southern African fauna.

Etymology

The species is named for Mr A. J. Carrington of De Beers Consolidated

Diamond Mines.

Family Buccinidae

Burnupena rogersi sp. nov.

Fig. 23

Material

Holotype. SAM—PQ-HB539, 58,1 x 34,5 mm (body whorl plus 22 whorls,

remainder of spire missing), Hondeklip Zone 3, 50 m Complex.

Paratypes. SAM-PQ-HBS540, 14 specimens, 48.3 x 30,5 mm to 14,7 x

8,7 mm (larger specimens with outer lip and/or body whorl damaged and spire

missing; smaller specimens complete). Hondeklip Zone 3, 50m Complex.

USNM 400987, 5 specimens, 38,4 x 24,3 mm to 15,7 x 8,9 mm, Hondeklip

Zone 3, 50 m Complex.

182 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 23. Burnupena rogersi. Holotype at upper left corner, remainder of specimens

paratypes. Scale = 20 mm.

FOSSIL MOLLUSCA OF HONDEKLIP 183

Description

Shell about 1,5 times longer than wide. Protoconch unknown; at least

5 postnatal whorls. Profile of earlier whorls evenly convex; of last 3 whorls with

distinct, rounded shoulder. Sculpture of fine spiral bands separated by narrow

incised grooves; 7—8 on third from last whorl, 14-15 on penultimate whorl, 35—42

on body whorl, extending on to outer surface of anterior canal, flattened bands

becoming obscure, divided by shallow incised lines on lower body whorl. No axial

sculpture apart from faint growth lines, and several aperture margin ridges in

larger specimens. Narrow gutter formed at suture line on last whorl, becoming

closed on earlier whorls. Inner surface of outer lip with about 10 faint ridges.

Base narrowly umbilicate; columella lacking pleats; with weak callus; anterior

canal short, flexed slightly to left.

Remarks

The present material, showing a strongly sunken suture, bears some

resemblance to the genus Babylonia. Burnupena rogersi, however, is plumper

than the two South African species of Babylonia, neither of which possesses fine

spiral sculpture.

While the generic position of the present material cannot be established in

the absence of radula, soft parts, and operculum, the species does bear a general

resemblance to species of Burnupena, and especially to B. papyracea (Bruguiére,

1789), which sometimes does show a fairly sunken suture.

Etymology

The species is named for Dr John Rogers of the Department of Geology,

University of Cape Town.

Burnupena aestus sp. nov.

Figs 24, 25

Material

Holotype. SAM-PQ—AV541, 69,9 x 31,0 mm, Avontuur A, 50 m Complex.

Paratypes. SAM-—PQ-AV542, 3 specimens, 62,4 x 30,3 mm (apex and outer

lip damaged), 67,1 x 29,9 mm, 68,4 x 32,0 mm, Avontuur A, 50 m Complex.

SAM-—PO-HB543, ?66,6 X 29,7 mm (apex damaged), Hondeklip Zone 12, 50 m

Complex. SAM—K1445, 61,3 x 27,0 mm (body whorl damaged), Swartlintjies,

coll. A. J. Carrington, 1967. SAM-K1446, 67,1 x 34,6 mm (body whorl dam-

aged), Swartlintjies, coll. A. J. Carrington, 1967.

Description

Shell elongate, relatively slender, spire longer than aperture. At least

5 postnatal whorls. Profile of whorls convex, with slight hollowing just below

suture. Columella smooth, basally flexed slightly to left, with rounded ridge just

184 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 24. Burnupena aestus. Holotype. Scale = 10 mm.

below suture; base rounded. Inner surface of outer lip with 20-22 rounded ridges.

Sculpture consisting of fine spiral lirae, occasionally doubled on lower whorls;

spiral lirae narrowly rounded to carinate. Second whorl with 10-12 lirae; third

with 12-15; fourth with about 20; body whorl with up to 56 lirae.

Remarks

In general proportions, the present species most resembles Afrocominella

capensis (Dunker, 1844), but lacks the obscure axial sculpture of the early whorls

typical of this species. The spiral lines are finer and more numerous than in

A. capensis, while the aperture is generally subequal in length to the spire. The

present species attains a much larger size than either A. capensis or Burnupena

papyracea, the other species to which it bears some similarity. This latter species

is much squatter, with fewer spiral lines.

FOSSIL MOLLUSCA OF HONDEKLIP 185

Fig. 25. Burnupena aestus. Paratypes. Scale = 20 mm.

There is a possibility of confusing small specimens of this species with large

specimens of Triumphis dilemma (see below). The two species may, however, be

easily separated on the following features: columella smooth in B. aestus, with

2-4 tubercles in T. dilemma; aperture longer than spire in 7. dilemma, shorter

than spire in B. aestus; outer lip frequently thickened in T. dilemma, unthickened

in B. aestus; spiral cords on body whorl and anterior canal often doubled,

186 ANNALS OF THE SOUTH AFRICAN MUSEUM

normally rounded to carinate in B. aestus, spiral cords single, rounded to tabulate

in T. dilemma.

Etymology

The specific name is derived from the Latin ‘aestus’, of the surf, and refers to

the probable habitat of the species.

Triumphis dilemma Kilburn & Tankard, 1975

Fig. 26

Triumphis dilemma Kilburn & Tankard, 1975: 200, fig. 10.

Material

Holotype. SAM-—K4565, 16,3 x 10,4 mm, Langebaan, Early Pleistocene,

coll. A. J. Tankard.

Non-type material. SAM—PQ-HB544, 10 specimens, 27,2 x 17,8 mm to

39,1 x 22,0 mm, Hondeklip Zone 4A, 50m Complex. SAM—PQ-AV545,

6 specimens, 25,3 X 15,0 mm to 33,0 x 19,9 mm, Avontuur A, 50 m Complex.

SAM-K4862, 4 specimens (2 damaged), 24,9 x 13,2 mm, 30,9 x 16,3 mm, Som-

naas SN4, Hondeklip Bay, coll. A. J. Tankard.

Previous records

Early Pleistocene, Langebaan, 9,5 m beach.

Remarks

With a wider range of material available, the original description can be

supplemented: parietal area with single tooth in most cases; 2 teeth (as in

holotype) only very occasionally seen. Outer lip only occasionally not thickened;

with 10-17 internal ridges. Columella with 2—5 basal tubercles, 2 most common.

Spiral cords rounded to tabulate. Aperture longer than spire, generally not as

constricted as in holotype. Kilburn & Tankard (1975) mentioned two damaged

specimens considerably larger than the holotype, about which they had

reservations. With good material in this size range, it can be seen that the two

earlier specimens are in fact this species, and that the higher counts of spiral cords

are typical of the species.

Family Nassariidae

Bullia annulata (Lamarck, 1816)

Bullia annulata: Barnard, 1959: 127, figs 25f, 27e; 1962: 189.

Bullia magna Haughton, 1932: 46, pl. 5 (figs 1, 4, 5). Barnard, 1959: 129; 1962: 182, 189.

Material

SAM-—9897 (syntypes of B. magna), 2 incomplete specimens, Graauwe-

duinen, Vanrhynsdorp District coast, Cape Province. SAM—7998, (syntypes of

B. magna), 16 incomplete specimens, Doornbaai, south of Olifants River mouth,

FOSSIL MOLLUSCA OF HONDEKLIP 187

Fig. 26. Triumphis dilemma. Range of material from Hondeklip, 50 m

Complex. Scale = 30 mm.

Cape Province. SAM-—K4863, 1 specimen, 67,6 X 32,2 mm, Swartlintjies SL2,

Hondeklip Bay, coll. A.J. Tankard. SAM-—PQ-HB210, 3 specimens,

63,7 X 30,1 mm, 68,7 X 32,2 mm, 55,6 X ? mm (body whorl incomplete), Honde-

klip Zone 12, 50 m Complex. SAM—PQ—AV233, 7 specimens (3 incomplete),

64,3 x 30,7 mm, 60,8 x 30,8 mm, 59,9 x 30,3 mm, 51,2 x 26,9 mm, Avontuur,

50 m Complex. SAM-PQ-AV588, 12 specimens, Avontuur, 50 m Complex.

SAM-POQ-HB613, 1specimen, Hondeklip Zone 4A, 50m Complex.

SAM-—PO-HB8839, 2 specimens, Hondeklip Zone 12, 50 m Complex.

188 ANNALS OF THE SOUTH AFRICAN MUSEUM

Remarks

Barnard (1962) expressed the view that Bullia magna was merely a worn and

fossilized B. annulata. All the syntypic material is extremely worn, only one

specimen showing faint spiral sculpture.

The present material, while friable, shows good preservation of detail, and

possesses the distinctive spiral lirae characteristic of B. annulata.

While only one of the specimens from the 50 m Complex shows the

development of a shoulder on the upper whorl, approaching the condition in

typical B. annulata, Barnard (1962) does note that some recent specimens show

only a weak development of the shoulder. The present material removes any

doubt that B. magna really is B. annulata.

The Hondeklip specimens occurred in a very fine sand, in a layer dominated

by specimens of the bivalve Dosinia sicarisinus sp. nov., often with both valves in

life position.

Distribution

Living B. annulata occurs from Saldanha Bay to Mozambique, from the low-

tide mark, but more usually infratidally, to 100 m (Barnard 1958; Kilburn &

Rippey 1982).

Fossil B. annulata has been recorded from Langebaan, Milnerton, Swart-

kops, Redhouse and Coega (Barnard 1959, 1962).

Nassarius cf. kochianus (Dunker, 1846)

Nassarius kochianus (Dunker), Kilburn & Rippey, 1982: 100, 213, pl. 23 (fig. 8).

Nassa kochiana: Barnard, 1959: 104, figs 22b, 23b; 1962: 189.

Material

SAM-PQ-KN546, 3 specimens, 10,96,1 mm, 9,1x5,1mm, 7,6

4,1 mm, Koingnaas KL south face.

Previous records

Living: False Bay to Transkei, intertidal to shallow infratidal.

Fossil: Algoa Bay.

Remarks

With slightly fewer axial ribs (13-14 per whorl) than is usual for

N. kochianus (15-18), the present material, however, does have only a single

columella nodule, and in profile more closely resembles this species than it does

N. signatus (Dunker, 1847). More specimens are needed to decide the range of

variation in the number of axial ribs, before a more definite identification can be

attempted.

FOSSIL MOLLUSCA OF HONDEKLIP 189

Family Fasciolariidae

Fasciolaria dinglei sp. nov.

Figs 27, 28

Fasciolaria sp. Carrington & Kensley, 1969: 193, pl. 18 (?partim).

Material

Holotype. SAM—PQ-AV547, 101 x 41 mm (apex and upper whorls eroded),

Avontuur A, 50 m Complex.

Paratypes. SAM—PQ-AV548, 3 body whorl fragments, 7 specimens,

39,4 x 19,8 mm to 83,4 x 40,3 mm, Avontuur A, 50 m Complex.

Description

Shell with aperture longer than spire. Four to five postnatal whorls. Profile of

whorls evenly convex. Third whorl with 12-13 fine spiral lirae; fourth with 15-

16 lirae; body whorl with 50-55 lirae running on to anterior canal; lirae on upper

Fig. 27. Fasciolaria dinglei. Holotype. Scale = 10 mm.

190 ANNALS OF THE SOUTH AFRICAN MUSEUM

body whorl near suture having finer intermediates appearing; lirae on base of

body whorl becoming coarser and broader. Columella with 3 pleats, sometimes

obscured in large specimens; low ridge on upper columella just below suture.

Anterior canal about half length of rest of aperture, flexed to left. Inner surface of

outer lip with up to 25 rounded ridges.

Remarks

Of the two specimens of Fasciolaria figured by Carrington & Kensley (1969,

pl. 18c), the smaller may well be the present species. The larger, however, is

undoubtedly a large Burnupena, lacking as it does an elongate anterior canal.

Carrington & Kensley (1969) noted the similarity of their specimen to

Fasciolaria lugubris (Reeve, 1847). This similarity in overall proportions and

general sculpture is again noted, but several differences easily separate the two

species. The most distinctive feature of the new species is the presence of three

pleats on the columella (F. lugubris is unpleated). The spiral sculpture of

F. dinglei is much finer and the lines more numerous (up to 55 on the body whorl)

than in F. lugubris (up to 30 on body whorl, and more raised).

Fig. 28. Fasciolaria dinglei. A. Range of paratypes. Scale=10 mm. B. Range of paratypes.

Scale = 30 mm.

FOSSIL MOLLUSCA OF HONDEKLIP 191

Etymology

The species is named for Professor Richard V. Dingle, Department of

Geology, University of Cape Town.

Family Olividae

Melapium hawthornei sp. nov.

Fig. 29

Material

Holotype. SAM—PQ-KNS549, 27,5 x 24,0 mm, Koingnaas KN-1.

Paratypes. SAM—PQ-KN550, 12 specimens, 23,4 x 16,7 mm to 26,1 x

20,1 mm, Koingnaas KN-1, 50m Complex. USNM 400988, 4 specimens,

23,4 x 17,1 mm to 27,0 x 18,9 mm, Koingnaas KN-1, 50 m Complex.

Non-type material. SAM—PQ-KNS551, 9 specimens, all worn and damaged,

Koingnaas KL south face. SAM—PQ—-KN366, 16 specimens, all damaged,

Koingnaas KN-—1, 50 m Complex.

Description

Shell 1,2 times longer than wide, of 4 postnatal whorls. Spire somewhat

sunken, but still protruding. Body whorl globose, profile evenly convex, with very

faint, shallow, spiral lines, and irregular axial growth lines. Parietal callus

extending from body-whorl suture in evenly convex line to lower columella.

Posterior canal short, but with distinct notch; anterior canal flexed to left; lower

columella narrowed, with fasciole strong but barely visible in apertural view;

outer lip evenly convex to notch at anterior canal.

Remarks

Two living and one fossil species of Melapium have been recorded from

southern Africa, all of which differ from the present species.

Melapium lineatum (Lamarck, 1822), known from Still Bay to Zululand in

depths of 30-160 m, is of comparable size to M. hawthornei, but possesses a more

elevated spire; also, the lower columella is broader, with the fasciole being strong

and visible in apertural view. Melapium lineatum also has a distinct concavity just

above the fasciole origin, and below the bulge of the body whorl; M. hawthorne,

by contrast, is evenly convex in this region.

Melapium elatum (Schubert & Wagner, 1829), from deeper water off Natal

and Mozambique, has a much larger shell, with the lower columella very broad

and with a very strong fasciole, and a distinct shoulder on the upper body whorl.

Melapium patersonae Bullen Newton, 1913, from the Neogene of Bredas-

dorp to Alexandria, Cape Province, is also a large shell, up to 70 mm in length,

has a sunken spire, a well-developed posterior canal with the outer lip reaching

posteriorly above the spire, and a very broad columella with a very strong

carinate fasciole.

ANNALS OF THE

SOUTH AFRICAN MUSEUM

Fig. 29. Melapium hawthornei. Holotype at upper left corner, remainder of specimens paratypes. Scale = 20 mm.

FOSSIL MOLLUSCA OF HONDEKLIP 193

Etymology

The species is named for Mr J. B. Hawthorne of De Beers Consolidated

Diamond Mines.

Pseudoliva lutulenta sp. nov.

Fig. 30

Material

Holotype. SAM—PQ-KNS552, 31,6 x 22,7 mm, Koingnaas KN-1.

Paratypes. SAM—PQ-KN553, 8 complete specimens, 20,7 x 15,8 mm to

29,2 x 21,2 mm, Koingnaas KN-1. USNM 400989, 3 complete specimens,

23,1 < 18,1 mm to 28,3 x 21,5 mm, Koingnaas KN-1.

Non-type material. SAM—PQ-KN373, 13 specimens (damaged), Koingnaas

KN-1.

Description

Shell globose, thick-walled, longer than wide, of 4 whorls. Spire short. Whor!l

profiles evenly convex; suture somewhat incised. Columella callus smooth,

convex, reaching posteriorly to level of suture, outline evenly convex on inner-lip

body whorl; parietal node somewhat elongate, forming shallow groove-like

posterior canal. Base with strong furrow running from shallow sinus in anterior

outer lip, around lower body whorl, to midpoint of columella callus; second, less

clearly defined groove anterior to strong groove, neither groove marked on inner

surface of outer lip.

Remarks

Three major differences separate the present material from the apparently

very similar living Pseudoliva crassa (Gmelin), known only from Angola, and

figured and described as P. plumbea (Chemnitz) by Nicklés (1950: 107, fig. 189).

The living species possesses two strong well-defined grooves on the anterior body

whorl, the upper of which is marked by a narrow ‘excroissance’ on the internal

surface of the outer lip. In the present species there is one strong groove and a far

less well-defined more anterior groove. Neither groove is reflected by any

structure on the internal surface of the outer lip. The columella callus in P. crassa

is concave, and posteriorly narrowed to the posterior canal. In P. /utulenta the

columella callus is convex, and posteriorly broadly rounded.

Bohm (1926) described two species of Pseudoliva from the Cretaceous of

Bogenfels, South West Africa—Namibia, neither of which bears much resem-

blance to the present species. Pseudoliva thielei B6hm shows enormous columella

callus development, and has several spiral ridges on the shell base; P. /eutweini

Bohm is a much larger species than P. Jutulenta, with a massive upper columella

callus.

194

ANNALS OF THE SOUTH AFRICAN MUSEUM

20 mm.

der of specimens paratypes. Scale

, remain

Fig. 30. Pseudoliva lutulenta. Holotype at upper left corner

FOSSIL MOLLUSCA OF HONDEKLIP 195

Etymology

The specific name is derived from the Latin ‘lutulentus’, covered with mud,

and alludes to the fine sediments in which the species was found.

Family Turridae

Drillia tempestae sp. nov.

Fig. 31

Material

Holotype. SAM—PQ-HBS554, 12,1 5,1 mm, Hondeklip Zone 3, 50m

Complex.

Paratypes. SAM-—PQ-HB555, 10 specimens, 6,3 x3,4 mm to 19,4~x

7,/mm, Hondeklip Zone 3, 50 m Complex. USNM 400990, 10 specimens,

6,3 x 2,0 mm to 16,4 x 7,2 mm, Hondeklip Zone 3, 50 m Complex.

Non-type material. SAM—PQ-SL260, 1 specimen, 19,9 x 8,4 mm, Swart-

lintjies River, spoil heap. SAM—PQ-—HB556, approx. 50 specimens, Hondeklip

Zone 3, 50 m Complex.

Description

Protoconch of 13 smooth whorls; 63-7 postnatal whorls. Spire 1,3—1,5 times

length of aperture. Sculpture consisting of rounded axial ribs and fine spiral lirae;

axial ribs slightly oblique, occasionally perpendicular, rounded, with distinct

shoulder well below upper suture line, reaching lower suture. Postnatal whorl 3

with 9 axial ribs, 6—9 spiral lirae; ribs increasing to 11 on body whorl, lirae to

10-18 on penultimate whorl; outer lip with 20—24 spiral lirae running on to outer

surface of siphonal canal. Lirae equal in width on spire, becoming broader on

body whorl and siphonal canal. Anal sinus in outer lip shallow. Columella callus

smooth; anterior siphonal canal short, well marked.

Remarks

The present species resembles Drillia caffra (Smith, 1882) in overall

proportions, but differs in details. The earlier species has more axial ribs per

whorl (11-12 on early whorls, 15-18 on later whorls), more spiral lirae (6 on

early whorls, up to 30 on last whorl), and possesses a deep anal sinus.

As generic placement in the Turridae is so dependent on radula structure,

the present material is placed in Drillia purely because of an overall resemblance

to some species in that genus.

Etymology

The specific name, from the Latin for ‘of a storm’, alludes to the storm

conditions that probably caused some of the Hondeklip fossil accumulations.

196 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 31. Drillia tempestae. A. Holotype. Scale=2 mm. B. Range of paratypes.

Scale = 5 mm.

FOSSIL MOLLUSCA OF HONDEKLIP 197

Family Terebridae

Terebra canisaxi sp. nov.

Fig. 32

Material

Holotype. SAM—PQ-HB557, 21,4 6,5 mm, Hondeklip Zone 12, 50 m

Complex.

Paratypes. SAM-—PQ-HB558, 10 specimens, 15,2x4,9 mm to

23,1 X 6,9 mm, Hondeklip Zone 12, 50 m Complex. USNM 400991, 10 speci-

mens, 15,7 X 5,6 mm to 23,1 x 6,4 mm, Hondeklip Zone 12, 50 m Complex.

Fig. 32. Terebra canisaxi. Holotype at upper left corner, remainder of specimens paratypes.

Scale = 10 mm.

198 ANNALS OF THE SOUTH AFRICAN MUSEUM

Non-type material. SAM—PQ-HB559, 4 specimens, 14,6x4,9 mm to

17,7X5,7 mm, many fragments, Hondeklip Zone 12, 50m Complex.

SAM-—PQ-HB331, 30 specimens, Hondeklip Zone 12, 50 m Complex.

Description

Shell of protoconch of 23 smooth whorls, plus 7 postnatal whorls. Postnatal

whorl sculpture consisting of strong, rounded, axial ribs running from upper to

lower suture, sometimes alternating from one whorl to next, sometimes

continuous. Postnatal whorl 1 with 9 ribs, following 6 whorls each with 8—9 ribs,

body whorl with 9-10 ribs. Anterior canal short, open, flexed to left. Base

rounded. Columella lacking pleats, smooth; low fasciole present. Spiral groove

absent.

Remarks

This species closely resembles Terebra lightfooti Smith, 1903, recorded alive

from Saldanha Bay to the Agulhas Bank to a depth of 62 m. Barnard (1969: 596)

described this species as having eight postnatal whorls. Terebra lightfooti is a

marginally more slender species than T. canisaxi, with slightly more numerous

axial ribs (11-12 on last three whorls, as against 8—9 in T. canisaxi).

Etymology

The specific name is the Latinized form of ‘dog stone’, a direct translation of

the name of the type-locality, Hondeklip.

Class BIVALVIA

Family Glycymeridae

Glycymeris fulleri sp. nov.

Figs 33, 34

Material

Holotype. SAM-—PQ-SLS561, both valves, 58,0 x 61,5 mm, Swartlintyies

SL-20, 50 m Complex.

Paratypes. SAM—PQ-SL562, both valves, 38,0 x 40,5 mm, Swartlintjies

SL-20, 50 m Complex. SAM—PQ-KNS560, right valve, 40,2 x 44,1 mm, Koing-

naas KN-1, 50 m Complex. SAM—PQ—-KN563, 3 left valves, 25,0 x 26,8 mm,

36,2 X 40,0 mm, 46,3 x 46,0 mm (distorted), 1 right valve, 37,1 x ? mm, Koing-

naas KN-1, 50m Complex. USNM 400992, 2 left valves, 23,6 x 24,9 mm,

42,9 x ? mm, 1 right valve, 34,8 x ? mm, Koingnaas KN-1, 50 m Complex.

Description

All material in poor condition, with external surfaces exfoliating, and shape

frequently deformed. Shell slightly inequilateral, posterior margin slightly more

pointed than anterior margin; slightly wider than high. Sculpture of about

40 radiating, apparently flattened ribs, becoming obsolete in anterior and

FOSSIL MOLLUSCA OF HONDEKLIP 199

Fig. 33. Glycymeris fulleri. WHolotype, external and internal views of both valves.

Scale = 30 mm.

posterior part of shell. Ventral margin faintly crenulate in external view,

internally more strongly crenulate. Anterior and posterior adductor muscle scars

situated on faint ledge. Hinge area with teeth discontinuous below umbo, divided

into two groups, each of 7—9 teeth.

Remarks

Glycymeris fulleri resembles the living southern African east-coast species

G. queketti (Sowerby, 1897) in having the muscle scars situated on faint ledges,

but is a smaller and more inequilateral species. Barnard (1962: 183) mentions two

specimens referred to G. queketti from Skulpfontein Point, Hondeklipbaai area,

by Krige (1927), but questioned the identification.

200 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 34. Glycymeris fulleri. Range of paratypes. Scale = 20 mm.

Glycymeris connollyi Tomlin, 1923, known living from Table Bay to Natal, is

smaller than the present species, and more trigonal in outline.

Glycymeris borgesi (Cox, 1946) (= G. africana Cox, 1939), known from the

Neogene of Ysterplaat, Cape Province (Tankard 1975a) as well as from the

Alexandria Formation in the Port Elizabeth area, is a much larger species (up to

100 mm diameter), with the hinge teeth in a continuous band.

Glycymeris ovata (Broderip), recorded from the Pliocene and Pleistocene of

Chile by Herm (1969) also shows the tooth row of the hinge divided into two

parts, but this species is less markedly inequilateral and somewhat more trigonal

than the present species.

Etymology

The species is named for Professor A. O. Fuller, Department of Geology,

University of Cape Town.

FOSSIL MOLLUSCA OF HONDEKLIP 201

Family Ostreidae

Ostrea cf. subradiosa Bohm

Fig. 35

Ostrea digitilina, non Dubois, Bohm & Weisfermel, 1913: 61, fig. 1a, pl. 8 (fig. 2).

Ostrea subradiosa Bohm, 1926: 56, text table A (figs 1-3).

Material

SAM-PQ-AV571, one left valve, 93 x 83 mm, Avontuur A, 50 m Complex.

Fig. 35. Ostrea cf. subradiosa. External and internal view of left valve. Scale = 10 mm.

Remarks

The present specimen has been compared with material in the South African

Museum (SAM-—K4936) from Bogenfels, as well as with the original descriptions

and figures. Neither of the figures of Bohm & Weisfermel (1913) shows a row of

pits in the shell margin close to the hinge area. The figure of BOhm (1926),

however, shows such a row of pits. The present specimen, while having the strong

radiating external ribs seen in O. subradiosa, lacks these marginal pits. A more

definite identification on the basis of a single specimen is not possible.

202 ANNALS OF THE SOUTH AFRICAN MUSEUM

Family Isognomonidae

Isognomon gariesensis sp. nov.

Fig. 36A

Material

Holotype. SAM—PQ-HB564, incomplete right valve, upper (dorsal) part of

valve missing, greatest length 158mm, greatest width (across hinge area)

90,9 mm, Hondeklip A Block, 50 m Complex.

Paratypes. SAM—PQ-HB263, 1 incomplete left valve, Hondeklip A Block,

50 m Complex. SAM—PQ-KN565, 2 incomplete left valves (one 145 mm in

length), Koingnaas KL south face.

Non-type material. SAM—PQ-KN368, numerous friable fragments, Koing-

naas KN-1. |

Description

Shell very thick (up to 40mm on ventral margin), becoming thinner

posterodorsally. Hinge with 11 elongate ligamental grooves; hinge width

36,5 mm.

Remarks

The genus /sognomon is largely a warm-water form. On the west African

coast, living Isognomon occurs as far south as the Congo (Nicklés 1950: 172). On

the east African coast, I. anomioides (Reeve, 1858) commonly occurs in rock

pools’as far south as the Transkei coast. Isognomon perna occurs occasionally in

Natal. Both these latter species reach a total length of about 66 mm and are thin-

shelled.

Isognomon cf. gaudichaudi (d’Orbigny, 1842) (from the Miocene of Chile)

was recorded from Needs Camp, Cape Province (Woods 1908; Newton 1913) and

originally was thought to be of Cretaceous age, but later was referred to the

Cenozoic. The species has also been recorded from Redhouse, Koega, and

_Bushmans River, Cape Province. The specimens examined in the South African

Museum are from Redhouse and Swartkops (Fig. 36B). Newton (1913) compared

the Cape Province specimens with J. maxillata (Lamarck, 1801) from Europe,

I. conradi (d’Orbigny, 1842) from Virginia, U.S.A., and JI. gaudichaudi

(d’Orbigny) from Chile, and on the basis of the hinge similarities decided that the

South African specimens most closely resembled the South American species.

The present west-coast material is similar to the material from the Eastern

Province in hinge structure, but is much thicker and heavier. The ventral marginal

sinuosity of the Isognomon cf. gaudichaudi is more marked than in the Hondeklip

material. While it is unlikely that either the present samples or the Eastern

Province material is conspecific with the Chilean species, there is precedent for

this distribution pattern. Kensley & Penrith (1970) recorded three species of

mytilid bivalve molluscs, and one brachiopod species from northern South West

FOSSIL MOLLUSCA OF HONDEKLIP 203

Fig. 36. A. Isognomon gariesensis. Holotype. B. Isognomon ‘gaudichaudi’. Three specimens

from Redhouse, Cape Province. Scales = 10 mm.

204 ANNALS OF THE SOUTH AFRICAN MUSEUM

Africa—Namibia and Angola that occur in Pacific South America, while Kensley

(1985b) recorded the South American thaidid Concholepas as a fossil from the

west coast of South Africa. Until more complete material is available, it is wiser

to give the present material new specific status, rather than confuse it with

earlier-described species.

Etymology

The specific name is derived from Garies, the district in which the Hondeklip

and Koingnaas localities are found.

Family Carditidae

Cardita unica sp. nov.

ewe, Sy)

Material

Holotype. SAM—PQ-KN566, 1 right valve, 44,0 x 67,5 mm, Koingnaas KL

south face.

Description

Shell robust, thick, oblong, anterior margin rounded, ventral margin

flattened, posterior margin more pointed than anterior. External sculpture

consisting of 27 radiating ribs, becoming broadly flattened near ventral margin,

separated by narrow grooves; ribs on posterior part of shell narrower than on rest

of shell. Concentric growth lines irregular, wavy. Lower anterior and posterior

margin and ventral margin on shell interior crenulate, crenulations largest at

posteroventral corner. Anterior adductor scar oval; small circular hollow dorsal

to anterior muscle scar, hidden under anterior hinge line. Posterior adductor scar

anteriorly truncate, posteriorly rounded. Hinge line broad, solid; anterior lateral

tooth almost vertical, short, separated from anterior cardinal tooth by triangular

pit; anterior cardinal tooth acutely triangular with posterior margin twice length

of anterior margin, separated from posterior cardinal tooth by narrowly

triangular oblique groove; posterior cardinal tooth elongate, with horizontal

dorsal part and oblique ventral part. Ventral margin of hinge area sinuous.

Remarks

The present specimen bears no resemblance to any carditid, fossil or living,

recorded from southern or west Africa.

Etymology

The specific name, derived from the Latin ‘unicus’, meaning unique, refers to

the fact that only a single valve of this species has been found, and that it bears no

resemblance to any African carditid.

FOSSIL MOLLUSCA OF HONDEKLIP 205

Fig. 37. Cardita unica. Holotype, external and

internal view of right valve. Scale = 10 mm.

Cuna aquaedulcensis Kensley, 1977

Fig. 38

Cuna aquaedulcensis Kensley, 1977: 203, fig. 15.

Material

SAM-—PO-HB567, about 50 valves, largest specimen 7,5 x 7,1 mm, Honde-

klip, 30 m Complex.

Description

Right valve, hinge with elongate anterior cardinal tooth, median tooth

narrowly triangular, separated from anterior tooth by narrow triangular gutter.

206 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 38. Cuna aquaedulcensis. Hinge details

of left (upper) and right (lower) valves.

Posterior cardinal tooth reduced to short low rounded ridge. Left valve, hinge

with elongate rounded ridge on anterior margin, running from anterior adductor

muscle scar to umbo; 2 median teeth, anteriormost larger, smaller tooth narrow

and lower; posterior tooth separated from margin by relatively deep groove.

Previous records

Quartzose Sand Member, Langebaanweg, Pliocene.

Remarks

As only two right valves were available when the species was described, it

was thought useful to supplement the description, now that far more material is

available.

Family Veneridae

Dosinia (Dosinia) sicarisinus sp. nov.

Figs 39, 40

Material

Holotype. SAM-—K4877, both valves, 75,7 x 72,4 mm, thickness of two

valves together 27,1 mm, Somnaas SN4, Hondeklip, 50 m Complex, coll. A. J.

Tankard.

Paratypes. SAM—PQ-HB568, 4 left valves, 66,7 X 65,5 mm, 63,1 X 65,0 mm,

57,4 57,7 mm, ?X66,9mm, 1 right valve, 67,4 70,7 mm, Hondeklip

FOSSIL MOLLUSCA OF HONDEKLIP 207

Fig. 39. Dosinia sicarisinus. A. Holotype, external and internal

view of both valves. B. Paratype. Scale = 20 mm.

208 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 40. Dosinia sicarisinus. Hinge details of right (upper) and

left (lower) valves.

Zone 12, 50 m Complex. SAM—PQ-HB569, both valves (hinge of right valve

missing), 62,8 X 68,0 mm, Hondeklip Zone 4A, 50 m Complex. USNM 400993,

both valves, 63,6 X 62,5 mm, 2 left valves, 66,3 x 64,5 mm, 61,8 X 63,2 mm,

Hondeklip, 50 m Complex.

Description

Shell proportions somewhat variable, generally longer than wide but

occasionally wider than long. Sculpture of concentric lines, barely lamellose

anteriorly and posteriorly. Lunule well defined; escutcheon lacking. Pallial sinus

horizontal to slightly ascending, apically narrowed. Left and right hinges with

small anterior lateral tooth set at angle to anterior cardinal tooth; posterior

cardinal tooth in right valve bifid; anterior cardinal tooth in left valve much less

markedly bifid, with very narrow longitudinal slit.

Remarks

The present material bears little resemblance to the extant species of Dosinia

from southern Africa. Apart from attaining a much larger size than any of the

living species, D. sicarisinus also differs in the very distinctive narrowly triangular

pallial sinus that reaches to below the umbo. Dosinia exoleta (Linnaeus) of the

Mediterranean, and also known from Norway to North Africa, has a broader

pallial sinus, and possesses lamellar concentric rings.

FOSSIL MOLLUSCA OF HONDEKLIP 209

In general appearance and size, the present material resembles some of the

Tertiary species from New Zealand, especially D. (Raina) bartrami Laws, 1930.

This latter, however, has a shallow pallial sinus, and the shallow lunule of the

subgenus (see Keen 1969: N679). Dosinia sicarisinus in placed in the subgenus

Dosinia for lacking an escutcheon and lamellose concentric rings.

Etymology

The specific name is derived from the Latin ‘sicarius’, a murderer, and

‘sinus’, a bay, and alludes to Moordenaarsbaai, a coastal embayment close to the

Hondeklip type localities.

Family Pholadidae

Barnea truncata (Say, 1822)

Barnea truncata: Nicklés, 1950: 232, fig. 454. Barnard, 1964: 565. Kilburn & Rippey, 1982: 203.

Kensley, 1985a: 116.

Material

SAM-—PO-KN356, 2 specimens, left and right valves, approx. 60 mm and

75 mm in length; right valve, approx. 50 mm in length, Koingnaas KN-1. SAM-

PQ-HB26, 3 worn hinges, Hondeklip, 30 m Complex.

Previous records

Living: Senegal to Angola; east coast of U.S.A.; numerous fresh dead shells

from Table Bay in South African Museum collection.

Fossil: Milnerton, Table Bay (Late Pleistocene); borehole in Kuiseb River,

South West Africa—Namibia, 9 miles (14,5 km) from sea (Geological Survey,

determined by K. H. Barnard).

Class SCAPHOPODA

Family Dentaliidae

Dentalium sp.

Fig. 41

Material

SAM-PQ-KN570, fragments, Koingnaas KN—1. SAM—POQ-KN376, several

fragments, Koingnaas KN-1.

Description

Shell gently convex; with 16 rounded ribs at narrower end (1,8 mm

diameter); ribs equal in width to concave furrows between them; at about 3,0 mm

diameter, much narrower intermediate rounded ribs (1 between 2 larger ribs)

becoming apparent; 26 ribs present.

210 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 41. Dentalium sp. Fragments. Scale = 5 mm.

Remarks

The present material bears little resemblance to any living species of

Dentalium described from southern Africa.

DISCUSSION

FAUNAL COMPOSITION

The following table is a broad-scale analysis of the faunal composition of the

fossil molluscs from the Hondeklip area collected during this study. Material

identified to generic level only is excluded (15 genera).

Totalnumber of Species 4 :.%-3.. eet ee eee 92

otal numberokextincispeciess= ee eee eee 45 (49%)

Lyellian percentace (percentage extant) meee ieee ier UY

Total number of species still living on east and west coasts....... 27

Total number of species living on east coastonly ............... 11

Total number of species with West African—Mediterranean

ALAMITOS: 35 25s Pe D eae Sea aie oe eee a 9

FOSSIL MOLLUSCA OF HONDEKLIP 211

Extinct species also present in east-coast deposits .............. 4

Total number of species from 50 m Complex .................. 78

Total number of extinct species from 50 m Complex ............ 37 (47%)

[Bye llianmgencenlage We ey ee eh ya ee wok baie 53 %

50 m Complex species recorded by Carrington & Kensley

(1969) but not found during course of thisstudy............ 4

(of which 3 are extinct)

otal number ofspecies trom 30m Complex .................. 48

Total number of extinct species from 30 m Complex ............ 19 (40%)

HB ye Ilia ENC EMCAGES ee) yas Agee erie, Govors alee hah dees ta on 60 %

Number of species common to 50 m and 30 m Complexes ....... 34

PNumbenor extinchspectes in'COMMON . 22.24... 22. ee eee ese: fe

From the previous summary, several topics require further comment.

Notably, the 50 m Complex shows a higher diversity than the 30 m Complex

(78 species as against 48). A glance at the species list (Table 1), which records the

depositional environment of the beds from which the molluscs were obtained,

shows that a greater range of environments was available in the 50 m Complex

(viz. back-barrier, tidal-inlet, and distal lower-shoreface deposits, as against only

near-shore open-coast deposits in the 30 m Complex). It is thus tempting to

attribute the higher diversity of the 50 m Complex to the wider range of habitats.

However, exclusion of specimens obtained only from 50 m Complex tidal-inlet

and back-barrier deposits from the species list removes only nine species (two

extinct), and the comparative diversity becomes 69 species (35 extinct) as against

48 species (19 extinct). This reflects the fact that the shelly beds in the 50 m

Complex back-barrier-related facies are transported assemblages consisting of a

mixture of calm-water and open-coast forms. Additionally, calm-water species

may also inhabit the substrate below average wave-base, and become incorpor-

ated into open-coast facies, provided the open-coast temperature regime is within

species tolerances.

As shown for the Late Pleistocene at Verlorevlei (Tankard 1975), thermally

anomalous back-barrier environments increase species-diversity by the addition

of warm-water taxa to the faunal composition. A modern example of such

localized habitats is Sandvis Lagoon (Sandwich Harbour) near Walvis Bay

(Kensley & Penrith 1977). However, in such cases the contrast between the

back-barrier and open-coast assemblages is clear-cut. This is evidently not the

case with the Hondeklip fauna, but this observation is qualified by the absence of

large life assemblages preserved in the 50 m Complex back-barrier facies.

In the 50 m Complex back-barrier facies, only Dosinia sicarisinus, Donax

haughtoni, Phaxas decipiens, Standella namaquensis, and Tivela cf. compressa

have been found in life positions. In open-coast facies, Lutraria sp. has been

found in life position around the lower-shoreface—upper-shoreface facies bound-

ary in the 50 m Complex; in the 30 m Complex, Donax rogersi and Standella

22 ANNALS OF THE SOUTH AFRICAN MUSEUM

namaquensis occur in life positions in the upper shoreface. Striostrea margariticea

has been found attached to rocks in open-coast contexts in both complexes.

In Table 1 the species obtained from deposits on the properties of Koingnaas

and Swartlintjiesrivier have been indicated separately owing to the unique addi-

tions to the faunal list those beds have contributed. In the case of Koingnaas (KN)

and Swartlintjiesrivier (SL) samples, this is further warranted since the beds are

distal lower-shoreface deposits of the 50 m Complex. The curious mix of intertidal

and deeper-dwelling forms (e.g. Dentalium, Ringicula, Tugali, and Turritella

declivis) in these samples, together with the muddy nature of the enclosing

sediment, is consistent with a depositional environment transitional to offshore

conditions. As such, the KN sample has contributed most to environmentally-

based diversity in the 50 m Complex.

Removing species unique to these samples (KN, SL, and KL) from the

faunal list results in a 50 m—30 m Complex diversity contrast of 64 species

(30 extinct) as against 39 species (14 extinct), respectively. Excluding these

examples, as well as species from the 50 m Complex tidal-inlet and back-barrier

facies from the faunal list, results in a diversity contrast of 54 species (28 extinct)

as against 39 species (124 extinct) for the 50 m and 30 m Complexes, respect-

ively. Only by not excluding the KL sample from the latter calculation is the

diversity contrast narrowed: 54 species (50 m Complex) as against 48 (30 m

Complex). Thus the higher diversity of the 50 m Complex is either real, or an

Fig. 42. Mactrid hinge remains from Koingnaas KL south face. Scale = 30 mm.

FOSSIL MOLLUSCA OF HONDEKLIP 213

unavoidable bias due to the nature of the available exposures. In addition, any

reworked forms possibly present in the 30 m Complex will reduce the diversity

contrast.

TABLE 1

Species of Mollusca obtained from four properties in the Hondeklip area of the Namaqualand

coast, reflecting the complex and depositional facies from which they were obtained.

Habitat: R = Rocky, S = Sandy, RiS = Rock in sand, M= Muddy. W/E: Species confined to

either west (W) or east (E) coasts. TIL =Tidal-inlet facies. BBR = Back-barrier facies.

USH = Upper-shoreface facies. LSH = Lower-shoreface facies. DLS = Distal lower-shoreface

facies. KN=Koingnaas KN. KL=Koingnaas KL. SL=Swartlintjies SL. += Extinct.

50 m Complex 30 m Complex

Habi- DLS LSH

Species tat | WE|TIL BBR USH LSH (KN/SL)|USH LSH (KL)

GASTROPODA

Afrocominella capensis.......... R spor SP

(Dunker in Philippi, 1844)

Amblychilepas scutellum ........ R in ‘tate ty ake +P

(Gmelin, 1791)

tArgobuccinum casus sp. nov. .... PSP ab

}+Bolma anoropha sp. nov......... +

Bullia annulata (Lamarck, 1816).. | S qe OR

Bullia digitalis (Dillwyn, 1817) ... | S seo PE + +

Bullia laevissima (Gmelin, 1791) | S/M +

{+ Burnupena aestus sp.nov. ....... ap Sp

Burnupena papyracea.......... aR St ayelcte oat

(Bruguiére, 1789)

+ Burnupena rogersi sp.nov. ...... =P

+Calliostoma depressa ........... =F

Carrington & Kensley, 1969

Cantharidus (Jujubinus) striatus . . W op

(Linnaeus, 1758)

Calyptraea helicoidea ........... R E =

Sowerby, 1883

+Calyptraea kilburninom. nov..... | ?R ap PE ap SP AP

+ Calyptraea viridarena ........... oR + +

Carrington & Kensley, 1969

+Clanculus lutosus sp.nov. ....... ate

NClanculus MUTTAYL.............. + + + +

Carrington & Kensley, 1969

Conusimozambicusl. \ 265). 4..... R <p tF

Hwass in Bruguiére, 1789

Crepidula porcellana............ R | W + =P

Lamarck, 1801

}Crepidula deprima sp.nov. ...... R ap BR

Cylichna tubulosa Gould, 1859 ... E a7

Diodora elevata (Dunker, 1846) .. | R ipo Pa 4e sep

| Drillia tempestae sp. nov......... cP) oF

+Epitonium lycocephalum sp. nov. ate ate

214 ANNALS OF THE SOUTH AFRICAN MUSEUM

50 m Complex 30 m Complex

Habi- DLS LSH

Species tat | WE | TIL BBR USH LSH (KN/SL)| USH LSH (KL)

GASTROPODA (contd)

t Fasciolaria dinglei sp. nov........ +

WIDOT OUD BUGEE co ccaccecoedcas ?R apr

Carrington & Kensley, 1969

WHESSLILELLGIKODUS(O. Henne R + + + +

Sowerby, 1892

Fissurciiidea apenas nr er R oP

(Sowerby, 1825)

FUSUSQUTEL vccio: os 1 eae Simhat 8 ah ?

Barnard, 1959 (Koingnaas)

Gibbula zonata patulassp. nov. ..| R qo OoP

PHAlOUS SAldanhads ene R + +

Kensley, 1972

EICICIOW SD ae es See ee R

+ Hespererato oppenheimeri....... SE ESTan cr ot aah ge ? a

Carrington & Kensley, 1969

EMMONNASD sw ee R “lake

Marginellasps ae ae ee AP +

Melanellaspre a0) ern ee +

+Melapium hawthornei sp. nov. ... ate ==

+Namamurex odontostoma ....... + + +

Carrington & Kensley, 1969

Nassarius kochianus ............ E +

(Dunker, 1846)

+ Nassarius litorafontis ........... ies TF ® helenh ary) ? +

Carrington & Kensley, 1969

Naticaciradansont =) eee S/M | W +

Blainville, 1824

Nucella dubia (Krauss, 1848)..... R <P 25

+Nucella praecingulata ........... tp BB SP + + +

(Haughton, 1932)

+Ocenebra bonaccorsii........... 4° ae

Carrington & Kensley, 1969

Ocenebra purpuroides .......... R +

(Reeve, 1845)

+ Ocenebra petrocyon sp. nov...... at

Oxystele sinensis (Gmelin, 1791)

Patella argenvillei Krauss, 1848

Patella barbara Linnaeus, 1758...

Patella granatina Linnaeus, 1758. .

| Patella hendeyisp. nov. .........

+ Patella hoffmani sp.nov. ........

Patella miniata Born, 1778.......

Patellaspe Ra eee ener te ate

| Pseudoliva lutulenta sp. nov. ..... +

Ringicula turtoni Bartsch, 1915... E ar

+ ++ 4

ARAAAARAR

FOSSIL MOLLUSCA OF HONDEKLIP BUNS)

50 m Complex 30 m Complex

Habi- DLS LSH

Species tat | WE | TIL BBR USH LSH (KN/SL)|USH LSH (KL)

GASTROPODA (contd)

Sinum concavum (Lamarck, 1822) W +

} Terebra canisaxisp.nov......... fpr =F

(HU GIS\GTENAGESP. MOVs 925022205: Bp

Tricolia capensis (Dunker, 1846) |} R ap aR

TOAD OWS CHQTUING coe 600006050" po ARS AE

Kilburn & Tankard, 1975

+ Trophon carringtoni sp. nov...... ae

Tugali barnardi (Tomlin, 1932)... E +

Turbo cidaris Gmelin, 1791 ...... R + + + + + + + +

itor nitellaicaninifenGd ae eee S oP se

Lamarck, 1822

MunnitelladeGlvisin) saga ee one. E +

Adams & Reeve, 1848

MLERISIUSTONVULIGe Mera eee ae ee ? ? yj ? y

Carrington & Kensley, 1969

(Koingnaas)

VAC CLUSISID. Mero nins ne Be oye ees oP OP aF oP aR

BIVALVIA

Arca avellana Lamarck, 1819 .... | ?R E

PATCAILAIMIYGUS J eee ane ?R Ra Mi ate eaten ?

Carrington & Kensley, 1969

(Strandfontein)

Arca noae Linnaeus, 1758 ....... 2R | W +

Aulacomya ater (Molina, 1782)... | R ae =F

Barnea truncata (Say, 1822)...... R | W 5° <P

CQNAUGUNICASD4MOVe =)... 45.4. +

+Carditella calipsamma .......... Ss ae + +

Carrington & Kensley, 1969

CONUS ae we ne oe eee +

+Chamelea krigei Haughton, 1926] S eee |e ude

GCHIGTAYS SP ae Orin Sagteie 28: ate

Choromytilus meridionalis....... R +

(Krauss, 1848)

jiConbulapalacgialus a... 6045: UE Ne cen ame ?

(Carrington & Kensley, 1969)

(Strandfontein)

+ Cuna aquaedulcensis............ a

Kensley, 1977

WDOMGEBE (POW UOM oo oo65s000e004. S deo dp Rp

Carrington & Kensley, 1969

{Donax rogersi Haughton, 1926 ... S stile

| Dosinia sicarisinus sp. nov. ...... S/M =F ar oF

GaStrand fUOTOSG) - 4.6.8.5... 2. S ap

Kilburn & Tankard, 1975

216

Species

BIVALVIA (contd)

iGGStan@ TOS. AlIGhme en eee nee

Carrington & Kensley, 1969

GaSiTQnG Spohn ee ee ee

+Glycymeris fullerisp. nov. .......

VEliatellasp yaa. oe ee

FLINNUCSISPAy serach eiet eerne

}Isognomon gariesensis sp. NOV... .

Leporimetis hanleyi.............

(Dunker, 1853)

LEULFATIO SD. Oa era in ta ae

WG Ct, COMUVAD 000006500006

Bohm & Weisfermel, 1913

MelliteryxCapensisimern ence

(Sowerby, 1889)

+Notocallista schwarzi ...........

(Newton, 1913)

Nuculana bicuspidata ...........

(Gould, 1845)

11 OSERCALCEASUDIGGIOSA A ee ae

(Bohm & Weisfermel, 1913

Perna perna (Linnaeus, 1758) ....

PA OMEDUE DRIVE so 000900066000000

Kilburn & Tankard, 1975

Phaxas decipiens (Smith, 1904)...

Scissodesma spengleri...........

(Linnaeus, 1767)

+ Standella namaquensis ..........

Carrington & Kensley, 1969

Striostrea margaritacea ..........

(Lamarck, 1819)

Kellinaponsonbyie eee eeer

(Sowerby, 1889)

Tellina trilatera Gmelin, 1791 ....

UNCOTASD.. a sei eo Ne ee

livelactacompressa pene

(Sowerby, 1851)

Venus verrucosa Linnaeus, 1758 . .

SCAPHOPODA

Dentalium span eee

POLYPLACOPHORA

Chaetopleura pertusa ...........

(Reeve, 1847)

Ch Chiionspy she Ses epee

Ci Dino plaxisp aan a eer:

Habi-

tat

ANNALS OF THE SOUTH AFRICAN MUSEUM

50 m Complex 30 m Complex

DLS

W/E | TIL BBR USH LSH (KN/SL)| USH LSH (KL)

LSH

FOSSIL MOLLUSCA OF HONDEKLIP DAG

Table 1 indicates the preferred habitats of most species. In a number of cases

there is uncertainty about habitat, especially for extinct species. While most are

open-coast forms, there are calm-water components present (e.g. Bullia

laevissima, Phaxas decipiens, Dosinia sicarisinus, Lutraria sp., Leporimetis

hanleyi). Among the open-coast forms, rocky-shore species outnumber sand and

mud inhabitants almost two to one. Among the rocky-shore forms are a number

of grazers and algal inhabitants such as Haliotis, Crepidula, Oxystele, Turbo,

Calliostoma, and Patella. Predators such as Argobuccinum, Namamurex, Thais,

Ocenebra, Conus, and Terebra are also abundant. Filter feeders such as

Turritella, scavengers like Bullia and Nassarius, and parasitic forms such as

Epitonium and Melanella are also present, indicating a rich and diverse

environment with numerous habitats. Of the sand and mud dwellers there are

suspension and filter feeders (Donax, Dosinia, Scissodesma, Venus, Lutraria), as

well as deposit feeders (Tellina).

Of the species recorded by Carrington & Kensley (1969), only four have not

been found during the present study. These are Fusus faurei, Turris nigrovitta,

Corbula palaegialus, and Arca halmyrus, all recorded from the 50 m Complex.

Nassarius litorafontis and Hespererato oppenheimeri, recorded from the 50 m

Complex by Carrington & Kensley (1969), have been found in the 30 m Complex

in this study.

ZOOGEOGRAPHIC AFFINITIES

There is a small west African—Mediterranean component of nine species in

the faunal list: Arca noae, Barnea truncata, Cantharidus (Jujubinus) striatus,

Crepidula porcellana, Leporimetis hanleyi, Nuculana bicuspidata, Natica cf.

adansoni, Sinum concavum, Venus verrucosa. Of these, Arca noae, Cantharidus

(Jujubinus) striatus and Natica cf. adansoni are strictly West Africa—Mediterra-

nean in distribution; the present fossil records are the most southerly for all three.

Nuculana bicuspidata occurs live in West Africa, and is known as a Pleistocene

fossil from Velddrif (Kruispad), Table Bay (Milnerton), and Port Elizabeth

(Redhouse) (See Kilburn & Tankard 1975: 206). Leporimetis hanleyi is known

live from Luanda, while an isolated population occurs at Sandvis, just south of

Walvis Bay. It is recorded as fossil in the Late Pleistocene in the Saldanha area,

and also at Redhouse, Knysna, Sedgefield, and Klein Brak River. Venus

verrucosa and Crepidula procellana both occur from North Africa, around the

Cape, to Natal. Barnea truncata is known from the Late Pleistocene of Table Bay

(Kensley 1985a) and also from fresh dead shells washed ashore at the same

locality. Sinum concavum is an extant West African species, known from Senegal

to Angola (Nicklés 1950). This so-called West African component thus represents

a mix of widespread temperature-tolerant species, as well as less temperature-

tolerant forms now restricted to more tropical areas.

The present-day distribution of living species represented in the faunal list

shows a majority of forms (27 species, 29%) occurring on both east and west

coasts of southern Africa. Only 11 species are at present restricted to the east

218 ANNALS OF THE SOUTH AFRICAN MUSEUM

coast: Calyptraea helicoidea, Melliteryx capensis, Nassarius kochianus, Oxystele

sinensis, Ringicula turtoni, Arca avellana, Striostrea margaritacea, Tellina

ponsonbyi, Tugali barnardi, Turritella declivis, Cylichna tubulosa. Of these, only

the Indo-Pacific Arca avellana has a distribution extending north of southern

Mozambique.

The inferring of past sea-temperature regimes based on the known

temperature ranges of living species is a complex issue fraught with pitfalls. Some

guidelines do exist. Tankard (1975b: 33) presented temperature minima for a

range of Pleistocene molluscs of South Africa, and indicated a warm-water

affinity for a number of west-coast fossils. Several of these occur in the present

suite:

Temperature

minima

ISTCTOAD (QHOVISDUGIIO. on 502900089500000005 LETC

SCISOGESINGSDEHOICTINN nna arene er 14°C

HAVE /DORSIOMOY 5 oo bacco ekeucsccsabeones 14°C

LGV OATES (WUDIL 5 66 on ben o06 0605065008 00 ie

Kilburn & Rippey (1982) mention that Striostrea margaritacea spawns when

monthly average temperatures exceed 20°C.

Examining the nine ‘West African’, and the 11 east-coast forms (with their

implied warmer-water requirements), it is seen that 7 species, including Striostrea

margaritacea and Nuculana bicuspidata, occur in both the 50m and 30m

Complexes. Fourteen of the 20 species occur in the 50 m Complex, 12 1n the 30 m

Complex. It is thus inadvisable to attribute a colder temperature regime to the

30 m Complex and a warmer regime to the 50 m Complex. Nevertheless, there is

undeniably a component showing a warm-water bias in the present suite of fossils,

deduced from known temperature ranges of living species. The temperature

requirements for the extinct forms, with the greater preponderance of species in

the 50 m Complex, can only be guessed at. The reasons for the extinction of these

forms are probably related in part to sea-temperature changes.

An occulinid coral, found encrusting rocks at Hondeklip in 50 m Complex

exposures, and tentatively identified as Schizoculina fissipara (Milne Edwards &

Haime), casts further light on the question of temperature regimes. According to

Laborel (1974), modern S. fissipara is adapted to low-salinity, warm, Guinean

waters, and has a ramose morphology. However, encrusting to subramose forms

(as in the present case) occur at the extremes of its range where periodic

upwelling takes place. This instance probably represents the most southerly

occurrence of S. fissipara in the fossil record and implies the extension of tropical

water southward to Hondeklip latitudes. Lithological evidence of upwelling off

Hondeklip during 50 m Complex times comes from phosphorite rinds inter-

bedded with regressive storm gravels deposited in the bedrock-defined embay-

ment during the earlier stages of bedrock emergence, prior to the establishment

of back-barrier conditions. The thin phosphorite rinds mark intervening

FOSSIL MOLLUSCA OF HONDEKLIP 219

fairweather periods during the deposition of the storm gravel. Hondeklip was

probably situated adjacent to a marine regime characterized by the interaction of

upwelling and south-flowing tropical currents during the period of deposition of

the 50 m Complex.

The warm-water species common to both the 50 m and the 30 m Complexes,

especially the abundant Striostrea margaritacea, undoubtedly point to a sea-

temperature regime considerably higher than that prevailing on the west coast at

present. Over the period of deposition of both complexes, however, a cooling

trend must be inferred from the presence of numerous species living at present on

the west coast. The apparent reduction in diversity reflected by the 30 m Complex

may be indicative of cooling. Significantly, Choromytilus meridionalis evidently

first appears during 30 m Complex times (Early Pleistocene). The cooling trend is

also indicated by the extinction of Striostrea margaritacea on the west coast

subsequent to the Early Pleistocene.

The Hondeklip fossil fauna probably represents a mixture of forms at

different points in their history, reflecting both the influence of the cold Benguela

system and subtropical waters, fluctuating, glacially controlled sea-levels and

temperatures, along with varying temperature requirements. Thus, some

specimens may represent the last members of a population that was dying out

owing to decreasing temperatures, others a population adapting to fluctuating

conditions, others an ‘experimental’ pioneering stock having tenuous repro-

ductive success, and still other temperature-tolerant and actively reproducing

populations.

FAUNAL COMPARISONS

Evaluation of the fauna from Hondeklip and vicinity must necessarily take

the following into consideration.

The faunal list is undoubtedly incomplete. An unavoidable bias is, of course,

the very nature of the geological record. Only deposits accumulated during

regression from the sea-level maxima are available for examination; deposits

closer to the regressive maxima are not available, nor are deposits of the

transgressions. Thus the sedimentary packages available for sampling were

deposited during the earlier stages of recovery from major polar deglaciations. In

this sense successive complexes are somewhat directly comparable, but this may

also make for the reduction of faunal dissimilarity between complexes. Original

community patchiness, differential transport, preservational bias due to dissolu-

tion, and sampling bias (e.g. the tendency for small forms to be overlooked) all

contribute to an incomplete faunal list.

The latitudinal ranges of species in deposits equivalent to the 50 m and 30 m

Complexes are largely unknown or uncertain. Sampling was restricted to a few

localities in close proximity.

The temporal ranges of species are similarly unknown or uncertain. The Late

Tertiary and Middle Pleistocene are practically unsampled. The best comparative

faunal lists exist for the Late Pleistocene (e.g. Tankard 1975a; Kensley 1985a).

220 ANNALS OF THE SOUTH AFRICAN MUSEUM

In spite of these limitations, comparisons with other west-coast assemblages

are instructive.

Bogenfels Tertiary Deposits

While the overall fauna of the Tertiary deposits of Bogenfels, South West

Africa—Namibia, is distinctive and very different from any present-day assem-

blage (B6hm & Weisfermel 1913; B6hm 1926) there are a few elements that hint

at a distant relationship with the Hondeklip suite.

The presence of a species of Pseudoliva (with one living species in West

Africa, and two extinct species from Bogenfels), mactrid hinges closely

resembling Mactra (Barymactra) dernburgi Bohm & Weisfermel from Bogenfels

(Fig. 42), and a single oyster valve almost identical with Ostrea subradiosa Bohm

(Fig. 35) suggest that there may well be Tertiary stragglers or their descendants in

the Hondeklip assemblage. This suggestion, however, must be treated as highly

speculative.

Marine Neogene of Ysterplaat

Tankard (1975a) provided a list of 13 molluscs recovered from the Miocene

of Ysterplaat, Cape. Olson (1985), on the basis of fossil penguin material from

the same locality, placed the assemblage in the Early Pliocene. The bivalves

Cardium edgari Newton, 1913, and Glycymeris borgesi (Cox, 1939) support a

Neogene age. Also taken from the Ysterplaat site were Donax serra Dillwyn,

1817, Scissodesma spengleri, Dosinia lupinus and a Pitar, since identified as

Notocallista schwarzi. Of these all but Donax serra and Dosinia lupinus occur at

Hondeklip. While most of the Ysterplaat Dosinia specimens are within the usual

size range of D. /upinus, a few reach the size of D. sicarisinus described above.

Unfortunately, the Ysterplaat material is all mouldic, and a clear impression of

hinge and mantle details is lacking, making a positive identification difficult.

Donax serra is discussed in the next section.

Early Pliocene molluscs from Langebaanweg, Cape

Kensley (1972, 1977) recorded a number of molluscs from Langebaanweg,

several of which also occur at Hondeklip. These are Cuna aquaedulcensis, Bullia

digitalis, B. laevissima, Tricolia capensis, Haliotis saldanhae, and Thais dubia.

Cuna aquaedulcensis is locally abundant in the 30 m Complex at Hondeklip.

Bullia digitalis is abundant in both the 30 m and 50 m Complexes; its living range

from South West Africa—Namibia to Transkei would indicate that this is a

temperature-tolerant species, and its presence in deposits from the Pliocene to

the Holocene is not surprising.

The fossil distribution of Donax serra, however, is difficult to explain. It has

been recorded from the Pliocene of Ysterplaat (Tankard 1975a), and Langebaan-

weg (Kensley 1972, 1977), and the Pleistocene of Liideritz, Orange River mouth,

Velddrif, Sedgefield (Barnard 1962), and Table Bay (Kensley 1985a). It has not

been seen in the Hondeklip deposits under discussion, where D. rogersi (30 m

FOSSIL MOLLUSCA OF HONDEKLIP Dai

Complex) and D. haughtoni (50 m Complex) are both abundant. Its present-day

distribution is from South West Africa—Namibia to Transkei, while De Villiers

(1975) reported the species to reproduce in a temperature range of 13—17°C.

Inappropriate temperature thus seems unlikely to be the reason for its absence

from the Hondeklip area. It is possible that local competition from the two extinct

species during the Early Pleistocene precluded the establishment of a population

of D. serra in the Hondeklip area.

Verlorevlei—Saldanha Pleistocene deposits

A number of species recorded from the Pleistocene of the south-western

Cape around Verlorevlei and Saldanha (Kilburn & Tankard 1975) are present in

the Hondeklip assemblage. The extinct species Petricola prava, Nucella prae-

cingulata, Triumphis dilemma and Fissurella robusta were obtained from deposits

at 10 masl at Saldanha Bay. The latter two species occur in the 50 m Complex and

the former two in both the 50m and the 30m Complexes at Hondeklip.

Cerithidea bifurcata Kilburn & Tankard, 1975, an extinct species also obtained

from the 10 masl deposit, has not yet been found in Namaqualand.

Tankard (1975c) correlated the Saldanha deposit at 10 masl with the 50 m

Complex in Namaqualand on the basis of the four species mentioned above. The

presence of Fissurella robusta and Triumphis dilemma suggests that this

correlation may be correct but it should be verified by additional criteria.

Accepting the equivalance of the deposits, comparison of the faunal lists shows an

additional 13 extant species in common with the 50 m Complex at Hondeklip, all

of which occur on both west and east coasts. Of the Saldanha fauna, only Patella

tabularis Krauss, 1848, P. concolor Krauss, 1848, and Peristernia nassatula

(Lamarck) are south- and east-coast species not present off Saldanha today.

Tankard (1975c) attributed the absence of Striostrea margaritacea and Donax

haughtoni from the Saldanha deposit to colder water conditions there during the

50 m Complex times.

Gastrana fibrosa, a probably extinct species recorded from the Late

Pleistocene in the south-western Cape, extends back into the Late Pliocene in

central Namaqualand, as shown by its occurrence in the 50 m Complex. The

extinct Late Pleistocene species Crepidula capensis praerugulosa Kilburn &

Tankard, 1975, is not present in the Hondeklip assemblage, though Crepidula

porcellana is.

CONCLUDING REMARKS

Significantly, the warm-—cold distinction between the 50m and 30m

Complexes respectively (or across the Plio—Pleistocene boundary), is not clear-

cut on the basis of existing data, due to the considerable faunal similarity between

the complexes. Notably, there are 34 species (11 extinct) common to both

complexes. Of the species unique to the 30 m Complex, only five are relatively

abundant, viz. Donax rogersi, Fissurella glarea, Choromytilus meridionalis, Cuna

222 ANNALS OF THE SOUTH AFRICAN MUSEUM

aquaedulcensis, Ocenebra petrocyon. The reduction of diversity and the appear-

ance of Choromytilus meridionalis may suggest a cooler regime during 30 m

Complex times, relative to the 50 m Complex.

The value of Donax haughtoni and Donax rogersi as zone fossils for the 50 m

and 30 m Complexes, respectively, is verified in this study. Donax haughtoni is

present at least as far south as the Olifants River. Donax rogersi, the 30 m

Complex zone fossil, is present at least as far south as Doringbaai and has been

found as far north as Walvis Bay (pers. obs.). However, there are no faunal lists

from 30 m Complex correlates in the south-western Cape for comparison. It

appears that only subsequent to the Early Pleistocene did the Namaqualand

west-coast mollusc fauna more closely resemble the modern fauna. This probably

reflects the increasing dominance of the Benguela system and restriction of

warmer waters to the north, the cooler regime facilitating a species radiation

northwards from the south-western Cape. The ecological niche of Striostrea

margaritacea was usurped, to some extent, by the mytilids and patellids, the latter

to become an important food resource of early man on the west coast. The first

appearance of Patella compressa, the kelp limpet, already present by the Late

Pleistocene, is an important datum yet to be established.

Sedimentological aspects of the complexes suggest that, in the Hondeklip

area during 50 m Complex times, the coast was characterized by a sea-level

interaction with the bedrock topography that resulted in an embayed coast

concomitant with conditions for the development of extensive back-barrier

environments. In contrast, the 30m Complex is characterized by relative

insulation from antecedent topographic effects, resulting in a more exposed

coastal regime. Superimposed upon these littoral influences are the effects of the

adjacent oceanographic temperature regime. Thus a secure database from which

to deduce Late Cenozoic palaeoclimatic influences will only exist once the

regressive packages are examined sedimentologically and faunistically sampled

on a regional basis in order to resolve the large-scale changes in the boundaries of

marine zoogeographic provinces in time, to recognize the environmental! vectors

such as depositional palaeodepth and possibly thermally anomalous lagoons, and

allied to the latter, to evaluate the roles of antecedent topography and sediment

supply.

In conclusion, this study suggests that molluscs have potential for a

significant contribution in the unravelling of Late Cenozoic history. More extinct

molluscs were found than expected and the relatively high level of endemicity

holds promise that more zone fossils may emerge. The extant component of the

fauna permits some extrapolations of temperature tolerances and environmental

preferences. An overall cooling trend is reflected in the approach of successively

younger assemblages towards the modern west-coast faunal composition.

Comparison of Palaeogene, Neogene, Quaternary, and extant faunas

suggests that extinction, speciation, and migratory colonizations were staggered

as a reflection of differing temperature tolerances, habitat creation—destruction,

and competition. However, a regional mollusc-assemblage zonation scheme must

FOSSIL MOLLUSCA OF HONDEKLIP 223

in time necessarily be supplemented by zonations of ostracode and benthic

foraminiferal assemblages. Planktonic Foraminifera are evidently very scarce in

these littoral deposits, while nannoplankton has not yet been collected, but

efforts to recover these microfossils may aid correlation with deep-sea data and

the global record. Resolution of the tectonic component of the coastal-plain

marine record awaits reliable regional correlations.

ACKNOWLEDGEMENTS

We thank Mr Clive Booth and Miss Elsabé Pretorius of the South African

Museum, and Mr Michael Carpenter of the Smithsonian Institution, for the

photographs used in this work.

We are grateful to Mr F. Hoffman of Transhex, for access to the mine

exposures at Hondeklip and Avontuur, and to Mr J. B. Hawthorne and Mr

R. Molyneaux of De Beers Consolidated Diamond Mines for arranging access to

the Koingnaas and Swartlintjies River exposures.

Dr George Steyskal of the Smithsonian Institution kindly checked the new

names and etymologies for correctness.

The paper benefited considerably from the comments and criticisms of Dr

R. N. Kilburn of the Natal Museum, and Dr J. Rogers of the University of Cape

Town.

We thank the Council for Scientific and Industrial Research for financial

assistance during the course of this work, and the Council of the South African

Museum for assistance and hospitality to the first author during several visits.

This work benefited immeasurably from the hospitality and co-operation of,

and discussions with, Dr Q. B. Hendey of the South African Museum, whose

encouragement to both authors is much appreciated.

REFERENCES

AGuIRRE, E. & PAsini, G. 1985. The Pliocene—Pleistocene boundary. Episodes 8 (2):

116-120.

BARNARD, K. H. 1958. Contributions to the knowledge of South African marine Mollusca.

Part 1. Gastropoda: Prosobranchiata: Toxoglossa. Annals of the South African Museum 44

(4): 73-163.

BARNARD, K. H. 1959. Contributions to the knowledge of South African marine Mollusca.

Part 2. Gastropoda: Prosobranchiata: Rhachiglossa. Annals of the South African Museum

45 (1): 1-237.

BARNARD, K. H. 1962. Revised list of South African Late Tertiary and Pleistocene marine

Mollusca. Transactions of the Royal Society of South Africa 36 (4): 179-196.

BARNARD, K. H. 1963. Contributions to the knowledge of South African marine Mollusca.

Part 3. Gastropoda: Prosobranchiata: Taenioglossa. Annals of the South African Museum

47 (1): 1-199.

BARNARD, K. H. 1964. Contributions to the knowledge of South African marine Mollusca.

Part 5. Lamellibranchiata. Annals of the South African Museum 47 (3): 361-593.

BARNARD, K. H. 1969. Contributions to the knowledge of South African marine Mollusca.

Part 6. Supplement. Annals of the South African Museum 47 (4): 595-661.

BEARD, J. M., SANGREE, J. B. & Smitu, L. A:. 1982. Quaternary chronology, paleoclimate,

depositional sequences, and eustatic cycles. . Bulletin of the American Association of

Petroleum Geologists 66 (2): 158-169.

224 ANNALS OF THE SOUTH AFRICAN MUSEUM

Boum, J. 1926. Uber Tertiare Versteinerungen von den Bogenfelser Diamantfeldern. Jn:

Kalser, E. ed. Die Diamantwiiste Stidwest-Afrikas 2: 55-87. Berlin: Dietrich Reimer.

Boum, J. & WEISFERMEL, W. 1913. Uber Tertiare Versteinerungen von den Bogenfelser

Diamantfeldern. Jn: Lotz, H., BOHM, J. & WEISFERMEL, W. Geologische und Palaonto-

logische Beitrage zur Kenntnis der Lideritzbuchter Diamantablagerungen. Beitrdge zur

geologischen Erforschung der Deutschen Schutzgebiete 5: 57-83.

CaRRINGTON, A. J. & KENSLEY, B. 1969. Pleistocene molluscs from the Namaqualand coast.

Annals of the South African Museum 52 (9): 189-223.

DE ViLueRS, G. 1975. Reproduction of the white sand mussel Donax serra Roding.

Investigational Reports, Sea Fisheries Branch, South Africa 102: 1-33.

DINGLE, R. V., SIESSER, W. G. & Newton, A. R. 1983. Mesozoic and Tertiary geology of

southern Africa. Rotterdam: A. A. Balkema.

FLEMMING, B. W. 1977. Langebaan Lagoon: a mixed carbonate-siliciclastic tidal environment

in a semi-arid region. Sedimentary Geology 18: 61-95.

Haucuton, S. H. 1932. The Late Tertiary and Recent deposits of the west coast of South

Africa. Transactions of the Geological Society of South Africa 34: 19-57.

HENDEY, Q. B. 198la. Palaeoecology of the Late Tertiary fossil occurrences in ‘E’ Quarry,

Langebaanweg, South Africa, and a reinterpretation of their geological context. Annals of

the South African Museum 84 (1): 1-104.

HENDEY, Q. B. 1981b. Geological succession at Langebaanweg, Cape Province, and global

events of the Late Tertiary. South African Journal of Science 77: 33-38.

HENDEY, Q. B. & Cooke, H. B. S. 1985. Kolpochoerus paiceae (Mammalia, Suidae) from

Skurwerug, near Saldanha, South Africa, and its palaeoenvironmental implications. Annals

of the South African Museum 97 (2): 9-56.

HENDEY, Q. B. & VotmaNn, T. P. 1986. Last interglacial sea-levels and coastal caves in the

Cape Province, South Africa. Quaternary Research 25 (2): 189-198.

Herm, D. 1969. Marines Pliozén und Pleistozan in Nord- und Mittel-Chile unter besonderer

Bericksichtigung der Entwicklung der Mollusken-Faunen. Zitteliana 2: 1-159.

KEEN, M. 1969. Veneracea, pp. N670—N690. In: Moors, R. C., ed. Treatise on Invertebrate

Paleontology, Part N, Vol. 2. Geological Society of America, Inc., and The University of

Kansas.

KENSLEY, B. 1972. Pliocene marine invertebrates from Langebaanweg, Cape Province. Annals

of the South African Museum 60 (4): 173-190.

KENSLEY, B. 1973. Sea-shells of southern Africa. Gastropods. Cape Town: Maskew Miller

Limited.

KENSLEY, B. 1977. A second assemblage of Pliocene invertebrate fossils from Langebaanweg,

Cape. Annals of the South African Museum 72 (10): 189-210.

KENSLEY, B. 1985a. The faunal deposits of a Late Pleistocene raised beach at Milnerton, Cape

Province, South Africa. Annals of the South African Museum 95 (2): 111-122.

KENSLEY, B. 1985b. The fossil occurrence in southern Africa of the South American intertidal

mollusc Concholepas concholepas. Annals of the South African Museum 97 (1): 1-7.

KENSLEY, B. & PENRITH, M. J. 1977. Biological survey of Sandvis 1, Introduction and faunal

list. Madoqua 10 (3): 181-190.

KENSLEY, B. & PENRITH, M.-L. 1970. New records of Mytilidae from the northern South West

African coast. Annals of the South African Museum 57 (2): 15-24.

Kitpurn, R. N. 1980. Taxonomic studies on the marine Mollusca of southern Africa and

Mozambique. Part 2. Annals of the Natal Museum 24 (1): 193-200.

KiLBurNn, R., & Rippey, E. 1982. Sea shells of southern Africa. Johannesburg: Macmillan

South Africa.

Kitpurn, R. & TANKARD, A. J. 1975. Pleistocene molluscs from the west and south coasts of

the Cape Province, South Africa. Annals of the South African Museum 67 (6): 183-226.

Krice, A. V. 1927. An examination of the Tertiary and Quaternary changes of sea-level in

South Africa, with special stress on the evidence in favour of a recent world-wide sinking of

ocean-level. Annals of the University of Stellenbosch 5 (A1): 1-81.

LaporEL, J. 1974. West African reef corals: an hypothesis on their origin. Proceedings of the

Second International Coral Reef Symposium. 1. Great Barrier Reef Committee, Brisbane.

1974, pp. 425-443.

Linpsay, E. H., Oppykxe, N. D. & JoHNson, N. M. 1980. Pliocene dispersal of the horse

Equus and Late Cenozoic mammalian dispersal events. Nature, London 287: 135-138.

FOSSIL MOLLUSCA OF HONDEKLIP DDS

Newton, R. BULLEN. 1913. On some Kainozoic shells from South Africa. Records of the

Albany Museum 2 (5): 315-352.

NickLés, M. 1950. Manuels ouest-Africains. II. Mollusques testacés marins de la céte occidentale

d’ Afrique. Paris: Paul Lechevalier.

Otson, S. 1985. An Early Pliocene marine avifauna from Duinefontein, Cape Province, South

Africa. Annals of the South African Museum 95 (4): 147-164.

PAES DA Franca, M. L. 1960. Contribugao para o conhecimento da Fauna Malacoldgica de

Angola. Trabalhos do Centro de Biologia Piscatoria 26: 1—40.

SIESSER, W. G. & DINGLE, R. V. 1981. Tertiary sea-level movements around southern Africa.

Journal of Geology 89: 83-96.

TANKARD, A. J. 1975a. Thermally anomalous Late Pleistocene molluscs from the south-

western Cape Province, South Africa. Annals of the South African Museum 69 (2): 17-45.

TANKARD, A. J. 19756. The marine Neogene Saldanha Formation. Transactions of the

Geological Society of South Africa 78: 257-264.

TANKARD, A. J. 1975c. The late Cenozoic history and palaeoenvironments of the coastal

margin of the north-western Cape Province, South Africa. Rhodes University: Unpublished

Ph.D. thesis.

VaIL, P. R. & HARDENBOL, J. 1979. Sea-level changes during the Tertiary. Oceanus 22: 71-79.

Woops, H. 1908. Echinoidea, Brachiopoda, and Lamellibranchia from the Upper Cretaceous

limestone of Need’s Camp, Buffalo River. Annals of the South African Museum 7: 13-20.

6. SYSTEMATIC papers must conform to the Jnternational code of zoological nomenclature (particu-

larly Articles 22 and 51).

Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed

by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov.,

ete:

An author’s name when cited must follow the name of the taxon without intervening punctuation

and not be abbreviated; if the year is added, a comma must separate author’s name and year. The

author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is trans-

ferred from its original genus. The name of a subsequent user of a scientific name must be separated

from the scientific name by a colon.

Synonymy arrangement should be according to chronology of names, i.e. all published scientific

names by which the species previously has been designated are listed in chronological order, with all

references to that name following in chronological order, e.g.:

Family Nuculanidae

Nuculana (Lembulus) bicuspidata (Gould, 1845)

Figs 14-15A

Nucula (Leda) bicuspidata Gould, 1845: 37.

Leda plicifera A. Adams, 1856: 50.

Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b).

Nucula largillierti Philippi, 1861: 87.

Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9.

Note punctuation in the above example:

comma separates author’s name and year

semicolon separates more than one reference by the same author

full stop separates references by different authors

figures of plates are enclosed in parentheses to distinguish them from text-figures

dash, not comma, separates consecutive numbers.

Synonymy arrangement according to chronology of bibliographic references, whereby the year is

placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable.

In describing new species, one specimen must be designated as the holotype; other specimens

mentioned in the original description are to be designated paratypes; additional material not regarded

as paratypes should be listed separately. The complete data (registration number, depository, descrip-

tion of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.:

Holotype

SAM-A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach, Port Eliza-

beth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973.

Note standard form of writing South African Museum registration numbers and date.

7. SPECIAL HOUSE RULES

Capital initial letters

(a) The Figures, Maps and Tables of the paper when referred to in the text

e.g. ‘.. . the Figure depicting C. namacolus ...’: ‘. . . in C. namacolus (Fig. 10)...’

(b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by

initials or full names

e.g. DuToit but A.L. du Toit; Von Huene but F. von Huene

e.g. Therocephalia, but therocephalian

Punctuation should be loose, omitting all not strictly necessary

Reference to the author should preferably be expressed in the third person

Roman numerals should be converted to arabic, except when forming part of the title of a book or

article, such as

‘Revision of the Crustacea. Part VIII. The Amphipoda.’

Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial

capital letter, provided the same generic name is used consecutively. The generic name should

not be abbreviated at the beginning of a sentence or paragraph.

Name of new genus or species is not to be included in the title; it should be included in the abstract,

counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts.

BRIAN KENSLEY

JOHN PETHER

LATE TERTIARY AND EARLY QUATERNARY

FOSSIL MOLLUSCA OF THE HONDEKLIP AREA,

CAPE PROVINCE, SOUTH AFRICA

| 97 PART 7 AUGUST 1986

ISSN 0303-2515

OF THE SOUTH AFRICAN

MUSEUM

CAPE TOWN

INSTRUCTIONS TO AUTHORS

1. MATERIAL should be original and not published elsewhere, in whole or in part.

2. LAYOUT should be as follows:

(a) Centred masthead to consist of

Title: informative but concise, without abbreviations and not including the names of new genera or species

Author’s(s’) name(s)

Address(es) of author(s) (institution where work was carried out)

Number of illustrations (figures, enumerated maps and tables, in this order)

(b) Abstract of not more than 200 words, intelligible to the reader without reference to the text

(d) Introduction ;

(e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents

(f) Summary, if paper is lengthy

(g) Acknowledgements

(h) References

(i) Abbreviations, where these are numerous.

3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with

3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for

illustrations should be typed separately, their positions indicated in the text. All pages should be num-

bered consecutively.

Major headings of the paper are centred capitals; first subheadings are shouldered small capitals;

second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further

subdivisions should be avoided, as also enumeration (never roman numerals) of headings and

abbreviations.

Footnotes should be avoided unless they are short and essential.

Only generic and specific names should be underlined to indicate italics; all other marking up

should be left to editor and publisher.

4. ILLUSTRATIONS should be reducible to a size not exceeding 12 x 18cm (19 cm including ~

caption); the reduction or enlargement required should be indicated (and preferably uniform); orig-

inals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and

final size. A metric scale should appear with all illustrations, otherwise magnification or reduction

should be given in the caption; if the latter, then the final reduction or enlargement should be taken

into consideration.

All illustrations, whether line drawings or photographs, should be termed figures (plates are not

printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items

of composite figures should be designated by capital letters; lettering of figures is not set in type and

should be in lower-case letters. If Letraset is used authors are requested to use Helvetica-style letter-

ing, if possible.

The number of the figure should be lightly marked in pencil on the back of each illustration.

5. REFERENCES cited in text and synonymies should all be included in the list at the end of the

paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable):

(a) Author’s name and year of publication given in text, e.g.:

‘Smith (1969) describes .. .’

‘Smith (1969: 36, fig. 16) describes. . .’

‘As described (Smith 1969a, 1969b; Jones 1971)’

‘As described (Haughton & Broom 1927)...’

‘As described (Haughton ef al. 1927) ..

Note: no comma separating name and year

pagination indicated by colon, not p.

names of joint authors connected by ampersand

et al. in text for more than two joint authors, but names of all authors given in list of references.

(b) Full references at the end of the paper, arranged alphabetically by names, chronologically within

each name, with suffixes a, b, etc., to the year for more than one paper by the same author in

that year, e.g. Smith (1969a, 19695) and not Smith (1969, 1969a).

For books give title in italics, edition, volume number, place of publication, publisher.

For journal article give title of article, title of journal in italics (according to the World list of scientific periodicals. 4th ed.

London: Butterworths, 1963), series in parentheses, volume number, part number in parentheses, pagination (first and

last pages of article).

Examples (note capitalization and punctuation)

BuLtouGu, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan.

FISCHER, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140.

FiscHeR, P. H., DuvaL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie

expérimentale et générale 74 (33): 627-634.

Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and

Magazine of Natural History (13) 2 (17): 309-320.

Koun, A. J. 19606. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of

the Bingham Oceanographic Collection, Yale University 17 (4): 1-S1.

THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthro-

pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Stid-Afrika ausgefiihrt in den Jahren

1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270.

(continued inside back cover)

ANNALS OF THE SOUTH AFRICAN MUSEUM

ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

Volume 97 Band

August 1986 Augustus

Part 7 Deel

Alig, AWS

LANTERNFISHES OF

THE SOUTHERN BENGUELA REGION

PART 1

FAUNAL COMPLEXITY AND DISTRIBUTION

P. ALEXANDER HULLEY

Cape Town Kaapstad

The ANNALS OF THE SOUTH AFRICAN MUSEUM

are issued in parts at irregular intervals as material

becomes available

Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000

Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

word uitgegee in dele op ongereelde tye na gelang van die

beskikbaarheid van stof

Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000

OUT OF PRINT/UIT DRUK

MAGES), 5) BG, 25.8, haos)), SOLS, 5, 7D),

Gay to), HOD), 8, CAD, 2), IOC),

11(1-2, 5, 7, t.--p.i.), 14(1-2), 15(4-5), 24(2), 27, 31(1-3), 32(5), 33, 36(2), 45(1)

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 084 2

Printed in South Africa by In Suid-Afrika gedruk deur

The Rustica Press, Pty., Ltd., Die Rustica-pers, Edms., Bpk..,

Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap

LANTERNFISHES OF THE SOUTHERN BENGUELA REGION

PART 1

FAUNAL COMPLEXITY AND DISTRIBUTION

P. ALEXANDER HULLEY

Department of Marine Biology, South African Museum, Cape Town

(With 2 figures and 8 tables)

[MS accepted 27 February 1986]

ABSTRACT

Various sampling cruises in the southern Benguela region (28°40’S—40°00’S

10°00’E-20°00’E) yielded a total of more than 17 000 lanternfishes (family Myctophidae). These

comprise 65 species in 23 genera. Their horizontal distribution in the region is discussed in terms

of their known Atlantic and Indo-Pacific ranges and their vertical distribution in relation to

oceanic and pseudoceanic zonality. Results indicate that inshore of the 300m isobath,

lanternfishes are represented by a single, pseudoceanic species, Lampanyctodes hectoris. Oceanic

species occur where bottom depths exceed this value. The sampling strategies employed preclude

investigation of the diurnal relationship between mesopelagic and bathypelagic species. The

southern Benguela region can be characterized as a transition zone rather than a subtropical

zone. There is a strong intrusion of convergence and semisubantarctic species in association with

cold-core eddies from the south, and a much weaker advection of tropical and broadly tropical

species in Agulhas Water. A first estimate of the offshore lanternfish stock for the eastern South

Atlantic is calculated as 8-12 x 10° tonnes which represents 50-70 per cent of the total estimated

mesopelagic fish stock of the offshore eastern South Atlantic.

CONTENTS

PAGE

LINGO GM CHON Ree Re MAR a nee nn ace Ml Sead a, 227

Matenalrandimethods. 555 sens ec soe ete le hoes eka ee ews 229

CSUltSee Ine re er rs ee a Me 230

DISCUSSION een UE MN 8 a rR he Ct A 238

INCKNOWIECESIMEMUS Dey ty- neces cronies eg asianinye nsleha ale ts 246

ENGL TCH Ce Siar ane hea ene moh cera Re ee eet een a 246

INTRODUCTION

In order to assess the status of commercially exploited marine resources, the

Sea Fisheries Research Institute (SFRI), Cape Town, has an ongoing series of

routine sampling cruises in the eastern South Atlantic. Material and data arising

from these cruises play a significant role in the CSIR-SANCOR-sponsored

Benguela Ecology Programme, in which the structure and dynamics of the

Mpay]

Ann. S. Afr. Mus. 97 (7), 1986: 227-249, 2 figs, 8 tables.

228 ANNALS OF THE SOUTH AFRICAN MUSEUM

Benguela Upwelling System is being investigated. Lanternfishes (family

Myctophidae) are a regular and abundant component of the catch from both the

pelagic and benthic sampling. They represent not only an important group of

trophic organisms within the system, but also represent an alternative fishery

resource to more conventional species like anchovy (Engraulis capensis) and

pilchard (Sardinops ocellata).

Ahlstrom et al. (1976) reported that in the eastern Atlantic myctophids

comprise nearly 10 per cent of all fish larvae caught between 19° and 26°S, with

larval abundance values greater than 10 larvae/m? for the period August 1973 to

April 1974. Recent surveys have indicated a larger proportion of myctophid

larvae (29%) in Bongo net catches off the west coast (R. A. Cruickshank pers.

comm.). Lanternfishes (mainly Lampanyctodes hectoris) were first recorded in

the South African purse-seine catches in 1969, when 1 134 tons were taken (0,3%

of the total catch). Subsequently, the catch has fluctuated, with a maximum of

42 560 tons being taken in 1973 (10,45%) (De Villiers 1982). A fishing quota of

50 000 tonnes has been allocated for the 1985 season.

Hulley (in press) has pointed out that there are 28 genera and 125 species of

myctophid known, or likely to be found, in the southern African region, and has

given a general account of their taxonomy and distribution. Some results have

also been presented on the myctophid fauna of the southern Benguela region

(Hulley 1972a, 1972b, 1981, 1986), but data from recent cruises allow for a more

critical investigation of the structure and distribution of the lanternfish fauna.

Rubiés (1985) has recently examined the myctophid fauna off the South West

African—Namibian coast, recording a total of 41 species. Twenty-five of these

species were taken only from the Valdivia Bank area on the Walvis Ridge,

ten species only from the Benguela area, and six species were common to both

areas.

The purpose of this paper is threefold: firstly, to establish the myctophid

species complexity in the southern Benguela region, i.e. between 28°40’S (mouth

of the Orange River) and 40°00’S, and between 10°00’E and 20°00’E; secondly, to

examine the nature of the distribution patterns of these species, including both

the oceanic and pseudoceanic zones; and thirdly, to obtain a first estimate of

lanternfish abundance in the offshore area (greater than 100 miles offshore) of

the South-east Atlantic. Here the total offshore mesopelagic fish stock has been

estimated at 16 x 10° tonnes (Gjdsaeter & Kawaguchi 1980).

Investigations of the biology of Lampanyctodes hectoris, leading to estimates

of inshore lanternfish abundance for the same region, are to be made

independently by scientists at the Sea Fisheries Research Institute, Cape Town;

these will be reported separately.

The areal choice in the present paper should also allow for the examination

of distributional phenomena across the frontal zone(s) between upwelled

Benguela Water and South Atlantic Central Water in the region of the shelf break

at 300-500 m (Shannon 1985), but this aspect will be more fully developed in

later publications.

SOUTHERN BENGUELA LANTERNFISHES 229

MATERIAL AND METHODS

Specimens and data on specimens from the following cruises and from

stations occupied with the following gear within the region have been incorpor-

ated into the analysis:

Walther Herwig 1971—Transect II (WH-71) MT-—1600 (March)

Walther Herwig 1971—Transect II David Net (March)

SFRI Hake Survey 1984 (HJUL84) BT-180 (July)

SFRI Hake Survey 1985 (HJAN85) BT-180 (January)

SFRI Phyllosoma Survey 1982 (PAUG82) RMT-2 (August)

SFRI Phyllosoma Survey 1983 (PAUG83) RMT-8 (August)

SFRI Phyllosoma Survey 1984 (PAUG84) RMT-8 (August)

SFRI Anchovy Acoustic Survey 1983 (ANAC83) RMT-8 (May)

In addition, the lanternfish material in the collections of the South African

Museum, Cape Town, taken within the defined southern Benguela region, has

been re-examined for the purposes of this paper. These specimens were taken

with a variety of gear-types: neuston net; N100B; N200B; IKMT.

Gear-type abbreviations are as follows:

BT-180: German bottom trawl with 180’ headline and stocking

David net: modified David neuston sampler

IKMT: 10’ Isaacs-Kidd midwater trawl

MT-—1600: Engel midwater trawl with 1 600-mesh circumference

RMT-2: rectangular midwater trawl with 2 m* mouth opening

RMT-8: rectangular midwater trawl with 8 m? mouth opening

All specimens were identified to species. Standard lengths (SL), and in

certain instances preserved wet weights, were taken for each specimen, resulting

in data on more than 17 000 specimens. Taxonomic details for individual species

are not given. However, the Smith’s sea fishes (SFSA) species number (Hulley in

press) is given, so that readers may refer to that publication for those details.

For all cruises, daytime hauls were distinguished from night hauls on the

basis of commencing after 06h00 or before 18h00 (local time). Myctophids were

absent from six (86%) of the day hauls from PAUG82 and eight (57%) day hauls

from PAUG83; no day hauls were undertaken during PAUG84. For the purposes

of stock estimation, distribution pattern and subpattern catch rates (specimens/

hour) were calculated for night hauls only, according to the method of

Hulley & Krefft (1985), for species from each of the RMT-2 (PAUG82) and

RMT-8 (PAUG83 and PAUG84) nets, and for both day (1 haul, fishing

depth >1 000 m) and night hauls for species from the MT-1600 (WH-—71) net.

These catch rates were converted into abundances in specimens/1 000 m’,

assuming that each gear was 100 per cent effective for the duration of the haul,

that the fishes were evenly distributed in the upper 1 000 m, and that the mouth

areas and mean trawling speeds were 2 m* and 2,5 knots (RMT-2), 8 m* and

2,5 knots (RMT-8), and 300 m? and 3,5 knots (MT-—1600). On the basis of SL/

weight scatter plots for Lampanyctodes hectoris (Fig. 1) and inspection of the

230 ANNALS OF THE SOUTH AFRICAN MUSEUM

standard lengths of specimens taken by each gear, the mean weight of a fish from

the RMT-2 samples was accorded a value of 0,2 g, from the RMT-8 samples a

value of 0,5 g, and from the MT-—1600 samples a value of 1,0 g. The area of the

offshore South-east Atlantic region is taken as 160 Xx 10!! m* (Gjdésaeter &

Kawaguchi 1980).

RESULTS

Haul data, which includes discrimination by depth and/or time of day, is

given in Tables 1 to 6. Positive hauls indicate the presence of myctophids. Table 7

is a species list of Myctophidae for the southern Benguela region and incorporates

the Atlantic distribution pattern and subpattern placement of each species

according to Hulley (1981), the number of specimens examined for each species,

and the SFSA species number.

BT-180

Catch data for this gear (Tables 1, 2) indicate that although sampling was

carried out from depths less than 101 m to greater than 500 m, the major fishing

effort was concentrated between depths of 101-200 m (35% of total) and

210-300 m (30-35%). During both cruises, stations were occupied mainly during

daylight hours (95% of total number). Therefore, although the results for night

hauls should be regarded as tentative, it would appear that the epibenthic

(pseudoceanic) lanternfish fauna exhibits diurnal migration into the water column

during this period, since no specimens were taken during the night at depths of

maximum daytime abundance (101-300 m). Furthermore during the day

myctophids were taken only in 41-52 per cent of the hauls, pointing to the possi-

bility of an extremely patchy distribution (see below). The major component of

the lanternfish catch throughout the year was Lampanyctodes hectoris, whose

highest mean catch rates were in the 101—200 m and 210-300 m depth ranges.

Day mean catch rates at these depths were 4,9 and 5,6 specimens/hour

respectively in winter, and >153,9 and 73,6 specimens/hour respectively in

summer, suggesting a marked seasonal variation in availability and/or distribu-

tion. More recent unpublished data from the winter (July) 1985 Hake Survey

support this suggestion. In the 301—400 m, 401-500 m and >500 m depth strata,

mean catch rates for Lampanyctodes hectoris were considerably lower (winter:

7,7; 1,1; 0,3 specimens/hour; summer: >13,8; >2,9; 0,0 specimens/hour respec-

tively). Other myctophid species were taken in those bottom hauls fished at

depths greater than 301m (Diaphus hudsoni (2 specimens), D. meadi (1),

D. ostenfeldi (1), Electrona risso (2), Gymnoscopelus (Nasolychnus) piabilis (2),

Scopelopsis multipunctatus (1), Symbolophorus barnardi (4), S. boops (7)) but

these probably represent contaminants as the net is heaved from depth, since

mean catch rates for the individual species range only between 0,1 and

1,0 specimens/hour (x =0,3 specimens/hour) and there is a decrease in the

percentage of negative hauls at depths below 301 m. The high percentage of

773)

SOUTHERN BENGUELA LANTERNFISHES

ddS_ dS ddS dS #

Ww QOS—LOV WwW 0Or-10E

0090-0081

S8sL< 9 LOL) S8NVIH

008T—0090

0090-008T

v8 TO fH

008T—0090

ddS dS #| ddS_ dS

asin

W 00E—-L07 wW 0Oc—-TOL

‘suouttoods Jo 1aquinu = gd§ ‘soiseds jo Joquinu = gS ‘s[ney Jo Joquinu = #

‘sk9AING OYLH [TWAS sulnp j0u Q8T—Lg@ Woy eyep prydojoAu Jo siskjeuy

7 AAV,

0090-008T

S8NVIH

008T—0090

SOF 3c Se WM 8 0 O 0 iE () tf

G 0 & 0 © 0 GY “@

ve S 6¢ G6 “Ve v ID GSE Bs 0081—0090

See

(1090})

0090-008T

p38 TOfH

asinia

ul Qos < WwW 00S—-T0P W OOr—-T0E WwW (Q0E—-107 wW 00C-I0T Ww 00T—-0 | ea | aul]

‘sIney oANesou Jo Joquinu = — ‘s[ney sATyIsod Jo 1oquinu = + ‘s[ney Jo Joquinu = #

‘sA9AINS OAL [YAS suLNp j0u QZT—Lgq Wor eyep prydojoAw Jo siskjeuy

] q1avL

232 ANNALS OF THE SOUTH AFRICAN MUSEUM

25 38 50 63 iS

OL (mm)

Fig. 1. Scatter plot of standard length (SL) versus preserved wet weight for Lampanyctodes

hectoris.

negative hauls in the 101-200 m (71-73%) and 201-300 m (66-83%) depth

strata can therefore only be accounted for by the absence of a single species,

Lampanyctodes hectoris, and Index of Dispersion (ID) values (Wormuth &

Roper 1983) considerably greater than 1,0 indicate that the distribution of

Lampanyctodes hectoris is strongly patchy within its principal epibenthic

distributional depth range (101-300 m). Further development of the data for this

species is Outside the scope of the paper, although it should be mentioned that

Lampanyctodes hectoris was also taken in pelagic hauls, particularly those inside

or immediately adjacent to the frontal system developed by the upwelled

Benguela Water. These specimens are probably associated with the system of

cold-core eddies generated at the front (Lutjeharms 1981a).

RMT-2 and RMT-8&

Catch data for these nets from the SFRI Phyllosoma Surveys (PAUG82,

PAUG83 and PAUG§884) are given in Tables 3 and 4. The major sampling effort

was directed at depths between 200-0 m and 75-0 m (63% of total) and only one

haul was made below 500 m (day). RMT-2 nets also sampled the 10—0 m (11%)

and 50—0 m (16%) strata—depths not specifically fished during the 1983 and 1984

RMT-—8 sampling programme, although nine oblique hauls were made with the

latter gear from a maximum depth of 250 m.

233

SOUTHERN BENGUELA LANTERNFISHES

suoumoads 98 | 0 0 0 0 0 0 0 0 0 0 S 0

saroeds TT] | 0 0 0 0 0 0 0 0 0 0 0 0

siney pT | 0 0 0 0 0 0 0 0 0 0 é 0

Vv dnouD |0 0 0 0 0 0 0 0 0 0 0 p

suowoads EQ I | 0 0 0 0 0 0 0 0 0 0 0 0

soroads (¢ | 0 0 0 0 0 0 0 0 0 0 0 8

siney 61 |0 0 0 0 0 0 0 0 0 0 i 0

ad dnouo |0 0 0 0 0 0 0 0 0 0 0 8

0 0 0 0 0 0 0 0 0 I 0

0 0 0 0 0 0 0 0 0 0 0 8

0 0 0 0 0 0 0 0 0 0 I 0

0 0 0 0 0 0 0 0 0 S i 0

0 0 0 0 0 0 0 0 I 0 0 0

suoumtoeds [gg Z 0 0 0 0 0 0 0 0 Z 0 0 0

saroeds gp 0 0 0 0 0 0 0 0 0 0 0 6

siney LL 61 8 0 0 0 0 0 0 0 0 0 0

9 dNOUD 0 0 7 I 9 0 0 0 6 0 0 0

0 0 0 0 0 0 0 0 I 0 0 0

0 0 0 0 0 0 0 0 0 0 0 8

0 0 0 0 0 0 0 0 0 8 G 0

0 0 0 0 0 0 0 0 0 0 0 I

0 0 0 0 0 0 0 0 € 0 0 0

0 0 0 0 0 0 0

suowmloads ¢eq | 0 0 0 0 0 0 0 0

soroads 6] 0 0 0 0 0 0 0

siney TT 0 0 0 0 0 0 0

d dnouo 0 0 0 0 0 I 0

0 0 0 0 0 0 0

suourtoads ()

saoads ¢ ‘jney T

qH dnowdo

dnoi9 |\ddS_ dS ddS_ dS ddS_ dS rs ddS_ dS Vv

7861 C861 C861

8-LNA 8-LNA C~ LNG

‘suourtoads Jo Joquinu = gd§ ‘soteds jo 1oquinu =

‘yjdap 3urysy 0} sjney onbijqo Jo Joquinu = gq ‘poj1y dolAop ae ieee fe 2h sjney Jo Joquinu = V

‘sAOAINS BUIOSOT[AUG [TYAS suLINp sjou S—TINY pue Z—LINY Woy eiep prydojsAw Jo sishjeuy

€ ATav |,

0-01

0-S7¢

0-0S

O0-SL

0-08

0-001

OT—-00T

Sc—-00T

O01

0-0C1

0c—-0CT

OT—-OST

OST

0-L61

L6l

00T—002

0-0S¢

OST—OS7

00C—00€

00-007

0-00S

O00vV—00S

00S-009

(uw)

yidaq

ANNALS OF THE SOUTH AFRICAN MUSEUM

234

‘s[ney sanesou Jo Joquinu = — ‘sney sarzIsod Jo Joquinu = + ‘sjney Jo Joquinu = #

‘skaAIng eUIOSOT[AUg TYAS suLInp sjou g—TIWY pue Z—LINY woy eyep prydojoAw jo siskjeuy

CCKAD

0090-008T

008T—0090

0090-0081

0081-0090

0090-008T

008T—0090

SOUTHERN BENGUELA LANTERNFISHES D5)

RMT-2 data reveal a high proportion of negative daytime hauls (86% of

PAUG82 cruise), the single positive sample (09h50: 150-0 m) yielding four

specimens each of Diaphus hudsoni and Lampanyctus lepidolychnus. Negative

daylight hauls with the RMT-8 were lower (57%) during the 1983 cruise

(PAUG83); no daytime sampling was undertaken during PAUG84. Two positive

daytime hauls (P 011: 06h31—07h31; P 024: 15h00-16h00) were made in the

200-0 m depth range during PAUG83 and yielded a total of 38 specimens

(Benthosema suborbitale, Ceratoscopelus warmingil, Diaphus hudsoni, D. meadi,

Lampanyctus pusillus, — Lobianchia dofleini, Metelectrona _ ventralis,

Symbolophorus boops), while 102 specimens were recorded from the four

positive daytime stations in the 500—0 m depth range during the same cruise. The

species recorded at these latter stations were Benthosema suborbitale, Cerato-

scopelus warmingii, Diaphus effulgens, D. hudsoni, D. lucidus, D. meadi,

D. metopoclampus, D. ostenfeldi, Diogenichthys atlanticus, Hygophum hygomit,

Lampadena notialis, Lampanyctus alatus, L. pusillus, Lobianchia dofleini,

Protomyctophum (Hierops) subparallelum, Scopelopsis multipunctatus and

Symbolophorus barnardi.

For night sampling, 89 per cent and 92—96 per cent of stations occupied with

the RMT-2 and RMIT-8 respectively, yielded lanternfishes. No definite

scattering layer was observed at station 9 (4)—the single negative night station of

the PAUG84 cruise, while the single negative night haul (P 003: 04h13-05h13)

from the PAUG83 cruise was in 120-0 m. Seven negative night hauls were

recorded during PAUG82, five (71%) of which were 10—0 m or at the surface; the

net bar was bent during the single negative night haul (Station Number 002030) in

the 50-0 m depth range. The number of hauls, number of species, and number of

specimens for both gear during the PAUG82, PAUG83, and PAUG§84 cruises is

summarized in Table 3.

RMT-8 data from ANAC83 are included in Table 6. A total of 2 248 speci-

mens was taken during this cruise, in which the net was aimed at target species,

the duration of the haul being one hour, and the maximum fishing depth not

exceeding 100 m. Thirty-eight specimens of Lampanyctodes hectoris from four

hauls lack accurate depth data but have been included at the 0 m depth only for

the sake of completeness. The four hauls in the 50-0 m fishing-depth range

yielded 643 specimens of one species (Lampanyctodes hectoris), but 35 hauls in

the 100-—O m range caught 1 563 specimens of Lampanyctodes hectoris, two

Diaphus hudsoni, and one specimen of each of Diaphus meadi and Hygophum

hanseni. Further analysis of these data has revealed that species other than

Lampanyctodes hectoris were caught in the upper 100 m only at those stations

where the bottom soundings were in excess of about 500 m.

For Lampanyctodes hectoris, the highest mean catch rate (130,14 specimens/

station; number of stations=7) was obtained inside of the 100 m isobath.

Between the 100m and 200m isobaths the mean catch rate was

52,76 specimens/station (n= 21); between 200 m and 300 m, 34,00 specimens/

station (n=3); between 300m and 400m, 15,25 specimens/station (n= 4);

236 ANNALS OF THE SOUTH AFRICAN MUSEUM

between 400 m and 500 m, 11,33 specimens/station (n=3); and greater than

500 m, 3,33 specimens/station (n = 3). These data suggest that, for RMT-8 hauls

at least, the major pelagic concentrations of Lampanyctodes hectoris are shore-

ward of the 100 m isobath and that ID values are considerably greater than 1,

i.e. a marked patchiness.

During this same cruise, two hauls were made at the surface with a neuston

net and yielded a total of 147 Symbolophorus boops, five Symbolophorus

barnardi, and one Lampanyctodes hectoris. Both hauls were deployed over

bottom depths of 510—640 m.

MT-1600

During the course of Transect II of the 1971 cruise of FRV Walther Herwig,

six stations were occupied in the defined area with an MT-1600 net (Table 5).

Shallower hauls (112—0 m, 305—0 m, 592-0 m) were made at night, while a single

deep haul (>1 000-0 m) was made during the day. A total of 2 310 specimens in

46 species was taken. Of these species, only Diaphus richardsoni, Electrona

carlsbergi, Lampanyctus lineatus and L. nobilis were not collected by the other

types of gear. Lampanyctodes hectoris was not recorded from any of the

MT-—1600 stations, which were situated at distances of greater than 95 sea miles

offshore, where bottom depths exceeded 3500m. Hulley (1981) gave a

breakdown of the species distributions at these stations.

Other gear

The myctophid specimens taken in the southern Benguela region during the

deep-sea cruises of Africana II (Table 6: N200B, IKMT.) have been described by

Hulley (1972a). The nine IKMT oblique hauls (5 day; 2 night; 1 day—night

(15h25-19h10); and 1 time unknown) from 1 000-0 m yielded 138 specimens

comprising 29 species, while the single IKMT oblique haul (day) from 1 400—0 m

yielded 16 specimens comprising nine species. The species Lobianchia dofleini,

Protomyctophum (Protomyctophum) andriashevi, and Scopelopsis multipuncta-

tus, taken in the latter haul, were not recorded from the shallower hauls. The

single N200B haul (day) from 823-0 m yielded one specimen each of Diogenich-

thys atlanticus and Lampanyctus alatus, both species being recorded also from the

IKMT samples of this cruise.

A haul analysis for the lanternfishes from the South African Museum’s IK

stations in the region (Grindley & Penrith 1965) is given in Table 6 (IKMT,). The

26 stations (11 day; 15 night) occupied with the gear during the sampling

programme yielded 141 specimens (18 species). Most stations (22) were

positioned west of Slangkop (34°09’S 18°19’E). This material was reworked

because of errors in identification and nomenclature (see synonymies in Hulley

in press). No lanternfish specimens were obtained from the day hauls, fished

obliquely to a maximum of 500 m. Myctophids were caught only at five of the

night stations, the 82 specimens from the 0-50 m depth strata consisting of a

single species, Lampanyctodes hectoris. The following species were taken in hauls

23,

SOUTHERN BENGUELA LANTERNFISHES

praeq

13910

‘LMI

"LAM

a007N

uo}snoN

aaa io Waa ian a>)

SQ So 29° &

C80VNV

8-— LNA

‘e1ep yidep ou = , ‘suouttoeds jo Joquinu = gd§ ‘soiseds Jo Joquinu = g§ ‘s[ney Jo Joquinu =

‘UOISOI gjonsuag uJoyINos sy} UI padojdeap sjou snoleA Wo eJep prydo}oAWI Jo es

w 000 I <—0 vay)

Q ATaV |,

0090-0081 TL61

009T-LINN

008T—0090

uw oool <

‘suouttoads Jo Joquinu = ggg ‘sotdeds jo Joquinu = gg ‘s[ney jo Joquinu = #

“UOISaI ejaNSUSg UIOYINOs UI sUONIRIS SimMiazy 4aYyJ]VA, WOT} JOU YOOT-LIN Wosy eIep prydoyoAu jo siskjeuy

¢ ATaV

238 ANNALS OF THE SOUTH AFRICAN MUSEUM

0-200 m and 0-500 m: Diaphus brachycephalus, D. hudsoni, D. lucidus,

D. meadi, D. metopoclampus, D. mollis, D. ostenfeldi, Hygophum hanseni,

H. hygomii, Lampanyctus alatus, L. australis, L. lepidolychnus, Lobianchia

dofleini, Protomyctophum (Protomyctophum) normani, Scopelopsis multipuncta-

tus, Symbolophorus barnardi and Triphoturus nigrescens.

Additional lanternfish material in the SAM collections (Table 6: Other) was

obtained from 11 hauls occupied with a variety of gear including bottom trawls

and N100B nets. Depth data (and in some instances gear type) were not available

for eight of these hauls, but the 346 specimens taken by these hauls have been

included in Table 6. These specimens comprise the following species: Diaphus

meadi (1), Gonichthys barnesi (293), Lampanyctodes hectoris (3), Lampanyctus

lepidolychnus (1), Symbolophorus barnardi (1), and S. boops (47). The single

species recorded from the two N100B hauls (0-200 m) and the one BT haul

(424 m) was Lampanyctodes hectoris. During the 1971 transect of FRV Walther

Herwig, 50 lanternfish specimens were taken at the surface with a David net

(Table 6): Gonichthys barnesi (42), Myctophum asperum (3), M. nitidulum (1)

and M. spinosum (4).

DISCUSSION

Any discussion of distribution in the family Myctophidae should take into

account that major ecological differences exist and that parameters affecting

distribution patterns in one community may not necessarily be the same in

another community. The community structure terminology used in this paper

follows Hulley (1981), in which it is suggested that myctophids may be divided

into (1) oceanic (=high-oceanic) mesopelagic and bathypelagic communities,

and (2) pseudoceanic epibenthic and pelagic communities. However, the

groupings within the pseudoceanic zone may not be as distinct. For example, it

would appear from the above results that Lampanyctodes hectoris, which is at

present regarded as a member of the pseudoceanic pelagic community, may be

taken in abundance in bottom trawls during the day but may move off the bottom

and into the upper 50m of the water column during the night. A similar

behaviour pattern has been reported for the Pacific pseudoceanic species Diaphus

watasei in Suruga Bay (Kawaguchi & Shimizu 1978). Lanternfish community

structure is further complicated by the fact that certain oceanic species, for

example Gymnoscopelus bolini, G. braueri, G. nicholsi, G. piabilis, Noto-

scopelus kroeyerii, may be caught (? seasonally) on upper-slope and outer-shelf

regions in potentially economic quantities (Dubrovskaya & Makorov 1969;

Hulley & Krefft 1985). In addition, other oceanic species (Diaphus dumerili,

Lampanyctus australis) may possess pseudoceanic populations (Hulley 1981;

Rubiés 1985). Comprehensive data on the life histories and reproductive biology

of these species is sparse, and more research will be required before delineation

of community structure in myctophids can be fully developed.

The interpretation of species complexity and general distribution in this

paper is constrained both by a ‘division of labour’ aspect (i.e. SFRI is

SOUTHERN BENGUELA LANTERNFISHES 239

investigating the biology of Lampanyctodes hectoris) and by the variety of

sampling strategies employed. Depths below 1 000m were not well sampled

(4 daylight hauls), and only 12 pelagic hauls (8 day; 3 night; 1 time unknown)

and 6 bottom hauls (all during the day) were made to depths between 500 m and

1 000 m. Therefore, since bathypelagic species are poorly represented in the

collections, the main thrust of the discussion will focus on the oceanic

mesopelagic community.

The results of the present paper, particularly those from the BT—180

samples, confirm the distinction of at least a pseudoceanic lanternfish community

and an oceanic community, the former characterized by Lampanyctodes hectoris.

The data does not allow for closer inspection of the relationships between

mesopelagic and bathypelagic species, except to indicate that there may be a

degree of depth separation at night (see Lampanyctus achirus below). However,

penetration of the upper 500 m of the water column by this bathypelagic species

does take place in the southern Benguela region.

The physical structure of the seas around South Africa, in particular the

South-east Atlantic and Benguela Upwelling Region, has been repeatedly

described (see Shannon 1985) and does not need to be given here. Only features

that are relevant to the interpretation of the distribution will be discussed.

Firstly, specimens of Indo-Pacific species are advected into the South-east

Atlantic in water that originates from the Agulhas Current (Heydorn 1959;

Krefft 1974; Weikert 1975; Hulley 1981, 1986; Bekker 1983; De Decker 1984).

Thermal infra-red imagery has indicated that at least two mechanisms exist for

this advection: (1) the growth, decay and dispersion of shearing eddies on the

northern border of the Agulhas Current, with the subsequent advection of

fragments into the South Atlantic; and (2) the production of Agulhas Water rings

from the Agulhas Current retroflection area south-west of South Africa

(Lutjeharms & Valentine 1981). On the basis of temperature recordings from a

satellite-tracked buoy placed in an Agulhas Current fragment, these authors

reported that the advected water lost its temperature characteristics (from 17°C

to 14°C) over a period of 40 days, owing to mixing with colder South Atlantic

water. Although Darbyshire (1966) has suggested that this advected Agulhas

Water may be detected as far north as 23°S, its distinction from aged, upwelled

Benguela Water at lower latitudes is difficult (Jones 1971). Shannon (1966) has

indicated that the advection of this Agulhas Water is at a maximum during

summer and autumn, and at a minimum during winter and spring.

Secondly, it would appear that some authors have characterized the offshore

region of the eastern South Atlantic, north of the Subtropical Convergence, as

subtropical (Boden 1951; Abrams et al. 1984; Abrams 1985). However, there is a

substantial intrusion of convergence and subantarctic faunal elements into the

southern Benguela region and even on to the South African shelf (Krefft 1974,

1978; Hulley 1981; McGinnis 1982; Bekker 1983; De Decker 1984), in association

with the north-east deflection of the isotherms and isohalines (D.H.I. Monatskar-

ten 1971). Drift-card observations have confirmed the northward flow of surface

240 ANNALS OF THE SOUTH AFRICAN MUSEUM

water from the Convergence towards the South African coast during the winter

months (Shannon et al. 1973), while subsurface investigations have demonstrated

the existence of a complex cold-core eddy system north of the Convergence to

about 31°S (Visser 1969; Shannon & Van Rijswijck 1969; Welsh & Visser 1970;

Henry 1975; Lenz 1975; Allanson etal. 1981; Lutjeharms & Emery 1983;

Lutjeharms et al. 1985). These cold-core eddies, which are formed from planetary

waves moving along the Convergence becoming unstable and losing their tops,

drift northward (Lutjeharms 19815). Although they extend to a depth well below

500 m and can be detected at 1 000 m, they become well mixed with South

Atlantic water (Welsh & Visser 1970). This suggests the probability of mixed

subtropical—temperate faunas in the study region and the possible existence of a

transitional zone similar to that reported off Peru (Parin et al. 1973).

Bathypelagic species (Table 7)

Of the five bathypelagic species known from the Atlantic and Indian oceans

(Gymnoscopelus opisthopterus, Lampadena anomala, Taaningichthys pauro-

lychnus, T. bathyphilus, Lampanyctus achirus), only the latter two species should

occur in the southern Benguela region (Hulley 1981; Bekker 1983). Isolated

specimens of the Antarctic species Gymnoscopelus opisthopterus have been taken

in trawls fished to below 2 000 m as far north as 40°20’S (Hulley 1981), but

sampling at these depths has not yet been undertaken in the southern Benguela

region. The deepest haul included in the present survey (WH 417/71) was fished

to 1550m. One specimen of Taaningichthys bathyphilus (58mm SL) was

recorded from 1 000-0 m at 30°19’S 10°08’E. Lampanyctus achirus was taken

throughout the region (31°19’S—39°06’S and 10°08’E-—17°11'E) and, except for a

single case, in depths greater than 592 m. However, Hulley (1981) has pointed

out that this species may be caught in depths shallower than 500 m in areas of

pronounced upwelling, which may account for the record of the specimen (48 mm

SL) at 33°44’S 17°11’E from 120 m.

Pseudoceanic species (Table 7)

These are species that are associated with land environments and land-

orientated food chains and are distributed on or over continental shelf and slope

regions and in the neighbourhood of oceanic islands. On a global basis, the

following pseudoceanic species of Myctophidae have been recognized: Diaphus

adenomus, D. coeruleus, D. garmani, D. knappi, D. minax, D. roei, D. sagami-

ensis, D. suborbitalis, D. taaningi, D. umbroculus. D. watasei, Idiolychnus

urolampus, Lampadena pontifex, Lampanyctodes hectoris, and possibly Diaphus

burtoni and Myctophum fissunovi (Krefft 1970; Nafpaktitis & Paxton 1968;

Kawaguchi & Shimizu 1978; Nafpaktitis 1978; Hulley 1981; Bekker 1983; Rubiés

1985). Two of these species, Diaphus garmani and Lampanyctodes hectoris, have

been taken in the southern Benguela region. The former species possesses a West

tropical Subpattern in the Atlantic (Nafpaktitis et al. 1977; Hulley 1981) and in

SOUTHERN BENGUELA LANTERNFISHES 241

the Indian Ocean has been recorded from the coast of East Africa, the Comoro

Islands, the west coast of Madagascar, off Mozambique, and to about 26°S

(Nafpaktitis 1978; Gjdsaeter & Beck 1981; Hulley 1984). The single specimen

taken during the present surveys, at 33°34’'S 17°32’E (bottom depth 404 m),

represents the first record of Diaphus garmani in the eastern South Atlantic. Its

presence here appears to be associated with the advection of Agulhas Water into

the region rather than an association with the bottom depth. An additional record

at 36°33’S 20°01’E serves to link the distribution to that in the Agulhas Current.

Diaphus taaningi is known from the Mauritanian Upwelling Region, Gulf of

Guinea, and south to about 24°S (O’Toole 1976; Hulley 1981; Rubiés 1985;

SAM data), while Lampadena pontifex has been caught at 24°26’S 13°30’E

(Karrer 1975), between 23°30'S 12°45’E and 25°30’'S 12°27'E (SAM data),

and ‘sporadically’ to about 28°30’S (Rubiés 1985). These records suggest that the

two species might be expected within the southern Benguela region as defined

here.

Oceanic species (Table 7)

In addition to the four bathypelagic and pseudoceanic species, 61 oceanic

mesopelagic species of Myctophidae were taken during the sampling in

the southern Benguela region. These are listed in Table 7, which is based

on the Atlantic Ocean distribution patterns and subpatterns given by Hulley

(1981).

Bekker (1983) recorded the following species from the region, but no

specimens of these were taken during the sampling cruises of the present survey:

Bolinichthys photothorax, Diaphus holti, D. termophilus, D. problematicus,

Electrona antarctica, Gonichthys venetus, Gymnoscopelus fraseri, G. braueri,

Lampadena- dea, L. chavesi, L. urophaos, Myctophum obtusirostre, Taaning-

ichthys miminus. However, recent investigations indicate that Diaphus holti,

Lampadena dea and L. chavesi are known from off the South West African—

Namibian coast (Rubiés 1985).

Species that have a Broadly Tropical distribution pattern in the Indian Ocean

(Diaphus parri, D. richardsoni, Diogenichthys panurgus, Lampanyctus turneri,

Myctophum spinosum, Triphoturus nigrescens) manifest themselves in the

Agulhas Subpattern of the Atlantic Ocean. These species, together with Diaphus

diadematus (Extended Agulhas Subpattern), are advected in Agulhas Water

pockets into the southern Benguela region and also, in the case of the latter

species, may be associated with aged upwelled Benguela Water. None of these

species was recorded by Rubiés (1985) off the South West African—Namibian

coast. In a similar manner, Atlantic holotropical species (Bolinichthys supra-

lateralis, Diaphus luetkeni, Lampanyctus nobilis)—which are more widely

distributed in the Indo-West Pacific—thermophilic-eurytropical species (Diaphus

brachycephalus, OD. lucidus,_ D. perspicillatus, D. splendidus, Myctophum

nitidulum), and possibly certain holoeurytropical species (Myctophum selenops),

penetrate the southern Benguela region with Agulhas Water.

242

ANNALS OF THE SOUTH AFRICAN MUSEUM

TABLE 7

Analysis of myctophid data from all types of gear. Distributional patterns and subpatterns after

Pattern

OCEANIC

BATHYPELAGIC

Widespread

Temperate

PSEUDOCEANIC

PELAGIC

Tropical

Benguela

OCEANIC

MESOPELAGIC

Widespread

Broadly Tropical

Tropical

Subpattern

South Temperate

Subantarctic

West Tropical

Holoeurytropical

Thermophilic-

eurytropical

Thermophobic-

eurytropical

Agulhas

Extended Agulhas

Holotropical

Hulley (1981).

Species

Taaningichthys bathyphilus

Lampanyctus achirus

Diaphus garmani

Lampanyctodes hectoris

Diogenichthys atlanticus

Electrona risso

Lobianchia dofleini

Notolychnus valdiviae

Ceratoscopelus warmingii

Diaphus mollis

Lampanyctus photonotus

Lobianchia gemellarii

Myctophum selenops

Notoscopelus resplendens

Diaphus brachycephalus

Diaphus lucidus

Diaphus perspicillatus

Diaphus splendidus

Myctophum nitidulum

Benthosema suborbitale

Lampanyctus alatus

Lampanyctus lineatus

Diaphus parri

Diaphus richardsoni

Diogenichthys panurgus

Lampanyctus turneri

Myctophum spinosum

Triphoturus nigrescens

Diaphus diadematus

Bolinichthys supralateralis

Diaphus luetkeni

Lampanyctus nobilis

Myctophum asperum

No.

specimens

> 10 842

WN WN

SFSA

No.

86.123

86.73

86.19

86.72

SOUTHERN BENGUELA LANTERNFISHES 243

OCEANIC

MESOPELAGIC (contd)

Subtropical Bisubtropical Bolinichthys indicus 20 86.4

Diaphus effulgens 14 86.16

Diaphus metopoclampus 46 86.27

Hygophum hygomii 179 86.61

Lampanyctus ater aS) 86.75

Lampanyctus festivus 333) 86.77

Lampanyctus pusillus 306 86.84

South Myctophum phengodes 13) 86.103

subtropical Scopelopsis multipunctatus 17 86.118

Symbolophorus barnardi 162 86.119

Temperate Bitemperate Lampadena speculigera 12 86.71

Lampanyctus intricarius 9 86.78

Lampanyctus macdonaldi 1 86.81

Loweina interrupta 1 86.94

South temperate Diaphus meadi 1 436 86.26

Convergence Diaphus ostenfeldi 15) 86.30

Gonichthys barnesi 335 86.49

Hygophum hanseni jj 86.60

Lampadena notialis aD 86.69

Lampanyctus australis 420 86.76

Lampanyctus lepidolychnus 278 86.79

Lampanyctus sp. B 9 86.89

Lampichthys procerus 287 86.90

Protomyctophum normani 4 86.116

Protomyctophum

subparallelum 11 86.110

Subantarctic Electrona carlsbergi 2 86.45

holosubantarctic Protomyctophum andriashevi 1 86.111

Protomyctophum parallelum 1 86.109

Subantarctic Diaphus hudsoni 472 86.20

semisubantarctic Gymnoscopelus piabilis 8 86.58

Metelectrona ventralis 45 86.96

Symbolophorus boops 316 86.120

Table 8 gives catch rates for each of the pattern types on the basis of four

cruises, during which RMT-2 (PAUG82), RMT-8 (PAUG83, PAUG84), and

MT-1600 (WH-71) nets were deployed. The percentage contribution for each of

the patterns and subpatterns, based on the mean catch rate (specimens per

1 000 m*) is given in Figure 2.

Catch rates for the Thermophilic-eurytropical, Agulhas, Extended Agulhas

and Holotropical subpatterns are low (Table 8), suggesting that all of the above

species show a ‘tailing’ distribution (i.e. a decline in density with increasing

latitude—see Hulley 1981, figs 39, 56, 67, 95) in the Agulhas Current, with a

correlated low abundance value in the eastern South Atlantic. It should be noted

that specimens of Myctophum asperum, M. nitidulum and M. spinosum were

taken at the surface during sampling with a David net. The seasonality of

occurrence of such species in this region will be examined in a later paper.

244 ANNALS OF THE SOUTH AFRICAN MUSEUM

TABLE 8

Catch rates, percentage contribution and stock estimate for oceanic mesopelagic species of

Myctophidae from SFRI Phyllosoma Surveys and from Walther Herwig stations in the southern

Benguela region. Patterns and subpatterns after Hulley (1981).

Catch rate

(specimens/1 000 m?)

PAUG82 PAUG83/84 WH-71

Pattern Subpattern RMT-2 RMT-8 MT-1600 %

Widespread 0,3188 0,0372 0,1192 0,1586 | 9,67

Broadly Tropical | Holoeurytropical 0,1889 0,2079 0,0599 0,1522 | 9,28

Thermophilic-eurytropical 0,0202 0,0124 0,0216 0,0181 | 1,10

Thermophobic-eurytropical 0.2986 0,0365 0,0035 0,1129 | 6,88

Agulhas 0,0101 0,0022 0,0041 0,0055 | 0,34

Extended Agulhas 0,0186 0,0080 0,0261 0,0176 | 1,07

Tropical Holotropical 0,0034 0,0022 0,0009 0,0022 | 0,13

Subtropical Bisubtropical 0,3876 0,0496 0,0968 0,1780 | 10,85

South subtropical 0,3475 0,0285 0,0285 0,1348 | 8,22

Temperate Bitemperate 0,0219 0,0015 0,0017 0,0084 | 0,51

South Temperate (Con-

vergence) 0,9144 0,4676 0,3548 0,5789 | 35,29

Subantarctic (Holosubantarctic) 0,0000 0,0000 0,0007 0,0002 | 0,01

Subantarctic (Semisubantarctic) 0,6866 0,1160 0,0163 0,2730 | 16,64

Gearcatchorate aera ener eee eee 32166 0,9696 0,7341

Mean'specimeniweight(g) hrereesaeeeeeee eee eee eee 0,2 0,5 1,0

Stockestimatel((onnes) Peete ree eee 10 x 10° 8 x 10° 12 x 10°

As pointed out above, the hydrography of the southern Benguela region

offshore of the continental shelf break suggests that in the main there should be a

mixed temperate—subtropical mesopelagic fauna. The results indicate that this is

true and that temperate species (52%) and subtropical species (19%) comprise

about 71 per cent of the total lanternfish fauna. The compositional value for the

broadly tropical element—i.e. those species distributed in both tropical and

subtropical waters—approximates that of the subtropical element, while wide-

spread species comprise about 10 per cent of the myctophid fauna.

All Atlantic widespread species are found in the southern Benguela region,

namely Diogenichthys atlanticus, Electrona risso, Lobianchia dofleini and

Notolychnus valdiviae. The Warm Water Group, which comprises about 37 per

cent of the fauna, includes Holoeurytropical, Thermophobic-eurytropical,

Bisubtropical and South subtropical Subpattern species. The percentage contri-

bution of each of these subpatterns to the myctophid fauna ranges between 7 and

11 per cent (Fig. 2). Of the 12 holoeurytropical species known from the Atlantic

Ocean, six species (Centrobranchus nigroocellatus, Diaphus subtilis, Gonichthys

cocco, Hygophum reinhardtii, Loweina rara, Notoscopelus caudispinosus) have

not been recorded in the southern Benguela region. Lampadena chavesi and

SOUTHERN BENGUELA LANTERNFISHES 245

WIDESPREAD

SEMISUBANTARCTIC

HOLOEURYTROPICAL

HOLOSUBANTARCTIC~

THERMOPHILIC- EURYTROPICAL

THERMOPHOBIC—EURYTROPICAL

-AGULHAS

EXTENDED AGULHAS

HOLOTROPICAL

CONVERGENCE

“BITEMPERATE

BROADLY TROPICAL

TEMPERATE

ij SUBTROPICAL

Fig. 2. Percentage composition by pattern and subpattern of high-oceanic species of

Myctophidae, based on catch rates of RMT-2, RMT-8 and MT-1600 nets from SFRI

Phyllosoma Surveys and from Walther Herwig stations in the southern Benguela region.

Taaningichthys minimus (Bisubtropical Subpattern: 9 Atlantic species) and

Diaphus anderseni (South subtropical Subpattern: 4 Atlantic species) have also

not been recorded. All Thermophobic-eurytropical Subpattern species known

from the Atlantic have been taken in the region. The somewhat low percentage

contribution of species of this subpattern is difficult to interpret at this stage; it

may be due either to a sampling artifact or related to the small number of species

involved (Lampanyctus lineatus is an uncommon species with a night distribution

at 150-350 m and 900-1 000 m—Nafpaktitis et al. 1977), or both.

Cool Water Group temperate species, which comprise about 52 per cent of

the myctophid fauna in the southern Benguela region, include bitemperate

(<1%), convergence (35%), semisubantarctic (17%), and holosubantarctic

(<1%) species (Table 8). This confirms the strong intrusion of southern elements

into the southern Benguela region, particularly those species associated with the

Subtropical Convergence and with the region between the Convergence and the

Subantarctic Divergence, and agrees with the conceptual image of the region

246 ANNALS OF THE SOUTH AFRICAN MUSEUM

derived from physical studies. Whether this compositional structure changes with

decreasing latitude is difficult to assess, since comparative data are not given by

Rubiés (1985) for the South West African—Namibian myctophids. However, he

has pointed out (p. 581) that most are subtropical species and that ‘subantarctic’

species, i.e. ‘coming from the Convergence area’, were caught in very small

numbers.

Stock estimate

Mean catch rates for each subpattern (specimens/1 000 m°?) for the PAUG82

(RMT-2), PAUG83 and PAUG84 (RMT-8) and WH-71 (MT-1600) cruises

have been calculated from abundance values (specimens/hour). These are

presented in Table 8. Estimates of the stock of lanternfishes in the offshore area

of the eastern South Atlantic are calculated as 10 x 10° tonnes in the case of

sampling with an RMT-2; 8 x 10° tonnes with an RMT-—8; and 12 Xx 10° tonnes

with an MT-1600 net. The mean value (10 X 10° tonnes) represents about 62 per

cent of the abundance estimate for mesopelagic fishes of the South-east Atlantic

(Gjdésaeter & Kawaguchi 1980) and compares with the lanternfish composition by

number (56%) for Transect II of the 1971 cruise of FRV Walther Herwig (Hulley

1981). However, this transect also encompassed subtropical and tropical regions.

Since Hulley & Krefft (1985) have demonstrated that smaller populations of

myctophids occur in warm-water species than in cold-water species, the mean

stock estimate of 10 x 10° tonnes appears to be realistic.

ACKNOWLEDGEMENTS

I would like to express my thanks to S. X. Kannemeyer, M. Bougaardt and

C. Goliath of the South African Museum for assistance with data capture, and to

those captains, crews and scientists of the Sea Fisheries Research Institute, Cape

Town, who helped to collect the specimens. In particular, my thanks are due to

Drs A. Payne and D. Pollock, and Mr I. Hampton of the Sea Fisheries Research

Institute for making their material available for study and for providing the

necessary adjunct information. This study was made possible through grants from

the CSIR Foundation for Research Development and the CSIR-SANCOR

Benguela Ecology Programme, as well as support from the South African

Museum.

REFERENCES

ABRAMS, R. W. 1985. Energy and food requirements of pelagic aerial seabirds in different

regions of the African sector of the Southern Ocean. Jn: SIEGFRIED, W. R., Conpy, P. R.

& Laws, R. M. eds. Antarctic nutrient cycles and food webs: 466-472. Berlin: Springer.

ABRAMS, R. W., Cooper, J. & Enticorr, J. W. 1984. Species diversity and trophic

relationships of the marine avifauna in prescribed sectors of the Southern Ocean. Final

Report to SASCAR. Cape Town: Percy Fitzpatrick Institute of African Ornithology: 1-35

(mimeograph).

SOUTHERN BENGUELA LANTERNFISHES 247

AuHLstTroM, E. H., Moser, H. G. & O’TooLe, M. J. 1976. Development and distribution of

larvae and early juveniles of the commercial lanternfish Lampanyctodes hectoris (Gunther)

off the west coast of southern Africa, with a discussion of phylogenetic relationships of the

genus. Bulletin of the Southern California Academy of Sciences 75 (2): 138-152.

ALLANSON, B. R., Hart, R. C. & LUTJEHARMS, J. R. E. 1981. Observations on the nutrients,

chlorophyll and primary production of the Southern Ocean south of Africa. South African

Journal of Antarctic Research 10/11: 3-14.

BEKKER, V. E. 1983. Myctophids of the world ocean. Moscow: Academy of Sciences of SSSR

(in Russian).

BopeEN, B. P. 1951. The euphaustiid crustaceans of southern African waters. Transactions of

the Royal Society of South Africa 34 (1): 181-243.

DaRBYSHIRE, M. 1966. The surface water near the coast of southern Africa. Deep-Sea

Research 13 (1): 57-81.

De Decker, A. H. B. 1984. Near-surface copepod distribution in the south-western Indian and

south-eastern Atlantic Ocean. Annals of the South African Museum 93 (5): 303-370.

De ViILuierS, G. (ed.) 1982. A report on alternative resources available in the Benguela

System. Internal Report of the Benguela Ecology Programme 3: 1-27 (cited by permission).

D.H.I. 1971. Monatskarten ftir den stidatlantischen Ozean. Dritte Auflage. Hamburg:

Deutsches Hydrographisches Institut.

DusrovskayYA, T. A. & Maxorov, O. E. 1969. Chemical characteristics of some Antarctic

fishes. Trudy Vsesoyuznogo nauchno-issledovatel’-skogo instituta morskogo rybnogo

khozyaistva i okeanografii 66: 311-317 (in Russian).

GIJOSAETER, J. & Beck, I.-M. 1981. Mesopelagic fish off Mozambique. Fiskdirektoratets

Skrifter. Serie Havundersokelser 17: 253-265.

GJOSAETER, J. & KAwacucul, K. 1980. A review of the world resources of mesopelagic fish.

FAO Fisheries Technical Paper 193: 1-151.

GRINDLEY, J. R. & PENRITH, M. J. 1965. Notes on the bathypelagic fauna of the seas around

South Africa. Zoologica africana 1 (2): 275-295.

Henry, A. E. 1975. Hydrology and nutrient salts of the south-east Atlantic and south-west

Indian oceans in 1968. Investigational Report. Division of Sea Fisheries, Republic of South

Africa 95: 1-66.

Heyporn, A. E. F. 1959. The Chaetognatha off the west coast of the Union of South Africa.

Investigational Report. Division of Fisheries, Union of South Africa 36: 1-56.

Hu.tey, P. A. 1972a. A report on the mesopelagic fishes collected during the deep-sea cruises

of R.S. ‘Africana II’, 1961-1966. Annals of the South African Museum 60 (6): 197-236.

Hu .ey, P. A. 1972b. Mesopelagic fishes from Vema Seamount (IK Station 52). Annals of the

South African Museum 60 (7): 237-244.

Huey, P. A. 1981. Results of the research cruises of FRV ‘Walther Herwig’ to South

America. LVIII. Family Myctophidae (Osteichthyes, Myctophiformes). Archiv fiir

Fischereiwissenschaft 31 (1): 1-300.

Hu.iey, P. A. 1984. The South African Museum’s Meiring Naude cruises. Part 14. Family

Myctophidae (Osteichthyes, Myctophiformes). Annals of the South African Museum 93 (2):

53=96.

Huey, P. A. 1986. A taxonomic review of the lanternfish genus 7riphoturus Fraser-Brunner,

1949 (Myctophidae, Osteichthyes). Annals of the South African Museum 97 (4): 71-95.

Hu.tey, P. A. In press. Family No. 86: Myctophidae. Jn: Smita, M. M. & HEeEmstra, P. C.

eds. Smith’s sea fishes.

Hutiey, P. A. & Krerrr, G. 1975. A zoogeographic analysis of the fishes of the family

Myctophidae (Osteichthyes, Myctophiformes) from the 1979—Sargasso Sea Expedition of

R.V. Anton Dohrn. Annals of the South African Museum 96 (2): 19-53.

Jones, P. G. W. 1971. The southern Benguela Current region in February 1966. Part I.

Chemical observations with particular reference to upwelling. Deep-Sea Research 18 (2):

193-208.

Karrer, C. 1975. Uber Fische aus dem Siidostatlantik (Teil 2). Mitteilungen aus dem

Zoologischen Museum in Berlin 51 (1): 63-82.

Kawacucul, K. & SHimizu, H. 1978. Taxonomy and distribution of the lanternfishes, genus

Diaphus (Pisces, Myctophidae) in the western Pacific, eastern Indian Oceans and the

Southeast Asian Seas. Bulletin of the Ocean Research Institute, University of Tokyo 10:

1-145.

248 ANNALS OF THE SOUTH AFRICAN MUSEUM

KrerFt, G. 1970. Zur Systematik und Verbreitung der Gattung Lampadena Goode & Bean,

1896 (Osteichthyes, Myctophoidei, Myctophidae) im Atlantischen Ozean, mit Beschreibung

einer neuen Art. Berichte der Deutschen Wissenschaftlichen Kommission fiir Meeresfors-

chung 21 (1-4): 271-289.

Krerrr, G. 1974. Investigations on midwater fish in the Atlantic Ocean. Berichte der

Deutschen Wissenschaftlichen Kommission fiir Meeresforschung 23 (3): 226-254.

Krerrr, G. 1978. Distribution patterns of oceanic fishes in the Atlantic Ocean. Selected

problems. Revue des travaux. Institut des Péches maritimes 40 (3-4): 439-460.

Lenz, W. 1975. Untersuchungen zur inneren hydrographischen Struktur des siidlichen und

mittleren Atlantiks (0Q—2 000 m Tiefe) mit zoogeographischen Anmerkungen. Berichte der

Deutschen Wissenschaftlichen Kommission fiir Meeresforschung 24 (1): 1-22.

LUTJEHARMS, J. R. E. 198la. Features of the southern Agulhas Current circulation from

satellite remote sensing. South African Journal of Science 177: 231-236.

LUTJEHARMS, J. R. E. 19816. Satellite studies of the South Atlantic Upwelling System. In:

Gower, J. F. R. ed. Oceanography from space. Marine Science 13: 195-199.

LUTJEHARMS, J. R. E. & Emery, W. J. 1983. The detailed thermal structure of the upper lave: -

between Cape Town and Antarctica during the period Jan.—Feb. 1978. South African

Journal of Antarctic Research 13: 3-14.

LUTJEHARMS, J. R. E. & VALENTINE, H. R. 1981. Ocean circulation studies in the vicinity of

southern Africa: preliminary results using FGGE drifters and remote sensing. Advances in

Space Research 1: 211-223.

LUTJEHARMS, J. R. E., WALTERS, N. M. & ALLANSON, B. R. 1985. Oceanic frontal systems

and biological enhancement. Jn: SIEGFRIED, W. R., Conny, P. R. & Laws, R. M. eds.

Antarctic nutrient cycles and food webs: 11-21. Berl’n: Springer.

McGinnis, R. F. 1982. Biogeography of lanternfishes (Myctophicae) south of 30°S. Antarctic

Research Series. National Academy of Sciences, Washington 35: 1-110.

NaFPAKTITIS, B. G. E. 1978. Systematics and distribution of lanternfishes of the genera

Lobianchia and Diaphus (Myctophidae) in the Indian Ocean. Science Bulletin. Natural

History Museum of Los Angeles County 30: 1-92.

NAFPAKTITIS, B. G., BAcKUS, R. H., CrAppock, J. E., HAEDRICH, R. L., Ropison, B. H. &

KARNELLA, C. 1977. Family Myctophidae. Memoirs of the Sears Foundation for Marine

Research 1 (7): 13-265.

NAFPAKTITIS, B. G. & PaxTon, J. R. 1968. Review of the lanternfish genus Lampadena with a

description of a new species. Contributions in Science. Los Angeles County Museum 138:

1-29.

O’TooLE, M. J. 1976. Incidental collections of small and juvenile fishes from egg and larval

surveys off South West Africa. Fisheries Bulletin. Division of Sea Fisheries, Republic of

South Africa 8: 23-33.

ParIN, N. V., BEKKER, V. E., BORODULINA, O. D. & TcHuvasov, V. M. 1973. Deep-sea

pelagic fishes of the South-eastern Pacific and adjacent waters. Trudy Instituta okeanologii,

Akademiya nauk SSSR 94: 71-172 (in Russian).

Rusigs, P. 1985. Zoogeography of the lanternfishes (Osteichthyes, Myctophidae) of Southwest

Africa. Simposio internacional sobre las areas de afloramiento mas importantes del Oeste

Africano (Cabo Blanco y Benguela): 573-586. Barcelona: Instituto de Investigaciones

Pesqueras.

SHANNON, L. V. 1966. Hydrology of the south and west coasts of South Africa. Investigational

Report. Division of Sea Fisheries, Republic of South Africa 58: 1-22.

SHANNON, L. V. 1985. The Benguela ecosystem. Part I. Evolution of the Benguela, physical

features and processes. Oceanography and Marine Biology. An Annual Review 23:

105-182.

SHANNON, L. V., STANDER, G. H. & CAMPBELL, J. A. 1973. Oceanic circulation deduced from

plastic drift cards. Investigational Report. Sea Fisheries Branch, Republic of South Africa

108: 1-31.

SHANNON, L. V. & VAN Ruswisck, M. 1969. Physical oceanography of the Walvis Ridge

region. Investigational Report. Division of Sea Fisheries, Republic of South Africa 70: 1-19.

VissER, G. A. 1969. Hydrological observations in the South-east Atlantic Ocean. 1. The

Schmitt-Ott Seamount area. Investigational Report. Division of Sea Fisheries, Republic of

South Africa 77: 1-23.

SOUTHERN BENGUELA LANTERNFISHES 249

WEIKERT, H. 1975. Distribution and occurrence of pontellids (Copepoda, Calanoida) in the

Central and South Atlantic Ocean. Berichte der Deutschen Wissenschaftlichen Kommission

fiir Meeresforschung 24 (2-3): 134-150.

WELSH, F. G. & VissER, G. A. 1970. Hydrological observations in the South-east Atlantic

Ocean. 2. The Cape Basin. Investigational Report. Division of Sea Fisheries, Republic of

South Africa 83: 1-24.

WormutH, J. H. & Roper, C. F. E. 1983. Quantitative sampling of oceanic cephalopods by

nets: problems and recommendations. Biological Oceanography 2 (2-3-4): 357-377.

——_—— — — gE gay en ae ee Te eT ee ete bs re eri hee ee Metin gn eel fre Ee me tne at eer ee et a —:

—— SGRTRGUNIS SPI ee as

7 SSIES

)

;

)

z * y

| 1

; \

7 :

ite “1 j

& , :

he :

6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu-

larly Articles 22 and 51).

Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed

by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov.,

etc:

An author’s name when cited must follow the name of the taxon without intervening punctuation

and not be abbreviated; if the year is added, a comma must separate author’s name and year. The

author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is trans-

ferred from its original genus. The name of a subsequent user of a scientific name must be separated

from the scientific name by a colon.

Synonymy arrangement should be according to chronology of names, i.e. all published scientific

names by which the species previously has been designated are listed in chronological order, with all

references to that name following in chronological order, e.g.:

Family Nuculanidae

Nuculana (Lembulus) bicuspidata (Gould, 1845)

Figs 14-15A

Nucula (Leda) bicuspidata Gould, 1845: 37.

Leda plicifera A. Adams, 1856: 50.

Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b).

Nucula largillierti Philippi, 1861: 87.

Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9.

Note punctuation in the above example:

comma separates author’s name and year

semicolon separates more than one reference by the same author

full stop separates references by different authors

figures of plates are enclosed in parentheses to distinguish them from text-figures

dash, not comma, separates consecutive numbers.

Synonymy arrangement according to chronology of bibliographic references, whereby the year is

placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable.

In describing new species, one specimen must be designated as the holotype; other specimens

mentioned in the original description are to be designated paratypes; additional material not regarded

as paratypes should be listed separately. The complete data (registration number, depository, descrip-

tion of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.:

Holotype

SAM-A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach, Port Eliza-

beth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973.

Note standard form of writing South African Museum registration numbers and date.

7. SPECIAL HOUSE RULES

Capital initial letters

(a) The Figures, Maps and Tables of the paper when referred to in the text

e.g. ‘. . . the Figure depicting C. namacolus .. .’: ‘. . . in C. namacolus (Fig. 10). . .’

(b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by

initials or full names

e.g. Du Toit but A.L. du Toit; Von Huene but F. von Huene

e.g. Therocephalia, but therocephalian

Punctuation should be loose, omitting all not strictly necessary

Reference to the author should preferably be expressed in the third person

Roman numerals should be converted to arabic, except when forming part of the title of a book or

article, such as

‘Revision of the Crustacea. Part VIII. The Amphipoda.’

Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial

capital letter, provided the same generic name is used consecutively. The generic name should

not be abbreviated at the beginning of a sentence or paragraph.

Name of new genus or species is not to be included in the title; it should be included in the abstract,

counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts.

P. ALEXANDER HULLEY

| LANTERNFISHES OF

THE SOUTHERN BENGUELA REGION

PART 1

FAUNAL COMPLEXITY AND DISTRIBUTION

S67X% 97 PART 8 AUGUST 1986 ISSN 0303-2515

be 542 ¢

OF THE SOUTH AFRICAN >

“NAIM

CAPE ‘TOWN

INSTRUCTIONS TO AUTHORS

1. MATERIAL should be original and not published elsewhere, in whole or in part.

2. LAYOUT should be as follows:

(a) Centred masthead to consist of

Title: informative but concise, without abbreviations and not including the names of new genera or species

Author’s(s’) name(s)

Address(es) of author(s) (institution where work was carried out)

Number of illustrations (figures, enumerated maps and tables, in this order)

(b) Abstract of not more than 200 words, intelligible to the reader without reference to the text

(d) Introduction

(e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents

(f) Summary, if paper is lengthy

(g) Acknowledgements

(h) References

(i) Abbreviations, where these are numerous.

3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with

3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for

illustrations should be typed separately, their positions indicated in the text. All pages should be num-

bered consecutively.

Major headings of the paper are centred capitals; first subheadings are shouldered small capitals;

second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further

subdivisions should be avoided, as also enumeration (never roman numerals) of headings and

abbreviations.

Footnotes should be avoided unless they are short and essential.

Only generic and specific names should be underlined to indicate italics; all other marking up

should be left to editor and publisher.

4. ILLUSTRATIONS should be reducible to a size not exceeding 12 x 18cm (19 cm including

caption); the reduction or enlargement required should be indicated (and preferably uniform); orig-

inals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and

final size. A metric scale should appear with all illustrations, otherwise magnification or reduction

should be given in the caption; if the latter, then the final reduction or enlargement should be taken

into consideration.

All illustrations, whether line drawings or photographs, should be termed figures (plates are not

printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items

of composite figures should be designated by capital letters; lettering of figures is not set in type and

should be in lower-case letters. If Letraset is used authors are requested to use Helvetica-style letter-

ing, if possible.

The number of the figure should be lightly marked in pencil on the back of each illustration.

5. REFERENCES cited in text and synonymies should all be included in the list at the end of the

paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable):

(a) Author’s name and year of publication given in text, e.g.:

‘Smith (1969) describes...”

‘Smith (1969: 36, fig. 16) describes...’

“As described (Smith 1969a, 19696; Jones 1971)’

“As described (Haughton & Broom 1927)...’

‘As described (Haughton er al. 1927)... .’

Note: no comma separating name and year

pagination indicated by colon, not p.

names of joint authors connected by ampersand

et al. in text for more than two joint authors, but names of all authors given in list of references.

(b) Full references at the end of the paper, arranged alphabetically by names, chronologically within

each name, with suffixes a, b, etc., to the year for more than one paper by the same author in

that year, e.g. Smith (1969a, 1969b) and not Smith (1969, 1969a).

For books give title in italics, edition, volume number, place of publication, publisher.

For journal article give title of article, title of journal in italics (according to the World list of scientific periodicals. 4th ed.

London: Butterworths, 1963), series in parentheses, volume number, part number in parentheses, pagination (first and

last pages of article).

Examples (note capitalization and punctuation)

BuLLouGu, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan.

FiscHER, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140.

FiscHER, P. H., DuvAL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie

expérimentale et générale 74 (33): 627-634.

Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and

Magazine of Natural History (13) 2 (17): 309-320.

Koun, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of

the Bingham Oceanographic Collection, Yale University 17 (4): 1-51.

THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthro-

pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Stid-Afrika ausgeftihrt in den Jahren

1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270.

(continued inside back cover)

ANNALS OF THE SOUTH AFRICAN MUSEUM

ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

Volume 97 Band

August 1986 Augustus

Part 8 Deel

NOTES ON THE NOMENCLATURE

OF THE KHOISAN

M. L. WILSON

Cape Town Kaapstad

The ANNALS OF THE SOUTH AFRICAN MUSEUM

are issued in parts at irregular intervals as material

becomes available

Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000

Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

word uitgegee in dele op ongereelde tye na gelang van die

beskikbaarheid van stof

Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000

OUT OF PRINT/UIT DRUK

de 203, 528), 3122), 42508) prinys 50-3) Sa-9))

6G tod), TOD), B, CAD, 7), 18),

ICED. ite), IED), ICES), WO), 27), SES), 2G), 28, 25O), 25)

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 085 0

Printed in South Africa by In Suid-Afrika gedruk deur

The Rustica Press, Pty., Ltd., Die Rustica-pers, Edms., Bpk.,

Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap

NOTES ON THE NOMENCLATURE OF THE KHOISAN

M. L. WILSON

Department of Archaeology, South African Museum, Cape Town

[MS accepted 10 April 1986]

ABSTRACT

The origin, meaning and use of the various names and terms applied to the indigenous non-

Negro peoples of southern Africa are discussed. It is concluded that, while the various other

branches of anthropology should use a common set of names or terms, archaeologists should, in

general, restrict themselves to the use of terms such as ‘herder’ and ‘hunter-gatherer’.

CONTENTS

PAGE

MMtKOGUCHOME se aioe as fins fh wonatdc a Sibtie sae buns Bale Dwi

URI OTIS VOT Senet IS as. Pate ee ot OE Alcs NA ee Cn 252

SVEIDN cosis aonerghs Reed She bp ah ah Sires ne gen ae ao 254

FAOLCENTO INET a eee ie. he Ca ie ae Ral tele wars ie eS 256

J BUI MATE 6S Bete oS Bhc eee ie Re wie at 5 aN one ee oe Aes ve a 257

NGO IS Ane Pee Ey Wr Par ae sei oe aie av Gie aon we 259

nlerderstandshuntersisae ee ec he os lee eee Soe carewee « 260

Discusstonvandiconclusions. 6.5... ee ee cc ces 261

PXCKMOWIECUSEMENMES cera biel. zh eas aie e cin berets ee the ele wl eles 264

IRGIEneM COSI Terre errr t cosa one oA ek eit t < lly tdi wie ecatiohs 264

INTRODUCTION

What's in a name? that which we call a rose

By any other name would smell as sweet.

Shakespeare: Romeo and Juliet

During the 1971 conference on “The Peoples of Southern Africa’, sponsored

by the Royal Society of South Africa and the South African Institute for Medical

Research, a session was devoted to nomenclature. The majority of the

participants agreed to the adoption of a tripartite terminology to cover the three

main areas of research. Excluding the terminology for the Negroes, which is not

relevant here, the consensus was that ‘San’ and ‘Khoikhov should be applied to

biological entities; ‘Bushman’ and ‘Hottentot’ to languages; and ‘hunters’ or

‘hunter-gatherers’ and ‘herders’ or ‘pastoralists’ to economy or way of life

(Jenkins & Tobias 1977: 51). These names or terms provide suitable headings for

discussion although it is not always possible to keep them entirely separate. The

term ‘Khoisan’ is included since it is also in current use.

Ann. S. Afr. Mus. 97 (8), 1986: 251-266.

DIS) ANNALS OF THE SOUTH AFRICAN MUSEUM

KHOIKHOI

The first recorded use of a cognate of this name is in the entry of 9 January

1653 in the journal of Jan van Riebeeck, where it is rendered as Quena (Thom

1952: 127). The name was, however, not commonly used and did not gain

currency among the early settlers and travellers.

Nienaber (1963: 310-311, 378-380, 384-385) lists, among others, the

following early usages and their attributed meanings: Witsen, 1691, quena, the

Hottentot nation, k’quique, a man; Valentyn, 1705, t-hoekoe and thdekoe, a

Hottentot, guoique, a man; Kolb(en), 1708, q~-ena, the Hottentot nation,

q~uoique, a man; Thunberg, 1773, Keuna, people; Le Vaillant, 1780-83, khoé-

khoep, Hottentot man; Von Winkelmann, 1788, xkeukoe, (the) people; Barrow,

1797, Quaiquae, the name ‘by which the whole nation was distinguished, and

which at this moment they bear among themselves in every part of the country’,

quaina, man. (The dates given by Nienaber refer to the observation, not to the

publication.) To these may be added the observation by Robert Jacob Gordon,

who died in 1795 and was probably the most widely travelled man up to that time,

that Queuna was the plural of Quoi, person (Forbes in Sparrman 1975: 181 n23).

More recently, Stow (1905: 31) commented that: “These people, who were

severally known to the old colonists and early writers as the Bosjesmans, the

Boschimans, and Bushmen, appear to have adopted among themselves the name

of ’Khuai, which is also the same as that given to the natural apron for which the

women of pure Bushman and Hottentot races are distinguished; and it was thus

probable that the appellation ’Quae-’quae, or perhaps more properly ’Khuai-

’Khuai or ’Khuai-’quae—the people of the Apron, was derived.’ Stow’s comment

is of interest in that he applied the name to Bushmen, whereas it is generally

taken to have been the name of the Hottentots. Schapera (1930: 428) gave the

Naron (= Nharo, one of the ‘Hottentot’ languages) kwe//kwa, men, and kwene,

people, and the Nama khoii, person, and khoin, people. Kroenlein’s Nama

dictionary gives ‘Khoi-khoin’ as meaning die Naman, the Nama (Rust 1969: 238).

According to Nienaber (1963) these names represent dialectal variants, those

given by Witsen, Kolb and Valentyn being in the ‘Cape’ dialect, the others in his

list being ‘Eastern’. Schapera’s and Kroenlein’s should be ‘Northern’, but

according to Nienaber’s classification Gordon’s Queuna (Forbes in Sparrman

1975) could be either ‘Cape’ (cf. Witsen and Kolb above) or ‘Eastern’ (cf.

Thunberg and Barrow above), while Stow’s (1905) seem closer to the latter.

Nienaber (1963: 311) observed that when the language became more

normalized the name was written Khoi-khdi-n in Nama and, according to

Meinhof, khoe-khoe-n-a in Korana, both with the meaning ‘Hottentots’. This is

based on the (masculine) singular khoi-b, man or person, which in reduplicative

apposition became the people’s name, meaning ‘the people of people’, or ‘the

real (or “‘true’”’) people’. Nienaber does not cite a source for this translation, but

Nienaber & Raper (1977: 31) cite the missionary Knudsen as having recorded in

1842 that ‘Khoi-khoin’ meant ‘Mensch-Menschen oder Mensch der Menschen’.

The literal translation of these, ‘person-people’ and ‘person of the people’ (or

NOMENCLATURE OF THE KHOISAN 253

even ‘of humankind’ ), is awkward. The second of Nienaber’s translations is more

suitable—better yet that of Hahn (1881: 2), ‘men of men, i.e. men par

excellence’. It seems legitimate to wonder whether these translations are

semantically correct, particularly in view of the wide range of early meanings

given above, and perhaps more specifically with regard to the meaning attributed

to the Nharo kwe//kwa. If the translations of Knudsen and others (Hahn was the

son of a missionary) were no more than pious attempts to uplift the people from

the degradation into which they had fallen (or been thrust), after more than a

century of these meanings being applied it is probable that today no Nama would

know differently. W. H. G. Haacke, a linguist researching the Nama language

and one of the contributors to a new Nama dictionary currently being compiled,

considers Hahn’s translation to be dubious, but says that further research is

needed (1985 pers. comm.).

In the current orthography for the Nama—Damara language accepted by the

South West African Department of Bantu Education (1977: 29) it is laid down

that: ‘All words which in the past were spelt with the letters -oi- are spelt with

-oe- in the standardised orthography. . . .’ Haacke (1982: 78) commented that

‘Khoi-khov’ is antiquated and used instead ‘Khoékhoe’. Haacke (1985 in litt. and

pers. comm.) has pointed out that Nama has no sound corresponding to the -oi-

diphthong. The two vowels were originally separated by a now-elided consonant

(probably -w-), and are pronounced separately and are different tonally.

Approximate phonetic equivalents are as in ‘port’ and ‘fare’ (International

Phonetic Alphabet symbols 9 and ea).

What is important is whether, however the name is spelt, this Nama dialectal

variant can properly be applied to all the people, past and present, to whom it

refers. The now-extinct speakers of the “Cape’ dialect seem to have called

themselves Khoina or something similar and it does not seem proper to call these

people by a name they would not have called themselves. However, if the oral

traditions of the Korana (Engelbrecht 1936: 2 ff.) are correct in asserting that

they, or some of their number, earlier lived closer to the Cape (i.e. modern Cape

Town), then it is arguable that ‘Khoekhoe’ (omitting the tonal marks, as will

inevitably happen if this word is accepted) may be a modern variant of an earlier

‘Cape’ dialectal form. Whether or not ‘Khoina’ and ‘Khoekhoe’ had or have the

Same meaning is a matter for linguists to determine. It seems unlikely, however,

that the Nama would have two words, Khoen and Khoekhoen, with precisely the

same meaning, ‘people’.

Two historians, B6eseken and Elphick, conducted a scholarly debate for

some years on the meaning, origin and use of ‘Khoikhoi’, ‘San’ and ‘Khoisan’

(Béeseken 1972, 1974, 1975; Elphick 1974, 1975). The arguments on both sides

are persuasive but in the end one is obliged to agree with Boeseken that there is

no documentary evidence from the historical period for the common use of

‘Khoikhov (or its cognates) as a generic by the people themselves.

G. Klinghardt (1985 pers. comm.) has pointed out that, in considering the

various names, it is important to bear in mind the different levels of identity on

254 ANNALS OF THE SOUTH AFRICAN MUSEUM

which a person can operate. These can range from: individual; member of a

smaller group (e.g. clan); member of a larger group (e.g. tribe); to member of the

largest group (e.g. nation, humankind, or all those who are considered to be ‘the

same’ ). He has also drawn attention to the fact that in the historical records there

is little indication that the Khoikhoi had a sense of nationhood, the tribe usually

being the largest cohesive unit (see also Elphick 1985: 68). The possibility thus

exists that, while ‘Khoikhoi’ may have been understood as a generic (‘us’, as

distinct from ‘them’ ), it may simply have been a general term for ‘people’.

Orthographically, the apocopic use of “Khoikhov’ or ‘Khoekhoe’ without the

gender:number suffix, and perhaps also the omission of the tonal marks, should

not be condoned; and the use of the abbreviated forms, ‘Khor and ‘Khoe’, is even

less excusable. Nienaber (1985 in litt.) has pointed out that ‘Khoe’ simply means

‘person’. Haacke (1985 in litt.) has, however, suggested that because of the

possibility of confusion or error—there are different suffixes for masculine,

feminine, singular, plural, dual and common plural—it is perhaps preferable to

drop the suffix unless actually writing the language.

A point that is less frivolous than it may seem is that, while the replacement

of the long-ingrained ‘Khoikhor by ‘Khoekhoe’ may cause little concern to

serious anthropologists, the introduction of this word into the ‘popular’ literature

without any indication of its correct pronunciation, and which so resembles the

Afrikaans ‘koekoek’ and its English counterpart ‘cuckoo’, is liable to bring it into

the same sort of disrepute as that into which ‘Hottentot’ is said to have fallen (e.g.

Wilson 1969: 40). It was not serious researchers who degraded ‘Hottentot’ but the

ignorant and uninformed or, following Elphick (1985: xv), ‘the intellectual

climate of eighteenth-century Europe’.

The questions to be answered in this regard are, then, whether ‘Khoikhov’ or

‘Khoekhoe’ were ever commonly used by the people as a (generic) name for

themselves; whether what appears to be a dialectal form is correctly applicable to

all the people, some of whom may have used another form; and whether

‘Khoekhoe’ is not liable to bring the name into derision. The actual, or original,

meaning of the name may be of lesser importance, especially as there is no

evidence of any derogatory connotation.

SAN

The first recorded use of a cognate of the name ‘San’ is the same as that for

Quena given above, where it is written Soaqua and applied to the elusive Visman

(‘Fishermen’ or perhaps just ‘Fishmen’) by the ‘Saldanhars’ (at that time

‘Hottentots’ sensu lato) (Thom 1952: 127). Unlike Quena, however, the name

occurs fairly frequently in various forms in the records of the second half of the

seventeenth century (see, for example, references under ‘Soaqua’ in the index of

Moodie 1960a-c).

Hahn (1881: 3), while admitting that the meaning of the Nama word San was

not quite intelligible, traced its origin to the root sd, which he interpreted as

NOMENCLATURE OF THE KHOISAN PISS)

meaning ‘aborigines’ or ‘settlers proper’. However, in the vocabulary of

Hottentot and Bushman words he provided, Hahn (1881: 7) gave the Hottentot

sab and Bushman ts@ as meaning ‘buchu’. Smith (1966: 165-166), a botanist,

suggested that ‘Sab’, plural ‘San’, meant ‘bushes’ and that the name was applied

because the people so called used the aromatic leaves of the buchu plant (see

Smith 1966: 135-141 under “boegoe’ for discussion of the wide range of species to

which the name has been applied). Westphal (1971: 369) translated the root sa as

meaning ‘gather food’. However, Nienaber (1985 in Jitt.) is of the opinion that the

various translations are ‘oogfilologies’—based on the visual similarities of

words— and fortuitous. It thus seems likely that the ascription by Elphick (1977:

27-28, 1985: 27-28) of a ‘San’ suffix to the names of various tribes, such as

Horisans, Cochesons and Kamesons, may be equally fortuitous. (The early

records are not clear as to the identities of these peoples. For example, in the only

two references to the Kamesons, the journal of Van der Stel’s expedition to

Namaqualand (Moodie 1960a: 409-410) records that ‘we found among the

Amaquas several of those Hottentots who are called Kamesons’. Later, members

of the party happened upon a group of five men, who were called ‘Sonquas’, one

of whom was captured and taken back to Van der Stel and ‘the Commander

sounded the Sonqua, who now said he was no Sonqua, but one of the tribe called

Kamesons’. From this, it seems not improbable that the early settlers and

travellers termed “‘Sonquas’ any people they found without domestic stock;

though in this case the party that captured the so-called ‘Sonqua’ included four of

the expedition’s Hottentots, who ought to have recognized him as one of their

own kind, if, in fact, he was. G. Klinghardt (1986 pers. comm.) is of the opinion

that the man was a !Gami-#nun (Bondelzwart, one of the major divisions of the

Nama) and that the ‘Sonqua’ with whom he was living were clients of the

‘Amaquas’ (= Nama) in that area.)

Haacke (1985 pers. comm.) has mentioned that there are tonal differences in

the pronunciation of the two words and that it is thus doubtful that San,

‘Bushmen’, derives from sd, ‘gather’. He has also pointed out that if the

gender:number suffix is to be dropped from ‘Khoekhoe’, for the sake of

conformity this should also be done with ‘San’: thus, Sa, or in the new Nama

orthography, Saa.

In the early records the use of this name or its variants is often coupled with

some pejorative comment, though it is sometimes also used neutrally, as an

identifier. Examples of the former are ‘Soanqua who are robbers, and are

enemies to them and all the Saldanhars’ (Thom 1954: 49); ‘Souquas or bandits’

(Thom 1958: 24); ‘It is from the Soncquas that we have to expect the greatest

mischief towards the burgers. A people of this sort called Obicquas . . . took the

cattle of a party of burgers. . . and killed one of the party. . . . These are a kind

of men who live in the mountains, who maintain themselves entirely by robbery,

and have nothing to lose by war; they do not hesitate, upon a favourable

opportunity, to steal the cattle of their own nation, under whose dominion they

live, and therefore the [Hottentot] Captains are always looking out for

256 ANNALS OF THE SOUTH AFRICAN MUSEUM

opportunities for revenge’ (Moodie 1960a: 392). Examples of neutral usage are

‘the Souquas (the hill people, who subsist chiefly by the game they kill)’ and ‘we

found no Souquas here’ (Moodie 1960a: 225, 231).

Hahn (1881: 3) commented: “The word sa(b) has also acquired a low

meaning and is not considered very complimentary. The Khoikhoi speak of /Uri-

San (white Bushmen) and mean the low white vagabonds and runaway sailors

who visit their country as traders. One also often hears ‘““Khoikhoi tamab, Sab

ké’’, he is no Khoikhoi, he is a Sa, which means to say, “‘he is no gentleman, he is

of low extraction, or he is a rascal’ (Hahn’s emphasis). Elsewhere Hahn (1881:

101) noted that the Nama called others of their kind ‘Bushmen’, ‘especially

when they are servants, or if they lead a Bushman’s life, and have no cattle and

sheep’.

Lee (1979: 29-30), commented that ‘these Southern African peoples

[Bushmen and Hottentots] were tagged with names not of their own choosing that

bore little or no relation to the names they used for themselves. Further, such

names often had derogatory or racist connotations.’ He was none the less content

to accept the use of ‘San’, which he described as ‘a comparatively neutral term

originally applied by the Khoi-khoi to their hunting and gathering neighbours’.

This was in spite of his further observations that “even San is not an entirely

satisfactory term. First, it too has a derogatory connotation, meaning “‘a rascal”

in Khoi-khoi. Second, because of the tremendous linguistic diversity among the

Kalahari San of today, over a dozen self-applied terms are in use by various

peoples, and San is not one of them!’ Lee’s unacknowledged source of the

connotation was probably Hahn (1881: 3), perhaps by way of Dornan (1925: 43);

and in the nineteenth century ‘rascal’ had a much stronger force than the

somewhat jocular connotation that attaches to it today. Lee’s attempt to justify

the use of the name, despite its not being the people’s own name for themselves

and its having a derogatory connotation, is as unacceptable as his defence of the

use of *!Kung’ in place of the name the people use for themselves: ‘!Kung is easier

to pronounce than Zu/’hoasi’ (Lee 1979: 31).

The evidence is thus that ‘San’, of which the meaning is not known, is not,

and apparently never has been, the peoples’ own name for themselves (but see

the observations by Burchell (1967a) and Smith (1830) under ‘Bushmen’,

pp. 258-259) and, however much it might have been accepted by anthropologists,

it has, and apparently always has had, a derogatory connotation. Its use should

not therefore be condoned, but the problem is to find an acceptable alternative.

HOTTENTOT

In 1620 De Beaulieu wrote of the people met on the shores of Table Bay:

‘Their usual greeting on meeting us is to dance a song, of which the beginning, the

middle, and the end is hautitow’ (Raven-Hart 1967: 101). In 1623 Olafsson

repeated a similar tale, but used the word ‘Hottentott’ (Raven-Hart 1967: 112). In

1627 Herbert called the people ‘Hatten-totes’ (Raven-Hart 1967: 119) and in 1640

NOMENCLATURE OF THE KHOISAN USI

De Graaf applied to them the name by which they were to become known in the

succeeding centuries, ‘Hottentots’ (Raven-Hart 1967: 154).

Although this name (or a variant of it: Bosman & Thom (1952: 20) gave

‘Ottento’, Thom (1952: 21) ‘Hottentot’ and Moodie (1960a: 9) ‘Ottentoo’) was

first recorded by the white settlers on 7 April 1652, the use of other names such as

‘Saldanhars’ or ‘Strandloopers’ was almost more common. Later, as the names of

the tribes became known, these were generally used; but the name ‘Hottentot’

was commonly used as a sort of generic for all the (non-Negro) peoples with

whom the settlers and early travellers came into contact. This included people

who would now be classified as ‘San’, for example, ‘Some Hottentots of the

Obiqua tribe’ (Moodie 1960a: 391) and ‘Sonquase Hottentots’ (Schrijver, 1689, in

Mossop 1931: 226). The name has also been applied to the Gonaqua of the

eastern Cape, who had a long history of interbreeding with their Negro

neighbours, the Xhosa (Masson 1776: 284-296; Soga 1930: 94-95; Maingard

1931: 500-501, 504; Harinck 1969: 153-159).

Objections to the use of ‘Hottentot’ are that it is not the people’s name for

themselves and that it has acquired derogatory connotations (Wilson 1969: 40;

Elphick 1977: xv, 1985: xv; Lee 1979: 29; etc.). Against the first objection it can

be argued that ‘American’ is not an indigenous name; that the name ‘Indian’, as

applied to the indigenous inhabitants of North and South America, is neither

indigenous nor correct; that ‘English’ is correctly applicable to only some of the

inhabitants of England; and that the ‘Germans’ do not call themselves by that

name when speaking their own language. What matters, surely, is whether any

name is acceptable to the people to whom it is applied. Against the second

objection there can be little argument, although Nienaber & Raper (1977: 33)

have observed that the name ‘Hottentot’ is used by the people themselves, and

without a sense of derogation. However, G. Klinghardt (1985 pers. comm.), who

has carried out research in Namaqualand, has mentioned that the name “‘Hotnot’

is generally applied to persons considered to be of lower status and never to

oneself or to equals except humorously. ‘Hottentot’ is applied specifically to the

offspring of Nama and Dama parents.

Where reference is made to ‘Hottentots’ in the early records or the literature,

it is not always possible to be sure that this term can be replaced by ‘Khoikhov’ or

some other preferred term, and in these cases the name should be retained. The

pros and cons of the use of ‘Hottentot’ and ‘Khoekhoen’ are discussed more fully

by Nienaber & Raper (1977: 29-34). The conclusion seems to be that there are

cases where the retention of ‘Hottentot’ is justified, others where it is not.

BUSHMAN

This name was first recorded in 1682 by Bergh (in Mossop 1931: 85) with

reference to people met along the Berg River: ‘We came to some Hottentots,

they being Somquaas alias Bushmen.’ In the entry in the Company’s journal for

31 October 1685 (see Elphick (1972: 58) for correction of the date given by

258 ANNALS OF THE SOUTH AFRICAN MUSEUM

Moodie) there is a reference to ‘Sonquas, commonly called Bosjesmans’ (Moodie

1960a: 399, under Oct. 3). However, despite the assertion that the Sonqua were

commonly called Bosjesmans, there is little evidence of this during the remainder

of the seventeenth century, it still being more usual to write of ‘Sonquas’,

‘Obiquas’ or variants of these (e.g. Moodie 1960a: 400-446). In the early part of

the following century Kolb mentions the “Odiquas’ and ‘Sonquas’ as well as, but

apparently distinct from them, ‘a Sort of Hottentot Banditti . . . called Buschies or

Highway Men’ (Kolb 1738: 63, 75-77, 90—Kolb’s emphasis). These last were, in

Kolb’s view, Hottentots who had become renegades and robbers, ‘finding the

Laws and Customs of their Countries to be too great Restraints upon their

Inclinations’, and not a separate race. Valentyn (1973: 60), who visited the Cape

between 1685 and 1714, merely lists the “Sonquaas’ among the Hottentot tribes,

although his transcription of Starrenburg’s journal (Valentyn 1973: 20) does

include the name ‘Bosjesmans’.

By 1770 the official documents tended to use the term ‘Bosjesmans’ or, in

translation, ‘Bushmen’ (e.g. Moodie 19605: 3, 11). In 1775 Sparrman (1975: 194)

commented that “There is another species of Hottentots, who have got the name

of Boshies-men, from dwelling in woody or mountainous places’ (Sparrman’s

emphasis). This suggests that by this time the opinion that the two groups were

different had not yet been formalized. Indeed, in 1779 Field Sergeant Charl

Marais recorded the allocation to his men of a number of ‘little Bushmen’ who

had been captured, comprising ‘a girl. . .a Hottentot. . .a Hottentot. . . alittle

Hottentot ...a girl... a little boy’ (Moodie 1960b: 81).

In 1809 Colonel Collins (Moodie 1960c: 2 ff.) referred to the ‘Bosjesmen’

without any suggestion that they were a part of the ‘Hottentot nation’. However,

in 1811 Burchell (1967a: 64) commented on the name ‘Bushman’ that “This is

often written Bosjesman, and Buschman, which merely being Dutch words

signifying men living wild among the bushes, and applied generally to several

tribes of the Hottentot race, I have preferred using the English orthography,

viewing it rather as a descriptive, than as a proper, name. They call themselves

Sdaqua; those, at least, who inhabit the country southward of the Gariep [Orange

River]. Yet it is difficult to avoid inaccuracy, in the application of one collective

name to a race of people who divide themselves into so many separate tribes.’

Smith (1830: 171-175, 179) asserted that: ‘The Aborigines of South Africa

. . will be found to have consisted, and still to consist, only of two distinct races,

namely those of the Hottentot and Caffer. ... Among these, one division has

always held, and still continue to hold, a most conspicuous position, and has ever

been proverbial with the rest, on account of its troublesome character and

universally outrageous conduct. To this the other tribes, as well as its own

member(s], apply the name of Saap or Saan, and history describes a portion

thereof under the appellation of Bushmen. . . . The majority of the Bushmen

[sic] population, according to the restricted sense in which the term is here to be

understood, consists of pure Hottentots; and the remainder of blacks either the

offspring of an intercourse with the former and other coloured persons, or else

ES EW ln cin

NOMENCLATURE OF THE KHOISAN 259

the actual outcasts of other races themselves. ... The Hottentot Bushman

presents most of the physical characters of the race as exemplified in other

situations, and the mixed description according to circumstances, exhibits more

or less of the appearances of the Negro or Caffer. . . . The language spoken by

the Bushmen, is decidedly a dialect or dialects of that in use by the Hottentots

elsewhere; but in most situations is so altered and modified, as that its origin and

dependance [sic] can scarcely be traced.’

Smith’s opinions were probably representative of the majority of ‘scientific’

opinion during the nineteenth century, i.e. that the Bushmen were biologically

Hottentots or of mixed race. However, towards the end of the century Stow

(1905: 11) suggested ‘that both Hottentots and Bushmen were descended from

the same original stock seems more likely’ and he saw the differences between

them as cultural rather than biological.

In his previously mentioned discussion of the use of ‘San’, Lee (1979: 30)

asserted that “The term Bushmen has both racist and sexist connotations’ (Lee’s

emphases). In contrast, Marshall (1976: xxi) observed: ‘While I believe that the

name Bushman was definitely derogatory when it was first applied by European

voyagers and settlers in South Africa, I have felt that it need not continue to be

derogatory unless the speaker is imputing derogation in using it. In my feelings, I

accord “‘Bushman”’ the dignity of any dignified race name.’

The objections to the use of ‘Bushmen’ are the same as those that apply to

the use of ‘Hottentot’ (see above), and in both cases it seems clear that the names

were applied to groups who were not biologically and/or culturally distinct.

KHOISAN

The term Koisan was coined by Schultze (1928: 211) and introduced into the

English-language literature as ‘Khoisan’ by Schapera (1930: 5). Schultze carried

out a biometric study of ‘Hottentot’ and ‘Bushman’ samples, from which he

concluded that, while there was sufficient to distinguish these people from all the

other races of mankind, he was unable to find differences between the two groups

that were adequate to justify their being kept separate. For various reasons (see

Wilson in press) Schultze’s study is not acceptable today, but the name he coined

has gained currency as a sort of ‘blanket generic’ for the peoples discussed in this

paper.

Westphal (1963: 243) asserted that ‘The term ‘“‘Khoisan”’ is of no linguistic

usefulness whatever.’ Elsewhere (Westphal 1971: 368-369) he stated that “Today

the majority of linguists who have actually been working on these languages do

not accept this classification’ of the non-Bantu languages of southern Africa as

belonging to the ‘Khoisan family’ of languages. He also commented that

‘“Khoisan” ... means “the food-gathering people” viz. Khoi-“person’’, sa

“gather food’’, -n “plur. common suffix’”’’. This is orthographically incorrect: the

order of the first two parts would be reversed in the formation of a word with the

meaning Westphal attributes to ‘Khoisan’, as, for example, in “‘Soaqua’, ‘Sonqua’,

260 ANNALS OF THE SOUTH AFRICAN MUSEUM

‘Saan’ or ‘Sana’, the -n(a) suffix having the same force and meaning as -qua

(‘people’) as, for example in ‘Namaqua’, ‘Naman(a)’.

More recently, Westphal (1986 in litt.) has commented: ‘(1) The word

Khoisan = Khoesan has no linguistic merit. It has not been shown that the

Khoisan people spoke a uniform language nor that their languages derive from a

single common ancestor. (2) The terms for the Janguages are derived from the

word for “people”... as in the case of the word BANTU. Thus KHOE, ZHU (or

whatever orthographic compromise is used), TA’A, !wr are all roots meaning

“people” and could be linguistically useful. In the case of the zHU language the

language name !Xt could be used for they say: Mi okx’oi !xi=I speak !X@.’

(Westphal’s emphases. The three dots after ‘people’ in (2) do not indicate an

ellipsis but are in Westphal’s letter.)

Contrary to Westphal’s first point, Hiernaux (1974: 98) stated that “There is

no basis for linguistic distinction between the two groups [Hottentots and

Bushmen]: the language of the Hottentots is just one of the central South African

languages, close to that spoken by the Naron Bushmen.’ It could, however, be

argued that the Naron (= Nharo) may be the descendants of Khoikhoi-speaking

herders who lost their stock and were obliged to subsist by hunting and gathering.

An alternative possibility is that they may have been clients of the Nama and, like

the Dama, adopted (and adapted) the Nama language.

Although Westphal is correct in stating that it has not been shown that the

people spoke a common language nor that their languages derive from a single

common ancestor, it has equally not been shown that they did not. The ‘Khoisan’

languages are more like each other than they are like any of the other languages

of Africa, possibly even including the ‘click’-using languages of the Hadza and

Sandawe of East Africa. This suggests that they are descended from a common

ancestor, or that they are amalgams of a common ancestral language with a

number of other languages. Scientific linguistic studies of the ‘Khoisan’ languages

do not extend back more than about a century and it must be borne in mind that

languages do evolve, converging or diverging according to the degree of contact

or isolation of their speakers. The distinctiveness of the ‘Khoisan’ languages from

the other African languages should therefore entitle them to be included in a

‘Khoisan family’ of languages.

Westphal’s second point is well made, but refers only to individual

languages: what is needed is a corporate term to include all the languages, as with

the peoples themselves—hence ‘Khoisan’.

HERDERS AND HUNTERS

These terms, popularized by Wilson (1969: 41-74) as alternatives to the

disreputable ‘Hottentots’ and ‘Bushmen’, at first glance appear to be nice,

succinct terms, useful for archaeologists, who have no linguistic or, as yet,

cultural or physical anthropological evidence on which to base any ascription of

difference of race or culture to the people whose remains they study. However, as

NOMENCLATURE OF THE KHOISAN 261

Wilson was well aware and Marks (1972), Elphick (1977, 1985) and Schrire (1980)

have more recently demonstrated (or reiterated), this neat dichotomy is specious.

As is abundantly clear from the early records, for the herders, by whatever

names they were called, hunting, gathering, collecting, and in some cases fishing

and even scavenging were an integral part of their socio-economic existence. It is

thus improper to exclude, even by inference, these essential elements of the

herder way of life, by calling the non-pastoralist peoples ‘hunters’ or any version

of this name that is expanded to include one or more of the aspects mentioned

above. The processes by which pastoralists lost, and sometimes regained, their

stock and non-pastoralists acquired domestic animals for other than immediate

consumption are discussed by the above-named authors and need not be

reiterated here.

Although anthropologists engaged in southern African prehistoric and

protohistoric research are aware that the herder:hunter dichotomy is not as

absolute as the terms might imply, those outside the field may not be. Terms like

‘herder’, “pastoralist’, ‘hunter-gatherer’ are part of anthropology’s jargon and

thus need to be explained for the benefit of those not familiar with the discipline’s

terminology. There seems, otherwise, no good reason for replacing them with

terms that, while they might be more accurate, might also be more cumbersome

or even more confusing.

DISCUSSION AND CONCLUSIONS

The principal question to be answered is: what name or names can

archaeologists use for the people whose remains—physical and cultural—they

study from the sites they excavate? A corollary to this is that, since archaeology of

necessity involves itself in other branches of anthropology (history, ethnology,

physical and social anthropology, for example) in its attempts to elucidate its

data, is there a terminology that can be adopted that will be acceptable to

anthropology as a whole? The consensus reached at the 1971 conference,

mentioned in the introduction to this paper, suggests that there is not. It is not

clear, however, whether this decision was based on a genuine need to have three

distinct sets of terms or was merely the result of a reluctance on the part of some

of the participants to replace their particular preferences with those of others, or

to identify themselves too closely with the other branches of the discipline. There

seems no good reason why, for example, physical anthropologists and linguists

cannot use the same names or terms, as long as it is made clear that there is no

necessary correlation between physical type and language—or, for that matter,

any aspect of culture. Haacke (1985 in litt.) has, however, mentioned that the

Nama-Damara language is referred to in scientific circles as Khoekhoegowab.

When it is necessary to refer to speakers of other ‘Hottentot’ dialects, for example

the Hai//om, Haacke calls them ‘Khoekhoe-speakers’.

Elphick (1977: xxi—xxii, 1985: xxi—xxii) considered ‘a Khoikhoi to be any

person accepted as a full (i.e. not a subordinate) member of a Khoikhoi

262 ANNALS OF THE SOUTH AFRICAN MUSEUM

community. A Khoikhoi community was one where a dialect of the Khoikhoi

language was spoken and where pastoralism was the preferred mode of economic

life’ (Elphick’s emphases). These three criteria, though fundamental, are

minimal. Probably justifiably, Elphick avoided ascription of a physical type (see

Elphick 1977: 8-10, 1985: 8-10 for his views) but there are many other cultural

factors besides language and ‘preferred mode of economic life’ that are

important, though perhaps not to a historian. By changing the name and

substituting ‘hunting’ for ‘pastoralism’, Elphick’s criteria could be used to define

the archetypal San. Archetypes are, however, generally useful only for the

definition of the opposite poles of a range, and in the human situation these rarely

exist.

Elphick’s criteria are not useful for archaeologists who, in the specifically

Khoisan context, would probably not be able to identify the social status of the

individuals whose skeletal remains might be excavated. Of language there would,

of course, be no trace, and the ‘preferred mode of economic life’ could only be

inferred. An example of the need for caution in interpreting archaeological data is

that of Kaabi’s kraal, visited in 1812 by Burchell (19675: 197, pl. 4). Kaabi and his

people were called Bushmen, probably on information from Burchell’s Hottentot

servants and guides. The illustration of the kraal shows it in many respects to be

typical of a Hottentot kraal: hemispherical mat-and-wattle huts distributed in a

rough circle around an open area (the kraal); and the text indicates that the

people had fifty oxen and about two hundred sheep. They also had clay pots

(Burchell 1967b: 45-47). That the stock were probably stolen is irrelevant:

according to Elphick (1977: 11-12, 1985: 11-12) that is possibly how the ‘proto-

Khov acquired their first stock ‘and became by that act the first Khoikhoi’. An

archaeologist who excavated the site of Kaabi’s kraal might correctly deduce it to

be a pastoralist camp-site, but would be incorrect in inferring from this that it was

a Khoikhoi site.

Another example of the problem of successfully applying Elphick’s three

criteria is the case of the Bitterbos clan of the Korana (Engelbrecht 1936: 17). On

their own information, they were originally Bushmen and some of them later

married into the Kats clan. These were considered to be hybrids by the others,

who saw themselves still as ‘pure’ Bushmen. The Korana are classified as

Hottentots and there is no evidence that the Bitterbosse, ‘pure’ or ‘hybrid’, were

not accepted as full members of the Korana community and therefore as

Hottentots, even though the ‘pure’ segment considered themselves Bushmen.

Elphick (1977: xxi—xxii, 1985: xxi—xxii), for reasons given, eschewed use of

the name ‘San’, preferring instead to use terms such as ‘hunters’ or ‘aborigines’.

This places him in the somewhat invidious position of having a proper name,

Khoikhoi, for one group, and common names for the others. This is analogous to

referring on the one hand to ‘Americans’ and on the other to ‘bankers’ or

‘natives’. The fact that hunting was an integral part of the herder and thus

Khoikhoi way of life has already been mentioned, and there is no proof that the

Khoikhoi were not aborigines: it has yet to be demonstrated that it was

NOMENCLATURE OF THE KHOISAN 263

pastoralists who moved into areas formerly only occupied by hunter-gatherers, or

whether it was the practice of pastoralism that was adopted by various groups and

thus moved into the areas in which it was observed from the fifteenth century

onward.

Elphick (1977: 11, 1985: 11), following Westphal (1963: 259), postulated an

area of origin in north-eastern Botswana for the ‘proto-Khoi’. If this was indeed

the case, it follows that these people differed biologically and (possibly)

linguistically very little, if at all, from the other hunter-gatherer bands in the area.

Assuming that it was pastoralists and not just pastoralism that moved, unless the

‘proto-Khoi bands were sufficiently large to provide an adequate number of

spouses, those moving into a region would have had to acquire spouses (probably

wives) from among the hunter-gatherer groups of the new region. The

consequence of this would be increasing genetic and possibly linguistic diversity.

The Khoikhoi pastoralists probably came in the course of time to regard

themselves as superior to their non-pastoralist neighbours and would thus have

chosen their marriage partners from other pastoralist communities. This would

have had the effect of tending to isolate the Khoikhoi from the non-pastoralists

and to increase biological, social, cultural, and linguistic similarity within their

communities while increasing the differences between them and non-Khoikhoi

communities. It seems likely, therefore, that any biological and linguistic

distinctions that can now be observed must result from a long and complex

process of interaction and/or isolation. It should consequently be anticipated that

archaeological sites will cover every part of the spectrum that ranges between the

archetypal pastoralist kraal and the archetypal hunter-gatherer camp, although it

can be argued that the archetypal herder site does not exist, pastoralism being an

addition to, rather than a replacement of, the hunter-gatherer way of life.

Parkington (1984) has suggested that there is archaeological evidence for

changes in the life-style of the hunter-gatherer populations of the Cape west-coast

region subsequent to the arrival of the pastoralists in that area, and again in the

early historical period, when the pastoralists lost their stock to, and were

displaced by, the white settlers. He has used the terms ‘soaqua’ and ‘bushman’ to

denote these altered life-styles, which he considers should be differentiated from

that of the aboriginal hunter-gatherers in their pristine habitat. It is felt, however,

that the adoption of these terms will result in confusion, especially when they

appear in papers that also use ‘Soaqua’, or one of its variant spellings, and/or

‘Bushman’.

In conclusion, while it seems that there is no real justification for the

retention of three separate sets of terms as advocated at the 1971 conference, it is

considered that archaeologists, qua archaeologists, should confine themselves to

using (socio-)economic terms such as ‘herder’ or ‘pastoralist’ and ‘hunter-

gatherer’ when dealing with material from prehistoric sites. It is considered that

the term ‘forager’, now gaining currency as a substitute for ‘hunter-gatherer’,

(e.g. Lee 1979) should be abandoned. The Concise Oxford Dictionary (1978: 410)

gives the following meanings for ‘forage’: ‘food for horses and cattle, esp. for

264 ANNALS OF THE SOUTH AFRICAN MUSEUM

horses in army; ... collect forage from, ravage; supply with forage, get by

foraging ... search for forage; search for (thing), rummage’. The word is

therefore better applied to the herders than to the hunters; and the implication of

‘rummage’ does less than justice to peoples whose modern descendants are well

known to have an acute knowledge of the availability and location of the

resources of their environment. When it is necessary to deal with the period from

1488 to the present, whether on the basis of sites or documentary records, the

choice of terms such as ‘Hottentot’ or ‘Khoikhov’ or ‘Bushman’ or ‘San’ should be

dictated by the context or individual preference. Where necessary, the usage

should be explained and justified.

ACKNOWLEDGEMENTS

Thanks are due, for the provision of information and comment, to:

Mr W. H. G. Haacke, Department of African Languages, The Academy,

Windhoek; Professor G. S. Nienaber, Institute for Research into Language and

the Arts, Human Sciences Research Council, Pretoria; Messrs G. Avery and

G. Klinghardt, South African Museum; Professor E. O. J. Westphal, formerly of

the Department of African Languages, University of Cape Town; and to

Mrs S. Saven, also of the South African Museum, for typing the manuscript.

REFERENCES

BOESEKEN, A. J. 1972. The meaning, origin and use of the terms Khoikhoi, San and Khoisan.

Cabo 1 (1): 5-10.

BGOESEKEN, A. J. 1974. The meaning, origin and use of the terms Khoikhoi, San and Khoisan:

Dr A. J. Boeseken replies to Prof. R. H. Elphick. Cabo 2 (2): 8-10.

BGOESEKEN, A. J. 1975. On changing terminology in history. Cabo 2 (3): 16-18.

Bosman, D. B. & THoM, H. B. eds. 1952. Daghregister gehouden by den Oppercoopman Jan

Anthonisz. van Riebeeck. I: 1651-1655. Cape Town: A. A. Balkema for the Van Riebeeck

Society.

BURCHELL, W. J. 1967a. Travels in the interior of southern Africa 1. Cape Town: C. Struik.

BURCHELL, W. J. 1967b. Travels in the interior of southern Africa 2. Cape Town: C. Struik.

CoNCISE OxForRD DicTIONARY. 1978. J. B. Sykes ed. Sixth edition. Oxford: Oxford University

Press.

DEPARTMENT OF BANTU EpucaTION, S.W.A. 1977. Nama/Damara: Orthography no. 2. (?Pre-

toria: Government Printer.)

Dornan, S. S. 1925. Pygmies and Bushmen of the Kalahari. London: Seeley, Service & Co.

Ltd. Photostatic reprint 1975. Cape Town: C. Struik.

Evpuickx, R. H. 1972. The Cape Khoi and the first phase of South African race relations. Ph.D.

dissertation, Harvard University. Ann Arbor, Mich.: Xerox University Microfilms.

Evpuick, R. H. 1974. The meaning, origin and use of the terms Khoikhoi, San and Khoisan:

Prof. R. H. Elphick replies to Dr A. J. Boeseken. Cabo 2 (2): 3-7.

Evpuick, R. H. 1975. The meaning, origin and use of the terms Khoikhoi, San and Khoisan:

Prof. R. H. Elphick’s final rebuttal of Dr Boeseken’s argument. Cabo 2 (3): 12-15.

Evpuick, R. H. 1977. Kraal and castle: Khoikhoi and the founding of white South Africa. New

Haven & London: Yale University Press.

Evpuick, R. H. 1985. Khoikhoi and the founding of white South Africa. Johannesburg: Ravan

Press.

ENGELBRECHT, J. A. 1936. The Korana: an account of their customs and their history, with

texts. Cape Town: Maskew Miller.

NOMENCLATURE OF THE KHOISAN 265

HaackeE, W. H. G. 1982. Traditional hut-building technique of the Nama (with some related

terminology). Cimbebasia (B) 3 (2): 77-98.

Haun, T. 1881. Tsuni-//Goam: the supreme being of the Khoi-khoi. London: Trubner & Co.

HarINcK, G. 1969. Interaction between Xhosa and Khoi: emphasis on the period 1620-1750.

In: THompson, L. ed. African societies in southern Africa: 145-170. London: Heinemann.

HIERNAUX, T. 1974. The people of Africa. London: Weidenfeld & Nicolson.

JENKINS, T. & Tosias, P. V. 1977. Nomenclature of population groups in southern Africa.

African studies 36 (1): 49-55.

KoB(EN), P. 1738. The present state of the Cape of Good Hope \. (Transl. G. Medley)

London: Innis & Manby.

LEE, R. B. 1979. San, Bushman, Basarwa: a question of names. In: LEE, R. B. The !/Kung San:

men, women and work in a foraging society: 29-38. Cambridge: Cambridge University

Press.

MAINGARD, L. F. 1931. The lost tribes of the Cape. South African Journal of Science 28:

487-504.

Marks, S. 1972. Khoisan resistance to the Dutch in the seventeenth and eighteenth centuries.

Journal of African History 13 (1): 55-80.

MarsHaL, L. 1976. The !Kung of Nyae Nyae. Cambridge, Mass. & London: Harvard

University Press.

Masson, F. 1776. Mr Masson’s botanical travels. Philosophical Transactions of the Royal

Society of London 66: 269-317.

MoopiE, D. ed. 1960a. The Record: Part I. 1649-1720. Amsterdam & Cape Town: A. A.

Balkema.

MooprE, D. ed. 1960b. The Record: Part III. 1769-1795. Amsterdam & Cape Town: A. A.

Balkema.

MooprE, D. ed. 1960c. The Record: Part V: 1808-1819. Amsterdam & Cape Town: A. A.

Balkema.

Mossop, E. E. ed. 1931. Journals of the expeditions of the Honourable Ensign Olof Bergh (1682

and 1683) and Isaq Schrijver (1689). Cape Town: Van Riebeeck Society (First Series no.

12).

NIENABER, G. S. 1963. Hottentots. Pretoria: J. L. van Schaik.

NIENABER, G. S. & Raper, P. E. 1977. Toponymica Hottentottica I: A-G. Pretoria: S.A.

Naamkundesentrum, Raad vir Geesteswetenskaplike Navorsing.

PARKINGTON, J. E. 1984. Soaqua and Bushmen: hunters and robbers. Jn: SCHRIRE, C. ed. Past

and present in hunter gatherer studies: 151-174. New York: Academic Press.

RAvEN-Hart, R. 1967. Before Van Riebeeck: callers at South Africa from 1488 to 1652. Cape

Town: C. Struik.

Rust, F. ed. 1969. Nama Worterbuch. Pietermaritzburg: University of Natal Press.

SCHAPERA, I. 1930. The Khoisan peoples of South Africa: Bushmen and Hottentots. London:

George Routledge & Sons.

SCHRIRE, C. 1980. An inquiry into the evolutionary status and apparent identity of San hunter-

gatherers. Human Ecology 8 (1): 9-32.

SCHULTZE, L. 1928. Zur Kenntnis des Korpers der Hottentotten und Buschmdnner. Jena:

Gustav Fischer Verlag (Jenaische Denkschrift 18).

SmiTH, A. 1830. Observations relative to the origin and history of the Bushmen. The South

African Quarterly Journal 1: 171-189.

SmiTH, C. A. 1966. Common names of South African plants. (Edited by E. P. Phillips & E. van

Hoepen.) Botanical Survey Memoir 35: 1-642.

Soca, J. H. 1930. The south-eastern Bantu. Johannesburg: Witwatersrand University Press.

SPARRMAN, A. 1975. A voyage to the Cape of Good Hope, towards the Antarctic Polar Circle,

round the world, and to the country of the Hottentots and the Caffres, from the year

1772-1776 1. (Edited by V. S. Forbes.) Cape Town: Van Riebeeck Society (Second Series

no. 6).

Stow, G. W. 1905. The native races of South Africa. (Edited by G. M. Theal.) London: Swan

Sonnenschein; New York: The Macmillan Co.

Tuom, H. B. ed. 1952. Journal of Jan van Riebeeck: I: 1651-1655. Cape Town & Amsterdam:

A. A. Balkema for the Van Riebeeck Society.

Tuom, H. B. ed. 1954. Journal of Jan van Riebeeck: I: 1656-1658. Cape Town & Amsterdam:

A. A. Balkema for the Van Riebeeck Society.

266 ANNALS OF THE SOUTH AFRICAN MUSEUM

THomM, H. B. ed. 1958. Journal of Jan van Riebeeck: III: 1659-1662. Cape Town &

Amsterdam: A. A. Balkema for the Van Riebeeck Society.

VALENTYN, F. 1973. Description of the Cape of Good Hope with the matters concerning it.

Amsterdam 1726. Part II. (Edited by E. H. Raidt. English transl. R. Raven-Hart.) Cape

Town: Van Riebeeck Society (Second Series no. 4).

WESTPHAL, E. O. J. 1963. The linguistic prehistory of southern Africa: Bush, Kwadi,

Hottentot, and Bantu linguistic relationships. Africa 33: 237-265.

WESTPHAL, E. O. J. 1971. The click languages of southern and eastern Africa. Jn: Berry, J. &

GREENBERG, J. H. eds. Linguistics in sub-Saharan Africa: 367—420. The Hague: Mouton

(Current Trends in Linguistics 7).

Witson, M. 1969. The hunters and herders. /n: Witson, M. & THompson, L. eds. The Oxford

History of South Africa 1: 40-74. Oxford: The Clarendon Press.

Witson, M. L. In press. Khoisanosis: the question of separate identities for Khoi and San. In:

SINGER, R. & Lunpy, J. K. eds. Variation, culture and evolution in African populations:

papers in honour of Dr Hertha de Villiers. Johannesburg: Witwatersrand University Press.

6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu-

larly Articles 22 and 51).

Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed

by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov.,

etc.

An author’s name when cited must follow the name of the taxon without intervening punctuation

and not be abbreviated; if the year is added, a comma must separate author’s name and year. The

author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is trans-

ferred from its original genus. The name of a subsequent user of a scientific name must be separated

from the scientific name by a colon.

Synonymy arrangement should be according to chronology of names, i.e. alli published scientific

names by which the species previously has been designated are listed in chronological order, with all

references to that name following in chronological order, e.g.:

Family Nuculanidae

Nuculana (Lembulus) bicuspidata (Gould, 1845)

Figs 14-15A

Nucula (Leda) bicuspidata Gould, 1845: 37.

Leda plicifera A. Adams, 1856: 50.

Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b).

Nucula largillierti Philippi, 1861: 87.

Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9.

Note punctuation in the above example:

comma separates author’s name and year

semicolon separates more than one reference by the same author

full stop separates references by different authors

figures of plates are enclosed in parentheses to distinguish them from text-figures

dash, not comma, separates consecutive numbers.

Synonymy arrangement according to chronology of bibliographic references, whereby the year is

placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable.

In describing new species, one specimen must be designated as the holotype; other specimens

mentioned in the original description are to be designated paratypes; additional material not regarded

as paratypes should be listed separately. The complete data (registration number, depository, descrip-

tion of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.:

Holotype

SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach, Port Eliza-

beth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973.

Note standard form of writing South African Museum registration numbers and date.

7. SPECIAL HOUSE RULES

Capital initial letters

(a) The Figures, Maps and Tables of the paper when referred to in the text

e.g. “. .. the Figure depicting C. namacolus .. .’: ‘. . . in C. namacolus (Fig. 10) .. .’

(b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by

initials or full names

e.g. Du Toit but A.L. du Toit; Von Huene but F. von Huene

e.g. Therocephalia, but therocephalian

Punctuation should be loose, omitting all not strictly necessary

Reference to the author should preferably be expressed in the third person

Roman numerals should be converted to arabic, except when forming part of the title of a book or

article, such as

‘Revision of the Crustacea. Part VIII. The Amphipoda.’

Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial

capital letter, provided the same generic name is used consecutively. The generic name should

not be abbreviated at the beginning of a sentence or paragraph.

Name of new genus or species is not to be included in the title; it should be included in the abstract,

counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts.

M. L. WILSON

NOTES ON THE NOMENCLATURE

OF THE KHOISAN

Toa: 72, /

ae SL

S67X |E 97 PART 9 JANUARY 1987 ISSN 0303-2515

NH |

ITHSON 2

cc a 2 ee "

6M MAN

‘ MAR 30 1987

we ~SIBRARIES

ANNALS

OF THE SOUTH AFRICAN |

MUSEUM

CAPE TOWN

INSTRUCTIONS TO AUTHORS

1. MATERIAL should be original and not published elsewhere, in whole or in part.

2. LAYOUT should be as follows:

(a) Centred masthead to consist of

Title: informative but concise, without abbreviations and not including the names of new genera or species

Author’s(s’) name(s) :

Address(es) of author(s) (institution where work was carried out)

Number of illustrations (figures, enumerated maps and tables, in this order)

(b) Abstract of not more than 200 words, inteiligible to the reader without reference to the text

(d) Introduction

(e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents

(f) Summary, if paper is lengthy

(g) Acknowledgements

(h) References

(i) Abbreviations, where these are numerous.

3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with

3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for

illustrations should be typed separately, their positions indicated in the text. All pages should be num-

bered consecutively.

Major headings of the paper are centred capitals; first subheadings are shouldered small capitals;

second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further

subdivisions should be avoided, as also enumeration (never roman numerals) of headings and

abbreviations.

Footnotes should be avoided unless they are short and essential.

Only generic and specific names should be underlined to indicate italics; all other marking up

should be left to editor and publisher.

4. ILLUSTRATIONS should be reducible to a size not exceeding 12x 18cm (19 cm including

caption); the reduction or enlargement required should be indicated (and preferably uniform); orig-

inals larger than 35 X 47 cm should not be submitted; photographs should be rectangular in shape and

final size. A metric scale should appear with all illustrations, otherwise magnification or reduction

should be given in the caption; if the latter, then the final reduction or enlargement should be taken

into consideration.

All illustrations, whether line drawings or photographs, should be termed figures (plates are not

printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items

of composite figures should be designated by capital letters; lettering of figures is not set in type and

should be in lower-case letters. If Letraset is used authors are requested to use Helvetica-style letter-

ing, if possible.

The number of the figure should be lightly marked in pencil on the back of each illustration.

5. REFERENCES cited in text and synonymies should all be included in the list at the end of the

paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable):

(a) Author’s name and year of publication given in text, e.g.:

‘Smith (1969) describes... .’

‘Smith (1969: 36, fig. 16) describes . . .’

‘As described (Smith 1969a, 1969b; Jones 1971)’

‘As described (Haughton & Broom 1927)...’

‘As described (Haughton et al. 1927)...’

Note: no comma separating name and year

Pagination indicated by colon, not p.

names of joint authors connected by ampersand

et al. in text for more than two joint authors, but names of all authors given in list of references.

(b) Full references at the end of the paper, arranged alphabetically by names, chronologically within

each name, with suffixes a, b, etc., to the year for more than one paper by the same author in

that year, e.g. Smith (1969a, 1969b) and not Smith (1969, 1969a).

For books give title in italics, edition, volume number, place of publication, publisher.

For journal article give title of article, title of journal in italics (according to the World list of scientific periodicals. 4th ed.

London: Butterworths, 1963), series in parentheses, volume number, part number in parentheses, pagination (first and

last pages of article).

Examples (note capitalization and punctuation)

BuLLouGu, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan.

FiscHER, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140.

FiscHER, P. H., Duvat, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie

expérimentale et générale 74 (33): 627-634.

Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and

Magazine of Natural History (13) 2 (17): 309-320.

Koun, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of

the Bingham Oceanographic Collection, Yale University 17 (4): 1-51.

THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthro-

pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika ausgefiihrt in den Jahren

1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270.

(continued inside back cover)

ANNALS OF THE SOUTH AFRICAN MUSEUM

ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

Volume 97 Band

January 1987 Januarie

ante oe Deel

A RE-EVALUATION

OF THE SOUTH AFRICAN SPECIES OF

LEMBOIDES STEBBING AND LEMBOS BATE

(AMPHIPODA, AORIDAE)

DESCRIBED BY K. H. BARNARD (1916)

A. A. MYERS

J. LYONS

Cape Town Kaapstad

The ANNALS OF THE SOUTH AFRICAN MUSEUM

are issued in parts at irregular intervals as material

becomes available

Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000

Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

word uitgegee in dele op ongereelde tye na gelang van die

beskikbaarheid van stof

Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000

OUT OF PRINT/UIT DRUK

I GES, SEN ED ES yA), SUES), 5, ED),

6(1, t-—p.i.), 7(1-4), 8, 9(1-2, 7), 10(1-3),

111-2, 5,7. tpi), 1402), 155), 240), 277 3103) 326)s sanse@4s)

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 088 5

Printed in South Africa by In Suid-Afrika gedruk deur

The Rustica Press, Pty., Ltd., Die Rustica-pers, Edms., Bpk.,

Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap

A RE-EVALUATION OF THE SOUTH AFRICAN SPECIES OF

LEMBOIDES STEBBING AND LEMBOS BATE

(AMPHIPODA, AORIDAE) DESCRIBED BY

K. H. BARNARD (1916)

A. A. Myers & J. LYONS

Department of Zoology, University College, Cork, Ireland

(With 9 figures)

[MS accepted 30 June 1986]

ABSTRACT

Barnard (1916) described three new species of aorid amphipod, Lembos hypacanthus,

Lemboides acanthiger and Lemboides crenatipalma from South Africa. These were poorly

figured, and have not since been fully figured or redescribed, although their distribution in South

African waters has been assessed by Griffiths (1974a, 1974b, 1974c, 1975). Until now, none was

known from outside South Africa.

In the present work, all three species are described and figured, Lemboides crenatipalma is

transferred to the genus Aorchoides Ledoyer (family Isaeidae) and Lembos hypacanthus is

recorded for the first time from North America. Lemboides afer Stebbing, a South African

endemic, and the type-species of the genus, is also described and figured for comparison with

L. acanthiger.

CONTENTS

PAGE

GRO GU CHORE rr ae ee ere soe Mingo Goma he ties cigs 267

SWSUCEI AICS meer Mata ne cee ta erie eit Sin boobies SOIR RS avermee 268

VAOKEHOIGCS ERCNGUPAIING@ yr. ia a ei Se ee es on: 268

WD CrD OLACHACANUNICCE |. Mac tn eee a> oA seek ae anne Di

WE CIMDOLACS GO [Chin = na tey nae eet | ee en we Be bs AUS)

MW CINDOS RY DGCQRERUS) 2) .0 ne ANS dh oad nie ae cee 278

[DIS SUISSTOYD. 8 Gee te Mee LO i 281

NCknoawledeeinents Lisle ci 5 ah Ween wn eeh ewe. eee ae 281

IREHOREMICES 5g bs wie SO OED SEO ie ate oa eee ee 281

INTRODUCTION

The genus Lemboides was described by Stebbing (1895) to include the single

South African species L. afer Stebbing. Later, K. H. Barnard (1916) described

two further South African species, L. acanthiger and L. crenatipalma, which he

ascribed to the same genus. To date, no further South African species have been

added to the genus. The genus has been recorded from outside South Africa

(Australia and Madagascar), but the relationships of these species are not

considered here (see Myers in press).

In the present work, material of all three South African species of Lemboides

Zo)

Ann. S. Afr. Mus. 97 (9), 1987: 267-282, 9 figs.

268 ANNALS OF THE SOUTH AFRICAN MUSEUM

was examined, together with material of the poorly known aorid Lembos

hypacanthus Barnard. Preliminary investigation revealed that Lemboides crenati-

palma should be removed from the family Aoridae and transferred to the

Isaeidae. It is described and figured herein and placed in the genus Aorchoides

Ledoyer. Lemboides acanthiger is difficult to assign and brings into question the

validity of the genus Lemboides. Stebbing (1895) gives only a single character

separating Lemboides from.Lembos, viz.: male gnathopod 1 ‘fifth joint is much

broader and longer than sixth’. Whilst this is true of Lemboides afer, the type-

species, it does not apply to L. acanthiger, in which the carpus and propodus of

the male gnathopod 1 are subequal, just as they are in Lembos hypacanthus.

However, synonymization of Lemboides with Lembos would not be advisable at

this stage, since Lemboides differs from the type-species of Lembos, i.e.

L. websteri (though not from most other species of the genus), by the short

peduncle of uropod 3. The splitting of the genus Lembos into several component

genera is being proposed elsewhere (Myers in press). A character of phylogenetic

significance is exhibited by the maxilliped of L. hypacanthus. Wing-like exten-

sions occur on the anterior margin of the inner and outer plate of this appendage.

This character is also present in some, but not all, Microdeutopus species, and in

L. websteri. Lembos websteri is unique among Lembos in having the uropod 3

rami subequal and equal in length with the elongate peduncle, a character shared

with Microdeutopus sensu stricto. Thus L. websteri in its maxilliped and uropod 3

structure aligns itself with Microdeutopus rather than with other Lembos species.

This is nomenclaturally confusing, since L. websteri is the type-species of the

genus. Lembos hypacanthus, which forms a geminate pair with L. websteri, is

here shown to be transatlantic. In the East Atlantic L. websteri and L. hypacan-

thus are widely allopatric. In the West Atlantic their distributions have not yet

been worked out, but they are probably closely allopatric or parapatric.

SYSTEMATIC SECTION

Family Isaeidae Dana, 1853

Genus Aorchoides Ledoyer, 1972

Aorchoides crenatipalma (K. H. Barnard, 1916) comb. nov.

Figs 1A, 2-3

Lemboides crenatipalma K. H. Barnard, 1916: 240, pl. 28 (figs 9-10). Griffiths, 1976: 34,

fig. 18C.

Type locality

Baboon Point ENE, distant 13 miles (off Saldanha Bay), 32 fathoms.

Material examined

SAM-A18949, 13 6, 24 2, Skrywershoek, Langebaan Lagoon, mud,

26 April 1949, UCT LB189P. SAM-A19396, 1 3, 1 2, 26°35’S 15°01’E, 71 m,

rock, 10 June 1963, UCT SWD13T. SAM-—A19397, 1 gd, 1 9, 25°51’'S 14°50,7'E,

60 m, 13 November 1948, UCT AFR1335A. SAM-A19398, 3 6, 1 2, 32°44’S

18°01’E, 11 m, 22 September 1960, UCT WCD68B.

269

SOUTH AFRICAN SPECIES OF LEMBOIDES AND LEMBOS

Fig. 1. A. Aorchoides crenatipalma (K. H. Barnard), male, 9,0 mm, Skrywershoek.

B. Lemboides acanthiger K. H. Barnard, male, 5,5 mm, Natal.

C. Lembos hypacanthus K. H. Barnard, male, 6,0 mm, Dalebrook (False Bay).

270 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 2. Aorchoides crenatipalma (K.H. Barnard), male, 7,0 mm, Langebaan Lagoon.

A. Gnathopod 2. B. Gnathopod 2 palm (enlarged). C. Gnathopod 1. D. Gnathopod 1 palm

(enlarged). E. Maxilla2. F. Mandible. G.Labrum. H. Labium. I, J, L. Maxilliped.

K. Maxilla 1.

SOUTH AFRICAN SPECIES OF LEMBOIDES AND LEMBOS BEN

K \\

Fig. 3. Aorchoides crenatipalma (K.H. Barnard), female, 8,0 mm, Langebaan Lagoon.

A. Gnathopod 1. B. Gnathopod 1 palm (enlarged). C. Gnathopod 2. D. Gnathopod 2 palm

(enlarged). E. Pereopod 3. F. Pereopod 4. G. Pereopod 5. H. Pereopod 6. I. Pereopod 7.

J. Uropod 2. K. Telson. L. Uropod 3. M. Uropod 1.

22 ANNALS OF THE SOUTH AFRICAN MUSEUM

Description

Head with eye lobes strongly produced, subocular margin strongly excavate,

eye situated in ocular lobe. Antennae 1 and 2 subequal, setose, accessory

flagellum 3-articulate, antenna 2 slender. Labium mandibular lobes only moder-

ately produced, subacute. Mandibular palp article 2 longer than article 3, article 3

truncate, approximately parallel-sided. Coxae 1-4 longer than broad, coxa 3

longest, produced forward distally. Gnathopod 1 ¢ basis enlarged, posterior

distal margin with long setae, carpus slender, longer than propodus, propodus

slender, palm evenly rounded (subadults) or produced into weak triangular

defining tooth (hyperadults). Gnathopod 1 @ slender, carpus longer than

propodus. Gnathopod 2 ¢ robust, basis maximally two-thirds as broad as long,

carpus longer than propodus, palm defined by triangular tooth. Gnathopod 2 2

slender, propodus a little longer than carpus, palm irregular. Pereopod 7 scarcely

longer than pereopod 6. Epimera 1—3 rounded. Uropod 1 peduncle longer than

rami with interramal tooth. Uropod 2 inner ramus longer than outer and longer

than peduncle, which lacks an interramal tooth. Uropod 3 peduncle short, inner

ramus one and a half times length of peduncle. Telsonic crests each with a stout

spine.

Remarks

This species shows numerous characters that indicate its isaeid ancestry.

Notably, the setose subequal antennae, slender antenna 2, extended eye lobes,

strongly regressed subocular margin. The deep coxae and short pereopod 7

immediately distinguish it from Lembos and Lemboides species.

Distribution

Southern African endemic.

Family Aoridae Stebbing, 1899

Genus Lemboides Stebbing, 1895

Lemboides acanthiger K. H. Barnard, 1916

Figs 1B, 4—5

Lemboides acanthiger K. H. Barnard, 1916: 239, pl. 28 (figs 7-8). Griffiths, 1976: 34, fig. 18A.

Type locality

Umvoti River mouth N by W i W, distant 15 miles (Natal), 56 fathoms.

Material examined

SAM-—A18947, 2 36,1 2, Natal, 29°53’S 31°06’E, 71 m, mud, 13 July 1959,

UCT NAD27C.

Description

Head with eye lobes moderately produced, eye small. Antennae missing in

all material examined. Labium mandibular lobes acute. Mandibular palp article 3

SOUTH AFRICAN SPECIES OF LEMBOIDES AND LEMBOS DAS

Fig. 4. Lemboides acanthiger K. H. Barnard, female, 6,0 mm, Natal. A. Gnathopod 2 palm

(enlarged). B. Gnathopod 2. C. Labrum. D.Labium. E. Gnathopod 1. F. Gnathopod 1

palm (enlarged). G. Mandible. H. Mandible palp article 3 (enlarged). I. Maxilla 1.

J, M, N. Maxilliped. K. Mandible—spine row and lacinia mobilis (enlarged). L. Maxilla 2.

274 ANNALS OF THE SOUTH AFRICAN MUSEUM

LW)

{,]

WN:

eS HY fa

— SS

Fig. 5. Lemboides acanthiger K. H. Barnard, male, 6,0 mm, Natal. A. Gnathopod 1 palm

(enlarged). B. Gnathopod 1. C. Gnathopod 2. D. Gnathopod 2 palm (enlarged).

E. Uropod 1. F. Uropod 2. G. Uropod 3. H. Telson.

SOUTH AFRICAN SPECIES OF LEMBOIDES AND LEMBOS 275

longest, terminally falcate. Coxae 1—4 shallow, coxa 1 elongate, broader than

deep, produced forward, subacute. Gnathopod 1 ¢d basis robust, carpus and

propodus subequal in length, propodus palm irregular, defined by strong acute

tooth, dactylus very strongly overlapping palm. Gnathopod 1 & slender, pro-

podus a little longer than carpus, palm weakly sinuous and crenulate, dactylus

elongate, overlapping palm. Gnathopod 2 ¢d carpus and propodus subequal in

length but propodus broader, palm weakly concave, crenulate, dactylus overlap-

ping palm. Gnathopod 2 ¢ similar to that of d but basis more slender, palm

regular. Pereopods 5—7 missing in all material examined. Epimera 1—2 rounded,

epimeron 3 with small posterodistal tooth. Uropods 1-2 with strong interramal

tooth. Uropod 1 peduncle a little shorter than subequal rami. Uropod 2 peduncle

very short, rami more than twice length of peduncle, telsonic crests each with a

pair of setae.

Remarks

Lemboides acanthiger and L. afer together resemble Atlanto—East Pacific

Microdeutopus (raised to generic level in Myers in press) in the very short

peduncle and spinose rami of uropod 3. By contrast, East Atlantic-Mediterra-

nean Microdeutopus have elongate uropod 3 peduncles, whilst Indo-Pacific

Lembos sensu lato have elongate setae on the rami. For the present, the genus

Lemboides is retained for these two species.

Distribution

Southern African endemic.

Lemboides afer Stebbing, 1895

Figs 6-7

Lemboides afer Stebbing, 1895: 209, pls 9-10. K. H. Barnard, 1932: 222, fig. 137. Griffiths,

1976: 34, fig. 18B.

Type locality

Cape of Good Hope.

Material examined

SAM-A19399, 3 3, 2 2, off Oatland Point, 34°12’S 18°29'E, 33-36 m,

5 March 1952, UCT FAL31H. SAM-A19400, 1d, 1 2, 34°19,3’S 18°32,3’E,

58 m, 8 September 1956, UCT TRA111P. SAM-—A18948, 20 3, 26 2, Skrywers-

hoek, Langebaan Lagoon, LWS and below, 26 April 1949, UCT LB191S.

Description

Head with eye lobes moderately produced, eye small. Antennae weakly

setiferous; antenna 1 short, about one-third body length, antenna 2 subpediform,

shorter than antenna 1. Labium mandibular lobes acute. Mandibular palp

ANNALS OF THE SOUTH AFRICAN MUSEUM

276

AI\\

—_~

59}

=e)

fae}

Se 4

~~.

D. Maxilla 1.

H. Gnathopod 1 palm (enlarged). I. Mandible. J. Mandible palp article 3

Lemboides afer Stebbing, female,

C. Maxilla 2.

(enlarged). K. Mandible—spine row and lacinia mobilis (enlarged). L—N. Maxilliped.

Fig. 6

B. Labrum.

G. Gnathopod 1.

SOUTH AFRICAN SPECIES OF LEMBOIDES AND LEMBOS DT

articles 2 and 3 subequal in length, article 3 with distal part of anterior margin

weakly concave. Coxae 1-4 shallow; coxal elongate, broader than deep,

rounded. Gnathopod 1 6 basis robust, carpus enormous with posterodistal

margin weakly produced into irregular lobe, propodus short, a little over one-half

length of carpus, palm defined by two irregularly triangular processes, dactylus

short. Gnathopod 1 2 carpus only a little longer than propodus, propodus with

crenulated palm. Gnathopod 2 d basis with anterior margin produced into a

crenulated flange, carpus enlarged, anterior margin with long setae, propodus

short, a little over one-half length of carpus, palm with V-shaped crenulate-sided

excavation defined by a triangular tooth bearing a spine, dactylus short.

Gnathopod 2 @ basis elongate, anterior margin with long setae, basis swollen

medially, anterior margin with very long sieve-setae, propodus elongate and

slender, a little longer than carpus, palm short, irregular, defined by two spines,

dactylus short, fitting palm. Pereopod 7 distinctly longer than pereopod 6.

Epimera 1-3 rounded. Uropods 1-2 with interramal peduncular tooth. Uro-

pod 1 peduncle and rami subequal. Uropod 2 inner ramus longer than outer

We SS EEE

YY SS

IISKE

Fig. 7. Lemboides afer Stebbing, male, 6,0 mm, Oatland Point. A. Uropod 1. B. Uropod 2.

C. Telson. D. Uropod 3. E. Gnathopod 1. F. Gnathopod 1 palm (enlarged).

G. Gnathopod 2. H. Gnathopod 2 palm (enlarged).

278 ANNALS OF THE SOUTH AFRICAN MUSEUM

ramus and longer than peduncle. Uropod 3 peduncle very short, inner ramus

longer than outer and more than twice length of peduncle. Telsonic crests each

with 4—5 setae.

Remarks

For relationships, see comments under previous species.

Distribution

Southern African endemic.

Lembos hypacanthus K. H. Barnard, 1916

Figs 1C, 8-9

Lembos hypacanthus K. H. Barnard, 1916: 237, pl. 28 (figs 5—6). Griffiths, 1976: 34, fig. 18D.

Lembos sp. Fox, 1978: 162.

Type locality

Sea Point, near Cape Town.

Material examined

SAM-A18951, 1 5, 1 2, Dalebrook (False Bay), intertidal, 14 September

1955, UCT CP439E. 67 3, 59 2, 21 immature, Brevard C, Florida. 13 6, 24 9,

North Carolina.

Description

Head with eye lobes moderately produced; eye of medium size. Antennae

weakly setiferous; antenna 1 about two-thirds body length, antenna 2 sub-

pediform, shorter than antenna 1. Labium mandibular lobes acute. Mandibular

palp article 3 longer than 2, distinctly falcate. Coxae 1—4 shallow; coxa 1 antero-

distal margin produced, acute. Gnathopod 1 ¢ basis robust, inner face with a

proximal tooth, carpus and propodus subequal in length and breadth, anterior

margin densely setose, propodus with a short crenulate palm, followed by a deep

excavation posterior to which is a slender tooth and a defining hump bearing a

spine, dactylus short. Gnathopod 1 2 propodus slightly longer than carpus, palm

irregularly rounded, defined by a spine. Gnathopod 2 d basis with falcate

anterodistal tooth, carpus and propodus elongate, especially so in American

material, anterior margin of both podomeres densely setose, dactylus short but

overlapping short palm. Gnathopod 2 2 similar to that of d but a little less

setose. Pereopod 7 distinctly longer than pereopod 6. Epimera 1-2 rounded.

Epimeron 3 with small posterodistal tooth. Uropod 1 peduncle and inner ramus

subequal, inner ramus longer than outer. Uropod 2 peduncle shorter than inner

ramus, inner ramus longer than outer. Uropod 3 peduncle unexpanded, rami

subequal, a little longer than peduncle. Telsonic crests each with a pair of unequal

length setae.

SOUTH AFRICAN SPECIES OF LEMBOIDES AND LEMBOS 279

Fig. 8. Lembos hypacanthus K. H. Barnard, male, 6,0 mm, False Bay. A. Gnathopod 1 palm

(enlarged). B.Gnathopod 1. C. Labium. D. Gnathopod 2 palm (enlarged). E. Labrum.

F. Gnathopod 2. G-—I. Maxilliped. J. Maxilla1. K. Mandible palp article 3 (enlarged).

L. Mandible—spine row and lacinia mobilis (enlarged). M. Mandible.

ANNALS OF THE SOUTH AFRICAN MUSEUM

280

False Bay. A. Uropod 1.

—tip of inner ramus.

B. Gnathopod 1 palm (enlarged). C. Gnathopod 1. D. Uropod. E. Telson. F. Gnathopod 2

I. Uropod 3

G. Gnathopod 2. H. Uropod 3

J—L. Male, 6,0 mm, North Carolina. J. Gnathopod 1. K. Head. L. Gnathopod 2.

Fig. 9. Lembos hypacanthus K. H. Barnard. A-I. Female, 5,5 mm

palm (enlarged).

SOUTH AFRICAN SPECIES OF LEMBOIDES AND LEMBOS 281

Remarks

The maxilliped flanges and form of the male gnathopod 1 clearly place this

species close to L. websteri, the type-species of the genus.

Distribution

Originally thought to be a South African endemic, this species is now known

to occur also in the West Atlantic.

DISCUSSION

The present study has revealed a close biogeographical relationship between

the Lemboides and Lembos species of southern Africa and those of the Atlanto—

East Pacific. Lemboides sensu stricto has its closest relationships with Atlanto—

East Pacific Microdeutopus, and Lembos hypacanthus is now shown to be

transatlantic. Lembos websteri, the sister species closest to L. hypacanthus, is also

Known from both eastern North America and the north-eastern Atlantic—

Mediterranean. These distributions may be best explained as relictual, resulting

from the opening-up of the Atlantic by tectonic activity in the early Tertiary. This

contrasts with what is known of the distribution of the genus Aora, which is aiso

present in southern Africa and appears to have Gondwanian or transantarctic

affinities.

ACKNOWLEDGEMENTS

Thanks are due to Dr C. L. Griffiths, Zoology Department, University of

Cape Town, Mrs M. G. van der Merwe, South African Museum, and Dr R. Fox,

Lander College, Greenwood, South Carolina, for generous loans of material

upon which this study was based.

REFERENCES

BARNARD, K. H. 1916. Contributions to the crustacean fauna of South Africa. 5. The

Amphipoda. Annals of the South African Museum 15 (3): 105-302.

BARNARD, K. H. 1932. Amphipoda. Discovery Report 5: 1-326.

Fox, R. S. 1978. Order Amphipoda. Jn: ZincMaRK, R. G. ed. An annotated checklist of the

biota of the coastal zone of South Carolina: 161-166. Columbia: University of South

Carolina Press.

GrirFiTrHs, C. L. 1974a. The Amphipoda of southern Africa. 2. The Gammaridea and

Caprellidea of South West Africa south of 20°S. Annals of the South African Museum

62 (6): 169-208.

GrirFirHs, C. L. 1974b. The Amphipoda of southern Africa. 3. The Gammaridea and

Caprellidea of Natal. Annals of the South African Museum 62 (7): 209-264.

GrirFiTHs, C. L. 1974c. The Amphipoda of southern Africa. 4. The Gammaridea and

Caprellidea of the Cape Province east of Cape Agulhas. Annals of the South African

Museum 65 (9): 251-336.

GrirFirHs, C. L. 1975. The Amphipoda of southern Africa. 5. The Gammaridea and

Caprellidea of the Cape Province west of Cape Agulhas. Annals of the South African

Museum 67 (5): 91-181.

282 ANNALS OF THE SOUTH AFRICAN MUSEUM

GriFFITHS, C. L. 1976. Guide to the benthic marine amphipods of southern Africa. Cape Town:

South African Museum.

Myers, A. A. in press. A cladistic and biogeographic analysis of the Aorinae sub-family nov.

Sixth International Colloquium on Amphipoda, 1985. Crustaceana suppl.

STEBBING, T. R. R. 1895. Notes on Amphipoda old and new. Annals and Magazine of Natural

History (6) 16 (30): 205-213.

6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu-

larly Articles 22 and 51).

Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed

by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov.,

ete:

An author’s name when cited must follow the name of the taxon without intervening punctuation

and not be abbreviated; if the year is added, a comma must separate author’s name and year. The

author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is trans-

ferred from its original genus. The name of a subsequent user of a scientific name must be separated

from the scientific name by a colon.

Synonymy arrangement should be according to chronology of names, i.e. all published scientific

names by which the species previously has been designated are listed in chronological order, with all

references to that name following in chronological order, e.g.:

Family Nuculanidae

Nuculana (Lembulus) bicuspidata (Gould, 1845)

Figs 14-15A

Nucula (Leda) bicuspidata Gould, 1845: 37.

Leda plicifera A. Adams, 1856: 50.

Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b).

Nucula largillierti Philippi, 1861: 87.

Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9.

Note punctuation in the above example:

comma separates author’s name and year

semicolon separates more than one reference by the same author

full stop separates references by different authors

figures of plates are enclosed in parentheses to distinguish them from text-figures

dash, not comma, separates consecutive numbers.

Synonymy arrangement according to chronology of bibliographic references, whereby the year is

placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable.

In describing new species, one specimen must be designated as the holotype; other specimens

mentioned in the original description are to be designated paratypes; additional material not regarded

as paratypes should be listed separately. The complete data (registration number, depository, descrip-

tion of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.:

Holotype

SAM-A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach, Port Eliza-

beth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973.

Note standard form of writing South African Museum registration numbers and date.

7. SPECIAL HOUSE RULES

Capital initial letters

(a) The Figures, Maps and Tables of the paper when referred to in the text

e.g. ‘. .. the Figure depicting C. namacolus ...’: ‘. . . in C. namacolus (Fig. 10)...’

(b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by

initials or full names

e.g. Du Toit but A.L. du Toit; Von Huene but F. von Huene

e.g. Therocephalia, but therocephalian

Punctuation should be loose, omitting all not strictly necessary

Reference to the author should preferably be expressed in the third person

Roman numerals should be converted to arabic, except when forming part of the title of a book or

article, such as

‘Revision of the Crustacea. Part VIII. The Amphipoda.’

Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial

capital letter, provided the same generic name is used consecutively. The generic name should

not be abbreviated at the beginning of a sentence or paragraph.

Name of new genus or species is not to be included in the title; it should be included in the abstract,

counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts.

A. A. MYERS

J. LYONS

A RE-EVALUATION

OF THE SOUTH AFRICAN SPECIES OF

LEMBOIDES STEBBING AND LEMBOS BATE

(AMPHIPODA, AORIDAE)

DESCRIBED BY K. H. BARNARD (1916)

cat

a JME 97 PART 10. ~—- OCTOBER 1987 ISSN 0303-2515

S67X

p NH

OU, See for ete tie

jes Ud aera HY? y

wet o Z, Py *

wey )

JAN &1 1988

ag, Lippany Li Hel Se

OF THE SOUTH AFRICAN —

SEUM —

CAPE TOWN

INSTRUCTIONS TO AUTHORS

|. MATERIAL should be original and not published elsewhere, in whole or in part.

. LAYOUT should be as follows:

(a) Centred masthead to consist of

Title: informative but concise, without abbreviations and not including the names of new genera or species

Author’s(s’) name(s)

Address(es) of author(s) (institution where work was carried out)

Number of illustrations (figures, enumerated maps and tables, in this order)

(b) Abstract of not more than 200 words, intelligible to the reader without reference to the text

(d) Introduction

(e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents

(f) Summary, if paper is lengthy

(g) Acknowledgements

(th) References

(i) Abbreviations, where these are numerous

3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced

with 2,5 cm margins all round. First lines of paragraphs should be indented. Tables and a list of

legends for illustrations should be typed separately, their positions indicated in the text. All

pages should be numbered consecutively.

Major headings of the paper are centred capitals; first subheadings are shouldered small

capitals; second subheadings are shouldered italics; third subheadings are indented, shouldered

italics. Further subdivisions should be avoided, as also enumeration (never roman numerals)

of headings and abbreviations.

Footnotes should be avoided unless they are short and essential.

Only generic and specific names should be underlined to indicate italics; all other marking

up should be left to editor and publisher.

4. ILLUSTRATIONS should be reducible to a size not exceeding 12 = 18 cm (19 cm including

legend); the reduction or enlargement required should be indicated; originals larger than

35 x 47 cm should not be submitted; photographs should be rectangular in shape and final

size. A metric scale should appear with all illustrations, otherwise magnification or reduction

should be given in the legend; if the latter, then the final reduction or enlargement should be

taken into consideration.

All illustrations, whether line drawings or photographs, should be termed figures (plates

are not printed; half-tones will appear in their proper place in the text) and numbered in a

single series. Items of composite figures should be designated by capital letters; lettering of

figures is not set in type and should be in lower-case letters.

The number of the figure should be lightly marked in pencil on the back of each illustration.

5. REFERENCES cited in text and synonymies should all be included in the list at the end of

the paper, using the Harvard System (ibid., idem, ioc. cit., op. cit. are not acceptable):

(a) Author’s name and year of publication given in text, e.g.:

‘Smith (1969) describes...’

‘Smith (1969: 36, fig. 16) describes. . .’

‘As described (Smith 1969a, 1969b; Jones 1971)’

“As described (Haughton & Broom 1927)...’

‘As described (Haughton et al. 1927)...’

Note: no comma separating name and year

Dagination indicated by colon, not p.

names of joint authors connected by ampersand

et al. in text for more than two joint authors, but names of all authors given in list of references.

(b) Full references at the end of the paper, arranged alphabetically by names, chronologically

within each name, with suffixes a, b, etc. to the year for more than one paper by the same

author in that year, e.g. Smith (1969a, 19695) and not Smith (1969, 1969a).

For books give title in italics, edition, volume number, place of publication, publisher.

For journal article give title of article, title of journal in italics (abbreviated according to the World list o,

Scientific periodicals. 4th ed. London: Butterworths, 1963), series in parentheses, volume number, part

number (only if independently paged) in parentheses, pagination (first and last pages of article).

Examples (note capitalization and punctuation)

BULLOUGH, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan.

FISCHER, P.—H. 1948. Données sur la résistance et de le vitalité des mollusques. J. Conch., Paris 88: 100-140.

FIscHER, P.-H., DuvAL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archs

Zool. exp. gén. 74: 627-634.

Konn, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon.

Ann. Mag. nat. Hist. (13) 2: 309-320.

Koun, A. J. 19606. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean.

Bull. Bingham oceanogr. Coll. 17 (4): 1-51.

THIELE, J. 1910. Mollusca: B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische

und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-Afrika 4: 269-270.

Jena: Fischer. Denkschr. med.-naturw. Ges. Jena 16: 269-270.

(continued inside back cover)

ANNALS OF THE SOUTH AFRICAN MUSEUM

ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

Volume 97 Band

October 1987 Oktober

Part 10 #£zDeel

MIOCENE SUIDAE FROM ARRISDRIFT,

SOUTH WEST AFRICA-NAMIBIA

MARTIN PICKFORD

Cape Town Kaapstad

The ANNALS OF THE SOUTH AFRICAN MUSEUM

are issued in parts at irregular intervals as material

becomes available

Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000

Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

word uitgegee in dele op ongereelde tye na gelang van die

beskikbaarheid van stof

Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000

OUT OF PRINT/UIT DRUK

il, WES, 59), SOW, 45, B, tan.) SCES, 5, 7),

AG, tost)), IO), 8, O02, 7), U0GES),

GE, Sy 1, i sjoae), IED), IXCLS), DO), Wi, SEB, 2216), 23, 35@), SM)

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 089 3

Printed in South Africa by In Suid-Afrika gedruk deur

The Rustica’ Press, Pty:, td: Die Rustica-pers, Edms., Bpk.,

Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap

MIOCENE SUIDAE FROM ARRISDRIFT, SOUTH WEST AFRICA-—

NAMIBIA

MARTIN PICKFORD

Institut de Paléontologie, 75005, Paris, France

(With 3 figs and 3 tables)

[MS accepted 30 April 1987]

ABSTRACT

Two taxa of small suids occur at the early middle Miocene locality of Arrisdrift, South West

Africa—Namibia. Of these, the more abundantly represented is Lopholistriodon moruoroti, a

small listriodont previously known from equatorial sites in Kenya. The second, poorly

represented species is a small pig with bunodont dentition, possibly related to the subfamily

Tetraconodontinae. The relationships of the two species and their implications for biostrati-

graphy and palaeozoogeography are explored, and it is concluded that the Arrisdrift strata post-

date the first of the Neogene Dispersal Phases of Thomas, dated about 17,5 m.y. ago, but that

they are unlikely to be as young as 14 m.y. old.

CONTENTS

PAGE

GME OYENUKCVOLI LS Sh Sis aes’ Sos ole co Ete NS Gynt GE ne et ney a 283

Systematic @escnipulomigans sacs. vey ce ss ene eae yee et ee 284

ID ISCUSSIO Merete te ey ote Nereis Gite Bens ets Saas « 293

PNCKMOWIECSEMIETIUS Hi anaes fe alt han Rete oaks mand Gad y se oa 294

TRRENSRET CESS. 6 sy oie sore, deo Gee aed, Cette i eer on ee oe ee 295

INTRODUCTION

Arrisdrift, a locality near the Orange River in South West Africa—Namibia,

is the only known middle Miocene African fossil locality south of the Nyanza Rift

Valley sites of Maboko and Nyakach, Kenya, which occur a few kilometres south

of the equator (Pickford 1984). It is evidently somewhat younger, on the basis of

its faunal content, than other fossiliferous strata in South West Africa—Namibia

(Stromer 1926) at Langental, Elizabethfeld, Bogenfels and Fiskus (Hendey

1984). The distance between the equatorial Kenyan sites and Arrisdrift is more

than 3 000 km; yet some of the fossil Suidae from the two regions appear, on the

basis of available evidence, to be comparable, which suggests that the two regions

were possibly biostratigraphically and ecologically similar.

Corvinus & Hendey (1978) and Hendey (1978, 1984) suggested, on the basis

of the aspect of the entire mammalian fauna, that Arrisdrift may be about

16-15 m.y. old, an estimate that seems reasonable. The ruminants and hyracoids

indicate an age slightly younger than Maboko and Buluk, Kenya. The latter site

has been radiometrically dated to be about 17,2 m.y. (MacDougall & Watkins

1985), while the former is known to be older than 13 m.y. (Bishop et al. 1969).

283

Ann. S. Afr. Mus. 97 (10), 1987: 283-295, 3 figs, 3 tables.

284 ANNALS OF THE SOUTH AFRICAN MUSEUM

Newly discovered strata at Nachola, northern Kenya, have yielded a comparable

fauna, dated about 15,5 m.y. old (work in progress). The Arrisdrift faunas are

probably older than those from Fort Ternan, Kenya, dated to be about 14 m.y.

(Bishop ef al. 1969) but possibly somewhat younger (work in progress).

There can be little doubt, therefore, that the Arrisdrift fauna is, in a broad

sense, early middle Miocene in age, 1.e. somewhere between 17,2 and 14 m.y.

old. Its importance is very great, providing, as it does, the only glimpse of a

middle Miocene terrestrial fauna for virtually the entire subequatorial expanse of

Africa. The zoogeographic and biostratigraphic implications of this important site

cannot be overstressed. This detailed systematic description of the suid remains

from Arrisdrift will hopefully go some way towards augmenting the already

valuable data and preliminary interpretations of the mammal fauna of the site.

The fossil suid material includes cranio-dental fragments of a small

lophodont species, teeth and mandible fragments of an unidentified bunodont

suid, and postcranial bones that, on the basis of their size, could belong to either

of the two forms. The material is in the collections of the South African Museum

and bears the catalogue prefix SAM-—PQ, which is omitted from the accession

numbers given in the text.

SYSTEMATIC DESCRIPTION

Family Suidae Gray, 1821

Subfamily Listriodontinae Simpson, 1945

Genus Lopholistriodon Pickford & Wilkinson, 1975

Type-species: Lopholistriodon kidogosana Pickford & Wilkinson, 1975.

Diagnosis

A genus that differs from other genera of the subfamily by its small size and

the extreme development of the transverse crests in the molars and fourth

premolars, with the suppression of accessory cusps. The upper premolars possess

enlarged cingula and wide cingular platforms. The nasal ridge is narrow. (After

Pickford 1986.)

Lopholistriodon moruoroti Wilkinson, 1976

Figs 1, 2A—C, 3D

Listriodon sp. Harris & Watkins, 1974: 576-577.

Lopholistriodon sp. D Pickford & Wilkinson, 1975: 133.

Lopholistriodon moruoroti Wilkinson, 1976: 242-245, pl. 9 (fig. B). Hendey, 1978: 23. Pickford,

1986: 56-58.

Diagnosis

A small species of Lopholistriodon in which the diastemata are rather short,

P* immediately behind C, P? separated from P’*. Hypoconid crests better

developed than in L. kidogosana.

MIOCENE SUIDAE FROM ARRISDRIFT 285

Holotype

KNM-MO 5, mandible with roots of left I,_3, complete C,, P3_4,, and M,_3;

right I,_5, root I, complete C,, P,-M; (Wilkinson 1976, pl. 9 (fig. B)), housed in

the Kenya National Museum, Nairobi, Kenya.

Horizon

Early middle Miocene, possibly about 16 million years.

Type locality

Moruorot, Kenya.

Distribution

Moruorot, Buluk and Maboko, Kenya; Arrisdrift, South West Africa—

Namibia.

Material

AD49: upper canine; AD135: left M;; AD136: left maxilla with P*—M?;

AD138+316+317: snout with left and right I’, left I, left and right C, and roots

of left P’ and right I°; AD636: left M3; AD769: lower molar fragment; AD990:

upper canine; AD1727: right I,; AD1753: left M*; AD2196: right M; broken, in

mandible fragment; AD2411: left M'; AD2535: mesial part of right dM;

AD2565: left M’ and broken M*; AD2658: left dM, in mandible fragment;

AD2692: distal portion of right M3; AD2927: right M; and distal part of M, in

mandible fragment; AD3014: right upper canine; AD3015: right I,; AD3276:

distal portion of right dMg,.

Description

Fossil listriodont material from Arrisdrift consists of at least 21 cranio-dental

fragments, which can confidently be assigned to a small lophodont pig species that

was identified by Hendey (1978) as Lopholistriodon moruoroti. In addition there

are five postcranial fragments that may belong to this species (see separate section

in which the postcranial elements are described).

The Arrisdrift specimens include a snout, hitherto not represented among

previously described material; the bulk of the sample duplicates existing

information.

Snout. AD138 is the only known snout of this species; it can be compared

with that of Lopholistriodon kidogosana from Ngorora, Kenya. The Arrisdrift

specimen is crushed and warped (Fig. 1!A—D) but details of its anatomy can be

discerned reasonably satisfactorily. The nasal bones are broken anteriorly but are

wide and stout. The premaxilla is slightly spatulate in palatal view, the area in

front of the I°—C diastema being wider than the width at the diastema. The roots

of the central incisors are widely separated although the mesial edges of the

crowns met mesially in life. The roots of I’~* are close together.

286 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 1. Lopholistriodon moruoroti. A-D. SAM—PQ-—AD138: snout with left and right I’,

left I°, both canines, and roots of left P!. A. Occlusal view. B. Dorsal view. C. Anterior

view. D. Left lateral view. E-F. SAM-PQ-AD136: left maxilla with P*-M?. E. Lingual view.

F. Occlusal view. G-H. SAM-PQ-AD2927: right mandible fragment with part of M, and

complete M3. G. Occlusal view. H. Lingual view. Scales =1 cm.

MIOCENE SUIDAE FROM ARRISDRIFT 287

TABLE 1

Dimensions (in mm) of the teeth of Lopholistriodon moruoroti.

Teeth Specimen Length Breadth

Upper dentition

Left I! AD316 10,5 4,6

Right AD316 12,0 4,8

Left I? AD317 5) 3,6

eit € AD138 10,0 7,8

Right C AD3014 7,8 5,6

Right C AD138 10,3 UP

Right C AD49 8,7 7,9

Left P* AD136 S55) 8,4

Left P* Mor 6,8 7,9

Left M! AD136 9) 5) 9,1

Left M! AD2411 9,9 9,3

Left M! AD2565 9,8 Oeil

Left M! Mor 9+ 8,8

Left M? AD2565 — 11,0

Left M? AD136 10,7 10,5

Left M? AD1753 10,9 10,2

Left M? Mor 10,5 10,4

Left M? AD136 11,3 10,0

Left M? Mor 11,0 10,2

Lower dentition

Right Ip AD1727 4,3 3,2

Right I, AD3015 — 4,2

Left M3 AD636 12,5 8,1

Left M3 AD135 14,0 8,8

Right M3 AD2927 15,0 8,2

Right M3 AD2196 14,0 —

Right M3 Mor 14,5 8,0

Left dM, AD2658 12,1 5,

Right dM4 AD3276 — 4,0

The premaxillary diastema that housed the lower canine during occlusion has a

sharp-edged dorsal ridge, which reaches upwards and laterally above the upper

canine to form a supracanine flange, as in L. kidogosana.

The upper canines point antero-laterally and slightly downwards, suggesting

the possibility that this individual was a female. The palate between the two

canines is rather broken but seems originally to have been quite flat. The P’ roots

are close to the upper canine and there is a substantial ridge forming the alveolar

process for P’. Apart from size, the only major difference from L. kidogosana is

the presence of P' roots close to the upper canines.

The upper dentition. 1’ is well preserved in the snout (AD138). It is

mesiodistally elongated with a large lingual cingulum; a lingual fossette above the

cingulum is wide and quite deep. The occlusal edge is worn, exposing dentine,

288 ANNALS OF THE SOUTH AFRICAN MUSEUM

and the pattern of dentine exposure and the wavy labial surface of the enamel

suggest that this tooth, when unworn, had terminal pectinations as in Lopho-

listriodon kidogosana and Listriodon splendens.

I' roots are long and housed in prominent juga, which form the lower lateral

parts of the nasal aperture. The roots of I*-°, in contrast, are very small. I* is a

very small tooth, with a lingual cingulum, a triangular crown in lingual view, with

its main wear facet anteriorly situated. The incisive foramina are large and

situated in line with the I* roots, far forward in the palate.

The canines are closed-rooted, tusk-like teeth tapering from the cervix both

crownwards and rootwards. A single large anterior facet is worn by occlusion with

the lower canine (Fig. 3D). This facet is almost vertically oriented, and its

palato-dorsal dimension is greater than its mesiodistal width.

Maxilla and posterior upper dentition. AD136 is a left maxilla with P*—M?

and part of the zygomatic process of the maxilla (Fig. 1E—F). The distal surface of

the zygomatic root is opposite M°* and there is a small projection of maxilla

behind M?. The greater palatine foramen is close to the edge of the palate on a

level with the front of M?.

P* is a markedly lophodont tooth, surrounded anteriorly, buccally and .

distally by a cingulum. The disto-buccal cusp is greatly reduced, appearing as a

slight cusplet on the disto-buccal crest. The molars are all strongly bilophodont,

but there are ridges leading into the median valley from both lophs, forming

characteristic low antero-posterior crests in the midline of the teeth. All the upper

molars bear peripheral cingula.

Mandibles and lower dentition. AD2658 is a juvenile mandible with left dM,

and part of the symphysis up to the midline (Fig. 2A—B), which indicates that this

species had a spatulate symphysis as in the genus Listriodon. The dM, is

comprised of three lophs, the distal loph being the widest. Antero-posterior crests

run into the median and distal transverse valleys of the tooth.

AD2927 (Fig. 1G—-H) and 2196 are small mandibular fragments each

containing M3. The mandibles are gracile, the roots of M3 are fused bucco-

lingually, and the ascending ramus is well behind M; as in Listriodon species, in

contrast to most other suids, where it hides the rear of M; in lateral view.

The lower third molars are comprised of two main lophs, behind which is a

well-developed, centrally positioned talonid. The anterior lophs are extremely

lophodont, with most of the grooves and crests suppressed, but the median

accessory cusp is joined by a well-developed crest to the hypoconid. In the latter

respect, this tooth differs from that of L. kidogosana, in which the homologous

crest is almost completely suppressed. The median valley is deep and U-shaped,

and is divided into two portions separated by the median accessory cusp. The

talonid is comprised of a single centrally positioned cusp, which is probably an

enlarged hypoconulid. On either side of it, mesially and laterally, it has low

cingula.

In Mp, which is poorly represented in the Arrisdrift collection (Fig. 1H), the

hypoconulid is small and positioned close behind the distal loph.

MIOCENE SUIDAE FROM ARRISDRIFT 289

Fig. 2. A-C. Lopholistriodon moruoroti. A-B. SAM—PQ-AD2658: juvenile left mandible

(with mirror image) containing dM,. A. Occlusal view. B. Lingual view. C. SAM—P—AD3015:

right I,, lingual view.

D-G. Gen. et sp. indet. D-E. SAM—PQ-AD 1697: juvenile right mandible (with mirror image)

containing dI,-I,. D. Occlusal view. E. Lingual view. F. SAM—PQ-AD102d: left I,, occlusal

view. G. SAM-—PQ-AD20d: lower central incisor. Scales = 1 cm.

290 ANNALS OF THE SOUTH AFRICAN MUSEUM

The I, (AD1727) has a distal scoop typical of listriodonts, bordered centrally

by a lingual ridge. It has a very light lingual cingulum, but its crown is appreciably

longer from cervix to tip than are the corresponding crowns in L. kidogosana or

species of Listriodon (see Pickford 1986).

Remarks

The listriodont fossils from Arrisdrift, Maboko and Moruorot are morpho-

logically and metrically very similar. There can be little doubt that they should be

classified in the same species. What is perhaps more difficult to understand is the

phyletic position of Lopholistriodon moruoroti. It is one of the most derived

listriodonts in terms of the perfection of its lophodonty, yet it seems to be one of

the earliest known members of the subfamily. It does possess some primitive

features, including the retention of a two-rooted P' situated close to the canine,

and a short diastema. Its precursors are unknown, and there seems little

possibility that it was derived from any of the early Miocene East African suids,

which are now quite well known (Pickford 1986). The alternative is that they were

derived from a small Eurasian precursor—perhaps one of the Palaeochoerus-like

suids of the early Miocene of Europe—and that they migrated into Africa about

17,5 m.y. ago at the time when many African taxa were translocating to Europe

and Asia (Thomas 1985). However, there are no obvious links in the fossil record.

Whatever the case may be, it seems that Lopholistriodon was an immigrant

to East Africa and South West Africa—Namibia, since there is no sign of

listriodont ancestry in the early Miocene deposits of Africa. The presence of

Lopholistriodon in a fossil fauna is therefore probably good evidence that one is

dealing with strata less than 17,5 m.y. old, i.e. middle Miocene rather than early

Miocene.

Lopholistriodon kidogosana is known from upper middle Miocene sites in

East Africa—such as Ngorora, Kenya—the youngest specimens being about

11 m.y. old. The genus has not yet been found in the same deposits as Hipparion,

suggesting that, like the genus Listriodon, its range was entirely within the middle

Miocene period. As such it is a useful genus for broad _ biostratigraphic

correlations.

Family Suidae: gen. et sp. indet.

Figs 2D-—G, 3A—C

Material

Sce- Mable 2:

Description

Right M* (AD1795) has marked lingual and buccal flare, and a simple talon,

which is little more than a distal cingulum (Fig. 3A). It has a fairly large anterior

cingulum and anterior accessory cusp; its main cusps are close together and

possess subdued wrinkling, suggestive of the presence of thick enamel.

MIOCENE SUIDAE FROM ARRISDRIFT 291

TABLE 2

Dimensions (in mm) of bunodont suid teeth (gen. et sp. indet.) from Arrisdrift.

Dentition Specimen Length Breadth

Right M? AD1795 16,0 13,1

P* fragment AD2821 — 9,0

Left I; AD1693 oats 4.8

I AD2355 — —

Right I; AD20d 4,5 4,7

Left I AD102d 4,7 5,0

Left mandible with half dMaM: (M1) AD631 ili 8,9

Right mandible with dIi2 (dh) AD1697 Spill 2,8)

(dI2) 3,4 2,9

Fig. 3. A-C. Gen. et sp. indet. A. SAM-—PQ-AD1795: right M3, occlusal view.

B-C. SAM-PQ-AD631: left mandible with half dM,M, (or half M,-M,). B. Buccal view.

C. Occlusal view.

D. Lopholistriodon moruoroti. SAM-—PQ-AD49: right upper canine.

Scales = 1 cm.

292 ANNALS OF THE SOUTH AFRICAN MUSEUM

AD631 (Fig. 3B—C), a juvenile left mandible with a fragment of M, and a

complete M, (or possibly dM,—M,), is comparable to Nguruwe kijivium in the

morphology of its molar crown, although the tooth is higher crowned than is

typical of Nguruwe.

AD1697 (Fig. 2D-E) is an infant mandible with dI,_,, and alveoli of

dI;—dM;. The symphysis, which is complete to the midline, is narrow anteriorly,

contrasting with that of Lopholistriodon, which is markedly spatulate. The

deciduous incisors in the mandible are typically suine, being narrow with concave

lingual surfaces bordered by low anterior and posterior ridges. The tip of dl, is

beaded or crenulated.

AD20d is a rootless, unworn lower central incisor (Fig. 2G), which is

pectinate at the tip and which has a central ridge running from crown tip to cervix,

as well as mesial and distal lingual ridges.

AD102d is a left I, (Fig. 2F), lightly worn with a characteristic distal scoop,

lingual rib, and mesial and distal cingular ribs on the lingual side of the crown.

The root is long, and in lingual aspect the tooth is gently curved from crown to

root, the cervix being more distal than either the root or the crown tips.

Remarks

The eight specimens tentatively assigned to this indeterminate bunodont suid

(Table 2) comprise a sample that is insufficient for the purpose of unequivocal

identification, bunodonty being a primitive and widespread characteristic of the

family. However, the great degree of flare in the upper molars and the lack of

buccal and lingual cingula indicate that it is probably not a kubanochoerine, and

its affinities possibly lie with genera such as Palaeochoerus or Conohyus, both of

which are well known in Miocene deposits of Europe. More substantial material

is required before a positive identification can be made.

POSTCRANIAL BONES

Material

See Table 3.

Remarks

On the basis of their overall sizes, the five postcranial bones from Arrisdrift

could belong either to Lopholistriodon moruoroti or to the bunodont suid. A

decision cannot confidently be made at this time, although—on the basis of

quantities of specimens—the chances seem to be higher that they belong to the

former rather than the latter. Whatever the final outcome, the available

postcranial elements indicate a morphological pattern close to extant suids, from

which it is suggested that in its postcranium, the Arrisdrift suid to which these

bones belonged was typically suine. Although it was appreciably smaller than any

living African suid, the Arrisdrift species was not smaller than the pygmy hog of

the Himalaya foothills (Sus salvinia).

MIOCENE SUIDAE FROM ARRISDRIFT 293

TABLE 3

Postcranial bones of suid (gen. et sp. indet.) from Arrisdrift (dimensions in mm).

Bone Specimen Size

Proximal end left femur AD3044

antero-posterior head diameter 1555

distance from head to greater trochanter S8)53)

Complete right tibia AD2696

total length LEZ

height of proximal end Ppl

breadth of proximal end 24,5

height of distal end JUL)

breadth of distal end 15,6

Distal end of right tibia AD646

height of distal end 11,6

breadth of distal end 15,0

Distal end of metapodial AD878

height of distal end 10,5

breadth of distal end 9,5

Distal end of metapodial AD2711

height of distal end 3)

breadth of distal end 8,9

DISCUSSION

The presence of two small suid species at Arrisdrift, in early middle Miocene

deposits, is interesting from the point of view of palaeozoogeography, because

both may represent immigrants from Eurasia. The subfamily Listriodontinae in

particular seems to be absent in lower Miocene deposits of Africa, the only

subfamily known in Africa at that time being Kubanochoerinae, which seems to

be an unlikely precursor for the listriodonts on a number of morphological

grounds (Pickford 1986). On this basis it is suggested that listriodonts migrated

into Africa from Eurasia about 17,5 million years ago, and that their origins may

be found in the late Oligocene and early Miocene palaeochoerines of Europe.

The bunodont Arrisdrift suid is enigmatic, in that it could represent a

peculiar form of kubanochoerine, but I consider it more likely that its affinities lie

with the Tetraconodontinae or Palaeochoerinae, both of which are well

represented in European deposits of early and middle Miocene age. If this is so

then the bunodont Arrisdrift pig would indicate that a second lineage of suids

migrated into Africa at the beginning of the early middle Miocene. However, in

view of the uncertainty about the identification of this small bunodont pig, it

would be unwise to be too dogmatic about this point.

It is probably worth mentioning that isolated teeth have recently been found

at Maboko and Nyakach, both of which are early middle Miocene sites in Kenya,

which also indicate the probability that Tetraconodontinae were present in Africa

294 ANNALS OF THE SOUTH AFRICAN MUSEUM

at that time. Unfortunately, none of the specimens is complete enough nor

diagnostic enough for the purposes of positive identification, and I prefer to await

the discovery of such evidence before using it for biostratigraphic or palaeo-

zoogeographic reconstructions.

Suids often seem to have been in the vanguard of major mammal migrations

(Ginsberg, Institut de Paléontologie, Paris, pers. comm.), which makes them

interesting for reconstructing the sequence and timing of palaeozoogeographic

events. It has been suggested on several occasions (Thomas 1985) that initial

crossings of the Tethys Seaway occurred about 19 m.y. ago, a period usually

referred to as the ‘proboscidean datum’. Suids seem not to have partaken in this

particular migration, which has been called the first Neogene Dispersal Phase

(NDP 1) by Thomas (1985), but undoubtedly reached Africa before it began,

indeed earlier than 20 m.y. ago, since they occur at sites such as Legetet and

Songhor in Kenya, and Napak in Uganda, all dated about 20 m.y. Indeed the

main groups of mammals implicated in the First Dispersal Phase of Thomas are

either very large such as proboscideans, were amphibious such as anthracotheres,

or were very small such as rodents and insectivores. This peculiar composition of

the migrant fauna suggests that a filter was active in restricting the migrations of

medium-sized mammals.

In contrast, numerous medium-sized mammals are implicated in Thomas’

second Neogene Dispersal Phase (NDP 2), including bovids, hyracoids, tubuli-

dentates, creodonts and suids. These forms are envisaged as having crossed the

Tethys region about 17,5 m.y. ago.

It is probable that, as Africa moved northwards due to plate tectonic

processes, it approached close enough to Eurasia about 19 m.y. ago to uplift the

floor of the Tethys Sea, making it rather shallow in parts, possibly with islands in

certain sectors. At this time, filtered migrations of the first Neogene Dispersal

Phase would have been possible. As Africa continued moving northwards, parts

of the Tethys sea-floor were elevated above sea-level in continuous strips,

effectively providing dry-land crossings from Eurasia to Africa. At this time

circulation of sea-water between the Atlantic and Indian oceans through the

Tethys was severed, which probably had marked effects on local and global

climates. The more pervasive migration of mammals in this second Neogene

Dispersal Phase dates this event to about 17,5 m.y. Having migrated from

Eurasia into Africa, the suids, particularly Lopholistriodon, apparently spread

quickly throughout the continent, having been found in early middle Miocene

sediments in equatorial and southern Africa.

ACKNOWLEDGEMENTS

I thank Drs Q. B. Hendey and M. A. Cluver of the South African Museum

for permission to study the Arrisdrift fossils. Mr Clive Booth took the

photographs. I also wish particularly to thank Professor Taquet (Institut de

Paléontologie, Paris) for permission to travel to Africa.

MIOCENE SUIDAE FROM ARRISDRIFT 295

REFERENCES

BisHop, W. W., Miiter, J. A. & Fircu, F. J. 1969. New potassium-argon age determinations

relevant to the Miocene fossil mammal sequence in East Africa. American Journal of

Science 267: 669-699.

Corvinus, G. & HENpDEy, OQ. B. 1978. A new Miocene vertebrate locality at Arrisdrift in

Southwest Africa (Namibia). Neues Jahrbuch fiir Paldontologie Monatshefte, Stuttgart

1978 (4): 193-205.

Harris, J. M. & WarKINS, R. 1974. New early Miocene vertebrate fauna near Lake Rudolf,

Kenya. Nature, London 252 (5484): 576-577.

HENDEY, Q. B. 1978. Preliminary report on the Miocene vertebrates from Arrisdrift, South

West Africa. Annals of the South African Museum 76 (1): 1-41.

HENDEY, Q. B. 1984. Southern African late Tertiary vertebrates. In: KLEIN, R. ed. Southern

African prehistory and palaeoenvironments: 81-106. Rotterdam: Balkema.

MacDouGaL.t, I. & WarTkINnS, R. 1985. Age of hominoid-bearing sequence at Buluk, northern

Kenya. Nature, London 318 (6042): 175-178.

PickrorD, M. 1984. Kenya Palaeontology Gazetteer. Volume 1. Western Kenya. Special

Publication National Museums of Kenya: 1-282.

PickForD, M. 1986. A revision of the Miocene Suidae of Africa. Tertiary Research, Special

Paper 7: 1-81.

PIcKFORD, M. & WiLkinson, A. 1975. Stratigraphic and phylogenetic implications of new

Listriodontinae from Kenya. Netherlands Journal of Zoology 25 (1): 132-141.

STROMER, E. 1926. Reste Land- und Susswasser-Bewohnender Wirbeltiere aus den Diamant-

feldern Deutsch-Sudwestafrika. In: KAIsER, E. ed. Die Diamantenwuste Sudwestafrikas 2:

107-153. Berlin: D. Reimer.

Tuomas, H. 1985. The lower and. middle Miocene land connection of the Afro-Arabian plate

and Asia: a major event for hominoid dispersal? Jn: DELSoN, E. ed. Ancestors—the hard

evidence: 42-50. New York: Alan R. Liss.

WILKINSON, A. 1976. The lower Miocene Suidae of Africa. Jn: SAVAGE, R. J. & CorYNDON,

S. C. eds. Fossil vertebrates of Africa 4: 173-282. London: Academic Press.

6. SYSTEMATIC papers must conform to the International code of zoological nomenclature

(particularly Articles 22 and 51).

Names of new taxa, combinations, synonyms, etc., when used for the first time, must be

followed by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb.

nov., Syn. nov., etc. :

An author’s name when cited must follow the name of the taxon without intervening

punctuation and not be abbreviated; if the year is added, a comma must separate author’s

name and year. The author’s name (and date, if cited) must be placed in parentheses if a

species or subspecies is transferred from its original genus. The name of a subsequent user of

a scientific name must be separated from the scientific name by a colon.

Synonymy arrangement should be according to chronology of names, i.e. all published

scientific names by which the species previously has been designated are listed in chronological

order, with all references to that name following in chronological order, e.g.:

Family Nuculanidae

Nuculana (Lembulus) bicuspidata (Gould, 1845)

Figs 14-15A

Nucula (Leda) bicuspidata Gould, 1845: 37.

Leda plicifera A. Adams, 1856: 50.

Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b).

Nucula largillierti Philippi, 1861: 87.

Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9.

Note punctuation in the above example:

comma separates author’s name and year

semicolon separates more than one reference by the same author

full stop separates references by different authors

figures of plates are enclosed in parentheses to distinguish them from text-figures

dash, not comma, separates consecutive numbers

Synonymy arrangement according to chronology of bibliographic references, whereby

the year is placed in front of each entry, and the synonym repeated in full for each entry, is

not acceptable.

In describing new Species, One specimen must be designated as the holotype; other speci-

mens mentioned in the original description are to be designated paratypes; additional material

not regarded as paratypes should be listed separately. The complete data (registration number,

depository, description of specimen, locality, collector, date) of the holotype and paratypes

must be recorded, e.g.:

Holotype :

SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach

Port Elizabeth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973.

Note standard form of writing South African Museum registration numbers and date.

7. SPECIAL HOUSE RULES

Capital initial letters

(a) The Figures, Maps and Tables of the paper when referred to in the text ’ .

e.g. “... the Figure depicting C. namacolus...; ‘...in C. namacolus (Fig. 10)...’

(b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded

by initials or full names

e.g. Du Toit but A.L.du Toit; Von Huene but F. von Huene

e.g. Therocephalia, but therocephalian

Punctuation should be loose, omitting all not strictly necessary

Reference to the author should be expressed in the third person

Roman numerals should be converted to arabic, except when forming part of the title of a

book or article, such as

“Revision of the Crustacea. Part VIII. The Amphipoda.’ ;

Specific name must not stand alone, but be preceded by the generic name or its abbreviation

to initial capital letter, provided the same generic name is used consecutively.

Name of new genus or species is not to be included in the title: it should be included in the

abstract, counter to Recommendation 23 of the Code, to meet the requirements of

Biological Abstracts.

MARTIN PICKFORD

MIOCENE SUIDAE FROM ARRISDRIFT,

SOUTH WEST AFRICA-NAMIBIA

JME 97 PART 11 JANUARY 1988 ISSN 0303-2515

' ie ‘ “Aas

APR 15 1989 |

~ OF THE SOUTH AFRICAN —

CAPE TOWN

INSTRUCTIONS TO AUTHORS

1. MATERIAL should be original and not published elsewhere, in whole or in part.

2. LAYOUT should be as follows:

(a) Centred masthead to consist of

Title: informative but concise, without abbreviations and not including the names of new genera or species

Author’s(s’) name(s)

Address(es) of author(s) (institution where work was carried out)

Number of illustrations (figures, enumerated maps and tables, in this order)

(b) Abstract of not more than 200 words, intelligible to the reader without reference to the text

(d) Introduction

(e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents

(f) Summary, if paper is lengthy

(g) Acknowledgements

(h) References

(i) Abbreviations, where these are numerous.

3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with

3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for

illustrations should be typed separately, their positions indicated in the text. All pages should be num-

bered consecutively.

Major headings of the paper are centred capitals; first subheadings are shouldered small capitals;

second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further

subdivisions should be avoided, as also enumeration (never roman numerals) of headings and

abbreviations.

Footnotes should be avoided unless they are short and essential.

Only generic and specific names should be underlined to indicate italics; all other marking up

should be left to editor and publisher.

4. ILLUSTRATIONS should be reducible to a size not exceeding 12 x 18 cm (19 cm including

caption); the reduction or enlargement required should be indicated (and preferably uniform); orig-

inals larger than 35 x 47 cm should not be submitted; photographs should be rectangular in shape and

final size. A metric scale should appear with all illustrations, otherwise magnification or reduction

should be given in the caption; if the latter, then the final reduction or enlargement should be taken

into consideration.

All illustrations, whether line drawings or photographs, should be termed figures (plates are not

printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items

of composite figures should be designated by capital letters; lettering of figures is not set in type and

should be in lower-case letters. If Letraset is used authors are requested to use Helvetica-style letter-

ing, if possible.

The number of the figure should be lightly marked in pencil on the back of each illustration.

5. REFERENCES cited in text and synonymies should all be included in the list at the end of the

paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable):

(a) Author’s name and year of publication given in text, e.g.:

‘Smith (1969) describes...’

‘Smith (1969: 36, fig. 16) describes... .’

‘As described (Smith 1969a, 19695; Jones 1971)’

‘As described (Haughton & Broom 1927)...’

‘As described (Haughton et al. 1927)... .’

Note: no comma separating name and year

pagination indicated by colon, not p.

names of joint authors connected by ampersand

et al. in text for more than two joint authors, but names of all authors given in list of references.

(b) Full references at the end of the paper, arranged alphabetically by names, chronologically within

each name, with suffixes a, b, etc., to the year for more than one paper by the same author in

that year, e.g. Smith (1969a, 19695) and not Smith (1969, 19692).

For books give title in italics, edition, volume number, place of publication, publisher.

For journal article give title of article, title of journal in italics (according to the World list of scientific periodicals. 4th ed.

London: Butterworths, 1963), series in parentheses, volume number, part number in parentheses, pagination (first and

last pages of article).

Examples (note capitalization and punctuation)

BuLLouGu, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan.

Fiscuer, P. H. 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140.

FIscHER, P. H., DuvaL, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie

expérimentale et générale 74 (33): 627-634.

Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and

Magazine of Natural History (13) 2 (17): 309-320.

Koun, A. J. 1960b. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of

the Bingham Oceanographic Collection, Yale University 17 (4): 1-51.

THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: SCHULTZE, L. Zoologische und anthro-

pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Stid-Afrika ausgefiihrt in den Jahren

1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270.

(continued inside back cover)

ANNALS OF THE SOUTH AFRICAN MUSEUM

ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

Volume 97 Band

January 1988 Januarie

Rant) hie Deel

A NEW FOSSIL STORK

(AVES, CICONIIDAE) FROM THE

LATE TERTIARY OF LANGEBAANWEG,

SOUTH AFRICA

P. J. HAARHOFF

Cape Town Kaapstad

The ANNALS OF THE SOUTH AFRICAN MUSEUM

are issued in parts at irregular intervals as material

becomes available

Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000

Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

word uitgegee in dele op ongereelde tye na gelang van die

beskikbaarheid van stof

Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000

OUT OF PRINT/UIT DRUK

il, AGES, 52), HOD, 4-5, 8, isa.,)), SUES, 5, ED),

AG, to.) TO), 8, OAD, 7), IGS,

GED, Si, 7, tos), WEED), 1SCS), DA), 27), FB}, 226), 23, 25), 25H)

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 091 5

Printed in South Africa by In Suid-Afrika gedruk deur

dihe Rustica bressy kiya ieide Die Rustica-pers, Edms., Bpk.,

Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap

A NEW FOSSIL STORK (AVES, CICONHDAE) FROM THE

LATE TERTIARY OF LANGEBAANWEG, SOUTH AFRICA

P. J. HAARHOFF

Department of Cenozoic Palaeontology,

South African Museum, Cape Town

(With 3 figures and 4 tables)

[MS accepted 2 June 1987]

ABSTRACT

A new species of stork, Ciconia kahli, from the early Pliocene Varswater Formation, Cape

Province, South Africa, is described on the basis of a partial associated skeleton and three other

referred specimens. This species has the greatest similarity to the living white stork, Ciconia

ciconia, but also resembles the saddlebill stork, Ephippiorhynchus senegalensis, and black-

necked stork, E. asiaticus, in certain characters. It is nevertheless very distinctive and probably

represents an extinct lineage within the Ciconiidae.

CONTENTS

PAGE

JEONG CENT eS eS SSS om SA eT net er OL 297

SW SUC A ULC Spier enn et cnc mere Ree reir tas CIR IM A aa Sonn une tae 299

IDI SCUSSIOl mp on hie ee irr rea ee R ore i ute nmy aie eu. 310

PNCKMOWHE CP EIMEINS ern ee ieee alte cee ee RES Ay clatlartua ain cease 4 Sulit

INCHCTCTICS Sarre ee nnr ee TEL Ure Ne eee a i oet es cI I cha alace ln 311

INTRODUCTION

The early Pliocene (c. 5 Ma) Varswater Formation at Langebaanweg,

approximately 110 km north-north-west of Cape Town, has yielded a great

number and variety of fossils (Hendey 1981a, 19815). In a preliminary report on

the avian remains from this site Rich (1980) listed a minimum of 61 taxa.

Further research on the collection has raised the number of species to 81.

Simpson (1971, 1975, 1979) studied the penguin material, which is now being re-

examined by Olson (1983, in prep.). Olson (1984) has also described a new

species of the ciconiiform family Scopidae, Scopus xenopus, and has studied the

Procellariiformes (1985a) and the ciconiiform family Plataleidae (19855). Rich &

Haarhoff (1985) have described a new species of the family Coliidae, Colius

hendeyi. The associated stork remains described in this paper are from the

Quartzose Sand Member (QSM), the lower of the two highly fossiliferous units

of the Varswater Formation. This unit accumulated behind a sand-bar in and

adjacent to the estuary of the Berg River, which now has a more northerly

course. The OSM contains fluviatile, estuarine, floodplain, marsh, tidal mudflat,

2S)

Ann. S. Afr. Mus. 97 (11), 1988: 297-313, 3 figs, 4 tables.

298 ANNALS OF THE SOUTH AFRICAN MUSEUM

and pond facies (Hendey 1976). The partial stork skeleton, SAM—PO-L22164,

was found in situ, in the floodplain deposits (QSM 1) of the Quartzose Sand

Member. Three other bones of uncertain provenance have been referred to the

Same species.

Kahl’s (1972) classification of living storks is followed here (Table 1),

although it requires revision (see p. 310). Comparisons with living species indi-

cate that, on the basis of the premaxilla and mandible, the fossil stork belongs to

the tribe Ciconini. The postcranial material is sufficiently different from all the

living species and all known fossil species to warrant assignment to a new

species. The new fossil stork does bear certain resemblances to Ciconia ciconia

and to a lesser extent to Ephippiorhynchus senegalensis and E. asiaticus.

The material described is housed in the Department of Cenozoic Palaeont-

ology, South African Museum. Catalogue numbers of fossil material are

prefixed SAM—PQ, with the additional prefix L indicating material from Lange-

TABLE 1

Classification of the Ciconiidae according to Kahl (1972).

Family Ciconiidae

Tribe Mycteriuni

Mycteria americana Linnaeus American wood stork

Mycteria cinerea (Raffles) milky stork

Mycteria ibis (Linnaeus) yellowbilled stork

Mycteria leucocephala (Pennant) painted stork

Anastomus oscitans (Boddaert) Asian openbill stork

Anastomus lamelligerus Temminck African openbill stork

A. I. lamelligerus Temminck

A. I. madagascariensis Milne Edwards

Tribe Ciconiini

Ciconia nigra (Linnaeus) black stork

Ciconia abdimii Lichtenstein Abdim’s stork

Ciconia episcopus (Boddaert) woollynecked stork

C. e. episcopus (Boddaert)

C. e. microscelis G. R. Gray

C. e. stormi (Blasius)*

Ciconia maguari (Gmelin) maguari stork

Ciconia ciconia (Linnaeus) white stork

C. c. ciconia (Linnaeus)

C. c. asiatica Severtzov

C. c. boyciana Swinhoe*

Tribe Leptoptilini

Ephippiorhynchus asiaticus (Latham) blacknecked stork

E. a. asiaticus (Latham)

E. a. australis (Shaw)

Ephippiorhynchus senegalensis (Shaw) saddlebill stork

Jabiru mycteria (Lichtenstein) jabiru stork

Leptoptilos javanicus (Horsfield) lesser adjutant stork

Leptoptilos dubius (Gmelin) greater adjutant stork

Leptoptilos crumeniferus (Lesson) marabou stork

*Possibly has attained specific status; further study needed.

FOSSIL STORK FROM LANGEBAANWEG 299

baanweg. Modern comparative material in the South African Museum is

distinguished by the prefix SAM-—ZO.

Institutional abbreviations used in the text are as follows:

AMNH American Museum of Natural History, New York.

AM-S Australian Museum, Sydney.

ANWC Australian National Wildlife Collection, Canberra, Australia

(CSIRO).

BMNH _ British Museum (Natural History), London.

LACM Natural History Museum of Los Angeles County, California.

NMV National Museum of Victoria, Australia.

OT-B Otago Museum, Dunedin, New Zealand.

SAM South African Museum, Cape Town.

™ Transvaal Museum, Pretoria.

UCMP University of California, Museum of Paleontology, Berkeley.

Most measurements (in mm) were taken following Von den Driesch (1976)

and Olson (19855). Anatomical terminology follows that of Howard (1929).

Comparative material examined

Mycteria americana, SAM-—ZO57920; M. leucocephala, BMNH 396A,

BMNH 396B; M. ibis, AM-S 1235; Anastomus lamelligerus, TM 33382,

AMNH 5292; Ciconia nigra, SAM-—ZO56944, OT-B76:1; C. abdimii,

T™ 33336; C. episcopus, TM 33391; C. maguari, AM-S 551; C. ciconia,

SAM-ZOS56181, SAM-—ZO57363, SAM-—ZO57471, AM-S 1126; Ephippio-

rhynchus asiaticus, BMNH 955B, ANWC 108, ANWC 686, ANWC 4139,

ANWC 1508, NMV-B6753; E. senegalensis, LACM 90275, AMNH 2903;

Leptoptilos dubius, NMVV-B11426, NMV-—R2203, NMV—W5083; L. crumeni-

ferus, TM 33412; L. javanicus, NMV—B736, AMNH 5059; Jabiru mycteria,

AMNH 2659, UCMP 133932.

SYSTEMATICS

Family Ciconiidae Gray, 1840

The fossil specimens can be referred to the Ciconiidae on the basis of the

following characters: (1) sternum excavated on anterior carinal margin; (2) last

thoracic vertebra not fused to synsacrum; (3) synsacrum, cranial view, pre-

zygapophysis not joined to diapophysis at posterior edge; (4) scapular facet of

coracoid deep and well rounded; (5) carpal tuberosity of ulna pointed, with liga-

mental attachment on caudal surface; (6) intercotylar prominence on

tarsometatarsus rises well above cotylae, which are more or less of equal height;

and (7) trochleae II and IV rather flattened anteriorly, not rounded. Characters

1, 4, 5, 6, and 7 separate the Ciconiidae from the Scopidae, Ardeidae and Plata-

leidae, character 2 from the Scopidae, and character 3 from the Ardeidae and

Plataleidae.

300 ANNALS OF THE SOUTH AFRICAN MUSEUM

Genus Ciconia Brisson, 1760

The fossil species described below is referable to the genus Ciconia by

having the combination of: (1) tips of maxilla and mandible straight; (2) anterior

carinal margin of sternum unperforated; (3) coracoid with a ridge between head

and midpoint of shaft in ventral view; (4) deep depression between the proximal

articulating surface of the tarsometatarsus and the hypotarsus; and (5) lack of

significantly pneumatic postcranial elements.

Characters 1—5 distinguish the genus Ciconia from the other five Recent

genera of Kahl (1972) and therefore the following putative extinct species within

those genera: Mycteria wetmorei Howard, 1935; Leptoptilos falconeri (Davies,

1880) (Lydekker 1884; Harrison 1974); L. siwalikensis Harrison, 1974; L. richae

(Harrison, 1974); L. titan Wetmore, 1940; L. pliocenicus Zubareva, 1948;

Leptoptilos sp. Hill & Walker, 1979; Ephippiorhynchus pakistanensis Harrison

& Walker, 1982.

The fossils from Langebaanweg were also compared with and found to be

distinct from the following extinct genera: Palaeoephippiorhynchus dietrichi

Lambrecht, 1930; Grallavis edwardsi (Lydekker, 1891) (Lambrecht 1933;

Cheneval 1984); ‘Propelargus’ olseni Brodkorb, 1963 (requires revision,

Cheneval pers. comm.); ‘“Dissouroides’ milleri Short, 1966 (requires revision,

Olson pers. comm.); Cryptociconia indica Harrison, 1974; Pelargosteon tothi

Kretzoi, 1962; “Prociconia’ lydekkeri Ameghino, 1891 (requires revision, Olson

pers. comm.). Ciconiopsis antarctica Ameghino, 1899, was not studied by the

author but, since it is from the early Oligocene of Argentina, it seems unlikely

that the fossils from Langebaanweg should be assigned to this genus. Also,

according to Olson (1986), it requires restudy before it can be accepted as a stork.

Ciconia kahli sp. nov.

Figs 1-3

Material

Holotype. SAM—PQ-L22164, partial associated skeleton consisting of the

following elements: fragment of premaxilla (length 53,7 mm), mandibular sym-

physis (length 80,7 mm) plus associated fragments; part of sternal manubrium

and coracoidal sulcus; complete furcular process plus fragmented pieces; com-

plete left proximal and incomplete right proximal scapulae; incomplete left and

right coracoids; incomplete left distal humerus; fragments of head, shaft, and

distal end of right humerus; almost complete right ulna; external cotyla and

crushed fragments of shaft of left ulna; complete right radius (slightly damaged

proximal end); right scapholunar; right cuneiform; right carpometacarpus; right

phalanx 1 and 2 of major alar digit; incomplete right femur; incomplete distal

end of left tibiotarsus and shaft; incomplete right fibula; incomplete left tarso-

metatarsus (lacks distal end); pes—incomplete ungual phalanx, possibly of digit 3;

first and second thoracic vertebrae; eleventh and twelfth cervical vertebrae

(pathological), plus numerous vertebral fragments.

FOSSIL STORK FROM LANGEBAANWEG 301

Fig. 1. Ciconia kahli sp. nov., holotype, SAM-—PQ-L22164. A-B. Coracoid.

C-D. Carpometacarpus. E. Sternum. F. Maxilla and mandible. All x 1.

302 ANNALS OF THE SOUTH AFRICAN MUSEUM

3 ,

“ay

Fig. 2. Ciconia kahli sp. nov., holotype, SAM-PQ-L22164. A. Radius. B. Ulna.

C. Humerus. D. Proximal end of humerus. E. Humerus (proximal view). F. Distal end

of tibiotarsus (internal view). G. Femur (posterior view). H. Femur (internal view).

I. Tibiotarsus. A-B.x0,5. C-H.x1. I. x 0,6.

FOSSIL STORK FROM LANGEBAANWEG

Fig. 2 (contd)

304 ANNALS OF THE SOUTH AFRICAN MUSEUM

Paratypes. SAM—PQ-L41955, synsacrum lacking a few caudal vertebrae.

SAM-PQ-L42157, right ulna lacking proximal end. SAM—PQ-L-6508, com-

plete distal end of right tarsometatarsus (Fig. 3). Including the holotype, the

minimum number of individuals is two.

Measurements of holotype (an mm)

Humerus: actual length, 180,0; estimated greatest length, 274,0; estimated

distal width, 41,2; smallest width of shaft, 19,1. Ulna: actual length of proximal

portion, 133,4; actual length of distal portion, 183,0; estimated greatest length,

320,0; greatest diagonal of distal end, 20,9; smallest width of shaft, 9,5. Radius:

greatest length, 304,0; least and greatest diameter of shaft at midpoint,

5,1 X 8,7; greatest distal diameter, 18,4. Carpometacarpus: greatest length,

144.3; depth through process of metacarpal 1, 29,0; greatest proximal width

through trochleae, 12,6; width and depth of major metacarpal at midpoint,

9,7 X 6,9; greatest distal diameter, 16,3. Coracoid: medial length, 93,4; esti-

mated greatest length, 103,0; greatest depth of sternal facet, 12,6. Sternum:

greatest width of excavation on anterior carinal margin, 12,0; greatest breadth of

coracoidal sulcus, 14,1. Scapula: greatest cranial diagonal, 26,6. Femur: actual

length, 32,5; estimated greatest length, 117,9; approximate shaft width and

depth, 15,2 x 15,4. Tibiotarsus: actual length, 267,7; estimated greatest length,

348,0; width and depth of shaft at approximate midpoint, 14,6 x 11,2; depth of

anterior internal condyle, 13,8; length of supratendinal bridge, 6,3. Fibula:

actual length, 51,4; estimated greatest length, 230,0. Tarsometatarsus: actual

length, 292,1; estimated greatest length, 321,7; proximal width, 22,5; approxi-

mate width and depth of shaft at approximate midpoint, 9,7 x 9,4.

Measurements of paratypes (in mm)

Synsacrum, SAM—PQ-L41955: actual length, 96,1; depth at midpoint of

second synsacral thoracic vertebra, 36,4; width of first synsacral thoracic verte-

bra at midpoint, 7,4. Ulna, SAM—PQ-L42157: actual length, 158,6; greatest

diagonal of distal end, 20,9. Tarsometatarsus, SAM—PQ-L6508: distal width,

24,5; width and depth of middle trochlea, 7,9 x 11,7.

Diagnosis

Larger than all five extant species of Ciconia (Table 2) and the extinct

species Ciconia minor Harrison, 1980, Ciconia sarmatica Grigorescu & Kessler,

1977, and Ciconia nana (De Vis, 1888) (Rich & Van Tets 1982). Also larger

than Ciconia sp. Harrison, 1980, and Ciconia sp. Ono, 1984. Ciconia sp. Stehlin,

1923, is juvenile and was not considered. Smaller than Ciconia gaudryi Lam-

brecht, 1933. Within size range (Table 3) of the extinct species Ciconia maltha

Miller, 1910, but morphologically distinct as follows: sternum—excavation on

anterior carinal margin shallower and wider; coracoid—internal view, less

inflated below depression beneath brachial tuberosity and brachial tuberosity

less erect; humerus—shaft more robust (see also unique features below);

FOSSIL STORK FROM LANGEBAANWEG 305

capmer tte

Fig. 3. Ciconia kahli sp. nov. A-C. Holotype, SAM-—PQ-L22164. Tarsometatarsus.

| A. Proximal view. 8B. Anterior view. C. Internal view. D-F. Paratype,

SAM-PQ-L6508. Tarsometatarsus. D. Anterior view. E. Posterior view. F. Distal

view. A, C-F. x1. B. x 0,6.

306 ANNALS OF THE SOUTH AFRICAN MUSEUM

TABLE 2

Length measurements (mm) of major elements of Ciconia kahli sp. nov. compared with extant

species of Ciconia.

C. kahli C.nigra C.abdimii C. episcopus C.maguari C. ciconia

(n= 1) (n= 2) (n= 1) (n= 1) (n= 1) (n= 4)

Coracoid 103* 77,5-85,5 O3}5) 66,4 97,9 80,5—90,9

Humerus 274* 191,4-197,1 IS 158,1 WMS) 196,8-213,8

Ulna 320* 218,4—230,9 174,0 185,7 265 ,4 227 ,1-—253,0

Radius 304 209 ,4—220,0 166,3 179,0 254,3 216,1—242,8

Carpometacarpus 144,3 110,7-113,5 84,0 85,4 119,5 107,7-114,6

Femur 117,9* 94 ,5-94,8 71,6 Bs 113,0 90,1-110,7

Tibiotarsus 348* 239,1-246,6 EO 202,6 SA 238 ,2—260,8

Tarsometatarsus 321,7* 192,4—200,1 11333}-5) 150,2 DORA 198,1-225,1

* = estimated

tarsometatarsus—intercotylar prominence narrower, in distal view trochleae

less arched and trochlea III less raised and rounded in external view in C. kahit.

Ciconia kahli differs morphologically from the extant species of Ciconia as

follows: sternum—expanded coracoidal sulcus; coracoid—neck proportionally

longer; radius—shaft less curved; carpometacarpus—inner carpal trochlea,

internal view, flattened anteriorly; tibiotarsus—internal ligamental prominence

placed further away from internal margin; tarsometatarsus—depression

between hypotarsus and proximal articulating surface deeper in C. kahii.

Certain features of the humerus of C. kahli are different from all species of

Ciconiidae studied for this report: (1) head, palmar view, relatively flattened,

sloping gradually toward external side; (2) proximal view, head broader near

external tuberosity; (3) capital groove shallow; (4) groove between ectepicon-

dylar prominence and external condyle deep.

Distribution

Early Pliocene Varswater Formation (Quartzose Sand Member Unit I and

possibly Pelletal Phosphorite Member bed 3aS) at Langebaanweg, south-

western Cape Province, South Africa.

Etymology

This species is named in honour of Dr M. Philip Kahl, who has done so

much work on the Recent genera of the family Ciconiidae.

Remarks

Although Ciconia kahli has the greatest similarity to C. ciconia, it shares

the following characters with Ephippiorhynchus: (1) twelfth cervical vertebra not

longitudinally compressed; (2) first and second thoracic vertebrae with hypapo-

physes; (3) ulna, in palmar view, flattened immediately below the proximal radial

depression; (4) carpometacarpus, in internal view, with a deep depression between

metacarpal I and the pisiform process; (5) internal condyle of tibiotarsus,

307

FOSSIL STORK FROM LANGEBAANWEG

LLOL A9[SSOM 2 NISoIOSIIH “9 “ZpHT PleMOH “¢ “Apnys JUaseIg “p “E€HT IYSOIQUILT “¢ ‘OSGI UOSHIET] “Z “PRG OUD ‘[ :seoUdIAFoy *afApuoo

[BU19}UI 1ol19]Ue Jo yIdop = step ‘aspriq [euIpusjeidns Jo yIsUs] = qs] ‘YyIpeorq [eISIP = pq ‘yISUI] 1Sso}v0I3 = [3 ‘yIpeoIq JeuNTxosd = dq ‘pojyeulnss = ,

S °C

ah WES

8°€l

4 8VE

€°9

Omar

«£01

9° €7

Sol Ocl

SiGe

Gane

OZE-8°6S7

S‘17-0'8T

ie Sle

6 LC GO

7 1OI-r'78

ClGI=G GC

SEE-STE

087-057

0° L7-¥' 97

091-9 rT

CULO

L'y-9'p

Ole

DIUOIID “dD

JO oZIs

«007 [3

DIUOIID ‘+

JO 9ZIS €/7 Pq

SNSICJEJOWOSIC TL

SNSIeJOIQL],

Inui

ploooied

sndivoejouodied

bul)

sniouiny}y

ejndeos

See ee ee ee

POLITY INOS

eIUPWIOY

ouepy

uosd10

epHol4y

EDIOMIED

a a I ee ee ee

140

DIUOIID

gDI1JDULADS

DIUOIID

sDY)]DUL

DIUODND

elyelsny 90001 eAuoy eAuoy ueder

(Gu)

pouvu -idupno3s z'ds 74OUIM , ds

DIUOII) DIUOIID DIUOIID DIUOINID DIUOII)

ee ne Nn Ow Se) eS a ee aly aoe Saree re eh ee yh ee RO eee Reet a Ss

‘Aou “ds 7Yyvy vIUOIID pu vIUOIID Jo saroeds {Isso} 10J (WUT) s}UOWIOINSeoW sATeIedUIOD

¢ ATaV

i Th sprees sees TS Sten

ANNALS OF THE SOUTH AFRICAN MUSEUM

308

snyoukysoiddiydy

sndoosida piuod1g

NVISVIVULSNV

‘ds piuoo1y

uvjy sojydojdaT

pjpydad0ona]

puvUu vIUOIID (sig]) vuajoc py

snignp sojydojdaT

snoiupavl

sojydojdaT

Snoyvisp

snyoudysoiddiy dq

DIUOIID DIUOIID

DASIU DIUOIID

sndoogsida pmuo0s1y

SUDJIISO SNUOJSDUY

pjvpydad0ona]

DIUAJIA

DadaUID DIUAJIAW

snoypisp

TVINAITYO

sndafiuawinsd

sojdojdaT]

sisuajvsauas

snyoudysoiddiydy

DIUOIII DIUOIID

sndoosida viuod1yd

CED DISD)

DASIU DIUOII)

SndasyjauUv]

snwojspuy

$1G1 DIUAaJIA PW

NVIdOIHLA

eS SR

‘ds piuoo1g

Men

uoajsosivjad

s1qi (S1Q]) viualnKw

DASIU DIUOIID)

DIUOIID DIUOIID

p1uodn1d VIUOIID

DASIU DIUOIID

JILDUVATVd

l4ayyapa]

D1U0I190dq,

(P10)

DUDILAAUD DIUAJIA

(PTO)

pidajokud ndigvoe

MONSDUL DIUOI1)

DIUAJIAU NAIGD

MONSDUL DIUOIID

DUDIIAIUD DIUAJIA

TVOIdOUL OAN

DYJJVU DIUOIID

1aAOUJaM DIAAIIA PL

DUDIIAIUD DIUAJIA

DUDIMAWUD DIAAIK

OILDUVAN

‘y10da1 SIy 1OJ PIIPN}s Sy10}s SUIAT] PUR [ISSO] [Je JO UONNQIISIP pue ody

p ITaVv |,

ALV'I

ANHOOLSIA Id

OTYOLSIH

309

FOSSIL STORK FROM LANGEBAANWEG

‘7861 IONE ZW UOStIIeF{—wUour90ds possozor A[OANLIUDL, “OS6T JOPIOMAeI 6 ‘LL6L

IQ|SSOS 7 NISIIOSUD *8 “6L6T JOMICM 7 IH “L “OLOL OPEAA 7 JOUPIV “9 “GT86I “PI86T AopusH “¢ “(‘uiuOD ‘siod) uostIeH “pf “pL6y SULIZANOD URA 2?

udIS3IIg “€ “LO6T BIOONPS “7 “6L6T [[oqdwieD “| :ssoudIoJoy ‘sosoyjUssed Ul UDAIS UDEq DAY JUASOId dIOJOq SIedA JO UOT Ut sayep ‘a[qIssod s10y A

1YdMjaIp

SnYDUAYA DINIADIUD ATHYVA ANAOODITO

-o1ddiydaoavjvg (S1Sdo01uod1),

AOUNU DIUOIID ISpAvMpa S1AD]JVAD) ATUNVA

6(0°ST-S‘€T)

‘ds y1uo91) luasjo snsivjadoid, AIGCIWN

ANADOIN

(0°21)

4 SISUIYNDMIS DINDUADS DIUOIID

sojudojdaT apyaid sojydojdaT ALV1

sisuaunjstyod SIT) (0'6)

snysudysoiddiydy ‘ds sojydojdaT = =— 1kuppnv8 viuoo1p

s(S‘b-0'p) s(0°S)

(snyoudysouay ) ‘aou ‘ds 14a] ]1U ATUVaA

snyoudysoiddiy dy ¢ 11YDY DIUOIID SAplOANOSSIC{,

snoiuasoyd

sojydojdaT ATAdIN

re(0°E-8'T)

DOIpUul

piuon120jdky ANADOr Id

»e(0°E-8°T)

uaUuogvf ALV'I

sojdojdaT

y‘c(0€-8'T)

sisuayyoMis (LZ0 + ¥'€)

sojdojdaT DYYVU VIUOIID

310 ANNALS OF THE SOUTH AFRICAN MUSEUM

in distal view, with convex external margin; (6) depression between hypotarsus

and proximal articulating surface; (7) trochleae, in distal view, form fairly flat,

wide, arch; (8) similar shape of trochlea III and position of trochlea IV.

Proportional differences observed include the following: (1) Leptoptilos

(apart from L. javanicus) is the only genus in which the ulna is longer than the

tibiotarsus. Ephippiorhynchus is extreme in that the tibiotarsus is considerably

longer than the ulna. In the relative length of the ulna and tibiotarsus, C. kahli

is more similar to the Ciconia group and Jabiru mycteria. (2) Ciconia kahli

differs from Recent storks in having the tarsometatarsus and tibiotarsus about

equal in length, whereas in all other species the tibiotarsus is longer, especially

so in Leptoptilos crumeniferus, L. dubius, and Mycteria leucocephala.

DISCUSSION

Wood (1984) suggested from his phenetic analysis, based on major skeletal

elements of the stork, that Jabiru mycteria be included in the genus Ephippio-

rhynchus and that Ephippiorhynchus should be transferred to the tribe Ciconiini.

Observations in this report on the similarities between Ciconia and Ephippio-

rhynchus support this view. Also, on the basis of their comparisons of the

genetic material, DNA, Sibley & Ahlquist (1985) place Ciconia and Ephippio-

rhynchus on the same branch of their phylogram. However, Cheneval (1984)

found that, although the fossil stork Grallavis edwardsi is most closely related to

Ephippiorhynchus, it also has some characters in common with Leptoptilos, so

the fossil record may require some revision of Kahl’s (1972) classification.

Ciconia kahli was a tall, fairly robust bird with shorter wings but longer legs

than Leptoptilos crumeniferus. Its most distinctive features are in the humerus,

radius, carpometacarpus, tibiotarsus and tarsometatarsus. Olson (pers. comm.)

has pointed out that the greater size of C. kahli is paralleled by some of the Plio-

cene and Pleistocene species of Leptoptilos that were much larger than any

modern species of that genus. As shown in Table 4, Ciconia is the most common

and widespread genus of stork in Tertiary and Historic times. Just how the

species of Ciconia are related to one another is as yet undetermined, but

Ciconia kahli seems sufficiently distinct morphologically as not to have been

ancestral to any living species and thus may represent an extinct lineage within

the Cicontidae.

Two species of stork occur today in the southern and south-western Cape

Province of South Africa. They are Ciconia c. ciconia and C. nigra. Neither is

common in the Langebaanweg area. Ciconia c. ciconia is a Palaearctic migrant,

although a few pairs breed in the southern Cape. Ciconia nigra is a locally

migratory species recorded as breeding in the eastern and southern Cape

(Clancey 1980).

The inferred habitat (floodplain close to a river with marshy areas) is in

keeping with that preferred by some contemporary species of the family Ciconii-

FOSSIL STORK FROM LANGEBAANWEG 311

dae. Others, including C. ciconia, are opportunistic and take advantage of fires

and burned areas. This could have been true of the Langebaanweg stork, for

such conditions were also present at the time of deposition (Hendey 1981b).

ACKNOWLEDGEMENTS

I am most grateful to Dr Q. B. Hendey (South African Museum), for allow-

ing me to work on the material. I am indebted to Dr P. V. Rich (Monash

University) for her encouragement and much needed supervision. I am also

grateful to Professor Bruce Hobbs and his staff at Monash University, Earth Sci-

ences Department, for providing facilities and help during the preparation of

this manuscript. Financial aid was provided by the South African Museum

enabling me to begin this study under the guidance of Dr P. V. Rich. I had

useful discussions and/or comments from the following people: Mr G. Avery

and Dr Q. B. Hendey (South African Museum), Prof. P. Brodkorb (University

of Florida), Mr R. K. Brooke and Dr A. Guillet (Percy FitzPatrick Institute of

African Ornithology), Dr K. Campbell (Natural History Museum of Los

Angeles County), Dr J. Cheneval (Université de Lyon), Dr C. J. O. Harrison

(British Museum (Natural History)), Dr M. P. Kahl (Naples, Florida), Dr S. L.

Olson (Smithsonian Institution). I thank Mr C. Balouet (Institut de Paléontologie,

Paris), Dr K. Campbell (Natural History Museum of Los Angeles County),

Mr G. S. Cowles, Mr C. A. Walker (British Museum (Natural History)), Dr

B. E. Hastings (University of California, Berkeley), Prof. D. Janossy (Magyar

Nemzeti Muzeum), Dr A. C. Kemp (Transvaal Museum), Mrs M. Le Croy, Dr

D. Russell (American Museum of Natural History), Mr A. McEvey (National

Museum of Victoria), Dr J. van Tets (C.S.I.R.O., Wildlife and Rangelands

Management, Canberra), for loan material. Dr C. J. O. Harrison provided a

date for the age of the Siwalik material. Figures 1-3 were prepared by Ms

E. Pretorius and Mr C. Hunter (South African Museum). I thank Mrs

R. Coppin, Mrs P. Eedes, Mr F. Ponticelli, Mrs E. Pullum, Mrs T. Reyneke,

Mrs S. Saven, Mrs M. Scheiner and Mrs R. Sheehan for their assistance in the

preparation of the manuscript. Help with translations was given by Dr A. H. B.

De Decker, Mr A. Hohmann and Mrs M. Wright. Especial thanks are due to

Drs Pat and Tom Rich for their much valued support, and also to Dr Ewan

Fordyce, Dr John Long, Mrs Betty Thompson (Monash University) and Paddy

and Libby McCarthy, (Cape Town).

REFERENCES

ARCHER, M. & WapbeE, M. 1976. Results of the Ray E. Lemley Expeditions. Part I. The

Allingham formation and a new Pliocene vertebrate fauna from northern Queensland,

Australia. Memoirs of the Queensland Museum 17 (3): 379-397.

BERGGREN, W. A. & VAN COUVERING, J. A. 1974. The Late Neogene. Amsterdam: Elsevier.

BLACKWELDER, B. W. 1980. Late Cenozoic marine deposition in the United States Atlantic

coastal plain related to tectonism and global climate. Palaeogeography, Palaeoclima-

tology, Palaeoecology 34 (1/2): 87-114.

SID ANNALS OF THE SOUTH AFRICAN MUSEUM

CAMPBELL, K. E., JR. 1979. The non-passerine Pleistocene avifauna of the Talara Tar Seeps,

northwestern Peru. Contributions. Life Sciences Division, Royal Ontario Museum 118:

1-203.

CHENEVAL, J. 1984. Grallavis edwardsi (Lydekker, 1891), nouveau genre d’oiseau (Ciconii-

formes) du gisement aquitanien de Saint-Gérard-le-Puy (Allier, France). Bulletin men-

suel de la Société linnéenne de Lyon 53 (2): 43-60.

Crancey, P. A. ed. 1980. S.A.O.S. checklist of southern African birds. Johannesburg:

Southern African Ornithological Society.

De Vis, C. W. 1888. A glimpse of the post-Tertiary avifauna of Queensland. Proceedings of

the Linnean Society of New South Wales (2) 3 (2): 1227-1292.

DriescH, A. VON DEN. 1976. A guide to the measurement of animal bones from archaeologi-

cal sites. Peabody Museum Bulletin. Peabody Museum of Archaeology and Ethnology 1:

103-136.

FepucciA, J. A. 1967. Ciconia maltha and Grus americana from the Upper Pliocene of Idaho.

Wilson Bulletin 79 (3): 316-318.

Gricorescu, D. & KEessLerR, E. 1977. The Middle Sarmatian avian fauna of south Dobrogea.

Revue roumaine de Géologie, Géophysique et Géographie (Géologie) 201: 93-108.

Harrison, C. J.O. 1974. A re-examination of material of the extinct marabou stork, Lepto-

ptilos falconeri: with descriptions of some new species. Bulletin of the British Ornith-

ologists’ Club 94 (2): 42-49.

Harrison, C. J. O. 1980. Fossil birds from afrotropical Africa in the collection of the British

Museum (Natural History). Ostrich 51 (2): 92-98.

Harrison, C. J. O. & WALKER, C. A. 1982. Fossil birds from the upper Miocene of northern

Pakistan. Tertiary Research 4 (2): 53-69.

HENDEY, Q. B. 1976. The Pliocene fossil occurrences in ‘E’ Quarry, Langebaanweg, South

Africa. Annals of the South African Museum 69 (9): 215-247.

HENDEY, Q. B. 198la. Geological succession at Langebaanweg, Cape Province, and global

events of the Late Tertiary. South African Journal of Science 77 (1): 33-38.

HENDEY, Q. B. 1981b. Palaeoecology of the Late Tertiary fossil occurrences in ‘E’ Quarry,

Langebaanweg, South Africa, and a reinterpretation of their geological context. Annals

of the South African Museum 84 (1): 1-104.

Hitt, A. & WALKER, A. 1979. A fossil marabou (Aves: Ciconiidae) from the Miocene

Ngorora Formation, Baringo District, Kenya. Netherlands Journal of Zoology 29 (2):

215-220.

Howarp, H. 1929. The avifauna of Emeryville Shellmound. University of California Publica-

tions in Zoology 32 (2): 301-394.

Howarpb, H. 1942. A review of the American fossil storks. Publications Carnegie Institution

of Washington (Contributions to Palaeontology no. 7) 530: 188-203.

Kau, M. P. 1972. A revision of the family Ciconiidae (Aves). Journal of Zoology, London

167 (4): 451-461.

LAMBRECHT, K. 1933. Handbuch der Palaeornithologie. Berlin: Gebrider Borntraeger.

LYDEKKER, R. 1884. Indian Tertiary and post-Tertiary Vertebrata. Siwalik birds. Memoirs of

the Geological Survey of India (10) 3 (4): 135-147.

Miter, L. H. 1910. Wading birds from the asphalt beds of Rancho La Brea. University of

California Publications. Bulletin of the Department of Geology 5 (30): 439-448.

Otson, S. L. 1983. Fossil seabirds and changing marine environments in the late Tertiary of

South Africa. South African Journal of Science 79 (10): 399-402.

Orson, S. L. 1984. A hamerkop from the Early Pliocene of South Africa (Aves: Scopidae).

Proceedings of the Biological Society of Washington 97 (4): 736-740.

Orson, S. L. 1985a. Early Pliocene Procellariiformes (Aves) from Langebaanweg, south-

western Cape Province, South Africa. Annals of the South African Museum 95 (3):

123-145.

Otson, S.L. 1985b. Early Pliocene ibises (Aves, Plataleidae) from south-western Cape Prov-

ince, South Africa. Annals of the South African Museum 97 (3): 57-69.

Oxson, S. L. 1986. The fossil record of birds. Jn: Farner, D. S., KING, J. R. & PARKES,

K. C. eds. Avian Biology 8: 79-238.

Ono, K. 1984. Fossil wading birds from northeastern Honshu, Japan. Memoirs of the

National Science Museum, Tokyo 17: 39-46.

FOSSIL STORK FROM LANGEBAANWEG 313

RicH, P. V. 1980. Preliminary report on the fossil avian remains from late Tertiary sediments

at Langebaanweg (Cape Province), South Africa. South African Journal of Science

76 (1): 166-170.

RicH, P. V. & HAARHorFF, P. J. 1985. Early Pliocene Coliidae (Aves, Coliiformes) from

Langebaanweg (Cape Province), South Africa. Ostrich 56 (1, 2, 3): 20-41.

Ricu, P. V. & vAN Tets, G. F. 1982. Fossil birds of Australia and New Guinea: their biogeo-

graphic, phylogenetic, and biostratigraphic input. Jn: Ricu, P. V. & THompson, E. M.

eds. The fossil vertebrate record of Australasia: 235-384. Clayton, Victoria: Monash

University.

S1BLEY, C. G. & AuHLQuist, J. E. 1985. The relationships of some groups of African birds,

based on comparisons of the genetic material, DNA. Jn: SCHUCHMANN, K-L. ed. Pro-

ceedings of the International Symposium on African vertebrates. Systematics, phylogeny

and evolutionary ecology: 114-161. Bonn: Zoologisches Forschungsinstitut und Museum

Alexander Koenig.

Simpson, G. G. 1971. Fossil penguin from the Late Cenozoic of South Africa. Science 171

(3976): 1144-1145.

Simpson, G. G. 1975. Notes on variation in penguins and on fossil penguins from the Plio-

cene of Langebaanweg, Cape Province, South Africa. Annals of the South African

Museum 69 (4): 59-72.

Simpson, G. G. 1979. A new genus of Late Tertiary penguin from Langebaanweg, South

Africa. Annals of the South African Museum 78 (1): 1-9.

STEHLIN, H. G. 1923. Die oberpliocaene Fauna von Senéze (Haute-Loire). Eclogae geologi-

cae Helvetiae 18 (2): 268-281.

Woop, D. S. 1984. Concordance between classifications of the Ciconiidae based on

behavioural and morphological data. Journal ftir Ornithologie 125 (1): 25-37.

i ‘ fi 4

io jes

es ait

si »

r % + 4)

7 ‘ n

= ees te es -

i = i Jig Bret py rf

L} i

4 -

= ”

‘ y

' a i

U cere at i ra

cae - 4 Lois

macy ve |

, - ¢ cs ’ |

: €. 5

: .

. z ' ‘

‘

* '

7 of '

a .

~ P,

tee 8

6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu-

larly Articles 22 and 51).

Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed

by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov.,

etc.

An author’s name when cited must follow the name of the taxon without intervening punctuation

and not be abbreviated; if the year is added, a comma must separate author’s name and year. The

author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is trans-

ferred from its original genus. The name of a subsequent user of a scientific name must be separated

from the scientific name by a colon.

Synonymy arrangement should be according to chronology of names, i.e. all published scientific

names by which the species previously has been designated are listed in chronological order, with all

references to that name following in chronological order, e.g.:

Family Nuculanidae

Nuculana (Lembulus) bicuspidata (Gould, 1845)

Figs 14-15A

Nucula (Leda) bicuspidata Gould, 1845: 37.

Leda plicifera A. Adams, 1856: 50.

Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b).

Nucula largillierti Philippi, 1861: 87.

Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9.

Note punctuation in the above example:

comma separates author’s name and year

semicolon separates more than one reference by the same author

full stop separates references by different authors

figures of plates are enclosed in parentheses to distinguish them from text-figures

dash, not comma, separates consecutive numbers.

Synonymy arrangement according to chronology of bibliographic references, whereby the year is

placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable.

In describing new species, one specimen must be designated as the holotype; other specimens

mentioned in the original description are to be designated paratypes; additional material not regarded

as paratypes should be listed separately. The complete data (registration number, depository, descrip-

tion of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.:

Holotype

SAM-—A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach, Port Eliza-

beth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973.

Note standard form of writing South African Museum registration numbers and date.

7. SPECIAL HOUSE RULES

Capital initial letters

(a) The Figures, Maps and Tables of the paper when referred to in the text

e.g. ‘.. . the Figure depicting C. namacolus .. .’: “. . . in C. namacolus (Fig. 10). . .”

(b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by

initials or full names

e.g. Du Toit but A.L. du Toit; Von Huene but F. von Huene

e.g. Therocephalia, but therocephalian

Punctuation should be loose, omitting all not strictly necessary

Reference to the author should preferably be expressed in the third person

Roman numerals should be converted to arabic, except when forming part of the title of a book or

article, such as

‘Revision of the Crustacea. Part VIII. The Amphipoda.’

Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial

capital letter, provided the same generic name is used consecutively. The generic name should

not be abbreviated at the beginning of a sentence or paragraph.

Name of new genus or species is not to be included in the title; it should be included in the abstract,

counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts.

P. J. HAARHOFF

A NEW FOSSIL STORK (AVES, CICONITDAE)

FROM THE LATE TERTIARY OF

LANGEBAANWEG, SOUTH AFRICA

_ VOLUME 97 PART 12 JULY 1988 | ISSN 0303-2515

OF THE SOUTH AF RICAN

MUSEUM

CAPE TOWN

INSTRUCTIONS TO AUTHORS

1. MATERIAL should be original and not published elsewhere, in whole or in part.

2. LAYOUT should be as follows:

(a) Centred masthead to consist of

Title: informative but concise, without abbreviations and not including the names of new genera or species

Author’s(s’) name(s)

Address(es) of author(s) (institution where work was carried out)

Number of illustrations (figures, enumerated maps and tables, in this order)

(b) Abstract of not more than 200 words, intelligible to the reader without reference to the text

(d) Introduction

(e) Subject-matter of the paper, divided into sections to correspond with those given in table of contents

(f) Summary, if paper is lengthy

(g) Acknowledgements

(h) References

(i) Abbreviations, where these are numerous.

3. MANUSCRIPT, to be submitted in triplicate, should be typewritten and neat, double spaced with

3 cm margins all round. First lines of paragraphs should be indented. Tables and a list of captions for

illustrations should be typed separately, their positions indicated in the text. All pages should be num-

bered consecutively.

Major headings of the paper are centred capitals; first subheadings are shouldered small capitals;

second subheadings are shouldered italics; third subheadings are indented, shouldered italics. Further

subdivisions should be avoided, as also enumeration (never roman numerals) of headings and

abbreviations.

Footnotes should be avoided unless they are short and essential.

Only generic and specific names should be underlined to indicate italics; all other marking up

should be left to editor and publisher.

4. ILLUSTRATIONS should be reducible to a size not exceeding 12 xX 18 cm (19 cm including

caption); the reduction or enlargement required should be indicated (and preferably uniform); orig-

inals larger than 35 X 47 cm should not be submitted; photographs should be rectangular in shape and

final size. A metric scale should appear with all illustrations, otherwise magnification or reduction

should be given in the caption; if the latter, then the final reduction or enlargement should be taken

into consideration.

All illustrations, whether line drawings or photographs, should be termed figures (plates are not

printed; half-tones will appear in their proper place in the text) and numbered in a single series. Items

of composite figures should be designated by capital letters; lettering of figures is not set in type and

should be in lower-case letters. If Letraset is used authors are requested to use Helvetica-style letter-

ing, if possible.

The number of the figure should be lightly marked in pencil on the back of each illustration.

5. REFERENCES cited in text and synonymies should all be included in the list at the end of the

paper, using the Harvard System (ibid., idem, loc. cit., op. cit. are not acceptable):

(a) Author’s name and year of publication given in text, e.g.:

‘Smith (1969) describes...”

‘Smith (1969: 36, fig. 16) describes...’

‘As described (Smith 1969a, 1969b; Jones 1971)’

“As described (Haughton & Broom 1927)...’

‘As described (Haughton ef al. 1927)...’

Note: no comma separating name and year

pagination indicated by colon, not p.

names of joint authors connected by ampersand

et al. in text for more than two joint authors, but names of all authors given in list of references.

(b) Full references at the end of the paper, arranged alphabetically by names, chronologically within

each name, with suffixes a, b, etc., to the year for more than one paper by the same author in

that year, e.g. Smith (1969a, 1969b) and not Smith (1969, 1969a).

For books give title in italics, edition, volume number, place of publication, publisher

For journal article give title of article, title of journal in italics (according to the World list of scientific periodicals. 4th ed.

London: Butterworths, 1963), series in parentheses, volume number, part number in parentheses, pagination (first and

last pages of article).

Examples (note capitalization and punctuation)

BuLLouGu, W. S. 1960. Practical invertebrate anatomy. 2nd ed. London: Macmillan.

FiscHER, P. H: 1948. Données sur la résistance et de la vitalité des mollusques. Journal de conchyliologie 88 (3): 100-140.

FiscHer, P. H., Duvat, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines. Archives de zoologie

expérimentale et générale 74 (33): 627-634.

Koun, A. J. 1960a. Ecological notes on Conus (Mollusca: Gastropoda) in the Trincomalee region of Ceylon. Annals and

Magazine of Natural History (13) 2 (17): 309-320.

Koun, A. J. 19606. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean. Bulletin of

the Bingham Oceanographic Collection, Yale University 17 (4): 1-51.

THIELE, J. 1910. Mollusca. B. Polyplacophora, Gastropoda marina, Bivalvia. In: ScHuLTZzE, L. Zoologische und anthro-

pologische Ergebnisse einer Forschungsreise im westlichen und zentralen Stid-Afrika ausgeftihrt in den Jahren

1903-1905 4 (15). Denkschriften der medizinisch-naturwissenschaftlichen Gesellschaft zu Jena 16: 269-270.

(continued inside back cover)

ANNALS OF THE SOUTH AFRICAN MUSEUM

ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

Volume 97 Band

July 1988 Julie

A REDESCRIPTION OF THE SOFT CORAL

ALCYONIUM VALDIVIAE KUKENTHAL, 1906,

WITH THE DESCRIPTION OF A NEW

SPECIES OF LITOPHYTON FORSKAL, 1775,

FROM SOUTHERN AFRICA

(OCTOCORALLIA, ALCYONACEA)

J. VERSEVELDT & GARY C. WILLIAMS

Cape Town Kaapstad

The ANNALS OF THE SOUTH AFRICAN MUSEUM

are issued in parts at irregular intervals as material

becomes available

Obtainable from the South African Museum, P.O. Box 61, Cape Town 8000

Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

word uitgegee in dele op ongereelde tye na gelang van die

beskikbaarheid van stof

Verkrygbaar van die Suid-Afrikaanse Museum, Posbus 61, Kaapstad 8000

OUT OF PRINT/UIT DRUK

il, HOS, 5-8), 202, 4S, &, to.i,), SOS, 5, 7-9),

A, io.t.), HO), 8, O02, 7), MCS),

GED, 5, Tint), WAGD), 15S), VAC), 27, SGE8), BAG), 38, 352), 45)

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 095 8

Printed in South Africa by In Suid-Afrika gedruk deur

The Rustica: Press, Pty lide Die Rustica-pers, Edms., Bpk.,

Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap

A REDESCRIPTION OF THE SOFT CORAL ALCYONIUM VALDIVIAE

KUKENTHAL, 1906, WITH THE DESCRIPTION OF A NEW SPECIES OF

LITOPHYTON FORSKAL, 1775, FROM SOUTHERN AFRICA

(OCTOCORALLIA, ALCYONACEA)

*J. VERSEVELDT

Weteringpark 7—48, 8025 AM Zwolle, The Netherlands

GARY C. WILLIAMS

Department of Marine Biology, South African Museum, Cape Town

(With 8 figures)

[MS accepted 1 December 1987]

ABSTRACT

Two presumably endemic species of soft corals are described from the shallow sublittoral of

southern Africa. A redescription of Alcyonium valdiviae Kikenthal, 1906 (family Alcyoniidae),

is presented, with a brief assessment of its variability, as well as a comparison with related

southern African taxa. Previous authors have confused this species with Cladiella pachyclados

(Klunzinger, 1877). The present re-evaluation, based on recently collected material, confirms

that C. pachyclados is not part of the fauna of the south coast of South Africa.

A recently discovered species of the genus Litophyton Forskal, 1775 (family Nephtheidae),

is here described as Litophyton liltvedi sp. nov., and is compared with other sympatric and

superficially similar species. This description represents the first record of the genus for the

subcontinent.

CONTENTS

PAGE

NOWGROLIUNC TON oc. Ba. srt ele oi Rais OURO nR etn aa ea ee SUS)

SVSUCilati CAC COUN ey ee ise AAI 4S sus waitlndn enies 316

ANCKNOWIEGPENIEMSH a qari n= ame si he nae ok eitnad ages 328

IRYSNTSILCMIGES 5, che dene tye rere SRE ee o| 6o8)

INTRODUCTION

Recent exploration of the shallow sublittoral of South Africa by means of

SCUBA and dredge has yielded many new species and records of soft corals as

well as newly acquired material of many previously described species that have

not been collected for the past 60—90 years.

From Dr G. C. Williams, Curator of Lower Invertebrates, South African

Museum, Cape Town, Dr J. Verseveldt received some alcyonacean specimens for

examination. Among these were five colonies of a soft coral, similar specimens of

* Deceased, manuscript published posthumously.

Ann. S. Afr. Mus. 97 (12), 1988: 315-328, 8 figs.

316 ANNALS OF THE SOUTH AFRICAN MUSEUM

which were previously identified as Alcyonium pachyclados Klunzinger, 1877, by

Hickson (1900: 72) and J. Stuart Thomson (1910: 570; 1921: 50). The generic

designation for A. pachyclados is now recognized as Cladiella Gray, 1869, based

on the form and distribution of sclerites (Tixier-Durivault 1966: 50). It was found

that the recently collected material mentioned above is conspecific with

Alcyonium valdiviae Kukenthal, 1906.

Another colony received represents a new species of Litophyton Forskal,

1775, an Indo-Pacific genus previously unrecorded from the coasts of southern

Africa.

Since the death of Dr Verseveldt on 29 March 1987 precluded the possibility

of the referees’ comments being incorporated into the original manuscript,

Dr J. C. den Hartog (Curator of Coelenterata at the Rijksmuseum van

Natuurlijke Historie, Leiden), long-time associate and friend of Dr Verseveldt,

suggested that Dr Williams make the necessary revisions of the manuscript and

act as co-author. This has been effected, together with the redrawing of the

figures portraying the polyps and sclerites, and the inclusion of scanning electron

micrographs as well as a map showing the geographic distribution of the two

species.

SYSTEMATIC ACCOUNT

Family Alcyoniidae Lamouroux, 1812

Alcyonium Linnaeus, 1758

Alcyonium valdiviae Kiikenthal, 1906

Figs 1, 2A, C-E, 3

Alcyonium valdiviae Kikenthal, 1906: 42, pl. 3 (fig. 11), pl. 8 (figs 39-41).

Alcyonium pachyclados (non Klunzinger, 1877): Hickson, 1900: 72. J. Stuart Thomson, 1910:

570, pl. 2 (fig. 14), pl. 4 (figs 33, 34); 1921: 155-156, pl. 5 (figs 6-8).

Material

SAM-—H3347, 1 colony (Fig. 1A, B), reef 2 km off Bird Rock, Algoa Bay,

33°50'S 25°40'E, depth 15m, SCUBA, 17 May 1984, coll. G. C. Williams.

SAM-—H3661, 1 colony, Gonubie, Eastern Cape Province, Station XX38,

32°51,2'S 28°02,8'E, depth 30 m, dredge, 17 July 1981, coll. G. C. Williams

(RV Meiring Naude). SAM—H3384, 1 colony (Fig. 1D), off Gans Bay, Western

Cape Province, 34°36,6’S 19°12,6’E, depth 78 m, 11 October 1983, coll. Sea

Fisheries Research Institute. SAM—H3732, 1 colony (Fig. 1C), Hottentots

Huisie, Cape Peninsula, 33°59’S 18°21’E, depth 14 m, SCUBA, 22 March 1984,

coll. G. C. Williams. SAM—H3409, 1 colony (Fig. 1E), off Danger Point, Cape

Province, 34°40’S 19°17’E, depth 42m, SCUBA, 13 April 1984, coll.

W. R. Liltved.

ALCYONACEAN CORALS FROM SOUTH AFRICA Sigh

"tg ” AP YD

ee ©

hai Be iB

wT8

Fig. 1. Alcyonium valdiviae Kikenthal, 1906. A-B. SAM-H3347.

C. SAM-H3732. D. SAM-H3384. E. SAM—H3409. Scale = 40 mm.

318 ANNALS OF THE SOUTH AFRICAN MUSEUM

Description

External features (Figs 1A—E, 2A)

The colony (SAM-—H3347) (Fig. 1A, B) measures 53 mm in total height, of

which the stalk comprises 10 mm. The latter is laterally flattened and measures

22 mm at its greatest width. The polyparium 1s also slightly flattened. It consists of

numerous closely set, round, sometimes rather conical, small lobes, 2—4 mm in

diameter; these lobes are united into groups at the ends of twigs.

Many polyps are extended. Owing to their white colour they contrast

distinctly with the dark-red colour of the polyparium. The anthocodiae protrude

above the surface of the lobes for a distance of up to 0,20 mm (Fig. 2A). The

tentacles are 0,40—0,50 mm long. At their bases they are swollen, about 0,13 mm

wide, tapering distally to a sharp point. On either side they bear one row of

10-15 thin pinnules, 0,12—0,14 mm long. The polyps are entirely devoid of

sclerites.

Sclerites (Figs 2C-—D, 3A-I)

The rigid surface layer of the polyparium and stalk is densely packed with

small capstans, which are primarily eight-radiates and six-radiates. The length of

these sclerites is usually 0,05 mm and does not exceed 0,07 mm. There are also

some four-radiates, as well as some three-radiates, 0,025—0,04 mm in length. The

interior of the colony is devoid of sclerites.

Colour

The whole colony is orange-red, the polyps are white.

Variability

The five colonies differ in shape and in colour. The colony depicted in

Figure 1C has a long, broad stalk. The specimens shown in Figure 1D—E have

short stalks. The lobes may be short and conical, long and conical, or finger-

shaped (Fig. 1C—E). The colony represented in Figure 1C is orange distally, but

towards the base is light pink. The colony shown in Figure 1D is yellowish-grey;

that in Figure 1E is pinkish-beige. The specimen from Gonubie (SAM-—H3661—

not illustrated) is bright yellow; this colony has a few extended polyps.

Distribution (Fig. 2E)

South coast of South Africa.

Remarks

It is unknown to the authors whether Ktkenthal’s type specimen is still kept

in some museum or collection, so it could not be investigated.

Some of the specimens may superficially resemble members of the genus

Cladiella but, for the following reasons, cannot be included in that genus:

Cladiella species contain double heads (see Bayer et al. 1983, figs 159, 160)

ALCYONACEAN CORALS FROM SOUTH AFRICA 319

East LondonsA

£\

Fig. 2. A. Alcyonium valdiviae Kikenthal, 1906. A single polyp, 0,6 mm in _ height.

B. Litophyton liltvedi sp. nov. A terminal lappet of four mature and two budding polyps, showing

arrangement and relative density of sclerites in the polyp walls and tentacles. Total height of figure

represents 3,3 mm. C-—D. Sclerites of Alcyonium valdiviae. C. Sclerites from the surface of a

lobe from the polyparium. D. Sclerites from the surface of the stalk. Total length of scale bar for

C and D=0,1 mm. E. Map of southern Africa showing collecting stations for Alcyonium

valdiviae (A) and Litophyton liltvedi (@). Arrows indicate type localities.

320 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 3. Alcyonium valdiviae Kiikenthal, 1906. Scanning electron micrographs of sclerites.

A-D. Sclerites from the surface of a lobe from the polyparium. A. 0,053 mm. B. 0,042 mm.

C. 0,053 mm. D.0,048 mm. E-I. Sclerites from the surface of the stalk. E. 0,049 mm.

F. 0,037 mm. G. 0,048 mm. H. 0,058 mm. I. 0,052 mm.

ALCYONACEAN CORALS FROM SOUTH AFRICA By

without terminal clusters of warts; the present specimens have capstans (see

Bayer et al. 1983, figs 148-150). In all Cladiella species the interior contains

sclerites; these are absent in the present specimens. Cladiella species have typical

finger-biscuit-like sclerites in the surface layer of the lobes and/or in the polyps

and/or in the tentacles; these are absent in the present specimens.

Hickson (1900) and J. Stuart Thomson (1910, 1921) referred the specimens

examined by them to Alcyonium pachyclados Klunzinger, 1877 (a species that

correctly belongs to the genus Cladiella). Although accurate in assigning the

generic name Alcyonium to their material, they were incorrect as regards the

specific identity for the above-stated reasons and since Cladiella pachyclados

possesses coenenchymal sclerites that are much larger (up to 0,12 mm in length).

Alcyonium rutilum (Tixier-Durivault, 1954) from northern Natal closely

resembles A. valdiviae. The two species can be differentiated as follows:

A. valdiviae varies in colour from white to pink, yellow, orange, or orange-red,

and possesses sclerites of capstans mostly 0,05—0,07 mm in length; A. rutilum, on

the other hand, is deep wine-red or dark red and the sclerites (also capstans) do

not exceed 0,04 mm in length. Alcyonium rutilum is known only from the

holotype. Additional material is necessary to assess the extent of the intraspecific

variation and to determine if it is truly distinct from A. valdiviae.

Family Nephtheidae Gray, 1862

Litophyton Forskal, 1775

Litophyton liltvedi sp. nov.

Figs 2B, E, 4-8

Material

SAM-—H3393, holotype, off Danger Point, Cape Province, 34°38’S 19°20'E,

depth 31 m, SCUBA, 10 April 1984, coll. W. R. Liltved.

Description

External features (Figs 2B, 4A—-B)

The roughly circular, more or less flat, soft colony has diameters of 80 mm

and 90 mm. On one side, where the excentrically situated stalk is found, the

thickness of the colony is 35 mm; on the opposite side it is 10 mm thick. The

holdfast is oval in shape, with diameters of 12-18 mm and 40 mm. (In Fig. 4B it is

the dark patch at the lower side of the photograph.) From the stalk two stems

arise: a short one, which passes into a small part of the polyparium, and a larger

one, which is much branched and bears a larger part of the polyparium.

The upper side of the colony is covered with lobes. The larger ones are

10-15 mm wide. The extended polyps measure 1,60—2,50 mm in total height, the

anthocodiae are 0,95—1,30 mm long (Fig. 2B). The extended tentacles, which are

0,60—0,70 mm long, have a single row of 5—7 pinnules on each side.

322 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 4. Litophyton liltvedi sp. nov. SAM—H3393, holotype. A. Seen from above. B. Seen

from below. Scale = 40 mm.

ALCYONACEAN CORALS FROM SOUTH AFRICA 38)

Fig. 5. Litophyton liltvedi sp. nov., holotype. A. Sclerites from the polyp wall. B. Sclerites

from the tentacles. Entire scale bar represents 0,2 mm.

Sclerites (Figs 5—8)

The tentacles contain many spiny and wed spindles 0,05—0,16 mm in length

(Figs 5B, 7A-F); in the base of the rachis the spindles are longitudinally

arranged, while distally they are shorter and are arranged more or less en

chevron.

324 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 6. Litophyton liltvedi sp. nov., holotype. A. Sclerites from the surface of a lobe from the

polyparium. B. Sclerites from the surface of the stalk. Entire length of scale bar represents

0,15 mm.

The anthocodial wall contains spiny spindles (Figs 5A, 7G—L), usually more

or less longitudinally placed, but they often lie crisscross, and thus do not present

a typical crown-and-points arrangement (Fig. 2B). Proximally their length is

Q,20—0,28 mm; towards the tentacles they are shorter, about 0,05—0,10 mm.

ALCYONACEAN CORALS FROM SOUTH AFRICA 325

Fig. 7. Litophyton liltvedi sp. nov. Scanning electron micrographs of sclerites. A-—F. Sclerites from the

tentacles. A. 0,074 mm. B. 0,120 mm. C. 0,063 mm. D. 0,080 mm. E. 0,105 mm. F. 0,108 mm.

G-L. Sclerites from the polyp wall. G.0,212 mm. H.0,163 mm. I.0,150mm. J. 0,150 mm.

K. 0,155mm. L. 0,200 mm.

The surface layer of the polyparium and stalk contains many tiny double stars

(usually 0,075 mm long, but ranging between 0,025 mm and 0,12 mm) with a

proportionally long ‘waist’ (Figs 6A, B, 8E—N). In addition, in the surface layer

of the terminal branches there are some spiny spindles, 0,08—0,20 mm long

(Figs 6A, 8A—D). The two types of sclerites are not particularly densely set; in

the interior of the colony sclerites are very scarce or absent.

Colour

In alcohol the colony is cream.

ANNALS OF THE SOUTH AFRICAN MUSEUM

ALCYONACEAN CORALS FROM SOUTH AFRICA 327)

Etymology

Litophyton liltvedi is named for Mr W. R. Liltved, who collected the

specimen on which this description is based.

Distribution (Fig. 2E)

South coast of South Africa.

Remarks

At first sight the colony resembles the type specimen of the alcyoniid

Alcyonium fauri, as illustrated by J. Stuart Thomson (1910, pl. 1 (fig. 5)).

However, in important respects the present specimen differs from the type of that

species. According to Thomson the colony of A. fauri is encrusting, fairly hard,

and the lobes are closely adjacent. The colony of L. liltvedi sp. nov. is not

encrusting, and it is not hard but weak and flexible. The lobes do not arise from a

lamella but rather from a fleshy polyparium, and the latter is provided with few

rather than numerous spicules. Finally, in A. fauri the sclerites are up to 0,12 x

0,09 mm. Unfortunately, Thomson’s figures of these sclerites (1910, pl. 4

(fig. 44)) are far from clear; with difficulty they can be discerned as double heads

or dumb-bells with a very short waist. They are totally different from the double

stars in L. liltvedi sp. nov. Examination of recently collected material referable to

A. fauri has shown that a variety of coarsely tuberculated coenenchymal sclerites

are evident, including robust spindles, wart clubs, double heads, and eight-

radiates. Double stars are absent altogether.

The only other southern African soft coral that resembles L. Jiltvedi sp. nov.

in its cauliflower-like appearance, is the nephtheid Capnella thyrsoidea Verrill,

1865. The two are distinguished by the fact that the genus Capnella possesses

many coenenchymal sclerites that are leaf clubs, whereas Litophyton is altogether

lacking in leaf clubs.

The new species is referred to the genus Litophyton on the following

grounds: (a) the colony branches repeatedly, and (b) the polyps are not

retractile, they lack a supporting bundle, they are arranged in terminal catkins,

and the anthocodial sclerites are not leaf clubs.

Small sclerites, but a little larger than the tiny double stars in the holotype,

also occur in other Litophyton species, e.g. L. stuhlmanni (May, 1899) and

L. confertum Kikenthal, 1903.

Fig. 8. Litophyton liltvedi sp. nov. Scanning electron micrographs of sclerites. A—H. Sclerites

from the surface of a lobe from the polyparium. A. 0,210 mm. B. 0,155 mm. C. 0,180 mm.

D. 0,125 mm. E. 0,085 mm. F. 0,090 mm. G. 0,083 mm. H. 0,075 mm. I-N. Sclerites from

the surface of the stalk. I.0,090 mm. J.0,075 mm. K.0,078 mm. L. 0,065 mm.

M. 0,090 mm. N. 0,090 mm.

328 ANNALS OF THE SOUTH AFRICAN MUSEUM

ACKNOWLEDGEMENTS

I thank Dr G. C. Williams, Department of Marine Biology, South African

Museum, Cape Town, for the loan of the material; he also put me on the track of

Alcyonium valdiviae. As always I am indebted to my friends Mr G. J. Vrijmoeth,

for taking the photographs of the corals, and Mr W. ter Spill, for reading the text

GEA):

I am grateful to Klaus Schultes and Dane Gerneke of the Electron

Microscope Unit, University of Cape Town, for technical assistance with the

preparation of scanning electron micrographs; Michelle van der Merwe and

Elizabeth Hoenson, South African Museum, for assistance with the preparation

of Figures 3, 7 and 8, and for curatorial work; and Elizabeth Louw, Editor, South

African Museum, for comments and suggestions (G.C.W.).

REFERENCES

Bayer, F. M., GrassHorr, M. & VERSEVELDT, J. 1983. Illustrated trilingual glossary of

morphological and anatomical terms applied to Octocorralia. Leiden: E. J. Brill & Dr

W. Backhuys.

Gray, J. E. 1869. Notes on the fleshy alcyonoid corals (Alcyonium, Linn., or Zoophytaria

carnosa). Annals and Magazine of Natural History (4) 3: 117-131.

Hickson, S. J. 1900. The Alcyonaria and Hydrocorrallinae of the Cape of Good Hope. Marine

Investigations in South Africa 1 (5): 67-96.

KLUNZINGER, C. B. 1877. Die Korallthiere des Rothen Meeres, I. Die Alcyonarien und

Malacodermen. Berlin: Verlag der Gutmannschen Buchhandlung.

KUKENTHAL, W. 1903. Versuch einer Revision der Alcyonarien. II. Die Familie der

Nephthyiden. le Teil. Zoologische Jahrbiicher, Zeitschfrift fiir Systematik, Geographie und

Biologie der Thiere 19 (1): 99-172.

KUKENTHAL, W. 1906. Alcyonacea. Wissenschaftliche Ergebnisse der Deutschen Tiefsee-

Expedition ‘Valdivia’ 13 (1): 1-111.

May, W. 1899. Beitrage zur Systematik und Chorologie der Alcyonaceen. Jenaische Zeitschrift

ftir Naturwissenschaft 33: 1-180.

THOMSON, J. Stuart. 1910. Alcyonacea. Transactions of the Royal Society of Edinburgh

47 (3): 549-589.

THOMSON, J. STUART. 1921. South African Alcyonacea. Transactions of the Royal Society of

South Africa 9 (2): 149-175.

TIxIER-DuRIVAULT, A. 1954. Les Octocoralliaires d’Afrique du Sud (I. Alcyonacea). Bulletin

du Muséum national da histoire naturelle (2) 26 (2): 261-268.

TrxiER-DuRIVAULT, A. 1966. Octocoralliaires. Faune de Madagascar 21: 1—456.

VERRILL, A. D. 1865. Synopsis of the polyps and corals of the North Pacific Exploring

Expedition, under Commodore C. Ringgold and Captain John Rodgers, U.S.N., from 1853

to 1856. Collected by Dr Wm. Stimpson, naturalist to the Expedition. With descriptions of

some additional species from the west coast of North America. Proceedings of the Essex

Institute Salem 4: 181-196.

6. SYSTEMATIC papers must conform to the International code of zoological nomenclature (particu-

larly Articles 22 and 51).

Names of new taxa, combinations, synonyms, etc., when used for the first time, must be followed

by the appropriate Latin (not English) abbreviation, e.g. gen. nov., sp. nov., comb. nov., syn. nov.,

ete.

An author’s name when cited must follow the name of the taxon without intervening punctuation

and not be abbreviated; if the year is added, a comma must separate author’s name and year. The

author’s name (and date, if cited) must be placed in parentheses if a species or subspecies is trans-

ferred from its original genus. The name of a subsequent user of a scientific name must be separated

from the scientific name by a colon.

Synonymy arrangement should be according to chronology of names, i.e. all published scientific

names by which the species previously has been designated are listed in chronological order, with all

references to that name following in chronological order, e.g.:

Family Nuculanidae

Nuculana (Lembulus) bicuspidata (Gould, 1845)

Figs 14-15A

Nucula (Leda) bicuspidata Gould, 1845: 37.

Leda plicifera A. Adams, 1856: 50.

Laeda bicuspidata Hanley, 1859: 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b).

Nucula largillierti Philippi, 1861: 87. }

Leda bicuspidata: Nicklés, 1950: 163, fig. 301; 1955: 110. Barnard, 1964: 234, figs 8-9.

Note punctuation in the above example:

comma separates author’s name and year

semicolon separates more than one reference by the same author

full stop separates references by different authors

figures of plates are enclosed in parentheses to distinguish them from text-figures

dash, not comma, separates consecutive numbers.

Synonymy arrangement according to chronology of bibliographic references, whereby the year is

placed in front of each entry, and the synonym repeated in full for each entry, is not acceptable.

In describing new species, one specimen must be designated as the holotype; other specimens

mentioned in the original description are to be designated paratypes; additional material not regarded

as paratypes should be listed separately. The complete data (registration number, depository, descrip-

tion of specimen, locality, collector, date) of the holotype and paratypes must be recorded, e.g.:

Holotype

SAM-A13535 in the South African Museum, Cape Town. Adult female from mid-tide region, King’s Beach, Port Eliza-

beth (33°51’S 25°39’E), collected by A. Smith, 15 January 1973.

Note standard form of writing South African Museum registration numbers and date.

7. SPECIAL HOUSE RULES

Capital initial letters

(a) The Figures, Maps and Tables of the paper when referred to in the text

e.g. ‘. . . the Figure depicting C. namacolus .. .’: ‘. . . in C. namacolus (Fig. 10) .. .’

(b) The prefixes of prefixed surnames in all languages, when used in the text, if not preceded by

initials or full names

e.g. Du Toit but A.L. du Toit; Von Huene but F. von Huene

e.g. Therocephalia, but therocephalian

Punctuation should be loose, omitting all not strictly necessary

Reference to the author should preferably be expressed in the third person

Roman numerals should be converted to arabic, except when forming part of the title of a book or

article, such as

‘Revision of the Crustacea. Part VIII. The Amphipoda.’

Specific name must not stand alone, but be preceded by the generic name or its abbreviation to initial

capital letter, provided the same generic name is used consecutively. The generic name should

not be abbreviated at the beginning of a sentence or paragraph.

Name of new genus or species is not to be included in the title; it should be included in the abstract,

counter to Recommendation 23 of the Code, to meet the requirements of Biological Abstracts.

J. VERSEVELDT & GARY C. WILLIAMS

A REDESCRIPTION OF THE SOFT CORAL

ALCYONIUM VALDIVIAE KUKENTHAL, 1906, WITH

THE DESCRIPTION OF A NEW SPECIES OF

LITOPHYTON FORSKAL, 1775, FROM SOUTHERN

AFRICA (OCTOCORALLIA, ALCYONACEA)

NOILALIL

LIBRAR

NOILN LIS

SHLINS S3IYVYEGIT LIBRARIES SMITHSONIAN

AN — w —

La 0 Ewes

~ ze Vy

~~ a w =

SONIAN INSTITUTION NOILALILSNI NVINOSHLIWS

SHLINS S3INVUGIT LIBRARIES SMITHSONIAN

SMITHSONIAN

NWINOSH.LIWS

SMITHSONIAN

YS 5 3 = Sy

SONIANT INSTITUTION | NOLLA.LUSNI”NVINOSHLINS

We E

is Z

SHLINS SSIYVYUEIT LIBRARIES

NVINOSHLINS S3IuVvadl

NVINOSHLINS

SMITHSONIAN

ONIAN INSTITUTION NOILNLILSNI_ NVINOSHLIWS

NOILNLILSNI

LIBRARIES

NOILNLILSNI

HLINS SJIYVYUEIT LIBRARIES SMITHSONIAN

~“

“iff

INSTITUTION

S3iyuvyugit

INSTITUTION

AG .

“

NVINOSHLIWS

SMITHSONIAN

4UIWS S3INVYSIT LIBRARIES

NOILNLILSNI

INIAN

1_ LIBRARIES SMITHSONIAN

TtRBRARTES

INSTITUTION NOILNLILSNI

LIBRAR

NOILALI

INSTITUTION NOILOLILSNI

sg1uvugin

INSTITUTION

SSIUVYUEIT LIBRA

NVINOSHIIWS

SMITHSONIAN

INSTITUTION NOILNLIL

LIBRARIES

S3IYVUYEIT LIBRARIES

INSTITUTION NOILALILSNI

INSTITUTION NOILNLILSNI

Saluvadit

INSTITUTION NOILNILILSNI

SMITHSONIAN

NVINOSHLIWS

SMITHSONIAN

NOILNLILSNI

INSTITUTION NOILNLILSNI

S2IUVYdIT LIBRARI

INSTITUTION

ONIAN INSTITUTION NOILALILSNI NVINOSHLINS S3INVYUGIT LIBRA

NVINOSHLINS SJIiuyvuslt LIBRARIES

NVINOSHLIWS

SMITHSONIAN

INSTITUTION NOILOALIL

NOILNLILSNI

1 LIBRARIES

N NOILALILSNI

Uppy

TLIBRARLES

NVINOSHLINS SJ!IYVUdIT LIBRAR IES

ac 25

: ‘ek

cO . = |

ck ro)

a a

$3)

z 2

: :

a, 5

0] —

= an

igp|

w” =

= Z

=i S

n a

O ae

= Ee

2 a

NVINOSHLINS S3I

SMITHSONIAN _INS1

SFIUWUGIT LIBRARIES

NVINOSHLINS Sal

NVINOSHLIWS

LIBRARIES SMITHSONIAN INST

NVINOSHLINS S31!

; =

NS fo)

YY: i

4 vo Wo 5

ba =

\ =

SMITHSONIAN INS’

SMITHSONIAN

Jul?

NVINOSHLIWS S3JI

SMITHSONIAN _INS)

NOILALIL

LIBRAR!

NOILNLIL

LIBRARI

NOLLM LIL

LIBRARI

LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3!IYVYUGIT LIBRA

a = ts = Fis Zz rm

A A

> =; > ra > 4 = ie

mn ” m n ma ” a

= z m m

wn = 7, = n fa ee wn

| NOILNLILSNI S3!YVYEIT LIBRARIES SMITHSONIAN INSTITUTION NOILNI

w Zz Ww = oor i?) z= > Ww

. & ia = Pe eS < w . =

EE Z : 2 WS 5 2S 5

‘> = > = eae = = : >

ae w“) z= wY A ese W - z ;

| poMITHSONIAN INSTITUTION NOLLALILSNI NVINOSHLINS S3IiuvugIyT LIBRA

> 3 3 pee 2) >

» za <a is AS ~

~ | “3 rz = XY z “

= < 3 = < 5 WA < a

= = = a = NS = 2

fe) = , Oo = ra) & S

| ke a za = Fd Ny z

Wm NOILNLILSNI NVINOSHLINS S3!1YVUdIT LIBRARIES SMITHSONIAN_INSTITUTION NOILNL

i ie = Bs = : a S

{ — ow — wo ae w -_—

y ae = ae 9 ~~ x5 ~

E = ANE : = = :

\ = as INES jew ss a Ee a

| LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS SAldvuall LIBRA

; ¥ = Z . 2 2 & < oy

= z 4 4 z NN

-: S g z 5 N

Oo cE eS O zs Ys

Zz E 2, = = aN

> ; = ~~

2 5 5 3 e >

NVINOSHLIWS LIBRARIES

SMITHSONIAN INSTITUTION NOILN

NOILNLILSNI

NOILALILSNI

NOILNLILSNI

LIBRARIES

LIBRARIES SMITHSONIAN

INSTITUTION NOILOLILSNI

LIBRARIES SMITHSONIAN NVINOSHLINS S3IYVYUEIT LIBRA

SUIYVUGIT LIBRARIES

NVINOSHLINS SAluvddlT LIBRARIES SMITHSONIAN

ia = z : = rc

1 9 = = ss faa =

| =. > eS 5 2

( =e LS ie

7 b b = =

\ a Zz z > m

Ww = = aaa uw

| NOILNLILSN! SSIYVYNEIT LIBRARIES INSTITUTION NOILNI

ig z Z gn ae z 2

£E = SVE ope = =

iE : 2 Gay = \\N : S

i be eS TEOONIEN 2 INSTITUTION NOILNLILSNI NVINOSHLINS =S3 iuvydag as

| on zs w

a z be rf AY, ‘ a. ow ef aa =

se “ YZ. c a Cc foe =

Bz mh S a 2 diane S

_NOILNLILSNI NVINOSHLINS SJ IyVvu_d a ae LIBRARI ES SMITHSON AN_INSTITUTION a NOILAL

Ayu ws

ae ae ie cis ae we . SMITHSONIAN INSTITUTION LIBRARIES

APM eh wt B 4 Mg ; ‘ i :

Wohin OTN) EB So rr 3 ; }

3 9088 01206 6940

DEM ch Fit pret yisy

Va aie ses Y ¥

i a ere eS

are Rs

tin Cops Ms See AON Ty

iy Geak

te tea

NFia 8 SRO, Hy Re isk Ray's

VR peer tees

YL Mange

We en teek

Uae

fs Ge cee meat Rig¥e. Rane PON TRAD Cp BHT g

RRS WT OM FN TA Qe A

AFR FREER a TN PTI

UME WAP TEAR UY Sieg eau yee?

> Ue Ee Ey

BAe Way ws

ASAD Ai Cee

Ley Bi i ee ee

Bry A eS

aie Me ee Mactan tn i ee TE a ae

VOM AREA, Tear BT

Oe i Ee or

1S eer ey cag

2 PE ay By AN fed

No Paeons 5 Sy Pe AN wD Aysg ee

PERSE TA car = ast WTB AD 0d QE Ana

Paes ae as

eRe tad”

t wy ‘ 4 WA ys

CREP RTI SRR BUY SOUS Bear yk By) ee wy 2

Pi Ra hea

eye aa

AOE OD Wg ere

A Nk BL aie et tie a Le ee ek ea a a eT

FOG SAY ae See Ap P

Wy See ee Sie ay e+

Re els iy

ED BY RE BRN LVS. CIT G Ay,

Were yy

i ie ee i ST | ah “1

UERNY Ee ely wre

tak decd

a We = FO Car ee ear toe a iy Sy

ad es Say ay Vee

aay

ta to oe ee

whee: Dita eye a

Ca We et

i eer

DOLL E ES Se yt:

fe ee a

Po 4y gays

SPRL eS aS Re

rea

BIA

Lo eR ak eee

PEE RIS eeu. yp

Ore

Wake Weg Wet e ey re ye

i ieths Siow vip

Piake ieee ae es

EY Hn te

Mae a ice a ae)

ratsy)

5 ney ely

LUE ET OE IOI NS ry

AMEE Oe eee :

ha Linge re

REPS SYS EES Qa S

E rd ier