ata Berne lap sane ys
Ror ners eae he
Vi97 us Paseo:

1 Eatie ty eanOre gh”

SA gon ue :
bath elbeg = Li OMe ss ae cer 5 Ta te

ede ates pan te - ‘ a > ase

PEA ee

‘

Ns ar
AL AAG

Wate RIA eee
Pe He Ore
fos SANA

ae.

Re snuegts

Tee.
Nes
A nase

ie

Sore

es

pete

at

2 Nee om SINT 8 Neem Non ty noe
— — (op) — 127) —
1 LIBRAR| ES SMITHSONIAN INSTITUTION NOILNLILSNI _NVINOSHLINS SAluvudit LIBRA
= = = = 4 <x = , ae
— = Y fy = > eS
S s S = Uy & = 5 N
2 8 2 b Uf? e OBS
2 = 2,9 Me 2 en
o Se: a = a 5 So
| NOILALILSNI_ NVINOSHLINS LIBRARIES SMITHSONIAN INSTITUTION NOILN.
— Y
aos B ea fa tT) = WW
< . oy
oc S . - i Ee =
a = m = om. = foe)
5 2 es 2 2 - =
—J
1 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3I1YVYSIT LIBRE
Re aS iE fe z 2
oo = <w = ee 2)
0) = 1 tof = 0] me oe on ea
> - wf Z > 2 > | = >
a Be Hy, ci = a = 2)
7 0 Vl & D = . =
N NOILALILSNI NVINOSHLINS S31IYVYGIT LIBRARIES SMITHSONIAN INSTITUTION
4 z g Be ane g ge
= 'S Se = = NSS =
ae O a pi oe: RC ok ro) ‘< ® =
” 7) Disy PY ” N w n NG \ nw
Misehy O
Z = 2 “fy F 2 = WY 9
See os =e aallint: _ = ,
7_LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYSIT_ LIBR
z he Z hi S ey ” Zz
2 z, : : = : ;
S =” : : S = S
= faa) aS je) — oO _
O aes ee O — rs
Fa af 2 ie - my =
N_ NOILNLILSNI_NVINOSHLINS S3IYVYSIT LIBRARIES INSTITUTION NOILA
g Eee MO = g : g
b= cal : <A be > ra 4 > -
= 2 ON os oy bes =
— — (ep) = =
1_LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3luvugi7_LIBR
a) g =e ” z n : rage
xe = iy = < = <= NN
= = - No 4 bar = Yl ‘ = NX
a wes D \ ow ‘ D a Vp ye a;
EF NOY 2 = A 2, ie 2 JY Fa
Y NOILALILSNI_NVINOSHLINS S3INYVYGIT_LIBRARIES SMITHSONIAN INSTITUTION | NOLL.
Z > wv) >
ad ” am Ww Uy, : uw
: z g = Yy,* =
my c ; a < Yi fori <
or ec ow Ce Me ire css wige
in = a + Oc Sais 4 oo
= iS) ees ro} ing je) =
I = us] z= iy rd ical
1 LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNI NVINOSHLINS SZIYVUEIT LIBR!
Js = Y Zz z Ls
w < oO = ~ w
: — IY ® : : :
ze) cate 2 Ps) a — ue
Bs sa sai Ee me
~ Zz wp Zz Zz a

N NOILNLILSNI

S3IYVYaIT LIBRARIES

V4 oh y if
bh pp

SMITHSONIAN INSTITUTION NOILN

ve

Qs
AW

=
<< OS SON 7

4 a= 44
oh IB (a,
tf fe ,,
-FA fs Oo {F

a\viSVM

2

SHLINS S3ZIYVYUStI

SHLIWS
nst>
SHLIWS
SONIAN
sad
SONIAN
te
ol
SHLIWS
aN

WAY - \c fot aah Na 5 Te EN a NG Z : a MA
NY @ S —-X ey i aie, A a — \iSabes/ NAdaere MY
SS i?) Sw. ASAT rm tINGs ys Ww m by Moe > Tp) Ss Waste’ SX
"ae = Ww z= Ww rod Pe ts

SMITHSONIAN

INSTITUTION NOILNILILSNI NVINOSHLINS S31YvVYGIT_LIBRARIES_ SMITHSOI

2
”
< 2 | z 2 a 2
— = li Z — a VW A; = =<
z = y pl hae s. Yo: = 4
2 NANG LY : we Joe :
WQ FH RUM AS Oo ye o> Oo
F Ws 2, fe = Buy ee) NIN 2
aS > >" = ays, >
= aa = = 2S ” »S _
SJIMYVHYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_ NWINOSH
z e 2 ct = *
2 ~ @ - ZN
=. a. ©
oc = oc et
a 4 4 “2
— m.- —_— fea) = DSoN
O ~ oO ~ Oo Ss
= 2 ot =

SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS SAI1YVYSIT LIBRARIES SMITHSC

SaIuVYSIT LIBRARIES

° = Ks O an O
— _ NS _ = Pee A
A ue * ee) Th
F We 2 of G
ta ~ m A nee m e/a
NVINOSHLINS S4IYVYSIT_ LIBRARIES SMITHSONIAN INSTITUTION NOILALILSNi_ NVINOSH
| = ps N = W = =
z = = Rs ay =
B DQ ; Sole w NN a EA =
NRA, OMe ag O oO Wy Yi: as
2 5 2 z Uy E
i > = > > ——
P25 op) Zz ze ep)

SMITHSONIAN INSTITUTION NOILNLILSNI LIBRARIES SMiITHSO

a
<
ra
O
Y
=
=
a
al Me me A
We Z A u Z ul
C S C = = ee
= ~ & = “LY
S m 2 co = a Uf
Ye O — @) aa S ae
= 4 Z See Fa Ph
NVINOSHLINS__S3IYVUGIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_NVINOSH
aS ~~
\y 2 m9 ° Ye S a
if S& EE zs — = a »
i i oe = — 2
N\ SE: = a * = D a
ae = 7) = a) = 7) py
SMITHSONIAN INSTITUTION NVINOSHLINS S31uvugi7T_ LIBRARIES SMITHSC
vi a op) z ep) Pz ”
=e = < = oS s
z = ee + z oa
O a f (©) ra 8 O ae
wo a@ Ww ” 2) \ ”
3 ae © ge @) ae \ SS fe)
Ze = = = WE
. e => Poa >
3 a z a ws

NVINOSHLIWS LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI_ NVINOSt

7
:
;

NOILNLILSNI
LIBRARIES
NOILALILSNI
LIBRARIES
NOILNLILSNI

INSTITUTION NOILMLILSNI NVINOSHLINS S3IYVYUEIT LIBRARIES SMITHS!

Z Z

Ys

INSTITUTION NOILNLILSNI NVINOSI

SAIUVUGIT LIBRARIES SMIT

INSTITUTION

INSTITUTION
Saiu¥vudil

INSTITUTION

a

NVINOSHLINS S31YVYSIT LIBRARIES

uel F Ws
S AWS

5 a)
1k3 &. &
I> a NWS N\

SMITHSONIAN

\wS¥m
oo 54

SHLINS SAIYVUEI)

FH LIWS
NO?

Ads
JONTAN
YY

fe Wh
SHLIWS
.
S
SONIAN
r77 >

EMEP ee) Bie:
SONIAN

ee | |©6UANNALS OF THE ANNALE VAN DIE
aa SOUTH AFRICAN MUSEUM SUID-AFRIKAANSE MUSEUM

VOLUME 90 BAND 90

FEB 28 034
LIBRARIES.

ANNALS OF THE SOUTH AFRICAN MUSEUM
ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

VOLUME 90 BAND

OY,
2, S
Alig Ah
vio novi

REE SVRUSIEES OF THE DIE TRUSTEES VAN DIE
SOUTH AFRICAN MUSEUM SUID-AFRIKAANSE MUSEUM
CAPE TOWN KAAPSTAD

1982-1983

ci
eile
Ti
= i
=
{
i
ie
= %
F ,
1s \
>
oy
‘
t
“1
’

9

Lisi OF CONTENDS

FERRARA, F. see Tali, S.

KENNEDY, W. J., WRIGHT, C. W. & KLINGER, H. C.
Cretaceous faunas from Zululand and Natal, South Africa. The ammonite subfamily
Bankoisiceratmac Basse, 1947. (Published March 1983:).. >... 240) -.sshe ee.

KENSLEY, B.
Revision of the southern African Anthuridea (Crustacea, Isopoda). (Published Sep-
RETMDSI NOEs tA etcetera emi te eee Pai pam emoe teices TPA de a bo tar lee

KLEIN, R. G.
Patterns of ungulate mortality and ungulate mortality profiles from Langebaanweg
(early Pliocene) and Elandsfontein (Middle Pleistocene), south-western Cape
Erovince, South) Airica. (Published August 1982.).. 2.5. 29.520. 50-52) 66 2 eee

KLINGER, H. C.
Revision of Ancyloceras bipunctatum Schliter, 1872 (Cephalopoda, Ammonoidea)
and discussion of the validity, phylogeny and limits of the genus Neancyloceras
Seamielo2on(Rublished: @Octoberml982.)) exes ms. sec) Ge te ey eee

KLINGER, H. C. see KENNEDY, W. J.

Prins, A. J.
Morphological and biological notes on six South African blow-flies (Diptera, Cal-
liphoridae) and their immature stages. (Published September 1982.) ...........

TaiTi, S. & FERRARA, F.
Revision of the family Philosciidae (Crustacea, Isopoda, Oniscoidea) from South
Paincom Gaolished July A982 a) ess or NGS Slee uj) ene ae oe ee

Wricut, C. W. see KENNEDY, W. J.

Page

241

95

49

219

201

NEW GENERIC NAMES PROPOSED IN THIS VOLUME

Barnardoscia Taiti & Ferrara, 1982

3
= :
i [
rt
¥
iy
\
af
i
a
ie
i
¥

fe ”

‘if an Lane i
= oe yt,

VOLUME 90 PART 1 JULY 1982 ISSN 0303-2515

ms QHY

1
S67X
NH

LIBRE RIES.

A rn

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
(c) Table of contents giving hierarchy of headings and subheadings
(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 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, 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; Thue ED
‘As described (Haughton & Broom aie

‘As described (Haughton et al. 1927) .

Note: no comma separating name andl 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. & RarFFy, A. 1933. Etudes sur les échanges respiratoires des littorines. Archs
Zool. exp. Zen. 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.

Konn, A. J. 1960b. 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 Stid-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 90 + +#£Band
July 1982 Julie
Part 1 Deel

AAG (ESS
TAR TION
2S B:9.8:3.2

?
OD \ 4 A
> S

REVISION OF THE FAMILY PHILOSCIIDAE
(CRUSTACEA, ISOPODA, ONISCOIDEA)
FROM SOUTH AFRICA

By

STEFANO TAITI
&
FRANCO FERRARA

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, AB, 5-9), AUD, 4-8, 8, cpult.), SEES, 5, TO),
GC. epi) 704), 8 9d=2s 7), 101-3),
INGER, 5, 7, eos), ICS), 21), V7, SIGS), OG), 33), 25Q). 25)

EDITOR/REDAKTRISE
Ione Rudner

Copyright enquiries to the South African Museum

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 037 0

Printed in South Africa by In Suid-Afrika gedruk deur
the wRustica ress, Pity slide Die Rustica-pers, Edms., Bpk.,
Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap

REVISION OF THE FAMILY PHILOSCIIDAE
(CRUSTACEA, ISOPODA, ONISCOIDEA) FROM SOUTH AFRICA

By
STEFANO TAITI
&
FRANCO FERRARA

Centro di Studio per la Faunistica ed Ecologia Tropicali del
Consiglio Nazionale delle Ricerche, Florence

(With 25 figures and 1 table)

[MS accepted 25 March 1982]

ABSTRACT

This revision includes a list of 13 species representing 6 genera and the description of one
new genus Barnardoscia, and 5 new species, Nahia louwi, Natalscia thomsoni, N. rotundata,
N. appletoni, and Barnardoscia maculata. All the previously known genera and species are
redefined. Also given are keys to the South African genera and species and a synopsis of all the
Afrotropical genera of Philosciidae, as well as the distribution of the South African philosciid
fauna and its affinities.

CONTENTS

PAGE

PSHE OCR LUO pepe eeg re ane oc OPIS Map as Gey Si haere Oa 1
Key to the South African genera of Philosciidae............... 2
GCHUSVA DIT OSC. Oa seins cs Res cites eens se a Ane 3
GemUSPDCHINANODS ee iV Acer. Pio we ar a oN tee eon iene nee 7
GEHUSP ALT OPI OSCLULS Soom oie Heo urs Dane Mee gee ek 11
CGYETTTTIS IAG UTE Steer ar eee ere See ear me eed Santas oe 14
Keytowthe ispecies OL NGNIG ws so. es Beas ee ee 14
remo IN ASAISCIO oe oir bn Gye 5 2 eet ee 19
Key tothe species Of Natalscid.2 2062 62. soe eee oe 21
GERUSPNATMATAOSCIA NOV. hoe Sen oid one as ae eS 34
Keynowthe Specics Of BalmardOsci@: 55. ees san eee 35
SSte MI ALI CISCUSSION sac So aie ee ee OO ee 4]
Moeccorraphic- discussion! \42%... 3 eee ok Ps) Se 42
Synopsis of the Afrotropical genera of Philosciidae ........... 47
pac kao WIE PEM CHISS 5 choses 2s, wie ayes ee sun tae ave cle Serer 47
PCIE TCT CES ein Neie ARC ES marie tine nee Bets eM ea au ai 47

INTRODUCTION

Taiti & Ferrara (1980) tentatively revised the family Philosciidae from the
Afrotropical region, defining all the known genera (though some of them in a
provisional way, due to the lack of material) according to modern criteria, and
describing or redescribing several species.

Ann. S. Afr. Mus. 90 (1), 1982: 1-48, 25 figs, 1 table.

D ANNALS OF THE SOUTH AFRICAN MUSEUM

Unfortunately neither the South African endemic genera (Nahia Budde-
Lund, 1908, and Benthanops Barnard, 1932) nor the species had been
examined directly, thus the generic diagnoses and species list were based on the
available literature.

S. Taiti was recently given the opportunity of collecting several philosciids
during his stay in South Africa on a fellowship provided by the Department of
National Education of the Republic of South Africa. Materials preserved in the
South African Museum, Cape Town, and Natal Museum, Pietermaritzburg,
and some types in the British Museum (Natural History) were also studied or
revised.

Abbreviations used throughout the text:

b distance of the nodulus lateralis from the posterior margin of the
pereon segment

b/c ratio between b and c

BM British Museum (Natural History)

C length of the pereon tergite

d distance of the nodulus lateralis from the lateral margin of the pereon
segment

d/c ratio between d and c

juv. juvenile(s)

MZUF Museo Zoologico dell’ Universita, Florence

NM Natal Museum, Pietermaritzburg

NRS Naturhistoriska Riksmuseet, Stockholm

ovig. ovigerous

SAM South African Museum, Cape Town.

KEY TO THE SOUTH AFRICAN GENERA OF PHILOSCHDAE

1... .Frontal! line:present (hig. IO) a. 05.9 i 4 cee seo St ae ee ee Aphiloscia
=/+ Frontal line missing i. 3 2204025445852 See ee ee ee z
2 Eye withasinglelarceommatidiruny (higa SA) oe aa ee eee Benthanops
= - Eye with several‘omimatidia 4.5.0.8. 4 e4..5 © 8 cine sa aa © 6 oo ea ek ee 3

3. Insertion of uropod exo- and endopodites at the same level (Fig. 6B); d/c co-
ordinates of noduli laterales with maxima on pereon segments 2 and 4 (Fig. 6A)
Afrophiloscia
— Insertion of endopodites more or less proximal to those of exopodites (Figs 8E, 12C,
20D); d/c co-ordinates without a maximum on pereon segment 2 (Figs 8A, 12A,

ZA ao cs laa ice 8 Vintowr ti ae eee, os ak eos tea Sateen ON CaS Gr 4
4. d/e co-ordinates without evident maximal (Figs SA) 932. ..5--5.0 4-0-4. se eee Nahia

d/c co-ordinates with maxima on pereon segments 1 and 4 (Figs 12A, 20A) ......... 5
5. Two noduli laterales on pereon segment 7; endite of maxilliped without a penicil

(Figg 12 BY: vis cee en oe hol cet ae Se Sere ell cele ae a tee Ce ere Natalscia

— Four noduli laterales on pereon segment 7; endite of maxilliped bearing a penicil
GR. 20B is odie kates Uae: 2 3 Pasi eee ae ee i See eee ae Barnardoscia

REVISION OF THE FAMILY PHILOSCIIDAE 3

Genus Aphiloscia Budde-Lund, 1908

Diagnosis

Sulcus marginalis and gland pores present. Two series of noduli laterales
on each side of pereon segments (Fig. 1B). Frontal and supra-antennal lines
present (Fig. 1C). Pleon epimera very produced. Molar penicil of mandible
consisting of a single unbranched seta; outer branch of maxilla 1 with 4+ 6
(5 cleft) teeth; endite of maxilliped without a penicil (Fig. 2B). Pleopod exopo-
dites with respiratory areas (Fig. 2F, H). Uropod protopodite with a triangular
depression on outer margin; insertion of endopodite proximal to that of
exopodite (Fig. 2C).

Type species

Philoscia guttulata Gerstaecker, 1873, from Tanzania.

1mm 0.5mm B

Fig. 1. Aphiloscia vilis (Budde-Lund, 1885). A. Adult female. B. Right half of pereon

segment 4: b—distance of the nodulus lateralis (n.1.) from the posterior margin of the segment,

c—length of the pereon tergite, d—distance of the nodulus lateralis from the lateral margin of

the pereon segment, g.p.—gland pore, s.m.—sulcus marginalis. C. Frontal view of cephalon:
f.1.—frontal line, s.a.l1—supra-antennal line.

c ANNALS OF THE SOUTH AFRICAN MUSEUM

Remarks

Aphiloscia is distinguished by the presence of two series of noduli laterales
on each side of pereon segments, a frontal line, very produced pleon epimera,
and respiratory areas in the pleopod exopodites.

Besides A. guttulata and A. vilis, the genus includes the following species:
A. annulicornis (Budde-Lund, 1885) widely distributed throughout Madagascar
and surrounding islands; A. maculicornis (Budde-Lund, 1898) from Uganda,
Tanzania, Zambia; A. sordida Arcangeli, 1950, A. congolensis congolensis
Arcangeli, 1950, A. congolensis damasi Arcangeli, 1950, all from Zaire;
A. montana Taiti & Ferrara, 1980, from Zimbabwe; A. trifasciata Taiti &
Ferrara, 1980, from Kenya; and A. digitata Taiti & Ferrara, 1980, from
Tanzania, Malawi, and Mozambique.

Aphiloscia vilis (Budde-Lund, 1885)
Figs 1-2
Philoscia vilis Budde-Lund, 1885: 210. Dollfus, 18956: 351. Barnard, 1937: 164. Brian, 1953: 9.
Philoscia (Aphiloscia) vilis: Budde-Lund, 1908: 292. Barnard, 1932: 239, figs 16g, i, IL-—n, u,
17a, 18d, 19c. Barnard, 1960a: 47. Appleton, 1974: 51. Lawrence, 1977: 175.
Aphiloscia vilis: Stebbing, 1910: 443. Arcangeli, 1950: 66. ?Barnard, 1956: 436. Barnard,
1960b: 505, 508. Ferrara & Taiti, 1979: 112. Taiti & Ferrara, 1980: 58.
Philoscia dilectum Collinge, 1917: 579, pl. 42 (figs 21-31). Collinge, 1920: 478. Collinge,
1945: 345. Brian, 1953: 9.

Material

Barnard Collection. Transvaal: 5 29, Louis Trichardt, Hanglip forest,
4500 ft alt., leg. R. F. Lawrence, February 1960, NM-6493; 4 2°, Louis
Trichardt, Hanglip forest, 3 000 ft alt., leg. R. F. Lawrence, February 1960,
NM-6497; 1 3, 3 22, Entabeni forest, Louis Trichardt, leg. R. F. Lawrence,
February 1960, NM-6501; 2 2°, Barberton, leg. R. F. Lawrence, March
1960, NM-6518; 16, 1, 6juv., Louis Trichardt, Hanglip, leg. R. F.
Lawrence, February 1960, NM-6519.

New material. Transvaal: 1 2, Gladdespruit, near Nelspruit, leg. C. C.
Appleton, 18 July 1969, No. 2L, SAM-A16843; 4 dd, 16 29, same data,
October 1969—October 1971, No. 165G—H, SAM-—A16844; 2 2°, same data,
No. 169S, SAM-A16845; 1 2, same data, No. 129H, SAM-A16846; 1 d,
4 29, same data, No. 1770, SAM-A13162; 1°, same data, No. 165Z,
SAM-A16847. Natal: 16 dd, 26 292, Pietermaritzburg, leg. S. Taiti, 19 April
1980, MZUF-995; 3 dd Gillits, leg. S. Taiti and K. C. Thomson, 26 April
1980, MZUF-996; 3 22, Durban, leg. G. Cosnett, 28 March 1980,
MZUF-997; 34 356, 32 22, Umdloti, leg. S. Taiti and K. C. Thomson, 26
April 1980, MZUF-998. Zululand: 3 dd, 12 29, near Lake Sibaya Research
Station, coastal dune forest, leg. C. C. Appleton, date ?, No. 18L,
SAM-A16848; 16, 4¢9°, same data, 10 October 1973, No. 76E,
SAM-A16849; 4 dd, 2 22, same data, 18 December 1973, No. 13E, SAM-
A16850; 6 66, 4 22, same data, 29 July 1973, No. 49L, SAM-A16851;

REVISION OF THE FAMILY PHILOSCIIDAE 5

Aaa Sa
dy = *
1.5 o--~ \
/ \
7 \
/ \
/ \
é »

1 Se a 9

—_

| Wom | 6IM Veo OVI A | nom Ww Vwi Vil

Ls

2mm
0,2mm 0,2: my

0.5mm

Fig. 2. Aphiloscia vilis (Budde-Lund, 1885). A. b/c and d/c co-ordinates. B. Apex of

maxilliped. C. Telson and uropods. D. Spine of pereopod 1 carpus (d). E. Pereopod 7

ischium and merus (d). F. Pleopod 1 exopodite (d): r.a.—respiratory area. G. Pleopod 1
endopodite (¢). H. Pleopod 2 (¢).

6 ANNALS OF THE SOUTH AFRICAN MUSEUM

1d, 2 22, same data, date?, No. 18D, SAM-A16852; 2 od, Lake Sibaya
Research Station grounds, leg. C. C. Appleton, 18 September 1973, No. 63C,
SAM-A16853; 2 $2, Lake Sibaya Research Station, garden, leg. C. C.
Appleton, 7 January 1973, No. 8L, SAM-A16854.

Description

9 mm long (according to Barnard, 11 x 4,5 mm). Colour extremely vari-
able, usually plumbeous with lighter mottling; pale spot at the base of pereon
epimera; epimera dark with external margin orange; joints 1-3 and basal half of
joint 5 of antenna orange; basis of pereopods with a dark spot; uropods orange.
Eye with twenty-five to thirty ommatidia. Each pereon segment with numerous
gland pores (about thirty) per side, arranged along the whole sulcus marginalis.
_ Noduli laterales with b/c and d/c co-ordinates as in Figure 2A. Telson (Fig. 2C)
with concave sides, dorsally impressed, narrowly rounded apex. Antenna: ratio
of flagellum joints 4:3: 3.

Male

Pereopods 1-4 merus and carpus with brushes of trifid spines (Fig. 2D).
Pereopod 7 ischium with a narrow and shallow depression on the rostral surface
(Fig. 2E). Pleopod 1 exopodite with subquadrate posterior lobe (Fig. 2F);
endopodite with apex slightly bent outward and equipped with fine setae
(Fig. 2G). Pleopod 2 as in Figure 2H.

Remarks

A. vilis is very close to A. montana Taiti & Ferrara, 1980, from Mt.
Selinda (Zimbabwe). It is distinguished in the male (cf. Taiti & Ferrara 1980,
figs 7-11) by the less enlarged pereopod 7 ischium which is not excavated at the
base; the shorter posterior lobe of pleopod 1 exopodite; the presence of setae
on pleopod 1 endopodite apex; the pleopod 2 endopodite that is only slightly
longer than exopodite.

The exact locality of the type material is unknown. Budde-Lund
(1885: 210) quotes ‘Caput Bonae Spei’ but, as pointed out by Barnard
(1932: 240) the specimen was collected by the botanist Drege, who travelled
both in the Cape and in Natal. Most probably the record refers to the latter
region, where A. vilis is widely distributed. It is definitely not found in the
Cape Peninsula.

This species occurs throughout southern Mozambique, Zimbabwe, eastern
Transvaal, Zululand, and Natal, and is also known in East London (Cape
Province) and Mafa (Ovamboland). In our opinion the latter record needs
further confirmation as this is the only occasion that A. vilis has been found
outside south-eastern Africa.

REVISION OF THE FAMILY PHILOSCIIDAE 7

Genus Benthanops Barnard, 1932

Diagnosis

A few gland pores are present on the anterior part of the pereon segments.
One series of noduli laterales on each side of pereon segments 1-6; two noduli
laterales on each side of pereon segment 7; d/c co-ordinates with a maximum
on segment 4. Eye with a single large ommatidium. Frontal line absent;
supra-antennal line present. Pleon epimera reduced, with very small posterior
points visible in dorsal view. Molar penicil of mandible consisting of a tuft of
plumose setae each arising separately (Fig. 3B); outer branch of maxilla 1 with
4+5 serrate teeth (Fig. 3C); endite of maxilliped without a penicil (Fig. 3D).
Pereopods without dactylar seta. Pleopod exopodites without respiratory areas.
Uropod protopodite grooved on outer margin; insertion of endo- and exopodite
almost at the same level.

Type species
Philoscia (Benthanops) fulva Barnard, 1932, from South Africa.

Remarks

Benthanops is very close to the South American genera Benthana
Budde-Lund, 1908, Benthanoscia Lemos de Castro, 1958a, and Benthanoides
Lemos de Castro, 19585, from which it is distinguished by the presence of a
single ommatidium and the different shape of the molar penicil of mandible.
Even greater affinities exist between this and Ctenoscia Verhoeff, 1928, from
which Benthanops differs in the number and position of the noduli laterales (cf.
Fig. 3A, and E) as well as the shape of the molar penicil.

The genus includes only the type-species.

Benthanops fulva Barnard, 1932
Figs 3A-D, 4, 5A

Philoscia (Benthanops) fulva Barnard, 1932: 247, figs 16c, f, r, 18e, 19f, 20. Brian, 1953: 9.
Benthanops fulva: Ferrara & Taiti, 1979: 112. Taiti & Ferrara, 1980: 87.

Material

Barnard Collection. Cape Province: 1 °, Caledon, leg. K. H. Barnard,
March 1908, SAM-A5894; 1 3, 17 22, Noordhoek forest, leg. K. H. Bar-
nard, 19 January 1928, SAM-A7329; 6 22, Kalk Bay Mt., leg. K. H.
Barnard, January 1929, SAM-—A7880; 2 dd, 1 2, Noordhoek forest, leg.
K. H. Barnard, 16 June 1929, SAM-A7881; 5 6d, 45 292, Table Mt., leg.
K. H. Barnard, September 1919, SAM-A7883; 1 2, Kleinmond, leg.
K. H. Barnard, 1927, SAM-A7946; 2 2°, Hout Bay, leg. ?, 9 August 1931,
SAM-A10364.

New material. Cape Province: many dd and 2°, Cape of Good Hope
Nature Reserve, leg. S. Taiti, 10 April 1980, MZUF-999; 34 dd, 35 22,

ANNALS OF THE SOUTH AFRICAN MUSEUM

0.5

Vil

E

vi Vi

a

m.p. : ‘e.. - .
0,05 mm | 0,05 mm

Fig. 3. A-D. Benthanops fulva Barnard, 1932. A. b/c and d/c co-ordinates. B. Apex of

D. Apex of maxilliped.

0,05 mm

C. Outer branch of maxilla 1.

mandible: m.p.—molar penicil.
E. Ctenoscia minima (Dollfus, 1892). b/c and d/c co-ordinates.

REVISION OF THE FAMILY PHILOSCIIDAE 9

0,1 mm

0,1 mm

Fig. 4. Benthanops fulva Barnard, 1932. Telson and left uropod. B. Spine of pereopod 1
carpus (d). C. Pereopod 7 (d). D. Pleopod 1 exopodite (d). E. Pleopod 1 endopodite (¢).
F. Pleopod 2 (¢).

10 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 5. A. Benthanops fulva Barnard, 1932. Adult female. B. Afrophiloscia, ocellata (Barnard,
1960). Adult female.

Table Mt., Skeleton Gorge, leg. S. Taiti, 13 April 1980, MZUF-—1000; 1 d,
3 2°, Betty’s Bay, leg. J. Hoy, 23 April 1980, MZUF-1001.

Description

645mm long; 2 7mm long. Yellowish brown, with more or less
evident transversal brown stripes at the anterior margins of segments. Noduli
laterales with b/c and d/c co-ordinates as in Figure 3A. Telson (Fig. 4A) with
feebly concave sides, rounded apex. Antenna: ratio of flagellum joints 5:4: 4.
Uropod exopodite very long (about three times longer than endopodite).

REVISION OF THE FAMILY PHILOSCIIDAE 11

Male

Pereopod 1-2 merus and carpus with very sparse brushes of spines
(Fig. 4B). Pereopod 7 without specializations (Fig. 4C). Pleopod 1 exopodite
(Fig. 4D) with outer margin feebly concave, rounded apex; endopodite
(Fig. 4E) equipped with some short spines on the medial surface and triangular
lobe on the external surface. Pleopod 2 as in Figure 4F.

Remarks

Benthanops fulva differs from all the other South African (and Afrotropi-
cal) philosciids by the presence of a single ommatidium.

Barnard (1932: 249) states that this species occurs on mountains but not at
low levels. In the Cape Peninsula it is abundant in bushes close to the sea. It is
common between Cape Town and Caledon (Cape Province).

Genus Afrophiloscia Taiti & Ferrara, 1980

Diagnosis

Sulcus marginalis and gland pores absent; one series of noduli laterales on
each side of pereon segments; d/c co-ordinates with maxima on pereon seg-
ments 2 and 4. Frontal line absent; supra-antennal line present. Pleon epimera
reduced, adpressed. Molar penicil of mandible consisting of a single
unbranched seta; outer branch of maxilla 1 with 4+ 6 (5 cleft) teeth; endite of
maxilliped without a penicil. Pleopod exopodites without respiratory areas.
Uropod protopodite grooved on outer margin; insertion of endo- and exopodite
at the same level.

Type species

Afrophiloscia kinolensis Taiti & Ferrara, 1980, from Tanzania.

Remarks

This genus, recently instituted and discussed (Taiti & Ferrara 1980), differs
from the other South African genera of Philosciidae by maxima on pereon
segments 2 and 4 of the noduli laterales d/c co-ordinates; the setose apex and
typical indentation on the medial margin of the maxilliped endite (see Fig. 6D)
and insertion of the uropod exo- and endopodites at the same level.

Afrophiloscia, besides A. kinolensis and A. ocellata, includes the following
species: A. uncinata (Ferrara, 1974) and A. brevicauda Taiti & Ferrara, 1980,
from Tanzania; A. africana (Schmoelzer, 1974), A. rotundata Taiti & Ferrara,
1980, and A. kenyensis Taiti & Ferrara, 1980, all from Kenya.

12 ANNALS OF THE SOUTH AFRICAN MUSEUM

Afrophiloscia ocellata (Barnard, 1960)
Figs. 5B-6

Philoscia (Setaphora) ocellata Barnard, 1960a: 48. Lawrence, 1977: 175.
Setaphora ocellata: Ferrara & Taiti, 1979: 119. Taiti & Ferrara, 1980: 83.

Material

Barnard Collection. Transvaal: 2 dd, 3 292 (types), Magoebaskloof,
4 000 ft alt., leg. R. F. Lawrence, March 1960, NM-6512; 2 22, Mariepskop,
6 000 ft alt., leg. R. F. Lawrence, March 1960, NM-6504; 1 9, 1 juv., Gras-
kop, leg. R. F. Lawrence, March 1960, NM-6508; 3 22, Malta forest, Selati
Estate, leg. R. F. Lawrence, NM-6516.

New material. Transvaal: 1 2, Gladdespruit, near Nelspruit, leg. C. C.
_ Appleton, October 1969—October 1971, No. 174T, SAM-A16855.

Description

9,5 mm long (according to Barnard 7 X 2,5—2,75 mm). Brown with pale
irroration on head and pereon; pleon almost a uniform brown; large oval pale
spot at the base of pereon epimera. Eyes with about sixteen ommatidia. Noduli
laterales with b/c and d/c co-ordinates as in Figure 6A. Pleon epimera
adpressed with very short posterior points, not visible in dorsal view. Telson
(Fig. 6B—C) with slightly concave sides, narrowly rounded or subacute apex.
Antenna with ratio of flagellum joints 5:3:3. Endite of maxilliped (Fig. 6D) as
in the other species of Afrophiloscia.

Male

Pereopods 1-2 are missing in the males observed. Pereopod 7 without any
evident specialization (Fig. 6E). Pleopod 1 exopodite (Fig. 6F) with triangular
posterior point bent outwards; endopodite (Fig. 6G-H) with an oval lobe
equipped with fine setae. Pleopod 2 exopodite much shorter than endopodite
(Fig. 61).

Remarks

Re-examination of the type material showed that these specimens belong
to the genus Afrophiloscia Taiti & Ferrara, 1980. A. ocellata differs from
A. africana and A. uncinata by the lack of specializations (hooks) on pereopod
7 merus 3; and from those and from all the other species of Afrophiloscia by
the swollen setose apex of pleopod 1 endopodite 6. The distinctive coloration
is a useful trait in identifying this species.

Afrophiloscia ocellata occurs throughout the eastern Transvaal. Its
presence in South Africa considerably enlarges the range of the genus Afro-
philoscia previously known only in Kenya and Tanzania.

REVISION OF THE FAMILY PHILOSCIIDAE is

| to om Ww eVoVi Vil A | to ot Ww VeVi VI

0,2 mm 0,2mm 0,2 mm

\
a
Fig. 6. Afrophiloscia ocellata (Barnard, 1960). A. b/c and d/c co-ordinates. B. Telson and
right uropod. C. Telson of a female specimen from Graskop. D. Apex of maxilliped.
E. Pereopod 7 ischium and merus (¢). F. Pleopod 1 exopodite (3). G. Pleopod 1 endopo-
dite (3). H. Apex of pleopod 1 endopodite (3), ventral view. I. Pleopod 2 (<d).

14 ANNALS OF THE SOUTH AFRICAN MUSEUM

Genus Nahia Budde-Lund, 1908

Diagnosis

Sulcus marginalis and gland pores present. One series of noduli laterales
on each side of pereon segments; their d/c co-ordinates lacking evident maxima
(Figs 8A, 10A). Frontal line absent and supra-antennal line present. Pleon
epimera reduced. Molar penicil of mandible consisting of a single unbranched
seta (Fig. 8B); outer branch of maxilla 1 with 4+ 6 (5 cleft) teeth (Fig. 8C);
endite of maxilliped without a penicil (Fig. 8D). Pereopods with a flagellar
dactylar seta. Pleopod exopodites without respiratory areas. Uropod protopo-
dite grooved on outer margin; insertion of endopodite slightly proximal to that
of exopodite.

Type species
Philoscia hirsuta Budde-Lund, 1906, from South Africa.

Remarks

An incomplete diagnosis of Nahia was given by Taiti & Ferrara (1980: 87),
based on the diagnoses by Budde-Lund (1906) and Barnard (1932), as well as
on the description and figures of Philoscia warreni (= Nahia hirsuta according
to Barnard 1932) by Collinge (1917).

The study of several specimens of Nahia (N. hirsuta and N. louwi n.sp.)
resulted in a complete diagnosis of the genus and in the demonstration that—as
will be discussed below—P. warreni is not a junior synonym of Nahia hirsuta as
proposed by Barnard (1932).

Nahia differs from all the afrotropical genera in the absence of peaks in the
d/c curve. This trait, which places it in the Plymophiloscia group, gives it
affinity to the genus Bilawrencia Vandel, 1973b, from which it differs in the
absence of a penicil on the maxilliped and in the simple, rather than dichoto-
mized, molar penicil of the mandible. Nahia is also very close to the genus
Australophiloscia Vandel, 1973a, from which it differs essentially in the position
of the noduli laterales.

The genus Nahia includes only two species, both from South Africa.

KEY TO THE SPECIES OF NAHIA

1. Pereopod 7 ¢: merus with a setose protrusion at the base (Fig. 8H). Pleopod 1 3d:
exopodite with outer margin truncated and distally deeply excised (Fig. 9A); endopodite
with spines and a crest-shaped lobe close to the apex (Fig. 9B—C) :

N. hirsuta (Budde-Lund)

— Pereopod 7 ¢: merus without setose protrusion at the base (Fig. 10C). Pleopod 1 ¢:
exopodite with outer margin not truncated and not excised (Fig. 10D); endopodite with
eoblet-shaped apex (Fig. wl0E)" ss dv. aah ene. ce eign Govcance. N. louwi sp. nov.

REVISION OF THE FAMILY PHILOSCIIDAE 15

Nahia hirsuta (Budde-Lund, 1906)
Figs 7-9

Philoscia hirsuta Budde-Lund, 1906: 89, pl. 3 (figs 42-52). Barnard, 1937: 164. Barnard,
1949: 403.

Philoscia (Nahia) hirsuta: Budde-Lund, 1908: 290. Budde-Lund, 1909: 64. Barnard, 1932: 245,
figs 16j—-k, p, v, 18c, 19f, partim. Brian, 1953: 9.

Nahia hirsuta: Stebbing, 1910: 442. Ferrara & Taiti, 1979: 115. Taiti & Ferrara, 1980: 88.

Anchiphiloscia karongae: Stebbing, 1922: 6.

Material

Barnard Collection. Cape Province: 12 dd, 34 929, Zeekoevlei, Cape
Flats, leg. K. H. Barnard, 29 October 1929, SAM-A7397; 14 6d, 16 29,
Keurbooms River, leg. K. H. Barnard, January 1931, SAM-A7858; 3 66,
4 2°, same data, October 1930-January 1931, SAM-A7856; 1 6, 9 929,
same data, January 1931, SAM-A7842; 2 dd, 6 2%, Noordhoekvlei, leg.
K. H. Barnard, 24 January 1928, SAM—A7332; 1 d, 3 22, Houw Hoek, leg.
K. H. Barnard, 1931, SAM-A8021; 2 22, Doorn River, leg. S. H. Haughton
and C. W. Thorne, 1931, SAM-A7957; 8 6d, 14 2°, Forebay, leg. K. H.
Barnard, January 1931, SAM-A7869; 3 dd, 3 29, Wilderness, leg. S. H.
Haughton and C. W. Thorne, 1931, SAM-A7950; 2 22, Knysna, leg. R. F.
Lawrence, 1929, SAM-A7390.

New material. Cape Province: 4 dd, 3 29, Queenstown, leg. V. B.
Whitehead, 21 June 1975, SAM-A16856; 1 6, Die Bosch, Bredasdorp, leg.
B. F. Kensley, 1969, SAM-—A16857; 1 °, near Calitzdorp, 7 November 1971,
SAM-A16858; 27 2°, Zoutpansklipheuwel on Olifants River, 3 January 1975,
SAM-A16859; 18 35d, 37 22, Cape of Good Hope Nature Reserve, leg. S.
Taiti, 10 April 1980, MZUF-1002; 8 dd, 41 29, Rondeberg, 65km N of
Cape Town, leg. S. Taiti, 20 March 1980, MZUF-1003; 1 d, Table Mt.,
Skeleton Gorge, leg. S. Taiti, 13 April 1980, MZUF-—1004.

Description

3 6,5 mm long; & ovig. 8 mm long. Slate-grey mottled with lighter spots,
a narrow white and a wide black stripe at the base of pereon epimera; joints
2-5 of peduncle of antenna with a white ring at the base. Eye with about
twenty-two ommatidia. Back equipped with numerous upright setae. Each
pereon segment with ten to fifteen gland pores per side arranged along the
whole sulcus marginalis. Noduli laterales with b/c and d/c co-ordinates as in
Figure 8A. Telson (Fig. 8E—-F) short, with straight or slightly concave sides,
rounded apex. Antenna: ratio of flagellum joints 4:3: 4.

Male

Pereopods 1-3 merus and carpus with brushes of spines as in Figure 8G.
Pereopod 7 ischium with a deep concavity; merus with an evident setose
protrusion at the base (Fig. 8H). Pleopod 1 exopodite with outer margin
truncated and distally deeply excised (Fig. 9A); endopodite with apex equipped
with spines and a crest-shaped lobe (Fig. 9B—C). Pleopod 2 as in Figure 9D.

ANNALS OF THE SOUTH AFRICAN MUSEUM

a

nas

Fig. 7. Nahia hirsuta (Budde-Lund, 1906). Adult female.

REVISION OF THE FAMILY PHILOSCIIDAE Ly

iS i %&

ue os. ee

I ite IN VO SV VII A | oe Ie eV Vil

0,1 mm
0,05 mm
0,05 mm
|
Ul Va
G \ {\ | Gol BS
\ Z
\ : 7
\ SS
\W
H SS
0,2 mm
i
/
0,5mm F f

Fig. 8. Nahia hirsuta (Budde-Lund, 1906). A. b/c and d/c co-ordinates. B. Apex of mandible:

m.p.—molar penicil. C. Outer branch of maxilla 1. D. Apex of maxilliped. E. Telson and

uropods. F. Telson of another specimen. G. Spine of pereopod 1 carpus (6). H. Pereopod 7
ischium and merus (¢).

18 ANNALS OF THE SOUTH AFRICAN MUSEUM

0,05 mm

0,1 mm

0,2 mm

Fig. 9. Nahia hirsuta (Budde-Lund, 1906). A. Pleopod 1 exopodite (¢d). B. Pleopod 1
endopodite (¢). C. Apex of pleopod 1 endopodite (¢), ventral view. D. Pleopod 2 (¢).

Remarks

A study of part of Barnard’s collection showed that he had identified two
different species as Nahia hirsuta. The specimens from Addo Bush (Barnard
1932: 246) belong, in fact, to a new species of Nahia (see below). Consequently
only the verified material can be considered as correctly assigned, while all the
other records must be considered as sub judice.

Furthermore, as Philoscia warreni Collinge, 1917, from Natal is not a
synonym of N. hirsuta, as affirmed by Barnard (1932: 247), all his records from

REVISION OF THE FAMILY PHILOSCIIDAE 19

this region must be considered as doubtful. In our opinion this species occurs
only in the southern part of Cape Province.

It is likely that Chaetophiloscia elongata from Cape Town (Dollfus
1895b: 350) refers to N. hirsuta, as its colour pattern and general body shape
are very similar to this species.

Nahia louwi sp. nov.
Fig. 10
Philoscia (Nahia) hirsuta: Barnard, 1932: 245, partim (the specimens from Addo Bush).

Material

Barnard Collection. Cape Province: 1 d, Addo Bush, leg. J. Drury, July
1919, holotype SAM-A16860; 94d, 139%, same data, paratypes
SAM-A6067.

Description

6 7,5mm long; ° 8mm long. Colour faded by long conservation. Eye
with about twenty-two ommatidia. Back with numerous upright setae. Number
and disposition of gland pores as in N. hirsuta. Noduli laterales with b/c and d/c
co-ordinates as in Figure 10A. Telson (Fig. 10B) with concave sides, rounded
apex. Antenna: ratio of flagellum joints 5:4: 4.

Male

Pereopods 1-3 merus and carpus with sparse brushes of spines as in
N. hirsuta. Pereopod 7 ischium very similar to that of the preceding species,
merus lacking the setose protuberance (Fig. 10C). Pleopod 1 exopodite with
outer margin not excised (Fig. 10D); endopodite with goblet-shaped apex,
equipped with many setae (Fig. 10E). Pleopod 2 asin N. hirsuta.

Etymology

The species is named for Prof. G. N. Louw, Head of the Department of
Zoology of the University of Cape Town.

Remarks

These specimens were identified as Nahia hirsuta by Barnard. The new
species, even if closely related to N. hirsuta, is readily distinguished in the male
by the lack of a meral protrusion on pereopod 7, the non-truncated and
non-excised outer margin of pleopod 1 exopodite, and goblet-shaped apex of
pleopod 1 endopodite.

Genus Natalscia Verhoeff, 1942
Diagnosis

Sulcus marginalis and gland pores present; one series of noduli laterales on
each side of pereon segments; d/c co-ordinates with a maximum on pereon

20 ANNALS OF THE SOUTH AFRICAN MUSEUM

0.5 0.5 e*_ a ae

I ib ot iM Veovi Vt A 1 it ot Ww Voevie Vil

Fig. 10. Nahia louwi sp. nov. A. bic and d/c co-ordinates. B. Telson and left uropod.
C. Pereopod 7 ischium and merus (d). D. Pleopod 1 exopodite (d). E. Pleopod 1 endo-
podite (¢).

REVISION OF THE FAMILY PHILOSCIIDAE 7]|

segment 4 and a lower one on segment 1 (Figs 12A, 14A, 15A, 16A, 17A).
Frontal line absent, supra-antennal line present. Pleon epimera reduced, with
small posterior points visible in dorsal view. Molar penicil of mandible consist-
ing of a single unbranched seta; outer ramus of maxilla 1 with 4+ 6 (5 cleft)
teeth; endite of maxilliped with tiny setae at the apex and without penicil
(Fig. 12B). A long flagellar dactylar seta is present. Pleopod exopodites with-
out respiratory areas. Uropod protopodite grooved on outer margin; insertion
of endopodite proximal to that of exopodite.

Type species
Philoscia warreni Collinge, 1917 = Philoscia mina Budde-Lund, 1885, from
South Africa.

Remarks

Natalscia was established by Verhoeff (1942: 64) for Philoscia warreni
Collinge, 1917. As Barnard (1932: 247) considered this to be a synonym of
Nahia hirsuta (Budde-Lund, 1906), the genus Natalscia also became synony-
mous with Nahia. Instead, a study of Collinge’s material showed that P.
warreni and N. hirsuta belong to distinct genera, and thus Natalscia is to be
considered as a valid genus. However, Verhoeff’s diagnosis—based exclusively
on Collinge’s description and figures (often erroneous)—is incomplete and
incorrect in at least two points: the frontal line is absent (present according to
Verhoeff), and the outer branch of maxilla 1 has ten instead of eight teeth.
Thus it proved necessary to redefine the genus Natalscia.

It is very close to the new genus Barnardoscia from which it differs by the
presence of two, rather than four, noduli laterales on pereon segment 7 and the
lack of a penicil on the maxilliped. It differs from Nahia in the different
position of the noduli laterales (compare Figs 8A and 12A for b/c and d/c
co-ordinates), and from Setaphora Budde-Lund, 1908, in the position of the
noduli laterales (compare Fig. 12A with Fig. 83 in Taiti & Ferrara 1980), the
lack of a maxillipedal penicil, and the different level of insertion of the uropod
endo- and exopodites.

A comparison of the type material showed that P. warreni is a junior
synonym of P. mina Budde-Lund, 1885.

The genus Natalscia includes five species, all from South Africa. A sixth
form (Natalscia sp.) is known from Transvaal.

KEY TO THE SPECIES OF NATALSCIA

ieeltelsonisemucincular (Fig. 16B). 2... 222 ee eh eee kee N. rotundata sp. nov.
=emneisonitnancular (Figs 12C, 14B, 15B, 17B)... .... . eee cans ee ee tees eek oe eee py)
2. Pereopods 1-2 carpus and merus ¢ without brushes of spines ..... N. mina (Budde-Lund)
— Pereopods 1-2 carpus and merus ¢ with brushes of spines (Figs 15C, 17C) ......... 3)
3. Pleopod 1 d: exopodite with an evident triangular protrusion on outer margin

(Fig. 1SE); endopodite with an elongate hyaline lobe at the apex (Fig. 15G)
N. thomsoni sp. nov.
— Pleopod 1 6: exopodite without triangular protrusion on outer margin (Figs 14E,
17E); endopodite without terminal hyaline lobe ........................0..0 2005. 4

Dips ANNALS OF THE SOUTH AFRICAN MUSEUM

4. Pleopod 1 6: endopodite with styliform apical part, apex equipped with tiny setae
(Big ATE )iig-g sa Ds.c ape Mite eats aie. cates a ee eee ee a N. appletoni sp. nov.

— Pleopod 1 6: endopodite without styliform apical part, apex without setae (Fig. 14F)
N. cingulata (Barnard)

Natalscia mina (Budde-Lund, 1885)
Figs 11-13
Philoscia mina Budde-Lund, 1885: 219. Dollfus, 18955: 351. Stebbing, 1910: 443. Barnard,
1937: 164. Barnard, 1949: 402.
Philoscia warreni Collinge, 1917: 578, pl. 42 (figs 10-20). Collinge, 1920: 477, pl. 27 (fig. 6).
Collinge, 1945: 346.
Philoscia (Setaphora) mina: Barnard, 1932: 242, figs 18b, 19e. Brian, 1953: 9.
Natalscia warreni: Verhoeff, 1942: 64.
Philoscia Warreni: Brian, 1953: 9.
Setaphora mina: Ferrara & Taiti, 1979: 119. Taiti & Ferrara, 1980: 83.
nec Philoscia (Setaphora) mina: Barnard, 1960a: 47.

Material

Budde-Lund Collection. 2 dd, 1 9, Cape, leg. J. F. Drege, BM-1921:
10: 18: 1899-1901 (syntypes).

Collinge Collection. 3 dd, 9 29, Umbilo Bush, near Durban, Natal,
leg. E. Warren, BM-1919: 4: 26: 469-478 (P. warreni; paratypes).

New material. Natal: 4 dd, 1 2, Pietermaritzburg, leg. S. Taiti, 19 April
1980, MZUF-1005; 2 dd, Gillits, leg. S. Taiti, 26 April 1980, MZUF-1006;
1 3d, 2 22, Umhlanga Rocks, coastal forest, leg. S. Taiti and K. C. Thomson,
26 April 1980, MZUF-1007; 17 dd, 30 22, mouth of Mdloti River, man-
grove forest, leg. S. Taiti and K. C. Thomson, 26 April 1980, MZUF-1008;
25 dod, 34 22, Umdloti, leg. S. Taiti and K. C. Thomson, 26 April 1980,
MZUF-1009.

Transkei: ? 1 36, ? 8 99, Umgazana, near Port St. Johns, leg. B. F.
Kensley, 18 August 1974, SAM-A16861.

Description

11 x 4,5 mm long (according to Barnard the maximum dimensions are
13 x 5 mm). Brown colour, more or less suffused with yellowish; a white oval
spot at the base of pereon epimera and medial dark spot on the posterior part
of pereon segments; pleon with a light stripe in the mid-line (Fig. 11); appen-
dages pigmented. Though the light and dark spots are always visible, very pale
specimens (due to isolated chromatophores) are rather common. Eye with
about twenty-two ommatidia. Each pereon segment with ten to twenty gland
pores per side. Noduli laterales with b/c and d/c co-ordinates as in Figure 12A.
Telson (Fig. 12C) with subacute apex. Antenna: ratio of flagellum joints
5 : 3: 3. Pereopod with a flagellar dactylar seta (Fig. 12D).

Male

Anterior pereopods without brushes of spines. Pereopod 7 ischium with a
strong spine on sternal margin (Fig. 12E). Pleopod 1 exopodite with a long

P28,

REVISION OF THE FAMILY PHILOSCIIDAE

m

Fig. 11. Natalscia mina (Budde-Lund, 1885). Adult female

24 ANNALS OF THE SOUTH AFRICAN MUSEUM

q,

0.5 0.5

! nom Mw VeVi Vil A | to ot Ww VeVi vil

i
wy! ! \ re!
ly 1

0,05 mm

0.2mm

Fig. 12. Natalscia mina (Budde-Lund, 1885). A. b/c and d/c co-ordinates. B. Apex of
maxilliped. C. Telson and uropods. D. Pereopod 1 dactylus: d.s.—dactylar seta. E. Pereopod
7 ischium and merus (<¢).

REVISION OF THE FAMILY PHILOSCIIDAE 125)

( \penoacsooH

0,2 mm 0,2mm

Fig. 13. Natalscia mina (Budde-Lund, 1885). A. Pleopod 1 exopodite (¢). B. Pleopod 1
endopodite (d). C. Pleopod 2 (¢). D. Pleopod 1 exopodite in the male from Umgazana.

triangular point bent outward, outer margin sinuose (Fig. 13A); endopodite
with a row of spines and a narrow hyaline lobe at the apex (Fig. 13B). Pleopod
2 as in Figure 13C.

Remarks

As mentioned above, a comparison of type material of Philoscia warreni
and P. mina showed these to be the same species. The original material of
P. mina collected by Drege is labelled ‘Cape’ but, as pointed out by Barnard
(1932: 244), these specimens probably come from Natal.

26 ANNALS OF THE SOUTH AFRICAN MUSEUM

As with Nahia hirsuta, the uncertainties and errors in identification by the
early authors make it impossible to define the exact distribution of Natalscia
mina. It is definitely common in southern Natal.

The male from Umgazana has the outer margin of pleopod 1 exopodite
truncated instead of rounded (Fig. 13D), while all the other characters coincide
with the typical N. mina. This could be a specific difference but the authors
cannot reach a firm conclusion due to the insufficient material.

Natalscia cingulata (Barnard, 1932)
Fig. 14
Philoscia (Setaphora) cingulata Barnard, 1932: 244, figs 160, 17b, 18b, 19b, Brian, 1953: 9.

Philoscia cingulata: Barnard, 1949: 403.
Setaphora cingulata: Ferrara & Taiti, 1979: 118. Taiti & Ferrara, 1980: 83.

_ Material

Barnard Collection. Natal: 1 6, 9 29, Port Shepstone, leg. K. H. Bar-
nard, 1912, BM—1933:1:25:87-92 (syntypes).

New material. Transkei: ? 10 29, Port St. Johns, leg. B. F. Kensley, 14
August 1974, SAM-A16862.

Description

3 6mm long; ¢ ovig. 7 mm long. Colour as described by Barnard: ‘Pale
yellowish, with broad greyish bands across front of head between eyes, and
across the peraeon and pleon segments, on the latter usually interrupted in the
middle line; antennae pale greyish, legs pale without gray marks.’ Eye with
eighteen to twenty ommatidia. Each pereon segment with about ten gland
pores per side arranged along the whole sulcus marginalis. Noduli laterales with
b/c and d/c co-ordinates as in Figure 14A. Telson with pointed apex (Fig. 14B).
Antenna: ratio of flagellum joints 6:5:5. Dactylar seta of pereopods
flagelliform, slightly enlarged at apex (Fig. 14C). Uropod as in Figure 14D.

Male

Pereopods 1-2 carpus and merus with a brush of spines as in N. thomsoni.
Pereopods 3—7 are missing in the only male. Pleopod 1 exopodite with short
triangular posterior point (Fig. 14E); endopodite dorsally with a row of long
spines close to the apex (Fig. 14F). Pleopod 2 endopodite much longer than
exopodite (Fig. 14G).

Remarks

The examination of the type material showed that this species belongs to
the genus Natalscia. N. cingulata closely resembles N. mina from which it
differs by the smaller size, presence of a brush of spines on the anterior
pereopods and the shorter posterior point of pleopod 1 exopodite d. The
colour pattern is also a distinguishing character.

REVISION OF THE FAMILY PHILOSCIIDAE 27

0.5 0.5

| oo Ive VME VU A | nom Ww Vv oVi Vi

0,2mm

0,2mm

Fig. 14. Natalscia cingulata (Barnard, 1932). A. b/c and d/c co-ordinates. B. Telson.
C. Pereopod 1 dactylus: d.s.—dactylar seta. D. Left uropod. E. Pleopod 1 exopodite (¢).
F. Pleopod 1 endopodite (¢). G. Pleopod 2 (¢).

28 ANNALS OF THE SOUTH AFRICAN MUSEUM

The specimens from Port St. Johns are tentatively ascribed to this species,
due to the lack of males.
N. cingulata is recorded from Howick and Port Shepstone.

Natalscia thomsoni sp. nov.
Fig. 15

Material

Natal: 1d, Claridge, near Pietermaritzburg, leg. S. Taiti and K. C.
Thomson, 27 April 1980, holotype SAM-A16863; 2 dd, 2 $9, same data,
paratypes SAM-A16864; 33 dd, 28 2°, same data, paratypes MZUF-1010;
1 ¢, Town Bush, leg. S. Taiti and J. H. Londt, 22 April 1980, paratype
MZUF-1011; 2 dd, 2 2, Karkloof, leg. S. Taiti and J. H. Londt, 24 April
1980, paratypes MZUF-1012.

_ Description

36 8mm long; 2 9mm long. Colour extremely variable: whitish with
isolated, more or less clustered chromatophores; or yellowish with three
longitudinal black stripes: a median one from cephalon to pleon segment 3 and
two at the base of pereon epimera; the lateral stripes have a row of white spots
in the middle; pleon segments 4-5 and telson black; antennae with a white ring
at the base of joints 4 and 5; pereopods, pleopods and uropods slightly
pigmented. Eye with about twenty-five ommatidia. Each pereon segment with
about fifteen gland pores per side arranged along the whole sulcus marginalis.
Co-ordinates of noduli laterales as in Figure 15A. Telson (Fig. 15B) with
almost straight sides. Antenna: ratio of flagellum joints 8:6:5. Pereopods with
a dactylar seta as in Figure 15D.

Male

Pereopods 1-2 merus and carpus with thick brushes of spines (Fig. 15C).
Pereopod 3 merus and carpus with sparse brushes. Pereopod 7 (Fig. 15D)
ischium with concave sternal surface equipped with short spines, closer together
distally; merus with the same spines at the base. Pleopod 1 exopodite
(Fig. 11E) with posterior point bent outwards and with one spine at apex; outer
margin distally concave with a small triangular protrusion, missing in juveniles
(Fig. 11F); endopodite with an elongated hyaline lobe at apex (Fig. 11G).
Pleopod 2 as in Figure 11H.

Etymology
The new species is named for Mr K. C. Thomson, Pietermaritzburg, for his
invaluable help in collecting material.

Remarks

N. thomsoni differs from N. mina by the presence of thick brushes of
spines on pereopods 1-2 6 carpus and merus and the distinctly concave sternal

REVISION OF THE FAMILY PHILOSCIIDAE 29

a

IV Vv vi Vil IV Vv vi Vil

0,2 mm _92mm \ at .

0,2mm

Fig. 15. Natalscia thomsoni sp. nov. A. b/c and d/c co-ordinates. B. Telson and left uropod.

C. Spine of pereopod 1 carpus (3). D. Pereopod 7 (¢): d.s——dactylar seta. E. Pleopod 1

exopodite (d). F. Pleopod 1 exopodite (¢) juvenile. G. Apex of pleopod 1 endopodite (<).
H. Pleopod 2 (¢).

30 ANNALS OF THE SOUTH AFRICAN MUSEUM

margin of pereopod 7 ischium which is straight in N. mina; from N. cingulata
which lacks both a hyaline lobe on the pleopod 1 d endopodite apex and a
triangular protrusion on the outer margin of pleopod 1 d exopodite; from
N. appletoni sp. nov. by the shape of pereopod 7 ¢ ischium (cf. Figs 15D and
17D) and pleopod 1 6 (cf. Figs 15E, G and 17E-F); from N. rotundata sp.
nov. which has a semicircular rather than a triangular telson.

Natalscia rotundata sp. nov.
Fig. 16

Material

Natal: 1 6, Umdloti, leg. S. Taiti and K. C. Thomson, 26 April 1980,
holotype SAM-A16865; 16, 19, same data, paratypes SAM-A16866;
3 dd, 6 $%, same data, paratypes MZUF-1013.

Description

6 5,5mm long; 2 6mm long. Yellowish colour mottled with brown;
antenna with joints 1-3 of peduncle reddish; pereopods and pleopods slightly
pigmented. Eye with sixteen to eighteen ommatidia. Back equipped with
upright setae, closer together on pleon; each pereon segment with five to ten
gland pores per side arranged along the whole sulcus marginalis; b/c and d/c
co-ordinates of noduli laterales as in Figure 16A. Telson semicircular
(Fig. 16B). Antenna: ratio of flagellum joints 10: 7:8. Dactylar seta
(Fig. 16C) as in N. thomsoni.

Male

Pereopods 1—2 merus and carpus with sparse brushes of spines similar to
N. thomsoni. Pereopod 7 (Fig. 16C) ischium with sternal margin swollen in the
middle. Pleopod 1 exopodite (Fig. 16D) with acute posterior point; endopodite
with some spines close to apex (Fig. 16E). Pleopod 2 as in Figure 16F.

Etymology

The specific name refers to the rounded telson, a character which imme-
diately separates N. rotundata from all the other species of Natalscia.

Natalscia appletoni sp. nov.
Fig. 17

Material

Zululand: 1 6, Lake Sibaya Research Station, garden, leg. C. C. Apple-
ton, 7 January 1973, No. 8L, holotype SAM-A16867; 6 dd, 3 2%, same
data, paratypes SAM-A16868; 7 6d, 13 2°, near Lake Sibaya Research
Station, coastal dune forest, leg. C. C. Appleton, 18 December 1973, No. 13E,
paratypes SAM-A16869; 1 2, same data, 29 July 1973, No. 49L, paratype

REVISION OF THE FAMILY PHILOSCIIDAE 31

15 by 1.5 dy.
1 1
0.5 0.5
A T7scaae Seana
| 1

I nom Ww VeVi Vil iv voevi Vil

0,2 mm

Fig. 16. Natalscia rotundata sp. nov. A. b/c and d/c co-ordinates. B. Telson and right
uropod. C. Pereopod 7 (d): d.s.—dactylar seta. D. Pleopod 1 exopodite (¢d). E. Apex of
pleopod 1 endopodite (¢). F. Pleopod 2 (¢).

32 ANNALS OF THE SOUTH AFRICAN MUSEUM

0.5 0.5

I fo ot 6 6IV VeooVE OVI A cb i em Ww oVoVEi Vit

Fig. 17. Natalscia appletoni sp. nov. A. b/c and d/c co-ordinates. B. Telson and right uropod.
C. Spine of pereopod 1 carpus (¢). D. Pereopod 7 ischium and merus (d). E. Pleopod 1
exopodite (d). F. Pleopod 1 endopodite (¢). G. Pleopod 2 (¢).

REVISION OF THE FAMILY PHILOSCIIDAE 33

SAM-A16870; 3 dd, 21292, same data, date ?, No. 18D, paratypes
SAM-A16871; 6 2°, same data, date ?, No. 18L, paratypes SAM-—A16872;
1 2, same data, 10 October 1973, No. 76E, paratypes SAM-—A16873; 2 2°,
same data, date ?, No. 62F, paratypes SAM-A16874; 2 2°, same data, date
?, No. 71Q, paratypes SAM-A16875.

Description

6 mm long. Pale brown colour; antenna with joints 1-3 and proximal part
of 5 yellowish; pereopods and pleopods pigmented. Eye with twenty to
twenty-two ommatidia. Each pereon segment with about ter gland pores per
side arranged along the whole sulcus marginalis; b/c and d/c co-ordinates of
noduli laterales as in Figure 17A. Telson (Fig. 17B) with apex at an obtuse
angle. Antenna with subequal flagellum joints. Dactylar seta of pereopods as in
N. rotundata.

Male

Pereopods 1-2 carpus with brushes of spines as in Figure 17C. Pereopod 7
ischium with sternal margin almost straight and a semicircular depression on
rostral surface (Fig. 17D). Pleopod 1 exopodite (Fig. 17E) with rounded pos-
terior point, outer margin feebly concave; endopodite (Fig. 17F) with a long
styliform distal part, apex equipped with tiny setae. Pleopod 2 as in Figure
£7/G.

Etymology

This species is named for Dr C. C. Appleton, Congella, who collected
these specimens.

Remarks

N. appletoni differs from all the other species of Natalscia by the presence
of a semicircular depression on pereopod 7 ¢d ischium; the largely rounded
posterior point of pleopod 1 6d exopodite and the styliform pleopod 1 6
endopodite apex.

Natalscia sp.
Fig. 18

Philoscia (Setaphora) mina: Barnard, 1960a: 47. Lawrence, 1977: 175.

Material
Bamard Collection.. Transvaal: 16, 3 29, Graskop, leg. R: F.

Lawrence, March 1960, NM-6509; 3 2°, Marieskop, +6 000 ft, leg. R. F.
Lawrence, March 1960, NM-—6505.

34 ANNALS OF THE SOUTH AFRICAN MUSEUM

0.2mm

0,5mm

0,2 mm

Fig. 18. Natalscia sp. A. Telson. B. Pleopod 1 exopodite (d). C. Pleopod 1 endopodite (¢).

Remarks

Barnard (1960a) identified these specimens as Philoscia (Setaphora) mina.
A check of his material showed that these belong to the genus Natalscia but are
not conspecific with N. mina. In fact, the shape of the telson (Fig. 18A) and
general shape of pleopod 1 6 endopodite (Fig. 18C) differ notably from those
of N. mina. The only male available (5 mm long) is probably a juvenile, as
suggested by the shape of pleopod 1 exopodite (Fig. 18B), and has the apices of
pleopod 1 endopodites damaged. Thus a distinctive description is not possible.

In discussing this material, Barnard (1960a: 48) stated that he also saw
‘specimens from Inhaca Island (Delagoa Bay) which appeared conspecific’. It is
not clear whether he referred to the typical N. mina or to the specimens from
Transvaal. The re-examination of the material from Inhaca Island (Mozam-
bique) is necessary for a correct identification.

Genus Barnardoscia nov.

Diagnosis

Sulcus marginalis and gland pores present; one series of noduli laterales on
each side of pereon segments 1-6; two noduli laterales on each side of pereon
segment 7; d/c co-ordinates with a maximum on pereon segment 4 and a lower
one on segment 1 (Figs 20A and 22A). Frontal line absent; supra-antennal line
present. Pereon segments 5-7 with acute posterior angles (Fig. 20C). Pleon
epimera reduced but with long posterior points. Molar penicil of mandible
consisting of a single unbranched seta; outer branch of maxilla 1 with 4+ 6 (5
cleft) teeth; endite of maxilliped with a small penicil (Fig. 20B). Pereopods

REVISION OF THE FAMILY PHILOSCIIDAE 35

with a flagellar dactylar seta. Pleopod exopodites without respiratory areas.
Uropod protopodite grooved on outer margin; insertion of endopodite proxi-
mal to that of exopodite.

Type species
Philoscia (Setaphora) demarcata Barnard, 1932, from South Africa.

Etymology
The genus is named for Dr K. H. Barnard to whom we are indebted for
many important papers on South African terrestrial Isopoda.

Remarks
The re-examination of the type material of Philoscia (Setaphora) demar-
cata showed that this species belongs to a new genus, Barnardoscia, akin to
Natalscia from which it is distinguished by the presence of four noduli laterales
on the pereon segment 7 and the presence of a penicil on maxilliped.
Barnardoscia includes two species, both from South Africa.

KEY TO THE SPECIES OF BARNARDOSCIA

1. Pereopod 7 6: ischium with a depression on the middle of the rostral surface (Fig. 20G).
Pleopod 1 6: exopodite with posterior point equipped with one or two apical spines
(VEE, DIC) 0 \o-cg este Ceska nar ara OI ne na co i a a ee B. demarcata (Barnard)

— Pereopod 7 ¢: ischium with a depression close to the base of the rostral surface (Fig. 22C).
Pleopod 1 6: exopodite without posterior point and spines (Fig. 22D) B. maculata sp. nov.

Barnardoscia demarcata (Barnard, 1932)
Figs 19-21
Philoscia (Setaphora) demarcata Barnard, 1932: 244, figs 16q, 17c, 18a, 19b.

Philoscia demarcata: Barnard, 1937: 164. Barnard, 1949: 402. Brian, 1953: 9.
Setaphora demarcata: Ferrara & Taiti, 1979: 118. Taiti & Ferrara, 1980: 83.

Material

Barnard Collection. Natal: 4 66, 3 2%, (in fragments), Pietermaritz-
burg, leg. K. H. Barnard, 1917, BM—1933:1:25:93—97 (syntypes).

New material. Natal: 17 dd, 42 22, Claridge, Pietermaritzburg, leg.
S. Taiti, 27 April 1980, MZUF-1014; 5 dd, 25 22, Cedara, near Pieter-
maritzburg, leg. S. Taiti, 20 April 1980, MZUF-1015; 23 6d, 34 22, Kar-
kloof, near Howick, leg. S. Taiti and J. H. Londt, 24 April 1980, MZUF-1016;
1 3d, 3 22, Town Bush, near Pietermaritzburg, leg. S. Taiti and J. H. Londt, 22
April 1980; MZUF-1017; 1 36, 1 2, Lundys Hill, about 70 km W of Pieter-
maritzburg, leg. S. Taiti and K. C. Thomson, 25 April 1980, MZUF- 1018; 1 d,
Bulwer, leg. S. Taiti and K. C. Thomson, 25 April 1980, MZUF-1019.

Description

3 7,5 mm long; 2 9 mm long. Dark brown with light mottling. Eye with
about twenty large ommatidia. Each pereon segment with seventeen to twenty

36 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 19. Barnardoscia demarcata (Barnard, 1932). Adult female.

gland pores per side; b/c and d/c co-ordinates of noduli laterales as in Figure
20A. Pereon segments 2—4 demarcated in some female specimens. Telson
(Fig. 20D) with concave sides, narrowly rounded apex. Flagellum joints of
antenna subequal.

Male

Cephalon with a bulbous profrons (compare Fig. 20E, ¢°, and Fig. 20F,
36). Pereopods 1-3 merus and carpus with sparse brushes of spines similar to
those of Natalscia thomsoni. Pereopod 7 ischium (Fig. 20G) with a large

REVISION OF THE FAMILY PHILOSCIIDAE 37

I to om WwW VeVi Vil A I to ot Ww VeVi Vil

0,05 mm

0,5mm

0.5mm

0.5mm

Fig. 20. Barnardoscia demarcata (Barnard, 1932). A. b/c and d/c co-ordinates. B. Apex of
maxilliped: p.—penicil. C. Pereon epimera 4~7, right. D. Telson and left uropod. E. Cepha-
lon from above (2). F. Cephalon from above (6). G. Pereopod 7 ischium and merus (<¢).

38 ANNALS OF THE SOUTH AFRICAN MUSEUM

0,05 mm
0,2mm
0,2 mm
B
D
0,2 mm
0,2 mm
E a

0,2 mm

0,2 mm

0,2 mm

Fig. 21. Barnardoscia demarcata (Barnard, 1932). A. Pleopod 1 exopodite (d). B. Pleopod 1

exopodite (d) in another specimen. C. Pleopod 1 exopodite (¢) in one syntype from

Pietermaritzburg. D. Pleopod 1 exopodite (¢), juvenile 4 mm long. E. Pleopod 1 exopodite

(3), juvenile 4,7 mm long. F. Pleopod 1 endopodite (d). G. Apex of pleopod 1 endopodite
(3). H. Pleopod 2 (d). I. Pleopod 5 exopodite (¢).

REVISION OF THE FAMILY PHILOSCIIDAE 39

depression on the middle of the rostral surface. Pleopod 1: exopodite varies in
shape (Fig. 21A—C). In some specimens the posterior, rather than the external,
point, is the most developed (Fig. 21A), while in other specimens the posterior
point is more rounded and less pronounced, and the external point is the most
developed (Fig. 21B). During the course of development (Fig. 21D-E) the
external point progressively increases in size with respect to the posterior point.

Apex of pleopod 1 endopodite (Fig. 21F—G) with spines and a small
triangular lobe. Pleopod 2 as in Figure 21H. Pleopod 5 exopodite with straight
medial edge (Fig. 211).

Remarks

The suture at the base of the pereon epimera 2-4 @ is a sporadic trait, as it
is in every other species of terrestrial isopod presenting this character.

The external point of the pleopod 1 exopodite d develops progressively
with the growth of the specimen. This appendage also varies greatly in shape in
fully adult males (see Fig. 21A—C) of the same lot.

Barnardoscia maculata sp. nov.
Bigs 22

Material

Nataly ig, near Bulwer, leg. S. Waitt, 25 April 1980, holotype
SAM-A16876; 2 66, 2 22, same data, paratypes SAM-A16877; 7 do,
12 22, 20 juv., same data, paratypes MZUF-1020.

Description

6 7mm long; ¢ 7,5 mm long. Iron-grey with lighter mottling, a white
stripe at the base and a pale spot in the middle of pereon epimera. Eye with
about twenty-five small ommatidia. Each pereon segment with ten to twelve
gland pores per side, arranged along the whole length; b/c and d/c co-ordinates
of noduli laterales as in Figure 22A. None of the females with pereon epimera
2-4 demarcated. Telson with almost straight sides, rounded apex (Fig. 22B).
Antenna with flagellum joints subequal.

Male

Cephalon as in B. demarcata. Pereopods 1-2 carpus and merus with sparse
spines of the same type as in Natalscia thomsoni. Pereopod 7 ischium with a
depression close to the base on the rostral surface (Fig. 22C). Pleopod 1
exopodite (Fig. 22D-E) typically without posterior point and apical spine;
endopodite (Fig. 22F) similar to that of B. demarcata. Pleopod 5 exopodite
with oblique medial edge (Fig. 22G).

Etymology
The specific name refers to the characteristic coloration of this species.

40 ANNALS OF THE SOUTH AFRICAN MUSEUM

1
0.5 0.5

i WM VoVI Vil A es | ee || |

0,05 mm

Fig. 22. Barnardoscia maculata sp. nov. <A. b/c and d/c co-ordinates. B. Telson.
C. Pereopod 7 ischium and merus (d). D. Pleopod 1 exopodite (¢). E. Pleopod 1 exopodite
(3), juvenile 4,2 mm long. F. Apex of pleopod 1 endopodite (¢). G. Pleopod 5 exopodite (¢).

REVISION OF THE FAMILY PHILOSCIIDAE 41

Remarks

The new species is very close to B. demarcata from which it differs by the
(i) different colour pattern, (11) smaller and more numerous ommatidia,
(iii) fewer number of gland pores, and (iv) male modifications: pereopod 7
ischium with a smaller depression much closer to the base, pleopod 1 exopodite
with truncated apex, shape of pleopod 5 exopodite.

SYSTEMATIC DISCUSSION

To date the following species were known in South Africa:

Aphiloscia vilis (Budde-Lund, 1885)
Benthanops fulva Barnard, 1932

Nahia hirsuta (Budde-Lund, 1906)
Setaphora mina (Budde-Lund, 1885)
Setaphora cingulata Barnard, 1932
Setaphora demarcata Barnard, 1932
Setaphora ocellata Barnard, 1960
Chaetophiloscia elongata (Dollfus, 1884)
Philoscia muscorum (Scopoli, 1763)

All the species listed were re-examined, with the exception of the last two
which are of Mediterranean origin and whose presence in South Africa—if
confirmed—is due to importation by man. In the authors’ opinion the Chaeto-
philoscia elongata quoted by Dollfus (1895) corresponds to Nahia hirsuta, while
the records of Philoscia muscorum from Natal, ‘Hilton Road and Mid-Illovo’
(Collinge 1920), are definitely a misidentification (see also Verhoeff 1942).

Five new species have also been described. Of even greater interest is the
fact that it was possible to assign each species to the correct genus, each of
which was redefined according to modern criteria.

As a result, Benthanops Barnard, 1932, and Ctenoscia Verhoeff, 1928,
which appeared to be identical according to the early descriptions, proved to be
distinct even if extremely close genera.

The study of most of Barnard’s material allowed the authors to redefine
the genus Nahia Budde-Lund, 1908, clarify the identifying traits of the only
species known to date, N. hirsuta, and describe a new one, N. louwi.

The task of correctly collocating the four species assigned by Barnard
(1932, 1960a) to Setaphora Budde-Lund, 1908, was complicated above all by
the uncertain diagnosis of this genus. It was established by Budde-Lund (1908)
for Philoscia suarezi Dollfus, 1895, from Diego-Suarez in Madagascar.
Budde-Lund reportedly found this species not only in Madagascar but on
several islands of the west Indian Ocean, and later (1910) in east Africa
(Kibonoto, Meru and Kibosho). In 1913, Budde-Lund published a list of
species belonging to the genus Setaphora, including two new species from the
Seychelles: S. ovata and S. pallidemaculata. Herold (1931) redefined the genus

42 ANNALS OF THE SOUTH AFRICAN MUSEUM

and described some species from the Sunda islands without, however, adding
any truly unequivocable traits. The same can be said for the diagnosis
presented by Barnard (1932). While Taiti & Ferrara (1980) and Ferrara & Taiti
(in press) presented a diagnosis of this genus based on the study of the syntypes
of S. ovata and S. pallidemaculata, they affirmed that only the study of the type
species S. suarezi could provide an exact diagnosis.

As the problem of identifying the genus Setaphora reappeared with the
South African material, it was decided to check the specimens of Philoscia
suarezi on which Budde-Lund (1908) had based the new genus. The study of
this material (BM No. 1921:10:18:2351—2358) did little clarify the problem as
the specimens contained in the tube (i) come from two different collections
(Voeltzkow and Alluaud) and two different localities (Nossi Bé and Diego-
Suarez) (Ellis & Lincoln 1975: 92), (ii) are in poor condition and unable to
withstand much handling, (iii) belong to two distinct species, one of which
probably corresponds to the Philoscia suarezi described by Dollfus and the
other to the Philoscia suarezi intended by Budde-Lund, and (iv) allow no
conclusions on the genus (or genera?) to which they belong. Furthermore, the
Danish author had established Setaphora on specimens—which he identified as
P. suarezi—from other islands of the western Indian Ocean (Réunion,
Comoro, Fundu) as well, but which probably belong to other species. It is
indicative that the east African specimens identified by Budde-Lund (1910) as
S. suarezi (NRS Is. 5826: Meru) not only do not correspond to the description
of P. suarezi by Dollfus (1895a) but are not conspecific with the British
Museum specimens. In fact, they belong to Afrophiloscia uncinata (Ferrara,
1974).

Until further information is available, the authors consider their diagnosis of
Setaphora (Taiti & Ferrara 1980; Ferrara & Taiti in press) as valid. The four
species assigned by Barnard to the genus belong to three distinct genera, none of
which fits the proposed diagnosis of Setaphora. In fact, Setaphora ocellata belongs
to the genus Afrophiloscia Taiti & Ferrara, 1980, S. demarcata is the type species
of the new genus Barnardoscia, which also includes B. maculata sp. nov., while S.
mina and S. cingulata belong to the genus Natalscia Verhoeff, 1942, together with
the new species N. thomsoni, N. rotundata, and N. appletoni.

ZOOGEOGRAPHIC DISCUSSION

At present the list of South African Philosciidae—excluding Chaetophi-
loscia elongata and Philoscia muscorum for the above-mentioned reasons—is as
follows:

Aphiloscia vilis (Budde-Lund, 1885) Mozambique, Zimbabwe, Trans-
vaal, Natal, Zululand, Cape Pro-
vince, ?0vamboland

Benthanops fulva Barnard, 1932 Cape Province

43

REVISION OF THE FAMILY PHILOSCIIDAE

‘Sp1O991 POWIYUOSUN IO [NJIJQnop oa}eOIpUI

J o82

Ce OZ Od)

Oo
|

uopuoy 4sn5 43 2

NVIDO a JINIAOYd Jdvd

unqing

Bingzjinwiajarg Pee, Be

NVIGNI

lajuofwaojg
O

JIVL1S J4us JONVAO

YoldsdV HLNOS

Bingsauunyor o

oyndow HP ® DI104914 O
® ve
®

IVVASNVJUL

INOIIWVYZOW
VNVMSLO4d

okpmoing &

IMAVAaWIZ

AANgsijoS
fe)

syieu uonsanb ayy, ‘saiseds viynxy pue vjv]Ja90 vIosopydosy ‘sya viosopydy Jo sp10d01 UONNGIsIp BuUIMOYs voy UsOYINos Jo dep ‘¢z ‘SY

Ao9l

uMO| adn+

es) SSN ic
DeNSa IG ONY N im
DJO]J990 DIDsSOjIydoisy Ww

SIJIA DIDSO|IYdY @

NVjIDO

|
(vid WYN) DILNVILV

‘VYM'S So¥Z

uior1idpy\ yo dIdo4y

POUPUIM O

44 ANNALS OF THE SOUTH AFRICAN MUSEUM

Afrophiloscia ocellata Barnard, 1960) Transvaal

Nahia hirsuta (Budde-Lund, 1906) ?Natal, ?Zululand, Cape Pro-
vince

Nahia louwi sp. nov. Cape Province

Natalscia mina (Budde-Lund, 1885) Natal, ?Transkei, ?Cape Province

Natalscia cingulata (Barnard, 1932) Natal, ?Transkei

Natalscia thomsoni sp. nov. Natal

Natalscia rotundata sp. nov. Natal

Natalscia appletoni sp. nov. Zululand

Natalscia sp. Transvaal

Barnardoscia demarcata (Barnard, 1932) Natal

Barnardoscia maculata sp. nov. Natal

The recorded distribution of each species is shown in Figures 23-25. The
records of Aphiloscia vilis from Ovamboland, Nahia hirsuta from Natal and
Zululand, and Natalscia mina from Cape Province are doubtful due to the
uncertainties and errors in identification by the early authors. Because of the
insufficient material the records of Natalscia mina and N. cingulata from
Transkei need confirmation. Thus far, only Aphiloscia vilis has been found
outside South Africa as well.

The distribution of South African philosciids is of particular interest: of the
six known genera, four (Benthanops, Nahia, Natalscia, and Barnardoscia) are
exclusive to this region. Benthanops, however, shows many affinities to the

A)

CAPE TOWN : @ Stellenbosch

A
© Caledon

Fig. 24. Map of the south-western Cape Province showing distribution records of Benthanops
fulva.

REVISION OF THE FAMILY PHILOSCIIDAE 45

MOZAMBIQUE

O Pretoria

O Johannesburg

TRANSVAAL SWAZILAND

ORANGE FREE STATE
ZULULAND

fe) Bloemfontein

e ietermaritzburg

A
=

@ Natalscia mina
ee re AN. cingulata
AN. thomsoni
WN. rotundata
32°S INDIAN VN. appletoni
CAPE OCEAN EIN. sp.
PROVINCE ® Barnardoscia demarcata

)
(X East London OB. maculata

Q Port Elizabeth

34°S
ZOE

Fig. 25. Map of eastern South Africa showing distribution records of Natalscia and Barnard-
oscia species. The question marks indicate doubtful or unconfirmed records.

46 ANNALS OF THE SOUTH AFRICAN MUSEUM

South American Benthana Budde-Lund, 1908, and allied genera and to the
palearctic genus Ctenoscia, while Nahia belongs to the Plymophiloscia group
(Vandel 1973a, 19736) which occurs throughout the Orient and Australia. Of
this group, only Bilawrencia occidentalis Ferrara & Taiti, in press, from
Seychelles is known in Africa and surrounding islands. Natalscia and Barnardo-
scia are very close to each other and appear to be isolated from all the other
African philosciids. Aphiloscia and Afrophiloscia are the only genera found in
other African countries. The former is distributed throughout central and east
Africa, Madagascar and other islands of the western Indian Ocean, while the
latter occurs in Kenya and Tanzania. The fact that both genera are found only
in the eastern part of South Africa is proof of the faunistic continuity of this
area with east Africa.

TABLE 1
Synopsis of the Afrotropical genera of Philosctidae.

0) ° aS)
7, n 7 & = a fe)
& x) — = oH 3 = a S
3 3 ios} — 3 ie) a 1S oOo S
ee we iS) te = — a a
n S & ~ & 3 = v ‘5 mS = = <= s
os Se Ss = e 9 — a = secs
CO ws eb ee = = 4} = =| = oO. Oo v ae oO o «Of
asses =) iF a Bs 2 S&S %S = Ss cone
useuove ¢ S ¢@ = &¢ So & £5
= Som © = §€ § S22 =) 27a.
s SOG, FM wv o &2 © Os 7) pea a ©
Mmrnegrotrg s gc 3 CG Be & @ 2. 25a0
Aphiloscia + + sp aD S c - + F¢
Buddelundiscus eaten arn ? ? =p EOD 2 c = = &
Massaiscia Sea sr lie & e ap ae YD ? c = a
Komatia Pere wees oa ? ? arp) S € =:
Perinetia Uap ee Gee x z spe it Ss. SCL) ye we
Pleopodoscia LPT area ? ? a ae d c = = =
Didima (3) + +r = ss &£ dQ)
Rennelloscia + + I, IV = oe ad S Cc + - =
Sechelloscia —- + II, IV =) SE S S + - =
Afrophiloscia = oF IW ee ee ete ak S c =)
Setaphora =F OE IV = sp. S c ——- =
Barnardoscia (3) + + I, IV Sap S c sp
Natalscia (3) <p oF I, IV = spr S € = = =
Uluguroscia spr IV = sp. it S c =
Bilawrencia = sr = ee I d c oe = SE
Helenoscia (3) —- + v = os Ff d c +
Nahia (3) =e SF = = op ff S c =) eee
Benthanops (3) (4) + =e IV = ote sf d©)imeguse = = =
Congophiloscia + = II, IV = = F&F d c Se
Gabunoscia + + II, IV - - d Cc - - ¢
Togoscia aF ap II, IV ae eee d Cc ee
Vandelophiloscia — +f (ID). IY = d S - = #
Zebrascia — + II, IV a a Ne S Cc Se =
Arcangeloscia + =F IV = = Ff d c a a
EE
+, present; —, absent; =, at the same level; +, spaced; c, cleft; d, dichotomized; p, pro-

duced; r, reduced; s, simple; se, serrate.
(1) 3+2 teeth; (2) 2 additional plumose teeth; (3) a dactylar seta is present; (4) a single
large ommatidium; (5) a tuft of plumose setae each arising separately.

REVISION OF THE FAMILY PHILOSCIIDAE 47

SYNOPSIS OF THE AFROTROPICAL GENERA OF PHILOSCIIDAE

In a previous paper on African philosciids (Taiti & Ferrara 1980), the
diagnosis of four genera was either incomplete (Benthanops, Nahia) or non-
existent (Natalscia, Barnardoscia) due to the lack of material, and thus it was
thought useful to present a new synopsis of the known afrotropical genera
(Table 1).

ACKNOWLEDGEMENTS

We are indebted to the Department of National Education of the Republic
of South Africa for having granted a fellowship to one of us (S. Taiti). We wish
to thank the following persons for the loan of material: Drs T. H. Barry and
P. A. Hulley (South African Museum), B. H. Lamoral and J. H. Londt (Natal
Museum), and Miss J. P. Ellis (British Museum (Natural History)). We are
grateful to Dr A. Andersson of the Naturhistoriska Riksmuseet, Stockholm, for
his help during a visit to the Museum. Particular thanks are due to Mr P. Paoli
(Centro di Studio per la Faunistica ed Ecologia Tropicali, Florence) for the
drawings.

REFERENCES

APPLETON, C. C. 1974. A check-list of the flora and fauna of the Gladdespruit, Nelspruit district
eastern Transvaal. News]. limnol. Soc. sth Afr. 22: 49-58.

ARCANGELI, A. 1950. Isopodi terrestri. Explor. Parc natn. Albert. Miss. H. Damas 15: 1-80.

BARNARD, K. H. 1932. Contributions to the crustacean fauna of South Africa. 11. Terrestrial
Isopoda. Ann. S. Afr. Mus. 30: 179-388.

BARNARD, K. H. 1937. New South African woodlice (Isopoda Terrestria). Ann. Natal Mus.
8: 155-165.

BARNARD. K. H. 1949. Descriptions of new species and records of woodlice from South Africa
and Southern Rhodesia. Ann. Natal Mus. 11: 395-403.

BARNARD, K. H. 1956. A new species of Akermania (Isopoda) from Southern Rhodesia. Ann.
Natal Mus. 13: 435-436.

BARNARD, K. H. 1960a. Terrestrial Isopoda from the Transvaal. Ann. Natal Mus. 15: 45-55.

BARNARD, K. H. 19606. A collection of terrestrial Isopoda from Mt. Gorongoza, Portuguese E.
Africa. Ann. Natal Mus. 15: 505-511.

BriAN, A. 1953. Determinazione di Isopodi marini e terrestri provenienti dall’Angola,
collaggiunta della descrizione di una rara specie di Porcellio raccolta nella Guinea Porto-
ghese. Isopodi d’Angola raccolti dal Prof. Dartevelle. Genova: Tip. C. Badiall.

BupDDE-LUND, G. 1885. Crustacea Isopoda terrestria per Familias et Genera et species descripta.
Hauniae.

BuppE-LunD, G. 1898. Die Land-Isopoden Ost-Afrikas. Thierwelt Deutsch Ost-Afrika
4: 1-10.

BuppDE-LunpD, G. 1906. Die Landisopoden der deutschen Stidpolar-Expedition 1901-1903. Mit
Diagnosen verwandter Arten. Dt. Siidpol.-Exped. 9(2): 69-92.

BuppE-Lunpb, G. 1908. Isopoda von Madagascar und Ostafrika mit Diagnosen verwandter
Arten. Wiss. Ergebn. Reise Ostafr. 2(4): 263-308.

BuppE-LunpD, G. 1909. Land-Isopoden. Jn: ScHuLTzE, L. Zoologische und Anthropologische
Ergebnisse einen Forschungsreise in westlischen und zentralen Stidafrika 2: 53-70. Denk-
schr. med.-naturw. Ges. Jena 14.

BuppeE-Lunp, G. 1910. Isopoda. In: Sjastedt’s Kilimandjaro-Meru Expedition. 21. Crustacea.
3: 3-20. Stockholm.

48 ANNALS OF THE SOUTH AFRICAN MUSEUM

BubDDE-LuUND, G. 1913. The Percy Sladen Trust Expedition to the Indian Ocean in 1905, under
the leadership of Mr. J. Stanley Gardiner. (IV. No. 22). Terrestrial Isopoda particularly
considered in relation to the distribution of the Southern Indo-Pacific species. Trans. Linn.
Soc. Lond. (Zool.) 15: 367-394.

CoLuincE, W. E. 1917. Contributions to a knowledge of the terrestrial Isopoda of Natal. Part I.
Ann. Natal Mus. 3: 567-585.

CoLLinGE, W. E. 1920. Contributions to a knowledge of the terrestrial Isopoda of Natal. Part
III. Ann. Natal Mus. 4: 471-490.

CoLLinGE, W. E. 1945. Note on some South African Terrestrial Isopoda. Ann. Mag. nat. Hist.
(11) 12: 344-347.

Do.tFus, A. 1884. Les Espéces frangaises du genre Philoscia Latreille (Crustacés isopodes du
groupe des Cloportides). Bull. Soc. Etud. scient. Paris 7: 1-4.

Do tFus, A. 1895a. Mission scientifique de M. Ch. Alluaud dans le territoire de Diégo-Suarez
(Madagascar-Nord). Avril-Aotit 1893. Isopodes terrestres recueillis 4 Diégo-Suarez, 4
Tamatave et ala Réunion. Mém. Soc. zool. Fr. 8: 180-188.

Do.iFrus, A. 18955. Voyage de M. E. Simon dans 1|’Afrique Australe (Janvier-Avril 1893).
Crustacés Isopodes terrestres. Mém. Soc. zool. Fr. 8: 345-352.

Exuts, J. P. & LINcoLn, R. J. 1975. Catalogue of the types of terrestrial isopods (Oniscoidea)
in the collections of the British Museum (Natural History). II. Oniscoidea, excluding
Pseudotracheata. Bull. Br. Mus. Nat. Hist. (Zool.) 28: 65-100 (a-e).

FERRARA, F. 1974. On some terrestrial Isopods from Tanzania. Monitore zool. ital. (n.s.) Suppl.
5: 309-324.

FERRARA, F. & Tait1, S. 1979. A check-list of terrestrial Isopods from Africa (south of the
Sahara). Monitore zool. ital. (n.s.) Suppl. 12: 89-215.

FERRARA, F. & Tait1, S. In press. Contribution a4 l’étude de la faune terrestre des iles
granitiques de l’archipel des Séchelles (Mission P. L. G. Benoit-J. J. van Mol 1972).
Isopodi terrestri. Annls Mus. r. Afr. cent. Sér. 8vo (Sci. Zool.) 235.

GERSTAECKER, A. 1873. Gliederthiere (Insecten, Arachniden, Myriapoden und Isopoden).
Isopoda. In: Baron Carl Claus von der Decken’s Reisen in Ost-Africa in den Jahren
1859-1865. 3(2): 525-528. Leipzig und Heidelberg.

HEROLD, W. 1931. Land-Isopoden von den Sunda-Inseln. Ausbeuten der Deutschen Limnologi-
schen Expedition und der Sunda-Expedition Rensch. Arch. Hydrobiol. Suppl. 9: 306-393.

LAWRENCE, R. F. 1977. Woodlice (Crustacea, Isopoda terrestria) recorded by the late K. H.
Barnard from Kruger National Park and its neighbourhood. Koedoe 20: 175-177.

Lemos DE Castro, A. 1958a. Benthanoscia longicaudata, a new genus and species of terrestrial
isopod of the family Oniscidae (Isopoda, Oniscoidea). Am. Mus. Novit. 1884: 1-7.

LEMos DE Castro, A. 19585. Reviséo do género Benthana Budde-Lund, 1908 (Isopoda,
Oniscidae). Archos Mus. nac., Rio de J. 46: 85-118.

SCHMOELZER, K. 1974. Landisopoden aus Zentral- und Ostafrika (Isopoda, Oniscoidea). Sber.
Akad. Wiss. Wien (Math.-nat. Kl., Abt. 1) 182: 147-200.

ScopoLi, G. A. 1763. Entomologia Carniolica exhibens Insecta Carnioliae indigena et distributa
in Ordines, Genera, species, varietates. Methodo Linnaeana. Vindobonae.

STEBBING, T. R. R. 1910. General Catalogue of South African Crustacea. Ann. S. Afr. Mus.
6: 437-447.

STEBBING, T. R. R. 1922. Isopoda and Amphipoda from Angola and South Africa. Géteborgs
K. Vetensk.-o. VitterhSamh. Handl. (4) 25(2): 1-16.

TaiT1, S. & FERRARA, F. 1980. The family Philosciidae (Crustacea Oniscoidea) in Africa, south
of the Sahara. Monitore zool. ital. (n.s.) Suppl. 13: 53-98.

VANDEL, A. 1973a. Les Isopodes terrestres de |’Australie. Etude systématique et biogéogra-
phique. Mém. Mus. natn. Hist. nat., Paris (n.s.) (Sér. A, Zool.) 82: 1-171.

VANDEL, A. 1973b. Les Isopodes terrestres (Oniscoidea) de la Mélanésie. Zool. Verh.
125: 1-160.

VERHOEFF, K. W. 1942. Athiopische Isopoda terrestria des Hamburger Zoologischen Museums.
84. Isopoden-Aufsatz. Zool. Anz. 140: 1-26 (Teil 1), 61-87 (Teil 2), 149-163 (Teil 3,
Schluss).

———
—
7
7 x
my
oe

i =
+
<4 2
be .
«
= te, a
, ;
3 : t

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 14-15SA
Nucula (Leda) bicuspidata Gould, 1845: 37.
Leda plicifera A. Adams, 1856: SO.
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

(c) Scientific names, but not their vernacular derivatives
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.

STEFANO TAITI
&
FRANCO FERRARA

~ REVISION OF THE FAMILY PHILOSCIIDAE

(CRUSTACEA, ISOPODA, ONISCOIDEA)
FROM SOUTH AFRICA

90 PART 2. AUGUST 1982 ISSN 0303-2515

OF THE SOUTH AFRICAN

— ee

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
(c) Table of contents giving hierarchy of headings and subheadings
(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
(Gi) 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, 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
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.

FiscHerR, P.-H., DuvAL, M. & RaAFFy, 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.

Konn, 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 90 Band
August 1982 Augustus
Part 2 IDee!

PATTERNS OF UNGULATE MORTALITY
AND UNGULATE MORTALITY PROFILES
FROM LANGEBAANWEG (EARLY PLIOCENE)
AND ELANDSFONTEIN (MIDDLE PLEISTOCENE),
SOUTH-WESTERN CAPE PROVINCE,
SOUTH AFRICA

By
RICHARD G. KLEIN

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

ily WEES), S-3), SCD, 4-8. Guat, SUES, 5, 7D).
GG, toi), HOM), 8, O02, 1), 1S),
NCD, S. 7, eat), ICS), 240), 27, SUB), AG), 26, EO), 4S)

EDITOR/REDAKTRISE
Ione Rudner

Copyright enquiries to the South African Museum

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 038 9

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, kKaap

PATTERNS OF UNGULATE MORTALITY AND UNGULATE
MORTALITY PROFILES FROM LANGEBAANWEG (EARLY
PLIOCENE) AND ELANDSFONTEIN
(MIDDLE PLEISTOCENB),
SOUTH-WESTERN CAPE PROVINCE, SOUTH AFRICA

By
RICHARD G. KLEIN

Department of Anthropology, University of Chicago

(With 16 figures and 4 tables)

[MS accepted 27 May 1982]

ABSTRACT

An ungulate species in which females usually have one young (or less) per year will have a
catastrophic (= survivorship = ‘lx’) age profile with a down-staircase shape in which successively
older age classes contain progressively fewer individuals. The corresponding attritional
(= mortality = ‘dx’) profile will tend to be U-shaped, with a large peak in the youngest age class
and a second smaller peak beyond 40-50 per cent of potential individual lifespan, or L-shaped,
with a large peak in the youngest age class and no obvious peaks thereafter. A species in which
females regularly produce more than one young per year will have catastrophic and attritional
profiles that both exhibit the down-staircase form. In samples of fossil ungulates, reasonably
accurate age (mortality) profiles may be calculated from dental crown heights, using quadratic
formulae that assume that the rate of dental wear slows with age. Without sound contextual
information, available from Langebaanweg (early Pliocene) but not from Elandsfontein (Mid-
dle Pleistocene), age profiles may be difficult to interpret, even when their form (catastrophic v.
attritional) is clear. Establishing whether bones accumulated seasonally or not and determining
the sex ratio in fossil ungulate samples is far more difficult than establishing age profiles. In
conjunction with contextual information, age profiles in fossil samples may be used to infer
basic biological facts in a species, such as the common rate of reproduction. The major
impediment to broader application of age (mortality) profile analysis in palaeontology and
archaeology is the rarity of suitably large samples.

CONTENTS

PAGE
Woe aS EY uO PE Me et Avot de ga icc ah eo ng) eel shave say che) c Rens shel rere a oremerorea ei eta oe casio 50
Some general principles of population dynamics in large mammals...................... 50
Age structure and mortality in extant populations of large mammals .................... 54
Age structure and mortality in some fossil populations of large mammals ................ 58
Age profiles produced by predation on some populations of large mammals.............. 60
Peamanescie acc Ot a fossil unoulate at death... Wee ye eee bye eee et eee ee 64
PEC aaliVie CHOSGIL SITE Misha ie ah bye UW DE ie Seno re baache Boor dye ciel aan oe 68
TILE TELE ICSI OVTIIS TIN HOSES ICT ee are ees nee Pe eR ere eed Orne en Pts SEMIN nares a Aa 70
The Langebaanweg and Elandsfontein ungulate samples and age profiles................ TAL
Interpretation of the Langebaanweg and Elandsfontein age profiles..................... 81
Season of bone accumulation at Langebaanweg and Elandsfontein...................... 87
eA SHELLS EOSSII| GAMIPICS) ar ov eetasiers cio ae es Seyi ey es eeaece Arete es ous Stes 88
TIS O ISIE cose ete ee ER een ditd Le Girls sGean. ok SHEE Pane Sain AL REO ER cancer 4 90
POMC HE CINETILG Meer rat ee Ter. Soo. te ni Cena tamnIN NEN eee mnR Eee A tae 8 91
eae GWE Pies eh PRIA DAS datte g RAE Ue Sis cb ie davon Oar am Plana, Wiehe UeagaNe Sauce AUR aa OP epee 92

49

Ann. S. Afr. Mus. 90 (2), 1982: 49-94, 16 figs, 4 tables.

50 ANNALS OF THE SOUTH AFRICAN MUSEUM

INTRODUCTION

The purpose of this paper is twofold: first, to review those aspects of
ungulate mortality that are of special interest to palaeontologists and second, to
construct and interpret ungulate mortality profiles from the fossil sites of
Langebaanweg (early Pliocene) and Elandsfontein (mid-Pleistocene) in the
south-western Cape Province of South Africa. Both are key sites for under-
standing Late Cenozoic mammalian evolution in Africa, and much of their
importance derives from the fact that they have provided unusually large
samples of well-preserved bones. These samples are useful not only for under-
standing the origins and meaning of the Langebaanweg and Elandsfontein fossil
accumulations but also for illustrating the potential and the limitations of
mortality profile analysis based on fossil ungulate remains.

Mortality profile analysis is essentially an aspect of the field of population
~dynamics whose central tool is generally the ‘life table’ as discussed, for
example, by Deevey (1947). The emphasis here, however, will not be on ‘life
tables’ but on mortality profile shape as a means of determining cause of death
in fossil samples. The discussion here builds on and expands concepts and
insights provided by many other investigators, particularly Deevey (1947) and
Spinage (1972a) with respect to mortality patterns in general, and Kurtén
(1953) and Voorhies (1969) with respect to mortality patterns in fossil mam-
mals.

SOME GENERAL PRINCIPLES OF POPULATION DYNAMICS
IN LARGE MAMMALS

A convenient way of illustrating some basic principles of large mammal
population dynamics is to formulate a hypothetical population such as the one
whose age structure is presented in the column labelled ‘Ix’ in Table 1.
Maximum individual longevity in this population is 9 years. Each year 500 new
individuals are born.

If we require that the overall population size remain constant from year to
year, the 500 births will have to be balanced by 500 deaths. If we further
require that the age structure remain unchanged, the 500 deaths will have to be
strictly apportioned among the various age classes. More specifically, the
number of deaths in each age class will have to equal the nusuver of individuals
that survive to that age class minus the number that survive to the next age
class. The resulting death tallies are listed in the ‘dx’ column of Table 1, and
the relationship between corresponding ‘lx’ and ‘dx’ values is illustrated graphi-
cally in Figure 1. Although the requirements of a constant population size and
constant age structure may appear arbitrary, natural populations adapting to
stable environments probably do tend to both a constant size and a constant
age structure.

The ‘dx’ entries in Table 1 are measures of absolute mortality in successive
age classes. Dividing each ‘dx’ entry by its corresponding ‘Ix’ entry provides a

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN Sil

TABLE 1

The age structure of a hypothetical population of large mammals in which females give birth
once a year, total births are 500 per year, and potential individual longevity is 9 years.

NUMBER OF INDIVIDUALS

Age (x) Number of Number of Rate of
(years) live dead mortality

individuals individuals %

(Ix) (dx) (qx)

0-1 500 250 50

1-2 250 25) 10

2-3 US jy) 10

3-4 203 20 10

4-5 183 37) 20

5-6 146 44 30

6-7 102 41 40

7-8 61 3 60

8-9 24 24 100

9-10 0 0
500 500
400 400
300 300
200 200
100 100
0,5 20 49 65 £85 0:3) -2;0,, .4,9. 6,0: 8

CLASS MIDPOINTS (years)

Fig. 1. Left. Blank bars: the age structure of a hypothetical population of large mammals in
which females give birth once a year, total births are 500 per year, and potential individual
longevity is 9 years. Hatched bars: the number of individuals of each age who must die each
year if the population size and age structure are to remain unchanged.
Right: A separate display of the hatched bars, reflecting the age profile of those individuals
who die each year in the hypothetical population.

measure of relative mortality or what is commonly called ‘the rate of mortality’.
Rates of mortality for successive age classes are listed in the ‘qx’ column of
Table 1. In population biology, ‘Ix’, ‘dx’, and ‘qx’ are conventional notations
for tallies of live individuals, dead individuals, and rates of mortality respec-
tively. When these notations are used, age is designated by ‘x’.

As a rule, individual large mammals achieve sexual maturity at an age that
is roughly 10 per cent of maximum potential longevity. In our hypothetical

52 ANNALS OF THE SOUTH AFRICAN MUSEUM

population this would mean that essentially all individuals over 1 year of age
would be sexually mature. The total of such individuals (from the ‘Ix’ column of
Table 1) is 1 194. Assuming that half of these individuals (597) are female and
females can produce only one offspring per year, then 84 per cent of the
females (397) would have to give birth every year to maintain a constant
population.

It is interesting to determine what would happen to the hypothetical
population if the key constraints were removed. Suppose, for example, that we
wanted to force the population into decline. The most direct way would be to
increase mortality in the youngest sexually mature age classes. This is because
there are only 597 females to produce 500 young every year. Even a slight rise
in mortality in the youngest sexually mature age classes would drive the total
number of sexually mature females below 500, not only immediately reducing
_ the size of the population but also reducing the number of new births in the
coming year. An increase in mortality in the youngest sexually mature females
can be particularly disastrous under natural circumstances, since not all sexually
mature females in natural populations will actually be capable of breeding.
Some will certainly be barren, particularly in the oldest age classes. Often,
older individuals also tend to be infirm, so that they suffer higher rates of
mortality, a circumstance which has been assumed for our hypothetical popula-
tion.

If, instead of decline, we wanted to promote population growth, the most
direct way would be to reduce mortality in the first age class, since even a small
percentage decrease would significantly increase the number of individuals that
reach sexual maturity. There would be a cumulative effect, since this would
also increase the number of births in the following year. However, even though
Table 1 shows that current first-year mortality is 50 per cent, which seems quite
high, a reduction in rate may be very difficult to achieve. This is because the
very young in many species are extremely vulnerable to death by predation,
disease, accidents, starvation, etc. First-year mortality rates in the vicinity of 50
per cent are, in fact, common in many naturally occurring populations of large
mammals.

So far we have assumed that each female is capable of producing only one
young per year. It is instructive to see what happens if the number of potential
young is increased to two per year. A constant population size can now be
maintained with only half as many females as before. If, in fact, population size
is to remain unchanged, mortality rates must rise in the youngest age classes to
reduce the number of sexually mature individuals. Table 2 provides ‘lx’, ‘dx’,
and ‘qx’ values for a hypothetical population similar to our original one in
constant overall size, but characterized by 1 000 births per year from 500
females. The relationship between corresponding ‘Ix’ and ‘dx’ values in Table 2
is illustrated graphically in Figure 2.

To summarize, populations of large mammals that are constant in size and
age structure will exhibit age profiles in which successively older age classes

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 53

TABLE 2

The age structure of a hypothetical population of large mammals in which females give birth
twice a year, total births are 1 000 per year, and potential individual longevity is 9 years.

NUMBER OF INDIVIDUALS

Age (x) Number of Number of Rate of
(years) live dead mortality

individuals individuals %

(Ix) (dx) (qx)

0-1 1 000 700 70

1—2 300 150 50

2-3 150 75 50

3-4 1S 30 40

4-5 45 9 20

5-6 36 7 20

6-7 29 6 20

7-8 23 12 50

8-9 11 11 100
9-10 0 0
1000 Lereye)
900 900
800 800
700 700
600 600
500 500
400 400
300 300
200 200
100 100

O15 2,9 4,5 (S\5) 8,5 0,5 ES 49 6,5 8,5

CLASS MIDPOINTS (years)

Fig. 2. Left. Blank bars: the age structure of a hypothetical population of large mammals in
which females give birth twice a year, total births are 1 000 per year, and potential individual
longevity is 9 years. Hatched bars: the number of individuals of each age who must die each
year if the population size and age structure are to remain unchanged.
Right. A separate display of the hatched bars, reflecting the age profiles of those
individuals who die each year in the hypothetical population.

contain progressively fewer individuals (Figs 1 and 2, left). If the females in the
population can have only one young per year, the age profile of those
individuals who die each year will generally be U-shaped (Fig. 1, right),
reflecting the fact that the rates of mortality are generally highest in the very
young and in the old. If, as may be the case in some natural populations, the
rate of mortality does not rise sharply in older age classes, the age profile of
those individuals who die each year may be L-shaped, with no obvious peaks

54 ANNALS OF THE SOUTH AFRICAN MUSEUM

after the one in the youngest age class (see below). Whatever the case,
however, the age profiles characterizing the live and dead segments of the
population in any given year will be formally quite distinct (Fig. 1, left v. Fig. 1,
right.)

If the females in the population produce more than one young per year,
the age profile of those individuals who die each year will be quite similar in
shape to the age profile of those who remain alive, reflecting the fact that rates
of mortality remain high into middle life. The main point is that, unlike
populations in which females produce only one young per year, the age profiles
characterizing the live and dead segments of the population will be difficult, if
not impossible, to distinguish on form alone (Fig. 2, left v. Fig. 2, right).

AGE STRUCTURE AND MORTALITY IN EXTANT
POPULATIONS OF LARGE MAMMALS

Characteristic age structures and mortality patterns have been established
for remarkably few populations of large mammals One major reason is that
many populations are fluctuating rapidly in size or rapidly declining, often as a
result of human activity. There is no way of determining the typical age
structure and mortality pattern of a population whose overall size is changing
rapidly. Another problem is that, even for stable populations, it is difficult to
collect reliable data. Three basic methods exist (Deevey 1947). First, survivor-
ship to successive age classes may be recorded in a cohort of individuals born at
more or less the same time; second, a census may be conducted of all those
individuals alive at any one time; and third, individuals that die within a
restricted time may be aged and tallied. In the conventional terminology of
population biology, the first two methods produce ‘Ix’ data, while the third
produces ‘dx’ data. In theory, all three methods will provide the same age
structure and mortality pattern for a population with a fixed size and age
structure.

In practice, the first method—tallying the numbers of individuals from a
single birth cohort that survive to successive ages—is all but impossible to apply
to free-ranging populations of large mammals. It will also not yield results
quickly, since individual large mammals have relatively long potential lifespans.

The second method—live censusing—is more practical, but it requires
either that an entire population be studied or that there be some way of
obtaining a reasonably large, unbiased sample. It further requires a method for
ageing live individuals reliably, often only by observing them from a distance.
The third method—tallying deaths—is, in fact, usually the most practical, since
it may be based on skeletal material, which is relatively easy to sample and age.

Some representative mortality (‘dx’) profiles obtained by the death-tally
method are presented here in Figure 3. All the profiles have been cast so that
the number of age classes in each is approximately equal, while the second age
class generally contains the youngest sexually mature individuals. Each profile

‘(OL6T PBPpOH) Aled [BuoNeN OARS], UT (s74091g $0499)
SOIODOUTYI YOR|q pur {(QO6] SMET) Ye YJoqezy_ usanO ur (smiqiydwp snwvjododdij{) snue,ododdty ‘(996] SMT) SALeq S]fe4 Uostyoin pue Yyoqezify

udenO ul (vuvoUfv vjuopoxoT) yueydoyo ‘(LL6] IepUIS) yIeg jeUONeN Nosussog sy) ur oyeyjnq odeD ‘(OLE eeuIds) yIeg Yloqezi[q uoonO UI (snu
-wdkudisdyja sngoy) yonqiciem {(DZL6] asvutds) yreg [eUOHEN esoseyy ul (snoidoiyjav snssoyroovyq) BOY-1AeM pue ‘(I7jayo4ng snnbq) vAgez s[]oyoing
‘(sndwmjau sosaokday) eyedun ‘(uaffvo snaaaudg) oyeyyng oded ‘(pp6l SUNIL) Wed [euOneN AopUryoy] yUNoW! Ut (17/p S14) doaays [[eq :s90Ino0s
‘syenplAipul suNnoA AIOA JO suOT]

-pyuasoidassopun oy} .Jo0I109, 0} suonrjndod sat] uo suoNRAsasgo asn 0} jduIA}}e s.jshjeUe [BUISIIO OY} B}BOIPUT yIeg [eUOHeN eloseyy Wo soy-11eM pur
‘eyedunt ‘ojeyyng odeg ay} 10J sopyoid oy} ul seq poyOd ‘AISAOIAI IIOJOQ 9jLIBOIUISIP 0} ATOYI] POU Iv SUTRUAI [RJO[OYS toy] 9snevd9q pojussoidoiiopun
aie (ssejo ase ys1y oy) ur) sjenprlAIpul 3uNOoA AyoA ‘safyoud ayy qe uy ‘aoejd yoo) Ayyeyow usyM potted oy} sulmMp uorntsoduros ase pue dZIS
ul a]quis ATYSnor seo] Je 919M Jey) SUONLINdod wo sMel 10 s[[NYs payooyJoo AyjeoNeUIs}sAs UO paseq oIe safyoid oy} [TY “Yore Ul owes 94} Ajoyeurxoidde
SI sassejo a8e JO JaqUINU OY) 1eY) OS Ised Uddq dARY soTyosd ayy “sfeWWeU o81e] SuIsUeI-s91J Jo suoNR[Ndod ur sajyord Ayyejou dsanejussaiday “¢ ‘314

SIVWAVW J9YV7 ONISNVY-33dY¥4 JO SNOILVINdOd AWOS NI SANSOUd (,*P,) ALITVLYOW

(saDaA ul S}UIOdpIW SSDI9) FOV
be 92 8il Ol 2 co ve 92 Bt A C2 9 IS 6€ le Sl ¢€ €9 IS 6€ le Si ¢€ ls Wah EI) SS eet ae = ih

319)

CM

JUL UHL QUO

STUANOVOCTATOOQONN QQUEUOUL SUUUEL THUUHUT
IMMUN

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN

(Cfb=N) (LLb=N) (€e1=N) (921 =N)
pueboyy OADS| ysaqozi|3 uaenh S}|D4 UOSIYIINW yyeqozljq uaaNnhH piaboyy oS -y
BoujysDM SOJBIDOUIYS YOIDIG snwpysododdiy yuDYda|a suDYydaja Diqez s,|Jayoung eo Lal
8)
: y a) Oo
mM
3 za
|
O
TN

S'z1 s'ol S's S'9 Gy s'z s‘o S‘Ol s‘8 S'9 S‘b Sz S'0 (4 ee See

======—= ===== Sa = 2===2= ==S====— = os
ef eee | See, ge 2 eee oY eS Zon =
== ae <
7a = wo)
Ov ‘S
(66=N) (O€1 =N) >
yyeqgoz!j3 usEeNnhH pisboyy 14ya6uasas piaboyVy Aa|UIyOW IW os os

(22) 4ONGsI9{DM (22) DjDdw! 0jD}3nq adp9 0|Dj3nq adDo daays ||0G

y

56 ANNALS OF THE SOUTH AFRICAN MUSEUM

is based on systematically collected skulls or jaws from individuals that died of
‘natural’ causes such as predation, accidents, and endemic diseases. Age was
established by counting the annual rings on horns (Dall sheep) or by estimation
from dental eruption and wear (remaining species). The base populations differ
in the extent to which they were stable in overall size and age structure during
the period when mortality took place, but all approximated stability sufficiently
for present purposes.

Very young individuals (in the first age class) are underrepresented in all
the profiles, because their skulls are more fragile than those of older individ-
uals. The extent of underrepresentation probably varies from profile to profile
depending upon local preservational conditions and on differences among
species in skull architecture. Dotted bars in Figure 3 indicate the original
analyst’s attempts to ‘correct’ for underrepresentation of the very young in the
mortality profiles of buffalo (at Akagera), impala, waterbuck, and wart-hog.

Keeping in mind the universal, though not necessarily equal, underrepre-
sentation of very young individuals, the profiles of all the species but wart-hog
are broadly similar in form, suggesting that in each case very young individuals
and older adults suffer substantially higher rates of mortality than do younger
(young sexually mature) adults. The pattern is very similar to the ‘dx’ (mortal-
ity) pattern in the hypothetical population of Figure 1 (right), in which females
were assumed to have no more than one young per year.

In all the species but wart-hog, females also have no more than one young
per year, and it was with these real species in mind that the hypothetical
population was devised. The ‘real’ profiles suggest that species in which females
bear one young (or less) per year are, in fact, often characterized by especially
high rates of mortality among the very young and the old. It is possible,
however, that the definition of ‘old’ may vary from population to population.
Figure 3 suggests that in Dall sheep, waterbuck, and hippopotamus a substan-
tial increase in the rate of mortality occurs in adults that are at 60—70 per cent
of potential lifespan, while in the other species it-tends to occur among
somewhat younger adults at about 40-50 per cent of potential lifespan. Many
more observations will be necessary to determine if this difference and other
apparent differences among the profiles are real, perhaps due to differences in
environment or species biology, or whether they are spurious, perhaps due to
chance (‘sampling error’), to differences in the extent to which the various
populations were truly stable, or to differences in the extent to which individ-
uals in different species can be accurately aged from skeletal remains. Since
mortality patterns may differ somewhat between males and females in the same
population (e.g. Spinage 1970 on waterbuck, or Sinclair 1977 on buffalo), it is
further possible that some of the differences among profiles in Figure 3 reflect
differences in the sex composition of the samples.

The wart-hog mortality profile is radically different from all the others in
that it indicates very high mortality rates not only for very young individuals
but also for young (sexually mature) adults. As the original analyst (Spinage

FREQUENCY

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN >i)

1972a) points out, the distinctive pattern of the wart-hog profile almost cer-
tainly reflects the fact that, unlike females in the other species, wart-hog
females regularly produce more than one young per year (the common pattern
is for a litter of 2 to 7). The wart-hog profile is, in fact, very similar in shape to
the ‘dx’ (mortality) profile of the hypothetical population in Figure 2 (right) in
which females were assumed to produce more than one young per year.

Though often less reliable than death-tally profiles such as those in Figure
3, live censuses of large mammal populations support the generalizations that
the death-tally profiles suggest. A representative ‘census’ based on non-
selective shooting and observations of live female Himalayan thar (Hemitragus
jemlahicus) in New Zealand (Caughley 1966) is presented in Figure 4 (left).
The population was probably roughly stable in size and age structure when the
‘census’ was taken. Female thar generally have no more than one young per
year, and the inferred mortality (‘dx’) profile is obviously broadly similar to
that of the species in Figure 3 in which females also have one young (or less)
per year.

= SURVIVORSHIP (100 Ix) gg MORTALITY (100 dy)
90 90
80 HIMALAYAN THAR ¢? 80
a (N=623) a6
60
50
40

30

OSmeaon 45° 865°" 8,5 10/5. 12;5 O19" e2\a) NEOMG SITs 2105
CLASS MIDPOINTS (years)
Fig. 4. Survivorship (‘Ix’) and mortality (‘dx’) profiles in a population of female Himalayan thar
(Hemitragus jemlahicus) in New Zealand, as reported by Caughley (1966). The survivorship
profile is based on random shooting and live counts, while the mortality profile was inferred

from the survivorship count. The mortality profile exhibits a tendency for the rate of mortality
to rise fairly sharply in individuals at 40-50 per cent of potential lifespan.

In sum, the available data on extant large ungulate populations suggest
that ones in which females have one young (or less) per year will tend to exhibit
a mortality pattern in which the rate of mortality is much higher in very young
individuals and in older adults than in younger ones. Depending upon the
population, ‘older’ adults may be ones at an age between 40-50 per cent and
60-70 per cent of potential lifespan. The live (‘Ix’) age structure of such a
population will be similar to that of the hypothetical population of Figure 1
(left), with a shape similar to a down staircase in which a very steep first step is
followed by a series of much less steep but steadily steepening ones. The

eS

58 ANNALS OF THE SOUTH AFRICAN MUSEUM

mortality (‘dx’) profile of the population will tend to be U-shaped as in Figure 1
(right).

The limited data presented here on large mammals that have more than
one young per year may be supplemented by data on fast-breeding small
mammals (references in Spinage 1972a) to support the conclusion drawn here
that high mortality rates will persist into young adult classes in such popula-
tions. The live age structure of the population will be similar to that of a down
staircase in which the initial three or four very steep steps are followed by a
series of barely perceptible ones. The corresponding mortality profile will recall
a very similar staircase, though the very last, barely perceptible, steps will tend
to ascend rather than descend (Fig. 2, right). Since the difference in shape
between the corresponding live (‘Ix’ or survivorship) and death (‘dx’ or mortal-
ity) profiles is so subtle, a very large sample will be necessary to show which
profile is represented by an age structure drawn from material whose context
did not allow an a priori determination. The significance of this point will
become clear below.

AGE STRUCTURE AND MORTALITY IN SOME FOSSIL
POPULATIONS OF LARGE MAMMALS

There is a sense in which the age profile of a fossil population is always a
mortality or ‘dx’ profile, since the individuals involved are all dead. However, it
will only resemble the ‘dx’ profile of an extant population if the fossil individ-
uals died as a result of accidents, predation, endemic disease, and other routine
attritional factors that ordinarily have their greatest impact on the very young
and the old. If, instead, the fossil individuals died as the result of a great flood,
volcanic eruption, epidemic disease, or other catastrophic event that affected
individuals of all ages to the same extent, the mortality profile of the fossil
population will directly reflect its original live age structure. In other words, it
will, in fact, be a survivorship or ‘Ix’ profile. In practice then, the age profile of
a fossil population may reflect either mortality (‘dx’) or survivorship (‘Ix’) as
they are commonly defined.

The observations on extant populations presented in the last section
suggest that shape alone may be sufficient to determine whether a large
mammal mortality profile reflects catastrophic (‘Ix’) or attritional (‘dx’) mortal-
ity, at least in species where females commonly have one young or less per
year. In such species a ‘catastrophic’ (= survivorship = ‘Ix’) profile will tend to
be shaped like a down staircase, with the first and near-last steps being the
steepest. An ‘attritional’ (= true mortality = ‘dx’) profile will tend to be U-
shaped, with a large peak or mode in the youngest age class, followed by a
marked dip, and then a second, smaller peak in individuals at 40-50 per cent to
60-70 per cent of potential longevity.

In many instances, it is, of course, possible to infer mode of death
(catastrophic or attritional) independently of profile shape. For example, cata-

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 59

strophic death by drowning is a likely explanation for a fossil example derived
from ancient fluvial deposits packed with bones belonging mainly to only one or
two species. Catastrophic death is probably also implied in most cases where an
age distribution is clearly discontinuous, that is, composed of discrete age
classes, each separated from the next by an age gap into which no individuals
fall. This could happen only if birth and death were both seasonally restricted
events, and a priori, seasonally restricted death (perhaps by flood or brush fire)
is more likely to be catastrophic than attritional. This is particularly true if the
dead animals accumulated in a deposit over a period of several years, so that
their deaths had to occur in the same season each year.

Since it is often possible to infer mode of death (catastrophic or attritional)
in a fossil sample by criteria other than age profile shape, age profiles in fossil
samples may be used to help establish the general characteristics of survivorship
and mortality in large mammals. In this regard they may even prove more
suitable than many samples from extant populations, since they are often larger
(or can be enlarged quickly with relatively little effort) and they frequently
derive from a substantial period of time during which short-term fluctuations in
population size and age structure will tend to have cancelled each other out.
This means that a fossil sample is more likely to produce an age profile that is
truly typical of the population.

Unquestionably the most famous studies of age structure and mortality in
fossil populations of large mammals are those by Kurtén (1953) and Voorhies
(1969) on samples from late Miocene fluvial deposits in north China and
Nebraska respectively. Voorhies carefully excavated his sample himself, while
Kurtén analysed material collected by others under conditions that were
probably much less well controlled. There is no reason to suppose, however,
that collecting seriously biased the age profiles of any of the species that Kurtén
studied.

Voorhies’s sample was heavily dominated by the extinct pronghorn ante-
lope Merycodus furcatus, with the three-toed horse Protohippus cf. perditus a
distant second, and other species still less common. Kurtén’s samples came
from several localities, at each of which one species tended to dominate
heavily, though not necessarily the same species at each locality. In their
analyses of dental eruption and wear in the principal species in their samples,
both Kurtén and Voorhies found that individuals fell into discrete age modes or
clusters, each cluster probably representing a cohort of individuals born at
more or less the same time and killed more or less simultaneously. Together
with the sedimentary context, this indicates that mortality in the various species
studied by Kurtén and Voorhies was ‘catastrophic’ rather than ‘attritional’.

The catastrophic (‘Ix’) profiles of M. furcatus and P. cf. perditus from
Nebraska and of the ovibovine bovids Plesiaddax depereti and Urmiatherium
intermedium and the gazelle Gazella dorcadoides from north China are dis-
played on the left-hand side of Figure 5 here. In M. furcatus the youngest
individuals represented were already old enough at death for their jaws to be

60 ANNALS OF THE SOUTH AFRICAN MUSEUM

about as durable as those of adults. As a consequence, there is no reason to
suppose serious. underrepresentation of very young M. furcatus in the age
profile. In the other four species, however, the jaws of individuals in the
youngest (first) age class, and in the ovibovines also in the second age class,
were relatively fragile, and such individuals are probably seriously underrepre-
sented in the age profiles. The dotted bars in Figure 5 represent the present
writer's attempt to correct for this underrepresentation in the gazelle and
ovibovine profiles by assuming roughly 50 per cent first-year mortality in all
three species and roughly 10 per cent second-year mortality in the ovibovines.
In all five species involved, females probably had no more than one young a
year. Keeping in mind that very young individuals are seriously underrepre-
sented in four of the profiles, they are all very similar in form to those of extant
species with the same reproductive potential.

The ‘dx’ profiles that are inferable from the ‘Ix’ ones are shown on the
right-hand side of Figure 5. They are also broadly similar to the ‘dx’ profiles of
the extant species, though in general they do not exhibit a clear second peak
among older individuals. The absence of a clear second peak may reflect chance
(‘sampling error’) or it may indicate that in some populations of large mammals
the rate of mortality does not rise sharply among older adults, leading to an
L-shaped (v. U-shaped) shaped mortality profile. The possibility that ‘dx’
profiles may be L-shaped is not widely recognized but must clearly be borne in
mind when interpreting mortality profiles such as those from Langebaanweg
and Elandsfontein considered below.

AGE PROFILES PRODUCED BY PREDATION ON SOME
POPULATIONS OF LARGE MAMMALS

Figure 6 presents age profiles of Burchell’s zebra and Cape buffalo killed
by lions on the Serengeti Plain (Schaller 1972), of chamois (Rupicapra rupicap-
ra) shot by recent hunters in the Alps (Kurtén 1953 using data from Bourliére
1951), and of pronghorn antelope (Antilocapra americana) killed by the late
prehistoric occupants of the Eden—Farson site, Wyoming (Nimmo 1971; Frison
1978a). In each case, individual age was established from dental eruption and
wear. Very young zebra, buffalo, and chamois are all seriously underrepre-
sented in Figure 6. Lions consume very young zebra and buffalo too quickly
and completely for observers to estimate accurately the number of very young
killed, while the hunters who shot the chamois were not interested in very
young animals because they make poor trophies. If it had been possible to
record all kills of zebra and buffalo, very young individuals would certainly be
the best represented age class in the kill profiles. Similarly, very young animals
would have been the most common individuals in a completely random shot
sample of chamois. Only in the pronghorn kill sample, derived by careful
excavation of well-preserved bone, are very young animals probably present in
more or less their proper proportion.

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 61

survivorship ( ‘I,’ ) mortality (‘dy’)

50 Merycogus furcatus 50 Merycodus furcatus

ee, 26 48 6; OS 28° 48° “68

50 FProtohippus near perditus

39 6©64500—CiGDS(tsCSS

Urmiatherium intermedium

ep)

=

<6

= 60

= 50 Plesiaddax devereti
5 40 (N=76)
= 30

uu 20-—-=.--
a

— ———

Fzs

Lid

©)

od

LiJ

oO

10,5

Gaze//a dorcadoides

275 475 675 O75 275 475 6,75

CLASS MIDPOINTS (years)

Fig. 5. Catastrophic (= survivorship = ‘Ix’) profiles (left) and attritional (death = “dx’) profiles
(right) in Merycodus furcatus and Protohippus near perditus from late Miocene deposits at
Verdigre Quarry, Nebraska (Voorhies 1969), and in Plesiaddax depereti, Urmiatherium interme-
dium, and Gazella dorcadoides from various late Miocene localities in north China (Kurtén
1953). The dotted bars represent the present author’s attempts to ‘correct’ for the underrepre-
sentation of young individuals in the profiles of the north Chinese species. In general. the
‘death’ (= ‘dx’) profiles in this Figure contrast with those in Figures 3 and 4 in lacking a clear
second peak among older adults. This may be due to chance, or it may indicate that such a
peak is not a necessary feature of attritional mortality in large mammals.

62 ANNALS OF THE SOUTH AFRICAN MUSEUM

100 100
90 90
80 80
70 70
60 60

Equus burchelli

Syncerus caffer
30 (N= 174) 30 )

: 100
90 90
80 80
70 ( ) 70
60 ( 60

Rupicapra rupicapra 50 Artilocapra americana
(N=422) (N=84)

PERCENT OF INDIVIDUALS
fe)
(e)

[3a So Gon SerSet Talo) 03° 23 084s nes

CLASS MIDPOINTS (years)

Fig. 6. Age profiles resulting from predation by lions on Burchell’s zebra and Cape
buffalo (Serengeti Plain) (Schaller 1972); from unrestrained, non-selective shooting of
chamois in the Alps (Kurtén 1953 using data from Bourliére 1951); and from prehistoric
American Indian hunting of pronghorn antelope (Eden—Farson site, Wyoming) (Nimmo
1971). The numbers of very young zebra and buffalo killed by lions cannot be accurately
estimated, but very young individuals would certainly dominate the kill profiles if
accurate counts could be made. The hunters who shot the chamois essentially ignored
very young individuals because they do not make good trophies. Only in the pronghorn
profile are very young animals probably represented in more or less their proper (true)
proportion.

The zebra data have been reorganized from the original presentation, with individ-
uals in what were sometimes broader age classes distributed evenly among the two-year

age classes used here. All other data are as presented in the original sources.

In all four species, females generally bear no more than one young per
year, which means that the profiles in Figure 6 are best compared with those in
Figure 1, representing catastrophic (= ‘Ix’ =survivorship) and _attritional
(= ‘dx’ = mortality) profiles of a hypothetical population with the same basic
reproductive potential. Taking into account the severe underrepresentation of
very young individuals in the zebra, buffalo, and chamois profiles, and assum-
ing that attritional profiles need not display a sharp rise in the rate of mortality

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 63

among older individuals, the zebra and buffalo profiles are essentially attri-
tional, while the chamois and pronghorn ones are clearly catastrophic.

The reason that the buffalo profile has an attritional shape is because the
physical condition, large size, and social organization of young adult buffalo
make them largely immune to lion predation. When lions hunt buffalo they are
forced to concentrate on the very young and the old. It is probable that similar
features of size, condition, and social organization also make young adult zebra
less vulnerable to lion predation, though the difference in shape between the
zebra and buffalo profiles in Figure 6 suggests the possibility that lions find it
easier to kill young adult zebra than young adult buffalo. Whatever the case,
lions apparently kill zebra and buffalo of various ages in broadly the same
proportions as they would die anyway. Such a pattern of predation is perhaps
to be expected in a situation where predator and prey populations have long
coexisted, since the prey populations could sustain it indefinitely.

In contrast to the attritional shapes of the zebra and buffalo profiles, the
catastrophic shape of the chamois profile reflects the activity of a predator (men
with rifles) that was not forced to concentrate on any particular age class. In
further contrast to predation on the zebra and buffalo, predation on the
chamois could not have continued indefinitely. This is because the elimination
of so many reproductively active young adults would depress the birthrate,
leading to a decline in numbers and perhaps even extinction. In fact, historic
observations indicate that hunting did seriously reduce the chamois population
on which the profile in Figure 6 is based.

The method by which the pronghorn were hunted must be inferred, since it
could not be directly observed. Given aboriginal American hunting technology,
the method most likely to produce a catastrophic profile would be the commu-
nal drive, which was, in fact, observed ethnohistorically among Indians hunting
pronghorn (Frison 1978a). As in the case of the chamois, a method netting so
many young (sexually mature) adults would lead to a significant decline in
overall pronghorn population size, if it were relentlessly employed. Histori-
cally, drives did apparently eliminate local pronghorn populations, which then
took several years to recuperate through immigration from neighbouring
regions. The point to be made is that hunting methods resulting in catastrophic
kills must be restrained, voluntarily or otherwise, if they are not to reduce prey
and thus ultimately predator numbers.

In sum, the examples presented here show that predation may produce
either attritional (‘dx’) or catastrophic (‘lx’) mortality profiles in prey, depend-
ing upon the characteristics of prey and predator. An attritional profile,
indicating that the predator was largely restricted to very young and old
individuals, is perhaps to be expected in any situation where predator and prey
have long coexisted. Catastrophic prey profiles, indicating that the predator
possessed a capture method to which individuals of all ages were about equally
vulnerable, will generally reflect unstable predator-prey relationships, probably
mainly ones in which recent people were the predators. If methods leading to

64 ANNALS OF THE SOUTH AFRICAN MUSEUM

catastrophic kills are not restrained, they can be counter-productive, since they
will eventually cause prey populations to decline or even disappear.

ESTIMATING THE AGE OF A FOSSIL UNGULATE AT DEATH

In general, the most useful skeletal elements for estimating the age of
individual ungulates at death are the teeth. In many species the age of a
recently dead (or still living) individual may be estimated by counting the
number of ‘annuli’ in the cementum covering dental roots (see Morris 1972 or
Spinage 1973 for overviews). The cementum annulus method is generally not
useful for estimating the age at death of fossil individuals, however, because
fossil cementum often does not retain the annulus structure, at least in a way
that can be detected with standard preparation techniques (Spiess 1979).
_ Additionally, specimen preparation is destructive, rendering teeth all but
useless for further study.

With fossil teeth, the only truly practical method of estimating individual
age is to evaluate the state of eruption and wear. This can be done subjectively
by comparing fossil dentitions to ones from known-age individuals, if a species is
still extant, or objectively by measuring a dental dimension that clearly changes
with age. In some fossil samples, such as those studied by Kurtén (1953) and
Voorhies (1969), the seasonally restricted birth and death of the species involved
allows subjective arrangement of the dentitions into a series of discrete eruption
and wear classes. These reflect discrete age clusters, each separated from the last
by approximately one year. Assuming that the youngest age class is made up of
individuals in the first year of life, it is then possible to assign approximate
(modal) ages to each of the successive eruption and wear classes.

In general the subjective method of age determination has the disadvan-
tage that different investigators may estimate different ages from the same
fossil dentition and also that it usually requires complete demi-mandibles or
maxillae. In many, perhaps most, fossil samples, many individuals are repre-
sented only by isolated teeth. The objective method, based on measurement,
has the advantage that it produces easily replicable results. Perhaps even more
important, it can also be readily applied to isolated teeth, as long as their
former position in the mouth can be determined. In sum, for fossil material, the
measurement method of estimating individual age from teeth is almost certainly
the best. At least with regard to high-crowned ungulates, the most obvious
dental dimension to measure is crown height.

The mathematical relationship between advancing age and decreasing
crown height has not been thoroughly established for any high-crowned ungu-
late species. The principal reason is the rarity of large samples of known-age
dentitions, particularly ones from a wide variety of age classes. In the absence
of a sound body of empirical observations the estimation of age from crown
height requires some ‘theoretical’ assumptions on the nature and rate of crown
attrition. In work to this point, the author has made the following assumptions:

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 65

1. That reduction in crown height is roughly constant through the life of a
tooth, that is, that the relationship between decreasing crown height and
advancing age is approximately linear.

2. That for a deciduous tooth, the chronological age of complete crown
reduction—when the crown is all but worn away—is the age when the tooth is
replaced by its permanent counterpart. For a permanent tooth, the chronologi-
cal age of complete crown reduction is the age past which no individuals survive
in the wild, sometimes known as ‘potential ecological longevity’. For most
species, the dental eruption and replacement schedule and potential ecological
longevity may be obtained directly from wildlife biology publications or
indirectly by inference from publications on closely related species of similar
SIZe.

3. That the amount of crown height lost per unit time on a deciduous
tooth equals the initial unworn crown height divided by the time interval
between age of eruption (usually birth) and age of replacement by a permanent
tooth. The amount of crown height lost per unit time on a permanent tooth
equals the initial unworn crown height divided by the time interval between age
of eruption and age of potential ecological longevity. Initial unworn crown
height may usually be estimated from unworn or lightly worn teeth occurring in
any sample that is large enough to warrant mortality profile construction. In the
author’s experience, individual variability in unworn crown heights within a
local population of a species tends to be very limited, so that the mean of a
fossil sample is an adequate estimate, even if it is based on very few specimens.

Klein et al. (1981) showed that these assumptions lead to the following
age-prediction formulae:

for a deciduous tooth:
AGE = AGEs—(AGEs/CHo) (CROWN HEIGHT)

for a permanent tooth:
AGE = AGEpel—((AGEpel—AGEe)/CHo) (CROWN HEIGHT)

where AGEs is the age at which a deciduous tooth is shed, AGEe is the
age at which a permanent tooth erupts, AGEpel is the age past which no
individuals appear to survive in the wild (‘potential ecological longevity’),
and CHo is initial (unworn) crown height.

Among the various assumptions perhaps the most questionable one is that
the rate of crown attrition is constant throughout the life of a tooth. Various
studies have suggested that wear is more rapid early in the life of a tooth than
later on (see, for example, Grimsdell 1973 on extant Cape buffalo, or Kurtén
1953 and Voorhies 1969 on the fossil ungulate species whose age profiles are
presented in Figure 5). Spinage (1971, 1972b, 1973) has pointed out that known
changes in occlusal topography—roughest at the very beginning, smoothest
near the end—imply that attrition will slow with age. Only in mid-life, when the
topography tends to remain relatively constant, will attrition occur at a more or
less constant rate. In order to accommodate a variable rate of wear of the kind

66 ANNALS OF THE SOUTH AFRICAN MUSEUM

the changes in occlusal topography suggest, Spinage has proposed age estima-
tion formulae that have the following form, when they are expressed using the
terms above:

for a deciduous tooth:

AGE = AGEs ((CROWN HEIGHT—CHo)/CHo)?

for a permanent tooth:

AGE = (AGEpel—AGEe) ((CROWN HEIGHT—CHo)/CHo)* + AGEe

In a study involving a relatively large sample (N=170) of known-age
wapiti (Cervus elaphus canadensis), Klein et al. (in press) found that Spinage’s
formulae do, in fact, provide more accurate age estimates from crown height
than do the formulae based on a constant wear rate. Perhaps even more
important, Spinage’s formulae tend to underestimate true age about as often as
‘they overestimate it, while the linear formulae mostly overestimate it. In the
language of statistics Spinage’s formulae are much less biased estimators. The
important point is that they are more likely to produce a truly accurate age
profile from a sample of crown heights, since overestimated ages would tend to
be balanced out by underestimated ones.

The principal disadvantage of Spinage’s formulae is that they are more
cumbersome to calculate than those based on a constant wear rate. However,
this is not a serious problem if a programable calculator or computer is
available. An additional problem is that changes in the values assigned to the
constants AGEs, AGEe, AGEpel, and CHo in Spinage’s formulae have a
greater effect on estimated ages than do comparable changes in the linear
formulae. This means that the values of the constants must be selected as
carefully as possible when Spinage’s formulae are applied.

Whichever formulae are used, it is advisable to group the ages estimated
from crown heights into classes that are at least as broad as the average error in
age estimation. In virtually all instances, a class interval based on 10 per cent of
potential ecological longevity will probably suffice. An important characteristic
of such an interval is that age profiles based on it will be immediately
comparable even among species with very different potential longevities.

In the case of species such as wapiti in which potential lifespan (AGEpel)
is long relative to initial (unworn) crown height (CHo), both the linear and
Spinage’s formulae will produce quite similar age profiles, when the class
interval is 10 per cent of potential lifespan, especially if the permanent tooth
used to estimate age is one whose crown height is still measurably greater than
‘0’ when potential longevity is approached. In most species, the tooth that
meets this requirement best is the third molar (M3). In very hypsodont species
such as equids in which potential lifespan is much shorter relative to initial
(unworn) crown height, the linear formulae and Spinage’s formulae may
produce quite different age profiles, regardless of what tooth is used (Fig. 7). In
the case of such species it is imperative that Spinage’s formulae be used or the
age profiles that result may be quite misleading.

67

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN

“QUO JJeINOOe IJIOUL

oy} Ajure}199 Jsouye st Uonenbs onespenb sy) Aq popraoid apyoid ay} ‘1x9} dy) UI possnosip suoseal 104 ‘Yysty st (Jady|Oy) AwAosuoy [enusjod 0} (OF{D)
}YsIoOY UMOIO (UIOMUN) [LIIUT JO ONBI OY} YOIYM Ul sisuadvd snnby se yons saioods ve 10} sayyoid jussazJIp AIDA sonpoid skemye JSOW]e [JIM Je[NUIIO]
JUdIOIJIP UL, “(T86T) 72 Ja ulopy Aq poysossns eynuoOJ IesUT] e pur (gZL6[) a8eUIdg Aq pojsadsns e[NUIIOJ ONvIpenb dy BuIsn s}YysI9Y UMOIO 9)
WOIJ poALop soyyoid (a8e) Ayypewoyy :oAogy ‘UloJUOJspuL]_ Wo sisuadvd snnbq jo syysioy UMOID puke syIpedIg UMOIO EJ PUL ZqP :MOTIG */ ‘SI4

LHSISH NMOYD cW

%O00! 06 08 OZ O09 OS OF O

— 02

OH
Fp (939V-leda9y) — |ed39y = 39V

wo

(WW) SINIOdGIN SSV19
Gh 85 CS EH =O ol

OHO

8349V + lea = <| (9399 - 19d399V)=39V
2

SAWdOY"d ALIIVLYOW

08

LHSISH NMOXD oW
NYOMNN G3LVWILSS

STVNGIAIGNI 40 YSEWNN

zie L922 Li2

—--

(8b=N) 2S 2FHE'9Z

HLOVAYd

NMONO ¢W

/Ecdcy ~ JENS}

Soe oe

Ol S| O02 G2 Of

LHSISH NMOUND adP
NYOMNN GALVWILSS

~—|—-

(Z=N) II'€FOS'02

LH9I3H
NMOUD 2dP

HLavaua
NMOU2 2dP

sisuadDa snnb7z

NISLNOASONV 13

68 ANNALS OF THE SOUTH AFRICAN MUSEUM

All the age profiles presented below were calculated using Spinage’s
formulae. For the sake of consistency, some age profiles that were presented in
a previous paper where the linear formulae were used (Klein 1981a), have been
recalculated, even though their essential form remains the same, since the
species involved were low-crowned relative to potential lifespan. All calcula-
tions were performed on an IBM Personal Computer.

THE LANGEBAANWEG FOSSIL SITE

Langebaanweg (32°58’S 18°9’E) is located approximately 110 km north-
north-west of Cape Town (Fig. 8). Fossils were first reported from Langebaan-
weg in 1958 (Singer & Hooijer 1958; Singer 1961). In 1969 Q. B. Hendey of the
South African Museum initiated a research programme at the site, which has
led to a vast accumulation of vertebrate fossils belonging to more than 150

¥Elands Bay Cave
Sea Harvestoo_ ANGE BAANWEG

.ELANDSFONTEIN
| Swartklip eee
Die Tp ts | Klasies River Mouth
B Nelson Bay Cave
-_——;—__|Byneskranskop ae

[5° Le IS Z\q (ys Zor + 2% » ZORie ES 33°

Map showing the approximate locations of sites mentioned in the text.

Fig. 8. The approximate locations of the sites mentioned in the text.

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 69

species, primarily mammals and birds. Publications on the site include numer-
ous taxonomic and phylogenetic studies by Hendey and others, and overviews
focusing on geologic context and palaeoecology (Hendey 1981la, 19816, with
references).

The overwhelming majority of Langebaanweg fossils, including all those
considered here, come from the Quartzose Sand Member (QSM) (older) and
Pelletal Phosphorite Member (PPM) (younger) of the Varswater Formation, as
exposed in Langebaanweg ‘E’ Quarry. There are no materials suitable for
chronometric dating, but Hendey’s analysis of the Langebaanweg stratigraphic
succession in relation to global sea-level and climatic events, together with the
vertebrate taxa represented, indicate that the QSM and PPM both accumulated
approximately 5 million years ago, or during the early Pliocene as this term is
presently defined.

The QSM consists primarily of fine-grained, white quartz sands laid down
on the estuarine floodplain of a river that probably entered the sea just to the
south-west of ‘E’ Quarry. The PPM consists of relatively coarse well-sorted
sands laid down primarily in the channel of the same river as its bed moved
progressively northward. Within the PPM there are two distinct, especially
fossiliferous channel fills, known as bed 3aS (further south) and bed 3aN
(further north) respectively.

Essentially the same mammalian taxa occur in both the QSM and the
PPM, but their mode of occurrence and their relative abundance differ dramati-
cally between the two units. In the QSM partial, semi-articulated skeletons are
common and abraded bones are rare. In the PPM partial, semi-articulated
skeletons are rare and abraded bones are common. In the QSM several species
are more or less equally represented, with no single species dominating over-
whelmingly. In the PPM alcelaphine antelopes (Damalacra acalla and D.
neanica) and giraffids (Giraffa sp. and especially Sivatherium hendeyi) are
superabundant compared to all other species, the alcelaphines dominating
particularly in bed 3aS and the giraffids in bed 3aN.

In an earlier paper Klein (1981a) showed that various species in the QSM
are characterized by attritional mortality profiles, while the superabundant
giraffids of bed 3aN are characterized by catastrophic profiles. Together with
the sedimentary context and mode of fossil occurrence, the mortality profiles
clearly suggest that individuals represented in the QSM died mainly of attri-
tional factors such as predation, accidents and endemic disease. Subsequently,
their skeletons were disarticulated and partially removed or destroyed by
scavengers and other biological agents, but the bones probably lie very near the
points of death. In contrast, the superabundant giraffids of bed 3aN probably
died mainly by drowning during high floods, perhaps some distance from where
the bones occur today, judging by the lack of articulated skeletons and the
frequency of abrasion. In this paper the mortality profile analysis is extended to
other QSM and PPM species, especially the alcelaphines of PPM beds 3aS and
3aN.

70 ANNALS OF THE SOUTH AFRICAN MUSEUM

THE ELANDSFONTEIN FOSSIL SITE

Elandsfontein (= ‘Saldanha’ = ‘Hopefield’) (33°05’S 18°15'E) is located
approximately 100 km north-north-west of Cape Town and roughly 10 km from
Langebaanweg (Fig. 8). Covering an area of approximately 1,5 x 3 km, the site
consists of a series of large barchan dunes separated by ‘bays’ in which deflation
has exposed numerous mammalian fossils and occasional artefacts. Singer
(1957) initiated scientific research, consisting mainly of surface collecting, at the
site in 1951. Limited excavations were conducted in the early 1960s, most
notably at the ‘Cutting 10 Acheulean Site’ (Singer & Wymer 1968; Klein
1978a). Except for occasional one-day status checks by staff of the South
African Museum or the University of Stellenbosch, the site has been largely
neglected since 1966. In 1981 G. and D. M. Avery of the South African
Museum reinitiated scientific research at the site.

Most of the fossils found at Elandsfontein occur in or on a nodular
calcareous duricrust that Butzer (1973) interprets as a subsoil manifestation of a
former land-surface. It is possible that more than one such duricrust, and thus
more than one former land-surface, are represented. Some fossils are also
associated with a stratigraphically higher ferruginous duricrust, related to a
somewhat more recent palaeo-surface. Finally, a small proportion of the fossils
(including all those from the ‘Bone Circle’ of Inskeep & Hendey 1966) have
been found in loose sands above the ferruginous duricrust. Butzer believes the
Elandsfontein geologic sequence reflects climatic oscillations between wetter
and drier during a large portion of the Middle and Upper Pleistocene.

The Elandsfontein fauna comprises one or two species of tortoise (Chelo-
nia), a small number of bird species (most prominently ostrich, Struthio
camelus), and more than fifty species of mammals (Hendey 1974), including
‘Saldanha Man’ represented by a skull cap and a mandible fragment found on
the surface in 1953 (Drennan 1953, 1955; Singer 1954, 1958; Rightmire 1976).
Unfortunately, most of the bones that are presently available for study were
collected from the surface without any record of their position within the site or
of their fossil associations, so it is impossible now to relate them to each other
or to the geology. Still, on taxonomic grounds, it is clear that there are at least
two faunal stratigraphic units represented, a Middle Pleistocene one and an
Upper Pleistocene one (Hendey 1974; Klein 1978a). Mid-Pleistocene fossils,
probably including the ‘Saldanha Skull’, predominate heavily.

The Middle Pleistocene fossils were probably all associated with the
calcareous duricrusts, while the Upper Pleistocene ones probably came from
the ferruginous crust or from the overlying loose sands. In keeping with the
possibility of multiple calcareous crusts, the taxonomic composition of the
Middle Pleistocene fauna suggests it may be a composite of faunas spanning a
large portion of the Middle Pleistocene, as this is now commonly defined (from
the beginning of the Brunhes Normal Palaeomagnetic Epoch, approximately
700 000 years ago, to the beginning of the Last Interglacial, approximately
130 000 years ago).

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN TA

Although the contexts of most of the Elandsfontein bones presently
available for study were not recorded, it is probable that they occurred in
basically the same contexts as the numerous bones that have been exposed by
deflation at the site since 1966, when compilation of the collection essentially
ceased. At the site today clusters of bones clearly gnawed by porcupines are
common (as are porcupine-gnawed specimens in the collected sample). Perhaps
equally frequent are semi-articulated portions of skeletons, particularly crania,
vertebrae, ribs, and distal extremities of a giant buffalo (Pelorovis sp.), which is
perhaps the most abundant mammalian species at the site. In both the already
collected and presently exposed samples, stone artefacts are relatively rare, as
are bones unquestionably cut or bashed by artefacts. Hyena coprolites, and
bones exhibiting damage from carnivore teeth are probably more common. The
overall impression is that the site is a complex aggregation of occurrences—por-
cupine burrows, ‘natural’ deaths, hyena burrows, and occasional hominid
camps or kills—all now resting on the same deflation surfaces. Animals were
probably drawn to the locality by the availability of perennial surface water
throughout much, if not all, of the Middle and Upper Pleistocene.

The ungulate mortality profiles presented in this paper represent the first
attempt to determine whether the available Elandsfontein collections can be
made to provide palaeoecological information, in spite of the less-than-ideal
way in which they were compiled and the obvious complexity of the site.

-THE LANGEBAANWEG AND ELANDSFONTEIN UNGULATE
SAMPEES AND AGE PROFILES

Table 3 lists the Langebaanweg and Elandsfontein ungulate taxa whose
mortality profiles are considered here. In order to be included, a taxon had to
be represented by a minimum of ten individuals that could be aged from teeth.
Table 3 also presents the dental eruption and wear parameters (AGEe, AGEs,
and AGEpel) that were used to calculate individual ages from crown heights.
The initial (unworn) crown heights (CHo) used are presented in Table 4.

A complication in the Langebaanweg samples is that teeth cannot always
be identified to species. Thus, while it is clear that two species of Damalacra
are represented by horn-cores (Gentry 1980), the species cannot be readily
separated even on complete dentitions. Similarly, although most of the teeth
assigned to Giraffa sp. probably come from a single species, some probably
derive from a dentally very similar species of palaeotragine, which is repre-
sented at Langebaanweg by occasional limb bones and ossicones (Hendey
1981a). Finally, although most of the Nyanzachoerus teeth probably come from
a single species (N. cf. pattersoni), some (in the PPM) probably come from a
dentally very similar second species (N. cf. jaegeri) (Hendey 1981a).

The choice was then either to exclude the Damalacra, Giraffa and Nyanza-
choerus samples from consideration or to analyse them as if only one species
were represented in each case. The second alternative was chosen, since it

VL ANNALS OF THE SOUTH AFRICAN MUSEUM

TABLE 3

The ages of dental eruption (AGEe), shedding (AGEs), and potential ecological longevity
(AGEpel) used to calculate age from crown height in the Langebaanweg and Elandsfontein
ungulate samples considered in this paper. For extinct species, ages were inferred from
published ages for closely related extant species. Where extinct species differed in size from
their extant counterparts, the ages for the extinct species were adjusted by a factor reflecting

the size difference. All quoted ages are in years.

Sources on pertinent ages in the extant species are: black rhinoceros (Diceros
bicornis)—Goddard (1970); Burchell’s zebra (Equus burchelli)—Klingel (1965), Klingel &
Klingel (1966), Smuts (1974); mountain zebra (Equus zebra)—Joubert (1972); giraffe (Giraffa
camelopardalis )}—Hall-Martin (1976); bush-pig (Potamochoerus porcus)—Sowls & Phelps
(1968); Cape buffalo (Syncerus caffer)—Grimsdell (1973), Sinclair (1977); greater kudu (Trage-
laphus strepsiceros)—Simpson (1966); Lichtenstein’s hartebeest (Alcelaphus lichtensteini)—Mit-
chell (1965); tsessebe (Damaliscus lunatus)—Huntley (1973); black wildebeest (Connochaetes
gnou)—Von Richter (1971, 1974); eland (Taurotragus oryx)—Kerr & Roth (1970); and Cape
erysbok (Raphicerus melanotis)—Manson (1974), and personal observation. Mentis (1972) was

consulted for estimates of potential ecological longevity in all relevant extant species.

LANGEBAANWEG rv fi
dP4 P4
AGEe AGEs AGEe AGEpel Source

Ceratotherium praecox 0 6 6 35. Inference from dentally very
similar black rhinoceros.

dP2 P2
AGEe AGEs AGEe AGEpel

Hipparion cf. baardi 0 3 3 20 Assumption that H. cf. baardi
parameters would be roughly
10% smaller than those for Bur-
chell’s zebra and mountain
zebra.

dP4 M3
AGEe AGEs AGEe AGEpel

Giraffa sp. 0 4.5 35 28 Inference from dentally very
similar modern giraffe.

Sivatherium hendeyi 0 6 4.67 37,24 Assumption that S. hendeyi
parameters would be roughly
33% larger than those of mod-
ern giraffe.

Mesembriportax acrae 0 Jy) Z 18 Inference from extant bovids of
similar size, such as greater
kudu, Lichtenstein’s hartebeest,
and black wildebeest.

Simatherium demissum 0 3,0 DS 20 Assumption that S$. demissum
parameters would be roughly
25 % smaller than those of Cape
buffalo.

Damalacra spp. 0 M3) Z 18 Inference from extant bovids of
similar size, such as greater
kudu, Lichtenstein’s hartebeest,
tsessebe, and black wildebeest.

Nyanzachoerus sp(p). 0 7p) 2 18 Assumption that Nyanzachoerus

parameters would be roughly
20% greater than those for
bush-pig.

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 13

ELANDSFONTEIN z is
dP2 M3
AGEe AGEs AGEe AGEpel Source
Equus capensis 0 3,6 3,6 26,4 Assumption that E. capensis
parameters would be roughly
20% larger than those for Bur-
chell’s zebra and mountain
zebra.
dP4 M3
AGEe AGEs AGEe AGEpel
Pelorovis sp. 0 4,35 3,28 26,4 Assumption that Pelorovis para-
meters would be roughly 10%
greater than those for Cape buf-
falo.
Taurotragus oryx 0 310 7635) 20 Data on modern eland.
Hippotragus gigas 0 4 3 24 Assumption that H. gigas par-
ameters would be similar to
those of Cape buffalo.
dP4 M1
AGEe AGEs AGEe AGEpel
Raphicerus sp. 0 725) 0,25 6 Inference from closely related

Cape grysbok.

seemed unlikely that mixed samples would produce interpretable age profiles

unless one member of a pair heavily dominated a sample or both members

were, in fact, characterized by the same profile shape to begin with.

In order to obtain a complete age (mortality) profile for each taxon, it was
necessary to make crown height measurements on a deciduous tooth and on a
permanent one that erupted before the deciduous tooth was shed. For each
taxon, the choice of an appropriate deciduous-permanent pair was determined
by:

(i) the abundance of various teeth (probably dictated mainly by relative
durability; for example, in most species, mandibular teeth are more
durable than maxillary ones, and molars are more durable than premolars);

(ii) the extent to which it was possible to determine the former place in the
mouth of an isolated tooth (for example, in most species, isolated M1’s are
difficult to tell from M2’s from the same jaw, while isolated M3’s are
immediately recognizable);

(iii) the ease of measurement (particularly important in the case of a species
represented by a number of teeth still mounted in jaws; in such an instance
the base of the crown of a late-erupted tooth (e.g. M3) will often be
masked by jaw-bone, and crown height can be measured only on a tooth
(e.g. M1) that entered occlusion considerably earlier).

The crown heights and age profiles (calculated by the linear formulae) for
Langebaanweg Sivatherium hendeyi (bed 3aN), Giraffa sp. (and palaeotragine)
(bed 3aN), Ceratotherium praecox (QSM), Mesembriportax acrae (QSM and
bed 3aN), and Simatherium demissum (bed 3aN) were presented in an earlier
paper (Klein 1981a). The age profiles alone, recalculated using the quadratic

74 ANNALS OF THE SOUTH AFRICAN MUSEUM

TABLE 4

The initial (unworn) crown heights (in mm) used to calculate the age profiles of the Langebaan-
weg and Elandsfontein ungulate species considered in this paper.

dP4 P4
Ceratotherium praecox 45,0 65,1
dP2 P2
Hipparion cf. baardi ZnO) 60,0
dP2 M3
Equus capensis 30,0 90,0
dP4 M1
Raphicerus sp. 5,4 10,6
dP4 M3
Giraffa sp. 16,3 27,8
Sivatherium hendeyi 23,4 45,1
Nyanzachoerus sp(p). —* 45,1
Damalacra spp. 15,0 39,6
Mesembriportax acrae 10,0 333-0
Simatherium demissum 16,3 36,0
Pelorovis sp. DD) TGS
Hippotragus gigas 23.9) 54,7

Taurotragus oryx 19,0 448

* No specimens

formulae, are presented here in Figure 15. The crown heights and age profiles
(calculated by the quadratric formulae) for Langebaanweg Damalacra spp.
(beds 3aS and 3aN), Nyanzachoerus sp(p). (QSM and PPM) and Hipparion cf.
baardi (QSM and PPM), and for Elandsfontein Equus capensis, Hippotragus
gigas, Taurotragus oryx, Raphicerus sp., and Pelorovis sp. are presented here in
Figure 7 and Figures 9-13. Figure 14 presents a recasting of the Langebaanweg
C. praecox crown heights and age profile, based on a somewhat larger sample
than in the earlier paper (and with the age profile calculated by the quadratic
formulae).

The various Figures also illustrate the crown height dimension that was
measured in each case. In general, this dimension was the minimum distance
between the occlusal surface of a tooth and the line separating the enamel of
the crown from the dentine of the roots, measured on the buccal surface of
mandibular teeth and on the lingual surface of maxillary ones. On multilobed
(or multilophed) teeth the measurement was made on the first (anteriormost)
lobe (or loph), except in samples specifically noted in the Figures, where it was
necessary to measure the second lobe because the first one was so often
damaged or missing.

Finally, the Figures present and illustrate crown breadth measurements for
each taxon. These were analysed for excessive variability and for deviations
from statistical normality that may indicate the underlying samples were mixed
or heterogeneous. In no instance was a significant departure from normality
detectable. This is a particularly important finding in the case of these dental
samples—especially the large one of Damalacra from Langebaanweg—that
probably or certainly derive from two closely related species. If the crown

WS

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN

"S}YSIOY UMOIO JY} WOT] poyepnoyeo
sopyoid (AylyeqOW) ade oy) pue ‘UIa]UOJspuL[Y WO ssid snsv4joddiyyT pue xXk410 snsp1JOAND I, JO S\YSIOY UMOID pu SYyIpPedIq UMOIO FJ pue FgP 6 ‘3I4

o's2 O'O€ O'se 0'Ob

NWdS34I17 JO LNS0Y3d

(WW) SINIOdGIN SSV19
O'Sb o'0S o'ss al S'91 6 bl

| Tse

1LH9ISH NMOYO EW (8=N) ELOFEb'9I
NYOMNN O3SLVWILSS

Hiavaue
NMOUD EW

%00| 06 08 OL 09 OS Ob OF OZ O! L'vs
= syyuow 9F + | ————— ¢ (9¢-882) =39V
z L'pS-HO=W
aS é
= ‘
> sysuow oH 6 Ben 9¢=59V
31140ud 39V 51? 2 6'€2- HOvdP
(WW) SINIOdGIN SSV10
o'o! o's! 0'02 ose O'O¢ o'se 0'Ob O'Sb o'os 6s! ¢4 Z's
LHOIZH NMOND EW (6=N) Ib'lFZ6Z1
NYOMNN G3LVWILS3
HLiGvayEg
NMONO £W
LHOISH NMONO EW
z Ss
NVdS3411 JO LN39Y3d
%00! 06 08 OZ 09 0S Ob OF OZ O! g'bb
2 sysyuoOW ee = =
3 ysuow Og + Paar (O€- Ob2)= 39
< A
So
(<= ‘
= sysuow pe lOKS lis fb = 39V
340d J9V a1? 0'61- HO "dP

rd

1H9I3H NMOND tap (Z=N) O21 F bS‘Ol
NYOMNN Q3LVWILS3

LH9I3H HLOV3aue
NMOUD Yap NMOUD Yap
sobib snbo1yoddify

S
s9 ss sol s‘zi S‘bl s‘9I Sel ¢‘Zl s'il La
LHOISH NMOYD *gP (62N)S¢'O FIS'II
NYOMNN G3ILVAILSS
LH9ISH HLovaue
NMOUOD bdp NMOUD Yap

XA10 SNBDNOINDL

NISLNOIJSGNV 13

ANNALS OF THE SOUTH AFRICAN MUSEUM

76

‘}]X9] 94} UI Passnosip suoseal
IO} ‘s1A04OjJag poyuny ATIaNov s[dood ssy sUO}S IOATY Solsepy oy) 1eY} sjsossns opyoid JoATY soisepy oy) ul (uUedsos Jo %OT
}SIY OY} UL) SfenprArpur suNoOA Ar9A Jo sOURpUNGe 19}v919 YONU IY], ‘SIYSIOY UMOIO OY) WOIJ poyepNoyeo soryoid (Ay]e1OWL) a3e oy}
pue ‘T dAP_D YINOP[ JOATY Solsepy pue Ulo}UOJspuL[_A WO siAosojag JO SJYysIOy UMOID pu syIpPedIg UMOIO FW pue Fgp ‘Ol ‘34

(UW) SLNIOdGIN SSV19

Ol S| oz Se o¢€ SE Ob Sb OS GG 109 SS! "OZ cGy cise s02 Ol S| O02 Sz ozl OSI OF!
a =
(1 =u) 92 (9z=u) 2641 Faz" (62=u)
81=U4) S2'‘1FH9'e7 92=U) 2S‘1 F8z‘ez7 62=U) 8Z'O F9b ‘El
WY 1H913H NMOND WHY»
bdP NYOMNN
Ol S| Oz G2 o'2zl o's! O'¢!
‘
j 5 (p=U) Oz'zFOl'e2 (S=4)00'2F8b 9
(2=4) S82 FLb IZ 41H9OISH NMOYO
LHSISH NMOS bdP NYOMNN
€W NYOMNN
133 143
—+—_
(8S=U) 22'2FeS'be
LHSISH HLOV3Yad

LHOISH NMOND Yap NMOUD “ap
€
NMOYDEW 9
Higv34ua [-
a NMONO £W | a,
6: a wo

wd

z

(<=

=

ios)

Sil m

Po)

(eo)

wv

OZ _

z

io)

<

cZo

>

(143) (WHY) HLNOW =
NISLNOJISONV 13 Y3AIN S3ISVIN Oe
GE

S3AWsdOY"d ALIMVLYOW

™

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN

‘ds snsao1ydDy

"s}ystay UMOIO
dy} Woy payejnojeo syyosd (Aqyeyow) o8e oy) pue ‘ulajUOJspueyy wor ‘ds snsaoiydpy Jo syysioy UMOIO puR sYyIpedIg UMOID TJ puke PdP “TI ‘SI4

(WW) SINIOdGIN SSV 19

%00!1 O

NvdS3djI71 JO LN390Y3d
6 08 O2 09 OS Ob O€ OZ OI

31sd0uUd J4SV

STVNGIAIONI

O

0"!

so sii s‘z se S‘b S'S

LHOISH NMOYD
t4p NYOMNN G3LVWILS3

o'2 O'¢ O'b O'S O'9 OZ O'S 06 OO! O'll

$5528.20

EERO
OC OL?

LHOISH NMOUD
IW NYOMNN G3LVWILSS

Or Be HE HE Ze O'€

——

(€l=N) S8°2 F pS'Se

HiGv34a NMONO tap

(ZS=N) 9b°EFOS'6b

HiLOvV3YNa NMOND |W

NISLNOISGNV 13

ANNALS OF THE SOUTH AFRICAN MUSEUM

78

sopyoid (Ayyej1owl) ad8e oy) pue ‘Somueeqosur’y] je Nee pue See spog wos ‘dds niovjnwog jo s}YsI9Yy UMOID pu syIpesIg UMOID EJ PUL FP ‘ZI

o's sz o'ol s‘2i o's! s'2l

BSS OS

o's SZ o'ol s‘2l O's! (97)

o* x SORRY
o

NWdS34!7 JO LN30Y¥3d
%00!1_ 06 O08 OL 09 OS Ow OF O2 Ol

(182 = N)
NDE B SOE Speg

o'0z

SES ee
TVNGIAIGNI JO Y3ASWNN

‘S}YsIoy UMOIO OY} WO] poyepNoyes

(WwW) SLNIOdGIN SSV19

o'oe s'2e O'se s‘ze

Sz O'Sb

SS II I I

oS
OY

oS
xy

(Z8=N) 9Z'0F 12‘

£W spe aaa

(WwW) SLNIOdGIN SSV 19

i 6s! I's! el s'e! Ms

o's2 EONS 2 vi Eel| El

s‘22 O'Sb

<
CO S22 L
Y

2
Q
Q
Q

(pleN) €2'SFLG6'6E
LH9SISH NMOYD £W NYOMNN

fw NDE pag

— 0o¢

(€El=N) B2Z'OF Z2'I1

Sasoud FO9v,,

NWdS34I1 JO LN30Y43d NvdS34I1 JO LN39Y3d

%001 06 08 O4 09 OS Ob OF OZ OI
— =
———— ES eS a

== —— Ol z
——==_ 0 oe

o¢ i oc &

or 2 ov

os z 5

09 <

u 8 ae

(6SI=N) NDE peg gt (Z2l2N) Soe peg are

“B14
os sto) vzs oss PREBBLE HL OL,
-¢
ae
(612 N) #v'O FOSL
Li \
° f ( (| 8q0})
| yypoasq 1psoq
(Wy ) ae
ox AMT
(| ©qo}) /
syBiey uMosd [| \\ 1 )
—\F Yup spe aad
o's sil Gt Ge SL ee ore
——1G)
—+-
(pizN) SEO FESL
|
1LH9I3H NMOUD
dP NYOMNN G3LVWILS3
tap NDE Gag

VY DV VW

\ WN
\ . \
\\ i pp /

SSAIMNVVESONV 1

WM,

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN

‘s}ySIOYy UMOID OY} WIT poyefnofeo sopyoud (Ajyej1oW!) ode oy) pue ‘domuveqosueyT]
wor Uuouvddiyy JO s\ysIoy UMOIO puR syIpeeIg UMOID Zq pue ZAP ‘MOTIG ‘SIYSIOY UMOID 9Y} Wos, pojepno[eo sojyoid
(Ajyeqiour) o8e oy) pue ‘Zomueeqosuey] wor sndaoyovzuvAN JO s]YSIoy UMOID puk SY]PeIIG UMOIO EY] pue PdP :9A0QY ‘ET “SIA

(WW) SLNIOdGIN SSV19

S Ol sl oz Gz of cE Ov Sb Os Ss 09 o'oe ~=s‘'zz~—s0'Sz Ss (o) S| 02 sz o'o2 §GLi o's!
+ WSO WSO
(C#N) 6b'OF C6'SS
S (e)] SI 02 G2 o¢ ce Ov Sb OS Ss 09 o'o2 s‘22z o's2 S (e)} S| (oy sz o'o2 = SLI o'si
| onl Nad Wdd
LH9I3H NMOND zd NYOMNN (L=N) 89'|FO8'Z2 LH9I3H NMOS 2dP NYGMNN

OaLVWIisa G31VWILS3
ae FoR Hidvaue
LHSISH NMOYD od J HLOV3ye
its 2 ivsv8 od LH9I3H NMOND adP IwSva 2dP
NWdS3411 JO LNS0Yad NVdS34I7 JO LN30Y3d NvdS34I1 40 LN39Y43ad
% 001 06 08 OL 09 OS Ob OF 02 OI %001 06 08 OL 09 0S Ob OF OZ OI % 001 06 08 OL 09 OS Ob OF OZ OI UOlMO COL}
J
Wdd '8 WSO Wdd WSO

pool delOlsle} BR hy,

(WwW) SLNIOdGIN SSV19

o's ov] o'ol S‘2l o's| S‘2l o'o2 s‘2zz o's2 sue ose size O'O¢ sizz. 90 0'S2
8 % : x eae 0 H H U u H (SN3WID39dS ON) (SNSWIO3dS ON)
ral ee WSO WSO
(plan) 29'2FoS'Oe
Bin UG OlOlu iG ellaeOrGh = Gull sOrccsmiGrecs mOiSe sue o'se size o'og siz o'sz
x x s x x x H 4 Lr (SNIWID3dS ON) (SNAWID3dS ON)
a
—_|— (9=N) €S'bF#80'ZE Wdd Wdd
(G°N) €O'NF9E'22
LHOISH NMOYD £W NYOMNN —¥ v y
sate f $ HiGv34a LHOSISH NMOYND ‘dP HLOVSYE WvSve "dP
af €
—) LHOIH NMOND £W Jn aN
NvWdS3417 JO LN39Y3d NvdS34I71 dO LN30Y3Sd NVdS34I71 JO LN3DYad
%00I 06 08 OL 09 OS Ob OF OZ OI! %001 06 08 OL 09 OS Ob OF OZ O| %00!1 06 08 OL 09 OS OF OF OZ O! SNIBOYIOZUDANY
Wdd WSO

Wdd 8 WSO
SA ndOdd ADV,

80 ANNALS OF THE SOUTH AFRICAN MUSEUM

METACARPAL II

AGE PROFILE’

54,49+2,96 (n=70)

—

‘Sex Profile’

NUMBER OF INDIVIDUALS

specimens

basal breadth

70,42£3.05 (n=47)

~~ lO=

c estimated unworn
@® :

= crown height

> S—

a

77)

Comey 70) 75 65 60 55 50 45 40 35 30 25 20 IS
class midpoints (mm) class midpoints (mm)

3 dp 4

2

a crown

5 height

5

57,74 +296 (n=l3) timated
cee “OHA Ceratotherium praecox

a 5— |

Cc

@

£

oO

a

CY 55 60 65 45 40 35 30 25 20 IS

class midpoints (mm) class midpoints (mm)

LANGEBAANWEG

Fig. 14. Below: dP4 and P4 crown heights and crown breadths of Ceratotherium praecox from

Langebaanweg. Above right: the age (mortality) profile calculated from the crown heights.

Above left: the distribution of measurements of the minimum mediolateral diameter (‘mini-

mum breadth’) of the third metacarpal of Ceratotherium praecox from Langebaanweg. The two
modes probably represent the sexes, as discussed in the text.

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 81

breadth distribution had departed significantly from normality, then the use of
a single set of constants to calculate ages from crown heights in the mixed
samples would be open to serious question.

All crown heights and breadths were measured to the nearest tenth of a
millimetre with a single pair of Helios dial-reading calipers.

INTERPRETATION OF THE LANGEBAANWEG AND
ELANDSFONTEIN AGE PROFILES

Figure 15 presents the Langebaanweg and Elandsfontein age profiles in
directly comparable summary form. With the exception of females in Nyanza-
choerus (Langebaanweg) and Raphicerus (Elandsfontein), females in all the
taxa involved probably had no more than one young per year, based on
observations of their living representatives or of their closest living relatives.
With the exceptions of Nyanzachoerus and Raphicerus then, all the taxa were
probably characterized by catastrophic (‘Ix’) and attritional (‘dx’) profiles that
were very different in form. More specifically, each species would have a
catastrophic profile in which successively older age classes contained progres-
sively fewer individuals. It would have an attritional age profile that was either
U-shaped, with a large mode in the youngest class and a second smaller one in
an age class beyond 40 per cent of potential lifespan, or possibly L-shaped, with
a large mode in the youngest age class and no obvious modes thereafter.

From Figure 15, it is apparent that the relevant profiles do separate fairly
clearly between the expected attritional and catastrophic types. Thus, the
profiles for Langebaanweg Ceratotherium praecox (QSM), Mesembriportax
acrae (QSM), Simatherium demissum (bed 3aN), and Hipparion cf. baardi
(QSM and PPM composite) are all basically attritional, while those for Lange-
baanweg Sivatherium hendeyi (bed 3aN), Giraffa sp. (bed 3aN), and Damalacra
spp. (beds 3aS and 3aN), and for Elandsfontein Equus capensis, Hippotragus
gigas, and Taurotragus oryx are all basically catastrophic.

A problem arises with respect to the profiles of Langebaanweg Mesembri-
portax acrae (PPM 3aN) and Elandsfontein Pelorovis sp., which could be
interpreted as either catastrophic or L-shaped attritional. Since the Pelorovis
profile is based on a relatively large sample and since its form contrasts with the
clearly catastrophic form of the profile for Equus capensis, based on a sample
of similar size, it seems most likely that the Pelorovis profile is attritional or
perhaps mixed attritional-catastrophic. There is also a problem with regard to
the Langebaanweg Hipparion cf. baardi profile, which contains enough individ-
uals for consideration only when material from two very different sedimentary
units is lumped. Although its form appears clearly attritional, it will be ignored
in what follows.

The profiles that appear to be clearly attritional are not completely
identical to each other, nor are those that appear to be clearly catastrophic.
Some of the differences are probably due to chance (‘sampling error’), particu-

82 ANNALS OF THE SOUTH AFRICAN MUSEUM

Ceratotherium praecox

20 40 60 80 100

Sivatherium hendeyi
(LBW 3aN)(N=552)

20 40 60 80 100

100
80
60
40
20

INDIVIDUALS

Oamalacra spp.
(LBW 3aS)(N#122)

100
80
60
40
20

PERCENT OF

Nyanzachoerus sp(p.)
(LBW QSM) (N#1I7)

20 40 60 80 100

100
80

604 Hippotragus gigas
40 (EFT) (N# 29)

20 40 60 80 100

Mesembriportax acrae
(LBW QSM) (N=20)

20 40 60 80 100

100

Giraffa sp.
(& palaetragine)
(LBW 3aN)(N=II0)

20 40 60 80 I00

~

Damatlacra spp.
(LBW 3aN)(N#1I5S9)

24

20 40 60 80 100

100
80

60 Pelorovis sp.
404 (EFT) (N=83)

20 40 60 80 100

100
80
60
40
20

Tavrotragus oryx
(EFT) (N=28)

20 40 60 80 100

PERCENT OF LIFESPAN

100
80

60 Mesembriportax acrae
4o4 (LBW 3aN)(N=28)

20

20 40 60 80 100

.
oC |

Simatherl/um demissum
404 (LBW 30N)(N#!7)

20

20 40 60 80 100

Hipparion cf. board
404 (LBW QSM & PPM (N#12)

20 40 60 80 1|00

100

80

oe Equus capensis
404 (EFT)(N=89)

20

20 40 60 80 100

100

80 Cc VS

604 Raphicerus sp.
(EFT){(N#81)

40

20 40 60 80 100

Fig. 15. Age (mortality) profiles of all the Langebaanweg and Elandsfontein ungulate
species discussed in the text.

= a

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 83

larly likely to affect some of the smaller samples. Other differences are
probably due to errors in the constants used in the age estimation formulae.
Such errors are almost certainly too small to affect overall profile shape, but
could well affect the precise content of particular age classes.

Neither chance nor the constants used in the formulae, however, can
explain the fact that all the profiles except the ones for Sivatherium hendeyi and
Giraffa sp. (and palaeotragine) contain many fewer very young individuals (in
the first 10% of lifespan) than would be expected in either a catastrophic or an
attritional age profile. At both Langebaanweg and Elandsfontein bone pres-
ervation is good to excellent, so that selective post-depositional destruction of
very young (relatively fragile) jaws and teeth is probably not the reason for
their frequent underrepresentation. Much more likely is pre-depositional
destruction or removal, primarily by scavengers.

It was pointed out above that selective destruction or removal of the bones
of very young individuals almost certainly accounts for their underrepresenta-
tion in mortality profiles based on jaws collected recently by wildlife biologists.
The smaller the species, the greater the underrepresentation, which probably
accounts for the fact that very young individuals are so poorly represented in
Damalacra spp. as against their abundance in the giraffids, though in both
cases, the basic mortality pattern is catastrophic. In order to avoid confronta-
tions with other scavengers attracted to catastrophic death sites, individual
scavengers probably often carried off the carcasses of very young Damalacra,
but had to feed in place on the much larger (and relatively less fragile)
carcasses of very young giraffids.

The catastrophic age profiles for Sivatherium, Giraffa, and Damalacra from
the Langebaanweg PPM (beds 3aS and 3aN) suggest that death occurred by
drowning during peak or flash flooding. This suggests in turn that Damalacra and
the giraffids are so much more abundant than other species in the PPM channel
fills because their feeding habits tied them to the ancient river even during those
periods (?seasons) when increased precipitation meant that surface water and
green vegetation were widespread. For the giraffids, the river margins might
have been fatally attractive because the trees or high bush on which they fed
were restricted to this environment. For Damalacra spp. the attraction might
have been floodplain meadows or the ecotone between riverine forest and
fringing grassland, habitats clearly favoured today by their close living relative
the tsessebe (Damaliscus lunatus) (Smithers 1971; Child et al. 1972).

Hendey (1981a) has suggested that between the deposition of bed 3aS
(earlier) and bed 3aN (later), grassland increased at the expense of bush and
forest in the ancient Langebaanweg environment. By 3aN times bush and forest
might have been all but absent away from the river margins. This could account
for the remarkable increase in giraffids compared to Damalacra in the bed 3aN
(compared to 3aS) channel fill. The nutritional stress that habitat reduction
induced in the bed 3aN sivatheres may explain the hypoplasia that is sometimes
obvious in their dental enamel (Hendey 1981).

84 ANNALS OF THE SOUTH AFRICAN MUSEUM

The attritional profiles of Ceratotherium praecox and Mesembriportax acrae
from the Langebaanweg OSM suggest that death occurred mainly as a result of
predation, disease, starvation and other attritional factors, in keeping with a
mode of bone occurrence and sedimentary context that imply subaerial death
very near the places where the bones lie. The attritional profiles of Simatherium
demissum and possibly of Mesembriportax acrae from the Langebaanweg PPM
channel fills are more difficult to understand. Given the rarity of these species
in the PPM, it is possible their bones were either swept by the river from the
adjacent floodplain or reworked from deposits of a more ancient floodplain on
which attritional mortality was the rule. It is further possible that the ambigu-
ous, neither clearly catastrophic nor clearly attritional, shape of the PPM 3aN
M. acrae profile reflects a mixture of catastrophic mortality by drowning and
- attritional mortality on the adjacent floodplain.

The probably attritional shape of the Pelorovis age profile from Elandsfon-
tein is in keeping with the fact that this species is commonly represented by
semi-articulated skeletons scattered across the ancient land-surface(s) now
exposed by deflation at the site. It is not difficult to imagine that Pelorovis
deaths were due mainly to attritional factors at or near the ancient waterholes
that probably led to concentrations of animals at Elandsfontein.

Whatever the precise causes of Pelorovis mortality at Elandsfontein, the
rarity of very young individuals is interesting in so far as these are well
represented in an attritional profile of Pelorovis from the Middle Stone Age
(early Upper Pleistocene) archaeological site of Klasies River Mouth (Klein
1978b) (Fig. 10 here). Since the rarity of very young Pelorovis at Elandsfontein
probably reflects pre-depositional destruction or removal by carnivores and
scavengers, their abundance in an archaeological site implies either that people
were able to locate the carcasses of very young individuals before other
predators or that they actively hunted the very young. A priori, the second
alternative seems more likely.

With the Pelorovis profile in mind, it is puzzling that the other relatively
abundant large ungulates at the site—Equus capensis, Hippotragus gigas, and
Taurotragus oryx—are characterized by catastrophic rather than attritional
profiles. In the case of H. gigas and T. oryx it is possible that larger samples
would turn what now appear to be catastrophic profiles into L-shaped attri-
tional ones, but the Equus capensis sample is clearly too large to suppose that
even a substantial increase in size would transform what is an almost classic
catastrophic profile into an attritional one.

At Elandsfontein today, the bones of Equus capensis (and of most other
species) appear to occur differently than do Pelorovis bones, with fewer
instances of semi-articulated skeletons. This may reflect the fact that E.
capensis (and most other species) lacked the massive skulls and vertebral
columns of Pelorovis sp., so that semi-articulated E. capensis skeletons are both
less likely to occur and less likely to be noticed, or it may mean that the E.
capensis bones are coming mainly from deflated porcupine or carnivore lairs.

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 85

An indeterminate, albeit smaller, proportion of the Pelorovis bones may also
come from such lairs, adding a pseudo-catastrophic overlay to an otherwise
essentially attritional profile. Skeletal element counts from the fossil carnivore
(probable hyena) lairs at Swartklip and elsewhere in southern Africa show size
or weight sorting of large ungulate bones that may bias age profiles based only
on teeth towards the catastrophic side (Klein 1975, and unpublished).

As a final alternative, it is possible that the E. capensis bones derive mainly
from predation by a creature that had the ability to kill individuals of all ages in
roughly their live proportions. Further speculation on this would be fruitless,
and perhaps the major point to be made is that the interpretative potential of
mortality profile analysis is obviously dependent upon the availability of good
contextual information.

With respect to the Nyanzachoerus profile (Langebaanweg QSM) and the
Raphicerus profile (Elandsfontein), sample profile shape alone is inadequate for
determining whether death was largely attritional or catastrophic. This is
because female Nyanzachoerus probably had litters of two or more, as have
extant African suids (Mentis 1972). Extant Raphicerus females generally bear
just one young at a time, but are fully capable of producing two surviving
young per year (Mentis 1972). Given this reproductive potential, high rates of
mortality in the fossil populations of Nyanzachoerus and Raphicerus could well
have continued into young adult age classes, leading to catastrophic and
attritional profiles that would both have the down-staircase form of catastrophic
mortality in a species where females have a much lower reproductive rate.

Figure 15 shows that the Nyanzachoerus and Raphicerus profiles both
exhibit down-staircase shapes. Since the Nyanzachoerus material was drawn
from a sedimentary unit (Langebaanweg QSM) in which context and the
mortality profiles of other species imply that attritional mortality was the rule,
it seems most likely that the Nyanzachoerus profile, in fact, reflects attritional
mortality in a species that regularly bore two young or more at a time.

It is more difficult to argue that attritional mortality is probably reflected in
the Elandsfontein Raphicerus profile, because it is unclear from modern obser-
vations whether female Raphicerus do commonly produce two young per year.
The problem is that Raphicerus spp. (grysbok and steenbok) are solitary,
secretive, and inconspicuous, in short, difficult to observe systematically. It is
pertinent, however, that Raphicerus mortality profiles from a wide variety of
southern African sites, including both archaeological ones (Die Kelders, Nelson
Bay, Klasies River Mouth, and others) and carnivore (probable hyena) ones
(Swartklip, Sea Harvest, and others), all display the same down-staircase shape
as the Elandsfontein profile (Klein 19815) (Fig. 16 here). None of the profiles
displays the classic U- or L-shape that would reflect attritional mortality in a
species characterized by a birth rate of one young (or less) per female per year.

The essential similarity in form of the Raphicerus profiles from such a wide
variety of contexts, where both attritional and catastrophic mortality are
probably represented, strongly suggests that Raphicerus populations are,

86

ANNALS OF THE SOUTH AFRICAN MUSEUM

70

60

507 ELANDSFONTEIN
40 (N=81)

70 70 70
60 60 60 ease ae
504 SWARTKLIP | 504 SEA HARVEST

(N= 41) (N =19) fe

401 & = 401 8
so] A = 20] EL
20 E == ==
v == == SES
~==—= == ==
4. ES 2=| = ° LER B
ras 40 60 60 100 20 40 60 60 100 20 40 60 80 100
=
OQ 70
° 70 70
= 60 60 60
rs)
505 =
| BYNESKRANSKOP | °° F) DIE KELDERS | °°] NELSON BAY CAVE
4044 (LSA) (Nre7) go El (LSA) (NE277) go (LSA) (N= 114)
Wi 30 30 a5
2 oles == EE
20-4 20 + ==
re == == ei==
Qa 10K 10- ==
=== === =—S/
20 40 60 60 100 20 40 60 80 100 20 40 60 80 100
70 70 70
60 =
ELANDS BAY cAVE ©° FA pie KELDERS! ©°7 KLASIES RIVER MOUTH |
sof4 (LSA)(N=130) 59} (MSA) (N=227) (MSA) (N= 59)
404 404 401 FB
so 30-44 30
20 2044 20 +443
Mzzs oH of
==s=_ ==2.— =
20 40 60 80 100 20 40 60 80 100 20 40 60 80 100

PERCENT OF LIFESPAN

Fig. 16. Age (mortality) profiles of Raphicerus from Elandsfontein and from the
fossil carnivore (probable hyena) lairs at Swartklip, Sea Harvest, and Equus
Cave, the Middle Stone Age levels of Die Kelders Cave 1 and Klasies River
Mouth Cave 1, and the Late Stone Age levels of Byneskranskop Cave 1, Nelson
Bay Cave, Elands Bay Cave, and Die Kelders Cave 1. Very young individuals are
probably underrepresented in several of the profiles because of the relative
fragility of their jaws and teeth. Keeping this fact in mind, the profiles all display
a general tendency for successively older age classes to contain progressively
fewer individuals. Since some (?most) of the profiles are probably attritional,
while others may be catastrophic, the implication is that Raphicerus females
commonly bear more than one young per year, as discussed in the text.

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 87

indeed, ones in which catastrophic and attritional profiles do not contrast
notably in form. This would be the case only if female Raphicerus regularly
produce two young per year. The Raphicerus example points up the fact that
age profiles based on fossil samples may be used to draw basic biological
inferences—in this case, the common rate of reproduction—that may be
difficult to determine from observations of live populations.

SEASON OF BONE ACCUMULATION AT LANGEBAANWEG AND
ELANDSFONTEIN

It was pointed out above that individuals from a fossil population in which
both birth and death were seasonally restricted events would fall into discrete
age classes or clusters, each separated from the next by an age gap in which no
individuals occur. Kurtén (1953) was probably the first to point out that crown
heights of a tooth drawn from a population with a discontinuous age distribu-
tion may also form clusters or modes, each cluster representing a group of
individuals born at about the same time. In his study of the same north Chinese
ungulates discussed above (age profiles in Figure 5), he believed he could
detect crown height modes reflecting seasonal birth or death, the latter prob-
ably by flooding (drowning).

In his study of the extinct pronghorn antelope, Merycodus furcatus, Voor-
hies (1969) also found crown height multimodality. It is more convincing than
in Kurtén’s samples, since Voorhies’s sample was so much larger. Variably
convincing crown height multimodality, taken as evidence for seasonal birth
and death, has also been reported for Bison spp. from several archaeological
sites in the western United States (Frison 19785, with references).

Both the crown height distributions and the distributions of estimated
individual ages (ungrouped) of all the ungulate samples reported here were
examined for multimodality that would reflect seasonal birth and death. The
possibility of finding such multimodality seemed particularly good for the
giraffids and Damalacra spp. from the Langebaanweg PPM channel fills, both
because the flooding that was the probable cause of death might well have been
seasonal and because the dental samples involved are quite large. However, a
convincing pattern of multimodality was not found.

For the giraffids, the problem may be that, like their close living relative,
Giraffa camelopardalis (Mentis 1972; Foster & Dagg 1972), they bore their
young more or less throughout the year, without clear seasonal peaks. Addi-
tionally, giraffid teeth are low crowned relative to potential individual lifespan,
so that crown height modes, even if they existed, would be very close together
and difficult to detect, except in truly enormous samples. In this context it is
pertinent that Klein et al. (1981) failed to find convincing crown height
multimodality in a large sample of wapiti with a similarly low crown height to
lifespan ratio, though seasonal birth and death in the wapiti population was
historically documented.

88 ANNALS OF THE SOUTH AFRICAN MUSEUM

It is also possible that Damalacra spp. were not seasonal breeders, though
their closest living relatives, various alcelaphine antelopes, generally are (Men-
tis 1972). Furthermore, individuals of Damalacra spp. had much higher crowns
and shorter potential lifespans than the giraffids. Perhaps the reason the
Damalacra samples fail to exhibit clear multimodality is that they comprise two
species, each with a somewhat different birth season or perhaps somewhat
different initial crown heights and eruption schedules.

In sum, it remains possible that bones did accumulate in the Langebaan-
weg PPM channel fills seasonally, but it will take either larger samples, or more
homogeneous ones, or both to show this from crown heights.

SEX RATIOS IN THE FOSSIL SAMPLES

Although ungulate species usually exhibit sexual dimorphism in bone
morphology or size, or both, there are serious practical obstacles to estimating
the sex ratio in most fossil samples. The principal problem is that the skeletal
parts that reflect sex the best are generally not very durable, and they are often
less durable in one sex than in the other. In bovids, for example, the skeletal
part from which sex can be most readily determined is the frontlet or part of
the skull that bears the horns. In most species female frontlets either lack horns
or exhibit ones that are quite different in size or shape from male horns. The
problem is that frontlets are among the least durable parts of the skeleton, and
the pre- and post-depositional destructive pressures that have affected most
fossil samples remove frontlets selectively. Thus in most samples they are too
rare for the reliable estimation of a sex ratio. Additionally, male frontlets tend
to be more robust than female ones, so that even when the number of frontlets
is large enough to estimate a sex ratio, there is a likely bias in favour of males.

In the present context the relative fragility of frontlets, particularly female
ones, almost certainly accounts for the fact that frontlets are much less common
than other bovid skeletal elements (particularly jaws and teeth) at both Lange-
baanweg and Elandsfontein, while male frontlets clearly predominate.

In equids the skeletal parts from which sex can be most readily determined
are the anterior portions of the jaws, which usually bear large canines in males
and very small (‘vestigial’) canines or none in females. Unfortunately, the
relevant parts of equid jaws are not very durable, while in females they are
even less durable due to the lack of large canines. Again, in the fossil equid
samples considered here, the relevant parts of the jaws are relatively rare and
male specimens dominate heavily.

An alternative and, in general, less biased way to estimate the sex ratio
derives from the fact that in most species, probably including all those consid-
ered here, bones of males are larger on average than their homologues in
females (Boessneck & Von den Driesch 1978, with references). The search for
size bimodality reflecting sex must generally be limited to bones on which the
epiphyses are fused, or modes reflecting the sexes may be confounded with

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 89

ones reflecting different age groups. In any case, bones on which the epiphyses
are unfused tend to be rare in many fossil samples, probably because of their
relative fragility. In general, among bones with fused epiphyses, the most
fruitful ones to scrutinize for size bimodality are those that bear weight, since
they will probably reflect larger male size most strongly.

With regard to the samples considered here, three—those of Ceratotherium
praecox from Langebaanweg and of Equus capensis and Pelorovis sp. from
Elandsfontein—contain a sufficient number of mature, weight-bearing bones to
undertake a search for size bimodality. In each case the most suitable bones in
terms of sample size and the likelihood of showing a sex difference were the
central metapodials (III or III/IV). (The numbers of central metacarpals and
metatarsals were respectively 80 and 65 for C. praecox, 76 and 87 for E.
capensis, and 127 and 96 for Pelorovis sp.) In so far as bone completeness
permitted, a total of six measurements was made on each metapodial (maxi-
mum total length, maximum proximal mediolateral diameter, maximum proxi-
mal anteroposterior diameter, maximum distal mediolateral diameter, maxi-
mum distal anteroposterior diameter, and minimum mediolateral shaft
diameter).

The measurements on both the metacarpals and metatarsals in each sample
were examined for size bimodality, using the histogram-creating capability of
IDA (Interactive Data Analysis—Ling & Roberts 1982), as implemented on
the University of Chicago’s Amdahl 470 computer. The advantage of using
IDA is not only the speed with which it produces histograms from
measurements but its ability to produce a histogram based on any desired class
breadth or distribution midpoint. Sometimes a narrower class breadth, chosen
second, will reveal bimodality that a broader one, chosen first, had masked.

No histogram of Equus capensis metacarpal or metatarsal measurements
displayed convincing bimodality, while the only Pelorovis histogram to display
any was that for total length of the metacarpal, and even in this case the
bimodality was relatively weak. The problem is probably not that the sexes in
E. capensis or Pelorovis sp. were the same in average size but rather that the
difference in average size would show up clearly only in much larger meta-
podial samples.

The results for Ceratotherium praecox were more gratifying, in that a plot
of the minimum mediolateral shaft diameter of metacarpal HI showed clear and
convincing bimodality, as reproduced here in Figure 14. Assuming the individ-
uals represented by the lower mode of the plot are females and those repre-
sented by the higher mode are males, the plot suggests that male and female C.
praecox are about equally represented. This compares with apparent parity in
the adult sex ratios of the closest living relatives of C. praecox, the white
rhinoceros (C. simum), and the black rhinoceros (Diceros bicornis) (Mentis
1972). The implication is that the pattern of attritional mortality reflected in the
C. praecox age structure did not differ significantly between the sexes, as it
apparently does not in the living relatives of C. praecox.

90 ANNALS OF THE SOUTH AFRICAN MUSEUM

In sum, a reliable sex ratio is much more difficult to establish than a
reliable age profile from fossil material. An additional complication, which
should be noted here, is that even given a reliable sex ratio from a fossil
sample, interpretation may prove difficult. This is because modern observations
indicate that the sex ratio in ungulates varies considerably, both among species
and among populations of the same species.

CONCLUSIONS

The principal conclusions of this paper may be listed as follows:

1. An ungulate species in which females regularly produce one young (or
less) per year will have a catastrophic (= ‘lx’) age profile with a down-staircase
shape in which successively older age classes contain progressively fewer
individuals. The corresponding attritional (= ‘dx’) profile will tend to be
U- shaped, with a large peak in the youngest age class and a second smaller
peak beyond 40-50 per cent of potential individual lifespan, or L-shaped, with
a large peak in the youngest age class and no obvious peaks thereafter.

2. An ungulate species in which females regularly produce more than one
young per year will have catastrophic and attritional mortality profiles that both
exhibit the down-staircase form. In such species, a difference in form between the
profiles will occur mainly at the ‘tail’ of old individuals, which will tend to exhibit a
small rise (or peak) in attritional profiles but not in catastrophic ones. Detecting
such a rise and demonstrating its statistical reality will probably require samples
larger than the ones palaeontologists or archaeologists can usually obtain.

3. Very young ungulates will generally be underrepresented (in contrast to
their live abundance) in mortality profiles obtained from skeletal remains,
whether these are recent or fossil. The reason is that the bones of very young
individuals are more fragile than those of older ones. In fossil sites where bone
preservation is good, so that post-depositional destruction was probably mini-
mal, the underrepresentation of very young individuals is most likely due to
pre-depositional destruction or removal by carnivores and scavengers. It fol-
lows that sites in which very young individuals are abundant must either be
ones at which carnivores and scavengers were absent, or ones to which they
took bones but did not consume them. In the case of a hominid (archaeologi-
cal) site where bones of very young individuals are abundant, the implication is
either that the people were able to locate the carcasses of very young individ-
uals before other predators, or that they actively hunted the very young
individuals. A priori, the second alternative seems more likely.

4. Predation on ungulates may produce either attritional or catastrophic
mortality profiles. Attritional mortality profiles will tend to characterize long-
standing predator-prey relationships. It is probable that only recent man
regularly produces catastrophic prey profiles. Predation that leads to cata-
strophic profiles must be restrained, voluntarily, or otherwise, if it is not to
reduce or extinguish prey populations.

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 91

5. The interpretation of mortality profiles derived from fossil samples is
dependent upon the availability of sound contextual information. Thus, the
conclusion that the giraffid and Damalacra individuals in the Langebaanweg
channel fills died by drowning during flash or peak floods is based both on the
mortality profiles that characterize the samples and on the sedimentary facies in
which they occur. The lack of contextual information for most of the bones
collected at Elandsfontein makes it difficult, if not impossible, to interpret the
Elandsfontein mortality profiles meaningfully.

6. Mortality profiles based on fossil samples may be useful for establishing
basic biological facts about certain species that are difficult to establish by
observation of live populations. Thus the fact that mortality profiles of Raphi-
cerus from a wide variety of sites always exhibit the same down-staircase form,
in spite of the fact that mortality at some (?most) of the sites was probably
attritional, implies that Raphicerus females commonly bear more than one
young per year. More generally, many fossil samples are probably derived from
populations in which short-term fluctuations in overall size and age structure
have balanced each other out, so that the survivorship and mortality patterns
that the samples exhibit are the long-term ones that biologists seek in order to
gauge the ecological status or health of living populations.

7. In theory, the crown height distributions of a fossil population that was
characterized by seasonally restricted birth and death will exhibit patterned
multimodality reflecting the discontinuity of the underlying age distribution. In
practice, the discovery of convincing multimodality will require much larger
samples than most palaeontologists or archaeologists generally obtain, prefer-
ably from species in which dental crowns were very high relative to potential
individual lifespan.

8. In fossil samples, sex ratios are generally more difficult to establish and
more difficult to interpret than age profiles. Measurements on weight-bearing
bones that are likely to reflect size differences provide the most widely
applicable means of estimating the sex ratio in fossil ungulate samples.

9. Although ageing animals from skeletal remains can be problematic, for
many high-crowned ungulate species ageing from teeth using quadratic formu-
lae such as those employed in this paper will probably provide age profiles that
are reasonably accurate. The wider application of mortality profile analysis in
palaeontology and archaeology is not limited by problems of ageing but by the
frequent lack of sufficiently large samples.

ACKNOWLEDGEMENTS

The author thanks the South African Council for Scientific and Industrial
Research, the South African Museum, and the United States National Science
Foundation for research support. K. Cruz-Uribe helped to draft the figures.
Q. B. Hendey and K. Cruz-Uribe kindly commented on a preliminary version
of the manuscript.

92 ANNALS OF THE SOUTH AFRICAN MUSEUM

REFERENCES

BoEssNECK, J. & A. DRIESCH, VON DEN. 1978. The significance of measuring animal bones from
archaeological sites. Peabody Museum of Archaeology (Harvard University) Bulletin 2:
25-39.

BourtikrE, F. 1951. Vie et moeurs des mammiferes. Paris: Payot.

Burtzer, K. W. 1973. Re-evaluation of the geology of the Elandsfontein (Hopefield) site,
south-western Cape, South Africa. S. Afr. J. Sci. 69: 234-238.

CAUGHLEY, G. 1966. Mortality patterns in mammals. Ecology 47: 906-918.

CuHILD, G., RopBpeLt, H. & HEpBurN, C. P. 1972. Observations on the biology of tsessebe,
Damaliscus lunatus lunatus, in northern Botswana. Mammalia 36: 342-388.

DeeEvey, E. S. 1947. Life tables for natural populations of animals. Q. Rev. Biol. 22: 283-314.

DRENNAN, M. R. 1953. The Saldanha Skull and its associations. Nature, Lond. 172: 791.

DRENNAN, M. R. 1955. The special features and status of the Saldanha Skull. Am. J. Phys.
Anthrop. 13: 625-634.

Foster, J. B. & Dace, A. I. 1972. Notes on the biology of the giraffe. E. Afr. Wildl. J. 10:
1-16.

Frison, G. C. 1978a. Prehistoric hunters of the High Plains. New York: Academic Press.

Frison, G. C. 19785. Animal population studies and cultural inference. Plains Anthrop. 23
(82,2): 44-52.

Gentry, A. W. 1980. Fossil Bovidae (Mammalia) from Langebaanweg, South Africa. Ann. S.
Afr. Mus. 79: 213-337.

GoppDarD, J. 1970. Age Criteria and vital statistics of a black rhinoceros population. E. Afr.
Wildl. J. 8: 105-121.

GRIMSDELL, J. J. R. 1973. Age determination of the African buffalo, Syncerus caffer Sparrman.
E. Afr. Wildl. J. 11: 31-53.

Hati-Martin, A. J. 1976. Dentition and age determination of the giraffe. J. Zool., Lond. 180:
263-289.

HENDEY, Q. B. 1974. The Late Cenozoic Carnivora of the south-Western Cape Province. Ann.
S. Afr. Mus. 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. Ann. S.
Afr. Mus. 84: 1-104.

HENDEY, Q. B. 1981b. Geological succession at Langebaanweg, Cape Province, and global
events of the Late Tertiary. S. Afr. J. Sci. 77: 33-38:

Hun t_ey, B. J. 1973. Ageing criteria for tsessebe (Damaliscus |. lunatus). J. sth. Afr. Wildl.
Mgmt Ass. 3: 24-27.

INSKEEP, R. R. & HENDEY, Q. B. 1966. An interesting association of bones from the
Elandsfontein fossil site. Publnes Mus. arqueol. Santa Cruz de Tenerife 6: 109-124.

JOUBERT, E. 1972. Tooth development and age determination in the Hartmann Zebra Equus
zebra hartmannae. Madoqua 1: 5-16.

Kerr, M. A. & Rotn, H. H. 1970. Studies on the agricultural utilisation of semi-domesticated
eland (Taurotragus oryx) in Rhodesia. Horn development and tooth eruption as indicators
of age. Rhod. J. agric. Res. 8: 149-155.

KLEIN, R. G. 1975. Paleoanthropological implications of the non-archeological bone assem-
blage from Swartklip 1, south-western Cape Provirce, South Africa. Quaternary Res., N.
Y. 5: 275-288.

KLEIN, R. G. 1978a. The fauna and overall interpretation of the “Cutting 10’ Acheulean site at
Elandsfontein (Hopefield), south-western Cape Province, South Africa. Quaternary Res.,
N. Y. 10: 69-83.

KLEIN, R. G. 19785. Stone age predation on large African bovids. J. archaeol. Sci. 5: 195-217.

KLEIN, R. G. 1981a. Ungulate mortality and sedimentary facies in the late Tertiary Varswater
Formation, Langebaanweg, south-western Cape Province, South Africa. Ann. S. Afr.
Mus. 81: 233-254.

KLEIN, R. G. 1981b. Stone age predation on small African bovids. S. Afr. archaeol. Bull. 36: in
press.

KLEIN, R. G., ALLWARDEN, K. & WOLF, C. In press. The calculation and interpretation of
ungulate age profiles from dental crown heights. Jn: BatLeEy, G. ed. Hunter—gatherer
economies in Late Pleistocene Europe. Cambridge: University of Cambridge Press.

UNGULATE MORTALITY, LANGEBAANWEG AND ELANDSFONTEIN 93

KLEIN, R. G., Wor, C., FREEMAN, L. G. & ALLWARDEN, K. 1981. The use of dental crown
heights for constructing age profiles of red deer and similar species in archaeological
samples. J. archaeol. Sci. 8: 1-31.

KLINGEL, H. 1965. Notes on tooth development and ageing criteria in the plains zebra Equus
quagga boehmi Matschie. E. Afr. Wildl. J. 3: 127-129.

KLINGEL, H. & KLINGEL, U. 1966. Tooth development and age determination in the plains
zebra (Equus quagga boehmi Matschie). Zool. Gart. Frankfurt 33: 34-54.

KurtTen, B. 1953. On the variation and population dynamics of fossil and recent mammal
populations. Acta zool. fenn. 76: 1-122.

Laws, R. M. 1966. Age criteria for the African elephant, Loxodonta africana africana. E. Afr.
Wildl. J. 4: 1-37.

Laws, R. M. 1968. Dentition and ageing of the hippopotamus. E. Afr. Wildl. J. 6: 19-51.

Linc, R. G. & Roserts, H. V. 1982. IDA: a user’s guide to the IDA Interactive Data Analysis
and Forecasting System. New York: McGraw Hill.

Manson, J. 1974. Aspects of the biology and behaviour of the Cape grysbok Raphicerus
melanotis Thunberg. Unpublished M.A. thesis, University of Stellenbosch.

MeEntis, M. T. 1972. A review of some life history features of the large herbivores of Africa.
Lammergeyer 16: 1-89.

MirTcHELL, B. L. 1965. Breeding, growth and ageing criteria of Lichtenstein’s hartebeest. Puku
3: 97-104.

Morris, P. 1972. A review of mammalian age determination methods. Mammal Rev. 2: 69-104.

Murie, A. 1944. The wolves of Mount McKinley. Fauna natn. Pks U.S. 5: 1-238.

Nimmo, B. W. 1971. Population dynamics of a Wyoming pronghorn cohort from the Eden-
Farson site, 48W304. Plains Anthrop. 16: 285-288.

RICHTER, W. von. 1971. Observations on the biology and ecology of the black wildebeest
(Connochaetes gnou). J. sth. Afr. Wildl. Mgmt. Ass. 1: 3-16.

RICHTER, W. von. 1974. Connochaetes gnou. Mammalian Spec. 50: 1-6.

RIGHTMIRE, G. P. 1976. Relationships of Middle and Upper Pleistocene hominids from
sub-Saharan Africa. Nature, Lond. 260: 238-240.

SCHALLER, G. B. 1972. The Serengeti Lion. Chicago: University of Chicago Press.

Simpson, C. D. 1966. Tooth eruption, growth and ageing criteria in the greater kudu—Tragela-
phus strepsiceros Pallas. Arnoldia (Rhodesia) 2: 1-12.

SincLair, A. R. E. 1977. The African buffalo. Chicago: University of Chicago Press.

SINGER, R. 1954. The Saldanha Skull from Hopefield, South Africa. Am. J. Phys. Anthrop. 12:
345-362.

SINGER, R. 1957. Investigations at the Hopefield Side. Jn: CLark, J. D. ed. Prehistory: Third
Pan-African Congress Livingstone 1955: 175-182. London: Chatto & Windhus.

SINGER, R. 1958. The Rhodesian, Florisbad and Saldanha skulls. Jn: KOENIGSWALD, G. H. R.
von, ed. Neanderthal Centenary: 52-62. Utrecht: Kemink en Zoon.

SINGER, R. 1961. The new fossil sites at Langebaanweg, South Africa. Curr. Anthrop. 2:
385-387.

SINGER, R. & Hoover, D. A. 1958. A Stegolophodon from South Africa. Nature, Lond. 182:
101-102.

SINGER, R. & Wyner, J. 1968. Archaeological investigations at the Saldanha Skull site in South
Africa. S. Afr. archaeol. Bull. 25: 63-74.

SMITHERS, R. H. N. 1971. The mammals of Botswana. Mus. Mem. natn. Mus. Monuments
Rhod. 4: 1-340.

Smuts, G. L. 1974. Age determination in Burchell’s Zebra (Equus burchelli antiquorum) from
the Kruger National Park. J. sth. Afr. Wildl. Mgmt. Ass. 4: 103-115.

Sowts, L. & PHELPs, R. J. 1968. Observations on the African bushpig Potamochoerus porcus
Linn. in Rhodesia. Zoologica 53: 75-83.

Spiess, A. 1979. Reindeer and caribou hunters: an archaeological study. New York: Academic
Press.

SPINAGE, C. A. 1970. Population dynamics of the Uganda defassa waterbuck (Kobus defassa
ugandae Neumann) in the Queen Elizabeth Park. J. anim. Ecol. 39: 51-78.

SPINAGE, C. A. 1971. Geratodontology and horn growth of the impala (Aepyceros melampus).
J. Zool., Lond. 164: 209-225.

SPINAGE, C. A. 1972a. African ungulate life tables. Ecology 53: 645-652.

94 ANNALS OF THE SOUTH AFRICAN MUSEUM

SPINAGE, C. A. 1972b. Age estimation of zebra. E. Afr. Wildl. J. 10: 273-277.

SPINAGE, C. A. 1973. A review of the age determination of mammals by means of teeth, with
especial reference to Africa. E. Afr. Wildl. J. 11: 165-187.

VooruiEs, M. R. 1969. Taphonomy and population dynamics of an early Pliocene vertebrate
fauna, Knox County, Nebraska. Contr. Geol. spec. Pap. Univ. Wyoming 1: 1-69.

6. SYSTEMATIC papers must conform to the /nternational 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 eee 1845: 37.
Leda plicifera A. Adams, : 50.
Laeda bicuspidata Hanley, See 118, pl. 228 (fig. 73). Sowerby, 1871: pl. 2 (fig. 8a—b).
Nucula largillierti Philippi, 1861:
Leda bicuspidata: Nicklés, 1950: 165, 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, it not preceded
by initials or full names
e.g. Du Toit but A.L.du Toit; Von Huene but F. von Huene

(c) Scientific names, but not their vernacular derivatives
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.

RICHARD G. KLEIN

PATTERNS OF UNGULATE MORTALITY

AND UNGULATE MORTALITY PROFILES

FROM LANGEBAANWEG (EARLY PLIOCENE)
AND ELANDSFONTEIN (MIDDLE PLEISTOCENE),
SOUTH-WESTERN CAPE PROVINCE,

SOUTH AFRICA

3 SEPTEMBER 1982 ae wets - ISSN 0303-2515

APE 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
(c) Table of contents giving hierarchy of headings and subheadings
(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
(Gi) 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 x 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, 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 a/. 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.

Konn, A. J. 19606. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean.
Bull. Bingham oceanogr. Coll. 17 (4): 1-51.

THELE, 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 90 Band
September 1982 September
Part 5/3) ~ Weel

“lp
Coun now KS

REVISION OF THE
SOUTHERN AFRICAN ANTHURIDEA
(CRUSTACEA, ISOPODA)

By
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 WEES, 58), 10D, 4-8, 3, cof), SOS, 5, 7S).
AG tans), HOD), 8 OCD, 1), 1Ge8),
11(1-2, 5, 7, t-—p.i.), 15(4-5), 24(2), 27, 31(1-3), 32(5), 33, 36(2), 45(1)

EDITOR/REDAKTRISE
Ione Rudner

Copyright enquiries to the South African Museum

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 039 7

In Suid-Afrika gedruk deur
Die Rustica-pers, Edms., Bpk.,
Courtweg, Wynberg, Kaap

Printed in South Africa by
The Rustica Press, Pty., Ltd.,
Court Road, Wynberg, Cape

REVISION OF THE SOUTHERN AFRICAN ANTHURIDEA
(CRUSTACEA, ISOPODA)

By
BRIAN KENSLEY

Smithsonian Institution, Washington, D.C.

(With 64 figures)

[MS accepted 28 April 1982]

ABSTRACT

The South African anthuridean isopod fauna, which includes thirty-seven species in sixteen
genera and two families, is reviewed. Full synonymies, diagnoses, and all available material
examined have been included for each species. The following new species have been described:
Haliophasma austroafricana, Malacanthura schotteae, M. transkei, and Mesanthura dimorpha.

The following nomenclatural changes have been made: Natalanthura foveolata is placed in
the genus Apanthuroides; Haliophasma caecus in Centranthura; Exanthura filiformis (sensu
Barnard), and E. macrura in Haliophasma; Horoloanthura capensis in Kupellonura;
Haliophasma coronicauda, H. foveolata, H. hermani, H. ornata, H. pseudocarinata, and
Agulanthura serenasinus in Malacanthura; Anthelura remipes in Quantanthura; and Zulanthura
laevitelson in Accalathura.

The distribution patterns of the southern African anthurideans are compared with those of
other geographical areas.

CONTENTS
PAGE
WARE CNS CET OME fasion Ben iosnies oa 9. 5d bs heh a ote EON on RN OD ead i cet 96
SWStemiatic GISCUSSION: 1.2 noni) tee xls Cee oe eg ee Si
Ney to thefamilies of the-Anthuridea:< 252 4) sane se oe a oe 97
Key to the South African genera of the family Anthuridae....... 97
naniily Amghuni Gaetan. ait. satis eet chy cteeeeeeemes ute eden 98
ApanihunastepomelO00).. 6 ete cree ccs ee en ee oes fe 98
mApanthuroidessNenzies é& Glynn>) 1968. 4. o=9seene 1. one 106
GeniranthuraW agelce (OSG nae Jp. ee etn ees Caso nae, oe 109
CGyathura Normané& Stebbing,, 1886) 5.0 ee tao eio ee ce 113
idalophasmatdaswe ly USS.) tis seer oe ar Cee ee 116
upelonura se anmarda O25 serene eo: sae re ee ee ae 126
Malacanthura Barmarde925 55 anne ee Cee ee 128
Mesanthuraipannard alOl4 se. eee os ee ee 153
INGO ESLVG ENGI es Spaces eis Gener aee Taree aeecle oan G 159
AnathuraBagnand. W925. oh va. eee coon. en «se es ec ee 159
Ouantanthura Menzies: & George, 1972 een. on eee ee 164
Key to the South African genera of the family Paranthuridae .... 169
amily, Paranthunidae. . eases aioe ee Ce erento Nee 169
AccalathuraBarnard’ 192555) ce oo ue oe oe ne ee 169
GolanthuraRichardsoms 1900s. 2) eye oe ae se ITAL
HECDIANIRUTA SATS MA SOO Mere ti een och ow a eke ee Wee 174
Paranthura Bate & Westwood, 1868 .- 2.6... ct 185
Eseudanthura Richardson. 911 4%. cf a oe poe 191
Zooceorrapmicalicomments: ©... oc. es a hewae ir olhy priser ae sens 197
PCknOWwiled Bemme Nise coi 4 a sf 5a Rid oe ote re was ~ 197
PRE TERENCE S here erate rer ot Nici ec Se creat a ial sha cn, BK umn ed eee ct etn 198

95

Ann. S. Afr. Mus. 90 (3), 1982: 95-200, 64 figs.

96 ANNALS OF THE SOUTH AFRICAN MUSEUM

INTRODUCTION

Between 1925, when Barnard summarized the knowledge of anthuridean
isopod systematics, and the early 1970s, no major study of the group has
appeared. At present, however, there are several workers concentrating on the
group, which is consequently approaching some degree of generic stability.
Because of Barnard’s interest in the anthurideans and because of much
subsequent collecting, the southern African area is now represented by
thirty-seven species, more than any other region in the world. It was thus felt
that a systematic review of the southern African anthurideans would be useful,
especially in updating generic and specific diagnoses.

For this study, the entire holdings of the South African Museum and the
University of Cape Town’s Department of Zoology have been examined, as
well as some material in the British Museum of Natural History, the
Smithsonian Institution, the Muséum National d’Histoire Naturelle, Paris, and
the Zoological Museum, Copenhagen. Also included is material from the 1978
and 1979 Meiring Naude cruises of the South African Museum.

Full synonymies are provided for each species. Full descriptions of new or
poorly-known species are given; for species already well documented, only a
diagnosis containing the species’ most distinctive features is given. The material
examined has been divided into two sections, viz. type material and other
material. For a few species the whereabouts of the types are unknown. Those
described by Stimpson from the United States North Pacific Exploring
Expedition, collected in False Bay, Cape, are presumed to have been lost in the
Chicago fire of 1871 (see Rathbun 1907). Localities, along with depth records
where known, are given for all material.

Keys to families, genera, and species have been constructed for ease of
use, and do not necessarily use phylogenetically significant features.

In the Material sections, distinction is made between ovigerous and
non-ovigerous females, mature males (bearing whorls of filiform aesthetascs on
the antennular flagella), submales (representing a stage in the protogynous
development from female to male, and recognized by the lack of antennular
aesthetascs), juveniles, and mancas (the latter having not yet developed the
seventh pair of pereopods).

For most of the species, a figure of the entire animal (usually a mature
female) is provided, plus line figures of the most significant features. In
addition, scanning electron micrographs have been used to illustrate some
details. Occasionally, a feature may be illustrated both by line drawings and
SEMs. In such cases, the SEMs are regarded as essential for interpretation of
fine surface detail, as well as for three-dimensional configuration and
orientation. It is felt that an abundance of figures, especially SEMs, will give a
better understanding of functional morphology of the group, and cast light on
aspects of adaptive radiation and phylogeny.

The present treatment of the southern African Anthuridea is certainly not
the final word, as the author has examined material that has not been included.

SOUTHERN AFRICAN ANTHURIDEA 97

For example, an undescribed species of Mesanthura from Lideritz (represented

by

a single specimen), and several immature or damaged specimens from the

continental shelf off the east coast have been omitted. Only with more
specimens can this material be adequately described. Reinterpretation of some
genera and species (e.g. Apanthura and Leptanthura) will almost certainly
become necessary.

The following abbreviations are used throughout the paper:

BMNH British Museum (Natural History)

SAM South African Museum

SIO Scripps Institution of Oceanography

USNM United States National Museum of Natural History, Smithsonian
Institution

ZMC Zoological Museum, Copenhagen

juv.(s) juvenile(s)

Ovig. ovigerous

Be total length

All locality references to False Bay refer to False Bay, Cape, and not False

Bay, Natal. In the sections Other material, all localities are given from west to
east.

oD |

—

SYSTEMATIC DISCUSSION

KEY TO THE FAMILIES OF THE ANTHURIDEA
Mouthparts adapted for biting and cutting, i.e. mandible possessing (usually) molar and

mMetsoientaihHa with:several distal SpiMnes ~ . 2. e. eeeee setae Sees ee er eee Anthuridae
Mouthparts adapted for piercing and sucking, i.e. mandible lancet-like, lacking molar and
PRIETO tpmerY PUNE ACS UY PROMI 2850 ey, Pe ah Se iss «eR ee eR eS He ale A ote Paranthuridae

KEY TO THE SOUTH AFRICAN GENERA OF THE FAMILY ANTHURIDAE

eM E CSO OMECE hye ae yee aig Se ace © Ue RIA EN Oe hi eis pgs 2
rene Sea nC OMNECS. I 9) TOSCO ia 21), a5 Se eis he lek ee Ieee Ae I as eee SAL +
. Pleonites 1-6 free-elongate
Exeapecolplecopod 1 non-operculiform ..¢.0024.05 85.2 oe seen dew tae tees eens 3
At least pleonites 1-5 free, short
Pxoanneromplcopod. 1 operculiform 4): . ..o 6. sins ents ese ee Sata ee he del Panathura
. Telson dorsoventrally flattened
PR ARMNNIBIC CLOW OISCOIMCMES, Statism seats spk etna? wale MG ks ASR Se Kupellonura
Telson spiniform, terete
Ra aeeeABIE CL ONMCSCOICMEG certs ects oh ae ak eee Ae co eptitrece ey Neohyssura
PME ONIMPIC CRON OISEPINCINS 2 ou Bo oe cscee Bajos RE OE ee ees 3 Re ERE oe Quantanthura
Maawhececontcwer than'G ScomientS, 20.) 2. ees eee cee kent ee Fe ees ene 5
ee CHO SC PINCHES) epee eh Cie Khe ete ne nee hah eeeradid SH Ee Leis Pome Ree 6
Maxtlhined of fewer than 5 SComemS 2.225 68a he See kis ek oe ena oe ee ene ne 9
. Persistent brown-black dorsal pigment pattern present..................-.. Mesanthura
Botistcnupioment pattem aAUSENb ag o.0s<. f sctc. Gc os Ua ak Fel ueles onde ad oo eee 7

. Pereopods 1-3 subsimilar, barely subchelate

Motarol mandible absent on one side’. 2.0... ee es desk eee oe ore Apanthuroides

98 ANNALS OF THE SOUTH AFRICAN MUSEUM

Pereopod 1 subchelate, markedly larger than following pereopods

Molanot mandible normaly presention both’sidess-. oe ne ee eee 8
8: Pereopods 4—7 with triangular carpuse: aaocme ooo a oe eee eee Apanthura
= Pereopods 4-7 with rectangular carpus™..44.5-42500- ooo ono eee eee Malacanthura
9 Maxillipedioi4rsepmicmts) 3 Sidaiier ph toes Series cate a ish eh orn chliy so ee eae 10
= sMaxillipedioti's SC pmemts) iii on carrer tee en seas cw ieyee tain een ao eee Centranthura
LO; Pereopods 4-7 with tiangularcarpus:.)2 = 5. ee) se ee ee ae ee Cyathura
= 'Pereopods 4-7 with rectangular carpus sen e ese 452 en oe eee Haliophasma
Family Anthuridae
Apanthura Stebbing, 1900
Diagnosis

_ Antennular flagellum of two to four articles; antennal flagellum of two to
four articles. Mandible with three-segmented palp; incisor, lamina dentata, and
molar present. Maxilliped five-segmented, endite present or absent. Pereopod 1
subchelate, propodus expanded. Pereopods 4-7 with triangular carpus
underriding propodus. Pleopod 1 exopod operculiform. Pleonites 1-5 fused,
pleonite 6 free. Telson with two basal statocysts.

Type species
Apanthura sandalensis Stebbing, 1900.

KEY TO THE SOUTH AFRICAN SPECIES OF APANTHURA

{. Uropodal exopod strongly notched\ 72: 445.-44 2-5 oes. Oe ee dubia
= Uropodalexoped margin entire, convexor SiInuOUS).....-.. .00- 45. 400 e nee 2
2. Eyes present; telson with distal raised lozenge-shaped area ..................... africana
= Eves absent telson with taint mid=dorsalendgern 4.444 japon ei ee insignifica

Apanthura africana Barnard, 1914
Figs 1-2
Apanthura africana Barnard, 1914: 340a, pl. 28C; 1925a: 142; 1940: 490, 498. Nierstrasz,

1941: 241. Day, Field & Penrith, 1970: 47. Kensley, 1975a: 38; 1978a: 46, fig. 20G-I.
Apanthura cf. sandalensis: Penrith & Kensley, 1970: 226.

Diagnosis

Integument lacking sculpture. Pereonites 4-6 each with mid-dorsal pit.
Pleonites 1-5 fused, dorsal lines of fusion complete between pleonites 1-4,
incomplete between pleonites 4 and 5; pleonite 6 free, with mid-dorsal notch in
posterior margin. Telson slightly raised, lozenge-shaped in distal half; distally
rounded. Pereopod 1 propodus expanded; palmar margin setose, with low
rounded lobe at midpoint. Uropodal exopod with outer margin slightly sinuous
distally.

Type material
Holotype, non-ovig. 2, SAM-A63, 17,0 mm, off Saldanha Bay, 160 m.

SOUTHERN AFRICAN ANTHURIDEA

Fig. 1. Apanthura africana. A. @, dorsal view. B. Maxil-
liped. C. Mandible. D. Uropodal exopod. Scale in mm.

99

100 ANNALS OF THE SOUTH AFRICAN MUSEUM

Other material

SAM-A12615, 2 non-ovig. ¢, Ltideritz, intertidal SAM-—A12630, 1 juv.,
Liideritz, intertidal. SAM-—A12739, 3 juvs, Liideritz, intertidal. SAM—A5961, 1
?, off Saldanha Bay, 174 m. SAM-A14040, 1 non-ovig. 2, off Saldanha Bay,
16m. SAM-A14046, 1 subd, off Saldanha Bay, 146m. SAM-A14047, 1
non-ovig. 2, off Saldanha Bay, 148 m. SAM-A14048, 1 subd, 1 non-ovig. &,
off Saldanha Bay, 79 m. SAM-A14052, 1 subd, 27 non-ovig. 2, 13 juvs, off
Saldanha Bay, 11m. SAM-A14053, 4 juvs, off Saldanha Bay, 91m.
SAM-A14054, 1 non-ovig. °, off Saldanha Bay, 172m. SAM-A14055, 6
non-ovig. 2, 12 juvs, off Saldanha Bay, 146m. SAM-A14056, 1 subd, 4
non-ovig. 2, 4 juvs, off Saldanha Bay, 142 m. SAM-—A14057, 1 non-ovig. 2,
off Saldanha Bay. SAM-—A14067, 2 6d, 4 non-ovig. ¢, 13 juvs, off Saldanha
Bay, 146m. SAM-A14113, 1 non-ovig. 92, off Saldanha Bay, 13 m.
SAM-A14118, 3 non-ovig. 2, 2 juvs, off Saldanha Bay, 15 m. SAM-—A14325, 1
non-ovig. ¢, off Saldanha Bay. SAM-A14326, 1 6, off Saldanha Bay.
SAM-A14327, 1 subd, 1 non-ovig. °, off Saldanha Bay. SAM-A14348, 1
non-ovig. ¢, off Saldanha Bay, 148 m. SAM-A14060, 1 juv., False Bay, 35 m.
SAM-A14043, 2 non-ovig. ¢, 1 juv., off Still Bay, 200 m. SAM—A14051, 1
non-ovig. °, off Jeffreys Bay, 32 m. SAM—A14063, 1 non-ovig. °, off Jeffreys
Bay, 25 m.

Distribution
Liideritz to Jeffreys Bay, intertidal to 200 m.

Apanthura dubia Barnard, 1914
Figs 3-4

Apanthura dubia Barnard, 1914; 342a, pl. 28D; 1955: 5.

Apanthura sandalensis non Stebbing, Barnard, 1925a: 141; 1940: 490, 498. Nierstrasz,
1941: 241 (partim). Day, Field, & Penrith, 1970: 47. Kensley, 1978a: 46, fig. 20J—K;
1980: 12, fig. 7. .

Diagnosis

Pleonites 1-5 fused, three complete fusion line-grooves between pleonites
1-4, incomplete line between pleonites 4 and 5. Telson widest at midlength,
with low rounded proximal mid-dorsal ridge, becoming obsolete distally.
Antennular flagellum of two articles. Antennal flagellum of single article.
Maxilliped five-segmented, with small triangular endite. Pereopod 1 in male
and female with corneous proximal tooth on propodal palm; carpus triangular,
with distal corneous tooth, more marked in male. Uropodal exopod with deep
distal notch.

Type material

Syntypes, SAM-8826, 2 non-ovig. 2, 10,0 mm (both damaged), St. James,
False Bay, intertidal.

SOUTHERN AFRICAN ANTHURIDEA 101

er agg

as Pee deund
%

Fig. 2. Apanthura africana. A. ¢ cephalon, dorsal view. B. 2 cephalon, ventral view.
C. Subd cephalon. D. Pleon and telson. E. Telson, dorsal view. F. Statocyst apertures.

102 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 3. Apanthura dubia. A. Antenna. B. Antennule. C. Mandibular palp. D. Pereopod 1
@. E. Pereopod 7. F. Pereopod 1 6. G. Mandibular incisor and lamina dentata.
H. Maxilliped. I. Uropodal peduncle and endopod. J. Uropodal exopod.

SOUTHERN AFRICAN ANTHURIDEA 103

Other material

SAM-A17494, 1 non-ovig. 2, 11,0 mm, Saldanha Bay. SAM-A14069, 2
non-ovig. 2, Agulhas Bank, 36-54 m. SAM-A14353, 1 non-ovig. °, 11,0 mm,
Agulhas Bank, 27 m. SAM-A14401, 1 non-ovig. 2, 8,0 mm, Agulhas Bank,
27 m. SAM-A14068, 1 6d, off Still Bay, 200 m. SAM—A14070, 5 non-ovig. 2,
1 6, off Still Bay, 200 m. SAM-A17495, 1 non-ovig. 2, off Mossel Bay.
SAM-A14072, 2 non-ovig. °, off Jeffreys Bay. SAM-A17496, 3 juvs, off
Transkei, 150-200 m.

Distribution
Saldanha Bay to Transkei, intertidal to 200 m.

Fig. 4. Apanthura dubia. A. 2 cephalon, dorsal view. B. 2 cephalon, ventral view. C. Pleon and
telson, dorsal view. D. Pleon and telson, lateral view.

104 ANNALS OF THE SOUTH AFRICAN MUSEUM

Remarks

Apanthura dubia has for many years masqueraded under the name of A.
sandalensis Stebbing, following Barnard’s synonymizing his own species.
Apanthura sandalensis Stebbing (1900), from the Loyalty Islands, south-
western Pacific Ocean, however, does not have a triangular process on the
carpus or propodal palm of pereopod 1. (Barnard, 1925a, does mention that a
co-type in the British Museum (Natural History) has a palmar tooth as in A.
dubia, which casts doubt on the identity of the specimen.) The telson of A.
dubia is distally narrower, the antennal flagellum has fewer articles, and the
uropodal exopod is broader than in A. sandalensis.

Barnard’s (1935) A. sandalensis from Travancore, Kerala, India, and
Pillai’s (1966) A. sandalensis from Kerala, India, probably belong to neither of
these species. Pillai’s species shows a uropodal endopod too slender, a more
setose telson, and antennae having too many flagellar articles.

Chilton (1924) recorded A. sandalensis from Chilka Lake, India, but the
very slender antenna and antennule, the former with a three-articulate
flagellum, makes this identification doubtful. Larwood (1940, fig. 7) recorded
A. sandalensis (which he too synonymized with A. dubia) from Alexandria,
Mediterranean Sea, but the sinus of the uropodal exopod is not as deep as in A.
dubia, while the uropodal endopod is less elongate. Larwood’s material thus
probably does not belong to either species.

Apanthura insignifica Kensley, 1978
Figs 5-6
Apanthura insignifica Kensley, 1978b: 2, figs 1-2.

Diagnosis

Eyes weakly pigmented. Telson elliptical-oval, with weak mid-dorsal
longitudinal ridge, distal margin broadly rounded. Uropodal exopod oval,
margin entire.

Type material

Holotype, SAM-A15646, 1 non-ovig. 2, 5,9mm, off Natal, 690 m.
Paratypes, SAM—A15647, 1 non-ovig. 2, 4,5 mm. 1 subd, 5,6 mm, off Natal,
850 m. Paratype, SAM-A15646, 1 non-ovig. 2, 5,2 mm, off Natal, 690 m.
Paratypes, USNM 170542, 2 non-ovig. 2, 5,4-5,6 mm, off Natal, 850 m.

Other material

SAM-A17497, 1 non-ovig. °, 4,5mm, off East London, 630m.
SAM-A17499, 2 non-ovig. 2, 4,3 mm, 4 juvs, south of East London, 90 m.
SAM-A17500, 1 3, 6,3 mm, off Natal, 850 m. SAM-A17498, 1 6, 5,0 mm, 1
non-ovig. 2, 4,6 mm, off Zululand, 550 m.

SOUTHERN AFRICAN ANTHURIDEA 105

Fig. 5. Apanthura insignifica. A. 2 dorsal view. B. Pereopod 1 2. C. Pleopod 2 endopod
3. D. Mandible. E. Maxilliped. F. Uropodal exopod. Scale in mm.

106 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 6. Apanthura insignifica. A. 2 cephalon, dorsal view. B. Pereopod 1 &.

Distribution
South of East London to Zululand, 90-850 m.

Apanthuroides Menzies & Glynn, 1968
Natalanthura Kensley, 19785: 5.

Diagnosis

Eyes present. Mandibular palp three-segmented; molar slender and
spike-like on one side, absent on other. Maxilliped five-segmented, endite
present. Pereopods 1-3 subsimilar, barely subchelate. Pereopods 4-7 with
rectangular carpus. Pleopod 1 exopod and endopod together forming
operculum. Pleonites 1-5 fused; pleonite 6 fused with telson. Telson with basal
statocysts. Integument pitted.

Type species
Apanthuroides millae Menzies & Glynn, 1968.

Remarks

Re-examination of the types of Apanthuroides millae from Puerto Rico,
and fresh material from Belize, Central America, has shown that the pleonal
segmentation, and especially the unusual mandibular structure, is identical to
that of Natalanthura.

Apanthuroides foveolata (Kensley, 1978)
Figs 7-8

Natalanthura foveolata Kensley, 1978a: 6, figs 3—4; 1980: 3, 32. Wagele 1981: 88-90.
Natalanthura natalensis Kensley, 1979: 823, (laps. cal.)

SOUTHERN AFRICAN ANTHURIDEA 107

Lif /,

F

Fig. 7. Apanthuroides foveolata. A. 2 dorsal view. B. Right mandible. C. Left mandible.
D. Maxilliped. E. Pereopod 1. F. Pereopod 7. Scale in mm.

: C
\
D
i ee a

108 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 8. Apanthuroides foveolata. A. ° cephalon. B. Rostrum and antennal bases. C. Pleon in
dorsal view. D. Telson and uropods.

Diagnosis
Telson with strong mid-dorsal longitudinal ridge. Telsonic and uropodal

margins serrate. Antennular flagellum of three articles; antennal flagellum of
six articles. Eyes feebly pigmented.

Type material

Holotype, SAM—A15648, 1 non-ovig. 2, 5,8 mm, off Zululand, 550 m.
Paratypes, SAM-—A15648, 2 non-ovig. 2, 3,5-5,4 mm, off Zululand, 550 m.
Paratype, SAM-A15649, 1 juv., 3,4mm, off Natal, 690m. Paratype,

SOUTHERN AFRICAN ANTHURIDEA 109

SAM-A15650, 1 ovig. 2, 6,9 mm, off Natal, 850 m. Paratypes, USNM 170543,
3 non-ovig. 2, 4,3-4,6 mm, off Zululand, 550 m.

Other material

SAM-A17501, 1 non-ovig. 2, off Transkei, 630 m. SAM-—A17502, 1 ovig.
2, 3 non-ovig. 2, 1 subd, 1 juv., off Transkei, 710-775 m. SAM-—A17503, 1
juv., off Transkei, 560—620 m.

Distribution
Zululand to Transkei, 550-850 m.

Centranthura Wagele, 1981

Centranthura Wagele, 1981: 113.
Haliophasma: Kensley, 1975b, partim.

Diagnosis

Eyes absent. Antennular flagellum of three articles; antennal flagellum of
two articles. Maxilliped three-segmented; endite absent. Pereopod 1 sub-
chelate, propodus expanded. Pereopods 4~7 with narrow rectangular carpus.
Pleonites 1-5 fused, pleonite 6 free. Telson with two basal statocysts.

Type species
Haliophasma caecus Kensley, 1975).

Centranthura caeca (Kensley, 1975)
Figs 9-10

Haliophasma caecus Kensley, 1975b: 209, figs 1-2; 1978a: 49, fig. 21 E.
Centranthura caecus: Wagele, 1981: 113.

Diagnosis

Pereonites 4-6 with mid-dorsal pit. Telson widest at midlength, with low
rounded proximal ridge, distally narrowed, distal margin broadly rounded.
Antennule with three-articulate flagellum; antenna with two-articulate flagellum.
Pereopod 1 subchelate, propodus expanded, propodal palm sinuous in female,
with peg-like tooth in male. Pereopods 4~7 with narrow rectangular carpus not
underriding propodus. Uropodal exopod oval, outer margin sinuous, apex acute.

Type material

Holotype, SAM-A13626, 1 6, 18,2 mm, Lambert’s Bay. Allotype,
SAM-A13626 1 non-ovig. 2, 16,0mm, Lambert’s Bay. Paratypes, SAM-—
A13627, 1 d, 21,0 mm, 5 non-ovig. 2, 12,5—-17,5 mm, Langebaan Lagoon.

110 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 9. Centranthura caeca. A. 2 dorsal view. B. Mandible. C. Pereopod 7. Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA 111

Other material

SAM-A14419, 15 non-ovig. °, 17 juvs, Lambert’s Bay. SAM-A14064, 1
juv., off Saldanha Bay, 51m. SAM-—A14106, 1 non-ovig. 9, 12 juvs, off
Saldanha Bay, 22m. SAM-A14107, 1 6, off Saldanha Bay, 172 m.
SAM-A14108, 2 non-ovig. 2, Saldanha Bay, 31 m. SAM-A14109, 1 non-ovig.
2, off Saldanha Bay, 18 m. SAM-A14110, 3 non-ovig. 2, off Saldanha Bay,
62 m. SAM-A14114, 2 non-ovig. °, off Saldanha Bay, 62 m. SAM-A14116, 2
6, 7 non-ovig. 2, 1 juv., off Saldanha Bay, 51 m. SAM-A14122, 1 6d, 3 juvs,
off Saldanha Bay, 15m. SAM-A14123, 1 non-ovig. ¢, off Saldanha Bay,
54m. SAM-A14407, 1 non-ovig. 2, Saldanha Bay, 27 m. SAM—A14408, 4
non-ovig. 2, 4 juvs, off Saldanha Bay, 68 m. SAM—A14409, 1 non-ovig. 2, 2
juvs, Saldanha Bay, 9m. SAM-—A14413, 1 non-ovig. °, off Saldanha Bay,
32m. SAM-A14417, 4 non-ovig. @, 16 juvs, Saldanha Bay, 31m.
SAM-A14421, 9 non-ovig. 2, off Saldanha Bay, 27m. SAM-A14422, 1
non-ovig. ¢, Saldanha Bay, 5m. SAM-A14847, 1 juv., off Saldanha Bay,
68 m. SAM-A17469, 2 non-ovig. 2, 3 juvs, off Saldanha Bay. SAM-A17470, 1
non-ovig. 2, 1 juv., off Saldanha Bay. SAM-—A17471, 1 non-ovig. 2, off
Saldanha Bay. SAM-A17472, 7 non-ovig. ¢, 4 juvs, off Saldanha Bay.
SAM-A17473, 1 non-ovig. 2, off Saldanha Bay. SAM-A14410, 3 non-ovig. 2,
2 juvs, Langebaan Lagoon. SAM-A14412, 11 non-ovig. 2, Langebaan
Lagoon. SAM-A14414, 2 6, 3 non-ovig. 2, Langebaan Lagoon. SAM-
A14423, 1 3d, 1 non-ovig. 2, Langebaan Lagoon. SAM-—A14424, 1 non-ovig.
2, Langebaan Lagoon. SAM-—A14425, 1 non-ovig. 2, Langebaan Lagoon.
SAM-A14426, 1 non-ovig. 2, Langebaan Lagoon. SAM-A14428, 1 non-ovig.
2, Langebaan Lagoon. SAM-A14429, 4 non-ovig. 2, Langebaan Lagoon.
SAM-A14430, 1 non-ovig. 2, Langebaan Lagoon. SAM-A14431, 1 6, 2
non-ovig. ¢, Langebaan Lagoon. SAM-A14432, 1 non-ovig. 2, Langebaan
Lagoon. SAM-A14427, 1 6, 1 non-ovig. 9, 1 juv., Langebaan Lagoon.
SAM-A14112, 1 3, False Bay, 9 m. SAM-A14121, 1 non-ovig. 2, False Bay,
7-9 m. SAM-A14126, 1 non-ovig. 2, False Bay, 68 m. SAM-A14406, 1 juv.,
False Bay, 31 m. SAM-A14415, 1 subd, False Bay, 31 m. SAM-A14416, 1
non-ovig. 2, False Bay, 31m. SAM-A14411, 1 63, Agulhas Bank, 44 m.
SAM-A14041, 1 non-ovig. 2, Still Bay, 80 m. SAM-—A14420, 1 non-ovig. 2,
Mossel Bay.

Distribution
Lambert’s Bay to Mossel Bay, 5-68 m.

Remarks

Wagele (1981) placed Haliophasma caecus into the new genus Centran-
thura for the following reasons: the ‘primitive’ antennule bearing four
aesthetascs, the telson which is neither significantly elongate nor dorsally
strongly keeled, and the antenna, which has a shorter flagellum than most

112 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 10. Centranthura caeca. A. Mandible. B. Maxilla. C. Maxilliped, external surface.
D. Maxilliped, internal surface.

species of Haliophasma. It is now felt that H. caecus should, indeed, be in a
separate genus, but not for the reasons given by Wagele.

In the original description, Kensley (19755) indicated that the maxilliped
was three- or four-segmented, and that the terminal segment was not distinct.
In the scanning electron micrograph (Fig. 10) a distinct groove is visible,
delimiting an obliquely-inserted fourth segment bearing five setae. In a cleared
and mounted specimen, however, this groove is seen to be not a true suture,
but a superficial fold perhaps indicating a line of fusion. The first maxillipedal
palp segment was indicated (19755, fig. 1g) as having an indistinct groove in its

SOUTHERN AFRICAN ANTHURIDEA 73

proximal third. In the scanning electron micrograph this groove is seen on the
external face of the segment, but barely indicated on the internal face, once
again indicating a line of fusion. The cleared and mounted specimen again
shows this not to be a true suture. The maxilliped is therefore regarded as being
three-segmented, as in Anthura, Exallanthura, Pendanthura, Ptilanthura,
Venezanthura, and Xenanthura.

Centranthura differs from these genera in the following respects: it does
not have the six free elongate pleonites of Xenanthura; it lacks the very
shortened pleon and maxillipedal endite of Pendanthura; it does not have a
single-segmented mandibular palp as seen in Ptilanthura and Exallanthura; it
lacks the maxillipedal endite and two-segmented mandibular palp of
Venezanthura. Wagele (1981) separates Centranthura from Anthura on the
basis of the five distal maxillipedal setae being inserted in a shallow distal
indentation, rather than laterally as in Anthura. Further differences include the
multidentate lamina dentata and the spiculate molar of C. caeca, and the
reduced antennal flagellum.

Centranthura caeca perhaps represents a form in the evolution of reduction
of the maxillipedal segments from five (as in Malacanthura) and four (as in
Haliophasma). This reduction is probably correlated with a specialized mode of
feeding as part of the infauna of fine-sediment embayments. The marginally
setose and broad segments of pereopods 2-7 are probably also correlated with
this habitat choice.

Cyathura Norman & Stebbing, 1886

Diagnosis

Eyes present or absent. Mandibular palp 3-segmented. Maxilliped 4-
segmented; endite absent. Pereopod 1 subchelate, propodus expanded;
pereopods 2-3 ambulatory; pereopods 4~7 with triangular carpus underriding
propodus. Pleopod 1 exopod operculiform. Pleonites 1-5 fused, pleonite 6 free
or fused to telson. Telson with two basal statocysts.

Type species
Anthura carinata Kroyer, 1847.

Cyathura estuaria Barnard, 1914
Figs 11-12

Cyathura estuaria Barnard, 1914: 334a, pl. 27D; 1955: 5. Miller & Burbanck, 1961: 62, 65.
Burbanck & Burbanck, 1972: 274. Kensley, 1978a: 47, fig. 21A—B.

Cyathura carinata non Kroyer, Barnard, 1940: 490. Day, Millard, & Harrison, 1952: 408. Day,
Millard, & Broekhuysen, 1954: 152. Millard & Harrison, 1954: 176. Day, 1959: 532;
1969: 78, fig. (unnumbered). Boltt, 1969: 253, 255, 259, 261.

Cyathura sp., Day, 1967: figs 3-4.

ANNALS OF THE SOUTH AFRICAN MUSEUM

114

~
Y 7]
y |

Wk

S

nw
fro ttt
Ne

ANON

Ss Ale

iS
x

~

\

——

WEsN
wk

Cone

Ce

N
AS
Ne \\ NN *s
S

Yess

ae)
°
a.
fo}
oO
te
W?)
a
a)
N
se)
e)
2.
°
oO
ian
>)
a
1S)
wy
3
|
|
>)
~
|
<
aa)
3
ae
>
oS
2)
il
)
«e)
O+
<
S
S
S
=
>>)
is)
S
&
S
=
=
S)
a
Sl
op
om
(Ly

&
=|
g
2
a9}
oO
(79)
S
Ke
S
a6
a)
2
No}
=|
S
Oo
ao)
oO
{er
kK
=
ea
ao)
©
Q.
©
*
oO
o
Mo)
ie)
QO.
je)
=)
5)
jaa
~

SOUTHERN AFRICAN ANTHURIDEA IS

Diagnosis

Integument moderately setose. Telson parallel-sided for three-quarters of
length, distally rounded, fused mediodorsally with pleonite 6. Copulatory stylet
of pleopod 2 endopod in male not extending beyond ramus. Pereopod 1 female
with rounded tooth on propodal palm; carpus distally produced, rounded.

Type material

Syntype, SAM-A68, 1 non-ovig. 2, 27,5 mm, Buffalo River estuary.
Syntypes, SAM-A2269, 14 juvs, 3,2-8,0mm, Zwartkops River estuary.
Syntypes, SAM-—A14073, 3 juvs, 7,0 mm. Zwartkops River estuary.

Fig. 12. Cyathura estuaria. A. & cephalon. B. Pereopod 1, dactylus and propodus.
C. Pleon, lateral view. D. Telson and uropods, dorsal view.

116 ANNALS OF THE SOUTH AFRICAN MUSEUM

Other material

SAM-A6289, 1 juv., St. Lucia estuary. University College of Zululand, 1
non-ovig. 2, 20,1 mm, Lake Msingazi, Zululand.

Distribution

Langebaan Lagoon; east coast estuaries and lakes from Zwartkops River,
Port Elizabeth, to Zululand.

Remarks

The species has been taken from estuarine muds in salinities ranging from
0 to 35%o.

Haliophasma Haswell, 1881
Exanthura Barnard, 1914: 336a.

Diagnosis

Integument usually indurate, often with scattered small pits. Pereonites
4-6 each with single mid-dorsal pit. Pleonites 1-5 fused, pleonite 6 free, or
rarely more or less fused with telson. Telson often sculptured, indurate, with
pair of basal statocysts. Antennular flagellum of two to six articles. Antennal
flagellum of four to seven articles. Mandibular palp three-segmented.
Maxilliped four -segmented, rarely with very reduced endite. Pereopod 1
propodus expanded. Pereopods 2-3 ambulatory, propodi not expanded.
Pereopods 4-7 with rectangular carpi, not underriding propodi. Pleopod 1
exopod operculiform, often indurate.

Type species
Haliophasma purpureum Haswell, 1881.

Remarks

Since the diagnosis of Haliophasma provided by Poore (1975), it has
become obvious that several of the South African species formerly in this genus
really belong to Malacanthura with its characteristically five-segmented
maxilliped. Only H. tricarinatum remains of the seven southern African species
previously placed in Haliophasma.

The genus Exanthura Barnard, 1914, was created for E. macrura Barnard,
E. filiformis (Lucas) being added later. On the basis of E. macrura however,
there is little justification for separating this genus from Haliophasma. The
maxilliped of E. macrura does carry a tiny endite, not seen in Haliophasma s.s.,
while the basal antennular segment bears a strong recurved spiniform process
that becomes obsolete in mature specimens. This antennular feature is regarded
as being only of specific value.

The position of E. filiformis (Lucas) is uncertain, as the type has not been
located. The description and figures given by Larwood (1940) for E. filiformis

SOUTHERN AFRICAN ANTHURIDEA 1637

Fig. 13. Haliophasma austroafricana. A. 2 dorsal view. B. Antennule. C. Antenna.
D. Maxilliped. E. Maxilla. F. Mandible. G. Distal mandibular palp segment. H. Uropod.
I. Pleopod 1. Scale in mm.

118 ANNALS OF THE SOUTH AFRICAN MUSEUM

from the Mediterranean, however, show a five-segmented maxilliped. As the
South African and Mediterranean material previously assigned to this species
obviously do not belong to the same species, a new name is provided for the
former.

KEY TO THE SOUTH AFRICAN SPECIES OF HALIOPHASMA

i lelsonidorsally;smoothydistally truncate ea are eee macrurum
= felson dorsally ndeedydistally roundedr 47.25) 625s ee eee 2
2. Kelson withisinglemmid-dorsalnidsemamaae nee nc eee austroafricanum
=~ Lelson-wath:3idorsalimdoese: < .) snee 5 ee ene oe rie ke ele eee tricarinatum

Haliophasma austroafricanum sp. nov.
Figs 13-15

Exanthura filiformis (non Lucas): Barnard, 1920: 340; 1925b: 388; 1925a: 131, pl. 4, fig. 22;
1940: 490, 497; 1959: 715, fig. 1. Nierstrasz, 1941: 239 (partim). Day, Field & Penrith,
1970: 47. Kensley, 1975a: 38; 1978a: 47, fig. 21C. Wagele, 1981: 114.

Diagnosis

Body slender, parallel-sided; pereonites 4-6 each with strong mid-dorsal
pit. Basal segment of antennule with strong conical, posteriorly-directed
tooth-like process. Telson with strong mid-dorsal carina. Mature female
unknown.

Description

Non-ovigerous female. Integument moderately indurate; numerous short
setae arising from shallow ‘dimples’ scattered over entire integument. Body
proportions: C<1=2>3<4<5>6>7=P. Cephalon with strong triangular
rostrum slightly overreaching anterolateral corners. Large dorsolateral eyes.
Pereonites 4-6 each with strong mid-dorsal elongate-oval pit. Pleonites 1-5
fused, fusion lines marked by dorsolateral grooves; pleonite 6 free, very short,
with mid-dorsal notch in posterior margin. Telson with lateral margins
subparallel, faintly sinuous; distal margin truncate-rounded, densely setose;
strong longitudinal mid-dorsal carina present.

Basal antennular peduncle segment longer and broader than two following
segments, with strong tapering posteriorly-directed tooth-like process; flagellum
of six articles, four distal articles each with single aesthetasc. Antenna bearing
numerous short setae, second peduncle segment longer than segment 3;
segments 3 and 4 subequal, segment 5 one-third longer than 4; flagellum of
seven setose articles. Mandibular palp segment 2 longest, segment 3 with
fourteen simple distal spines and one short and one elongate serrate spine;
incisor of two cusps; lamina dentata of nine serrations; molar reduced,
rounded. Maxilla with one strong and six more slender distal spines. Maxilliped
four-segmented, distal segment rounded, setose, set obliquely on segment 3;
endite absent, segments 2 and 3 subequal in length. Pereopod 1 subchelate,
expanded; unguis strong, more than half length of rest of dactylus; propodal

119

SOUTHERN AFRICAN ANTHURIDEA

‘L podosiog ‘q ‘7 podosisg ‘Dd ‘Pp [ podosiog ‘q “4 [ podosiog “YW ‘wnuvdiufoosnv pusvydoyvy ‘pl ‘314

O

120 ANNALS OF THE SOUTH AFRICAN MUSEUM

palm with low rounded scale-bearing lobe in proximal half, few simple marginal
setae; inner surface near palm with several fringed spines; carpus triangular,
short. Pereopods 4—7 with unguis about one-fifth length of dactylus; propodus
elongate-rectangular, with short spine at posterodistal angle; carpus
rectangular, with short spine at posterodistal angle. Pleopod 1 exopod
operculiform, distal margin densely fringed with plumose setae, outer surface
with groove close to lateral margin, shorter ridge set back from medial margin;
endopod tapering, not reaching exopod apex; basis with seven or eight coupling
hooks, Uropodal exopod reaching beyond base of endopod, margin finely
denticulate, densely setose, distally acute, lateral margin with strong distal
sinuousity; endopod not reaching telsonic apex, margins finely denticulate,
strongly setose, tapering distally, apically rounded.

Male. Antennular flagellum of twenty aesthetasc-bearing articles. Pereo-
pod 1 propodus with dense band of simple setae on inner surface near palm.
Pleopod 2 endopod with copulatory stylet reaching distal margin of ramus,
distally spooned.

Type material

Holotype, SAM-A14078, non-ovig. 2, 26,5 mm, 33°50’S 25°47’E (near
Port Elizabeth), 36m. Paratype, SAM-—A5965, non-ovig. 2, 21,5 mm, off
Cape Infanta, 86 m. Paratype, SAM-—A4012, non-ovig. 2, 22,5 mm, off Cape
Peninsula, 460 m.

Other material

SAM-A14079, 1 juv., False Bay, 68 m. SAM-A14081, 1 juv., False Bay,
75 m. SAM-A14086, 1 juv., False Bay, 33 m. SAM—A14088, 3 juvs, False Bay,
42 m. SAM-A14089, 1 subd, False Bay, 87 m. SAM-A14090, 1 juv., False
Bay, 73 m. SAM-A14095, 1 juv., False Bay, 36 m. SAM-A14096, 1 juv., False
Bay, 31 m. SAM-A14098, 1 juv., False Bay, 18 m. SAM-A14102, 1 juv., False
Bay. SAM-A14075, 1 non-ovig. °, Agulhas Bank, 45m. SAM-A14077, 1
non-ovig. 2, Agulhas Bank, 44m. SAM-—A14085, 1 juv., Agulhas Bank,
200 m. SAM-A14091, 1 juv., Agulhas Bank, 42 m. SAM-A14092, 1 juv.,
Agulhas Bank, 183 m. SAM-A17504, 2 juvs, off Natal-Transkei, 80 m.
SAM-A17505, 1 juv., off Natal-Transkei, 90 m. SAM-A17506, 1 juv., off
Natal-Transkei, 560-620 m. SAM-—A17507, 1 juv., off Natal—Transkei,
150-200 m. SAM-A17508, 1 juv., off Natal-Transkei, 550 m.

Distribution
Saldanha Bay to Natal, 19-620 m.

Etymology

The specific name refers to South Africa, where the species appears to be
endemic.

SOUTHERN AFRICAN ANTHURIDEA 11

Fig. 15. Haliophasma austroafricanum. A. Antennular base. B. Cephalon, ventral view.
©, Pleon: DD: ‘Velson:

Haliophasma macrurum (Barnard, 1914)
Figs 16-18

Exanthura macrura Barnard, 1914: 337a, pl. 28A; 1925a: 131, fig. 7; 1940: 490, 497; 1955: 5.
Penrith & Kensley, 1970: 227. Kensley, 1978a: 47, fig. 21D. Wagele, 1981: 114.
Exanthura macruron (sic): Nierstrasz, 1941: 239.

Diagnosis
Body widening posteriorly, pereonites 4-6 each with strong mid-dorsal pit.

Telson widening posteriorly, distal margin truncate. Uropods and telson
together forming cup-shaped protective ‘operculum’, margins densely setose.

122 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 16. Haliophasma macrurum. A. 2 dorsal view. B. Maxilliped. C. Pereopod 1. Scale in
mm.

SOUTHERN AFRICAN ANTHURIDEA 123

Antennule with basal segment bearing conical, posteriorly-directed tooth-like
process. Male unknown.

Type material

Holotype, SAM-A2667, non-ovig. 2, +22 mm TL (telson sectioned by
Barnard), Sea Point, Table Bay.

Other material
SAM-A12740, 2 juvs, 1 non-ovig. 2, 28,0 mm, Liideritz, intertidal.
SAM-A14105, 1 juv., 15,2 mm, Strandfontein, Cape, intertidal. SAM—A14104,

Fig. 17. Haliophasma macrurum. A. 2 cephalon, dorsal view. B. Cephalon, lateral view.
©. \Pleon., D:' Telson:

124 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 18. Haliophasma macrurum. A. Pereopod 7. B. Pleopod 1 exopod. C. Statocyst apertures.
D. Rounded bosses at base of pleopod 1.

1 juv., 17,0 mm, Kommetjie, Cape. SAM—A14103, 1 juv., 12,0 mm, locality
unknown.

Distribution
Liideritz to False Bay.

Remarks

Haliophasma macrurum has been found on several occasions in the tubes
of the intertidal reef-building polychaete worm Gunnerea capensis, on which it
probably preys. The cup-shaped and sclerotized tail fan neatly closes the tube
mouth, affording the isopod protection while it feeds.

SOUTHERN AFRICAN ANTHURIDEA 125

Fig. 19. Haliophasma tricarinatum. A. 2 dorsal view. B. Pereopod 1 2. C. Dactylus and
propodal palm of pereopod 1 2. D. Maxilliped. Scale in mm.

126 ANNALS OF THE SOUTH AFRICAN MUSEUM

Haliophasma tricarinatum Barnard, 1925
Figs 19-20
Haliophasma tricarinata Barnard, 1925a: 132, pl. 4 (fig. 2); 1925b: 385; 1940: 490, 498;

1955: 50. Day, Field & Penrith, 1970: 47. Kensley, 1978a: 50, fig. 22D.
Haliophasma tricarinatum: Nierstrasz, 1941: 239. Poore, 1975: 532.

Diagnosis

Body slender. Pereonites 4-6 each with mid-dorsal pit. Telson with three
dorsal rounded longitudinal ridges, with scattered pits between; strong
longitudinal midventral keel; posterior margin broadly rounded. Basal
antennular segment with outer margin rounded. Pereopod 1 propodal palm a
broadly rounded lobe with rows of fine acute scales along margin. Male
unknown.

Type material
Holotype, SAM-—A5968, non-ovig. ¢, 14,0 mm, Agulhas Bank, 80 m.

Other material

SAM-A14188, 1 juv., off Saldanha Bay, 70 m. SAM—A14076, 1 juv., False
Bay, 53 m. SAM-A14083, 1 juv., False Bay, 80 m. SAM-A14084, 1 juv., False
Bay, 39 m. SAM-A14186, 1 non-ovig. 2, False Bay. SAM—A14191, 1 juv.,
False Bay. SAM—A14192, 1 juv., False Bay, 48 m. SAM-A14093, 1 juv.,
Agulhas Bank, 183 m. SAM-A14101, 1 juv., Agulhas Bank, 73 m. SAM-
A14190, 1 non-ovig. 2, Agulhas Bank, 84 m. SAM-A5967, 1 non-ovig. °, off
Cape St. Blaize, 84 m. SAM-—A17509, 1 juv., off East London, 90 m.

Distribution
Saldanha Bay to south of East London, 48-183 m.

Kupellonura Barnard, 1925a

Diagnosis

Eyes small or absent. Mandibular palp three-segmented; molar relatively
short. Maxilliped seven-segmented; endite well developed. Pereopods 1-3
subsimilar, subchelate; pereopods 4-7 with triangular carpus underriding
propodus. Pleonites 1-6 elongate, free. Pleopod 1 similar to following
pleopods, not operculiform.

Type species
Kupellonura mediterranea Barnard, 1925a.

Kupellonura capensis (Kensley, 1975a)
Fig. 21
Holoroanthura (sic) capensis Kensley, 1975a: 75, figs 19-20.

Horoloanthura capensis: Kensley, 1978a: 50, fig. 22 E-G.
Kensleyanthura capensis: Wagele, 1981: 106.

SOUTHERN AFRICAN ANTHURIDEA 127

Fig. 20. Haliophasma tricarinatum. A. 2 cephalon, dorsal view. B. Mouthparts.
C. Pereopod 1. D. Telson.

Diagnosis
Eyes absent. Telson somewhat broadened in posterior half, then tapering

to narrowly rounded apex; margin denticulate. Uropodal exopod folding over
telson, with narrow distal extension; medial margin dentate.

Type material

Holotype, SAM-—A13555, 1 6, 3,8mm, off Lambert’s Bay, 128 m.
Allotype, SAM-—A13624, 1 non-ovig. ¢, 6,1 mm, off Lambert’s Bay, 400 m.
Paratypes, SAM-A13625, 1 subd, 3,1 mm, 5 non-ovig. 2, 4,2-5,0 mm, off
Lambert’s Bay, 172 m.

128 ANNALS OF THE SOUTH AFRICAN MUSEUM

Other material

SAM-A14193, 1 non-ovig. 2, off Saldanha Bay, 141 m. SAM-A14194, 2
non-ovig. ¢, off Saldanha Bay, 183m. SAM-A14195, 4 non-ovig. @, off
Saldanha Bay, 183m. SAM-A14196, 1 non-ovig. 9, off Lambert’s Bay,
128 m. SAM-A14197, 1 non-ovig. 2, Agulhas Bank, 183 m.

Distribution
Lambert’s Bay to Agulhas Bank, 128-400 m.

Remarks

Horoloanthura irpex Menzies and Frankenberg, the type species of
Horoloanthura, was incorrectly described and figured as having six free
elongate pleonites and a five-segmented maxilliped lacking an endite. In fact,
pleonite 6 is fused to the spiculate telson, and the maxilliped is
seven-segmented, with a well-developed endite. Wagele (1981) correctly states
that H. irpex should be in the genus Neohyssura. Thus H. capensis, which is not
a Neohyssura, must be accommodated elsewhere. Wagele (1981) erected the
new genus Kensleyanthura for H. capensis. Unfortunately, Kensley’s
description was inaccurate in the configuration of the maxilliped, which is
seven-segmented with an endite. In fact, H. capensis agrees closely with
Kupellonura mediterranea Barnard, and K. serritelson Wagele, 1981, in all
features of the mouthparts, antennules, pleon, telson, and uropods.

Kupellonura capensis differs from K. serritelson in having a digitiform lobe
on the uropodal exopod, and in having the uropodal endopod margins entire
(the outer margin is serrulate in the Bermudan species). The telson of K.
capensis is widest at about the distal third and is apically narrowly rounded.
The telson of K. serritelson is widest at the midlength and distally broadly
rounded.

Kupellonura mediterranea, of which the type from the Copenhagen
Museum has been examined, has a telson similar in shape to K. capensis but
with entire margins, except for a shallow subterminal notch on each side, and a
uropodal exopod as in K. serritelson.

Malacanthura Barnard, 1925

Agulanthura Kensley, 1975a: 72.
Haliophasma: Barnard, 1925a: 131; 1940: 382; 1955: 50 (partim). Kensley, 1975a: 72. (partim).

Diagnosis

Eyes present. Antennular flagellum of three to six articles. Antennal
flagellum of four to seven articles. Mandible with three-segmented palp;
incisor, lamina dentata, and molar present. Maxilliped five-segmented, endite
small to rudimentary, or absent. Pereopod 1 subchelate, propodus expanded.
Pereopods 2-3 smaller than pereopod 1, ambulatory. Pereopods 47 with
rectangular carpus not underriding propodus. Pleopod 1 exopod operculiform.

SOUTHERN AFRICAN ANTHURIDEA 129

Fig. 21. Kupellonura capensis. A. @ dorsal view. B. Right mandible. C. Left mandibular
incisor, lamina dentata, and molar. D. Maxilliped. E. Pereopod 1. F. Pereopod 2.
G. Pleopod 1. H. Pleopod 2 ¢. I. Telson. J. Uropod. K. Pereopod 7. Scale in mm.

130 ANNALS OF THE SOUTH AFRICAN MUSEUM

Pleonites 1-5 fused, pleonite 6 free. Telson with two basal statocysts. Pleon in
male frequently more elongate than in female.

Remarks.

With Poore’s 1975 redefinition of Haliophasma, it has become obvious that
several of Barnard’s species of Haliophasma belong to Malacanthura as now
defined. Similarly, Kensley’s Agulanthura, viewed in the light of the above
redefinitions, is now regarded as a specialized Malacanthura.

Type species
Apanthura linguicauda Barnard, 1920.

KEY TO THE SOUTH AFRICAN SPECIES OF MALACANTHURA

i) Integument with numerous smallipits| 59452 esses ane oe oe ae eee 2
= Integument lacking numerous small pitss-4-- 4.554)... s00-ce eee 4
2. Kelsonunpitted, withsingle mid-dorsalridge!=...--.0- 5.) eee 3
= elsonipitted, obscurely tm-ndgeds. an ake ee ee oo eee foveolata
3) Uropodaliexopodidistally stronglymotchediasea- aan ore eee schotteae
= Uropodal exopod ovaltanceolate;iunnotched ii: 44,442.42 ae ea eee transkei
4. Felson unsculptured . : = isc. . eceadi sc oe hee nea e See | ee tae ee eee 5
= Telson with mid-dorsal-and/or lateral ridges).24- 4-4-0 404 oe eee 6
5. Telson dorsally convex; uropodal exopod closely adpressed to telson.......... serenasinus
— Telson dorsally flat; uropodal exopod freestanding, not adpressed to telson .... linguicauda
6. Telson dorsally concave, with very strong mid-dorsal ridge .................... hermani
= Welson‘dorsally flat or'convex... 24) 6.3.5 ieccbeee dente hee eee ee eee 7
7. Telson with faint lateral ridges, lacking mid-dorsal ridge.......................4. ornata
= Melson with faint lateral and mid-dorsalinidges 252.54... +--+ 44-442 - aoe eee 8
S| Telson-almostas widevas lone isc gacs oo) oho ate oer ee eee pseudocarinata
= wWelson almosttwice as long asswider: an. a> «chschaene iere: aeeoe aie oe coronicauda

Malacanthura coronicauda (Barnard, 1925)
Figs 22-23
Haliophasma_ coronicauda Barnard, 1925a: 132; 1925b: 386; 1940: 490, 498; 1955: 50.

Nierstrasz, 1941: 239. Day, Field & Penrith, 1970: 47. Poore, 1975: 532. Kensley,
1975a: 38; 1978a: 49, fig. 21F. Christie, 1976: 155.

Diagnosis

Telson with mid-dorsal raised area, faintly tri-ridged. Pereopod 1 propodal
palm with arcuate proximal lobe, numerous simple setae along inner palmar
margin.

Description

Non-ovigerous female. Integument moderately indurate, dorsally smooth,
lacking pits. Proportions: C<1=2=3>4<5>6>7. Cephalon with low
rostrum, rounded anterolateral corners, large dorsolateral eyes. Pleonites 1-5
fused, no indication of lines of fusion in dorsal view; posterior margin of
pleonite 5 with slight mid-dorsal point; pleonite 6 free, with mid-dorsal notch in
posterior margin. Telson with 2 basal statocysts, widest at midlength, distally

SOUTHERN AFRICAN ANTHURIDEA 131

Fig. 22. Malacanthura coronicauda. A. 2 dorsal view. B. Antennule. C. Antenna.
D. Mandible. E. Maxilliped. F. Pereopod 1. Scale in mm.

132 ANNALS OF THE SOUTH AFRICAN MUSEUM

broadly rounded with very low rounded mid-dorsal longitudinal ridge, and
stronger lateral ridges.

Basal antennular segment equal to two distal segments together; flagellum
of five articles. Antenna with segment 5 slightly longer than 4; flagellum of six
articles. Mandibular palp three-segmented; molar low, rounded. Maxilliped
five-segmented, terminal segment short, set obliquely on penultimate segment;
latter with row of short medial setae; endite lacking. Pereopod 1 unguis more
than half length of remainder of dactylus; propodus expanded, palm with low
convex lobe proximally bearing setae, inner palmar margin with numerous
simple setae. Pereopods 2-3 ambulatory; pereopods 4~7 with rectangular carpi;
propodi, carpi, and meri with elongate setae on posterior margins. Pleopod 1
exopod operculiform, bearing numerous plumose setae distally, with strong
groove on outer surface close to median line. Uropodal exopod just reaching
distal margin of basis, apically narrowly rounded, with distinct sinuosity
distally; endopod triangular, distally rounded.

Type material

Syntypes, SAM-A5962, 2 non-ovig. 2, 16,0mm, off Saldanha Bay,
174 m.

Other material

SAM-A14131, 1 juv., off Saldanha Bay, 82 m. SAM—A14135, 1 juv., off
Saldanha Bay, 84m. SAM-A14141, 1 6, off Saldanha Bay 141m.
SAM-A14144, 1 6, off Saldanha Bay, 146 m. SAM-A14146, 1 d, 2 non-ovig.
2, off Saldanha Bay, 146 m. SAM-A14169, 1 non-ovig. 2, off Saldanha Bay,
79 m. SAM-A14171, 1 non-ovig. 2, off Saldanha Bay, 148 m. SAM-A14177,
1 6, off Saldanha Bay, 148 m. SAM-A14119, 1 non-ovig. °, False Bay, 26 m.
SAM-A14125, 1 juv., False Bay, 81 m. SAM—A14127, 1 3, False Bay, 87 m.
SAM-A14129, 2 non-ovig. 2, False Bay, 80m. SAM-A14130, 1 subd, 1
non-ovig. ¢°, False Bay. SAM-—-A14132, 1 non-ovig. 2, False Bay, 102 m.
SAM-A14134, 1 6, 1 non-ovig. 2°, False Bay. SAM—A14136, 1 6, 3 non-ovig.
2, False Bay, 82m. SAM-A14137, 2 non-ovig. 9, False Bay 71m.
SAM-A14138, 1 6d, False Bay. SAM-A14140, 1 non-ovig. ¢, False Bay.
SAM-A14142, 1 6, 1 non-ovig. 2, False Bay SAM—A14143, 4 non-ovig. @, 1
juv., False Bay, SAM—A14145, 2 6, False Bay. SAM—A14147, 1 d, False Bay.
SAM-A14149, 4 non-ovig. 2, 2 juvs, False Bay. SAM-A14150, 2 6, 4
non-ovig. 2, False Bay. SAM-A14151, 1 juv., False Bay. SAM-A14152, 1
non-ovig. 2, False Bay. SAM—A14167, 3 non-ovig. ¢, 1 juv., False Bay, 53 m.
SAM-A14168, 1 non-ovig. ¢, 1 juv., False Bay. SAM—A14170, 5 non-ovig. °,
False Bay. SAM-A14172, 1 juv., False Bay, 53 m. SAM-A14173, 1 d, False
Bay. SAM-A14174, 3 non-ovig. @, False Bay, 85m. SAM-A14178, 2
non-ovig. 2, False Bay, 87 m. SAM—A14179, 1 non-ovig. 2, False Bay, 73 m.
SAM-A14111, 1 non-ovig. @, Still Bay, 80 m. SAM-A14124, 1 non-ovig. 2,
Still Bay, 80 m. SAM-A14153, 1 d, 2 juvs, Agulhas Bank, 97 m. ZMC 51.68,

SOUTHERN AFRICAN ANTHURIDEA 133

C

Fig. 23. Malacanthura coronicauda. A. 2 cephalon, dorsal view. B. © cephalon, ventral view.
C. Maxilliped. D. Telson and uropods.

Th. Mortensen Expedition, 1 6, False Bay. ZMC, Galathea Expedition,
station 165, 1 6, 1 juv., False Bay.

Distribution
Saldanha Bay to Agulhas Bank, 26-174 m.

Malacanthura foveolata (Barnard, 1940)
Figs 24-25

Haliophasma foveolata Barnard, 1940: 384, 490, 498, fig. 2; 1955: 50, fig. 24 a-c. Day, Field &
Penrith, 1970: 47. Poore, 1975: 532. Kensley, 1975a: 38; 1978a: 49. Wagele, 1981: 86.

134 ANNALS OF THE SOUTH AFRICAN MUSEUM

50 0088

6.
Oops 26. 9-OP:2

Fig. 24. Malacanthura foveolata. A. 2 dorsal view. B. Cephalon, lateral view. C. ¢
telson. D. Pereopod 1 6. E. Pereopod 7. F. Maxilliped. G. Maxilla. H. Mandible.
I. Pereopod 1 ¢. Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA 135

Diagnosis

Body strongly indurate, with numerous small scattered pits. Telson in
female constricted at midlength, with rounded mid-dorsal longitudinal and
lower lateral ridges, several pits between ridges. Telson in male similar in
outline to female, but ridges less marked, pits lacking. Pereopod 1 in female
with propodal palm somewhat sinuous, setose; in male, propodal palm deeply
concave, with distal broadly triangular process; band of simple setae on inner
surface near palmar margin.

Type material

Holotype, SAM-A8280, 1 non-ovig. ¢, 12,1mm, Port Elizabeth,
intertidal.

Other material

SAM-A14158, 8 juvs, Saldanha Bay, 5 m. SAM-A14162, 1 non-ovig. &,
Saldanha Bay, 24 m. SAM-—A14180, 1 6, Saldanha Bay, 35 m. SAM-—A14322,
1 non-ovig. 2, off Saldanha Bay. SAM—A14397, 2 non-ovig. ¢, off Saldanha
Bay, 79 m. SAM-A14849, 1 subd, Langebaan Lagoon. SAM-A14155, 1 6,
False Bay. SAM-A14159, 1 non-ovig. 2, False Bay, 31 m. SAM-A14160, 1
non-ovig. 2, False Bay, 73m. SAM-A14161, 3 juvs, False Bay. SAM-
A14163, 1 non-ovig. 2, False Bay, 5m. SAM-A14164, 1 subd, False Bay,
40 m. SAM-A14165, 1 non-ovig. 2, False Bay. SAM—A14166, 1 non-ovig. 2,
False Bay, 22 m. SAM-A14846, 1 non-ovig. 2, False Bay. SAM-A14950, 1
non-ovig. 2, False Bay. SAM—A14156, 1 non-ovig. 2, Agulhas Bank, 27 m.
SAM-A14157, 1 non-ovig. 2, off Still Bay, 120 m.

Fig. 25. Malacanthura foveolata. A. Cephalon ¢ dorsal view. B. Telson d.

136 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 26. Malacanthura hermani. A. 2 dorsal view. B. Antennule. C. Antenna.
D. Mandible (distal palp segment missing). E. Pereopod 1. F. Maxilla. G. Maxilliped.
H. Pereopod 7. I. Pleon lateral view. Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA 137

Distribution
Saldanha Bay to Port Elizabeth, intertidal to 120 m.

Remarks

There is a marked loss of integumental pits as specimens increase in size,
especially in the transition from female to submale to male.

Malacanthura hermani (Barnard, 1940)
Fig. 26

Haliophasma hermani Barnard, 1940: 383, 490, 498, fig. 1; 1955: 50. Poore, 1975: 532.
Kensley, 1978a: 49, fig. 22A.

Diagnosis

Mid-dorsal pits on pereonites 1-6. Telson distally broadly rounded,
dorsally concave, with strong mid-dorsal longitudinal carina, and strong
midventral carina. Pereopod 1 propodal palm almost straight, sparsely setose.
Posterior margins of uropodal rami finely denticulate. Male unknown.

Type material
Holotype, SAM-A8076, non-ovig. 2, 20,0 mm, Hermanus, Cape.

Distribution
Known from type locality only.

Remarks

The single known specimen was taken from a cavity in the base of an
Allopora sp. coral.

Malacanthura linguicauda (Barnard, 1920)
Fig. 27

Anthura linguicauda Barnard, 1920: 338.

Malacanthura linguicauda: Barnard, 1925a: 133 (partim); 1940: 497 (partim). Nierstrasz,
1941: 240 (partim). Kensley, 1978a: 52, fig. 23A.

Non Malacanthura linguicauda: Barnard, 1925b: 388.

Diagnosis

Integument hardly indurate. Pereonites 4-6 each with mid-dorsal elongate
pit. Pleonites 1-5 fused, lines of fusion indicated by faint lateral grooves;
pleonite 6 free. Telson oval in outline, distally rounded, dorsally very gently
convex. Maxilliped with short endite. Uropodal exopod with sinuous outer
distal margin.

Description

Male. Integument hardly indurate. Proportions: C<1=2>3<4=6>7.
Cephalon with large dorsolateral eyes. Anterior margin of pereonites 2 and 3
with rectangular depression for articulation. Slit-like mid-dorsal pit present on

138

aT

SE

=

ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 27. Malacanthura linguicauda. A. ¢ dorsal view. B. 6 lateral view. C. Pereopod 1
3. D. Pereopod 2. E. Antenna. F. Pereopod 7. G. Maxilliped. H. Uropod. I. Pleopod
1. J. Pleopod 2 ¢ endopod. Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA 139

pereonites 4-6. Pleonites 1-5 fused, pleonites indicated ventrolaterally by short
slits; pleonite 6 free, with small mid-dorsal notch in posterior margin. Telson
elongate-oval, posteriorly rounded, dorsally gently convex, with two proximal
statocysts.

Antennular peduncle three-segmented, basal segment broadest; flagellum
of twenty to twenty-two articles bearing whorls of aesthetascs, reaching back to
pereonite 3. Antennal peduncle five-segmented, segment 2 broadest, grooved
to accommodate antennule; flagellum of four articles. Maxilliped five-
segmented, terminal segment set obliquely on segment 4, with five setae on
medial margin; segment 4 with few setae on medial margin; endite on inner
surface reaching base of segment 4, triangular, with single distal seta. Pereopod
1 unguis one-third length of rest of dactylus; dactylus with small posterior lobe
on inner margin; propodal palm convex, with distal notch and row of six to
eight setae. Pereopod 2 unguis one-fourth length of rest of dactylus, with two
small basal spines; propodus elongate-rectangular, with several setae on
posterior margin; carpus short, triangular. Pereopods 4-7 with posterodistal
spine on carpus and propodus; carpus half length of, and not underriding
propodus. Pleopod 1 exopod operculiform; endopod half width and
three-fourths length of exopod; basis with eight retinaculae. Pleopod 2 endopod
with simple rod-like copulatory stylet on medial margin not reaching distal end
of ramus; basis with four retinaculae. Uropodal exopod reaching to distal
margin of basis, outer margin distally sinuous, bearing numerous plumose
setae; endopod oval, bearing numerous plumose setae.

Type material

Holotype, SAM-A4172, 6, 11,0 mm, off Umhlangakulu River, Natal,
100 m.

Distribution
Off Natal, 100 m.

Remarks

The specimen recorded by Barnard (19256: 388) from off the Cape
Peninsula, bears four short proximal lobes on the inner margin of the dactylus
of pereopod 1, as well as a strong posteriorly-directed spine-like process on the
basal antennular segment. Neither of these features is present in‘the holotype
of M. linguicauda. As this specimen is a submale, and shows these differences,
it cannot be considered the same species.

The holotype was dissected, probably by Barnard, and the mandible and
maxilla have not been located.

Malacanthura ornata (Barnard, 1957)
Fig. 28
Haliophasma ornatum Barnard, 1957: 3, fig. 2. Poore, 1975: 531. Kensley, 1978a: 50, fig. 22B.

140 ANNALS OF THE SOUTH AFRICAN MUSEUM

a

——_

a

.
id
f

sale

Fig. 28. Malacanthura ornata. A. @ dorsal view. B. Antenna. C. Antennule.
D. Pereopod 1. E. Pereopod 2. F. Pereopod 7. G. Maxilliped. H. Mandible. I. Pleopod 1.
J. Uropod. Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA 141

Diagnosis

Body lacking sculpture. Telson dorsally gently convex, lacking mid-dorsal
ridge, with slight lateral flange, distally broadly rounded. Pereopod 1, unguis
only a little shorter than rest of dactylus; propodus broadly expanded, palm
with low proximal convexity, bearing few setae and spines. Maxilliped with
small bisetose endite. Male unknown.

Type material

Holotype, SAM-A10599, non-ovig. 2 (with eggs in body cavity),
10,0 mm, Mouille Point, Table Bay, intertidal.

Other material
SAM-A14181, 1 non-ovig. 2, 9,5 mm, off Cape Point.

Distribution
Table Bay, west coast of Cape Peninsula, intertidal.

Malacanthura pseudocarinata (Barnard, 1940)
Fig. 29

Haliophasma pseudocarinata Barnard, 1940: 385, 490, 498, fig. 3a—c; 1955: 5, 50. Day, Field &
Penrith, 1970: 47. Poore, 1975: 531. Kensley, 1978a: 50, fig. 22C.

Diagnosis

Integument indurate, with small pits only noticeable on cephalon, pleon,
and first pleopods. Pleonite 6 fused with telson dorsally, fusion line visible.
Telson broad, with strong lateral ridge and barely noticeable mid-dorsal
longitudinal ridge. Pereopod 1 propodal palm faintly sinuous, denticulate; row
of finely fringed setae on inner propodal surface near palm in female. Male
unknown.

Type material
Holotype, SAM—A8281, non-ovig. °, 17,0 mm, Port Elizabeth, intertidal.

Other material

SAM-A14323, 1 juv., 9,1 mm, off Saldanha Bay. SAM-A14184, 1 juv.,
9,6mm, False Bay. SAM-A14154, 1 juv., 6,0 mm, Port Elizabeth. SAM-
A14182, 1 non-ovig. 2, 19,2 mm, locality unknown.

Distribution
Saldanha Bay to Port Elizabeth, intertidal to 4 m.

Remarks

The single specimen from off Saldanha Bay has numerous rhizocephalan
parasites attached to the ventral pereon.

142 ANNALS OF THE SOUTH AFRICAN MUSEUM

Jol

Fig. 29. Malacanthura pseudocarinata. A. ¢ dorsal view. B. Pereopod 1 inner view.

C. Pereopod 1 outer view. D. Pereopod 7. E. Maxilliped. F. Distal mandibular palp

segment. G. Uropodal exopod. H. Mandible. I. Uropodal basis and endopod. J. Pleopod
1 exopod. Scale in mm.

143

SOUTHERN AFRICAN ANTHURIDEA

fare)

“2@ e (e}
seogoRes ®

o 2% eee
226 0 2a y
R 2°. 68h rs

< ; ‘ible bo :

Hi ©°e
°

C. Antennule.
Ht; Eleopods 2) <G-

B. Antenna.

G. Mandible.

A. 2 dorsal view.
F. Maxilla
I. Pereopod 7. Scale in mm.

EB Bereopod™ 2:

Fig. 30. Malacanthura_ schotteae.

De Lercopod 1 od.

————S——EE

144 ANNALS OF THE SOUTH AFRICAN MUSEUM

Malacanthura schotteae sp. nov.
Figs 30-31

Diagnosis

Integument indurate, bearing numerous pits. Pereopod 1 propodus greatly
expanded proximally in female; dactylus with rounded scaled lobe at base of
unguis in female, absent in male. Telson narrow, widest at proximal third, with
strong rounded mid-dorsal longitudinal ridge; posterior margin broadly
rounded. Uropodal exopod with strong notch in outer margin.

Description

Non-ovigerous female. Integument indurate, with numerous small pits,
~each containing single sensory seta; pits less numerous on posterior pereonites.
Proportions: C<1=2=3<4>5>6>7. Cephalon with large eyes; well-
developed rounded rostrum. Pereonites 4-6 with large mid-dorsal slit-like pit.
Pleonites 1-5 fused, fusion indicated laterally by shallow grooves; pleonite 6
free, with mid-dorsal slit in posterior margin. Telson narrowed in posterior
half, posterior margin broadly rounded, with strong rounded longitudinal
mid-dorsal ridge; cuticular scales becoming more developed in posterior half;
margin with many finely fringed setae.

Antennule with three-segmented peduncle; flagellum of three articles,
article 2 longer than 1 and 3 together; distal article bearing two aesthetascs.
Antennal peduncle segment 2 strongly grooved; segment 5 longest; flagellum of
seven setose articles. Mandibular palp three-segmented, distal segment bearing
twelve fringed spines, subterminal spine longest; incisor of two cusps; lamina
dentata with five serrations; molar rounded. Maxilla with eight distal spines.
Maxilliped five-segmented, distal segment semicircular, fringed with setae.
Pereopod 1 strongly subchelate; unguis almost half length of rest of dactylus;
latter with rounded lobe on inner margin bearing fringed scales at base of unguis;
propodus expanded, elongate, pitted, with proximal rounded lobe of palm bearing
fringed scales. Pereopod 2 much less robust than pereopod 1. Pereopods 5—7 with
carpus roughly rectangular. Pleopod 1 exopod operculiform, outer surface pitted
and bearing longitudinal groove; fringed with elongate plumose setae.

Male. Body indurate, relatively more slender than female; telson not as
markedly narrowed posteriorly as in female. Eyes relatively larger than in
female. Antennule elongate, reaching posteriorly to end of pereonite 1, with
numerous whorls of filiform aesthetascs. Pereopod 1 not robust, dactylus and
propodus lacking scaled rounded lobe as in female; propodal inner surface with
numerous minutely fringed setae. Pleopod 2 endopod with slender rod-like
copulatory stylet not quite reaching distal end of ramus.

Type material

Holotype, SAM-—A17525, non-ovig. 2, 12,5 mm, south of East London,
90 m. Allotype, SAM—A17526, 1 3, 8,4 mm, south of East London, 90 m.

SOUTHERN AFRICAN ANTHURIDEA 145

Fig. 31. Malacanthura schotteae. A. Cephalon 2 dorsal view. B. Cephalon 2 ventral view.
C. Maxilliped. D. Pereopod 1. E. Dactylus pereopod 1. F. Ventral pleon.

146 ANNALS OF THE SOUTH AFRICAN MUSEUM

Paratype, SAM-A17527, 1 non-ovig. ¢, 6,5 mm, 10 juvs, south of East
London, 90 m. Paratype, USNM 189055, 1 non-ovig. 2, 7,6 mm, south of East
London, 90 m. Paratype, SAM—A17528, 1 non-ovig. 2, 8,3 mm, off Transkei,
150-200 m. Paratype, USNM 189056, 1 non-ovig. 2, 9,0 mm, off Transkei,
150-200 m.

Other material

SAM-A14094, 1 non-ovig. 2, 10,8mm, off Still Bay, 80m. SAM-—
A14080, 1 non-ovig. ¢, 11,5 mm, off Still Bay, 120m. SAM-—A14099,
1 juv., off Still Bay, 120m. SAM-A17259, 1 non-ovig. @, off Transkei,
710-775 m.

' Distribution
Still Bay to Transkei, 80-775 m.

Etymology

The species is named for Marilyn Schotte, of the Department of
Invertebrate Zoology, Smithsonian Institution, in appreciation of the many
illustrations she executed for this paper.

Remarks

Of the seven earlier-described species of Malacanthura, only M. foveolata
has an integument as densely pitted as in the two new species described here.
The telson of M. foveolata, however, is obscurely tri-ridged, with pits between
the ridges, while M. schotteae and M. transkei have a single mid-dorsal ridge
and no pits on the telson.

Malacanthura schotteae has a strongly notched uropodal exopod, a
posteriorly constricted telson, and scaled lobes on the dactylus and propodal
palm, while M. transkei has an unnotched oval-lanceolate uropodal exopod,
evenly convex telsonic margins, and lacks propodal and dactylar lobes in
pereopod 1. Malacanthura transkei also has mid-dorsal depressions on
pereonites 2 and 3, not seen in M. schotteae.

Malacanthura serenasinus (Kensley, 1975)
Figs 32-33
Agulanthura serenasinus Kensley, 1975a: 72, fig. 18; 1978: 46, fig. 20B-E.

Diagnosis

Body profile smooth; uropodal exopods closely adpressed to telson; latter
with gentle ventral curvature in posterior half, lanceolate, dorsally convex,
posterior margin narrowly rounded. Eyes small, weakly pigmented. Antennular

SOUTHERN AFRICAN ANTHURIDEA 147

Fig. 32. Malacanthura serenasinus. A. ¢ dorsal view. B. Pereonites 6 and 7, and pleon
lateral view. C. Antennule. D. Antenna. E. Pereopod 7. F. Pereopod 1. G. Uropodal
basis and endopod. H. Mandible. I. Maxilliped. Scale in mm.

148 ANNALS OF THE SOUTH AFRICAN MUSEUM

flagellum in female with two articles. Antennal flagellum of four articles.
Pereopod 1 propodus expanded, palm slightly concave in female, slightly
convex in male. Pereopods 4-7 with carpi and meri expanded, rectangular.
Maxilliped five-segmented with small endite.

Type material

Holotype, SAM-—A13553, 1 6, 11,0mm, False Bay, 62m. Allotype,
SAM-A13554, 1 non-ovig. °, 14,5mm, False Bay, 29m. Paratype,
SAM-A13621, 1 non-ovig. ¢, 13,0mm, Agulhas Bank, 22m. Paratype,
SAM-A13622, 1 non-ovig. ¢, 14,0mm, Saldanha Bay, 5m. Paratypes,
SAM-A13623, 1 35, 11,0 mm, 3 non-ovig. 2, 6,9-12,9 mm, off Still Bay, 20 m.

_ Other material

SAM-A17790, 6 non-ovig. 2, off Saldanha Bay, 79 m. SAM-A14029, 1
juv., False Bay, 56m. SAM-A14030, 1 non-ovig. 2, False Bay, 29 m.
SAM-A14031, 1 non-ovig. 2, False Bay, 26m. SAM-A14032, 1 non-ovig. @,
False Bay, 61m. SAM-—A14033, 1 non-ovig. 9, False Bay, 42m. SAM-—
A14034, 1 6, False Bay, 66 m. SAM-A14035, 1 non-ovig. 2, False Bay, 75 m.
SAM-A14036, 1 juv., False Bay, 48 m. SAM-A14037, 1 non-ovig. 2, Agulhas
Bank, 125m. SAM-A14038, 1 non-ovig. @, Agulhas Bank, 97 m.
SAM-A14062, 1 non-ovig. 2, Agulhas Bank, 36 m. SAM-A14039, 5 non-ovig.
2, off Still Bay, 15 m.

Distribution
Saldanha Bay to Still Bay, 5-125 m.

Remarks

When originally described as the type species of Agulanthura, the species
was placed in a new genus on the basis of the smooth body profile, the lack of
pereonal pits, and the expanded carpi and meri of the posterior pereopods.
With rediagnosis of the genera Haliophasma and Malacanthura this species is
now regarded as a specialized fossorial Malacanthura. The five-segmented
maxilliped with weak endite, the antennular and antennal structure in the
female, and the presence of small ventrolateral pits on the posterior pereonites
support this placement.

The smooth body profile, the reduced and weakly pigmented eyes, the
uropodal exopods closely adpressed to the telson, and the expanded carpi and
meri of the posterior pereopods could all be adaptations for a fossorial mode of
life. Almost all the specimens recorded above were taken from fairly fine
sediments by grab, another indication of the burrowing habit.

In the original description (Kensley 1975a: 72), the appendages figured
were those of a male. With examination of a female, it would seem that with
the change from female to male the mandible undergoes some reduction and
loss of function.

SOUTHERN AFRICAN ANTHURIDEA 149

Fig. 33. Malacanthura serenasinus. A. ° cephalon dorsal view. B. 2 cephalon ventral
view. C. Maxilliped inner view. D. Integumental scales on pereopod 1 propodus.
E. Ventrodistal spines on pereopod 2 propodus. F. Telson dorsal view.

150 ANNALS OF THE SOUTH AFRICAN MUSEUM

Malacanthura transkei sp. nov.
Figs 34-35

Diagnosis

Integument indurate, heavily pitted, Pereopod 1 propodus expanded.
Telson narrow, extending beyond uropodal endopod, widest at about midpoint,
posterior margin broadly rounded, finely denticulate; strong rounded mid-
dorsal longitudinal ridge present.

Description

Female. Integument indurate, heavily pitted dorsally and ventrally.
Proportions: C<1<2>3<4=5>6>7. Cephalon with rounded rostrum
extending as far as anterolateral lobes; shallow groove present between anterior
margin and well-developed eyes. Pereonite 1 with bilobed posterior margin;
pereonite 2 with shallow mid-dorsal area ending posteriorly in shallow circular
depression; pereonites 3-5 with mid-dorsal elongate pit in shallow depression.
Pleonites 1-5 fused, fusion lines indicated by dorsolateral grooves. Pleonite 6
free, posterior margin with short mid-dorsal notch. Telson widest at about
midpoint, lateral margins gently convex, posterior margin broadly rounded,
denticulate.

Basal antennular segment somewhat shorter than segments 2 and 3
together; flagellum three-articulate, second article more than twice length of
first or third; distal article bearing three aesthetascs. Antenna with numerous
setae on all segments; peduncle segment two strongly grooved; segment 5
longest; flagellum five-articulate. Mandibular palp three-segmented, terminal
segment with seven fringed spines; incisor of three cusps; lamina dentata with
four teeth; molar conical. Maxilla with six distal spines. Maxilliped
five-segmented, distal segment oval, set obliquely on segment 4, with numerous
fine setules; endite absent. Pereopod 1 unguis more than half length of
dactylus; propodus expanded proximally, palm gently concave, with five
submarginal spines; carpus triangular, distal rounded lobe bearing spine-scales,
Pereopod 2 unguis half length of dactylus, with strong auxiliary spine at base;
propodus rectangular, with strong sensory spine at posterodistal corner,
posterior margin bearing fringed scales; carpus triangular, with strong
posterodistal spine. Pleopod 1 exopod operculiform, subequal in length and
about three times width of endopod. Uropodal exopod elongate-oval, outer
margin dentate, setose, distally acute; endopod broadly oval, margin setose,
finely dentate.

Male. Antennule with ten-articulate flagellum. Pereopod 1 propodal
palm straight, with dense band of fringed spines on inner surface. Pleopod 2
endopod with slender distally expanded copulatory stylet reaching slightly
beyond ramus. Pleon, telson, and uropods relatively more elongate than in
female.

SOUTHERN AFRICAN ANTHURIDEA 11

Fig. 34. Malacanthura transkei. A. 2 dorsal view. B. Antennule. C. Antenna.
D. Mandible. E. Maxilliped. F. Maxilla. G. Pleopod 2 ¢. H. Uropodal endopod and
basis. I. Uropodal exopod. J. Pleopod 1. Scale in mm.

152 ANNALS OF THE SOUTH AFRICAN MUSEUM

c)

Fig. 35. Malacanthura transkei. A. Pereopod 1 °. B. Pereopod 1 6. C. Pereopod 2..
D. Pereopod 7.

Type material

Holotype, SAM-A17531, 1 6, 7,5 mm, off Transkei, 710-775 m.
Allotype, SAM-—A17532, 1 non-ovig. 2, 7,5 mm, off Transkei, 710-775 m.
Paratype, SAM—A17533, 1 non-ovig. 2, 6,1 mm, off Transkei, 710-775 m.
Paratype, USNM 184667, 1 non-ovig. 2, 6,3 mm, off Transkei, 710-775 m.

Etymology

The specific name derives from the area of the coastline of southern
Africa, off which the species was collected.

Remarks

See discussion at end of M. schotteae discussion.

SOUTHERN AFRICAN ANTHURIDEA 153

Mesanthura Barnard, 1914

Diagnosis

Eyes present. Antennular flagellum of two articles; antennal flagellum of
three articles. Mandibular palp three-segmented. Maxilliped five-segmented,
endite rudimentary or absent. Pereopod 1 expanded, subchelate, sometimes
highly modified in male. Pereopods 2-3 ambulatory; pereopods 4~7 with
triangular carpus underriding propodus. Pleopod 1 exopod operculiform.
Pleonites 1—5 fused, 6 free. Telson with two basal statocysts. Dorsal integument
with chromatophores in species-specific pattern.

Type species
Anthura catenula Stimpson, 1855.

KEY TO THE SOUTH AFRICAN SPECIES OF MESANTHURA

1. Dorsal pigment in solid triangle on cephalon, posterior bars on pereonites 4-7 .. dimorpha
— Dorsal pigment in open rings or rectangles on cephalon and all pereonites ....... catenula

Mesanthura catenula (Stimpson, 1855)
Figs 36-37

Anthura catenula Stimpson, 1855: 393. Beddard, 1886: 143. Stebbing, 1910: 420.

Mesanthura catenula: Barnard, 1914: 343a, pl. 29A; 1925a: 143, fig. 9a; 1940: 490; 1955: 5.
Nierstrasz, 1941: 242. Day, Field & Penrith, 1970: 47. Kensley, 1978a: 52, fig. 23B—C;
1978b: 1.

Diagnosis

Pigment pattern: hollow rectangular patches on cephalon and pereonites
1-7; broad transverse band on pleon; dense oval patches on uropodal exopod,
endopod, and telson. Pereopod 1 propodus expanded, palm sinuous, with
hyaline margin. Uropodal exopod with distinct distal notch.

Type material

Stimpson’s type from Simon’s Bay in False Bay was apparently lost,
perhaps in the Chicago fire of 1871.

Other material

SAM-A2719, 1 juv., Mouille Point, Table Bay, intertidal. SAM-8825, 1
non-ovig. °, 3 juvs, St. James, False Bay. SAM-A250, 2 non-ovig. &,
10,5—11,8 mm, Kalk Bay in False Bay, intertidal. SAM—A2106, 1 non-ovig. 2,
St. James, False Bay; SAM—A3302, 1 juv., 8,0 mm, Buffels Bay in False Bay.
SAM-A14350, 1 ovig. 2, 14,0 mm, False Bay, 4m. SAM-A4171, 1 ovig. °,
18,4 mm, East London, intertidal. SAM—A14352, 2 non-ovig. 2, 14,5 mm,
locality unknown.

Distribution
Table Bay to East London, intertidal to 4 m.

154 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 36. Mesanthura catenula. A. ¢ dorsal view. B. Antennule. C. Antenna.
D. Maxilliped. E. Pereopod 7. F. Pereopod 1. G. Mandible. H. Uropodal exopod.
I. Uropodal basis and endopod. Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA 155

Fig. 37. Mesanthura catenula. A. Spine cluster at dactylar-propodal articulation of pereopod 1.
B. Pereopod 2 propodal scales.

Remarks

In some of the specimens listed above, which were collected in the early
part of this century, the colour pattern has faded, but in material collected since
1932 the pigment persists. Some variation in the pattern has been noted e.g.,
occasionally there is a simple broad bar across the cephalon posterior to the
eyes, while pereonite 1 sometimes has a solid patch of pigment. In spite of
these variations the overall pattern makes the species unmistakable, and it may
easily be separated from M. dimorpha by the structure of pereopod 1 in the
male and female.

Mesanthura dimorpha sp. nov.
Figs 38-39

Diagnosis

Pigment pattern: triangle on cephalon; ring on pereonites 1-3; posterior
bar on pereonites 4~7; bar on pleonite 6; oval on telson. Cephalon in male with
midventral lamellar process. Pereopod 1 in female, propodal palm with low
tooth at midpoint, carpus distally acute; pereopod 1 in male, propodal palm
concave, carpus distally strongly produced into bilobed process.

Description

Female. Integument moderately indurate. Body proportions:
C<1=2=3<4=5>6>7. Cephalon with low rostral point; eyes dorso-
lateral. Fused pleonites 1-5 subequal in length to pereonite 7; pleonite 6 free,
with tiny mid-dorsal notch in posterior margin. Telson posteriorly evenly

156 ANNALS OF THE SOUTH AFRICAN MUSEUM

I Ne

Fig. 38. Mesanthura dimorpha. A. 2 dorsal view. B. Antennule. C. Antenna.
D. Pereopod 1 &. E. Pereopod 2 ¢. F. Pleopod 1. G. Maxilla. H. Maxilliped.
I. Mandible. J. Uropodal exopod. Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA 157

rounded, central pigmented part dorsally gently convex. Antennule with basal
peduncular segment equal in length to, but broader than two distal segments;
flagellum of two articles. Antennal peduncle with second segment grooved to
accommodate antennule; flagellum of one article. Mandibular palp
three-segmented, segment 2 twice length of basal segment, with two simple
distal setae; terminal segment with four distal spines; incisor of three cusps;
lamina dentata of eight serrations. Maxilla with seven distal spines. Maxilliped
five-segmented, lacking endite; subterminal segment with two short setae on
median margin; terminal segment with five setae on median margin. Pereopod
1 subchelate, expanded; unguis about one-third length of dactylus; propodal
palm with blunt tooth at midpoint; carpus triangular, with corneous triangular
extension distally. Pereopods 2—3 ambulatory, with strong sensory spine at
posterodistal corner of propodus. Pereopods 4~7 with short roughly triangular
carpus. Pleopod 1 exopod operculiform; endopod less than half width and
shorter than exopod; basis with four retinaculae. Uropodal exopod with deep
notch in distal margin; endopod not reaching telsonic apex.

Male. Antennule with multiarticulate flagellum bearing filiform
aesthetascs. Midventral keel of cephalon expanded into slightly hollowed
platelike process. Pereopod 1 strongly indurate; unguis almost half length of
dactylus; propodus with concave palm bearing distal keeled tooth; inner surface
with single row of spines and strong proximal tooth; carpus triangular, distally
produced well beyond propodal base into narrowly rounded bilobed process.
Pereopods 2-3 with dactylus meeting distal extension of triangular carpus.
Pleopod 2 endopod with simple rod-like copulatory stylet not reaching apex of
ramus.

Colour pattern

Similar in male and female. Cephalon with triangular patch of pigment
with base between eyes. Pereonite 1 with narrow hoop of pigment in anterior
half. Pereonites 2—3 with hollow rectangle of pigment. Pereonites 4-7 with
transverse band of pigment near posterior margin. Pleonite 6 with transverse
band. Telson with oval reticulation of pigment.

Type material

Holotype, SAM-A17534, 6, 6,3 mm. Allotype, non-ovig. 2, 5,9 mm.
Paratypes, 2 non-ovig. 2, 4,8-5,4mm, 2 juvs, off East London, 90 m.
Paratype, SAM-A14347, 6, 9,0 mm, off Natal. Paratype, SAM-—A14354, 6,
7,5 mm, off Port Elizabeth, 84m. Paratypes, USNM 184666, 3, 6,0 mm, 1
non-ovig. 2, 5,0 mm, off East London, 90 m.

Distribution
Natal to Port Elizabeth, 84—90 m.

158 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 39. Mesanthura dimorpha. A. Cephalon <6 lateral view. B. Pereopod 2 4a.
C. Pereopod 7 ¢. D. Pereopod 1 ¢ imner surface. E. Pereopod 1 6 outer surface.
F. Pleopod 2 6.

Etymology

The specific name derives from the dimorphic nature of the first pereonite
and first pereopods.

SOUTHERN AFRICAN ANTHURIDEA 159

Remarks

Of the described species of Mesanthura, the present species resembles only
M. ocellata Barnard, 1925a, from Thailand, especially in having rings of
pigment on the anterior pereonites. Barnard’s species, however, also has rings
on the posterior pereonites. Examination of Barnard’s type material reveals
further differences. The proportions of the mandibular palp segments, the
pereopods 1 of the female, and the uropods all show differences from the
present species.

Neohyssura Amar, 1952

Diagnosis

Eyes present or absent. Antennular flagellum of three to five articles;
antennal flagellum of seven articles. Mandibular palp three-segmented.
Maxilliped seven-segmented; endite present. Pereopods 1-3 similar, sub-
chelate; pereopods 4~7 with triangular carpus underriding propodus. Pleonites
1-5 free, 6 fused with telson. Pleopod 1 non-operculiform. Telson indurate,
spiniform.

Type species
Hyssura spinicauda Walker, 1901.

Neohyssura skolops Kensley, 1978
Figs 40-41
Neohyssura skolops Kensley, 1978b: 9, figs 5-6.

Diagnosis

Telson terete, spike-like. Uropodal exopod dentate on mesial margin.
Maxilliped with thin-walled endite. Eyes present.

Type material
Holotype, SAM-—A15651, 1 non-ovig. 2, 5,8 mm, off Natal, 850 m.

Other material
SAM-—A17535, 2 juvs, 3,1 mm, off East London, 90 m.

Distribution
East London to Natal, 90-850 m.

Panathura Barnard, 1925a

Diagnosis
Eyes present. Antennular flagellum of two to five articles; antennal
flagellum of two to three articles. Mandible with three-segmented palp.

160 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 40. Neohyssura skolops.
? dorsal view. Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA 161

Fig. 41. Neohyssura skolops. A. ¢ cephalon dorsal view. B. Mouthparts and pereopod 1
palm. C. Pereopod 1 propodus and dactylus. D. Pereopod 1 propodal palm spine and
lobe. E. Pereopod 1 apex of carpus. F. Pleon and telson.

162 ANNALS OF THE SOUTH AFRICAN MUSEUM

Maxilliped six- (seven) segmented; endite well developed, distally rounded or
acute. Pereopods 1-3 subchelate, subsimilar. Pereopods 4-7 with triangular
carpus underriding propodus. Pleopod 1 exopod operculiform. Pleonites 1-6
free, short. Telson lacking statocysts.

Type species

Apanthura serricauda Barnard, 1920.

KEY TO THE SOUTH AFRICAN SPECIES OF PANATHURA

1. Pereopod 1 propodal palm with strong triangular proximal tooth; telson distally narrowly

TOURCE: Jin ecg nats 25 Wa see Pe eet Ree mares ee te ent amstelodami
— Pereopod 1 propodal palm with several low rounded teeth; telson distally broadly and
» Gevenly rounded). kot gaelsaiee cs a rseiniran er algr ete tenetacte aut ok Airs tv eo ee serricauda

Panathura amstelodami Kensley, 1976
Fig. 42

Panathura amstelodami Kensley, 1976: 277, figs 4-5; 1977: 239; 1978a: 54, fig. 23D; 1980: 32.
Expanathura amstelodami: Wagele, 1981: 121.

Diagnosis

Telson and uropods not indurate; telson posteriorly tapering to broadly
rounded apex, margins finely serrate. Maxillipedal endite distally narrowed.
Lamina dentata of mandible with finely denticulate part; molar reduced.
Pereopod 1 propodal palm with single strong proximal tooth, followed by two
smaller teeth.

Type material

Holotype, Muséum National d’Histoire Naturelle, Paris, IS.1002, 1 6,
5,0 mm, Amsterdam Island. Paratypes, M.N.H.N., Paris, IS.1003, 2 non-ovig.
2, 3,2-3,8 mm, Amsterdam Island, upper infralittoral to 80 m. Paratypes,
SAM-A14994, 2 3, 3,0-3,2 mm, 1 non-ovig. 2, 4,4 mm, Amsterdam Island.

Other material

SAM-A14356, 1 ovig. 2, 4,9 mm, off Natal. SAM—A17536, 1 ¢, 3,1 mm.
1 ovig. 2, 2,9 mm, 1 non-ovig. 2, 3,0 mm, Walter’s Shoal, 33°13’S 45°51’E,
38-46 m. USNM 171731, 11 6, 6 ovig. 2, 31 non-ovig. 2, south of Beira,
Mozambique, 62 m. USNM 171732, 1 ovig. 2, 4 non-ovig 2, 1 juv., south
coast of Madagascar, 46 m. USNM 171733, 1 6, 3 ovig. 2, 8 non-ovig. 9, 6
Juvs, south coast of Madagascar, 38 m.

Distribution

St. Paul and Amsterdam Islands, southern Indian Ocean; Walter’s Shoal,
south-western Indian Ocean; Natal to Mozambique and Madagascar, upper
infratidal to 80 m.

SOUTHERN AFRICAN ANTHURIDEA 163

i

Fig. 42. Panathura amstelodami. A. 2 dorsal view. B. Pereopod 1. C. Pereopod 2.
D. Antennule dé. E. Mandible. F. Maxilliped. Scale in mm.

164 ANNALS OF THE SOUTH AFRICAN MUSEUM

Panathura serricauda (Barnard, 1920)
Figs 43-44

Apanthura serricauda Barnard, 1920: 339, pl. 15 (figs 11-12).

Panathura serricauda: Barnard, 1925a: 143; 1940: 490; 1955: 5. Nierstrasz, 1941: 241. Penrith
& Kensley, 1970: 228, Day, Field & Penrith, 1970: 47. Kensley, 1976: 264, fig. 6;
1978a: 54, fig. 23E-—G. Wagele, 1981: 118.

Panathura serricaudata (sic): Day, 1969: 78.

Diagnosis

Uropods and telson indurate. Telson distally evenly rounded, margins
serrate. Maxillipedal endite distally obtuse. Pereopod 1 propodal palm bearing
several blunt, almost rectangular teeth.

Type material

Syntypes, SAM-—A2620, 1 non-ovig. ¢, damaged, Mouille Point, Table
Bay, intertidal. Syntypes, SAM—A2698, 2 non-ovig. 2, Mouille Point, Table
Bay, intertidal. Syntypes, SAM—-A2692, 1 non-ovig. 2, St. James, False Bay,
intertidal.

Other material

SAM-A12741, 2 ovig. 2, 4,5 mm, 1 non-ovig. 2, 4,5 mm, 1 juv.,
Liideritz, intertidal. SAM-—A14355, 1 ovig. 2, 5,0 mm, 1 non-ovig. 2, 3,9 mm,
Langebaan Lagoon. SAM-—A14359, 3 ovig. 9%, 4,0-5,0mm, 3 juvs, 15
non-ovig. 2, Schaapen Island, Langebaan. SAM-—A14358, 2 non-ovig. &,
False Bay, 42m. SAM-—A14357, 1 non-ovig. ¢, Agulhas Bank, 49 m.
SAM-A17550, 7 ovig. ¢, 3,0-4,0 mm, 9 non-ovig. 2, 2,5-3,2 mm, south of
East London, 90 m. SAM-—A15511, 2 non-ovig. 2, St. Paul Island.

Distribution

Liideritz to East London, intertidal to 90 m; St. Paul Island, southern
Indian Ocean.

Remarks

Panathura serricauda is most frequently collected from the holdfasts of the
kelp Ecklonia maxima.

Quantanthura Menzies & George, 1972

Diagnosis

Eyes present or absent. Antennular flagellum of five to seven articles;
antennal flagellum of four to nine articles. Mandibular palp three-segmented.
Maxilliped six-segmented; endite well developed. Pereopod 1 subchelate,
propodus expanded; pereopods 2-3 smaller than pereopod 1; pereopods 4~7
with rectangular carpus. Pleopod 1 exopod operculiform. Pleonites 1—5 fused,
pleonite 6 free. Telson with two basal statocysts.

165

SOUTHERN AFRICAN ANTHURIDEA

= yy

= J ala \.,
facl aly

\
: \
f

A. 2 dorsal view. B. Pereopod 1. C. Pereopod 2.

Fig. 43. Panathura serricauda.
D. Maxilliped. E. Mandible. Scale in mm.

166 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 44. Panathura serricauda. A. Maxilliped. B. Pleon dorsal view.

Type species
Quantanthura globitelson Menzies & George, 1972.

Quantanthura remipes (Barnard, 1914)
Figs 45—46

Anthelura remipes Barnard, 1914: 338a, pl. 28B; 1925a: 135; 1940: 490. Nierstrasz, 1941: 240.
Kensley, 1977: 239; 1978a: 46, fig. 20F.
Non Anthelura remipes Kensley, 19785: 1.

Diagnosis

Eyes absent. Antennular flagellum of five articles. Antennal flagellum of
four to five articles, reduced. Maxilliped with broad, distally rounded endite.
Pereopod 1 propodus expanded, palm convex, with groove on inner surface.
Telson elliptical, posteriorly narrowly rounded, dorsally convex. Uropodal
exopod ovoid, outer margin sinuous.

Type material
Holotype, SAM-A58, 1 non-ovig. ¢, 28,1 mm, (damaged), off Cape
Peninsula, 312 m.

Other material |
SAM-A14402, 1 non-ovig. ¢, 10,2 mm, Agulhas Bank, 78m. SAM-—
A14993, 1 non-ovig. 2, 15,9 mm, Lambert’s Bay.

Distribution
Lambert’s Bay to Agulhas Bank, 78-312 m.

SOUTHERN AFRICAN ANTHURIDEA 167

Fig. 45. Quantanthura remipes. A. Holotype @ dorsal view. B. Pleon ¢ dorsal view.
C. Pleon lateral view. D. Antenna. E. Antennule. F. Mandible. G. Maxilla.
H. Maxilliped. Scale in mm.

168 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 46. Quantanthura remipes. A. Pereopod 1. B. Pereopod 2. C. Pereopod 7.
D. Pereopod 7 dactylus and propodus outer surface. E. Pereopod 7 dactylus and propodus
inner view.

SOUTHERN AFRICAN ANTHURIDEA 169

Remarks

Since the revision of the diagnosis of Anthelura (Kensley 1978d: 787) and
Quantanthura (Kensley and Koening, 1979: 953), there can be little doubt that
the present species is a member of the latter genus.

KEY TO SOUTH AFRICAN GENERA OF THE FAMILY PARANTHURIDAE

1. Antennular and antennal flagella of more than 10 articles................... Accalathura
— Antennular and antennal flagella with fewer than 10 articles ...................... 2
PEICREMMOUMBADSCI Grr in is es ence Bale wig ers av a a yep aR Re Re ee Colanthura
= PeCeODOd T (CRISIS Sak enn Oe Soren rae, sn Rae ahaa A oenr airmen of Hc 3)
seropodalicxopod reduced, endopod and basis fused ..........2.-....+---+- Pseudanthura
=LOnodalcexopod not reduced, endopod free’... :... 25.2.5 2e eee be deen eee: +
Aaeevesrapsent: pereopod 7 with triangular carpus ...........5.....0.000e sere Leptanthura
SEeevesmncsent- percopod / with rectangular carpus .....-...... 22-6525 e eee oe 8: Paranthura
Family Paranthuridae
Accalathura Barnard, 1925a
Diagnosis

Eyes present. Flagella of antennule and antenna multiarticulate. Mandi-
bular palp three-segmented. Maxillipedal endite reaching almost to end of palp;
latter of two segments. Pereopod 1 subchelate, propodus expanded; pereopods
2-3 ambulatory; pereopods 4~7 with rectangular carpus. Telson with or without
statocyst.

Type species
Calathura crenulata Richardson, 1905.

Accalathura indica (Nierstrasz, 1941)
Fig. 47
Metanthura indica Nierstrasz, 1941: 247, figs 15-24.
Accalathura indica: Kensley, 1977: 250, fig. 8; 1978a: 44, fig. 20A. Poore, 1980: 59.
Diagnosis
Antennular and antennal flagella of about twenty-eight and thirty-five
articles respectively. Eyes well developed. Pleonite 6 longest, bilobed. Telson

elongate-lanceolate, with single basal statocyst. Uropodal exopod elongate-
oval.

Type material

Whereabouts unknown.

Other material

SAM-A15348, 1 ovig. 2, off Mozambique, 100 m. SAM-—A17577, 1 ovig.
2, off Natal, 90 m. SAM-A17578, 1 6, off Sodwana Reef, Zululand, 17 m.

170 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 47. Accalathura indica. 2 dorsal view.
Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA WAL

Distribution
Natal, Mozambique, to Java Sea, 17-100 m.

Accalathura laevitelson (Kensley, 1975)
Fig. 48

Katanthura laevitelson Kensley, 1975a: 69, fig. 17; 1978a: 50, fig. 22H-I.
Zulanthura laevitelson: Poore, 1980: 65.

Diagnosis

Eyes present. Pleonites 1-6 free. Maxilliped with strong endite, palp
two-segmented. Pereopods 2-3 smaller than pereopod 1; pereopods 4-6 with
rectangular carpus not underriding propodus. Single telsonic statocyst present.

Type material
Holotype, SAM—A13552, 1 manca, 6,4 mm, off Still Bay, 30 m.

Distribution
Off Still Bay, 30 m.

Remarks

G. Poore (1981 pers. comm.) has examined this specimen, and is
convinced that it is a manca of an Accalathura species, almost certainly not A.
indica. Because the single specimen is immature, a key to the two species of
southern African Accalathura is not included.

Colanthura Richardson, 1902

Diagnosis

Eyes present or absent. Mandibular palp absent. Antennular and antennal
flagella with few short articles. Maxilliped lacking endite or palp segments.
Pereopod 1 subchelate, propodus expanded; pereopods 4-6 with rectangular
carpi. Pereonite 7 very short, lacking pereopod. Pleonites 1-6 short, free.
Pleopod 1 exopod operculiform. Telson lacking statocyst.

Type species
Colanthura tenuis Richardson, 1902.

Colanthura uncinata Kensley, 1978
Figs 49-50
Colanthura uncinata Kensley, 1978b: 12, figs 7-8.

eZ ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 48. Accalathura laevitelson. Holotype
dorsal view. Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA 173

Diagnosis

Telson posteriorly evenly rounded, with low mid-dorsal rounded ridge.
Pleopod 2 of male, apex of copulatory stylet with single recurved hook.
Pereopod 1 male with fourteen fringed spines on inner propodal surface;
female with six spines. Uropodal exopod narrowly oval, endopod more broadly
oval.

Fig. 49. Colanthura uncinata. A. 3 dorsal view. B. Antennule ¢. C. Antennule 9°.
D. Antenna. E. Maxilliped. F. Pereopod 1 6. G. Uropodal exopod. H. Uropodal basis
and endopod. Scale in mm.

174 ANNALS OF THE SOUTH AFRICAN MUSEUM

Type material

Holotype, SAM-A15652, 1 3, 3,9 mm. Allotype, SAM—A15652, 1 ovig.
2, 4,5 mm, off Natal, 28°31’S 32°34’E, 680 m. Paratypes, SAM-A15653, 2 6,
3,4mm, 1 ovig. ?, 4,0 mm, off Natal, 27°59’S 32°40’E, 550 m. Paratypes,
USNM 170544, 2 3, 3,9 mm, 1 ovig. 2, 4,5 mm, off Natal, 28°31’S 32°34’E,
680 m.

Fig. 50. Colanthura uncinata. A. Cephalon dorsal view. B. Pleon dorsal view.

Other material

SAM-A17581, 1 6, off Transkei, 32°28’S 28°58’E, 710-775 m.
SAM-A17582, 1 6, 2 ovig. 2, off Transkei, 31°59’S 29°22'E, 150-200 m.
SAM-A17579, 1 6, 2 subd, 2 non-ovig. 2, 4 ovig. &, off Natal, 27°59’S
32°40’E, 550 m. SAM-A17580, 6 6, 3 subd, 5 ovig. 2, 4 juvs, off Natal,
28°31'S 32°34’E, 680 m.

Distribution
Northern Natal to Transkei, 150-775 m.

Leptanthura Sars, 1899

Diagnosis

Eyes absent. Flagella of antennule and antenna reduced. Mandibular palp
three-segmented. Maxillipedal endite rudimentary or absent; palp of one to
three articles, with one to two terminal spines only. Pereopod 1 subchelate;
pereopods 2-3 similar to but smaller than pereopod 1; pereopods 4~7 with

SOUTHERN AFRICAN ANTHURIDEA 175

triangular carpus lacking free anterior margin. Pleonites free. Pleopod 1
exopod operculiform. Uropodal exopod often broadly oval. Telson with single
statocyst.

Type species
Anthura tenuis Sars, 1873.

KEY TO THE SOUTH AFRICAN SPECIES OF LEPTANTHURA

Pe roOpodalexopod margin Serrate ©. 0:60. 2 eo ee ee ce eee ee eees urospinosa
Sire et AmCEOPOd Maron CMTE <2. 2. Fi. ee bees wes Cees be ee ewok Tee ce ee

meeecanipostcriony broadly rounded... :..-...0... 220.08 b eet eee ee meee ese ees 3
SE isensmastenoriy tapered, apically acute .. 22 42....0.5e. enced seta weee eee agulhasensis
3. Uropodal exopod with strong notch in distal margin .....................2.05. laevigata
Cane ANeXOPOC UNNOtCHE . 2.2 cb sows Se ees oeka sos Atnes alg ue ee eant Meas +
4. Uropodal endopod triangular; midventral processes absent from pereon........... minuta
— Uropodal endopod narrow; midventral processes present on some pereonites ... natalensis

Leptanthura agulhasensis Kensley, 1975
Big: 51
Leptanthura agulhasensis Kensley, 1975a: 64, figs 14-15; 1978a: 52, fig. 22M. Poore, 1978: 141,
144.

Diagnosis

Telson parallel-sided, posterior third tapering to acute apex; dorsal
surface with few scattered fringed scales. Uropodal exopod narrow,
lanceolate, not reaching endopod. Maxilliped five-segmented, three distal
segments tiny.

Type material

Holotype, SAM~A13550, 1 6, 9,0mm, False Bay, 66m. Allotype,
SAM-A13551, 1 non-ovig. ¢°, 8,0mm, off Still Bay, 80m. Paratypes,
SAM-A13617, 2 d, 9,0-9,1 mm, False Bay, 66 m. Paratypes, SAM-—A13618, 2
non-ovig. 2, 6,5-7,9 mm, Agulhas Bank, 183 m. SAM-—A17583, 2 ovig. 2, 1
non-ovig. 2, south of East London, 90 m.

Other material

SAM-A14198, 1 ovig. 2, off Saldanha Bay, 320m. SAM-A14328, 1
non-ovig. 2, off Saldanha Bay. SAM—A14343, 1 ovig. 2, 1 non-ovig. 2, False
Bay, 75 m. SAM-A14344, 2 non-ovig. 2, False Bay, 26 m. SAM-A14346, 3
non-ovig. ¢, False Bay, 39 m. SAM-A14345, 1 non-ovig. 2, Agulhas Bank,
50 m. SAM-A14199, 1 non-ovig. ¢, Still Bay, 183 m. SAM-—A17584, 1 ovig.
2, Still Bay, 80 m. SAM-A14933, 1 6, 1 ovig. 2, locality unknown.

Distribution
Saldanha Bay to East London, 26-320 m.

176 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 51. Leptanthura agulhasensis. A. 3 dorsal view. B. Uropodal exopod.
C. Maxilliped. Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA wa

Leptanthura laevigata (Stimpson, 1855)
Figs 52-53
Anthura laevigata Stimpson, 1855: 393.
Leptanthura faurei Barnard, 1914: 345a, pl. 29B.

Leptanthura laevigata: Vanhoffen, 1914: 492, fig. 30. Barnard, 1925a: 151. Kensley, 1975a: 38;
1978a: 52, fig. 22N; 1980: 2. Poore, 1978: 138; 1980: 62.

Diagnosis

Telson posteriorly evenly rounded; dorsal surface covered with numerous
regular imbricate scales. Uropodal exopod almost circular, with strong notch in
distal margin.

Type material

Whereabouts unknown, Type locality: Simon’s Bay in False Bay, ‘12
fathoms’.

Other material

SAM-A14208, 1 6, 1 non-ovig. ¢, off Orange River mouth, 170 m.
SAM-A14298, 1 non-ovig. ¢, off Orange River mouth. SAM-A14234, 1
non-ovig. 2, Saldanha Bay. SAM-—A14273, 1 6, Saldanha Bay, 88 m.
SAM-A14284, 1 non-ovig. ¢, Saldanha Bay. SAM-A14329, 3 juvs, off
Saldanha Bay. SAM-A14331, 2 ¢6, Saldanha Bay. SAM—A14333, 1 non-ovig.
2, 2 juvs, Saldanha Bay. SAM-A14335, 1 6, 5 juvs, Saldanha Bay.
SAM-A14336, 1 non-ovig. 2, 2 juvs, Saldanha Bay. SAM—A14337, 1 subd, 2
juvs, Saldanha Bay. SAM-A14231, 1 ovig. 2, 1 non-ovig. 2, Langebaan.
SAM-A14839, 1 juv., Langebaan. SAM-A14246, 16, off Cape Peninsula.
SAM-A14205, 1 3, 4 non-ovig. 2, 1 juv., False Bay, 57 m. SAM-A14206, 1
non-ovig. 2, False Bay, 66 m. SAM-—A14207, 1 non-ovig. 2, False Bay, 54 m.
SAM-A14209, 1 6, False Bay, 26-29 m. SAM-A14210, 2 non-ovig. @, 1 juv.,
False Bay, 22 m. SAM-A14211, 3 d, 1 subd, 1 ovig. 2, 4 non-ovig. °, 7
juvs, False Bay 75 m. SAM-A14212, 3 non-ovig. 9, 1 juv., False Bay, 66 m.
SAM-A14213, 1 non-ovig. 2, False Bay, 29 m. SAM—A14214, 1 non-ovig. &,
False Bay, 36m. SAM-A14215, 2 6, 1 non-ovig. 2, False Bay, 87m.
SAM-A14216, 1 3, 4 non-ovig. 2, False Bay 40 m. SAM-A14217, 2 non-ovig.
2, 1 juv., False Bay, 17 m. SAM-A14219, 2 d, 1 non-ovig. ¢, False Bay,
42 m. SAM-A14221, 2 6, 2 ovig. 2, 2 non-ovig. 2, 2 juvs, False Bay, 87 m.
SAM-A14222, 5 6, 2 subd, 3 ovig. 2, 14 non-ovig. 2, 25 juvs, False Bay,
61m. SAM-A14224, 2 non-ovig. 2, False Bay 33m. SAM-A14225, 3
non-ovig. 2, False Bay, 26 m. SAM—A14226, 1 6, 1 ovig. 2, 6 non-ovig. 2, 2
juvs, False Bay, 66 m. SAM—A14227, 1 juv., False Bay, 44 m. SAM—A14288, 1
juv., False Bay, 26 m. SAM-A14233, 1 6, 2 non-ovig. 2, 6 juvs, False Bay.
SAM-A14235, 1 non-ovig. 2, False Bay, 48 m. SAM-A14237, 1 subd, False
Bay. SAM-A14239, 1 non-ovig. 2, False Bay, 13m. SAM-A14240, 16, 1
subd, 1 non-ovig. 2, 1 juv., False Bay. SAM—-A14241, 1 non-ovig. 2, False

178

ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 52. Leptanthura laevigata. A. 2 dorsal view. B. Uropod.
Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA 179

Fig. 53. Leptanthura laevigata. A. Cephalon @ ventral view. B. Cephalon 6d ventral view.
C. Telson. D. Telsonic scales enlarged.

Bay, 53 m. SAM-A14242, 2 non-ovig. 2, False Bay, 27 m. SAM-A14243, 1
non-ovig. ¢, False Bay, 48 m. SAM-A14244, 3 non-ovig. 2, 1 juv., False
Bay, 42 m. SAM-A14245, 1 non-ovig. 2, False Bay, 50 m. SAM-A14253, 1
ovig. ¢, False Bay, 68m. SAM-A14254, 1 ovig. ¢, False Bay, 64m.
SAM-A14255, 1 juv., False Bay, 44m. SAM-—A14256, 1 non-ovig. ¢, False
Bay. SAM-A14257, 1 non-ovig. ¢, False Bay. SAM—-A14259, 1 juv., False
Bay. SAM-A14260, 2 non-ovig. 2, False Bay, 42 m. SAM-A14261, 2 ovig. 9,
3 non-ovig. 2, False Bay, 81 m. SAM—A14262, 2 6, 2 ovig. 2, 4 non-ovig. 2,
False Bay. SAM-A14268, 1 ovig. 9, 1 non-ovig. ¢, False Bay, 50m.

180 ANNALS OF THE SOUTH AFRICAN MUSEUM

SAM-A14269, 1 juv., False Bay. SAM-—A14271, 1 subd, 1 ovig. 2, False Bay.
SAM-A14272, 1 dg, 1 non-ovig. 2, 4 juvs, False Bay. SAM—A14277, 1 6, 1
ovig. 2, 4 non-ovig. 2, False Bay, 82 m. SAM-A14279, 1 non-ovig. 2, False
Bay, 79 m. SAM-A14280, 1 6d, False Bay, 40 m. SAM-A14283, 2 juvs, False
Bay, 23m. SAM-A14287, 2 ovig. 2, 6 non-ovig. 9, False Bay, 58 m.
SAM-A14289, 2 non-ovig. 2, 1 juv., False Bay, 82m. SAM-A14290, 1
non-ovig. 2, False Bay, 36 m. SAM-A14291, 1 subd, 2 ovig. 2, 1 non-ovig.
2, False Bay, 59m. SAM-A14292, 1 6, 3 juvs, False Bay, 80m.
SAM-A14293, 3 non-ovig. 2, False Bay, 58 m. SAM—-A14295, 11 juvs, False
Bay, 5m. SAM-A14297, 1 6, 1 non-ovig. 2, False Bay, 59 m. SAM—-A14299,
1 3, 5 non-ovig. 2, 6 juvs, False Bay, 39 m. SAM-A14300, 1 ovig. 9, 1
non-ovig. ¢, 2 juvs, False Bay, 53 m. SAM-A14301, 1 ovig. 2, 1 juv., False
Bay, 82m. SAM-A14302, 1 6, 2 non-ovig. @, False Bay, 53 m.
SAM-A14303, 1 non-ovig. 2, False Bay, 16m. SAM-—A14304, 1 d, 1 subd,
False Bay, 44 m. SAM-A14305, 1 juv., False Bay. SAM—A14306, 1 ovig. 2, 4
juvs, False Bay, 19 m. SAM-A14307, 1 6, 1 subd, 2 ovig. 2, 3 non-ovig. 2,
False Bay, 82 m. SAM-A14308, 1 non-ovig. 2, 5 juvs, False Bay, 18 m.
SAM-A14309, 1 63, False Bay, 38 m. SAM-A14312, 2 63, 4 non-ovig. 9, 2
juvs, False Bay, 87m. SAM-A14314, 1 6, 3 juvs, False Bay, 56m.
SAM-A14313, 1 non-ovig. @. 1 juv., False Bay, 59m. SAM-—A14315, 4
non-ovig. 2, False Bay, 38 m. SAM-A14317, 3 non-ovig. 2, 2 juvs, False
Bay, 23m. SAM-A14318, 2 ovig. @, 2 non-ovig. 9°, False Bay, 58m.
SAM-A14320, 2 6, 1 non-ovig. 2, 4 juvs, False Bay, 58 m. SAM—A14321, 1
6, 1 non-ovig. 2, False Bay, 40 m. SAM—A14330, 1 d, 1 subd, 2 non-ovig.
2, 3 juvs, False Bay, 40m. SAM-A14332, 1 juv., False Bay, 38 m.
SAM-A14338, 1 6, 2 ovig. 2, 5 non-ovig. 2, 1 juv., False Bay.
SAM-A14340, 1 non-ovig. 2, False Bay, 26m. SAM—A14836, 3 juvs, False
Bay. SAM-A14838, 2 d, 4 ovig. 2, 6 non-ovig. 2, 3 juvs, False Bay, 40 m.
SAM-A14218, 1 ovig. 2, 1 non-ovig. 2, Agulhas Bank, 44 m. SAM-A14229,
1 ovig. 2, 2 non-ovig. 2, 1 juv., Agulhas Bank, 75 m. SAM-A14236, 2 juvs,
Agulhas Bank, 84m. SAM-A14247, 1 non-ovig. 2, Agulhas Bank, 100 m.
SAM-A14248, 1 juv., Agulhas Bank, 27 m. SAM-A14249, 1 non-ovig. &,
Agulhas Bank, 26m. SAM-A14251, 1 juv., Agulhas Bank, 172m.
SAM-A14252, 1 non-ovig. 2, Agulhas Bank, 183 m. SAM-A14263, 1 6, 1
non-ovig. 2, Agulhas Bank, 124m. SAM-A14264, 1 subd, Agulhas Bank,
93 m. SAM-A14265, 1 non-ovig. 2, Agulhas Bank, 97 m. SAM-A14267, 1
non-ovig. 2, Agulhas Bank, 84m. SAM-—A14270, 1 non-ovig. 2, Agulhas
Bank, 108 m. SAM-A14274, 1 ovig. 2, 2 non-ovig. 2, Agulhas Bank, 106 m.
SAM-A14275, 6 juvs, Agulhas Bank, 107 m. SAM-A14276, 4 non-ovig. ,
Agulhas Bank, 183m. SAM-A14278, 1 ovig. 2, Agulhas Bank, 46 m.
SAM-A14282, 2 non-ovig. 2, 1 juv., Agulhas Bank, 125 m. SAM-A14288, 1
non-ovig. 2, Agulhas Bank, 11 m. SAM-A14294 1 6, 1 non-ovig. 2, 1 juv.,
Agulhas Bank, 44m. SAM-A14296, 1 ovig. 2, Agulhas Bank, 300 m.
SAM-A14310, 1 non-ovig. ¢, Agulhas Bank, 44 m. SAM-A14316, 1 ovig. 2,

SOUTHERN AFRICAN ANTHURIDEA 181

1 non-ovig. 2, Agulhas Bank, 44m. SAM-A14837, 2 non-ovig. 2, Mossel
Bay. SAM-A14230, 1 non-ovig. ¢, off Knysna. SAM-—A14200, 1 6, 12
non-ovig. 2, 15 juvs, Still Bay, 20 m. SAM—A14201, 2 non-ovig. 2, Still Bay,
80 m. SAM-A14202, 1 non-ovig. 2, 1 juv., Still Bay, 15 m. SAM—A14203, 3
non-ovig. &, Still Bay, 15m. SAM-—A14204, 1 juv., Still Bay, 200m.
SAM-A14286, 3 juvs, off Port Elizabeth. SAM-—AS57, 2 6, 2 non-ovig. &, off
East London, 86m. SAM-A62, 1 ovig. 2, off Keiskamma Point, 66 m.
SAM-A64, 1 ovig. 2, off Cape St. Francis, 50 m. SAM—A2745, 3 non-ovig. °,
off East London, 104m. SAM-—A4169, 1 juv., off Cape Seal, 160 m.
SAM-A5959, 2 ovig. ¢, 2 non-ovig. ¢, off Duminy Point, 174 m.
SAM-A5960, 1 ovig. 2, off Cape St. Blaize, 84 m.

Distribution

Orange River mouth to Agulhas Bank; Durban to Mozambique Channel,
42-1 360 m.

Remarks

The material recorded by Kensley (1980) from Madagascar, Mauritius, and
Sumatra, although superficially similar to L. laevigata in uropodal and telsonic
shape, tends to be ovigerous at a smaller size and lacks imbricate scales on the
telson. These differences would indicate that at least one other species is
involved in the western Indian Ocean.

Leptanthura minuta Kensley, 1978
Fig. 54
Leptanthura minuta Kensley, 1978b: 16, figs 9-10.

Diagnosis

Adult male and female less than 5,0 mm total length. Integument not
indurate. Telson elliptical, posteriorly broadly rounded. Uropodal exopod
broadly oval, margin crenate.

Type material

Holotype, SAM-A15654, 1 6, 4,6mm. Allotype, 1 ovig. 2, 4,5 mm.
Paratypes, 1 3, 1 ovig. 2, 1 non-ovig. @, off Natal, 27°59’S 32°40’E, 550 m.
Paratypes, USNM 170545, 1 6, 1 ovig. 2, 1 non-ovig. 2, off Natal, 27°59’S
32°40’E, 550 m.

Other material

SAM-A15655, 1 ovig. 2, off Natal, 27°31’S 32°50'E, 750m. SAM-—
A15656, 2 3, 1 non-ovig. 2, off Natal, 30°53’S 30°31’E, 850 m.

Distribution
Off Natal, 550-850 m.

182 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 54. Leptanthura minuta. A. 3 dorsal view. B. Maxilliped. C. Antennule 92.
D. Antenna. E. Telson and uropod. F. Pereopod 1 d. Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA 183

Fig. 55. Leptanthura natalensis. A. 3 lateral view. B. Pereonite 5 ¢6 lateral view.
C. Uropodal endopod and basis. D. Maxilliped. E. Telson. F. Uropodal exopod. Scale in
mm.

184 ANNALS OF THE SOUTH AFRICAN MUSEUM

Leptanthura natalensis Kensley, 1978
Figs 55-56
Leptanthura natalensis Kensley, 1978b: 20, figs 11-12.

Diagnosis

Integument moderately indurate. Telson posteriorly evenly rounded.
Uropodal exopod broadly oval-subcircular. Short posteroventral lobes present
on some pereonites.

Type material

Holotype, SAM-A15657, 1 6, 20,6mm, off Natal, 27°09'S 32°58’E,
800 m. Paratypes, SAM-—A15658, 2 6, 6,7-6,8 mm, 1 juv., off Natal, 26°51’S
33°12'E, 720 m. Paratypes, SAM—A15659, 2 6, 6,7—7,5 mm, off Natal, 27°10’S
32°58'E, 820m. Paratypes, USNM 170546, 1 6, 8,0mm, 1 non-ovig. &,
5,7 mm, off Natal, 30°17’S 31°10’E, 820 m.

Other material

SAM-A15660, 1 6, 1 juv., off Natal, 30°33’S 30°48’E, 690 m.
SAM-A15661, 7 non-ovig. 2, off Natal, 30°53’S 30°31’E, 850 m.

Distribution
Off Natal, 690-850 m.

Fig. 56. Leptanthura natalensis. A. Cephalon é dorsal view. B. Pereopod 1 inner surface.

SOUTHERN AFRICAN ANTHURIDEA 185

Leptanthura urospinosa Kensley, 1975
Fig. 57
Leptanthura urospinosa Kensley, 1975a: 67, fig. 16; 1978a: 52, fig. 22K—L. Poore, 1980: 62.

Diagnosis
Telson parallel-sided, posterior third tapering to acute apex. Uropodal
exopod broadly oval, with dentate medial margin. Maxilliped four-segmented.

Type material

Holotype, SAM-A13619, 1 non-ovig. 2, 10,5 mm, False Bay, 26m.
Paratype, SAM—A13620, 1 non-ovig. 2, 13,3 mm, False Bay, 5 m.

Other material

SAM-A17611, 1 ovig. 2, 8,8 mm, 1 non-ovig. 2°, 9,6 mm, False Bay,
75m. SAM-A17612, 3 non-ovig. 2, 6,0-8,5 mm, False Bay, 39 m. SAM-—
A14341, 1 non-ovig. ¢, 8,0mm, off Still Bay, 120m. SAM-—A14342, 1
non-ovig. ¢, 5,4mm, off Still Bay, 200m. SAM-A17613, 1 non-ovig. 2,
5,2 mm, Agulhas Bank, 50 m.

Distribution
False Bay to Still Bay, 5-200 m.

Paranthura Bate & Westwood, 1868

Diagnosis

Eyes present. Pleonites more or less distinct. Antennular flagellum shorter
than peduncle. Antennal flagellum of one or more articles, short, flattened.
Mandibular palp three-segmented. Maxillipedal endite small to obsolete; palp
of one or two articles. Pereopod 1 propodus expanded, subchelate; pereopods
2-3 smaller than 1; pereopods 4-7 with rectangular carpi, anterior margins
subequal to posterior margins. Pleopod 1 exopod operculiform. Telson lacking
statocyst.

Type species
Paranthura costana Bate & Westwoad, 1868.

Remarks

Paranthura is probably the most speciose genus of the Paranthuridae, with
many of the species being superficially very similar. Species reported to have
wide geographical distributions should be viewed with suspicion; close
examination and attention to fine details of tail-fan structure especially will
probably reveal several restricted species masquerading under long-established
names.

186 ANNALS OF THE SOUTH AFRICAN MUSEUM

A. 2 dorsal view.

Fig. 57. Leptanthura_ urospinosa.
B. Maxilliped. C. Uropodal exopod. Scale in mm.

SOUTHERN AFRICAN ANTHURIDEA 187

KEY TO THE SOUTH AFRICAN SPECIES OF PARANTHURA

1. Telson densely setose; ischium and basis of pereopods 4~7 subcircular............. latipes
— Telson only sparsely setose; ischium and basis of pereopods 4~7 not expanded ... punctata

Paranthura latipes Barnard, 1955
Fig. 58

Paranthura latipes Barnard, 1955: 51, fig. 24d-f. Kensley, 1978a: 54, fig. 23H-I. Poore,
1980: 63.

Diagnosis

Telson ovate-lanceolate, bearing elongate finely plumose setae. Uropodal
exopod ovate, bearing plumose setae. Pereopod 1 with basal tooth on propodal
palm ‘not prominent’ (Barnard 1955: 51). Ischia and bases of pereopods 4-7
broad, subcircular.

Type material

Holotype, ¢d, 7,5 mm, Maxixe, Inhambane Bay, Mozambique, intertidal.
This specimen was not received by the South African Museum, and is
apparently lost.

Distribution

Inhambane, Mozambique; intertidal.

Remarks

This species has not been recorded since the original description.

Paranthura punctata (Stimpson, 1855)
Figs 59-60

Anthura punctata Stimpson, 1855: 393. Stebbing, 1910: 419.

Paranthura punctata: Hilgendorf, 1878: 847. Barnard, 1914: 348a, pl. 29C; 1920: 343;
1925a: 154; 1940: 490. Nierstrasz, 1941: 252. Thomson, 1951: 2. Hurley, 1961: 283.
Grindley & Kensley, 1966: 8. Kensley, 1975a: 39; 1978a: 55, fig. 23J; 1978b: 2. Chapman
& Lewis, 1976: 162, fig. 105. Poore, 1980: 64.

Diagnosis

Antennular flagellum of four articles; antennal flagellum of one article.
Mandibular palp segment 3 with comb of fourteen to sixteen fringed spines.
Maxilliped with rudimentary endite, palp of single setose segment. Pereopod 1
propodus expanded. Pleon shortened, pleonites free; pleonite 6 strongly
incised. Telson lanceolate, posteriorly rounded. Uropodal exopod narrowly
oval-lanceolate.

Type material
Whereabouts unknown. Type locality: False Bay, ‘20 fathoms’.

188 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 58. Paranthura latipes. A. Pereopod 5. B. Uropod and telson (from Barnard 1955).

SOUTHERN AFRICAN ANTHURIDEA 189

Fig. 59. Paranthura punctata. A. @ dorsal view. B. Antennule. C. Mandible.
D. Uropodal exopod. E. Uropodal basis and endopod. F. Maxilliped. Scale in mm.

190 ANNALS OF THE SOUTH AFRICAN MUSEUM

Other material

SAM-A12256, 1 non-ovig. ¢, 1 juv., off Orange River mouth.
SAM-A14371, 1 non-ovig. 2, off Saldanha Bay, 50-54 m. SAM-A2612, 2
non-ovig. 2, 1 juv., Mouille Point, Table Bay. SAM-—A14361, 1 non-ovig. &,
False Bay, 44m. SAM-A14363, 1 non-ovig. ¢°, False Bay, 39 m. SAM-—
A14364, 8 ovig. 2, 3 non-ovig. 2, 14 juvs, False Bay, 42 m. SAM-A14366, 1
non-ovig. 2, False Bay, 42 m. SAM—A14368, 1 non-ovig, 2, False Bay, 13 m.
SAM-A14369, 1 non-ovig. 2, False Bay. SAM-—A14377, 5 non-ovig. 2, False
Bay, 17m. SAM-A14382, 1 non-ovig. ¢, 1 juv., False Bay, 40m.

Fig. 60. Paranthura punctata. A. Cephalon ? dorsal view. B. Cephalon @ ventral view.
C. Antennal flagellum. D. Apex of oral cone, with projecting tips of maxillae.

SOUTHERN AFRICAN ANTHURIDEA 191

SAM-A14383, 1 non-ovig. 2, False Bay, 42 m. SAM-A14384, 1 non-ovig. &,
False Bay, 27 m. SAM-A14386, 1 ovig. @, 1 non-ovig. 2, False Bay, 66 m.
SAM-A14387, 2 ovig. 2, 2 juvs, False Bay. SAM-A14388, 1 non-ovig. 2,
False Bay, 35 m. SAM-A14389, 1 juv., False Bay, 51 m. SAM-A14391, 1 juv.,
False Bay, 7m. SAM-—-A14392, 1 non-ovig. 2, 2 juvs, False Bay, 18 m.
SAM-A14394, 2 ovig. 2, False Bay, 19 m. SAM-A14372, 2 juvs, Still Bay,
80 m. SAM—A14396, 1 non-ovig. ¢, Still Bay, 120 m. SAM-A14365, 1 juv.,
Agulhas Bank, 183 m. SAM-—A14367, 1 non-ovig. 2, Agulhas Bank, 10 m.
SAM-A14370, 1 juv., Agulhas Bank, 11-18 m. SAM-—A14374, 1 ovig. 2, 1
juv., Agulhas Bank, 121 m. SAM-—A14375, 1 non-ovig. 2, Agulhas Bank,
110 m. SAM-A14379, 1 ovig. 2, 2 juvs, Agulhas Bank, 84 m. SAM—A14380, 1
juv., Agulhas Bank, 84m. SAM-A14390, 1 ovig. 2°, Agulhas Bank, 84 m.
SAM-A14393, 1 non-ovig. 2, Agulhas Bank, 88 m. SAM-—A2555, 1 non-ovig.
2, off Umhlangakulu River, Natal, 100 m.

Distribution
Orange River mouth to Natal, 7-200 m.

Remarks

The specimens recorded from Australia (Thomson 1951) and New Zealand
(Hurley 1961; Chapman & Lewis 1976) need to be re-examined, as few
anthurideans have been found to have such wide distributions.

Barnard (1920) recorded a specimen of P. punctata being taken from a
Leuconia sp. sponge.

Pseudanthura Richardson, 1911

Diagnosis

Eyes absent. Pereonite 7 considerably shorter than preceding segments.
Pleonites and telson fused, anterior five pleonites indicated by shallow grooves,
pleonite 6 indistinguishable from telson. Statocyst absent. Maxilliped four-
segmented, endite present. Pereopod 1 subchelate, propodus expanded;
pereopods 2-3 ambulatory; pereopods 4-7 with rectangular carpi, not
underriding propodi. Pleopod 1 exopod operculiform, indurate; endopod
reduced. Uropodal exopod reduced to short triangular structure; endopod and
basis tending towards fusion. Brood pouch in female of four oostegites.

Type species
Pseudanthura lateralis Richardson, 1911.

KEY TO THE SOUTH AFRICAN SPECIES OF PSEUDANTHURA

1. Pleopod 1 endopod one-fourth length of exopod. Faint articulation between uropodal basis
ASM ODO triage corny Reta coe Nac coe hse ected ay NRA eM anE tee ala sah 4 5 tenuis
— Pleopod 1 endopod one-third length of exopod. No articulation between uropodal basis and
CHAO POC Bigeye ests Sea ree tee I Falster scene EMde RR ee eres, Ae Ya ea lateralis

192 ANNALS OF THE SOUTH AFRICAN MUSEUM

Pseudanthura lateralis Richardson, 1911
Figs 61-62

Pseudanthura lateralis Richardson, 1911: 7. Barnard, 1920: 343, pl. 15 (figs 13-16); 1925a: 157,
figs Is, 3e, 5d; 1940: 490, 497. Nierstrasz, 1941: 252. Menzies, 1962: 191, fig. 70. Kensley,
1978a: 55, fig. 23K—L; 1978c: 229, figs 5-6; 1982: 42, figs 40-42. Poore, 1980: 64.

Diagnosis
Pleopod 1 endopod one-third length of exopod, triangular. No articulation
between uropodal basis and endopod; exopod short, triangular.

Type material

Holotype, whereabouts unknown. Paratype, USNM 42171, 17,5 mm, off
Dakar, west Africa, 930-3 200 m.

Other material

SAM-A3832, 1 6, 15,5mm, 1 non-ovig. 2, 15,6mm, 1 ovig. 2
(damaged) + 16,0 mm, off Cape Point, 1 800-2 000 m. SIO station WHOI 191,
1 6, 160mm, 2 non-ovig. °, off South West Africa (Namibia), 23°04'S
12°31'E, 1553 m. USNM 185042, 2 6, 15,4-20,4 mm, 7 non-ovig. 2, 10 juvs,
off South West Africa (Namibia), 23°01’'S 12°19’E, 2136 m.

Distribution
West Africa to Cape Point, South Africa, 930-3 200 m.

Pseudanthura tenuis Kensley, 1978
Figs 63-64
Pseudanthura tenuis Kensley, 1978a: 224, figs 1-2; 1978b: 23, fig. 13. Poore, 1980: 64.

Diagnosis

Pleopod 1 endopod one-fourth length of exopod, triangular. Faint
articulation between uropodal basis and endopod; exopod short, triangular.
Telsonic margins straight. Male unknown.

Type material

Holotype, 1 non-ovig. 2, SAM—A15664, 25,3 mm. Paratypes, 3 non-ovig.
2, SAM-A15664, 6,0 17,1 23,4 mm, off Natal, 28°31’S 32°34’E, 680 m.
Paratype, 1 non-ovig. 2, SAM—A15665, 9,9 mm, off Natal, 26°51’S 33°12’E,
720 m. Paratypes, 2 non-ovig. 2, USNM 170272, 6,0 18,9 mm, off Natal,
28°31'S 32°34’E, 680 m.

Other material

SAM-A17614, 1 juv., off Natal, 26°51’S 33°12’E, 720 m. SAM-A17615, 2
non-ovig. 2, 14,0 18,5mm, 3 juvs, off Natal, 30°53’S 30°31’E, 850 m.

193

SOUTHERN AFRICAN ANTHURIDEA

CCCckcececcce

Jal

G

Fig. 61. Pseudanthura lateralis.
D. Antenna. E. Antennule 2. F. Pleopod 1. G. Maxilla. H. Maxilliped.
J. Uropod. Scale in mm.

A. 2 dorsal view. B. @ lateral view. C. Antennule a.
I. Mandible.

194 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 62. Pseudanthura lateralis. A. ¢ cephalon dorsal view. B. 6 cephalon dorsal view.
C. 2 pleon dorsal view. D. ¢ pleon ventral view; note copulatory stylet. E. ¢ pleon
dorsal view. F. ¢ pleon ventral view.

SOUTHERN AFRICAN ANTHURIDEA

Fig. 63. Pseudanthura tenuis. A. @ dorsal view. B. 2 lateral view.

C. Uropod. D. Pleopod 1. Scale in mm.

195

196 ANNALS OF THE SOUTH AFRICAN MUSEUM

SAM-A17616, 2 non-ovig. 92, 14,2 mm (damaged) off Transkei, 32°14’S
29°10’E, 560-620 m.

Distribution
Northern Natal to Transkei, 560-850 m.

Fig. 64. Pseudanthura tenuis. A. ¢ cephalon dorsal view. B. @ cephalon ventral view.
C. Pleon dorsal view. D. Pleon lateral view, uropod in situ.

ZOOGEOGRAPHICAL COMMENTS

In the following brief discussion, the genera and species of both the Anthuridae
and Paranthuridae are dealt with as a unit. The southern African region is defined as
being from the intertidal to the outer edge of the continental shelf, between the

SOUTHERN AFRICAN ANTHURIDEA 197

mouth of the Kunene River on the west coast to Vilanculos in Mozambique on
the east coast.

Of the 37 species here dealt with, 15 have been recorded from deeper than
200 m, while 5 of these range from considerably less than, to well beyond
200 m. Of these 15 deepwater species, only Pseudanthura lateralis, originally
described from West Africa, has been recorded outside the region under
discussion. When dealing with species occurring in deep water, little regarding
geographic distribution may be deduced, such deep areas being seldom well
sampled. Certainly, discussion of endemism in this context is futile. In fact,
several species of deep-water Atlantic anthurideans have much wider geogra-
phical distribution than most shallow-water species (Kensley 1982).

The shallow-water forms display a distributional pattern also seen in other
areas (e.g. southern Australia and the Caribbean), in which the anthurideans
have received any attention, viz. very high specific, and very low generic
endemism. Of the sixteen genera, only Centranthura (a recently created mono-
typic genus) is limited to the southern African region. Almost all the remaining
genera have been recorded from tropical to cold-temperate regions of most of
the world’s oceans. Of the twenty-two shallow-water species, only Accalathura
indica (occurring widely in the Indian Ocean) and Panathura amstelodami
(known from the St. Paul and Amsterdam islands, northern Mozambique, and
Madagascar) are not endemics.

ACKNOWLEDGEMENTS

My sincere thanks are due to Dr T. E. Bowman of the Smithsonian
Institution and Dr G. C. B. Poore of the National Museum of Victoria for
reading a draft of this paper, and for their many comments and criticisms.

I am grateful to Dr R. Lincoln and Miss J. Ellis of the British Museum
(Natural History) and Dr J. Forest of the Muséum National d’Histoire
Naturelle, Paris, for their hospitality during my visits.

Mr M. Carpenter, Miss M. Schotte, and the staff of the Smithsonian
Institution’s Electron Microscope Laboratory all assisted with the preparation
of illustrations—I am very grateful to them.

Dr P. E. Reavell of the University College of Zululand supplied a
specimen of Cyathura estuaria for study.

My sincere thanks are due to the Director and staff of the South African
Museum, and especially Miss A. E. Louw and Mrs M. van der Merwe, for
allowing me to examine material and for assistance during my 1981 visit; and
Mrs Ione Rudner, Editor, for her meticulous checking of the manuscript and
proofs.

198 ANNALS OF THE SOUTH AFRICAN MUSEUM

REFERENCES

Amar, R. 1952. Isopodes marins du littoral Corse. Bull. Soc. zool. Fr. 77: 349-355.

BARNARD, K. H. 1914. Contributions to the crustacean fauna of South Africa. 3. Additions to
the marine Isopoda, with notes on some previously incompletely known species. Ann. S.
Afr. Mus. 10: 325a—442.

BARNARD, K. H. 1920. Contributions to the crustacean fauna of South Africa. 6. Further
additions to the list of marine Isopoda. Ann. S. Afr. Mus. 17: 319-438.

BARNARD, K. H. 1925a. A revision of the family Anthuridae (Crustacea Isopoda) with remarks
on certain morphological peculiarities. J. Linn. Soc. 36: 109-160.

BARNARD, K. H. 1925b. Contributions to the crustacean fauna of South Africa. 9. Further
additions to the list of Isopoda. Ann. S. Afr. Mus. 20: 381-412.

BARNARD, K. H. 1935. Report on some Amphipoda, Isopoda, and Tanaidacea in the collections
of the Indian Museum. Rec. Indian Mus. 37: 279-319.

BARNARD, K. H. 1940. Contributions to the crustacean fauna of South Africa. 12. Further
additions to the Tanaidacea, Isopoda, and Amphipoda, together with keys for the
identification of the hitherto recorded marine and freshwater species. Ann. S. Afr. Mus.
32: 381-543.

BARNARD, K. H. 1955. Additions to the fauna-list of South African Crustacea and Pycnogonida.
Ann. S. Afr. Mus. 43: 1-107.

BARNARD, K. H. 1957. Additions to the fauna-list of South African Crustacea. Ann. Mag. nat.
Hist.(12) 10: 1-12.

BARNARD, K. H. 1959. New and little-known South African marine isopods (Crustacea). Ann.
Mag. nat. Hist. (13) 1: 715-720.

BaTE, C. S. & WEstTwoop, J. O. 1868. A history of the British sessile-eyed Crustacea 2. London:
John van Voorst.

BEDDARD, F. E. 1886. Report on the Isopoda collected by H. M. S. ‘Challenger’ during the
years 1873-1876. Second Part. Rep. Voy. Challenger 17: 1-178.

Bott, R. E. 1969. The Benthos of some southern African lakes. Part II. The epifauna and
infauna of the Benthos of Lake Sibayi. Trans. R. Soc. S. Afr. 38: 249-269.

BuRBANCK, M. P. & BuRBANCK, W. D. 1972. Distinguishing characteristics of Cyathura estuaria
Barnard (Crustacea, Isopoda) from South Africa. Bull. Ass. Syst. Biol. 19: 58.

CHILTON, C. 1924. Fauna of the Chilka Lake. Tanaidacea and Isopoda. Mem. Indian Mus. 5:
875-895.

CHAPMAN, M. A. & Lewis, M. 1976. Introduction to the freshwater Crustacea of New Zealand.
Auckland, New Zealand: Collins.

CurisTIE, N. D. 1976. A numerical analysis of the distribution of a shallow sublittoral sand
macrofauna along a transect at Lambert’s Bay, South Africa. Trans. R. Soc. S. Afr. 42:
149-172.

Day, J. H. 1959. The biology of Langebaan Lagoon: a study of the effect of shelter from wave
action. Trans. R. Soc. S. Afr. 35: 475-547.

Day, J. H. 1967. The biology of Knysna Estuary, South Africa. In: Laurr, G. H., ed.
Estuaries: 397-407. Washington, D.C.: American Association for the Advancement of
Science. (A.A.A.S. Publication 83.)

Day, J. H. 1969. A guide to the marine life on South African shores. Cape Town: Balkema.

Day, J. H., FIELD, J. G. & PENritTH, M. J. 1970. The benthic fauna and fishes of False Bay,
South Africa. Trans. R. Soc. S. Afr. 39: 1-108.

Day, J. H., MiLtarp, N. A. H. & BROEKHUYSEN, G. J. 1954. The ecology of South African
estuaries. Part IV. The St. Lucia System. Trans. R. Soc. S. Afr. 34: 129-156.

Day, J. H., MiLtarp, N. A. H. & Harrison, A. D. 1952. The ecology of South African
Estuaries. Part III. Knysna: a clear open estuary. Trans. R. Soc. S. Afr. 33: 367-413.
GRINDLEY, J. R. & KENSLEY, B. F. 1966. Benthonic marine fauna obtained off the Cranes

River mouth by the diamond dredger Emerson-K. Cimbebasia 16: 1-14.

HasweELL, W. A. 1881. On some new Australian marine Isopoda. Part I. Proc. Linn. Soc.
N.S. W. 5: 470-481.

HILGENDoRF, F. 1878. Die von Hrn. W. Peters in Mocambique gesammalten Crustaceen. Mber.
K. Akad. Wiss. Berl. 1878: 782-850.

Hurtey, D. E. 1961. A checklist and key to the Crustacea Isopoda of New Zealand and the
Subantarctic Islands. Trans. R. Soc. N. Z. (Zool.) 1: 259-292.

SOUTHERN AFRICAN ANTHURIDEA 199

KENSLEY, B. F. 1975a. Marine Isopoda from the continental shelf of South Africa. Ann. S. Afr.
Mus. 67: 35-89.

KENSLEY, B. F. 19755. A new species of Anthurid isopod from the Cape. Zool. afr. 10:
209-213.

KENSLEY, B. F. 1976. Isopodan and tanaidacean Crustacea from the St. Paul and Amsterdam
Islands, southern Indian Ocean. Ann. S. Afr. Mus. 69: 261-323.

KENSLEY, B. F. 1977. New records of marine Crustacea Isopoda from South Africa. Ann. S.
Afr. Mus. 72: 239-265.

KENSLEY, B. F. 1978a. Guide to the marine isopods of southern Africa. Cape Town: Trustees of
the South African Museum.

KENSLEY, B. F. 1978b. The South African Museum’s Meiring Naude Cruises. Part 8. Isopoda
Anthuridea. Ann. S. Afr. Mus. 77: 1-25.

KENSLEY, B. F. 1978c. Two new species of the genus Pseudanthura Richardson (Crustacea:
Isopoda: Anthuridea). Proc. biol. Soc. Wash. 91: 222-233.

KENSLEY, B. F. 1978d. Five new genera of anthurid isopod crustaceans. Proc. biol. Soc. Wash.
91: 775-792.

KENSLEY, B. F. 1979. New species of anthurideans from the Cook and Fiji Islands (Crustacea:
Isopoda: Anthuridea). Proc. biol. Soc. Wash. 92: 814-836.

KENSLEY, B. F. 1980. Anthuridean isopod crustaceans from the International Indian Ocean
Expedition, 1960-1965, in the Smithsonian Collections. Smithson. Contr. Zool. 304: 1-37.

KENSLEY, B. F. 1982. Deep-water Atlantic Anthuridea (Crustacea: Isopoda). Smithson. Contr.
Zool. 346: 1-60.

KENSLEY, B. F. & KoeEntInG, M. L. 1978. Two new species of Quantanthura from Brasil
(Crustacea: Isopoda: Anthuridae). Proc. biol. Soc. Wash. 91: 953-962.

Kroyer, H. 1847. Karcinologiske bidrag. Naturh. Tidsskr. 2: 1-123, 366-446.

Larwoop, H. J. 1940. The fishery grounds near Alexandria. XXI. Tanaidacea and Isopoda.
Notes Mems. hydrobiol. Dep. U. A. R. 35: 1-72.

MENZIES, R. J. 1962. The isopods of abyssal depths in the Atlantic Ocean. Vemma Res. Ser. 1:
79-206.

MeENzIES, R. J. & GeorGE, R. Y. 1972. Isopod Crustacea of the Peru-Chile Trench. Anton
Bruun Rep. 9: 1-124.

MENzIES, R. J. & GLyNnn, P. W. 1968. The common marine isopod Crustacea of Puerto Rico.
A handbook for marine biologists. Stud. Fauna Curacao 27: 1-133.

MILLARD, N. A. H. & Harrison, A. D. 1954. The Ecology of South African Estuaries. Part 5.
Richards Bay. Trans. R. Soc. S. Afr. 34: 157-180.

MILterR, M. A. & BurRBANCK, W. D. 1961. Systematics and distribution of an estuarine isopod
crustacean, Cyathura polita (Stimpson, 1855), new comb., from the Gulf and Atlantic
Seaboard of the United States. Biol. Bull. 120: 62-84.

NrerstrasZ, H. F. 1941. Die Isopoden der Siboga-Expedition, IV: Isopoda Genuina, III:
Gnathiidea, Anthuridea, Valvifera, Asellota, Phreatocoidea. Siboga Exped. monogr. 32d:
1-308.

NorMan, A. M. & StTesBING, T. R. R. 1886. On the Crustacea Isopoda of the ‘Lightning’,
‘Porcupine’, and ‘Valorous’ Expeditions. Trans. zool. Soc. Lond. 12: 77-141.

PENRITH, M.-L. & KENSLEY, B. F. 1970. The constitution of the intertidal fauna of rocky shores
of South West Africa, Part 1. Liideritzbucht. Cimbebasia (A) 1: 191-239.

Pittal, N. K. 1966. Littoral and parasitic isopods from Kerala: Family Anthuridae—1.
J. Bombay nat. hist. Soc. 63: 152-161.

PoorE, G. C. B. 1975. Australian species of Haliophasma (Crustacea: Isopoda: Anthuridae).
Rec. Aust. Mus. 29: 503-533.

Poore, G. C. B. 1978. Leptanthura and related genera (Crustacea, Isopoda, Anthuridea) from
Eastern Australia. Mem. Nat. Mus. Vic. 39: 135-169.

PooreE, G. C. B. 1980. A revision of the genera of the Paranthuridae (Crustacea: Isopoda:
Anthuridea) with a catalogue of species. Zool. J. Linn. Soc. 68: 53-67.

RATHBUN, M. J. 1907. Introductory note. Jn: Stimpson, W. Report on the Crustacea
(Brachyura and Anomura) collected by the North Pacific Exploring Expedition,
1853-1856. Smithson. Misc. Coll. 49: 3-240.

RICHARDSON, H. 1902. The marine and terrestrial isopods of the Bermudas, with descriptions of
new genera and species. Trans. Conn. Acad. Sci. 11: 277-310.

200 ANNALS OF THE SOUTH AFRICAN MUSEUM

RicHARDSON, H. 1905. A monograph on the isopods of North America. Bull. U.S. natn. Mus.
54: 1-727.

RICHARDSON, H. 1911. Les Crustacés isopodes du Travailleur et du Talisman; formes nouvelles.
Bull. Mus. Hist. nat., Paris 7: 518-534.

Sars, G. O. 1873. Bidrag til Kundskaben om Dyrelivet paa vore havbbanken. Forh. Vidensk.
Krist. 1872: 73-119.

Sars, G. O. 1899. An account of the Crustacea of Norway. 2. Isopoda. Bergen: Bergen
Museum.

STEBBING, T. R. R. 1900. On Crustacea brought by Dr Willey from the South Seas. Willey’s
Zool. Res. 5: 605-690.

STEBBING, T. R. R. 1910. General catalogue of South African Crustacea. Ann. S. Afr. Mus. 6:
281-593.

Stimpson, W. 1855. Descriptions of some new marine Invertebrata. Proc. Acad. nat. sci.
Philad. 7: 375-397.

TuHomson, J. M. 1951. The fauna of Rottnest Island. X. Anthuridae. J. R. Soc. West. Austr. 35:
1-8.

VANHOFFEN, E. 1914. Die Isopoden der Deutschen Siidpolar-Expedition 1901-1903. Dt.
Stidpol.-Exped. 15: 447-598.

WAGELE, J. W. 1981. Zur Phylogenie der Anthuridea (Crustacea, Isopoda). Mit Beitragen zur
Lebensweise, Morphologie, Anatomie und Taxonomie. Zoologica 132: 1-127.

WALKER, A. O. 1901. Contributions to the Malacostracan Fauna of the Mediterranean. J. Linn.
Soc. Lond. 28: 290-306.

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, 1.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.

oe punctuation in the above example:
comma separates author’s name ang 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

b] 6

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

(c) Scientific names, but not their vernacular derivatives
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.

BRIAN KENSLEY

REVISION OF THE
SOUTHERN AFRICAN ANTHURIDEA
(CRUSTACEA, ISOPODA)

ART 4 SEPTEMBER 1982 Seal fa ISSN 0303-2515

OF THE SOUTH AFRICAN
MUSEUM

CAPE TOWN 2 an SOF &

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
(c) Table of contents giving hierarchy of headings and subheadings
(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 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, 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
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.

Konn, A. J. 19606. Spawning behaviour, egg masses and larval development in Conus from the Indian Ocean.
Bull. Bingham oceanogr. Coll. 17 (4): 1-51.

THEELE, 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 90 Band
September 1982 September
Part. 4. . Deel

SS
CS
pracy iS

MORPHOLOGICAL AND BIOLOGICAL NOTES
ON SIX SOUTH AFRICAN BLOW-FLIES
(DIPTERA, CALLIPHORIDAE)

AND THEIR IMMATURE STAGES

By
A. J. PRINS

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), S=O)), UD, ZS), Geo), SUES, 5, I);
GG, rot), GED), &, OGD, 7), 1OGES)).
GED, 5 7, tan), CED), MAO), HL, NCES, 26), 25, 2EO), 25)

EDITOR/REDAKTRISE
Ione Rudner

Copyright enquiries to the South African Museum

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 040 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

MORPHOLOGICAL AND BIOLOGICAL NOTES ON SIX SOUTH
AFRICAN BLOW-FLIES (DIPTERA, CALLIPHORIDAE) AND
THEIR IMMATURE STAGES

By
A. J. Prins
South African Museum, Cape Town
(With 5 figures)
[MS accepted 8 July 1982|

ABSTRACT

Short biological notes and illustrations are given on Lucilia sericata (Meigen), Chrysomya
albiceps (Wiedemann) and C. chloropyga (Wiedemann). Larvae and pupae of Calliphora
croceipalpis Jaennicke, Chrysomya regalis Robineau-Desvoidy, and C. megacephala (Fabricius)
are described and illustrated with notes on their biology and association with decaying organic
matter. A key for the identification of the larvae is provided.

CONTENTS

PAGE
| SDCTRO CIES TOY ON es eee 2 cee a: Oem Se RES RE pares MN ec 201
EEE ESCHICALO INICIO EM) raised aie a odie otis 2 nd aie Reta eetelai ees 202
Chrysomya albiceps (Wiedemann) .......-. 52220-2205. eaeee 204
Chrysomya chloropyga (Wiedemann): .. 2. . 2255802. -2.026- 204
@alliphora croceipalpis Wacmmicke 0s .ae fat = Gees bee ee 205
Chrysomya regalis Robineau-Desvoidy ...................--- 209
Ginysomya megacephala (Fabricius) —. 21025. - ase. ee 2 ae 213
Key foridentification of larvae . 2252S]. . 02-250. ees phen 216
PACKMOWIEMPEMENES -2. ice aac cuide Hae ees sede Se iegeteass ees, cies ZA)
TRE HCE SU Po are hoe an Meera eaves AR Rete een Aee ot ace ce AY ZAG.
PROUREVIAUONS oi. wre te es | ek ne ga ee eee 217

INTRODUCTION

The blow-flies commonly known as bluebottles and greenbottles are
mostly carrion breeders and, apart from the newly introduced oriental latrine
fly, Chrysomya megacephala (Fabricius), the large bluebottle, C. regalis
Robineau-Desvoidy, and the cadaver fly, Calliphora croceipalpis Jaennicke, are
usually the first flies to appear at more or less fresh carcasses and cadavers in
the south-western parts of the Cape Province. Both the last-named species have
been involved in the past in myiasis in man, C. regalis in traumatic myiasis and
C. croceipalpis in wound and enteric myiasis (Porter 1924; Zumpt 1956).

According to observations made during surveys along the coastal areas
from Mossel Bay to Port Nolloth, C. regalis infests mainly large animals such as
eland and gnu, whereas C. croceipalpis was seen to infest bird carcasses and
some of the smaller mammals such as rats and dassies; occasionally, however,
its larvae were recovered from larger animals such as gnu, but very few adults

201

Ann. S. Afr. Mus. 90 (4), 1982: 201-217, 5 figs.

202 ANNALS OF THE SOUTH AFRICAN MUSEUM

were reared in these cases due to competition by C. albiceps (Wiedemann), C.
chloropyga (Wiedemann), and C. megacephala. Both these species infest
human cadavers, C. croceipalpis apparently mostly during the winter and early
spring, and C. regalis mostly during the warmer autumn months.

Chrysomya megacephala may also be regarded as a first-wave fly at
decaying carcasses; however, it still seems to be restricted to the Cape
Town-Yzerfontein and Durban areas as it has not been observed elsewhere. It
is commonly found at the Cape Town docks, breeding in ships’ holds carrying
fish-meal, and this is probably also one of the ways in which it was introduced
into this country; the larvae are often found during the late autumn crawling
along the railway lines where they breed in dead rats and pigeons.

The green-tailed blow-fly, Chrysomya chloropyga (Wiedemann), and the
green blow-fly, Lucilia sericata (Meigen), are widely distributed in South Africa
and are often associated with the above-mentioned species as first- and
second-wave flies respectively. Both these flies are facultative parasites and are
responsible for sheep myiasis, the first one being subordinate to the second.
The final instar larvae of a third species, the banded blow-fly, Chrysomya
albiceps (Wiedemann), are mostly predaceous on the larvae of the others and
they usually develop in carcasses in an advanced state of decomposition.

It is interesting to note that adult blow-flies obtain their carbohydrate
requirements from nectar and honeydew, and therefore play an important part
in the pollination of flowers such as Stapelia (Ryke 1969). Specimens of
Calliphora croceipalpis, Chrysomya chloropyga, and Sarcophaga spp. coloured
red on the head and thorax by pollen of the plant Ferraria crispa (Iridaceae)
were often observed in the sandy areas along the west coast. These flies
apparently also serve as pollinators of this plant.

In order to study the different instars, the eggs or larvae were placed on
rotting flesh in glass jars covered with gauze cloth and containing a layer of
about 75 mm of clean, damp sand. Drawings were made with a camera lucida
from specimens preserved in 80 per cent alcohol.

Lucilia sericata (Meigen)

Fig. 1A

Adult and larval instars were described by Zumpt (1956, 1965). This
brilliant green, shiny blow-fly is apparently attracted by a different set of
odours than is the cadaver fly (Calliphora croceipalpis) and in most of the cases
that were examined it appeared as a second-wave fly. In a few instances,
however, particularly in gutted animals, it occurred as a first-wave invader
together with one of the flesh flies (Sarcophaga sp.). It is one of the most
common blow-flies on the islands along the west coast and often breeds in
decaying kelp on the beach together with the kelp fly, Coelopa africana
Malloch.

SOUTH AFRICAN BLOW-FLIES 203

Fig. 1. A. Lucilia sericata, female. B. Chrysomya chloropyga, female. C. C. albiceps, female.
D. C. megacephala, female. E. C. megacephala, head of male from above. F. C. megace-
phala, head of male from the front. G. C. chloropyga, head of male from above.

H. C. chloropyga, head of male from the front.

204 ANNALS OF THE SOUTH AFRICAN MUSEUM

Eggs collected on human corpses measure, 1,2-1,36 xX 0,32-0,36 mm, and
closely resemble those of Chrysomya megacephala (Fig. 3M), being creamy
white, smooth and shiny, and almost without any visible reticulation. The
ledges are narrow, as close together as in C. megacephala, and extend for
nearly seven-eighths of the length of the egg. In most cases the eggs hatched 14
hours after oviposition.

In one test in the laboratory at a room temperature of 25-28 °C, the first
instar occupied 14-15 hours and the second 14-17 hours, the larvae reaching a
length of 3,4-4 mm 13-16 hours after hatching. The third instar lasted 143-150
hours and the pupal stage about 120 hours, the total larval lifespan in this case
was nearly 170-180 hours. In another instance, however, the first and second
instars occupied almost 48 hours at 26 °C and the third instar 140-160 hours.
The total larval lifespan was 187-204 hours and the pupal stage 192 hours.

Chrysomya albiceps (Wiedemann)
len UC

Adult and larval instars were described by Zumpt (1956, 1965). The fly
varies from 5 to 10 mm in length and is metallic green with a black transverse
band across the posterior border of each abdominal segment. It is very widely
distributed and as a sheep myiasis producer it is a secondary fly; as far as is
known, it has not been recorded on human beings. It has been observed on a
wide range of carcasses and cadavers and is usually the last blow-fly to be
attracted before the skin-and-hide beetles appear.

Eggs collected on decaying seal carcasses along the west coast vary from
1,4 to 1,5 mm by 0,30 to 0,34 mm, and are very similar to those of Lucilia
sericata but are duller and with a more pronounced reticulation. All eggs
examined hatched 21 hours after oviposition (at temperatures of 25-28 °C).

In the laboratory the first instar lasted 15-20 hours at room temperatures
of 25-28 °C, and reached a length of 3 mm about 12 hours after hatching. The
second instar occupied 26-30 hours after which they were 6-7 mm long. The
third instar varied from 153 hours to 158 hours, the total larval lifespan being
199 hours to 204 hours. The pupal period was 96 hours.

Chrysomya chloropyga (Wiedemann)
Fig. 1B, G-H

Adult and larval instars were described by Zumpt (1956, 1965). A metallic
greenish-blue fly with the posterior part of the abdomen greenish yellow, and
easily recognized by the _| _ marks on the presutural area of the mesonotum,
which are absent in the very similar C. megacephala. The eyes of the males
without distinct separation of upper and lower facets (Fig. 1H).

The eggs collected on a dead cow measure about 1,6mm X 0,36 mm and
are very similar to those of C. megacephala, being also shiny white with an
indistinct reticulation.

SOUTH AFRICAN BLOW-FLIES 205

In two laboratory tests the first instars lasted 20-22 hours at a room
temperature of 22—25 °C, after which they measured about 3,5 mm in length.
They increased in size reaching 7,9-8,7 mm about 48 hours after hatching; this
second instar occupied at least 25-27 hours. The third instar larvae matured
after 114-177 hours, the total larval lifespan being 162-230 hours, depending on
food availability, temperature and humidity; the mature larvae reached a length
of 16-17 mm. The pupal stage lasted from 188 to 204 hours.

In the case of third instar larvae (12-13 mm long), which were removed 85
hours after hatching from their original food source, cessation of feeding was
observed and pupation occurred about 38 hours after removal. The total larval
lifespan was 85-86 hours. The pupal stage lasted 144-168 hours, the flies
emerging being of normal size, but the time necessary to straighten their wings
and to assume their normal colour was much longer than in the other tests
where the larvae were left undisturbed to mature on their original food supply.

Calliphora croceipalpis Jaennicke
Adult (Fig. 2A—C)

For a detailed description of the adult see Zumpt (1956, 1965). Head and
thorax blackish blue to steel blue, almost dull; abdomen more greenish blue
and almost shiny and with whitish pollinosity forming large patterns; dark
iridescent transverse band present on posterior border of each abdominal
segment. Tuft of strong setae present on each side of first visible abdominal
segment, just before its posterior border. Four short distinct vittae present on
anterior half of presutural area of mesonotum, indistinct over rest of
mesonotum. Reddish colour of basal parts of third antennal segments and
orange colour of palpi very characteristic. In the males, frontal stripe narrowed
by eyes and there are only two vertical bristles, outer verticals (present in
female) being absent. Normal length 9-12 mm.

Widely distributed throughout the Subsaharan region, except probably the
western parts of Africa. It is common in South Africa and also occurs on some
of the islands along the west coast such as Dassen Island.

Larva

First Instar

Similar to mature larva, but spinose girdles on segments 7 and 8 and on
posterior side of 11 much less distinct. Cephalopharyngeal skeleton (Fig. 2J)
different from that of second and third instars; basal piece widely and deeply
emarginate behind; dorsal and ventral cornuae narrow and connected by
narrow central piece; anterior dorsal bridge very narrow and widely separating
the two dorsal cornuae; parastomal sclerite absent. Mouth-hooks rather weak,
and connected to median piece as shown in Figure 2J. Anterior spiracles very
indistinct and branches absent. Posterior spiracles (Fig. 2P) with partly
developed peritremal ring, orifices ovate and touching ventrally. Just before
moulting larvae reach a length of 3,3-4,0 mm.

206 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 2. Calliphora croceipalpis. A. Adult female. B. Head, frontal view of female. C. Head,
frontal view of male. D. Caudal aspect of puparium. E. Puparium, dorsal view. F. Puparium,
left lateral view. G. Egg. H. Third stage larva, left lateral view. I. Cephalopharyngeal
skeleton of third instar larva. J. Cephalopharyngeal skeleton of first instar larva. K. First two
segments of third instar larva, left lateral view. L. First two segments of third instar larva,
dorsal view. M. Spines highly magnified. N. Cephalopharyngeal skeleton of second instar
larva. O. Caudal aspect of third instar larva. P. Caudal aspect of first instar larva. Q. Caudal
aspect of second instar larva.

SOUTH AFRICAN BLOW-FLIES 207

Second Instar

Very similar to final instar, including number of branches on anterior
spiracles and spinose girdles on body segments. However, cephalopharyngeal
skeleton (Fig. 2N) much more heavily sclerotized than in first instar and rather
similar to that of third instar larva, except for absence of oral sclerite.
Mouth-hooks are more robust and dorsal projections longer and more acute
than those of first instar. Parastomal bars or sclerites slender, almost horizontal
in specimens seen, and directed upward at extreme apex. Posterior spiracles
(Fig. 2Q) much larger and each one with two ovate orifices and an open
peritremal ring. Larvae vary from 4,6 to 8mm in length before moulting
occurs.

Third Instar (Fig. 2H-I, K-M, O)

Length 16,5—-18 mm when full-grown and creamy white to reddish white,
depending on food. Segments 2-8 with complete anterior spinose girdles
(sometimes also on segment 9); segments 9-11 with spinose girdles on ventral
and lateral sides only; posterior girdles also on ventral side of segments 6-11,
that on 11 complete; lateral spines not very distinct in some specimens; segment
12 with anterior ventral girdle and some spines surrounding anus. Spines
typically with rounded apical dents, those on more posterior segments with
rather broader dents, while larger spines on ventral parts of segments 8-12 are
more or less triangular. Head deeply emarginate in front, forming oval lobe on
each side, which bears small but rather conspicuous antennae and maxillary
palpi. First spinose girdle with dorsal and ventral spines well developed, those
placed laterally rather inconspicuous.

Cephalopharyngeal skeleton (Fig. 21) with ventral cornuae of basal piece
shorter than dorsal ones, each with light area with less sclerotization; dorsal
ones with lightly sclerotized dorsal area and also with small oval, less sclero-
tized area near posterior border. When viewed from above the two cornuae
fairly parallel or slightly diverging posteriorly. Anterior dorsal bridge complete,
somewhat truncate in front and apparently without anterior and posterior
projecting lobes. Parastomal bars slender and directed slightly upward particu-
larly at apex. Median piece with complete arched ventral bridge and almost
H-shaped when viewed from below. Mouth-hooks fairly wide apart, upper
surface in lateral view fairly straight and forming a rounded projection poster-
iorly. Dental sclerites present; accessory oral sclerite wider than in C. vicina
(see Zumpt 1965), its basal lobes broad and lightly sclerotized. When seen from
below it forms a Y-shaped structure, its basal lobes connected to mouth-hooks,
forming posterior arms.

Last abdominal segment (Fig. 20) with twelve conspicuous fleshy
tubercles as well as two smaller inconspicuous tubercles located in posterior
cavity some distance above the two median ventral tubercles. Each yellowish
posterior spiracle consists of three elongate, brownish oval slits surrounded by
closed black peritremal ring with button on inner margin; distance between

208 ANNALS OF THE SOUTH AFRICAN MUSEUM

spiracles is about same as diameter of a spiracle. Anterior spiracles with eight
to nine branches.

Puparium (Fig. 2D-F)

Smooth, dull or slightly shiny, finely and transversely striate; posterior
face finely rugulose. Light red when freshly formed, becoming brownish red
after day or so. Somewhat constricted just behind anterior spiracular openings,
which are small and circular. Respiratory horns absent. Anterior lateral ridges
stretch over second and third segment. Twelve small, but conspicuous tubercles
surround posterior spiracular plates (Fig. 2D). Posterior spiracles with three
distinct elongate slits. Just below each spiracle there is a darker patch and
below this patch an almost circular to oval dark spot, slightly smaller than
spiracles and absent or inconspicuous in some specimens; between these two
patches are two tiny tubercles. Also short black tubercle on each side of anal
opening; latter about same size or slightly larger than spiracles. Posterior face
convex; spiracles flush with surface and fairly high above longitudinal axis.
Normally puparia measure about 9-10 mm.

Biology

This blow-fly was observed around Cape Town almost throughout the
year, but was found to be more active during the winter months and early
spring. It attacks carcasses of both birds and mammals and, as already stated, is
usually the first fly to be attracted to human cadavers. In the laboratory eggs
were deposited on the surface of fresh meat, but the flies refused to lay eggs on
fairly decayed meat.

Eggs (Fig. 2G) vary from 1,7 x 0,44 mm to 1,8 x 0,44 mm and are white in
colour, almost matt, with the surface finely reticulate; the lateral ledges or
flanges are narrow and close together, extending for almost the whole length of
the egg, the median area between the flanges is narrow. In most cases the eggs
hatched within 21-23 hours after oviposition at a room temperature ranging
from 18 to 23 °C during August to September. When the temperature was raised
to 26 or 28 °C, some of the eggs hatched within 18 hours and in one case three
eggs hatched after 15 hours.

The amount of food available and the temperature play an important part
in the development of the larvae. However, the available information indicates
that the duration of the first stage larva is 23-24 hours, after which the first
moult occurs. Just before each moult the spiracles of the next stage become
visible through the integument so that it is fairly easy to establish the time of
the next moult.

The lifespan of the second instar larva is 37-40 hours at a room tempera-
ture of 18-23°C. The third instar, which may occur about 60 hours after
hatching, feeds voraciously for 5—6 days, after which it enters the soil to rest for
another 2-5 days before pupation occurs. In most cases the total lifespan of
third instar larva was about 8,5 days. The total lifespan of the larvae from egg

SOUTH AFRICAN BLOW-FLIES 209

to pupa may vary from about 8 days during October to December to about 13
days during July to August, depending on the prevailing humidity and
temperature; in tests during August with a room temperature of 18 °C the
lifespan of the larvae was 10,75 days (258 hours).

Puparia are formed in the soil without the formation of cocoons and the
flies appear after about 14 days from October to December and after 20 days
from July to September. The whole cycle from egg to adult is about 25 days
during September to November, to about 33 days during July to August.

As only two cases have been reported in the past in which this oviparous
species had been identified as causing dermal and intestinal myiasis in man
(Porter 1924), it would seem to be only an occasional facultative parasite,
probably because it is easily disturbed and because of its habit of alighting and
flying around under such circumstances rather than settling down.

Chrysomya regalis Robineau-Desvoidy

Adult (Fig. 3A—C)

This was previously described by Smit (1929) and the genitalia of the male
by Zumpt (1956). Colour bright bluish, presutural area of mesonotum
somewhat lighter in colour than rest of body, its extreme anterior margin black
and with two paramedian longitudinal lines of lighter colour. First visible
segment of abdomen blackish, other abdominal segments each with posterior
black transverse band; median area of each segment with white pollinosity,
showing up as paler areas in incident light. Eyes in life bright red. Parafrontalia
and genae yellowish due to presence of fine pubescence and longer fine
yellowish hairs, which are blackish round ocellar triangle. Frontal stripe,
antennae and palpi somewhat reddish. Vibrissae consisting of only two long
black setae.

Females with outer and inner verticals and with no separation between
upper and lower facets of eyes. In males only inner verticals present and eyes
contiguous in middle; large upper facets of eyes clearly separated from lower
smaller facets, as in C. megacephala. Easily distinguished from latter species by
white anterior thoracic spiracles (blackish brown in C. megacephala) and by
dark costal margin of wing. Length 9,9-11,5 mm.

Widely distributed over Subsaharan region and also present in Arabia,
certain parts of India and the Madagascar region. Although present in most
areas surveyed, it is not common.

Larva

Third Instar (Figs 3F, H-K, 4C, 5A)

Length 14-18 mm, similarly coloured to that of C. croceipalpis, with
complete anterior spinose girdles on segments 2-11, segment 12 with ventral
band only and some spines around anus. Anterolateral swelling present on
segments 5-10; these adjoin spinose girdles and bear some spines. Head

210 ANNALS OF THE SOUTH AFRICAN MUSEUM

‘ee serena

sh ¢
<a

Fig. 3. A-K. Chrysomya regalis. A. Female. B. Head of female from front. C. Head of
male from front. D. Puparium, dorsal view. E. Puparium, left lateral view. F. Larva, left
lateral view. G. Posterior view of puparium. H. First two segments of third instar larva, left
lateral view. I. First two segments of third instar larva, dorsal view. J. Posterior view of third
instar larva. K. Cephalopharyngeal skeleton of third instar larva. L-T. Chrysomya megace-
phala. L. First two segments of third instar larva, left lateral view. M. Egg. N. Cephalopha-
ryngeal skeleton of third instar larva. O. Right posterior spiracle of third instar larva.
P. Cephalopharyngeal skeleton of second instar larva. Q. Cephalopharyngeal skeleton of first
instar larva. R. Cephalopharyngeal skeleton of first instar larva, dorsal view. S. Right
posterior spiracle of first instar larva. T. Right posterior spiracle of second instar larva.

SOUTH AFRICAN BLOW-FLIES 211

Oras]
a
mm
4 i>)
aid af
oe nes
2 aes
, So
sa p B
Ona
ee)
mm

Fig. 4. A. Mature larva of Chrysomya chloropyga. B. Mature larva of C. megacephala.
C. Mature larva of C. regalis. D. Mouth-hook of C. chloropyga, left lateral view.
E. Cephalopharyngeal skeleton of Lucilia sericata.

DD ANNALS OF THE SOUTH AFRICAN MUSEUM

divided into two lobes, each bearing short antenna and maxillary palp. First
spinose girdle almost surrounding second segment, spines on its lateral and
ventral side small and inconspicuous. All spines on all segments much thicker
than in C. croceipalpis and almost spoon-shaped. Antennae and maxillary palpi
clearly visible.

Posterior spiracles large and yellowish, distance between them slightly less
than half the diameter of spiracle, each with open thin black peritremal ring
and with three elongate brownish oval slits; each slit with a projection with less
sclerotized central area and also with sclerotized elongate sclerite between two
inner slits, with circular opening dorsally. These projections, rather character-
istic of this species, also present in C. megacephala, but spiracles are smaller
and projections therefore less noticeable; apparently absent in C. chloropyga
and C. albiceps. Anterior spiracles with eleven to fifteen short branches, the
spiracles short and wider than in C. croceipalpis.

Cephalopharyngeal skeleton (Fig. 3K) with ventral cornuae almost as long
as dorsal ones; anterior dorsal bridge complete, somewhat rounded when seen
from above and without any projections; cornuae slightly diverging towards
back and forming a narrow V. Parastomal bars slender, fairly straight, some-
what wider in front than behind and somewhat inclined anteriorly, also
extended slightly upwards at extreme apex in some specimens (e.g. C. crocei-
palpis).

Median piece with complete, arched ventral bridge, rather similar to that
of C. croceipalpis. Mouth-hooks without any posterior projections or with short
projections, shorter than in C. croceipalpis; dorsal surface fairly straight in
lateral view, in some specimens slightly raised in middle. Small V-shaped dental
sclerite present laterally and only V-shaped basal part of accessory oral sclerite
present. Last body segment (Fig. 3J) rather similar to that of C. croceipalpis,
but two small tubercles present in latter could not be traced.

Puparium (Fig. 3D-E, G)

Smooth, dull, fairly coarsely and transversely striate, striations forming oval
rugulose patterns laterally, posterior face also rugulose in certain areas. Colour
light reddish when newly formed, but later becoming dark blackish brown.
Very small respiratory horns present and puparium somewhat constricted just
behind these horns. Anterior lateral ridges not very conspicuous, extending
over second and third segments and located on or just below longitudinal axis.
Anterior spiracles somewhat fan-shaped and yellow. Spinose bands present on
almost all segments, those on segment 9 inconspicuous and narrower than those
on others. Posterior face (Fig. 3G) rather truncate, spiracles large, somewhat
protruding and enclosed in black peritremal ring, slits yellowish brown to-
almost orange. Twelve tubercles surrounding spiracular plates not very distinct.
Oval shallow depression present below each spiracle, latter situated above
longitudinal axis. Anal opening similar to that of C. croceipalpis and with short
projections on each side. Puparium measures 8,9-10 mm long.

SOUTH AFRICAN BLOW-FLIES DAS)

Biology

According to Ullyett (1950) Chrysomya chloropyga, C. albiceps, and
Lucilia have a distinct advantage over C. regalis in interspecific competition for
food, as the total growth period of the latter species is longer than any of the
above-mentioned species. He also states that C. regalis is the least well adapted
to withstand the adverse conditions engendered by larval overcrowding.
Because of its larger size and considerably longer period of growth, combined
with the essentially limited nature of available carrion in the field, it is not well
favoured for the production of large populations of adults, as is possible in the
case of Lucilia. This statement does not seem to be altogether true, as in three
carcasses of large mammals examined during the present investigation (particu-
larly that of an eland) very large populations of this fly were produced, with a
fairly low rate of mortality, and in all the samples taken (some of which
comprised up to 8 000 larvae) dwarfing was not observed; although a fairly
large number of eggs was produced by C. albiceps, none survived. Apart from
this, C. regalis is a first-wave blow-fly and, by the time the C. albiceps adults
were attracted, large numbers of C. regalis larvae had already hatched, and due
to their vigorousness they managed to compete for survival both intra- and
interspecifically. C. regalis may, however, be subordinate to C. chloropyga,
also a first-wave blow-fly and, as in the case of the latter, appears to be
attracted to fairly fresh meat, but has not been found during these surveys in
association with other blow-flies except C. albiceps.

The larval lifespan in all cases examined was about 11 days during late
summer (compare with C. chloropyga, in which it was 6-9 days (summer) and C.
croceipalpis, with a larval span of 8-9 days (summer)). The pupal stage lasts
about 9 days during summer and about 14 days during spring (C. chloropyga 5-8
days and C. croceipalpis about 14 days (early summer) to 20 days (late winter)).

This blow-fly seems to be restricted to the autumn months as most of the
infestations occurred during March and April, the fly being almost absent
during the summer months, which is in agreement with the findings of Smit
(1931). The puparia are usually formed under the carcass or in the soil without
the formation of cocoons.

Chrysomya megacephala (Fabricius)
Adult (Fig. 1D-F)

Various stages previously described by Zumpt (1965) and Prins (1979).
Colour metallic greenish; easily separated from our other South African species
(albiceps, chloropyga and regalis) by the blackish-brown anterior thoracic
spiracles (white in the other species). In females there is no separation between
facets of the upper and lower halves of eyes but in males lower, smaller facets are
clearly separated from larger ones in upper half of eyes (Fig. 1F). Otherwise very
similar to C. regalis, but lacking dark costal area of the wings. Length
8,9-10 mm.

214 ANNALS OF THE SOUTH AFRICAN MUSEUM

Widely distributed in Oriental, Madagascan, Australasian and Palaearctic
regions; recently also found in Brazil, Senegal, and the west coast and Durban
areas of South Africa.

Larva

First Instar (Fig. 3Q-S)

Very similar to last-stage larva except for size and shape of posterior
spiracles (Fig. 3S), which have two elongate slits touching each other ventrally.
Peritreme somewhat stronger developed on ventrolateral side and rather simi-
lar to that of Calliphora croceipalpis. Cephalopharyngeal skeleton (Fig. 3Q—-R)
also similar to that of first instar of C. croceipalpis; however, mouth-hooks
broader in lateral view and with definite short dorsal projections. Central piece
connecting the two cornuae broader than in C. croceipalpis; the sclerite when
seen from above not so pointed in front as in that species.

Second Instar (Fig. 3P, T)

Very similar to mature larva, except for absence of anterior lateral
swellings, which are present on segments 5-8 in mature larvae. Anterior
spiracles with eight to nine small branches. Posterior spiracles (Fig. 3T) similar
to those of C. croceipalpis but smaller, slits oval and elongate. Cephalopharyn-
geal skeleton (Fig. 3P) with basal piece similar to that of third instar larva;
anterior dorsal bridge complete and with short posterior projections; mouth-
hooks in lateral view narrower than those of C. croceipalpis and directed
slightly upwards; posteriorly a small, almost acute projection also present.
Ventral bridge of middle piece more arcuate and therefore more clearly visible
laterally than in C. croceipalpis. Small dental sclerite present on each side.

Third Instar (Figs 3L, O, 4B, 5B, E)

Very similar to that of C. regalis, measuring 13,5-15 mm in length.
Anterior spinose girdles present on all segments, those on segments 2-8
complete, that on segment 9 interrupted laterally. Segment 10 without dorsal
spines; segment 11 with posterior girdle; segment 12 with ventral band and
some spines around anus. Segments 5-8 with anterior lateral spinose swellings
similar to those of C. regalis. First spinose girdle on second segment with dorsal
spines strongly developed, lateral ones small and inconspicuous; ventral ones
somewhat better developed. Head divided as in C. regalis. Antennae fairly
conspicuous. Posterior face of abdomen hollowed out as in latter species, the
fleshy tubercles surrounding spiracular plates and those on each side of anus
also as in that species. Spines on body integument mostly bi- or tridentate,
including some smaller ones, others spoon-shaped as in C. regalis.

Anterior spiracles with 8-10 short branches (Zumpt (1965) gives the
number as 11-13); posterior yellowish brown spiracles (Fig. 30) similar to
those of C. regalis but smaller, projections not so conspicuous as in that species
and distance between spiracles about half or slightly more than half diameter of
spiracle.

SOUTH AFRICAN BLOW-FLIES DAS

025

D

Fig. 5. A. Posterior view of mature larva of Chrysomya regalis. B. Posterior view of mature

larva of C. megacephala. C. Left anterior spiracle of mature larva of C. chloropyga.

D. Posterior view of mature larva of C. chloropyga. E. Right anterior spiracle of mature
larva of C. megacephala.

Cephalopharyngeal skeleton (Fig. 3N) similar to that of C. regalis, but
basal part of oral sclerite smaller and not V-shaped in most specimens seen;
dorsal outline of mouth-hooks in lateral view more convex, the concave
posterior area smaller. Anterior dorsal bridge complete and with short poster-
ior projection on each side, almost rounded in front in some specimens.

Puparium

Colour reddish brown to mahogany brown, with anterior spiracle fan-
shaped and yellow. Rather similar to that of C. regalis. but oval, rugulose areas
absent on sides. Ridge around the spiracular plates formed by surrounding
tubercles not so conspicuous as in latter species and spiracles much smaller, their
diameter being about one-fifth diameter of posterior area surrounded by
tubercles (in C. regalis diameter of spiracles is about one-fourth diameter of

216 ANNALS OF THE SOUTH AFRICAN MUSEUM

posterior area). As in the latter, two oval depressions are present below
spiracles, which may be confluent in some specimens. Respiratory horns very
small as in C. regalis and in most specimens examined anterior part of puparium
is somewhat broader and shorter when seen from above than in latter species.

Biology

Adult flies were observed mostly during the late summer and early
autumn and were attracted to both mammal and bird carcasses as well as to
human excrement. The eggs (Fig. 3M) are very similar to those of C. croceipal-
pis but are smaller, measuring about 1,38 x 0,33 mm, and fairly shiny, with the
reticulation very superficial and visible only under certain light conditions. As
in C. croceipalpis, the ledges are narrow, close together and extend for almost
the whole length of the egg. The incubation period is short, only about 14 hours
ale ZO AC,

In breeding tests in the laboratory the duration of the first larval instar was
about 23 hours, during which period they reached a length of 3-4 mm; the
second instars lasted about 21 hours at 26 °C. When the second moult occurs
they are usually about 6 mm long and by this time the spiracles of the last instar
are clearly visible through the integument. After reaching the third instar
(about 44 hours after hatching), they feed for 60-72 hours and then burrow into
the soil to pupate. The total larval lifespan at 26 °C was 140-148 hours in the
laboratory. According to Wijesundara (1957) the life cycle from egg to adult
takes 204 hours in Sri Lanka; in Cape Town it varied from 276 to 306 hours, the
pupal stage lasting 136-144 hours.

KEY FOR IDENTIFICATION OF LARVAE

The following key will assist in the identification of the full-grown larvae
of the six common blow-flies described above.

f. Warvae with fleshy processes on most'segments.- = 522-024. 32-- 2565" Chrysomya albiceps
— Larvae without fleshy processes on most segments. .

2. Posterior spiracles with closed peritremalinme (Fie. 20) 2 362 4. sae) oa ee 3)
— Posterior spiracles with open peritremal ring.

3. Posterior spiracles large in relation to posterior face (Fig. 5A), the projections between

slits, particularly the one between the inner two slits, distinct. Dorsal spines of girdles on at
least anterior half of body broadly rounded, not cleft (Fig. 3H). Segments 2-11 with
complete spinose girdles (Fig. 4C). Mouth-hooks with small accessory oral sclerite
PV ESCItee Sra OE eee Meson ta ai cpa ie AP peo 1) ni arr set nn ha ecete Chrysomya regalis
— Posterior spiracles smaller in relation to posterior face (Fig. 5B, D), the projections
between slits, particularly the one between the inner two slits, indistinct. Most of dorsal
spines of girdles on at least anterior half of body bi- or tridentate (Fig. 3L). Segments 2-9
with complete spinose eirdlesi 2) 3.2... $28 deal SOLS ee 6 ee 4
4. Segments 2-7 and sometimes 8 with obviously complete spinose girdles (Fig. 4A).
Mouth-hooks devoid of accessory oral sclerite (Fig. 4D), and anterior spiracles with 10-12
branches(Picw 5 @)n ence ee recite seie Negetee sic Chrysomya chloropyga
— Segments 2-8 and sometimes 9 with obviously complete spinose girdles (Fig. 4B).
Mouth-hooks with small accessory oral sclerite (Fig. 3N), and anterior spiracles with 8-10
short branches:(F1e. SE) ae ae hoses ee ae See eon ha Chrysomya megacephala
Mouth-hooks with accessory oral sclerite (Fig. 21) ................ Calliphora croceipalpis
— Mouth-hooks without accessory oral sclerite (Fig. 4E) .................. Lucilia sericata

er

SOUTH AFRICAN BLOW-FLIES DAG

ACKNOWLEDGEMENTS

I wish to express my thanks to Dr V. B. Whitehead of the South African
Museum for his advice and suggested improvements, and to Mr V. Branco of
the same institution for drawing Figure 1A-—C.

REFERENCES

Porter, A. 1924. Notes on some insect larvae that may occur in man in South-Africa. S. Afr. J.
Sci. 21: 373-377.

Prins, A. J. 1979 The discovery of the oriental latrine fly, Chrysomyia megacephala (Fab.)
along the south-western coast of South Africa. Ann. S. Afr. Mus. 78: 39-47.

RykE, P. A. J. 1969. Ekologie. Potchefstroom: Pro Rege.

Smit, B. 1929. The sheep blow-flies of South Africa. Bull. Dep. Agri. Un. S. Afr. 47: 1-27.

Smit, B. 1931. A study of the sheep blow-flies of South Africa. Rep. vet. Res. Un. S. Afr. 7:
299-421.

Utiyetr, G. C. 1950. Competition for food and allied phenomena in sheep blow-fly
populations. Phil. Trans. R. Soc. (B) 234: 77-174.

WIJESUNDARA, D. P. 1957. The life history and bionomics of Chrysomyia megacephala (Fab.).
Ceylon J. Sci. (B) 25: 169-185.

ZumMPT, F. 1956. CALLIPHORIDAE I: CALLIPHORINI AND CHRYSOMYIINI. Explor. Parc. natn. Albert
Miss. G. F. de Witte 87: 1-200.

ZumPT, F. 1965. Myiasis in man and animals in the Old World. London: Butterworths.

ABBREVIATIONS
ant antenna mpp middle piece or
as anterior spiracle hypostomal sclerite
bu button os accessory oral sclerite
db anterior dorsal bridge pab parastomal bar
dc dorsal cornua per peritremal ring
ds dental sclerite sl slit
mh mouth-hook tu tubercle

mp maxillary palp vc ventral cornua

6. SYSTEMATIC papers must conform to the /nfernational code of zoological nomenclature
(particularly Articles 22 and S51).

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 Seine: of prefixed surnames in all ignores. Shen 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

(c) Scientific names, but not their vernacular derivatives
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.

A. J. PRINS

MORPHOLOGICAL AND BIOLOGICAL NOTES
ON SIX SOUTH AFRICAN BLOW-FLIES
(DIPTERA, CALLIPHORIDAE)

AND THEIR IMMATURE STAGES

QH

1 .

S67X [5 OCTOBER 1982 0 eee ISSN 0303-2515
a oh po,

NH

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
(c) Table of contents giving hierarchy of headings and subheadings
(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 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, 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 eft 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.

Konn, 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 90 Band
October 1982 Oktober
Part 5 Deel

REVISION OF ANCYLOCERAS BIPUNCTATUM
SCHLUMERe le 72
(CEPHALOPODA, AMMONOIDEA) AND
DISCUSSION OF THE VALIDITY, PHYLOGENY
AND LIMITS OF THE GENUS
NEANCYLOCERAS SPATH, 1926

By
HERBERT CHRISTIAN KLINGER

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, AOS, 5-8), 402, “5, 8, to), SCS, 5, 5).
GG eset) EY, & CED, 7), 11003)).
11(1=2 5, 7, t=p.i.), 1545), 240), 27, 3103), 3205), 332 36): 45a)

EDITOR/REDAKTRISE

Ione Rudner

Copyright enquiries to the South African Museum

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 041 9

Printed in South Africa by In Suid-Afrika gedruk deur
he WRustica, Press; Pty) ide Die Rustica-pers, Edms., Bpk.,
Court Road, Wynberg, Cape Courtweg, Wynberg, Kaap

REVISION OF ANCYLOCERAS BIPUNCTATUM SCHLUTER, 1872
(CEPHALOPODA, AMMONOIDEA) AND DISCUSSION OF THE
WACIDITY: PHYLOGENY AND LIMITS‘ OF THE GENUS
NEANCYLOCERAS SPATH, 1926

By
HERBERT CHRISTIAN KLINGER

South African Museum, Cape Town

(With 10 figures)

[MS accepted 27 July 1982]

ABSTRACT

The syntypes and topotype material of Ancyloceras bipunctatum Schliiter, 1872, the type
species of the genus Neancyloceras Spath, 1926, are described and illustrated photographically.
Schluter’s figures are misleading, being partially based on idealized reconstructions. Even if the
effects of post-depositional deformation are taken into consideration, it appears that coiling in
Ancyloceras bipunctatum as here interpreted may vary from open ancyloceratid to aspinocera-
tid or crioceratitid. Neancyloceras does not merit separation from Exiteloceras Hyatt, 1894, nor
do Axonoceras Stephenson, 1941, or Exicrioceras Anderson, 1958. Affinities of Exiteloceras
with Neocrioceras Spath, 1921, and Pseudoxybeloceras Wright & Matsumoto, 1954, are
discussed and a possible common origin in the Turonian is postulated. Species previously
referred to Neancyloceras are reviewed.

CONTENTS

PAGE
BRERA CLIO Ten Pues <<. eaters eta eden rete athe, ae eRe, Serena te esc) e hee 219
USCC IN ATIC CESCHIP LION 5 inks, site aersets <a ee es ee tS Re een 221
MP CHMIMION OT ENE SPECIES 2. = as ac San SS eee ae ee ee i 229
Mee MIC ASLAUUS Es outa hey fe ei ae SRE ee te es GR ee 232
Sreinzana phylosenyike vss kt oe oo eee eee. ee ote te 234
PSTD OL Seep ee A ge I Are eB aR ge Cea Ohne eRe Peep EE pr 235i
PREKMOWICCOEMENTS. 222M 2 Sue ss os eee ee eRe RT een: ba ease 23h
Ree CRE MCE Sate hn hee AT ane chee GEO: IEE RC dae re aR ee ae 238

INTRODUCTION

Two or, possibly, all three of Schluter’s figured syntypes of Ancyloceras
bipunctatum (Schliiter 1872: 98, pl. 29 (figs 1-3)) (Fig. 1 (1-3 herein)) as well
as a number of other possible syntypes and topotype material were located in
the collections of the Geologisch-Palaontologisches Institut und Museum der
Georg-August Universitat, Gottingen. Examination of Schltter’s original and
topotype material allows better definition of the species and discussion of the
validity, origin, phylogeny, and limits of the genus Neancyloceras Spath, 1926,
of which it is the type species. |

29

Ann. S. Afr. Mus. 90 (5), 1982: 219-239, 10 figs.

ANNALS OF THE SOUTH AFRICAN MUSEUM

220

Se aon

septacainge

By AGH Bares ee ek ee

nee

ad

ahhh. init sibial ibal tide:
Mb : ny '

4

1s

inal plate 29. x 0,67.

Copy of Schliiter’s (1872) orig

fe

Fig.

ANCYLOCERAS BIPUNCTATUM AND NEANCYLOCERAS Zak
SYSTEMATIC DESCRIPTION

Ancyloceras bipunctatum Schliter, 1872

Figs 1 (1-3), 2-8 A-E, 9

Ancyloceras bipunctatum Schliiter, 1872: 98, pl. 29 (figs 1-3) (erroneously spelt bipunctum in
explanation to plate figures).

Ancyloceras bipunctatum Schliiter: Wegner, 1905: 210. Nalivaiko, 1936: 35, pl. 16 (fig. 39).
Michailov, 1951: 88, pl. 16 (figs 66-71). Pasternak, 1954: 157, text-fig. p. 158.

Neancyloceras bipunctatum (Schliiter): Naidin, 1959: 182, pl. 3 (fig. 6). Hancock, 1961: 30.
Giers, 1964: 283. Blank er al. 1974: 169.

Type

The lectotype, herein designated, is the specimen figured by Schliter
(1872, pl. 29 (fig. 1)) from the Campanian of Ahlten, near Hanover, West
Germany, housed in the collections of the Geologisch-Palaontologisches Insti-
tut und Museum der Georg-August Universitat, Gottingen, GPIG Orig. 65-10.

Material

The paralectotype figured by Schliiter (1872, pl. 29 (fig. 3)) (catalogue
number GPIG Orig. pending), a specimen tentatively identifiable with the third
figured specimen in Schliiter (1872, pl. 29 (fig. 2)) (GPIG Orig. 65-11), and
forty-six other specimens, all from the same locality and in the same collec-
tions, were available for study. Schluter’s descriptions were based on five
specimens in the collections at Gottingen, and on a sketch of one of the
specimens in the collection of Mr Witte of Hanover. The latter collection was
subsequently presented to the above-mentioned Institute in GOttingen (see
Schliiter 1876: 181 footnote) and incorporated in their collections. Apart from
the lectotype and paralectotype(s), it is impossible to determine from the
labelling which specimens were Schltter’s original syntypes, and which
belonged to the Witte collection.

Descriptions
Lectotype GPIG Orig. 65-10 (Fig. 2A)

Schliter’s illustration of the lectotype is reversed and much restored. Only
half of the outer whorl and less than a quarter of the inner whorls are actually
preserved as internal moulds. The rest of the specimen is indicated only by
brown (?limonitic) stained impressions. The double row of ventral tubercles in
the actual specimen is by no means as pronounced as in the lithograph, nor are
both rows as clearly exposed as the figures suggest. The strongly flared ribbing
(as indicated on the lower half of Schliiter’s figure) is also idealized. The last
part of the whorl seems to detach itself from the open spiral. The last septum is
in the last half of this whorl.

DDD. ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 2. Exiteloceras bipunctatum (Schliter, 1872).
A. The lectotype, figured in reverse by Schliiter (1872, pl. 29 (fig. 1)) GPIG Orig. 65-10.
B. A large, slightly curved body chamber fragment with differentiated ribbing. GPIG-8.
Both specimens from the Upper Campanian of Ahlten, West Germany. x 1.

ANCYLOCERAS BIPUNCTATUM AND NEANCYLOCERAS 223

Paralectotype GPIG Orig. pending (Figs 3, 4A)

According to Schliiter (1872, pl. 29 (fig. 3)) the figure of the paralectotype
was based on a sketch of a complete specimen in the possession of Mr Witte of
Hannover. Part of the shaft and open hook in the figure can be identified with a
crushed specimen (GPIG Orig. pending). The loosely coiled inner whorls of the
sketch on which the lithograph was based are very similar to those of GPIG—2
(Fig. 5A), but this specimen does not belong to the same individual as GPIG
Orig. pending, nor do any of the other available specimens. It must be assumed
either that the initial inner whorls of paralectotype GPIG Orig. pending are
lost, or that Witte’s sketch from which Schluter’s lithograph was constructed
was, in fact, a synthetograph based on GPIG Orig. pending and GPIG-—2 or
another specimen similar to it such as GPIG—3 (Fig. 6). Whatever the case may
be, Schliter’s illustration of the shaft and hook is misleading. Ribbing is not
uniform throughout as the illustration suggests, but stronger ribs separated by a
variable number of normal ribs are present. These major ribs are bituberculate,
like the normal ribs. At the adapical end of the shaft the major ribs are
separated by about eight normal ribs, but towards the hook the major ribs
become more prominent, their tuberculation less distinct, and intermediaries
decrease to as little as two. Traces of the last septa are visible near the lower
end of the straight shaft.

?Paralectotype GPIG Orig. 65-11

A slightly crushed specimen, GPIG Orig. 65-11 (Fig. 5B—D) is of the same
size as the third specimen figured by Schltiter (1872 pl. 29 (fig. 2)), but due to
lack of documentation the author is unable to confirm whether or not this is
indeed the figured specimen in its unrestored form. None of the available
specimens shows the perfect bilateral symmetry as seen in Schluter’s figures,
suggesting that asymmetrical whorl section and ornament were the norm rather
than the exception, as will be discussed further.

Topotype material

Two specimens, GPIG-—2 (Fig. 5A) and GPIG-—3 (Fig. 6) show the early,
irregular crioceratitid coiling as seen in the lectotype. GPIG—3 is the most
complete and has a somewhat irregular mode of coiling. Three-quarters of the
outer whorl is occupied by body chamber. The last part of the body chamber
appears to uncoil slightly and then recoil again. Only the last rib is thickened.
GPIG-2 (Fig. 5A) has a virtually circular inner whorl as found in the lectotype,
but a body chamber that uncoils and then recoils similar to that of GPIG-3.
Here only the last half whorl is body chamber, and thickened ribs do not occur.

All the specimens are deformed to some extent; however, it does appear
that the whorl section and ornament are asymmetrical in the majority, with
tuberculation displaced towards one side. This suggests that the inner phragmo-
cone whorls were coiled in a low, perhaps irregular or elliptical helix rather
than in one plane. At the smallest whorl diameter preserved (6 mm), the

224

ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 3. Exiteloceras bipunctatum (Schluter, 1872).
The paralectotype, GPIG Orig. pending upon which
Schliiter’s (1872, pl. 29 (fig. 3)) reconstruction was partially
and incorrectly based. From the Upper Campanian of
Ahlten, West Germany. X 0,77.

ANCYLOCERAS BIPUNCTATUM AND NEANCYLOCERAS

eee
Dae re 5

gh

&,

Fig. 4. Exiteloceras bipunctatum (Schliter, 1872).
A. The paralectotype, GPIG Orig. pending upon which Schliiter’s (1872, pl. 29
(fig. 3)) reconstruction was partially and incorrectly based. B. Topotype spe-
cimen GPIG-—5; a straight body chamber fragment. C. Topotype specimen
GPIG-12 showing shallow, wide constriction.
All specimens from the Upper Campanian of Ahlten, West Germany. A xX 0,77;
B-C x 1.

ANNALS OF THE SOUTH AFRICAN MUSEUM

226

‘T x ‘AUBULIOD ISO ‘UayY Jo uetueduiey soddy oy} wos suauiseds yOg
‘((Z 3) 67 Id
‘Z/Q] JOIM[YOS 99S) [[-S9 “B8UO OIdH edAjoyojered ajqissog ‘q-q ‘winjdas a]qIsiA jse] 0} s}uIod MOIIW *7-DJ[dH usunoods odkjodoy “yv
‘(ZLET ‘lomnqyos) wnywjaundig svsaoojanxy °S “si4

ANCYLOCERAS BIPUNCTATUM AND NEANCYLOCERAS 22

Fig. 6. Exiteloceras bipunctatum (Schliter, 1872). Topotype specimen GPIG-3. Arrow points
to last visible septum. From Ahlten, West Germany. x 1.

228 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 7. Exiteloceras bipunctatum (Schliter, 1872).
A-C. Topotype specimen GPIG-9. D. Straight body chamber fragment GPIG-—10.
Both specimens from the Upper Campanian of Ahlten, West Germany. x 1.

ANCYLOCERAS BIPUNCTATUM AND NEANCYLOCERAS 229

whorls are still coiled, and do not show straight or otherwise aberrant early
coiling. GPIG—16 (Fig. 8D—-E) shows labeceratid coiling but this is probably
due to post-mortem deformation.

On the phragmocone, rib density is generally between four and five per
maximum whorl diameter, but may be as high as eight. Some specimens, e.g.
GPIG-12 (Fig. 4C) and GPIG-—13 (Fig. 9B), show slight irregularities in ribbing
which may be interpreted as shallow constrictions. The sharp-crested, slightly
rursiradiate, bituberculate and even ribbing 1s characteristic of the major part
of the shell. Stronger ribs appear only on the last part of the phragmocone and
on the body chamber in some specimens.

Straight and curved body chamber fragments can be identified. GPIG—5
(Fig. 4B), GPIG-10 (Fig. 7D), GPIG-—11 (Fig. 9D), and GPIG-—17 (Fig. 9A)
are straight or slightly curved body chamber fragments identifiable with the
open ancyloceratid paralectotype, GPIG pending (Figs 3, 4A). GPIG-—11 (Fig.
9D) is of interest in not having major ribs. GPIG-8 (Fig. 2B) and GPIG—9 (Fig.
7A-C) are large, curved body chamber fragments; the former has distinct
differentiated ribbing. GPIG-—17 (Fig. 9A) is a body chamber fragment much
larger than any of the other specimens.

Indistinct traces of trifid lobes are present on the lower ends of the
paralectotype GPIG pending and GPIG-3.

DEFINIRION OF THE SPECIES

As can be seen from the description of the original and topotype material,
there is considerable variation in the coiling, ornament, and size of Ancyloceras
bipunctatum. Even if the effects of the post-mortem deformation are taken into
consideration, coiling of the body chamber ranges from distinctly open ancylo-
ceratid to possibly aspinoceratid or crioceratitid. Unmistakable ancyloceratid
forms include the paralectotype GPIG pending (Figs 3, 4A), and parts of
straight shafts such as GPIG-—5 (Fig. 4B), GPIG—7 (Fig. 9C), GPIG-—10 (Fig.
7D), and GPIG-11 (Fig. 9D). The last septum in the paralectotype GPIG
pending is located in the lower end of the straight shaft. In the lectotype
GPIG—Orig. 65-10 (Fig. 2A) half of the outer whorl is body chamber, whereas
in GPIG-3 (Fig. 6) three-quarters of the outer whorl is body chamber. This,
together with the presence of such large body chamber fragments as GPIG-8
(Fig. 2B) and GPIG-9 (Fig. 7A—C), suggests crioceratitid or aspinoceratid
coiling for some specimens rather than distinct open ancyloceratid coiling as
suggested by Schluter’s reconstructed figure.

The relationship between these different modes of coiling is obscure. It is
partially due to post-mortem deformation, but may also be a manifestation of
dimorphism or, less probably, extreme intraspecific variation due to the fact
that the material is from more than one stratigraphic horizon. Schliiter (1876:
248 and footnote) had already expressed doubts on whether or not the material
from Ahlten all belonged to one zone.

ANNALS OF THE SOUTH AFRICAN MUSEUM

230

‘| x ‘AuRulIod jsomy ‘UdI[YY Jo uetueduieg sisddyQ ay) wos suaumoads [Ty ‘¢{[-S9 “3UG O[dH AdAjouow Aq adAjojoy ayy, ‘(ZL8T
‘IOINYIS) snidnasajui (SvsdIOUa]OSvADG) SvAIIOJagAxXOpNnasg ‘H-A ‘UOHeULIOJOp 0} oNp SuTjlod pyeissaqe] yWM usuTIDads ‘g]-NI[gqHN ‘(ZL81
‘TaIN[YS) wWnywjIundig svdgdojaNxXY “A-C ‘pI-OldH ‘wouwse1y suosowsesyd jeoiddy (ZL8] ‘Joinpyos) wnywjoundig spsadojanx_y “D-V “8 ‘314

ANCYLOCERAS BIPUNCTATUM AND NEANCYLOCERAS Dn

Fig. 9. Exiteloceras bipunctatum (Schluter, 1872).

A. Gigantic body chamber fragment GPIG—17. B. Body chamber fragment with differentiated
ribbing and shallow constriction. GPIG—-13. C. Slightly curved body chamber fragment
GPIG-7. D. Straight body chamber fragment GPIG-11.

All from the Upper Campanian of Ahlten, West Germany. x 1.

V3 ANNALS OF THE SOUTH AFRICAN MUSEUM

On the basis of the material available, it is not possible to give a definitive
answer to these questions, and all the specimens are included in the same
species. Specimens figured by Michailov (1951: 88, pl. 16 (figs 66-71)) as
Ancyloceras bipunctatum also suggest an open, crioceratitid or aspinoceratid
mode of coiling in some specimens, although a straight fragment was also
figured.

From the descriptions above, Ancyloceras bipunctatum may be defined as
probably being dimorphic. The shell probably has an initial, loosely coiled,
shallow helix with similar ribbing, and tubercles displaced slightly to one side.
Wide, shallow constrictions may occur. This is followed in smaller specimens by
an open, ancyloceratid shaft and hook, and in the larger specimens by irregu-
lar, crioceratitid or aspinoceratid coils. Major ribs occur variably on the last
part of the phragmocone and on the body chamber.

GENERIC STATUS

Spath (1926: 80) erected the genus Neancyloceras merely by citing Ancy-
loceras bipunctatum Schltter, 1872, as type species and added that ‘this stock of
the mucronata zone is obviously distinct from Oxybeloceras Hyatt, with which
the writer (loc. cit. Zululand, 1921, p. 254) had formerly united it’. Later,
Spath (1953: 16) added that ‘Oxybeloceras and Neancyloceras are distinguished
by their regular and very sharp costation’ and (Spath 1953: 17) ‘Ancyioceras (?)
pseudoarmatum Schliter ... 1s probably a somewhat homoeomorphous devel-
opment of Neancyloceras’.

Wright (1957: L227) interpreted Neancyloceras as differing ‘from open-
whorled species of Glyptoxoceras in less regular coiling and bituberculate
periphery’. Comparisons with Glyptoxoceras as suggested by Wright are super-
ficial, as no true glyptoxoceratid is bituberculate, and coiling is by no means as
regular as is claimed. Comparisons with Neoglyptoxoceras Collignon can be
ignored on the same grounds.

Interpreted in terms of the type species as here defined, Neancyloceras
bears closer affinity to the Upper Cretaceous genera Exiteloceras Hyatt, 1894,
Axonoceras Stephenson, 1941, and Exicrioceras Anderson, 1958. Differences
between these taxa are slight and it is doubtful if they even merit generic
separation. All generally have simple bituberculate ribbing.

Axonoceras Stephenson, 1941, (type species Axonoceras multicostatum
Stephenson, 1941) is best known from Texas (Stephenson 1941), ?California
(Anderson 1958), Angola (Haas 1943), and Madagascar (Collignon 1971). All
specimens are small—less than 50 mm in diameter. Axonoceras is identified
mainly by the irregular coiling of the inner whorls, that is, uncoiling and
recoiling, resulting in a polygonal outline and irregular spaces between succes-
sive whorls. Stephenson (1941: 422) initially regarded the genus as having
planispiral coils, but Haas (1943: 9) found that the Angolan species A.
angolanum Haas was coiled helically. Most ribs in Axonoceras are bitubercu-

ANCYLOCERAS BIPUNCTATUM AND NEANCYLOCERAS 233

late, but non-tuberculate intermediaries do occur in the Texan species. In
addition, Collignon (1971: 13) reported the occurrence of occasional stronger
ribs in A. multicostatum ellipticum Collignon. Apart from the uncoiling and the
recoiling of the inner whorls, the general shell shape of Axonoceras seems to be
that of a low helix or loosely coiled spiral. Straight shafts and recurved hooks
are unknown.

Scott & Cobban (1965) provided a reconstruction of the type species of
Exiteloceras, E. jennyi. Gill & Cobban (1966: A32) and Scott & Cobban (1970:
D73) defined the species as ‘an aberrant ammonite that has juvenile whorls as
straight limbs connected by semicircular bends, and later whorls loosely coiled in a
plane without contact between adjacent whorls. Ornamentation consists of moder-
ately coarse ribs, each of which terminates in a node at the margin of the venter’.

The irregularly coiled inner whorls of Axonoceras are compatible with the
straight early limbs of Exiteloceras. Ornament and coiling in later whorls in
both genera are similar, and it seems a reasonable procedure to regard
Axonoceras as a junior synonym of Exiteloceras, as tentatively suggested by
Wiedmann (1962: 198), and advocated by Matsumoto (1967: 340), Lewy
(1969: 123), and Klinger (1976: 76).

Lewy (1969) described two new species of Exiteloceras from the Late
Campanian of Israel, E. unciforme and E. etegense. The first shows scaphitoid
coiling (uncoiling or recoiling) of the body chamber, and the latter irregular
tuberculation on the venter, analogous to that found in Neocrioceras s.s. Spath,
1921. (Matsumoto & Morozumi (1980: 18), however, regard FE. eteqense and
other species described by Lewy as possibly representing a new genus allied to
Neocrioceras (Schlueterella) Wiedmann, 1962.) Wiedmann (1962: 206) also
referred an uncoiled body chamber fragment from the Upper Campanian of
Spain to Exiteloceras.

Thus interpreted, Exiteloceras cannot satisfactorily be separated from
Neancyloceras as represented by the type species Ancyloceras bipunctatum.
Furthermore the irregular crioceratitid or aspinoceratid coiling in some Ancy-
locerus bipunctatum is similar to that found in Exicrioceras Anderson, 1958,
type species Exicrioceras ortigalitense Anderson, 1958. The scaphitoid coiling of
the body chamber of Exiteloceras unciforme Lewy easily connects the ancylo-
ceratid coiling of some Ancyloceras bipunctatum with the closer, crioceratitid
coiling of typical Exiteloceras. The only real difference between Neancyloceras
and Exiteloceras as interpreted above seems to be the differentiation of ribbing
near or on the body chamber, but it is doubtful whether this merits separation,
even at subgeneric level. If they were to be separated on this basis, Axonoceras
multicostatum ellipticum Collignon with differentiated ribbing on the inner
whorls would also have to be separated, thereby just increasing the list of
heteromorph taxa based on insignificant differences.

Given that there appears to be no real difference between Evxiteloceras,
Axonoceras, Exicrioceras, and Neancyloceras, it is probably best to refer them
all to the oldest available name, Exiteloceras.

234 ANNALS OF THE SOUTH AFRICAN MUSEUM

Unfortunately detailed stratigraphic data are lacking, but it would be
interesting to see the relationship between the ancyloceratid (‘Neancyloceras’
pars), ellipsoceratid (‘Exicrioceras’), scaphitoid (Exiteloceras unciforme) and
closer-coiled crioceratitid forms (e.g. Exiteloceras jennyi). Is there a trend
towards recoiling, as observed in numerous heteromorph groups (see e.g.
Wiedmann 1969) or is this merely part of intraspecific or generic variation or
dimorphism?

ORIGIN AND PHYLOGENY

Matsumoto (1967: 339-40) considered it possible to derive Exiteloceras and
Axonoceras from Nostoceras through widening of the apical angle of the helix.
Indeed, forms such as ‘Bostrychoceras polyplocum Roemer ? var. doneziana
Michailov’ (Michailov 1951: 53, pl. 4 (figs 23-24)), which are coiled in a loose
helix and with bituberculate ornament, are remarkably similar to the phragmo-
cone whorls of Exiteloceras bipunctatum. Similarly, the early straight whorls of
forms such as Didymoceras cf. D. nebrascense (Lewy 1969: 116, pl. 1 (fig. 2))
are comparable to the early whorls of Exiteloceras jennyi.

As far as coiling and ornament are concerned, there are a number of
features in common between Exiteloceras, Neocrioceras, and Pseudoxybelo-
ceras that appear to be too related to be considered as merely homoeomorphic
development (as tentatively suggested by Spath 1953: 17), and the author
would rather derive these taxa from a common ancestor than from the contem-
poraneous nostoceratid forms as advocated by Matsumoto (1967).

Neocrioceras Spath, 1921, as interpreted by Wiedmann (1962: 205) com-
prises two subgenera, Neocrioceras s.s.,, type species Crioceras spinigerum
Jimbo, 1894, with crioceratitid coiling, simple ribbing and irregular tubercula-
tion over the venter; and N. (Schlueterella), type species Ancyloceras pseudo-
armatum Schliiter, 1872. N. (Schlueterella) is a rather heterogenous group, but
includes ancyloceratid to polyptychoceratid forms in which the tubercles are
situated either on stronger or on looped ribs. Generally Neocrioceras s. |. can
be separated from Exiteloceras by the presence of four rows of tubercles in the
former but only two in the latter, but, as will be seen in Pseudoxybeloceras, the
presence or absence of lateral tubercles may be of low systematic value. Points
of similarity between Exiteloceras and Neocrioceras are: early helical coils occur
in some Exiteloceras and in some Neocrioceras, e.g. Neocriceras cf. spinigerum
(Spath 1921: 52, pl. 7 (fig. 6)); ancyloceratid coiling occurs in some Exiteloceras
bipunctatum as well as N. (Schlueterella) pseudoarmatum (Schltter), although
admittedly the latter already tends towards polyptychoceratid coiling; and-
irregular ornament occurs in both Exiteloceras eteqense Lewy and N. (Neocrio-
ceras) spinigerum (Jimbo).

Closer resemblance is to be found in the genus Pseudoxybeloceras Wright
& Matsumoto, 1954, as interpreted by Klinger (1976: 75), Matsumoto &
Morozumi (1980), and (in essence) Ward & Mallory (1977), which includes

ANCYLOCERAS BIPUNCTATUM AND NEANCYLOCERAS D5

Christophoceras Collignon, 1969 (type species Christophoceras ramboulai
Collignon, 1969) and Parasolenoceras Collignon, 1969 (type species Parasoleno-
ceras splendens Collignon, 1969) as subgenera. The early whorls of
Pseudoxybeloceras are unknown; thus far only J-shaped fragments have been
found, but there is a distinct trend towards polyptychoceratid coiling in the
later stages. Pseudoxybeloceras s.s. is quadrituberculate on every rib in the
adult stage, but Matsumoto (1977: 346) has recently shown that the early stages
of Pseudoxybeloceras (P.) quadrinodosum (Jimbo) lack the lateral tubercles,
and that ornament in this respect is similar to that of Exiteloceras. Pseudoxybel-
oceras (Parasolenoceras) is apparently bituberculate throughout, like the juve-
nile stages of Pseudoxybeloceras (Pseudoxybeloceras) quadrinodosum, but
already has distinct polyptychoceratid coiling in the early stages of growth.
Pseudoxybeloceras (Christophoceras) has similar coiling and _ bituberculate
ornament on the phragmocone, but develops major ribs on the body chamber
and acquires lateral tubercles on these. The bituberculate ornament in P.
(Parasolenoceras) and P. (Christophoceras) is analogous to that found in
Exiteloceras in general, whereas differentiation of ornament on the body
chamber of P. (Christophoceras) is similar to that of Exiteloceras bipunctatum.

The essential ingredients for the ‘Bauplan’ of these predominantly San-
tonian-Campanian genera Fxiteloceras, Neocrioceras, and Pseudoxybeloceras
were already present in the Turonian.

Parts of the early whorls of Exiteloceras bipunctatum are indistinguishable
from the predominantly Turonian genus Allocrioceras Spath, 1926. Spath
(1926: 81) had already noticed this, but regarded it as ‘only superficial resem-
blance’. Allocrioceras consists mainly of helically coiled forms with undiffer-
entiated, bituberculate ribbing. Differentiated ribbing does, however, occur in
forms such as Allocrioceras cuvieri (Schluter, 1872) and A. turoniense (Schliter,
1872). Ancyloceras paderbornense Schltter (1872: 97, pl. 30 (figs 1-2)) already
has distinct Neocrioceras (Schlueterella) ornament. Hamites multinodosus
Schliiter (1872: 106, pl. 32 (figs 1-2)) is part of a straight body chamber
fragment, and shows ornament similar to that of Christophoceras.

Ward & Mallory (1977, test-fig. 2) show a lineage starting with Neocrio-
ceras and Pseudoxybeloceras in the Turonian, trending towards more polypty-
choceratid coiling through ‘Cyphoceras’ Ward & Mallory, 1977 (= P. (Christo-
phoceras) and P. (Parasolenoceras)) to Solenoceras in the Maastrichtian. Exite-
loceras first appears in the Upper Campanian, and it seems unlikely that its
Origins can be traced directly to Neocrioceras or Pseudoxybeloceras of the
Turonian—Coniacian. Matsumoto (1959: 162) originally regarded Exiteloceras
and Pseudoxybeloceras as having evolved in parallel or sister relationship from
a ‘plastic genus Hyphantoceras’.

Exiteloceras was probably derived from Pseudoxybeloceras s.s. in parallel
with P. (Christophoceras) and P. (Parasolenoceras) in the Campanian (Fig. 10).
In the latter two subgenera the trend is towards acquisition of polyptychocera-
tid coiling and reduction of the helical stage, whereas Exiteloceras retains the

236 ANNALS OF THE SOUTH AFRICAN MUSEUM

[= 4 Exiteloceras
=
W
<x Solenoceras
:
U i)
Exiteloceras

‘Axonoceras’ unciforme ‘Neancyloceras’ /
74
<q
=
<q
U
za 4
fe)
[=
2 ay
q
WP |

N. (Schlueterella)
a Q
VU
<
Zz
‘e)
U
Pseudoxybeloceras
2 | iy
Neocrioceras s.s.

1e)
[34
=
—

Allocrioceras

Fig. 10. Inferred phylogenetic relationship of Exiteloceras. Part of the diagram after Ward &
Mallory (1977, text-fig. 2).

ANCYLOCERAS BIPUNCTATUM AND NEANCYLOCERAS 237),

helical or crioceratitid stage to greater diameters and may possibly have solved
hydrodynamic requirements by trending towards planispiral coiling.

LIMITS

Several species that have been referred to ‘Neancyloceras’ probably do not
belong there (see e.g. Naidin 1959; Wiedmann 1962; Blank et al. 1974), e.g.
Hamites wernicket Wolleman, 1902, Ancyloceras retrorsum Schliiter, 1872,
Hamites interruptus Schliter, 1872, and Hamites phaleratus Griepenkerl, 1889.

Ancyloceras retrorsum lacks ventral tubercles throughout and is apparently
a large glyptoxoceratid, as examination of Schluter’s types has shown, a view
supported by Atabekian & Khakimov (1976: 61) and Blank et al. (1974: 168).

Hamites interruptus is thus far known only from small recurved fragments.
The holotype (Fig. 8F—G) is still septate at the larger end, thus indicating that it
is not a body chamber hook. The possibility cannot be excluded that this
represents the early whorls of a form similar to Exiteloceras jennyi, but because
of the polyptychoceratid coiling it is probably best placed in Pseudoxybeloceras
(Parasolenoceras). Blank et al. (1974: 167) place it in Solenoceras.

Hamites wernickei is difficult to interpret. The small specimen with two
shafts in contact, figured by Wolleman (1902, pl. 4 (fig. 5)) under that name,
was previously regarded (Klinger 1976: 73) as better placed in Solenoceras.
Should that specimen be conspecific with the larger hooks figured under that
name by Wolleman (1902, pl. 4 (fig. 4), pl. 5 (figs 1-2)) or Pervinquiére
(1907: 86, pl. 3 (fig. 33)), the species may also be referred to Pseudoxybelo-
ceras (Parasolenoceras).

Collignon (1971: 11, pl. 644 (fig. 2380)) described a small straight fragment
from the Maastrichtian of Madagascar as Neancyloceras ambindense. In that
specimen only every third rib is tuberculate. Collignon’s generic allocation
seems correct.

Blank et al. (1974: 169) refer Hamites phaleratus to Neancyloceras, but the
figures of Griepenkerl (1889, pl. 11 (fig. 3), pl. 12 (figs 3-4)) suggest coiling to
be polyptychoceratid, which would place it closer to Pseudoxybeloceras (Para-
solenoceras) than to Exiteloceras.

Spath (1953: 49) mentions Neancyloceras from Angola, but this material
has never been described, and the generic allocation cannot be verified.

ACKNOWLEDGEMENTS

I thank Dr S. Ritzkowski of the Geologisch-Palaontologisches Institut und
Museum der Georg-August Universitat, GOttingen, for access to the collections
and for the loan of material on which these descriptions are based. Financial
aid from the Deutsche Forschungsgemeinschaft, and from the Alexander von
Humboldt-Stiftung during the tenure of a research fellowship at Tiibingen is
gratefully acknowledged. I am grateful to Prof. Dr J. Wiedmann (Tubingen)

238 ANNALS OF THE SOUTH AFRICAN MUSEUM

for useful discussions. Dr W. J. Kennedy (Oxford) read the manuscript and
suggested several constructive changes. Mr W. Wetzel (Tiibingen) kindly
photographed the specimens.

REFERENCES

ANDERSON, F. M. 1958. Upper Cretaceous of the Pacific Coast. Mem. geol. Soc. Am. 71:
1-378.

ATABEKIAN, A. A. & KuHakimov, F. K. 1976. (Campanian and Maastrichtian ammonites of
Central Asia) (in Russian). Tridy Inst. Geol. Dushanbe: 1-146.

BLANK, M. YA., KRrimHoLTs, G. YaA., Natpin, D. P. & SAvTScHINSKAJA, O. V. 1974. (Atlas
Upper Cretaceous fauna Donbassa) (in Russian). Moscow.

CoLLIGNoN, M. 1969. Atlas des fossiles caractéristiques de Madagascar (Ammonites) XV
(Campanien inférieur). Tananarive: Service Géologique.

CoLLIGNoN, M. 1971. Atlas des fossiles caractéristiques de Madagascar (Ammonites) XVII
(Maastrichtién). Tananarive: Service Géologique.

Giers, R. 1964. Die Grossfauna der Mukronatenkreide (unteres Obercampan) in Ostlichen
Miunsterland. Fortschr. Geol. Rheinld. Westf. 7: 213-294.

GILL, J. R. & CosBan, W. A. 1966. The Red Bird Section of the Upper Cretaceous Pierre
Shale in Wyoming. Prof. Pap. U. S. Geol. Surv. 393-A: Al—A73.

GRIEPENKERL, O. 1889. Die Versteinerungen der Senonen Kreide von K6nigslutter im Herzog-
thum Braunschweig. Paldont. Abh. Berlin 4: 1-117.

Haas, O. 1943. Some abnormally coiled ammonites from the Upper Cretaceous of Angola.
Am. Mus. Novit. 1222: 1-17.

Hancock, J. 1961. The Cretaceous System of Northern Ireland. Q. Jl. geol. Soc. Lond. 117:
11-36.

Hyatt, A. 1894. Phylogeny of an acquired characteristic. Proc. Am. Phil. Soc. 32: 349-647.

Jimpo, K. 1894. Beitrage zur Kenntniss der Fauna der Kreide-formation von Hokkaido.
Paldont. Abh. Berlin 6: 149-194.

KLINGER, H. C. 1976. Cretaceous heteromorph ammonites from Zululand. Mem. geol. Surv.
Rep. S. Afr. 69: 1-142.

Lewy, Z. 1967. Some Late Campanian nostoceratid ammonites from southern Israel. Israel J.
Earth Sci. 16: 165-173.

Lewy, Z. 1969. Late Campanian heteromorph ammonites from southern Israel. Israel J. Earth
Sci. 18: 109-135.

Matsumoto, T. 1959. Upper Cretaceous ammonites of California II]. Mem. Fac. Sci. Kyushu
Univ. (D). Special Volume 1: 1-172.

Matsumoto, T. 1967. Evolution of the Nostoceratidae (Cretaceous heteromorph ammonites).
Mem. Fac. Sci. Kyushu Univ. (D) 18: 331-347.

Matsumoto, T. 1977. Some heteromorph ammonites from the Cretaceous of Hokkaido. Mem.
Fac. Sci. Kyushu Univ. (D) 23: 303-366.

Matsumoto, T. & Morozumti, Y. 1980. Late Cretaceous ammonites from the Izumi Mountains,
Southwest Japan. Bull. Osaka Mus. nat. Hist. 33: 1-31.

MicnaliLov, N. P. 1951. (The Upper Cretaceous ammonites of the southern part of the USSR
and their significance for zonal stratigraphy (Campanian, Maastrichtian)) (in Russian).
Triidy Inst. geol. NAUK. USSR 129, geol, ser. 50: 1-143.

NaipIn, D. P. 1959. In: Mosxvina, M. M., ed. (Atlas Upper Cretaceous southern Caucasus and
Crimea, Cephalopoda) (in Russian). Moscow.

NALIVAIKO, L. E. 1936. Macrofauna gorishnokredovich pokladiv pivdennoi okraini donbasi.
Kiev.

PASTERNAK, S. I. 1954. Ancyloceras bipunctatum Schliter iz Maastrichta Volino-Pdolskoi pliti.
Geol. Sb. L’vov 1: 157-169.

PERVINQUIERE, L. 1907. Etudes de Paléontologie Tunisienne. II. Cephalopodes des terrains
secondaires. Carte Géol. Tunisie. Direct. Gén. Trav. Publ.: 1-438.

SCHLUTER, C. 1871-7. Die Cephalopoden der oberen deutschen Kreide. Palaeontographica 21:
1-24 (1871); 25-120 (1872); 24: 121-262 (1876).

ANCYLOCERAS BIPUNCTATUM AND NEANCYLOCERAS 239

Scott, G. R. & CosBan, W. A. 1965. Geologic and biostratigraphic map of the Pierre Shale
between Jarre Creek and Loveland, Colorado. Miscellaneous Geologic Investigations. Map
1439.

Scotr, G. R. & CosBan, W. A. 1970. Occurrence of the Late Cretaceous ammonites
Didymoceras stevensoni (Whitfield) and Exiteloceras jennyi (Whitfield) in Delaware. Prof.
Pap. U. S. Geol. Surv. 700-D: D71-—D76.

SpaTH, L. F. 1921. On the Upper Cretaceous Ammonoidea from Pondoland. Ann. Durban
Mus. 3: 39-56.

SpaTH, L. F. 1926. On new ammonites from the English Chalk. Geol. Mag. 63: 77-83.

SpaTH, L. F. 1953. The Upper Cretaceous cephalopod fauna of Graham Land. Sci. Rep. Falkld
Isl. Depend. Surv. 3: 1-60.

STEPHENSON, L. W. 1941. The larger invertebrate fossils of the Navarro Group, Texas. Bull.
Univ. Tex. Bur. econ. Geol. Tech. 4101: 1-641.

Warp, P. & MALLory, S. M. 1977. Taxonomy and evolution of the lytoceratid genus
Pseudoxybeloceras and relationship to the genus Solenoceras. J. Paleont. 51: 606-618.
WEGNER, T. 1905. Die Granulatenkreide des westlichen Munsterlandes. Z. dt. geol. Ges. 57:

112-231.

WIEDMANN, J. 1962. Ammoniten aus der Vascogotischen Kreide (Nordspanien). Phyllocera-
tina, Lytoceratina. Palaeontographica 118A: 119-237.

WIEDMANN, J. 1969. The heteromorphs and ammonoid extinction. Biol. Rev. 44: 563-602.

WoLLEMAN, A. 1902. Die Fauna der Liineburger Kreide. Abh. preuss. geol. Landesanstalt 37:
1-129.

Wricut, C. W. 1957. In: Moore, R. C., ed. Treatise on invertebrate paleontology. Part L.
Mollusca, Cephalopoda. Boulder and Lawrence: Geological Society of America and
University Kansas Press.

WriGut, C. W. & Matsumoto, T. 1954. Some doubtful Cretaceous ammonite genera from
Japan and Saghalien. Mem. Fac. Sci. Kyushu Univ. (D) 4: 107-143.

Me # ,
iJ
q =
: )
; ?
{
oy Ly
= ‘i I 7 ”
av
t ~
. Spo
a
i "4 s i E c
7 i
} i i
1 tt a"
\
: 1
: me
nf
y/
i
i i us :
a
, ZT ‘
i
>| ‘ My
ti y
P =
¢
y
‘
= 1
3)
a C
: ;
H
'
i
f
.
| é
39
i
r u 4
=
{
2
i
{
(ne
%
¢
y - i h *
i #
ca eS j
bo €
| . a -
F ' — 5 7
; y -
7 *
U3 ’ a)
I Y 1

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

(c) Scientific names, but not their vernacular derivatives
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.

HERBERT CHRISTIAN KLINGER

REVISION OF ANCYLOCERAS BIPUNCTATUM
SCHLUTER, 1872

(CEPHALOPODA, AMMONOIDEA) AND
DISCUSSION OF THE VALIDITY, PHYLOGENY
AND LIMITS OF THE GENUS
NEANCYLOCERAS SPATH, 1926

-~VOLUME 90 PART 6 MARCH 1983 ISSN 0303-2515

ee gt
1

|. S67X
ce ae
|

| f w/ :
|
Pe Lh 4 he iC}
: i ; te ae ee

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
(c) Table of contents giving hierarchy of headings and subheadings
(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 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, 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
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.

Konn, A. J. 196056. 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 90 + +#£4Band
March 1983 Maart
Part 6 Deel

vy,
D> S
Alig, WS
vio now

CRETACEOUS FAUNAS FROM ZULULAND
AND NATAL, SOUTH AFRICA
THE AMMONITE SUBFAMILY
BARROISICERATINAE BASSE, 1947

By
WILLIAM JAMES KENNEDY,
CLAUD WILLIAM WRIGHT
&
HERBERT CHRISTIAN KLINGER

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

is AOS O)), OD, C5). Os eo Ce), Si 7D),
Gil, Baas), HOD), 8, SOD, 1), MOS).
11(1-2, 5, 7, t.-p.i.), 15(4-5), 24(2), 27, 31(1-3), 32(5), 33, 36(2), 45(1)

EDITOR/REDAKTRISE
Ione Rudner

Copyright enquiries to the South African Museum

Kopieregnavrae aan die Suid-Afrikaanse Museum

ISBN 0 86813 045 1

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

CRETACEOUS FAUNAS FROM ZULULAND AND NATAL,
SOUTH AFRICA
THE AMMONITE SUBFAMILY BARROISICERATINAE BASSE, 1947

By
WILLIAM JAMES KENNEDY,

CLAUD WILLIAM WRIGHT
Geological Collections, University Museum, Oxford

&

HERBERT CHRISTIAN KLINGER

South African Museum, Cape Town

(With 51 figures)

[MS accepted 23 September 1982]

ABSTRACT

Representatives of the subfamily Barroisiceratinae are locally common in the Coniacian of
Zululand. The following are described: Reesideoceras lornae (van Hoepen, 1968), Forresteria
(Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907), to which all previously named
South African F. (Forresteria) and many other species are referred, F. (F.) madagascariensis
(Collignon, 1965), F. (F.) cf. hobsoni (Reeside, 1932), Yabeiceras orientale Tokunaga &
Shimizu, 1926, Y. cf. orientale, Y. transiens sp. nov., Y. ankinatsyense Collignon, 1965, Y.
costatum Collignon, 1965, Y. manasoaense Collignon, 1965, Y. aff. manasoaense, Y. crassiorna-
tum sp. nov., Y. cobbani sp. nov. and Yabeiceras sp. indet.

Itwebeoceras van Hoepen, 1968, is shown to be a synonym of Reesidites Wright &
Matsumoto, 1954; Zumpangoceras Basse, 1947, Collignonella van Hoepen, 1957, Basseoceras
van Hoepen, 1968, Eedenoceras van Hoepen, 1968, and Neokanabiceras Collignon, 1965, are
synonyms of Forresteria (Forresteria), and it is suggested that Harleites Reeside, 1932, is a
senior synonym of Reesideoceras Basse, 1947.

CONTENTS

PAGE
MIREROCUCHI ON s 8 ten some reenter enc eae 241
WOCAtON,OhSDECIMENS.4- 4. | ae ae eo 242
RCIA OCAlItTS See eee eects, ak ra a eel Certain 242
DIMENSIONSIOMSHECIMIENS eee hae ee 242
SutunestenminOlogy nea. a5 ee ee ee eee 243
Systematic palacontolopys jane eee erers ee 243
PXCKNOWIECMSEMICINES) oo iy fs atay eres yore oe eee SYD
INCLCR ENCES ERG re oe Chee aE ee ee ee 322

INTRODUCTION

The Barroisiceratinae are a small subfamily of the Collignoniceratidae
derived from late Collignoniceratinae, perhaps via Subprionocyclus Shimizu.
They vary from compressed and involute to evolute and inflated, and either

241

Ann. S. Afr. Mus. 90 (6), 1983: 241-324, 51 figs.

242 ANNALS OF THE SOUTH AFRICAN MUSEUM

have strong to weak ribs arising singly or in groups from bullae, or not.
Ventrolateral and siphonal clavi are generally developed at some stage during
ontogeny, the latter sometimes fusing into a keel. A lateral tubercle develops in
some taxa and several lose virtually all ornament on the body whorl, which may
be either tabular or compressed with fastigiate or tabulate venter. The sutures
are generally simple with short, moderately incised saddles.

Within several genera large and small species with modified body chambers
occur and are in some cases afforded subgeneric status. This may represent an
as yet poorly perceived dimorphism: the evidence is still inconclusive.

The group as a whole differs from the ancestral Collignoniceratinae in
having only one rather than two rows of ventrolateral nodes. On this basis
Reesidites Wright & Matsumoto, 1954, belongs to this subfamily, which thus
ranges from Upper Turonian through the Coniacian. The subfamily has a wide
geographic distribution, and most members are taken as stratigraphic indicators
of the Coniacian. As is clear from recent discussions by Hancock & Kennedy
(1981) and Matsumoto (in Matsumoto et al. 1981: 63) amongst others, the true
stratigraphic distribution of species and genera is still unsettled. The present
firmly dated records from Zululand are thus of more than local significance.
Added to this, the local abundance of one genus, Forresteria Reeside, 1932,
allows a discussion of ontogenetic development and intraspecific variation that
clarifies several taxonomic problems.

LOCATION OF SPECIMENS

The following abbreviations are used to indicate the repositories of the
material studied:
BMNH British Museum (Natural History)
SAM South African Museum, Cape Town
SAS South African Geological Survey, Pretoria
USNM _ National Museum of Natural History, Washington, D.C.

FIELD LOCALITIES

Details of localities mentioned in the text are given by Kennedy & Klinger
(1975); fuller descriptions of sections are deposited in the Palaeontology
Department of the British Museum (Natural History), London; Geological
Survey, Pretoria; and the South African Museum, Cape Town.

DIMENSIONS OF SPECIMENS

All dimensions are given in millimetres:

D = diameter, Wb=whorl breadth, Wh=whorl height, U = umbilical
diameter; c and ic refer to costal and intercostal measurements respectively.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 243

SUTURE TERMINOLOGY

The suture terminology of Wedekind (1916), reviewed by Kullman &
Wiedmann (1970) is followed here:
I = internal lobe, U = umbilical lobe, L = lateral lobe, E = external lobe.

SYSTEMATIC PALAEONTOLOGY

Superfamily ACANTHOCERATACEAE de Grossouvre, 1894
Family Collignoniceratidae Wright & Wright, 1951
Subfamily Barroisiceratinae Basse, 1947

Genus Reesidites Wright & Matsumoto, 1954
(= Itwebeoceras van Hoepen, 1968)

Type species

Barroisiceras minimum Hayasaka & Fukada (1951: 325 (ex Yabe 1925,
nom. nud)), from the Upper Turonian of Japan, by the original designation of
Wright & Matsumoto (1954: 130).

Diagnosis

Involute, compressed, flat-sided sinuous ribs arise in groups of two or three
from variably developed umbilical bullae. Shorter intercalated ribs are inserted
on the middle to outer flank and all ribs bear strong ventrolateral clavi from
which they project forward to prominent siphonal clavi. Ribs broaden and
flatten with increasing age.

Suture with markedly asymmetric E/L in the type species at least.

Discussion

Reesidites has been discussed at length by Wright & Matsumoto
(1954: 130), Obata (1965), and Matsumoto (1965: 61). There seems little doubt
that Reesidites minimus and R. elegans Matsumoto & Inoma (1971: 139, pl. 23
(figs 1-3), text-figs 5-7) are derived from an involute collignoniceratid, prob-
ably Subprionocyclus (see e.g. Reyment 1975). The ornament of the two
genera differs only in the presence of both inner and outer ventrolateral
tubercles in the latter. The suture of Subprionocyclus (e.g. Matsumoto 1965,
text-figs 28-30) already shows the asymmetry of E/L, that is so distinctively
developed in Reesidites.

The ornament of Reesidites minimus differs little from that of Barroisiceras
haberfellneri (von Hauer) (compare Fig. 1A and Fig. 1B—D), being a little more
flexuous in Reesidites and projected strongly forward on the ventrolateral
shoulder rather than straight. The sutures differ more obviously; in B. haber-
fellneri there is no comparable asymmetry (see Fig. 2). However, the type
specimen of /twebeoceras lornae van Hoepen, 1968, described below, has the
ornament of Reesidites without sutural asymmetry. Taken together these obser-

244 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 1. A. Barroisiceras haberfellneri (von Hauer, 1866). Geological Survey of Austria Collections
3764, from the Gosau Beds of Gams near Hieflau Austria. (Original of Von Hauer 1866: 30, pl. 1
(figs 1-2).) x1. B-D. Reesidites minimus (Hayasaka & Fukada, 1951). B. University of Kyushu
Collections H4089G bis, from the Upper Turonian of the Ikushumbets, Hokkaido. x1. C-D.
Holotype, Hokkaido University Collections, from the Upper Turonian of the Ikushumbets. x1.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 245

L
NeVAvae

Se
E

TAR

E L

2 8a,

Cc \
:

D

Fig. 2. A-C. Reesidites lornae (van Hoepen, 1968). Sutures of the holotype SAS Z1128.

D. Barroisiceras haberfellneri paeon (Redtenbacher, 1873). Sutures of a specimen in the

Geological Survey of Austria Collections 3481, from Ofenwald near Strdébl-Weissenbach,

Austria. (Original of Redtenbacher 1873: 103, pl. 23 (fig. 3c-e).) E. Reesidites minimus

(Hayasaka & Fukada, 1951). Sutures of the holotype. (After Matsumoto 1965: 65, text-fig. 36.)
All x2.

246 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 3. Reesidites lornae (van Hoepen, 1968). Holotype SAS Z1128. x1.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 247

vations suggest that Reesidites may best be treated as a subgenus or even strict
synonym of Barroisiceras. However, until the type material of B. haberfellneri
is redescribed and its stratigraphic position clarified, Reesidites and Barroisi-
ceras are here maintained as separate taxa.

Itwebeoceras van Hoepen, 1968 (type species Itwebeoceras lornae van
Hoepen, 1968), from the ?Upper Coniacian of Zululand, has the same
ornament as Reesidites, differing only in detail from that of R. minimus. It does
not show the same asymmetry of E/L but is here regarded as a synonym.
Buenoceras Etayo-Serna (1979: 101; type species B. loboi Etayo-Serna
(1979: 101, pl. 14 (fig. 2), text-figs 9R, U)) from the Coniacian of Colombia is
based on a fragment only and may also be a synonym.

Occurrence

Upper Turonian of Japan and Armenia; Coniacian of Colombia, Vene-
zuela, and Zululand.

Reesidites lornae (van Hoepen, 1968)

Figs 2A-C, 3-4.
Itwebeoceras lornae van Hoepen: 1968a: 184, pl. 4.

Holotype

By monotypy, SAS Z1128, from the St. Lucia Formation, (Coniacian
271V), locality 13 of Van Hoepen (1968a, 1968b) = locality 73 of Kennedy &
Klinger (1975).

Dimensions
D Wb Wh Wb: Wh U
c. 74,5(100) 21,0(—) 31,5(—) 22,7(—)
46,5(100) 14,1(30,3) 20,0(43,0) 0,71 13,2(28,3)
36,6(100) i GOs3) 16,0(43,7) 0,69 10,0(27,3)
Description

The holotype is a wholly septate internal mould, somewhat corroded on
the ventral region of the outer whorl; the estimated maximum diameter is
approximately 75 mm.

The smallest diameter at which the specimen can be examined is 17 mm
(Fig. 4). At this size the coiling is rather involute, half the previous whorl being
covered. The whorl section is compressed (whorl breadth to height ratio is 0,8),
with the greatest breadth at the umbilical bullae. The flanks are flattened and
convergent, the ventrolateral shoulders narrowly rounded, and the venter
fastigiate. There are twelve prominent umbilical bullae per whorl; these give
rise to pairs of low, broad prorsiradiate ribs strengthened into an oblique
ventrolateral clavus from which they sweep forward across the ventrolateral
shoulder, declining as they do so. There is an initially entire siphonal keel

248 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 4. Reesidites lornae (van Hoepen, 1968). Holotype SAS Z1128. x2.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 249

Fig. 5. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907). A-C. A para-
type of ‘Basseoceras krameri’ van Hoepen, 1968, SAS Z1437. D-E. Another paratype, SAS
Z978. All X1.

250

ANNALS OF THE SOUTH AFRICAN MUSEUM

, 1907). Holotype of ‘Basseoceras krameri’ van Hoepen, 1968, SAS Z935.

évenin

Z

Fig. 6. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Th

—

CRETACEOUS FAUNAS FROM SOUTH AFRICA Wes\\\

flanked by faint, shallow grooves, which subsequently develops low crenula-
tions, each of which corresponds to a rib. As size increases, ribbing and
tuberculation becomes increasingly differentiated. At a diameter of 36 mm the
coiling has become a little more evolute (umbilicus is 27,3 % of diameter) and
higher-whorled (whorl breadth to height ratio of 0,69). There are fifteen
medium-sized umbilical bullae per whorl. These give rise to pairs of low,
relatively broad, flexuous prorsiradiate ribs that sweep forward across the inner
flank, are feebly convex across the mid-flank and feebly concave across the
outer flank. Occasional intercalated ribs are inserted around mid-flank and all
the ribs develop a small, conical ventrolateral tubercle. These give rise to broad
prorsiradiate extensions of the ribs that sweep forward to link with elongate
siphonal clavi borne on the siphonal keel. There are thirty-eight ribs per whorl
at this diameter, corresponding to a slightly smaller number of ventral clavi.

This style of ornament extends to the greatest diameter preserved where
there are seventeen to eighteen umbilical bullae per whorl and a total of
forty-two ribs.

The suture line is shown in Fig. 2A—C. Elements are relatively simple and
little subdivided, with a broad, slightly asymmetric bifid E/L, narrower bifid L,
narrow, slightly asymmetric L/U,, small U, and broad, simple U,/U3.

Discussion

Reesidites lornae resembles R. minimus (see Matsumoto 1965: 63, pl. 14
(fig. 1), pl. 15 (figs 1-3), text-figs 34-39) but differs in being more evolute, with
stronger bullae, broader whorls, narrower, better differentiated ribs, and an
essentially symmetrical rather than asymmetric E/L. R. elegans Matsumoto &
Inoma (1971: 139, pl. 23 (figs 1-3), text-figs 5-7) is much more delicately and
flexuously ribbed, with more secondary ribs. R. subtuberculatus (Gerhardt)
(1897: 156, pl. 3 (fig. 12)) has far fewer ribs, lacks bullae, and has a broader
venter without obvious ribs but rather with striae between ventrolateral and
siphonal clavi.

There is a striking similarity between R. lornae and ‘Schlénbachia (Gau-
thiericeras)’ crioceratiformis Liithy (1918: 43, pl. 3 (fig. 2)) from the Coniacian
of Peru. The outer whorl of the wholly septate holotype of this species uncoils,
perhaps due to pathological disturbance of growth, and further material is
needed before placing /ornae in synonymy with crioceratiformis.

Occurrence
St. Lucia Formation of Zululand. Precise age unknown but certainly
Coniacian and possibly Coniacian IV.

Genus Forresteria Reeside, 1932

Type species
Barroisiceras (Forresteria) forresteri Reeside (1932: 17, pl. 5 (figs 2-7)), by
the subsequent designation of Wright (1957: L432) = Acanthoceras (Prionotro-

Dy ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 7. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907). Slender variants. A-C.

SAM-D1187G. D-F. ‘Barroisiceras (Harleites) castellense’ Reeside, 1932. The holotype USNM 73758

from the Mancos Shale, some 61 m (200 ft) above the top of the Ferron Sandstone in Sevier County,
Utah. G. SAS Z913, large septate fragment. All x1.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 253

pis) alluaudi Boule, Lemoine & Thévenin (1907: 32, pl. 8 (figs 6-7), text-fig.
17).

Diagnosis

Both small species (adult at 80-100 mm) and large ones (adult at 250 mm)
are known. Coiling varies from involute to moderately evolute, the whorls from
compressed to depressed. Early whorls bear weak umbilical bullae, strong to
weak lateral tubercles or spines linked by primary ribs, the latter giving rise to
pairs of weak to strong secondary ribs which, with intercalatories, are linked to
variably developed ventrolateral and siphonal clavi. Ornament may be lost at
small diameters, with a smooth or feebly ornamented lanceolate or tabulate
venter on the body whorl, or persist, with the tubercles declining only at
maturity. The suture is simple with a broad E, narrow E/L and L, with large U,
in many species.

Occurrence

Coniacian of France, Austria, Czechoslovakia, Israel, west Africa, Zulu-
land, Madagascar, Japan, Colombia, Mexico, and the United States (Utah and
Wyoming).

Subgenus Forresteria (Forresteria) Reeside, 1932

(= ?Zumpangoceras Basse, 1947: 144; Collignonella van
Hoepen, 1957: 350 (pro Collignoniceras van Hoepen, 1955:
361), non Breistroffer, 1947: unpaged; Basseoceras van
Hoepen, 1968b: 162 (non Collignon, 1965: 73); Eedenoceras
van Hoepen, 19685: 171; Neokanabiceras Collignon, 1965: 42).

Diagnosis

Variable, early whorls as in Forresteria sensu lato. In compressed variants
the umbilical and lateral tubercles are lost at an early growth stage (‘Basseo-
ceras van Hoepen) leaving feeble flank ribs, ventral and siphonal clavi, and a
high angular whorl section that persists to over 200 mm. These grade into
massively whorled hypernodose individuals with persistent ribs and tubercles.
The lateral and ventrolateral tubercles fuse in some species, in others they
survive separate to the beginning of the body chamber; thereafter tubercles
decline and ribs dominate ornament.

Discussion

The type species of Forresteria (Forresteria) was based on a single juvenile
specimen (see Fig. 14E-H) from the Mancos Shale of Sevier County, Utah,
some 61 m (200 ft) above the top of the Ferron Sandstone. Reeside (1932: 17)
remarked that it ‘seems very close indeed to the form from the Senonian of
Diego-Suarez, Madagascar, described by Boule, Lemoine and Thévenin as
Acanthoceras (Prionotropis) alluaudi, which seems to the writer to be a species

254 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 8. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907). A-B. SAM—.
D1187H. C-D. SAM-D1187F. E-G. SAM-D1187G. All x1.

agp vas

CRETACEOUS FAUNAS FROM SOUTH AFRICA 255

of Barroisiceras. B. forresteri differs chiefly in its fewer ribs and nodes’. It is
believed that the types of F. (F.) forresteri and F. (F.) alluaudi are so similar
that they cannot be separated specifically, as discussed below.

Subsequent accounts of Forresteria (Forresteria) have been hampered by
the small number of specimens generally available and as a result numerous
species, subgenera, and genera have been introduced for a range of ammonites
that the large Zululand collections described here suggest belong to no more
than a single species.

Initial discussion is restricted to the type material only. The holotype of F.
(F.) forresteri occurred with a compressed ammonite of similar size and
ornament, described by Reeside (1932: 16, pl. 4 (figs 4-8)) as Barroisiceras
(Alstadenites) sevierense (the holotype, and only specimen described, is illus-
trated here, Fig. 14A—D), and an even more compressed, involute and feebly
ornamented species, also represented by only one specimen and described by
Reeside as Barroisiceras (Harleites) castellense Reeside (1932: 19, pl. 6 (figs
1-5) (see Fig. 7D-F herein). These three species, referred to three subgenera,
in fact represent no more than variable juveniles of a single species, as W. A.
Cobban has demonstrated to the authors on the basis of his unpublished new
collections. Furthermore, the new American material shows that Barroisiceras
(Forresteria) stantoni Reeside (1932: 17, pl. 7 (figs 1-7)) is yet a further variant
of the same species (Figs 15A—B, 35C-E).

This same variablity of nuclei is confirmed by the South African material
described here. Not only do no two nuclei match, but the same range from
compressed ‘castellense’ forms (e.g. Fig. 7A—C) through feebly ribbed (e.g. Fig.
11A-B) to depressed spinose juveniles (e.g. Fig. 11C—N) can be demonstrated.

Medium sized and large individuals preserve this same wide variation, and
the type material of Basseoceras van Hoepen, 1968b (type species Basseoceras
krameri van Hoepen, 19685: 164, pl. 7, (fig 2b-g), non Basseoceras Collignon,
1965) corresponds to the slender variant with feeble ornament, while Collig-
nonella van Hoepen, 1957 (= Collignoniceras van Hoepen, 1955, type species
Collignoniceras hammersleyi van Hoepen, 1955: 361, figs 7-9, non Breistroffer
1947) is no more than a large adult of the coarsely ornamented, stout-whorled
form (Figs 16A—C, 17A-C).

Size apart, there are no significant or diagnostic differences between this
adult and the holotype and only described specimen of Eedenoceras van
Hoepen, 1968b (type species Eedenoceras multicostatum, van Hoepen
1968b: 171, pl. 12, text-fig. 4b) (re-illustrated here as Figs 12A-B, 13E); it is
also a synonym of Forresteria (Forresteria).

Although specifically separable from F. (F.) alluaudi, the type species of
Neokanabiceras Collignon, 1965, N. madagascariense Collignon (1965: 42, pl.
432 (figs 1784-1786), including var. ankinatsyense (fig. 1787)), is a perfectly
good Forresteria (Forresteria) aS is apparent from Wiedmann’s recent re-
illustration of the type material (in Herm, Kauffman & Wiedmann 1979: 44, pl.
7A-B, text-figs 7D, 8). There are small umbilical bullae linked by primary ribs

256

ANNALS OF THE SOUTH AFRICAN MUSEUM

, 1907). BMNH C8333. x0,7.

évenin

Z

Fig. 9A-B. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Th

CRETACEOUS FAUNAS FROM SOUTH AFRICA DT,

Fig. 10. A-B, E-F. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907). A-B.
SAM-D1187L. E-F. SAM-Z167. C-D. Forresteria (Forresteria) madagascariensis (Collignon,
1965). SAM-D1187I. All x1.

258 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 11. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907). A-B. SAS Z2195.
C-E. SAS Z1453A. F-H. SAS Z1453B. I-K. SAS Z1438B. L-N. SAS 2946. All x1.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 259

to massive lateral nodes that give rise to pairs of ribs linking to ventral clavi,
exactly as in Forresteria; the only difference appears to be that the siphonal
clavi are raised on a low siphonal keel. Several specimens of F. (F.) alluaudi
that have been examined show this feature developed weakly at some stage in
ontogeny (e.g. Fig. 13A—B): it is regarded here as no more than a specific
difference.

Zumpangoceras Basse, 1947 (type species Zumpangoceras burckhardti
Basse, 1947: 144) was based on the ammonites described by Burckhardt
(1919: 99-108) as Barroisiceras, and later illustrated by him (1921, pl. 22 (fig.
16), pl. 23 (figs 1-2), pls 24-25). Basse originally thought this to be a subgenus
of Barroisiceras and noted the very fine ornament—especially the well-
preserved growth striae—and the presence of a lateral tubercle. Etayo-Serna
(1979: 99) has designated the original of Burckhardt (1921, pl. 23 (fig. 1))
lectotype of the species. This is a large fragment with lateral (+ ?umbilical)
tubercles and more numerous ventral and siphonal clavi linked by ribs and
dense growth striae. It appears to be a fragment of a typical Forresteria
(Forresteria), so that Zumpangoceras also falls into synonymy.

Harleites Reeside, 1932, type species by original designation Barroisiceras
haberfellneri var. harlei de Grossouvre (1894: 56, pl. 2 (figs 2, 7-8)), is based on
specimens from the basal Coniacian of the Dordogne, France, now preserved in
the collections of the Sorbonne, Paris. This was originally described as a
subgenus of Barroisiceras, but Basse (1947) noted the presence of a tiny
umbilical and inner lateral tubercle on the inner whorls of some of the type
material which, taken with the more numerous but feeble ventrolateral and
siphonal clavi, led her to treat it as a subgenus of Forresteria, a view followed
by Wright (1957). Both these authors regarded Alstadenites Reeside, 1932 (type
species, herein designated, Ammonites alstadenensis Schluter, 1876: 151, pl. 40
(figs 13-16); lectotype, herein designated, the original of Schliiter’s pl. 40 (figs
13-16)) as a synonym, a view followed here (indeed, the type species are
probably synonymous).

Matsumoto (1969: 327) has, however, afforded Harleites full generic status
(following Parnes 1964: 21) on the basis of the more involute shell and weak
lateral ornament with finer, more numerous and more persistent ventrolateral
and siphonal tubercles.

As discussed above, the compressed variants of Forresteria (Forresteria)
alluaudi in both North America and Zululand are Harleites-like in morphology;
however, what is the position of the type species?

Re-examination of De Grossouvre’s material shows that the large, almost
smooth holotype of H. harlei (e.g. De Grossouvre 1894, pl. 2 (fig. 2)) shows no
trace of tubercles on the earliest part of the outer whorl (in part, perhaps, due
to abrasion), but weak umbilical bullae appear on the body chamber (the
specimen is septate to a diameter of approximately 60 mm). The same is true of
the smaller paratype (De Grossouvre 1894, pl. 2 (fig. 8)), but there are
specimens such as that figured in De Grossouvre’s pl. 2 (fig. 7) that have an

ANNALS OF THE SOUTH AFRICAN MUSEUM

260

‘TX d-D ‘SL'0x d-V “16SZ SVS ‘A-d ‘16SZ SVS ‘G-D ‘7L6Z SVS ‘8961
‘uodoopy] UeA .Winwjsooujnu spradouapaq, Jo adAjojOH “G-V ‘(L061 ‘UUoASYL 2% WUIOWIT ‘a[nNog) 1pnvN]YY (visajsat110J) VidaJSaddOY “7 “BLA

CRETACEOUS FAUNAS FROM SOUTH AFRICA 261

Fig. 13. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907). A-B. SAS Z987.
C_D. SAS Z1438G. E. Holotype of ‘Eedenoceras multicostatum’ van Hoepen, 1968, SAS Z972.
All x1.

262 ANNALS OF THE SOUTH AFRICAN MUSEUM

ee ee ee

Fig. 14. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907). A—D. Holo-

type of ‘Barroisiceras (Alstadenites) sevierense’ Reeside, 1932, USNM 73756, from the Mancos

Shale, some 61 m (200 ft) above the Ferron Sandstone, Sevier County, Utah. E-H. Holotype

of ‘Barroisiceras (Forresteria) forresteri’ Reeside, 1932, from the same locality and horizon as ;
the original of Figure 14A—D. All x2.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 263

early stage (up to 25 mm) with tiny umbilical and stronger lateral tubercles
before the smooth harlei stage. These connect via specimens such as De
Grossouvre’s var. alstadenensis (1894, pl. 2 (fig. 4)) to the variable but
strongly ribbed and tuberculate phragmocones of typical Ammonites petroco-
riensis Coquand (1859: 995) (= Reesideoceras gallicum Basse, 1947: 133), the
type species of Reesideoceras Basse, 1947. The largest French Harleites
described retain a sharp fastigiate venter with distinct ventral and siphonal
clavi to the beginning of the adult body chamber, with renascent umbilical
bullae, as noted above. Unfortunately the venter of the holotype of harlei is
damaged so that the mature venter is not recognizable. According to the
figure of Schliiter (1876, pl. 40 (fig. 14)), the lectotype of H. alstadenensis
shows persistent ventrolateral clavi with a broadly arched flattened venter. In
the largest known Reesideoceras petrocoriensis (De Grossouvre 1894, pl. 1
(fig. 2)) the siphonal clavi are lost at the beginning of the body chamber and
the ventral clavi are raised above a flattened venter, with all tubercles lost on
the last quarter whorl.

It is concluded that the types of Harleites harlei lie in the same relationship
to the type of Reesideoceras petrocoriense and other comparably ribbed speci-
mens, as does the holotype of Harleites castellense to the holotype of Forres-
teria forresteri (=alluaudi) in the United States, or Basseoceras krameri to
Forresteria alluaudi in South Africa. Reesideoceras is thus a synonym of
Harleites, the type species of the latter being Ammonites petrocoriensis Co-
quand, 1859 = Ammonites alstadenensis Schliiter, 1876, = Barroisiceras haber-
fellneri de Grossouvre, 1894 (non von Hauer) (including varieties harlei and
alstadenensis but not desmoulinsi de Grossouvre) = Reesideoceras gallicum
Basse, 1947.

Harleites (= Reesideoceras) is regarded as a subgenus of Forresteria, distin-
guished from F. (Forresteria) on the basis of smaller adult size, fusion of
mediolateral and umbilical tubercles, and early loss of siphonal clavi leaving a
flat or concave venter on the body chamber, on the latter parts of which the
ventral clavi may also disappear.

We are not sure how many of the Harleites species described by previous
authors are referable to the subgenus as interpreted here; Japanese and
Madagascan examples may well be no more than compressed and feebly
ornamented variants of F. (Forresteria).

Because F. (Forresteria) and F. (Harleites) have very different geographi-
cal distributions the present authors do not believe them to be macro- and
microconchs.

Forresteria (Muramotoa) Matsumoto, 1969, type species by original desig-
nation F. (M.) yezoensis Matsumoto (1969: 317, pl. 42 (figs 1-2), text-figs 8-9)
is based on three specimens only from the Coniacian of Hokkaido, Japan. The
inner whorls are identical in style and proportions to those of F. (Forresteria),
but the ornament is rapidly lost at maturity, leaving a body chamber smooth
but for a faintly serrated siphonal ridge. The holotype of the type species is

ANNALS OF THE SOUTH AFRICAN MUSEUM

264

a at Sa OF A

‘LX IIV “ALSIIZ SVS ‘(S961 ‘UoUusTTOD) sis
-UAMDISVSVPDUL (V1AIJSAAAOJ) VIAIJSALAOY “Q—D “SUIWIOKA, ‘JapUR'T JO jsed-YINOS “UOHeWIOT JoNUOIA 94} WOIJ “T9LEL WNSN “ZE6I “OPIseey
duojuyys (didajJsaddOq) Sbdsaoisioing, JO IdA\OJOH “(LO6T ‘UlUsAsY, 2 SUIOWIT ‘gInog) IpnonyY (VidaJSasdO4) VIAAISAAAOY “G-V “ST ‘S14

1

265

CRETACEOUS FAUNAS FROM SOUTH AFRICA

SVS °SS6

[ ‘uodaoyH] ueA iMajsiawuvy sp1vzIUoUsyjoOD

?

SOX OZ
Jo odAjofoH “(LO6T “UtuaAgYL, 2 oUIOWOT

‘g[NOg) IpNoN]Y (V14aJSas44OJ) VIAIJSALAOY “QT “BLA

ANNALS OF THE SOUTH AFRICAN MUSEUM

266

‘LX ‘OIZ SVS ‘Sc6I ‘usdoox ueA
MasdauUuDvy spdanuousiyjo), JO sdAjojoy IY} JO sj1OYM JOUUT *(ZN6T ‘UTUSADYT, 2 CULOWOT ‘g[Nog) ipnoN]Y (V1dajSa4dO,J) VidaSaddOY *L] “BLy

CRETACEOUS FAUNAS FROM SOUTH AFRICA 267

only 75 mm in diameter and the second Japanese species, F. (M.) muramotoi
Matsumoto (1969: 320, pl. 43 (fig. 1), text-fig. 11), is equally diminutive.

If adult, Forresteria (M.) yezoensis and F. (M.) muramotoi may represent
a specialized micromorph offshoot of F. (Forresteria). That no such specimens
are known outside Japan again suggests them to be micromorph, rather than
microconch.

We cannot agree with Matsumoto (1969) that Basseoceras van Hoepen,
1968 (non Collignon, 1965) is a synonym of Muramotoa. The type species
Basseoceras krameri retains ventrolateral and siphonal clavi to a large size and
is simply a feebly ribbed, tuberculate compressed and involute variant of F.
(F.) alluaudi.

Forresteria (Forresteria) is easily distinguished from genera such as
Barroisiceras de Grossouvre, 1894 (and subgenera), Pseudobarroisiceras Shi-
mizu, 1932, and Niceforoceras Basse, 1948, on the basis of the presence of
strong lateral tubercles.

Solgerites Reeside, 1932 (= Piveteauoceras Basse, 1947) may have lateral
tubercles, but the body chamber becomes rounded with persistent siphonal
clavi in some, or strong ventrolateral nodes. It could represent a derivative of
F. (Forresteria) of the hobsoni group but, not having good material for study,
the position of the genus is unclear to the present authors.

Occurrence

Forresteria (Forresteria) is known from the Coniacian of Israel, west
Africa, Zululand, Madagascar, Japan, Mexico, Colombia, and the United
States (Utah, Colorado and Wyoming).

Forresteria (Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907)

Figs 5—9, 10A-B, E-F, 11-14, 15A-B, 16-31, 33-34, 35C-E, 40D-E

Acanthoceras (Prionotropis) alluaudi Boule, Lemoine & Thévenin, 1907: 12, pl. 1 (figs 6-7),
text-fig. 17.

Prionotropis alluaudi Boule, Lemoine & Thévenin, 1907: Lisson, 1908: 17, pl. 17. Briggen,
1910: 772. Basse, 1935: 90. Hourcq, 1936: 6.

Barroisiceras (Forresteria) alluaudi (Boule, Lemoine & Thévenin): Reeside, 1932: 12, 14.
Benavides-Cacéres, 1956: 478, pl. 61 (fig. 1).

Barroisiceras (Alstadenites) sevierense Reeside, 1932: 16, pl. 4 (figs 4-8).

Barroisiceras (Forresteria) forresteri Reeside, 1932: 17, pl. 5 (figs 2-7).

Barroisiceras (Forresteria) stantoni Reeside, 1932: 17, pl. 7 (figs 1-7), pl. 8 (figs 1-3), pl. 9 (fig. 1).

Barroisiceras (Harleites) castellense Reeside, 1932: 19, pl. 6 (figs 1-5).

Mortoniceras vinassai Venzo, 1935: 88, pl. 7 (fig. 12).

Forresteria alluaudi bit. [sic]: Basse, 1947: 128, pl. 8 (fig. 3), pl. 9, (fig. 2).

Forresteria sevierense Reeside: Basse, 1947: 131.

Harleites (?) castellensis Reeside: Basse, 1947: 140.

Collignoniceras hammersleyi van Hoepen, 1955: 361, figs 7-9.

Collignoniceras peregrinator van Hoepen, 1955: 364, figs 10-11.

Forresteria (Forresteria) alluaudi Boule, Lemoine & Thévenin: Wright, 1957: L432, fig. 551 (2).

? Forresteria cf. alluaudi (BLT) [sic]: Parnes, 1964: 23, pl. 2 (figs 7-8).

Forresteria alluaudi B.L.Th [sic]: Collignon, 1965: 76, pl. 448 (fig. 1828).

268 ANNALS OF THE SOUTH AFRICAN MUSEUM

Forresteria razafiniparyi Collignon, 1965: 78, pl. 449 (figs 1829-1831).

Forresteria costata Collignon, 1965: 80, pl. 450 (figs 1833-1834).

Basseoceras krameri van Hoepen, 1968: 164, pl. 7, text-fig. 2b-g.

Forresteria hammersleyi (van Hoepen, 1955): van Hoepen, 1968: 166, pl. 8.

Forresteria vandenbergi van Hoepen, 1968: 167, pl. 9, text-fig. 3a—b.

Forresteria itwebae van Hoepen, 1968: 169, pl. 10, text-fig. 3c.

Forresteria reymenti van Hoepen, 1968: 169, pl. 9, text-figs 3d—-e, 4a.

Eedenoceras multicostatum van Hoepen, 1968: 171, pl. 12, text-fig. 4b.

Forresteria (Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907): Matsumoto, 1969: 308,
pl. 40 (figs 1-4), text-figs 5-7. Gonzalez-Arreole, 1977: 171, text-fig. 2f-h.

Forresteria (Forresteria) armata Matsumoto, 1969: 313, pl. 41 (fig. 1), text-fig. 8.

? Zumpangoceras ospinai Etayo-Serna, 1979: 99, pl. 14 (fig. 8), text-fig. 9O-P.

Types

The smaller of the two specimens from Madagascar figured by Boule,
-Lemoine & Thévenin (1907, pl. 1 (fig. 7), text-fig. 17), is here designated
lectotype of the species.

Material

More than a hundred specimens in the SAM, SAS, and BMNH collec-
tions, from the St. Lucia Formation, Coniacian II of localities 10, 13, 92, 93,
and 145, including the holotypes of Collignoniceras hammersleyi van Hoepen,
1955, C. peregrinator van Hoepen, 1955, the holotype and syntypes of Basseo-
ceras krameri van Hoepen, 19685, the holotype of Forresteria vandenbergi van
Hoepen, 1968b, F. itwebae van Hoepen, 1968b, F. reymenti van Hoepen,
1968b, and Eedenoceras multicostatum van Hoepen, 1968).

Dimensions
D Wb Wh Wb: Wh U ut! ve
SAS Z935, holotype of
B. krameri c 107,5(100) 41,2(26.1) 54,9(51,1) 0,75 17,8(16,6)
SAS Z167 c 89 ,3(100) 37,5(42,0) 43,7(48,9) 0,86 19.5(@1k8) 5 2S esas
SAS Z1429 c 81,2(100) 35,5(43.7) 37,8(46,6) 0,94 17,9(22,0) 12 24-25
SAS Z1437 c 74 ,2(100) 30,7(41,5) 34,6(46,6) 0,89 13,8(18,6) 12 25
ic 74,2(100) 29,3(39,4) 32,2(43,4) 0,91
SAS Z978 c 93 ,5(100) == ©) 45,0(48,1) _ 18,9(20,0) 13/14 26/27
SAS Z987 c 43,0(100) 20,8(48.4) 20,2(47,0) 1,03 9,6(22,3) 9 19
ic 18,0(41,9) 18,5(43,0) 0,97
SAS Z591 45 ,4(100) 21,6(47,6) 21,0(47,6) 1,0 10,6(23,3) 13 223
18,2(40,1) 19,8(43,6) 0,92
SAS 1773 c 51,3(100) = (©) 23,0(44,8) as 13,5(26,3) = =
ic = ©) 22,3(43,5) ==
SAS Z1583, larger
specimen C 44,0(100) 24,0(54,5) 20,6(46,8) 1,16 10,9(24,8) 12 =
ic 21,4(48,6) 17,9(46,7) 1,20
SAS Z513 c 51,5(100) 33,2(64,5) 23,5(45,6) 1,41 15,3(29:7), 2B a eee
ic 27,7(53,8) 20,2(39,2) 37
SAS Z514 c 59,7(100) 36,3(60,8) 26,5(44,4) fear 18,3(32,6) 13 =
ic 31,4(52,6) 23,0(38,5) 1,36
SAS Z7124 C 92,0(100) 51,0(55,4) 42,5(46,2) 12 24,5(26,6) = =
ic 41,5(45,1) 40,3(43,8)
SAS Z1687 C 103,0(100) 66,7(64.7) 50,0(48,5) 1,33 31,5(30,6) 9 phe
ic 54,0(52.4) ey (=)
SAS Z922 holotype of
F. reymenti C 110,0(100) =e) 52,5(47,7) 27,5(25,0) 10 24
ic = ©) 50,0(45,5)
c 87,3(100) 60,5(69,3) 46 ,3(53,0) 1,31 23,0(26,3) = =
ic 50,0(57,3) 42,0(48,1) 1,19

ut'=umbilical tubercles; vi’=ventral tubercles

EE ——————————

CRETACEOUS FAUNAS FROM SOUTH AFRICA 269

D Wb Wh Wb: Wh U ut! ve

SAS Z923 c 109,0(100) 39 ,5(36,2) 50,3(46,1) 0,79 Pi NOS) 14 26/27
ic 38,0(34.9) 49 3(45,2) 0,77
c 83,5(100) 39,5(47,3) 38,5(46,1) 1,03 19,6(23,4) — —
ic 36,3(43.5) 36,3(43,5) 1,00
c — (—) 30,8(—) 27,5(—) iP
ic 26,5(—) 25,3(—) 1,05

SAM-D1187N- c 99 ,0(100) 41,0(41,4) 43 ,3(43,7) 0,95 28,0(28,3) 9/10 ==
ic 39 ,2(39.6) 40,5(40,9) 0,97

SAS Z972 holotype of

E. multicostatum c 140,0(—) — (—) 64,5(46,0) 36,5(26,1) 13 —
ic — (—) 61,5(43,9)

SAS Z16 holotype of

F. hammersleyi c 106,5(100) 60,0(56,3) 52,3(49,1) oS 28,5(26,8) 11 24
ic 56,0(52.6) 48 ,5(45,5) HIS

SAS Z16 c 150,0(100) 76,5(51,0) — (—) _ 39,0(26,0) 12 28
ic 70,5(47,0) — (—) —
€ 230,0(—) 106,0(46,0) 102,0(44,3) 1,04 65,0(28,2) 13/14 32/33
ic 96 ,0(41,7) 100,0(43,5) 0,96

ut! = umbilical tubercles; vf = ventral tubercles

Description

Previous accounts of Forresteria (Forresteria) alluaudi and other species
have been based on small numbers of specimens, and a number of species are
based on single specimens. Over a hundred specimens were available, all from
a narrow Stratigraphic interval and, in a number of cases, associated specimens
from single lines of concretions. These show that F. (F.) alluaudi is variable in
the extreme, and that there is widespread co-variance between depth and
relative diameter of umbilicus, whorl height, and strength and rate of dis-
appearance of ornament. On the basis of this material the authors place in
synonymy all the previously named South African Forresteria, all those
described from Madagascar (except F. (F.) madagascariensis previously refer-
red to Neokanabiceras), plus a number of other Japanese and North American
forms.

The compressed, involute, feebly ornamented members of the species are
represented by specimens formerly referred to as Basseoceras krameri van
Hoepen. All specimens studied here are wholly septate juveniles. The smallest
specimen seen is SAS Z2195 (Fig. 11A-B), which retains some body chamber
and has about twelve primary ribs per whorl, corresponding to twenty-two
ventral clavi. The whorl section is compressed with a breadth to height ratio of
0,72, the greatest breadth being at the mid-lateral tubercle. This is linked to
SAS Z249, which still retains a full complement of ribs and tubercles at a
diameter of 61,2 mm.

SAS Z1429 is somewhat stouter, with a breadth to height ratio of 0,87 at a
diameter of 82 mm. There are thirteen to fourteen umbilical bullae at this
diameter, and twenty-four ventral and siphonal clavi. The flank ornament is
beginning to decline, with the lateral tubercle represented by a slight strength-
ening of the ribs only from about 55 mm onwards.

The holotype of Basseoceras krameri, SAS Z935 (Fig. 6), retains the
lateral tubercle and feeble ribs at the smallest diameter visible, being only a
little more compressed than Z1429. It is wholly septate and so worn that the

ANNALS OF THE SOUTH AFRICAN MUSEUM

270

rial

CRETACEOUS FAUNAS FROM SOUTH AFRICA

‘LX ‘ZZ6Z SVS ‘896, ‘usdooH uRA Yuawidas Didajsas10J, JO IdA\O[OH “(L061 “UlUSAIYL, 2 GUIOWIT

6

a[nog) 1pnon

110 (D1daJSAAdOJ) DIAAJSALIOT “6 “BIA

Die,

ANNALS OF THE SOUTH AFRICAN MUSEUM

10ns

ty Collecti

1Versl

.

Un

, 1907). Holotype of ‘Forresteria armata’ Matsumoto, 1969,
lal oI, SI,

u

évenin

Z

from the Ikushumbets area of Hokkaido, Kyush

Fig. 20. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Th

CRETACEOUS FAUNAS FROM SOUTH AFRICA DEB)

ornament is artificially subdued. Umbilical bullae, ribs, and ventral and siphonal
clavi are retained throughout, however, and lateral tubercles might well have
originally survived to a larger diameter than at present. This is certainly the case
in SAS Z913, a very large, wholly septate fragment with a whorl height of 65 mm
that still retains low, broad ribs and subdued mid-laterals (Figure 7G).

Amongst other individuals that are relatively compressed is the figured
(Van Hoepen 19685, pl. 6) paratype of Basseoceras krameri, SAS Z1437 (Fig.
5A-C), which is strongly ribbed with distinct tubercles to a diameter of 75 mm
(costal whorl breadth to height ratio 0,87; intercostal 0,79). SAS Z978 is a
rather more evolute specimen, with U = 21 per cent as opposed to 16,6 per cent
in the holotype. It is especially well preserved (Fig. 5D-E), retaining aragonitic
shell, and is wholly septate. The lateral tubercles are distinct to a diameter of
about 60 mm but decline thereafter. SAS Z949 is similar, with stronger decora-
tion to the inner whorls but a similar decline beyond 60-65 mm diameter.

The authors have many small specimens of up to about 50-60 mm di-
ameter that show a progressive increase in strength of ornament with increasing
whorl inflation, e.g. SAS Z591 (Fig. 12C—-D).

Similar transitions at a somewhat greater diameter—up to 120 mm—link
SAS eZo7o0(Fic. 26), Z969 (Fig. 27), and 2922 (Fic. 19, the holotype of
Forresteria reymenti).

Within this range of ornament and size fall the types of Forresteria alluaudi
(Boule, Lemoine & Thévenin), F. forresteri (Reeside), F. razafiniparyi Collig-
non and F. costata Collignon, all of which can be matched in this series.

There are also specimens that show sparse ribbing—SAS Z1782 (Fig. 21)
and the holotype of F. peregrinator (van Hoepen) (Fig. 25A—B). Although at
first sight appearing distinct, they are simply a further variant of the species and
these small individuals match the inner whorls of the large specimens described
as F. vandenbergi van Hoepen (Fig. 23A-B) and F. itwebae van Hoepen (Figs
31, 40D-E). F. armata Matsumoto (Fig. 20) is an individual of this type, and
also probably a synonym.

With increasing diameter, changes in ornament suggesting the approach of
maturity are shown by a number of specimens. SAS Z979 (Fig. 29) and SAS
Z923 (Fig. 28) have inner whorls only moderately inflated, ribbed, and tubercu-
lated up to a diameter of about 65 mm, whereafter the whorls become more
evolute, lateral tubercles decline, and ribs crowd. This to a degree follows the
changes seen in feebly ribbed Basseoceras krameri variants. The more inflated
holotype of Eedenoceras multicostatum van Hoepen (Figs 12A—-B, 13E) shows
typical strong, inflated inner whorls to a diameter of about 70-80 mm; there-
after the lateral tubercles decline and one or two shorter intercalated ribs
become prominent on the last half whorl. This is essentially what is shown by
SAS Z923 but with more intercalatories. This is regarded as individual body
chamber variation in a small adult.

Contrasting with these specimens that appear adult at 120-160 mm are
specimens that remain septate to beyond this diameter; this applies to feebly

ANNALS OF THE SOUTH AFRICAN MUSEUM

274

CRETACEOUS FAUNAS FROM SOUTH AFRICA 275

ornamented specimens, e.g. SAS Z913, rather compressed and sparsely ribbed
individuals, e.g. the type of Forresteria vandenbergi, and more inflated and
strongly decorated individuals such as SAS Z1981. The most complete of these
large adults is the holotype of F. hammersleyi (van Hoepen) (Figs 16-17). This
shows intercalated ribs, singly or in pairs, as in the smaller adult holotype of
‘Eedenoceras’ multicostatum, but with traces of the lateral tubercle to the end
of the clearly adult body chamber at almost 230 mm diameter.

The sutures are variable, depending on whorl section and position relative
to ribs and tubercles, as shown in Figures 33-34.

Discussion

The description given above indicates that all the species originally
described from Zululand (Basseoceras krameri van Hoepen, Forresteria ham-
mersleyi, (van Hoepen), F. peregrinator (van Hoepen), F. vinassai (Venzo), F.
vandenbergi van Hoepen, F. itwebae van Hoepen, F. reymenti van Hoepen,
Eedenoceras multicostatum van Hoepen, and the Madagascan Forresteria raza-
finiparyi Collignon, F. costata Collignon, and F. alluaudi (Boule, Lemoine and
Thévenin)) can be matched in the present collection and that they are to be
treated as no more than variants of a single, apparently dimorphic species, for
which the name F. (F.) alluaudi has priority. It has been recognized that the
North American F. forresteri Reeside (1932: 17, pl. 5 (figs 2-7)) (Fig. 14E-H)
was also a synonym of F. (F.) alluaudi, and this is confirmed by the Zululand
material. F. forresteri co-occurs with Barroisiceras (Harleites) castellense Ree-
side (1932: 19, pl. 6 (figs 1-5)) (Fig. 7D-F herein) which is a very compressed
variant of this species.

Forresteria stantoni Reeside (1932: 17, pl. 7 (figs 1-7); pl. 8 (figs 1-3), pl. 9
(fig. 1)) (Fig. 35C—E herein) is no more than a variant of F. alluaudi as is
Barroisiceras (Alstadenites) sevierense Reeside (1932: 16, pl. 4 (figs 4-8))
(Fig. 14A—-D herein). In contrast, Forresteria hobsoni Reeside (1932: 18, pl. 9
(figs 2-4), pl. 10 (figs 1-2)) (Fig. 35A-B herein) is a quite distinct species,
discussed further below. It is more evolute and slender-whorled, and shows a
development of sparse ventrolateral horns developed by fusion of lateral and
ventrolateral tubercles, amongst other differences. Neokanabiceras madagasca-
riensis Collignon (1965: 42, pl. 432 (figs 1784-1787)) is regarded here as a
distinct species of Forresteria only, rather than meriting generic separation. It
differs from the variable F. alluaudi in being more slowly expanding and in the
development of very strong horns from the lateral tubercle with increasing
diameter. The horns bear rounded ribs that link them to pairs of rather small
transverse or oblique ventrolateral tubercles. The siphonal clavi are borne on a
low ridge. It is thus closer to F. hobsoni.

Forresteria nwalii (Reyment) (= Barroisiceras nwalii Reyment, in Offo-
dile & Reyment 1976: 61, fig. 14a-b), from the Coniacian of Nigeria, the
holotype and only known specimen, refigured here as Figure 39, is also a
distinct species allied to F. madagascariensis. It differs very obviously from

276 ANNALS OF THE SOUTH AFRICAN MUSEUM

, 1907). SAS Z1456. x 1.

évenin

4

Fig. 22. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Th

277

CRETACEOUS FAUNAS FROM SOUTH AFRICA

ULA 1B4IQUAPUDA D1AIJSIsAOL],

‘LX “€ISZ SVS ‘G-O SL‘0X ‘0LZ SVS ‘8961 ‘uedsoHy
jo odAjojoxY “G-V ‘(LO6I ‘UIUDADYL 2 UIOWAT ‘a[nog) IpnYN]D (vIUa}sad410.]) DVIdsaISasso4

(6

rh

278 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 24. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907). A-B. SAS Z949.
C-D. SAS H196/2. All x1.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 279

F. (F.) alluaudi in developing massive pinched lateral horns that bear looped
ribs connecting to ventral clavi and lacks an umbilical bulla (whether the
lateral horn develops from the fusion of an umbilical bulla and lateral tubercle
on the inner whorls or represents merely the outward migration and accentua-
tion of the umbilical bulla above is not clear, due to artificial carving of ribs
on the inner whorl of the type specimen). F. serrata Reyment (1955: 69, pl.
15 fig. 3a—b)), from the Coniacian Awgu-Ndeaboh Shales near Enugu,
Nigeria, is known from the holotype only, a juvenile 36,5 mm in diameter.
The flanks are flattened and subparallel, with a rounded fastigiate venter.
Weak umbilical bullae give rise to narrow primary ribs that develop into a
bullate tubercle on the ventrolateral shoulder, where they are joined by one
or two non-tuberculate intercalatories. The ribs terminate in a blunt ventro-
lateral bulla separated by a smooth zone from a row of weak siphonal clavi.
This ornament is very different from that of most F. (Forresteria) species, and
is transitional towards Yabeiceras.

Reyment (1958: 68, pl. 6 (fig. 1)) referred Reesideoceras camerounense
Basse, 1947 (= Barroisiceras haberfelineri von Hauer var. alstadenensis
(Schliiter) de Grossouvre of Solger, 1904: 170, pl. 5 (fig. 6), text-figs 56-57) to
Forresteria (Forresteria). It appears rather to be a F. (Reesideoceras). The
Forresteria cf. allaudi |sic] of Parnes (1964: 23, pl. 2 (figs 7-8)) may belong here
or to F. (F.) peruanum (Briiggen) (see below); it is crushed. F. (F.) peruana
(= Gauthiericeras margae var. peruanum Briggen, 1910: 720, pl. 27 (fig. 3)
=F. (F.) bassae Benavides-Cacéres (1956: 477, pl. 58 (fig. 5)) from the
Coniacian of South America appears to be a distinctive compressed species,
perhaps closer to F. (F.) hobsoni. The best representation of this species is by
Liithy (1918: 41, pl. 1 (fig. 2a—b)). It does not find a match in the material here
referred to F. (F.) alluaudi.

Zumpangoceras ospinai Etayo-Serna (1979: 90, pl. 74 (fig. 8), text-figs 90,
9P) from Colombia may be a synonym of F. (F.) alluaudi but is figured in side
view only. F. (F.) armata Matsumoto (1969: 313, pl. 41, (fig. 1), text-fig 8) is a
clear synonym, as discussed above.

Occurrence

This species characterizes the Coniacian in many areas of the world and is
known from Zululand, Madagascar, Japan, Mexico, the United States Western
Interior, ? Colombia, Peru, and perhaps Israel. In South Africa it is restricted
to Coniacian II of Kennedy & Klinger (1975).

Forresteria (Forresteria) cf. hobsoni (Reeside, 1932)
Fig. 32
Compare

Barroisiceras (Forresteria) hobsoni Reeside, 1932: 18, pl. 9 (figs 2-4), pl. 10
(figs 1-2).

ANNALS OF THE SOUTH AFRICAN MUSEUM

280

4

A-B. Holotype

1907)
SAS Z714. All x1.

»)

nin
D

Z

éve
SAS Z18. C-

b)

Lemoine & Th

(Boule,
, 1968

ia) alluaudi
tor’ van Hoepen

(Forrester
ina

ia
iceras peregr

Forrester
Collignon

4

2D)
of

Fig

CRETACEOUS FAUNAS FROM SOUTH AFRICA 281

Material
SAM-D1187J, from locality 145 at Morrisvale, Zululand, St. Lucia Forma-
tion, Coniacian II.

Dimensions
D Wb Wh Wb: Wh U
182,0(100) —(—) 69,5(38,2) — 65,0(35,7)
Description

This specimen is a worn internal mould retaining traces of recrystallized
shell. Approximately two-thirds of the outer whorl is body chamber.

The coiling is very evolute, the umbilicus comprises 35,7 per cent of the
diameter and only 20 per cent of the previous whorl is covered. The umbilicus
is shallow, the umbilical wall low and vertical. The whorl section is compressed,
with an estimated whorl breadth to height ratio of 0,85-0,90, the greatest
breadth being at the lateral tubercle in costal section and close to the umbilical
shoulder in intercostal section.

There are seventeen primary ribs on the outer whorl. On the first half,
which is the best preserved, the low vertical umbilical wall terminates in an
abruptly rounded umbilical shoulder bearing small and variably developed
umbilical bullae.

From these arise broad and somewhat weak, straight prorsiradiate ribs that
extend to mid-flank, where they bear midlateral tubercles that vary irregularly
from weak to massive. These give rise to a pair of straight ribs that connect
with small obliquely placed ventrolateral clavi from which low, broad ribs
sweep forward in a broad chevron, the apex of which is a strong, elongate
siphonal clavus.

Similar ornament is locally preserved on the very worn body chamber,
where the irregular development of lateral tubercles is especially noticeable.

Where the inner whorls are visible the ribs seem to be relatively stronger
and coarser with respect to the tuberculation than on the outer whorls.

The sutures are too worn for useful description.

Remarks

This specimen has closely similar proportions to, and shows the same style
of ribbing and tuberculation as, the unique holotype of Forresteria (Forresteria)
hobsoni (Reeside) (1932: 18, pl. 9 (figs 2-4), pl. 10 (figs 1-2)) (Fig. 35A-B
herein). The latter is more evolute, however, with an outward sloping umbilical
wall. To what degree this and other minor differences reflect the different
preservations (distorted chalk composite internal mould in F. (F.) hobsoni,
wor sandstone mould in the African specimen) cannot be ascertained with
such limited material, as a result of which the present specimen is identified as
F. (F.) cf. hobsoni. This specimen shows the same style of ornament as the
more numerous F. (F.) madagascariensis with which it occurs, especially in

ANNALS OF THE SOUTH AFRICAN MUSEUM

282

WA Qo
p1dajsa4oq, JO sdXjojoH “(LO6T ‘UIUSASYL, 27 OUTOWIOT ‘o[nog)

IpnonyY (V1daISALLOJ) VIdASAAdOY “OT “BUA

283

CRETACEOUS FAUNAS FROM SOUTH AFRICA

LX “696Z SVS ‘(LOGI ‘UIUoADYT, 2 DUIOW] ‘a]Nog) IpnoN]JY (ViLaSaL40,]) DIMaISAL4O4 *LZ “BLA

————— a

ANNALS OF THE SOUTH AFRICAN MUSEUM

284

—-

—_——

es

SS orl

a ree

CRETACEOUS FAUNAS FROM SOUTH AFRICA 285

terms of variable lateral tuberculation and ventrolateral ribbing and tubercula-
tion. It may, indeed, be no more than the compressed and relatively weakly
ornamented end of the variation range in the species, but it is not possible to
tell on the basis of the available material; if true, the specific name hobsoni has
priority over madagascariensis.

Occurrence

St. Lucia Formation, Coniacian II, of Zululand.

Forresteria (Forresteria) madagascariensis (Collignon, 1965)

Figs 10C-D, 15 C-D, 36-38

Neokanabiceras madagascariense Collignon, 1965: 42, pl. 432 (figs 1784-1786, 1787 (var.
ankinatsyensis)). Wiedmann in Herm, Kauffman & Wiedmann, 1979: 44, pl. 7 (figs A-B);
text-fig. 7D.

Neokanabiceras apambiense Collignon, 1965: 44, pl. 432 (fig. 1788).

Holotype

The original of Collignon (1965: 42, pl. 432 (fig. 1784)) from the middle
Coniacian Zone of Kossmaticeras theobaldi and Barroisiceras onilahyense of
Ankinatsy (Belo sur Tsiribihina), Madagascar.

Material

SAM-D1187A-I and SAS Z1084A, from ‘Pisana’, on the Msindusi River,
presumably equivalent to locality 145 of Kennedy & Klinger (1975), St. Lucia
Formation, Coniacian II.

Dimensions
D Wb Wh Wb: Wh U
SAM-D1187H c 43 ,2(100) 24,3(56,3) 17,8(41,2) 137) 12,7(29,4)
ic 20,5(47,5) 15,8(36,6) 1,30
SAM-D1187I 43,9(100) 27,7(63,0) 18,9(43,0) 1,47 11,8(26,9)
SAM-D1187D c 93,0(100) 50,5(54,3) 36,0(38,7) 1,40 32,3(34,7)
ic 39,3(42,3) 34,2(36,8) 115
at D= c 78,5(100) 46,2(58,9) 30,0(38,2) 1,54 28,3(36,1)
ic 35,8(45,6) 27,8(35,4) 1,29
SAM-D1187C c 105,5(100) —(—) 41,7(39,5) —_ 36,1(34,2)
128,0(100) 68,8(53,8) 50,0(39,0) 1,38 41,2(32,2)
54,5(42,6) 46,8(36,6) 1,16
Description

All the available specimens are internal moulds retaining traces of shell.
The largest, SAM—D1187A, is still septate at a diameter of 130 mm.

The earliest visible stages are shown by SAM-D1187I (Fig. 10C—-D). The
coiling is evolute, with a deep umbilicus (26-34% of diameter). The whorl
section is depressed, coronate (whorl breadth to height ratio is 1,3 to 1,4 in
intercostal, and 1,37 to 1,47 in costal section) the maximum breadth at the
massive lateral tubercles in costal section and low on the flank in intercostal

ANNALS OF THE SOUTH AFRICAN MUSEUM

286

CRETACEOUS FAUNAS FROM SOUTH AFRICA 287

A B

Fig. 30. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907). A pathological
body chamber, SAS Z977. x1.

288 ANNALS OF THE SOUTH AFRICAN MUSEUM

~~

A B

Fig. 31. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907). Holotype of
‘Forresteria itwebae’ van Hoepen, 1968, SAS Z250. x1.

=

CRETACEOUS FAUNAS FROM SOUTH AFRICA 289

section. Twelve or thirteen strong, broad, rounded ribs arise at the umbilical
seam. They develop umbilical bullae of variable strength and pass slightly
prorsiradiate across the flanks to massive conical mid-lateral tubercles. These
give rise to a pair of straight or feebly concave ribs that link to strong oblique
ventral clavi twice as numerous as the lateral tubercles. These in turn give rise
to weaker, broad prorsiradiate ribs that form a chevron over the venter with a
strong siphonal clavus at the apex of the ‘V’.

Even at this diameter there is great variation in the relative strengths of
tubercles and ribs, the former dominating in SAM-—D1187G, the latter in
SAM-D 11871.

Two slightly larger fragments, SAM-—D1187E and F, show essentially
similar ornament at estimated diameters of 70mm; SAM-—D1187E (Fig.
36A-B) shows the beginning of differentiation in strength of the lateral
tubercles, every fourth one of which is much stronger than the others, while
SAM-—D1187F has shell preserved on sharp, high siphonal clavi (Fig. 15C—D).

SAM-—D1187D shows the ornamentation at a diameter of 93 mm. There
are twelve to thirteen strong ribs per whorl, bearing weak to strong umbilical
bullae that link to variable strong to massive lateral tubercles. These give rise to
pairs of strong ribs linking to strong ventral clavi from which weaker, broad ribs
form a chevron linking to the strong, sharp siphonal clavi.

SAM-D1187D (Fig. 36C—D) is a comparable specimen with approximately
110 mm diameter. The style of ornament on the first half of the outer whorl is
similar, if less hypernodose, while on the outer whorl ribbing is relatively better
developed with respect to the tubercles. Both these specimens and
SAM-D1187B, a rather poorly preserved individual, also show occasional
primary ribs without umbilical bullae or lateral tubercles. The largest specimen,
SAM-D1187A (Fig. 37A—-B), maintains the same style of ornament described
above, with thirteen strong, tuberculate primary ribs per whorl.

The suture line is incompletely exposed on SAM-—D1187C only. E/L is
elongate, rectangular, little incised, and asymmetrically bifid, L is narrow and
deep.

Discussion

The specimens referred to this species vary greatly in proportions and
relative development of ribs and tubercles, but are linked by coronate form,
Sparse primary ribs, massive, if variable lateral tubercles, paired outer flank
ribs, and ventral and siphonal clavi. They match closely with the types of
Neokanabiceras madagascariense Collignon (1965: 42, pl. 432 (figs 1784-1786))
and encompass the predominantly ribbed var. ankinatsyensis Collignon
(1965: 43, pl. 432 (fig. 1878)). The authors have little doubt that, given a
somewhat larger collection, Neokanabiceras apambiense Collignon (1965: 43,
pl. 432 (fig. 1788)) would prove to be no more than a very depressed variant of
this species.

290 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fig. 32. Forresteria (Forresteria) cf. hobsoni (Reeside, 1932). SAM-—D1187. x0,78.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 291

Fig. 33. Forresteria (Forresteria) alluaudi
(Boule, Lemoine & Thévenin, 1907). External
suture of an inflated specimen, SAS Z1438. x2.

L "2
E TA
if
|
A
L U
E r
z :
bn
B
L Au ~
U
U, E A
\ 3 /

| x :
| i !

Fig. 34. Forresteria (Forresteria) alluaudi (Boule, Lemoine & Thévenin, 1907). Showing external
sutures. A. A compressed specimen, SAS Z1429. B. A compressed specimen, SAS Z978. C. A
pathological specimen, SAS Z1523. All x2.

ANNALS OF THE SOUTH AFRICAN MUSEUM

292

a — - — a ———

‘Ix ‘((L-€ s8y) ZL ‘1d ‘ZE61) APIsooy JUMOIUYIS (DI4a]SAAAOJ) SDAIIISIOAADG

jo odAjeied v “(Z06T ‘UtUoADYL, 2 DUIOWIST ‘a[nNog) IpnHoN]JDY (vid4ajsa140J) DIdaIsas4d0oq “F—D ‘LOOX “ZOLEL WNSN ‘opesojo9d
‘sdutidg ayyjieD vou ouojsoury seduny, oy) wory adAjopoy oy, “(ZEGT “OpIseoy) MMosqoy (vILaISa140J) DIdaISa14o. “F-W SE “BIA

CRETACEOUS FAUNAS FROM SOUTH AFRICA 293

There is close resemblance to Forresteria (Forresteria) hobsoni Reeside
(1932: 18, pl. 9 (figs 2-4), pl. 10 (figs 1-2)), a species originally described from
the Coniacian of the United States Western Interior and refigured here as
Figure 35A-B. A single specimen compared with this species occurs with the
collection of F. (F.) madagascariensis, but differs in being compressed rather
than depressed and with more primary ribs per whorl. It does, however, show a
similar strong and variably developed lateral tubercle that gives rise to pairs of
ribs linking to ventral clavi that in turn link to a ventral chevron and row of
siphonal clavi. This specimen may be simply a compressed variant of a
population that encompasses both F. (F.) madagascariensis and F. (F.) hob-
soni (the latter name having priority), but as the type of F. (F.) hobsoni is the
only known specimen from the United States, this cannot be certain, and they
are maintained separate.

There are also similarities to Yabeiceras transiens sp. nov. (Fig. 46A-C),
described and discussed below, which points to the close affinities of Yabeiceras
and Forresteria.

When compared to the variable Forresteria (Forresteria) alluaudi described
above, the most obvious differences are to be found in the combination of
massive lateral tubercles and strong ribs, plus the presence of a ridge support-
ing the siphonal clavi.

There is a closer similarity to Forresteria (Forresteria) nwalii (Reyment,
1976) (in Offodile & Reyment 1976: 61, fig. 14A—B), the holotype of which is
re-illustrated here as Figure 39A-B. There are a similar number of strong
primary ribs in both species with a weaker umbilical than lateral tubercle, the
latter variably and massively developed. In F. (F.) nwalii, however, it is
pinched and narrow, the pairs of ribs linking to the ventral clavi are stronger
and not uncommonly accompanied by one or two weaker additional ribs
looping between lateral and ventral tubercles. In the Nigerian form, which has
an altogether clumsier ornament, the ventral clavi are relatively larger, ventral
ribs stronger, and chevron angle larger.

Occurrence
St. Lucia Formation, Coniacian I, of Zululand; Coniacian of Madagascar.

Genus Yabeiceras Tokunaga & Shimizu, 1926
(= Eboroceras Basse, 1946)

Type species

Yabeiceras orientale Tokunaga & Shimizu (1926: 20, pl. 22 (fig. 7), pl. 27
(fig. 1)) from the Coniacian of Japan, by the original designation of Tokunaga
and Shimizu (1926: 20).

Diagnosis

Very evolute. Inner whorls depressed, coronate, with lateral tubercles,
oblique ventrolateral tubercles twice as numerous as the lateral, and a rounded

294 ANNALS OF THE SOUTH AFRICAN MUSEUM

siphonal ridge with low clavi. These persist into middle growth in some species.
In others the tubercles decline and are replaced by ribs; the venter has a
rounded siphonal ridge flanked by variably developed grooves, the outer edges
of which may be slightly raised. Ornament persists in some species, in others
the outer whorl is smooth or has a ventral keel and sulci only. Smooth body
chambers develop at disparate sizes, suggesting a size dimorphism.

Suture with rather narrow, deeply incised elements.

Discussion

The inner whorls of Yabeiceras are so similar to those of Forresteria
(Forresteria) that close affinity cannot be doubted; it is a member of the
subfamily Barroisiceratinae (compare Figs 11A—K, 41A—D) as Matsumoto et al.
(1964) and Matsumoto (1969) have shown. Yabeiceras transiens sp. nov.,
described below (p. 303), shows an intermediate morphology even in middle
growth (Fig. 46). Forresteria (Harleites) appears at the base of the Coniacian
(Hancock & Kennedy 1981; Kennedy, Wright & Hancock 1982) and the line of
descent is taken here to be F. (Harleites) — F. (Forresteria) — Yabeiceras.

Eboroceras Basse (1946: 73), type species Eboroceras magnumbilicatum
Basse (1946: 73, pl. 2 (fig. 2), text-fig. 2), is a clear synonym. Yabeiceras is a
very rare genus in terms of numbers and, as is commonly the case, many
species have been described that differ only in details:

1. Yabeiceras orientale Tokunaga & Shimizu, 1926: 20, pl. 22 (fig. 7).
Shimizu, 1926: 547. Matsumoto et al. 1964: 323, pl. 48 (figs 1-2), text-figs 1-3.
Matsumoto, 1969: 324, pl. 44 (figs 1-2), pl. 45 (fig. 1), text-figs 12-13. The
unique holotype from the upper reaches of the Sakurazawa in Oriki, Japan,
was destroyed in World War II. Two specimens are known from the Futaba
area of north-eastern Japan (Matsumoto et al. 1964) and two more are
described by Matsumoto (1969).

2. Yabeiceras kotoi Tokunaga & Shimizu, 1926: 202, pl. 22 (fig. 8), pl. 28
(fig. 16). This was based on a single specimen from the same horizon and
locality as the holotype of Y. orientale, and was also destroyed by fire.

3. Yabeiceras himuroi Tokunaga & Shimizu, 1926: 203, pl. 22 (fig. 9), pl.
27 (fig. 2). This was based on a single fragment with the same origin and fate as
the previous species. It is best treated as a nomen dubium.

4. Yabeiceras magnumbilicatum (Basse, 1946): Basse, 1946: 73, pl. 2, (fig.
2), text-fig. 2. Collignon, 1965: 82, pl. 451 (fig. 1835): 84, pl. 452 (fig. 1837).
This species was described on the basis of the holotype only. Two other
specimens are described by Collignon. All are from Madagascar.

5. Yabeiceras bituberculatum Collignon, 1965: 82, pl. 451 (fig. 1836): 84,
pl. 452 (fig. 1838). Two specimens only have been illustrated, from the
Kossmaticeras theobaldi and Barroisiceras onilahyense Zone of Ampozaloaka,
Madagascar.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 295

6. Yabeiceras manasoaense Collignon, 1965: 84, pl. 452 (fig. 1839). Matsu-
moto, 1971: 144, pl. 24 (fig. 2), text-fig. 9. Klinger et al., 1976: 162, figs 1-4.
The holotype is from the same horizon as the previous species at Manasoa
(Betioky), Madagascar. Two others, from Japan and the South African off-
shore Alphard Group, have also been described.

7. Yabeiceras menabense Collignon, 1965: 86, pl. 453 (fig. 1840). The
holotype is the only described specimen and is from the Peroniceras dravidicum
Zone of Ankinatsy-Souroumaraina (Belo sur Tsiribihina), Madagascar.

8. Yabeiceras costatum Collignon, 1965: 87, pl. 454 (fig. 1841). The holo-
type is the only described specimen and is from the same horizon and locality as
the previous species.

9. Yabeiceras ankinatsyense Collignon, 1965: 87, pl. 454 (fig. 1842). Only
the holotype was described and is from the same horizon and locality as the
previous species.

It will be seen that there are only 17 described specimens of Yabeiceras, 8
from Japan, referred to 4 species, 8 from Madagascar, referred to 5 species,
and 1 from the South African offshore Alphard Group.

The present collection of fourteen specimens from Zululand nearly doubles
the described material of this genus. Unfortunately it complicates rather than
clarifies the taxonomy; because there are so few specimens, the authors cannot
assess the limits of intraspecific variability nor gauge the effect of suspected yet
unconfirmed dimorphism. In contrast to the genus Forresteria, where the extent
of intraspecific variation could satisfactorily be demonstrated, a conservative
morphological taxonomy is therefore followed below, with many species that
may prove to be no more than fragments of one or a few variable species.

Occurrence

Coniacian of Japan, Madagascar, Zululand, and the South African off-
shore Alphard Group.

Yabeiceras orientale Tokunaga & Shimizu, 1926
Figs 41, 42G-I, 45

Yabeiceras orientale Tokunaga & Shimizu, 1926: 20, pl. 22 (fig. 7), pl. 27 (fig. 1). Shimizu,
1926: 547. Matsumoto et al., 1964: 323, pl. 48 (figs 1-2), text-figs la-d, 2a-e, 3. Matsu-
moto 1969: 324, pl. 44 (figs 1-2), pl. 45 (fig. 1), text-figs 12-13.

Holotype

The original of Tokunaga & Shimizu (1926: 20, pl. 22 (fig. 7), pl. 27 (fig.
la—c)), from the Coniacian Futaba Formation of the upper reaches of the
Sakurazawa in the Oriki, north-eastern Japan, destroyed in World War II.

Material

SAS H196/1, from the west bank of the Hluhluwe River, Zululand,
32°19'30"E 28°5'30"S, St. Lucia Formation, Coniacian II.

296 ANNALS OF THE SOUTH AFRICAN MUSEUM

‘
3
'
|

Fig. 36. Forresteria (Forresteria) madagascariensis (Collignon, 1965). A-B. SAM-D1187E.
C-D. SAM-D1187D. All x1.

mii)

CRETACEOUS FAUNAS FROM SOUTH AFRICA

1x

VL8IT

d-WVS

(S961 UOUSIT[OD) sisuatDISVSDpHOUL (D14I]SALIOJ ) D1AA]SALLOT

Ee

314

ANNALS OF THE SOUTH AFRICAN MUSEUM

298

‘LX ‘OL8LIG-INVS (S961 ‘UoUsTTIOD) sisuatuvosvsvpoU (VI4a]SaL10J) DIAAJSALIOY “QE “SIA

Vv

CRETACEOUS FAUNAS FROM SOUTH AFRICA 299

Dimensions

D Wb Wh Wb: Wh U
Cc 64,0(100) 26,5(41,4) 18,5(28,9) 143 29,3(45,8)
IC 64,0(100) 23 ,8(37,2) 18,5(28,9) 1,29 29 ,3(45,8)
Description

The earliest stage visible is at a diameter of 20 mm (Fig. 41C-D). The
coiling is evolute with a deep umbilicus, the whorl section depressed, with the
greatest breadth low on the flank intercostally and at the lateral tubercle
costally. There are thirteen to fourteen strong primary ribs per whorl, arising
on the umbilical shoulder. They develop into strong bullate to conical umbilico-
lateral tubercles that increase rapidly in strength as size increases. At the
smallest diameter visible these tubercles give rise to pairs of strong prorsiradi-
ate ribs that terminate in oblique ventral clavi. A shallow depression separates
these from a corresponding number of siphonal clavi. This same style of
ornament continues to 20 mm but the ribs become broader and lower. Beyond
this the whorl section becomes very depressed, the tubercles conical and
progressively stronger to a diameter of about 50mm, at the same time
migrating outwards from the umbilicus (Fig. 42G—I). The ribs progressively
decline and are effaced on the outer flank, the ventral grooves are weak, and a
line of low, weak siphonal clavi is still visible.

The coiling is very evolute on the outer whorl, less than 25 per cent of the
previous whorl being covered. The whorl section is depressed and reniform,
with an intercostal whorl breadth to height ratio of 1,29 and a costal ratio of
1,43, the greatest breadth being close to the umbilical shoulder.

The flanks are reduced and the venter is very broad and rounded. The
umbilicus is of moderate width, comprising 45 per cent of the diameter, with a
low, rounded, outward-sloping wall indented to accommodate the tubercles of
the preceding whorl.

Ornament consists of eighteen strong, broad primary ribs arising at the
umbilical wall and developing into strong umbilicolateral tubercles. These are
initially conical but become progressively bullate as size increases. Single ribs or
pairs of low, rounded, concave, prorsiradiate ribs arise from these tubercles,
sweep forward over the ventrolateral shoulders and become increasingly
weakened as size increases, terminating in weak, obliquely placed ventral clavi.
These are separated by a feeble, broad groove from a spiral siphonal swelling
that bears low, rounded clavi corresponding to the ventrolaterals.

The suture line is well exposed (Fig. 45).

Discussion

This interesting specimen shows the ontogenetic development from a di-
ameter similar to that of specimens of Yabeiceras orientale described by
Matsumoto ef al. (1964), to which it is essentially identical. The change in
tuberculation round the outer whorl is comparable to that in Y. orientale,

ANNALS OF THE SOUTH AFRICAN MUSEUM

300

‘LX ‘RLIOSIN ‘oleIg eIquieuy ‘n8e[eyN Wosy sdAjojoy 2YL “(9L6T ‘WuowAoY) mjomu (D149]SALdO]) D1LIJSALLOT

"6¢ “SI

CRETACEOUS FAUNAS FROM SOUTH AFRICA 301

Fig. 40. A-C. Yabeiceras costatum Collignon, 1965. SAM-D1188C. D-E. Forresteria (Forres-
teria) alluaudi (Boule, Lemoine & Thévenin, 1907). Inner whorls of the holotype of ‘Forresteria
itwebae’ van Hoepen, 1968, SAS Z250. All x1.

302 ANNALS OF THE SOUTH AFRICAN MUSEUM

D

C

Fig. 41. Yabeiceras orientale Tokunaga & Shimizu, 1926. Inner whorls of SAS H196/1. x2.

although occurring at a slightly different diameter. It is too small to show the
change from depressed tuberculate whorls to high, compressed, smooth whorls
of adult Y. orientale (e.g. Figs 43-44), but in spite of this the authors believe it
to represent the same species.

Occurrence

The species is known from the Futaba area of north-eastern Japan and
from Hokkaido. The present specimen is from the St. Lucia Formation,
Coniacian II, of Zululand.

Yabeiceras cf. orientale Tokunaga & Shimizu, 1926
Fig. 42A—C
Compare
Yabeiceras orientale Tokunaga & Shimizu, 1926: 20, pl. 22 (fig. 7), pl. 27
(fig. 1).

Material

SAM-D1188F, from locality 145 in the Morrisvale area to the north of
Neweni, Zululand, St. Lucia Formation, Coniacian II.

Discussion

This poorly preserved juvenile has sixteen primary ribs with rather strong
umbilicolateral tubercles. It most closely recalls Y. orientale, discussed above.

Occurrence

St. Lucia Formation, Coniacian II, of Zululand.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 303

Yabeiceras transiens sp. nov.
Fig. 46

Holotype

SAM-D1188A, by monotypy, from locality 145 in the Morrisvale area to
the north of Ngweni, Zululand, St. Lucia Formation, Coniacian II.

Etymology

Refers to the transitional features of the species between Forresteria and
Yabeiceras.

Dimensions
Cc 83,9(100) 40,5(48,3) 32,0(38,1) 127 30,3(36,1)
ic 35,0(41,7) 31,8(37,9) fet
68,5(100) 32,9(48,0) 25 ,4(37,1) 1,30 24,4(35,6)
29,5(43,1) 25,0(36,5) hati:
Description

The holotype and only known specimen is a well-preserved, wholly septate
mould retaining traces of recrystallized shell. Coiling is moderately evolute with
approximately 40 per cent of the previous whorl being covered. The whorl
section is depressed; at the beginning of the outermost whorl the whorl breadth
to height ratio is 1,65, the costal section polygonal, and the intercostal section a
depressed oval. At the greatest preserved diameter the whorl breadth to height
ratio is 1,27 costally and 1,1 intercostally, the greatest breadth being at the
lateral tubercle on the ribs and close to the umbilical shoulder intercostally.
The umbilicus comprises 36,1 per cent of the diameter, is relatively deep with a
flattened, vertical umbilical wall and abruptly rounded umbilical shoulder.
There are eighteen primary ribs on the outer whorl. These bear well-developed
umbilical bullae of variable strength. They are prorsiradiate and straight or
slightly concave, strong and bar-like to mid-flank where they bear a lateral
tubercle. This is strong and spinate at the smallest diameter visible but declines
around the outer whorl, becoming bullate. Each lateral tubercle gives rise to a
pair of somewhat weaker, broader, rounded ribs that bear strong, obliquely
placed ventrolateral clavi, an estimated total of thirty to thirty-three per whorl.
These project forward over the venter where they are interrupted by a broad
shallow groove on either side of a rounded siphonal keel. This bears distinct
siphonal clavi at the smallest diameter visible, corresponding in number to, but
displaced adaperturally of, the ventrolateral clavi. These decline around the
outer whorl giving a low, undulose Keel at the greatest diameter preserved.

The suture line is incompletely exposed. E/L is broad and bifid with
moderate incision; L deep, narrow and bifid; L/U, smaller and symmetrically
bifid.

304 ANNALS OF THE SOUTH AFRICAN MUSEUM :

]
:
:

Fig. 42. A-C. Yabeiceras cf. orientale Tokunaga & Shimizu, 1926, SAM-D1188F. D-E. Yabeiceras
costatum Collignon, 1965, SAM-D1188E. F. Yabeiceras sp. indet., SAM-— D1182. G-I. Yabeiceras
orientale Tokunaga & Shimizu, 1926, SAS H196/1. All x1.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 305

A B

Fig. 43. Yabeiceras orientale Tokunaga & Shimizu, 1926. From the upper reaches of the
Ikushumbets, Hokkaido, Japan, Kyisht University Collections H5624. x1.

Discussion

As suggested by the name, this new species combines features of both
Yabeiceras and Forresteria, having inner whorls with strong ribs and tubercles
like the latter and a carinate-bisulcate venter like Yabeiceras that retains traces
of siphonal clavi to a much greater diameter than any other species referred to
the genus.

The only described species with which it is likely to be confused is
Yabeiceras bituberculatum Collignon (1965: 82, pl. 341 (fig. 1836), pl. 342 (fig.
1838)). They differ in that the ribs are relatively stronger and more numerous
in the new species and bear umbilical bullae throughout, while ribbing and
tuberculation are retained to a diameter where Y. bituberculatum has already
developed a smooth, constricted body chamber. Among Forresteria (Forres-
teria) species, the closest resemblance is to F. (F.) madagascariensis with which
it occurs (see above). All the specimens of this species are stronger ribbed with
massive, variable lateral tubercles, stronger ventrolateral and siphonal clavi and
no obvious ventral sulci. In the F. (Forresteria)-like inner and smoothing outer

306

ANNALS OF THE SOUTH AFRICAN MUSEUM

IGPS 35342, from the Bibai, Ishikari Province,

ty Collections

1versl

, 1926. Tohuko Un

imizu

°

entale Tokunaga & Sh

.

wceras orl

Fig. 44. Yabe

Hokkaido, Japan. x1.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 307

Fig. 45. Yabeiceras orientale Tokunaga & Shimizu,
1926. External suture of SAS H196/1. x2.

whorls there is a resemblance to F. (Muramotoa), but all described species are
more involute and high-whorled and show a rapid loss of ornament.

It may be that Yabeiceras transiens is a micromorph and perhaps a
microconch of some other barroisiceratid, for, as has been noted above, both
micro- and macromorph taxa occur throughout the family (see p. 242).

Occurrence
St. Lucia Formation, Coniacian II, of Zululand.

Yabeiceras ankinatsyense Collignon, 1965
Fig. 5|0C-—D
Yabeiceras ankinatsyense Collignon, 1965: 87, pl. 454 (fig. 1842).

Holotype

The original of Collignon (1965: 87, pl. 454 (fig. 1842)), from the Peroni-
ceras dravidicum Zone of Ankinatsy (Belo sur Tsiribihina), Madagascar.

Material

SAS H196/3, from the west bank of the Hluhluwe River, 32°19'30"E
28°5'30"S Zululand, St. Lucia Formation, Coniacian II.

Dimensions

D Wb Wh Wb: Wh U
SAS H196/3 34,2(100) 15,5(45,3) 12,8(37,4) 12 8,2(24,0)
Holotype (after
Collignon) 70,0(100) 24,0(34) 24,0(34) 1,0 27,0(39)
Description

This specimen is only 34,2 mm in diameter, moderately evolute with a
rather smail umbilicus comprising 24 per cent of the diameter. The whorl
section is a depressed oval in intercostal section, with the greatest width low on
the flanks and at the umbilical bullae in costal section. The umbilical wall is
subvertical, undercut, and of moderate elevation, the umbilical shoulder
abruptly rounded, the inner flanks moderately inflated, the outer flattened,
with a broadly rounded, flattened venter.

ANNALS OF THE SOUTH AFRICAN MUSEUM

308

‘Op “SIF

oS a, ee

CRETACEOUS FAUNAS FROM SOUTH AFRICA 309

There are sixteen strong, narrow ribs per whorl. These arise as mere striae
at the umbilical seam but strengthen at the umbilical shoulder, developing
feebly defined bullae. They are strong, rounded and narrow on the inner half of
the flanks, straight and prorsiradiate, and bear a strong, sharp, bullate lateral
tubercle. The lateral tubercles usually give rise to pairs of weak, concave
secondaries that are accompanied by a few intercalated secondaries to give a
total of thirty ribs per whorl on the outer parts of the flanks. These sweep
forward across the ventrolateral shoulder where they terminate in low, swollen,
oblique ventrolateral clavi. There is a low, broad, rounded undulose siphonal
ridge flanked by shallow grooves, developing ill-defined siphonal clavi corre-
sponding in position and number to the ventrolateral clavi.

The sutures are not exposed.

Discussion

Few Yabeiceras species are known at this diameter but all those that are
show the umbilicolateral, ventrolateral, and siphonal tuberculation of the
present specimen to various degrees. Y. orientale Tokunaga & Shimizu is more
depressed with strong lateral tubercles. Strength of tubercles also distinguishes
Y. magnumbilicatum, manasoaense, menabense, transiens, and bituberculatum
at this size as far as can be seen from the inner whorls of the much larger type
specimens, while there are also differences in number of ribs. The costal whorl
sections of Y. bituberculatum and Y. transiens are markedly polygonal and
easily distinguished. The inner whorls of the holotype of Y. costatum are much
more coarsely and closely ribbed with stronger flank tubercles. The inner
whorls of Y. ankinatsyense thus provide the closest comparisons—rib density
and strength are similar, as far as is visible—but the umbilicus at the admittedly
greater diameter (70 compared to 34,5 mm) is much wider (39% compared to
24%) and the whorls are as wide as high rather than depressed. These
differences may well be entirely due to the disparate sizes of our specimen and
the holotype.

Occurrence

St. Lucia Formation, Coniacian II, of Zululand.

Yabeiceras costatum Collignon, 1965
Figs 40A—C, 42D-E
Yabeiceras costatum Collignon, 1965: 87, pl. 454 (fig. 1841).

Material

SAM-D1180C and E, both from locality 145 in the Morrisvale area north
of Ngweni, Zululand, St. Lucia Formation, Coniacian II.

310 ANNALS OF THE SOUTH AFRICAN MUSEUM

Dimensions

D Wb Wh Wb: Wh U
SAM-D1188C 68;,0(100) . 25;5(37;5) . 21,5G1,6) 1419, _292Gs%a)
Holotype (after
Collignon 1965) 74,0(100) 29,0(39) 24 ,0(32) 1,20 36,0(49,0)

Description

SAM-—D1180C is a well-preserved specimen, largely septate, and retaining
recrystallized shell. The coiling is very evolute, with a broad umbilicus com-
prising 43 per cent of the diameter. The umbilical wall is relatively low,
rounded, and undercut. The whorl section is depressed (whorl breadth to
height ratio is 1,19) with the greatest breadth at the lateral tubercles, where

present, or at mid-flank. The flanks are flattened and converge to a broadly
rounded venter.

There are twenty-five to twenty-six primary ribs per whorl. These arise at
the umbilical seam, pass backward across the umbilical wall and strengthen into
an umbilical bulla of variable, generally weak development at the umbilical
shoulder. The ribs are straight, broad, rounded, and prorsiradiate, strengthen
across the inner flank and bear a well-defined pointed midlateral tubercle up to
a diameter of 45 mm, beyond which it progressively declines. The ribs weaken
across the outer flank; on the first half of the outer whorl, to a diameter of
around 55 m, they strengthen as they sweep slightly forward across the ventro-
lateral shoulder, giving the appearance of an ill-defined oblique, rounded
ventral clavus that terminates against a shallow, broad ventral groove flanking
the siphonal keel. Beyond 55 mm diameter the ribs weaken markedly over the
ventrolateral shoulder.

The siphonal Keel is rounded, obscurely undulose at the smallest diameter
visible, becoming continuous over the last half whorl.

SAM-D1188E, a partially septate specimen with an estimated original
diameter of 75 mm, shows the continuing disappearance of the midlateral
tubercle and weakening of the ribs over the ventrolateral shoulder.

The sutures are not exposed.

Discussion

Style of ornament, in particular the predominance of ribbing and early
decline of the midlateral tubercle, link this specimen with the holotype of Y.
costatum Collignon (1965: 87, pl. 454 (fig. 1841)); it differs only in having a
slightly narrower umbilicus (43 % compared to 49).

Y. costatum most closely resembles Y. ankinatsyense Collignon (1965: 87,
pl. 354 (fig. 1842)) from the same horizon at Ankinatsy (Belo sur Tsiribihina),
which has slightly different proportions (D=70 (100); Wb = 24(34,0);
Wh = 24(34); U = 27(39,0)), in particular a whorl section as wide as high and a
narrower umbilicus (39% of diameter). There are few ribs—fifteen to sixteen
per whorl, concave and, according to Collignon, lacking tubercles. The holo-
type is the only known specimen.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 311

Y. menabense Collignon (1965: 86, pl. 453 (fig. 1840)) and Y. manasoaense
Collignon (1965: 84, pl. 452 (fig. 1839)) are easily distinguished by their
massive whorls and the development of large umbilicolateral tubercles at
diameters comparable to our specimens.

The holotype of Y. bituberculatum Collignon (1965: 82, pl. 451 (fig.
136a—c)) again has massive umbilical tubercles; the smaller paratype referred to
the species (Collignon 1965: 84, pl. 452 (fig. 1836)) has a polygonal whorl, with
massive lateral and prominent ventral tubercles, the latter twice as numerous as
the former.

Y. magnumbilicatum (Basse) (1946: 73, pl. 2 (fig. 2a—b) text-fig. 2) has
strong, sparse umbilical and lateral tubercles, and loses all ornament on the
body chamber.

Of Yabeiceras described from Japan, Y. orientale Tokunaga & Shimizu,
1926, has fewer, stronger lateral tubercles that give rise to pairs of secondaries
with ventral clavi in middle growth and a smooth body chamber.

Y. kotoi Tokunaga and Shimizu (1926: 302, pl. 22 (fig. 8), pl. 26 (fig. 15)),
the types of which were destroyed in World War II, has not been discussed by
subsequent Japanese workers. According to the original account it is com-
pressed (whorl breadth to height ratio is 0,83 at a diameter of 80 mm).

Occurrence

St. Lucia Formation, Coniacian II, of Zululand.

Yabeiceras manasoaense Collignon, 1965

Figs 47, 51A-C

Yabeiceras manasoaense Collignon, 1965: 84, pl. 452 (fig. 1839). Matsumoto 1971: 144, pl. 24
(fig. 2), text-fig. 9. Klinger, Kennedy & Siesser, 1976: 162, figs 1-4.

Holotype

The original of Collignon (1965: 84, pl. 452 (fig. 1839)) from the zone of
Kossmaticeras theobaldi and Barroisiceras onilahyense of Manasoa, Madagas-
Car.

Material

BMNH C81465 and C81472, both from locality 145 in the Morrisvale area
north of Ngweni, Zululand, St. Lucia Formation, Coniacian II.

Dimensions
D Wb Wh Wb: Wh U
BMNH C81472 c 65,9(100) 35 (53,1) 19,7(29,9) a8 31,6(48,0)
ic 31,8(48,3) 19,7(29,9) 1,61
BMNH C81465 c 114,5(100) 42,0(36,7) 33,3(29,1) 1,26 54,0(47,2)
ic 40,8(35,6) 33,3(29,1) (12
c 89,8(100) 39,3(43,8) 26,1(29,1) 1,50 41,6(46,3)
ic 35,7(39,8) 26,1(29,1) 137

ch SOE. ly CED = PS meD Wi

ANNALS OF THE SOUTH AFRICAN MUSEUM

312

CRETACEOUS FAUNAS FROM SOUTH AFRICA 313

Description

The two specimens both retain recrystallized shell; the smaller shows details
of early development previously unknown in the species. Coiling is evolute (the
umbilicus comprises 48% of the diameter) with the strong tubercles exposed
within the umbilicus. The whorl section is depressed (breadth to height ratio is
1,61 intercostally, 1,78 at the tubercles), with the greatest breadth low on the
flank between the ribs and at the strong tubercles, the cross-section reniform,
with a broadly rounded venter and almost no flanks. The umbilicus is broad
(48 % of the diameter), relatively deep, with a rounded wall.

There are seventeen massive conical to bullate umbilicolateral tubercles
per whorl. At the smallest diameters visible these tubercles give rise to groups
of two or three low, broad, straight prorsiradiate ribs that sweep forward across
the ventrolateral shoulders and venter, terminating at the edge of the broad,
shallow ventral grooves on either side of the low, broad, rounded siphonal
ridge. The termination of the ribs is thickened but never develops into a
tubercle; as size increases the ribs broaden and weaken, leaving the venter
essentially smooth or with only faint undulations.

In the larger specimen the strong umbilical tubercles decline abruptly on
the body chamber from a diameter of 95 mm onwards, beyond which ornament
consists of distant, low, broad, concave flank ribs that decline and disappear
over the venter. At the same time the whorl section becomes less depressed
costally and intercostally, and the coiling becomes slightly eccentric.

The sutures are not exposed.

Discussion

The combination of massive, coronate, tuberculate inner whorls with a
decline to a ribbed and smoothing adult whorl match well with the Madagascan
holotype and the larger specimen from Japan described by Matsumoto (in
Matsumoto & Inoma 1971: 144, pl. 24 (fig. 2), text-fig. 9). When compared
with other species the massive tubercles of early-middle growth immediately
distinguish this species from comparably sized Yabeiceras magnumbilicatum,
Y. costatum, and Y. ankinatsyense. The tubercles of Y. menabense are spatulate
rather than conical. In Y. bituberculatum the umbilicolateral tubercles are
smaller, ribbing stronger, and ventrolateral tubercles better developed.
Y. transiens sp. nov. has a polygonal whorl section, ribbing strongly developed,
and umbilical, lateral, and ventrolateral tubercles. Y. orientale is less depressed
and has relatively weaker tubercles and stronger ribs.

Occurrence

The types are from the Zone of Kossmaticeras theobaldi and Barroisiceras
onilahyense of Manasoa (Betioky), Madagascar; the present specimens are
from the St. Lucia Formation, Coniacian II, of Zululand, and it is also recorded
from the Alphard Group off the southern Cape coast. It is also known from
Japan, occurring in Teshio Province, Hokkaido.

ANNALS OF THE SOUTH AFRICAN MUSEUM

314

CRETACEOUS FAUNAS FROM SOUTH AFRICA 315

Yabeiceras aff. manasoaense Collignon, 1965
Figs 48-49

Compare
Yabeiceras manasoaense Collignon, 1965: 84, pl. 452 (fig. 1839).

Material

SAM-D1188B-C from locality 145 in the Morrisvale area north of Ngweni,
Zululand, St. Lucia Formation, Coniacian II.

Dimensions
D Wb Wh Wb: Wh U
SAM-D1188B c 101,9(100) 37,8(37) 30,3(29,7) 1,25 49 8(48,9)
ic 34,5(33,9) 30,3(29,7) 1,14
82,0(100) 32,5(39,6) 24,6(30) 1,32 37,8(46,1)
29.0(35,4) 24,6(30) fs
SAM-D1188C 97,4(100) =) 27,5(28,2) ees 45,5(46,6)
c 90,0(100) 35,6(39,6) 26,0(28,9) 1,37 41,0(45,6)
ic 32,3(35,9) 26,0(28.9) 1,24
Description

These two specimens are both wholly septate and retain recrystallized
shell.

The coiling is very evolute, approximately 25 per cent of the previous
whorl being covered. The umbilicus is broad (up to 49,8% of the diameter),
shallow, with a low, rounded, outward sloping wall. The whorl section is
depressed (costal breadth to height range 1,37—1,25; intercostal 1,24~1,14),
with the greatest intercostal breadth low on the flank, the greatest costal
breadth at the lateral tubercles.

Low, broad primary ribs, fifteen to sixteen per whorl at the smallest
diameters visible (approximately 60 mm in SAM-D1188B and 50mm in
SAM- D1188C), arise at the umbilical seam and bear strong conical umbilico-
lateral tubercles housed in notches in the umbilical wall of the succeeding
whorl. As size increases this tubercle migrates outward to a lower lateral
position. The ventrolateral and ventral region is visible only from a diameter
of 60 mm; in SAM-—D1188C the tubercles are seen to give rise to single ribs
or, more rarely, pairs of ribs, which terminate in a blunt, low, oblique
ventrolateral clavus.

The outer whorl bears twenty primary ribs arising at the umbilical seam
and bearing a strong conical lateral tubercle at the beginning of the whorl. This
declines progressively, leaving only strong inner flank ribs in SAM—D1188B at a
diameter of 100 mm, and a small, somewhat bullate midlateral tubercle in
SAM-—D1188C at the same diameter. In both specimens this tubercle gives rise
to a broad, prorsiradiate concave rib that declines and effaces over the
ventrolateral shoulder, leaving a smooth or gently undulose zone on either side

ANNALS OF THE SOUTH AFRICAN MUSEUM

316

‘LX ‘O88LLG-NVS ‘S961 ‘UousT[oD asuavospunu

q

‘Jye spsa91aqv Xx

6b “SI

CRETACEOUS FAUNAS FROM SOUTH AFRICA Bile

of the strong, rounded siphonal keel and shallow flanking ridges, the outer
sides of which are raised into weaker ridges.
The sutures are not exposed.

Discussion

Rounded rather than polygonal whorl section, absence of umbilical
tubercles, loss of ventral tuberculation and entire, rather than nodate, siphonal
keel readily distinguish this species from Yabeiceras transiens sp. nov.,
described above.

Y. crassiornatum sp. nov., described below, is more coarsely ribbed and
tuberculate and retains its ribs and ventral tubercles to a much larger diameter.
Y. orientale has rather similar inner whorls, as far as comparison is possible, but
at a diameter of 80-90 mm loses all ornament and becomes compressed,
whereas the present species retains its ornament. This loss of ornament and
acquisition of smooth outer whorls at a relatively small size also allows our
specimens to be distinguished from Y. magnumbilicatum (Basse), which also
has rather fewer primary ribs per whorl in the early stages (thirteen in the
specimen figured by Collignon 1965, pl. 451 (fig. 1835)).

Y. bituberculatum has altogether distinct, smooth adult whorls at a rela-
tively small size, and utterly distinctive inner whorls. Y. menabense differs in its
curious spatulate tubercles on the inner whorls and loss of ribbing at a size
where these are retained by our specimens.

Y. ankinatsyense is more delicately ornamented with the ribs dominant
over tubercles at the same size, as in Y. costatum.

The closest comparisons thus appear to be with Y. manasoaense Collignon;
this, however, has much coarser lateral tubercles and a more depressed whorl
section. In the holotype (Collignon 1965: 84, pl. 452 (fig. 1839)), coarse
tubercles, fourteen to fifteen per whorl, persist to a diameter where the present
specimens already show a decline and predominance of ribs or feebly tubercu-
late ribs that number 20 per whorl.

The larger specimen figured by Matsumoto (1971, pl. 24 (fig. 2)) has
eighteen tubercles per whorl at a diameter of 100 mm and retains blunt ribs,
thus more closely resembling our specimens. When compared with the two
specimens from the same locality that have been referred to Y. manasoaense
above, the coarser, more massive tubercles and whorls and distinctively differ-
ent ontogenetic changes around the last whorl are again distinctive.

With so few specimens, we cannot tell if these differences are within the
range of intraspecific variation of the species Y. manasoaense or whether a
distinctive form is present. They are therefore referred to as Yabeiceras aff.
manasoaense, acknowledging both the similarities to and differences from the
Madagascan holotype and other material.

Occurrence

St. Lucia Formation, Coniacian II, of Zululand.

ANNALS OF THE SOUTH AFRICAN MUSEUM

318

|
|

CRETACEOUS FAUNAS FROM SOUTH AFRICA 319

Yabeiceras crassiornatum sp. nov.
Fig. 5|0A—B

Holotype
BMNH C81501, from locality 145 in the Morrisvale area, north of Ngweni,
Zululand, St. Lucia Formation, Coniacian II.

Etymology

Refers to the coarse ornament.

Dimensions
D Wb Wh Wb: Wh U
(é 137,00(100) 5255(6833) 44,0(32,1) 1,19 61,7(45,0)
IC 50,5(36,9) 44 .0(32,1) (SES
Description

The coiling is very evolute, serpenticone, with a broad, rather shallow
umbilicus that comprises 45 per cent of the diameter. The whorls are slightly
depressed (whorl breadth to height ratio is 1,19 costally and 1,15 intercostally),
the greatest breadth being well below midflank. The inner flanks are strongly
rounded, the outer flanks flattened, converging to the broad venter. The inner
whorls bear fifteen strong, broad, coarse primary ribs per whorl. These bear
strong clavate tubercles that migrate progressively outward from an umbilico-
lateral to lower lateral position. At the smallest diameter visible, the beginning
of the outer whorl, these tubercles give rise to broad, coarse, single ribs that
terminate in blunt ventrolateral tubercles. There is a strong, broad, rounded
siphonal keel flanked by broad, shallow grooves.

Traced around the outer whorl, the ribs remain coarse, distant and strong,
totalling 18 or 19 per whorl. The lateral tubercles, initially strong, decline over
the last third of a whorl and disappear, leaving only concave ribs on the flank.
In contrast the ventral tubercles strengthen and persist as obliquely placed clavi
on either side of the persistent coarse keel and shallow flanking grooves.

The sutures are not exposed.

Discussion

The persistence of coarse ribs and coarse ventral clavi to a very large
diameter combined with the strong lateral tubercles in middle growth distin-
guishes this specimen from all other described forms; it is consequently
afforded specific status.

Occurrence

St. Lucia Formation, Coniacian II, of Zululand.

320 ANNALS OF THE SOUTH AFRICAN MUSEUM

C D E

Fig. 51. A-C. Yabeiceras manasoaense Collignon, 1965. BMNH C81472. D-E. Yabeiceras cobbani
sp. nov. BMNH C81542. All x1.

CRETACEOUS FAUNAS FROM SOUTH AFRICA 321

Yabeiceras cobbani sp. nov.
Fig. 51D-E

Holotype

BMNH C81542, by monotypy, from locality 145 in the Morrisvale area
north of Ngweni, Zululand, St. Lucia Formation, Coniacian II.

Etymology
Named for Dr W. A. Cobban of the United States Geological Survey,
Denver.

Material
Only the holotype.

Description

The single specimen is distorted, but retains recrystallized shell and has a
maximum preserved diameter of 60 mm. The coiling is evolute, the umbilicus
deep, and comprises an estimated 35-40 per cent of the diameter. The whorl
section is depressed reniform in intercostal section with the greatest breadth
low on the flanks. In costal section the greatest breadth is at the lateral
tubercle; the estimated whorl breadth to height ratio is 1,7. There are an
estimated eighteen primary ribs per whorl. These arise from umbilical tubercles
of variable strength at the smallest diameter visible and migrate to a low lateral
position around the outer whorl. They give rise to pairs of broad, coarse,
concave prorsiradiate ribs that terminate in coarse, obliquely placed ventro-
lateral clavi. These are separated by a smooth, shallow groove from a broad,
rounded siphonal keel that bears blunt clavi corresponding approximately to
the ventral clavi. Interspaces between ribs are periodically deepened into
constrictions, strong across the flanks but absent over the venter. There are
four or five of these on the outer whorl.

Discussion

None of the described Yabeiceras species combines a coarse ornament of
umbilicolateral, ventrolateral, and siphonal tubercles in the manner shown by
this specimen and none develops constrictions.

Occurrence
St. Lucia Formation, Coniacian II, of Zululand.

Yabeiceras sp. indet.
Fig. 42F

Material

SAM-D1182, from locality 145 in the Morrisvale area north of Ngweni,
Zululand, St. Lucia Formation, Coniacian II.

322 ANNALS OF THE SOUTH AFRICAN MUSEUM

Description

This fragment shows an inner whorl with medium-sized conical lateral
tubercles. The outer whorl has a whorl breadth to height ratio of 1,15 with
rounded flanks and a broadly rounded venter. The flanks bear closely spaced
concave flank ribs that die out across the ventrolateral shoulders. There is a
strong, rounded ventral keel flanked by broad grooves, in turn flanked by slight
lateral Keels.

Discussion

The fragment has typically tuberculate inner whorls, as seen in many
specimens of Yabeiceras, combined with a ribbed phase recalling Y. costatum at
small diameters (e.g. Fig. 40A—C). It is, however, specifically indeterminate.

Occurrence
St. Lucia Formation, Coniacian II, of Zululand.

ACKNOWLEDGEMENTS

We thank Dr M. K. Howarth and Mr D. Phillips of the British Museum
(Natural History) (London) and the staff of the Geological Collections, Uni-
versity Museum (Oxford), and South African Museum (Cape Town) for their
assistance. The financial support of the Sir Henry Strakosch Bequest, the
Natural Environment Research Council, Wolfson College (Oxford), and the
South African Council for Scientific and Industrial Research is gratefully
acknowledged.

REFERENCES

BASSE, E. 1935. Etude géologique du Sud-Ouest de Madagascar. Mém. Soc. géol. Fr. 4: 1-153.

Basse, E. 1946. Sur deux ammonites nouvelles du Sud-ouest de Madagascar: Subbarroisiceras
n.g. mahafalense n. sp. et Eboroceras n.g. magnumbilicatum n. sp. Bull. Soc. géol. Fr. (5)
16: 71-76.

Basse, E. 1947. Les peuplements Malgaches de Barroisiceras (Révision du genre Barroisiceras
DE Gross.). Paléontologie de Madagascar 26. Annls Paléont. 22: 97-190 (1-82).

BENAVIDES-CACERES, V. E. 1956. Cretaceous System in northern Peru. Bull. Am. Mus. nat.
Hist. 108: 353-494.

BouLe, M., LEMOINE, P., & THEVENIN, A. 1906-1907. Paléontologie de Madagascar III.
Céphalopodes crétacés des environs de Diego-Suarez. Annls Paléont. 1: 173-192 (1-20); 2:
1-56 (21-76).

BREISTROFFER, M. 1947. Notes de nomenclature paléozoologique. P-v. Soc. Sci. Dauphiné 26:
5 pp. (unpaginated).

BruGGEN, H. 1910. Die Fauna des unteren Senon von Nord-Peru. Jn: STEINMANN, G. Beitrage
zur Geologie und Palaeontologie von Stidamerika. Neues Jb. Miner. Geol. Paldont.
BeilBd. 30: 717-788.

BURCKHARDT, C. 1919-21. Faunas Jurassicas de Symon (Zacatecas) y faunas Cretacicas de
Zumpango del Rio (Guerro). Boln Inst. geol. Mex. 33: 1-138 (1918); atlas (1921).

CoLLiGNon, M. 1965. Atlas des fossiles caractéristiques de Madagascar (Ammonites), XIII
(Coniacien). Tananarive: Service Géologique.

Coquanpb, H. 1859. Synopsis des animaux et des végétaux fossiles observés dans la formation
crétacée du Sud Ouest de la France. Bull. Soc. géol. Fr. (2) 16: 545-1023.

—s~

CRETACEOUS FAUNAS FROM SOUTH AFRICA 323

Erayo-SERNA, F. 1979. Zonation of the Cretaceous of central Colombia by Ammonites.
Publnes Geol. espec. Ingeominas 2: 1-186.

GERHARDT, K. 1897. Beitrag zur Kenntniss der Kreide Formation in Columbien. Neues Jb.
Miner. Geol. Paldont. BeilBd 11: 118-208.

GONZALEZ-ARREOLE, C. 1977. Ammonitas del Coniaciano (Cretacico superior) de la region de
Tepetlapa, Estado de Guerro. Revta Inst. Geol. Univ. nac. auton. Mex. 1: 167-173.

GrossouvrE, A. DE 1894. Recherches sur la Craie Supérieure, 2, Paléontologie. Les ammonites
de la Craie Supérieure. Mém. Serv. Carte géol. dét. Fr.: 1-264.

Hancock, J. M. & KENNEDY, W. J. 1981. Upper Cretaceous ammonite stratigraphy: some
current problems. Systematics Association Spec. Publ. 18: 531-553.

Hauer, F. von, 1866. Neue Cephalopoden aus den Gosaugebilden der Alpen. Sber. Akad.
Wiss. Wien 53: 300-308.

HayasakA, I. & FuKaApbA, A. 1951. On the ontogeny of Barroisiceras minimum Yabe from the
Upper Ammonite Bed in Hokkaido. J. Fac. Sci. Hokkaido Univ. (4) 7: 324-330.

HeErM, D., KAUFFMAN, E. G. & WIEDMANN, J. 1979. The age and depositional environment of
the ‘Gosau’-Group (Coniacian-Santonian), Brandenberg/Tirol, Austria. Mitt. bayer. St.
Paldont. Hist. Geol. 19: 27-92.

Hourca, V. 1936. Notice explicative de la feuille de Belo-sur-Tsiribihina. Tananarive: Service
Géologique.

KENNEDY, W. J. & KLINGER, H. C. 1975. Cretaceous faunas from Zululand and Natal, South
Africa. Introduction, Stratigraphy. Bull. Br. Mus. nat. Hist. (Geol.) 25: 265-315.

KENNEDY, W. J., WricHT, C. W. & Hancock, J. M. 1982. Zonations et corrélations par les
ammonites du Cénomanien terminal et du Turonien du sud d’ Angleterre et des régions
stratotypiques (Sarthe et Touraine—France). (In press.)

KLINGER, H. C., KENNEDY, W. J. & SIESSER, W. G. 1976. Yabeiceras (Coniacian ammonite)
from the Alphard Group off the southern Cape Coast. Ann. S. Afr. Mus. 69: 161-168.
KULLMAN, J. & WIEDMANN, J. 1970. Significance of sutures in phylogeny of Ammonoidea.

Paleont. Contr. Univ. Kans. 44: 1-32.

Lisson, C. I. 1908. Contribucién al conocimiento sobre algunos ammonites del Péru. Lima:
Tipografia el Peru.

Lttny, J. 1918. Beitrage zur Geologie und Palaeontologie von Péru. Abh. schweiz. paldont.
Ges. 43: 1-87.

Matsumoto, T. 1965. A monograph of the Collignoniceratidae from Hokkaido. Part 1. Mem.
Fac. Sci. Kyushu Univ. (D) 16: 1-80.

Matsumoto, T. 1969. A monograph of the Collignoniceratidae from Hokkaido. Part 3. Mem.
Fac. Sci. Kyushu Univ. (D) 19: 297-330.

Matsumoto, T. & INoMA, A. 1971. In: Matsumoto, T. A monograph of the Collignonicerati-
dae from Hokkaido. Part 5. Mem. Fac. Sci. Kyushu Univ. (D) 21: 129-162.

Matsumoto, T., Oats, I., MAepaA, S. & Sato, T. 1964. Yabeiceras (Cretaceous ammonite)
from Futaba, northeast Japan. Trans. Proc. palaeont. Soc. Japan 56: 322-331.

Matsumoto T., Muramorto, K., HIRANO, H. & TAKAHASHI, T. 1981. Some Coniacian ammon-
ites from Hokkaido. Trans. Proc. palaeont. Soc. Japan 121: 51-73.

Opata, I. 1965. Allometry of Reesidites minimus, a Cretaceous ammonite species. Trans. Proc
palaeont. Soc. Japan 58: 39-63.

OFFODILE, M. E. & REYMENT, R. A. 1976. Stratigraphy of the middle Benue region of Nigeria.
Bull. geol. Instn Univ. Uppsala 7: 37-66.

ParNES, A. 1964. Coniacian ammonites from the Negev (southern Israel). Bull. geol. Surv.
Israel 39: 1-42.

REESIDE, J. B. 1932. The Upper Cretaceous ammonite genus Barroisiceras in the United States.
Prof. Pap. U.S. geol. Surv. 170-B: 9-29.

REYMENT, R. A. 1955. The Cretaceous Ammonoidea of Nigeria and the Southern Cameroons.
Bull. geol. Surv. Nigeria 25: 1-112.

REYMENT, R. A. 1975. Analysis of a generic level transition in Cretaceous ammonites.
Evolution 28: 665-676.

SCHLUTER, C. 1876. Cephalopoden der oberen deutschen Kreide. Palaeontographica 24: 1-144
(121-264).

SHIMIZU, S. 1926. Three interesting Cretaceous ammonites recently acquired from Hokkaido
and Saghalin. Proc. imp. Acad. Japan 2: 547-550.

324 ANNALS OF THE SOUTH AFRICAN MUSEUM

SoLGER, F. 1904. Die Fossilien der Mungokreide in Kamerun und ihre geologische Bedeutung.
In: Escu, E. ed. Beitrdge zur Geologie von Kamerun. Schweizerbart: Stuttgart.

TOKUNAGA, S. & SuHimizu, S. 1926. The Cretaceous Formation of Futaba in Iwaki and its
fossils. J. Fac. Sci. Tokyo Univ. 1: 181-212.

VAN Hoepen, E. C. N. 1955. Turonian-Coniacian ammonites from Zululand. S. Afr. J. Sci. 51:
361-377.

VAN Hoepen, E. C. N. 1957. The deposits on the Umsinene River. C. R. CCTA. Géol.
Tananarive, Serv. géol. Madagascar 2: 349-350.

VAN Hoepen, E. C. N. 1968a. New ammonites from Zululand. Ann. geol. Surv. S. Afr. 4
(1965): 183-191.

VAN HoeEpEN, E. C. N. 1968b. New and little known Zululand and Pondoland ammonites. Ann.
Geol. Surv. S. Afr. 4 (1965): 157-181.

VENZO, S. 1935. Cefalopodi del Cretaceo médio-superiore dello Zululand. Palaeontogr. ital. 36:
59-133.

WEDEKIND, R. 1916. Uber Lobus, Suturallobus und Inzision. Zentbl. Miner. Geol. Paldont.
1916: 185-195.

WricutT, C. W. 1957. In: Moore, R. C. ed. Treatise on invertebrate palaeontology, Part L,
Mollusca, Cephalopoda, Ammonoidea. Geological Society of America and University of
Kansas Press.

WriGHT, C. W. & Matsumoto, T. 1954. Some doubtful Cretaceous ammonite genera from
Japan and Saghalien. Mem. Fac. Sci. Kyushu Univ. (D) 4: 107-134.

WrIGHT, C. W. & WriGuT, E. V. 1951. A survey of the fossil Cephalopoda of Great Britain.
Palaeontogr. Soc. Monogr.: 1-40.

YaBE, H. 1925. Japanese Cretaceous ammonites belonging to Prionotropidae, 1. Sci. Rep.
Toéhoku Univ. 7: 125-138.

\}

iad

j
{
at
HN
|
A
{|
y
|
x '
|
~
al
= Um
}
|
. | i
. a
Bin
|
:
| /
| \
Went
| -
;
Biel
Dt ©
wt fi
ive ;
ii a

fa

6. SYSTEMATIC papers must conform to the International code of zoological nomenclature
(particularly Articles 22 and S51).

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

’ 6

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

(c) Scientific names, but not their vernacular derivatives
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 HEUTE of
Biological Abstracts.

WILLIAM JAMES KENNEDY,
CLAUD WILLIAM WRIGHT

&

HERBERT CHRISTIAN KLINGER

CRETACEOUS FAUNAS FROM ZULULAND
AND NATAL, SOUTH AFRICA

THE AMMONITE SUBFAMILY
BARROISICERATINAE BASSE, 1947

en a Us — Y) tat a
SONIAN INSTITUTION oe SN eS IYVYEIT LIBRARI ES SMITHSONIAI
(ap) wa n nee Ww ud Vl
: 2 : 2 Yim,
sm < a =F Ye ZA
e w = ow & Cs
O = O = Oo
Za Sj <2 = Ps

SHLINS SAZIYVYPIT LIBRARIES SMITHSONIAN INSTITUTION NOILOLILSNI NVINOSHLIW

Sada y da ial
INSTITUTION

“yy

INSTITUTION NOILNLILSNI NVINOSHLINS S3IYVYEIT LIBRARIES

SONIAN

INSTITUTION

Sa1uvugin
INSTITUTION

NVINOSHLINS SZiu¥vadil_ LIBRAR!

= w ze o2) Zz
< Qe = < = x
> WS 5 ‘S = ws
oO NS . ae O x. O :
aN NSE 2 3 2
- YQ 2 é = =
toe ; : 5)
THINS 33 l\Yuvuad Valve SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIW
OL WJ < uJ = lJ 2
as ~ = “ z 4
<a = te a =
\N < a < % a < a
LY a : = ca :
a 2 £ S ws Oo
sua « Lal | —l = *
SONIAN INSTITUTION ~ NOILNLILSNI NVINOSHLINS gee luYVvdd ee LIBRARI ES suns
HY, oD 2 ae S) = ) 4
Ly, = = s F = =
iy it i= iis = = a
p z te te 2 S
“= wn = Ww =
SHLINS (S3 l\yvyual Si B RARI ES SMITHSONIAN INSTITUTION NOILALILSNI NVWINOSHLIA
2 YY: : 2 X\y i Gi
é Ae as a oe oO t’ > ND
Se ie Z i
SS = OS Bey =
me Y) oN Pa ~” +s =
SONIAN INSTITUTION ee NVINOSHLIWS S43 | YVaatT LI BRARI ES SMITHSONIA
Zz = ety ud = Ww Yy
= z Z e = = Vig
Q a. = eo M = a <x Yo G7
: : ; : : :
SHLINS SJIYVYUEIT LIBRARIES SMITHSONIAN INSTITUTION NOILONLILSNI NVINOSHLIA
PE lie = By. z rc
ee O = = : 2 =
Se 2 = ea = ee
A | a. = a
we N\ ra Be ey > as > %
Se = ‘ - 2
4 2 " 2 m = m
. 2 = o” = ”
SONIAN INSTITUTION NOILOALILSNI S3i1Y¥VYEIT LIBRARIES
z a n a wn z hae ”
ie ees < = _< Es
S ‘< N a By S ae ae z
SY ff 2 S z é
= Z “iy E 2 E S
= & GD = > = q >
(ep) Fue io) a Ww = Ps

SHLINS SaluVUdIT LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLIM

SRARIES
INLILSNI
LALILSNI
SRARIES
INLILSNI

SRARIES
NS
A

—-
—

UTION NOILNLIZSNI NVINOSAIING sslGvVadil LIGRARIECS SMIINSONIAN INSTITU LIVIN

a = i 2 Ovi 2
: z : = = Uy," :
= a Ce = <x ZZ Uj 7° = <
a = ce 5 a. a ee &
cs S = S = 5 2
a5 Za = z ae Zz =
74911 LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S3I1YVvadly
Be z ce Z ~ roe z Sp
w = “O S aan YS a
bk , 9 = ~y She kK xy
= =e aS = . \ a)
= Jae Ky Y = r- pasa = >
2 - “Gyp = = Z = Z
m o La m = m & m
top) — ” = Ww = 1S)
UTION NOILNLILSN!I NVINOSHLINS S3IYVYEIT_ LIBRARIES
wn z n Zz ee ” > =
~ = ons = is : = = =
-- 2 5 = owe =: 5 a
ey 2 gy = g ie B
> g E 27 e 2 NE =
Z a Z a Se a in

jugiy SMITHSONIAN INSTITUTION NOILNLILSNI NVINOSHLINS S31YVYUdII

S

NOILNLILSNI
LIBRARIES

UTION NOILALILSNI NVINOSHLINS S3SI¥VYUdIT LIBRARIES

INSTITUTION NOILNLILSNI
INSTITUTION NOILNLILSNI
SJIMVUSIT LIBRARIES

INSTITUTION NOILNLILSNI

SS1YvVYd!ITLIBRARIE

Sana Vel-delal

INSTITUTION

LIBRARIES SMITHSONIAN INSTITUTION NOILNLILSNI_ NVINOSHLIWS Saluvusl

“lf
IWINOSHLINS S3IYVYSIT_LIBRARIES

we

wn un ; =
< = z z < = fy, =
= Zz
. 6 = 5 2 2S = 5
2 8 : 8 2 g :
= z a 2, cs = =
> ae Sj =
= _ 2 5 A = w”
UTION NOILNLILSNI NVINOSHLIWS LIBRARIES
uu (op) Lu ” tu Ww Li
e = hie 5 = a= a
a S) ea o i O e
ain = — = — z at
VYGIT LIBRARIES SMITHSONIAN INSTITUTION NOILONLILSNI NVINOSHLINS S3IYVUEIT
7L Co z _ a
= S z ©
ee oe By, at 0 = ke
a) = 1 Upp Y =) =
> H-Yf ff > - -
a . “Yi a E =
m ” w”
Ds w = z

SMITHSONIAN INSTITUTION

NVINOSHLINS S31YVddl
NVINOSHLINS S3IYVYEtItT

NVINOSHLIWS
NVINOSHLIWS
SMITHSONIAN
SMITHSONIAN

SMITHSONIA

SMITHSONIAN INSTITUTION NOILNLILSNI

xttSON/2

‘> 1
= a
> 4

4

5

5
ARIES
LILSNI
ARIES

LILSNI
VMS
"3
ewe
ARIES
"3
wy
LILSNI

2
wr

US fo Ca) ee CR Re ery z -
WME OSE BAS aT, AER AT . ~4 t Pas 4 3 é

SHAD AAT ve ae : :

= .

SMITHSONIAN INSTITUTION LIBRARIES ;

Pity hind HEA Sea ne . ens
Shcnla pet ely j : f !
; Cheb tae ‘et eee : :
FBO Marae Sg agit, : ’ * Ra ’ ‘ ¥s My toR Te ‘ ot at ab of ‘
ey geen pray aT Nas Bet “6 ng i Set s pa holena i » 8 Ma ph EB
PANEA RTM h ee be rege ea emrnee EAS ; Mage i ae . j isis gels s 5 Liipleliadioy A ofiheie : a
bars ip) ges at raed keris Bey ut s Stat ft i Fs rig . i ; : tae, Lear se adic
i REED ID , ; , i : hia at Wek ORe! ‘ : us
Pores ered a eee ee 2 F, ¥ ¢ uy fie ot I i u '
Ut ehty ts A . : Ls ou aes . . am 3 Ci eat
he ee dite bx he reH : tt 1 tyne
Sti aaah HESS) Abas Bacvetet sd! f rhe ga ; ;
ns 2 fat oe teste ait 7 aah , ; he Sears aby ries San Px ; ‘ oe 5
MME Me NSB Cmte diate: G0 ink : , ; rok tt : cars soe i & y
Mone Bed, ‘ } a i ny ahd “, t : ?, Bs ae ‘ F
Siect, Fe bye c y : is : Eien Wee
Dad ghee sis hoth ace, ae yy Ws ess : “ I Fi / : A wee} i ‘ye ‘ ‘
OE At rcnr'Cy races oot : : AAS ? r 5
reel aa bp cig 4 5 Weisee Seah ihr actn rae, 3 ,
ak ais de by Z5: 9 oF 9, Fees R A bata f eae: zy i
Pease hs De tA NNR CAN ava Fp Avi) bit EF: = F
Pa ee RE ON At Ae : oe) : rh Hey ps ees
fhilet a nense pcan . an Pct {he iy : 2 tah pon AR BC it hs 1B 15 Myl- Fe yd ‘
SaCapays ; eoripe : a ri
Ld int ee ‘ of et : :
Ca es eee : . 5 $ : Stain t Lom ' 3 Fi
Cit Ri te 7 : if ‘ veal ‘ ’
FPO pe 4 > 4 % LAPOY 4 eat ' ‘ ‘ £
(bale, PPM TCR, eeha ts by Fatal Ae: i! Arp! : é
chm wath RAG tS pate thes Cf i 1A by are aie ae @ yet
LOY ore np op nor prety aE bab Skee, Dag ' Phy ‘ oil Oats: 5 . .
“Tym nt FoR, DIG / f q U Lee, ‘ * * God . .
Z ASI EG ET 2 sei te PS : vere ates 1 che Sayon ra 2 t aA H F
SWS hy ny ser, tied 4 raha 4 :
eee: fphs i H . ; , wih « . ‘ ‘ . ' ' ’ ‘
ORIEL Ui a Sen Tete ‘¢ fete agit if} : . ree
OREM AS ¥y : gta it ; . ENED ty an r ae i ‘ ‘ A ihe
iu PUBS Fog is 7. ‘ ; E Lg eae mul igh capris weet eee hit rll,
AEGEAN RET ee paige pi ek ee Def her steny ay rR GS iy f
LP iin hit Miah ee
TRU Ebon beseranpder ag a4 | 9 : ‘ Ot ELE ons ,
Fhe apt ea Frere ahaa ot) olla yo IF ms : , }
; APN A NITR I " GEN Gh sted as ctl ‘ :
PEPE Rede Gd piney ey Tee Le Shen 4
CO Sal te, : a7! see * iin tH ;
. ta 4 Ta : por ate! i
: A Tah diatates } (at phate
Sree orn wy rat i
been ‘ ‘we <} Fag Y 2
Wer ot : i re Peg : J
, Op Wich dae oe
ey (aun bean ; oe :
PP hep ity fe the thent i ' ’ : A
Fe Pere 4 ‘ ,
ft: ag J ‘ ‘
i ‘ , ry, , he .
BPs : Pept ; F 3
is
Neg } Orne is ote b :
Le eae eT Salt “1 Ae
re Pitas Feat ite cee i
inp Cis ey . a - . 4
EA TI wk ‘\ ae Ok! :
ba ar POS a “
MPR RWING Pou pia sous Buia n PAed GAD, Ls :
[ ; ¢ Feds Orr at boat ‘ “ :
Leer a pees rs Ee eps Hy . lay : 1 el ae .
hei S f hi 7 , a OEY : BAN i lk rSrepe : ;
es 4 f ee ? ‘ Park ee Ok va . : 5 ' y ; Mi ;
Ore 5 p can 5 mile yb fe a J ‘ d
eee Seah a 2 ‘ : a ‘i ts ey Ie, at \ \
ney Ge AU iy 4 Fi i : x ‘ Meas Ha tae ‘
pisern Y SPM i: eae \ ,
te Eden : epee y ea tire aR ae e el: . He. Big ty de of vers
US EARS EPD VV yliek vgs 4 ge igs é > : Ce ' ' é ‘
babobeyas B } i : hey je th ‘ Phage K
ad eae . ie ‘, A t b ‘ . } °
UPlORY ees Oe 3 : if 5 : ea
tah: he Wee ie : y a Sara kn Ad open ; ‘
SSP Set ee. ri tne ,, . , 2
We Say hoe ny ee re w ; ch 4 :
“eSBs ap anere, . i vey aN ’ e ¥
i a SPS tie H ter aL a a ‘ x ‘
° PEPE UE PP a eg Log. aa
¢ NP aw Sy ay ‘ = Sale i « ee 7 . «
ASP dY Hyak ‘ edly ‘ wae ; 4 : : oF : f ut i ‘ 1s be ‘
pe was okaiiy ‘ aa. i itp : : pooh ed I ‘ 3 A
4 WR oath gy ab pe R : 3 “4 me Ralbrey, ue 4 ,
Bonners ut Up ik AE te ry A re (os y E ‘ R - ‘ c at ‘ yy ‘
ane RAE Lee ae Hae gi hal 7 aha i Wey eet NEN : i
oa ete ie ee ‘ b teh y t Se WYN yy, . ' . . aA cs & -
s- VP orate y: 75 y AYA dee PER iota or " bie yh . ;
Bt Ck eter Wa fe BON ee ease jetta! " rw hy hey : i . ‘
FYE PEST LP SAGE he BOE BIS ye iy, ao tenn ' i 2 a
PRE RD UE ice ee et 5 1 ‘ cet ‘
WAL Woivaginy : thee Chak a) £
a ae ae Leaky Up. Dap . : . :
oor. Sontag k. v
Vesa Me wile i ¥ ‘ , Petqalivg “ i
i RE BEE ie Lp Rub dy Wet p if ) si X ‘ é r aan, (! M b , . pa
pie EEA TOPE VET EIAYS eUT SAI ERT eae : rac : } EN 5 oO 3 Fa oe ‘ y
REO OP Uy VERS MEE S Ob eg USS aie \ j | Mig | : ‘aa < .
aa SNL Aye waa i : . " 3 ‘ mph PH 5 :
WARE iin tee { 4 3 ; : ‘ Meare Sa '
L tree a Yee ° 5 A i ‘ re : F
WERE WS WE Kaien Ue aes r : : a : ( é
OL BME SIs fi , (ies i ;
ieee eee a al Sea . OS:
POUUEN Rg gets lye all i
Urey ube es, fe ae Baek LPT ERG : i .
wien Le ut i “4 .
EAL oot ‘ re 4 3 ‘ ‘ i i
HU Ty
Pat