Full text of "Annals of the South African Museum = Annale van die Suid-Afrikaanse Museum"

See other formats

ea ey

I ee eee cepuninsthaite the the “ait Mivtnettartbe the: Ba

mate rat Sean tharth te ae tin taints
tor tee teree mann f

ee puss dmieaaceenaarabeenaa nee
athe ee Dynes tae lh Vee Some toe Map hte ities et hag HR Ht
wien arte geipiine ar tinsle Tie eatin eho Metine Sahih ae Ve Sahat

ne
OL yaathamtienaie

are es
meio

ON ne ern

ee te TNS

ire —notind ee tee en
ritchie in Pe ante Thee em Meth co ee EPO eA
™ arteth

pa el
in Aenede>

ab ind

eg pe eg ag in oe
nn gO meg went
err snows

on “es

; \

gt apy ot - . ore a ~ ~~ ~ w yew
ee ’ a on ee here re rn
us eh Mo ynacas

fh
|
|

“4
j as
H2E FF isi

T oe

i yi

oie HELE \
Lt om
etl! a “ al tes
" YES

Bee
SS %,
WX Sea

saat

BAA ACY

ee
te
OLIN,

* eH
aa,
a

peat
BOT eT aay

S uy

J So Mail se %, Madi ot , Sece)
oe hoy en “ii iil aS Oy, a I os
= = «

_
SS My, =

Poets
ire
Soul’

40> ANNALS OF THE
ea SOUTH AFRICAN MUSEUM _ re a

VOLUME XLIV |

‘

CAR PRALSURIA) CLARE Aine et Mieaae
‘ ¢ _ ¥ vad 1 ba tiad|
hut
|
f ‘
; }
NM t

ANNALS

OF THE

SOUTH AFRICAN MUSEUM

VOLUME XLIV

TRUSTEES OF THE SOUTH AFRICAN MUSEUM
ie aE BY

ALOE ALTA en ae
® PRINTED IN THE UNION OF

THE RUSTICA PRESS, PTY., LTD.,

507,68

Vv. 44
195 7- SF

LIST OF CONTRIBUTORS

BARNARD, K. H.

Contributions to the knowledge of South African Marine Mollusca. Part I.
Gastropoda: Prosobranchiata: Toxoglossa (published July 1958)

- Boonstra, L. D. |
| The Moschopid Skulls in the South African Museum (published October 1957) ..

BREUNING, S.

Beitrag zur Kenntnis der Lamuiden Stidafrikas (Cerambycidae, Coleoptera) (pub-
lished November 1958)

BROEKHUYSEN, G. J. and Tayior, H.

The ecology of South African estuaries. Part VIII. Kosi Bay estuary system
(published April 1959) S me Me

Crompton, A. W. and ELLENBERGER, F.

On a New Cynodont from the Molteno Beds and the Origin of the Tritylodontids
(published October 1957) .. , ny 5 % a: iat

ELLENBERGER, F.
See Crompton, A. W.

Miciarp, N. A. H.

Hydrozoa from the coasts of Natal and Portuguese East Africa. Part I. oa
blastea (published July 1958)

Hydrozoa from the coasts of Natal and Portuguese Dee ie, Pane Us eae

blastea (published April 1959)
- Murray, D.

The genitalia of the genus Phasis and allied genera (Lepid. Lycaenidae) (published

November 1958) hs i ‘ie ae a ou of ae
SCHEIN, H.

Bestimmungstabellen der Heterochelides (Coleoptera, Lamellicornia, Hopliini) mit
Ausnahme von Heterochelus Burm. und Ischnochelus Burm. Mit Neubeschrei-
bungen und Bemerkungen (published November 1958) ..

Scott, K. M. F.

Some new Caddis Flies (Trichoptera) from the Western Cape Province — II

(published May 1958)
TAytor, H.
See BROEKHUYSEN, G. J.

WINTERBOTTOM, J. M.

Review of the Races of the Spike-heeled Lark, Certhilauda albofasciata Lafresnaye
(published May 1958)

On the Races of Lybius leucomelas (Bodd.) in South Africa (published May 1958) ..
A review of the subspecies of the Yellow Canary, Serinus flaviventris (Swainson)
(published April 1959) :

SMITHSONIAN

Page

279

LIST OF GENERA AND SUBGENERA

Acrobela
Aglaophenia ..
Aloeides
Amphisbetia ..
Antennella
Anthene
Antimitra
Asthenotoma ..
Athripsodes

Avenantia

Bela
Bela‘...
Bellardiella
Bimeria
Bizanus

Campanularia
Certhilauda
Chrysoritis
Cladocoryne ..
Clavatula
Clionella
Clionella

Clytia ..
Conus .. re
Corydendrium
Corymorpha ..
Cricodon
Crudaria
Cylindrocrates
Cymatosyrinx
Cythara

Daphnella
Delphinognathus
Desmolycaena
Diademodon ..
Diaplochelus ..
Diarthrognathus
Dichelus
Dicranocnemus
Diplomeriza ..
Dirphya

Drillia ..
Dynamena

PAGE

254, 266

170

275

oF ko BOO
94, III, 140, 143
sh a 141
143

172

301

299

Senne 397s

246, 263

f 138

115, 150

160
15, 33
276
ay 7
255, 266

230, 257
250, 264

227

94, 118, 127
183

PAGE
Eudendrium .. 302
iH f
Filellum 175
G x
Gemmula 109
Genotia Lh ; 112
Gomphodontosuchus Me bi 10
Goniaspidius .. 247, 263
H
Halecium 168
Halicornaria . 215,
Halopteris 200
Hastula 78
Hebella 176
Hincksella 181
Hydractinia 305,
Hydrichthys .. 309,
K
Kirchenpaueria oe 203
Li
Lamiasaurus . 15
Leptecho 48
Leptocerus 40
Lienardia 116
Lybius. . 69°
Lytocarpus 220
M
Macrodicranocnemus 250, 264
Mangilia 151
Melatoma 98
Mitre morpha. . 162
Monostaechas 204
Moschognathus 15 35
Moschoides 36.
Moschops 15, 29
N
Nanniscus 254, 266
. O
Obelia. . 174
Oberea ‘ 227
Oligokyphus . 12
Omacrates 24.0, 260
Omocnemus .. 249, 264.

us ate mh oe fa Pee ed, a ‘ ie *®) ee let Ay

NEW GENERIC AND SUBGENERIC NAMES PROPOSED
IN THIS VOLUME

_ Page

:
|
|

Paragattya..

Pennaria ae
Phasis ..- aS
Philbertia

_ Phytoecia
Pleurotoma ..
Pleurotomella
Plumularia ..
Pnigalion the

Podocoryne ..
Poecilmitis ..
Protacmon ..
Pseudoconizonia
Pycnotheca

Rectoscutaria..

Salacia oe
Scalenodon ..
Scalenodontoides

LIST OF GENERA.

—

: 208°
.) 306
: 269
117, 158
ai i, eee
126, 14.7
BaD oY.
ev, ZOO
> 15535
; 307
- 274
. )
Se 7
MPa 0
248, 263
Bd 186
A 7

Sertularella ..
Sertularia (71),
Sophronica .. _
Stegopoma ..-
Stereothecay —--
Surcula Ba,
SUC te an ae
Synthecram 0a

Werebra vue ae
Thecocarpus ..
Thesbia Hl at aie
Thyroscyphus
Traversodon ..
Tribolodon ..
Trirachodon ..
Tubularia

Turris . . ps

Zygophylax

; ore South African Museum. By cee Dirk ae ox : :
ae Galagee) NN ee a eee

Dry

\ , A
inl ,a oa
YY ig : o.
: i rs oa at
5 by
‘ BE 4 2
he
3 Pa
nau 4
" Lae
th
3 ke
ye
- ‘
a e
2
? ‘
:
‘
i XS
{ty <o'D
a
i a 3
Pe
\ s
r
y be
a >» ¢
ee -

ON A NEW CYNODONT FROM THE MOLTENO BEDS AND
THE ORIGIN OF THE TRITYLODONTIDS

A. W. Crompton ane F. Ellenberger
S.A. Museum, Cape Town Université de Paris

(With 5 figures in the text and 1 plate)

CONTENTS
Page
Introduction : : s : 2 . I
Description of mandible . 2 : | 3 : 2
Relationships : : , : é : 7
Origin of tritylodontids . : : ; ; II
Summary. : - mete ; - : 13
References : : : : : : 13
INTRODUCTION

Early in 1956 three fossiliferous sites were discovered on Morobong Hill
in Southern Basutoland. This hill is situated on the north-east bank of the
Orange River a few miles north of the point of exit of this river from Basutoland.

At site A, a little way below the P.E.M.S. school on the northern slope of
the hill, several fragments of bone were found which fitted together to form the
cynodont jaw described in this paper. All the fragments were found in an area
not larger than 5 sq. ft. At site B, approximately 100 yards west of site A, several

fragments of what appear to be a cynodont snout were found in a ploughed field.
These fragments have not yet been prepared. At site C, on the southern slope
of the hill, large dinosaur bones were found. This site appears to be of the same
height as sites A and B. It appears that all these sites occur well below the
typical Red Bed facies indicating that they occur in the uppermost layers of the
Molteno Beds. Cynodonts are well known from the Cistecephalus, Lystrosaurus
and Cynognathus zones, but only one specimen of Cynidiognathus longiceps
(Boonstra, 194.7) has been described from the lower Molteno Beds of Basutoland.
_ This species also occurs in the Cynognathus zone. The only therapsid remains
which have been described from the Red Beds are Tritylodon, Tritheledon,
Pachygenelus and Lycorhinus. Several cynodonts have been described by Parring-
ton (1946), von Huene (1950), Boonstra (1953) and Crompton (1955) from
the Manda Beds of Tanganyika. These beds appear to be of the same age as
_ the Molteno Beds of Southern Africa. Cynodonts of the same age have also
been described by von Huene (1935-42) from the Rio do Rasto Beds of Brazil.

I
VOL. XLIV. PART I.

2 ANNALS OF THE SOUTH AFRICAN MUSEUM

The discovery of a well-preserved cynodont jaw from the upper Molteno
Beds, which have been assumed to be barren, is of importance and warrants a
full description.

We are indebted to the Paramount Chief of Basutoland for permission
to undertake a series of excavations. The senior author wishes to record his
cordial thanks to the Council for Scientific and Industrial Research for a
grant enabling him to visit the Eastern Free State and Basutoland. Special
thanks are due to Mr. F. R. Parrington for his help and advice.

DESCRIPTION OF MANDIBLE

Order THERAPSIDA

Suborder THERIODONTIA

Infraorder GYNODONTIA

Family TRAVERSODONTIDAE

Genus Scalenodontoides gen. nov.

Genotype: Scalenodontoides macrodontes gen. et sp. nov.

Generic description. A large specialized gomphodont cynodont with
procumbent incisors and lower postcanine teeth larger, but similar to those of
Scalenodon angustifrons. ‘The postcanine teeth are larger than those of any known
gomphodont cynodont. The type specimen is at present in Paris, but it has
not yet been decided in which institution it will be placed.

Scalenodontoides macrodontes gen. et sp. nov.

Only the tooth-bearing portions of the lower jaw could be fitted together
from the numerous fragments. The jaw is exceptionally large and complete
would probably have measured 30 cm. in length. The matrix consists of a
hard calcareous mudstone, grey to blue in colour. A hard layer containing a
high percentage of iron oxide lies between the matrix and the outer surface of
the bone. This made preparation extremely difficult as this layer could not be
cleanly removed from the bone.

A marked characteristic of the jaw is the massive symphysis with an oblique
anterior surface (fig. 2, plate 1) and procumbent incisors. The posterior surface
of the symphysis is indented by a deep impression. The lower edges of the rami
have been deflected outwards as a result of dorsal pressure.

Dentition

Only the roots of the first and second incisors are preserved on the left, but
the crowns of the right incisors are fairly well preserved. The cross-section of
the first left incisor at the alveolar border is circular to oval (anterior-posterior
measurement 1:5 cm., medio-lateral measurement 2:0 cm.). The remnant of —

NEW CYNODONT FROM MOLTENO BEDS AND ORIGIN OF TRITYLODONTIDS 3

the root in the second left incisor is badly preserved and the third left alveolus
is empty. The damaged tip of a replacing incisor is visible in the first right
alveolus. In cross-section this tip is crescent-shaped with a marked ridge in
the centre of the posterior surface. The enamel of the anterior surface of the
crown is thick in contrast to the exceptionally thin enamel on the posterior
surface. The second right incisor is the best preserved of the series. This tooth

a
ae SY aS N r
) \
i Ly - a a
eres \ G71
BS

a
77
el

= eae

Vv ha ‘\
sof \\
- - H
= Narr]
SN ae

a
3)

Fic. 1. Scalenodontoides macrodontes gen. et sp. nov. Dorsal view of lower jaw,
x4. U.P.C., Unerupted postcanines; R.I., Replacing incisor.

extends almost directly forwards from the mandible and this direction does not
appear to be the result of distortion. This incisor is chisel-shaped with a fairly
flat anterior surface and rounded posterior surface. The length of the crown
measured from the posterior border of the alveolus to the apex is 3-3 cm. The
anterior-posterior measurement at the alveolar border is 1-5 cm. and the
medio-lateral 2-0 cm. The central ridge visible on the posterior border of the

4 ANNALS OF THE SOUTH AFRICAN MUSEUM

replacing incisor in the adjacent alveolus is not present in the second incisor
which appears to indicate extensive wear. |
A fracture through the mandibular symphysis has exposed a section
through the inner portion of the alveolus of the second right incisor. A shallow
pocket is visible above the root of the functional incisor in which the crown of a

i
h\i\\ \ ys
) i DD

Fic. 2. Scalenodontoides macrodontes gen. et sp. nov. A, lateral view, and B, inner view of lower jaw,
x4. D.I., Developing incisor; P.G.D., Postdentary groove; U.P.C., Unerupted postcanines.

replacing tooth is visible (R.I., fig. 2B). Only the medial edge of this tooth is
visible and in order to expose the crown a considerable portion of the mandible
would have to be removed. |

The crown of the third right incisor is fairly badly damaged, but its form
appears to be identical to that of the second.

The incisors appear to be adapted to a herbivorous diet and are rather
similar to the cropping incisors of ungulate mammals. The presence of replace-

NEW CYNODONT FROM MOLTENO BEDS AND ORIGIN OF TRITYLODONTIDS 5

ment in so large a specimen is of interest. It suggests that, as in Scalenodon
(Crompton, 1955), the incisors may have been replaced a number of times.

The fact that the crowns of the canines only partially fill the alveoli appears
to indicate that the incisors have recently been replaced. The apices of the
crowns of both the canines have been destroyed. In lateral view the crown
appears to be slightly recurved.

The canines are ovoid in section and faint ridges are present upon the
anterior and posterior edges of the crown. The maximum anterior-posterior
measurement of the canine alveoli is 2-9 cm. and the medio-lateral measurement
2°2 cm.

A fracture passing vertically through the left ramus immediately behind
the canine has exposed the crown of a partially developed canine medial to the

Fic. 3. Scalenodontoides
macrodontes gen. et sp.
nov. Lower view of
typical ? left lower post-
canine, X 2.

Se
ES

bi
ty

= =
a

root of the functional canine. It appears therefore that the canines are replaced
several times.

A marked feature of the postcanine series, consisting of six functional and
two unerupted teeth, is that they all have the same crown pattern and there is
very little difference in size between the anterior and posterior teeth. This is
in contrast to the Diademodontidae where there is a marked differentiation of
the postcanine row both in form and size.

The crown of a typical postcanine tooth (fig. 3) is roughly rectangular
when viewed from above. The anterior region is dominated by two massive
crescent-shaped cusps; the labial higher than the lingual and directed more
dorsally than posteriorly, whereas the lingual cusp is directed more posteriorly
than dorsally (see fig. 2B, 6th postcanine). A well-defined V-shaped depression
is present on the anterior surface of the crown between the two cusps. There is
no evidence of an anterior cingulum. A broad, slightly concave heel is present
behind the crown. This heel is flanked by two ridges which extend backwards
alongside the edge of the heel from the apices of the main anterior cusps. The
posterior edge of the heel is deflected upwards to lie against the anterior edge
of the succeeding tooth. There is no evidence of any minor cusps on the edge
of the heel. As in the case of the incisors the postcanine teeth are characterized by
thick enamel on the anterior surface of the two main cusps, whereas the enamel
on the posterior surface of these cusps and upon the heel is exceptionally thin.

The postcanine teeth of the right ramus are badly preserved and the
following description is based upon those of the left.

6 ANNALS OF THE SOUTH AFRICAN MUSEUM

The anterior postcanines have been slightly dislocated from their sockets.
The length of the diastema measured from the anterior border of the first
postcanine alveolus to the posterior border of the canine alveolus is 1-2 cm.
The length of the functional postcanine row of six teeth is 11-3 cm. In the
anterior postcanine both the anterior cusps have been destroyed, but sufficient
remains to indicate that these cusps were present. The labial portion of a
broad heel is well preserved as is the ridge extending backwards along the
outer side of the heel. In the second postcanine the apex of the labial cusp and
the entire lingual cusp have been destroyed. Sufficient remains of the labial
cusp to indicate that it was not as high as the labial cusps of the more posterior
teeth. The heel of this tooth is well preserved. The V-shaped depression of the
anterior surface is not as prominent as in the posterior postcanines.

The apex of the labial cusp of the third postcanine has been destroyed, but
the lingual cusp is well preserved. The length of the heel behind the anterior
cusps is slightly less than in the preceding two teeth. The lingual cusp of the
fourth postcanine is well preserved, but the labial cusp is badly damaged. Well
shown in this tooth is the ridge extending backwards alongside the heel from
the lingual cusp. The V-shaped depression on the anterior surface of the crown
is prominent. In the fifth and sixth postcanine teeth (plate 1, fig. 2) all the
features are well preserved except that the posterior edges of the heels are
damaged in both. A series of small ridges are present upon the inner surface of
the crown behind the apex of the lingual cusp.

Lying partially behind and below the sixth postcanine on both sides of the
jaw is the crown of an unerupted tooth. This tooth lies medial to the coronoid
process which starts adjacent to the sixth postcanine. Below the crown of this
unerupted postcanine the mandible is exceptionally thin, indicating that no
root to this tooth has as yet been formed. The crown of this tooth has the same
pattern as the functional teeth, except that the labial cusp is considerably
higher than those of the functional teeth. The apex of this cusp lies below the
posterio-lateral corner of the sixth postcanine. There are indications of the
labial cusp of an additional unerupted postcanine behind the seventh. Unfor-
tunately the specimen is badly damaged in this region. The following are the
measurements of the postcanine teeth:

Maximum length in cm. Maximum breadth in cm.
if 1°8 1-6
2 18 ey
ae 1°8 D7
4. 1:8 17
5. 2:0 1°7
6. 2°0 ie |
7: iy ire
8. — —_

NEW CYNODONT FROM MOLTENO BEDS AND ORIGIN OF TRITYLODONTIDS 7

Although the crowns of the postcanine teeth are very similar in both size
and form, progressive changes can be observed in a posterior direction.

(1) There is a gradual increase in the size of the teeth, the height of the cusps
and the prominence of the V-shaped depression, on the anterior surface
of the crown.

(2) There is a gradual diminution in the size of the heel. The ridge extending
along the labial side of the crown from the labial cusp arises more steeply
in the posterior than anterior postcanines.

Increase in the size of the cusps in a posterior direction is a normal feature
of several cynodont dentitions, e.g. Scalenodon and Trirachodon, but although
present, it is not marked in Scalenodontoides. The other progressive changes are
probably the result of wear, but this feature is not marked. The fact that the
wear pattern increases progressively in an anterior direction appears to indicate
that there has been no recent replacement of the postcanine teeth. This is in
contrast to the incisor and canine regions, where active replacement is still
taking place. Fracture through the rami exposed the roots of the postcanines
in several places, but no replacing teeth were observed.

The roots of the functional teeth are stout and are directed slightly forwards.

RELATIONSHIPS

Taxonomic position of Scalenodontoides

One of the most reliable diagnostic features of cynodonts is the crown
pattern of the postcanine teeth. In figure 4 the crown views of the main types
of cynodont teeth have been illustrated. The only cynodonts which have lower
postcanines with two anterior cusps and a heel behind are Scalenodon (fig. 4U),
Traversodon (fig. 4V) and Scalenodontoides (fig. 4W). Scalenodon is from the Manda
Beds of ‘Tanganyika, Traversodon from the Rio do Rasto Beds of Brazil and
Scalenodontoides from the Molteno Beds of Basutoland. All these deposits are
--eonsidered to be of approximately the same age, i.e. Middle Triassic. There is
a marked difference in the actual size of the postcanine teeth of these three
genera and they occur in widely separated areas, but the crown pattern is so
similar in all three that it is reasonable to conclude that they are related. The
mandibular postcanines of Scalenodontoides are the largest known gomphodont
cynodont teeth.

There is no evidence in Scalenodontoides or Traversodon of the anterior
cingulum or heel cusps found in Scalenodon. This can possibly be explained as a
result of wear as they are lost in older Scalenodon specimens. An additional
similarity between the three forms is that, with the exception of the small
sectorial postcanines at the back of the postcanine row in the older Scalenodon
specimens, the crown pattern of all the postcanine teeth is the same and there is
a gradual increase in the size of the teeth in a posterior direction. This is in
contrast to the condition found in Diademodon. In both Scalenodon and Scaleno-

8 ANNALS OF THE SOUTH AFRICAN MUSEUM

RED BEDS
RHAE TIC

a Ce

AAw) AAI) Zu) Yu) BBuw)
MOLTENO BEDS
MANDA BEDS ;
RIO DO RASTO SERIES
Wy 1) Res
Sw) Sic)
@ O i
Ri)
Pa) Qw)

CYNOGNATHUS

LYSTROSAURUS

Bsc
1 ©
a &

CISTECEPHALUS

Q &

> &
0e |z &

Fic. 4. Crown view of typical postcanines of cynodonts, and tritylodontids, x 14. (U), maxillary,
and (L), mandibular postcanine.

A, Levachia. B, Silphedestes and Silphedocynodon. C, Baurocynodon. D, Nanictosuchus and Proto-
cynodon. E, Glochinodon and Galesaurus. F, Microconodon. G, Notictosaurus. H, Thrinaxodon and
Nythosaurus. 1, Sysphinctostoma. J, Cistecynodon and Nythosaurus. K, Cynognathus. LL, Tribolodon.
M, Diademodon. N, Trirachodon. O, Cynidiognathus. P, Unidentified cynodont from Manda Beds.
Q, Unidentified cynodont from Manda Beds. R, Unidentified cynodont from Manda Beds.
S, Gomphodontosuchus. T, Cricodon. U, Scalenodon. V, Traversodon. W, Scalenodontoides. X, Tritylodon.
Y, Bienotherium. Z, Sterognathus. AA, Oligokyphus. BB, Pachygenelus.

NEW CYNODONT FROM MOLTENO BEDS AND ORIGIN OF TRITYLODONTIDS 9

dontoides there is little or no evidence of tooth replacement of the postcanine teeth
whereas replacement appears to occur frequently in the incisor and canine regions.

Classification of gomphodont cynodonts

Several trends of development of the postcanine teeth can be traced in
fig. 4. The Cistecephalus and Lystrosaurus zone cynodonts (A—H) of the families
Procynosuchidae and Galesauridae (Thrinaxodontidae) have either simple
conical postcanines or shearing postcanines with minor cuspules anterior and
posterior to the main cusp. Sysphinctostoma (1) and Cistecynodon (J) appear to be
surviving members of the Galesauridae in the Cynognathus zone. The large
shearing postcanines of the Cynognathidae found in both the Cynognathus zone
(K) and Molteno Beds (O) could have developed from the type of teeth found
in the Galesauridae. Pachygenelus (BB) from the Red Beds appears to be a
surviving member of either the Galesauridae or the Cynognathidae.

The postcanines of forms such as Tribolodon (L) and the unnamed cynodont
from the Manda Beds (P) appear to have developed from the Galesauridae.
In these forms the crown has extended slightly lingually to form a cingulum
supporting minor cuspules.

Watson and Romer (1956) have placed the gomphodont cynodonts in two
families; the Diademodontidae and the Gomphodontosuchidae, the latter for
a single South American genus. Von Huene (1956) and Haughton and Brink
(1954) have placed all the African gomphodont cynodonts in the family
Diademodontidae. The two South American forms Gomphodontosuchus and
Traversodon, von Huene (1950) has placed in the family Traversodontidae. The
postcanine teeth appear to indicate that gomphodont cynodonts developed in
two independent directions from cynodonts with simple postcanine teeth and
therefore should be divided into at least two families.

In both cases there is a tendency to widen the teeth transversely. In
Diademodon (M) and Protacmon there are several cusps in addition to the main
cusp in the lingual and labial edges of the crowns of the maxillary postcanines.
A series of ridges extends transversely across the crown, but there is no evidence
of a central cusp. The mandibular postcanines are similar, but the accessory
cusps may not be so well developed on the labial and lingual borders and the
crown is more circular in outline compared with the oval shape of the typical
maxillary postcanines. An additional characteristic of Diademodon is that there
is a marked differentiation of the postcanines; the anterior teeth are simple
cones, those of the middle are transversely widened and the posterior teeth tend
to be sectorial. The postcanines of the unidentified cynodont from the Manda
Beds (R) are clearly similar to those of Diademodon. It is probable that the
Diademodon-type of postcanine developed from the Thrinaxodon-type through
stages represented by forms such as Tribolodon (L) and the unidentified cynodont
from the Manda Beds (P).

In the mandibular postcanines of the unidentified cynodont from the
Manda Beds (Q) there are two main cusps on both the lingual and labial

Io ANNALS OF THE SOUTH AFRICAN MUSEUM

borders, but the centre of the crown is hollow and the tooth is placed obliquely
in the jaw. The development of more than one main cusp on the outer and inner
edges and the absence of a central cusp appear to indicate that this type of
crown pattern could be developed from that of the Dzademodon-type. The
postcanines of Gomphodontosuchus (S) are badly preserved but they appear to be
of the same type as those of the Manda Beds form. Watson and Romer (1956)
have placed Gomphodontosuchus in the separate family, the Gomphodonto-
suchidae.
The other direction in which gomphodont cynodonts appear to have
developed is illustrated by Trirachodon (N) from the Cynognathus zone.
Unfortunately only the maxillary postcanines of this form are completely
known. In the maxillary postcanines there is only one main cusp on the lingual
and labial borders and a well-developed central cusp is present. A row of small
cuspules is present on the anterior and posterior borders of the crown. The
Trirachodon-type of crown pattern is also found in the Manda Beds genus
Cricodon (T). In this form the mandibular postcanines have the same crown
pattern as the maxillary postcanines. A feature of Cricodon and Trirachodon is
that the postcanine row is not differentiated to the marked degree it is in
Diademodon. In Cricodon two sectorial teeth are present at the posterior end of
the postcanine row, but this type of tooth has not been recorded in Trirachodon.
The presence of only one main cusp on the outer and inner edges of the tooth
of the Trirachodon-type appears to indicate that this type of gomphodont tooth
was developed from a simple conical tooth by transverse widening rather than
from the Thrinaxodon-type, where the development was initially longitudinal.
A variation of the Trirachodon-type postcanine is that found in Scalenodon
(U). In the maxillary postcanines of this form there is only one main cusp on
the lingual and labial edges and one central cusp. The labial cusp has a vertical
inner surface and the central cusp lies towards the lingual edge of the crown,
but on the basis that this crown has only three main cusps arranged upon the
same transverse plane, it appears to be related to Trirachodon and Cricodon. The
mandibular postcanines, however, are fundamentally different from those of
Trirachodon and the crown consists of two high anterior cusps and a broad heel
behind. It has already been shown that, on the basis of the mandibular post-
canines, Scalenodon, Traversodon and Scalenodontoides appear to be closely related.
On the basis of the form of the crowns of the postcanine teeth it appears
that the gomphodont cynodonts can be divided into two families:

(A) Diademodontidae

More than one cusp on the lingual and labial borders of the crown and no
central cusp.
(a) With low ridges connecting the lingual and labial crowns, e.g.
Diademodon, Protacmon, etc.
(b) Centre of the crown hollow, e.g. Gomphodontosuchus and the unnamed
specimen from the Manda Beds (fig. 4Q). Watson and Romer (1956)

NEW CYNODONT FROM MOLTENO BEDS AND ORIGIN OF TRITYLODONTIDS II

are perhaps justified in placing this type in a separate family, but
nevertheless it appears to be closely related to Diademodon.

(B) Traversodontidae

One cusp on both the lingual and labial borders of the maxillary post-
canines and a central cusp. All three cusps arranged upon the same transverse
dlane.

(a2) Upper and lower postcanines of the same pattern with minor cuspules
upon the anterior and posterior margins of the crown, e.g. Trirachodon
and Cricodon.

(6) Maxillary postcanines with three cusps, but mandibular consisting of
two anterior cusps with a heel behind, Scalenodon, Traversodon and
Scalenodontoides.

ORIGIN or TRITYLODONTIDS

Both Watson (1942) and Kiihne (1956) have stressed the similarity between
the skulls of tritylodontids and cynodonts and have concluded that the tritylo-
dontids were derived from the cynodonts, but a more precise statement could
not be made. Haughton and Brink (1954) have classified tritylodontids as
cynodonts. A study of the postcanine teeth of gomphodont cynodonts appears
to throw some light on the origin of the tritylodontids.

The maxillary postcanine teeth of the tritylodontids consist of a variable
number (not more than four) of transverse rows of three cusps. The majority
of these cusps are crescent-shaped with the concave surface directed forwards.
The mandibular postcanines consist of a variable number (not more than four)
of transverse rows of two cusps. The majority of these cusps are crescent-shaped
with the concave surface directed backwards. Kiihne (1956) and Butler (1939)
have shown that the number of transverse rows varies in a single dentition and
that the number of rows tends to decrease in a posterior direction. Within the
Tritylodontidae the number of transverse rows varies considerably. In Stereogna-
thus (Simpson, 1928) for example. there are only two transverse rows in the
maxillary postcanines.

In Trirachodon and Scalenodon the crown pattern of the maxillary post-
canines consists of one transverse row of three cusps. This has been given as a
diagnostic feature for the cynodont family Traversodontidae and it therefore
appears that this family is more closely related to the tritylodontids than the
Diademodontidae. In Scalenodon the relationship with the tritylodontids is more
marked for in this genus not only do the maxillary postcanines have three cusps
in a transverse row, but the mandibular postcanines have two crescent-shaped
cusps arranged upon the same transverse plane with the concave surfaces of the
cusps directed posteriorly. .

Scalenodon, Traversodon and Scalenodontoides are the only known cynodonts
which have crescent-shaped cusps which are similar to those found in the tritylo-

I2 ANNALS OF THE SOUTH AFRICAN MUSEUM

dontidae. The anterior region of the mandibular postcanines of the tritylodonts
and especially Oligokyphus is almost identical to the postcanines of Scalenodon.
The lateral views of the mandibular postcanines of Scalenodon and Oligokyphus
have been compared in figure 5. In both there are two high anterior cusps
with a deep depression between them. In both the anterior surfaces of the cusps
arise steeply and are convex in contrast to the gentle sloping concave posterior
surface. In Scalenodon there is a minor cuspule on the anterior surface of the

Fic. 5. Lateral views of the postcanines
of: A, Scalenodon, and B, Oligokyphus.

labial cusp and in Oligokyphus there is a minor cuspule on both the lingual and
labial cusps. Superficially it would appear possible to derive the double-rooted
tritylodontid cheek teeth by the fusion of two or more teeth of the Scalenodon-
type. On the other hand the additional cusps of the tritylodontid teeth could be
developed from the additional cusps on the heel of the postcanines of Scalenodon.
Kiihne (1956) has shown that at the end of the postcanine row of Oligokyphus a
progressive reduction of the elements takes place. In the ultimate tooth the
crown consists of two anterior crescent-shaped cusps behind which there is a
fairly broad heel. On the posterior edge of this heel two small cusps which are
not crescent-shaped are present. It appears, therefore, that one row of cusps
can be considered as a unit of which there are more in the more anterior teeth.
The ultimate tooth of Oligokyphus is fundamentally the same as the postcanines
of Scalenodon and consequently there is a complete overlap in the range of
variation of the postcanine crown pattern of Oligokyphus and Scalenodon, and this
appears to indicate a relationship between Scalenodon and the tritylodontids.

An additional similarity between Scalenodon and Oligokyphus is that in both
the anterior cheek tooth (Kiihne’s zero tooth) is lost during growth.

On the basis of the above evidence it appears reasonable to conclude that
the cynodonts with the type of postcanines found in Scalenodon are more closely
related to the tritylodontids than any other known gomphodont cynodont. As
these cynodonts occur in older beds than those in which tritylodontids are
found it is possible that the tritylodontids were derived from:this or a closely
related group of cynodonts.

In an earlier paper (Crompton, 1957) it was concluded that Diarthrognathus
broomi and Broom’s Ictidosaurian A were derived from the scaloposaurids. If

1 dais,

bet MA ep Sr

NEW CYNODONT FROM MOLTENO BEDS AND ORIGIN OF TRITYLODONTIDS 13

this is the case the Tritylodontidae which appear to have been derived from the
cynodonts should be removed from the infraorder Ictidosauria as this infraorder
was created for the classification of Broom’s Ictidosaurians.

The close similarity of several of the features of the skulls of Diarthrognathus
and Tritylodontidae appears to be the result of parallel evolution and does not
indicate a true relationship between the two groups. The Tritylodontidae
should be either placed in a new infraorder or, as has been done by Haughton
and Brink (1954), classified as aberrant cynodonts.

SUMMARY

(1) The lower jaw of a new cynodont, Scalenodontoides macrodontes gen. et
sp. nov. from the upper part of the Middle Triassic Molteno Beds of Basutoland
has been prepared and described. Cynidiognathus longiceps from the lower Molteno
Beds and Scalenodontoides macrodontes are the only cynodonts known from these

beds.

(2) On the basis of the structure of the postcanine teeth of S. macrodontes
it is concluded that this form is closely related to the Brazilian genus from the
Rio do Rasto Series, Traversodon, and the East African genus from the Manda
Beds, Scalenodon. All these beds are considered to be of Middle Triassic age.

(3) The crown patterns of postcanine teeth of cynodonts have been com-
pared. On the basis of the form of the postcanine teeth it is concluded that the
gomphodont cynodonts can be divided into two families, the Diademodontidae
and the Traversodontidae. The Brazilian genus Gomphodontosuchus may perhaps
be included in a separate family, the Gomphodontosuchidae.

(4) It is concluded that the Tritylodontidae could be derived from those
members of the Traversodontidae which have postcanine teeth similar to those
of Scalenodon, Scalenodontoides and Traversodon.

(5) It is concluded that as the Tritylodontidae cannot be classified as
-Ictidosauria, they should be placed.in a new infraorder or considered as
aberrant cynodonts.

REFERENCES

Boonstra, L. D. 1947. Notes on some Stormberg fossils from Basutoland. In G. M. Stockley’s
Report on the geology of Basutoland, 94-5.

Boonstra, L. D. 1953. A report of a collection of fossil bones from Tanganyika Territory.
Ann. S. Afr. Mus., 42, 5-18.

Butler, P. M. 1939. The postcanine teeth of Tritylodon longaevus Owen. Ann. Mag. nat. Hist.,
(II) 4, 514-20.

Crompton, A. W. 1955. On some Triassic cynodonts from Tanganyika. Proc. zool. Soc. Lond.,
125, 617-609.

Crompton, A. W. 1957. The cranial morphology of a new genus and species of Ichdosaurian.
Proc. zool. Soc. Lond. (in press).

Haughton, S. H., and Brink, A. S. 1954. A bibliographical list of Reptilia from the Karroo
Beds of Africa. Palaeont. Afr., 2, 1-187.

I4 ANNALS OF THE SOUTH AFRICAN MUSEUM

Huene, F. von. 1950. Die Theriodontier des ostafrikanishen Ruhuhu-Gebiets in der Tiibinger
Sammlung. Nueus 7b. Min. Geol. Paléont., 92, Heftl. 47.

Huene, F. von. 1956. Paldontologie und Phylogenie der Niederen Tetrapoden. Jena: Gustav Fisher
Verlag.

Kiihne, W. G. 1956. The Liassic therapsid Oligokyphus. London: British Museum (Natural
History).

Parrington, F. R. 1946. On the cranial anatomy of cynodonts. Proc. zool. Soc. Lond., 116, 707.

Seeley, H. G. 1895. Researches on the structure, organization and classification of fossil
Reptilia. Part IX, sec. 3. On the Gomphodontia. Phil. Trans., (B), 186, 1.

Simpson, G. G. 1928. A catalogue of the Mesozoic Mammalia in the Geological Department of the
British Museum. London: British Museum (Natural History).

Watson, D. M. S. 1942. On Permian and Triassic tetrapods. Geol. Mag., 79, 81.

Watson, D. M. S., and Romer, A. S. 1956. A classification of therapsid reptiles. Bull. Mus.
comp. Kool. Harv., 114, 38.

Ann. S. Afr. Mus. Vol. XLIV

Scalenodontoides macrodontes gen. et sp. nov. 1, dorsal view of the lower jaw, x3. 2,
of the 4th to 6th postcanines viewed slightly from above, natural size.

Plate 1

inner view

THE MOSCHOPID SKULLS IN THE SOUTH AFRICAN
MUSEUM

Lieuwe Dirk Boonstra
South African Museum

(With eleven text-figures)

CONTENTS

Page
Introduction : ; : : 15
Historical . : : , 15
General shape and form of skull. au oe 16
Bones of dorsal and lateral surfaces : 16
The occiput , ; ? Di
Ventral surface of skull. : : 24.
Taxonomic ‘ : : : ‘ 28
Discussion : : : : ; : 36
References : : : 37

INTRODUCTION

All the known specimens of the South African Moschopids, with the single
exception of Moschoides—housed in the Walker Museum of the University of
Chicago, have been personally examined. In addition to these there are in the
collection of the South African Museum a number of cranial specimens found
as isolated specimens at various locations in the Koup, and lastly the skulls of
at least twenty individuals found together at Kruisvlei in a thin layer of argil-
laceous sandstone.

HIsTORICAL

Delphinognathus conocephalus, described in 1892 by Seeley, was the first
South African Moschopid to become known. Although some of the structures
were misinterpreted, Seeley’s woodcuts excellently portray the parts of the skull
preserved. Of this skull Broom in 1910 published a restored figure which is
inaccurate in a number of respects.

Nineteen years later the second genus, Moschops, was described by Broom
(1911), the type being an incomplete skull. This Spitskop material, found by
Whaits and sold by Broom to the American Museum, includes a number of
topotype skulls. These skulls were subsequently studied by Gregory in 1926
and by the author in 1936.

The third genus, Moschognathus, was founded by Broom in 1914 on a
lower jaw found by Whaits near Beaufort West and this jaw, together with the
postcranial skeleton, was restudied by Gregory in 1926.

In 1914 Watson founded two additional genera, Pnigalion and Lamiasaurus,
on rather poor material. As Lamiasaurus has as type specimen a snout which

16 ANNALS OF THE SOUTH AFRICAN MUSEUM

is Titanosuchian and a postorbital skull piece which is Moschopid, it has been
proposed that only the snout be considered as the type and this has been found
to be at most a doubtful species of the genus Tztanosuchus. ‘The postorbital
skull piece is best left unnamed.

In 1937 Byrne founded the genus Moschoides on a fair skeleton with which
is associated four fragments with teeth. Finally in 1952 I described a fairly good
skull from Kruisvlei under the name Avenantia kruisvleiensis.

GENERAL SHAPE AND FORM OF THE SKULL

The Moschopid skull is fairly light to massive and of moderate size (length
308-423 mm., width 200-395 mm., height 216-309 mm.). The snout is
relatively weak, very short and fairly high (preorbital length 160-195 mm.,
height 108-162 mm., width 114-222 mm.).

The orbit, alee is fairly small to fairly large, does not lie wholly i in the
posterior half of the skull (with the possible exception of Delphinognathus). The
bones of the preorbital part of the skull are only moderately thick, whereas those
of the posterior half, particularly of the cranial roof, are fairly to greatly
pachyostosed: in the latter the surface is fairly smooth to rough, whereas in the
former it is smooth to fairly rough. In most, except Delphinognathus, the ‘cheek’
is also thickened and fairly rough. The pachyostosis of the cranial roof is in
Delphinognathus greatest round the pineal foramen, but in the other genera it is
general and no distinct naso-frontal boss is developed. There is also no step
between the facial and cranial surfaces, as these surfaces run into each other in
an even convexity. The postfrontal bone is fairly light to massive and never has
a boss. The orbits, except in Delphinognathus, are fairly small and look forwards
and outwards, dorsally overhung and posteriorly bounded by the greatly
thickened orbital rim.

The temporal fossa is wide and roomy in Avenantia, with a moderate antero-
posterior diameter in Delphinognathus and narrow and slit-like in Moschops.

The interparietal width is small to large (54-198 mm.), with a parietal
crista present in Avenaniia.

Due to the forward position of the quadrate and the short snout, the lower
jaw is short and the gape small.

BONES OF THE DorSAL AND LATERAL SURFACES

The premaxillaries (P.M.) form the anterior part of the upper median
surface; they stretch backwards from the anterior alveolar border to end well
anterior to the plane in which the anterior orbital border lies; posteriorly they
lie in a groove in the nasals which bones further posteriorly, meeting each other
in the median line, prevent the premaxillaries from meeting the frontals
(my 1936 account on this feature in A.M.N.H. 5553 is thus incorrect and
Efremov’s criticism justified). Broad anteriorly up to the anterior edge of the
naris, the premaxillaries narrow abruptly and then taper in posterior direction,

a he ee Se

THE MOSCHOPID SKULLS IN THE SOUTH AFRICAN MUSEUM EZ

but the internarial bar is strong. The surface of the premaxillaries in the median
line curves evenly dorso-posteriorly and this slope is continued, without any
step, on to the nasals and frontals.

The nasals (N.) together present a U-shaped surface with the limbs of the
U directed anteriorly and the premaxillaries wedged in between the limbs and
posteriorly lying in a groove in the nasals, anteriorly each nasal forms the
posterior border of the naris and meets the posterior edge of the septomaxilla;
posterior to the limits of the premaxillaries the two nasals meet in the median
_ line and meet the anterior ends of the frontals, but although the bones are here
greatly thickened there is no indication of any distinct naso-frontal boss, and
the sweep of the upper surface is a gentle convexity with no indication of a step
between the facial and cranial surfaces; there is no depression in the posterior
nasal surface for a facial gland as there is in most of the Tapinocephalidae.
Laterally the swollen surface of the nasal curves evenly into that of the strongly
swollen prefrontal. A tongue of the frontal sometimes separates the posterior
part of the prefrontal from the nasal.

The septomaxilla (S.M.) has a small facial exposure; it forms the lateral
and part of the anterior and posterior borders of the naris; its surface within the
naris, of which it forms the floor, is extensive. I have not located a septomaxillary
foramen.

The maxilla (M.) is the largest bone of the face; although fairly short it
is quite high; its posterior edge is weakly indented to receive the very short
lacrimal, so that both the dorsal and the ventral processes appear to be short;
the dorsal process is short and wide and posteriorly truncated to meet the
prefrontal in a long suture; the ventral process is longer and tapers to a point
along its short suture with the jugal; the anterior half of the alveolar edge
curves gently upwards to where it meets the premaxilla.

The lacrimal (L.) has a small facial exposure; it is very short, but high and
in outline presents a face which is a crescent with the tips cut off; it thus forms
most of the anterior orbital rim, which though rounded is not much thickened.
_ It is only slightly overhung by the prefrontals and has no contact with the nasal
because of the intercalation of the strong dorsal process of the maxilla. No
foramen is visible on the facial surface as this probably lies within the orbit.

The jugal (J.) is quite a strong bone but like the bones of the snout is but
little pachystosed and its surface is fairly smooth; it forms the ventral, relatively
unthickened, border of the orbit. From its contact with the maxilla, its ventral
edge sweeps down sharply postero-ventrally to where it makes a small contact
with the upper edge of the quadratojugal (but in Delphinognathus the posterior
part of its ventral edge is deeply excavated to form a deep and roomy indenta-
tion breaking the even sweep of the antero-ventral border of the jugal on to
the anterior face of the quadratojugal, whereas in Moschops and Avenantia this
sweep is only interrupted by a slight nick at the junction of these two bones
(this is also evident in some Struthiocephalids). The posterior edge of the jugal
is slightly (Avenantia) or deeply indented by the anterior process of the squamosal

18 ANNALS OF THE SOUTH AFRICAN MUSEUM

so that the face of the postero-ventral process of the jugal is long and narrow,
except in Avenantia when it is short and broad. Its contact with the postorbital
is along a short curved suture.

The prefrontal (Pr.F.) is only moderately to strongly thickened; it forms
the moderate to thick antero-dorsal part of the orbital border which overhangs
the orbit moderately to strongly. In its anterior part, where it meets the lacrimal
and maxilla, it is less swollen and its surface is smoother than in its posterior
parts, and it does not strongly overhang the lacrimal. In all the Moschopid
genera the prefrontal meets the maxilla and the lacrimal is thus excluded from
contact with the nasal. In dorsal view its width is seen to vary, this also in
specimens of the same genus. Posteriorly it usually does not meet the post-
frontal, but in Avenantia it just meets the postfrontal, so the frontal just enters
the orbital border or is just excluded from it. Medially a varying tongue of the
frontal is sometimes wedged in between the nasal and prefrontal, but this tongue
is sometimes absent. The swollen surface of the prefrontal passes evenly on to
the swollen surfaces of the nasal and frontal and no groove is ever developed to
prevent this confluence as in some Tapinocephalids and Struthiocephalids.

The frontal (F.) is the largest bone of the dorsal skull roof, being particularly
large in Moschops koupensis. The pair together forms a roughly rectangular
surface as the main body of the two bones, with, laterally, a tongue which just
enters or is just excluded from the orbital border and anteriorly, there is in some
specimens a varying tongue wedged in between the posterior end of the nasal
and the prefrontal. The frontals meet the parietals in a nearly straight trans-
verse suture, but in Moschops koupensis the parietals form a strong anteriorly
directed wedge into the frontals. Sometimes a postero-laterally directed tongue
separates the postfrontal from the parietal, but usually there is a fair to small
contact between parietal and postfrontal. Anteriorly the thickened frontal
surface flows with a gentle curve on to that of the nasal so that there is no
indication whatever of a distinct naso-frontal boss. The frontal surface is
similarly confluent with that of the parietals and postfrontals. Although in
general the frontal face is domed there is in some forms a depression along the
middle line, slight in some specimens of M. capensis, but in Avenantia it forms a
decided wide but fairly shallow longitudinal groove. The pachyostosis along
the line of junction of the two frontals is thus, in some specimens, less developed
than more laterally.

The postfrontal (Po.F.) is a fair-sized bone moderately to very strongly
thickened; it forms a small to fairly large part of the dorso-posterior section of
the orbital border; it sometimes meets the prefrontal but more often the lateral
tongue of the frontal is intercalated between these two bones; posteriorly it
meets the parietal in most specimens, but sometimes the postero-lateral tongue
of the frontal intervenes; stretching from the orbit to the temporal fossa it forms
the upper part of the postorbital bar lying above the postorbital and latero-
ventral to the frontal. The postfrontal, though moderately to very strongly
pachyostosed, never forms a distinct boss, but has its rough outer surface

THE MOSCHOPID SKULLS IN THE SOUTH AFRICAN MUSEUM 19

confluent with the swelling of the frontal, parietal and postorbital in an even
gentle convexity.

The parietal (P.) of the Moschopids is a remarkably variable element, not
- so much in regard to its connections with and relationships to the contiguous
bones as in the shape and topography of its upper face.

In Moschops koupensis and in some specimens of M. capensis it forms and
completes the general dome-shaped convexity of the upper cranial roof
coalescing in gentle curves with the greatly pachyostosed surrounding bones.
Whereas in other specimens of Moschops capensis there is a tendency, more or
less pronounced, for it to form a mound around the pineal foramen, and, in one
greatly distorted and probably young skull, this mound approaches the conical
shape seen in Delphinognathus.

In Avenantia the parietal is but little pachyostosed; there is a circular and
prominent wall around the pineal foramen, with a lateral excavation bounded
by a ridge lying still further laterally; posteriorly the bone forms a low, fairly
narrow crista which curves down towards the occiput, whereas in Moschops the
parietal posterior to the foramen tends to form a groove along the median line
more or less strongly developed.

In Moschops koupensis the parietal forms the posterior edge of the skull
between the interparietal and the tabular, whereas in all the other forms the
posterior edge is wholly formed by the interparietal and the two tabulars; this
is due to the parietal curving down postero-ventrally much more strongly.

As a pair the parietals form a much smaller part of the cranial roof than
do the two frontals. A moderate to large pineal foramen lies very near to fairly
far away from the frontal suture but it is always well anterior of the occipital
edge. The size of the pineal foramen varies within the same species and a
greater size does not appear to be correlated with juvenility.

There is considerable variation in antero-posterior length of the parietal;
the fronto-parietal suture is either a fairly straight transverse line or forms an
anteriorly directed obtuse or sharp V; the parietal meets or does not meet the
- postfrontal. Posteriorly the parietals meet the interparietal and tabulars in a
fairly straight or slightly curved suture, but in M. koupensis this general curve is
indented where it makes contact with the interparietal.

The parietal width in the Moschopids shows considerable variation, which,
within the species Moschops capensis, appears to be due to differences of age in
some cases, but where other features exclude this explanation it can only be due
to differences in sex. In M. capensis the parietal width varies from 108 to 172 mm.;
in M. koupensis it is 198 mm.; in Delphinognathus 120 mm., and in Avenantia
84 mm. (with the crista 48 mm. at its narrowest part). In some skulls of M.
capensis there is some indication of a pinching-in across the parietals to form a
dorsal bay to the temporal fossa as in some species of Struthiocephalus and Kerato-
cephalus.

The parietal forms a rounded ill-defined edge to the upper border of the
temporal fossa in all the forms, except in Avenantia where the edge of the crista

20 ANNALS OF THE SOUTH AFRICAN MUSEUM

forms this edge; lateral to this the parietal forms a more or less vertical face
within the temporal fossa with its ventral edge meeting the posterior end of the
postorbital and the dorsal edge of the upsweeping squamosal.

Except in Avenantia the parietal of the Moschopids does not send a tapering
process on to the posttemporal arch intercalated between the tabular and the
squamosal, as is the case in Tapinocephalids and Struthiocephalids. This
confirms my observation (1936) on the Spitskop material in the American
Museum and Efremov’s (1940) stricture on this observation is thus groundless.
The mistake made by Efremov is that he made no allowance for the possibility
that the postero-lateral tongue of the parietal may, by increased pachyostosis of
the contiguous bones, be overgrown by them. This in fact is just what has taken
place in Moschops.

The postorbital (P.O.) is a moderate to very massive bone forming the
lower part of the postorbital bar and with a much more lightly built posterior
flange lining part of the upper wall of the temporal fossa; anteriorly it forms
a small to great part of the thickened posterior orbital border, and posteriorly
most of the thick to very thick anterior border of the temporal fossa. Ventrally
it abuts against the squamosal in a long curved suture and, except in Avenantia,
meets the jugal along a short curved suture. Dorsally it meets the lower edge
of the postfrontal along a curved suture running across the postorbital bar.
Within the temporal fossa the postorbital sends a short to moderately long
lightly built flange upwards and backwards to flank the lateral face of the
parietal and to meet the dorsal edge of the upsweeping squamosal within the
temporal fossa. These two bones together with the lateral face of the parietal
thus form the median lining of the temporal fossa.

The squamosal (Sq.) is large and is the main constituent bone of the
‘cheek’; it is little to greatly thickened with its outer face fairly flat and smooth
in Delphinognathus, but more or less swollen and rough in the other genera.
Anteriorly it meets the jugal in a squamous suture, fairly straight in Avenantia,
moderately indented in Pnigalion and deeply indented in Moschops and Delphino-
gnathus so that it lies over the upper edge of the postero-ventrally directed
process of the jugal. In its antero-ventral corner it receives the quadratojugal
overlapping the posterior part of this bone’s outer face. Dorsally its anterior
part forms the base of the postorbital bar and here it is overlapped by the
postorbital and further posteriorly the squamosal forms the fairly thin to thick
lower border of the temporal fossa and then, sweeping upwards, forms most of
the posttemporal arch and the posterior border of the temporal fossa. From the
posterior edge of the temporal fossa the squamosal curves inwards and forwards
to meet the parietal and postorbital inside the upper part of the temporal
fossa.

On the outer face of the posttemporal arch the squamosal meets the tabular
and in the Moschopids there is no long tongue-like process of the parietal inter-
calated between the squamosal and the tabular, but in Avenantia there is a short
wedge of the parietal between these two bones.

eS ee ee ee ee ee

THE MOSCHOPID SKULLS IN THE SOUTH AFRICAN MUSEUM 21

Postero-ventrally the thick rounded edge of the squamosal forms the edge
of the skull and then on the posterior face carries the ‘auditory groove’, which
is bounded medially by a scroll-like ridge.

The tabular (T.) in dorsal view presents only its upper edge, which is thin
in Avenantia but thick in Moschops where it has overgrown the postero-lateral
tongue of the parietal. In lateral view the tabular is seen to form most of the
posterior edge of the skull above the squamosal.

The interparietal (I.P.) in dorsal view presents only its dorsal edge which
_ forms the median part of the occipital edge. The dorsal interparietal edge is
thin in Avenaniia, fairly thick in Moschops capensis and in M. koupensis forms only
a small part of the dorsal edge in the median line where the dorsal surface of
the skull is grooved; laterally to it the parietal forms the posterior edge until
it meets the tabular.

The quadratojugal (Q.J.) in lateral view, presents a narrow dorso-ventrally
elongated face in Moschops where it is directed antero-ventrally. Dorsally it
forms a weak contact with the jugal, with a slight narrow notch between these
two bones anteriorly. In most specimens of Moschops the squamosal overlaps
the posterior border of the quadratojugal, but in one specimen the squamosal
also clasps the quadratojugal dorsally so that it is partly wedged in to the
_squamosal, but the quadratojugal never forms a narrow deep wedge into the
squamosal as in ‘T'apinocephalids.

In Delphinognathus the anterior half of the upper end of the quadratojugal
is deeply notched and concomitantly the anterior half of the lower end of the
jugal is also notched so that a deep broad notch is formed between these two
bones. The posterior half of the upper end of the quadratojugal meets the
posterior half of the lower end of the jugal as in Moschops. The lateral surface
of the quadratojugal is thus not narrow and elongated as in Moschops but can
be described as an elongated dorso-ventrally directed surface with wedge-
shaped anterior process. According to Watson (1914) the quadratojugal in
Pnigalion presents a large, roughly quadrangular outer surface meeting the
_ jugal along a fairly long suture and not wedged into the squamosal.

The quadrate (Q.) can in lateral view be seen lying medially to the quad-
ratojugal with the outer cotylus showing below the ventral limit of the quadrato-
jugal, but in M. koupensis much of the antero-lateral surface of the quadrate
above the cotylus is also visible in lateral view.

THE OccIPuT

The occiput in the Moschopidae presents a fairly large surface, which is
at least twice as broad as high and is shallowly concave from side to side. The
dorsal edge of the occipital plate lies further posteriorly than the ventral edge.
Along the median line there is a weak to fairly weak ridge on the interparietal.
On either side of this ridge lies a shallow to moderately deep depression. The
upper edge of the occipital plate proper lies high in most forms with a rounded
edge, but in Moschops koupensis both the upper and lateral edges form a rampart-

22 ANNALS OF THE SOUTH AFRICAN MUSEUM

like wall enclosing the inner part of the plate and encroaching on it so that the
occipital plate proper becomes very small. The condyle is directed postero-
ventrally so that the skull forms a sharp angle with the neck; it is a stout
rounded knob dorsally excavated by a groove, sometimes quite deep, for the
medulla; the lateral parts of the knob are formed by the exoccipitals.

The foramen magnum is large and oval or is fairly small with its sides pinched
in. The posttemporal fossae are small with, in one case, a median groove
leading into them; they are bounded by the supraoccipital and paroccipital.
The lateral border of the skull is formed by the swollen squamosal which more
medially is excavated to form the ‘auditory groove’ and this is bounded medially
by a prominent scroll-like ridge wholly formed by the squamosal. Ventrally,
the condyles of the quadrates lie in a plane far anterior of the plane of the
occiput.

Viewing the occiput at right angles to the plane in which the alveolar
borders lie, much of the surface of the parietal, postfrontals, postorbitals and
frontal is seen and the ventral edge is formed by the squamosal, quadratojugal,
quadrate, quadrate ramus of the pterygoid, exoccipital and basioccipital.

The basioccipital (B.O.) in occipital view shows only a small surface, viz.
an elongated strip along the middle line of the condyle; its upper part is
excavated and this groove leads into the foramen magnum; near its lower border
there lies a notochordal pit.

The exoccipitals (E.O.) form the lateral surfaces of the condyle separated
by the above-mentioned strip of the basioccipital (but in Avenantia the two
exoccipitals meet to form the lower border of the foramen magnum); from the
condylar surface the exoccipital extends forwards and upwards to enter the
occipital plate forming an irregular triangular surface lateral to the foramen
magnum overlapping the supraoccipital and paroccipital; the ventral border of
this sheet of the exoccipital is notched to allow for the exit of the tenth nerve;
the exoccipital forms most of the lateral rim of the foramen magnum.

The supraoccipital (S.O.) is a low but broad bone; in its lower median part
it forms the upper and sometimes also much of the lateral rim of the foramen
magnum. Laterally it extends to the outer corner of the posttemporal fossa, or
well lateral of this fossa, and forms the dorsal border of this fossa and above this
it meets the tabular in an undulating suture; dorsally it meets the interparietal
in a fairly straight to curved suture. Along its ventral edge it makes contact
with the paroccipital on both sides of the posttemporal fossa (but in Moschops
koupensis only medially of this fossa). The median ridge is low or absent on the
supraoccipital, but is fairly prominent on the interparietal.

The interparietal or dermosupraoccipital (I.P.) is a moderately to large-
sized element forming the upper part of the middle of the occipital plate. It is
a rectangular element with the corners developed as wedge-like processes ;
along the median line it carries a fairly strong ridge and lateral to this the bone
is slightly or quite deeply excavated. Dorsally the bone forms a strong rounded
edge curving over to meet the parietal, but in Moschops koupensis the upper part

a a ee es a ee

THE MOSCHOPID SKULLS IN THE SOUTH AFRICAN MUSEUM 23

of the bone forms a very strong overhanging rampart, which is laterally con-
tinued by a similar overhanging of the tabular.
The tabular (T.) has the largest surface of all the occipital bones. Its

- outer edge, which forms the lateral and part of the dorsal limits of the occipital

plate proper, is weakly rounded or is developed into a very massive overhanging
rampart, which greatly encroaches on to the more inner part of the occipital
plate greatly reducing this area. (This condition in Moschops koupensis is a much
more advanced development than that shown in Yapinocephalus, where it is

_ only incipient.)

Dorsally the tabular meets the parietal on the dorsal surface of the skull
and is just excluded from or just enters the edge of the posttemporal fossa. Its
outer edge meets the upsweeping dorsal process of the squamosal but is excluded
from the ‘auditory ridge’ which lies wholly on the squamosal. The latero-
ventral corner of the tabular overlaps the paroccipital lateral to the posttemporal
fossa. Internally it meets the interparietal and supraoccipital along a long
undulating suture and is just or well excluded from the edge of the posttemporal
fossa.

The paroccipital (P.O.) is, in occipital view, seen to be a strong bar medially

_ abutting against the basioccipital and overlapped by the flange of the exoccipital

and laterally extends to meet the squamosal and here it is overlapped by the
downsweeping corner of the large tabular and itself overlaps on to the posterior
face of the quadrate, which it supports very firmly. The medio-veniral corner
of the paroccipital is seen to descend and form the posterior part of the rim of the
Jenestra ovalis. 1ts dorsal edge meets the supraoccipital except at the posttemporal
fossa. In Moschops koupensis there is a groove medial to and leading into the
fossa. In occipital view the paroccipital is seen to obscure the stapes to a smaller
or greater extent.

The quadrate (Q.) shows a small to fair-sized surface when viewed at right
angles to the plane in which the alveolar borders of the maxillaries lie. Its
ventral edge has two fairly strong oval knobs, separated by a notch, which form

_ the articulatory condyles for the reception of the articular. In its dorsal part

the posterior face of the quadrate is overlapped by the distal end of the paroc-
cipital firmly applied to it. Medially the short anterior process of the quadrate
is overlapped by the distal end of the quadrate ramus of the pterygoid, which is
firmly applied to it and which passes further backwards on to the posterior face
of the quadrate above the internal cotylus. Above the pterygoid the postero-
distal process of the stapes abuts securely against the posterior face of the
quadrate. Just below or just distal of the postero-distal corner of the stapes
there is a distinct low mound on the quadrate.

Dorso-laterally the quadrate meets the quadratojugal and along the suture
these two bones are notched to form a fair-sized rounded quadratic foramen.

The quadratojugal (Q.J.) in this view presents a low but broad face
between the suture with the quadrate and the edge of the downsweeping over-
lapping squamosal.

24 ANNALS OF THE SOUTH AFRICAN MUSEUM

The stapes (St.) is, in occipital view, obscured by the paroccipital, slightly
in Moschops capensis but greatly in. Avenantia and Moschops koupensis. It is a
moderately strong rod-like element with a slight waist and an expanded distal
end, and is in its middle portion pierced by a large oval stapedial foramen.
Its postero-distal process, directed towards the tubercle on the quadrate, is
applied to the inner part of the posterior face of the quadrate, and this is firmly
wedged in between the paroccipital above and the quadrate ramus of the
pterygoid below. Proximally the footplate of the stapes passes anteriorly to the
downwardly directed medio-ventral corner of the paroccipital to fit into the
ventrally situated fenestra ovalts.

The stapes lies diagonally with its proximal end appreciably higher than
its distal end.

THE VENTRAL SURFACE OF THE SKULL

The occiput being sloping, with its dorsal edge lying well posterior to its
ventral border, is visible in ventral view. Post-mortem dorso-ventral pressure in
the rock tends to accentuate this slope and in figures 9 and 11 correction to
neutralize this should be allowed for. |

Except for a slight upward tilt anterior to the transverse pterygoidal rami,
the palate and basis cranii lie in practically the same plane, with the weak to
fairly weak lateral pterygoidal rami lying moderately far below this plane and
the suspensorium lying still further ventrally.

The articulating condyles of the quadrates lie far to very far anteriorly in
a plane far anterior to the basioccipital condyle, nearly half-way up the ventral
surface of the skull. The subtemporal fossae are small, short, but fairly wide;
the choanae are fairly short but wide; the interpterygoid vacuity is a
fairly narrow and short slit anteriorly bounded by the prevomer and
there is a very small suborbital foramen on the suture between the palatine
and transversum.

The basioccipital (B.O.) is, in ventral view, seen to form the greater part
of the occipital condyle. The whole condyle is usually a more or less rounded
strong knob, but in some cases it is more pear-shaped and even triangular in
outline.

The basioccipital forms the anterior and medial portion of the condyle but
in Avenantia the exoccipitals meet each other posteriorly. The notochordal pit
faces postero-ventrally and the skull would in life hang downwards from its
articulation with the atlas.

Anterior to its condylar part the basioccipital has a short but fairly wide
to wide face which lies on a higher level than the condyle. This face is only
slightly tilted downwards anteriorly and thus meets the basisphenoid in a very
large angle. In its median part the basioccipital carries a pair of anteriorly
converging tuberous ridges in Moschops koupensis, which tend to coalesce in
M. capensis to form a low keel whereas in Avenantia two weak ridges are developed
anteriorly with the posterior part of the surface flat.

THE MOSCHOPID SKULLS IN THE SOUTH AFRICAN MUSEUM 25

Anteriorly the basisphenoidal tubera underlie the basioccipital which is
met in a slightly concave transverse suture.

Laterally the basioccipital forms the thickened ventral rim of the fenestra
- ovalis, which thus lies far ventrally. Further backwards it abuts against the
paroccipital and is underlain by the antero-medial process of the paroccipital,
which forms the posterior part of the rim of the fenestra ovalis, and still farther
backwards it is underlain by the thin sheet of the exoccipital.

The basisphenoid (B.S.) carries a low but fairly sharp median keel, lateral
_to which there are two posteriorly diverging fairly strongly swollen tubera

(flattish in Avenantia); postero-lateral to the tubera the basisphenoid forms the
anterior border of the fenestra ovalts.

Anteriorly the basisphenoid tapers and is wedged in between the quadrate
rami of the pterygoids and its median keel is continued anteriorly by those of
the two pterygoids. Immediately anterior to the tubera and lateral of the
median keel lie the internal carotid foramina and lateral to the tubera lie the
external carotid foramina.

In general the ventral faces of the basisphenoid and basioccipital lie in
nearly the same plane, as these faces subtend only a very obtuse angle because
the anterior part of the basioccipital is only very weakly inclined downwards

in anterior direction.

The paroccipital (P.Oc.) is a large, strong and massive bone which forms
a firm and strong connecting link between the cranial base and the laterally
situated quadrate, squamosal and tabular. Medially the paroccipital abuts
firmly against the basioccipital and is underlain by the thin flange of the
exoccipital; its antero-median corner usually forms a strong process, strongly
thickened where it forms the ventral and posterior parts of the rim of the
fenestra ovalis. Its antero-lateral corner is very firmly applied to the postero-
median face of the quadrate, and here the postero-distal process of the stapes is
firmly wedged in between it and the quadrate. Its postero-lateral corner over-
laps and is firmly applied to the medio-posterior face of the squamosal just

- medial of the ‘auditory ridge’, and also meets the ventral process of the tabular.
(But in Avenaniza the lateral end of the supraoccipital is intercalated between the
paroccipital and the tabular.) Posteriorly it meets the supraoccipital on both
sides of the small slit-like posttemporal fenestra in Avenantia and Moschops
capensis, but only medially of this fenestra in Moschops koupensis.

Concomitant with the forward shift of the quadrate condyles the parocci-
pital has rotated on its long axis with the result that it presents a much greater
face in ventral view than it does in occipital view.

The antero-ventral edge of the paroccipital is free and rounded and forms
the posterior border of the fairly large fenestra lying between the basisphenoid,
quadrate ramus of the pterygoid, quadrate and paroccipital.

On the suture between the paroccipital and the exoccipital lie two small
foramina, the anterior being the jugular foramen and the posterior one for the
exit of the tenth nerve.

26 ANNALS OF THE SOUTH AFRICAN MUSEUM

The pterygoid (Pt.) is a bone of moderate size and with its fellow forms the
middle portion of the ventral surface of the skull. Its most prominent com-
ponents are a strong, deep but short quadrate ramus and a short and fairly weak
transverse ramus. In the median line its antero-posterior length is short mainly
because there is little left of the anterior process, which is so well developed in
the ancestral Pelycosaurs and contemporary Therapsids.

Anteriorly the pterygoids are separated in the median line by a long and
fairly open interpterygoid slit; posterior to this the pair of pterygoids meet to
form a sharp keel which is posteriorly continued on the basisphenoid; the
pterygoid keel is wedged into the basisphenoid.

Lateral to this median keel the pterygoid is deeply excavated and becomes
very thin to form a wide and deep groove lying diagonally in the skull; in this
groove the pterygoid meets the basisphenoid along a diagonally directed, fairly
straight suture. This deep groove is laterally bounded by the prominent
quadrate ramus, which is a deep sheet of bone lying nearly at right angles to
the plane of the palate; above, where it meets the basisphenoid, it is thin, but
its ventral edge is thickened and rounded and thus forms a strong girder, which
posteriorly supports the quadrate.

The quadrate rami are short and diverging in posterior direction; each
is firmly applied to the mesial face of the short anterior process of the quadrate
and its extremity extends well posterior to the quadratic condyles. The upper
edge of the end of the ramus lies below and overlaps the anterior part of the
distal end of the stapes, which passing above it is firmly abutted against the
mesial face of the quadrate. The quadrate ramus is connected with the lateral
ramus by a strong, thickly rounded web of bone and it is here that part of the
pterygoid muscle was attached. The edge of the quadrate ramus and of this
web of bone form the mesial and anterior border of the subtemporal fenestra.

The lateral ramus of the pterygoid is weak, does not descend much ventrally
nor extend much laterally. In Avenantia its ventral edge is narrow and sharp,
in Moschops capensis it is a little thicker and in Moschops koupensis laterally swollen;
postero-medially its edge curves to join the median keel. Laterally the ramus
is met by the transversum and anteriorly by the palatine, and it has a small
contact with the prevomer in the median line.

The transversum or ectopterygoid (Tr.) is a fairly small but quite strong
bone linking the transverse ramus of the pterygoid with the bones of the side-
wall of the skull; it is firmly joined to the inner face of the jugal and flanked
by the maxilla; anteriorly its edge meets the palatine and here there is a small
suborbital foramen. Due to the forward position of the transverse ramus the
transversum lies lateral and even somewhat posterior to the extremity of the
ramus, whereas primitively it lies anterior to the ramus.

The palatine (Pal.) is short and extends antero-laterally from its contact
with the pterygoid as a sheet of fairly thick bone to form much of the lateral
border of the choana and a varying part of its posterior border; it is laterally
applied to the inner maxillary face, where it flanks the alveolar border. In the

es a Oe ne migie conte Bibi Bei eee ae ee — noe

THE MOSCHOPID SKULLS IN THE SOUTH AFRICAN MUSEUM 27

median line a tongue of the prevomer prevents it from meeting its fellow; here
the palatine carries a low mound of varying shape which is not dentigerous.
In Avenantia this mound is continued on to the short anterior ramus of the
- pterygoid. 7

The prevomers (vomers P.V.) are strong but short bones which together
form a massive interchoanal bar. Their anterior bevelled edges are applied to
the inner face of the premaxillaries. Posterior to the choanae they widen to meet
the palatines and in the median line they lie between the mesial edges of the
palatines and meet the pterygoids. Anteriorly along the median line there is
a groove and along the lateral edge there is an undulating ridge. Dorsally the
prevomers form a narrow median ridge which is posteriorly continued by the
pterygoids as part of a median septum.

The premaxillary (P.M.) has a massive alveolar face and although poorly
exposed in most specimens, appears always to have housed three fairly large
teeth, with indications, in some specimens, of replacing teeth lying lingually.

The maxilla (M.) has the anterior part of its alveolar border massive and
wide but in posterior direction it tapers fairly rapidly and, posterior to the
teeth, the maxillary edge is continuous with the sharp ventral edge of the jugal.
Little is known of the teeth, but the anterior ones are large and they then rapidly
decrease in size backwards. There would appear to be from ten to possibly
thirteen maxillary teeth.

The jugal (J.) shows, besides its internal face which is anteriorly overlain
by the transversum, a sharp to somewhat rounded ventral edge, which curves
downwards to have its end abutted against the quadratojugal in Moschops with
a nick at the junction in some skulls; but in Delphinognathus there is a wide and
deep notch preventing the ventral edges of the jugal and quadratojugal from
meeting, but these bones do meet at a higher level. ‘The squamosal is never
wholly intercalated between jugal and quadratojugal as in some of the advanced
Pelycosaurs and most other Therapsids.

The quadrate (Q.) in ventral view presents its articulatory surface as a
_ very prominent feature. The quadratic condyle is bipartite, with two fairly
robust ovoid cotyli separated medially by a shallow open groove. The edges
of both cotyli in ventral view overhang both the anterior and posterior faces
of the upper part of the quadrate. The short anterior quadratic process lies
obliquely in the skull, being directed antero-internally; it is against the mesial
face of this process that the quadratic process of the pterygoid is very firmly
applied. Lateral to the outer cotylus there is a step up before the quadrate
meets the quadratojugal in a fairly long and nearly straight suture. A third of
the way up along this suture lies a fairly small foramen quadrati. Due to the
forward inclination of the quadrate much of its posterior face can be seen in
ventral view. On this otherwise featureless face there is a small low tubercle
lying some distance above the inner cotylus. More medially the quadrate
receives the distal surface of the stapes firmly applied to it. In ventral view it is
evident how firmly the more dorsal part of the posterior face of the quadrate is

28 ANNALS OF THE SOUTH AFRICAN MUSEUM

overlapped by the downsweeping process of the squamosal and more internally
by the robust quadratic process of the paroccipital, which is here also wedged
in between the two faces of the postero-lateral stapedial process

The quadratojugal (Q.J.) is seen, in ventral view, to be a fairly small bone
wedged in between the outer face of the quadrate and the lower overlapping
edge of the downsweeping process of the squamosal A small foramen quadrati
notches the mesial edge of the quadratojugal a third of the way up along the
quadrate-quadratojugal suture.

The squamosal (Sq.) is seen, in ventral view, to form the postero-lateral
corner of the skull, and sweeping downwards and forwards overlaps on to the
posterior faces of the quadrate and quadratojugal. Internally the squamosal
meets the paroccipital ventrally and the tabular dorsally. Near its junction
with the paroccipital and the tabular the squamosal carries a fairly to very
prominent ‘auditory ridge’ with an ‘auditory groove’ lying lateral to this ridge.

The stapes (St.) lies more or less obliquely in the skull. From its proximal
end, which fits into the ventrally situated fenestra ovalis, it is inclined both
forwardly and downwardly to abut against the quadrate. In ventral view the
stapes shows two faces, viz. a ventral and a posterior. The ventral surface is
elongate with a central waist and expanded ends. The proximal end is knob-like
and is firmly fixed into the fenestra ovalis which is situated low down in the skull;
the distal expansion has a long tapering postero-lateral process, the extremity
of which is applied to the low tubercle on the posterior face of the quadrate;
more anteriorly the main surface of the distal end abuts firmly against the
quadrate and is here wedged in between the quadrate and the quadratic process
of the pterygoid and the quadratic process of the paroccipital. The posterior
face of the stapes, which is pierced by a fairly large oval stapedial foramen, is
triangular in outline with its base applied to the quadrate after passing above
the quadratic process of the paroccipital, which wedges the stapes firmly against
the quadrate. The distal end of the stapes is thus seen to be firmly wedged
against the quadrate with but little possibility of movement.

The tabulars (T.), interparietal (I.P.) and the supraoccipital (S.O.) are
to a greater or less extent visible in ventral view because of the variable forward
inclination of the occipital surface from above downwards.

These bones lie more or less posterior of the basioccipital condyle with the
interparietal and the two tabulars forming the posterior edge of the skull. In
ventral view the posterior edge is slightly concave in Avenantia but gently
concave in Moschops.

'TAXONOMIC

MoscHOPIDAE
Skull characters of the family
Skull of medium size (length 308-423 mm.; breadth 200-395 mm.)
relatively short and broad to long and narrow (length varies from 105 to
182 per cent of the breadth); relatively low to fairly high (height varies from

THE MOSCHOPID SKULLS IN THE SOUTH AFRICAN MUSEUM 20

55 to 80 per cent of the width); snout very short to fairly short (snout length
varies from 45 to 61 per cent of the total skull length); the snout is fairly high
to high and fairly narrow to broad (height of snout from 81 to 156 per cent of
the width of the snout).

The orbit does not lie wholly in the posterior half of the skull except in
Delphinognathus.

The transition from the facial to the cranial surface is by a gentle, even
curve and never by an abrupt step. There is no depression in the surface of
_ the nasal as in the Tapinocephalidae. The dorsal cranial bones are moderately
to strongly pachyostosed (except in Delphinognathus) with the centres of thickening
coalesced and running into the thickening of the dorsal facial bones.

The postorbital bar is fairly slender and light to very wide and massive;
the postfrontal does not form a prominent boss but its surface curves evenly on
to the dorsal surface. The temporal fossa is fairly small to small with its antero-
posterior diameter fairly small to very small. The inter-temporal region is
narrow to wide (54-172 mm.) with the parietals entering the upper border of
the temporal fossa. The frontal is not excluded from the orbital border; the
lacrimal has no contact with the nasal. The quadrate ramus of the pterygoid
is very short. The dentition is well developed but undifferentiated.

MoscHops

Skull characters of the genus

The preorbital length is 45-49 per cent of the total median length and the
orbit does not, thus, lie wholly in the posterior half of the skull; the snout is
thus short (preorbital length 160-190 mm.) and fairly broad to broad (156—
227 mm.) and fairly low to fairly high (108-162 mm.) with the height 74-95?
per cent of the width.

The dorsal cranial bones are strongly pachyostosed.

The prefrontal is greatly thickened and this is confluent with the pachyosto-
sis of the nasal and frontal, but this does not greatly overhang the orbit although
the orbital border is thick.

The area around the pineal foramen is greatly thickened but does not stand
out above the general skull surface.

The postorbital bar is fairly to very wide and moderately to very
massive.

The posttemporal arch is moderately to very thick and massive and the
temporal fossa is moderate to small with its antero-posterior diameter fairly to
very greatly reduced.

The dorsal parietal surface is broad to very broad (142-172 mm.) and the
interorbital width is 70-100 per cent of the intertemporal width.

The intersquamosal width is moderate to great (210-395 mm.) and the
median length is 103-159 per cent of the width. (The low figure for the width
in one specimen is undoubtedly due to lateral compression and the average

30 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fic. 7

Moschops capensis. Lateral view of the posterior
two-thirds of a skull from Kruisvlei, Beaufort West.
S.A.M. 11972. 1/6. Note. All the figures in this
paper are projections and not perspective drawings.
Lateral view indicates an orthoprojection of the lateral
surface on to the median (sagittal) plane; dorsal
view—the dorsal surface projected on to the plane
in which the alveolar borders of the maxillaries lie;
occipital view—the occiput projected on to a plane
at right angles to the above two; ventral view—a
projection on to the plane in which the alveolar
borders of the maxillaries lie.

Moschops capensis. Occipital view of S.A.M. 11972
x 1/6.

a2 P<
- Pree

s
eile
\
.
-
?
.
-
-
“.
:
¢

“\
1 ‘
/ s H : ‘ “ \
1
,
U
U

Moschops, capensis. Ventral view of S.A.M. 11972
x 1/6.

.
-

-
-
,

.
+
’
aK
wer
-

PE ey CIO
Soe

Ny
es So cy

“9

Selita he aad
.
ae

Fic. 4

Moschops capensis. Dorsal view of an imperfect skull
from Groot Kruidfontein, Prince Albert. S.A.M.
5010. xX 1/6.

THE MOSCHOPID SKULLS IN THE

Fic. 5

Moschops koupensis nov. sp. Lateral view of a very
good skull from Die Krans, Prince Albert. S.A.M.
Pee. x 1/6.

Fic. 6

Moschops koupensis nov. sp. Lateral view of S.A.M.
11582 X 1/6 with the slight dorso-ventral compres-
sion corrected.

SOUTH AFRICAN MUSEUM 31

Fic. 7

Moschops koupensis nov. sp. Dorsal view of S.A.M.
11582 X 1/6 with the bilateral symmetry restored.

Fic. 8

Moschops koupensis nov. sp. Occipital view of S.A.M.
11582 x 1/6 with the symmetry restored, but not
the effects of a slight dorso-ventral compression.

32 a ANNALS OF THE SOUTH AFRICAN MUSEUM

proportion appears to be about 112 per cent.) The Moschops skull is thus fairly
short and broad.

Genotype
Moschops capensis, Broom 1911 (figs. 1-4)

Specific diagnosis: |

Medium to wide across parietals, with the interorbital width 73-100 per
cent of the intertemporal width. Snout relatively wide. Occipital surface
unreduced and transverse pterygoidal rami weak.

Holotype A.M.N.H. 5550. Nearly complete skull with mandible which
probably belongs to the same individual.

Spitskop Laingsburg Low ? Tapinocephalus zone Coll. Whaits.

Topotypes A.M.N.H. 5551-5557. Parts of eight skeletons.

Referred specimens:
S.A.M. 5010. Posterior two-thirds of a skull. Groot Kruidfontein, Prince

Albert.
Low Tapinocephalus zone. Coll. Haughton.

S.A.M. 11291. A practically complete skull but distorted by dorso-ventral
compression.
' Kruisvlei, Beaufort West. Low Tapinocephalus zone. Coll.
Boonstra.

S.A.M. 11295. A fairly good posterior two-thirds of a skull.
Koringplaas, Laingsburg.
Low ? Tapinocephalus zone. Coll. Boonstra. _
S.A.M. 11970. Substantial pieces of ten separate skulls showing various parts
of the cranial structure.
Kruisvlei, Beaufort West.
Low Tapinocephalus zone. Coll. Boonstra.
S.A.M. 11972. An excellent occiput and posterior half of a skull.
Kruisvlei, Beaufort West.
Low Tapinocephalus zone. Coll. Boonstra.
S.A.M. 11973. Good posterior two-thirds of a skull.
Kruisvlei, Beaufort West.
Low Tapinocephalus zone. Coll. Boonstra.
S.A.M. 11974. A good skull which has been subjected to a little dorso-ventral
pressure.
Kruisvlei, Beaufort West.
Low Tapinocephalus zone. Coll. Boonstra.

Moschops koupensis sp. nov. (figs. 5-9)
Specific diagnosis:
Very wide across the parietals with the interorbital width 70 per cent of
the intertemporal width. Snout relatively narrow. Occipital surface greatly

THE MOSCHOPID SKULLS IN THE SOUTH AFRICAN MUSEUM 33

reduced by overgrowth from above and from the sides. The transverse ptery-
goidal rami are strong.
Holotype S.A.M. 11582. A very good skull with part of the mandibles.
Die Krans, Prince Albert.
Low ? Tapinocephalus zone. Coll. Boonstra and Botma.

DELPHINOGNATHUS
Skull characters of the genus

The preorbital Jength is 61 per cent of the total median length (as recon-
structed); the snout is thus apparently fairly long (preorbital length 195 ? mm.)
and it is narrow (125 ? mm.) and fairly high (108 mm.) with the height 86 per
cent of the width. The dorsal cranial bones are not very strongly pachyostosed.

The prefrontal is not greatly thickened.

The area around the pineal foramen is greatly raised to form a conical
mound standing well above the general dorsal surface.

The postorbital bar is fairly narrow and lightly built.

The posttemporal arch is apparently fairly narrow and lightly built and
the temporal fossa is fairly large with its antero-posterior diameter not greatly
reduced.

, The dorsal parietal surface is only moderately broad (120 mm.) and the
interorbital width is 96 per cent of the intertemporal width.

The intersquamosal width is apparently fairly small (200 mm. as recon-
structed) and the median length is probably about 160 per cent of the width and
the skull is thus relatively long and narrow.

There is a very distinctive notch between the jugal and the quadratojugal.

Genotype |
Delphinognathus conocephalus, Seeley 1892 (fig. 10)

Specific diagnosis as for the genus.
Holotype S.A.M. 713. An incomplete, somewhat distorted skull.
Locality. unknown, but probably from high up in the Tapinocephalus zone
near Beaufort West. Coll. T. Bain.
Referred specimens: ' .
S.A.M. 11971. A fair skull much distorted by lateral pressure.
Kruisvlei, Beaufort West.
Low Tapinocephalus zone. Coll. Boonstra.

AVENANTIA
Skull characters of the genus
The preorbital length (as reconstructed) is 45 per cent of the total median
length and the orbit thus does not lie wholly in the posterior half of the skull;
the snout (as reconstructed) is thus short (preorbital length 174 mm.) and
broad (210 mm.) and fairly low (126 mm.) with the height 60 per cent of the
width.

34 ANNALS OF THE SOUTH AFRICAN MUSEUM

Moschops koupensis nov. sp. Ventral view of S.A.M.
11582 Xx 1/6, with symmetry restored, but the slight
forward shift of the suspensorium due to some
dorso-ventral pressure is not corrected.

Fic. 10

Delphinognathus conocephalus. Lateral view of the
type skull S.A.M. 713 1/6, clearly showing the
notch between the jugal and the quadratojugal.

Fic. 11 es
Avenantia kruisvleiensis. The type skull S.A.M. 9166 xX
1/6: Top-lateral view; middle-dorsal view; bottom-
ventral view. The symmetry has been restored but more
allowance should have been made for the effects of the
derso-ventral pressure that the skull has undergone. This
applies mainly to the too forward position of the
suspensorium.

THE MOSCHOPID SKULLS IN THE SOUTH AFRICAN MUSEUM 35

The dorsal cranial bones are strongly pachyostosed (but moderate in the
intertemporal region and along the median line).

The prefrontal is greatly thickened and this is confluent with the pachyosto-
sis of the nasal and frontal, but this does not greatly overhang the orbit although
the orbital border is fairly thick.

A well-defined ringwall is developed round the pineal foramen, but this
does not rise well above the general dorsal surface.

The postorbital bar is moderately wide and massive.

The posttemporal arch has its upper edge narrow and lightly built and the
temporal fossa is fairly large with a large antero-posterior diameter.

The dorsal parietal surface is narrow (54 mm.) with the development of a
low crista curving down posteriorly, and the interorbital width is 333 per cent
of the intertemporal width.

The intersquamosal width is great (348 mm.) and the median length (as
reconstructed) 112 per cent of the width and the skull is thus short and broad.

Genotype
Avenantia kruisvleiensis, Boonstra 1952 (fig. 11)

Specific characters as for the genus.

Holotype $.A.M. 9166. A good skull lacking only the tip of the snout and
the lower jaw.

Kruisvlei, Beaufort West.

Low Tapinocephalus zone. Coll. Boonstra.

MoscHOGNATHUS

Skull characters of the genus
Only fragments of the skull and a good mandible are known, but much of
the postcranial skeleton is preserved. No skull characterization can thus be
given here.
Moschognathus whaitsi, Broom 1914

Holotype A.M.N.H. 5602. Good mandibles with which are associated part
of the skull and much of the skeleton.

Beaufort West district.

High ? Tapinocephalus zone. Coll. Whaits.

PNIGALION

Skull characters of the genus

The dorsal cranial bones are not very strongly pachyostosed. The area
around the pineal foramen is thickened to form a low mound standing above
the general dorsal surface. The postorbital bar is not wide and is fairly lightly
built. The posttemporal arch is moderately wide and massive and the temporal
fossa is fairly large with its antero-posterior diameter not much reduced. The
dorsal parietal surface is only moderately broad (120 mm.). The intersquamosal

36 ANNALS OF THE SOUTH AFRICAN MUSEUM

width is fairly great (375 mm.). The quadratojugal as identified by Watson
(1914) shows a large squarish outer face. There is a distinctive step at the
junction of the squamosal and tabular.

Genotype
Pnigalion owent, Watson 1914

Holotype B.M.(N.H.) R3596. A good posterior two-thirds of the skull and
dentaries.

De Cypher, Beaufort West.

Low Tapinocephalus zone. Coll. Seeley.

MoscHowes
Skull characters of the genus
As only part of the lower jaw is known no skull characterization can be
given.

Genotype
Moschoides romeri, Byrne, 1937

Holotype. Walker Museum, Chicago. A fairly good skeleton.
Hottentotsrivier, Beaufort West.
Low ? Tapinocephalus zone. Coll. Romer and Miller.

DIscUSSION

As the cranial material of Moschognathus, Pnigalion and Moschoides is so
inadequate and as, moreover, they could very well at this stage be considered
co-generic with Moschops, only three Moschopid genera, viz. Delphinognathus,
Moschops and Avenantia, need be discussed here.

In the genus Moschops the more heavily built or more pachyostotic skulls
may be considered to represent either (a) males or (b) mature and old indivi-
duals male or female, and the more lightly built or less pachyostotic skulls to
be of either (a) females or (5) juveniles of either sex.

Now, in the Deinocephalia generally it is clear that in the more primitive
forms (Moschosaurus, Agnosaurus, Delphinognathus) there is little pachyostosis and
in the more advanced forms the pachyostosis becomes more and more developed.
The pachyostosis is thus a phyletic feature. But within certain species there is
also sufficient evidence that the pachyostosis is also a function of age in the
individual. There is, however, little real evidence that the pachyostosis is
sexually determined.

Thus, the degree of pachyostosis in those skulls referred to the genus
Moschops should be considered to be dependent on age only.

In the known two skulls of Delphinognathus the pachyostosis is moderate,
being chiefly confined to the conical mound around the pineal foramen and a
little thickening of the supra- and postorbital borders. There is practically no

THE MOSCHOPID SKULLS IN THE SOUTH AFRICAN MUSEUM 37

thickening, or very little, in the posttemporal, postorbital and infratemporal
and infraorbital arcades.

Now, in no lightly built Moschops skull do we find this localization of the
pachyostosis, these skulls being only slightly larger than that of Delphinognathus.
If the lightly built Moschops skulls are correctly considered to be those of young
animals then Delphinognathus cannot be considered a juvenile of Moschops and is
thus rightly considered to be a distinct genus.

In Avenantia the pachyostosis is well developed in the postorbital bar, the
dorsal and dorso-anterior orbital rim and the roof bones up to the fronto-
parietal suture. But in the parietal region the pachyostosis is little developed,
the posttemporal arch remains lightly built and the temporal fossa remains
roomy. No process of simple ageing can possibly be thought capable of trans-
forming the Avenaniia skull into a Moschops skull.

Efremov (1940) has maintained that the Moschops skull is a growth stage of
Mormosaurus with Tapinocephalus as the final product. He does not mention the
Struthiocephalids and one wonders where he would fit them in this process of
ageing.

In all the known heavy Moschops skulls the sutures are closed which allows
one to conclude that they are mature. But in some Tapinocephalus skulls the
sutures are still open and these skulls are of the same size as one in which the
sutures are closed. The large skulls with open sutures may thus be considered
immature. They and Moschops cannot thus both be considered immature
Tapinocephalus.

Moreover, in Tapinocephalus the dorsal and lateral borders of the occiput
slightly overgrow and encroach on to the planum occipitale reducing its size. Now,
in most heavy Moschops skulls there is no or very little overgrowth and encroach-
ment, but in the skull of M. koupensis, which is somewhat smaller than the heavy
M. capensis skulls, this overgrowth and encroachment is very much greater than
in the very much larger skull of Tapznocephalus. Obviously the skull of M. kou-
pensis cannot possibly be a young Tapinocephalus. In any case, Mormosaurus, with
its longer snout, cannot be intermediate between Moschops and Tapinocephalus
in both of which the snout is short. Similarly the long-snouted Aeratocephalus
and Struthiocephalids (except Struthionops) cannot be intermediate between
Moschops and Tapinocephalus.

In our present stage of knowledge the Moschopids, Struthiocephalids and
Tapinocephalids must be considered as distinct directions in the evolution of the
‘Tapinocephalia.

REFERENCES

Boonstra, L. D. 1936. ‘Some features of the cranial morphology of the tapinocephalid deino-
cephalians. Bull. Amer. Mus. nat. Hist., 72, 2, 75-98.

Boonstra, L. D. 1952. ’n Uiters interessante nuwe deinocephaliér, Avenantia kruisvleiensis, gen.
et sp. nov.’ Tydskr. Wet. Kuns, 12, 2, 250-5.

Broom, R. 1910. A comparison of the Permian reptiles of North America with those of South
Africa. Bull. Amer. Mus. nat. Hist., 28, 20, 197-234.

38 ANNALS OF THE SOUTH AFRICAN MUSEUM

Broom, R. 1911. On some new South African Permian reptiles. Proc. Zool. Soc. Lond., 1911,
1073-82.

Broom, R. 1914. A further comparison of the South African dinocephalians with the American
pelycosaurs. Bull. Amer. Mus. nat. Hist., 33, 9, 135-41.

Broom, R. 1932. The mammal-like reptiles of South Africa and the origin of mammals. London:
Witherby. :
Byrne, F. 1937. A preliminary report on a new mammal-like reptile from the Permian of South

Africa. Trans. Kans. Acad. Sci., 40, 221-4.

Byrne, F. 1940. Notes on the evolution of the mammal-like reptiles. Trans. Kans. Acad. Sci., 43,
291-6.

Efremov, J. A. 1940. Ulemosaurus svijagensis Riab.—ein Dinocephale aus den Ablagerungen des
Perm der UdSSR. Nova Acta Leop. Carol., N.F. 9, 59, 155-205.

Gregory, W. K. 1926. The skeleton of Moschops capensis Broom, a dinocephalian reptile from
the Permian of South Africa. Bull. Amer. Mus. nat. Hist., 56, 3, 179-251.

Seeley, H. G. 1892. On Delphinognathus conocephalus (Seeley) from the Middle Karroo beds,
Cape Colony, preserved in the South African Museum, Cape Town. Quart. F. geol. Soc.
Lond., 48, 469-75. “*

Watson, D. M. S. 1914. The Deinocephalia, an order of mammal-like reptiles. Proc. Zool.
Soc. Lond., 1914, 749-86.

The ANNALS OF THE SOUTH AFRICAN MUSEUM are issued in parts at

intervals as material becomes available.

Some volumes and parts are out of print, and others are only sold as parts of: a set, or ina |

respectively. The prices of parts published prior to 1940 have been increased.
Out of print: Vols. I, II (Parts 1-3, 5, 7, 8), III (Part 1), V Bho 1, 2, 5, 8,9), VI

Index), VII P (Parts I, 2, 3, 5), VIII, IX (Parts 1, 2), X (Part 2), XI

(Part 1), XXIV (Part 2), XXXI (Parts I-3).

Vol.

1900-1902 Zoology and Geology
1903-1905 Zoology .. as oe és .- (excl. Part. 1)
1903-1908 Palaeontology a
1906-1910 Geology, paca Zoology, Anthropology (eae
Parts 1, 2, 5, 8, 9)
1908-1910 Zoology a AS i eaee ‘Batts I, dee
1908-1913 Palaeontology ae ae . (Parts 4 and 6 only)
Ig11-1918 Botany a “ efeiced Parts 1 and 2)
1911-1914 Zoology .. ae oy : . (excl. Part 2)
1911-1918 Zoology (excl. Parts 2, 7, Index)
1913-1924 Palaeontology and Galas An oe oe ve
1913-1923 pee oe and ae Hadi Ble ban ep hniie Je
1915-1924 Zoology : : os & . ss oe
1914-1916 Zoology .. an “9 ee oe a we
1917-1933 Botany “e ae o7 2 ee «. ne
1917-1920 Zoology .. 4 a oy oe es a
1921 Zoology .. AP ae ee o. ae ee
1924-1925 Zoology .. .«. He Pe oe Ms Pe
1924-1926 Zoology .. es ae ia a oar! ee
1925-1927 Zoology nu .-(excl. Part 1)
1925-1928 Pilacnendaey a in aie ae v. <a
1925-1926 Zoology .. aA ip ee
1929-1938 Anthropology and Ethnology .. (excl. Part 2)
1927-1928 Zoology .. ie ve oe “a ae
1928 Zoology .. Sp ae Ve ey
1929 Anthropology ie a =) -*
1929-1932 Palaeontology of es ie a oie
1929-1931 Zoology
1931-1935 Zoology ; ‘ as
of papers, authors, and stibjects, Published in Vols. I-XXX re
1934-1950 Palaeontology . (Part 4 only)
1935-1940 Zoology .. oe ay se oa + ste
1939 Zoology .. os i - ae an os
1938 Zoology .. ie va a on vi f
1956 Zoology .. ava ee 58 big oe 5%
1942-1948 Zoology .. “4 Be 4h oe we a
1947-1952 Archaeology yy as ay a yi oe
1950 Zoology .. “A oy a wis as ss
1952 Zoology .. se ie og wie ps re
1952-1956 Botany os oe * o- oe ee ve $e
1952-1955 Zoology .. ie af ‘a ine ue os
1953-1956 Palaeontology ip an ee ae a
1955-1957 Zoology and Palnsossolaay Ag is ba 3

(excl. Parts 1-3, 5, 7, 8)

Copies may be obtained from—

The LIBRARIAN, Soutn Arrican Museum, Cape Town.

(Parts 2, 7, Index.

ROP NOM HANHNHWOOW NDNA me De Na DOR WAOW AR OA DN
~
wo

P 4 4
Yin
. +.
‘ .. pe ¥
pag
Hor
4.3

ae t
ore Sou

La d. 4

I 2 BS cust

1 16 Oo
3 7 3

16 PS

_
o>
SCMOMMOGCHCGCDOOCIOMIOOCHWOOCOOUWOWwW OWE

val
f *
, Z
+

: ve 4 2 f
5c tata eames .
fy é *
; of a ty : ts »
¢ * Ret 8
Mera Se Mets aat dae ty CUTE Naira Be ints nea
Bos pip Liam ae Mined Hae reece Meer ;

OF THE

= Some new Caddis Flies ( Trichoptera) from the Western Cape Prstiea Te By
i M. F. Scort, Ph. ae F.R.E.S. (With 4 peed in the text.)

Ia i

. fe 2

wuzest 1

ISSUED MAY 1958 PRICE 2s.

a ‘PRINTED FOR THE 2
_ TRUSTEES OF THE SOUTH AFRICAN MUSEUM
BY THE RUSTICA PRESS (PTY.) LIMITED, COURT ROAD, WYNBERG, CAPE

PM ii Maal Srha ot

SOME NEW CADDIS FLIES (TRICHOPTERA) FROM
THE WESTERN CAPE PROVINCE — II

K. M. F. Scort, Ph.D., F.R.ES.
Department of Zoology, University of Cape Town

(With 4 figures in the text)

INTRODUCTION

This is the second paper in the present series, in which new species of
caddis flies from the Western Cape (mainly from the Great Berg River) are
being described. The imagos of some new Leptoceridae are described in this
paper; their immature stages will be described in Part III of the series.

Methods and other introductory matter have been given in Part I (Scott,
1955). All drawings for text-figures 2-4 were made to the same scale with the
aid of camera lucida attachments (in the case of wings a binocular stereoscopic
microscope was used at a magnification of 12:5 diameters; in the case of
genitalia specimens cleared in a mixture of parachlorophenol and chloral
hydrate were drawn at a magnification of 80 diameters, using a research
microscope). For text-figure 1 the wings were drawn to a smaller scale as the
species is larger. The wing notation and the terminology used in the description
of genitalia are those employed by Mosely and Kimmins (1953). Holotypes
will be lodged in the South African Museum; paratypes will be sent to the
British Museum (Natural History).

Detailed descriptions of the habitats of the larvae of the new species will
be found in Harrison and Elsworth (in press, 1958) and Harrison (in press,
1958).

Grateful thanks are again due to the Council for Scientific and Industrial
Research for financial assistance in the form of a Senior Bursary, which has
made this work possible. The author would also like to express her gratitude
to Dr. Kimmins of the British Museum (Natural History) for helpful comments,
as well as to Mr. A. D. Harrison of the Council for Scientific and Industrial
Research, Dr. K. H. Barnard of the South African Museum, and Professor
J. H. Day of this Department.

DESCRIPTION
Genus ATHRIPSODES Billberg

Billberg, 1820, Enum. Ins. Mus. Billberg: 94.
Milne, 1934, Stud. N. Amer. Trich., 1: 18.

40 ANNALS OF THE SOUTH AFRICAN MUSEUM

Ross, 1938, lll. Nat. Hist. Surv. Bull., 21: 155-7
Kimmins, 1949, Entomologist, LX XXII, No. 1030; 201-4

= Leptocerus auctt. nec Leach
Athripsodes prioni sp. n.
Fig. 1, A-K

A brownish species with wide hind-wings and distinctive rather complex
male genitalia bearing numerous short orange-coloured spines; inferior
appendages trilobed. The larvae are swimmers; older larvae live in cases
made from dead palmiet leaves (Prionium serratum (L.f.) Drége), whence the
trivial name is derived.

Imago (in alcohol). Head: light chestnut brown with mingled brown and
yellow setae; antennae annulate brown and yellowish; eyes black. Mazxillary
and labial palps fuscous, covered with yellowish pubescence. Thorax: chestnut-
brown, mesonotum with short recumbent yellowish setae covering the mid-
dorsal area, flanked on each side by a double line of long erect setae; sternites
light chestnut; membranous parts pale yellow. Legs: tibial spurs 2, 2, 2;
the anterior pair short, the second and third pairs long; inner spur of second
and third pairs considerably longer than outer. Femora and tibiae yellow-
brown; tarsi, particularly of forelegs, somewhat darker. Abdomen pale yellow,
tergites brownish.

Wings: 3 9:0-9°5 mm.; 2 8-5 mm. (fig. 1 A-C). ¢ fore-wing: membrane
brownish, darker and thickened along R and Sc, particularly in the region of
the pterostigma; a slight fold between R and Sc. Pubescence brownish with
a slightly brindled appearance due to an admixture of grey hairs; a brown
patch over the wing-spot; a short dark fringe along the posterior border;
forks 1 and 5 present (1, 3 and 5 in 9); stalk of fork 1 about half length of Rg.
¢$ hind-wing: fawn, thinly pubescent, a slight fold just above Cu,; broadest
at base, much broader than fore-wing; folded under along 2A; fringe short
except along turned-in portion where it is very long and silky, there are also
long hairs springing from 3A and from the wing between 3A and the margin;
Sc is thickened and there is a thickened patch at the base of the wing; forks 1
and 5 present, stalk of fork 1 approximately twice length of R,; a false vein
present between 1A and 2A. Q as 3, excepting that in fore-wing stalk of fork 1
is sub-equal to R, and fork 3 is present; the hind-wing though broad is not as
broad as in the g.

Genitalia: 3 (fig. 1 D-G). The ninth segment is narrow dorsally, broad
ventrally, the lateral margins forming short side-pieces; the apical margin is
produced to form a pair of long preanal (superior) appendages, fringed ventrally

NEW CADDIS FLIES (TRICHOPTERA) FROM WESTERN CAPE 4I

Fic. 1. Athripsodes prionii sp. n.

A, B, fore- and hind-wings of g. C, tip of fore-wing of 9. D, E, F, G, dorsal, posterior, ventral
and lateral views of ¢ genitalia (in D the right dorsal process is omitted and in E the tip of the
penis). H, J, K, lateral, posterior and ventral views of 2 genitalia.

42 ANNALS OF THE SOUTH AFRICAN MUSEUM

with stout orange-coloured spines and narrowing abruptly almost midway.
Between these appendages the tenth segment forms an oval membranous dorsal
plate, produced downwards laterally as a thin flange on each side of the penis;
each flange bears a brush of strong orange spines. The penis is stout, unarmed,
with an expanded membranous tip, and appears to have thin chitinous lower
penis-covers closely adhering to its ventral side. The paired claspers (inferior
appendages) are trilobed. The lower lobe arises near the mid-ventral line
below the penis, and is somewhat bowl-shaped, with an irregular, setose margin;
it curls inwards to end in a tuft of orange spines beneath the penis, and outwards
where it is produced upwards to form the second, lateral, lobe, a narrow
sinuous finger-like projection. The third branch arises from the base of the
first, as a tube which opens out apically into a flattened trumpet-like expansion;
this curls inward, and its inner margin is armed with 10-12 stout claw-like
orange spines; near the apex of the trumpet is a small projection bearing several
very long colourless setae, and there are a few similar setae on the upper corner
of the main lobe. The shape of the claspers is somewhat reminiscent of some
Triaenodes species, but the wing neuration is that of Athripsodes. The ninth
segment from beneath has a U-shaped excision along its anterior margin,
paired lateral projections, and a dense band of hairs along the posterior edge.

Genitalia: 9 (fig. 1 H, J, K). Sternite of eighth segment modified to form
a flat sub-genital plate with lateral margins expanded as chitinous flaps (these
form an angle of about 45° with the ventral surface of the body and are fore-
shortened in ventral view); through this plate the vagina can be seen in a
cleared specimen. Sternite of eighth segment separated from ninth by a
bilobed transverse flap of thin chitin which is connected to the lateral flaps of
the eighth by a spoon-shaped chitinized lobe on each side. The ninth sternite
consists of an oval flattened area flanked by a pair of trilobed setose cerci;
lateral to these the margins form thickened chitinous flaps (in posterior view
these are seen to curl round the cerci); between the ninth and tenth sternites
is a transverse chitinous bar, and posterior to the bar a small semicircular flap.
Ninth tergite normal, slightly extended and curled over posteriorly. Tenth
segment small; appears to have a central opening surrounded by a thickened
margin and flanked by a pair of small cerci tipped with a few hairs; this
opening appears to connect dorsally with the vaginal opening in the single 9
available.

The species is easily separable from other African species of Athripsodes by
the very characteristic male genitalia, particularly the strong armature of
orange-coloured spines and the trumpet-shaped lobe of the claspers.

Locality: Palmiet River, Elgin (A.D.H.), November 1952, 2 gd, 1 9, bred
out from larvae in the laboratory. Larvae were collected from Elgin in October
and November, and from the Great Berg River at Assegaibos, Driefontein,
Groot Drakenstein and Bridgetown (April to October).

NEW CADDIS FLIES (TRICHOPTERA) FROM WESTERN CAPE 43

Athripsodes bergensis sp. n.
Fig. 2 A-K

A small greyish insect with strongly striped front legs and annulate
antennae; the larvae are black-headed crawlers living in long slender cases
made from coarse sand grains, to which sticks or bits of charcoal are often
attached. The trivial name is taken from the Great Berg River, in Zone IITA
of which the larvae abound.

Imago (description from dry, freshly killed insects, made before preservation
in alcohol). Sexes similar, but 9 slightly smaller than g. Head: blackish, with
strong white or greyish hairs; antennae with basal part annulate cream and
black, distal part grey; eyes black. Maxillary and labial palps black with grey
pubescence. Thorax: black, pruinose, mesonotum with short, recumbent white
setae on mid-dorsal area, flanked on each side by erect black setae; sternites
blackish, pruinose. Legs: tibial spurs 2, 2, 2, sizes as in A. prionii; anterior pair
of legs: femora and tibiae black, with black setae on inner side and white setae
on the outer, tarsi annulate black and white, spurs whitish; middle pair of
legs: similar but duller, femora browner, spurs pale fawn with white pubescence;
hind legs yellowish, with slightly darker annulations on tarsal joints, pubescence
whitish. The legs of different specimens vary somewhat in intensity of coloura-
tion. Abdomen green.

(Specimens in alcohol: general appearance brownish; thorax sepia;
wings mottled brown and grey; legs yellowish, markings dull and hardly
visible; antennae still conspicuously annulate.)

Wings: 3 5:5-6°5 mm. (Elgin specimen 7°5 mm.); 9 5:5-5°6 mm. (fig. 2
A-C). Fore-wing ¢: brindled dark grey and white; membrane brownish,
darker and thickened along Sc and R and at the pterostigma, with clear streaks
along M and Cu, and 1A, and sometimes a few clear spots as well; forks 1
and 5 present (1, 3 and 5 in 9); stalk of fork 1 approximately two-thirds of
length of R,. Hind-wing J: membrane dusky, with sparse brownish pubescence,
broadest in middle, broader than fore-wing, a small portion folded under along
2A; fringe long and silky, particularly along lower margin of wing and along
turned-in portion; Sc thickened and a thickened patch at base of wing. Forks
I and 5 present, fork 1 small, stalk five to six times length of R,. (In @ Rg is
very short and stalk is about six times length of Rg.)

Genitalia: 3 (fig. 2 D-G). Ninth segment much narrower dorsally than
ventrally; side-pieces slightly produced and with heavily chitinized posterior
margins; ventral plate ends posteriorly in a short triangular point. The whole
ninth segment is covered with microtrichia, which are slightly longer on the
ventral process than elsewhere. Dorsal plate of tenth segment wide, bilobed,
setose; from it a broad median dorsal process projects backwards; this process
is irregularly sub-triangular as seen from above and is somewhat folded proxi-

44 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fic. 2. Athripsodes bergensis sp. n.

A, B, fore- and hind-wings of g. C, tip of fore-wing of 9. D, E, F, dorsal, ventral and lateral
views of ¢ genitalia (seta and microtrichia further enlarged to show relative sizes). G, lateral
view of f genitalia with clasper removed. H, J, K, dorsal, lateral and ventral views of 2 genitalia.

—=_—— — a. =

NEW CADDIS FLIES (TRICHOPTERA) FROM WESTERN CAPE 45

mally, the distal end bears a few stout bristles. The dorsal process is hollowed
out ventrally and beneath it lies a complex set of processes which seem from
their basal attachments to represent upper penis-covers; these include a pair
of stout lateral processes each of which bears a pair of strong bristles and a
ventro-lateral patch of fine pubescence; also two membranous lobes apparently
lying one above the other, the lower one just above the penis, the upper one
partly within the hollow of the dorsal process. On each side of these, mesial to
the lateral processes, is a stout socketed spine. The penis is stout and mem-
branous, with a chitinized sclerite strengthening the tip; on each side of it is
another socketed spine, and below it, closely applied to it, a pair of chitinized
lower penis-covers. Each of the four socketed spinous processes appears to
consist of a bundle of fused setae. The claspers are trilobed as seen in lateral
view, with a stout keeled base covered with microtrichia. The anterior branch
is slender, with a truncate tip bearing three long setae; the middle branch as
seen in dorsal view consists of two rounded inwardly projecting setose lobes;
the posterior branch is the stoutest and also curls inwards. In ventral view it is
seen that the claspers each have an additional basal process just beneath the
penis.

Gemtalia: 2 (fig. 2 H, J, K). Seventh segment normal; eighth segment
normal, but with the posterior corners of the sternite slightly produced; in a
cleared specimen the vaginal structure is visible through these two segments.
Ninth segment with finely pubescent tergite; sternite subdivided into three: a
pair of finely folded lateral lobes, and a narrow tongue-like central plate. Tenth
segment with a bilobed setose dorsal plate very like that of the 4, posterior to
this the tergite narrows and curls downwards; it is joined by vertical side-
pieces to a flat rounded sternite with a large U-shaped apical excision. There
is also a pair of small leaf-like cerci; each of these arises from a bridge-like side-
piece (probably belonging to the ninth segment) anterior to which there is a
patch of small fine chitinous points.

This species is easily separated from other African species so far described
by the genitalia, both ¢ and 9.

Locality: Imagos: Great Berg River near Driefontein (A.D.H., 1 3, 2 99,
bred out from larvae, March 1953; K.M.F.S., 7 3d, 6 99, bred out from larvae
and pupae, October-December, 1956); Palmiet River, Elgin (A.D.H., 1 3,
bred out from larva, November 1952). Larvae: collected from the Great Berg
River near Driefontein and Groot Drakenstein from March to October.

Athripsodes tuckert Barnard ? var.
Fig. 3 A-J

This may be a distinct species, or only a variety of Athripsodes tuckert
Barnard (1934). It is a small species with male genitalia close to those of

46 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fic. 3. Athripsodes tuckeri Barnard ? var. and A. tuckeri Barnard

A, B, fore- and hind-wings of ¢, A. tuckeri ? var. C, D, dorsal and lateral views of 3 genitalia,
A. tuckeri ? var. E, lateral view of $ genitalia, A. tuckeri ? var. (clasper removed to show penis).
E,, tip of spinous process to show spinules. F, left clasper of § from another specimen of
A. tuckeri ? var., to show variation in shape (projection marked by arrow seen by transparency).
G, H, J, dorsal and lateral views of 3 genitalia of A. tuckeri, and clasper of another specimen
(from Jonkershoek); in G right clasper is out of place and only the lower part has been shown;
in H the clasper has been removed; in J the projection marked by the arrow is seen by
transparency.

NEW CADDIS FLIES (TRICHOPTERA) FROM WESTERN CAPE 47

A. tuckeri, the most conspicuous differences being seen in the shape of the dorsal
plate and dorsal process of the tenth segment.

3 imago (in alcohol). Head: face yellowish, vertex light brown; antennae.
annulate brown and yellow basally, distal portion dusky; eyes black. Palps
fuscous. Thorax: pronotum yellowish, mesonotum chestnut, with setae
arranged as in A. prioni, scutellum outlined in darker brown, metanotum paler
brown, sternites brownish-yellow. Legs: tibial spurs 2, 2, 2; legs yellowish-
brown, tarsi of fore-legs darkened distally, particularly along the anterior side.
Abdomen pale yellow.

Wings: 3g 6°5-7°5 mm. (fig. 3 A, B). ¢ fore-wing chequered brown and
gold, giving a golden-brown appearance; there is a whitish spot at the arculus
and a dark streak along the margin from the arculus to the apex of the wing.
Fringe short. Membrane brownish, darkened and slightly folded along R and
Sc, and thickened at the pterostigma. Forks 1 and 5 present; stalk of fork 1
subequal to R,. Hind-wing sparsely pubescent, brownish, broadest in middle,
broader than fore-wing, folded under along 2A. Fringe short except along
folded-in portion. Sc thickened and a thickened area at base of wing; forks 1
and 5 present, fork 1 small, stalk five to six times length of Rg.

Gemtalia: 3 (fig. 3 C-F). Ninth segment narrow dorsally, broad ventrally;
side-pieces somewhat produced, each with a small projection near the dorsal
side; sternite ending in a small rounded point. Tenth segment with a bilobed
transverse dorsal plate, posterior to which is a stout median dorsal process armed
with colourless bristles; in dorsal view the process is slightly sinuous with a
rounded (not expanded) tip, in lateral view it is a blunt hatchet shape. Below
the dorsal process, and attached to it for the proximal half of its length, are
paired upper penis-covers which extend downwards lateral to the penis, each
ending ventrally in a strong curved spinous process with two or three spinules
near its tip. Under the upper penis-covers is a median membranous hood. The
penis is stout, the membranous part folded and the tip strengthened with a spur-
shaped chitinous bar; on each side of it there is a stout down-curved spine or
titillator, and below it are paired lower penis-covers. The claspers are strong,
leaf-shaped, with a stout supporting flange; dorsal margin formed of five small
setose lobes, of which the central three usually project inwards and are not
visible in lateral view; one or more triangular lobes at the postero-ventral
corner.

Remarks. The main differences between this variety and Athripsodes tuckert
Barnard lie in the shape of the transverse dorsal plate and of the median dorsal
process (cf. fig. 3 C, D and E with G and H); note also the extent of attachment
of the dorsal process to the upper penis-covers and the thickness of the spinous
processes. These differences were constant in the three males available, and
appear to be sufficiently marked to warrant a full description being given, in

48 ANNALS OF THE SOUTH AFRICAN MUSEUM

case this should prove to be a new species when more material is available. The

figures of A. tuckeri given for comparison were drawn from type material kindly
loaned by the South African Museum (fig. 3 G-J). The claspers also show
differences, but these are less marked as different specimens show a considerable
amount of variation; on the whole however those of the variety seem to be
consistently wider and stronger than those of A. tuckeri. The number of spinules
on the spinous processes varies, in both, from one to three. Wings are similar,
though in the hind-wing of A. tuckert fork 1 may be minute or absent.

Should this eventually prove only to be a variety, the figures given will
serve as an indication of the range of variation within the species.

Locality: Great Berg River, Driefontein (A.D.H. and K.M.F.S., December
1952 and January 1955, in each case one 4, bred out in the laboratory from
material collected from backwaters). Also one ¢ imago caught flying at dusk
at Groot Drakenstein, lower down on the Great Berg River (A.D.H., Novem-

ber 1953).

Genus LEpTECcHO Barnard

Barnard, 1934, Trans. Roy. Soc. S. Afr., XXI: 349
Barnard, 1940, Ann. S. Afr. Mus., XXXII: 647.
Kimmins, 1956, Trans. R. ent. Soc. Lond., 108: 117-46

The new species described below falls into the genus Leptecho Barnard
according to the key given by Dr. Kimmins (1956). The only difference from
Athripsodes appears to be the absence of fork 3 in the 2 fore-wing, the wing neura-
tion being similar in both 9 and 3, and it seems very probable, as Dr. Kimmins
suggests, that Leptecho will eventually have to be sunk in Athripsodes. In the
meantime, however, the species described here joins Barnard’s two species,
L. scirpi and L. lupi, in the genus Leptecho.

Leptecho helicotheca sp. n.
Fig. 4 A-N

A fairly small, slender, grey caddis fly with annulate antennae. The ¢ is
larger than the 9 and has considerably longer antennae, but the sexes are similar
in colouring. The larvae are crawlers with bright brown heads, living in neat
snail-shaped cases made of sand grains. The larvae resemble Athripsodes larvae,
but the cases are remarkably similar to the larval cases made by species of
Helicopsyche; the pupal cases however are straight, not coiled.

Imago (description made from dry, freshly killed insects, before preservation
in alcohol). Head: face brownish; vertex and back of head leaden in colour
(actually dark brown to black but heavily pruinose), setae mingled white and
grey; eyes brown to black; antennae very long (about one and a half times

NEW CADDIS FLIES (TRICHOPTERA) FROM WESTERN CAPE 49

body length in 99, more than twice body length in $3), proximally annulate
black and white, distally grey. Palps leaden with silvery pubescence. Thorax:
tergites leaden, with silvery-white hairs; pleura and sternites brownish,
pruinose, with sparse silvery hairs. Legs: yellowish with silvery pubescence
except as indicated: fore-legs: lower part of anterior side of femora and
anterior side of tibiae, grey, tarsi black and white annulate; mid-legs: anterior
side of tibiae and tarsi greyish, tarsi with faint annulations; hind legs: posterior
side of tibiae and tarsi greyish. (Leg colouring varies somewhat in intensity in
different specimens.) Spurs on tibiae as in Athripsodes: 2, 2, 2, and of similar
proportions. Abdomen dark green, pruinose, genitalia yellowish.

Wings: 5 7:°5-8:0 mm.; 9 5°5—-6-0 mm. (fig. 4 A, B). ¢ fore-wing: mem-
brane yellowish-brown, iridescent, densely covered with brownish-grey and
silvery hairs (in most specimens the proximal half of the wing is browner and
the distal half more silvery, in one 3 however the dorsal half was more silvery
and the ventral half browner); a silvery spot at the arculus; fringe short,
brindled. Forks 1 and 5 only present in both g and 9; Sc, R and the ptero-
stigma slightly thickened; stem of fork 1 sub-equal to Ry. ¢ hind-wing:
membrane pale fawn with sparse silvery pubescence; fork 5 only present in
both ¢ and 9; hind-wing broader than in 9, in both sexes the wing is turned
under along 3A and this portion bears a very long silky fringe.

(Specimens in alcohol: general appearance light golden-brown with
chestnut head and thorax; antennae annulate yellow and brown; wings
brownish-grey brindle; legs yellowish.)

Genitalia: ¢ (fig. 4 C-J). Ninth segment somewhat narrower dorsally
than ventrally, bearing long setae, specially on the sternite; side-pieces with
strongly chitinized posterior margins. The tenth segment comprises a mem-
branous dorsal plate terminating in a small, oblong, median process tipped
with three long setae, and, lateral to the dorsal plate, a pair of large, finely
pubescent lobes studded with sword-like stalked setae. The median projection
overlies a membranous hood, lateral to which are paired upper penis-covers each
bearing several stout setae and with an obliquely truncate apex. The penis
itself is short, with an expanded bifid tip and a horseshoe-shaped chitinous
support; beneath it there are short lower penis-covers with rounded down-
curved ends. The claspers are short and strong, trilobed in lateral view, with
a slight ventro-lateral keel. In dorsal or ventral view it is seen that the lower
lobes of the claspers are incurved, setose, with sinuous inner edges and with a
small inwardly projecting tooth near the tip. The setose median lobes project
inwards; the upper lobes are truncate and tipped with two or three long setae.

Genitalia: 2 (fig. 4 K—-N). Eighth segment slightly flattened ventrally ;
vagina visible through this in cleared specimens. Ninth segment with a rounded
setose dorsal plate; sternite slightly hollowed forming a sub-genital plate with
curved lateral margins and a flat, bifid, backwardly projecting central plate.

50 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fic. 4. Leptecho helicotheca sp. n.

A, B, fore- and hind-wings of 3. C, D, dorsal and ventral views of ¢ genitalia. E, detail of upper
penis-covers etc., dorsal view. F, G, lateral views of 3 genitalia (IF with clasper removed).
H, dorsal view of claspers. J, lateral view of 3 genitalia (another specimen to show variation
in shape of clasper and detail of upper penis-cover). K, L, M, N, lateral, ventral, dorsal and
posterior views of 9 genitalia (in L the chitinous bar of the hood is seen by transparency).

NEW CADDIS FLIES (TRICHOPTERA) FROM WESTERN CAPE HE

Tenth segment hood-like, with a posterior opening strengthened by a chitinous
bar which appears semicircular from behind but W-shaped in dorsal view.
This hood is joined laterally to a flat, bilobed transverse plate which is larger
than the hood but transparent and therefore difficult to see. The tenth segment
bears two large, slightly hairy cerci, which, together with the flat transverse
plate and the small bifid process, surround the genital opening.

Remarks. This species can easily be distinguished from Leptecho scirpi and
L. lupi by the structure of the male genitalia, in particular by the absence of
titillators, the shape of the claspers and the dorsal plate, and the existence of a
median dorsal process.

Locality: Great Berg River near Driefontein (2 §¢ March 1951, A.D.H.;
2 3d; 2 29 May 1957, K.M.F.S.; all bred out from larvae. Larvae were also
collected there from November to March.) Great Berg River, Groot Draken-
stein (gg and 99 collected in June 1951, A.D.H.).

SUMMARY

Three new species of caddis (Trichoptera: Leptoceridae) are des-
cribed from South Africa: Athripsodes prionii, A. bergensis and Leptecho helicotheca ;
also a variety of Athripsodes tuckert Barnard.

REFERENCES

Barnard, K. H. 1934. South African Caddis-flies (Trichoptera). Trans. roy. Soc. S. Afr. 21,
291-394.

Barnard, K. H. 1940. Additional Records, and Descriptions of New Species, of South African
Alder-flies (Megaloptera), May-flies (Ephemeroptera), Caddis-flies (Trichoptera), Stone-
flies (Perlaria) and Dragon-flies (Odonata). Ann. S. Afr. Mus. 32, 609-661.

Barnard, K. H. 1941. May-flies and Caddis-flies from Natal, Basutoland and Pondoland.
Ann. Durban Mus. 3, 105-108.

Harrison, A. D., and Elsworth, J. F. 1958. Hydrobiological Studies of the Great Berg River,
Western Cape Province. Part I. Trans. roy. Soc. S. Afr. (in press).

Harrison, A. D. 1958. ibid., Part II. Trans. roy. Soc. S. Afr. (in press).

Kimmins, D. E. 1949. Some Changes in Generic Names in the Family Leptoceridae (Order
Trichoptera). Entomologist, 82, 201-204.

Kimmins, D. E. 1953. Trichoptera collected by Miss R. H. Lowe in Uganda, with descriptions
of three new species of Leptoceridae. Entomologist, 86, 274-278.

Kimmins, D. E. 1956. New and little known species of the Leptocerinae (Trichoptera) from
the African Mainland (south of the Mediterranean region). Trans. R. ent. Soc. Lond. 108,
117-146.

Mosely, M. E. 1933. The Genus Pseudoleptocerus Ulmer (Trichoptera). Ann. Mag. nat. Hist.
Ser. 10, 11, 537-547.

Mosely, M. E. 1936. New African Trichoptera, I. Ann. Mag. nat. Hist. Ser. 10, Vol. 17, 429-451.

Mosely, M. E. 1939. Trichoptera. Ruwenzori Expedition 1934-5. British Museum (Nat.
Hist.), III, Part 1, pp. 1-40.

Mosely, M. E. 1939. New African Caddis Flies (Trichoptera). Ann. Mag. nat. Hist. Ser. 11,
Vol. 3, 1-28.

Mosely, M. E., and Kimmins, D. E. 1953. The Trichoptera (Caddis Flies) of Australia and New
Kealand. British Museum, London.

52 ANNALS OF THE SOUTH AFRICAN MUSEUM

Navas, R. P. L. 1930. Insectes du Congo belge. Rev. Zool. Bot. afr. 19, 323-336.

Scott, K. M. F. 1955. Some new Caddis Flies (Trichoptera) from the Western Cape Province
—I. Ann. S. Afr. Mus., 41, 367-380.

Ulmer, G. 1905. Neue und wenig bekannte Trichopteren der Museen zu Briissel und Paris.
Ann. Soc. ent. Belg. 49, 17-42.

Ulmer, G. 1913. Trichopteren von Aquatorial-Afrika. Wiss. Ergeb. D. Zentral-Afrika Exp. IV.
Zool. 2, 81-125.

Ulmer, G. 1922. Trichopteren aus dem Agyptischen Sudan und aus Kamerun.’ Mitt. miinch.
ent. Ges. 12, 47-63.

ACKNOWLEDGEMENT

The publication of this paper was made possible by a grant from we
South African Council for Scientific and Industrial Research.

The ANNALS OF THE SOUTH APRICAN museum are : issued inp -
intervals as material becomes available. : sc

Some volumes and parts are out BE nAAG ed es are only ae as cats ofa re t ;
respectively. The prices of parts published prior to 1940 have been increased. ag

. Out of print: Vols. I, II (Parts 1-3, 5, 7, 8), III (Part 1), V
r Index), VII (Parts 1, 2, 3, 5), VIII, LX (Parts 1, 2), X on 2), XI (Parts 2, ae i
(Part 1), XXIV (Part 2), sae ts (Parts 1-3).

Vol.

4%,
= :
TR AS > Fahy
; CS hat
; ane ce % fio es
* * ay aa
€a' td eae nai ’ 1A
+ sS hae. ¥ ee
ay Nat e

(Parts 1, 2, 5, 8, 9)>.

vues 1-3, 53 75 7 cf

1900-1902 Zoology and Geology
1903-1905 Zoology .. Sony une t us . (excl. Part. 1) See
1903—1908 Palaeontology :
1906-1910 Geology, Palaeontology, Zoology, Anthropology (excl, »
Parts’1, 2.°5,,0; 9) 2h) ss
1908-1910 Zoology... ¢ Meek Parks i; "Ise 2
1908-1913 Palaeontology . (Parts 4’and 6 only)
1911-1918 Botany eek Parts1 and2) 2
1911-1914 Zoology ue . (excl. Part 2) 4
1g11-r9g18 Zoology (excl. Parts 2, 7, Index) 2
1913-1924 Palaeontology and Cacia | : (oe
IQI3—1923 spires h and coe Reet kc plave “s <a ee
1915-1924 Zoology : ae Be e 3 ie. 3
1914-1916 Zoology .. Bh Se e i sd re i
1917-1933 Botany ue At ihokiags % e ae a eae
1917-1920 Zoology .. ig eat a a.
1921 Zoology .. i ie ae i os (S Ceg ee
1924-1925 Zoology .. te ue =o “fe Raper a
1924-1926 Zoology oy 33 . 2 ee
1925-1927 Zoology . (excl. Part 1) > xe
1925-1928 Palaeontotany <p ae i st
1925-1926 Zoology. .. an uf Sage
1929-1938 Anthropology and Ethnology .. (excl. Part 2) > 229 a0
1927-1928 Zoology : si i. aoe
1928 Zoology .. By wes oF id Pa ine I 10 0 pe
1929 Anthropology By Ze ma ty a .. oan
1929-1932 Palaeontology 1 oo) a
1929-1931 Zoology ie “i an ay te 2. ioe
1931-1935 Zoology... o 0: SASS
of papers, authors, and oe. naples G in ala. XXX 2. So ena
1934-1950 Palaeontology . (Part4 only) 014 0
1935-1940 Zoology .. me a = - “s Bide ec,
1939 Zoology 2250.5
1938 Zoology 2. Bou
1956 Zoology 6 0 OW
1942-1948 Zoology 2 4T of
1947-1952 Archaeology te ie opr at Rent ee) Hass
1950 Zoology .. st ita SR a ee y Pg 4 + \. 3 oa
1952 Zoology i 214 6 #
1952-1956 Botany se i a i iy ee 2) pe
1952-1955 Zoology .. si: fi te a ..) ye ye
1953-1956 Palaeontology i F 3 3 hie ae
1955-1957 Zoology and Palacontolinn os 2 0 oO.
1957- Parts 1,2 & 3, Zoology and Palaeontology 10 6

The LIBRARIAN, Soutu Arrican Museum, Care Town.

Copies may be obtained from—

‘ s

PRINTED FOR THE

REVIEW OF THE RACES OF THE SPIKE-HEELED LARK,
CERTHILAUDA ALBOFASCIATA LAFRESNAYE

J. M. WINTERBOTTOM
(South African Museum)

CONTENTS
PAGE PAGE
INTRODUCTION. : 4 : ‘ 53 SUMMARY OF RACES. ; : : 58
ACKNOWLEDGEMENTS : : d 54. PROPORTIONS . : : ‘ 5 66
REVIEW OF MATERIAL. : : 55
INTRODUCTION

In 1956, the List Committee of the South African Ornithological Society
asked me to make a preliminary investigation on their behalf of the races of
Certhilauda albofasciata. This paper is the result of work begun for that
purpose. )

The races of Certhilauda albofasciata have been most recently reviewed by
Meinertzhagen (1951) and, in respect of the western forms, by Macdonald (1953).
These two reviews do not agree with one another, though both recognize
fewer races than does Roberts (1940) and, following him, Vincent (1952). The
comparative data are set out in Table I. Neither Meinertzhagen nor Macdonald
defines the ranges of the subspecies they recognize with precision (though this
only applies to the eastern races in Macdonald’s case), but it is clear that in
some cases these ranges do not correspond to the joint ranges of the various
“Roberts-recognized’ subspecies fused in the synonymy. The problem is not
rendered any easier by doubts as to the correct names to be applied to the
different races; and, in particular, as to which form should be called C. albo-
fasciata Lafres. The type of C. albofasciata is now at Harvard. The locality is
said to be ‘Cape of Good Hope’, but, apart from the fact that there is no record
of the species occurring within 70 miles of the Cape, this locality was given at
that time to a wide, but indefinite, area, covering much of the present Cape
Province. Comparisons of the type specimen with others of known origin
suggest that it belongs to the population found in the north-central and north-
eastern Cape and southern and eastern Orange Free State; and Macdonald
has recently (1958) proposed Deelfontein as the type locality.

Andrew Smith described garrula in Ill. Zool. S. Afr., Aves, 1846, pl. 106.
In the description, he gives a reference to a prior description of his own in

54 ANNALS OF THE SOUTH AFRICAN MUSEUM

Proc. S. Afr. Inst., 1833. These Proceedings are contained in the S. Afr. Quart.
Journ. and no such description appears in the period 1830-6. Smith gives the
_ range as ‘the northern districts of the Colony’, which might be almost anywhere
north of the Olifants River and Little Karoo. Macdonald (1953), pointing out
that Smith’s figure is of a darker bird than that shown by Lafresnaye, has
accordingly applied the name garrula to the dark, south-western population and
fixed Vanrhynsdorp as the type locality. Actually, the colouring of Smith’s
plate appears to me so unlike any Certhilauda albofasciata specimen I have ever
seen that I think it might have been wiser to have rejected the name altogether
as indeterminable; but to do so now would cause a great deal of confusion and
I propose to accept it.

It must be noted, however, that the consequences of accepting these
findings of Macdonald’s are that some of the birds called albofasciata by Roberts,
Vincent and Meinertzhagen must now be called garrula; and some, at least, of
the birds called subpallida or alticola by these authors now become albofasciata
(allowing for differences in recognition of subspecies).

ACKNOWLEDGEMENTS

The total number of skins assembled for the present study was 374. I am
much indebted to the Directors and staff of the following Museums for the loan
of material: British Museum (Natural History), London; Coryndon Museum,
Nairobi; Durban Museum and Art Gallery; East London Museum; Kaffra-
rian Museum, King William’s Town; National Museum of Southern Rhodesia,
Bulawayo; Port Elizabeth Museum and Snake Park; and the Transvaal
Museum, Pretoria. I am also indebted to Messrs. J. D. Macdonald and
C. M. N. White for giving me their views on some of the problems raised,
though they must not be held responsible for any of the opinions expressed here;
and to my colleagues of the List Committee (Messrs. R. H. N. Smithers,
J. Vincent, P. A. Clancey and R. Liversidge) for much helpful criticism of my
initial review. I must extend my sincere thanks, too, to Mr. Smithers and
Mrs. B. P. Hall for making available the fine series of Spike-heels collected in
southern Bechuanaland by the joint British Museum-National Museum
expedition to that area, as noted below; and to Mr. P. A. Clancey for a
preview of the material collected by the Durban Museum in the northern
Cape.

I am most grateful to Mr. A. N. Rowan for statistical treatment of the
measurements; and to Mr. E. H. J. Middlemiss for information on the
geography and vegetation in the Johannesburg-Pretoria area, where variation
among Spike-heels is so confusing.

The work on which this paper is based was done while the author was

holding a Senior Bursary of the South African Council for Scientific and
Industrial Research.

REVIEW OF THE RACES OF THE SPIKE-HEELED LARK 55

REVIEW OF THE MATERIAL

The west coast from the Olifants valley to Little Namaqualand and inland
as far east as Calvinia is occupied by the race garrula Smith, a dark bird, notably
darker and redder above than albofasciata and much darker below.

To the south and east of this, from Karoopoort, in the Ceres District, east
through Beaufort West and southern Aberdeen to Bedford, is a race, greyer
than garrula and colder and more vinaceous below. This race is without a
name and I propose to call it macdonald, after Mr. J. D. Macdonald, who first
pointed out that it was distinct. East of this, the birds become redder and lose
the cold tone. They most closely resemble the Griqualand West series, from
which they are separated by the lighter nominate form; but they are not so
dark as those northern birds. I name them below latimerae, after Miss M.
Courtenay-Latimer, whose collections from the Cape Province have done so
much to clear up the races in that area.

Inland from the coast in the north-west Cape, three races have been
described, calviniensis (Roberts), bushmanensis (Roberts) and meinertzhagent
Macdonald. Macdonald places the first two in the synonymy of garrula;
Meinertzhagen, who does not refer to garrula, puts the first as a synonym of
albofasciata and the second of bowent.

To take calviniensis first, I have seen ten skins from Calvinia itself, besides
one from 50 miles south and another four from Lokenburg, west of Calvinia.
All these are garrula. But the East London and Durban Museums’ series from
Brandvlei, some 80 miles north-east of Calvinia, are quite different. Now
Roberts’s type locality for calviniensis is given as ‘35 miles east of Calvinia on the
Brandvlei road’; and the phrase I have italicized is omitted from the citation by
both Macdonald and Meinterzhagen. The omission is important because, in
point of fact, the road from Calvinia to Brandvlei, though it starts by going
east, swings round through go° and ends by going north. At 35 miles, it has
already started this swing; and two Transvaal Museum specimens from the
type locality are much closer to the Brandvlei birds than to garrula. They are,
however, somewhat yellower and less pink than any of the Brandvlei series, but
this may be because they are in worn plumage.

The type locality of bushmanensis is ‘border of the Bushman Flats and
Little Namaqualand on the road from Goodhouse to Steinkopf’. I have seen
the type, which can be matched by examples from Brandvlei. I am of the
opinion that only one subspecies can be recognized from this area. It is
singularly unfortunate that both the described races are from the edges of the
range, but bushmanensis has page priority and is slightly less inappropriate, so
I propose to use it.

To the north-east of bushmanensis, there are two skins from Pofadder and
five from the Putzonderwater and Britstown areas in the East London collection
and four Transvaal Museum skins from Putzonderwater and Vanwyksvlei, in

esents

specimens has been examined. Unnamed areas—specimens—

indeterminable (see text).

ANNALS OF THE SOUTH AFRICAN MUSEUM

fuossmry:

imate distribution of the races of Certhilauda albofasciata Lafresnaye. Each dot repr

a locality from which one or more

Approx

REVIEW OF THE RACES OF THE SPIKE-HEELED LARK 57

which the intensity of the central dark streak in the back feathers is considerably
reduced and the red tint of the rest of the feather is much deeper than in any
other examples known to me. In the first of these characters, they approach
bradfieldi of the region west of Upington, but they are much redder; while the
underparts are dark—darker than either bradfieldi or bushmanensis and as dark
as garrula. These birds are meinertzhageni Macdonald. In a series of eight
Durban Museum skins from Vanwyksvlei, however, only one agrees with
meinertzhagent and the others are bushmanensis.

Passing north from the ranges of bushmanensis, meinertzhageni and garrula,
we come to the ranges of three more named races. On the east, to the west of
Upington, is bradfieldi (Roberts), a subspecies of very restricted range. I have
seen eight skins of this bird, all from 20 miles west of Upington and forming a
well-marked and compact group with the dark centres of the back feathers much
reduced, so that the effect is of a rufous bird, though, in series, yellower and less
red than meinertzhagent.

To the west of bradfieldi are two named races, barbiensis (Roberts) and
arenaria Reichenow. I am unable to find any distinction between them and
agree with Macdonald and Meinertzhagen that barbiensis is a synonym.
C’. a. arenaria is paler, with narrower dark centres to the back feathers, than
garrula to the south of it: but darker than bradfieldi above, though lighter below.

North-west of the range of arenaria is bowenit (de Schauensee), of which
namibensis (Roberts) has been accepted by practically everybody, including its
author (but not, oddly enough, by Meinertzhagen), as a synonym. It is a pale,
sandy bird, with wide light margins to the feathers of the back and the dark
centres reduced and lighter or almost absent. It is also smaller.

Still further north occurs erikssoni Hartert, the lightest race of all, though
the dark centres to the back feathers are more distinct and much darker than
in bowen. ‘The birds of the area east of arenaria and erikssoni have been described
as another race, kalahariae O.-Grant. Meinertzhagen fuses this race with
erikssont, saying, ‘I believe kalahariae to be the grey phase of erikssoni’. Only
three skins of erzkssoni were available to me, but I was fortunate to have at my
disposal the long series of kalahariae collected by the British Museum-National
Museum expedition to southern Bechuanaland in 1957, and was later able to
see ten more skins of erikssoni in the British Museum collection. These certainly
do not support Meinertzhagen’s belief.

The British Museum-National Museum expedition also revealed the
presence, east of the pale kalahariae and west of the much darker Griqualand
birds, of a population resembling arenaria but less heavily streaked; and
Liversidge has recently collected two similar birds from between Stella and
Vryburg. They constitute the race bathoeni Smithers and Paterson.

East of these races and north of latiemerae and macdonald occurs a form, paler
and less rufous than macdonald: though darker than arenaria and much darker
than erikssont. This form Macdonald has shown must be called albofasciata. Its

58 ANNALS OF THE SOUTH AFRICAN MUSEUM

range extends north over most of the Orange Free State and east, according to
Macdonald, to northern Natal (no Natal skins were available to me). Westward,
eight out of nine Durban Museum skins from Riverton, Kimberley, belong to
this form, though Bulawayo and Cape Town skins from the Kimberley area
agree with the next race.

West of this, from De Aar north to Keine and west to Kuruman, the birds
are darker and show an approach to altzcola (with the exception noted below).
They represent baddeleyi Clancey. From the Vaal valley, separating the range
of baddeleyi from alticola, come five quite different skins. Two, from Christiana
and Wolmaransstad, are yellowish brown; the other three, from Kroonstad,
are a cold grey. What the status of these five is it is impossible to say in the
absence of more material, but one of the National Museum skins from Riverton
is a near match to the first two.

From the Witwatersrand-Pretoria area, two races have been described,
alticola (Roberts) and roberts: Macdonald. I am unable to distinguish between
them. Birds from this area are black and deep rufous above, darker than
griquensis and much darker than albofasciata. But, as Macdonald has pointed
out, birds from Fountains Blue and Zwartkops, between Johannesburg and
Pretoria, are quite distinct, since they lack the dark centres to the feathers and
are wholly dark rufous. They are old skins and I do not propose to name them,
but the position in this area can only be described as fantastic. On a practically
straight line from Boksburg through Pretoria to Zoutpan Road, we have the
following forms succeeding each other:

Boksburg: alticola (type locality).

Modderfontein (¢. 11 miles): subpallida.

Fountains Blue (c. 6 miles) and Zwartkop (c. 20 miles): rufous birds, as
noted above.

Fountains (c. 7 miles): robertsi (type locality) (=altzcola).

Zoutpan Road (c. 30 miles): subpallida.

I can find no correlation between this distribution and either the vegetation
as given by Acocks (1953) or the 1955 geological map.

The extreme north-east of the species’ range is occupied, from Zoutpans-
berg south to near Pretoria and to Lake Chrissie, by subpallida (Roberts),
lighter above and below than alticola, greyer above and less red below than
albofasciata.

Far to the north of any other form of the species, in central Angola, occurs
the very dark obscurata Hartert.

SUMMARY OF THE RACES

The forms recognized may be summarized:

1. Certhilauda albofasciata albofasciata Lafresnaye, Mag. de Zool., 1836: pl. 85.
Type locality, Deelfontein.

REVIEW OF THE RACES OF THE SPIKE-HEELED LARK 59

Range: North-eastern Cape Province, most of the Orange Free State and
north-western Natal. Meets baddeleyi in the Kimberley region, where the
detailed delimitation of the respective ranges needs further study.

Moderately pale and not very red.

Average measurements :
13 $3: Wing, go-1 mm. (83-97); tail, 48-5 (44-53); culmen, 23-2 (20-25);
tarsus, 25:0 (22~29); hind-claw, 13°8 (11-21).
10 99: Wing, 84:0 (81-91); tail, 43-7 (39-52); culmen, 20°3 (18-24);
tarsus, 24°4 (22-27); hind-claw, 12°5 (11-15).

Material examined:

South African Museum, 12 (Deelfontein, Hanover, Richmond, Winburg,
Middelburg, Vredefort road, Wagenaarskraal); Transvaal Museum, 16

- (Dealesville, De Brug, Zastron-Rouxville road, Bethulie, Bloemfontein,
Heilbron, Springfontein, Excelsior, Meadows, Rhenosterspruit); Durban
Museum, 3 (Elandshoek, Steynsburg); East London Museum, 1 (Elands-
hoek); Kaffrarian Museum, 1 (Colesberg).

2. Certhilauda albofasciata macdonaldi subsp. nov. Type: in South African
Museum, Cape Town, No. 20340, 3, 23 miles north-east of Karoopoort,
Ceres District, c. 33° S., 20° E.

Range: From Beaufort West and Aberdeen to the southern borders of
the Karoo. Intergrades with albofasciata in southern Victoria West (Wagenaars-
kraal) and with latimerae at Klipplaat, in the Sundays River valley.

Greyer above than albofasciata and garrula, colder and more vinaceous
below.

Average measurements :
32 gd: Wing, 90°6 mm. (82-97); tail, 46-4 (40-56); culmen, 23-4
(18-28); tarsus, 25:6 (22-31); hind-claw, 13°7 (10-18).
21 92: Wing, 85°5 (78-95); tail, 43°5 (38-55); culmen, 20-3 (17-25);
tarsus, 23°4 (20-27); hind-claw, 14°3 (7-18).
Maiterial examined:
South African Museum, 24 (Karoopoort (type), Beaufort West, Dwyka,
Oudtshoorn, Barrydale, Matjesfontein, Laingsburg, Touws River, Klaar-
stroom, Fraserburg road, Aberdeen); East London Museum, 29 (Beaufort
West, Bedford-Grahamstown road, Oudtshoorn, Skietkuil); Durban
Museum, 4 (Oudtshoorn, Murraysburg, Skietkuil); Coryndon Museum, 1
(Laingsburg); Port Elizabeth Museum, 1 (Klipplaat—intermediate
between macdonald: and latimerae).

3. Certhilauda albofasciata latimerae subsp. nov. ‘Type: in East London Museum,
No. 3235, 6, Cofimvaba, Transkei, c. 32° S., 27° 30’ E.

60 ANNALS OF THE SOUTH AFRICAN MUSEUM

Range: From the Transkei through Glen Grey, Queenstown and Tarka-
stad to the Sundays River valley, where it intergrades with the preceding.

Redder than macdonaldi above and lighter and less vinaceous below;
darker than albofasciata; lighter than baddeleyz.

Average measurements :

5 6d: Wing, 90-0 mm. (88-93); tail, 49-0 (46-52); culmen, 23-0 (21-24);
tarsus, 24°6 (23-27); hind-claw, 16-6 (15-17).

4 99: Wing, 82-2 (80-91); tail, 41-5 (35-47); culmen, 19:5 (18-23);
tarsus, 23:2 (20-27); hind-claw, 14.0 (10-15).

Material examined:

South African Museum, 2 (Queenstown, Glen Grey); East London
Museum, 5 (Mostertshoek, Cofimvaba (type)); Durban Museum, 3 (Fish
River, Tarkastad); Kaffrarian Museum, 1 (Bolotwa).

4. Certhilauda albofasciata garrula A. Smith, Ill. Zool. S. Afr., Aves, 1846, pl. 106.
Type locality, Vanrhynsdorp.

Range: The western Cape from the lower Olifants River to Port Nolloth,
east to Calvinia and Springbok.

Darker and redder above than albofasciata and much darker below. Redder
above and below than macdonald.

Average measurements :

26 gg: Wing, 92°6mm. (81-98); tail, 47-5 (41-63) ; culmen, 23-7 (21-27);
tarsus, 27:1 (22-30); hind-claw, 14:4 (11-18).

23 29: Wing, 83°8 (79-92); tail, 46-9 (38-52); culmen, 196 (18-24);
tarsus, 24°7 (21-28); hind-claw, 13-4 (11-19).

Material examined:

South African Museum, to (Liebendal, Springbok-Goodhouse road,
Lokenburg, Papendorp, 50 miles south of Calvinia); East London
Museum, 16 (Clanwilliam, Clanwilliam-Calvinia Road, Sandkraal, Bitter-
fontein, Nieuwerust-Bitterfontein Road, Springbok, Calvinia); Durban
Museum, 13 (Vanrhynsdorp, Nieuwerust, Nieuwerust-Bitterfontein road,
Okiep, Calvinia); Transvaal Museum, 9 (Klaver, Klipfontein, Nieuwe-
rust, Port Nolloth); National Museum, Bulawayo, 1 (Vanrhynsdorp).

5. Certhilauda albofasciata bushmanensis (Roberts), The Ostrich, 1937: 99. Type
locality, border of Bushman Flats and Little Namaqualand, on the road
from Goodhouse to Steinkopf, c. 29° N., 18° 30’ E.

Synonym: C. a. calviniensis (Roberts), The Ostrich, 1937: 100.

REVIEW OF THE RACES OF THE SPIKE-HEELED LARK 61

Range: Bushman Flats and Kakamas south to 35 miles north-east of
Calvinia and to Williston.
Lighter and redder above than garrula and macdonald.

Average measurements :
15 6d: Wing, 90-7 mm. (85-95); tail, 46-9 (41-52); culmen, 24-1 (20-27);
tarsus, 26-2 (23-28); hind-claw, 12-5 (9-18).
15 99: Wing, 86-3 (80-94); tail, 43°6 (38-48); culmen, 19-9 (18-24);
tarsus, 23°6 (20-28); hind-claw, 11-5 (6-15).

Material examined:

Transvaal Museum, 4 (Bushman Flats (type), Brandvlei, 35 miles east of
Calvinia); East London Museum, 19 (Brandvlei, Brandvlei-Kenhardt
Road, Williston, Kenhardt, Kakamas-Kenhardt road, Williston-Fraser-
burg Road, Fraserburg); Durban Museum, 13 (Brandvlei, Vanwyksvlei).

6. Certhilauda albofasciata meinertzhagem Macdonald, Proc. Zool. Soc., 122, 1953:
maw. ype locality, Pofadder, c. 29° N., 20° E.

Range: A narrow strip along the south side of the Orange River valley
from Pofadder to Britstown and Vanwyksvlei, where it occurs alongside
bushmanensis.

Differs from bushmanensis in that the central dark stripes of the feathers of
the upper parts are considerably reduced and the red tint of the rest of each
feather is much deeper. Much redder than bradfieldi. Underparts darker than
either of these two races and similar to garrula. |

Average measurements : |
743: Wing, 92:1 mm. (89-95); tail, 48-1 (46-52); culmen, 23-0 (20-25);
tarsus, 25:0 (23-27); hind-claw, 12-0 (11-193).
2 99: Wing, 83:7 (83-84); tail, 43°7 (42-46); culmen, 18-0 (17-19);
tarsus, 23-0 (21-25); hind-claw, 8-7 (7-10).

Material examined:

Transvaal Museum, 4 (Putzonderwater, Vanwyksvlei); East London
Museum, 6 (Pofadder, Putzonderwater, Marydale, Kenhardt-Putzonder-
water road, Britstown-Merriman road); Durban Museum, 2 (Marydale,
Vanwyksvlei).

7. Certhilauda albofasciata baddeleyi Clancey, Durb. Mus. Novit. 5, 1957: 43. Type
locality, Rietfontein, Griqualand West, ¢. 29° 15’ S., 23° E.

Synonym: C. a. griquensis Winterbottom, nom. nud., in Clancey, op. cit.

Range: From De Aar north and west to Kanye and Niekerkshoop and
east to Boshoff and Koffiefontein in the Orange Free State.

62 ANNALS OF THE SOUTH AFRICAN MUSEUM

Intermediate in colouration above between alticola and albofasciata. Below,
rather darker than albofasciata, matching alticola.

Average measurements :
22 gd: Wing 91.2 mm. (86-97); tail, 49:0 (40-56.5); culmen, 23-4
(20-26); tarsus, 26-6 (23-30); hind-claw, 13-2 (10-16).
9 99: Wing, 82-4 (79-86); tail, 42-0 (37-47); culmen, 19°5 (18-21);
tarsus, 25°3 (24-27); hind-claw, 10-8 (10-12).
Material examined:

South African Museum, 3 (Fourteen Streams, Kimberley, Boshoff) ;
Transvaal Museum, 8 (Barkly West, De Aar, Petrusville, Fourteen Streams,
Vryburg, Fauresmith, Kanye, Niekerkshoop); National Museum,
Bulawayo, 8 (Riverton, Langley, Kuruman); Coryndon Museum, 4
(Koffiefontein, Luckhoff); Durban Museum, 10 (Riverton, Rietfontein
(type)); Port Elizabeth Museum, 2 (Vryburg-Mafeking road).

8. Certhilauda albofasciata bradfieldi (Roberts), Ann. Tvl. Mus., 15, 1932: 28.
Type locality, Langklip, 60 miles west of Upington, c. 28° 30’ S., 20° 30’ E,
Range: 20-60 miles west of Upington.

Yellower, less red, than meinertzhageni but, like it, with the dark centres of
the back feathers reduced. Redder than albofasciata, lighter above and darker
below than arenaria, yellower above and lighter below than baddeleyr.

Average measurements :
6 gg: Wing, 92-0 mm. (89-94); tail, 47-7 (42-53); culmen, 24°3 (22-26);
tarsus, 26-0 (25-27); hind-claw, 13:2 (12-14).
2 99: Wing, 82-0 (80-84); tail, 41-5 (40-43); culmen, 210: (21); tarsus,
24:0 (24); hind-claw, 12-0 (11-13).
Material examined:

Transvaal Museum, 8 (20 miles west of Upington).

9. Certhilauda albofasciata arenaria Reichenow, Vég. Afr., 3, 1904: 354. Type

locality, Rehoboth.

Synonym: C. a. barbiensis (Roberts), The Ostrich, 8, 1937: 2.

Range: From the Orange River (reaching the south bank at Aggenys and
Bladgrond) north to Windhoek.

Paler above, with narrower dark centres to the feathers, than garrula.
Darker above and lighter below than bradfieldi. Less red above than bushmanensis.

Average measurements :
13 3g: Wing, 90:0 mm. (86-97); tail, 45°3 (40-51); culmen, 23°2 (21-27);
tarsus, 25°3 (22-29); hind-claw, 12°1 (10-15).

REVIEW OF THE RACES OF THE SPIKE-HEELED LARK 63

9 29: Wing, 81-9 (78-93); tail, 40-6 (36-45); culmen, 18-9 (17-21);
tarsus, 23°2 (21-26); hind claw, 11-2 (10—19).

Material examined :
British Museum, 6 (40 miles north of Kleinkaras, Witputs, Lower Barby,
69 miles south of Windhoek); East London Museum, 6 (Kenhardt,
Aggenys, Bladgrond, Bladgrond-Kakamas road); Transvaal Museum, 3
(Gobabis, Barby, Karibib); Durban Museum, 5 (Aggenys, 30 miles from
Kakamas on Kenhardt road, Bladgrond); Port Elizabeth Museum, 3
(Kenhardt).

10. Certhilauda albofasciata boweni (de Schauensee), Proc. Acad. Nat. Sci. Philad.,
83, 1931: 5. Type locality, Spitzkopjie.
Synonym: C. a. namibensis (Roberts), Ann. Trans. Mus., 14, 1931: 243.
Range: From Swakopmund north to Omaruru.
A pale, reddish-sandy bird, with wide light margins to the back feathers
and the dark centres reduced and lighter or almost absent. Smaller.

Average measurements :
10 6g: Wing, 85:8 mm. (81-91); tail, 45-2 (40-51); culmen, 22-2 (21-23);
tarsus, 23°7 (21-26); hind-claw, 13-0 (11-16).
8 O°: Wing, 78-0 mm. (75-81); tail, 41-9 (38-44); culmen, 18-4 (17-20);
tarsus, 21-8 (20-25); hind-claw, 11-8 (9-15).

Material examined:

British Museum, 14 (Ebony, Wilson’s Fountain, Karibib, Okomuhe) ;
Transvaal Museum, 2 (Ebony); National Museum, Bulawayo, 3 (Namib).

11. Certhilauda albofasciata kalahariae Ogilvie-Grant, The Ibis, 1912: 375. Type
locality, Lehututu, central Kalahari.
Range: The Kalahari desert.
‘Much greyer than boweni, the dark centres of the feathers of the back very
narrow but blackish. Below, rather darker and more vinaceous.

Average measurements :
25 6d: Wing, 90°4 mm. (83-95); tail, 50°6 (43-54); culmen, 23-4
(20-27); tarsus, 24-9 (22-28); hind-claw, 13-0 (8-16).
20 99: Wing, 83°5 (75-90); tail, 44°5 (39-53); culmen, 20°8 (18-25);
tarsus, 24°2 (21-28); hind-claw, 10-7 (8-14).

Material examined:
British Museum, 1 (Fort Rietfontein); British Museum-National Museum

Expedition, 44 (Tsabong, Lochahane, Bosho Bohulu, Tsane, Chawe,
Kome, Kakia).

64 ANNALS OF THE SOUTH AFRICAN MUSEUM

12. Certhilauda albofasciata erikssont Hartert, Bull. B.O.C., 19, 1907: 82. Type
locality, Okahokahana, Ovamboland.
Range: The Outjo District and Ovamboland.
Somewhat paler than kalahariae above and below; greyer than boweni.

Average measurements :
8 $3: Wing, 87-4 mm. (81-95); tail, 44-1 (37-45); culmen, 20-6 (19-22) ;
tarsus, 25°6 (24-27); hind-claw, 13-6 (12-16).
522: Wing, 77°6 mm. (75-81); tail, 40-4 (37-47); culmen, 18-2 (17-19);
tarsus, 22-4 (18-24); hind-claw, 11-4 (9-14).
Material examined:

South African Museum, 1 (Okahokahana); British Museum 11 (Kalk-
rand, Ondonga).

13. Certhilauda albofasciata bathoent Smithers and Paterson (description in the

press). Type locality, west of Kakia, B.P., ¢. 25° S., 24° E.

Range: A narrow strip of grass country in south-eastern Bechuanaland,
south to the Stella-Vryburg road, between the limestone country inhabited by
kalahariae and the Acacia country inhabited by baddeleyz.

Darker and redder than kalahariae, but much lighter than baddeley:. Most
like arenaria, but less heavily streaked above.

Average measurements :
5 6g: Wing, 91-8 mm. (86-97); tail, 49-0 (46-53); culmen, 25:2 (24-26);
tarsus, 26:0 (24-28); hind-claw, 13°6 (12-17).
8 99: Wing, 81-9 (80-85); tail, 44:7 (42-47); culmen, 20°8 (19-22);
tarsus, 23-9 (22-26); hind-claw, 10-6 (9-13).
Material examined:

British Museum-National Museum Expedition, 13 (53-57 miles east of
Kakia); Port Elizabeth Museum, 2 (Stella-Vryburg road).

14. Certhilauda albofasciata alticola (Roberts), Ann. Tol. Mus., 15, 1932: 28.
Type locality, Van Dyk Mine, Boksburg.
Synonym: C. a. robertsi, Macdonald, P.<.S., 122, 1953: 1003.
Range: From Potchefstroom east to Belfast and Carolina and north to
Pretoria.

The darkest of the South African races, black and dark rufous above; but
below, no darker than baddeley1.
Average measurements :

5 dd: Wing, 88-4 mm. (85-go); tail, 44:6 (41-50); culmen, 23:6 (22-25);
tarsus, 26-4 (24-28); hind-claw, 14:8 (12-16).

REVIEW OF THE RACES OF THE SPIKE-HEELED LARK 65

g 92: Wing, 83-3 (75-90); tail, 41-7 (36-50); culmen, 19°8 (18-22);
tarsus, 24°4 (21-27); hind-claw, 12-8 (8-16).

Maiterial examined:

Transvaal Museum, 15 (Boksburg, Witwatersrand, Enkeldoorn, Zwartkop,
Belfast, Carolina, Fountains Blue, Potchefstroom); Durban Museum, 1
(Witbank); South African Museum, 3 (Zwartkop, between Florida and
Roodepoort).

15. Certhilauda albofasciata subpallida (Roberts), Ann. Tol. Mus., 15, 1932: 29.
Type locality, Marabastad, Pietersburg District.

Range: From Modderfontein and Lake Chrissie north to the borders of
the Bushveld.

Paler above and below than alticola and very close to baddeleyi, but paler
below and yellower, less rufous, above.

Average measurements :
443: Wing, 87-2 mm. (86-88); tail, 43-0 (41-46); culmen, 22-2 (21-23);
tarsus, 26-5 (26-28); hind-claw, 14-2 (13-16).
2 99: Wing, 78:5 (78-79); tail, 39:0 (38-40); culmen, 20-0 (19-21);
tarsus, 25°5 (25-26); hind-claw, 13-0 (12-14).
Material examined:

Transvaal Museum, 6 (Marabastad (type), Modderfontein, Pretoria,
Lake Chrissie, Zoutpan road).

16. Certhilauda albofasciata obscurata Hartert, Bull. B.O.C., 19, 1907: 83.
Type locality, Bulu-Bulu, Bihe District, Angola.
Range: The grasslands of central Angola.
Back very dark, the feathers with whitish edges; head black, with rufous

margins to feathers; below deep rufous. Hind-claw very long, probably in
correlation with habitat.

Average measurements :

6 gg: Wing, 86-7 mm. (81-89); tail, 40-3 (37-44); culmen, 21-8 (21-23);
tarsus, 25:6 (22-28); hind-claw, 17-3 (16-19).

4 29: Wing, 80-8 mm. (74-91); tail, 36-0 (32-41); culmen, 19-2 (18-22);
tarsus, 24:0 (22-30); hind-claw, 15:8 (13-19).

Material examined:

British Museum, 7 (Bure-bure, Missao de Luz); Hall Collection, 3 (Silva
Porto).

66 ANNALS OF THE SOUTH AFRICAN MUSEUM

In addition, as noted above, the following populations cannot be fitted
into any of the above races at present and may prove to be worth naming when
adequate material is to hand:

(2) Birds from Christiana and Wolmaransstad, South-West Transvaal.
(b) Birds from Kroonstad, northern Orange Free State.

The birds from Fountains Blue and Zwartkops, included in alticola above, may
also warrant separation.

PROPORTIONS

Macdonald (1953) has shown that the sum of the lengths of the wing, tail
and bill can be used as a rough measure of size and that the ratio of the means
of these quantities between the sexes changes according to the locality, the
females representing 85 per cent of the males south-west of Windhoek, 90 per
cent over several other South-West African localities and intermediate else-
where. Working with average figures for each race, I obtained ratios varying
from 88 per cent for boweni and arenaria to 95 per cent for macdonaldi, the pattern
roughly conforming to decreasing dimorphism as one goes east and south from
west-central South-West Africa. Within races, figures are naturally less reliable
since the numbers are smaller, but variation is shown by the following popula-
tions of garrula:

Klaver-Vanrhynsdorp (5 $4, 5 22) . do, 164°4; 9, 148-2 = 90%

Calvinia (7 3d, 3 29) ; : . &, 163°3; 9 1429°o)— eae
Brandvlei (bushmanensis) birds give:
Brandvlei (11 $3, 8 99) . . =>. 6 159'03° 2) t19-2)——ene

showing an approach to equality of size, while bradfieldi gives:
20 m. W. of Upington (6 gg, 2 92) . dg, 163-0; 2 144-5) eee
more like the Calvinia population than the intervening one at Brandvlei.
In macdonaldi, we can contrast three local populations:

' Skietkuil, Murraysburg (15 gd, 7 99) do) 154-3; 9; 143°2)-— aan
Beaufort West (5 3d, 3 99) - 515725 9, 154°3 = 98%
Oudtshoorn (6 gg, 19) . . > 164°0; 9, 14970 = aie

It will be noticed that Skietkuil males are small—the same size as Beaufort West
females.

Other populations for which more or less adequate data exist are the total
population of bathoeni, all taken within a few miles of one another, and single
populations of baddeley: and kalahariae:

50-60 m. E. of Kakia (5 gg, 8 99) . og, 166-0; 9, 147-4 = 890%

Kimberley (4 $3, 3 99) - . 6g, 161-7; 9, 144-3 = 89%

Kakia (4 3d, 3 29). : é . 4, 165:03 9, 148-0 —=sgone
These all agree with the neighbouring population of bradfieldi in proportions.

REVIEW OF THE RACES OF THE SPIKE-HEELED LARK 67

Mr. A. N. Rowan, who investigated the wing-lengths statistically, tells me
that in the males only bowen: differs significantly from the average of the species.
Among the females, arenaria and macdonaldi may be significantly different. In all
races except macdonaldi, the differences between male and female are highly
significant; in macdonald, the difference may be significant. Note that the
statistical analysis of wing-measurements confirms the conclusion drawn from
Macdonald’s ‘size index’ that sexual dimorphism is less marked in macdonald
than in the other races.

REFERENCES

Acocks, J. H. P. 1953. Weld Types of South Africa. Bot. Surv. S. Afr., Mem. 28.

Macdonald, J. D. 1953. Some Aspects of Variation in the Spike-heeled Lark (Certhilauda
albofasciata). Proc. zool. Soc., 122, 985-1006.

Macdonald, J. D. 1958. Note on the Spike-heeled Larks of Cape Province. The Ostrich, 28,195-6.

Meinertzhagen, R. 1951. Review of the Alaudidae. Proc. zool. Soc., 121%, 81-132.

Roberts, A. 1940. The Birds of South Africa.

Vincent, J. 1952. A Check List of the Birds of South Africa.

TABLE I
NAMES USED FOR RACES OF Certhilauda albofasciata

Race (those in heavy type Roberts (1940) and Meinertzhagen Macdonald

recognized in this study) Vincent (1952) (1951) (1953)
albofasciata Lafres. . albofasciata albofasciata albofasciata
caluiniensis Rbts. ; calviniensis (=albofasciata) (=albofasciata)
garrula Smith .. — — garrula
alticola Rbts. . . alticola — alticola alticola
subpallida Rbts. . subpallida (=alticola) subpallida
bradfieldi Rbts.. . bdradfieldi bradfieldt bradfieldr
arenaria Rchw. . . arenaria arenaria arenaria
barbiensis Rbts. . . barbiensis (=arenaria) (=arenaria)
namibensis Rbts. . . (=dowent) namibensis (=bowent)
boweni de Schauen. . doweni bowent bowen
bushmanensis Rbts.. bushmanensis (=bowent) (=garrula)
erikssoni Hart. . . ertkssont erikssont ertkssont
kalahariae O.-Grant. kalahariae (=ertkssont) kalaharvae
meinertzhageni Macd. — — meinertzhagent
robertst Macd. : — — robertst

obscurata Hart.. . (extra-limital) obscurata (extra-limital)

bie Wain nee eae? 2)

; aie so a At i he
hn OR ab a lg a

ay f ;

; , \ ar
| , fh OS Say a
\ by
é us hoe di
9 1 (
4 7
4
‘
)
2 |
» Vy
Hy eas
1 f
' ig
i rial | \ i
. Y" )
Q ‘
»,
S
{
(
;
,
‘
i
\ *
;
y
i
i
J
F
a)

ON THE RACES OF LYBIUS LEUCOMELAS (BODD.)
IN SOUTH AFRICA

J. M. WINTERBOTTOM
(South African Museum)

Six races of the Barbet Lybius leucomelas are recognized by Roberts (1940)
and Vincent (1952) from within South African limits. The races are the same
in both, though Roberts refers to his own zuluensis as tongensis in his book, as he
does also in his description of L. 1. centralis. Thanks to the co-operation of the
Director and Mrs. Campbell of the Transvaal Museum and the Directors of
the East London and Durban Museums and the National Museum of Southern
Rhodesia, a series of 110 specimens was assembled for study; and I subsequently
also examined a series of 28 of the specimens in the British Museum (Natural
History) on which Macdonald (1957) based his findings.

My original purpose in investigating the races of this species was to check
the validity and the range, if valid, of L. 1. namaqua (W. L. Sclater), and I shall
deal with this race first.

The sole distinction given by Sclater between namaqua and leucomelas is that
the former has ‘the underparts thickly spotted with oval spots of black’. The
type locality of leucomelas is given as the Cape of Good Hope; that of namaqua
is Klipfontein. I found that most birds, though not quite all, in a coastal strip
from Namaqualand to Albany, were heavily spotted. Birds from Committee’s
Drift, Albany and Mostertshoek, Tarkastad, varied, some being heavily spotted,
others not. Clancey (1954) had already drawn attention to this and Macdonald
(1957) also noted it. One of two birds from Clanwilliam was virtually unspotted
and so was the only available specimen from Hopefield. But birds from Klaver,
Cape L’Agulhas, Stilbaai and another Clanwilliam bird were all spotted. I
noted also that one of the two British Museum birds from Klipfontein, the type
locality of namaqua, was a lightly spotted form. Unless, therefore, there is an
unspotted form confined to the area west of the Hottentots Holland Mountains
and south of the Olifants River Mountains—which is just possible but highly
improbable—we must take it that heavily spotted underparts are a normal
feature of southern and western birds.

L. 1. centralis was described by Roberts (1932) as having the underparts of
the body pale yellow (though in 1940 he calls it ‘clearer white below’). There is
certainly a yellow tinge, of very variable depth, in all the birds I have seen from
within the given range of centralis; but it can be matched by many birds from
further south —skins from Pearston and Beaufort West, for example, are rather
_ yellower than the average of South-West African and Bechuanaland birds. _

70 ANNALS OF THE SOUTH AFRICAN MUSEUM

Both Roberts and Vincent consider Eastern Cape birds to be affinis Shelley,
described from Weenen, Natal. The characters given by Shelley are, as Sclater
(1924) pointed out, those of a young bird; but in reviving the name, Roberts
substituted lesser or absent striping on the flanks and smaller size than L. 1.
leucomelas or L. 1. namaqua. He gives no diagnosis of the differences between
affinis and centralis. I have seen only one Natal bird, from Colenso, and it is
yellow below, almost as deep as Swaziland birds, but it can be matched by
specimens from further south. The size given by Roberts for the two subspecies
(centralis and affinis) is virtually identical. ;

L. l. nkatiensis (Roberts) is described as ‘silkier and whiter below’ than
centralis and much smaller. But Roberts appears to have had only one skin, a
female—and he gives the measurements of male centralis for comparison. A
series of five Bechuanaland skins, four of them topotypical for nkatiensis, in the
National Museum of Southern Rhodesia have wings 78-83 mm., the range
Roberts gives for centralis. I am unable to perceive the whiter underparts which
Roberts gives as the other characteristic of this race—indeed, as compared with
a series from South-West Africa and the northern Cape, Bechuanaland birds
appear to me to be slightly greyer, but it is not of a degree which would allow
of certain separation of individual skins. _

I am therefore of the opinion that only one subspecies of this Barbet is
recognizable from the whole of this area, and the next problem is to decide what
name to apply. If Boddaert’s bird came from what is today called the Cape of
Good Hope, then we must call the western and southern birds leucomelas, with
namaqua a synonym, and the population east and north of this affinis Shelley,
which is the next available name. There is, however, no certainty that
Boddaert’s bird did come from the Cape itself and a considerable balance of
probability that it did not; for ifit did, as we have seen, it would probably have
been spotted below. Most of the other South African birds described by
Boddaert ex Daubenton as from the ‘Cape of Good Hope’ are inconclusive on
this point, but one of them, Euplectes progne, is decisive that some, at least, of
them came from further north and east. I therefore propose to amend the type
locality of Lybius leucomelas from the Cape of Good Hope to Beaufort West,
Cape Province. This enables us to retain namaqua for the coastal birds and
leucomelas for the bulk of the population, that inland of the range of namaqua.

L. 1. zuluensis (Roberts) is described as yellower below, and smaller, than
affinis. The yellow tinge below is clear in a series but not 100 per cent evident
in individuals; the smaller size seems constant. I therefore recognize this
race.

The subspecies of Lybius leucomelas within South African limits thus become:

1. Lybius leucomelas leucomelas (Bodd.)

Bucco leucomelas Boddaert, Tabl. Pl. Enlum., 1783: 43—Cape of Good Hope;
amended herein to Beaufort West, Cape Province.

ON THE RACES OF LYBIUS LEUCOMELAS (BODD.) IN S.A.

SasIWOTTY O09
SAW OOF

APPROXIMATE DISTRIBUTION OF THE RACES OF Lybius leucomelas (Boddaert)

has been examined.

imens

Each dot represents a locality from which one or more spec

Type localities of described races are named and indicated by an open square.

72 ANNALS OF THE SOUTH AFRICAN MUSEUM

Synonyms: Pogonorhynchus affinis Shelley, Proc. Zool. Soc., 44, 1879: 680—
Weenen; WNotopogonius leucomelas centralis Roberts, Ann. Tul. Mus., 15, 1932: 26
—Rustenburg; and Notopogonius leucomelas nkatiensis Roberts, Ann. Tol. Mus.,
15, 1932: 27—Nkate.

White or yellowish white below, without spots.

Range: The whole of southern Africa except the south-western and eastern —
coastal strips.

2. Lybius leucomelas namaqua (W. L. Sclater)

Tricholaema leucomelan namaqua W. L. Sclater, Bull. B.O.C., 42, 1922: 63 —
Klipfontein.

Heavily spotted black on whitish below.

Range: The western Cape Province from Little Namaqualand inland to
south-western Kenhardt and to Calvinia; and east along the coastal low
country to Cradock, Tarkastad and Albany. Intermediates between this and
the nominate form are found at Grootderm (Orange River mouth), Deelfontein

and Albany.

3. Lybius leucomelas zuluensis (Roberts)

Notopogonius leucomelas zuluensis Roberts, Ann. Tol. Mus., 14, 1931: 240—
Mkuzi River, Zululand.

Synonym: Notopogonius leucomelas tongensis Roberts, Ann. Tol. Mus., 15,
1932: 26—Mkuzi River.

Yellowish below, without spots. Smaller than the other two races (wing,
73-83 mm., as against 78-87 mm.).

Range: Northern Zululand, the Lowveld of Swaziland and the Transvaal,
Portuguese East Africa and the Sabi Valley of Southern Rhodesia.

The work on which this paper is based was done while the author was
holding a Senior Bursary of the South African Council for Scientific and
Industrial Research.

REFERENCES

Clancey, P. A., 1954, The Ostrich, 25, 40.

Macdonald, J. D., 1957, A Contribution to the Ornithology of Western South Africa.
Roberts, A., 1940, The Birds of South Africa.

Sclater, W. L., 1924, Syst. Av. AEth., 1, 276.

Vincent, J., 1952, A Check List of the Birds of South Africa.

‘The ANNALS OF THE. 5
“interval as taaterial pecans av

Out of print: Vols. I, II (Parts 1-3, 5, 7, 8), WI (Part SY )
Index), VII (Parts 1, 2, 3, 5), VIII, IX ae a; ra x acs

Bee Fe (Part 1), XXIV (Part 2), XXXI (Parts baht
ane se PSE VOL aie we : | ;
ne II. 1900-1902 Zooloes and Genieay
III. 1903-1905 Zoology .. ree Se one
‘ IV. 1903-1908 -Palaeontology .. Heo te Sh Sia
V. 1906-1910 Geology, ie senna ty Zoology, Anthropology
<a : ie Parts 1, 2, 5, 8 Odes pe ee
i “VI. 1908-1910 Zoology .. Roig see a (excl. ‘Part ce In
VII. 1908-1913 Palaeontology : Siuaee te A .. (Parts 4 gad 6
: 1X, SiGhI-1H18) “Botany ss o7 ao aes sie (excl. Parts 1 and
X. 1911-1914 Zoology .. is pal pee ek (excl. Part
: XI. s911-1918 Zoology .. a as segs Parts 2, 7: inde
XII. 1913-1924. Palaeontology and Geakouy Ps pe Me oie
XIII. 1913-1923 Archaeology and Zoology Rte up ak 3
- XIV. 1915-1924 Zoology .. ne a Serge Mere
. XVi Agi4-1916 -;. Zoology G8 eo a ee ee Wee
: XVI. 1917-1933 Botany... si aie eee te son Sy
‘e MVE." 1917-1920 | Zoolosy Se eo sk ee pale 5 alt rpm ora He
, XVIII. 1924. "Zoology. 620" ya iag chat ees ood ae
XIX. 1924-1925 Zoology .. Bin A teagan Wael eR aie nee me
MX: 1924-1996 Zoology. sg pa eee ea eg al A
XXI. 1925-1927 Zoology .. herder V. . (excl. Part 1
are AIT, 1925-1988, sPalacontdlogy, i cas we iis
pet y, XXIII. 1925-1926 Zoology .. — Lebar Are a
ead 3 XXIV. 1929-1938 Anthropology and Ethnology ater tt (excl, Part ¢
gra XXV. 1927-1928 Zoology .. a op se Byard aie
He Se AVI. 1928 Zoology .. ae 1 te eo tee
XXVII. 1929 Anthropology 3 ‘i va mney Fay
XXVIII. 1929-1932 Palaeontology i a oe ene
XXIX. 1929-1931 Zoology .. a Da nates co sida kes
iy XXX. 1931-1935 Zoology .. ae ho ao ina een)
3 INDEX of papers, authors, and subjects, published in Vols. LXXX Saas
XXXI. 1934-1950 Palaeontology ae RH 5 . (Part 4 only)
XXXII. 1935-1940 Zoology .. + er he or ae aa 5
XXXITI. 1939 Zoology .. a caches op ea on
XXXIV. YORB “ZOOL Pa wars a te Se a as
se XXXV. 1950 = *Z00lOgy *)) veo Wows Hs asa settee es
| XXXVI. 1942-1948 Zoology WEA cA ea eee te ie rainy
XXXVII. 1947-1952 Aechaedlocy’ ae a huh tea aie oi exten cae A
XXXVIII. 1950 Zoology .. Be ss Pisa gh age eis
XXXIX. 1952 Zoology... De teh ae ui aM Lago
XL. 1952-1956 Botany... an in i Kee
XLI. 1952-1955 Zoology .. rb eae cre AF ene me hy
XLII. 1953-1956 Palaeontology esha hates ae oi Sema
Ka XLIII. 1955-1957 Zoology and Pulacdniolowy , va
j XLIV. Part 1, 1 & m, Palaeontology and Zoology as

1957

The LIBRARIAN, Sourn Arrican Museum, CAPE ‘Town. Be

Copies may be obtained rome

— hd eo i. aed ‘ 7" a
Fe AY et oo a Sts
CL e Re eee
eG ‘ ea 28
} ‘ *«

me - ANNATS

OF THE

~ SOUTH AFRICAN MUSEUM

VOLUME XLIV

PART IV, containing: —

_ Contributions to the knowledge of South African Marine Mollusca. Part I.
Gastropoda: Prosobranchiata: Toxoglossa. By K. H. Barnarp. (With
30 figures in the text and 1 plate.) | |

ISSUED JULY 1958 PRICE 15s.

; PRINTED FOR THE
TRUSTEES OF THE S APE
BY THE RUSTICA PRESS (PTY.)

CONTRIBUTIONS TO THE KNOWLEDGE OF
SOUTH AFRICAN MARINE MOLLUSCA

Part 1. GASTROPODA: PROSOBRANCHIATA: TOXOGLOSSA
By
K. H. BARNARD

(With 30 figures in the text and 1 plate)

INTRODUCTION

A vast amount of research will be necessary before a comprehensive and
reliable monograph on South African Marine Mollusca can be produced.
Therefore the following work is only a preliminary account of some of the species
known to inhabit South African seas, and of their distribution around the coast.

It is based mainly on the collections in the South African Museum, and the
collection made during the University of Cape Town’s Ecological Survey. Of
the former the most important is the collection made by the Cape Government
trawler s.s. Pieter Faure. The latter comprises material, collected alive, from the
intertidal zone and the estuaries, and by dredging in a few localities.

In South Africa, perhaps more than in other regions, the study of mollusca
has suffered from the enthusiasm of shell-collectors, most of whom have been
beach-combers with little or no interest in the living animals. Large numbers
of dead shells have been described and given scientific names by overseas
specialists, often without fully appreciating the possibilities of variation in
coloration and pattern, or of the alteration in shape and sculpture which can
be produced by beach-rolling.

‘The fauna of a few favourite and accessible strands is adequately
known, mainly of course from dead and more or less beach-worn material.
When “‘live’”’ examples, dredged or otherwise, do turn up, it is frequently
a matter of the greatest difficulty to appraise their relationship with the
worn types or battered series which constitute all that is hitherto available.’
(Tomlin. Ann. S. Afr. Mus., xxv, p. 313, 1928.)

Thus a check-list would contain many hundreds of names; but how many
of these names denote true species—species which a zoogeographer can regard
as components of the South African fauna, and can accept for comparison with
the faunas of other regions?

In past years, in addition to the persistent describing of ‘species abrasae
litoris’, erroneous identifications have swelled the list. On the other hand,

1Tomlin. 1922. 7. Conch., 16, 256-8.

74 ANNALS OF THE SOUTH AFRICAN MUSEUM

energetic collecting of living molluscs has drastically reduced it, e.g. in the genus
Patella.?

Dead shells recorded from the greater part of the South African coast—
from the Cape to Pondoland or southern Natal—may be accepted in most cases
as genuine inhabitants of these waters, especially species whose shells occur in
abundance, because the habitats of the living animals are likely to be close at
hand.

Very different, however, is the case of records from Natal and Zululand,
whose shores are washed by the southward flowing Mozambique current,
rendering possible the existence of many tropical Indo-Pacific species. Here an
element of doubt creeps in, because most collectors do not confine themselves
rigidly to one area, but include in their collections shells from neighbouring
regions, e.g. the east coast of Africa or Mauritius. They may have spent a
holiday in those areas, or accepted shells from friends; in the latter case the
locality is hearsay. How many of the Indo-Pacific species of Conus or Cypraca
(collectors’ ‘favourites’), which have been recorded from Durban or Natal, are
known to live in South African waters and thus entitled to inclusion in the
fauna-list ?3

Another source of error, known to have occurred sometimes in entries in
Museum accession registers, and suspected in some published records of
localities, lies in recording the domicile of the collector (or donor) as the habitat
of the mollusc.*

These remarks are not intended to discredit collectors, but to show that the
only (scientifically) good mollusc is a live one.

A full historical account of the growth of our knowledge of South African
molluscs must wait for inclusion in a monograph, but a few of the more impor-
tant steps may be mentioned here.

Dr. Ferdinand Krauss visited the Cape and Natal in 1840. He himself
collected and described many molluscs, and the beautiful illustrations in his
work have scarcely been surpassed.® He did no dredging, but it is obvious that,
for the most part, he collected living specimens.

In later years several beach-combers made collections Whig were submitted
to overseas conchologists, e.g. Dr. E. Fritsch (1874), S. D. Bairstow, Dr. H.
Becker, and Lt.-Col. W. H. Turton.

After his earlier collections had been described (by Edgar Smith, and
Bartsch), Turton himself undertook the description of his main collection. His
work may be described as a tour de force, but it is a stumbling-block to students
of South African molluscs because it obviously contains many synonyms, which

* Tomlin & Stephenson. 1942. Proc. Mal. Soc., 25,4; Koch. 1949. Ann. Natal Mus., XI, 487.

3 cf. Sowerby. 1889. 7. Conch., 6, 10.

* Bairstow lived at Port Elizabeth, but did the recorded specimen of Purpura trigona really come
from Port Elizabeth? Did Mrs. Trotter’s example of Voluta festiva really come from Natal?

5 Die siidafrikanischen Mollusken. Stuttgart, 1848.

CONTRIBUTIONS TO KNOWLEDGE OF S:A. MARINE MOLLUSCA 75

can only be correctly assigned by the re-examination of his collection (at
Oxford) by someone thoroughly conversant with the South African fauna.

H. C. Burnup was a most careful and assiduous collector, especially of
living material. The animals he sent to Professor Gwatkin, who extracted the
radulae, but no shells were sent for verification.® In this manner errors are liable
to occur, and indeed have occurred (e.g. Euthria queketiz).

The littoral fauna, however, is but a small part of the fauna of a marine
province. Several overseas expeditions have ‘fished’ molluscs in our waters.
The U.S. North Pacific Exploring Expedition visited Simon’s Bay in 1853, and
the naturalist W. Stimpson collected both on shore and by dredging.’

H.M.S. Samarang, H.M.S. Sulphur and H.M.S. Challenger added to our
knowledge. Later, with improved methods, and probably more care in sorting
the material, the German Deep-sea Expedition (Valdivia) obtained many new
species.®

All these expeditions were passing voyages. The first systematic exploration
of the coastal seas around South Africa was carried out by the Cape Government
under the direction of Dr. J. D. F. Gilchrist (Government Biologist, 1895-1907).
The object was primarily economic: to discover payable fishing-grounds for
_ steam-trawlers, but the scientific aspect was by no means neglected. During the
years 1897-1907 the Cape Government trawler s.s. Pieter Faure operated around
the coast from St. Helena Bay to Zululand, mostly in shallow and moderate
depths, but down to 400 fathoms off East London and Natal, and 1,000 fathoms
off Cape Point.

Large collections of marine organisms were obtained which, on the
discontinuance of the Survey, were transferred to the ownership of the South
African Museum (1910).

As regards the Mollusca the ‘cream’ of the collection was reported on by
G. B. Sowerby (Third) before the collection came to the South African Museum.?
After registration further material was submitted to the late J. R. le B. Tomlin.”

The reports by these two specialists are almost entirely conchological. In
only very few instances did Sowerby mention the anatomy or the radula,
although several of the novelties were described from specimens containing the
animal. There is evidence to show that Professor Gwatkin extracted the radulae
of some of these animals.' Excepting two species, all the material sent to
Tomlin was dry.

® Cooke. 1919. Proc. Malac. Soc., 13, 109.

7See: Bartsch. 1915. Bull. U.S. Nat. Mus., no. 91, p. 2.

8 Von Martens. 1903. Wiss. Ergebn. D. Tiefsee Exp., 7; Thiele. 1925. ibid., 17; Thiele &
Jaeckel. 1931. ibid., 2/.

® Marine Investigations in South Africa, i-v, 1898-1908. N.B. The dates of publication of the
reports in this work are frequently earlier than the dates on the title-pages of the respective
volumes. See: Barnard. 1950. 7. Soc. Bibl. Nat. Hist., 2, 6, 187.

1 Ann. S. Afr. Mus., 20, 1925; 25, 1927-8; 29, 1931; 30, 1932 & 1934.

11 e.g. Conus patens. See: p. 90.

76 ANNALS OF THE SOUTH AFRICAN MUSEUM

The greater part of the Pieter Faure collection of Mollusca, when transferred
to the Museum, was preserved in formalin (later changed to alcohol), the
exceptions being the specimens returned dry by Sowerby. On two occasions
(one before the present writer was appointed to the staff, the other when he was
overseas on leave), the collection was encroached upon to provide a representa-
tive series of South African molluscs for the public exhibition galleries. There
would have been no objection to this procedure if the animals extracted from
their shells had been replaced in the jars and preserved. They were not, though
the opercula were neatly mounted in the customary manner. Thus, most.
regrettably, several species, the animals of which would have supplied most
desirable taxonomic data, remain known to science by their shells alone.

Nevertheless the hitherto unknown radulae of several species have been
obtained, and are described in the present work. Not unexpectedly, some
species are shown to have been wrongly classified.

Recent examination of the Preter Faure bottom-samples has resulted in
finding not only several additions to the fauna-list (e.g. in the family Turritidae),
but also further examples of species hitherto known only from singletons (e.g.
Drillia fossata Sow., Solartella persculpta Sow., Heliacus petasus Tomlin) and unworn
specimens of species known only from beach-worn examples (e.g. Afritrophon
insignis). Moreover these bottom-samples have supplied data for plotting the
distribution of some of the species; for this purpose broken examples and even
small fragments, if the sculpture is distinctive, are useful.

The Museum Molluscan collections comprise also several private collec-
tions which have been acquired in the course of years by gift and bequest. While
most of the shells in these collections are of little use for scientific purposes,
examination of large quantities, sometimes hundreds of one and the same
species, often provides successive stages in weathering and abrasion and thus a
clue to the manner in which natural agencies can produce ‘new species’ for the
over-enthusiastic conchologist.

The collection bequeathed by the late Dr. John Muir of Riversdale must
be specially mentioned. In his retirement he patiently sorted through bagfuls
of sand collected at Still Bay, and thereby obtained an extremely useful collec-
tion of the smaller species (Turbonilla, Eulimella, Alvania, etc.).

At the suggestion of the present writer he concentrated on tracing the
growth, from protoconch upwards, of the commoner species. These series have
been of great value in the present work, in which special attention has been paid
to the protoconch and early whorls with a view to providing more precise
definitions of the species.

Some of the features of the South African marine province have been
discussed by von Martens (1903) and Tomlin (1922). Tomlin has cleared the
ground by excluding records of European species based on erroneous identifica-
tions, due probably to beach-worn material. But comparison with other marine
provinces is scarcely justified until at least the great majority of the species have
been correctly classified. As an example, the Antarctic genus Glypteuthria has

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 77

been recorded from Cape waters; but one of the species assigned to this genus
has been shown to belong to a different family (Barnard, 1957), and the other
is a species inquirenda. ‘Therefore this genus cannot be claimed as indicating an
Antarctic element in the South African fauna.

The recent identification (Knudsen, 1956) of a West African Nassa
(Nassarius) with a Natal species (V. desmouliozdes); the extraordinary likeness of
the West African Pleurotoma spiralis Smith and the South African P. fulioni Sow. ;
and the discovery of shells off the coasts of West Africa and South West Africa
apparently identical with the Italian Pliocene-Miocene Cancellaria (Sveltia)
lyrata, open up tantalizing zoogeographical questions.

Mention may be made of one feature alluded to by von Martens (1903,
p. 56), viz.: the pink coloration of several South African shells. Von Martens
gave examples showing that it is ‘characteristic’ of shells in different families.
Here again the beach-worn specimen has given a false impression. Gzbbula rosea
and Coralliophila rosacea may have a beautiful pink colour when picked up on the
beach, but they are not at all pink when alive. A pink coloration may be normal
in some species, but in others is obviously the result of weathering and chemical
action. In the South African Museum collection there is a Conus which is aurora
on one side and Javendulus on the other; a clear proof of the identity of the two
so-called ‘species’ !

Mauve or lavender is another noticeable tint occurring chiefly on the
protoconch and early whorls of some shells, e.g. Eugyrina pustulata and Demoulia
retusa.

TAXONOMY

The present work admittedly shows some conservatism in retaining some
of the old-established and well-known generic names in preference to adopting
a more modern taxonomy, in which a multiplicity of genera has been founded
often on characters which seem to be meagre or artificial. In the case of the
Cowries this splitting-up has been carried to such a fantastic extent that, of the
several hundreds of known species, only one remains in the original genus Cypraea.

It seems doubtful whether this increasingly complex classification is of any
real help to zoology, in particular to faunistic zoology.

Thiele (e.g. 1929, p. 368) objects to the use of the name of a fossil genus for
a recent genus, however much alike they may be conchologically. The anatomy
of the extinct animal can never be known. This very sound principle applies
equally to species. In the above-mentioned case of Cancellaria lyrata the present
writer considers it would be preferable to give the recent examples a separate
specific name.!* Zoogeographical conclusions based on the assumption that the
fossil and Recent examples are conspecific may be erroneous and are certainly
unscientific.

12 Perhaps a trinomial addition would meet the case, e.g. Cancellaria lyrata vivans.

78 ANNALS OF THE SOUTH AFRICAN MUSEUM

ACKNOWLEDGEMENTS

Acknowledgements and thanks are herewith expressed to the South African
Council for Scientific and Industrial Research for a Research Grant enabling
me to undertake this work; to Dr. A. W. Crompton, Director of the South
African Museum, for permitting me to continue my study of the South African
Museum collections; to Prof. J. H. Day, University of Cape Town, for submit-
ting for identification and comparison the large collection of molluscs preserved
in the Zoology Department under his charge; to the Professor of Zoology and
Mrs. Kalk, University of the Witwatersrand, for the opportunity of examining
their collection from Delagoa Bay; to Mr. A. E. Salisbury of the Conchological
Society of Great Britain, for comparing many specimens and giving me the
benefit of his knowledge and experience; and to my College friend H. Watson
of Cambridge, well known for his malacological researches.

Tribute must also be given to H. C. Burnup (1928) and J. R. le B. Tomlin
(t1954), with whom I discussed synonymies and other matters relating to
South African Molluscs for many years.

ABBREVIATIONS
P.F. . . The Cape Government trawler s.s. Pieter Faure.
U.C.T. . University of Cape Town Ecological Survey.
U.W. . . University of the Witwatersrand Zoology Department.

Specimens taken by the s.s. Africana, Fisheries Survey, Union of South
Africa, were submitted through Prof. Day of the University of Cape Town.

Fam. TEREBRIDAE

1928. Tomlin. Ann. S. Afr. Mus., xxv, p. 329 (‘records . . . so scanty that any
additional information is worth chronicling’).

Species recorded from South Africa, but whose occurrence in South African
waters has yet to be confirmed:

Hastula apicina (Desh.) Terebra macandrewi Smith
5, cuspidata (Hinds) » moms @. SoG,
Terebra archimedis (Desh.) », pertusa Born.
», babylonia Lam. 5, raphanula Lam.
» eingulifera Lam. ,, subulata Linn.

textilis Hinds
tiarella Desh. ? a Columbellid

» grayt Smith
» livida Rve.
Species inquirendae (only one specimen of each known):

Terebra filmerae Sow.
5, kowwensis Turton (v. infra)

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 79

For convenience the shells may be separated into two groups:

No spiral groove: Hastula apicitincta Sow., casta var. natalensis Smith, Terebra
diversa Smith, filmerae Sow., kowiensis Turton, lightfooti Smith, planecosta
n. sp., thielei n. sp.

With spiral groove: Dzplomeriza duplicata (Linn.), fictzlis (Hinds), Terebra affinis
Gray, babylonia Lam., capensis Smith, cingulifera Lam., circinata Desh.,
dimidiata Linn., evoluta Desh., laevigata Gray, loisae Smith, longiscata Desh..,
monilis Q. & G., myuros Lam., nebulosa Sow., spectabilis Hinds, straminea
Gray, subulata Linn., suspensa Smith.

Terebra circinata Desh.
Delagoa Bay. One animal examined: no radula.

Terebra myuros Lam.

Off O’Neil Peak (Zululand), 55 fathoms (S. Afr. Mus. P.F. Coll.); Natal
(S. Afr. Mus.).

The badly worn P.F. specimen is stouter than normal (less ‘rat-tail’-like).
The Natal specimen is slightly worn, but retains its colouring. The exact locality
is not recorded, but the P.F. specimen allows this species to be added to the
South African fauna-list.

S. Afr. Mus. also has specimens collected by E. L. Layard (on board
H.M.S. Castor, 1856) at Farquhar Island (10° S., 51° E.).

Terebra straminea Gray

Off O’Neil Peak (Zululand), 55 fathoms (S. Afr. Mus. P.F. Coll.). Two

somewhat worn specimens.
Terebra lightfoott Smith
Bie...t ic

Table Bay, 22 fathoms, 11 dead; twelve miles S.E. of Cape Point, 45
fathoms, 2 dead but fresh; Brown’s Bank (approx. 364° S., 21° E.), 80-100
fathoms, 1 dead. (S. Afr. Mus. P.F. Coll.) Living: False Bay, 52 metres
(ELC“T.).

Apex corroded in all specimens seen, but protoconch seemingly 2-whorled.
Terebra capensis Smith
Fie..1 ¢

Protoconch 2 whorls, alt. and diam. 0-5 mm., smooth. Very fine spiral
striae in the intervals between ribs, c. 10 on early whorls, becoming fewer and
irregularly spaced on later whorls.

Still Bay, dead shells common (S. Afr. Mus. Muir Coll.).

Off Cape St. Blaize, 37 fathoms, 1 dead (S. Afr. Mus. P.F. Coll.).

The record from Karachi (1901. Melvill and Standen. Proc. Zool. Soc. Lond.,
ll, p. 428) is probably a misidentification.

80 _ ANNALS OF THE SOUTH AFRICAN MUSEUM

Terebra suspensa Srnith

Protoconch 2-24 whorls, alt. and diam. 0:75 mm., smooth.

Pleistocene deposits at Sedgefield, Groenvlei (Knysna district). (S. Afr.
Mus.) See:A: R. H. Martin. S. Afr. 7: Se 52, ps 187, 1950:

Still Bay, dead shells (S. Afr. Mus. Muir Coll.).

Terebra diversa Smith

Protoconch 2 whorls, alt. and diam. 0-5 mm., smooth.

No records west of Pondoland (S. Afr. Mus.) except the single specimen
recorded by Bartsch (1915, p. 11) as collected by Stimpson in False Bay during
the U.S. North Pacific Exploring Expedition. Like other species collected on
this expedition the specimen bears a low register number in the U.S. Nat. Mus.
Catalogue (see Bartsch, passim), and its provenance can therefore be accepted.

Terebra affinis Gray

S. Afr. Mus. has 2 (one very worn) from Delagoa Bay. Also collected by
Layard (on board H.M.S. Castor, 1856) at Farquhar Island.

Terebra spectabilis Hinds

Two specimens in good condition but evidently not alive when dredged:
off Amatikulu River (Zululand), 24 fathoms; and off Cape Natal (Durban),
54 fathoms (S. Afr. Mus. P.F. Coll.).

The 32 mm. Zululand specimen was identified by Sowerby.

T. geminata Desh. 1859 was considered a synonym by Sowerby (1897).

Terebra longiscata Desh.
Off Cape Natal (Durban), 54 fathoms, 1 dead (S. Afr. Mus. P.F. Coll.).
Terebra casta Hinds

Typical form (with flat ribs): Nacala, north of Mozambique Island
(U.W.).

In var. natalensis the protoconch has 44-5 whorls, alt. 0-75 mm., diam.
ist whorl ¢. 0°15, 5th whorl o-5 mm., smooth, dark maroon or purplish-brown.

The variety seems to merit the epithet ‘tinted apex’ far more than Sowerby’s
aprcitincta.

No animal available for determining the correct genus.

Terebra dimidiata Linn.

Living: Delagoa Bay (U.W.). Animal (as preserved) yellowish-brown.

S. Afr. Mus. has it from Ferndo Veloso Bay, north of Mozambique Island
(from the collection of P. Ross Frames).

Terebra thielei n. sp.
Fig. 1d
? 1925. Thiele. D. Tiefsee Exp., xvii, p. 256, pl: 41 (29), fig. 17. (Terebra sp.,
part). !

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 81

Four specimens, 4-6-5 mm. in length, with 3 or 4 postnatal whorls; 11 axial
ribs on 1st whorl, increasing to 14-15 (16), the ribs are broadly rounded with
narrower intervals, contrasting with the narrowly rounded ribs and wide
intervals of capensis; no spiral groove or spiral striae.

33° 50’ S., 25° 48’ E. (depth?), and Algoa Bay, 155 metres (Thiele).

Algoa Bay, 67 fathoms, 4 dead specimens (S. Afr. Mus. A.8657, P.F. Coll.).

Fic. 1. a. Terebra planecosta n. sp. 5 mm. fragment above, one whorl of g mm. Type below.
b, c, d, e. Diagrammatic cross-sections of T. planecosta n. sp., lightfooti Smith, thielei n. sp., and
capensis Smith, respectively. f One pair of teeth from radula of Diplomeriza fictilis (Hinds).

One specimen 5°25 X 2°25 mm., one 5°25X2 mm., one fragment 4 mm.
long, one with 4 postnatal whorls 6-5 mm. long. All four appear to be con-
specific, although the two complete specimens differ in proportionate width.

Thiele had one specimen from both of the above localities, and also a third
from Gt. Fish Bay, Angola; but I am inclined to query the identity of the latter.

Terebra planecosta n. sp.
Piss a7d

Protoconch 2 whorls, smooth. Postnatal whorls 8, early whorls with
11 straight axial ribs, broad and flat, with narrow, shallow intervening grooves,
increasing to 12 or 13 on later whorls; no spiral groove or spiral striae.

Type, protoconch plus 6 whorls 9 mm., another specimen consisting of the
3rd to 8th whorls 12-5 mm.; fragment protoconch and 5 whorls 5 mm.

Glossy, pale buff, a faint darker band on upper half of whorls.

82 ANNALS OF THE SOUTH AFRICAN MUSEUM

Off Cove Rock (East London area), 44 miles, 22 fathoms, one dead
(Type, A8659); Cape St. Blaize, N. x E., 73 miles, 125 fathoms, two broken
specimens, A8658 (S. Afr. Mus. P.F. Coll.).

Terebra kowiensis Turton

1932. Turton. Mar. Sh. Pt. Alfred, p. 11, pl. 2, no. 92.

The figure shows an 18 mm. shell with 6 whorls; nearly straight, broad
axial ribs with narrow intervals, 7 on 2nd whorl, about Io or 11 on last whorl;
the photograph does not seem to have been retouched (as have some of Turton’s
photographs), so one may reckon there should be about 24 axial ribs on the
complete circumference of the last whorl.

Such as it is, the description does not correspond: 7 whorls, and about
40 ribs on the last whorl. The number 40 is probably a misprint. —

In having no spiral sculpture, Turton’s shell has some resemblance to
thielec and planecosta; to the former species in regard to the number of ribs, to
the latter in regard to their (apparent) flatness and narrow intervals. But the

kowwensis photograph shows a greater number of ribs, and an altogether larger
shell.

Diplomeriza fictilis (Hinds)

Higa iy:
1845. Hinds in Sowerby. Thes. Conch., i, p. 183, pl. 45, figs. 109, 110 ( Terebraf.).
1928. Tomlin. Ann. S. Afr. Mus., xxv, p. 329 (Terebra f.).

Protoconch 14 (?2) whorls, alt. 0-2, diam. 0-25 mm., smooth (all specimens.
slightly corroded).

Three of the shells mentioned by Tomlin have been examined and found
to possess a radula comprising 24 pairs of solid, slightly curved teeth. This
species must therefore be removed from Terebra s.s.

In contrast with Thiele’s (1929, Handbuch, i) figure 469 (after Troschel) of
a radula tooth of D. duplicata, the denticle or flange is on the anterior margin in

fictilis.
Diplomeriza duplicata (Linn.)

S. Afr. Mus. has examples obtained by P. Ross Frames from Fernao Veloso
Bay, north of Mozambique Island. There is every probability that the species
occurs farther south.

Fam. CONIDAE
1937. Tomlin. Proc. Malac. Soc., xxii, pp. 205-330 (list of Recent and fossil

species).

The following species have been recorded from South Africa (Cape and
Natal), but their presence as living components of the fauna has yet to be
confirmed: aplustre, arachnoideus, bandanus, capitaneus, consors, conspersus, crotchit
(Saldanha Bay!), eumitos Tomlin 1926 (? =panniculus), figulinus (Durban: S. Afr.

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 83

Mus.), geographus, giluus (Saldanha Bay!), glans, gubernator, guttatus, lamarckit,
legatus, lineatus, litteratus, namocanus, nimbosus, obscurus, pauperculus, plumbeus,
primula, punctatus, quercinus, tulipa.

The other species may be accepted as South African, subject to future
research on synonymy (e.g. elongatus-mozambicus and simplex).

Conus eucoronatus Sow. (3rd)

1903. Sowerby. Mar. Invest. S. Afr., ii, p. 217, pl. 3, fig. 9.
te3q-. Peile. Proc. Mal. Soc., xxiii, p. 354 (radula).

As the figure shows: the width ‘34’ is a misprint for 24 mm.

The second specimen recorded from Natal by Sowerby was a dead example.
A third example, also dead, 24 mm. long, was taken off Durnford Point (Zulu-
land) in 90 fathoms; and a fragment with the distinctive sculpture, from off
O’Neil Peak (Zululand), 90 fathoms. (S. Afr. Mus. P.F. Coll.)

The Type, now in S. Afr. Mus., was presumably a living specimen, the
animal of which was extracted by Sowerby and sent to Gwatkin, because Peile
records a Gwatkin slide of the radula (in Brit. Mus.). The radula teeth are
similar to those of aurora. There is, however, no certainty that this radula
belongs to eucoronatus.

Conus papillaris Ad. & Rve.

1848. Adams and Reeve. Zool. H.M.S. ‘Samarang’ Moll., p. 17, pl. 5, figs. 7a,

6 (coloured).

1873. Sowerby. Proc. Zool. Soc. Lond., p. 146, pl. 15, fig. 4 (coloured) (alti-
spiratus).

Shoulder. angular, with moderately numerous and prominent, slightly
oblique knobs, spire high, turreted, angle 70°-80°, whorls above shoulder with
axial lines of growth but no spiral striae. Periostracum thin, adherent.
49 X 23 mm.

Operculum narrow, ovate-cuneiform, margin entire, 5°5 1°75 mm. in
49 mm. shell.

Agulhas Bank (Sowerby, 2nd); False Bay, 18-23 fathoms (S. Afr. Mus.
Eb Goll.).

Although described in the Samarang Report, the type specimen of papillaris
was taken during Sir Edward Belcher’s previous voyage in H.M.S. Sulphur. No
note of its locality was attached to it. Nevertheless there is no improbability in
its having been dredged on the Agulhas Bank, together with several other
molluscs which were described by Hinds in the Sulphur Report. This is supported
by the almost exact resemblance of the largest S. Afr. Mus. specimen to Adams
& Reeve’s figure; it is slightly larger (if the figure of papillaris is natural size:
42 mm.), and not so strongly coronate, but otherwise might well have posed
for the original illustration.

Moreover, although Sowerby (grd) identified this one specimen as his
father’s altispiratus, he identified other P.F. specimens from the Agulhas Bank

84 ANNALS OF THE SOUTH AFRICAN MUSEUM

as papillaris, without however recording them in print. These latter, being non-
ccoronate, I regard as gradatulus (q.v.).

At first sight there is more resemblance between the figures of turritus Sow.
1870 and papillaris than between those of altispiratus and papillaris; and Sowerby —
(1870. Proc. Zool. Soc. Lond., p. 256) said that turritus was slightly suggestive of
papillaris but was non-coronate. Sowerby’s description of altispiratus does not
mention the shoulder knobs, but his figure shows them. (His comparison with

_franciscanus seems inexplicable because the latter is quite different in shape.)

Possibly this is a case of a species having coronate and non-coronate forms,
analogous to that suspected in levidus-flavidus. In C. semisulcatus Sow. the early
whorls are coronate, the later ones non-coronate. See Illustr. Zool. R.I.M.S.
‘Investigator’ Moll., pl. 14, figs. 1, 1a, 1907; also Smith’s remarks on C. aculeiformis
Rve. in Ann. Mag. Nat. Hist. (7), xiii, p. 455, 1904.

The largest specimen was taken alive, and retains the operculum; Sowerby
probably retained the animal. Another specimen, 40 (protoconch missing) Xx
18 mm., with shoulder and spire slightly worn and corroded, closely resembles
the figure of altispiratus. )

Five dead juveniles from Still Bay (S. Afr. Mus. Muir Coll.) are coronate
-and may perhaps be this species; but the early whorls have 2-3 spiral striae
above the shoulder, and there are no connecting stages.

Conus umperialis Lam.
1906. Smith. Ann. Natal Mus., i, p. 22, pl. 7, fig. 1 (quekettz). |

In spite of Smith saying ‘a very distinct species and not comparable with
any of the known forms’, I venture to suggest that queketti is synonymous with
imperialis.

S. Afr. Mus, has examples of zmperialis from Mozambique Island (Ross
Frames Coll.), and Diego Garcia.

Conus ebraeus Linn.
1939. Peile. Proc. Mal. Soc., xxiii, p. 352, fig. 16 (radula).

Living examples known from: Scottburgh (Natal) (S. Afr. Mus. Coll.
Burnup); Port St. Johns and Umgazana (Pondoland), Umtwalumi (Natal)
(U.C.T.); Delagoa Bay (U.W.).

Peile says the radula teeth have neither barb, blade, nor serrations.

Conus arenatus Brug.
1929. Thiele. Handbuch, 1, fig. 463 (radula).
1939. Peile. Proc. Mal. Soc., xxiii, p. 352, fig. 24 (radula).
Living: Durban (S. Afr. Mus. Coll. Burnup); Delagoa Bay (U.W.).
The radula tooth of the Delagoa Bay specimen conforms with Peile’s figure.
Conus ceylanensis Brug.
1939. Peile. Proc. Mal. Soc., xxiii, p. 352 (radula).

Living: Scottburgh (Natal) (S. Afr. Mus. Coll. Burnup); Delagoa Bay
(U.W.).

One of the Scottburgh specimens is subscalariform.

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 85.

Conus miliaris Brug.

1939. Peile. Proc. Mal. Soc., xxiii, p. 352 (radula).
Living: Scottburgh (Natal) (S. Afr. Mus. Coll. Burnup); Delagoa Bay
(U.W.). |
Conus minimus Linn.
Living: Isipingo, Natal (S. Afr. Mus. Coll. Burnup); Delagoa Bay (U.W.);
Durban Bay, and a dead but fresh specimen from Umgazana, south of Port
St. Johns (U.C.T.).

Conus zeylanicus Gmelin

1939. Peile. Proc. Mal. Soc., xxiii, p. 352, fig. 25 (radula).

Living: Durban (S. Afr. Mus. Coll. Burnup).

Conus lividus Brug.
Fig. 2 ¢
1939. Peile. Proc. Mal. Soc., xxiii, p. 352, fig. 15 (radula).

Living: Scottburgh (S. Afr. Mus. Coll. Burnup); Isipingo and Durban.
(U.C.T.); Delagoa Bay (K.H.B., also U.W.).

The radula of a Delagoa Bay specimen is not long-shafted, has a barb, a.
narrow blade, and serrations extending to nearly midway on the shaft. This
is in conflict with Peile’s description and figure of a Seychelles specimen.

Conus flavidus Lam.
1939. Peile. Proc. Mal. Soc., xxiii, p. 352 (radula).
1952. Braga. Anais fF. Invest. Ultramar., vii, 3, p. 70, pl. 1, fig. 8.

It has been suggested that this is the non-coronate form of lividus. The spiral
striae on the whorls of the spire do not seem to be nearly so well marked as in
lividus.

Conus textile Linn.
1939. Peile. Proc. Mal. Soc., xxiii, p. 350, fig. 9 (radula).

Living: Delagoa Bay (U.W.).

Conus betulinus Linn.
1939. Peile. Proc. Mal. Soc., xxiii, p. 352, fig. 21 (radula).
1952. Braga. Anais 7. Invest. Ultramar., vii, 3, p. 69, pl. 1, fig. 9.
Living: Delagoa Bay (U.W.).
Conus miles Linn.
1939. Peile. Proc. Mal. Soc., xxiii, p. 354, fig. 26 (radula).
S. Afr. Mus. P.F. Coll. has a specimen retaining its periostracum, without.
precise locality.
Conus catus Brug.

1939. Peile. Proc. Mal. Soc., xxiii, p. 349 (radula).
Living: Scottburgh (S. Afr. Mus. Coll. Burnup).

86 ANNALS OF THE SOUTH AFRICAN MUSEUM

d e f g h
Fic. 2. Radula teeth of Conus. a. infrenatus, b. aurora, c. elongatus, d. gilchristi, e. natalis, f. vexillum,

g. lividus, h. gradatulus, d, e. copies of drawings by H. Watson of teeth in his collection, mounted
by Gwatkin; the mount of gilchristi is labelled [received from] ‘Sowerby 1903’.

Conus ratius Brug.

1939. Peile. Proc. Mal. Soc., xxiii, p. 351 (radula).
Living: Scottburgh (S. Afr. Mus. Coll. Burnup).

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 87

Conus vexillum Gmelin

Fig. 2 f
1939. Peile. Proc. Mal. Soc., xxiii, p. 350, fig. 13 (radula).

Living: Off Cape St. Lucia (Zululand), 27 metres. 2 specimens (s.s.
Africana, 15/5/48).

Shell thick, angle of spire 110°. Yellowish or pinkish, pale band around
middle, darker band above and below; 2-3 narrow dark bands near shoulder,
spire mottled; periostracum pale brown. Operculum narrow, margin entire.
56 x 28 mm.

Radula tooth with slender elongate shaft, small barb, bevelled blade, and
serrations extending 4/5 length of shaft, ending in an obscure spur. |

These specimens seem referable to this species as they are narrower than
typical namocanus, and the brown spiral lines characteristic of the latter are
absent.

Conus tessulatus Born

1939. Peile. Proc. Mal. Soc., xxiii, p. 352, fig. 23 (radula).
Living: Delagoa Bay (U.W.). |
The radula tooth of this specimen agrees with Peile’s figure.
Conus piperatus Dillwyn

1939. Peile. Proc. Mal. Soc., xxiii, p. 352, fig. 20 (radula).

Living: Scottburgh (S. Afr. Mus. Coll. Burnup).

Conus infrenatus Rve.
Fis: 204
1848. Reeve. Conch. Icon. Suppl., pl. 3, sp. 285.
1889. Sowerby. 7. Conch., vi, p. 9, pl. 1, fig. 12 (bairstowt).

The S. Afr. Mus. series shows transition from the typical infrenatus, with
numerous spots, the more prominent of which are arranged in at least 10 spiral
series, 2 series in the middle usually enclosing a darker brown band; to bairstowi
with pale unspotted ground-colour and only 6 well-spaced series of brown spots.

Living: Algoa Bay (U.C.T. infrenatus; and S. Afr. Mus. P.F. Coll:
bairstow?).

A radula of a typical infrenatus has 15 pairs of teeth, barb very fine, blade
very narrow and inconspicuous, serrations extending not quite to midway on
shaft (3/7), ending in a small spur.

Conus gilchristi Sow. (3rd)
Plate II and fig. 2 d

1939. Peile. Proc. Mal. Soc., xxiii, p. 348 (radula).

Spire low, angle 130°. 52x27 mm.

The Type and only known specimen, now in S. Afr. Mus., was taken alive.
The animal was presumably passed to Gwatkin, and, again presumably, it was
a Gwatkin slide which Peile examined.

‘

88 ANNALS OF THE SOUTH AFRICAN MUSEUM

There exists another mounted tooth presented by Gwatkin to my friend
H. Watson, Cambridge (in litt. 5/4/57). Shaft slender elongate (Watson:
3°65 mm.), barb small, blade short and bevelled, minute serrations extending
at least to midway along shaft.

Conus natalis Sow. (2nd)

Plate II and fig. 2 e
Laps. cal. natalensis Sow. (3rd).

The shape varies: slender forms, somewhat elliptical in outline, with
flatly rounded shoulder and rather high spire, angle go° (fresh) to 105° or
110° (worn) ; plump forms, with more prominent shoulder and low spire,
angle 120°-130° (or 135° worn).

Radula of a Port Shepstone specimen collected by Burnup, in coll. Watson,
Cambridge: shaft 3:65 mm. long, similar to that of gilchristi (supra).

The most south-westerly locality for beach-worn specimens seems to be the
Kowie (Port Alfred), whence S. Afr. Mus. has one example with the gilchristz
pattern.

Placed side by side, a typical natalis would appear to be a species quite
distinct from gilchristt. But the S. Afr. Mus. series indicates that they are only
forms of one variable species. The accompanying photograph (Plate II) of
a few specimens selected from the series demonstrates the variation in form
(allowance being made for some examples which are beach-worn) and colour-
pattern. The third shell from the right in the lower row has the gilchristi shape
with the natalis pattern. The radula teeth are the same in the two forms.

The Pieter Faure obtained no specimens, dead or alive, of natalis. This may
indicate that the natalis form is found in the littoral zone or very shallow water,
while gilchristz is a deeper water form. It must, however, be borne in mind that
trawling close inshore on the Natal coast is difficult; except at two or three
places, e.g. Tugela River mouth, the Pieter Faure did scarcely any trawling
within the 25-fathom contour.

A transition in colour pattern from natalis to infrenatus is almost possible,
except that the latter is ‘dotted’ and never has the slightest trace of triangular or
arrow-head markings. Moreover the radulae are quite different.

Burnup, with whom I discussed (1914) the matter, agreed that on the
available evidence gilchristi should be regarded as a synonym of natalis. For the
time being, however, I retain them here as two species.

Conus pictus Rve.

1843. Reeve. Conch. Icon., i, pl. 18, sp. 98.
1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 14.
1937. Tomlin, ‘Proc, Mal. Soc., xxi, ps 20%,

Pale buff or rosaceous, sometimes with faint darker spots, two paler bands,
varying in width, articulated with darker spots or angular axial streaks, a third
similar band at the anterior end, spire with irregular flames. 36 x20 mm.

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 89

Six specimens received from the Albany Museum, Grahamstown, probably
collected by Dr. H. Becker at the Kowie (Port Alfred), and labelled pictus Rve.
This identification has been confirmed by A. E. Salisbury (1956). The shells
are very like beckert Sow. (1911), which both Tomlin (1937, loc. cit.) and Salis-
bury consider as a variety of aurora.

Bartsch recorded it (presumably dead) from Port Alfred and Gt. Fish
River mouth.

A dead but fresh specimen was dredged in Algoa Bay (U.C.T.).

Conus scitulus Rve.

1848. Kiener. Cog. viv., p. 218, pl. 55, fig. 4 (taspideus, non Gmelin).

1848. Krauss. Sidafr. Moll., p. 131 (jaspideus Kien. non Gmelin).

1849. Reeve. Conch. Icon. Suppl., pl. 9, sp. 283.

1858. Crosse. Mag. Zool. (2), x, p. 122 (danieli, nom. nov. for jaspideus Kien.).
1903. Smith. Proc. Mal. Soc., v, p. 362.

1937. Tomlin. ibid., xxii, pp. 237, 263, 306.

Angle of spire go°. A continuous reddish or orange-brown band below the
shoulder, the rest of the whorl mottled orange on white ground, the mottling
denser on base, spire with brown marks between white patches which latter are
more or less lobate, descending to or very slightly over the shoulder (but not
nearly so far as in simplex). 23 X12 mm.

Algoa Bay (Kiener); Cape (Krauss, Smith); Hermanus (S. Afr. Mus.).

The worn Hermanus specimens were identified by J. H. Ponsonby, and
recently (1956) confirmed by A. E. Salisbury.

The white patches around the shoulder give the species a strong resemblance
to simplex.

Conus gradatulus Weink.

Fig. 2h

1870.. Sowerby. Proc. Zool. Soc. Lond., p. 256, pl. 22, fig. 14 (coloured) (turritus,
non Lam., fossil).

1875. Weinkauff in Mart. Chemn., p. 356, pl. 66, fig. 15 (copy).

1903. Von Martens. D. Tiefsee Exp., vii, p. 22.

Spire high, turreted, angle 70°-go°, shoulder angular, non-coronate;
operculum narrow oval, margin entire; periostracum thin, fimbriate around
the shoulders, adherent. Up to 54x24 mm.

Creamy-white, with faint indications of 3 fulvous bands, one below shoulder,
one in middle and one between latter and base, aperture within faintly pink,
periostracum pale buff or yellowish.

Radula tooth rather slender in distal two-thirds, barb small, blade very
narrow, serrations extending about one-third length of shaft, no spur.

Agulhas Bank (Sowerby, von Martens).

go ) ANNALS OF THE SOUTH AFRICAN MUSEUM

Living: on Agulhas Bank from approx 27° E. to around Cape Point and
off west coast of Cape Peninsula as far north as approx. latitude of Cape Town,
44-256 fathoms (S. Afr. Mus. P.F. Coll.).

Recorded also from Gt. Fish Bay, Angola, by von Martens.
Conus patens Sow. (3rd)

1939. Peile. Proc. Mal. Soc., xxiii, p. 351 (radula).

Spire rather low, angle 105° (Type) to 110° (another specimen). Shoulder
rounded-angular; operculum oval, margin entire; periostracum rather thick
and rough.

According to Peile the radula teeth are similar to those of vextllum. Pre-
sumably the radula he examined was in the Gwatkin collection in the Brit.
Mus. and derived from the animal of the unique Type shell.

Type in S. Afr. Mus. In the P.F. collection there is a smaller specimen,
36 mm. long, from off west coast of Cape Peninsula, 156 fathoms, identified by
Sowerby. With this identification I agree. But Sowerby also identified several
other specimens as patens, which I refer to gradatulus on account of their high
spires. A curious feature confirms this: the thickish periostracum of both the
Type and the smaller specimen of patens flakes off in patches, whereas the thin
periostracum of gradatulus (and also papillaris) adheres closely to the shell.

Conus simplex Sow. (2nd)

1843. Reeve. Conch. Icon., pl. 5, sp. 24. (tnformis, non Brug.).
1857/8. Sowerby. Thes. Conch., iii, p. 31, pl. 9 (195), fig. 199.
1903. Smith. Proc. Mal. Soc., v, p. 362 (quotes “Thes. 222’).
1937. Tomlin. ibid., xxii, p. 308 (quotes “Thes. 72).
Spire high, angle 90°. Operculum oval, margin entire. Up to 54 x 24 mm.
Brown with a series of white irregular spots around middle and a series of
crescentic or zigzag white marks in basal half, shoulder with white lobate blotches;
or the white predominating as ground-colour, the brown forming zigzag axial
streaks or flames. Animal dark grey.
Radula with 18-19 pairs of teeth which are similar to those of elongatus
(q-y.)).
Living: False Bay, o-11 fathoms (S. Afr. Mus. P.F. Coll., also U.C.T.).
Recorded from East Indies by Sowerby. S. Afr. Mus. has two specimens
from Mauritius.
The conspicuous white lobate patches around the shoulder are a distinctive
feature, though somewhat similar but smaller patches occur in seztulus.

Conus elongatus Chemn.
Fig. 2.¢

? Chemnitz. Conch., x, pl. 144A, figs. i & k (quoted from Lamarck, 1810).

1798. Lamarck. Tabl. Encycl., pl. 337, figs. 1, 2 (mozambicus Brug.).
1810. id. Ann. Mus. Paris, xv, p. 281 (mozambicus Brug.).

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA gti

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 13, pl. 1, fig. 12 (alfredensis).
1932. Turton. Mar. Sh. Pt. Alfred, p. 12, pl. 3, no. 97 (juv. subscalariform).
1937- Tomlin. Proc. Mal. Soc., xxii, pp. 210, 244, 278.

Protoconch 14-2 whorls, diam. 1-5 mm. In young shells with 1, 2 or 3
(sometimes 4) whorls spiral striae cover the whole whorl, but in older shells are
confined to the base. Elongate, shoulder flatly rounded, spire high, angle c. 85°;
operculum oval, margin entire. Up to 70x31 mm. (S. Afr. Mus. worn),
68 x 33 mm. (U.C.T. living).

Variable brown mottling on white ground colour, sometimes nearly con-
tinuous around body whorl, sometimes broken up into axial zigzags and flames,
often a predominantly white spiral band around middle, sometimes spiral series
of small brown rectangular spots; shoulder irregularly brown and white;
inside of aperture faintly violaceous; periostracum yellow-brown, in fresh
specimens almost concealing the shell pattern. Animal (as preserved) blackish.

Radula with 14-15 pairs of teeth, with barb, narrow blade, serrations
extending to about midway along shaft, ending in a spur; the proximal part of
the serrated ridge is bordered with minute scabrosities.

Living (or in fresh condition): Still Bay, Danger Point, Hermanus, False
Bay, west coast of Cape Peninsula, Dassen Island, Langebaan (Saldanha Bay),
Hondeklip Bay, Port Nolloth (S. Afr. Mus. and U.C.T.).

Recorded (dead) from Port Alfred and Port Elizabeth.

The Pieter Faure obtained the large 70 mm. worn specimen from 130 fathoms
off the west coast of the Cape Peninsula.

C. alfredensis Bartsch. seems to me synonymous. Tomlin (1937, p. 210)
makes it a synonym of aurora; but perhaps the words ‘=aurora’ were intended
to be inserted in the next entry, namely algoensis. The shape of alfredensis is more
like that of elongatus than of aurora.

S. Afr. Mus. has a subscalariform example 40 x17 mm. from Hermanus.
cf. Turton’s juv. 12 <5°5 mm. from Port Alfred.

Tomlin in his 1937 list of Cones does not mention an elongatus of Chemnitz.

The identity of mozambicus and elongatus, or its occurrence on the east coast
of Africa, has not yet been confirmed, so far as I am aware.

Conus aurora Lam.
Fig. 2}

1939. Peile. Proc. Mal. Soc., xxii, p. 352, fig. 19 (radula).

This common species has numerous synonyms: algoensis, tinianus, caffer,
loveni, secutor, fulvus, beckeri, lavendulus, krausst.

Shoulder rounded, spire moderate, angle 105°-115°. Up to 65 X33 mm.

Colour and pattern very variable, ranging from almost uniform rose
(aurora) and fulvous or chocolate (algoensis, fuluus) to the variously speckled,
lined, and banded forms (loveni, lavendulus). The rose, lavender, or mauve
coloration seems to be due, at least to some extent, to beach wear and weathering.

Q2 ANNALS OF THE SOUTH AFRICAN MUSEUM

One specimen in 8S. Afr. Mus. has mauve ground colour on one side (lavendulus)
and orange-pink on the other (aurora).

Radula with 14 pairs of teeth, o-g-1 mm. long in a 35 mm. shell, with
barb, narrow but rather long blade bevelled off, serrations extending to midway
along shaft, ending in a small spur. Confirmed by a radula in coll. H. Watson,
Cambridge (in litt. 5 Apr. 1957) with teeth 0-65 mm. long. Peile describes a
tooth 0-5 mm. long with the serrations not extending beyond the blade.

Living: Richmond (Alexandria Division), and Still Bay (U.C.T.); Peile’s
and Watson’s specimens of radulae from Algoa Bay examples.

Records of dead examples: from Tongaat (30 miles north of Durban) to
Still Bay (S. Afr. Mus.).

Bartsch records an algoensis collected by the U.S. North Pacific Exploring
Expedition in False Bay; this is probably an error because S. Afr. Mus. has no
records of localities west of Still Bay, and U.C.T. has not found this species in
False Bay.

The Pieter Faure obtained no examples in the course of all her trawling in
Algoa Bay and on the Agulhas Bank.

There are plump and slender forms (cf. lovent and caffer). Sowerby figured
a subscalariform beckert, and Turton’s No. 97 may be either aurora or elongatus.
S. Afr. Mus. has a subscalariform example of the ornate variety mentioned
below. This slipping of the body whorl away from the preceding whorl is.
particularly well shown in one freak specimen, 54 X 33 mm., which in addition
has two bulbous expansions on the shoulder, the last one causing a wide bulge
on the outer lip.

S. Afr. Mus. has several examples of a particularly ornate colour variety,
unfortunately without precise locality: red-brown, nearly uniform or mottled
to a varying extent with white, either irregular patches or axial zigzags, always
a white band a little below the middle, with dark red-brown patches or zig-
zags, base with brown and white patches; sometimes faint spiral interrupted
dark lines above and below the middle band; the latter is constant, though the ©
relative amounts of brown and white vary.

Fam. TURRITIDAE

1929. Thiele. Handbuch, i, pp. 357-72 (Conidae part).

Thiele’s Conidae embraced three subfamilies of ‘Pleurotomids’ and a fourth:
the true Cones. The distinctive facies of the shell of the latter seems to require
them to be kept in a separate family.

The family Turritidae is well represented in the South African region,
especially by ‘species’ founded on beach-worn shells. The animals of most of
them are unknown, and the species have been assigned to various genera on
account of the similarity of the shells to typical representatives of these genera.
Examination of radulae, however, has shown that some species, e.g. Clavatula

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 93

tumida, Pleurotoma (Drillia) scitecostata and fultoni, have been assigned to entirely
wrong genera (or even subfamilies).

For these reasons in the present work a fourth category has been added to
Thiele’s three subfamilies: Species incertae sedis. In most cases, maybe, concholo-
gists have been guessing correctly in which genus to put a species, but until the
animals have been examined we do not know the correct genus.

This fourth category is merely a temporary dumping-ground for ‘concho-
logical species’. It contains the great majority of the South African species, and
indicates the pressing need for collecting living examples, especially of the
littoral and inshore species. In some cases a species occurring in South Africa
but well known from another region (e.g. Turris cingulifera) has been assumed
to have been assigned to its correct genus in order to facilitate comparison with
the shells of closely allied species.

The lateral plate of the radula of Clavatula and Turris seems to need
investigation.

Thiele (1903. D. Tiefsee Exp., vii, p. 173, pl. 9(4), fig. 7. Pleurotoma
_(=Turris)) describes the lateral plate as ‘slipper’-shaped, i.e. resembling a
picture of a slipper in one plane. He does not suggest the plate is actually slipper-
like, i.e. a hollow cone, sharply pointed at one end, obliquely truncate at the
other end. In C. sinuata the plate is a solid 3-sided pyramid with the basal half
obliquely sheared off. Associated with the plate is a thin oval lamella (‘wing’:
Thiele). Is this lamella attached to the plate at the distal end of the bevelled
portion? Is it an actual appendage (‘Anhang’: Thiele, 1925. D. Tiefsee Exp.,
Xvil, p. 206; and 1929. Handbuch, i, p. 357)?

Thiele’s figures 13 and 14 (1925. loc. cit.) show it as such; but figure 13
(stnuata) is certainly incorrectly drawn. In arranging a radula on a slide, this
lamella can easily be displaced, and is then seen to have no point of attachment,
being rounded at both ends, sometimes more narrowly rounded or subacute at
the inner or median end.

If the lamella is not attached, the further question arises: does it lie inside
the bevelled portion of the plate or outside? By manipulation of the radula
during mounting, I am inclined to adopt the latter view. If this is correct,
cannot this lamella be regarded as a degenerate 2nd lateral or a marginal plate?
Or, following Cooke (1895. Cambr. Nat. Hist., iii, pp. 218, 219) the lamella is
the degenerate lateral, the conical plate the marginal or uncinus (in either case,
formula 1.1.1.1.1).

In indica, gilchristi, and lobata I have not succeeded in separating an acces-
sory (or concomitant) lamella from a ‘lateral’ plate; in fact I doubt whether
there is any accessory lamella.

The distal half of the plate is conical, possibly triquetral at the apex; the
proximal half is a flat ‘handle’, which is about half the thickness of the conical
part in side view, but as broad as the latter in face view. In side view an acces-
sory lamella often appears to be present, but in face view this appearance is
seen to be due to the sides of the ‘handle’ curving upwards (and slightly inwards).

Q4 ANNALS OF THE SOUTH AFRICAN MUSEUM

Thus the proximal half of the plate appears to be a hollow half-cylinder, the
distal half a solid cone.
The forked appearance seen in some preparations in more or less face view
(fig. 3 g, hand cf. Thiele, 1925. figs. 17, 19) is due to the transmitted light having
to penetrate a greater thickness of ‘chitin’, the flanges being seen in edge-view.
The above suggestions should be investigated by section-cutting and more
refined methods of preparation than are available to me.

Subfam. TurrinAE Thiele (? emend: TuRRITINAE)

Gen. DRILLIA Gray

Of the numerous South African species hitherto assigned to this genus I
have been able to examine the animals of only four: stolida, grayi, scitecostata,
fultom. As a result one goes into a different genus and three into a different
subfamily.

Drillia falsa n. sp.
Figs. 3 a, 4a

Spire subtending an angle of c. 30°. Aperture 14 in spire. Protoconch
2 whorls, diam. 1, alt. 0-75 mm., smooth. Postnatal whorls 5. Axial ribs 10
on Ist whorl, 9 on each of the others, oblique, protractive, narrower than inter-
vals, not crossing sulcus, and petering out on base of last whorl. Growth-lines
well marked. No spiral striae on whorls and only 3-4 very faint striae near
extremity of base. Sulcus wide; lip sinus deep. Canal short. 9°5 x 3:75 mm.
Operculum oval, nucleus apical. Pale fawn, operculum amber.

Radula with 34 rows, central plate narrow, lateral (intermediate) with
10-11 denticles, marginal with a small rounded lobe on postero-external
corner.

False Bay (Tra. 139. s.s. Africana per U.C.T.).

Remarks. Thiele’s species ancilla was 4°75 mm. long, with 4 whorls, diam.
protoconch 0-55 mm. (in figure), and, judging by the figure, the lip sinus was
already developed; the present specimen agrees with the characters given by
Thiele, except the ribs do not cross the sulcus; but it is a larger species with
distinctly larger protoconch.

Resembles hottentota in having no spiral sculpture, but seems to be a smaller
species with more prominent shoulder and consequently more convex profile,
and fewer axial ribs.

Clavatula taxus Chemn.
Figs. 3 b, 4.6
1923. Odhner. Med. Goteb. Mus., xxiii, p. 7 (taxus, and bimarginata non Lam.).
1926. Tomlin. Ann. Natal Mus., v, p. 289 (bimarginata, apud Odhner).

1932. Turton. Mar. Sh. Pt. Alfred, p. 18, and var. affinis, p. 18, pl. 3, no. 142.
1932. id. ibid., p. 18, pl. 4, no. 144 (rufanensis).

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 95

3 . =~
ZW

Fic. 3. Radula teeth of: a. Drillia falsa n. sp., central, lateral and marginal. 6. Clavatula taxus

Chemn. central and lateral. ¢. C. gravis (Hinds), three centrals and laterals, showing ‘accessory

lamellae’ displaced while arranging radula on slide for mounting. d. C. sinuata Born., central

and lateral, face and edge views of latter. e. Turris stolida (Hinds), lateral. f. 7. saldanhae n.

sp., lateral. g. T. indica Bolten, face and edge views. h. T. gilchristi (Sow.), face and edge views.
it. T. lobata (Sow.), from lobate East London specimen. j. from Cape Point specimen.

Protoconch 2 whorls, diam. 1:3 mm., smooth. Postnatal whorls 1o.
Oblique ridges above suture usually distinct in early whorls, obsolete in later
whorls. Fine spiral striae. Cingulum usually prominently tumid, sometimes,
chiefly in early whorls, feebly nodulose, and demarcated by a slight groove from
the smooth area of sulcus, which is narrower than cingulum. Lip sinus narrow
and very deep. 87 (without protoconch) x 29mm. Up to 100 mm. (Turton).

Operculum oval, nucleus nearly at middle of inner margin, which is
thickened or duplicated on outer surface; 16 X5°5 mm. in a shell 48 mm. long.

Periostracum yellowish-brown, dark chestnut-brown, or blackish brown.
Animal (as preserved) dull reddish-brown.

Radula with 42-50 rows, central plate small, narrow, with median cusp,
lateral plate with wing-like appendage.

Off Cape Barracouta, 40 fathoms (Odhner); St. Sebastian Bay, 40 fathoms
(coll. K.H.B.); off East London, and Algoa Bay to off Cape St. Blaize, 19-52
fathoms (S. Afr. Mus. P.F. Coll.); Simon’s Bay (dredged) (S. Afr. Mus.) ;
False Bay (U.C.T.).

Remarks. Odhner’s ‘bimarginata’ was 25 mm. long. The Simon’s Bay
example (identified as bimarginata by J. H. Ponsonby) is 32 mm. long, and

96 ANNALS OF THE SOUTH AFRICAN MUSEUM

obviously a young taxus. There are two smaller examples, 21 and 15 mm.,
which show the oblique nodules and the spiral striae very clearly.

Juveniles of taxus and gravis may be liable to confusion; taxus has a slightly
larger protoconch.

I have seen one example of the West African bimarginata Lam. (35 mm.,
from Goree). It has the whole base of the body whorl below shoulder with
strong spiral lirae which completely obliterate the oblique axial ribs, reducing
these to knobs on the shoulder; and the curve of the growth-lines on the sulcus
is unsymmetrical, almost horizontally retractive below the cingulum and
obliquely protractive above the shoulder knobs.

Melvill (1917. Proc. Mal. Soc., xii, p. 165), quoting Reeve, says that the
salmon-pink hue of this species (bimarginata) is unique in the genus. Presumably
that applies to the cleaned shell, because the periostracum in the above specimen
is dark chestnut or umber-brown.

There are broad and narrow individuals, e.g. 45 X17 mm., 44% 15 mm.
and 48 x 18 mm. (the latter two taken in the same haul) (width across shoulder,
excl. aperture).

I consider rufanensis ‘Turton a young worn example of taxus.

Clavatula impages (Ad. & Rve.)

1848. Adams and Reeve. “Samarang’, p. 39, pl. 9, fig. 1 (Pleurotoma 1.).

1903. Von Martens. D. Tiefsee Exp., vii, p. 23 (Clionella impages, non Ad. &
Rve.).

1932. Turton. Mar. Sh. Pt. Alfred, p. 18 (impages ?).

First recorded in South Africa by von Martens from 35° 16’ S., 22° 26’ E.,
155 metres, and from Port Elizabeth (dead).

One of these specimens was stated by E. A. Smith to be similar to the
type of impages. Von Martens doubted the locality ‘China Sea’ given by Adams
and Reeve, because of the likeness to taxus, and the restriction of the genus
Clionella to South Africa (the latter not at all a conclusive reason!).

Turton said he saw specimens at the British Museum labelled taxus var. _

Both von Martens and Turton record their specimens as being narrower
than taxus. The S. Afr. Mus. series, short as it is, shows that this is not a specific
character. Moreover the original figure of impages shows none of the features
characteristic of taxus.

I propose therefore to delete impages Lam. from the South African fauna-
list, and to add the recorded localities to those of taxus.

Clavatula gravis (Hinds)

PISS. (90, 4.6
1903. Sowerby. Mar. Invest. S. Afr., ii, p. 229.
1903. Von Martens. D. Tiefsee Exp., vii, p. 23.
1925. Thiele. ibid., xvii, p. 213, pl. 35(23), fig. 12.

[Not Turton. 1932. p. 19, pl. 4, no. 146=very worn tumida.]

97 : CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA

OQ’)

Fic. 4. a. Drillia falsa n. sp. 6. Clavatula taxus Chemn. c. C. gravis (Hinds). d. tripartita
Weink. e. ‘Clavatula’ tumida (Sow.). b-e. semidiagrammatic to show shape of lip sinus.

Protoconch 2 whorls, diam. 1 mm., smooth. Postnatal whorls 8-9. Fine
spiral striae on all whorls, somewhat variable in strength. Nodules above
shoulder appear subcircular to naked eye but slightly oblique under a lens;
suture runs close below these nodules, sometimes encroaching on them, but
never completely absorbing or concealing them (cf. Thiele’s figure). Tumid
cingulum below suture may be more prominent than the nodules above, and
may itself be slightly nodulose, but is never demarcated by a groove from the
smooth area of the sulcus, which is broader than the cingulum; lip sinus broad
and rather deep. On last whorl the nodules extend basally as oblique ridges,
sometimes with an additional ridge or a series of feeble nodules below the
shoulder (cf. Thiele’s figure). Up to 45 mm. long.

Operculum oval, nucleus at middle of inner margin, which is thickened ;
5 X 2-75 mm. in shell 23 mm. long.

Cream under the pale buff periostracum, with faint indications of slightly
darker marks, end of rostrum pinkish-brown; operculum amber.

Radula with 50 rows (23 mm. shell), central plate small, narrow, acicular,
lateral with wing-like appendage.

Cape Agulhas, 43 fathoms (Hinds); Agulhas Bank, 47 and 17 fathoms
(Sowerby); St. Francis Bay, 80-100 metres (von Martens); off Durnford Point
(Zululand), 13 fathoms; off Umhloti and Umkomaas Rivers (Natal), 27
fathoms and 40 fathoms; off East London, 45 fathoms, and Algoa Bay to False
Bay, 10-47 fathoms (S. Afr. Mus. P.F. Coll., all dead but fresh). 35° S.,
20° 49’ E., g1 metres, 1 dead, 1 living (s.s. Africana). False Bay and Mossel Bay,
living (U.C.T.).

Reported from the estuary of the Congo (Dautzenberg. 1912. Ann. Inst.
Océan., 5, fasc. 3, p. 10). The identification should be checked.

98 ie ANNALS OF THE SOUTH AFRICAN MUSEUM

Remarks. The ‘bead-necklace’ appearance, caused by the nodules above
and the cingulum below the suture, is distinctive. The absence of a groove
between cingulum and sulcus distinguishes this species, especially in the case of
juveniles, from taxus.

Clavatula (Melatoma) sinuata Born
Figs. 3 d, 5a

1877. Smith. Ann. Mag. Nat. Hist. (4), xix, p. 499 (borni).

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 17, pl. 2, fig. 2 (turtont).
1915. id. ibid., p. 18 (senuata and bornit).

1925. Thiele. D. Tiefsee Exp., xvii, p. 204, fig. 13 (radula).

1932. Vurton.’ Mor Sh. Pi Alfred). 17, ple 35 no. 17

Length of aperture 14-2 times in spire. Protoconch ? Postnatal whorls 8
(?9). Oblique ribs 11-12 on early whorls, increasing to 15-16 on later whorls,
sometimes 18 on body whorl of large specimens, extending to or almost to canal.
Suture undulate, embracing lower ends of ribs; a slightly nodulose cingulum
below suture; sulcus shallow; outer lip with narrow and rather deep sinus,
which at the formation of each successive rib produces a nodule slightly more
prominent than the rib below it. Growth-lines across sulcus only slightly oblique to
the suture. Spiral striae (when present) numerous, inconspicuous, best seen on
early whorls. Up to 60 mm. long.

Operculum oval, nucleus in middle of inner margin, which is duplicated
on outer surface; 84 mm. in shell 25 mm., 10 x5 in shell 40 mm.

Brown, with dark brown or blackish periostracum; or pale brown with
yellowish-brown periostracum (bornii, turtoni).

Radula with 50-60 rows, central plate small, squarish, with median cusp,
lateral rather slender, with wing-like appendage.

Port Alfred, Hermanus, False Bay, west coast of Cape Peninsula, Saldanha
Bay, and northwards to Buffelsrivier (S. Afr. Mus. and U.C.T.). Recorded
also from Natal by Krauss.

Littoral and shallow inshore waters. Not obtained in any of the Preter Faure
dredgings.

The pale form occurs together with the typical dark form at Port Alfred,
but at the Cape only the latter.

Remarks. ‘The lip sinus, not being in the concavity of the sulcus but forming
a series of raised nodules corresponding with the ribs, is distinctive; in slightly
worn examples these abraded nodules form a series of white beads separated by
grooves in which the dark periostracum persists. ,

The apex and early whorls are much subject to corrosion, so that the normal
number of whorls is uncertain. I have seen no specimen with its protoconch.

There are plump and slender individuals, e.g. 35 X14, 43X14, 45 X12;
and 50 X15 mm.

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 99°

One specimen, 25 mm., from Saldanha Bay (U.C.T.), has ribs on 54 whorls,
then after an injury, continues with extremely feeble ribs which become obsolete
on the body whorl; the sinus, however, continues to form the characteristic keel.

Other specimens become ribless from the 3rd or 4th whorl onwards (cf.
sigillata).

var. sigillata Rve.
Fig. 5 6

Length of aperture subequal to, or a little shorter than spire (but apex
broken in all specimens). Protoconch? Postnatal whorls 6-64. Whorls smooth,
faint indications of ribs on the upper 4 or 5 whorls, and even less defined on
earlier part of 6th whorl; a feeble spiral lira immediately below suture, and
another marking the lower boundary of sulcus and middle of the lip sinus, thus.
corresponding with the series of nodules in typical sinuata; below this 2 or 3
additional very feeble lirae on body whorl, best seen where they have become
white from abrasion. Sulcus very shallowly concave; lip sinus moderately wide,
shallow, growth-lines very slightly curved. 19 (apex broken <x 9g mm. (S. Afr.
Mus.); 20 (protoconch missing) x 8 mm. (U.C.T.).

Fic. 5. a. Clavatula sinuata Born. 6. var. sigillata Rve., worn, especially the apical whorls..
c. C. confusa Smith. d. C. kraussi (Smith). e. C. subventricosa (Smith). f. C. semicostata Kiener..

100 ANNALS OF THE SOUTH AFRICAN MUSEUM

Operculum as in sinuata, 5 X 2°75 mm. in shell 16 mm. long.

Greyish-brown, mottled and streaked with white due mostly to abrasion;
operculum horny; periostracum yellowish.

Radula as in sinuata.

West and east coasts of Cape Peninsula (S. Afr. Mus. and U.C.T.).

Remarks. Seems to be merely a smooth and less turriform variety of sinuata.
Clavatula krausst (Smith)
Fig. 5 d

1877. Smith. Ann. Mag. Nat. Hist. (4), xix, p. 500 (Pleurotoma (Clionella) k.).
1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 14 (kraussiz).
1932. Turton. Mar. Sh. Pt. Alfred, p. 15.

Length of aperture about 14 times in spire. Protoconch ? Postnatal whorls
7-8 (?9). Oblique ribs 11-12 on early whorls, 12-13 on later whorls, starting
from suture above and on body whorl extending halfway across base; shoulder
slightly below middle of whorl; suture not, or only very slightly embracing lower
ends of ribs; no nodular or only a very feeble cingulum below suture; sulcus.
not strongly concave. Fine spiral striae close together on sulcus, farther apart
on rest of whorl, about 4 on early whorls, 6 on later whorls, usually only visible
between the ribs. Outer lip with deep and narrow sinus, a little distance from
suture. 33 (without protoconch) x 12mm. Turton: up to 42 mm.

Operculum oval, nucleus at middle of inner margin, which is thickened.

White with purplish-brown dots and zigzag lines following the growth-
lines, especially noticeable on the sulcus, inner surface of outer lip white, oper-
culum horny, periostracum pale yellowish.

Radula similar to that of sinnata, with c. 50 rows, central plate small,
quadrangular, with acicular median cusp, lateral with wing-like appendage.

Living: East London, Richmond (Alexandria Division); Port Elizabeth;
Jeffreys Bay; False Bay (all U.C.T.).

Remarks. The 33 mm. example has on the body whorl some inconspicuous
nodules on the ribs below the shoulder. One specimen (S. Afr. Mus., locality ?),
26 x9 mm., has 16 ribs on each of the last four whorls, the early whorls worn.

There is a curious resemblance in colour pattern and spiral sculpture to
Drillia albotessellata (q.v., p. 122).

Clavatula subventricosa (Smith)
Fig. 5 ¢

1877. Smith. Ann. Mag. Nat. Hist. (4), xix, p. 500 (Pleurotoma (Clionella) s.).
1892. Sowerby. Mar. Sh. S. Afr., p. 6, pl. 4, fig. 76 (bad).
1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 16, pl. 2, fig. 8 (nereia).

Shoulder above middle of whorl. Protoconch ? Postnatal whorls 7 (? 8).
Oblique ribs 14. Suture more or less embracing lower ends of ribs; cingulum

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA IOI

not very noticeable and only sometimes slightly nodulose. Ribs as prominent at
their lower ends as at the shoulder, and on the body whorl ending below rather
abruptly, the profile being thus squarish. Spiral striae in sulcus fewer and
farther apart than in krausst, below shoulder 3-4 on early whorls, 5-6 on later
whorls. 23 <9 mm.

Operculum oval, nucleus at middle of inner margin, which is thickened.

White with orange-brown or purplish brown irregular spots and marks on
sulcus; a pale band crossing the ribs and grooves, below which the base is
brown, inner surface of outer lip consequently white above, followed by a brown
band, canal white; operculum horny, periostracum greyish-brown.

Radula similar to that of sinuata, with c. 50 rows, central plate small,
quadrangular, with acicular median cusp, lateral with wing-like appendage.

Living: East London, Richmond (Alexandria Division); Kleinmond
(Bathurst Division); Port Elizabeth (all U.C.T.).

Dead: Still Bay (S. Afr. Mus.).

Remarks. Sometimes difficult to distinguish from semicostata and kraussi, but
the squarish profile and the brown band inside the aperture seem to be
distinctive. i

The Still Bay specimens (coll. Muir) are all worn, though two retain traces
of the periostracum. —

Sowerby’s bad figure shows a dark band in the middle of the whorl.

In 8. Afr. Mus. 7 specimens, without locality, identified by Tomlin as.
nereia. ‘They seem to me to be examples of subventricosa in which the squarish
profile of the whorls has been worn to a more evenly convex curve as in Bartsch’s
figure.

Clavatula hottentota (Smith)

Only one specimen with the animal is available. It was collected at
Lambert’s Bay (U.C.T., L.B.375.V.), and like many examples of C. sinuata and
species of Burnupena etc. is much corroded, surface with minute perforations
probably caused by Cliona. The ribs are scarcely traceable. Nevertheless the
specimen does not seem referable to any other species. The shape and the lip.
sinus correspond with those of hottentota. 12 x 4-5 mm. Operculum (damaged)
appears to have the nucleus more apical than lateral.

Radula similar to that of sinuata, with 25 rows, small central plate, elongate
lateral plate with appendage.

This one specimen scarcely justifies transferring hottentota definitely to the
genus Clavatula. Moreover there are no records of hottentota from the west coast,
not even from the west coast of the Gape Peninsula or Table Bay.

For remarks on synonymy see infra, p. 120.

Turris stolida (Hinds)
Fig. 3¢
1843. Hinds. Proc. Zool. Soc. Lond., p. 37 (Pleurotoma s.).
1844. id. Zool. Voy. ‘Sulphur’ Moll., p. 15, pl. 5, fig. 5 (Pleurotoma s.).

102 ANNALS OF THE SOUTH AFRICAN MUSEUM

1923. Odhner. Med. Géteb. Mus., xxiii, p. 7 (Drillia s.).
1926. Tomlin. Ann. Natal Mus., v, p. 290 (Drillia s.).

Length of aperture 14-14 times in spire. Protoconch high, 24 whorls,
smooth. Postnatal whorls 9-10. Oblique ribs 11-12 on early whorls, 13-14 on
later whorls, prominent at shoulder but petering out below, scarcely or only
just reaching suture of following whorl. Fine spiral striae over whole whorl.
No cingulum below suture. Outer lip with broad and deep sinus. 68 x 20 mm.
(protoconch and 1 whorl missing, width across shoulder of last whorl).

Operculum ovate, nucleus apical, 11 x 4 mm. in shell 51 mm. long.

Uniform cream under the pale buff periostracum, operculum horny-
amber. Animal white with black specks (K.H.B.).

Radula with 38 rows, no central plate, lateral with wing-like appendage.

Agulhas Bank, 40-43 fathoms (Hinds, Odhner). Glendower Beacon
(Port Alfred area) to False Bay, 32-73 fathoms (S. Afr. Mus. P.F. Coll.).
St. Sebastian Bay, 40 fathoms (coll. K.H.B.). False Bay (U.C.T.).

Slender form: slightly more slender, the ribs consequently closer together,
with 5 ribs visible in face view!’ instead of 4. Yellowish brown (darker than
typical form). 5215 mm.

Off Cape Hangklip, 73 fathoms, 1 together with 1 typical; off Glendower
Beacon (Port Alfred area), 66 fathoms, 2. (S. Afr. Mus. P.F. Coll.)

Remarks. The 60 mm. Pleurotoma (Surcula) margaritae Smith (1904. Ann.
Mag. Nat. Hist. (7), xiii, p. 458) from off the Andaman Islands, 405 fathoms,
appears from the figure (1907. Illustr. Zool. R.ILM.S. ‘Investigator’. Moll., pl. 14,
figs. 2, 2a) to be almost indistinguishable from stolzda.

Turris lignaria (Sow.)

1903. Sowerby. Mar. Invest. S. Afr., 11, p. 215, pl. 3, fig. 4 (Pleurotoma (Clavus) 1.).

1903. Von Martens. D. Tiefsee Exp., vii, p. 24 (Clionella semicostata var., non
Kiener).

1925. Thiele. ibid., xvii, p. 212, pl. 35 (23), fig. 10 (Clconella semicostata var.,
non Kiener).

Very close to stolida, but more compact in shape, the spire less tapering,
canal shorter and rostrum blunter. Protoconch 2 whorls, smooth (fide Sowerby).
Postnatal whorls 6 (? 7). Oblique axial ribs 10 on 1st whorl (as preserved,
possibly this is the 2nd), increasing to 11 (12) on body whorl. Fine spiral striae
over whole whorl. No cingulum. Lip sinus broad and deep. Columella
slightly rimate at rostrum. 22 <9 mm. (Sowerby).

Pale brown, columella white. Yellowish-brown, columella whitish (Thiele).

Off west coast of Cape Peninsula, 136 fathoms (Sowerby); Agulhas Bank
(34° 51'S., 19° 37’ E.), 80 metres (von Martens); False Bay, 32 fathoms, 1 dead
(S. Afr. Mus. P.F. Coll.).

13 j.e. not counting the profile rib on either side.

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 103

Remarks. Might almost be regarded as a squat form of stolida. I have seen
only two specimens. One returned by Sowerby with his label; this cannot be
regarded as the type although it measures 22 <9 mm. because it lacks the
2-whorled protoconch (possibly also the 1st postnatal whorl). Sowerby gave
the total number of whorls as 103; this seems excessive (cf. his figure) and is
probably a misprint.

The False Bay example has the same width as this specimen, but has only
the 4 last whorls.

Both these examples differ from equal-sized stolida by the rimate columella;
only in large examples (40 mm. upwards) of the latter is the margin of the
columella glaze slightly raised as a free edge, but there is no (umbilical) indent.

Although the operculum and radula are unknown, this species is included
provisionally in Turris for the sake of comparison with stolida.

Turris cingulifera (Lam.)

Fig. 6 a and profile

1897. Sowerby. Append. Mar. Sh. S. Afr., p. 2 (Pleurotoma c.).
1917. Melvill. Proc. Mal. Soc., xii, p. 162 (Surcula c.).

Anterior canal short. Length of aperture less than half total length.
Protoconch 33 (4) whorls, diam. 1:25 mm., alt. 1°75 mm., apex smooth,
following whorls with slightly curved axial riblets, c. 26 on last whorl, junction
with Ist postnatal whorl abrupt. Below suture 2 (in early whorls) —5 (later
whorls) spiral lirae followed by a deep narrow groove, and then 2 lirae; lip
sinus forming the moderately prominent shoulder, with 2 (3) lirae, followed by
2 lirae with finer intervening lirae; on base 5-6 lirae with intermediates.
Sometimes axial striae forming a fine cancellate sculpture. 55 x 15 mm. (minus
protoconch, width across shoulder).

White, lip sinus (shoulder) with irregularly spaced brown spots, the other
lirae with numerous brown dots.

Dead: Durban (Sowerby); Natal coast and Mozambique Island (S. Afr.
Mus.); off Umkomaas River, 40 fathoms, 1 juv. dead but unworn. (S. Afr.
Mus. P.F. Coll.)

Distribution. Farquhar Island (S. Afr. Mus.), Mauritius, Madagascar,
Seychelles, Indo-Pacific.

Remarks. Protoconch described from a 13:5 mm. juvenile.
Turris acuta (Perry)
Fig. 6 profile

1811. Perry. Conchology, pl. 54, fig. 5.

1822. Lamarck. Anim. sans Vert., vii, p. 95 (Pleurotoma tigrina).
1897. Sowerby. Append. Mar. Sh. S. Afr., p. 2 (P. tigrina).
1917. Melvill. Proc. Mal. Soc., xii, p. 142.

104 ANNALS OF THE SOUTH AFRICAN MUSEUM

Bet a Tong WN
SS SSS
SSS ee

oY 8

pes
We

sy S58

Fic. 6. Protoconchs of: a. Turris cingulifera (Lam.). b. T. indica Bolten.

c. T. multiseriata (Smith). d. T. gilchristi (Sow.). Profiles of (left to right):

cingulifera, indica, acuta, gilchristi, lobata, multiseriata; position of sinus marked
by a double line.

Anterior canal long. Length of aperture half total length.

Below suture a sharp keel, followed by several lirae all of about the same
strength; lip sinus forming a prominent shoulder, square in profile, below which
on body whorl 3 (4) sharp lirae with finer intermediaries.

57 X15 mm. (width across shoulder).

Operculum ovate, nucleus apical, 9 x 5 mm. in shell 57 mm. long.

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 105

Biscuit colour, with irregularly spaced dark brown spots on the keel below
suture, and numerous dots on the margins of the lip sinus (shoulder) and the
other lirae.

Dead: Durban (Sowerby); 2 worn specimens probably from Durban

(S. Afr. Mus.).

Distribution. Farquhar Island (S. Afr. Mus.); Mauritius, Madagascar,
Seychelles, Indo-Pacific.

Remarks. Distinguished from indica by the larger spots being on the keel
below suture, the lirae between the keel and shoulder all of about same strength,
and shoulder square in profile.

Turris indica Bolten-R6éding
Figs. 3 g, 6 6 and profile

1798. Bolten-Réding. Mus. Bolt., p. 124, no. 1594 (zndica).

1822. Lamarck. Anim. sans Vert., vii, p. 95 (marmorata).

1843. Reeve. Conch. Icon., i, pl. 3, fig. 21 a and b (var. maculata).

1902. Sowerby. Mar. Invest. S. Afr., ii, p. 100 (marmorata).

1917. Melvill. Proc. Mal. Soc., xii, p. 143.

1926. Tomlin. Ann. Natal Mus., v, p. 289 (marmorata).

1942. Gravely. Bull. Madras Govt. Mus., n.s. V, no. 2, p. 73 (in key), fig. 14(1)-

Anterior canal very long. Length of aperture a little more than half total
length. Protoconch 24 whorls, apex smooth, followed by 12-13 axial riblets,
diam. 1°25, alt. 1-5 mm.

Postnatal whorls 10. Suture slightly underriding preceding whorl. Below
suture a slightly prominent keel followed by several lirae of varying strength;
lip sinus forming the shoulder, the upper edge of which forms a very prominent
keel, below sinus 4 or more sharp lirae with finer intermediaries. 59 x 16 mm.
(protoconch and end of canal broken, width across shoulder; in proportion to
an unbroken Philippine Is. specimen 52 x 13 mm. the full length would have
been 64 mm.).

Operculum ovate, nucleus apical, 5°75 X 3 mm. in shell 32 mm. long.

Marbled brown and white in varying proportions, but the larger dark
brown spots or marks are on the shoulder sinus. South African specimens are
var. maculata: white or biscuit colour, with brown or orange-brown spots on
shoulder and numerous smaller spots and dots on the other lirae. Albino
specimens are known.

Radula with about 45 (specimen incomplete) pairs of teeth, no central
plate, lateral without wing-like appendage (see p. 93).

Off Tugela River (Natal), 55 fathoms (Sowerby) ; Natal, from fish stomachs
(Tomlin); off Tugela River, 40-73 fathoms, off Cape Vidal and O’Neil Peak
(Zululand), 55-100 fathoms (S. Afr. Mus. P.F. Coll.). Most specimens dead,
but living ones were taken off the Tugela River.

Distribution. Cargados Islands, Red Sea, Indo-Pacific.

106 ; ANNALS OF THE SOUTH AFRICAN MUSEUM

Remarks. Distinguished from acuta by the larger spots being on the shoulder,
the varying strength of the lirae between suture and shoulder, and the projecting
upper edge of the latter.

Turris gilchristi (Sow.)
Figs. 3 h, 6 d and profile

1897. Sowerby. Append. Mar. Sh. S. Afr., p. 2 (monilifera non Pease).
1902. id. Mar. Invest. S. Afr., ii, p. 99, pl. 2, fig. g (Pleurotoma g.).
1903. Smith. Proc. Mal. Soc., v, p. 362.

1917. Melvill. ibid., xii, p. 145.

Anterior canal very long. Length of aperture equal to, or a little more or
a little less than half total length. Protoconch 34 whorls, diam. 1:3, alt. 175mm. ,
apex smooth, followed by 20-25 axial riblets on last whorl and a half. Post-
natal whorls 9-10. Suture slightly canaliculate. Below suture one strong keel
followed by 3-4 less strong lirae; lip sinus forming an outstanding girdle with
numerous tubercles, oblong in axial direction; below shoulder of last whorl
3-4 main keels with smaller intermediary lirae; oblique axial growth lines well
marked. 61 x18 mm.

Operculum ovate, nucleus apical, 7 x 3 mm. in 39 mm. shell.

White, some specimens with orange-brown spots in the hollows between
the shoulder knobs, and orange spots or suffusions on the lirae above and below.

Radula with about 70 pairs of teeth, no central plate, lateral without wing-
like appendage. The radula of a Farquhar Island specimen corresponds with
this.

Off Tugela River mouth (Natal), 55 fathoms (Sowerby); Zululand and
Natal coast 27-90 fathoms, and as far south as off Cape Natal (Durban),
185-200 fathoms; off Hood Point (East London), 49 fathoms. (S. Afr. Mus.
P.F. Coll.)

Distribution. Farquhar Island (S. Afr. Mus.); Mekran coast (S. Persia),
180 fathoms (Melvill).

Remarks. Smith suggested that the specimen recorded by Sowerby as
monilifera was probably a gilchristi. Pease’s description of the Sandwich Island
species, as far as it goes, fits galchristi; but actual specimens should be compared.

Sowerby in 1902 compared his species with the Californian gemmata Hinds
and the Chinese Azenert Doumet. He did not specify the differences but referred
to the size, which may be a misleading character; Sowerby saw no specimens
of gilchristi larger than 32 mm.

Although broken, the apex (protoconch plus 5 whorls) from the Hood Point
locality is unworn, indicating that the species may occur living as far south as
the East London area. Very few bottom samples are available between here
and the Natal coast.

Plump and slim forms occur, e.g. 38x 13 mm. and 39 XII mm.; some-
times both forms were taken in the same haul. The rather striking difference

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 107

in appearance is due mainly to the greater or lesser prominence of the tuberculate
shoulder band; and in the slim forms not only are the tubercles less prominent
but on the later whorls they tend to degenerate into a double keel crossed by
coarse growth-lines (cf. von Martens’ remarks on P. carinata Gray, 1903.
D. Tiefsee Exp., vii, p. 77; also Smith on vagata, 1904. Ann. Mag. Nat. Hist. (7),
Xiii, p. 456). Such specimens appear at first sight to be quite different from
strongly tuberculate specimens.

Further, the spiral lirae and the oblique growth-lines vary in intensity;
the former may be high and sharp, with crinkly edges, especially the uppermost
lira forming the lower margin of the sutural canal; and on the base of the body
whorl the growth lines may produce an almost cancellate sculpture.

This species should be compared with P. (Gemmula) carinata Gray, a figure
of which is given in I/lustr. Zool. R.I.M.S. ‘Investigator’, Moll., pl. 20, figs. 3, 4,
1908. The specimens from off the Somaliland coast and Nicobars, identified
as carinata by von Martens (1903. loc. cit., p. 76), were later regarded by Thiele
as a separate species: valdiviae Thiele (1925. D. Tiefsee Exp., xvii, p. 208, pl. 35
(23), fig. 1). Thiele stated (p. 208) that the figure of vagata Smith 1895 and 1904
(Illustr. Zool. R.ILM.S. ‘Investigator’ Moll., pl. 14, figs. 3, 3a, 1907) was similar to
valdiviae; and likewise P. sibogae Schepman.

P. (Gemmula) gemmulina von Martens 1902 (1903. loc. cit., p. 77, pl. 1,
figs. 2, 2a) from the eastern Indian Ocean is also not dissimilar; and is compared
by von Martens (p. 78) with praesignis Smith 1895 from Ceylon and (1906) the
Coromandel coast.

P. aethiopica Thiele (1925. loc. cit., p. 208, pl. 34 (22), fig. 25), from off the
East African coast, and P. fusiformis Thiele (1925. loc. cit., p. 210, pl. 34 (22),
fig. 24) from the East Indies, also invite comparison.

Melvill compared his specimens with ceylonica Smith: the tubercles are
smaller and more compact in gilchristi; and the latter is narrower than carinata
Gray =granosa (Helb.). .

All these species can be more or less closely matched among the small
series (60) of gilchristi in 8. Afr. Mus.; but it is inconceivable that so many
separate species, including gilchristt, exist together in one small area off the
Natal-Zululand coast.

For the present I maintain gilchristi, the radula of which is now known,
though eventually it will have to become a synonym of one of the above species.

Turris lobata (Sow.)
Figs. 3 2, 7, 6 profile
1903. Sowerby. Mar. Invest. S. Afr., ii, p. 213, pl. 4, fig. 9 (Pleurotoma (Surcula) 1.).

1906. Smith. Ann. Natal Mus., i, p. 24 (Pleurotoma, not Surcula).
1925. Thiele. D. Tiefsee Exp., xvii, p. 210 (similarity of shell with bisinuata).

108 ANNALS OF THE SOUTH AFRICAN MUSEUM

Anterior canal short. Length of aperture 14-14 in spire. Protoconch ?
Postnatal whorls 10. Below suture one rather prominent sharp keel, followed
by one feeble lira (or 2 lirae); lip sinus forming an outstanding nodulose girdle,
nodules rounded or very slightly oblong in an axial direction; only a portion of
protoconch preserved in one specimen, but nodules beginning apparently
immediately on first postnatal whorl. Below girdle 3 spiral lirae, the lowermost
one in some specimens becoming prominent and coalescing with the middle lira
to form a strong costa forming a sinus on margin of outer lip. 31 x 12 mm. lobate
specimen; 32X11 mm. non-lobate; 35x14 mm. lobate; 39 x13 mm., costa
present but lobe not formed. (In all cases protoconch missing.)

Operculum oval, nucleus apical, 9 x 4 mm. in 35 mm. shell.

Radula (of an East London specimen) with c. 75 pairs of teeth, no central
plate, lateral broadly cuneiform, one margin sharply angular (in edge view),
no appendage.

Off Cape Natal (Durban), 440 fathoms, and (dead shells) off Buffalo River
(East London), 310 fathoms (Sowerby).

S. Afr. Mus. P.F. Coll.: Co-types (topotypes) from above localities, one
of the East London specimens taken alive.

Off Cape Point, 380—g00 fathoms (3 living and 5 dead shells).

The largest of the Cape Point shells (apex corroded, only 6 whorls remaining)
is 40X14.mm. These examples are more corroded than those from the Natal-
East London area. There are 3-4 prominent lirae between the keel below the
suture and the nodular girdle, at least in the earlier whorls; growth-lines much
more prominent and on the later whorls tending to obliterate the spiral lirae.
No indication of the formation of a basal costa and additional sinus in outer lip.
Operculum as above.

Radula with about 70 pairs of teeth, the lateral plate broadly cuneiform
as in the East London example, but no angular margin (in edge view).

Conchologically the Cape Point examples are not separable from the Natal-
East London specimens. The slight difference in the radula teeth is scarcely of
specific importance; but there is only one radula of each for comparison.

Remarks. The formation of a strong basal costa and an additional sinus on
the outer lip is paralleled in Ptychosyrinx bisinuata (von Martens, 1901) (see:
1903. D. Tiefsee Exp., vii, p. 82, pl. 1, fig. 8; juvenile, ibid., pl. 1, fig. 3, as
rotatilis; and 1925. Thiele. ibid., xvii, p. 210, pl. 35 (23), fig. 4), from off the
East African and Somaliland coast. The first 3 whorls of this species are axially
ribbed, but so far as the embryonic and early whorls are preserved in the
corroded South African specimens there is no indication of such ribbing in
lobata. In other respects there are no conchological differences; in fact,
von Martens’s figure 8 might almost have been drawn from one of the East
London examples. |

The genus Ptychosyrinx, however, is distinguished by the concentric oper-
culum, and by the large central plate in the radula.

This case illustrates the danger of attempting to classify dead shells.

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA I09Q

Melvill (1917. Proc. Mal. Soc., xii, p. 150) records the formation of a some-
what similar lobe on the outer lip of Drillia athyrma Melv. & Stand. 1901, from
the Persian Gulf.

Turris (Gemmula) multiseriata (Smith)
Fig. 6c and profile

1877. Smith. Ann. Mag. Nat. Hist. (4), xix, p. 491.

1go1. Melvill & Standen. Proc. Zool. Soc. Lond., ii, p. 434 (Pleurotoma (Gemmula)
1917. Melvill. Proc. Mal. Soc., xii, p. 145, pl. 8, fig. 3 (Turris (Gemmula) m.).

Aperture 14-14 times in spire. Protoconch 23 whorls, diam. 0-5 mm., apex
smooth, followed by a spiral keel, which is continued on to shoulder nodules on
Ist postnatal whorl. Postnatal whorls 7. Curved axial plicae (c. 21), retractive
and varicoid between suture and lip sinus, and below sinus forming a shoulder
with strong nodules (oblong in axial direction), continued on base nearly to
extremity; crossed by spiral lirae, 2 on the varicoid nodules, 2-3 in lip sinus,
2 on shoulder nodules, 2-3 below, with c. 10 additional ones on base; on base
axial and spiral lirae form a more or less cancellate sculpture, slightly nodulose
at the intersections. Lip sinus deep, remote from suture. Canal short.
15 X5°5 mm. Red-brown to ochraceous (Melvill).

Off Umvoti and Umhloti River mouths (Natal) 27 fathoms, 5 fresh; off
Tongaat River, 36 fathoms, 1 dead; Algoa Bay, 21 fathoms, 1 dead. (S. Afr.
Mus. P.F. Coll.)

Distribution. Ceylon, Persian Gulf, Karachi. From the latter locality very
large specimens 3 inch long. Also China Seas.

Remarks. Three of the South African specimens were identified by Sowerby.
At first sight somewhat similar to gilchristi on account of the axially oblong
tubercles, but the protoconchs are quite different.

Operculum and radula not present in the South African examples. It is
here assumed that the radula has been examined (by Melvill or some other
author) and that the species is correctly classified in the genus Turris.

Turris saldanhae n. sp.

Fiss. 3 f, 7

Aperture subequal to spire, or a little shorter. Protoconch broken. Post-
natal whorls 73-8. Shoulder somewhat angular (but frequently corroded), a
little above middle of whorl; oblique axial ribs from shoulder to suture below,
petering out on base, 12-14 on earlier whorls, 15-16 on later whorls, subequal
in width to intervening grooves; crossed by spiral lirae, 2 or 3 on 3rd whorl,
3 or 4 on 4th, increasing to 7 or 8 (g) on last whorl, 12-14 additional lirae on
base; growth-lines not conspicuous. Outer lip with broad and moderately

IIo ANNALS OF THE SOUTH AFRICAN MUSEUM

deep sinus; canal moderately long. 33-34 (protoconch broken) x 12 mm.;
and 46X15 mm. -

Operculum ovate, nucleus apical, 9 x 4
mm. in 33 mm. shell.

White with yellowish-brown _ perio-
stracum, operculum amber.

Radula with 60 pairs of teeth, no
central plate, lateral with wing-like appen-
dage. 3

Type locality. Off Baboon Point (Saldanha
Bay), 31 fathoms (S. Afr. Mus. A1738, P.F.
Coll.).

26° 33’ S., 15° E. (off Liideritzbucht),
55 metres (Fisheries Survey vessel. Africana
II, AFR.1224); also AFR.1263, 26° 39’ S.,
14° 17’ E., 311 metres, and AFR rareniga-
D. 14 95 E., 168 metres.

Remarks. All the specimens taken by
the Pieter Faure are more or less corroded,
the protoconch broken off and, even in the
smallest example (22 mm.), the apex stopped

(ite Gh Ute ce a bree o. with secondary shelly substance. The Ist
with protoconch further enlarged. and 2nd whorls appear to have been smooth,
the ribs beginning on the 3rd whorl.

The 4 Africana examples, 188 mm. up to 46X15 mm., though more
slender, are obviously conspecific. The protoconchs are broken off and the
early whorls corroded, though not so much as in the Saldanha Bay examples.
Consequently the lirae are more prominent, and owing to the slight lengthening
of the whorls there is an additional spiral lira on each whorl, and 15-20 lirae
on the base. At first glance there appears to be a likeness between the Africana
46 mm. example and the enlarged figure of macilenta Melvill (see Drillia platy-
stoma, p. 125), but the proportion of aperture to spire, and number of ribs and
spiral lirae, are different.

In sculpture this species is rather similar to the worn 18-6 mm. Drillia
halidoma Bartsch 1915 from the ‘Cape of Good Hope’. The latter, however, has
fewer spiral lirae (see Clavatula semicostata).

Three specimens, 26, 27 and 33 mm. long, from off Cape Point, 250-700
fathoms (S. Afr. Mus. A362—A364, P.F. Coll.), agree with the above except the
spiral lirae are finer and more numerous, due to the development of inter-
mediaries; this is especially noticeable on the base.

All three are dead shells, white, without protoconchs, slightly corroded,
and without any trace of periostracum.

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 1If

Subfam. BRACHYTOMINAE Thiele

‘Clavatula’ tumida (Sow.)
Figs. 4e, 8a, 9a

1870.: Sowerby. Proc. Zool. Soc. Lond., p. 253 (Clavatula t.).

1892. id., Mar. Sh. S. Afr.; p. 5, pl. 5, fig. 101 (Pleurotoma t.).

1903. Von Martens. D. Tiefsee Exp., vii, p. 24.

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 19, pl. 2, fig. 3 (Clavatula haliplex).

1932. Turton. Mar. Sh. Pt. Alfred, p. 19 (haliplex), and p. 19, pl. 4, no. 146
(gravis, non Hinds).

Protoconch 14-2 whorls (iunction with 1st postnatal whorl not clear)
diam. 1°75-2 mm., alt. 2-2:25 mm., usually corroded. Postnatal whorls 7-8.
Oblique nodules above suture longer than in gravis, better described as oblique
ridges. Cingulum below suture not very timid, obscurely nodulose; or in other
words: the arcuate oblique ridges cross the whole whorl but are interrupted
by the sulcus groove, above which they are less conspicuous (often quite
obsolete) than they are below; no indication of additional nodules or a ridge
below shoulder on body whorl. Spiral striae on early whorls (if not corroded),
about 10 on Ist whorl, 12 on 2nd, microscopically granulose at intersections
with growth-lines; striae usually obsolete on later whorls. Lip sinus narrow
and shallow. 60 (minus protoconch) x 20 mm. Another shell: 56x20 mm.
Sowerby’s figure, if natural size, is 61 mm.

Operculum oval, nucleus slightly below middle of inner margin, which is
thickened on outer surface; 11 x6 mm. in 48 mm. shell.

Cream under the yellowish-brown periostracum, operculum horny. Animal
flesh-coloured.

Radula without basal membrane, 20-24 teeth (3 radulae examined),
elongate, slender, basal half slightly thicker, apex acute, one apical barb,
another on opposite side a little farther proximally.

Agulhas Bank (Sowerby). Off Cape Morgan, 34 fathoms, and Agulhas
Bank to False Bay, 27-55 fathoms (S. Afr. Mus. P.F. Coll.). False Bay, 54
metres (U.C.T.).

Remarks. ‘Two live specimens taken in the same haul (P.F. Coll.) represent
stout and slim forms: 49X17 mm. and 52 X16 mm. respectively.

Bartsch’s haliplex appears to be a young worn specimen of this species.
There is one in S. Afr. Mus. almost exactly the same size as Bartsch’s type which
corresponds with his description and figure. It has the upper part of the whorls
white where the periostracum is worn off.

This species combines a Clavatula-like operculum with a radula composed
of a bunch of barbed teeth without basal membrane as in the Brachytominae.

I12 ANNALS OF THE SOUTH AFRICAN MUSEUM

e b

ee L B ins
ey
7 f I A
d

Fic. 8. Radula teeth of: a. ‘Clavatula’ tumida (Sow.). 6. ‘Genotia’ belaeformis

(Sow.). ¢. Asthenotoma vertebrata (Smith). d. Cythara africana (Sow.). e. Lienardia

grayi (Rve.). f. ‘Drillia’ scitecostata (Sow.). g. ‘Drillia’ fultoni (Sow.). h. Philbertia
capensis (Smith).

‘Genotia’ belaeformis (Sow.)
Figs. 8b, 9 b

1903. Sowerby. Mar. Invest. S. Afr., ii, p. 216, pl. 4, fig. 8 (Pleurotoma (Genoitia) b.).
1906. Smith. Ann. Natal Mus., i, p. 24 (Genotia b.).

Anterior canal short. Length of aperture subequal to half total length.
Protoconch 14 whorls, mammillate, diam. 1-75, alt. 1-5 mm., smooth (but partly
corroded). Postnatal whorls 5. Whole whorl with fine close spiral striae, crossed
by oblique growth-lines, the latter immediately below the suture forming a
narrow crimped band above the sulcus. Outer lip prominent, sinus broad and

deep. 24X11 mm.
Operculum ovate, nucleus apical (apex broken in the only example).

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA Il3

Radula without basal membrane, teeth narrow, aciculate, shallowly
grooved nearly to apex, margins thickened.

Off Cape Point, 230 fathoms (Sowerby). Two cotypes in S. Afr. Mus.

Off Cape Point, 190 fathoms, 1 live, 2 dead: 36° 40’ S., 21° 26’ E.,
200 fathoms, 1 dead but fresh. (S. Afr. Mus. P.F. Coll.)

Remarks. The cotypes are corroded; two of the other Cape Point specimens
are slightly corroded at the apex, but the third (15 mm. long) has an unbroken
protoconch. The largest specimen is from the southern slope of the Agulhas
Bank; it also has an unbroken protoconch, and is more strongly sculptured,
some of the sigmoid growth-lines, especially across the sulcus, being very sharp.
Sowerby’s expression ‘a punctured groove a little below the suture’ refers to
the most concave portion of the sulcus immediately below the crimped band.

Generic position doubtful; the radula excludes it from the Turrinae, and
the possession of an operculum from Genota (Cytharinae). The radula is very like
that of Thesbia nana (see: Sars. 1878. Moll Reg. Arct. Norv., pl. viii, figs. 3
¢, d), another Cytharine genus without operculum.

Gen. AsTHENOTOMA Harr. & Burr.

1942. Gravely. Bull. Madras Govt. Mus., n.s. v, no. 2, pp. 71, 72 (key to 3 closely
allied species).
Asthenotoma vertebrata (Smith)
Figs. 8c, 9¢

1875. Smith. Ann. Mag. Nat. Hist. (4), xv, p. 416 (Pleurotoma v.).

1903. id. Proc. Mal. Soc., v, p. 363.

1917. Melvill. ibid., xii, p. 149, pl. 8, fig. 4 (Turris (Tomopleura) v.).

1942. Gravely. loc. cit., p. 72, figs. 13 a (inverted), and 14 (2) (not good).

Aperture 1} times in spire. Protoconch 13 or 2 whorls (incomplete), diam.
0-5 mm., smooth, with a few feeble axial plicae before junction with Ist post-
natal whorl. Postnatal whorls 10}; 1st with one spiral keel in middle, on later
whorls successively 3, 4, and 5 keels; one slightly below suture forming upper
boundary of sulcus, keel in middle of whorl the most prominent (Gravely:
‘lower cardinal spiral’), and forming lower boundary of sulcus; below this
3 keels, and additional lirae on base. Axial plicae between the keels; each plica
in the sulcus is slightly nodulose forming a feeble moniliform lira in middle of
sulcus (Gravely: ‘intracardinal spiral’). Columella with a more or less prominent
pleat. Canal short. 18-5 5:5 mm.

Operculum ovoid, nucleus at the reclangal apex, inner margin straight
for greater part of length.

Pale greyish, interior of aperture violaceous.

Radula without basal membrane, c. 48 teeth (i.e. 24 pairs), elongate,
slender, with a very delicate flange (not barbed) on either side of apex.

Durban (Smith, 1903).

II4 ANNALS OF THE SOUTH AFRICAN MUSEUM

eA Bde

geet Awa
a

TLL

SSS

wes % otro

ee ESOS
)

a}
Sis
BAST SA

Fic. 9. Protoconchs of: a. ‘Clavatula’ tumida (Sow.). 6. ‘Genotia’ belaeformis (Sow.).—
c. Asthenotoma vertebrata (Smith).

_. Off Umhlanga River (north of Durban), 22-26 fathoms, 1 dead (S. Afr.
Mus. P.F. Coll.). tee
Delagoa Bay, 1 dead (S. Afr. Mus. Coll. K.H.B.).
Morrumbene estuary, Inhambane, Portuguese East Africa, 2 living
(URE. 1):

Distribution. Karachi, Madras, and Japan.

Remarks. Melvill, contrary to some authors, keeps vertebrata separate from.
nivea (Phil.) and its var. violacea Hinds.

From the description and figure it is impossible to say how this species
compares with the Mauritian Daphnella elata Sow. (1893. Proc. Zool. Soc. Lond..,

p- 490, pl. 38, figs. 19, 20).
Asthenotoma eva (Thiele)
His. 21 @

1925. Thiele. D. Tiefsee Exp., xvii, p. 227, pl. 37, (25), fig..12 (Bela ee

Protoconch 14 whorls, smooth. Postnatal whorls 3. Spiral keels 2 on each
whorl, the upper one peripheral and more prominent; distinct sharp, close-set
axial pliculae or growth-lines across sulcus and between the keels; on base 4
additional distinct lirae and 3-4 obscure ones. 3°75 X1°5 mm. Thiele: 48x
2°2 mm.

Operculum (apud Thiele) oval, narrowed below.

Radula (apud Thiele) without basal membrane, teeth ‘arrow-like’ (‘pfeil-
zahne’, not figured by Thiele). ,

35° 19’ S., 20° 15’ E., 126 metres (Thiele).

Off Cape Recife, 56 fathoms, 1 dead (S. Afr. Mus. A8563, P.F. Coll.).

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA ny

Remarks. Thiele placed his species in Bela because it had an operculum and
‘pfeilzahne’, and from the character of the shell suggested the genus Drilliola.
In 1929, however, he put Bela as a synonym of Mangelia (Cytharinae, without
operculum), and Drilliola as subgen. of Asthenotoma in the Brachytominae.

The sculpture is not unlike that of Asthenotoma species. The axial pliculae
are more conspicuous on the present specimen than in Thiele’s figure.

Subfam. GyTHARINAE
Cythara africana (Sow.)
Figs. 8d, 10 a

1903. Sowerby. Mar. Invest. S. Afr., 11, p. 216, pl. 5, fig. 9 (Mangilia (Eucythara)
a.).
1906. Smith. Ann. Natal Mus., i, p. 27 (listed).
Protoconch 1} or 2 whorls, diam. 0-3-0-4, alt. 0-4: mm. (but nucleus broken),
smooth, last half whorl with rather closely set, narrow axial plicae. Postnatal
whorls 7. Axial ribs 11 on Ist whorl, increasing to 12 on penultimate, and

es SIE: Zee
a

]
dpdt
oe a a
Wi lf He

Fic. 10. Protoconchs of: a. Cythara africana (Sow.). 6. Lienardia grayi (Rve.).

12-13 (14) on last whorl, from suture to suture, and extending to extremity of
base, crossed by numerous very fine spiral striae over whole whorl. Columella
with numerous (at least 25 in the Type) plicae. Outer lip in adult with sharp
margin, internally plicate (c. 25), externally with varix. 20x 7 mm.

No operculum.

Coloration faded except a faint mauve band on the sulcus and another
below middle of body whorl, some dull orange spots on the lip varix.

Radula without basal membrane, number of teeth ? (only 10 observed
during treatment with KOH); tooth forming a v-shaped channel, with acute
non-barbed apex, base pear-shaped.

116 ANNALS OF THE SOUTH AFRICAN MUSEUM

Off Umhloti River mouth (Natal), 25-27 fathoms. Type, cotype, and
‘6 other specimens in S. Afr. Mus. (P.F. Coll.).

Remarks. Only eight specimens were obtained in three hauls in the same
-area, the actual bearings being: Umbhloti River mouth NW. x W2W. 22 miles,
25 fathoms; NNW. 14 miles, 27 fathoms; and N.XxE., 2 miles, 27 fathoms.

The cotype is 18 mm. long, with thin outer lip.

The cotype and one of the others were taken alive. In all the specimens the
actual nucleus of the protoconch is broken.

The radula teeth seem shaped to conduct poison into a wounded prey.
‘The true shape is not seen in the final mount, but was happily seen at an earlier
stage by delicately sliding the cover-slip so as to roll the teeth into different
positions. Thiele (1925. D. Tiefsee Exp., xvii, p. 207) mentions the ‘rinnen-
formige Endteil . . . mit ibergebogenen Lamellen’ in Mangelia (Benthomangelia) ;
and the same shape probably occurs in Mangelia costata and the species of ‘Bela’
figured by Sars (1878. Moll. Reg. Arct. Norv., pl. viii, figs. 7, etc., especially 12).
In Thesbia nana Sars shows the tooth with a simple shallow groove (loc. cit.,
fig. 3) and in Clathurella leufroy (fig. 2) with narrow marginal flanges.

Lienardia gray (Rve.)
Figs. 8 e, 105

1845. Reeve. Proc. Zool. Soc. Lond., p. 114 (Pleurotoma g.).

1901. Sowerby. Proc. Mal. Soc., iv, p. 214, pl. 22, fig. 20 (Drillia rugisculpta).

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 27, pl. 3, fig. 8 (Mangilia arata).

1915. id. ibid., p. 29 (Mangilia g.).

1932. Turton. Mar. Sh. Pt. Alfred, p. 28 (Mangilia g.); and p. 28, pl. 6, no. 211
(grayt var. assimilans).

Protoconch 2 whorls, diam. and alt. 0-5 mm., smooth but in living examples
very faintly shagreened and spirally striate, last three-quarters of 2nd whorl with
a spiral lira, continued as the upper of the 2 lirae on ist postnatal whorl.
Postnatal whorls 4. Axial riblets 10-11 on 1st whorl, increasing to (12)13—14(15)
on last whorl, more or less traceable across sulcus, rounded at shoulder; crossed
by 2 spiral lirae on 1st whorl, 2-3 on 2nd, increasing to 5-6(7) on 4th, with
13-15 additional lirae on base, usually 2-3 very fine lirae on sulcus above
shoulder. Columella with 2-3 plicae; outer lip in adult thickened, inner
margin denticulate. 8-5 x 3-75 mm.

No operculum.

Castaneous brown; beach specimens castaneous, amber, buff, white,
unicolorous or with pale band, broad or narrow, around middle of last whorl.

Only one animal available for dissection. The tentacles were short,
rounded lobes (? due to injury), the eyes in an expansion on outer margin.

Radula without basal membrane, about 35 pairs of slender acicular teeth,
expanded at base; shallowly grooved.

a ss ae

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA Ti.

Table Bay, False Bay, Still Bay (S. Afr. Mus.); off Cove Rock (East
London), 27 fathoms (S. Afr. Mus. P.F. Coll.); Still Bay (U.C.T.); living
example from Cape Peninsula (U.C.T.) examined.

Remarks. ‘Two lots, both from the Cape Peninsula, were labelled respectively
grayt and rugisculpta by Tomlin. I fail to see any differences.

The species seems to be a typical Lienardia. The Tertiary genus Glyphostoma
has precedence, but as Thiele (1929) remarked it is not feasible to include living
molluscs in fossil genera of which the anatomy is unknown.

Glyphostoma siren Smith (1904. 7. Malac., xi, p. 28, pl. 2, fig. 7), with pale
band around middle, appears to be very similar, but with slightly fewer axial
riblets and spiral] striae: resp. c. 10, and 3-4, with 6 additional lirae on base.
Probably only a casual variation. cf. also Mangilia helena Turton (1932. p. 28,
pl. 6, no. 208). |

Philbertia capensis (Smith)
Figs. 8h, 29 5b

1882. Smith. Ann. Mag. Nat. Hist. (5), x, p. 296.

1892. Sowerby. Mar. Sh. S. Afr., p. 6, pl. 4, fig. 84 (not good) (Defrancia c.)..
1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 23 (Mangilia c.).

1925. Thiele. D. Tiefsee Exp., xvii, p. 230 (Clathurella c.).

1925. id. ibid., p. 231, pl. 40 (28), fig. 21 (Pleurotomella ida).

1932. Turton. Mar. Sh. Pt. Alfred, p. 24, pl. 5, no. 182 (Mangiliac.). .

Protoconch 34 whorls, diam. 0-4, alt. 0-5 mm., nucleus smooth, 1st whorl
with faint oblique axial striae, next whorl with criss-cross microsculpture, last
half whorl with a spiral keel in middle and a faint lira below. Postnatal whorls
6; Ist with 3 spiral keels, the upper one being the continuation of the keel on.
protoconch, and forming a prominent shoulder; a fine subsidiary lira may some-
times develop between ist and 2nd keels (sometimes between other pairs;
v. infra); base with 8—9(10) additional keels. Narrow sharp axial ribs 9-10 on
early whorls, increasing to 11-12 on body whorl, sometimes to 13 or 14 on
plump examples; intersections with spiral keels sharp-pointed, especially on
the shoulder keel; on base ribs extend to columella, forming a lozenge-shaped
reticulation (nodulose in worn specimens). Lip sinus deep, adjoining suture.
Outer lip (when coinciding with formation of a rib) thickened, plicate within.

17 (minus protoconch) x 6-5 mm.

No operculum.

Buff, a (faint) darker band below middle of body whorl, or whole lower
half of whorl brown.

Radula without basal membrane, ¢. 40 teeth, rather stout, shallowly
grooved, margins thickened.

Kalk Bay to Port Alfred (previous records, and S. Afr. Mus.); St. Francis.
Bay, 80 metres (Thiele); 35° 19’ S., 20° 12’ E., 126 metres (Thiele: P. ida).

118 ANNALS OF THE SOUTH AFRICAN MUSEUM

Off East London, 32 fathoms, and off Cove Rock, 27 fathoms; Algoa Bay,
36 fathoms; off Umhloti River (Natal), 47 fathoms; off Tugela River, 65-80
fathoms (S. Afr. Mus. P.F. Coll.).

Living: False Bay, 55 metres (U.C.T.).

There is a single specimen in 8S. Afr. Mus. from Table Bay, but the record
is unreliable.

Remarks. The sulcus between suture and shoulder keel is nearly horizontal

on the early whorls, but on later whorls slopes to a varying degree, consequently
there are squat forms (U.C.T., TRA.133N) and elongate forms (S. Afr. Mus.
A4952). | :
Two specimens from False Bay (FB.952.4.U.C.T.) differ slightly in
appearance: the shoulder keel is less prominent and does not form the widest
part of the whorl, the profile of which is thus more evenly convex from suture
to suture.

In one of these two specimens the intermediate lira between the shoulder
and peripheral keels becomes from about the half of the 3rd whorl as strong as
the primary keels, so that on the 4th whorl there are 4 keels; also on 4th whorl
the lira between suture and shoulder keel becomes a keel demarcating the
sulcus, and another keel develops at the bottom of the whorl immediately above
the suture, thus the profile of the 5th whorl shows 6 keels.

The development of these additional keels is, in my opinion, only a casual
variation. .

A 5 mm. long juvenile with 4 postnatal whorls, from Algoa Bay, indicates
that Pleurotomella ida Thiele is the juvenile of capensis, although Thiele’s figure
shows only 2 spiral keels on 1st and 2nd whorls, and the 3rd whorl is a little
ambiguous at the profile.

Trophon ornatus ‘Turton (1932. p. 75, pl. 18, no. 544), 1 mm. long, is
certainly the protoconch of this or an allied species.

‘Drillia’ fultont (Sow.)
Fig. 8 g

1888. Sowerby. Proc. Zool. Soc. Lond., p. 210, pl. x1, fig. 17 (Pleurotoma f.).
1889. id. 7. Conch., vi, p. 7 (Pleurotoma f.).

1892. id. Mar. Sh. S. Afr., p. 5 (Pleurotoma (Drillia) f.).

1903. Von Martens. D. Tiefsee Exp., vii, p. 23 (Drillia f.).

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 18 (Turris f.).

1932. Turton. Mar. Sh. Pt. Alfred, p. 18 (Turris f.).

Protoconch smooth (present only in one specimen in S. Afr. Mus., and not
well preserved). Postnatal whorls 10-11. Three spiral keels, one below the
upper suture, one in middle of whorl, and one above the lower suture; the
uppermost keel not always well developed, more like a low cingulum than an:
upstanding keel; below the lowermost keel on base 4 smaller additional keels.
Between middle and lower keels minute spiral striae which form a micro-

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 119

cancellate sculpture with the fine close-set growth-iines; cancellation distinct
on base, but not in the sulcus. Periostracum hides the cancellation, and shows
only close fine pleating, retractive between suture and middle keel, protractive
below the latter. Columella with a slight swelling (not a pleat). Lip sinus
remote from suture, adjoining the middle keel. 29 (minus protoconch) x tomm.

No operculum.

Yellowish or buff, with white keels, a narrow brown band immediately
below suture, columella suffused, rostrum and edge of canal brown, perio-
stracum pale buff.

Radula without basal membrane, 15 pairs of narrow, acicular teeth, slightly
swollen in distal third, shallowly grooved, margins thickened.

Presumably dead: Port Elizabeth (Sowerby); Port Alfred (Bartsch,
Turton). Dead: St. Francis Bay, 80-100 metres (von Martens).

Off Kowie (Port Alfred), 40 fathoms, dead; Algoa Bay, 37 fathoms,
1 fresh, 1 alive; off Cape St. Blaize, 39 fathoms, 1 fresh; off Cape Infanta,
46 fathoms, 1 broken apex (S. Afr. Mus. P.F. Coll.). False Bay, living (U.C.T.).

Remarks. The radula shows that this species has been entirely misplaced
conchologically.

In general appearance there is a strong resemblance to the West African
Pleurotoma spiralis Smith 1871, and Knudsen’s description (1952. Vid. Medd.
Dansk. For., cxiv, p. 164, pl. 1, fig. 2) intensifies the likeness to such an extent
that fulton might be considered a synonym. But Knudsen places spiralis in
Asthenotoma; thus presumably it has an operculum; /fultont has no operculum
and therefore must go into a Cytharine genus.

‘Drillia’ scztecostata (Sow.)
Figs. 8 f, 20 (left)

1903. Sowerby. Mar. Invest. S. Afr., ii, p. 214, pl. 4, fig. 10 (Pleurotoma (Drillia)
$2).
1906. Smith. Ann. Natal Mus., i, p. 25 (Drillia s.).

Protoconch 2 whorls, diam. and alt. 1:25 mm., smooth. Postnatal whorls 6.
Oblique ribs slightly broader than the deep intervals, 14-15 on early whorls,
increasing to 25-26 on last whorl; upper ends coronate, slightly projecting
above the smooth sulcus, becoming obsolete on lower half of base. Base spirally
lirate. Lip sinus rather broad, moderately deep, adjoining the suture. Columella
nearly straight, canal short. 26 x8 mm.

No operculum. Buff.

Radula without basal membrane, c. 33 pairs of slender, — teeth,
shallowly grooved nearly to apex, margins thickened.

Off Glendower Beacon (near Port Alfred), 100 fathoms (Sowerby). Type
(presumably) and cotype in S. Afr. Mus.

Off East London, 80-130 fathoms, 3 dead, 1 alive; Algoa Bay, 56 fathoms,
2 dead; False Bay, 20 fathoms, 1 dead (S. Afr. Mus. P.F. Coll.).

120 ANNALS OF THE SOUTH AFRICAN MUSEUM

Remarks. Like fultoni this species must be removed from Drillia on account
of the absence of an operculum and the character of the radula.
False Bay specimen has only 21 ribs on the last whorl.

INCERTAE SEDIS— ANIMALS UNKNOWN
Drillia hottentota (Smith)

1882. Smith. Ann. Mag. Nat. Hist. (5), x, p. 208 (Pleurotoma (Clavus) h.).
1892. Sowerby. Mar. Sh. S. Afr., p. 5, pl. 4, fig. 81.

1897. id. Append. Mar. Sh. S. Afr., p. 3, pl. 8, figs. 1, 2 (burnupr).

1915. «Bartsch. Bull. U.S. Nat, Mus, 191, p21.

1921. Sowerby. Proc. Mal. Soc., xiv, p. 127 (var. fuscescens).

1932. Turton. Mar. Sh. Pt. Alfred, p. 21, pl. 4, nos. 161, 162:

1932. id. ibid., p. 21, pl. 4, no. 163 (neptunt).

Drillia layard: Sow.

1886. Sowerby. 7. Conch., v, p. 5 (Pleurotoma castanea, non Rve.).

1897. id. Append. Mar. Sh. S. Afr., p. 2, pl. 8, fig. 3 (Plewrotoma (Drillia) 1.).
1903. Von Martens. D. Tiefsee Exp., vil, p. 23.

1915. Bartsehs/loc:-ciewp. 21:

1992. LurtonsMloc emp s22)

S. Afr. Mus. has examples of both these ‘species’ identified by Tomlin.
Unless some slip has occurred, there is a transition in colouring from typical
hottentota (brown with white base, and brown with pale band) to the uniform
brown var. fuscescens and layardi. Conchologically there are no differences, and
therefore I propose to make layardi a synonym of hotientota.

Protoconch 2 whorls, diam. and alt. 1 mm., smooth but with ¢. 10 straight
axial plicae on last half whorl; 1st postnatal whorl starts with oblique riblets.
Postnatal whorls 5. Axial riblets on 1st whorl 10-11, increasing to 13—15(16)
on 5th whorl. In some specimens, both typical hottentota and fuscescens-layard1,
some (not all) of the riblets cross the sulcus and reach or almost reach the suture
(burnupi). Spiral striae absent except on the base (and here often visible only
on the lower part). Occasionally on the lower part of the intervals between the
ribs there are very faint indications of 2 (3) spiral striae.

Up to 18 mm. long (Turton).

False Bay to Port Alfred (previous records, and S. Afr. Mus.).

Off Cove Rock (East London), 22 fathoms, 1 dead but fairly fresh (brown:
Suscescens-layard:); Algoa Bay, 10 fathoms, 1 dead (hottentota coloration). (S. Afr.
Mus. P.F. Coll.)

Remarks. The absence of spiral striae distinguishes this species from
subcontracta.

D. burnupi, 10 X 4. mm. with 8 [sic, =2-+6] whorls, was stated by Sowerby
to be allied to layardi, but with the axial riblets running across the sulcus to
suture. A very worn specimen in 8S. Afr. Mus. with only 3 whorls from Tongaat

’
’

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA I2T

(go miles north of Durban) shows this character. It is dull yellowish with a
brown band below the suture, and a very faint narrow white band below the
brown, as in some examples of hottentota.

Drillia ancilla Thiele

BeeG. iluecle. D. Siefsee Exp., xvi, p. 228, pl. 37 (25), fig. 10.
1932. Turton. Mar. Sh. Pt. Alfred, p. 29, pl. 6, no. 214. (Mangilia innotabilts).

Protoconch 14 whorls, diam. and alt. 0-5-o-6 mm., smooth. Postnatal
whorls 4. Axial riblets 9 on each whorl, oblique, crossing sulcus. No spiral
striae. Lip sinus deep, adjoining suture, but with columella callus interposed
when fully developed. 4°75 X 2:1 mm.

33° 50’ S., 25° 48’ E., and 35° 26’ S., 20° 56’ E., no depth stated (Thiele).
Port Alfred (Turton).

Off Cove Rock (East London area), 22 fathoms, 2 dead (S. Afr. Mus. P.F.
Coll.).

Remarks. Agrees with hottentota and falsa in the absence of spiral striae, but
seems to be a smaller species than either of these.

Drillia subcontracta Smith

1904. Smith. 7. Malac., xi, p. 26, pl. 2, fig. 2.

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 22.

1915. id. ibid., p. 28, pl. 7, fig. 7 (Mangilia herilda).
1932. Turton. Mar. Sh. Pt. Alfred, p. 22.

1932. id. ibid., p. 29, pl. 6, no. 215 (Mangilia nereia).

Protoconch 13-2 whorls, smooth (but worn). Postnatal whorls 6. Axial
riblets 14-15 on Ist whorl, 13-14 on middle whorls, 15-16 on 5th and 6th
whorls, more or less traceable across sulcus, petering out on base. Spiral striae
over whole whorl, not crossing riblets (but no living specimens seen), 3-4 on
sulcus, (8)9—10 below in intervals between riblets on last whorl, 12-15 additional
stronger striae on base. 12 X 4°5 mm.

Amber-brown, worn specimens with paler or white ribs.

Port Alfred (Smith, Bartsch, Turton, S. Afr. Mus.).

Remarks. The S. Afr. Mus. specimens were identified by Tomlin. The
axial ribs are more numerous, and the spiral striae fewer than in diversa (Smith).

A specimen in S. Afr. Mus., identified by Tomlin as herilda, is a young
subcontracta bleached white.

Drillia diversa (Smith)
1882. Smith. Ann. Mag. Nat. Hist. (5), x, p. 207 (Pleurotoma (Clavus) d.).
Drillia bairstowi (Sow.)

1886. Sowerby. 7. Conch., v, p. 6 (Pleurotoma b.).
pog2. id: loc. cit.; p. 6, pl. 1, fig. "6.

122 ANNALS OF THE SOUTH AFRICAN MUSEUM

Drillia albonodulosa Smith
1904. Smith. 7. Malac., xi, p. 27, pl. 2, fig. 3.

S. Afr. Mus. has 3 diversa (identified by Tomlin), 1 bairstowi (identified by
Tomlin) and 3 topotypes of albonodulosa; all water-worn.

All specimens have 10 axial ribs on the whorls; the 5th whorl in bairstowi
has 11; the body whorl in albonodulosa only 6 or 7, the last part of the whorl
being ribless. In the latter the white mark on the ribs is continued as a white
band on the smooth part, but it is superficial and not deep-seated.

Spiral striae over whole whorl, 5-6 on sulcus, 14-15 (or more) below
(visible in worn specimens only between the ribs) and c. 15 stronger and more
widely spaced striae on base.

Turton (1932, p. 22) says all three species are ‘quite distinct’: ‘orange with
faint ribs and dark colouring in between’, ‘white with dark reddish streaks
between the ribs’, and ‘dark brown with a row of white nodules’. The same
colouring expressed in different words, and referring to water-worn, not
conchological, characters.

Possibly the actual types of these three ‘species’ may differ, but unless and
until the conchological differences (if any) have been stated, the above synonymy
is suggested.

See also Clavatula halistrepta Bartsch (infra, p. 140).

Drillia roust (Sow.)

1886. Sowerby. 7. Conch., v, p. 6 (Pleurotoma r.).

1892. id. Mar. Sh. S. Afr., p. 5, pl. 1, fig. 3 (bad) (Pleurotoma r.).

1906. Smith. Ann. Natal Mus., i, p. 26, pl. 7, fig. 3 (not good) (albotessellata).

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 16, pl. 4, fig. 1 (Cloonella eli zabethae).

1932. Turton. Mar. Sh. Pt. Alfred, p. 20, pl. 4, no. 154 (large specimen; ? not
the juv.).

Axial ribs 9-10 on last 2 whorls, broad, rounded, shouldered or subcuspidate
above, extending halfway across base on last whorl. Spiral striae 4—5 on last
2 whorls, and 5-6 additional on base, none on sulcus.

Sowerby: 208 mm. (4 whorls in fig.); Bartsch: 13 x 7 mm. (34 whorls
in fig.); Smith: 18 x6 mm. (3 whorls in fig.). S. Afr. Mus.: 15 x6 mm. (feebly
shouldered), 15 x 7 mm. (strongly shouldered).

Port Elizabeth (Sowerby, Bartsch); Port Shepstone, Natal (Smith) ;
Port Alfred (Turton).

Remarks. It seems clear that Sowerby, Bartsch and Turton have described
and figured examples of one and the same species. Four topotypes of albotessellata
(S. Afr. Mus. coll. Burnup) justify including this also as a synonym.

All four have only 3 whorls with portion of an earlier whorl, and Burnup’s
label says ‘always decollated’; Sowerby’s original description included the
word ‘decollata’. Smith’s ‘anfractus circiter 8’ appears to be a misprint for 3,
as shown in his figure.

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 123

Smith’s figure is not good, as the profile of the whorls is evenly convex,
without any prominent shoulder. It can, however, be nearly matched by one
of the S. Afr. Mus. specimens. Two others have strong shoulders; one of them
closely resembles Bartsch’s figure, with the colour pattern shown in Turton’s
photograph. The fourth specimen has moderate shoulders.

Turton also figured a 3 mm. specimen stated to be a juvenile of rouse. It
had protoconch 2 whorls plus 3 postnatal whorls, the 3rd whorl showing 4 axial
ribs in face view; spiral striae not visible in the photograph. Presumably

Turton had intermediate sizes connecting the juveniles with the decollated
adults.

Drillia omia n. sp.

Fig. 11 a

S. Afr. Mus. No. A8651. Eleven specimens, locality ? Port Alfred. All
more or less worn, uniform pale or dark brown, without any indication of a
darker band.

Smallest 7 mm. long with protoconch plus 3 whorls, therefore not com-
parable with Turton’s juvenile mentioned above under rousz; largest 10 mm.
long with protoconch plus 4 whorls; none with parietal callus or lip sinus.

Profile strongly shouldered and convex below the slightly concave sulcus.
Axial ribs broad, rounded, slightly protractive, g-10 on Ist and 2nd whorls,

* AS
WAS
i

\\i

\
AV

‘

Fic. 11. a. Drillia omian. sp. 6. D. sowerbyi Turton. Protoconchs of: c. D. caffra (Smith).
d. D. thetis (Smith).

124 ANNALS OF THE SOUTH AFRICAN MUSEUM

10-11 on 3rd and 4th, extending halfway across base on body whorl; 4 spiral
striae in sulcus on last whorl, 8 in the intervals between the ribs, and c. 12
additional on base (contrast rousz).

Drillia sowerbyi ‘Turton
Fig. 11 b
1932. Turton. Mar. Sh. Pt. Alfred, p. 22, pl. 4, no. 170.

Aperture 1} in spire. Protoconch 14 whorls, diam. 1 mm., smooth (worn).
Postnatal whorls 44; profile almost straight, a very slight convexity (cingulum)
below suture, and a very slightly concave sulcus, remainder slightly convex. .
10 feeble slightly oblique, protractive, axial ribs on 3rd whorl; body whorl
with 6, thereafter obsolete. Whole whorl, including sulcus, with very fine close-
set spiral striae, c. 15 on 2nd whorl, 20 on grd and 25-30 on 4th, c. 15 additional
striae (slightly farther apart) on base. Parietal callus well developed, lip sinus
deep. 1 1.X4:5)mm.

Buff, remains of periostracum dark brown. Turton: very dark brown.

Port Alfred (Turton). Off Cove Rock (East rains 22 fathoms, 1 dead
(S. Afr. Mus. P.F. Coll.).

Remarks. This single specimen (A354) is slightly worn but corresponds so
closely with Turton’s figure that it can be referred to his species. Turton makes.
no mention of the spiral striae, and the figure scarcely shows them. His specimen -
was 8 X 3°5 mm., with 2 plus 4 whorls (Turton reckoned 1 plus 5).

Drillia caffra (Smith)
Fig. 11i¢

1882. Smith. Ann. Mag. Nat. Hist. (5), x, p. 209 (Pleurotoma (Clavus) c¢.).
1892. Sowerby. Mar. Sh. S. Afr., p. 6, pl. 4, fig. 80 (bad).

1904. Smith. 7. Malac., xi, p. 27, pl. 2, fig. 4 (praetermissa).

1904. id. ibid., p. 27, pl. 2, fig. 5 (nzvosa).

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 21 (Drillia c.).

1915+ id.” ibid!) p. 22, pla. tig. 4) (lara),

1931. Tomlin. Ann. Natal Mus., vi, p. 440 (Clionella c.).

1932. Turton. Mar. Sh. Pt. Alfred, p. 20.

Aperture about 1} in spire. Protoconch 14-2 whorls (junction with 1st
postnatal whorl not clear), diam. and alt. 1 mm., smooth. Postnatal whorls 6;
profile slightly concave at sulcus, convex below, widest at the shoulder above
middle of whorl. Axial ribs oblique, 11-12 on early whorls, increasing to.
15-18 on last whorl, most prominent at shoulder, not indicated on sulcus,
petering out at middle on body whorl. No cingulum at suture, or only a very
slight one. Fine spiral striae over whole whorl, 6 at start of 1st whorl, becoming
numerous on following whorls, at least 30 on last whorl, in fresh specimens.
crossing the ribs. Lip sinus deep, with (in adult) parietal callus. Columella
reflected at the short and broad canal. 29 x9 mm. (across shoulder).

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA I25

Pale brown, mottled; beach specimens buff, pink, white, often a few dark
spots below suture. |

Port Elizabeth (Smith, Sowerby); Port Alfred (Bartsch, Turton).

Off Cape Natal (Durban), 55 fathoms, dead but fresh, and off Port Shep-
stone, Natal, 36 fathoms, 3 unicolorous pink, grey, white, dead, but 2 showing
striae; off Umhloti River mouth, 40 fathoms, dead; off Cove Rock (East
London) 80-130 fathoms, 2 dead, discoloured; off Nieca River (south of East
London), 43 fathoms, 1 dead; off Glendower Beacon (Port Alfred), 66 fathoms,
2 dead (S. Afr. Mus. P.F. Coll.).

Remarks. Specimens from Port Alfred identified as praetermissa and lara
indicate that both these are synonyms of caffra, and I would include also nivosa,
as did Tomlin, 1931.

With profile similar to that of subcontracta, but far fewer spiral striae.

The two from the East London area have very strong auriculate and
varicoid outer lips.

Drillia thetis (Smith) —
Fig. 11d

1904. Smith. 7. Malac., xi, p. 26, pl. 2, fig. 1.

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 22.

1932. Turton. Mar. Sh. Pt. Alfred, p. 21, pl. 4, no. 164 (pretiosa).
1932. id. ibid., p. 23, pl. 5, no. 175.

Aperture 14-14 in spire. Protoconch 14-2 whorls, diam. and alt. 0-6 mm.,
smooth. Postnatal whorls 5; profile convex from suture to suture (no concave
sulcus). Oblique axial ribs 11 on 1st whorl, increasing to 12-13 on last whorl,
from suture to suture, obsolescent on base. Fine spiral striae 6-7 on 1st whorl,
increasing to c. 25 (between sutures) on last whorl, crossing the ribs. No
cingulum. Lip sinus deep, with parietal callus in adult. Columella reflected
at short broad canal. 10X14 mm. (incl. outer lip).

Port Alfred (Smith, Bartsch, Turton). 33° 3’ S., 27° 57’ E. (East London
area), 32 fathoms, 2 dead but fresh; off Cove Rock (East London area), 80-100
fathoms, 3 dead (S. Afr. Mus. P.F. Coll.).

_ Remarks. A smaller species than caffra (compare protoconchs) and dis-
tinguished by the ribs starting at the suture above.

Drillia platystoma (Smith)

1877. Smith. Ann. Mag. Nat. Hist. (4), xix, p. 501 (Pleurotoma (Clionella) p.).
1889. Sowerby. 7. Conch., vi, p. 7, pl. 1, fig. 21 (Pleurotoma wilkiae).

1892. id. Mar. Sh. S. Afr., p. 4, pl. 1, fig. 4 (wilkiae).

1892. id. ibid., p. 5, pl. 4, fig. 82 (platystoma). |

1903. Smith. Proc. Mal. Soc., v, p. 363 (Drillia p.).

1903. Von Martens. D. Tiefsee Exp., vii, p. 23 (castanea, non Rve.).

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 18 (Clionella ? p.).

126 ANNALS OF THE SOUTH AFRICAN MUSEUM

1923. Melvill. Proc. Malac. Soc., xv, p. 168, pl. 5, fig. 13 (Surcula macilenta).
1925. Thiele. D. Tiefsee Exp., xvii, p. 212, pl. 35 (23), fig. 17 (distincta).
1931. Tomlin. Ann. Natal Mus., vi, p. 438 (macilenta).

1932. Turton. Mar. Sh. Pt. Alfred, p. 23 (platystoma and var. wilkiae).

Aperture 14 times in spire. Protoconch 14-2 whorls, diam. and alt. 1 mm.
(or a little less), smooth, glossy, ending with an evenly curved plica preceded
by 2 obscure ones (cf. fig. 12a). Postnatal whorls 7. Oblique ribs weak,
13-14 on body whorl. Close, fine spiral striae over whole whorl, slightly less
distinct on sulcus, on last whorl 8-9 in sulcus, 12-13 below shoulder, and c. 24
additional striae on base. 16X5:5 mm.; 175 (macilenta).

Yellowish, fuscous, pale brown or buff.

Port Elizabeth (Sowerby); Port Alfred (Bartsch, Turton); St. Francis
Bay, 80-100 metres, several dead (von Martens).

Off Umhloti River mouth, 40 fathoms, 2 juv. dead; off Cape Natal
(Durban), 62 and 85 fathoms, 3 dead; off Great Fish Point Lighthouse, 49
fathoms, 1 dead (S. Afr. Mus. P.F. Coll.).

Remarks. The Great Fish Point specimen seems to be an exact counterpart
of the Valdivia specimen figured by Thiele, but even a trifle more slender. Its
protoconch measures diam. and alt. o-8 mm.

The Natal specimens were identified as wilkiae by Sowerby.

The oblique ribs vary in strength according to the condition of the shell:
in unworn specimens they are obvious from the shoulder almost to bottom of
whorl (macilenta), in worn specimens they are reduced to more or less con-
spicuous knobs at the shoulder (platystoma and wilkiae, as far as can be judged
from Sowerby’s 1892 feeble figures).

Mangilia benjamin Bartsch (v. infra) is a broader shell (width less than 3
in length), with more numerous whorls at a shorter length.

Cf. Pleurotoma? paula Thiele
Fig. 12 a

1925. Thiele. D. Tiefsee Exp., xvii, p. 229, pl. 37 (25); fig. 2.

Two specimens are very similar to platystoma, but differ in having fewer
spiral lirae.

Aperture 14 in spire. Protoconch 2 whorls, smooth, glossy, ending with a
distinct axial, evenly curved plica preceded by 2 feeble ones. Postnatal whorls 5.
Twelve to fourteen very feeble oblique knobs traceable on 3rd to 5th whorls
at the shoulder, which is slightly above middle of whorl. Spiral striae over
whole whorl, 4 above shoulder on 2nd and 3rd, 5 on 4th and 5th whorls, 1-2
at shoulder visible between the knobs, 4 below shoulder on 2nd and grd, 5 on
4th and 5th whorls, 6 additional striae on base of same strength as those above,
followed by 6 feebler ones; the intervening lirae flat; the lira between suture
and first stria wider and stronger than those following, forming a slight cingulum,
Growth-lines sigmoid, sometimes rather prominent, due possibly to surface wear

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 127

on either side of them. In places the striae are slightly punctate (or the lirae
slightly beaded) at the intersections with growth-lines. 11 <x 4 mm.

Algoa Bay, depth not recorded, one; off Nanquas Peak (eastern part of
Algoa Bay), 49 fathoms, one (S. Afr. Mus. A8591, P.F. Coll.).

Remarks. In the small (6-3 mm.) Pleurotoma? paula Thiele, from 35° 16’ S.,
22° 26’ E., 155 metres, similar prominent sigmoid growth-lines occur at regular

ii |
inp
df Ee
Y

Fic. 12. a. Pleurotoma ? paula Thiele, body whorl, apex and protoconch further
enlarged. b. Drillia flavidula (Lam.), apex. c. D. laterculoides n. sp., body whorl and
protoconch.

intervals on the 3 postnatal whorls, which are otherwise devoid of any sculpture.
The Valdivia example was worn; possibly the present unworn specimens are
conspecific.

A worn specimen resembling Thiele’s specimen but consisting of only the
2-whorled protoconch plus one postnatal whorl was taken off Cape Recife
(34° S., 25° 44’ E. (approx. 30 fathoms)). It measures 3°25 X 1-75 mm., the
protoconch alt. 1-25, diam. 1 mm. The postnatal whorl has 9 sigmoid axial
grooves. (S. Afr. Mus. A8754, P.F. Coll.)

Drillia variabilis Smith

1877. Smith. Ann. Mag. Nat. Hist. (4), xix, p. 495.
1901. Melvill and Standen. Proc. Zool. Soc. Lond., ii, p. 441.
1917. Melvill. Proc. Mal. Soc., xii, p. 159, pl. 8, fig. 8.

128 ANNALS OF THE SOUTH AFRICAN MUSEUM

Aperture 1} times in spire. Protoconch missing. Postnatal whorls 10.
Oblique axial ribs 8 on early, 12-13 on later whorls, extending over base nearly
to canal. A feeble, narrow cingulum below suture, inconspicuously nodulose.
Below sulcus 8-10 spiral lirae crossing ribs and grooves, slightly and irregularly
nodulose, visible chiefly on base where there are c. 12 additional lirae, with finer
intermediaries. Growth-lines rather distinct on body whorl, forming a semi-
cancellate sculpture. Lip sinus broad, separated from suture by cingulum.
32 X10 mm.

Buff, with a few orange patches, and numerous orange spots on the
cingulum and lirae. ‘Freckled with pale brown’ (Melvill).

Off Cape Vidal, Zululand, 22 fathoms, 1 dead (S. Afr. Mus. P.F. Coll.).

Distribution. Farquhar Island (S. Afr. Mus.); Persian Gulf, 6-15 fathoms;
Karachi, 3 fathoms; off Bombay, 47 fathoms; Red Sea; Andaman Is.

Remarks. The Zululand specimen was identified by Sowerby.

Drillia flavidula (Lam.)
Fig. 12 b
1822. Lamarch. Anim. sans Vert., vii, p. 92.

1843. Reeve. Conch. Icon., i, pl. 8, fig. 66.
1917. Melvill. Proc. Mal. Soc., xii, p. 152.

Slender, spire elongate, aperture 1} in spire. Protoconch 2 whorls, smooth,
diam. 0°75, alt. o-8 mm., junction with 1st postnatal whorl distinct. Postnatal
whorls 10 (as preserved). Oblique axial ribs 9 on Ist, 10 on 2nd, increasing to
12 on 10th whorl, rounded at shoulder, petering out on anterior half of base.
Crossed by spiral lirae, 3-4 fine ones in sulcus, on ribs 4 (3 in early whorls)
larger followed by 4 finer lirae; c. 15 additional lirae on base, with an inter-
mediate between each of the upper 6 or 7 pairs. Cingulum present from Ist
whorl onwards, forming a somewhat angular keel below and an irregularly
crimped or beaded sutural margin above. Lip sinus moderately deep, situated
in the sulcus. Growth-lines forming the sutural crimping but indistinct across
cingulum and between the ribs, distinct in sulcus and on the base. Canal
narrow (rostral point broken). 29 x8 mm.

Pale buff, no dark blotches visible between the ribs.

Off Tongaat River (Natal), 36 fathoms, 1 dead but unworn (S. Afr. Mus.
atoll), ,

Distribution. Red Sea, Persian Gulf, China.

Remarks. Melvill (p. 153) considered flavidula a variable species. The
present specimen, described above, agrees with 2 Hong Kong examples (5. Afr.
Mus. Ross Frames don.) but has a stronger cingulum and more conspicuous
crimped sutural margin. The latter feature indicates intertincta Smith (cf.
Gravely. 1942. Bull. Madras Govt. Mus. n. s. Nat. Hist., v, 2, Pp. 75, in key), but
the present shell is much more slender than Melvill’s figure (1917. loc. cit.,
pl. 8, fig. 6), which shows moreover 3 spaced lirae crossing the ribs.

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 129

Drillia laterculoides n. sp.
Fis. 12/6

Aperture 14 in spire. Protoconch 3 whorls, alt. 1, diam. 0-8 mm., smooth,
junction with 1st postnatal whorl distinct in one specimen, not in the other.
Postnatal whorls 7, profile convex with slight shoulder. Oblique axial ribs
narrower than intervals, 10 on 1st whorl, 9 on each of the others, on 1st whorl
the first 4 or 5 are closer together than the following ribs, crossing sulcus,
petering out below periphery on base; spiral lirae obscure on first half of 1st
whorl, 4. on second half, and on each succeeding whorl, distinct on ribs but
obscure or obsolete in the intervals, 1st bounding the sulcus, 2nd peripheral,
4th almost concealed by following whorl; very fine spiral striae on sulcus and
between main lirae; on base 2 additional lirae (on the ribs) and numerous fine
striae. Growth-lines strongly sigmoid. Sulcus concave only in the intervals
between ribs. Lip sinus deep, adjoining suture. 17 <6 mm. Buff.

Off Hood Point (East London area), 49 fathoms, dead, one complete and
one apex (S. Afr. Mus. A8709, P.F. Coll.).

Remarks. Very similar to Drillia laterculata Sow. 1870 from China and
N. Australia. Corresponds better with Watson’s figure (Challenger Rep., xv,
pl. 18, fig. 5) than with Smith’s figure (Zool. H.M.S. Alert, pl. 4, figs. E, E’),
but the profile is more rounded; moreover Smith said the two (main) lirae are
continuous between and across the ribs. |

Similar also in shape to variabilis, but the axial ribs definitely cross the sulcus,
reaching the suture above.

Drillia collina n. sp.
ie aa

Aperture 13 in spire. Protoconch 2 whorls, smooth with median spiral
lira in last half whorl. Postnatal whorls 5, profile convex. Axial ribs 9 on Ist
whorl, increasing to 10-11 on last, retractive across sulcus, protractive below,
narrower than intervals; crossed by spiral lirae 2 on 1st whorl, 3 on each of
the others, the upper one in the sulcus, the middle one strongest and peripheral,
grd stronger than rst, one fine intermediary between Ist lira and suture, one
between 2nd and rd lirae, one below 3rd; 8-g additional fine lirae on base;
complanate nodules at intersections of ribs and lirae. Growth-lines sigmoid.
Sulcus not concave, with a subsutural lira forming a narrow but distinct
cingulum. Lip sinus deep, adjoining suture. 7 x 2-8-3 mm. Buff.

Off East London, 32 fathoms, one; off Hood Point (East London area),
49 fathoms, 2 and 2 apices; off Cape Natal (Durban), 85 fathoms, two (S. Afr.
Mus. P.F. Coll.). :

Remarks. The description is taken from the East London specimens
(A8587, A8710);. the Natal specimens (A8711) appear to be conspecific.

130 ANNALS OF THE SOUTH AFRICAN MUSEUM

With some similarity to Pleurotomella ida Thiele, but distinctly narrower,
and with ribs crossing the sulcus. Thiele’s figure seems to indicate a cingulum,
but it is not mentioned in the description.

Also rather like Bellardiella alfredensis ‘Turton 1932, but the latter is only
4 mm. long with 4 whorls (figure; description says 5). 3

Drillia bruchia n. sp.
Fig. 13 b “

A broken specimen consisting of three whorls, apex (? 2 or 3 whorls
missing. |

Profile sharply angular. Axial ribs 11 on 1st two whorls, 12 on the last,
scarcely indicated on sulcus except near the shoulder, petering out on lower part

Fic. 13. a. Drillia collinan. sp. b. D. bruchian. sp. c. D. latisulcus n. sp.

of base; crossed by 2 spiral lirae, with an intermediary beginning on later part
of earliest whorl, continued on the two later ones, c. 12 additional lirae on base;
lirae thin, forming sharp complanate nodules at intersections with ribs; sulcus
with distinct narrow cingulum, crossed by growth-lines; lip sinus deep,
adjoining suture. 83:5 mm. Buff.

Off Cape Natal (Durban), 440 fathoms, one dead (S. Afr. Mus. A8717,
P.F, Coil.)

Remarks. Although similar to D. collina (A8711, and A8587, A8710), this.
specimen is distinguished by the better marked cingulum, absence of lirae in
the sulcus, and the axial ribs not crossing the sulcus.

Drillia tholos n. sp.
Fig. 14
Aperture subequal to spire. Protoconch 24 whorls, dome-like, 0-3-0-4 alt.,
diam. 0-7-0°8 mm., apex smooth, 5-6 feeble close-set axial pliculae on last

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 131

quarter, glossy. Postnatal whorls 3-34, profile convex with slight rounded
shoulder. Axial ribs 13 on Ist, 11-12 on 2nd and grd whorls, crossing sulcus
and extending halfway across base; narrow, sharp; crossed by narrow, sharp
spiral lirae 2 on beginning of 1st whorl, increasing to 3, 5 on 2nd and 3rd, the
uppermost in the sulcus, the second forming the shoulder, forming tiny points
at intersections with ribs; c. 12 additional lirae on base. Growth-lines fine,
forming, with extremely fine spiral striae, a microcancellate sculpture. Lip
sinus moderate, adjoining suture. 5x2 mm. Pale brown.

Off Hood Point (East London area), 49 fathoms, 14 dead; off Cape Natal
(Durban), 11 dead (S. Afr. Mus. A8735, A8759, P.F. Coll.).

Remarks. There is no clear
distinction between protoconch
and first post-natal whorl except
a rather more conspicuous
riblet; some of the 5 or 6
pliculae may belong to the Ist
postnatal whorl.

Most of the specimens are
about 3°75 X 1°75 mm.

Fic. 14. Protoconch and Ist postnatal whorl of Drillia Distinguished from Glyphos-
tholos n. sp., with portion of upper part of last whorl ; ; :
further enlarged. toma siren Smith by the micro-

cancellate sculpture, and from
Mangilia shepstonensis by the non-tabulate axial ribs.
Drillia latisulcus n. sp.
Fig. 13 ¢

Aperture 14-14 in spire. Protoconch 2 whorls, diam. and alt. 0-5 mm.
smooth. Postnatal whorls 5, profile convex with slight shoulder. Axial ribs
(below shoulder) 11 on 1st whorl, increasing to 13-14, but evanescent on last
part of last whorl; crossed by spiral lirae 2 on Ist, 2nd and 3rd whorls, 2 and
a fine 3rd near lower suture on 4th whorl, 3 on 5th, the lowest fine, also an
intermediary between Ist and 2nd lirae, at least 12 additional lirae on base;
slight complanate nodules at intersections. Growth-lines fine; sulcus rather
wide, scarcely concave, no cingulum, without spiral lirae and not crossed by the
ribs. Lip sinus deep, adjoining suture. Outer lip submarginally incrassate.
6-5-7 X 2°3-2°5 mm. Pale brown. : |

Off Tugela River (Zululand), 65-80 fathoms, one; off Tongaat River
(Natal), 36 fathoms, one; off Umhloti River (Natal), 40 fathoms, two; off
Illovo River (Natal), 27-30 fathoms, one; off Cape Natal (Durban), 54 fathoms,
3 dead; off Cape Morgan, 47 fathoms, three; off Hood Point (East London
area), 49 fathoms, six (S. Afr. Mus. A8718-A8723, A8760, P.F. Coll.).

Remarks. In some of the specimens the intermediary lira between the Ist
and 2nd main lirae becomes as strong as the others on 4th and 5th whorls,

132 ANNALS OF THE SOUTH AFRICAN MUSEUM —

noticeably so in the Tugela River specimen and one of the Cape Morgan
specimens.

In one of the Hood Point specimens and the Tugela River specimen the
axial ribs are broader and do not exceed 12 in number.

The figured specimen is one of the Hood Point lot (A8718), which may be
regarded as type material.

Drillia perfluans n. sp.
Fig. 15 a
Aperture 14 in spire. Protoconch 14 whorls, diam. and alt. 0-75 mm.,
smooth, glossy, junction with Ist postnatal whorl distinct. Postnatal whorls 4.

al Cc

Fic. 15. a. Drillia perfluans n. sp. b. D. falcicosta n. sp. c. D. oneili n. sp.

Axial ribs 11 on 1st and 2nd whorls, 11-12 on 3rd, 12 on 4th, crossing (perfluans)
sulcus; crossed by very faintly impressed spiral striae 3 on 2nd, 4 on 3rd, 5 on
4th whorl, the first at shoulder; about 12 additional striae on base, obscure in
upper part, more distinct on lower part. Sulcus with one fine stria below suture,
and one or two others very obscure near shoulder on last whorl. Lip sinus deep,
adjoining suture. 5:75 < 2:3 mm. Pale buff. - oe

Off Hood Point (East London area), 49 fathoms, one dead (S. Afr. Mus.
A8716, P.F. Coll.).

Drillia falcicosta n. sp.
Fig. 15 D

Aperture slightly less than spire. Protoconch 2 whorls, alt. 0:5, diam.
0-75 mm., smooth, glossy, junction with rst postnatal whorl distinct. Postnatal
whorls 3, profile convex with very slight shoulder. Axial ribs 11-12 on Ist,
12-13 on 2nd, 13-15(16) on 3rd whorl, narrow, sharp, crossing sulcus and
slightly enlarged or raised at suture, slightly enlarged at shoulder but not

CONTRIBUTIONS TO KNOWLEDGE OF 8.A. MARINE MOLLUSCA 133

nodular; extremely fine and faint spiral striae between ribs, chiefly visible in
the sulcus near suture; upper part of base smooth, lower part with 6—7 lirae;
sulcus slightly concave between ribs; lip sinus deep, adjoining suture. 6 X2°5
mm.; both the Zululand specimens slightly larger, with 34-4 whorls: 7 x 3 mm.
Pale buff.

Off Umhloti River (Natal), 40 fathoms, 2 unworn:
and one apex; off O’Neil Peak (Zululand), 90 fathoms,
2 worn (S. Afr. Mus. A8724, A8725, P.F. Coll.).

Described from the unworn Natal specimens
(A8724).

Drillia morgana n. sp.
Fig. 16

Aperture 14 in spire. Protoconch 2 whorls, nucleus
depressed, alt. 0-8, diam. 1 mm., smooth, junction with
Ist postnatal whorl distinct. Postnatal whorls 34, profile
evenly convex, sutures rather deep. Axial ribs 15-16 on
Ist, 190n 2nd, ¢. 25 on 3rd whorl, narrow, not conspicuous,
on last whorl sometimes coalescent (making an exact
count difficult), crossing the undefined sulcus. No spiral
sculpture except c. 10 lirae on lower part of base. 7°5 x 3
mm. Creamy white. |
| Off Cape Morgan, 47 fathoms, one dead (S. Afr.
Fic. 16. Drillia mor- Mus. A8739, P.F. Coll.).
gana n. sp., with pro-

toconch further en-
larged.

Drillia oneili n. sp.
Fig. 15 ¢

Aperture slightly less than spire (protoconch incomplete). Protoconch
2 whorls, apex corroded (but only the nucleus seems to be missing), smooth.
Postnatal whorls 4, profile convex, scarcely any shoulder. Axial ribs 11 on Ist
whorl, 12 on 2nd and 3rd, 13 on 4th, slightly oblique, rounded, about as broad
as intervals, crossing sulcus but not the cingulum, petering out on base; crossed
by narrow spiral lirae (on 1st whorl ?), 2 on 2nd, 3 on 3rd and 4th, c. 14 addi-
additional lirae on base; in addition very fine spiral lirae over whole whorl,
c. 6 in sulcus on 4th whorl. Sulcus not concave, with narrow but distinct
cingulum. Lip sinus deep, adjoining suture. Growth-lines very faint, not
forming any cancellation with the lirae. 6:5 x2 mm. Pale buff.

Off O’Neil Peak (Zululand), 90 fathoms, one slightly worn (S. Afr. Mus.
A8731, P.F. Coll.).

Drillia pleonastica n. sp.
Pig. 17 4

Aperture slightly less than spire. Protoconch 14 whorls, alt. 0-75, diam.
I mm., smooth, junction with 1st postnatal whorl not distinct. Postnatal whorls

134 ANNALS OF THE SOUTH AFRICAN MUSEUM

4, profile with slight peripheral angle. Spiral lirae 3 on first half of 1st whorl,
middle one the most prominent and continued slightly below middle of whorl
as the peripheral lira on succeeding whorls; above peripheral lira 4 lirae on
2nd whorl, 6 on grd, 7-8 on 4th; below periphery 2 on 2nd, 3 on 3rd, 4-5 on
4th whorl; c. 15 additional lirae on base. No axial ribs, but growth-lines
distinct, strongly sigmoid, producing a cancellate sculpture on the sulcus, and a

Ns testes

eso SN

Sy SSN
Lat
Ce Ee

Fic. 17. a. Drillia pleonastica n. sp. 6. D. spiralis n. sp. Protoconchs further enlarged.

=
Y \)

beaded sculpture below periphery and on base. Lip sinus deep, remote from
suture. 7X2°5 mm. Pale buff.

34° 26’ S., 25° 42’ E., 124 fathoms, one dead (S. Afr. Mus. A8565, P.F.
Coll.).

Drillia spiralis n. sp.
Fig. 17 b

Aperture subequal to spire. Protoconch 24 whorls, very finely spirally
striate. Postnatal whorls 3, profile convex, with slight but definite shoulder.
Obliquely protractive axial ribs 10 on 1st whorl, 9 on 2nd and 3rd whorls,
traceable across base, not crossing sulcus; crossed by spiral lirae (below sulcus)
6 on 1st whorl, c. 10 on 2nd, ¢. 12-14 on 3rd, additional lirae on base at least
20; sulcus scarcely concave, with 4 spiral lirae on 2nd whorl, 6 on 3rd. Lip
sinus deep, adjoining suture. 72:5 mm. Buff.

Off Cape St. Blaize, 125 fathoms, one dead (S. Afr. Mus. A8583, P.F.
Coll.).

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 135

Remarks. Differs from Haedropleura dora Thiele (see Cythara alfredi) by the
more prominent shoulder, less prominent ribs on last whorl, and the ribs not
crossing the sulcus.

Drillia fossata Sow.
Fig. 18 a

1903. Sowerby. Mar. Invest. S. Afr., 11, p. 214, pl. 3, fig. 5 (Pleurotoma (Drillia) f.).

Aperture subequal to spire. Protoconch 2 whorls, diam. and alt. 1 mm.,
smooth. Postnatal whorls 6. First whorl with 14 short oblique ribs, broader
than intervals, increasing to 30 on body whorl, forming a sharp, coronate or
undulate keel above, bounding the deeply concave sulcus; on middle whorls
the ribs cross the 1st and more or
less the 2nd spiral lirae, but on
later whorls peter out below, and
on body whorl are scarcely indi-
cated even on ist lira. 2-3 strong,
spaced spiral lirae below keel, and
c. 10 additional ones on base;

ad mre lirae in profile tabulate above,
Fic. 18. a. Drillia fossata Sow. b. D. simplici-

cingula n. sp., protoconchs sloping below. Lip sinus deep,
adjoining suture. 216-5 mm.

(Sowerby: 22 X 7 mm..).

Pale buff, the markings on the type described by Sowerby now faded.

Off Cape Vidal, Zululand, 80-100 fathoms (Sowerby). Type (with
Sowerby’s label) (fresh) and cotype (dead) in S. Afr. Museum, also broken
apices of 4 examples and fragments of others; off Cape Natal, 85 fathoms,
fragments; off Hood Point (East London), 49 fathoms, 5 broken apices;
34° 3'S., 25° 10’ E. (St. Francis Bay), 24-34 fathoms, 1 dead but fresh (S. Afr.
Mus. P.F. Coll.).

Remarks. The presence of a specimen in fresh condition as far west as
St. Francis Bay is very surprising. There were no fragments of any other
examples in the bottom-sample (dredged March 1899), as there were in the
Cape Vidal sample (Febr. 1901) from which the type and cotype had been
originally picked out. The Cape Natal (Dec. 1900) and Hood Point (July 1901)
samples contained only fragments.

Drillia simplicicingula n. sp.
Fig. 18 5

Similar in shape to fossata. Aperture equal to spire. Protoconch 14 whorls,
diam. and alt. 1 mm., smooth. Postnatal whorls 4; ist sharply demarcated
from protoconch, with 6 spiral lirae, of which the uppermost continues on all
whorls as a sharp keel below the suture, the 2 lirae below this (in the sulcus)
soon disappear, the lower 3 continue on to 2nd whorl; on 3rd whorl 4 lirae
below sulcus, on last whorl 7, with c. 15 additional ones on base; lirae in profile

136 ANNALS OF THE SOUTH AFRICAN MUSEUM

tabulate above, sloping below (as in fossata). Lip sinus broad and deep,
separated from suture by the subsutural keel. 13 X5 mm.

Pale buff, columella orange-brown, in one specimen aperture internally
pinkish, protoconch glossy.

Off Hood Point (East London), 49 fathoms, 2 dead but unworn, 1 broken;
off Sandy Point (Cape Morgan, East London area) 57 fathoms, 1 dead but
unworn; off Cape Morgan, 47 fathoms, one and 2 fragments; off Great Fish

Fic. 19. a. Drillia dolorosa Thiele. 6. D. diasi n. sp.

Point, 51 fathoms, 1 broken ((S. Afr. Mus. A3471, A8666, A8667, A8708, P.F.
Coll.).

Remarks. Differs from fossata by the complete absence of any axial ribs, and
by the different spiral sculpture.

The largest specimen (A8666, Hood Point) may be taken as the type; the
smaller Sandy Point specimen (A3471), also with 4 whorls but only 10 X 4 mm.,
as a cotype. The broken Great Fish Point specimen had 5 postnatal whorls.

Drillia lea Thiele

1925. Thiele. D. Tiefsee Exp., xvii, p. 226, pl. 37 (25), fig. 5.

Protoconch 14 whorls, smooth. Postnatal whorls 4 (6 mm. shell), 54
(8 mm. shell). Spiral lirae 3 on 1st whorl, increasing to 5 on 4th and 6-7 on
5th, 15-20 additional lirae on base of 5th whorl. No axial riblets, growth-lines
indistinct. Sulcus 4 length of whorl. Lip sinus deep, adjoining suture.

CONTRIBUTIONS TO KNOWLEDGE OF 8.A. MARINE MOLLUSCA 137

6X 2°5 mm. (4 whorls), 8:5 x3°5 mm. (fragment: 2nd—5th whorls); Thiele:
8 x 3 (54 whorls).

Be to §., 22° 26’ E., 155; metres. (Thiele). 94° 26’ S., 25° 42’ E., 124
fathoms, one; and off Cape St. Blaize, 125 fathoms, one and one fragment
(S. Afr. Mus. A8567, A8582, P.F. Coll.).

Remarks. Like the Valdivia specimens, the P.F. specimens from bottom
samples are corroded, and the spiral lirae rather difficult to count.

Another specimen (A8585) from the same bottom sample off Cape St. Blaize
is distinctly broader than typical Jee, and has only 2 spiral lirae on the 1st and
2nd whorls, 3 on 3rd, and 4 on 4th, with 2 very fine intermediaries. 6 X 2°75 mm.

Drillia dolorosa Thiele
Fig. 19 a

1925. Thiele. D. Tiefsee Exp., xvii, p. 227, pl. 37 (25), figs. 6, 6a.

Protoconch 2 whorls, diam. 0-6 mm., smooth. Postnatal whorls 4 (Thiele:
4%), profile convex, shoulder not prominent. No axial ribs. Spiral lirae (below
sulcus) 3 on 1st whorl, 4 on 2nd, 5 on 3rd and 6 on 4th, ¢. 15 additional lirae
on base, very faint towards extremity. Growth-lines obscure. Sulcus scarcely
concave, with 2 fine spiral lirae on 2nd whorl, 3-4 on 3rd and 4 on 4th. Lip
sinus deep, adjoining suture. 7°3 X2:75mm.; Thiele: 10-6 x 3:5 mm. Pale buff.

35° 16’ S., 22° 26’ E., 155 metres (Thiele).

Off East London, 32 fathoms, one dead (S. Afr. Mus. A8588, P.F. Coll.).

Drillia diasi n. sp.
Fig. 19 b

Aperture 1} in spire. Protoconch 2 whorls, diam. and alt. 0-75 mm.,
smooth. Postnatal whorls 4. No axial ribs. Spiral lirae (below sulcus) 2 on Ist
whorl, 3 on 2nd, 4 on 3rd, becoming 5 on the later part, 5 on 4th, at least 15
additional lirae on base, faint towards extremity. Growth-lines fine. Sulcus
broad, nearly 4 length of whorl, scarcely concave, with fine spiral lirae, 40n
body whorl. Lip sinus deep, adjoining suture. 93°75 and 62-5 mm.
Creamy white. ;

34° 26’ S., 25° 42’ E., 124 fathoms, two dead (S. Afr. Mus. A8566, P.F.
Coll.).

Remarks. Close to lea but with a broader sulcus, a slight but definite
shoulder, and a larger protoconch; has only 3 whorls at same length as lea with
4 whorls.

Drillia armilla n. sp.

Protoconch 24 whorls, diam. 0-75 mm., smooth. Postnatal whorls 4,
profile convex, without shoulder. No axial ribs. Spiral lirae 3 on 1st whorl,
3 on 2nd becoming 4 on later part, 4 on 3rd becoming 5 on later part, 5 on 4th
whorl, 18-20 additional lirae on base. Growth-lines very fine, subordinate to

138 ANNALS OF THE SOUTH AFRICAN MUSEUM ©

the lirae. Sulcus § length of whorl, scarcely concave, with fine spiral lirae 2 on
ist and 2nd whorls, 3 on 3rd and 4th. Lip sinus deep, adjoining suture. Outer
lip submarginally incrassate. 7:5 3:5 mm. Buff.

Off Cove Rock (East London area), 80-130 fathoms, one dead (S. Afr.
Mus. A8714, P.F. Coll.).

Remarks. Similar to lea but proportionately broader; the larger proto-
conch also indicates a different species. The lirae are sharply defined and well
separated, without intermediaries.

Close to diast but whorls more convex.

Drillia pselia n. sp.

Aperture subequal to spire. Protoconch 2 whorls, diam. 1 mm., smooth.
Postnatal whorls 4, profile convex. Spiral lirae 2 on first quarter of 1st whorl,
3 on remainder, 3 on 2nd, 4 on 3rd, and by duplication and interpolation 6 on
last whorl; slight thickenings or nodules appear on the lirae on 2nd and 3rd
whorls, but there are no definite axial ribs; at least 25 fine additional lirae on
base. Growth-lines for the most part obscure. Sulcus about 4 length of whorl,
with 2 spiral lirae on 1st whorl, 3 on 2nd, 4 on 3rd and 4th. Lip sinus deep,
adjoining suture. 11 X4°5 mm. Buff.

Off Cape St. Blaize, 125 fathoms, one and 2 broken specimens (S. Afr.
Mus. A8715, P.F. Coll.).

Remarks. A species with sculpture similar to that of /ea but larger, and with
larger protoconch.
Drillia (Cymatosyrinx) eva Thiele
Fig. 20 (right)
1925. Thiele. D. Tiefsee Exp., xvii, p. 228, pl. 37 (25), fig. 9.

Aperture 1? in spire.
Protoconch 14 whorls, diam.
0-5, alt. o-6 mm., smooth,
polished. Postnatal whorls
5. Broad oblique axial ribs,
wider than the deep intervals,
12 on last whorl, upper ends
becoming increasingly pro-
minent forming a coronate
shoulder. No spiral sculp-
ture. Sulcus deeply concave.
Lip sinus deep, adjoining
suture. 72:5 mm. (Thiele

and S. Afr. Mus.).

Fic. 20. Right: Drillia eva Thiele. Left: apex of ; :
‘Drillia’ scitecostata (Sow.), to same scale, 35° 16’ 8., 22° 20 E.

155 metres (Thiele). Off
Cape St. Blaize, 125 fathoms, 1 dead (S. Afr. Mus. A8561, P.F. Coll.).

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 139

Remarks. At first sight the sculpturing suggests a young scztecostata, but the
protoconch is smaller and the shell narrower.

Bela alma Thiele
Fig. 21 a
1925. Thiele. D. Tiefsee Exp., xvii, p. 227, pl. 37 (25), fig. 13.

Aperture 14 in spire. Protoconch 2 whorls, smooth, last half whorl with
median spiral lira. Postnatal whorls 4. Axial ribs slightly oblique below sulcus,
¢. 10 (obscure) on 1st whorl, 12 on 2nd, 15-16 on last whorl; crossed by spiral
lirae on each whorl, feeble knobs at intersections, on body whorl an additional
lira at top of base with small knobs, below which the ribs cease; g—10 fine
additional lirae on base. Growth-lines obscure. Sulcus not concave, not crossed
by the ribs, no cingulum. Lip sinus deep, adjoining suture. 72:5 mm. Pale
brown.

35° 26’ S., 20° 56’ E., depth not recorded (Thiele).

Off Illovo River (Natal), 27-30 fathoms, one; off Tugela River (Zulu-
land), 65-80 fathoms, two (S. Afr. Mus. A8712, A8713, P.F. Coll.).

Remarks. Thiele does not mention the actual number of spiral lirae and
axial ribs; his figure shows 4 noduliferous lirae on body whorl. In spite of this
the present specimens, which agree in dimensions, may be provisionally
referred to his species.

Fic. 21. a. Bela alma Thiele, with protoconch. 6. B. anna Thiele. c. B. bella n. sp.
d. Asthenotoma eva (Thiele). e. Acrobela acus n. sp., with protoconch.

140 ANNALS OF THE SOUTH AFRICAN MUSEUM

Bela anna Thiele
Fig. 21 5
1925. Thiele. D. Tuefsee Exp., xvu, p. 228, pl: 97 (25),4ie. 1

Aperture 14 in spire (Thiele’s figure: 14). Protoconch 14 whorls, diam.
0-5 mm., smooth (corroded). Postnatal whorls 4 (shell 6 mm.), Thiele: 5}
(shell 10-5 mm.), profile convex with slight shoulder. Axial ribs below shoulder
low and rounded, 13 on 1st and 2nd whorls, 15 on 3rd, slightly curved and
protractive, becoming obscure on 4th whorl; crossed by spiral lirae 2 on Ist
whorl, the upper forming the shoulder, increasing to 5 on 4th whorl (Thiele’s
figure: 6 or 7 on 5th whorl), c. 7 additional lirae on upper part of base (lower
part corroded); growth-lines distinct. 62:5mm.; Thiele: 10°5 4mm. Buff.

5 10 9.) 22> 26 E.. 155, mctres)( Mnicle)”

Off Cape St. Blaize, 125 fathoms, one dead (S. Afr. Mus. A8584, P.F.
Coll.).

Remarks. The Pieter Faure specimen seems to be identical with, though
smaller than, the Valdivia one.

Bela bella n. sp.
Fig. 21 ¢

Aperture 1% in spire. Protoconch 2 whorls, diam. 0-75 mm., smooth
(corroded). Postnatal whorls 5, profile convex with slight shoulder. Growth-
lines forming narrow, sharp axial ribs 14-15 on 1st whorl (more or less corroded),
increasing to 20-22 on body whorl, strongly curved in sulcus, protractive below;
spiral lirae below sulcus 4 on 1st and 2nd whorls, 4-5 on 3rd, 6—7 on 4th and
7-8 on 5th, c. 15 additional lirae on base; sulcus scarcely concave, with 2 lirae
on 2nd whorl, 3 on grd, 3-4 on 4th and 4 on 5th. Lip sinus deep, adjoining
suture. 10X3°5 mm. Pale buff.

Brown’s Bank (approx. 364° S., 21° E.) on Aguihas Bank, 80-100 fathoms,
3 dead (S. Afr. Mus. A8639, P.F. Coll.).

Remarks. Somewhat resembling anna, but distinguished by the sharper,
more oblique axial ribs, which cross the sulcus.

Clavatula halistrepta Bartsch

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 19, pl. 2, fig. 5.
1932. Turton. Mar. Sh. Pt. Alfred, p. 19, and var. albocincta, pl. 4, no. 149.
1932. id. ibid., p. 19, pl. 4, no. 150 (hera).

One very worn topotype Port Alfred (ex Turton), in S. Afr. Mus. I have
seen one specimen from Jeffrey’s Bay agreeing with Bartsch’s description:
6 whorls. 28 x 10 (across outer lip) mm., aperture 12 mm.

Although larger there are sculptural resemblances between this and Drillia
diversa (Smith).

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 141

Clavatula semicostata Kiener

Fig. 5 f
1848. Krauss. Stidafr. Moll., p. 109.
1892. Sowerby. Mar. Sh. S. Afr., p. 6.
1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 17.
1915. id. ibid., p. 23, pl. 2, fig. g (Drillia halidoma).
1932. Turton. Mar. Sh. Pt. Alfred, p. 15.

Not: Von Martens. 1903. D. Tiefsee Exp., vii, p. 24. } Dols 2
Thiele. 1925. ibid., xvii, p. 212, pl. 35 (23), fig. 10.

Aperture 14 to nearly 2 in spire. Protoconch ? Postnatal whorls g (10).
Oblique axial ribs 11-12 on early whorls, 13-14 on later whorls, beginning at
suture, on body whorl extending halfway across base, shoulder slightly above
middle of whorl; suture embracing lower ends of ribs, producing a more or less
nodular cingulum; sulcus rather deeply concave; fine spiral striae, close
together on sulcus, farther apart below, 4 on early whorls, increasing to about
10 on later whorls, usually visible only in the grooves between ribs. Lip sinus
deep and narrow, a little distance from suture. 48 X15 mm.

White or pinkish, periostracum yellow-brown.

Cape (Krauss, Bartsch); Port Alfred (Turton). Durban, Port St. Johns,
Hermanus, Kalk Bay (False Bay), and False Bay, 9 fathoms (S. Afr. Mus.).

Remarks. The Durban and the 48 mm. False Bay specimens are fresh and
retain the periostracum.

Bartsch’s 18-6 mm. halidoma can be matched with juvenile shells in S. Afr.
Museum.

Bergh (1895. Nov. Act. K. Leop-Carol. D. Ak. Naturf., xv, no. 2, p. 192,
pl. 13, figs. 293-6) gave a full description of the anatomy of three specimens
from Saldanha Bay. The radula has a central and a lateral plate resembling
those of Clavatula taxus and sinuata, but Bergh makes no mention of an appendage
to the lateral plate. The species seems to be correctly placed in Clavatula
(Clionella), if the identification (or alternatively the locality) is correct. Up to
the present there are no other records of this species from the west coast, where
the only species seems to be sinuata and its var. sigillata.

Clionella confusa Smith
Fig. 5 ¢

1906. Smith. Ann. Natal Mus., i, p. 23, pl. 7, fig. 2.

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 15 (? part only of colour vars.).
1932. Turton. Mar. Sh. Pi. Alfred, p. 16 (? pl. 3, no. 130, juv.).

1932. id. ibid., p. 16, pl. 3, no. 132 (kowvensis), and p. 17 (k. var. viridis).

Aperture 1} to nearly 2 in spire. Protoconch ? Postnatal whorls 8 (9),
profile nearly straight. Oblique axial ribs 14-16 on early whorls, 18-20 on

142 ANNALS OF THE SOUTH AFRICAN MUSEUM

later whorls, on body whorl extending across base to canal. Suture undulate,
embracing lower ends of ribs, producing a strong, slightly nodular cingulum.
Sulcus deep. Lip sinus narrow, forming the deepest part of the concave sulcus.

Growth-lines on cingulum more oblique to the suture than in sinuata.
Spiral striae (when traceable) numerous but inconspicuous, best seen on early
whorls. Up to 40 mm. (apex missing); Smith and Turton give 45 mm.; a
specimen (minus protoconch and ? 2 whorls) 36 x 10 mm.

Brown, weathering to orange-red, the cingulum in weathered specimens
usually paler, periostracum dark brown; var. viridis olive-green with white
sutural band.

Port Elizabeth (Smith, Bartsch); Port Alfred (Bartsch, Turton). Port
St. Johns, Port Alfred, Jeffreys Bay, Mossel Bay (S. Afr. Mus.).

Remarks. Described from a worn specimen. All S. Afr. Mus. specimens are
also worn, but some retain portions of the periostracum.

In general, a narrower shell than sinuata, with usually a few more ribs, but
best distinguished by the lip sinus. Smallest specimen in S. Afr. Mus. 15 mm.

Clionella rosaria Rve.

1848. Krauss. Stidafr. Moll., p. 109.

1892. Sowerby. Mar. Sh. S. Afr., p. 6.

1906. Smith. Ann. Natal Mus., i, p. 23 (comparison with confusa).

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 15.

1915. id. ibid., p. 15, pl. 7, fig. 8 (sybaritica). ;

1932. Turton. Mar. Sh. Pt. Alfred, p. 16, pl. 3, no. 131 (sinistral specimen).
1932. id. ibid., p. 17 (sybaritica).

Distinguished conchologically from confusa by the slightly narrower (in
equal-sized specimens) cingulum, and stronger development of spiral striae.
These are numerous (at least a dozen on middle whorls) and in live or fresh
specimens probably cross the ribs.

The ‘grating’ effect (microcancellation) referred to by Bartsch and Turton
in sybaritica is not a specific character; the strength of growth-lines and striae
may vary, but the effect is due to weathering; it may be seen in typical examples
of rosaria.

Protoconch ? Postnatal whorls 7 (8). Oblique axial ribs 14 on early
whorls, on later whorls not exceeding 16 normally; Bartsch mentions an
increase from 14 on 6th whorl to ‘about 20’ on 7th, but ribs are often duplicated,
which probably explains the high number 20. |

28 (64 whorls present, apex broken) X 10 mm.

Weathered specimens variously coloured, usually rose or salmon, and
flecked with darker red or brown spots, cingulum white with brown marks;
rarely unicolorous rose with white cingulum.

Cape (Krauss); Port Alfred (Bartsch, Turton, S. Afr. Mus.).

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 143

Remarks. Knudsen (1952. Vid. Medd. Danks. For., cxiv, p. 135) states,
without reference, that Krauss recorded rosacea Rve. [sic] from Algoa Bay. In
the above reference Krauss recorded rosaria from the Cape coast.

Clavatula (Clonella) tripartita Weink.

Fig. 4 d
1877. Smith. Ann. Mag. Nat. Hist. (4), xix (bipartiia).
1892. Sowerby. Mar. Sh. S. Afr., p. 6, pl. 4, fig. 83.
1902. Smith. 7. Conch., x, p. 115, pl. 1, fig. 7 (parilis).
1903. id. Proc. Mal. Soc., v, p. 362 (parilis).
1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 14 (bipartita) and p. 248 (tripartita).
1915. id. ibid., p. 20, pl. 8, fig. 2 (helena) (figs. 2 and 3 transposed).
1932. Turton. Mar. Sh. Pt. Alfred, p. 15, pl. 3, no. 128 (bipartita).

Protoconch 14 whorls, alt. and diam. 1 mm., with fine close-set axial
pliculae, junction with Ist postnatal whorl abrupt.!* Postnatal whorls 8 (9).
Oblique axial ribs well marked on shoulder of whorls, especially on early
whorls. Fine spiral striae on early whorls (if not too worn). Cingulum more
(parilis) or less (tripartita) tumid, a little wider than the smooth sulcus, the latter
well defined (éripartita) or less defined (parilis). Lip sinus deep and narrow.
Lirae on base of last whorl sometimes feebly nodulose; in one parilis specimen
the base is nodulose in axial series, white nodules on orange ground-colour,
in other words the axial ribs peter out below the shoulder but reappear as
nodules on the lower base. 46 (minus protoconch) x 16 mm.

Operculum (in Sowerby’s figure) oval, nucleus in middle of inner margin.

Buff, greyish, or dull orange, fine arcuate axial red-brown or orange lines
and flames, often more conspicuous on cingulum than on rest of whorl.

Off Durban, 40 fathoms, from fish stomach (Smith: parilis); Port Alfred
(Bartsch, Turton); Port St. Johns (S. Afr. Mus.). Also several specimens
without locality, most of them shiny, none with protoconch (S. Afr. Mus.) ;
3 typical parilis slightly worn, probably ex pisce off Durban (S. Afr. Mus.).

Off Umtwalumi, Umkomaas, and Itongazi (Natal), 13-50 fathoms; and
Algoa Bay, 10-16 fathoms, all dead. (S. Afr. Mus. P.F. Coll.)

Remarks. The series in S. Afr. Mus. leaves no doubt that parilis is synony-
mous. There are plump and slim forms.

The species seems to have its main habitat in the Natal area, where it is
obtained from fish stomachs by the trawlers. More or less worn specimens are
washed up farther south at Port St. Johns, Port Alfred, and Port Elizabeth.

Turton (p. 16) says he found it alive in the Lagoon at Port Alfred—a rather
remarkable statement which needs confirmation.

Dautzenberg (1912. Ann. Inst. Océan., vol. 5, fasc. 3, p. 10) records tripartita
from south of Senegal, and from Tiger Bay (Angola). This distribution is
unexpected, but cf. Nassa demoulioides.

14 Only one specimen, slightly worn, showing traces of axial pliculae. In two other specimens
the protoconch is worn quite smooth.

144 ANNALS OF THE SOUTH AFRICAN MUSEUM

Clavatula turriplana Sow.

1903. Sowerby. Mar. Invest. S. Afr., 11, p. 215, pl. 3, fig. 6 (Pleurotoma (Clavatula)
t.)

1906. Smith. Ann. Natal Mus., i, p. 24 (listed).

Aperture 13 times inspire. Protoconch 24 whorls, smooth, but with growth-
lines on last half whorl, diam. 1, alt. 1-25 mm., junction with Ist postnatal whorl
distinct. Postnatal whorls 10. Oblique axial ribs below sulcus on first 4 or
5(6) whorls, but becoming obsolete on later whorls. Fine spiral striae on all
whorls. No cingulum below suture. Lip sinus broad, moderately deep.
42x11 mm. (Type); another specimen minus protoconch 43 x 12 mm.

Buff, shoulder of each whorl with a very indistinct pale or white band
(Type) or an irregular series of pale patches; protoconch white, glossy.

Off Cape St. Blaize, 85 fathoms (Sowerby). Off Cape St. Francis, 75
fathoms (43 mm. fresh); off Cape Point, 80 fathoms (24 mm., corroded);
36° 40’ S., 21° 26’ E., 200 fathoms (24mm. corroded) (S. Afr. Mus. P.F. Coll.).

Type in S. Afr. Mus.

Remarks. The corroded specimen from 200 fathoms has 6 whorls rather
angulate at the shoulder, the last (body) whorl smoothly rounded as in the other
specimens.

Genus uncertain. The shape of the shell resembles some of the dextral

species (e.g. thalaea Dall) of Antiplanes (west coast of N. America).
Clavatula (Surcula) opulenta Thiele
1925. Thiele. D. Tiefsee Exp., xvii, p. 226, pl. 36 (24), fig. 15.

Aperture subequal to spire. Protoconch (apud Thiele) smooth. Anterior
canal long. Shoulder slightly above middle of whorl, slightly angular in early
whorls, rounded in later whorls. Oblique axial ribs from shoulder, petering out
on base, 15 on early whorls, increasing to 18-20, width subequal to grooves;
fine close spiral striae crossing ribs, slightly stronger on base; growth-lines not
conspicuous. Lip sinus broad and deep. 39 (protoconch and 2 or 3 whorls
missing, only 6 remaining) X 12 mm.

35° 16’ S., 22° 26’ E., 155 metres (Thiele). Off Cape Point, 230 fathoms,
1 broken and corroded; 3 without locality (probably off Cape Point),
18-39 mm., also broken and corroded. (S. Afr. Mus. P.F. Coll.)

Remarks. Very like anteridion, but with more numerous, sharper, and longer
ribs.

Clavatula (Surcula) anteridion Watson
1881. Watson. 7. Linn. Soc. Lond., xv, p. 399.
1886. id. ‘Challenger’ Rep., xv, p. 295, pl. 19, fig. 6.
1903. Smith. Proc. Mal. Soc., v, p. 363 (listed).
1925. Thiele. D. Tiefsee Exp., xvii, p. 223, pl. 37 (25), fig. 3 (Surcula sp. juv.).
? not Thiele. ibid., p. 222, pl. 36 (24), fig. 13 (? anteridion from New Amsterdam)-

—aSTY

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 145,

Aperture subequal to spire. Anterior canal long. Protoconch 2 (24)
whorls, diam. and alt. 1:-3-1-5 mm., smooth, but with fine growth-lines, junction
with Ist postnatal whorl distinct. Postnatal whorls 8-9. Shoulder in middle of
whorl, angular in early whorls, rounded in later whorls. Oblique axial ribs from
shoulder, petering out on base, 13 on 1st whorl, increasing to 15 on last whorl,
subequal in width to the grooves; fine close spiral striae crossing ribs, slightly
stronger on base; growth-lines not conspicuous. Lip sinus broad and deep.
42 X11 (across shoulders of last whorl, 14 if aperture included).

Operculum ovate, nucleus apical, 6 x 3-5 mm. in 36 mm. shell.

Cream or pale buff; protoconch white, glossy; operculum horny.

Off Cape Point, 150 fathoms (Watson); 35° 32’ S., 18° 20’ E., 2,750 metres.
(Thiele).

Off Cape Point, west coast of Cape Peninsula, and Table Bay, 180-230
fathoms. 15 specimens, smallest 15 mm. long, most of them corroded, one with

operculum (S. Afr. Mus. P.F. Coll.).

Remarks. The identity of these specimens is not in doubt. They are dis-
tinguishable at a glance from opulenta by the fewer ribs. In this respect they
agree with Watson’s figure, though he said in his description there were ‘about
nineteen of these ribs and hollows on the last whorl’; but 19 is the number
found in opulenta! I prefer to accept his illustration.

Unfortunately the animals in the fresh specimens, even the one retaining
its operculum, were completely decomposed, and no radula was obtained.

Surcula sulcicancellata n. sp.
Pig429; ¢

Aperture subequal to spire. Protoconch 134 whorls, smooth. Postnatal
whorls 6; profile rounded but with a slight shoulder. Oblique axial ribs 18
on early whorls, 20-21 on later whorls, not quite reaching suture of succeeding
whorl, and becoming feeble and obsolescent on body whorl, upper ends
projecting above sulcus and somewhat tubercular; spiral lirae 6 on early
whorls, becoming more numerous on later whorls, with (especially on body
whorl) finer intermediaries. Sulcus concave, with regular retrorse curved axial
plicae, 3-5 times as numerous as the ribs, crossed by spiral lirae forming a
cancellate sculpture, somewhat more conspicuous on the early whorls. Lip sinus
wide but shallow. Canal short. 31X12 mm. White.

Off Cape Point, 130-300 fathoms, 5 dead (S. Afr. Mus. A361, A3477,
A3481, A3495, P.F. Coll.). ;

Remarks. These five specimens are very similar to S. obliquicosta von Martens.
(1903. D. Tiefsee Exp., vii, p. 80, pl. 2, fig. 1) from off Sumatra, 1,143 metres,
but rather broader proportionately to length, and with distinct cancellate
sculpture on the sulcus. Von Martens said the retrorse pleats on the sulcus were
characteristic of his species, but similar pleats occur in S. profundorum Smith 1896.

146 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fic. 22. a. Surcula amplisulcus n. sp. b. S. faurein. sp. c. S. sulcicancellata n. sp. d. S. scalaria n. sp.

(and 1898. Illustr. R.I.M.S. ‘Investigator’ Moll., pl. 7, figs. 2, 2a), from the
Maldive Islands, 719 fathoms.

Judging from von Martens’s figure these pleats do not seem to be so
numerous as in the present species, and they do not extend across the whole
sulcus to the tubercle-like ends of the oblique ribs; consequently the sulcus is
not cancellate; nor is it cancellate in profundorum.

Surcula scalaria n. sp.

Fig. 22 /d

Aperture a little shorter than spire. Protoconch 24 whorls, diam. and alt.
c. 0-6 mm., apex very small, diam. 0:25 mm., smooth. Postnatal whorls 8,
turreted, with prominent tubercular shoulder. Tubercles 12-13 on early
whorls, 11-12 on middle whorls, increasing to 14-15 on body whorl, oval or
slightly oblong (axially) strongest at the shoulder, weaker below and just
reaching suture of succeeding whorl, indicated on base by slight swellings on
the spiral lirae; crossed on early whorls by 2 (in one specimen 3) spiral lirae,
which become obsolete on later whorls, but the lower one traceable and making
the profile of the tubercles slightly rectangular; spiral lirae well spaced on upper
part of base, closer together below, no intermediaries. Suture undulate. Sulcus
broad, forming nearly half the whorl, crossed by fine growth-lines only. Lip
sinus broad and rather deep. Canal short. 24 10mm. White.

Off Cape Point, 480-800 fathoms, 7 dead (S. Afr. Mus. A358, A360,
A1789, A3474, A3476, P.F. Coll.). |

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 147

Surcula amplisulcus n. sp.
Fig. 22 a

Aperture subequal to spire. Protoconch 14 whorls, diam. and alt. 1 mm.,
smooth (but somewhat corroded). Postnatal whorls 7; profile of early whorls
prominently shouldered, less prominently on later whorls, and rounded on body
whorl. Oblique axial ribs c. 18 on middle whorls, not reaching suture below;
crossed by well-marked spiral lirae, 2 on 3rd and 4th whorls, increasing to
5-6 on last whorl, c. 10 additional lirae on base, without or with only feeble
intermediaries. On body whorl the axial ribs indicated only by slight thickenings
on the lirae. Sulcus broad, forming half the whorl in early whorls, nearly half
in later whorls, crossed by fine growth-lines only. Lip sinus broad and
moderately deep. Canal long. 29x11 mm. White.

Off Table Bay, 196 fathoms; off west coast of Cape Peninsula, 120
fathoms; 36° 40’ S., 21° 26’ E. 80-100 fathoms. 5 dead (S. Afr. Mus. A1680,
mages, 499907, PF. Coll.)

Surcula faurei n. sp.
Fig. 22 5

Aperture subequal to spire. Protoconch corroded. Postnatal whorls 5,
profile rounded, without shoulder. 1st whorl corroded; slightly oblique axial
ribs 11 on 2nd whorl, 12 on 3rd, 15 on 4th and 16-17 on 5th (obscure on later
half of whorl, but 2 clearly marked at outer lip), reaching suture below, obsolete
on base; crossed by spiral lirae 4 on 2nd whorl, 5 on 3rd, 9 on 4th, and ro plus
intermediaries on 5th, c. 14 additional fine lirae on base passing into ¢c. 10
stronger and more widely spaced ones on rostrum. Suture undulate. Sulcus
with numerous distinctly marked growth-lines, crossed by 1, 2 or (on last whorl)
3 fine spiral lirae. Lip sinus deep. Canal short. 21 x9°3 mm. Pale buff.

Brown’s Bank, Agulhas Bank (approx. 364° S., 21° E.), 80-100 fathoms,
1 dead, but in good condition. (S. Afr. Mus. A8611, P.F. Coll.).

Pleurotoma (Surcula) dissimilis Watson
Fig. 23 a
1886. Watson. ‘Challenger’ Rep., xv, p. 298, pl. 26, fig. 3.

Aperture subequal to spire. Protoconch 2 whorls (lip broken) diam.
1-5 mm., with very fine oblique plicae, passing into the more sigmoid and less
conspicuous growth-lines on Ist postnatal whorl, the junction marked by change
of colour (brown to white). Postnatal whorls 64 (7), profile evenly rounded,
scarcely any shoulder. Growth-lines not conspicuous. Very numerous fine
spiral striae: below shoulder 3 at beginning of 1st whorl, increasing to 6, 7-8
on 2nd increasing to 12 on 4th and 18-20 on 6th, c. 26 on 7th; at least 55
. additional striae on base. Growth-lines on sulcus perpendicular to the suture,
where they form a slight crimping (see Watson’s figure), feebly concave across

148 ANNALS OF THE SOUTH AFRICAN MUSEUM

the sulcus, but strongly protractive from lower border of sulcus. Lip sinus deep.
50 (probably 54 if canal complete) x 17 mm., 49 (with protoconch) x 16 mm.,
two others 45 X14 mm. Creamy-white, glossy, protoconch pale brown.

Off Cape Point, 660-900 fathoms, green mud. 5 dead, but in good con-
dition. (S. Afr. Mus. P.F. Coll.)

Distribution. SE. of the Philippine Islands, 500 fathoms (Watson).

Remarks. There is a close resemblance between this species and S. alberti
Dautzenberg & Fischer (1906. Res. Sct. Camp. Monaco, fasc. 32, p. 16, pl. 1, figs.
8-10) from off Cape Verde, 3,890 metres.

Soba,

TD.
}

Fic. 23. a. Surcula dissimilis Watson, protoconch. b. Turris ambages
n. sp., one whorl and protoconch.

Comparison of the figures shows that the lower part of the columella is a
little more curved and the aperture broader in alberti than in dissimilis. The
Cape specimens agree better with the latter.

For this reason it seems better to identify the Cape specimens with dissimilts,
although evidence of a faunistic relationship, dating perhaps from Miocene
times, between the north and south Atlantic off the west coast of Africa is
accumulating. (See: Cancellaria (Sveltia) lyrata.)

The course of the growth-lines is not shown very clearly in the figures of
either dissimilis or alberti, certainly not the almost horizontal direction at the
lower boundary of the sulcus which is so conspicuous in the present specimens.

The animal of alberti was reported to be closely related to that of the genus.
Bela (not Bela of Thiele 1929), but no figure of its radula was given. Presumably
it had no operculum.

Turris ambages n. sp.
Fig. 23 5
Protoconch 3 whorls, diam. 1°25-1°3, alt. 1°3-1°5 mm., with axial riblets.

(as in Turris). Postnatal whorls with 4 strong spiral keels (profile of whorl

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA “149

-4-lobate), 1st rather sharp, forming the cingulum, 2nd rounded forming the lip
sinus, with growth-lines producing more or less conspicuous squamiform
nodules, 3rd the most prominent, rounded, forming the periphery, below this
-the smaller 4th keel; base with ? c. 10 additional lirae. 17 x 7 mm. (protoconch
and 7 whorls); 25 8 (protoconch and 8 whorls); a broken specimen 30 mm.
long, with 5 whorls corresponding to the 5th—8th whorls plus a portion of the
gth whorl.

The 17 mm. example is whitish, with traces of irregular brown spots and
streaks.

Off Cape Natal (Durban), 54 fathoms, 2 broken specimens; off Umkomaas
River, 40 fathoms, 1 (17 mm. long); off O’Neil Peak (Zululand), 90 fathoms,
1 (25 mm. long) (S. Afr. Mus. A8683—A8685, P.F. Coll.).

Remarks. Adams & Reeves’ figure of fagina (1850. Voy. ‘“Samarang’ Moll.,
pl. 9, figs. 2 a, b) shows 4 keels, but the 2nd separated from the 1st by a wide
groove, and at least as wide and prominent as the 3rd. The accuracy of the
drawing is confirmed by Yen’s figure (1942. Proc. Malac. Soc., xxiv, pl. 25, fig.
184). The position of the lip sinus is not absolutely clear in either figure.

Moreover fagina is a large species (78 mm. in Yen’s figure if nat. size),
and presumably has a larger protoconch (? 2 mm. diam. in Yen’s figure) than

the present specimen. |

According to Melvill (1917. loc. cit., p. 147) Pleurotoma annulata Rve.
(Conch. Icon., i, pl. 5, fig. 55) is near akin to fagina. |

Acrobela acus n. sp.
Fig. 21 e

Protoconch 44 whorls, first 2 smooth, the following with criss-cross sculp-
ture, alt. 0-7, diam. 0-4 mm. Postnatal whorls 34; 1st whorl with one peripheral
keel at start, but soon with one above and one below, following whorls with
3 keels, the median one peripheral and strongest; no distinct axial sculpture
except the protractive curved growth-lines in the sulcus; additional lirae on
base 4 or 5 rather strong and 8-9 finer ones. 4°75 X 2 mm.

Off Cape Recife, 256 fathoms, 2 and one fragment, dead (type material) ;
off Cape St. Blaize, 125 fathoms, one dead (S. Afr. Mus. A8748 and A8562,
Fe: Coll.).

Remarks. Nearest to A. sansibarica Thiele (1925. D. Tiefsee Exp., Xv, p. 239,
pl. 37 (25), fig. 21), but the latter has a cingulum below the suture and only one
spiral keel.

A. circumvertens (M. & S.), with which A. optima Thiele 1925 appears to be

synonymous, from Gulf of Oman and East African coast, also has a cael
and, in the later whorls, 2 spiral keels.

Taranis miranda Thiele (1925. loc. cit., p. 254, pl. 32 (20), fig. 7), from
34° 51' S., 19° 37’ E., 80 metres, has a low protoconch (14 whorls), 2-3 rather

150 ANNALS OF THE SOUTH AFRICAN MUSEUM

broad keels which are crossed (in the figure) by pliculae, retractive across the
sulcus.

Distinguished from ‘Bela’ eva (supra, p. 114) by the three spiral keels, the
absence of axial pliculae (except in the sulcus), the growth-lines in the sulcus
being protractive, and by the very different protoconch.

Cythara alfredi (Smith)
Fig. 25 a

1892. Sowerby. Mar. Sh. S. Afr., p. 7 (Mangilia costata, non Donovan).

1897. id. Append. Mar. Sh. S. Afr., p. 31 (costata var. coarctata, non Forbes).
1904. Smith. 7. Malac., xi, p. 29, pl. 2, fig. 9.

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 31 (alfredensis typ. err.).

1925. Thiele. D. Tiefsee Exp., xvii, p. 229, pl. 37 (25), fig. 22 (Haedropleura dora).
1925. id. ibid., p. 232, pl. 40 (28), fig. 1 (Mangelia misera).

1932. Turton. Mar. Sh. Pt. Alfred, p. 29, pl. 6, no. 218.

1932. id. ibid., p. 30, pl. 6, no. 219 (thetis) (not Drillia thetis Smith).

Protoconch 24 whorls, with faint axial pliculae towards junction with Ist
postnatal whorl. Postnatal whorls usually 4, sometimes 5. Axial ribs 6 (7) on
ist whorl, 7 (sometimes 8) on following whorls; spiral lirae c. 8 on 1st whorl
increasing to ¢. 25-30 on last whorl, about the same number of additional lirae
on base. 7°8 X 2°75 mm.

Fulvous, 4th whorl usually paler with a fulvous band in middle.

Recorded (dead) from Kalk Bay (False Bay) to Port Alfred. S. Afr. Mus.
has one specimen alleged to come from Table Bay, but the provenance is
doubtful.

Thiele’s H. dora from Algoa Bay, depth not stated; and M. misera from
35° 16’ S., 22° 26’ E., 155 metres.

Off Illovo River (Natal), 27-30 fathoms; off Cove Rock (East London
area), 22 fathoms; Algoa Bay, 67 fathoms; off Cape Recife, 124 fathoms;
off Cape St. Blaize, 37 fathoms; all dead (S. Afr. Mus. P.F. Coll.).

Cythara ima Bartsch

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 31, pl. 3, fig. 1.
1932. Turton. Mar. Sh. Pt. Alfred, p. 30.

Protoconch 2 whorls, smooth. Postnatal whorls 5. Axial ribs 10 on Ist—
grd whorls, 12 on 4th and 5th; spiral lirae c. 24 on last whorl, plus c. 30 on base.
8-1 X 3°5 mm. (Bartsch).

Recorded from Simon’s Bay and Port Alfred.

S. Afr. Mus. has 2 worn specimens from Port Alfred, identified and
presented by Turton. The larger measures 9 (apex worn, 5 whorls) 3°75 mm.
Neither specimen corresponds with Bartsch’s description; they appear to be
merely 8-ribbed examples of alfredi.

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA I51

Cythara deliciosa n. sp.
Fig. 24

Protoconch 23 whorls, alt. 0-6, diam. 0:75 mm., smooth at start, last half
or three-quarters with a series of peripheral axially oblique knobs, and an axial
plica about at the junction (which is obscure) with 1st postnatal whorl. Post-
natal whorls 34, with 7 sharp axial ribs on each whorl, and numerous fine spiral
lirae, c. 20 on 1st whorl and 25 on body whorl (excluding base), last rib with at
least 40 lirae; lirae with very minute prickles on both sides, so that the inter-
vening striae appear punctate. 5:3 2:3 mm. Uniform pale brown.

ia.
HSS
earse
SsSy
RAS a
4
xX gl
ae

Fic. 24. Cythara deliciosa n. sp., body whorl and protoconch.

Off Umhloti River (Natal), 40 fathoms. 6 dead but unworn specimens
(S. Afr. Mus. A86q2, P.F. Coll.).

Remarks. Distinguished from alfred: by the lower, more squat protoconch,
with nodules, the more numerous spiral lirae on the postnatal whorls, and the
sharper axial ribs.

Mangilia amplexa Gould
Fig. 25 5

1860. Gould. Proc. Boston Soc. N.H., vii, p. 338.

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 30, pl. 2, fig. 10, pl. 7, fig. 6.
1915. id. ibid., p. 30, pl. 2, fig. 6 (Aumerosa).

1925. Thiele. D. Tiefsee Exp., xvii, p. 229, pl. 37 (25), fig. 23 (Haedropleura thea).
1932. Turton. Mar. Sh. Pt. Alfred, p. 27, pl. 6, no. 205 (rietensis).

19s2. id. ibid., p. 20.

152 ANNALS OF THE SOUTH AFRICAN MUSEUM

Protoconch 2} whorls, smooth. Postnatal whorls 44-5. Axial ribs 10 on
ist whorl, 10-11 on 2nd and grd, 11-13(14) on 4th; spiral lirae 6-7 on 1st
whorl, increasing to 18-20 on last whorl, 18-20 additional lirae on base.
9°5 X3°5 mam. :

Simon’s Bay (Gould), Port Elizabeth (Sowerby), Port Alfred (Bartsch,
Turton). ane

Algoa Bay (Thiele).

Table Bay, Kalk Bay and Gordon’s Bay (False Bay), Still Bay (S. Afr.
Mus.).

Fic. 25. Protoconchs of: a. Cythara alfredi (Smith), and
b. Mangilia amplexa Gould.

Algoa Bay, 67 fathoms; off Keiskamma Point, 33 fathoms; and off East
London, 27-32 fathoms; all dead (S. Afr. Mus. P.F. Coll.).

Remarks. The numbers ‘14’ and ‘12’ (axial ribs) in Bartsch’s description
of amplexa seem to be transposed; in any case the numbers are too high (see his
figures).

Mangilia eucosmia Bartsch

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 28, pl. 2, fig. 7.
1932. Turton. Mar. Sh. Pt. Alfred, p. 27.
1932. id. ibid., p. 28, pl. 6, no. 212 (rufanensis).

Protoconch and coloration as in amplexa. Distinguished by the more
convex (shouldered) whorls and deeper sutures, fewer and more oblique axial
ribs. Nevertheless the two ‘species’ are doubtfully distinct. On two occasions
both were found together in the same bottom sample.

According to topotype examples received from Turton, the numbers of
ribs stated by Bartsch are much too high: the present topotypes have only
10-11 on the last whorl, which is fewer than in amplexa. In face view only 3 ribs
appear between those forming the profile on either side, whereas in amplexa 4
are visible. In this respect Bartsch’s figure corresponds neither with the ei
nor with the Pieter Faure specimens.

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 153

Port Alfred (Bartsch, Turton). Common in the lagoon (estuary) at Port
Alfred (Turton).

Off East London, 32 fathoms; off Keiskamma Point, 33 fathoms. (S. Afr.
Mus. P.F. Coll.)

: Mangilia muri n. sp.

Fig. 26 a

Protoconch 3 whorls, smooth (but probably worn), an obscure spiral keel
at end of 3rd whorl. First postnatal whorl with 2 feeble spiral keels which develop
into spiral lirae with knobs at intersections with the 13-14 axial ribs; the latter

starting from suture, where they are slightly enlarged.
Still Bay (S. Afr. Mus. A8647, Muir Coll.).

Fic. 26. a. Apex of Mangilia muiri n. sp. b. M. kowiensis Turton.
c. M. shepstonensis Smith, with protoconch and lip sinus further enlarged.

Remarks. There are 8 specimens from 1 mm. long (protoconch and 1st
whorl) to 2°75 x 1:5 mm. (protoconch and 2 whorls).

Mangilia benjamini Bartsch

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 26, pl. 7, fig. 5.
False Bay (Bartsch; collected by the U.S. Exploring Exp., 1853).

Remarks. This species, based on ? only one specimen, is broader than
Surcula macilenta Melv., the width less than 3 in the length (1: 2-7), but otherwise
there is considerable resemblance (see Drillia platystoma, p. 125). It was described
as having 9 postnuclear whorls at a length of 15:3 mm.

Thiele described two species from the Agulhas Bank also very similar to
benjamim, having the whole whorl covered with fine spiral lirae: Pleurotoma?
vilma (1925. D. Tiefsee Exp., xvii, p. 229, pl. 37 (25), fig. 1) and Crassispira?
agulhasensis (ibid., p. 213, pl. 35 (23), fig. 11). In both of these, however, the
axial ribs cross the sulcus to the suture above.

154 ANNALS OF THE SOUTH AFRICAN MUSEUM

Mangilia kowvensis Turton
Fig. 26 b
1932. Turton. Mar. Sh. Pi. Alfred, p. 25, pl. 5, no. 190.

Protoconch 14 whorls, smooth. Postnatal whorls 3, each with 4 strong
spiral lirae separated by deep grooves. Shoulder strongly tabulate. Axial ribs
not conspicuous, but indicated by 10 or 11 nodules on the lirae on 2nd and
3rd whorls; when viewed apically the periphery is undulate. 7—8 additional
lirae on base, decreasing in size anteriorly. 3:3 X 1-5 mm.

Port Alfred (Turton).

34° S., 25° 45’ E., depth not recorded, 1 dead (S. Afr. Mus. P.F. Coll.).
Mangilia shepstonensis Smith
Fig. 26 ¢
1914. Smith. Ann. Natal Mus., iii, p. 1, pl. 1, fig. 1.

Aperture subequal to spire. Protoconch 24~3 whorls, alt. 0-3, diam.
0-5 mm., smooth, glossy. Postnatal whorls 4, profile convex with rounded,
almost tabulate shoulder. Axial ribs 18 close together on 1st (or last half whorl
of protoconch plus Ist postnatal whorl), 12 on 2nd, 3rd and 4th, not quite as
wide as intervals, crossing sulcus and extending half-way across base; crossed
by 4 narrow spiral lirae on each whorl, the 4th lira low down near suture, and
also by extremely fine spiral striae, c. 10-12 of the latter in sulcus on last whorl,
and c. 5 between each pair of lirae; c. 14 additional lirae on base, with inter-
vening striae. Sulcus concave between the ribs. Lip sinus deep, adjoining
suture. 5 2 (4 whorls) and 4°5 X 1:75 mm. (3 whorls); Smith: 4-32 mm.

Pale buff, a faint brown mark in middle of the incrassate outer lip.
Yellowish, with a narrow rufous interrupted band around middle of last whorl
(Smith).

Port Shepstone (Natal) (Smith).

Off Umhloti River (Natal), 40 fathoms, one fresh, one slightly worn; off
Umkomaas River (Natal), 40 fathoms, one worn; off Cove Rock (East London
area), 22 fathoms, 2 unworn (S. Afr. Mus. P.F. Coll.).

Remarks. The narrow spiral lirae are continuous across, but subordinate to,
the axial ribs. In addition to the semicancellate macro-sculpture, the growth-
lines and the spiral striae produce a very fine micro-cancellate sculpture in each
hollow.

The junction of protoconch and 1st postnatal whorl is indistinct; probably
5 or 6 of the more closely set initial axial ribs belong to the last half whorl of the
protoconch. , |

There seems to be no great difference between this species and Glyphostoma
siren Smith 1904. The latter has only 10 ribs, which are not ‘superne rotunde
tabulati’; and Smith makes no mention of the presence of any microsculpture.

Contrast also Drillia tholos (supra).

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 155

Fic. 27. a. Mangilia translucens n. sp., and protoconch. 6. M. phoxos n. sp. c. M. exstans n. sp.

Mangilia translucens n. sp.
Fis 27 2

Aperture a little longer than spire. Protoconch 3 whorls, alt. 0-5 diam.,
0-75 mm., smooth, glossy. Postnatal whorls 3, profile convex with distinct but
rounded shoulder. Axial ribs 12-13 on each whorl, narrow, sharp when fresh,
extending across sulcus and across base; no spiral sculpture except a mere trace
here and there of one or two striae. 5 2:3 mm. Pale buff, semitranslucent,
glossy.

Off Tugela River, 65-80 fathoms, one glossy; off Umhloti River (Natal),
40 fathoms, 12 specimens, some glossy, some dull, some broken (type material) ;
off Umkomaas River (Natal), 40 fathoms, one dull; off East London, 32
fathoms, 1 glossy, 3 dull; all dead (S. Afr. Mus. A8755, A8586, A8726, A8732,
EE. Coll.).

Remarks. The ribs where they cross the sulcus and form the shoulder are
usually more prominent than shown in the figure.

Mangilia phoxos n. sp.
Fig. 27 5
Aperture 1} in spire. Protoconch 4 whorls, alt. and diam. 0-75 mm.,
first 3 whorls smooth, fine close-set axial pliculae on last three-quarters of
4th whorl, glossy. Postnatal whorls 3, profile convex with slight shoulder.

Axial ribs 11 on Ist, 12 on 2nd and 3rd whorls, fine but distinct, curved on
sulcus and slightly oblique below shoulder, petering out on base; crossed by

156 ANNALS OF THE SOUTH AFRICAN MUSEUM

slight and narrow spiral lirae 2 on Ist, 3 on 2nd and 3rd whorls, the upper one
defining the sulcus and forming the shoulder, forming slight complanate nodules
at intersections, ¢. 18 additional lirae on base; in addition extremely fine spiral
lirae over whole whorl, 6 in sulcus on 1st whorl increasing to 10 on 3rd, 4
between each of the two pairs of main lirae, and forming intermediaries
between the lirae on base; these fine lirae together with the fine growth-lines
form a microgranulate or microcancellate sculpture. Lip sinus deep, adjoining
suture. 5°82 mm. Pale buff, protoconch glossy.

Off Umhloti River (Natal), 40 fathoms, one dead (S. Afr. Mus. A8730,
P.F. Coll.). })

Remarks. Similar to M. (Paraclathurella) padangensis Thiele 1925 from
Sumatra, but with finer spiral lirae and no regularly granulate whorl near apex
(on last part of protoconch: Thiele).

Mangilia exstans n. sp.
Fig. 27 ¢

Protoconch 14 whorls, smooth. Postnatal whorls 3, each with 2 spiral
lirae, the upper being the more prominent, especially on 3rd whorl; in sulcus
one additional lira on 1st whorl, 2 lirae on 2nd and 3rd; _no axial ribs, but lirae
with complanate nodules: 12 on the peripheral lira on 2nd whorl, 14 on 3rd,
the lower lira and those in sulcus also nodulose; 7 additional lirae on base,
rather strong and well spaced, more or less feebly nodulose. 2-5 1-3 mm.
(1-5 including lirae).

Off Gove Rock (East London area), 22 fathoms, one dead (S. Afr. Mus.
A8756, P.F. Coll.).

Remarks. Distinguished by the prominent peripheral lira and absence os
axial ribs. Protoconch not so elevated as in tranquilla (infra).

Mangilia nisga Bartsch
Fig. 28 d
1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 25, pl. 7, fig. 1.

Protoconch 14 whorls, smooth. Postnatal whorls 3, each with 3 spiral lirae.
Axial ribs beginning on later half of 1st whorl, 10 on 2nd, 11 on grd, forming
rounded knobs at intersections with the lirae. 7-8 additional lirae on base,
feebly nodulose. An additional lira with feeble knobs in the sulcus on 2nd and
3rd whorls. 3°5 x 1°75 mm.

Port Alfred bPartegh Turton).

34° 20’ S., 25° 42’ E., 124 fathoms, and 33° 3’ S., 27°.57 BE. » 32 fathoms.

3 dead (S. Afr. Mus. P.F. Coll.).

Remarks. Bartsch said there were 14 ribs on the last whorl, which seems
scarcely compatible with his figure. The 4th spiral lira shown in Bartsch’s figure
as above the posterior end of the aperture, is in these specimens on a level with
or below it, and definitely on the base. |

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 157

_- Fic. 28. a. Mangilia minuscula Smith (nodulose form), protoconch. 5. M. tranquilla n. sp.
c. M. sciolan. sp. d. M. nisga Bartsch, with protoconch further enlarged.

Mangilia sciola n. sp.
Fig. 28 ¢

Protoconch 14 whorls, smooth, junction between protoconch and 1st whorl
obscure. Postnatal whorls 2, each with 2 spiral lirae. Axial ribs beginning a
little later than the spiral lirae, 11 on rst, 12 on 2nd whorl, forming rounded
knobs at intersections with lirae. 6 additional lirae on base, the 3 posterior ones
nodulose. No lira in sulcus. 3 x 1:5 mm.

34° 27’ S., 25° 42’ E., 256 fathoms, and 33° 6’ S., 27° 55’ E., 43 fathoms.
2 dead (S. Afr. Mus. A8642, A8643, P.F. Coll.).

Remarks. Deceptively like nisga, but with only 2 spiral lirae. The specimen
from shallower water is fresh, but that from deep water is somewhat worn.

Mangilia ponsonbyi (Sow.)

1892. Sowerby. Mar. Sh. S. Afr., p. 7, pl. 1, fig. 5 (not good) (Defrancia p.).
1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 31.

1925. Thiele. D. Tiefsee Exp., xvii, p. 230 (Clathurella p.).

1932. Turton. Mar. Sh. Pt. Alfred, p. 25.

Sowerby gives 10 axial ribs on last whorl on shell 6-5 x 3 mm. with 6 (total)
whorls, and 4 spiral lirae. |

Three worn specimens in S. Afr. Mus. (coll. Turton), largest with 5 whorls
(minus protoconch) 8-5 x 3-5 mm., have 13 ribs on 3rd, 4th and 5th whorls,
crossing sulcus, and continued across base; and 6 spiral lirae (2 fine ones on
sulcus, 4. stronger ones between shoulder and suture below) on 4th and 5th
whorls, slightly nodulose at intersections with ribs; 8 (9) additional lirae on
base. Columella with 2-3 pleats and inner margin of outer lip 6-7 plicae.

Turton gave the size as 10 X 3 mm.

Recorded (dead) from Port Elizabeth and Algoa Bay, and Port Alfred.

158 ANNALS OF THE SOUTH AFRICAN MUSEUM

Mangilia tranquilla n. sp.
Fig. 28 b

Protoconch 14 whorls, elevated, smooth. Postnatal whorls 2. Spiral lirae
2, slightly nodulose at intersections with axial plicae, of which there are 13-14
on Ist, and 15-16 on 2nd whorl; plicae traceable across sulcus, with slight
nodular enlargements on 2nd whorl; lattice hollows between 1st and 2nd lirae
square.

Protoconch plus 1st whorl 1-30-75 mm.; protoconch plus 2 whorls
2X15 mm. White.

Still Bay, 2 specimens (S. Afr. Mus. A8648, Muir Coll.).

Cf. Mangilia gemmula Turton 1932, which has 12 axial plicae.

The protoconch is smaller and proportionately narrower than in Tritonalia
scrobiculata. )

Mangilia minuscula Smith
Fig. 28 a

1910. Smith. Ann. Natal Mus., ii, p. 191, pl. 7, fig. 4.
1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 24, pl. 7, fig. 3 (Mangilia gisna).
Protoconch 14 whorls, elevated, smooth. Postnatal whorls 3. Two spiral
lirae at beginning of 1st whorl, on later half a third lira develops between
shoulder and suture below, on 2nd and 3rd whorls 3 lirae. Axial ribs subordinate
to the lirae but forming conspicuous knobs at intersections, and connecting
riblets between the lirae, 12 on 1st whorl, 16 on 2nd and ¢. 20 on 3rd (Bartsch’s
“28” is too high a number for the present specimens, and conflicts with his figure;
? type err. for 20); 7-8 additional lirae on base. Protoconch alt. 0-5, diam.
0°35 mm.; protoconch plus half 1st whorl 1 x0-5 mm.; 3 X 1-3 mm.
Recorded from Port Elizabeth and Port Alfred. Kalk Bay (False Bay) and
Still Bay (S. Afr. Mus.). False Bay (U.C.T.).

Remarks. Specimens from the Muir collection from protoconch up to
3-whorled examples.

Turton says minuscula and gisna are easily distinguished, but specimens of
both identified by him are indistinguishable; gisna is the strongly nodulose form.

Similar to verrucosa Sow., but narrower and with more numerous axial ribs
(knobs): 12-20 instead of 11-14.

M. helga Bartsch differs in having the spire distinctly longer than aperture,
4 spiral lirae on last half of 1st whorl, and on 2nd and 3rd whorls, and no axial
plicae or knobs on the lirae.

Philbertia natalensis n. sp.
Fig. 29 a

Protoconch 33—4 whorls, diam. and alt. 0-5 mm., with criss-cross sculpture.
Postnatal whorls 3, profile angular slightly below middle. Axial riblets 9 on
each whorl, rounded, narrower than intervals, extending above almost to

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 159

suture, obsolete on base; crossed by spiral lirae, 3 on 1st whorl, the lowest the
strongest and peripheral, 4 on 2nd, 2 above and one below the peripheral lira,
on body whorl 4—5 above and 3-4 below, ¢. 12 additional lirae on base. Sulcus

not concave, a very feeble cingulum below suture. Lip sinus adjoining suture.
| 5 X 2°3 mm.

Fic. 29. a. Philbertia natalensis n. sp., protoconch. b. P. capensis (Smith), two views of
protoconch. c. P. alfredensis (Turton), with protoconch.

Off Umhloti River mouth, Natal, 40 fathoms, 1 dead; off Cape Natal
(Durban), 54 fathoms, 1 dead; (S. Afr. Mus. A8654, A8758, P.F. Coll.).

Remarks. Protoconch broader than in P. capensis (p. 117), Ist postnatal
whorl not abruptly wider than protoconch, and sculpture less outstandingly
cancellate.

Philbertia alfredensis (Turton)
Fig. 29 ¢

? 1925. Thiele. D. Tiefsee Exp., xvii, p. 240, pl. 38 (26), figs. 4, 4@ (Bellardiella
sultana).
1932. Turton. Mar. Sh. Pt. Alfred, p. 24, pl. 5, no. 181 (Bellardiella alfredensis).
Aperture 14 in spire. Protoconch 4 whorls, alt. 0-75, diam. 0-5 mm., 2nd
to 4th whorls in profile angular slightly below middle, but not sharply keeled,
minutely and closely punctate in spiral lines (and also in axial series). Postnatal
whorls 3. Axial ribs 10 on 1st and 2nd whorls, 11 on 3rd whorl, feebly indicated
across sulcus, obsolete on lower half of base, narrower than intervals; spiral
lirae narrower than ribs, 2 on each whorl, intersections slightly nodulose, 7
additional lirae on base; in the sulcus one fine lira on 2nd whorl, 2 on 3rd
whorl. Lip sinus deep, adjoining suture. 3°75 X1°5 mm.
Off Tugela River (Natal—Zululand), 65-80 fathoms, one dead (S. Afr.
Mus. A8761, P.F. Coll.).

Remarks. Seems to be referrable to Turton’s species, based on a very worn
specimen, 4. 1-8 mm., from Port Alfred. Turton compared his species with
Beilardiella sp. Thiele 1925 (loc. cit., pl. 38 (26), fig. 7) from East Africa.

160. ; ANNALS OF THE SOUTH AFRICAN MUSEUM

_ Thiele described B. sultana and other species from the same area. I should
be inclined to refer the P.F. specimen to suliana (72:75 mm.), but for the
presence of the fine lirae in the sulcus, which Thiele neither mentions nor
figures. -

Daphnella sulcata (Sow.)
Fig. 30 5

1892. Sowerby. Mar. Sh. S. Afr., p. 11, pl. 1, fig. 10 (Cominella ? s.).
1904. Smith. 7. Malac., xi, p. 28 (Daphnella ? s.).

1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 32 (Daphnella? s.).

1932. Turton. Mar. Sh. Pt. Alfred, p. 30.

Fic. 30. a. Daphnella recifensis n. sp., with protoconch. b. D. sulcata (Sow.), two views of proto-
conch. ¢. Worn specimen referred to Thesbia algoensis Thiele. d. Pleurotomella ursula Thiele, two
views of protoconch of 9 mm. specimen.

Protoconch 14 whorls, smooth, junction with 1st whorl oblique. Postnatal
whorls 4 (shell 9 mm., Sowerby: total whorls 6, shell 10 mm.). Spiral lirae
(5)6—7 on 1st whorl, 6-8 on 2nd and 3rd, 7-8 on 4th, width subequal to the
grooves which are crossed by fine growth-lines; 8—g additional lirae on base;
3-5 denticles on inner side of outer lip. 7:3 x 2:5 mm., Sowerby: 10 X 3. ,

Pale fulvous, obscurely mottled or uniform (Sowerby) ; yellow, spotted with
brown, or uniform yellowish-brown (Turton).

Recorded (dead) from Port Elizabeth and Port Alfred. Still Bay, dead
(S. Afr. Mus. Muir Coll.). Off Cape St. Blaize, 125 fathoms, one broken; off
Cove Rock (East London area), 22 fathoms, one dead; off Umkomaas River
(Natal), 40 fathoms, one and three fragments; off Cape Vidal (Zululand),
80-100 fathoms, one broken (S. Afr. Mus. P.F. Coll.).

Remarks. Sometimes the spiral lirae have slight thickenings —‘subgranose’
(Smith) —making their margins feebly undulate. : |

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 161

Three worn specimens (S. Afr. Mus. loc.?), however, show a definite
cancellate sculpture, due to well-marked axial plicae, slightly nodulose at inter-
sections with the lirae; the latter are fewer in number: 5 on 1st whorl, 6 on

the others. Two of these specimens have a series of brown spots around the
middle as in some typical specimens.

Of the Natal specimens, two of the fragments are normal, with growth-lines

‘in the grooves, some-of the growth-lines stronger than others; the third frag-
ment and the complete specimen are definitely cancellate, the axial ribs as
‘strong as the spiral lirae. This complete specimen, which has the characteristic
obliquely and abruptly ending protoconch, would certainly be regarded as a
separate species but for the connecting ‘subgranose’ specimens.

Daphnella recifensis n. sp.
Fig. 30 a

_ Protoconch 14 whorls, moderately elevated, very finely spirally striate,
junction with 1st postnatal whorl distinct. Postnatal whorls 3. Axial ribs 16
on Ist, 20 on 2nd, 27 on 3rd whorl, from suture to suture, petering out on base;
spiral lirae 3 on Ist, 4-5 on 2nd, 6 on 3rd whorl, 10 additional on base; ribs
predominant over the lirae, nodulose at intersections. 4°5 X 1°75 mm. White.

Off Cape Recife (34° 27’ S., 25° 42’ E.), 256 fathoms, one dead (S. Afr.
Mus. A8757, P.F. Coll.).

Remarks. A much smaller species than alfredensis Bartsch. Smaller and
broader than Daphnellopsis lamellosa Schepman 1913 and Daphnella subuloides
Schepman 1913, to which there is some similarity in sculpture.

Daphnella alfredensis Bartsch
1915. Bartsch. Bull. U.S. Nat. Mus., 91, p. 32, pl. 8, fig. 3.

In an 8 mm. topotype (top of protoconch worn) with 3 postnatal whorls,
there are c. 12, 16 and 21 axial riblets on 1st, 2nd and 3rd whorls respectively,
i.e. more than Bartsch gave in his description (his figure shows too few!); spiral
lirae 3 on 1st whorl, 6 on 2nd, 16-17 on 3rd, including those on sulcus (Bartsch
said ‘appressed portion’ appears to be free of spiral sculpture).

Thesbia algoensis Thiele
Fig. 30 ¢
1925. Thiele. D. Tiefsee Exp., xvii, p. 232, pl. 40 (28), fig. 14.

Aperture subequal to spire. Protoconch 14 whorls, somewhat lopsided
(corroded). Postnatal whorls 3, profile evenly convex. Axial ribs (1st whorl
corroded), c. 19 on 2nd whorl, 21 on 3rd, not strongly curved, crossing sulcus
which is not defined, subequal in width to intervals; spiral lirae 5-6 on 2nd
whorl, 6—7 on 3rd, showing only in the intervals between ribs, c: 12 additional
lirae on base. 4°5 2:25 mm. Thiele: 3-5 1-7 mm. Buff.

162 ANNALS OF THE SOUTH AFRICAN MUSEUM

Algoa Bay (33° 50’ S., 25° 48’ E.), depth not stated (Thiele). Off Cape
Morgan, 47 fathoms, one abraded specimen (S. Afr. Mus. A8734, P.F. Coll.)..

Remarks. Thiele did not state the number of ribs and lirae; his figure seems
to show a few more than in the present specimen, but the slight discrepancy does
not preclude referring the Pieter Faure specimen to his species.

Although Thiele’s figures of J. algoensis and Columbella dianae appear
different, there is considerable difficulty in practice in assigning several Pieter
Faure shells to one or the other. The above specimen is much more like T. algoensis
in having a broad body whorl, but other specimens are more like C. dianae, and
have been reserved for further study.

Pleurotomella ursula Thiele
Fig. 30 d
1925. Thiele. D. Tiefsee Exp., xvii, p. 231, pl. 41 (29), fig. 3.

Aperture subequal to spire (Thiele’s figure), slightly longer than spire
(15 mm. specimen), or 1} in spire (9 mm. specimen). Protoconch 14 whorls,
alt. and diam. o-6 mm. (0:5 mm. in g9 mm. specimen), apically smooth, last
half whorl with close-set fine axial pliculae. Postnatal whorls 5. Axial ribs 11
on Ist whorl, increasing to 15 on last whorl; crossed by spiral lirae 4, increasing
to 5-6 on 4th and 5th whorls, intermediaries on later part of 3rd and on 4th
whorl (g mm. specimen), in 15 mm. specimen intermediaries as strong as
primaries, total 12-13. Sulcus deeply sunken, with well-marked growth-lines.
Lip sinus very deep, adjoining suture; outer lip expanding below sulcus. Canal
short, but longer than in Thiele’s figure. 6-75 x 3 mm. (across shoulder of body
whorl, 3:5 if outer lip included), 93-5 mm. (resp. 4°5), and 156 mm.
(resp. 7). ‘“Uiniele: 6x 3; mom

35° 16’ S., 22° 26’ E., 155 metres (Thiele).

Cape St. Blaize, N. x E., 73 miles, 125 fathoms, two dead; off Cove Rock
(East London area), 80-100 fathoms, one dead (S. Afr. Mus. A8559, A87o01, |
P.F. Coll.).

Remarks. The largest specimen came from the East London area; all three
are larger than the Valdivia specimen.

The protoconch does not appear quite so high as in Thiele’s figure. He
described it as somewhat rough. According to his 1929 diagnosis, the genus has
criss-cross sculpture.

Gen. MirromorpHa A. Adams
Subgen. Antimitra Iredale
Mitromorpha (Antimitra) hewitti Tomlin

1904. Smith. 7. Malac. xi, p. 31, pl. 2, fig. 13 (volva Sow. var.).
1921. Tomlin. 7. Conch., xvi, p. 156.
1932. Turton. Mar. Sh. Pt. Alfred, p. 48, pl. xi, no. 359 (striolata). —

CONTRIBUTIONS TO KNOWLEDGE OF S.A. MARINE MOLLUSCA 163

Recorded from Port Alfred and East London.
Off Cove Rock (East London area), 22 fathoms, 2 specimens from bottom
sample (S. Afr. Mus. P.F. Coll.).

Mitromorpha apollinis Thiele

1925. Thiele. D. Tiefsee Exp., xvii, p. 255, pl. 31 (19), fig. 14.

Protoconch 2 whorls, alt. and diam. 0-5 mm., smooth, junction with 1st
postnatal whorl distinct. Postnatal whorls 5. Axial ribs 13-14 on Ist and 2nd
whorls, increasing to 15-16 on 3rd, but evanescent on later part of 3rd and
obsolete on 4th and 5th, crossing the sulcus but not the cingulum; crossed by
flattened spiral lirae 3 on 1st whorl, 3-4 on 2nd, 4(5) on 3rd, 5 on 4th and 5th
whorls, c. 16 additional lirae on base. Sulcus with narrow cingulum. Fine
growth-lines across sulcus, and between the lirae. 9°5 x 3:5, and 10°5 X 3°75 mm.
Pale fawn. }

34° 51’ S., 19° 37’ E., 80 metres (Thiele).

Cape St. Blaize N. x E., 73 miles, 125 fathoms, 2 dead (S. Afr. Mus. P.F.
Coll.).

Remarks. Superficially like Daphnella sulcata, but with smaller protoconch,
flatter whorls, and spiral lirae fewer on the whorls but more numerous on the
rostrum. With 5 whorls the above two specimens are larger than the four
obtained by the Valdivia. Thiele’s description says 6 spiral lirae on penultimate
(i.e. 3rd whorl); this includes the cingulum, but even so the present specimens
appear to have one lira less because the lowest one is weaker than the others
and more or less occluded by the suture of the following whorl.

M. jovis Thiele 1925, from the same Valdivia station, in spite of its slightly
larger protoconch (according to the figure: 0-6 or 0-7 mm.) is extraordinarily
hike apollinis.

The third South African species described by Thiele: M. neptuni Thiele
1925, is very similar in sculpture to Mangilia nympha Bartsch 1915, but is a little
larger with the same number of whorls.

‘ysiyopiod JO sdA J, oy. SULaq oUO

JSP] ou) “SULIOT poards-Moy »>MOd ut0}}Oq ul 6 ode uO IJITY [, ‘SSULY IV UL pue wutOy Ul yeordA) SU)DIDU :MOd soddn UI JATY YS. “TSIAYIPLO S1]D]DU SNUO’ )

>
oomal
—
b
%
nd
oO
=
a
cS
—
Pd)

Ann. S. A

The ANNALS OF THE SOUTH AFRICAN MUSEUM are issued in parts at ieee ;

intervals as material becomes available.

Some volumes and parts are out of print, and others are only sold as parts of a set, or a:

respectively. The prices of parts published prior to 1940 have been increased.

Out of print: Vols. I, II (Parts 1-3, 5, 7, 8), III (Part 1), V (Parts 1, 2, 5, 8,9), VI (Part 1,
Index), VII (Parts 1, 2, 3, 5), VIII, [X (Parts 1, 2), X (Part 2), XI (Parts 2, 7, Index), XXI
(Part 1), XXIV (Part 2), XXXI (Parts I-3).

Vol.

XXXIX.

1900-1902
1903-1905
1903—1908
1906-1910

1908-1910
1908-1913
IQII-1918
IQII-I1QI4
1911-1918
1913-1924
1913-1923
1915-1924
1914-1916
1917-1933
1917-1920

1921
1924-1925
1924-1926
1925-1927
1925-1928
1925-1926
1929-1938
1927-1928

1928

1929
1929-1932
1929-193!
1931-1935

Zoology and Geology (excl. Parts 1-3, 5, 7, 8)

Zoology .. (excl. Part. 1)

Palaeontology ns

Geology, eee Zoology, Anthropology (ext
Parts 1, 2,5,8,9) .-

Zoology .. an

Palaeontology

Botany oe ae °°

Zoology

Palaeontology and Gentes

fee and mine

Zoology a

Zoology

Botany fe ae es ie He

Zoology

Palacontaloay

Zoology 4

Anthropology and Ethnology

Zoology ,

Zoology i ae oe wil SF ves

Anthropology oye af asa 2% ns pie

Palaeontology ‘

Zoology

osama by te ith “ind

. (Parts 4 and 6 only)
‘Gane Parts 1 and 2)

. (excl. Part 2)

ey pe 2, 7, Index)

..(excl. Part 1)

..(excl. Part 2)

of papers, authors, and pieces’ ‘eublished in “a ie XXX

1934-1950
1935-1940
1939
1938
1956
1942-1948
1947-1952
1950
1952
1952-1956
1952-1955
1953-1956
1955-1957
1957-

Palaeontology

Zoology

Archaeology

Zoology

Botany

Zoology

Palaeontology .
Zoology and falacoutalegy A es
Parts 1-5, Zoology and Palaeontology ..

. (Part 4 only)

Copies may be obtained from—

The LIBRARIAN, Sout Arrican Museum, Care Town.

= NOP HKHNHWO NAN NWOOWN Nm me ae ND Nan DWH OW AO OR DAN

Si
I

Se,
2

woof

= =
Ooo OO) ~7 09

hee

POMIARCOCOCOSCNOHMIOOWO OO SUW OWN

loc)

owuMwn

As
ytd
a?
a.
at
ae
i
‘ge
aN
=
he
i)
|

VOLUME XLIV

1 | S . the coasts res Natal and Portuguese East Africa. Part 1. Calypto-
* ek By N. eo Muniarp, Pu.D. (With 16 figures in the text.)

ial

| etm MES ag

MUSEUM
\ -WYNBERG, CAPE,

po.

HYDROZOA FROM THE COASTS OF NATAL AND PORTUGUESE
EAST AFRICA. Parr I. CALYPTOBLASTEA.

N. A. H. Mitiarp, Px.D.
Koology Department, University of Cape Town

(With 16 figures in the text)

INTRODUCTION

This paper represents the second of a series describing the Hydrozoa of
the coasts of South Africa. The species from Natal and the southern part of
Portuguese East Africa are dealt with together since from a provisional inspec-
tion they appear to represent a fairly comprehensive geographical unit bathed
by the warm waters of the southward-flowing Mozambique Current. They
represent the subtropical component of the fauna of the South African coast.

How far this apparent geographical unit is a real one, and how sharp a boundary

there is between this unit and the more southerly one, will only be answered
when the species of the south coast have been described.

The material dealt with comes from several sources. Part was collected
in the early years of the century by the government survey vessel, the s.s. Pieter
Faure, and is lodged in the South African Museum. Most of it is of more recent
origin and forms part of the Ecological Survey collection in the Zoology
Department of the University of Cape Town. Material has also been submitted
for examination by Mrs. M. Kalk of the University of the Witwatersrand, and
by Messrs. B. R. Allanson and P. Zoutendyk, to whom I wish to express my
thanks. The details of localities, etc., are listed below.

My thanks are also due to Dr. W. R. Rees for enabling me to examine
certain material in the British Museum, and to the Zoologische Sammlung des
Bayerischen Staates, Munich, for the loan of types.

All type material described in this paper has been deposited in the
South African Museum.

Financial aid from the C.S.I.R. and the Carnegie Corporation made
possible the ecological survey, and grants from the Staff Research Fund of the
University of Cape Town have assisted greatly in the collection of the necessary
literature and microscopic apparatus. The cost of publication was partly
defrayed by a grant from the publications fund of the University of Cape Town.

165

166 ANNALS OF THE SOUTH AFRICAN MUSEUM

STATION List

AFR Material collected by the government research vessel R.S. Africana off
the Natal coast.

Date Position Depth Bottom
AFR 1028 15/5/48 28°28’S./32°25'8’E. 27 m. Fine sand and rock
AFR 1098 1/6/48 29°57'S./31°11’E. 333 m. Green mud
D_ Intertidal material from Isipingo and other localities near Durban,
collected in July 1935 and July 1936. .
DNB Material from Durban Bay (a land-locked bay).
DBN 2 7/7/50 From floating jetty
DBN 62-70 July 1950 Intertidal
DBN 89 8/1/51 Intertidal
DBN 119 12/1/51 Dredged just outside bay (south of S. Breakwater), from atleast 8m.
DBN 130 15/1/51 From ships’ hulls operating only in Durban Bay

DBN 199, 200 ~—- 3/10/51 Intertidal
DBN 238-329 = April 1952 Intertidal
DBN 381 1/5/52 Dredged from 12 m.

IN Material from Inhaca Island, Delagoa Bay, Portuguese East Africa.
Collected by Mrs. M. Kalk.

IN 41 Date unknown

IN 49 1954. Intertidal

IN o1 July 1955 _—sIntertidal

IN 100 18/9/55 Intertidal

IN 136 1/4/56 Brought up by diver from 12 m.

IN 137-138 14/7/56 Canal debris
IN 139-140 =. 20/7/56 Intertidal

MOR Material from Morrumbane Estuary, inland from Inhambane,
Portuguese East Africa.

MOR 7 14/1/54 Dredged in channel at Mongue Ferry, up to 3 m.

MOR 40 20/1/54. Linga Linga, intertidal

MOR 216-218 July 1954 Linga Linga, from hull of wreck

NA Material from intertidal zone on Natal coast, collected partly by Prof.
J. H. Day, partly by B. R. Allanson.

NA 111 16/7/46 Umhlanga Rocks
NA 112 16/7/46 Tiger Rocks, Isipingo
NA 114 16/7/46 Umhlanga Rocks
NA 116-117 18/7/49 Kosi Bay reef

NA 184 July 1950 Kosi Bay reef

NA 212 14/7/46 Inyoni Rocks

PF Material dredged off the Natal coast by the s.s. Pieter Faure. Collection
lodged in South African Museum. The positions were given in the
original records as compass bearings off salient points on the coast, and
were probably not very accurate. These have been converted into
latitude and longitude and given to the nearest minute.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA

Date Position Depth Bottom
PF 10781 = 17/12/00 29°53’S./31°11’E. (off Cape Natal) 155m. s. sh.
PF 11141 = 31/12/00 30°15’S./30°54’E. (off Umkomaas River) 73m. brk. sh. st.
PF 11323 = 10/t/or 29°27’S./31°34’E. (off Tugela River) 7 pis ees se
PF 11803 A 8/2/o1 29°0’S./31°49’E. (off Durnford Point) 241m. §./Sb.
PF 12028 =27/2/01 28°41’S./32°22’E. (off Cone Point) 62m. __ brk. sh.
PF 12308 = 14/3/01 30°53’S./30°28’E. (off Port Shepstone) 66m. _ brk. sh. st.
PF 12392 )=s-114/3/0o1 = 31°2°5’S./30°18’E. (off Itongazi River) 46m: rst.
PF 12456 22/3/01 30°32’S./30°38-5’E. (off Umtwalumi R.) 46m. brk. sh.

167

PZ Material collected by P. Zoutendyk from Morrumbane Estuary, inland

from Inhambane, Portuguese East Africa.

PZ 13

16/7/54 Linga Linga Point, dredged in 3-5 m.

RHB Material from Richard’s Bay estuary, Natal.

RHB 51 1/2/49

RHB 52 27/1/49
RHB 76 14/7/49
RHB 85 16/7/49
RHB 113-114) 25/1/51

U and UU
U1 21/12/38
UU 2 18/7/40

Umbhlatuzi Lake, under 1 m.

Lower Channel, drifting material brought in by tide.
Lower Channel, drifting material.

Upper Channel, drifting material.

Lower Channel, drifting material.

Intertidal material from Umbhlali, Natal.

List oF SPECIES

Haleciidae
Halecium beani (Johnston)
Halecium inhacae n. sp.

Campanulariidae

Campanularia ?crenata (Hartlaub)

Campanularia integra Mac-

Gillivray

Campanularia morgansi Millard

Campanularia sp. -

Clytia gracilis (M. Sars)

Clytza johnston: (Alder)

Clytia serrata n. sp.

Obelia bicuspidata Clarke

Obelia dichotoma (Linnaeus)
Campanulinidae

Stegopoma fastigiata (Alder)
Lafoeidae

Filellum ?antarcticum (Hartlaub)

Hebella scandens (Bale)

Scandia mutabilis (Ritchie)

LKygophylax ?brarmata Billard

Kygophylax geminocarpa n. sp.
Kygophylax infundibulum n. sp.

Sertulariidae

Amphisbetia bidens (Bale)
Amphisbetia minima (Thompson)
Amphisbetia operculata (Linnaeus)
Dynamena crisioides Lamouroux
Dynamena obliqua Lamouroux
Dynamena quadridentata (Ellis and
Solander)
Salacia articulata (Pallas)
Sertularella arbuscula (Lamouroux)
Sertularella diaphana (Allman)
Sertularella dubia Billard
Sertularella flabellum (Allman)
Sertularella mediterranea Hartlaub
Sertularella polyzonias (Linnaeus)
Sertularella spp. (2)
Sertularia acuta (Stechow)
Sertularia distans (Lamouroux)
Sertularia ligulata Thornely
Sertularia linealis Warren
Sertularia turbinata (Lamouroux)
Stereotheca acanthostoma (Bale)
Thyroscyphus aequalis Warren
Thyroscyphus fruticosus (Esper)

168 ANNALS OF THE SOUTH AFRICAN MUSEUM

Syntheciidae
Hincksella corrugata n. sp.
Synthecium ?elegans Allman

Plumulariidae Plumularia warreni Stechow
Antennella secundaria (Gmelin) Pycnotheca mirabilis (Allman)
Halopteris glutinosa (Lamouroux) Aglaophenia late-carinata Allman
Kirchenpaueria adhaerens n. sp. Aglaophenia pluma (Linnaeus)
Monostaechas faurei n. sp. Halicornaria africana n. sp.
Monostaechas natalensis n. sp. Halicornaria arcuata (Lamouroux)
Paragatiya heurieli (Billard) Halicornarta gracilicaults ( Jaderholm)
Paragattya intermedia Warren Halicornaria hians (Busk)
Plumularia filicaulis Kirchenpauer Lytocarpus filamentosus (Lamarck)
Plumularia.irregularis n. sp. Lytocarpus philippinus (Kirchenpauer)
Plumularia setacea (Ellis & Solander) Thecocarpus formosus (Busk)
Plumularia spinulosa Bale Thecocarpus giardi Billard

Family Haleciidae

Halecium beani (Johnston) 1838

Halecium Beanii Hincks 1868, p. 224, pl. 43, fig. 2. Broch 1918, p. 38, fig. 13.
Halecium beanii Millard 1957, p. 188.

Records. MOR 218 C. DBN 329 J (reported by Day and Morgans, 1956,

as Halecium cf. halecinum). |

Description. ‘Two small colonies without gonangia, reaching a maximum
height of 2-1 cm. (MOR) and 3:1 cm. (DBN).

_ Halecium inhacae n.sp.
Big i

Holotype. IN 140 H.

Description. A single fairly rich colony growing on weed. Hydrorhiza with
internal thickenings of perisarc. Stem low, reaching a maximum height of 4 mm.,
unfascicled and generally unbranched. Nodes usually not visible; when present,
oblique. Internodes of variable length. Perisarc with internal thickenings
similar to those of hydrorhiza, giving a corrugated appearance, but no definite
annulations.

Hydrothecae alternate, the two rows in one plane. Primary hydrothecae
sessile, each borne on a broad apophysis arising from the distal half of a stem
internode but well below its terminal node. Margin of primary hydrotheca
perpendicular to axis of stem. Secondary hydrophores asymmetrical: peduncle
convex on adcauline side, straight or concave on abcauline side, with the
result that the whole structure bends away from stem. Margin of secondary
hydrotheca not perpendicular to hydrophore axis, but tilted obliquely towards
the adcauline side. Hydrotheca widening slightly to margin; walls straight
and not bent outwards. Diaphragm distinct, attached to the upper side of a
powerful thickening of the hydrophore wall, which is more heavily developed

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 169

eettaphee ee OS SIN:

Fic. 1. Halecium inhacae n. sp.

A, a single stem. B, portion of stem with 2 hydrophores. C, a female
gonangium containing a branching blastostyle and planula larva.

on the adcauline side. Row of puncta about midway between diaphragm and
margin. Hydranth with about 19 tentacles.

Female gonangia borne abundantly on hydrorhiza, and sometimes on
stem, where they always emerge from within the hydrothecae. Gonotheca
broadly oval in side view, flattened from side to side, very distinctly demarcated
from a short pedicel of one segment. Distal end broad, slightly convex, or
occasionally dished in the centre, bearing a terminal aperture at the end of a
raised collar on one side. Containing a branching blastostyle and a single
planula larva. No hydranths visible.

Measurements (mm.)

Internode, length see ve ie af .. 0°20—0°29
diameter across node P Bi 7. O°FO-0'13
Hydrotheca, depth, diaphragm to margin... .. 0°025-0°04
diameter at level of diaphragm .. br --. O'f1-0°16
diameter at margin .. un = a (O13 O°l 7
Gonotheca, length, excluding nenee St, a .. 0:67-0:78

maximum diameter fees os me ». 0°43-0°55

170 ' ANNALS OF THE SOUTH AFRICAN MUSEUM

Remarks. ‘The gonangia show resemblances to those of H. halecinum, but
are much broader at the base and do not taper downwards to the pedicel. The
general form of the colony, however, is different, and the shape of the secondary
hydrophores differs completely from that described by Broch 1918 and Vervoort
1946.

The secondary hydrophores resemble in shape those of H. expansum
Trebilcock 1928 from New Zealand, but the walls of the hydrothecae are
always thickened below and not at the level of the diaphragm. In H. expan-
sum, moreover, the two rows of hydrothecae are not in the same plane, and
there are no perisarcal thickenings on the stem.

Family Campanulariidae

Campanularia ?crenata (Hartl.) 1901

Fig. 2A-C, E

Eucopella crenata Hartl. 1901a, p. 364, pl. 22, figs. 27-31, 33-35. ?Billard 1906b, p. 170, fig. 3.
?Ritchie 1909, p. 73.
Orthopyxis crenata Bale 1924, p. 232, fig. 3.

Records. NA 117A, 184 E. IN 49 M.

Description. Colonies growing on weed. Pedicel arising direct from hydro-
rhiza; annulated at base and immediately below hydrotheca, smooth or wavy
in centre; perisarc of variable thickness.

Hydrotheca variable in shape and thickness of perisarc. Generally
triangular, with depth less than marginal diameter, but sometimes deeper and
narrower. Wall sometimes thin, sometimes greatly thickened on two opposite
sides to just below margin. Margin with 11-15 low, rounded teeth.

Gonophores absent.

Measurements (mm.)

Pedicel length .. i aD is a .. | (Oh 2=4eme
maximum diameter ae Me a, .. 0°08-0'°14
Hydrotheca, depth Ey ay i Es .. 0°33-0°50
diameter at margin .. A, a i .. 0*31-0°68

Remarks. This species cannot be identified with certainty in the absence
of gonophores and in view of certain differences from the description of the
type from New Zealand. In the latter the diameter of the hydrotheca is
always less than the depth, and the pedicel has less distinct annulations or none
at all. The South African material, however, resembles more closely Billard’s
specimens from the Sargasso Sea and Ritchie’s specimens from Cape Verde,
both of which were assigned with a query to the species. In Richie’s material
the diameter of the hydrotheca was equal to, or greater than, the depth.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 17!

Campanularia integra MacGill. 1842

Campanularia caliculata Warren 1908, p. 338, fig. 19.
Campanularia integra Broch 1918, p. 159. Millard 1957, p. 193.

Records. D173. IN 49 J.

E-F
a a ee |
0:5mn.

Fic. 2. Campanularia ?crenata (Hartl.) (A-C, E), and Campanularia sp. (D, F)

A-B, a normal thin-walled and a thick-walled hydrotheca from NA 184E. C, a narrower hydro-
theca from IN 49M. D,ahydrotheca from DBN 28. E, F, the diaphragms of A and D respectively.

Campanularia morgansi Millard 1957
Campanularia morgansi Millard 1957, p. 195, fig. 6.
Records. PF 12028 D.

Description. Colony growing on Sertularella dubia var. magna. Measurements
completely within range of those of the type, but hydrothecae with fewer
marginal teeth (8-11). No gonophores.

Campanularia sp.
Fie 2Di)F
Records. DBN 28, 270 W (reported by Day and Morgans 1956 as Clytia sp.).

Description. ‘Two small colonies growing on other hydroids. Pedicel wavy
or irregularly corrugated, with one distinct spherule below hydrotheca.

72 ANNALS OF THE SOUTH AFRICAN MUSEUM

Hydrotheca rather small and delicate, widening very slightly to margin.
Margin with 10-12 tongue-shaped teeth. No diaphragm, but an anuivar
thickening near base and a rather flattened basal chamber.

One rather damaged and empty gonotheca present, barrel-shaped, with
about 5 annulations and a truncated distal end.

Remarks. It is safer to leave the final identification of this species until
further material is available. The hydrotheca is similar to that of C. africana
Stechow, but smaller and more delicate. The gonotheca is different.

The absence of a true diaphragm excludes it from Clytia johnston (Alder), —
for as yet no South African specimens of the latter show variations so com-
pletely approaching the Campanularia type. The pedicel is also different.

Clytia gracilis (M. Sars) 1851
Fig. 3B, E, G

Gonothyraea gracilis Hincks 1868, p. 183, pl. 36, fig. 1.
Clytia gracilis Stechow 1925, p. 431, figs. 9-10. Millard 1957, p. 196.

Records. RHB 51 (recorded by Millard and Harrison 1954 as Clytia sp.).

Measurements (mm.)

Pedicel lengthy ae ia: ee ne .. 1:64-3:84
maximum diameter a si N .. O*10—0-12
Hydrotheca, depth we me: Be Ks .. 0:48-0:80
maximum diameter ae oe a -. 0°24-0°49
length/diameter a ait By. Ai .. 1°47 -2:0@
Gonophore, length ag As i mM .. 0°84-0°97
maximum diameter ve a ae .. 0°43-0°67

Clytia johnston (Alder) 1857
Fig. 3A, D, F
Clytia johnstont Hincks 1868, p. 143, pl. 24, fig. 1. Billard 1928, p. 456.

Campanularia johnstoni Broch 1918, p. 163. 1933, p. 93, fig. 40.
Campanularia raridentata Stechow 19192, p. 58, fig. Q

Records. IN 49 L.

Description. Hydrorhiza creeping on weed, giving off pedicels which are
generally unbranched, but occasionally branched to bear 2 hydrothecae.
Pedicel closely annulated at base and at distal end, smooth or with scattered
groups of annulations in middle portion.

Hydrotheca deep, expanding gently to margin. Wall smooth. Marginal
teeth 10-13 in number, triangular, with rounded tips, separated by deep,
rounded bays. Diaphragm well marked, of two distinct portions.

Gonotheca barrel-shaped, annulated, truncated at distal oe, with a
distinct collar.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 173

= = =
E = =
rsp) 9) 4)
Co) (@) (@}

A-C
D-E
F-H

Fic. 3. Clytia johnstoni (Alder) (A, D, F); Clytia gracilis (M. Sars) (B, E, G); and
Clytia serrata Nn. sp: (Cy Eh):

A-C, the hydrothecae. D-E, gonothecae. F—H, diaphragms.

Measurements (mm.)

Redicel lenoth * ).. at 2 ne oe .. 0°42-1'58
maximum diameter me ay by .. 0°07—0°09

Hydrotheca, depth my ie a ts .. 0°31-0°565
diameter at margin .. ce se oe .. O*2I-0°40
length/diameter a oe ae Bo, J) 1 Al—1-00

‘Gonotheca, length me Ne Ais a .. 0°55-0°88
maximum diameter “ He of 1) 0-32-00

Remarks. This is the first record of this cosmopolitan species from South
Africa. Due to the difficulty of distinguishing C. gracilis and C. johnstoni in the
absence of gonophores, only fruiting material of the two species has been
included.

Clytia serrata n. sp.
Bie. 19s) ot

Holotype. MOR 216 C.

Description. A small colony with 8 hydrothecae growing on the stem of
another hydroid. Hydrorhiza creeping, giving rise directly to the pedicels of
the hydrothecae.

174 ANNALS OF THE SOUTH AFRICAN MUSEUM

Pedicel unbranched, closely annulated at base and distal end, smooth or
with scattered groups of annulations in middle part.

Hydrotheca deep and narrow, about 3 times as deep as wide, with the
walls practically parallel in the distal half. Marginal teeth 8, very narrow and
needle-like, each bearing a longitudinal ridge which is continued down the
wall of the hydrotheca for a quarter to halfits length. Diaphragm very delicate.
Basal chamber deep and quadrangular.

Gonophores absent.

Measurements (mm.)

Pedicely length’ 42. Be a ve Ae .. 0°76-1-46
maximum diameter i - oF .. 0°05-0°065,

Hydrotheca, depth a iy si He .. 0°45-0°62
diameter at margin .. He A ae .. 0*16-0°20
length/diameter we 2h i 2°65-3°10

Remarks. ‘This species is very similar to Clytza kincaid: (Nutting) 1899,
differing from it only in its smaller size (the hydrotheca of C. kincaidi measures.
0-57—-0°65 mm. as judged from Fraser’s diagrams, 1944) and less extensive
longitudinal ridges. C. kincaidi, however, has only been found on the coasts of
N. America, and it is most probable that the gonophores of the two species will
be found to differ.

Obelia bicuspidata Clarke 1875

Laomedea bicuspidata Vervoort 1946, p. 298, fig. 132. 1946a, p. 344, fig. 10.
Records. DBN 89 S (reported by Day and Morgans, 1956). RHB 51 B.

Description. Slender, monosiphonic, sterile colonies reaching a maximum.
height of 7 mm. Stem occasionally branched in form of dichasium. Hydro-
thecal pedicel annulated throughout, or with a smooth, somewhat swollen
region in middle. Hydrotheca long and slender, with 8-11 pairs of slender,
rodlike teeth. In microscopic preparations striations are visible near the sides.
only.

Gonophores absent.

Measurements (mm.) | Durban Bay Richard’s Bay

Hydrotheca, depth... a By .. 0°38-0°51 0°3'7—0°46
maximum diameter mr ie ». O°19-0°27 0:18-0:22
length/diameter .. as NS ». 1°67-2°45 1°85—2°21

Remarks. This species has a more or less cosmopolitan distribution, but
this is the first record from South Africa.

Obelia dichotoma (Linn.) 1758
Obelia dichotoma Millard 1957, p. 198. |
Records. DBN 62 QC, 130 E (reported by Day and Morgans, 1956).
IN 100 B.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 175

Family CGampanulinidae
Stegopoma fastigiata (Alder) 1860
Calycella fastigiata Hincks 1868, p. 208; fig. 25; pl. 39, fig. 3.
Stegopoma fastigiata Fraser 1944, p. 178; pl. 32, fig. 153.

Records. AFR 1008 L.

Description. Hydrorhiza creeping on stem of <ygophylax ?biarmata. Hydro-
theca as in previous descriptions. Pedicel of variable length, either longer or
shorter than hydrotheca. Margin of hydrotheca often reduplicated, and a node,
indicative of regeneration, sometimes present in pedicel. Hydranth with about
11-13 tentacles. No gonophores.

Measurements (mm.)

Peaicel length .. 2H i bes eu .. 0°67-1°35
diameter near base bi - i .. 0°07—0:09
diameter at junction with hydrotheca.. .. O*10-O'15

Hydrotheca, depth sg si Sele eee .. 1:02—1°84
diameter at margin .. ie ab on: 53) 4 0294-0799

Remarks. This is the first record of the species from South Africa.

Family Lafoeidae
Filellum ?antarcticum (Hartl.) 1904

Lafoea antarctica Vanh6ffen 1910, p. 311, fig. 31.
Filellum antarcticum Stechow 1925a, p. 214.
Reticularia antarctica Totton 1930, p. 160.

Records. PE yri4k DD.

Description. Colony growing on Sertularella arbuscula. Margin of hydro-
theca slightly everted and usually reduplicated. Many of the hydrothecae free
for entire length. No coppiniae.

Measurements (mm.)

Hydrotheca, length of free part, without reduplications 0-36—0-9o0
length of free part, with reduplications .. .. 042-102
diameter at margin .. ft es aff .. O°16—-0-22

Remarks. There is little to distinguish F. antarcticum from F. serpens in the
absence of coppiniae, although Vanhdoffen states that the hydrothecae of the
former are ‘higher’ than those of the latter, and his measurements are much
greater than those given by Stechow 1925 for F. serpens. The measurements of
the present material are greater than those of the material from False Bay
(Millard 1957, Reticularia serpens), and it is solely on this basis that it is placed
provisionally in F’. antarcticum rather than F. serpens. Both species have been
recorded from South Africa, though the record of F. antarcticum is somewhat
doubtful (Stechow 19252).

I 76 ANNALS OF THE SOUTH AFRICAN MUSEUM

Hebella scandens (Bale) 1888
Hebella scandens Millard 1957, p. 202.
Records. DBN 267 F. PF 11323 B, 12392 B, 12456 H.

Scandia mutabilis (Ritchie) 1907
Scandia mutabilis Millard 1957, p. 202.

Records. IN 140 J.

Description. Colony growing on weeds and other hydroids. Pedicel
variable in length, with 3-9 annulations. Hydrothecae rather longer and
narrower than those reported from False Bay. Gonophores absent.

Measurements (mm., without reduplications. Base of hydrotheca taken as
first annulation below diaphragm)

Pedicel length” 2 oM a ae as .. 0°24-0°93
maximum diameter “i ah ie .. O-I4=o7e2
Hydrotheca, depth se ae ff oe .. Ys40-O°91
diameter at margin a af pit ». 0°50=0ng
length/diameter oh bs oe et .. T7Q=2556

Kygophylax ?biarmata Billard 1905
Fig. 4A

Kygophylax biarmata Billard 1906b, p. 180, fig. 8. Broch 1918, p. 24.
Records. AFR 1008 E.

Description. A colony of 4 fascicled branching stems reaching a maximum
height of 3:5 cm., in very poor condition and overgrown with sponges and
worm-tubes. Stem giving off alternate hydrothecae, and branches which are
roughly alternate and more or less in one plane, but the whole scheme is very
irregular. Each branch usually arises from the axial tube below a hydrotheca,
but in some cases no axillary hydrotheca is present. Peripheral tubes may
reconnect with the branches close to the origin of the latter, and may separate
in bundles from the stem to accompany worm-tubes and eventually connect
up with other parts. They may also bear irregular hydrothecae, nematothecae
and even branches.

Hydrotheca borne on apophysis of stem or branch, and separated from it
by a partial or complete node. Pedicel short, hydrotheca curved away from
stem, with adcauline wall convex. Many reduplications of margin. Diaphragm
present.

Nematotheca tubular with everted rim, separated from its short pedicel
by a delicate diaphragm, often much elongated by reduplications of margin.
Typically two on each apophysis, although many missing.

Gonophores absent. .

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA ERT

Measurements (mm., exclusive of reduplications)

Final branches, diameter - els ie .. 0:07-0:08
distance between 2 hydrothecae .. ei .. 0°42-0°58
Hydrothecal pedicel, length adcauline = .. 006-011
Hydrotheca, length adcauline Ap a .. O°3I-O'41
length abcauline.... aye ae ae Ran) O72 7-000
diameter at mouth .. ri a ae .. O°13-0°16
diameter at level of diaphragm .. x. .. 0:06-0:07
Nematotheca, length, including pedicel oe .. 0:08—-0°14
diameter at mouth .. a 26s a ». 0°035-0°05

Remarks. The exact identification of this species cannot be certain due to.
the poor condition of the material and the absence of gonophores. The hydro-
theca has a similar shape to that of <. africana Stechow, but is larger and
proportionally wider. The measurements are within range for those given for
XK. biarmaia and &. armata (Ritchie). <. biarmata has previously been reported
from islands in the Indian Ocean (Jarvis 1922), but never from the Union. It
can only be distinguished with certainty from <. armata by the structure of the
gonophores.

Kygophylax geminocarpa n. sp.
Fig. 4D-G

Holotype. PF 12308 A.

Description. Colony consisting of about 8 branching stems reaching a.
maximum height of 11:1 cm. Hydrorhiza forming a tangled mass up to 1 cm.
in diameter, clasping quantities of sand grains and other debris amongst its
tubes. Stem heavily fascicled, reaching a diameter of 3 mm., and branching in.
one plane. Stem bearing alternate hydrothecae, and alternate branches
arising generally at the base of every third and fourth hydrotheca. Branches.
varying greatly in thickness, a few heavily fascicled and rebranching, others.
unfascicled and unbranched. Each branch arising from an apophysis of the
axial tube of the stem and separated from it by a distinct node. ‘Axillary’
hydrotheca shifted on to upper side of apophysis. The branches are of two
types:

(i) Found mainly on the lower part of the stem, but also occurring irregularly
amongst type (ii). These branches may or may not be fascicled, they give
off alternate hydrothecae, and sometimes alternate subbranches from the
base of every third and fourth hydrotheca. The two rows of hydrothecae
and subbranches are in one plane, the same as that on the stem.

(ii) Forked branches found mainly on the upper part of the stem, and always
unfascicled. Each of these branches divides dichotomously immediately
beyond the apophysis which bears it, and the two limbs of the fork are in a.
plane at right angles to the normal branching plane of the colony. At the
angle of the fork and on its lower side (i.e. towards the base of the colony).

178 ANNALS OF THE SOUTH AFRICAN MUSEUM

is a single hydrotheca. After this each limb bears alternate hydrothecae,
of which the two rows are not in one plane, but shifted on to the inner
surface (so that the hydrothecae of the two limbs of the fork face towards
one another).

All branches with occasional transverse nodes at irregular intervals.
Occasionally the whole arrangement is complicated by the presence cf more
than one axial branching tube in the stem.

Hydrotheca resting on apophysis of stem or branch. Pedicel short, much
narrower than apophysis at its base, widening to diaphragm distally. Hydro-
theca with abcauline wall practically straight, adcauline wall convex for most
of its length, concave immediately below margin. Margin slightly everted,
forming an angle of approximately 45° with branch. Diaphragm formed by
thickened perisarcal ring. Cauline hydrothecae almost completely immersed
in the peripheral tubes of the stem.

Nematotheca tubular, widest at margin or just below, borne on a short
pedicel from which it is separated by a delicate diaphragm. 1-4 on each
hydrothecal apophysis (usually 2), and an irregular number on the peripheral
tubes of the stem.

Gonothecae not collected in coppiniae, but attached to one another in
pairs, and arranged in dense clusters around the main stem and principal
branches. Each gonotheca very large, elongated, round in section, tapering
to the base, and, more rapidly, to the tip, fused to its twin for about ? of length
and then free. Scattered nematothecae borne on lower half. The gonothecae
are not fully mature and have no openings to the exterior, nor can the sex be
determined. It is probable that an opening will develop on the inner surface
of the free distal part where a flattened area is present and where the end of
the blastostyle is pressed against the perisarc. The gonothecae are borne by
the peripheral tubes of the stem. Although most of them are arranged in pairs,
there are occasional solitary individuals or groups of three.

Measurements (mm.)

Final branches, diameter Ae ‘ Bs .. Ot12—0°22
distance between 2 hydrothecae .. + .. 0°365-0°485
Hydrothecal pedicel, length abcauline . . ay .. 0°06-0°12
Hydrotheca, length abcauline .. oi os .. 0°35-0°43
length adcauline... si a rf .. 0°34-0°42
diameter at mouth .. “by M i .. (ON7=Oezt
diameter at level of diaphragm .. ae .. O°10-0°15
Nematotheca, length including pedicel. . an /.° 0*09-=Orne
maximum diameter Ne By ih .. 0°05-0:06
Gonotheca, length re 4 MS haa .. 3°78—-4°15
maximum diameter across pair... af, 2. 146-1°75

Remarks. This species is unique in the dichotomous division of the branches,
and the twin gonothecae.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 179

ee US ee Ses ein:

8 mm.

B,O

Fic. 4. <ygophylax ?biarmata Billard (A); <ygophylax infundibulum n. sp. (B-C); and <ygophylax
geminocarpa n. sp. ( ve

B and D, portion of stem to show branching. A, C, E, F, portion of branch on larger
scale to show hydrothecae and nematothecae (C and F in side view). G, gonothecae.

180 ANNALS OF THE SOUTH AFRICAN MUSEUM

LK ygophylax infundibulum n. sp.
Fig. 4B-C

Holotype. PF 10781 B.

Description. A number of damaged stems reaching a maximum height of
6 cm. Hydrorhiza embedded in sponge. Stem fascicled, giving off alternate
hydrothecae, and alternate branches which generally arise from the base of
every third and fourth hydrotheca, though there are many irregularities.
Branches often subdividing in the same manner. All branches in one plane.
Final branches unfascicled; each borne on an apophysis of the axial tube of the
stem or branch, from which it is separated by a distinct node; bearing alternate
hydrothecae; unsegmented or with a rare transverse node midway between two
hydrothecae. The two rows of hydrothecae on stem and branches not in the
same plane, but shifted on to the anterior surface and forming a sharp angle
(about 15-40°) between them. The hydrotheca at the origin of each branch
not exactly in the axil but on the anterior surface of the apophysis.

Hydrotheca borne on a short apophysis of stem or branch and separated
from it by a transverse node. Pedicel about half length of hydrotheca, widening
slightly in distal region and not sharply demarcated from hydrotheca. Hydro-
theca elongated, curving upwards in the distal part, with abcauline wall
convex, and adcauline wall convex in proximal two-thirds and concave in
distal third. Margin facing upwards but also twisted slightly outwards, so that
its plane forms a right angle with the margin of a hydrotheca in the opposite
row. Diaphragm in form of thickened perisarcal ring, with inner edge bent
towards distal end of hydrotheca in form of inverted funnel; abcauline side
better developed than adcauline.

Nematotheca tubular, widening from base to two-thirds of height, then
narrowing to margin. One nematotheca on the apophysis next to each hydro-
thecal pedicel, but many are missing. None observed on the stem. At the
origin of a branch the nematotheca of the ‘axillary’ hydrotheca is seated on
the branch apophysis instead of on the hydrothecal apophysis.

The colony shows much evidence of regeneration following damage.
This may occur immediately beyond an apophysis resulting in a second short
segment, in the pedicel of the hydrotheca, or in the hydrotheca itself resulting
in a repetition of the diaphragm or reduplication of the margin.

Gonophores absent.

Measurements (mm.)

Final branches, distance between 2 hydrothecae .. 0°99 anal
diameter, above hydrotheca * se .. 0064007

Pedicel, length adcauline a i ue .. 0'18—Or88

Hydrotheca, total length. . ca big Eee .. 0°41-0°485
length adcauline.. ai nt i ». 0°29-0°38
diameter at mouth .. fr ha oy} .. OtLI-O'14
diameter at level of diaphragm .. np .. 0:06-0:07

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 18r

Nematotheca, length _.... am si 2: .. 0:06-0:07
maximum diameter EA 7 me ». 0°03-0°035,

Remarks. ‘This species is very close to <. unilateralis ‘Totton from New
Zealand, but differs from it in the proportions of the hydrothecae, which are
more slender. Very characteristic is the funnel-shaped diaphragm, which
_ rather resembles that of <. curvitheca Stechow 1913@.

Family Syntheciidae

Fitncksella corrugata n. sp.
Fig. 5
Holotype. PF 12456 J.

Description. Wydrorhiza creeping on
weed, forming a branching network.
Stem unbranched, unfascicled, reaching
a maximum height of 44 mm. and bearing
up to 5 hydrothecae. Basal part irregu-
larly corrugated or annulated, remainder
divided into internodes of rather irregular
length by oblique nodes. Each internode
bearing a hydrotheca, and with 1-3 corru-
gations immediately above node.

Hydrothecae alternate, the two rows
in the same plane, tubular, bending
slightly outwards, with about half the
adcauline wall adnate. Walls smooth or
with faint transverse corrugations. Margin
entire, round, slightly everted, forming an
angle of about 45° with stem, sometimes
reduplicated.

Gonophores absent.

Imm.

Measurements (mm.)

Stem, internode length ». 0°50-0°75
diameter at node .. .. O°15—-0°20
Hydrotheca, total length, in
GENIre .... se Orn Orme
length abcauline .. .. 0°53-0°69
length adcauline, adnate part 0:35-0:45
length adcauline, free part 0-28-0-42 pensar me
adnate part/adcauline length 0:47-0:61
diameter at margin -+ O°41-0°45 Fic. 5. Hincksella corrugata n. sp.

Remarks. This species is close to H. cylindrica (Bale), but differs in the
proportions of the hydrothecae, which are shorter and wider. In this character

182 ANNALS OF THE SOUTH AFRICAN MUSEUM ©

H. corrugata is intermediate between H. cylindrica and H. sibogae Billard (the
proportion of thecal diameter to length is approximately 1-2 in H. sibogae, 0-7
in H. corrugata, and 0-3-0-4 in H. cylindrica). H. corrugata also differs from
Hi, cylindrica in the greater adnate portion of the adcauline thecal wall (about
one-half as against one-third), in the less delicate wall to the hydrotheca, and
in the presence of corrugations on the stem and occasionally on the hydrotheca
as well.

Synthecium elegans Allman 1872
Synthecium ramosum Billard 1907, p. 359, fig. 8.
Synthecium ?elegans Millard 1957, p. 203, fig. 9D.
Record. PF 12308 F.

Description. A young colony of 9 simple stems reaching a maximum height
of 1:6 cm. Two of the stems end in stolons, and in 3 of them stolons arise from
within the hydrothecae. Shape and arrangement of hydrothecae similar to
the simple stems described from False Bay (Millard 1957), except that the
measurements are greater and a smaller proportion of the hydrotheca is adnate
to the stem.

Gonophores absent.

Measurements (mm.)

Stem, internode length .. By! sic « .. O°QO-1'17
Hydrotheca, length abcauline .. zk i. .. 0°44-0°76
length adcauline, adnate part ie oe .. 0'55-0°67
length adcauline, free part .. oe a .. 0°18-0°44
diameter near base .. oe Bs a .. OBt ony
diameter at margin .. ts ss As . 0:27-6°8q

Remarks. The measurements are very similar to ie of the material
from Madagascar, described by Billard 1907 as S. ramosum but assigned to
S. elegans in 1910.

Family Sertulariidae

Amphisbetia bidens (Bale) 1884

Sertularia bidens Bale 1884, p. 70; pl. 6, fig. 6; pl. 19, fig. 1. Warren 1908, p. 310, fig. 10.
Amphisbetia bidens Millard 1957, p. 220.

Records. RHB 76 B, 85 C, 114 E (recorded by Millard and Harrison 1954
as Thuiaria bidens). DBN 119 H (recorded by Day and Morgans 1956).

Description. Rich colonies reaching a maximum height of 13 cm. Nodes
of stem very distinct. No oblique nodes separating basal part of stem from
distal pinnate part (as described by Warren). Basal part of stem short (always
shorter than distal part) and sometimes branched, consisting of internodes of
irregular length, of which the distal one may bear hydrothecae, terminated by
a normal transverse node. Distal portion bearing alternate hydrocladia, one
to each internode. On the hydrocladia the two rows of hydrothecae are not
always in the same plane, but sometimes shifted on to one side, although never

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 183

in contact with one another. Marginal teeth of hydrothecae much longer than
in False Bay material.

Gonophores triangular in section, with 2 long spines produced from two
angles of the triangle, and very occasionally a third; present in January and

July.
Amphisbetia minima (Thompson) 1879
Amphisbetia minima Millard 1957, p. 221.
Records. IN 49 F, 140 F.

Description. Rich colonies growing on weed. Stems bearing up to I! pairs
of hydrothecae and reaching a maximum height of 0.4 cm.
Male gonophores present.

Amphisbetia operculata (Linn.) 1758 |

Amphisbetia operculata Millard 1957, p. 221.

Records. RHB 76 E, 85 D (recorded me Millard and Harrison 1954 as
Sertularia operculata). DBN 2 P, 119 J, 130 A (recorded by Day and Morgans

1956).
Dynamena crisioides Lamx. 1824

Dynamena tubuliformis Marktanner-Turneretscher 1890, p. 238; pl. 4, fig. 10.
Thuiaria tubuliformis Warren 1908, p. 314, fig. 12.
Dynamena crisioides Billard 1925, p. 181, figs. 36-37; pl. 7, fig. 21.
Records. D 170. DBN 267 E (recorded by Day and Morgans 1956).
IN 49 P. MOR 40 F.

Description. Stems reaching a maximum height of 2-1 cm., zigzag, pinnate.
Detailed structure similar to that in previous descriptions. Pairs of hydrothecae
on pinnae well separated and not overlapping, as illustrated by Warren.

Gonothecae of the typical shape for the species, arising from below the
hydrothecae, or, occasionally, from within the hydrothecae, present in January
and April.

Measurements. See var. gigantea.

var. giganiea Billard 1925
Fig. 6c

Thuzaria interrupta Allman 1886, p. 145; pl. 16, figs. 8-10.
Dynamena crisioides var. gigantea Billard 1925, p. 186, fig. 38F; pl. 8, fig. 24. Wedveurt 1941,

Pp. 210, fig. 4.
Records. MOR 7 X.

Description. A single colony, with long, straight stems reaching a maximum
height of 10 cm., and a diameter of 0-53 mm. near base. Stem divided into
internodes, each bearing 2-15 subopposite hydrothecae (usually about 9), and
one pinna arising below the first hydrotheca. Pinnae alternate, reaching a

184 ANNALS OF THE SOUTH AFRICAN MUSEUM

length of 15 mm., sometimes ending in tendrils, divided into internodes, each
bearing 2-5 pairs subopposite hydrothecae (usually 4).

Hydrothecae closely set and members of a group usually overlapping,
with only a very small portion of adcauline wall free.

Gonothecae present.

Measurements (mm.) Typical form v. gigantea
Hydrotheca, total height to top of adcauline :
tooth .. 0-5 1—-0-68 0°55—0°70
length abcauline .. 0°34-0°51 0-36—0°48
length adcauline, adnate part 0:36-0:65 0°53-0°67
length adcauline, free part 0°07-0°24, 0:05-0'1I
adnate part/adcauline length 0:63-0:88 0:84—-0-92
diameter at mouth O-12—-0'21 0-16-0-20
Gonotheca, length 0-98—1-37 2°13—2°16
maximum diameter 0°44—0°69 0-79-0°85

Remarks. This material differs from the typical form of the species mainly
in its growth-form and long straight stems. The hydrothecal pairs are also
more closely set, those on the pinnae usually overlapping one another. The
hydrothecae are a little longer than in the typical material, and have a smaller
proportion of the adcauline wall free. The gonothecae are larger, though of
the typical shape for the species.

Dynamena obliqua Lamx.
Fig. 64

Pasythea quadridentata Bale 1888, p. 770; pl. 14, figs. 6—7.
Pasythea quadridentata var. Balei Billard 1907, p. 355, fig. 6.
Dynamena obliqua Billard 1925, p. 198 (synonymy).

Records. PF 11803 AK, 12308 G.

Description. Two colonies reaching a maximum height of 11 mm. Hydro-
rhiza with internal ridges of perisarc in some areas only. Basal part of stem
terminated by 1 or 2 oblique ‘hinge-joints’. Remainder divided into inter-
nodes, each bearing 1 or 2 pairs of hydrothecae, but generally only one. Nodes
generally transverse and very indistinct, but occasionally oblique and distinct.

Members of a pair of hydrothecae in contact in front (except near base of
stem), separate behind. Hydrotheca broad with free part narrowing evenly to
margin. Margin facing outwards and slightly upwards, bearing two triangular
lateral teeth and a small adcauline one. Minute internal teeth present in some
hydrothecae only. Operculum of 2 valves, abcauline the larger. Hydranth
with no abcauline blind pouch.

Gonophores absent.

a a 2

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 185

Measurements (mm., adcauline length in isolated pairs only)

Distance between 2 pairs of hydrothecae ote oo 042-002
Hydrotheca, length abcauline (without teeth). . .. O'2I-0°34.
length adcauline, adnate part ae e: oy 0:27-0:36
length adcauline, free part he a .. O°12—0°16
diameter at mouth .. ie ae a .. Or12-0'14
diameter at base across pair =: ae .. 0°24-0°39

Fic. 6. Dynamena obliqua Lamx. (A); Dynamena quadridentata (Ellis & Sol.), var. nodosa Hargitt
(B); and Dynamena crisioides Lamx., var. gigantea Billard (C).

A, upper part of stem showing normal arrangement of hydrothecae from PF 12308G. B, an
unusual stem with small groups of hydrothecae from DBN 381G for comparison with A.
Opercular valves in dotted lines.

Remarks. Billard (1925) is of the opinion that D. obliqgua must remain
separate from D. quadridentata, and assigns to it material that has previously
been included in the latter species. In the South African collections both
species seem to be present, D. obliqua differing from D. quadridentata in the shape
of the hydrotheca, the tendency for a grouping in solitary pairs, and in the
measurements. Further, D. quadridentata is typically a littoral form and
D. obliqua occurs at depths of 24-66 m.

186 ANNALS OF THE SOUTH AFRICAN MUSEUM

D. obliqua is known from Australia (Bale), Mozambique (Billard 1907) and
possibly Japan (Stechow). This is the first record from the Union of South
Africa.

Dynamena quadridentata (Ell. & Sol.) 1786
var. nodosa Hargitt 1908
Fig. 6B
Pasythea quadridentata Warren 1908, p. 312, fig. 11.
Dynamena quadridentata var. nodosa Billard 1925, p. 197, fig. 43E. Leloup 1935, p. 43, fig. 25.

Records. NA 116 A. DBN 267 CG, 381 G (recorded by Day and Morgans
1956). IN 49 N.

Description. Wydrorhiza with internal ridges of perisarc in certain areas
only. Stems reaching a height of 0.6 cm. and bearing hydrothecae in groups
of 1-4 pairs. Basal part of stem of variable length, terminated by oblique
‘hinge-joint’. Nodes oblique, but sometimes the distal end of an internode is
cut off by an indistinct transverse node, and sometimes there are only these
transverse nodes to separate the groups of hydrothecae. Stem occasionally
terminating in stolon.

Hydrotheca slender and elongated, the lowest pair in a group swollen at
base as illustrated by Leloup. Internal teeth absent, or present (including
one adcauline and one or two abcauline). Operculum of 2 valves, abcauline
the larger. Hydranth without abcauline blind pouch.

Gonotheca with 4 transverse annulations and broad operculate aperture,
present in April (only one seen).

Measurements (mm., adcauline length in isolated pairs only)
Durban Bay — Kost Bay

Hydrotheca, length abcauline e .. O*17-0°34 0°19-0°35
length adcauline, adnate part .. .. 0°23-0°25
length adcauline, free part “i ». 0*°09-0°14
diameter at mouth ue ft .. 0°07—0:09 0:09—0°13
diameter across pair at base... .. 0*19—0:26 0:26-0°31
Gonotheca, length ... vs iM at: 0-98
maximum diameter hes ys ae 0-70

Remarks. The varieties of this species have been defined by Billard, and
var. nodosa again by Leloup. The species has previously been recorded from
the Natal coast by Warren 1908 and Vervoort 1946a. From the diagrams,
Warren’s material probably also belongs to var. nodosa. Vervoort’s material
was not illustrated.

Salacia articulata (Pallas) 1766
Salacia articulata Millard 1957, p. 208.
Records. PF 11803 AG, 12308 D, 12392 F.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 187

Genus SERTULARELLA

In a recent paper Picard (1956) has united Sertularella lagenoides Stechow
1g919a, S. mediterranea Hartl. 1901, S. ellis: (M.-Edw.) 1836, and S. fusiformis
(Hincks) 1861 (including f. glabra Broch 1933 and f. ornata Broch 1933), under
the single name of S. ellistt on the basis of intergrading forms occurring in the
Mediterranean.

He uses as his main diagnostic character the shape of the hydrotheca
which in his composite species is said to be bent towards the distal end of the
stem (i.e. the margin is not perpendicular to the axis of the hydrotheca but
tilted towards the adcauline side). This character is used to distinguish
S. ellista from a group of species (including S. polyzonias (Linn.) 1758) in which
the hydrotheca is bent towards the base of the stem.

I do not feel that this character has so great a diagnostic value as assigned
to it by Picard. A study of the literature shows that the hydrotheca of S. fusi-
formis is normally mid-way between the two types (i.e. with the margin perpen-
dicular to the hydrothecal axis), but can vary in both directions, as is clearly
shown by Broch’s figures (1933), and as confirmed by my own observations on
the South African material. Further, in the South African material of S. poly-
zonias and S. falsa Millard 1957 the hydrotheca is normally bent towards the
base of the stem, but occasional hydrothecae tend to approach the ellisiz type
by an elongation of the abcauline marginal tooth and a consequent tilting of
the margin towards the adcauline side.

Picard rejects the diagnostic value of the internal hydrothecal teeth, for
he unites in one species forms which have 3, 4, or none at all. Admittedly this
feature is variable in some species (e.g. S. fusiformis), but where it is constant in
all hydrothecae of colonies which obviously belong to the same species I can see
no reason why it should not have diagnostic value.

If we are to reject both hydrothecal shape and internal teeth as diagnostic
features, we would have to unite the South African species S. mediterranea,
S. fusiformis, S. falsa, S. capensis, S. africana and S. polyzonias, for all are unfascicled
_and have hydrothecae of similar size. This does not seem reasonable since
each has its own distinctive characters (Millard 1957).

Hitherto my diagnosis has depended firstly on the presence or absence of
internal teeth and their number and size, and only secondly on shape, recog-
nizing the fact that the latter may vary in certain species. For the internal
teeth do appear to be constant in the South African material, and even in S. fusi-
formis, where the number may vary, the teeth are always small and alternate
with the marginal teeth. S. mediterranea is one of the most constant, both as to
hydrothecal shape and its three internal teeth, and I have seen no variations
approaching the fusiformis type.

Although more extensive information may make it necessary in the future
to reduce some of these species to varieties, for the present it is safer, and
certainly can do no harm, to keep the species separate.

188 ANNALS OF THE SOUTH AFRICAN MUSEUM

Sertularella arbuscula (Lamx.) 1816

Sertularella arbuscula Millard 1957, p. 208, figs. 10B, 11.
Records. PF 10781 A, 11141 A, 11323 A, 11803 AH, 12028 C, 12392 A.

Description. Only one of the samples (PF 12392 A) bears gonophores, and
these are annulated throughout the distal $ of their length, the annulations
being very much more distinct than is usual for the species.

Sertularella diaphana (Allman) 1886
Fig. 7c, D

Sertularella lata Billard 1907, p. 346, fig. 4.
Sertularella diaphana Billard 1925, p. 157, fig. 22; pl. 7, figs. 12-13. ?Stechow 1925a, p. 226,
fig. H.

Records. IN 136A.

Description. A single bushy colony reaching 15 cm. in height, with a woody
rooting mass 3} cm. in diameter. Stem and principle branches fascicled,
branches given off irregularly. Distal parts unfascicled, divided into regular
internodes by oblique nodes, which slope alternately to right and left. Each
internode bearing one hydrotheca on one side, and two hydrothecae and a
hydrocladium between them on the other.

Hydrocladium bearing alternate hydrothecae; internodes of irregular
length, usually bearing 11-12 hydrothecae each, but occasionally as few as 2
or as many as 14; nodes oblique and often indistinct. The two rows of hydro-
thecae not diametrically opposite but shifted on to the anterior face, this con-
dition being more pronounced on the stem than on the hydrocladia. |

Hydrotheca completely adnate or with a very short free part, very slightly
turgid near base and then curved outwards. Margin facing outwards and
upwards, forming an angle of 25-45° with hydrocladium. Margin with 4 low
teeth. Operculum of 4 valves, usually missing. Abcauline wall with perisarcal
thickening below margin. No internal teeth.

Gonophores absent.

Measurements (mm.)

Hydrotheca (hydrocladial), length adcauline .. -. 0°43-0°48
length of free part .. st 4 gg .. 0:00-0:08
diameter at mouth .. ie ist fi -. 0°20-0°24,

Remarks. S. diaphana is difficult to distinguish from S. /ata (Bale) in the
absence of gonophores. Fertile material of S. diaphana has been recorded from
Portuguese East Africa by Billard 1907 and Jarvis 1922, and this fact, and the
angle of the hydrotheca margin (which is said to be almost parallel to the
hydrocladium in S. lata), makes it fairly certain that the present material
should be allocated to this species.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 189

Sertularella dubia Billard 1907
var. magna n. var.
Fig. 7A
Sertularella dubia Billard 1907, p. 344, fig. 3; pl. 25, fig. 1. (typical form).

Types and Records. Holotype PF 12028 B. Further records: PF ieeee H,
PF 12456 D.

0-5 mm.

A,B,D

Fic. ‘7. Sertularella dubia Billard, var. magna n. var. (A); Sertularella mediterranea Hartl.,
var. asymmetrica n. var. (B); and Sertularella diaphana (Allman) (C, D).
4, two hydrothecae from the holotype. B, part of a stem in side view.
C, part of a stem showing origin of hydrocladia. D, part of a hydrocladium.

190 ANNALS OF THE SOUTH AFRICAN MUSEUM

Description. Stiff, erect colonies reaching a maximum height of 10-6 cm..,.
and branching in one plane. Stem fascicled, bearing alternate branches usually
separated by 3 hydrothecae. Branches arising from immediately below hydro-
thecae. Nodes visible at distal ends of branches only.

Hydrothecae swollen, narrowing to margin, adnate for about 4 length of
adcauline wall, with margin not perpendicular to hydrothecal axis, but tilted
slightly away from stem. Abcauline wall straight to concave, with pronounced
perisarcal thickening at about 2 of its height. No internal teeth. Margin with

4 low teeth.
Gonophores absent.

Measurements (mm.) Holotype PF 12308 H

Internode, length 0°58—0°72 0°53-0°72
diameter across node 0°22—0°38 0°27-0'51

Hydrotheca, length abcauline 0-50-0°61 0°46-0°55
length adcauline, adnate part 0°37—-0°45 0°39-0°45
length adcauline, free part 0°40-0°49 0°39—0°54.
adnate part/adcauline length 0°45-0°52 0°43—0°52
diameter at mouth 0°25—0°29 0°30—-0°33
maximum diameter 0°34—0°38 0°39-0°43

Remarks. The appearance of the colony and the detailed structure agree
exactly with the description of the typical form from Macalonga (P.E. Africa).
Var. magna differs only in the dimensions of the hydrothecae which are about
14 times as great. The species is here recorded for the first time from the
Union of South Africa.

S. dubia resembles S. arbuscula in the shape of the hydrotheca, and in the
presence of a perisarcal thickening on the abcauline thecal wall, but differs in
the absence of internal teeth.

The size and shape of the hydrotheca also resemble those of S. xantha
Stechow 1923b, but the form of the colony is quite different. S. xantha has a
straggling, flexuous stem, and S. dubia a stiff, erect stem with shorter internodes.
The abcauline thecal thickening, which is pronounced in S. dubia and some
distance from the margin, is less pronounced or absent in S. xantha and, if
present, close to the margin. The hydrothecae of the latter also have a greater
proportion of the adcauline wall adnate. |

Sertularella flabellum (Allman) 1886
Sertularella flabellum Millard 1957, p. 212, figs. 10G, 11G.
Records. PF 12308 J.

Sertularella mediterranea Hartl. 1901
Sertularella mediterranea Millard 1957, p. 215, figs. 10E, 11B.
Records. D 169 (recorded by Eyre and Stephenson 1938). NA 114.
PF 11803 AJ.

Measurements. See var. asymmetrica.

ES ee
.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA IOI

var. asymmeirica n. var.
Fig. 7B
Holotype. IN 49 K.

Description. Small, unbranched stems growing on weed and reaching a
maximum height of 0-6 cm. The two rows of hydrothecae not in the same
plane but shifted towards one side of stem.

Hydrotheca similar to the typical form with 3 internal teeth and elongated
abcauline wall; but smaller in all its dimensions and with a smaller proportion
of the adcauline wall adnate. Many of the hydrothecae tend to be asymmetrical,
with one of the lateral marginal teeth very much longer than the other.
Operculum with 4 valves.

Female gonothecae present, shorter and more truncated than the male
of the typical form, with external marsupium and no marginal teeth.

Measurements (mm.) | var.
Typical asymmetrica
Internode, length 0°42—0°80 0°31-0-42
diameter across node 0°17—0°29 0°09-0°13
Hydrotheca, length abcauline 0°57—0°71 0°33-0°54.
length adcauline, adnate part 0°30—0°37 O-11-O'17
length adcauline, free part 0°34—0°42 0°24—0°40
adnate part/adcauline length 0°43-0°51 0°22—0°37
diameter at mouth 0°21-0:27 O°14-0°21
maximum diameter 0:29-0°35 0°22-0°29
Gonotheca, female, length 1:08-1:29
maximum diameter 0-75-0°89
Gonotheca, male, length 2°02—2:28
maximum diameter 0°83—0-94.

Remarks. At first appearance this material looks very different from the
typical S. mediterranea, but intermediate forms from other parts of the coast
show that there are no grounds for specific differentiation. The asymmetry of
the marginal hydrothecal teeth is found towards the distal parts of the stems,
but is usually not evident in the lower regions.

Sertularella polyzonias (Linn.) 1758

Sertularella polyzonias Millard 1957, p. 217, figs. 10J, 11H.
Records. AFR 1028 C. PF 12456 G.

Description. Immature colonies growing on other hydroids and reaching
a maximum height of 1-8 cm. Gonophores absent.
Measurements (mm.)
Internode, length
diameter across node

0°40-0°54
O-11—O°17

192 ANNALS OF THE SOUTH AFRICAN MUSEUM

Hydrotheca, length abcauline .. Bi 43 .. 0°34-0'52
length adcauline, adnate part fs A .» O25=oren
length adcauline, free part i. ne .. 0°23-0°31
adnate part/adcauline length be i .. 0°46-0°57
diameter at mouth .. = a o .. O0°18-0-22
maximum diameter P oe ae -. | OF23=0597

Remarks. The measurements of the internodes and hydrothecae are
‘somewhat less than those of the material from False Bay.

Sertularella sp.
Records. PF 12308 K.

Description. A fragment of an unfascicled stem with very large hydrothecae
reaching 1-42 mm. in abcauline length. Hydrotheca bent outwards, with just
under half adcauline wall adnate and 3 large internal teeth.

Remarks. ‘This is possibly a new species, but cannot be described as such
until more material is available.

Sertularella sp.
Records. PF 12456 E.

Description. Several rooted, unfascicled and unbranched stems reaching a
maximum height of 3:7 cm. Hydrothecae very large, reaching 1-92 mm. in
‘abcauline length, with just over half adcauline wall adnate, free part narrow
and tubular with sides almost parallel. No internal teeth.

Remarks. The shape of the hydrotheca resembles that of S. leiocarpa (All-
man) 1888, but a greater proportion is adnate, and the size is greater. In fact,
there appears to be no described species of Sertularella with such large hydro-
thecae. The material is probably that of a young colony, and without a know-
ledge of the form of the mature colony and the gonothecae I do not think it
advisable to establish a new species.

Sertularia acuta (Stechow) 1921a
Fig. 8a, F
Sertularia loculosa Warren 1908, p. 306, fig. 8; pl. 48, fig. 37. Bale 1913, p. 121; pl. 12, figs. 7-8.

Records. RHB 52 C, 85 G (recorded by Millard and Harrison 1954 as
Sertularia turbinata).

Description. Numerous unbranched stems growing on weed and reaching
a maximum height of 6mm. Stem with 1 or 2 spiral turns at base, remainder
divided into thecate internodes bounded by transverse nodes, with occasional
athecate internodes bounded distally by oblique nodes. Hydrothecae in
opposite pairs, members of a pair in contact in front, free behind.

Hydrotheca squat and swollen, with intrathecal ridge about half-way up
abcauline wall and extending round the sides for about # diameter. Margin
facing outwards and upwards, with 2 blunt lateral teeth and a very small

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 193°

adcauline one. Operculum of 2 valves, the abcauline the longer. No internal.
teeth.
Gonotheca with 7-8 annulations and broad distal aperture, not com--

pressed.

Measurements (mm.)

Thecate internode, length ny : .. 0°39-0°54.

Hydrotheca, length abcauline, without eth. .. O*16-0:21
length adcauline, contiguous part Sd ». 0°05-0°16
length adcauline, adnate part ee a -. O°19-0°23
length adcauline, free part ae a .. 0°08-0°13
adnate part/adcauline length o ais .. 0°62-0°71
diameter at margin .. ae she sf -. 0°0Q-0°12
diameter across pair (maximum) .. x .. 0°45-0°55

Gonotheca, length ms See ss os .. I*3I—1+50
maximum diameter a. i te .. 0:84-0:98

Remarks. See Sertularia turbinata.

Sertularia distans (Lamx.) 1816
var. gracilis Hassall 1848

Seriularia distans var. gracilis Millard 1957, p. 221, fig. 12.
Records. IN 140 G.

Description. A single colony with stems reaching a maximum height of
3 mm. and bearing up to 7 pairs hydrothecae. Detailed structure similar to
False Bay material, but measurements somewhat less. An internal tooth
present on abcauline wall of hydrotheca. Gonophores absent.

Measurements (mm., without lateral marginal teeth or reduplications)

Stem, length of basal part As a #3 .. 0*20—0°40
Internode, thecate, length an 40 ae .. 0°30-0°41
thecate, diameter at node aN ay -. 0°04-0:07
athecate, length of gi - ee .. O*°10—0°22
Hydrotheca, length, abcauline .. oe ER .. O*1Q—-0°21
length adcauline, adnate part es is 12) 10-17—0:20
length adcauline, free part be a .. OrrI—0-16
length adcauline, contiguous part m .. O13-O°15
diameter at base... oh aft on .. 0°07—0°09
diameter at margin .. oe be oe -. 0°07-0:08.

Sertularia ligulata Thornely 1904
Fig. 8c; 9A, B
Sertularia ligulata Billard 1925, p. 178, fig. 35. Leloup 1937, p. 44, fig. 30.
Records. PZ 13 E.

194 ANNALS OF THE SOUTH AFRICAN MUSEUM

Fic. 8. Sertularia acuta (Stechow) (A, F); Sertularia turbinata (Lamx.) (B); Sertularia ligulaat
Thornely (C); Sertularia linealis Warren (D, G); and Sertularia linealis var. longa n. var. (E).
A and F, stem and gonotheca from RHB 52C. C from PZ 13E. D and G,
stem and male gonophore from IN 49H. Opercular valves in dotted lines.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 195

Description. A number of stems reaching a maximum height of 7 mm.

- growing on floating weed. Stems unbranched, with the exception of one,
which has a single branch bearing two pairs of hydrothecae. Some stems
ending in stolons. ,

Structure agreeing exactly with Billard’s description, even to the finest
details, except that the stem is somewhat thicker in the region of the nodes,
which are not visible, and that the intrathecal ridge usually runs obliquely
downwards rather than obliquely upwards round the anterior surface of the
hydrotheca. Ligula not well preserved, but clearly visible in some of the
hydrothecae.

Gonophores absent.

Measurements (mm.) Thornely’s
PZ IDE type

Stem, distance between 2 consecutive pairs of
Hyarothecac).... = .. 0°49-0°67 0°55—-0°79
diameter in region of node O-12-0°2I 0145-021
Hydrotheca, length abcauline 0°20—0'29 0-22—0°26
length adcauline, adnate part 0°27—0°39 0°33-0°38
length adcauline, free part 0-12—0°20 0-10—0-16
adnate part/adcauline length 0-58—0-76 0:68-0:78
contiguous part 0*14—0°27 0-21—0°30
diameter at margin O°1I-O'15 0:16-0-19
maximum diameter across pair .. 0-48—0-62 0°48—0-64.

Remarks. I have seen a prepared slide of Thornely’s type material of this
species (fig. 9), and although the slide is unstained and the margins of most of
the hydrothecae damaged, the resemblance to my own material and to
Billard’s is close. Nodes are not visible or only very faintly indicated at the
edge of the stem, and the distance between consecutive pairs of hydrothecae is
variable, being greater on the stem than on the branches. In only the lowest
pair of hydrothecae can the margin be said to be ‘turned a little towards the
base’ as stated by Billard. In all others it faces outwards and slightly upwards.
The diameter at the margin is slightly greater than in my material or Billard’s.
See also remarks on S. turbinata.

Sertularia linealis Warren 1908
Fig. 8p, G
Sertularia linealis Warren 1908, p. 308, fig. 9. Jarvis 1922, p. 339.
Records: NA 117 B. IN 49 H.

Description. Two colonies growing on weed, with the hydrorhiza arranged

in the typical longitudinal lines. Stem reaching a maximum length of 5 mm.,

and bearing up to 8 pairs hydrothecae. 1 or 2 oblique hinge-joints at base, and
beyond this the internodes at first indistinct, and later distinct and oblique.

196 ANNALS OF THE SOUTH AFRICAN MUSEUM

Hydrothecae in opposite pairs; members of a pair contiguous in front,
except sometimes for the basal pair, separate behind. Hydrotheca bent out-
wards, narrowing to margin, margin facing outwards and slightly upwards.
Margin with 2 rounded lateral teeth, with perisarcal thickening around inner
edge and usually 3 internal teeth, 1 adcauline and 2 abcauline. Free part of
adcauline wall concave below margin. The free adcauline walls of a pair of

Sa er ee eee OI
at

A B

Fic. 9. Sertularia ligulata Thornely.

A, a portion of the stem, and B, a portion of a branch
from the holotype in the British Museum
(B.M. 1907.8.27.5).

hydrothecae forming more or less a straight line at right angles to the axis of

the stem.
Male gonophores as described by Warren: smooth, subglobular, and with

broad, operculate margin. Present in July.
Measurements. See var. longa.

Remarks. From Warren’s description it appears that the internal teeth
are not a constant feature, and this is supported by the present material. The
internal teeth are seated on the internal perisarcal thickening, and in the lower
parts of many stems tend to become minute and indistinguishable.

The species is known from Kosi Bay, Natal (Warren).

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 197

var. longa n. var.
Fig 85
Holotype. IN 140 E.

Description. Stem longer than typical form, reaching 9g mm., and bearing
up to 17 pairs hydrothecae. Nodes more distinct and more oblique. Inter-
nodes of variable length, longer towards base of stem. Perisarc in general much
thinner than in typical form, and marginal thickening of hydrotheca less
pronounced. Internal teeth generally absent, though observed in one hydro-
theca. Marginal teeth longer. A suggestion of an intrathecal ridge present,
visible on the lateral sides of the hydrotheca, but not on the abcauline wall.
Gonophores absent.

Measurements (mm.) Typical form var. longa
IN4g Hf NA117 B

Internode, length a .. 0°35-0°44. 0°37-0°44 0°36-0°48

Hydrotheca, length abcauline (without

teeth) 4 5 . O°13-0°20 0°13-0:20 0-16—-0-24

length adcauline, contiguous part .. 0:00-0:13 0:00—0°13 0:00-0:21

length adcauline, adnate part . 0°16-0:20 0:18-0:20 0-18-0-24.

length adcauline, free part . 0:08-0:14. 0-10-0°16 o-TI-0-18

diameter at margin JO:00—O7 FE) O° rO—-O-LE O-00-O"1s

diameter across pair Hinchicieis teeth) 0-35-0754. 0°36-0°53 0:43-0°63
Gonotheca, male, length .. 0°94-I°0I 0:90-0:96
maximum diameter . 0:48-0:70 0:68-0:80

Remarks. This material appears to be a variety of the typical S. linealis.
The absence of internal teeth would probably go together with the thinner
perisarc of the whole colony. The presence of an intrathecal ridge, however,
might be expected to occur in colonies with heavier perisarc, and this feature,
if constant, might justify the elevation of the variety to specific rank.

Sertularia turbinata (Lamx.) 1816
Fig 8B

Seriularia turbinata Billard 1925, p. 177, fig. 34.
Sertularia loculosa Jarvis 1922, p. 340.
non Sertularia turbinata Jarvis 1922, p. 341.

Records. PZ 13 B.

Description. Stem simple, reaching 2°5 cm. in length. Basal part unseg-
mented, terminated by hinge-joint. Remainder divided into thecate inter-
nodes. Nodes oblique, but often indistinct.

Members of a pair of hydrothecae in contact in front (except for one or
two pairs at base), separate behind. Hydrotheca swollen in lower part, free
part tapering to margin, intrathecal ridge present on abcauline wall and
continuing about half-way round sides. Aperture directed outwards. Margin

198 ANNALS OF THE SOUTH AFRICAN MUSEUM

with 2 bluntly pointed lateral teeth and a small median adcauline one;
generally thickened on abcauline side. Operculum of 2 valves, abcauline the
larger.

Gonophores absent.

Measurements (mm.)

Internode, length x a si a .. 0°66-0°85
diameter atnode .. a .. 0°0Q—0"14
Hydrotheca, length abcauline eathone tectange -. 0°25-0°31
length adcauline, adnate part oe ai ». 0°24-0°34.
length adcauline, contiguous part .. ae ». 0°00-0°25
length adcauline, free part . . sa: A .. 0*16-0:27
adnate part/total adcauline length By .. 0*49-0:68
diameter at margin . E i 2 ». O'L2—-0°17
diameter across pair Cael Voth) sh .. 0°68-0-86

Remarks. There has been considerable confusion in the literature over a
number of closely related species of Sertularia, but the position has been clarified
by Billard 1925, who figures Lamouroux’s type material of S. turbinata. He
recognizes the following three species.

(i) S. turbinata (Lamx.) = S. loculosa Busk = S. brevicyathus (Versluys).
(ii) S. acuta (Stechow) = S. loculosa Bale 1913 and Warren 1908.
(iti) S. ligulata Thornely = S. turbinata Bale 1913 (= S. turbinata (Stechow)
1925a?).
All three species occur on the South African coast. S. turbinata was pre-

viously reported from Zanzibar by Jarvis 1922 as S. brevicyathus, and S. acuta
from Natal by Warren 1908 as S. loculosa.

S. acuta can be distinguished from S. turbinata only by the shorter internodes,
and the hydrothecae which are more squat and which narrow more abruptly
towards the margin. These differences are illustrated by the accompanying
measurements of internode length, abcauline thecal length, and diameter of
margin in the two species. Another difference shown by the present material
is that S. turbinata has oblique nodes, and S.. acuta straight ones (excluding those
terminating athecate internodes). Billard suggests that these two species may
be synonymous, but until the gonophores of S. turbinata are known it is wiser
to keep them separate.

S. ligulata can be distinguished by its more upright hydrothecae, of which
a greater proportion is adnate to the stem, by its more poorly developed margi-
nal teeth and by the presence of a ligula attached to the hydranth.

I have seen a sample of Jarvis’s material from Zanzibar recorded as
S. turbinata and ascribed by Billard 1925 to S. ligulata, and it certainly does
not belong to either of these species. It has no abcauline intrathecal ridge and

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 199

the hydranth is without an abcauline blind pouch, which makes it a Dynamena,
and it appears to be closely related to D. cornicina McCrady.

Jarvis’s material recorded as S. loculosa Bale from Amirante in the Indian
Ocean, which I have also seen, can be included in S. turbinata, since the inter-
node length and the measurements of the hydrotheca agree well with this
species. The margin of the hydrotheca, however, is rather narrower (0-09
0°13 mm.).

Stereotheca acanthostoma (Bale) 1882

Sertularia acanthostoma Bale 1884, p. 85; pl. 4, figs. 7-8. 1913, p. 131. Billard 1907, p. 352.
Warren 1908, p. 303, fig. 7; pl. 46, figs. 23-26.

Records. D 168. RHB 52 B, 76 C, 85 E (recorded by Millard and Harrison,
1954).

Description. Stems reaching a maximum height of 4:2 cm. Detailed
structure as described by Billard and Warren. Gonophores absent.

Thyroscyphus aequalis Warren 1908

Thyroscyphus aequalis Warren 1908, p. 344, fig. 23; pl. 48, figs. 38-40.
Cnidoscyphus aequalis Splettstésser 1929, pp. 82, 124, figs. 78-82.

Records. AFR 1028 OA. PF 12028 A, 12308 E, 12456 A.

Description. Well-developed colonies reaching a maximum height of 20 cm.
Root-stock massive, up to 3 cm. in diameter, consisting of a mass of tangled
hydrorhizal tubes. Stem fascicled at base, branching in rather an irregular
manner, though usually alternate. Gonophores absent.

Thyroscyphus fruticosus (Esper) 1788-1830
Thyroscyphus fruticosus Splettst6sser 1929, pp. 7-30, 122, figs. 1-7. Vervoort 1941, p. 202.
Records. IN 49 D.

Description. Colony reaching 7:2 cm. in height, stem unfascicled, branching
in an irregularly alternate manner, and occasionally rebranching. Stem and
branches in one plane. Nodes not visible on stem and main branches,
occasionally faintly indicated near ends of smaller branches. Shape and
detailed structure of hydrothecae as described by Splettstésser, and measure-
ments well within range. Operculum present in young hydrothecae only.
Gonophores absent.

Family Plumulariidae

Antennella secundaria (Gmelin) 1788-1793

Antennella natalensis Warren 1908, p. 318, fig. 14.
Antenella secundaria Stechow 1925, p. 493. Broch 1933, p. 19.

Records. IN g1 Q.

200 ANNALS OF THE SOUTH AFRICAN MUSEUM

Description. One small colony without gonophores reaching a maximum
height of 0-5 cm. Stem simple. Basal athecate part of stem with 1-5 trans-
verse nodes and bearing 1-6 nematothecae. Intermediate internodes long and
slender, usually with 2 nematothecae, but occasionally with only one. Quite
often the basal part of the internode is cut off by an extra transverse node.
Gonophores absent.

Remarks. This is the first record of the species from Portuguese East Africa,
although it has been reported to the north on the Tanganyika coast, to the
east in Madagascar, and to the south in Natal.

Halopteris glutinosa (Lamx.) 1816
Fig. 10A—D

Heteroplon pluma Allman 1883, p. 32; pl. 8, figs. 1-3.
Plumularia glutinosa Billard 1910, p. 36, fig. 16. Stechow 1925, p. 502.

Records. D 155. DBN 70 P (recorded by Day and Morgans 1956 as
Halopteris sp.). RHB 52 E (recorded by Millard and Harrison 1954 as Plumu-
laria glutinosa). IN 49 G, 139 A.

Description. Stem unfascicled, pinnate, reaching a maximum height of
1°7 cm. Basal part without hydrothecae or hydrocladia, containing a few
irregular transverse nodes and bearing a few nematothecae. Distal part
divided by oblique nodes into thecate internodes, of which the basal 1 or 2
bear a pair of hydrocladia each, and the rest one each, alternately on the
right and the left. The nodes may be indistinct in some regions, or the first
2 or 3 nodes of this region may be extra well defined and resemble hinge-
joints.

Hydrocladium arising next to a hydrotheca immediately below the
lateral nematotheca, bearing up to 7 hydrothecae. Consisting of a short
athecate internode devoid of nematothecae, a longer athecate internode with
one mesial nematotheca, and then thecate internodes separated by oblique
nodes. Towards the distal end of the hydrocladium the upper part of each
internode is usually cut off by a transverse node to form an intermediate
athecate internode.

Hydrotheca adnate to a varying degree, but free for at least a third of its.
height, walls straight or flaring very slightly to margin, depth equal to, or
slightly less than, diameter.

Nematothecae, typical arrangement per thecate internode: one immove-
able, 2-chambered, mesial, inferior nematotheca below hydrotheca; one pair
moveable, 2-chambered, lateral nematothecae on processes arising just below
top of adnate part of hydrotheca and not quite reaching margin; one or two
mesial, superior nematothecae above hydrotheca, either 1- or 2-chambered.

Gonophores (not previously described) borne on hydrocladia below hydro-
thecae to one side of the inferior mesial nematothecae. Male elongated-oval,
present in July.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 201

Measurements (mm.) Da55 DEN 70P RAB 54> E. IN a9 G
Stem, internode length .. 0°40-0°47 0°56-0°74. 0°26-0°30 0°49-0°77
diameter .. ae ». O13-0°25 O°13-0°18 0:22-0:26 OI 1-020
Hydrocladium, internode length,
Ist short athecate .. ... O°10-0°12 0°06-O-II 0°05-0°09 0:07-0:10
ist long athecate .. .. 0°27-0°32 0°29-0°33 0°15—-0°22 0:20-0°30
normal thecate .. .. 0°27-0°45 O°2I-0°54 0°30-0°38 0:21-0:25
normal athecate (where
present) y O*19-0'21 0°19-0°33 0:06 0-16—-0°22

Hydrotheca, length abcauline.. 0-20—-0°26 0:23-0°25 O°15-O°17 0*2I1—0°24
length adcauline, free part 0-09—0°135 0°-08—0-10 0°09-0°12 0:08-0:09
free part/abcauline length 0-43-0-55 0°33-0°43 0-60-0°73 0°35-0°43

diameter at mouth .. 0:26-0:29 0°24. 0:°185-0:20 0:22-0:26
Gonophore, length, male .. 0°43-0°50
diameter .. a .. O°15-0°22

Remarks. This species appears to be very much more variable than is
indicated in previous descriptions, particularly in the presence or absence of
athecate internodes, and in the arrangement of nematothecae. The variability
can be best illustrated by the description of a single hydrocladium chosen from
D 155 (fig. 10a).

Thecate internodes nos. 1 and 2 have a short distal region (beyond the
hydrotheca) of 0.07—0.08 mm. (measured on anterior surface). The superior
mesial nematotheca is minute (0-035 mm.), 1-chambered, and situated in the
exact angle between the adcauline hydrothecal wall and the internode.

In internode no. 3 the superior mesial nematotheca is longer (0-05 mm.)
with an indistinct septum visible.

In internode no. 4 the distal region is longer (0-11 mm.) with a transverse
node faintly indicated immediately above the adnate part of the hydrotheca.
The superior mesial nematotheca (0-05 mm.) is distinctly 2-chambered, and
there is a supplementary mesial 2-chambered nematotheca above it (0-04 mm..).

In internodes nos. 5 and 6 the transverse node is quite distinct and cuts off
the distal part as an intermediate internode which carries the supplementary
nematotheca (0-04 mm. in no. 5 and 0:07 mm. in no. 6). In both internodes
the normal superior mesial nematotheca is absent, apparently a normal con-
dition, for there is no scar for its attachment. (Distal region including inter-
mediate internode measures 0-12 and 0:15 mm. respectively.)

A similar variation is found in the stem where the distal internodes may
have their upper regions cut off as athecate internodes. :

The cauline nematothecae are similar to those on the hydrocladia, although
in the lower parts of the stem there may be an extra pair of lateral nematothecae
or an extra superior mesial nematotheca above the level of the hydrothecal
margin.

In DBN 70 P, IN 49 C and IN 1909 A all the hydrocladial internodes are
as nos. 5 and 6 above (fig. 108). In RHB 52 E (fig. 10c) intermediate inter-

202 ANNALS OF THE SOUTH AFRICAN MUSEUM

O:5 mm.

A E

Fic. 10. Halopteris glutinosa (Lamx.) (A-D); and Paragattya heurteli (Billard) (E).
A, a hydrocladium from D 155 showing its 2nd to 5th thecate internodes to illustrate arrangement
of nematothecae as described in text, p. 201. B and C, portions of hydrocladia from IN 49C
and RHB 52E respectively. D, a male gonophore from D 155.

nodes are only rarely present, but the superior mesial nematothecae are missing
altogether.

One case of a branching hydrocladium was seen (D 155). The branch
arises from the first thecate internode of the hydrocladium, below the hydro-
theca and next to the mesial inferior nematotheca.

The species is closely related to H. valdiviae, differing from it in the shorter
lateral nematothecae and in the greater length of the free part of the hydro-

a a oe

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 203

- theca. It has been reported from Francis Bay on the south coast (Stechow),

but this is the first record from Portuguese East Africa.

Kirchenpaueria adhaerens n. sp.
Fig. 13F, G
Holotype. RHB 52 G.

Description. A minute epizootic form with its stem adherent to the posterior
surface of the stem of Paragattya intermedia Warren. Stem with no visible
segmentation, giving off alternate hydrocladia, each from a long apophysis.

Hydrocladia bearing up to 6 hydrothecae on the surface facing the stem
of the host (i.e. the hydrothecae face the same way as do the cauline hydro-
thecae of the host), divided by oblique nodes into thecate internodes, with an
occasional athecate internode at the base or between consecutive thecate
internodes.

Hydrotheca adnate for over half height, then free, margin facing forwards

_ and upwards, and forming an angle of 45-65° with the hydrocladium, width

slightly exceeding depth. Abcauline wall straight or slightly concave, free part
of adcauline wall concave. Hydranth with about 16 tentacles.

Each thecate internode with one very shallow, saucer-shaped, mono-
thalamic nematotheca below hydrotheca, and one naked sarcostyle above it.
A similar rudimentary nematotheca on the upper surface of each hydrocladial
apophysis. Nematothecae very delicate and often missing. .

Gonophores absent.

Kirchen-

Measurements (mm.) paueria Plumularia
adhaerens nova*

Internode length (posterior surface) .. 0-22—0°28 0-68-1°30
diameter near distal end 0:04-0:06 O° 10—0°12
Hydrotheca, length abcauline 0:06-0:07 0-15—-0-19

length of free part 0°02—0°03

free part/abcauline length 0-29—-0°50
diameter at mouth 0:07-0:10 0°15—0°20

Remarks. K. adhaerens has a similar growth-form to Plumularia nova Jarvis
1922 from Zanzibar. The two species are undoubtedly closely related, but
P. nova has greater measurements (the internodes are over three times as long,
and the hydrothecae over twice as deep, as those of K. adhaerens), and its hydro-
thecae are completely adnate and held at a different angle (the margin is
nearly at right angles to the internode).

In the author’s opinion the present species with its rudimentary inferior
nematothecae is more reminiscent of Oswaldella than Kirchenpaueria, and this
probably applies to P. nova as well. However, as long as branching hydro-

* Measurements calculated from diagrams given by Jarvis, 1922.

‘204 ANNALS OF THE SOUTH AFRICAN MUSEUM

cladia are taken as the diagnostic feature of Oswaldella the species must remain
in Kirchenpaueria. |

Monostaechas faurei n. sp.
Fig. 11
Holotype. PF 12028 F.

Description. Material consisting of 6 upright stems without rootstock,
reaching a maximum height of 2-4 cm., and a number of fragments.

Fic. 11. Monostaechas faurei n. sp.

A, a single hydrocladium to show method of branching. B, a single stem. C and
D, lateral and anterior views of 2 segments of a hydrocladium, C with a gonotheca.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 205,

Stem fascicled, branching in a roughly dichotomous manner. The tubes.
in the fascicled stem branch at intervals and are connected with one another
by pores containing bridges of coenosarc. On the surface they bear numerous.
nematothecae. The hydrocladia are formed from the continuation of these
tubes and arise from all surfaces of the stem in an irregular fashion, sometimes.
singly, sometimes in pairs and sometimes in clusters.

Hydrocladia branching in a sympodial manner. Each subsidiary hydro-.
cladium arises from the posterior surface of the previous one just below the
level of its first hydrotheca. The ‘axis’ so-formed (which consists of the basal
parts of successive hydrocladia) curves gently backwards, and the distal parts.
of the hydrocladia lie in one plane, but diverge slightly from one another.
Each hydrocladium accompanies, and is fused to, its predecessor for a short.
distance, and the coenosarc of the two is in communication through an opening
in the perisarc just below the second hydrotheca of the latter. The basal part
of each hydrocladium is of variable length and contains one or two transverse
nodes; in primary hydrocladia there are usually two soon after the origin from.
the stem, followed by a long internode without hydrothecae but bearing two.
rows of nematothecae; in subsidiary hydrocladia there is one transverse node
immediately beyond the communication with the previous hydrocladium, and
the following athecate internode is relatively short. The remainder of the
hydrocladium is divided by oblique nodes into thecate internodes.

Hydrotheca deep, completely adnate, with margin even and at right
angles to hydrocladium, broad just below the centre, slightly narrowed above
this and flaring again at the margin. The basal hydrotheca in each hydro-.
cladium shorter than the following ones.

Nematothecae bithalamic and moveable, 9 to each internode: one median
below hydrotheca and 4 pairs of lateral. The first lateral seated on a long
process arising near base of hydrotheca, the second in the angle between the
process and the hydrocladium, the third arising directly from the hydro-.
cladium near the upper part of the hydrotheca and just overtopping its margin,
and. the fourth arising directly from the hydrocladium above the level of the
hydrotheca. Occasionally the distal end of an internode is cut off by a trans-.
verse node which may cause some irregularity in the nematothecae.

Male gonophores borne on hydrocladia below hydrothecae, each is pear-.
shaped, about twice length of hydrotheca, borne on a short pedicel of 1 seg-.
ment, and bears one pair of nematothecae.

Measurements (mm.)

Internode length (on posterior surface) af ». 0°4.0-0°54
diameter (above hydrotheca) ie 4; -. 0:07—0°12
Hydrotheca, height a oe ae oe -» O°18—0:26
diameter at margin .. be ai ine 2. O°12-0°145
Nematotheca, height .. : | se -+ 0°07—0°095,
Gonotheca, male, length Guitheut pedieob) bt .. 0°36—0°46

maximum diameter | ims is ds es 0:20-0:26.

206 ANNALS OF THE SOUTH AFRICAN MUSEUM

Remarks. So far only three species of Monostaechas have been described
(M. quadridens (McCrady) = M. dichotoma Allman, M. fischeri Nutting, and
M. sibogae Billard), none of which has been recorded from South Africa.
M. faure: differs from all of these in the fascicled stem, the adnate hydrothecae,
and in the arrangement of the nematophores.

Monostaechas natalensis, n. sp.

Fig. 12

Types. Holotype PF 12456 C. Paratypes: PF 11803 AF and PF 12392 G.

Description (Holotype). A fairly rich colony reaching a maximum height
of 2-1 cm. Hydrorhiza reticular. Stem fascicled, short and stubby, often
embedded in sponge or ascidian. Hydrocladia arising from the stem or
directly from the hydrorhiza. Tubes of stem connected with one another by
pores and cross-branches, without nematothecae. |
The stem and hydrocladia often branch, but in rather an irregular fashion.
The tubes of the stem may divide dichotomously, give off lateral branches, or
give rise to tufts of hydrocladia at certain levels. The hydrocladia themselves
may be simple or may branch in a sympodial manner, a subsidiary hydro-
cladium arising from the posterior surface of its predecessor at any level, either
from the basal athecate part or from the distal thecate part. A hydrocladium
has not been seen to branch more than twice. A subsidiary hydrocladium may
be connected to its predecessor by a pore soon after its origin, as in M. faurei,
but this is the exception rather than the rule.
Basal part of hydrocladium athecate, consisting of 2-3 internodes separated
by transverse nodes, and bearing a double row of nematothecae. Remainder
consisting of thecate internodes separated by oblique nodes.
Hydrotheca moderately deep, adnate for over half its height, and then free,
walls flaring slightly towards margin. Margin even, not at right angles to hydro-
cladium but directed obliquely away from it, forming an angle of about 65°.
Nematothecae bithalamic and moveable, the number per internode
varying from 6 to 10, of which only the basal 5 are constant in position. They
include in all:
one median below hydrotheca on proximal end of internode,
one pair of laterals borne on long processes arising just below centre of hydro-
theca,
one pair of laterals arising from angle between above-mentioned processes and
internode,

one or a pair of supracalycines arising behind free part of hydrotheca and
overtopping its margin,

0, I, 2 or 3 median arising above level of hydrothecal margin on distal end of
internode. :

Occasionally the distal end of an internode is cut off by a transverse node
above the level of the hydrotheca to form an athecate internode, taking with
it one or two median nematothecae.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 207

0-5 mm.

Fic. 12. Monostaechas natalensis n. sp.

A from the paratype PF 11803AF, B-F from the holotype. A, a group of hydrocladia to show

method of branching. B, a branching stem and its hydrocladia. Cand D, a single segment from

the proximal and the distal ends respectively of the same hydrocladium in anterior view to show

the arrangement of nematothecae. E and F, lateral views of different hydrocladia, E with a
gonotheca.

208 ANNALS OF THE SOUTH AFRICAN MUSEUM

Gonothecae (female) arising from hydrocladia immediately below hydro-
thecae. Each is about three times the length of the hydrotheca, pear-shaped,
with wide distal aperture provided with operculum, bears 2 nematothecae,

and is seated on a short pedicel of 2 segments.

Measurements (mm.) Holotype Paratypes
Internode length (on posterior surface) 0:45-0:87 0°41—0:60
diameter (above hydrotheca) 0:08-0:14 0:08-0-12
Hydrotheca, height 0-17—0°26 0-16—0°22
height of free part 0-03-O'11 0°04—0:09-
free part/total length 0:17—0'46 0°22-0°43
diameter at margin 0°15—0:20 0-16—0°215,
Nematotheca, length i. 0:06-0:10 0:07-0:09
Gonotheca, female, length Gatton, pedicel 0:58-0:67
maximum diameter 0°32—0°39

Remarks. An interesting feature of this species is the variability in the
number and arrangement of nematothecae in hydrocladia of the same colony.
Unlike M. faurei, however, any nematothecae arising above the level of the
hydrothecal margin are single and not paired.

M. natalensis resembles M. fauret in the presence of a fascicled stem, but
does not possess the same graceful scheme of branching. It is clearly more
primitive, and, possessing both simple and branched hydrocladia, suggests the
manner of origin of the Monostaechas form. It is possible that the primitive
Antennella form gave rise to the Halopteris form on the one hand by lateral
branching of the hydrocladium and the formation of a pinnate plume, and to.
the Monostaechas form on the other hand by branching from the posterior
surface and the formation of a sympodium. This demonstrates the advisability
of keeping the three genera separate, for although Antennella could conceivably
be combined with Halopteris or with Monostaechas, due to the presence of inter-
grading forms, under no circumstances could Monostaechas be combined with
Halopteris.

Paragattya heurteli (Billard) 1907
Fig. 10E

Plumularia Heurteli Billard 1907, p. 360, figs. 9-10.
Plumularia quadridentata Jarvis 1922, p. 348; pl. 26, fig. 22.

Records. PF 12308 C.

Description. Five upright stems without rootstock, reaching a maximum.
height of 10 cm., one of them with two branches. Lower parts of stem and
branches slightly fascicled.

Hydrocladia-bearing ramules with 1-4 basal athecate internodes, of which
the distal one bears 2 bithalamic, moveable nematothecae similar to the
lateral ones. The rest divided into thecate, hydrocladia-bearing internodes by

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 209

oblique nodes. Of the latter the basal two are much better defined than the
rest and resemble hinge-joints.

Arrangement of hydrocladia and nematothecae as described by Billard
and Jarvis. The hydrothecae are set on the anterior surface of the hydrocladia
as in Aglaophenia, and not on the upper surface as in Plumularia. Some of the
hydrocladia end in swollen stolons.

Gonophores absent.

Measurements (mm.)
Hydrocladia-bearing ramules, internode length (on

posterior surface) .. Bes i Es 1 96-30-0798
diameter (above hydrotheca) i. ns .. O*II—0°20
Hydrocladia, internode length (on posterior surface).. 0°30—0°44
diameter (above hydrotheca) re ms .. 0:06-0:08
Hydrotheca, height, abcauline .. i sé .. O*115—0:26
diameter at mouth .. = al bstiyebs .. Or16-0-21

, Remarks. ‘This species has previously been recorded from Macalonga,
Portuguese East Africa (Billard) and from Pemba Is. (Jarvis), but it is a new
record for the Union of South Africa.

Paragattya intermedia Warren 1908

Paragattya intermedia Warren 1908, p. 323, fig. 16; pl. 47, fig. 27. Millard 1957, p. 230.
Records. RHB 52 F.

Plumularia filicaulis Kirchenpauer 1876
Fig. 13D, E
Plumularia filicaulis Bale 1884, p. 134, pl. 11, figs. 6-7, pl. 19, figs. 41-42. Leloup 1934, p. 4.
Records. IN 49 G.

Description. A fairly rich colony growing on an alga and containing both
simple and pinnate forms. Pinnate stems reaching a maximum height of
5 mm., simple stems reaching 2 mm. and bearing up to 5 hydrothecae.
Detailed structure exactly as in Bale’s description, to which the following
points may be added.

First 2 internodes of pinnate stem without hydrocladia, the second with
one nematotheca, the two separated by a rather indistinct oblique node, and
the second terminated by an oblique node more sharply demarcated than the
following ones. Stem decreasing in diameter towards tip, with very thick
perisarc on anterior surface. The two rows of hydrocladia not in the same
plane, but shifted slightly to the anterior side. Hydrothecae not borne on the
upper face of the hydrocladium as is usual in Plumularia, but on the anterior
face as in Aglaophenia.

First internode of hydrocladium and first internode of simple stem sterile,
with no hydrotheca or nematotheca, this followed by a normal thecate inter-
node on the hydrocladium, and a normal athecate internode on the simple

210 ANNALS OF THE SOUTH AFRICAN MUSEUM

stem. Perisarc on anterior surface very thick, and at times equal to the
diameter of the internode.

Lateral nematothecae trumpet-shaped, with diaphragm very close to
base. Only 2 cauline nematothecae visible on each stem internode, one near
proximal end, and one adjacent to origin of hydrocladial apophysis.

Ripe male gonophores present, each containing a series of dense masses
of spermatozoa. Transverse annulations very indistinct.

Measurements (mm., internode lengths taken on posterior surface)

Pinnate stem, diameter .. 1 if re -. 0°06-0°15
internode length .. oe ate Be .. 0°26-0°36
Hydrocladium, Ist internode, length .. 1% .. 0°08—-0-10
thecate internode, length .. ie i .. O*1G—-0°23
athecate internode, length bs ah .. O*10—-O'14
Simple stem, Ist internode, length a, ae .. _O*19-0:20
thecate internode, length .. ae i: .. | 02-0726
athecate internode, length .. #43 at .. O°13-0°19
Hydrotheca, depth in centre... si ae .. 0°09-0-12
diameter at margin .. .. O°175-0°21
Gonophore, including adherent expansion, ieneth .. 1:03-1°49
maximum diameter ag: aye ae .. 0*49-0:60

Remarks. This comparatively rare species has been reported only once
from South Africa (Table Bay: Leloup 1934). For Portuguese East Africa it
is a new record.

Plumularia irregularis n. sp.
Fig. 13A—C

Holotype. DBN 70 Q (recorded by Day and Morgans 1956 as Plumu-
laria sp.).

Description. Material consisting of 5 upright stems reaching a maximum
height of 1-2 cm., one with a single lateral branch. Stem slightly fascicled at
base, divided by transverse nodes into long internodes, each bearing a hydro-
cladium from an apophysis near the distal end.

Hydrocladia alternate, the two rows in one plane, bearing up to 5 hydro-
thecae, and very irregularly segmented. In the normal condition a hydro-
cladium appears to consist only of thecate internodes separated by transverse
nodes, but many intermediate internodes may occur, as many as three at the
base, and up to two between consecutive thecate internodes. Occasionally an
oblique node is present immediately behind the hydrotheca and in a line with
its posterior wall. Internodes very long and slender, hydrothecae borne near
the distal ends. |

Hydrotheca widening towards margin, adnate for a varying extent, but
with at least part of the adcauline wall free.

ee ee en See | O-5mm.

SO) Sinn:

sD |

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA

Fic. 13. Plumularta irregularis n. sp. (A-C); Plumularia filicaulis Kirch. (D, E); and Kirchenpaueria
adhaerens n. sp. (F, G).

A-C, portions of a colony to show variations in segmentation. D, two segments of a hydro-

cladium of a pinnate stem. E, a male gonophore containing masses of spermatogenic cells.

F, a hydrocladium in lateral view. G, anterior view of stem of Paragattya intermedia Warren with
a colony of Kirchenpaueria adhaerens on its posterior surface.

212 ANNALS OF THE SOUTH AFRICAN MUSEUM

Nematothecae bithalamic, with diaphragm close to base, outer surface
convex, margin oblique. Two to each thecate internode, one mesial inferior
and one mesial superior, and one on each stem internode immediately above
hydrocladial apophysis. One ‘mamelon’ on upper surface of each apophysis.

Gonophores absent.

Measurements (mm.)

Stem internode, length .. * Sf AM! .. O°3I-O°51
diameter (at distal end) .. WY Dae .. 0°09-0°10
Hydrocladium, thecate internode, length _.... .. 0°21-0°67
diameter (at distalend) .. bs Ne ». 0°045-0°055
Hydrotheca, height abcauline .. Me AE .. 0:08-O'11
height of free part .. Ne ae Ae .. 0°02—0:075
free part/abcauline height .. A i .. 0°18-0°75
diameter at mouth .. a Se. ue .. 0°095—-0'12
Nematotheca, length .. ie ot a .. 0°03-0°055

Remarks. Plumularia irregularis is related to a group of species which possess
only two mesial nematothecae to each hydrocladial internode, and in which
athecate internodes are normally missing, namely P. inermis Nutting 1900,
P. triangulata Totton 1930 and P. bonnevieae Billard 1906b. From these it
differs in having a hydrotheca which is not competely adnate. From the two
latter it also differs in the short distal part of the hydrocladial internode, in the
bithalamic nematothecae, and in the arrangement of the cauline nematothecae.

Plumularia setacea (Ellis & Sol.) 1755
Plumularia setacea Hincks 1868, p. 296; pl. 66, fig. 1. Millard 1957, p. 232.

Records. D 99, 154. B. DBN 199 J, 238 R (reported by Day and Morgans
1956). NA 212 M.

Remarks. The material from the Natal coast mostly falls within Broch’s
forma microtheca (1914), but intermediate stages between this and the typical
form also occur. On the whole Broch’s contention is supported that forma
microtheca is typical of tropical and subtropical waters, and the typical form of
temperate waters.

Plumularia spinulosa Bale 1882

var. typica Stechow 1923¢
Plumularia spinulosa Bale 1882, p. 30; pl. 15, fig. 8. 1884, p. 139; pl. 12, figs. 11-12. Warren
1908, p. 320.
Records. D 52, 154 A.

Description. Minute colonies reaching 0:25 cm. in height, and bearing up
to 13 hydrocladia. Lower part of stem with irregular internodes without
hydrocladia or nematothecae. Internodal septa very well marked, two at
base and one at distal end of each stem internode, one on hydrocladial apophy-
sis, one in first internode of hydrocladium and two in second internode behind

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 213

hydrotheca. Hydrocladial.apophysis arising from middle of stem internode.
Spine on end of hydrocladium long and sharp and reaching above margin of
hydrotheca. Gonophores absent.

Plumularia warrent Stechow 1919a
Plumularia tenuis Warren 1908, p. 316, fig. 13.

Records. D174. NA 111, 116 B, 184 A. U1 B (pp.). IN 140 A.

Description. Fairly abundant colonies reaching a maximum height of
2:2 cm. Stems occasionally branching near base. Gonophores present in July
and December.

Remarks. This species is impossible to distinguish from P. setacea in the
absence of gonophores, and to avoid mistakes only the fruiting material of the
two species has been considered.

Pycnotheca mirabilis (Allman) 1883

Pycnotheca mirabilis Millard 1957, p. 234.
Records. PF 12456 K.

Description. One plume 10 mm. in length, without rootstock or gonophores.

var. warreni ‘Totton 1930

Kirchenpaueria mirabilis Warren 1908, p. 321, fig. 15.
Pycnotheca mirabilis subspecies warreni Totton 1930, p. 216.

Records. RHB 52 D (recorded by Millard and Harrison 1954 at Kirchen-
pauerta mirabilis). IN 49 B, 140 B. PF 11803 AE.

Description. A fragment 0-7 cm. long from Richard’s Bay, a number of
stems reaching 1-5 cm. from Inhaca Is. and 4 stems reaching 0-9 cm. from off
Durnford Point, Natal. Ends of hydrocladia sometimes forming tendrils. The
material from Inhaca bears gonophores on the hydrorhiza (July) ; these contain
one elongated body, which is either an embryo or a mass of gonadial tissue.

Aglaophenia late-carinata Allman 1877

Fig. 14.
Aglaophenia late-carinata Allman 1886, p. 151; pl. 23, figs. 5-6.
Aglaophenia minuta Nutting 1900, p. 96; pl. 21, figs. 1-3. Billard 1906b, p. 230, fig. 19.
Aglaophenia latecarinata var. madagascariensis Billard 1907, p. 387; pl. 26, figs. 18-19.

Records. PF 11803 AB, 12456 B.

Description. Numerous plumes growing on sponges, worm-tubes and
ascidians, reaching a maximum height of 1-9 cm. Hydrorhiza without the
annulations described by Nutting. Basal part of stem without hydrocladia,
but with a single row of nematothecae on anterior surface, terminated by two
closely placed oblique hinge-joints with one nematotheca between them.
Remainder bearing alternate hydrocladia. Segmentation usually visible in
distal region only.

214 ANNALS OF THE SOUTH AFRICAN MUSEUM

Hydrocladial internodes and hydrothecae as previously described.
Hydrothecal margin with 9 teeth, one solid median and four pairs of lateral;
of the latter the 1st and 4th are narrow and triangular, and the 2nd and 3rd
broad and rounded. ‘Keel’ on anterior surface of hydrotheca slightly longer
than median tooth. Perisarcal thickening on abcauline wall of median
nematotheca of variable thickness, sometimes very pronounced, sometimes
scarcely visible.

Cauline nematothecae: one large below hydrocladial apophysis, one
axillary anterior, one axillary posterior, and one mamelon on anterior surface
of apophysis.

Fic. 14. Aglaophenia late-carinata Allman.

A, lateral view, and B, anterior view of two segments of a
hydrocladium in PF 11803AB. C, lateral view in PF 12456B.

One corbula present (in February), about 2:3 mm. long and 1 mm. wide,
and bearing 10 pairs of ribs. Detailed structure as described by Nutting.

Measurements (mm.) PF 11803 AB PF 12456 B

Hydrocladium, internode length ae ». O123—-0°25 0°29—-0°37
width in centre .. ay aie .. 0:06-0:085 0:08—0°10

Hydrotheca, height .. At a .. 0°24-0'27 0:30-0:36
width at mouth (excluding keel) .. O'12-O'14 0°135-0°16

Remarks. This species is common in the western tropical and subtropical
regions of the Atlantic, where it generally grows on floating weed. It has,
however, been recorded by Jarvis 1922 from Cargados in the Indian Ocean

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 215

(as A. minuta), and a separate variety has been described by Billard 1907 (var.
madagascariensis) from Madagascar. This is the first record from South Africa.

Billard’s variety differs from the type in the nature of the hydrothecal
margin which has 2 pairs of bifid lateral teeth instead of 4 pairs of single teeth.
The present material seems to combine features of the typical form (described
and illustrated by Billard 1906b) and the variety from Madagascar. The
lateral hydrothecal teeth and the arrangement of cauline nematothecae are
closer to the typical form, but the shape of the median hydrothecal tooth and
the size of the median nematotheca are closer to the form from Madagascar.

The measurements are rather larger in one sample (PF 12456 B) than
in the other (PF 11803 AB), but on the whole the range extends over both the
typical and the Madagascan types.

Aglaophema pluma (Linn.) 1758
var. typica Bedot 1919
Aglaophenia pluma, var. typica Millard 1957, p. 238, fig. 15A.
Records. PF 12392 E.

Description. One fragment 1-2 cm. in length, without rootstock or corbulae.

var. parvula Bale 1882
Aglaophenia pluma var. parvula Millard 1957, p. 239, fig. 15D-F.
Records. D 203. NA 112. DBN 200 Q (reported by Day and Morgans
1956).

Measurements (mm.)

Maximum height of colony a a ays en tAcOO
Stem internode, length .. we Bi a .. 0*16—0°26
Hydrocladial internode, length .. Ae a .. . 0°21-0:28
Hydrotheca, depth y ie te ne 0°20-0'27
length above median nematotheca By .. 0:06-0:09
diameter at margin a a. bs ». O*15-0°19
Corbula, male, length .. ae sb a! .. ItQ—1°46
male, width .. ve s a ie i 0770-0797
female, length Me ue ae ih ve 1722
female, width i By ls th Wf 0°84

Remarks. The colonies from Natal are smaller than those reported from
False Bay, and the dimensions of the internodes, hydrothecae and corbulae
are also less.

Halicornaria africana n. sp.
Fig. 15A—-C
Holotype. AFR 1028 B.

Description. A single colony of about a dozen monosiphonic stems arising
from a matted hydrorhiza and reaching a maximum height of 16-5 cm. Basal

216 ANNALS OF THE SOUTH AFRICAN MUSEUM

part of stem without hydrocladia, divided into internodes of irregular length
by transverse nodes, bearing scattered nematothecae. Distal part divided into
regular internodes by transverse nodes, each bearing a pair of opposite or
subopposite hydrocladia. Three cauline nematothecae around the base of each
hydrocladium, one below the origin, one posterior axillary, and one anterior
axillary. Towards the distal end of the stem the axillary nematothecae have
part of their margins drawn out into long slender tubes.

Hydrocladia reaching 11 mm. in length and bearing up to 39 hydro-
thecae. The two rows in one plane or shifted slightly on to the anterior surface.
Each divided by very slightly oblique nodes into regular thecate internodes.
No internodal septa.

Hydrotheca deep, widening to margin, its central axis forming an angle
of 40-50° with hydrocladium. No intrathecal ridge. Margin with three pairs
of low lateral teeth and one anterior tooth. No posterior tooth. Towards the
basal part of the hydrocladium the lateral teeth are indistinct and the margin
appears sinuated.

Median nematotheca long and gently curved, adnate as far as the anterior
tooth, then free, tapering to tip which may be closed, but is usually open.
Lateral nematotheca saccular, not reaching margin of hydrotheca, with a single
aperture of which the lateral corner may be raised into a short tubular structure
towards the distal end of the hydrocladium.

Gonophores absent, though scars for their attachment are present on the
anterior faces of the hydrocladial apophyses.

Measurements (mm.)

Stem, diameter \ i By ah oe .. 0°285-0°74
imternoce length > °- st sh Be .. 0°46—0-68
Hydrocladium, diameter at node Kf ne .. O°12-0°24
internode length .. 0° of se .. O27 =O-8%
Hydrotheca, length oe He a Ph .. 0°20-0°25
diameter at mouth, lateral view .. ne .. _0'16—OFng
diameter at mouth, frontal view .. 2 i 0°19—0'24.
Median nematotheca, length of free part - .. OTG=naea

Remarks. ‘This species is very close to H. regalis Totton 1930 from New
Zealand, but differs in the shape of the hydrocladial internode, which is
shorter and thicker, and in the nature of the hydrothecal margin. In Totton’s
diagram the 3rd lateral tooth is the largest of the three, but in H. africana this
tooth is always the smallest. The anterior tooth of H. africana is larger and more
upright than in H. regalis.

H. africana differs from H. arcuata (Lamx.) in the absence of the posterior
tooth on the hydrothecal margin, the more erect position of the anterior tooth,
and in the different shape of the median nematotheca.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA ay

O-3mm.

B= Cc

Fic. 15. Halicornaria africana n. sp. (A-C); Halicornaria arcuata (Lamx.) (D, E); Halicornaria
arcuata, var. epizootica n. var. (F); Halicornaria hians (Busk) (G, H); and Halicornaria gracilicaulis

(Jad.) (1, J).

A, a single stem. B-—J, two segments of a hydrocladium. B and C, and D and E, from the distal
and proximal ends respectively of the same hydrocladium. I from PF 11141B. J from PF 12028E.

218 ANNALS OF THE SOUTH AFRICAN MUSEUM

Halicornaria arcuata (Lamx.) 1816
Fig. 15D, E
Halicornaria arcuata Billard 1907, p. 366, fig. 13. 1910, p. 46. Bale 1913, p. 141; pl. 13, figs. 1-4.
Records. PF 11803 AC.

Description. A small colony of 9 unbranched stems reaching a maximum
height of 1-8 cm. Stem internodes short and broad. Hydrocladia alternate,
one to each internode.

Hydrotheca deep and narrow, without a true intrathecal ridge but with
basal part of abcauline wall projecting inwards and bearing 2 or 3 small
denticles. Margin with an incurved anterior tooth, an incurved posterior
tooth, and 2 or 3 pairs lateral teeth. The median lateral tooth is always smaller
than the others and often absent altogether.

Median nematotheca always open at tip, with its distal end pointing
obliquely forwards. Lateral nematotheca with two openings, of which one
may be produced into a tube.

Gonophores absent.

Remarks. This material combines some of the features of the young
colonies and some of the old colonies described by Billard 1907 from Mada-
gascar. Bale (1913) doubted whether Billard’s material all belonged to the
same species, but the present material seems to show that Billard was right,
and that the species is variable in the nature of the hydrothecal margin, and
in the presence or absence of an opening at the end of the median nematotheca.
H. ascidioides is a separate species as shown by Bale.

var. epizoolica n. var.
Fig. 15F

HAHolotype. PF 12392 D.

Description. Wydrorhiza epizootic on back of stem of Thecocarpus formosus
(Busk), giving rise directly to a number of solitary hydrocladia and to one
upright stem of 1-g mm. in length.

Stem internodes longer and narrower than those of typical variety, each
giving rise to one hydrocladium. Hydrocladia alternate, each bearing up to
3 hydrothecae.

Solitary hydrocladia alternating with those of host, bearing up to 18
hydrothecae and reaching a length of 4:2 mm.

Hydrotheca exactly like those of typical form. Median nematotheca open
at tip, often rather short. Lateral nematotheca with one of its apertures
forming a tubular structure towards the distal end of the hydrocladium.

Remarks. In view of the similarity in structure to H. arcuata it is not
justifiable to create a separate species for this material. An almost parallel —
case within the genus is to be found in H. longirostris Kirch. 1872 from Australia,
which can exist in a parasitic and in a free-living form.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 219

Halicornaria gracilicaulis (Jad.) 1903

Fig. 15 1, J
Lytocarpus gracilicaulis Jad. 1903, p. 299; pl. 14, figs. 3-4.
Hralicornaria gracilicaulis Billard 1907, p. 364, fig. 12; pl. 25, fig. 7. 1913, p. 63.
Records. PF 11141 B, 12028 E. :

Description. Both samples without rootstock, one consisting of two strongly
fascicled, branched stems reaching a maximum height of 9:6 cm., and the
other of a few detached fragments.

Appearance and composition of colony similar to that described and
figured by Billard 1907 from Macalonga. Branches and pinnae of stem all
in one plane; pinnae alternate, often giving rise to alternate sub-pinnae.
Branches and pinnae fascicled practically throughout, only the tips of the
smaller pinnae unfascicled. Pinnae without hinge-joints at base.

Hydrocladia borne on stem and pinnae, bearing up to 6 hydrothecae.
Internodal septa very well marked and varying in number from 4 to 6. The
arrangement with 4 septa resembles that described by Billard in the material
from Macalonga. When 6 septa occur there is one opposite the adcauline
intrathecal septum, one opposite the base of the lateral nematotheca, two
between these, one at the proximal end of the internode and one at the
‘distal end. ;

Hydrotheca with a thick, but not very long, adcauline intrathecal septum, .
and a very pronounced perisarcal thickening at the point of curvature on the
abcauline wall, the condition approaching that of H. longicornis (Busk). Margin
smooth or sinuated with indications of three low teeth on each side, no posterior
‘tooth.

_ Median nematotheca of variable length, sometimes not quite reaching to
margin of hydrotheca, and sometimes very much longer as in Billard’s var.
armata (1913).

Gonothecae present (in February), flattened and disc-shaped, with a slit-
shaped terminal aperture and no distal horns.

Remarks. Although this species has been recorded from Portuguese East
Africa, this is the first record from the Union. The material shows certain
features which approach the condition in H. longicornis (Busk), namely the
pronounced perisarcal thickening on the abcauline thecal wall, and the
sinuated thecal margin, and it is possible that these two species may eventually
have to be united. This must wait, however, until the gonophores of the
latter have been described.

Halicornaria hians (Busk) 1852
Fig. 15G, H
Halicornaria hians Bale 1884, p. 179; pl. 13, fig. 6; pl. 16, fig. 7. Billard 1913, p. 68. Vervoort
1941, p. 222, figs. 7-8.
Halicornaria hians var. profunda Ritchie 1910, p. 24; pl. 4, figs. 13-14.

220 ANNALS OF THE SOUTH AFRICAN MUSEUM

Records. PF 11141 C.

Description. One colony including 3 pinnate stems reaching a maximum
length of o-g cm. Hydrorhiza epizootic on Halicornaria gracilicaulis ( Jad.) and
penetrating between the tubes of its fascicled stem. Stem delicate, unfascicled,
bearing alternate hydrocladia, one to each internode. Hydrocladia about
2 mm. long and bearing up to 6 hydrothecae, distinctly segmented.

Hydrothecae well separated, margin with 3 distinct teeth on each side.
Abcauline intrathecal septum thickened and upturned at end, and bearing a
number of minute denticles.

Median nematotheca of variable length, adnate part not reaching level of
intrathecal septum and tip not reaching thecal margin in lower parts of hydro-
cladia, adnate part reaching beyond level of septum and tip overreaching level
of marginal teeth in upper parts. Abcauline wall convex, with perisarcal
thickening usually present. Lateral nematotheca not reaching thecal margin.

Gonophores absent.

Measurements (mm.)

Hydrocladium, internode length Be is .+ O30=G74m
diameter at node +... ee Hf an ... 0*09-0°13

Hydrotheca, height Ae fis ae a .. 0°13-0°265
diameter at margin .. ae a ay: .. 0*16-0°22

Remarks. This species is well known from the Dutch East Indies, from
which region it extends east to the Pacific Ocean, south to Australia and west
to the Bay of Bengal. Var. bale: has also been reported from the Red Sea.
This is the first record from South Africa, and the delicacy of the stems suggests.
a young colony.

The structure of the hydrotheca agrees well with that of the typical form
as described by Bale and Ritchie, but the material differs in the presence of
only one hydrocladium to each internode of the stem. Marktanner 1890,
however, reports that var. balei does rarely possess one hydrocladium to an
internode, although the normal condition is two or more.

Lytocarpus filamentosus (Lamarck) 1816

Lytocarpus filamentosus Millard 1957, p. 241.
Records. DBN 119 L (recorded by Day and Morgans 1956). PF 11803 AD.

Lytocarpus philippinus (Kirch.) 1872

Lytocarpus Phillipinus Bale 1888, p. 786; pl. 21, figs. 5-7.
Lytocarpus philippinus Markt. 1890, p. 274; pl. 6, fig. 16.

Records. IN 41, 49 A, 137 A, 138 A, 140 GC. PZ 13 A.

Description. Colonies with root-stock, reaching 14:0 cm. in height.
Gonophores present in July.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 221

Thecocarpus formosus (Busk) 1850
Aglaophenia formosa Markt. 1890, p. 264; pl. 6, fig. 11.
Thecocarpus formosus Billard 1907, p. 378, figs. 19-20. Vervoort 1946a, p. 332, fig. 7.
Aglaophenia parasitica Warren 1908, p. 332, fig. 17; pl. 48, figs. 28-32.
Records. DBN 119 K (recorded by Day and Morgans 1956). RHB 52 A,
76 D, 85 A, 113 K (recorded by Millard and Harrison 1954.as T. formosus and
Meenarasiicus). D o7 A. UU 2 J. PF.11803 AA, 12392 C.

Description. Colonies abundant along the Natal coast, often growing on
Corallines or other algae, reaching a maximum height of 11-6 cm. Hydro-
rhiza reticular and creeping, producing suckers which can penetrate into the
substratum.

Corbulae of two kinds. The first, which is probably female, is closed, and
has up to 12 pairs of ribs and may reach 4 mm. in length. In addition to the
series of lateral leaflets described and figured by Warren (pl. 48, fig. 31), each
rib may bear two more lateral leaflets, the second arising about half-way up
its length and the third near the distal end. The length of the leaflets often
exceeds that of the ribs, giving to the corbula a spinous appearance.

The second type of corbula, presumably male, is generally longer, and
more open, the distal ends of the ribs being separated by pores, and easily
forced apart. Only one series of small lateral leaflets is present, and the
appearance resembles Warren’s diagram (pl. 48, fig. 30).

As suspected by Vervoort the corbula of both the parasitic and the normal
type bears a hydrotheca near the base of each rib.

Remarks. ‘Totton 1930 and Vervoort 1946a consider A. parasitica to be a
synonym for 7. formosus. The former is characterized by suckers on the hydro-
rhiza which penetrate into the tissues of coralline algae. Totton has observed
similar suckers on the lectotype of 7. formosus. The author has found that
suckers are invariably present, whether the colony is growing on algae, or on
worm-tubes, ascidian tests, etc., and that these are capable of penetrating a
considerable distance into pebbles and barnacle shells. Under the latter con-
ditions they can, of course, act as holdfasts only, which is probably their
primary function, but there is no reason to doubt they they can also obtain
nourishment from host algae. They must certainly secrete some substance
capable of dissolving calcium.

Thecocarpus giard: Billard 1907
Fig. 164

Thecocarpus Giardi Billard 1907, p. 381, fig. 21; pl. 25, fig. 9; pl. 26, figs. 11-16.
? Aglaophenia(?) bifida Stechow 1923b, p. 117. 1925, p. 515, fig. 53.
Thecocarpus giardi Vervoort 1946a, p. 335. Millard 1957, p. 240.

Records. PF 12308 B.

Description. One colony with a large rooting mass and a single stem 9°5 cm.
in length. Lateral hydrothecal teeth bifurcated as in Billard’s description. No.
corbulae.

ae ANNALS OF THE SOUTH AFRICAN MUSEUM

Measurements (mm.) False Bay* Natal A. bifidat
Hydrocladium, internode length .. 0:30—0-36 0:25-0:29 0°34—-0°39
Hydrotheca, height (to tip of inner

point of median tooth) .. 0-26—0-30 0:28-0°32 0°32—0°36
diameter at margin (inside) .. 0-17-0-21 0°15-0'18 0°22—-0'23
Median nematotheca, length .. 0:21-0:27 0-14-0°18 0°19-0°23
length of free part My, .. 0°04-0:07 0°02—0°04. 0:04-0:05

Remarks. See var. solidus.

0:5 mm.

Fic. 16. Thecocarpus giardi Billard (A); and Thecocarpus giardi, var. solidus n. var. (B, C).
B and C from the holotype, B showing a rare hydrotheca with a hollow ‘keel’.

var. solidus, n. var.
Fig. 168, c

Types and Records. Holotype AFR 1028 A. Further record: PF 12308 L.

Description (Holotype). A well-developed colony with a broad rooting
mass and about g fascicled stems, reaching a height of 55 cm. Stem branching
in sympodial manner to form alternate pinnae, the whole twisted in the spiral
manner typical of the species. Pinna divided into a basal part devoid of
hydrocladia, and a distal part bearing alternate hydrocladia, the boundary
between the two commonly marked by 1 or 2 hinge-joints.

* From material described by Millard 1957.
+ From type material loaned by the Munich Museum. The measurements are not quite

comparable— the hydrotheca height was taken to the tip of the outer point of the median tooth,
and the diameter is an outside one.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 223
Hydrotheca with g teeth, 1 median and 4 pairs lateral. Median tooth
bifurcated, with a small inner point and a longer outer one which is of variable
length and practically always solid. First and second pairs of lateral teeth not
bifurcated.
Corbulae exactly as described by Billard for the typical form, present in.

May.
Measurements (mm.) Holotype PF 12308 L
Hydrocladium, internode length .. 0°27-0°34 0°30—0°35
Hydrotheca, height (to tip of inner point of
median tooth) ae .. 0°26-0°31 0°27—-0°34
diameter at margin (inside) 0-15—0°20 0-14-0719
Median nematotheca, length .. 0-18—0°24 O-19—0°25
length of free part 0°03—0°05 0:04—0°05
Corbula, length 5°04—-7°14
diameter 0-98-12

Remarks. As in many other Indian Ocean species T. giardi shows con--
siderable variability. The stems always branch in the form of a sympodium
(each pinna arising from the anterior surface of the preceding one), and are
spirally twisted. The appearance, however, varies from a tight spiral (as
illustrated by Billard, pl. 25, fig. 9) with a geniculate ‘main axis’, in which the
spaces between consecutive pinnae are approximately 3 mm., to a much
looser spiral and a less geniculate axis in which the pinnae may be as much as
15 mm. apart.

Each pinna, after its origin from its predecessor, is divided into a basal
part without hydrocladia and with one nematotheca to each internode, and a.
distal part bearing alternate hydrocladia. The transition between these two
parts may occur after the origin of the next pinna, or before it (in which case
the ‘main axis’ will bear hydrocladia). Both arrangements are shown occurring
on the same colony in Billard’s fig. 21, but in my experience one arrangement
is usually characteristic of a colony. Further, at the boundary between the two
regions there may occur I, 2 or 3 hinge-joints. In AFR 1028 A and
PF 12308 L hinge-joints occur on practically every pinna, and always distal
to the origin of the subsequent pinna. In many cases the pinna appears to
have broken off at this point and later regenerated, the regenerated portion
being of lesser diameter and sometimes bearing a second group of hinge-
joints. In another colony (to be described in a later paper) a few groups of
hinge-joints occur on the ‘main axis’ itself, i.e. proximal to the origin of the
next pinna. So far hinge-joints have only been observed in var. solidus.

Variation also occurs in the marginal teeth of the hydrotheca. Generally
there is one median tooth and 4 pairs of lateral ones, but the first and second
paired teeth may be bifurcated. This, however, has not been observed in var.
solidus. ‘The median tooth is always double, but the length of the outer point
or ‘keel’ varies considerably. In the typical form the ‘keel’ is usually hollow,

224 ANNALS OF THE SOUTH AFRICAN MUSEUM

although solid in occasional hydrothecae. Sometimes the tip appears to be
worn down, so that the cavity is open to the exterior. In var. solidus the ‘keel’
is generally solid, but occasionally hollow. The size of the median nemato-
theca is also variable, in some cases it does not reach to the level of the intra-
thecal septum, in others well beyond it. The extreme forms of the hydrotheca
are very different in appearance, but as all the features show intergrading
stages there does not seem to be any grounds for specific differentiation.

The corbulae are generally larger than those described by Billard. They
show variation in the development of the ‘crest’ at the base of each rib, but are
identical in all other details. One case of a double corbula occurs (AFR 1028 A)
in which the pedicel bifurcates in the 5th segment giving rise to 2 corbulae
which face one another and are fused together to the level of the 2nd rib. Only
closed corbulae have been observed.

Aglaophenia(?) bifida Stechow 1923b from the Agulhas Bank is possibly a
synonym for J. giardi. Stechow mentions ‘einzelne Fiedern’ reaching 6 cm.
in length. These may be either young colonies or detached portions of an older
colony. I have seen a prepared slide of Stechow’s material loaned by the
Munich Museum, and the measurements are well within range.

DIscussION

It is not proposed to deal with the geographical distribution in any detail,
as this can be done more profitably when the species of the south and west
coasts of the Union have been described.

Ten species and five varieties of known species described in this paper
are new to science, and it is noteworthy that of these, six species and three
varieties have been taken in dredgings off the coast of Natal by the s.s. Pieter
Faure and r.s. Africana in depths ranging from 24 to 155 metres. The shallow
infratidal waters off Natal and Portuguese East Africa are still very poorly
known and further dredgings in this area may be expected to yield many more
new species and records.

In addition eight species represent completely new records from the coast
of Africa south of the equator. Five of these have a very scattered distribution
and may be considered cosmopolitan (Campanularia crenata (Hartl.), Clytia
johnston. (Alder), Obelia bicuspidata Clarke, Stegopoma fastigiata (Alder) and
Kygophylax biarmata Billard). The other three (Sertularia ligulata Thornely,
Aglaophenia late-carinata Allman and Halicornaria hians (Busk)) have a more
restricted distribution but are known from the Indian Ocean, so their presence
is not remarkable.

Four species are recorded for the first time from Natal waters (all taken by
the s.s. Pieter Faure), although previously reported from Portuguese East Africa,
so their known range is extended further south. These include Dynamena
obliqua Lamx., Sertularella dubia Billard, Paragattya heurieli (Billard) and Hali-
cornaria gracilicaulis ( Jad.).

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 22 5

Finally six species are recorded for the first time from Portuguese East
Africa, although known previously from the Union, so that their known range
is extended further north, namely Dynamena quadridentata (Ell. & Sol.), Anten-
nella secundaria (Gmelin), Halopteris glutinosa (Lamx.), Plumularia _filicaults
Kirch., P. warreni Stechow and Pycnotheca mirabilis (Allman).

SUMMARY

This paper records a total of sixty-six species of Hydrozoa from the coasts
of Natal and Portuguese East Africa. Ten of these are new to science, and
include Halecium inhacae, Clytia serrata, Xygophylax geminocarpa, X<ygophylax
infundibulum, Hincksella corrugata, Kirchenpaueria adhaerens, Monostaechas fauret,
Monostaechas natalensis, Plumularia irregularis and Halicornaria africana. In
addition there are eight records new to Africa south of the equator, four new to
the Union of South Africa and six new to Portuguese East Africa.

REFERENCES

Allman, G. J., 1883. Report on the Hydroida dredged by H.M.S. Challenger during the years
1873-76. Part I. Plumularidae. Rep. Sci. Res. H.M.S. Challenger, Zool., 7, (20) 1-54.
Allman, G. J., 1886. Description of Australian, Cape and other Hydroida, mostly new, from
the collection of Miss H. Gatty. 7. Linn. Soc. Lond., Zool., 19, 132-161.

Bale, W. M., 1882. On the Hydroida of South-Eastern Lonieia with descriptions of supposed
new sreves and notes on the genus Aglaophenia. 7. Micr. Soc. Victoria, 2, 15-48.

Bale, W. M., 1884. Catalogue of the Australian Hydroid Zoophytes. Sydney.

Bale, W. M., 1888. On some new and rare Hydroida in the Australian Museum collection.
Proc. Linn. Soc. N.S. Wales, (2), 3, 745-799.

Bale, W. M., 1913. Further Notes on Australian Hydroids. II. Proc. Roy. Soc. Victoria (n.s.),
26, 114-147.

Bale, W. M., 1924. Report on some Hydroids from the New Zealand coast, with notes on New
Zealand Hydroida generally, supplementing Farquhar’s list. Trans. N.Z. Inst., 55, 225-268.

Billard, A., t906b. Hydroides. Expéd. Sci. ‘Travailleur’ et ‘Talisman’, 8, 153-241.

Billard, A., 1907. Hydroides de Madagascar et du Sud-Est de l’Afrique. Arch. Zool. exp. gén.,
(4), 7, 335-396.

Billard, A., 1910. Revision d’une partie de la collection des Hydroides du British Museum.
Ann. Sci. Nat. Paris, Zool., (9), 11, 1-67.

Billard, A., 1913. Les Hydroides de Expédition du Siboga. I. Plumulariidae. Szboga-
Expeditie, monogr. 7a, I-115.

Billard, A., 1925. ‘Les Hydroides de PExpédition du Siboga. II. Synthecidae et Sertularidae.
Siboga-Expeditie, monogr. 7b, 1-116.

Billard, A., 1928. Clytia Fohnstoni Alder, Campanularia raridentata Alder et Thaumantias inconspicua
Forbes. Bull. Mus. Nat. d’ Hist. nat., no. 6, 456-457.

Broch, H., 1914. Hydrozoa benthonica. Michaelsen’s Beitrége zur Kenntnis der Meeresfauna West-
afrikas, 1, 19-50.

Broch, H., 1918. Hydroida II. Danish Ingolf Exped., 5, 1-205.

Broch, H., 1933. Zur Kenntnis der adriatischen Hydroidenfauna von Split. Norske Videns.-
Akad. Oslo, I. Mat.-Naturv. Klasse, no. 4, 1-115.

Day, J. H., and Morgans, J. F. C., 1956. The Ecology of South African Estuaries. Part 8. The
Biology of Durban Bay. Ann. Natal Mus., 13, 259-312.

Eyre, J., and Stephenson, T. A., 1938. The South African Intertidal Zone and its relation to
Ocean Currents. V. A Sub-tropical Indian Ocean Shore. Ann. Natal Mus., 9, 21-46.

Fraser, C. McL., 1944. Hydroids of the Atlantic Coast of North America. Toronto.

Hartlaub, C., rg01a. Hydroiden aus dem Stillen Ocean. ool. Farhb., 14, 349-379.

Hincks, T., 1868. A History of the British Hydroid Zoophytes. London.

226 ANNALS OF THE SOUTH AFRICAN MUSEUM

Jaderholm, E., 1903. Aussereuropdische Hydroiden im schwedischen Reichsmuseum. Ark.
Kool., X, 259-312.

Jarvis, F. E., 1922. The Hydroids from the Chagos, Seychelles and other Islands and from the
coasts of British East Africa and Zanzibar. Trans. Linn. Soc. London, 18, 331-360.

Leloup, E., 1934. Trois Hydropolypes de la Baie de la Table, Afrique Australe. Bull. Mus.
Hist. Nat. Belg., 10, (19), 1-7.

Leloup, E., 1935. Hydraires Calyptoblastiques des Indes Occidentales. Mém. Mus. Hist. Nat.
Belg., sér. 2, fasc. 2, 1-73.

Leloup, E., 1937. Hydropolypes et Scyphopolypes recueillis par C. Dawydoff sur les cétes de
l’Indochine Francaise. Mem. Mus. Hist. Nat. Belg., sér 2, fasc. 12, 1-73.

Marktanner-Turneretscher, G., 1890. Die Hydroiden des k.k. naturhistorischen Hofmuseums.
Ann. k.k. naturh. Hofmus., 5, 195-286.

Millard, N. A. H., and Harrison, A. D., 1954. The Ecology of South African Estuaries. Part V>
Richard’s Bay. Trans. Roy. Soc. S. Afr., 34, 157-179.

Millard, N. A. H., 1957. “The Hydrozoa of False Bay, South Africa. Ann. S. Afr. Mus., 43, 173—
243.

Nutting, C. C., 1900. American Hydroids. Part I. The Plumularidae. Spec. Bull. Smithsonian
Inst., Washington, no. 4, 1-285.

Picard, J., 1956. Les espéces et formes méditerranéennes du genre Sertularella. Vie et Milieu,
7, (2), 258-266.

Ritchie, J., 1909. Supplementary Report on the Hydroids of the Scottish National Antarctic
Expedition. Trans. Roy. Soc. Edinburgh, 47, 65-101.

Ritchie, J., 1910. The Hydroids of the Indian Museum. Rec. Ind. Mus., 5, 1-30.

Splettstésser, W., 1929. Beitrage zur Kenntnis der Sertulariiden. Thyroscyphus Allm., Cnido-
scyphus nov. gen., Parascyphus Ritchie. Zool. Fahrb., Syst., 58, 1-134.

Stechow, E., 1913a. Hydroidpolypen der japanischen Ostkiiste. II. Teil: Campanularidae,
Halecidae, Lafoeidae, Campanulinidae und Sertularidae nebst Erganzungen zu den
Athecata und Plumularidae. Abh. bayer. Akad. Wiss., Math.-phys. Kl., suppl. 3, 1-162.

Stechow, E., 1g919a. Zur Kenntnis der Hydroidenfauna des Mittelmeeres, Amerikas und
anderer Gebiete. ool. Fahrb., Syst., 42, 1-172.

Stechow, E., 1923b. Ueber Hydroiden der Deutschen Tiefsee-Expedition,. mene Bemerkungen
iiber einige andre Formen. Zool. Anz., 56, 97-1109.

Stechow, E., 1925. Hydroiden der Denechen Tiefsee-Expedition. Wassenschaftliche Ergebnisse
der Deutschen Tiefsee-Exped. auf dem Dampfer ‘Valdivia’ 1898-1899, 17, 383-546.

Stechow, E., 1925a. Hydroiden von West- und Siidwest Australien nach den Sammlungen
von Prof. Dr. Michaelsen und Prof. Dr. Hartmeyer. Zool. Fahrb., 50, 191-269.

Totton, A. K., 1930. Coelenterata. Part V. Hydroida. Brit. Antarct. (‘Terra Nova’) Exped.,
1910. Nat. Hist. Rep., Zool., 5, 131-252.

Trebilcock, R. E., 1928. Notes on New Zealand Hydroida. Proc. Roy. Soc. Victoria (n.s.), 4%, 1-31.

Vanhoffen, E., 1910. Die Hydroiden der Deutschen Siidpolar-Expedition 1901-1903. Deutsche
Stidpolar-Exped. 1901-1903, 11, Zool., 269-340.

Vervoort, W., 1941. Biological Results of the Snellius Expedition. XI. The Hydroida of the
Snellius Expedition (Milleporidae and Stylasteridae excluded). Temminckia, 6, 186-240.

Vervoort, W., 1946. Fauna van Nederland. Aflevering XIV. Hydrozoa (C I). A. Hydro-
polypen. Leiden.

Vervoort, W., 1946a. Exotic Hydroids in the collections of the Rijksmuseum van Natuurlijke
Historie and the Zoological Museum at Amsterdam. ool. Meded., 26, 287-351.

Warren, E., 1908. On a collection of Hydroids, mostly from the Natal coast. Ann. Natal Mus.,

1, 269-355.

Bake
A

The ANNALS OF THE SOUTH AFRICAN MUSEUM are issued in parts at irregular
intervals as material becomes available.

Some volumes and parts are out of print, and others are only sold as parts of a set, or volume,
respectively. The prices of parts published prior to 1940 have been increased.

Out of print: Vols. I, II (Parts 1-3, 5, 7, 8), III (Part 1), V (Parts 1, 2, 5,8, 9), VI (Part 1
Index), VII (Parts 1, 2, 3, 5), VIII, LX (Parts 1, 2), X (Part 2), XI (Parts 2, 7, Index), XXI

(Part 1), XXIV (Part 2), XXXI (Parts I-3).

Vol.
II. 1go0-1902 Zoology and Geology (excl. Parts 1-3, 5, 7, 8)
III. 1rg03-1905 Zoology . (excl. Part. 1)
IV. 1903-1908 Palaeontology be
V. 1906-1910 Geology, Palaeontology, Zosloey, Anthropology (excl;
Parts 1, '2,'5;. 05:9).% ‘a’ a
VI. 1908-1910 Zoology SP heel ‘Parts Ts ‘Peden
VII. 1908-1913 Palaeontology Sie Ae . (Parts 4 and 6 only)
IX. 1911-1918 Botany Say aie ia Genk Parts 1 and 2)
X. «911-1914 Zoology , .. (excl. Part 2)
XI. srg11-1918 Zoology (ene: Parts 2, 7, Index)
XII. 1913-1924 Palaeontology and eens
XIII. 1913-1923 ee and Zoology
XIV. 1915-1924 Zoology ; a a
XV. 1914-1916 Zoology
XVI. 1917-1933 Botany
XVII. 1917-1920 Zoology
XVIII 1921 Zoology
XIX. 1924-1925 Zoology
XX. 1924-1926 Zoology is Be a
XXI. 1925-1927 Zoology .. (excl. Part. 1)
XXII. 1925-1928 Palacantelugy
XXIII. 1925-1926 Zoology .. wk ss =
XXIV. 1929-1938 Anthropology and Ethnology ..(excl. Part 2)
XXV./ 1927-1928 Zoology
XXVI. 1928 Zoology
XXVIII. 1929 Anthropology
XXVIII. 1929-1932 Palaeontology
XXIX. 1929-1931 Zoology
XXX. 1931-1935 Zoology.
INDEX of papers, authors, and elena published in Maks I XXX A
XXXI. 1934-1950 Palaeontology Bidets 4 a
XXXII. 1935-1940 Zoology ;
XXXITI. 1939 Zoology
XXXIV. 1938 Zoology
XXXV. 1956 Zoology .. re A he ‘
XXXVI. 1942-1948 Zoology rr ais we ck es
XXXVII. 1947-1952 hichaanlagy: ie:
XXXVIII. 1950 Zoology
XXXIX. 1952 Zoology
XL. 1952-1956 Botany
XLI. 1952-1955 Zoology
XLII. 1953-1956 Palaeontology :
XLII. 1955-1957 Zoology and Pelawmatalaey :
XLIV. 1957- Parts 1-5, Zoology and Palatontology. .

Copies may be obtained from—

The LIBRARIAN, Souru African Muszum, Cape Town,

CO mm m thy

MH HOP HHO HENAN NW OOW NN a men DH De DOHRWO AO OR DAN

5.
2

bal ae ca
ow om O08 SIO oo o

-—

DAIOARIAHMRCCOCOCOCCI CO MOOWOO00OuUW Ooww Ow MNwr

hoki’
we
or
aye
sm
ay
4
Pa
3

°_—_—

& ~~

Ta Ws Oo

if =

ISSUED ao, = PRICE 11s.

cas (ere ¥ UMITED, couRT ROAD, , WYNBERG, CAPE

BEITRAG ZUR KENNTNIS DER LAMIIDEN SUDAFRIKAS
(CERAMBYCIDAE, GOLEOPTERA)

VON

S. BREUNING,

Paris

Dank dem freundlichen Entgegenkommen von H. Andreae konnte ich
eine Anzahl Exemplare des Cap-Museums untersuchen, welche zu den nachste-
henden Bemerkungen Anlass geben. Die Typen der neuen Formen befinden
sich durchwegs im Cap-Museum.

1. Sophronica carbonaria Pasc. m. rufa nov.

Wie die Stammform, aber der Kérper durchwegs rot mit gelben abste-
henden Haaren.

Typ von Transvaal: Louis Trichardt. Weitere Stiicke von Natal: Malvern
(British Museum) und Moosplats (? Mooiplaats) (coll. Lepesme).

2. Nach Ansicht des Typs von Dirphya gigantea Nonfr. 1m Museum von
Berlin konnte ich feststellen, dass ich diese Art falsch gedeutet hatte. Gigantea
Nonfr. ist identisch mit der Form, welche ich unter dem Namen Duirphya
griseipennis Breun. m. nigrosternalis beschrieben habe (1956, Longicornia, III,
P- 590); nzgrosternalis Breun. ist daher als Synonym einzuziehen, und grisezpennis
Breun. sowie aériventris Breun. (l.c.) sind als Varietaten von gigantea Nonfr.
anzusehen.

Dagegen ist an genannter Stelle gigantea Nonfr. zu streichen und similis
Gah. als gute Art anzusehen mit der einzigen Varietat flavoabdominalis Breun.

3. Oberea mgrocincta Auriv. m. andreaei nov.

Wie die Stammform, aber das Metasternum ohne schwarze Makeln.
Typ ein von der Delagoa Bai, Rikatla.

4. Pseudoconizonia bicolor Pér., dessen Typ ich nunmehr vergleichen konnte,
ist als Synonym von basalis Gah. anzusehen. Fir die Form, welche ich irriger-
weise als m. bicolor Pér. identifizierte (1956, Longicornia, III, p. 612), schlage
ich den Namen v. rufoampliata nov. var. vor.

Typ ein 2 von Basutoland: Likhoele, Mrs. Dieterlein.

5. Die Art, welche ich mit Phytoecia (Blepisanis) maculicollis Per. identifiziert
hatte (1951, Ent. Arb. Mus. Frey, H, p. 119), ist eine von maculicollis abwei-
chende Art, fiir welche der Name anterufa Breun. einzutreten hat (welch
letztere die fiir die m. anterufa Breun. angegebene Pubeszenz aufweist).

227

SmMilHSONIAN ad
Ms SNSTITUTION JAN 2 7 195%

228 ANNALS OF THE SOUTH AFRICAN MUSEUM

6. Phytoecia (Blepisanis) maculicollis Per. —
Nitocris maculicolis Péringuey, 1888, Trans. 8. Afr. Philos. Soc., IV, p. 184.

Sehr lang. Fiihler massig dick, um ein Viertel langer (3) oder etwas langer
(2) als der Korper, das erste Glied wenig lang und dick, das dritte etwas langer
als das vierte, merklich langer als das erste, das vierte etwas langer als das
fiinfte. Untere Augenloben 2 mal so lang (¢) oder um die Halfte langer (9)
als die Wangen. Stirn etwas breiter (3) oder fast 2 mal so breit () als einer
dieser Loben. Kopf und Halsschild sehr dicht und fein punktiert. Halsschild
etwas langer als breit (¢) oder so lang als breit (9), seitlich schwach verrundet,
mit vier kleinen glatten runden Erhabenheiten auf der Scheibe, die beiden
premedianen der Mittellinie genahert, die beiden postmedianen seitlich
gelagert. Schildchen quer, apikal verrundet. Decken sehr lang, apikal schmal
sehr schwach abgestutzt, dicht und massig grob punktiert, die Punkte gereiht,
im apikalen Viertel feiner und unregelmassig gelagert. Das zweite Abdominal-
segment des ¢ mit einem sehr kleinen Mitteldorn an seinem Hinterrand.

Schwarz, fein hellgrau tomentiert. Kopf rot mit Ausnahme des vorderen
Mittelteiles der Stirn. Halsschild rot, alle Rander schmal schwarz. Schildchen
dicht weisslich tomentiert.

Lange 7-11 mm.; Breite 1+-2 mm.

Von Péringuey nach Stiicken von der Kap-Kolonie beschrieben.

m. trilobata nov.

Wie die Stammform, der mittlere Teil des Scheitels zuweilen schwarzlich;
der Halsschild schwarz mit Ausnahme einer sehr breiten nach vorn zu drei-
lappigen, postmedianen roten Makel, die premedianen glatten Erhabenheiten ©
daher ebenfalls schwarz; die Deckenscheibe zuweilen teilweise rotlich.

Typ ein § von der Kap-Provinz: East of Pakhuis Pass. Ein Allotyp idem;
ein Paratyp (2) von Klaarstroom—Prince Albert.

Maculicollis Pér. unterscheidet sich auf den ersten Blick von anterufa Breun.
durch viel weniger grob punktierte Fliigeldecken.

BESTIMMUNGSTABELLEN DER HETEROCHELIDES

(COLEOPTERA, LAMELLICORNIA, HOPLIINI) MIT AUSNAHME
VON HETEROCHELUS BURM. UND ISCHNOCHELUS BURM.

Mit Neubeschreibungen und Bemerkungen

VON

Hans SCHEIN
Miinchen

INHALT

Einleitung . : - ay py 229
Bemerkungen und Neubeschreibungen p. 230
Bestimmungstabellen der Gattungen p. 256
Bestimmungstabellen der Arten aM cs Did

EINLEITUNG

Eine der interessantesten Kafergruppen Siidafrikas bilden die Hoplini, die
zuletzt im Jahre 1902 Péringuey zusammenhangend systematisch bearbeitet
hat. Seitdem sind viele neue Arten gefunden worden, daB es an der Zeit ist,
Péringueys Arbeit auf den neuesten Stand zu bringen und neue Bestimmungs-
tabellen aufzustellen. Meine erste Arbeit in dieser Richtung befaBt sich mit
den Pachycnemini; sie wird in der Zeitschrift ,, Entomologische Arbeiten aus dem
Museum G. Frey Tutzing bei Miinchen”’ herauskommen. In der vorliegenden
Arbeit behandle ich die Halfte der Heterochelides; die andere Halfte —die sehr
viele Arten umfassenden Untergattungen Heterochelus Burm. und Ischnochelus
Burm. —soll einer gesonderten Behandlung vorbehalten bleiben.

In dem Bestreben, meiner Arbeit eine sichere Grundlage zu geben, habe
ich versucht, die Typen und ein méglichst groBes Kafermaterial zusammen zu
tragen, was mir weitgehend gelungen ist. Es ist mir ein Herzensbediirfnis, auch
hier meinen besten Dank den Mannern auszudriicken, die mir dabei geholfen
haben. Es sind Dr. A. J. Hesse und Dr. H. Andreae des Siidafrika Museums
in Kapstadt, die mir die Typen Péringueys geliehen haben; Professor Dr. J. O.
Hising des zoologischen Institutes der Martin-Luther-Universitat in Halle, der
es mir erméglicht hat, die Typen Burmeisters zu sehen; Professor Dr. René
Malaise des Museums in Stockholm, der Typen Bohemans gesandt hat;
Professor Dr. K. Delkeskamp der Humboldt-Universitat in Berlin, dem ich
Typen v. Harolds, Nonfrieds und Mosers verdanke; Fabrikant Georg Frey in
Tutzing (Museum G. Fr.) und Direktor Dr. W. Forster und Hauptkonservator

229

230 ANNALS OF THE SOUTH AFRICAN MUSEUM

H. Freude der zoologischen Sammlung des bayerischen Staates in Miinchen,
die mir das ganze Material der von ihnen betreuten Sammlungen zur Verfi-
gung gestellt haben. Besonderen Dank schulde ich dem Kurator C. Koch des
Transvaal Museums in Pretoria, der mir das ganze sehr reiche Material seines
Museums zur Bearbeitung geschickt hat.

Damit die Arbeit nicht zu umfangreich wird, habe ich darauf verzichtet,
Ausfiihrungen alter Autoren, besonders Beschreibungen, zu wiederholen. Die
wenigen in Betracht kommenden Werke stehen Interessenten in jeder gréBeren
Museumsbibliothek zur Verfiigung: Der ,,Descriptive Catalogue of the Coleop-
tera of South Africa” von L. Péringuey, 1902, das ,, Handbuch der Entomologie”
4, Band von Hermann Burmeister, 1844, der ,,Coleopterorum Catalogus Pars
50 Hopliini’” von Junk-K.W. Dalla Torre 1912/1913 und in geringerem Umfang
die ,,Insecta Caffrariae”’ von Boheman 1857. Ich setze sie als bekannt voraus.

Den Bestimmungstabellen schicke ich in der bei ihnen gegebenen Reihen- ~
folge Beschreibungen neuer Gattungen und Arten und kurze Bemerkungen
voraus, in welchen ich Berichtigungen von Autoren, erganzende Angaben
besonders iiber 99 und Angaben tiber den Verbleib von Typen bringe. Die
oben im zweiten Absatz genannten Museen und Universitatsinstitute bezeichne
ich dabei der Abkiirzung halber mit ihrem Ort: Kapstadt bedeutet also
Siidafrika Museum in Kapstadt u.s.w.

Ein nomemklatorisch dunkler Punkt bleiben die Namen Ecklons. Dieser,
Apotheker in Hamburg und Autor eines mit Zeyher herausgegebenen Werkes
uber siidafrikanische Pflanzen, hat mit Kafern gehandelt und eine Preisliste
versandt, die gr6Btenteils nomina nuda, aber auch einige kurze Beschreibungen
enthalten hat. Burmeister zitiert die Liste an verschiedenen Stellen mit
,,Ecklon Cat’’. Diese Beschreibungen Ecklons haben die Prioritat, wenn sie zur
Erkennung der Art geniigen. Die Liste ist aber nicht mehr aufzutreiben; meine
Anfragen bei den groBen entomologischen Bibliotheken hatten keinen Erfolg.
Wiirde sie gefunden, kénnte der Mangel an Typen ihrer Auswertung entgegen-
stehen.

Alle GréBenangaben im folgenden Text sind ohne Beine zu verstehen.

Ortsangaben ohne Bezeichnung des Teilstaates der Siidafrikanischen Union
beziehen sich auf die Kap-Provinz.

BEMERKUNGEN
1. DicHELus Serv.

Serville hat 1825 nur die Zahl, nicht auch die Lange der Krallen der
Hinterbeine als Kriterium beniitzt; alle Arten mit einer Kralle hat er zu
Monochelus gestellt. Burmeister hat 1844 dies als unnatiirlich bezeichnet und
Dichelus und Monochelus in eine einzige Gattung Heterochelus zusammengezogen.
Lacordaire ist 1856 dem entgegengetreten. Péringuey hat 1902 als erster nicht
nur die Zahl, sondern auch die Lange der Krallen zur Einteilung beniitzt und
die Gattung Dichelus (Genotyp dentipes F.) auf die Arten mit 2 gleichlangen

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 231

Krallen beschrankt, wahrend er die Arten mit 2 ungleichen Krallen und mit
nur einer Kralle zu Heterochelus gestellt hat. Diese Lésung hat leider den
Schénheitsfehler, da& auch bei Dichelus Arten mit 2 ungleichen Krallen vor-
kommen. Péringuey selbst muBte denticeps Wied. als Ausnahme nennen,
nitidissimus Burm. und kochi n. sp. kommen dazu. Ich halte es deshalb fir
richtig, mit Burmeister nur eine einzige Gattung anzunehmen. Diese muB
allerdings Dichelus Serv. sensu lato heiBen. Sie zerfallt in die drei Untergattungen
Dichelus Serv. sensu stricto (Genotyp dentipes F.), Heterochelus Burm. und
Ischnochelus Burm. Hier wird nur die Untergattung Dichelus behandelt. Genotyp
dentipes F.

Infolge des sehr einheitlichen Baues der 99, welche die Differenzierung der
dd nicht mitmachen, lat sich bei Dichelus eine brauchbare Bestimmungs-
tabelle nur fiir die §¢ aufstellen. Wenn bei manchen Arten die 99 noch nicht
mit Sicherheit bekannt sind, ist diese Einheitlichkeit der 29 einer der Griinde.

Die Arten lassen sich nach den Hinterschienen der ¢¢ zwanglos in drei
Gruppen einteilen:

denticeps-Gruppe: Die Hinterschienen nahern sich dem weiblichen Typ, indem
sie ziemlich gleichma8ig an Breite zunehmen und am Ende sehr schrag
abgeschnitten sind. Hierher gehéren simplicipes Burm., denticeps Wied.,
pallidipennis Blanch. und flavimanus Burm.

dentipes-Gruppe: die Schienen tragen unten zwischen dem Zahnchen beim Knie
und dem Apikalmukro einen zahnartigen Lappen. Hierher zahlen Jaticollis
Burm., ntidissimus Burm., dentipes F., expansus Pér., acanthopus Burm.,
villosus Burm., luteopygus n. ssp., minor n. ssp., lucidus Pér., vittatus Burm.,
peringueyi n. sp., duplosquamosus n. sp., pseudovittatus n. sp. und holosquamosus
n. sp.

holosericeus-Gruppe: Hinterschienen schmaler, starker gebogen, fast parallel-
seitig ohne den zahnartigen Lappen. Arten: holosericeus Burm., albolineatus
n. sp., platynotus Burm., luctuosus Pér., kocht n. sp. und pseudoluctuosus n. sp.
Wohin zuluanus Pér. (nur 2) gehort, ist unsicher.

Zu den Dichelus-Arten:

Der Coleopterorum Catalogus zahlt 24 Arten auf, weitere Arten sind nicht
beschrieben. Der Katalog ist zu berichtigen:

latipes und subpilosus Nonfried sind zu streichen; ich habe die Typen gesehen.
Ersterer ist eine Chasme decora Wied., letzterer ein abgeriebener villosus Burm.

expositus Har. ist ein Platychelus mit ungleichen doppelten Krallen an allen
Beinen. Da die langere Kralle bis zum Grunde gespalten ist, war die
Verwechslung moglich.

flavipennis Blanch. geh6rt zu Omocrates.

mgra Wied. hat nach der Beschreibung an allen Fi®en doppelte ungleiche
Krallen und ist kein Dichelus. Ich habe die Type noch nicht gesehen.

232 ANNALS OF THE SOUTH AFRICAN MUSEUM

quadratus Wiedemann ist ein Heterochelus.

soricinus Blanch. geh6rt ebenfalls zu Heterochelus.

Auf femoratus Thunberg komme ich weiter unten bei holosericeus Burm.
zuriick. Thunberg, der Nachfolger Linnés in Uppsala, hat viele Kafer in der
damals tiblichen fliichtigen Weise beschrieben, von denen viele ohne die Typen
kaum richtig zu deuten sind. Es ist schade, daB sich niemand seiner Sammlung
annimmt, in der sicher noch eine ganze Anzahl von Typen ermittelt und fest-
gelegt werden kénnten.

Zu den nach Bereinigung des Katalogs verbleibenden Arten ist zu
bemerken:

(1) D. simplicipes Burmeister 1844

Kommt im Aussehen den Arten der holosericeus-Gruppe nahe, die Hinter-
beine des ¢ sind aber sehr verschieden. 9 mit Sicherheit nicht bekannt. Typus
in Halle. Lange 5 mm. Malmesbury, Paarl, Caledon.

(2) D. pallidipennis Blanchard 1850

Da das Pariser Museum grundsatzlich Typen nicht mehr ausleiht, habe
ich nur 2 mit diesem Namen versehene Stiicke des Siidafrika-Museums gesehen,
die 9° sind und den holosericeus-92 nahe zu stehen scheinen. Die Einreihung
dieser Art in die Tabelle konnte ich nur nach der Beschreibung vornehmen.
Beim ¢ ist auffallend, daB es einen Sporn an den Hinterschienen hat. Lange
5 mm. Malmesbury. Typen im Musé National d’Histoire Naturelle in Paris.

(3) D. denticeps Wiedemann 1821

Das ¢ erinnert durch seine Hinterschienen etwas an Diaplochelus squamulatus
Burmeister, der einen abgerundeten Kopfschild hat. Das 2 ist oben etwas
dunkler als das g und diesem dhnlich; sein Pygidium ist wie beim ¢ ohne
Schuppen und glatt. Lange 5-5,5 mm. Cold Bokkeveld, Ceres Distr., Cape
Town Rondebosch. Typus im Zoologischen Museum der Universitat Kopen-
hagen.

(4) D. flavimanus Burmeister 1855

Das ¢ hat groBe Ahnlichkeit mit holosericeus Burmeister in der Form, dem
Seidenglanz und der Farbe, ist aber davon durch die zitronengelben Fihler,
Taster und Vorderbeine und die anders gebauten Hinterschienen leicht zu
unterscheiden. Das @ ist noch unbekannt. Ich habe nur die Type gesehen, die
Art scheint also selten zu sein, ihr genauer Fundort ist nicht bekannt. Lange
5 mm. ,,Kaffernland’”. Typus in Halle.

(5) D. laticollis Burmeister 1844

Diese Art fallt durch ihre langere Form, den nach vorn weniger ver-
schmalerten Halsschild, dessen unregelmaBige grobe Punktierung und die
verhaltnismaBig kurzen Beine aus dem Rahmen der Gattung; Fihler, Taster

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 233

und Vorderbeine sind hellrot. Das 2 ist dem ¢ sehr ahnlich, aber durch das
gewolbte Pygidium und den fehlenden Mucro an den Hinterschienen, die einen
Sporn tragen, leicht zu erkennen. Lange 5-7 mm. Cape Town, Somerset West,
Stellenbosch, Caledon, Heidelberg.

Die Type ist verloren gegangen. In Halle steckt zwar ein so etikettiertes
2 in der Sammlung Burmeister, es ist aber mit Sicherheit nicht laticollis; ich
halte es fiir villosus Burm. Ich nehme an, daB der Autor die Type nach der
Beschreibung wieder an den Einsender Van Winthem in Hamburg zuriick-
gegeben hat. Dafiir spricht der Sprachgebrauch bei Burmeister, der bei dieser
Art die Wendung gebraucht ,,von Herrn v. Winthem mitgeteilt”, wahrend er
bei den noch in der Sammlung vorhandenen Arten schreibt ,,von...
erstanden”’. Auch seine platynotus fehlt, die inm ebenfalls ,,mitgeteilt’? worden
ist. Die Sammlung Van Winthem ist in das Hamburger Museum iibergegangen
und, wie mir der jetzige Leiter Prof. Dr. Weidner mitgeteilt hat, im letzten
Krieg mit diesem Museum vollig zerst6ért worden. Ich habe deshalb ein Paar
laticollis des Siidafrika- Museums als Neo-Holo-und Allotypus bezeichnet.

(6) D. nitidissimus Burmeister 1844

Im Gegensatz zur Beschreibung haben die Hinterschienen des ¢ unten
iiber der Mitte einen Zahn. Trotz den ungleichen Krallen passt die Art dem
Habitus nach besser zu Dichelus. Das Q2 ist unbekannt. Lange 5,5 mm.
,sudafrika”’. Typus in Halle.

(7) D. dentipes Fabricius 1781

Das 9 dieser altbekannten groBen Art ist meist etwas heller gefarbt als das
6; ich habe aber auch, wenn auch viel seltener, Stiicke mit sehr dunkelbraunen
Fliigeldecken gesehen; sein Pygidium ist immer einfarbig hell beschuppt und
ungefleckt. Man findet manchma! Stiicke des acanthopus Burmeister und seines
villosus, die ihre Schuppenbinden fast ganz verloren haben, als dentipes bestimmt.
Sie kénnen an den Hinterschienen des ¢ unterschieden werden: diese sind
unten im basalen Teil bei dentipes breit und gerillt, bei villosus breit und nicht
gerillt, bei acanthopus zusammengedriickt, soda8 der scharfe Kiel der Kante
auffallt. Bei vzllosus sind die Schuppenbinden in deutliche, breite Rillen
gelagert, acanthopus hat eine solche schmdlere und weniger deutliche Rille nur
langs der Naht. Lange 6-7, 5 mm. Cold Bokkeveld Ceres Distr., Malmesbury,
Paardekop, Melkbosch Strand, Gape Town Orange Kloof, Camps Bay,
Kirstenbosch, Constantia Nek, Clovelly Cape Peninsula, Fish Hoek Cape
Peninsula, Cape Flats, Banhoek Valley Stellenbosch, Franschhoek, Worcester,
Cloete Pass, Robinson Pass. Typus im Zoologischen Museum der Universitat
Kopenhagen.

(8) D. expansus Péringuey 1902
Das Pygidium des 9° tragt helle Schuppen mit 2 runden dunklen Flecken
in der basalen Halfte, sonst gleicht das 2 dem von dentipes. Einige ¢¢ aus

234. ANNALS OF THE SOUTH AFRICAN MUSEUM

Sneeugat Valley in Tulbagh haben matte, tief dunkelbraune Fliigeldecken,
werden aber durch die mitgefangenen 99 als expansus ausgewiesen. Lange
6-7 mm. Ceres, Sneeugat Valley, Tulbagh, Koeberg, Noordhoek, Somerset
West, Stellenbosch, Caledon, Worcester. Typus in Kapstadt.

(9) D. acanthopus Burmeister 1844

Die 99 sind von denen der vorigen Art nicht zu unterscheiden. In der
Sammlung des Siidafrika-Museums sind die ,,acanthoscelis Ecklon’’ bezeichneten
Stiicke hier eingereiht; gut erhaltene Exemplare zeigen aber mehrere Schup-
penbinden auf den Fliigeldecken und damit die nahere Verwandtschaft mit
villosus Burmeister, welcher Autor sie auch richtig dorthin gestellt hat, siehe
folgende Ziffer 10 b. Lange 4,5 — 6,5 mm. Ceres, Tulbagh, Cape Town,
Somerset West, Banhoek Valley Stellenbosch, Tradouw Pass Swellendam Distr.,
George Distr. Typen in Halle.

(10) D. villosus Burmeister 1844

Die Koérperform ist etwas schmaler als bei den vorausgegangenen Arten,
die dunkle Grundbehaarung ist deutlicher, die hellen Schuppen der Langs-
binden, deren 4uBere oft nur rudimentar oder ganz verschwunden sind, sind
haarahnlicher. !

Die 99 sind meist dunkler als bei dentipes und folgenden, ihr Pygidium ist,
wie Burmeister im Gegensatz zu Péringuey richtig angibt, iberwiegend dunkel
mit schmaler weiBgelber Schuppenbinde langs der Mitte. Lange 7—7,5 mm.
Darling, Malmesbury, Cape Town, Camps Bay, Tafelberg Blinkwater,
Rondebosch, Somerset West, Banhoek Valley Stellenbosch, Koegelberg in den
Hottentots Holland Bergen. Typus in Halle.

(10a) JD. villosus subspecies luteopygus nova

Ein wenig breiter gebaut, in Skulptur und Beinbildung aber ganz wie
villosus Burmeister, in der Farbe abweichend. Die 3 Schuppenbinden auf jeder
Decke, der Halsschildhinterrand, die oberen Rander der Seitenteile der Brust
und die Enden der Bauchringe sind zwar weiblich wie bei villosus, die Grund-
farbe ist aber itiberall tiefschwarz und das Schildchen und der Hinterrand des
sonst schwarzen Propygidiums sowie das ganze Pygidium sind auffallend
orangegelb beschuppt. Das 9 ist ebenfalls tiefschwarz, etwas mehr glanzend,
das Pygidium ist bei ihm heller gelb, die dunklen Seitenflecken schimmern nur
leicht durch die helle Beschuppung durch. Gr6éBe wie villosus. Lange 7—7,5 mm.
Jakalswater Bushmanland, Gifberg Vanrhynsdorp Distr., Somerset West,
Stellenbosch, Worcester, George Distr. Typus in Kapstadt.

(10b) JD. villosus subspecies minor nova

Es handelt sich um den im Handbuch Burmeisters Band IV 1 Seite 110
am SchluB der Diagnose des villosus als ,,var. min. Monoch. acanthoscelis Eckl.
Cat. No. 410” erwahnten Kafer. Sollte Ecklon eine Beschreibung ver6ffentlicht

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 235

haben, hatte sein Name die Prioritat vor minor. Will manin dem Wort minor eine
giltige Beschreibung erkennen, ist als Autor Burmeister zu zitieren. Um keine
Verwirrung zu schaffen, habe ich den Burmeisterschen Namen aufrecht erhalten.

Subspecies mznor ist eine villosus-Rasse mit starkerem griinlichem Metall-
schimmer des Halsschildes, hellbraunen Fliigeldecken und konstant geringerer
GréBe, sowie anderer Fleckung des Pygidiums des 9. Ferm, Skulptur und
Beinbildung wie vzllosus. Kopf und Halsschild sind aufrecht grau behaart. Der
Metallschimmer erinnert an Heterochelus viridicollis Blanchard. Die langs der
Mitte des Halsschildes durchlaufende Rille ist vorn sehr seicht, hinten tief und
nicht beschuppt wie auch zwei quere schmale Eindriicke an seinen Seiten neben
dem Hinterrand. Die hellbraunen Fliigeldecken sind sehr fein anliegend dunkel
behaart und lassen je eine ziemlich breite Rille neben der Naht und neben der
Mitte der Scheibe erkennen, die wie auch das Schildchen langliche weife, den
Grund nicht ganz verdeckende Schuppen tragen; eine weitere solche Schuppen-
binde lauft am Beginn des Seitenabfalles. Die Nahtbinde ist immer vorhanden,
die anderen sind oft rudimentar oder ganz verschwunden. Weil beschuppt
sind auch der Pygidialteil, die Seitenstiicke der Brust und die Enden der
Bauchringe. Die Beine sind braun oder schwArzlich, alle Beinpaare gleichfarbig.

Das 2 gleicht dem ¢ in der Farbung; die Behaarung des Halsschildes ist
heller; das Pygidium ist weiBgrau beschuppt mit 2 groBen keilformigen Kahl-
stellen beiderseits der Mitte und einer runden Kahlstelle am Apex; die
beschuppte Flache ist groBer als bei villosus. Lange 5-5,5 mm. Ceres, Cape
Town, Camps Bay, Stellenbosch, Zwartberg Pass Prince Albert Distr. Typus
in Kapstadt.

(11) D. lucidus Péringuey 1902

Die Fliigeldecken sind heller braun und glanzen starker als beim vorigen,
der zahnartige Lappen oberhalb der Mitte des unteren Randes der Hinter-
schienen des ¢ ist stumpfer und fallt gegen das Schienenende allmahlich ab.
Die Vorder- und Mittelbeine und die Tarsen der Hinterbeine sind braun, die
Hinterschenkel und -schienen pechschwarz. Die Borstchen an den Hinter-
schienen und besonders an den -tarsen sind braungelb im Gegensatz zu villosus
und minor, wo sie pechschwarz sind. Auf dem Schildchen im Gegensatz zur
Beschreibung einige helle Schuppen.

Ich habe nur die Type und ein weiteres ¢ gesehen. 2 unbekannt. Lange
4-4,5 mm. Caledon. Typus in Kapstadt.

(12) D. vittatus Burmeister 1844 (nec Péringuey 1902)

Ich kenne diese Art nicht in natura. Sie fehlt in der Sammlung Burmeister
in Halle. Die Type ist mit gro68ter Wahrscheinlichkeit verloren wie bei laticollis
oben Ziffer 5, siehe dort. 9 unbekannt. Lange 8,7 mm. ,,Siidafrika.”’

(13) D. péringueyi nova species
Das Stiick in Kapstadt ist als vettatus Burm. bezettelt, passt aber nicht ganz
zur Beschreibung dieser Art. Wahrend dort die Behaarung auf Kopf und

236 ANNALS OF THE SOUTH AFRICAN MUSEUM

Halsschild als tief schwarz und die Krallen als ,,fast’’ gleich bezeichnet sind,
sind diese Haare hier hell und die Krallen gleich. Auch sind die Beine rotbraun,
nicht schwarz. Die 4 Zahne des Kopfschildvorderrandes sind gleich, bei
vittatus sind die mittleren kleiner. Im tibrigen beziehe ich mich auf die Beschrei-
bung des vittatus Péringueys in seinem ,,Descriptive Catalogue of the Coleoptera
of South Africa 1902 Seite 702”. Lange 7—7,5 mm. Ein Fundortzettel ist an
der Type nicht vorhanden. 2 Paratypen in meiner Sammlung sind leider auch
nur ,,Caffraria” bezettelt. 9 unbekannt. Typus in Kapstadt.

(14) D. duplosquamosus nova species

Schwarzer Dichelus aus der Verwandtschaft des vittatus Burm. mit je 3
Langsbinden aus sehr dicht stehenden, kleinen runden weifgelben Schuppen
und 2 Rippen dazwischen, die mit weiBgelben aufrechten lang-kegelférmigen,
weniger dicht stehenden Schuppen besetzt sind.

Kopf und Halsschild schwarz, Kopfschild mit 4 Zahnen, von denen die
mittleren kleiner sind, runzlig punktiert. Halsschild nach vorn gerundet
verschmalert, fein punktiert, schwach gewélbt, in der hinteren Halfte mit
Langsfurche, mit kleinen, weiBgelben, tiber die ganze Flache verstreuten, am
Hinterrand dichteren Schuppen und abstehenden hellen Harchen. Schildchen
dicht beschuppt. Fliigeldecken mit rotbraunem Grund, der an den nackten
Schulterbeulen und in schwacherem MaBe an den Rippen sichtbar wird. Die
doppelte Form der Schuppen ist auffallend, doch treten die Binden nicht sehr
hervor, weil die Schuppen der Rippen wie die der Binden gefarbt sind. Der
ganze Pygidialteil ist dicht weiBgelb beschuppt. Die Unterseite ist dicht zottig
ziemlich lang gelbweif behaart. Die Beine sind wie bei villosus gebaut; die
vorderen sind pechschwarz, die anderen rotbraun, gelbweiB behaart und
beborstet. Alle Krallen sind doppelt, gleichlang und gespalten. 9 unbekannt. ~
Lange 5-6 mm. Cold Bokkeveld Ceres. ‘Typus in Kapstadt.

(15) D. pseudovittatus nova species

Schwarzer Dichelus aus der Verwandtschaft des vittatus Burm. mit tief
schwarz behaartem, am Hinterrand orangegelb beschupptem Halsschild, mit
je drei dichten orangegelben Schuppenbinden auf den dunkelbraunen Fligel-
decken und schmalen Zwischenr4éumen mit Schuppen von der Farbe des
Grundes zwischen den Binden.

Kopf und Halsschild schwarz und kurz abstehend tief schwarz behaart,
dicht runzlig punktiert, die 4 Zahne des Kopfschildes klein und gleich. Der
Hinterrand des Halsschildes bis kurz vor die Ecken und der hinterste Teil der
Langsfurche dicht orangegelb beschuppt wie auch das Schildchen. Fligel-
decken mit braunem bis schwarzem Grund, die nackte Schulterbeule heller
rotbraun, mit je 3 orangegelben Binden aus dichten runden kleinen Schuppen,
von denen die Naht- und Randbinden hinten zusammenhangen und die
schmalere mittlere Binde hinten verkiirzt ist. Die Zwischenraume tragen runde
kleine glanzende Schuppen von der Farbe des Untergrundes, Der ganze

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 237

Pygidialteil ist dicht beschuppt, die Schuppen sind etwas heller als die der
Binden. Auch die Enden der Bauchringe sind so beschuppt, die ibrige Unterseite
schwarz behaart. Die Beine sind schwarz bis pechschwarz mit rotbraunen
Tarsen, die Hinterbeine sind manchmal rotbraun. Das Q ist schlanker, der
Halsschild ist nicht beschuppt und braunlich behaart, die gelben Binden der
Fligeldecken sind viel dinner und schmialer, das gelblich beschuppte Pygidium
hat 2 seitliche keilfo6rmige schwarze Flecken, deren Spitzen zum Apex zeigen.
Die schlanken Beine sind schwarzbraun mit gelbbraunen Hinterschienen und
Tarsen. Lange 5-5,5 mm. Banhoek Valley Stellenbosch, Assegaibosch-
La Motte bei Humansdorp. Typus in Kapstadt.

(16) D. holosquamosus nova species

Schwarzer Dichelus aus der Verwandtschaft des vittatus Burm. mit Hals-
schild wie duplosquamosus n. sp. und gleichmaBig gelblich fein beschuppten
Fliigeldecken.
Kopf, Halsschild und Schildchen in Form, Punktierung, Beschuppung und
Behaarung wie bei duplosquamosus n. sp. Fliigeldecken mit gelbbraunem Grund
und 2 schwachen Rippen, die sehr kleinen runden weiBgelben Schuppen stehen
dicht und sind gleichmaBig ititber die ganze Flache der Decken einschlieBlich
der Rippen verteilt. Auf den Rippen stehen nur feine gereihte gelbliche
Boérstchen zwischen den Schuppen. Der Pygidialteil ist dicht weiBgelb
beschuppt. Das Abdomen und die Brust sind ziemlich dicht und lang abstehend
weiB behaart, der schwarze Grund bleibt aber sichtbar. Die Beine sind pech-
braun, die vorderen etwas dunkler, die Tarsen rotbraun. Schenkel und Schienen
sind dicht und fein abstehend weiflich behaart, die Boérstchen der Tarsen
rotbraun.
Dem duplosquamosus ahnlich, aber Fliigeldecken heller und anders beschuppt
und Beine heller. Lange 6,5 mm. Malmesbury Paardekop. 92 unbekannt.
Typus in Kapstadt.

(17) D. holosericeus Burmeister 1844

Mein Versuch, mir die Type des femoratus Thunberg 1818 zu verschaffen,
von der Burmeister vermutete, daB sie synonym sein kénnte, ist gescheitert; sie
wurde in der Sammlung Thunberg in Uppsala nicht gefunden. Die Beschrei-
bung allein laBt die Frage der Synonymie nicht eindeutig lésen.

Den Seidenglanz, auf den der Name anspielt, hat die Art mit mehreren
anderen gemeinsam. Charakteristisch ist das Zahnchen unten am Hinter-
schenkel neben der Trochanterspitze.

Die Fliigeldecken der ¢¢ sind braun in verschiedenen Ténen oder schwarz.

Die 9° unterscheiden sich durch hellere gelbbraune Fliigeldecken und
rotbraunes, dicht gelb beschupptes Pygidium und Abdomen ohne Flecken auf
dem Pygidium. Die Beine sind rotbraun.

Lange 4,5-5,5 mm.

238 ANNALS OF THE SOUTH AFRICAN MUSEUM

Stiicke mit braunen Fliigeldecken: Cape Town, Witsands (White Sands),
Somerset West, Banhoek Valley Stellenbosch, Houhoek, Oudebosch, Rivier
Zonderend, Ashton, George Distr., Knysna, Langekloof.

Stiicke mit schwarzen Fligeldecken: Gifberg Vanrhynsdorp Distr.,
Sneeugat Valley Tulbagh, Somerset West, Stellenbosch, Algoa Bay.

Typus in Halle.

(18) D. albolineatus nova species

Diese Art ist schwarzen holosericeus sehr ahnlich, hat insbesondere den
kleinen Schenkeldorn neben der Trochanterspitze und die gleiche Form und
Skulptur. Sie unterscheidet sich aber auffallig dadurch, daB die Fliigeldecken
je eine weifBe Schuppenbinde an der Naht und in der Mitte der Scheibe tragen,
von denen die Nahtbinde vorn, die Scheibenbinde hinten verkiirzt ist. Die
Binden stehen in Rinnen. AuBerdem sind das Schildchen, der Rand des
Propygidiums, die Enden der Bauchringe und der obere Rand der Hinterhiiften
weiB beschuppt. Die Brust ist lang abstehend weil behaart. Die Beine sind
schwarz, der Fihlerstiel pechbraun. Kopf und Halsschild sind lang und diinn
abstehend dunkel behaart. Lange 5 mm. Mamre. Nur 2 @¢ liegen vor.
Typus in Pretoria.

(19) D. platynotus Burmeister 1844.

Lange 6 mm. ,,Siidafrika’’, Cold Bokkeveld?

Die Type ist verloren gegangen wie die von vittatus Burm. und laticollis
Burm., siehe oben bei Ziffer 5. Sie war auch von Van Winthem ,,mitgeteilt’’.

Ich sehe davon ab, das einzige Stiick des Siidafrika-Museums, das hierher
gehoren kénnte und in der Tabelle erwahnt wird, als Neo-Typus zu bezeichnen,
weil es in einigen Punkten von der Beschreibung abweicht. 9 unbekannt.

(20) D. luctuosus Péringuey 1902

Warum der Autor diese Art mit denticeps Wied. vergleicht, zu dem wenig
Ahnlichkeit besteht, nicht aber mit holosericeus Burm., verstehe ich nicht. Sie
unterscheidet sich von diesem nur durch Grofe, den weifen Schuppenstreifen
entlang der Naht der Fliigeldecken und das fehlende Zahnchen neben der
Trochanterspitze. Das ist schwarz mit braunlichem Schimmer, dicht weiBlich
beschupptem Propygidium und weniger dicht fein gelblich beschupptem, nicht
geflecktem Pygidium. Die Enden der Bauchringe tragen weiBliche Schuppen-
haare. Die Beine sind pechschwarz. Lange 6 mm. Cape Town. Typus in
Kapstadt.

(21) D. kocht nova species

Diese Art hat zwar ungleiche Krallen, aber ganz den Habitus der holosericeus-
Gruppe.

Schwarzer Dichelus dieser Gruppe, der durch je 2 weiBe Schuppenbinden
auf jeder Fliigeldecke, weiBbeschuppten Pygidialteil und ungleiche Krallen
gekennzeichnet ist.

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 239

Glanzend, starker als der Seidenglanz des holosericeus Burm. Kopfschild
mit 4 gleichen Zahnchen. Kopf fein gerunzelt, kurz aufrecht dunkel behaart.
Fihler und Taster pechbraun. Halsschild maBig gewélbt mit ziemlich tiefer
Langsfurche uber die hinteren zwei Drittel. Ohne Schuppen, unregelmabig
punktiert, aufstehend dunkel behaart, Seiten mit dunklem Borstensaum.
Pedunkulus und Schildchen dicht weiBlich beschuppt. Fligeldecken unregel-
maBig punktiert, mit einer flachen Rille langs der Naht und einer hinten
verkiirzten Rille langs der Mitte der Scheibe; in diesen Rillen weifBe langliche
Schuppen. Die weibe Nahtbinde lauft vom Schildchen bis zum Apex, die auf
der Scheibe hort hinter der Mitte auf. Der ganze Pygidialteil dicht weib
beschuppt, ebenso die Seitenstiicke der Brust und die Enden der Bauchringe.
Brust sonst abstehend wei behaart. Vorder- und Mittelbeine dunkelrotbraun,
Hinterbeine braunschwarz. Trochanterdorn lang, frei. Ecken der Schenkel
beiderseits des Kniees scharf. Schienen wie bei den anderen Arten der Gruppe.
Krallen braun, alle gespalten und ungleich. Lange 5-6 mm. Vanrhynsdorp.
Q unbekannt. Typus in Transvaa] Museum in Pretoria.

Dem Betreuer der Kafer in diesem Museum C. Koch gewidmet.

(22) D. pseudoluctuosus nova species

Schwarzer Dichelus der holosericeus-Gruppe ohne Schuppen auf Fliigeldecken
und Pygidium und ohne Schenkelzahnchen neben dem Trochanterdorn.

6S ganz schwarz, nur Krallen rotbraun, maBig glanzend. Kopfschild mit
4 gleichgroBen Zahnchen, von denen die mittleren einander gendhert sind.
Kopf kornig punktiert, kurz aufrecht schwarz behaart. Fiihler und Taster mit
braunlichem Schimmer, Facher schwarz. Halsschild maBig gewélbt, unregel-
maBig kornig punktiert, auf der Scheibe fast glatt, mit vorn fast erloschener,
hinten deutlicher Langsfurche in der Mitte; Seiten bis zur Mitte fast parallel,
dann starker verschmdlert. Ohne Schuppen. Schildchen wei beschuppt.
Fligeldecken mit unregelmaBig ziemlich weitlaufig stehenden Punkten, die
stellenweise gereiht sind; ohne Schuppen. Propygidium, obere Seitenstiicke
der Brust und Enden der Bauchringe dicht weiB beschuppt, Pygidium ohne
Schuppen, sehr glatt und glanzend. Unterseite kurz schwarz behaart. Beine
schwarz mit rotbraunen, bei einigen Stiicken dunkleren Krallen, Schenkel
manchmal mit braunlichem Schimmer, stark nach unten verbreitert, sodaB ein
Dreieck entsteht, dessen Basis die obere Kante des Schenkels bildet und dessen
Scheitel an dem Punkte liegt, an dem der Trochanterdorn frei wird. Daneben
kein Zahnchen. Lange 6 mm.

Das 2 kommt mit schwarzen und mit kaffeebraunen Fliigeldecken vor und
ist starker glanzend. Die Oberseite ist starker punktiert. WeiB beschuppt sind
das Schildchen und die Seiten der Bauchringe, gelblich beschuppt der Pygidial-
teil ohne Flecken auf dem Pygidium, die sonstige Unterseite ist ziemlich lang
weiBlich abstehend behaart. Beine pechschwarz, bei den braungefliigelten
Stiicken sind die Hinterbeine und alle Tarsen ebenfalls braun. Lange 5,5-6 mm.
Willowmore, Uniondale Distr. Typus im Transvaal Museum in Pretoria.

240 ANNALS OF THE SOUTH AFRICAN MUSEUM

(23) D. zuluanus Péringuey 1902

Diese Natal-Art ist nur in einem 9 bekannt. Lange 6,75 mm. Eshowe
Natal. Typus in Kapstadt.

2. OmocrATEs Burmeister 1844

Als Genotyp bezeichne ich axillaris Burm. 1844 und glaube dabei im Sinne
dieses Autors zu handeln, der nach der Schulterbildung dieser Art den Gattungs-
namen gewahlt hat. Wahrend Burmeister die Gattung nur auf Arten mit
sexuell verschiedenen Hinterschienen beschrankt hat, spricht Péringuey diesem
Merkmal generischen Wert ab und zieht die zweite Gruppe der Burmeisterschen
Gattung Gonzaspidius zu Omocrates, worin ich ihm folge. Zur Abgrenzung gegen
die neuen Gattungen Cylindrocrates und Omocnemus muB noch die nach hinten
verschmalerte Form der Fligeldecken und die einspitzige Form der Vorder-
schienen herangezogen werden. Damit sind die Merkmale der Gattung
Omocrates das groBe, dreieckige Schildchen, die schmale Koérperform, die nach
hinten verschmdlerten Fliigeldecken, die einspitzigen Vorderschienen und ihre
rechtwinklig abstehenden Seitenzahne, die in beliebiger Zahl vorhanden sein
konnen.

Der ,,Coleopterorum Catalogus’’ verzeichnet 17 Arten. Weitere sind seither
nicht beschrieben worden.

Zu den Omocrates Arten

(1) O. misellus Péringuey 1902
Die 4 Zahne des Kopfschildes sind gleichgroB, ebenso die 2 Seitenzahne
der Vorderschienen. Die Furche des Halsschildes ist tief, aber viel schmdler als

bei canaliculatus Blanch. Ich habe nur die Type, ein 2, gesehen. Lange 5,25 mm.
Namaqualand. Type in Kapstadt.

(2) O. karrooanus nova species

GroBer Omocrates mit 4-zahnigem Kopfschild, 3-zahnigen Vorderschienen
und einzelnen einfachen Krallen an den Hinterbeinen, schwarz mit blaBgelben,
seitlich angedunkelten Fligeldecken und dicht graugelb beschupptem
Pygidialteil.

3g. Kopfschild quer, etwas nach vorn schmdler werdend, von den 4 Zahnen
sind die 2 mittleren kleiner, wenig scharf und manchmal zu Kornchen reduziert.
Eine stumpfe zahnformige Erhebung iiber dem Fihleransatz. Fiihler schwarz-
lich. Halsschild fast so lang als breit, mit abgerundeten Hinterecken, dann in
den hinteren zwei Dritteln fast parallel und anschlieBend nach vorn ver-
schmalert. In der hinteren Halfte eine schwache Langsfurche. Wie der Kopf
ziemlich fein und dicht kérnig punktiert und aufrecht lang und dicht fein
gelblich behaart. Das groBe schwarze kérnige Schildchen mit weiblichen, den
Grund nicht ganz verdeckenden Schuppen. Fliigeldecken mit betonten
Schultern, nach hinten verschmalert, blaBgelb, Beulen und abfallende Teile
bis zum Nahtende angedunkelt, mit kurzen anliegenden diinnstehenden hellen

BESTIMMUNGSTABELLEN DER HETEROCHELIDES .- 241

Harchen. Innen neben der Schulter ein Eindruck, der fast bis zur Apikalbeule
reicht. Apikalrand um die Beule hinten herum dicht dunkelgelb beschuppt.
Propygidium schmal, wie das Pygidium geschlossen graugelb beschuppt mit
einigen hell behaarten Kahlpunkten. Abdomen grauweif beschuppt und wie
die ganze Unterseite dicht und lang weiBlich behaart, die Haare sind an den
Seiten besonders lang. Beine schwarz, Tarsen am Ende rot, bei einem Exemplar
aus Klaarstroom Beine dunkelpechrot. Die 3 Zahne der Vorderschienen
stehen wie bei den Heterochelus rechtwinklig ab. Hinterschenkel dick, Hinter-
schienen unter dem Knie am breitesten mit stumpfer Ecke, zum Apex schmialer,
am inneren Ende mit kurzem spitzem Mukro, am auBeren Ende mit scharfer
Ecke. Beine fein wei behaart. Vorder- und Mittelkrallen doppelt und
ungleich, Hinterkralle einzeln und einfach.

Q. Wie das 4, Fliigeldecken ohne dunklen Rand, Zahne des Kopfschildes
sehr undeutlich, Pygidium nur lang behaart, ohne Schuppen, Beine braun.
Lange 6-6,5 mm. Dikbome Merweville Koup, Klaarstroom Prince Albert.
Typen in Kapstadt.

(3) O. pauxillus Péringuey 1902

Die mittleren Zahne des Kopfschildes genahert. Oben nur der Kopf
schwarz, Halsschild rotbraun, fast so lang wie die Fliigeldecken, in den hinteren
drei Vierteln parallelseitig, dann verschmdlert. Vorderschienen nur mit 1
Seitenzahn. Nur Type (2) bekannt. Lange 4,5 mm. Hex River. Type in
Kapstadt.

(4) O. andreaei nova species

Schwarzer Omocrates mit 3-zahnigem Kopfschild, blaBgelben Fliigeldecken,
dicht gelb beschupptem Pygidialteil, braunen Hinterschienen und Tarsen und
4-zahnigen Vorderschienen. Lang und schmal. Basis des Kopfschildes kurz
verschmialert, dann ist er quadratisch, vorn mit 3 spitzen Zahnen, deren mitt-
lerer etwas kleiner ist. Kopf fein eng gekérnt, mit sehr kurzen abstehenden
Boérstchen. Halsschild so lang wie breit, in der hinteren Halfte parallelseitig,
grob punktiert, mit abstehenden kurzen feinen Harchen, am Hinterrand mit
feiner gelb beschuppter Rille. Schildchen grob gerunzelt und gelb beschuppt.
Fliigeldecken blaBgelb, stark nach hinten verschmalert, von den Hiiften an die
Seiten des Abdomen nicht verdeckend, mit kurzen langlichen Eindriicken innen
neben den Schultern und einem gemeinsamen Eindruck auf der Scheibe, in den
Eindriicken mafig dicht punktiert, sonst glatt. Propygidium breit, wie das
Pygidium beim ¢ geschlossen goldgelb beschuppt. Bauchringe mit Binden aus
weiBen Schuppenhaaren iiber die Mitte, von oben sichtbar. Brust hell behaart.
Vorder- und Mittelbeine pechschwarz wie auch die Hinterschenkel, Hinter-
schienen und -tarsen dunkelrotbraun. Vorderschienen mit 4 Zahnen, der apikale
Zahn schrag nach vorn, die anderen 3 Zahne rechtwinklig abstehend, die 2
basalen Zahne gleichstark, der Zahn zwischen diesen und dem apikalen Zahn
nur halb so lang. Diese Zahnung erinnert an Ischnochelus Burm., welche Unter-

242 ANNALS OF THE SOUTH AFRICAN MUSEUM

gattung aber ein kleines Schildchen hat. Hintertarsen langer als die Schienen,
diese einfach. Alle Beine mit 2 sehr ungleichen Krallen, beide Krallen der
HinterfiiBe gespalten, die grdéBere sehr tief, die kleine ganz fein an der
Spitze. :

@ dem ¢ sehr ahnlich, der letzte Bauchring dicht goldgelb fein beschuppt,
das Abdomen sonst wenig dicht anliegend weiblich behaart. Lange 4-5 mm.
Bosluis Pass. Gamkas Poort. Typen in Kapstadt.

(5) O. plausibilis Péringuey 1902

Die Art ist schon an dem Fleck tiber dem Schildchen aus sehr dichten
kleinen ovalen dunkelgelben Schuppen, an der sehr kornigen Struktur des
Halsschildes und an der langen dichten weiBen Behaarung der Halsschildseiten,
der Unterseite und der Schienen leicht zu erkennen. Lange 5,5 mm. Burghers-
dorp. Typen in Kapstadt.

(6) O. pygidialis nova species

Schwarzer Omocrates mit gelbbraunen Fliigeldecken, 3-zahnigem Kopf-
schild und 3-zahnigen Vorderschienen, bei dem das Propygidium und Pygidium
des 3 im Basalteil dicht dunkelbraun, auf dem Rest der Flache hellgelb
beschuppt ist. |

6. Kopfschild breiter als lang, der mittlere Zahn bei einigen Exemplaren
kleiner als die anderen Zahne. Stirn flach eingedriickt, runzlig punktiert,
abstehend fein hell behaart. Fiihler pechschwarz. Halsschild so lang als breit,
k6rnig runzlig, schwach glanzend, fein abstehend gelb behaart, an den Seiten
mit langeren weiBen Haaren, am Basalrand mit weiblichem Schuppensaum.
Mit seichter Furche im hinteren Teil. Schildchen runzlig mit schwach gekielter
Mitte, beiderseits davon gelbweif beschuppt. Fliigeldecken stark verschmalert,
sodaB das Abdomen seitlich sichtbar wird, mit deutlicher Grube innen neben
der Schulter und schwacher Rille neben der Naht, wenig dicht grob punktiert
mit kurzen feinen hellen anliegenden Borstchen, am Apikalrand mit weifem
Schuppensaum. Propygidium und Pygidium geschlossen beschuppt, die
Schuppen sind hellgelb und in der Basalgegend dunkelbraun. Nur bei einem
einzigen ¢ sind sie einfarbig hellgelb (forma col. zmmaculatus nov.). Abdomen
schwarz mit weiBen Binden iiber die Mitte der Ringe, deren Rander sichtbar
bleiben, die 3 letzten Ringe braunlich. Brust dicht und lang weif behaart.
Vorderbeine pechschwarz, Mittelbeine pechbraun, Hinterbeine rotbraun.
Vorderschienen mit schragem Apikalzahn und 2 genaherten gleichlangen
Seitenzahnen, der Apikalzahn zeigt am Grunde eine scharfe Ecke (Vorstufe
eines weiteren Zahnes?). Hinterschienen mit kurzen Sporen. Alle Krallen
doppelt, ungleich, die gréBere seitlich gespalten. Alle Beine wei beborstet.

2. Wie das 3, Halsschild langer behaart, der glanzende, schwarzbraune
Untergrund des Pygidialteiles durch eine feine helle Behaarung sichtbar. Alle
Beine braun, die hinteren heller als die vorderen, Tarsen diinner, Sporn der
Hinterbeine langer. Lange 5-5,5 mm. Middelburg Div. Typen in Kapstadt.

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 243

(7) O. variabilis Burmeister 1844

Die Fliigeldecken dieser Art sind im Grunde pechschwarz oder hellbraun,
die Bérstchen darauf sind starker als beim vorigen. Der Pygidialteil ist einfarbig
gelbweiB. Sonst dem vorigen ahnlich, kleiner. Lange 3,5-4,5 mm. Port
Elizabeth, Uitenhage, Dunbrody. Typen in Halle.

(8) O. axillaris Burmeister 1844

Genotyp. Schon an der GroSe und der Form der Hinterschienen des 3
sofort zu erkennen. Das Pygidium des 9 tragt im Gegensatz zu dem einfarbig
goldgelb beschuppten Pygidium des ¢ zwei kleine dunkle Flecken. Lange
6,5-7 mm. Stellenbosch, Koeberg, Malmesbury, Clanwilliam. Type in Halle.

(9) O. lobipes Burmeister 1844

Dem vorigen ahnlich, kleiner und Hinterschienen des ¢ anders. Lange
4,5 mm. Algoa Bay. Die Type befindet sich nicht mehr in Halle und ist
offenbar verloren gegangen. Neotypus in Kapstadt.

(10) O. modestus Péringuey 1902

Das Propygidium ist hier auffallend breit, der ganze Pygidialteil dicht
graugelb beschuppt. Lange 5 mm. Kowie (Port Alfred). Type im Pariser
Nationalmuseum.

(11) O. hesset nova species

Sehr lang und schmal gebauter schwarzer Omocrates mit blaBgelben, an den
Beulen angedunkelten Fliigeldecken und bei G2 orangegelb beschupptem
Pygidialteil, der sich durch 2-zaéhnigen Kopfschild und 4-zahnige Vorder-
schienen auszeichnet. Der Kopfschild ist parallelseitig und verhaltnismaBig
schmal, die Ecken sind stark aufgebogen, der Rand zwischen ihnen zeigt eine
sehr schwache Woélbung (Vorstufe eines dritten Zahnes), der ganze Kopf ist
grob runzlig-kérnig punktiert. Fiihlerstiel braun, Facher schwarz. Der Hals-
schild ist etwas langer als breit, hinten wenig breiter als vorn und stark zum
Schildchen vorgezogen, mit abgerundeten Ecken und schwach gebogenen
Seiten. Er ist verhaltnismaBig flach, vorn unregelmaBig punktiert, hinten fast
glatt, jeder Punkt tragt ein gelbes, an den hinteren Seiten weifes Haar.
Schildchen grob punktiert mit glatter Mittellinie, seitlich weiBlich behaart.
Fliigeldecken hinter den Schultern stark verschmdlert, auf der Scheibe fast
glatt, in dem Eindruck neben der Schulter und seitlich unregelmaBig punktiert,
mit schmaler Rille neben der Naht, mit unregelmaBig gereihten kurzen halb
anliegenden weiBen Bérstchen. Am Beginn des Propygidiums ein Kranz gelber,
seitlich weiBer Boérstchen, dieses sehr breit und wie das Pygidium bei ¢9°
geschlossen orangefarben beschuppt und dazwischen behaart, beim ¢ sind die
Schuppen rund, beim @ langlich. Das Pygidium fallt beim ¢ schrag nach vorn,
beim 9 schrag nach hinten ab. Abdomen beim ¢ schwarz, beim 2 rotbraun,
die Ringe oben weiBlich, sonst gelb behaart. Brust beim ¢ wei8, beim 92

244 ANNALS OF THE SOUTH AFRICAN MUSEUM

weniger dicht gelblich behaart. Vorder- und Mittelbeine beim 3 schwarz mit
braunlichem Schimmer, Hinterbeine des ¢ und alle Beine des 2 rotbraun. An
den Vorderschienen ist der Zahn nach dem Apikalzahn kiirzer.

Die Art hat Ahnlichkeit mit O. andreaei n. sp., ist aber gréBer und hat nur
2 Kopfschildzahne. Lange 5-6 mm. Tankwa Karroo Waterval. ‘Typen in
Kapstadt.

(12) O. depressus Blanchard 1850

Durch die einschlieBlich des Halsschildes gelb beschuppte Oberseite
hervorgehoben. Von placens Pér. durch parallele Halsschildform und dunkel-
gelbe Farbe der Schuppen verschieden. Lange 4-4,5 mm. Cape Town,
Stellenbosch. Typus im Pariser Nationalmuseum.

(13) O. placens Péringuey 1902
Sehr ahnlich dem vorigen, aber Halsschild gleichmafig verschmalert und

Schuppenfarbe mehr grau. Lange 3,5-4 mm. Stellenbosch, Caledon. Type 1 in
Kapstadt.

(14) O. spatulipennis Blanchard 1850 (=J/epidus Boheman 1857)

Die Type blieb mir unbekannt. Type des lepzdus Boh. im Museum —
holm. Lange 4,5-6mm. Natal, Orange Freistaat. —

(15) O. liwidipennis Boheman 1857

Als Goniaspidius beschrieben, der Autor vergleicht die Art aber mit seinem
lepidus. Lange 5 mm. ,,Ganz Caffraria.”” Typus im Museum Stockholm.

(16) O. humilis Péringuey 1902

Durch die Form der Zahne der Vorderschienen von allen 4hnlichen kleinen
Arten verschieden. Lange 4,5 mm. Hex River. Typus in Kapstadt

(17) O. placidus Péringuey 1902
Der Halsschild ist auffallend grob weitlaufig punktiert und hat hinten

keine Furche. Dem luridipennis Burm. ahnlich, Pygidium aber einfarbig. Lange
4,5-5,5 mm. Hex River. Type in Kapstadt.

(18) O. pseudoplacidus nova species

Schwarzer Omocrates mit blaBgelben, hinten angedunkelten Fliigeldecken
aus der Verwandtschaft des placidus Pér., der sich davon besonders durch den
tief gefurchten Halsschild unterscheidet.

Form ahnlich luridipennis Burm., Kopfschild vorn gerade, etwas aufgebogen
mit ein wenig vorstehenden Ecken. Die blaBgelben Fliigeldecken an dén
Schultern rotbraun, hinten mit einem pechschwarzen, verkehrt-keilformigen
auf der Apikalbeule endenden Fleck. Der Halsschild ist gew6lbter und dichter
und feiner punktiert als bei placidus und hat eine fast bis zum Vorderrand
reichende tiefe Furche. Am Hinterrand ein Schuppensaum, sonst ist der
Halsschild lang dicht abstehend braunlich behaart. Schildchen mit Mittelkiel,

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 245

- daneben gelb beschuppt. Auf den Fliigeldecken lauft der Eindruck neben der
Schulter fast bis zum Apex, neben der Naht eine feine Rille. Oberflache fast
glatt, in den Eindriicken punktiert, auf den abfallenden Teilen gestrichelt, mit
zerstreuten, im apikalen Teil dichteren feinen anliegenden weifen Harchen,
hinter den Apikalbeulen mit Schuppenbindchen. Propygidium breit, wie das
Pygidium beim ¢ dicht goldgelb beschuppt. Bauchringe dicht gelb behaart,
Brustseiten mit langeren buschigen weiSen Haaren. Alle Beine rotbraun, die
4 vorderen etwas dunkler. Schenkel und Schienen lang gelb beborstet. Vorder-
schienen mit schragem, am Grund eckigem Apikalzahn, von den 2 recht-
winklig abstehenden Seitenzahnen ist der basale sehr klein. Hinterschienen mit
1 Sporn. Alle Schienen mit doppelten Krallen, deren langere gespalten ist.
Nur gd bekannt. Lange 5:5-6 mm. Stellenbosch. Type in Pretoria.

(19) O. luridipenns Burmeister 1844

Dem axillaris ahnlich, Kopfschild aber nur mit Eckz4hnen. Die Fliigel-
decken der Typen sind blaBgelb, an den Beulen angedunkelt. 4 Individuen des
_ Siidafrika-Museums sind dunkelbraun, fast schwarz; da sie 1883 gefangen
worden sind, halte ich eine postmortale Veranderung der Farbe fiir wahr-
scheinlich. Das auffallendste Kennzeichen der Art sind die 4 (beim 9 2)
dunkelbraunen Flecken auf dem Pygidialteil. Lange 4,5-5 mm. Cape Town,
Stellenbosch, Koeberg, Malmesbury, Paleisheuwel. Typen in Halle.

(20) O. mendax Péringuey 1902

Den 3 vorausgehenden Arten und axillaris Burm. ahnlich, durch die nur
2-zahnigen Vorderschienen sofort zu unterscheiden. Pygidialteil des ¢ einfarbig
gelb, das Propygidium des 9 hat 2 sich wenig abhebende dunklere Flecken.
Lange 6-6,5 mm. Namaqualand. Type in Kapstadt.

(21) O. cylindricus Burmeister 1844 (=flavipennis Blanchard 1850)

Bekannte Art mit rudimentarer Halsschildfurche. Lange 4 mm. Cape
Town, Camps Bay, Stellenbosch, Tulbagh. Typen in Halle.

(22) O. namaquensis nova species

Schwarzer Omocrates mit gelbbraunen Fliigeldecken aus der Verwandt-
schaft des cylindricus, der durch breiteren 2-zahnigen Kopfschild, gewélbteren
Halsschild und nur sparlich beschuppten Pygidialteil unterschieden ist.

Wahrend der Kopfschild des cylindricus Burm. verschmalert ist, bleibt er
hier vorn ungefahr so breit wie an der Basis, die Ecken sind schrag aufgebogen.
Der grobgerunzelte Kopf ist auf der Stirn kurz hellbraun beborstet. Der Hals-
schild ist gew6lbter und fallt zum Kopf fast so steil ab wie bei den Dicranocnemus.
Er ist ein wenig breiter als lang, bis zur Mitte parallelseitig, dann allmahlich
verschmalert, fein gerunzelt, dicht lang aufrecht bradunlich behaart, ohne
Schuppen. Hinten mit rudimentarer Furche. Das glanzend schwarze Schild-
chen zeigt keine Schuppen, allerdings ist bei dem einzigen vorhandenen

246 ANNALS OF THE SOUTH AFRICAN MUSEUM

Exemplar die Nadel durch das Schildchen gesteckt, wodurch Schuppen
verdeckt worden sein kénnten. Die Fliigeldecken sind glanzend gelbbraun, mit
tiefen Eindriicken neben der Schulter bis zur Mitte der Scheibe und neben der
Naht, auf den erhobenen Stellen fast glatt, sonst zerstreut punktiert, unbehaart
und unbeschuppt. Pygidialteil und Abdomen glanzend schwarz, mit kleinen
borstenférmigen hellen Schuppen, groBtenteils aber unbeschuppt, fein punk-
tiert. Bauchringe zur Halfte hell beschuppt. Brust dicht und lang gelb behaart.
Alle Beine pechschwarz, alle Krallen doppelt, ungleich und die gr6Bere deutlich
gespalten. Nuri jg bekannt. Lange 4mm. Bowesdorp Namaqualand. Typus
in Kapstadt.

(23) O. elongatus Blanchard 1850

Blieb mir in natura unbekannt und ist nur nach der Beschreibung ein-
geordnet. Lange 5,25 mm. ,,Kap.” Typus im Pariser Nationalmuseum.

(24) O. canaliculatus Blanchard 1850

Diese etwas kiirzer als die anderen kleinen Arten gebaute Art ist sehr leicht
an der tiefen, den ganzen Halsschild durchlaufenden Furche zu erkennen, welche
das ganze innere Drittel des Halsschildes ausfillt. Lange 4 mm. Hopefield.
Typus im Nationalmuseum Paris.

3. CGYLINDROCRATES novum genus

Diese neue Gattung besitzt die Eigenschaften der Gattung Omocrates bis auf
eine, die nach hinten verschmalerten Fliigeldecken, welche das Abdomen nicht
ganz bedecken. Das groBe Schildchen ist dreieckig, die Korperform schmal,
die Vorderschienen sind einspitzig und ihre Seitenzahne stehen rechtwinklig ab.
Die Fliigeldecken sind fast parallelseitig und bedecken das Abdomen ganz, ihr
Seitenausschnitt ist ein sehr flacher Bogen. Das Propygidium fallt beim 3
senkrecht ab, das Pygidium schrag nach vorn. Das 9 ist noch unbekannt.

Die Zahl der Seitenzahne halte ich fiir kein Gattungsmerkmal. Genotyp
und bisher einzige Art ist parallelus nova species.

C. PARALLELUS nova species

Schwarz, Grundfarbe der hinteren Halfte des Halsschildes rotbraun, mit
hellbrauner geschlossener feiner Beschuppung auf Hinterkopf, vorderem Drittel
des Halsschildes und seinem Hinterrand, Schildchen und Fliigeldecken; mit
geschlossener orangeroter Beschuppung des Pygidialteiles und letzen Bauch-
ringes, mit dichter weiBer Beschuppung der Restflache des Abdomen und der
Seitenstiicke der Brust und mit weiBer Behaarung der Brustseiten.

Kopfschild glanzend schwarz, verschmalert, fein runzlig-kornig, vorn
gerade, an den Ecken in schmale spitzige Zahne aufgebogen. Stirn starker
punktiert und hellbraun beschuppt, ohne Haare. Halsschild etwas langer als
breit, hinten bogig zum Schildchen vorgezogen, mit verrundeten Hinterecken,
in den hinteren zwei Dritteln fast parallel, dann verschmalert, vorn gerade

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 247

abgeschnitten. Der gr6éBte Teil der glanzenden, stark punktierten, ziemlich
flachen Oberflache ist unbeschuppt, vorn schwarz, hinten rotbraun. Die
abfallenden Seiten des vorderen Drittels und ein schmaler Hinterrand sind dicht
hellbraun beschuppt. Die Schuppen sind langlich. Das groBe gleichseitig-
dreieckige kérnige Schildchen ist dicht beschuppt mit Ausnahme der Mitte der
Basis auf eine kurze Strecke. Die Fliigeldecken sind 14 mal so lang als zusammen
an den Schultern breit, fast parallel, flach, mit steil abfallenden Seiten, deshalb
ist der flache Seitenausschnitt von oben nicht zu sehen. Apex gemeinsam
abgerundet mit rechtwinkligen Nahtecken. Schulter- und Apikalbeulen nicht
betont. Auffallend ist eine schwarze eingegrabene Linie, die in gerader
Richtung von der oberen Ecke des Schildchens schrag nach auBen zum Beginn
des Seitenausschnitts lauft und nicht beschuppt ist. Die Seiten des Abdomen
sind von den Fliigeldecken ganz bedeckt und nicht sichtbar. Ohne Haare.
Hautfliigel rauchbraun. Die orangeroten Schuppen des Pygidialteiles sind noch
kleiner als die der Fliigeldecken. Die Vorderbeine sind pechschwarz, die
anderen Beine rotbraun. Die Vorderschienen zeigen 6 Zahne, auf den schrag-
abwarts gerichteten Apikalzahn folgen nach einer Liicke die rechtwinklig und
etwas nach unten gebogenen Seitenzahne, zunachst ein groBer, dicht daneben
ein nur wenig kiirzerer und dann die 3 restigen wesentlich kiirzeren und etwas
weiter auseinander stehenden Zahne, deren letzter sehr kurz ist. Schenkel mit
weifBen, Schienen mit gelblichen Borstenhaaren. Alle Beine mit doppelten
ungleichen Krallen. Die groBere ist tief gespalten. Lange 5,5, Breite 2 mm.
Wallekraal, Namaqualand. Nur 1 3 liegt vor. Typus in Kapstadt.

4. Gontaspipius Burmeister 1844 (nec Péringuey 1902). Genotyp brevis Burm. 1844

Schon durch den Vergleich der Beschreibungen wird klar, daB jeder dieser
Autoren einen anderen Kafer meint. Besonders beim Schildchen und in der
Zahl der Krallen der Hinterbiene wird dies deutlich und der Vergleich der
Typen Burmeisters mit den von Péringuey als brevis bestimmten Individuen des
Siidafrika-Museums beseitigt alle Zweifel. Das Schildchen bei unserer Gattung
ist ein gleichschenkliges Dreieck, dessen Basis ein wenig kiirzer ist als die Seiten.
Im Gegensatz dazu ist es bei der Péringueyschen Gattung in der Hauptsache
ein gewolbtes Viereck, dem hinten ein sehr flaches stumpfwinkliges Dreieck
angehanet ist.

Die sonstigen Eigenschaften teilt die Gattung mit Omocrates mit Ausnahme
der kurzen breiten Koérperform; der Autor sagt sie verhalte sich zu Omocrates
wie diese Gattung zu den breiten Heterochelus, aber mit groBem Schildchen.
Zu den Gontaspidius-Arten:

(1) G. brevis Burmeister 1844 (nec Péringuey 1902)

Genotyp. Durch einkrallige Hinterschienen gekennzeichnet. Kopfschild
mit 3 starken Zahnen. Schwarz, Fliigeldecken des ¢ dunkelrotbraun mit
schwarzem Saum oder ganz schwarz, beim @ Fliigeldecken, Abdomen und Beine
ganz rotbraun. Vorderschienen mit schrag abwarts gebogenem Apikalzahn

248 ANNALS OF THE SOUTH AFRICAN MUSEUM

und 2 senkrecht abstehenden Seitenzahnen. Uberall ohne Bindenbildung lang
abstehend dunkel behaart. ‘Tarsenglieder zylindrisch. Lange 6,5 mm.
,sudafrika.”” Ich habe nur die Typen gesehen. Typen in Halle.
(2) G. lebist (Schein)

Ich habe diese Art als Omocrates in einer Arbeit fiir das Institut des Parcs
Nationaux du Congo Belge beschrieben, deren Manuskript wohl vor dieser
Arbeit gedruckt werden wird.

Hinterbeine mit 2 Krallen. Kopfschild 3-zahnig. Schwarz, Fliigeldecken
rotbraun mit 2 unbeschuppten Rippen und 3 Langsbinden aus dicken weiBen
Harchen, Pygidialteil von rundlichen wenig dichten gelben Schuppen bedeckt,
Vorderschienen 2-zahnig. Lange 3 mm. Congo Belge, Upemba Park. Type
in dem genannten Institut des Parcs Nationaux du Congo Belge.

(3) G. angolensis nova species

Schwarzer Gonzaspidius mit rotbraunen ohne Bindenbildung weiB behaarten
matten Fligeldecken, 2-zahnigen Vorderschienen und 2-kralligen Hinterbeinen.

Schwarz, Fligeldecken und Beine rotbraun. Kopfschild quer, parallel-
seltig, vorn mit 3 aufgebogenen Zahnen, wie der Kopf fein punktiert, mit
dunklen Harchen. Halsschild so lang als breit, bogig zum Schildchen vor-
gezogen, mit stumpfen Hinterecken, in den hinteren zwei Dritteln wenig, dann
stark verschmAlert, vorn gerade, die spitzen Vorderecken unter die Augen ein-
gebogen, im hinteren Teil stark gewélbt, mit von der Mitte an tiefer Furche
und schmalen Eindriicken am Hinterrand. Sehr fein punktiert, matt, in der
vorderen Halfte aufrecht gelb behaart, am Seiten- und Hinterrand mit kurzem
gelbem Schuppensaum. Das grofe Schildchen gleichseitig-dreieckig, runzlig,
matt, ohne Haare oder Schuppen. Fliigeldecken so lang als zusammen breit,
verschmalert und das Abdomen nicht bedeckend, Nahtecken rechtwinklig.
Durch Eindriicke neben Schulter und Naht und auf der Scheibe uneben.
Uberall mit gereihten, nicht dichten kurzen weiBen Bérstchen. Apikalrand mit
feinem gelbem Schuppensaum. Pygidialteil fein gelb beschuppt und dazwischen
behaart. Das Pygidium fallt bemm 4 senkrecht, beim 9 schrag nach hinten ab.
Abdomen schwarz, an den letzten 3 Ringen braun mit dichten gelben Borsten-
schuppen. Brust schwarz, seitlich gelb behaart. Vorderschienen mit gerade
vorgestrecktem Apikalzahn und einem schrag nach unten gerichteten Seiten-
zahn. Hinterschienen bei $2 mit Sporn. Alle Krallen doppelt und ungleich,
die langere gespalten. Lange 5 mm. Angola, Kuangu. Typen im Museum
Georg Frey in Tutzing vor Miinchen.

5. RECTOSCUTARIA nomen novum, fiir Goniaspidius Péringuey 1902 (nec Bur-
meister 1844)

Genotyp péringueyt nom. nov. |
Da, wie oben ausgefiihrt, Péringuey einen anderen Kafer fiir Goniaspidius
brevis Burm. gehalten hat, miissen seine Gattung und seine Art neue Namen

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 249

erhalten. Thre Beschreibung steht im ,,Descriptive Catalogue of the Coleoptera
of South Africa” von Péringuey Seite 785/6. Hauptmerkmal ist die Form des
groBen Schildchens. Es stellt ein Viereck dar, dessen untere Seite die Basis eines
gleichschenkligen Dreiecks mit sehr geringer Hohe ist, weshalb das Schildchen
sehr stumpfwinklig endet.

Die bekannten Arten haben am Vorderrand des Kopfschildes drei starke
Zahne, an den Vorderschienen auBer dem Apikalzahn zwei Seitenzahne, an
allen Beinen zwei ungleiche Krallen und bei ¢@ einen Sporn an den Hinter-
schienen.

Zu den Rectoscuraria-Arten:
(1) R. péringueyx nomen novum fiir brevis Péringuey nec Burmeister.

Die Fliigeldecken dieser schwarzen Art sind sehr uneben und schwarz oder
ganz oder teilweise braun, das Pygidium ist gelb beschuppt, die Hintertarsen
der 3¢ sind etwas langer als die Schienen, die Tarsenglieder sind am Ende in
scharfe Ecken ausgezogen, nicht zylindrisch wie bei der echten brev:s. Nach
der Beschuppung der Oberseite lassen sich drei Rassen unterscheiden:

Nominaiform: Die weiBen anliegenden haarahnlichen Schuppen bilden 3
Langsbinden auf jeder Fliigeldecke. Lange 5-6 mm. Touwsrivier, Dikbome
Merweville Koup, Klaarstroom Prince Albert, Lammerskraal Prince Albert,
Wilowmore. |

Rasse lunata nov. Jede der 3 Binden ist vorn und hinten derart verkirzt,
daB alle 6 Binden zusammen einen nach vorn offenen Bogen bilden. Meist ist
der Grund der Fliigeldecken im Bereich dieser bogigen unterbrochenen Quer-
binde rotbraun. Lange 5-6 mm. Brandkop Nieuwoudtville, Willowmore.

Rasse uniformis nov. Diese hat keine Binden, sondern auf der ganzen Flache
halb abstehende feine graue Harchen. Lange 5-6 mm. Willowmore. Typen
der Nominatform und lunata in Kapstadt, der unzformis in Pretoria.

(2) R. simplex Péringuey 1902

Schwarz, Fliigeldecken dunkelrotbraun mit angedunkelten Seiten und
Beulen, Fliigeldecken weniger uneben, nur unterhalb des Schildchens flach
eingedriickt. Der Hauptunterschied liegt in der Form des Apikalzahnes, der
mehr umgebogen und fast parallel zu den Seitenzahnen ist. Lange 5,6-7 mm.
Touwsrivier. Type in Kapstadt.

6. OmocNEMUS genus novum

Diese Gattung steht zwischen Omocrates und Dicranocnemus. Von jener hat
sie das groBe Schildchen, von dieser die beim ¢ zweispitzigen Vorderschienen.
Sie ist die einzige Gattung der Heterochelides, bei der die Zahnung der Vorder-
schienen nach dem Geschlecht verschieden ist. Diese Schienen besitzen 3 scharfe
Zahne; beim ¢ sind die beiden apikalen Zahne gendhert, parallel gestellt und
am Grunde verwachsen, beim @ sind sie getrennt, der mittlere Zahn steht vom

250 ANNALS OF THE SOUTH AFRICAN MUSEUM

apikalen weiter ab und zeigt fast rechtwinklig nach auBen. Der basale Zahn
ist bei beiden Geschlechtern ein wenig kiirzer.

Der Halsschild ist flach, das Schildchen ein gleichseitiges Dreieck mit etwas
ausgebogenen Seiten. Das Pygidium fallt beim ¢ schraég nach vorn, beim 92
steil nach hinten ab. Genotyp und bisher einzige Art ist kochi nova species.

O. kochi nova species

Schwarzer Omocnemus mit rotbraunen, hinten angedunkelten, nach hinten
verschmalerten im hinteren Teil weiBgelb behaarten Fliigeldecken, gelb
beschupptem Pygidium, pechschwarzen Vorderbeinen, rotbraunen wbrigen
Beinen und stumpf-dreizahnigem Kopfschild.

Letzterer ist quer, fast parallelseitig mit stumpf aufgebogenen Ecken,
dazwischen ein breiter oben abgerundeter aufgebogener Lappen, der aus 2
zusammengewachsenen Zahnen entstanden sein diirfte. Der ganze Kopf ist
kérnig-runzlig. Die Fihler sind braungelb mit schwarzem Facher. Der Hals-
schild ist hinten so breit als lang, hinten bogig zum Schildchen ausgezogen, mit
rechtwinkligen, die Schultern umfassenden Hinterecken, nach vorn gleich-
maBig verschmalert, sehr fein punktiert und glanzend, unbehaart, iber dem
Schildchen mit gelblichem Schuppensaum, sonst ohne Schuppen. Schildchen
runzlig, mit Ausnahme der Mittellinie dicht weiB beschuppt. Fliigeldecken mit
starkem Seitenausschnitt und betonten Beulen, Seiten steil abfallend, am Apex
einzeln abgerundet, mit Eindriicken neben der Schulter, auf der Scheibe und
langs der Naht. Die ganze Flache ist sehr fein runzlig-punktiert und schwach
elanzend, in der Nahtdepression, auf den abfallenden Seiten und auf den
hinteren zwei Dritteln der Fliigeldecken sind dichte feine anliegende weibgelbe
Harchen vorhanden. Propygidium sammetschwarz mit gelbem Apikalrand,
Pygidium des ¢ dicht orangegelb, des 9 heller gelb beschuppt. Das Abdomen,
dessen Seiten von oben sichtbar sind, ist oben wei, sonst gelblich beschuppt.
Die Seitenstiicke der Brust sind wei beschuppt, dazwischen hell behaart. Die
Tarsen der pechschwarzen Vorderbeine sind braun, die anderen 4 Beine sind
rotbraun, die Beine sind schlank ohne besondere Anhange. Alle Krallen sind
doppelt und ungleich, alle gespalten. Das 9 ist—abgesehen von der Form des
Pygidiums, des Abdomen und der Zahnung der Vorderschienen—dem ¢ sehr
ahnlich. Lange 5-6 mm. Aus, Great Namaland. Herrn C. Koch in Pretoria
gewidmet. T'ypen in Pretoria.

7. DicRANOCNEMUS Burmeister 1844

Die zweispitzigen Vorderschienen und der gewoélbte Halsschild sind die
Merkmale dieser Gattung. Burmeister hat nur Arten mit kleinem Schildchen
gekannt; inzwischen sind auch solche mit groBem Schildchen wie bei Omocrates
gefunden worden, auf welche ich die Untergattung Macrodicranocnemus novum
subgenus errichte. Ihr einziges Unterscheidungsmerkmal gegen Dicranocnemus
Burm. sensu stricto ist das groBe Schildchen. Subgenotyp andreaei n. sp.

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 251

Zu den Macrodicranocnemus-Arten:

Bisher sind zwei Arten gefunden worden, deren eine in 2 Rassen vorkommt.

(1) D. andreaei nova species

Schwarzer Macrodicranocnemus mit braunen Fliigeldecken, 2-zahnigem
Kopfschild, lang und dicht aufrecht behaartem Halsschild, gelb beschupptem
Pygidialteil und schwarzen Beinen. Der nach vorn stark verschmd4lerte Kopf-
schild ist in 2 starke Zacken aufgebogen. Der Kopf dicht gekérnt, oben lang
und dicht rotbraun behaart, seitlich abstehend schwarz beborstet. Fiihler
schwarz. Der stark gewélbte Halsschild ist hinten bogig abgeschnitten, hat
stumpfe Hinterecken, ist in den hinteren zwei Dritteln parallelseitig und dann
verschmalert und besitzt eine tiefe Furche, die fast bis zum Vorderrand reicht.
Er ist dicht und fein punktiert und dicht und lang aufrecht gelblich behaart.
Das groBe dreieckige Schildchen schwarz mit gelben nicht dichten Schuppen.
Fliigeldecken hellbraun mit unregelmaBig verteilten gelben Schuppen, die quer
uber die Scheibe beider Fliigeldecken 4 Fleckchen und hinter den Apikalbeulen
bogige Bindchen bilden. Mit Eindriicken neben der Schulter und auf der
Scheibe, narbig punktiert, schwach glanzend. Pygidialteil sehr dicht dunkelgelb
beschuppt und abstehend gelb behaart, wie auch die Seiten des sonst schwarzen
Abdomen. Brust lang braunlich behaart. Beine schwarz mit braunen Bérstchen
besonders an den Hinterschienen, die bei $9 einen langen Sporn tragen. Alle
Krallen doppelt und ungleich, die langere gespalten. Lange 6-7,5 mm. Seven
Weeks Poort Berg Ladismith, Riversdale, Tradouw Pass Swellendam. Type in
Kapstadt.

(1a) D. andreaei subspec. zumpti nova

Diese Rasse unterscheidet sich durch dunkelrotbraune, an den Schultern
und am Apex angedunkelte Fliigeldecken und dunkelbraune, vorn fast schwarze
abstehende Behaarung des Halsschildes. Sonst von der Nominatform nicht
verschieden. Lange 6—7,5 mm. Robinson Pass nordwestlich der Mossel Bay.
Type im Museum Georg Frey in Tutzing vor Miinchen.

(2) D. hirtipes nova species

Schwarzer Macrodicranocnemus mit braunschwarzen Fliigeldecken mit gelber
Schuppenzeichnung, der sich durch lange buschige Behaarung der Hinter-
schienen und ihrer Tarsen auszeichnet. Kopf und Halsschild sehr 4hnlich dem
vorigen, tiefschwarz mit kiirzerer braunschwarzer Behaarung. Fliigeldecken
schwarz, auf der Scheibe dunkelrotbraun, mit einer Zeichnung aus hellgelben
Schuppen; sie besteht aus je einem kurzen Streifen neben der Mitte der Naht,
je einem doppelt so langen Streifen auf der Scheibe und je 2 rundlichen kleinen
Flecken an und neben dem Ende der Naht. Der Pygidialteil ist hellgelb
beschuppt, das Propygidium dichter als das Pygidium. Ebenso beschuppt sind
die Seiten des schwarzen Abdomen. Brust schwarzlich behaart. Beine schwarz
mit braunen Krallen, die Hinterschienen mit ihren Tarsen sind auffallend dicht
und lang buschig schwarz behaart. Alle Krallen sind doppelt und ungleich.

252 ANNALS OF THE SOUTH AFRICAN MUSEUM

Die 99 gleichen den $3, die wenigen Exemplare sind kleiner als diese, was
ein Zufall sein kann. Lange 5-6 mm. Leipoldtville Elandsbaai, Graafwater.
Typen in Kapstadt.

Zu den Arten von Dicranocnemus sensu stricto. Genotyp sulcicollis Wied.
(3) D. natalensis Péringuey 1902

Die Art ist schon durch den flacheren Halsschild von den anderen sehr
verschieden. Lange 5,25-5,5 mm. Natal, Estcourt. Cape Province, Durbrody.
Type in Kapstadt.

(4) D. spiniceps Péringuey 1904
Einzige Art mit kurzem Trochanterdorn beim g. Lange 4,5 mm. Willow-
more, Ashton. Type in Kapstadt.

(5) D. pulcher Péringuey 1902

Diese und die sulcicollis-Formen haben ein sehr 4hnliches Zeichnungsmuster,
das sich auch bei hypocrita findet. Man unterscheidet sie am sichersten am
Kopfschild. Dieser ist bei pulcher breiter, mit rechtwinkligen, nicht oder héch-
stens als kleines Ko6rnchen aufgebogenen Ecken. Bei den sulcicollis-Formen ist
er nach vorn verschmdlert und sind seine Vorderecken in scharfe Zacken
aufgebogen. Bei hypocrita sind die Ecken abgerundet und gar nicht aufgebogen.
Lange 5-5,5 mm. Barrydale, George, Grahamstown, Port Alfred. Typen in
Kapstadt.

(6) D. sulcicollis Wiedemann 1821

Halsschild stark aufrecht hell behaart und in der Furche und hinten an den
Seiten gelb beschuppt. Die braunen Fliigeldecken haben auBer dem Nahtfleck
mehrere gelbe Schuppenstreifen. Das 9 ist 4hnlich, nur ohne Schuppen auf dem
Halsschild. Lange 4,5-6,5 mm. Cape Town, Salt River, Zeekoe Vlei, Cape
Flats, Stellenbosch, Hex River. Typus im Museum Kopenhagen.

(6a) D. sulcicollis subspecies niger nova

Man kG6nnte diese Rasse fiir eine gute Art halten, weil sie durch die tief- —
schwarze Grundfarbe des ganzen Ké6rpers und nur einen sehr hellgelben
Schuppenstreifen auf der Scheibe der Fliigeldecken sehr abweichend aussieht.
Da sie aber alle sonstigen Merkmale mit der Stammform gemeinsam hat, ist
sie besser als Rasse zu werten. Der schwarz behaarte Halsschild ist in der Furche
und hinten an den Seiten beim ¢ beschuppt, das sehr ahnliche 2 hat auf dem
Halsschild fast keine Schuppen. Die Beine sind schwarz, nur die Tarsen und
die Spitzen der Seitenzahne der Vorderschienen sind rotbraun. Lange 5—5,5 mm.
Obere Quellen des Olifants River in Ceres, Elandsbaai, Kamieskroon in Nama-
qualand. Typen in Kapstadt, Paratypen im Museum Georg Frey in Tutzing.

(6b) D. sulcicollis subspecies firscht nova

Diese Rasse unterscheidet sich von der Nominatform durch langere und
heller gelbe Behaarung des Halsschildes, den konstant fehlenden auBeren

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 253

Schuppenstreifen der Fliigeldecken und die mehr graugelbe Beschuppung des
Pygidiums; in allen sonstigen Merkmalen gleicht sie dem sulcicollis. Dem
Spezialisten fiir Coccinelliden Fiirsch in Miinchen gewidmet. Lange 5 mm.
Saldanha Bay. Type in Kapstadt.

(7) D. nudus nova species 3

Schwarzer Dicranocnemus aus der Verwandtschaft des sulcicollis Wied. mit
gelbbraunen Fliigeldecken ohne Zeichnung, weiBgelb behaartem Propygidium
und nur sehr diinn staubférmig hell beschupptem, fast nackt erscheinendem
Pygidium.

Kopf und Halsschild ahnlich sulcicollis, doch sind die Ecken des Kopf-
schildes weniger hoch aufgebogen. Die Halsschildfurche ist hier bei g9
unbeschuppt, ebenso das schwarze Schildchen. Der Kopf ist fein gekérnt, der
Halsschild fein dicht punktiert, maBig glanzend und nicht sehr lang dicht
aufrecht gelb behaart. Ebenso das Schildchen. Die hell gelbbraunen Fliigel-
decken sind nackt, maBig glanzend und tragen nur am Rande einige helle
Harchen. Die weiBgelbe Behaarung des Propygidiums ist ziemlich dicht und
- liegt an, das Pygidium ist glanzend pechschwarz, die feinen staubférmigen
Schuppen sieht man mit bloBem Auge kaum. AuBer ihnen sind einige wenige
abstehende weibliche Haare vorhanden. Das schwarze Abdomen ist an den
Seiten ahnlich behaart wie das Propygidium. Auch die Brust tragt weiBliche
Haare. Die Beine sind braun, die Hinterscheinen haben bei ¢@ einen Sporn,
alle Krallen sind doppelt und ungleich. Das 9 ist dem ¢ sehr Ahnlich. Die
Art ist von dem oberflachlich ahnlichen natalensis durch den gezahnten
Kopfschild, den gewolbten Halsschild und die unbeschuppten und unbehaarten
Fligeldecken leicht zu unterscheiden. Lange 4,5 mm. Saldanha Bay. Typen
in Kapstadt.

(8) D. mendicus Péringuey 1902

Die Art ist an dem mit Ausnahme des vorderen Teiles dicht beschuppten
Halsschild und an den grauweiBen Schuppenstreifen auf den sonst braunen
Fliigeldecken leicht zu erkennen. Nach der mir vorliegenden Reihe scheint die
GroBe der Art von Westen nach Osten abzunehmen. Alle Exemplare aus
Fort Beaufort messen nur 4 mm. Beim @ sind die Schuppen durch Harchen
ersetzt. Lange 4-4,5 mm. Klaarstroom Prince Albert, Patentie Humansdorp,
Uitenhage, Port Elizabeth, Algoa Bay, Dunbrody, Albany Resolution, Grahams-
town, Fort Beaufort. Typen in Kapstadt.

(9) D. hypocrita Péringuey 1902

Ahnlich pulcher und sulcicollis gezeichnet, ist die Art an den abgerundeten
Ecken des Kopfschildes zu erkennen. Lange 4 mm. Villiersdorp, Hawston,
Port Elizabeth. Typen in Kapstadt.
(10) D. burchelli Arrow 1917

Diese mir in natura unbekannte Art ist nur nach der Beschreibung ein-
gereiht. Lange 4,5-5,5 mm. Uitenhage. Typen Britisches Museum in London.

254 ANNALS OF THE SOUTH AFRICAN MUSEUM

(11) D. arduus Péringuey 1904
Auch diese Art kenne ich nicht in natura. Lange 4 mm. Uitenhage
Port Elizabeth. 2 Typen in Kapstadt.

(12) D. pulverulentus Burmeister 1844

Ahnlich mendicus, aber Fligeldecken einfarbig hellgelb beschuppt. Lange
4,5 mm. Knysna, Patentie Humansdorp, Uitenhage, Grahamstown. Typen in
Halle.

(13) D. squamulatus Burmeister 1844

Von dem folgenden am leichtesten durch die deutlichen Kopfschildecken
und die fast zeichnungslosen Fligeldecken zu unterscheiden. Lange 4-5 mm.
Port Elizabeth, Uitenhage, Dunbrody, Grahamstown. Typen in Halle.

(14) D. squamosus Burmeister 1844

Die Ecken des Kopfschilds sind hier abgerundet, die Fliigeldecken haben
eine Zeichnung 4hnlich sulcicollis, die stark variiert. Lange 4-5,5 mm. Cape
Town, Stellenbosch, Mossel Bay, Keurbooms River Knysna, Patentie Humans-
dorp, Port Elizabeth, Algoa Bay, Dunbrody, Kowie (Port Alfred), Grahams-
town. Type in Halle.

8. NAnniscus Burmeister 1844

Diese Gattung steht zwischen Dizcranocnemus, von der sie die zweispitzigen
Vorderschienen hat, und Diaplochelus, mit welcher Gattung sie den para-
bolischen Kopfschild gemein hat. Der flache Halsschild hat rechtwinklige
Hinterecken und ist zunadchst schwacher, dann starker nach vorn verschméalert.
Das kleine Schildchen ist herzformig. Die verhaltnismaBig kurzen Fliigel-
decken bedecken die Seiten des Abdomen vom Seitenausschnitt an nicht.

Einzige Art:
N. pulicarius Burmeister 1844

Schwarz, Fliigeldecken hellgelb, Beine hellbraun. Oben von feinen fast
anliegenden kurzen weiBen Harchen bedeckt, die ein seidiges Aussehen verur-
sachen und den Grund nicht verdecken. Ich habe nur die beiden Typen
gesehen, deren Hintertarsen, bis auf die 2 basalen Glieder bei einem Bein,
abgebrochen sind. Diese 2 Glieder sind sehr lang, dreimal so lang als breit.
_ An der Nadel steckt eine kleine Skizze des Autors, die zeigt, daB die Hinter-
krallen doppelt, sehr ungleich und beide gespalten sind. Die Hinterschienen
sind sehr schrag abgeschnitten und tragen einen langen spitzen Sporn. Lange
2,5 mm. Ohne nahere Heimatangabe. 2 Typen in Halle.

g. BizAnus Péringuey 1902

Merkmale sind der parabolische, vorn von einem erhobenen Rand
umgebene Kopfschild, die nach hinten verschmalerten, das Abdomen nicht

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 255

. ganz bedeckenden Fliigeldecken und der Anhang am Grund der Kralle der
Mittelschienen des 4, der in rudimentarer Form auch beim @ vorhanden ist.
Bei Nanniscus und Diaplochelus fehlt dieser Anhang, letztere Gattung hat fast
parallelseitige Fliigeldecken. Die Hinterbeine der $3 sind robuster als die der
09. Genotyp caliginosus Pér.

Zu den Bizanus-Arten:
(1) B. caliginosus Péringuey 1902

Schwarz mit schokoladebraunen Fliigeldecken und beim ¢ roten Vorder-
beinen. Oben beim ¢ grau beschuppt, mit gelbbeschupptem Propygidium und
glattem, unbeschupptem Pygidium. 9 oben mit dunkler, Jangerer Behaarung,
auch auf dem Pygidium, die manchmal] grauweif wird. Lange 3,25-4 mm.
Obere Quellen des Olifants River, Tulbagh, Ceres, Touwsrivier. Typen in
Kapstadt.

(2) B. vansoni nova species

Schwarze Bizanus-Art mit hellbraunen Fliigeldecken ohne deutliche
Rippen, mit feiner staubartiger grauweiSer Beschuppung der Oberseite, gelber
Beschuppung des Pygidialteiles und dichter weiBer Beschuppung des Abdomen,
mit roten Vorderbeinen und tiefschwarzen Hinterbeinen.

Kopfschild parabolisch, vorn fein gerandet, fein gerunzelt, fast matt, fein
abstehend dunkel behaart. Halsschild so lang als breit, etwas oberhalb der
Mitte am breitesten, von dort nach vorn starker verengt als nach hinten, mit
stumpfen Hinterecken, hinten etwas zum Schildchen vorgezogen, wenig
gewolbt, mit Furche im hinteren Teil und je einem rillenartigen Eindruck
beiderseits der Mitte langs des Hinterrandes. Sehr fein punktiert, auf den
Seitenteilen staubartig grauweiB beschuppt, mit einigen abstehenden hellen
Haaren dazwischen, schwach glanzend. Das kleine schmale Schildchen dicht
gelbweiB beschuppt. Die hellbraunen Fliigeldecken sind nach hinten ver-
schmalert und bedecken die Seiten des Abdomen nicht. Sie sind staubférmig
weiBlich beschuppt, der Untergrund bleibt sichtbar. Das schmale Propygidium
und das Abdomen sind geschlossen wei®lich beschuppt, wahrend das Pygidium
staubartig gelb beschuppt ist und bei ihm der Grund sichtbar bleibt. Die Brust
ist diinn hell behaart. Die mittleren und hinteren Beine, deren tiefschwarze
Farbe stark gegen die roten Vorderbeine kontrastiert, sind verdickt, wenn auch
seitlich mafBig zusammengedriickt. Die Hinterschienen haben auf der oberen
Kante mehrere Sagezihne mit Dornen. Die Vorderkrallen sind doppelt und
ungleich, gespalten, die Mittelkralle ist einzeln und gespalten und tragt den
charakteristischen Anhang an der Basis, die Hinterkrallen sind einzeln und
einfach. Nur $4 liegen vor. Lange 4 mm. Vanrhyns Pass. Type in Pretoria.

10. DiaPLocHELus Burmeister 1844 |

Merkmale sind der parabolische Kopfschild und die das Abdomen ganz
bedeckenden Fliigeldecken. Genotyp longipes F.

2 56 ANNALS OF THE SOUTH AFRICAN MUSEUM

Zu den Diaplochelus-Arten:
(1) D. crassipes Burmeister 1844

GroéBte, allbekannte Art. Péringuey irrt, wenn er squamulatus Burm. damit
synonym gestellt hat. Burmeister schreibt in seinem Handbuch ganz richtig,
daB crassipes an den Hinterbeinen nur eine Kralle, squamulatus aber doppelte und
ungleiche Krallen hat, was ich an der Type nachpriifen konnte. Lange 8-10 mm.
Gifberg Vanrhynsdorp Distr., Malmesbury, Ceres, Cape Town, Franschhoek,
Caledon, Cogmans Kloof Montagu, Bechuanaland. Type in Halle.

(2) D. squamulatus Burmeister 1844 (=transvaalensis Péringuey 1902)

Es ist logisch, daB Péringuey nach Synonymstellung der Burmeisterschen
Art denselben Kafer neu beschrieben hat. Ein mildernder Umstand ftr ihn
ist, daB Burmeister in seiner Beschreibung die charakteristischen Sagezahne auf
der unteren Kante der Hinterschienen nicht erwahnt hat, sodaB Péringuey der
Meinung sein konnte, eine neue Art gefunden zu haben. Ich habe die Typen
vor mir gehabt, sie besitzen dieses Merkmal sehr deutlich. Alle Dzaplochelus-
Arten kommen mit gelbbraunen Fliigeldecken und Beinen, mit schwarzen
Decken und Beinen und in Ubergangen dazwischen vor. Obwohl es sich um
Farbformen handelt, hat man ibnen Namen gegeben, die Burmeister zitiert, so
hier castaneus Ecklon fiir Stiicke mit braunem Halsschild, maculicollis fur Stiicke
mit schwarzem, braungeflecktem Halsschild. Zwischen rufipes Ecklon und der
Nominatform finde ich keinen Unterschied. Man konnte éransvaalensis als
Bezeichnung einer ganz gelbbraunen Farbform beniitzen. Lange 5,5—7 mm.
Oudebosch Zonderend, Caledon, Transvaal Johannesburg. Typen in Halle.

(3) D. longipes Fabricius 1787

Auch bei dieser altbekannten Art gibt es Namen ftir abweichende Farb-
formen, die Burmeister zitiert (rufiventris Ecklon, pallidipennis Ecklon, luridipennis
Ecklon, obscurus Ecklon). Sie haben keinen systematischen Wert.

Lange 5,5-7 mm. Gifberg Vanrhynsdorp Distr., Clanwilliam, Malmes-
bury, Great Winterhoek, Tulbagh, Cape Town, Noordhoek, Schusters Kraal,
Cape Flats, Somerset West, Stellenbosch, Franschhoek, Caledon, Riversdale,
Knysna, Willowmore. Type im Museum Kopenhagen.

BESTIMMUNGSTABELLEN

GATTUNGSTABELLE DER HETEROCHELIDES

1 (22) Kopfschildvorderrand gerade abgestutzt, meist mit 2-4 aufgebogenen Zahnen.
2 (15) Spitze der Vorderschienen nicht gespalten, entweder parallel zu den Seitenzahnen
umgebogen oder schrag nach vorne-auBen zeigend.
3 (8) Schildchen klein (etwa 4 der Naht), herzformig.
Dichelus Serv. sensu lato (pp. 230, 257)

4 (5) Hinterbeine mit 2 gleichlangen Krallen.

Subg. Dichelus Serv. sensu stricto (pp. 231, 257)
Hinterbeine mit 2 in der Linge stark verschiedenen Krallen oder mit nur einer Kralle.

uw
-
>
~~

6 (7)
fo e(6)
8 (3)
9 (14)
10 (13)
Ir (12)
12 (11)
13 (10)
14 (9)

15 (2) 1
16 (17)
17 (16)
18 (19)

19 (18)
20 (21)

21 (20)

22 (1)
23 (24)

24 (23)

Lin 8)
2 (3)

(2)
(7)
(6)

Or He 09

6 (5)

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 257

Vorderschienen mit 3 Zahnen. : , , Subg. Heterochelus Burm. (p. 231)
Vorderschienen mit mehr als 3 Zahnen. ; Subg. Ischnochelus Burm. (p. 231)
Schildchen groB (mindestens } der Naht).
Schildchen dreieckig.
K6rperform langlich, mindestens doppelt so lang als breit.
Fliigeldecken deutlich nach hinten verschméalert. . Omocrates Burm. (pp. 240, 260)
Fliigeldecken parallelseitig. . : 3 Cylindrocrates nov. gen. (pp. 246, 263)
Korperform kiirzer, gedrungen. Goniaspidius Burm. (nec Péringuey) (pp. 247, 263)
Schildchen im basalen Teil parallelseitig, im apikalen Teil stumpfwinklig.
Rectoscutaria nov. gen. fiir Goniaspidius Pér. nec Burm. (pp. 248, 263)
ieedeechieaca mindestens beim ¢ zweispitzig.
Vorderschienen beim ¢ dadurch zweispitzig erscheinend, da der apikale Seitenzahn
dem Spitzenzahn genahert und parallel steht, wahrend er beim Q weiter absteht und
nicht parallel ist... 2 : : . Omocnemus nov. gen. (pp. 249, 264)
Vorderschienen bei 39 else
Schildchen gro wie bei den Omocrates-artigen.
Dicranocnemus subgen. Macrodicranocnemus nov. (pp. 250, 264)
Schildchen klein, herzformig.
Halsschild stark gewolbt, vorn und hinten steil abfallend.
‘ Dicranocnemus Burm. sensu stricto (pp. 252, 264)
Halsschild ands Pei ne . WNanniscus Burm. (pp. 254, 266)
Kopfschildvorderrand He gecrekor Bes eeoieene
Fliigeldecken nach hinten verschmalert, die Seiten des Abdomen nicht bedeckend,
Kralie der Mittelbeine an der Basis mit zahnartigem Anhang.
§ Bizanus Pér. (pp. 254, 266)
Fliigeldecken die Seiten des cde Sollepeend, Krallen ohne Anhang.
: : Diaplochelus Burm. (pp. 255, 266)

Dichelus Serville 1825
subgenus Dichelus Serv. sensu stricto
(= Trichidius Billberg 1820)
Bestimmungstabelle der §¢

Hinterschienen vom Knie bis zum Apex mehr oder weniger gleichmaBig verbreitert
und am Ende schrag abgeschnitten.
Trochanterdorn der Hinterschenkel sehr deutlich in eine lange freie Spitze ausgezogen.
Fligeldecken mit verkiirzten weiBen Schuppenbinden langs der Naht und auf der
Scheibe.
Schwarz, Fliigeldecken braun mit einer vorn verktirzten weiBen Schuppenbinde
neben der Naht und einer hinten verkiirzten solchen auf der Scheibe. Propygidium
ganz, untere Halfte des Pygidiums und Seiten der Bauchringe weiB beschuppt.
5 mm. Malmesbury, Paarl, Caledon.
1. simplicipes Burmeister 1844 (p. 232)
Trochanterdornen eae, zu svbennen. Fliigeldecken ohne Schuppenbinden.
Am Kopfschildvorderrand sind die 2 4uBeren Zahne groBer.
Krallen der Hinterbeine gleichlang, die mittleren Zahne des Kopfschildes noch
erkennbar, Hinterschienen mit Sporen, Pygidium mit 2 dunklen Flecken. 5 mm.
Malmesbury ? : 2. pallidipennis Blanchard 1850 (?) (p. 232)
Krallen der Pameoeneree maeieen lang, die mittleren Zahne des Kopfschildes erloschen,
Hinterschienen ohne Sporn.
Schwarz, Fiihler, Fliigeldecken und Beine rotbraun, Hinterschienen kurz und am
Ende fast so breit wie bei Jatzcollis. 5-5,5 mm. Cold Bokkeveld Ceres District,
Cape Town, Rondebosch. : ‘ 3. denticeps Wiedemann 1821 (p. 232)

258

7 > @

12 (27)

13 (18)
14 (15)

15 (14)
16 (17)

17 (16)

18 (13)

21 (24)
22 (23)

ANNALS OF THE SOUTH AFRICAN MUSEUM

Alle 4 Zahne des Kopfschildvorderrandes gleich.

Schwarz mit schwachem Seidenglanz, Fiihler, Taster und Vorderbeine zitronen-
gelb, die anderen Beine schwarzbraun, nur Schildchen und Seiten der Bauchringe
hell beschuppt, Pygidialteil unbeschuppt. 5 mm. ,,Kaffernland.”

‘ : : 4. flavimanus Burmeister 1855 (p. 232)

Hinterschienen nicht gleichmafig verbreitert.

Hinterschienen unten tiber der Mitte (also zwischen dem kleinen Zahn am Knie und

dem Mukro am Ende) mit zahnartigem Lappen.

Halsschild in der vorderen Halfte so breit wie in der hinteren, mit weitlaufig und

unregelmaBig verteilten groben Punkten, Hinterschienen kurz und am Ende mit einem

rechtwinklig nach innen abstehenden langen stumpfen Mukro.
Schwarz, manchmal mit griinem Schimmer, fast matt; Pygidialteil schwarz, ohne
Schuppen. Fihler, Taster und Vorderbeine hellrot. 5-7 mm. Cape Town,
Somerset West, Stellenbosch, Caledon, Heidelberg.
5. laticollis Burmeister 1844 (p. 232)

Halsschild nach vorn verschmdalert, gleichmaBig punktiert, Beine, besonders die

Hinterschienen langer und schlanker, ihr Mukro weniger stumpf und mehr nach unten

gerichtet, meist auch ktirzer.

Fliigeldecken ohne Schuppen oder mit schmalen Langsbinden aus diinn stehenden

hellen langlichen Schuppen.

Fligeldecken ohne Schuppen.

Fliigeldecken stark glanzend hell rotbraun, hintere Krallen ungleich.

Kopf, Halsschild und Abdomen schwarz, Fiihler (auch der Facher) und Beine
rotbraun, Halsschild abstehend dunkel behaart, der basale Teil der Mittelfurche
und Hinterrand des Halsschildes, das Schildchen, die Seiten des Abdomen und
das Propygidium gelblich beschuppt, Pygidium nur in der basalen Halfte mit
Ausnahme der Rander ebenso beschuppt. Am Ende des Pygidiums ein Kranz
langer weiBer Haare. Die langen Trochanterdornen etwas gebogen. 5,5 mm.
, sudafrika.”’ , 5 : 6. nitidissimus Burmeister 1844 (p. 233)

Fliigeldecken maBig alansend oder matt, hintere Krallen gleich.

Fliigeldecken maBig glanzend dunkel rotbraun oder dunkelbraun, Beine dunkelbraun

oder fast schwarz.

Genotyp. Schildchen, Pygidialteil und obere Enden der Bauchringe weibgelb
beschuppt. Untere Kante der Hinterschienen breit und im basalen Teil gerillt.
6-7,5 mm. Cold Bokkeveld Ceres District, Malmesbury Paardekop, Melkbosch
Strand, Cape Town Orange Kloof, Camps Bay, Kirstenbosch, Constantia Nek,
Clovelly Cape Peninsula, Fishhoek Cape Peninsula, Cape Flats, Banhoek Valley
Stellenbosch, Franschhoek, Worcester Cloete Pass, Robinson Pass.

Fhe dentibes Fabricius 1781 (p. 233)

Fliigeldecken matt hess eka paeceny Bene rotbraun, dem vorigen sonst ahnlich.
6-7 mm. Ceres, Sneeugat Valley, Tulbagh, ioe sh Noordhoek, Somerset West,
Stellenbosch, Caledon, Worcester. . . 8. expansus Pecinenee 1902 (p. 233)

Fliigeldecken mit langlichen weiBen, den Grund nicht verdeckenden Schuppen, die

in Langsbinden angeordnet sind.

Fliigeldecken nur mit einer solchen Binde, die langs der Naht angeordnet ist.
Fliigeldecken braun oder pechschwarz, Beine dunkel. 4,5-6,5 mm. Ceres,
Tulbagh, Cape Town, Somerset West, Stellenbosch, Banhoek Valley, 'Tradouw
Pass Swellendam District, Tee District.

g. acanthopus Burmeister 1844 (p. 234)

Fliigeldecken auBer der Naheoinde mit 1-2 oft rudimentaren weiteren Langsbinden.

Grofe Formen von 7—7,5 mm.

Fliigeldecken braun, Schuppen des Schildchens und des Pygidiums weiBlich.

Die Binden langs der Naht und auf der Scheibe der Fliigeldecken laufen in
breiten Rillen. Unterkante der Hinterschienen breit, nicht gerillt. Darling,
Malmesbury, Cape Town, Camps Bay, Tafelberg Blinkwater, Rondebosch,
Somerset West, Stellenbosch Banhoek Valley, Koegelberg in den Hottentots
Holland Bergen, ; , , ‘ 10, villosus Burmeister 1844 (p. 234)

23 (22)

24 (21)
25, (26)

26 (25)

27 (12)

28 (35)
29 (34)
30 (31)

31 (30)
32 (33)

33 (32)

34 (29)

35 (28)

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 259

Fliigeldecken tiefschwarz wie der ganze Kafer, Schuppen des Schildchens und des
Pygidiums orangegelb.
Sonst wie der vorige. Bushmanland Jakkalswater, Gifberg bei Vanrhynsdorp
Somerset West, Sicllenboseh. Worcester, George District.
5 : : 10a. villosus subspec. luteopygus nova subspecies (p. 234)
Kleine Formen unter 6 mm.
Fliigeldecken braun, halbmatt, Borstchen der Tarsenglieder pechschwarz.
Kleine Form des villosus, die Binden auBer der Nahtbinde oft ganz abgerieben.
5-5.5 mm. Ceres, Cape Town, Camps Bay, Stellenbosch, Zwartberg Pass Prince
Albert District. . . rob. villosus subsp. minor (Burm. nom. nud. 1844) n. ssp.
(=acanthoscelis Ecklon, nom. nud.?) (p. 234)
Fliigeldecken braun, glanzend, Borstchen der Glieder der Hintertarsen rotgelb.
Etwas kleiner als der vorige, Pygidialteil wei8gelb beschuppt, Vorder- und
Mittelbeine gelbbraun, Hinterbeine pechschwarz. 4-4,5 mm. Caledon.
; 11. lucidus Péringuey 1902 (p. 235)
Fliigeldecken mit dichten, aoa enand feeeiee cadena Schuppen in Langsbinden oder
auf der ganzen Flache.
Diese Schuppen bilden breite Langsstreifen.
Farbe der Schuppen hell gelbgrau.
Kopf und Halsschild rein schwarz behaart, Beine schwarz, Krallen der Hinterbeine
nicht ganz gleichlang.
Jede Fliigeldecke mit 3 Schuppenlangsbinden, die Raume zwischen ihnen schwach
glanzend, dicht mit kurzen angebogenen Borsten besetzt. Pygidialteil ganz
beschuppt. 8,7 mm. ,,Siidafrika.”’ ‘ 12. vittatus Burmeister 1844 (p. 235)
Kopf und Halsschild hell behaart.
Die Raume zwischen den 3 Langsbinden jeder Fliigeldecke unbeschuppt und mit
Borstchen in der Farbe des dunklen Grundes besetzt.
Kopf und Halsschild auch hell fein beschuppt, vorn diinner, nach hinten zu dichter.
Pygidialteil und Abdomen dicht gelbgrau beschuppt, Brust hell behaart. Beine
rotbraun oder dunkelbraun. 7—7,5 mm. Kap-provinz (Worcester ?), ,,Caffraria’’.
é 13. péringueyt nova species (—vittatus Péringuey, nec Burm.) (p. 235)
Die Raumne zwischen den 3 Langsbinden jeder Fliigeldecke mit aufstehenden hellen
stabchenformigen Schuppen ausgefiillt, die im Gegensatz zu den runden Schuppen der
Binden den Grund nicht ganz bedecken.
Abgesehen von der Beschuppung dem vorigen ahnlich, Beine dunkelbraun, mit
feinen hellen Harchen. 5-6 mm. Cold Bokkeveld Ceres.
‘ ; é 14. duplosquamosus nova species (p. 236)
Farbe der Schuppen dunkel orangegelb.
Kopf und Halsschild pechschwarz kurz behaart, letzterer nur langs des Hinter-
randes und im basalen Teil der Mittelfurche orangegelb beschuppt. Fliigeldecken
tief dunkelrotbraun, fast schwarz wie das Abdomen und die 2 vorderen Beinpaare;
die schuppenfreie Schulterbeule ist rotbraun, die Naht- und Randbinde, die hinten
zusammenlaufen, sind mehr als doppelt so breit als die angrenzenden Zwischen-
raume; die zwischen ihnen liegende Binde ist schmaler und hinten verkiirzt.
Die Zwischenraume tragen runde kleine glanzende Schuppen von der Farbe des
Untergrundes. Pygidialteil und Enden der Bauchringe dicht beschuppt, Brust
schwarz, Hinterbeine rotbraun. 5-5,5 mm. Stellenbosch Banhoek Valley,
Assegaibosch-La Motte bei Humansdorp.
Bei einigen Exemplaren von letzterem Fundort sind die Hinterbeine und der
Grund der Fliigeldecken ganz schwarz. . 15. pseudovittatus nova species (p. 236)
Die ganze Flache der Fliigeldecken gleichmaBig und gleichartig dicht beschuppt, nur
die Rippen mit feinen Borstenreihen.
Abgesehen von der Beschuppung dem duplosquamosus gleichend. Halsschild und
Kopf lang abstehend hell behaart, dazwischen ziemlich dicht beschuppt, sodaB
die dichte Beschuppung des Hinterrandes nicht absticht. Abdomen mit gr6éBerer
Schuppenflache und wie der Brustteil dicht weiBlich behaart. Beine pechschwarz,
Tarsen rotbraun. 6,5 mm. Malmesbury Paardekop.
16. holosquamosus nova species (p. 237)

260

36 (9)
37 (42)

38 (39)

39 (38)
40 (41)

41 (40)

42 (37)
43 (46)
44 (45)

45 (44)

46 (43)

ANNALS OF THE SOUTH AFRICAN MUSEUM

Hinterschienen unten tiber der Mitte ohne zahnartigen Lappen, stark gebogen, fast
in der ganzen Lange gleichbreit, ‘Trochanterdornen lang und spitzig.
Hinterschenkel kurz hinter der Trochanterspitze mit einem Zahnchen wie bei Mono-
chelus laetus Péringuey.
Fliigeldecken ohne weiBe Schuppenbinde. Pygidium nackt.
Fliigeldecken rotbraun, schokoladebraun oder schwarz, mit schwachem Seiden-
glanz, Schildchen, Propygidium, Ecken der Bauchringe und Seitenteile der Brust
weiB beschuppt, Beine wie die Fliigeldecken gefarbt. 4,5-5,5 mm.
Exemplare mit braunen Decken liegen vor aus Cape Town, Witsands (White
Sands), Somerset West, Stellenbosch Banhoek Valley, Houhcek, Oudebosch
R. Zonderend, Ashton, George District, Knysna, Langekloof.
Solche mit schwarzen Decken aus Gifberg bei Vanrhynsdorp, Sneeugat Valley
Tulbagh, Somerset West, Stellenbosch, Algoa Bay.
: 17. holosericeus Burmeister 1844 (p. 237)
Fliigeldecken mit alfa Sgihygnan lee esipiadlon.
Fliigeldecken mit je 2 solchen Binden, Pygidium nackt.
Die Nahtbinde ist vorn, die Scheibenbinde hinten verktirzt; sonst wie der vorige.
5mm. Mamre. Nur 2 3d. : : 18. albolineatus nov. spec. (p. 238)
Fliigeldecken mit je 3 diinnen weifen Rides, Pygidialteil hell beschuppt.
Etwas flacher als holosericeus, Schuppen und Haare gelb, Beine hellrot. 6 mm.
,sudafrika.”’? (Cold Bokkeveld?) : 19. platynotus Burmeister 1844 (p. 238)
Hinterschenkel nur mit Trochanterdorn, ohne Zahn daneben.
Fliigeldecken mit Langsbinden.
Nur eine Binde, die neben der Naht lauft.
Schwarz, die Binde diinn und vorn verkiirzt. Schildchen, Brustseiten und Enden
der pene weiB beschuppt, Haare dunkel, Beine schwarz. 6 mm. Cape
Town. : : : ; 20. luctuosus Péringuey 1902 (p. 238)
Jede Decke aa 2 Bigdem
Krallen ungleich. Schwarz, etwas glanzend, die 4uBere Binde vorn verkiirzt,
Schildchen, Enden der Bauchringe und der ganze Pygidialteil wei8 beschuppt,
Brustseiten wei behaart. Hinterbeine schwarz, die anderen rot. 5-6 mm.
Vanrhynsdorp. . i - « . @1. kocht novspeeyi(paess)
Fliigeldecken ohne ieestinden,
Schwarz mit Seidenglanz, Schildchen, Rand des Propygidiums, Brustseiten vor
und hinter der Schulter und Enden der Bauchringe wei8 beschuppt, Pygidium
nackt. Haare schwarzlich. Beine schwarz, die 4 vorderen mit braunem Schimmer.
6-6,5 mm. Willowmore, Uniondale Distrikt.
i 3 : 22. pseudoluctuosus nov. spec. (p. 239)
Nicht aufgefiihrt konnte werden. ; ; 23. zuluanus Péringuey 1902 (p. 240),
da nur @ bekannt. Glanzend schwarz, Schildchen, Rand des Propygidiums, Brust-
seiten und Ecken der Bauchringe mit gelblichen Schuppen. 6,75 mm. Eshowe in
Natal.

Omocrates Burmeister 1844

(=Goniaspidius Burmeister partim)

Grundfarbe bei allen Arten schwarz, Fliigeldecken braun, soweit nicht
anders bemerkt. Abdomen der 99 meist ganz oder im apikalen Teil rotbraun.
Genotyp axillaris Burm.

1 (6,18) Kopfschildvorderrand mit 4 Zahnen.

2 (5)
3 (4)

Vorderschienen mit 3 Zahnen.
Mittlere Zahne des Kopfschildes so groB und lan wie die auBeren.
Decken einfarbig gelbbraun. 5,25 mm. Namaqualand.
1, misellus Péringuey 1902 (p. 240)

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 261

(3) Mittlere Zahne des Kopfschildes kleiner und weniger scharf als die 4uBeren.
Decken blaBgelb, beim g mit angedunkeltem Rand. Pygidialteil graugelb
beschuppt, beim 2 behaart. 6-6,5 mm. Dikbome Merweville Koup, Klaar-
stroom Prince Albert. . : - , . 2. karrooanus nov. spec. (p. 240)

t (2) Vorderschienen mit 2 Zahnen.

Kopf, Schildchen und Halsschildvorderrand schwarz, Halsschild gelbrot, ohne
Furche, fast glatt und sehr lang. Fliigeldecken hellbraun, fein weiB behaart.
Pygidialteil dicht goldgelb beschuppt, Abdomen weiBlich beschuppt und behaart,
Beine gelbbraun, 4,5 mm. Hex River. . 3. pauxillus Péringuey 1902 (p. 241)
6G (1,18) Kopfschildvorderrand mit 3 Zahnen.
(8, 13) Vorderschienen mit 4 Zahnen.

Lang und schmal, Fliigeldecken blaBgelb, Pygidialteil bei $Q dicht goldgelb
beschuppt. Beine pechschwarz, Hinterschienen und Tarsen braun, alle Krallen
doppelt, der Zahn hinter dem Apikalzahn der Vorderschienen ist kiirzer als die
anderen Zahne. 4-5 mm. Bosluis Pass, Gamkas Poort.

: 4. andreaei nov. spec. (p. 241)
8 (7,13) Vorderschienen mit 3 Zahnen.
g (10) Basalhalfte des Halsschildes iiber dem Schildchen mit einem groBen querrechteckigen

goldgelben Schuppenfileck.
Fliigeldecken und Beine gelbbraun, Halsschild grau, an den Seiten langer weiB
behaart, Schildchen graugelb beschuppt, Fliigeldecken mit gelben, am Apex
weiBlichen Schuppen, Pygidialteil graugelb beschuppt. Abdomen und Unterseite
weiBlich behaart, wie auch die Hinterschienen. 5,5 mm. Burghersdorp.
: 5. plausibilis Péringuey 1902 (p. 242)
10 (g) Halsschild behaart, pee am eae ein Schuppensaum.
t1 (12) Propygidium und Pygidium des ¢ mit dichter dunkelbrauner Schuppenbinde im
basalen Teil, ihre Restflachen hellgelb beschuppt.
Fliigeldecken mit feinen anliegenden Bérstchen. 5-5,5 mm. Middelburg Div.
. 6. pygidialis nov. spec. (p. 242)
Farbform mit einfarbig weiBgelbem Pygidialteil.
6a. f. col. immaculatus nov. (p. 242)
12 (11) Pygidialteil des ¢ einfarbig he bacat
Fliigeldecken mit starkeren, mehr schuppenadhnlichen Bérstchen. Grund der
Decken pechschwarz oder braun. 3,5-4,5 mm. Port Elizabeth, Uitenhage,
Dunbrody. : : : 7. variabilis Burmeister 1844 (p. 243)
13 (7,8) Vorderschienen mit 2 Zabmen.
14 (17) Obere Kante der seitlich stark zusammengedriickten Hinterschienen des 3 mit
Ausbuchtungen.

15 (16) Diese Kante bildet 13 Wellenberge. Hintere Krallen einfach.

Schultern durch Eindruck daneben sehr betont. Fliigeldecken blaBbraun mit
schwarzem Rand. Pygidialteil des ¢ einfarbig goldgelb, des 2 mit 2 braunen
Flecken. 6,5—-7 mm. Stellenbosch, Koeberg, Malmesbury, Clanwilliam.

: é 8. axillaris Burmeister 1844 (p. 243)

16 (15) Diese Kante bildet nur Pace vor idler Apex einen aufrechten breiten, abgerundeten

Lappen. Hintere Kralle gespalten.
Dem vorigen ahnlich, Pygidium einfarbig weibgelb.
2 unbekannt. 4,5 mm. Algoa Bay. : g. lobipes Burmeister 1844 (p. 243)

17 (14) Hinterschienen des ¢ oben ohne Ausbuchtungen.

Den beiden vorigen ahnlich, Propygidium des ¢ viel breiter, der ganze Pygidialteil
dicht graugelb peti neil 5 mm. Kowie (Port Alfred), Algoa Bay.
10. modestus Péringuey 1902 (p. 243)

18 (1,6) Kopfschildvorderrand nur mit Ee Sahifaemisen Ecken, dazwischen kein Zahn.

19 (20, 33) Vorderschienen mit 4 Zahnen.

Sehr schmal gebaut. Schwarz, Fliigeldecken blaBgelb, hinten geschwarzt,
Pygidialteil und 2-3 folgende Bauchringe dicht orangegelb beschuppt, Beine
pechschwarz mit rotbraunen Hinterschienen und Tarsen. Der auf den apikalen
Zahn folgende Zahn ist nur halb so pes als die anderen Zahne. 5-6 mm.
Tankwa Karroo Waterval. ; ‘ II. hesset nov. spec. (p. 243)

262

ANNALS OF THE SOUTH AFRICAN MUSEUM

20 (19, 33) Vorderschienen mit 3 Zahnen.

21 (28)
22 (25)
23 (24)

24 (23)

25 (22)
26 (27)

27 (26)

28 (21)
29 (30)

30 (29)
31 (32)

32 (31)

Beide basale Zahne ungefahr gleichlang.
Auch Halsschild beschuppt.
Schuppen der Oberseite einschlieBlich Pygidialteil dunkelgelb, auf dem Halsschild
den Grund nicht ganz verdeckend.
Halsschild in den hinteren zwei Dritteln fast parallel, dann stark verschmalert,
Fliigeldecken, Abdomen und Beine am Grunde rotbraun. 4-4,5 mm. Cape
Town, Stellenbosch, Tradouw Pass Swellendam.
: 12. depressus Blanchard 1850 (p. 244)
Schuppen mehr grau, an dem fey etre’ graugelb, auf dem Halsschild den Grund
verdeckend.
Halsschild langer und gleichmaBig verschmalert. Sonst dem vorigen 4hnlich.
3,5-4 mm. Stellenbosch, Caledon. é 13. placens Péringuey 1902 (p. 244)
Halsschild nur behaart.
Pygidialteil und Abdomen gelb beschuppt, Behaarung der Fliigeldecken feiner, Beine
dunkel.
Halsschild grau behaart, die hellbraunen Fliigeldecken mit undeutlich gereihten
feinen weifen Schuppenhaaren, Hinterkrallen doppelt, gespalten. 4,5-6 mm.
Natal: Durban, Frere, Malvern, Park Rynie. Orange Freistaat: Smithfield.
14. spatulipennis Blanchard 1850 (=J/epidus Boheman 1857) (p. 244)
Pyzidialteil ur Abdomen wei beschuppt, Haare der Fliigeldecken kraftiger, Beine
rostrot.
Dem OE sehr ahnlich, Haare der Decken hinten dichter. 5 mm. ,,Ganz
Caffraria.” ; ‘ 15. lividipennis Boheman 1857 (p. 244)
Der basale Zahn der Vorderschienen viel kiirzer als der mittlere Zahn.
Fliigeldecken einfarbig, kleinere Art.
Halsschild grau behaart, Decken gelbbraun, Pygidialteil beim 3 gelb beschuppt,
beim @ gelb behaart. 4,5 mm. Hex River. . 16. humilis Péringuey 1902 (p. 244)
Fliigeldecken in der Gegend der Beulen angedunkelt, groBere Arten.
Halsschild ohne deutliche Langsfurche.
Halsschild weiter und gréber punktiert, dem luridipennis Burm. sehr ahnlich, aber
Pygidialteil des § einfarbig. 5 mm. Hex River.
17. placidus Péringuey 1902 (p. 244)
Halsschild mit deutlicher, ee Ne zum PNardereat durchlaufender Langsfurche.
Halsschild enger und feiner punktiert, dem vorigen ahnlich. 5,5 mm. Stellenbosch. ,
18. pseudoplacidus nov. spec. (p. 244)

33 (19, 20) Vorderschienen mit 2 Zahnen.

34 (37)
35 (36)

6 (35)

37 (34)
38 (43)
9 (42)

40 (41)

41 (40)

42 (39)

Fliigeldecken in der Gegend der Beulen angedunkelt.
Propygidium und Pygidium jeweils an der Basis bei §2 mit dunkelbraunen Flecken,
sonst gelb.
Sonst dem gr6Beren axillaris Burm. sehr ahnlich. 5 mm. Stellenbosch, Cape Town,
Koeberg, Malmesbury, Paleisheuwel. . 19. luridipennis Burmeister 1844 (p. 245)
Pygidialteil beim J einfarbig gelb, beim 2? nur Propygidium mit 2 rundendunklen Flecken.
Sonst dem vorigen ahnlich. 6-6,5 mm. Namaqualand.
20. mendax Péringuey 1902 (p. 245)
Fliigeldecken einfarbig.
Halsschild ohne oder nur mit schwacher und schmaler Langsfurche.
Kopfschild zwischen den scharfen aufgebogenen Ecken gerade. Halsschild schwach
gefurcht. Pygidialteil dicht gelb beschuppt.
Kopfschild starker verschmélert, Halsschild weniger gewolbt.
Fliigeldecken hellgelb, mit apikalem Schuppenrand, mit Langseindriicken neben
Naht und Schultern und auf der Scheibe, Hinterkrallen doppelt. 4 mm. Cape
Town, Stellenbosch, Tulbagh. ; 21. cylindricus Burmeister 1844 (p. 245)
Kopfschild weniger verschmalert, Halsschild starker gewolbt.
Sonst dem vorigen 4hnlich. 4 mm. Namaqualand.
; 22. namaquensis nov. spec. (p. 245)
Kopfschild zwischen den 2 Zahnen dtiscenandes, Halsschild nicht gefurcht, Pygidialteil
grau behaart.

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 263

Schmdler als die beiden verausgegangenen, Fliigeldecken rotbraun, auch Hals-
schild und Abdomen grau behaart, Beine schwarz, Hinterbeine mit braunem
Endteil. 5,25 mm. ,,Kap.” . Lapis: elongatus Blanchard 1850 (p. 246)
43 (38) Halsschild mit tiefer, durchlaufender eee die 4 seiner Breite einnimmt.
Breiter und kiirzer gebaut als cylindricus. Schildchen pechschwarz, Fliigeldecken
gelbbraun, Pygidialteil gelb beschuppt. Glanzend, fast glatt. Beine und Abdomen
rotbraun. 4mm. Hopefield. . . 24. canaliculatus Blanchard 1850 (p. 246)

Cylindrocrates nov. gen.

Einzige Art: parallelus nov. spec. 5,5 mm. Wallekraal, Namaqualand.
(p. 246.)

Goniaspidius Burmeister 1844 nec Péringuey 1902
Genotypus brevis Burm.

t (2) Hinterbeine mit einer Kralle.. Schildchen ein groBes gleichschenkliges Dreieck, dessen
Basis etwas kiirzer ist als die Seiten. Vorderschienen dreizahnig.
Schwarz, Fliigeldecken des g dunkelrotbraun mit schwarzem Saum oder ganz
schwarz, des 2 mit Abdomen und Beinen ganz rotbraun. Kopfschild mit 3 groBen
Zahnen. Uberall lang abstehend dunkel behaart. Tarsenglieder zylindrisch.
6,5 mm. ,,Siidafrika.”’ : brevis Burmeister 1844 (p. 247)
2 (1) Hinterbeine mit 2 Krallen. Schildchen ein groBes gleichseitiges Dreieck. WVorder-
schienen zweizahnig. Pygidialteil gelblich beschuppt.
3 (4) Halsschild maBig glanzend, ohne Langsfurche.
Schwarz, Fligeldecken dunkelrotbraun mit 3 Langsbinden aus anliegenden dicken
weiBen Harchen. 3 mm. Congo Belge, Upemba-Park.
lebist (Schein) 1958 (als Omocrates) (p. 248)
4 (3) Halsschild ganz matt, oa Cree ee
Schwarz, Fliigeldecken rotbraun mit gereihten, keine Binden bildenden weiBen
Boérstchen und feinem hellem Apikalsaum. 5 mm. Angola, Kuangu.
angolensis nov. spec. (p. 248)

Rectoscutaria nomen nov. fiir Goniaspidius Péringuey nec Burmeister
Genotypus péringueyi nom. nov.

1 (6) Der Apikalzahn der Vorderschienen zeigt schrag nach vornaufen, Fliigeldecken sehr
uneben.
2 (5) Fliigeldecken mit deutlichen Langsbinden aus anliegenden weiBlichen Schuppenhaaren.
3 (4) Die auBeren Binden sind vollstandig, die mittlere ist hinten verkiirzt.
Schwarz, Fliigeldecken schwarz oder braun oder schwarz mit braunem Fleck.
5-6 mm. Touwsrivier, Dikbome Merweville Koup, Klaarstroom Prince Albert Div.,
Lammerskraal Prince Albert Div., Willowmore.
péringueyi nomen nov. fiir brevis Péringuey nec Burmeister (p. 249)
4 (3) Alle aeion sind vorn und hinten verkiirzt und in einem nach vorn offenen Bogen quer
iiber die Fliigeldecken angeordnet.
Schwarz, meist mit dunkelrotbraunem Fleck im Zuge des Bindenbogens. 5-5,5 mm.

Nieuwoudtville, Willowmore. . . . éringueyi subspec. lunata nov. (p. 249)
5 (2) Fliigeldecken mit halbaufstehenden feinen hellen Haaren, die nicht zu Binden zusammen-
gefaBt sind.

Farbe und ica wie saceuiial 5-6 mm. Willowmore.
péringueyi subspec. uniformis nov. (p. 249)

264.

ANNALS OF THE SOUTH AFRICAN MUSEUM

6 (1) Der Apikalzahn der Vorderschienen ist starker nach den Seiten umgebogen, soda er
fast parallel zu den Seitenzahnen wird, Fliigeldecken nur mit flachem Eindruck unterhalb
des Schildchens.

Schwarz, Fliigeldecken dunkelrotbraun mit angedunkelten Seiten und Beulen.
5.5-7 mm. Touwsrivier. Willowmore. . . simplex Péringuey 1902 (p. 249)

Omocnemus nov. gen.

Einzige Art kocht nov. spec. 5-6 mm. GroB-Namaland, Aus. (p. 250.)

Dicranocnemus Burmeister 1844

Macrodicranocnemus nov. subgen. Subgenotyp andreaei n. sp.

1 (2) Halsschild hinten schwach gefurcht, Hinterbeine normal beborstet.

Schwarz, Fliigeldecken hellbraun, Halsschild lang aufstehend gelblich behaart,
Fliigeldecken mit unregelmaBig verteilten, nur hinter den Apikalbeulen kurze
Binden bildenden gelben Schuppen, Schildchen, Pygidialteil und Abdomen gelb
beschuppt und behaart, Beine braun, gelb beborstet. 6—7,5 mm. Seven Weeks
Poort Berg, Ladismith, Riversdale, Tradouw Pass Swellendam.

I. andreaei nov. spec. (p. 251)
Rasse dieser Art:

Fligeldecken dunkelrotbraun, an Schulter und Apex angedunkelt, Haare des
Halsschildes besonders im vorderen Teil dunkelbraun, Schuppen und Haare des
Pygidialteiles und Abdomens weifblich. 6—7,5 mm. Robinson Pass nordéstlich
Mossel Bay. : : Ia. andreaei subspec. zumpti nov. (p. 251)

2 (1) Halsschild im hinteren TESA tief perineli Hinterschienen mit Tarsen buschig schwarz
behaart.

L (24)
2 (15)

g(G)
4 (5)

5 (4)

Schwarz, Fliigeldecken braunschwarz mit gelben Schuppenflecken auf Scheibe und
Apex, Schildchen und Pygidialteil dicht gelb beschuppt. Beine braunschwarz.
5-6 mm. Leipoldtville, Elands Bay, Graafwater. . 2. hirtipes nov. spec. (p. 251)

Dicranocnemus Burmeister 1844 sensu stricto
Genotypus sulcicollis Wiedemann

GroBere Kralle der Mittelbeine ohne zahnartigen Anhang.
Vorderrand des Kopfschildes mit rechtwinkligen oder in zahnartige Zacken auf-
gebogenen Vorderecken.
Halsschild weniger gewolbt, besonders im basalen Teil, Langsfurche nur angedeutet.
Ecken zipfelformig aufgebogen, ihr Abfall nach innen reicht fast bis zur Mitte des
Vorderrandes. Schenkel ohne Trochanterdornen.
Schwarz, Halsschild hellgrau behaart, Fliigeldecken gelb beschuppt ohne Zeich-
nungsmuster, Pygidium und Abdomen gelb beschuppt. 5,25-5,5 mm. Natal,
Estcourt. Dunbrody C.P. ‘ : 3. natalensis Péringuey 1902 (p. 252)
Ecken steil aufsteigend, nach innen senkrecht abfallend, Hinterschenkel mit Trochan-
terdornen.
Schwarz, Fliigeldecken schwarzbraun, langs der Basis noch dunkler, mit schmaler
gelber Schuppenbinde langs der Naht um den Apex herum bis zur Schulter und
mit 4 quer angeordneten grauen Schuppenflecken. Propygidium und Abdomen
weiBgelb, Pygidium ebenso mit 2 groBen gelben Flecken. 2 dhnlich, aber statt
der Schuppen Haare. 4,5 mm. Willowmore, pee
: 4. spiniceps Péringuey 1904 (p. 252)

6 (3)
72 (8)

8 (7)
9 (14)
10 (11)

11 (10)
12 (13)

13 (12)

14 (9)

B52 (2)
16 (17)

17 (16)

18 (21)
19 (20)

20 (19)

a1 (18)
22 (23)

23 (22)

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 265

Halsschild besonders im basalen Teil starker gewolbt, hinten mit tiefer Langsfurche.
Kopfschild mit rechtwinkligen, nicht aufgebogenen, héchstens k6rnchenfo6rmigen Ecken.
Schwarz, Fliigeldecken schwarzbraun mit gelber Zeichnung, Halsschild am
Seiten- und Hinterrand und in der Furche gelb beschuppt. 9 mit reduzierter, aus
Harchen bestehender Zeichnung. 5-5,5 mm. Barrydale, George, Grahamstown,
Port Alfred: 4 : . 5. pulcher Péringuey 1902 (p. 252)

Kopfschildecken deutlich i Peshogee,

Fliigeldecken mit Zeichnung, Kopfschild starker verengt.

Fligeldecken auBer dem Nahtstreif mit mehreren gelben Langsstreifen.

Schwarz, Fliigeldecken braun mit gelber Zeichnung, Halsschild lang dicht aufrecht
gelb behaart und beim ¢ auf den hinteren Seitenteilen und in der Furche gelb
beschuppt. 4,5-6,5 mm. Cape Town, Salt River, Zeekoe Vlei, Cape Flats,
Stellenbosch, Hex River. 3 : 6. sulcicollis Wiedemann 1821 (p. 252)

Fliigeldecken auBer dem Nahtstreif mit nur einem gelben Langsstreifen.

Grundfarbe des ganzen Kafers tiefschwarz.

Sonst wie sulcicollis, doch ist die Farbe der Schuppenzeichnung weniger gelb und
mehr weiBlich. 5-5,5 mm. Upper sources Olifants River, Elandsbaai, Kamies-
kroon Namaqualand. : ; ; 6a. sulcicollis subspec. niger nov. (p. 252)

Grundfarbe der Fliigeldecken dunkelbraun.

Abgesehen vom Zeichnungsmuster dem sulcicollis sehr ahnlich. 5 mm. (ein
einzelnes Stick 4 mm.). Saldanha Bay.
6b. sulcicollis subspec. ftirscht nov. (p. 252)
Fligeldecken ohne jede 7 ighrae. fast nackt.
Kopfschild wenig verschmdlert, Ecken schwacher aufgebogen als bei der vorigen
Art. Schwarz, Fliigeldecken gelbbraun, Pygidium pechschwarz, ohne Schuppen,
Halsschild gelb behaart und Pygidium mit einigen Harchen, Beine braun. 4,5 mm.
Saldanha Bay. . . 7. nudus nov. spec. (p. 253)

Vorderrand des Tas Recaees mit shuamlar oil: neeruade en Ecken.

Halsschild des § fast ganz dicht und fein von kleinen hellen Schuppen bedeckt.
Schwarz, Fliigeldecken rotbraun, Fliigeldecken mit 2 Rippen, gelb, in den
Zwischenraumen hellgrau beschuppt, auch der Fleck an der Naht hellgrau. Beim
@ sind die Schuppen durch Harchen ersetzt. 4-4,5 mm. Klaarstroom Prince
Albert, Patentie Humansdorp, Uitenhage, Port Elizabeth, Algoa Bay, Dunbrody,
Resolution Albany, Grahamstown, Fort Beaufort.

8. mendicus Péringuey 1902 (p. 253)

Halsschild hochstens an den Seiten antl an der Basis mit Schuppensaum, sonst nur

behaart.

Fliigeldecken des ¢ mit deutlicher gelblicher Zeichnung.

Die Zeichnung besteht auBer der Nahtbinde aus mehreren Langsbinden und dem

Nahtfleck. Pygidium einfarbig gelb.

Dem sulcicollis sehr ahnlich. 4 mm. Villiersdorp, Hawston, Port Elizabeth.
g. hypocrita Péringuey 1902 (p. 253)

Die Zeichnung besteht AnBer dee Naiweinde und oe Nahtfleck aus nur einer Langs-

binde. Pygidium gelb mit dunklem querem Fleck an der Basis.

Dunkelbraun, Fliigeldecken und Beine rot, Halsschild mit Furche, Schuppensaum
und lohfarbener Behaarung, Schildchen wei beschuppt, Propygidium und
Pygidium dicht orangegelb beschuppt, ersteres mit dunklem Basalfleck. 2 nur
behaart. 4,5-5,5 mm. Uitenhage. . ._ 10. burchelli Arrow 1917 (p. 253)

Fliigeldecken ohne deutliche Zeichnung.

Halsschild ohne Schuppen, Pygidialteil mit 2 gelbbraunen Flecken.

Mir in natura unbekannt (ob Form des vorigen?). 4mm. Uitenhage, Port Eliza-
beth. ; : 11. arduus Péringuey 1904 (p. 254)

Halsschild an der Basis mit Sen tspehen ire Pygidialteil einfarbig.

Halsschild auch in der Furche und auf den hinteren Seitenteilen gelb beschuppt,
Schildchen heller beschuppt als die Fliigeldecken, diese einfarbig hellgelb
beschuppt, nur der Nahtfleck ist manchmal ein wenig heller. 4,5 mm. Knysna,
Patentie Humansdorp, Uitenhage, Grahamstown.

: 12. pulverulentus Burmeister 1844 {(p. 254)

266 ANNALS OF THE SOUTH AFRICAN MUSEUM

24 (1) Die groBere Kralle der Mittelbeine des J im Basalteil mit einem zahnartigen Anhang.
25 (26) Vorderecken des Kopfschildes deutlich, Fliigeldecken mit dichten haarahnlichen hellen
Schuppen ohne Zeichnung.
Halsschild mit Schuppen nur in der Furche und als schmaler Hinterrandsaum,
Fliigeldecken einfarbig hell beschuppt, nur der Nahtfleck ist manchmal ein wenig
heller. 4-5 mm. Port Elizabeth, Uitenhage, Dunbrody, Grahamstown.
: 13. squamulatus Burmeister 1844 (p. 254)
26 (25) Vorderecken des lepicliiees sige dey Fliigeldecken mit deutlicher gelber
Zeichnung.
Schuppenrand an der Basis des Halsschildes breiter, Furche weniger tief, die
Schuppenzeichnung der Fliigeldecken ist wie bei sulcicollis und etwas variabel,
manchmal mit zusatzlichen Schuppen, da die Raume zwischen den Langsbinden
fast ausgeftillt werden, manchmal reduziert mit breiten Zwischenraumen.
4-5,5 mm. Cape Town, Stellenbosch, Mossel Bay, Keurbooms River Knysna,
Patentie Humansdorp, Port Elizabeth, Algoa Bay, Dunbrody, Kowie (Port
Alfred), Grahamstown. . 5 : 14. squamosus Burmeister 1844 (p. 254)

Nanniscus Burmeister 1844

Emzige Art. =: : ; : “ : : 5 . pulicarius Burmeister 1844 (p. 254)

Offenbar selten, da ich nur die Type gesehen habe. Schwarz, Fliigel-
decken hellgelb, iiberall anliegend weif behaart. 2,5 mm. Nahere Heimat
unbekannt.

Bizanus Péringuey 1902

Vorderbeine rot, die anderen Beine schwarz, Krallen der Mittelbeine des
gd mit zahnartigem Anhang. Genotyp caliginosus Pér.

1 (2) Fliigeldecken schokoladebraun, 3 mit flaumartigen grauen Schuppen, Propygidium und
Abdomen gelb beschuppt, Pygidium nicht beschuppt, pechschwarz, 2 oben und am
Pygidialteil lang behaart. 3,25-4 mm. Ceres Obere Quellen des Olifants River, Tulbagh,
Touwsrivier. ; : caliginosus Péringuey 1902 (p. 255)

2 (1) Fliigeldecken fe fibrenem a emmiaieen weiblichen Harchen, Propygidium und
Abdomen dicht wei beschuppt, Pygidium weniger dicht (Grund bleibt noch etwas
sichtbar) gelb beschuppt. 3,5 mm. Vanrhyns Pass. : vansoni nov. spec. (p. 255)

Diaplochelus Burmeister 1844
Genotyp longipes F.

D. squamulatus Burmeister ist nicht synonym zu crassipes Burm., wie
Péringuey angenommen hat, sondern hat 2 Krallen an den Hinterbeinen.
D. transvaalensis Péringuey ist synonym mit squamulatus Burm.

Alle Arten sind schwarz mit gelbbraunen Fliigeldecken und rotbraunen
Beinen und kommen gelegentlich ganz schwarz und in Ubergangen zu dieser
Form vor.

1 (2) Hinterbeine mit einer Kralle, groBe Art. 8-10 mm. Gifberg Vanrhynsdorp Distr.,
Malmesbury, Ceres, oe Town, Franschhoek, Caledon, Cogman’s Kloof Montagu,
Bechuanaland. : ; : 1. crassipes Burmeister 1844 (p. 256)

BESTIMMUNGSTABELLEN DER HETEROCHELIDES 26 ap

2 (1) Hinterbeine mit 2 Krallen.
' 3 (4) Untere Kante der Hinterschienen des g mit mehreren Sagezahnen. 5,5~7 mm. Oude-
bosch Zonderend, Caledon. Transvaal, Johannesburg.
2. squamulatus Burmeister 1844 (p. 256) (=¢ransvaalensis Péringuey 1902)
4 (3) Untere Kante der Hinterschienen des g glatt. 5,5-7 mm. Gifberg Vanrhynsdorp Distr.,
Clanwilliam, Great Winterhoek Tulbagh, Malmesbury, Cape Town, Noord Hoek,
Schuster’s Kraal, Cape Flats, Somerset West, Stellenbosch, Franschhoek, Caledon,
Riversdale, Knysna, Willowmore. : ¥ , 3. longipes Fabricius 1787 (p. 256)

>’ a ye) ae
P= x te ee ee 1
c a oe A d
i the
- + 4 LF
" ‘ vii a
: 4 Vig “%
: = 1)
, Pes if

hee a vite
ade ae Hes:
Wie ‘ ely
hi au pbb eokesic

\ f A plies at
Lae, ae Ge Ayr , a al A a 5)

THE GENITALIA OF THE GENUS PHASIS AND ALLIED GENERA,
(LEPID. LYCAENIDAE)

DESMOND MurRRAY

(With 2 plates)

CONTENTS
Introduction .. A on af nS PAU we, 6s)
Genera and species dealt with ia oy Bf neat
The two sections based on the genitalia ee ee 7 2
Food-plants .. re me af He s mehael Ji
References Chen a ete aye ee ie SE L2T7
INTRODUCTION

Aurivillius (in Seitz) divides the Phasis group into two sections or three
groups of species and sub-ordinate genera. There are approximately 46 species
and subspecies. Below is given the classification as given in Seitz.

1. HW., two small tails, one at end of 1b and another at end of 2. FW 12 veins, vein 7 termi-
nating into margin. Phasis Hubner.
2. HW.., at end of 1, 1b angular or tailed, but always without tail at end of 2.

(a) FW., above without black discal band or separate black discal spots— always 12 veins,
vein 7 terminating into margin. Palpi below coarsely scaled, without bristly hairs.
Aloeides Hubner.

(6) (i) F.W., and often also HW above with black discal spots on light (orange) ground,
or F.W above with coherent black discal band being only united with black marginal
band in area tb and at costal margin. Poecilmitis Butler.

(ui) FW., 11 veins Chrysoritis Butler.
N.B.—P. zeuxo and chrysantas also have only 11 veins.

Although the above can be taken as the classification now recognized,
the genitalia have not so far been studied.

The principle laid down by Stempffer that ‘today all authors should
publish descriptions and figures of the genital armature of the species they
study’, is certainly necessary with the Lycaenidae and more especially with
the Phasis group, because in most cases the valves are alike. All the species
available have been so studied and lead one to a somewhat different division
of the group. Besides the venation and other points noted in the above extract,
the valves are found to be very similar in most cases; the anellus and juxta
though distinct in some are not alone sufficient to separate the subordinate
genera or the species.

269

270 ANNALS OF THE SOUTH AFRICAN MUSEUM

A close study of the genitalia however shows that there is a natural
division into two sections when the “‘falces’ (the two hooks of the gnathos) are
compared. Some are ‘straight’, others are ‘elbowed’, dividing the whole group
into about equal numbers. This difference is therefore used here to separate
the numerous species into two sections. There can be little doubt that those
included in each section show close affinity to each other. Apart from wing-
colouring and size there seems to be no other character to distinguish them;
the form of the aedoeagus is also helpful in distinguishing the species.

There are often many varieties among the species themselves, so that it is
very difficult to separate them, especially with pierus, thyra, taikosama and others.
Some have been named as osbecki from thysbe, but the former is now known to
be distinct from the other. These differences in wing-colouring are not con-
stant. Take for example the three species pierus, thyra and taikosama, they can
seldom be separated for certain by wing-colouring and sometimes appear at
different times or together. Riley (1938) says in the case of thyra that ‘exami-
nation of the genitalia in seven widely different males (Gape Town) provided
no character whatever on which to separate the various varieties, all seem
exactly alike’. In another example he says ‘a cursory (dry) examination of the
genitalia confirms this finding’. But such an examination is hardly sufficient
with such critical material, the very minutest comparison has to be made.
That used by Stempffer, in all his papers, seems to be by far the best. By this,
or what is called the ‘open’ method, the uncus holding the falces is separated
from the tegumen, the valves and aedoeagus are also separated, so that each
part can be studied separately and compared with the same in other species.

Polymorphism, E. B. Ford (Butterflies (New Nature Series)) says, ‘is the
occurrence together, in the same habitat of two or more distinct forms of a
species, in such proportions that the rarest of them cannot be maintained by
recurrent mutation’. This may help us to understand something about
varieties in a species.

The genus Phasis is almost exclusively confined to southern Africa, only a
few species being found farther north. They are generally small and brilliant
coloured butterflies and in the sunshine shine like jewels.

Perhaps it will be best first to enumerate the species into the present
recognized groups and then give short descriptions of each from a genitalic
point of view.

The writer has to acknowledge the very generous help and encouragement
of the following, without whose assistance this paper could not have been put
together:

Messrs. C. G. C. Dickson, K. M. Pennington, Gowan C. Clark,
D. A. Swanepoel and especially Dr. G. van Son of the Transvaal Museum,
Pretoria, who very kindly went to some trouble (with a former paper on the
same subject) pointing out the slips and errors as well as the difficulties to be
faced by anyone who would venture to tackle this difficult group of South
African insects.

THE GENITALIA OF THE GENUS PHASIS AND ALLIED GENERA 271

That was some years ago when two papers were appearing on the genitalia

of other groups of South African Lycaenidae (Murray, 1944, 1948). Since then
a great deal of new material has come forward and the Phasis group has not
been undertaken by anyone. It is to be hoped that this contribution, though
by no means complete, or without errors, will help to bring the study nearer
completion. There is a lot more still to be learnt, especially from the study of
the early stages, from which alone we can be certain of distinction with critical
species. Mr. C. G. Dickson and Mr. Gowan Clark have given us some valuable
papers on this section.

At the end of this paper is given a list of food-plants, compiled for the
most part from papers noticed in the references. This information is very
valuable both to collectors and to breeders from whom (especially the latter)
further knowledge is still to be gained from the study of these beautiful insects.

GENERA AND SPECIES DEALT WITH

Gen. Phasis Gen. Poecilmitis
thero (genotype) lycegenes (genotype)
clavum (subspecies) lyncurtum
sardonyx tstno (subspecies nov.)
arg yraspis chrysantas
wallengrenit aethon
malagrida lycia
dicksoni chrysaor

phosphor

Gen. Aloeides - pyroeis
pierus (genotype) felthami
egerides ZeUuxoO ZeUxO
pees Zeuxo Zonarius
simplex penningtoni
damarensts turnert
molomo aridus
almeida palmus
thyra thysbe
pallida osbecki
dentatis nigricans
thyra maseruna (subspecies) trement (subspecies)
taikosama brookst
orthus pelion
barklyi pyramus

Chrysorites oreas
Crudaria leroma

272 ANNALS OF THE SOUTH AFRICAN MUSEUM

Several other species are awaiting description, including Poecilmitis tsino.
Among the above Aloeides pallida was not available for dissection and among
the Poecilmitis, turnert and brooksi, including also the subspecies.

THe Two SECTIONS BASED ON THE GENITALIA

Falces ‘straight’

Phasis thero and the five other species
named under Phasis. Aloeides pierus
and all named under this genus, with
the exception of orthus which has the

Falces ‘elbowed’

Poecilmitis lycegenes and ali other
species named under this genus, as
well as A. orthus, excluding in both
columns species not available for

falces ‘elbowed’. examination and the _ subspecies

named.

Taken all together the total amounts to 46 species including the sub-
species. It should be noted that orthus is distinctly misplaced among the puerus-
thyra group, as it has the falces ‘elbowed’, i.e. from the genitalic view.

The figures of the genitalia have been made by the ‘open’ method and
more or less uniform in size for ready comparison and with the aid of the
camera lucida, to ensure accuracy.

There follow short descriptions of the genitalia. The data of all the
species examined are given below.

SECTION I (FALCEs ‘STRAIGHT’)
I. Phasis, Butler (1869) U = uncus; V = valve; A = aedoeagus.

1. P. thero L. Type. (Pl. III, 1.) The genitalia of this and the next three
species approximate to Capys alphaeus Cram. U. broad, well developed. V. large
bottle-shaped, apex a straight serrated edge. Anellus large. A. broad, saccus
produced into a hooked point. Cape Flats. g/30.

P. clavum, Murr. (subspecies). The genitalia are almost the same as thero
but the wing expanse smaller; forewing more rounded and the marking
differing considerably; it has one tail only, not two as in thero. Taken at
Nieuwoudtville, Vanrhynsdorp Dist., Cape, by Dr. G. van Son, 8/27. Since
then it has been found at several places around Boskloof near De Wet in the
Worcester Dist. by Mr. D. A. Swanepoel and others, all over the Karoo.

2. P. sardonyx Tr. (Pl. III, 2.) V. shaped somewhat like a boomerang,
inner edge produced into a fine blunt point. Anellus large, divides into two
long arms. A. broad with pointed apex holding cornuti; saccus produced.
Cape Prov. 1941 (Gowan C. Clark).

3. P. argyraspis Tr. (Pl. III, 3.) Genitalia similar to the last except that the
apex of the valve is convex instead of concave ending in a blunt knob instead

THE GENITALIA OF THE GENUS PHASIS AND ALLIED GENERA 273

_ of a point; anellus two large pointed processes. Graaff-Reinet, Cape. 3/40
(Swanepoel).

4. P. wallengreni Tr. (Pl. III, 4.) U. broad with rounded pads. V. pointed.
A. very large and broad, divided at apex. Anellus small, shaped like a triangle.
Mamre Dist., Cape. 3/37 (Dickson).

5. P. malagrida Walleng. (Pl. III, 5). U. broad. V. very similar in shape
to pierus but smaller Anellus large triangular-shaped. A. large and broad.
Lion’s Head 2/35 (Dickson).

6. P. dicksont Gabr. (Pl. III, 6.) U. broad. V. large and broad. Anellus
small, triangular-shaped. A. large, ending in a fine point. The male of this
species was first found by Mr. C. G. Dickson near Melkbosstrand in 9/46; the
female the following year 8/47.

II. Aloeides Hubner (1816).

7. A. egerides Riley. (Pl. III, 7.) U. narrow. V. narrow and pointed with a
distinct indentation about middle. Anellus small. A. long, the apex bulbous.
Philadelphia, Cape. 3/36 (Dickson). .

gaa. prerus Cram. (Pl. III, 7a.) U. a narrow, hairy pad. V. large,
restricted at apex, broad about middle, rounded off at base. Anellus a large
broad triangular process. A. large, tapering to a blunt point. Wynberg, Cape.
11/36. Stellenbosch, Cape. 10/36.

8. A. aranda Walleng. (Pl. III, 8.) U. narrow. V. same as malagrida.
Anellus a small triangle. A. large, broad. Natal, 1941.

g. A. simplex Tr. (Pl. III, 9.) U. narrow. V. large, ending in a blunt point.
Anellus large, apex two short points. A. large with broad apex. Tsaborg,
Bechuanaland. 12/55 (Pennington). |

10. A. damarensis Tr. (Pl. III, 10). Genitalia similar to molomo. A. broad
with blunt apex. Anellus smaller. Zululand and Natal coast Dist.

mee. wolome Fr. (Pl. Til, "113) U: broad: V: broad, bottle-shaped.
A. tapering to a point at apex. Anellus a large process, apex and base pointed.
Johannesburg, Transvaal. 10/02 (Feltham Coll., Johannesburg University).

12. A. almeida Feld. (Pl. III, 12.) U. narrow. V. long and narrow, ending
in a blunt point. Anellus a large triangular process. A. large and broad with
a blunt apex. Weenen, Natal. 11/31.

13. A. thyra L. (PI. III, 13.) U. narrow. V. bottle-shaped like pierus, but
smaller apex a blunt point. Anellus a large triangular process. A. large and
broad, apex blunt. Cape and Colenso, Natal. 1928.

14. A. taikosama Walleng. (Pl. ITI, 14.) U. narrow. V. and A. very similar
to those of thyra. Anellus smaller. Brakpan, Transvaal. 10/30; Springs, Trans-
vaal. 2/37.

274 ANNALS OF THE SOUTH AFRICAN MUSEUM

15. A. barkly: Tr. (Pl. IIT, 15.) U. narrow. V. differs considerably from
the other species, a long narrow arm with a rounded apex, there is also a
broad point attachment about the middle. Anellus has apex produced into
two fine points. A. moderate in size, broad with several cornuti at apex.
Namaqualand, Aughrabies. 9/49 (Swanepoel).

Species not available for dissection, pallida.

SECTION II (Fatces “‘ELBOWED’)

III. Poecilmitis Butler (1899).

1. P. lycegenes Tr. (genotype). (Pl. III, 16.) U. broad with two prominent
hairy pads. V. broad at centre and base narrowing to a blunt point at apex.
Anellus small, triangular-shaped at top but broadening at base. A. long,
broad, ending in a fine point. Natal, 1941 (G. Clark).

2. P. lyncurium Tr. (Pl. III, 17.) Very similar to last species but V. smaller
and not so broad. A. more narrow. Mbulu, Transkei, Cape. 1/33 (Pennington).

It is unnecessary to describe all the species given under this section as
they are very similar to each other, though differing slightly, as will be seen
from the drawings.

3. P. chrysantas Tr. (Pl. III, 18.) Naauwpoort, Cape, 12/31 (Gowan C.
Clark).

4. P. aethon Tr. (Pl. III, 19.) Pilgrim’s Rest, N. Transvaal. 12/45 (Swane-
poel).

5. P. lycia Riley. (Pl. III, 20.) Matjesfontein, Cape. 3/52 (Swanepoel).

6. P. chrysaor Tr. (Pl. III, 21.) Riebeek Kasteel Mts., Cape. 4/37 (Dickson).

7. P. phosphor Tr. (Pl. IV, 22.) Newstead, Balgowan, Natal. 5/34 (Pen-
nington).

S. P. pyrocs Tr, {Pl 1V..29)) Cape, 11/a6qavicksen):

g. P. feltham Tr. (Pl. 1V, 24.) Cape Flats, Cape. 11/37.

10. P. zeuxo zeuxo L. (Pl. IV, 25.) Cape Pen. 2/27; Oakford, Natal. 2/38.

11. P. penningtoni Riley. (Pl. 1V, 26.) Gaikaskop, Cape. 1/35 (Pennington).

12. P. aridus Penn. (Pl. IV, 27.) Springbok, Cape. 10/55 (Pennington).

13. P. palmus Cram. (Pl. IV, 28.) Tygerberg Hills, Gape. 11/45 (Dickson).

14. P. thysbe L. (Pl. IV, 29.) Cape Dist. 4/36 (Dickson).

15. Aloeides orthus 'Tr. (Pl. IV, 30.) Mountain Drive, Second Pass, Basuto-
land. 1/57 (Pennington).

16. P. pyramus Penn. (Pl. IV, 31.) Zwartberg Pass, Cape. 11/46 (Pen-
nington).

17. P. tsino. (Pl. IV, 32.) Mbulu Ridge, Cape. 11/54 (Swanepoel). Not
yet described but very near to P. lyncurium.
P. lurnert was not available for examination.

THE GENITALIA OF THE-GENUS PHASIS AND ALLIED GENERA 295

IV. Chrysoritis Butler (1898).

18. C. oreas Tr. (Pl. IV, 33.) This species was rediscovered by Mr. M.
Pennington after a lapse of some 70 years or more. Niginya, Natal. 12/27
(Pennington).

V. Crudaria Wallengren. Recognized as a distinct genus by Aurivillius.

19. C. leroma Walleng. (Pl. IV, 34.) The genitalia of this species differ
from those of all the Phaszs and so it has been placed in a genus of its own. De
Wildt, Pretoria, Transvaal. 12/32.

VI. Anthene Doubleday (1847).

This genus is included here in order to complete the study of all the
South African Lycaenids; the other groups have been dealt with previously
(Murray, 1944, 1948, 1956).

The genus was created by Doubleday in 1847, afterward placed under
_ Lycaenesthes Moore in 1865, the genotype being A. larydas Cram. It is a very
large genus but there are only a few representatives in South Africa. Among
other characiers it is distinguished by the two or three slender pencils of hair
which nearly always occur on the hind-wings at the extremities of veins 1, 2
and 3. The genitalia are very distinct, the aedoeagus always long and well
developed. Bethune-Baker revised the whole African genus in 1910, including
most of our species, though the figures he gives of the genitalia, from photo-
graphs, are not very helpful in distinguishing the various species. Very few
life histories, including the food-plants are so far known. The descriptions
given below are very brief.

1. Anthene larydas Cram. (genotype). (Pl. IV, 35.) U. rounded into two
large hairy pads. V. of moderate size, bottle-shaped, apex blunt. A. very long
and broad. Anellus large, forming a three-pronged process; saccus produced.
Durban, Natal. 6/31.

2. A. amarah Guer. (Pl. IV, 36.) U. pads small. V. oblong in shape,
inner apex pointed, outer rounded. A. very long, blunt at orifice. Anellus a
crossed process, resembling a wide X. Brakpan, Transvaal. 5/51; Naboom-
spruit, N. Transvaal. 4/32.

3. A. lida Butl. (Pl. 1V, 37.) V. of moderate size, broad, divided at apex
into two pointed teeth. A. long, blunt at orifice. Anellus small. Pretoria,
Transvaal. 9/45.

4. A. definiia Butl. (Pl. IV, 38.) V. very large and broad, divided at apex
into fine pointed teeth. A. long and narrow. Anellus small; saccus produced.
Warmbaths, Transvaal. 4/32; Pretoria, Transvaal. 10/37.

5. A. otacilia Tr. (Pl. IV, 39.) V. round at base with a very long bent
arm, ending in a fine point, holding a blunt knob near middle with two fine
spines. A. long and narrow. Anellus small. Natal, 1941.

276 ANNALS OF THE SOUTH AFRICAN MUSEUM

6. A. princeps Butl. (neglecta Tr.) (Pl. IV, 40.) V. very large and broad,
almost as large as the whole organ, apex divided into a series of strong teeth;
about middle a long, pointed arm, also with teeth. A. long and narrow,
bulbous at apex. Anellus very small; saccus produced. Howick, Natal. 3/34
(Pennington).

7. A. lemnos Hewit. (Pl. IV, 41.) V. very large, almost round in shape,
apex minutely divided. A. very long, broad, ending in a blunt point. Saccus
ending in a long fine point. Anellus two curved arms. Durban, Natal. 12/34.

8. A. millari Tr. (Pl. IV, 42.) V. of moderate size, heart-shaped, upper
edge serrate. A. very long and narrow. Anellus small. Weenen, Natal (Tring
Museum).

g. A. minima Tr. (Pl. IV, 43.) V. divided at middle, where the inner edge
is produced to a fine sharp point; upper half of V. a long, curved serrate arm.
A. very long, ending in a fine point. Anellus pear-shaped, divided into two
points. Saccus produced to a point. Northdene, Natal. 1/33 (H. Millar).

10. A. talboti Stempff. (1936). (Pl. IV, 44, redrawn from Stempffer’s
figure.) This species was found some years ago at Chuckomaas, Natal, by
Mr. L. S. Higgins; it also occurs in Kenya. The genitalia are similar to definita
but the V. is divided into distinct long, pointed spears. Anelius very large’
and broad.

11. A. contrata taken by Pennington in 4/53. Natal?

VII. Desmolycaena Trimen. (1898).

Desmolycaena Mazoensis Tr. (Pl. IV, 45.) Uncus broad. Falces small.
Valve oblong in shape. Aedoeagus broad of moderate size, blunt at apex.
N. Transvaal, 1947 (Swanepoel).

One or two other species have been found in recent years but they are not
available.

SUMMARY

The genitalia of the genus Phasis (Lepid. Lycaenidae) and its subordinate
genera have been worked out, accompanied by figures of all the species avail-
able, as these have not so far been given by any other worker. The paper
suggests a natural division of the approximate number of 46 species into two
sections of equal numbers by means of the different shaped ‘falces’ or hooks of
the gnathos. There is added the small number of South African species of the
genus Anthene to complete the whole study of the South African Lycaenidae;
others being dealt with elsewhere. At the end a list of food-plants and necessary
references are added.

THE

Phasis —
I. zeuxo zeuxo Chrysanthemoides
moniliferum
C. incana
2. lycia Royena hirsuta and
other plants
3. chrysaor Zygophyllum sessi-

lifolium and Rhus sp.

Z. sessilifolium
Z. flexuosum
Aspalathus sp.
Ifloga laricifolia

4. felthami

5. malagrida

Senecio pubigerus
Danthonia stricta
Leucadendron sp.

6. wallengrenii
7. dicksoni

Pelargonium flavum.

Felicia muricata
Aspalathus sp.

8. taikosama

g. turneri Zygophyllum sp.
10. thysbe Aspalathus sp.
Chrysanthemoides
II. nigricans Senecio sp.

12. palmus Aspalathus sarcantha
Berzelia sp.
13. thero Melianthus major

Rhus sp.

GENITALIA OF THE GENUS

PHASIS

Foop-PLANTS

Compositae

Ebenaceae
Zygophyllaceae
Zygophyllaceae

Leguminosae
Compositae

Compositae
Gramineae
Proteaceae
Geraniaceae
Compositae
Leguminosae

Zygophyllaceae

Leguminosae
Compositae

Compositae
Leguminosae
Bruniaceae
Melianthaceae

AND ALLIED GENERA

7);

Cape (Dickson, 1953). Eggs gene-
rally laid singly on fresh or
withered leaves, also on stem and
bark.

Cape (Swanepoel, 1953).

Cape (Dickson, 1943). Eggs laid
singly on underside of leaf.
Cape (Dickson, 1940).

Cape (Dickson, 1940).

Cape (Dickson, 1940). Eggs gene-
rally laid singly, sometimes in
pairs on stem.

Cape (Dickson, 1953).

(Cape (Dickson, 1953).

Cape (Dickson, 1953).

Transvaal (Murray, 1937). Reared
on by G. C. Clark.

Western Karoo (Dickson, 1953).
Cape (Dickson). Reared on this
plant.

Eggs laid singly on leaf and stem.
(Dickson).

Cape (Dickson, 1947).

Cape (Dickson, 1953).

Cape (Murray, 1939).

The following species have been reared on Aspalathus sp. by Mr. G. C. Clark: pierus, aranda
thyra, almeida and damarensis, and Crudaria leroma on Acacia Karroo.

Anthene—

1. definita Albizzia lophanta
Acacia, Rhus and
other plants

2. lemnos

3. minima See Refer. (Clark,
1940)

4. livida Kalanchoe crenata

Desmolycaena Acacia sp.
mazoensis

Leguminosae

Micrococca berberidea Euphorbiaceae

Crassulaceae

Leguminosae

REFERENCES

Cape (Penfold). Natal (Clark) on
flower buds and fruit.

Natal (Dickson, 1954).
Natal (Clark, 1940).

Kenya (Jackson, 1937).
Zululand (Pennington, 1953 and
1956).

Bethune-Baker, G. T. (1910). Revision of the African species of Lycaenesthes-group of the

Lycaenidae.

Trans. roy. ent. Soc., Lond., p. 1.

Clark, Gowan C. 1940. On the life-histories of some South African Lepidoptera. 7. ent. Soc.,

S. Afr., 5 3, 42.

1951. The life history of Crudaria leroma. F. ent. Soc., S. Afr., 14, 127.
1952. Some peculiarities in the early stages of some Butterflies. 7. ent. Soc., S. Afr., 15, 96.

Clark, Gowan C, and Dickson, C. G. C. 1952. The life history of Phasis thysbe in Some South

African Butterflies.

278 ANNALS OF THE SOUTH AERICAN MUSEUM

Dickson, C. G. C. 1940. Notes on the early stages of Phasis felthami Trim., a Lycaenid Butterfly
from the Cape Peninsula, and a list of some recently determined food-plants of some other
South African Butterflies. Ann. S. Afr. Mus., 32, 545.

1943. The life history of Phasis chrysaor (Trim.). 7. ent. Soc., S. Afr., 6, 37.

1946. The life history of Phasis thysbe L. var. nigricans Aur. (Lepidoptera: Lycaenidae).
fj. ent. Soc., S. Afr., 9, 178.

1947. Recently observed food-plants of some Cape Lepidopterous larvae. 7. ent. Soc.,
S. Afr., 10. 126.

1952. The life history of Phasis zeuxo zeuxo L. Trans. roy. Soc., S. Aff., 33, 447-

1953. Recently observed food-plants of some Cape Lepidopterous larvae (4th series).
J. ent. Soc., S. Afr., 16, 73. (Other series in vols. 7 (1944), 8 (1945) and xo (1947)
respectively.)

1954. Note on prises of Thestor basutus (Wllg.). F. ent. Soc., S. Afr., 17. 140.
Gabriel, A. G. 1947. Phasis dicksoni sp. n. of South Africa. Entomologist, 80, 60.

Jackson, T. H. E. 1937. The early stages of some African Lycaenidae (Lepidoptera), with an
account of the larval habits. Trans. roy. ent. Soc., Lond., 86, 201.

Murray, D. P. 1939. The early stages of Phasis thero Linn. (Lepidoptera, Lycaenidae). 7. ent.

SOG., O- Affen 2s h-

1944. The genus Cupido (Lepidoptera: Lycaenidae) in South Africa. 7. ent. Soc.. S. Afr.,
7, 82.

1948. The genitalia of some South African Lycaenids (Lepidoptera: Lycaenidae). Ff. ent.
Soc., S. Afr., 10, 182.

1953. South African Butterflies, a monograph of the South Afvitad Lycaenidae. London.

1956. The species Lepidochrysops, Euchrysops and Cupidopsis of Southern Africa. The

genitalia Lep. Lycaenidae. Ann. S. Afr. Mus., 43, 109.
Pennington, K. M. 1953. New species and subspecies and other additions to the butterflies
of Southern Africa, together with new records of little-known species. 7. ent. Soc., S. Afr.,
16, 102-9 and plates I and II.

1956. A new species of Thestor from the Western Province. 7. ent. Soc., S. Afr., 19, 33-
Platt, E. E. 1921. Food-plants of Lepidopterous larvae. S. Afr. J. nat. Hist., 3, 65.

Riley, N. D. 1938. Descriptions of new or little-known South African Rhopalocera. Trans.
roy. ent. Soc., Lond., 87, 238 and plate I.

Seitz (Aonawalline C.). 1925 Macrolepidoptera of the World, 13.

Stempffer, H. 1936. Contribution a etude des Lycaenidae éthiopiens. Bull. Soc. ent. Fr., 41,
284.

Swanepoel, D. A. 1953. Butterflies of South Africa. Cape Town.

Ann. S. Afr. Mus., Vol. XLIV.

Plate II

urray ded.

/2 CDEP
1. Phasis thero Linn. (genotype). 2. P. sardonyx Tr. 3. P. argyraspis Tr. 4. P. wallengrenii Tr.
5. P. malagrida Walleng. 6. P. dicksoni Gabr. 7. Aloeides egerides Riley. 7a. A. pierus Cram.
(genotype). 8. A. aranda Walleng. 9. A. simplex Tr. 10. A. damarensis Tr. 11. A. moloma Tr.
12. A. almeida Feld. 13. A. thyra Linn. 14. A. taikosama Walleng. 15. A. barklyi Tr. 16. Poecil-
mitis lycegenes ‘Tr. (genotype). 17. P. lyncurium Tr. 18. P. chrysantas Tr.

19. P. aethon Tr.
20. P. lycia Riley. 21. P. chrysaor Tr.

Ann. S. Afr. Mus., Vol. XLIV Plate IV

PYaeAy Gel
VIS/-

22. Poecilmitis phosphor Tr. 23. P. pyroeis Tr. 24. P. felthami Tr. 25. P. zeuxo zeuxo Linn.
26. P. penningtoni Riley. 27. P. aridus Penn. 28. P. palmus Cram. 29. P. thysbe Linn. 30. A. orthus.
31. P. pyramus Penn. 32. P. tsino (subsp. nov.) (not yet described). 33. Chrysoritis oreas ‘Tr.
34. Crudaria lerona Walleng. 35. Anthene larydas Cram. 36. A. amarah Guer. 37. A. livida Butl.
38. A. definita Butl. 39. A. otacilia Tr. 40. A. princeps Butl. (neglecta Tr.). 41. A. lemnos Hewit.
42. A. millari Tr. 43. A. minima Tr. 44. A. talbott Stempff. 45. Desmolycaena mazoensis Tr.

Re ate DOiit of print: Vols. i Ul oe 1-3, 5 - ne ey Il t (Part 1), V(
Jeok -@ index), VIT (Parts tr, 2, 35), Atl, Tx (Barts)1, 2), X Sieg 2

=i (Part 1), XXIV (Part 2), XXXI (Parts 1-3). ae ae, Se

Vol. ) a
II. 1900-1 902 Zoology aad Geology cel
III. 1903-1905 Zoology - ei eed 2 ‘
IV. 1903-1908 | Palacentalany nage .
V. 1906-1910 Geology, Palaeontology, Zoology, “Anthropolog
. Gann | Parts 1, 2, 5, 8, Q) 7 ta ae ec nee
Vi... 1908-1910", Zoology re) (62) pa . (excl. Par ce
VII. 1908-1913 Palaeontology Hae re ih e fe pitas Ae a d ¢
IX. 1911-1918 Botany ei Soe Mee, it
M711 19rd: Loolopy ur. 2s pe eh
XI. 1911-1918 Zoology .. Ae Se cate
XII. 1913-1924 Palaeontology and Geology A a
XIII. 1913-1923 Archaeology and Zoology % ak td
XIV. 1915-1924 Zoology .. oH ae LE
XV. 1914-1916 Zoology ..- .. ; os te
XVI. 1917-1933 Botany .. .. Lae ae
XVII. 1917-1920 Zoology .. Scie s ey!
XVIII 1921) Zoology. 1747.5) tres acs ala
MIM; 1924-1925 = Zoology. 6.09) =~ sea ae ia
4 XX. 1924-1926 Zoology .. ry eek oe Be
XXI. 1925-1927 Zoology .. Ag PBR
¢ ~ XXIT, 1925-1928 Palaeontology ... .. --
fh XXIII. 1925-1926 Zoology .. Eee ees a
; XXIV. 1929-1938 Anthropology and Ethnology ..
XXV. 1927-1928 Zoology .. uk ue fe a ee
§ RO 1s 1928 Zoology .. oe Ap Ae ae
‘ XXVII 1929 Anthropology .. rs Sat tate
XXVIII. 1929-1932 Palaeontology .. .. an Ars
XXIX. 1929-1931 Zoology .. Re eres
XXX. 1931-1935 Zoology .. a AU eeetrmetteh ss %s.
INDEX of papers, authors, and ances published i in Vola: Besa
XXXI. 1934-1950 Palaeontology .. ors ace
XXXII. 1935-1940 Zoology .. stb Saati sa
XXXIII. 1939 Zoology a Ae Ne Ke
o¥ XXXIV 1938 Zoology’) a.° 55. a as
XXXV. 1956 Zoology .. Rae EN a ee
XXXVI. 1942-1948 Zoology .. Ms = sis
XXXVII. 1947-1952 Archaeology a Nek of
; XXXVITI 1950 Zoology .. ne ea ia
XXXIX. 1952 Zoology .. ey a asa
: XL. 1952-1956 Botany .. oe eS hal
XLI. 1952-1955 Zoology .. oe are es
XLII. 1953-1956 Palaeontology + cha : oe :
XLIII. 1955-1957 Zoology and Palancntolony is os y
XLIV. 1957- Parts 1-6, Zoology and Palaeontology . nia

Copies may be obtained from— __ : x
The LIBRARIAN, SoutrH AFRICAN Museum, Carz Town. a oe

pee Mowliraiainn

Bie) of South African estuaries. Part X: Kon Bay estuary system. By
_ G. J. Broexuuysen and H. Taytor. (With 1 figure i in the text and
| Plates Mi. and Ve ie

a { te blastea. By N. A. H. Mitzaro. (With 4 figures it in the text.)

ed of the subspecies of the Ye allows Gnear, Serinus flaviventris (Swainson) i. ee “ a
By J. M. WINTERBOTTOM. earn ee : °g

ISSUED APRIL 1959 —“~PRICE 8s. 6d.

ee PRINTED FOR THE . ae
TRUSTEES OF THE ne AFRICAN MUSEUM Rng

os
«

es 4
5 .
ote |
Py
PR
f ers
Fi | | '
at)
“ee , : :

‘
;

__ —j =. rn
ar a, ae) x
- -
et 4
—

THE ECOLOGY OF SOUTH AFRICAN ESTUARIES.
Parr Vill. KOS! BAY ESTUARY SYSFEM
By

G. J. BRoEKHUYSEN AND H. TAYLOR
Department of Xoology, University of Cape Town

(With 1 figure in the text and Plates V and VI)

The topography of the Kosi Bay Estuary System is described and observations on the
physical and chemical factors are tabulated. The estuary is partly tidal and there is a salinity
gradient ranging from 3°3%, in the top part to 33%, near the mouth. On the basis of these
factors and ecological and topographical factors, the system is divided into seven regions. ‘The

fauna and vegetation of six of these regions are described. A total of 216 species of animals was

found and is listed in the appendices. The composition and distribution of the fauna are discussed
and compared with those of Richard’s Bay. It is concluded that the distribution is similar but
the composition is very different, probably due to difference in substrata as salinity and other
conditions are rather similar in both.

CONTENTS
INTRODUCTION .. if ae ee 270) NoTEs ON THE Birps Ae =e MOT
TOPOGRAPHY OF THE SYSTEM IN ’ Discusston af ay sis sway) Coty)
GENERAL .. su bi ee ZOE REFERENCES .. bs op ue, (28S
RESULTS OF THE SURVEY... oe eee APPENDICES... He Ris err ZOO)
NOTEs ON THE FISHES “ie xe (266
INTRODUCTION

In July 1949 a team of six biologists from the Department of Zoology of
the University of Cape Town revisited the St. Lucia Estuary and Richard’s Bay
in order to investigate any differences which might have resulted from the
floods since the previous visits. On that occasion the present authors joined the
Third Tongoland Expedition organized by the Natal Society for Preservation
of Wild Life and Natural Resorts, to carry out an ecological survey of Kosi Bay
Estuary. It was thought that the results of such a survey might be of con-
siderable interest in connection with work already carried out on St. Lucia
and Richard’s Bay Estuaries (Day, Millard and Broekhuysen, 1953; Millard
and Harrison, 1953).

Kosi Bay was reached in the evening of 11 July and work was started the
next day and continued until the 19th when we had to leave to rejoin our team
at St. Lucia. Due to the short time available (eight days), the difficulties with
boats, and the fact that the. work had to be done by only two people, the

219

280 ANNALS OF THE SOUTH AFRICAN MUSEUM

26.12.45"

28 2€

Of 2¢

KOSI-BAY iF
aS LA RY SIHLANDE i han

LAKE De f hi ash.
BASED ON 1 :50000 SERIES ee eo
OF SOUTH AFRICA eee ae ae
pz WATER i ie
wo MARSH SE ic
ue fe NRA
UGULU Le ee
Sp 4 MANGROVE S AM > ae as
ms \
RiGee. o SCATTERED BUSH
A map of the §
Kosi Bay Estuary ei SAND
System based on ie Deg RIC ON tan é
the I : 50,000 ae oie imino
series of South ine : LAKE»)
Africa. ks ROCK nas Hapa
The depth is CC ae 5) MRE be, alt
: 2 2 MILES \LARE N
given in feet. ae 8 Be
i s 2 ai i. a
joe
P < Ca Sian a, ee 2d
A | CNKANINI ei
ane Ta RIV. md ni an
SOT a. s

27.00.00

SHENGETE
LAKE

=p
3,
:

THE ECOLOGY OF SOUTH AFRICAN ESTUARIES: KOSI BAY 281

. extent of the survey was limited. As the boat and canoe were too light for
_ dredging, we could only work the shores and shallow water. We feel, however,
that a fairly comprehensive knowledge of the main biological characteristics of
_ these shores and shallow waters has been obtained. Mr. G. D. Campbell
_ collected the majority of the species of fish given in the Appendix C (Gampbell
and Allanson, 1952).

TOPOGRAPHY OF THE Kosi Bay EsTuARY SYSTEM

It consists chiefly of three lakes, e.g. Nhlange Lake, Sifungo Lake and
Mpunowini Lake (fig. 1). The system runs from south-west to north-east and
opens into the sea 2 to 3 miles south of Oro Point (indicated but not named on
the map) which is in Portuguese territory. A narrow channel broadening into
a shallow tidal basin connects Mpunowini Lake with the sea. Four rivers flow
into the system: (a) the Tombeni River which enters Nhlange Lake at the
south-west point, (b) the Nkanini River which enters the same lake on the west
side, (c) the Ugulu River which enters the northern tidal basin at a point

south-west of the Mission Station and (d) a small stream from Sihlande Lake
which enters the same basin just east of Noisy Point. The whole system from
the southern shore of Nhlange Lake to the mouth is 7 to 74 miles long.

RESULTS OF THE SURVEY

The system can be divided into seven sections, based on the physical,
chemical and biological factors.
These are as follows:
I. The mouth of the estuary (the section east of Noisy Pt.).

Il. The tidal basin, which is the section between Noisy Pt. and the mouth
of the Ugulu River.

Til. The shallows between the Ugulu River mouth and Mpunowini Lake.
IV. Mpunowini Lake.

V. Sifungo Lake.

VI. The winding narrow channel between Sifungo Lake and Nhlange Lake.
VII. Nhlange Lake.
During the eight days at our disposal we only managed to cover the first five
sections though a few observations were made in the other two.

The physical and chemical properties of the water in the different sections
have been tabulated in Appendix A.

I. Tse Mout or THE EsTruAarRy

A short straight channel, about 15 to 20 yards wide, running from west to
east, formed the connexion with the sea. This channel formed the outlet of a

SMOTTUTION JUN 1 1988

282 ANNALS OF THE SOUTH AFRICAN MUSEUM

basin bordered by sandy shores except on the south-east side where there was
an outcrop of limestone and more to the south a beach covered with limestone
pebbles. In the centre was a sandbank exposed during low water. By marking
on a stick the lowest and the highest water level which occurred during eight
days between spring and neap tides, it was found that the maximum tidal
range opposite the mouth was 2 to 24 feet. As can be seen from Appendix A
the salinity varied considerably but approaching that of ordinary sea water.
For the pH and temperature we can also refer to Appendix A. The water was
extraordinarily clear as in all parts of the system. The ‘rocky outcrop’ and the
‘pebble beach’ were particularly rich in animal life. A large variety of molluscs
and crabs were found, some of which were restricted to this area and the pebble
beach was rich in Polychaet worms. For detailed information the reader is
referred to Appendixes B and C. The sandy shores and the central sandbank
were rather poor and even the sand-crab Ocypode ceratophthalmus was not
common.

Il. Tue Twa Basin BETWEEN Noisy Pr. AND THE UGuLu RIVER

As can be seen from figure 1, this area was much larger than the previous
one and there was a striking difference between the two shores. Except for a
narrow twisting channel between 3 and 4 feet deep at low water, the basin was
extremely shallow. Moreover Native fishtraps (pl. V, A) stretched across the
whole area and made navigation, even in a small craft, extremely difficult
The tidal range was about 13 to 2 feet. The water was brackish (see Appendix
A), while the pH and temperature in shallow pools cut off at low water were

high.
The East shore

The northern part was a short narrow stretch of sandy beach with scattered
mangroves. Southwards the shore widened out and formed extensive shallow
pools and banks of sandy mud with large patches of dense mangroves (pl. V, B)
including Avicennia officinalis and Bruguieria gymnorhiza. At a slightly higher
level there was a zone of rushes — Juncus kraussi—which in part was very wide
and stretched up to where the bush started to grow.

The three following zones were clearly marked:

(i) Juncus zone. Inhabited by many crabs burrowing in the ground and
considerable numbers of Littorina scabra living on the leaves.

(ii) Mangrove zone. This was fairly extensive. It included the dense
mangrove growth in the middle and the higher parts of muddy sandbanks
separated by shallow water. The most striking features were the many crabs
(up to 31 holes per square yard under the mangroves) and the two whelks
Pyrazus palustris and Cassidula labrella. P. palustris was common just below the
dense mangroves, but was rather patchy (Pi. VI, A). C. labrella lived among the
thickest mangroves and seemed to prefer the shade. There was a certain

THE ECOLOGY OF SOUTH AFRICAN ESTUARIES: KOSI BAY 283

amount of overlap between the two. The barnacle—Balanus amphitrite—was

quite prominent on the mangrove trunks and fishtrap sticks.

(iii) Zone below mangroves. Consisted of gently sloping sandy mud with
hardly any vegetation. This area was fairly rich with many Polychaets (see
Appendix B), the Sipunculid Siphostoma australe and the bivalve Loripes clausus
~ which must have been very abundant in the past judging by the many empty
shells. The hermit-crab Clibanarius longitarsus and the prawn Panaeus japonicus
were both common.

The West shore

This was a narrow beach bordered by a belt of Yuncus which on the land-
ward side was replaced by Phragmites growing in a swampy surrounding. Just
above the reeds scattered young mangroves occurred. Most of the crabs found
on the east shore also occurred here among the Juncus and some of the Sesarma
eulimene were in berry. Seining in the shallows revealed the prawn P. japonicus
and young of several fish such as Mugil sp., Therapon jarbua, Gobius giuris and
_ Ambassis commersoni.

Ill. THe SHALLOWS BETWEEN THE UcuLu River MoutH AND MPuUNOWINI
LAKE

These shallows form the connexion between the tidal part of the estuary
and Mpunowini Lake. Although some mixing between fresh and brackish
water seemed to occur in this region, it was definitely much less than in the
area previously described (see Appendix A). The tidal range was probably
not more than 4 to 5 inches. The channel was 4 to 6 feet deep and the shallow
water bordering it was obstructed by many fishtraps.

The eastern shore was very similar to that of the previous section, with a
wide zone of dense mangroves at the water’s edge and Juncus at higher levels.
Near Mpunowini Lake there were only few mangroves and Juncus grew down
to the edge of the water while a green filamentous alga became quite common
along the margins at the entrance to the lake. On the western shore the man-
groves were rather patchy but the Juncus zone remained distinct. About half-
way along the shallows Phragmites appeared at the water’s edge. These reed-
beds later became quite extensive and at the entrance of the lake they were
several yards wide with a forest of mangroves behind and above them. The
bivalve L. clausus was not found alive but many empty shells were embedded
in the sandy mud at the northern end of the shallows. The large whelk Pyrazus
palustris, although common at the northern part of the shallows, petered out
towards Mpunowini Lake and the individuals which did occur were stunted.
C. labrella, so characteristic on the mud under dense mangroves of the previous
section, was gradually replaced by Assiminea bifasciata. Balanus amphitrite
disappeared just south of the Ugulu River mouth. Littorina scabra, however,
persisted at and above extreme high-water level and on the leaves of mangroves

254 ANNALS OF THE SOUTH AFRICAN MUSEUM

almost to the entrance of Mpunowini Lake. The sand-prawn Callianassa
Kraussi first appeared near the northern end of the shallows and became abun-
dant about half-way down, where 45 to 180 holes per square yard were counted.
Large numbers of Polychaet worms were noticed at this point. Crabs were
common not only under the mangroves in the northern part but also in the
Juncus zone on the western shore Hymenosoma orbiculare made its first appearance
in the narrow channel Amphipods such as Chiltonia capensis, Melita zeylanica
and a species of Grandidierella and Isopoda including Cirolana fluviatilis, Dies
monodt and Synidothea variegata also appeared and soon became common. The
first specimen of the bivalve Modtzolus capensis appeared here and the presence
of Chironomid larvae in the sand indicated a strong fresh-water influence.

IV. Mpunowini LAKE

This lake was the smallest of the three and the northern part was mainly
shallow (about 3 feet deep). The deepest part was along the eastern shore
where a depth of 21 feet was recorded. We also found a definite layering in
these deeper waters with a surface salinity less than half that of the bottom.
There was also a vertical temperature difference of at least 1$° C. Details of
the physical and chemical conditions are given in Appendix A. There were
indications of a tidal range of about 4 inches at the northern entrance.

The northern shore was a narrow sandy beach with Juncus, grass and
scattered palms growing on the bank above. This vegetation harboured many
large crabs (Sesarma meinerti). ‘The shallow water contained more algae
including two filamentous green species and one brownish one and amphipods
and isopods were numerous.

The eastern shore consisted of reed-beds with scattered mangroves in
between. Young fish (Therapon jarbua), prawns (L. pacificus) and many isopods
and amphipods were seen in the shallows. Above and beyond the reeds were
numerous holes inhabited by S. meznertz.

The western shore was covered with mangroves, two species of rushes,
palms and grass growing at and above high-water level. At the southern end,
near the connexion with Sifungo Lake the brack grass Ruppia maritima made its
first appearance. Littorina scabra was still common along this shore and Modtolus
capensis had become more numerous. Callianassa krausst was common or
abundant along the whole of this shore and 153 to 162 holes per square yard
were counted.

V. Srrunco LAKE

The connexion between Mpunowini Lake and Sifungo Lake consisted of
two short channels separated from each other by an island covered with dense
mangroves. The large crab S. meinerti occurred in great numbers on this
island and was collected by Natives as food, while the small periwinkle
A. bifasciata, common in the area between the Ugulu River mouth and

THE ECOLOGY OF SOUTH AFRICAN ESTUARIES: KOSI BAY 285

Mpunowini Lake, was present in its characteristic habitat. ‘The two channels
were fairly deep at the northern end but where they ran into Sifungo Lake
they were blocked by a shallow sandbank only covered by a few inches of
water.

The north-western part of the lake was only a few feet deep, but towards
the middle of the eastern side depths up to 46 feet were recorded. The water
was exceptionally clear. Salinity determinations of surface- and bottom water
samples indicated the existence of some layering but not as striking as in the
previous lake (Appendix A). The bottom water was 3° C. warmer than the
surface.

The northern and southern shores were narrow beaches of clear sand while
the western and eastern shores were fringed by dense reeds with mangroves
erowing behind them. The three algae seen in Mpunowini Lake were present
but the one like a broad Enteromorpha seemed to peter out and a new green alga
—Chara macropogon—appeared for the first time. R. maritima now became
common. Two species of isopods were common and three species of amphipods
_ were collected one of which (Urothoé serrulidactylus) was restricted to this area.
Modiolus capensis had now become fairly common in places and A. bifasciata
was still present in the north-western corner of the lake. The bivalve Psam-
mobia ornata appeared to be common in the shallow water among the reeds
and it is remarkable that the only other place where this species was collected
was on the sandy shore of the mouth of the estuary (see Appendix B). There
was one common Polychaet. The sand-prawn Callianassa was still abundant
and from 117 to 171 holes per square yard were counted.

VI. CHANNEL BETWEEN SIFUNGO LAKE AND NHLANGE LAKE AND NHLANGE
LAKE ITSELF

Since only one short visit was made to these regions, information regarding
them is incomplete. As shown in pl. VI, B, a narrow winding channel fringed by
tall reeds connected the two lakes. Its depth averaged 6 feet or even 11 feet
in places. Nhlange Lake, which had a diameter of approximately 3 to 4 miles,
is separated from the sea on the south-eastern side by a low sandbar only a
few hundred yards wide. We heard rumours that the lake was over 60 fathoms
deep in one spot but we know of no published records. We, therefore, took a
series of depth-soundings which have been entered in figure 1. From this it can
be seen that the maximum depth we found was 52 feet. The greater part of
the lake appeared to be 9g feet or less. It is still possible, that greater depths
occur in the western part of the lake where soundings were not taken. The
figures in Appendix A indicate that the water had a very low salinity and
appeared to be uniform from surface to bottom. The sole collection made was
where the channel entered the lake. Large rushes and R. maritima were very
common here. The mussel Modiolus capensis was abundant, attached to Ruppia,
also a fresh-water sponge, possibly an undescribed species of Desmospongia. The
fresh-water crab Rhynchoplax bovis and the shrimp Caridina nilotica were common,

286 ANNALS OF THE SOUTH AFRICAN MUSEUM

also two amphipods (Melita zeylanica and Talorchestia ancheidos\. ‘The only
isopod was Sphaeroma annandale:, which was restricted to this lake (Appendix B).

We may summarize by saying that the Kosi Bay System is an exceptionally
clear estuary. Although on the whole rather shallow, there are some very deep
parts and some of these show a distinct vertical Jayering. The mouth was
strongly saline and the ‘rocky outcrop’ and ‘pebble beach’ are very rich. The
Tidal Basin, which shows a considerable drop in salinity, is characterized by a
typical mangrove fauna and burrowing animals of tidal mud flats. There are
indications of serious silting in recent times, probably accelerated, if not
originally caused, by large numbers of Native fishtraps. Some of these have
stimulated the growth of mangroves. The shallows between Ugulu River
mouth and Mpunowini Lake have a small tidal range, a low salinity and a
brack-water fauna. Mpunowini Lake showed a distinct layering in the deeper
parts. The surface salinity was low. There was a slight increase of algal growth
and a further decrease in the number of species of animals. Sifungo Lake was
generally similar to Mpunowini Lake but the vertical salinity gradient was
not striking although the temperature gradient suggested layering of the
water. ‘The surface salinity is somewhat lower than that of Mpunowini
Lake. The fauna of the two lakes is essentially the same with minor changes
in abundance.

NOTES ON THE FISHES

Although we did a little seining, when time permitted, few records of fish
could be obtained. However, members of the Tongoland Expedition and
especially Mr. G. D. Campbell concentrated on the collecting of different
species of fish by angling and seining. Their results from Kosi Bay and some
other areas such as Lake Sibayi, Nyamiti and Kangazini Pans, etc., have
already been published (Campbell and Allanson, 1952).

In Appendix C all records of fish caught in the Kosi Bay Estuary System,
excluding the rivers running into it, have been listed. It is interesting to com-
pare this list with the list of species recorded from Richard’s Bay by Millard
and Harrison (1952). It appears that only 22 species occur in both estuaries;
38 occur at Kosi Bay but not at Richard’s Bay and 54 occur in Richard’s Bay
and not at Kosi Bay. Of the 38 species restricted to Kosi Bay, 19 were collected
near or on the ‘rocky outcrop’ at the mouth of the estuary. The relatively
large number of species which occurred either in the one or in the other but
not in both estuaries is interesting. The fact that Richard’s Bay had extensive
Kostera beds which provided shelter and food while no Zostera occurred at
Kosi Bay, and the fact that Kosi Bay had small rocky and pebble patches,
which did not occur at Richard’s Bay may account for the difference in species
in the two estuaries. It should also be remembered that the water in the Kosi
Bay System seemed to be clearer than that of Richard’s Bay and on the whole
was very much less saline.

‘

THE ECOLOGY OF SOUTH AFRICAN ESTUARIES: KOSI BAY 287

NOTES ON THE BriRpDs

Aquatic birds may be important in the ecology of estuaries. During the
short time in which the present survey was carried out, any birds wholly or
partly dependent on water were recorded. The total number of species
tabulated in Appendix D is 36. The figures in the different columns are the
maximum numbers seen in the area at any one time. The number of species
is relatively low possibly because the Kosi Bay System was visited during the
winter when most palaearctic waders had left for their northern breeding
quarters. From Appendix D it will be seen that the mouth of the estuary was
the poorest in bird life. The large numbers of Avocets and relatively large
numbers of Whimbrels are interesting.

DIsCcussION

Of all the estuaries which have been investigated along the coast of Natal
(Durban Bay, St. Lucia and Richard’s Bay) none is actually comparable with
the Kosi Bay System. Although Kosi Bay shows a gradual salinity gradient
from slightly brack water at the top to almost sea water near and at the mouth,
it is unique in that the major part contains brack water and that it is divided
in such distinct parts interconnected by narrows. This seems to be the first
time that the fauna of long stretches of brack water has been studied in the
ecology of South African estuaries.

Of the other Natal estuaries studied Richard’s Bay perhaps comes nearest
to Kosi Bay, although the topography is very different and the largest part
contained water which was much more saline. Both, however, have a gradual
salinity gradient and are relatively ‘clear water’ estuaries. Moreover they
are near enough to each other to expect considerable numbers of identical
species.

Comparing the two estuaries we find that the total number of animal
species —excluding birds—recorded from Kosi Bay is 173 and for Richard’s
Bay 183.

Although the value of these figures is limited as they are so dependent on
the thoroughness of the Kosi Bay survey, their similarity is rather striking.

In Table I the distribution of the different phyla over the different parts
ot the Kosi Bay System has been analysed. No information for Nhlange Lake
is given as this area was very inadequately covered.

From this table it is clear that the largest number of species occurred in
‘the mouth of the estuary’. Millard and Harrison (1952) found that at Richard’s
Bay ‘the middle reaches’ were by far the richest. As they point out (p. 174),
number of species alone cannot give an idea of richness of the population.
Reference to Appendices B and C of the present paper and Appendix B of the
Richard’s Bay paper will show that many species were common or limited to
the regions under consideration.

288 ANNALS OF THE SOUTH AFRICAN MUSEUM

TABLE [. DiIsTRIBUTION OF DIFFERENT PHYLA

Estuary Za AL ka ee
os a =a
Mouth =. < 9 S
2 | oT= S 5
wae
S| e | 2 (ea) so
5 ion | 2 Ses er he =9 IS
ow oi. eS SS |) Byeaiee 5 ©
eon hegre io Se 3, £. *
< oO <<] } © om et io. oO
= o ° E PE) Bs sf bad
) a) = eo ro. an
Lar) se) 5 > o @)) =) io)
@) 2 | a | Cc = < C
Sf so)
© og = & 09
co » &
—O | Oo =
| <5 1 &
|
=x Lise ee ems | | 2
Porifera ) a el) Oo ) fe) Oo
Coelenterata .. O 6) 3 O fe) oO oO
Annelida I Gy i Ao 8 5 I I
Crustacea AG it leg Al ee ae IO QI 9 13
Mollusca a A ae Bi lii23 6 9 2 3
Ascidia i. oe We, Ags | i
Pisces sh Hs Ae See iain a 18 6 2 2
et Babel 3 |
OWAT S his ni a i Peer he 205 39 14 20
eee -————> |
|

The difference in distribution of the animals in the two systems is rather
striking and asks for an explanation. Considering the salinity of the water, the
‘mouth of the estuary’ at Kosi Bay seems to be similar to the ‘middle reaches’ of
Richard’s Bay. In the case of Kosi Bay there is no Zostera providing shelter
and food, but there are small patches of rocky and pebble substrata which
provide shelter and suitable attachment for sedentary species. In Richard’s
Bay there is extensive Zostera growth but no rocky and pebble substrata.

It therefore seems, that salinity together with the presence of shelter, food
and suitable attachment are responsible for the abundance of animals in the
regions under consideration.

It seems interesting to compare the actual species occurring in both
systems. In Table II the species occurring in both estuaries and those occurring
in one of the two only have been tabulated.

TABLE II. COMPARISON OF SPECIES AT Kost BAy AND RICHARD’s BAY

NUMBER OF SPECIES
PHyLuM ; i
Present in both Present in Richard’s | Present in Kosi Bay
Bay only only
Porifera a Aa a) | é O 3
Coelenterata .. ne ? 8 or II ?
Annelida sil By I =o) 12
Crustacea... 5 23 34 30
Insecta She fy O hale 2
Mollusca... ay 5 12 32
Ascidia oy ih O O I
Pisces ¥ ‘ss 22 54 38
TOSAL,:. i 51 123 118

THE ECOLOGY OF SOUTH AFRICAN ESTUARIES: KOSI BAY 289

From this table it is obvious that the fauna of both systems is very different
and that only 51 species occurred in both the estuaries. This is the more
striking as both estuaries as regards salinity conditions are to a certain extent
similar. This suggests the conclusion that type of substratum may be more
important than salinity especially in Mollusca.

ACKNOWLEDGEMENTS

The work was made possible by the co-operation of the Natal Society for
Preservation of Wild Life and Natural Resorts and the assistance given and
interest shown by the other members of the Third Tongoland Expedition.

To these, and the systematists who have helped with the identification of
the material the authors tender their sincere thanks. Professor J. H. Day has
been kind enough to criticize the manuscript, and his helpful suggestions are
much appreciated. The method of work was on the whole similar to that
described for other papers in the series.

REFERENCES

Campbell, G. D., and Allanson, B. R. 1952. “The Fishes of the 1947, 1948 and 1949 Scientific
Investigation of the Kosi area, organized by the Natal Society for the Preservation of Wild
Life and Natural Resorts.’ Magaz. Natal Soc. f. t. Preserv. 0. Wild Life and Natural Resorts,
pp. 1-8.

Day, J. H., Millard, N. A. H., and Broekhuysen, G. J. 1953. “The Ecology of South African
Estuaries. Part IV: The St. Lucia System.’ Tvans. Roy. Soc. S. Africa, vol. XXXIV,
pp. 129-56.

Millard, N. A. H., and Harrison, A. D. 1953. “The Ecology of South African Estuaries. Part
V: Richard’s Bay.’ Trans. Roy. Soc. S. Africa, vol. XXXIV, pp. 157-79.

APPENDIX A

The physical and chemical factors of Kosi Bay Estuary. In each column the lowest and
the highest figure represent the extremes, the other in the middle is the average. The number
of records on which this average figure is based is shown as superior figure. Where surface and
bottom samples were taken, these figures are given separately. The pH figures are without the
salt-error and therefore somewhat too high.
* Shallows cut off between mangrove banks.
** Boggy place just above H.W.S.

Area

Mouth of Estuary

Salinity

| 21-2—28-73-392-9

Water tempera-
ture am °F.

19°2—20°517—-22°0

8-5-8587

Turbidity

(Secchi Disc.)

Tidal Basin 9°9-13°3°-16°6 | 21°:3-23°8°—26-0* | 7-4**—-6-4'-g-2*
Shallows between |
Uegulu River mouth
and Mpunowini Lake | 9°91 18-0-18:74-19°9 | 8-3-8-4°-8'5
Mpunowini Lake 7°4-7°8?-8'1 18-8—19°64-20°5 8-31
[Surt.s yee. 18°8
_ Bottom: 18°3 | 20°2 |
Sifungo Lake | Surf.: 6-6 [19-3-20-77-23°0 ||| 96-970. uae 23 feet
_ Bottom: 7-6 | Surf.: 19°3 Spe
75 Bottom: 23-0 .7°5 (HS smell)
BIE |
Nhlange Lake 3\9-3°4°=3°5 | 20°2-90797-20°4. | 8-0-8-37--Br 14 feet
| Surf.: 3°3 20°2
Bottom: 3°5 20°0
APPENDIX B
Comparative list of the fauna of the Kosi Bay Estuary System.
P = present, C = common, A = abundant.
Estuary ZH Pe a) A
Mouth | OB) ° teria S ey ee
| 2) ‘oe ee 5 sy)
| | | =| 2 5 | 8 | da
S a a erg = 6. = im 3
=) or ©. ast =e =} in
alo) e 8° 8) eee ee ee
< me il Set za. ae a a
“A Sprites, o
S| o | 26 2 (are
oO oO = lo O | ee Baio a
5 i) 5 le onl o +] o
L.9 4 Bil ce er eae
Peg uke ate |
U | o's] elas
g.9| Be.
£5} 7 ©
PORIFERA
Desmospongia sp. (KOS 6B) C
Haliclona raphidiophora (Lendenfeld) ig
Higginsia coralloides Higgin Lg
COELENTERATA: Actinozoa
Small brown species (KOS 21A) P
Light coloured sp. (KOS 30A) oy r
Small burrowing anemone (KOS 33D) C

290

THE ECOLOGY OF SOUTH AFRICAN

ANNELIDA: Polychaeta
Ceratonereis keiskama Day
Dasybranchus caducus (Gr.)
Dendronereis arborifera Peters
Loimia medusa Sav.

Lycastis indica Southern ..
Nerine cirratulus (D. Ch.)
Notomastus abberans Day. .
Orbinia bioreti Fauvel
Perineries capensis (Kbg.)

_ Pomatoleois kraussui (Baird)

Scolecolepsis indica Fauvel

ANNELIDA: Szpunculoidea
Phascolosoma scolops Sel et de Man
oy stephensont Stephen
Siphonostoma australe :

Crustacea: Cirrepedia
Balanus amphitrite v. denticulata Broch. ..
5 trigonus Darwin

Chthamalus sp. (KOS 18A)

Tetraclita squamosa (Brug.) v. rufotincta Pilsbry.

CrusTAcEA: Tanaidacea
Tanais philetaerus Stebb.

CrustTAcEA: Jsopoda
Cirolana fluviatilis Stebb.. .
Dies monodi Brnrd. :
Pontogeloides latipes Brnrd.
Sphaeroma annandalei Stebb.
Synidothea variegata Cllge.

Crustacea: Amphipoda
Afrochiltonia capensis (Brnrd.)
Grandidierella sp. (KOS 62D)
Melita zeylanica Stebb. ..
Talorchestia ancheidos Brnrd.
Urothoé serrulidactylus Brnrd.

Crustacea: Macrura
Alpheus sp. (KOS 32B) ..
Athanas sp. (KOS 33)

Caridina nilotica (P. Roux) Oe
Metapenaeus (? monoceros) (Fabr.)
Palaemon pacificus Stimps.
Penaeus indicus M. Edw...

» japonicus Bate

55 monodon Fabr.

ESTUARIES :

Estuary
Mouth
gl y|2
ales) &
SE | ie | Se
Cee
Ko |

ile P
P
P
B
Cc 2}
Ie

KOSI BAY

Zo) we QZ
<i Eee ee || |
fe 42/8) S| a
eee | S| |e
=) |
gs S84. ©. fa
ie? Ieee) &
> 3) & |
Sica he!
#3) 52
25 2 5
R
C
PR Cc Leas
G
P
RE
P
Chit yes.
Bayi) gab
CHG uk
FE
C | P
© LAP
C G
| G
Buse iG
Cet
2
A A Gist
P Guleey| A
P
| A
P
Chine
We
Chime

292 ANNALS OF THE SOUTH AFRICAN MUSEUM

CRUSTACEA: Anomura
Calcinus laevimanus (Randall)
Callianassa kraussi Stebb.
Clibanarius longitarsus (de Haan)

A; virescens (Krss.)
Coenobita cavipes Stimps.

CrustAceEA: Brachyura

Actaea depressa (White)
Actumnus setifer (de Haan) 4
Cyclograpsus punctatus M. Edw. ..
Epixanthus frontalis (M. Edw.)
Eriphia smithii McLeay :
Euruppellia annulipes (M. Edw. ‘
Grapsus strigosus (Herbst)
Hymenosoma orbiculare Desm.
Liomera bellus (Dana)
Matuta lunaris (Forsk.)
Metapograpsus messor (Forsk.)
Ocypode ceratophthalmus (Pallas)
Ozius regulosis Stimps. :
Pseudograpsus erythraeus Kossman
Pseudo zius caystrus (Ad. & ere
Rhyncoplax bovis Brnrd.
Sarmatium ? crassum Dana
Scylla serrata (Forsk.)
Sesarma catenata Ortm.

,, eulimene de Man

, meinerti de Man
Uca annulipes (M. Edw.) ale
Uca chlorophthalmus (M. Edw.)
Xantho quinquedentatus Kr. ie
Kozymodes xanthoides (Kr.)

INSECTA
Chironomid larvae
Hydrometridae

Mo.t.usca: Pelecypoda
Barbatia decussata (Sow.)
Crassostrea cuculata (Born)
Dosinia hepatica (Lam.) .
Isognomon dentifera (Krss.)
Loripes clausus Phil.

Modiolus (Brachydontes) papiabilie Krss, i,

Modiolus capensis Krss.
Mytilus perna (Linn.)
Psammobia ornata Desh...
Pteria natalensis Jameson
Tellina queketti Sow.

Selbastas,

aroys Apues |

Estuary
Mouth

Qe Oi Otte @ eter

peels = Raa el @

yorsq 219994

doisyno AYI0Yy |

PHO

‘Iq ASION
TePLL

YINoul JIA
uIseq

njnsQ pue
u99M49q

nei? aoa]

ayey TIM
-oundyy pue yynour JAN

ao)

POs Go GS ee

n[NsAQ, wsamjoq sMoOTTeYS

aye'y turmoundyy

oayeyT osunyig

yey osuelyN

THE ECOLOGY OF SOUTH AFRICAN ESTUARIES:

KOSI BAY

293

Mo tusca: Gastropoda
Assiminea bifasciata (Nevill)
Cassidula labrella Desh.

Cellana capensis (Gmelin)

Cerithidia decollata (Linn.)

Cerithium morus Lam. ..

Coralliophila sp. (KOS 26U)

Cypraea annulus Linn.

Drupa squamosa (Pease) .. ae
,, tuberculata (Blainv.) Le

Engina (Pusiostoma) mendicaria (Linn.) “

_Littorina obesa Sow.
Littorina scabra Linn.
Mitra litterata Lam. :
Natica marochiensis Gmelin
Nerita albicilla Linn.

», plicata Linn.

;, umlaasiana Krss. ..
Planaxis sulcata Q. & G.
Polynices mamilla Lam. .
Pyramidella mitralis A. Adams
Pyrazus palustris Linn.
Siphonaria anneae Tomlin

pe capensis Q. and G.
“3 oculus Krss.
Tricolia bicarinata (Dunker)

Vermetus sp. (KOS-37A)

ASCIDIA
Styela aequatorialis Michaelson. .

a1oys Apurs |

Estuary
Mouth

ywaq sIqQqeq

(YP aclige) alae) ae u/@ll@) acl@lact@)

doisjno Ayo xy |

you sJaATYy

Id ASION

nynsp, pure

@)reled

ulseq

ayer] 1UIM
-oundyy puv yynow Ary -
nyNsAQ usamjzoq sMoOTTeYyS

u29M}9q

oye'yT turmoundyy

Oye] Osunjig

oyey osurlyN

294 ANNALS OF THE SOUTH AFRICAN MUSEUM

APPENDIX CO

Comparative list of the fishes of the Kosi Bay Estuary System
P = present, C = common, A = abundant.

Estuar Zl ae eZ
ett g. = <5. 5 = =
Slate le eee
e)e| FFrolge2| sles
O..) 0 © p)) Su on) age ere eee
Qu. a yr |S a| eS oO oa — ©
TS a » Bee 2 |S eeaaG
o o ° E 5 Cs < ©
2/8 | si" ieee
°F 8 |” ae) ae
O.)) . og deems
£.9| Be
a= 1 &
Abudefduf saxtilis Forsk. is oy, .. |<—|-——| P -—=>
Acanthopagrus berda (Forsk.) in ie C C
Acanthurus fulgiginosus (Lesson) Me .. |<-P |—+
e triostegus (Linn.) 1?
Alticops oryx (Cuv. ) ie oe Ip
Ambassis commersoni Cuv. & Val. ms ak C
ny safga (Forsk.) .. ‘ a7 oh P C
Antennablennius bifilum (Gunth. ,) sts .. [<—}——] CG |—>
Arothron aerostaticus (Jenyns) ; P
Bothus pantherinus (Rupp.) P
Callyodon guttatus (Schneider) C
Caranx melampygus Cuv. nie of. ..) |<—|-——| P |——=>
» sexpascratus O. E.G. Sp] —
Chaetodon lunula (Lacep.) : P
Coracinus multifasciatus (Pellegr. ) |i?
Diplodus sargus Linn. Cc
Dules taeniurus Cuv. Cc
Echidna nebulosa (Ahl.) Ljocalijty Un/known
Eleotris fusca (Bloch) ..
Ellochelon vaigiensis (Q. & G. . “5 Liocalijty Unknown
Elops saurus Linn. ifs .. |<—| P =
Enneapterygius obtusirostre Klusinger 3 - C
Epinephelus areolatus (Forsk.) .. zie ae C
Epinephelus tauvina (Forsk.) AS Se at P
Fissilabrus dimidiatus (Val.) Bi 45 3 |e
Fistularia petimba Lacep. Ke ty, a les
Gerres acinaces Bleeker .. “i Ay: +0 Ljocali|ty Un|known
oyena (Forsk.) te af oh es C
5 rappi (Brnrd. \ ; 2 [<— P —> )
Gilchristella aestuarius (Gilch. & ‘Thomp, ) ae Bletwejen Sjifungo jand M/punjowinii
Gobius giuris Hamil. .. = Cc P CG a) ae,
Hepsetia pinguis (Lacep.) i in Liocalijty Unjknown
Hyporamphus delagoae (Brnrd. ee a ee Liocalijty Unknown
Johnius hololepidotus anes ee He .. |<—\—| P -——
Kuhlia taeniura (C. & V.) : oe 4 ig
Lethrinus nebulosus (Forsk.) me be sn Cc
Lithognathus mormyrus (Linn.) .. oh .. |<——| P ——
Liza ‘macrolepis (Smith) .. , aye .. |<—/—| A >
Lutianus argentimaculatus (Forsk. ) om yk Ig C
», Jfulviflamma (Forsk.) .. os i, Liocalijty Unjknown
“ vaigiensis (Q. & G.) a Cc
Monodactylus argenteus (Linn.) .. 2 .. |<———}| A |
Salciformis Lacep. . . iy .. |<—}+—]} A a
Mugil cephalus Linn. .. Me yh .. |<—}-——| A —>-
» vopustus Gunth: ” ,; i * .. [<——| PF —_>
oe ey hy be a a si Cc C Cc

a rn ET LL LL

APPENDIX D

Birds occurring in the near vicinity of the water.

Nore.— The figures appearing in the different columns are the maximum number seen at any
time in that particular area. Therefore the figures for the same species in different columns may

refer to the same birds but appearing in different areas.

Actitis hypoleucos (Common Sandpiper)
Anas undulata (Yellow-billed Duck)
Anhinga rufa (Snake Bird) oe
Ardea cinerea (Grey Heron)
» purpurea (Purple Heron)

Bubulcus ibis (Cattle Egret) :
Burhinus vermiculatus (Water Dikkop)
Butorides striatus (Green-backed Heron)
Ceryle rudis (Pied Kingfisher) :
Charadrius hiaticula (Ringed Plover)

ais marginatus (White-fronted Sandplover)

a tricollaris (Three-banded Sandplover)

yinour Arenysy

‘Id ASION
[ePLL

uIseq

YINOUW JOATY

nyns~Q, pue
u29M}9q

Localit

AR < ce
26| =>
ee 5
@
ee) ie
toy a ae =
Sister US) )
ma Oo a
foe sai Se
aee| &
Oia @
oS
= 09
2S
ss
I
< 4,———>
y Unkjnown
10 10
I
12
2
<—| I>
I 3)
I
4.
2 2

oye] osuelyN

296 ANNALS OF THE SOUTH AFRICAN MUSEUM

Circus ranivorus (African Marsh Harrier)
Corythornis cristata (Malachite Kingfisher)
Egretta garzetta (Littel Egret) ..
Gypohierax angolensis (Vulturine Fish Eagle)
Hagedashia hagedash (Hadedah) . :
Haliaétus vocifer (Fish Eagle) oP
Himantopus himantopus (Black-winged Stilt)
Larus cirrocephalus (Grey-headed Gull) ..
Megaceryle maxima (Giant Kingfisher)
Motacilla capensis (Cape Wagtail)
Numenius arquata (Curlew) a
a: phaeopus (Whimbrel)

Nycticorax nycticorax (Night Heron)
Pandion haliaétus (Osprey)
Phalacrocorax africanus (Reed Duiker) :

ha carbo (White-breasted eee
Phoenicopterus sp. (Flamingo). :
Porhyrio porhyrio (Purple Gallinule) a
Psalidoprocne holomelacena (Black Saw-wing

Swallow)
Pseudohirundo griseopyga (Grey-rumped Swallow)
Recurvirostra avosetta (Avocet) . .
Riparia riparia (African Sand Martin)
Sterna bergu (Swift Tern).. a
Tringa nebularia (Greenshank)

yjnour Arenysy

25) Pe)
(6) jst me
=. Ou <
na» o-=
< aye
ad 33
ey a) 236
Se Olea
o° 8/259

I 2
I I
II 10

Locality Unkjnown

282

ye] asuelyN

Ann. S. Af. Mus. Vol. XLIV Plate V

A. A typical Native fishtrap just opposite the mouth of the Ugulu River.
Photographed by G. F. Broekhuysen.

B. The southern part of the east shore of the Tidal Basin between Noisy Pt. and the Ugulu
River mouth. The extensive growth of mangroves and large shallow pools in between as in
the picture was characteristic.

Photographed by G. 7. Broekhuysen,

Ann. S. Af. Mus. Vol. XLIV. Plate VI

Syaat iy

A. A dense patch of the whelk Pyrasus pallustris typical for the lower part of the mangrove zone
on the east shore of the tidal basin.

Photographed by G. F. Broekhuysen.

B. The channel connecting Sifungo Lake with Nhlange Lake. The growth of Phragmites reeds
all along the edges was characteristic and indicated a very low salinity.

Photographed by G. F, Broekhuysen.

HYDROZOA FROM THE COASTS OF NATAL AND
PORTUGUESE EAST AFRICA

PART II: GYMNOBLASTEA
By

N. A. H. Mritiarp, Px.D.
Koology Department, University of Cape Town

(With 4 figures in the text)

INTRODUCTION

This paper represents the second part of an account of the Hydrozoa from
the east coast of South Africa. The first part, dealing with the Calyptoblastea,
was published in the same journal in 1958.

The material includes a relatively small number of species, only 12 in all,
of which 2 are new species, and 3 are new records for the country. All of them
come from the littoral area or shallow water to a depth of about 7 m. There
were no identifiable gymnoblasts present in the collection trawled by the
s.s. Pieter Faure off the coast, although calyptoblasts were plentiful. However,
one must bear in mind the fact that the Pieter Faure material was collected nearly
sixty years ago and the preservation was not good. Under such circumstances
the gymnoblasts are the first to suffer, and in fact a few decayed remnants
indicate that several species were originally present.

The classification adopted is based on that of Rees 1957, who has attempted
to integrate the traditional dual classification of hydroids and their medusae
into one coherent system. Such a system should be the ultimate aim of all
systematists, and a revision of the group along these lines has long been overdue.
I am indebted to Dr. Rees for valuable advice on the position of problematical
species.

Also following on the lines laid down by Rees I have retained ‘separate
genera for hydroids with fixed gonophores and for hydroids with free medusae’.
In some cases this distinction is clear-cut and straightforward, but in the case
of Hydractinia and Podocoryne it is not so easy, for there are various intermediate
forms which produce degenerate medusae with a very short free-swimming life.
Although previously following Kramp, who combines the two genera, I have
now kept them apart, and have used Podocoryne to include forms which produce
medusae of a recognizable Podocoryne type, i.e. with 4 oral tentacles, and

297

298 ANNALS OF THE SOUTH AFRICAN MUSEUM

Hydractinia for forms with fixed gonophores of the styloid to eumedusoid type
without oral tentacles, even though they may be freed for a short while.

The structure of the nematocysts of hydroids has been receiving increasing
attention since the classic work of Weill in 1934. It is difficult as yet to estimate
the value of nematocysts in systematics, for so few species have been fully
described, but there is evidence that in some families at least they provide a
useful method of distinguishing otherwise closely related species. The difficulty
is that a systematist almost invariably has to work on preserved material
brought in by expeditions, where the chance of finding discharged capsules is
remote and where the shape is liable to be distorted. It is felt that any informa-
tion at this stage, however scanty, is of value, and will serve as a basis for a
future full assessment of the value of these structures in classification. Accordingly
descriptions and measurements of nematocysts have been included wherever
possible, using the categories suggested by Weill. Measurements are given to
the nearest half-micron.

The source of the material was detailed in Part I of the paper, and I wish
to thank once again those who contributed to the collection. I am also extremely
grateful to the Naturhistoriske Riksmuseet, Stockholm, and to the Zoological
Survey of India, for the loan of type material. The cost of publication was
partly defrayed by a grant from the Editorial Board of the University of Cape
Town.

As before, the type material of new species has been deposited in the
South African Museum (S.A.M.).

STATION LIsT

Db; Intertidal material from Isipingo and other localities near Durban,
collected in July 1935 and July 1936.

DBN. Material from Durban Bay (a land-locked bay).

DBN 2. 7/7/50. From floating jetty, Salisbury Island.

DBN 25. 13/7/50. From dead branch below low tide mark, Salisbury Island.

DBN 47. 18/7/50. Intertidal, on mud flats.

DBN 94. 8/1/51. Seined from shallow water north of Salisbury Island.

DBN 112. 10/1/51. Intertidal, on central sand bank.

DBN 130. 15/1/51. From ships’ hulls operating only in Durban Bay.

DBN 140. 16/1/51. From rotting branch trawled in o-7 m. north of Salisbury
Island.

DBN 191. 2/10/51. Intertidal, on causeway.

DBN 249. 24/4/52. From ship’s hull.

DBN 270. 26/4/52 Intertidal, on public ferry pier.

DBN 320. 28/4/52 From sublittoral fringe on caisson, Salisbury Island.

IN. Material from Inhaca Island, Delagoa Bay, Portuguese East Africa.
Collected by Mrs. M. Kalk.

IN 49 1954. Intertidal.
IN 112. 18/9/55. Punta Torres.
IN 140. 20/7/56. East shore rocks, intertidal.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 299

- MOR. Material from Morrumbene Estuary, inland from Inhambane, Portu-
guese East Africa.

MOR 34. 20/1/54. From hull of wreck at Linga Linga.

MOR 51. 21/1/54. Dredged from 3-5 m. off mouth of Rio Coche.
MOR 216. 15/7/54. From hull of wreck at Linga Linga.

MOR 217. 12/7/54 On weed in 2 m. of water at Linga Linga.
MOR 218. 13/7/54. From hull of wreck at Linga Linga.

NA. Material from intertidal zone on Natal coast.

NA 184. July 1950. Kosi Bay, reef, intertidal. Collected by B. R. Allanson.
NA 218. 18/1/58. Wentworth Beach, Durban.

STL. Material from St. Lucia Estuary, Natal.

STL 174. 12/7/49. ‘Channel’ area, south of Mpate River mouth, on aerial roots of
mangroves.

List OF SPECIES

Corymorphidae Eudendriidae
Corymorpha sp. Eudendrium carneum Clarke.
E um ?
Tubulariidae udendrium ?parvum Warren
Tubularia warren Ewer. Hydractiniidae
; Hydractinia diogenes n. sp.
Cladocorynidae Aydractima kaffraria Millard
Cladocoryne floccosa Rotch. Podocoryne nassa n. sp.
Pennariidae Pandeidae
Pennaria disticha Goldfuss Hydrichthys boycet Warren
Clavidae Bougainvilliidae
Corydendrium parasiticum (Linn.) Bimeria fluminalis Annandale

Family Corymorphidae
Corymorpha sp.
Records. DBN 47P.

Description. A single sterile specimen growing in the mud and reaching a total
length of 14 mm. Hydranths with at least 18 filiform tentacles in each group.

Remarks. In the absence of gonophores the species cannot be determined. This
is probably a young individual.

Family Tubulariidae
Tubularia warrem Ewer 1953
Tubularia warreni Ewer 1953, p- 351, figs. 1-4.

Records. DBN 2.0, 130D (reported by Day and Morgans 1956).
Description. Rich colonies on floating jetty and ships’ hulls, reaching a maximum

300 ANNALS OF THE SOUTH AFRICAN MUSEUM

height of 5 cm., and with abundant gonophores. Structure agreeing with
Ewer’s description, with the following additions.

The longitudinal septa within the stem are a feature of the species, but are
not constant in number and size. Some stems have 2 and some 3. In some
cases they meet in the centre as in 7. mesembryanthemum, but often they do not,
and sometimes they are merely very low ridges on the internal surface.

The dilation of the stem on which the hydranth rests is surrounded at its
widest part by a shallow transverse groove, above which the thickened ectoderm
protrudes in a pendulous flap, as described for 7. bethae Warren 1908. The
ectoderm is also thickened on the lower surface of the dilation as described for
T. crocea by Ritchie (19100), but unlike the latter there is a definite thickening
and differentiation of the endoderm lining the dilation, though not to the same
extent as in T. bethae.

Hydranth tentacles often fewer than stated by Ewer, varying from 18 to
29 in the proximal row, and 15 to 24 in the oral row in mature specimens.
Oral tentacles in one row, but occasionally alternate ones are slightly displaced.

Blastostyles not so constant in number as indicated by Ewer, and primary
and secondary pedicels often difficult to distinguish. Actinula with as many
as 6 oral tentacles at time of liberation.

Family Cladocorynidae

Cladocoryne floccosa Rotch. 1871

Cladocoryne floccosa. Allman 1872, p. 380, fig. 82. Warren 1908, p. 284. Vervoort 1941, p. 190.
Records. NA 184D.

Description. A single colony creeping on weed. Stems simple or with one lateral
branch; increasing in diameter from base to summit; annulated at base,
smooth or roughly corrugated for remainder. Hydranths very poorly preserved.
No gonophores.

Family Pennariidae

Pennaria disticha Goldfuss 1820, var. australis Bale 1884

Halocordyle cooperi Warren 1906, p. 73; Pl. 9. 19074, p. 209.
Pennaria australis var. cooperi. Warren 1908, p. 282.
Halocordyle disticha var. australis. Vervoort 1941, p. 192. 1946a, p. 290.

Records. D 39. NA 184B. DBN 2Q, 130B, 191D (recorded by Day and Mor-
gans 1956). MOR 34J, 216A, 218A. IN 140D.

Description. Rich colonies from intertidal rocks, ships’ hulls, etc., reaching a
maximum height of 13:9 cm. Structure exactly as described by Warren.
Hydrotheca-bearing ramules increasing in diameter towards distal end, and
generally annulated in basal portion only, as characteristic of variety, but
occasionally with annulations in distal portion as well. Gonophores observed
in January, July and October.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 30I

Nematocysts of at least 3 kinds:
(i) Stenoteles.

(a) Large, reaching a maximum size of 47:0 X 25-0, present only on
capitate tentacles.

(b) Small, varying in size from about 8 x 5p to 14 < 10p, on filiform and
capitate tentacles. Very abundant.

(ii) Desmonemes, 6-5 x 5:5, on capitate and filiform tentacles.

(iii) Undetermined heteronemes, 11-5 x 5:0p, on filiform and capitate tentacles.
Scarce. Capsule similar to the microbasic mastigophores described for
P. tiarella by Weill 1934, fig. 21, each containing a large refringent sphere
in the basal part.

Remarks. Four categories of nematocysts have been described by Weill 1934
for P. tiarella, and the maximum size for the stenoteles is given as 18 XIIp.
This species thus lacks the enormous stenoteles on the capitate tentacles which
_ are characteristic of P. disticha (cf. also Warren 1906), and this may be contribu-
tary evidence for keeping the two species separate and not combining them as
has been done by Vervoort 1941.

The undetermined heteronemes of the present material are almost certainly
microbasic mastigophores, which occur in both P. tiarella and P. cavolini
(= P. disticha), and there is possibly a fourth category which can easily be
overlooked in preserved material.

Family Clavidae
Corydendrium parasiticum (Linn.) 1767

Soleniopsis dendriformis Ritchie 1907, p. 495, figs. 142, 143; PI. 26, fig. 1.
Corydendrium parasiticum. Vervoort 19464, p. 292.

Records. DBN 2R, 140F.

Description. A well-developed colony reaching a maximum height of 7-2 cm.
and a smaller one of poorly preserved material. Stem fascicled at base with a
diameter of about 2 mm. Branching, and origin of hydranths as described by
Ritchie, except that the branches are not strictly in one plane. Perisarc thick,
smooth or faintly wrinkled, sometimes transversely folded below distal margin
possibly due to contraction of the hydranth.

Hydranths with about 22-209 scattered, filiform tentacles (though this may
be an underestimate since some of the tentacles may be retracted within the
perisarc), and the typical swelling below the base inside the perisarc.

Gonophores absent.

Nematocysts of 2 kinds:

(i) Undetermined heteronemes, 7-0 X 4:0 wy.
(ii) Desmonemes, 5:0 X 3°5 pL.

302 ANNALS OF THE SOUTH AFRICAN MUSEUM

Measurements (preserved material, mm.)

Hydranth pedicel, diameter. : . 0°2'7-0'48
Hydranth, length . : . O*5I-I°10
diameter : ; .” | 020-0737

Remarks. ‘The identification of this species is fairly definite in spite of the absence
of gonophores, since the method of branching is exactly similar to that described
by Ritchie 1907.

Vervoort 1946a has established fairly definitely that C. dendriforme (Ritchie)
is a synonym for C. parasiticum, but I am not convinced of the wisdom of including
C. sessile Ritchie 1910a as well, as has been done by Leloup 1937. Young
hydranth pedicels in C. parasiticum are necessarily adnate due to the method
of branching, but they do not remain so, and it is only the distal 2 or 3 pedicels
in a stem which have this arrangement. Ritchie’s material of C. sessile was a
well-developed colony of 37 mm. in height, and the pedicels were apparently
adnate throughout.

C. parasiticum is known from the Mediterranean, Cape Verde Islands,
French Indo-China and the Dutch East Indies. This is the first record from
South Africa. C. sessile has been reported from several localities in the Indian
Ocean (Mergui Archipelago, Cargados and Amiranie).

Family Eudendriidae
Eudendrium carneum Clarke 1882
Fig. 1, A-F

Eudendrium carneum Clarke 1882, p. 137; Pl. 7, figs. 10-17.
E. cunninghami Kirkpatrick 1910, p. 127; Pl. 7, figs. 1-3.
E. carneum. Vannucci 1954, p. 101; Pl. 1, figs. 1-9; Pl. 2, fig. 8; Pl. 4, figs. 2-5.

Records. DBN 25J, 1300, 249K, 270V, 320Q (reported by Day and Morgans
1956 as Eudendrium ?racemosum). MOR 34K, 216B, 217A, 218B.

Description. Rich, tree-like colonies, common on ships’ hulls, wrecks and piers,
etc., to a depth of about 6 ft. below water-level. Colonies reaching a maximum
height of 16:2 cm. Branching irregular, main stem thick and fascicled, up to
3°5 cm. in diameter, larger branches also fascicled. Groups of annulations
present on origins of branches, and rarely at intervals on main stem; hydranth-
bearing ramules with scattered groups of annulations or completely annulated.
Hydranth with pseudohydrotheca covering basal part and terminating in
annular groove, with about 30 tentacles (26—33 in 20 counts). No cnidophores.

Female gonophores borne in a whorl on a tentacular blastostyle, becoming
unevenly spaced along the length at a later stage when the blastostyle lengthens
and the tentacles are lost. Gonophores with a bifurcating spadix, which is later
shed. Developing embryo enclosed in a transparent perisarcal capsule which is
basket-shaped when empty.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 303

Fic. 1. Eudendrium carneum Clarke (A-F), and Eudendrium ?paroum Warren (G—H)

A, a large isorhiza. B and C, undischarged and discharged capsules of heterotrichous, micro-
basic euryteles. D, old female gonophores showing the basket-shaped capsules surrounding the
embryos. E and F, earlier stages in the development of female gonophores showing the bifurcating
spadices and the tentacles of the blastostyle. G, a female blastostyle with young gonophores.
H, part of colony with female blastostyles.

Male gonophores borne in an umbel-shaped whorl on a blastostyle devoid

of tentacles; 3-4 chambered.

(i)

Gonophores observed in January, April and July.
Nematocysts of 2 kinds:

Large isorhizas, probably atrichous. Capsule pear-shaped, slightly curved,
with operculum off-centre, measuring 2411p. Tube in many coils,
forming figures of 8 in the longitudinal axis. Only undischarged capsules
observed. Scattered irregularly on hydranth body and manubrium, but
most abundant in the ‘nettle-ring’.

Small, heterotrichous, microbasic euryteles, similar to those of E. vaginatum
(see Weill 1934). Capsule pear-shaped, narrowing at summit, measuring
9xX4p. Butt about two-thirds length of capsule, bearing 3 large spines on
the swollen distal end. Terminal tube bearing spiral ridges, coiled obliquely
in undischarged capsule. Present abundantly on tips of tentacles, and also
scattered irregularly in the ‘nettle-ring’.

304. ANNALS OF THE SOUTH AFRICAN MUSEUM

Remarks. ‘This material agrees in all its essential features, and particularly in
the nature of the female gonophores, with EL. carneum, though the nematocysts
of the latter have not been described.

It is also very similar to E. racemosum (Cavolini), another species with a
bifurcating spadix, but differs from it in the absence of cnidophores, in the
basket-shaped capsule of the female gonophore, and in the nature of the
nematocysts.

E. carneum is known from the east and west coasts of North America, from
Brazil, and from St. Helena. It is a new record for South Africa, and has
possibly been introduced on the hulls of ships, for it has been found only in
Durban Bay and Morrumbene Estuary. |

Eudendrium ?parvoum Warren 1908
Fig. 1, G-H
Eudendrium parvum Warren 1908, p. 272, fig. 1; Pl. 45, figs. 1-4.
Records. IN 49E.

Description. Colonies growing on weed and reaching a maximum height of
0-5 cm. Stem unbranched or branching irregularly, annulated at base, on the
origin of the branches, and at other irregular intervals.

Female blastostyles present, borne on annulated pedicels arising from stem
or direct from hydrorhiza. Blastostyle well formed, with manubrium and about
15 tentacles, bearing the gonophores in a verticil below the tentacles. Spadix of
gonophore unbranched, arching round egg in form of question-mark, later shed,
leaving the developing embryo enclosed in a very thin perisarcal capsule.

No large nematocysts present. Small ones (probably heteronemes)
measuring 5°5 X 2 present on tentacles and body.

Measurements (mm.)

Hydrorhiza, diameter : . 0'09-0°13
Stem, diameter ; - 0°09—0°12

Remarks. Vannucci 1954 has included E. parvum in the synonymy of E. capzllare
Alder, but I feel that this is a dangerous assumption at the present state of our
knowledge. Warren states that his species differs from £. capillare in ‘the three-
chambered condition of the male gonophore, the absence of a well-defined
terminal tubercle to the gonophore, and the extension of the perisare over the
base of the polyp’. Vervoort 1946 (p. 146) also states that E. capillare has no
nettle-ring or annular groove round the base of the hydranth, the latter being
present in EL. parvum.

The present material cannot be determined with certainty in the absence
of male gonophores. Warren has not described the female. The general
appearance of the colony is very similar to Warren’s material, but the hydranths
are too badly preserved to determine the details of structure.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 305

Several other small Eudendrium colonies are present in the collections from
this coast, but in the total absence of gonophores identification has not been
attempted.

Family Hydractiniidae
Hydractinia diogenes n. sp.
Fig. 2
Holotype. MOR 51H. (S.A.M. registered number H123).

Description. Colonies covering 5 gastropod shells occupied by hermits (Diogenes
costatus). Hydrorhiza a network of perisarcal tubes, covered by a layer of free
coenosarc, but clearly visible through it in the thinner regions. Spines smooth,
hollow and horn-coloured, of medium length, reaching a maximum height of
o-6 mm.

Gastrozooids reaching a height of about 2 mm. (preserved), with a long
hypostome capable of great distension, and 11-26 tentacles. No spiral zooids
or tentacular filaments observed.

| mm.

‘ 7. ‘ DN eet ra
NGNSs Z MAGES eM ee
Cue a ae Sey SR ON

Fic. 2. Hydractinia diogenes n. sp.

A, portion of female colony with gastrozooids and gonozooids, drawn from preserved material.
B, l.s. mature male gonophore showing circular canal. C, t.s. female gonophore showing radial
canals. D, l.s. female gonophore showing circular canal.

306 ANNALS OF THE SOUTH AFRICAN MUSEUM

Gonozooids variable in size, but generally smaller than the gastrozocids,
with 7-15 tentacles. The gonozooids are particularly abundant and well
developed around the openings and siphons of the host shells, and may com-
pletely fill the latter. In this region they are quite as long as the gastrozooids,
but have slender columns and never achieve the same robustness. Some have
a well-developed crown of tentacles, but in others, probably due to reproductive
exhaustion, the tentacles are reduced to mere stumps.

Gonozooid bearing a circle of up to 6 shortly stalked gonophores near distal
end of column. Male and female on different colonies, but colonies of opposite
sex may occupy the same shell. Gonophores in the form of fixed sporosacs
which have no free-living life (several partly empty ones observed still attached),
more or less spherical or with diameter slightly exceeding height when fully
mature, with no vestige of marginal tentacles. Spadix well developed, with its
internal cavity fairly spacious in the proximal third, but reduced to a crevice
in the distal region. Pedicel 0-02—0-07 mm. in length.

Male gonophore swollen with sexual products, which may protrude
through the distal aperture together with the end of the spadix, reaching a
maximum depth of 0-45 mm. and a maximum diameter of 0-46 mm. Circular
canal present around aperture, radial canals not visible.

Female gonophore containing 5-13 large eggs generally in 1 or 2 tiers
around a central spadix, reaching a maximum depth of 0-39 mm. and a
maximum diameter of 0:43 mm. Radial canals and circular canal present.
Though difficult to see in whole specimens the canals are easily visible in empty
gonophores and in sections.

Nematocysts of 2 types:

(i) Desmonemes, 5:0 X 3°5 p.
(ii) Microbasic euryteles, 9:0 X 4:0 p.

Remarks. Only three species of Hydractinia with smooth spines are known from

the southern hemisphere, namely H. parvispina Hartlaub 1905, H. subinermis —

Jaderholm 1923, and H. altispina Millard 1955. ,

H. diogenes is closely related to H. parvispina, but since Hartlaub states quite
definitely that there are no radial canals in the female gonophores of the latter,
the two cannot be synonymous. Attempts were made to obtain material of
H. parvispina, but unfortunately Hartlaub’s type material (deposited in the
Hamburg museum) was destroyed during the war. The species was subsequently
reported by Jaderholm 1905 from the Antarctic, and prepared sections of
this material were loaned to the author by the Naturhistoriske Riksmuseet,
Stockholm. The gonophores present were all male and immature, and no
further information could be obtained therefrom. No radial canals could be
found.

H. subinermis likewise has no radial canals in the gonophores, and further
the gonozooids are atrophied with only 2-4 tentacles, and the spines are very
short (0:16-0'19 mm.).

+ a eryeeecte

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 307

H. altispina is clearly distinct from H. diogenes, and the differences between
the two are summarized in the following table.

Colony growing:

Perisarcal reticulation
of hydrorhiza:

Spines:
Gastrozooids, number

of tentacles:

Gonozooids:

Male gonophores with:

Female gonophores
with:

1. diogenes

On gastropod shells occu-
pied by hermits.

Clearly visible through
coenosarc.

Of medium length reach-
ing 0-6 mm.

Up to 26.

Not particularly reduced,
with up to 15 tentacles.

Spermatogenic cells not
divided into groups and
no radial canals visible.

Large eggs, up to 13 in
number.

Hi, altispina

On occupied shells of
Thais squamosa.

Not visible.

Long, reaching 1-0 mm.

Uprto,12.

Much smaller than gastro-
zooids, with only 3-5
tentacles.

Spermatogenic cells divi-
ded into 4 groups and
radial canals clearly
visible.

Small eggs, up to 32 in
number.

Sterile colonies are very similar to those of Podocoryne carnea, but have

somewhat longer spines.

Hydractinia kaffraria Millard 1955
Hydractinia kaffraria Millard 1955, p. 217, fig. 2.

Records. DBN 47M, 94G, 112E (reported by Millard 1955, and by Day and
Morgans 1956).

Podocoryne nassa n. sp.

Figg

FAfolotype. IN 112. (S.A.M. registered number H122).
Description. Colony epizootic on the shell of the gastropod Nassa (Nassarius)
fenestrata. Hydrorhiza for the most part reticular, following the depressions of
the host-shell and covered with a thin layer of perisarc; but in the denser areas
closely packed stolons may be cemented together. No spines, spiral zooids or
tentacular filaments.

Hydranths columnar, but narrowed at base, reaching a height of 1-4 mm.
(preserved), with a conical hypostome and 8-16 tentacles. A very thin exten-

308 ANNALS OF THE SOUTH AFRICAN MUSEUM

sion of perisarc sometimes present over the base. Young hydranths have a
single verticil of about 8 tentacles, older ones have a second verticil of smaller

tentacles developing slightly proximal to the first and alternating with them.

Gonophores arising separately from the hydrorhiza on slender pedicels,
developing into free medusae.

The young gonophore is pear-shaped, flattened at the summit and
narrowing towards the base where it merges gradually into the pedicel which is
only about 0-02 mm. in diameter. The whole structure is completely enveloped

(eee et nm

tee,
‘. =

0-3 mm.

Fic. 3. Podocoryne nassa n. sp.

A, portion of colony on snail-shell, drawn from preserved material.
B-D, stages in the development of the medusa.

CES

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 309

-in a membranous extension of the perisarc. The older gonophore is more
rounded and distinctly demarcated from the pedicel. The oldest gonophores
present are in the form of young medusae with the diameter approximately
equal to the height (about 0-25 mm.), and with the tentacles still coiled up in
the umbrella cavity. They have become detached from the pedicel but are still
enclosed in a perisarcal membrane.

Medusa with 4 radial canals and a circular, 4 marginal tentacles, a large
hypostome not quite reaching the margin of the bell and bearing 4 oral tentacles
armed with terminal clusters of nematocysts. 7

Nematocysts of 2 kinds:

(1) Undetermined heteronemes, 8-0 X 3:0 pL.

(11) ?Desmonemes, 6°5 X 3°5 p.
Remarks. ‘This is the only species of Podocoryne known to the author without
spines and with gonophores arising separately from the hydrorhiza.

Family Pandeidae
Hydrichthys boycet Warren 1916
Hydrichthys boycet Warren 1916, pp. 172-187, fig. 12; Pl. 17-20.
Records. NA 218.

Description. Several small colonies parasitic on Chaetodon lunula. Structure
exactly as described by Warren. Medusa buds present in various stages of
development.

Family Bougainvilliidae
Bimeria fluminalis Annandale 1915

Fig. 4
Bimeria fluminalis Annandale 1915, p. 111, fig. 10; Pl. 9, figs. 3, 3a. 1917, p. 111, fig. 1.

Record. STL 174 (reported by Day, Millard and Broekhuysen 1954 as Bimeria
sp.).

Description. Rich colonies growing on the aerial roots of mangroves and for the
most part thickly covered with diatoms and other epibiotic forms. Hydrorhiza
forming a matted feltwork. Stem reaching a maximum height of 6-3 cm.,
unfascicled, often slightly geniculate in the distal regions, giving rise to alternate
branches or hydranth pedicels. Occasionally two hydranth pedicels arise from
the same level], either opposite one another or from the same side. Perisarc on
stem and branches fairly thick, with adhering silt and diatoms, annulated at
base of stem, on origin of branches and on part or all of hydranth pedicels;
continued over the base of the hydranth to form a thick ‘pseudohydrotheca’.
Sections show that the pseudohydrotheca terminates abruptly on the base of
the tentacles (fig. 4, H, I), and that the hydranth can be partially retracted into

310 ANNALS OF THE SOUTH AFRICAN MUSEUM i

5 cm.

Ee
E
”

Oo
!
uJ

Fic. 4. Bimeria fluminalis Annandale.

A, portion of colony. B, young male gonophore. C, old female gonophore containing
planula. D, a single hydranth and two young female gonophores, one with an egg.
E, t.s. male gonophore. F, t.s. young female gonophore with egg and spadix.
G, t.s. old female gonophore with planula and remains of spadix. H and I, Ls.
hydranths, I more contracted than H. J, a heterotrichous microbasic eurytele.
(e, egg. ect, ectoderm. f, folded ectoderm. m, male generative cells. , perisarc.
5, adhering silt. sp, spadix of gonophore.)

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 3II

_ the pseudohydrotheca, but that in the process the ectoderm at the margin is
thrown into a distinct fold (f). A very thin layer of perisarc appears to con-
tinue for a short distance over the tentacles, but it is very difficult to detect and
is only visible in isolated regions—there is no definite sheath over the tentacles
as in typical members of the genus. Hydranths with 10-12 tentacles.

Gonophores in the form of fixed sporosacs, borne singly or in clusters on the
hydranth pedicels, male and female on separate stems. Gonophore completely
surrounded by perisarc, with a very short pedicel, and no radial canals or
tentacle rudiments.

Male gonophore ovoid or spherical when young, becoming more elongated
when mature, bearing a mass of spermagenic material around a os strongly
developed, and sometimes curved, spadix.

Female gonophore ovoid, producing a single egg, which is flattened against
the spadix, displacing it to one side and causing it to arch over the egg in the
manner described by Annandale. After fertilization the egg develops in situ.
Later the spadix degenerates completely leaving a large oval planula still
enclosed in a perisarcal membrane.

Nematocysts of 2 types:

(1) Desmonemes, 3°5 X 2°5 p.

(ii) Heterotrichous microbasic euryteles, reaching a maximum size

of 7:0 X4°5 p.

Annandale’ s

Measuremenis (rmm., preserved) STL 174 Type
Material
Stem, diameter ; 2) (0282-60-42 0°13—0°20
Branch, diameter. : . O° 12—0°20 0-09-0°16

Pseudohydrotheca (mostly contracted) :

length. : . 018-038 0°13-0°22
diameter . : 2 02 5-0-92 0°13—-0°29

Gonophores, male: length reaching 0-51 0°52

diameter . reaching 0°30 0°23

Gonophores, female: length . reaching 0°40 0°30

diameter . reaching 0°29 0°24.

Remarks. 1 have seen a sample of Annandale’s type material of B. fluminalis from
Calcutta kindly loaned to me by the Zoological Survey of India, and am satisfied
that the South African material belongs to the same species. The general form
of the colony is the same, and the nature of the gonophores leaves no room for
doubt. Measurements of the type material (a small sample only) are included
above.

The nematocysts of the type material are undischarged, but there are two
types (heteronemes and desmonemes) whose measurements agree closely with
the South African material.

312 ANNALS OF THE SOUTH AFRICAN MUSEUM

I am somewhat doubtful about the inclusion of this species in the genus
Bimerta. According to Rees (1938 and personal communication) Bimeria and
Garveia can be distinguished by the presence of tubular sheaths of perisarc over
the bases of the tentacles in the former and their absence in the latter. The
variable nature of this character in B. fluminalis has been remarked on by
Annandale, and in the South African material the perisarcal sheath is practically
non-existent. Other things being equal it hardly seems a reliable character for
generic diagnosis.

B. fluminalis is known only from brackish and temporarily fresh water
connected with the Bay of Bengal and Gulf of Siam in specific gravities varying
from 1-000 to 102575. It is therefore interesting that it should appear in South
Africa, also in an estuary. Conditions in St. Lucia estuary are, however, rather
different than those described by Annandale, for in times of drought the salinity
may rise above that of the sea. The material was found in the ‘channel’ area
just south of the opening of the Mpate River. At the time the salinity in the
area was 28°/,,, but it has been known to rise to 35-7°/,,, and probably falls
very low when the river is in flood. Whether or not the species can survive
such drastic changes from year to year is unknown.

SUMMARY

A total of 12 species of gymnoblastic hydroids is recorded. Two of these are
new species, namely Hydractinia diogenes, and Podocoryne nassa, and three are
new records for South Africa, namely Corydendrium parasiticum (Linn.), Euden-
drium carneum Clarke, and Bimeria fluminalis Annandale.

REFERENCES

Allman, G. J., 1871-2. A Monograph of the Gymnoblastic or Tubularian Hydroids. London.

Annandale, N., 1915. Fauna of the Chilka Lake. The Coelenterates of the Lake, with an account
of the Actiniaria of brackish water in the Gangetic Delta. Mem. Indian Mus., 5, 65-114.

Annandale, N., 1917. Zoological Results of a Tour in the far East. Hydrozoa and Ctenophora.
Mem. Asiatic Soc. Bengal, 6, 101-117.

Clarke, S. F., 1882. New and interesting Hydroids from Chesapeake Bay. Mem. Boston Soc. Nat.
Fist., 3, 135-142.

Day, J. H., Millard, N. A. H., and Broekhuysen, G. J., 1954. The Ecology of South African
Estuaries. Part IV: The St. Lucia System. Trans. Roy. Soc. S. Afr., 34, 129-156.

Day, J. H., and Morgans, J. F. C., 1956. The Ecology of South African Estuaries. Part 8: The
Biology of Durban Bay. Ann. Natal Mus., 13, 259-312.

Ewer, D. W., 1953. On anew Tubularian Hydroid from Natal. Ann. Natal Mus., 12, 351-357.

Hartlaub, C., 1905. Die Hydroiden der magalhaensischen Region und chilenischen Kiiste.
Kool. Fahrb., Suppl. 6 Fauna Chilensis, 3, 497-714.

Jaderholm, E., 1905. Hydroiden aus antarktischen und subantarktischen Meeren. Wiss. Erg.
Schwed. Stidpolar-Exped. 1901-1903, 5, I-41.

Jaderholm, E., 1923. Hydroids from West and South Africa. Meddel. Géteborgs Mus., Zool.
No. 26, 1-7.

Kirkpatrick, R., 1910. Description of new species of Hydrozoa and Porifera. Ex Cunningham,
J. T.: On the Marine Fishes and Invertebrates of St. Helena. Proc. Zool. Soc. Lond., 1910,
127-130.

Leloup, E., 1937. Hydropolypes et Schyphopolypes recueillis par C. Dawydoff sur les cétes de
’Indochine Frangaise. Mém. Mus. Roy. D’ Hist. Nat. Belgique, (2), 12, 1-73.

HYDROZOA FROM COASTS OF NATAL AND PORTUGUESE EAST AFRICA 313

Millard, N. A. H., 1955. New Species of Hydrozoa from South Africa. Ann. S. Afr. Mus., 41,
E5222.

Millard, N. A. H., 1958. Hydrozoa from the Coasts of Natal and Portuguese East Africa. Part I.
Calyptoblastea. Ann. S. Afr. Mus., 44, 165-226.

Rees, W. J., 1938. Observations on British and Norwegian Hydroids and their Medusae.
j- Mar. Biol. Ass. U. K., 23, 1-42.

Rees, W. J., 1957. Evolutionary Trends in the Classification of Capitate Hydroids and Medusae.
Bull. Brit. Mus. (Nat. Hist.), Zool., 4, 4.55-534-

Ritchie, J., 1907. On Collections of the Cape Verde Islands Marine Fauna, made by Cyril
Crossland, M.A.(Cantab.), B.Sc.(Lond.), F.Z.S., of St. Andrews University, July to
September, 1904. Proc. Zool. Soc. Lond., 1907, 488-514.

Ritchie, J., 1910@. The Marine Fauna of the Mergui Archipelago, Lower Burma, collected by
Jas. J. Simpson, M.A., B.Sc., and R. N. Rudmose-Brown, D.Sc., University of Aberdeen,
February to May 1907.— The Hydroids. Proc. Zool. Soc. Lond., 1910, 799-825.

Ritchie, J., 1910. Hydroids from Christmas Island, Indian Ocean, collected by C. W. Andrews,
Wese., E-R-S., F.Z:S., in 1908. Proc» Zool. Soc. Lond., 1910, 826-836.

Vannucci, M., 1954. Hidrozoa e Scyphozoa existentes no Instituto Cceanografico. II. Bol.
Inst. Oceanogr., 5, 95-138.

Vervoort, W., 1941. Biological Results of the Snellius Expedition. XI. The Hydroida of the
Snellius Expedition (Milleporidae and Stylasteridae excluded). Temminckia, 6, 186-240.

Vervoort, W., 1946. Fauna van Nederland. Afl. 14. Hydrozoa (C1). Leiden.

Vervoort, W., 1946a. Exotic Hydroids in the Collections of the Ryksmuseum van Natuurliyke
Historie and the Zoological Museum at Amsterdam. ool. Meded., 26, 287-351.

Warren, E., 1906. On Halocordyle cooperi, sp. n.; a Hydroid from the Natal Coast. Ann. Naial
Mus., t, 73-81. i

Warren, E., 1907a. Note on the Variation in the arrangement of the Capitate Tentacles in the
Hydroid, Halocordyle cooperi Warren. Ann. Natal Mus., 1, 209-213.

Warren, E., 1908. On a Collection of Hydroids, mostly from the Natal Coast. Ann. Natal Mus.,
1, 269-355. .

Warren, E., 1916. On Aydrichthys boycet, a Hydroid parasitic on Fishes. Aan. Durban Mus., 1,
172-187.

Weill, R., 1934. Contribution a étude des Cnidaires et de leurs Nématocystes, 1 and 2. Trav.
Stat. Kool. Wimereux, 1 and 2, 1-701.

nit i a ¥ at Se AF ae Anat

K a

A
‘ i PHAGE
h
: Fce
‘ } fhe
5
iy
:
4 :
t ; i
,
7
*
’
|
|
)
|
|
,

A REVIEW OF THE SUBSPECIES OF THE YELLOW CANARY,
SERINUS FLAVIVENTRIS (SWAINSON)

J. M. WINTERBOTTOM
South African Museum

INTRODUCTION

Five subspecies of the Yellow Canary, Serinus flaviventris, have been
described: flaviventris (Swainson) from ‘South Africa’; marshall: Shelley from
Potchefstroom; guillarmod: (Roberts) from Sanqubetu Valley, Basutoland;
damarensis Roberts from Windhoek; and aurescens Clancey from between
Brandvlei and Kenhardt. Macdonald (1957) observed that birds from ‘the
limited area of the Sclerophyll Region in the extreme south-west Cape are
slightly darker and may eventually be separated as a distinct race’. Material
recently received in the South African Museum from Mr. R. W. Rankine,
collected near the Kalahari Game Reserve, suggested that the range of damarensis
was wider than current literature indicated. Under the circumstances, it
seemed advisable to assemble the available material and review the whole
species; and to make a special point of obtaining a series of breeding birds from
the restricted type locality of flaviventris, the Berg River (this being the first
locality mentioned by Andrew Smith in his field notes).

Altogether, 152 skins were examined; and my thanks are due to the
following for the loan of material: the Director, Durban Museum & Art
Gallery; the Director, East London Museum; the Director, Kaffrarian Museum;
the Director and Mr. M. P. S. Irwin, National Museum of Southern Rhodesia;
the Director and Mr. R. Liversidge, Port Elizabeth Museum and Snake Park;
the Director and Mrs. Campbell, Transvaal Museum. The work on which this
paper is based was done while I was holding a Senior Bursary of the South
African Council for Scientific and Industrial Research.

REVIEW OF THE MATERIAL

The broad pattern of variation in the Yellow Canary follows expectations
in being lightest in the hot north and darkest in the colder south. Within that
broad pattern, however, several more local patterns are visible.

The birds of the north-east were separated by Shelley as marshalli on the
grounds that the males had slightly larger bills and yellower upper parts while

oS)

316 ANNALS OF THE SOUTH AFRICAN MUSEUM

the females were paler and less heavily streaked below. I have found no signifi-
cant difference, however, in the size of the bill; but the other distinctions hold.

Roberts then separated the north-west population as damarensis, in which
the males were a clearer yellow than those of marshalli and the females were
unstriped below, but often with a yellow wash. This is confirmed in the series
I have examined.

Occupying an area south of the range of damarensis is a population in which
the males are duller and darker above than damarensis but yellower than
marshalli1; and richer yellow below than either; while the females are less green
above and white below. This is aurescens of Clancey.

South of aurescens, occupying the main plateau of the Karoo from east to
west, is another form, the males of which are darker above and paler below
and the females more heavily streaked than any of the preceding subspecies.
This population gives place in the coastal strip from Still Bay to Oranjemund
to another, which is smaller, the males paler, more lemon yellow below, and
the females even more heavily streaked. The last is the typical flaviventris; and
and Karoo population is without a name. I name it below, quzntoni.

From the mountains of Basutoland, Roberts described guillarmod: as being
darker above than /flaviventris in the male and more heavily streaked below in
the female, in which, too, the sides of the face are darker grey. He also noted
that it, and a bird from Carnarvon, were larger than coastal birds from Tulbagh
to Vanrhynsdorp. Only three males (one of them immature) and five females
were available to me, but the heavy streaking below of the females is broadly
confirmed by these specimens, which are more heavily streaked than quzntont,
but no more so than flaviventris. They also appear to have less green suffusion
above than either qguintoni or flaviventris. The two adult males, however, appear
to me to be indistinguishable from flaviventris, except on size.

CONCLUSIONS

The following subspecies are accordingly recognized:

I. Serinus flaviventris flaviventris (Swainson)

Loxia flaviventris Swainson, ool. Fourn., 3, 1828: 348—South Africa
(restricted to Berg River, S.W. Cape).

Range. The coastal strip from Oranjemund, just north of the mouth of
the Orange River, to Still Bay, Riversdale District.

Characters. ‘The male is dark above, like quintoni, from which it differs in
being paler yellow, more lemon and less orange, below. The female is also
similar above to quintoni but more heavily streaked below. Both sexes average
smaller than the other forms.

Measurements :

23 gd: Wing, 68-76, av. 71°6 mm.; tail, 46-56, av. 49°5; tarsus, 15-22,
av. 18:8; culmen, 13-14, av. 13:8.

Sanaworly 009 0S OOL JaawoTy

IAIN

mS Se OOF od¢
—_=
cf re 3
By
&
=I
ssf
oa
oO
Sos re
i (0)
ro)
fe)
ro)
S
is}
gj

S7@

°
acer a

4. damarensis
5. marshalli
6. guillarmodi_

1. flaviventris
2. quintoni
3. aurescens

Type localities of described forms are named.

DISTRIBUTION OF THE SUBSPECIES OF Serinus flaviventris (Swains.)

REVIEW OF THE SUBSPECIES OF THE YELLOW CANARY
Each dot represents a locality from which one or more specimens

318 ANNALS OF THE SOUTH AFRICAN MUSEUM

10 99: Wing, 66-72, av. 68-9 mm.; tail, 49-58, av. 51:2; tarsus, 17-21,
av. 18:9; culmen, 13-15, av. 14:0.

It will be noted that there is a considerable overlap in wing-lengths
between this form and the next; but the males with wing-lengths in excess of
73 mm. and the females with wing-lengths in excess of 70 mm. are all from the
eastern edge of the range—Springbok, Monazite Mine, Berg River; except
one, from L’Agulhas.

Material examined:

South African Museum: 24 (Berg River, Cape Town, Swartklip, Zoetendals-
viei, Still Bay, Gape Flats, Wynberg, Langebaan, Leipoldtville, Monazite
Mine, Annisfontein).

East London Museum: 7 (Malmesbury, Springbok, Sandkraal, Vanrhynsdorp-—
Bitterfontein).

Port Elizabeth Museum: 2 (Oranjemund).

National Museum of Southern Rhodesia: 1 (Langebaan).

2. Serinus flaviventris quinton subsp. nov.

Range. The Karoo areas from Calvinia, the Cold Bokkeveld and Touws
River east to Skietkuil, Murraysburg District, and Fish River Station. Inter-
grades with marshall: at Elandshoek, Aliwal North District, and with aurescens
at Vanwyksvlei.

Characters. The male is dark above, like flaviveniris, from which it differs in
being richer yellow, more orange and less lemon, below. The female is also
similar above to flaviventris but less heavily streaked below. Both sexes are larger
than flaviventris. From marshalli and aurescens, the male differs in being more
heavily streaked and less yellow above and not so rich a yellow as aurescens
below. The female is darker above and more heavily streaked below than either
marshallt or aurescens.

Type. In the South African Museum, Cape Town; 4, Hillmore, Beaufort
West, 22 September 1955, collected by J. M. Winterbottom; collector’s —
number, 250; S.A.M. number, 20188.

Measurements :

25 gd: Wing, 71-82, av. 75°38 mm.; tail, 49-59, av. 53°2; tarsus, 16-23,
av. 18:9; culmen, 13-15, av. 14°1.

799: Wing, 71-79, av. 73°9 mm.; tail, 49-52, av. 50°2; tarsus, 16-20,
av. 18:3; culmen, 13-15, av. 13°9.

Material examined:

South African Museum: 7 (Touws River, Hanover, Beaufort West (type),

Cold Bokkeveld).

East London Museum: 23 (Calvinia, Fraserburg, Williston, Brandvlei,

Brandvlei-Kenhardt, Touws River, Skietkuil, Fish River Station).
Durban Museum: 1 (Calvinia).

REVIEW OF THE SUBSPECIES OF THE YELLOW CANARY 319

Rankine Collection:* 1 (Beaufort West).

Note. I name this subspecies after my friend Mr. W. F. Quinton, a keen
ornithologist, on whose farm the type was collected.

3. Serinus flaviventris aurescens Clancey

Serinus flaviventris aurescens Clancey, Durb. Mus. Novtt., 5, 1958: 104-10
miles from Kenhardt, on Brandvlei Road.

Range. From Kenhardt and Olyvenhout Drift, east and north to Riverton,
near Kimberley, and Kuruman; intergrading with quintom at Vanwyksvlei,
with marshall: at Riverton and with damarensis at Olyvenhout Drift and Kuruman.

Characters. Male less heavily streaked and lighter above than quintonz;
female, lighter above and less heavily streaked below. From marshalli, the male
is distinguished by being yellower and less streaked above and richer, deeper
yellow below; the female is greyer above and whiter, less streaked, below.

Measurements:

8 gd: Wing, 73-77, av. 74°7 mm.; tail, 50-56, av. 53:1; tarsus, 16-20,
. av. 17°13; culmen, 14-15, av. 14°.
299: Wing, 72-73, av. 72°5mm.; tail, 47-52, av. 49°5; tarsus, 16, av. 16-0;
culmen, 14, av. 14:0.
Material examined:

Durban Museum: 5 (Riverton, Kenhardt (type), Niekerkshoop, Niekerkshoop—
Griquatown).

East London Museum: 4 (24 miles south of Kakamas, Klipput, Kuruman-—
Askham).

National Museum of Southern Rhodesia: 1 (Kuruman).
Intermediates, quintoni-aurescens :

Durban Museum: 4 (Vanwyksvlei).
Intermediates, aurescens-damarensis:

South African Museum: 2 (Olyvenhout Drift).

4. Serinus flaviventris damarensis Roberts

Serinus flaviventris damarensis Roberts, Ann. Tvl. Mus., 8, 1922: 264—
Windhoek.

Range. South West Africa (except the area at the mouth of the Orange
River), Bechuanaland Protectorate, the extreme north-west corner of Southern
Rhodesia and the extreme north and west of Griqualand West; intergrading
with aurescens at Olyvenhout Drift and Kuruman.

Characters. ‘The male is brighter and lighter above and paler yellow below
than aurescens; and a clearer, brighter yellow below than marshalli. The female

* The Rankine collection will be divided between the South African Museum and the National
Museum of Southern Rhodesia.

- ¢
4
ae
%

320 ANNALS OF THE SOUTH AFRICAN MUSEUM

is greener above and yellower below than aurescens; and yellower and less —
striped below than marshall.
Measurements:
19 gd: Wing, 71-78, av. 75:2 mm.; tail, 47-58, av. 50-7; tarsus, 17-20,
av. 18-1; culmen, 13-16, av. 14:0.
729: Wing, 70-77, av. 73°1 mm.; tail, 50-55, av. 52°23: (farcue -1O,
av. 16-7; culmen, 12-14, av. 13°1.
Material examined:

National Museum of Southern Rhodesia: 18 (Kakia, Ghanzi, 200 miles south
of Francistown, Tsabong, 55 miles west of Kanye, Tsane, Lehututu,
Molepole-Lephepe, 23 miles south of Letlaking). |

Transvaal Museum: 5 (Windhoek, Okahandja, Matetsi).

Durban Museum: 2 (Otjomassu).

South African Museum: 1 (Keetmanshoop).

East London Museum: 1 (Kalahari Game Reserve).

Rankine Collection*: 8 (Noeniput).

5. Serinus flaviventris marshall: Shelley
Serinus marshallt Shelley, Bds. Afr., 3, 1902: 200— Potchefstroom.

Range. Western Transvaal, Orange Free State and the adjacent parts of
the Cape Province (Fourteen Streams, Kimberley, Hopetown, Colesberg) ;
intergrading with aurescens at Riverton and with quintont at Elandshoek, Aliwal
North District.

Characters. ‘The male is greener and less heavily streaked above than
quintom; and the female is paler and with the underparts less heavily streaked.
The male is greener, less yellow, above than aurescens and damarensis, and paler
yellow below; the female is less streaked below than damarensis and greener
above than aurescens.

Measurements : |
26 gg: Wing, 70-78, av. 75:1 mm.; tail, 47-59, av. 52°8; tarsus, 15-20,
av. 18-3; culmen, 13-16, av. 14:0. |
6 99: Wing, 70-77, av. 73°2 mm.; tail, 51-54, av. 52°53; tarsus, 17—20,
av. 18-8; culmen, 14, av. 14:0.
Material examined:
South African Museum: 5 (Orange River Station, Potchefstroom).
Transvaal Museum: 12 (Brandfort, Bloemfontein, Wolmaransstad, Venters-
kroon, Bothaville, Fourteen Streams, Rustenburg, Hoeningspruit).
East London Museum: 10 (Rooipoort, Bloemfontein).
Durban Museum: 3 (Riverton, Glen).

* The Rankine collection will be divided between the South African Museum and the National
Museum of Southern Rhodesia,

REVIEW OF THE SUBSPECIES OF THE YELLOW CANARY 321

Port Elizabeth Museum: 1 (Bloemfontein).
_Kaffrarian Museum: 1 (Colesberg).
National Museum of Southern Rhodesia: 1 (Rustenburg).

6. Serinus flaviventris guillarmod: (Roberts)

Serinops flaviventris guillarmod: Roberts, Ann. Tvl. Mus., 18, 1936: 256—
Sanqubetu Valley, Basutoland.

Range. The high mountains of Basutoland (S. f. marshalli occurs along the
Caledon River valley).

Characters. Broadly speaking, a flaviventris swelled to the size of quintoni;
but the upper parts of the female are darker, with less green wash, than those
of flaviventris. It averages larger than any of the other races.

Measurements :

246: Wing, 78-79, av. 78:5 mm.; tail, 50-55, av. 52°5; tarsus, 19, av. 19°0;
culmen, 13-14, av. 13°5.

5 22: Wing, 76-80, av. 77°70 mm.; tail, 53-58, av. 56-0; tarsus, 18-20,
av. 18°8; culmen, 14-15, av. 14°4.
Material examined:

Transvaal Museum: 3 (Sanqubetu Valley (Type), Maluti Mts.).
Durban Museum: 5 (40 miles east of Maseru).

REFERENCE

Macdonald, J. D. 1957. A Contribution to the Ornithology of Western South Africa, 1957.

a, hi ai a ar
Re Gi iii i it) | e ie
jak aay ih a) me Digetae e
pie COs Ny ri" Rr a «ti aria, $

POE EE ER Cie Tn) iO Rat letinaetnd

a* , 1
r a. : t ’
s atk hee Prod 4 iS i, SATS OGY. im

s a Z i . | i i

ta PSE CRE, RATS he op
EMD or Ley ihe hn ae a)

~~ i
Pat { As ty ‘
% ta
‘ # [: 4
j
i
Lh ic
~
af a Pa
> " ‘ F
?
i pits i ae
=)
?
ie
/
i ;
er
; \
j
b ‘
,

The ANNALS OF THE SOUTH AFRICAN MUSEUM are issued in parts at irregular

intervals as material becomes available.

Some volumes and parts are out of print, and others are only sold as parts ofa set, or valine

‘ respectively. The prices of parts published prior to 1940 have been increased.

Out of print: Vols. I, II (Parts 1-3, 5,7, 8), III (Part 1), V (Parts 1, 2, 5,8, 9), VI (Bare me

Index), VII (Parts 1, 2, 3, 5), VIII, 1X (Parts 1, 2), X (Part 2), XI (Parts 2, 7, Index), XXI
(Part 1), XXIV (Part 2), XXX (Parts I-3).

Vol.
aT,
III.
IV.
Vis

VI.
VII.
IX.

XI.
XII.
XIII.
XIV.
XV.
XVI.
XVIT.
XVIII.
XIX.
XX.
XXII.
XXIT,
XXIII.
XXIV.
XXV.
XXVI.
XXVIT.
AXVIILI.
XXIX.
XXX.
INDEX
XXXI.
XXXII.
SXOCXITI.

XXXIV.

KKK,
SICK IL.
XXXVII.
XXXVIIL
SOCK IX,
xc,

XI.

XLII.

. XLIV.

1900-1902
1903-1905
19031908
1906-1910

1908-1910
1908-1913
IQII—1918
IQLI-I914
IQII-1918
1913-1924
1913-1923
IG15-1924
IQ14—1916
1917-1933
1917-1920

1921
1924-1925
1924-1926
1925-1927
1925-1928
1925-1926
1929-1938
1927-1928

1928

1929
1929-1932
1929-1931!
1931-1935

(excl. Parts 1-3, 5, 7, 8)
HOY (ee se! Part. r)

Zoology and Geology

Zoology
Palaeontology

Geology, Palaeontology, Zoology, Anthropology (excl

Parts 1, 2, 5, 8, 9)
Zoology
Palaeontology
Botany
Zoology
Zoology
Palaeontology and cane
ede and ca
Zoology ci
Zoology
Botany
Zoology
Zoology a5 “3 is ee oe
Zoology .. a) we Me ae ¥ Ry
Zoology eA Ae we
Zoology eS ..(excl. Part 1)
PAlaconiolosy, ae be A te oie a
Zoology :
Anthropology and Ldhnology
Zoology
Zoology ie ae An Bs se ve
Anthropology hs ia a Big ps he
Palaeontology As 44 aie Sy Sie ot
Zoology
Zoology

veda Part 1 iy ‘inde

. (Parts 4 and 6 only)
mene Parts 1 and 2)

. (excl. Part 2)

(xd: Pan Digs Index)

of papers, authors, and ec raven ae in Vel LXXX

1934-1950
1935-1940
1939
5938
1956
1942-1948
1947-1952
1950
1952

195271956
| £:1952-1955 ,
XLII.

1953-1956

4 1955-1957

1957-

Palaeontology
Zoology
Zoology
Zoology
Zoology
Zoology
Archaeology
Zoology
Zoology
Botany
Zoology .

. (Part 4 only)

‘ Palaeontology;

Zoology and Palaeontology : o's
Parts 1-7, Zoology and Palaeontology .

Copies may be obtained from—

The LIBRARIAN, Sourn Arrican Museum, Cape Town. \

..(excl. Part 2)

I
I

tS)

HNYNOP KRDO RN ONNW COW NN se we ew De Ne DO KRWO KOO AR PAWN

S.
2

_ Lead 5 ‘
Ou Oo 6 SF 09

Sor
COMSMHRGCGDAPOFDOCONWOMOSCHWSOGCAGCOUWOWW OWMwWSAT OL

—

woof

Pe ININ A LS

OF THE

SOUTH AFRICAN MUSEUM

VOLUME XLIV

PART IV (supplement), containing the References of:—

CONTRIBUTIONS TO THE KNOWLEDGE OF SOUTH AFRICAN
MARINE: MOLLUSCGA.
PART I. GASTROPODA: PROSOBRANCHIATA: TOXOGLOSSA

K. H. BARNARD

ISSUED NOVEMBER 10969 PRICE 30c

| PRINTED FOR THE
TRUSTEES OF THE SOUTH AFRICAN MUSEUM

BY THE RUSTICA PRESS (PTY.) LIMITED, COURT ROAD, WYNBERG, CAPE

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

Die ANNALE VAN DIE SUID-AFRIKAANSE MUSEUM

word uitgegee in dele op ongereelde tye na beskikbaarheid
van stof

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

OUT OF PRINT/UIT DRUK

1, 2(1, 3, 5, 7-8), 3(1-2, 5, t.-p.i.), 5(2, 5) 7-9),

6(1, t.—p.i.), 7(1, 3), 8, g(I-2), 10(1-3),

DI(I—2, “Fanvte Pal.) 12h.) 242) 2 75 oh i—3) aos
44(4).

Price of this part/Prys van hierdie deel

30c

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

CONTRIBUTIONS TO THE KNOWLEDGE OF SOUTH AFRICAN
MARINE MOLLUSCA

PART I. GASTROPODA: PROSOBRANCHIATA: TOXOGLOSSA

By
tK. H. BARNARD

REFERENCES

(Compiled by Brian Kensley, South African Museum)

Apams, A. & REEVE, L. 1848-50. The zoology of the voyage of H.M.S. SAMARANG . . . during the
years 1843-46. Mollusca. London: Reeve, Benham & Reeve.

BARNARD, K. H. 1950. The dates of issue of the ‘Illustrations of the Zoology of South Africa’
and the ‘Marine Investigations in South Africa’. 7. Soc. Bibl. nat. Hist. 2: 187-189.

BARNARD, K. H. 1957. Glypteuthria capensis: a generic correction. 7. Conch. 24: 210.

BartscH, P. 1915. Report on the Turton collection of South African marine mollusks, with
additional notes on other South African shells contained in the United States National
Museum. Bull. U.S. nat. Mus. gt: 1-305.

BeRGH, R. 1895. Beitrage zur Kenntniss der Coniden. Nova Acta Leop. Carol. 65: 67-214.

BOLTEN-RODING see RODING.

BraGA, J. M. 1952. Materias para o estudo da fauna malacolédgica de Mocambique. Ann. Fta
Miss. geogr., Lisboa 7: 63-127.

Cooke, A. H. 1895. Cambridge natural history 3. Molluscs. London: Macmillan.

Cooks, A. H. 1919. The radula in Thais, Drupa, Morula, Concholepas, Cronia, Iopas, and the
allied genera. Proc. malac. Soc. Lond. 13: 91-109.

Crosse, H. 1858. Observations sur le genre cone et description de trois espéces nouvelles, avec
un catalogue alphabétique des cénes actuellement connus. Rev. Mag. Zool. 10: 113-127,
150-157, 199.

DAUTZENBERG, P. 1912. Mission Gruvel sur la céte occidentale d’Afrique (1909-1910).
Mollusques marins. Ann. Inst. océanogr. Monaco 5(3): 1-111.

DAUTZENBERG, P. & FiscHER, H. 1906. Mollusques provenant des dragoges effectués a l’ouest
de l’Afrique pendant les campagnes scientifiques de S.A.S. le Prince de Monaco. Résult.
Camp. sci. Monaco 32: 1-125.

DEsHAYES, G. P. 1859. A general review of the genus Terebra and a description of new species.
Proc. zool. Soc. Lond. 1859: 270-321.

GouLp, A. A. 1860. Descriptions of shells collected in the North Pacific Exploring Expedition.
Proc. Boston Soc. nat. Hist. 7: 323-340.

GRAVELY, F. H. 1942. Shells and other animal remains found on the Madras beach. II. Snails,
etc. (Mollusca, Gastropoda). Bull. Madras Govt. Mus. (n.s. Nat. Hist.) 5(2): 1-110.

Hinps, R. B. 1843. Descriptions of new shells from the collection of Captain Belcher. Ann. nat.

| Mist. 11: 16-21, 255-257.

Hinps, R. B. 1844. The zoology of the voyage of H.M.S. suLPHUR . . . during the years 1836-42. 2.
Mollusca. London: Smith, Elder.

ILLUSTRATIONS OF THE ZOOLOGY OF THE ROYAL INDIAN MARINE SuRVEY SHIP ‘INVESTIGATOR’.
1898, 1907. Mollusca, Part 2. Mollusca, Part 4. Calcutta: Indian Museum.

KrenER, L. C. 1835-49. Spécies général et iconographie des coquilles vivantes, comprenant la collection du
Muséum d’ Histoire Naturelle de Paris... Livr. 10-16, 59-71, 124-137. Paris.

KNupDsEN, J. 1952. Marine prosobranchs of tropical west Africa collected by the ‘Atlantide’
Expedition, 1945-46. Vidensk. Medd. dansk. naturh. Foren. Kbh. 114: 129-185.

KNupDsEN, J. 1956. Marine prosobranchs of tropical west Africa (Stenoglossa). Aélantide Rep.
4: Q-II0.

Kocu, H. J. 1949. A review of the South African representatives of the genus Patella Linnaeus.
Ann. Natal Mus. 11: 487-517.

Krauss, F. 1848. Die siidafrikanischen Mollusken. Ein Beitrag zur Kenntniss des Kap- und Natallandes
und zur geographischen Verbreitung derselben, mit Beschreibung und Abbildung der neuen Arten.
Stuttgart: Ebner & Seubert.

LAMARCK, J. B. P. 1798. Tableau encyclopédique . . . des trois régnes de la nature. Vingt-uniéme partie.
Mollusques testacés. Paris: Agasse.

Lamarck, J. B. P. 1810. Suite de la détermination des espéces de mollusques testacés: continua-
tion du genre Porcelaine et la genre Ovule. Ann. Mus. Hist. nat., Paris. 16: 89-114, 300-328.

Lamarck, J. B. P. 1815-1822. Histoire naturelle des animaux sans peti Paris.

Martens, E. von. 1902. Einige neue Arten von Meer-Conchylien aus den Sammlungen kes
deutschen Tiefsee-Expedition. S.B. Ges. naturf. Fr. Berl. 1902: 237-244.

Martens, E. von. 1903. Die beschalten Gastropoden der deutschen Tiefsee-Expedition 1898—
1899. A. Systematisch-geographischer Teil. Wiss. Ergebn. ‘Valdivia’. 7: 1-146.

Martin, A. R. H. 1956. The ecology and history of Groenvlei. S. Afr. 7. Sct. 52: 187-192.

MELVILL, J. C. 1917. A revision of the Turridae (Pleurotomidae) occurring in the Persian Gulf,
Gulf of Oman and north Arabian Sea, as evidenced mostly through the results of dredging
carried out by Mr. F. W. Townsend, 1893-1914. Proc. malac. Soc. Lond. 12: 140-201.

ME LviLL, J. C. 1923. Descriptions of twenty-one species of Turridae (Pleurotomidae) from
various localities in the collections of Mr. E. R. Sykes. Proc. malac. Soc. Lond. 15: 162-171.

MELVILLE, J. C. & STANDEN, R. 1901. The Mollusca of the Persian Gulf, Gulf of Oman, and
Arabian Sea. Proc. zool. Soc. Lond. 1901: 327-460.

Opun_er, N. H. 1923. Contribution to the marine molluscan faunas of South and west Africa.
Géteborgs Vetensk.Samh. Handl. 26('7): 1-39.

PEILE, A. J. 1939. Radula notes. 8. Conus. Proc. malac. Soc. Lond. 23: 348-355.

Perry, G. 1811. Conchology, or a natural history of shells: containing a new arrangement of the genera
and species. London.

ReeEvE, L. A. 1843-78. Conchologia iconica. London.

REEVE, L. 1845. Descriptions of eighty-nine new species of Mitra, chiefly from the collection of
H. Cuming, Esq. Proc. zool. Soc. Lond. 1845: 45-61.

Ropine, P. F. 1798. Museum Botienianum;. . . Pars secunda continens conchylia. Hamburg.
Sars, G. O. 1878. Bidrag til Kundskaben om Norges artiske fauna. 1. Mollusca regionis articae Norvegiae.
Christiania.

ScHepMAN, M. M. 1913. The Prosobranchia of the Siboga-Expedition. 5. Toxoglossa. Siboga
Exped. monogr. 491°: 365-452.

SmitrH, E. A. 1871. A list of species from west Africa, with descriptions of those hitherto unde-
scribed. Proc. zool. Soc. Lond. 1871: 727-739.

Situ, E. A. 1877. Diagnoses of new species of Pleurotomidae in the British Museum. Ann.
Mag. nat. Hist. (4) 19: 488-501.

SmitH, E. A. 1882. Diagnoses of new species of Pleurotomidae in the British Museum. Ann.
Mag. nat. Hist. (5) 10: 206-218.

SmiTH, E. A. 1884. Mollusca. In Report on the zoological collections made in the Indo-Pacific Ocean
during the voyage of H.M.S. ‘Alert? 1881-2: 34-116. London: British Museum.

SmitH, E. A. 1895. Natural history notes from H.M. Indian Marine Survey steamer “Investi-
gator’... Series II, no. 19. Report upon the Mollusca dredged in the Bay of Bengal and
the Arabian Sea during 1893-94. Ann. Mag. nat. Hist. (6) 16: 1-19.

SmitH, E. A. 1896. Natural history notes from H.M. Indian Marine Survey steamer ‘Investi-
gator’... Series II, no. 22. Descriptions of new deep-sea Mollusca. Ann. Mag. nat. Hist.
(6) 18: 367-375.

Smitu, E. A. 1902. Descriptions of new species of marine shells from South Africa. 7. Conch.
10: 248-319.

SmitH, E. A. 1903. A list of species of Mollusca from South Africa, forming an appendix to
G. B. Sowerby’s ‘Marine shells of South Africa’. Proc. malac. Soc. Lond. 5: 354-402.

Smitu, E. A. 1904a. Natural history notes from H.M. Indian Marine Survey steamer ‘Investi-
gator’. Series III, no. 1. On Mollusca from the Bay of Bengal and the Arabian Sea. Ann.
Mag. nat. Hist. (7) 13: 453-473.

Sir, E. A. 19045. On a collection of marine shells from Port Alfred, Cape Colony. 7. Malacol.
II: 21-44.

lil

SmiTH, E. A. 1904c. Natural history notes from H.M. Indian Marine Survey steamer ‘Investi-
gator’. Series III, no. 1. On Mollusca from the Bay of Bengal and the Arabian Sea. Ann.
Mag. nat. Hist. (7) 14: 1-14.

SmiTH, E. A. 1906a. Natural history notes from H.M. Indian Marine Survey steamer ‘Investi-

gator’. Series III, no. 10. On Mollusca from the Bay of Bengal and the Arabian Sea. Ann.
Mag. nat. Hist. (7) 18: 245-264.

SmitH, E. A. 19064. On South African marine Mollusca, with descriptions of new species.
Ann. Natal Mus. 1: 19-71.

SmirH, E. A. 1910. On South African marine Mollusca, with descriptions of new species. Ann.
Natal Mus. 2: 175-2109.

SmitTH, E. A. 1914. Descriptions of some South African marine shells. Ann. Natal Mus. 3: 1-6.

SowErRBy, G. B. 1842-87. Thesaurus conchyliorum, or Monographs of genera of shells. London:
Sowerby.

Sowersy, G. B. 1870. Descriptions of forty-eight new species of shells. Proc. zool. Soc. Lond.
1870: 249-259.

SoweErsy, G. B. 1873. Descriptions of five new cones. Proc. zool. Soc. Lond. 1873: 145-146.

SOwERBY, G. B. 1886. Marine shells of South Africa, collected at Port Elizabeth, with descrip-
tions of some species. 7. Conch. 5: 1-13.

Sow_ERsBy, G. B. 1888. Descriptions of sixteen new species of shells. Proc. zool. Soc. Lond. 1888:
207-213.

SOWERBY, G. B. 1889. Some further notes on marine shells collected at Port Elizabeth, South
Africa, with descriptions of some new species. 7. Conch. 6: 6-15.

SOWERBY, G. B. 1892. Marine shells of South Africa. A catalogue of all the marine species with reference
to figures in various works, descriptions of new species, and figures of such as are new, little known, or
hitherto unfigured. London: Sowerby.

SowERsBy, G. B. 1893. Descriptions of fifteen new species of shells of the family Pleurotomidae.
Proc. zool. Soc. Lond. 1893: 487-492.

SOwERBY, G. B. 1897. Appendix to Marine shells of South Africa. London: Sowerby.

SowERBy, G. B. 1898. Description of a new South African marine gasteropod. Mar. Invest. S.
Afr. 1: 5-7.

SOWERBY, G. B. 1901. On seven new species of marine Mollusca collected by Dr. H. Becker at
“The Kowie’, South Africa. Proc. malac. Soc. Lond. 4: 213-215.

SOWERBY, G. B. 1903. Mollusca of South Africa. Mar. Invest. S. Afr. 2: 213-232.

SOWERBY, G. B. 1911. Description of a new species of the genus Conus from South Africa. Proc.
malac. Soc. Lond. 9: 352.

SOWERBY, G. B. 1921. New shells from Port Alfred, collected by Lieut.-Col. W. H. Turton.
Proc. malac. Soc. Lond. 14: 125-127.

THIELE, J. 1903. Die beschalten Gastropoden der deutschen Tiefsee-Expedition 1898-1899.
B. Anatomisch-systematische Untersuchungen einiger Gastropoden. Wiss. Ergebn. ‘Valdivia’
7: 148-180.

THIELE, J. 1925. Gastropoda der deutschen Tiefsee-Expedition. II. Teil. Wiss. Ergebn. ‘Valdivia’
Beye) 97 —9O2.

THIELE, J. 1929-31. Handbuch der systematischen Weichtierkunde. 1. Jena: Fischer.

THIELE, J. & JAECKEL, S. 1931. Muscheln der deutschen Tiefsee-Expedition. Wiss. Ergebn.
‘Valdivia’ 21: 161-268.

Tomuin, J. R. LE B. 1922. Some remarks on the Cape marine province. 7. Conch. 16: 255-262.

Tomtin, J. R. LE B. 1925. Reports on the marine Mollusca in the collections of the South African
Museum. I. Turritellidae. Ann. S. Afr. Mus. 20: 309-316.

Tomuin, J. R. LE B. 1926. On South African marine Mollusca, with descriptions of new species.
Ann. Natal. Mus. 5: 283-301.

Tomun, J. R. LE B. 1927. Reports on the marine Mollusca in the collections of the South
African Museum. II. Families Abyssochrysidae, Odcorythidae, Haliotidae, Tonnidae.
Ann. S. Afr. Mus. 25: 77-83.

Tomiin, J. R. LE B. 1928. Reports on the marine Mollusca in the collections of the South African
Museum. III. Revision of the South African Nassariidae (olim Nassidae). IV. Families
Terebridae, Columbariidae, Thaididae, Architectonicidae. Ann. S. Afr. Mus. 25: 313-335.

Tomi, J. R. LE B. 19314. On South African marine Mollusca, with descriptions of new genera
and species. Ann. Natal Mus. 6: 415-450.

ToMLIn, J. R. Le B. 19315. Reports on the marine Mollusca in the collections of the South
African Museum. V. The Scaphopoda. Ann. S. Afr. Mus. 29: 337-340.

Tomiin, J. R. LE B. 1932. Reports on the marine Mollusca in the collections of the South African
Museum. VI-VIII. Ann. S. Afr. Mus. 30: 157-169.

ToMLin, J. R. LE B. 1937. Catalogue of recent and fossil cones. Proc. malac. Soc. Lond. 22: 205-330.

Tomun, J. R. te B. & STEPHENSON, T. A. 1942. South African Patellidae. Proc. malac. Soc.

Lond. 25: 4-9.
Turton, W. H. 1932. The marine shells of Port Alfred, South Africa. Oxford: Oxford Univarsity
Press.

Watson, R. B. 1881. Mollusca of the ‘Challenger’ Expedition. Part VIII. Pleurotomidae.
J. Linn. Soc.(Zool.) 15: 388-412.

Watson, R. B. 1886. Report on the Scaphopoda and Gasteropoda collected by H.M.S. Challenger
during the years 1873-1876. Rep. Voy. Challenger 1873-76. 15(2): i-iii, 1-756.

WErnKaurFr, H. C. 1875. Systematisches Conchylien-Cabinet von Martini und Chemnitz. Neu herausgegeben

. von H. C. Kuster... Bd. 4, Abth. 2, Lief. 233-244; 309-412. Coneae, oder Conidae.

I. Conus. Niirnberg.

Yen, T.-C. 1942. A review of Chinese gastropods in the British Museum. Proc. malac. Soc.
Lond. 24: 170-289.

INSTRUCTIONS TO AUTHORS

Based on

CONFERENCE OF BIOLOGICAL EDITORS, COMMITTEE ON FORM AND STYLE. 1960.

Style manual for biological journals. Washington: American Institute of Biological Sciences.

MANUSCRIPT

To be typewritten, double spaced, with good margins, arranged in the following order:
(1) Heading, consisting of informative but brief title, name(s) of author(s), address(es) of
author(s), number of illustrations (plates, figures, enumerated maps and tables) in the article.
(2) Contents. (3) The main text, divided into principal divisions with major headings; sub-
headings to be used sparingly and enumeration of headings to be avoided. (4) Summary.
(5) Acknowledgements. (6) References, as below. (7) Key to lettering of figures. (8) Explana-
tion to plates.

ILLUSTRATIONS

To be reducible to 4$ in. x 7 in. (7$ in. including caption). A metric scale to appear with
all photographs.

REFERENCES

Harvard system (name and year) to be used: author’s name and year of publication given
in text; full references at the end of the article, arranged alphabetically by names, chronologi-
cally within each name, with suffixes a, b, etc. to the year for more than one paper by the
same author in that year.

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

For journal articles give title of article, title of journal in italics (abbreviated according to
the World list of scientific periodicals. 4th ed. London: Butterworths, 1963), series in
parentheses, volume number, part number (only if independently paged) in parentheses,
pagination.

Examples (note capitalization and punctuation)

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

FiscHER, P.-H. 1948. Données sur la résistance et de le vitalité des mollusques. 7. Conch., Paris
88: 100-140.

FiscHER, P.-H., Duvau, M. & Rarry, A. 1933. Etudes sur les échanges respiratoires des littorines.
Archs Zool. exp. gén. 74: 627-634.

Koun, 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. 19605. 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. Jn Schultze, L.
Koologische und anthropologische Ergebnisse einer Forschungsreise im westlichen und zentralen Siid-
Afrika. 4: 269-270. Jena: Fischer. Denkschr. med.-naturw. Ges. Jena 16: 269-270.

ZOOLOGICAL NOMENCLATURE

To be governed by the rulings of the latest International code of zoological nomenclature issued
by the International Trust for Zoological Nomenclature (particularly articles 22 and 51).
The Harvard system of reference to be used in the synonymy lists, with the full references
incorporated in the list at the end of the article, and not given in contracted form in the
synonymy list.
Example
Scalaria coronata Lamarck, 1816: pl. 451, figs 5 a, b; Liste: 11. Turton, 1932: 80.

is oe cP
; Ve an
Bathe 4
; My -R TM! ic.
E- . Fee Oy alah el
ae LTRS SUiwe 9 Ml ay au ae gre
i hee as ve sak

$4 "ss * fa)

A ASP ee wg, Pere rs. | PO Re

Whi. 5 grea OOP: aay

i Wy :

J
PRS!

‘is
cS, a
uutumiiee ¢ i
fad i i
i att
: v

pLite rs

Sls

erent
SE
Sano a

Zhe strats

SJ,

ht
et
eee

)
ay.
oe

oe
Rg

yh
it,
hi
wet
tt

ant

rat ia i
i
ap

erry
md reret
re
css catyt gs
TNBON
are

it
ars
Wy

a
we

Fase
Hat

ae
HI

Serr
ae, inca
(ies

aye } i \ Nes “yor ait

S capes Ch Sp aoe EY

Ee ae Ysa pany
Cee Eee SS
sinatonee” i Se

©)

)

“e, “ sill —

on ee
“ill

le
| Hie Ht i] j
| QD
j Mi | |
wal TM & sh

t
|i!
Bh a nae inicae sa “>

iti i |
aa
O

it
| ff i i
l|
Rll

4 i
r “ zal

ao f eee rT Pita
Kg, gps,

om 4° .

ae
2
ee
mp
oS
24 |

Thy
inte
SF

Ptrteys)
7 *
mt

act
ace

rf sa
=
eo.
=
La

ty,

+ 2
Ww Ay

?
gf
of

asaniis oe ; ee pea : “a0 , ; a xp
Pati . P: ered ge fi Poa Si Ss. fr

2 def bul- lg oc La ; ft § oH ft. ery cat ‘ “sl i
aN fe : Dh % & GOP
Bs: ‘ the — fy : bas (i

i
Hi

3 Be a & F z iY a a. : Sa
! 3 rat ns E; <p 4 gh ti _ Lap tiles Fp g f 4 hi ae
Nh Ean pee ro ys

seriou nae

mat

sealer

“logs

SMITHSONIAN INSTITUTION LIBRARIES ‘
ITA
3 9088 01206 5785

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Annals of the South African Museum = Annale van die Suid-Afrikaanse Museum"

See other formats