'1 7?

\ 01?

BULLETIN OF

THE BRITISH MUSEUM
(NATURAL HISTORY)

ZOOLOGY

Vol. 30
1976-1977

BRITISH MUSEUM (NATURAL HISTORY)
LONDON: 1977

DATES OF PUBLICATION OF THE PARTS

No. I . . . . -24 June 1976

No. 2 . . . . . -24 June 1976

No. 3 . . . . -23 September 1976

No. 4 . . . . -23 September 1976

No. 5 ..... 28 October 1976

No. 6 . . . . -25 November 1976

No. 7 . . . . -25 November 1976

No. 8 ..... 27 January 1977

No. 9 ..... 27 January 1977

ISSN 0007-1498

Printed in Great Britain by John Wright and Sons Ltd. at The Stonebridge Press, Bristol BS4 5NU

CONTENTS

ZOOLOGY VOLUME 30

PAGE

No. i. Bats referred to Hesperoptenus Peters, 1869 (Chiroptera : Vesper-
tilionidae) with the description of a new subgenus. By J. E. HILL
(Pis 1-4) i

No. 2. Taxonomy and distribution of the Metapheretima elongata species-
complex of Indo-Australasian earthworms (Megascolecidae : Oligo-
chaeta). By E. G. EASTON ....... 29

No. 3. The taxonomy of Conchoecia (Ostracoda, Halocyprididae) of the
gaussi and edentata groups from the northeast Atlantic with a note
on their ecology. By A. J. GOODAY ...... 55

No. 4. Echinoderms from the northern region of the Great Barrier Reef,

Australia. By P. E. GIBBS, A. M. CLARK and C. M. CLARK (PI. i) . 101

No. 5. Miscellanea

Tilapiine fishes from crater-lakes of Lake Malawi. By E. TREWAVAS

(Pis 1-3) 149

A new species of Hemichromis (Pisces, Cichlidae) of Sierra Leone and
Liberia. By A. I. PAYNE and E. TREWAVAS (PI. i) . . 159

A possible intergeneric cyprinid hybrid from Lake Tanganyika. By

K. E. BANISTER (Pis 1-2) ....... 171

Two new species of large Barbus (Pisces, Cyprinidae) from central
Africa. By K. E. BANISTER (PL i) 191

No. 6. Miscellanea

A new genus and two new species of copepod parasitic on freshwater
fishes. By G. A. BOXSHALL 209

A new species of Chonopeltis (Crustacea : Branchiura) from southern
Africa. By G. A. BOXSHALL ....... 217

Microlaophonte trisetosa sp. nov., a new harpacticoid copepod from
southern Florida. By G. A. BOXSHALL ..... 223

A new species of Amphithoe (Pleonexes) (Amphipoda : Amphithoidae)
from the northeast Atlantic with a redescription of A. (P.) gam-
maroides (Bate). By R. J. LINCOLN 229

The first zoeal stage of Scopimera crabricauda Alcock (Crustacea,
Brachyura, Ocypodidae). By A. L. RICE 243

Asterozoa from Amsterdam and St. Paul Islands, southern Indian
Ocean. By AILSA M. CLARK (Pis 1-6) ..... 247

Restriction of the family Poraniidae, sensu Spencer & Wright, 1966
(Echinodermata : Asteroidea). By F. H. C. HOTCHKISS and AILSA
M. CLARK (Pis 1-3) ......... 263

No. 7. Variation in Eurasian shrews of the genus Crocidura (Insectivora :

Soricidae). By P. D. JENKINS ....... 269

No. 8. Biology and fine structure of Cryptodifflugia oviformis (Rhizopodea :
Protozoa). By R. H. HEDLEY, C. G. OGDEN and N. J. MORDAN
(Pis 1-6) 311

No. 9. The larval and post-larval development of the scorpion spider crab,
Inachus dorsettensis (Pennant) (Family : Majidae), reared in the
laboratory. By R. W. INGLE ....... 329

BATS REFERRED TO

HESPEROPTENUS PETERS, 1869

(CHIROPTERA : VESPERTILIONIDAE)

WITH THE DESCRIPTION OF A

NEW SUBGENUS

J. E. HILL

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. i

LONDON: 1976

•BATS REFERRED TO HESPEROPTENUS
PETERS, 1869 (CHIROPTERA : VESPERTILIONIDAE)
WITH THE DESCRIPTION OF A NEW SUBGENUS

BY

JOHN EDWARDS HILL

Pp. 1-28 ; 4 Plates ; 5 Text-figures

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. i

LONDON: 1976

THE BULLETIN OF THE BRITISH MUSEUM

(NATURAL HISTORY), instituted in 1949, is
issued in five series corresponding to the Scientific
Departments of the Museum, and an Historical series.

Parts will appear at irregular intervals as they
become ready. Volumes will contain about three or
four hundred pages, and will not necessarily be
completed within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 30, No. i, of the Zoology
series. The abbreviated titles of periodicals cited
follow those of the World List of Scientific Periodicals.

World List abbreviation :
Bull. Br. Mus. not. Hist. (Zool.)

ISSN 0007-1498

Trustees of the British Museum (Natural History), 1976

BRITISH MUSEUM (NATURAL HISTORY)

Issued 24 June, 1976 Price £3-20

BATS REFERRED TO HESPEROPTENUS PETERS,

1869 (CHIROPTERA:VESPERTILIONIDAE) WITH

THE DESCRIPTION OF A NEW SUBGENUS

By J. E. HILL

CONTENTS

Page

SYNOPSIS ........... 3

INTRODUCTION ........... 3

HISTORY ............ 4

Genus Hesperoptenus Peters, 1869 ....... 5

Hesperoptenus (Hesperoptenus) Peters, 1869 ..... 6

Hesperoptenus doriae (Peters, 1869) ...... 7

Hesperoptenus (Milithronycteris) subgen. nov. . . . . . 14

Hesperoptenus tickelli (Blyth, 1851) . . . . . . 15

Hesperoptenus tomesi Thomas, 1905 . . . . . . 19

Hesperoptenus blanfordi (Dobson, 1877). ..... 21

RELATIONSHIPS .......... 24

ACKNOWLEDGEMENTS ......... 25

REFERENCES ........... 26

SYNOPSIS

Species currently allocated to the vespertilionid genus Hesperoptenus are reviewed in detail
and divided into two subgenera, one, Hesperoptenus (Milithronycteris} with type species Nyctice-
jus tickelli Blyth, 1851, proposed as new, to include also H. blanfordi (Dobson, 1877) and H. tomesi
Thomas, 1905. The nominate subgenus contains only its type species H. doriae (Peters, 1869),
here reported for the first time from Malaya, this being the second record of the species, hitherto
known only from its holotype from Sarawak, Borneo.

INTRODUCTION

BATS of the Asiatic genus Hesperoptenus are, in general, rarely collected and therefore
do not feature extensively in museum collections. Consequently, they remain for
the most part rather poorly known and, apart from one Indian species, have received
only scant attention in the literature. As it is currently recognized, Hesperoptenus
is held to include four species, doriae, tickelli, tomesi and blanfordi. Their association
together has been considered open to doubt by some authors while the most recent
definition of the genus is based upon tickelli rather than upon the type species
doriae. In this account the four species are examined and compared and although
the number of specimens available remains small in some it has been possible
nevertheless to describe the features of each species in detail and to discuss their
differences and possible relationships. As a result, Hesperoptenus as previously
understood is divided into two subgenera, one described as new,

4 J. E. HILL

HISTORY

The name Hesperoptenus was proposed by Peters (i86ga : 626) for a subgenus of
Vesperus, with type species V. (H.) doriae described concurrently from Sarawak,
Borneo. Shortly afterwards, Peters (i869b : 638) described V. (H.) kraussi from
Yoruba, Lagos, Nigeria, nowadays considered a synonym of Glauconycteris poensis
(Dobson, 1878 : 255 ; Thomas, 1913 : 145 ; Rosevear, 1965 : 275). The two species
described by Peters are figured together in his posthumous publication of illustrations
of Chiroptera from the Zoologische Museum, Berlin (1906, pi. 26), and in this are
referred to Vesperus, with no mention of Hesperoptenus.

Dobson (1876 : 112) in a monographic study of the Asiatic Chiroptera placed
doriae in the genus Vesperugo, largely repeating the description by Peters, in
translation. Dobson remarked that he thought doriae to be closely allied to
Vesperugo tickelli (Blyth, 1851) and that when specimens of both could be compared
it would be found that they could be included together in the subgenus Hesperop-
tenus. Later, Dobson (1877 : 312) described Vesperugo (Hesperoptenus) blanfordi
and so implicitly recognized Hesperoptenus as a subgenus. Subsequently, he
(1878 : 239) defined Hesperoptenus as a subgenus of Vesperugo, to include doriae,
tickelli and blanfordi. In the same work (p. 255) Dobson reduced Vesperus (Hes-
peroptenus) kraussi Peters, 1869 to the synonymy of Glauconycteris poensis, where it
has since remained. Eventually, Hesperoptenus as understood by Dobson was
raised to generic rank (Jentink, 1887 : 278 ; 1888 : 182 ; Wroughton, 1897 : 723 ;
Trouessart, 1897 : 116 ; 1904 : 83), with Thomas (1905 : 575) describing a fourth
species, tomesi.

The essentials of the classification proposed by Dobson have persisted until the
present day. Miller, the monographer of bat genera, retained (1907 : 211) generic
rank for Hesperoptenus but examined only tickelli, which he employed as the basis
for his generic diagnosis. This definition moved from the concept established by
Dobson (1878 : 239), who, following Peters, had characterized the subgenus Hes-
peroptenus by 'outer upper incisors very small, scarcely raised above the gum,
generally posterior to the toothrow ; penis with a distinct bone ; post-calcaneal
lobe well-developed ; wings from the base of the toes'. Dobson had not seen doriae
which he stated differed from tickelli and blanfordi in that the outer upper incisors
were in the toothrow and not posterior to it as in these species. Miller (1907 : 211)
diagnosed Hesperoptenus almost entirely by the features of the dentition, relying
largely on the configuration and position of the upper incisors. Since, like Dobson,
Miller did not examine doriae, he employed Hesperoptenus with some hesitation,
considering that the description of doriae by Peters indicated certain features
suggesting that it might be generically distinct from both tickelli and blanfordi.
Miller did note, however, that peculiarities of the prepuce and the presence of a
large bone in the penis might indicate (as Dobson had thought with regard to the
os penis or baculum alone) that the three species formed a distinct group (Miller did
not include tomesi, of which he appears to have been unaware) but remarked that
the value of these characters was not then well understood.

More recently, Tate (1942 : 268) considered Miller correct in the supposition that
more than one genus was involved, but was unable to study either doriae or blanfordi.

BATS REFERRED TO HESPEROPTEN US 5

This author referred to the position of the outer upper incisor in doriae and suggested
further that the specialized thumb of blanfordi might be an indication that it too
differed widely from its supposed congeners. Ellerman & Morrison-Scott (1951 : 174),
who drew attention to this situation, pointed out that the holotype of doriae ap-
parently remained unexamined and that the genus Hesperoptenus might prove
untenable in the sense in which it was then accepted. Davis (1962 : 42) referred a
specimen from Sabah, Borneo, to doriae, noting that in this example the outer
upper incisor was displaced posteriorly. This circumstance led Medway (1965 : 65)
to consider that the type species thus conformed to the current generic definition,
thereby rendering groundless the speculation by Ellerman & Morrison-Scott. The
example from Sabah, however, does not represent the type species but instead should
be referred to tomesi, of which it is a juvenile male (Hill, 1972 : 38 ; vide infra).

It is clear that the taxonomy of Hesperoptenus, so far unresolved, can be clarified
only by the detailed examination of the holotype of doriae, upon which any inter-
pretation of the genus must turn. The original description by Peters and a brief
note of forearm length, colour and incisive dentition by Hill (1972 : 38) appear to be
the only published accounts based upon direct examination of the holotype, which is
in the Museo Civico di Storia Naturale 'Giacomo Doria', Genoa. This specimen,
until recently the only definitively known example of the species, was examined also
by Oldfield Thomas, who, in a copy of Dobson's Catalogue of the Chiroptera in the
Collection of the British Museum, 1878, now in the Library of the British Museum
(Natural History), has annotated the remark 'Specimen sent over 12/1901' to the
account of doriae. Thomas has added manuscript notes on the external features,
baculum and incisive dentition, with some measurements, to the account by Dobson.
Shortly after the holotype was studied during the early preparation of this paper, a
second specimen of doriae was obtained quite by chance in Malaya by Lord Medway,
almost exactly one hundred years after Peters first acknowledged the species.

Genus HESPEROPTENUS Peters, 1869

Hesperoptenus Peters, i86ga : 626 (as a subgenus of Vesperus). Type species Vesperus
(Hesperoptenus} doriae Peters, 1869, from Sarawak, Borneo.

DIAGNOSIS. Similar to Eptesicus but differing from this genus in relatively
wider palate with wide anterior emargination ; in the presence of well-developed
basioccipital pits ; in larger, caniniform inner upper incisor (i2), its basal area equal
almost to one half that of the upper canine (c1) ; and by the structure of the outer
upper incisor (i3) which is low, its tip reaching barely to the cingulum of i2 or of c1,
with basin-like cingulum and low central cusp. Dentally similar to Glauconycteris
but differing from this genus and from the closely related Chalinolobus in having the
posterior margin of the ear terminating on the head at a point not much anterior to
the insertion of the anterior margin, not extending almost to the corner of the
mouth, and in the absence of a fleshy lappet on the lower lip near this point. The
majority of species of Hesperoptenus differ further from Eptesicus, Glauconycteris
and Chalinolobus in the marked inward displacement of i3 from the toothrow.

6 J. E. HILL

DESCRIPTION. Muzzle low and wide, sparsely furred anteriorly ; narial openings
semi-lunate or nearly circular, opening sub-laterally ; usually a small wart-like
protuberance above the anterior canthus of the eye, bearing a few long whiskers ;
lower lip with broad median pad ; a small naked or nearly naked area immediately
beneath the symphysis menti, preceding a low wart which bears a few long hairs.
Ear subquadrangular ; anterior margin with rounded or very slightly pointed
posteriorly directed basal lobe ; tip of ear rounded ; posterior margin with promi-
nently developed antitragal lobe effecting a quadrate junction with the side of the
head behind and level with the angle of the mouth, from which it is separated by a
low wart bearing a few longer hairs. Tragus slightly hatchet-shaped, its upper part
prolonged anteriorly. Metacarpals of third, fourth and fifth digits nearly equal in
length, the third usually very slightly the longest ; a well-developed post-calcarial
lobe, sometimes supported by a cartilaginous spur ; penis with wide, saccular
prepuce with large terminal opening.

Skull with wide braincase, the frontal region not especially inflated ; rostrum
short ; palate short, wide, the lateral margins of the pre-palatal emargination not
extending greatly beyond the inner margins of the inner upper incisors (i2-2) and
separated from the canines (c1"1) by a relatively wide ledge of bone ; evident basioc-

.21 i 3

cipital pits. Dental formula i -, c -, pm -, m - = 32 ; inner upper incisor (i2)

massive, caniniform, unicuspid, with prominent cingulum, the tips of the inner
incisors inwardly directed and slightly convergent ; outer upper incisor (i3) low,
with basin-like cingulum from which arises a small pyramidal cusp, extending barely
to the cingulum either of i2 or of the upper canine (c1) in height ; i3 sometimes
displaced inwardly from the toothrow to the extent that it may lie behind or almost
behind i2 ; third upper molar (m3) reduced but retaining mesostyle, metacone and
three commissures ; third lower molar (m3) usually only slightly reduced, its
posterior triangle a little smaller than its anterior triangle, but considerably so in
one species.

DISTRIBUTION. The genus is distributed through India, Sri Lanka, Burma and
the Andaman Islands to Thailand and Malaysia, including Borneo.

Subgenus HESPEROPTENUS Peters, 1869

DIAGNOSIS. Ear and tragus membranaceous ; second phalange of third digit
markedly longer than first phalange ; braincase slightly inflated, rather globose ;
rostrum not especially widened ; supraorbital region only very slightly inflated ;
lower maxillary margin of orbit narrow, forming a narrow, tapering ledge alongside
the toothrow ; maxillary toothrows nearly parallel ; outer upper incisor (i3) in
toothrow or nearly so, not markedly intruded.

DISTRIBUTION AND SPECIES. The subgenus is known so far only from Malaysia
and includes solely the species Hesperoptenus doriae (Peters, 1869), described
originally from Sarawak, Borneo, and here reported for the first time from Malaya.

BATS REFERRED TO HESPEROPTENUS 7

Hesperoptenus doriae (Peters, 1869)

Vesperus (Hesperoptenus} doriae Peters, i86ga : 626. Sarawak, Borneo.

Vesperugo (Vesperus) doriae, Dobson, 1876 : 104, 122.

Vesperugo (Hesperoptenus) doriae, Dobson, 1878 : 239.

Vesperugo doriae, Dobson, 1880 : 187 ; Hose, 1893 : 40.

Vesperugo donriae Hose, 1893 : 40. Lapsus.

Hesperoptenus doriae, Trouessart, 1897 : n6 ; 1904 : 83 ; Miller, 1907 : 211 ; Chasen, 1940 : 52 ;
Tate, 1942 : 269 ; Ellerman & Morrison-Scott, 1951 : 174 ; Davis, 1962 : 42 (erratum, based
on tomesi) ; Medway, 1965 : 65 (erratum, based on tomesi) ; Hill, 1972 : 38.

HOLOTYPE. Adult male CE 40094 in the Museo Civico di Storia Naturale
'Giacomo Doria', Genoa, Italy. In alcohol, skull removed.

DESCRIPTION. Muzzle with a shallow longitudinal groove immediately behind
the widely separated narial openings which are nearly circular ; a few sparse hairs
along the upper lip between the nostrils, the internarial region otherwise nearly
naked and with only a sparse covering of short hairs above the nostrils and inter-
narial pad ; prominent wart above anterior canthus of eye, bearing a few long
whiskers ; lower lip with a broad median pad with no more than a trace of a median
dividing groove at its posterior border only ; a small naked area beneath the
symphysis menti, preceding a low wart which bears a few long hairs. Ears (PI. i)
membranaceous, not thick or fleshy, rounded, more or less subquadrangular, the
external surface of the conch quite densely clothed with hair for about one half the
length of the ear, a few sparse hairs on the inner surface ; anterior margin of ear
strongly convex, with rounded, posteriorly directed lobe at its base ; tip rounded ;
posterior margin shallowly concave distally but otherwise strongly convex ; well-
developed, fleshy antitragal lobe thickened along its outer margin, terminating
quadrately behind and level with the corner of the mouth but separated from it by
a low wart bearing a tuft of longer hairs. Tragus (Figs la, ib ; PL i) membrana-
ceous, about one half the length of the ear, its anterior margin straight proximally,
concave distally ; tip rounded, directed anteriorly ; upper margin nearly hori-
zontal, sloping only slightly posteriorly, curving convexly to the slightly convex
posterior margin which has a large, more or less triangular lobe at its base.

5 mm

FIG. i. Tragus of : (a) Hesperoptenus doriae, holotype, Sarawak ; (b) H. doriae,
Malaya ; (c) H. tickelli ; (d) H. tomesi; (e) H. Uanfordi.

8 J. E. HILL

Thumb long, its upper surface sparsely haired, with a strong claw ; forearm naked
except for a few hairs proximally ; upper surface of tibia and of toes with a few
scattered hairs ; fur extending sparsely on to the dorsal surface of the uropatagium
for about one half its width, the ventral surface similarly but more thinly haired.
Third metacarpal the longest, second and fourth very slightly shorter, the fifth the
shortest ; second phalange of third digit longer than first phalange ; endopatagium
and uropatagium slightly reticulated, remainder of flight membrane striated ; a
rounded post-calcarial lobe supported by a poorly defined cartilaginous prop ; penis
with saccular prepuce, sparsely haired, with V-shaped preputial opening. Holotype
dorsally pale brown, hairs dark brown at the base but pale straw for much of their
length, the ventral surface similar but paler, the specimen evidently much bleached
by long immersion in alcohol. A second specimen (in alcohol) from Malaya is dark
brown above and below, the membranes blackish brown.

Skull (PI. i) small, with inflated, slightly globose braincase, more elevated
posteriorly than anteriorly, the frontal region slightly inflated ; slight lambdoid
crests ; postorbital region sharply constricted ; rostrum short, narrow, with very
slight supraorbital ridges, the supraorbital region only very slightly inflated ; a
shallow depressed area immediately above anteorbital foramen ; anterior margin
of orbit in frontal aspect sloping slightly inwards towards upper surface of rostrum ;
shallow median rostral depression. Narial emargination wide, deep, extending
posteriorly to a line joining the anterior edges of the anteorbital foramina, in the
holotype V-shaped, with rounded apex, in a Malayan specimen U-shaped.

Zygomata slender ; anteorbital foramen closed by a narrow bar of bone ; lower
maxillary margin of orbit narrow, not especially flange-like ; pre-palatal emargin-
ation wide, extending laterally a little beyond the inner faces of the inner upper
incisors (i2~2) and posteriorly to a line joining the posterior faces of the canines
(c1-1), rounded posteriorly, its lateral margins forming a narrow bony shelf. Palate
rather short and wide, the maxillary toothrows nearly parallel ; post-palatal
extension long, its lateral margins initially more or less parallel, with a slight convex
expansion a little more than halfway along the extension, the margins then conver-
gent, the hamulars convergent posteriorly ; post-palatal spine prominent, broad at
the base. Soft palate (Fig. 2a) with five post-canine ridges, the first unbroken, the
remainder each medianly divided, curved convexly on each side of the palate towards
its centre. Basioccipital narrow, with deep, narrow basioccipital pits, separated by
a narrow ridge, the bullae relatively large.

Baculum (Fig. 3a) of holotype damaged, but has evidently a long, relatively wide,
flattened shaft with little sign of any pronounced upward curvature. At the base
the baculum is expanded to form paired basal flanges, thick and massive, separated
above and below by shallow median fissures. The shaft itself is parallel-sided ; in
section its upper surface is slightly convex, the ventral surface in the proximal and
central parts of the shaft slightly concave. Towards the tip the shaft is thickened
vertically, the tip itself rounded

Inner upper incisor (i2) caniniform, large, unicuspid, its basal area one half that
of the upper canine (c1), with prominent cingulum. Anterior face rounded, tooth
with slight lateral edges, postero-lateral faces scooped out at each side of a posterior

BATS REFERRED TO HESPEROPTENUS

2 mm

FIG. 2. Palate ridges of : (a) Hesperoptenus doriae, Malaya ; (b) H. tickelli
(c) H. tomesi ; (d) H. blanfordi.

10

J. E. HILL

C- •^,^."- JV.. ...,i,. .,n.JUii 'j,.!...,.
£^v.i:;f;.^v

€

I I

2 mm

FIG. 3. Baculum of : (a) Hesperoptenus doriae, holotype ; (b) H. tickelli
(c) H. tomesi, holotype.

BATS REFERRED TO HESPEROPTEN US II

median knife-like ridge. Outer upper incisor (i3) simple, basal area little more
than one quarter that of i2, with basin-like cingulum and simple central cusp, i3
basally slightly ovate, the longer axis lying diagonally to the toothrow ; not
displaced or only very slightly intruded, in contact or nearly so with i2 and c1,
which is massive, with small anterior and posterior cingulum cusps. Upper pre-
molar (pm4) twice as wide as long, not compressed in the toothrow : first and second
upper molars (m1-2) with low hypocones, their lingual margins rounded ; third
upper molar (m3) about one half crown area of m1, with metacone and three com-
missures, the third not greatly reduced. Lower incisors (i!-3) tricuspid, linear,
slightly imbricated, i3 a little thickened ; anterior lower premolar (pm2) about one
third the crown area of the second lower premolar (pm4) ; entoconid and hypoconid
of third lower molar (m3) not greatly reduced, one half the height of the anterior
cusps, the tooth only slightly narrowed posteriorly (Pis 2 and 4d).

Measurements of the holotype and of a female example from Malaya appear in
Tables i and 2.

DISTRIBUTION. MALAYA (Selangor) ; BORNEO (Sarawak). Hesperoptenus doriae
is known apparently only from the holotype from Sarawak and from a female
example (in alcohol, skull extracted) obtained by Lord Medway on 14 February 1969
at the Field Studies Centre, Ulu Gombok, Selangor, Malaya. This further specimen
is now in the collections of the British Museum (Natural History). No direct
comparison with the holotype has been possible but as far as can be determined the
two specimens are in close agreement except for the more rounded, less angular
narial emargination of the Malayan example.

REMARKS. The inflated, slightly globose braincase of H. doriae removes it rather
sharply from Eptesicus and suggests Glauconycteris or the closely allied Chalinolobus,
which are thought to be subgenerically related by Ryan (1966 : 86, 89) and by
Koopman (1971 : i). However, these differ from Hesperoptenus in the insertion of
the posterior margin of the ear and in the presence of a lappet on the lower lip near
the angle of the mouth, and the rostrum in general is more shortened. Externally,
H. doriae is strongly reminiscent of Eptesicus or Pipistrellus and indeed its baculum
is of the long-shafted type associated with the latter genus. Dentally, the species
has some features in common with Glauconycteris in which the outer upper incisor
(i3) has a wide cingulum with simple central cusp and stands in the toothrow,
although in some species a small degree of extrusion is found, to approach the
condition usual in Chalinolobus in which the four upper incisors lie on a line in most
species. In H. doriae the tendency is for i3, although in the toothrow, to lie towards
its inner or lingual border : the inner margins of the inner upper incisor (i2), of i3
and of the upper canine (c1) lie on a line, with the outer margins of i2 and c1 extending
considerably beyond the outer margin of i3. Although on occasion a similar situ-
ation occurs in Glauconycteris (cf. G. gleni, vide Peterson & Smith, 1973 : 3, fig. 2,
centre) the tendency in this genus seems to be for i3 to become extruded. Hesperop-
tenus doriae differs further from Glauconycteris in its rounded and not linear post-
calcarial lobe and in having the second and third lower incisors (i2-3) linear and
imbricated, lacking any low posterior cusp, rather than standing along the line of the

" J. E. HILL

g o N ,

CD OTj-Tj-iOrOOOiO'^-

CO Tf- CO O M

N N M
I I I I I I I I

Q) 00 ^ *^" vO u~) CT* O *O
op N N (S W

to

s *t

•S •B^BIBJ^ ._j_ o

gz^'i'i'Z j\[g 'PPV^PT^^^^P ^

^ SdA^OpJJ lOHTJ-Tt-^H^Tt-Tt- **1

% '•-

^ I

'— ' rt P
O _ * * * * * * *

Cu O Qs ^t" w ^" CO *O O^ "

^-^ <D ....... Q

p ^H ^t" c^ O^ O ci c*l O^ CO *ri

B ^<>OMTt-U-)WNTj-^

p "** • V

rt) •••••••• rt O

hn OfO't'POioiOMN 173 u

gVOHiOiONNiOiO rt M

H g fe 3 I I I I I I I I ^ co

^ rH ••..,... ^H

V\ C*l t^» 00 M M ^ t^ O O
<U •^-MTj--<J-NNTj-Tj-j3S

b o o

3 O^-i

rt c! •<*• rt 'O

§ .s |

1 "2* o

"^ Q C/r-VT'C"/ -TIT*— r- * -4- M CO O^ O^ I""* M 00 *O W ^^

>Th romw(v»rom ^ H

fT\ — Vi y /-\ I—.

- •« S3 vo ^

W ...... v, ,.,. ,4, pj

^O -H

33 uinasnm •EOUSO T Pf^^^PP0?

OO 00 u~J vO >O O^ VO "^" c/3

fO CO CO H M fO CO

oooooooo

flj o S S m o o
,_! _! J J J J J

S S Q

D •• ^ *-*

c ^ a •

H 9. Jl •
JJ 53 T3 a

^2 C _ C I

BATS REFERRED TO HESPEROPTEN US

d p M o p

"§« 1) rniNj"' "* <O •

bO M M HI ' 00 M

-^ € i i i i i i i i i i i i i i i

f\s t^ OO t^» O* 00 OO O^ CO O O >O O O* C^ OO

H OO O HI O •<*•

Ha '6 « ^ 2s "° °° 2

Tf (T) N
c, "*t b&F'F^ J\[g 'P ^ <S (S

S [JH

f-j M^

a

Ll rt ^

M «- i§ sp % $ $

.< z - a i i i i

100*000 MO* oooooor^-^-o>o-^--^-oo

I *K . . - T

f^gi'?^ jAig '£ ^" J7 H? N
° I

tti JIBABlBg N N POOO

uin9snj\[

4)

5 3

'-P 13

** --e is g

^-. ... «^^ *—• w^

O X5

2 *" rt "^ <U S

"S-S "^0J_c!'^I^ "^2

bO d

I

c

s

a" S

I4 J. E. HILL

toothrow with a low supporting cusp behind the central one of the three incisive
cusps either of i2 or of i3. It differs quite sharply from Chalinolobus in the presence
of a triangular lobule at the base of the posterior margin of the tragus rather than a
low convexity ; in its broad, saccular rather than smooth, prepuce which has a
rounded, slightly V-shaped, preputial orifice and not a narrow vertical aperture ;
in wide, not narrow, pre-palatal emargination ; in the structure of i3, a simple cusp
with broad cingulum rather than a broader, hollowed tooth and in the position of
this tooth in the row rather than extruded from it. Despite some similarities,
therefore, there appears to be no direct connection between this species and con-
sequently Hesperoptenus with either Glauconycteris or Chalinolobus.

Hesperoptenus doriae differs sufficiently from the other species hitherto assigned
to the genus as to justify their separation in a distinct subgenus. For this I propose :

MILITHRONYCTERIS subgen. nov.

TYPE SPECIES. Nycticejus tickelli Blyth, 1851.

INCLUDED SPECIES. Vesperugo (Hesperoptenus) blanfordi Dobson, 1877 ;
peroptenus tomesi Thomas, 1905.

ETYMOLOGY. The name of this newly separated subgenus has been chosen to
commemorate the late G. S. Miller, the monographer of bat genera, at one time of the
United States National Museum, now the National Museum of Natural History,
Washington. The name is derived from [jivXwdpas, a master miller, and wKrepis,
a bat.

DIAGNOSIS. Like Hesperoptenus (Hesperoptenus) but ears and tragus fleshy
rather than membranaceous ; second phalange of third digit reduced, its length
about equal to or less than the length of the first phalange ; braincase lacking any
marked inflation, elongate rather than globose ; rostrum much broadened supra-
orbitally ; lower maxillary margin of orbit heavy, flange-like, viewed ventrally
forming a wide ledge alongside the toothrow ; maxillary toothrows convergent
anteriorly ; and outer upper incisor intruded from toothrow to the extent that on
occasion it may lie behind the inner upper incisor.

DESCRIPTION. Similar externally to Hesperoptenus (Hesperoptenus) but internarial
region sometimes densely covered with short hairs ; facial warts sometimes low and
imperceptible ; ear and tragus fleshy, the tip of the tragus sometimes considerably
attenuated and prolonged anteriorly.

Skull with uninflated, elongate, slightly broadened braincase ; frontal region
barely elevated, the cranial profile straight or nearly so ; rostrum relatively short,
wide, the supraorbital region inflated and swollen ; rostral depression very shallow
or absent ; an almost imperceptible depressed area immediately above the anterior
part of the orbit ; anteorbital foramen large ; lower maxillary margin of orbit wide,
flange-like ; narial emargination deep, rounded posteriorly ; pre-palatal emargin-
ation deep, wide, its lateral margins extending beyond the inner faces of the inner
upper incisors (i1"1), extending posteriorly to or almost to a line joining the posterior

BATS REFERRED TO HESPEROPTENUS 15

faces of the upper canines (c1-1) ; post-palatal extension short, wide, with prominent
median spine ; shallow but evident basioccipital pits ; bullae of moderate size.

Dentition much as in the nominate subgenus but outer upper incisor (i3) intruded
from toothrow so that the centre of the tooth lies almost exactly behind the centre
of the inner upper incisor (i2), with its anterior face lying approximately on a line
joining the anterior faces of the upper canines (c1-1), the tooth in contact or nearly
in contact with the posterior face of i2 and with the antero-internal face of c1 ;
anterior lower premolar (pm2) about one half or a little less the basal area of the
second lower premolar (pm4), slightly compressed in the toothrow.

DISTRIBUTION. India, Sri Lanka and the Andaman Islands east to Thailand,
Malaya and Sabah, Borneo ; possibly occurs also in China.

REMARKS. As here understood, the subgenus Milithronycteris includes three
species, tickelli, tomesi and blanfordi. The larger species tickelli and tomesi are
clearly more closely related to each other than either is to the very much smaller
blanfordi, from which they differ quite sharply in a number of features. Neverthe-
less, in blanfordi as in the larger species, the braincase is broad, low and elongate,
the rostrum expanded supraorbitally, with a broad sub-orbital flange, and the
outer upper incisor (i3) is intruded to lie behind the inner tooth (i2). The three
species may be keyed :

1 Large, length of forearm 48-6-60-0 mm ........ 2

- Small, length of forearm 25-9-27-5 mm ...... blanfordi (p. 21)

2 Dorsally yellowish brown to straw brown, ventrally a little paler ; anterior dentition

to premolars not especially massive ; lower incisors (ii-3) not greatly imbricated

tickelli (p. 15)
— Rich chocolate brown both dorsally and ventrally ; anterior dentition to premolars

heavy and massive ; lower incisors (ix-3) much imbricated . . . tomesi (p. 19)

Hesperoptenus tickelli (Blyth, 1851)

Nycticejus isabellinus Kelaart (ex Blyth), 1850 '.317 (pagination of reprint of 1890). Nomen
nudum. Sri Lanka.

Nycticejus tickelli Blyth, 1851 : 157. 'Central India, Ceylon and the intervening hilly country.'
Restricted to Chaibassa, Bihar (Anderson, 1881 : 132, Wroughton, 1918 : 593).

Nycticejus isabellinus Horsfield (ex Blyth), 1851 : 38. Central India.

Nycticejus tickelli, Kelaart, 1852 : 24 ; Wagner, 1855 : 765, footnote, 766 ; Blyth, 1863 : 31.

Nycticejus isabellinus, Kelaart, 1852 : 24 ; Wagner, 1855 : 765, footnote ; Fitzinger, 1870 :
363 ; Dobson, 1876 : 113 ; 1878 : 240 ; Blanford, 1891 : 317.

N. [ycticejus] isabellinus, Blyth, 1863 : 31.

Nycticejus tickellii Fitzinger, 1870 : 362. Lapsus.

Vesperugo (Vesperus) tickelli, Dobson, 1876 : 104, 113, figs a, b, 208.

Vesperugo (Hesperoptenus) tickelli, Dobson, 1877 : 312 ; 1878 : 239, 240, pi. 12, fig. 3.

Vesperugo (Hesperopterus) (sic) tickelli, Anderson, 1881 : 132.

Vesperugo tickelli, Sterndale, 1884 : 63 ; Blanford, 1891 : 317, fig. 97.

Hesperoptenus tickelli, Jentink, 1887 : 278 ; Wroughton, 1897 : 723 ; Trouessart, 1897 : 116
1904 : 83 ; Thomas, 1905 : 575 ; Miller, 1907 : 211 ; Wroughton, 1912 : 1180 ; 1913 : 37
I9i5a : 86 ; igisb : 103 ; 1917 : 66 ; 1918 : 593 ; Phillips, i932a : 345 ; I932b : 349 ; 19320
133 J i932d : 137 ; 1933 : 240 ; 1935 : 118 ; Tate, 1942 : 269 ; Ellerman & Morrison-Scott,
1951 : 174 ; Brosset, 1962 : 729 ; Hill, 1967 : 7 ; Hill & Thonglongya, 1972 : 191.

16 J. E. HILL

Hesperoptenus tickelii Jentink, 1888 : 182. Lapsus.
{Hesperoptenus'] isabellinus, Trouessart, 1897 : 116.
Hesperoptenus isabellinus, Tate, 1942 : 269.

SYNTYPES. Dobson (1876 : 114 ; 1878 : 241) stated 'Type in the collection of the
Indian Museum, Calcutta' but did not designate a specimen. Anderson (1881 : 132)
listed as types three specimens in the Indian Museum. These are skins from
Chaibasa (= Chaibassa, Bihar, India) presented in 1842 by S. R. Tickell to the
Asiatic Society of Bengal, whose Museum later became a part of the Indian Museum.
Listed as I56a-c by Anderson (1881 : 132) in the Indian Museum collections, these
specimens are numbered 95A-C in Blyth's Catalogue (1863 : 31) of the mammals
in the Museum of the Asiatic Society of Bengal and by Dobson (1876 : 208) as
474-476 in a catalogue of the specimens of Chiroptera in the collection of the Indian
Museum which he provided in his Monograph of the Asiatic Chiroptera. Blyth
(1863 : 31) also listed without data a skull (95D) later recorded (as 477) by Dobson
(1876 : 208) as having the same data as the three skins, but Anderson (1881 : 132)
lists this skull (as I56d) without locality but as originating from S. R. Tickell in
1842.

OTHER TYPE MATERIAL. Horsfield (1851 : 38) lists only a single specimen under
Nycticejus isabellinus, a dried example from Central India, in the Museum of the
Honourable East India Company, presented by the Asiatic Society of Bengal.
There is little doubt that this specimen is that now registered as BM 60.5.4.12,
presented to the British Museum (Natural History) by the Secretary of State for
India, and from the East India House collection. It is a skin with skull in situ, but
with the upper incisors visible, labelled in Blyth's handwriting 'Nycticejus isabellinus
Bl. Central India' and by Dobson 'Vesperugo tickelii, Blyth. Syn. V. isabellinus,
Blyth'. The specimen is therefore apparently the holotype of Nycticejus
isabellinus Horsfield, 1851 : it exactly resembles Hesperoptenus tickelii as Dcbson
recognized.

DESCRIPTION. Muzzle much as in Hesperoptenus doriae, the longitudinal depres-
sions behind nostrils shallow and faint ; membrane of ear and tragus thick and fleshy,
fur extending densely on outer surface of conch for about one half the length of the
ear, the inner surface with a few sparse hairs ; anterior margin of ear with a well-
developed, slightly pointed posteriorly directed basal lobe ; proximally the anterior
margin is strongly convex, distally nearly straight, the tip of the ear broadly rounded,
the posterior margin convex. Tragus (Fig. ic) about one half the length of the ear,
slightly hatchet-shaped, widest at about one half its length, its tip narrow, rounded,
anteriorly directed, the upper margin nearly horizontal, then sharply convex to join
the rather less convex posterior margin.

Thumb relatively long, with a slight sprinkling of short hairs along its dorsal
surface and with a strong claw ; forearm naked ; dorsal surface of tibia and foot
with a few sparse, short hairs ; fur extending very sparsely on to the dorsal surface
of the uropatagium for about one half its width, its ventral surface with a few
scattered short hairs ; metacarpals of approximately equal length but the third
usually slightly the longest ; post-calcarial lobe linear or slightly rounded at its

BATS REFERRED TO HESPEROPTEN US

central point, lacking a cartilaginous support ; penis saccular, with U-shaped
preputial opening.

Dorsally yellowish brown to straw brown, brighter posteriorly than anteriorly, the
ventral surface similar but paler. Specimens from Sri Lanka and the Andaman
Islands are a little darker than are those from the Indian mainland, and are more
reddish brown dorsally, especially posteriorly. They also average a little smaller
than those from the mainland. The differences, however, are small and scarcely
justify subspecific recognition : in any event, specimens from Thailand are similar
to Indian examples in colour but in size to those from Sri Lanka and the Andaman
Islands. The differences in size are compared in Table 3.

TABLE 3

Dimensions (in mm) of Hesperoptenus tickelli from India, Sri Lanka,
the Andaman Islands and Thailand

Length of forearm

Condylobasal length

India
Sri Lanka
Andaman Islands
Thailand

India
Sri Lanka
Andaman Islands
Thailand

n

Range

Mean

s.d.

15

54-3-60-

o

57-13

1-92

25

49-1-56-

8

52-24

i -80

51-1

3

52-8-54-

2

53-43

m

3_m3

n

Range

Mean

s.d.

12

8-9-10-

2

9-60

0-38

13

8-5- 9-

5

9-07

0-28

8-8, 8-9

3

9-1- 9-

3

9-23

s.e.
0-51
0-36

s.e.
o-ii
0-08

n

Range

Mean

s.d.

ii

18-3-19-6

18-96

0-38

13

17-0-18-3

17-70

0-42

18-5

3

17-8-18-0

17-90

c-m3

n

Range

Mean

s.d.

12

7-3-8-1

7-79

0-26

17

7-1-7-7

7-38

0-19

7-2, 7-2

3

7-3-7-6

7-40

s.e.

O-I2
O-I2

s.e.
0-08
0-05

n, number of specimens; s.d., standard deviation; s.e., standard error of mean.

Skull (PI. 3a-c) large, with well-developed sagittal and lambdoidal crests, the
braincase broad, uninflated ; postorbital region sharply constricted, its width less
than the width across the canines at the alveoli ; supraorbital region considerably
expanded, the supraorbital ridges terminating in small inflations ; anterior orbital
margin sharply defined, in frontal view sloping slightly inwards ; anteorbital foramen
of moderate size, enclosed by a moderate bony bar ; maxillary margin of orbit
heavy, flange-like ; zygomata slender ; narial emargination usually V-shaped but
with rounded apex, sometimes more or less U-shaped ; post-palatal extension short,
with widely based but otherwise narrow median spine. Soft palate (Fig. 2b) with
six post-canine ridges, the first two unbroken, remainder curving convexly to the
median line, the last small, not reaching laterally to the teeth. Basioccipital pits
large, shallow, sharply incised into the basisphenoid.

Baculum (Fig. 3b) with strong upward deflection, its shaft long, nearly parallel-
sided, curved upwards at its middle part, its tip unthickened, rounded, shaft slightly
narrowed towards the base, the basal part of the baculum expanded to form paired
basal flanges, separated above and below by median V-shaped fissures.

J. E. HILL

FIG. 4. Hesperoptenus tickelli, distribution.

Outer upper incisor (i3) ovate, with wide, basin-like cingulum and three-faced
cusp ; labial part of tooth interposed between inner upper incisor (i2) and canine (c1)
but tooth otherwise intruded from row, its anterior face in contact with the posterior
face of i2 and lying on or a little in advance of a line joining the anterior faces of c1"1,
its postero-lateral face in contact or nearly so with the antero-lateral face of c1 ;
i2 and c1 separated by a narrow space equal to or a little less than one half the basal
diameter of i2, the centre of i3 lying in advance of a line joining the centres of c1-1 ;
lower incisors (^-3) thick but linear, not greatly imbricated, i3 not especially massive
(Pis 3a, c, 4e).

Measurements of Hesperoptenus tickelli appear in Tables i and 2.

DISTRIBUTION (Fig. 4). INDIA (Rajasthan : Dobson, 1878 : 241 ; Blanford,
1891 : 317. Uttar Pradesh : Wroughton, 1913 : 37. West Bengal : Wroughton,
1917 : 66. Bombay, Dobson, 1878 : 241 ; Wroughton, 1897 : 723 ; Brosset,
1962 : 729. Madhya Pradesh : Dobson, 1876 : 114, 208 ; Anderson, 1881 : 132, 133 ;
Sterndale, 1884 : 63. Bihar : Dobson, 1876 : 114, 208 ; Anderson, 1881 : 132 ;
Sterndale, 1884 : 63 ; Wroughton, igisb : 103. Madras, Wroughton, 1918 : 593.
Mysore : Wroughton, 1912 : 1180. Andaman Islands : Dobson, 1876 : 208 ; Ander-
son, 1881 : 132 ; Hill, 1967 : 7). SRI LANKA (Kelaart, 1850 : 317 (separate pagina-
tion) ; 1852 : 24 ; Blyth, 1863 : 31 ; Dobson, 1878 : 241 ; Wroughton, 19153. : 86 ;
Phillips, i932a : 345 ; ig32b : 349 ; ig32c : 133 ; i932d : 137 ; 1933 : 240 ; 1935 :

BATS REFERRED TO HESPEROPTEN US 19

118). BURMA (Dobson, 1877 : 312 ; Anderson, 1881 : 133 ; Blanford, 1891 : 317).
THAILAND (Hill & Thonglongya, 1972 : 191). CHINA (Jentink, 1887 : 278 ; 1888 :
182, from 'Kolongsu', not traced).

REMARKS. Nycticejus isabellinus Horsfield, 1851. Initially a manuscript name
from Blyth, isabellinus is validated by Horsfield (1851 : 38) (contrary to the assertion
by Blanford, 1891 : 317, who says of this citation 'no description') with the remark
'Of the size of N. temminckii [= Scotophilus kuhlii], clearly characterised by an
uniform isabellina tint both above and underneath'. The earliest use of isabellinus,
by Kelaart (1850 : 317, pagination from a reprint dated 1890, in the British Museum
(Natural History), differing apparently from the original), is without description in a
Catalogue of mammals from Sri Lanka. Kelaart (1852 : 24) and Blyth (1863 : 31)
list it as a synonym of tickelli, as does Fitzinger (1870 : 362), Dobson (1876 : 113 ;
1878 : 240), Blanford (1891 : 317) and Trouessart (1897 : 116). Tate (1942 : 269)
listed it in parentheses as [Hesperoptenus isabellinus (Horsfield)] and repeated the
description by Horsfield.

The names Nycticejus tickelli Blyth and Nycticejus isabellinus Horsfield both
appeared in 1851. In neither of the publications concerned, the Journal of the
Asiatic Society of Bengal or Horsfield's Catalogue of the Mammalia in the Museum of
the Hon. East-India Company does there appear evidence to establish a definite
date of publication. All that can be said is that the latter certainly appeared after
18 August 1851, the date of its prefatory list of contributors to the Museum ; Blyth's
description is in the second issue of the four of the Journal for that year and probably
appeared before the account by Horsfield. In these circumstances the name adopted
is that employed by the first reviewer, namely Dobson (1876 : 113) ; Fitzinger
(1870 : 362) gave a detailed description of the species and might be considered a
prior reviewer.

Hesperoptenus tomesi Thomas, 1905

Hesperoptenus tomesi Thomas, 1905 : 575. Malacca, Malaya.

Hesperoptenus tomesi, Kloss, 1908 : 158 ; Chasen, 1940 : 53 ; Tate, 1942 : 269 ; Medway, 1969 :

38 ; Hill, 1972 : 38.
Hesperoptenus doriae, Davis, 1962 : 42 ; Medway, 1965 : 65.

HOLOTYPE. Adult male BM 7.1.1.428. Skin and skull, rear and lower part of
cranium missing. From the Tomes Collection.

DIAGNOSIS. Muzzle a little more densely haired than in H. tickelli ; tragus
(Fig. id) slightly more acutely pointed and directed a little further anteriorly ;
differs sharply in rich, dark chocolate brown coloration both above and below ;
skull larger with more prominent supraorbital ridges terminating in small tubercles ;
anterior orbital margin more nearly vertical ; zygomata more widely expanded ;
cingulum of outer upper incisor (i3) a little narrower, tooth further intruded from row
and only a little interposed between the inner upper incisor (i2) and the canine (c1).

DESCRIPTION. Apart from colour the external features of H. tomesi resemble
those of H. tickelli very closely ; the metacarpals are a little more graduated, with

2*

20 J. E. HILL

the third the longest, the fourth usually a little shorter, and the fifth the shortest,
although the differences in length are small. The post-calcarial lobe has a rounded
central part but is otherwise linear, and lacks a cartilaginous supporting prop ;
the thumb, forearm and tibia are naked or nearly so.

Skull (PI. 3d-f) similar to that of H. tickelli but braincase a little more elevated,
especially posteriorly, with a more pronounced occipital 'helmet' ; postorbital
region narrow, its width less than the width across the upper canines (c1-1) at the
alveoli ; anterior orbital margin heavy, flange-like, more sharply defined than in
tickelli ; anteorbital foramen large, closed by a strong flange of bone ; zygomata
relatively massive anteriorly, heavier than in tickelli ; narial emargination wide,
more or less U-shaped ; soft palate (Fig. 2c) much as in tickelli ; broad post-palatal
spine ; basioccipital pits a little deeper than in tickelli.

Baculum (Fig. 3c) long, with slender, cylindrical shaft, lacking any upward
curvature or deflection, tip slightly thickened vertically, upper surface of shaft
flattened proximally, the base widened to form paired basal flanges separated by a
deep V-shaped aperture. The baculum is longer than in H. tickelli and differs in its
lack of upward curvature and in narrower basal flanges which are more widely
separated.

Inner upper incisor (i2) relatively larger than in H. tickelli ; outer upper incisor
(i3) more or less triangular in basal outline, intruded from the toothrow to lie directly
behind i2, its anterior face in contact with the posterior face of that tooth and lying
behind a line joining the anterior faces of the canines (c1"1), the tip of its roughly
triangular base interposed between the postero-lateral face of i2 and the antero-
lateral face of c1 ; centres of i3"3 lying on a line joining the centres of c1"1, with i2 and
c1 almost in contact labially. Remaining dentition much as in tickelli but third
upper molar (m3) a little further flattened and platelet-like, its second and third
commissures more reduced ; lower incisors (ii-a) much imbricated, relatively more
massive than in tickelli, i2 with small posterior cingulum cusp behind central cusp,
i3 much thickened, with massive crown in which individual cusps are obscured but
with prominent anterior and posterior cingulum cusps, its lateral diameter less than
its diameter from front to back. In general, the dentition is considerably heavier
than in tickelli (Pis 3d, f, 41).

Measurements of Hesperoptenus tomesi are given in Tables i and 2.

DISTRIBUTION. MALAYA (Malacca ; also two examples from an unknown
Malayan locality) ; BORNEO (Sabah).

REMARKS. Davis (1962 : 42) recorded as Hesperoptenus doriae a specimen from
the Sapagaya Forest Reserve, Sandakan, Sabah, Borneo (5°37'N n8°04'E).
Through the kindness of Dr J. C. Moore, lately of the Field Museum of Natural
History, Chicago, I have been able to examine this specimen, now Field Museum
77025, collected by D. D. Davis on 29 July 1950. As Davis pointed out, it is a
juvenile male, but is, however, much larger than the adult holotype of doriae, with
length of forearm 42-8 mm and c-m3 7-3 mm, and, although badly damaged with
much of the cranium lost, the skull agrees closely with that of tomesi to which the
specimen must be referred. It is thus the first of the species to be reported from

BATS REFERRED TO HESPEROPTEN US 21

Borneo. The specimen differs slightly from the Malayan holotype of tomesi in the
positional relationship of the upper incisors, the outer tooth (i3) being a little less
intruded from the toothrow, with the inner upper incisor (i2) and the canine (c1)
separated by a slightly wider interspace. A brief account of this specimen can be
found in Hill (1972 : 38).

Hesperoptenus tomesi is clearly separated from H. tickelli by differences in color-
ation, cranial architecture, baculum and dentition. That it represents a distinct
species cannot be doubted, as Tate (1942 : 268) pointed out from a consideration of
the anterior part of the orbit in tickelli and tomesi. The unique features of the species
were first recognized by Tomes who has labelled the holotype 'V. dedalion n. s.', a
circumstance remarked by Thomas (1905 : 576) who would have used the name
suggested by Tomes had it matched euphoniously with the generic epithet Hesperop-
tenus.

Hesperoptenus blanfordi (Dobson, 1877)

Vesperugo (Hesperoptenus) blanfordi Dobson, 1877 : 312. Tenasserim.

Vesperugo (Hesperoptenus) blanfordi, Dobson, 1878 : 242.

Vesperugo blanfordi, Anderson, 1881 : 133 ; Blanford, 1891 : 317.

Hesperoptenus blanfordi, Jentink, 1888 : 182 ; Trouessart, 1897 : 116 ; 1904 : 83 ; Miller, 1907 :
211 ; Kloss, 1908:158; Robinson & Kloss, 1915 : 116; Thomas, 1916:2; Wroughton,
1918 : 393 ; Gyldenstolpe, 1919 : 137 ; Chasen, 1940 : 52 ; Tate, 1942 : 270 ; Ellerman &
Morrison- Scott, 1951 : 174 ; Medway, 1969 : 38 ; Hill, 1972 : 37 ; Hill & Thonglongya, 1972 :
191.

HOLOTYPE. Adult male No. i57a in the Indian Museum, Calcutta. In alcohol.
Presented by W. T. Blanford.

TYPE LOCALITY. The holotype is said by Blanford (1891 : 318) to have been
obtained east of Moulmein in Burma.

DIAGNOSIS. Much smaller than H. tickelli or H. tomesi ; internarial region
densely haired ; tragus (Fig. ic) less acutely pointed ; facial warts imperceptible;
upper surface of forearm densely haired ; a broad, cushion-like pad at base of thumb ;
braincase flattened ; postorbital region relatively much wider ; outer upper incisor
(i3) greatly reduced ; third molars (m |) more reduced.

DESCRIPTION. Muzzle very low, wide, moderately furred, nostrils semi-lunate,
opening sublaterally, no perceptible grooves behind each nostril ; internarial
region densely clothed with short hairs to lower edge of nostrils ; lower lip naked
except for a few sparse hairs ; a prominent naked area beneath the symphysis menti
preceding a low or almost imperceptible wart ; facial warts almost imperceptible or
absent ; ear nearly subquadrate but rather more triangular than in H . tickelli or
H. tomesi, its anterior margin strongly convex proximally, with well-developed,
roundly pointed basal lobe, distally nearly straight to broadly rounded tip ; posterior
margin of ear slightly convex in its distal third, its lower part similarly convex but
with a shallow central concavity ; antitragal lobe thickened ; a low, barely percep-
tible wart, with a few longer hairs, just above the junction of the posterior margin of
the ear with the side of the head ; outer surface of conch furred for about one third

22 J. E. HILL

of its length, a few sparse hairs internally. Tragus (Fig. le) slightly hatchet-shaped,
the inner margin sharply concave to a rounded, anteriorly directed tip, upper
margin nearly horizontal, posterior margin straight or slightly convex, with a large
triangular lobe at its base.

Thumb very short with strong claw, the upper surface of the thumb clothed with
short hairs ; a broad, well-developed pad or cushion extending over the base of the
thumb and of the second metacarpal ; third and fourth metacarpals usually sub-
equal, fifth a little shorter ; upper surface of forearm densely clothed with fine, short
hairs ; upper surface of tibia naked ; feet very small, the sole thickened, a few sparse
hairs on the digits ; prominent, rounded post-calcarial lobe supported by a supple-
mentary cartilaginous calcarial spur. No male example in alcohol is available but
Dobson (1877 : 313) states that the penis of the holotype resembles that of H.
tickelli. Dorsal surface bright chestnut brown, ventral surface similar in colour.

Skull (PI. 4a-c) very small, braincase flattened, wide, with lambdoid ridges but no
sagittal crest ; frontal region barely elevated, the cranial profile almost straight ;
postorbital region wide, its width exceeding the width across the canines at the
alveoli ; a trace only of supraorbital ridges ; slightly inflated lateral rostral swellings
with small supraorbital tubercles ; a shallow median rostral depression ; narial
emargination deep, extending posteriorly to a line joining the posterior margins of
the anteorbital foramina, U-shaped ; anteorbital foramen large, enclosed by a
narrow bar ; zygomata slender ; lower maxillary margin of orbit flange-like ;
pre-palatal emargination wide anteriorly, its lateral margins extending beyond the
inner faces of the inner upper incisors (i2~2), the emargination extending posteriorly
almost to a line joining the posterior faces of the canines (c1-1) ; posterior edge of the
emargination sometimes with a narrow tongue-like extension into the palate, on
occasion extending as far as a line joining the mesostyles of the first upper molars
(m1-1). Soft palate (Fig. 2d) with six post-canine ridges, the first straight, unbroken,
the second slightly curved, narrowly separated, remainder curved, medianly divided,
the last small. Post-palatal extension short, wide, with prominent, broad post-
palatal spine ; shallow basioccipital pits not excised sharply into the basisphenoid,
separated by a broad ridge.

Upper incisive dentition similar to that of H. tomesi but outer incisor (i3) very
small, sometimes minute, one quarter or less the basal area of the inner tooth (i2),
with prominent cingulum and small central cusp, intruded to lie directly behind i2,
its anterior face in contact with the posterior face of that tooth, its outer face in
contact with the inner face of the canine (c1), i2 and c1 barely separated. Upper
premolar (pm4) not compressed in the toothrow, its length and width almost equal ;
third upper molar (m3) less than one half the crown area of first upper molar (m1),
with small metacone and much reduced second and third commissures ; lower
incisors (^-3) linear, only slightly if at all imbricated, i3 a little more massive than
ij-2 ; anterior lower premolar (pm2) small, its crown area about one quarter that of
the second lower premolar (pm4), rather compressed in the toothrow ; posterior
triangle of third lower molar (m3) much reduced, the hypoconid small, the entoconid
almost obsolete, the posterior part of the tooth about one quarter to one third the
area of the anterior triangle (PI. 4a, c, g).

BATS REFERRED TO HESPEROPTEN US

FIG. 5. Hesperoptenus blanfordi, distribution.

Measurements of H. blanfordi appear in Tables i and 2.

DISTRIBUTION (Fig. 5). BURMA (Dobson, 1877 : 312 ; 1878 : 242 ; Anderson,
1881 : 133 ; Blanford, 1891 : 318) ; THAILAND (Jentink, 1888 : 182 (Jongo Hills,
perhaps = Chongo Hills, io°i7'N 99°oi'E) ; Robinson & Kloss, 1915 : 116 ; Thomas,
1916 : 2 ; Gyldenstolpe, 1919 : 137 ; Hill & Thonglongya, 1972 : 191 ); MALAYA
(Anderson, 1881 : 133 ; Blanford, 1891 : 318 ; Thomas, 1916 : 2 ; Hill, 1972 : 37).

REMARKS. Flattening and broadening of the braincase, with a wide postorbital
region and rostrum, is characteristic of the vespertilionine genera Tylonycteris and
Mimetillus, which also have pads on the thumbs and feet. In these, however, the
braincase is yet more flattened than in H. blanfordi and no extensive shortening of the
rostrum has occurred. Rostral inflation in H. blanfordi is of the same order as that
in Tylonycteris, with the supraorbital area swollen to develop sometimes a small
tubercle. In Mimetillus the whole of this area is inflated, together with the upper
part of the bar enclosing the anteorbital foramen and the region immediately above
this aperture, but with no definite tubercle. Although Tylonycteris and Mimetillus
have the same dental formula as Hesperoptenus, the inner upper incisor (i2) is bicuspid
and the outer upper incisor (i3) remains in the toothrow, separated from the canine

24 J. E. HILL

by a small interspace. Further, in Mimetillus the wing is much reduced by shorten-
ing of the third and fifth digits ; in H. blanfordi the digits are not reduced and indeed
are relatively rather long.

Of other vespertilionine bats with pads on thumbs and feet, Eudiscopus denti-
culus has an extreme degree of flattening of the skull with uninflated, unshortened,
rostrum and retains the anteriormost upper premolar (pm2) with a minute second
lower premolar (pm3) intruded between the anterior (pm2) and posterior (pm4)
lower premolars. The braincase is higher in Myotis rosseti which has a caniniform
outer upper incisor (i3) exceeding the inner tooth (i2) in basal area and which also
retains the anteriormost upper premolar (pm2) ; Glischropus tylopus cranially is like
Pipistrellus, with the rostrum unexpanded and not much shortened although i3 is
extruded to lie alongside i2, the four incisors forming a straight line, and with pm2
present and retained in the toothrow. In all of these, i2 remains rather elongate as
it is in Tylonycteris and Mimetillus, and is similarly bicuspid. The massive, uni-
cuspid i2 and displaced i3 ally blanfordi with H. tickelli and H. tomesi which it almost
exactly resembles in the features of its upper incisive dentition although in a number
of other features it differs quite sharply from either of these species.

RELATIONSHIPS

It is clear that Hesperoptenus as here understood embodies two rather widely
separated trends, one represented by a single species (doriae), the other by three
(tickelli, tomesi and blanfordi). At the same time, one species of the latter group
(blanfordi} is quite sharply removed from its associates and apparently represents
a further course of adaptation.

Thomas (1902 : 220) thought Hesperoptenus related to Philetor but apart from any
other considerations the upper incisors of Philetor differ widely, the inner tooth (i2)
being elongate, long and bicuspid, with the outer tooth (i3) not displaced. Indeed,
Philetor is closely related to Pipistrellus (Hill, 1966 : 380 ; 1971 : 143). Miller
(1907 : 211) compared the dentition of Hesperoptenus with Eptesicus and Vespertilio
but examined only tickelli, and Tate (1942 : 232, 233, 268) included the genus among
the 'pipistrelloid' genera. Tate added (p. 269) that in tickelli (he did not examine
either doriae or blanfordi) the upper incisors show an interesting intermediate con-
dition between the Pipistrellus-Uke genera which have mostly a bicuspid inner tooth
(i2), with the outer tooth (i3) present, and the Scotophilus-like genera in which i2 is
unicuspid and i3 is lost. In doriae, tickelli, tomesi and blanfordi i2 is massive and
unicuspid and in all but the first of these i3 is markedly displaced inwardly from the
toothrow, this feature reaching an extreme in tomesi and blanfordi. Hesperoptenus
doriae indeed stands nearest to the pipistrelline genera and has a number of similari-
ties to Glauconycteris and, to a lesser extent, to Chalinolobus. Its braincase is rather
inflated, its rostrum not much shortened or broadened to the extent that it is in
tickelli, tomesi or blanfordi, and its outer upper incisors are not greatly displaced as
they are in its congeners. It is therefore considered the least modified member of
Hesperoptenus.

The remaining species tickelli, tomesi and blanfordi in many ways approach the
'nycticeine' genera (Tate, 1942 : 280) allied to Scotophilus, and, indeed, in these the

BATS REFERRED TO H ESPEROPTEN US 25

elongated braincase and greatly broadened rostrum are features strongly reminiscent
of that genus or of Scotomanes. The large species tickelli and tomesi are similar to
Scotophilus in the general appearance of the head and ears and also in the shape of
the tragus, which is widened to some extent at its centre, rather than elongate and
more or less spatulate as in the pipistrelline genera. In fact, the early authors
referred tickelli to Nycticejus (in part = Scotophilus) and, on occasion, even relatively
modern specimens referred by previous identifiers to Hesperoptenus have been found
to be examples of Scotophilus.

The nycticeine genera are characterized among other features by the unicuspid
inner upper incisor (i2) and by the loss of the outer upper incisor (i3) and of the small
anteriormost upper premolar (pm2) although at the base of their sequence a rudi-
mentary i3 is sometimes present in Scotozous which also retains pm2 ; the latter tooth
also occurs sometimes in Scotoecus. Scotozous has been thought to be pipistrelline
rather than nycticeine (Tate, 1942 : 259) but, if the African rueppellii is excluded,
then, as Tate (p. 260) postulated, the genus with its remaining Indian species
dormer i is transitional between Pipistrellus and the nycticeine genera in which i3
and pm2 are lost. In Hesperoptenus, contrary to the usual nycticeine condition, i3
is present and furthermore usually well developed. However, shortening of the
rostrum has led to its intrusion from the toothrow in three of four species, and, as in
most nycticeines, pm2 has been lost. It is effectively further separated from
Scotophilus by the lack of any distortion of the W-pattern of the first and second
upper molars (m1-2) and from Scotomanes by its less widely expanded rostrum, the
anterior orbital margin not swollen or beaded (although the flange-like margin of
tomesi tends towards this condition), and its lack of a pronounced rostral depression.

Hesperoptenus provides, therefore, a further example of a tendency apparent
among some Vespertilionidae to shorten and broaden the rostrum with concomitant
reduction of the dentition, in some by the loss of the outer upper incisor (i3) with
consequent enlargement of the inner upper incisor (i2), the loss of the anterior upper
premolar (pm2), and some reduction of the third upper molar (m3). The process
has not progressed as far in Hesperoptenus as in the genera most modified in these
respects and, indeed, has taken a different course, with the removal of i3 from the
toothrow rather than its loss. One species (doriae] remains relatively unmodified,
and one (blanfordi) has clearly diverged rather widely with further modification of
the skull and with the development of pads on the thumbs. The latter, however,
occur quite independently in various sections of the Vespertilioninae, in the 'myotine'
type in Eudiscopus and Myotis, the 'pipistrelline' type in Glischropus, Tylonycteris
and Mimetillus, and in blanfordi in a species that in many ways approaches the
'nycticeine' type. The conclusion seems unavoidable that Hesperoptenus must be
placed with the genera Scotozous, Scotoecus and Nycticeius that link the more truly
pipistrelline bats to the more modified, extreme nycticeine bats such as Scotomanes
and Scotophilus.

ACKNOWLEDGEMENTS

My thanks are due to Professor Enrico Tortonese of the Museo Civico di Storia
Naturale 'Giacomo Doria', Genoa, Italy, who made possible my examination of the

26 J. E. HILL

holotype of Hesperoptenus doriae (Peters, 1869), a specimen so vital to this study
that without it the work could not have been properly undertaken. I am indebted
also to Dr J. C. Moore, until lately of the Field Museum of Natural History, Chicago,
U.S.A., for the loan of an important specimen from Sabah now in that collection ;
to Mr Lim Boo Liat, A.M.N., of the Institute for Medical Research, Kuala Lumpur,
Malaya, for specimens of Hesperoptenus tomesi and H. blanfordi and to Dr K. F.
Koopman, of the Department of Mammalogy, the American Museum of Natural
History, New York, U.S.A., for much valuable discussion.

REFERENCES

ANDERSON, J. 1881. Catalogue of Mammalia in the Indian Museum, Calcutta. Calcutta.
BLANFORD, W. T. 1891. The Fauna of British India including Ceylon and Burma. Mammalia.

Part II. London.
BLYTH, E. 1851. Report on the Mammalia and more remarkable species of birds inhabiting

Ceylon. /. Asiat. Soc. Beng. 20 : 153-185.

1863. Catalogue of 'the Mammalia in the Museum of the Asiatic Society of Bengal. Calcutta.

BROSSET, A. 1962. The bats of central and western India. Part II. J .Bombay nat. Hist. Soc.

59 : 583-624, 4 pis, 9 maps.
CHASEN, F. N. 1940. A handlist of Malaysian mammals. Bull. Raffles Mus. No. 15 : i-xx,

1-209, map.
DAVIS, D. D. 1962. Mammals of the lowland rain-forest of North Borneo. Bull. natn. Mus.

St. Singapore, No. 31 : 1-129, 20 figs, 23 pis.
DOBSON, G. E. 1876. Monograph of the Asiatic Chiroptera and Catalogue of the species of bats

in the Collection of the Indian Museum, Calcutta. London.

— 1877. Notes on a collection of Chiroptera from India and Burma, with descriptions of new
species. /. Asiat. Soc. Beng. 46, (2) : 310-313.

— 1878. Catalogue of the Chiroptera in the Collection of the British Museum. London.
1880. Report on accessions to our knowledge of the Chiroptera during the past two years

(1878-80). Report of the British Association 169-197.
ELLERMAN, J. R. & MORRISON-SCOTT, T. C. S. 1951. Checklist of Palaearctic and Indian

mammals, 1758-1946. London.
FITZINGER, L. J. 1870. Kritische Durchsicht der Ordnung der Flatterthiere oder Handfliigler

(Chiroptera). Familie der Fledermause ( Vespertiliones) . V. Abtheilung. Sber. Akad

Wiss. Wien 62, (i) : 353-438.
GYLDENSTOLPE, N. 1919. A list of the mammals at present known to inhabit Siam. /. nat.

Hist. Soc. Siam 3, (3) : 127-175.
HILL, J. E. 1966. A review of the genus Philetor (Chiroptera : Vespertilionidae) . Bull. Br. Mus.

nat. Hist. (Zool.) 14 : 373-382, 3 figs.

— 1967. The bats of the Andaman and Nicobar Islands. /. Bombay nat. Hist. Soc. 64 : 1-9.

— 1971. The status of Vespertilio brachypterus Temminck (Chiroptera : Vespertilionidae).
Zool. Meded. Leiden 45 : 141-146.

— 1972. New records of Malayan bats, with taxonomic notes and the description of a new
Pipistrellus. Bull. Br. Mus. nat. Hist. (Zool.) 23 : 21-42, 3 tabs.

— & THONGLONGYA, K. 1972. Bats from Thailand and Cambodia. Bull. Br. Mus. nat.
Hist. (Zool.) 22 : 171-196, 4 figs, 2 tabs.

HORSFIELD, T. 1851. A Catalogue of the Mammalia in the Museum of the Hon. East-India

Company. London.

HOSE, C. 1893. A descriptive account of the mammals of Borneo. London.
JENTINK, F. A. 1887. In Schlegel, H. & Jentink, F. A. Musdum d'Histoire Naturelle des

Pays-Bas. Revue methodique et critique des Collections deposees dans cet Etablissement.

Tome 10. Catalogue Qsteologique des mammiferes. Leiden.

BATS REFERRED TO HESPEROPTEN US 27

JENTINK, F. A. 1888. In Schlegel, H. & Jentink, F. A. Museum d'Histoire Naturelle des

Pays-Bas. Revue mdthodique et critique des Collections dtposdes dans cet Etablissement. Tome

12. Catalogue Systdmatique des mammiferes (Rongeurs, Insectivores, Cheiropteres , Edentis et

Marsupiaux}. Leiden.
KELAART, E. F. 1850. List of Mammalia observed or collected in Ceylon. // R. Asiat. Soc.

Ceylon 2 (2) : 316-320 (pagination from reprint of 1890).

1852. Prodromus Faunae Zeylanicae : being contributions to the Zoology of Ceylon. Colombo.

KLOSS, C. B. 1908. A list of the bats occurring in the Peninsular Region with a key to the

genera. /. fed. Malay St. Mus. 2 : 151-161.
KOOPMAN, K. F. 1971. Taxonomic notes on Chalinolobus and Glauconycteris (Chiroptera,

Vespertilionidae) . Am. Mus. Novit. No. 2451 : i-io, i fig.
MEDWAY, LORD. 1965. Mammals of Borneo. Field keys and an annotated checklist. /.

Malay. Brch R. Asiat. Soc. (1963), 36 : i-xiv, 1-193, 9 ngs. 34 pis, 5 tabs, map.
1969- The wild mammals of Malaya and offshore Islands including Singapore. Kuala

Lumpur ; London.
MILLER, G. S. 1907. The families and genera of bats. Bull. U.S. natn. Mus. 57 : i-xvii,

1-282, 49 figs, 14 pis.
PETERS, W. i869a. Mittheilung iiber die von dem Hon. Marquis Giacomo Doria in Sarawak

auf Borneo gesammelten Flederthiere. Mber. K. preuss. Akad. Wiss. (1868) : 626-627.
i86gb. Mittheilung iiber neue Saugethiere (Colobus, Rhinolophus, Vesperus) und

neue oder weniger bekannte Amphibien (Hemidactylus, Herpetodryas, Spilotes, Elaphis,

Lamprophis, Erythrolamprus) . Mber. K. preuss. Akad. Wiss. (1868) : 637-643.
1906. Chiroptera of Berlin Museum. Berlin

PETERSON, R. L. & SMITH, D. A. 1973. A new species of Glauconycteris (Vespertilionidae,

Chiroptera). Life Sci. Occ. Pap. R. Ont. Mus. No. 22 : 1-9, 4 figs.
PHILLIPS, W. W. A. i932a. Survey of the distribution of mammals in Ceylon. Report No. 2.

Ceylon J. Sci. B, 16 : 341-347.
I932b. Survey of the distribution of mammals in Ceylon. Report No. 3. Ceylon J. Sci.

B, 16 : 347-351-
I932C. Survey of the distribution of mammals in Ceylon. Report No. 5. Ceylon J. Sci.

B, 17: 131-134-
I932d. Survey of the distribution of mammals in Ceylon. Report No. 7. Ceylon J. Sci.

B, 17: 136-138.
1933- Survey of the distribution of mammals in Ceylon. Report No. 8. Ceylon J. Sci.

B, 17: 237-242.

1935- Manual of the mammals of Ceylon. Colombo.

ROBINSON, H. C. & KLOSS, C. B. 1915. On a collection of mammals from the Siamese Province

of Bandon, N. E. Malay Peninsula. J. fed. Malay St. Mus. 5 : 111-127.
ROSEVEAR, D. R. 1965. The bats of West Africa. London.
RYAN, R. M. 1966. A new and some imperfectly known Australian Chalinolobus and the

taxonomic status of African Glauconycteris. J. Mammal. 47 : 86-91, i tab.
STERNDALE, R. A. 1884. Natural History of the Mammalia of India and Ceylon. Calcutta.
TATE, G. H. H. 1942. Review of the vespertilionine bats, with special attention to genera

and species of the Archbold Collections. Bull. Am. Mus. nat. Hist. 80 : 221-297, 5 figs.
THOMAS, O. 1902. On a new genus of vespertilionine bat from New Guinea. Ann. Mag. nat.

Hist. (7), 9 : 220-222.

1905. A new genus and two new species of bats. Ann. Mag. nat. Hist. (7), 16 : 572-576.

1913. On some specimens of Glauconycteris from the Cameroons. Ann. Mag. nat. Hist.

(8), 11 : 144-145.
1916. List of Microchiroptera, other than leaf-nose bats, in the collection of the Federated

Malay States Museums. /. fed. Malay St. Mus. 7 : 1-6.
TROUESSART, E-L. 1897. Catalogus mammalium tarn viventium quam fossilium. I. Primates,

Prosimiae, Chiroptera, Insectivora, Carnivora, Rodentia, Pinnipedia. Berlin.

28 J. E. HILL

TROUESSART, E-L. 1904. Catalogus mammalium tarn viventium quam fossilium. Quinquen-

nale supplementum. Berlin.
WAGNER, J. A. 1855. Die Saugethiere in Abbildungen nach der Natur mil Beschreibungen von

Dr. Johann Christian Daniel von Schreber. Supplementband 5. Leipzig.

WROUGHTON, R. C. 1897. Some Konkan bats. /. Bombay nat. Hist. Soc. 12 : 716-725, i pi.
1912. Bombay Natural History Society's Mammal Survey of India. No. 5. Dharwar.

/. Bombay nat. Hist. Soc. 21 : 1170-1195.

— 1913. Bombay Natural History Society's Mammal Survey of India. No. 6. Kanara.
/. Bombay nat. Hist. Soc. 22 : 29-44.

— igisa. Bombay Natural History Society's Mammal Survey of India, Burma and Ceylon.
Report No. 18. Ceylon. /. Bombay nat. Hist. Soc. 24 : 79-96.

— I9i5b. Bombay Natural History Society's Mammal Survey of India, Burma and Ceylon.

Report No. 19. Bengal, Bihar and Orissa. /. Bombay nat. Hist. Soc. 24 : 96-110.
1917. Bombay Natural History Society's Mammal Survey of India, Burma and Ceylon.

Report No. 27. Bhutan Duars. /. Bombay nat. Hist. Soc. 25 : 62-71.
1918. Summary of the results from the Indian Mammal Survey of the Bombay Natural

History Society. /. Bombay nat. Hist. Soc. 25 : 547-598.

J. E. HILL

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD,

LONDON swy 530

PLATE i

Hesperoptenus doriae, holotype, Genoa Museum CE 40094, £, Sarawak. Ear and tragus.
Skull, approx. x 4 ; lingual aspect of mandible, approx. x 5-5.

Bull. Br. Mus. nat. Hist. (Zool.) 30, i

PLATE i

PLATE 2
Hesperoptenus doriae, holotype, Genoa Museum CE 40094, 5*, Sarawak. Dentition.

Bull. Br. Mus. nat. Hist. (Zool.) 30, i

PLATE 2

PLATE 3

Skull of: (a-c) Hesperoptenus tickelli, BM 12.11.28.46, $, India, approx. X2; (d-f) H.
tomesi, holotype, BM 7.1.1.428, <$, Malaya, approx. X2.

Bull. Br. Mus. nat. Hist. (Zool.) 30, i

PLATE 3

PLATE 4

Skull of: (a-c) Hesperoptenus blanfordi, BM 65.345, ?, Malaya, approx. x 3. Maxillary
dentition of: (d) H. doriae, BM 75.1873, ?, Malaya, approx. X5'5; (e) H. tickelli, BM 12.11.28.46,
approx. x 3-5 ; (f) H. tomesi, holotype, BM 7.1.1.428, approx. x 4 ; (g) H. blanfordi, BM 65.345,
approx. x6-5.

Bull. Br. Mus. nat. Hist. (Zool.) 30, i

PLATE

A LIST OF SUPPLEMENTS
TO THE ZOOLOGICAL SERIES

OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

1. KAY, E. ALISON. Marine Molluscs in the Cuming Collection British Museum
(Natural History) described by William Harper Pease. Pp. 96 ; 14 Plates.
1965. (Out of Print.)

2. WHITEHEAD, P. J. P. The Clupeoid Fishes described by Lacepede, Cuvier and
Valenciennes. Pp. 180 ; n Plates, 15 Text-figures. 1967. £4.

3. TAYLOR, J. D., KENNEDY, W. J. & HALL, A. The Shell Structure and Mineralogy
of the Bivalvia. Introduction. Nuculacea-Trigonacea. Pp. 125 ; 29 Plates,
77 Text-figures. 1969. £4.50.

4. HAYNES, J. R. Cardigan Bay Recent Foraminifera (Cruises of the R.V. Antur)
1962-1964. Pp. 245 ; 33 Plates, 47 Text-figures. 1973. £10.80.

5. WHITEHEAD, P. J. P. The Clupeoid Fishes of the Guianas. Pp. 227 ; 72
Text-figures. 1973. £9.70.

6. GREENWOOD, P. H. The Cichlid Fishes of Lake Victoria, East Africa : the
Biology and Evolution of a Species Flock. Pp. 134 ; i Plate, 77 Text-figures.
1974. £3-75. Hardback edition £6.

Printed in Great Britain by John Wright and Sons Ltd. at The Stonebridge Press, Bristol BS4 5NU

TAXONOMY AND DISTRIBUTION

OF THE METAPHERETIMA
ELONGATA SPECIES-COMPLEX
OF INDO-AUSTRALASIAN /

EARTHWORMS !

(MEGASCOLECIDAE : OLIGOCHAETA)

E. G. EASTON

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. 2

LONDON: 1976

TAXONOMY AND DISTRIBUTION OF THE
METAPHERETIMA ELONGATA SPECIES-
COMPLEX OF INDO-AUSTRALASIAN

EARTHWORMS
(MEGASCOLECIDAE : OLIGOCHAETA)

*, <»

-$,.

BY

EDWARD GLYNN EASTON

Pp. 29-53 ; 6 Text-figures

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)

ZOOLOGY Vol. 30 No. 2

LONDON : 1976

THE BULLETIN OF THE BRITISH MUSEUM

(NATURAL HISTORY), instituted in 1949, is
issued in five series corresponding to the Scientific
Departments of the Museum, and an Historical series.

Parts will appear at irregular intervals as they
become ready. Volumes will contain about three or
four hundred pages, and will not necessarily be
completed within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 30 No. 2 of the Zoology series. The
abbreviated titles of periodicals cited follow those of the
World List of Scientific Periodicals.

World List abbreviation :
Bull. Br. Mus. nat. Hist. (Zool.)

ISSN 0007-1498

Trustees of the British Museum (Natural History), 1976

BRITISH MUSEUM (NATURAL HISTORY)

Issued 24 June, 1976 Price £1.95

TAXONOMY AND DISTRIBUTION OF THE
METAPHERETIMA ELONGATA SPECIES-
COMPLEX OF INDO-AUSTRALASIAN

EARTHWORMS
(MEGASCOLECIDAE : OLIGOCHAETA)

By E. G. EASTON

CONTENTS

Page
SYNOPSIS ........... 31

INTRODUCTION ........... 31

METHODS. ........... 32

RESULTS ............ 33

NOMENCLATURAL CONSIDERATIONS ....... 37

TAXONOMY ........... 38

DISTRIBUTION ........... 45

ACKNOWLEDGEMENTS ......... 48

REFERENCES ........... 48

APPENDIX. Material examined and other records from the Indo-Australasian
Archipelago .......... 50

SYNOPSIS

Following studies on the setal and spermathecal systems of the Metapheretima elongata
species-complex five species, elongata, everetti, kinabaluensis , phacellotheca and stelleri are recog-
nized. The taxonomy and indigenous distributions of each are described.

INTRODUCTION

DURING their numerical revision of the oriental earthworm genus Pheretima auct.
Sims & Easton (1972 : 252) proposed that the Indo-Australasian earthworm Meta-
pheretima elongata (Perrier, 1872) and its allies should be recognized as a single
species-complex. They assigned eighteen nominal species and subspecies to the
complex and tabulated the variation in expression of several characters. The lack
of obvious correlations, however, prevented the delineation of discrete groups of
taxa from among the relatively short series of material which they examined.

Subsequently more detailed investigations have continued into the taxonomy of
the acaecate members of the Pheretima group of genera (Archipheretima, Ephemitra,
Metapheretima and Planapheretima] and in particular into the members of the
Metapheretima elongata species-complex which have received special scrutiny.
During this latter study the material was located on which many of the taxa and
records of the elongata species-complex were based. Additionally material of the
species Metapheretima phacellotheca (not previously included in the elongata group
of species) was also examined from which it was concluded that this species too is

32 E. G. EASTON

closely allied to the complex. Recent additions to the collections of the British
Museum (Natural History) also included many examples of the species-complex.
Examination of all this material has provided the data presented in this report
from which it is possible to recognize five species and to establish the indigenous
distributions of each.

METHODS

During revision of the genus Metapheretima it became evident that discrete
taxonomic assemblages could be recognized in the elongata species-complex on the
expression of three characters :

1. The volume of a single segment. This value is derived from measurements of
the diameter and length of the segment. In specimens that have been dis-
sected the circumference is measured instead of the diameter. For convenience
during the study, the segment selected for measurement was vii since the
volume determination could be assessed when making the setal count (see
below) so that a direct correlation could be established between segmental
volume and setal number.

2. Number of setae. During preliminary studies it was observed that the setal
number of vii, in addition to being easier to determine, was also less variable
than that of xx, the other segment sometimes selected for study.

3. Number of spermathecae. The spermathecae of members of the elongata
species-complex, although occasionally absent, are usually arranged in paired,
ventrolateral batteries of up to 28 spermathecae, opening into intersegmental
furrows 5/6 and/or 6/7. The number of pores in each spermathecal battery
was recorded.

The three characters employed here as taxonomic criteria differ somewhat from
those utilized by Sims & Easton (1972 : 256). These authors used the length of
an individual as an expression of size since this value was readily available in pub-
lished descriptions. During the present study it became evident that segmental
volume provided a more accurate indication of size owing to the vicissitudes of
relaxation and fixing techniques causing wide fluctuations in overall length in
similarly proportioned individuals. Initially segment number was also employed
in this study but the large proportion of the specimens available for examination
proscribed by damage or regeneration caused the data to lack significance. Some
authors (Beddard & Fedarb, 1895 ; Michaelsen, 1899, 1922) considered the number
of segments bearing genital markings to be of taxonomic importance for distinguish-
ing the species of the elongata species-complex. However, when this character was
assessed and correlated with the other results the degree of overlap of the groups
made it impractical to utilize it as a taxonomic criterion although several groups of
individuals could be recognized which were related to those established using the
criteria discussed above.

The majority of the taxa included within the elongata species-complex by Sims &
Easton (1972) are known only from one or two series and whenever possible all the
members of those series were examined in this study. Only in the cases of the

METAPHERETIMA ELONGATA SPECIES-COMPLEX 33

common forms biserialis and elongata, often recorded from outside the Pheretima
domain, were the number of specimens examined restricted by sampling. These
two forms are usually considered synonymous and Sims & Easton (1972 : 256) sug-
gested that they originated probably in the area bounded by Sumatra, Philippines,
Celebes and Java. They concluded that all other records were the result of intro-
duction. Gates (1972 : 182) however was of the opinion that these forms had
originated in a smaller area : Borneo, Celebes and Philippines. All the records
that could be traced from these areas were included in the study as well as a selection
from several new series collected recently in Bali, Lombok, Sumbawa and Komodo.
For many years phacellotheca has been known only from the type-series which is now
unsuitable for study. It was found however that two specimens from Buru identi-
fied by Michaelsen (i934a) as stelleri bonensis so closely resembled the original
description of phacellotheca as to require re-identifying with the result that they
were included in the study as representatives of this species.

RESULTS

Correlation between each pair of characters was studied separately.

i. Setal number : segmental volume. The relationship between setal numbers and
the volume of segment mi was tested (Fig. i) and found to provide a means of

100-
90-
80-
70-

60-

*«,-

o°

o o

1 2 3 4 5 6 7 8 9 10 20 40 60 80 100 200

Seqmental volume (mm3)

FIG. i. Relationship between the setal number and segmental volume of segment vii
(scales logarithmic). Origins of material: O Mt Kinabalu (above 2 loom) ; • other
localities.

34 E. G. EASTON

recognizing two distinct groups of individuals. The smaller cluster (depicted by
circular symbols with star-shaped centres) includes only examples from the upper
slopes of Mt Kinabalu, Sabah (series A : Sims & Easton, 1972 : 253), while the
larger cluster (depicted by solid circular symbols) contains examples from all other
parts of the elongata species-complex range. Examples from the larger cluster have
considerably higher setal numbers for a given segmental volume than those of the
smaller cluster. Examination of the other individuals from the samples identified as
series A revealed that the majority of them belong to the smaller cluster.

In both groups there is a positive correlation between setal number and seg-
mental volume. The size of an individual, and therefore the segmental volume,
increases during an individual's life and it may be deduced from Fig. i that the
number of setae also increases. This conclusion has a wide significance in view of
the diagnostic value usually attributed to setal number. It is implicit from the
literature that the setal number of a segment is assumed to remain fairly constant
(allowing for loss, etc.) throughout the life of an individual ; indeed, many of the
'species' in the Pheretima-complex have been distinguished by smaller differences in
the setal numbers than are recorded here.

2. Anterior spermathecal batteries : posterior spermathecal batteries. In the speci-
mens studied each spermathecal battery of an individual rarely contained an
identical number of pores although the discrepancies occurring between the right
and left batteries within a furrow were small and apparently haphazard. The
relationship between the mean number of spermathecae in a battery in furrow 5/6
and in 6/7 was tested (Fig. 2). A positive correlation was established between the
sizes of the anterior and posterior batteries for most of the specimens with
the posterior pair containing, on average, 20% more spermathecal pores than the
anterior pair.

Two specimens, both from Buru (depicted by star-shaped symbols), depart sig-
nificantly from the general trend. Both these specimens lack spermathecal batteries
in furrow 6/7 (confirmed by internal examination) although there are large anterior
batteries. Some of the other specimens also lacked posterior spermathecal bat-
teries, but in all these individuals the anterior spermathecal batteries were very
small or absent.

Some signs of clustering may be detected in the larger group of specimens, i.e.
those with the usual spermathecal arrangement. Specimens from Sumbawa and
Komodo (solid square symbols) and Lombok (open square symbols) have small
spermathecal batteries, individuals from Borneo (solid circular symbols) and Celebes
(open triangular symbols) have batteries of intermediate size while those from
Balabac Island (open circular symbols) and Sangihe (solid triangular symbols) have
large spermathecal batteries. The individuals from Mt Kinabalu (noted in Fig. i
because of their low setal numbers) were indistinguishable from other specimens
from Borneo.

3. Setal number : size of spermathecal batteries. Because of the strong positive
correlation noted between the relative sizes of the anterior and posterior sperma-
thecal batteries in the main cluster in Fig. 2 the regression line of the main cluster
has been utilized as an axis in Fig. 3. Since this line passes close to the origin of

METAPHERETIMA ELONGATA SPECIES-COMPLEX 35

• A

• O A

*• • O

-
*•••

D

A A 00*

DD D ••

D

D • ••••>
•,••••• •• *•

•!•••••

•"••••

am am _

Mean number of spermathecae in posterior batteries

FIG. -2. Relationship between the mean number of spermathecae in the anterior (5/6) and
posterior (6/7) batteries. Origins of material : • Sumbawa and Komodo ; Q Lombok ;
• Borneo ; O Balabac Island ; A Sangihe ; A Kepulaud Taulaud ; A Celebes ;

Fig. 2 the positions of specimens along this axis are determined by the distance
each occurs from the origin. (The positions of specimens along this axis may be
calculated directly from the formula x = -\/(a2 + b2), where a and b are the mean
sizes of the anterior and posterior spermathecal batteries.) To facilitate the recog-
nition of clusters, members of each sample are linked.

Most of the specimens plotted in Fig. 3 form two discrete clusters. The remain-
ing individuals are scattered in an area, peripheral to the larger of these clusters,
which is characterized by large spermathecal batteries and low setal numbers. The
first cluster, characterized by high setal numbers and small spermathecal batteries,
contains all of the specimens from Java, Bali, Sumbawa and Komodo (depicted by
solid square symbols). The series from which these specimens were selected were
remarkable in the high proportion of athecate individuals present : approximately
half the clitellate portion of each series lacked spermathecae and therefore had to
be excluded from this figure. Setal numbers of athecate individuals agreed closely
with the thecate individuals plotted. The second cluster, containing individuals

E. G. EASTON

Spermathecal number

FIG. 3. Relationship between the setal numberof segment vii and the spermathecal number,
V(«2 + &2) (scales logarithmic). Members of each series are linked. Origins of material :
| Sumbawa, Komodo, Bali and Java ; D Lombok ; • Borneo (individuals with a high
setae : volume ratio) ; O Borneo (individuals with a low setae : volume ratio) ; O Balabac
Island ; A Sangihe ; A Kepulaud Taulaud ; A Celebes.

with a greater range of setal numbers and larger spermathecal batteries, includes all
of the specimens from Balabac Island (open circular symbols), Celebes (open tri-
angular symbols) and Borneo (solid circular symbols) with the exception of the
individuals from Mt Kinabalu (circular symbols with star-shaped centres) which
were noted (Fi) for their low setal numbers.

Specimens from Lombok (open square symbols) occur in both clusters. Indi-
viduals from low altitudes (c. 50 m) however belong to cluster i while, with one
exception, those from high altitudes (c. 400 m) belong to cluster 2.

The specimens situated peripherally to cluster 2 are those from Sangihe (solid
triangular symbols), Kepulaud Taulaud (partially filled triangular symbols) and
Mt Kinabalu (circular symbols with star-shaped centres). Since these may be
distinguished by the difference in their setal development it is considered that two
further clusters should be recognized among material from this area ; one containing
the specimens from Sangihe and Kepulaud Taullaud (cluster 3) and the other the Mt
Kinabalu specimens (cluster^ . Although the individuals of cluster 4 resemble those
of cluster 3 in their relationships between setal and spermathecal development, it
should be recalled (Fig. i) that the setal development, compared with the segmental
volume, is reduced in members of cluster 4, while the relationship of the segmental

METAPHERETIMA ELONGATA SPECIES-COMPLEX

37

volume to the spermathecal development in individuals in this cluster is closer to
that found in cluster 2.

The two specimens from Buru (depicted by star-shaped symbols in Fig. 2) have
not been plotted in Fig. 3 since they did not occur in the main cluster in Fig. 2. They
most closely resemble specimens from Sangihe and Kepulaud Taulaud (cluster 3) on
setal numbers and the size of the anterior spermathecal batteries.

The marker attributes of the five groups of individuals recognized above have
been tabulated in Table i. Morphologically groups i, 2 and 3 form a series in which
the only differences noted are increases in the sizes of spermathecal batteries in
individuals of a similar size. Group 4 resembles group 2 but has lower setal num-
bers. Group 5 differs from group 3 only in lacking spermathecal batteries in
furrow 6/7.

TABLE i

Marker attributes of the morphological groups

Group

1. (elongata)

2. (evevetti)

3. (stelleri)

4. (kinabaluensis)

5. (phacellotheca)

* See text.

Setae : segmental
volume (Fig. i)

high
high
high
low
high

Positions of spermathecal
pores (Fig. 2)

5/6 and 6/7
5/6 and 6/7
5/6 and 6/7
5/6 and 6/7
5/6 only

Size of spermathecal
batteries (Fig. 3)

small (cluster i)
intermediate (cluster 2)
large (cluster 3)
intermediate* (cluster 4)
large

NOMENCLATURAL CONSIDERATIONS

The type-series of nine of the eighteen nominal species and subspecies assigned to
the elongata species-complex were examined during the preparation of this report,
thus permitting these names to be readily allocated to the appropriate morpho-
logical group. The types of elongata Perrier, 1872 and biserialis Perrier, 1875 were
not examined, but Michaelsen (igioc) had previously synonymized the taxa and
there is no evidence to dispute this decision. On the basis of Michaelsen 's published
description, as well as those of more recent workers (Stephenson, 1923 ; Gates,
1972) the types as well as the majority of records of both elongata and biserialis can
be assigned to group i as also may one other taxon, acystis Beddard, 1895.

Although the type-series of seven other taxa could not be located during the
preparation of this report, it is possible to estimate approximate co-ordinates in
Figs i, 2 and 3 for each of the taxa from the original descriptions. None of the
original descriptions recorded segmental volume but approximate values have been
calculated based on the mean segmental lengths (total length/segment number) and
the maximum diameters. The estimated co-ordinates and the group to which each
taxon is assigned are shown in Table 2 where data regarding stelleri everetti :
Michaelsen, 1899 are also recorded.

In Table 2 it is uncertain to which group kinabaluensis Beddard & Fedarb, 1895
should be assigned. The original description lacks any data regarding setal numbers

E. G. EASTON

TABLE 2

Estimated co-ordinates based on published data

Taxon

baritoensis

beranensis beranensis
kinabaluensis
phacellotheca
stelleri annectens
stelleri bonensis
stelleri everetti

(sensu Michaelsen, 1899)
stelleri koroensis

Segmental
volume

(mm3)

"•5
i-i

47'7
14-2
28-8
95'4

126-2
176-6

Setal

number

on vii

49

47

46
40

85
60

Positions of

spermathecal

batteries

5/6/7

5/6/7

5/6/7

5/6

5/6/7

5/6/7

5/6/7
5/6/7

Size of

spermathecal
batteries

4'7
8-1

IO-I

_ *
26

28

8

Group

2

2

2, 3 or 4
5
3

2

3

2

* Size of spermathecal batteries not recorded.

so it could belong to any of groups 2, 3 or 4. It seems unlikely to belong to group 3
because the type locality of kinabaluensis is Tamburungare, Mt Kinabalu, Sabah,
at an altitude of 2350 m (7700 ft) and only groups 2 and 4 are known from Borneo.
It is however proposed to assign kinabaluensis to group 4 since it is evident from
Sims & Easton (1972) that group 4 (series A) is the commonest form at altitudes of
more than 2100 m.

Information on three records from the Borneo-Celebes area, stelleri : Michaelsen,
1896 : 202 ; stelleri (typica) : Ude, 1932 : 155 and elongata : Ude, 1932 : 155, lacks
sufficient detail to establish even approximate co-ordinates for this material. Never-
theless, as the two records of stelleri are from Borneo at altitudes at which group 2
is believed to be the only representative of the complex present, both are assigned
to that group, while the record of elongata, from south Celebes, is allocated to
group i. Although all other records from Celebes have been assigned to groups 2,
3 or 5, they are from the north and west which are ecologically distinct from the
south which resembles Lombok and Sumbawa where group i is indigenous (see
Distribution).

Each morphological group may be recognized as forming a distinct taxonomic
unit. The status of each and its relationships with the others still require con-
siderable study but since there is no evidence of hybridization, although several
samples studied contain representatives of more than one group, it is proposed to
consider each as a distinct species while retaining them within the elongata species-
complex. The senior synonym allocated to each group is given in Table i.

TAXONOMY

Metapheretima elongata species-complex

Metapheretima elongata species-complex : Sims & Easton, 1972 : 252.
Metapheretima phacellotheca : Sims & Easton, 1972 : 233.

METAPHERETIMA ELONGATA SPECIES-COMPLEX 39

DIAGNOSIS. Metapheretima with male pores within copulatory pouches lacking
stalked glands. Spermathecae in paired, ventrolateral batteries opening in inter-
segmental furrows 5/6 and/or 6/7. Genital markings simple, large, paired, presetal,
on xix and successive segments in line with or slightly ventral to the male pores.

DESCRIPTION. External characters. Length 40-360 mm, diameter 1-5-10 mm.
100-220 segments. Clitellum xiv-xvi. First dorsal pore 12/13. Setae regularly
distributed around each segment, 20-130 on vii, 36-86 on xx ; setal distance
aa = i-2ab ; occasionally setae a and b enlarged.

Male pores paired on short penes within shallow copulatory pouches on xviii,
c. 0-25 body circumference apart. Female pore single ; midventral xiv. Sperma-
thecal pores small, numerous, arranged in paired, ventrolateral batteries each com-
prising up to 28 pores, intersegmental in 5/6 and/or 6/7.

Genital markings (Fig. 5a) : paired, simple, large, oval presetal papillae on xix
and successive segments in line with or slightly ventral to the male pores ;
occasionally present on vi, vii and xvii.

Internal characters. Septa 4/5-7/8 thickened, 8/9 membranous, 9/10 absent,
10/11-13/14 thickened. Intestine begins in xv. Lateral hearts in x, xi, xii, usually
xiii.

Holandric ; testes sacs large, paired, extending to the dorsal line in x and xi ;
seminal vesicles paired in xi and xii, the anterior pair enclosed in the posterior testes
sacs. Coelomic sacs in xiii, xiv. Spermathecae (Fig. 5b) numerous, in paired
batteries of up to 28 in vi and/or vii. Copulatory pouches shallow, confined to the
body wall of xviii.

DISTRIBUTION. Indigenous records : Kepulaud Taulaud ; Sangihe ; Celebes ;
Buru ; Balabac Island, Palawan ; Borneo ; ? Madura ; east Java ; Bali ; Lom-
bok ; Sumbawa ; Komodo ; ? Gt Bastard Island, Flores.

Four of the five species included in this species-complex are known only from the
indigenous range. The fifth, elongata, has been introduced into many parts of the
world.

REMARKS. This species-complex is readily recognized by the positions of the
spermathecal pores and the form of the male pores. Three other species of Meta-
pheretima with similar features may be distinguished either by the absence of genital
markings in the case of M. annamanensis , a bithecal species known only from the
Lang Biang Peaks, Vietnam, or by the form of the genital markings in the cases of
aringeana from Kelantan, Malaya, and grata from New Guinea. In aringeana the
postclitellar genital markings are paired and occur only on xvii and xix as circular
pads with many small pores while in grata they are small and numerous, being
present only in the region of the spermathecal pores. This latter species also has
stalked glands discharging into the copulatory pouches which are large and invade
the coelom. Clitellate individuals from the indigenous range of the species-complex
are rarely athecate but the majority of specimens examined from introduced popu-
lations lack Spermathecae. In these populations athecate individuals may be
identified by the arrangement of the genital markings and the form of the copulatory
pouch. Outside of the Pheretima domain, identification is easier since only one

4o

E. G. EASTON

elongata

everetti & phacellotheca

kinabaluensis & stellen

Spermathecal number

FIG. 4. Metapheretima elongata species-complex. Relationship between the setal number
of segment vii and the spermathecal number, A/(«2-f62), for the member species.

other species of Metapheretima is commonly encountered. This species, M. tapro-
banae, has paired spermathecal pores in furrow 7/8 and genital markings on vi-ix
and xviii-xxii. Additionally it is considerably smaller than the members of the
elongata species-complex, rarely exceeding 140 mm in length.

The species of the elongata species-complex may be distinguished from one another
by comparison of the setal number on vii and the size of the spermathecal batteries
(Fig. 4). (The size of the spermathecal batteries is calculated by substituting the
mean size of anterior and posterior spermathecal batteries of an individual for a
and b in the formula x — -\/(az + b2).)

Two pairs of species, everetti-phacellotheca and stelleri-kinabaluensis, need ad-
ditional characters to be assessed for differentiation. In the first pair, M. everetti
has spermathecae in furrows 5/6 and 6/7 while phacellotheca has spermathecae in
furrow 5/6 only. In the second, M. stelleri has very numerous setae on vii (up to
130) while kinabaluensis has few setae (less than 40).

Metapheretima elongata (Perrier, 1872)

Perichaeta elongata Perrier, 1872 : 124 ; Beddard, 1895 : 431.
Megascolex elongata : Vaillant, 1889 : 81.
Amynthas elongatus : Beddard, 1900 : 650.

Pheretima elongata : Michaelsen, 1900 : 265 ; Cognetti, igosa : 33 ; Michaelsen, igioa : 252 ;
( = biserialis) Michaelsen, igioc : 84 ; Michaelsen, 1913 : 262 ; Michaelsen, 1920 : 68 ;

METAPHERETIMA ELONGATA SPECIES-COMPLEX 41

Stephenson, 1920 : 222 ; Michaelsen, 1922 : 32 ; Stephenson, 1922 : 433 ; Stephenson,
1923 : 298 ; Stephenson, 1924 : 339 ; Gates, ig26a : 208 ; Gates, ig26b : 444 ; Gates,
I926c : 183 ; Gates, ig26d : 153 ; Stephenson, 1926 : 256 ; Stephenson, 1929 : 237 ; Gates,
1930 : 309 ; Gates, 1931 : 378 ; Stephenson, 1931 : 58 ; Gates, 1932 : 391 ; Gates, 1933 :
525 ; Cernosvitov, 1934 : 49 ; Cernosvitov, 1935 : I][ : Gates, 1936 : 413 ; Gates, I937a : 318 ;
Gates, I937b : 352 ; Gates, I937C : 201 ; Gates, 1939 : 87 ; Daweini, 1940 : no ; Gates, 1954 :
86 ; Gates, 1959 : 9 ; Gates, ig6oa : 57 ; Gates, i96ob : 256 ; Gates, i96ia : 62 ; Gates,
ig6ib : 308 ; Gates, 1972 : 182 ; [non Ohfuchi, 1956 : 148 ( = Amynthas morrisi}].

Pheretima (Pheretima} elongata : ? Ude, 1932 : 155 ; Michaelsen, i934a : 108.

Perichaeta biserialis Perrier, 1875 : 1044 ; Beddard, 1890 : 63 ; Beddard, 1895 : 430 ; Michael-
sen, 1897 : 226 ; (= acystis} Horst, 1899 : 202 ; Beddard & Fedarb, 1899 : 803.

Megascolex (Perrier a) biserialis : Vaillant, 1889 : 76.

Amynthas biserialis : Michaelsen, 1899 : 21 ; Beddard, 1900 : 638.

Pheretima biserialis : Michaelsen, 1900 : 256 ; Michaelsen, 1902 : 9 ; Michaelsen, 1904 : 285 ;
Cognetti, igo5a : 30 ; Cognetti, igo5b : 2 ; Ude, 1905 : 471 ; Michaelsen, 1907 : 44 ; Michael-
sen, igoSa : 187 ; Michaelsen, igoSb : 14 ; Ohfuchi, 1940 : 13 ; Ohfuchi, 1956 : 151.

Perichaeta acystis [nom. nov. pro biserialis : Beddard, 1890 (non Perrier, 1872)] Beddard, 1895 :

Perichaeta monocystis (lapsus pro acystis} Horst, 1899 : 202.

Metapheretima elongata species-complex : Sims & Easton, 1972 : 256, footnote i.

Metapheretima elongata : Jamieson (in press).

TYPE LOCALITY. Peru.

DIAGNOSIS. M. elongata species complex with numerous setae (usually about 80
on mi but up to 130 on large individuals) and small spermathecal batteries (rarely
more than 3 spermathecae in a battery) in furrows 5/6/7, or 5/6 or 6/7 only or
absent. Spermathecal batteries usually present in about 50% of the clitellate
portion of indigenous populations to about only 15% of the clitellate portion of
introduced populations.

DISTRIBUTION. Indigenous records (Fig. 6). ? Madura ; east Java ; Bali ;
Lombok ; Sumbawa ; Komodo ; ? Gt Bastard Island, Flores ; ? south-east
Celebes. The establishment of the indigenous range of this species is discussed
below (see Distribution). The specimens identified from Bali, Lombok, Sumbawa
and Komodo represent new indigenous records for the species.

Introduced records. This species has been introduced into many parts of the
world ; for a summary of these records see Gates (1972). During the preparation
of this paper specimens from Queensland, Australia ; Tahiti ; Antigua, West
Indies ; and Oman were identified. These represent new introduced records for
the species.

Metapheretima everetti (Beddard & Fedarb, 1895)

Perichaeta everetti Beddard & Fedarb, 1895 : 69 ; Beddard, 1895 : 428 (non Amynthas stelleri

everetti : Michaelsen, 1899 : 43).

Perichaeta papillata Beddard & Fedarb, 1895 : 71 > Beddard, 1895 : 428.
Perichaeta sarawacensis Beddard & Fedarb, 1895 : 71 ; Beddard, 1895 : 429.
Perichaeta bar ami Michaelsen, 1896 : 203.
Amynthas stelleri bar ami : Michaelsen, 1899 '.41.

Perichaeta stelleri : Michaelsen, 1896 : 202 ; (non Michaelsen, 1891 : 39) ; Horst, 1899 : 205.
Amynthas stelleri (typica) : Michaelsen, 1899 : 83.

E. G. EASTON

146.

Pheretima stelleri (typica} : Michaelsen, 1922 : 25.

? Pheretima (Pheretima) stelleri (typica): Ude, 1932

Amynthas stelleri seriatus : Michaelsen, 1899 : 44.

Amynthas stelleri klabatensis Michaelsen, 1899 : 46.

Amynthas stelleri bonensis MUhaelsen, 1899 : 45.

Pheretima stelleri bonensis : Michaelsen, 1900 : 307.

[non Pheretima (Pheretima) stelleri bonensis : Michaelsen, i934a : 108 ( = M. phacellotheca)]

Pheretima stelleri koroensis Michaelsen, igiob : 109.

Pheretima stelleri mahakkami Michaelsen, 1922 : 25.

Pheretima (Pheretima) beranensis Michaelsen, 1928

Pheretima (Pheretima) baritoensis Michaelsen, 1932

? Pheretima (Pheretima) elongata : Ude, 1932 : 155 ;

Pheretima (Polypheretima) beranensis tinjarana Michaelsen, 19345 : 25.

Metapheretima elongata species-complex [series B (part)] : Sims & Easton, 1972 : 253.

TYPE LOCALITY. Balabac Island, Palawan. Beddard & Fedarb reported that
the type locality of Perichaeta everetti is Mt Kinabalu, Sabah. The type series of
this species in the collections of the British Museum (Natural History) is labelled as
originating from Balabac Island, Palawan. Although Beddard & Fedarb wrote in
their introduction that the collection they were reporting was collected from both
Borneo and Palawan, no reference to Palawan is made in the general text. It is

23.

9.

(non Perichaeta elongata Perrier, 1872).

FIG. 5. Metapheretima everetti. (a) Diagram of the papillae pattern on the ventral
surface of the anterior region of the body, (b) Spermatheca.

METAPHERETIMA ELONGATA SPECIES-COMPLEX 43

proposed to follow Recommendation 72E of the International Code of Zoological
Nomenclature and redesignate Balabac Island, Palawan as the type locality for
Perichaeta everetti Beddard & Fedarb, 1895.

DIAGNOSIS. M. elongata species-complex with numerous setae (up to 130 on mi
in large individuals) and intermediate-sized spermathecal batteries (usually 6-12
spermathecae in each battery of large individuals) in furrows 5/6/7. Spermathecal
batteries present in aclitellate as well as clitellate individuals.

DISTRIBUTION. Indigenous records (Fig. 6) : north and west Celebes ; Balabac
Island, Palawan ; Borneo (up to altitudes of 2400 m) ; Lombok (c. 350-450 m
only).

This species is unknown outside its indigenous range.

REMARKS. On Lombok this species has been recorded only from localities at
altitudes between 350 and 450 m. (Samples from lower altitudes contained the
species elongata which is rare at the higher altitudes. No samples of earthworms
are known from above 450 m on Lombok.) Although altitude would appear to be
the primary factor governing the distribution of both everetti and elongata, it should
be noted that, on Lombok, the samples of everetti were from natural woodland
while those of elongata were from cultivated land. All the records of everetti, stelleri
and phacellotheca from Celebes are from the northern and western region, an area of
forest, while the single record of elongata (as yet unconfirmed as an indigenous
population) is from the southeast which is an area of savannah. On Mt Kinabalu,
Borneo, the species everetti occurs at altitudes up to 2400 m, being sympatric from
2 100 m with the high altitude species kinabaluensis.

Although on Lombok and Mt Kinabalu everetti is sympatric with other members
of the elongata species-complex, there is no indication of hybridization among them.

Metapheretima kinabaluensis (Beddard & Fedarb, 1895)

Perichaeta kinabaluensis Beddard & Fedarb, 1895 : 71 ; Beddard, 1895 : 429-

Metapheretima elongata species-complex [series A & series B (part)]: Sims & Easton, 1972 : 253.

TYPE LOCALITY. Tamburgare, Mt Kinabalu, Sabah, 2350 m (7700 ft).

DIAGNOSIS. M . elongata species-complex with few setae (less than 40 on vii in
large individuals) and intermediate-sized spermathecal batteries (usually 6-12
spermathecae in each battery of large individuals) in furrows 5/6/7. Spermathecal
batteries present in aclitellate as well as clitellate individuals.

DISTRIBUTION. Indigenous records (Fig. 6) : Mt Kinabalu, Sabah (2100-2750 m).
This species has not been recorded outside its indigenous range.

REMARKS. This species is known only from high altitudes on Mt Kinabalu where
the fauna and flora are of a distinct montane type. It is probable that future
investigations of the high-altitude earthworm fauna in the Borneo-Celebes area will
reveal other population of this species. Although samples collected at 2100 and
2400 m on Mt Kinabalu contained both everetti and kinabaluensis there is no indi-
cation of hybridization between these two forms.

44 E. G. EASTON

Metapheretima phacellotheca (Michaelsen, 1899)

Amynthas phakellotheca (sic) Michaelsen, 1899 : 47.
Amynthas phacellotheca : Beddard, 1900 : 640.

Pheretima phacellotheca : Michaelsen, 1900 : 293 ; Gates, 1961 : 304.
Pheretima (Polypheretima) phacellotheca : Michaelsen, 19345 : 15.
Metapheretima phacellotheca : Sims & Easton, 1972 : 233.

Pheretima (Pheretima} stelleri bonensis : Michaelsen, 19343 : 108 ; (non Amynthas stelleri bonensis
Michaelsen, 1899 : 45).

TYPE LOCALITY. Mt Masarang, above Tomohon, northeast Celebes.

DIAGNOSIS. M. elongata species-complex with numerous setae (up to 80 on vii
in large individuals) and large spermathecal batteries (9-12 spermathecae in each
battery in large individuals) in furrow 5/6. Spermathecal batteries present in
aclitellate as well as clitellate specimens.

DISTRIBUTION. Indigenous records (Fig. 6) : northeast Celebes, Buru.
This species has not been recorded outside its indigenous range.

REMARKS. This species is the only member of the elongata species-complex
known from Buru which is the most easterly indigenous record of the complex.
Morphologically it most closely resembles stelleri from which it may be distinguished
only by the restriction of spermathecal batteries to furrow 5/6. Its relationships
with everetti and stelleri, which also occur in north Celebes, are unknown.

Metapheretima stelleri (Michaelsen, 1891)

Perichaeta stelleri Michaelsen, 1891 : 39 ; (non Michaelsen, 1896 : 202 ; Horst, 1899 : 205).
[non Amynthas stelleri (typica) : Michaelsen, 1899 : 83.]
[non Pheretima stelleri (typica) : Michaelsen, 1922 : 25.]

Pheretima (Pheretima) stelleri (typica) : Michaelsen, i934a : 108 (non Ude, 1932 : 146).
Amynthas stelleri annectens : Michaelsen, 1899 : 42.

Amynthas stelleri everetti : Michaelsen, 1899 : 43 ; (non Perichaeta everetti Beddard & Fedarb,
1895 : 69).

TYPE LOCALITY. Sangihe.

DIAGNOSIS. M. elongata species-complex with very numerous setae (up to 130
on vii in large individuals) and large spermathecal batteries (up to 28 spermathecae
in each battery of large individuals) in furrows 5/6/7. Spermathecal batteries
present in aclitellate as well as clitellate individuals.

DISTRIBUTION. Indigenous records (Fig. 6) : Kepulaud Taulaud ; Sangihe ;
north Celebes.

This species has not been recorded outside its indigenous range.

REMARKS. In Celebes stelleri has been recorded from Bone Valley (c. 400 m) and
the Matinang range (c. 1500 m). In both of these areas everetti has also been re-
corded but the relationship between the two species and the factors governing their
distribution are unknown. No information is available regarding the relationship
between stelleri and phacellotheca.

METAPHERETIMA ELONGATA SPECIES-COMPLEX 45

DISTRIBUTION

All the records attributed to everetti, stelleri, kinabaluensis and phacellotheca
(groups 2, 3, 4 and 5) are from Borneo, Celebes or nearby islands and it is con-
sidered that these records indicate the indigenous ranges of the species. The species
elongata (group i) has been recorded from islands as far apart as Luzon, New Guinea
and Sumatra in the Indo-Australasian Archipelago (as well as many other localities
throughout the world) but it is thought that this distribution is not entirely natural
and that it has been augmented through accidental introduction by man. In the
area adjacent to the range of the other members of the complex elongata has been
recorded from Sumatra, Java, Madura, Bali, Lombok, Sumbawa, Komodo, Gt
Bastard Island, Celebes, Kepulaud Taulaud and the Philippines and it is suspected
that at least some of these records are from indigenous populations. Among these
localities Lombok and Celebes are of particular interest since the most closely related
member of the complex, everetti, has also been recorded from these islands.

Although the majority of species of the Pheretima group possess normal her-
maphrodite reproductive systems and are assumed to reproduce only sexually,
several species are known to occur as morphs lacking portions of the reproductive
system. Gates (1972) recognized four principal morphs : H morph, the normal
hermaphrodite form ; A morph, lacking spermathecae ; R morph, lacking male
terminalia ; and Z morph, lacking testes. (In addition to these basic forms he also
recognized AR and ARZ morphs.) He considered that, with the exception of the
H morph, all these forms reproduced by parthenogenesis. He recorded 70 species
of the Pheretima group from Burma of which 42 were known only from Burma and
the adjacent area while 19 were peregrine ; only 6% of the indigenous species but
47% of the peregrine forms are known to possess non-hermaphrodite morphs. This
correlation between non-hermaphrodite morphs and peregrine distributions led
Gates (1956) to suggest that the peregrine species Metaphire anomala (Michaelsen,
1907) had originated from eastern Burma where the H morph was more common.

TABLE 3

Population characteristics of M. elongata from selected islands

Number of clitellate Proportion of

Island specimens examined H morphs (%)

Jamaica 84 18

Sumatra 47 15

west & central Java 13 15

east Java 22 50

Madura 3 (2) *

Bali 27 44

Lombok 19 26

Sumbawa 59 "\

Komodo 4 J

Kepulaud Taulaud i (o)*

Philippines 9 (i) *

* For small samples (less than 10 individuals) the number of thecate individuals (rather than the
percentage) has been recorded and placed in parentheses.

46 E. G. EASTON

Metapheretima elongata similarly occurs as H and A morphs so if the hypothesis
is valid then the H morph will be more common in its indigenous range than else-
where. Samples of island populations were examined and the incidence of H morphs
was calculated from among the clitellate portion of each series. The results of this
study are shown in Table 3. (Java was treated as two regions since it may be con-
sidered to be composed of two ecologically distinct areas, tropical rain forest in
central and west Java and savannah in east Java.) To obtain a comparative value
for a population which was known to be introduced, the incidence of H morphs was
assessed in a population from Jamaica. The low incidence of H morphs in the
introduced Jamaican population (18%) agrees very well with that of the popu-
lations from central and west Java and Sumatra (both 15%) but contrasts strongly
with populations from east Java, Bali, Sumbawa and Komodo (44-51%). The
value calculated for Lombok (26%) is intermediate between the two classes and
somewhat anomalous in view of the geographical position of the island. Possibly
the small size of the sample may be the causative factor. Assuming the basic
assumption to be valid, these calculations indicate that elongata is indigenous in east
Java, Bali, Lombok, Sumbawa and Komodo, while the form which is morpho-
logically the most similar, everetti, occurs at high altitudes (c. 400 m) in Lombok.
These values are consistent in the degree of magnitude with those recorded by
Gates (1956) when he recorded 20% H morphs of anomala in eastern Burma but only
0-7% in other parts of Burma.

Samples from Kepulaud Taulaud, Philippines and Madura were too small to
derive any reliable conclusions from them but it seems unlikely that elongata is
indigenous in either Kepulaud Taulaud or the Philippines, although it may possibly
occur naturally in Madura. There have been single inconclusive records of elongata
from south-east Celebes and Gt Bastard Island but in the former case the specimens
cannot be located and in the latter the single extant specimen proved to be indeter-
minate and the occurrence of indigenous populations of elongata in both of these
islands has yet to be confirmed. The known indigenous distributions of the mem-
bers of the elongata species-complex are shown in Fig. 6. The earthworm fauna of
south-east Asia has been extensively studied (Gates, 1972) and data from this work
and other recent collections in the British Museum (Natural History) have allowed
the establishment of the western limit of the indigenous range of the complex.
Unfortunately the earthworm faunae of the Philippines, Celebes, Moluccas and
Lesser Sunda Islands still require considerable study and it is not practical to estab-
lish the northern and eastern limits of the distribution of the complex. However,
recent extensive collecting in New Guinea (Easton, in preparation) indicates that
the complex is absent from this island except for introduced populations of elongata.

The form kinabaluensis is known only from Mt Kinabalu, Sabah, at altitudes of
over 2100 m. It is probable that future collections will reveal that it also occurs
on some of the other high mountains in Borneo and possibly even Celebes. Likewise
the occurrence of everetti in Lombok at altitudes of 400 m may indicate that it occurs
at high altitudes throughout the Lesser Sunda Islands.

Metapheretima stelleri has been recorded from Sangihe, Kepulaud Taulaud and
north Celebes. Michaelsen (1934^ was of the opinion that this species had been

METAPHERETIMA ELONGATA SPECIES-COMPLEX

47

FIG. 6. Metapheretima elongata species-complex. Endemic distributions : • elongata ;
<D everetti ; (§) hinabaluensis ; @ phacellotheca ; 0 stelleri ; ? limits of range uncertain ;
o isolated or minor localities.

introduced into both Sangihe and Kepulaud Taulaud, but this view is inconsistent
with the species' restricted distribution in Celebes. (At the time of Michaelsen's
paper the species was thought to be widespread in both Borneo and Celebes.) The
wide distribution of the species-complex, crossing both the Flores Sea and the
Makassar Strait, suggests that the complex is of considerable antiquity and M.
stelleri may have reached both Sangihe and Kepulaud Taulaud at a time when land
bridges existed between these islands and Celebes.

In Celebes the ranges of each of the three forms occurring there as well as the
natural occurrence of elongata have still to be established.

48 E. G. EASTON

ACKNOWLEDGEMENTS

This study has only been made possible by the examination of specimens from the
collections of several museums and I must record my gratitude for the loan of
material to : Dr M. Dzwillo, Zoologisches Staatsinstitut und Zoologisches Museum,
Hamburg ; Dr G. Hartwich, Institut fur Spezielle Zoologie und Zoologisches
Museum, Berlin ; Dr S. Kadarsan, Museum Zoologicum Bogoriense, Bogor ; Dr
J. van der Land, Rijksmuseum van Natuurlijke Historic, Leiden ; Dr J. Renaud-
Morant, Laboratoire de Zoologie (Vers), Museum National d'Histoire Naturelle,
Paris ; Dr J. van der Spoel, Zoologisch Museum, Universiteit van Amsterdam. I
am also grateful to Mr R. W. Sims for allowing me access to his field notes, made
while collecting in Jamaica, and for reading the manuscript. Finally, I would like
to thank Miss L. Bellenger and Miss D. Feldman for their assistance with sorting
material and assembling data.

REFERENCES

BEDDARD, F. E. 1890. Observations upon an American species of Perichaeta and upon some
other members of the genus. Proc. zool. Soc. Lond. 1890 : 52-69.

- 1895. A Monograph of the Order Oligochaeta, pp. i-xii, 1-769. Oxford.

- 1900. A revision of the earthworms of the genus Amyntas (Perichaeta). Proc. zool. Soc.
Lond. 1900 : 609-652.

- & FEDARB, S. M. 1895. On some Perichaetidae from the Eastern Archipelago collected
by Mr. Everett. Ann. Mag. nat. Hist. (ser. 6), 16 : 69-73.

-- 1899. Notes on two earthworms, Perichaeta biserialis and Trichochaeta hesperidium.

Proc. zool. Soc. Lond. 1899 : 803-809.
CERNOSVITOV, L. 1934. L68 Oligochetes de la Guyane francaise et d'autres pays de 1'Amerique

du Sud. Bull. Mus. natn. Hist. nat. Paris 2 (6) : 45-59.

- 1935- Oligochaeten aus dem tropischen Siid-Amerika. Capita zool. 6 (i) : 5-51.
CHEN, Y. 1931. On the terrestrial oligochaeta from Szechuan. Contr. biol. Lab. Sci. Soc.

China (zool.) 7 : 117-171.

- 1936. On the terrestrial oligochaeta from Szechuan II with the notes on Gates' types.
Contr. biol. Lab. Sci. Soc. China (zool.) 11 : 269-306.

- 1946. On the terrestrial oligochaeta from Szechwan III. Jl W. China Border Res. Soc.
16 : 83-141.

COGNETTI DE MARTiis, L. igosa. Gli Oligochaeti della regione neotropicale. Memorie Accad.
Sci. Torino 55 : 1-72.

Oligocheti raccolti nel Darien dal Dr. E. Festa. Boll. Musei Zool. Anat. comp.

R. Univ. Torino 20 (495) : 1 — 7.
1911. Nuove specie dei genera Pheretima e Dichogaster (Diagn. prel.). Boll. Musei Zool.

Anat. comp. R. Univ. Torino 26 (641) : 1-7.
DAWEINI, K. EL. 1940. The earthworms of Egypt. Bull. Inst. Egypte 22 : 99-122.
GATES, G. E. ig26a. The earthworms of Rangoon. /. Burma Res. Soc. 15 : 197-221.

— i926b. Notes on the Rangoon earthworms. The peregrine species. Ann. Mag. nat.
Hist. (ser. 9) 17 : 439- 473.

I926c. Notes on the seasonal occurrence of Rangoon earthworms. /. Bombay nat. Hist.

Soc. 38 : 180-185.

— I926d. Notes on the earthworms from various places in the province of Burma, with
descriptions of two new species. Rec. Indian Mus. 28 : 141-170.

1930. The earthworms of Burma I. Rec. Indian Mus. 32 : 257-356.

1931. The earthworms of Burma II. Rec. Indian Mus. 33 : 327-442.

1932. The earthworms of Burma III. The Megascolecidae. Rec. Indian Mus. 34 :

357-549-

METAPHERETIMA ELONGATA SPECIES-COMPLEX 49

GATES, G. E. 1933. The earthworms of Burma IV. Rec. Indian Mus. 35 : 413-606.

1936. The earthworms of Burma V. Rec. Indian Mus. 38 : 377-468.

I937a. Notes on some species of Drawida and Pheretima with descriptions of three new

species of Pheretima. Bull. Mus. Comp. Zool. Haw. 80 : 305-334.
19370. The genus Pheretima in North America. Bull. Mus. comp. Zool. Ham. 80 :

339-373-

19370. Indian earthworms I. The genus Pheretima. Rec. Indian Mus. 39 : 175-212.

1939. Thai earthworms. /. Thailand Res. Soc. 12 : 65-114.

1954- The earthworms of Burma VI. Rec. Indian Mus. 52 : 55-93.

1956. Reproductive organ polymorphism in earthworms of the oriental Megascolecine

genus Pheretima Kinberg, 1867. Evolution, Lancaster, Pa. 10 : 213 — 227.

I959- On some earthworms from Taiwan. Am. Mus. Novit. 1941 : 1-19.

ig6oa. Earthworms of Burma. Burma Res. Soc. ^oth Ann. Pub. 1 : 51-58.

19600. On Burmese earthworms of the family Megascolecidae. Bull. Mus. comp. Zool.

Haw. 123 : 203-282.
I96ia. Ecology of some earthworms with special reference to seasonal activity. Am.

Midi. Nat. 66: 61-86.
ig6ib. On some species of the Oriental earthworm genus Pheretima Kinberg, 1867.

Zool. Meded. Leiden 37 : 293-312.
1972. Burmese earthworms, an introduction to the systematics and biology of Megadrile

Oligochaetes with special reference to southeast Asia. Trans. Am. phil. Soc. 62 (7) : 1-326.
HORST, R. 1899. On the variability of characters in Perichaetidae. Notes Leyden Mus. 20 :

201-209.
JAMIESON, B. G. M. (in press). Catalogue of the named Megascolecidae (Oligochaeta : Annelida)

in the National Museum of Natural History, Paris.
LEE, K. E. 1967. Pheretima tumulifaciens (Oligochaeta : Megascolecidae). A new species of

earthworm from the Sepik district, New Guinea. Trans. R. Soc. S. Aust. 91 : 59-63.
MICHAELSEN, W. 1891. Oligochaeten des Naturhistorischen Museum in Hamburg, IV.

Jb. hamb. wiss. Anst. 8 (2) : 3-42.
1896. Oligochaten Kiikenthal-Ergebnisse einer zool. Forschungsreise in den Molukken

und in Borneo. Abh. senckenb. naturforsch. Ges. 23 : 192—243.
1897. Die Terricolen des Madagassischen Inselgebiets. Abh. senckenb. naturforsch. Ges.

21 : 219-252.
1899. Terricolen von verschiedenen Gebieten der Erde. Jb. hamb. wiss. Anst. 16 (2) :

3-122.

1900. Oligochaeta. Das Tierreich 10: 1-575. Berlin.

1902. Neue Oligoschaten und neue Fundort alt-bekannter. Jb. hamb. wiss. Anst. 19 (2) :

3-54-
1904. Catalogo de los Oligoquetos del territorio Chileno-Magallanica i descripcidn de

especias neuvas. Revta chil. Hist, nat., 8 : 262-292.
1907. Oligochaten von Madagaska, den Comoren und anderen Inseln des westlichen

Indischen Ozeans. Wiss. Ergebn. Reise Ostafr. 2 : 41-50.
igoSa. The Oligochaeta of India, Nepal, Ceylon, Burma and the Andaman Islands.

Mem. Indian Mus. 1 : 103-253.

igoSb. Die Oligochaten Westindiens. Zool. Jb. (supp.) 11 : 13-32.

igioa. Oligochaten von den Aru und Kei Inseln. Abh. senckenb. naturforsch. Ges. 33 :

251-261.

igiob. Oligochaten von verschiedenen Gebieten. Jb. hamb. wiss. Anst. 27 : 47-169.

igioc. Die Oligochatenfauna der vorderindisch-Ceylonischen Region. Abh. Geb. Naturw.

Hamburg, 19 (5) : 1-108.
1913. Die Oligochaten von Neu-Caledonien und den benachbarten Inselgruppen. Nova

Caledonia 1 (3) : 173-280.

1920. Oligochaten von westlichen Vorderindien und ihre Beziehungen zur Oligochaten-
fauna von Madagaskar und den Seychellen. Jb. hamb. wiss. Anst. 38 : 27-68.

5o E. G. EASTON

MICHAELSEN, W. 1922. Oligochaten aus dem Rijks-Museum van Natuurlijke Historic zu
Leiden. Capita zool. 1 (3) : 1-67.

— 1928. Die Oligochaten Borneos. Ark. Zool. 20 (3) : 1-60.

— 1932. Neue Oligochaten von Bali und Borneo. Bull. Mus. r. Hist. nat. Belg. 8 (32) : i -i i.
— i934a. Oligochaten von Niederlandisch-Indien. Arch, neerl. Zool. 1 : 100-117.

— 19340. Oligochaeta from Sarawak. Quart. J. microsc. Sci. 77 : 1-47.

OHFUCHI, S. 1940. On the Oligochaeta from the East and West Carolines, South Sea Islands,
I. Palao trop. biol. Stn. Stud. 2 (i) : 7-28.

— 1956. On a collection of terrestrial Oligochaeta obtained from the various localities in
the Riu-Kiu Islands, together with the considerations of their geographical distribution.
(Part i.) /. agric. Sci. Tokyo 3: 131-176.

PERKIER, E. 1872. Rescherches pour servir a 1'histoire des Lombriciens terrestres. Nouv.
Archs Mus. Hist. nat. Paris 8 : 5-198.

— 1875. Sur les vers de terre des iles Philippines et de la Cochinchine. C. r. hebd. Seanc.
Acad. Sci. Paris (D), 81 : 1043-1046.

SIMS, R. W. & EASTON, E. G. 1972. A numerical revision of the earthworm genus Pheretima
auct. (Megascolecidae : Oligochaeta) with the recognition of new genera and an appendix
on the earthworms collected by the Royal Society North Borneo Expedition. Biol. J. Linn.
Soc. 4 (3) : 169-268.

STEPHENSON, J. 1920. On a collection of Oligochaeta from the lesser known parts of India and
from eastern Persia. Mem. Indian Mus. 7 : 191-261.

1922. Some earthworms from Kashmir, Bombay and other parts of India. Rec. Indian

Mus. 24 : 427-443.

1923. Oligochaeta. Fauna Br. India 1-518.

— 1924. On some Indian Oligochaeta, with a description of two new genera of Ocnerodrilinae.
Rec. Indian Mus. 26 : 317-375.

— 1926. Descriptions of Indian Oligochaeta. Rec. Indian Mus. 28 : 149-168.

— 1929. The Oligochaeta of the Indawagyi lake (Upper Burma). Rec. Indian Mus. 31 :
225-239.

— 1930. The Oligochaeta, pp. 1-978. Oxford.

— 1931. Oligochaeta from Burma, Kenya and other parts of the world. Proc. zool. Soc.
Land. 1931 : 33-92.

UDE, H. 1905. Terricole Oligochaten von den Inseln der Siidsee und von verschiedenen andern

Gebieten der Erde. Z. wiss. Zool. 83 : 405-501.
1932. Beitrage zur Kenntnis der Gattung Pheretima und ihrergeographischen Verbreitung.

Arch. Naturgesch. 1 : 114-190.
VAILLANT, M. L. 1889. Lombriciniens, Hirudiens, Bdellomorphs, Teretulariens et Planariens.

In : Histoire naturelle des Anneles marins et d'Eau douce 3 : 1-766. Paris.

APPENDIX
MATERIAL EXAMINED AND OTHER RECORDS FROM THE INDO -AUSTRALASIAN ARCHIPELAGO

Museums :

Amsterdam Zoologisch Museum, Universiteit van Amsterdam

Berlin Institut fur Spezielle Zoologie und Zoologisches Museum, Berlin

BM(NH) British Museum (Natural History), London

Bogor Museum Zoologicum Bogoriense, Bogor

Hamburg Zoologisches Staatsinstitut und Zoologisches Museum, Hamburg

Leiden Rijksmuseum van Natuurlijke Historic, Leiden

Paris Museum National d'Histoire Naturelle, Paris

The suffixes C and A refer to the numbers of clitellate and aclitellate individuals in each series
respectively.

METAPHERETIMA ELONGATA SPECIES-COMPLEX 51

Metapheretima elongata (Perrier, 1872)

1. MATERIAL COLLECTED BY DR G. A. LINCOLN DURING THE BRITISH UNIVERSITY DRAGON
EXPEDITION, May -September 1973. BM(NH) 1975.7.59-302.

Sumatra : Simabur, iC ; Sungalamas, 46 5A ; Kalianda, 6C.

Java : village 16 km north of Balaradja, iC ; plantation n km west of Tomo, nC 5 A ; Krijan,
3C ; cultivated area 6 km east of Bangil, loC 2 A ; Pasir Putih, ?C 4 A ; Baluran, iC ; Sukamade,

2C.

Bali : Udjung, 156 lA ; foothills 8 km north of Klungkung, I2C I2A.
Lombok : Surawadi, iC ; Songian, 26C nA

Sumbawa : Alas, 156 ; Kanar, 146 ; Lape, jC ; Lampang, gC ; Napa, gC lA ; cultivated area
5 km east of Dompu, loC nA ; forest 8 km west of Bolo, 76 lA.

Komodo ; Komodo village, 46.

2. OTHER MATERIAL EXAMINED.

Philippines : Bataan, Luzon, 46, Paris AE 634, 638, 643 (elongata : Jamieson, in press) ;
Mindoro, 6C, Paris AE 639-642, 644* (elongata : Jamieson, in press) ; Manila, Luzon, 2C lA,
BM(NH) 1904.10.5.1-2 (biserialis : Beddard, 1890 ; syntypes of acystis).

Kepulaud Taulaud : Salibaboe Liroeng, Salibaboe Island, iC i A, Amsterdam V.ol 254 (elongata :
Michaelsen, i934a).

Sumatra : Bindjey Estate, 3C lA, Hamburg v.352 (biserialis : Michaelsen, 1899) ; north-east
Sumatra, 2C lA, Hamburg v.5o6o ; Poeloe-Weh, 330 n A, Leiden 1879, 1882, 1887, 1888
(elongata : Michaelsen, 1922) ; Nias Island, lA, Leiden.

Madura : 36 lA, Leiden 1880, 1883 (elongata : Michaelsen, 1922).

Java : Cimaja, Pelabuhanratu, 156 lA, Bogor Ann. 090 & BM(NH) 1975.5.1-4.

Flores : Groot Bastaard Island, i macerated specimen, Leiden 1886 (elongata : Michaelsen, 1922).

Jamaica: 846, 99 A ; BM(NH).

3. OTHER RECORDS FROM THE INDO- AUSTRALASIAN ARCHIPELAGO.
Kei Islands : Elat, Gross Kei (elongata : Michaelsen, 1910).

Java : Malang (elongata : Michaelsen, 1922).

Celebes : Boeton (elongata : Ude, 1932).

New Guinea : Kila Kila, Madang district (elongata : Sims & Easton, 1972).

Metapheretima everetti (Beddard & Fedarb, 1895)

1. MATERIAL COLLECTED BY DR G. A. LINCOLN DURING THE BRITISH UNIVERSITY DRAGON
EXPEDITION, August 1973. BM(NH) 1975.57-58, 303-313.

Lombok : Bentak, nC 2A.

2. OTHER MATERIAL EXAMINED.

Celebes : Minahassa, iC, Hamburg v.3836 (barami : Michaelsen, 1896) ; — ,t 2C lA, Hamburg
v-5197, 5198 (types of stelleri seriatus) ; Klabat, iC, Hamburg v.5ig6 (holotype of stelleri
klabatensis) .

* A specimen from this series was examined by Sims & Easton (1972 : 254) and referred to as a syn-
type of Perichaeta biserialis Perrier, 1875. Further information has now been obtained and it is now
certain that the specimen is not typical since it was not collected until 1876.

f The material on which the subspecies stelleri seriatus was described came from the Uangkahulu
valley; north side of Matinang range; south side of Matinang range and Buol. There is no indication
from which of these localities the two series in the Hamburg Museum were collected.

52 E. G. EASTON

Palawan : Balabac Island, 2C, BM(NH) 1904.10.5.38-40 (syntypes of everetti).

Borneo : Mt Kinabalu, Sabah, 6C 3A, BM(NH) 1971.19.12-20 [series A (part), series B (part) :
Sims & Easton, 1972] ; Baram river, Sarawak, iC, Hamburg v.3835 (holotype of barami) ;
Labuan, iC, BM(NH) 1904.10.5.150 (holotype of sarawacensis) ; Sarawak, 8C 6A, BM(NH)
1904.10.5.1336-1340 ; Long Lejok, Sarawak, 6C I3A, BM(NH) 1933.10.6.36-42 & Hamburg
v.i 1956 (stelleri everetti : Michaelsen, 1934) '> Tinjar river, Long Lejok, Sarawak, 8C loA,
BM(NH) 1933.10.6.12-20 & Hamburg v. 11951 (syntypes of beranensis tinjarana] ; Merabah,
Sarawak, 146 I2A, BM(NH) 1904.10.5.1265-70 (syntypes of papillata) ; Putussibau, Kali-
mantan, loC I3A, Leiden 1091 (stelleri : Horst, 1899) ; Nangaraun, Kalimanton, 5C gA, Leiden
1903 (stelleri : Horst, 1899) ; 'the Liang Koeboeng' (the exact location of this locality could not
be established), iC, Leiden 1904 (stelleri ; Horst, 1899) ; Bendjermasin, Kalimanton, iC, Ham-
burg V.4O24 [stelleri (typica) : Michaelsen, 1899] ; Bo river, Kalimanton, 30, Leiden 1900 &
Hamburg V.93O7 [stelleri (typica) : Michaelsen, 1922] ; above Mahakkam rivers, Kalimanton,
iC, Leiden 1897 (holotype of stelleri mahakkami) ; Birang river, Beran district, Kalimanton,
fragments of one or more specimens, Hamburg v. 10576 [type(s) of beranensis}.

Lombok : Swela, iC, Berlin 7214 [this specimen is here separated from the type series of Meta-
pheretima badia (Ude, 1932)].

3. OTHER RECORDS.
Celebes : Koro valley [type(s) of stelleri koroensis] ; Bone valley (types of stelleri bonensis).

Borneo : Baram river, Sarawak (stelleri : Michaelsen, 1896) ; Dorf Poh Trap, Brunei [stelleri
(typica) : Ude, 1932] ; Boentok, on the Barito river, Kalimanton [type(s) of baritoensis].

Metapheretima kinabaluensis (Beddard & Fedarb, 1895)

1. MATERIAL EXAMINED.

Borneo : Mt Kinabalu, Sabah, 126 2A, BM(NH) 1971.19.1-11, 21-23 [series A (part), series B
(part) : Sims & Easton, 1972].

2. OTHER RECORDS.

Borneo : Tamburungare, Mt Kinabalu, Sabah [type(s) of kinabaluensis].

Metapheretima phacellotheca (Michaelsen, 1899)

MATERIAL EXAMINED.

Celebes : Mt Masarang, above Tomohon, several spermathecae and fragments of body wall,
Hamburg v.5195 [type(s) of phacellotheca].

Buru : between Mnges'wain and Leksula, iC, Amsterdam V.ol 302 (stelleri bonensis : Michaelsen,
1934) ; Leksula, iC, Amsterdam.

Metapheretima stelleri (Michaelsen, 1891)

1. MATERIAL EXAMINED.

Sangihe : — , 76 lA, Hamburg v.338, BM(NH) 1904.10.5.162 & Leiden 1903 (syntypes of
stelleri) .

Kepulaud Taulaud : Lobo, Karakelang, iC, Amsterdam V.ol 303.

2. OTHER RECORDS.

Celebes : Matinang range (stelleri everetti : Michaelsen, 1899) ; Bone valley [type(s) of stelleri
annectens].

METAPHERETIMA ELONGATA SPECIES-COMPLEX 53

E. G. EASTON, B.Sc.

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

LONDON SWy

A LIST OF SUPPLEMENTS
TO THE ZOOLOGICAL SERIES

OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

1. KAY, E. ALISON. Marine Molluscs in the Cuming Collection British Museum
(Natural History) described by William Harper Pease. Pp. 96 ; 14 Plates.
1965. (Out of Print.)

2. WHITEHEAD, P. J. P. The Clupeoid Fishes described by Lacepede, Cuvier and
Valenciennes. Pp. 180 ; n Plates, 15 Text-figures. 1967. £4.

3. TAYLOR, J. D., KENNEDY, W. J. & HALL, A. The Shell Structure and Mineralogy
of the Bivalvia. Introduction. Nuculacea-Trigonacea. Pp. 125 ; 29 Plates,
77 Text-figures. 1969. £4-50.

4. HAYNES, J. R. Cardigan Bay Recent Foraminifera (Cruises of the R.V. Antur)
1962-1964. Pp. 245 ; 33 Plates, 47 Text-figures. 1973. £10-80.

5. WHITEHEAD, P. J. P. The Clupeoid Fishes of the Guianas. Pp. 227;
72 Text-figures. 1973. £9-70.

6. GREENWOOD, P. H. The Cichlid Fishes of Lake Victoria, East Africa : the
Biology and Evolution of a Species Flock. Pp. 134 ; i Plate, 77 Text-figures.
1974. £3'75- Hardback edition £6.

Printed in Great Britain by John Wright and Sons Ltd. at The Stonebndge Press, Bristol BS4 5NU

GENERAL
2 4 SEP 197
\ ^ LIBRARY

THE TAXONOMY OF CONCHOECIA&*L«0
(OSTRACODA, HALOCYPRIDIDAE)
OF THE GAUSSI AND EDENTATA
GROUPS FROM THE NORTHEAST
ATLANTIC WITH A NOTE ON
THEIR ECOLOGY

A. J. GOODAY

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. 3

LONDON: 1976

GENERAL

2 4 SEP 197

THE TAXONOMY OF CONCHOECIA

(OSTRACODA, HALOCYPRIDIDAE) OF THE

GAUSSI AND EDENTATA GROUPS FROM THE

NORTHEAST ATLANTIC WITH A NOTE ON

THEIR ECOLOGY

BY

ANDREW JOHN GOOD AY

Institute of Oceanographic Sciences

Pp 55-100 ; 22 Text-figures

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. 3

LONDON: 1976

THE BULLETIN OF THE BRITISH MUSEUM

(NATURAL HISTORY), instituted in 1949, is
issued in five series corresponding to the Scientific
Departments of the Museum, and an Historical
series.

Parts will appear at irregular intervals as they
become ready. Volumes will contain about three or
four hundred pages, and will not necessarily be
completed within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 30 No. 3 of the Zoology series.
The abbreviated titles of periodicals cited follow those
of the World List of Scientific Periodicals.

World List abbreviation
Bull. Br. Mus. nat. Hist. (Zool.)

ISSN 0007-1498

Trustees of the British Museum (Natural History), 1976

BRITISH MUSEUM (NATURAL HISTORY)

Issued 23 September, 1976 Price £3.70

THE TAXONOMY OF CONCHOECIA

(OSTRACODA, HALOCYPRIDIDAE) OF THE

GAUSSI AND EDENTATA GROUPS FROM THE

NORTHEAST ATLANTIC WITH A NOTE ON

THEIR ECOLOGY

By A. J. GOODAY

CONTENTS

Page

SYNOPSIS ........... 57

INTRODUCTION .......... 57

TAXONOMIC SECTION ......... 59

edentata group .......... 59

Conchoecia subedentata sp. n. ....... 60

Conchoecia aff. edentata ........ 69

gaussi group .......... 77

Conchoecia incisa ......... 78

Conchoecia gaussi ......... 85

ECOLOGY. ........... 93

ACKNOWLEDGEMENTS ......... 99

REFERENCES ........... 99

SYNOPSIS

Conchoecia subedentata sp. n. and C. aff. edentata are described from the Northeast Atlantic
and placed in a new edentata species group based on C. edentata, an Antarctic species. C.
edentata was previously assigned to the gaussi group which is here restricted to C. gaussi and
C. incisa, both of which are redescribed.

C. subedentata sp. n. and C. aff. edentata are commonest at 6o°N and decrease in abundance
southwards. C. subedentata sp. n. occurs mainly between 300 m and 900 m depth, C. aff.
edentata is most frequent below 800 m. C. gaussi is most abundant at 53°N and C. incisa at
3O°N. C. incisa males are rare and restricted to below 1250 m, while females and juveniles
have two peaks of abundance, below 800 m and between 600 m and the surface. C. gaussi
occurs mainly below 1000 m. None of the species appears to undergo diurnal vertical migration
except for females and juveniles of C. incisa where all specimens in the top 200 m were taken
at night.

Taken together adults and juveniles of C. gaussi and C. incisa fall into three or four size
classes which may have some trophic significance.

INTRODUCTION

SKOGSBERG (1920 : 676) proposed the gaussi group of species to include Conchoecia
incisa Muller, 1906 and Conchoecia gaussi Muller, 1908. Later Rudyakov (1962)
placed C. edentata Muller, 1906 close to C. gaussi and Deevey (1974 : 368) included it

A. J. GOODAY

TABLE i

Station details and gear used

STATION

LOCATION IN
NORTHEAST ATLANTIC

58i4

58i6

5818
5819

* 28°N,

I4°W

5823

5825..

6i55"

6156

>• 28°N. 14

°w

6164

i

6176

6l8A 34°

6i93 / 34

N,

So

°W

6674 11°

35'N,

2I°26'W

7089 i 8°

N,

25

°W

7406")

7478 }- 48°

N,

20

°W

748oJ

7709 60°
7711 53°

N,
N,

20
20

°W
°W

}

8262
8264
8265
8270
8271
8272

i8°N, 25°W

32°N, i6°W"
32°N, 24°W
32°N, 27°W
34°N, 34°W
34°N, 39°W
34°N, 43°W,

DATE

Sept. to Dec. 1965

>Oct. to Nov. 1966

Jan. to April 1968
Oct. to Dec. 1969

Sept. to Nov. 1970

April to June 1971
Feb. to April 1972

Feb. to March 1973

Discovery
CRUISE No.

GEAR

Nn3

15

21
30

36

39
45

52

RMT 1+8
RMT i + 8

RMT i +8

in the gaussi group, while in a revision of Conchoecia Poulsen (1973 : 103) established
the genus Gaussicia for these three species. Unfortunately some of the genera into
which Poulsen subdivided Conchoecia are questionable. In this paper I therefore
disregard Gaussicia until Poulsen's taxonomy can be thoroughly assessed. I also
believe it is wrong to combine C. edentata with C. gaussi and C. incisa. In my view
C. edentata and its allies described in this paper differ sufficiently from C. gaussi
and C. incisa to warrant their placement in a new edentata group.

Plankton samples collected during recent years in the Northeast Atlantic by the
RRS Discovery have yielded abundant material of the gaussi and edentata groups.
Specimens resembling C. edentata predominate and faU into two size classes The
larger form is closer to Miiller's (1906) original description of C. edentata and is assigned
here to C, aff. edentata while the smaller form is described as C. subedentata sp. n.
In addition, the two poorly known species of the gaussi group, C. gaussi and C. incisa,
are redescribed, and the ecology of members of the gaussi and edentata groups is
discussed.

CONCHOECIA FROM THE NORTHEAST ATLANTIC 59

Material and methods

The material was obtained from 27 stations in the Northeast Atlantic during
seven cruises by RRS Discovery between 1965 and 1973. Details of the stations
are summarized in Table i. On earlier cruises, aim2 net (Nii3) fitted with a
catch dividing bucket was employed (Foxton, 1969). This was superseded in 1969
by the combination net (RMTi + 8) described by Baker, Clarke & Harris (1973).
Most of the ostracods were caught in the RMT I part of the combination net although
a few adult specimens of C. incisa and C. gaussi were picked out from the RMT 8
samples.

A few characters measured must be briefly defined here. Carapace length
includes the rostrum ; carapace height and breadth are maximum values.
Segments i and 2 of the first antenna are measured along their dorsal margins and
the first segment is taken to begin at the base of the frontal organ. Protopodite
length includes the anterior hook and segments 2-8 of the second antenna exopodite
are measured from the distal end of segment i and not from the proximal end of
segment 2.

TAXONOMIC SECTION

edentata group

DIAGNOSIS. Both valves bear median ventral glands ; asymmetric glands are
in usual positions. Male first antenna with long a and c setae and pad on b seta ; e
seta has proximally pointing spines which are difficult to distinguish. Female first
antenna has very short dorsal seta, a-d setae are slightly longer or just shorter than
e seta which bears group of proximally pointing spines on distal flattened and
widened part. Male second antenna with g seta slightly widened and flattened
distally ; in female, g seta is rather wider and sword-shaped. Serrated teeth of
mandibular basale low with weak primary cusps. Maxilla with only four posterior
setae. Labrum has weak hyaline plate and small weak teeth. Female sixth limb
with characteristically long and flexible ventral seta and very short dorsal seta on
final segment.

REMARKS. Conchoecia edentata of Miiller (1906) was included with C. gaussi and
C. incisa in the new genus Gaussicia (= gaussi group) by Poulsen (1973). However
this species, together with C. aff. edentata and C. subedentata sp. n. described here,
differs from C. gaussi and C. incisa in numerous characters which are summarized
in Table 5. In contrast the points of similarity are rather few : the presence of the
ventral gland, the general character of the b, d and e seta armature on the male
first antenna, the shape of the female g seta (second antenna) and e seta (first
antenna), the development of minute spines on various segments of the female first
antenna and the male and female second antenna and perhaps the same general
structure of the cutting edge and tooth lists of the mandibular coxale.

Because of the large number of differences (Table 5) it seems sensible to remove
C. edentata, C. aff. edentata and C. subedentata sp. n. from the gaussi group and place
them in a new edentata group. The arrangement is supported by the fact that many

60 A. J. GOODAY

differences between the two groups are concerned with characters of the mandible,
maxilla, fifth and sixth limbs and labrum. These limbs are more conservative in
their structure than the first and second antennae and so their characteristics must
carry more taxonomic weight.

The edentata group is characterized by several features, (i) The development
of a median ventral gland ; this is also present in C. gaussi and in C. prosadena
the right asymmetric gland (not homologous with the ventral gland in the edentata
and gaussi groups) is situated at the posterior end of the ventral margin, (ii) The
a-d setae of the female first antenna are slightly shorter or slightly longer than the
e seta. In no other species of Conchoecia do these setae approach the length of the
e seta. The group of proximally pointing spines with expanded bases on the distal
part of the e seta are probably also peculiar to this group, (iii) On the terminal
segment of the female sixth limb the ventral a seta is very long, thin and flexible,
the claw-like median b seta rather longer or shorter and the dorsal c seta is very
short. These setae are probably diagnostic for the edentata group although
characters of the sixth limb have generally been ignored in species descriptions.
However, in all species of Conchoecia for which information is available the terminal
setae of this limb are claw-like and shorter than the two ventral setae of the edentata
group.

The edentata group is probably closely related to the gaussi group. It also shares
a number of characters with the mollis group : distally widened g setae on the male
and female second antennae and a distally widened e seta on the female first antenna,
long a and c setae of the male first antenna and a pad on the b seta of this limb.
C. congolensis Poulsen, 1969, which Poulsen (igGga : 116) felt was related to the
bispinosa and gaussi groups, may be closer to the edentata group. In the only known
specimen, a female, the dorsal seta is short and the a-d setae are long, although not
as long as in the edentata species group. However the long, rod-shaped capitulum
and the absence of a ventral gland distinguish C. congolensis from species of both the
edentata and gaussi groups.

Conchoecia subedentata sp. n.

(Figs 2-5)

DIAGNOSIS. Species of edentata group characterized by small size (male carapace
length 1-123-1-280 mm, mean 1-21210-030 mm) and setation of final segment of
female limb where ventral a seta is longer than middle claw b seta.

MATERIAL. 1447 specimens comprising 388 females, 392 males and 667 juveniles.

HOLOTYPE AND PARATYPE. The holotype, a male stained with lignin pink and
mounted on slides in Euparal, is deposited in the British Museum (Natural History),
reg. no. 1974.736. The type locality is Discovery station 7711, haul 23 ; 52°57'i'-
52°53-i'N, 20°o-o'-20°3-4'W. Depth 200-300 m. Time 2342-0142 hr. Date 19-
20 May 1971. Gear RMTi + 8. The paratype material comprises a dissected
female reg. no. 1974.737 and some undissected male and female specimens in 70%
alcohol, reg. nos 1974.738-748.

CONCHOECIA FROM THE NORTHEAST ATLANTIC

61

10

Numbers 50

40
30
20
10

aff. edentata p

IV V VI |

n r-i n r-n i"i D i tlj 1 rTf^^^r^Ti irFl^lfon n

0.75 1.00

n1-25 1.50 1.75

" pi

prr,

[j:::::;| Adult Females

subedentata J

I 1

^ Adult Males

/ y
/ 0

V VI

M

/ /

W

!H Stage VI Males

IV

I Tl

Eg;j

1

: •: ~] ~] Other Juveniles

^ ^ x

<??'.' V • •

m ^-r

J_ ;;

- fi^

;

. '. v n

dltoJ]

1

| ^^

•

ft,

0-75

1.00 / 1-25 v

Length (mm)

1.50

FIG. i. Histograms showing adult and juvenile carapace lengths of Conchoecia aff. edentata

and Conchoecia subedentata sp. n.

1.0

FIG. 2. Conchoecia subedentata sp. n. Lateral and ventral carapace outlines of adults

and final three juvenile instars.

62 A. J. GOODAY

DESCRIPTION OF MALE. Carapace (Fig. 2). The carapace lengths of 216 speci-
mens range from 1-123 mm to 1-280 mm with a mean value of 1-212 + 0-030
(Fig. i). Height is less than half the length, and breadth is about 80% of the
height. In lateral view the posterior margin is gently curved, joining the dorsal
margin at a rounded angle and the ventral margin evenly. The ventral margin is
either straight or slightly curved and the lateral outline narrows slightly towards
the anterior end. There is no surface ornamentation. The right asymmetric
gland is displaced up the dorsal margin, opening at about f of the carapace height.
The median ventral gland is as large or rather larger than the right asymmetric
gland.

Frontal organ (Figs 3 A, C). The stalk extends just beyond the end of the first
antenna. The proximal half of the capitulum tapers forwards, while distally the
dorsal margin is nearly straight, the ventral margin slightly convex and the end
asymmetrically rounded. Proximally, the lower surface of the capitulum bears
numerous prominent forward directed spines and the distal part has a few scattered
spines.

First antenna (Figs 3 A, B). Segments 1-4 are usually bare but occasionally
have a few scattered spines. The first segment is shorter than the second. The a
seta is longer than the limb and swollen near the base ; it loops down proximally
and then lies parallel to the limb for the rest of its length. The b seta is slightly
shorter than the d seta and bears a distal pad ; adjacent to the pad on the anterior
side of the seta are 8-n closely spaced spines preceded proximally by 6-10 widely
spaced spines. The c seta is somewhat shorter than the a seta and usually lies
parallel to the antenna. The d seta carries 3-5 small anterior spines at a point level
with the b seta pad, and sometimes has 2-3 posterior spines. The e seta armature
is difficult to discern. In side view the posterior margin of the seta is raised into
17-22 low 'bumps', in front view the 'bumps' appear as vague paired patches from
which arise indistinct and proximally directed spines. This main armature is
preceded by 4-7 small anterior spines.

Second antenna (Figs 3D-G). The protopodite surface is bare but the first
exopodite segment has numerous minute spines concentrated proximaUy where they
form a dense covering visible only at high magnification. On the first endopodite
segment, the processus mammillaris has a rounded end and no terminal tubercule.
The b seta is about f the length of the a seta and the long slender upward curving
c seta is some three times as long as the downward curving d seta ; the e seta is a
small spine. The a, b and c setae bear fine hairs. The / seta is over | the length
of the g seta ; it is thin walled for much of its length and carries 4-7 minute anterior
spines. The g seta is flattened and slightly widened distally. On the proximal
tubular part it has 7-9 anterior and 8-10 posterior spines. The h, i and/ setae are
shorter than the / seta ; the j seta is rather swollen at its base and bears prominent
spines, most on the swollen part but some below. The left hook appendage has a
short proximal section, a rounded angle of about 90°, a long slightly curved distal
section and a rounded tip with a few indistinct subterminal ridges. The right hook
appendage is larger with an angle of about 135° between the proximal and distal
sections and about 8 prominent subterminal ridges.

FIG. 3. Male dimorphic parts of Conchoecia subedentata sp. n. ; arrow indicates anterior.
(A) First antenna and frontal organ. (B) Main armature of b and e setae of first antenna.
(C) Capitulum of frontal organ. (D) Second antenna. (E) Second and third segments
of endopodite of right second antenna. (F) Endopodite of left second antenna. (G) g
seta of second antenna. (H) Penis. (I) Sixth limb.

5*

FIG. 4. Non-dimorphic parts of Conchoecia subedentata sp. n. ; arrow indicates anterior.
(A) From top to bottom, cutting edge, distal and proximal tooth lists of mandibular
coxale. (B) Cutting edge and adjacent setae of mandibular basale. (C) Basale, endo-
podite and exopodite of mandible. (D) Labrum. (E) Maxilla. (F) Caudal furca.
(G) Fifth limb.

CONCHOECIA FROM THE NORTHEAST ATLANTIC 65

Mandible (Figs 4A-C). The toothed edge of the coxale slopes down posteriorly.
It has 19-23 teeth comprising one elongate anterior tooth, 5-6 fairly large teeth and
13-17 small teeth which decrease in size posteriorly ; the posterior £ of the toothed
edge lacks teeth. The distal list has a large posterior tooth and 16-19 regular teeth
of equal size. The proximal list has a large posterior tooth and 15-19 smaller teeth
of rather variable sizes ; the inner surface of this list is covered with papillae. The
toothed edge of the basale has the usual arrangement of two spine teeth, six serrate
teeth and an unusually long and low outer tooth. Only the posterior two serrate
teeth have well-defined main cusps ; on the remaining teeth, the main cusps are
only slightly larger or are indistinguishable from the secondary cusps. There are
two short setae near the spine teeth on the inner surface of the basale. The second
exopodite segment has some fine hairs on the outer distal edge, the third segment has
a more extensive distal area of hairs.

Maxilla (Fig. 4E). The anterior margin has four long setae, the distal two longer
than the proximal two. The posterior margin has three setae. The basal seta
extends to the end of the limb.

Fifth limb (Fig. 4G). On the dorsal side of the first exopodite segment, the distal
seta extends only as far as the base of the dorsal seta on the next segment.

Sixth limb (Fig. 3!). This is typical for the genus.

Labrum (Fig. 4D). The hyaline membrane is fragile with a deep V-shaped
notch. On each side of the notch are about nine small inward facing teeth which
become more closely spaced towards the inside.

Caudal furca (Fig. 4F). The caudal lamellae are covered with fine hairs.

Penis (Fig. 3H). The upper margin is almost straight and the lower margin
curved. There are five transverse muscle bands.

DESCRIPTION OF FEMALE. Carapace (Fig. 2). The carapace lengths of 190
specimens range from 1-261 mm to 1-438 mm with a mean value of 1-345 ± 0-041 mm
(Fig. i). Apart from being rather more elongate, the lateral outline is identical to
that of the male. There is no surface ornamentation.

Frontal organ (Fig. $C). The capitulum is distinct from the shaft. It tapers
slightly anteriorly, with a rounded end and sometimes a slight concavity below the
end. The capitulum is armed with long prominent spines on the ventral half and
smaller, rather less, extensive spines dorsally.

First antenna (Figs 5A, B). The second segment is longer than the first and one
or both may have a few scattered spines, visible only at high magnification. The
dorsal seta is very short ; the a-d setae are longer than the limb and the a seta is
somewhat expanded proximally. The e seta is rather shorter than the more
proximal setae. On the posterior side it bears 29-30 spines of varying sizes. The
proximal tubular part of the seta has 16-19 spines of which the first 6-8 are small and
difficult to distinguish and the remainder long and prominent. The distal flattened
part of the seta bears 3-5 long distally pointing spines followed by 8-12 proximally
pointing spines with unusually wide bases and 3-4 minute distally directed spines.
On the anterior side are 10-13 spines, about % on the distal flattened part of the seta.

Second antenna (Figs 5D-F). The armature of the second exopodite segment
and the a and b setae are as for the male limb. The /seta is over £ the length of the

FIG. 5. Female dimorphic parts of Conchoecia subedentata sp. n. ; arrow indicates anterior.
(A) First antenna and frontal organ. (B) e seta of first antenna. (C) Capitulum of
frontal organ. (D) Second antenna. (E) Endopodite of left second antenna. (F) g
seta of second antenna. (G) Terminal segment and setae of sixth limb.

CONCHOECIA FROM THE NORTHEAST ATLANTIC

67

g seta and in some specimens has 1-2 minute anterior spines. The g seta is flattened
and widened in its distal half. Posteriorly it bears 19-26 spines of which the
proximal 5-8 lie on the tubular part of the seta and on the anterior side it carries
6-8 similar spines, all on the tubular section. The h, i and j setae are slightly
shorter than the / seta and the j seta bears a few spines proximally.

Sixth limb (Fig. 56). On the final segment the three setae are here designated
a, b and c from ventral to dorsal (Fig. 5G) and their relative lengths are summarized
in Table 4. The a seta is thin, long and flexible, the b seta is claw-like and shorter
than the a seta while the c seta is a short spine.

JUVENILES (Fig. 2). The range of carapace lengths in 24 Stage IV juveniles is
0-709-0-808 mm (mean 0-745 ± 0-014 mm) ; 57 Stage V juveniles have a length
range of 0-827-0-926 mm (mean 0-888 + 0-021 mm). Male and female Stage VI
instars may be distinguished by the presence of immature eggs in the females and a
rudimentary penis and testes in the male. The males are rather smaller (1-024-
1-123 mm ; mean 1-066 ± 0-013 mm in 44 specimens) than the females (1-084-
1-182 mm ; mean 1-1241 0-019 mm m 37 specimens). Juvenile and adult carapace
lengths are summarized in a histogram (Fig. i).

TABLE 2
Meristic characters of the males of the four described species

N
Length: range (mm)
Length: mean, s.d. (mm)

C. subedentata
sp. n.

216
1-123-1-280
1-210 0-030

C. aff.
edentata

4°
1-400-1-517
1-460 0-027

C. incisa C. gaussi

8 10
1-822-1-911 2-955-3-330
1-877 0-030 3-138 0-107

N

20

10

8

6

X

s.d.

X

s.d.

X

s.d.

X

Height

46-27

0-90

47-47

1-26

43-66

0-76

44-64

Breadth

40-83

1-24

40-39

I-O2

38-49

1-42

40-93

FO shaft

32-30

1-29

31-53

0-69

28-10

o-44

30-33

capitulum

11-51

0-36

n-57

0-69

18-63

1-17

16-52

ist Antenna segment i

10-07

0-31

11-49

0-70

11-15

0-27

10-99

segment 2

18-09

o-45

16-38

o-43

10-79

0-16

13-93

total

28-84

0-78

27-87

0-62

21-95

0-38

24-92

a seta

36-87

1-49

35-89

i-75

38-44

I-I5

22-68

b seta

42-26

0-92

40-40

1'33

46-27

0-84

48-79

c seta

30-63

1-46

30-96

1-92

12-68

o-73

13-21

d seta

42-44

1-03

4I-39

1-09

39-86

o-95

49-18

e seta

50-91

1-34

51-81

i-39

51-80

2-25

52-00

2nd Antenna protopodite

53-04

1-32

52-09

0-51

47-72

1-23

46-82

exopodite segment i

16-80

0-39

I7-55

0-27

17-54

0-36

19-26

segments 2-8

8-19

0-31

8-32

0-22

7-95

0-27

7-19

LSS

48-28

1-56

47-70

i-39

48-96

0-28

42-54

g seta

45-96

1-07

48-96

0-48

52-86

1-29

42-77

/seta

24-70

0-94

28-13

1-78

39-46

2-16

39-98

h-j setae

21-44

1-78

22-59

1-29

20-57

1-66

27-31

All measurements, other than the carapace length, are expressed as percentages of the carapace
length. The measurements are mean values and are followed by standard deviations. N is the number
of specimens measured. LSS is the longest swimming seta. FO is the frontal organ.

68 A. J. GOODAY

REMARKS. C. subedentata sp. n. is identical in many respects to C. aff. edentata.
It is compared and contrasted with this form in detail below but the main points
may be summarized here, (i) The mean carapace length in C. subedentata sp. n.
is 81% (?) to 83% (cJ) that of C. aff. edentata ; (ii) the / and g setae of the male and
female second antennae are proportionally shorter in C. subedentata sp. n. (Tables 2

TABLE 3

Meristic characters of the females of the four described species
C. subedentata C. aff.

N
Length: range (mm)
Length: mean, s.d. (mm)

sp. n.

190
1-261-1-438
1-345 0-041

edentata

1-576-1-734
1-651 0-067

C. incisa

39
2-462-2-837
2-624 0-062

C. gaussi
4

3-462

N

20

10

8

4

X

s.d.

X

s.d.

X

s.d.

X

Height

46-08

0-99

47-20

1-29

49-54

1-23

46-94

Breadth

38-37

0-85

38-77

I-I2

4I-65

1-49

41-08

FO total length

29-83

0-69

28-77

0-74

28-92

0-91

35-58

ist Antenna segment i

6-75

0-36

6-89

0-52

4-88

0-18

5-73

segment 2

10-33

0-69

9-54

0-70

6-71

0-14

8-25

total

16-98

0-58

16-66

0-65

11-58

0-26

I3-I3

a-d setae

29-51

1-57

29-15

1-78

11-78

0-84

14-77

e seta

28-77

0-77

33-87

0-O2

32-22

0-64

37-02

2nd Antenna protopodite

47-32

0-80

44-96

0-70

38-83

0-83

42-21

exopodite segment i

15-04

0-72

15-67

o-34

14-23

0-24

17-47

segments 2-8

7-23

0-29

7-25

0-19

6-46

0-17

6-98

LSS

41-42

1-28

42-18

0-84

34-58

1-50

36-47

g seta

32-24

o-34

34-79

0-66

28-14

0-49

33-98

/seta

18-18

0-81

20-27

0-67

19-64

0-91

26-34

h -j setae

22-36

1-05

24-52

1-40

18-49

0-66

28-01

All measurements, other than the carapace length, are expressed as percentages of the carapace length.
The measurements are mean values and are followed by standard deviations. N is the number of speci-
mens measured. LSS is the longest swimming seta. FO is the frontal organ.

TABLE 4

Meristic characters of the sixth limb in males and females of C. subedentata sp. n. and

C aff. edentata

C. subedentata sp. n.

C. aff. edentata

$

<j

$

c?

N

9

6

6

6

Segment i

I4-32

12-58

14-02

11-97

Segment 2

19-09

17-96

18-71

18-01

Total

33-48

30-55

32-73

29-98

a seta

11-09^1

I2-59"|

b seta

16-51 i.

37-69

12-56 J.

40-50

c seta

*-55j

3-o6j

All measurements are mean values and are expressed as percentages of the carapace length. N is the
number of specimens measured.

CONCHOECIA FROM THE NORTHEAST ATLANTIC 69

and 3) and the female g seta is relatively wider distally ; (iii) the right hook appendage
extends to just above the base of the c seta in C. subedentata sp. n. but well above this
point in C. aff. edentata ; (iv) on the inner side of the mandibular basale there are
two setae near the spine teeth but only one in C. aff. edentata ; (v) the thin ventral a
seta of the female sixth limb extends well beyond the end of the middle claw b seta
in C. subedentata but not to the end of the b seta in C. aff. edentata and the short
dorsal c seta in C. subedentata sp. n. is about twice the proportional length of this
seta in C. aff. edentata (Table 4).

The characters which separate C. subedentata sp. n. and C. aff. edentata also
distinguish C. subedentata sp. n. from previously described and figured specimens
of C. edentata. Thus C. subedentata sp. n. is much smaller than C. edentata and
differs from the figured male of Miiller (1906 : pi. XV, fig. 2a) and Rudyakov
(1962 : fig. 12), in having a shorter right hook appendage and from the females of
Barney (1921 : fig. 5F) and Rudyakov (1962 : fig. nE) in the relative lengths of the
terminal setae on the sixth limb and from Rudyakov's (1962 : fig. 136) figured
specimens in the presence of two setae rather than one next to the spine teeth on the
mandibular basale.

GEOGRAPHICAL DISTRIBUTION. In the Northeast Atlantic this previously un-
recorded species was found at the following stations ; the number of specimens
caught at each station is given in parentheses : Station 7709, 6o°N, 20°W (708) ;
Station 7711, 53°N, 20°W (408) ; Stations 7406 (245), 7478 (14) and 7480 (i) all
40°N, 20°W ; Stations 8262 (i), 8264 (2), 8265 (i), 8270 (4), 8271 (4), 8272 (3), all
along 32°N, ; Station 7856, 30°N, 23°W (63), Station 7089, i8°N, 25°W (3). The
depth distribution at different latitudes is summarized in Fig. 19.

Conchoecia aff. edentata Miiller 1906
(Figs 6-9)

MATERIAL. Ninety-nine specimens comprising 26 females, 43 males and 30
juveniles. Dissected male and female specimens, stained with lignin pink and
mounted in Euparal, are deposited in the British Museum (Natural History), reg.
nos 1974.749 and 750.

DESCRIPTION OF MALE. Carapace (Fig. 6). The carapace lengths of 40 specimens
range from 1-399 mm to I'5I7 mm witn a mean value of 1-45910-027 mm (Fig. i).
The lateral outline and gland positions are as for C. subedentata sp. n. There is no
surface ornamentation.

Frontal organ (Fig. 7D). As for C. subedentata sp. n.

First antenna (Figs 7A, C). Segments i and 2 are together proportionally rather
longer than in C. subedentata sp. n., are usually bare but may have a few scattered
spines. The a-e setae have about the same proportional lengths as in C. subedentata
sp. n. and similar armature. The b seta bears 9-12 widely spaced spines and 8-10
more closely spaced and rather weaker spines level with the pad. The d seta carries
6-12 weaker anterior spines and sometimes a few similar posterior spines while the
e seta has 19-25 paired 'bumps' from which arise indistinct spines.

A. J. GOODAY

1.0

FIG. 6. Conchoecia aff. edentata. Lateral and ventral carapace outlines of adults.

Second antenna (Figs 76, E-G). The proportional lengths and armature of the
limb segments are as for C. subedentata sp. n. but the /and g setae are proportionally
longer (Table 2). The / seta bears 5-7 minute anterior spines. The g seta has
6- 1 1 small proximal and median posterior spines and 6-9 similar, mainly median
spines on the anterior surface ; the distal flattened part of this seta is bare. The
remaining setae are as described for C. subedentata sp. n. Also the right hook
appendage is markedly longer in C. aff. edentata, extending well above the bases
of the c and d setae, but only just above this point in C. subedentata sp. n.

Mandible (Figs 8A-C). The toothed edge, the proximal and distal lists of the
coxale and the basale toothed edge have the same structure as in C. subedentata sp.n.
The toothed edge of the coxale bears 15-19 teeth comprising a broad posterior tooth,
five large teeth and 10-13 small teeth. The distal list has a large curved posterior
tooth and 18-21 smaller teeth of equal size, while the proximal list has a large
posterior tooth and 15-22 smaller teeth of rather variable size. As in C. subedentata
sp. n. the second and third exopodite segments carry fine hairs. There is only a
single short seta near the spine teeth on the inner surface of the basale compared
with two setae here in C. subedentata sp. n. Also the three small setae at the base
of the dorsal claw seta of the third exopodite segment are appreciably shorter than
in C. subedentata sp. n.

Maxilla (Fig. 8F) and furca (Fig. 8D). As for C. subedentata sp. n.

Labrum (Fig. 8E). This differs from the labrum of C. subedentata in having only
very faint lines on the hyaline plate.

Penis (Fig. 7!!). There are usually six, but occasionally only five, transverse
muscle bands.

Sixth limb (Fig. 7!). As in C. subedentata sp. n. except for the proportionally
rather longer setae on the final segment (Table 4).

DESCRIPTION OF FEMALE. Carapace (Fig. 6). The carapace lengths of 31 speci-
mens range from 1-576 mm to 1-734 mm with a mean value of 1-651 ±0-067 mm

FIG. 7. Male dimorphic parts of Conchoecia afif. edentata ; arrows indicate anterior.
(A) First antenna and frontal organ. (B) g seta of second antenna. (C) Main armature
of (from left to right) d, e and 6 setae of first antenna. (D) Capitulum of frontal organ.
(E) Second antenna. (F) Endopodite of left second antenna. (G) Second and third
segments of endopodite of right second antenna. (H) Penis. (I) Sixth limb.

FIG. 8. Non-dimorphic parts of Conchoecia aff. edentata ; arrows indicate anterior.
(A) Basale, endopodite and exopodite of mandible. (B) From top to bottom, cutting
edge, distal and proximal tooth lists of mandibular coxale. (C) Cutting edge and adjacent
setae of mandibular basale. (D) Caudal furca. (E) Labrum. (F) Maxilla. (G) Fifth
limb.

FIG. 9. Female dimorphic parts of Conchoecia aff. edentata ; arrows indicate anterior.
(A) First antenna and frontal organ. (B) e seta of first antenna. (C) Capitulum of
frontal organ. (D) Second antenna. (E) Endopodite of right second antenna. (F) g
seta of second antenna. (G) Final segment and setae of sixth limb.

74 A. J. GOODAY

(Fig. i). The lateral outline and gland positions are similar to C. subedentata sp. n.
There is no surface ornamentation.

Frontal organ (Fig. gC). As for C. subedentata sp. n.

First antenna (Figs gA, B). As for C. subedentata sp. n. except that the e seta
is proportionally longer (Table 2) and bears 36-46 posterior spines. These comprise
18-24 spines on the proximal tubular part of the seta (11-12 of these are small and
difficult to distinguish), 7-10 long distally pointing spines on the distal flattened
section followed by 7-9 proximally pointing spines with expanded bases and 3-5
weaker distally pointing spines.

Second antenna (Figs gD-F). The protopodite is proportionally shorter and the
f-j setae are proportionally rather longer than in C. subedentata sp. n. (Table 3).
The first exopodite segment usually bears a few small scattered spines. The / seta
has up to six small anterior spines. The distal flattened part of the g seta is more
slender than in C. subedentata sp. n. On the anterior side of this seta are 19-26
spines of which the proximal 5-8 lie on the tubular part of the seta ; the posterior
side bears 6-8 similar spines, all on the tubular part. The _; seta carries a few proxi-
mal spines.

Sixth limb (Fig. qG). The ventral a seta is considerably shorter than in C.
subedentata sp. n. and does not extend beyond the b seta. The c seta is about twice
the proportional length of this seta in C. subedentata sp. n. (Table 4).

JUVENILES. Earlier (IV and V) juvenile stages of C. aff. edentata overlap in
length with the corresponding stages of C. subedentata sp. n. and so are difficult to
distinguish. The following lengths are of a few specimens which exceed the
maximum size for the corresponding instar of C. subedentata and so probably belong
to C. aff. edentata (Fig. i) : Stage IV, 0-808-0-847 mm, mean 0-827 mm (3 specimens) ;
Stage V, 0-985-1-083 mm, mean 1-047 mm (6 specimens). Male Stage VI juveniles
range from 1-162 mm to 1-300 mm, mean 1-281 (5 specimens) and females from
1-281 mm to 1-320 mm, mean 1-311 mm (7 specimens). Stage VI juveniles of
C. aff. edentata are thus comparable in size to adult males and small adult females
of C. subedentata sp. n. and so easily distinguishable from the final juvenile instars
of this species.

REMARKS. This Northeast Atlantic form is at variance with Miiller's original
description of C. edentata (1906 : 76, pi. 15, figs 24-29) from the Antarctic in three
important respects, (i) Miiller's single adult specimen, a male, is longer (1-70 mm)
than Atlantic males, (ii) The carapace of Miiller's specimens has 'ziemlich auffalliger'
dorsal and anterior striations, 'ungewohnlich umfangreich' asymmetrical glands and
an 'umfangreiche sehr auffallige' group of gland cells in the middle of the ventral
margin (Miiller 1906 : 77). In contrast the Discovery specimens have no surface
ornamentation and the glands are much smaller, (iii) The c seta of the first antenna
is considerably longer in the Discovery material than in Miiller's male (pi. XV,
fig. 27).

There is insufficient evidence to decide whether these differences warrant separating
the North Atlantic and Antarctic forms as distinct species or whether they are only
intraspecific variations. Until Antarctic specimens are redescribed, it seems best

CONCHOECIA FROM THE NORTHEAST ATLANTIC 75

to leave the nomenclature of the North Atlantic form open and assign it to
C. aff. edentata.

Descriptions of C. edentata by later workers also differ more or less from the form
described here and, like Miiller's, the material of these authors was caught outside
the North Atlantic. Barney (1921 : 183, 184) first described the female (from the
Antarctic) and found the ventral gland to be less noticeable than in the male'
although his fig. 5a shows it to be larger than in the Discovery material. Also 'the
anterioventral curve [of the outline] viewed from the inside has an imbricate
appearance' in Barney's specimen. Although there are no other differences,
Barney's figures and descriptions are too inadequate for a meaningful comparison
to be made. However, it is significant that his specimen (1-62 mm) is comparable
to the mean carapace length of North Atlantic females (i'65i + 0-067 mm).

Rudyakov (1962) records three adults and two juveniles from the Northwest
Pacific. His description and figures agree well with the form described here.
There are a few minor differences : Rudyakov's male has no small spines on the
first exopodite segment of the male second antenna, and the d seta of this limb is
shorter (see Rudyakov, figs 126, G), the proximal mandibular tooth list has more
teeth, there are no hairs above the spine teeth of the mandibular basale (according
to Rudyakov, fig. 136), in fig. nG of Rudyakov the basal seta, the lateral and three
posterior setae of the maxilla are shorter than in my specimens and the hyaline
plate of the labrum has fewer teeth (Rudyakov, fig. 12). These differences are
probably only of intraspecific significance. Of more importance are the carapace
lengths of Rudyakov's adult female (1-83 mm) and two Stage VI females (1-43 and
1-45 mm) which are much greater than the corresponding lengths for the North
Atlantic form, although his two adult males (lengths also 1-43 and 1-45 mm) measure
close to the mean carapace length of North Atlantic males. This size difference
between the females but not the males is puzzling and makes it difficult to assess the
conspecificity or otherwise of the two forms.

Poulsen (1973 : 107, 108) attributes a juvenile female from the Pacific off New
Caledonia to C. edentata. This individual differs in several respects from C. aff.
edentata. It is much longer (1-50 mm) than Discovery Stage VI juvenile females,
the dorsal seta of the first antenna is longer (Poulsen, fig. 53g) ; there are 'medium-
long' hairs on the lunular bulge of the second antenna (fig. 53c) ; on the mandibular
basale, the spine teeth are longer and the primary cusps of the serrated teeth are
more pronounced (fig. 53d) ; the edge of the mandibular coxale lacks teeth (fig. 53d),
the proximal and distal tooth lists have two large posterior teeth (fig. 53d). These
differences are considerable and Poulsen's specimen probably belongs to a distinct
species.

Finally Deevey (1974 : 368) records three females (1-65, 1-73 and 1-73 mm long)
and two juveniles (1-40 and 1-45 mm long) from the South Atlantic. The females
fall just within the size range of C. aff. edentata although the juveniles are rather
larger than Stage VI specimens in the Discovery material. In the absence of
additional details the comparison can be taken no farther. Other authors (Miiller,
1908, 1912 ; lies, 1953 ; Hillman, 1967, 1969) give no descriptions, figures or
lengths.

76

A. J. GOODAY

GEOGRAPHICAL DISTRIBUTION. Miiller (1906 : 76), 1$, i juvenile ? ?, Indian Ocean
sector of Antarctic, 55°S ; Miiller (1908 : 65), Antarctic, 6i°S and Indian Ocean, o° ;
Barney (1921 : 163), 2? (i juvenile), 2$ (i juvenile), Antarctic, 72°S, o-iooo m ;
lies (1953 : 263), i$, Atlantic 23°S, 500-250 m ; Rudyakov (1962), 2$, 3$ (2 juvenile),
Northwest Pacific, 44°N (1568-1900 m), 46°N (1190-2500 m), 49°N (580-1300 m) ;
Hillman (1967, 1969), Antarctic, numerous stations between 5o°S and 70°$ ; Deevey
(1974 : 368) 3$, 2 juveniles, South Atlantic between 23°S and 55°S. As discussed
above, most of these specimens are regarded as only questionable synonyms of
C. aff. edentata.

In the Northeast Atlantic, this form was found at the following stations ; the
number of specimens caught at each station is given in parentheses : Station 7709,
6o°N, 20°W (37) ; Station 7711, 53°N, 20°W (21) ; Stations 7406 (18), 7478 (3),
both 40°N, 20°W ; Station 7856, 3O°N, 23°W (i) ; Station 7089, i8°N, 25°W (23) ;
Station 6665, n°N, 20°W (2). The depth distribution at different stations is
summarized in Fig. 20.

TABLE 5
Summary of the main differences between the edentata and gaussi species groups of Conchoecia

edentata group

1. Capitulum of <$ and $ frontal organ
shorter (Tables 2 and 3)

2. c? and $ first antenna longer (Tables 2
and 3)

3. Pad on b seta of <$ first antenna relatively
short

4. c seta of (J first antenna longer than limb
(Table 2)

5. a-d setae of $ first antenna longer or
slightly shorter than e seta (Table 3)

6. e seta of $ first antenna with distal,
proximally directed spines

7. Dorsal seta of $ first antenna very short

8. Protopodite of second antenna longer,
particularly in $ (Tables 2 and 3)

9. No hairs on protopodite

10. h and i seta of $ and <$ second antenna
spineless

11. / seta of £ second antenna slightly longer
than h-j setae (Table 2)

12. Ventral surface of mandibular basale
lacks hairs

13. Only one marginal seta on inside of first
mandibular exopodite segment

14. 4 posterior setae on maxilla endopodite

15. Labrum with small weak teeth

1 6. Short distal seta on dorsal surface of first
exopodite segment of fifth limb

17. $ sixth limb with long flexible a seta,
claw-like b seta and very short c seta

gaussi group

Pad elongate

Short and wide

These setae much shorter than e seta

All spines point distally

Long and extends well beyond limb

Hairs present behind endopodite
Spines near base

Much longer than h-j setae

Carries fine hairs

3 additional smaller setae inside margin

6 setae

Teeth larger and stronger

This seta long, extending to final segment

All 3 setae are claw-like and a seta is
shortest

CONCHOECIA FROM THE NORTHEAST ATLANTIC 77

gaussi group

gaussi group Skogsberg, 1920 : 676, 677.
Gaussicia Poulsen, 1973 : 103, 104 (in part).

DIAGNOSIS. Both valves with median glands ; asymmetric glands are in usual
place. Male first antenna with a seta reaching to base of limb (C. incisa} or to end
of first segment (C. gaussi) ; c seta is short, flat and widened with short continuation
above horizontal basal stalk ; armature of b, d and e setae is as in edentata group.
Dorsal seta of female first antenna is long, a-d setae are short. Male second antenna
has blunt ended (C. incisa) or narrow and tapering (C. gaussi) f seta, and g seta is
blunt ended (C. incisa) or slightly widened and then tapered distally (C. gaussi).
Maxilla bears six anterior setae. Labrum with numerous large teeth. Female
sixth limb carries three terminal claw setae.

REMARKS. The gaussi group is restricted in this paper to C. gaussi and C. incisa,
the two species originally included by Skogsberg (1920). This group is probably
related to the bispinosa group to which Miiller (1906, 1908) assigned C. incisa and
C. gaussi. These species, particularly C. gaussi, resemble typical members of the
bispinosa group (i.e. C. bispinosa Glaus, 1890, C. secernenda Vavra, 1906, C. haddoni
Brady & Norman, 1896 and C. striola Miiller, 1906 as redescribed by Angel, 1970 and
Poulsen, 1973) in a number of the characters which were used in Table 5 to distinguish
them from the edentata group. The main points of similarity are the short a-d setae
and long dorsal seta of the female first antenna, the short hairs behind the endopodite
on the female second antenna, the development of spines in both sexes near the bases
of the h and i setae of the second antennae, the four setae on the inside of the first
mandibular exopodite segment and the ventral hairs on the basale of this limb, the
six anterior setae of the first endopodite segment of the maxilla, the rounded notch
and well-developed teeth of the labrum, the long distal dorsal seta on the first
exopodite segment of the fifth limb and the three stiff claw setae on the final segment
of the female sixth limb. In addition, C. gaussi was described by Miiller (1908) as
having several enlarged gland cells near the middle of the posterior margin resembling
the gland cells which he regarded as typifying the bispinosa group. These cells
were also observed by Skogsberg (1920) in C. gaussi but were not visible in any
Discovery specimens. Angel (igGSb) reported that glands in this region release a
bioluminescent secretion in C. secernenda and C. bispinosa.

The gaussi group is distinguished from the bispinosa group by the presence of the
ventral gland, the absence of the c seta and the distally widened g seta (thin and
tapered in bispinosa group) on the female second antenna, the shorter c seta of the
male second antenna and the g seta of this limb which is tapered in the bispinosa
group but square ended in C. incisa and in C. gaussi is widened distally before tapering
and finally the absence of a terminal spine on the hook appendages (see Skogsberg,
1920 and Granata & di Caporiacco, 1949 : 12, 13). Also the c seta of the male first
antenna is flattened and widened and extends dorsally above the basal stalk in the
gaussi group ; this seems to be a unique feature.

78 A. J. GOODAY

Conchoecia incisa Miiller, 1906
(Figs 10-14)

Conchoecia incisa Miiller, 1906 : 94, pi. 19, figs i-n ; Miiller, 1912 : 80 ; Angel, 1969 : 541 ;

Poulsen, 1969 : 163, 164 ; Deevey, 1974 : 368, 369.
Conchoecia gaussi ? (Miiller) Angel, ig68a : 308 (non Miiller, 1908).

DIAGNOSIS. Species of gaussi group characterized by carapace length (<$ 1-822-
1-911 mm, mean 1-877 ± 0-030 mm ; ? 2-462-2-837 mm, mean 2-64410-024 mm)
and tapering lateral outline of female ; male frontal organ long and slender with
long hairs, a seta of $ first antenna has lateral lobes near base and e seta bears
25-33 pairs of spines ; /and g setae of <$ second antennae are square ended ; cutting
edge of mandibular coxale lacks teeth.

MATERIAL. One hundred and twenty-four specimens comprising 31 females,
8 males and 85 juveniles. Dissected male and female specimens, stained with lignin
pink and mounted on slides in Euparal, are deposited in the British Museum (Natural
History) reg. nos 1974.751 and 752.

DESCRIPTION OF MALE. Carapace (Fig. n). The carapace lengths of eight
specimens range from 1-822 mm to 1-911 mm with a mean value of 1-877 ± 0-030 mm
(Fig. 10). In lateral view the posterior margin is gently curved and joins the ventral
margin at a slightly truncated posterioventral corner. The ventral margin is almost
straight and converges only slightly with the dorsal margin anteriorly. In ventral
view, the carapace outline does not curve evenly towards the posterior end but is
chisel shaped, the sides of the carapace in the posterior I or so being straight or even
slightly concave. There is no surface ornamentation. The right asymmetric gland
opens above £ of the maximum carapace height. The median ventral gland is
fairly well developed.

Frontal organ (Fig. I2D). The shaft extends beyond the first antenna, and the
capitulum is about f the length of the shaft. The capitulum is narrow, elongate and
often curves slightly forwards ; the distal | is rather expanded and the end is
asymmetrically rounded. The proximal f of the capitulum bears long, fine hairs
mainly ventrally.

First antenna (Figs 12 A- C). The first segment is slightly longer than the second
and the limb lacks armature. The a seta is rather less than twice the length of the

10
5

Numbers
s

Vl incisa

I ITh n • P •tJ=H33Tn n FSJI B^P^fa H R

Adult females
Adult males
Stage VI males
Other juvenile

150 2.00 2.50 •

D
gaussi

V VI

m rj m n M | ra ro I»RI ul ia ra

1.50

ZOO Z$0 MO

Length (mm)

150

FIG. 10. Histograms showing carapace lengths of the adults and last two juvenile instars
of Conchoecia incisa and Conchoecia gaussi.

CONCHOECIA FROM THE NORTHEAST ATLANTIC

79

Stage VI 9

to

Stage V

FIG. ii. Conchoecia incisa.

Lateral and ventral carapace outlines of adults and last two
juvenile instars.

limb ; proximally it loops down and up and then lies parallel to the antenna. Near
its base, the a seta is expanded and bears side lobes, usually three dorsal and one
lateral. The b, d and e setae are thin walled distal of their main armature. The
b seta is rather longer than the d seta and carries a number of anterior spines, 4-6
are widely spaced and 7-13 closely spaced and adjacent to an elongate narrow pad ;
on the thin- walled part of the seta below the pad are 8-10 smaller, closely spaced
pairs of spines followed by up to 36 minute spines, usually on the anterior side. The
c seta is short and wide with a basal horizontal stalk above which is a short dorsal
continuation. The d seta carries 4-18 small spines usually on each side. The
main e seta armature comprises 25-33 paired posterior spines and there may also
be a few small anterior spines above the main armature.

Second antenna (Figs I2E-H). The limb has no additional armature. The
processus mammillaris is conical with a terminal tubercle. The swimming setae
are distally thin walled and flaccid. The b seta is rather more than •£ the length of the
a seta, the c seta is the same length or rather shorter than the b seta and the d seta
is just under -| the length of the c seta. There are a few short hairs on the a and b
setae. The / and g setae are thin walled for approximately their distal £ and have
blunt ends, often with a filament. Both setae are bare or have 1-2 small median

H

FIG. 12. Male dimorphic parts of Conchoecia incisa ; arrows indicate anterior. (A) First
antenna and frontal organ. (B) Proximal part of a seta of first antenna showing lateral
lobes. (C) Main armature of (from left to right) e, b and d setae of first antenna. (D)
Capitulum of frontal organ. (E) Second antenna. (F) Second and third segments of
endopodite of left second antenna. (G) Second and third segments of endopodite of
right second antenna. (H) / and g setae of second antenna. (I) Penis. (J) Sixth limb.

FIG. 13. Non-dimorphic parts of Conchoecia incisa ; arrows indicate anterior. (A) From
top to bottom, cutting edge, distal and proximal tooth lists of mandibular coxale. (B)
Basale, endopodite and exopodite of mandible. (C) Cutting edge and adjacent setae of
mandibular basale. (D) Maxilla. (E) Caudal furca. (F) Labrum. (G) Fifth limb.

82 A. J. GOODAY

spines. The h and i setae bear numerous spines above the proximal constriction
while the j seta has rather longer spines above the constriction and a few shorter
spines below. The right hook appendage has a short proximal section and a long
distal section which increases in curvature towards the end where there are 9-10
subterminal ridges. The left hook appendage is smaller and has an almost straight
distal section with fewer subterminal ridges.

Mandible (Figs I3A-C). The cutting edge of the coxale lacks teeth and slopes
down posteriorly. On the distal tooth list is a large posterior tooth followed by (in
three specimens) 17, 19 and 22 smaller teeth of rather variable size. The proximal
list has a large posterior tooth and a rather smaller tooth before the mid-point ;
between these teeth the list slopes up and is either straight or toothed. The median
tooth is followed anteriorly by (in three specimens) 12, 15 and 18 small teeth of rather
variable sizes. The inner surface of the proximal list bears papillae. The cutting
edge of the basale has two posterior spine teeth, one long and pointed, the other
shorter and blunt, six serrated teeth and a conical inner tooth at the anterior end.
The basale has a small patch of hairs above the spine teeth and a larger area on the
ventral surface. There are also numerous hairs on the distal part of the third
exopodite segment and a small distoventral patch of hairs on the second segment.
The first exopodite segment has one long seta arising from the inner margin and three
shorter setae just inside the margin.

Maxilla (Fig. I3D). The anterior margin has five setae, which decrease in size
ventrally, and one shorter seta arising just inside the margin. The posterior margin
has three setae and the basal seta extends beyond the end of the limb. The five
setae on the final segment are long and slender. There are numerous minute spines
on the bottom of the first endopodite segment.

Fifth limb (Fig. i^G). The dorsal seta of the first exopodite segment is long and
extends to the final segment.

Sixth limb (Fig. 12 J). This is typical for the genus.

Labrum (Fig. isF). The hyaline membrane has four large outer teeth, 3-4
rounded teeth and 3-4 small inner teeth. The notch is rounded.

Furca (Fig. I3E). The caudal lamellae are covered with fine hairs.

Penis (Fig. 12!). The penis tapers distally. In two specimens there are six
transverse muscle bands and in a third specimen, seven bands.

DESCRIPTION OF THE FEMALE. Carapace (Fig. n). The carapace lengths of 39
specimens range from 2-462 mm to 2-837 mm with a mean value of 2-624 ± 0-062 mm
(Fig. 9). The posterior margin is curved and joins the ventral margin evenly at the
rather truncated posterioventral corner. The ventral margin is slightly concave
or straight and converges strongly with the dorsal margin anteriorly. There is no
surface ornamentation. The right asymmetric gland opens at about ^ of the
maximum shell height and the median ventral gland is fairly well developed.

Frontal organ (Fig. 146) . The shaft extends beyond the end of the first antenna
and the capitulum is about f the length of the shaft. The proximal f of the capitulum
is parallel-sided or very slightly tapered, the distal part is slightly expanded and the
end is asymmetrically rounded. There are prominent ventral spines on the proximal
f of the capitulum and a few similar proximal spines dorsally.

FIG. 14. Female dimorphic parts of Conchoecia incisa ; arrows indicate anterior. (A)
First antenna and frontal organ. (B) Capitulum of frontal organ. (C) e seta of first
antenna. (D) Second antenna. (E) g seta of second antenna. (F) Second and third
segments of endopodite of left second antenna. (G) Sixth limb.

84 A. J. GOODAY

First antenna (Figs I4A, C). The second segment is longer than the first and
both have a few small ventral spines. The dorsal seta is long and armed with short
hairs. The a-d setae are about as long as the first two limb segments combined
while the e seta is just under three times the length of a-d setae and its anterior side
bears 20-39 (usually 24-30) small, flat-lying hairs. On the posterior side are 16-17
small hairs, 30-35 stout spines, mostly in pairs, and 7-11 smaller spines on the
distal flattened and slightly widened part of the e seta.

Second antenna (Figs 140 -F). The protopodite has a patch of short hairs
behind the endopodite and the first exopodite segment has some small proximal
outer spines. The processus mammillaris, a and b setae are as for the male. The
/seta is rather longer than the h-j setae, with a number of proximal spines and 1-5
median and distal anterior spines. Proximally, the g seta also has a concentration
of small spines. On the anterior side, these are followed by 15-19 spines of which
5-9 are on the distal flattened and widened part of the seta ; the posterior side bears
7-9 similar spines of which only 1-2, if any, are on the flattened section. Both h
and i setae have numerous spines above and below the proximal constriction. The
j seta has rather longer spines, 7-8 arising from slight 'bumps' on the inside of the
seta, above the constriction.

Sixth limb (Fig. 140). The three setae on the final segment are slender and
claw-like. The dorsal c seta is longer than the b seta and rather less than twice the
length of the ventral a seta.

JUVENILES (Fig. n). Nine Stage V juveniles range from 1-280 mm to 1-379 mm
with a mean length of 1-30810-031 mm. Male and female Stage VI juveniles are
readily distinguishable ; the males are shorter (1-694-1-812 mm, mean length
1-773 ±0-03 mm in 13 specimens) than the females (1-793-2-009 mm, mean of
1-87810-05 mm in 25 specimens). The males only increase slightly in size at the
final moult while the size increase is much greater for females. Juvenile and adult
lengths of C. incisa are summarized in Fig. 10.

REMARKS. This form differs somewhat from Miiller's (1906) original description
of C. incisa which was based on two specimens from the Indian Ocean, (i) Miiller's
male is larger (2-15 mm) than the North Atlantic males (1-822-1-911 mm, mean
1-877 mm) and the carapace has a much more tapered lateral outline (Miiller : pi. XIV,
fig. i). On the other hand, Miiller's female specimen compares well with the
Atlantic form in size and outline, (ii) The original male specimen has 'wenig
auffallige' surface striations although the female shell lacks ornamentation (Miiller :
94). (iii) The a seta of the male first antenna extends only just beyond the main
joint of the limb (Miiller : pi. XIX, fig. 4) but at least to the base of the limb in the
North Atlantic form, (iv) The e seta of the female first antenna seems rather shorter
in Miiller's female (pi. XIX, fig. 3). (v) On the second antenna, only two of the
'tube setae' (h-j) have basal spines in Miiller's female (Miiller : 94) while all three
of these setae are spinose in Discovery females.

Apart from points (iv) and (v), which are relatively minor, Miiller's description
and figures of his female specimen agree well with the Discovery females particularly
in carapace length, outline and lack of surface ornamentation. The differences
between Miiller's male from the Indian Ocean and the North Atlantic males are more

CONCHOECIA FROM THE NORTHEAST ATLANTIC 85

pronounced. The degree to which the two forms differ in carapace length and outline
is particularly significant and this possibly warrants a taxonomic distinction between
them. They are not separated here because females from the North Atlantic
closely resemble Miiller's female.

Another possibility is that while the females in the Discovery material belong to
C. incisa, the males are of a distinct species. This would explain the considerable
size difference between the two sexes. However, it is unlikely because (i) the
non-dimorphic characters of males and females are closely similar and (ii) in such a
comprehensive series of samples, it is improbable that of two sympatric species, one
would be represented by males alone and the other by females. The conspecificity
of Miiller's male with his female and with the Discovery material should however be
regarded as somewhat questionable.

Poulsen's (igGga : 163, 164) brief description of three females of C. incisa from the
Atlantic near the equator agrees well with Discovery females although his later (1973)
account of this species from the Pacific, Indian and Atlantic Oceans differs in a few
respects. The proximal and distal tooth lists of the mandibular coxale both have a
large anterior tooth which is absent from Discovery specimens and the structure of
the labrum is rather different. These are probably only intraspecific variations.
Deevey (1974 : 368) records a single female from just south of the equator in the
Atlantic ; this specimen is only just shorter than the Discovery females.

The juveniles and females of C. incisa reported by Angel (1969 : 541) are included
in my material. Angel (i968a : 308) earlier tentatively identified two juveniles
caught above and within the thermocline at 34°N off Morocco as C. gaussi. On
re-examination, these proved to be Stage V juveniles of C. incisa.

GEOGRAPHICAL DISTRIBUTION. Miiller (1906, 1912), i$, i<$, Indian Ocean, o°
and 26°S ; Angel (i_968a : 308), North Atlantic, 34°N, 2 juveniles ; Angel (1969 : 541),
23$ and juveniles, North Atlantic off Canaries, 28°N ; Poulsen (1969 : 163, 164),
3$, Atlantic, just south of equator, about 2500 m ; Poulsen (1973), 103$, Pacific,
about 32° to 9°S, Indonesian Seas, about 4° to about 23°N, Indian Ocean, about 24°
to about 3°N, Atlantic, about 4° to about 38°N, most common at 1000-2500 m ;
Deevey (1974 : 368), i$, Atlantic Ocean, 3°4o'S, 3O°W, 200-0 m.

In the Northeast Atlantic this species was found at the following stations ; the
number of specimens at each station is given in parentheses : Station 7709, 6o°N,
20°W (2) ; Station 7406, 4o°N, 2O°W (3) ; Stations 8263 (2) and 8270 (i), both
32°N, 27°W; Stations 7803 (i), 7856 (37), both 3O°N, 23°W ; Stations 5814 (7),
5816 (i), 5818 (9), 5819 (2), 5825 (5), 6155 (12) 6156 (10), 6164 (14), 6176 (8), 6183 (i),
6193 (i) all 28°N, I4°W ; Station 7089, i8°N, 25|°W (10). The depth distribution
at different latitudes is summarized in Fig. 21.

Conchoecia gaussi Miiller 1908
(Figs 10, 15-18)

Conchoecia gaussi Muller, 1908 : 71, 72, pi. IX, figs 14-16, pi. X, figs 9-12 ; Miiller, 1912 : 80 ;

Skogsberg, 1920:677-681, figs CXXIX i-n ; Poulsen, 1969:164; Poulsen, 1973:106,

107, fig. 52 ; Deevey, 1974 : 369.
Non Conchoecia gaussi ? (Miiller) Angel, ig68a : 308 ( = C. incisa).

86

A. J. GOODAY

Stage V

FIG. 15. Conchoecia gaussi.

Lateral and ventral carapace outlines of adults and final
two juvenile instars.

DIAGNOSIS. Species of gaussi group characterized by large size (<$ 2-955-3-329 mm,
mean 3-13810-101 mm ; $ 3-369-3-546 mm, mean 3-462 mm) and subrectangular
lateral outline with truncated posterioventral corner ; a seta of <$ first antenna is
shorter than limb, b and d setae are only slighter shorter than e seta which bears
44-49 paired spines ; on $ second antenna, / and g setae both taper to whip-like
extremities and / seta is only slightly shorter than g seta.

MATERIAL. Forty-one specimens, comprising 5 females, 12 males and 24 juveniles.
Dissected male and female specimens, stained with lignin pink and mounted on slides
in Euparal, are deposited in the British Museum (Natural History) reg. nos 1974.753
and 754.

DESCRIPTION OF MALE. Carapace (Fig. 15) . The carapace lengths of 10 specimens
range from 2-955 to 3-329 mm with a mean value of 3-138 + 0-107 mm (Fig- 10). In
lateral view the carapace is subrectangular and the dorsal margin usually curves

H

FIG. 1 6. Male dimorphic parts of Conchoecia gaussi ; arrows indicate anterior. (A) First
antenna. (B) Capitulum of frontal organ. (C) Main armature of (from left to right)
d, e and b setae of first antenna. (D) Second antenna. (E) / and g setae of second
antenna. (F) Second and third segments of endopodite of right second antenna. (G)
Endopodite of left second antenna. (H) Sixth limb. (I) Penis.

88 A. J. GOODAY

upwards towards the posterior end. The posterior margin is vertical, slightly
curved in the upper part and truncated posterioventrally where it joins the ventral
margin evenly ; the ventral margin is straight or very slightly concave and parallel
to the dorsal margin. There is no surface ornamentation. The right asymmetric
gland opens just about J of the maximum shell height and is larger than the ventral
gland which discharges behind \ the carapace length.

Frontal organ (Fig. i6B). The shaft extends beyond the end of the first antenna.
The capitulum is strongly tapered proximally and almost parallel-sided distally,
usually with a distinct angle between the dorsal edges of the two sections.
ProximaUy, the capitulum has numerous strong closely spaced spines, the distal
section bears much smaller spines.

First antenna (Figs i6A, C). The second segment is longer than the first and
often carries numerous small anterioventral spines decreasing in size ventrally. The
a seta is shorter than the limb and somewhat expanded beyond the short horizontal
basal stalk. It S-bends before extending back to just past the main joint. The
b seta is usually rather shorter than the d seta and has a narrow elongate pad at
about f of its length. On its anterior side the b seta bears 17-22 spines, 8-12 are
closely spaced and below the pad and the remainder larger and more proximal,
those adjacent to the pad being particularly large and prominent. There may also
be some minute spines on the distal, thin-walled part of this seta. The c seta
resembles that of C. incisa. The d seta has 15-20 anterior pairs of spines which
increase in size distally and may be followed by minute spines like those on the b
seta. Both b and c setae are only slightly shorter than the e seta (Table 2). The
main e seta armature comprises 44-49 large paired spines, preceded by 14-23 small
anterior spines and followed by 4-5 long stiff spines on both sides of the bend below
the main armature.

Second antenna (Figs i6D-G). The protopodite has a patch of large hairs
behind the first endopodite segment. The first exopodite segment has numerous
very fine proximal spines on its anterior (outer) surface and the first endopodite
segment has similar spines on the process which carries the a and b setae. The
processus mammillaris is narrow and triangular. The a seta is 1-5-1 -6 times the
length of the b seta ; the c seta is slender and rather longer than the a seta, the d seta
is just over ^ the length of the c seta ; the e seta is a short proximally directed spine
less than -| the length of the d seta. The a-c setae bear short hairs. The /seta is
somewhat shorter than the g seta and thin walled for f or so of its length. In the
median part it is widened slightly and then tapers distally to a point ; the / seta is
bare or has 1-2 anterior spines in the proximal ^. The g seta is thin walled and
widened into a lancet shape in the distal ^. On the posterior side are 9-10 small
proximal spines and the anterior side bears 33-38 spines of which 13-16 lie on the
proximal tubular part and 20-24 on the distal widened part of the seta. The h
and i setae have scattered spines above the proximal constriction and sometimes a
few spines below ; the /seta has numerous rather larger spines above the constriction
and scattered spines beyond this point. The proximal section of the right hook
appendage is more than | the length of the gently curved distal section ; the angle
between the two sections is less than 90° and there are 8-9 subterminal ridges. The

FIG. 17. Non-dimorphic parts of Conchoecia gaussi ; arrows indicate anterior. (A) From
top to bottom, cutting edges, distal and proximal tooth lists of two specimens. (B)
Basale, endopodite and exopodite of mandible. (C) Cutting edge and adjacent setae
of mandibular basale. (D) Maxilla. (E) Caudal furca. (F) Labrum. (G) Fifth limb.

go A. J. GOODAY

left hook appendage has a basal angle of about 45° and a long straight distal
section.

Mandible (Figs lyA-C). The cutting edge of the coxale carries 17-18 teeth,
decreasing in size posteriorly, and is straight near both ends. The proximal tooth
list has a large pointed posterior tooth, followed by 4-5 smaller teeth, a second large
pointed tooth and 10-12 smaller anterior teeth of rather variable size. The distal
list has a rather rounded posterior tooth and 19-21 smaller teeth also of variable
size. The cutting edge of the basale has the usual structure ; the inner tooth is
broad and raised into a small central point. The basale has an area of hairs extending
from the posterior margin above the spine teeth as far as the broad inner tooth.
There are also numerous hairs above the ventral margin of the basale, on the distal
part of the third exopodite segment and a few hairs on the outer distal edge of the
second segment. The first exopodite segment bears one long marginal inner seta
and three smaller setae arising inside the margin and decreasing in size proximally.

Maxilla (Fig. 170). The first endopodite segment bears five anterior marginal
setae, one seta just inside the anterior margin, three posterior marginal setae and a
lateral seta. The basal seta extends to near or just beyond the end of the limb.

Fifth limb (Fig. 170). As for C. incisa.

Sixth limb (Fig. i6H). As for genus.

Labrum (Fig. I7F). The hyaline plate bears 9-10 prominent teeth and the notch
is V-shaped and fairly deep.

Caudal furca (Fig. I7E). The caudal lamellae are covered with fine hairs.

Penis (Fig. i6I). Slender and distally tapered with six transverse muscle bands.

DESCRIPTION OF FEMALE. Carapace (Fig. 15). The carapace lengths of four
specimens range from 3-369 mm to 3-546 mm with a mean value of 3-462 mm (Fig.
10). Gland positions and lateral outline are as for the male except that the outline
is rather more elongated and tapers slightly towards the anterior end.

Frontal organ (Fig. i8C). The shaft extends beyond the first antenna. The
capitulum is elongate and tapers anteriorly, the proximal ^ is rather expanded.
Along the proximal f to f of the ventral surface are numerous strong spines ; there
are similar spines on the proximal \ to | of the dorsal surface but only a few scattered
lateral spines. Distally the capitulum bears much smaller spines.

First antenna (Figs i8A, B). The first segment has a few small distal spines on
the ventral surface and is shorter than the second segment which carries numerous
small ventral spines. The dorsal seta is as long as the first two limb segments. The
e seta is less than twice the length of the a-d setae and is flattened and widened
distally. It bears numerous spines which start on both sides near the base and are
most concentrated and prominent near the middle of the seta. On the anterior
surface are 80-100 spines which do not extend far onto the distal section of the seta,
while posteriorly there are at least 130-140 spines extending onto the widened part.
The spines are arranged in at least two rows for much of their extent.

Second antenna (Figs i8D-G). The a and b setae and the armature of the proto-
podite, the first exopodite and endopodite segments are as for the male. The / seta
is rather more than f the length of the g seta and bears fairly numerous basal anterior
spines. The g seta has 7-10 rather larger basal, anterior spines followed by 28-40

FIG. 1 8. Female dimorphic parts of Conchoecia gaussi ; arrows indicate anterior. (A)
First antenna and frontal organ. (B) e seta of first antenna. (C) Capitulum of frontal
organ. (D) Second antenna. (E) Armature on proximal parts of /-/ setae of second
antenna. (F) Endopodite of right second antenna. (G) g seta of second antennae.
(H) Sixth limb.

92 A. J. GOODAY

anterior spines of which 12-18 lie on the distal flattened part ; there are no posterior
spines on the g seta. The h and i setae are rather longer than the / seta and have
scattered spines above and below the proximal constriction. The j seta has rather
more spines above and below the constriction.

Sixth limb (Fig. i8H). The three setae on the final segment are slender and
claw-like. The dorsal c seta is slightly longer than the b seta and the ventral a seta
is about | the length of the b seta.

JUVENILES (Fig. 15). Five Stage V juveniles range in length from 1-753 mm to
1-931 mm (mean 1-836 mm). Seven female Stage VI instars measure 2-502 mm to
2-719 mm long (mean 2-620 mm) and three male Stage VI instars have lengths of
2-778 mm, 2-541 mm and 2-502 mm. The Stage V and VI instars of C. gaussi are
thus the same size as Stage VI juvenile females and adult females respectively of
C. incisa. However, juvenile instars of the two species can be distinguished by their
characteristic and differing outlines. Adult and juvenile lengths for C. gaussi are
summarized in a histogram (Fig. 10).

REMARKS. The Discovery material is closely similar to the male described by
Miiller (1908 : 71-72) and the female of Skogsberg (1920 : 677-681). There are only
a few differences, (i) The posterioventral corner of Miiller's specimen (pi. X, fig. 9)
is less truncated and more rightangular than in North Atlantic males, (ii) The
carapace of Miiller's specimen bears dorsal and anterior striations. (iii) There are
more teeth on the cutting edge of the mandibular coxa figured by Skogsberg
(fig. CXXIX 6) and the teeth have a greater size range than in the Discovery
specimens, (iv) Skogsberg noted that the joints of the female first antenna are
'smooth' and so presumably lack the small spines present in the North Atlantic
material. These differences are minor and the Discovery material is undoubtedly
conspecific with that of Miiller and Skogsberg.

Rudyakov (1962) has described the female of C. gaussi curilensis, a new subspecies
from the Northwest Pacific, characterized mainly by its large size (4-08-4-18 mm).
Two females reported by Poulsen (19693. : 164) from the South Atlantic are almost
comparable in size (4-0 mm) to Rudyakov's subspecies and Poulsen's male specimen
is also larger (3-30 mm) than Discovery males. Size seems to be an important
taxonomic character in halocyprids, so this large form may be a new species rather
than just a subspecies. It is also conceivable (although unlikely) that it belongs to
a second adult instar of C. gaussi similar to the large specimens of Gigantocypris
agassizi Miiller reported by Poulsen (1962 : 34 and Table 4, p. 54 ; see also Angel,
1970 : 164).

Poulsen (1973 : 106) described a male and two females of C. gaussi from the Pacific
which agree well in size and other characteristics with the North Atlantic material.
Recently Deevey (1974 : 369) has reported two female specimens from the South
Atlantic one of which is comparable in size to Discovery females (3-45 mm) and the
other rather larger (3-85 mm). As noted above the two juveniles of C. gaussi ?
reported by Angel (ig68a : 308) are juveniles of C. incisa.

Apart from its larger size and dissimilar carapace outline, C. gaussi differs from
C. incisa as follows, (i) The capitulum of the male frontal organ is shorter, broader

CONCHOECIA FROM THE NORTHEAST ATLANTIC 93

and a different shape in C. gaussi and is armed with spines rather than long hairs as
in C. incisa. (ii) The a seta of the male first antenna is shorter than the limb in
C. gaussi but longer than the limb in C. incisa (Table 2), where it also bears lateral
lobes near the base, (iii) The b and d setae of this limb are proportionally much
longer in C. gaussi and are almost as long as the e seta (Table 2). (iv) The e seta of
the female first antenna is proportionally longer in C. gaussi (Table 3). (v) The first
exopodite segment of the male and female second antennae and the protopodite of
the female second antenna are proportionally longer in C. gaussi (Tables 2 and 3).
(vi) The swimming setae of the male second antenna are proportionally longer in
C. incisa (Table 2) and the distal flattened part is thinner walled and more flaccid
than that of C. incisa. (vii) In C. gaussi the /and g setae of this limb both taper to
whip-like extremities while those of C. incisa are square-ended. Also the g seta is
proportionally much shorter in C. gaussi and only slightly longer than the / seta,
(viii) The c seta of this limb is more slender and proportionally longer in C. gaussi.
(ix) The h-j setae of the male and female second antennae are proportionally much
shorter in C. incisa (Table 2). (x) In C. gaussi, the cutting edge of the mandibular
coxale is toothed, but in C. incisa it lacks teeth.

The Antarctic species C. hettacra Miiller, 1906 is remarkably similar in outline to
C. gaussi, particularly in the male. However, this species is smaller, the female has
a rather more tapered outline and the prominent right asymmetric gland opens
much further down, on the posterioventral corner.

GEOGRAPHICAL DISTRIBUTION. Miiller (1908 : 72), ij, South Atlantic, 35°S, 2°E ;
Skogsberg (1920:681), i$, South Atlantic 5o°S, 50°W, 2700-0 m ; Poulsen
(1969 : 164), 2$, i(?, Atlantic, 6°35'S (725 m) and 2°S (2500 m) ; Poulsen (1973 : 106),
1$, 2$, near New Guinea, about 9°S, off New Zealand, about 35°S ; Deevey (1974 :
369), 2$, South Atlantic, 32°3o'S, 750 m wire, 5i°3o'S, 3000 m wire.

In the Northeast Atlantic C. gaussi was found at the following stations ; the
number of specimens at each station is given in parentheses : Station 7709, 6o°N,
20°W (8) ; Station 7711, 53°N, 2O°W (24) ; Stations 7482 (i), 7406 (4), both 4O°N,
20°W ; Station 7856, 30°N, 23°W (2) ; Stations 7089, i8°N, 25°W (3) ; Station 6674,
u°N, 2i|°W (i). The depth distribution at different latitudes is summarized in
Fig. 22.

ECOLOGY

The vertical distributions of the four species described here at different latitudes
in the Northeast Atlantic are shown in Figs 19-22. These sections include only
data from stations where the top 2000 m of the water column was comprehensively
sampled with the RMT 1 + 8. Angel & Fasham (1975) have analysed associations
of ostracod species in this region.

Conchoecia subedentata sp. n. (Fig. 19) is a high-latitude species which is more
abundant at the most northerly station sampled (6o°N). It becomes steadily less
common southwards and is very rare at i8°N and absent at n°N. This species
was caught mainly between 300 m and 900 m but was never present in the top 200 m
and only occasionally taken below 100 m. There is a slight tendency for C.

94

A. J. GOODAY

subedentata sp. n. to live deeper further south. C. afL edentata has a similar geographic
distribution and is also commonest at 6o°N, although there is a secondary peak of
abundance at i8°N (Fig. 20). This species was usually caught at greater depths
than C. subedentata sp. n. It was never present above 400 m and occurred most
frequently below 800 m. In neither species of the edentata group is there any
evidence for diurnal vertical migrations.

The two species of the gaussi group tend to occur further south ; C. gaussi is
most abundant at 53°N, and C. incisa at 3O°N (Figs 21, 22). The vertical distribution
of C. incisa is rather anomalous. Males are rare and none was caught above 1250 m.
The females and juveniles tend to be rather shallower (below 800 m) but may some-
times be taken quite near the surface. This was particularly true at 28°N where

60°N

DEPTH

(m)

u-

•

DAY

i:

3'
U

-

<

i

75
2C
163

I']
3:

'
(

,

>2

NIGHT DAY

1

45

10J

K -1

>3 3S
'7 10

« 4J

?4 95
1 4d
2 1SJ

\

fl 15
^3 1<3

n 73

•1 49'

ri 53°

173

~

3
2C

NIGHT DAY

I
1

j
.

2 <
80
3 0
•
( 2«<
1 44C
903
h
S 9(
• 8C
2 7J

1
6
•6
13

'I

•
'1
2

|
1

NIGHT
DAY

m

rs

33

NIGHT

i

•

72

> 15

'16 1°
5 2J
1
?» 135

l"32 50
30 58 J
k
?4 40?
i"10 33C
7 1150
\
2<
4
,
.

21 <
M 19C
180

(

II

'3
3

>
!9
•18
82
•
?16
r29
16
•
?43

r74

80

i

!
i

•
•
?i

n

49

500-

H

448

323

285

91

131

7

lUUU

1500

?onn

4

:
4

;

s I

r2

1

1?

1

?3

>4

H
M

6

9
5

3 63 245 408 708

FIG. 19. Conchoecia subedentata sp. n. Depth distribution at various latitudes in the
Northeast Atlantic Ocean. The number of males, females and juveniles caught with
the RMT 1 + 8 during day and night in 100 m horizons down to 1000 m and wider hori-
zons between 1000 m and 2000 m are shown. The figures along the bottom of the dia-
gram are the total numbers of specimens occurring at each latitude and the figures in
the right-hand column give the total numbers of specimens caught in each depth horizon.

CONCHOECIA FROM THE NORTHEAST ATLANTIC

95

11

18

30°

40

53

60° N

u

500-

DEPTH

(m)

1000-

icnn

DAY

«
7:

j

NIGHT DAY

i
I

NIGHT DAY
1

NIGHT DAY

'«

1S

I

1C

U

!<

1

?

; f

3 3.

4!
1C

•

1:

NIGHT
DAY

!2
[
' 2<

;

K
n

•

. ^

6d
20 J

;

•

I-

r

NIGHT

!

F

1

-

|

'

15

<

1C
1C

«<
2.

* .
• •

IS

?

'

?' i,
rl

•

1

' *

J
\

e

4

5

3

16

10

43

5

IDUU

2000

1

2 23 1 18 21 37

FIG. 20. Conchoecia ail. edentata. Depth distribution at various latitudes in the Northeast
Atlantic Ocean. For explanation see Fig. 19.

TABLE 6

Vertical distribution during the day and night of females and juveniles of Conchoecia incisa

at 28°N

DEPTH (m)

o-ioo
100-200
200-300
300-400
400-500
500-600
600-700
700-800
800-900

DAY

NIGHT

FEMALES JUVENILES FEMALES JUVENILES

i 3

i i 7

96 A. J. GOODAY

most females and juveniles were caught between 600 m and the surface with a
concentration of juveniles at 500-600 m (Table 4 : these data were obtained from
Nii3 catches and are not comparable with the data in Fig. 21 which are based on
RMT 1 + 8 samples) . This vertical distribution pattern resembles that of Halocypris
globosa Claus, 1874. Here the males are deep-living while the females and juveniles
have a bimodal distribution with peaks of abundance between 1500 m and 3000 m
and in the top 100 m (Poulsen, i96gb : 49, 60). Almost all specimens of C. incisa
caught near the surface (0-200 m) were taken at night (Table 4 and Fig. 21) which
suggests that shallower members of this species migrate diurnally. However, there
is no evidence for diurnal migrations by individuals living below 1000 m (Fig. 21).

C. gaussi occurs mainly below 1000 m although a few specimens were caught at
shallower depths and a juvenile was obtained from the top 100 m at 53°N (Fig. 22).
This species does not seem to migrate vertically although the evidence is sparse.

DEPTH

(m)

u-

DAY

NIGHT CAY 1, NIGHT DAY

NIGHT

NIGHT

-

V

j

rl

2

500-

H

i

2>

3

i<

1!

'

2

2S

r

2

I

—

1? 91

J3

7

3J1

4

1000-

—

<9

6

7

ii1

11

273

1500-

+

—

293

4<?3

'2

12

2000-

10 37 3 2

FIG. 21. Conchoecia incisa. Depth distribution at various latitudes in the Northeast
Atlantic Ocean. For explanation see Fig. 19.

CONCHOECIA FROM THE NORTHEAST ATLANTIC

11° 18T 30° 40° 53° 60° N

97

DEPTH

(m)

DAY

DAY

DAY

NIGHT DAY
i:

NIGHT
DAY

1

1C

2J

•
r

3

500-

H

c

:

M
1

3

1C

•

1

1000-

—

J1

I

1

1

r

I
152

12

1500-
2000-

-

1C

-

f

2 91 ,<
1 0*5 ,
2J1

!
•

21

32 4 24 8

FIG. 22. Conchoecia gaussi. Depth distribution of various latitudes in the Northeast
Atlantic Ocean. For explanation see Fig. 19.

It is notable that most species of Conchoecia occurring in the Arctic have Antarctic
counterparts (C. borealis Sars, 1866, C. antipoda Miiller, 1906, C. obtusata Sars, 1866,
C. elegans Sars, 1866). Only C. maxima Brady & Norman, 1896 is endemic to the
Arctic region although this species is closely related to C. borealis 1C. antipoda. On
the other hand, several species appear to be endemic to the Antarctic area (C. belgica
Miiller, 1906, C. chuni Miiller, 1906, C. hettacra Miiller, 1906, C. isocheira Miiller, 1906,
C. schwabenlandica Hartmann, 1962, C. serrulata Claus, 1874). Should C. aff.
edentata prove to be conspecific with Miiller 's C. edentata, then this will be another
species having a bipolar distribution.

C. subedentata sp. n. and C. aff. edentata are to some extent sympatric and this
raises the question of how competition between these two closely related species is
reduced far enough to allow their coexistence. The mandibular structure is virtually
identical in both species and so they presumably eat the same sort of food. The
main factor may thus be the larger body size of C. aff. edentata allowing it to eat

98 A. J. GOODAY

larger food particles than C. subedentata sp. n. (Angel pers. comm.). This may also
apply to juvenile instars since the lengths of the few measured C. aff. edentata
juveniles tend to fall between the peaks on the length histogram for adult and
juvenile stages of C. subedentata sp. n. (Fig. i) ; however, more measurements are
needed to confirm this pattern. A similar situation is well known in calanoid
copepods where the coexistence of closely related sympatric species has been
attributed in part to size differences (for example, Sandercock, 1967 ; Hutchinson,
1967 : 680-684 '< Hammer & Sawchyn, 1968). Competition must also be reduced
by the partial depth separation of the two species (Figs 19, 20).

There is a different pattern of size distribution in the gaussi group. Here, adults
and juveniles of C. gaussi and C. incisa do not interdigitate but fall into three or
four fairly distinct size classes (Fig. 10) ; (i) Stage V juveniles of C. incisa ; (ii)
adult males and Stage VI juveniles of C. incisa and Stage V instars of C. gaussi ;
(iu) adult females of C. incisa and Stage VI juveniles of C. gaussi ; (iv) adult males
and females of C. gaussi may be regarded as forming a fourth group although the
size range here is greater and the group poorly defined. As in the edentata group,
competition between the size groups is probably reduced by the difference in size
of the food particles that members of each group are able to eat. However, groups
(ii) and (iii) include both juveniles of C. gaussi and adults and juveniles of C. incisa
and since these species coexist to some extent (Figs 21, 22) some factor other than
size must reduce competition between them. In C. incisa the cutting edge of the
mandibular coxale is straight while in C. gaussi it is toothed, hence a difference in
diet reflected by the dissimilar mandibular structure seems the most likely way that
competition is minimized. Again an analogy may be drawn with freshwater
calanoid copepods where Maly & Maly (1974) found that two closely related
sympatric species with different mandibular structure had dissimilar diets.

From the above discussion three possible methods of avoiding or reducing
competition between closely related sympatric species of Conchoecia may be
suggested, (i) Depth separation, as is partially the case with C. subedentata sp. n.
and C. aff. edentata. (ii) Differences in the size of food particle eaten as reflected by
differences in body size ; here one would expect the adult and juvenile size ranges
of closely related species to interdigitate as apparently happens in the edentata
group, (iii) Differences in diet reflected by dissimilar mandibular structure ; this
would allow closely related species of similar size to coexist as with adults and
juveniles of C. incisa and juveniles of C. gaussi.

ADDENDUM

After this paper was submitted, Dr G. B. Deevey kindly sent me two females of
C. edentata from Stations 18 and 25 of the Hudson 70 Expedition (Deevey 1974).
The specimens are mounted on slides and before dissection measured 1-70 mm and
1-77 mm long. The relative lengths of the setae on the final segment of the sixth
limb and the single seta near the spine teeth on the mandibular basale suggest that
Deevey's specimens are probably my C. aff. edentata. Dr Deevey also sent me some
material identified as C. edentata from Station S off Bermuda. It comprises some

CONCHOECIA FROM THE NORTHEAST ATLANTIC 99

undissected specimens in formalin and four mounted males and females which were
collected in 1969 and 1970 and so not included in Deevey's (1968) paper. These
specimens clearly belong to C. subedentata and the occurrence of this species near
Bermuda is confirmed by its presence at Discovery station 8281.

Dr Angel has restudied Fowler's (1909) material in the British Museum (Natural
History). He reports (pers. comm.) that Fowler's (1909 : 255, pi. 25, figs 258-262)
female specimen of Conchoecia C belongs in C. subedentata sp. n.

ACKNOWLEDGEMENTS

I gratefully acknowledge the help of Dr M. V. Angel who introduced me to this
project and critically read the manuscript and suggested a number of improvements.
The work was supported by a N.E.R.C. Research Fellowship.

REFERENCES

ANGEL, M. V. i968a. The thermocline as an ecological boundary. Sarsia 34 : 299-312.

i968b. Bioluminescence in planktonic halocyprid ostracods. /. mar. biol. Ass. U.K.

48 : 255-257.

1969. Planktonic ostracods from the Canary Island region ; their depth distribution,

diurnal migrations and community organisation. /. mar. biol. Ass. U.K. 49 : 515-553.

1970. The redescription of Conchoecia bispinosa Claus, C. haddoni Brady & Norman and

C. secernenda Vavra from the North Atlantic. Crustaceana 18 (2) : 147-166.

& FASHAM, M. 1975. Analysis of the vertical and geographical distribution of the

abundant species of planktonic ostracods in the north-east Atlantic. /. mar. biol. Ass.
U.K. 55 : 739-757-

BAKER, A. DE C., CLARKE, M. R. & HARRIS, M. J. 1973. The N.I.O. combination net (RMT

1 + 8) and further developments of rectangular midwater trawls. /. mar. biol. Ass. U.K.

53 : 167-184.
BARNEY, R. W. 1921. Crustacea, Pt. V — Ostracoda. British Antarctic ("Terra Nova")

Expedition, igio, Zoology 3 (7) : 175-190.
DEEVEY, G. B. 1974. Planktonic ostracods collected on the Hudson 70 between the Equator

and 55°S in the Atlantic. Proc. biol. Soc. Wash. 87 (2) : 351-380.
FOWLER, G. H. 1909. Biscayan plankton Pt. XII. The Ostracoda. Trans. Linn. Soc.

Lond. Zool. ser. 2, 10 : 219-358.
FOXTON, P. 1969. SOND Cruise 1965 : biological sampling methods and procedures. /. mar.

biol. Ass. U.K. 49 : 603-620.
GRANATA, L. & CAPORIACCO, L. DI 1949. Ostracodes marins recueillis pendant les croisieres

du Prince Albert ler de Monaco. Res. Camp. sci. Monaco 109 : 1-51.
HAMMER, U. T. & SAWCHYN, W. W. 1968. Seasonal succession and congeneric associations

of Diaptomus spp. in Saskatchewan ponds. Limnol. Oceanogr. 13 : 476-484.
HILLMAN, N. S. 1967. Ecology of Antarctic pelagic Ostracoda. Ant. Journ. U.S. 2 (5) : 199-

200.

1969. Halocypridae. In: Hedgpeth, J. W. (ed.). Distribution of selected groups of

marine invertebrates in waters south of 35°S latitude. Antarctic Map Folio Ser. 11 : 29-
30, pis 15-16.

HUTCHINSON, G. E. 1967. A Treatise on Limnology, vol. 2, New York.

ILES, E. J. 1953. A preliminary report on the Ostracoda of the Benguela Current. 'Dis-
covery' Rep. 57 : 259-280.

MALY, E. J. & MALY, M. P. 1974. Dietary differences between two co-occurring calanoid
copepod species. Oecologia 17 : 325-333.

ioo A. J. GOODAY

MULLER, G. W. 1906. Ostracoda. Wiss. Ergebn. Deutsch. Tiefsee-Exped. 8 : 29-154.

- 1908. Die Ostracoden der Deutschen Siidpolar Expedition 1901-1903. Siidpolar Exped.
10 : 51-182.

- 1912. Ostracoden. Das Tierreich 31 : 1-434.

POULSEN, E. M. 1962. Ostracoda-Myodocopa, i. Cypridiniformes - Cypridinidae. Dana
Rep. 57 : 1-414.

I969a. Ostracoda-Myodocopa from the eastern tropical Atlantic. Viddensk. Meddr.

dansk naturh. Foren. 132 : 129-197.

19690. Ostracoda-Myodocopa, 3a. Halocypriformes - Thaumatocypridae and Halo-

cypridae. Dana Rep. 75 : i-ioo.

I973- Ostracoda-Myodocopa, 30. Halocypriformes - Halocypridae, Conchoecinae.

Dana Rep. 84 : 1-223.

RUDYAKOV, Yu. A. 1962. Ostracoda Myodocopida (family Halocypridae) from the N.W.

Pacific Ocean. Trudy Inst. Okeanol. 58 : 172-201.
SANDERCOCK, G. A. 1967. A study of selected mechanisms for coexistence of Diaptomus

species in Clarke Lake, Ontario. Limnol. Oceanol. 12 : 97-112.
SKOGSBERG, T. 1920. Studies on marine ostracodes, i. Cypridinids, halocyprids and poly-

copids. Uppsala Univ. Zool. Bidrag Suppl.-Bd 1 : 1-784.

A. J. GOODAY

INSTITUTE OF OCEANOGRAPHIC SCIENCES

WORMLEY

GODALMING

SURREY

A LIST OF SUPPLEMENTS
TO THE ZOOLOGICAL SERIES

OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

1. KAY, E. ALISON. Marine Molluscs in the Cuming Collection British Museum
(Natural History) described by William Harper Pease. Pp. 96 ; 14 Plates.
1965. (Out of Print.)

2. WHITEHEAD, P. J. P. The Clupeoid Fishes described by Lacepede, Cuvier and
Valenciennes. Pp. 180 ; n Plates, 15 Text-figures. 1967. £4.80.

3. TAYLOR, J. D., KENNEDY, W. J. & HALL, A. The Shell Structure and Mineralogy
of the Bivalvia. Introduction. Nuculacea-Trigonacea. Pp. 125 ; 29 Plates,
77 Text-figures. 1969. £5.40.

4. HAYNES, J. R. Cardigan Bay Recent Foraminifera (Cruises of the R.V. Antur)
1962-1964. Pp. 245 ; 33 Plates, 47 Text-figures. 1973. £11.90.

5. WHITEHEAD, P. J. P. The Clupeoid Fishes of the Guianas. Pp. 227 ;
72 Text-figures. 1973. £10.65.

6. GREENWOOD, P. H. The Cichlid Fishes of Lake Victoria, East Africa : the
Biology and Evolution of a Species Flock. Pp. 134 ; i Plate, 77 Text-figures.
1974. £3-75. Hardback edition £6.

Printed in Great Britain by John Wright and Sons Ltd. at The Stonebridge Press, Bristol 884 jNU

- xv- -«£,•>,

•» swau*
24SEPW76

ECHINODERMS FROM THE X
NORTHERN REGION OF THE
GREAT BARRIER REEF, AUSTRALIA

P. E. GIBBS,
A.M. CLARK AND C.M.CLARK

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. 4

LONDON : 1976

ECHINODERMS FROM THE

NORTHERN REGION OF THE

GREAT BARRIER REEF, AUSTRALIA

BY

PETER E. GIBBS

Marine Biological Association of the U.K.

AILSA M. CLARK AND CHRISTINE M. CLARK

British Museum (Natural History)

Pp 101-144 ; i Plate ; 3 Text-figures ; I Map

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30, No. 4

LONDON: 1976

THE BULLETIN OF THE BRITISH MUSEUM

(NATURAL HISTORY), instituted in 1949, is
issued in five series corresponding to the Scientific
Departments of the Museum, and an Historical series.

Parts will appear at irregular intervals as they
become ready. Volumes will contain about three of
four hundred pages, and will not necessarily be
completed within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 30 No. 4 of the Zoology series.
The abbreviated titles of periodicals cited follow those
of the World List of Scientific Periodicals.

World List abbreviation
Bull. Br. Mus. nat. Hist. (Zool.)

ISSN 0007-1498

Trustees of the British Museum (Natural History), 1976

BRITISH MUSEUM (NATURAL HISTORY)

Issued 23 September 1976 Price £3-35

ECHINODERMS FROM THE

NORTHERN REGION OF THE

GREAT BARRIER REEF, AUSTRALIA

By P. E. GIBBS, AILSA M. CLARK & CHRISTINE M. CLARK

SYNOPSIS

This paper gives records of about 140 species of echinoderms from the northern region of the
Great Barrier Reef collected during the Royal Society-Universities of Queensland Expedition
in 1973. One ophiuroid species is new to science, namely Amphiura phrixocantha sp. nov.
(p. 121). Ten other ophiuroids, two echinoids and one holothurian are new to the Australian
fauna and the ranges of eight more ophiuroids and one holothurian are extended to the Great
Barrier Reef Province from other parts of Australia. In addition, a few unpublished records from
relevant material in the British Museum collections are included and notes are given on several
species of especial interest. In contrast to most shallow-water tropical collections, the one from
this Expedition includes a high proportion of burrowing species from soft substrates. A list of
some animal associates of the specimens collected is appended.

INTRODUCTION

EARLY investigations of the echinoderm fauna of the Great Barrier Reef were largely
centred on the Torres Strait region and essentially began with the visit of the
Challenger Expedition in 1874. The material collected, mostly from dredging,
together with that taken by the Alert Expedition in 1881, provided records for a
basic list of about 80 species. Subsequently, two land-based expeditions, firstly
that of R. Semon in 1892 and later that of the Carnegie Institution of Washington
in 1913, increased the number of species known in the area to about 250 (H. L. Clark,
1915, 1921).

South of Torres Strait along the east coast of Queensland early records of
echinoderms were provided by the observations and collections of Tenison- Woods
(i88oa, b) made in the Port Douglas -Trinity Bay area and by the Alert Expedition,
collecting particularly from Port Curtis, Port Molle and Port Denison (see Bell, 1884).
The next substantial collection was not made until 1924 when W. E. J. Paradice
took about 60 species, chiefly between the latitudes of 17° and I9°S (H. L. Clark,
1926 ; McNeill & Livingstone, 1926). This collection also included specimens from
Low Isles (i6°23'S), the echinoderm fauna of which is now well known as a result of
the surveys by the Great Barrier Reef Expedition in 1928-29 (H. L. Clark, 1932 ;
Livingstone, 1932) and the Great Barrier Reef Committee in 1954 (Endean, 1956).
In recent years, further echinoderm records (excluding crinoids) have been compiled
by Endean (1953, 1961, 1965), while crinoids from the Swain Reefs Expedition of
1962 have been dealt with by A. M. Clark (1975).

The Great Barrier Reef as a zoogeographical region has been discussed in detail
by such authors as Whitley (1932), H. L. Clark (1946) and Endean, Kenny &
Stephenson (1956). In the present paper, the Great Barrier Reef Province is taken

104 GIBBS, CLARK & CLARK

to include the sea area east of longitude I42°E (just west of Torres Strait) down to a
depth of 200 m, from Anchor and Bramble Cays (between 9°oo'S and g°2o'S) in the
north to Fraser Island including Hervey Bay (effectively the latitude of 25°S) in the
south. Echinoderm records from the Queensland coast are summarized by H. L.
Clark (1946) and, except for crinoids, by Endean (1957) with later additions also by
Endean (1961, 1965). Within the geographical limits denned above it appears that
about 310 species of echinoderms have been recorded from shallow- water ; however,
many of these species are known from relatively few localities and/or specimens and
records are particularly sparse for the northern part of the Great Barrier Reef
outside the Torres Strait region. Accordingly a list of the echinoderms collected in
this area during the 1973 expedition should be of some zoogeographic value,
especially as very little systematic dredging or trawling has previously been under-
taken in the region as a whole (Dall & Stephenson, 1953).

The echinoderm records given in the present paper result from the material
collected during the Royal Society- Universities of Queensland Expedition to the
Great Barrier Reef which took place from July to November 1 973 . The investigations
of Phases I and II of the Expedition were mostly concerned with the structure and
ecology of islands and reefs between Cairns and Pipon Island (approximately I7°S
to I4°S) but Phase III operations extended northwards to Raine Island (u°36'S).

Most of the records given below, particularly those of ophiuroids, originate from
a benthic survey carried out using naturalist and anchor dredges (Forster, 1953)
and involving bottom sampling at 62 stations in depths of 4- 64m between the
latitudes of i6°42'S to n°4o'S. Station data are given in Table i.

TABLE i

The positions of the stations at which dredge samples were taken in the northern part of the

Great Barrier Reef, with depth, description of deposit and date of collection also given

STATION LATITUDE LONGITUDE DEPTH DEPOSIT DATE

(S) (E) (m) (1973)

Di i4°25-2' I44°47'9' 15 Muddy sand with many forams 5 Aug.

D2 I4°25'i' i44°48-6' 15 Muddy sand with many forams 5 Aug.

D3 I4°24'9' i44°5o-i' 16 Muddy sand with many forams 6 Aug.

04 i4°27-i' i44°53-8' 20 Muddy sand with many forams 6 Aug.

05 i4°28-7' I44°55'7' 18 Muddy sand with many forams 6 Aug.
D6 i4°29'2' I44°57'8' 26 Muddy sand with many forams 6 Aug.
D7 i4°28-8' i45°o-2' 26 Muddy sand with many forams 7 Aug.
D8 i4°28-o' I45°o-o/ 26 Muddy sand with many forams 7 Aug.
Dg i4°26-7' i44°58-9' 37 Shell gravel 7 Aug.
Dio i4°29-2' i44°57-i' 20 Shell gravel 7 Aug.
Dn I4°2i-i' I44°53'6' 22 Muddy sand with many forams 8 Aug.
Di2 I4°i7'9' i44°53-4' 24 Muddy sand with shells 8 Aug.
Di3 i4°i4-8' i44°53-2' 22 Shell gravel 8 Aug.
014 I4°i5'8' i4405i-i' 33 Muddy fine sand with shells 8 Aug.
Di5 i4°22-6' i44°5o-2' 16 Muddy sand with forams 8 Aug.
Di6 i4°32-2' i44°53'4' n Muddy sand with shells 9 Aug.
Di7 I4°34'2' i44°52-2' 4 Clean medium sand with shells 9 Aug.

ECHINODERMS FROM THE GREAT BARRIER REEF

TABLE i (cont.)

105

STATION

LATITUDE

LONGITUDE

DEPTH

(S)

(E)

(m)

Di8

i4°34-°'

i44°5o-5'

4

Dig

i4°3i-8'

i44°5o-7'

9

D20

i6°4O'4'

I45°57'8/

22

D2I

i6°4O*4'

i45°59-i'

24

D22

i6°40-3'

I46°O'2'

27

D23

i6°3g-2'

I46°i'2'

27

024

i6°40-i'

I46°i'2

29

025

i6°40'7'

I46°i-o'

27

D26

i6°4i-i'

i46°o-8'

22

027

i6°38-i'

I45°57<8/

27

D28

i6°39-2'

I45°57'6/

27

029

i6°4i-6'

I45°57'3'

9

D3o

i5°44-3'

I45°27-8'

ii

D3i

I5°44'3'

i45°27-7'

ii

D32

I5°44'0'

i45°27-i'

18

D33

I5°43>2/

i45°27-4'

16

D34

i5°43-8'

i45°29-2'

27

D35

i5°44-4'

i45°3i-i'

27

D36

I5°45<0/

i45°33-o'

37

D37

i5°45-6'

I45°35'°'

29

D38

I5042-g'

i45°25-6'

21

D39

I5°40-g'

i45°22-9'

2O

040

I5°39'3'

I45°22'0'

18

041

I5°37'8/

I45°2i-o'

16

042

I5°39'5'

i45°2g-4'

16

D43

I5°39'0'

i45028-8'

18

D44

I5°38'3/

I45°28-0'

16

D45

i5°37-3'

i45°26-5'

22

046

I4044-o'

i45°n-6'

9

D47

I4°43'7'

I45°i2-i'

9

D48

i4°43-47

i450ii-o'

7

049

i4°42-6'

I45°io-i'

7

050

i4°36-8'

i45°28-7'

22

051

i4°37-2'

i45°26-3'

22

052

I4°39'3'

i45°25-7'

24

Ki(A)

I3°28-2'

I43042'0/

22

K2(A)

13*12-2'

I43°34'I/

II

K3(A)

n°50-o'

I43°34'2'

64

K3(N)

,,

,,

,,

K4(A)

1 1 40*4

I43°57'8/

35

K5(A)

i i 40*4

I43058>5/

34

K6(A)

i i 40*4

i43°59-3'

36

K7(N)

II°40-2'

i44°o-2'

22

K8(N)

n°48-4'

i43°2i-4/

18

Kg(N)

I2°g-o'

I43°I3.2'

24

Kio(N)

I2°l8-o'

i43°ig-6'

34

DEPOSIT

Clean coarse sand with calcareous algae

Muddy sand with shells

Silty mud with many forams

Silty mud with many forams

Calcareous mud with few forams

Muddy sand with many forams

Calcareous mud with many bivalves

Calcareous mud with many bivalves

Muddy sand with many forams

Silty mud with many forams

Silty mud with many forams

Clean medium sand

Muddy fine sand

Muddy fine sand

Muddy fine sand with forams

Sandy silt - few forams

Mud with coral fragments

Mud

Mud with coral and shell

Calcareous mud

Muddy coarse sand with forams

Muddy sand with many shells

Muddy sand with much shell

Muddy sand with shells

Calcareous mud with coral

Calcareous mud with coral

Coral blocks

Muddy sand with many forams

Muddy sand with few forams

Muddy sand with few forams

Muddy sand with few forams

Muddy sand with few forams

Halimeda sand

Silty sand with forams

Silty sand with forams

Sticky black mud

Sticky black mud

Silty fine sand with shells

Halimeda sand
Silty sand with Halimeda
Halimeda sand
Medium -coarse sand
Silty sand with forams
Black mud with shells
Black mud with shells

DATE
(1973)

9 Aug.

9 Aug.
16 Aug.
16 Aug.

16 Aug.

17 Aug.
17 Aug.
17 Aug.

17 Aug.

1 8 Aug.
18 Aug.
1 8 Aug.
3i Aug.
3i Aug.
3 1 Aug.
3i Aug.

Sept.

Sept.

Sept.

Sept.
2 Sept.
2 Sept.
2 Sept.

2 Sept.
4 Sept.
4 Sept.
4 Sept.

4 Sept.
i Oct.
i Oct.
i Oct.
i Oct.

21 Oct.
21 Oct.

21 Oct.

28 Oct.

29 Oct.
3i Oct.

5 Nov.

3 Nov.
3 Nov.
3 Nov.
3 Nov.

6 Nov.

7 Nov.

8 Nov.

Stations Di-D52 were sampled with a naturalist dredge, Stations Ki-Kio with anchor (A) and/or
naturalist (N) dredge.

io6

GIBBS, CLARK & CLARK

TABLE 2

Positions of islands and reefs where collections of littoral echinoderms were made.
Dates of collection and survey station numbers are given

LOCALITY

Beesley

Bewick

Bird

Hampton

Howick

Ingram

Lark Passage

(reef to N of )
Lizard
Low

Mid Reef
Noble
Pelican
Pipon
Three

Turtle I
Turtle IV
Two
Watson

LATITUDE (S)

I2°I4-5'
I4°26-0'

I4°33'8'

I4°30-2'

I4°30-o'
I5°o6-o'
I4°40-o'

I6°23'2'

I4°30-V

I4°07-4'

I5°07-o'

I4I44'°'
I5°oi'3'

I4°28-0'

LONGITUDE (E)

I43°I2-0'

i44°48-6'

I44°53'2'
i44°58-4'

I44°52-8'

I45°44'0'

DATE

7 Nov.
9 Aug.
6 Nov

ii Aug

8 Aug
22/23 Oct

25 Sept

SURVEY STATIONS

i45°28-o'

21 Oct

i45°34-o'

23/30 Aug

i44°57'5'

10 Aug

i44°46-o'

23 Oct

I43°50-o'

29 Oct

I44°3i'5'

25 Oct

I45°25'5'

6/28 Sept

i45°n-3'

29 Sept/i4 Oct

I45°I2'2'

8 Oct

i45°26-6'

20/22 Sept

i44°53-5'

28 July

IN.i-IN.3
Z4-Z5

LD/Z3
Li-Lj

3-I-3-H

ZI-Z2

TIA-TIC

2-I-2-4

TABLE 3

British Museum echinoderm collection : additional unpublished localities

LOCALITY
(REEF)

Farquharson
Lodestone
Sudbury
Wheeler

LATITUDE (S) LONGITUDE (E)

i? 47'5
i8°4i-5'
I7°oo-o'
i8°48-5'

i46°3i-o'
I47°o6-o'
I46°i3-o'

DEPTH

(m)

3-6
3-6

Littoral investigations were chiefly directed towards a survey of the fauna
inhabiting the intertidal sand and mud flats on those islands and reefs visited by
the Expedition. Stations are prefixed by the initial letter (s) of the island name
(Table 2) : details of this survey will be published elsewhere (Gibbs, 1976). In
addition, collections of echinoderms were also made on reef flats ; for the most part
these species were living either under coral boulders or in crevices. Sublittoral
crinoids were kindly sampled by Mr L. Zell at five localities (designated Z stations),
all at 3-5 m depth, as follows :

Zi - Three Isles - north end, off sand cay, September 23.

Z2 - Three Isles - south-east, off mangrove island, September 26.

Z3 - Lizard Island - off north-east promontory, October 21.

Z4 - Ingram Island - south-west of sand cay, October 22.

Z5 - Ingram Island - north-west of sand cay, October 23.

ECHINODERMS FROM THE GREAT BARRIER REEF

107

Map showing collecting localities

The Expedition collection is composed of about 450 specimens consisting of about
140 species plus about 12 samples of immature or incomplete specimens not positively
identifiable to species. About two-thirds of the species are represented by only one
or two specimens but this need not imply rarity in many cases since the larger
littoral forms for example, particularly the species of holothurians, were purposely
collected only once or twice, although generally common and widespread. The
class Crinoidea is represented by 15 species, Asteroidea by 17, Ophiuroidea by 56,

io8 GIBBS, CLARK & CLARK

Echinoidea by 21 and Holothurioidea by 29 species. Of these totals it would appear
that ii ophiuroids (including i new species), 2 echinoids and 2 holothurians* are
recorded for the first time from Australian waters and 8 further ophiuroids and I
holothurian newly from the Great Barrier Reef Province. These species are indicated
in the systematic account. Twenty-two of the species are not listed by A. M. Clark
& Rowe (1971) as occurring in some part of the whole of northern Australia, so that
the number of species from this area is increased to about 460 ; the total number of
species for the Great Barrier Reef Province rises to about 330.

Considerable faunistic information is available for Low Isles as a result of the
surveys of 1928-29 and 1954. The 1973 expedition spent just seven days at Low
Isles but it is worth noting that seven species of echinoderms, previously unrecorded
for these islands, were taken. These are Gomophia egyptiaca, Amphioplus (Lyman-
ella) bocki, Amphiura diacritica, Schizaster lacunosus, Leptosynapta latipatina,
Opheodesoma grisea and Chiridota rigida.

The bulk of the echinoderm material collected on the Royal Society- Universities
of Queensland Expedition has been deposited in the collections of the British
Museum (Natural History). Selected specimens donated to the Australian Museum,
Sydney, are indicated in the text by the numbers in square brackets [ ]. The
paratype of Amphiura phrixocantha sp. nov. only, is in the Queensland Museum.

This report was initiated by one of us (P. E. G.) who drew up the introduction,
list of associated animals (see Table 4, p. 141) and record data for the species
using identifications made by the others. The notes on individual species of
crinoids are by A. M. C., those on echinoids and holothurians by C. M. C., the two
combining on ophiuroids so that the new species is the responsibility of both Clarks.

SPECIES COLLECTED

For each species the locality or station and number of specimens are given, details
being found in Tables i and 2. Sight records of distinctive species (not collected)
are indicated by 'S'. In the case of those species for which few locality or specimen
records from the Great Barrier Reef Province exist, these are cited together with
any unpublished records from the British Museum collections (Table 3). The
remaining species are considered to be widespread in the Province (see Endean,
1957) and a blanket distribution for them can be assumed. References are kept to
a minimum by citing, wherever possible, H. L. Clark (1946), or an alternative
reference from which the original description can be traced. Synonyms and other
combinations used by H. L. Clark, Livingstone and Endean are included.

Class CRINOIDEA
COMASTERIDAE
Capillaster multiradiatus (Linnaeus, 1758)

See : H. L. Clark, 1946 : 27.
052-1; K3-i.

* One of these is the species here named 'Rynkatorpa sp. nov.'. Since this paper went to press Dr Rowe
has provided a specific name and description, to be published shortly in this Bulletin.

ECHINODERMS FROM THE GREAT BARRIER REEF 109

Comantheria briareus (Bell, 1882)

See : H. L. Clark, 1946 : 34.

Three Is (Zz) - 4[i].

Previous records : Port Denison ; Outer Reef between i7°S and I9°S (H. L. Clark,

1946).

Comantheria or Comanthus sp.

DSI - i.

This specimen is immature with only 18 very slender arms, c. 70 mm long, and few
division series - hence the uncertainty as to genus. Three arms arise direct from the
first (IBr) division series. The proximal ossicles are smooth. The centrodorsal is
very reduced, in the form of a flat pentagon marked by subradial clefts. It bears a
single cirrus socket and four small 'buds' of cirri. One IIIBr series is present ;
this has four ossicles (like all seven IIBr series) and is an outer one. On this account
the specimen is more like Comanthus parvicirrus timorensis than Comantheria
briareus, which usually has only two ossicles in the IIIBr series.

Comanthina schlegeli (P. H. Carpenter, 1881)

See : H. L. Clark, 1946 : 33.

Three Is (Zz) - i ; Lizard I. (Za) - i.

Previous records : Murray Is (3 specimens) ; Percy Is (i specimen), (H. L. Clark,

1946).

Comanthus bennetti (J. Miiller, 1841)

See : H. L. Clark, 1946 : 36 (as Cenolia bennetti) ; A. M. Clark & Rowe, 1971 : 16

(as Comanthus bennetti) .
Lizard I. (Za) - i.

B.M. collections : Wheeler Reef, near Townsville (3 specimens).
Previous records : Port Denison ; north of N Direction I. (2 specimens) (H. L.

Clark, 1946).

Comanthus parvicirrus (J. Miiller, 1841)

See : H. L. Clark, 1946 : 39 ; A. M. Clark, 1975 : 395.

SYNONYMS : Comanthus annulata (Bell, 1882) H. L. Clark, 1921 ; Comanthus
callipepla H. L. Clark, 1915 ; Comanthus timorensis (J. Miiller, 1841) H. L.
Clark, 1946.

Howick I. (north side at 3 metres depth) - [i] ; Three Is (Zi) - 2 ; Ingram I.
(Z5) - 2 ; Lark Passage (reef flat) - 2.

6*

no GIBBS, CLARK & CLARK

Comanthus samoanus A. H. Clark, 1909

See : H. L. Clark, 1946 : 36 (as Cenolia samoana) ; A. M. Clark & Rowe, 1971 : 16 (as

Comanthus samoanus}.
Ingram I. (Z^) - i ; (Z5) - 2 ; Lark Passage (reef flat) - i.

Comatella nigra (P. H. Carpenter, 1888)

See : H. L. Clark, 1946 : 25 ; A. M. Clark, 1975 : 392.
052 - 1.

B.M. collections : Wheeler Reef, near Townsville (2 specimens).
Previous records : Murray Is ; Lizard I. (H. L. Clark, 1946) ; Swain Reefs (A. M.
Clark, 1975).

This specimen does not help to solve the problem of the distinction between
Comatella nigra and Comatella stelligera mentioned by A. M. Clark (1975). It has
only 28 arms, so running down to C. stelligera, but the arm length is only c. 55 mm
so the infrequency of them could be accountable to immaturity. One of the few
mature cirri remaining has as many as 26 segments, the division series are well
separated laterally and smooth and the arms feel smooth when stroked distalwards,
since only fine grooving develops on the middle and distal brachials, all characters
justifying its identification as C. nigra.

However, seven specimens of Comatella from the Easter Group, Abrolhos Islands,
Western Australia (collected by a recent Aquinas College expedition) do fall easily
into two distinct groups. Three of them have c. 54, 59 and c. 60 arms, smooth to the
touch, and up to c. 31, 31 (or ? 32) and 35 cirrus segments (cirrals) - agreeing with
C. nigra - as opposed to 28, 29, 29 and 29 arms, rough in texture, and up to only
21, 22, 23 and 23 cirrals respectively in the remaining four specimens, which are
referable to C. stelligera. The arm length is 75-90 mm in the former (nigra) and
90-110 mm in the latter, giving no significant size difference.

The cirri are remarkably reduced in this Queensland specimen, only VII mature
ones remaining besides about ix more or less reduced immature ones and about
six obsolete cirrus sockets. The number is usually XX-XXX.

Comatella stelligera (P. H. Carpenter, 1880)
See : H. L. Clark, 1946 : 26.
Three Is. (Zi) - i.

This specimen has 29 fairly crowded arms, with little space between the division
series laterally and up to only 23 cirrus segments, agreeing with C. stelligera.

Comatula pectinata (Linnaeus, 1758)

See : H. L. Clark, 1946 : 31 ; A. M. Clark, 1975 : 394.

K3-i.

ECHINODERMS FROM THE GREAT BARRIER REEF

Comatula purpurea (J. Miiller, 1843)

See : A. M. Clark, 1975 : 394.
D8-i.

Comatula rotalaria Lamarck, 1816

See : H. L. Clark, 1946 : 29 (as Validia rotalaria) ; A. M. Clark, 1975 : 393 (as

Comatula (Validia} rotalaria}.
07-1.

HIMEROMETRIDAE

Himerometra robustipinna (P. H. Carpenter, 1881)

See : H. L. Clark, 1946 : 42.

Bewick I. (7 m depth) - 2 ; Three Is (Z2) - i ; Lizard I. (Z^) - 3.

B.M. collections : Farquharson Reef (4 specimens) ; Sudbury Reef (i specimen).

Previous records : Ellison Reef (2 specimens) (H. L. Clark, 1946).

One of the three specimens from Lizard Island is aberrant in three characters.
Firstly, its enlarged basal pinnules are almost smooth in profile, the segments not
being individually flared at their distal ends, while they number from 26 to as many
as 32 on the few that are still intact. In H. robustipinna the maximum number
given by A. H. Clark is 24 segments. Secondly, of the up to 39 segments of the
peripheral cirri, as many as 20 of the distal ones have a fairly well-developed blunt,
nose-like dorsal spine. In H. robustipinna usually less than ten distal segments are
described as having a small median tubercle and sometimes this may be negligible.
Thirdly, the middle brachials are more markedly flared and more rugose at their
distal ends than in the remaining specimens, though A. H. Clark and others appear
to give little weight to this character, which is hardly mentioned in the published
descriptions of the species of Himerometra.

This aberrant specimen seems to approximate to Himerometra bartschi A. H. Clark,
1908, known from Singapore to the Kei Islands, in the relatively large number of
segments in the proximal pinnules (though as many as 40 are described for H.
bartschi} which are also fairly smooth, and in the greater development of dorsal
spines on the distal cirrus segments (about 25 segments being spinose in H. bartschi,
described as 'small but prominent') . However, A. H. Clark considers H. robustipinna
to be a particularly variable species and in the holotype itself the proximal pinnules
are relatively smooth. The difference in the cirri may be more significant but better
sampling is needed to show whether or not a second species of Himerometra can be
recognized from Australian waters.*

* Since completion of this paper, another specimen from Lizard Island (North Reef, 18 m, collected
by Neville Coleman) has come to hand. This too has relatively smooth basal pinnules, stout basally
but slender and flagellate terminally, PD having up to 27 segments. Two intact mature cirri with 31
and 34 segments have dorsal spines from about the twelfth. Both these characters agree with the
aberrant specimen but the brachials, though becoming slightly flared, are not rugose.

112 GIBBS, CLARK & CLARK

COLOBOMETRIDAE

Colobometra perspinosa (P. H. Carpenter, 1881)

See : H. L. Clark, 1946 : 52.
Lizard I. Z - i.

Oligometra carpenteri (Bell, 1884)
See : H. L. Clark, 1946 : 49.

D2I - I.

TROPIOMETRIDAE

Tropiometra afra (Hartlaub, 1890)

See : H. L. Clark, 1946 : 53.

Ingram I. (Z4) - 6 ; (Z5) - i.

B.M. collections : Lodestone Reef, NE of Townsville (i specimen) ; Wheeler Reef,

near Townsville (2 specimens).
Previous records : Murray Is (3 specimens) ; Bowen (2 specimens) (H. L. Clark,

1946).

Subclass ASTEROIDEA

ASTROPECTINIDAE

Astropecten granulatus Miiller & Troschel, 1842

See : H. L. Clark, 1946 : 76.
Kg -i.

ARCHASTERIDAE

Archaster typicus Miiller & Troschel, 1840

See : H. L. Clark, 1946 : 79.

Hampton I. (reef flat) - 3 [2] ; Low Is (I^) - 3 ; Three Is (L5) - S.

GONIASTERIDAE

Stellaster equestris (Retzius, 1805)

See : H. L. Clark, 1946 : 97 (as Stellaster incei) ; A. M. Clark & Rowe, 1971 : 49 (as
Stellaster equestris).

Stellaster sp. juv. aff. S. equestris (Retzius)
K6-i; Kg -i.

ECHINODERMS FROM THE GREAT BARRIER REEF 113

OREASTERIDAE

Culcita novaeguineae Miiller & Troschel, 1842

See : H. L. Clark, 1946 : 108.
Low Is. - S ; Three Is - S.

Pentaceraster regulus (Miiller & Troschel, 1842)

See : H. L. Clark, 1946 : 107 (as P. australis) ; Endean, 1953 : 54 (as P. australis} ;

A. M. Clark & Rowe, 1971 : 55 (as P. regulus).
Di7-2; Di8-i.

Protoreaster nodosus (Linnaeus, 1758)

See : H. L. Clark, 1946 : 106.
Di8 - 4.

OPHIDIASTERIDAE

Fromia sp. juv.

Lizard I. (Z3) - i (on crinoid Comanthus bennetti}.

Gomophia egyptiaca Gray, 1840

See : Endean, 1965 : 230.

Low Is (off NE shore at 3 m depth) - i.

Previous record : Heron I. (2 specimens), (Endean, 1965).

Linckia taevigata (Linnaeus, 1758)

See : H. L. Clark, 1946 : 117.

Mid Reef - S ; Three Is - S ; Lark Passage - S.

Nardoa mamillifera Livingstone, 1930

See : H. L. Clark, 1946 : 116.

044-1.

Previous record : Murray Is (Livingstone, 1930).

Nardoa pauciforis (von Martens, 1866)

See : H. L. Clark, 1946 : 115.
Mid Reef - S ; Three Is - S.

Ophidiaster granifer Lutken, 1872

See : H. L. Clark, 1946 : 121.
Turtle IV I. (reef flat) - 2.

n4 GIBBS, CLARK & CLARK

METRODIRIDAE

Metrodira subulata Gray, 1840

048 - i (juvenile identified by Mrs Loisette M. Marsh).

ASTEROPSEIDAE

Asteropsis carinifera (Lamarck, 1816)

See : H. L. Clark, 1946 : 109 (as Asterope carinifera) ; Endean, 1956 : 124 (as

Asterope carinifera} ; A. M. Clark & Rowe, 1971 : 65 (as Asteropsis carinifera}.
Low Is (Tripneiistes spit) - i ; Lark Passage (reef flat) - i.

ASTERINIDAE

Nepanthia brevis (Perrier, 1875)

See : H. L. Clark, 1946 : 141.

North of Snake Reef at I4°26-2'S : i45°o-7'E (Grab st. 115), 24m depth, mud- i.

Patiriella pseudoexigua Dartnall, 1971

See : H. L. Clark, 1946 : 136 (as Patiriella exigua [part]) ; Endean, 1953 : 54 (as
Patiriella exigua) ; 1956 : 125 (as Patiriella exigua} ; Dartnall, 1971 : 43 (as
Patiriella pseudoexigua}. [Non Asterias exigua Lamarck, 1816.]

Three Is (3.7) - [i] ; Turtle I I. (TIC) - 3.

ACANTHASTERIDAE

Acanthaster planci (Linnaeus, 1758)

See : H. L. Clark, 1946 : 150.
Ingram I. (Z4, Zs) - S.

Subclass OPHIUROIDEA

EURYALIDAE

Euryale aspera Lamarck, 1816

See : H. L. Clark, 1946 : 172.
K3-2.

OPHIOMYXIDAE

Ophiomyxa australis Liitken, 1869

See : H. L. Clark, 1946 : 170.
K6-i.

ECHINODERMS FROM THE GREAT BARRIER REEF 115

OPHIACANTHIDAE

Ophiacantha Iconfusa Koehler, 1905

See : H. L. Clark, 1946 : 184.
K3-i.

AMPHIURIDAE

The disc of many burrowing amphiurids is very vulnerable to loss during collection
and accordingly several of these specimens cannot be determined as to species.

Amphiodia sp. juv.
D2 - i ; D2i -2.

Amphioplus (Amphioplus) didymus H. L. Clark, 1938 (Fig. la)

See : H. L. Clark, 1946 : 206 ; A. M. Clark & Rowe, 1971 : 101, fig. 48d, e.
Di - i ; 03-1; Di5 - i(no disc).

These specimens provide a new record for the Great Barrier Reef Province. The
holotype was taken at Broome, NW Australia, and a paratype at Darwin, the type-
locality of the very similar Amphioplus stenaspis H. L. Clark. H. L. Clark's descrip-
tions and figures of these two species (1938) indicate that they can be distinguished
by the contiguity of the radial and adoral shields and the arrangement of the oral
papillae. However, his description contradicts his figure of A. didymus with regard
to the adoral shields, which are in fact broadly contiguous in the holotype, as in
A. stenaspis, while the present material from Queensland indicates that the radial
shields may also be contiguous distally in A. stenaspis. (Incidentally the second
point throws further doubt on the validity of Amphioplus iuxtus Murakami, 1943
from the Caroline Islands as distinct from A. stenaspis, commented on by A. M.
Clark in 1971.) The character of the arrangement of the oral papillae in a concave
row (in A . didymus) as opposed to a straight line concealing more of the oral tentacle
scale (in A . stenaspis) used in the 1971 key to distinguish the two, is also now found
to be variable. This leaves only the distinctive bihamulate shape of the second from
lowest arm spine beyond the arm base in A. didymus as a reliable diagnostic
character ; this spine in A . stenaspis is simply blunt distally (Fig. ib) .

Amphioplus didymus is also liable to be confused with Amphioplus (Lymanella)
bocki Koehler (see below) which similarly has two very large tentacle scales and the
second arm spine modified - though in the case of A . bocki no more than broad
spatulate in the Queensland specimens. The most obvious distinctions are that
A. bocki has only three arm spines proximally, the arms themselves are relatively
longer (about 12 times the disc diameter), the radial shields are relatively broader
and more fully contiguous and the dorsal arm plates are elliptical rather than
blunted rectangular in shape.

Ii6 GIBBS, CLARK & CLARK

Amphioplus (Amphioplus) impressus (Ljungman, 1867)

See : A. M. Clark, 1970 : 63, fig. 9!, m.
D22-2; 036-1; D40-[i].

This is a new record for Australian waters. The species was previously known
from various localities in the East Indies, including Timor and New Guinea
approximating to Australia, from the Philippines and from Marovo lagoon, Solomon
Islands, collected by Gibbs in 1965 (unpublished).

Amphioplus impressus may be confused with the following species, Amphioplus
intermedius, since both are unusual among species of Amphioplus s.s. in having the
distalmost oral papilla markedly broadened. Also the disc is fully scaled, often
with a sharp edge, the radial shields are more or less fully contiguous, the arm spines
number three and the two tentacle scales are moderate in size. The most obvious
difference is that A . impressus has coarser, very well-defined disc scales (a minimum
of six across the shortest line between two radial shields interradially in a specimen
with d.d. 4-5 mm compared with c. 10 in A. intermedius with d.d. 2-5 mm). The
rosette is also conspicuous in specimens of impressus with original (as opposed to
regenerated) discs, the dorsal arm plates of that species also have a median distal
angle and the first ventral arm plate is completely superficial, fitting up against the
distal edges of the two outermost oral papillae rather than being overlain by them
proximally.

Amphioplus (Amphioplus) intermedius (Koehler, 1905)
See : A. M. Clark, 1970 : 45, 64.

D2-I.

This too is a new record for Australian waters. The species was previously known
from the East Indies off Java and the Borneo Bank. Comparative remarks with
A. impressus are given above.

Amphioplus sp. aff. A. intermedius (Koehler)
D39- 2 (no discs).

Amphioplus (Amphioplus) lucidus Koehler, 1922

See : H. L. Clark, 1946 : 204.
D8 - i ; 041 - i ; K$ - i.

A new record for the Great Barrier Reef Province. The species was previously
known to the west from Broome, NW Australia, and to the south-east from Broken
Bay, N.S.W.

Amphioplus (Amphioplus) stenaspis H. L. Clark, 1938 (Fig. ib, c)

See : H. L. Clark, 1946 : 206 ; A. M. Clark & Rowe, 1971 : 101, fig. 53c-e.
D38-i(nodisc) ; K6-i.

ECHINODERMS FROM THE GREAT BARRIER REEF 117

A new record for the Great Barrier Reef Province. The species was previously
known only from Darwin.

Comparative remarks with A . didymus are given under that heading.

Amphioplus (Lymanella) bocki Koehler, 1910 (Fig. id-f)

See : Devaney, 1974 : 106.

Low Is : (L5) - i ; (L6) - [i] ; (Ly) - i ; Three Is (3.2) - 3.

These records provide an extension of range to Australian waters for this species,
recently revived by Devaney from the synonymy of Amphioplus (Lymanella) laevis
(Lyman). The species was previously known from Fiji, the Society Islands, the
Tuamotu Archipelago and the Cook Islands, as well as from Marovo lagoon, Solomon
Islands, collected by Gibbs in 1965 (unpublished).

A specimen from Three Isles has d.d. 5-5 mm and a.l. c. 70 mm, a ratio of c. i : 12.
After about a year in alcohol some of the arm spines are stih1 banded with brown and
there is a brown spot in the centre of each oral shield ; otherwise the specimen is pale.

The disc scaling is rather fine with no distinct rosette ; the margin is distinctly
angular. The radial shields are fully contiguous, length : breadth c. 3 : i and the
length is about two-fifths of the disc radius, that is slightly smaller than shown in
Koehler's photograph of 1927, but conforming with the range in Devaney's material.

The oral shields are spearhead-shaped, longer than broad, with a short distal lobe
and a sharp proximal angle.

The dorsal arm plates are broad elliptical, the distal side straight to convex with
no trace of a median angle (Fig. le). The ventral arm plates are pentagonal with
the distal side straight. The arm spines number three throughout ; beyond the arm
base the second one becomes flattened with the tip abruptly truncated. The two
tentacle scales are very large, the one on the ventral arm plate sometimes extending
distally beyond the plate.

The complete absence of any sort of median distal angle to the dorsal arm plates
and the bluntness of the middle arm spine fully justify Devaney's restoration of
A. bocki from the synonymy of A. laevis, provisionally recorded here from Australian
waters, which has conspicuously slender, sharp pointed arm spines. There is clearly
some variation in the degree of modification of the second arm spine in A. bocki.
Koehler describes the spines as obtuse and rounded at the tips, the middle one being
evidently not distinguished ; Devaney's from SE Polynesia are bluntly pointed, the
middle spine widest, which description agrees with the two specimens from the
Solomon Islands (Fig. if). The spatulate form of the middle spine in these
Queensland specimens (Fig. id) therefore appears to be excessive but does not on
its own justify a taxonomic distinction.

Amphioplus (Lymanella) depressus (Ljungman, 1867)

See : H. L. Clark, 1946 : 205.

Di - i ; D2 - i ; D6 - i ; D8 - i ; Dig - i[i] ; K3 - i ; K8 - 2.

6**

u8

GIBBS, CLARK & CLARK

FIG. i. Twentieth arm segments of Amphioptus species, (a) Amphioplus (Amphioplus)
didymus H. L. Clark, d.d. c. 4 mm (actual disc lost), st. Di5, ventral view, (b) and (c)
Amphioplus (Amphioplus) stenaspis H. L. Clark, d.d. 3-8 mm, st. K6, ventral view (b)
and the corresponding dorsal arm plate (c). (d) and (e) Amphioplus (Lymanella) bocki
Koehler, d.d. 5-5 mm, Low Is, ventral view (d) and the corresponding dorsal arm plate (e).
(f ) Amphioplus (Lymanella) bocki Koehler, d.d. 6-0 mm, Marovo lagoon, Solomon Is,
ventral view, (g) and (h) Amphioplus sp. ? A. (Lymanella) laevis (Lyman), d.d. c. 5 mm
(actual disc lost), st. Ka, ventral view (g) and the corresponding dorsal arm plate (h).
[The scale represents i mm.]

These appear to be the first records of Amphioplus depressus from south of Torres
Strait.

D.d. in the largest specimen seen by H. L. Clark from Australian waters is 6 mm.
One of the present collection has the disc i mm broader.

Amphioplus sp. ? A. (Lymanella) laevis (Lyman, 1874) (Fig. ig, h)

See : A. M. Clark & Rowe, 1971 : 102, fig. 48b, c.
K2 - 2 (no discs).

The slight median angle on the dorsal arm plates, relatively slender pointed arm
spines and less exaggeratedly large tentacle scales than usual in Lymanella leave
little doubt of the identity of these two specimens, but it should be confirmed by
intact specimens with radial shields before the range of the species can definitely be
extended to Australia. A. laevis was previously known from the Red Sea to the
Philippines, the Gilbert Islands and various localities in the East Indies, besides also
being taken at Marovo lagoon, Solomon Islands, with Amphioplus (Lymanella) bocki
by Gibbs in 1965.

ECHINODERMS FROM THE GREAT BARRIER REEF 119

Amphioplus sp. aff. A. (Amphioplus) pectinatus Mortensen, 1933
KIO-I.

This small delicate specimen does retain the disc, though it is partly detached.
The identification is only provisional on two counts, firstly because Amphioplus
pectinatus is only known from as far away as Natal, South Africa, and secondly
because there appear to be two distinct forms there, designated A and B by A. M. Clark
(1974 : 456) and it is not yet certain to which one the holotype belongs.

D.d. in this specimen is c. 4-5 mm. The disc scaling is fairly fine, the radial shields
are very long and narrow and barely contiguous distally, the oral shields are pear-
shaped, the adorals only contiguous interradially on the jaw with the madreporic
oral shield, the dorsal arm plates thin and broad fan-shaped, the arm spines slender
and tapering, numbering four or five proximally for a few segments, then three, the
second from lowest developing a small, distally directed hook and the tentacle
scales numbering two, moderate in size. These characters conform with those of
A. pectinatus but none of them are particularly distinctive, even in combination.
However, the specimen also has two unusual features : there are spinose processes
distal to each radial shield and there is a second oral tentacle scale on the side of the
oral plate on the same level as the usual scale above (behind) the two middle oral
papillae. The former occurs in both forms of A. pectinatus but the latter only in
form B (A. M. Clark, 1974, fig. 8f-i).

Amphipholis squamata (Delle Chiaje, 1829)

See : H. L. Clark, 1946 : 202.
05-1; D6-i; 046-2; 047-1.

Amphiura (Amphiura) ambigua Koehler, 1905

See : Koehler, 1905 : 39 ; H. L. Clark, 1946 : 196.
KIO-I.

This is a new record for the Great Barrier Reef Province, though a specimen from
Port Curtis in the British Museum collections is now also identified as Amphiura
ambigua. The species was previously known in Australia from Darwin (H. L. Clark,

1938).

The Kio specimen has d.d. only 3 mm ; all the arms are badly broken. The
distal oral papilla is small and peg-like, there are seven arm spines proximally, the
lowest and uppermost spatulate, the second and third from lowest bihamulate and
the others with only distally pointing hooks and there are two large tentacle scales.
These characters agree with Koehler's description of the holotype of A. ambigua,
from the Sulu Archipelago, Philippines, but not the ventral side of the disc which is
naked, not scaled. However, in this it agrees with H. L. Clark's key, although he
does not mention the anomaly elsewhere ; presumably his Darwin specimens are
partly naked. In contrast, the Port Curtis specimen has fine scaling both dorsally

120 GIBBS, CLARK & CLARK

and ventrally ; d.d. is 6 mm and there are eight arm spines, the middle ones hooked ;
the distal oral papilla is relatively larger and more rounded than in the smaller
specimen.

Several species of amphiurids are already known to show considerable variation
in the extent of the disc scaling, notably Amphiura acrystata H. L. Clark, 1911,
from the North Pacific, so this character alone is no barrier to considering
ventrally naked Australian specimens to be conspecific with A . ambigua.

Amphiura (Amphiura) bidentata H. L. Clark, 1938

See : H. L. Clark, 1946 : 197.
K3-2.

Amphiura (Amphiura) catephes H. L. Clark, 1938

See : H. L. Clark, 1946 : 193.
D2 - i ; D6 - i.

This is a new record for the Great Barrier Reef Province. The species was
previously known only from two specimens taken at Port Jackson, N.S.W.

Amphiura (Amphiura) constricta Lyman, 1879

See : H. L. Clark, 1946 : 194.
Di6 - i ; 046 - i.

Amphiura (Amphiura) diacritica H. L. Clark, 1938

See : H. L. Clark, 1946 : 196.
Low Is. (L6) - i.

The holotype and only other recorded specimen was taken at Black Island,
Whitsunday Group (H. L. Clark, 1938). The Low Isles specimen was found in
muddy sand at low water mark.

Amphiura leptotata H. L. Clark, 1915

See : A. M. Clark & Rowe, 1971 : 95, fig. 5oa-c.

Di - 2 ; D2 - 3 ; Dn - 2 ; Di6 - 2 ; D2i - [2] ; 023 - 3 ; 034 - i ; 039 - i ;
048 - i ; 049 - i.

These records extend between i4°2i'S and i6°4o'S and provide an extension of
range to Australia. The type-locality is in the Philippines and the species was also
collected by Gibbs at Marovo lagoon, Solomon Islands, in 1965. It may have
escaped attention because of the small size, d.d. in the holotype being only 3-5 mm
and not exceeding 3 mm in the other specimens.

At d.d. 3 mm the arm length is 30+ mm. The disc is covered with very fine
scales above but on the ventral side this gives way proximally to naked skin with

ECHINODERMS FROM THE GREAT BARRIER REEF 121

only scattered scales, appearing dark brown in colour. The radial shields are
banana-shaped, two-fifths to half as long as the disc radius and with length : breadth
c. 3 : i ; they are completely separated by a row of four or five scales.

The arm spines number four proximally, falling to three. The second spine from
below is much broader than the others with an abruptly truncated tip, not simply
tapering to a blunt tip. There are no tentacle scales.

The absence of tentacle scales distinguishes this species from all those included in
H. L. Clark's key to the Australian species of Amphiura. However, with the
partially bare underside of the disc it is shared by a species which H. L. Clark
(1946 : 201) referred to Ophionephthys, namely Ophiolepis perplexa Stimpson, 1855,
from Port Jackson, referred by Lyman in 1865 to Amphiura, to which it should be
restored following A. M. Clark's restriction of Ophionephthys (1970). Amphiura
perplexa does not appear to have been recorded from the Port Jackson area since
1855. If further material from N.S.W. proves to be conspecific with that from
Queensland, then the name Amphiura leptotata must become a synonym.

Amphiura (Amphiura) magnisquama H. L. Clark, 1938
See : H. L. Clark, 1946 : 195.

This is a new record for the Great Barrier Reef Province. The species was
previously known from off Botany Bay, N.S.W. (H. L. Clark, 1938).

Amphiura phrixocantha sp. nov.
(Fig. 2, PL i, figs 1-3)

TYPE MATERIAL. 036 - 2. i5°45'S : i45°33'E (between Endeavour and Cairns
Reefs, opposite Papuan Pass) at 37 m depth. Holotype B.M. reg. no. 1975.6.27.44 ;
paratype Queensland Museum no G. 9297.

DESCRIPTION. The holotype has d.d. 8mm, a.l. 30+ mm. The arms taper
rather abruptly distally ; two of them are regenerating.

The disc is covered with relatively large, very thick scales, numbering five to seven
across each interradius between adjacent radial shields. No primary rosette is
distinguishable. The disc is strongly contracted interradially and on the ventral
side and this probably accounts for the almost vertical alignment of the ventral
scales. The radial shields are markedly convex abradially, 1-9-2-0 mm in length,
with length : breadth c. 2-0 : i and length about half the disc radius. The two of
each pair are completely separated by three (or in one case four) large scales.

The oral shields are keyhole-shaped, the main part almost circular but with a
well-developed distal lobe ; length : breadth is 1-5:1. The adoral shields are
broadly contiguous interradially with a rounded end radially ; from the middle half
of the proximal side of each arises a relatively huge, very broad rounded distal oral
papilla, about as long as the oral plate and about half again as broad, sometimes
slightly prolonged interradially towards the block-like infradental papillae. The

GIBBS, CLARK & CLARK

first oral tentacle scale is compressed vertically and appears very narrow and pointed
in ventral view. A most unusual additional scale borders the adradial side of the
second (more superficial) oral tentacle pore like a vertical crest, its ventral end
curling horizontally to a small extent just behind the large distal oral papilla, so
giving the effect of a small conical additional oral papilla. It probably arises from
the edge of the first ventral arm plate.

The dorsal arm plates are relatively narrow, covering only the midline of the arms,
rounded octagonal in shape and broadly contiguous, the distal edge straight or
slightly convex, about as broad as long proximally but becoming half as broad again
by the twentieth segment. The ventral arm plates are broader than long,
rectangular, with blunt distal corners, the proximal and distal sides straight.

The number of arm spines is nine proximally, falling to seven by the thirtieth
segment but still as many as five on the distalmost segments remaining. Except
for the three uppermost, which are simply flattened and blunt, and the lowermost
one, which is pointed, all the spines are elaborately hooked, hence the specific name.
The hooks are glassy and hyaline. The second from lowest spine is the longest and
on the basal segments has a large hook directed proximally ; beyond the disc it
becomes bihamulate. The next three spines are mostly also bihamulate but the
sixth usually has only the distal-pointing hook.

The tentacle scales are unusual. On the first four arm segments there are two
large scales on each pore, a rounded proximal one on the lateral arm plate overlying
a more elongated but truncated scale projecting at right angles from the side of the
ventral arm plate. From about the fifth arm segment, the latter scale becomes
gradually reduced in size and inset so as to become barely distinguishable from about
the twelfth segment.

The paratype has d.d. c 5-5 mm. The disc is again very contracted interradially.
The rosette is distinct among the large rounded scales. Proximally there are not

FIG. 2. Amphiura phrixocantha sp. nov. Holotype. (a) Twelfth arm segment in dorsal
view, (b) Fifth segment in ventral view, (c) and (d) Parts of twelfth and twenty-fifth
segments in ventral view, (e) Oblique distal ventrolateral view of seventh arm segment,
the lower spines (shown uppermost in the figure) somewhat foreshortened. [The scale
represents i mm.]

ECHINODERMS FROM THE GREAT BARRIER REEF 123

more than six arm spines. Otherwise it is similar to the holotype, allowing for the
smaller size.

Both specimens, after about a year in alcohol, still show a yellow colour on the
dorsal arm plates, interrupted at intervals by usually one to three unpigmented
plates.

AFFINITIES. This species has most in common with Amphiura dolia H. L. Clark,
1938, from Port Jackson, to which it runs down in H. L. Clark's Amphiura key
of 1946 (p. 191) since it has two tentacle scales, fully scaled disc with coarse scales
and numerous arm spines. The very broad distal oral papilla is shared by both,
while the disc scaling is also thick and the form of the radial shields similar in
A . dolia, allowing for the much larger size of the holotype, d.d. 11-12 mm. However,
the arm spines are still only up to eight in number at this large size and are un-
specialized, 'pointed but not acute', also the second tentacle scale maintains its size,
or even exceeds the proximal scale.

Amphiura (Fellaria) octacantha (H. L. Clark, 1915)

See : H. L. Clark, 1946 : 200 (as Ophionephthys octacantha} ; A. M. Clark & Rowe,

1971 : 95, fig. 26b (as Amphiura (Fellaria) octacantha).
SYNONYM : Ophionephthys decacantha H. L. Clark, 1938.
K3-2.
Previous records : Torres Strait.

Dougaloplus echinatus (Ljungman, 1867)

See : H. L. Clark, 1946 : 203 (as Ophiocnida echinata) ; A. M. Clark & Rowe,

1971 : 100 (as Dougaloplus echinatus).
03 - i ; 023 - [i] ; 027 - i.
Previous record : Papuan Pass (i specimen) (H. L. Clark, 1932).

Ophiocentrus verticillatus (Doderlein, 1896)

See : H. L. Clark, 1946 : 198.
D6-i.

A new record for the Great Barrier Reef Province. Previously the species was
known from Broome, NW Australia.

OPHIAGTIDAE

Ophiactis savignyi Muller & Troschel, 1842

See : H. L. Clark, 1946 : 210.
Turtle I I. (reef flat) - 7.

Ophiactis sp. juv.
04-1 ; D6-2; Di6-i.

124 GIBBS, CLARK & CLARK

OPHIOTRICHIDAE

Macrophiothrix belli (Doderlein, 1896)

See : H. L. Clark, 1946 : 221.

Mid Reef (reef flat) - i ; Turtle IV I. (reef flat) - i.

Previous record : Torres Strait.

Macrophiothrix koehleri A. M. Clark, 1968

See : Koehler, 1907 : 333 (as Ophiothrix galatheae [part]) ; H. L. Clark, 1932 : 204

(as Ophiothrix longipeda [part]) ; A. M. Clark & Rowe, 1971 : 114, fig. 37p (as

Macrophiothrix koehleri) .
Three Is (reef flat) - 2; Three Is (3.11) - i.

In H. L. Clark's key to the species of Macrophiothrix (1946 : 218) this species runs
down to the vicinity of M. spinifera and M. scotia (EE) because of the presence of
spinelets on its oral shields, whereas its arm structure agrees far better with that
of M . belli, the dorsal arm plates having sharp latero-distal angles and the longer arm
spines smooth shafts and clavate tips.
Previous records : Probably Koehler's specimen from Torres Strait as well as

H. L. Clark's Barrier Reef Expedition one from Low Isles are conspecific with

M. koehleri.

Macrophiothrix lorioli A. M. Clark, 1968

See : A. M. Clark & Rowe, 1971 : 115, fig. 37 m.
Mid Reef (reef flat) - i.

This specimen provides a new record for Australian waters. The species was
previously known from the South China Sea, the Solomon Islands and the Tonga
Islands.

In H. L. Clark's 1946 key, M. lorioli runs down to the vicinity of M . callizona,
M. calyptaspis and M. michaelseni (GG) because of the unarmed oral shields and
non-trapezoidal dorsal arm plates, these being fan-shaped with more or less distinct
lateral angles. M. callizona differs in having elliptical dorsal arm plates and the
distal side of the ventral arm plates concave rather than convex. M. calyptaspis
has the arms relatively short, less than ten times the disc diameter rather than
14-19 times in M. lorioli (17-5 times in this specimen), while M. michaelseni from
the west coast of Australia (as opposed to the north-west) has the radial shields more
or less bare rather than covered with trifid stumps.

Macrophiothrix megapoma H. L. Clark, 1938
See : H. L. Clark, 1946 : 219.

ECHINODERMS FROM THE GREAT BARRIER REEF 125

Macrophiothrix sp. aff. M. longipeda (Lamarck, 1816)
See : H. L. Clark, 1946 : 221.

D2I-I.

This specimen agrees in all respects with M. longipeda except that it has the
lowermost arm spine hook-shaped with only one or two accessory spines within the
main hook, rather than comb-like with multiple accessory spines. There are also
only up to six arm spines (d.d. is 15 mm), an unusually small number for M.

longipeda.

Ophiomaza cacaotica Lyman, 1871

See : H. L. Clark, 1946 : 234.

07 - i (on crinoid Comatula rotalaria) .

H. L. Clark (1921) records this commensal (or parasitic) ophiuroid from Comatula
purpurea and Comanthus annulatus (i.e. Comanthus parvicirrus) . Other specimens
in the British Museum collections from Queensland were from Comatula Solaris,
Comantheria perplexa and Zygometra microdiscus.

Ophiothela danae Verrill, 1869

See : H. L. Clark, 1946 : 229 (as Ophiothela hadra} ; A. M. Clark & Rowe, 1971 : 116,

pi. 14, fig. 5.
South-west of Long Reef at I5°O5-2'S : i45°33-o'E (Grab st. 3I4A), 31 m depth -

innumerable specimens on a gorgonian [5].

The arm length in these specimens appears to be 3€5-5*o times the d.d., their
colour in alcohol is blue on white (the gorgonian contrasts in vermilion but the match
may have been better in life) and morphologically also there is no reason why they
should not be referred to 0. danae, with which A. M. Clark provisionally synonymized
Ophiothela hadra in 1971. Although H. L. Clark (1915) originally distinguished
0. hadra by the arms being only about twice the d.d., in 1946 he amended this to
'hardly' four times the d.d. It is possible that relative arm length may be correlated
with the nature of the host, the types of 0. hadra being from a sponge where shorter
arms may be at an advantage. This remains to be seen from field studies, when the
extent of correlation between ophiuroid and the colour of the host can also be
determined, following from Japanese work on these lines.

Ophiothrix (Ophiothrix) foveolata Marktanner-Turneretscher, 1887

See : A. M. Clark & Rowe, 1971 : no, pi. 15, fig. 3.
D28-2; 051-3.

These records represent an extension of range to Australia. The species was
previously known from the Aru and Kei Islands, the Sunda Islands and Sulu
Archipelago.

126 GIBBS, CLARK & CLARK

Because of the completely bare radial shields and arms of moderate length,
O.foveolata runs down to Placophiothrix s.l. in H. L. Clark's key to the Ophiotrichidae
(1946 : 213). The very pretty colour pattern with the radial shields and distal ends
of the arm segments (ventrally as well as dorsally) outlined by narrow bands
(dark-red in preserved specimens) serves to distinguish it from the species which H. L.
Clark included in Ophiothrix and Placophiothrix while the narrowly contiguous
dorsal arm plates exclude it from Placophiothrix. The disc armament is relatively
reduced dorsally with only a few scattered spinelets.

Four of these specimens have d.d. c. 6 mm and the arm length 5-0-6-7 times as
much, while the fifth at d.d. 4 mm has arms 30 mm long, a ratio of i : 7-5.

Ophiothrix (Acanthophiothrix) artnata Koehler, 1905

See : H. L. Clark, 1932 : 205 (part, st. 25) (as Ophiothrix stelligera) ; A. M. Clark &
Rowe, 1971 : 84, 90, in (as Ophiothrix (Acanthophiothrix) armata).

South-east of Long Reef at i5°04'6'S : I45°37-2'E (Grab st. 3I5A) at 31 m depth - 2
(on coral Seriatopora) ; 045 - i.

Previous record : Papuan Pass (i specimen) (H. L. Clark, 1932).

Since H. L. Clark (1932) did not distinguish the small Barrier Reef Expedition
specimen from Ophiothrix stelligera (i.e. 0. ciliaris), Ophiothrix armata is not included
in his keys of 1946. It would fall within Placophiothrix in his wide sense, because
of the conspicuous, mostly bare radial shields. In the Placophiothrix key (1946 : 223)
it runs down to EE, P. elegans (with a single longitudinal dark line on each arm) and
P. trilineata, said to have three dark lines separated by two light ones. However,
this last observation is surely a mistake since other descriptions of trilineata, which
A. M. Clark has referred back to Ophiothrix s.s., give three light lines and two dark
ones between them, the limits of the outer light ones sometimes irregular or ill-
defined.

All four specimens from the Great Barrier Reef are small with slender arms and
needle-like spines, the arms being marked with two dark lines without a background
of lighter colour.

Ophiothrix (Acanthophiothrix) proteus Koehler, 1905

See : A. M. Clark & Rowe, 1971 : in, pi. 15, fig. 5.
K6-i.

This record provides an extension of range to Australia. The species has previously
been recorded in the Pacific from the South China Sea, the Sunda Islands, Kei
Islands and New Caledonia.

The single specimen belongs to the colour form of 0. proteus which has the double
dark lines on each arm green rather than purple, though the radial shields and
central disc plates are still pinkish after preservation in alcohol ; the keel of the
dorsal arm plates between the green lines is white. The longest arm spine is five
to six times the segment length.

ECHINODERMS FROM THE GREAT BARRIER REEF 127

Like the species above, 0. (Acanthophiothrix) proteus is likely to run down to
section EE of H. L. Clark's 1946 key to the species of Placophiothrix. Again it is
a much more slender species than trilineata.

Ophiothrix (Acanthophiothrix) purpurea von Martens, 1867
See : A. M. Clark & Rowe, 1971 : 112, fig. 35d, pi. 15, figs 4, n.

This record provides an extension of range to Australian waters. The species
has previously been recorded through most of the Indo-West Pacific, including the
Banda Sea, New Hebrides and Solomon Islands in the neighbourhood of Australia.

The specimen is small with d.d. only c. 1-5 mm but has the distinctive narrow dark
line on each arm and very slender form. Again it runs down to section EE of
H. L. Clark's Placophiothrix key but differs from P. elegans, which A. M. Clark has
referred back to Ophiothrix s.s., recorded from Disaster Bay, N.S.W., in the narrow
hexagonal rather than broad rhombic dorsal arm plates and more needle-like
armament.

Ophiothrix (Keystonea) propinqua Lyman, 1861

See : H. L. Clark, 1946 : 232 (as Ophiotrichoides propinqua) ; A. M. Clark & Rowe,

1971 : 107, pi. 15, fig. 7 (as Ophiothrix (Keystonea) propinqua).
K4-i.
Previous record : Murray Is (H. L. Clark, 1946).

Ophiothrix (Theophrix) pusilla Lyman, 1874

See : A. M. Clark & Rowe, 1971 : 107, fig. 35c, pi. 15, fig. 8.
D6 - i ; 036 - i ; K3 - i.

These records provide an extension of range to Australian waters. The species
has previously been recorded from the South China Sea, Philippines, Sunda and
Kei Islands.

The three specimens have d.d. 2-6-3-5 mm and d.d. : a.l. i : 2-3-3-7, the relative
length increasing with size. The colour in alcohol is pale with a bluish tinge and
faint blue irregular bands on the arms.

This is a small but distinctive species with its thick, elongate, oval dorsal arm
plates, much narrower than long, relatively short arm spines not exceeding twice
the segment length and complete covering of the disc including the radial shields
with trifid stumps. The structure and shortness of the arms serve to distinguish it
from all the species included in H. L. Clark's key for Ophiothrix (1946 : 214).

OPHIOCOMIDAE

Ophiarthrum elegans Peters, 1851

See : H. L. Clark, 1946 : 252.

Mid Reef (reef flat) - 8 [2] ; Three Is (reef flat) - 2 ; K4 - i.

128 GIBBS, CLARK & CLARK

Ophiocoma dentata Miiller & Troschel, 1842

See : H. L. Clark, 1946 : 246 (as Ophiocoma insularis var. variegata) ; Endean,

J953 : 55 » J956 : 126 (both as 0. insularia var. variegata} ; A. M. Clark & Rowe,

1971 : 119 (as 0. dentata}.
Mid Reef (reef flat) - 8[i] ; Lark Passage (reef flat) - 2.

Ophiocoma erinaceus Miiller & Troschel, 1842

See : H. L. Clark, 1946 : 244.
Turtle IV L (reef flat) - i.

Ophiocoma scolopendrina (Lamarck, 1816)

See : H. L. Clark, 1946 : 243.

Three Is (reef flat) - 7[i] ; Lark Passage (reef flat) - i.

Ophiocomella sexradia (Duncan, 1887)

See : H. L. Clark, 1946 : 206 (as Amphiacantha sexradia} and 247 (as Ophiocoma

parva} ; A. M. Clark & Rowe, 1971 : 118, fig. 38c-f (as Ophiocomella sexradia}.
Turtle I I. (reef flat) - i.
Previous records : Mer, Murray Is [also Lord Howe I.] (H. L. Clark, 1946).

OPHIONEREIDAE

Ophionereis dubia (Muller & Troschel, 1842)

See : H. L. Clark, 1946 : 240 (as Ophionereis stigma and 0. dubia}.

DS - i ; D6 - [i] ; Dig - i.

As noted in 1953 (Proc. zool. Soc. Lond. 123 : 82) although H. L. Clark's supposition
that the 'Alert' specimens from Torres Strait named Ophionereis dubia by Bell are
in fact 0. semoni (Doderlein) is largely correct, two of them do have the disc scaling,
oral shields and proximal ventral plates unobscured by skin and the arm spines
relatively long, thereby justifying the inclusion of 0. dubia in the Australian fauna
even if it were not for the inclusion of the north-western Australian Ophionereis
stigma H. L. Clark in the synonymy.

The present records extend the range south from Torres Strait.

Ophionereis porrecta Lyman, 1860

See : H. L. Clark, 1946 : 238.

K4-i.

Previous record: Murray Is (H. L. Clark, 1921).

ECHINODERMS FROM THE GREAT BARRIER REEF 129

Ophionereis semoni (Doderlein, 1896)
See : H. L. Clark, 1946 : 240.

Previous records : Lindeman I., near Mackay (H. L. Clark, 1938) ; east of Snake
Reef (H. L. Clark, 1932) ; Low Is (Endean, 1956).

OPHIODERMATIDAE

Ophiarachna incrassata (Lamarck, 1816)

See : H. L. Clark, 1946 : 253.

Watson I. (reef flat) - I ; Mid Reef (reef flat) - 2.

Ophiarachnella gorgonia (Miiller & Troschel, 1842)

See : H. L. Clark, 1946 : 260.
Mid Reef -4[i].

Ophiarachnella infernalis (Miiller & Troschel, 1842)

See : H. L. Clark, 1946 : 262.
027 - i ; D5i - i.

Ophiochaeta hirsuta Liitken, 1869

See : A. M. Clark & Rowe, 1971 : 127, fig. 44a, b.
D6-i.

This record provides an extension of range to Australian waters. The species has
previously been recorded from Fiji and the Gilbert Islands and there are specimens
in the British Museum collections from Aldabra (Indian Ocean) and Tetel (or
Gaskell) Island in the Solomon Islands (collected by Dr H. G. Vevers in 1965).

Though small (d.d. 4 mm) the specimen shows the mixture of extremely fine
spinelets and indented granules on the disc and the spiniform rather than granuliform
armament of the oral plates, which serve to distinguish Ophiochaeta hirsuta from all
the ophiodermatids included in H. L. Clark's key (1946 : 252). The arm spines are
short and appressed and the oral shields are bare. Unusually in this specimen the
radial shields are also bare but this may be a juvenile character. The colour in
alcohol consists of a light brown central area on the disc bordered with dark brown
and then a white peripheral area ; ventrally the disc and arm bases are white but the
rest of the arms are banded with brown on both sides.

Ophiochasma stellatum (Ljungman, 1867)

See : H. L. Clark, 1946 : 259.

i; 026-3; D27-[i]; K3-2[i].

130 GIBBS, CLARK & CLARK

Ophiopeza spinosa (Ljungman, 1867)

See : H. L. Clark, 1946 : 258 (as Ophiopezella spinosa) ; A. M. Clark & Rowe,

1971 : 127, fig. 446 (as Ophiopeza spinosa}.
KS-I; K6-i.
Previous records : Torres Strait ; Murray Is (H. L. Clark, 1921).

Ophiopsammus yoldii (Liitken, 1856)

See : H. L. Clark, 1946 : 257 (as Pectinura yoldii) ; A. M. Clark & Rowe, 1971 : 127,

pi. 21, figs 7, 8 (as Ophiopsammus yoldii).
K3-u[2].

Ophiodermatid sp. juv.
KS-I.

OPHIURIDAE

Ophiolepis superba H. L. Clark, 1915

See : H. L. Clark, 1946 : 272.
Two Is (reef flat) - 3.

Ophiura kinbergi (Ljungman, 1867)

See : H. L. Clark, 1946 : 270.

D2i - i ; 023 - i ; 1)29 - 2[i] ; 031 - 5.

Previous record : Murray Is (Endean, 1957).

Class ECHINOIDEA

CIDARIDAE

Prionocidaris bispinosa (Lamarck, 1816)
See : H. L. Clark, 1946 : 286.
DI7-3M-

DIADEMATIDAE

Diadema setosum (Leske, 1778)

See : H. L. Clark, 1946 : 297 (as Centrechinus setosus) ; Endean, 1956 : 127 (as

Diadema setosum).
Two Is (reef flat) - i ; Three Is (reef flat) - S.

Mespilia globulus (Linnaeus, 1758)

See : H. L. Clark, 1946 : 315.
Turtle I I. (reef flat) - i ; D5o - 5[i] ; K6 - i.

Previous records : Port Denison (Mortensen, 1943) ; Trinity Bay (off Cairns)
(Tenison- Woods, i88ob) ; Low Is ; Hardy Reef ; Heron I. (Endean, 1961).

Temnopleurus alexandri (Bell, 1884)

See : H. L. Clark, 1946 : 311 (as Salmacis virgulata var. alexandri} ; Endean, 1956 :

128 (as Temnopleurus alexandri}.
DI-I.

Temnotrema phoenissa H. L. Clark, 1926

See : H. L. Clark, 1946 : 314.

D50 - 1.

This small echinoid, h.d. (horizontal diameter) 15 mm, v.d. 10 mm, is still very
distinctively coloured in alcohol. The apical system is greenish, the test aborally
dark purplish-red, darker in the interambulacral areas, the aboral spines red with
white tips and the primary ones also with white bases ; orally the test is white and
the spines pale green. This combination of colours agrees not only with that of the
holotype and only recorded specimen of Temnotrema phoenissa from Ellison Reef
(h.d. only 6-5 mm) but also with Temnotrema pulchellum (Mortensen), known
from the Philippines to the Kei Islands. Mortensen (1943 : 268) has already
suggested that the two nominal species may prove to be conspecific. Without
material of T. pulchellum to hand this problem cannot be resolved.

Temnopleurid sp.
04-1; 039-3.

TOXOPNEUSTIDAE

Gymnechinus epistichus H. L. Clark, 1912
See : H. L. Clark, 1946 : 326.

D2I-I.

Tripneustes gratilla (Linnaeus, 1758)

See : H. L. Clark, 1946 : 326.
Three Is (reef flat) - i.

I32 GIBBS, CLARK & CLARK

PARASALENIIDAE

Parasalenia sp.

K4 - i (test only and therefore not positively identifiable).

ECHINOMETRIDAE

Echinometra mathaei (de Blainville, 1825)

See : H. L. Clark, 1946 : 332.
Three Is (reef flat) - 2.

ECHINONEIDAE

Echinoneus abnormalis de Loriol, 1883

See : A. M. Clark & Rowe, 1971 : 158.
K4-i.

This record provides an extension of range to Australia. The specimen differs
from Echinoneus cyclostomus Leske, 1778, known from Low Isles and Heron Island
(Endean, 1956) and from Lord Howe Island (H. L. Clark, 1938), in having the spine
tubercles perforated and in lacking glassy knobs on the test. E. abnormalis has
previously been recorded from Mauritius (the type locality), the Kei Islands, Rotuma
in the EUice Islands, Palmyra Island and the Hawaiian Islands ; there is also a
specimen in the British Museum collections from Aldabra. Since the species is
sympatric in most, if not all, of these localities with the much more common E.
cyclostomus, it is possible that E. abnormalis may have gone unrecognized in other
samples.

FIBULARIIDAE

Fibularia ovulum Lamarck, 1816

See : A. M. Clark & Rowe, 1971 : 161, 167.
K7-i.

This record represents an extension of range to Australia. The species has
previously been recorded from most parts of the tropical Indo-West Pacific except
from the more remote Pacific islands.

Although Mortensen (1948) reckons that most of H. L. Clark's records of Fibularia
craniolaris (Leske, 1778) are conspecific with Fibularia ovulum, he makes an
exception of the Australian ones, referring them instead to Fibularia oblonga Gray,
1851, which he recorded from 'South, West and North Australia', without precise
localities (1948 : 223). However, in H. L. Clark's key to the Australian species of
Fibularia (1946 : 347) the present specimen does tend to run down to F. craniolaris
on account of its high ovoid test (length : height 4-5 : 3-5 mm) and the rounded

ECHINODERMS FROM THE GREAT BARRIER REEF 133

pores forming distinct petals on the aboral side (though they are not appreciably
smaller than the genital pores). In these same characters it also agrees with
F. oblonga, recorded by Endean (1961 : 294) from Caloundra, southern Queensland,
but in intact specimens like this one, as opposed to dead tests, there are important
differences in the plating of the peristome and periproct. In F. ovulum the peri-
stomial membrane is devoid of plates and the periproct has five triangular plates,
as this specimen shows, whereas F. oblonga has multiple smah1 irregular plates in
both peristome and periproct. However, dead tests of F. ovulum should be
recognizable as having the round or quadrangular periproct as broad as or broader
than long, whereas in F. oblonga it is normally elongate oval.

According to H. L. Clark the common Fibularia in northern Australia is F. volva
Agassiz & Desor, 1847, which should be distinguishable from both F. ovulum and
F. oblonga by the relatively low test (length : height c. 2 : i) with more or less tapering
ends.

Fibulariid sp.

037 - 2 (tests only).

LAGANIDAE

Laganum decagonale (de Blainville, 1827)

See : A. M. Clark & Rowe, 1971 : 162.

D4i - 1.

Previous records : Torres Strait (Agassiz, 1881) ; Bowen (Mortensen, 1948).

This species was excluded from the Australian fauna by H. L. Clark (1946) although
in 1925 (p. 156) he cited a Challenger specimen of Laganum decagonale from Torres
Strait, following Agassiz. There are now a total of eight Challenger laganids from
Torres Strait in the British Museum collections, some acquired since 1925. Although
they show considerable variation in the number and development of the genital
pores and the shape of the test, several of them agree with Mortensen's description
of L. decagonale (1948 : 331) in having five genital pores and an S-shaped groove for
the madreporic pores on the apical system, both characters distinguishing them from
Peronella. However, the extent of the variation throws some doubt on the generic
weight of these characters. The remaining Challenger specimens bridge the gap
with, and may even run down to, Peronella lesueuri (Valenciennes, 1841) to which
H. L. Clark (1946 : 346) referred them.

Laganum depressum Lesson in L. Agassiz, 1841

See : H. L. Clark, 1946 : 343 (Laganum depressum) and 344 (Laganum dyscritum) ;

Endean, 1956 : 129, 134 (L. depressum).

Low Is : (L5) - 5 ; (L6) - 4 ; Di7 - i ; 027 - i ; 030 - [2] ; 031 - i ; 033 - i ;
- i ; D46 - i.

134 GIBBS, CLARK & CLARK

This species is very common in muddy sand close to the Anchorage at Low Isles,
as also noted by Endean (1956) who agreed with Mortensen (1948) that Laganum
dyscritum H. L. Clark, 1932, of which Low Isles in the type locality, is synonymous
with Laganum depressum. H. L. Clark had acknowledged the likelihood of confusion
between the two but considered that those he called L. dyscritum have significantly
broader tests, more posteriorly situated periprocts and relatively larger petaloid
areas, besides having more elongated spinelets and a purplish brown coloration,
rather than the yellowish-khaki usual in L. depressum. Mortensen attributed the
colour difference to the type material of L. dyscritum possibly having been preserved
together with comatulids and it is certainly true that most Queensland comatulids
are very liable to release copious amounts of purple pigment in alcohol. None of
the laganids collected at Low Isles by Gibbs are at all purple. The morphological
differences Mortensen discounted as being within the range of variation of L.
depressum. His conclusion is supported by a series of measurements made on the
present collection together with the material from the Great Barrier Reef Expedition
which H. L. Clark referred to both L. depressum and L. dyscritum as well as specimens
of L. depressum from various other parts of its extensive Indo-Pacific range.

Although H. L. Clark recorded four specimens of Laganum from Low Isles as
L. depressum, these are all large spineless tests, length 60 mm or more, and do not
compare in size with the numerous specimens from Low Isles which he referred to
L. dyscritum where length is 28-42 mm. The three Great Barrier Reef Expedition
specimens from st. XXIV (NE of Pasco Reef) with length 32-45 mm are
unfortunately not in the British Museum collections and must be either at Harvard
or the Australian Museum. H. L. Clark's values for test breadth of L. depressum
of 0-80-0-85 (or 80-85%) °f length and c. 20% for distance of the periproct from the
posterior edge of the test appear to have been taken from the two tests from the
Anchorage area, now measured as 71 and 65 mm in length with breadths 79 and
89% and periproctal distance 19 and 18%. However, 30 specimens of L. depressum
with test length c. 25-40 mm from various localities show a much greater breadth
range, from 83-93% with a mean of 89%. This compares with a range of 80-94%
and again a mean of 89% for 30 of the specimens which H. L. Clark named L.
dyscritum, suggesting that his figure of 90-95% was biased and probably derived
from only a few specimens. The mean periproctal distance also practically coincides
when these two series are compared, nor is there a significant difference in the relative
length of the petaloid area.

Laganid sp. juv.
K3 - i ; K8 - i.

Peronella lesueuri (Valenciennes in L. Agassiz, 1841)

See : H. L. Clark, 1946 : 346.
K8-i.

135

Peronella orbicularis (Leske, 1778)

See : H. L. Clark, 1946 : 346.

D2 - i ; 07 - 2 ; Di4 - i ; Di6 - i ; Dig - 2 ; 036 - i ; 048 - i ; 049 - i ;

D52-[2]; K2-2; K3 - i.

Peronella sp. juv.
DI-I.

SPATANGIDAE

Maretia planulata (Lamarck, 1816)

See : H. L. Clark, 1946 : 380 (as Maretia ovata] [non Spatangus ovatm Leske, 1778] ;

Endean, 1956 : 129 (as Maretia planulata}.

Low Is (Ls) - 15 ; Low Is (Anchorage at 3 m depth) - 2 ; 023 - i ; D26 - [i] ;
029-4; Ki-6; K3-8[6].

Like Laganum depressum this species is common in muddy sand close to the
Anchorage on Low Isles. During the period of exposure on day-time spring tides
it often comes to the surface of the sand. This habit has also been noted in Lovenia
elongata in the Red Sea and elsewhere.

LOVENIIDAE

Lovenia elongata (Gray, 1845)

See : H. L. Clark, 1946 : 381.
K7-i.

SCHIZASTERIDAE

Schizaster lacunosus (Linnaeus, 1758)

See : H. L. Clark, 1946 : 368.

Low Is (L2) - i. South of Arlington Reef at i6°42-5'S : i45°57-o'E (Grab ARL 16)

at 51 m depth - i. [Both from muddy sand.]
Previous records : Torres Strait ; Turtle Is (H. L. Clark, 1946).

BRISSIDAE

Brissus latecarinatus (Leske, 1778) (Fig. 3b)

See : H. L. Clark, 1946 : 375.

Pelican I. (strand line) - 4 (tests only) ; Dig - i.

One of the four specimens from Pelican Island agrees in some ways more with
Brissus agassizi Doderlein, 1885, having the posterior end of the test vertically
truncated, the posterior interambulacrum being only slightly carinate aborally and

136 GIBBS, CLARK & CLARK

FIG. 3. Profiles of (a) Brissus agassizi Doderlein and (b) B. latecarinatus.

not prolonged backwards to overhang the periproct and conceal it from dorsal
view (Fig. 3a), in contrast to B. latecarinatus (Fig. 3b). Endean (1956 : 129) records
as 'Brissus (? latecarinatus)' a dead test from Heron Island which approximates to
some extent with this specimen, lacking the overhang above the periproct, though
it does have the posterior end of the test 'obliquely truncate as in typical B.
latecarinatus'. In the absence of pedicellariae from this specimen it is impossible
to determine if it is specifically distinct, the globiferous ones of B. agassizi being
unusual in terminating in a slit bordered with long narrow teeth rather than a
rounded opening with numerous small teeth, as in B. latecarinatus.

Metalia spatagus (Linnaeus, 1758)

See : H. L. Clark, 1946 : 372.

Low Is (L6) - i.

Previous records : Torres Strait ; Low Is (H. L. Clark, 1946).

Metalia sternalis (Lamarck, 1816)

See : H. L. Clark, 1946 : 372.

Low Is : (L6) - i ; (Ly) - i ; 029 - [i].

These three specimens and the single Metalia spatagus were all living in muddy
sand.
Previous record : Low Is (H. L. Clark, 1946 ; Endean, 1957).

Class HOLOTHURIOIDEA

HOLOTHURIIDAE

Actinopyga Imiliaris (Quoy & Gaimard, 1833)

Bird I. (sand flat) - i ; Three Is (reef flat) - i ; Lark Passage (reef flat) - S.

The two specimens collected both have extraordinarily few spicules and their
provisional identification as A. miliaris is mainly because of the uniformly dark
colour.

ECHINODERMS FROM THE GREAT BARRIER REEF 137

Bohadschia argus Jaeger, 1833

See : H. L. Clark, 1946 : 425 (as Holothuria argus) ; A. M. Clark & Rowe, 1971 : 176

(as Bohadschia argus},
Three Is (reef flat) - i.

Bohadschia bivittata (Mitsukuri, 1912)

See : A. M. Clark & Rowe, 1971 : 194.
Two Is (2.4) - i.

This record provides an extension of range to Australian waters. The species
has previously been recorded from southern Japan to the Philippines, Caroline
Islands, Fiji, Samoa and the Bismarck Archipelago, north of New Guinea.

The single large individual was found almost buried below the surface of muddy
sand. After preservation in alcohol for about a year it still shows the very distinctive
colour pattern of Bohadschia bivittata with two broad dark brown transverse bands
across the otherwise pale yellow upper side. After contraction in preservation the
specimen measures 170 mm in length. The more anterior brown band is c. 30 mm
wide but is not sharply delimited anteriorly, the area up to and around the ring of
20 tentacles being brownish. The median pale 'saddle' is c. 25 mm wide and the
posterior brown band c. 40 mm wide, not sharply delimited posteriorly. The under-
side is uniformly pale.

H. L. Clark did not distinguish Bohadschia from Holothuria. In his key to
Holothuria (1946 : 422) this species, with its rosettes in the body wall, runs down
to section C, which includes Bohadschia argus marked with spots and B. marmorata
with irregular brown blotches on the upper side.

Bohadschia marmorata Jaeger, 1833

See : H. L. Clark, 1946 : 426 (as Holothuria marmorata) ; A. M. Clark & Rowe,

1971 : 176 (as Bohadschia marmorata).
Low Is (L3) - i ; Two Is (2.4) - i.

Holothuria (Acanthotrapeza) coluber Semper, 1868

See : H. L. Clark, 1946 : 428.
Three Is (reef flat) - i.

Holothuria (Cystipus) rigida (Selenka, 1867)

See : H. L. Clark, 1946 : 432.

Two Is (reef flat) - i.

Previous records : Murray Is ; Green I. (H. L. Clark, 1946 ; Endean, 1957),

138 GIBBS, CLARK & CLARK

Holothuria (Halodeima) atra Jaeger, 1833

See : H. L. Clark, 1946 : 427.

Mid Reef - S ; Low Is (L3) - 3 ; Three Is (reef flat) - 4[>] ; Three Is (3.5) - i ;
Lark Passage (reef flat) - i.

Holothuria (Lessonothuria) pardalis Selenka, 1867

See : H. L. Clark, 1946 : 437.
Three Is (reef flat) - 2.

Holothuria (Mertensiothuria) leucospilota (Brandt, 1835)

See : H. L. Clark, 1946 : 438.

Mid Reef - S ; Three Is (reef flat) - 4[i].

Holothuria (Mertensiothuria) ? leucospilota (Brandt)

See : H. L. Clark, 1921 : 179.

Low Is (L5) - [i] ; Three Is (3.5) - i.

When collected these two specimens were dark brown, as usual in H. leucospilota
but after preservation in alcohol they turned quite black, whereas the specimens
positively identified as H. leucospilota are purplish-brown. In addition, these two
were found buried on a sand flat, not on a reef flat where H. leucospilota is usually so
conspicuous stretching out from under stones.

However, H. L. Clark (1921) notes that older individuals of this species are much
more black in colour ; though very contracted now, the two specimens were almost
certainly distinctly larger than the rest in life. Their body walls, spicules and cal-
careous ring agree very well.

Holothuria (Metriatyla) scabra Jaeger, 1833

See : H. L. Clark, 1946 : 430.

Pipon Is (sand flat) - 6[i] ; Bird I. (sand flat) - 3.

Holothuria (Theelothuria) sp. juv.
032 - 1.

This small specimen has the tack-like spicules characteristic of the subgenus
Theelothuria but cannot be identified to species.

Holothuria (Thymiosycia) arenicola Semper, 1868 (PL i, fig. 4)

See : H. L. Clark, 1946 : 438.

Low Is (L4) - 5 ; Three Is (3.2 & 3.5) - 2 ; Two Is (2.2) - [i].

ECHINODERMS FROM THE GREAT BARRIER REEF 139

This species is one of the most conspicuous members of the infauna of the sand
flats, burrowing to a depth of about 25 cm. One specimen harboured the commensal
fish Carapus homei Richardson (see Mukerji, 1932; Jangoux, 1974).

Even when preserved in alcohol, these specimens show a conspicuously abrupt
change in colour at about 10-20 mm from the anus, from the yellowish-brown with
lines of small dark blotches, usually described for this species, to pure white, though
after two years this is much less distinct than in the photograph from life shown in
Plate i. Among the 55 specimens of H. arenicola preserved in the British Museum
collections, two show a distinctly lighter posterior end ; one was collected by
Gibbs in the Solomon Islands in 1965 and the other, from the 'Java Sea', was received
as long ago as 1889. As far as we can ascertain, this abrupt colour change to white
posteriorly has not been recorded in the literature.

Holothuria (Thymiosycia) impatiens (Forskaal, 1775)

See : H. L. Clark, 1946 : 434.
Three Is (reef flat) - 2.

Labidodemas semperianum Selenka, 1867

See : H. L. Clark, 1946 : 421.

Low Is (L3) - 2.

Previous records : Murray Is ; Capricorn Group (H. L. Clark, 1946).

STICHOPODIDAE

Stichopus chloronotus Brandt, 1835

See : H. L. Clark, 1946 : 417.

Mid Reef (reef flat) — S ; Low Is (L3) - I ; Three Is (reef flat) - S ; Lark Passage
(reef flat) - S ; Two Is (reef flat) - S.

Stichopus horrens Selenka, 1867

See : H. L. Clark, 1946 : 418.
Three Is (reef flat) - I.

Stichopus variegatus Semper, 1868

See : H. L. Clark, 1946 : 418.

Three Is (reef flat) - i ; Di7 - i ; K4 - i.

GUGUMARIIDAE

Havelockia sp. juv.
D3 - i ; D2i - i.

It is likely that this species would have been included in Thyone by H. L. Clark.

140 GIBBS, CLARK & CLARK

Pentacta australis (Ludwig, 1875)
See : H. L. Clark, 1946 : 392.

D2I -2.

Pentacta cucumis (Semper, 1868)

See : H. L. Clark, 1946 : 393.

07-1.

Previous records : Torres Strait (H. L. Clark, 1946) ; west of N Direction I. (H. L.

Clark, 1932) ; Coppersmith I. (south of Whitsunday Group) (Endean, 1956).

Thyone papuensis Theel, 1886

See : H. L. Clark, 1946 : 399.
K8-i.

PHYLLOPHORIDAE

Afrocucumis africana (Semper, 1868)

See : H. L. Clark, 1946 : 404 (as Discucumaria africana) ; Endean, 1956 : 130 (as
Discucumaria africana} ; A. M. Clark & Rowe, 1971 : 182 (as Afrocucumis africana}.
Three Is (reef flat) - [2] Three Is (3.7) - 3 ; Noble I. (reef flat) - I.

An additional specimen from the Three Islands reef flat lacks spicules in the body
wall and cannot be positively identified.

Actinocucumis typicus Ludwig, 1874

See : H. L. Clark, 1946 : 403.
D47-i.

Phyllophorus (Phyllothuria) cebuensis (Semper, 1868)

See : H. L. Clark, 1946 : 409.
D48-[i].

We are indebted to Dr F. W. E. Rowe of the Australian Museum for identifying
this very small specimen.

This record provides an extension of range to Queensland waters from Western
Australia, H. L. Clark (1938) having named as P. cebuensis a small specimen from
Dongarra, near Geraldton, though expressing a little doubt. Previous records are
from the Philippines (the type locality) and Indonesia.

SYNAPTIDAE
Leptosynapta latipatina H. L. Clark, 1921

See : H. L. Clark, 1946 : 454 (Leptosynapta parvipatina [lapsus]).
Low Is (Li) - i.

ECHINODERMS FROM THE GREAT BARRIER REEF 141

The single specimen was found among Mesochaetopterus tubes in muddy sand.
The species was previously known only from the incomplete holotype, taken at
Friday Island, Torres Strait (H. L. Clark, 1921).

Opheodesoma grisea (Semper, 1868)

See : H. L. Clark, 1946 : 449.
Low Is (Tripneustes spit) - i.

Rynkatorpa sp. nov. (to be described by F. W. E. Rowe; see p. 108)
D33 - 2 pieces [z] (very small).

Synaptula recta (Semper, 1868)

See : H. L. Clark, 1946 : 453 (as Chondrocloea recta) ; A. M. Clark & Rowe, 1971 : 188

(as Synaptula recta}.
Beesley I. (sand flat) - I.

CHIRIDOTIDAE

Chiridota rigida (Semper, 1868)

See : H. L. Clark, 1946 : 457.
Low Is: (L3)-3; (L7) - [i].

ACKNOWLEDGEMENTS

One of us (P. E. G.) would like to thank the Royal Society for the opportunity to
participate in the Expedition to the Great Barrier Reef ; and also Dr D. R.
Stoddart (Leader) and other Expedition members, particularly Dr R. F. McLean,
Dr J. E. N. Veron and Mr L. Zell, for much assistance in the field. The skilful help
of the Masters and crews of the M.V. Calypso (J. Lawlor) and R.V. James Kirby
(D. Duncan) in carrying out the dredge survey, often under difficult conditions, is
gratefully acknowledged.

TABLE 4

List of some associates of these echinoderms, collected simultaneously

HOST ASSOCIATE

CRINOIDEA

Comantheria briar eus Paradyte crinoidicola (Potts) Polychaete

Myzostomum sp. Myzostome

Comantheria or Comanthus sp. Myzostomum sp. Myzostome

Comanthina schlegeli Paradyte crinoidicola Polychaete

Myzostomum sp. Myzostome

Comanthus bennetti Paradyte crinoidicola Polychaete

Galathea elegans Adams & White Decapod

142

GIBBS, CLARK & CLARK

HOST
Comanthus parvicirrus

Comanthus samoanus

Comatella stelligera

Comatula rotalaria
Himerometra robustipinna

Colobometra perspinosa
Tropiometra afra

ASTEROIDEA

At -chaster typicus
Pentacer aster regulus

OPHIUROIDEA

Ophiarthrum elegans
Ophiocoma dentata
Macrophiothrix belli

HOLOTHURIOIDEA

Actinopyga sp.
Bohadschia argus
Holothuria coluber
Holothuria atra
Holothuria leucospilota
Holothuria arenicola
Stichopus chloronotus
Chiridota rigida

TABLE 4 (cont.)

ASSOCIATE

Myzostomum polycyclus Atkins Myzostome

Myzostomum sp. Myzostome

Paradyte crinoidicola Polychaete

Myzostomum sp. Myzostome

Galathea australiensis Stimpson Decapod

Myzostomum sp. Myzostome

Eulimid sp. Gastropod

Ophiomaza cacaotica Lyman Ophiuroid

Paradyte crinoidicola Polychaete

Myzostomum costatum Leuckart Myzostome

Myzostomum crosslandi Boulenger Myzostome

Galathea elegans Decapod

Paradyte crinoidicola Polychaete

Paradyte crinoidicola Polychaete

Galathea elegans Decapod

Ophiodromus sp. Polychaete

Hololepidella nigropunctata (Horst) Polychaete

Hololepidella nigropunctata Polychaete

Hololepidella nigropunctata Polychaete

Hololepidella nigropunctata Polychaete

Gastrolepidia clavigera Schmarda Polychaete

Gastrolepidia clavigera Polychaete

Gastrolepidia clavigera Polychaete

Gastrolepidia clavigera Polychaete

Gastrolepidia clavigera Polychaete

Carapus homei Richardson Fish

Gastrolepidia clavigera Polychaete

'Leptonid' sp. Bivalve

REFERENCES

AGASSIZ, A. 1881. Echinoidea. Rep. sclent. Results Voy. Challenger (Zool.) 3 : 1-321, 45 pis.
BELL, F. J. 1884. Echinodermata. In Coppinger, R. W. Report on the zoological collections

made in the Indo-Pacific Ocean during the voyage of H. M.S. 'Alert' , 1881-2. London : 117-

i?7. 509-5*2, pis 8-17, 45.
CLARK, A. M. 1970. Notes on the family Amphiuridae. Bull. Br. Mus. not. Hist. (Zool.)

19: 1-81.

1974- Notes on some echinoderms from southern Africa. Bull. Br. Mus. nat. Hist.

(Zool.) 26 (6) : 421-487, 3 pis.

1975- The Swain Reefs Expedition : Crinoidea. Rec. Aust. Mus. 29 : 391-406.

& ROWE, F. W. E. 1971. Monograph of shallow-water Indo-West Pacific echinoderms.

London. 238 pp., 100 figs, 31 pis.

ECHINODERMS FROM THE GREAT BARRIER REEF 143

CLARK, H. L. 1915. The comatulids of Torres Strait with special reference to their habits
and reactions. Pap. Dep. mar. Biol. Carnegie Instn Wash. no. 8 : 67-125.

1921. The echinoderm fauna of Torres Strait : its composition and its origin. Pap.

Dep. mar. Biol. Carnegie Instn Wash. no. 10 : 1-223, 3^ pis.

1925. A catalogue of the recent sea-urchins (Echinoidea) in the collection of the British

Museum (Natural History}. London: xxviii + 25o, 12 pis.

1926. Notes on a collection of echinoderms from the Australian Museum. Rec. Aust.

Mus. 15 : 183-192.

1932. Echinodermata (other than Asteroidea). Scient. Rep. Gt Barrier Reef Exped.

4 : 197-239, i pi.

1938. Echinoderms from Australia. Mem. Mus. comp. Zool. Harv. 55 : 1-596, 63 figs,

28 pis.

1946. The echinoderm fauna of Australia : its composition and its origin. Publs

Carnegie Instn, no. 566 : 1-567.

DALL, W. & STEPHENSON, W. 1953. A bibliography of the marine invertebrates of Queensland.

Pap. Dep. Zool. Univ. Qd 1 : 21-49.
DARTNALL, A. J. 1971. Australian sea-stars of the genus Patiriella (Asteroidea, Asterinidae) .

Proc. Linn. Soc. N.S.W. 96 : 39-49, i fig., pis 3, 4.
DEVANEY, D. M. 1974. Shallow-water Asterozoans of southeastern Polynesia. 2. Ophiu-

roidea. Micronesica 1C : 105-214.
ENDEAN, R. 1953. Queensland faunistic records. 3. Echinodermata (excluding Crinoidea).

Pap. Dep. Zool. Univ. Qd 1 : 53-60.

1956. Queensland faunistic records. 4. Further records of Echinodermata (excluding

Crinoidea). Pap. Dep. Zool. Univ. Qd 1 : 123-140.

1957- The biogeography of Queensland's shallow-water echinoderm fauna (excluding

Crinoidea), with a rearrangement of the faunistic provinces of tropical Australia. Aust. J.
Mar. Freshwat. Res. 8 : 233-273.

1961. Queensland faunistic records. 7. Additional records of Echinodermata (excluding

Crinoidea). Pap. Dep. Zool. Univ. Qd 1 : 289-298.

1965. Queensland faunistic records. 8. Further records of Echinodermata (excluding

Crinoidea) from southern Queensland. Pap. Dep. Zool. Univ. Qd 2 : 229-235.

KENNY, R. & STEPHENSON, 1956. The ecology and distribution of intertidal organisms

on the rocky shores of the Queensland mainland. Aust. J. Mar. Freshwat. Res. 7 : 88-146.

FORSTER, G. R. 1953. A new dredge for collecting burrowing animals. /. mar. biol. Ass.

U.K. 32: 193-198.
GIBBS, P. E. 1976. The macrofauna of the intertidal sand flats on low wooded islands in the

northern region of the Great Barrier Reef Province. Phil. Trans R. Soc. B (in press).
JANGOUX, M. 1974. Sur 1" 'association" entre certaines Asteries (Echinodermata) et des

Poissons Carapidae. Revue zool. afr. 88 : 789-796.
KOEHLER, R. 1905. Ophiures de 1'Expedition du Siboga. 2. Ophiures littorales. Siboga

Exped. 45b : 1-142, 18 pis.

1907. Revision de la collection des Ophiures du Museum d'Histoire Naturelle de Paris.

Bull. sci. Fr. Belg. 41 : 279-351, 5 pis.

1927. Ophiures recueillis aux lies Gilbert, Marshall et Fiji. Goteborgs K. Vetensk.- o

vitterhSamh. Handl. 33 (3) : 1-13, i pi.

LIVINGSTONE, A. A. 1930. On some new and little-known Australian asteroids. Rec. Aust.
Mus. 18 : 15-24, 5 pis.

1932. Asteroidea. Scient. Rep. Gt Barrier Reef Exped. 4 : 241-265, 2 figs, 12 pis.

McNeiLL, F. A. & LIVINGSTONE, A. A. 1926. A supplementary list of the echinoderms col-
lected by Surgeon Lieutenant-Commander W. E. J. Paradice, R.A.N., in Queensland and
North Australia. Rec. Aust. Mus. 15 : 193-199.

MORTENSEN, T. 1943. Monograph of the Echinoidea. 3(2). Camarodonta I. Copenhagen.
533 PP-, 321 figs, 56 pis.

I44 GIBBS, CLARK & CLARK

MORTENSEN, T. 1948. Monograph of the Echinoidea. 4 (2). Clypeastroida. Copenhagen.

471 pp., 256 figs, 72 pis.
MUKERJI, D. D. 1932. Biological observations on and instances of commensalism of an

ophioid fish with echinoderms from the Andaman Islands. Rec. Indian Mus. 34 : 567-569.
POTTS, F. A. 1915. The fauna associated with the crinoids of a tropical coral reef : with

special reference to its colour variations. Publs Carnegie Instn no. 212 : 71-96.
STEPHENSON, W., ENDEAN, R. & BENNETT, I. 1958. An ecological survey of the marine

fauna of Low Is., Queensland. Aust. J. Mar. Freshwat. Res. 9 : 261-318.
TENISON-WOODS, J. E. i88oa. On some of the littoral marine fauna of northeast Australia.

Proc. Linn. Soc. N.S.W. 5 : 106-131. [Echinoderms pp. 125-131.]

i88ob. On the habits of some Australian echini. Proc. Linn. Soc. N.S.W. 5 : 193-204.

WHITLEY, G. 1932. Marine zoogeographical regions of Australia. Aust. nat. 8 : 166.

DR P. E. GIBBS

MARINE BIOLOGICAL ASSOCIATION OF THE U. K.

PLYMOUTH

AILSA M. CLARK

CHRISTINE M. CLARK

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON SW7 5BD

PLATE i

Amphiura phrixocantha sp. nov. Holotype Top. Dorsal view ( x i).

Centre, left and right. Dorsal and ventral views ( x 4).

Bottom. Holothuria (Thymioscyia) arenicola Semper, photographed in life ( x

Bull. Br. Mus. nat. Hist. (Zool.) 30, 4

PLATE i

A LIST OF SUPPLEMENTS
TO THE ZOOLOGICAL SERIES

OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

1. KAY, E. ALISON. Marine Molluscs in the Cuming Collection British Museum
(Natural History) described by William Harper Pease. Pp. 96 ; 14 Plates.
1965. (Out of Print.)

2. WHITEHEAD, P. J. P. The Clupeoid Fishes described by Lacepede, Cuvier and
Valenciennes. Pp. 180 ; n Plates, 15 Text-figures. 1967. £4.80

3. TAYLOR, J. D., KENNEDY, W. J. & HALL, A. The Shell Structure and
Mineralogy of the Bivalvia. Introduction. Nuculacea-Trigonacea. Pp. 125 ;
29 Plates, 77 Text-figures. 1969. £5.40.

4. HAYNES, J. R. Cardigan Bay Recent Foraminifera (Cruises of the R.V. Antur]
1962-1964. Pp. 245 ; 33 Plates, 47 Text-figures. 1973. £11.90.

5. WHITEHEAD, P. J. P. The Clupeoid Fishes of the Guianas. Pp. 227 ; 72
Text-figures. 1973. £10.65.

6. GREENWOOD, P. H. The Cichlid Fishes of Lake Victoria, East Africa : the
Biology and Evolution of a Species Flock. Pp. 134 ; i Plate, 77 Text-figures.
1974. £3.75. Hardback edition £6.

Printed in Great Britain by John Wright and Sons Ltd, at The Stonebridge Press, Bristol BS4 sNU

270CT197<
\.

MISCELLANEA

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. 5

LONDON: 1976

MISCELLANEA

Pp 145-203 ; 7 Plates ; 25 Text-figures

GENERAL

270CT1976

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)

ZOOLOGY Vol. 30, No. 5

LONDON: 1976

THE BULLETIN OF THE BRITISH MUSEUM

(NATURAL HISTORY), instituted in 1949, is
issued in five series corresponding to the Scientific
Departments of the Museum, and an Historical
series.

Parts will appear at irregular intervals as they
become ready. Volumes will contain about three or
four hundred pages, and will not necessarily be
completed within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 30 No. 5 of the Zoology series.
The abbreviated titles of periodicals cited follow those
of the World List of Scientific Periodicals.

World List abbreviation
Bull. Br. Mus. not. Hist. (Zool.).

ISSN 0007-1498

Trustees of the British Museum (Natural History), 1976

BRITISH MUSEUM (NATURAL HISTORY)

Issued 28 October, 1976 Price £5-35

CONTENTS

Page

Tilapiine fishes from crater-lakes north of Lake Malawi. By E. TREWAVAS 149

A new species of Hemichromis (Pisces, Cichlidae) of Sierra Leone and Liberia.
By A. I. PAYNE & E. TREWAVAS 159

A possible intergeneric cyprinid hybrid from Lake Tanganyika. By K. E.
BANISTER 171

Two new species of large Barbus (Pisces, Cyprinidae) from central Africa.
By K. E. BANISTER 191

7*

TILAPIINE FISHES FROM CRATER-LAKES
NORTH OF LAKE MALAWI

By E. TREWAVAS

SYNOPSIS

Tilapia chungruruensis Ahl of a crater-lake in the Rungwe Mountains, about 32 km north
of Lake Malawi, is redescribed from the holotype and ten of the 'cotypes' (paratypes). The
population is related to the Sarotherodon squamipinnis group of Lake Malawi. The holotype
cannot be assigned to any of the four species of this group, nor to the related species of Lakes
Mweru and Rukwa, and consequently the name of neither of these can be regarded as its
synonym. It may represent the result of hybridization, although this cannot be definitely
assumed. Some of the 'cotypes', however, are probably 5. lidole (Trewavas) modified by the
environment.

Changes in the level of Lake Malawi may have permitted the natural populating of the
crater-lake or it may have been stocked by man. All the specimens have relatively bigger
heads than their Lake Malawi relatives and are sexually mature at a much smaller size. A
comparison is drawn with the Sarotherodon species of a crater-lake in West Cameroon and with
isolated populations of other species.

Lake Tschungruru is probably the same lake as that now known as Masoko. Two specimens
of 5. lidole from Lake Kingiri, a crater-lake nearer to Lake Malawi, are also described.

INTRODUCTION

THE main object of this paper is to redescribe the types of Tilapia chungruruensis
Ahl, 1924, and to relate them to other tilapiine fishes of the genus Sarotherodon
(formerly subgenus) to which they belong. I have also considered two specimens
of this genus collected by Dr H. Albrecht in Lake Kingiri, north of Lake Malawi,
and some juveniles collected in Lake Masoko by Miss C. K. Ricardo (later Dr
Ricardo-Bertram) in 1937.

I have been privileged to examine the holotype and eleven of the twenty-nine
'cotypes' (paratypes) of T. chungruruensis, thanks to the kindness of Dr K. Deckert,
who was in charge of the fish collections in the Berlin Museum at the time of the loan.

The meristic characters, the black colour and tasselled genital papilla of the males,
the long pectoral fin, the long, slender caudal peduncle and the shape of the lower
pharyngeal bone collectively relate these fishes to the Malawi species-flock of the
5. squamipinnis group and distinguish them from 5. rukwaensis and S. macrochir,
tasselled Sarotherodon of neighbouring basins. At the time of Ahl's description and
Ricardo's determination the four species of the Malawi group had not been
distinguished. On the same morphological criteria used for the Malawi group
(Trewavas, 1941 ; Lowe, 1952) I find that the holotype and two of the paratypes
of 5. chungruruensis belong to one species, not exactly identifiable with either of the
Lake Malawi species, eight others to a second species which I identify with S. lidole.
(The eleventh 'cotype' is a juvenile of 42 mm SL, too small to be determined.) I

Bull. BY. Mus. nat. Hist. (Zool.) 30, 5 Issued 28 October 1976

150 E. TREWAVAS

therefore describe them under two headings, grouping the Lake Masoko juveniles
tentatively with 5. chungruruensis and the Lake Kingiri specimens with 5. lidole.

DESCRIPTION OF THE LOCALITIES

Ahl (1924) quoted Fiilleborn's account of the Tschungruru-See. He stated that
it is the Kiungvuvu of Cross, near the mission station of Neu-Wangermannshohe.
This name is not found on recent maps, but in the Atlas of Stuhlmann's Deutsch
Ost-Afrika it is marked at 9°i7'S, 33°5o'E, north of Lake Malawi but within its
drainage basin. The crater-lake Masoko lies a few kilometres NW of this at about
9°i6'S, 33°43'E and is probably the same lake under another name.

Fiilleborn described Lake Tschungruru as a crater-lake about 500 m in diameter
and 45 m deep with no surface outlet. The water did not taste brackish.

Lake Kingiri is a crater-lake about 8 km north of Lake Malawi and 3 km north
of Ipinda Ferry.

DESCRIPTION OF SPECIES

Sarotherodon chungruruensis (Ahl)

Tilapia chungruruensis Ahl, 1924 : 86.

Tilapia squamipinnis (part., nee Giinther ?) ; Ricardo, 1939 : 655.

TYPE MATERIAL. Holotype : the 'type' by original designation is the largest
specimen, a mature male of total length 193 mm (153 + 40 mm), one of thirty
numbered 208 in the Zoological Museum of the Humboldt University, Berlin,
collected by Professor Fiilleborn about 1923 in Lake Tschungruru, a crater-lake in
the Rungwe Mountains, north of Lake Malawi.

Lectoparatypes i1 two specimens, respectively 51 and 69 mm in SL. I assign
the eight other 'cotypes' examined to 5. lidole (p. 151).

DESCRIPTION. Some of the variation is consistent with the usual allometries, so
that the highest ratios for depth of body, length of snout, depth of preorbital,
interorbital width and length of jaw are those of the holotype. So also is that of
head-length, a less usual trend, but found also in some other phytoplankton feeders
(S. linnellii and 5. caroli of Lake Barombi-Mbo, Cameroons), and the length of the
pectoral fin. The eye, as usual, is negatively allometric.

Proportions as % SL : depth of body 37-6-38-5 ; length of head 37-3-40-5 ;
length of pectoral fin 35-42 ; length of caudal peduncle 13-8-16-0 (1-15-1 -3 times its
depth).

Proportions as % length of head : snout 29-38 ; eye 29-0-20-5 ; depth of preorbital
16-8-24-2 ; interorbital width 31-6-39-7 ; length of lower jaw 32, 32-8 in the small
specimens, 39 in the holotype.

Teeth in 3-4 series in upper jaw, 4-5 in lower ; 42-78 in outer row of upper
jaw ; with slender shafts and curved crowns, the outer bicuspid in the young,

1 Since the holotype is not a lectotype I think that this, and not 'paralectotypes' is the correct title
for those 'cotypes' that I recognize as conspecific with the holotype.

TILAPIINE FISHES OF CRATER LAKES 151

mixed bi- and tri-cuspid in the holotype, which has many of the crowns broken off ;
inner tricuspid, but also many of these are broken in the holotype.

Gill-rakers on first arch 2 + 1 + 17.

Lower pharyngeal bone : see Tables I and 2 and PI. 3.

Scales in 2 or 3 series on the cheek ; 32 in the lateral line series, 4 or 4^ from origin
of dorsal to lateral line.

Dorsal XIV 12 (holotype), XV n or XV 12. Anal III 10-11. Caudal truncate
or very slightly emarginate, not densely scaled, but with small scales extending
part-way along upper and lower rays.

Genital papilla small in young ; in the holotype with several short branches
arising in two groups from a club-shaped base.

Vertebrae 30 (14 + 16) in the holotype and one other.

Colour. The holotype, a ripening male, is dark all over, including the fins. The
colour is now dark brown, probably faded from black. Even the dorsal lappets are
now dark. The 69 mm specimen is pale, without markings, the vertical fins faintly
dusky. In the 51 mm specimen the dorsal fin and back are dark ; along the middle
of the flank are four dark spots produced vertically.

ADDITIONAL SPECIMENS. Four young of 35-5-61-0 mm SL were collected in
crater-lake Masoko in 1937 by Miss C. K. Ricardo (later Dr Ricardo-Bertram).
These also belong to the 5. squamipinnis group and both jaws and pharynx are too
well toothed for 5. lidole. There are 32 or 33 scales in the lateral line series, 8 or 9
soft rays in the anal fin, 17 or 18 gill-rakers on the lower part of the first arch. The
dorsal formula is XV 10 (f.i), XV n (f.2) or XVI 10 (f.i). The three of 56-61 mm
have 4-5 rows of teeth in the upper jaw, 5-6 in the lower ; the length of head is
36% SL. There are 6 dark vertical bars on the body, accentuated above the lateral
line.

We have not so far succeeded in distinguishing between the four Malawi species
at such a small size and in placing these in 5. chungruruensis. I am merely indicat-
ing that like the types they are not 5. lidole and may be one of the other species
or the result of hybridization between some of them.

Sar other odon lidole (Trewavas)

Tilapia lidole Trewavas, 1941 : 297 ; Bertram, Borley & Trewavas, 1942 : 34, fig. 46 & D ;
fig. 6C ; Lowe, 1952 : 2-65.

DESCRIPTION (of eight specimens from Lake Tschungruru and, in brackets, two
from Lake Kingiri).

The Tschungruru specimens are eight of the paratypes of T. chungruruensis,
nos 2 and 4-10 in Table i. The two Kingiri fishes measure 139 + 31-5 and 162 +
39 mm to the end of the middle caudal rays ; they are listed at the end of Table i.

Proportions as % SL : depth of body 34-8-40-0 (40-5, 38-4) ; length of head
39-4-42-4 (38-5, 38-6) ; length of pectoral fin 38-0-43-3 (41-5, 40-0) ; length of caudal
peduncle 15-3-16-8 (16-3, 15-7) ; length of lower jaw 11-6-14-8.

Depth of caudal peduncle contained 1-27-1-5 times in its length.

152 E. TREWAVAS

Proportions as % length of head : length of snout 34-2-39-0 (32-6, 26-8) ; diameter
of eye 21-3-24-4 (24-2, 21-0) with no evidence of allometry ; depth of preorbital
18-7-23-8 (22-4) ; interorbital width 34-0-44-5 (45-2, 44-7) ; length of lower jaw
27-8-39-6 (31-8-32-3).

Proportions of lower pharyngeal bone : see Tables i and 2. Pharyngeal teeth
very fine, in a restricted area.

Teeth in 3 rows in both jaws with an incomplete fourth in a few ; with long,
slender shafts and expanded crowns ; those of outer row bicuspid in small specimens,
in some bigger fishes with an admixture of tricuspid and occasionally a few unicuspid,
often but not always by wear ; a few posterior often simple. Teeth of inner rows
tricuspid unless with worn crowns. Number of outer teeth in upper jaw 52 at
SL 65-5 mm, 70-82 from 86 to 162 mm.

Gill-rakers 2-4 + 1 + 17-20 on first arch.

Scales on cheek in 2-3 rows, occasionally a few small scales of a fourth row ;
lateral line series 33-34 ; 4, 4| or 5 from origin of dorsal fin to lateral line, about
20 around caudal peduncle.

Dorsal XIV-XV 10-11 (XVI n, XVII 10 in Lake Kingiri) ; anal III 8-10.

Vertebrae 31 in three counted.

Genital papilla of a male of SL 143 mm large, with two groups of short papillae
at its distal end, the pore between them. A female of SL 123 mm has enlarged
ovaries, in which the biggest egg is 3 mm in long diameter.

TABLE i

Data from the holotype and ten paratypes of Sarotherodon chungruruensis (Ahl) in order

of size

No. of

SL

Head

Ph.l.

Ph.w.

Rows teeth in

specimen

(mm)

%SL

% head

% head

Ph.l/w

Ph.bl./dent.

lower jaw

1-

51

373

316

316

10

13

4-5

2.

65-5

40-7

36-3

26-2

!-4

1-8

3

3-

690

37-7

346

32-7

1-06

1-2

4

4-

86-0

42-0

36-4

27-8

i'3

2'5

3

5-

93'5

40-6

37-o

26-3

i'4

2-06

3

6.

123-0

39'4

37'5

26-0

J-4

3'4

3

7-

125-0

42-4

39-6

27-0

1-47

3-88

3

8.

126-0

40-8

34-o

25-6

1-32

3'4

3

9-

141-0

40-4

38-0

25-0

i-5

3-0

3-4

10.

143-0

39-6

40-7

28-6

1-42

2-7

3

11.

1540

425

35-5

28-6

124

14

4-5

LAKE KINGIRI

139-0

38-5

37'6

27-2

1-38

1-88

3

162-0

38-6

36-0

26-2

i'37

2-2

3

No. ii is the holotype and nos i and 3 resemble it in the proportions of the lower pharyngeal bone
and number of rows of teeth in the jaws ; data for these are in bold type. The other specimens are
determined as 5. lidole. The measurements of the lower pharyngeal bone are its median length (Ph.l.),
greatest width (Ph.w.), both expressed as % length of head ; the median length divided by the width
(Ph.l/w.) and the length of the anterior blade divided by the median length of the dentigerous area
(Ph.bl./dent.). For comparison the same data are given for two specimens of T. lidole from Lake
Kingiri.

TILAPIINE FISHES OF CRATER LAKES 153

TABLE 2
Summary of Table i

5. chungruruensis Other 'cotypes', Two

holotype and lectoparatypes determined as S. lidole S. lidole from

(Specimens n, i and 3) (Specimens 2, 4-10) Lake Kingiri

Ph.l. % head 3i-6-35'5 34'°-4°-7 36-0,37-6

Ph.w. % head 28-6-32-7 25-6-28-6 26-2, 27-2

Ph.l./w. 1-0-1-24 I-3-I-5 x>37. 1*38

Ph.l./dent. 2-2-2-4 2-8-4-8 2-8,3-2

TABLE 3
Pharyngeal proportions in Lake Malawi

S. lidole S. squamipinnis and S. saka S. karongae

N 12 5 4

SL (mm) 230-285 200-250 150-235

Phar.l./w. 1-33-1-47 1-22-1-38 1-29-1-35

Phar.l./l.dent. 3-0-4-5 2-4-3-1 2-16-2-25

Colour. Small specimens and females with five dark blotches, vertically drawn
out, along the side and some with two fainter on the caudal peduncle ; vaguer
markings in a more dorsal row ; one male almost uniformly dark. Males with broad
dark margin on dorsal fin and dark anal fin, in one with narrow white margin (as
preserved). Females with no marginal dorsal band. Smaller specimens with a
tilapia-mark.

Caudal clear in small specimens ; in mature males of Lake Tschungruru dark or
dusky with a white margin or upper corner. Male of Lake Kingiri with caudal
dark proximally, distally with a dark reticulum and a pale margin.

COMPARISON WITH SPECIES OF LAKE MALAWI

S. lidole

i. Resemblances

I have identified eight Tschungruru specimens with this species because of the
proportions of the pharyngeal, the narrow bands of teeth in the jaws and the general
shape.

(a) Pharyngeal. In adults of Lake Malawi, all bigger than the crater-lake
specimens, the median length of the pharyngeal bone is 1-38-1-5 times its greatest
width. In the crater-lake the ratio is 1-3-1-5 (Tables i and 2), including specimens
down to 65-5 mm SL, thus smaller than the sizes at which the species has been
recognized in Lake Malawi. The dentigerous area is restricted and the blade is
long, also characters distinguishing 5. lidole.

(b) Jaw teeth. The inner teeth of the jaws are set in irregular rows and the
formula '3-4', including the outer row, expresses this irregularity rather than
variation within the size-range of the crater-lake sample. In Lake Malawi, even in
the much bigger adults, 3-5 is the range, in contrast to 4-7 in S. saka, S. squamipinnis
and 5. karongae.

154 E- TREWAVAS

(c) Plates I and 2 show the characteristic position of the greatest depth of body
in front of the dorsal fin in contrast to the other species.

2. Differences from the Malawi population

(a) Length of head. 5. lidole in Lake Malawi is credited by the fishermen with a
big head and the ratio there shows a higher range, though with a wide overlap, in
comparison with that of the other species. It is 34'0-38'5% SL in Lake Malawi in
contrast to 39-4-42-4 in the crater-lake. It is true that the Malawi specimens are
bigger, but smaller Malawi fishes of indeterminate species, probably including
5. lidole, have head ratios in the lower part of the range. The biggest ratios are
found in much bigger, mature fishes. The significance of this will be discussed below.

(b) Size at sexual maturity. Lowe (1952 : 24) gives the minimum size at sexual
maturity for both males and females of S. lidole as 28 cm TL (about 240 mm SL) .
In contrast, the crater-lake male of SL 143 mm (TL about 17-5 cm) already has an
enlarged genital papilla terminating in a short tassel, and a female of SL 123 mm
(TL 15 cm) has an ovarian egg of 3 mm long diameter.

The small size and early age of sexual maturity in pond populations of other
species of Sarotherodon are well known (Lowe, 1955 : 365 ; 1958 ; Hickling, 1962,

1963)-

In the other crater-lake, Kingiri, the two available specimens have a head ratio
at the upper end of the Malawi range. Their gonads were removed, but a short
tassel on the papilla of the male shows that here too sexual maturity occurs at a
smaller size than in the Great Lake.

S. chungruruensis

The holotype contrasts with the 5. lidole of the same crater-lake in the shape of
the pharyngeal bone as expressed in its length : width ratio as well as the bigger
dentigerous area relative to the length of blade (PI. 3 and Table 2). It also has more
rows of teeth in the jaws and that this is not only because of its greater size is
confirmed by the small lectoparatypes, which also combine a wider tooth-band with
a general shape contrasting with that of S. lidole (Pis I and 2).

The pharyngeal bone of the holotype contrasts more with that of the S. lidole of
Lake Tschungruru than with those of Lake Malawi as figured by Bertram et al.
(1942, fig. 6C). Column B of that figure, labelled T. squamipinnis , includes pharyn-
geals also from S. saka (Lowe) (at that time not recognized as distinct), but from
what we now know of S. saka these would be the pharyngeals with the broadest
dentigerous area, and those with a concave-sided dentigerous area would be S.
squamipinnis. It is the latter that most resemble the pharyngeals of 5.
chungruruensis, but the breeding male of 5. squamipinnis is pale blue with a white
head, that of 5. saka black, as S. chungruruensis must have been.

The character of the pharyngeal bone may well have changed in the environment
of the crater-lake, so this is not a final argument for not identifying 5. chungruruensis
with S. saka. The holotype (and indeed some of the others) may have been the
result of hybridization between, for example, 5. saka and 5. lidole.

TILAPIINE FISHES OF CRATER LAKES 155

SIGNIFICANCE OF THE BIG HEAD

Not only is there a high head : body ratio in the S. lidole of the crater-lake, but
the same ratio in the type and lectoparatypes of T. chungruruensis is in the highest
part of the range for all the species of this group in Lake Malawi.

Big heads have been found in small populations of other species of Sar other odon.
Two lagoon populations of the type species, 5. melanotheron , have head ratios
ranging higher than the range for the rest of this widespread species. The type
specimens of S. niloticus eduardianus occupied a small crater-lake on the lower
slopes of Ruwenzori, near enough to the shore of Lake George to have received its
fish population from that lake during an exceptionally wet season. They had
evidently become more crowded as time went on and when caught were stranded on
the shore of the shrinking lake. Their head-ratios ranged from 36-6 to 39-8% SL,
nearly all greater than in the parent population. The 5. niloticus in Buhuku Lagoon
when it was temporarily isolated from Lake Albert (see Lowe, 1958) had head-lengths
ranging from 36-4 to 41-0% SL in contrast to 34-3 to 39-6% in a sample of comparable
size from the open lake. The S. niloticus vulcani of Crater-lake A on the central
island of Lake Rudolf provide a further example. Most of the endemic cichlid
species of the Cameroons crater-lake, Barombi Mbo, are characterized by big heads,
and S. linnellii and 5. caroli provide an interesting parallel to 5. lidole and the rest
of the Malawi species-flock, both structurally and ecologically (Trewavas et al.,
1972).

The relative growth phenomenon resulting in a big head is known in pisciculture
in the 'hunger-form' of the common carp.

THE ORIGIN OF THE TSCHUNGRURU (? = MASOKO) FISH POPULATION

Dixey (1926) gives the evidence for previous surface levels of Lake Malawi. The
level has been falling since the Tertiary, for which epoch two levels have been deter-
mined, respectively 300 m and 210 m above the present level. The fall is attributed
mainly to subsidence of the rift floor. The Tertiary lake consisted only of the
northern part, which extended from about 32 km to the north-west of the present
northern end to about the latitude of Deep Bay (io°24'S). During the Quaternary
the lake extended southwards and when it was 120 m above its present level still
included an area 80 km north of the present north end.

Just where the limit was at the time, relatively recent, of the volcanic activity
that produced the Rungwe Mountains, including the Masoko crater, is not certain,
but it seems quite possible that it was near enough to be supplied naturally with its
fish fauna from Lake Malawi or one of its affluent rivers.

If it was stocked by man the source would be the same, namely the fry that abound
among the grasses and reeds of the banks ; they might have survived a journey of at
the most 30 km in a calabash or a cool clay pot.

Lowe (1952, 1953) has described the breeding places, seasons and colours that
characterize the three tasselled tilapias of the Malawi flock at the south end of the
lake. Her stay of one month (March) at Karonga near the north end was not long
enough to establish such distinctions, and she found there some specimens that

156 E. TREWAVAS

seemed 'equivalent to T. squamipinnis and T. saka at the south end of the lake',
that is to a single representative of these two species, as well as some that could be
assigned to each of them. The northern species, S. karongae, was also present and
there was evidence that S. lidole was caught at a different time of the year (October).

None of the examined types of S. chungruruensis has a pharyngeal bone big
enough for 5. karongae so that the source-species must be 5. lidole and one or both
of the other two.

The seasons and topography in the north of Malawi do not provide the same
framework as that within which the three southern forms preserve their identity.
Still more is this true of Lake Masoko and although S. lidole has kept its chara~ter-
istics with but slight modification, I find it impossible to be sure that S.
chungruruensis is identical with either 5. squamipinnis or S. saka, or to justify the
synonymizing of its name with either the older 5. squamipinnis or the junior S. saka.
It seems more expedient to retain the name chungruruensis for the section of the
crater-lake population that may be a hybrid between two or more of the southern
species or a modified population of 5. saka, and to refrain from suppressing the latter
name in favour of that of such a doubtful form.

A fresh examination of the Lake Masoko population and its ecology with these
possibilities in mind would be interesting. The specimens collected by Ricardo
in 1937 are too young to provide evidence, but they do show that fish were still in
the lake at that time.

REFERENCES

AHL, E. 1924. t)ber einen neuen Cichliden aus Ost-Afrika. Zool. Anz. 59 : 86-87.
BERTRAM, C. K. RICARDO, BORLEY, H. J. H. & TREWAVAS, E. 1942. Report on the fish and

fisheries of Lake Nyasa. London, Crown Agents.
DIXEY, F. 1926. The Nyasaland section of the Great Rift Valley. Geogr. J. London. 68 : 117-

140.
HICKLING, C. F. 1962. Fish culture. London.

1963. The cultivation of Tilapia. Scient. American 208 : 143-152.

LOWE, R. H. 1952. Report on the Tilapia and other fish and fisheries of Lake Nyasa, 1945-47.
London, H. M. Stationery Office.

1953. Notes on the ecology and evolution of Nyasa fishes of the genus Tilapia, with a

description of T. saka Lowe. Proc. zool. Soc. Lond. 122 : 1035-1041, i pi.

1955. New species of Tilapia (Pisces, Cichlidae) from Lake Jipe and the Pangani River,

East Africa. Bull. Br. Mus. nat. Hist. (Zool.) 2 : 349-368, pis 13-17.

1958. Observations on the biology of Tilapia nilotica Linne in East African waters.

Rev. Zool. Bot. afr. 57 : 129-170.
RICARDO, C. K. 1939. The fishes of Lake Rukwa. /. Linn. Soc. Lond. Zool. 11 : 625-657.
TREWAVAS, E. 1941. Nyasa fishes of the genus Tilapia and a new species from Portuguese

East Africa. Ann. Mag. nat. Hist, (n) 7 : 294-306.
GREEN, J. & CORBET, S. A. 1972. Ecological studies on crater lakes in West Cameroon.

Fishes of Barombi Mbo. /. Zool. Lond. 167 : 41-95, i pi.

DR E. TREWAVAS

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON SW7 5BD

PLATE i

Above, Sarotherodon lidole of SL 125 mm from Lake Tschungruru. Below, S. chungruruensis,
holotype.

Bull. Br. Mus. nat. Hist. (Zool.) 30, 5

PLATE i

PLATE 2

Above, 5. lidole of SL 65-5 mm from Lake Tschungruru. Below, 5. chungruruensis, a lecto-
paratype of SL 69 mm.

Bull. Br. Mus. nat. Hist. (Zool.) 30, 5

PLATE

PLATE 3

Lower pharyngeal bones of, above, the holotype of 5. chungruruensis ; below, a specimen of
S. lidole of SL 125 mm from Lake Tschungruru. ( x 8-25).

Bull. BY. Mus. nat. Hist. (Zool.) 30, 5

PLATE 3

A NEW SPECIES OF

HEMICHROMIS (PISCES, CICHLIDAE)

OF SIERRA LEONE AND LIBERIA

By A. I. PAYNE AND E. TREWAVAS

INTRODUCTION

WHILE the junior author (A. I. P.) was making a general survey of the freshwater
fishes of Sierra Leone, he found a hitherto undescribed species of Hemichromis.
Some specimens were taken in baited plexiglass traps similar to that described by
Breder (1960) as a fry-trap ; others were caught by hook and line.

Among earlier collections from Sierra Leone in the British Museum (Natural
History) we found that those of Mr N. W. Thomas, received in 1915, and Mr T. S.
Jones, agricultural officer in the territory in the 19503, also contained examples of
the new species. The juveniles from Mr Jones had been found sufficiently striking
to have been labelled with a query as to species. Two from Liberia presented by
Mr E. Roloff in 1972 and provisionally labelled 'Hemichromis sp/ also belong to the
new species.

The description below is based mainly on preserved material, but the colour is
described from life.

DESCRIPTION OF NEW SPECIES

Hemichromis fugax sp. n.

HOLOTYPE. <J, 72 + 19 mm, from a forest stream, Kassewe Forest Reserve,
Sierra Leone.

PARATYPES. Five specimens, 41-76 mm in SL from Kassewe, the Gbangbar
system near Moyamba and from a small stream with muddy bottom near Njala, as
listed on p. 166.

DESCRIPTION. Details of these specimens (except the smallest), two from
Victoria, Sierra Leone, and one from a stream about 50 km from Monrovia, Liberia
(coll. Roloff), are given in Table I. The 41 mm paratype, having been preserved
with the mouth thrust forward, is unsuitable for measurement of proportions.

Upper profile of head straight, lower jaw slightly projecting.

Cheek with 3 or 4 horizontal rows of scales, in some i or 2 small scales in addition.

Teeth of jaws unicuspid, in one row in lower jaw (rarely i or 2 inner teeth), one
and a short second row in upper jaw, the lengths grading smoothly from longer
anterior to shorter posterior.

Gill-rakers short, 2 + 1 + 5-7 on the first arch, the i or 2 lowest often abruptly
smaller than the others.

Bull. Br. Mus. nat. Hist. (Zool.) 30, 5 Issued 28 October 1976

9

i6o

A. I. PAYNE AND E. TREWAVAS

SL (mm)
As % SL
Depth
Head

TABLE i

Proportions and some meristic characters in H. fugax

Holotype
Kassewe

72

Paratypes

Kassewe
76-5 66-5

36-1

36-8

34'6
38-5

34'6
39-8

«'5

n-8

12-9

9-i5

9-3

9-8

15-3

15-9

15-8

i3-9
16-8
24-6

14-4
15-7
25-5

I3-5
15-0

22-6

3i-8
26-8

30-6

22-O

32-5
23-0

n-5

15-9

I5-I

25-3

24-0

24-5

34-8

33-9

34'°

42-5
26

3i

XIV 1 1

41-4

27
3i

XIV 12

39'7

26 ?

3
XIV 1 1

8

9

8

Moyamba
70

35'7
39-3

12-2

9'45
16-5
11-7
16-8
25-0

31-0
22-5
12-7
24-0
33-o
41-8
26

3i

XIV 1 1

Njala
43

Victoria
80 53'5

Liberia
59

sn
int
l.j.
c.p.l.
c.p.d.
P

As % head
sn
eye

pr

int

u.j.

l.j.

Sc.1.1.
Sc.D-1.1.
D
A (III + )

Head = length of head ; sn = snout, from middle of upper lip to edge of orbit ; pr = depth of pre-
orbital bone ; int = interorbital width, the roof of the skull between the eyes ; u.j. = upper jaw, from
middle of upper lip to end of maxillary ; l.j = lower jaw, from middle of lower lip to end of retro-
articular ; c.p.l. and c.p.d. = length and depth of caudal peduncle ; P = length of pectoral fin ; Sc.1.1.
= no. of scales in lateral line series, resuming at the end of the upper lateral line with the lower lateral
line scale (pierced or not) in the oblique row behind, the row used sloping downwards and forwards ;
D = dorsal fin-rays ; A = soft anal rays. SL = Standard Length.

34-8

34'4

34'6

35-6

38-8

38-2

39-8

39-2

10-5

I2-I

13-1

I2-O

8-6

9-25

9'35

8-5

17-0

15-3

16-8

16-1

n-9

12-5

14-0

12-9

16-6

16-4

16-8

16-3

26-0

23-7

25-6

24-6

27-0

31-7

32-8

30-2

28-8

22-8

26-3

28-1

I2-O

16-35

15-5

n-7

22-0

24-0

23-5

21-3

30-6

32-7

35-2

3i-i

43-7

39'9

42-2

39'4

25

26

—

27

3*

4

4

4

XIV ii XIII 12 XIII 12 XIV ii

2mm

FIG. i. The lower pharyngeal bones of H. bimaculatus (left) and H. fugax (right).

NEW SPECIES OF HEMICHROMIS 161

Scales cycloid with gothic circuli, 25 to 27 in the lateral line series, 3, 3^ or 4 from
origin of dorsal to lateral line.

Caudal fin rounded. Length of caudal peduncle 0-75-0-9 of its depth.

Lower pharyngeal bone hastate (Fig. i) with rather long lateral apophyses.
Teeth slender, but the posterior ones of the middle rows may be worn down ; lateral
posterior and several others with a crest of 1-4 small brown points in front of the
main cusp. In the Liberian specimen some teeth with crest not divided into points,
others with one, two or three in addition to the main cusp ; anterior teeth unicuspid.
Teeth of posterior row 16-20.

Colour in life (taken straight from the water). General appearance olive brown
with three approximately equidistant black blotches on each side, the first on
operculum, the third at base of caudal ; much red on the fins.

Head with variable gold and red streaks on cheek and gill-cover and a greenish
gold rim to the opercular spot. A circumpupillar gold ring, rest of iris grey or (in
two of four observed) with red lower quadrant. Lips and chin sometimes tinged
red. Lappets of dorsal fin, apex of soft fin and often also its upper edge bright
red ; a clear submarginal band ; rest of fin membrane golden-yellow to pinkish
with clear or pale blue spots. Anal fin clear proximally, reddish distally. Caudal
fin with distal quarter to three-quarters bright red, proximally yellowish ; a clear
submarginal streak underlining the red upper corner. Pectoral transparent,
colourless or yellow. Pel vies clear with red or black leading edge.

After a minute or two out of water small pale turquoise spots may appear on the
flank, but these are much smaller and less iridescent than those of H. bimaculatus
Gill. With time reddish centres to the scales may also appear.

After two days in a tank one fish became much more sombre and the middle
flank spot disappeared. A dark bar through the eye became pronounced and eight
faint vertical bars could be switched on.

Preserved specimens show the three lateral blotches and often traces of the
vertical bars.

DIAGNOSTIC FEATURES. Distinguished from H. bimaculatus by the long, acute
snout and narrow interorbital space, giving the ratio of Table 2 (p. 164) and from
H. fasciatus by lower numbers of scales (29-30 in H. fasciatus] and the fact that the
anterior teeth are not abruptly contrasted in length with the rest but grade smoothly
into the series.

ECOLOGY. The fish has been caught in fairly fast-flowing forest streams, where it
tends to lie in the lee of weed-beds, stones or logs, and also in swampy areas. In
five localities sampled it was accompanied by H. thomasi (Boulenger) (three places)
or H. fasciatus Peters (two places), but not by H. bimaculatus. Although both
species were sent by N. W. Thomas in 1915 from 'Victoria' we have no details of the
locality or localities in which they were taken.

DISTRIBUTION. The sites where H. fugax has been found in Sierra Leone, along
with the records for H. bimaculatus, are indicated in Fig. 3. H. fugax appears to be
restricted to the river systems east of the Rokel and the Freetown Peninsula.
Samples from the most easterly rivers of Sierra Leone, Moa and Mano, have not yet

162 A. I. PAYNE AND E. TREWAVAS

been obtained, but some juveniles described below were taken not far west of River
Moa and no doubt it will be found in suitable places between here and the Liberian
locality.

H. bimaculatus also occurs in the east of Sierra Leone but in the territory so far
it has been located only in the physiologically more demanding brackish water areas
of Sherbro Island and the coastal region opposite, including Lake Kwarko, a
lagoon in the waterways parallel to the coast into which Rivers Waanje and Sewa
drain. H. bimaculatus may therefore be physiologically more adaptable than
H . fugax. It is widespread in other African territories.

The Liberian specimen of H. fugax mentioned above and a smaller one collected
by Herr Roloff at the same time show that the range of H. fugax must extend
eastwards, but the limits have yet to be determined. We have not found it among
the material from the rest of Africa in the BMNH.

JUVENILES OF THE SPECIES OF HEMICHROMIS

Young H. fugax of about 15 mm SL obtained at Kassewe have a black median
longitudinal stripe from the operculum to the end of the caudal peduncle, in
preserved specimens ending in a slightly darker blotch at the base of the caudal fin.

Seven young of 21-0-22-5 mm SL from Potoru (a few kilometres west of the
expanded part of River Moa), preserved in the BMNH, show this broad band
expanded at the position of the mid-lateral blotch. At the base of the caudal fin
it ends as an intense black blotch, narrowly extended on the middle caudal rays to
the tip of the fin (Fig. 2). These have the acute snout and triradiate pharyngeal
bone characteristic of H. fugax, in contrast to young of H. bimaculatus of the same
size.

Young of H. bimaculatus from Bonthe, Sherbro Island (where adults have also
been collected), show in contrast separate opercular, mid-lateral and caudal blotches
with no extension on to the caudal rays (Fig. 2). They have the more robust
pharyngeal bone characteristic of this species. In Lake Kwarko (or Kwako) the
young are similar.

The difference in the pigmentation of the caudal between the two samples of
H. fugax suggests that this pattern may not be strictly specific ; or possibly the fact
that those lacking the caudal streak were smaller may be significant. In H.
bimaculatus too there is some variation. In a sample of thirteen, 15-5-44-5 mm in
SL, from Kiyawa River near Katagum, Nigeria, the two smallest have the caudal
blotch produced a short distance on the caudal fin and between the mid-lateral and
caudal blotches is a dusky band. The same is true of a 20-5 mm juvenile from
about 19 km north of Monrovia taken with two adult H. bimacidatus. It has the
appearance of H. bimaculatus, but the pharyngeal bone is equivocal. (This sample
suggests that a borderline between the distribution of H. fugax and H. bimaculatus
may lie between 20 and 50 km from Monrovia.)

Some fry of H . fasciatus from Sherbro Island have the lateral band emphasized
at the positions of the future lateral blotches and continued on the middle caudal
rays. This caudal extension was not figured for young of this species at Yangambi,

NEW SPECIES OF HEMICHROMIS

FIG. 2. Juveniles of (above) H. fugax from Potoru and (below) H. bimaculatus from
Bonthe with their pharyngeal bones.

Congo, by Gosse (1963, fig. 17), but the two 'forms', A and B, of H. fasciatus are
found in Sierra Leone and the fry may give a clue to their distribution there.

A recent account of the breeding and rearing of young of H. fasciatus from
Liberia (Nieuwenhuizen, 1975) shows, among other handsome photographs, a shoal
of 3-week-old fry. Their colour-pattern is almost exactly that of our fry from
Potoru (Fig. 2), but with a more pronounced upper stripe on each side of the base
of the dorsal fin. The identity of our juveniles with H. fugax has been confirmed
by the meristic characters (scales in lateral line series 26-27, against 29-30 in
H. fasciatus, both 'forms') and that of Nieuwenhuizen's fry by the photograph of
the parent, which has five black blotches along the side behind the opercular spot.

164 A. I. PAYNE AND E. TREWAVAS

This is probably the 'B form' of H.fasciatus (? =H. elongatus Guichenot in Dumeril,
1859 cf. Loiselle, quoted in Trewavas, 1974) since the 'A form' is notoriously so
aggressive that it is almost impossible to rear it in an aquarium.

RELATIONSHIPS OF H. FUGAX

In meristic characters this species cannot be distinguished from H. bimaculatus
Gill, also abundant in Sierra Leone, but it is easily recognized by the different colour-
ing, especially the lack (or lower brilliance) of the turquoise spots on the flanks, and
by the acute snout (PI. i and Fig. 2). The difference in snout length between the
two species (Table 2 and Fig. 3) is slight, with some overlap, but the interorbital

TABLE 2

Length of snout (Sn) and interorbital width (Int) in Hemichromis fugax and H. bimaculatus
and the ratio of snout to interorbital width

H. fugax (N = 13) H. bimaculatus (N = 19)

Range Mean Range Mean

SL (mm) 40-7 -80-5 61-21 44-0 -100-0 61-45

Sn (% head) 26-0 -30-6 28-33 22-6 - 28-6 25-55

(% SL) 9-5 -11-9 10-88 8-45- 10-2 9-27

Int (% head) 20-0 -23-0 21-20 22-2 - 26-8 24-73

(% SL) 7-6 - 8-95 8-18 8-2 - 9-85 8-96

Sn/Int 1-23- 1-50 1-33 0-86- 1-19 1-04

width is less in H. fugax and the snout /inter orbital ratio consequently gives a
reliable difference between the two species. They are also distinguished by the
pharyngeal teeth, fewer and more slender in H . fugax, sometimes with up to four
minor cusps, whereas in H. bimaculatus some of the median teeth are stout and
blunt. In both adult and juvenile the pharyngeal bone is more robust in H.
bimaculatus. The teeth of the jaws are variable in both species, an inner series being
present or absent, when present of a few teeth only. In neither species are the
anterior so sharply contrasted in size with the others as in H. fasciatus Peters, but
the shape of the snout and mouth in H. fugax approaches that in H. fasciatus.

The significance of the difference in colour-pattern between our samples of
juveniles can only be judged after tests of variation between populations and
developmental stages. Possibly the juvenile pattern is another feature in which
H. fugax resembles H. fasciatus and differs from H. bimaculatus.

We have mentioned H. thomasi as though it were a species of Hemichromis, as
Loiselle & Welcomme have treated it (1972). This is, in any case, an isolated species
not liable to be confused with any of the currently recognized species of Hemichromis
and certainly not with H. fugax, having a very short, blunt snout and smaller, more
numerous pharyngeal teeth.

We have also examined syntypes (or topotypes) of H. letourneuxi Sauvage
(i88oa) from Lake Mareotis, Nile delta ; of H. rolandi Sauvage (1881) from Zibane,

NEW SPECIES OF HEMICHROMIS

165

10

10

o

A
O

SNOUT

50
S L (mm)

60

70

80

90

100

O

o

O

q A

INTERORB.

FIG. 3. Comparison between H.fugax (A) and H. bimaculatus (O) in length of snout and
interorbital width (in mm).

Constantine, Algeria ; H. saharae Sauvage (i88ob) from near Touggourt ; and the
large specimens that are the syntypes of H. guttatus Giinther, of unknown locality.
These names have been regarded as synonyms of H. bimaculatus (by Pellegrin,
1904 : 219, and Boulenger, 1915 : 431) and we are satisfied that H. fugax is distinct
from them all.

ACKNOWLEDGEMENTS

We are grateful to Mr Paul L. Loiselle for communicating to us his views on the
species of Hemichromis and particularly his opinion that the type of H. bimaculatus,
which he has seen, is the species to which we have attributed the name. We also
thank Dr G. Corbet and Dr P. H. Greenwood for reading the script and suggesting
improvements.

1 66

A. I. PAYNE AND E. TREWAVAS

MATERIAL EXAMINED

Hemichromis fugax SL (mm)

BMNH 1976.1.28.1 (holotype) 72

BMNH 1976.1.28.6-7 (paratypes) 66, 76

BMNH 1976.1.28.2-3 (paratypes) 41, 70

BMNH 1971.8.13.24 (paratype) 43

BMNH 1976.1.28.4-5 45, 46-5

BMNH 1976.1.28.8 72

BMNH 1976.1.28.9 53

BMNH 1976.1.28.10-13 52-65

BMNH 1915.5.27.21-23 40-80

BMNH 1958.9.18.224-9 21-22-5

BMNH 1972.3.16.11-12 28, 59

Hemichromis bimaculatus

BMNH 1888.10.19.23-32 24-74-5

BMNH 1899.11.25.4 100

BMNH 1915.5.19.28-30 48-5-72

BMNH 1915.5.27.24 63

BMNH 1950.9.22.50-70 22-5-71

BMNH 1958.9.18.219-222 45'5-89

BMNH 1958.9.18.223 25-3

BMNH 1958.9.18.230-239 22-36

BMNH 1958.9.18.240-1 27, 33

BMNH 1969.11.19.20-21 (3 fishes) 20-5, 62-5, 6

BMNH 1928.7.3.112-116 29'5~45'5
BMNH 1930.3.22.276-284 (13 fishes) 15-5-44-5

BMNH 1884.5.2.3-4 (syntypes of 39'5, 41*0

H. saharae)

BMNH 1884.5.2.5-7 (syntypes of 41-2-44-0

H. rolandi)

BMNH 1898.2.15.1 (syntypes of 47'5> 51>o

H. letourneuxi)

BMNH 1860.4.19.3-4 (syntypes of 90-7, 95-5

H. guttatus)

Locality
Forest stream, Kassewe

Forest Reserve, S.L.
Forest stream, Kassewe

Forest Reserve, S.L.
Gbangbar system nr

Moyamba, S.L.
Stream nr Njala, S.L.
nr Moyamba, S.L.
River Taia, nr Njala,

S.L.

Kassewe, S.L.
River Tabe, S.L.
'Victoria, S.L.'
Potoru, S.L.
50 km from Monrovia,

Liberia

Freetown, S.L.
Mountain stream,

Freetown, S.L.
North Sherbro District,

S.L.

'Victoria, S.L.'
Lake Kwarko, S.L.

(7°i8'N, n°59'W)
Black 'River', Sherbro

Is., S.L.
Bonthe, S.L.
Bonthe, S.L.
RoKupr, S.L.
19 km from Monrovia,

Liberia
Kiyawa River, nr

Katagum, Nigeria
Kiyawa River, nr

Katagum, Nigeria
'Oued Rhir Constantine'

'Oued Rhir, Constantine,

Algeria'
Lake Mareotis, Egypt

Collector
A. I. Payne

A. I. Payne
A. I. Payne

A. I. Payne
A. I. Payne
A. I. Payne

A. I. Payne
A. I. Payne
N. W. Thomas
T. S. Jones
E. Roloff

R. Dinzey
Hopkins

N. W. Thomas

N. W. Thomas
T. S. Jones

T. S. Jones

T. S. Jones
T. S. Jones
T. S. Jones
C. Steiner

LI. Lloyd
LI. Lloyd

Mus. Hist.
Nat. Paris

Mus. Hist.
Nat. Paris

Mus. Hist.
Nat. Paris

Stevens (pur-
chased)

NOTE ON THE LOCALITIES

Most of the localities mentioned can be found on a modern map and by reference
to our Fig. 4, but with two we had difficulty.

'Victoria, Sierra Leone.' The Official Standard Names Gazetteer gives the
position of three such places, but in the maps consulted by us only those of an early
Handbook of Sierra Leone (Goddard, 1925) mark any of them, and this only one,

NEW SPECIES OF HEMICHROMIS

167

nearest to the Gazetteer reference 7°39'N, i2°2o'W. This is a very small place a
short way inland at the latitude of the north end of Sherbro Island, north of the
delta of River Jong. Among the other fishes sent from this address was one
'Psettus sebae' , which confirms it as a locality near the coast.

Potoru. This also has three references in the Gazetteer, but the probable one is
some 8 km west of River Moa at its expanded part. This is the only one marked

FIG. 4. The site records of H. fugax (A) and H. bimaculatus (•) in Sierra Leone. At
one site (?) fishes that were probably H. bimaculatus were observed but no specimens
were taken.

168 A. I. PAYNE AND E. TREWAVAS

on a map of 1950 (Director of Colonial Surveys, London), the period when Mr Jones
was resident in Sierra Leone. It is also marked on the American i : i ooo ooo map
NB 29 ser. 1301.

'Oued Rhir, Constantine.' This is the locality given in the register of BMNH
for specimens of both H. saharae and H. rolandi. The original descriptions give
for H. saharae 'Sahara, aux environs de Touggourth' and for H. rolandi 'Zibans,
Sahara, Province de Constantine.' Both these localities are probably in the
catchment area of Oued Rhir and Chott Melrhir. These and the specimens of
H. letourneuxi in the BMNH are listed as 'types' by Boulenger (1915). Other
syntypes remain in Paris.

REFERENCES

BOULENGER, G. A. 1915. Catalogue of African freshwater fishes. Vol. III. London.

BREDER, C. M. 1960. Design of a fry trap. Zoologica N.Y. 45 : 155-159.

DUMERIL, A. H. A. 1859 Reptiles et poissons de 1'Afrique occidentale. Arch. Mus. nat.

Hist. Paris 10 : 137-268, pis xiii-xxiii.
GILL, T. 1862. On the West Africa genus Hemichromis and descriptions of new species in

the museums of the Academy and Smithsonian Institutions. Proc. Acad. nat. Sci. Philad.

1862 : 134-139.

GODDARD, T. N. 1925. Handbook of Sierra Leone. London.
GOSSE, J. P. 1963. Le milieu aquatique et 1'ecologie des poissons dans la region de Yangambi.

Ann. Mus. r. Afr. centr. no. 116 : 113-249, pis i-x.
GUNTHER, A. 1862. Catalogue of the Acanthopterygii Pharyngognathi in the collection of the

British Museum. Vol. IV. London.
LOISELLE, P. L. & WELCOMME, R. L. 1972. Description of a new genus of cichlid fish from

West Africa. Rev. Zool. Bot. afr. 85 : 37-57.
NIEUWENHUIZEN, A. VAN DEN, 1975. Hemichromis fasciatus - gouden schoonheid uit Afrika.

II. Het Aquarium Jg. 46 : 30-34, 7 figs.

Official Standard names Gazetteer No. 101. Sierra Leone. U.S. Board on Geographical Names.
PELLEGRIN, J. 1904. Contribution a 1'etude anatomique, biologique et taxonomique des

poissons de la famille des cichlides. Mem. Soc. zool. Fr. 16 : 41-401, pis iv-vii.
PETERS, W. C. H. 1857. Neue Chromiden Gattung. Mber. K. preuss. Akad. Wiss. 1857 : 403.
SAUVAGE, E. i88oa. Notes sur quelques poissons recueillis par M. Letourneux en Epire, a

Corfu et dans le Lac Mareotis. Bull. Soc. philom. Paris (7) 4 : 211-215.

i88ob. Description de quelques poissons d'especes nouvelles dans la collection du Musee

d'Histoire Naturelle. t.c. 220-228.

1 88 1. Description de quelques poissons d'especes nouvelles dans la collection du Musee

d'Histoire Naturelle. Bull. Soc. philom. Paris (7) 5 : 101-107.

TREWAVAS, E. 1974. The freshwater fishes of Rivers Mungo and Meme and Lakes Kotto,
Mboandong and Soden, West Cameroon. Bull. Br. Mus. nat. Nist. (Zool.) 26 : 331-419,
4 pis.

DR A. I. PAYNE
LANCHESTER POLYTECHNIC
PRIORY STREET
COVENTRY

DR E. TREWAVAS

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON SW7 5BD

PLATE i

(A) Hemichromis bimaculatus from Bonthe. (B) H. fugax holotype. (C) H. fugax, a Liberian
specimen. All to the same scale. Photo BMNH.

Bull. Br. Mus. nat. Hist. (Zool.) 30, 5

PLATE i

•••••••••••••••••••••••••••••••••••••••••••••••III

By K. E. BANISTER

CONTENTS

Page
SYNOPSIS . . . . . . . . . . . .171

INTRODUCTION ........... 171

NOTES ON COUNTS AND MEASUREMENTS . . . . . .172

THE PUTATIVE HYBRIDS . . . . . . . . . . IJ2

THE PUTATIVE PARENT SPECIES . . . . . . . .176

Varicorhinus tanganicae . . . . . . . . .176

Barbus tropidolepis . . . . . . . . . .180

DISCUSSION ........... 183

ACKNOWLEDGEMENTS . . . . . . . . . .189

REFERENCES . . . . . . . . . . .189

SYNOPSIS

A series of fishes from Lunkungwe (Lake Tanganyika) were found to have characters inter-
mediate between two endemic species, Barbus tropidolepis and Varicorhinus tanganicae. The
likelihood of a hybrid origin for the Lunkungwe fishes is discussed and comments are made on
the validity of the generic separation of the two putative parent species.

INTRODUCTION

DURING 1968 and 1969, Dr E. K. Balon of the Zoology Department, University of
Guelph, Canada, collected eleven specimens of a cyprinid fish from Lunkungwe
(5°45'S, 29°55'E), Lake Tanganyika. The largest specimen (133 mm SL) is now
in the Museum of Zoology, University of Michigan (MZM 191573). The other ten
specimens are in the Royal Ontario Museum (ROM 28132 and 28167).

These fishes are not referable to any known cyprinid species from Lake Tanganyika
but are morphologically intermediate between Varicorhinus tanganicae and Barbus
tropidolepis, species endemic to the Lake Tanganyika basin. From the data
presented below it will be seen that the eleven Balon fishes are unlikely to represent
extreme examples of either Varicorhinus tanganicae or Barbus tropidolepis.

In this paper I intend to put forward the hypothesis that these eleven fishes are
hybrids between Barbus tropidolepis and Varicorhinus tanganicae.

Much of the evidence and discussion is, of necessity, similar to that given in
Banister (1972). In that case the discovery of some fishes morphologically inter-
mediate between Barbus somereni and Varicorhinus ruwenzorii was used as evidence
for the hypothesized hybridization of these two species. Without breeding
experiments, hybridization cannot be proved, but the circumstantial evidence is
worth examining. Previously (Banister, 1972) I noted that the boundary between
the African representatives of the genus Barbus and the genus Varicorhinus is

Bull. Br. Mus. nat. Hist. (Zool.) 30, 5 Issued 28 October 1976

172 K. E. BANISTER

vague. It has been shown (Groenewald, 1958 ; Gaigher, 1975) that under certain
conditions some species of Barbus can change their mouth form and simulate the
mouth type present in Varicorhinus species. Without an examination of more
trenchant characters and without a knowledge of the previous history of the
specimen, an individual fish could be placed with equal justification in either genus
as the two genera are currently accepted (e.g. the definitions of Boulenger, 1909,
and Jubb, 1968). These two definitions are, in the light of the known variation of
some species, inadequate. The morphological intermediates described here, and
Barbus alluaudi (Banister, 1972), raise further doubts about whether some species
have been correctly assigned to the genus Varicorhinus. Work is in progress on the
limits of the genera Barbus and Varicorhinus.

NOTES ON COUNTS AND MEASUREMENTS

The standard length (SL) was taken in the usual manner. The lateral line count
(LL) was taken from the first pore-bearing scale behind the head to the scale lying
lateral to the end of the hypurals. The body depth (D) is the maximum body depth,
usually to be found just in front of the dorsal fin. The anterior limit for the head
length (H) and snout length (Snt) was the premaxillary symphysis, with the pre-
maxillae retracted. The posterior limit for the snout is the anterior margin of the
orbit and for the head length is the most posterior part of the bony edge of the
operculum. The term mouth width (MW) refers to the width across the lower jaw
at the level of the angle of the mouth. The pectoral fin length (Pet) is the total
length of the fin, measured in a straight line from the base of the first fin ray to the
distal extremity of the fin. The measurements were taken in this way because of
the ease of so doing with dial calipers, which were used on all fish except the smallest
when dividers were used. The caudal peduncle length (CP1) is the horizontal
distance from the posterior angle at the base of the last anal fin ray to the end of the
hypurals and the caudal peduncle depth is the least depth of that part. The
interorbital width (IO) was measured as the least distance between the bony edges
of the interorbital space. The eye diameter (I) is the horizontal diameter of the
visible part of the eye. Dsp symbolizes the height of the dorsal fin from the base
of the first spine to the distal extremity of the longest fin ray. With any measure-
ments which were repeatable on both sides of the fish (e.g. the anterior barbel,
Ab or posterior barbel, Pb) the larger was taken except in cases where a deformity
was obvious.

The majority of measurements were taken to the nearest millimetre, the exceptions
were some measurements on small fishes ; these were taken to the nearest half
millimetre. The mean is symbolized by x, the standard deviation by s.d., the
standard error by s.e. and the number of fish in the sample by n.

THE PUTATIVE HYBRIDS

DESCRIPTION. The description is based upon n fishes of 133, 105, 77, 69, 66, 59,
59. 58, 55, 55, 55 mm SL from Lunkungwe, Lake Tanganyika, ROM 28132, 28167
and MZM 191573. The largest specimen is shown in Fig. i.

AN INTERGENERIC CYPRINID HYBRID

2cm

FIG. i. The largest putative hybrid, a fish of 133 mm SL.

X

s.d.

s.e.

range

SL

-

-

-

55 -133 mm

D

25-8

i'5

0-4

23-5-28-6

H

27-0

1-7

0'5

23-5-29-3

I

8-5

o-5

O-2

7-8- 9-3

10

7'5

0-6

0-2

6-8- 8-8

MW

6-9

0-6

0-2

6-4- 8-5

Pet

21-3

i-i

0-3

19-6-28-6

CP1

18-9

i'5

o-5

15-3-21-0

CPd

10-9

0-7

0-2

9-8-11-7

Snt

7-8

i-i

0-6

6-8- 9-8

Ab

1-8

0'3

O-I

i-5- 2-3

Pb

3'2

o-5

O-2

2-5- 3-8

Unless stated otherwise, all measurements are expressed as a percentage of the
standard length.

The body is compressed, with smooth dorsal and ventral profiles. The deepest
part of the body is just in front of the dorsal fin. The eye is large, protuberant and
just visible in ventral view. The fleshy, rounded snout terminates in a rostral flap
which is level with, but does not overhang, the mouth.

The anterior margin of the lower jaw is gently curved ; it has a sharp edge but
lacks the conspicuous horny sheath present in specimens of Varicorhinus tanganicae
(PL 2). Nevertheless, it has a shiny, padded appearance suggesting that a tough
integument is present. The lower jaw is only slightly wider at its articulation than
it is distally. The ventral face of the lower jaw is concave and the skin in this and
the contiguous gular region bears numerous small papillae.

No lateral line pores are visible on the ventral surface of the lower jaw. There
are two pairs of barbels. The short anterior barbels are usually hidden in a groove
between the ventral edge of the lachrymal bone and lips. In the second largest

I74 K- E- BANISTER

specimen (105 mm SL) a few isolated tubercles are present on the skin of the snout
and the dorsal half of the lachrymal bone. The tubercles are about four times the
diameter of the lateral line pores of the lachrymal bone.

In nine of the specimens the gill filaments could be seen clearly through a trans-
parent zone in the centre of the opercular bone. In the other two fishes a dark
patch of pigment was present through which the gill filaments could only just be
discerned.

The peritoneum is dark grey /brown in the two specimens examined. The
alimentary canal is narrow and much convoluted. There are 20 + 21 (fi), 20 + 22
(13), 21 + 20 (fi), 21 + 21 (12), 21 + 22 (13) or22 + 2i (fi) vertebrae including those
comprising the Weberian mechanism.

From radiographs it can be seen that the angle of insertion of the pectoral fin is
different in the three samples of fishes. The angle of insertion of the pectoral fin is
here defined as the angle that a line through the mid-point of the first and last
pectoral radials subtends with a line from the mid-point of the occipital condyle to
the mid-point of the ural centrum. In the largest fish this angle is 30° ; in the other
specimens it varied from 21° to 41°.

Dorsal fin. The dorsal fin has 4 simple rays and 9 (fn) branched rays. The last
simple ray is thickened into a straight smooth spine (x = 26-3, s.d. = 1-5, s.e. = 0-5,
range = 23-8- 28-5). A low sheath of scales is present at the base of the dorsal fin.

FIG. 2. The right pharyngeal bone of the largest hybrid.

AN INTERGENERIC CYPRINID HYBRID

175

FIG. 3. Occlusal views of the pharyngeal bones of hybrids of 133 and 59 mm SL to show

the arrangement of the tooth rows.

This sheath is highest anteriorly and is reduced to the level of the body contour by
about the seventh or eighth branched ray. The origin of the dorsal fin is in advance
of the pelvic fin.

Squamation. In the lateral line series there are 46 (fi), 47 (fi), 49 (f2), 50 (fi),
51 (fi), 52 (fa), 53 (fi), 54 (fi) or 56 (fi) scales. From the dorsal mid-line to the
lateral line there are io| (f5) or n| (f6) scales. From the lateral line to the ventral
mid-line there are io£ (12), n£ (f3) or 12^ (13) scales. This count was unobtainable
on some specimens. Between the lateral line and the base of the pelvic fin there are
5 1 (f5), 7 (fi) or 7! (f4). Around the mid-point of the caudal peduncle there are
22 (f5) or 24 (f4) scales. The scales bear slightly diverging striations (Fig. 5).

Pharyngeal bones and teeth. The pharyngeal teeth number 2.3.5-5.3.2. The right
pharyngeal bone of the largest specimen is shown in Fig. 2. The first (anterior)
tooth of the inner row is short and slender and its alignment is more or less parallel
with the adjacent edge of the second tooth. This tooth is large and stout with a
mamilliform crown surmounted by a small recurved hook. The third, fourth and
fifth teeth become progressively more slender and spatulate. The teeth of the second
and third rows are smaller versions of the third, fourth and fifth teeth of the inner
row. From the occlusal view of the pharyngeal bones the alignment of the middle
and outer tooth rows can be seen. In about half of the specimens these two rows
diverge from the first tooth of the middle row. In the rest of the specimens the
rows are parallel. In two specimens one pattern is present on one bone and the
others on its partner.

The mean tip-to-tip length of the pharyngeal bones of a specimen 105 mm SL is
10 mm and the mean length in a specimen of 59 mm SL is 6-3 mm. These figures

10

I76 K. E. BANISTER

should be compared with the measurements of the pharyngeal bones of equal-sized
Barbus tropidolepis and Varicorhinus tanganicae (pp. 179 and 183).

Gill-rakers. There are 15 (f5), 16 (f5) or 18 (fi) gill-rakers on the lower limb of the
first gill-arch. The shape of the gill-rakers is intermediate between those of the
putative parents ; i.e. they are moderately thick with a clubbed or T-shaped
extremity.

Coloration. These colour notes are based on alcohol-preserved specimens. The
smaller fishes are lighter than the larger fishes. The body colour is yellow-brown,
darker on the back than below the lateral line. The belly in some specimens has a
pink tinge. There is a dark, broad mid-lateral stripe 3 or 4 scales deep caused by a
concentration of pigment along the posterior edge of alternate scales in the rows at
the edge of the band and on all the scales of the central rows. A distinct mid-dorsal
stripe is present in several specimens. These stripes become less distinct in larger
fishes as the body colour becomes progressively darker. All but two of the specimens
have a clear spot on the operculum. The dorsal fin has a dark margin in small
fishes but this becomes progressively fainter in larger fishes.

DISTRIBUTION. All the specimens came from Lunkungwe, Lake Tanganyika.
Dr Balon kindly provided the following details of the locality. The Lunkungwe
River empties into a large bay on the eastern shore of central Lake Tanganyika
about 100 km south of Kigoma in Tanzania. The river flows from the Kingwe
mountains in a valley flanked by very steep foothills. The river gradient is steep.
The width varies from 4 to 10 m, and there are rapids between rock outcrops and large
boulders on the river bed. The fish were sampled about 2 km before the river
empties into the lake. The river at this point flows swiftly but there are some pools
with a plant resembling Fontinalis sp. The river is muddy when it empties into
the lake and is bordered with reed thickets. The fish were poisoned with toxophene.

THE PUTATIVE PARENT SPECIES

Varicorhinus tanganicae (Boulenger, 1900)

Capoeta tanganicae Boulenger 1900 : 478.

Varicorhinus tanganicae : Boulenger, 1905 : 43 ; Poll, 1953 : 97.

LECTOTYPE. A fish of 282 mm SL from the north end of Lake Tanganyika, one
of the two syntypes, BMNH reg. no. 1906.9.6:9-10. The lectotype is the largest
of the two specimens, the smaller specimen being designated a paralectotype.

MATERIAL EXAMINED. Apart from the syntypes the following specimens were used
in this study : BMNHi936. 6.15:731-751 ; 1955.12.20:900-903 ; 1955.12.20:881.

DESCRIPTION. Two separate samples of Varicorhinus tanganicae from Lake
Tanganyika have been measured. One sample, of 24 fish from 48 to 282 mm SL, is
intended to show the morphometric and meristic variation of the species. The
other sample consists of five fishes of 59, 59, 80, 104 and 139 mm SL which are all the
specimens available close to the size range of the hybrids. These are included to
enable a detailed comparison to be made limiting, as far as possible, any distortion
from allometry.

AN INTERGENERIC CYPRINID HYBRID 177

Sample i (n = 24)

x s.d. s.e. range

SL 48 -282 mm

D 25-9 2-1 0-4 23-0-30-0

H 22-7 2-4 0-5 14-6-26-0

I 6-3 0-9 0-2 4-6- 8-3

IO 9-1 0-9 9-2 7-4-10-6

MW 8-0 0-5 o-i 7-1- 9-5

Pet 19-7 1-2 0-2 16-6-21-4

CP1 17-2 0-9 0-2 15-1-18-9

CPd 12-8 0-9 0-3 10-4-14-7

Snt 7-3 0-6 o-i 6-4- 8-3
Ab +
Pb +

Dsp 19-8 3-0 0-6 12-6-24-8

It should be noted that although two pairs of barbels are invariably present in the
specimens examined they are minute in the large fishes. Consequently barbel
lengths are not included in this sample but are included in the 59-139 mm SL
sample below.

Sample 2 (n = 5)

x s.d. s.e. range

SL 59 -139 mm

D 26-3 1-2 0-5 25-0-28-0

H 24-8 2-0 0-9 22-3-27-1

I 7-3 0-3 o-i 6-9- 7-6

IO 8-8 0-3 o-i 8-5- 9-3

MW 8-7 0-5 0-2 7-9- 9-3

Pet 20-5 1-2 0-5 18-9-22-0

CP1 15-4 1-4 0-6 I3-5-I7-3

CPd 10-5 0-5 0-2 lo-o-ii-o

Snt 6-8 0-5 0-2 6-3- 7-5

Ab (n = 4) 1-3 i-o 0-5 o-i- 2-6

Pb 3-1 1-3 0-6 1-4- 5-1

Dsp 23-9 2-3 i-o 21-2-27-3

Unless stated otherwise all measurements are expressed as a percentage of the
standard length.

The body is compressed and streamlined, with smooth dorsal and ventral profiles.
The snout is fleshy and bluntly rounded. The rostral flap almost overlaps the margin
of the upper j aw on the ventral surface of the snout. The eye is large and protuberant
usually visible from below, a feature that is more conspicuous in smaller specimens.
In fishes of less than about 60 mm SL the anterior edge of the lower jaw is gently
curved and lacks a horny covering. The lower jaw is straight-edged in larger
fishes and has a sharp horny cutting edge.

A series of six or seven lateral line pores is usually conspicuous on the skin of the
ventral surface of each ramus of the lower jaw. Two pairs of small barbels are

I78

K. E. BANISTER

FIG. 4. Varichorhinus tanganicae (from Boulenger, 1901).

present. The anterior barbels can best be described as small prolongations of the
postero-ventral corners of the rostral flap. The groove below the lachrymal bone
is much narrower than in the Balon specimens. A few of the larger (more than
about 200 mm SL) specimens have a scattering of small, poorly defined tubercles on
the skin covering the lachrymal bone. The tubercles are non-contiguous and are
about the same absolute size as those in the putative hybrids.

The peritoneum is dark grey-brown and the intestine is more convoluted than in
the putative hybrids.

There are 20 + 21 (fi), 21 + 21 (12), 21 + 22 (14), 22 + 21 (12) or 22 -I- 22 (fi) vertebrae
in the ten specimens radiographed.

The angle of insertion of the pectoral fin varies between 37° and 51° in the ten
specimens examined.

Dorsal fin. There are 4 unbranched rays and 8 (14), 9 (fiy) or 10 (13) branched
rays. The fourth simple ray is ossified into a straight smooth spine. The shape of

FIG. 5. A scale of a specimen of Varichorhinus tanganicae of 139 mm SL (left) a hybrid
of 133 mm SL (centre) and a Barbus tropidolepis of 134 mm SL (right).

AN INTERGENERIC CYPRINID HYBRID

179

FIG. 6. The right pharyngeal bone of Varicorhinus tanganicae of 139 mm SL.

the dorsal fin is shown in Fig. 4. The anterior half of the base of the dorsal fin is
surrounded by a low sheath of scales. The origin of the dorsal fin is in front of the
origin of the pelvic fin.

Squamation. There are from 57 to 67 scales in the lateral line series : 57 (£2),
58 (fi), 60 (£2), 61 (14), 62 (f2), 63 (£4), 64 (£4), 65 (£4), 67 (fi). There are i2| (£12),
13 (£9) or 14-! (£3) scale rows from the dorsal mid-line to the lateral line. Between
the lateral line and the ventral mid-line there are I2| (fi), 13^ (£5), 14! (£6), 15^ (£5),
i6| (£2) or 17^ (f i) scale rows. These scales could not be counted on some specimens.
From the lateral line to the base of the pelvic fin there are 8 (fio), 9 (£12), 10 (fi) or
ii (fi) scales. Around the least circumference of the caudal peduncle there are
29 (fi), 30 (£5), 31 (£3), 32 (£6) or 33 (£2) scales.

The striations on the scales are radiate, but they tend to be more divergent in
smaller fishes than in larger ones.

Pharyngeal bones and teeth. The pharyngeal teeth number 2.3.5-5.3.2 (Fig. 6).
The teeth are slender and the posterior teeth in each row have spatulate crowns.
The teeth are arranged in three parallel rows. The mean tip-to-tip measurement
of the pharyngeal bones of a specimen of 104 mm SL is 8-9 mm and of a 60 mm SL
specimen 6-0 mm.

i8o

K. E. BANISTER

Gill-rakers. In the 14 specimens examined the number of gill rakers on the lower
limb of the first gill arch is 17 (fi), 18 (f4), 19 (fy) or 20 (f2). The gill rakers are
slender, conical and bear small lateral projections.

Coloration. Live fishes were described by Poll (1953) as brown to grey on the back,
paler ventrally. A silver sheen is usually present and the fins are pale grey.
Preserved specimens are usually a pale yellow/brown in colour although the larger
specimens tend to be greyer.

DISTRIBUTION. This species is confined to the Lake Tanganyika basin. Accord-
ing to Poll (1953) it is found in both the lake and the affluent rivers, even in the
rapids of the inflowing streams. In the lake it is commonest in rocky areas near to
the mouths of streams. It ascends rivers to breed. Small fishes (less than 80 mm
SL) are rare in the lake and seem to spend the early part of their lives in the streams.
In both the lake and the feeder streams Varicorhinus tanganicae is found feeding in
sandy, gravelly or rocky regions, rummaging for insect larvae. Ostracods, worms
and diatoms mixed with grains of sand are also found in the alimentary canal
(Balon, pers. comm.).

Barbus tropidolepis Boulenger, 1900
For synonymy see Banister (1973).

MATERIAL EXAMINED. In addition to the syntypes, the following specimens were
examined : BMNHi955. 12. 20:737-740 ; 1955.12.20:775-792 ; 1955.12.20:795-799 ;
1955.12.20:806-815 ; 1955.12.20:1169-1171.

DESCRIPTION. Two separate samples of Barbus tropidolepis were measured. The
first, a sample of 47 fishes from Lake Tanganyika, are between 99 and 365 mm SL
and serve to show the variation of morphometric characters in the species (taken
from Banister, 1973). The second sample is of ten fishes and represents all the
available specimens within the size range of the putative hybrids. These ten fishes
are of 56 (2), 59 (2), 60, 66, 69, 75, 105 and 134 mm SL.

FIG. 7. Barbus tropidolepis (from Boulenger, 1901).

AN INTERGENERIC CYPRINID HYBRID 181

Sample I (11 = 47)

X

s.d.

s.e.

range

SL

-

-

-

99 -365 mm

D

28-8

2-7

0-6

25-0-33-7

H

27-0

1-4

0-3

24-8-30-1

I

9-4

i'3

o-3

7-0-11-0

10

9-0

i'5

9'3

7-4-11-4

MW

6-4

0-8

0-2

5-2- 8-8

Pet

20-8

i-o

O-2

19-1-23-9

CP1

14-9

1*2

0-3

12-9-17-7

CPd

11-7

0-9

O-2

10-1-13-7

Snt

8-4

i-o

0-2

5-2- 9-3

Dsp

25-8

2-7

0-6

20-8-30-0

Sample 2 (n = 10)

X

s.d.

s.e.

range

SL

-

-

-

56 -134 mm

D

26-3

2-2

0-7

23-4-30-6

H

27-2

0-7

O-2

25-9-28-8

I

8-8

0-6

O-2

8-2-IO-2

IO

8-7

0-6

0-2

7-6- 9-7

MW

5-9

o-5

0-1

5-1- 6-7

Pet

20-4

1-7

o-5

16-6-22-6

CPU

M-3

I-O

o-3

12-0-15-3

CPd

II-2

0-6

0-2

10-5-12-1

Snt

7-3

0-6

O-2

6-5- 8-2

Dsp

28-2

i'5

o-5

25-0-30-6

All measurements are expressed as a percentage of the standard length unless
stated otherwise.

The body has a compressed, fusiform shape. The blunt snout lacks a rostral
flap. Only a very shallow groove is present between the edge of the lachrymal
bone and the snout. The eye is protuberant and conspicuous in ventral view.

The mouth is subterminal. The lower jaw is narrow with a soft, curved, anterior
margin. The lower lip is often discontinuous medially. A few specimens have the
lips slightly thickened. The gular surface is deeply folded. There is no anterior
barbel and the posterior barbel is represented by a very small papilla. Although
invariably present, this barbel is too short to measure accurately.

Small, scattered tubercles are present on the skin covering the dorsal part of the
lachrymal bone as well as between the orbit and the nostrils. The tubercles are
smaller than those of Varicorhinus tanganicae or the Balon specimens. In common
with Varicorhinus tanganicae a series of 6 or 7 lateral line pores is conspicuous on the
ventral surface of each ramus of the lower jaw. The peritoneum is mid-brown in
colour.

In the 14 specimens radiographed there are 18 + 22 (fi), 19 + 21 (£4), 19 + 22 (f8)
or 19 + 23 (f i) vertebrae. The angle of insertion of the pectoral fin varies from 12°
to 23° in the specimens radiographed.

Dorsal fin. There are 4 unbranched rays (not 3 as reported by Boulenger, 1911,
and Worthington & Ricardo, 1937). The last unbranched ray forms a smooth

i82

K. E. BANISTER

straight spine. There are 9 (rarely 10) branched rays. The dorsal margin of the
dorsal fin has a characteristic shape which can be seen in Fig. 7. The origin of the
dorsal fin is directly above or slightly in advance of the pelvic fins.

Squamation. There are from 39 to 44 scales in the lateral line series : 39 (fa),
40 (fg), 41 (fn), 42 (fi3), 43 (fg), 44 (13). There are 8| (rarely 7^) scale rows
between the dorsal mid-line and the lateral line and 8| (rarely 7^, very rarely g%)
scale rows between the lateral line and the ventral mid-line. Around the least
circumference of the caudal peduncle there are 16 (f34), 17 (f6) or 18 (17) scales.
The ridges of fat that may occur on the scales of this species (see Banister, 1973) do
not occur in the putative hybrids nor in Varicorhinus tanganicae. The scale
striations are more or less parallel. The posterior margin of each scale is markedly
lobate.

Pharyngeal bones and teeth. The pharyngeal teeth usually number 2.3.5-5.3.2.
The first tooth of the inner row is very small and may be absent. When absent,
its locus is marked by a small pinnacle of bone. The second tooth of the inner row
is stout with a molariform crown. The remaining three teeth in this row become
progressively thinner. The crowns of especially the third and fourth teeth are
almost as wide as that of the second tooth but are not so long, i.e. they are kidney-
shaped when viewed from an occlusal position (see Fig. 9).

The alignment of the second and third rows is characteristic in that the first tooth
of the second row is displaced slightly so that the second and third tooth rows

FIG. 8. The right pharyngeal bone of Barbus tropidolepis (from Banister, 1973)-

AN INTERGENERIC CYPRINID HYBRID 183

5mm

FIG. 9. Occlusal view of the right pharyngeal bone of Barbus tropidolepis (from Banister,

1973).

appear to radiate from it, rather than lying parallel to each other. The mean
tip-to-tip length of the pharyngeal bones from a specimen of 105 mm SL is 13-5 mm
and the mean for the pharyngeal bones from a fish of 60 mm SL is 7-5 mm.

Gill-rakers. There are 13 (fio), 14 (f4) or 15 (f2) gill-rakers on the lower limb of
the first gill-arch in the 16 specimens examined. The gill-rakers are broad based
with the outside edge produced forwards as a hook.

Coloration. The colour of live fishes, as described by Poll (1953), is grey or olive
on the back, lighter ventrally. There is no noticeable colour pattern. Preserved
fishes are brown on the body. Small specimens (less than about 90 mm) have a
dark margin to the dorsal fin ; otherwise the fins are pale brown.

Habitat. This species is found in Lake Tanganyika to a depth of 25 m. It also
occurs in feeder streams, especially in the breeding season (Poll, 1953).

DISCUSSION

There are three possible ways of regarding Balon's anomalous fishes. They could
represent a new species, they could be hybrids or they could represent extreme
representatives of a known species. The Balon specimens are very much rarer in
the wild than either Barbus tropidolepis or Varicorhinus tanganicae and they have
only been found at one locality. Hubbs, Hubbs & Johnson (1943) quote, for two
species of North American Catostomidae captured in the same river system, a
proportion of hybrids to parents of 4-2 and 6-4%. They also state that some hybrids
are much rarer and the proportion may be about i%.

There is no quantitative information available about the Balon specimens.
However, Balon (pers. comm.) has stated that they occur with Barbus tropidolepis
and Varicorhinus tanganicae and are very much rarer than either. This information,
coupled with the conspicuous morphological intermediacy between Barbus tropidolepis

184 K. E. BANISTER

and Varicorhinus tanganicae possessed by the Balon fish, suggests it is unlikely that
they represent a true species.

The consistent differences between Barbus tropidolepis and Varicorhinus tanganicae
in morphometric and meristic characters leave no doubt as to their specific distinct-
ness. The differences in mouth form and associated feeding structures could
arguably be the result of particular responses to the environment (Groenewald, 1958),
but this is unlikely because of the discrete differences in other, and apparently
unrelated, characters.

The morphometric intermediacy of aquarium-bred, and thus indisputable, hybrids
has often been noted (e.g. Holcik & de Witt, ig62a, b ; Hubbs & Miller, 1952).
Hubbs (1955) described the morphological intermediacy of hybrids in the wild.
Berry & Low (1970) and Hubbs (1955) advocate intermediacy as a very significant
character of hybridization.

Unfortunately, the supporting evidence for hybridization that was used in the
case of Barbus alluaudi (Banister, 1972) has not been available in this instance. For
example, I have not been able to examine the gonads because of their inadequate
preservation and I have but little information of the frequency of the occurrence
of the Balon fishes in the wild. Therefore, the case rests solely on morphological
evidence.

The ii Balon fishes have 45-56 scales in the lateral line series. This is a wide
range for a small sample. There are only five species of large Barbus and
Varicorhinus in Lake Tanganyika. Barbus tropidolepis has 39-44 (n = 47) scales ;
Varicorhinus tanganicae has 57-67 (n = 24) ; Varicorhinus leleupanus has 42 or 43
(n = 2) ; Barbus platyrhinus has 38-41 (n = 7) ; and Barbus caudovittatus has 24-30
(n = 47). Considering just the lateral line scale counts, no overlap exists between
any of the recognized species and the Balon specimens ; which alone tends to
render unlikely the hypothesis that these specimens represent extreme forms of any
known species.

Only one of the five species of Varicorhinus and large Barbus in Lake Tanganyika
has smaller scales than the fishes under consideration, i.e. Varicorhinus tanganicae
with 57-67 scales in the lateral line series. This species, therefore, is a certain
candidate as one of the parents of the putative hybrids.

On the basis of the scales alone, Barbus caudovittatus can surely be eliminated as
the other parent, but this still leaves Barbus platyrhinus, Barbus tropidolepis and
Varicorhinus leleupanus. The bulky body of Barbus platyrhinus is not reflected in
the streamline shape of the Balon fishes. A hybrid between Varicorhinus leleupanus
and Varicorhinus tanganicae would be very likely to have a wide, horny-edged
'sector' mouth. The Balon fishes do not have a wide, horny-edged 'sector' mouth.
Furthermore, Varicorhinus leleupanus has not been recorded from the Lunkungwe
area of Lake Tanganyika. In the wild, the Balon fishes occur together with both
Barbus tropidolepis and Varicorhinus tanganicae.

The range, mean and one standard deviation each side of the mean for seven
morphometric characters of the putative hybrids and probable parent species are
given in Table 2. The characters chosen (H, I, IO, MW, Dsp, Ab, Pb) are the
same as were considered when discussing the hybrid origin of Barbus alluaudi

AN INTERGENERIC CYPRINID HYBRID

185

(Banister, 1972). The choice was made partly for comparison with the Barbus
alluaudi situation and partly because these are the only morphometric characters in
which the putative parent species differ significantly from each other and/or from
the hybrid. The samples of Barbus tropidokpis and Varicorhinus tanganicae are
from the same size range as the putative hybrids.

The number of vertebrae and the number of gill-rakers on the lower limb of the
first gill-arch in the Balon fishes are intermediate between those of the putative
parents (Table i). The pharyngeal bone length and the angle of insertion of the

Total no. of vertebrae

B.t. 40-42 (n = 14)
Hyb. 40-43 (n = u)
V.t. 41-44 (n = 10)

TABLE i

Gill-rakers

13-15 (n = 16)
15-18 (n = n)
17-20 (n = 14)

Pectoral insertion angle

12-23° (n = 9)
21-40° (n = n)
37-51° (n = 12)

Mean length of pharyngeal bones

SL

Ph. length

(mm)

(mm)

V.t.

60

6-0

V.t.

104

8-9

Hyb.

59

6-3

Hyb.

105

10-0

B.t.

60

7'5

B.t.

105

13-5

pectoral fin show the same intermediacy, as does the alignment of the pharyngeal
teeth.

The possibility that the Balon fishes represent an extreme example of either
putative parent species must now be considered. The techniques of Schultz &
Schaeffer (1936) and of Hubbs, Hubbs & Johnson (1943) have been used to evaluate
the probability of a character of the sample under examination falling within the

35-
30-
25-
20-
15-
10-
5-

16 1718 19 2021 22232425 262728293031 3233

FIG. 10. A histogram of the frequencies of the lateral line scale counts in Barbus tropidolepis
(left), the hybrids (black, centre) and Varicorhinus tanganicae (right).

i86

K. E. BANISTER

12-
10-
8-
6-
4-
2-

384042444648505254565860626466

FIG. ii. A histogram of the frequencies of the scale counts around the least circumference
of the caudal peduncle in Barbus tropidolepis (left), the hybrids (black, centre) and
V aricorhinus tanganicae (right).

25 26 27 28

H

23 24 25 26 27

FIG. 12. A comparison of the range, mean and standard deviations of the head length
in the three forms. The base line represents the range, the vertical line the mean and
the black area covers one standard deviation either side of the mean. The figures are
the lengths of the heads expressed as a percentage of the standard length. B = Barbus
tropidolepis, V = Varicorhinus tanganicae, H = hybrids.

8-0 8-5 9-0 9-5

H

8O 8-5 9-0

7-0 7-5 8-0 8-5

7-0 7-5

8-5 90

FIG. 13. Comparisons of the eye diameter (left) and interorbital width (right) in the three

forms. Details as in Fig. 12.

AN INTERGENERIC CYPRINID HYBRID 187

range of that character in either of the putative parent species. The results are
shown in Table 2.

TABLE 2

Character n x s.d. % B % V P

H

V. t. 5 24-82 2-05 0-4

Hyb. ii 26-97 I-72 91 9

B. t. 10 27-18 0-77 0-7

I

V. t. 5 7-30 0-30 o-i

Hyb. ii 8-54 0-55 81 19

B. t. 10 8-83 0-61 0-25

IO

V. t. 5 8-78 0-31 o-i

Hyb. ii 7-46 0-63

B. t. 10 8-66 0-66 o-i

MW

V. t. 5 8-72 0-55 o-i

Hyb. ii 6-86 0-63 52 48

B. t. 10 5-90 0-51 o-i

Dsp

V. t. 5 23-94 2-35 0-4

Hyb. ii 26.34 1-52 44 56

B. t. 10 28-2 1-59 o-i

Ab

V. t. 5 1-35 1-06 o-i

Hyb. ii i -80 0-28

B. t. 10

Pb

V. t. 5 3-06 1-36 0-78

Hyb. ii 3-24 0-54

B. t. 10

For the characters H, I, MW and Dsp the mean for the hybrid lies between the
means for the other two species. The % B refers to the percentage proximity
of this mean to the mean for Barbus tropidolepis and the % V refers to the percentage
proximity to the Varicorhinus tanganicae mean. The probability, P, refers to the
probability of observing a mean equal to the Balon specimen value if the sample
had come from one of the other two species.

The morphometric measurements of the putative hybrids are satisfactorily
intermediate for the characters MW and Dsp. However, the head length results
are not as significant as they were for Barbus alluaudi (Banister, 1972). In that
case the percentage likeness of the head length to each of the putative parents is

i88

K. E. BANISTER

1-5 2-0

2-5 3-0 3-5

O5 1O 1-5 2-0 2-5

1-5 2-0 2-5 3-0 3-5 4-0 4-5 5-0

FIG. 14. Comparisons of the anterior (left) and posterior (right) barbel lengths in the

three forms. Details as in Fig. 12.

25 26 27 28 29 30

5-0 5-5 6-0 6-5

26 27

80 8-5

FIG. 15. Comparisons of the dorsal spine length (left) and the mouth width (right) in
the three forms. Details as in Fig. 12.

69 and 31% respectively. It can be seen from Table 2 that the head-length of the
Balon fish is much closer to that of Barbus tropidolepis than to that of Varicorhinus
tanganicae (% 6 = 91 ; % V = 9). One striking similarity between Barbus alluaudi
and the Balon fishes can be seen in the interorbital width. In both cases, the hybrid
interorbital width is less than that of the putative parents, the latter differing but
little from each other in this character.

Although generally it is true to say that African Barbus species have longer
barbels than African Varicorhinus species, it is not true in this case. What little
difference exists between the putative hybrids and Varicorhinus tanganicae shows
that the hybrid has slightly longer barbels.

The results of the morphometric characters analysed in Table i and the characters
described in the earlier part of the text indicate that it is unlikely that the Balon
specimens are extreme forms of either Barbus tropidolepis or Varicorhinus tanganicae.
The degree of intermediacy hints most strongly at a hybrid origin.

The evidence for the hybrid origin of the Balon fish is circumstantial and rests
solely on the fact that in many characters these specimens are intermediate between
Barbus tropidolepis and Varicorhinus tanganicae. This is the second example of
hybridization between species of the genera Barbus and Varicorhinus in Africa.
Such intergeneric hybridization raises doubts about the validity of the two genera.
The mouth shape is the main character used to distinguish the two genera and yet
this has been shown (Groenewald, 1958 ; Banister, 1973) to be a variable feature of
little value at the generic level. Some Barbus species show extra-limital variation
and achieve the Varicorhinus facies, or at least a close approximation to it. In

AN INTERGENERIC CYPRINID HYBRID 189

all probability the true situation is that the genus Varicorhinus is polyphyletic and
contains some species whose phyletic status may be best expressed by placing them
in the genus Bar bus. A full revision of the genus Varicorhinus is needed to clarify
these anomalies. This work is in progress.

ACKNOWLEDGEMENTS

My thanks go to Dr Eugene Balon for kindly allowing me to examine the specimens
he collected. Dr P. H. Greenwood read the manuscript and offered much valuable
advice. Mrs M. Clarke processed the statistical data under the guidance of Dr M.
Hills. The uncredited drawings of whole fish are the work of Miss M. Holloway.
The pharyngeal bones were drawn by G. Howes, who with the assistance of Mrs
Clarke provided me with the radiographs.

REFERENCES

BANISTER, K. E. 1972. On the cyprinid fish Barbus alluaudi Pellegrini : a possible inter-
generic hybrid from Africa. Bull. Br. Mus. nat. Hist. (Zool.) 24 (5) : 261-290.

1973- A revision of the large Barbus (Pisces, Cyprinidae) of East and Central Africa.

Bull. Br. Mus. nat. Hist. (Zool.) 26 (i) : 1-148.

BERRY, P. Y. & Low, M. P. 1970. Comparative studies on some aspects of the morphology
and histology of Ctenopharyngodon idellus, Aristichthys nobilis and their hybrid (Cyprinidae).
Copeia (4) : 708-727.

BOULENGER, G. A. 1909. Catalogue of the freshwater fishes of Africa in the British Museum
(Natural History) 1. London.

1911. Catalogue of the freshwater fishes of Africa in the British Museum (Natural History)

2. London.

GAIGHER, I. G. 1975. The occurrence of 'Varicorhinus' and 'rubberlip' mouth forms in the

small scaled yellowfish, Barbus polylepis. Piscator (92) : 162.
GROENEWALD, A. A. v. J. 1958. A revision of the genera Barbus and Varicorhinus (Pisces,

Cyprinidae) in Transvaal. Ann. Transv. Mus. 23 (3) : 263-330.
HOLCIK, J. & DE WITT, J. J. D. ig62a. Character of Rhodeus ocellatus and Acheilognathus

lanceolatus hybrids. Copeia (2) : 277-390.

ig62b. The taxonomic characters of the hybrid Rhodeus. Copeia (4) : 777-778.

HUBBS, CARL L. 1955. Hybridisation between fish species in nature. Syst. Zool. 4 (i) : 1-20.

HUBBS, LAURA & JOHNSON, R. E. 1943. Hybridisation in nature between species of

Catostomid fishes. Contr. Lab. vertebr. Biol. Univ. Mich. 22 : 1-76.

MILLER, R. R. 1952. Hybridisation in nature between the fish genera Catostomus and

Xyrauchen. Pap. Mich. Acad. Sci. 38 : 207-233.

JUBB, R. A. 1968. The Barbus and Varicorhinus species (Pisces : Cyprinidae) of Transvaal.

Ann. Transv. Mus. 26 (4) : 79-97.
POLL, M. 1953. Lac Tanganika (poissons non Cichlidae). Result, scient. Explor. hydrobiol.

lac Tanganika 3 (5 A) : 1-251.
SCHULTZ, L. P. & SCHAEFFER, M. B. 1936. Descriptions of new intergeneric hybrids between

certain cyprinid fishes of the northwestern United States. Proc. biol. Soc. Wash. 49 : i-io.
WORTHINGTON, E. B. & RiCARDO, C. K. 1937- The fish of Lake Tanganyika (other than

Cichlidae). Proc. zool. Soc. Lond. (4) : 1061-1112.

DR K. E. BANISTER

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON SW7 5BD

PLATE i

A hybrid (top) of 133 mm SL, Varicorhinus tanganicae (centre) and Barbus tropidolepis

(bottom).

A LIST OF SUPPLEMENTS
TO THE ZOOLOGICAL SERIES

OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

1. KAY, E. ALISON. Marine Molluscs in the Cuming Collection British Museum
(Natural History) described by William Harper Pease. Pp. 96 ; 14 Plates.
1965. (Out of Print.)

2. WHITEHEAD, P. J. P. The Clupeoid Fishes described by Lacepede, Cuvier and
Valenciennes. Pp. 180 ; n Plates, 15 Text-figures. 1967. £4.80.

3. TAYLOR, J. D., KENNEDY, W. J. & HALL, A. The Shell Structure and Mineralogy
of the Bivalvia. Introduction. Nuculacea-Trigonacea. Pp. 125 ; 29 Plates,
77 Text-figures. 1969. £5.40.

4. HAYNES, J. R. Cardigan Bay Recent Foraminifera (Cruises of the R.V. Antur)
1962-1964. Pp. 245 ; 33 Plates, 47 Text-figures. 1973. £11.90.

5. WHITEHEAD, P. J. P. The Clupeoid Fishes of the Guianas. Pp. 227 ; 72
Text-figures. 1973. £10.65.

6. GREENWOOD, P. H. The Cichlid Fishes of Lake Victoria, East Africa : the
Biology and Evolution of a Species Flock. Pp. 134 ; i Plate, 77 Text-figures.
1974. £3.75. Hardback edition £6.00.

Printed in Great Britain by John Wright and Sons Ltd. at The Stonebridge Press, Bristol BS4 5NU

Bull. Br. Mus. nat. Hist. (Zool.) 30, 5

PLATE i

PLATE 2

Ventral views of the head of Barbus tropidolepis (left), Varicorhinus tanganicae (centre)

and a hybrid of 133 mm SL (right).

Bull. Br. Mus. nat. Hist. (Zool.) 30, 5

PLATE 2

TWO NEW SPECIES OF LARGE

BARBUS (PISCES, CYPRINIDAE)

FROM CENTRAL AFRICA

By K. E. BANISTER

SYNOPSIS

Two new species of large Barbus from central Africa are described : Barbus huloti from the
Lake Albert basin and Barbus humphri from the Ituri system. Barbus huloti is aligned with
Barbus somereni and Barbus mirabilis but the affinities of Barbus humphri are uncertain.

INTRODUCTION

DURING a recent visit to the Institute Royale des Sciences Naturelles, Brussels, I
found in the collections two series of specimens labelled Barbus altianalis from
localities in which this species was unknown. Further examination showed that
each series represented a new species. The series from the Vuda River (a feeder
stream of Lake Albert) is particularly interesting because it represents only the
second known species of large Barbus from that lake basin.

Notes on counts and measurements

The standard length (SL) was taken in the usual manner. The lateral line count
(LL) was taken from the first pore-bearing scale behind the head to the scale lying
lateral to the end of the hypurals. The body depth (D) is the maximum body depth,
usually to be found just in front of the dorsal fin. The anterior limit for the head
length (H) and snout length (Snt) was the premaxillary symphysis, with the pre-
maxillae retracted. The posterior limit for the snout is the anterior margin of the
orbit and for the head length is the most posterior part of the bony edge of the
operculum. The term 'mouth width' (MW) refers to the width across the lower jaw
at the level of the angle of the mouth. The pectoral fin length (Pet) is the total
length of the fin, measured in a straight line from the base of the first fin ray to the
distal extremity of the fin. The measurements were taken in this way because of the
ease of so doing with dial calipers, which were used on all fish except the smallest,
when dividers were used. The caudal peduncle length (CP1) is the horizontal
distance from the posterior angle of the base of the last anal ray to the end of the
hypurals and the caudal peduncle depth (CPd) is the least depth of that part. The
interorbital width (IO) was measured as the least distance between the bony edges
of the interorbital space. The eye diameter (I) is the horizontal diameter of the
visible part of the eye. Dsp symbolizes the height of the dorsal fin from the base
of the first spine to the distal extremity of the longest fin ray. With any measure-
ments which were repeatable on both sides of the fish (e.g. the anterior barbel, Ab,

Bull. Br. Mus. nat. Hist. (Zool.) 30, 5 Issued 28 October 1976

ii

192 K. E. BANISTER

or posterior barbel, Pb) the larger was taken except in cases where a deformity was
obvious.

The majority of measurements were taken to the nearest millimetre ; the exceptions
were some measurements made on small fishes ; these were taken to the nearest
half millimetre.

The mean is symbolized by x, the standard deviation by s.d., the standard error
by s.e. and the number of fish in the sample by n.

DESCRIPTION OF SPECIES

Barbus huloti sp. nov.

HOLOTYPE. The holotype is a fish of 140 mm SL, no. 13289.558 in the collections
of the Institute Royale des Sciences Naturelles, Brussels. Ten specimens ex
no. 13288 are paratypes, as is BMNH. 1975.4. 30 :i. These fishes were collected by
M. Hulot on 4 August 1953 at Zega on the Vuda River, Lake Albert ( = Lake Mobuto
Sese Seko) basin, Zaire, i°44'N, 3O°45'E.

DESCRIPTION. The description is based on 36 specimens, 68-282 mm SL.

n

X

s.d.

s.e.

range

SL

-

-

-

-

68 -282 mm

D

35

28-1

1-8

o-3

25-0-33-7

H

36

27-5

i'3

0-2

22-6-29-4

I

36

6-4

0-8

O-I

4-2- 7-9

10

36

7-8

0-7

O-I

6-6- 9-6

MW

35

6-5

0-6

O-I

5-8- 8-2

Pet

36

22-5

1-4

0-2

19-5-26-1

CP1

36

16-9

i'3

0-2

13-3-19-1

CPd

36

n-9

0-6

O-I

10-8-13-5

Snt

36

8-6

0-9

O-I

6-3-10-0

Ab

36

6-1

1*4

0-2

3-1- 8-6

Pb

36

7'3

i-i

O-2

4-8-10-2

Dsp

34

25-4

2-8

o-5

19-1-30-9

S-d

12

49-3

i'5

0-4

46-3-52-8

S-p

12

52-1

i'4

0-4

50-8-55-4

Unless stated otherwise all measurements are expressed as a percentage of the
standard length.

The body is compressed and shallow with the ventral profile gently convex from
the gill isthmus to the anal fin base (Fig. i). The mouth (PI. i) is subterminal or
ventral. The lower jaw is long with a curved buccal edge. Usually a thin mental
lobe is present but in a few specimens this is greatly reduced, although the lip is
always continuous. The skin of the lips, snout, base of the barbels and adjacent
gular regions has a velvety texture caused by large numbers of minute papillae.

Sections were cut of the papillae but the state of preservation of the tissues was
inadequate for detailed histological examination. The larger specimens (above
about 120 mm SL) have tubercle scars on the skin of the cheeks. In a few of the
largest fishes two rows of tubercles are present on the anal fin and small scattered

TWO NEW SPECIES OF BARBUS

193

2cm

FIG. i. Barbus huloti, holotype, 140 mm SL.

tubercles, or their scars, are present on the top of the head. The eyes are
protuberant in most specimens but this may be a post-mortem effect of perhaps the
result of the fishes having been collected by dynamiting. A few specimens (see
PL i) have a few poorly developed rows of pit organs on their cheeks.

The vertebrae in the twelve specimens radiographed number 18 + 21 (fi), 19 + 20
(13), 19 + 21 (14), 20 + 20 (13) or 20 + 21 (fi). This count includes the vertebrae
incorporated in the Weberian mechanism.

Dorsal fin. There are 4 unbranched rays and 8 (17) or 9 (129) branched rays.
The last unbranched ray is thickened into a smooth, straight spine (£ = 25-4;
s.d. = 2-8; s.e. = o-5 ; range 19-1-30-9). Most specimens have a sheath of scales
covering the bases of the simple rays and the first five or six branched rays. The
dorsal fin origin is in advance of the pelvic fin origin.

The anal fin has 3 unbranched rays and 4 branched rays (136).

Squamation. In the lateral line series there are 24 (12), 25 (14), 26 (17), 27 (fn),
28 (19) or 29 (13) scales. From the dorsal mid-line to the lateral line there are
4^ (f6) or 5^ (130) scales and from the lateral line to the mid-ventral line there are
4| (115), 5 (fi) or 5 1 (fn) scales. Between the lateral line and the base of the
ventral fin there are 2 (fi), 2\ (122) or 3 (17) scales. Scale counts are not obtainable
on all specimens. There are 12 (136) scales around the least circumference of the
caudal peduncle.

The scale striations are disposed radially on the anterior scales but are parallel
or slightly convergent on the posterior scales (Fig. 2).

Pharyngeal bones and teeth (Fig. 3). The pharyngeal teeth number 2.3.5-5.3.2 in
all the 12 specimens examined. There is no significant change in the shape of the
pharyngeal teeth with an increase in the size of the fish. The second tooth of the
inner row is always mammiliform and never develops a molariform crown. The

194

K. E. BANISTER

2 mm

FIG. 2. Barnus huloti, scale striations. Above, the third scale from the row above the
lateral line ; below, the fifteenth scale from the same row.

mean tip-to-tip length of the pharyngeal bones of a specimen of 164 mm SL is
17-0 mm and that of a specimen 216 mm SL is 23-5 mm.

Gill-rakers. In 12 specimens examined there are 10 (fi), n (f2), 12 (f3) or 13 (f6)
gill-rakers on the lower limb of the first gill-arch. The gill-rakers are slightly
curved and obtusely pointed.

TWO NEW SPECIES OF BARBUS

195

Coloration. Alcohol-preserved specimens are dark brown dorsally and paler on
the flanks and belly. The centre of each scale is lighter than the edges. Diffuse
dark pigment is present in the proximal part of the membrane of all fins. The
lateral line is more conspicuous on the caudal peduncle where it is paler than the
background.

ETYMOLOGY. This species is named in honour of the collector, M. Hulot.

DISTRIBUTION. This species is known only from pot-holes in the bed of the Vuda
River, a feeder stream of Lake Albert, i°44'N, 30°45'E, Zaire.

DIAGNOSIS AND AFFINITIES. The only other species of large Barbus recorded
from the Lake Albert basin is Barbus bynni (Banister, 1973 : 27). This nilotic

5mm

FIG. 3. Barbus huloti : left pharyngeal bone of the holotype.

ig6 K. E. BANISTER

species is, by its massive dorsal spine, deep body and parallel, sinuous striations on
the scales, easily distinguished from Barbus huloti.

Superficially, Barbus huloti resembles Barbus altianalis, subspecies of which are
found in Lake Victoria (including Lake Kyoga), Lake Kivu and Lakes Edward and
George and their feeder streams. Meristic and morphometric differences between
the three subspecies are small (see Banister, 1973), but meristic and morphometric
data for specimens of two of the three subspecies within the size range of the Barbus
huloti sample are given in Tables i and 2. The main differences between Barbus

TABLE i

Comparisons of the mean and range of certain characters in Barbus altianalis and B. huloti

B. a.

radcliffii

B. a. eduardianus

B.

huloti

n

= 18

n = 33

n

= 36

X

range

x range

X

range

SL (mm)

90-282

68-280

68-282

I

7-9

5-0-10-0

8-8 6-4-11-5

6-4

4-2- 7-9

Ab

5'3

3-6- 8-6

4-2 2-2- 6-4

6-2

3-1- 8-6

Pb

6-4

4'5- 9-6

5'7 3'9- 7'7

7'3

4-8-10-2

Dsp 19-9 13-6-30-4 18-9 13-2-25-0 25-4 19-1-30-9

TABLE 2

A comparison of the number of scales in the lateral line series in Barbus altianalis radcliffii
(B. a. r.), B. a. eduardianus (B. a. e.) and B. huloti (B. h.)

No. of scales 24 25 26 27 28 29 30 31 32 33 34 35 36

B. a. r. 1353 4

B. a. e. 362644 i

B. h. 2 4 7 ii 9 3

altianalis and Barbus huloti are in scale number as well as in the eye, barbel length
and length of dorsal spine.

There are, in addition, some less quantifiable differences between Barbus huloti
and Barbus altianalis. The latter species has never been observed to develop
tubercles, nor to have papillae around the mouth. The scale striations of Barbus
altianalis are more or less parallel and their direction does not vary with the position
of the scale on the body. The pharyngeal bones and teeth of Barbus altianalis
vary in shape and size (with the diet), a phenomenon not yet observed in Barbus
huloti.

An attempt to explain the absence of Barbus altianalis from Lake Albert was made
by Banister (1973 : 22) and the arguments given there form an essential background
to the current problem. The central part of this problem is to determine the
likelihood of Barbus altianalis populations having moved from Lake Edward via
the Semliki River into Lake Albert in the last 8000-10 ooo years. This date
coincides with the cessation of the Katwe vulcanicity (Bishop, 1969), an event
contemporaneous with the fish extinctions in Lake Edward (Greenwood, 1959 ',
Kendall, 1969). Hitherto there has been no evidence of any faunal interchange via

TWO NEW SPECIES OF BARBUS 197

the Semliki River since that time. This is rather puzzling, especially if the latter
river is compared with the Ruzizi River.

The Ruzizi River dates, in more or less its present form, from about 13 ooo B.P.
(see evidence summarized in Banister, 1973) so that in each case the time scale is
of the same order of magnitude. However, the slope of the two rivers is very
different ; the Ruzizi drops approximately 590 m in 120 km whilst the Semliki only
drops approximately 290 m in 220 km. Yet one fish, Barilius moori (see Poll, 1952),
is known to have ascended the steeper Ruzizi River. Barbus altianalis altianalis
has not descended the Ruzizi River into Lake Tanganyika although it has been
reported from the Sange River, a tributary of the Ruzizi (Poll, 1952). A known
ascent of the Ruzizi River would suggest that current speed alone is unlikely to be a
barrier to migration along the gentler Semliki River. Fryer (1968) postulated that
cold water flowing down the Ruwenzori mountains into the Semliki may form an
effective barrier. However, Beadle (1974 : 139, 177, 181) points out that the main
drop in the Semliki River bed is confined to a small distance and he thinks that it is
these rapids and not the temperature that are the main barrier to the movement
of fishes.

Another factor to consider is the rate of evolution of Barbus altianalis. It is
assumed that this species was originally confined to the headwaters of the River
Zaire before tectonic movements resulted in the formation of the lakes (Banister,
1973). The three subspecies have diverged but little from each other since their
initial isolation. The time of their separation and isolation is in the region of
100 ooo B.P. (the date of the Bufumbiro volcanoes - see Banister, 1973). Within
the three subspecies there is no evidence of a potential for tachytelic evolution such
that a population with Barbus altianalis characters would manifest Barbus huloti
characters within 10 ooo years. There is no morphological nor palaeozoogeo-
graphical evidence that indicates a sufficiently close relationship between Barbus
altianalis and Barbus huloti to justify the inclusion of these species within the same
supraspecific complex.

Barbus somereni Boulenger, 1911 and Barbus mirabilis Pappenheim & Boulenger,
1914 live in the streams of, respectively, the Ruwenzori mountains and the Ituri
system. They both have scales on which the striation pattern differs with the
position of the scale on the body. Barbus huloti displays the same phenomenon.
There are certain characters present in Barbus huloti that are not found in the other
two species ; these include papillae around the mouth, a long dorsal spine, a slender
body and tubercles on the cheeks. Barbus somereni and Barbus mirabilis both have
bulky bodies and a higher modal number of branched rays in the dorsal fin (9 or 10
compared with 8 or 9). The presence of variable scale striation patterns in the
three species is regarded as a synapomorphic character indicating that Barbus huloti
is more closely related to Barbus somereni and Barbus mirabilis than to Barbus
altianalis. This character is interpreted as synapomorphic because (a) it is of
infrequent occurrence and (b) it is theoretically derivable from the commoner
condition of sinuous, parallel striations.

Geographically, there is nothing inconsistent in such a relationship. Barbus
somereni lives in fast-flowing streams up to altitudes of 1750 m (Greenwood, 1966).

I98 K. E. BANISTER

Barbus mirabilis comes from the Ituri River to the west of the rift valley. Barbus
huloti is found in a fast-flowing feeder stream of Lake Albert on the floor of the rift
valley. It was supposed (Banister, 1973) that Barbus somereni and Barbus mirabilis
are the descendants of a once homogeneous population found in the upper reaches of
the formerly more extensive Zaire system. Barbus huloti could, therefore, represent
a segment of this population that became isolated on the floor of the rift valley.

The fishes of the Ituri system are poorly known (see below). The Ituri catchment
is largely separated from the Lake Albert catchment area by the scarp wall of the
rift valley. Only at the south-west part of the lake and towards the Semliki River
is the scarp less severe. There the feeder streams of Lake Albert extend westwards
and the relatively low watershed offers the possibility that at some time in the past
there may have been movement of fishes between Lake Albert and the Ituri system.
Barbus somereni has not, to date, been recorded from Ruwenzori streams flowing into
the Semliki River.

On the available information I align Barbus huloti with Barbus somereni and
Barbus mirabilis. Whether each species represents an isolated population of a
widespread ancestral species or whether Barbus huloti is an evolved isolated population
of one of the other two species cannot, at the moment, be determined.

Barbus humphri sp. nov.

HOLOTYPE. The holotype (Fig. 4) is a fish of 145 mm SL, no. 13289.559 in the
collections of the Institute Royale des Sciences Naturelles, Brussels. Ten specimens
ex no. 13289 are paratypes, as is BMNH. 1975. 4.30:2. These specimens were collected
with some 500 others in the River Tabie, North Kivu district, Zaire, about 25 km
south of Beni.

DESCRIPTION. The description is based on 12 specimens of 82-214 mm SL.

X

SL

D 25-8

H 27-2

I 5-4

IO 7-8

MW 6-6

Pet 19-4

CP1 16-2

CPd 10-8

Snt 8-7

Ab 6-9

Pb 7-8

Dsp 22-9

S-d 46-8

S-p 51-1

Unless stated otherwise all measurements are expressed as a percentage of the
standard length.

The body is shallow, thick and terete. Almost all the specimens have a
conspicuous coating of mucus. The snout is blunt with, in larger fishes, a few very

s.d.

s.e.

range

-

-

82 -214 mm

1-8

°'5

24-0-30-3

i*5

0-4

24-9-30-3

0-8

O-2

4'2- 7'3

0-6

O-2

6-9- 9-0

0-7

O-2

5-9- 7-8

1-3

0-4

17-3-22-1

i'3

0-4

13-4-18-2

0-4

o-i

9-7-11-3

0-5

o-i

7'9- 9-5

1-3

0-4

4-7- 8-8

0-9

0-3

6-1- 9-2

1-8

0-9

15-9-28-0

1-8

°'5

44-9-50-7

1-8

°'5

48-3-54-0

TWO NEW SPECIES OF BARBUS

199

2cm

FIG. 4. Barbus humphri, holotype, 145 mm SL.

small tubercles on its lateral faces. The horseshoe-shaped mouth is subterminal
and has an uninterrupted lower lip. A small mental lobe is occasionally present.
The vertebrae in 12 specimens radiographed number 18 + 21 (fi), 19 +19 (£2),
19 + 20 (12), 19 + 21 (12), 20 + 19 (£2), 21 + 18 (12) or 21 + 19 (i).

Dorsal fin. There are 4 simple rays (fi2), the last of these a stout, rather blunt
spine, sometimes shorter than the longest branched rays. A raised sheath of scales
is always present at the base of the simple rays, and in a few specimens it extends
posteriorly to about the mid-point of the fin. There are 9 (fi2) branched rays. The
dorsal fin is in front of the pelvic fin origin.

Squamation. In the lateral line series there are 22 (f4), 23 (13), 24 (f4) or 25 (fi)
scales. There are 4! (fi2) scale rows between the dorsal mid-line and the lateral
line. Between the lateral line and the ventral mid-line there are 4^ (fi2) scale
rows. From the lateral line to the base of the pelvic fin there are 2,\ (fi2) scale
rows. There are 12 (fi2) scales around the least circumference of the caudal
peduncle. The scales bear parallel striations (Fig. 5).

Pharyngeal bones and teeth. The pharyngeal teeth number 2.3.5-5.3.2 in the 12
specimens examined. The first tooth (Fig. 6) of the inner row is small and com-
pressed, i.e. longer than wide (length taken in the plane of the anterior edentulous
process, width horizontally at 90° to that plane). The second tooth is also
compressed, little enlarged and with a mamilliform crown. The third, fourth and
fifth teeth approach a circular cross-section and become progressively more recurvUL
The teeth of the second and third rows resemble the posterior teeth of the inner row.
The mean tip-to-tip measurement of the pharyngeal bones of a fish of 165 mm SL
is 12-8 mm and that of a specimen of 214 mm SL is 19-2 mm.

Gill-rakers. In the 12 specimens examined there are 10 (13), n (f8) or 12 (fi)
gill-rakers on the lower limb of the first gill-arch. The gill-rakers are scimitar
shaped with small conical processes on the leading edge.

K. E. BANISTER

2mm

FIG. 5. Barbus humphri : fifth scale from the row above the lateral line of a specimen of

145 mm SL.

Coloration. Alcohol-preserved specimens are dark brown above, paler beneath.
The scales on the flanks are dark edged. Traces of dark pigment remain on the
pelvic, pectoral and anal fins.

ETYMOLOGY. This species is named in honour of Dr P. Humphry Greenwood for
his services to African ichthyology.

DISTRIBUTION. This species is known only from the type locality, the River Tabie
about 25 km south of Beni, North Kivu district, Zaire.

FIG. 6. Barbus humphri : left pharyngeal bone.

TWO NEW SPECIES OF BARBUS 201

DIAGNOSIS AND AFFINITIES. The large Barbus species of the head waters of the
Zaire system are poorly known, especially in the Ituri region. Only four species,
Barbus fasolt Pappenheim, Barbus mawambi Pappenheim, Barbus mirabilis
Pappenheim & Boulenger, and Barbus mawambiensis Steindachner, are recorded
from the Ituri region. Barbus paucisquamatus Pellegrin and Barbus longifilis
Pellegrin are known from rivers in the Lowa system to the west of Lake Kivu (and
to the south of the Ituri system).

The outstanding feature of Barbus humphri is the low number of lateral line scales.
Barbus fasolt can also be easily distinguished from Barbus humphri because it has a
flexible last simple ray in the dorsal fin. Barbus mawambi has more scales in the
lateral line series (29 fide Pappenheim & Boulenger, 1914 ; cf. 22-25 m B- humphri)
and 10 branched rays in the dorsal fin (cf. 9). Barbus mirabilis is a bulky-bodied
fish with 28 scales in the lateral line series and a variable pattern of striations on the
scales (see p. 197).

Barbus mawambiensis, the only other species recorded from the Ituri system,
presents a problem. This species was described as Barbus hindii var. mawambiensis
by Steindachner (1911) on the basis of seven, now untraceable, specimens. His
description, as well as his determination of this form as a variety of Barbus hindii,
draws attention to the superficial resemblance between these specimens and Barbus
hindii ( = Barbus oxyrhynchus) from the Athi-Tana system in East Africa. The
differences given by Steindachner between Barbus hindii var. mawambiensis and
Barbus hindii are a deeper caudal peduncle in the former (Cpl/Cpd 1-0-1-9 '> c^-
1-5-1-67) and more (sic) scales in the lateral line (21-24) than in Barbus hindii.
(Boulenger, 1902, gives 25-29 scales in Barbus hindii.) This is perplexing informa-
tion, since the range of the Cpl/Cpd ratios for the four syntypes of Barbus hindii was
quoted by Boulenger (1902) as i^-if, while my own measurements give a mean of
1-38. The mean Cpl/Cpd ratio for all the Barbus oxrhynchus specimens used in an
earlier study (n = io8) was 1-32 (Banister, 1973). The syntypes of Barbus hindii
have 25 (f2) or 26 (f2) scales in the lateral line series ; the range for Barbus oxyrhynchus
is 21-28 scales : 21 (fi), 22 (fio), 23 (fi5), 24 (£36), 25 (130), 26 (f8), 27 (f6) or 28 (fi).
Unfortunately, the types of Barbus hindii var. mawambiensis are untraceable, which
prevents the corroboration of Steindachner 's data. The specimens were deposited
in the Vienna Museum, but a recent search by the curator, Dr Kahsbauer, and myself
failed to find them.

In 1912 Steindachner redescribed Barbus hindii var. mawambiensis as Barbus
mawanbiensis based on a sample of fish from the Ja River, Cameroon. The spelling
(mawanbiensis) for the specific name must be regarded as a misprint for mawambiensis
There are inconsistencies between his 1911 and 1912 accounts, e.g. in 1912 he states
that there are fewer scales in the lateral line series of Barbus mawambiensis than in
Barbus hindii - the reverse, but correct, interpretation of the 1911 information. It
is, of course, possible that Steindachner muddled scale size with scale number. He
does not give the size, nor the number of the specimens examined, merely their
locality - the Ja River, Cameroon.

In 1914 Steindachner published another paper on a collection of fishes from the
Ja River. Included in this is a further description of Barbus mawambiensis based

202 K. E. BANISTER

on (in translation) 'three small examples from the Ja River, via Dr Haberer and from
the Ituri via Herr Grauer'. The confusion is further increased by the statement
that the largest of the three fishes is illustrated, although the scale on the figure
shows that the smallest of the three fish is figured. (Boulenger, 1916, rightly pointed
out that on Steindachner's pi. 3, figs i and 3 have been transposed.) Tables of
measurements for each of the three specimens are given but which specimen or
specimens was or were from the Ja River, and which from the Ituri, is not stated.
The Ja and Ituri rivers are both in the Zaire system and are tributaries of respectively
the Sangha and Aruwimi rivers which flow into the north side of the Zaire some
1000 km apart. Generally, the large Barbus spp. are found in the upper reaches of
the tributaries, not in the main stream (pers. obs.). The two localities are some
1600 km distant and are separated by a wide sluggish river with few favourable
habitats for large Barbus spp. Trewavas (1974 : 344) comments that the fauna
of the Ja River has much in common with the rivers of Cameroon (therefore not
with the Zaire system) and is also of the opinion that Steindachner's Barbus
mawambiensis from the Ja resemble Barbus batesii. Although it is not known which
of Steindachner's specimens came from which locality (see above), I agree with
Trewavas that the Barbus spp. from the Ja are more likely to be related to those of
west Africa (e.g. Barbus batesii} than to those of central Africa.

Thys van den Audenaerde (1966 : 90) compared the fauna of the upper Ja (above
Molundu) with that of the Nyong River and from the faunistic and geographical
evidence concluded that the upper part of the Ja had been captured by the Sangha
from the upper part of the Nyong. He describes the fauna below Molundu as
typically central Congolese ( = Zairoise) . The only information given by Steindachner
is that his specimens came from the Molundu district.

It would seem that the presence of the same species of large Barbus in the Ituri
and Ja rivers is most unlikely and would present some zoogeographical enigmas.

The concept of Barbus mawambiensis as based on the points of agreement in
Steindachner's three descriptions seems to indicate that there is a species of fish in
the Ituri River which superficially resembles Barbus hindii but has a low number of
lateral line scales. There are two specimens in the collections of the British Museum,
nos. 1944.12.4:2-3 collected by Ricardo and Owen in the Epulu River (Ituri system)
which have these characters and have been identified as Barbus mawambiensis. If
this determination is correct then they can be distinguished from Barbus humphri
by a stronger dorsal spine and a much deeper, more compressed body. These two
fish also closely resemble Steindachner's figured specimen. However, the
identification of these two fishes must remain tentative until the syntypes of Barbus
mawambiensis are found.

At the moment I can find no close relatives to Barbus humphri ; further collections
from the Ituri may help to settle its relationships.

ACKNOWLEDGEMENTS

I am most grateful to M. J. P. Gosse of the Institute Royale des Sciences Naturelles
in Brussels for drawing my attention to these fishes and for giving me the opportunity

TWO NEW SPECIES OF BARBUS 203

to examine them. Valuable comments on the manuscript were provided by Drs
P. H. Greenwood and P. J. P. Whitehead. Mrs M. Clarke assisted most efficiently
with the calculations. The drawings are the work of Miss M. Holloway and Mr
G. Howes.

REFERENCES

BANISTER, K. E. 1973. A revision of the large Barbus (Pisces, Cyprinidae) of East and

Central Africa. Studies on African Cyprinidae II. Bull. Br. Mus. nat. Hist. (Zool) 26

(i): 1-148.

BEADLE, L. C. 1974. The inland waters of tropical Africa. London.
BISHOP, W. W. 1969. Pleistocene stratigraphy in Uganda. Mem. geol. Surv. Uganda

10: 1-128.
BOULENGER, G. A. 1902. On the fishes collected by Mr. S. L. Hinde in the Kenya district,

East Africa with descriptions of four new species. Proc. zool. Soc. Lond. : 221-224.

1916. Catalogite of the freshwater fishes of Africa in the British Museum (Natural History)

4. London.

FRYER, G. 1968. The parasitic Crustacea of African freshwater fishes : their biology and

evolution. /. Zool. Lond. 156 : 45-95.
GREENWOOD, P. H. 1959. Quaternary fish fossils. Explor. Pare natn. Albert Miss. J. de

Heinzlin de Braucourt 4 (i) : 1-80.

1966. The fishes of Uganda, 2nd edn. Kampala, the Uganda Society.

KENDALL, R. L. 1969. An ecological history of the Lake Victoria basin. Ecol. Monogr.

39 (2) : 121-176.
PAPPENHEIM, P. & BOULENGER, A. G. 1914. Fische. Wiss. Ergebn. dt. ZentAfr. Exped.

Zool. (5), Zoologie 3 : 225-260.
POLL, M. 1952. Poissons de rivieres de la region des lacs Tanganika et Kivu recueillis par

G. Marlier. Rev. Zool. Bot. afr. 46 (3-4) : 221-236.
STEINDACHNER, F. 1911. . . . iiber einige neue und seltene afrikanische siisswasser Fische.

Anz. Akad. Wiss. Wien 48 : 531-535.

1912. Zur Fischfauna des Dscha, eines sekundaren Nebenflusses des Congo im Bezirke

Molundu des siidlichen Kamerun. Anz. Akad. Wiss. Wien 49 : 443-449.

1914. Zur Fischfauna des Dscha, eines sekundaren Nebenflusses des Congo im Bezirke

Molundu Kamerun. Denkschr. Akad. Wiss. Wien 89 : 1-64.

THYS VAN DEN AUDENAERDE D. F. E. 1966. Les Tilapia (Pisces, Cichlidae) du Sud-Cameroun
et du Gabon, etude systematique. Ann. Mus. r. Afr. entr. 8° Zool. 153 : 1-95.

TREWAVAS, E. 1974. The freshwater fishes of rivers Mungo and Meme and lakes Kotto,
Mboandong and Soden, West Cameroon. Bull. Br. Mus. nat. Hist. (Zool.) 26 (5) : 331-419.

DR K. E. BANISTER

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON SW7

PLATE i

Barbus huloti : ventral view of the head of a specimen of 214 mm SL to show the mouth,
the papillae and the pit lines on the cheeks.

Bull. Br. Mus. nat. Hist. (Zool.) 30, 5

PLATE i

GENERAL

2 6 NOV i97(

MISCELLANEA

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. 6

LONDON: 1976

MISCELLANEA

Pp 205-268 ; 9 Plates ; 19 Text-figures

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. 6

LONDON: 1976

THE BULLETIN OF THE BRITISH MUSEUM

(NATURAL HISTORY), instituted in 1949, is
issued in five series corresponding to the Scientific
Departments of the Museum, and an Historical series.

Parts will appear at irregular intervals as they
become ready. Volumes will contain about three or
four hundred pages, and will not necessarily be
completed within one calender year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 30 No. 6 of the Zoology series.
The abbreviated titles of periodicals cited follow those
of the World List of Scientific Periodicals.

World List abbreviation :
Bull. Br. Mus. nat. Hist. (Zool.)

ISSN 0007-1498

Trustees of the British Museum (Natural History), 1976

BRITISH MUSEUM (NATURAL HISTORY)

Issued 25 November, 1976 Price £6-75

CONTENTS

Page
A new genus and two new species of copepod parasitic on freshwater fishes.

By G. A. BOXSHALL .......... 209

A new species of Chonopeltis (Crustacea : Branchiura) from southern Africa.

By G. A. BOXSHALL .......... 217

Microlaophonte trisetosa sp. nov., a new harpacticoid copepod from southern

Florida. By G. A. BOXSHALL ........ 223

A new species of Amphithoe (Pleonexes) (Amphipoda : Amphithoidae) from

the north-east Atlantic with a redescription of A. (P.) gammaroides (Bate).

By R. J. LINCOLN 229

The first zoeal stage of Scopimera crabricauda Alcock (Crustacea, Brachyura,

Ocypodidae). By A. L. RICE ........ 243

Asterozoa from Amsterdam and St Paul Islands, southern Indian Ocean.

By AILSA M. CLARK .......... 247

Restriction of the family Poraniidae, sensu Spencer & Wright, 1966

(Echinodermata : Asteroidea). By F. H. C. HOTCHKISS & AILSA M. CLARK 263

A NEW GENUS

AND TWO NEW SPECIES OF

COPEPOD PARASITIC ON FRESHWATER FISHES

By G. A. BOXSHALL

INTRODUCTION

THE distribution of known species of lernaeid parasites of freshwater fishes in the
Old World is notably patchy, with the majority of species known either from Africa
or from eastern Asia. Between these two foci, in the region extending from India to
the Mediterranean Sea, relatively few species have been recorded. It is probable
that this paucity merely reflects the lack of fieldwork in this region ; thus the
discovery of two new species, one from Iraq and the other from India, belonging to a
new genus of lernaeid comes as no surprise.

DESCRIPTION OF NEW GENUS AND SPECIES

PSEUDOLAMPROGLENA gen. nov.

DIAGNOSIS. Lernaeidae. First pedigerous somite partially incorporated into
cephalothorax. Second to fourth pedigerous somites distinct, each subdivided into
anterior and posterior portions by a transverse groove. Genital complex comprising
fused fifth pedigerous somite and genital somite. Genital apertures on dorsal surface
of genital complex. Maxilla 2-segmented, inflated distal segment bearing single
claw medially. Maxilliped small, flattened and lying in groove posterior and lateral
to maxilla; 3-segmented with terminal segment bearing weak setiform spine (s)
medially. Maxilliped bases connected by fleshy ridge. Thoracic legs 1-4 biramous,
with 3-segmented exopod and 2-segmented endopod. Leg 5 a simple lobe with 2
apical setae. Parasitic on gills of freshwater teleosts. Male unknown.

Type species P. annulata sp. nov.

Pseudolamproglena annulata sp. nov.

(Figs la-g, 2a-e)

DIAGNOSIS, Body cylindrical and distinctly segmented (Fig. la, b). Cephalo-
thorax broad, dorsal surface concave in holotype, comprising 20-25% °f total body
length. Anterior portion of first pedigerous somite incorporated into cephalothorax ;
posterior portion free, rounded dorsally and narrower than both cephalothorax and
second pedigerous somite. Second to fourth pedigerous somites separated by
intersegmental sutures and subdivided into anterior and posterior portions by a
transverse groove. Thoracic legs located anterior to groove. Thoracic somites
increasing in size posteriorly. Genital complex small, narrower than fourth

Bull. Br. Mus. nat. Hist. (Zool.1 30, 6 Issued 25 November 1976

G. A. BOXSHALL

O2mm

g

FIG. I. Pseudolamproglena annulata sp. nov. female : (a) dorsal entire ; (b) lateral
entire ; (c) cephalothorax, postero-ventral ; (d) antennule, ventral ; (e) oral region,
ventral ; (f) maxilla, lateral ; (g) maxilla, ventral. Bar scale : a, b, 0-5 mm ; c, 0-2
mm ; e, o-i mm ; d, f, g, 50 (j.m.

COPEPOD PARASITES OF FRESHWATER FISHES 211

pedigerous somite and with genital apertures situated dorso-laterally. Abdomen
elongate, consisting of 2 indistinctly divided somites. Posterior margin of anal
somite bilobed bearing medially directed furcal rami. Total body length from
i '74 to 2-20 mm, with a mean of 1-97 mm (based on 2 specimens).

Egg strings uniseriate, containing up to 14 eggs. Eggs maturing in oviduct
visible through body wall of second to fourth pedigerous somites.

Antennule (Fig. id) situated on ventral surface near anterior margin of cephalo-
thorax, directed postero-laterally. Segmentation obscure with only the apical
segment clearly delimited. Armature comprising 14 setae on anterior margin of
proximal segment, 4 setae on distal segment and 2 aesthetasc-like structures, one on
each segment. Antenna apparently absent.

Oral region occupied by large trilobed structure (Fig. le), probably representing the
labrum, anterior to which lies a transverse ridge. Maxillule absent. Maxilla
(Fig. if, g) 2-segmented, proximal segment providing a broad base for swollen,
gobular distal segment. Entire surface of distal segment ornamented with minute
tubercles. Single, dorsally curved spine present on medial surface of segment.
Maxilliped inconspicuous, dorso-ventrally flattened and lying in a groove in cephalo-
thorax (Fig. ic, 2a). Indistinctly 3-segmented with a broad base, elongated middle
segment and subquadrate terminal segment ; single setiform spine located on
medial surface. Maxilliped bases connected by ridge of tissue.

Thoracic legs 1-4 (Fig. 2b, c) similar, biramous. Sympod projecting from
body surface, bearing single seta lateral to exopod base. Endopod indistinctly
2-segmented, exopod 3-segmented. Leg i armature comprising 2 unequal setae at
tip endopod, 2 at tip exopod and i lateral spine on basal segment of exopod. Legs
2-4 unarmed apart from i lateral spine on basal segment of exopod. Leg 5 (Fig. 2e) a
simple process with 2 apical setae.

Furcal ramus bearing 4 setae (Fig. 2d) ; i on lateral margin, i on medial margin
and i at each posterior corner. Two small papillae present on distal margin.

ETYMOLOGY. The generic name proposed indicates a close affinity to the genus
Lamproglena Nordmann. The specific name alludes to the very distinct segmen-
tation in this species.

MATERIAL EXAMINED. Type material 2 ovigerous $$. Holotype $ reg. no.
1975.800 ; paratype reg. no. 1975.801 deposited in the collections of the British
Museum (Natural History). The material was obtained from the gills of Cyprinion
macrostomus (caught in the river Tigris at Mosul, Iraq) present in the British Museum
(Natural History) collections. Additional material : 5 ??, all either damaged or
incomplete, collected by Mr Z. I. Fattohy from the gills of C. macrostomus from the
same locality. Three of these specimens donated to the British Museum (Natural
History) by Mr Fattohy and Dr M. H. Kasim (Mosul University).

Pseudolamproglena simplex sp. nov.
(Figs 2f-i, 3a-d)

DIAGNOSIS. Body segmentation and tagmosis as in P. annulata but less distinct
and without intersegmental sutures between the pedigerous somites (Fig. 2f, g).

G. A. BOXSHALL

50pm

0-5mm

FIG. 2. Pseudolamproglena anmilata sp. nov. female : (a) maxilliped, ventral ; (b) leg i,
ventral ; (c) leg 4, ventral ; (d) furcal ramus, ventral ; (e) genital complex, ventral.
Pseudolamproglena simplex sp. nov. female : (f ) dorsal entire ; (g) lateral entire ; (h)
antennule, ventral ; (i) antenna, lateral. Bar scale : a-d, h, i, 50 [xm ; e, 0-2 mm ;
f, g, 0-5 mm.

COPEPOD PARASITES OF FRESHWATER FISHES

213

50>inn

FIG. 3. Pseudolamproglena simplex sp. nov. female : (a) cephalothorax, ventral ; (b)
maxilla, ventral ; (c) maxilliped, ventral ; (d) furcal ramus, ventral. Bar scale : a,
0-2 mm ; b-d, 50 (im.

Cephalothorax comprising about 20% of total body length. Cephalothorax with
prominent anterior ridge visible from the dorsal aspect. Second to fourth pedigerous
somites similar in width. Genital complex of holotype (a gravid female) with
conspicuous dorsal swellings marking the genital apertures. Total body length from
1-93 to 2-45 mm, with a mean of 2-25 mm (based on 5 specimens).

Only i specimen ovigerous, egg string containing 20 eggs. Mature eggs in oviduct
visible in second to fourth pedigerous somites.

Position of all cephalic appendages illustrated in Fig. 3a. Antennule 2-segmented
(Fig. 2h) ; large basal segment bearing 10 setae on anterior margin and i on posterior
margin, apical segment armed with 8 setae and an aesthetasc-like structure. Antenna
(Fig. 21) curved posteriorly, indistinctly 2-segmented and carrying an apical armature
of about 5 setae.

2i4 G- A- BOXSHALL

Oral region occupied by simple hemispherical lobe, probably representing the
labrum (Fig. 3a). Transverse ridge present on ventral surface anterior to labrum.
Maxillule (Fig. 3a) an unarmed fleshy process situated lateral to labrum. Maxilla
2-segmented with broad basal segment and massive, inflated distal segment. Distal
segment (Fig. 3b) marked by transverse constriction and armed with single curved
spine on medial surface. Maxilliped indistinctly 3-segmented (Fig. 3c). Basal
segments of maxilliped pair connected by transverse ridge on ventral surface.
Middle segment elongate, cylindrical. Terminal segment armed with 2 basal setae
on postero-medial surface.

Thoracic legs 1-4 similar to those of P. annulata both in segmentation and
armature. Leg 5 positioned anteriorly on ventral surface of genital complex,
comprising a simple process bearing 2 apical setae.

Furcal rami (Fig. 31) armed with 6 marginal setae ; i lateral, i medial and 4 distal.

ETYMOLOGY. The specific name refers to the simple, hemispherical structure of
the lobe in the oral region.

MATERIAL EXAMINED. 6$$ (i ovigerous). Holotype $ reg. no. 1975.803; 5
paratypes reg. nos 1975.804-808. deposited in the collections of the British Museum
(Natural History). All specimens collected by Mr M. K. Jyoti from the gills of
Labeo diplostomus caught in Dal Lake, Kashmir, India.

HOST-PARASITE RELATIONSHIPS

The two attached P. annulata observed were located on the anterior surface of the
posterior hemibranch of gills 2 and 3. A single primary gill filament had been
grasped between the maxillae and secured by the insertion of the spines. P. simplex
attaches to the host in an identical manner but nearer to the distal tip of the gill
filament than P. annulata. In both species the maxilla appears to be the only
attachment organ and the massive development of this limb presumably compensates
for the reduction of the maxilliped.

The host tissue response to the presence of the parasite was negligible in P. simplex
infections. Different degrees of epithelial hypertrophy were observed associated
with the two specimens of P. annulata, with a localized, minor response in one case
and, in the other, a massive response which resulted in the parasite being completely
enclosed by host tissue to the level of the genital complex. It has been suggested
(Sproston et al., 1950) that envelopment of the parasite provides additional security
to the attachment of the parasite.

REMARKS

The new genus can be readily assigned to the subfamily Lamprogleninae of the
family Lernaeidae. Of the two other genera in this subfamily, Lamproglena
Nordmann and Lamproglenoides Fryer, the new genus is more closely allied to the
former. Although the genus Lamproglena comprises a quite heterogeneous group of
species the two new species could not be included within this genus without an
unacceptable broadening of the generic diagnosis. The genus Pseudolamproglena is

COPEPOD PARASITES OF FRESHWATER FISHES 215

erected to contain these two species which form a well-defined group distinguishable
from species of Lamproglena by the structure and armature of the maxillae and
maxillipeds and by their distinctive somitic segmentation.

Fryer (1968) recognized several evolutionary trends within the Lernaeidae.
Amongst these are the suppression of external signs of segmentation and tagmosis,
the development of a 'neck', reduction of antennules and antennae and the reduction
of the number of terminal spines on the maxilliped. Pseudolamproglena exhibits a
mosaic of primitive and advanced characters with respect to these trends. It
retains a primitive degree of external somitic segmentation and tagmosis combined
with the specialized condition of the maxilla and maxilliped and, in P. annulata, with
the apparent absence of the antenna. This combination of characters suggests that
Pseudolamproglena diverged from the ancestral Lamproglena-like stock at an early
date and that specialized adaptations of the mouthparts evolved independently.

ACKNOWLEDGEMENTS

I would like to thank Dr G. Fryer for his valuable comments on some of these
specimens and Dr R. J. Lincoln for his criticism of the manuscript. I am grateful
to Mr Z. Fattohy and Dr M. H. Kasim (Mosul University, Iraq) whose donation
first brought P. annulata to my attention.

REFERENCES

FRYER, G. 1968. The parasitic Crustacea of African freshwater fishes ; their biology and

distribution. /. Zool., Lond. 156 : 45-95.
SPROSTON, N. G., YIN, W. Y. & Hu, Y. T. 1950. The genus Lamproglena (Copepoda para-

sitica). The discovery of the life histories and males of two Chinese species from food

fishes, revealing their relationship with Lernaea and of both to the Cyclopoidea. Sinensia,

Shanghai (n.s.) 1 : 51-84.

DR G. A. BOXSHALL

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON swy 560

A NEW SPECIES OF

CHONOPELTIS (CRUSTACEA: BRANCHIURA)
FROM SOUTHERN AFRICA

By G. A. BOXSHALL

INTRODUCTION

THE genus Chonopeltis Thiele, 1901 is endemic to Africa and comprises seven known
species. The southernmost record of this genus is that of C. meridionalis Fryer from
the Limpopo river system in Rhodesia (Fryer, 1968). It was interesting, therefore,
to discover some specimens of Chonopeltis in a collection of crustaceans parasitic on
freshwater fishes of the Orange river system. The specimens represent a new species
which is described below.

DESCRIPTION OF NEW SPECIES

Chonopeltis australis sp. nov.

DIAGNOSIS. With characters of genus. Female body squat, with 2 longitudinal
bands of pigment spots on thorax. Second maxilla with scale-covered prominence
on proximal portion. Leg 3 less than twice as long as leg 4. Male lacking posterior
projections on leg 2 sympod and lacking process on exopod of leg 4.

DESCRIPTION. Adult female. General appearance squat (Figs la, 2a), large
trifoliate carapace comprising about 54% of total body length. Lateral lobes of
carapace extending posteriorly to cover the bases of leg 2 ; oval respiratory areas on
ventral surface of each lateral lobe similar to other species of genus. Chitinous
supporting rods present in anterior lobe of carapace. Eyes small ; ocellus small,
distance between ocellus and eyes equal to interocular distance. Segmentation of
thorax distinct. Two longitudinal bands of dark pigment spots present on thorax,
dorsal to uteri. Abdomen comprising 20-24% of total body length. Furcal rami
minute. Spermathecae short, extending to level of bifurcation of abdomen in some
specimens.

First antenna (Fig. 2e) slender, indistinctly 5-segmented ; apical segment bearing
3 spines. Mandible (Fig. 2b) curved, bearing 3-4 rows of denticles on convex surface
and a row of spinules on concave margin. Suckers (first maxillae) as in other species
of the genus. Second maxillae (Fig. 2d) with a 2-segmented proximal portion and
3-segmented distal portion ; scale-covered prominence present on posterior surface
near apex of second segment opposing a small elevated region on third segment ;
2 small claws situated in depression at tip of terminal segment. Legs 1-4 biramous ;
sympod and both rami armed with single rows of setae ; legs 1-3 similar in size,
leg 4 about two-thirds as large as leg 3. Natatory lobes of leg 4 (Fig. 2c) extensive,
meeting at ventral midline, with spinules present on posterior and medial margins.

Bull. Br. Mus. nat. Hist. (Zool.) 30, 6 Issued 25 November 1976

218

G. A. BOXSHALL

FIG. i. Chonopeltis australis sp. nov. : (a) adult female, dorsal ; (b) adult male, dorsal.

A NEW SPECIES OF CHONOPELTIS

219

0-1mm

•2mm

FIG. 2. Chonopeltis australis sp. nov. : (a) adult female, ventral ; (b) mandible, ventral ;
(c) natatory lobes female, ventral ; (d) second maxilla female, ventral ; (e) first antenna
female, ventral.

G. A. BOXSHALL

FIG. 3. Chonopeltis australis sp. nov. male : (a) leg 2 ; (b) leg 3 ; (c) leg 4.

Body length 5-0-8-9 mm, with a mean of 6-8 mm (based on 8 specimens).

Adult male. Carapace comprising about 57% and abdomen about 34% of total
body length (Fig. ib). Anterior appendages as in female. Legs 2-4 showing sexual
characters ; leg 2 without any projections on posterior margin of sympod (Fig. 3a) ;
sympod of leg 3 (Fig. 3b) bearing an extensive marginal flap posteriorly ; leg 4
(Fig. 3c) with an antero-laterally directed conical process on sympod, exopod simple
without any processes.

A NEW SPECIES OF CHONOPELTIS 221

Body length 3-8-5-8 mm, with a mean of 4-5 mm (based on 8 specimens).

MATERIAL EXAMINED. 9$$ (i damaged) and 8<$$ specimens. Holotype $ reg. no.
1975.1092 ; paratypes reg. nos 1975.1093-1108 ; deposited in the collections of the
British Museum (Natural History).

DISTRIBUTION. C. australis was collected from the body surface of Labeo capensis
Smith and L. rosae Steindachner caught at two sites : near Vereeniging on the Vaal
river and near Potchefstroom, in the Boskop Dam reservoir on the Mooi river. Both
the Vaal and the Mooi are part of the Orange river system.

REMARKS. The new species can readily be placed in a group of species, together
with Chonopeltis brevis Fryer and C. meridionalis, which is characterized by the
possession of two longitudinal bands of pigment spots on the thorax. Also in
common with these species C. australis is found externally on the host. This is in
keeping with the suggestion made by Fryer (1964) that the pigment bands serve to
protect developing eggs in the uteri from intense illumination and are therefore found
only in species of Chonopeltis which are external body surface parasites.

The males of C. australis can be distinguished from those of C. brevis and C.
meridionalis by the absence both of any posterior projections on the sympod of leg 2
and of any process on the exopod of leg 4. The females of these three species are
very similar and trivial morphological characters must be used to distinguish between
them. In C. brevis leg 3 is more than twice as long as leg 4 whereas in C. australis and
C. meridionalis it is less than twice as long. The natatory lobes of leg 4 in C.
meridionalis are rounded and less extensive laterally than in C. australis. In addition
the scale-covered prominence on the proximal portion of the second maxilla of
C. australis is a distinctive character.

The structure of the second maxilla of C. australis suggests that it is able to function
in a prehensile manner although it is not so obviously prehensile as in C. schoudeteni
Brian and C.flaccifrons Fryer, both of which possess a large digitiform process on the
proximal portion of the appendage.

ACKNOWLEDGEMENTS

I would like to thank Dr H. Schoonbee (Rand Afrikaans University, Johannesburg)
for donating the material to the Museum and for providing data on the distribution
of the species and Dr G. Fryer (Freshwater Biological Association) for reading and
commenting upon the manuscript.

REFERENCES

FRYER, G. 1964. Further studies on the parasitic Crustacea of African freshwater fishes.
Proc. zool. Soc. Lond. 143 : 79-102.

1968. The parasitic Crustacea of African freshwater fishes ; their biology and distri-
bution. /. Zool., Lond. 156 : 45-95.

DR G. A. BOXSHALL

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON SWy 5BD

13

MICROLAOPHONTE TRISETOSA SP. NOV.,

A NEW HARPACTICOID COPEPOD

FROM SOUTHERN FLORIDA

By G. A. BOXSHALL

INTRODUCTION

THE copepod described below was isolated by Dr J. D. George (British Museum
(Natural History)) from laboratory cultures of the polychaete Capitellides giardi
Mesnil. Dr J. Fell and Mrs R. Hendrix (University of Miami) originally established
the cultures by inoculating agar plates with decomposing mangrove leaves during
investigations into the fungi and nematodes of the southern Florida red mangrove
forest detrital system. The cultures have been successfully maintained since August
1972 and the culture technique, together with certain aspects of the biology of the
new harpacticoid species, is described in detail by George (1976).

The material upon which this study is based was provided by Dr George in July
1975. It was examined both by phase contrast microscopy and by scanning
electron microscopy (SEM). Drawings were made with the aid of a camera lucida
and from SEM photographs.

DESCRIPTION OF NEW SPECIES

Microlaophonte trisetosa sp. nov.

TYPES. Holotype $ registration no. 1975.1255, allotype <$ registration no.
1975.1256, paratypes registration nos 1975.1257-1267 ; deposited in the collections
of the British Museum (Natural History). Cultures established using material
collected near Matheson Hammock, Miami, Florida, U.S.A.

ADULT FEMALE. Body slender, cylindrical, with somites clearly denned (Fig.
la-c). Cephalothorax about as long as 3 succeeding somites combined, with
sensillae around its lateral and posterior margins. Rostrum short, with 2
sensillae. Each thoracic somite with a row of sensillae dorsally and laterally along
the posterior margin. Genital double somite subdivided by a continuous dorsal and
lateral chitinous strip armed with sensillae, posterior margin of double somite also
provided with sensillae except on mid-ventral portion. Antepenultimate somite
similarly provided with sensillae ; penultimate somite without sensillae but with a
continuous ventro-lateral and ventral row of spinules. Anal somite with a sensilla on
each side of the operculum and with a row of spinules ventrally and ventro-laterally
around the bases of the caudal rami.

Bull. BY. Mus. nat. Hist. (Zool.) 30, 6 Issued 25 November 1976

224

G. A. BOXSHALL

FIG. i. Microlaophonte trisetosa sp. nov. : (a) female, dorsal ; (b) abdomen, ventral
(c) abdomen, lateral ; (d) male, dorsal ; (e) abdomen, ventral ; (f ) abdomen, lateral.

A NEW HARPACTICOID FROM FLORIDA

225

FIG. 2. Microlaophonte trisetosa sp. nov. : (a) female first antenna, dorsal ; (b) second

antenna, ventral ; (c) mandible, posterior ; (d) first maxilla, posterior ; (e) second

maxilla, posterior ; (f ) maxilliped, ventral ; (g) leg i, anterior ; (h) leg 2, anterior ;
(i) leg 3, anterior ; (j) leg 4, anterior.

226

FIG. 3. Microlaophonte trisetosa sp. nov. : (a) female leg 5, anterior ; (b) caudal ramus,
ventral ; (c) male leg 3, anterior ; (d) first antenna, dorsal ; (e) leg 5, ventral.

Caudal rami (Fig. 3b) not markedly divergent, about 2 times longer than broad ;
with a pointed process at the postero-lateral angle ; whole surface with densely
scattered minute spinules.

First antenna (Fig. 2a) 6-segmented, third segment longest, about 3 times longer
than fourth which carries a long aesthetasc ; last segment longer than 2 preceding
segments combined.

Second antenna (Fig. 2b). Coxa unarmed ; allobasis with a plumose seta on
outer margin ; exopod small bearing i lateral and 2 apical plumose setae. Endopod
becoming flattened distally, with a dense covering of spinules and i distal spine on
anterior margin, 4 strong spines and 2 geniculate setae on the distal margin and
2 spinule rows near apex on ventral surface.

Mandible (Fig. 2c). Praecoxa provided with a bidentate pars incisiva and a
strong spine terminally and a slender seta subterminally. Palp i-segmented with

1 apical and 3 lateral setae.

First maxilla (Fig. 2d). Arthrite of praecoxa with a proximal row of slender
spinules and, along the distal margin, i very robust spine with spinules, about
5 slender spines and a spinulate seta. Coxa with a proximal row of spinules and

2 terminal setae, i spinulate distally ; basis with 3 apical setae, i strong and distally
spinulate ; endopod represented by 4 setae.

Second maxilla (Fig. 2e). Syncoxa with 3 endites : proximal endite short with a
single seta ; middle endite with 2 stout setae ; distal endite with 3 setae. A row of

A NEW HARPACTICOID FROM FLORIDA 227

spinules and a patch of minute denticles also present at outer corner of syncoxa.
Basis drawn out into a slightly curved claw accompanied by i anterior, i posterior
and 2 dorsal setae ; endopod represented by 3 setae.

Maxilliped (Fig. 21). Basis with a single seta ; endopod 2-segmented with apical
segment forming a slender claw bearing a spinule basally.

Leg i (Fig. 2g) . Coxa with spinule row along outer margin ; basis with proximal
spinule row around a small anterior tubercle, a spinule row near base of endopod,
another row near base of exopod and 2 setae, i located on outer margin proximal to
the exopod and the other on the anterior surface near the inner margin. Exopod
i-segmented, although in some specimens an incipient subdivision is apparent, with
i proximal and 2 distal setae on the outer margin, i short and 2 long setae on the
distal margin and a distal spinule row. Endopod well developed ; first segment long
with spinules on proximal portion of inner margin, second segment with spinules on
inner and outer edges ; third segment forming a curved claw.

Legs 2-4 (Fig. 2h-j). Coxa unarmed ; basis with an outer seta and a single spinule
row on ventral surface between bases of rami. Exopod 3-segmented ; first and
second segments similar in length and longer than third ; rows of spinules present on
all exopod segments. Endopod 2-segmented ; first segment larger, with minute
spinules on lateral surface and longer hairs on inner surface ; second segment of leg 2
with a conspicuous proximal spinule ; endopod of leg 4 shorter, not reaching beyond
end of second exopod segment. Spine and seta formula :

EXOPOD

ENDOPOD

i

2

3

i

2

o

I

212

o

2 2 O

o

I

222

o

3 2 I

o

I

222

o

221

Leg 2
Legs
Leg 4

Spines and setae closely approximated and exact point of insertion on internal or
distal margin sometimes difficult to discern.

Leg 5 (Fig. 3a). Baseoendopod longer than exopod ; anterior surface with many
rows of spinules ; outer margin with a single seta, inner margin with 2 proximal
spinules, i seta at the base of the endopod portion and a spine and a long seta
distally ; distal margin with i seta and 2 spines. Exopod with 4 spinules on anterior
surface and with 6 setae around the outer and distal margins.

Ovisac with 6-10 eggs.

Body length of female ranging from 0-47 to 0-51 mm, with a mean of 0-49 mm.

ADULT MALE. Body more slender and shorter than female (Fig. id-f). Ornamen-
tation of abdominal somites similar to female but with additional ventral spinule
rows located anterior to the sensillae on the ante- and penultimate somites.

First antenna (Fig. 3d) 6-segmented, subchirocerate ; aesthetasc present on fourth
segment.

Legs 2 and 4 as in female but outer spines on exopod longer and more robust ;
inner seta on second exopod segment usually absent from leg 4. Leg 3 (Fig. 3c) with
3-segmented endopod ; second segment with a mucroniform process at outer distal
corner, third segment with 4 setae.

228 G. A. BOXSHALL

Leg 5 (Fig. 3e). Baseoendopod with a spinule row on inner margin, a long seta on
outer margin and some scattered spinules located anterior and lateral to base of
exopod; exopod broader than long, with a row of spinules, i seta at outer distal
corner and 3 distal setae.

Leg 6. Each leg comprising a long outer seta and a short inner seta.

Mean body length 0-40 mm, with a range of 0-38 to 0-43 mm.

VARIATION. Ten specimens of each sex were dissected to examine possible
variation in the armature of legs 2-4. No variation was found in the female. In the
male the inner seta of exopod segment 2 was present on leg 2 in all 10 specimens, on
leg 3 in 5 specimens and on leg 4 in only 2 specimens. No other variability was
observed.

REMARKS

The new species is provisionally assigned to the genus Microlaophonte Vervoort,
1964 in the subfamily Laophontinae. There are marked differences between the new
species and M. spongicola Vervoort, 1964, the type and only species of the genus, but
these are not sufficient to justify the establishment of a new genus. However,
M. spongicola is known only from the female (Vervoort, 1964) and the eventual
discovery of the male may necessitate the removal of M. trisetosa from the genus if
the male secondary sexual characters are found to differ significantly from those of
M. trisetosa.

The spine and seta formula of legs 2-4 is of diagnostic importance at the generic
level in the Laophontidae. The formulae of M. spongicola and M. trisetosa differ
primarily in the presence of an additional internal seta on the second endopod
segment of legs 2 and 3 in the latter species. However, the absence of the small
innermost seta on this segment in M. spongicola may only be apparent as Vervoort
(1964) states that small setae may have been obscured by the hairs on the endopod.

The two species can be readily separated by the presence in M. trisetosa of 3 setae
on the exopod of the second antenna (the character alluded to in the specific name), of
6 setae on the exopod of leg 5 in the female and by the i-segmented nature of the
exopod of leg I.

ACKNOWLEDGEMENTS

I am indebted to Dr George for providing the material for this study and to Dr
R. J. Lincoln for his critical reading of the manuscript.

REFERENCES

GEORGE, J D. (1976) The mass culture of benthic polychaetes and harpacticoid copepods
on agar. Proceedings of the tenth European Symposium on Marine Biology.

VERVOORT, W. 1964. Free-living Copepoda from Ifaluk Atoll in the Caroline Islands. Bull.
U.S. natn. Mus. no. 236 : 1-431.

DR G. A. BOXSHALL

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON SWy 5BD

A NEW SPECIES OF AMPHITHOE (PLEONEXES)

(AMPHIPODA: AMPHITHOIDAE)

FROM THE NORTH-EAST ATLANTIC

WITH A REDESCRIPTION OF

A. (P.) GAMMAROIDES (BATE)

By ROGER J. LINCOLN

INTRODUCTION

A SHORT review of the subgenus Pleonexes is given in a recent paper by Mateus &
Afonso (1974) together with a table of comparative characters for the six recognized
species. Pleonexes was only recently relegated from the familiar generic status to
that of a subgenus by Barnard (1970) on the evidence of a gradation of characters
between Amphithoe and Pleonexes. The subgenus is identified in the Atlantic by the
marked expansion of the propodus of pereopods 5-7 and the pair of prominent hooked
spines on the distal margin of the telson, although an examination of all Amphithoe
species reveals a gradation of these characters which suggests that Pleonexes may not
be a valid subgenus.

Only three species of Amphithoe (Pleonexes) are recognized from the north-east
Atlantic and Mediterranean, namely A. (P.) bicuspis (Heller) from the Adriatic,
A. (P.) pomboi Mateus & Afonso from the Azores and A. (P.) gammaroides (Bate),
which is the only widespread species recorded along the coast of Europe from southern
Norway to the western Mediterranean and including the British Isles.

After examining the amphipods in the British Museum (Natural History) as part
of the preparation for a new handbook on the amphipod fauna of the British Isles, it
is clear that there are in fact two quite distinct species of Pleonexes in collections from
British localities. The two species differ considerably in body size and especially in
the relative proportions and robust nature of the antennae, as well as other structural
characters. Collections from more northerly areas, especially Shetland Isles and the
west coast of Scotland, consist only of the larger of the two species which is the true
gammaroides, while material from the south coast of England, southern and south-
western Ireland belongs mostly to the new and smaller species which is described
below as Amphithoe (Pleonexes) neglectus n. sp.

The confusion over the identification of the two species can probably be explained
by the fact that Sars (1894), in his excellent monograph on the amphipods of Norway,
figured the female of the true gammaroides (actually a specimen from southern
Norway originally ascribed to Sunamphithoe hamulus by Boeck but now recognized
to be a synonym of gammaroides) , while his figures of the male are not of gammaroides
but the new species, neglectus. Fortunately, Sars offers an explanation of this in his

Bull. Br. Mus. nat. Hist. (Zool.) 30, 6 Issued 25 November 1976

230 ROGER J. LINCOLN

text (page 584) where he states that he had no male specimen of gammaroides from
Norway and had illustrated material collected in France.

Two other species should be referred to the synonymy of gammaroides, namely
Sunamphithoe longicornis Boeck and Sunamphithoe hamulus Bate. An examination
of the type 5. hamulus has shown it to be the female of gammaroides and not of
Sunamphithoe conformata Bate as proposed by Stebbing (1906).

DESCRIPTION OF SPECIES

Amphithoe (Pleonexes) neglectus n. sp.

(Figs la ; 2a-h ; sa-g ; 4a-g)
Pleonexes gammaroides : Sars, 1894 : 5^2 (Part), pi- 207 (cj) ; Stebbing, 1906 : 642 (part).

DIAGNOSIS. Length up to 5 mm ; antennae subequal, moderately setose, equal to
half body length or less. Gnathopod I propodus elongate oval. Gnathopod 2
ischium without distinct lobe on anterior margin ; in male propodus broadly oval,
tumid, palmar surface concave with distinct inner and outer palmar margin,
delimited from posterior margin by rounded angle. Pereopod 5 basis very much
expanded, broader than long. Telson with prominent marginal hooks and pair of
mediolateral setae.

DESCRIPTION. Length 4-5 mm, body rather slender, compressed, urosome
segment I with pair of dorsal setules (Fig. la) ; colour in spirit whitish with scattered
dark chromatophores on body, coxal plates, basal segments of pereopods and
antennal peduncles. Coxal plates 1-5 moderately large, distal margins rounded with
numerous small setules ; coxal plate i produced slightly forwards ; plate 2 with
anterior margin very broadly rounded ; plate 4 entire, not excavate posteriorly ;
plate 5 with small posterior lobe, and large anterior lobe about equal to length of
plate 4. Epimeral plates 1-3 rounded ; plates 2-3 (Fig. 4c) with small setule inset in
posterodistal margin. Head with lateral lobes broadly convex, only moderately
produced ; eyes small and rounded, visual elements distinct. Antennae (Fig. 2a, b)
relatively short and setose, equal or little less than half body length, of subequal
length or with antenna i little longer than 2 ; accessory flagellum absent ; antenna
i peduncle article i robust with 2-3 distoventral, and 1-2 small mid-ventral, spinules ;
article 2 little shorter than i ; article 3 about half length of 2 ; flagellum about
15 to i8-articulate, each article with distinct elongate aesthetasc ; antenna 2 more
robust than i especially in male, peduncle articles 4 and 5 subequal, or article 5
slightly longer than 4 ; flagellum moderately setose, shorter or equal to peduncle
article 5, about Q-articulate with proximal articles often swollen (Fig. 2b). Upper
lip entire, margin rounded, setulose. Mandible with distinct molar ; left mandible
(Fig. 2f) with 5 spines in spine-row, right mandible (Fig. 2g) with 4 spines ; palp
3-articulate, article i short, article 2 with single large distal seta, article 3 with group
of about 9 large terminal setae. Lower lip (Fig. 2c) with inner and outer lobes
distinct, setulose. Maxilla i (Fig. 2d) inner plate very small with single small seta,
outer plate with about 9 robust spines ; palp 2-articulate, article i very short,
article 2 elongate and curved, reaching beyond apex of outer plate with 2-3 small

NEW SPECIES OF AMPHITHOE

231

FIG. i. (a) Amphithoe (Pleonexes) neglectus n. sp. male, entire ; (b) Amphithoe (Pleonexes)

gammaroides (Bate), male, entire.

terminal spines. Maxilla 2 (Fig. 2e) inner plate with several long, inner-marginal,
plumose setae ; inner margin also fringed with fine short setules ; outer plate longer
than inner, outer and distal margins fringed with fine short setules, distal margin with
several long setae. Maxilliped (Fig. 2h) inner plate short ; outer plate elongate oval
with inner margin finely serrate, marginal spines increasing in length distally ; palp
4-articulate, setose. Gnathopod i (Figs 3a, 41) generally similar in male and female ;

232

ROGER J. LINCOLN

a

0 -5mm

FIG. 2. Amphithce (Pleonexes) neglectus n. sp. male ; (a) head and antennae ; (b) head
and antennae, robust ; (c) lower lip ; (d) maxilla i ; (e) maxilla 2 ; (f ) left mandible ;
(g) right mandible ; (h) maxilliped.

NEW SPECIES OF AMPHITHOE

233

FIG. 3. Amphithoe (Pleonexes) neglectus n. sp. male ; (a) gnathopod i ; (b) gnathopod 2
(c) gnathopod 2, propodus ; (d) gnathopod 2, propodus, small male ; (e) pereopod 3
(f ) pereopod 4 ; (g) pereopod 5.

234 ROGER J. LINCOLN

basis robust, narrow proximally with posterior margin rather angular, anterodistal
angle with small rounded lobe ; merus with few distal setae ; carpus much shorter
than propodus, anterior margin convex with large median spine and 1-2 large distal
spines, posterior margin rounded with few long setae ; propodus elongate oval, palm
convex and poorly defined from posterior margin, delimited by single large spine,
distal margin of palm with single curved striated spine close to base of dactylus,
posterior and palmar margins moderately setose ; dactylus little longer than palm,
inner margin toothed. Gnathopod 2 female (Fig. 4g) little larger and more robust than
i ; basis with large anterodistal lobe bearing 1-2 small spinules ; carpus with
posterior lobe rather slender and produced ; propodus broadly oval, anterior margin
with 4-5 small groups of setae, palm convex, oblique, delimited from posterior margin
by obtuse angle and single large spine ; posterior and palmar margins strongly
setose, palmar margin with single curved, striated, spine close to base of dactylus.
Gnathopod 2 male (Fig. 3b) very much larger and more robust than in female ; basis
with very large anterodistal lobe bearing 2-3 small spines ; carpus very short with
slender posterior lobe, anterior margin with strong median and distal spines ;
propodus very broad and robust, anterior margin strongly convex with several small
groups of setae or spines, palm complex with concave surface between distinct inner
and outer palmar margins (Fig. 3c), delimited by obtuse angle and single stout spine ;
in smaller specimens the outer palmar margin may be indistinct (Fig. 3d) ; palm
weakly setose with single striated spine close to base of dactylus. Pereopods 3-4
(Fig. 36, f) basis narrowly oval, anterior and posterior margins weakly setose ;
merus broad distally ; carpus short ; propodus about equal to length of merus,
tapering distally ; dactylus weakly curved and about half length of propodus ;
merus, carpus and propodus with few long setae. Pereopod 5 (Fig. 3g) basis broader
than long with very large posterior lobe, anterior margin with several short spines ;
merus little longer than carpus ; propodus broad distally, palm with 3-4 short, and
i long curved, striated spines ; inner and outer palmar surface with small group of
long setae. Pereopod 6 (Fig. 4a) more elongate than 5 ; basis with moderately large
posterior lobe which narrows distally to produce sinuous posterior margin ; merus,
carpus and propodus similar to pereopod 5 only more elongate. Pereopod 7 (Fig. 4b)
slightly longer than 6 ; basis narrowly oval with only small posterior lobe and
convex posterior margin. Uropods (Fig. 4d) short and spinose. Uropod i peduncle
with 3 distomarginal spines and many long outer-marginal setae, distoventral angle
with short blunt process ; outer ramus with about 3 marginal spines and small group
of terminal spines ; inner ramus equal to length of outer, with only small group of
apical spines. Uropod 2 peduncle with 2 distomarginal spines ; outer ramus with 2
marginal spines ; inner ramus equal to length of outer, with only apical spines.
Uropod 3 (Fig. 4e) peduncle robust with single distal spine ; outer ramus with pair of
large curved spines and finely denticulate dorsal margin ; inner ramus with small
apical spine and about 6 long setae. Telson fleshy, entire, with distinct distolateral
hooks and pair of mediolateral setae.

HOLOTYPE. Portsmouth, <J collected from Sargassum. BM(NH) reg. no.
1975:467:1.

NEW SPECIES OF AMPHITHOE

235

FIG. 4. Amphithoe (Pleonexes) neglectus n. sp. a-e, male ; f-g, female : (a) pereopod 6 ;
(b) pereopod 7 ; (c) epimeral plates 2-3 ; (d) urosome and uropods ; (e) uropod 3,
inner aspect ; (f ) gnathopod i, female ; (g) gnathopod 2, female.

236 ROGER J. LINCOLN

OTHER MATERIAL EXAMINED

Portsmouth : 7^, 3$? collected from Sargassum. BM(NH) reg. no. 1974:365:10.
Loch Ine, Ireland: $<$<$, 30$$, 2ojuv. collected from C odium and Stilophora, all

specimens damaged. BM(NH) reg. no. 1975:464:55.
Loch Ine, Ireland : 5^, 18$$, 10 juv. collected from Codium and Stilophora.

BM(NH) reg. no. 1975:465:33-

Trevone Bay, Cornwall : 3$$ from algae at low-water. BM(NH) reg. no. 1974:366:3.
Portnafranca, Co. Mayo, Ireland : i$ BM (NH) reg. no. 1971:63:1.
Colieragh Bay, Bantry Bay, Ireland: i<$, 3$$ collected from low water. BM(NH)

reg. no. 1975:466:4.

REMARKS. The propodus of the second gnathopod shows a considerable range of
development in the male. In small specimens the palm is almost straight with only
little evidence of the concave surface and outer palmar margin (Fig. 3d) . However,
in large males the propodus is very robust and swollen with a broadly concave palmar
surface and distinct inner and outer margin. A small amount of variation is also
evident in the second antenna which may be slender in females and some males,
while in others the peduncle is quite robust and the basal articles of the flagellum are
distinctly swollen (Fig. 2b).

The new species, neglectus, is most closely allied to the Mediterranean bicuspis
figured by Giordani Soika (1950) which also has the broad palm on the male second
gnathopod. However, the two species can be separated by the detailed structure of
the male and female gnathopods, the setation of the pereopods and the arrangement
of spines on the propodal articles, as well as other differences in relative proportions
and setation.

Amphithoe (Pleonexes) gammaroides (Bate)
(Figs ib ; 5a-g ; 6a-e ; 7a-g)

Pleonexes gammaroides Bate, 1857 : 147 ; Stebbing, 1906 : 642 (part) ; Sars, 1894 : 582 (part)

pi. 207 (?) ; Chevreux & Fage, 1925 : 335, fig. 344.

Amphithoe gammaroides : Bate, 1862 : 235, pi. 41, fig. 4 ; Bate & Westwood, 1863 : 427.
Sunamphithoe hamulus Bate, 1857 : 148 ; 1862 : 250, pi. 43, fig. 5 ; Bate & Westwood, 1863 :

430 ; Boeck, 1872 : 594, pi. 27, fig. i.
Sunamphithoe longicornis Boeck, 1870 : 165 ; 1872 : 596, pi. 27, fig. 2.

DIAGNOSIS. Length up to about 8 mm ; antenna 2 longer than i, sparsely setose
especially in male, robust and much more than half body length. Gnathopod i
propodus rather rectangular, palm oblique. Gnathopod 2 ischium with large
asymmetrical lobe on anterior margin ; in male propodus very broad, palm straight
and finely toothed, delimited by distinct angle from posterior margin, distal part
of anterior margin expanded towards base of dactylus. Pereopod 5 basis only
moderately expanded, longer than broad. Telson with pronounced marginal hooks
and numerous dorsal setules.

DESCRIPTION. Length 6-8 mm ; urosome segment i with pair of dorsal setae ;
colour bright green with scattered dark chromatophores on sides of body. Coxal
plates moderately large, margins rounded with numerous short setules ; coxal plate i

NEW SPECIES OF AMPHITHOE

23?

FIG. 5. Amphithoe (Pleonexes) gammaroides (Bate), male : (a) head and antennae, large
male ; (b) lower lip ; (c) maxilla i ; (d) maxilla 2 ; (e) right mandible ; (f ) left man-
dible ; (g) maxilliped.

238 ROGER J. LINCOLN

produced slightly forwards ; plate 2 very broadly rounded ; plate 5 with anterior
lobe equal to length of plate 4. Epimeral plates 1-3 (Fig. 70) rounded. Head with
lateral lobes convex, weakly produced ; eyes small and rounded, visual elements
distinct. Antennae relatively long and robust, sparsely setose (Fig. 5a) ; antenna i
equal to, or often little more than, half body length, peduncle article i robust with
3-4 distoventral and 1-2 mid-ventral spinules, article 2 equal to length of i ;
flagellum up to about i8-articulate, each article with slender aesthetasc, setae very
sparse and small ; antenna 2 more robust and much longer than i, up to about
two-thirds body length, peduncle article 5 longer than 4, flagellum about 9 to
j-3-articulate, also with only few small setae, flagellar articles often very robust in
large specimens. Upper lip entire, setulose. Mandible with well developed molar,
palp 3-articulate, article 2 with single long distal seta, article 3 with group of long
apical setae ; left mandible (Fig. 51) with 5 spines in spine row, right mandible
(Fig. 5e) with 4 spines. Lower lip (Fig. 5b) with inner and outer lobes distinct,
setulose. Maxilla i (Fig. 5c) inner plate small with single seta ; outer plate with
about 9 robust spines ; palp 2-articulate, article 2 elongate, curved, with 3-4 small
apical spines. Maxilla 2 (Fig. 5d) inner plate with several long plumose setae, inner
margin fringed with fine setules ; outer plate longer than inner, outer and distal
margins fringed with fine short setules, distal margin with several long setae.
Maxilliped (Fig. 5g) inner plate short, setose ; outer plate elongate, oval, with inner
margin finely serrate, marginal spines increasing in length distally ; palp 4-articulate,
setose. Gnathopod i (Figs 6a, 71) generally similar in male and female ; basis robust,
narrow proximally, anterodistal angle with rounded lobe bearing 1-2 small spinules ;
carpus much shorter than propodus, anterior margin with median spine and 1-2 large
distal spines, posterior margin rounded, setose ; propodus elongate, palm convex
and delimited by single large spine, posterior margin about straight, setose ; palmar
margin setose with single curved striated spine close to base of dactylus. Gnathopod
2 female larger and more robust than i (Fig. 7g) ; basis with very large anterodistal
lobe bearing 2-3 small spinules ; ischium with large asymmetrical lobe on anterior
margin ; carpus with posterior lobe rather slender and produced ; propodus broadly
oval, palm convex, oblique, delimited by obtuse angle and single large spine ;
posterior and palmar margins strongly setose, palm with single striated spine close to
base of dactylus. Gnathopod 2 male (Fig. 6b) very much larger and more robust than
in female ; basis with large anterodistal lobe ; ischium with large asymmetrical lobe
on anterior margin ; carpus very short, anterior margin with about 5 strong spines,
posterior lobe slender and elongate ; propodus very broad, posterior margin about
straight, palm long, oblique, straight with margin finely toothed, palm delimited by
distinct angle and single stout spine ; anterior margin of propodus expanded distally
to produce a rounded lobe at base of dactylus. Pereopods 3-4 (Fig. 6c, d) basis
narrowly oval ; merus broad distally with anterior angle somewhat produced ;
carpus short ; propodus about equal to length of merus, tapering distally ; dactylus
only weakly curved and about half length of propodus. Pereopod 5 (Fig. 6e) basis
broadly expanded but still longer than wide ; merus longer than carpus ; propodus
broad distally, palm with 3-4 short spines and i long curved spine, inner and outer
palmar surface with small group of long setae. Pereopod 6 (Fig. 7a) longer than 5 ;

NEW SPECIES OF AMPHITHOE

239

FIG. 6. Amphithoe (Pleonexes) gammaroides (Bate), male : (a) gnathopod i ; (b) gnatho-
pod 2 ; (c) pereopod 3 ; (d) pereopod 4 ; (e) pereopod 5.

240

ROGER J. LINCOLN

FIG. 7. Amphithoe (Pleonexes) gammaroides (Bate), a-e, male ; f-g, female : (a) pereopod
6 ; (b) pereopod 7 ; (c) epimeral plates 2-3 ; (d) urosome and uropods ; (e) uropod 3,
inner aspect ; (f) gnathopod i, female ; (g) gnathopod 2, female.

NEW SPECIES OF AMPHITHOE 241

basis with moderately large posterior lobe which narrows distally ; merus, carpus
and propodus similar to pereopod 5 only more elongate. Pereopod 7 (Fig. yb) little
longer than 6 but generally similar ; basis only narrowly oval with weakly convex
posterior margin. Uropods (Fig. 7d) moderately elongate and spinose. Uropod i
peduncle with 4 distomarginal spines and numerous long outer-marginal setae,
distoventral angle with large blunt process ; outer ramus with about 4 marginal
spines and small group of apical spines ; inner ramus equal to length of outer with
only small group of apical spines. Uropod 2 peduncle with 3 marginal spines and
small ventrodistal process ; outer ramus with 3 marginal spines ; inner ramus equal
to length of outer with only small group of apical spines (occasionally small marginal
spine present). Uropod 3 (Fig. 7e) peduncle elongate with single distal spine ; outer
ramus with pair of large curved spines and finely denticulate dorsal margin ; inner
ramus with small apical spine and about 5 long setae. Telson fleshy entire, with
distinct distolateral hooks and several short dorsal setae.

DISTRIBUTION. This species is quite widely recorded in the north-east Atlantic
area, although probably confused with neglectus over part of its range : coast
of Europe from northern Norway (Vader, 1969, 1971), British Isles, to western
Mediterranean, Azores and Canary Isles (Chevreux & Fage, 1925) . It is possible that
Chevreux & Fage confused gammaroides and bicuspis, and that gammaroides does not
occur in the Mediterranean. The figured material was collected from the Shetland Isles.

REFERENCES

BARNARD, J. L. 1970. Sublittoral Gammaridea (Amphipoda) of the Hawaiian Islands.

Smithson. Contr. Zool. 34 : 1-286.
BATE, C. SPENCE 1857. A synopsis of the British Edriophthalmous Crustacea, Part i :

Amphipoda. Ann. Mag. nat. Hist. 19 : 135-152.

1862. Catalogue of the specimens of Amphipodous Crustacea in the collection of the British

Museum, pp. 1-399. London.

& WESTWOOD, J. O. 1863. A history of British sessile-eyed Crustacea. Vol. i, pp. 1-507.

London.

BOECK, A. 1870. Crustacea Amphipoda borealia et arctica. pp. 83-280. Christiania.

1872. De Skandinaviske og Arktiske Amphipoder. pp. 1-711. Christiania.

CHEVREUX, E. & FAGE, L. 1925. Faune de France 9. Amphipodes. pp. 1-488. Paris.
GIORDANI SOIKA, A. 1950. Gli Anfipodi Gammarini della Laguna di Venezia. Archo

Oceanogr. Limnol. 6 (2-3) : 165-212.
MATEUS, E. & AFONSO, O. 1974. Etude d'une collection d'Amphipoda des A$ores avec la

description d'une nouvelle espece. Anais Fac. Cienc. Porto 57 : 9-39.
SARS, G. O. 1894. An account of the Crustacea of Norway. Vol. i (24) pp. 517-711.

Christiania & Kjobenhavn.

STEBBING, T. R. R. 1906. Amphipoda i : Gammaridea. DasTierreich21. pp. I -806. Berlin'.
VADER, W. 1969. Notes on a collection of Amphipoda from the Trondheimsfjord area.

K. norske Vidensk. Selsk. Skr. 3 : 3-20.

1971- Additions to the Amphipoda of northern Norway. Astarte 4 (2) : 47-51.

DR R. J. LINCOLN

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON SW7 5BD

THE FIRST ZOEAL STAGE OF

SCOPIMERA CRABRICAUDA ALCOCK

(CRUSTACEA, BRACHYURA, OCYPODIDAE).

By A. L. RICE

ABSTRACT

The first zoeal stage of the ocypodid crab Scopimera crabricauda is described and compared
with previously described larvae of the same family.

INTRODUCTION

RICE (1975) described the first zoea of M acrophthalmus depressus Riippell and
reviewed the available information on larval development within the brachyuran
family Ocypodidae, pointing out that there seem to be clear differences between the
zoeae of the three recognized sub-families, particularly in the setation of the mouth-
parts. However, knowledge of one of these sub-families, the Scopimerinae, is based
on published descriptions of the larvae of only five species, belonging to three genera,
and most of these accounts are very inadequate. Within the genus Scopimera only
the first stage of 5. globosus has been described (Aikawa, 1929) so that when first
zoeae were obtained from female S. crabricauda collected in Bahrain, Arabian Gulf, a
detailed description seemed desirable, even though all the animals died without
moulting.

MATERIAL

Two ovigerous female S. crabricauda were collected from the foreshore at Jufair,
Bahrain, in April 1975 and were maintained in aquaria in the laboratory of the
adjacent Fisheries Resources Bureau at a salinity of 42%0 and a temperature of 19 °C
until the larvae hatched some ten days later. Although the zoeae were supplied
with freshly hatched Artemia nauplii, none of them moulted into the second stage.

The adults were fairly confidently identified as 5. crabricauda since they agree quite
well with Alcock's (1900) original description and Kemp's (1919) additional notes
based on material from Karachi. Stephensen (1945) identified specimens from
Bushire on the coast of Iran as S. crabricauda, but pointed out that the sculpturing
on the dorsal surface of the carapace was much less pronounced than in the specimens
from Pakistan. This sculpturing is similarly unclear in the Bahrain material and, in
addition, the blunt ridge on the inner face of the chela seems to be less well developed
than Kemp's description suggests. However, agreement in other characters is good,
and the identity of the Bahrain animals with Alcock's species is supported by
the presence in the same area of males which are apparently conspecific with the

Bull. Br. Mus. nat. Hist. (Zool.) 30, 6 Issued 25 November 1976

244 A. L. RICE

ovigerous females and which possess the racket-shaped abdomen so characteristic of
5. crabricauda.

The adult female crabs, and the larvae obtained from them, are deposited in the
collections of the British Museum (Natural History) under reg. nos 1976:71, 1976:72,
1976:73, 1976:74.

DESCRIPTION OF THE LARVAE

Dimensions. Carapace length (base of rostrum between the eyes to posterio-lateral
carapace margin) 0-46-0-50 mm. Tip of dorsal to tip of rostral carapace spines
2-46-2-65 mm. Carapace width (tip to tip of lateral spines) 0-44-0-53 mm.

Carapace (Fig. la, b). Long dorsal and rostral carapace spines ; dorsal spine more
than twice carapace length, rostral spine almost three times carapace length.
Lateral carapace spines short and straight. Posterio-lateral carapace margins
naked, but dorsal carapace surface with a pair of small setae anterior to the base of
the dorsal spine.

Eyes. Fused.

Antennule. Unsegmented, with 2 terminal aesthetascs and one or two setae.

Antenna (Fig. la, b, d). Spinous process about | length of rostrum, with a total of
about 15 spinules becoming longer distally. No endopod or exopod.

Maxillule (Fig. le). Endopod 2-segmented ; proximal segment naked, distal with

4 terminal setae.

Maxilla (Fig. if). Endopod bi-lobed, with 3 and 2 setae respectively ; basal
endite bi-lobed with 3 and 4 setae ; coxal endite weakly bi-lobed with a total of

5 setae. Scaphognathite with 4 marginal setae and a long, setose posterior process.
Maxilliped I (Fig. ig). Basis with 10 medial setae arranged 2, 2, 3, 3 ; endopod of

5 segments with 2, 2, i, 2 and 4 + 1 setae ; third segment with 2-6 very fine hairs on
the lateral margin. Exopod with 4 natatory setae.

Maxilliped 2 (Fig. ih). Basis with 3 medial setae ; endopod of 3 segments with
o, i and 6 setae respectively. Exopod with 4 natatory setae.

Abdomen (Fig. la, c). Somites 2 and 3 with small dorso-lateral process ; somites
3 and 4 with short posterio-lateral spines. Dorsal surfaces of somites 2-5 and of the
telson with many very small spinules arranged in a more or less symmetrical pattern.
Telson long and narrow ; forks long and divergent, each with a pronounced dorsal
spine. Posterior margin with a deep median cleft and with 3 pairs of setae less than
half the length of the forks.

DISCUSSION

The characterization on the basis of mouthpart setation of the larvae of the
Scopimerinae as opposed to the other ocypodid sub-families, suggested by Rice
(1975), is supported without change by this account of the first zoea of 5. crabricauda*
This short discussion will therefore be restricted to the Scopimerinae.

Although the larvae of 5. crabricauda resemble very closely Aikawa's (1929)
description of the zoea of 5. globosus, there are nevertheless clear differences between

* The setation of the endopod of the maxillule in S. globosus is o, 4, as in 5. crabricauda, and not o, 5
as given by Rice (1975) while a dorsal spine is present on each telson fork in both species.

FIRST ZOEA OF SCOPIMERA CRABRICAUDA

245

FIG. i. Scopimera crabricauda, stage I zoea : (a) lateral view ; (b) frontal view ; (c) abdo-
men ; (d) antenna ; (e) maxillule ; (f ) maxilla ; (g) first maxilliped ; (h) second maxil-
liped. The bar scale represents 0-5 mm for a and b, and o-i mm for c-h.

the two species. First, the dorsal and rostral carapace spines are much longer in
5. crabricauda ( > twice carapace length) than in S. globosus ( < twice carapace length).
Second, these spines are completely naked in 5. crabricauda whereas Aikawa describes
them as 'sparsely toothed' in 5. globosus. Finally, according to Aikawa's fig. 52,
abdominal somites 3 and 4 in 5. globosus lack the small but clear posterio-lateral
spines present in 5. crabricauda.

The available descriptions of the larval stages of other members of the Scopimerinae
(see Rice, 1975) are not sufficiently detailed to enable generic characters within the
sub-family to be discussed. Nevertheless, it seems that the zoeae of the three genera
Scopimera, Ilyoplax and Dotilla may be distinguished between by the lengths of their
dorsal and rostral carapace spines relative to the carapace length, these spines being

246 A. L. RICE

longest in Scopimera, shorter in Ilyoplax, and shortest in Dotilla. This variation in
carapace spine length is reflected in a difference in overall size, the first zoeae of
S. crabricauda and S. globosus being more than 2 mm from the tip of the dorsal spine
to the tip of the rostral spine, whereas in the described first stages of Ilyoplax and
Dotilla species this dimension is little more than i mm. Finally, the wing-like
expansions of the fifth abdominal somite in the zoeae of Dotilla species (see Ramadan,
1940 ; Gohar & Al-kholy, 1957 ; Rajabai, 1959) clearly distinguish them from the
known larvae of both Scopimera and Ilyoplax.

REFERENCES

AIKAWA, H. 1929. On the larval forms of some Brachyura. Rec. oceanogr. Wks Jap. 9 : 87-

162.
ALCOCK, A. 1900. Materials for a carcinological fauna of India. No. 6. Brachyura, Cato-

metopa or Grapsoidea. /. Asiat. Soc. Bengal 69 : 279-456.
GOHAR, H. A. F. & AL-KHOLY, A. A. 1957. The larvae of some brachyuran Crustacea.

Publs mar. biol. Stn. Ghardaqa 9 : 146-176.
KEMP, S. W. 1919. Notes on Crustacea Decapoda in the Indian Museum. XII. Scopi-

merinae. Rec. Indian Mus. 16 : 305-348.
RAJABAI, K. G. 1959. Studies on the larval development of the Brachyura. I. The early

and post larval development of Dotilla blanfordi Alcock. Ann. Mag. nat. Hist. ser. 13,

2 : 129-135.
RAMADAN, M. M. 1940. On the first zoeal stage of Dotilla sulcata (ForskSl). Ann. Mag. nat.

Hist. ser. n, 5 : 253-255.
RICE, A. L. 1975. The first zoeal stages of Cancer pagurus L., Pinnotheres pisum (Pennant)

and Macrophthalmus depressus Riippell (Crustacea, Decapoda, Brachyura). Bull. Br.

Mus. nat. Hist. (Zool.) 28 : 237-247.
STEPHENSEN, K. 1945. The Brachyura of the Iranian Gulf. Dan. Scient. Invest. Iran 4 : 57-

257-

DR A. L. RICE

INSTITUTE OF OCEANOGRAPHIC SCIENCES

WORMLEY

GODALMING

SURREY

ASTEROZOA FROM

AMSTERDAM AND ST PAUL ISLANDS,
SOUTHERN INDIAN OCEAN

By AILSA M. CLARK

SYNOPSIS

Asteroids and ophiuroids collected recently around Amsterdam and St Paul Islands by
French scientific collectors in association with lobster fishing vessels are reported. One new
species of asteroid, a goniasterid named Sphaeriodiscus mirabilis, is described and the systematic
position of another asteroid, named Culcita veneris by Perrier (1879) but now confirmed as
better placed in Spoladaster, is discussed at some length with comments on related Poraniidae.
Besides a third asteroid and one ophiuroid species previously recorded, four other ophiuroids
new to these islands are also found to occur there.

INTRODUCTION

THE interesting collection of Asterozoa which forms the basis for this report was sent
to the British Museum by Dr P. M. Arnaud of the Station Marine d'Endoume in
Marseille. The material was primarily collected by Dr J. Beurois in 1971-72 but a
few additional specimens originated from P. M. Arnaud, P. Noel, R. Vranckx,
M. Segonzac and the fishing vessel Maria Martina.

Financial support for the collectors came from Terres Australes et Antarctique
Frangaises and logistic support from S.A.P.M.E.R. (Societe Anonyme de Peche
Maritime et de Ravitaillement), to which institutions thanks are also due.

Hitherto the only published records of Asterozoa from these remote islands in the
Indian Ocean (c. 38°S, 77^°E) were based on material from two sources. Firstly the
French Transit of Venus Expedition, 1874-75 (naturalist C. Velain), from which two
^asteroids, named as Asterina exigua (Lamarck) and Culcita veneris sp. nov., were
reported by Perrier (1879). Secondly the German South Polar (Gauss) Expedition
of 1901-3, which evidently took only the first asteroid, recorded as Patiriella exigua
by Doderlein (1927), and one ophiuroid, recorded as Amphipholis minor by Hertz
(1927). Dartnall (1971) confirmed that the Gauss asteroids are conspecific with
Patiriella exigua (Lamarck) from South Africa and temperate Australia, when he
split off tropical records of P. exigua as a distinct species. In contrast, Perrier's
second asteroid and the new species now described appear to be endemic to these
isolated islands.

Thanks to offshore work by lobster fishing vessels on the shelf around these islands
to a depth of about 300 m and to painstaking collecting from washings of encrusting
organisms, one fine new species of goniasterid starfish and four additional species of
ophiuroids are included in the present material. Unfortunately most of the
ophiuroids are very small and not in good condition and one amphiurid is not

Bull. Br. Mus. nat. Hist. (Zool.) 30, 6 Issued 25 November 1976

15

248 A. M. CLARK

identifiable to species. However, the other three indicate an affinity with the
ophiuroid fauna of south and south-east Africa in similar latitudes up-current in the
West Wind Drift. This accords with the distribution of some other benthic
invertebrates.

DESCRIPTION OF SPECIES

ASTEROIDEA

Sphaeriodiscus mirabilis sp. nov.
(Pis i and 2)

MATERIAL. Vicinity of St Paul and Amsterdam Islands at 200-300 metres,
fishing vessel Maria Martina, 1971, the holotype, B.M. reg. no. 1976.1.12.1.

DESCRIPTION. R/r 70/45 mm = 1-56/1. The breadth of the swollen arm tip is
c. 26 mm.

The area of abactinal plates is almost pentagonal. After preservation in spirit it is
sunken below the raised rim of the superomarginal plates. Distally the abactinal
plates are hexagonal and fairly regular in arrangement in longitudinal and diagonal
series but interstitial plates are found proximally to about two-thirds R from the
centre and as these become larger and more numerous towards the centre the regular
arrangement is lost. A mid-radial row of plates is j ust discernible, though proximally
its constituent plates are all separated. Conversely the plates of the interradial rows
are nearly all contiguous.

All the plates are flat-topped and covered with a pavement of close polygonal
flat-topped granules, numbering about 70 on the larger plates. Papulae are wide-
spread except close to the superomarginals and in interradial bands.

The madreporite is triangular with two angles truncated to form an irregular
pentagon. It lies at about one-third r from the centre.

The superomarginal plates number only four in each series (i.e. eight on each of the
five sides) with the exception of one series where a short extra plate occurs between
the second and third. In some of the other series at this same position a triangular
abactinal plate is partially inserted between these two marginals, possibly reinforcing
this vulnerable area. The two inner superomarginals are elongated tangentially,
rectangular in outline and slightly raised at the joints, though in general with a
part-cylindrical section. The third plate is abruptly enlarged and swollen, extended
radially to meet its fellow ; together with the two smaller fourth superomarginals of
each ray these make the upper side of a discrete broad rounded knob. The sutures
between the third and fourth plates are oblique to the midradial line and the fourth
plates are unequal in size so that the sutures are asymmetrical in all five cases.

The inferomarginals match the upper series in general but individual plates may be
slightly misaligned ; also there is an additional fifth plate in each series, bordering
the small spout-like terminal plate. The third to fifth plates participate in the
knob-like expansion of the tip. The more convex areas of the swollen marginals
are mostly naked with only scattered granules over most of their surface and these
sunk into sockets in the plates ; only peripherally is the granulation continuous, as it

ASTEROZOA FROM THE SOUTHERN INDIAN OCEAN 249

is on the first two (interradial) plates. The suture between the two marginal series is
distinctly above the ambitus so that the inferomarginals are clearly visible from
above but this may be due to contraction of the upper side during preservation.

The actinal plates in the distal halves of the rays form regular longitudinal series
but proximally the arrangement becomes irregular because of the development and
enlargement of interstitial plates. All the plates are fully covered with polygonal
flat-topped granules distinctly coarser than the abactinal ones with the peripheral
row discrete, even to the naked eye.

The adambulacral plates mostly bear five short blunt-tipped furrow spines, aligned
vertically, sometimes four or (distally) six. These stand out slightly from the lower
surface but are backed by several series of shorter 'spines' transitional to granules, all
angular in section.

No pedicellariae can be distinguished.

AFFINITIES. This handsome goniasterid with its limited number of marginal
plates has a considerable resemblance to those individuals of Pentagonaster pulchellus
Gray, from temperate Australasia, with swollen arm tips (such as the holotype shown
in Gray's pi. 8, fig. 3, 1866). Pentagonaster differs however in having the abactinal
plates with only peripheral granules and their centres more or less markedly convex ;
also the individual marginal plates are rounded at the angles, not rectangular. There
is also a remarkable resemblance to the Jurassic Miopentagonaster calloviensis Mercier
in the very broad rounded arm tips, though again in that fossil species it is only the
distalmost marginal of each series (together with the enlarged terminal plate in this
case) that forms the expansion. The interradial marginals of M. calloviensis number
only four on each of the five sides and appear rectangular in dorsal view, as in S.
mirabilis.

However, judging from the great variation in the breadth of the arm tips in various
goniasterid species such as Pentagonaster pulchellus and Tosia australis I think it
likely that other specimens from the Amsterdam/St Paul area may prove to have
more pointed arm tips. This is the case in the holotype of Sphaeriodiscus scotocryptus
Fisher, 1913, from the Philippine Islands. It too has an almost complete granule
covering, except on the centres of the marginal plates, a basically pentagonal form
but for projecting tips to the rays and a limited number of marginal plates - in this
case at R 32 mm five superomarginals with the three rectangular proximal ones of
each series forming the sides of the body and the two distal ones contiguous mid-
radially with their corresponding plates and forming the arm tip. Apart from these
narrow arm tips and the additional marginal plate in each series at a smaller total
size, 5. scotocryptus differs from the new species in having seven to ten furrow spines
on each plate, besides numerous pedicellariae, though this last may not be a specific
character.

Of the other species of Sphaeriodiscus, the type species, Sphaeriodiscus bourgeti
(Perrier, 1885) from off north-west Africa, also has four superomarginals but the first
one is relatively short (tangentially) and the second is distinctly more swollen though
not meeting its fellow midradially, the third plate is still convex but smaller than the
second so that the outline is approximately pentagonal. The largest of Perrier's
specimens had R 35 mm. H. L. Clark (1926) has recorded a specimen with R 27 mm

250 A. M. CLARK

from off Natal in 420 m as S. bourgeti, expressing surprise about the extension of
range to the south-east side of South Africa. The fourth superomarginals are only
just beginning to appear and the arm tips are somewhat swollen, vertically if not
horizontally, at least, judging from the measurements. He compared it with a
syntype and with Perrier's figures and found no significant difference, so the Natal
specimen presumably also has the second superomarginal distinctly larger and more
swollen than the first, a condition also found by Madsen (1958) in a specimen
of 5. bourgeti from the Cape Verde area and difficult to reconcile with the close
similarity of the first two marginals in 5. mirabilis. Madsen compared 5. bourgeti
with a small specimen of 5. placenta (Miiller & Troschel, 1842), known from wide-
spread localities in the Atlantic and Mediterranean, the main difference in the latter
being the similarity of the consecutive superomarginals of which the fourth is already
well developed. 5. placenta is known to reach a size of nearly 90 mm R and at the
same time to increase the number of marginals in each series to a maximum of nine
superomarginals, though more often five to seven ; two to five of the distal ones may
be contiguous midradially and the arm tip may be broadened in some specimens as
shown in Tortonese's fig. 72 (1965) of a Mediterranean example.

In the Pacific, Sphaeriodiscus inaequalis (Gray, 1847), supposedly from Amboina
or New Guinea, has an even more diminutive first marginal in both upper and lower
series than 5. bourgeti in comparison with the second and third plates, which are
markedly elevated with more or less tabulate raised areas, also on the corresponding
inferomarginals, though at the same time the arms taper to a blunt point since the
distal superomarginals are hardly at all expanded horizontally. At R only 21 mm,
as in the holotype and only recorded specimen, there are only three or four supero-
marginals in each series. Owing probably to abrasion of the surface, this specimen
does not show the characteristic 'peppered' appearance of the barer parts of the
marginals of Sphaeriodiscus due to the recessed isolated granules, but this appearance
is very conspicuous in Fisher's photograph (1906) of the holotype of Sphaeriodiscus
ammophilus, from the Hawaiian Islands. At R only 24 mm this has but three
superomarginals and the third is relatively small yet, though the first is hardly at all
smaller than the second. The form is pentagonal.

LOCALITY. Unfortunately a small element of doubt exists concerning the type
locality of S. mirabilis. Although the specimen came to Dr Arnaud together with
invertebrates characteristic of the shelf in the vicinity of Amsterdam and St Paul
Islands at 200-300 metres, the Maria Martina had also worked off South Africa and
Reunion Islands. If it should prove that specimens of Sphaeriodiscus from Natal
are conspecific with S. mirabilis rather than with S. bourgeti and no further material
from these two remote islands is forthcoming, then the species may have to be
deleted from their fauna.

Patiriella exigua (Lamarck, 1816)

Psterias exigua Lamarck, 1816 : 554.

Asterina exigua : Perrier, 1879 : 47.

Aatiriella exigua : Doderlein, 1927 : 296 ; Dartnall, 1971 : 40-43, pi. 4, fig. c.

ASTEROZOA FROM THE SOUTHERN INDIAN OCEAN 251

MATERIAL. Amsterdam Island : mostly from the north side in the vicinity of
La Roche Godon, under or on mid- or sub-littoral stones or in pools at low water,
60 specimens.

St Paul Island : from the crater, mostly the north-east part near the entrance, on
or under sub-littoral or sometimes mid-littoral stones or from scrapings, usually with
white or green algae, ascidians or sponges and often with Amphipholis squamata and
Amphiura capensis, in specimens.

NOTES. Of the total of 171 specimens, two are four-rayed and two are six-rayed.
Six out of 48 specimens from Beurois' station St Paul 8a (from the north-east part of
the crater among lower mid-littoral and sub-littoral stones) appear more or less
deformed, one having abnormally short rays and the interradii convex, making an
ovate outline ; the other five have one or sometimes two rays aborted or possibly
regenerated, their other four (or three) rays aligned at right angles to each other.
The smallest specimens have R c. 0-5 mm and three pairs of tube feet in each ray.

Spoladaster veneris (Perrier)
(Fig. i ; PL 3, figs i, 2 ; PI. 4, figs i, 2 ; PL 5, figs i, 2 ; PL 6)

Culcita veneris Perrier, 1879 : 47-49, pi. 4 ; Mortensen, 1933 : 249, 250 ; Arnaud & Beurois,

1972 : 874 [Non C. veneris : Bell, 1905 : 248.]
Spoladaster ? veneris : Fisher, 1940 : 136 ; A. M. Clark, 1952 : 208.

MATERIAL. Amsterdam Island : off east coast at 'BMG', in lobster pot, depth
unknown, possibly 10-50 metres, with gastropods Ranella olearium and Charonia
lampas pustulata, i specimen ; no details, i specimen ; from washings of the large
bryozoan Turbicollopora redoutei, sponges, gorgonians, the scleractinian coral
Caryophyllia prof undo. , alcyonarians, etc. at 80-100 metres, i juvenile.

St Paul Island : off east coast in lobster pots, c. 300 metres, 3 specimens ; north-
west coast at La Chaussee du Phoques, 10-50 metres ?, 2 specimens ; near Roche
Nord, in lobster pot, i specimen ; various parts of the crater (the type locality) at
0-5-1 -0 metre and in sublittoral scrapings, sometimes with the gastropods Ranella
and Charonia and with echinoids, 5 specimens.

DESCRIPTION. A photograph of a live specimen (PL 6, fig. i) shows a distinctly
stellate form, R/r c. i-8/i, whereas the preserved specimens all appear more nearly
pentagonal with R/r i -2-1 -5/1. There must therefore be considerable contraction
across the interradii marginally in preservation, possibly correlated with bending
upwards of the arm tips, the resultant form having a close superficial resemblance to
a Culcita. The collectors of Perrier's type material noted a considerable ability of the
live animals to vary the shape of their outlines and degree of inflation of the upper
side, which was also noted by the recent collectors. When inflated the whole
appearance is said to be very like an orange, with which the bright colour of the upper
side also agrees ; the lower side is paler. In the live specimen photographed the
madreporite shows up clearly about midway from the centre to the indented inter-
radial edge but in preserved specimens it appears at a quarter to a third r from the
centre. The enlargement of an arm in life (PL 6, fig. 2) shows a very 'shaggy' effect

252 A. M. CLARK

produced by the extended papulae, which appear to arise all over the upper side
except for narrow bands, notably in the interradii and in the centre, with only the
slightest indication of grouping as in the meshes of a reticulum. In specimens
preserved in alcohol the papulae are more or less retracted. A rough texture can be
felt through the pustular skin, produced by numerous fine spinelets, which are
invisible unless the specimen is partially or completely dried. These are isolated in
the body wall, not based on skeletal plates. Indeed, X-radiography of an adult
shows that the only abactinal plates of any density remaining are a few at the tips of
the rays. The spinelets have the shape shown in Fig. i(a). On the upper surface
the spinelets are fairly evenly scattered but on the sides of the body they tend to
become aligned in vertical series. At the junction of abactinal and actinal areas
below the ambitus, where the papulae cease abruptly, this linear arrangement is even
more marked and the pustular skin covering forms folds with narrow grooves between
leading to the adambulacral furrows. Some of these series branch or anastomose.

In wet specimens there is usually a distinct change in the contours where the
actinal and abactinal areas meet, with a slight horizontal ridge where one would
expect to find the inferomarginal series of plates. Drying tends to emphasize this
and may bring out a second apparently reinforced band just above. However, there
is no trace of a separate intermarginal band of papulae, such as occurs in Porania.
X-radiographs show no marginal plates at all (PI. 5, figs i, 2), nor do drying, dissolv-
ing the body wall or xylol treatment reveal any vestiges of marginal plates, even
viewing from the inside of the body wall (PI. 3, fig. i), except at the tips of the rays.
The external change in contours must therefore be due to reinforcing bands of
connective or muscle tissue.

The actinal spinelets are similar in form to the abactinal ones and are similarly
isolated in the body wall. The X-radiographs taken fail to show up any actinal
plates but special treatment or viewing from inside reveals that in each V-shaped
interradial area adjoining the oral plates and the proximal adambulacrals is a small
triangular patch of hollow, extremely thin-walled, rounded actinal plates (Fig. ic).
More distally there are traces of two, possibly three, partial series of plates running
parallel to the inferomarginal series but these are not complete interradially in any
of the areas treated and usually only the youngest plates adjoining the adambulacral
plates are at all distinct. Presumably these actinal plates are progressively reduced
during growth. None of these vestigial plates bear any enlarged spinelets or spines.

The adradial ends of the actinal pustular folds envelop the subambulacral spines as
well ; these number two on one or two proximal plates but one on the rest. There
are two separately sheathed spines bordering the furrow on most plates. The oral
plates usually have four furrow spines and one larger subambulacral spine, though
larger specimens, R 60 + mm, may have two.

The juvenile specimen (Fig. ig, h) has R/r 1-5/1-3 mm (excluding the length of the
marginal fringe of spinelets). The margin is formed entirely by the relatively large
inferomarginal plates, four on each of the five sides, and the terminal plates. The
adjacent plates dorsally (presumably superomarginals) are markedly convex at their
outer edges. There are two midradial plates on each ray and a ring of five larger
primary interradial plates around a small central area including the anus. All these

ASTEROZOA FROM THE SOUTHERN INDIAN OCEAN

253

FIG. i. Spoladaster veneris (Perrier). (a-f) St. AMS-IO4- R c. 55 mm. (a) Abactinal
spinelet. (b) Jaw and adjacent four adambulacral plates, denuded except for major
spines, (c) Proximal apex of one ventral interradius partly dried and cleared with
xylene to show pustular folds (somewhat shrunken) with isolated spinelets and under-
lying reduced actinal plates, (d) Odontophore, viewed ventrally. (e, f) An isolated
proximal adambulacral plate from the left side of a ray in aboral and adoral views, the
furrow spines arose opposite the letter 'f (g, h) St. AMS-g6. Juvenile partly dried
(some abactinal spinelets probably lost). (The scale equals c. I mm for a, c. 4 mm for
b and c, c. 3 mm for d-f and 2 mm for g and h.)

plates bear spaced rugose spinelets, mostly about three times as long as broad and
slightly waisted. On the inferomarginals the spinelets graduate to the horizontal
fringe of spinelets bordering the entire animal, which is interrupted only at the distal
ends of the furrows. The ventral side is quite flat. The inferomarginals are com-
pletely naked ventrally and the two first ones in each interradius cover most of the

254 A- M- CLARK

interradial area together with the single actinal plate proximal to them. Two of the
five actinal plates bear a single median distal spinelet but the other three are com-
pletely bare. There are six pairs of podia in each ray. Except for the first plate, the
adambulacrals bear three spines in a more or less straight line at right angles to the
furrow. The first plates may have four spines, the second one aligned obliquely to
the first and clearly potentially a furrow spine. Presumably during further growth
the second spine on the succeeding plates also shifts to the furrow edge of the plate.
The oral plates already have four furrow spines and one suboral, as in most adults.

The general appearance of this specimen is remarkably like the holotype of
Marginaster fimbriatus Sladen, 1889, as shown in Sladen's pi. 58, figs 4-6 ; that has
R 6 mm and was collected to the north of Ireland. According to Ludwig it is
conspecific with Marginaster capreensis from the Mediterranean but I consider that its
affinities are much more likely to be with Poraniomorpha. The only difference
between M. fimbriatus and this juvenile from the Indian Ocean is the development of
a second row of actinal plates, which would be expected at the larger size.

There is no question that this small specimen could be a young Patiriella exigua
since there are examples of that species of comparable size and they show a much
more granuliform armament, with single shorter adambulacral spines. Also the
contours are quite different, the abactinal side rising vertically above the relatively
smaller and more numerous inferomarginal plates.

GENERIC POSITION. Perrier described this species as belonging to a separate
section of the genus Culcita (now included in the tropical shallow-water family
Oreasteridae of the order Valvatida), this section being characterized by the fleshy
body wall concealing the skeleton. He also noted affinities with Asteropsis (currently
aligned near the Poraniidae in the order Spinulosida) .

In 1905 Bell identified a fleshy pentagonal South African asteroid as Culcita
veneris but Mortensen (1933) referred this specimen to Cryaster brachyactis H. L.
Clark, 1923, also from South Africa. At the same time Mortensen pointed out that
Culcita veneris is not a Culcita but probably belongs to the Asteropidae or perhaps the
Cryasteridae, though in a footnote he doubts the distinction of these two nominal
families.

In 1940 Fisher demonstrated that Cryaster antarcticus Koehler, 1906, the type-
species of that genus, is congeneric with Perknaster fuscus Sladen, 1889 and so
referable to the family Ganeriidae. However, he recognized that this affiliation does
not extend to Cryaster brachyactis, for which he established a new nominal genus
Spoladaster, referring it to the Asteropidae 'in the vicinity of Tylaster' . On the
strength of Mortensen's comments, Fisher noted that Culcita veneris can probably
also be referred to Spoladaster, an idea shared by me in 1952, when I suggested also
that the South African Tylaster meridionalis Mortensen, 1933 might be based on
smaller specimens of Spoladaster brachyactis. If this last premise proves true, then
the problem is whether the species from South Africa and from Amsterdam and St
Paul Islands are congeneric with Tylaster willei Danielssen & Koren, 1881, from the
deep Norwegian Sea, this being the type species of Tylaster.

More immediately there is still a possibility that specimens from Amsterdam and
St Paul Islands could be conspecific with those from South Africa, in spite of

ASTEROZOA FROM THE SOUTHERN INDIAN OCEAN 255

Mortensen's denial of this (probably without seeing specimens from the islands).
Zoogeographically this would be quite likely in view of the fact that Patiriella
exigua, one of the two other asteroids known from these isolated islands, is also found
in southern Africa, though it occurs also in temperate Australia. In addition,
at least two species of ophiuroids, as well as the cosmopolitan Amphipholis
squamata, are here recorded as common to this island group and to either South
Africa or southern Mozambique.

However, the specimen of Spoladaster brachyactis in the British Museum collections
from False Bay shows a distinct grouping of abactinal papulae in the meshes of an
irregular reticulum, as described by H. L. Clark in the holotype. Also there is a
separate horizontal band of papulae laterally in what should be the intermarginal
area (PI. 3, fig. 3, top). H. L. Clark describes 'innumerable minute' abactinal plates
bearing the spinelets in the holotype, though Fisher found in the paratype that
abactinal plates of any sort are only present at the arm tips and I am unable to find
such plates in the general abactinal body wall either. H. L. Clark also describes a
series of 'rather large plates buried in the skin' between abactinal and actinal areas,
i.e. inferomarginal plates, while Fisher found that the paratype has 'fairly large
phanerozonid inferomarginal plates imbedded in the leather body wall'. However,
the X-radiograph taken of the British Museum specimen shows no trace of any
marginal plates (PI. 5, fig. 3), though the external view (PI. 4, fig. 3) shows that some
short inferomarginal spines are present. Unfortunately this is the only South
African specimen now in the British Museum collections and I hesitate to mutilate it
further to ascertain if any rudiments of marginal or actinal plates, as described by
Fisher, are visible on the inside of the body wall. In Mortensen's smaller holotype of
Tylaster meridionalis from South Africa with R 28 mm there is a continuous marginal
fringe of short spines, with a parallel series of a few similar actinal spines close by on the
distal part of each actinal area mid-interradially. In another specimen (presumably
of similar or smaller size since it was not selected as holotype) which he dried,
Mortensen made visible three or four irregular series of actinal plates. The Cape
Town University specimen with R c. 80 mm which I referred to Spoladaster brachyactis
in 1952 has numerous actinal spines, which would be much coarser than the micro-
scopical actinal spinelets of the Amsterdam/St Paul material. Finally, the
adambulacral plates of 5. brachyactis (and of Tylaster meridionalis - if distinct) bear
only one furrow spine apart from the subambulacral spine, whereas the Amsterdam/
St Paul specimens have two furrow spines. I have no doubt therefore that the latter
specimens are specifically distinct, so that 5. veneris is a species endemic to the islands.

The differences can be summed up as follows :

Spoladaster veneris Spoladaster brachyactis

No separate intermarginal band of An intermarginal band of papulae

papulae laterally distinct

No enlarged spines at the junction of Some inferomarginal spines usually

the abactinal and actinal areas, nor present and in smaller specimens (R

on the actinal areas c. 35 mm) also some actinal spines

Two furrow spines on most Furrow spines usually single

adambulacral plates

256 A. M. CLARK

The problem remains as to whether or not these two species should be regarded
as congeneric with one of the northern hemisphere genera such as Tylaster or
Poraniomorpha, both of Danielssen & Koren, 1881.

Unfortunately only a single young specimen of Tylaster willei is in the British
Museum collections. This is the smaller specimen from Norwegian North Atlantic
Expedition station 200 (7i°25'N ; i5°4i'E, 1134 metres). It has R only c. 12 mm.
The triangular arms are curled up dorsally but the form is markedly stellate rather
than pentagonal. There are prominent inferomarginal and actinal spines similar in
size to the adambulacral spines. The species was very well described by Danielssen
& Koren in 1884 (pp. 64-67, pi. n, figs 1-6), presumably on the basis of the single
larger specimen, which was from the same station as this one. R is given as 40 mm
and R/r can be calculated as i-6/i, the form again being markedly stellate. Drying
of the dorsal body wall revealed numerous isolated spinelets, described as needle-like
but appearing to be quite capitate in their pi. n, fig. 3. Viewing the body wall from
the inside showed a horizontal series of rudimentary isolated lateral plates which were
assumed to be the superomarginals and below these three series of more regular small
plates linked by connective tissue, also approximately horizontal, the upper one of
the three being taken for the inferomarginals and bearing externally a horizontal
series of marginal spines in groups of usually three corresponding to the plates.
Many of the actinal plates also bear similar but single spines forming several series,
while there are some scattered additional proximal actinal spines.

Tylaster willei thus seems intermediate in form and skeletal development between
the Spoladaster spp. and the wide-ranging genus Porania, the latter having a usually
stellate form and an articulated skeleton.

Not surprisingly, the Amsterdam/St Paul species is remarkably similar to some
North Atlantic pentagonal cushion-like specimens which were likewise ascribed to
Culcita under the name Culcita borealis Sussbach & Breckner, 1911, from the area of
the Shetlands, which name Farran (1913) also gave to some specimens from the west
of Ireland. In 1914 Verrill established a new nominal genus Cukitopsis for C.
borealis, pointing out that its affinities are with Tylaster and Chondraster of the family
Poraniidae (which he kept distinct from the Asteropidae) .

Ostergren (1904), Grieg (1907) and Koehler (1909) emphasized that Poraniomorpha
hispida (M. Sars in G. O. Sars, 1872) is a very variable species, Koehler including
records from both sides of the Atlantic in his lengthy synonymy. In 1927 Grieg
referred Culcitopsis borealis to the synonymy of P. hispida and Mortensen (1927)
agreed that 'there is hardly any doubt' this is correct. Most subsequent workers have
accepted Mortensen's judgement so that the synonymy of this species includes the
following :

Goniaster hispida M. Sars in G. O. Sars, 1872 : 28 (R of the holotype n mm) : Storm, 1881 : 90.

Asterina borealis Verrill, 1878 : 213 (R 12 mm).

Porania spinulosa Verrill, 1879 : 202 (R 40 mm).

Poraniomorpha rosea Danielssen & Koren, 1881 : 189 (Re. 27 mm, judging from the scale of

the figures).

Rhegaster murrayi Sladen, 1883 : 156 ; 1889 : 368 (R 14-3 mm).
Lasiaster villosus Sladen, 1889 : 372 (R 10-5 mm).

ASTEROZOA FROM THE SOUTHERN INDIAN OCEAN 257

Poraniomorpha hispida : Ostergren, 1904 : 615 ; Koehler, 1909 : 100 ; Grieg, 1927 : 129 (with

forma borealis Sussbach & Breckner) ; Mortensen, 1927 : 92.
Poraniomorpha (Lasiaster) hispida : Grieg, 1907 : 40.

Culcita borealis Sussbach & Breckner, 1911 : 217 (R 40 mm) ; Farran, 1913 : 15.
Poraniomorpha villosa : Farran, 1913 : 17.
Culcitopsis borealis (Sussbach & Breckner) Verrill, 1914 : 21.
Rhegaster borealis (Verrill) Verrill, 1914 : 17.

As shown by the R measurements for the holotype of each nominal species included
above, most of them are based on relatively small specimens, the exceptions being
Porania spinulosa, Poraniomorpha rosea and Culcita borealis. This is very unfortun-
ate since small specimens of poraniids may give an exaggerated impression of the
extent of skeletal development in the adult, as the evidence suggests there is often a
tendency for progressive resorption with growth. It is significant that there are
several other North Atlantic poraniids with markedly reduced skeletons, namely
C hondr aster grandis (Verrill, 1879), Chondraster hermanni Madsen, 1959 (treated as a
subgenus of Porania}, Pseudoporania stormi Dons, 1936, Sphaeraster berthae and
Sphaeraster bjorlykkei Dons, 1938 - the last nominal genus preoccupied and subse-
quently renamed Sphaeriaster by Dons, all based on relatively huge specimens with
R at least 80 mm and often 100 mm or even more. Madsen (1959) considers that all
these are congeneric. I believe that their large size is responsible at least in part
for their very reduced skeletons. The X-radiographs given by Dons and Madsen
indicate that their material completely lacks abactinal and actinal plates of any
substance and only Dons' Pseudoporania stormi has vestiges of supero- and infero-
marginal plates.

The holotype of Poraniomorpha hispida is described as pentagonal in form,
R/r i -2/i, flat below and convex above, with both marginal series of plates distinct
(probably by their contours). It came from the Lofoten Islands, off northern
Norway. A larger specimen, R 40 mm, from Trondheim Fjord further south,
presented to the British Museum by Storm, agrees well with this description (PI. 3,
fig. 4 ; PI. 4, fig. 4 ; PI. 5, fig. 4). It was X-radiographed for comparison with
Spoladaster brachyactis and 5. veneris, revealing very well-developed series of block-
like supero- and inferomarginal plates outlining what appears to be an almost
continuous pavement of slightly spaced plates, apparently mainly actinal plates but
probably also some superimposed abactinals. In contrast, a cushion-like specimen
in the British Museum collections, from the warm area of the Faeroe Channel, halved
and dried (PI. 4, fig. 5) shows only vestigial microscopic abactinal platelets on the
inside of the body wall, small isolated superomarginals barely distinguishable
interradially, slightly less reduced but still discontinuous inferomarginals and two
well-separated series of isolated actinal plates with a group of additional proximal
ones in each area. The inferomarginals and actinals are clearly in process of
resorption, most of them being crescentiform or ring-shaped.

This degree of skeletal reduction approximates to that shown by Spoladaster and
at the same time makes it difficult to regard C. borealis as conspecific with well-
calcified specimens of Poraniomorpha hispida such as the one from Trondheim Fjord.
Although he did not mention Poraniomorpha in his 1959 paper, Madsen's remarks on

258 A. M. CLARK

the variations of the northern Porania pulvillus suggest that the distinction between
that and Poraniomorpha is by no means sharp and needs further study, especially of
growth changes. Until this can be done for the northern poraniids and the various
taxa can be redefined, it seems unwise to alter further the nomenclature of these two
southern species, which I accordingly propose to leave under the name of Spoladaster.

OPHIUROIDEA

Amphipholis squamata (Delle Chiaje, 1829)

Asterias squamata Delle Chiaje, 1829 : 74, 77, figs 1-4.
Amphipholis minor : Hertz, 1927 : 35.
Amphipholis squamata : Mortensen, 1933 : 364-365.

MATERIAL. Amsterdam Island : mostly from mid- to sub-littoral, in algae or
coralligenous substrates or rock pools from various parts of the shore, some from
grab hauls off the east coast down to 80 metres, others from stomachs of fishes such as
Acantholatris monodactylus, 135 specimens.

St Paul Island : from washings of algae and Laminaria roots, east coast towards
Roche Quille, 3-4 metres, 24 specimens ; washed from Macrocystis roots, 62
specimens ; from various parts of the crater, mostly under sublittoral stones or in
washings from sponges, algae, bryozoa, compound ascidians, etc., 150 specimens.

Sea-mount south-east of St Paul Island : with compound ascidians, the scler-
actinian coral Caryophyllia profundi and bryozoa at 120 metres, i specimen.

NOTES. Most of these specimens are very small and in poor condition. Some
were mixed with Amphiura capensis and also with young Patiriella exigua.

Mortensen has already pointed out that the Amphipholis from St Paul Island
crater recorded by Hertz are not distinguishable from A . squamata, the only cosmo-
politan echinoderm.

Amphiura capensis Ljungman, 1867
Amphiura capensis Ljungman, 1867 : 320 ; Mortensen, 1933 : 348-350.

MATERIAL. Amsterdam Island : washed from coralligenous bottom at 35 metres,
north-east coast, i specimen ; from stranded laminarians, near La Cale, north coast,
3 specimens.

St Paul Island : from Macrocystis and Laminaria roots, 23 specimens ; from
various parts of the crater, usually under sublittoral stones, 29 specimens.

NOTES. Most of the specimens from the crater of St Paul Island were taken with
Amphipholis squamata and some also with young Patiriella exigua. All are small, the
disc diameter not exceeding 4 mm. The arm spines number up to five proximally
and are markedly flattened and spatulate. Some specimens have the disc scaling
deficient near the oral shields, as shown also in many South African specimens.

This species is easily distinguished from Amphipholis squamata by the radial
shields being only contiguous distally, the arm spines being more numerous and a
different shape and the distal oral papillae and tentacle scales being single.

ASTEROZOA FROM THE SOUTHERN INDIAN OCEAN 259

These records represent an extension of range from southern Africa, where
Amphiura capensis is known from Luderitz Bay to Natal.

Amphiura sp.

MATERIAL. St Paul Island : on coarse sandy bottom with Venus at 50-80 metres
off the north-east coast, 4 specimens.

Sea-mount south-east of St Paul Island : with compound ascidians, Caryophyllia
and bryozoa at 120 metres, 2 specimens.

NOTES. Unfortunately all these specimens are in poor condition. The disc was
probably fully scaled, the radial shields about a third as long as the disc radius and
contiguous for their distal third, inwardly divergent. The oral shields are as long as
broad or slightly longer, with the distal lobe shorter than the proximal angle. The
distal oral papilla is blunt. The dorsal arm plates are fan-shaped. There are four
tapering but blunt-tipped arm spines. The tentacle scales mostly number two but
one specimen has more pores with only a single small scale. None of these characters
are particularly distinctive.

This species does not agree with any so far recorded from southern Africa, judging
from these specimens.

Ophiactis sp. probably O. savignyi Miiller & Troschel, 1842

MATERIAL. Amsterdam Island : dredged on consolidated bottom at 40-50
metres off north coast, i specimen.

St Paul Island : washed from various coralligenous encrusting organisms, com-
pound ascidians, Caryophyllia, bryozoa, etc. at 60-80 metres off the north-east coast,
i specimen.

NOTES. Both specimens are fissiparous with six arms ; one has three arms larger
and three regenerating, the other has two and four.

Even the larger specimen has the disc diameter only just exceeding i mm. The
disc scales are relatively coarse at this small size and the proportions of the radial
shields not obvious - partly due to the regeneration. One specimen has disc spines,
the other has not. Some of the jaws have two distal oral papillae on at least one
side, elsewhere there is only one distinct papilla. The dorsal arm plates are fan-
shaped but the distal curve is flattened medially. There are four rugose arm spines,
all but the uppermost one truncated.

The occurrence of two distal oral papillae in some cases and the shape of the arm
spines make it almost certain that these two small specimens are referable to the
'tropicopolitan' Ophiactis savignyi, which ranges down the coast of East Africa to Natal.

Ophiocoma pusilla (Brock, 1888)

Ophiomastix pusilla Brock, 1888 : 499.
Ophiocoma pusilla : Devaney, 1970 : 25-28.

MATERIAL. Amsterdam Island : on consolidated bottom at 40-50 metres off the
north coast, i specimen.

260 A. M. CLARK

Sea-mount south-east of St Paul Island : with compound ascidians, Caryophyllia
and bryozoa at 120 metres, 10 specimens.

NOTES. The largest specimen has disc diameter 7-5 mm and is the only one to
show the swollen-tipped arm spines on a few segments beyond the arm bases so
characteristic of this diminutive species of Ophiocoma.

The granulation of the disc distinguishes this from any of the other species recorded
from these two islands.

Ophiocoma pusilla is known from isolated records throughout the tropical Indo-
West Pacific, as far south in East Africa as Inhaca, southern Mozambique. It is
usually found in coral heads, mostly in shallow water but possibly down to just over
100 metres. Because of its furtive habits and likelihood of confusion with other
species of Ophiocoma, it may well have been overlooked at greater depths.

One of the two localities at which this species was collected is common to the
Ophiactis discussed above. If this really is Ophiactis savignyi, as seems very likely,
then the extension of range to Amsterdam and St Paul Islands of another tropical
reef species is not surprising.

REFERENCES
Papers not marked with an asterisk relate only to North Atlantic Poraniidae.

*ARNAUD, P. M. & BEUROIS, J. 1971 [1972]. Premiere signalisation des genres Charonia et

Ranella aux iles St-Paul et Amsterdam (Ocean Indien) et revision du genre Argobuccinum

(Gastropoda Cymatiidae). Tethys 3 (4) : 865-874.
*BELL, F. J. 1905. The Echinoderma found off the coast of South Africa. 2. Asteroidea.

Marine Invest. S. Afr. 3 : 241-253.

*BROCK, J. 1888. Die Ophiuridenfauna des indischen Archipels. Z. wiss. Zool. 47 : 465-539.
*CLARK, A. M. 1952. Some Echinoderms from South Africa. Trans R. Soc. S. Afr. 33 : 193-

221, 3 figs, pi. 17.
*CLARK, H. L. 1923. The Echinoderm fauna of South Africa. Ann. S. Afr. Mus. 13 : 221-

435, 4 figs, pis 8-13.
* 1926. Echinoderms from the South African Fisheries and Marine Biological Survey.

2. Sea-stars (Asteroidea). Rep. Fish. mar. biol. Surv. Un. S. Afr. No. 4 : Spec. Rep.

No. 7 : 1-34, 7 pis.
DANIELSSEN, D. C. & KOREN, T. 1881 . Fra den norske Nordhavs-expedition : Echinodermer.

Nyt. Mag. Naturvid. 26 : 177-195, 2 pis.

— 1884. Asteroidea. Norwegian North- Atlantic Expedition, 1876-1878. Zoology.

n8pp., 15 pis. Christiana.
*DARTNALL, A. J. 1971. Australian sea stars of the genus Patiriella (Asteroidea, Asterinidae) .

Proc. Linn. Soc. N.S.W. 96 : 39-49, i fig., 2 pis.
*DELLE CHIAJE, S. 1829. Asteria. Memorie sulla storia e notomia degli animali senza vertebre

del Regno di Napoli, 3 : 74-81. Napoli.
*DEVANEY, D. M. 1970. Studies on ophiocomid brittlestars. i. Smithson. Contr. Zool.

No. 51 : 1-41, 49 pis.
*DODERLEIN, L. 1927. Die Seesterne der deutschen Siidpolar-Expedition, 1901-1903.

Dt. Sudpol. Eyped. 19, Zool. n : 291-301, 4 pis.
FARRAN, G. P. 1913. The deep-water Asteroidea, Ophiuroidea and Echinoidea of the west

coast of Ireland. Sclent. Invest. Fish. Brch Ire. 1912 (6) : 1-66, 17 figs, 2 pis.
*FISHER, W. K. 1906. The starfishes of the Hawaiian Islands. Bull. U.S. Fish. Comm.

23 (3) : 987-1130, 49 pis.

ASTEROZOA FROM THE SOUTHERN INDIAN OCEAN 261

*FISHER, W. K. 1940. Asteroidea. Discovery Rep. 20 : 69-306, 23 pis.

*GRAY, J. E. 1866. Synopsis of the species of starfish in the British Museum, iv+iy pp.,

1 6 pis. London.
GRIEG, J. A. 1907. Echinodermen von dem norwegischen Fischereidampfer "Michael Sars"

in den Jahren 1900-1903 gesammelt. Asteroidea. Bergens Mus. Aarb. 1906 (13) : i-

87, 10 figs, 2 pis.

1927. Echinoderms from the west coast of Norway. Nyt Mag. Naturvid. 65 : 127-

136, 6 figs.

*HERTZ, M. 1927. Die Ophiuriden der deutschen Siidpolar-Expedition, 1901-1903. Dt.

Sudpol. Exped. 19, Zool. n : 1-56, 9 pis.
KOEHLER, R. 1909. Echinoderms provenant des campagnes du yacht "Princesse Alice".

Result. Camp, scient. Prince Albert 34 : 1-317, 32 pis.
*LAMARCK, J. B. P. A. DE 1816. Radiare Echinodermes. Histoire naturelle des Animaux

sans Vertebres, Ed. i, 2 : 522-568 ; 3 : 1-76. Paris.
*LJUNGMAN, A. 1867. Ophiuroidea viventia hue usque cognita enumerat. Ofvers. K.

VetenskAkad. Fdrh. Stockh. 23 : 303-336.
*MADSEN, F. J. 1958. On Sphaeriodiscus placenta and a few other sea-stars from W. Africa.

Bull. Inst.fr. Afr. noire 20A : 90-94, 4 figs.

1959. On a new North Atlantic sea-star, Chondraster hermanni n. sp., with some re-
marks on related forms. Vidensk. Meddr dansk naturh. Foren. 121 : 153-160, 4 figs.

MORTENSEN, T. 1927. Handbook of the Echinoderms of the British Isles. 1x4-471 pp., 269
figs. London.

* 1933- Echinoderms of South Africa (Asteroidea and Ophiuroidea). Vidensk. Meddr

dansk naturh. Foren. 93 : 215-400, 91 figs, 12 pis.

*MULLER, J. & TROSCHEL, F. H. 1842. System der Asteriden. xx+134 pp., 12 pis. Braun-
schweig.

OSTERGREN, H. 1904. Ueber die arktischen Seesterne. Zool. Anz. 27 : 615-616.
*PERRIER, E. 1879. Les Stellerides de 1'ile Saint-Paul. Arch. zool. exp. gen. 31 : 47-49,

pi. 4.
SARS, G. O. 1872. Nye Echinodermer fra den Norske Kyst. Fork. VidenskSelsk. Krist.

1871 : 1-31.

SLADEN, W. P. 1883. The Asteroidea of H.M.S. "Challenger" Expedition. 2. /. Linn.
Soc. Zool. 17 : 214-269.

1889. Asteroidea. Rep. scient. Results Voy. Challenger. Zool. 30: 1-935, 118 pis.

STORM, V. 1881. Bidrag til Kundskab om Throndhjemsfjordens Fauna. 3. K. nor she

Vidensk. Selsk. Skr. 1880 : 73-96.
SUSSBACH, S. & BRECKNER, A. 1911. Die Seeigel, Seesterne und Schlangensterne der Nord-

und Ostsee. Wiss. Meeresunters. Kiel N.S. 12 : 167-300, 3 pis.

*TORTONESE, E. 1965. Echinodermata. Faunad' Italia. 6 : xv + 422 pp., i86figs. Bologna.
VERRILL, A. E. 1878. Notice of recent additions to the marine fauna of the eastern coast of

North America. Amer. J. Sci. 16 : 207-215, 371-378.

1879. Notice of recent additions to the marine Invertebrata of the northeastern coast

of America. Proc. U.S. natn. Mus. 2 : 165-205.

1914. Revision of some genera and species of starfishes, with descriptions of a few new

genera. Ann. Mag. nat. Hist. 14 : 13-22, i pi.

AILSA M. CLARK

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON SW7 530

PLATE i
Sphaeriodiscus mirabilis sp. nov. Holotype in dorsal view (scale in mm).

Bull. Br. Mus. nat. Hist. (Zool.) 30, 6

PLATE i

16

PLATE 2
Sphaeriodiscus mirabilis sp. nov. Holotype in ventral view (scale in mm).

Bull. Br. Mus. nat. Hist. (Zool.) 30, 6

PLATE 2

16*

PLATE 3

FIGS i & 2 (right). Spoladaster veneris (Perrier). B.M. reg. no. 1975.12.4.164, 165. St. ST
PAUL-yi, dried half specimen opened out, R 58 mm, viewed internally ; intact specimen,
R 52 mm, viewed dorsally.

FIG. 3 (top left). Spoladaster brachyactis (H. L. Clark). 1904.4.20.133, False Bay, South
Africa, R 58 mm, viewed dorsally.

FIG. 4 (bottom left). Poraniomorpha hispida (M. Sars). 91.4.15.1, Trondheim Fjord, Norway,
R 40 mm, viewed dorsally. (See PI. 4 for scale.)

Bull. Br. Mus. nat. Hist. (Zool.) 30. 6

PLATE 3

PLATE 4

FIGS i & 2 (right). Spoladaster veneris.
FIG. 3 (top left). Spoladaster brachyactis.
FIG. 4 (bottom left). Poraniomorpha hispida ; all as in PL 3 but in external or ventral views.

FIG. 5 (centre). Poraniomorpha hispida forma borealis (Sussbach & Breckner), 1974.1.4.1,
Faeroe Channel, west of Wyville Thomson Ridge, dried half specimen, side view (scale in mm).

Bull. Br. Mus. nat. Hist. (Zool.) 30, 6

PLATE 4

PLATE 5

FIGS i & 2 (right). Spoladaster veneris.
FIG. 3 (top left). Spoladaster brachyactis.
FIG. 4 (bottom left). Poraniomorpha hispida ; all as in PI. 3, X-radiographs.

Bull. BY. Mus. mat. Hist. (Zool.) 30, 6

PLATE 5

PLATE 6

Spoladaster veneris (Perrier). Live whole specimen and one ray in dorsal view, photographed
by Dr Beurois. (Superimposed on black.)

Bull. Br. Mus. nat. Hist. (Zool.) 30, 6

PLATE 6

RESTRICTION OF THE FAMILY PORANIIDAE,
SENSU SPENCER & WRIGHT, 1966
(ECHINODERMATA : ASTEROIDEA)

By FREDERICK H. C. HOTCHKISS AND AILSA M. CLARK

SYNOPSIS

The amalgamation of two families of asteroids under the name Poraniidae in the Asterozoa
section of the Treatise on invertebrate paleontology by Spencer & Wright (1966) is here rejected
on the basis of the fundamentally different arrangement of the actinal plates in the two groups
of genera concerned.

INTRODUCTION

THE initiative for this note and the observation on which it is based came from one of
us (F. H. C. H.), the history of the problem being dealt with by the other. The
specimens used to illustrate it were specially prepared (by treatment with KOH
followed by glycerine to increase the transparency of the normally opaque skin and
reveal the skeleton) by Mr R. H. Harris of the Histology and Preservation Section of
the Zoology Department, British Museum (Natural History), and photographed to
emphasize further the skeletal structure by Mr P. A. Richens of the Photographic
Unit of the Museum, to whom we are indebted.

HISTORY

In 1840 three nominal genera of stellate, thick-skinned asteroids were described :

(1) Aster ope Miiller & Troschel (i84oa) with type Asterias carinifera Lamarck,
1816, by monotypy, renamed Asteropsis by the same authors (i84ob) several months
later (without explanation but presumably because they had found the first name
preoccupied) ;

(2) Gymnasteria Gray for G. spinosa and G. inermis spp. nov. ; and

(3) Porania Gray for 'Asterias gibbosa Leach', said to be Leach 1817 but probably
from an MS catalogue of the British Museum collections. (This is not the same
species as Asterias gibbosa Pennant, 1777, which Gray gives as Asterina gibbosa on the
next page.) Gray's paper was published in December, 1840 and post-dated both of
Miiller & Troschel's (see Fisher, 1908).

In their main work, the System der Asteriden in 1842, Miiller and Troschel cited
both Gymnasteria and Porania as synonyms of Asteropsis. They included in the
genus A. carinifera (Lamarck), Asteropsis ctenacantha sp. nov. (referred to the
synonymy of Porania pulvillus (O. F. Miiller) by Perrier, 1875), Asterias pulvillus

Bull. BY. Mus. nat. Hist. (Zool.) 30, 6 Issued 25 November 1976

264 F- H- c- HOTCHKISS AND A. M. CLARK

O. F. Muller, 1788, with synonym Porania gibbosa Gray, and Asterias vernicina
Lamarck, 1816 ; also in a postscript Gray's Gymnasteria spinosa and G. inermis,
which they thought to be conspecific at least with each other.

After study of Gray's types of G. spinosa and G. inermis in the British Museum
collections, Perrier (1875 : 285) referred both to the synonymy of Asterias carinifera
Lamarck. This should automatically have relegated Gymnasteria Gray where it
belongs in the synonymy of Asteropsis Muller & Troschel. Although Perrier was
aware of Muller & Troschel's usage of Asteropsis in 1840 before Gray's of Gymnasteria
(see 1875 : 9-10), for some reason he retained the name Gymnasteria for A. carinifera
and at the same time improperly emended Asteropsis to include only Asterias
vernicina Lamarck, ineligible as type-species since not included in 1840. Since he
simultaneously synonymized Petricia punctata Gray, 1847, the type species of
Petricia Gray, with A . vernicina, this move dispensed with a valid generic name while
perpetuating at the same time an invalid one.

Unfortunately Sladen was apparently unaware that Muller & Troschel had used
Asteropsis already in 1840 and perpetuated Perrier's errors until they were pointed
out by Fisher (1908 and 1911 : 247, 248) when he noted that Asterope Muller &
Troschel, 1840 is not invalidated by Asterope Philippi, 1840, having been published at
least two months earlier. He therefore referred Gymnasteria Gray to the synonymy
of Asterope, rejected Asteropsis as superfluous and revived Petricia Gray from
synonymy.

As for the family position of these genera, in 1875 Perrier included Porania,
Asteropsis (emended), Dermasterias nov. and Gymnasteria at the end of the Goni-
asteridae, following them with the Asterinidae and implying some affinity with the
latter family. In 1879 Viguier referred the first three genera to the Asterinidae.
However, in 1884 Perrier divided the families of asteroids into named orders,
placing the Asterinidae in the order Spinulosae but establishing a new family
Gymnasteriadae in the order Valvulatae (p. 165) (although on p. 154 he lists a family
Asteropsidae in the order Valvatae !). Included in the Gymnasteriadae were
Gymnasteria, Porania, Asteropsis, Dermasterias and Marginaster gen. nov. In the
Challenger report, Sladen (1889 : 355-356) adopted Gymnasteriidae (with corrected
spelling) and added Tylaster and Poraniomorpha, both of Danielssen & Koren, 1881,
also Rhegaster Sladen, 1883 and Lasiaster gen. nov. ; he commented on the affinity of
some of these with the Asterinidae.

After discovery of the echinasterid which he called Poraniopsis, Perrier in 1891
(pp. 107, 163) removed Porania from the Gymnasteriidae to the Asterinidae, adjacent
to the Echinasteridae in the order Spinulosae. He made no mention in the discussion
of Porania that it should be referred to a family of its own but in 1894 (p. 163) took
this action, though citing 'Poraniidae Perrier, 1891'. This is more puzzling because
in his preliminary discussion on classification in the 1894 Travailleur and Talisman
report he makes not the slightest mention of Poraniidae and indeed lists Marginaster
pentagonus (the only relevant species collected by these vessels) under the family
Pentagonasteridae (i.e. Goniasteridae) on p. 30. On p. 163 he placed Poraniidae in
the Spinulosae and included in it Porania, Tylaster, Poraniomorpha, Marginaster,
Rhegaster and Lasiaster. This would leave only Gymnasteria, Asteropsis and

RESTRICTION OF PORANIIDAE (ASTEROIDEA) 265

Dermasterias in the family Gymnasteriidae, which is listed on p. 327 under the order
Valvata [sic]).

Having in 1908 renamed as Asteropidae Perrier's Gymnasteriidae, in 1911 Fisher
referred Porania and its relatives to the Asteropidae. Verrill (1914 : 304) expressed
doubt about this because of affinity of Porania with the Asterinidae and in 1919
Fisher separated them again, though removing also Dermasterias to the Poraniidae,
leaving only Asterope and Petricia in the Asteropidae.

In the years following, several new nominal genera were added to the Poraniidae,
while others were synonymized. Mortensen (1933 : 249) included under the heading
Asteropidae Tylaster meridionalis and 'Cry aster' brachyactis (both more closely
related to Porania than to Asterope) ; then Fisher (1940 : 136) named a new genus
Spoladaster for C. brachyactis, again giving the family as Asteropidae (while on
p. 154 he deals separately with the family Poraniidae) a treatment also followed by
A. M. Clark in 1952. H. L. Clark (1946 : 109) also evidently considered that
Poraniidae and Asteropidae can be kept distinct. However, Madsen (1959 1153)
refers to Porania under Asteropidae, order Spinulosa.

Spencer & Wright (1966 : U6g) in the Treatise on invertebrate paleontology again
unite the two groups of genera but under the name Poraniidae Perrier, 1894, in the
order Spinulosida. They also show that Asterope Miiller & Troschel is after all
invalid, being preoccupied by Asterope Hiibner, 1819, an insect, and must be replaced
by Asteropsis (a fact noted independently by A. M. Clark in 1967).

TAXONOMY

Clearly with so much anomalous treatment it is time that this problem was
re-examined by further study of the skeleton obscured beneath the thick skin of these
asteroids. The use of X-radiography by Dons (1936, 1938) and Madsen (1959) has
revealed the extent of reduction of the skeleton in some species related to Porania
and even more to Poraniomorpha but it is comparison between the better-calcified
Porania itself and Asteropsis that is needed.

Examination revealed that there are two fundamentally different arrangements of
the plates in the triangular actinal intermediate areas under each side of the arm
among the genera of Poraniidae sensu Spencer & Wright. In Porania itself and also
Poraniomorpha and Tylaster, for instance, the arrangement consists of rows of actinal
plates running parallel to the inferomarginals, the zone of plate addition is adjacent
to the adambulacrals and the shortest row is in the proximal part of the interradial
area near the oral plates (PI. i). Conversely, in Asteropsis, Dermasterias and Petricia
the arrangement consists of rows of plates running parallel to the adambulacrals, the
zone of addition is adjacent to the inferomarginals and the shortest row adjoins the
interradial inferomarginals in the distal part of the interradial area (PI. 2 ; PI. 3).
The second pattern is the usual one in the families of the order Valvatida (e.g.
Goniasteridae and Oreasteridae) as well as in some Spinulosida (e.g. Asterinidae).

It is therefore possible to divide the genera concerned into two distinct groups.
In the cases where the actinal plates are either few and irregular (e.g. the diminutive
species referred to Marginaster*) or extremely reduced (e.g. Spoladaster and Tylaster)

* Which I think may all prove to be based on juveniles of other Poraniid genera. A. M. C.

266

F. H. C. HOTCHKISS AND A. M. CLARK

by resorption in adults, the pattern can be estimated by the locus of plate formation,
judged by the sizes of the plates or the positions of the surviving plates when the
older ones have been resorbed.

The two groups of genera are as follows :

ASTEROPSEIDAE*Hotchkiss & Clark, 1976

Asteropsis Miiller & Troschel, 1840
Dermasterias Perrier, 1875
Petricia Gray, 1847

PORANIIDAE Perrier, 1894

Porania Gray, 1840

Chondraster Verrill, 1895

Marginaster Perrier, 1881

Poraniella Verrill, 1914

Poraniomorpha Danielssen & Koren, 1881

Poranisca Verrill, 1914

Pseudoporania Dons, 1936

Sphaeriaster Dons, 1939

Spoladaster Fisher, 1940

Tylaster Danielssen & Koren, 1881

The Triassic genus Trichasteropsis Eck, 1879 (illustrated by Spencer & Wright,
1966, fig. 43-3a-c) has poraniid actinal growth gradients and should be transferred
to the Poraniidae from the Palasterinidae (order Paxillosida) .

Concerning the Ordovician somasteroid Chinianaster levyi Thoral, Spencer's
reconstruction of this (1951, fig. 7 ; Spencer & Wright, 1966, fig. 39.4) shows poraniid-

I'

II

Fig. i.

Fig. 2.

FIG. i. Diagram showing orientation and polarity of growth gradients as observed by

Fell.

FIG. 2. Diagram showing orientation and polarity of growth gradients observed in

Poraniidae.

*This family name is based on Asteropsis Miiller & Troschel, 18405, which is the earliest available
name to replace Asterope Mtiller & Troschel, i84oa, and is a senior subjective synonym of Gymnasteria
Gray, (Dec.), 1840 (see Miiller & Troschel, 1842 : 65). Asteropsidae Perrier (1884:154) is not valid under
Article 55 of the Code because he exluded the the type-species, A. carinifera Lamarck, from the type-
genus, Asteropsis. Gymnasteriidae (as Gymnasteriadae) Perrier (1884 1165), based on the junior synonym,
though adopted by some nineteenth century authors, has not been used for nearly 70 years and may be
considered to have lost all claims to validity under Article 40. The inserted ' e ' in Asteropseidae derives
from the genitive singular root Asteropse-. We are indebted to Dr R. V. Melville for help with this
nomenclatorial problem.

RESTRICTION OF PORANIIDAE (ASTEROIDEA) 267

like growth gradients, the shortest longitudinal series adjoining the ambulacra
proximally. This contrasts with Fell's 'inferred appearance' of the same species
(1963, fig. nD), where minor differences in the sutures result in an asteropseid
arrangement, conforming with his theories on the phylogeny of asteroids (Fig. i). If
Spencer is proved correct, then Fell's growth gradient argument for crinoid ancestry
of the Asterozoa is seriously undermined.

REFERENCES

CLARK, A. M. 1952. Some echinoderms from South Africa. Trans. R. Soc. S. Afr. 33 (2) :
193-221, 3 figs, i pi.

1967. Echinoderms from the Red Sea. Part 2. Bull. Sea Fish Res. Stn Israel 41 : 26-

58, 5 figs-

CLARK, H. L. 1946. The Echinoderm fauna of Australia. Publs Carnegie Instn No. 566 : 567.
DONS, C. 1936. Zoologische Notizen. 26. Pseudoporania stormi n. gen., n. sp. K. norske

Vidensk. Selsk. Fork. 8 : 17-20, 4 figs.
1938. Zoologische Notizen. 35. Sphaeraster Berthae n. gen., n sp. 36. Sphaeraster

Bjorlykkei n. sp. K. norske Vidensk. Selsk. Fork. 10 : 161-164, figs I-6 ; 165-168, figs 1-6.
ECK, H. 1879. Bermerkungen zu den Mittheilungen des Herrn H. Pholig 'Aspidura ein meso-

izoisches Ophiuriden-genus' und iiber die Lagerstatte der Ophiuren im Muschelkalke.

Z. geol. Ges. 31 : 35-40.
FELL, H. B. 1963. The phylogeny of sea-stars. Phil. Trans. R. Soc. B246 : 381-435, 19

figs, 2 pis.
FISHER, W. K. 1908. Some necessary changes in the generic names of starfishes. Zool.

Anz. 33 : 356~359-

1911. Asteroidea of the North Pacific and adjacent waters, i. Phanerozonia and

Spinulosa. Bull. U.S. natn. Mus. 76 (i) : vi + 4i9, 122 pis.

1919. Starfishes of the Philippine Seas and adjacent waters. Bull. U.S. natn. Mus.

100(3) : 1-546, 156 pis.

1940. Asteroidea. Discovery Rep. 20 : 69-306, 23 pis.

GRAY, J. E. 1840. A synopsis of the genera and species of the class Hypostoma (Asterias,
Linn.). Ann. Mag. nat. Hist. 6 : 175-184 (Nov. ) ; 275-290 (Dec.).

1847. Descriptions of some new genera and species of Asteriadae. Proc. zool. Soc. Lond.

1847 : 72-83. (Also in : Ann. Mag. nat. Hist. 20 : 193-204.)

MADSEN, F. J. 1959. On a new North Atlantic sea-star, Chondraster hermanni n. sp., with some

remarks on related forms. Vidensk. Meddr dansk naturh. Foren. 121 : 153-160, 4 figs.
MORTENSEN, T. 1933- Echinoderms of South Africa (Asteroidea and Ophiuroidea) . Vidensk.

Meddr dansk naturh. Foren. 93 : 215-400, 91 figs, 12 pis.
MULLER, J. & TROSCHEL, F. H. i84oa [No title.] Mber. Akad. Wiss. Berl. 1840 : 100-106

(April).

i84ob. Ueber die Gattungen der Asterien. Arch. Naturgesch. 6 (i) : 318-326 (Sept. ?).

1842. System der Asteriden. xx+ 134 pp., 12 pis. Braunschweig.

PERRIER, E. 1875. Revision de la collection de Stelldrides du Museum d'Histoire Naturelle de

Paris. Paris. 384 pp. (Also in : Archs Zool. exp. gen. 4 (1875) : 265-450 ; 5 (1876) : i-

104, 209-309.)

1884. Memoire sur les Etoiles de Mer recueillies dans la Mer des Antilles et le Golfe de

Mexique durant les expeditions de dragage faites sous la direction de M. Alexandre Agassiz.
Nouv. Archs Mus. Hist. nat. Paris (2) 6 : 127-276, 10 pis.

1891. Echinodermes. i. Stellerides. Mission Sclent. Cap Horn 6. Zool. 1-198,

13 pis.

1894. Stellerides. Expeditions Scientifiques du Travailleur et du Talisman. 431 pp.,

26 pis. Paris.

268

F. H. C. HOTCHKISS AND A. M. CLARK

1889. Asteroidea. Rep. scient. Results Voy. Challenger. Zool. 30 11-935,
Early Palaeozoic starfish. Phil. Trans. R. Soc. 6235:87-129, 28

SLADEN, W. P.

118 pis.
SPENCER, W. K.

figs, 7 pis.
& WRIGHT, C. W. 1966. Asterozoans. In : Moore, R. C. (ed.) Treatise on invertebrate

paleontology. PtU. Echinodermata, 3 (i) Geological Society of America Inc. : University

of Kansas Press. U4-Uio7, 89 figs.
VERRILL, A. E. 1914. Monograph of the shallow-water starfishes of the North Pacific coast

from the Arctic Ocean to California. Harriman Alaska Ser. 14 : 1-408, no pis.
VIGUIER, C. 1879. Anatomic comparee du squelette des Stelle'rides. Archs Zool. exp. gin.

7 : 33-250, 12 pis.

FREDERICK H. C. HOTCHKISS

Division of Natural Sciences

STATE UNIVERSITY OF NEW YORK

COLLEGE AT PURCHASE

P.O. Box 337

PURCHASE,

NEW YORK 10577

U.S.A.

AILSA M. CLARK

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON SW7 530

PLATE i

Porania antarctica glabra Sladen. B.M. reg. no. 1948.3.16.408. Discovery Investigations
St. 366, South Sandwich Islands (x i).

Bull. By. Mus. nat. Hist. (Zool.) 30, 6

PLATE i

18

PLATE 2
Asteropsis carinifera (Lamarck). 1964.8.17.84. Red Sea (x i).

Bull. Br. Mus. nat. Hist. (Zool.) 30, 6

PLATE 2

PLATE 3
Petricia vernicina (Lamarck). 1963.4.19.6. Port Noarlunga, South Australia (x i).

Bull. Br. Mus. nat. Hist. (Zool.) 30, 6

PLATE 3

A LIST OF SUPPLEMENTS
TO THE ZOOLOGICAL SERIES

OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

1. KAY, E. ALISON. Marine Molluscs in the Cuming Collection British Museum
(Natural History) described by William Harper Pease. Pp. 96 ; 14 Plates.
1965. (Out of Print.)

2. WHITEHEAD, P. J. P. The Clupeoid Fishes described by Lacepede, Cuvier and
Valenciennes. Pp. 180 ; u Plates, 15 Text-figures. 1967. £4.80

3. TAYLOR, J. D., KENNEDY, W. J. & HALL, A. The Shell Structure and
Mineralogy of the Bivalvia. Introduction. Nuculacea-Trigonacea. Pp. 125 ;
29 Plates, 77 Text-figures. 1969. £5.40.

4. HAYNES, J. R. Cardigan Bay Recent Foraminifera (Cruises of the R.V. Antur)
1962-1964. Pp. 245 ; 33 Plates, 47 Text-figures. 1973. £11.90.

5. WHITEHEAD, P. J. P. The Clupeoid Fishes of the Guianas. Pp. 227 ; 72
Text-figures. 1973. £10.65.

6. GREENWOOD, P. H. The Cichlid Fishes of Lake Victoria, East Africa : the
Biology and Evolution of a Species Flock. Pp. 134 ; i Plate, 77 Text-figures.
1974. £3-75. Hardback edition £6.

Printed in Great Britain by John Wright and Sons Ltd. at The Stonebridge Press, Bristol BS4 5NU

2 6 NOV J976

LIBRARY

VARIATION IN EURASIAN

OF THE GENUS CROCIDURA
(INSECTIVORA : SORICIDAE)

P. D. JENKINS

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. 7

LONDON: 1976

VARIATION IN EURASIAN SHREWS OF THE
GENUS CROCIDURA (INSECTIVORA:SORICIDAE)

BY

PAULINA D. JENKINS

Pp 269-309 ; 14 Text-figures

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. 7

LONDON: 1976

THE BULLETIN OF THE BRITISH MUSEUM

(NATURAL HISTORY), instituted in 1949, is
issued in five series corresponding to the Scientific
Departments of the Museum, and an Historical series.

Parts will appear at irregular intervals as they
become ready. Volumes will contain about three or
four hundred pages, and will not necessarily be
completed within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 30 No. 7 of the Zoology series.
The abbreviated titles of periodicals cited follow those
of the World List of Scientific Periodicals.

World List abbreviation
Bull. Br. Mus. nat. Hist. (Zool.).

ISSN 0007-1498

Trustees of the British Museum (Natural History), 1976

BRITISH MUSEUM (NATURAL HISTORY)

Issued 25 November, 1976 Price {3.30

VARIATION IN EURASIAN SHREWS OF THE
GENUS CROCIDURA (INSECTIVORA:SORICIDAE)

By P. D. JENKINS

CONTENTS

Page

SYNOPSIS ............ 271

INTRODUCTION ........... 271

MATERIALS ........... 272

METHODS ............ 274

CANONICAL VARIATE ANALYSIS ........ 276

IDENTIFICATION .......... 279

VARIATION AND SYSTEMATICS ........ 282

C. fuliginosa ........... 282

C. horsfieldi ........... 285

C. suaveolens ........... 286

C. zarudnyi ........... 289

C. russula ........... 291

C. dsinezumi ........... 295

C. attenuata . . . . . . . . . . . 295

C. leucodon ........... 297

C. lasiura ........... 301

OTHER SPECIES RECORDED FROM THE AREAS OF STUDY BUT NOT INCLUDED

IN THE ANALYSIS .......... 3O2

ACKNOWLEDGEMENTS ......... 306

REFERENCES ........... 306

SYNOPSIS

An investigation including the use of a canonical variate analysis is made into the relationships
between various species of Crocidura occurring in Eurasia. The status of species and their
geographical variation are examined and keys are provided to aid segregation of species.

C. suaveolens is the most widespread species. C. zarudnyi is regarded as a distinct species
showing some affinity to C. suaveolens and to Asian C. russula. C. russula is divisible into two
main groups - Europe/ Algeria and western Asia. The latter group is close to some populations
of C. suaveolens. The European /Algerian group shows some association with western
Asian C. leucodon. The Japanese form, C. dsinezumi, while somewhat similar to some populations
of C. suaveolens and Asian C. russula, is sufficiently distinct to retain specific status. C. leucodon
is also divided into European and western Asian groups and the Iranian caspica is considered to
be a subspecies of the latter group. The affinities of the western Asian group of C. leucodon and
C. lasiura are discussed. C. fuliginosa trichura is assigned to C. attenuata. C. dracula is allocated
to C. fuliginosa and some of the forms recorded from Indonesia are considered to belong to
this species. C. horsfieldi is compared with C. suaveolens but retained as a distinct species.
Short notes are made on other species occurring in the area.

INTRODUCTION

WHITE-TOOTHED shrews of the genus Crocidura occur mainly southwards from
latitude 53°N in the Palaearctic, Oriental and Ethiopian regions. This report is
concerned with those species from the first two regions.

19*

272 P. D. JENKINS

Ellerman & Morrison-Scott (1966) recognized only fourteen species amongst the
forty four species names recorded from the Palaearctic and Indian regions. Chasen
(1940) listed twenty nine species names from the Malaysian subregion of the Oriental
region. It is possible that the same species occur under different names in different
areas, so comparison over the entire area was considered necessary.

New forms have frequently been described from single specimens but only rarely
figured and with inadequate descriptions. Information on colour of pelage, external
dimensions and a few skull measurements does not seem to provide adequate
diagnoses ; while diagnostic characters that vary from species to species, yet are
intraspecifically constant, are distinctly lacking in this genus.

This report describes geographical variation between populations occurring
throughout Eurasia. A canonical variate analysis has been used to investigate
relationships between populations. This type of analysis allows metrical characters
of a large number of samples to be handled simultaneously but is limited to those
specimens possessing all the necessary characters. The results have been correlated
with all other available information and some changes to the existing taxonomy are
suggested.

MATERIALS

The collections of the British Museum (Natural History) (BMNH) contain 740
specimens from Eurasia. All of these were examined and, for the sake of completeness,
specimens of North African C. russula and C. suaveolens (which are mainly distributed
in Europe and Asia) were included. In addition the following specimens from other
institutions have been examined : Harrison Zoological Museum, Sevenoaks : C.
russula - 28, C. russula caspica - 3, C. leucodon judaica - 2, C. lasia - i ; United
States National Museum : C. zarudnyi - 9, C. pergrisea - 2 ; Naturhistorisches
Museum, Vienna : C. pergrisea arispa - i.

C. olivieri, C. religiosa and C. floweri from Egypt have been excluded because they
are confined to Africa and are readily distinguished from Eurasian species. Since
the BMNH collections do not contain any of the eight species listed by Taylor (1934)
from the Philippine Islands these have also been excluded.

For the canonical variate analysis (CVA) the number of available specimens was
reduced to 526, since the method demanded specimens complete for all of the
characters included. In the CVA the group of specimens from each locality is
referred to by the term operational taxonomic unit (OTU) followed by the number
of the group (see Table i and Fig. i).

TABLE i

Populations entered in the canonical variate analysis

No. of
OTU Locality specimens Current species name

1 Isles of Scilly 10 C. suaveolens cassiteridum

2 Jersey and Sark 9 C. suaveolens

3 Guernsey and Alderney 15 C. russula peta

VARIATION IN EURASIAN CROCIDURA

273

TABLE i (Cont.)

No. of

OTU

Locality

Specimens

4

France

ii

5

France

J4

6

Spain

7

7

Portugal

8

8

Algeria

16

9

Algeria

4

10

Germany, Belgium and

17

Holland

ii

Germany

9

12

Switzerland

8

13

Switzerland

19

14

Switzerland

8

15

Sicily

5

16

Sardinia

9

17

Corsica

ii

18

Rumania and Hungary

5

19

Yugoslavia

15

20

Yugoslavia

21

21

Greece

8

22

Turkey

ii

23

Turkey

15

24

Lebanon

ii

25

Israel

16

26

Lebanon and Israel

7

27

Iran

7

28

Afghanistan

5

29

Pakistan

8

30

Russian Turkestan

18

31

Kashmir

6

32

Punjab

ii

33

Sri Lanka

J4

34

Darjeeling

M

35

Bhutan

7

36

Assam

20

37

Burma

8

38

Malaya

16

39

Borneo

7

40

Christmas Island

10

(io°3i'S. io5°33'E)

41

Yunnan

8

42

Yunnan

4

43

N. Vietnam

20

44

China

9

45

Korea

16

46

Korea

15

47

Japan

14

Current species name

C. russula russula
C. leucodon leucodon
C. russula pulchra
C. russula pulchra
C. russula agilis
C. suaveolens whitakeri
C. russula russula

C. leucodon leucodon

C. suaveolens mimula

C. russula russula

C. leucodon leucodon

C. sicula

C. russula ichnusae

C. russula cyrnensis

C. suaveolens mimula

C. suaveolens mimula

C. leucodon

C. suaveolens mimula

C. russula monacha

C. leucodon lasia

C. russula

C. russula

C. leucodon judaica

C. russula caspica

unnamed

C. zarudnyi

C. suaveolens ilensis

C. russula pullata

unnamed

C. horsfieldi

C. attenuata rubricosa

C. attenuata rubricosa

C. attenuata rubricosa

C. attenuata

C. malayana

C. fuliginosa baluensis

C. fuliginosa trichura

C. dracula

C. suaveolens

C. dracula

C. attenuata attenuata

C. suaveolens shantungensis

C. lasiura

C. russula dsinezumi

274

P. D. JENKINS

FIG. i. Distribution map of samples entered in the canonical variate analysis.

METHODS

Characters were taken exclusively from skins and skulls because there are very few
shrew skeletons in the BMNH collections, while the paucity and age of alcoholic
specimens preclude the use of anatomical characters such as those of the male
reproductive tract which have been used successfully by other authors (Vinogradov,
1958 ; Martin, 1967). The following characters were examined with the abbrevia-
tions used in the text in brackets.

Head and body length (HB)

Tail length (TL)

Hindfoot length

Ear length

Weight

Colour of pelage

Condylobasal length (CBL)

Length of upper toothrow (UTL)

Combined length of upper incisor and uni-

cuspids

Lingual length of upper unicuspids
Labial length of upper unicuspids at cingula

(LLU)

Width across maxillae at level of second

molar (MB)
Height of rostrum
Lacrymal breadth
Interorbital breadth (IB)
Postglenoid breadth (PB)
Braincase breadth (BB)
Height of braincase
Length of mandible (excluding incisor)
Height of mandible at coronoid process
Length of mandibular toothrow (excluding

incisor)
Shape of teeth

VARIATION IN EURASIAN CROCIDURA 275

Skin measurements were taken from collectors' notes on the specimen labels,
which are notoriously fallible because they are difficult to measure accurately and
consistently and the measuring technique varies from one collector to another.
However, skin measurements are still useful as long as these disadvantages are
taken into account and since they have been extensively used in the taxonomy of
this group their inclusion was considered necessary. The effect of using skin
measurements was checked by running the CVA twice, first including head and body,
tail and hindfoot length and secondly omitting these three characters. The results
were found to be comparable so only the former analysis is described here.

The toothrow and anterior part of the skull are less frequently damaged than the
posterior. Since the CVA requires possession of all characters by each specimen
either the number of specimens or number of characters had to be reduced ; the
latter course was considered preferable. The excluded characters were examined
and incorporated with the results of the CVA at a later stage. The following
characters were selected for the CVA and found to show little correlation with each
other (see Table 2) :

1. Length of upper toothrow ;

2. Combined length of upper incisor and unicuspids ;

3. Lingual length of upper unicuspids ;

4. Labial length of upper unicuspids at cingula ;

5. Width across maxillae at level of second molar ;

6. Head and body length ;

7. Tail length ;

8. Hindfoot length.

TABLE 2
Correlation matrix of characters used in the canonical variate analysis

Character No.

Length of upper toothrow (i) i-oo

Combined length of upper incisor (2) 0-43 i-oo

and unicuspids

Lingual length of upper unicuspids (3) 0-60 0-40 i-oo

Labial length of upper unicuspids (4) 0-55 0-38 0-60 i-oo

at cingula

Width of maxillae at level of second (5) 0-50 0-22 0-34 0-32 i-oo

molar

Head and body length (6) 0-20 0-09 0-12 0-16 0-27 i-oo

Tail length (7) 0-35 0-08 0-28 0-27 0-24 0-32 i-oo

Hindfoot length (8) 0-33 0-15 0-24 0-24 0-29 0-30 0-37 i-oo

(i) (2) (3) (4) (5) (6) (7) (8)

The CVA programs used were part of the CLASP package provided by J. C. Gower,
Statistics Department, Rothamsted Experimental Station. The data, which may
be thought of as a 'plot' in eight dimensions, are translated into visible form by the
following treatment. The CVA program seeks a plane in this eight-dimensional
space which, when the group means are projected on to the plane, best reproduces

276 P. D. JENKINS

the Mahalanobis' distances between groups. The Mahalanobis' generalized distances
(D2) are akin to phenetic distances but take some account of intraspecific variation
and correlation of characters (see Sneath & Sokal, 1973). If the variation between
the projected group means is almost as large as the total variation in eight dimensions
then the representation on the plane will be a good one. In the first analysis (a) the
variation on the plane was 84% of the total and therefore the plot of group means on
this plane gives a fairly accurate overall picture of the relationships between them.

The following relevant information is produced by the program : (i) Correlation
matrix between characters for all specimens. (2) Distance matrix (D2) or
Mahalanobis' distances which are an index of dissimilarity between OTUs (see
Table 8).

CANONICAL VARIATE ANALYSIS

(a) FIRST ANALYSIS OF ALL OTUs

The results of the first analysis are summarized in Fig. 2, in which OTUs are
plotted on a plane which accounts for 84% of the variation, suggesting a reasonable
representation of the actual relationships. Circles are calculated to include 90% of
the individual specimens that combine to give the group mean, while the boundaries
of currently recognized species groups are indicated by lines of differing texture (see
key on figure).

OTUs Malaya (38), Borneo (39), Yunnan (41) and N. Vietnam (40) overlap but
are well separated from all others and may be regarded as a distinct species, herein-
after referred to as C. fuliginosa. The remaining groups are disposed in a Y-shaped
pattern, those at the extreme of each limb being clearly distinct.

(1) Top right - Darjeeling (34), Bhutan (35), Assam (36), China (44) correspond
with C. attenuata, and Christmas Island specimens (40), C. fuliginosa trichura, are
associated with this group. Burma (37) is also usually thought to belong to this
species and is positioned closest to it despite showing no overlap at the 90% level.

(2) Lower centre - Isles of Scilly (i), Jersey and Sark (2), Rumania and Hungary
(18), Yugoslavia (19), Greece (21), Russian Turkestan (30), Yunnan (42) and Korea
(45) correspond with the classic concept of C. suaveolens. Showing overlap with this
group is Pakistan (29) currently regarded as a distinct species, C. zarudnyi. Sri
Lanka (33), C. horsfieldi, lies close to this group but does not overlap with it. C.
dsinezumi from Japan (47) overlaps with C. suaveolens, C. russula and C. zarudnyi.

(3) The remainder form a central group extending to the top left which corresponds
with the russula-leucodon complex, with leucodon falling at the extreme left. Those
in the upper portion of this group include specimens from Turkey (23) and also
Israel and Lebanon (26) and Iran (27) whose affinities are doubtful. The Korean
OTU (46), C. lasiura, overlaps with the group at the extreme upper left.

(b) SECOND ANALYSIS OF 42 OTUs (5 OMITTED)

The analysis was repeated but five OTUs were omitted to improve resolution
(see Fig. 3). Those excluded were Malaya (38), Borneo (39), Yunnan (41) and

VARIATION IN EURASIAN CROCIDURA

277

FIG. 2. First analysis of all OTUs.

C.fuliginosa - upper centre of figure ; C. russula - central group (stippled) ; C. leucodon -
extreme left of central group ; C. attenuata - extreme right of central group ; C.
suaveolens - lower centre of central group ; other species currently regarded as distinct -
broken circles.

P. D. JENKINS

-5-

— T

1 1 1 1 1 1 T

0 5

FIG. 3. Second analysis of 42 OTUs.

C. russula - central group (stippled) ; C. leucodon - upper left of central group ; C.
attenuata - lower right of central group ; C. suaveolens - lower left of central group ; other
species currently regarded as distinct - broken circles.

VARIATION IN EURASIAN CROCIDURA 279

N. Vietnam (42) which form a distinct group, C. fuliginosa, and also Burma (37)
which is separated from any other OTU.

The plane in this analysis includes 82% of the variation. The species groups still
overlap but to a lesser extent.

(1) Lower right - OTUs Darjeeling (34), Bhutan (35), Assam (36), China (44) form
the loosely knit C. attenuata, with which Christmas Island (40) is associated.

(2) Lower left - OTUs overlapping in the suaveolens group remain as before.
C. zarudnyi overlaps with this group and C. dsinezumi overlaps with both groups and
also with C. russula.

(3) The russula- leucodon complex is better spaced in this figure. C. russula falls
roughly into two subgroups with most European OTUs in the upper part of the
figure and Asian ones occurring lower down. C. leucodon lies at the top left.

IDENTIFICATION

All available information including that from the previous section is assembled
here and correlated with the existing taxonomic groupings. The species recognized
are :

Crocidura fuliginosa - including malayana, baluensis and dracula.

Crocidura horsfieldi

Crocidura suaveolens

Crocidura zarudnyi

Crocidura russula - two groups

Crocidura dsinezumi

Crocidura attenuata - including C. fuliginosa trichura

Crocidura leucodon - two groups

Crocidura lasiura

In any one locality where two or more species occur they are usually easy to
separate. An exception to this is found in Switzerland where three species occur.
C. suaveolens is readily distinguished but C. russula and C. leucodon are only clearly
separable on the basis of pelage colour.

In the majority of species some populations show overlap with at least one other
species and therefore an exclusive diagnosis covering the entire region is impossible.
Even when the range of the genus is divided into 'European', 'eastern Asian' and an
intermediate zone from Turkey east to Sri Lanka, the construction of keys presents
considerable difficulties. For example, the separation of fuliginosa from other
species of Crocidura depends mainly on skull length. Although the key is somewhat
unsatisfactory because of the overlap in size, the degree of separation is adequate in
most cases, as may be seen in Fig. 4. A similar situation obtains in other examples
of overlap.

Thus while the three keys given below are not a completely accurate identification
guide, they may be used to discriminate with a reasonable degree of confidence
between the species from a region, particularly when a series of specimens is available.

280 P. D. JENKINS

C.suaveolens

C.horslieldi

C.dsinezumi

C.russula

C.attenuata

C.lasiura

C.fuliginosa

15-0 2C.O -o

condylobasal length

FIG. 4. Condylobasal length in species of Crocidura from eastern Asia, showing mean
(vertical bar), range (horizontal line), one and two standard deviations (rectangles).

Confirmation of identification should be sought by reference to the additional data
provided in the text.

1. KEY TO THE EUROPEAN SPECIES OF CROCIDURA

1 Bicoloured, pale ventral surface sharply differentiated from dark dorsum along

flanks . ......... C. leucodon

- No sharp differentiation between dorsal and ventral colour ..... 2

2 Average smaller, labial length of upper unicuspids up to 2-0 mm . . C. suaveolens

- Average larger, labial length of upper unicuspids usually over 2-0 mm . C. russula

Note, (a) Where C. suaveolens and russula are sympatric, they may be distinguished on the basis of
size and usually by differences in the shape of the large upper premolar.

(b) C. russula from Corsica may be distinguished from those from elsewhere in Europe by the dis-
tinctive shape of the large upper premolar (see text and Fig. 5), and from C. suaveolens only by a com-
bination of several characters. However, C. suaveolens is not recorded from Corsica.

2. KEY TO THE SPECIES OF CROCIDURA OCCURRING FROM TURKEY IN THE WEST TO SRI LANKA

IN THE EAST

i Bicoloured, pale ventral colour differentiated from dark dorsal colour along flanks or,
when unicoloured, size large (upper toothrow over 8-8 mm, width across maxillae
at M2 over 6-0 mm) ......... C. leucodon

VARIATION IN EURASIAN CROCIDURA

281

(a)

(b)

(c)

(d)

FIG. 5. Large upper premolar of European species of Crocidura. (a) C. suaveolens -
Austria ; (b) C. russula - Corsica ; (c) C. russula - France ; (d) C. leucodon - France.

No sharp differentiation between dorsal and ventral colour ; smaller (upper toothrow

length less than 8-8 mm, width across maxillae at M2 less than 6-0 mm . . 2

2 Lingual region of large upper premolar relatively narrow and lobed (see Fig. 6) ;

labial length of upper unicuspids greater than 1-8 mm . . . C. russula

Premolar not so shaped, labial length of upper unicuspids less than i -9 mm . . 3

3 Sri Lanka. Skull small and narrow, width across maxillae at M? less than 5-1 mm.

Ratio of tail length to upper toothrow length 6-5-7-5 : i . . . C. horsfieldi
- Other localities. Width of maxillae at M2 greater than 5-0 mm. Ratio of tail length
to upper toothrow length 3-4-5-2 : i (suaveolens) or if greater (4-8-7-1 : i) colour
of pelage very pale (zarudnyi) ........ C. suaveolens

C. zarudnyi

Note, (a) The separation of C. russula and the unicoloured C. leucodon caspica from Iran is not complete
as there is a small overlap in size of the characters used. However, most specimens seen can be identified
by the key above.

(b) C. suaveolens and C. zarudnyi are difficult to distinguish, see discussion in text under C. zarudnyi.

FIG. 6. Large upper premolar of species of Crocidura occurring from Turkey in the west to
Sri Lanka in the east, (a) C. suaveolens - Russian Turkestan ; (b) C. russula - Turkey ;
(c) C. leucodon - Turkey ; (d) C. zarudnyi - Pakistan ; (e) C. horsfieldi - Sri Lanka.

282

3-

i

P. D. JENKINS

KEY TO EASTERN ASIAN SPECIES OF CROCID URA
Large, condylobasal length over 21 -o mm ...... C.fuliginosa

- Smaller, condylobasal length less than 21-6 mm, where greater than 21-0 mm, tail

48-58% of head and body length (lasiura) or 59-92% (fuliginosa) . Only one
specimen of attenuata has a condylobasal length greater than 21-0 mm - see Fig. 4
and text .............. 2

2 Upper toothrow length 9-0-10-2 mm, tail length 34-49 mm, 48-58% of head and

body length, ear margins densely furred . . . . . . . C. lasiura

- Upper toothrow length 7-8-9-8 mm, tail length 50-88 mm, 69-108% of head and

body length, ear margins not densely furred ..... C. attenuata

- Upper toothrow length 7-4-8-8 mm, tail length 37-51 mm, 54-76% of head and body

length, ear margins not densely furred ........ 3

3 Width of braincase over 8-8 mm ....... C. russula

- Width of braincase usually less than 8-8 mm . . . . . . . 4

4 Japanese, width of braincase 8-1-8-8 mm, condylobasal length over 16-8 mm C. dsinezumi

- Elsewhere, width of braincase less than 8-5 mm (if from Tsushima Islands or Korea,

width of braincase less than 8-2 mm and condylobasal length less than 16-9 mm) 5

5 Ratio of tail length to upper toothrow length less than 5-8 : i . C. suaveolens

- Ratio of tail length to upper toothrow length over 5-7:1 . . . C. horsfleldi

Note, (a) Separation of C. russula from C. attenuata depends on a combination of characters, see text,
(b) It is possible to distinguish C. dsinezumi from C. suaveolens occurring in nearby areas, see text.
C. dsinezumi is geographically separated from C. russula and C. horsfieldi and a combination of characters
serve to distinguish these species.

(e)

FIG. 7. Large upper premolar of Crocidura from eastern Asia, (a) C. suaveolens - Korea ;
(b) C. russula - Punjab ; (c) C. attenuata - Assam ; (d) C. lasiura - Korea ; (e) C.
dsinezumi - Japan ; (f) C. fuliginosa - N. Vietnam.

VARIATION AND SYSTEMATICS

Crocidura fuliginosa (Blyth, 1855)

This is the only group that is completely distinct from all others. It includes
three forms previously regarded as separate species but here associated with C.

VARIATION IN EURASIAN CROCIDURA

283

fuliginosa from Burma. These are C. malayana from Malaya (38), C. baluensis from
Borneo (39) and C. dracula from Yunnan (41) and N. Vietnam (43).

Populations from Yunnan and N. Vietnam are very similar to each other and are
currently known as C. dracula. They have a D2 value of 1-5 which is very low and
reflects their conspecificity. These populations resemble specimens from Borneo
and Malaya which, with a D2 value between them of 2-69, are also very similar. The
range of D2 values within the group goes up to 4-3 but this is still fairly low when the
geographical distance between populations and ecological barriers are taken into
consideration, see distribution map (Fig. 8).

FIG. 8. Distribution of C. fuliginosa. Broken lines - exact distribution uncertain.

The results of the CVA may be correlated with characters not employed in the
numerical analysis. This group contains some of the largest specimens in the genus.
Those from Borneo are the largest while the greatest size range occurs in Malaya.
Populations from Yunnan and N. Vietnam, are more similar in appearance and size
to each other than to populations from Malaya or Borneo. Size range is given in
Table 6.

A large number of distinct species have been described from many of the separate
islands off the Malay Peninsula. The BMNH houses most of the holotypes of these,
which have been compared with mainland specimens and those from Borneo. The

284 P. D. JENKINS

holotypes fall within the range of this group. Some of these island forms may not
even be distinct subspecies but no definite conclusions can be based on the few
specimens available.

Unfortunately I have not seen the holotype of C. fuliginosa which was collected
from Schwe Gyen, Pegu, Burma, and is lodged in the Calcutta Museum. Banks
(1931) recorded it as the commonest lowland shrew in Borneo and Chasen (1940)
accepted a brown form in the lowlands of Malaya as fuliginosa and a darker form as
malayana.

The name C. fuliginosa is the subject of some confusion. Blyth's (1855) original
description is rather scanty and there is some doubt whether the subsequent
description by Anderson (1881) refers to the same specimen, especially as different
terminology is used for measurements. The statement by Lindsay (1929) '. . . . the
specific name of fuliginosus which Blyth plainly applied to a Suncus' is erroneous
since Suncus is defined as possessing four upper unicuspids. Medway (1965) has
attempted to elucidate the situation but there is still some doubt about the identity
of the holotype, particularly as some label mixing may have occurred in the past.

This report confirms the view of Medway (1965) who allocated Malayan specimens
of malayana and Bornean foetida, kelabit and baluensis to C. fuliginosa. This name
is also applicable to the large specimens from the offshore islands of Malaysia as well
as to specimens of dracula from Yunnan and N. Vietnam.

C. fuliginosa is recorded from S. Vietnam by Van Peenan, Ryan & Light (1969)
which suggests a fairly continuous distribution from Yunnan in the north, lower
Burma, N. and S. Vietnam to the Malay Peninsula and Borneo (see Fig. 8). A
limited number of specimens indicate that this species is also present on the islands
of Sumatra, Java and Sulawesi.

The following subspecific groups are suggested in addition to C. f. fuliginosa from
Burma :

Crocidura fuliginosa malayana Robinson and Kloss, 1911. Mainland of Malaya.

Probable synonyms of C./. malayana from the offshore islands of the Malay Peninsula
are :

Crocidura aoris Robinson, 1912. Aor Island.

Crocidura klossii Robinson, 1912. Great Redang Island.

Crocidura negligens Robinson & Kloss, 1914. Samui Island.

Crocidura maporensis Robinson & Kloss, 1916. Mapor Island.

Crocidura gravida Kloss, 1917. Dayang Bunting Island.

Crocidura tionis Kloss, 1917. Tioman Island.

Crocidura aagaardi Kloss, 1917. Bang Nara, Patani, southern Thailand.

Crocidura fuliginosa baluensis Thomas, 1898. Borneo.

Probable synonyms are :

Crocidura foetida Peters, 1870. Borneo. This name would predate C. f. baluensis as the

correct subspecific name.
Crocidura fuliginosa kelabit Medway, 1963. Bario, Baram District, Sarawak.

VARIATION IN EURASIAN CROCIDURA 285

Crocidura fuliginosa dracula Thomas, 1912. Yunnan.

Crocidura praedax Thomas, 1923. Likiang Valley, Central Yunnan.

Crocidura grisescens Howell, 1928. Kuatun, Fokien, South-eastern China (listed as a synonym

of C. dracula by Ellerman & Morrison- Scott, 1966).
Crocidura dracula mansumensis Carter, 1942. Mansum, northern Burma.

Other species occurring in the same area as C. fuliginosa, but readily distinguished
by their small size, are C. suaveolens in Yunnan and N. Vietnam (condylobasal length
15-2-17-5 mm) and C. monticola in Borneo (condylobasal length of one specimen
only 15-2 mm). Condylobasal length of C. fuliginosa is over 21.0 mm. C. attenuata
is also sympatric at least in the northern part of the range but is usually distinguish-
able on the basis of small size. While there are no members of both species from the
same area in the BMNH collections, specimens from relatively close areas, such as
fuliginosa from Yunnan and N. Vietnam and Chinese attenuata, are distinct. Overlap
in size occurs in some large specimens of attenuata from Assam and Burma and a few
small fuliginosa from Malaya.

Crocidura horsfieldi (Tomes, 1856)

Only one OTU of this species, Sri Lanka (33), was entered in the CVA. It is rather
similar to C. suaveolens but can be distinguished by the following features : the skull is
relatively short and narrow and in these proportions it is most similar to populations
of C. suaveolens from Tsushima Island (Japan), N. Vietnam and Korea but consider-
ably smaller than the remaining populations (see Table 3). The tail is proportion-
ately much longer than in C. suaveolens, the ratio of tail length to upper toothrow
length ranging from 3-4 to 6-1 : i in the latter and from 6-5 to 7-5 : I in C. horsfieldi.
The upper premolar is small and comparatively narrow (see Fig. 6) .

The distribution of the nominate form and the other subspecies of C. horsfieldi
listed by Ellerman & Morrison-Scott (1966) is given in Fig. 9. Of these the holotype
of C. h. indochinensis, in the BMNH collections, agrees very well with specimens of
C. horsfieldi from Sri Lanka and there seems little doubt that it is correctly assigned
to this species. There is also a single specimen of C. watasei and this too is very
similar to Sri Lanka specimens ; Imaizumi (1970) gives the Japanese distribution
of this subspecies. Two skulls from Taiwan are larger than Sri Lanka specimens
(see Table 3). Another subspecies, C. h. wuchihensis, described from Hainan, differs
slightly from specimens from Sri Lanka (condylobasal length 15-5-15-7 mm, skull
width 7-8 mm, tail 67% of head and body length).

The subspecies recorded are :

C. h. horsfieldi (Tomes, 1856). Sri Lanka.

C. h. indochinensis Robinson and Kloss, 1922. Dalat, S. Annam.

C. h. watasei Kuroda, 1924. Liukiu Islands.

C. h. tadae Tokuda and Kano, 1936. Koto-sho, island E. of Taiwan.

C. h. wuchihensis Wang, 1966. Mt Wuchih, Hainan Island.

The only other species recorded from Sri Lanka is C. miya Phillips, 1929 which is a
much larger shrew, condylobasal length of holotype 19-7 mm, upper toothrow

286

P. D. JENKINS

length 8-7 mm, braincase damaged but width across maxilla at second molar
5-9 mm (see p. 302).

TABLE 3

Size range of C. horsfieldi, C. suaveolens, C. russula, C. zarudnyi and C. dsinezumi

C. horsfieldi Sri Lanka

Taiwan

C suaveolens Korea,
Tsushima, N. Vietnam
Europe, Algeria, Russian
Turkestan, Yunnan

C. russula Asia

Europe

C . zarudnyi Pakistan,
Punjab, Turkey, Iran
C. dsinezumi Japan

Upper

Width of

Condylobasal

toothrow

maxillae at M2,

Braincase

length, mm

length, mm

mm

breadth, mm

16-2-17-2

6-8-7-4

4-8-5-1

7-4-7-8

x 16-60

X7-05

x 5-00

x 7-60

17-8

7-5 and 8-0

5-5

8-5

15-2-16-9

6-7-7-7

4-7-5-3

7-0-8-2

x 16-03

f 7-17

x 4-98

X7-74

15-7-18-1

6-9-8-4

4-8-5-8

7-7-8-8

x 16-89

x 7-59

x 5-28

x 8-1

I7-5-I9-5

7-6-8-8

5-2-6-0

8-4-9-4

x 18-39

x 8-28

X5-65

x 8-91

17-5-20-1

7-6-9-2

5-4-6-5

8-5-9-5

x 18-79

x 8-48

x5-89

x8-97

16-1-18-8

7-1-8-3

S-o-5-5

7-8-8-5

x 17-69

X7-78

X5-3Q

x8-25

16-8-18-4

7-4-8-1

5-1-5-9

8-1-8-9

x 16-03

x 7-82

X5-5Q

x8-53

Crocidura suaveolens (Pallas, 1811)

The following OTUs are included in this species : Isles of Scilly (i), Jersey and
Sark (2), Algeria (9), Switzerland (12), Rumania and Hungary (18), Yugoslavia (19),
Greece (21), Russian Turkestan (30), Yunnan (42) and Korea (45).

The suaveolens group has a very large distributional range but geographical
variation is not very pronounced and most Mahalanobis' distances are corre-
spondingly low, varying from 2 to 3. The European group of Switzerland (12),
Rumania and Hungary (18), Yugoslavia (19) and Greece (21) have low D2 values of
2'03-2-87.

The Algerian, Russian Turkestan and Yunnan populations are similar in size to
European ones ; the remaining Asian populations from Korea, Tsushima (Japan)
and N. Vietnam are on average smaller, see Table 3.

In this species there is rather more colour variation than in most of the others.
Most are greyish-brown often with a slightly paler venter but certain populations,
notably those occurring in arid habitats such as Russian Turkestan, Algeria and
Israel, have a strikingly pale grey or fawn dorsum and a still paler venter.

Another character which is probably ecologically linked is the densely furred ear
margins which distinguish Korean specimens (though nearby specimens from
Tsushima, Japan, lack this character) and also specimens from Russian Turkestan,
especially those from Semiretchensk. In Korea this character may be subject to a
certain amount of seasonal variation ; the ears of specimens in summer pelage in

VARIATION IN EURASIAN CROCIDURA

287

C.suaveolens

FIG. 9. Distribution of C. suaveolens, C. zarudnyi and C. horsfieldi.

distribution uncertain.

Broken lines - exact

August, September and early October are only slightly haired, while by late
November, December and January specimens have winter pelage and densely furred
ears.

Geographical proximity does not always correspond with low values of D2,
especially where ecological barriers are involved. For example, island populations
such as the Isles of Scilly (i) and Jersey and Sark (2) are relatively close to each
other but the D2 value is 3-15. Specimens from the Isles of Scilly are most similar
to those from Yugoslavia (19), with a D2 value of 1-96, while those from Jersey and
Sark are most similar to Pakistan (29) specimens of C. zarudnyi with a D2 value of
2-94. Delany & Healy (1966), in their study of Crocidura of the Channel Islands and
the Isles of Scilly, also found that populations from Jersey and Sark can be distin-
guished from each other and that Channel Island populations are very distinct from
those from the Isles of Scilly.

In summary, European mainland populations are generally more homogeneous
than the island populations, the Algerian or the Asian populations which are
sampled from well-separated areas. However, despite this geographical variation
there is little doubt that the populations are truly representatives of one very
widespread species probably with an almost continuous distribution over its range
(see Fig. 9).

288

P. D. JENKINS

Of the subspecies listed from the Palaearctic and Indian region by Ellerman &
Morrison-Scott (1966), the U.S.S.R. by Gureev (1971), China by Allen (1938) (listed
under C. ilensis) and Japan by Imaizumi (1970), the following have been examined :

C. s. mimula Miller, igoib. Western Europe (12, 18, 19, 21).

C. s. cassiteridum Hinton, 1924. Isles of Scilly (i).

C. s. whitakeri de Winton, 1897. Morocco, Algeria (9).

C. s. ilensis Miller, igoic. Russian Turkestan to Mongolia (30).

C. s. shantungensis Miller, I90ic. Korea, Tsushima Islands (Japan), China (45).

Also samples from Jersey and Sark (2) and Yunnan (42).

Mahalanobis' distances between some populations of C. suaveolens (Isles of
Scilly (i), Yugoslavia (19), Greece (21), Russian Turkestan (30) and Yunnan (42))

9-0-

8-0-

x

X X

X

X

O

9 O O
O

T

Y Y

O

80 9-0

upper toothrow length

FIG. 10. Comparison of skull size.

C. russula (Asia and Corsica) • ; C. suaveolens (Isles of Scilly, Greece, Russian Turkestan,
Yunnan) T ; C. dsinezumi x ; C. zarudnyi O-

VARIATION IN EURASIAN CROCIDURA 289

and others of the Asian subgroup of russula (Lebanon (24), Israel (25), Afghanistan
(28) and Punjab (32) and also Corsica (17)) range from 2-25 to 5-48. Some of these
are fairly low but since the sample populations are geographically well separated
their similarity is unlikely to reflect any true relationship. Yunnan (42) , for example,
has D2 values under 3-0 with Afghanistan (28), Corsica (17), Punjab (32), Lebanon (24)
and Israel (25) ; Afghanistan has D2 values under 3.0 with the Isles of Scilly (i),
Russian Turkestan (30) and Yunnan (42).

Compared with Asian and Corsican russula these populations of suaveolens have
somewhat smaller skulls (see Fig. 10 and Table 3) and shorter tails (TL 27-48 mm,
x 35-3 mm ; in the Asian and Corsican russula TL 36-52 mm, x 44-79 mm).

Richter (1970) decided to treat the entire species of C. suaveolens as a third sub-
specific group of C. russula, the other two being russula and gueldenstaedti. He
postulated that the suaveolens subspecies group is merely a 'steppe form' of the
gueldenstaedti subspecies group and that suaveolens has only secondarily come to
inhabit the area occupied by the russula subspecies group in western Europe. While
it is true that some populations of suaveolens mentioned above are similar to some
members of Richter's gueldenstaedti subspecies group, it is possible to distinguish the
two species where they occur together. Furthermore, there is usually no difficulty
in distinguishing suaveolens from the subspecies that Richter places in the russula
subspecies group, especially where they are sympatric, so C. suaveolens should be
treated as a separate species.

Crocidura zarudnyi Ognev, 1928
Crocidura pergrisea Miller, 1913

One OTU of C. zarudnyi from Pakistan (29) was entered in the CVA but several
specimens from other localities were also examined. Because only two specimens of
C. pergrisea were seen, they were not placed in the CVA but they have been
examined and are briefly commented on here.

Specimens from Pakistan are currently regarded either as a distinct species, C.
zarudnyi Ognev, 1928, type locality Baluchistan border (see Hassinger, 1970 ;
Spitzenberger, 1971), or as a subspecies of C. pergrisea Miller, 1913 from Baltistan
(see Ellerman & Morrison-Scott, 1966). The Pakistan (29) specimens are geo-
graphically fairly close to specimens of C. suaveolens from Russian Turkestan (30)
and to Asian russula from Afghanistan (28) and Punjab (32), although they are
probably ecologically isolated by the rocky, desert habitat and the surrounding
mountainous terrain. The D2 values between Pakistan (29) and Punjab (32) and
Israel (25) are fairly low, 3-5, but the Pakistan specimens are most similar to those
of C. suaveolens from Jersey and Sark (2) with a D2 value of 2-94. Despite the
similarity to various populations of C. suaveolens from Greece, Russian Turkestan,
Yunnan, the Channel Islands and the Isles of Scilly, the only relevant comparison is
with the Russian Turkestan specimens of C. suaveolens which have shorter tails on
average (TL 27-40 mm, x 32-2 mm ; in C. zarudnyi 36-57 mm, x 49-2 mm). The
shape of the upper premolar is similar to that of specimens of Asian C. russula,
though it is slightly smaller (see Fig. 6).

2QO

P. D. JENKINS

Miller (1913) described four specimens from Skoro Loomba, Shigar, Baltistan
(2900 m), giving them the name C. pergrisea. Ognev (1928) first used the name
C. zarudnyi for a new species which he had formerly described under a preoccupied
name (1921). This new species, represented by one specimen, was collected in East
Iran on the Baluchistan border.

Ellerman & Morrison-Scott (1966) examined three specimens from Baluchistan
which they believed to be zarudnyi and placed this as a subspecies of pergrisea.
Lay (1967) allocated to C. pergrisea six specimens from Iran, which Hassinger (1970)
re-identified as C. zarudnyi. Hassinger made a preliminary study of the C. zarudnyi-
pergrisea group and concluded that they were distinct species. He also described a
new subspecies, C. zarudnyi streetorum, from Afghanistan.

Richter (1970) tentatively assigned both zarudnyi and pergrisea to the subspecies
group gueldenstaedti of C. russula. Spitzenberger (1971) gave measurements of three
subspecies of C. pergrisea (the nominate form C. p. pergrisea, C. p. serezkeyensis
Laptev, 1929 from the Pamirs and a new subspecies from Turkey, C. p. arispd) as
well as C. zarudnyi and distinguished them from both C. russula and C. suaveolens by
the light colour of the pelage and the relatively long tail. Gureev (1971) also
recorded C. zarudnyi as a separate species from C. pergrisea. Hassinger (1973)
amplified his previous discussion by comparing Afghan specimens with other species,
still concluding that C. pergrisea and C. zarudnyi are two separate species. He
distinguished C. zarudnyi from C. russula by its smaller size and from both C. russula
and C. suaveolens by the relatively long tail and pale coloration. The habitat of
C. zarudnyi in Afghanistan is reported to be drier than that of either C. russula or
C. suaveolens.

TABLE 4
Comparison of various samples of C. zarudnyi with C. pergrisea

C. zarudnyi

holotype 18-8

C. zarudnyi

Pakistan 16-1-18-7

n = ii x 17-6

C. zarudnyi

Iran n = 2 !7'4» 18-7

C. zarudnyi

Punjab n = i
C. pergrisea arispa

holotype Turkey 17-7

C. p. pergrisea

Baltistan original series 19-0- 19-4
C. p. pergrisea

Baltistan n = 2 19-2

Upper

Condylobasal toothrow
length length

mm mm

8-3

7-15-8-2
x 7-8

7-8, 7-9
7'5
7'5
8-6
8-4

Source

Length of when not
Width of mandible- measured
braincase excluding Ix personally

8-3 Ognev (1928)

8-7-9-2
x 9-0

8-0, 8-3 9-0, 9-2

8-9

8-25 9-0

8-8-8-9 Miller (1913)

8-8 9-6, 9-7

VARIATION IN EURASIAN CROCIDURA 291

Table 4 shows that Baltistan specimens of C. p. pergrisea (two specimens measured
personally plus measurements from the original description) are distinctly larger in
most characters than all populations of C. zarudnyi. On these criteria the holotype
of C. p. arispa from Turkey should be placed in the latter group as C. zarudnyi arispa.
Although the specimens from Baltistan might be considered as a montane form of
zarudnyi and, despite the fact that three specimens are not an adequate sample from
which to draw conclusions, they are probably best regarded as examples of a distinct
species, C. pergrisea.

Although rather similar to the Asian subgroup of C. russula, specimens of C.
zarudnyi may usually be separated on size (see Fig. 10 and Table 3) as well as colour
and labial length of upper unicuspids (1-45-1-9 mm, x 1-77 mm in C. zarudnyi ;
1-8-2-3 mm, x 2>I6 mm in Asian C. russula). However, separation from C. suaveolens
is not so easy, the only useful characters being the pale pelage colour of C. zarudnyi
which is paralleled by several populations of C. suaveolens from Russian Turkestan,
Algeria, Israel and also specimens from Central Gobi, Mongolia, described by Allen
(1938) under the name of C. ilensis lar (which Ellerman & Morrison-Scott (1966)
believed to be a subspecies of suaveolens}. This character may reflect adaptation to
local environment rather than phylogenetic affinity. The other character is the
relatively long tail which is also not an exclusive character (ratio of TL : UTL
3-4-5-2 : i in C. suaveolens ; 4-8-7-1 : I in C. zarudnyi). Both C. suaveolens and
C. zarudnyi have been recorded from Iran (Lay, 1967), Turkey (Spitzenberger, 1970,
1971) and Afghanistan (Hassinger, 1970, 1973) but not from the same habitats, and
while the mean skull size is usually greater for specimens of C. zarudnyi than C.
suaveolens the only distinct character is the greater tail length of the former. The
evidence suggests that C. zarudnyi is fairly closely related to C. suaveolens and may
be simply an ecological form of it, but for the present it is retained as a distinct
species.

Crocidura russula (Hermann, 1780)

The OTUs that belong to this species fall into two groups roughly equivalent to a
geographical division into European and Asian forms, except that the Corsican (17)
population appears to belong to the latter group and Algeria to the former. Apart
from these anomalous OTUs the European group consists of Guernsey and Alderney
(3), France (4), Spain (6), Portugal (7), Germany, Belgium and Holland (10),
Switzerland (13), Sicily (15), Sardinia (16) and also Algeria (8) ; the Asian group
includes Turkey (22), Lebanon (24), Israel (25), Afghanistan (28), Kashmir (31) and
Punjab (32).

Fig. ii gives the distribution of this species and shows the wide geographical
separation of the two subgroups. The main characters for distinguishing these
groups are the differences in relative tail length, length of the upper unicuspids
(LLU 2-0-2-5 mm, x 2-25 mm ; TL 28-46 mm, x 38-35 mm in the European/
Algerian group ; LLU 1-8-2-3 mm> x 2'16 mm ; TL 36-51 mm, x 45-33 mm in
the Asian group). The premolar in mainland European populations has a very
broad, squarish lingual portion in comparison with the narrow lobed appearance

2Q2

P. D. JENKINS

C.russula Europe/
Algeria

FIG. ii. Distribution of C. russula, C. dsinezumi and C. attenuata.

distribution uncertain.

Broken lines -exact

characteristic of the Asian subgroup and Corsica (see Figs 5, 6 and 7). Skull size
in the Asian subgroup is also on average slightly smaller than in the European/
Algerian group, see Table 3.

OTUs of the European subgroup are fairly similar to each other with D2 values
of 1-3. In particular Spain (6) and Portugal (7) are very alike with a D2 of 1-23,
as are France (4) and Germany, Belgium and Holland (10) with a D2 of 1-42. The
Algerian population, although separated from the European mainland and Sardinia,
is nevertheless very similar as reflected by the low D2 values of 2-42-2-86.

As in C. suaveolens the greatest variation is shown by island populations,
especially Sicily (15), Sardinia (16) and Corsica (17) where the D2 values range from
3-92 between Sicily and Corsica, 3-95 between Sicily and Sardinia and 5-74 between
Sardinia and Corsica. Specimens from Sicily have short narrow skulls in comparison
with those from Sardinia which also have fairly well-spaced upper unicuspids (see
Table 5). Specimens from Sicily are most similar to Spanish specimens with a D2
value of 1-96, while those from Sardinia are closest to those from France and Germany,
Belgium and Holland with D2 values respectively of 2-14 and 2-2.

In spite of their close proximity to Sardinia and the European mainland, Corsican
specimens show much more similarity to the Asian subgroup than to the European

VARIATION IN EURASIAN CROCIDURA 293

one. D2 values between Corsica and the Asian subgroup are very low, varying from
1-44 to 2-33, while the lowest D2 between Corsica and any member of the European
group is 2-89 for Spain. The Corsican population is compared with those from
Sicily and Sardinia in Table 5. However, the resemblance of the Corsican population
to members of the Asian subgroup is unlikely to indicate a correspondingly close
relationship. It is more likely to be allied to European mainland and nearby island
populations despite the dissimilarity, which is probably a reflection of an isolated
island population evolving along different lines.

TABLE 5
Variation between island populations of C. russula

Corsica Sicily Sardinia

(n = 15) (n = 5) (n = 9)

Condylobasal length, mm 17-5-19-1 18-0-18-3 18-3-19-1

x 18-10 x 18-10 x 18-75

Upper too throw length, mm 7-7-8-7 8-1-8-3 8-3-8-9

x 8-21 x 8-23 x 8-62

Labial length of unicuspids, mm 1-9-2-3 2-1-2-2 2-3-2-5

x 2-06 x 2-16 x 2-42

Braincase breadth, mm 8-4-8-9 8-5-8-7 9-o-9'3

x 8-72 x 8-63 x 9-16

Tail length, mm 40-49 3J-41 33-4°

x 44-8 x 35-0 x 36-7

The Asian subgroup of C. russula has a wide distribution but shows very low D2
values of 1-2. The adjacent Israel (25) and Lebanon (24) populations are very
similar, with a D2 value of 1-68, and although geographically separated, populations
from Lebanon and Punjab (32) are also similar with a D2 value of 1-83. Although
geographically fairly close together Punjab and Kashmir (31) differ slightly, for while
Punjab is fairly typical of the group Kashmir specimens are at the upper limit of the
size range (e.g. UTL 7-4-8-7 mm, x 8-25 mm for Punjab ; 8-6-8-8 mm, x 8-73 mm
for Kashmir).

The earliest name for the Asiatic group is C. gueldenstaedti Pallas, 1811 from the
Caucasus. This form has been considered conspecific with European C. russula
(from which it is geographically separated) by Kuzyakin (1944, 1965), Ellerman &
Morrison-Scott (1966), Spitzenberger (1970), Felten, Spitzenberger & Storch (1973)
and Richter (1970). The last placed East Mediterranean forms of russula in the
same subspecies group as gueldenstaedti.

C. gueldenstaedti has been regarded as specifically distinct by Vinogradov (1958),
Gureev (1971) and Kock (1974). Vinogradov presumably contrasted the structure
of the external genitalia of both European russula and gueldenstaedti, although only
producing evidence about the latter. The character he did use was the difference in
relative tail length.

Without actually examining any specimens of gueldenstaedti from the Caucasus, I
hesitate to use this name as the senior synonym of any of the subspecies listed here,

294

P. D. JENKINS

although accepting its usage by other authors for C. r. monacha of Turkey as probably
correct. For the same reason it seems incorrect to use it as a blanket name as
Richter did, especially as some of the subspecies he included in that group appear
fairly distinct. Also it is not necessarily correct to ally Asian forms with Mediter-
ranean island forms which may have originated from any of many possible sources
and which are presumably fairly isolated genetically. For example, although
Corsican animals very closely resemble populations from Lebanon and Punjab they
are also similar to Spanish specimens (see above, p. 293).

The following subspecies have been examined and fall into two groups ; European
group

C. r. russula (Hermann, 1780). European mainland (4, 10, 13).

C. r. pulchra Cabrera, 1907. Spain, Portugal (6, 7).

C. r. peta Montagu & Pickford, 1923. Guernsey, Herm and Alderney (3).

C. r. ichnusae Festa, 1912. Sardinia (16).

C. r. sicula Miller, igoia. Sicily (15).

C. r. cyrnensis Miller, 1907. Corsica (17).

Associated with this group is

C. r. agilis Levaillant, 1867. Algeria (8).

Asian group

C. r. monacha Thomas, 1906 (probably C. r. gueldenstaedti Pallas, 1811 see above).

Turkey (22).
C. r, pullata Miller, 1911. Kashmir (31).

Also samples from Lebanon (24), Israel (25) and Punjab (32).

The affinities of C. russula and C. suaveolens have been discussed previously under
C. suaveolens and those of C. russula and C. leucodon are dealt with in the account
of C. leucodon.

Punjab (32)

Dar.eeling (341

Bhutan (35)

FIG. 12. D2 values between the closest geographical populations of C. russula and

C. attenuata.

VARIATION IN EURASIAN CROCIDURA 295

C. attenuata may be distinguished from C. russula by the longer tail (TL 36-51 mm,
x 45-33 mm in Asian russula ; 50-79 mm, x 62-94 mm in C. attenuata). Kashmir
(31) and Punjab (32) populations of the Asian subgroup of russula mark the eastern
limit of this group, while Darjeeling (34) and Bhutan (35) are the westernmost
populations of C. attenuata. The D2 values between these four populations are
shown in Fig. 12.

The lowest figure, which is between Punjab (32) and Bhutan (35), does not in this
case represent great similarity as specimens from Punjab may usually be distin-
guished from Bhutan specimens by their slightly broader skulls and shorter tails
(IB 4-2-4-5 mm, PB 6-0-6-4 mm> TL 40-51 mm in Punjab specimens of C. russula ;
IB 3-9-4-2 mm, PB 5-7-6-1 mm, TL 52-63 mm in Bhutan specimens of C. attenuata).

Crocidura dsinezumi Temminck, 1844

Specimens from Japan (47), currently regarded either as a distinct species, C.
dsinezumi, or as a subspecies of C. russula, emerge in the CVA close to certain OTUs
of both the Asian subgroup of C. russula and C. suaveolens. The Japanese specimens
may be distinguished from their nearest neighbours of C. suaveolens occurring in
Korea and Tsushima Islands, which are smaller, see Table 3. Although geo-
graphically well separated from other samples of C. suaveolens, the Japanese popula-
tion shows low D2 values with some OTUs, such as Jersey and Sark (2) and Greece
(21), of 2-7. Low D2 values are also shown between Japan and the Asian subgroup
of C. russula, for example Punjab (32) with 2-7, Lebanon (24) with 2-8. The size
of the skull is within the upper limits of C. suaveolens but only just within the lower
limits of Asian russula (see Fig. 10), while the tail is relatively long and within the
range of Asian russula (TL 37-50 mm, x 44-6 mm).

The Japanese population was described as a distinct species, C. dsinezumi, but
was treated as a subspecies of C. russula by Ellerman & Morrison-Scott (1966).
Imaizumi (1970) regards C. dsinezumi as a distinct species and separable from
C. suaveolens occurring in China, Korea and the Tsushima Islands. This view seems
preferable and it is considered here to be a separate species.

Crocidura attenuata Milne-Edwards, 1872

This species includes the following OTUs : Darjeeling (34), Bhutan (35), Assam
(36), Burma (37), China (44) and also Christmas Island (40). Specimens from
Christmas Island have hitherto been assigned to C. fuliginosa trichura. The
distribution of C. attenuata is shown in Fig. n.

D2 values between Darjeeling, Bhutan, Assam and China range from 2-43 to 3-47,
which, while not as low as in some groups, nevertheless indicate a fairly high degree
of relationship, especially when the geographical distances and variation in ecological
conditions are taken into account. Darjeeling, Bhutan and Assam are at least
partially isolated from each other by the mountainous terrain and separated from
Burma and China by distance and differing habitats. The Burmese population is

296

P. D. JENKINS

separated from any group other than C. attenuata by very high values of D2. The
lowest D2 value is 4-84 with Assam and 4-91 with Darjeeling.

Christmas Island is a considerable distance from the main distribution area of
C. attenuata but it shows relatively low D2 values of 3-65-4-73 with Darjeeling,
Bhutan, Assam and Burma. It shows no similarity to any other group entered in
the CVA and may be readily distinguished from both the small and very large species
oi Crocidura occurring in Indonesia, which is the nearest area where shrews of this
genus occur. A single specimen from Java and three from Sumatra closely resemble
both Christmas Island specimens and other populations of C. attenuata, which
suggests that this species may also be present, under different names and rarely
recorded, in some parts of Indonesia. This would make the presence of attenuata in
Christmas Island more likely, as a representative at the extreme limit of a species
having a discontinuous distribution from Asia down through the Indonesian chain.

The Mahalanobis' distances are a reflection of the range of variation in this group,
for example, in populations from Bhutan and Christmas Island the anterior part of
the skull is fairly narrow (MB 5-2-5-9 mm, x 5-5 mm) but the posterior part of the
skull in the latter population is wider than in the former (PB 5-7-6-1 mm, x 5-84 mm
in Bhutan ; 6-1-6-5 mm> x 6-32 mm in Christmas Island) - see Fig. 13. Measure-
ments for the species are given in Table 6.

6-0-

O

*
*
* **

o

D

DO •
•

O

o

DOT

O

8

5-5

6-5

width across maxillae at

FIG. 13. Variation in skull width in populations of C. attenuata.

Darjeeling • ; Bhutan * ; Assam O ; Burma Q ; China T ; Christmas Island

VARIATION IN EURASIAN CROCIDURA

297

C. attenuata has apparently been confused with C. fuliginosa in the past and it
seems likely that the latter name has been used at various times to refer to quite
different animals. For example, Blanford (1888) synonymized C. rubricosa and
C. kingiana (currently regarded as subspecies of C. attenuata) with C. fuliginosa,
giving the distribution as Eastern Himalayas, Assam and Tenasserim (Burma).
Thomas (1889) gave the distribution of C. fuliginosa as Himalayas through Burma to
Java and in this paper Dobson described a new shrew from Christmas Island as a
geographical race - C. fuliginosa trichura. The holotype and ten other specimens
from Christmas Island are in the BMNH ; information given above indicates close
association with C. attenuata and certainly none with the group referred to as C.
fuliginosa in this report. The correct name for the Christmas Island Crocidura is
thus C. attenuata trichura,

The following subspecies have been examined :

C. a. attenuata Milne-Edwards, 1872. China and Hainan (44).
C. a rubricosa Anderson, 1877. Assam, Himalayas (34, 35, 36).
C. a. trichura Dobson, 1888. Christmas Island (40) - formerly attributed to C.
fuliginosa trichura.

Also a sample from Burma (37), an attenuata whose affinities are uncertain at sub-
specific level.

The affinities of this species with members of the Asian subgroup of C. russula
have been discussed previously under the latter species. Key 3 (p. 282) and Fig. 7
give the characters for separating C. attenuata from C. fuliginosa and others occurring
in the area.

TABLE 6

Size range of C. attenuata, C. leucodon, C. lasiura and C. fuliginosa

Upper

Width of

Braincase

Condylobasal

toothrow

maxillae at

breadth,

length, mm

length, mm

M2, mm

mm

18-3-20-6

7-8-9-8

5-2-6-5

8-2-10-0

x 19-95

x 8-79

xs-Sg

x 9-12

17-8-19-9

8-0-9-2

5-7-6-5

8-6-9-5

x 18-74

x 8-60

x 6-15

x 9-11

19-0-21-4

8-8-10-2

5-8-6-8

8-8-10-2

x 19-96

x 9-18

x 6-29

X9'45

18-6-21-6

9-2-10-3

6-2-7-0

9-1-10-2

x 20-77

X9-74

x6-85

X9-67

21-3-25-0

9-4-11-5

6-2-7-8

9-5-11-1

x 22-80

x 10-40

x 6-90

X IO-IO

C. aitenuata Asia

C. leucodon France, Germany,
Yugoslavia
Switzerland, Asia Minor

C. lasiura Korea

C. fuliginosa Asia, Malaysia

Crocidura leucodon (Hermann, 1780)

Fig. 14 gives the distribution of this species. The OTUs are : France (5),
Germany (n), Switzerland (14), Yugoslavia (20), Turkey (23), Israel and Lebanon

P. D. JENKINS

C.leucodon

C tas.ura

FIG. 14. Distribution of C. leucodon and C. lasiura.

uncertain.

Broken lines -exact distribution

(26) and also Iran (27) which has hitherto been regarded as a subspecies of
russula - C. r. caspica.

This is a loosely knit group exhibiting a large range in variation. Those showing
greatest association are European populations from France (5), Germany (n) and
Yugoslavia (20) which have D2 values between i and 3. French and German
populations in particular have crowded unicuspids (LLU 1-8-2-1 mm, x 1-93 mm),
while this measurement in the Yugoslavian population is 1-9-2-3 mm, x 2-05 mm.
Skins usually show a sharp delineation along the flanks between the dark dorsal and
pale ventral colour.

Although hitherto believed to belong to C. russula, Iranian specimens can be readily
distinguished from that group - see last section of this species account. The Iranian
population is very similar to that from Turkey with a D2 value of 2-4. Both Iranian
and Turkish specimens on average have larger skulls with uncrowded unicuspids
in comparison with their European counterparts (LLU 2-1-2-55 mm, x 2-23 mm).
Turkish specimens have a dark dorsum and paler venter although not so definitely
bicoloured as European specimens, while Iranian specimens differ markedly in their
very dark dorsal and ventral surface.

Specimens from Israel and Lebanon are most similar to those from Turkey and
Iran, although the D2 value is rather high. They fall within the range on skull size
and the pelage colour resembles that of Turkish specimens.

VARIATION IN EURASIAN CROCIDURA 299

Although Swiss specimens might be expected to be most similar to nearby
populations from France and Germany, Mahalanobis' distances are from 3 to 4,
while the D2 values between Swiss and Turkish populations is 2-26 and between
Switzerland and Iran 2-77.

Variation between the two subgroups is shown in Table 6.

Turkish specimens are associated with other OTUs in this group by D2 values of
2-5 and they represent C. leucodon lasia which has a chequered history. It was
originally described by Thomas (1906) as a subspecies of C. leucodon, though he later
(1907) declared it distinct enough to be a full species - C. lasia. Bobrinskii et al.
(1944) decided it was a subspecies of the Korean C. lasiura, an opinion with which
Ellerman & Morrison-Scott (1951) concurred, although in the second edition
Bobrinskii (1965) included it as a distinct species -- C. lasia. Vinogradov (1958),
examining the external genitalia, was unable to differentiate lasia from C. leucodon
although C. lasiura was found to be distinguishable. Lay (1967) and Gureev
(1971) also regard it as a subspecies of C. leucodon and this seems the best arrangement
since it is fairly similar to, though relatively larger than, the European members of
C. leucodon. It is quite similar to specimens of C. lasiura (D2 value of 3-03) but the
considerable discontinuity of distribution and Vinogradov's findings discourage the
view that they might belong to the same species.

Specimens from Iran, including the holotype of C. russula caspica, are most
similar to C. leucodon lasia from Turkey and other members of the leucodon group.
In 1951 Ellerman & Morrison-Scott noted that caspica might represent C. lasiura
(in which species they had placed C. leucodon lasia}. However, Bobrinskii (1965),
Lay (1967) and Richter (1970) placed it with C. russula in the gueldenstaedti subspecies
group with a number of other subspecies which in this report belong to the Asian
subgroup of C. russula and are readily distinguished from Iran specimens. Finally,
Gureev (1971) included C. r. caspica, together with C. r. monacha, in synonymy with
C. gueldenstaedti. This last seems very suspect since specimens of C. r. monacha
from Turkey may be distinguished from Iran specimens on many characters (see
Table 7) . Despite the marked colour difference I believe these specimens from Iran
should be regarded as a subspecies of C. leucodon - C. leucodon caspica - close to
C. leucodon lasia.

Another subspecies, C. leucodon persica, also occurs in Iran. The holotype and
one other specimen are in the BMNH. They resemble European members of
leucodon in size and colour and are easily distinguished from Iran specimens of
C. 1. caspica. However, while all specimens seen of C. I. caspica are from low alti-
tudes (the type series from under 25 m), C. I. persica appears to be a high altitude
subspecies since the holotype was collected at 1980 m.

The resemblance of specimens from Israel and Lebanon to C. russula from Central
Europe was noted by Thomas (1919) in his original description of C. russula judaica.
In 1951 Ellerman & Morrison-Scott suggested the possible inclusion of this form with
C. lasiura (along with lasia}. Harrison (1963) considered it to be a local form of
C. leucodon and this view is supported by the resemblance to specimens of C. leucodon
from Turkey and Iran (see Table 7).

3oo P. D. JENKINS

a

a

o

C

Q CTj *"rt

"2 « - ,li c
I 1 1 .1

^ il>£| ^M ^<»

o <j«rl ST*******

•w i-c on

O o

3 o

o

P< § -a

138 M S*°P VO °P M ^>

""Q £3 CM r ON N vO CM ON

fi 3 ^ rt I O i X I ^ I

aj '*••> <tf -Q (t " cp flp op

""•» en J ON iX 00 '* io '^ 00

g oMhJ M
w

5, 6 2 ^ o r> t^0?

O ^ClS.ONMVOMON

g S «0 I P I

00 O N I

iX ob '^ o '^ ON '^ °p
ro

5

<0 £^>

— O

»S ^

. TH

1 ON

2 ?

°P rf 6 ro

o •*• ;; >r>
. . w .

1 vO 1 ON

O^ _-.
« ro
Tf ro

O H

1

0H

M

ob

^5 «X • IX

S'X

o

o
3

I

*O

.2 ^l"

julations

Q en , • 1-1
S^0^

8 a ^c-

?lf 5

• 00 P «

V

M .

1 00

ob N
i •

O^-
in . M
vO ^O ON ON

N ^ ^

^ 'X ob '*

H M

VO Tf

a

0 1 "3 I

? c 1

M

n-'X

1

" S

00

s

S

£

•*i

00

^

0)

• j i

0)

.

i

W)

rt

CD

"o

&

B

'X X3

OH
OnX!

1

C

1

rt ^

6 o3
en ^

| f

l_!

en b

c3

OH

1

o

1

0 «

s a

£ 8

O rC

O

^^

t.

r> O

B

o

Q

1

^~

5 c

S 1

| 8

0

^ w

PC

VARIATION IN EURASIAN CROCIDURA 301

The following subspecies have been examined and fall into two main groups :

C. 1. leucodon (Hermann, 1780). Most of Europe (5, n, 20).

C. I. persica Thomas, 1907. Elburz Mountains, Iran.

The second subgroup includes :

C, I. lasia Thomas, 1906. Turkey, Transcaucasia (23).

C. I. judaica Thomas, 1919. Israel and Lebanon (26).

C. /. caspica Thomas, 1907. S. coast of Caspian Sea, Northern Iran (27).

A sample from Switzerland (14) of uncertain affinity.

Keys i and 2 give the characters that are useful in distinguishing C. leucodon from
other species occurring in the area. Despite the original naming of Iranian specimens
as C. russula, they are particularly similar to specimens of C. leucodon lasia from
Turkey mentioned above and can be readily distinguished from the Asian subgroup
of C. russula by their more robust, longer and broader skulls. They are rather
similar to the European subgroup of C. russula, but the skulls are on average slightly
larger and the resemblance is unlikely to be a true indication of relationship. Table 5
gives a comparison of these populations.

Swiss specimens of C. leucodon are most closely associated in the analysis with
Swiss specimens of C. russula at a very low D2 value of 1-64 and with other European
members of that species at D2 values of 2-3. In Switzerland individual specimens
of the two species are not readily distinguished from each other except on the basis
of colour. Specimens of leucodon fall in the upper part of the russula size range
(CBL 18-0-19-9 mm, x 19-01 mm ; BB 8-5-9-4 mm, x 9-13 mm in russula ; CBL
19-2-19-8 mm, x 19-5 mm ; BB 9-2-9-5 mm, x 9-38 mm in leucodon}. This problem
in Switzerland obviously requires very careful investigation into ecology, behaviour
and genetics to try to establish the true relationship between these species.

Crocidura lasiura Dobson, 1890

The distribution of this species, represented in this report by the Korean population
(46), is given in Fig. 14.

Korea is geographically isolated from Middle Asian populations of C. leucodon, yet
the D2 value between Korea and Turkey (23) is comparatively low, at 3-03. The
range in skull size is similar for Korean and Turkish populations but on average
Korean specimens have larger skulls, see Table 6. In Korean specimens the ventral
pelage colour is only slightly paler than the dark dorsum, in contrast to the more
marked differentiation in Turkish specimens.

In this study C. lasiura was represented by a large series from Korea, including
the holotype of C. I. thomasi. In the CVA this sample showed some affinity to
C. leucodon lasia (23) with a D2 value of 3-03, but diverged more from C. leucodon
caspica (27), a D2 value of 4-77, and C. leucodon judaica (26) a D2 value of 6-38.
Ellerman & Morrison-Scott (1966) also noted this similarity but the evidence of
dissimilarity in penis structure (Vinogradov, 1958) strongly suggests that it should
be retained as a distinct species.

302 P. D. JENKINS

The subspecies examined was :
C. lasiura thomasi Sowerby, 1917. Korea (46).

See Key (p. 282) for comparison with other species in the surrounding area.

OTHER SPECIES RECORDED FROM THE AREAS OF STUDY BUT NOT
INCLUDED IN THE ANALYSIS

The following species recorded from Eurasia are either absent from the collection
or there are too few specimens for inclusion in the previous account.

C. hispida Thomas, 1913. Middle Andaman Island, Bay of Bengal. The BMNH
houses the holotype, which is a very large animal, (CBL 27-4 mm) much larger
than any of the species described in this account and with the tail longer than the
head and body. It is also distinguished by its semi-spinous pelage.

C. andamanensis Miller, 1902. South Andaman Island, Bay of Bengal. No
specimens apart from a small hair sample which differs from other species in that
some of the hairs are flattened, although they are not so broad as the hairs of
C. hispida, and there is no mention in the original description of the pelage being
spinous. From the description of the holotype it appears to be comparable in
size, though slightly smaller than C. hispida (CBL 25-6 mm).

C. nicobarica Miller, 1902. Great Nicobar Island, Bay of Bengal. No specimens
apart from a small hair sample which contains flattened, broadened, hairs
resembling those of C. hispida but less spinous. The condylobasal length of the
holotype given in the original description is 27-0 mm.

C. miya Phillips, 1929. Sri Lanka. The skin and skull of the holotype are in the
collection of the BMNH. The specimen is distinguished by the proportionately
very long tail but in other respects it is very similar to C. attenuata (CBL 19-7 mm,
UTL 8-7 mm, MB 5-9 mm, HB 79 mm, TL 88 mm).

C. caudata Miller, igoia. Palermo, Sicily. No specimens examined. Kahmann &
Einlechner (1959) believe that this is a subspecies of C. russula.

C. balearica Miller, 1907. Minorca, Balearic Islands. Kahmann & Vesmanis (1974)
consider this to be a subspecies of C. suaveolens.

From the U.S.S.R. Gureev (1971) records the following species which I have not
seen :

C. ognevi Stroganov, 1956. Altai.

C. pamirensis Ognev, 1928. Lake Drum vicinity, southern slopes of Pamir Mountain

Range.

C. armenica Gureev, 1963. 14 km down river from Garni, Armenia.
C. dinniki Ognev, 1921. Stavropol'.

From Japan Imaizumi (1970) records :
C. orii Kuroda, 1924. Amami-Oshima, Riukiu Islands.

Preliminary examination of examples of the following forms recorded from
Sumatra and Java by Chasen suggests that they represent C. fuliginosa :

C. villosa Robinson & Kloss, 1918. Sumatra. Holotype in BMNH.

VARIATION IN EURASIAN CROCIDURA 303

C. brunnea Jentink, 1888. Sumatra, Java. Holotype of C. b. brunnea and ten

specimens of C. b. pudjonica loaned by Leiden Museum.
C. brevicauda Jentink, 1890. Java. Holotype and fifteen specimens loaned by

Leiden Museum.
C. orientals Jentink, 1890. Java. Holotype and eight specimens loaned by

Leiden Museum.
C. melanorhyncha Jentink, 1910. Java. Holotype and one other specimen loaned

by Leiden Museum.

The following species recorded from the Malaysian region have also been subject to
a preliminary examination and may be easily separated from C. fuliginosa by their
small size (CBL 15-2-19-5 mm).

C. monticola Peters, 1870. Java, Borneo, Lombok, Sumbawa, Komodo, Flores,
Sumba and ? Timor. There is only one specimen in the BMNH collections.

C. aequicauda Robinson & Kloss, 1918. Sumatra, Malay Peninsula. Holotype in
BMNH.

C. beccarii Dobson, 1886. Sumatra. One damaged specimen only.

C. bartelsii Jentink, 1910. Java. Four specimens loaned by Leiden Museum.

C. maxi Sody, 1936. Java. One damaged specimen loaned by Leiden Museum.

C. minuta Otten, 1917. Java. Two badly damaged specimens loaned by Leiden
Museum.

Of these bartelsii is probably conspecific with monticola. It is possible that maxi
and minuta might also belong to this species, although more specimens are required
to furnish any definite conclusions.

I have seen none of the remaining species listed by Chasen :

C. paradoxura Dobson, 1886. Sumatra.

C. neglecta Jentink, 1888. Sumatra.

C. weberi Jentink, 1890. Sumatra.

C. lepidura Lyon, 1908. Sumatra.

C. vosmaeri Jentink, 1888. Banka Island.

There are several unidentified specimens from Sulawesi which fall within the range
of C. fuliginosa although this species has not previously been described from there.
Laurie & Hill (1954) list the following species, none of which has been examined :
C. elongata Miller & Hollister, 1921. North eastern Sulawesi.
C. nigripes Miller & Hollister, 1921. North eastern Sulawesi.
C. rhoditis Miller & Hollister, 1921. North eastern Sulawesi.
C. lea Miller & Hollister, 1921. North eastern Sulawesi.
C. levicula Miller & Hollister, 1921. Middle Sulawesi.
C. tenuis (Miiller, 1839). Timor.

Taylor (1934) records eight species from the Philippines of which the BMNH
houses examples of only the first one :

C. grayi Dobson, 1890. Luzon and Benguet, Philippines. Holotype and two other
specimens in the BMNH (CBL 18-6-20-0 mm, UTL 8-5-9-1 mm, BB 9-3-9-5 mm,

304 P. D. JENKINS

TABLE 8

Mahalanobis' distances. The 47 OTUs are numbered as in Table i
3 4 5 6 7 8 9 10 ii 12 13 14 15 16

I

2

3
4
5
6
7
8

9
10

ii

12

13

o-o

4'9

6-2
6-2
4'5
4'4
5-8
3-6
6-6

5'5
2-7
6-0

o-o
6-4
7-9
7-6
6-4
6-4
8-1
5'3
8-4

6-9

7-5

o-o

2-O

4-8
1-9

2-O

3'7
6-3
2-6

4'9
6-5
2-7

o-o
4'4
2-0
2'5
2'5

7-1
7-8

2-O

o-o
4-8
5'5
4'7
6-1

4'5

2-8
8-1
3'5

o-o

1-2

2-4
5-9
2-9

5'2

2-6

o-o
2-8
5'9

5'3
5'7
3'3

o-o
6-3
2-9

5'4
7'4
3-0

o-o

7'5

5'3
4-0
6-8

o-o

4'9
8-3
1-8

o-o
6-9
4'3

o-o
7-6

o-o

14

6-9

8-5

4-0

3'2

3'4

4-1

4'7

3'9

7-5

2-4

4-2

8-7

1-6

o-o

15

4-6

6-5

2-7

3-0

4'7

2-O

2-6

3-3

5-6

3'3

4'9

6'2

2-7

3-8

o-o

16

7-2

9-2

3'5

2-1

6-0

3'i

2-9

2-8

8-1

2-2

6-4

8-5

3-4

4'4

4-0

o-o

17

3-1

4-2

3'4

4'5

5-7

2-9

3-2

4'7

5'7

5'2

5'9

4'9

4'5

5-8

3'9

57

18

3'2

4-1

6-4

7'4

7-4

5-8

5'4

6-9

3'4

7-9

6-4

2-2

7'3

8-4

5'7

8-1

19

2'0

3'4

5'3

6-7

7-2

5-0

4-6

6-5

3'9

7-2

6'2

2-O

6-8

7-9

5-2

7-5

20

5-6

7'5

4-2

3-6

1-9

3-8

4'5

3'5

6-2

3-6

3'3

7'7

2-9

3-0

3'9

5-o

21

3-1

3'9

4'5

5-6

6-5

3-8

3-8

5-6

4'5

6-2

6-0

3'5

5'5

6-8

3-6

6-4

22

4'7

5'5

3'5

4-1

6-1

2-8

3'4

4'5

6-9

4-8

6-7

6-1

4-0

5'3

3'7

5'2

23

8-0

9'3

5'3

4'4

3'4

5'3

6-2

4'9

8-6

4-0

5'3

9-9

3-1

2-3

5'4

5-8

24

3'7

4'4

3'3

4'5

6-3

3-1

3'3

5-0

6-2

5'4

6-5

5-0

4-8

6-2

4'3

5'7

25

3'3

4-6

4-0

5'0

6-7

3'2

3-1

4-8

5'9

5'7

6-7

4'4

5'3

6-7

4'5

5'7

26

9'3

II-2

5'9

4'i

5-6

5-7

6-0

4-2

9-7

3-8

6-8

10-9

4'4

4'3

6-5

4-0

27

7'4

8-5

4'7

3'7

4-0

4'4

5-1

4-2

8-1

3'7

5-6

9-0

2-5

2-8

4-6

4-8

28

2'6

3-6

3-6

4'7

4-6

3'3

3-6

4-8

4-6

5-1

4'3

4-2

4-1

5-1

3'7

6-0

29

3-7

2-9

6-9

8-2

8-6

6-4

6-4

8-0

6-4

8-9

8-4

3-8

8-1

9'3

7-2

9-1

30

2-7

4-0

6-1

7-0

5'9

5'5

5'5

6-2

3'4

7-1

4-8

3'3

6-2

6-9

5'3

7'9

31

4'7

5'9

3'9

4-i

4'9

3'3

3'7

4'i

6-7

4-7

57

6-5

3'9

5'i

4'7

5'3

32

3'5

3-8

4'i

5'2

6-0

3'7

3'9

5'3

5'9

5'7

6-0

4'5

4-8

6-1

4'5

6-3

33

6-0

5'3

8-7

10-2

n-5

8-5

8-2

10-2

8-2

II'O

II'O

5'2

10-6

I2'O

9-2

10-9

34

8-2

7-9

8-2

8-6

10-0

7-6

8-0

8-8

10-6

9'5

II-2

9-2

8-7

10-0

8-7

9-5

35

5'9

5-6

6-6

7'4

9-0

6-0

6-3

7-6

8-3

8'4

9-6

6-6

7-6

9-0

7-0

8-4

36

8-0

7'7

6-6

7-0

8-9

6-4

6-9

7'9

10-4

7-9

IO'O

9-2

7-2

8-5

7'5

8-1

37

10-8

10-7

9'5

9-8

11-7

9-6

9-9

10-5

12-9

10-5

I2'8

ii-9

10-3

II'2

10-8

10-7

38

10-9

ii-7

8-4

7'7

8-6

8-3

9-1

8-5

12-9

7-7

IO'2

12-9

7-5

7'5

9-1

8-6

39

13-1

13-6

10-4

9-7

IO-2

10-6

n-4

10-8

14-9

9-7

II-9

15-2

9'5

9'3

II'2

10-7

40

10-0

9-2

10-0

10-7

12-6

9-7

9-9

n-i

12-3

ii-7

13-5

10-4

II-O

12-4

10-8

"•5

41

11-7

n-8

9'3

8-9

9'4

9'5

10-3

10-0

13-3

9'3

II-2

13-5

8-7

8-9

10-3

10-3

42

2-4

3-7

3'9

5-2

6-0

3'7

3'2

5-2

4-2

5'7

5'i

2-9

5'2

6-4

4'3

6-0

43

I2'I

I2'O

9-8

9'5

9-8

IO-I

10-8

10-7

13-8

9-8

"•5

13-8

9-1

9-2

10-7

10-8

44

6-4

6-9

5'3

5'5

7-1

4-8

5'5

6-0

8-7

6-5

8-4

8-1

5'7

6-9

6-0

6-8

45

3'3

3'3

7'5

8-8

9-1

6-9

6-8

8-3

47

9'4

8'2

2-0

8-7

9-7

7-0

9-6

46

9-8

II'O

7-9

7-0

5'4

7'7

8-6

7-2

10-5

6-5

7-1

ii'9

5'7

4'7

7-6

8-2

47

3'3

2-7

5'i

6-5

7-3

4-8

5'i

6-8

5'7

7-2

7-2

4-0

6-2

7-5

4'9

7'8

VARIATION IN EURASIAN CROCIDURA 305

TABLE 8 (Cont.)

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31 32

17

o-o

i8

5-1

o-o

19

4-0

2-O

o-o

20

4-8

7-1

6-7

o-o

21

3'3

2-9

2-7

6-0

o-o

22

2-0

6-4

5-5

5'3

4-1

o-o

23

6-5

9-6

9-2

3'3

8-0

6-0

o-o

24

1-3

5-5

4'3

5'7

3-6

2-0

6-9

o-o

25

1-5

4-7

3'7

5-8

3'4

2-7

7-6

1-7

o-o

26

7-7

10-4

10-0

5-0

9-1

7-2

4'4

7-9

8-1

o-o

27

5'5

8-6

8-4

3-8

6-8

4-6

2-4

5-8

6-3

4-2

o-o

82

2-1

4'3

3-6

4'i

3-0

3'3

6-0

2-8

3-0

7'7

5'2

o-o

29

3-8

4-8

4-0

8-1

4-6

5'2

9-9

4-0

3'5

II-2

8-9

4-4

o-o

30

4-6

3-0

3'4

5-5

3-8

5-8

7-9

5"4

4'9

9'4

7-3

3-0

5-o

o-o

31

2-3

6-5

5'7

4-0

5'2

2-9

5-2

2-9

3-0

6-1

4'3

3-0

5-3

5-4

o-o

32

1*4

4'9

4-2

5'4

3'4

2-3

6-7

1-8

2-1

8-1

5'5

1-9

3-5

4'3

2-6 o-o

33

6-5

6-4

5'4

10-9

6-5

7-5

12-8

6-1

5-8

.13-6

n-7

7-4

3-8

7-9

8-3 6-5

34

5-7

9-6

8-8

9'3

7-9

5'3

9-7

5-6

5'7

10-7

8-3

7-4

6-1

9'5

5'9 5-7

35

3'9

7-2

6-4

8-4

5'5

3'9

9'3

3'5

3'7

10-3

7-9

5-6

3'9

7-5

5'i 3'9

36

5'i

9'5

8-6

8-3

7'3

4'4

8-3

4-8

5'5

9'4

6-9

6-7

6-7

9'3

5'2 5'2

37

8-4

12-5

11-4

n-i

10-7

7'9

10-8

7'9

8-5

II-2

10-0

10-0

9-2

12-4

8-2 8-7

38

8-4

12-8

n-9

7-8

10-7

7-7

6-3

8-7

9-2

7-4

6-6

9-2

ii-i

n-6

7-1 8-8

39

10-6

15-0

14-2

9'7

12-9

10-0

7-8

10-9

n-6

9-2

8-5

11-4

13-4

13-7

9-4 ii-o

40

7-9

n-i

10-3

I2-I

9'5

7'4

12-3

7'3

7'7

13-1

10-8

9-6

7-4

n-6

8-6 7-8

4i

9-0

13-5

12-7

9-1

11-4

8-4

7'4

9-1

10-0

9-1

7'5

10-0

n-3

12-5

8-0 9-3

42

2-9

3-1

2-2

5'5

2-7

4-2

7'7

3-0

2-7

8-3

6-7

2-2

4-1

3'4

4-1 2-8

43

9'5

13-9

I3-I

9'7

n-8

8-8

7-9

9-6

10-5

9-7

7'9

10-3

n-7

12-8

8-5 9-6

44

3'5

8-2

7-3

6-3

6-2

3-0

6-7

3-6

4-1

7-9

5'5

5'2

5'9

7'7

3'4 4'0

45

5'3

3-0

2-9

8-6

4-1

6-5

10-8

5-6

4'9

I2-I

9-9

5-1

3'4

4-1

7-1 5-2

46

8-4

u-6

II-2

5'2

10-1

8-0

3-0

9-1

9-6

6-4

4-8

7-9

n-6

9'4

7-i 8-6

47

2-8

4-6

3-8

6-8

2-8

3'5

8-3

2-9

3'3

IO-I

7-2

3'2

3-2

4'7

4-9 2-7

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

33

o-o

34

7-8

o-o

35

5-4

3-0

o-o

36

8-5

2-7

3-6

o-o

37

9'9

4'9

6-4

4-8

o-o

38

13-6

7'7

9'3

6-4

8-4

o-o

39

15-8

9-8

"'5

8-4

9-8

2-7

o-o

40

7-6

3'7

4'2

4-7

4'7

10-4

12-2

o-o

4i

13-6

7-2

9-1

5'9

7-1

3'7

3'7

9-2

o-o

42

6-2

8-0

5'7

7'5

10-4

10-7

12-9

9-7

11-4

o-o

43

14-0

7-9

9-6

6-4

7-6

4'3

4-0

9-6

1-5

n-7

o-o

44

8-2

3'3

3'5

2-4

6-1

6-3

8-5

6-1

6-4

6-2

7-1

o-o

45

4'5

8-8

6-2

9-2

n-8

I3-3

15-5

9-9

13-7

4'3

14-1

8-1

o-o

46

14-7

II-O

n-i

9-8

I2-I

6-1

7-2

I3-9

7'7

9.9

8-0

8-3

12-5

o-o

47

5'3

6-3

3-8

5'9

9-2

IO-2

12-3

7'7

10-4

3'7

10-7

5'i

3'9

10-1

o-o

306 P. D. JENKINS

HB of holotype 68 mm, TL of holotype 51 mm). This species seems very similar

to C. attenuata.

C. parvacauda Taylor, 1934. Cotabato, Philippines.
C. halconus Miller, 191 ib. Mindoro, Philippines.
C. beatus Miller, 191 ib. Mindanao, Philippines.
C. mindorus Miller, 191 ib. Mindoro, Philippines.
C. grandis Miller, 191 ib. Mindanao, Philippines.
C. palawanensis Taylor, 1934. Palawan, Philippines.
C. edwardsiana Trouessart, 1880. Jolo, Sulu Archipelago, Philippines.

ACKNOWLEDGEMENTS

For the loan of specimens vital to this study, I am indebted to Dr C. O. Handley
of the United States National Museum, Dr D. L. Harrison of the Harrison Zoological
Museum, Sevenoaks, Dr A. M. Husson of the Rijksmuseum van Natuurlijke Historic,
Leiden, and Dr F. Spitzenberger of the Naturhistorisches Museum, Vienna. My
thanks are due to Linda Jackson of Rothamsted Experimental Station, who was
responsible for running the computer program. I am especially grateful to Dr
M. Hills of the Biometrics Section, BMNH, who has given freely of his knowledge
and time in the mathematical studies involved, to Mr J. E. Hill and Mr I. R. Bishop
O.B.E. of the Mammal Section, BMNH, for their advice and encouragement at every
stage of the work and to Dr G. B. Corbet, Deputy Keeper of Zoology, BMNH, for his
valued criticism of the manuscript.

REFERENCES

ALLEN, G. M. 1938. The mammals of China and Mongolia. Parti. New York.
ANDERSON, J. 1877. Description of some new and little known Asiatic shrews in the Indian
Museum, Calcutta. /. Asiat. Soc. Beng. 46 (2) : 280.

1 88 1. Catalogue of Mammalia in the Indian Museum, Calcutta. Part i. Calcutta.

BANKS, E. 1931. A popular account of the mammals of Borneo. /. Malay. Brch R. Asiat.

Soc. 9 (2) : 1-139, 9 pis.

BLANFORD, W. T. 1888. The fauna of British India, including Ceylon and Burma. London.
BLYTH, E. 1855. Proceedings of the Asiatic Society of Bengal for May 1855. /. Asiat. Soc.

Beng. 24 (4) : 362.

1856. Memoir on the Indian species of shrews. Ann. Mag. nat. Hist. (2) 17 : 11-28.

BOBRINSKII, N. A., KUZNETZOV, B. A. & KUZYAKIN, A. P. 1944. [Mammals of the U.S.S.R.]

Moscow.

- 1965. [The key to the mammals of the U.S.S.R.] Moscow.

CABRERA, A. 1907. Three new Spanish Insectivores. Ann. Mag. nat. Hist. (7) 20 : 213.
CARTER, T. D. 1942. Three new mammals of the genera Crocidura, Callosciurus and Pteromys

from Northern Burma. Am. Mus. Novit. No. 1208 : 1-2.

CHASEN, F. N. 1940. A handlist of Malayan mammals. Bull. Raffles Mus. No. 15 : 1-209.
CHUGONOV, Y. D. In Flint, V. E., Chugonov, Y. D. & Smirin, V. M. 1965. [Mammals of the

U.S.S.R.] Moscow.
DELANY, M. J. & HEALY, M. J. R. 1966. Variation in the white-toothed shrews (Crocidura

spp.) in the British Isles. Proc. R. Soc. series B 164 (994) : 63-74.
de WINTON, W. E. 1898. On a collection of mammals from Morocco. Proc. zool. Soc. Lond.

1897 : 953-962.

VARIATION IN EURASIAN CROCIDURA 307

DOBSON, G. E. 1887. Description of new species of Soricidae in the collection of the Genoa
Civic Museum. Annali Mus. civ. Star. nat. Giacomo Doria (2) 4 : 1-4.

DOBSON, G. E. 1889. In Thomas, O. On the mammals of Christmas Island. Proc. zool. Soc.
Lond. 1888 pt. 4 : 532.

1890. Description of a new species of Crocidura from the Amur Region. Ann. Mag. nat.

Hist. (6) 5 : 31.

1890. Descriptions of new species of Crocidura. Ann. Mag. nat. Hist. (6) 6 : 494.

ELLERMAN, J. R. & MORRISON-SCOTT, T. C. S. 1951. Checklist of Palaearctic and Indian

mammals, 1758-1946. London.

1966. Ibid., 2nd edn.

FELTEN, H., SPITZENBERGER, F. & STORCH, G. 1973. Zur Kleinsaugerfauna West-Anatoliens.

Teil 2. Senkenberg. biol. 54 (4/6) : 227-290.
FESTA, E. 1912. Descrizione di una nuova specie del genere Crocidura di Sardegna. Boll.

Musei Zool. Anat. comp. R. Univ. Torino 27 (648) : i.
GUREEV, A. A. 1963. [Mammals of the fauna of the U.S.S.R. Part i.] Moscow, Leningrad.

1971. [World fauna of shrews (Soricidae).'] Leningrad.

HARRISON, D. L. 1963. Some observations on the white-toothed shrews (genus Crocidura
Wagler 1832) of Israel. Bull. Res. Coun. Israel. 11B (4) : 177-182.

1964. The mammals of Arabia. Vol. i. London.

HASSINGER, J. D. 1970. Shrews of the Crocidura zarudnyi-pergrisea group with descriptions
of a new subspecies. Fieldiana Zool. 58 (2) : 5-8.

I973- A survey of the mammals of Afghanistan resulting from the 1965 Street Expedition

(excluding bats). Fieldiana Zool. 60 : 1-195.

HERMANN. 1780. In Zimmermann, E. A. W. Geographische Geschichte des Menschen, 2 : 382.

Leipzig.
HINTON, M. A. C. 1924. On a new species of Crocidura from Scilly. Ann. Mag. nat. Hist. (9)

14 : 509-510.

HOWELL, A. B. 1928. A new white-toothed shrew from Fukien, China. /. Mammal. 9 : 60.
IMAIZUMI, Y. 1970. The handbook of Japanese land mammals. Vol. i. Tokyo.
JENTINK, F. A. 1888. On the shrews from the Malayan Archipelago. Notes Leyden Mus.

10: 161-167.

1890. In Weber, M. C. W. Zoologische Ergebnisse einer Reise in Niederldndisch Ost-

Indien. Vol. i. Leiden.

1910. Two undescribed shrews from Mount Pangerango, Java. Notes Leyden Mus.

32 : 197-198.

KAHMANN, H. & EINLECHNER, J. 1959. Bionomische Untersuchung an der Spitzmaus

(Crocidura) der Insel Sardinien. Zool. Anz. 162 (3/4) : 63-83.
& KAHMANN, E. 1954. La musaraigne de Corse. Mammalia 18 (2) : 129-158.

& VESMANIS, I. 1974. Morphometrische Untersuchungen an Wimperspitzmausen

(Crocidura) i. Die Gartenspitzmaus Crocidura suaveolens (Pallas, 1811) auf Menorca.
Saugetierk. Mitt. 22 (4) : 313-324.

KLOSS, C. B. 1917. On two new pygmy shrews from the Malay Peninsula. /. fed. Malay

St. Mus. 7 : 127.

- 1917. On five new mammals from Siam. /. nat. Hist. Soc. Siam 2 : 283.
KOCK, D. 1974. Zur Saugetierfauna der Insel Chios, Agais. Senckenberg. biol. 55 (1/3) : 1-19.
KURODA, N. 1924. New mammals from Riukiu Islands, i and 3. Tokyo.
KUZYAKIN, A. P. see Bobrinski, N. A., Kuznetzov, B. A. & Kuzyakin, A. P. 1944 and 1965.
LAPTEV, M. K. 1929. Opredelitel' mlekopitayushchikh Srednei Azii [Key to the mammals

of Central Asia.] Materialy Sredne-aziatskogo zoologischeskogo sada (i) : 16. [N.V.]
LAURIE, E. M. O. & HILL, J. E. 1954. List of land mammals of New Guinea, Celebes and

adjacent islands 1758-1952. London.
LAY, D. M. 1967. A study of the mammals of Iran resulting from the Street Expedition of

1962-63. Fieldiana Zool. 54 : 1-282.

3o8 P. D. JENKINS

LEVAILLANT. 1867. In Loche, V. Exploration Scientifique de I'Algerie, Atlas Mamm.

pi. 4 fig. 2.
LINDSAY, H. M. 1929. Scientific results from the mammal survey no. 48 Indian shrews.

/. Bombay nat. Hist. Soc. 33 (2) : 1-15.
LYON, M. W. 1908. Mammals collected in eastern Sumatra by Dr. W. L. Abbott during 1903,

1906 and 1907, with descriptions of new species and subspecies. Proc. U.S. natn. Mus.

34 : 619-679, 5 pis.
MARTIN, R. A. 1967. Notes on the male reproductive tract of Nectogale and other Soricid

Insectivores. /. Mammal. 48 (4) : 664-666.
MEDWAY, LORD. 1965. Mammals of Borneo, field keys and an annotated checklist. /.

Malay. Brch R. Asiat. Soc. (1963) 36 part 3 : 1-193, 34 Pls. J maP-

MILLER, G. S. igoia. Five new shrews from Europe. Proc. biol. Soc. Wash. 14 : 41-45.
A new shrew from Switzerland. Proc. biol. Soc. Wash. 14 : 95.
Descriptions of three new Asiatic shrews. Proc. biol. Soc. Wash. 14 : 157-159.

1902. The mammals of the Andaman and Nicobar Islands. Proc. U.S. natn. Mus.

24 : 751-795, 2 pis.

1907. Some new European Insectivora and Carnivora. Ann. Mag. nat. Hist. (7)

20 : 389-398.

191 la. Two new shrews from Kashmir. Proc. biol. Soc. Wash. 24 : 241.

191 ib. Descriptions of two new genera and sixteen new species of mammals from the

Philippine Islands. Proc. U.S. natn. Mus. 38 : 391-404, 3 pi.

1913. A new shrew from Baltistan. Proc. biol. Soc. Wash. 26 : 113.

& HOLLISTER, N. 1921. Twenty new mammals collected by H. C. Raven in Celebes.

Proc. biol. Soc. Wash. 34 : 93-104.

MILNE-EDWARDS, A. 1872. Recherches pour servir a I'histoire naturelle des Mammiferes.

Paris.
MONTAGU, I. G. S. & PICKFORD, G. 1923. On the Guernsey Crocidura. Proc. zool. Soc. Lond.

1923 : 1044.
MULLER. 1839. In TEMMINCK, C. J. Verhandeling over de Natuurlijke Geschiedenis der

Nederlandsche overzeesche bizittingen, door de Leden der Natuurkundige Commissie in Indie,

1 : 26, 50. Holland.
OGNEV, S. I. 1921. Materials for the systematics of insectivorous mammals of Russia.

Ezheg. zool. Muz. 22 : 311-350.

1928. [Mammals of eastern Europe and northern Asia. Vol. i. Insectivora and

Chiroptera.] Moscow, Leningrad.

OSBORN, D. J. 1965. Hedgehogs and shrews of Turkey. Proc. U.S. natn. Mus. 117

(3517) : 553-566.
OTTEN, L. 1917. Over de biologic van Mus concolor. Meded.burgerl.geneesk.Dienst.Ned.-Ind.

(6) : 82-119.

PALLAS, P. S. 1811. Zoographia Rosso-Asiatica i. Leningrad.
PETERS, W. 1870. Uber neue Arten von Spitzmausen des Konigl. zoologischen Museums

aus Ceylon, Malacca, Borneo, China, Luzon und Ostafrika. Mber. K. preuss. Akad. Wiss. :

584-596.

PHILLIPS, W. W. A. 1929. New and rare Ceylon shrews. Spolia zeylan. 15 : 113-118.
RICHTER, H. 1970. Zur Taxonomie und Verbreitung der palaearktischen Crociduren

(Mammalia, Insectivora, Soricidae). Zool. Abh. st. Mus. Tierk., Dresden 31 (17) : 293-304.
ROBINSON, H. C. 1912. On new mammals from the Islands of the Johore Archipelago, South

China Sea. Ann. Mag. nat. Hist. (8) 10 : 589.
— & KLOSS, C. B. 1911. On new mammals from the Malay Peninsula and adjacent islands.

/. fed. Malay St. Mus. 4 : 243.

1914. On new mammals, mainly from Bandon and the adjacent Islands, East

Coast of the Malay Peninsula. Ann. Mag. nat. Hist. (8) 13 : 232.

1916. A collection of mammals and birds from Pulau Panjang or Pulau Mapor, Rhio

Lingga Archipelago. /. fed. Malay St. Mus. 7 : 63.

VARIATION IN EURASIAN CROCIDURA 309

ROBINSON, H. C. & KLOSS, C. B. 1918. Results of an expedition to Korinchi Peak, Sumatra.

i. Mammals. /. fed. Malay St. Mus. 8 (2) : 1-80.
1922. New mammals from French Indo-China and Siam. Ann. Mag. nat. Hist.

(9) 9 : 87-99-
SHAW, T-H., WANG SUNG, Lu, C-K. & CHANG, L-K. 1966. A survey of the mammals of

Hainan Island, China. Ada zootax. sin. 3 : 260-276.

SNEATH, P. H. A. & SOKAL, R. R. 1973. Numerical taxonomy. San Francisco.
SODY, H. J. V. 1936. Seventeen new generic, specific and subspecific names for Dutch East

Indian mammals. Natuurk. Tijdschr. Ned.-Indie, 96 : 42-55.
SOWERBY, A. DE CARLE. 1917. On a new species of shrew from Corea. Ann. Mag. nat.

Hist. (8) 20 : 318.
SPITZENBERGER, F. 1970. Zur Verbreitung und Systematik tiirkischer Crocidurinae (Insecti-

vora, Mammalia). Annln. naturh. Mus. Wien. 74 : 233-252.

1971- Eine neue, tiergeographisch bemerkenswerte Crocidura (Insectivora, Mammalia)

aus der Tiirkei. Annln. naturh. Mus. Wien 75 : 539-552.

STROGANOV, S. U. 1956. Materials on the systematics of Siberian mammals. Trudy biol.

Inst. zapad.-sib. Fil. Akad. Nauk. SSSR 1 : 3-10.

TAYLOR, E. H. 1934. Philippine land mammals. Monograph 30. Manila.
TEMMINCK, C. J. 1844. In Siebold, F. Fauna Japonica Mammalia.
THOMAS, O. 1889. On the mammals of Christmas Island. Proc. zool. Soc. Lond. 1888 : 532-

534-

1898. Descriptions of new Bornean and Sumatran mammals. Ann. Mag. nat. Hist. (7)

2 : 247.

1906. New Insectivores and Voles collected by Mr. A. Robert near Trebizond. Ann.

Mag. nat. Hist. (7) 17 : 415-421.

1907. On mammals from Northern Persia, presented to the National Museum by Col.

A. C. Bailward. Ann. Mag. nat. Hist. (7) 20 : 196-202.

1912. New species of Crocidura and Petaurista from Yunnan. Ann. Mag. nat. Hist. (8)

9 : 686.

1913. A new shrew from the Andaman Islands. Ann. Mag. nat. Hist. (8) 11 : 468-469.

1919. The white-toothed shrew of Palestine. Ann. Mag. nat. Hist. (9) 3 : 32.

1923. On mammals from the Li-kiang Range, Yunnan, being a further collection obtained

by Mr. George Forrest. Ann. Mag. nat. Hist. (9) 11 : 655-663.

TOKUDA, M. & KANO, T. 1936. A bat and a new shrew from Koto-Sho (Botel Tobago).

Annotnes zool. jap. 15 : 427-432 2 figs.
TOMES, R. F. 1856. In Blyth, E. Memoir on the Indian species of shrews. Ann. Mag. nat.

Hist. (2) 17 : 23.
TROUESSART, E.-L. 1880. Description d'une espece nouvelle de musaraigne de la collection

du Museum de Paris. Naturaliste, 1 (42) : 330.
VAN PEENAN, P. F. D., RYAN, P. F. & LIGHT, R. H. 1969. Preliminary identification manual

for mammals of South Vietnam. Washington.
VINOGRADOV, B. S. 1958. On the structure of external genitalia in white-toothed shrews

(Genus Crocidura, Insectivora, Mammalia) as a classificatory character. Zool. Zh. 37

(8) : 1236-1243.
WANG, S. see SHAW, T-H., WANG SUNG, Lu, C-K. & CHANG, L-K. 1966.

P. D. JENKINS

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON SW7

A LIST OF SUPPLEMENTS
TO THE ZOOLOGICAL SERIES

OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

1. KAY, E. ALISON. Marine Molluscs in the Cuming Collection British Museum
(Natural History) described by William Harper Pease. Pp. 96 ; 14 Plates.
1965. (Out of Print.)

2. WHITEHEAD, P. J. P. The Clupeoid Fishes described by Lacepede, Cuvier and
Valenciennes. Pp. 180 ; n Plates, 15 Text-figures. 1967. £4.

3. TAYLOR, J. D., KENNEDY, W. J. & HALL, A. The Shell Structure and Mineralogy
of the Bivalvia. Introduction. Nuculacea-Trigonacea. Pp. 125 ; 29 Plates,
77 Text-figures. 1969. £4.50.

4. HAYNES, J. R. Cardigan Bay Recent Foraminifera (Cruises of the R.V. Antur)
1962-1964. Pp. 245 ; 33 Plates, 47 Text-figures. 1973. £10.80.

5. WHITEHEAD, P. J. P. The Clupeoid Fishes of the Guianas. Pp. 227 ; 72
Text-figures. 1973. £9-7O-

6. GREENWOOD, P. H. The Cichlid Fishes of Lake Victoria, East Africa : the
Biology and Evolution of a Species Flock. Pp. 134 ; i Plate, 77 Text-figures.
1974. £375. Hardback edition £6.

Printed in Great Britain by John Wright and Sons Ltd. at The Stonebridge Press, Bristol BS4 5NU

. <£- .

"*

BIOLOGY AND FINE STRUCTURED
OF CRYPTODIFFLUGIA OVIFORMIS
(RHIZOPODEA : PROTOZOA)

R. H. HEDLEY, C. G. OGDEN

AND

N. J. MORDAN

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. 8

LONDON: 1977

f , .

BIOLOGY AND FINE STRUCTURE OF

CRYPTODIFFLUGIA OVIFORMIS j&^
(RHIZOPODEA : PROTOZOA)

BY

RONALD HENDERSON HEDLEY, COLIN GERALD OGDEN

AND

NICOLA JANE MORDAN

Pp. 311-328; 6 Plates, 3 Text-figures

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. 8

LONDON: 1977

THE BULLETIN OF THE BRITISH MUSEUM

(NATURAL HISTORY), instituted in 1949, is
issued in five series corresponding to the Scientific
Departments of the Museum, and an Historical
series.

Parts will appear at irregular intervals as they become
ready. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 30, No. 8, of the Zoology series.
The abbreviated titles of periodicals cited follow those
of the World List of Scientific Periodicals.

World List abbreviation :
Bull. Br. Mus. nat. Hist. (Zool.)

ISSN 0007-1498

Trustees of the British Museum (Natural History), 1977

BRITISH MUSEUM (NATURAL HISTORY)

Issued 27 January, 1977 Price £2.75

BIOLOGY AND FINE STRUCTURE OF
, CRYPTODIFFLUGIA OVIFORMIS
(RHIZOPODEA : PROTOZOA)

By R. H. HEDLEY, C. G. OGDEN and N. J. MORDAN

CONTENTS

SYNOPSIS ........... 313

INTRODUCTION .......... 313

Previous work - taxonomy . . . . . . . . 314

Previous work - biology . . . . . . . . 315

MATERIAL AND METHODS . . . . . . . . . 316

DISTRIBUTION ........... 318

SHELL STRUCTURE AND COMPOSITION ....... 318

CYTOPLASM ........... 320

PSEUDOPODIA ........... 322

REPRODUCTION .......... 323

CYST ............ 326

ACKNOWLEDGEMENTS ......... 326

REFERENCES ........... 326

SYNOPSIS

Cryptodifflugia oviformis, a cosmopolitan freshwater, moss and soil inhabiting testacean has
been established in clonal culture with a doubling time of between 29 and 41 hours ; it is
capable of reproducing in a 14 parts per thousand saline solution. The shell is shown to have
two components, a thin outer organic layer and a thick inner, non-crystalline calcareous layer.
The process of reproduction by simple division and the fine structure of the vegetative stage
are described in detail. Ultrastructural features of special interest are the presence of calcareous
inclusions in the mitochondria ; and a storage area for polysaccharide material within the
cytoplasm.

Cryptodifflugia operculata of Page (1966) is considered to be a synonym of Cryptodifflugia
oviformis Penard (1890).

INTRODUCTION

THE main characters used in the classification of testate amoebae are the form,
structure and composition of the shell and the form of the pseudopodia. The
sheDs are generally reported as being either purely organic, arenaceous, or sili-
ceous with occasional reports of forms secreting calcareous shells being unsub-
stantiated. Until relatively recently information on the biology of many testate
amoebae has been known only from infrequent observations on living animals,
usually isolated as individuals during ecological or faunal surveys of freshwater
habitats. The establishment and study of clonal cultures in this laboratory of
certain species with siliceous shells (Hedley & Ogden, 1973, iQ74a ; Hedley et al.,
1974) have contributed to our understanding of their biology and shell structure.

3H R- H. HEDLEY, C. G. OGDEN & N. J. MORDAN

Recent ultrastructural studies on arenaceous forms (Netzel, iQ72a ; Eckert &
McGee-Russell, 1974) and on organic forms (Griffin, 1972 ; Netzel, 19720, I975a, b)
have contributed towards the understanding of their shell formation.

Species attributed to the genus Cryptodifflugia Penard, 1890, are possibly the
smallest and most widely distributed of testate amoebae. There has been some
doubt amongst protozoologists regarding the structure and composition of the
shell in these animals with previous descriptions (Penard, 1890 ; Deflandre, 1953 ;
Grospietsch, 1964 ; Page, 1966) reporting it as chitinous, smooth, hyaline and rigid.

The present work describes some aspects of the biology and shell structure of
Cryptodifflugia oviformis Penard, 1890, an unusual testacean in that it is now reported
as possessing a calcareous shell.

Previous work - taxonomy

The genus Cryptodifflugia was erected by Penard (1890) for the reception of a
new species C. oviformis. Penard (1902) later amended his initial description and
described two further species, C. compressa and C. sacculus. These three species
were redescribed and illustrated by Playfair (1917) with five new species, C. minuta,
C. angulata, C. valida, C. crenulata and C. pusilla, and four varieties, all from ponds
or swamps in Australia. Zacharias (1903) reported a new species, C. turfacea,
which sometimes had two apertures each having a distinct rim or collar. Another
new species, C. voigti, described by Schmidt (1926) was observed to have an apertural
rim and particles adhering to the shell surface.

Deflandre (1953) emended the generic diagnosis of both Cryptodifflugia Penard,
1890 and Difflugiella Cash, 1904, using the degree of compression of the shell as a
main character. These emendations included the transfer of the type species,
C. oviformis, of the genus Cryptodifflugia to the genus Difflugiella. Several authors,
Bonnet & Thomas (1955, 1960), Thomas (1959), Grospietsch (1964) and Schonborn
(1965), have agreed with Deflandre's changes. For example, Thomas (1959)
redescribed D. oviformis (Penard, 1890) and tabulated the previously reported
measurements of this species. In a review of the genera Cryptodifflugia and Difflu-
giella, Grospietsch (1964) agreed with Deflandre (1953) and gave keys to the species
of both genera. Whilst Schonborn (1965) described four new species of Difflu-
giella and redescribed D. crenulata (Playfair, 1917), D. oviformis (Penard, 1890)
and D. voigti (Schmidt, 1926).

Page (1966) reviewed the genus Cryptodifflugia and concluded that Deflandre's
emendations were unjustified. He based his opinion on the original descriptions
of Difflugiella and Cryptodifflugia which showed the former to have a flexible shell,
whilst the latter had a rigid shell. Recent ultrastructural studies on Difflugiella
sp. by Griffin (1972), and the present work, support the view that this feature
is a good taxonomic character.

Page (1966) used the presence of an operculum in the aperture of encysted animals
of a new species C. operculata to distinguish it from C. oviformis. This feature
had not been previously described and it is difficult to determine whether its omission
in earlier descriptions was due to oversight, as the operculum is often difficult to

BIOLOGY OF CRYPTODIFFLUGIA OVIFORMIS 315

see with the optical microscope, or to the absence of encysted animals in the material
examined by other authors.

As a result of an examination of the two available cultures of C. operculata and
C. oviformis (from this laboratory), Dr Page has suggested in a personal communi-
cation that the presence of an operculum may not be a good specific character, as it
appears to be present in most encysted animals. Our own studies show that the
structure of the shell and the fine structure of the cytoplasm in the two forms
are indistinguishable. It is suggested therefore that C. operculata be regarded as
a synonym of C. oviformis.

Previous work - biology

In his original description of Cryptodifflugia oviformis, Penard (1890) reported
that the cytoplasm rilled the shell and contained one nucleus, a contractile vacuole
close to the nucleus and 'numerous brilliant granules'. Similar 'brilliant' or 'coarse'
granules found in C. operculata have been shown by Page (1966) to react positively
with supravital Janus green from which he concluded that they are mitochondria.

Penard (1890) observed that the form of the small number of pseudopodia of C.
oviformis changed rapidly and later (Penard, 1902) described the pseudopodia as
straight structures without ramifications. The arrangement and length of the
pseudopodia in C. oviformis are described by Schonborn (1965), who also observed
that the rate of locomotion varied from 34 to 40 (jim/min. According to Page
(1966) there are between two and five pseudopodia in C. operculata that extend
from a mass of cytoplasm outside the aperture. Both this mass and the cytoplasm
appear to be free from coarse granules. Although these pseudopodia appear to
be filose all seem to have rounded ends. The rate of locomotion of ten individuals
ranged from 14 (jurn/min to 28 [jun/min.

United individuals were described as 'conjugating tests' by Penard (1902) in
which one shell always appears empty whilst the other shell, usually the larger,
contains a rounded mass of cytoplasm separated from the aperture by a 'chitinous'
diaphragm. Penard suggests that this could be a resting stage during which the
animal constructs a new shell that gradually hardens. Playfair (1917) has described
similar united individuals of C. oviformis as encysted forms. The process of binary
fission has been described for C. operculata by Page (1966), in which most of the
cytoplasm to form the daughter cell is extruded before nuclear division begins
and is followed by the formation of a new shell around the daughter. Nuclear
division from prophase to anaphase takes place in the aboral region of the parent
and the daughter nucleus moves into the extruded cytoplasm in late telophase.
Both the nuclear membrane and the nucleolus disappear before metaphase and no
centrioles are seen. It also appears that most of the granules pass into the extruded
cytoplasm and few are left in the parent.

Penard (1890, 1902) described an encysted form of C. oviformis in which the cyto-
plasm was concentrated and contained in the aboral region behind a cyst-membrane.
A similar cyst has been described by Schonborn (1965). Page (1966) reported this
type of cyst only once in C. operculata, but regularly found cysts in which the

316 R. H. HEDLEY, C. G. OGDEN & N. J. MORDAN

aperture was sealed. In these latter forms the cytoplasm filled the shell, the
contractile vacuole disappears and the aperture is blocked by a lens-shaped oper-
culum. At encystment the operculum is either rejected from the shell or ingested
within the cytoplasm.

MATERIALS AND METHODS

Cryptodifflugia oviformis was isolated from a sample of moss, Eurhynchium
praelongum, taken from the bark of a tree at Cilpost Farm, Whitland, Carmarthen-
shire, Wales, in July 1968. Rough cultures were obtained by aUowing small
portions of this sample to stand, covered by a shallow layer of culture medium, at
room temperature 18-20 °C. Agnotobiotic cultures were kept in small plastic
containers on a thin substrate of agar (i per cent agar agar in distilled water) with
a sterilized wheat grain added prior to setting and covered with a layer of culture
medium. The medium was a 5 per cent (w/v) solution of sterilized soil extract in
distilled water. Clonal cultures were established by isolating single active animals,
and one such clone has been used subsequently to produce the working cultures.
This clone has been deposited with, and is now maintained at, the Culture Centre
of Algae and Protozoa, The Natural Environment Research Council, Cambridge,
England (Reg. No. 1514/2).

Specimens of C. operculata obtained from the Culture Centre for Algae
and Protozoa (Reg. No. 1514/1), which were initially isolated by Page (1966),
have been maintained in a manner similar to C. oviformis. Under these conditions
C. oviformis remains active for approximately 4 weeks, after this time mobility
decreases and encysted forms predominate. Cultures may also be maintained
on an agar agar substrate, streaked with the bacterium Klebsiella pneumoniae,
but without a covering fluid. Such cultures, kept on agar slopes in closed universal
bottles at 9 °C, may be sub-cultured successfully after 6 months (Dr F. C. Page,
personal communication) .

Cultures have been established using a graded (v/v) series of seawater to fresh-
water culture medium as the overlying culture fluid to the normal agar substrate.
The initial experiments were made with i, 2, 5 and 10 per cent seawater to fresh-
water mixtures, and in all cases active reproducing animals were observed. In
subsequent experiments with higher concentrations of seawater it was found that
cultures could be maintained in media up to and including 40 per cent seawater,
which is the equivalent of a 14 parts per thousand saline solution.

Optical microscopy
Live animals were examined by both phase-contrast and bright-field illumination.

Scanning electron microscopy

Specimens taken from 4 or 5 weeks old cultures were cleaned by taking them
individually through several changes of distilled water, using a single-hair brush.

BIOLOGY OF CRYPTODIFFLUGIA OVIFORMIS 317

They were allowed to dry and then transferred onto a previously cleaned cover-slip.
These prepared cover-slips were mounted on a standard Stereoscan stub using an
electrically conductive paint, 'Silver Dag', prior to being coated evenly with gold.
The stubs were examined on a Cambridge Stereoscan Mkll operating at 10 kV and
the results recorded on Ilford H?4 film.

Preparations to display pseudopodia were obtained by fixing specimens in situ
using the following method. The agar substrate in a culture vessel was wetted
with a minimal quantity of culture medium, and a small number of specimens
from a fresh culture were placed on this surface. After a short time the animals
resumed their normal activities. The vessel was then quickly flooded with 3 per
cent glutaraldehyde in cacodylic acid buffer, and left for 15 minutes. The agar
surface was washed with the same buffer and selected squares were cut out and
allowed to air dry. These squares were mounted on a standard Stereoscan stub
using Araldite and processed as previously described.

Specimens from a culture vessel were air dried on pieces of spectographically
standardized carbon rod for the analysis of inorganic chemical elements. Previous
trials with this carbon rod had shown that it gave a minimal elemental background
count. The pieces of rod were attached to Stereoscan stubs with Araldite. Stubs
were then examined in a Cambridge Stereoscan coupled to an Ortec solid-state
energy dispersive X-ray analyser, operating at 15 kV for 400 or 1000 seconds.

Electron probe analysis was carried out on specimens taken from a culture vessel
and allowed to air dry onto a Stereoscan stub with a short shaft. They were
coated with a thin layer of carbon and examined in a modified Cambridge Geoscan,
operating at 20 kV with a specimen current of o-i x io~7 amps.

Transmission electron microscopy

Specimens taken from cultures, approximately 2 or 3 weeks old, were fixed at
room temperature for 12 minutes in i per cent glutaraldehyde in 0-025 M cacodylic
acid buffer, followed by 7 minutes in 3 per cent glutaraldehyde in the same buffer.
After several rinses in o-i M cacodylic acid buffer, they were post-fixed for 12
minutes in i per cent osmium tetroxide in distilled water. Some specimens were
conventionally fixed, but before dehydration were treated with a chelating agent,
5% EDTA, for 15 minutes. The material was dehydrated, and embedded in
Epon 812. Sections were cut on a Porter Blum MT2 ultramicrotome using a
Du Pont diamond knife, stained in a saturated alcoholic solution of uranyl acetate
and Reynold's lead citrate, and examined in an AEI 6B electron microscope operating
at 60 kV. The results were recorded on Ilford EM6 plates.

The periodic acid/thiosemicarbazide/silver proteinate technique (Thiery, 1967)
was carried out on thin sections mounted on gold grids. After treatment with
i per cent periodic acid for 20 minutes at room temperature followed by several
washes in distilled water, the sections were left for either 40 minutes or 18 hours
in a i per cent solution of thiosemicarbazide in 10 per cent acetic acid. They
were washed thoroughly in several changes of 10 per cent acetic acid followed by
several rinses in distilled water, before treatment with i per cent solution of silver

3i8 R. H. HEDLEY, C. G. OGDEN & N. J. MORDAN

proteinate for 30 minutes in the dark. The grids were washed in distilled water
before examination.

Thick, unstained sections, coated with carbon, were analysed for elemental
composition in an AEI EMMA-4 analytical electron microscope, equipped with a
minilens, operating at 80 kV with a probe size of 0-06 [zA.

DISTRIBUTION

C. oviformis is commonly found in damp and wet mosses, standing water, on
aquatic plants, various soils and forest litter. A list of locality records and references
was given by Bonnet & Thomas (1960). Additional localities were given by several
authors, Hoogenraad & Groot (1952), Stepanek (1963) and Bonnet (1966).

The following list shows the geographically widespread distribution :
EUROPE : England, Wales, Scotland, Ireland, France, Belgium, Holland,

Germany, Iceland, Lapland, Italy, Czechoslovakia, Hungary,
Bulgaria, Russia.

NORTH AMERICA : United States of America and Canada.
SOUTH AMERICA : Columbia and Chile.
AFRICA : Cameroun and the Congo.

ASIA : Siberia.

AUSTRALASIA : Australia and New Zealand.

SHELL STRUCTURE AND COMPOSITION

Based on measurements from a hundred specimens the ovoid, rigid and smooth
shells of C. oviformis vary in length from 14-5 to 22-2 (j,m and in diameter from
12-8 to 17-6 (Am. The single, terminal and circular aperture has a diameter of
between 3-2 and 6-4 pun and is surrounded by a thin collar (PL i, fig. A).

Ultrastructural observations show that the shell wall varies in thickness, between
individuals, from 150 to 260 nm and that it is divided into two distinct layers.
A thin outer organic layer, which is usually no more than 20 nm thick, and a thick
inner layer of electron-dense calcined material. These layers appear to have a
uniform thickness over most of the shell surface (Fig. i), but in the apertural region
there is a slight thickening of the calcified layer and an extension of the organic
layer to form a circular collar (Fig. 2). This organic collar is strengthened at the
internal face by a continuation of the calcified layer (PI. i, fig. B). Examination
of the complete shell wall at the ultrastructural level is difficult because the calcified
layer is not penetrated by the embedding resin, and as a result is either fractured
or torn in thin sections (PI. 2).

Whole shells, examined with an Ortec X-ray analytical attachment to the Stereo-
scan, showed significant elemental peaks for silicon, potassium and calcium, the
last mentioned being pronounced. Semi-quantitative electron probe microanalysis
of shells indicated that the calcium count-rate was high, and demonstrated a weak
line for silicon corresponding to not more than i per cent. Sectioned material
examined with the X-ray analytical electron microscope, EMMA, using a standard
of calcium carbonate, gave significant intensities from the calcified layer. This

BIOLOGY OF CRYPTODIFFLUGIA OVIFORMIS

319

microbody

contractile
vacuole

nucleus

'storage' area
Golgi

mitochondrion

dense endoplasmic

reticulum

food vacuoles

FIG. i. Diagram of the vegetative stage of C. oviformis showing the arrangement of the

main organelles.

organic collar

organic layer
calcareous layer

- shell wall

FIG. 2. Diagram of the apertural collar with a section cut away to show the relationship
of the organic and calcareous layers.

320 R. H. HEDLEY, C. G. OGDEN & N. J. MORDAN

layer also gave significant intensities for phosphorus, using sodium dihydrogen
orthophosphate as a standard. Further attempts to analyse the composition
of the calcified layer, using whole shells for X-ray diffraction and sectioned specimens
for electron diffraction, were inconclusive but suggest that this material is mainly
amorphous. Dr K. M. Towe (Smithsonian Institution, Washington, D.C., U.S.A.),
who kindly undertook the electron diffraction analysis, suggested that the material
was either amorphous or exceptionally beam-sensitive as it was unstable. He
ruled out the possibility of the beam sensitivity being associated with the presence
of an oxalate because there was no inversion to calcium oxide. In conclusion, we
suggest that in all probability the calcified layer is amorphous calcium phosphate.

When whole specimens are immersed in either dilute nitric or hydrochloric acids
or with 5% EDTA the major part of the shell dissolves. That which remains
is the thin organic layer, delimiting the shape of the shell, and the organic thick-
ening in the apertural region. In an attempt to demonstrate the fine structure
of the organic layer, animals were fixed normally but before dehydration were
decalcified by treatment with 5% EDTA for 10 minutes (PL i, fig. C). The organic
layer revealed by this method is a membrane-like structure which is thickened
and folded at the aperture to form a collar (PL i, fig. D). Specimens stained with
the periodic acid/thiosemicarbazide/silver proteinate technique to demonstrate
reactive hydroxyl groups show that the organic layer is composed of two thin,
closely opposed, distinct membrane-like structures (PL i, fig. E), with a high carbo-
hydrate content.

Abnormal forms occur in cultures and represent about 3 per cent of the population.
These are usually large forms, two or three times the size of a normal animal, and
have two or more apertures (PL i, fig. F). The shell of these animals is characterized
by an uneven distribution and thickness of the outer organic covering (PL i, fig. F ;
PL 6, fig. C).

CYTOPLASM

The cytoplasm normally occupies the whole of the shell cavity (Fig. i ; PL 2)
and is enclosed by a plasmalemma. The single nucleus is normally spherical,
between 4-3 and 5-1 (xm in diameter, and is enclosed by two tripartite membranes.
The nuclear matrix is granular with small, densely stained concentrations of
chromatin scattered throughout and a dense nucleolus that is usually situated
centrally. The nucleus is surrounded by a densely stained mass of granular endo-
plasmic reticulum (Fig. i). This region of endoplasmic reticulum appears more
electron-dense than the surrounding cytoplasm due to the concentration of
ribosomes. Two or three contractile vacuoles may be present in the mid-body
region (PL 3, fig. A). The area surrounding these vacuoles is free of ribosomes
but has numerous small vesicles which are often continuous with the lumen of the
vacuole. The contractile vacuoles lie at the periphery of the cytoplasm and dis-
charge directly into the shell cavity. Specimens in 2-week-old cultures discharged
their contractile vacuoles at periods that ranged from 25 to 33 seconds between
systole and diastole. These periods increased in animals from 6-week-old cultures

BIOLOGY OF CRYPTODIFFLUGIA OVIFORMIS 321

to range from 28 to 41 seconds. Numerous food vacuoles, which have a single
unit membrane and may contain food organisms or waste material, occupy the
apertural end of the cytoplasm (PI. 2).

The mitochondria are ovoid or spherical, with tubular cristae, a dense granular
matrix and they are distributed at random throughout the cytoplasm. Frequently
large, spherical, electron-dense inclusions occur between the outer and inner mito-
chondrial membranes (PL 3, figs A and B). These inclusions are naturally electron-
dense in sections from specimens fixed in glutaraldehyde but without subsequent
treatment with heavy metals, and show similar fracture patterns to those some-
times seen in the electron-dense shell wall. We believe that these mitochondrial
inclusions are the 'brilliant' or 'coarse' granules described by earlier workers.
Analysis of these inclusions using the X-ray analytical electron microscope EMMA
gave significant intensities for calcium and phosphorus which correspond to those
produced by the shell. These calcareous inclusions dissolve in 5% EDTA.

As many as 6 Golgi bodies may be present in one animal and they are usually
associated with concentrations of granular endoplasmic reticulum. Each dictyo-
some is composed of a stack of 3 to 6 saccules (PI. 3, fig. C) which are usually concave
and centrally flattened, whilst the matrix around these saccules may be noticeably
less dense than the surrounding cytoplasm. Condensations of a substance, which
is electron-dense with normal staining procedures, occur in the central areas of the
saccules. This material appears as if it may pass to the outer margins of
the saccules where they form spherical protrusions, which are budded off as smooth
membrane-bound vesicles (PI. 3, fig. C). At the inner face of the dictyosome,
vesicles containing an electron-lucent material are formed (PI. 3, fig. E). These
vesicles are slightly larger than those containing electron-dense material and are
lined by a thin layer of an electron-dense material (PI. 3, fig. D). A third type of
vesicle is often present in the dense granular endoplasmic reticulum (PI. 4, fig. A).
These vesicles are large, usually have electron-lucent contents, but are occasionally
seen to contain an electron-dense material (PI. 4, fig. D), and are often surrounded
by a band of endoplasmic reticulum. Similar vesicles to these have been previously
reported in Trinema linear e (Hedley & Ogden, I974a).

On the basis of their distribution within the cytoplasm it is postulated that these
three types of vesicles move through the cytoplasm along two distinct pathways.
The large electron-lucent vesicles would pass through the central cytoplasm, whilst
the smaller vesicles would pass around the less dense peripheral region to become
incorporated at sites that are considered to be possible storage areas (PI. 4, figs B
and D). In the trophic animal there may be as many as three such storage
areas, usually found on a level or posterior to the nucleus in the aboral region
(PI. 4, fig. C). These regions are electron-lucent and do not appear to have any
limiting membrane.

When sections are treated with the periodic acid/thiosemicarbazide/silver
proteinate technique the storage areas stain strongly, which might suggest that they
are regions rich in carbodyrate. The small electron-lucent vesicles also contain
the stain, but in these it is confined to the lining of the vesicles (PI. 4, fig. B). The
saccules of the Golgi apparatus are sometimes stained, as are the vesicles that are

322 R. H. HEDLEY, C. G. OGDEN & N. J. MORDAN

detached at their margins, but this staining is less positive (PL 4, fig. B). Similar
results after using this technique were obtained from specimens which had been
decalcified with EDTA prior to embedding.

Microbodies are present in most of the specimens examined. They are oval or
spherical, varying between 0-25 and 0-44 [xm in diameter, have a dense matrix
and a single unit membrane. They may contain either one or two tubular-like
elements, between 19 and 24 nm in diameter (PL 4, fig. E), or a rectangular structure
of parallel lines with a spatial distance of 8 nm between each line (PL 4, fig. F).
Similar tubular-like elements in microbodies have been reported in the testate
amoeba Trinema lineare (Hedley & Ogden, iQ74a), and in various Foraminifera by
Hedley & Wakefield (in Hruban & Rechcigl, 1969) and Febvre-Chevalier (1971).
Lattice-like structures within microbodies have been previously described in the
Foraminifera Shepheardella taeniformis and Globigerina bulloides by Hedley et al.
(1967) and Febvre-Chevalier (1971).

PSEUDOPODIA

In recent classifications (Loeblich & Tappan, 1961 ; Honigberg et al., 1964)
the class Rhizopodea has been divided - using the form of the pseudopodia as the
distinguishing character - into three subclasses ; the Lobosia, Filosia and Granulo-
reticulosa. In both classifications the superfamily Cryptodiffmgiacea is included
in the sub-class Lobosia. Previous studies (Penard, 1902 ; Page, 1966) demonstrate
that the pseudopodia of C. oviformis appear to be intermediate in form between
lobose and filose structures. Recognizing this difficulty, Deflandre (1953) erected
a new sub-order the Reticulolobosa, but this has not been adopted in subsequent
classifications.

In actively moving specimens of C. oviformis the pseudopodia are numerous,
relatively long and appear to taper and have small branches. All these features are
normally associated with filose pseudopodia (PL 5, fig. A).

Micrographs of whole animals which had been found with the pseudopodia
extended (PL 5, fig. C) indicate that in addition to the main filose pseudopodia as
seen with optical microscopy there are numerous ultrastructural pseudopodia or
pseudopodial strands. These may or may not be connected to form a reticulum.
These ultrastructural pseudopodia are similar to those seen in the granuloreticulose
pseudopodia described for various foraminiferans by Hedley et al. (1967), Marszalek
(1969) and Lengsfeld (1969).

Each pseudopodium appears to consist of ground-plasm, limited by a unit mem-
brane, often containing bands of thick microfilaments and small electron-dense
concentrations (PL 5, fig. B). In addition, some sections of pseudopodia contain
strands of endoplasmic reticulum, Golgi bodies, membrane-bound vesicles and
mitochondria, some of which may contain electron-dense calcium inclusions. The
bands of thick microfilaments vary from 25 to 33 nm in diameter and are about
0-4 |zm long (PL 5, fig. B). Microfilaments are sometimes found concentrated to
form adhesion plaques at points of contact between opposing pseudopodia (PL 5,
fig. D). We suggest that this reaction is probably restricted to contact between

BIOLOGY OF CRYPTODIFFLUGIA OVIFORMIS 323

pseudopodia from different animals. Similar adhesion plaques have been reported
in species of Euglypha by Hedley & Ogden (1973, 19740).

The fine structure of the pseudopodia of C. oviformis is similar to that reported
for other testate amoebae (Wohlman & Allen, 1968 ; Griffin, 1972 ; Eckert &
McGee-Russell, 1973 ; Hedley & Ogden, 1973, I974a) and naked amoebae (Comly,
1973 ; Taylor et al., 1973 ; Rinaldi & Hrebenda, 1975) in having microfilaments
as a structural element. Wohlman & Allen (1968) and Eckert & McGee-Russell
(1973) suggest that both thin and thick microfilaments are associated with
pseudopodia extension and contraction in Difflugia, and similar studies on slime
moulds and naked amoebae by Komnick et al. (1972, 1973) suggest that the uptake
and release of calcium ions by microfilaments plays an important role in cytoplasmic
streaming. Another role for both types of microfilaments, that of cellular wound
healing, has recently been suggested by Jeon & Jeon (1975) from studies on Amoeba
proteus.

REPRODUCTION

The length of time required to double the population (doubling time) was esti-
mated from three identical cultures established and maintained under similar
conditions. Growth curves produced from observations made at regular intervals
showed that the doubling time is between 29 and 41 hours.

The formation of a daughter cell by binary fission begins with an elongation
of the cytoplasm of the parent shell to form a thick trunk which emerges from the
aperture (Fig. 3-1). This trunk expands until the cytoplasm has reached about
half the size of the parent shell. There is then a slight contraction, followed by
further expansion until it reaches a size equivalent to that of the parent shell.
At about this time the first indication of the transfer of cytoplasmic organelles
becomes apparent, with the rapid passage of the dense granules into the daughter.
The parent cytoplasm contracts slightly and is less active for a period of approxi-
mately i minute. The daughter then collapses slightly and the granules return to
the parent. The rapid movement of the granules between the parent and daughter
is repeated, and finally most of the granules are passed into the daughter with
only one or two remaining in the parent (Fig. 3-6). The cytoplasm in the parent
is reduced, to about one-third its normal volume and is situated at the aboral
region of the shell, whilst the daughter cell appears to have reached full size. The
connection between the two animals is reduced to a thin thread (Fig. 3-8). As the
parental cytoplasm moves towards the apertural region of the shell, the connection
between the parent and daughter is severed. Pseudopodia emerge from the parent
and the two animals separate, the daughter remaining quiescent for periods in excess
of 45 minutes from the time of separation.

The process of granular exchange takes place between 15 and 25 minutes after
the first stages of division. Contractile vacuoles appear in the daughter after about
20 minutes, and the nuclei appear in both animals after about 30-35 minutes.

Often animals appear to start dividing only to cease at various stages, including
stages that are almost half the parental size. These abortive products are rejected

R. H. HEDLEY, C. G. OGDEN & N. J. MORDAN

start

22 mins.

15 mins.

8

23 mins.

16-5 mins.

24 mins.

10

1 7 mins.

# 29 mins.

11

18 mins.

^^-^' 35 mins.

19 mins.

FIG. 3. Diagram showing the progressive stages in the formation of a daughter cell. At
stage i the animal appears to rest for up to 15 minutes. Stages 2-7 represent the cyto-
plasmic contractions and rapid interchange of dense granules between the parent and
daughter. Stage 5 shows the first indication of a thin shell (pecked line) surrounding the
daughter, although this is probably present at an earlier stage. By stage 9 the cytoplasm
has been divided between the parent and daughter and at stage 12 the parent moves away.

BIOLOGY OF CRYPTODIFFLUGIA OVIFORMIS 325

by the parent to become isolated, rounded lumps of cytoplasm, which are often
seen in the cultures.

The first indication of a shell surrounding the newly formed daughter cell is a
thin organic layer, indications of which are first seen after the start of granular
exchange between parent and daughter. Nevertheless, it is possible that this
organic layer is present around the initial extrusion of cytoplasm at the beginning
of division. The formation of the shell in C. oviformis is probably dependent on
the accumulation within the parent of reservoirs of an acid mucopolysaccharide
and calcium ions. The organic material we believe is produced by both the Golgi
complexes and the dense endoplasmic reticulum, and is stored in the cytoplasm at
the storage areas (PI. 3, fig. A ; PL 4, fig. B). The fusion of vesicles to form storage
areas appears to represent a process of intake and assimilation. There is no evidence
to suggest how this material is distributed from this reservoir, but the storage areas
are absent in young daughter cells. A similar aggregation of vesicles rich in poly-
saccharides has been demonstrated by Holtrop (1972) in cartilage cells of the epiphy-
seal plate, where the protein polysaccharide complex was considered to assemble
before it was extruded into the matrix.

Calcification of the shell appears to start shortly after division. During the early
stages of calcification the shells often have an irregular outline (PL 6, fig. B), and
they appear to be flexible. This initial shell wall has a diffuse appearance, with
some small concentrations of electron-dense material on the inner surface (PL 5,
fig. E). When the parent and daughter cells have finally divided they separate
and the aperture of the daughter is sealed by a thin, fibrous organic membrane
(PL 6, fig. A). In the cytoplasm of the daughter cell the usual membrane-limited
profiles of endoplasmic reticulum are absent and it appears to be a mass of free
ribosomes. The small circles of ribosomes that are seen (PL 5, fig. E) are possibly
the initial stages in the reformation of these profiles.

In C. oviformis it has been possible to locate only one site where calcium is con-
centrated. This is between the outer and inner mitochondrial membranes where
it is stored as spherical inclusions of amorphous calcium phosphate (PL 3, fig. B).
Most of these mitochondria pass at division into the daughter cell. Several young
animals examined shortly after division contain mitochondria without inclusions
and in such cases the matrix of the mitochondria appears to be homogeneous, the
cristae are not readily resolved, and the outer mitochondrial membrane is seen to
have small swellings (PL 5, fig. F). It has been established by Wasserman &
Kallfelz (1970) that calcium ions are actively transported across mitochondrial
membranes, and a model for this transport has been proposed by Rasmussen (1966).
Previous investigations (Greenawalt et al., 1964 ; Greenawalt & Carafoli, 1966 ;
Thomas & Greenawalt, 1968) on calcium-rich granules in mitochondria suggest
that formation takes place in three stages. First at the inner membrane surfaces,
especially on the cristae, followed by free granules in the matrix and finally the
granules grow by accumulation.

Opinions differ over the factors that induce concentrations of calcium in the
mitochondria of the ciliated protozoan Spirostomum ambiguum, where they have
been considered to be a mobile endoskeleton (Bien, 1967), a metabolic store of

326 R. H. HEDLEY, C. G. OGDEN & N. J. MORDAN

phosphate (Pautard, 1970), or concentrations of waste calcium (Jones, 1967, 1969).
But in vertebrates it has been suggested that similar mitochondrial inclusions to
those demonstrated here in C. oviformis are involved in mineralization of deer antlers
(Sayegh et al., 1974) and eggshell formation in chickens (Hohman & Schraer, 1966).

CYST

After 4 or 5 weeks encysted animals appear in cultures. Such animals have a
distinct operculum forming a seal across the aperture (PL 6, fig. E) and the ultra-
structure of this operculum is a finely fibrillar matrix interspersed with small
naturally electron-dense bodies (PI. 6, fig. D). Results from periodic acid-Schiff
tests carried out by Page (1966) suggest that the operculum is mainly of a poly-
saccharide nature.

ACKNOWLEDGEMENTS

We would like to thank Dr F. Page of the Culture Centre for Algae and Protozoa,
NERC, Cambridge, for his helpful discussions on the identification of C. oviformis,
Dr K. M. Towe (Smithsonian Institution, Washington, D.C., U.S.A.) who offered
observations after electron diffraction examination of the shell, and our colleague
Dr C. R. Curds who kindly read and discussed the typescript.

REFERENCES

BIEN, S. M. 1967. High hydrostatic pressure effects on Spirostomum ambiguum. Calc.

Tiss. Res. 1 : 170-172.
BONNET, L. 1966. Le peuplement thecamoebien de quelques sols du Chili (I). Protistologica

2 : 113-139.

— & THOMAS, R. 1955. Etude sur les Thecamoebiens du sol (I). Bull. Soc. Hist. not.
Toulouse 90 : 411-428.

1960. Faune terrestre et d'eau douce des Pyrenees-Orientales. Thecamoebiens
du sol. Hermann, Paris. 103 pp.

COMLY, L. T. 1973. Microfilaments in Chaos carolinensis. Membrane association, distri-
bution, and heavy meromyosin binding in the glycerinated cell. /. Cell Biol. 58 : 230-237.

DEFLANDRE, G. 1953. Ordres des Testacealobosa, Testacaefilosa, Thalamia ou Thecamoe-
biens (Rhizopoda Testacea) . In Traiti de Zoologie, Vol. I, Fasc. II : 97-148. Masson, Paris.

ECKERT, B. S. & McGEE-RussELL, S. M. 1973- The patterned organization of thick and
thin microfilaments in the contracting pseudopod of Difflugia. J. Cell Sci. 13 : 727-739.
1974. Shell structure in Difflugia lobostoma observed by scanning and trans-
mission electron microscopy. Tissue Cell 6 : 215-221.

FEBVRE-CHEVALIER, C. 1971 . Constitution ultrastructurale de Globigerina bulloides d'Orbigny,
1826 (Rhizopoda-Foraminifera). Protistologica 7 : 311-324.

GREENAWALT, J. W. & CARAFOLI, E. 1966. Electron microscope studies on the active
accumulation of Sr++ by rat-liver mitochondria. /. Cell Biol. 29 : 37-61.

— Rossi, C. S. & LEHNINGER, A. L. 1964. Effect of accumulation of calcium and phos-
phate ions on the structure of rat liver mitochondria. /. Cell Biol. 23 : 21-38.

GRIFFIN, J. L. 1972. Movement, fine structure, and fusion of pseudopods of an enclosed

amoeba, Difflugiella sp. /. Cell Sci. 10 : 563-583.
GROSPIETSCH, TH. 1964. Die Gattungen Cryptodifflugia und Difflugiella (Rhizopoda testacea).

Zool. Anz. 172 : 243-257.

BIOLOGY OF CRYPTODIFFLUGIA OVIFORMIS 327

HEDLEY, R. H. & OGDEN, C. G. 1973- Biology and fine structure of Euglypha rotunda
(Testacea: Protozoa). Bull. Br. Mus. nat. Hist. (Zool.) 25: 119-137.

— I974&. Observations on Trinema lineare Penard (Testacea: Protozoa). Bull.
Br. Mus. nat. Hist. (Zool.) 26 : 187-199.

19740. Adhesion plaques associated with the production of a daughter cell in

Euglypha (Testacea; Protozoa). Cell. Tiss Res. 153 : 261-268.
& KRAFFT, J.I. 1974. Observations on clonal cultures of Euglypha acanthophora

and Euglypha strigosa (Testacea: Protozoa). Bull. Br. Mus. nat. Hist. (Zool.) 27 : 103-111.
PARRY, D. M. & WAKEFIELD, J. ST J. 1967. Fine structure of Shepheardella taeniformis

(Foraminifera: Protozoa). // R. microsoc. Soc. 87 : 445-456.
HOHMAN, W. & SCHRAER, H. 1966. The intracellular distribution of calcium in the mucosa

of the avian shell gland. /. Cell Biol. 30 : 317-331.
HOLTROP, M. E. 1972. The ultrastructure of the epiphyseal plate. I. The flattened

chondrocyte. Calc. Tiss. Res. 9 : 131-139.
HONIGBERG, B. M., BALAMUTH, W., BOVEE, E. C., CORLISS, J. O., GOJDICS, M., HALL, R. P.,

KUDO, R. R., LEVINE, N. D., LOEBLICH, A. R., WEISER, J. & WENRICH, D. H. 1964.

A revised classification of the phylum Protozoa. /. Protozool. 11 : 7-20.
HOOGENRAAD, H. R. & GROOT, A. A. DE 1952. Thekamobe Moosrhizopoden aus Asien.

Arch. Hydrobiol. 47 : 263-287.

HRUBAN, Z. & RECHCIGL, M. 1969. Microbodies and related particles : Morphology, bio-
chemistry, and physiology. Int. Rev. Cytol. Suppl. i. Academic Press, London. 296 pp.
JEON, K. W. & JEON, M. S. 1975. Cytoplasmic filaments and cellular wound healing in

Amoeba proteus. J. Cell Biol. 67 : 243-249.
JONES, A. R. 1967. Calcium and phosphorus accumulation in Spirostomum ambiguum.

J. Protozool. 14 : 220-225.

1969. Mitochondria, calcification and waste disposal. Calc. Tiss. Res. 3 : 363-365.
KOMNICK, H., STOCKEM, W. & WOHLFARTH-BOTTERMANN, K. E. 1972. Ursachen, Begleit-

phanomere und Steuerung zellularer Bewegungserscheinungen. Fortschr. Zool. 21 : 1-74.
— 1973- Cell motility : Mechanisms in protoplasmic streaming and ameboid

movement. Int. Rev. Cytol. 34 : 169-249.
LENGSFELD, A. M. 1969. Zum Feinbau der Foraminifere Allogromia laticollaris. I. Zellen

mit ausgestreckten und eingezogenen Rhizopodien. Helgolander wiss. Meeresunters

19 : 230-261.
LOEBLICH, A. R. & TAPPAN, H. 1961. Suprageneric classification of the Rhizopodea. /.

Paleont. 35 : 245-330.
MARSZALEK, D. S. 1969. Observations on Iridia diaphana, a marine foraminifer. /. Protozool.

16 : 599-611.
NETZEL, H. I972a. Die Schalenbildung bei Difflugia oviformis (Rhizopoda, Testacea).

Z. Zellforsch. 135 : 55-61.

I972b. Die Bildung der Gehausewand bei der Thekamobe Centropyxis discoides (Rhizo-
poda, Testacea). Z. Zellforsch. 135 : 45-54.
— i975a. Struktur und Ultrastruktur von Arcella vulgaris var. multinucleata (Rhizopoda,

Testacea). Arch. Protistenk. 117 : 219-245.
- I975b. Morphologic und Ultrastruktur von Centropyxis discoides (Rhizopoda, Testacea).

Arch. Protistenk. 117 : 369-392.
PAGE, F. C. 1966. Cryptodifflugia operculata n. sp. (Rhizopodea: Arcellinida, Cryptodifflu-

giidae) and the status of the genus Cryptodifflugia. Trans. Am. microsoc. Soc. 85 : 506-515.
PAUTARD, F. G. E. 1970. Calcification in unicellular organisms. In Biological Calcification :

Cellular and Molecular Aspects, pp. 105-201. North-Holland, Amsterdam.
PENARD, E. 1890. Etudes sue les Rhizopodes d'eau douce. Mem. Soc. Phys. Hist. nat.

Geneve 31 : 1-230.

— 1902. Faune Rhizopodique du Bassin du Ldman. Geneva. 700 pp.
PLAYFAIR, G. J. 1917. Rhizopods of Sidney and Lismore. Proc. Linn. Soc. N.S.W. 42 :

328 R. H. HEDLEY, C. G. OGDEN & N. J. MORDAN

RASMUSSEN, H. 1966. Mitochondrial ion transport : Mechanism and physiological signi-
ficance. Fed. Proc. 25 : 903-911.

RINALDI, R. A. & HREBENDA, R. 1975. Oriented thick and thin filaments in Amoeba proteus.
J. Cell Biol. 66 : 193-198.

SAYEGH, F. S., DAVIS, R. W. & SOLOMON, G. C. 1974. Mitochondrial role in cellular minerali-
zation. /. Dent. Res. 53 : 581-587.

SCHMIDT, H. 1926. Untersuchungen an Rhizopoden aus Buchenhohlen. Verh. naturh.
Ver. preuss. Rheinl. 82 : 218-226.

SCHONBORN, W. 1965. Studien iiber die Gattung Difflugiella Cash (Rhizopoda, Testacea).
Limnologica 3 : 315-328.

STEPANEK, M. 1963. Die Rhizopoden aus Katanga ( Kongo- Afrika). Annls Mus. r. Afr.
cent. 117 : 9-91.

TAYLOR, D. L., CONDEELIS, J. S., MOORE, P. L. & ALLEN, R. D. 1973. The contractile basis
of amoeboid movement. I. The chemical control of motility in isolated cytoplasm.
/. Cell Biol. 59 : 378-394.

THIERY, J. P. 1967. Mise en Evidence des polysaccharides sur coupes fines en microscope
eUectronique. /. Microsoc. Paris 6 : 987-1018.

THOMAS, R. 1959. Les Thecamoebiens muscicoles et terricoles : notions d'6cologie generate
et comparative. P.-v. Soc. linn. Bordeaux 89 : 27-53.

THOMAS, R. S. & GREENAWALT, J. W. 1968. Microincineration, electron microscopy, and
electron diffraction of calcium phosphate loaded mitochondria. /. Cell Biol. 39 : 55-76.

WASSERMAN, R. H. & KALLFELZ, F. A. 1970. Transport of calcium across biological mem-
branes. In Biological Calcification: Cellular and Molecular Aspects, pp. 313-345. North-
Holland, Amsterdam.

WOHLMAN, A. & ALLEN, R. D. 1968. Structural organization associated with pseudopod
extension and contraction during cell locomotion in Difflugia. J. Cell Sci. 3 : 105-114.

ZACHARIAS, O. 1903. Zurr Kenntnis der niederen Flora und Fauna holsteinischer Moorsumpfe.
ForschBer. biol. Stn Plon 10 : 223-289.

DR R. H. HEDLEY

C. G. OGDEN

NICOLA J. MORDAN

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON SW7 560

PLATE i
Cryptodifflugia oviformis

FIG. A. Specimen to illustrate the circular aperture and the apertural collar (arrowed) .
x 2500

FIG. B. Section showing one side of the apertural collar. Note that the calcareous layer
continues as a border (arrow) to the organic collar (oc). x 30 100

FIG. C. Section through the aperture of a specimen treated with EDTA to remove the
calcareous layer. Note the thin organic layer (ol) and thickened collar (oc). x 5800

FIG. D. Section through aperture of specimen treated with EDTA, to show the extension
of the organic layer (ol) to form the collar. x 27 ooo

FIG. E. Section of shell wall stained with thiosemicarbazide/silver proteinate, showing
silver deposited at the organic layer (ol). The gap (g) represents the calcareous layer removed
during the staining technique. x 24 ooo

FIG. F. Abnormal shell showing irregular patterning of the shell and two apertural openings
(arrowed). x 1600

Bull. Br. Mus. nat. Hist. (Zool.) 30, 8

PLATE i

PLATE 2
Cryptodifflugia oviformis

Section through the aperture of a whole specimen to show the nucleus (n), Golgi bodies (G),
granular endoplasmic reticulum (ger) and food vacuoles (fv). x 8600

Bull. Br. Mus. nat. Hist. (Zool.) 30, 8

PLATE 2

PLATE 3
Cryptodifflugia oviformis

FIG. A. Section to show the nucleus (n), contractile vacuoles (cv), storage area (sa) and
mitochondria containing calcareous inclusions (me) . x 7800

FIG. B. Section of a mitochondrion (m) with a calcareous inclusion. x 15 600

FIG. C. Section through a Golgi body to illustrate the formation of vesicles with electron-
dense contents at the margin of the saccules (arrowed). x 44 600

FIG. D. Section showing membrane-bound vesicles (mbv) with a thin electron-dense lining
and an electron-lucent centre. x 44 600

FIG. E. Two Golgi bodies (G) in a region of dense endoplasmic reticulum. Note the presence
of membrane-bound vesicles (mbv) at the inner face of the saccules. x 30 100

Bull. Br. Mus. nat. Hist. (Zool.) 30, 8

PLATE 3

PLATE 4
Cryptodifflugia oviformis

FIG. A. Section to illustrate a large electron-lucent vesicle (Iv) and the associated band of
endoplasmic reticulum (er). x n 500

FIG. B. Section stained with thiosemicarbazide/silver proteinate, showing silver deposited
at the storage areas (sa), Golgi bodies (G) and the membrane-bound vesicles (mbv). x 10 ooo

FIG. C. Transverse section showing the position of the storage areas (sa), food vacuole (fv),
contractile vacuoles (cv) and nucleus (n) . x 7800

FIG. D. A typical storage area showing three types of vesicles, electron-dense (ed),
membrane-bound (mbv) and large vesicles (Iv). Note that the large vesicles contain an electron-
dense material. x 13 500

FIG. E. Section through a microbody with tubular elements. x 75 600

FIG. F. Section through a microbody containing a lattice-like structure. x 90 600

Bull. Br. Mus. nat. Hist. (Zool.) 30, 8

PLATE 4

PLATE 5
Cryptodifflugia oviformis

FIG. A. Specimen on an agar surface to illustrate the pseudopodia extended. x 960

FIG. B. Section through a pseudopodium containing mitochondria (m), bundles of micro-
filaments (mf ) and small electron-dense concentrations (arrowed) . x 44 600

FIG. C. Section of a specimen to illustrate the range and abundance of pseudopodia.
X4300

FIG. D. Section through two pseudopodia showing that the opposed membranes are parallel
in the region of contact. Note the thickening of those membranes (arrowed) and the bundles
of microfilaments (mf). x 15 ooo

FIG. E. Section of partially calcified daughter cell, showing the organic layer (ol) with
small concentrations of electron-dense material at the inner surface. x 20 100

FIG. F. Section of a mitochondrian with a homogeneous matrix and apparent absence of
cristae. A swelling of the membrane is arrowed. x 30 100

Bull. Br. Mus. nal. Hist. (Zool.) 30

PLATE 6
Cryptodifflugia oviformis

FIG. A. Section through the aperture of a partially calcified daughter cell. Note the thin
shell wall (sw), the band of fibrillar organic material (fm) sealing the aperture and the vesicles
containing electron-dense material (arrowed). x 20 100

FIG. B. Section of a young daughter cell illustrating the irregular outline of the uncalcified
shell. x 5800

FIG. C. Section of shell wall of abnormal specimen. Note the folded organic layer (ol)
and the small concentrations of calcified material on the inner surface (arrowed). x 30 100

FIG. D. Section through an aperture (ap) sealed with an operculum (op). x 28 400

FIG. E. Light micrograph showing the operculum (op) in the apertural opening. x 2000

Bull. Br. Mus. nat. Hist. (Zool.) 30, 8

PLATE 6

sw

A LIST OF SUPPLEMENTS
TO THE ZOOLOGICAL SERIES

OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

1. KAY, E. ALISON. Marine Molluscs in the Cuming Collection British Museum
(Natural History) described by William Harper Pease. Pp. 96 ; 14 Plates.
1965. (Out of Print.)

2. WHITEHEAD, P. J. P. The Clupeoid Fishes described by Lacepede, Cuvier and
Valenciennes. Pp. 180 ; n Plates, 15 Text-figures. 1967. £4-80.

3. TAYLOR, J. D., KENNEDY, W. J. & HALL, A. The Shell Structure and
Mineralogy of the Bivalvia. Introduction. Nuculacea-Trigonacea. Pp. 125 ;
29 Plates, 77 Text-figures. 1969. £5-40.

4. HAYNES, J. R. Cardigan Bay Recent Foraminifera (Cruises of the R.V. Antur]
1962-1964. Pp. 245 ; 33 Plates, 47 Text-figures. 1973. £11-90.

5. WHITEHEAD, P. J. P. The Clupeoid Fishes of the Guianas. Pp. 227 ; 72
Text-figures . 1 973 . £10-65.

6. GREENWOOD, P. H. The Cichlid Fishes of Lake Victoria, East Africa : the
Biology and Evolution of a Species Flock. Pp. 134 ; i Plate, 77 Text-figures.
1974. £3-75. Hardback edition £6.

Printed in Great Britain by John Wright and Sons Ltd. at The Stonebridge Press, Bristol BS4 5NU

( 2iJANlOT

THE LARVAL AND POST-LARVAL^^
DEVELOPMENT OF THE SCORPION

SPIDER CRAB, INACHUS

DORSETTENSIS (PENNANT)

(FAMILY : MAJIDAE), REARED IN

THE LABORATORY

R. W. INGLE

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)
ZOOLOGY Vol. 30 No. 9

LONDON: 1977

V

25 JAN 1977

THE LARVAL AND POST-LARVAL .LIBRARY
DEVELOPMENT OF THE SCORPION SPI
CRAB, INACHUS DORSETTENSIS (PENNANT)
(FAMILY : MAJIDAE), REARED IN THE
LABORATORY

BY

R. W. INGLE

Pp. 329-348 ; 10 Text-figures

BULLETIN OF

THE BRITISH MUSEUM (NATURAL HISTORY)

ZOOLOGY Vol. 30 No. 9

LONDON: 197?

THE BULLETIN OF THE BRITISH MUSEUM

(NATURAL HISTORY), instituted in 1949, is
issued in Jive series corresponding to the Scientific
Departments of the Museum, and an Historical series.

Parts will appear at irregular intervals as they
become ready. Volumes will contain about three or
four hundred pages, and will not necessarily be com-
pleted within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 30, No. 9, of the Zoology series.
The abbreviated titles of periodicals cited follow those
of the World List of Scientific Periodicals.

World List abbreviation
Bull. Br. Mus. nat. Hist. (Zool.)

ISSN 0007-1498

Trustees of the British Museum (Natural History), 1977

BRITISH MUSEUM (NATURAL HISTORY)

Issued 27 January, 1977 Price £2.00

This issue ends Volume 30.

THE LARVAL AND POST-LARVAL

DEVELOPMENT OF THE SCORPION SPIDER

CRAB, INACHUS DORSETTENSIS (PENNANT)

(FAMILY : MAJIDAE), REARED IN THE

LABORATORY

By R. W. INGLE

INTRODUCTION

THE Scorpion Spider Crab, Inachus dorsettensis (Pennant), occurs from Norway,
through the north Atlantic to the Mediterranean and along the west and South
African coasts as far as Natal, being confined to littoral waters to depths of 550 m
(Monod, 1956 : 529). It is reported from hard to muddy substrates (Christiansen,
1969 : 102) although in British waters it has been taken chiefly in areas of sandy-
mud (Marine Biological Association, 1957 : 260 ; Hartnoll, 1963 : 434). In British
waters this species breeds throughout the year (Lebour, 1928 : 502).

Although adults of the three British Inachus species are readily distinguishable
from each other, the larval and post-larval stages cannot be reliably identified at
present because of the lack of adequate descriptions. Recently I. dorsettensis was
reared to ninth crab stage in the laboratory and the early developmental stages
are described in detail below.

MATERIAL AND METHODS

Larvae were hatched from crabs collected on or near the Eddystone Grounds, off
Plymouth, Devonshire, England, and were reared using the method described by
Rice & Ingle (1975 : 104).

Larvae and moults were preserved in 70% alcohol and in 5% buffered formal-
saline. Dissections were made in polyvinyl-lactophenol/lignin pink medium.
Drawings and measurements were made with the aid of a camera lucida. The
measurements taken were : dorsal spine length (D.S.) = distance between tip of
dorsal spine and rostrum, and carapace length (C.L.) = distance between base of
eyestalk and posterio-lateral margin of carapace. Between 5 and 10 suitable
specimens of each stage were measured. Sizes given under Dimensions are those
for the smallest and the largest specimens.

RESULTS

Most females hatched their total brood within a period of 36 hours ; a few took
up to 48 hours. All successfully reared larvae hatched as free-swimming zoeae.
The small percentage that hatched as pre-zoeae all died within 24 hours.

332

R. W. INGLE

FIG. i. Inachus dorsettensis, lateral and frontal aspects of zoeal stages : (a), (b) first zoea
and (c), (d) second zoea ; (e) basal part of dorsal spine of first zoea ; (f ) majid spine of
first zoea. Scales in millimetres.

LARVAL DEVELOPMENT OF INACHUS DORSETTENSIS

333

FIG. 2. Inachus dorsettensis, (a), (b) antennule ; (c), (d) antenna ; (e), (f ) mandible ;
(g), (h) maxillule ; (i), (j) maxilla of : (a), (c), (e), (g), (i) first zoea and (b), (d), (f), (h),
(j) second zoea. Scales in millimetres.

334 R- W. INGLE

DESCRIPTIONS

Inachus dorsettensis (Pennant)

LARVAL REFERENCES. Inachus scorpio Claus, 1876 : Taf. X, fig. 8 (ist zoea) ; non Inachus
scorpio Cano, 1893 : Taf. 35, fig. 7 (ist zoea) = /. thoracicus (Roux) ; Inachus dorsettensis
Lebour, 1927 : 795 ; PI. I, figs i, 4-6 (pre-zoea), fig. 7 (ist zoea), PI. II, figs i, 4-7, PL IV, figs
i, 3 (and zoea), PL II, fig. 10, PL III, figs i, 7, 8, PL IV, figs 2, 4 (megalopa), PL III, fig. 8 (ist-
3rd crab) ; 1928 : 546, PL III, fig. 5 (ist zoea, col. fig.), PL XV, fig. 4 (ist crab), fig. 5 (2nd
crab) ; Heegaard, 1963 : 471 (ist zoea), PL XVII (coloured fig.), text-figs 70-76.

FIRST ZOEA

Dimensions : D.S. 1-9-2-0 mm ; C.L. 0-9-1-0 mm.

Carapace (Fig. la, b) : Dorsal spine well developed, apically curved, armed with
minute spinules most numerous at base (Fig. le). Rostral and lateral spines
absent. Dorso-median elevation (D.E.) prominent ; majid spine (M.S.) plumose,
arising from inner ventral carapace margin and well developed (Fig. if), posterio-
ventral margin of carapace spinulate.
Eyes : Partly fused to carapace.

Antennule (Fig. 2a) : Unsegmented with two terminal aesthetascs and 2 setae.
Antenna (Fig. 2c) : Spinous process (S.P.) about f length of dorsal spine, distal £
invested with minute spinules. Exopod as long as spinous process with 2 median
spines, distal \ spinulate. Endopod about ^ length of spinous process.
Mandible (Fig. 2e) : Incisor and molar process developed, palp absent.
Maxillule (Fig. 2g) : Endopod 2-segmented, distal segment about 4x length of
proximal and with 4 distal setae ; basal endite (B.E.) with 5 setosed spines and
2 setae, coxal endite with 7 plumose setae.

Maxilla (Fig. 21) : Endopod (E.N.) narrow, apex truncate with 3 long and i (outer-
most) short plumose setae ; basal endite (B.E.) bilobed with 7-8 plumose setae ;
coxal endite bilobed with 7 plumose setae, inner margin invested with fine setules.
Scaphognathite with 10 plumose setae, distal one very stout, posterior lobe short.
First maxilliped (Fig. 3a) : Basis with 9 setae (arranged 2, 2, 2, 3), endopod 5-
segmented with 3, 2, i, 2, 4 + i setae respectively ; exopod incipiently 2-segmented,
distally with 4 terminal plumose setae.

Second maxilliped (Fig. 3b) : Basis devoid of setae, endopod 3-segmented with
o, i, 3 + i setae respectively, distal segment about 3 x length of proximal, exopod
incipiently segmented, distally with 4 plumose setae.

Third maxilliped (Fig. 30, MX) : Represented as a small unsegmented bud.
Peraeopods (Fig. 30) : Conspicuous, ist incipiently chelate, 2nd~5th unsegmented
as terminally acute finger-like projections.

Abdomen (Fig. 4a, b) : 5-segmented + telson. Segments 2-5 with incipient pleopod
buds ; 2nd segment with pair of forwardly directed dorso-lateral processes, segments
3-5 each with a pair of long posterior dorso-lateral spines and with posterior ventro-
lateral margins rounded to truncate. Posterio-dorsal margin of segments 2-5 each
with a pair of small setae. Telson forks long, not diverging posteriorly, lateral
spines (L.S.) long, nearly % length of forks, both invested with minute spinules ;
each half of telson posterior margin convex, with 3 plumose setae.

LARVAL DEVELOPMENT OF INACHUS DORSETTENSIS

335

FIG. 3. Inachus dorsettensis, (a), (d) first maxilliped ; (b), (e) second maxilliped ;
third maxilliped and peraeopods of : (a)-(c) first zoea and (d)-(f) second zoea.
in millimetres.

(c), (f)
Scale

336

R. W. INGLE

FIG. 4. Inachus dorsettensis, abdomen, dorsal (a), (c) and lateral (b), (d) aspects of (a),
(b) first zoea and (c), (d) second zoea ; (e) right half of telson of second zoea ; (f ) anten-
nule of megalopa ; (g) antennule of first crab. Scales in millimetres.

LARVAL DEVELOPMENT OF INACHUS DORSETTENSIS 337

SECOND ZOEA

Dimensions : D.S. 1-7-1-8 mm ; C.L. 1-2-1-3 mm.

Carapace (Fig. ic, d) : Dorsal spine slightly more curved and shorter than in first

stage, dorso-median elevation more pronounced and now with 2 rows of 3 setae ;

rostrum pronounced, ocular eaves developed.

Eyes : Now stalked.

Antennule (Fig. 2b) : Now with 3 terminal aesthetascs and 3 setae.

Antenna (Fig. 2d) : Endopod now about \ length of spinous process.

Mandible (Fig. 2f ) : Distal margin of incisor process cut into rounded teeth.

Maxillule (Fig. 2h) : Outer margin of basal endite now with a plumose seta, distal

margin now with 3 setae.

Maxilla (Fig. 2J) : Outermost plumose seta of endopod longer than in first stage ;

basal endite with 9 plumose setae ; scaphognathite with 17 plumose setae, distal

one no longer stout.

First maxilliped (Fig. 3d) : Exopod now with 6 terminal plumose setae.

Second maxilliped (Fig. 3e) : Exopod now with 6 terminal plumose setae.

Third maxilliped (Fig. 3f ) : Now a pronounced bilobed bud.

Peraeopods (Fig. 3f ) : 2nd~5th incipiently segmented.

Abdomen (Fig. 40, d) : Segments 2-5 with well-developed pleopod buds, dorso-

lateral processes on 2nd segment longer than in first stage and with terminal spinules ;

posterior dorso-lateral spines shorter than in first stage ; telson forks diverging

posteriorly, spinules (Fig. 46) conspicuous.

MEGALOPA

Dimensions : C.L. 1-1-1-2 mm.

Carapace (Fig. loa) : Longer than broad, narrowing anteriorly. Front with a pair
of widely spaced anteriorly directed acute frontal horns and medially with an
obtuse protuberance ; protogastric region with 3 transverse low tubercules, outer-
most 2 spined ; cardiac with 3 spines transversely placed, intestinal with a low
posteriorly placed tubercule.
Eyes : Large and long, cornea well developed.

Antennule (Fig. 4f) : Peduncle 3-segmented, exopod (EX) not demarcated from
peduncle, with 4 aesthetascs and 2 setae, endopod vestigial.

Antenna (Fig. 5d) : 6-segmented, flagellum not demarcated from peduncle. First
segment with ventrally directed spine and outwardly directed obtuse process, 3rd
segment with i seta, 5th segment longest with 4 terminal setae, 6th with 2 long
and i short terminal setae.

Mandible (Fig. 5a) : Incisor process well developed, molar reduced, now with 3-
segmented mandibular palp, distal segment with 1-2 setae.

Maxillule (Fig. 5b) : Endopod now reduced, unsegmented ; basal endite with 6
spines and 5-7 setae, spines thinner than in zoeal stages.

Maxilla (Fig. 5c) : Endopod reduced to terminally acute process ; basal endite
slightly bilobed with 9-10 plumose setae ; coxal endite slightly bilobed with 5
plumose setae. Scaphognathite with 22 plumose setae shorter than in zoeal stages.
First maxilliped (Fig. 6a) : Coxal endite (C.E.) bilobed, invested with i and 4 setae

338

R. W. INGLE

FIG. 5. Inachus dorsettensis, (a) mandible, (b) maxillule, (c) maxilla and (d) antenna of
megalopa ; (e) antenna and (f ) endopod of maxilla of first crab. Scale in millimetres.

LARVAL DEVELOPMENT OF INACHUS DORSETTENSIS

339

FIG. 6. Inachus dorsettensis, first maxilliped of (a) megalopa and (b) first crab ; second
maxilliped (c) of megalopa and (d) endopod of second maxilliped of first crab. Scale in
millimetres.

340 R. W. INGLE

respectively ; inner margin of basis (B.S.) with 5 setae. Exopod long, stout, 2-
segmented, distally with 4 long terminal plumose setae ; endopod curved with 1-2
subapical setae, epipod short, bilobed.

Second maxilliped (Fig. 6c) : Basal and coxal segments not separated, endopod 5-
segmented, ist segment not completely demarcated from 2nd, 3rd with i, 4th with
3 dorso-external setae, 5th with 4 setosed spines. Exopod long, stout, 2-segmented,
distal segment with 4 plumose setae, epipod short, bilobed.

Third maxilliped (Fig. 7a) : Basis slightly demarcated from coxa, inner margin of
coxal segment with i spine and i seta, ischium with 3-5 acute spines or processes
and 6 setae with additional 2 on outer surface ; merus with a small blunt anterio-
internal spine on inner surface, merus to dactylus setosed as shown in figure.
Peraeopods, cheliped (Fig. 7c) : Inner margin of ischium/basis with 3 stout curved
spines, merus with 3 stout spines, carpus inner margin with i small spine, propodus
distal half, inner margin sharp, slightly cristate, setal investment as shown in
figure. Peraeopods 2-5 (Fig. 8b) dactylus distally spinulate, much longer than
propodus + carpus, basis of peraeopods 2-3 with stout curved spine, merus of
peraeopods 2-4 and sometimes ischium with long spines decreasing in size distally ;
setal and spinule investment as shown in Fig. 8b.

Abdomen (Fig. 8a, loa) : Sub-cylindrical 5-segmented + telson. Posterio-dorsal
margins of segments 2-5 with pair of acute spines (P.D.) decreasing in size on suc-
cessive segments ; segment 3 with i pair and 4 and 5 with 2 pairs of additional
minute spinules ; dorsal surface of segments 2 and 3 with a median spine. A pair
of well-developed pleopods on segments 2-5, exopods (Fig. gc) long, each with 8
terminal plumose setae ; endopod short with 2 coupling hooks on each pleopod.

FIRST CRAB

Dimensions : C.L. 1-4-1-5 mm.

Carapace : Longer than broad, narrowing anteriorly, invested with numerous setae
of varying lengths (Fig. lob). Anteriorly directed frontal horns broader than in
megalopa, apices setosed. Hepatic and outer epibranchial regions swollen, each
with a spine ; meso-metagastric and cardiac regions swollen, latter with a pair of
small tubercules.

Eyes : Large and broad, cornea well developed.

Antennule (Fig. 4g) : Exopod now with 3 incipient segments, terminal with 5
aesthetascs and one seta ; endopod with 3 incipient segments, terminal with 3
setae.

Antenna (Fig. 5e) : Now y-segmented, flagellum not demarcated from peduncle ;
ist segment longest, inner margin with ventrally directed spine shorter than in
megalopa and 2 additional smaller spines on inner margin, outer margin with 2
plumose setae ; 2nd and 3rd segments each with 2 setae, 5th with i, 6th with 3,
7th with apex indistinctly segmented.
Mandibles, maxilhde and maxilla : As in megalopa.

First maxilliped (Fig. 6b) : Coxal segment with 8 setae, no longer bilobed, demar-
cated from basis which has 11-12 setae, endopod distally expanded into broad
lobe with 2-3 marginal setae, epipod longer than in megalopa with 1-2 setae.

LARVAL DEVELOPMENT OF INACHUS DORSETTENSIS

341

FIG. 7. Inachus dorsettensis, third maxilliped of (a) megalopa and (b) first crab ; cheliped
of (c) megalopa and (d) first crab. Scales in millimetres.

342

R. W. INGLE

FIG. 8. Inachus dorsettensis, (a) abdomen (b) peraeopods 2-5 of megalopa ; abdomen
(c) of second crab, (d) of third crab cj, (e) of fourth crab ? and (f ) of fourth crab <J. Scale
in millimetres.

LARVAL DEVELOPMENT OF INACHUS DORSETTENSIS 343

Second maxilliped : 2nd segment of endopod longer than in megalopa (Fig. 6d), 4th

segment now with 4 setae, 5th with 2 setae.

Third maxilliped (Fig. 7b) : Outer margin of basis with 7-10 setae, inner margin of

ischium with 9-12 acute or subacute spines and 9-12 setae ; anterio-internal spine

on merus now prominent, merus-dactylus setosed as shown in figure.

Peraeopods : Propodus of cheliped (Fig. yd) less inflated than in previous stage,

margins sharp. Peraeopods 2-5 similar, 2nd (Fig. gd) longest, dactylus of 2nd~4th

shorter than propodus, dactylus of 5th (Fig. ge) same length as propodus and with

inner margin bearing 1-2 broad spines, peraeopod setae long.

Abdomen (Fig. ga., b) : Sub-cylindrical, posterio-dorsal margins of segments 2-5

now with minute spinules and a pair of long setae ; a minute median spine on dorsal

surface of telson ; pleopods reduced.

SECOND CRAB

Dimensions : C.L. 1-8-2-0 mm.

Description : Unchanged except for abdomen (Fig. 8c) which is now strongly

dorso-ventrally flattened and broad. Pleopods on segments 2-5 now reduced to

minute buds.

THIRD CRAB

Dimensions : C.L. 2-3-2-5 mm.

Carapace : With a pair of well-developed hepatic spines (Fig. loc) ; setae investing
carapace and limbs now strongly hooked at apices.

Abdomen : Male : Narrow (Fig. 8d), well-developed pleopod buds now present near
posterior margins of segments i and 2, those on segment 3 minute ; no pleopods on
segments 4-5.

Female : slightly broader than that of male's ; biramous pleopods present on seg-
ments 2-5, but those on segments 4 and 5 very small.

FOURTH CRAB

Dimensions : C.L. 2-8-2-9 mm.

Carapace and limbs : As in previous stage.

Abdomen : Ventral surface with more setae and margin of 6th segment setosed.

Pleopods : Male - ist and 2nd pairs (Fig. 8f ) much longer than in previous stage.

Female - 2nd and 3rd pairs longer than in previous stage, 4th and 5th pairs still

vestigial (Fig. 8e).

FIFTH CRAB

Dimensions : C.L. 3-1-3-2 mm.

Carapace (Fig. rod) : Now with 4 transverse protogastric tubercules and i meso-

gastric ; a pair of mesobranchial and i gastric tubercule as in the adult form.

Pleopods : Male - ist and 2nd now well developed.

Female - Pleopods of the 4th and 5th abdominal segments now much longer and

strongly biramous.

344

R. W. INGLE

FIG. 9. Inachus dorsettensis, abdomen of first crab, (a) dorsal and (b) lateral aspects
respectively ; (c) ist pleopod of megalopa ; (d) ist peraeopod of first crab ; (e) 5th
peraeopod of first crab. Scales in millimetres.

LARVAL DEVELOPMENT OF INACHUS DORSETTENSIS

345

FIG. 10. Inachus dorsettensis, (a) megalopa ; (b) first crab ; (c) third crab ; (d) fifth
crab. Scales in millimetres.

346

R. W. INGLE
REMARKS

In Tables i and 2 the present material of I. dorsettensis is compared with previous
accounts of the larval and juvenile stages of this species. Some differences are
apparent in setal armature of certain appendages of the first zoeal stages, particu-
larly with reference to the maxillules, maxillae and maxillipeds of Heegaard's
Mediterranean material.

TABLE i

Glaus, 1876

Lebour, 1927 ;

Heegaard, 1963

Ingle, 1977

1928

ist zoea

Dorsal spine :

slightly longer

as long as

much longer than

slightly longer

than carapace

carapace, curved

carapace, curved

than carapace,

length, curved

apically curved

Majid spine :

not shown

not shown

well developed

well developed

Dorso-median

elevation of

carapace :

slight

pronounced

very pronounced

very pronounced

Antennule :

2 aesthetascs

2 aesthetascs +

2 aesthetascs +

2 aesthetascs +

i seta

2 setae

2 setae

Antenna, spinous

process :

shorter than

shorter than

much shorter

shorter than

dorsal spine

dorsal spine

than dorsal spine

dorsal spine

Antenna, endopod/

spinous process

length :

about J

about J

between J and £

about i

Maxillule, basal

endite :

not shown

not shown

with 4 setosed

with 5 setosed

spines and 2 setae

spines and 2 setae

Maxilla, basal

endite :

not shown

not shown

with 8 setae

with 7-8 setae

Maxilla, scaph-

ognathite :

not shown

not shown

with 7 setae,

with 10 setae,

posterior lobe

posterior lobe

long

short

First maxilliped,

basis :

without setae (?)

without setae (?)

with 3 setae

with 2, 2, 2, 3

setae

First maxilliped,

endopod :

o, i, 2, i, 3 + 1

(?) o, 2, o, 2, 3 + 1

2, 2, I, 2, 3 + 1

3, 2, i, 2, 4, + i

Second maxilliped,

endopod :

(?)

(?)

2-segmented

3-segmented

Second maxilliped,

endopod setae :

(?) 0, 2, 2

(?) o, o, 3

o, o, 2 + 1

o, i, 3 + 1

Abdomen, dorso-

lateral processes :

acute

acute

truncate

acute

These differences could be genotypic, resulting from widely separated geographical
populations, or phenotypic, since Heegaard's material was reared during spring and
summer when Mediterranean water temperatures are much higher than those of
British waters. Differences in setal armature are more apparent when Lebour 's

LARVAL DEVELOPMENT OF INACHUS DORS ETTE NS IS

347

Antennule :

Maxillule, basal endite :

Maxilla, basal endite :
Maxilla, scaphognathite :

Carapace :

Antennule, exopod :
Antennule, endopod :
Antenna, ist segment :
3rd segment :
5th segment :
terminal segment
Abdomen, dorsal spines :
ist segment :
3rd segment •

Carapace :

TABLE 2

Lebour 1927 ; 1928

2nd zoea
4 aesthetascs + i long seta

plumose seta on outer margin
not shown
with 8 setae
with 20 setae

Megalopa

(?) cardiac region with a
tubercule
with 5 aesthetascs
bud well developed
spine short, obtuse
without a seta
with i seta
with 2 setae
long

with lateral setae
without median spinule

ist crab
almost as broad as long

Ingle, 1977

3 aesthetascs + 2 long and i

short seta

a well-developed plumose seta

on outer margin

with 9 setae

with 17 setae

cardiac region with 3 spines

with 4 aesthetascs and 2 setae

bud vestigial

spine long, acute

with a seta

with 4 setae

with 3 setae

short

without lateral setae

with median spinule

much broader than long

account of the second zoeal stage is compared with the corresponding stage of the
present material and there are even greater dissimilarities between corresponding
megalopal and first crab stages (see Table 2). Such discrepancies are not easily
explained, particularly as both Lebour's larvae and the present material were
reared during the same months and hatched from females collected in the same area.
However, variation in laboratory reared material has received little attention (see
Ingle & Rice, 1971 : 284 ; Conor & Conor, 1973 : 245).

The first zoeal stage figured by Cano (1893) as /. scorpio was referred to /. dorset-
tensis by Lebour (1927) and Heegaard (1963). But this figure clearly shows a globu-
lar carapace that is without a dorso-median elevation which are two important
features of the first zoea of I. thoracicus and Cano's larva, therefore, clearly belongs
to this latter species.

At present the larval and early crab stages of /. dorsettensis cannot be distinguished
from those of /. phalangium ( = I. dorynchus) or from I. leptochirus. Although
Lebour (1928 : 548) states that the first zoea of /. leptochirus has longer lateral
telsonic spines than either of the other two species and is more brightly coloured
and that /. phalangium has a longer dorsal spine than I. dorsettensis, she was unable
to detect differences between the megalopal stages of these three species except for
the 'rostral horns' that are apparently closer together in /. phalangium and I. lepto-
chirus than in /. dorsettensis.

An interesting result of this study of I. dorsettensis is that the first pair of pleopod
buds of the male appear at the third crab stage and not at the second as described

348 R. W. INGLE

by Shen (1935) for the portunid Carcinus maenas. It remains to be seen whether
this will be the case when post-larval material of other Ma j ids and Portunids become
available for comparison.

ACKNOWLEDGEMENTS

I wish to thank the Director of the Marine Biological Association, Plymouth,
for providing laboratory facilities for holding ovigerous material, Mr T. R. Tozer
and Mr A. D. M. Mattacola for their generous assistance on many matters during
my visit to Plymouth, the crew of the Sepia for their help in collecting material
and Dr A. L. Rice for kindly reading the manuscript.

REFERENCES

CANO, G. 1893. Sviluppo e morfologia degli Oxyrhynchi. Mitt. zool. Stn. Neapel 10 : 527-

583-
CHRISTIANSEN, M. E. 1969. Marine Invertebrates of Scandinavia. No. 2. Crustacea Deca-

poda Brachyura. Universitetsforlaget, Oslo, 1-143.
CLAUS, C. 1876. Untersuchungen zur Erforschung der Genealogischen Grundlage des Crustaceen-

Systems. Wien, i-viii + 114 pp.
CONOR, J. J. & CONOR, S. L. 1973. Variations in appendage setal counts in zoea larvae of

four porcellanid crabs (Decapoda Anomura) from Oregon. Crustaceana 25 : 245-252.
HARTNOLL, R. G. 1963. The biology of Manx spider crabs. Proc. zool. Soc. Lond. 141 : 423-

496.
HEEGAARD, P. 1963. Decapod larvae from the Gulf of Napoli hatched in captivity. Vidensk.

Meddr dansk. naturh. Foren. 125 : 449-493.
INGLE, R. W. & RICE, A. L. 1971. The larval development of the masked crab, Corystes

cassivelaunus (Pennant) (Brachyura, Corystidae), reared in the laboratory. Crustaceana

20: 271-284.
LEBOUR, M. V. 1927. Studies of the Plymouth Brachyura. I. The rearing of crabs in

captivity, with a description of the larval stages of Inachus dorsettensis, Macropodia longi-

rostris and Maia squinado. J. mar. biol. Ass. U.K. 14 : 795-820.
- 1928. The larval stages of the Plymouth Brachyura. Proc. zool. Soc. Lond. 1928: 473-

560.

MARINE BIOLOGICAL ASSOCIATION 1957. Plymouth Marine Fauna, 3rd Ed. i-xliii, 457 pp.
MONOD, T. 1956. Hippidea et Brachyura ouest-africains. Mem. Inst.fr. Afr. noire 45 : i-

674.
RICE, A. L. & INGLE, R. W. 1975. The larval development of Carcinus maenas (L.) and

C. mediterraneus Czerniavsky (Crustacea, Brachyura, Portunidae) reared in the laboratory.

Bull. Br. Mus. nat. Hist. (Zool.) 28 (3) : 103-119.
SHEN, C. J. 1935. An investigation of the post-larval development of the shore-crab Carcinus

maenas, with special reference to the external secondary sexual characters. Proc. zool.

Soc. Lond. 1935: 1-33.

R. W. INGLE

Department of Zoology

BRITISH MUSEUM (NATURAL HISTORY)

CROMWELL ROAD

LONDON SW7 5BD

A LIST OF SUPPLEMENTS
TO THE ZOOLOGICAL SERIES

OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

1. KAY, E. ALISON. Marine Molluscs in the Cuming Collection British Museum
(Natural History) described by William Harper Pease. Pp. 96 ; 14 Plates.
1965. (Out of Print.)

2. WHITEHEAD, P. J. P. The Clupeoid Fishes described by Lacepede, Cuvier and
Valenciennes. Pp. 180 ; n Plates, 15 Text-figures. 1967. £4.

3. TAYLOR, J. D., KENNEDY, W. J. & HALL, A. The Shell Structure and Mineralogy
of the Bivalvia. Introduction. Nuculacea-Trigonacea. Pp. 125 ; 29 Plates,
77 Text-figures. 1969. £4'5o.

4. HAYNES, J. R. Cardigan Bay Recent Foraminifera (Cruises of the R.V. Antur)
1962-1964. Pp. 245 ; 33 Plates, 47 Text-figures. 1973. /io-8o.

5. WHITEHEAD, P. J. P. The Clupeoid Fishes of the Guianas. Pp. 227 ; 72
Text-figures. 1973. £9'70.

6. GREENWOOD, P. H. The Cichlid Fishes of Lake Victoria, East Africa : the
Biology and Evolution of a Species Flock. Pp. 134 ; i Plate, 77 Text-figures.
1974. £3-75. Hardback edition £6.

Printed in Great Britain by John Wright and Sons Ltd. at The Stonebridge Press, Bristol BS4 5NU