aaa Pe a WS

+ 2 z ® 1
jane Cape nee se tee

th iret eh toast
: Sone a en

Cea

ANNALS

OF THE

NATAL GOVERNMENT
MUSEUM

EDITED BY

ERNEST WARREN, D.Sc.(Lonp.)., Drtrecror.

PRINTED) BY ORDER OF THE TRUSTEES:

TONED. ONE:
ADLARD & SON, BARTHOLOMEW CLOSE.

1908.

Bg = yj
ee
~ ore

=

@
ey ba |
* ‘3
i > ie
.
a
i
A
a
=
ba. | :
‘
. ' ’ y :
s ~ f © ta? t
.¢* rad
= a)
ae |
i i

¥
ri

te

rs
ay oe
2 ae |
2
— 4
a 4
no
_ a io)
t : hr
‘ or
=i
\ |
ay ¥
a i =a 7
Sor
. »
iy ==
vee
6 ‘ 5
s ¥
; :
a
;
;
{
— »
te
C3 ,
\
; |
vi}
sau
4
a aa

CON WE IN Ts:

VOLUME I.

Part 1, rssueD JuNE, 1906.

Description of new or little-known Fishes from the Coast of
Natal. By C. Tare Recan, B.A. (With Plates I—V)

On Bertramia kirkmani sp. n.,a Myxosporidium occur-
ring in a South African Rotifer. By Ernest WARREN,
D.Se.Lond. (With Plate VI)

On South African Marine Mollusca, with Deseneuens of Naw
Species. By Epear A. Surrn, 1.8.0., F.Z.8. (With
Plates VII, VIIT) ;

On Halocordyle cooperi sp. u., a Hydroid from the
Natal Coast. By Ernest Warren, D.Sc.Lond. (With
Plate IX) .

On Tubularia solitaria sp. n., a Hedveidl from ‘whe Seal
Coast. By Ernest Warren, D.Sc.Lond. (With
Plates X and XI).

Notes on a New Species of Gymnoplea con ichuondl Natal
South Africa; Adiaptomus natalensis gq. e. sp.
By Arnotp W. Cooper, F.R.M.5., ete. (With oe
XIT) : :

Note on Convoluta moroettenets Graf collected on the
Natal Coast. By Ernest Warren, D.Sc.Lond. (With
Plate XIIT) 2

Note on the Abnormal Fonte of a Sheep. By ERNEST
Warren, D.Se.Lond. (With Plate XIV)

Parr 2, 1ssuED Marcu, 1907.

A Contribution to the Study of the Characteristics of Larvee
of Species of Anophelina in South Africa. By Ernest
Hitt, D.P.H.Camb., and L. G. Haypon, M.B., C.M.,
D.P.H.Aberd. (With Plates XV—XXVI)

Language of Colours amongst the Zulus expressed by their
Bead-work Ornaments ; and some General Notes on their
Personal Adornments and Clothing. By Rev. FarHer
Franz Mayr. (With Plate XXVII)

DEF

PAGE

“NI

ies)

73

97

105

109

111

159

lv CONTENTS.

On Two New Reptiles from the Karroo Beds of Natal. By
R. Broom, M.D., D.Sc., C.M.Z.S., Victoria College,
Stellenbosch. (With Plate XX VITT)

On Entomostraca collected in Natal by Mr. James Gibson.
(Part II). By G. Srewarpson Brapy, M.D., LL.D.,
D.Sc., F.R.S., C.MLZ.S. (With Plates XXIX—XXXII)

On Parawrightia robusta g. e. sp. n., a Hydroid from
the Natal Coast; and also an Account of a Supposed
Schizophyte occurring in the Gonophores. By Ernest
Warren, D.Sc.Lond. (With Plates XXX4II and
EXEXAY SE

Note on the Variation in the Aawaawerent at the Capiate
Tentacles im the Hydroid, Halocordyle cooperi
Warren. By Ernest Warren, D.Sc.Lond.

Note on the Larva of a Fly (Sarcophaga sp.) occurring in the
Human Intestine. By Ernest Warren, D.Sc.Lond.

Part 3, issueD May 15ru, 1908.

On a Collection of Fresh-water Fishes, Batrachians, and
Reptiles from Natal and Zululand, with descriptions of
New Species. By G. A. BouLencsr, F.R.S. (With
Plates XX XV—XXXVI1)

Note on Clarias capensis C.and V. By G. A. BouLenGER,
F.R.S.

A Collection of Fishes from ne Goan of Natal: Faluianel
and Cape Colony. By C. Tare Reean, M.A. (With
Plates XXX VII—XLII) . .

A Short Study on Zulu Music. By Rev. FarHer Franz
Mayr. (With Plates XLITI and XLIV)

On a Collection of Hydroids, mostly from the Natal Obast.
By Ernest Warren, D.Sc.Lond. (With Plates XLV—
NKUVALE))

IssueD Juty, 1908.

Title-page of Vol. I

Contents of Vol. I

Index of South African Marine Molina
General Index .

Errata .

PAGE

173

FISHES FROM THE COAST OF NATAL. 1

Descriptions of new or little-known Fishes from
the Coast of Natal.'

By
c. Vate Regan, B.A.

With Plates I—YV.

PLIOTREMA gen. nov.

Dirrers from Pristiophorus Mill. u. Henle in having six
aill-clefts on each side.

Pliotrema warreni sp. n. (Pl. I).

Body elongate, about as broad as deep, the flat lower sur-
face margined on each side by a ridge which is scarcely dis-
tinct on the trunk, but well-developed on the tail. Denticles
small, pointed, with a median keel and sometimes with a pair
of lateral keels, close set on the lower parts of the body and
the anterior margins of the fins; distal parts of the fins naked.
Length of snout in front of eye 42 times its breadth at the
base. Teeth of the rostrum compressed, pointed, unequal in
length, the larger ones with denticulated posterior edge.
Barbel, when laid back, reaching the nostril, which is shehtly
nearer to the angle of the mouth than to the root of the barbel.
Upper jaw with 40 to 44 series of teeth, lower Jaw with 31 to
34. First dorsal scarcely larger than the second, originating

! The fishes here described form part of a collection sent to the
British Museum by Dr. E. Warren, who has retained a duplicate set
for the Natal Government Museum, Pietermaritzburg. The specimens
were obtained by the Museum Collector, Mx. F. Toppin.

VOU. be PART I. ]

Y C. TATE REGAN.

above the extremity of the inner edge of the pectoral. Caudal
feebly heterocercal, with upper lobe well developed.

Two specimens, each about 750 mm. in total length, the one
from the coast of Natal, taken at a depth of forty fathoms, the
other from False Bay, Cape of Good Hope, received in 1899
from Dr. J. D. F. Gilchrist.

The presence of six gill-clefts in a Pristiophorid is a most
unexpected feature. As has been pointed out by Jaekel, the
Pristiophoride are closely allied to the Squalide (Spinacidee),
a family in which the gill-clefts are constantly five im number.
The anatomy of the fish described above is extremely similar to
that of Pristiophorus cirratus and there is no evidence
of relationship to the Notidanoids. It may be that the large
number of gill-clefts is a primitive feature retained in this
fish and lost in its allies,! in which case it would follow that
the reduction in number to five has been independently
arrived at in different groups. On the other hand, there is
the possibility that an increased number of gill-clefts may
sometimes be a feature of specialisation, just as in the case
of certain groups of Teleosts with an increased number of
vertebree or of pectoral pterygials, which are now regarded
as derived from forms with a lower number.

Raia ocellifera sp. n. (Pl. Il).

Snout with a short, obtuse, triangular projection. Diameter
of eye 34-82 in the length of snout and 12-13 in the inter-
orbital width. Distance between outer edges of nostrils
greater than their distance from the tip of the snout. Teeth
obtuse, in 48-50 series in the upper jaw. Anterior border of
pectoral undulated. Body smooth, except for some small
asperities along the anterior margin of each pectoral. Three

1 This view is supported by the presence in Raia of a structure which
has been interpreted as the remnant of a sixth gill-cleft, and by the
recent discovery of a sixth branchial arch in the embryonic Cestracion
(Mrs. Hawkes, ‘ Journ. Anat. and Phys.,’ xi, 1905, pp. 81-84).

FISHES FROM THE COAST OF NATAL. )

or four spines in front of and three behind the orbit. One or
two median spines on the back in front of the supra-scapulary
region. A median series of spines commencing on the pos-
terior part of the body and extending on to the tail; tail with
one or two series of spines on each side. A large bluish-black
white-edged ocellus near the middle of the base of each
pectoral.

Two specimens of nearly the same size, 460 and 480 mm. in
total length, a male from Algoa Bay, received in 1895 from
H. A. Spencer, Esq., and a female from the coast of Natal, at
a depth of forty fathoms.

The male specimen has mixopterygia which extend only a
little more than half the distance from their base to the origin
of the first dorsal fin, a group of small spines on the anterior
part of each side of the pectoral, and a double series of spines
on each pectoral near the outer angle of the fin. The median
series of spines commences scarcely before the tail, which has
only one series of spines on each side.

The female specimen has a group of small spines on the
posterior part of the pectoral. The median series of spines
commences further forward than in the male, and the tail has
two series of spines on each side.

Raia rhizacanthus sp.n. (Pl. III).

Snout with a short, obtuse, triangular projection. Diameter
of eye 34 in the length of snout and 1+ in the interorbital
width. Distance between the outer edges of nostrils a little
less than their distance from the tip of snout. Teeth more
or less distinctly pointed, in 36 series in the upper jaw.
Anterior border of pectoral undulated. Snout, interorbital
space and anterior parts of pectoral fins with numerous small,
four-rooted spines. Two strong spines in front of and three
behind each orbit. A series of strong spines along the
median line of the back and tail; a pair of supra-scapulary
spines. ‘l'ail with two series of small spines on each side.
Upper part of the body brownish, with some large oblong or

A C. TATE REGAN.

oval lighter spots. An oblong, blackish, ocellated spot near
the middle of the base of each pectoral.

A single specimen, a young male, 210 mm. in total length,
from the Coast of Natal, at a depth of forty fathoms.

Clupea durbanensis sp.n. (Pl. IV).

Depth of body 22 in the length, length of head 33. Snout
as long as eye, the diameter of which is 4 in the length of
head, interorbital width 34. Lower jaw shutting within the
upper ; maxillary extending to below middle of eye or slightly
beyond ; no teeth on the palate or tongue. Giull-rakers fine,
long, and very numerous; lower branch of the anterior
branchial arch scarcely bent, the two portions meeting ata
very obtuse angle. Scales regularly arranged, finely striated,
and with the margin more or less distinctly ciliated, 43 to
45 in a longitudinal, 14 in a transverse series; 12 scutes
behind the ventral fins. Dorsal III 14, equidistant from
the tip of snout and from the procurrent caudal rays; the
longest ray equal to the length of the base of the fin and twice
as long as the last ray. Anal III 17-18. Pectoral 3-3 the
length of head, extending {—? of the distance from its base to
the origin of ventral, which is a little in advance of the middle
of the dorsal. Caudal deeply forked. A bluish humeral spot ;
upper edge of the dorsal fin blackish.

Two specimens, 190 and 200 mm. in total length, fron
Durban Bay.

This species is nearest to the West African C. dorsalis,
which has more scales in a transverse series, the lower branch
of the anterior branchial arch more distinctly bent, and the
post-orbital part of the head considerably longer than the
rest of the head, instead of equal to it, as in the species
described above.

Merluccius capensis Casteln.

Depth of body about 6 in the length; length of head
about 34. Diameter of eye 44-5} in the length of head,

FISHES FROM THE COAST OF NATAL. 3)

length of snout 3-3, interorbital width 32-34. Lower jaw
projecting; maxillary extending to or beyond the vertical
from the posterior margin of pupil; 15 or 14 gill-rakers
on the lower part of the anterior arch. Dorsal 10-11,
38-40. Anal 38-39. Pectoral 2-3 the length of head,
extending beyond the origin of anal. Ventrals 7-rayed, 2
(adult) to 2 (young) the length of head. Caudal truncate.
About 150 scales in a longitudinal series above the lateral
line, about 12 in a transverse series between base of first ray
of anterior dorsal and lateral line. Greyish; inner surface
of pectoral blackish.

Three specimens, 160 to 370 mm. in total length, from the
Cape of Good Hope and from Natal. This species is very
distinct from the Kuropean M. vulgaris, with which it has
been confounded.

Scorpena natalensis sp. n. (Pl. V).

Depth of body 22 in the length, length of head 22. Snout
longer than eye, the diameter of which is 5 in the length
of head and equal to the interorbital width. Jaws equal
anteriorly ; maxillary extending to below the middle of eye;
palatine teeth present. Head naked, except for a few rudi-
mentary scales. Interorbital space concave, with a pair of
weak ridges ; occiput with a quadrangular depression. A pair
of nasal spines ; each supra-orbital ridge with 3 spines; a
pair of spines near the extremities of the interorbital ridges;
on each side 2 spines in the parieto-occipital region and
below them 2 temporal spines; sub-orbital ridge with 3
spines leading to a double spine on the preoperculum, below
which are 3 preopercular spines; 9 gill-rakers on the lower
part of the anterior arch. Dorsal XII 10, the third and
fourth spines the longest, as long as the longest soft rays
and nearly half the length of head. Anal III 5, the second
spine scarcely longer than the third, one third the length of
head. Pectoral with 9 branched and 10 simple rays, 2 the
length of head, not quite reaching the origin of anal. Ventrals

6 C. TATE REGAN.

extending to the vent. Caudal rounded. Scales 48,%.
Lateral line 26. Body and fins spotted and marbled with
blackish.

A single specimen, 255 mm. in total length, from the Coast
of Natal, at a depth of forty fathoms.

Genypterus capensis A. Smith.

Depth of body 8 im the length, length of head 5. Snout
a little longer than eye, the diameter of which is 7 in the
length of head, interorbital width 73. Maxillary extending
well beyond the vertical from the posterior margin of eye,
the width of its distal extremity sheghtly greater than the
diameter of eye. Longest gill-rakers half the diameter of
eye; 4 and some rudiments on the lower part of the anterior
arch. Dorsal commencing in advance of middle of pectoral ;
pectoral 2 the length of head; longest ray of ventral nearly
+ the length of head. Scales more or less distinct on cheeks
and opercles, wanting on the upper surface of the head.
Thirteen series of scales between anterior dorsal rays and
lateral lme. Uniformly greyish.

A single specimen, 460 mm. in total length. This is the
first spirit specimen of this species to reach the British
Museum.

Uy J Tap useay ‘pe

iid

et TORSTEN A

“INHHUVM VINES LOTPId

TTICA SRW D [eveN UUY

Pale
a*

d.Green del.et lith.

LIGEE

a |
ey!

RATAD OCr

mi Natal Ge Mine. Vol. ik

Pica,

Green del et lith.

+
oe.

NTHUS

ZACA

|
i

RH

IA

R PA

lo NWeaNy aaah vara Ms

YNZ + Tep usedyg'pP

rN Goal ‘TTOA STIN'D TReYEN UY

PANE

Mas. Vol. I.

G,

Ann. Natal

lith.

J.Green delet

NATALENSIS.

SCORPAENA

Ae

9)
to

7. yo es
ao . 2 ‘i
OO a ae
Pry ie. « o
at a ay

BERTRAMIA KIRKMANIT. 7

On Bertramia kirkmani sp. nov.; a Myxo-
sporidium occurring in a South African
Rotifer.

By
Ernest Warren, D.Sc.Lond.,
Director of the Natal Government Museum.

With Plate VI.

In March, 1905, the Hon. Thomas Kirkman, of Natal, who
was working at rotifers in the laboratory of the Natal
Government Museum, showed to me a living specimen which
was apparently parasitised by contaimimg a number of gran-
ular sausage-shaped bodies lying unattached in the body-
cavity. The rotifer in question is a species of Copeus of
the Notommatide; it possesses the curious sac of chalky
matter found in several members of this family, and it is
characterised by a curious hump towards the posterior end
of the body. I beheve Mr. Charles Rousselet is about to
describe the species. This rotifer appears to be local in its
occurrence: Mr. Kirkman, its discoverer, has found it in the
neighbourhood of the Botanical Gardens, Pietermaritzbure,
and also at Richmond, Natal, some twenty miles distant.

Nearly the whole of the following account had been written
before I discovered that a similar organism in rotifers had
been previously partially described. Prof. E. A. Minchin!
in his recent article on Sporozoa gives some original figures
of the organism. It was called by L. Cohn? (1902) Ber-

1 +A Treatise on Zoology, edited by Prof. E. Ray Lankester, Pt. I,
1903.

2 + Zool. Anzeiger,’ xxv, 1902, pp. 497-502.

8 ERNEST WARREN.

tramia asperospora (Fritsch), and was originally described
by Bertram! in 1892. It will be seen that the species about
to be described differs in some important characters from
B. asperospora; but for the present the generic name may
be retained, and the species will be called after its discoverer
—Bertramia kirkman1.

(1) Structure and growth of the organism.—The
species of rotifer in which alone this parasite was found
lives in shallow pools among conferya, etc. It is by no means
a common species, but the percentage of parasitised indi-
viduals was exceedingly high. Out of some thirty-five speci-
mens collected and examined during the months of April,
May, and June about twenty imfected individuals were
discovered; that is, nearly 60 per cent. were infected. In an
allied species, Copeus spicatus, which was very common in
the same water, the parasite was never found ; it appeared to
be confined to the rarer species. During the same period a
number of specimens were collected at Richmond, Natal, but
all these were non-infected. It would thus appear that the
disease 1s prevalent in the particular spot near Pietermaritz-
burg where the species has been found for several consecutive
years.

The youngest stage found in the life-history of the parasite
occurred in only a single specimen of rotifer. The body-
cavity of the host contained some thirty irregularly-shaped
bodies, which were the youngest trophozoites discovered.
Some of these, the more curiously shaped ones im particular,
are shown in fig. 5. The smallest trophozoite in this rotiter
possessed two nuclei, and is shown in section in fig. 1. The
nuclei are relatively large; they are provided with a definite
nuclear membrane and a karyosome. In the living condition
the general cytoplasm im this early stage is exceedingly hya-
line, and there is but little distinction between ectosare and
endosarc. No amceboid movements were observed in any
stage of growth of the parasite.

'“ Beitrage zur Kenntnis der Sarcosporidien,” * Zool. Jahrb. Abth.
f. Anat.,’ V.

BERTRAMIA KIRKMANT. )

With reference to the technique employed, it should be
mentioned that the rotifers were fixed with hot corrosive
sublimate solution ;' they were then carefully imbedded in
parafiin, and serial sections of 44 4 in thickness were cut.
The sections were stained with iron-heematoxylin or Delafield,
followed by orange, eosin, or fuchsin. Flemming’s solution
was employed on two occasions, but the parasite becomes
intensely blackened with this reagent.

The trophozoite grows and the number of nuclei increases
by karyokinesis. It is rather remarkable that in the younger
stages, if not later, all the nuclei in the trophozoite are
approximately in the same condition at the same time; that
is, they are either all in the resting condition or all are
undergome karyokinesis. This phenomenon would seem to
demonstrate the controlling action of the general cytoplasm
on the nuclei.

The individual shown in section in fig. 2 contained about
six nuclei, and all of these were in a state of division. In
the process of karyokinesis the nuclear membrane gradually
disappears, and the chromatin becomes concentrated around
the circumference of the karyosome (fig. 17, 1-3). The chro-
matin now divides into two granular sheets (4), which become
associated with two vesicles, formed apparently by the
division of the original karyosome, and are connected
together by achromatic fibres (5). The chromatic plate
extends over the surface of the vesicle (6), and, finally, the
chromatic substance becomes uniformly dispersed (7). In
fie. 17 (8) the nuclear membrane is being separated off from
the nucleus, and ultimately the karyosome diminishes very
considerably in size (9). In fig. 4 it is seen that all the
nuclei are preparing for division; the nuclear membrane has
disappeared, and the chromatin has become concentrated
around the periphery of the karyosome. In fig. 6 the con-
dition is shown where the nuclear membrane is reappearing.

' A half-saturated solution of corrosive sublimate in 30 per cent.
spirit with 13 per cent. acetic acid.

10 ERNEST’ WARREN.

It will be noticed that all the nuclei in the trophozoite are in
the same stage at the same time.

The trophozoite is able to reproduce by fission, and also
by a kind of budding. In fig. 5 (3 and 7) the trophozoite is
about to divide by binary fission. The Y-shaped trophozoites
break up into three individuals. These phenomena have
been termed plasmotomy by Doflein, and although their
occurrence has been denied, they undoubtedly take place in
Bertramia kirkmani. An individual undergoing binary
fission is shown in section in fig. 3, and it should be noticed
that the fission appears to be in no way directly connected
with the division df the nuclei. Plasmotomy appears to be
only possible in the early stages of growth.

The trophozoites grow into sausage-shaped bodies, which
may be equal to half the length or more of the entire rotifer.
The whole of the body-cavity of the rotifer may become
almost completely choked with these bodies, which vary very
much in size, and I have counted as many as forty individuals
in one host.

As the trophozoite grows the protoplasm becomes charged
with refringent granules, which at first tend to be confined
towards the centre; but later on they extend throughout its
substance, and the hyaline protoplasm ultimately becomes
densely granular and opaque.

Fig. 7 shows a portion of an individual which has nearly
completed its growth: the protoplasm is fairly granular and
the nuclei are numerous. Ina large individual the number
of nuclei would be considerably over a hundred. It may be
noticed that the nuclei of the adult trophozoite are only about
half the size of those of the youngest trophozoite (cf. fig. 1
and fig. 7).

The trophozoite next passes into the spore-producing stage.
In fig. 8 it will be seen that the cytoplasm contains very
numerous large, refringent granules. The karyosome has
apparently become swollen, so as to occupy the whole area of
the nucleus, and the nuclear membrane is not distinct from
it (cf. figs. 17, 7, 8). The nucleus is now surrounded by an

BERTRAMIA KIRKMANT. 1]

area of protoplasm clear of granules (fig. 8,¢. a). At this
stage the trophozoite may become surrounded by a membrane
of very various thickness ; sometimes it is almost or quite
imperceptible (fig. 10); or it is thin, as im fig. 9; or it may
form a thick cyst-wall, as in figs. 12 and 15. In the same
host and at the same time we can find all these conditions
among the different trophozoites.

In the case of fig. 12 the individual was surrounded by a
thick eyst-wall (cy); but whether or not this is produced, the
mode ot formation of the spores is the same (c f. figs. 10
and 13).

In fig. 12 it can be seen that the densely granular tropho-
zoite is becoming divided up by trabeculee in a manner some-
what similar to that which occurs in Sarcocystis. Ulti-
mately the trabecule are so formed that each nucleus,
together with a certain amount of the cytoplasm, becomes
shut off from the remainder (fig. 9). The trabeculee will form
the future spore-membrane (fig. 11). In fig. 12 it may be
observed that at the time of killing a certain amount of the
trophozoite at the two ends (un. p.) had not been invaded
by the trabecule. Such a condition may frequently be
observed in Sarcocystis.

During spore-formation the large granules in the cyto-
plasm tend to disappear, and the substance of the trophozoite
becomes rather homogeneous in appearance. In the great
majority of cases which I have observed the whole substance
of the trophozoite is completely divided up into compart-
ments, each of which is converted ito a single spore. In
one or two cases, however, a certain amount of the trophozoite
had not become involved in the spore-formation, although
such spores as had been formed were completely ripe (fig. 14,
r. p.) and had mostly escaped out of the thin cyst.

These isolated cases may perhaps be regarded as a rever-
sion to a more typical Myxosporidium ancestor, for in the
great majority of the specimens observed such residual
plasmodium is not present, and it cannot be regarded as
normal in this species.

12 ERNEST WARREN.

(2) The formation of the spore.—The wall of the
trabeculee becomes the spore-membrane; the nucleus has a
somewhat irregular outline. The chromatin becomes concen-
trated towards the periphery (figs. 9 and 16, x), and ultimately
separates from the nucleolar element either in the form of a
band or of a hoop (fig. 18). There appears, however, to be
no inherent difference in these; for in either case the band
or hoop divides into two directly without mitosis (fig. 18),
and ultimately into six, eight, or ten (figs. 4 and 8) masses

of chromatin, which stain readily with safranin or hema-

ce b

toxylin. The vesicle or “ vacuole ” which remains (fig. 19, v)
does not stain at* all readily ; iodine has no particular effect
on it, neither has it any marked affinity for aniline stains or
hematoxylin. A minute granule can generally be found in
the vacuole towards the more rounded end of the spore
(fig. 19,g). The spore-membrane has a kind of operculum at
the narrower end (/), and such may be seen in many Myxo-
sporidium spores. Doubtless the young sporozoite escapes
through the operculum.

Fig. 19 represents a ripe spore. In the sporoplasm can be
seen some ten nuclei, each surrounded by a clear area, and
the “ vacuole” is situated towards the broader end.

Average length of spore is 6°1 wand breadth 2°9 w; the spore
is shg¢htly curved or falciform in shape, thus recalling the
spore of Sarcocystis.

(3) Life-history of the species.—Several infested
rotifers were kept alive in a small bulk of water and ex-
amined daily. The following is an account of one specimen.
On July 5th the rotifer contained about fifteen of the para-
sitic bodies, which were of various sizes, the average of four
of the largest being about 60 uw in length and 18 » in breadth.
The outer border of the bodies consisted of hyaline proto-
plasm ; internally they were shehtly granular.

On July 6th the number of the bodies had increased by
plasmotomy to about 25.

On July 7th the bodies were becoming opaque, owing to
the development of granules in the protoplasm.

BERTRAMIA KIRKMANI. L3

On July 8th the average size of the four largest bodies was
96 2 in length and 24 im breadth. The specimen was now
unobserved until July 14th, when all the bodies except one
had become divided up into a mass of spores without encyst-
ing. One specimen had become encysted ; 1t was rounded in
shape, and about 39” im diameter. The thickness of the
cyst was about 5°O nu. The protoplasm had not become divided
up ito spores.

On July 24th the encysted specimen burst, and the spores
were liberated. ‘The non-encysted individuals had broken
up into loose spores several days previously. The rotifer had
died on the 20th of the month. The cyst-wall is highly
elastic, for on the liberation of the spores the empty cyst
contracted to half of its previous size. This elasticity of the
cyst-wall must act very effectively in forcing out the contained
spores. From this account we can see that the rotifer must
have lived for more than three weeks in a parasitised con-
dition.

In a few days the body of the rotifer disintegrated, and
the spores were scattered through the water. Into the water
a non-parasitised specimen was introduced; but the animal
died in about a fortnight and did not become infected.

Another specimen containing trophozoites was observed on
June 21st. The host was dead on the 25rd, and the parasites
had completely broken up into loose spores on the 27th.

In the material at my disposal | have been unable to trace
the path of the sporozoite into the body-cavity of the host.
It is, of course, probable that the spores are swallowed, and
that the sporozoites escape through the operculum of the
spore-membrane and penetrate some part of the alimentary
canal; but im all my sections no trace of sporozoites in the
wall of the alimentary canal could be detected.

The youngest trophozoite found contained, as already
stated, two nuclei; while the number of nuclei in the sporo
plasm amounts to six to ten, on the supposition that the
isolated masses of chromatin above described are to be
regarded as nuclei. It would consequently follow that the

14 ERNES'1 WARREN.

greater number of these nuclei must be absorbed or fuse
together, on the transformation of the sporozoite imto the
trophozoite.

It is highly remarkable that the formation of a cyst-wall
is such a variable character. ‘The cyst may be quite thin, or
it may attain a thickness of 6 4 or more. Encysted and non-
encysted specimens frequently occur in the same host. The
character of the trophozoite and the spores ultimately pro-
duced appear to be quite the same in either case.

I have noticed that in the same host those specimens which
are encysted tend to form their spores more slowly than those
which are non-encysted. For example, in one case there
were some twenty non-encysted bodies which had completely
divided up into spores, while the remaining eight bodies,
which were surrounded by a thick cyst-wall, had not become
split up by the trabeculee.

[ have not had sufficient material to test the matter in a
practical manner, but I suggest that the thick cysts are
formed so that if the pool should dry up in which the host is
living the trophozoite enclosed would be able to resist a
considerable amount of desiccation. On falling into water
the encysted trophozoite would, without doubt, quickly break
up into spores which would be liberated by the bursting of
the cyst.

The spores themselves are only enclosed by a very thin
spore-membrane, and it is extremely improbable that isolated
spores could resist desiccation,

The cyst-wall is apparently homogeneous and is tough and
elastic. It tends to break up under the knife of the microtome.

(4) Affinities of the organism.—The present observa-
tions throw considerable light on the affinities of the organism.
The genus has been placed under the provisional order of
Haplosporidia, which includes a number of allied forms.

Bertramia kirkmani exhibits undoubted afhnities with
the Sarcosporidia, but it differs from them in that the
trabecule divide the trophoplasm into single spores, and
not into pansporoblasts. No trace of a polar capsule, or even

BERTRAMIA KIRKMANI. 1655

the striated area described in the spores of Sarcocystis, could
be detected. It differs from a typical Myxosporidium in that
spore-formation typically terminates the trophozoite stage,
the whole of the trophoplasm being converted into the spores.
In two or three cases, however, in about 100 specimens, the
whole of the trophoplasm was not involved in the spore-
formation. I consider that these cases are to be regarded as
a reversion to a more typical Myxosporidium ancestor, where
the formation of spores is indefinitely continued. The organism
may be regarded as having become adapted to the relatively
brief life of its host.

The so-called “vacuole” at the broader end of the spore
recalls the iodinophilous vacuole found in so many spores.
In Bertramia kirkmani it shows no particular affinity for
iodine, and it appears to arise from the original nucleus.

It is interesting, also, to note that ten chromatic masses, or
nuclei, are frequently found in the sporoplasm, thus recalling,
for example, the ten nuclei which are formed in the pansporo-
blasts of Myxobolus.

To conclude, Bertramia kirkmani is to be regarded as
being derived from a Myxosporidium ancestor, in which the
typical structure and hfe-history have become modified and
simplified, in accordance with the special conditions occurring
in its comparatively short-lived host.

EXPLANATION OF PLATE VI,

Illustrating Dr. Ernest Warren’s paper on “ Bertramia
kirkmani sp. noy., a Myxosporidium occurring in a
South African Rotifer.”

Fre. 1.—x 2000 diameters. Youngest trophozoite found, possesses
two nuclei (v.) and distinct nuclear membrane (n. m.).

Fie, 2.—x 2000 diameters. Trophozoite in which the nuclei are in
a state of division. d. Dyaster stage showing achromatic fibres.
d.n. Daughter nuclei after the disappearance of the achromatic fibres ;
the chromatin is extending around the periphery of the nucleus.

16 ERNEST WARREN.

Fig. 3.— x 2000 diameters. Trophozoite undergoing fission at the
place of constriction. The trophoplasm is vacuolated, and granules
(s.gr.) are beginning to appear in it.

Fie. 4.— x 2000 diameters. Trophozoite in which all the nuclei are
preparing for division ; the nuclear membrane is disappearing (d.n. m.),
and the chromatin is becoming concentrated around the periphery oi
the karyosome (n.d.). Ectosare and endosare are distinguishable.

Fic. 5.— x 300 diameters. <A series of young trophozoites, all taken
from one host. These exhibit the phenomena of plasmotomy both in
the form of fission and of budding.

Fic. 6.—x 2000 diameters. Older trophozoite with numerous large
and small granules (l. gr. s. gv.) in the endosare. The nuclei are just
passing into a resting-condition, the nuclear membrane (7. m.f.) is being
formed.

Fic. 7.—x 2000 diameters. A portion of a nearly full-grown tropho-
zoite. Nucleus has a well-marked nuclear membrane and karyosome.
It should be noticed that the nuclei in the adult stage are not more
than half the size of those in the younger trophozoites (Figs. 1—6).

Fic. 8.— x 2000 diameters. A portion of a full-grown trophozoite.
The trophoplasm is densely crowded with granules (g7.).. The karyosome
of the nucleus has swollen, and the nuclear membrane has disappeared.
The nucleus is surrounded by a layer of clear protoplasm containing no
granules (c.a.). The opaque trophozoite is now in a preparatory con-
dition for the formation of spores.

Fie. 9.—x 2000 diameters. Section through trophozoite, showing
the formation of the trabecule (¢. and ¢.f.).. The whole of the tropho-
plasm had not become involved (wi. p.) at the time of killing and fixing
the specimen. In several of the areas two nuclei can be seen ; doubtless
these would subsequently have divided. Each area will produce one
spore. The nucleus is without nuclear membrane (7.), and the chromatin
is concentrated towards the periphery.

Fie. 10.—x 2000 diameters. Surface view where the whole of the
trophozoite has been converted into spores.

Fie. 11.—x 5500 diameters. Surface view, still further enlarged ;
the spores are nearly ripe, and are seen to contain a ‘“ vacuole ” (v.) and
small nuclei (7.).

Fic. 12— x 1100 diameters. Elongated trophozoite surrounded by
a thick cyst (cy.). The trophoplasm is beginning to be divided up by
trabecule (¢.f.). The two ends (wn. p.) are still undivided, thus some-
what recalling the condition seen in Sarcocystis.

Fia. 15.— x 2000 diameters. Rounded encysted specimen with ripe
spores.

BERTRAMIA KIRKMANI. Uz.

Fie. 14.— x 2000 diameters. This trophozoite was surrounded by a
thin cyst (f. cy.), and it contained ripe spores and also a considerable
amount of residual trophoplasm (7.p.). The presence of residual
trophoplasm is apparently quite abnormal in this organism ; and pro-
bably it should be regarded as a reversion to a more typical myxo-
sporidium ancestor.

Fra. 15.—x 2000. A cyst which has burst and liberated the spores
by its elasticity.

Fra. 16.— x 4000 diameters. Youngest spore, drawn from a specimen
in the same condition as Fig. 9. The nucleus (i.), which is the same as
the nuclei in Fig 8, 7., has no nuclear membrane, and the chromatin
lies around the periphery.

Fic. 17.—x 2000 diameters. (1) Nucleus in youngest trophozoite
(Fig. 1). (2) Nucleus preparing to divide, nuclear membrane dis-
appearing (Fig. 4). (3) Nucleus preparing to divide. (4) Nucleus pre-
paring to divide, the chromatin around the periphery dividing into two
plates (Fig. 4). (5) Chromatin plates separated and connected together
by achromatic spindle (Fig. 2). (6) One of the daughter nuclei (Fig. 2,
d.n.). (7) Daughter nucleus passing into the resting condition. (8)
Formation of nuclear membrane (Fig. 6). (9) Resting nucleus of adult
trophozoite; it is about half the size of (1) (Fig. 7). Preparatory to
the formation of spores the karyosome swells up until it reaches about
the diameter of the whole nucleus (7). The nuclear membrane is not
distinct (Fig. 8).

Fie. 18.—x 4000 diameters. (1-4) Stages in the formation of the
6-10 nuclei of the ripe spore; the chromatin first appearing in the form
of a band. (5-8) Stages in the formation of the 6-10 nuclei of the ripe
spore ; the chromatin first appearing in the form of a ring or hoop.

Fie, 19.— x 8000 diameters. Ripe spore showing operculum (/.), ten
nuclei (i.), so-called * vacuole” (v.), and problematical granule (g.).

EXPLANATION OF REFERENCE LETTERS.

c.a. Clear area around nucleus. cy. Thick-walled cyst. ch. b. Chro-
matie body of spore. d. Dyaster stage in dividing nucleus. d.n.
Daughter nuclei after division. d.n.m. Disappearing nuclear mem-
brane. ec. Ectoplasm. en. Endoplasm. g. Granule imbedded in
vacuole of spore. gv. Granules in endoplasm. k. Karyosome. J. Lid
or operculum of spore membrane. J/.gr. Large granule surrounded by
clear area. n. Nucleus. n.d. Nuclei preparing for karyokinesis. x. m.
Nuclear membrane. n.m.f. Nuclear membrane forming. +. p. Residual
trophoplasm. s.gr. Small granules. sp. Spore. sp.w. Spore-wall.
t. Trabecula. t.cy. Thin-walled cyst. ¢.f. Trabecule forming. wn. p.
Undivided trophoplasm. v. So-called vacuole of spore. v,. Vacuole of
trophoplasm.

WOE i, PARD |, 2

Ann. Natal G Mus. Vol.I.

KVI x 4000

GPoognoe@e®

Nae

XXVIL x 2000 ih

a»)
~

SOUTH AFRICAN MARINE MOLLUSCA. 19

‘

On South African Marine Mollusca, with
Descriptions of New Species.

By
Edgar A. Smith, 1.8.0., F.Z.8.

Plates VII, VIII.

In the ‘Proc. Malac. Soc., 1903, vol. v, pp. 354-402, I
gave a list of species from South Africa not recorded in
Mr. Sowerby’s work entitled ‘Marine Shells of South
Africa” In the present paper I propose to bring the
catalogue of known species up to date, and also to make
corrections of a few errors which, unfortunately, occur in
the lst referred to, and to supply some omissions. These
errors and omissions have mostly been very kindly pointed
out to me by my friend Mr. H. C. Burnup.

I now also quote in full, giving references, the species
mentioned and described by Mr. Sowerby im the ‘ Marine
Investigation in South Africa,’ vol. u, pp. 213-252, which
were merely listed in the above-mentioned paper.

By consulting, therefore, the present paper, that in the
‘Proc. Malac. Soc., and Mr. Sowerby’s work, a complete list
of the known South African species can be extracted.

Those not in Mr. Sowerby’s book or my former paper are
marked with an asterisk in the following pages.

The principal papers dealing with the South African
fauna published since 1903 are the following :

(1) “Mollusca of South Africa. Pelecypoda.” In ‘ Marine

20 EDGAR A. SMITH.

Investigations in South Africa,’ vol. iv, pp. 1-19, Pls. VI,
Vik By GiBy Sowerby.

(2) “Ona Collection of Marine Shells from Port Alfred,
Cape Colony.” In the ‘Journal of Malacology, vol. xi,
pp. 21-44, Pls. I, HI. By Edgar A. Smith.

(3) “Die beschalten Gastropoden der deutschen Tiefsee-
Expedition, 1898-1899,” in Chun’s ‘ Wissenschafthche Ergeb-
nisse der deutschen Tiefsee-Expedition auf dem Dampfer,’
“Valdivia,” 1898-1899, vol. vu, pp. 26-60, Pls. II-V. By
K. von Martens.

Cephalopoda.

*Sepia burnupi Hoyle.

Sepia burnupi Hoyle, ‘J. of Conch.,’ vol. xi, p. 27, Pl. I.
Hab.—Umkomaas, Natal, and Port Ehzabeth, Cape Colony.

Gastropoda.

*Oncidium peronii Cuvier.
Onchidium peronii Cuvier: Krauss, ‘Siidafrik. Moll.,’
p. 72.
Hab.—Natal coast (Krauss).

*Oncidium burnupi Collinge.
Onchidium burnupi Collinge, ‘J. of Malac.,’ vol. ix
Peet eos ls. 22
Hab.—Umlaas Lagoon, Natal.

*Ampullarina africana Smith.

Ampullarina africana Smith, ‘J.of Malac.,’ vol. x1,
238, Pl. SLY fie... 14.
Hab.—Port Alfred, Cape Colony.

—
Y

SOUTH AFRICAN MARINE MOLLUSCA.

bo

Retusa truncatula (Bruguwiere).
Retusa truncatula (Bruquiere): Smith, ‘J. of Malac.,’
vol. xi, p. 38.
Hab.—Port Alfred, Cape Colony.

Scaphander punctostriatus Mighels.

Scapander punctostriatus Mighels: Pilsbry, ‘Man.
Conch.,’ vol. xv, p. 246, Pl. XXXI, fig. 16; Sowerby, ‘ Marine
Invest.,’ vol. 11, p. 233.

Hab.—Off Cape Colony, 154 and 166 fathoms.

* Dolabella scapula (Martyn).

Dolabella scapula Martyn: Pilsbry, ‘Man. Conch.,’
vol. xvi, p. 152, P]. XX VI, figs. 26-28, Pl. XX VII, figs. 29, 30.

Hab.—Natal.

Martyn’s scapula has priority over D. rumphii, under
which name this species has been quoted by Krauss and
Sowerby.

*Terebra suspensa Smith.

Terebra suspensa Smith, ‘J. of Malac.,’ vol. xi, p. 30,
Bl foe 12:
Hab.—Port Alfred, Cape Colony.

*Conus geographus Linn.
Conus geographus Linn.: Reeve, ‘Conch. Icon.,’ vol. 1,
fig. 150.
Hab.—Bluff, Durban (G. W. Westcott).

* Conus punctatus Gmelin.
Conus punctatus Gmelin, ‘Syst. Nat.,’ p. 3389.
Conus augur, Hwass: Reeve, ‘Conch Icon.,’ vol. i, fig. 7;
Kiener, ‘ Cog. viv.,’ Pl. XVIII, fig. 3.
Hab.—Durban (Burnup), Ceylon, ete.
Conus punctatus of Hwass, also occurring in Natal, is

22. EDGAR A. SMITH.

a different species, and now bears the name of C. piperatus,
Dillwyn.

*Conus bandanus Hwass.
Conus bandanus Hwass: Reeve, ‘ Conch. Icon.,’ vol. i,
fig. 43.
Hab.—Bluff, Durban (G. W. Westcott).

Conus eucoronatus Sowerby.

Conus eucoronatus Sowerby, ‘Marine Invest.,’ vol. 1,
oh Calls TENS NS ace
Hab.—Off Cape St. Blaize, Cape Colony, 27 fathoms.

Conus gilchristi Sowerby.
Conus gilchristi Sowerby, ‘Marine Invest.” vol. un,
Oa He SES SUS steaites
Hab.—Off Natal, 50 fathoms.

Conus patens Sowerby.
Conus patens Sowerby, ‘Marine Invest.,’ vol. u, p. 218,
Pl fied.
Hab.—Off S. Africa, 85 fathoms.

*Conus queketti nasp. Pl. Vil fie 1

Testa parva, elongato-turbinata, supra depressa, coronata,
sordide flavescens lineis albis transversis ineequalibus nume-
rosis fusco punctatis ornata, transversim tenuiter sulcata,
sulcis confertim et minute punctatis, subsquidistantibus,
lineisque incrementi tenuibus striata; spira perpaulum elata,
ad apicem mucronata; anfractus 8-9, lente accrescentes,
supremi duo (protoconcha) convexi, ceteri angusti, fere
plani vel leviter concavi, coronati, stris spiralibus paucis
et lineis incrementi curvatis sculpti, ultimus antice oblique

sulcatus et fusco tinctus; apertura angusta, alba. Longit.
26 mm., diam. 12:5.

Hab.—Isezela, Natal.

SOUTH AFRICAN MARINE MOLLUSCA. 23

This is a very distinct species and not comparable with any
of the known forms. Of the transverse white lines dotted
with brown about a dozen are conspicuous to the naked eye,
but the narrower intervening ones are hardly visible except
with the aid of a lens. The spire, which is very little raised,
is whitish, streaked and spotted irregularly with brown.
The apex is peculiar, consisting of two convex convoluted
whorls which rise as a sort of mamilla above the rest.

*Clionella confusa n.sp. Pl. VII, fig. 2.

Testa C. rosariz similis, sed major, anfractibus plerumque
minus conyexis, striis spiralibus vix conspicuis, cingulo infra
suturam majori, colore diversa. Longit. 45 mm., diam. 14.
Aperture 12 longa, 6 lata.

Hab.—Port Elizabeth, Algoa Bay.

Shell elongate, of an uniform bright red or yellowish
colour ; whorls 10? (apex broken away), rather flat, con-
stricted at the upper part, leaving a rounded girdle beneath
the suture, with about 14 sheghtly oblique costw below the
constriction, by which they are interrupted, so that they are
only faintly indicated upon the cingulum above ; ribs upon
the body-whorl attenuated anteriorly, but disappearing upon
the snout; suture a little oblique and wavy ; aperture more
or less rosaceous within; columella curved, covered with a
thin whitish callus; outer lip distinctly but not deeply
notched at the constriction.

This species has been confounded with C. rosaria, Reeve,
because of its bright red colour. Reeve’s species, however,
appears invariably to have the cingulum at the upper part of
the whorls whitish, and more or less spotted with brown.
The whorls of that species are more convex, and are distinctly
transversely striated. It appears to be smaller also, the
largest specimen examined being only 25 mm. in length. It
is more variable in colour, sometimes bright red with a white
oirdle, sometimes brown, mottled with white, or yellowish,
flecked with white dots.

24, EDGAR A. SMITH.

The so-called species of Clionella are not always easy to
distinguish, and beyond a difference of colour and_ the
absence of spiral striation, there is little to distinguish the
present new form from that which I described under the
name of C. Bornii!. The remains of the periostracum in

©. confusa are dark or blackish brown.

* Clavatula impages (Adams and Reeve).
Clionella impages Ad. and f#.: Martens, ‘ Tiefsee-
Kixped.,’ p. 23.
Hab.—Agulhas Bank, 155 fathoms, and Port Elizabeth,
dead.
Smaller, but allied to C. taxus, Chemn.andtumida, Sow.

Pleurotoma lobata Sowerby.
Pleurotoma (Surcula) lobata Sowerby, ‘Marine
Imvest., vol.u, p. 213, Pl. LV, fig. 9.
This species, in my opinion, belongs to true Pleurotoma,
and should not be referred to the sub-genus Surcula.

Hab.—Off Cape Natal, 440 fathoms.

Clavatula turriplana (Sowerby).
Pleurotoma (Clavatula) turriplana Sowerby, ‘Marine
Invest. vol. up. 205) bl tie nes 6:
Hab.—Off Cape St. Blaize, Cape Colony, 85-90 fathoms.

Drillia (Clavus) lignaria (Sowerby).
Fleurotoma (Clavus) lignaria Sowerby, ‘Marine
Imvests, vol. u,p. Zi, i, tie. 4:
Hab.—Off S. Africa, 136 fathoms.

Genotia beleformis (Sowerby).
Pleurotoma (Genotia) belewformis Sowerby, ‘Marine
Invest.,’ vol. 1, p. 216, Pl. IV, fig. 8.
Hab.—Off S. Africa, 230 fathoms.

1 * Ann. Mag. Nat. Hist.,’ 1877, vol. xix, p. 499.

SOUTH AFRICAN MARINE MOLLUSCA. 25

*Drillia albonodulosa Smith.
Drillia albonodulosa Smith, ‘J. of Malac.,’ vol. xi, p. 27,
Blo phe 3.
Hab.—Port Alfred, Cape Colony.

Drillia scitecostata (Sowerby).
Pleurotoma (Drillia) scitecostata Sowerby, ‘Marine
Invest.,’ vol. 11, p. 214, Pl. IV, fig. 10.
Hab.—Near Port Alfred, Cape Colony. 100 fathoms.

Drillia fossata (Sowerby).
Pleurotoma (Drillia) fossata Sowerby, ‘Marine Invest.,’
"OL Hapa 2146 PIL fio. 3.
Hab.—Cape Vidal, Natal, 80-100 fathoms.

*Drillia nivosa Smith.
Drillia nivosa Smith, *J. of Malac., vol. xi, p. 27, Pl. I,
Heo.

Hab.—Port Alfred, Cape Colony.

* Drillia pretermissa Smith,
Drillia pretermissa Smith, ‘J. of Malac.,’ vol. x1, p. 27,
Pl. II, fig. 4.
Hab.—Port Alfred, Cape Colony.

* Drillia subcontracta Smith.
Drillia subcontracta Smith, ‘J. of Malac.,’ vol. x1, p. 26,
Pl U, fey 2:
Hab.—Port Alfred, Cape Colony.

* Drillia thetis Smith.
Drillia thetis Smith, ‘J. of Malac.,’ vol. xi, p. 26, Pl. II,
fig 1
Hab.—Port Alfred, Cape Colony.

26 EDGAR A. SMITH.

* Drillia albotessellata n.sp. Pl. VII, fig. 3.

Testa elongata, turrita, supra subacuminata, imperforata,
olivacea, maculis parvis subquadratis czeruleo-albidis tessel-
lata, circa medium anfractus ultimi olivaceo obscure zonata;
anfractus circiter 8, supra excavati, infra costis rotundatis
9-10 leviter obliquis instructi, supra costas striis transversis
(in anfr. penult, 4-5, in ultimo circiter 18) vix punctatis
sculpti, incrementi lineis tenuiter striati, ultimus infra
medium pallidus, ad extremitatem fusco tincta ; apertura intus
saturate fusca; labrum tenue, intus flavescens, supra medio-
criter profunde sinuatum; columella parum arcuata, callo
crasso flavo supra tuberculato induta. Longit. 18 mm.,
diam. 6. Apertura 7°5 longa, 2°5 lata.

Hab.—Port Shepstone (Burnup).

This Drillia is remarkable on account of its peculiar
coloration. The ground tint is a sort of olive upon which
there is a tesselation of squarish whitish dots. The concavity
at the upper part of the whorls is more or less brown, and the
shght thickening at the suture is minutely dotted and spotted
with brown. ‘The aperture is very deeply coloured within, so
that the labrum and columella, being yellowish, contrast
strongly. Around the middle of the body-whorl a more or
less obscure olivaceous zone is observable.

* Clathurella crassilirata Smith.
Clathurella crassilirata Smith, ‘J. of Malac.,’ vol. x1
Pai el Lite. 6:
Hab.—Port Alfred, Cape Colony.

* Glyphostoma siren Smith.
Glyphostoma siren Smith, ‘J. of Malac.,’ vol. xi, p. 28,
Pi, fice 7.
Hab.—Port Alfred, Cape Colony.

* Mangilia alfredi Smith.
Mangilia alfredi Smith, ‘J. of Malac., vol. xi, p.
Piri, fies s:
Hab.—Port Alfred, Cape Colony.

bo
Ne)

SOUTH AFRICAN MARINE MOLLUSCA.

bo
“NI

Mangilia africana Sowerby.
(Hucythara) africana
Inivest.;; vol.11, p--216, Pl. V; fig. 9

Mangilia

Sowerby, ‘ Marine

Hab.—Off mouth of Umhloti River, Natal, 25 fathoms.

Cancellaria producta Sowerby.
Cancellaria producta Sowerby, ‘ Marine Invest.,’ vol. u,
e220) el DVestion oS:

Hab.—Off Natal, 40 fathoms.

* Ancilla albozonata Smith.
Ancilla albozonata Smith, ‘J. of Malac.,’ vol. xi, p.29
BLL ties 9:

Hab.—Port Alfred, Cape Colony.

Ancilla bulloides (Reeve).
Ancilla bulloides

Reeve, Sowerby, ‘Marine Invest.,’
vol. 11, p. 228.

Hab.—South Africa, 190 fathoms.

Ancilla contusa (Reeve).
Ancilla econtusa Reeve

: Sowerby, ‘ Marine Invest.,’
wolenips22o, Pl, fie<3

A good figure of this species, described originally from a
somewhat abnormal specimen, is given by Mr. Sowerby.

* Ancilla reevei Smith.
Ancilla reevei Smith, ‘J. of Malac.,” vol. xi, p. 29
Pil, tie. 10.

A>}

Hab.—Port Alfred, Cape Colony.

* Ancilla ordinaria n.sp. Pl. VII, fig. 4.
Testa parva, subovata, supra acuminata, dilute flavescens,
stuigis vel lneis, pallidis, parum obliquis, obscuris, ornata,
interdum omnino nivea; anfractus 4—5, celeriter accrescentes;

28 EDGAR A. SMITH.

spira brevis, ad apicem obtusa; anfr. ultimus antice bisul-
catus, sulco superior supra labrum denticulum inconspicuum
formante, ad extremitatem alteram ad partem aperture
superiorem terminante ; labrum supra incrassatum, incurvum,
ad insertionem leviter incisum; columella antice callosa,
reflexa, leviter sulcata.

Longit. 13 mm., diam. 7; apertura 8 mm. longa, 3 lata.

Hab.—Port Shepstone (Burnup).

Rather like A. sarda Reeve in shape, but differmg in
having a second groove upon the body-whorl higher up than
the one around the anterior end of Reeve’s species. The
habitat of the latter was unknown to its author, but a nice
series of it from Zanzibar was received by the British Museum
from Mr. J.T. Last. Two specimens of the present species
have been examined—the one entirely snow white, the other,
viewed at a distance, apparently of an uniform pale-yellow
colour, a little darker at the suture and upon the body-whorl
below the upper groove. Ona closer inspection the yellow
tint is broken up by slightly oblique whitish streaks.

*Ancilla hasta (Martens).
Ancillaria hasta Martens, ‘Tiefsee-Exped., p. 37,
Prairie: Us:
Hab. —Agulhas current, 270 fathoms.

*Marginella labrosa Redfield.
Marginella labrosa Redfield, ‘Amer. Jour. Conch.,’
vol. vi, p. 239.
This name must be employed for the species catalogued by
Mr. Sowerby (‘ Marine Shells 8. A.,’ p. 21) as M. crassila-
brum, as that name was already twice preoccupied.

Marginella angustata Sowerby.
Ancilla angustata Sowerby, ‘Marine Invest., vol. u,
p. 229.
Hab.—Off Cape Point, 42 fathoms.
By a curious accident this species was wrongly recorded
by Mr. Sowerby under the genus Ancilla.

iw)
<©

SOUTH AFRICAN MARINE MOLLUSCA.

Marginella biplicata Krauss,

Marginella biplicata Krauss: Martens, ‘ Tiefsee-
Hecped.; Eps. 37,7 Ib. tie, 6.

Hab.—Cape Coast (Krauss).

This and the four following specics were figured for the
first time in the above work. M. chrysea of Watson
(Sowerby, ‘Marine Shells 8. A.,’? p. 20) is a synonym of
this species. It was described as having three columellar
folds, and on this account Martens hesitated to unite it with
biplicata. Having examined the types of chrysea and
compared them with the figure of biplicata given by
Martens, I feel no hesitation in pronouncing them the same
species. One of the two “ Challenger” examples of chrysea
has a very faint third fold, but the other specimen has
practically only the shghtest trace of it, so that if this existed
in Krauss’s examples, it was probably overlooked or not
considered worthy of notice. The yellowish tint described
by Watson has almost faded away, so that the shore speci-
mens, such as Krauss probably had before him, would most
likely be destitute of colour also.

Marginella zeyheri Krauss.

Marginella (Gibberula) z
‘ Tiefsee-Exped.,’ p. 34, Pl. III

Marginella pura Smith, ‘
Pe JUL Seared is)

Hab.—Offt Cape, 177 fathoms (Martens); Port Alfred
(Smith).

When describing this species, under the name of M. pura,

eyheri Krauss: Martens,
, fig. 4
J.

of Malac.,’ vol. xi, p. 31,

I had not seen Martens’ figure of zey heri, and from Krauss’s
original description, unaccompanied by any figure, it was
impossible to know what the form of his shell really was.

Marginella neglecta Sowerby.

Marginella (Gibberula) reevei Krauss : Martens,
‘Tiefsee-Exped.,’ p. 35, Pl. III, fig. 3.

30 EDGAR A. SMITH.

Hab.—Agulhas Bank, 84 fathoms.

M. reevei Krauss is certainly synonymous with the present

species, which has six years’ priority of publication.

Marginella multizonata Krauss.

Marginella (Volvarina) multizonata Krauss: Mar-

tens, ‘ Tiefsee-Exped.,’ p. 36, Pl. III, fig. 5.
Hab.—NSimon’s Bay, Cape Colony, 38 fathoms.

This species appears to be very hke M. cylindrica of

Sowerby.

*Marginella corusca Reeve.
Marginella corusca Reeve, ‘Conch. Icon.,’ vol.
figs. 143 a, b; Smith, ‘J. of Malac.,’ vol. xi, p. 23.
Hab.—Port Alfred, Cape Colony ; Singapore (Reeve).

*Margineélla differens Smith.

Marginella differens Smith, ‘J. of Malac.,’ vol.
pod, EL JEL) fig. L9;
Hab.—Port Alfred, Cape Colony.

*Marginella dulcis Smith.

Marginella dulcis Smith, ‘J. of Malac.,’ vol. xi, p.
Pl UL-fig. 20:
Hab.—Port Alfred.

*Marginella munda Smith.

Marginella munda Smith, ‘J. of Malac.,’ vol. x1, p.
Pie: 14.
Hab.—Port Alfred.

*Marginella pseustes Smith.

Marginella pseustes Smith, ‘J. of Malac.,’ vol. x1, p. :

Pl Us fie. 21.
Hab.—Port Alfred.

XV,

Sl,

Be

wie,

SOUTH AFRICAN MARINE MOLLUSCA. 3]

*Marginella ros Reeve.

Marginella ros Reeve, ‘Conch. Icon.,’ vol. xv, fig. 147;
Martens, ‘ Tiefsee-Exped.,’ p. 36 (in section Granula).
Hab.—Agulhas Bank, 85 fathoms.

* Marginella shepstonensis n.sp. PI. VII, fig. 5.

Testa parva, oblonga, ovata, alba, lineis longitudinalibus
undulatis vel angulatis, flavis, et zonis duabus transversis
interruptis ornata; spira obtusa, haud elata; anfractus tres,
ultimus convexiusculus, latere sinistro dextro convexiore ;
apertura angusta, alba; labrum intus incrassatum, tenuiter
liratum; columella callo tenui circumscripto induta, denti-
culis vel plicis circiter decem instructa.

Longit., 6 mm.; diam., 5°6 mm.

Hab.—Port Shepstone.

The coloration of this species is rather like that of M.
pulchella Kiener (‘ Coq. Viv.,’ Pl. EX, fig. 40), and the form
also is very similar. That species, however, is said to have a
smooth labrum and only four columellar folds. The angula-
tions of the zigzag lines form transverse zones, but only two
of these, one round the middle and one at the upper part of
the body-whorl, are specially conspicuous, being darker in
colour.

Marginella zonata Kiener.

The variety bilineata Krauss is said by Martens (‘Tiefsee-
Exped., p. 57) to be the Voluta biannulata of Fabricius.

Voluta (Volutocorbis) abyssicola Adams and Reeve.

Voluta (Ternivoluta) abyssicola Adams and Reeve:
Martens, ‘'Tiefsee-Exped.,’ p. 51.

As differing, in his opinion, from Volutilithes, to which
genus it has often been assigned, Martens placed this species

in his sub-genus Ternivoluta (1903), the character of its

3 EDGAR: A. SMITH.

radula being the same as that of that group. The style of
its sculpture, however, is very different. Volutocorbis,
Dall (1890).

Fusivoluta pyrrhostoma (Watson).
Neptuneopsis pyrrhostoma (Watson): Sowerby,
‘Marine Invest.,’ vol. ii, pp. 218, 226, Pl. III, fig. 1.
Fusivoluta pyrrhostoma (Watson) : Martens, ‘ Tiefsee-
Exped.,’ p. 32, Pl. III, fig. 15.

Mitra punctostriata A. Addams. Pl. VU, fig. 6.

Mitra punctostriata A. ddams: Smith, ‘Proc. Malac.
Soc.,’ vol. v, p. 366.

? Mitra cylindracea Reeve: Sowerby, ‘ Marine Invest.,’
vol. a,.p.22¢.

Considered by Sowerby to be the same as M. cylindracea
Reeve. The type of M. cylindracea, said to be in the
Cuming collection, cannot now be found. Judging from the
description and figure, I do not feel convinced of the identity
of these two forms. In the South African specimens, which I
have examined, the upper whorls exhibit five punctured lines,
whereas in Reeve’s figure of cylindracea only three are
indicated. Also the ‘interrupted band of spots round the
middle” of the body-whorl is wanting in these shells, which
agree in every detail with punctostriata, the type of which
is in the British Museum collection.

* Mitra paupercula (Linn).
Mitra paupercula Lunn.: Reeve, ‘Conch. Icon.,’ vol.
ul, fig. 84.
Hab.—Durban (Burnup); also recorded from the Red

~

Sea, E. Africa, Philippines, Polynesia.

Mitra schreteri Dillwyn.
This species is considered the same as M. picta Reeve
and tessellata Kiener by Martens (‘ Tiefsee-Exped.,’ p. 53).

SOUTH AFRICAN MARINE MOLLUSCA. 33

Dillwyn appears to have been the inventor of the specific
name schroeteri and not Chemnitz. At present I am not
convinced that the two forms schroeteri and picta as
determined by Reeve (‘ Conch. Icon.,’ vol. 11, figs. 167 and 123
respectively) are absolutely identical. The former seems
somewhat shorter and to have a broader aperture, the labrum
also having a tendency to be a little patulate. <A difficult
question arises as to which form is the true schreteri, but
at present I am inclined to agree with Martens that the shell
known as picta of Reeve is that figured by Schroeter and
afterwards called Voluta schreeteri by Dillwyn. If this
be correct, then the Mitra schroteri of Reeve, if distinct,
would require a fresh name. Until I have the opportunity of
examining a larger series of specimens in good condition I do
not think it advisable to attempt any further solution of the
difficulty.

Mitra simplex Dunker.

With this species Martens (‘Tiefsee-Exped.,’ p. 53) unites
cinnamomea A. Adams. The latter, however, I have
reason to beleve is Tasmanian, and does not occur in
Natal, as stated by Adams. There are numerous specimens
in the British Museum from Oyster Cove, Tasmania, agreeing
exactly with the types in every respect.

Mitra (Turricula) dedala Reeve.

Mitra dedala Reeve: ‘Conch. Icon.,’ vol. ii, fig. 281.

Mitra (Costellaria) dwedala: Sowerby, ‘Marine Invest.,’
vol. u, p. 227.

Hab.—Off Scottsburg, Natal, 92 fathoms; Philippine
Islands (Reeve).

*Fusus cingulatus Smith.

Fusus cingulatus Smith: ‘J. of Malac., vol. xi, p. 30,
Plaine. 1;
Hab.—Port Alfred.

you. J, parr, 1.

wy)

34 EDGAR A. SMITH.

Fasciolaria rutila Watson.

Fasciolaria rutila Watson: Sowerby, ‘Marine
Invest.,’ -vol. 11, p. 222, Pl. Il, fie, 2:
Hab.—Off Natal, 40 and 154 fathoms.
A young shell and the radula are figured by Sowerby.

“Latirus burnup: msp, Pl Vil hom7:

Testa breviter fusiformis, albida, periostraco fuscescente
tenui induta; spira acuminato-conica; anfractus 8-9, supra
declives et leviter concavi, dein convexi, costis 8 fortibus
rotundatis supra attenuatis instructi, liris tenuibus spiralibus,
confertis, infra suturam plus minus granulatis, ornati, ultimus
antice attenuatus, breviter caudatus, costis infra medium
sensim evanidis, lira unica paulo infra peripheriam czeteris
magis conspicua; apertura roseo-purpurea, intus tenuiter lirata,
cum canal longit. totius $ adequans; labrum pallidum,
subexpansum, antice fere patulum; canalis parum obliquus ;
columella callo roseo-purpureo induta, in medio plicis tribus
tenuibus instructa, tuberculo parvo unico superme munita.

Longit. 28 mm., diam. 11°5; apertura cum canali 14 longa,
5 lata.

Hab.—Port Shepstone (Burnup).

Somewhat resembling L. flavidus A. Adams and L.
marie Crosse, but apparently distinct. The termination of
the more conspicuous lira upon the body-whorl is near the
columellar tubercle at one end and is indicated at the other
by a faint denticle upon the lower edge of the outer lip. All
four specimens examined are rather worn at the apex and
upon the anterior rostrum. It looks as if the periostracum is
soon lost at these parts.

Cominella lagenaria (Lamarck).

Under this species Mr. Sowerby (‘ Marine Shells 8. A.,’
p. 10), has placed Purpura dubia Krauss as a variety.
This is incorrect, as Krauss’s species certainly belongs to

SOUTH AFRICAN MARINE MOLLUSCA. 30

Purpura, and is the same as the P. cataracta of Reeve.
The Bueccinum cataracta of Chemnitz has never been
satisfactorily identified.

Tritonidea carinifera (Kuster).
Tritonidea carinifera (Kiister), Smith, ‘Proc. Malac.
Soc.,’ vol. v, p. 371 (1903).
Tritonidea natalensis Smith: Sowerby, ‘ Marine In-
vest., vol. 11, p.. 229 (19038).
Kiister’s name carinifera has many years’ priority over
natalensis of Smith.

* Kuthria pura Martens.

Euthria pura Martens, ‘Tiefsee-Exped.,’ p. 25, Pl. H,
fie. 14.

Hab.—Agulhas Current, 273 fath.

JO

Kngina mendicaria (Lamarck).

This species, placed by Mr. Sowerby (‘Marine Shells 8. A.,’
p. 22) in the genus Columbella, should be removed to
Engina. It has been known for many years that its radula
is Buccinoid and very different from that of the Colum-

bellide.

* Phos roseatus Hinds.
Phos roseatus Hinds: Sowerby, ‘Thes. Conch.,’ vol. 111,
P]. CCXXI, figs. 1-3.
Hab.—Durban (Burnup), Philippines, Moluccas, ete.

Sylvanocochlis ancilla (Hanley).
Pseudolivaancilla Hanley: Sowerby, ‘Marine Invest.,’
vol. 11, p. 228.
Sylvanocochlis ancilla: Melvill, ‘J. of Conch.,’ vol. x,
p. 325, fig.; Smith, ‘J. of Malac.,’ vol. xi, p. 23.
Hab.—Off South Africa, 40 fathoms (Sowerby); Port
Alfred (Smith).

36 EDGAR A. SMITH.

* Nassa circumtexta Martens.

Nassa (Amycla) circumtexta Martens, ‘ Tiefsee-
Exped, “ps. 21, 4) fies 18.

Nassa trifasciata A. Adams: Sowerby, ‘Marine Invest.,’
volvan, 7p. 29s El RV", fist 92:

Hab.—Francis Bay, Algoa Bay, Agulhas Bank, Simon’s
Bay, 38—6-4 fathoms.

This species is the N. trifasciata of Adams. The types of
that species in the Cuming collection are certainly the same
as the South African shell, and I doubt very much if they
came from Spain. A species very like it—semistriata
Brocchi—does occur there, but it is not quite the same. The
name trifasciata was preoccupied by Gmelin for a species
belonging to the genus Nassa, which he described as a

Bueccinum.

Nassa analogica Sowerby.

Nassa analogica Sowerby, ‘Marine Invest.,’ vol. ii, p.
219 PIS ny, tieses:

Hab.—South Africa, 40 fathoms.

I can hardly believe that this species is distinct from the
preceding (circumtexta), notwithstanding the differences
pointed out by Mr. Sowerby.

Nassa desmoulioides Sowerby.

Nassa desmoulioides Sowerby: ‘Marine Invest.,’
vols, 3p. 29> Plies tios 1
Hab.—Off Natal, 100 fathoms.

* Nassa pecilosticta Smith.

Nassa pecilosticta: Smith, ‘J. of Malac., vol. xi, p. 33,
Piles lo:
Hab.—Port Aifred, Cape Colony.

“I

SOUTH AFRICAN MARINE MOLLUSCA. 3

%

Bullia trifasciata Smith.

Bullia trifasciata Smith, ‘J. of Malac.,’ vol. xi, p. 34,
Pai ie. 27,
Hab.—Port Alfred.

~pulliayancilletormis nm. sp. Pl Vilyness,

Testa parva, oblonga, supra acuminata, alba, infra suturam
fusco zonata, levis, lmeis incrementi tenuibus arcuatis
obliquis sculpta, obsolete spiraliter striata; anfractus 5—6,
celeriter accrescentes, fere plami, supra suturam callo tenu
induti, ultimus magnus, elongatus, leviter convexus, antice
oblique descendens; apertura supra acuminata, infra trun-
cata, lata; columella obliqua, paulo arcuata, infra oblique
truneata, callo tenui induta; labrum tenue, inferne haud
profunde smuatum. Longit. 19 mm., diam. 7°5; apertura
8°) mm. longa, 4 lata.

Hab.—Port Shepstone, about 70 miles south of Durban,
Natal (McBean).

This species is quite distinct from any of the other South
Africa forms, being remarkable on account of its pecuhar
shape, recalling somewhat that of certain species of Ancilla.
The thin callosity which spreads over the columella winds
up the spire and covers the greater part of the surface of the
whorls.

Columbella filmer Sowerby.
Columbella filmere, Sowerby: Smith, ‘Proc. Malac.
Soc.,’ vol. v, p. 374.
This species was wrongly quoted in the ‘Marine Shells of
South Africa’ as C. sagena Reeve, a Japanese form, not
yet known as South African.

* Columbella versicolor Sowerby.
Columbella versicolor Sowerby, ‘Thesaurus Conch.,’
vol. i, Pl. XXXVII, figs. 41-46; Reeve, ‘Conch. Icon.,’

vol. xi, figs. 51 a, b.

38 EDGAR A. SMITH.

Hab.—Natal (Burnup) ; Annaa, Philippines, etc.
This species is the C. scripta of Lamarck, a name pre-
occupied by Linné.

Columbella mitriformis 4. Adams.

This species should be removed from the South African
list, as the shells so named (Proc. Malac. Soc.,’ vol. v, p. 379)
prove to be C. leptalea Smith.

Murex axicornis Lamarck, var.?

Murex axicornis Lamarck: Reeve, ‘Conch. Icon.,’
vol. im, Pl. X, tig..37, Pl. XV, fig. 3%; Sowerby, “Marime
Invest.,’ vol. 1, p. 227.

Hab.—Off mouth of Umhloti River, Natal, 110 fathoms.

Murex carduus Broderip.
Trophon carduus Broderip: Sowerby, ‘ Marine Invest.’
vol. i; p. 224.
Hab.—Off Natal, 250 fathoms; Peru (Cuming).
The generic position of this shell appears to be rather
uncertain. It might, perhaps, be placed in Coralliophila.

*Ocinebra natalensis n.sp. Pl. VII, fig. 9.

Testa parva, ovato-fusiformis, albida, rimata, sex-varicosa,
ris spiralibus tuberculatis (in anfr. superioribus 2, in ultimo
circiter 8) instructa, inter liras lineis incrementi lamellosis
ornata ; anfractus 6, convexi, varicibus squamosis, gradati ;
apertura ovata, intus lineis mgris 5-6 picta; labrum ad
marginem acutum, extra varice ultimo expanso incrassatum ;
columella arcuata, callo libero induta; canalis angustatus,
brevis, obliquus. Longit. 14 mm., dian. 8°5. Apertura intus 4
longa, 2°75 lata.

Hab.—Umkomaas and Port Shepstone (Burnup).

This species is remarkable on account of the peculiarity of
the sculpture. The spiral ridges between the varices are

SOUTH AFRICAN. MARINE MOLLUSCA. 39

ornamented in a very unusual manner with close-set tubercles,
the varices upon the ridges are squamosely produced and
the lines of growth between the ridges are also somewhat
squamous. ‘he body-whorl exhibits six principal spirals and
two minor ones anteriorly, which are close together, also a
basal ridge which is scaled and forms the umbilical rimation.
The dark lines within the aperture correspond to the external
ridges. The edge of the labrum is somewhat frilled by the
termination of the ridges.

Urosalpinx contracta (Reeve).

Urosalpinx contracta (Reeve): Smith, ‘Proc. Malac.
Soc.,’ vol. v, p. 376.

This is merely a variety of the Ricinula heptagonalis,
Reeve, already quoted by Sowerby (‘Marine Shells S. A.,’
p. 15) as Sistrum heptagonale. According to Professor
Gwatkin, who has examined the radula, its position in the
genus Urosalpinx appears to be justified. The specific
name heptagonalis, having a few months’ priority, should
be retained.

Latiaxis tortilis H. and A. Adams.

Latiaxis tortilis A. Adams: Sowerby, ‘ Marine Invest.,’
vol. u, p. 228.

Hab.—South Africa, 166 fathoms.

The shell figured by the late G. B. Sowerby (‘Thesaurus
Conch.,’ vol. v, Pl. CCCCXXLYV, fig. 1) is not the actual type
specimen in the Cuming collection. It was described by H.
and A. Adams (not A. Adams only) and is probably merely
a white variety of L.idoleum Jonas as suggested by Gray
and Tryon. The name idoleum being a substantive cannot
be altered to idolea, as given by Tryon and Sowerby.

*Purpura castanea Kiister.
Purpura castanea Kiister: Smith, ‘J. of Malac.,’ vol. x1,
p. 95.

Hab.—Port Alfred.

40) EDGAR A. SMITH.

Purpura texturata Smith.

Purpura texturata Smith, ‘J. of Malac., vol. xi, p. 32,
Pra feo,
Hab.—Port Alfred.

*Pinaxia coronata A. Adams.

Pinaxia coronata A. Adams, ‘Proc. Zool. Soc.,’ 1853,
p. 185; H. and A. Adams, ‘Genera Moll., Pl. XIV, fig. 1.

Hab.—Umkomaas (Burnup): Ceylon, Philippines, Sand-
wich Islands, ete.

Sistrum cancellatum (Quoy and Gaimard) : Smith, ‘ Proc.

Malacs Soc vol. aa) pad.

This species is quoted in the ‘Marine Shell of 8. Africa,’
Appendix, p.6,as Sistrum elongatum Blainville. Reeve’s
figure, there quoted, does not represent Blainville’s species,
but another form described by that author under the name of
Purpura fenestrata, which is a synonym of the present
species (cancellata).

Sistrum concatenatum (Lamarck).

Sistrum concatenatum Lamarck: Sowerby, ‘Marin
Shells 8. A.,’? Appendix, p. 6.

It seems to be uncertain whether this species really is
South African. The specimens quoted as squamosum Pease,
var. (‘ Proc. Malac. Soc., vol. v, p. 377) may have been mis-
taken for it.

Argobuccinum (Fusitriton) murrayi (Smith).
Tritonium (Cryotritonium) murrayi Smith: ‘ Mar-
tens, ‘ Tiefsee-Exped.,’ p. 38, Pl. ITI, fig. 16.
Hab .—Off Cape, 97-270 fath.

SOUTH AFRICAN MARINE MOLLUSCA. 4]

* Bursa (Bufonaria) lampas (Lamarck).
Triton lampas Lamarck: Reeve, ‘Conch. Icon.,’ vol. u,
PP EX ne. 30a, Pi X, fig. 30:
Hab.—Bluff, Durban (G. W. Westcott).

Septa leucostoma (Lamarck) var.

Ranella leucostoma Lamarck, var. pccilostoma:
Martens, ‘ Tiefsee-Exped.,’ p. 56.

Martens has given this varietal name to the South African
specimens, which differ from Australian examples in having
“black markings on the ip” (Sowerby, ‘Marine Shell 8. A.,’
p: 9).

Colubraria crebrilirata (Sowerby).
Epidromus crebriliratus Sowerby, ‘Marine Invest.,’
WOleeiy 220, 1) IV fro Ay
Hab.—Off Port Alfred, 100 fathoms.

* Cassis pirum Lamarck.
Cassis pyrum Lamarck: Kiener, ‘Cog. Viv.,’ p. 39,
Pl. XIII, fig. 25, Pl. XV, fig. 30; Martens, ‘Tiefsee-Exped.,’
pp. 54, 56: note 12, var. intercedens.

Hab.—S. Africa.

Oniscia macandrewi1 Sowerby.
Oniscia macandrewi Sowerby, ‘Proc. Zool. Soc.,’ 1888,
p. 567, Pl. XXVIII, figs. 1, 2; ‘Marine Invest.,’ vol. n, p. 229.
Hab.—Off Natal, 27-250 fathoms.

Dolium fimbriatum Sowerby.

Var. natalensis n. var. Pl. VII, fig. 10.

Testa parva, ovato-globosa, rimata, solidiuscula, pallide
grisea vel dilute lilacea, costis fuscescentibus ornata, perios-
traco tenui deciduo flavescente induta; spira mediocriter
elata ; anfractus 6, superiores tres (protoconcha) fusco-cornel,

42 EDGAR A. SMITH.

leaves, politi, convexi, ceeteri spiraliter costati, costis angustis
(in anfractu penultimo 3, in ultimo 13-14); apertura irre-
gulariter elongato-pyriformis, intus fuscescens, labrum versus
albida; labrum intusincrassatum, album, denticulatum, ad
marginem tenue, fimbriatum, pone varice obliquo conspicuo
instructum ; columella callo tenui induta, antice plicis obliquis
paucis munita, subtuberculata ; canalis anticus brevis, haud
profundus, leviter recurvus. Longit. 41 mm., diameter 31.
Apertura cum labro 54, diam. intus 15.

Hab.—Durban, Bluff, Natal.

This variety was catalogued in Sowerby’s ‘ Marine shells
of South Africa,’ Appendix, p. 11, as D. fimbriatum without
any observation with regard to its differmg from the typical
form. It is much smaller, has a distinct varix outside the
labrum, and shows scarcely any trace of the spotting upon
the spiral ridges which is so characteristic of the normal

form.

Pedicularia sicula Swainson.

Pedicularia sicula Swainson: Tryon, ‘Man. Conch.,’
vol. vii, p. 241, Pl. I, figs. 1-8; Sowerby, ‘ Marine Invest.,’
vol. u1, p. 280.

Hab.—Off Cape St. Blaize, 116 fathoms.

Cyprea barclayi Reeve.

Cyprea barclayi Reeve: Sowerby, ‘Thesaurus Conch.,’
vol. iv, Pl. CCCV, figs. 91, 92; ‘Marine Invest.,’ vol. ii,
p. 230.

Hab.—Off Cape St. Blaize, Cape Colony, 55 fathoms.

Cyprea fultoni Sowerby.
Cyprea fultoni Sowerby, ‘ Marine Invest.,’ vol. ii, p. 218,
) Sy:
PLOIVS fie:
Hab.—S. Africa.

SOUTH AFRICAN MARINE MOLLUSCA. 43

*Trifora convexa Smith.
Trifora convexa Smith, ‘J. of Malac., vol. xi, p. 37,
Pl: EU, fie. 9.
Hab.—Port Alfred.

* Trifora fuscescens Smith.
Trifora fuscescens Smith, ‘J. of Malac.,’ vol. xi, p. 3
Pip TE ties, 6.
Hab.—Port Alfred.

~
~
“

*Trifora fuscomaculata Smith.

Trifora fuscomaculata Smith, ‘J. of Malac., vol. x,
Pestle os 7
Hab.—Port Alfred.

JPritoravcerea nm: sp. Fl. Vil, figs 11 ila:

Testa subulata, flavescens, nitida; anfractus 14 (7), convexi,
costis spiralibus tuberculatis quatuor cincti, duobus medianis
ceteris majoribus, inter costas oblique costulati, ultimus costis
sex instructus, duobus inferioribus vix tuberculatis ; columella
supra arcuata, callo albo crassiusculo induta; canalis brevis,
obliquus, recuryus, haud clausus; labrum subpatulum, extre-
mitatibus costarum leviter dentatum.

Longit. 10°5 mm., diam. 2°25. Apertura 1°25 longa.

Hab.—Port Shepstone (Burnup).

Of a uniform yellow wax colour, ornamented with four rows
of granules on each whorl, the lowest row being the smallest,
and the two central series rather more prominent than the
uppermost row.

*Trifora shepstonensis n.sp. Pl. VII, figs. 12, 12a.

Testa elongata, subulata, fuscescens; anfractus circiter 15,
plani, tricingulati, cingulis plus minus moniliformibus, mediani
ceteris minori, in sulcis liris longitudinalibus decussati,

AA, EDGAR A. SMITH.

ultimus liris 5 ornatus ; apertura parva, albida ; labrum tenue,
interdum productum, columellam antice attingens; columella
supra arcuata, callo crassiusculo reflexo induta ; canalis brevis,
obliquus, recurvus. Longit. 10 mm., diam, 2°5.

Hab.—Port Shepstone (Burnup).

The spiral ridges are crossed by oblique shallow sulci so as
to produce a somewhat beaded appearance.

Cerithium pingue (A. Adams).

Colina pinguis A. Adams, ‘ Proc. Zool. Soc., 1854, p. 86.

Cerithium pingue Sowerby, ‘Thesaurus Conch., vol. 11,
p: 8/7, Pl, CUXXXIV,, fio. 217 (1855).

Cerithium contractum Sowerby, |. c. p. 877, PI.
CLXXXIV, fig. 218.

Cerithium teniatum Sowerby, ‘Conch. Icon.,’ vol. xv,
fig. 119 (1865).

C. contractum and C. teniatum, regarded as species in
the ‘Marine Shells of 8. A., p. 35, are not, in my opinion,
worthy of even varietal rank. Both names were preoccupied,
the former by Bellardi for a fossil species, and the latter by
Quoy and Gaimard. ‘lhe name C. crumena was proposed
by Bayle in 1880 for Sowerby’s C. contractum.

The species has a wide range. Adams quoted it from the
Phihppines, and there are specimens from Muscat and the
Persian Gulf in the British Museum.

* Cerithiopsis trilineata (Philippi).

Cerithium trilineatum Philippi, ‘Enum. Moll. Sicil.,
Volo, p. 195, PF]. x tie: 13.

Cerithiopsis trilineata Smith, ‘J. of Malac.,
p. 24.

Hab.—Port Altred.

The specimens quoted under Cerithiopsis purpurea
Angas (‘ Marine Shells S. A.,’ p. 27) appear to belong to this
species. I am doubtful at present whether any constant dis-

SV) xed

tinctions between the two species can be pointed out.

SOUTH AFRICAN MARINE MOLLUSCA. 45

*Cerithiopsis insignis n. sp. Pl. VIL, fig. 13.

Testa minima, elongato-pupoidea, corneo-albida, rufo
unibalteata ; anfractus circiter 8, convexiusculh, lente accres-
centes, seriebus tribus tuberculorum ornati; series suprema
saturate rubra, mediana ceteris minor; anfr. ultimus seriebus
quatuor cinctus; apertura parva; columella brevis, callo
crasso induta. Longit. 3°25 mm., diam. 1.

Hab.—Port Shepstone (Burnup).

A very small species, but well characterised by its striking
colouring. The tubercles of the three series are joined by
longitudinal connections, so that the surface is, in fact, cancel-
lated. The median tubercles are much smaller than those of
the upper and lower rows. Quite distinct from C. pul-
chella C. B. Adams, from Jamaica, which is somewhat
similarly coloured.

*Cerithiopsis chapmaniana n.sp. Pl. VII, fie. 14.
I y Hg

Testa elongata, turrita, alba, subpellucida, nitens; an-
fractus 10, leviter convexi, costis longitudinalibus leviter
obliquis numerosis tenuibus lrisque spiralibus tribus, supra
costas granosis, cancellati, sutura profunda sejuncti; anfr.
ultimus liris quatuor cinctus, infra concavus, levis, sed lineis
incrementi tenuissimis striatus; apertura parva; labrum
tenue; columella in medio leviter arcuata, antice obliqua ;
canalis brevissimus, recurvus. Longit. 8 mm., diam. 2°75.
Apertura 1°75 min. longa, 1°25 lata.

Hab.—Isezela (Miss Chapman).

A pure white shell with three principal rows of granules
upon each whorl and sometimes a much finer row in the
spaces between. ‘I'he longitudinal coste are about twenty-
two in number, slender, a little oblique and slightly arcuate.
The apical whorls are broken away. .

Cerithium tricarinatum Pease!, badly figured by

1 * Proce. Zool. Soc.,’ 1860, p. 433.

4.6 EDGAR A. SMITH.

Sowerby !, is sculptured very similarly, and should, I think,
be removed to the genus Cerithiopsis. It is, however, a
httle more slender, of a light brown colour, more openly
latticed, has a different columellar fold, and is distinctly
spirally striated between the three principal spiral rows of
granules.

Turritella bacillum Kiener.
Turritella bacillum Kiener: Martens,‘ Tiefsee-Hxped.,’
p. 44.
Its occurrence in South Africa confirmed.

Turritella declivis Adams and Reeve, var.
Turritella declivis Adams and Reeve, var.: Martens,
‘Tiefsee-Hxped.,’ p. 44, Pl. IV, fig. 10.
Hab.—Simon’s Bay, 38 fathoms.

Turritella punctulata Sowerby.

Turritella punctulata Sowerby: Martens, ‘Tiefsee-
Exped.,’ p. 43, Pl. IV, figs. 9, 9 a, 6.

Hab.—Francis Bay, Cape Agulhas, Algoa Bay, 44-55

fathoms.

* Littorina scabra (Lmn.).

Littorina scabra Innn.: Reeve, ‘Conch. Icon., vol. x,
figs. 21 a-c.

The followimg so-called species, viz. L. angulifera
Lamarck, intermedia Philippi, ahenea Reeve, and
newcombi Reeve, appear to pass one into the other and
also into L. scabra, and without a very deep study of the
group it appears hopeless to attempt to give any definite
opinion upon the specific value of any of them. ‘They have
already been united by Tryon as forms of L. scabra.
L. ahenea, newcombi and intermedia are quoted in the
‘Marine Shells of S. A.’ as distinct species. Specimens
have been examined which seem to be inseparable from the
L. scabra as determined by Reeve, Weinkauff, etc.

1 *Conch. Icon.,’ vol. xv, fig. 127. Copied by Tryon, ‘Man. Conch.,’
Wola, pl Kock att.

SOUTH AFRICAN MARINE MOLLUSCA. 47

* Rissoa conspecta Smith.

Rissoa conspecta Smith, ‘J. of Malac.,’ vol. xi, p. 30,
le Le fig. 26;
Hab.—Port Alfred.

* Rissoa perspecta Smith.

U

~
~

Rissoa perspecta Smith, ‘J. of Malac., vol. xi, p. 3:
Plott, fig. 25.
Hab.—Port Alfred.

* Rissoina alfredi Smith.

Rissoina alfredi Smith, ‘J. of Malac.,’ vol. xi, p. 35,
Pl. II, fig. 24.
Hab.—Port Alfred.

* Rissoina durbanensis n.sp. Pl. VII, fig. 15.

‘Testa mediocriter elongata, albida ; anfractus 7—8, vix con-
vexi, costis circiter 16 leviter obliquis et lis transversis
supra costas tuberculatis (in anfractu penultimo quatuor, m
ultimo septem) instructi, sutura profunda sejuncti, ultimus
supra lram anticam, ceteris crassiorem, quasi sulcatus ;
apertura obliqua, irregulariter ovalis, antice haud profunde
sinuata; labrum varice crasso lato extra munitum; columella
in medio arcuata, callo albo, antice leviter incrassato, induta.

Longit. 45 mm.; diam. 15 mm. Apertura intus, | longa,
D lata.

Hab.—Durban (Burnup).

A small, prettily granuled-cancellated species, with a deep
suture, a strongly varixed Jabrum, and a conspicuous sulcus
around the base of the body-whorl, above the most anterior
of the transverse lire, which is rather thicker than those
above. It belongs to the same group as R. bicollaris and
R. fenestrata of Schwartz.

AS EDGAR A. SMITH.

*Rissoima shepstonensis n.sp. Pl. Vil, tie 10:

Testa elongata, alba, subpellucida, oblique costata, circa
basim anfr. ultimi transversim striata ; anfractus 9-10, superi-
ores duo rotundati, leaves, cater! mediocriter convex, costis
16-18 oblique arcuatis instructi,’ ultimus costis flexuosis,
infra medium transversim tenuiter liratus; apertura obliqua,
subovalis, antice late sinuata; labrum extra valde incrassa-
tum, ad marginem tenue; columella obliqua, callo, supra et
infra labro juncto, induta.

Longit. 775 mm., diam. 2°5 mm. Apertura 2 mm. longa,
1-25 mm. lata. :

Hab.—Port Shepstone (Burnup).

The convexity of the whorls diminishes as the shell in-
creases, the upper ones having almost a turreted appearance.

Vanikoro cancellata (Lamarck).

Vanikoro cancellata Lamarck: ‘Tryon, ‘Man. Conch.,’
vol. vii, p. 67, Pl. XXIX, figs. 60, 61; Sowerby, ‘ Marine
Invest.,’ vol. u, p. 229.

Hab.—Off Natal, 43 fathoms.

* Natica areolata Reécluz.

Natica areolata Récluz, ‘Proc. Zool. Soc.” 1843, p. 206 ;
Philippy, “Conch. Cabs’-p.67, Plo XL fig. 2; Tryon, “Man:
Conch.,’ vol. vin, PIO Vil fen 23:

Hab.—Scottsburg, Natal.

* Natica decipiens Smith.
Natica decipiens Smith, ‘J. of Malac., vol. xi, p. 34,
Pl iio ws:
Hab.—Port Alfred.

Natica forata Reeve.

This species, wrongly quoted as of Récluz both in the
‘Thesaurus Conch.’ and the ‘ Marine Shells of South Africa,’
according to Martens, is the same as N. pygmea Philippi.
At present I am unable to concur in that opinion.

SOUTH AFRICAN MARINE MOLLUSCA. 4.9

*Natica napus Smith.
Natica napus Smith, ‘J. of Malac.,’ vol. xi, p. 34, Pl. IT,
fig. 22.
Hab.—Port Alfred.

Natica sagraiana dOrbigny, var.
Natica sagraiana d’Orbigny: Reeve, ‘ Conch. Icon.,’
vol. ix, fig. lll a, b; Sowerby, ‘ Marine Invest.,’ vol. 11,
29
p. 229.
Hab.—NSaldanha Bay, Cape Coast, 28 fathoms.

*Scala bullata Sowerby.

Scalaria bullata Sowerby, ‘Thesaurus Conch.,’ vol. 1,
py oss, PIXEL, fic. 87; “Conch. Icon. vol. xix, fig. 8.
Hab.—Durban (Burnup) ; Philippine Islands.

Scala tenebrosa Sowerby.
Scala tenebrosa Sowerby, ‘ Marine Invest.,’ vol. 11, p. 224,
ELV, fic. 6.
Hab.—S. Africa.

*Scala durbanensis n.sp. Pl. VII, fig. 17.

Testa parva, elongata, alba, solidiuscula; anfractus 10,
supremi tres leves, convexi, ceeteri normales convexi, costis
obliquis 14 valde reflexis, supra spiram peroblique continuis,
instructa, costis in anfr. ultimo versus aperturam sensim
latioribus, inter costas minute spiraliter striati; apertura
oblique ovata.

Longit. 10 mm.; diam. 4 mm. Apertura intus 2 longa,
1-5 lata.

Hab.—Durban (Burnup).

Remarkable on account of the very reflexed and rather
numerous coste, a few of which upon the body-whorl, near
the aperture, are much broader than the rest. ‘The spiral,
close-set striz between the riblets are only visible under a
strong lens.

VoL. 1, PART |] 4.

50 EDGAR A. SMITH.

*Scala eborea n. sp..-Pl. VIEL fig. I.

Testa parva, elongata, acuminata, alba, nitida; anfractus 10,
supremi tres convexi, leves, ceteri convexi, sutura obliqua
sejuncti, costis tenuibus obliquis decem reflexis, costis in
anfractu ultimo versus aperturam sensim crassioribus ; aper-
tura rotunde ovata.

Longit. 9 mm.; diam. 3°75 mm. Apertura intus 2 longa,
15 lata.

Hab.—Port Shepstone and Durban (Burnup).

More acuminate than 8S. durbanensis, with fewer cost,
and without any transverse striz. The riblets are very fine,
and so much rolled back that they appear to form thread-like
hollow lire.

* Acrilla gracilis H. Adams.

Acrilla gracilis H. Adams: Smith, ‘J. of Malac.,’ vol. x1,
p. 24.

Scalaria minor Sowerby: Reeve’s ‘Conch. Icon.,’ vol.
p cibxGy Kone FAO)

Hab.—Port Alfred, Cape Colony.

Eulima dilecta Smith.

Eulima dilecta Smith: ‘Proc. Malac. Soc.,’ vol. v,
p. 386.

Under this species it should have been stated that the
Eh. solida of the ‘ Marine Shells of South Africa’ is the same
form, and distinct from the true solida, the locality of which
is said to be Sandwich Islands.

* Kulima distincta Smith.

Kulima distincta Smith: ‘J. of Malac.,’ vol. xi, p. 35,
Pin fies 1.
Hab.—Port Alfred.

SOUTH AFRICAN MARINE MOLLUSGA. 5]

* Niso interrupta Sowerby.
Niso interrupta Sowerby: Reeve’s ‘Conch. Icon.,’ vol.
xv, fig. 8 a, b; Smith, ‘J. of Malac.,’ vol. xi, p. 24.
Hab.—Port Alfred.

Turbonilla hofmani Angas.

Turbonilla hofmani Angas (1877): Smith, ‘J. of
Malac.,’ vol. xi, p. 24.

T. candida A. Adams: Sowerby, ‘ Marine Shells of South
Africa,’ p. 26 ; Smith, ‘ Proc. Malac. Soc.,’ vol. v, p. 386.

The shells formerly considered to belong to T. candida
are certainly distinct from that species. ‘They appear to be
inseparable from T.hofmani. The name lactea of Krauss
(1848), afterwards changed by Clessin (1900) to kraussi,
was preoccupied by Linnzeus for a northern species.

* Turbonilla decora Smith.
Turbonilla decora Smith: ‘J. of Malac.,’ vol xi, p. 36,
PRE fie: 5:
Hab.—Port Alfred.

*Turbonilla gemmula Smith.
Turbonilla gemmula Smith: ‘J. of Malac.,’ vol. xi,
paco Je) ThE ns. 4:
Hab.—Port Alfred.

* Mormula rissoina A. Adams. PI. VIII, fig. 2.
Mormula rissoina A. Adams: ‘Journ. Linn. Soc.,’ 1863,

vol. vil, p. 1; Smith, ‘J. of Malac.,’ vol. x1, p. 24.
Hab.—Port Alfred.

* Hlusa natalensis n.sp. Pl. VIII, fig. 3.
Testa elongata, subulata, ad apicem obtusa, alba, polita ;
anfractus 8, superiores duo convexi, leves, ceeteri fere plani,
superne tenuiter oblique striati, sutura fere canaliculata

a

|

52, EDGAR A. SMITH.

sejuncti, ultimus infra medium transversim striatus, antice
leviter ascendens; apertura piriformis, parva; peristoma
continuum, margine dextro incrassato, extra subvaricoso,
columellari supra anfractum posito, tenui, intus in medio
oblique uniplicato.

Longit. 7°25 mm., diam. 2°5. Apertura 2 mm. longa,
1-25 lata.

Hab.—Port Shepstone (Burnup).

A glossy white shell with oblique striz, which are stronger
at the upper part of the whorls than below, and with some
transverse striae upon the lower half of the body-whorl, those
quite at the anterior end being closer together than those
above. The inner lip might be described as a callus upon
the whorl, joining the basal margin of the labrum to its point
of insertion above. ‘To the naked eye the shell looks almost
smooth and glossy. E.aclis, A. Adams, described origin-
ally as a Pyramidella, is somewhat lke the present species
in form, but is distinctly costate.

* Kulimella minor Smith.

Eulimella minor Smith: ‘J. of Malac.,

Pl. III, fig. 3.
Hab.—Port Alfred.

vol. vi, p. 36,

* Kulimella nivea Smith.

Kulimella nivea Smith: ‘J. of Malac.,’

PS sti. 2:
Hab.—Port Alfred.

VOls Xie ps0;

Astralium andersoni Smith.

Astralium andersoni Smith: Sowerby, ‘ Marine Invest.,’
vol. nu, p- 230, Pl. V, fig. 5.

Hab.—Off south coast of Cape Colony, 36 fath.

A good figure of an adult shell is given by Mr. Sowerby.

Or
—e
—~

SOUTH AFRICAN MARINE MOLLUSCA.

Astralium gilchristi Sowerby.
Astralium (Cyclocantha) gilchristi Sowerby:
‘Marine Invest.,’ vol. 1, p. 221, Pl. V, fig: 6.
Hab.—Off Natal, 90 and 92 fathoms.

* Astralium henicus (Watson).

Turbo (Calcar) henicus /atson : ‘Challenger, Gastero-
poda, p. 130, Pl. VI, figs. 11 a-c; Martens, ‘ Tiefsee-Exped.,’
p. 46.

Hab.—Agulhas Bank, 55 fathoms; Sumatra and Fiji.

* Leptothyra armillata (A. Addams, Sowerby).
Leptothyra armillata A. Adams: Smith, ‘J. of Malac.,’
vol. xi, p. 24.
Hab.—Port Alfred.

* Kthalia africana Smith.
Kthalia africana Smith, ‘J. of Malac.,’ vol. xi, p. 38,
Pith sies2 10) lt.
Hab.—Port Alfred.

* Liotia bicarinata Martens.
Liotia bicarinata Martens, ‘ Tiefsee-Exped.,’ p. 46, Pl. V,
fig. 4.
Hab.—Near Agulhas Bank, 271 fathoms.

*Cyclostrema (Tubiola) semisculptum Martens.
Cyclostrema (Tubiola) semisculptum Martens,
‘Tiefsee-Exped.,’ p. 49, Pl. V, fig. 6.
Hab.—Outside Agulhas Bank, 1490 fathoms.

*Cynisca forticostata Smith.
Cynisca forticostata Smith, ‘J. of Malac.,’ vol. xi, p. 38,
Petites. i225 13:
Hab.—Port Alfred.

54 EDGAR A. SMITH.

Calliostoma perfragile Sowerby.
Calliostoma perfragile Sowerby, ‘Marine Invest.,’
VOlmiy Pp. 222s be. V jt. 3:
Hab.—Off Cape coast, 154 and 166 fathoms.

*Calliostoma bisculptum n.sp. Pl. VIII, fig. 4.

Testa acute conica, angulata, subrimata, griseo-albida,
fusco strigata, strigis plus minus duplicatis; anfractus 7,
superiores 1-2 convexi, leves, ceeteri fere plani liris tenuibus
numerosis spiralibus ornati, lneis obliquis incrementi sculpti,
ultimus ad peripheriam angulatus, infra liris concentricis
circiter 10, quam superioribus fortioribus cinctus, inter lras
transversim striatus ; apertura subquadrata; labrum acutum,
intus leviter incrassatum; columella leviter arcuata, alba,
reflexa, callo tenui labro juncta.

Diam. maj. 8 mm., min. 7, alt. 10.

Hab.—Durban (Burnup).

The spiralsabove the angle, about fifteen in number, aremuch
finer than those upon the base. The slight umbilical perfora-
tion is perhaps sometimes covered by the reflexed columella.

The colour may be variable, but im the unique example
examined the brown stripes are divided down the middle
by a whitish line. They are shghtly undulating and the
basal liree are spotted with the same colour, which is blackish
brown. The whorls are not quite flat, as the rounded keel,
which passes above the suture, causes a faint swelling at the
lower part.

Calliostoma granoliratum Sowerby.
Calliostoma (Lischkeia) granoliratum Sowerby,
‘Marine Invest.,’ vol. 1, p. 222, Pl: V, fig. 7.
Hab.—Off Cape Point, 45 fathoms.

Calliostoma (Astele) iridescens Sowerby.
Calliostoma (Astele) iridescens Sowerby, ‘ Marine
Invest.,’ vol. 1, p. 223, Pl. V, fig. 4.
Hab.—Off Cape Natal, 55 fath.

or
c

SOUTH AFRICAN MARINE MOLLUSCA.

Oxystele impervia Menke.

This species should stand under the name variegata
Anton, if both forms really belong to one and the same
species. Anton’s species was described in 1839 and Menke’s
about four years later. Krauss and Martens have united
them, but, on the other hand, Philippi decided to keep them
separate.

*Huchelus natalensis n.sp. Pl. VII, fig. 5.

Testa minima, rotunde turbinata, sordide albida, anguste
perforata, spiraliter carinata et inter carinas fortiter oblique
lirata vel lamellata; anfractus 4-43, supremi 14 rotundati,
fere leves, ad apicem involuti, penultimus carinis duobus
cinctus, ultimus carimis sex acutis prominentibus instructus ;
superficies inter carinas et liris obliquis margaritacea ; apertura
rotundata, intus margaritacea; labrum carinis extus dentatum ;
columella arcuata, callo reflexo induta.

Longi. 5 mm., diam. maj. 3; apertura 1°56 mm. longa, 1'3 lata.

Hab.—Durban (Burnup).

A beautiful little species allied to E. foveolatus A Adams,
from Lord Hood Island. It is, however, much sinaller than
that species, is more delicately sculptured, the spiral keels
being regular and simple, whereas in foveolatus they are
subspinose where the oblique lamella join them. Besides the
six carine mentioned above, a seventh is noticeable upon the
body-whorl close to the suture, and this may be traced upon
the penultimate volution also.

Solariella undata (Sowerby).

Minolia undata Sowerby: Martens, ‘'Tiefsee-Exped.,’
paced Vi. d10%- 5.
Hab.—Agulhas Bank and off the Cape, 84 and 173 fathoms.

Solariella congener (Sowerby).
Minolia (Nacheroplax) congener Sowerby, ‘Marine
Invest.,’ vol. 1, p. 223, Pl. V, fig. 2.

D6 EDGAR A. SMITH.

Hab.—Off 8. coast of Cape Colony, 37 fathoms.
Macheroplax is misprinted Nacheroplax in the
‘Marine Invest.’

* Solariella infundibulum (Watson).

Solariella infundibulum Watson: Martens, ‘ Tiefsee-
Hixped.,’ p. 48, Pl. IV, fig. 22.
Hab.—Outside Agulhas Bank, 1719 fathoms.

Solariella levissima Martens.

Solariella levissima Martens: ‘ Tiefsee-Exped.,’ p. 49,
Pl. -V, fig. 2; Smith, ‘Proc. Malac. Soc.,’ vol. vy, p. 390, as
Minolia; Sowerby, ‘ Marine Invest.,’ vol. i, p. 231, Pl. V,
fig. 2, as Minolia.

Solariella persculpta Sowerby.
Solariella persculpta Sowerby, ‘Marine Invest.,’ vol. u1,
Pa223. PLN aio. Ss.
Hab.—Off Cape Natal, 440 fathoms.

*Glyphis fuscocrenulata n.sp. Pl. VIII, fig. 6.

Testa parva ?, ovata, antice leviter angustata, mediocriter
elata, fusca vel purpureo-fusca, tenuiter cancellata, costis
radiantibus numerosis inequalibus lrisque concentricis cir-
citer 20, supra costas squamatis, instructa ; foramen parvum,
antemedianum, ad longitudinis 4 situm; superficies interna
callo tenui albo nitente induta, tenuiter crenulata vel denticu-
lata, et inter denticulos fusco punctata.

Longit. 16 mm., diam. 11°25 alt. 6.

Hab.—Port Shepstone and Umkomaas, Natal.

The radiating coste in this species are fine, and, as in
many other species, of different thicknesses. A few upon the
hinder half of the shell are more conspicuous than the rest.
The concentric lire form thickened scales upon the costz,
and, when a little worn, have a bead-like appearance. The
foramen, situated at about one third of the length from the

SOUTH AFRICAN MARINE MOLLUSCA. 57

front margin, is roundly ovate, small, and thickened within
with a white collar, which is truncate behind. The interior
of the shell has a thin deposit of white, glossy callus, through
which the external costae are visible, producing a radiate-
lineated appearance. The dotting upon the margin is formed
by the terminations of the cost, the dots being in minute
depressions between the denticulations. It is impossible to
say whether this species attains much larger dimensions
than those given above. However, the specimen described
appears to be fairly mature.

Puncturella noachina (Linn.).
Puncturella noachina Linn.: Sowerby, ‘Marine Invest.,’
VO). 11, p. 20.
Hab.—Off South Africa, 125 fathoms.
This northern species also occurs as far south as the
Straits of Magellan and Kerguelen Island.

*Puncturella fastigiata (A. Adams).

Cemoria fastigiata A. Adams; Sowerby’s ‘Thesaurus
Conch.,’ vol. i, p. 208, Pl. CCXLV, figs. 15, 16; Martens,
‘Tiefsee-Exped.,’ p. 50, as Puncturella.

€

Hab.—Simon’s Bay, 38 fathoms.
Hanleya sykesi (Sowerby).
Chiton (Hanleya) Sykesi Sowerby, ‘Marine Invest.,’

Vlei. a 2207 bly Vj hie. 13:
Hab.—South Africa. 166 and 210 fathoms.

SCAPHOPODA.

Dentalium plurifissuratum Sowerby.

Dentalium plurifissuratum Sowerby: Pilsbry, ‘ Man.
Conch.” vol. xvu, p. 82,°Pl: VIL figs: 87-89; Sowerby,
‘Marine Invest.,’ vol. 1, p. 231.

Hab.—Off Cape St. Blaize, 55-100 fathoms.

58 EDGAR A. SMITH.

Dentalium novemcostatum Lamarck.

Dentalium novemcostatum Lamarck: Pilsbry, ‘ Man.
Conch.,’ vol. xvii, p. 51, Pl. IX, figs. 44-48; Sowerby,
‘Marine Invest.,’ vol. un, p. 231.

Hab.—Off Cape St. Blaize, 85-90 fathoms.

A species occurring off the north Coast of France.

Dentalium africanum Sowerby.

Dentalium africanum Sowerby, ‘ Marine Invest.,’ vol. 11,
p. 224, Plo fie. 10.
Hab.—Off Natal, 25 fathoms.

Dentalium belcheri Sowerby.

Dentalium belcheri Sowerby: Pilsbry, ‘Man. Conch.,
vol, xvi, p. 60; Pl. XLY,. figs. 29, 30; Smith, ~Joum. yet
Malac.,’ vol. xi, p. 25; Sowerby, ‘Marine Invest.,’ vol. u,
p: 231:

Hab.—Port Alfred (Turton) ; off Cape 35 fathoms.

Dentalium exasperatum Sowerby.
Dentalium exasperatum Sowerby, ‘Marine Invest.,’
vol. 11; p. 225, PIV. figs t2.
Hab.—Off Natal, 27 fathoms.

Dentalium inflexum Sowerby.
Dentalium inflexum Sowerby, ‘Marine Invest.,’ vol. i,
pace24, Ply Ve fie le
Hab.—Mouth of Tugela River, Natal, 14 fathoms.

*Dentalium politum Linn.
Dentalium politum Linn.: Pilsbry, ‘Man. Conch.,’
vol. xvu, p. 128, Pl. XIX, figs. 18-21.
Hab.—Off Natal, 54 fathoms.

SOUTH AFRICAN MARINE MOLLUSCA. 59

PELECYPODA.

* Lima perfecta Smith.
Lima perfecta Smith, ‘Jour. of Malac., vol. xi, p. 43,
Ri wre 29.
Hab.—Port Alfred.

*Chlamys humilis Sowerby.
Chlamys humilis Sowerby, ‘Marine Invest.,’ vol. iv,
Deoy ele Wi hee 3:
Hab.—Off Cape Colony, 51 and 90 fathoms.

*Chlamys gilchristi Sowerby.
Chlamys gilchristi Sowerby, ‘Marine Invest.,’ vol. iv,
De eee tie. ©.
Hab.—False Bay, 230 fathoms.

*Chlamys fultoni Sowerby.
Chlamys fultoni Sowerby, ‘Marine Invest., vol. iv,
Peete. VL he.e 5,
Hab.—Off the Cape, 26 fathoms.

Chlamys tinctus (Reeve).

Pecten tinctus Reeve, ‘Conch. Icon.,’ vol. vin, fig. 106 ;
Smith, ‘Jour. of Malac., vol. xi, p. 25.

Pecten textilis Reeve, loc. cit., fig. 174.

Recten clulerens ieeve, loc. c1t., fig. Loo.

Pecten pusio Reeve (nec. Linn.), loc. cit., fig. 15
Sowerby, ‘Marine Shells 8. Africa,’ p. 66.

Hab.—Port Alfred, Algoa Bay, and Natal.

I am inclined to believe that this South African shell is
distinct from the European P. pusio (multistriatus Polt).
It does not exhibit any of the microscopical longitudinal
strie which are characteristic of that species.

lie

60 EDGAR A. SMITH.

* Chlamys natalensis n.sp. Pl. VIII, figs. 7, 8 a.

Testa parva, mediocriter compressa, ineequilateralis, fere
equivalvis, altior quam longa, grisea, radiis nigris interruptis
et maculis albis supra valvam sinistram picta; valva dextra
costis tenuibus circiter 40 fere levibus fusco notatis instructa,
in interstitis pulcherrime cancellatim sculpta, auricula antica
magna, pallida, costis circiter 9 squamatis instructa, postica
parva, costis 5-6 squamatis ornata; valva sinistra costis
squamis numerosis brevibus ; pagina interna alba nigro-fusco
interruptim hneata.

Longit. 20 mm., alt. 23°5, diam. 7:5.

Hab.—Durban (McBean).

This species is separable from Ch. tinctus on account of
its coarser and more equal ribs and the very different sculpture
in the intervening grooves. This consists of a fine cross-
hatching of crisscross striz, in some places looking lke the
remains of a periostracum. The partial absence of scales
upon the right valve is in some measure due to attrition.

* Pinna afra Sowerby.

Pinna afra Sowerby, ‘ Proc. Zool. Soc., 1835, p. 85.
Hab.—Cape of Good Hope.
Known only from the brief description.

* Pinna equilatera Martens.

Pinna equilatera Martens, ‘ Paetel’s Cat.’ (1890), 4th
ed., Abt. im, p. 208.

Hab.—“ Afric. mer.”

I have so far been unable to find any description of this
species.

* Pinna natalensis n.sp. Pl. VIII, fig. 9.

Pinna madida WSowerby (nec Reeve), ‘Marine Shells
S. A.,’ Appendix, p. 27.

Testa elongata, oblique triangularis, ad marginem ligamenti
recta, vel leviter incurva, ad marginem ventralem apicem

SOUTH AFRICAN MARINE MOLLUSCA. 6]

versus paulo incurva, deinde subexcurva, postice oblique
arcuatim truncata, versus umbones_ pallida, viridi-albida,
deinde fuscescens vel fumosa, costis radiantibus circiter 15,
haud squamulatis, instructa, lineis incrementi tenuissimis
sculpta; valve tenues, supra latus ventralem apicem versus
oblique corrugatee.

Longit. 155 mm., diam. obliqua max. 86, cross, 20.

Hab.—Durban. :

Although somewhat resembling P. madida Reeve, I think
this species may be separated. It is not quite of the same
form, the ventral margin beig less concave, the coloration
is different, the coste are more prominent, and the lines or
lamellee of growth are finer.

One of the two specimens at hand is more obliquely truncated
posteriorly. The ribs in the other specimen are in certain
hghts of an obscure golden tint.

* Modiola tenerrima Smith.
Modiola tenerrima Smith, ‘J. of Malac.,’ vol. xi, p. 42,
PES tees 26:
Hab.—Port Alfred.

* Hochstetteria velaini Smith.

Hochstetteria velaini Smith, ‘J. of Malac.,’ vol. xi,
p. 42, Pl. III, fig. 24.

Hab.—Port Alfred.

* Hochstetteria limoides Smith.

Hochstetteria limoides Smith, ‘J. of Malac.,’ vol. xi
p- 42, Pl. III, fig. 25.
Hab.—Port Alfred.

*Crenella striatissima Sowerby.
Crenella striatissima Sowerby, ‘ Marine Invest.,’ vol. iv,
Peoedels Vill stior oI
Hab.—Off Cape Colony, 56 and 100 fathoms.

62 EDGAR ‘A. SMITH.

* Arca (Scapharca) africana Sowerby.
Areca (Scapharca) africana Sowerby, ‘Marine Invest.,’
vol. iv, p. 4, Pl. VI, fig. 4.
Hab.—Off the mouth of the Tugela River, 46-55 fathoms.

* Limopsis pumilio Smith.
Limopsis pumilio Smith, ‘J. of Malac., vol. xi, p. 43,
PING, fies:27;/28:
Hab.—Port Alfred.

*Nucula sculpturata Sowerby.
Nucula sculpturata Sowerby, ‘Marine Invest.,’ vol. iv,
pe 7 ev es li:
Hab.—Off Cape Colony, 34 fathoms.

* Nucula irregularis Sowerby.
Nucula irregularis Sowerby, ‘Marine Invest.,’ vol. iv,
if .
Pesto Viliehio lo:
Hab.—Off Struis Point, Cape Colony, 48 fathoms.
®) p J?)

*Nuculana belcheri (Hinds).
Nuculana belcheri Hinds, Sowerby, ‘ Marine Invest.,’
VOI twp tye lena stig. 1.
Hab.—Off Cape Colony and Natal, 34-440 fathoms.

* Nuculana compta Sowerby.
Nuculana compta Sowerby, ‘Marine Invest.,’ vol. iv, p. 6,
Pl. Vietig. 10;
Hab.—Off Cape Natal, 440 fathoms.

*Nuculana gemmulata Sowerby.
Nuculana gemmulata Sowerby, ‘ Marine Invest.,’ vol. iv,
pao; elev Eig. 9.
Hab.—Off mouth of ‘ugela River, 37 fathoms; off
Umvoti River, 27 fathoms.

SOUTH AFRICAN MARINE MOLLUSCA. 63

*Nuculana lamellata Sowerby.
Nuculana lamellata Sowerby, ‘Marine Invest.,’ vol. iv,
poo, ple Vi fig. 8.
Hab.—Cape Natal, 54 and 85 fathoms.

* Cardita pulcherrima Sowerby.
Cardita pu!lcherrima Sowerby, ‘ Marine Invest.,’ vol. iv,
[Se Ue ells Nasir
Hab .-—Off Cape Natal, 54 fathoms.

* Cardita (?) minima Smith.
Cardita (?) minima Smith, ‘J. of Malac., vol. xi, p: 41,
PIER sto. 22.
Hab.—Port Alfred.

* Carditella laticosta Smith.
Carditella laticosta Smith, ‘J.of Malac., vol. x1, p. 41,
Prt tig? 23:
Hab.—Port Alfred.

* Crassatella abrupta Sowerby.
Crassatella abrupta Sowerby, ‘ Marine Invest.,’ vol. iv,
pal Pi Vie fie 15.
Hab.—Mouth of Umhloti River, 100 fathoms.

*Crassatella africana Sowerby.
Crassatella africana Sowerby, ‘Marine Invest.,’ vol. iv,
= ee,
Pence le Vil to. 13.
Hab.—Off Cape Infanta, 45 fathoms.

* Crassatella angulata Sowerby.
Crassatella angulata Sowerby, ‘ Marine Invest.,’ vol. iv,
Bees av tie. 16.
Hab.—Off mouth of Umhlangakulu River, 50 fathoms.

64 EDGAR A. SMITH.

*Crassatella gilchristi Sowerby.
Crassatella gilchristi Sowerby, ‘Marine Invest.,’
VOly av. Onl. V Dy tip, tae
Ha b.—Off Martha Point, 42 fathoms.

* Crassatella tenuis Sowerby.
Crassatella tenuis Sowerby, ‘Marine Invest.,’ vol. iv,
Delo, el. Vitec,
Hab.—Off Cape St. Blaize, 90 fathoms.

*Montacuta macandrewi (Fischer).

Kellia macandrewl, Fischer, ‘J. de Conch., 1867, p. 194,
Pl, 1X, fig. 1; Smith, “J. of Malac.” vol. x1,.p. 26 "Gis
Montacuta).

Hab.—Port Alfred: Spain, and Faro, Portugal.

*Tellimya similis Smith.
Tellimya similis Smith, ‘J. of Malac.,’ vol. xi, p. 41,
Py fie.
Hab.—Port Alfred.

*Lepton fortidentatus Smith.

Lepton fortidentatus Smith, ‘J. of Malac.,’ vol. x1,
ped, Plo 20;
Hab.—Port Alfred.

*Tridacna sp.

Hab.—Durban (McBean).
A young shell, probably belonging to T. elongata, was
found at the above locality.

*Cardium gilchristi Sowerby.
Cardium gilchristi Sowerby, ‘Marine Invest., vol. iv,
peel Vi atic, 1.
Hab .—Algoa Bay, 15 fathoms.

SOUTH AFRICAN MARINE MOLLUSCA. 65

Venus (Timoclea) arakana Nevill.
Erroneously quoted by me (‘Proc. Malac. Soc.,’ vol. v,
p. 397) as V. arakanensis.

* Venus (Anaitis) intersculpta Sowerby.
Venus (Anaitis) intersculpta Sowerby, ‘Marine
Invest.,’ vol. iv, p. i eran fig. 2.
Hab.—Offt Algoa Bay, 10-16 fathoms.

Tapes corrugatus (Gmelin).

Tapes corrugatus Deshayes: Sowerby, ‘ Marine Shells
S. Africa,’ p. 59.

The author of this species was Gmelin, and not Deshayes as
given by Reeve (Conch. Icon.,’ vol. xiv, sp. 72). Mr. Sowerby
may have been misled by Reeve’s mistake. It is curious
that Deshayes has quoted this species (‘ Cat. Conchifera Brit.
Mus., pp. 184, 185) both under the name corrugata
Gmelin and obsoleta Chemnitz, both being founded on the
same figure in the Conchylien-Cabinet.

Petricola robusta Sowerby.

With this species should be united P. typica Jonas.

*Donax madagascariensis Mood.

Donax madagascariensis Wood: Reeve, ‘Conch. Icon.,’
vol. vii, fig. 50; Pilsbry, ‘Proc. Acad. Nat. Sci.,’ Philad.;
1901, vol. li, p. 190.

Hab.—South Africa (Pilsbry) ; Madagascar, Mozambique.

This and the two following species are recorded by Mr.
Pilsbry as occurring ‘in ballast from South Africa,’ a rather
unsatisfactory locality.

*Donax erythreensis Bertin.

Donax erythreensis Bertin, ‘Nouv. Arch. Mus,,’ Paris
(2), vol. iv, p. 99, Pl. ITI, figs. 7a—d; Pilsbry, ‘Proc. Acad.
Nat. Sci., Philad., 1901, vol. liii, p. 190.

Hab.—South Africa (Pilsbry) ; Red Sea (Bertin).

voL. 1, part 1 5

66 EDGAR A. SMITH.

* Donax spiculum Reeve.

Donax spiculum Reeve: ‘Conch. Icon., vol. viii, figs.
67 a, b; Pilsbry, ‘Proc. Acad. Nat. Sci.,’? Philad., 1901, vol. liii,
p90:

Hab.—South Africa (Pilsbry).

* Cultellus decipiens Smith.

Cultellus decipiens Smith, ‘J. of Malac.,’ vol. xi, p. 39.
Hab.—Port Alfred.

* Ervilia scaliola Issel. Pl. VIII, figs. 10, 11.

Ervilia scaliola Issel: ‘ Malac. Mar. Rosso,’ 1869, p. 53,
lesbo.

Ervilia purpurea Deshayes : Sowerby, ‘Marine Invest.,’
vol. iv, p. 16.

Hab.—Buffalo River, two miles abovethe jetty (Sowerby).

E. purpurea, which is probably the same as Ervilia
scaliola of Issel, does not appear to have been described.
There are specimens in the British Museum from the Dahlac
Archipelago, Red Sea, named by Deshayes Ervilia purpurea,
but he did not, as far as I can ascertain, publish any descrip-
tion of them. They are elongate, mequilateral, narrower
behind than in front, sharply rounded at both ends, brownish-
purple, with two or three whitish rays, two down the middle
of the valves, and sometimes a third at the posterior end.
Valves moderately strong, sculptured with fine striz of
growth. Interior purplish-brown, obscurely rayed with
white, especially at the ventral margin. Muscular scars
moderately large, and the palhal sius extending 53, of the
length of the shell from the posterior end.

Length 12°5 mm., height 7, diam. 4°25,

The shells described by Issel were very small, only 5 mm.
in length, and were apparently of an uniform, pale-rose tint,
without any colour rays. Their form, however, was exactly
the same as that of the larger shells from the Dahlac Islands,

SOUTH AFRICAN MARINE MOLLUSCA. 67

so that Iam inclmed to believe that Issel’s specimens from
Suez were very young examples of the same species.

Loripes clausus (Philippi).

Loripes clausus Philippi: Smith, ‘J. of Malac.,’ vol. x1,
p. 40.
Hab.—Port Alfred (Smith) ; Natal (Sowerby).

* Lucina despecta Smith.

Lucina despecta Smith: ‘J. of Malac., vol. xi, p. 40.
Hab.—Port Alfred.

* Lucina valida Smith.
Lucina valida Smith, ‘J. of Malac.,’ vol. xi, p. 40,
Pie tras 19:
Hab.—Port Alfred.

* Cryptodon investigatoris Smith.
Cryptodon investigatoris Smith: Sowerby, ‘ Marine
Invest.,’ vol. iv, p. 12.

Hab.—Off Cape Point, 800 fathoms.

* Tellina analogica Sowerby.
Tellina analogica Sowerby: ‘Marine Invest.,’ vol. iv,
pela, Pl. WIT, fig: 4:
Hab.—Off Saldanha Bay, 55 fathoms.

*Tellina gilchristi Sowerby.
Tellina gilchristi Sowerby: ‘Marine. Invest.,’ vol. iv,
pe 2 Pl. VE, fie 3.
Hab.—Off Cape, 30-50 fath.

* Tellina regularis Smith.
Tellina regularis Smith, ‘J. of Malac.,’ vol. .xi, p. 39,
Biatiigne. ls:
Hab.—Port Alfred.

68 EDGAR A. SMITH.

*Tellina vidalensis Sowerby.
Tellina vidalensis Sowerby, ‘Marine Invest.,’ vol. iv,

pedesk |. Vill, fies 5.
Hab.—Off Cape Vidal, 15 fathoms.

* Macoma africana (Sowerby).
Tellina (Macoma) africana Sowerby, ‘ Marine Invest.,’
vol. iv, p. 14, Pl. VII, fig. 8.
Hab.—Algoa Bay, 16 fathoms.

* Macoma inclinata (Sowerby).
Tellina (Macoma) inclinata Sowerby, ‘Marine Invest.,’
vol. ty, p. 145 Pl Ville. 9:
Hab.—Off mouth of Tugela River, 27 fathoms.

* Macoma levior (Sowerby).
Tellina (Macoma) levior Sowerby, ‘Marine Invest.,’
VO ive .wio, bl. Vi hes 6:
Hab.—Off Amatikulu River, 26 fathoms, and off Tugela
River, 25 fathoms.

* Macoma ordinaria (Sowerby).
Tellina (Macoma) ordinaria Sowerby, ‘ Marine Invest.,’
wolsav,p: 145 Plo Vill for,
Hab.—Saldanha Bay, 10 fathoms.

* Theora ovalis Smith.
Theora ovalis Smith, ‘J. of Malac.,’ vol. xi, p. 39, Pl. ILI,
nes 17.
Hab.—Port Alfred.

* Semele capensis Smith.
Semele capensis Smith, ‘J. of Malac.,’ vol. xi, p. 39,
Pin Delphos: 5, 16.
Hab.—Port Alfred.

SOUTH AFRICAN MARINE MOLLUSCA. 69

* Cuspidaria nasuta Sowerby.
Cuspidaria nasuta Sowerby, ‘Marine Invest.,’ vol. iv,
p- 18; Pl. Vil, fig. 14.
Hab.—Off Cape Point, 85 fathoms.

* Cuspidaria optima Sowerby.
Cuspidaria optima Sowerby, ‘Marine Invest.,’ vol. iv,

Dede le Vt IG.
Hab.—Off Umtwalumi River, 50 fathoms.

* Cuspidaria (Cardiomya) forticostata Sowerby.
Cuspidaria (Cardiomya) forticostata Sowerby,

‘Marine Invest., vol. iv, p. 18, Pl. VIL, fig. 15.
Hab.—Off Cape Natal, 440 fathoms.

* Cuspidaria (Cardiomya) gilchristi Sowerby.
Cuspidaria (Cardiomya) gilchristi Sowerby, ‘ Marine
iInvest:, <vol.ay, p. 18, Pi. VIL, fig: 17.
Hab.—Off Cape Natal, 85 fathoms.

* Poromya curta Sowerby.
Poromya curta Sowerby, ‘ Marine Invest.,’ vol. iv, p. 17,
Blea ll, ties 13:
Hab.—Off Cape Natal, 440 fath.

* Poromya gilchristi Sowerby.
Poromya gilchristi Sowerby, ‘Marine Invest.,’ vol. iv,

petobl Vil, fie. 10:
Hab.-—Off mouth of Umtwalumi River, 50 fathoms.

* Poromya granosissima Sowerby.
Poromya granosissima Sowerby, ‘Marine Invest.,’
mOledvesp. LO, Pls Vit, fig. 12:

Hab.—Cape Natal, 54 fathoms.

70

Po

EDGAR A. SMITH.

* Poromya striata Sowerby.

romya striata Sowerby, ‘Marine Invest.,’ vol. iv, p. 16
Y; ’ ; )

PE Vl fies:
Hab.—Off False Bay, 166 fathoms.

EXPLANATIONS OF PLATES VII AND VIII,

Illustrating Mr. Edgar A. Smith’s paper ‘On South African
Marine Mollusca, with Descriptions of New Species.”

Fic.
Fria.

PLATE) Vit

.1—Conus queketti n. sp.

.2.—Clionella confusa n. sp.

.3.—Drillia albotessellata n. sp.

.4.—Aneilla ordinaria n. sp.

.5.—Marginella shepstonensis n. sp.

.6.—Mitra punctostriata A. Ad.

.7—Latirus burnupi n. sp.

.8—Bullia ancilleformis n. sp.

.9.—Ocinebra natalensis n. sp.

.10.—Dolium fimbriatum, var. natalensis n. var.
. 11.—Trifora cerea n. sp.

. lla.—Trifora cerea, sculpture magnified.

.12.—Trifora shepstonensis n. sp.

. 124.—Trifora shepstonensis, sculpture magnified.

. 13—Cerithiopsis insignis n. sp.

. 14.—Cerithiopsis chapmaniana n. sp.

.15.—Rissoina durbanensis n. sp.
.16.—Rissoina shepstonensis n. sp.

.17.—Scala durbanensis n. sp.

PLA aveuen:
1.—Scala eborea n. sp.

2—Mormula rissoina A. Ad.

Fig.
Fig.
Fia.
Fie.
Fie.
Fia.
Fig.
Fia.
Fic.
Fie.
Fie.

SOUTH AFRICAN MARINE MOLLUSCA. i

3.—Elusa natalensis n. sp.

4.—Calliostoma bisculptum n. sp.
5.—Euchelus natalensis n. sp.

6.—Glyphis fuscocrenulata n. sp.
7.—Chlamys natalensis n. sp., left valve.
7A.—Chlamys natalensis, sculpture magnified.
8.—Chlamys natalensis, right valve.
8a.—Chlamys natalensis, sculpture magnified.
9.—Pinna natalensis n. sp.

10.—Ervilia scaliola Issel.

11.—Ervilia sealiola, interior.

”

i.

=.

Wi Sherine

nie

’ oa i
af yt
J =
Y
A
'
1 7
‘y
' = t
‘
i
a i
4 i ’ r
- '
Fi
x,
' ‘ A
a ) 1
uy ' t
4
"i ites
= 25 i
. t
a é
5 1
1 Ly
f c i
; : Ae
7 i
; =
- 4
C
'
- i

WU

‘.
;
A)
'
7
a!
f

—

HES VO.

G.M

Ann.Natal

dea taa Tie
LOPLY LL

~J
..
~~

HALOCORDYLE COOPERI.

On Halocordyle cooperi sp. nov., a Hydroid
from the Natal Coast.

By
Ernest Warren, D.Sc.Lond.,
Director of the Natal Government Museum.

With Plate IX.

Tue hydroid was found at Scottsburg, Natal, in the rock
pools near low water. It occurs generally among the sea-
weeds, etc., covering polychet worm-tubes, especially old
disused tubes, which often become much encrusted with
many forms of plant life, sponges, tunicates, etc.

(1) TropHosome.—The trophosome possesses a creeping,
branched hydrorhiza from which upright stems (fig. 1) arise.
They may attain a height of 2 inches or more. ‘The main
stem curves gradually from base to summit, and is shghtly
ag-zag from right to left, where the lateral or primary
branches are given off alternately in a regularly distichous
manner. This condition of branching, as pointed out by
Allman, is common in the Sertularians, but rare among the
eymnoblastic hydroids. The hydranths are carried at the
summits of the main stem and primary branches, and are also
carried on secondary branches or ramuli, which are regularly
disposed along the distal border of the primary branches
(fig. 2).

The number and arrangement of the polyps on the upright

stems approximate to the following plan :

74, ERNEST WARREN.

APEX OF MAIN STEM (ONE TERMINAL POLYP).

First right primary branch . 1 polyp (terminal).

First left 4 iy ; . 2 polyps (last terminal).
* Second right 3 Poss r

Second left 3

Third right 4

Third left OD As

Fourth right ,, 5

Fourth left 6

Fifth right —,, 6

Fifth left es 6

Sixth right 6

Sixth left 5

Seventh right ,, 5

Seventh left 5

Highth right 5

A perisarc is well developed, and the main stem is
annulated with four or five rings immediately above the
origin of the primary branches. At the base, where the
main stem arises from the hydrorhiza, there may be as
many as twenty rings. Occasionally an internodal group
of annulations may be found (fig. 2, 7. a.).

The lateral or primary branches are annulated at their
base, and also just in front of the origin of the secondary
branches or ramuli bearing the polyps. There is con-
siderable variation in the nature of the annulation; some-
times it consists of true rings, at other times it is a spiral
ridge running three, four, or five times around the stem,
and the spiral may be either a right-handed or a left-handed
one. The secondary branches bearing the polyps are annu-
lated towards their base. There can be little doubt that a
mechanical principle is involved in the presence of these
annulations, and that they strengthen the stem, for they
appear just in those places where there is most strain.

The perisare is of considerable thickness. It is nearly
black over the main stem and greater part of the primary
branches ; but it becomes pale brown towards the extremi-

HALOCORDYLE COOPERI. 75

ties, and over the secondary branches bearing the polyps
it is very transparent and thin.

Polyp is flask-shaped; its average height from base to
summit is 0°844 mm. The endoderm of the swollen basal
portion is red in the living condition. There are two kinds
of tentacles:

(1) A single whorl of basal filiform tentacles ; the typical
number appears to be eight, although there may be mine
or ten.

(2) Typically there appear to be two whorls or verticils
of capitate tentacles, with four in each whorl; but there is
variation in this arrangement. The tentacles of the two whorls
are arranged alternately, and they are inserted on the sides
of the hypostome. The swollen extremities of the capitate
tentacles are provided with very large nematocysts, 0)°0387 mm.
in length and 0:017 mm. in breadth; while the filiform tentacles
and the general ectoderm of the polyp have nematocysts,
about 00112 mm. in length and 0:0054 mm. im breadth, but
they are far more variable in size than the large nematocysts,
and perhaps there are several kinds.

In Pennaria cavolinii Hhren. Allman describes the
filiform tentacles as being slightly swollen at their extremities;
such does not appear to be always the case in H. cooper.

In the fully expanded condition the mouth is large and
funnel-shaped.

(2) Hisroroay.—Ectoderm: The ectoderm is of the usual
type, and contains, as we have seen, two kinds of nematocysts.
The nematocysts appear to be confined to the ectoderm, as I
have been unable to find any in the endoderm; although in
the hydroid Tubularia solitaria, which is also described
in this journal, nematocysts are exceedingly abundant in the
inner layer.

The large nematocysts are mostly confined to the swollen
extremities of the capitate tentacles (fig. 6, l.n.), although I
have occasionally found them in the ectoderm of the coenosarc.
The small nematocysts occur on the capitate tentacles and
generally throughout the ectoderm.

76 ERNEST WARREN.

At the base of the hydranth the perisarc terminates in a
well-defined groove (fig. 6, p. g.), which is lined by a special
epithelium of granular cells. This groove is the remnant of
the chitin-secreting zone of ectoderm at the growing point.
It is from such a groove, or zone, that the perisarc is secreted
at the growing apices of stem, branches, or hydrorhiza.

Fig. 7 is a vertical section through the growing apex of a
branch of the hydrorhiza, and the zone of ectoderm marked
p. Ss. e€. 18 seen to be densely granular, and it is here that the
perisare is secreted. I do not beheve that the ordinary
ectoderm below (0. e.) takes any part in the production of the
perisare (p.). The apical portion of the growing branch is
naked (m. l.).

Endoderm.—The hypostome is lined by a regular columnar
epithelium, between the cells of which are wedged small,
vacuolated, and apparently glandular cells (fig. 6, v. ¢.).
They stain deeply with hematoxylin, and consequently stand
out sharply from the surrounding columnar epithelium.

Below the region of the hypostome the endoderm consists
of elongated, more or less amoeboid cells, with vacuoles.
Between these, sometimes in clusters and sometimes solitary,
are inserted large, irregularly-shaped, glandular cells (fig. 6,
g. e. c.), with large nuclei, and the cytoplasm is densely
crowded with large granules, which blacken with osmic acid.
These densely granular cells occur throughout the endoderm
of the general ccenosare.

At the base of the polyp the endoderm cells are taller and
constitute a more regular epithelium (fig. 6, b.c.) ; towards
their base, inserted on the mesoglea, the cells contain very
large vacuoles. Often in this region the endoderm does not
consist of a smgle layer of cells. A sheet of smaller, more
rounded cells, with denser protoplasm, may cover a consider-
able area of the tall columnar epithelium at the base of the
polyp (fig. 6, c.e.c.).

The endoderm of the filiform tentacles is of the usual
septate character. In the capitate tentacles the endoderm

scarcely enters the “head.”

HALOCORDYLE COOPERI. 7a

The chief food of the hydroid appears to be copepods. As
far as could be ascertained from my sections, digestion takes
place solely in the digestive cavity, the pieces of food never
being found in the epithelium. In some sections recently cut
of an HKudendrium, which feeds on similar organisms, the
separate egos of the ege-clusters of the devoured copepods
are seen to have been taken bodily into the endodermal
epithelium.

(5) Gonosome.—The female gonophore is long and ovate.
In P. cavolinii it is phanerocodonic, but im the present
species I have never found the umbrella cavity open to the
exterior, and it is therefore probably adelocodonic. The
hydroid is far from common, and consequently | have not
had an unlimited supply of material. Unfortunately, I have
not seen a gonophore empty of its contents, but I think there
can be httle doubt that it is adelocodonic. ‘The section of an
apparently ripe gonophore (fig. 4) gives no impression of any
further development. The gonophores arise from the base of
the polyp just above the verticil of filiform tentacles. There
are four radial canals, but no circular canal. The radial
canals are not connected together by an endodermal lamella
(fig. 5). The umbrella cavity is lined externally by a definite
ectodermal epithelium (e. e). Apically the radial canals slightly
expand, and contain a deeply staining gelatinous (7) mass
(fig. 4, g.m), which is probably of the nature of a basal bulb.
No rudiment of ocellus has been detected. The spadix is well
developed, and is covered with germinal epithelium four or
five layers thick. .

The germinal epithelium arises in situ as a thickening of
the ectoderm of the young gonophore bud (fig. 4, e. 7). I
have been unable to observe any migration or wandering of
sexual cells.

Since the above paragraph was written I have obtained
specimens of the male colony, which appears to be much
scarcer than the female. Out of some thirty colonies col-
lected eight were females, two were male, and the remainder
were undetermined as they were not producing gonophores.

78 ERNEST WARREN.

In one of the male colonies examined the filiform tentacles
of the polyps were distinctly swollen at their extremities ; but
there appeared to be no increase in the development of the
nematocysts, and I am not inclined to regard the condition as
of any special significance.

The average length of a mature male gonophore is 0°84 mm.
and breadth 0°60 mm. The female gonophores examined were
not quite mature, but it is probable that they are slightly
smaller.

The male gonophore is provided with four tentacular knobs

bearing large nematocysts. There is no circular canal. The

blind ends of the radial canals are occupied by a gelatinous,
deeply-staining mass which shows concentric lamination, and
is crowded with irregularly-shaped, somewhat refringent,
bodies (text-fignre 3, g), which are probably the granules
giving the endoderm its scarlet colour during hte. The
general ectoderm of the radial canals, etc., contains such
granules, but they are not present m great abundance. In
the distal portion of the radial canals of the female gono-
phore there is a special middle strip of swollen cells both
in the inner and outer layer of the canal, and here, also, the
granules are more abundant than elsewhere.

HALOCORDYLE COOPERI. 79

The male gonophore is adelocodonic, there being no open-
ing to the umbrella cavity.

The male gonophore is characterised by the presence of
four little papille of ectoderm projecting into the umbrella
cavity (text-figures 1 and 2, p). They arise opposite the
fourth radial canals (fig. 4) towards the upper end of the
gonophore. It may be noticed that the flat layer of ectoderm
lining the umbrella cavity thins off and apparently disappears
around the base of the mushroom-shaped papilla. It is not
possible, at present, to suggest any function or meaning to
these very curious structures. They consist of elongated
ectoderm cells with fairly conspicuous nuclei. They have
not been found in the female gonophore, but they occur in
all the mature male gonophores that have been examined.
IT am not aware that similar structures have been recorded in
other gonophores or meduse. Their position on the radial
canals recalls the gonads of calyptoblastic meduse.

The male gonophore dehisces by the expansion of the
endodermal spadix, which forces its way through the distal
extremity of the gonophore (fig. 2).

(4) Systematic position.—The present species differs
from P. cavolinii Hhren, and P. gibbosa Agass, and agrees
with Halocordyle tiarella (Ayres), from the Atlantic
shores of North America, in that the capitate tentacles tend
to be arranged in whorls instead of being irregularly scattered.
It is very interesting, however, to note that this verticillate
character seems to be on the point of being acquired. In
several colonies collected, which appear to differ in no other
character, I have found great irregularity in the disposition
of the capitate tentacles. In these colonies some of the
polyps may have a distinct distal verticil of four capitate ten-
tacles, while below these there are scattered seven or eight
capitate tentacles with no obvious arrangement. In a very
few of the polyps some twelve capitate tentacles are irregu-
larly scattered, and even the distal verticil is indistinguishable.
It is important to notice that this tendency to vary in the
arrangement of the tentacles concerns the colony as a whole,

80 ERNEST WARREN.

and it is not merely an individual polyp variation. Thus, all
the polyps of a colony tend to have the capitate tentacles in
whorls, or all tend to be irregular in the character.

In the accompanying diagram a view from above is sup-
posed to be taken. The circles are the “heads” of the
tentacles. Three distinct stages may be observed:

In fig. 1 some twelve tentacles are irregularly scattered.

In fig. 2 the distal verticil of four is acquired, and these are
joined together by a circle as in a floral diagram.

In fig. 3 the number of tentacles is reduced to eight, and
they are arranged in two whorls in the typical manner.

: &
a gS :

If more material comes to hand, a statistical mvestigation
with respect to this matter will be made.

The present species also agrees with Halocordyle
tiarella, in that the gonophores arise above the whorl of
filiform tentacles and not in the verticil.

It differs from Halocordyle tiarella and from Pen-
naria in that the gonophores are adelocodonic, and no
processes representing marginal tentacles on the umbrella
were found in the female gonophore.

The species Halocodyle ccoperi is named after my
friend Mr. Arnold Cooper, who was shore-collecting with me
at the time it was discovered.

HALOCORDYLE COOPERI. 8]

EXPLANATION OF PLATE IX,

Illustrating Dr. Ernest Warren’s paper “On Halocordyle
cooperi sp. nov., a Hydroid from the Natal Coast.”

Fie. 1—H. cooperi; natural size, showing three main stems arising
from the branched hydrorhiza.

Fie. 2.—x 7. Small portion of frond, with an internodal group of
annulations (7. a.).

Fie. 3—x 15. Primary branch with hydranths bearing female
gonophores, arising Just above the verticil of filiform tentacles. The
capitate tentacles are in two alternating whorls.

Fre. 4.—x 50. Female gonophores in longitudinal section in various
stages of development. e. uw. is the ectoderm from which the generative
epithelium and ectoderm lining umbrella cavity originate.

Fie. 5.— x 50. Female gonophore in transverse section, showing
absence of endodermal lamella between the radial canals.

Fie. 6—x 75. Longitudinal section of hydranth, showing its
general histological character and the perisarc-groove (p. 4.).

Fie. 7.—-x 75. Growing point of hydrorhiza, showing the epithe-
lium secreting the perisare (p.s.e.), and the naked terminal portion of
coenosare (nN. ¢.).

EXPLANATION OF REFERENCE LETTERS.

b. c. Cells of basal endodermal epithelium. c. e. ¢. Covering endo-
dermal cells. ¢. m. Closed mouth. d.c. Digestive cavity. e. Endoderm.
e.u. Eetoderm of umbrella (including germinal epithelium) in gono-
phore bud. e.e. External ectoderm of umbrella cavity. g. e. Generative
epithelium. g.e.c. Glandular endodermal cells. g.m. Gelatinous mass
(?) at termination of radial canal. h.e. Hypostomal endoderm. 7. a.
Internodal annulation. /.n. Large nematocyst. m. Mouth. m.s. Main
stem. #.c. Naked conosare. o0.e. Ordinary ectoderm. 0.m. Open
funnel-shaped mouth. p.b. Primary branch. jp. Perisare. p.g. Peri-
sarc-groove. p.s.e. Perisarc-secreting ectoderm. 7. Ramulus or
secondary branch. 7.¢. Radial canal. s. Spadix or manubrium. = s.e.
Septate endoderm of tentacle. s.. Small nematocyst. u.c. Umbrella
cavity. v.c. Vacuolated deeply staining cells of hypostome and basal
epithelium.

voL. 1, part 1. 6

ee is 1 oP el] ee ses

}

ps 5 a ° 2 Fx j 7 4
es ? 2 “« ; ; | :
| a a *-
~ < : 4 i t : sy &
a 7 < : | | | |
. 2 ay = | | |
2k é. JAic « - | a
‘ ‘ abe : . | .
i 7 ’ | F
a) , | |
| c : Ps
y | | ;
, a |
x :
‘
—,
-
a : | |
,
’ | |
rc . 7 : |
. : |
i a |
‘ ° 7 / | |
f : |
S | :
- | |
, i . |
at) ? | 7 :
i | |
; + « |

oy
~
~
~

Ann. Natal G. Mus. Vol. 1.

1 nat. size.

Huth, Lith? London.

s

ren, del.

Wa

TUBULARIA SOLITARIA. 83

On Tubularia solitaria sp. nov., a Hydroid
from the Natal Coast.

By
Ernest Warren, D.Sc.Lond.,
Director of the Natal Government Museum.

With Plates X and XI.

Tuts hydroid has been collected at several places along
the Natal Coast. It is solitary in habit, and I have only
found it attached to sponges. The species of sponge is not a
matter of indifference; for, although a certain siliceous scarlet
sponge is exceedingly common on this coast, yet I have never
found the hydroid attached to it. I have only discovered it
on a certain siliceous dark maroon coloured sponge and on a
siliceous white sponge. The hydroid occurs between the tide
marks, and is imbedded in sponges covering the surfaces of
rocks which happen to be somewhat protected from the
violence of the waves. The individual hydroids may generally
be seen in clumps irregularly scattered, and spreading over
a considerable area of sponge.

Trophosome.—The hydroid varies in height, but its
maximum length projecting beyond the general surface of
the sponge is about half an inch (PI. X, figs. l and 2). It
may be seen by the naked eye that the endoderm of the
hydranth and of the gonophores is of a beautiful rose-red
colour, while the tentacles and the hydrocaulus are trans-
lucent and whitish. The hydroid is fixed to a considerable

84. ERNEST WARREN.

depth in the substance of the sponge by a branching root or
hydrorhiza (figs. 3 and 4).

Around the mouth is a ring of oral or distal tentacles
sixteen to twenty-one in number. ‘These originate in as
many prominent ridges which run down the hypostome. It
is as if the tentacles spring from the sides of the hypostome
and adhere to it, becoming free around the mouth.

The hydranth is somewhat expanded towards the base, and
from this region there arises a single verticil of basal or
proximal tentacles about sixteen to twenty-one in number.

The gonophores arise in small clusters from short semi-
erect peduncles, which spring from the body of the hydranth
just above the verticil of basal tentacles (fig. 4). In each
cluster there are three or four gonophores.

Both male and female gonophores are produced by the
same hydranth; but as far as I have observed they are
invariably formed on distinct peduncles. This fact favours
the view of regarding the peduncles as blastostyles which
are either male or female.

The hydrocaulus is covered by a thin, soft perisare covering,
which commences in the perisarc-groove situated just below
the attachment of the hydranth (fig. 5 p.g.). As the hydro-
caulus elongates the perisarc substance is secreted by the
epithelium of the groove. The softness of the perisare
differentiates the present species from the typical tubularia.

The area of attachment of the hydranth and hydrccaulus
is small, and under certain conditions the hydranth appears
to become decrepit and withers or drops off. The blind end
(fig. 3 a, b) of the hydrocaulus can, without doubt, produce
another hydranth. I have not, however, been fortunate
enough to find a specimen in which such re-formation of the
hydranth was taking place; but in specimens in which the
hydranth is withering it seems certain that the life of the
individual is not drawing to a close, for at the same time it
may possess rich supplies of food-material stored up in the
hydrorhiza (fig. 4 ¢).

The basal end of the hydrocaulus is imbedded in the sponge

TUBULARIA SOLITARIA. 85

to a quite variable extent. At least half of the stem (fig. 3)
is usually below the surface. At the base of the imbedded
hvdrocaulus is the hydrorhiza, and from it may be given out
branches of two kinds: (1) thin rootlets which penetrate into
the substance of the sponge and fix the hydroid in its bed
(fig. 4, r); (2) thickened fleshy structures (fig. 4, ¢) in which
apparently nutritive substances are stored in the endoderm
in amanner recalling the storage of carbohydrates, ete., in
the tubers of a plant. These thickened branches are of a
characteristic yellow colour in preserved specimens. In some

2

individuals no special “tubers” are formed, but the hydro-
rhiza as a whole expands and takes on the function (fig. 3; 2
and 4).

The apices of the rootlets are swollen into a sort of cap
(fig. 3, a); this appearance, recalling the root-cap of a plant,
is due to a marked thickening in the ectoderm at the apex.

(2) Histology.—The ectoderm of the hydroid is of typical
structure ; but the endoderm shows much differentiation and
is highly specialised.

Both large and small nematocysts occur in the ectoderm,
and the endoderm is richly supplied with the large variety
(figs. 8 and 10, J. n.). The large nematocysts are nearly
spherical in shape; their average size in spirit specimens
is 12°1 » in length and 11°6 win breadth. The small nemato-
cysts are probably of several kinds, as they tend to vary con-
siderably in size and shape; their average size is 6°24 4 in
length and 4°6 w in breadth.

Fig. 5 is a vertical median section of the hydranth and
upper part of hydrocaulus. The oral or distal tentacles (o. ¢.)
adhere to the sides of the hypostome, and the ectoderm is, so
to speak, squeezed out between the endoderm of the tentacle
and that of the hypostome, and only a thin layer of mesoglea
remains (fig. 8, mes.). The endoderm of the oral tentacles is not
regularly septate in character. The endoderm of the hypostome
is markedly ridged, so that in transverse section (fig. 11, e. 7.)
eight to ten prominent projections may be seen. These
ridges hang down into the digestive cavity of the hydranth as

86 ERNEST WARREN.

conspicuous lobes (s.w.). The endoderm of the hypostome is
supplied with the usual deeply-staining secretory cells (figs. 5,
7, 8s. ¢.). The tall endoderm cells are not amceboid to any
extent.

Below the hypotome we pass into the digestive cavity
(fig. 5,d.c.). Here the ordinary endoderm cells are exceedingly
amoeboid, and are of very irregular outline (a.e.). Wedged
between the ordinary cells are special digestive or gland
cells (g.l.) which stain intensely with hematoxylin and other
stains ; also large cnidoblast cells are fairly abundant. At the
expanded basal region of the hydranth the endoderm is
markedly differentiated into three layers: (1), a covering
layer of somewhat short amoeboid cells (b. ep.) and gland
cells (fig. 5a, gl.c.) which are separated by a kind of base-
ment membrane or mesoglea from (2), a middle-thick layer
of highly vacuolated cells (r.ed.) and (3), smaller less
vacuolated cells which he in immediate contact with the
mesoglea (b. ed.).

The covering layer of amoeboid cells marked b. ep. is
shown enlarged in fig. 5 a, and it is especially charac-
terised by containing in greater abundance than the rest of
the endoderm large granular masses enclosed in vacuoles
(pr.). These masses appear to escape from the cells and
pass down into the hydrocaulus, where a couple may be
seen, marked pr., in fig. 5. I suggest that these bodies are
to be regarded as masses of worked-up food material which
are being distributed to different parts of the hydroid.

An open channel leads from the digestive cavity of the
hydranth into the ccelenteron of the hydrocaulus (fig. 5,c. ch.).
‘he upper portion of this channel is narrow, and is lined by
small amoeboid endoderm cells continuous with the layer of
covering cells on the floor of the digestive cavity (see fig. 12,
c.e.p.). As the passage approaches the diaphragm (fig. 5, d)
it greatly expands and is lined by long, tlat, non-amceboid
cells. The passage is greatly constricted as it passes through
the small aperture of the diaphragm.

The basal or proximal tentacles arise from the margin of

TUBULARIA SOLITARIA. 87

the basal expansion of the hydranth (figs. 5 and 12, 6. ¢.).
The endoderm of the tentacles is continuous with the middle
layer of the endoderm of the polyp (7. ed.). It is far from
regularly septate in character, although it approaches this
condition more closely than in the case of the distal tentacles.

The mesoglea projects inwards between the hydrocaulus
and the hydranth, forming a kind of diaphragm, and leaves
open only a small pore of communication (fig. 5,d.). The
possession of this diaphragm is perhaps associated with the
habit of renewing the hydranth. The wound resulting from
the casting off of the old hydranth would be reduced to a
minimum, and healing could very rapidly be effected.

Immediately below the hydranth can be seen the perisare
groove (fig. 5, p.g.). An enlarged view is shown in fig. 6.
The secretory cells (p.s.¢.) are richly granular, and towards
the free surface the perisare substance can be observed to
run in for some little distance between them. The perisare
groove is shown in horizontal section in fig. 13.

The modification of the endoderm in the hydrocaulus is
considerable. In the middle region a longitudinal banding,
due to the presence of endodermal canals, can readily be seen
in the preserved specimen (fig. 4). This banding is more
conspicuous in the living animal, and can be traced up to the
hydranth, in the immediate neighbourhood of which they
_ become closely applied to each other.

Fig. 10 is a transverse section of the middle region, showing
in this case sixteen endodermal canals. A reticulum of fine
branching endoderm cells with large nematocysts occupies
the greater part of the ceelenteron. An enlarged view of a
couple of canals is seen in fig. 15. Each canal consists of an
outer wall of regular columnar epithelium on which no cilia
could be detected with certainty. The inner surface of this
epithelium is concave, and the canal is completed by branch-
ing cells (c. c.) forming a thin roof.

These canals anastomose to some extent, and at the upper
end of the hydrocaulus they fuse into some nine to twelve
canals. At some little distance below the diaphragm the

88 ERNEST WARREN.

cavity of the canals, roofed over by thin branching cells,
expands considerably, and in the region of the perisare
groove these coverings of thin cells come into close contact
and constitute some nine to twelve radial septa (fig. 13, v. s.).
The cells in contact with the mesoglea do not appear columnar
in fig. 13 ed. owing to the fact that the cells, hanging down
from the diaphragm (fig. 5 ed.), are cut obliquely. The endo-
dermal canals (e. c.) open into the vesicular endoderm of the
hydranth by irregular channels, which pass round the edge of
the diaphragm-pore (fig. 5, 0. e. c.). Bodies similar to those
seen in the endoderm cells of the floor of the hydranth (figs. 5
and 5 a, pr.) can be observed in the hydrocaulus (fig. 13, g. m.,
and fig. 5, pr.).

The endodermal canals probably serve to convey nutritive
fluids. It must be remarked that cilia were not clearly seen
on the columnar cells, although the specimens were carefully
fixed.! The general histological condition of the sections was
very good. I believe, however, that cila or flagella are
present, which are of too delicate a nature to remain clearly
visible after the ordinary processes of fixation and imbedding,
etc. In some of the specimens examined the columnar cells
of the endodermal canals contained a considerable number
of globules similar to those which are so abundant in the
endoderm of the tuber-like expansions of the hydrorhiza.
Towards the base of the hydrocaulus the endodermal canals
gradually fuse together and constitute a regular layer of
columnar epithelium, and the canals, as such, disappear. As
the tuber-like expansions are approached, the endoderm cells
become taller and more crowded with globules. Fig. 16 is a
small piece of a transverse section of atuber. Here the endo-
derm is so tall that the lumen of the tuber is almost obliterated.
The cells are densely crowded with fairly large globules of a
wonderfully uniform size. The globules appear perfectly homo-
geneous, and without doubt consist of reserve food material.

! Fixation was accomplished by (1), Flemming’s solution, six hours ;
(2), half concentrated corrosive sublimate in 30 per cent. spirit and
13 per cent. acetic acid, applied hot.

TUBULARIA SOLITARIA. 89

I regard the canals as being for the purpose of conveying
down to the tuber nutritive fluids, which are elaborated by
the endoderm cells into the globules so abundantly stored
up in the tuber.

The ectoderm below the perisare groove is of considerable
thickness. In several specimens which were sectioned, but
not in all, the ectoderm, especially in this region, was crowded
with large granular bodies (fig. 14, g.m.), which stain with
much intensity with hematoxylin. The bodies were of irre-
gular shape, and some of them, especially those in the neigh-
bourhood of the mesoglea, were apparently in the condition
of breaking up into small granules. These bodies stain
similarly, and have quite the same aspect as the bodies
marked g: m. in fig. 13, or pre in fig. 5. I consider it
probable that active growth was taking place in this region,
and that the bodies in question are coagulated albumens or
some other substances, which had been passed out through
the mesoglea into the ectoderm.

In the ectoderm of the hydrorhiza and the tubers the
wedge-shaped cells are characterised by being finely granular
in nature, and in this respect they resemble the cells forming
the perisarc groove (fig. 16). It is probable that these cells
still retain their power of secreting perisarc substances.

It would be interesting to ascertain whether there is any
symbiotic or parasitic interaction between the hydroid and
the sponge further than the mechanical support rendered by
the latter to the former. The sponge appears to attempt to
shut itself off as much as possible from the hydroid; the
tissues of the sponge in the immediate neighbourhood of
the hydroid are denser and more fibrous than further in, thus
forming a kind of cyst-wall (fig. 16s. p.).

(3) Gonosome.—The gonophores spring in clumps of
three to five from short semi-erect peduncles (or blastostyles)
(fig. 5 ped.), which arise from the hydranth just above the
verticil of basal tentacles. A gonophore originates as a
swelling or out-pushing from the peduncle. The ectoderm
at the apex of the swelling thickens, and the endoderm grows

90 ERNEST WARREN.

up around it. This thickening of ectoderm will produce the
ectoderm-lining to the umbrella cavity and the germinal
tissue. I have not observed any migration or wandering of
the sexual elements. As we have already seen, male and
female gonophores are invariably produced on distinct
peduncles, but both kinds of peduncles occur on the same
hydranth. In a transverse section of a mature gonophore
(fig. 9) we find—(1) the outer ectoderm; (2) two thin layers
of endoderm with no development of radial canals; (3) a
thin layer of ectoderm, which is the outer ectodermal wall
of the umbrella cavity; (4) umbrella cavity, which is mostly
occupied by (5), germinal tissue covering (6), manubrium or
spadix.

When the gonophore is mature the umbrella cavity opens to
the exterior by a small apical aperture (fig. 19, 0), and it is
probable that fertilisation takes place through this opening.
No rudiments of umbrella tentacles in the form of processes
or knobs are produced. Thus, with the exception of the
small aperture to the exterior, the gonophore may be called
adelocodonic, since no radial canals and no marginal tentacles
are formed.

Among the clumps of gonophores one frequently meets
with abortive or semi-abortive specimens (fig. 18). These
abortive gonophores appear to be almost invariably the
terminal gonophore of the peduncle, although it must be
remembered that the terminal gonophore is not always abor-
tive. The terminal gonophore of a peduncle differs from the
lateralS in (1) the great thickness of the mesoglea, and in this
it agrees with that of the peduncle; (2) the great thickness of
the outer ectoderm. These two differences can be picked
out at once im the sections, and are shown in the comparison
of an abortive and normal gonophore in fig. 18.

An abortive gonophore, male or female, tends to be about
one half or two thirds of the size of a normal gonophore.
The endoderm of the spadix is relatively larger than in a
normal gonophore, and contains swollen nuclei. The umbrella
cavity is completely filled with a tissue of small stringy

TUBULARIA SOLITARTA. 91

ectoderm-cells continuous with a thick layer of ectoderm on
the outside of the gonophore.

In the fibrous mass of ectoderm (e.a. w.) there can occa-
sionally be distinguished one or more small rounded cells (a. 0.) ;
these can be regarded as abortive ova.

Abortive female gonophores of this nature are absolutely
sterile; but in the case of male gonophores the sterility may not
be complete. In the semi-abortive male gonophores a variable
number of ripe spermatozoa may be found scattered amongst
the fibrous or stringy ectoderm filling the umbrella cavity.

This difference in structure and the very frequent abortion
or semi-abortion of the terminal gonophore are remarkable ;
but before any safe conclusions can be arrived at a com-
parison must be made with other hydroids which bear gono-
phores on peduncles or blastostyles.

(4) Formation of the actinula.—I have only made a
few observations im connection with the development of the
actinula. Among the ripe female gonophores one can find
cases in which the ova are apparently fusing together
(fig. 20, g. e. m.). This may almost certainly be regarded as
the ingestion of ova by some one or more fertilised eggs. It
may be noticed that in fig. 20 the germinal epithelium has
shehtly extruded itself through the opening of the umbrella,
and it is from this point, where fertilisation would readily
occur, that the ingestion of ova is taking place.

Fig. 21 represents the youngest condition found. Here I
cannot say with certainty which is the fertilised nucleus ;
it 1s quite possible that the enlarged swollen nuclei sn, and
sng may both be fertilised nuclei, as it is extremely probable
that a considerable number of the ova would be fertilised.
The reason for regarding these nuclei as distinct from the
others may be explained by reference to fig. 25. The in-
gested ova (t.0.) can be seen, but the remarkably swollen
condition of the nucleus (fig. 21) is not apparent. The
germinal epithelium is ultimately fused into one or two
masses, according to whether one or two actinule will
ultimately be developed in the gonophore. Whether ferti-

92 ERNEST WARREN.

lised or unfertilised, the ova of the germinal epithelium fuse,
in the majority of cases, into a single egg-mass (fig. 22) ; but
occasionally there are two such masses. In a densely gran-
ular mass, with nuclei in all stages of disintegration (fig. 25),
I have been unable to distinguish with certainty the first
segmentation nucleus. The nearest approach is seen in
fig, 22, where the egg-mass contains about four nuclei, one of
which is shown in a state of division (dy.). Towards the
centre the egg-mass consists of vacuolated protoplasm, with a
number of large yolk masses and granules; towards the
periphery the protoplasm is dense and finely granular and
there are no vacuoles. The nuclei are contained in this outer,
denser portion. On the formation of the egg-mass the
manubrium shrinks considerably (figs. 22-24). Fig. 26 is an
enlarged piece of an egg-mass, similar to that in fig. 22, and
containing some four or six nuclei. The nucleus (7.) 1s sur-
rounded by a clear area, and is characterised by staming
exceedingly faintly, owing to the small amount of chromatin
it contains. The nucleus consists of a delicate nuclear mem-
brane and a fine reticulum of chromatin.

The egg-mass secretes a firm egg-membrane (fig. 26, e. m.).
In fig. 23 the embryo has developed further, and it is often
remarkably irregular in shape. At this period it consists of
some eighteen to twenty blastomeres, which are not separable
from one another by cell-outlines nor from the central
vacuolated mass containing innumerable granules and yolk-
masses.

The material at my disposal has not permitted further
observations on the development of the actinula.

In fig. 17 is represented a gonophore containing an actinula.
Such is shown in section in fig. 24. ‘he remainder of the
yolk-spheres are contained in the endoderm. ‘lhe number of
tentacles varies from about nine to twelve, although ten seems
to be the typical number.

It is probable that these tentacles correspond to the basal
or proximal tentacles of the hydranth. The actinula in fig. 17
has every appearance of being ready to emerge from the

TUBULARIA SOLITARIA. 93

gonophore, but I have not found a mouth or oral tentacles
developed in any of the specimens. In fig. 24 the part of the
actinula next to the manubrium will probably be the portion
which will elongate and form the hydrocaulus ; the opposite
pole would then be the future oral pole.

(5) Systematic position.—The present species is some-
what intermediate in its characters between the genera
Corymorpha and Tubularia, but the occurrence of an
actinula seems sufficient to place it in the genus Tubularia.

It agrees with Corymorpha in the following characters :

(1) Sohtary hydrocaulus.

(2) The softness of the perisare.

(3) The base of hydranth occupied by differentiated layers
of endoderm.

(4) The arrangement of the endodermal canals approaches
the condition seen in Corymorpha rather than in Tubularia.

It agrees with Tubularia in the following:

(1) Occurrence of an actinula.

(2) No papilhform or filamentary appendages at the
proximal end of hydrocaulus.

(5) The comparative fewness of the oral or distal tentacles—
sixteen to twenty-one instead of about eighty, as in Cory-
morpha.

(4) Pendulous endodermal lobes hanging down from
hypostome.

(5) The number of basal or proximal tentacles—sixteen to
twenty-one.

(6) The shape of the nematocysts.

On the whole it may be considered that in the trophosome
it is closely related with Corymorpha, and in the gonosome
with 'Tubularia; however, the presence of an actinula is
so characteristic of Tubularia that I have considered it
advisable not to found a new genus for the reception of the
present species.

94 ERNEST WARREN.

EXPLANATION OF PLATES X AND XI,

Illustrating Dr. Ernest Warren’s paper “On Tubularia
solitaria sp. nov., a Hydroid from the Natal Coast.”

Fic. 1.—Natural size. A group of hydroids attached to a white
siliceous sponge.

Fic. 2—x 5. Somewhat enlarged view of a specimen attached to
sponge.

Fig. 3—x 5. Six specimens showing the variable extent to which
the hydrocaulus is imbedded in the sponge, and also the variable
character of the hydrorhiza.

Fie. 3a—x 10. A specimen in which the hydranth has shrivelled
away or dropped off.

Fic. 4.— x 25. Enlarged drawing of side view of specimen. Notice
the ridges on the hypostome, which are the continuations of the oral
tentacles ; the gonophores arising in clusters from semi-erect peduncles,
which spring just above the verticil of basal tentacles; the longitudinal
banding, which is more particularly obvious over the middle region of
the hydrocaulus, due to the presence of the endodermal canals ; a thin
“rootlet” (vr) on the right of the hydrocaulus, anda thick “ tuber” filled
with reserve food on the left.

Fie. 5.—x 75.—Longitudinal vertical section of hydranth and upper
portion of stem. Notice: longitudinal ridge of hypostome projecting
down as a lobe (/.) into digestive cavity ; base of hydranth occupied by
endoderm, consisting of (1) basal endodermal epithelium (b. ep.) ; (2)
reticular layer of vesicular endoderm ; (3) smaller endoderm cells lying
next to the mesoglea (b. ed.); diaphragm (d.) with an endodermal epi-
thelium (ed.) on one side and a vertical septum (v. s.) on the other;
perisare groove (p. g.).

Fic. 5a.—x 300. Basal epithelium from the floor of the digestive
cavity of hydranth. It consists of amceboid cells enclosing masses of
apparently worked-up food material (p. 7.) and of glandular cells.

Fig. 6.—x 300. Perisarc-secreting groove showing the production
of the perisare (p).

Fie. 7.— x 300. Hypostome endoderm with the secreting cells, which
stain intensely. Such are usually found in the hypostome of the hydroid-
polyp.

Fie. 8.— x 300.—Cross section of hypostome region; it shows the
absence of ectoderm over the contact-plane of the tentacular ridge.

TUBULARIA SOLITARIA. 95

Fie. 9.— x 125.—Cross section of female gonophore; it shows the
absence of distinct radial canals in the endoderm of the umbrella wall.

Fre. 10.—x 75. Cross section towards the basal end of hydrocaulus
to show the endodermal canal system. The vesicular endoderm, occupy-
ing a considerable portion of the cavity, contains large nematocysts
(1. ”.).

Fie. 11.—x 50. Cross section of hypostome region to show the
tentacular and endodermal ridges.

Fie. 12.—x 50. Cross section through the base of polyp showing :
(1) Central channel lined by an irregular epithelium: (2) vesicular
endoderm ; (3) the origin of the basal or proximal tentacles.

Fie. 13.—x 100. Cross section through the region of the perisare-
secreting groove. It shows (1) expanded endodermal canals (e. c.) with
vertical radial septa (v. s.); (2) horizontal section through the endo-
dermal epithelium (ed.) hanging down from the diaphragm ; (3) the tall
glandular epithelium secreting the perisare.

Fie. 14.— x 200. Small piece of cross section just below the region
of the perisarc-secreting groove. In this specimen the ectoderm is
charged with large granular masses of deeply-staining substance, similar
in appearance to the masses marked pr. in the basal epithelium (Fig. 5a),
pr, Fig. 5, and g. m., Fig. 13. Such masses are by no means invariably
found in the ectoderm below the perisare groove.

Fic. 15—x 200. Small portion of Fig. 10 to show endodermal
canals.

Fie. 16.—x 200.—Cross section through the tuber-like outgrowth
from hydrorhiza, as it lies in situ in the sponge. The modified tissue
of sponge contiguous with the hydroid is also shown (sp.). The tall
endoderm cells are crowded with globules.

Fic. 17.—x 50. Gonophore with enclosed actinula and shrivelled
manubrium.

Fie. 18—x 75. Peduncle carrying terminally an abortive gonophore
and laterally an ordinary fertile gonophore.

Fie. 19.— x 200. Apical portion of gonophore showing the opening
of the umbrella to the exterior.

Fie, 20.—x 150. Longitudinal section of gonophore with the ova
being absorbed into an egg-mass (g. e. 7v.). The first segmentation
nucleus has not been identified with certainty.

Fie, 21— x 500. The beginning of the formation of the egg-mass.
It is suggested that sn, and sn, may be fertilised nuclei.

Fie. 22.—~x 150. Longitudinal section of gonophore occupied by
embryo surrounded by egg-membrane (e. m.), and possessing several
nuclei, one of which is in a state of division (dy.).

96 ERNEST WARREN.

Fie. 23.— x 150. Longitudinal section of gonophore with irregularly

shaped embryo, consisting of large blastomeres with nuclei which stain

with difficulty, and apparently contain but little chromatin. The yolk
is concentrated towards the centre.

Fic. 24—»x 150. Longitudinal section of gonophore with fully-
developed actinula.

Fig. 25.—x 350. Egg-mass with ingested ova. It may be noticed
that the nuclei are not swollen as in Fig. 21 sn, and sn,.

Fie. 26—x 300. Egg-mass with several nuclei, of which n is one.
The centre of the mass is occupied by vacuolated protoplasm and yolk.

EXPLANATORY REFERENCES.

a, Actinula. a.o. Abortive ovum. «a.e. Amceboid endoderm cells.
b. Truncated end of hydroid after the hydranth has dropped off. _b. ed.
Basal endoderm cells lying in contact with the mesoglea. b. ep. Basal
endodermal epithelium. 6. ¢. Basal or proximal tentacles. c¢.e. Covering
cells roofing over the endodermal canals. c. e. p. Central endoderm
passage. c.ch. Central channel. d. Diaphragm. dy. Dyaster. d.c.
Digestive cavity of hydranth. d.t. Distal tentacles. d.a. Digestive
cavity of actinula. ect. Ectoderm. end. Endoderm. e.c. Endodermal
canal. e.7. Endodermal ridge. e.rh. Ectoderm of hydrorhiza. e. a. w.
Ectoderm filling abortive umbrella-cavity. e. uw. Ectoderm of umbrella,
including both the germinal epithelium and the ectoderm lining the
umbrella-cavity. ed,. Endoderm covering diaphragm on the under
surface. ed. Endoderm running up under the diaphragm. e.m. Egg-
membrane. g.m. Granular masses of worked-up food material (?).
g.e.m. Generative epithelium fusing into an egg-mass. g.e. Germinal
epithelium. gl. Gland cell. gl.c. Glandular cell of endoderm. /. Endo-
dermal lobe. J. x. Large nematocyst. mes. Mesoglea. 7. 0. Ingested
ova. my,. The ordinary thin mesoglea of gonophore. m,. The thick
mesoglea of abortive gonophore. 2. Nucleus. 0. Opening of umbrella
cavity to the exterior. o.e.c. Opening of endodermal canals into
hydranth. o.¢. Adhering portion of oral or distal tentacles. p. Peri-
sare. pr. Mass of worked-up food material in floor of hydranth. p7r;.
Ditto in hydrocaulus. p.t. Proximal tentacles. p.g. Perisare groove.
pr. Proteid-like mass. ped. Peduncle. p.s.c. Perisarc-secreting cells.
r. Rootlet of hydrorhiza. 7. ed. Reticular endoderm. +. f.g. Reserve
food-globules of endoderm. s. a. Swollen apex of rootlet. sp. The
modified sponge-tissue contiguous with hydrorhiza. — s. c. Secreting
cells. s.c.p.g. Secreting cells of perisare-groove. s.. Small nemato-
cyst. s.m. Shrivelled manubrium, s.7,. Swollen nucleus with nucle-
olus. s.%. Swollen nucleus with chromatin reticulum. ¢. Tuber-like
portion of hydrorhiza. v.s. Vertical septum. v.ed. Vesicular endo-
derm. v.p. Vacuolated protoplasm. y. Yolk-mass.

Pl X.

Huth Lith? Landon

10 «75

Ann. Natal G. Mus. Vol.1

Warren del

Warren, del

>

Huth Jath* London

£

NOTES ON A NEW SPECIES OF GYMNOPLEA. 97

Notes on a New Species of Gymnoplea from
Richmond, Natal, South Africa; Adiaptomus
natalensis (gen. et sp. nov.).

By

Arnold W. Cooper, F.R.MLS., ete.

With Plate XII.

THE following notes on a new species of Gymnoplea found
by Mr. J. Y. Gibson and myself near Richmond in November,
1905, may be of interest. During the latter part of 1905 we
had been making periodical visits to several pools in a marsh
with the view of ascertaining what varieties of aquatic lfe
occur during the cycle of the year. During the months of
August, September, and October Diaptomus orientalis
was plentiful; the new species now being described first
appeared in November. I have no reason to think it had
been overlooked earlier because of its larger size and dis-
tinctive features. It is, however, rather local, not beme
found in all the pools examined, although its range has
extended since first being observed. This species and D.
orientalis are the only two Gymnoplea which have been
seen in this locality during our visits, which have been made
fortnightly. A marked peculiarity of the new species is that
both antennz in the female and the left antenna in the male
have twenty-six joints; I have not seen any Gymnoplea
described with more than twenty-five. There are other
differences from Diaptomus and Paradiaptomus, which will
be seen from the following description.

von I, PART <1, 7

98 - ARNOLD W. COOPER.

I have not seen any note mentioning the extreme elasticity
of the spermatophores which the following incident exem-
phifies. Having mounted a specimen in damar after the usual
fixation in perchloride of mercury and acetic acid, passing
through absolute alcohol after staining, and clearing in oil of
cloves, a sight accident happened to the mount. In trying
to re-arrange the specimen with a fine needle, one of the
spermatophores attached to the vulva became detached and
stretched to nearly twice its normal length, the two ends
being connected by a thread-like portion only ; in less than
an hour afterwards it had assumed nearly its former shape
and proportions.

Description:

Bopy ;

Total length of female 1-8-1‘9 mm. Male somewhat smaller.

Male.—Thoracic segments five, the anterior segment
being partially fused with the head, the last thoracic segment
being rounded along the posterior edge. Abdomen five
segments. Furca symmetrical, each fork with five plumous
bristles (fig. 2). Genital aperture at the anterior end of the
second abdominal segment. No median dorsal spine.

Female similar to the male, except that the last thoracic
seoment is drawn out into a right and left backwardly directed
flange (fig. 3). Abdomen consists of two segments, the
vulva being in the middle of the first. Furca symmetrical,
each fork bearing five short swollen plumous sete. No
median dorsal spine.

CEPHALIC APPENDAGES.

Antenne 1, Male.—The right antenna consists of twenty-
three joints (fig. 4), the terminal prehensile portion having four
joints, the first of these (twentieth joint) with well-developed
terminal hinges; joints fifteenth to nineteenth greatly swollen;
a hyaline membrane extends along the inner side of the
eighteenth, nineteenth, and twentieth joints.

NOTES ON A NEW SPECIES OF GYMNOPLEA. QQ

Female.—Right and left antenne symmetrical, similar to
the left antenna of the male, with twenty-six joints (fig. 5).

Antenne 2.—Similar in male and female. Basipodites
two joints. HEndopodite two joints, the first being shghtly
longer than the terminal joint, provided with two clumps of
five non-plumous bristles. Exopodite with seven joints, the
second and last being the longest; provided with three terminal
bristles and nine non-plumous lateral setz (fig. 6).

Mandibles.—Similar in male and female, the biting blade
being provided with one large anterior tooth and a posterior
serration of seven teeth, of which the anterior one is the
largest. Basipodite two-jointed, provided with three sete
on the inner edge. Exopodite not distinctly jointed off
from the Basipodite; provided with seven large bristles.
Endopodite bi-lobed, the terminal lobe provided with seven
bristles, the lower lobe with four (fig. 7).

Maxilla 1.—Basipodite not obviously divided into two
joints; provided on the inner edge with a clump of nine strong
bristles; on the outer side are two lobes, each provided with
about four bristles. Exopodite and Endopodite not
distinctly jointed off from Basipodite; Exopodite provided
with a clump of nine bristles; Endopodite considerably lobed,
each lobe provided with four or more bristles (fig. 8) (these
bristles are finely plumous; a good light and definition are
necessary to observe this).

Maxilla 2.—Basipodite provided with six lobes on
the inner edge, each with a pair of finely plumous bristles.
Endopodite, constituting the remaining portion of the
appendage, is not obviously jointed off from the Basipodite;
it carries five long, finely plumous bristles (fig. 9).

Maxilliped.—Basipodite consists of two long joints,
the basal proximal joint provided with a prominent keel
anteriorly directed. Endopodite three-jointed; these joints
are provided with small lobes which carry very finely plumous
bristles (fig. 10).

LK

100 ARNOLD W. COOPER.

THORACIC APPENDAGES.

First pair.—First thoracic appendages: Basipodite
two-jointed; Exopodite three-jointed ; first and third joints
provided on the outer edge each with a short seta bearing
four or five short spines; no seta present on the second joint ;
the inner edge of the third joint is provided with six plumous
bristles; the terminal two serrated on the outer edge.
Endopodite two-jointed; sete absent on the outer edge;
the inner edge bears seven long plumous bristles (fig. 11).

Second, third, and fourth pair.—Basipodite two
joints; Hxopodite three joints, each bearing on the outer
edge a swollen seta with spines; on the inner edge seven
plumous bristles and a terminal serrated seta. Endopodite
three joints, bearing along the outer edge and terminaliy
ten plumous bristles (fig. 12).

Fifth pair.—Male asymmetrical, the right appendage
being the larger, consisting of (1) Basipodite with two
joints, (2) Exopodite, two-jointed; the second joint with a
small spine on the outer edge, and long terminal curved claw
without serrations. Hndopodite three - jointed without
bristles. Left: Exopodite not obviously jointed off from
Basipodite, bearing on the inner edge two smooth pads or
cushions; no “appendage” present on the outer edge; two
claws, the outer being the larger. No endopodite has been
observed (fig. 13).

Female.—Right and left symmetrical. Basipodite two-
jointed, Exopodite two-jointed, the second joint bearing a
prominent claw serrated on the inner edge; the place of the
third joint is occupied by a prominent spine (s.), two accessory
spines (a. s.) on the posterior surface. Hndopodite one
jot, with two terminal stout spines (fig. 14).

SYSTEMATIC PosITION.
The characters in which the present species agrees with
Diaptomus are numerous, and it is scarcely necessary to
enumerate them; the characters in which it differs are:

NOTES ON A NEW SPECIES OF GYMNOPLEA. 101

(1) Abdomen of female two-jointed instead of three.

(2) Antenna 1, male, left, twenty-six joints; female, both
twenty-six instead of twenty-five.

(5) Mandible: Exopodite consists of one joint instead of five.

(4) Maxilla 2: Basipodite not obviously jointed, as im
Diaptomus; proximal and distal setae equal in length instead
of unequal.

(5) Thoracic 5, male, right, endopodite consists of three
joints instead of one. Left, exopodite one-jointed, two
cushions with smooth surfaces, and no “appendage.” (In
Diaptomus, exopodite two joints, the two cushions having a
fringe of fine spines, and an ‘“ appendage” is present.)
Endopodite not found, but described in Diaptomus. Female,
third joint of exopodite absent, being replaced by a spine (s.)
(a small third joint present in Diaptomus).

This species agrees with Paradiaptomus in :

(Ll) Abdomen of female two-jointed.

(2) Maxilliped: Endopodite three-jomted instead of five,
as in Diaptomus.
> on the outer
side of end joint of exopodite, present in Diaptomus.

(5) Thoracic five, male, left, “no appendage’

CHARACTER PEcuLIAR TO PRESENT SPECIES.

Antenna 1: Left of male, and right and left of female,
twenty-six joints. This very peculiar character appears to be
due to the division of the second joint. In the great majority
of the Gymnoplea this second joint is longer than the third,
fourth, ete., while in the new species the difference in size is
not observable, and the position of the spine (s,), fig. 5, at the
proximal end instead of in the middle of the third joint,
favours this view—that the additional joint is obtained by
the division of the second joint of a typical antenna, The
hypothesis is supported both by the size of the second joint
and the position of the spine in Diaptomus orientalis (see
HOE OO, G5)

It appears impossible, with these differences, to place this

102 ARNOLD W. COOPER.

species with Diaptomus or Paradiaptomus, and accordingly a
new genus (Adiaptomus) has been made for its reception.

The descriptions of Diaptomus and Paradiaptomus have
been taken from the Das Tierreich, Copepoda, Gymnoplea,
by W. Giesbrecht and D. Schmeil.

I am also indebted to Dr. Ernest Warren (Director of the
Museum) for much valuable assistance with regard to this
paper.

EXPLANATION OF PLATE XII,

Illustrating Mr. Arnold W. Cooper’s ‘Notes on a New
Species of Gymnoplea from Richmond, Natal, South
Africa; Adiaptomus natalensis (gen. et sp. nov.).”

Fie. 1—x 30. Side view of female of Adiaptomus natalensis.

Fie. 2.—~x 40. Abdomen, male, side view; note the rounded pos-
terior edge of the last thoracic segment.

Fie. 3.—x 40. Abdomen, female, ventral view; note the backwardly
directed flange of last thoracic segment.

Fie. 4—x 40. Antenna 1, male. Right, twenty-three joints; twen-
tieth joint double-hinged, fifteenth to nineteenth swollen. Left, twenty-
six joints like Fig. 4.

Fie. 5.—x 40. Antenna 1, female; twenty-six joints, the additional
joint apparently due to the division of the second joint of typical limb.
The spine (s.), which is usually in the middle of the second joint, is
found at the proximal end of the third joint.

Fie. 5a.—x 40. Proximal joints of antenna 1, female of Diap-
tomus orientalis, to show position of spine s., on second joint, and
to compare with the position of spine s., in Fig. 5.

Fic. 6.—x 80. Antenna 2, male.
Fie. 6 a.—Face view of terminal joint of endopodite.

Fie. 7.—x 80. Mandible, right ; exopodite, one joint, not separable
from basipodite.

Fie. 8.—x 80. Maxilla 1. left: note the two outer lobes as in
Diaptomus.

NOTES ON A NEW SPECIES OF GYMNOPLEA. 103
Fie. 9.—x 80. Mazxiila 2, left; endopodite not obviously jointed off
from basipodite.
Fie. 10.—~x 90. Maxilliped, left ; endopodite three-jointed.

Fie. 11.—~x 30. Thoracic leg 1; second joint of exopodite without
spine on outer edge.

Fie. 12.—x 30. Thoracic leg, second to fourth, inclusive.

Fie. 13.—~x 40. Thoracic leg 5, male; right, three-jointed endopo-
dite; left, one-jointed exopodite, two cushions, no “appendage.” no
endopodite found.

Fic. 14.—x 50. Thoracic leg 5, female; right and left symmetrical ;
third joint of exopodite absent and replaced by a spine (s.); two acces-
sory spines on posterior surface of second joint (@. s.).

Ann, Natal G Mus. Vol.1. ge Gal

vulva

— a
aperuLre = = WN
v my
by *

oh Fyaline _%
79§ \-menbrane 78%

bast

10 x 90

cushtons {

. Z ty a)
_ ln --- es
left limb lax } > ; it, 7
f iH fe " :
i 7 § @ i.---enad
right lunb A.

Z
a, fi
4

S

ena ---

20

+)

ms ha

7

CONVOLUTA ROSCOFFENSIS, GRAFF. 105

Note on Convoluta roscoffensis Graff.
collected on the Natal Coast.

By

Ernest Warren, D.Sc.Lond.,
Director of the Natal Government Museum.

With Plate XIII.

Ar Scottsburg, on the Natal Coast, about forty miles south
of Durban, a Convoluta was discovered in September, 1904,
and in May, 1905. It is of a bright green colour and lives
in the sand. The animals craw! on the surface of the sand
grains, and also penetrate between the grains to a depth of a
quarter of an inch or more. ‘They were found in the little pools
on the sand left by the tide, and occurred in such vast
numbers as to give the sand a bright green tint. The con-
voluta has only been found over a distance of a few hundred
yards. I have searched a considerable portion of the coast
both north and south of this spot, but the animal has not
been discovered elsewhere.

I have carefully examined the species, and I have not
been able to separate it from C. roscoffensis. It possesses
the same longitudinal nerve-bands, two eyes, otocyst, and
the same disposition of the generative organs, The only
marked difference appears to be that the mouth is situated
further forwards than is described by v. Graff in the typical
C. roscoffensis. The animal is capable of contracting its
body at the anterior end in such a manner as to form a
sucker-like depression or vestibule around the mouth. This

condition is especially noticeable when the animal has been

106 ERNEST WARREN.

fixed with hot corrosive or Flemming’s solution ; but it may
be observed during life. Very possibly this character is also
possessed by the typical C. roscoffensis.

In the vestibule the ectoderm is only thinly clothed with
cilia, and they are only about half the height of those on the
rest of the ectoderm (fig. 5,cf.,c.v.andc.b.). The general
ectoderm is so closely packed with cilia that they constitute
a definite layer (c.b.), which in section appears as thick or
thicker than the ectoderm cells themselves. The difference
in this respect between the ectoderm of the vestibule and the
general ectoderm is very apparent. The average size of
mature specimens alive is about 4 mm. in length and 0°35 mm.
in breadth. The usual length of the large sagittocysts 1s
about 30u. The animal undoubtedly feeds on diatoms (fig. 2
and fig. 3 f.) and other unicellular organisms, in addition to
any nutritive substances it may obtain from the chlorophyll
bodies.

All stages of growth were collected both in September and
in May. When very young the animal resembles a ciliated
planula (fig. 1) in general appearance, and I could observe
no obvious differentiation of organs or tissues, save into a
ciliated ectoderm and a homogeneous inner mass of cells.
The first sign of differentiation to appear is a row of four or
five cells on each side just behind the oral aperture. These
are the germinal cells which will develop into the ova (figs. 2
and 5,g.¢.). The germinal cells, which will form the scattered
testes, seem to originate at a later period in their definitive
position, and do not appear to arise from the two rows of cells
above mentioned. The ova, during their growth and matura-
tion, are remarkable for their branched and stellate condition
(fig. 3, d.o.). I suggest that this is associated with the fact
that there appear to be no follicle or feeding cells, and con-
sequently the ova themselves obtain their nourishment by
sending ont pseudopodia into the surrounding parenchyma.

The drawings on the accompanying plate have been made
chiefly for the purpose of illustrating the peculiar amoeboid
character of the ova.

CONVOLUTA ROSCOFFENSIS, GRAFF. 107

EXPLANATION OF PLATE XIII,

Illustrating Dr. Ernest Warren’s “Note on Convoluta
roscoffensis Graff; collected on the Natal Coast.”

Fre. 1—x 100. Young specimen resembling a planula.

Fie. 2.—x 100. Somewhat older specimen, showing the formation
of the oral vestibule (0. v.), and the germinal cells (g. ¢.). Some ingested
diatoms can be seen (f.).

Fie, 3—-x 100. Anterior portion of an individual nearly mature.
b. Brain. 6. s. Bursa seminalis. ch. Chitinous tube. ch. b. Chloro-
phyll bodies. d. 0. Developing ova in a remarkably branched condition.
e. Hye. 7. Food. f.0. Frontal organ. g.c. Germinal cells. m. Mouth.
0. Otocyst or statocyst. o.v. Oral vestibule. ¢. Testes. Qa. Female
aperture.

Fie. 4.—x 350. Chitinous tube of bursa seminalis.

Fie. 5.—~x 200. Horizontal section through the anterior end, show-
ing: f.0. Frontal organ. o.v. Oral vestibule, with its scattered short
cilia (¢.v.). g.e. Germinal cells. ch. b. Chlorophyll bodies. ¢.b. Cilia
forming a compact layer on the general surface of the body.

E poi ts <>. ee
nee oe a
a 8 Ue oA : ae
: - ca ae ae owe Pee aS a's 4) a
pate: ae Gis.
Seah oD | fhe a eg

—. te, rw ‘

oF hens fae of

i ae Now :

e

u
= = 4 .
= ;
> . - i
a ute
-. 4 :
: i
* :
| : Q ~
. 7
- i ~
“a =
n
). -
‘ a
> 2 i
* a ;
= - : .
7 r

Pl. XII.
Qa.

iO.

-
©

C.Yv.

j Sees
Oa ot aT
\

O.v
Ne,

~
~
~
XX

Ann. Natal G. Mus. Vol. 1.

NOTE ON THE ABNORMAL HOOFS OF A SHEEP. 109

Note on the Abnormal Hoofs of a Sheep.

By
Ernest Warren, D.Sc.Lond.,
Director of the Natal Government Museum.

With Plate XIV.

In 1905 Miss Shirley Moor, of Greystone, Estcourt, Natal,
sent to the Museum the left foot of a sheep exhibiting a
striking abnormality.

The annexed illustration shows the abnormal foot, and for
comparison a normal foot of a Kafir sheep is shown with it.

With reference to the specimen the donor writes: ‘The
sheep, so far as I can trace back, was a cross between the
Merino and the ordinary Kafir sheep. The hind feet were
the only ones so deformed; the other hind foot was some-
what broken, and was not so regularly formed as that which
I sent you.”

The sheep used all its feet for walking, apparently without
very great inconvenience, but the spirally twisted hoofs
exhibited little sign of wear.

On the occasion of the visit of the British Association to
Natal Dr. S. F. Harmer informed me that such spiral growth
of the hoofs is known in the horse, ox, and perhaps in the
sheep ; also Miss Moor informs me that she has heard of the
same or of a similar abnormality in the hoof of a bull from
the same locality (Estcourt) as that from which the sheep
came. The present specimen is, however, worthy of record
for the following two reasons :

(1) The spirals for the two hoofs both run in the same

vou. 1, part l. 8

110 ERNEST WARREN.

direction (i.e. looking from, above the spirals run from
left to right), and not in opposite directions. It would
be interesting to ascertain if such is generally the case
in abnormalities of this nature, since the shght spiral
tendencies in the normal hoofs of a foot are in opposite
directions (see figure). It is possible that the fact of the
spirals running in the same direction for the two abnormal
hoofs may be due to the essentially asymmetrical nature
of the two hoofs of a ruminant, as they are the third and
fourth digits of the typical pentadactyle limb.

On inquiry Miss Moor further informs me that im the case
of the right foot both of the twisted hoofs grew in the same
spiral, but that this spiral ran in an opposite direction (i.e.
from right to left) to that of the left foot.

(2) The rudimentary hoofs of the fetlock show a distinct
tendency for abnormal growth (see figure) ; thus the causes
which produced the abnormality were effective over the
whole of the foot, and the abnormality was not due to a
mere local disturbance in the growing bases of the hoofs
of the third and fourth digits.

HXPLANATION OF PLATE XIV,

Dlustrating Dr. Ernest Warren’s paper on “ Note on the
Abnormal Hoofs of a Sheep.”
Fie. 1.—Plantar surface of normal left hind foot of Kafir sheep.
(Natural size.)

Fic. 2.-—Plantar surface of abnormal left hind foot of supposed
cross hetween Merino and Kafir sheep. (Natural size.)

Pia Scie

(1)

Ann. Natat G. Mus., Vor. I, Parr 1.

Adlard & Son, Impr.

~~
‘
“ a
ae,
a @
.' q re
m!
a
0 ae @
~
&
a ,
Ms
I
1
i
0
* i yy
git
i
5

=- te.

})

Vol. I, part 2, issued March Sth, 1907.

CHARACTERISTICS OF LARV# OF ANOPHELINA. ele

A Contribution to the Study of the Character-
istics of Larve of Species of Anophelina in
South Africa.

By
Ernest Hill, D.P.H.Camb.,
and

L. G. Haydon, M.B., C.M., D.P.H.Aberd.

With Plates XV—XX VI.

PREFATORY NOTE.

WHEN this paper had been completed and was about to be sent
to press a letter was received from Colonel Giles, who had with his
usual kindness examined specimens of the species which we have
throughout called Myzomyia funesta, and states that they are
of Pyretophorus pitchfordi Power. Colonel Giles, however,
admits that the difference between the palpal banding of our
specimens and those sent him by Mr. Power is so marked that his
description of the latter is hardly appropriate to the former. We
have the greatest respect for the opinion of Colonel Giles on such a
matter, but this species which we have called funesta, as repre-
sented in our collection, appears to us to differ, both in respect of
the markings of wings and palpi, quite as much from Guiles’
description of pitchfordi as it does from Theobald’s description
of funesta, while we should have had no hesitation in describing
wing scales as mostly long and narrow, which is the chief point,
as we understand it, by which Myzomyia is distinguished by
Theobald from Pyretophorus, in which the scales are described
as lanceolate. Further, some while back, specimens were sent to
Mr. Power, who first despatched specimens of pitchfordi to
Giles, and he expressed himself as unable to determine them, with-

yous); Parr’ 2, 9

a2, ERNEST HILL AND L. G. HAYDON.

out reference to works on the subject, whereas it might surely be
anticipated that he would recognise a species first determined as
new by himself.

In the circumstances it seemed advisable to let the name
funesta stand in the paper with this note, indicating the doubts,
rather than to alter the naming, seeing that in the present confused
condition of the subject of Anophelina it might be within range
of possibility that pitchfordi might later on rank as a variety of
funesta or other species.

PART I—Inrropvcrion.

Fifteen species of the sub-family Anophelina have, to the
present, been found in the territory of the Colony of Natal.
These, grouped according to genera defined by Theobald in
‘A Monograph of the Culicidee of the World, are as follows :

Genus. Species.
Myzomyia . . funesta Giles.*
= ; . rhodesiensis Theobald.
Pyretophorus . . costalis Loew.*
a , . cinereus Theobald.*
‘p : - pirtchtord1 Powers
re : . superpictus Grassi.
5 . . ardensis Power.*?
pe : . marshalli Theobald.*
Nyssorhynchus . . maculipalpis Giles.
33 pretoriensis Theobald.*
Cellia . : . . squamosa Theobald.*
i ; ; ; . jacobi. New species.*?
Myzorhynchus. . . paludis Theobald.*
i: : . mauritianus Grandpre.*
a3 , . natalensis. New species.**

Those marked with an asterisk are represented in our collec-
tion, in many cases by a considerable number of specimens.
The species called jacobi and natalensis respectively

1) First recognised by Mr. H. 8. Power, and described by Col. Giles in his
revision of ‘Gnats and Mosquitoes,’ 1905.

* First recognised by Mr. H. 8. Power, of Natal; examined by Theobald.

%’ Named by the authors of this paper.

CHARACTERISTICS OF LARVA OF ANOPHELINA. 113

appear to conform quite satisfactorily to Theobald’s
definition of the respective genera, but are not identified
with any described species.

In a contribution to the ‘Journal of Hygiene’

by us
(vol. v, No. 4) on the subject of an epidemic of malarial
fever, statements were made as to relative prevalence and
habits of certain species of Anophelina. The determina-
tion of species was, in most instances, made for us. We
have since given attention to the matter, and have found it
necessary to revise the determination of three species. There
may thus be discrepancy between portions of that paper and
of this. For the present determination we are solely re-
sponsible; photographic illustrations of the wing and palpi
of each species herein noticed are put in, by which the
determination may be checked.

The original intention of this paper was to contribute to
the study of Anophelina by a description of the character-
istics of the larvee of some African species, as to which very
little, and that little very meagre, has been published. In
collecting and sorting material, it became increasingly borne
in upon us that Theobald’s grouping into genera is by no
means in all respects satisfactory, and that, admirable as are
his description of imagines, and notwithstanding that in the
introductory pages he briefly mentions that markings are not
constant, yet there is so much rigidity both in the definitions
of the size, and of the wing and leg markings, that great
difficulty confronts the ordinary collector, in attempting to
assign specimens to their proper species. This is a matter of
importance in the case of disease-carrying insects.

Generic distinctions are largely based on the shape of
scales. On this point we take the hberty of quoting from
‘A Monograph of the Anopheles Mosquitoes of India, by
James and Liston, a work to which we are indebted for much
assistance. On page 20 we find objections laid against
present generic grouping as follows:

(2) The classification is based in great part on the shape
and not on the presence or absence of scales. Scales of

114 ERNES! HILL AND L. G. HAYDON.

various shapes are present on different parts of the thorax,
abdomen, and especially the wings of ‘ Anopheles,’ and it is
a matter of great difficulty to decide in some instances what
form of scale predominates; nor does Mr. Theobald give us
any indication of what portion of a wing should be examined
to decide this point.

“ (3) The terms ‘lanceolate, ‘long and narrow,’ ‘true
scales,’ etc., are not sufficiently definite to permit of such
scales being easily distinguished from one another. . . . It
is obvious that the distinction between ‘ hair-like curved

‘narrow curved scales’ is not great, and also

scales’? and
that it would be difficult to decide whether the abdomen
is ‘hairy,’ or whether it is covered with ‘ hair-like scales,’
which apparently resemble hairs so closely that they cannot
be termed ‘true scales.’ As regards the wing again, it would
certainly be difficult to decide whether most of the scales
were ‘lanceolate,’ or whether they were ‘mostly long and
narrow’ especially as the part of the wing to be examined is
not stated, but on this decision alone depends the distinction
between the genera Anopheles and Myzomyia.

“ (4) One of the objects of classification is to simplify the
identification of species, but the new classification does not
aid this in any way. In practice, it will be found much
easier to determine the specific name of any specimen of
‘ Anopheles’ than its generic name, according to the new
system.”’

With this criticism we are for the most part in accord, but
some genera, particularly Myzorhynchus and Cellia,
appear to be sufficiently distinct ; Nyssorhynchus, how-
ever, touches closely on the latter, and both of the genera
are relatively uncommon.

A genus should possess one or more characteristics which
are prominent, and by which it can be readily and certainly
distinguished from other genera, and this characteristic or
these characteristics should be, at any rate in broad terms,
common to all species in that genus, and shared by none in
any other. A classification by genera which fails in these

CHARACTERISTICS OF LARVE OF ANOPHELINA. 115

requirements has neither scientific basis nor practical
advantage. The requirements are not complied with
throughout Theobald’s grouping, which is based also entirely
on imagines.

If the points of difference between the larvee of different
species of Anophelina generally were slight and unim-
portant, there might be sound reason for generic groupings
on the features of imagines alone, but the differences are
certainly as great and quite as definite as those between the
different imagines. It is therefore desirable that the larval
characteristics be given an equal weight with those of the fly.
In Theobald’s ‘ Monograph of the Culicidee of the World’
(vol. 11, p. 10) is found a table of genera with characteristics
which are here abbreviated :

Anopheles . . Thorax and abdomen with Wing scales lanceo-
hair-like curved scales late.
Myzomyia . .. Ditto Wing scales mostly
long and narrow.
Pyretophorus . Thoraxwithnarrowcurved Wing scales small,
scales ; abdomen hairy lanceolate, or nar-
rowed.
Nyssorhynehus. Thorax and abdomen with Wing scales bluntly
true scales ; lateral tufts lanceolate.

and dorsal patches of
small flat scales

Cellia . : . Thorax and abdomen true Ditto.
scales. Abdomen nearly
completely scaled with
lateral tufts

Myzorhynehus . Thorax with hair-like Wing scales broadly
curved scales. Abdo- lanceolate.
minal scales on venter
only, with distinct apical
tuft; no lateral tufts

Of the last two genera the characteristics appear to be
definite and distinctive, but the features of the first three are
very variable in practice, and such as to render distinction
very difficult. Nyssorhynchus, as far as can be judged
from description of different species, merges in Cellia on the

see. ERNEST HILL AND L. G. HAYDON.

one hand, and the group Anopheles, Myzomyia, and
Pyretophorus on the other, for in the species stephen sl
(vol. iii, p. 93) the abdomen is described as covered with
seales, while maculipalpis (vol. iii, p. 97) is credited with
a very few scales on the apical segment only (apparently a
characteristic also of Myzorhynchus), while in pretori-
ensis we can detect but some seven or eight in all on the last
segment. On page 111, vol. iti, albipes is called both
Cellia and Nyssorhynchus.

‘The shape of wing scales, as described, might be thought
sufficiently to distinguish the latter from Myzomyia, but
Theobald expresses doubt as to whether the species elegans
is not of the genus Nyssorhynchus rather than of
Myzomyia, to which he has allotted it, and says that it is
very near to stephensi, a species of which the abdomen is
thickly scaled (vol. ii, p. 53). Again, the descriptive shp on
Plate V in the same volume indicates that pretoriensis
barely escaped from the genus Pyretophorus.

From the table in ‘ Monograph of the Anopheles Mosquitoes
of India,’ and from illustrations and descriptions in ‘ Mono-
graph of the Culicide of the World, a table of larval
characteristics has been compiled to which are added in their
place the nine species identified by us :

|
Frontal hairs. Antenna Palmate
shaft. hairs.
Vv — OE OOOO | OO
External.
Genus. Species. : co) a
ao | S = 5 i)
Go.) : rm Re icctan | Micy anomie as
or | A PS ay LS ea sete eee
ao) = Pes res an a | @ loo |
Sia|) asst a eesl e g pa oe
ag] 4 & EP petals) Si ea
BRE = 2 oO Heal ola tA os
fon| < a a |A9 (48 & |Ae
———— — a = — —— =|
Anopheles «. - | macwilapenmisg) 4) ye |
JOmarerOryntss @)) i) ef aug Poly? lee
lindesayi — | + |—}— ] + | — | BP ?
| aitkeni. . .| +|)— >) —!]—jJ—j|+4+{]—)/ 4 |
Myzomyia . .|rossi’. .. .| =| >) —4 =) =
eculicifacies .|— | + | —}]—{—}+]4+)—
elegans: 282 ee ee
turkhudi ra eg ena Veen sme SF Se
funesta. . .) + |—/]|—/]—j;—]+)+/]—]
ethoOmilt 5 . 6 |) ==) ee jes | — | se |] es) =],

CHARACTERISTICS OF LARVZ OF ANOPHELINA. 117

Frontal hairs. | Antenna | Palmate

= A

a

| External. | |
—_— |

Genus. Species

No branched

hair.

Branched,
some or all.
All plain.
Dendriform.
Penniform.
Branched
Filamentous.
No definite
filaments.

Pyretophorus . jeyporensis
superpictus
cinereus
ardensis,
costalis .
Nyssorhynchus maculipalpis
maculatus.
fuliginosus .|
theobaldi .
/stephensi .
karwari.
annulipes .
pretoriensis .|
Celintan a eed|elbipes!
| pulcherrima.
squamosa
jacobi
Myzorhynchus_ barbirostis
‘nigerrimus
sinensis.
paludis.
natalensis.

eels

+{/+[++++]]]]] | +] +++] 4+

+{[ [| | terete ttertttertrett

+] [|] | t++¢4¢ eet | toe tte [cet
[++~ee] (itt eelttet lle

Pele lil lteeree lel |
SSS
|

Theobald writes (vol. im, p. 10): “It will be noticed that
by these characters (1. e. his generic features) a natural
grouping is formed, and that it, in the main, tallies with what
we know of grouping them by their larval structure. I do
not think that the minute structural differences in the larvee
should be taken as of greater value than specific characters ;
but it is of interest to find that classification by certain larval
and certain adult characters give the same result.” The
meaning of the last sentence is not quite clear, but from the
context generally it would appear to indicate that the generic
classification by scale structure is supported by the genera
having common larval characteristics. This must have been

* Either dendriform or penniform.

118 ERNEST HILL AND L. G. HAYDON.

written with a then limited acquaintance with larvee, for if,
from the above table of thirty-two species, we endeavour to
pick out certain characteristics forming natural groups cor-
responding with Theobald’s generic classification the difficulties
are even greater. Considering, in the first instance, the
antenne we see that a multibranched hair is found starting
from the intero-anterior aspect of five, its presence or absence
not specified in two, and that it is definitely absent from
twenty-five. Of the five, two belong to genus Anopheles,
three to Myzorhynchus. Even if it were possible to find
affinities between these two species of Anopheles and the
genus Myzorhynchus, which would justify their inclusion
in the latter genus, the difficulty would not be surmounted,
because the prominent dendriform external frontal hair
(Pl. XXII, fig. d) forms at least an equally distinctive
feature, and is found in Anopheles maculipennis, but
not in lindesayi,and again is absent from natalensis. In
six species of the six genera represented in the table, the
external frontal hair is dendriform (Pls. XX and XXII d).
One of these is an Anopheles, one a Cellia, and four
belong to genus Myzorhynchus; but of all the genera that
which has the most pronounced characteristics in the imago
is Cellia. Again, we find that of four mounted specimens of
Cellia squamosa, which we have kept, the external frontal
hair is penniform in one, dendriform in three. There can be
no doubt of the identity of the former, seeing that it differs
in no other shade from the latter. The distinction between
the two types is barely appreciable with a lower magnification
than x 100, and might readily be overlooked.

The most striking feature which we have seen in any larva
isa large dendriform hair found on the first abdominal segment
of a species of Myzorhynchus, which we take, despite the
difference as to the amount of white in the second hind tarsus
of imago, to be Theobald’s paludis (Pl. XXII, fig. f).
This hair certainly engendered expectation of its being of
some specific value, but we are forced to regard it as merely
an individual peculiarity. We have only seen it in three

CHARACTERISTICS OF LARVH® OF ANOPHELINA. 119

instances. On the first occasion it was observed in a living
larva. Six others collected at the same time were closely
examined, but no such hair was detected. It is translucent,
and overlies the palmate hair in a different plane, and in
hurried examination of a lively larva might readily be over-
looked, unless the observer were especially alert for it, when
it cannot be missed. Subsequently, in looking through some
mounted specimens, this hair was found in a second instance.
A large collection was made from the pool from which the
latter had been taken some months earlier. Thirty individuals
were carefully examined, and a similar hair was found in one
only. Both larvee developed into imagines. No difference was
observed between the markings of these two and of others,
greater than between any of the combinations given on page
149. Both proved to be males, but male insects were obtained
from larve on which this dendriform hair was not present.

The principle of grouping into a genus different species, of
which the points of resemblance in imago are no stronger
than the features of difference in larva is unsound, but it
must be readily admitted that any attempt to give equal con-
sideration to larval characters would make all attempts at
generic Classification hopeless.

James and Liston propose to divide Anophelina into
groups, taking into consideration the characters of egg, larva
and imago, together with habitat, habits and pathological
aflinities, particularly the last. This is doubtless satisfactory
for a country in which these particulars are known, but may
be misleading if applied to the sub-family generally. In
such an arrangement the South African species funesta,
ardensis, cinereus, pretoriensis andnatalensis would
be grouped with the malaria-carriers listoni (or christo-
phersi, according to Giles) and culicifacies, but costalis,
as the authors themselves state, with stephensi and rossi.
Funesta is a known malaria-carrier, but there is no evidence
against the other four, whereas costalis, although placed
with rossi, which apparently does not act as host to the
malaria parasite in natural conditions, is, as we have shown

120 ERNEST HILL AND L. G. HAYDON.

elsewhere, probably the premier agent in conveying the
disease in Natal. Further, until all details of an Ano-
phelina were ascertained, it would need go nameless, in
regard to group.

It is, on the whole, unfortunate that generic grouping on
narrow and refined differences was ever attempted. It
appears to us that the need is for a better representation
of specific characters. Descriptions which will adequately
convey the writers’ meaning to the reader are difficult to
draw up. Such terms as “spot” “ patch,” “small spot,”
etc., are too vague to give sound guidance. There is also
much variation in the markings of the wings of individuals
of a species. A photograph of a wing showing clearly all
markings with a tabular statement, giving variations met
with in a given number of species, is a much greater aid in
making determinations. This method we have adopted to
the extent admitted by the number of specimens of a species
in our possession.

Some points to which specific importance has been
attached in Theobald’s ‘ Monograph’ cannot be admitted to
have the value attributed to them. Stress is laid in descrip-
tions of Myzorhynchus paludis and mauritianus on
the proportion ef the second hind tarsus which is white, and
the presence or absence of a wing spot, as distinctive points.
As will be seen from particulars given in the systematic
portion of this paper, the wing spots vary much in size to the
point of total extinction, and the proportion of white m the
second hind tarsus from seven-eighths to one-fourth, but the
variations in the wing spots do not coincide with the amount
of white in the tarsus, nor have we been able to make out
any corresponding difference in the larve.

In our collection we have a single specimen of a female
imago, in which there is marked difference in the two wings.
The insect belongs to a species, which more nearly corre-
sponds to Theobald’s Myzomyia funesta than to any other
of his species, although of a larger size. On one wing, the
middle of the three main white or yellow spots, described as

CHARACTERISTICS OF LARVZ OF ANOPHELINA. IPA

marking costa and first longitudinal vein, is entirely absent,
on the other it is present. Some of Theobald’s species are
based on a single female. If this specimen had been defec-
tive in this spot on both wings, it would appear at least
possible for it to found a new species. There is, however, no
reason why such a variation should not occur on both equally
with one wing.

The Nyssorhynehus pretoriensis of Theobald is de-
scribed as 5 to 5°5 mm. long. This would make it, according
to him, to be of the same size as Myzorhynchus paludis.
In our collection, however, we find that the specimens of a
species, which accords in all important particulars with
Theobald’s pretoriensis, are very much smaller; while the
frontal hairs of the larvee correspond with Theobald’s state-
ment. We think it scarcely credible that two mosquitoes
from places near by on the same continent so very nearly
alike can belong to totally different species.

Individuals of the same species are subject to much varia-
tion in size, and in the pattern on the wings (although the
main costal spots are mostly constant) and the relative
breadth of the bands on the palpi. The most uniform of
those which are dealt with in this paper is Pyretophorus
costalis, the most variable Myzomyiafunesta. General
appearance is sometimes more helpful to distinction than
detailed description of markings. Theobald, in describing
Pyretophorus cinereus, says that at first sight it looks
hike a large funesta. We have a number of specimens
which we do not doubt are large funesta. This for a time
we took to be cinereus, although Theobald describes the
latter as having four white bands on the palpi against three
infunesta. He says elsewhere (vol. ii, p. 5) that specific
value cannot be attached to the palpal bandings. In some of
our specimens a few dark scales are visible in the middle of
the last segment, which in two is certainly divided by a dark
ring into two white bands. ‘The larva of this larger variety,
however, differs in no way, except in size, from that which
results in funesta of the described size. On the other hand,

122 ERNEST HILL AND L. G. HAYDON.

we have an imago, herein referred to as Pyretophorus
cinereus, which is much larger than the largest funesta,
and by its naked-eye appearance is obviously different, much
more than can be shown by any detailed description (Pl. XXV,
fig. c). The larva of this comes near in resemblance to
that of funesta, but the position of palmate hairs and the
shape at termination are different (Pls. XV, XVII, fig. B).

Variations in size are to some extent due to external cir-
cumstances. Nearly all our specimens were obtained from
larve kept in basins until the insects emerged from pupal
stage. Those which were young and small when captured,
more particularly in cool weather, when larval existence was
prolonged to seven weeks, developed generally into smaller
insects. Their growth was stunted, and they suffered from
debility, as a result of unnatural conditions. It was observed
that after some weeks’ captivity the metamorphosis into pupa
took place, when the larva appeared still some way short of
full grown; the pupa was smaller, the imago also, which
experienced much difticulty in getting free of pupa skin,
often losing hind legs, and not seldom life, in the process.
This did not happen to larvee which were full-grown at time
of capture, unless shaken up in travelling.

We have identified the larva of nine of the fifteen species
enumerated on page 112. ‘The most certain method of
identifying larvee of a species is to breed from the imago and
watch the larva from egg to final metamorphosis. This
proved impracticable, because conjugation did not take place
in captivity, and impregnated females were not captured when
required. Larvee were collected, examined alive, sorted into
groups, watched to metamorphosis, and specimens preserved
from time to time, and last larval moults mounted. With a
little practice familiarity is gained and distinction between
different species readily appreciated. The characters of each
species are sufficiently defined to leave little room for error if
sufficient care is taken.

The points of importance for differentiation of larve have

CHARACTERISTICS OF LARVA OF ANOPHELINA. 123

been set forth by Christophers and Stephens, and we do not
find specific value in any additional character.

(1) Antenna.—(a) Presence or absence of a branched
hair on the shaft. This hair is found on the antero-internal
aspect. There is in some species a spikelet on the outer side,

which is not constant in any species in our list, except possibly
Cellia jacobi; (b) number of branches or divisions of the
hair between the terminal spines.

The former is of value in that the larve in which it is
found form a numerically small group ; the latter may be used
as a supplementary point in final distinction between similar
larve of different species. For instance, ardensis and
natalensis come in some respects near together (Pls.
XVIII, XXIII, XXIV), but in ardensis the hair divides
into two, in natalensis into six or eight. The number,
however, is not constant. In most specimens of funesta
three divisions were observed, but in one half-grown larva
there were four, which isthe usual number in costalis. The
relative size, one to the other, of the two spines is not, in our
experience, of use—in fact, we find them practically equal in
all our species, variation being no greater than is found in
the two sides of a lobster’s claw, which they much resemble.

James and Liston lay stress on the importance of a branched
hair outside and parallel to the antenna, which they call
the “basal hair.” We do not feel quite sure what they mean
by this, because they make no mention of it in description of
individual species, except in the case of culiciformis, while
it is figured in plates of jeyporensis, culicifacies, and
aitkeni only in addition. We find a similar hair in all
species. It arises from a point outside the antenna, and on a
slightly lower plane only, and therefore comes into view under
the microscope with the antenna, and is shown accordingly in
the plates. This is present on the head of paludis, but im
all our specimens arises on a much lower plane than the
antenna. It is necessary in most instances to focus especially
for it; it cannot conveniently be drawn in relation to the
dorsal aspect of the larva, and is omitted on that account

124, ERNEST HILL AND U. G. HAYDON.

from the figure. If this be the ‘ basal hair” it is not of
differential import in our species.

(2) Frontal or Clypeal hairs.—In all larve examined
by us three pairs are found, two near to the anterior ex-
tremity of the head, and a third a little behind the median
pair. The posterior may be very small and require relatively
high magnification to detect in the first instance (Pl. XXII,
@,, 4). The frontal hairs are of prime specific importance,
and it is generally necessary to consider the three pairs
together. They exhibit much individual variation, especially
the posterior (vide Pls. XV—XXIII, fig. a). In funesta
they may be quite free of branches, while in pretoriensis
they invariably are. The lateral hairs also vary, though to
less extent (Pls. XVI, XVIII, XXI, XXIII); but it is to be
observed that in natalensis (Pl. XXIII, fig. a) on one
side is found a fine branched hair, on the opposite a straight
smooth bristle. Attention is needed both to structure and
to relative size of posterior to anterior median, which, for
instance, is much less in pretoriensis than in funesta or
cinereus. Relative position of point of origin is not avail-
able for comparison, because of the impracticability of
arranging a preserved specimen precisely flat in such a
manner that exact measurements can be made; it is not
possible to secure uniformity of plane without crushing the
larva, which is thereby distorted and the hairs frequently
broken and detached.

Thus it happens that in drawings of larvee the proportions,
except of palmate hairs, are approximate, although a micro-
meter scale was scrupulously applied to every detail as far as
circumstances admitted. This more particularly apples to
instances in which one or two specimens only were available
for drawing. In these instances, however, it so happened
that the anterior frontal hairs le in an uniform plane, and it
is mainly the segments of the bodies as to which there is a
margin of inaccuracy. Where several specimens are available
greater accuracy is attained. Palmate hairs can be detached
and mounted absolutely flat, and error entirely avoided.

CHARACTERISTICS OF LARVA OF ANOPHELINA. 125

(5) Thorax.—A difference in hair ornamentation as be-
tween Myzorhynchus paludis and others is observable,
but we hesitate to assign significance to this.

(4) Palmate hairs.—These are of decided specific im-
portance in regard to: (a) relative size; (b) relation of
filament to leaflet; (c) character of filament and of
“shoulder” (the term “ shoulder” is adopted from Christo-
phers and Stephens as meaning the abrupt widening at the
point where the filament runs into the leaflet—vide PI.
XXIV); (d) position in which palmate hairs are found.

A palmate hair may be found on the thorax and all the first

seven abdominal segments, or it may be absent from thorax,
first and occasionally second abdominal segments. Great
care in examination is necessary before concluding the
absence entirely from any segment, because on thorax it is
always, and on the first abdominal generally, rudimentary,
and in instances close scrutiny with a magnification of x 250
or so is required for.certainty where the hair is closely folded
in death (cf. Pl. XXI, fig.c, 1). If the larva be examined
alive the rudimentary hairs are found expanded, as the fully
developed, but they are generally quite translucent; the
leaflets, smooth and free from shoulder in most species, do not
terminate in any filament, but in smooth, narrow lanceolate
extremity (Pl. XXI, fig. c, 2). They appear, however, to
be functionally active (vide Pl. XXIV, fig. J, first and
second abdominal segments).

It is said that these hairs aid the larva in maintaining its
horizontal position, and that the lone plumose lateral hairs
act as balancers. They are, however, lable to be broken or
torn off, and the larva appears little, if at all, iacommoded by
their loss. On one occasion a larva of species paludis (in
which the palmate hairs are pigmented and easily discerned)
was watched for a considerable time ; there were no balancers
on the thorax, two broken stumps on the first abdominal
segments on one side, and none on the other, nor on the
second on either side ; palmate hairs lost from fourth, fifth, and
sixth abdominal segments on one side, and from third and

126 ERNEST HILL AND L. G. HAYDON.

fifth on the opposite, but the larva behaved in all respects,
and assumed a similar attitude to others, in which the hairs
were complete.

Relative length, in the average, of filament to leaflet is of
general value in determination of species, although the rela-
tion differs in individual leaflets and individual specimens,
but im some species, as funesta, it is more inconstant than
in others. This species, which exhibits the least constancy in
larval features, is also the most variable in the wing pattern
of imago.

In our hmited experience we find that, although larvee—
and some species more than others—are subject to much
individual variation in some essential points, yet the sum total
of the characters is sufficiently pronounced to enable the
identity of any one of the nine species which we present to
be readily recognised, even in the living larva.

We have not followed up larvee systematically from youth
to maturity, observing any changes of character in important
features, but from a limited number of measurements it
appears that i very young larve the frontal and palmate
hairs, though relatively large, are absolutely small, whereas,
when about half-growth is attained, both are in absolute
measurement about the size found in the mature larva. In
very young larve the characteristic shape of the palmate
leaflets, as a rule, is not defined, but in larve which have
passed through but a quarter of their career the features are
quite distinct.

Duration of life.—The length of time occupied in each
stage-varies with the temperature. In warm summer weather,
mean daily temperature about 73° F., very tiny larve will
arrive at pupation in twelve to fourteen days, the imago
emerging in a further period of two; but im autumn and
winter, mean daily temperature about 56° F., the larval stage
is prolonged to seven weeks, and the pupal existence to five
days. We have observed larve certainly not one third of
average mature length as long as six weeks after collection,
when certainly no insect has had access to the water.

CHARACTERISTICS OF LARVE OF ANOPHELINA. 127

PART I[J—Systematic.

Myzomyia funesta Giles.

The following description of certain features of the female
is given by Theobald (vol. 1, p. 178) :

“ Palpi black, with a white apex and two white rings, the
one nearest the apex sometimes involving both sides of the
last two joints.

“Legs dark brown to nearly black, with a few pale apical
scales to the metatarsi and tarsi, often indistinct unless seen
under the microscope.

“ Wings with the black costa with six pale creamy almost
white spots, the three apical ones extending on to the first
long vein, remaining veins with patches of white and black
scales as follows: One small and one large dusky patch on
each of the branches of the first submarginal cell, the greater
part of its stalk dark scaled, a dark patch at the base and
apex of the third long vein, two dark patches on each branch
of the second fork-cell, and the greater part of the root and
stem dusky, two dark patches on the upper and one on the
lower branch of the fifth vei, another at the fork, and
another at the base of the vein, two small dusky patches on
the sixth. . . . Fringe black, with pale spots at the junc-
tions of all the veins, except the sixth; apex mostly yellow,
but with a black spot between the two branches of the first
submarginal cell. Length, 3 to 3°5 mm.”

In vol. ii, p. 34, two varieties are described: “ um brosa—
costa black at base, unbroken by the typical small pale spot.
Veins with dusky scales predominating, the pale scaled areas
restricted to the region of the cross veins, and bases of the
fork-cells, and on the fifth long vein; the third long vein
dark, as in rhodesiensis. Wing fringe spotted as in the
type. Subumbrosa—costa black at the base, unbroken by
any pale spot. Dusky scales predominating, but not con-
trasted, as in the type, with the pale scaled areas. Third
long vein pale scaled in the middle, and pale scaled
areas also on the fourth, fifth, and sixth.”

VO. l, PART 2. 10

128 ERNEST HILL AND L. G. HAYDON.

This species has been found in Natal, from sea level to an
elevation of 4000 ft., above which we have not made search,
and the larve have been taken, and imagines obtained from
them throughout the year. It is the commonest and most
widely distributed of Anophelina.

Two definite varieties are found: a smaller, of which the
average measurement of detached wing is 3 mm., a larger
of average measurement 3°8 mm. Intermediate sizes are also
observed. Dark individuals, corresponding more or less with
umbrosa and subumbrosa, are very occasionally en-
countered, the larve being captured among the ordinary
type. Insects of greater and lesser measurement are about
equally common on higher grounds, as at Maritzburg, 2200 ft.,
but in our present collection of some fifty females we have
none of the larger variety from coast districts.

The width of the bands on the palpi vary considerably, and
the apical band is rarely as narrow as figured in ‘ Monograph
Culicide,’ vol. 11, p. 36, and on that account 1s inadmissible
as a specific distinction.

Theobald’s description of the wing does not tally with his
diagram on page 386, in which a lengthening of the second
yellow spot on the first long vein is shown, which is not men-
tioned in the text. In the species as found in Natal, this
feature is always present, generally as shown on Plate XXV,
fig. A, less frequently as a distimet spot, separated by a few,
or several, black scales from the spot beneath that on the
costa. The figure is representative of the markings of both
varieties, but considerable variation is found, as shown in the
table below, in which we have found it more convenient to
specify “ pale” spots rather than “ dusky ” spots, seeing that
black scales predominate in the second and fourth longitudinal
veins. ‘I'he three main white or cream spots on costa are fairly
constant, the apical is always present, but varies in size,
while at the base there is one spot only in 18 per cent. of our
specimens ; there is, in all, a pale spot on the shaft of the
second and fourth longitudinal veins, though it may be very
narrow, and, except in the dark varieties, there is no im-

CHARACTERISTICS OF LARVE OF ANOPHELINA. 129

portant deviation in the third, fifth, or sixth. In all we find
a pale spot at the point of bifurcation of the second and fourth
veins.

Table showing important variations in pale spots on palpi and
wines of female Myzomyia funesta.

ay
| fe | Presence or absence of Number in |
| Number of | spots! on : which spot |
| Spleselereioualss Be occurs on |
| Na hich pal.) ————— fringe oppo-|
| bands are: | "site veins < |
| ; | | 1st fork cell. | 2nd fork cell. | ~ cs
| Number of specimens. | |
| = 77, | )
Fc : H B ea B |
galeala| as a2 | 45 ao yrs
|
: —_ 2 | | eee
Theobald’s type . —|—|+ + + + Sa lee ec |
|
| | |
SS 2ales. |) act =) gt ua 3} 4] 1
| |
Large mosquitoes /4 |— 1,3] — + + ==) A SSO Ob eo
_ from upland dis- Ves ie Ne
tricts—high levels }2 |—| 2;—| — — se) | Gobo ihe Os ee
| Ce SS ay TE ey ee er | —
A | |
cha) | Me Se = + ap aa | 24
6: 2512 |"2e eee — | + — ji ae ea
Small mosquitoes : | |
10 from upland |}2 |—)\/—) 2 -- + + a eae C8) aro
| districts ; | | | |
_ 16 from coast 2 |—|2\/—| — == (8 eee 9 emt ete al
level. | |
me es — | + = || al
| |
|
6 eee OM |e = eel aes Wd |) hy), 2
| | | |
Neat | | |
SS —— — aaera| leer | ea —
i+ +—/+ —|/+—| | |
42 | 8 |21/18/21 21| 6 36/34 8 | 21 21) 4 | 26/12)

42 4.2 42 4.2 4.2 4.2

' The minute spot always present just at the point where the vein runs
into the fringe is not included. ‘The presence of a pale spot indicates that
there are two “dusky” spots; its absence that the whole branch is dusky
except at its immediate commencement as such and at its termination.

130 ERNEST HILL AND L. G. HAYDON.

The Larva (Pl. XV).—The species was determined from
over thirty specimens which were examined and identified
alive, and all subsequently emerged as the same species. The
drawings were made from a comparison of twelve specimens.

There are no structural differences between the larva of
the large and small variety, but there is an appreciable
difference in size. It is small, short, and appears rather
broad for its length. When about one third of growth is
attained, there is generally observable a brown central strip,
while the sides are translucent; this appearance is retained
to the time of pupation, if the larva is reared in captivity,
but larvee of the small variety, when full-grown or nearly so
at time of collection, are commonly deeply pigmented, fre-
quently quite black, whereas the larger are pale hke younger
specimens. This is probably attributable to character of food.

Antenna.—No branched hair on shaft, terminal spines
equal, and between them a hair, which divides into three
branches, generally, though four have been observed.

Frontal hairs.—Three pairs. Anterior constant in all
specimens examined; posterior pair long and prominent, and
subject to much variation (Pl. XV, fig.a). In one instance
in a live larva four branches were observed on one hair.

Palmate hairs.—On thorax and first seven abdominal
segments ; on thorax of good size but rudimentary in shape
(fig. c); on first abdominal intermediate ; on other segments
well developed and defined (fig. b, and Pl. XXIV, fig. a).
Radius, 0°127 mm. to 0138 mm.; relation of length of
filament to total length of leaflet, including filament, fairly
constant in any particular batch, varies considerably in
different batches; extreme average of leaflets of individual
specimens as 0°28 to 0°38 to one, and extremes of individual
leaflets as from 0°22 to 0°45 to one. The larva is, never-
theless, readily distinguished from others described in this
paper, by the general character of palmate hairs, position i
which they occur, and above all by the features of posterior
frontal; both are quite distinct from those of Nyssor-
hynchus pretoriensis (Pl. XIX, figs. a, b), and the former

CHARACTERISTICS OF LARVH OF ANOPHELINA. 181

from Pyretophorus cinereus (Pl. XVII, fig. a) which
it most nearly resembles.

James and Liston express an opinion that this species is
one with listoni.! They represent the latter as having
no posterior hairs. This would appear to negative such
assumption.

Habitat.—Elective spots for larve; under overhanging
banks of brisk streams, among bits of thick grass, etc., in
eddies of similar streams, but also among grass and rushes,
and in small spring waters. Not found in puddles or stagnant
waters.

Season.—Perennial. In October numbers were collected
from a tiny roadside spring-pool among dead bamboo leaves,
by pressing down the leaves and dipping up water, but no
other species was found. In January in the same spot a few
funesta were gathered and great numbers of Pyreto-
phorus costalis.

Relation to Malaria.—This larva is abundant both in

malarious and non-malarious districts, but in the malaria
season, in places where the disease is epidemic, we have
always found costalis also, which is absent in the cooler
months. The imago is found in houses, but cannot be said
to frequent them as costalis does. We have not found it in
dwellings of persons infected with malaria, but possibly
because it is more difficult to distinguish than costalis.
On the other hand, in Maritzburg a few indigenous cases of
malarial fever occurred in 1906. Costalis larve were not
found in the neighbourhood, but in the vicinity of a part
where some cases were reported, funesta larve were
collected; larvee of N. pretoriensis, and M. paludis,
however, were found at the same time, and in the same place.

Pyretophorus costalis Loew.

The following description is given by Theobald (vol. 1, p. 157)
of certain markings of the female imago:

1 If this species of which we describe the larva is, as Giles says, not
funesta, the latter may be one with listoni.

132 ERNEST HILL AND L. G. HAYDON.

“Palpi black scaled, apical joint yellowish white, and the
apices of the two preceding also banded white, the bands
being narrow.

“Lees with the femora and tibie brown, spotted and
mottled with yellowish scales; in the fore legs the joints are
broadly banded with yellow, the bands involving both sides
of the joints, in the mid and hind legs these bands are not
so marked, and only now and then partially spread on to the
bases of the joints, the major part of the bands being
apical.

“ Wines with four large and two small black spots on the
costa, the two median large ones are the longest; on the first
long vein there is a black mark under the apical spot and
two under the next costal spot and two under the next large
one, a single one under the fourth; a black patch at the tip
of each vein, and at the fork of the first submarginal cell.
On the stem of the first submarginal cell, just at the fork and
under the third costal spot, is a small patch, another on each
side of the mid cross vein on the third vein, another at the
fork of the fourth vein, and one on each branch of the second
posterior cell, and another past the cross veins; two on the
stem of the fifth, with three on its upper branch and three
on the sixth; first 'submarginal cell longer, but little
narrower than the second posterior cell. . . . Fringe
black with yellow patches, where the veins join the border
of the wing. Length, 3 to 45 mm.”

Variations in wing spots are noted, particularly in the first
long vein.

This is the commonest and most widely distributed of
Anophelina in Natal on low levels near the sea in the
warmer months of the year. We have not encountered it in
inland districts, and have not found the larva before October
nor later than early May.

The wing markings do not correspond in all respects with
Theobald’s description (vide Pl. XXV, fig. b). Twenty-four
females were carefully examined; the black spot at the fork
of the first submarginal cell was absent in three, and consisted

CHARACTERISTICS OF LARVE OF ANOPHELINA. 133

of few black or brown scales on other three; there was no
“black” spot at the fork of the second cell in any of the
twenty-four ; in six there was no pale spot on the wing fringe
opposite the sixth vein; in nine an additional spot was noted
between the fifth and sixth ; in seven a further spot on the basal
side of the sixth, and the last mentioned was observed in two,
in which there was no spot between the fifth and sixth veins.

The wing pattern of this species is, however, fairly constant,
and though differences are observable they are not sufficient
to justify any tabular representation. The species is very
readily identified. The dark scales are, in most individuals,
brown rather than black, giving a general colour appearance
of brown. ‘The wing viewed through a hand lens has a
rather unkempt appearance in contrast with the sleek, well-
groomed aspect of funesta, and the row of small dark spots
of even size at the extremity of the veins, just separated by a
yellow dot from the fringe, is noticeable. The mottled
appearance of femora and tibiz is, as Theobald says, the
chief character in identification. The word “ mottled” very
ably expresses the appearance. The legs of Pyretophorus
ardensis are also spotted, but the spots are brighter in colour
and sharper in outline, the distinction being enhanced by the
black ground-tint in this species.

Average length of detached wing, 3°3 mm.

The Larva (Pl. XVI)—Determined on over fifty speci-
mens, drawn from comparison of ten. This species can
generally be distinguished from others described in this
paper by the unaided eye. There is nothing particularly
noticeable in respect of length and breadth relative to other
species ; both might be stated as medium, the latter a little
under than over medium size; longer, but narrower, than
funesta, and narrow as compared with ardensis. There
is in nearly all specimens, when alive, an irregular central
patch of much deeper brown, which is particularly noticeable
in the second to the fifth abdominal segments. This appear-
ance is scarcely capable of description, but has been observed
in no other species.

134 ERNEST HILL AND L. G. HAYDON.

Antenna.—No branched hair on shaft, terminal spines
equal, hair divides into four.

Frontal hairs.—Three pairs. Anterior median straight
with fine spicular branches, often difficult to discern in living
larve ; antero-external and posterior small, fine, and in
instances bifurcate; subject to variation (vide Pl. XVI,
fig. a).

Palmate hairs.—Relative to larva very small; average
radius, 0°094 mm.; entirely absent from thorax, rudimentary
on first abdominal segment (Pl. XVI, fig. c) well developed
on—second to seventh (Pl. XVI, fig. b; Pl. XXIV, fig. 5b).
Relation of length of filament to length of leaflet, together
with filament, averages as 0°34 to 1; maximum, 0°37,
minimum, 0°29.

The combination of frontal and palmate hairs is quite
distinctive.

Habitat.—Preferential situation rather dirty water, by
reason of which this larva, much more than any other
Natal species, is found covered with vorticelle. It is found
in rain-water pools, cattle foot-prints by streams and marshes,
roadside puddles, especially where refreshed by a shght flow
from a temporary spring, and residual pools in rocky streams,
(but only very occasionally in the actual watercourse), drains
and trenches in fields and marshy areas.

Season.—Summer, from October to May. The summer
months happen to be the rainy season, and it may be the lack
of smtable water which limits the breeding. It is notice-
able, however, that, in general, the water in which it is
found is quite bare of shade, and commonly quite warm to
the hand.

Relation to Malaria.—Found throughout parts where
malarial fever is epidemic, but not encountered where the
disease 1s absent or sporadic; the appearance of larve pre-
cedes the rise in malaria incidence, which commences sharply
about five or six weeks later, and sinks to minimum a few
weeks after larvee cease to be found. Almost all the imagines
captured by us in infected habitations in 1905 were of this

CHARACTERISTICS OF LARVA OF ANOPHELINA. 135

species, and in 16 per cent. sporozoites indistinguishable from
those of malaria were demonstrated.

Pyretophorus cinereus Theobald.

Essential points in the markings of the female of this
species are thus described by Theobald (vol. i, p. 161, and
vol.11, p: 78) :

“ Palpi quite straight, of nearly equal thickness throughout,
but shghtly dilated at the base owing to long scales, dark
purplish brown, with four white rings, the last apical, the
three median ones involving both sides of the joints, apex
with a brush of golden bristles.

“Legs very thin, deep black, coxee and trochanters very
pale yellowish white, greatly contrasted with the dark
femora; apex of the femora and tibiae with a pure white
spot, apices of the fore and hind metatarsi and tarsi with
minute yellow apical bands, which seem absent in the mid
legs ; last larval joint of all three pairs rather paler than the
rest, hind metatarsi longer than the tibie.

“Wines with the costa black scaled, with three yellow
spots and yellow and black apical fringe ; the yellow spots
extend on to the first longitudinal, which has also yellow
scales at the base and a small spot in the middle, two small
ones on the upper branch of the first fork cell, one on
the lower branch and one at the base, the greater part of
the third vein pale with three black spots, two on the upper,
one on the lower branch of the second posterior cell, and one
at the base and another on the stem; two black spots on
the upper and one on the lower branches of the fifth long
vein, one at the fork, and another black spot near the base of
the fifth vein; the sixth mostly pale with three black spots ;
, . . fringe yellow atthe apex, with or without two small
dark spots, remainder brown, but the scales where the
branches of the fourth and fifth join the border somewhat
paler.

“Length, 5 to 5:2 mm.”

£36 ERNESt HILL AND L. G. HAYDON.

This species is rare in Natal, only less so than Myzo-
rhynchus natalensis. We have taken few specimens, most
of them in inland districts on higher levels, but some from
sea level also.’ Our specimens (PI. XXV, fig. ¢) differ in wing
pattern in some particulars from the description, principally
in having two small yellow spots at the base of the costa and
bright patches on the fringe, where the branches of all veins
except the sixth join the border. We have not sufficient
specimens to judge of the amount of individual variation.

Average length of detached wing, 5'°2 mm., the measure-
ment in Theobald’s figure being 5°5 mm.

The species is readily identified by its large size, the pre-
dominance of very pale yellow in the wing pattern, the long
thin palpi with four pale bands and the long black legs.
The imago has been taken two or three times in a house in
Maritzbure.

The Larva (Pl. XVII).—Determined on one specimen,
drawn from two. Owing to paucity of specimens we do not
write with the same absolute confidence of identity as in the
case of the other species. There was, however, no room for
doubt as to the larva being different from funesta, and the
imago, a male, appears to be certainly cinereus.

General aspect.—A medium length larva, with long,
large head, moderately pigmented in the middle strip of the
body.

Antenna.—No branched hair on shaft; terminal spines
equal, hair bifurcate.

Frontal hairs (Pl. XVII, fig. a).—Three pairs, all
straight and smooth, the posterior longer and finer than of
any other species of our knowledge.

Palmate hairs.—Absent from thorax, rudimentary on
the first abdominal segment (Pl. XVII, fig. c), intermediate in
character on second (that is, the hair as a whole is fairly
developed, but on most leaflets the notching of shoulder is
scarcely perceptible (c f. Pl. XXI, fig. c, 2), large, with well-
defined shoulder on third to seventh segments inclusive

(Pl. XVII, fig.b; Pl. XXIV, fig. c). Average radius 0°140 mm.

CHARACTERISTICS OF LARVH OF ANOPHELINA. 137

There is no proper filament, but a lanceolate ending tapering
to a point; average relation of lancet to leaflet with lancet as
0:20 to 1; maximum, 0°23; minimum, 0:19. There is not
much variation as between individual leaflets.

The larva of this species comes near to that of funesta,
particularly the large variety, and Jacobi. Points of dis-
tinction are that the posterior frontal hairs are relatively
longer and finer than those of funesta, in which also is a
rudimentary palmate hair of ten or twelve leaflets on the
thorax and a well-developed hair on the first abdominal
segment, and the palmate hairs are provided with definite
filaments, whereas in cinereus there is no hair on the thorax,
and the rudimentary hair on the first abdominal segment is
smaller. (In one specimen it is very small, indeed.) There 1s no
defined filament. The anterior median frontal hair of jacobi
has spicular branches, whereas that of cinereus is smooth.

Habitat.—The few larve collected were found in running
water—on sea level and at 2200 ft.

Season.—Perennial.

Relation to Malaria.—No evidence of any in Natal.

Pyretophorus ardensis Power.

This species is not described in 'Theobald’s ‘ Monograph.’
The following are distinctive features :

Palpi thickly clothed with dark brown scales, the tip of
pale cream, and three narrow bands of the same colour, the
second and third being far apart.

Legs black, all joints banded with pale yellow, the bands
being wider in the larger joints and the hind legs than in the
smaller joints and fore and mid legs. Coxa and trochanter
very pale. Femora, tibia, and metatarsus of all legs brilhantly
marked with clearly defined pale yellow or cream spots and
incomplete bands, which are appreciable by the naked eye.

Wings (Pl. XXV, fig. p).—The pattern of the markings is
sufficiently indicated in the figure. We have not observed
much variation. The small yellow spot on the first longi-

138 ERNEST HILL AND L. G. HAYDON.

tudinal vein, beneath the large median black patch on the
costa im some specimens merges in the adjoming yellow spot,
but in most is separated from it by a few dark scales. In
some a yellow spot is found in the middle of the posterior
branch of the first fork cell. In some specimens the whole
shaft of the fourth longitudinal vein is dark. ‘here is a
yellow spot on the dark brown fringe opposite the termina-
tion of all branches of the first five veins, except the posterior
of the second, and sometimes an additional spot on the
proximal side of the sixth.

Length of detached wing, 3°5 mm. to 4 mm.

This species is not common. We have found it only in
three places on the higher levels, and never at sea level. It
is captured in houses. Was first found by Mr. H. 8. Power
on his property forty miles from the coast, and we are indebted
to him for information that it has received generic and specific
title from Theobald. It is poorly represented in our collec-
tion, and there may be greater variety in wing patterns than
we have observed.

The spotted legs cause it to resemble slightly Nyssorhyn-
chus pretoriensis, from which the absence of white
hind tarsi enables it to be readily distinguished, apart from
other differences, and the dark wing scales are brown as
contrasted with black in pretoriensis.

The Larva (Pl. XVIII).—Determined on three speci-
mens, drawn from comparison of five, two being very small.

General appearance.—Of medium length, and rather
sturdy build, uniform pale brown colour.

Antenna.—No branched hair on shaft, terminal spines
equal, hair short, simple and undivided in the three instances
in which examination could be made.

Frontal hairs.—Three pairs. Antero-external and pos-
terior subject to variation (Pl. XVIII, fig. a; a, is the external
frontal hair of a third larva).

Palmate hairs.—Very small and rudimentary on thorax,
larger but still rudimentary on first abdominal segment
(Pl. XVITI, fig. c), very large and well defined on second to

CHARACTERISTICS OF LARVA OF ANOPHELINA. 139

seventh segments inclusive (Pl. XVIII, fig. b, Pl. XXIV,
fig. d). Average radius 0°166 mm. Average relation of
filament to leaflet and filament together as 0°26 to one;
maximum 0°34; minimum 0°18. The palmate hairs are
characteristic in exceptional length, breadth and number of
leaflets, and this together with the frontal hairs make the
species unmistakable.

Habitat.—Found ina quick-running streamlet under over-
hanging banks in company with the larger variety offunesta.

Season.—Fairly numerous in the one stream in April,
May, June and Sept., 1905, rare in Jan., Feb., May, 1906.

Relation to Malaria.—No evidence of any connection.

Nyssorhynchus pretoriensis Theobald.

Theobald (vol. III, p. 99) writes as follows :

“ Closely related to N. maculipalpis, but the palpi are
not mottled and are somewhat longer; the two white apical
bands are farther apart. ‘The hind tarsi have also different
ornamentation; the second hind tarsus has a small black
patch near its base; the metatarsus is mottled with white
and black, and has a broad white apical band hke the first
tarsal. The last two hind tarsi only being all white.

“Tt is subject to much variation.

“ One specimen in the series has the last palpal joint black
at the base, so there are four, not three, pale bands; the
wings are also darker scaled, the third long vein being
entirely black.”

The legs and wings of N. maculipalpis are thus
described (p. 97) :

“Lees black, with the femora and tibiw, and to some
extent the metatarsi, with small clear white spots, first fore
tarsi with apical white bands, mid-tarsi plaim; in the hind
legs the last three and apex of the first tarsi pure white.

“Wings clothed with mostly dusky-black narrow elongate
shehtly clavate scales, with some white areas, as follows: On
the costa five pure white spots, the two apical ones and the
fourth spreading on to the first long vein, which is also

140 ERNEST HILL AND L. G. HAYDON.

white at the base, a small white patch on the lower branch
of the first fork cell, a few white scales at the base of the
fork; a small white spot about the middle of the third long
vein, and another near its base; one small white spot on each
branch of the second fork cell, and one at the base; three
fair-sized white spots on the upper, and one large one on the
lower branch of the fifth, and two on the stem; apex and
base of the sixth white, and a broadish median white spot ;
fringe apparently all black.

“Length, 5°5 mm.”

This species is not frequently found in Natal, and its
breeding grounds appear restricted. It has occasionally
been found by us inland at level 2,200 ft., and rarely at sea
levei or thereabout. The wing (Pl. XXV, fig. e) corresponds
closely to Theobald’s description, except that there are pale
spots on the fringe opposite the terminations of all veins,
other than the lower branch of the first fork cell, and in one
instance, the sixth. The coast specimens are darker than
those from inland, and have one small white spot on first
longitudinal vein under the long black costal patch as
against two in inland specimens. The size varies somewhat,
bunt not so much as in funesta.

Average length of detached wing 3°8 mm.

Following table shows important variations in ten females:

Win hee Number with a |
Number of of Hanne spot on lower Size of spot in middle Proportion of second
specimens. SAR branch of first of third long vein. hind tarsus white. |
On Pa Dr: fork cell. |
|
a No: | Q me 2
Spot spot. | Large. Small. |Absent. so ple ae
| |
5) 3 3 2 | 3 1 1 3 | O 2
5 4 3 2 salem 2 1 2 OS |
| i
| ~ ~
6 Ail ee 3 2 5) 2 3
7 = SS SSS SS
10 10 10 10

CHARACTERISTICS OF LARVE OF ANOPHELINA. 141

The Larva (Pl. XIX).—Determined on fifteen specimens,
drawn from comparison of eight of different sizes and five
last moults.

General aspect.—About the same size and dimensions
as funesta; translucent at sides, brownish in middle.

Antenna.—No branched hair on shaft, terminal spines
equal, hair divides into three. Theobald presents it with six
branches, besides the termination of the stem. Weare unable
to make out more than three in any of our specimens in which
satisfactory examination is possible. Theobald also states
that the spines are unequal, which does not concur with our
experience. From this it follows either that this hair is
variable and of no specific value, or that, despite the close
similarity to his pretoriensis of the imago which we
represent, it is really a distinct species.

Frontal hairs.—Remarkably uniform; posterior pair
relatively short and fine as compared with unbranched
posterior hairs in funesta (Pl. XIX, fig. a; Pl. XV, fig. a).

Palmate hairs.—Absent entirely from thorax, rudimen-
tary on first abdominal segment (Pl. XIX, fig. c) well deve-
loped, but relatively small for size of larva on second to seventh
inclusive (Pl. XIX, fig. b; Pl. XXIV, fig.e). Average radius,
07088 mm. Average relation of length of filament to total
length of leaflet, together with filament, as 0°26 to one; maxi-
mum, 0°38 ; minimum, 0°20. The contour of the “shoulder ”
is somewhat variable in this species, and in some specimens
the filament is less tapered, and terminates in an abrupt
widening of the leaflet with fewer notches at the “ shoulder.”

This species is recognised by the three pairs of smooth
frontal hairs, the absence of any palmate hair on the thorax,
and the relatively small size of these hairs on the abdominal
segments. These features of the palmate hairs differentiate
it effectively from funesta.

Habitat.—Found in slowly-running water in ditches in
October, March, April, May, at one place near sea level and
in one locality at 2200 ft.

Relation to Malaria.—No evidence of any.

142 ERNEST HILL AND Il. G. HAYDON.

Cellia squamosa Theobald.

Theobald (vol. i, p. 167) thus describes certain features of
the female of this species:

‘“Palpi densely scaled with deep brown scales, which stand
out from the surface like A. sinensis, with a few white ones
here and there, forming also three narrow bands, one apical.

“Wings with the costa deep black with three distinct very
small white spots, two smaller basal ones and two small
apical ones, some of these spots pass but very indistinctly
into the first long vein; most of the veins dark scaled, the
lateral scales being clavate, always dark but paler on the
white areas of the vein; the median scales are creamy white
on most of the third long vein, the black forming three spots ;
a small patch on each branch of the second fork cell, a large
patch on each branch of the fifth vei, and another on the
stem, and two large patches on the sixth long vein; fringe
uniformly brown, the lateral scales, even on the pale areas,
are very dusky, so that the creamy spots on the wing field do
not show up very strongly.

“ Lees with the femora dark brown, tibize and metatarsi of
the fore legs mottled with numerous white scales, also the
apex of the first and second tarsal joints of the fore and mid
legs broadly white banded; hind legs with the femora
swollen, dark brown with white patches, one large patch near
the apex, the extreme apex white; tibiz mottled black and
white ; metatarsi apically banded, and also the next three
tarsal joints, the last black.

“Leneth, 5 to 5:5 mm.”

The genus Cellia is by no means common in Natal. We
have encountered two species; the one, of which a wing is
reproduced (Pl. XXV, fig. f) corresponds fairly accurately
with Theobald’s description of squamosa. On the first longi-
tudinal vein, however, are two minute additional white spots, not
mentioned in the description, and a large spot on the posterior
or lower branch of the first fork cell, of which no mention is

CHARACTERISTICS OF LARVE OF ANOPHELINA. 143

made, whereas in Theobald’s figure is shown a spot on the
anterior or upper branch which is not alluded to in the text.
The wing figured in ‘ Monograph of the Culicide’ is 3°9 mm.
in length, that of Natal specimens but little over 3mm. In
the light of the great variation in size which we find in
funesta and paludis, this appears to be no obstacle to
placing our specimens in the same species. We have but few,
and are unable to state how much variation in wing pattern
occurs ; but we may note that we have observed one specimen
with an additional white spot on the fringe opposite the
termination of the third longitudinal vein, and a second with
yet another spot opposite the anterior branch of the second
fork cell. There are also observed in some specimens four
narrow white bands on palpi instead of three.

The Larva (Pl. XX).—Determined on four specimens,
drawn from comparison of four.

General appearance.—A slender larva with much pig-
ment in median strip.

Antenna.—No branched hair on shaft, terminal spines
equal, hair divides into three branches.

Frontal hairs.—Three pairs; anterior median and pos-
terior vary (Pl. XX, fig. a). In three of the four specimens
the external hair is dendriform (Pl. XX, fig. d), in the fourth
(fig. e) itis rather penniformin shape. This larva was taken
from a small collection, of which some five or six developed
into imago of squamosa. The exact shape of the hair is
scarcely appreciable under a lower magnification than x 100,
and is readily overlooked, and can scarcely be differentiated
in the living larva. There is no other difference.

Palmate hairs.—Rudimentary on thorax (Pl. XX, fig.c),
well developed on the first abdominal segment and excep-
tionally large second to the seventh inclusive. The leaflets
are relatively narrow and few, about sixteen in number (as in
jacobi) in contrast to ardensis, in which they are broad
and numerous, about twenty-five. Average radius 0°144 mm.

Average relation of filament to total length of filament and
leaflet as 0°36 to one ; maximum, 0°40; minimum, 0°35,

VOL. 15 PART 2. 11

144 ERNEST HILL AND L. G. HAYDON.

Habitat.—Found occasionally on coast and at levels of
2000 ft. to 2800 ft., once in residual pools in a river bed, and
three times in marshy pools directly fed by small springs.

Season.—April to October.

Relation to Malaria.—No evidence of any.

Cellia jacobi n. sp.

We have not found any description tallying with this
species. It may be described in the following terms :

A large black and white mosquito with spotted legs.
Palpi thickly covered with bushy black scales, a few white
interspersed ; irregular white bands at apex and last joint, a
very narrow band about the middle, and a few white scales
in an incomplete rmg between that and the base. A tuft of
white hairs on the clypeus overhanging origin of palpi.

Thorax of sepia, with clothing of narrow curved white
scales forming three distinct longitudinal bands; three white
bands on lateral aspect of thorax; abdomen black, thickly
covered with narrow curved yellow scales and long golden
hairs ; a thick lateral tuftof black scales on second to seventh
segments.

Legs: Coxa and trochanter dark grey, flecked with white
scales. Femur and tibia thin, white scales predominating
over black, a few white flecks on black metatarsi; white
bands at apex of metatarsus, and all tarsi, except in the fore
and mid Jegs the third; tip of last tarsus white in all legs.

Wing (Pl. XXYV, fig. g): Costa is black; there are three
main white spots, which are small; a fourth still smaller at
the apex, and two white dots at the base. First longitudinal
vein black, one white dot at base, a white spot under each of
the four remaining costal spots, mostly smaller than the
latter, and a minute group of white scales in the two long
black strips. Second longitudinal vein black, a white spot at
the fork, anda large white patch on the posterior branch of
the first fork cell; third vein mostly white with a black spot
at each end; fourth vein black, a white spot near the fork,

CHARACTERISTICS OF LARVE OF ANOPHELINA. 145

which also is white, and a large white spot on each branch of
the second fork cell, and at the tip of each branch ; fifth vein
mostly white, a black patch on the stem and at the fork, two
on upper and one on lower branch; sixth vein white with
three black spots. On the second and fourth veins white
scales are interspersed with black in some specimens. Fringe
black, with a white spot opposite termination of all branches
of veins except second.

Length of detached wing, 4°5 mm.

The principal variations noted in nine females are: Very
few white scales at fork of second longitudinal, the stem of
the fourth vein almost all or entirely white, and three white
spots on the first vein additional to costal spots, instead
of two.

The Larva (Pl. XXI).—Determined on three specimens,
drawn from one preserved and two living.

General aspect.—Large, deeply pigmented.

Antenna.—No branched hair on shaft, but a prominent
curved spicule about one third of length from base on antero-
external aspect; spines equal, terminal hair divides im-
mediately into three.

Frontal hairs.—Variations as shown in Pl. XXI, fig. a;
the spicular branches on the anterior median are difficult to
discern with low powers.

Palmate hairs.—Absent from thorax, very fine and
rudimentary on first abdominal segment (Pl. XXI, fig. ¢, 1),
rudimentary but functionally active on second abdominal
(Pl. XXI, fig. c, 2), well developed and well defined with
broad leaflet on third to seventh inclusive. Average radius,
0°108 mm., leaflets fifteen to eighteen, relation of filament to
length of filament and leaflet as 0°20 to one, maximum 0°24,
minimum 0:14; in the latter there is really no filament
at all.

Habitat.—Found in small springs at sea level in one
neighbourhood only.

Season.—Cold weather.

Relation to Malaria.—No evidence of any.

146 ERNEST HILL AND L. G. HAYDON.

Myzorhynchus paludis Theobald.

In ‘ Monograph of the Culicide,’ the following descriptions
of certain parts of the female are found (vol. 1, p. 128) :

“ Palpi densely scaled with black scales, with four narrow
rings of white scales.

“Legs yellowish, with dark brown scales, the apices of the
metatarsi and first tarsal joints of the fore and mid legs with
a fine yellow band; hind legs with the metatarsus very long,
and the extreme tip of the first tarsal joint white, the other
three jomts pure white.

“Wings clothed with yellow and black scales; the costa
dark, broken by two small yellow spots, one near the apex,
and the other about a third of the length from the apex, the
apical spot extending on to the first long vein and the upper
fork of the second long vein, the other spot passing on to the
first long vein only; there isa yellowish spot farther back
on the first long vein, which does not, however, reach the
costa ; fringe just as in sinensis.”

Of the fringe of sinensis is written :

“Fringe yellow at the apex, a small black patch separating
it from the costal spot, a pale patch where the lower half of
the fifth vein joins the border, all the rest dusky violet
black, except the border scales, which are pale yellow in
reflected hght.

“Qength, 5 to 55 mm.”

On page 129, vol. i, is found the following on Anopheles
paludis, variety similis:

“ Palpi densely scaled with black scales, with two, three
or four rings of white scales ; the apex usually white.

“Lees ochraceous, clothed with deep brown scales, with
a bright ochraceous sheen in some hghts; the apex of the
metatarsi and first two tarsal joints of the fore and mid legs
apically banded white, and also, to some extent, may be seen
a white basal patch on the metatarsi. In the hind legs the
metatarsi are broadly white banded at the base and apex,

CHARACTERISTICS OF LARVE OF ANOPHELINA. 147

and the apex of the first is also banded white; the second
is almost all white, save a small black basal band, last two
pure white.

‘Wines densely clothed with black and yellow scales, the
former predominating, so that the wings look sooty black.

The black scaled costa has two small white spots,
one near the apex; this spot extends on to the first long vein
and the upper branch of the first submarginal cell; the
second costal spot just touches the first long vein. ‘ There are
also small pale scaled areas on the lower branch of the first
submarginal cell, on each of the branches of the second pos-
terior cell, a few pale scales on the bases of the fork cells,
and on the third long vein. On the upper branch of the
fifth are two small pale areas, the lower branch and stem
being mostly pale scaled. The sixth pale scaled with two
black spots. Wing fringe entirely brown, the border scales
yellowish.

“Length, 5 to 55 mm.”

Vol. ii, p. 85, this variety is elevated to the dignity of a
distinct species, Myzorhynchus mauritianus Grandpre,
and the following remarks are made :

“The species is quite distinct, and may be told by the four
palpal bands, by the last two hind tarsi only being all white,
and the absence of the pale wing spot, and thus differs from
paludis Theobald.”

It may be that the species are quite distinct to the describer,
thoroughly familiar with both, but the descriptions would cer-
tainly not enable a distinction to be made. Distinction of
colour as between white and yellow is not constant. In vol. 1
it is stated that paludis has four white palpal bands ; there
are three pale wing spots mentioned as found in paludis,
which are not mentioned in the account of paludis, variety
similis, viz: one on the first long vein further back than one
third of the length from the apex, one on the fringe at the
apex, and one at the termination of the lower branch of the
fifth vein. Which of these is the pale wing spot, of which
the absence distinguishes mauritianus from paludis? The

148 ERNEST HILL AND L. G. HAYDON.

apical spot, too, is absent from fig. 26, p. 129, representing
paludis. The remaining point is thatin paludis the whole
ot the second tarsus is white, but in mauritianus there is a
small black basal band. This appears to be a refinement.

One or other, or both, is, or are, largely represented in our
collection, and thirty-seven females have been carefully
examined and the results tabulated. The wings of some are
sooty black, of others of a deep sepia. There is much variation
in the length of the detached wing, the smallest 3°4 mm., the
longest 5°0 mm. (Pl. XXV, h,k,1l). In all save five there is a
spot at the apex, in some quite small, in others very broad ;
in three there is no spot at all on the costa near the apex; m
other two it is very minute. In eight the costal spot one
third of the length from the apex is absent, and in these and
further twenty-six there are no white scales on the first longi-
tudinal vein at this point. The proportion of the second hind
tarsus, which is white, varies from one sixth to seven eighths,
or almost all. Differences are set forth in the table, in which
it is not possible to represent the marked difference of size of
spots, which is very noteworthy.

The variations, as shown in the table, are characteristic of
this species in Natal, of which we have examined several
score of specimens. They are quite as great as between
Theobald’s paludis and his mauritianus, and between
specimens with any one spot and those deficient in it there is
a continuous gradation. Seemg that we have been able to
detect no difference in dozens of larvae’ examined, we are
forced to the conclusion that our collection represents one
species only. Whether that species is paludis or mauri-
tianus we are quite at a loss to determine, and have, there-
fore, retained the earlier designation of Theobald.

The species is very common and widely distributed, falling
little, if at all, short of funesta in this respect. It is not
infrequently found in houses with onset of cold weather.

The Larva (Pl. XXTI).—Determined on over twenty speci-
mens, drawn from comparison of ten of different sizes.

General aspect.—A large larva with small head, which,

149

LINA.

ny
vi

F ANOPHI

O

LARV A

OF

Ss

TERISTIC

CHARAC

‘SOOURISUL [TV UL MOT[OA AO OFIYM JUO}XO SSOT IO 1ozVaIH OJ ST UIOA OYJ Jo O1}UOD OTT, ¢
SOTVOS OFLA XIS IO BAY OUIOS Fo sJsTSUOD sooUASUI ysour ut gods si], ;

“LOTIOJUY TO JOU PUY AOTAE4sod WoO 9MO UT ‘.OTAEYSOd Jou puR
Iomeyur moO oO UT fYouRAG yore wo yods v 90 UT

‘polos oFLYM tO MOT[OA UILOA OATPUO SOUTH 5
“ATWO

“AXX ‘Td 7 puv y ‘ssy oredunog ,

j |
LE Lé | L& Le “e | Let Ze | 28 | 28, Ls | LE | Le LE L&E ~ STB4OL,
eee | | ——
Ne Ce leG a ROLaiesir (9G: lke ST Es
en icc ee | (eagle L | ae NG Comoe ae eS S +S) Sa a eee PLAS Gy
= — —_—— ali - A eotieed onal === | — =
O€ | GE] GT | Z | TE | 28 | SS Pl 1s AM VCCY GAS: MEAS | [eel ey anit) | ASE Fa
| —|T tT Z |Z = |) = |e ae leant = ih G et | (LG = i ae | I G Zz 4
| Com eL Ge eGMit tne hag Gm Nemes vile ene eel cel OM oer KS al & 135
=<" Jo Bh Ve) eet | fey Vp | |) fecal ete FS )). (Ty eee Lk Sl (me We a {ite a & (Bs
S| Sareea eta 1 I SFU Mes 3 Sel et (esol pee sea | Macaeal leak € ; rey babes zc é |:
SANG CaaS WE | NE TG PSN seine Siti 96 | 9G F GI | G 96 | 92 | 9G | DA | GT | 6 G Gs Aaaaee
| | | : ald pUnuy
| | « ce
| SUNUVIFLANVUL,,
ol) ee Seen eee gn cere eee ween err (eg eee Sees eee Stee tee || ee OCLAgI SED BO OoMtn|
| | | .. Stpnyed ,,
| ¢ :
te | | Ty |), ea gy eee || meat ies a hl] ae Heo] | fe af | —— | eee PedAa sipleqoouiy,
miler eo Pes ee ea eon iss hes] we ee Pale hace pea oa | en Eee ee ale oes
Se Cag ie eon ale Che Pano ee oie calc. | Sere Sn eae pe || is ie ete Rgere eoiaiieree | aoalatae | te
8 L ; G (Ee hE I ee ree | SL Vee al ey D ct a | oe = 5 2 ES 2 = 6
De eee cl te Wel tae Neva Bil ae le saints aa ese ce ESA ae lac |
ep =e CPi : ep > © eS & E & g ‘ 5 =
Ceiesate ee Slee Metal ee eal eaimee eel P
| , ® | | ; Say aligey ale cae lhe |
} nw . .
ig? ‘ULAA SUOT PLE atI89 ‘UIA HUOT IST “RISO:

OTIM SNSIRZ PUI Puy

yo wOLtodoad YOuUpA WE .caquImM N

YlOF PU

YAOF IST

"SOUIM TO 890d8 OF AA

‘uoeutdads

» SR U7
“SLp n ped O[BUTO YT JO suoutta0dg WOAOS- AUF, ul W1OFTB jo SUOLPBUTG Ul) JUILO LLP OULMOTLS OQRL

150 ERNEST HILL AND L: G. HAYDON.

when mature, has a sturdy appearance, but when immature
looks slender, in part due to deficiency of pigment at the
sides, a feature commonly retained throughout in captivity.
The full-grown larva is generally richly pigmented in its entire
breadth, and exhibits greater colour variations than any
other ; many specimens are dense black, others brown, brick-
red, yellow, green, and even occasionally cobalt. It does not
thrive in captivity, and, although living for many weeks,
grows very little, and seldom, if not nearly full-grown at time
of capture, undergoes metamorphosis to pupa. A feature in
its behaviour is quiet submission to microscopic examination,
a treatment commonly resented by other species.

Antenna.—A multibranched plumose hair on intero-
anterior aspect of shaft, about one third of distance from
origin. The number of divisions was eight in eleven of
thirteen counted, but in one specimen ten were counted on
one side and eleven on the other. (This specimen also ex-
hibited a peculiar hair on the first abdominal segment).
Terminal spines equal, and the hair commonly divides into
six, but of seven counted eight were found in two and ten in
one, and the number was not constantly symmetrical.

Frontal hairs.—Three pairs. Anterior (Pl. XXII,
figs. a, d) constant in all specimens, only shght modifications
in the branching of the dendriform being observed. Pos-
terior pair very small and difficult to distinguish, and
exhibiting variation (Pl. XXII, fig. a).

Thorax.—Median hairs overlapping the occiput, which
are constant in the other species which we describe, are
absent from this.

Palmate hairs.—Rudimentary but large on thorax
(Pl. XXII, fig. c), smaller and still rudimentary in character,
though stronger, on first and second abdominal segments,
well developed and with definite notching on third to seventh
segments inclusive (Pls. XXII, b, XXIV, h). Average radius,
0-116 mm. There is no filament, but a jagged lanceolate ter-
mination of characteristic. pattern. This is not observed in
rudimentary hairs on thorax and first two abdominal seg-

CHARACTERISTICS OF LARVE OF ANOPHELINA. 151

ments, the outline of which is quite smooth to the tip. Inthe
young larva in most, but not in all, cases, the palmate hairs
are quite transparent, with exceptionally sharp outline ; in the
mature larva on the third to seventh segments they are
highly pigmented, except in the distal one third, which
remains quite clear; the hairs on the first two abdominal
segments and the thorax do not acquire pigment, and owing
to this may readily be overlooked in the live larva. In the
dead the hairs on the lower segments are commonly expanded,
but on the anterior and thorax always collapsed, on which
account and by reason of flimsy texture they may not be
noticed. We have frequently needed to employ brilliant
illumination through the body and relatively high powers for
demonstration, but have invariably discovered them. PI.
XXIV, fig. 1, shows a larva of this species with palmate
hairs of the first four and the seventh abdominal segments in
action; the different character of the first two 1s well shown,
particularly by the shadow on the surface of the water. The
plate has been accidentally scratched, and an appearance as
of a thin hair arising from the thorax caused, but inspection
with a lens shows its nature.

A very curious feature has been observed in three indi-
viduals only, which appears to have no specific importance—
a large delicate dendriform hair on the first abdominal seg-
ment (Pl. XXII, fig. f). This overlies the palmate on a
higher plane, with origin nearer the middle line (the palmate
hair is shown on one side only of Pl. XXII); the branches
are quite transparent, and in examination of a living larva, if
the palmate came at once into focus, might probably be
missed. It was first observed in one larva of a small batch,
but not in other six gathered at the same time. The collec-
tion of preserved larvee was then examined minutely, and
this feature found in one only. More were collected from
the pool from which the latter had been taken, but of thirty
the dendriform hair was observed in one only, and in ten
gathered from another source in none. Imagines were ob-
tained from the two; both were males but no differences

U

152 ERNEST HILL AND L. G. HAYDON.

could be detected between them and other males in which its
absence had been ascertained. It is strange that so remark-
able a feature should have no specific or other importance.

The larva of this species, whether it be paludis or mauri-
tianus, differs from barbirostis, as figured by James and
Liston, in the relative thickness and number of branches in
the dendriform frontal, in the presence of posterior frontal
hairs, and in the character and position of palmate hairs.
We have not observed branching of median anterior hairs.
The two species must, therefore, be deemed quite distinct.

Habitat.—Rarely, if ever, found in water in which grass
and weeds were not growing, otherwise in stagnant (occa-
sionally), very slow or steadily flowing water, clean or polluted,
but not actually foul. It is generally possible by the appear-
ance of water to predict whether this species will be found or
no. For the most part one only, and rarely more than three
mature larve will be taken at a single dip. Widely dis-
tributed from sea level to 4000 ft.

Season.—Perennial.

Relation to Malaria.—Found in areas of epidemicity,
where costalis is more common in the summer, and in places
where the disease is of sporadic occurrence, in which, how-
ever, funesta is also found.

Myzorhynchus natalensis n. sp.

No description, even approximately, representing the
characters of this species is found in ‘Monograph of the
Culicidee,’ or Giles’ ‘ Revision of the Anopheline.’ It may be
thus described :

A medium-sized mosquito, of which the general aspect 1s
black with white and yellow markings; legs brilliantly spotted
and banded, hind tarsi white. The closely-set, broad lanceo-
late scales on the wings and the scaly palpi clearly place it
alongside of paludis, that is to say, in the genus Myzo-
rhynchus, although we are unable to detect any ventral

CHARACTERISTICS OF LARVAE OF ANOPHELINA. 193

scales, as specified in Theobald’s description of the generic
features.

The female.—Palpi black, scales long and copious at the
base, shorter and smoother in the distal portion. Tip white,
and in addition four narrow white bands, about equi-distant
from one another; first from apex narrower than the
second and fourth, and the third very thin indeed, and barely
perceptible in some specimens. There are a few white curved
scales on the head, but no tuft. Thorax black with the
dorsum grey, clothed with short golden hairs, scutellum dark
grey, halteres black at extremity. Abdomen black, clad with
long golden hairs. No scales observed on the ventral aspect.

Wing (Pl. XXV, fig. m; owing to density of scales the
contrast of dark and light cannot be adequately reproduced
in a photograph, as in most species).— Prevailing colour black
with white and yellow markings, the former on the anterior
border, the latter in the wing veins. Scales broadly lanceo-
late, rather longer than in paludis, to which it has the
further similarity that light scales are mixed with the dark
here and there without forming definite spots. Costa black ;
a small white spot one third or so of the length from the
base, a larger spot two thirds of the length, and a third,
which is small, at the commencement of the curve of the
wing; a fourth extending to the apex. On the first longi-
tudinal vein are two yellow spots near the base, and white
spots beneath the costal spots, with a small additional white
mark a little on the distal side of the first costal. Second
longitudinal vein black, two small yellow spots opposite to
the second costal spot; first fork cell long and narrow, a
yellow dot at the end of each branch. On the third longi-
tudinal vein black and yellow scales intermingle, the former
predominating in some specimens, the latter in others, hence
the number and size of definite spots vary, and is not
symmetrical. The stem of the fourth vein is black; there
are numerous yellow scales about the fork cell, which may, or
not, form definite spots at the bifurcation, and on either
branch. The stem of the fifth is mostly black with yellow

154 ERNEST HILL AND L. G. HAYDON.

patches near the base and at the bifurcation ; variable yellow
spots on the anterior, and a large yellow patch on the
posterior branch. Sixth longitudinal vein yellow and black
alternately. Fringe black, a golden spot opposite termina-
tion of all veins, and in some instances a light patch on the
proximal side of the sixth. The number of actual spots other
than on costa and first longitudinal vein is uncertain and
asymmetrical.

The male.—Wing is less thickly clad, and the proportion
of yellow scales is always higher.

Length of detached wing, 3°8 mm. to 4:2 mm.

Legs.—Black; femur and tibia of all brilliantly spotted
with pale yellow, and the metatarsus banded with three or
four bands of white or pale yellow. Onthe fore legs is a
broad white band at the apical extremity of the metatarsus,
and all tarsal segments, except the fourth; the hind legs
similarly adorned, but the apical band of the first tarsus very
broad, and the distal two thirds to one half of the second,
and the third and fourth snow-white. In the mid-legs there
is a barely perceptible apical ring on the tarsal segments.
Knees black in all legs.

This species 1s very rarely found, and we have insufficient
specimens on which to base a table of wing-markings. It is
quite unmistakable ; although at first sight, under a magnifi-
cation of x 10, it resembles Nyssorhynchus pretori-
ensis, the character of the wing scales immediately shows
the difference.

The Larva (Pl. XXIII).—Determined on ten specimens,
drawn from comparison of eight and five last larval moults.

General aspect—A large larva, deeply pigmented
generally with dark brown, very like in shape to a three-parts
grown paludis; in water it looks slim, but measurement
proves the thorax exceptionally broad as contrasted with
the small head, and in this respect also it resembles paludis.

Antenna.—There is no branched hair on shaft. Spines
equal, terminal hair divides generally into six or eight

branches.

CHARACTERISTICS OF LARVAE OF ANOPHELINA. 155

Frontal hairs (Pl. XXIII, fig. a)—Median anterior
fairly constant, with shght variations in number and position
of spicular branches shown in drawings; external and
posterior very fine and variable.

Palmate hairs.—Large with very long fine filament ;
rudimentary in shape but with numerous leaflets on thorax
(Pl. XXIII, fig. c), with a well-defined long filament on first
abdominal, but notching at shoulder indistinct and hair
generally folded in death; well developed on second to
seventh segments inclusive (Pls. XXIII b, XXIV k). Average
radius, 0°166 mm.; maximum 0°187 mm. ; minimum 0°154 mm.
Relation of filament to combined length of filament and
leaflet as 0°53 to one; maximum 0°38; minimum 0°28. It
comes near to ardensis, but the general shape is different,
and the filament of palmate hairs longer and finer, while the
terminal hair on the antenna of ardensis is simple.

Habitat.—Only found three times, once near sea level,
twice at 2200 ft.,im both situations in eddies in fast-running
streams, about shoots of grass and rushes.

Relation to Malaria.—No evidence of any.

DESCRIPTIONS OF PLATES XV—XXVI,

Illustrating Mr. Ernest Hill and Mr. L. G. Haydon’s paper
on “ A Contribution to the Study of the Characteristics
of Larvee of Species of Anophelina in South Africa.”

All the figures are on precisely uniform scale. All figures of larve
and parts are taken from mature specimens unless otherwise stated.

Drawings are taken from the dorsal aspect of larva, and all hairs,
which are fairly appreciable at the magnification employed, and which
can suitably be brought in from the dorsal aspect, are represented.

Measurements.—A micrometer scale was applied to every detail,
although lack of uniformity of plane in some precludes absolute
accuracy.

Palmate hairs——Radius measured from origin of stem on body
to extremity of filament or sharp end of leaflet in direct line with the
stem. Leaflet for comparison with filament from origin on stem to
end of filament. Filament from tip to level of first notch on shoulder

156 ERNEST HILL AND L. G. HAYDON.

on distal extremity. N.B.—Photographs of palmate hairs are neces-
sarily taken from dried specimens mounted in balsam. There is some
shrinkage in preparation, which makes a little difference in appearance.
The drawings are taken from specimens preserved in formalin. Some
points are brought out better in the latter, and on this account both
drawing and photograph are reproduced.

PLATES XV—XXIII.

From drawings. Main figure: dorsal aspect of head, thorax, and
first three abdominal segments of larva. Figures: (a) frontal hairs ;
hairs represented on opposite sides in corresponding positions of the
plate are taken from right and left of the same larva; (b) palmate
hairs; (c) rudimentary palmate hairs; (d,e) frontal hairs further
enlarged; (f) dendriform hair on first abdominal segment of rare
specimens of Myzorhynchus paludis; (¢1, ¢2) rudimentary hairs on
first and second abdominal segments of Cellia jacobi.

PEATE XOX.

Reproductions of photographs of (a—k) palmate hairs of larve. (a)
funesta; (b) costalis; (ce) cinereus; (d) ardensis; (e) pretori-
ensis; (f) squamosa; (gy) Jacobi; (h) paludis; (k) natalensis.
(/) larva of Myzorhynchus paludis at surface of water.

PLATE XXYV.

Reproductions of photographs of (a—m) wings of female imago. (a)
funesta; (b) costalis; (c) cinereus; (d) ardensis; (e) pretori-
ensis; (f) squamosa; (g) Jacobi; (h, k,l) paludis; (m) natal-
ensis.

PLATE XXVI.

Reproductions of photographs of : («) funesta: head and palpi; (b)
costalis: head and palpi; (¢c) cinereus: head, palpi, and part of hind
leg ; (d) ardensis, head, palpi, and legs; (e) pretoriensis:! head and
palpi, part hind and mid legs; (f) squamosa: head and palpi; (g)jJacobi:
head and palpi; (h) paludis: head and palpi; (#) paludis: hind tarsi;
(1) natalensis: head and palpi; (/2) natalensis: legs.

1 In specimens obtained from larve captured at sea level there is a broad
black band at the proximal extremity of the third hind tarsus, which has
not been found in specimens taken on higher levels. The photograph was
taken from a coast specimen.

CHARACTERISTICS OF LARVA OF ANOPHELINA. 157

Maegnifications.
Plates XV—X XIII, main figure x 30. (a) x 60. (b—f) x 180.
Plate XXIV (a—k) x 180; (l) x 3 (about). Plates XXV and XXVI,
all figures x 15.

Drawings from Immature Larve.
Plate X VITI.—(a) Lower figures; very small larve.
Plate X X.—(a) Two lower figures, one half, and three quarters grown.
Plate X XIII.—(a) Three lower, two one half, and one three quarters
erown.

o Seat

Dae "a a
irre tt eae >

Ann.Natal GMus Voll.

ff Fill, del.

C x180.

bx1so.

MYZOMYIA FUNESTA Gtles.

Huth Litht London.

=

\ a ie Ne Tia,
Vitus, VoL. eae VEG

&xeo.

E Fall del Huth Lith? London

PYRETOPHORUS COSPALIS Loew.

eo

VoLL.

fais

Gh

atal

V

Ann.)

axX60

xS3O

X180.

Cc

Huth Lith? London.

E.Hill, del.

PYRETOPHORUS CINEREUS Theobald.

Ann Natal G Mus. Vol 1. Pl XVII

E.Hill del.

ae SS

b x~180.

Huth Lith London,

PYRETOPHORUS ARDENS!S Power

Men Naval CG. Wie Voll: PLXIX

| &
ax6o.
x8oa
CxX180
bx180
E. Hill del Huth Lith? London.

NYSSORHYNCHUS PRETORIENSI!IS Theobald.

Ann. Natal G.Mus.Vol |. PR

' &X 60.

bx180.

E.Hill del. Huth Lith’ London.

CELLIA SQUAMOSA Theobald.

Ann.Natal G Mus Vol.1 Pi XX!

Cc, *180. C,x 180.
c,on first abdonunal c,on second abdonunal segment.
E. Hill del. Huth LithY London,

CEEEtA JACOB! 7usp:

Ann, Natal G Mus.Vol 1 .

fxiso. “

; bx18o0.
£ Hil ; Z 5 ;
peer a,a, postertor frontal hairs.

MYZORHYNCHUS PALUDIS Vheobald.

Huth Lith® L.ondon,

Ul

PLXX

G.Mus.Vol.L

Ann. Natal (

&x69.

bx180.

ith? London,

Huth L

E. Hill, del.

MYZORHYNCHUS NATALENSIS wy sp.

PaALMATE HAIRs. a.
d. P. ardensis. e.
jacobi. h.

Ann. Narat G. Mus., Vor. I.

Pin

Myzomyia funesta. 6. Pyretophorus costalis.

Nyssorhynchus pretoriensis. jf. Cellia squamosa. g. Cellia
Myzorhynchus paludis. k. Myzorhynchus natalensis. x 180.
/. Living larva ot Myzorhynchus paludis, showing palmate hairs spread out at the

surface of the water. x 3.

c. P. cinereus.

XXIV.

Ann. Narat G. Mos., Vou. I. Pipes

a
|" Wee ee

WINGS. a. funesta. b. costalis. c. cinereus. 4d. ardensis. e. pretoriensis.
7. squamosa. g. jacobi. h, k,l. paludis. m. natalensis. %X 15.

ddlard & Son, lpr.

Ann. Nara G. Mus., Vot. I. Prox Vie

PaLpl AND Leas. a. funesta. 6. costalis. c. cinereus. d. ardensis. e. pretoriensis.
J. squamosa g. jacobi. h,k. paludis. 7, lz, natalensis. x 15.

LANGUAGE OF COLOURS AMONGST THE ZULUS. 159

Language of Colours amongst the Zulus expressed
by their Bead-work Ornaments; and some
General Notes on their Personal Adornments

and Clothing.

By

?

Rev. Father Franz Mayr.

With Plate X XVII.

Ar the suggestion of Dr. Warren, Director of the Natal
Government Museum, I propose to place before the reader a
few notes on the personal ornaments used by Zulus, especially
with reference to the meanings assigned to the variously
coloured beads which are so generally used in certain of their
adornments.

Before speaking of the bead-work ornaments, now very
freely used by the Zulus in adorning their dusky bodies, it
will be interesting to indicate the kind of ornaments that
were formerly used before the advent of beads.

Ie

Previous to the arrival of the white man nature supplied
the Zulu with the materials for the following articles of
ornament :

(1) For adorning the head the Zulu man uses the well-
known head-ring (isicoco). Sewing fibre (uzi), obtained
from the bast of the wild fig and certain other trees, is
twisted to form a circlet, which is sewn into the woolly hair.

VoL. 1, parr 2. {2

160 REV. FATHER FRANZ MAYR.

This ring of fibre is then covered with the viscous material
(ungiyane) obtained from the sticky secretion of a scale-
imsect which hves on the thorn bush—Dalbergia obovata
(umzungulu). Afterwards it is polished with a pebble
until it shines like a well-polished boot.

The man with the head-ring is called ikehla. Formerly
it was the sign of attaining manhood, and it gave the young
man royal right to look for a wife. The time for putting on
the head-ring, known as ukutunga, was announced to the
young men by the Zulu king. To touch a man’s head-ring
disrespectfully was the greatest insult possible ; and formerly
such a deed was often revenged by putting the offender to
death.

Besides the head-ring, men of importance or warriors
would wear an ostrich feather ; each regiment would have a
uniform colour—all white, all black, ete.

Another head ornament which was, and still is, very much
sought after is a bunch of tail-feathers of the large Kafir
finch or long-tailed widow bird (isakabuli). The bunch of
feathers may cover almost the whole of the head, and is called
isidhlodhlo.

Headmen wore, and still wear, below the head-ring a circlet
of leopard, serval, otter, or other fur.

(2) As neck ornaments the men of the royal family wore
circlets of hon or leopard claws. Royal princesses wore stiff
collars of heavy, solid, brass rings, made by bending a brass
rod into a spiral of two or three turns. It was called
umnaka or ubedu. These collars must have been most
uncomfortable to wear, as the head could not be turned with-
out moving the whole body.

An ordinary Zulu women would frequently wear around
the neck a fibre string carrying a perforated brass ball
(indondo) about an inch in diameter. The ball would hang
about at the level of top of the sternum.

Sometimes a number of little sticks of the scented umtom-
boti tree (Excecaria africana) would be threaded on a
string to forma necklet ; or large beads made of scent-powder

LANGUAGE OF COLOURS AMONGST THE ZULUS. 16]

would be similarly threaded. The powder was prepared
from various scented plants, which were dried and pulverised,
and kneaded into balls.

The single brass ball ((ndondo) is no longer seen; but
the circlet of scented balls (amaka) and the necklet of
little sticks (ubande) of the scented umtomboti tree are
still in use.

(3) For adorning the arms and legs the young men used
the bushy ends of cattle tails (amatshoba), which were
fastened above the elbows and below the knees. As in
the case of the ostrich feathers, the various regiments were
distinguished by the colour of the tails

white, black, brown,
etc. Boys and girls satisfied themselves with grass wristlets
and anklets, but princesses were obliged to wear heavy brass
rings around the wrists and ankles (uamnaka or ubedu),
similar to those worn as necklets, only they were simple rings
without spiral turns.

(4) Lastly, for covering and adorning the body grass belts
(izifociya) were used, besides strips of skin of bucks or
domestic animals.

Married women used formerly, as at the present time, the
short petticoat (1sidwaba) of goat- or ox-skin.

Men wore the loin-dress (umutsha), consisting of bushy
tails in front (isinene) anda square or oblong piece of ox-
or buck-skin behind (ibetshu). In place of the bushy tails
strips of buck-skin were often used.

Grown-up girls clothed themselves with the ubendhle,
which was a tringed loin-covering encircling the body. It was
inade from the veld plant Gazania longiscapa.

Witch-doctors were distinguished by their iminqwambi,
which were strips of skin worn over the shoulders, and
fastened together at the middle, before and behind, something
like a pair of braces.

For the great umkosi, the Zulu king’s annual festival, the
warriors wore the state dress, consisting of three girdles or
kilts of ox-tails. One was worn low down over the buttocks,
another above the hips, and the third over the shoulders. In

162 REV. FATHER FRANZ MAYR.

this way the body was entirely covered from neck to knee.
This state dress was called imiqubula.

AT:

From the brief account given above we see that there has
always been a great variety of different dresses and orna-
ments, distinguishing royalty, warriors, witch-doctors, and
common folk; but when the European brought glittering
coloured beads the sympathy of the Zulu was at once aroused,
and he found a new field for his imagination and skill.

Beads were first brought by the Portugese, then by the
Dutch, and now by the English. The natives show very con-
siderable skill and taste in making ornaments and designing
patterns. There are established colours and kinds of beads in
use among the Zulus, and they will not look at any other kinds,
however pretty they may be, which are not established by
traditional use. Traders soon discover this peculiarity, and
take care to only have in stock the kinds of beads and colours
which are hked by the natives.

The natives have given each colour of beads a special name
and meaning; and they have invented a kind of language of
colour, whereby they can convey their thoughts from one to
another without speaking. How this is done will best be
explained by reference to the illustrations of bead work given
in the plate (Pl. XX VII) attached to this paper; but I will
first give a list of the names and literal meanings of the
different coloured beads :

White beads. = itambo (lit. bone).

Black beads. = isitimane (name of a_ regiment
formed by Mpande ; also, nick-name
for a very black person).

Blue beads : = 1juba (lit. dove).
Red beads = umgazi (from igazi, blood).
Yellow beads . = incombo (lit. young Kafir-corn still

yellow) or ipuzi (lit. bright yellow

.

native pumpkin).

LANGUAGE OF COLOURS AMONGST THE ZULUS. 165

Green beads = ubuhlalu obuluhlaza (from ulu-
hlaza, new grass).

Striped beads. = intotoviyane (lit. large striped
grasshopper).

Pink beads : = ubuhlalu obumpofu(fromimpofu,

poor, poverty).
Transparent brown

beads : = umlilwana (lit. a low fire).
Dark blue beads = inkankane (lit. Common Ibis).

Large-sized beads of

any colour. =amapohlo or amaqanda (ama-

qanda means eges).

Anyone uninitiated in the secret meaning of the different
beads, and seeing, for example, a number of white beads
followed by a few red, green, blue, and black, and then again
white, red, green, blue, and black in the same succession and
the same number of each kind of bead, might think that the
arrangement was simply for the sake of ornament and sym-
metry. But a kraal-native would say it was a letter, and
would call it so, i.e. inewadi, or better, ubala abuyise,
which means “one writes in order that the other should

reply.”

In this way an uneducated Zulu girl will present her
sweetheart with an ubala abuyise, and will expect his visit
mm return,

A variety of different bead ornaments are used as letters in
this way, the chief of which are illustrated in the accom-
panying plate.

Fig. 1, ingeje, a single bead string.

Fig. 2, umampapeni, one square of beads with one or
more bead strings.

Fig. 3, ulimi (lit. tongue), one long oblong piece of bead-
work with one or more bead strings.

Fig. 4, igcagcane, a necklet consisting of a number of
connected small squares of bead work.

These illustrations are taken from actual specimens in the
Natal Government Museum, and the colours represent, asnearly
as possible, the favourite shades established by custom.

164 REV. FATHER FRANZ MAYR.

The simplest form of ubala abuyise, or letter, is an
ingeje, or single bead string (fig. 1), consisting of only two
kinds of beads. In this specimen one half of the string con-
sists of white beads, and the other half of pink. The mean-
ing of this would be, that the girl’s heart is full of love (white
beads), but she tells her lover by the same number of pink
beads that his poverty is as great as her love towards him.
This implies, of course, an earnest appeal to him to work hard
in order to get the cattle for the lobola, or payment, neces-
sary to buy her from her father or guardian.

To express wealth yellow beads are used.

The remaining figures in the plate represent more com-
pleated letters; and a succession of white, red, green, blue
and black beads, for example, would be interpreted in the
following way :

Inhliziyo yami imhlope ngezinsuku ezikude (ex-
pressed by the numerous white beads). Kodwa sengi-
kubeke, amehlo ami az’abomvu (red beads). Nang’-
ubala abuyise, kepa sengizacile ngaza nga’luhlaza
(green beads). Uma bengi | ijuba, bengizaundiza
ngiyocotsha emyango kwenn (blue beads). Ngivin-
jelwe ubumnyama ngingeze ngaya kuwe (black beads).
Kodwa inhliziyo yami imhlope, ete. (white beads—the
same story 1s repeated again and again).

The Zulu idiom can never be expressed in English, and
in a translation the letter loses much of its force; but it
would run in some such fashion as the following. My heart
is pure and white in the long weary days (white beads). My
eyes are sore and red by looking out for you so long (red
beads). Nang’ubal’abuyise = here is my letter to you. I
have become quite lean and sickly (green beads). If I were
a dove I would fly to your home and pick up food at your
door (blue beads). Darkness prevents my coming to you
(black beads). But my heart is pure, etc., etc., and the
whole message would be repeated several times.

The actual pattern does not appear to have any defined
significance; it is rather the succession of colour, and the

LANGUAGE OF COLOURS AMONGSY! THE zULUS. 165

relative amounts of the different colours, that express the
tenor of the message.

In reading a letter of the umampapeni (fig. 2) and
ulimi (fig. 3) types, the string which passes round the neck,
beginning at the fastening, is taken first ; and on the whole
the string has the greater significance. In the case of the
square and oblong piece the letter would be read from
without inwards, but the edging or border is for the sake of
ornament simply, and, as a rule, has no special meaning. In
a string the number of successively placed yellow beads
may indicate the number of cattle owned by the recipient of
the letter.

Having thus obtained a key to the meanings assigned to
the colours, it is an easy matter to interpret any of these
epistles; and it may suffice to say that the Zulu lad is very
proud of them, and hangs them all round his neck and head
in order to show everyone how much he is loved by one or
a number of girls.

EXPLANATION OF PLATE XXVII,
Illustrating Rev. Father Mayr’s paper, “ Language of Colours
amongst the Zulus, etc.”

The illustrations are reduced to five-eighths of the actual size.

Fie. 1.—Ingeje. The tenor of the letter is that the girl realises that
the man is poor (pink), and she asks him to work for cattle, as she is in
love (white) with him.

Fie. 2—Umampapeni. The purport of the letter is that the
recipient is well to do (yellow), and the girl is weeping (red) on account
of his not going to her.

Fic. 3—Ulimi. The general tone is that the girl is greatly in love
(numerous white beads), and she thinks the man is sufficiently rich
(yellow) to marry her. The blue border at the bottom is pure ornamen-
tation, without special significance.

Fie. 4—Igeageane. The general tone is somewhat distressing.
The girl is fond of the boy (white), yet there is difficulty in going to him
(black), as he is poor (pink), and consequently she feels lean and sickly
(green).

ANS bf

7

ca

amare HL

bok
a
<

C66 o®

>
pt Tees o oe
o§ac8 al

“sw

{
oO =
Y

SAO UAMOTOrpe8s 00 i) > 6 CRM OMe 9.55 45 9 9% o®

RETEUEEAUEEEO LACE DMLLULAUISMSORICUUCIER US COESILEPCATECCOCIUCL SG TUEETINE IECTEECCOIGSETETCRVEPeSE ICEL RITIER SO LSEE Ut TT atte eisuveriyh mr rtrestner) LL. SesnvecunE SLO TiC cD Ls, a ool nan ita en aaa

TWO NEW REPTILES FROM THE KARROO BEDS OF NATAL. 167

On Two New Reptiles from the Karroo Beds of
Natal.

By
R. Broom, ™..D., D.Sc., C..I.Z.S.
Victoria College, Stellenbosch.

With Plate XXVIII.

THouaH for many years reptilian bones have been known
to occur in the Karroo beds of Natal, until a few months
ago nothing definite has been known concerning them. Mr.
William Anderson, lately government geologist of Natal,
recently submitted to me a small collection of reptilan
remains for examination. Though most of the specimens
were extremely fragmentary, those that could be identified
were all Anomodonts, and for the most part belonged to the
genus Lystrosaurus. <A second collection has just been
sent me for examination by Dr. Warren, and comprises speci-
mens belonging to the Government Museum, Pietermaritzburg.
Though the collection is small, and many of the specimens
very imperfect, almost all the fragments can be identified,
and the collection proves to be of considerable interest.
Most of the specimens belong toa large Dicynodon, which
proves to be a new species, and there is also a specimen of a
new species of Therocephalian. These two new species |
have much pleasure in describing. Three Natal species are
now known, and though they belong to well-known genera
they are all very distinct from the species known in Cape
Colony. This isa little remarkable, considering that the forms
from the Western Karroo beds are identical with those from
similar horizons in the Hastern Province. Very likely the

16

CO

R. BROOM.

Natal species may yet be found in Cape Colony when the beds
are more fully worked.

Dicynodon ingens n. sp. Pl. XXVIII, figs. 1-4.

This new species of Dicynodon is represented by a con-
siderable portion of the snout of an animal which is perhaps
an adult female; by the imperfect front portion of the skull
of an immature animal; by four tusks, two probably of adult
males and two of females; and by a number of fragments of
limb bones. The snout of the supposed adult female is taken
as the type. Of species already known only three at all
resemble the Natal form. Theseare D. leoniceps, D. simo-
cephalus, and D.latifrons. Neither D. leoniceps nor
D. latifrons have ever been discovered of nearly such large
size as D. ingens, and there are marked differences.

The specimen which is taken as the type is the fairly
well preserved palatal portion of a large skull, compris-
ing the maxillary bones, the pre-maxillary bones, most of
the vomer, and parts of the palatines. The tusks are
broken off at their bases. In general appearance the palate
resembles that of D. leoniceps in being much longer than
broad. Unfortunately the palate of D. leoniceps has not
been cleared of matrix, and it is impossible therefore to
institute detailed comparisons. The tusks in D.ingens are,
however, very much longer than in D.leoniceps. At the
point where broken across in the type specimen the tusk
measures 40 mm. in diameter, and is as near as may be
circular in section. The distance between the tusks is 82 mm.,
but there is a slight degree of lateral crushing. From the
line joining the two tusks to the front of the snout the dis-
tance is about 150 mm. At the hollow immediately in front
of the great dilatation caused by the tusk the transverse
measurement of the snout is 120 mm., and at the tusk region
probably 190 mm. The median ridge is very prominent, and
extends from about 35 mm. in front of the plane passing
along the anterior border of the tusks to near the plane pass-

TWO NEW REPTILES FROM THE KARROO BEDS OF NATAL. 169

ing along the posterior border. In front the palate has the
usual two premaxillary ridges well developed. The front of
the snout is unusually straight. It has alow median ridge
and a shght Jateral one on each side. The nostril is of large
size, and probably is within 50 mm. of the edge of the jaw.
Antero-posteriorly it measures at least 80 mm., and the lower
border is moderately straight.

There are tusks from the same locality, and most probably
of the same species, which are very much larger than those of
the type specimen. This is probably due to their being tusks
of males. One specimen measures at the root 57 mm. x 45 mm.
Tt is considerably crushed. Another specimen, which is also
crushed, measures 59 mm. x 43 mm.

The type specimen was discovered at Ennersdale, Natal, by
Mr. A. 8. Woodgate.

Scymnosaurus warreni n.sp. Pl. XXVIII, figs. 5-7.

The Therocephalian specimen is the snout of a moderate-
sized reptile. It is considerably crushed from above down-
wards, but otherwise is in good preservation. The front
part of both mandibles is preserved, and the right one has
been detached from the skull and cleared of matrix. The
animal has manifestly had a broad, yet deep, snout, while
probably the back part of the skull was relatively short.

The nasal bones are broad both in front and behind. Pos-
teriorly the two probably measure 45 mm. across, and in front
slightly more. The greatest length of the nasals is 96 mm.
apparently, but it is a little difficult to be sure of the pos-
terior suture. Superficially the bones are shghtly rugose.

The premaxillaries are well developed, but, as in other
Therocephalians, considerably covered over by the maxil-
laries. The inter-nasal process is relatively short and broad.
The extra-nasal process is larger and broader than usual.
The nostril measures about 17 mm. across, and the extra-
nasal process, so far as it is visible in front of the maxilla,
is 15 mm. wide. The depth of the premaxilla below the

170 R. BROOM.

nostril is about 16 mm. From the median suture to the
anterior part of the maxilla is 30 mm. There appear to be
five teeth in front of the large canine. On the right side
five teeth are visible, buta fairly wide gap is present between
the first and second. On the left side three teeth are seen,
and wide gaps are present before and behind the third.
If we look at the alveolar margins, however, we find reason
to believe that there are only five teeth in front of the large
canine. All five are, | believe, incisors, though it must
be borne in mind that in some Therocephalians the small
tooth in front of the large canine is also a maxillary tooth,
and therefore a canine. The first incisors of each side are
fairly equal in size, both measuring about 9 mm. xX 6 mm.
The second incisor is much smaller on the left side than on
the right, the latter measuring 10 mm. x 6 mm., and the
former 6mm. x 5mm. The third incisor of the right side
measures 10 mm.x 7mm. The fourth incisor of the right
side measures 6 mm. X 5 mm.; that of the left side 9mm. x
55mm. The fifth incisor measures 6°55 mm. x 5mm. The
differences in the sizes of the teeth on the two sides is pro-
bably owing to the fact that in Therocephalians there is
apparently a continuous succession of teeth, a new one always
replacing an injured or worn one. ‘The measurement in a
direct line from the front of the first incisor to the front of
the canine is 43 mm. on both sides. In no case is the crown
of the tooth preserved.

The maxillary bone has a very convex outer surface and
is very rugose, especially at the lower and anterior margins.
Its greatest depth is 50 mm. and the greatest length
probably 85 mm. The canine tooth measures 16 mm. x
12 mm. on both sides. Behind the canine are three small
molars. The first is 8mm. behind the canine, it measures
5 mm. x 35 mm. The other two are smaller. The three
together measure 15 mm.

From the back of the nostril to the front of the orbit
is 74 mm.

The anterior part of the dentary is strongly developed,

TWO NEW REPTILES FROM THE KARROO BEDS OF NATAL. 171

and bears four incisors, a large canine, and probably three
small molars. Fortunately the incisors and canines have
been extracted from the matrix. so that their crowns can be
examined. The first incisor is 19 mm. in height, 8 mm. at
the base from without in, and 5 mm. from side to side. On
passing upwards it curves shghtly backwards like a bird’s
claw, and ends ina somewhat obtuse point. Along each side
in the upper two thirds of the tooth runs a low ridge, which
curves with the tooth, but passes more to the inner side on
approaching the apex. On the front of the upper part of
the tooth are four faint ridges, and on the concave side are
also four ridges shehtly more marked. There is no evidence
of any serrations on the tooth. The second, third, and fourth
teeth are apparently all similar to the first, but they are all
more or less worn, and the ridges less distinct. The teeth
decrease in size from the first to the fourth. The fourth
measures 8 mm. in height. ‘The four incisors measure
together 20 mm. The canine is large and very like the
incisors in structure. There is a feeble ridge in front and
behind, and on the concave side five faint ridges. No ridges
are seen on the outside of the tooth, but this may be due in
part to wear. ‘The tooth measures in height probably 30 mm.
Antero-posteriorly it measures 11 mm., and in the other
direction 10 mm. There are probably three molars, but only
two are preserved, the first close against the canine, and the
supposed third 16 mm. from the canine. Though the jaw is
probably pretty deep in front, it rapidly becomes narrow, and
at the third molar it measures only 21 mm.

The dental formula of the species is i3c!m3. This is
probably identical with the formula of the specimen des-
cribed from Cape Colony as Scymnosaurus ferox Broom,
and for this reason I have provisionally placed the Natal
specimen in the same genus. The differences in the dentition
are sufficiently great to make it anew genus if it were a
mammal, but it seems better in the meantime to make the
reptilian genera rather comprehensive.

It is unfortunate that the locality of the Cape specimen

a R. BROOM.

of Scymnosaurus is unknown, but as it is a rather highly
specialised type of Therocephalian, it seems likely that it is
from the upper division of the Lower Beaufort beds. The
close resemblance of Seymnosaurus to the Russian genus
Inostranzewia (Amalitzky) is worthy of note. They are
apparently distinct, but closely allied.

I have much pleasure in naming the Natal specimen after
Dr. E. Warren, of the Government Museum, Pietermaritz-
burg.

The specimen was discovered at the Little Tugela River
by Mr. P. Paterson.

EXPLANATION OF PLATE XXVIII,

Illustrating Dr. R. Broom’s paper “On Two New Reptiles
from the Karroo Beds of Natal.”

Fie. 1.—Plate of Dicynodon ingens. x °3.

Fie. 2.—Side view of portion of skull of Dicynodon ingens. x3.

Fie. 3.—Canine of Dicynodon ingens. X ‘28.

Fia. 4.—Canine of Dicynodon ingens. X ‘28.

Fie. 5.—Under view of portion of skull of Seymnosaurus warreni.
x 56.

Fia. 6.—Side view of portion of skull of Seymnosaurus warreni,
slightly restored. x ‘56.

Fie. 7—Upper view of front of right mandible of Seymnosaurus
warreni. Nat. size.

Ann. Natal G. Mus. VoLL PL XXVIT.

Fig. x0-56.

Fig.6 xo-56.
Broom del Huth Lith? London
DICYNODON INGENS SBroom.Fig® 1-4.
SCYMNOSAURUS WARRENI Broom. Fié5 5-7,

a

a |
we
ae

ENTOMOSTRACA COLLECTED IN NATAL. ]

On Entomostraca Collected in Natal by
Mr. James Gibson. (Part II.)

By
G. Stewardson Brady, i.D., LL.D., D.Sc., F.R.S., C.M.Z.S8.

With Plates XXIX—XXXITI.

Axsour two years ago I published in the ‘ Proceedings of
the Zoological Society of London’ a short account of some
Natal Entomostraca, the material for which was kindly sent
to me by Mr. James Gibson. I am now indebted to him for
further gatherings from the same district. Many of these
have been already described by other authors, notably by
Professor G. O. Sars and Dr. G. W. Miiller, references to
whose papers are given under the heads of the various
species; others appear to be new. The new species I here
briefly describe and figure; I also figure, sufficiently I think
to ensure proper identification, some of the previously known
species, as far as my material enables me to do so. Some
others, e.g. Cypretta sarsi, Oncocypris voeltzkowi,
Broteas lamellatus, and Diaptomus orientalis, have
already been abundantly figured.

OsTRACODA.
Genus Cypris Miller.
Cypris intumescens n. sp. Pl. XXIX, figs. 1-5.

Shell seen laterally subreniform, greatest height in front
of the middle, and equal to about two thirds of the length
(fig. 1). Anterior extremity broadly rounded, posterior much
narrower and rather obliquely rounded, dorsal margin

174 G. STEWARDSON BRADY.

almost gibbous in front of the middle, thence sloping steeply
to the front and much more gently backward, ventral margin
almost straight. Seen from above (fig. 2) the outline is very
broadly ovate, obtusely pointed in front, rather abruptly
rounded behind, lateral margins evenly arcuate; greatest
width situated behind the middle, and equal to at least two
thirds of the length. Surface of the shell perfectly smooth,
without any trace of sculpture, the free margins fringed with
fine hairs; colour pale green. Length, 1°35 mm.

The swimming sete of the posterior antenne reach as far
as the apices of the terminal claws; margins of the labial
plates (fig. 3) very strongly toothed ; principal spines of the
first pair of maxille quite destitute of marginal processes ;
caudal rami (fig. 5) extremely slender, having two slender
apical sete, marginal seta very small, and not far removed
from the apex; genital plates well developed, the convoluted
spermathecal (?) tubes very large and conspicuous (fig. 4).

One example only of this species was seen; it occurred in
a gathering from Somkele, Zululand.

Genus Cypretta Vavra.
Cypretta sarsi G. S. Brady.

1901. Cypretta sarsi Brady, “On new or imperfectly-
known Ostracoda, chiefly from a Collection in the Zoological
Museum, Copenhagen” (‘ Trans. Zool. Soc., London,’ vol. xvi,
pi 1v), p. 1907 Pl XekV,, tres; 10K:

The type specimens were from St. Thomas, West Indies.
Those now noted were from Pietermaritzburg, m which
oathering it occurred abundantly.

Genus Cypria Zenker.
Cypria armata G. W. Miller. Pl. XXIX, figs. 6-11.

Cypria armata Miller ‘Die Ostracoden (Ergebnisse einer
Zoologischen Forschungsreise in Madagaskar und Ost-A frika,’
1889-1895), p. 261, Taf. 13, figs. 1-5, 12.

ENTOMOSTRACA COLLECTED IN NATAL. 175

Shell seen from the side oblong, subovate, nearly twice as
long as high (fig. 6), greatest height in the middle, anterior
extremity rounded off gently above but somewhat abruptly
below, posterior evenly rounded, dorsal margin well and
evenly arched, ventral slightly protuberant in the middle,
otherwise almost straight. Seen from above, the outline is
ovate, more than twice as long as broad, widest in the middle,
much compressed, posterior extremities rounded off, anterior
scarcely rounded, subacuminate (fig. 7), colour more or less
brown, surface smooth. Length, 1‘°2 mm. Swimming sete,
as usual in this genus, reaching very far beyond the ex-
tremity of the antennal claws (fig. 8) ; caudal rami very stout
(fig. 11), their terminal claws stout and moderately curved,
marginal seta long, attached a little beyond the middle of the
limb.

Hab.—Pietermaritzbure.

These specimens agree in all respects with those described
by Dr. G. W. Miiller from Madagascar.

Genus Cypridopsis Brady.
Cypridopsis punctillata n.sp. Pl. XXXI, figs. 7-15.

Shell seen from the side oblong, reniform, greatest height
situated in the middle and equal to rather more than half
the length (fig. 7), extremities evenly rounded, the anterior
rather the broader of the two, dorsal margin very boldly
arched, sloping gently toward the front, steeply arcuate
behind, ventral margin distinctly sinuated in the middle ;
seen dorsally the outline is ovate, about twice as long as
broad (fig. 8), tapermg toward the front, which is bluntly
pointed, wider and almost subtruncated behind. Surface
smooth, marked uniformly with very small, impressed puncta-
tions (fig. 11). Antero-ventral angle of the left (and in
some cases also of the right) valve very minutely denticu-
lated, and fringed with slender hairs (fig. 9), colour greenish.
Length 0°57 mm. Swimming sete of the posterior antenne

VoL. 1, PART 2. 13

176 G. STEWARDSON BRADY.

reaching much beyond the extremities of the terminal claws ,
branchial plates of the mandibles bearing only five or six
sete ; caudal rami simple, setiform (fig. 10).

Hab.—Pietermaritzburg.

This is a very small species presenting all the characters
of a typical Cypridopsis—somewhat lke C. newtoni in
general character—not nearly so compressed as C. villosa
nor so pubescent.

Genus Proteocypris Brady.

Proteocypris Brady. ‘Trans. Northumberland and Durham
Nat. Hist. Soc.’

This genus does not differ materially from Cy pridopsis
except in its bisexual character.

Proteocypris reniformis sp.n. Pl. XXX, figs. 1-9.

Shell seen laterally reniform, highest in the middle, height
equal to more than half the length (fig. 1), anterior margin
broadly rounded, and produced at the ventral angle so as to
form a rather prominent lip; posterior forming a steep curve
and rounded off at the ventral angle, dorsal margin boldly
and evenly arched, ventral sinuated in the middle; seen
dorsally the outline is compressed, ovate, acutely pointed in
front, rounded off behind, greatest width situated behind the
middle and equal to less than half the length (fig. 2), shell
smooth, shghtly pubescent, marked with rather coarse,
closely-set oblong or angular pittings (fig. 9). Length
0-5 mm. Natatory sete reaching far beyond the extremities
of the terminal claws (fig. 3); mandibular palp rather short
and stout (fig. 4); apices of posterior feet simply setiferous,
not forcipate; caudal rami (fig. 7) simple, setiform. Hjacu-
latory duct of male bearing about twelve whorls (fig. 8).

Hab.—Somkele.

ENTOMOSTRACA COLLECTED IN NATAL. Zz.

Proteocypris (?) globuloides sp.n. Pl. XXX, figs. 10-18.

Shell seen laterally subovate, height equal to two thirds of
the length (fig. 10), anterior extremity broadly rounded,
posterior somewhat narrower and more flattened, dorsal
margin well arched, highest in the middle, ventral shghtly
sinuated. Seen dorsally the outlme is very broadly oval,
almost circular, the width only shghtly less than the length
(fig. 11), natatory setee of the posterior antenne not reaching
beyond the terminal claws (fig. 12). Caudal rami simple,
bearing two apical sete (fig. 17). In other respects the
animal is essentially similar to P. reniformis. Shell quite
smooth, with a few marginal hairs, and showing by trans-
mitted light a rather coarsely reticulated structure. Colour,
brown. Length 0°45 mm.

Hab.— Richmond.

‘his is a minute species, easily distinguished from the pre-
ceding by its very tumid outline, and on dissection, by the
shorter length of the swimming setz, the character of the
caudal rami, and, in the male, by the more robust build of the
prehensile second maxille.

Genus Stenocypris G. O. Sars.
Stenocypris GO. Sars. ‘On some Freshwater Ostracoda

and Copepoda raised from Dried Australian Mud’ (Chris-
tiania, 1889).

Shell much compressed and elongated, valves subequal
“free edges smooth, inner duplications very large, especially
at the anterior part. Natatory sete of the posterior antenne
not reaching beyond the terminal claws. Palp of the first
pair of mawxille narrow, cyhndrical, last jomt small, masti-
catory lobes long and narrow. Caudal rami rather large,
more or less lamellitorm, dorsal edges sometimes pectinate,
claws very unequal, both coarsely denticulate, seta of dorsal
edge absent or very small, the apical one rather elongate.
Propagation exclusively parthenogenetical.”

178 G. STEWARDSON BRADY.

Apart from the structure of the shell, to which one need
not perhaps attach very great importance, the principal dis-
tinctive characters of this genus, as described by Sars, are
“ parthenogenetical ” reproduction, the unequal and pecti-
nated claws of the caudal rami, and the rather abnormal
build of the first maxillar palps. G. W. Miiller, however,
adds to these characters the asymmetrical structure of the
two caudal rami, though this want of symmetry is not found
in the types previously described—S. malcolmsoni and 8.
chevreuxii. A still more important divergence is the
bisexual character of S. sinuata as described by Miller.
In this species the two sexes are fully developed, so that the
“ exclusively parthenogenetical”’ character ascribed to the
genus by Sars cannot here apply. But no males have been
found in the case of S. aldabre, and in this species Miller
states that the receptacula seminis are always empty.
This accords with my own observation. Still, I find it diffi-
cult to believe that these organs should exist, as they do in
many true Cypridz, apart from any possible functional use,
and it seems to me not unlikely that some day or other males
may be found in species which we are at present constrained
to call “‘ parthenogenetical.” The present generic allocation
of the various species must be looked upon as provisional.

Stenocypris aldabre G. W. Miller. Pl. XXXII, figs. 1-6.

Stenocypris aldabre Muller, loc. cit., p. 275, Pl. VIL,

Shell seen laterally elongated, siliquose, highest in the
middle, height equal to less than half the length ; anterior
extremity well rounded, posterior sloping steeply and almost
in aright line to the ventral angle, which is rounded off ;
dorsal margin evenly arched, rather flattened, ventral sinuated
in the middle, otherwise straight (fig. 1); seen dorsally the
outline is much compressed, elongated, fully four times as
long as broad, with acutely pointed extremities (fig. 2).
Surface of the shell quite smooth; colour, grey. Length,

ENTOMOSTRACA COLLECTED IN NATAL. 179

23. mm. Swimming sete of the posterior antenne (fig. 3)
reaching *to the extremity of the terminal claws ; apices of
the last feet forcipate (fig. 4). Caudal rami bearing two
strong apical claws and one short seta (figs. 5 and 6), no
marginal seta, inner margins of the claws coarsely pectinated,
distal portions of the rami spinulose on the dorsal margins,
the spines of the left limb small, those of the right side much
larger, and arranged in progressively graduated groups of
three or four in each.

Hab.—Pietermaritzbure.

Only a few specimens—probably not fully grown—were
seen. The measurement given by G. W. Miiller is 2°9 mm.
to 3°5 mm. of specimens from Madagascar.

Genus Oncocyrris G. W. Miller.
Oncocypris voeltzkowi G. W. Miller.
Oncocypris voeltzkowi Miller, ‘Die Ostracoden,
Voeltzkow Reissergebnisse,’ p. 288, Pl. I, figs. 1-18.
Plentiful in gatherings from Richmond and Pietermaritz-

bure.

Cladocera.
Genus SmocePpHaLus Schoedler.

Simocephalus capensis? G.O. Sars. Pl. XXXII, fig. 8,

? Simocephalus capensis Sars. ‘On some South African
Entomostraca raised from Dried Mud’ (1895), p. 15, Pl. III.

This species occurred plentifully in a gathering from
Richmond. I ean scarcely doubt that the specimens belong
to Sars’ species, though they seem to differ shghtly in the
slightly spinulose character of the himder margins of the
valves, whereas Sars remarks that it is “nearly allied to the
European species S. serrulatus Koch, but differs in the
somewhat different form of the head, and in the circum-
stance that the hind edges of the valves, below the post-
erior prominence, are quite smooth, not, as in that species,
serrulate.” But different stages of growth are accountable

180 G. STEWARDSON BRADY.

for much difference in minor distinctive characters, and it
seems very likely that the specimens here noticed are not
quite fully grown. The most characteristic form of the
Richmond specimens is represented in fig. 8.

Genus Ceriodaphnia Dana.
Ceriodaphnia natalis sp.n. Pl. XXXII, figs. 3-7.

Carapace of female seen laterally tumid, rounded, sub-
quadrangular, postero-dorsal angle much produced and
sharply pointed, both dorsal and ventral margins strongly
convex (fig. 5), seen dorsally (fig. 6) the outline is ovate,
rather acutely pointed behind, about twice as long as broad ;
shell marked throughout with minute and closely-set circular
punctations ; antennules short, club shaped; eye moderately
large. Post-abdomen with a small dorsal process, terminal
portion rounded and rather wide, anal spinules eight or nine on
each side, gradated in size from the middle to each end of the
series (fig. 7), terminal claws bearing four short spinules near
the base; head much depressed, separated from the body by
a deep dorsal depression, rounded and prominent inferiorly.
Length, 0°77 mm.

The carapace of the male (fig. 38) seen laterally is much
narrower, and almost quadrangular, the dorsal and ventral
margins almost straight and sub-parallel; eye rather larger
than in the female; antennule large, bi-articulate, projecting
beyond the head, bearing a large setiform flagellum and a
brush of about four small sensory sete. Length, 0°66 mm.

This species was found rather plentifully im a netting from
Richmond. It seems to be quite distinct from any other
described form.

Genus Atona Baird.
Alona glabra G. O. Sars.

Alona glabra Sars, ‘Fresh-water Entomostraca of South
America,’ pt. I “ Cladocera,” p. 55, Pl. IX, figs. 6, 6a.

ENTOMOSTRACA COLLECTED IN NATAL. 181

Alona guttata G. O. Sars. Var. parvula Kurz.
Sars lee. cite, p. oly Pl. LX, figs. 3,54.

Both of these Alone occurred in the Richmond gatherings.

Genus A1noneLiA G. O. Sars.
Alonella eclathratula G. O. Sars.

Alonella clathratula Sars, loc. cit., p. 62, Pl. X,
figs. 5, 5a.
Taken at Richmond with the two species of Alona.

Alonopsis? sp., Pl. XXXI, figs. 14, 15.

A species probably referable to this genus occurred in the
Richmond nettings. Only one specimen was seen.

Genus Leydigia Kurz.
Leydigia propinqua G. O. Sars, Pl. XXXI, figs. 12, 13.

1893. Leydigia acanthocercoides Sars on ‘South
African Entomostraca raised from Dried Mud,’ p. 18, Pl. IV,
figs. 1-4.

1903. Leydigia propinqua Sars, ‘Fresh-water Ento-
mostraca from China and Sumatra,’ p. 14 (separate copy),
Pl. I, figs. 4, 4a.

1904. Leydigia africana Gurney, “ Fresh-water Ento-
mostraca trom South Africa” (‘Proc. Zool. Soe. Lond.’),
p. 300) Pl. XVIIL, figs. 5; 6:

Professor G. O. Sars remarks respecting this species:
‘“ Having now, through the kindness of Professor Lilljeborg,
had an opportunity of examining Swedish specimens of the
true L. acanthocercoides Fischer, I find that the South
African form described by me under this name is in reality
specifically distinct, for which reason I propose for it the
above name. It differs not only in the shape and sculpture
of the carapace, but also in the much smaller size of the
ocellus, which scarcely exceeds that of the eye, whereas in

182 G. STEWARDSON BRADY.

L. acanthocercoides it is considerably larger. Moreover,
the caudal part is somewhat less broad, and the terminal
claws have each at the base a very small denticle, omitted in
my previous figure.

Two specimes of L. propinqua were seen in the nettings
from Richmond. These are undoubtedly identical with Sars’
species, and are here figured (Pl. XX XJ, figs. 12,13). I think
that Mr. Gurney’s L. africana also is referable to the same
species.

Genus Pievroxus Baird.
Pleuroxus assimilis sp.n. Pl. XXXII, figs. 9, 10.

Body of the female seen laterally broadly subovate, broadly
rounded in front, abruptly narrowed and truncated behind,
height equal to nearly three fourths of the length (fig. 9) ;
anterior extremity produced below the middle into a long,
slender, acutely-pointed rostrum, posterior very narrow,
almost rectilinear, notched twice or thrice near the ventral
angle, dorsal margin very strongly arched, shghtly sinuated
behind, where it forms a sharp postero-dorsal angle; ventral
margin boldly convex in front, where it turns abruptly up-
wards, forming a deep sub-rostral sinus. Surface of the
valves faintly marked behind the middle with flexuous longi-
tudinal striz, plain in front. Ocellus smaller than the eye,
and situated nearly half way between that organ and the
point of the rostrum. Post-abdominal lamina with a pro-
duced and rather broadly rounded post-anal angle, of nearly
uniform width throughout (fig. 10), marginal spines nume-
rous, about twenty on each lamina, those of the distal angle
long and curved, the rest short and nearly equal; terminal
claw strong and deeply coloured, with two spinules at the base,
one long and one short. Length, 0°5 mm.

Hab.—Richmond. Only two specimens seen.

ENTOMOSTRACA COLLECTED IN NATAL. 183

Genus Cuyporus Leach.
Chydorus gibsoni sp.n. Pl. XXXII, figs. 1, 2.

Carapace seen from the side subglobose, broadly rounded
in front (fig. 1), narrower behind, dorsal and ventral margins
both strongly arcuate, no posterior marginal denticle, width
equal to more than three fourths of the length, rostrum curved
and acutely pointed, lip-plate rather prominent (fig. 1a),
smooth, without any trace of serration; ocellus considerably
smaller than the eye; post-abdominai lamina narrowed and
deeply sinuated distally (fig. 2), bearmg about ten small,
sharp, rather distantly-placed denticles; terminal claw
smooth, with two basal spinules, one long and one short.
Length, 0°5 mm.

Several specimens of this form occurred in the Richmond
gatherings. Its nearest allies appear to be Chydorus
barroisi Richard and C. poppei Richard. It may, how-
ever, be distinguished from the former by the absence of
any posterior marginal denticle, by the perfectly smooth
“lip-plate,’ and by the unsculptured carapace—from the
latter by the number and character of the caudal spimules
and the want of shell sculpture.

Copepoda.
Genus Broreas Loven.
Broteas lamellatus (G. O. Sars).

1895. Paradiaptomus lamellatus Sars. ‘On some South
African Entomostraca raised from Dried Mud, p. 46
(separate copy), Pls. VII and VIII.

Hab.—Richmond.

Professor Sars, in his paper “On the genus Broteas of
Loyén,” rejects his previously described genus “ Paradiap-
tomus” in favour of Lovén’s Broteas, which claims priority
of date. The type species B. falcifer, as well as B. lamel-
latus, is native to South Africa.

184

Frequent in gatherings from Somkele and Richmond.
This species was described many years ago from specimens
taken in Ceylon,! and has been since seen by other observers
in various places.

G. STEWARDSON BRADY.

Genus Diapromus Westwood.

Diaptomus orientalis G. S. Brady.

species raised from dried Australian mud.

Fig
Fig
Fie
Fie
Fig

Fie
Fic
Fie

ID,
HIG.
Fia.

Fia.
Fre.
Tes
Fia.
Fie.
1pieh
IDEs 7
Fia.
Fig.

Professor G. O. Sars has noticed it among

EXPLANATION OF PLATES XXIX—XXXII,

Illustrating Dr. G. S. Brady’s paper “On Entomostraca

Collected in Natal by Mr. James Gibson.”

PLATE XXIX.
Cypris intumescens n. sp.
1.—Shell seen from right side. x 40.
2.—Shell seen from above. x 40.
3.—Labrum. x 240.
4.—Genital plate. x 240.
5.—Caudal ramus. x 100.

Cypria armata G. W. Miiller.
6.—Shell seen from right side. x 84.
7.—Shell seen from above. x 84.
8.—Posterior antenna. x 200.
9.—Extremity of foot of second pair. x 240.
10.—Right copulative organ of male. x 350.
11.—Caudal ramus. x 240.

PLATE XXX.

Proteocypris reniformis n. sp.
1.—Shell seen from left side. x 84.
2.—Shell seen from above. x 84.
3.—Posterior antenna. x 300.
4.—Mandible and palp. x 300.
5.—Maxilla of first pair. x 300.
6.—Claw of second maxilla, gd. x 300.
Caudal ramus. x 440.
8.—Hjaculatory duct and copulative organ, ¢.
9.—Shell structure. x 240.

(.

x 240.

1 Linnean Society’s Journal, ‘ Zoology,’ vol. xix (1885).

Ftia.
Fie.
Fia.
Fia.
Fra.
Fia.
Fia.
Fie.
Fie.

Fie.
Fia.
Fic.

Fie

Fia.
Fig.
Fic.
Fig.
Fia.

Fia.
Fic.

Fia.
Idives

Fie.

Fia.

ENTOMOSTRACA COLLECTED IN NATAL.

Proteocypris globuloides n. sp.
10.—Shell seen from left side. x 84.

11.—Shell seen from above. xX 84.
12.—Posterior antenna. x 240.

13.—Claw of second maxilla, right side, ¢. x 3500.

14.—Second maxilla of left side, g. x 300.
15.—Foot of first pair. x 240.

16.—Foot of second pair. x 500.
17.—Caudal ramus. x 350.
18.—Copulative organ of right side. x 240.

PLATE XXXII.
Stenocypris aldabrxe G. W. Miiller.

1.—Shell seen from left side. x 31.
2.—Shell seen from above. x 3l.
3.—Posterior antenna. x 84.

. 4.—Extremity of second foot. x 240.
Fic.
Fic,

5.—Caudal rami. x 84.
6.—End of right ramus, enlarged. x 240.

Cypridopsis punctillata n. sp.
7._Shell seen from right side. x 80.
8.—Shell seen from above. x 80.
9.—Anterior border of left valve. x 8U.
10.—Caudal ramus. x 440.
11.—Shell sculpture. x 240.

Leydigia propinqua G. O. Sars.
12.—Female seen from left side. x 100.
13.—Caudal lamina. x 240.

Alonopsis sp?
14.—Female seen from left side. x 130.
15.—Caudal lamina. x 250.

PLATE XXXII.
Chydorus gibsoni n. sp.
1.—Female seen from left side. x 90.

14.—Lip-plate.
2.—Caudal lamina. x 250.

185

186 G. STEWARDSON BRADY.

Ceriodaphnia natalis n. sp.

Fie. 3.—Male seen from left side. x 84.
Fie. 4.—Antennule of same. x 100.

Fie. 5.—Female seen from left side. x 84.
Fic. 6.—Female seen dorsally. x 84.

Fie. 7.—End of caudal lamina. x 240.

Simocephalus capensis? G. O. Sars.

Fic. 8.—Female seen from right side. x 31.

Pleuroxus assimilis n. sp.

Fie. 9.—Female seen from left side. x 110.
Fie. 10.—Caudal lamina. x 320.

Ann Natal G.Mus.Voll. | PL.XXIX.

Wi i, 1x40.

Brady del. Huth Lith® London.

CYPRIS INTUMESCENS nov. sp Fig? 1-5 CYPRIA ARMATA G.WMiid@er Fig? 6-11.

Gos

a

a

Ann. Natal G.Mus Vol.

Brady del. Huth Lith® London.

PROTEOCYPRIS RENIFORMIS nov sp.FigS 1-9 PROTEOCYPRIS GLOBULOIDES nov. sp.Fig$10-18.

Ann.Natal G.Mus.Vol. 1.

1xa1.

Brady del. Huth Lith’ London.
STENOCYPRIS ALDABRA GWMiiller Fig2 1-6 CYPRIDOPSIS PUNCTILLATA nov sp.Fig§ 7-1.
LEYDIGIA PROPINQUA GOSars Fig§ 12-13 ALONOPSIS? sp. Fig® 14-15.

Ann. Natal G.Mus.Vol1. PL. XXXIT.

Brady del. Huth Lith? London,
CHYDORUS GIBSON! nov.sp.FigS1-2 CERIODAPHNIA NATALIS nov. sp. Fig? 3-7.
SIMOCEPHALUS CAPENSIS? GOSars hig,8 PLEUROXUS ASSIMILIS nov.sp Fig$ 9-10.

ON PARAWRIGHTIA ROBUSTA. 187

On Parawrightia robusta gen. et sp. nov., a
Hydroid from the Natal Coast; and also an
Account of a Supposed Schizophyte occur-
ring in the Gonophores.

By
Ernest Warren, D.Sc.Lond.
Director of the Natal Government Museum.

With Plates XX XIII and XXXIV.

15

THis hydroid has been found at several places on the Natal
coast. It was first collected at Park Rynie, and subsequently
at Isipingo and Scottsburg. These places are situated thirty
to forty miles south of Durban. The hydroid is not very
common; it occurs attached to sea-weeds and sponges in the
rock-pools near the low-water line.

(1) Trophosome.—lIt consists of a creeping stolon, about
0-25 mm. in diameter, forming an irregular reticulum on
the surface of the foreign body to which it is attached
(Pl. XX XIII, figs. 1, 2). From the reticulum there spring
upright stems 5—12 mm. in height, which usually bear a
single terminal polyp. Occasionally the stem may produce
one or even two short lateral branches, each carrying a
hydranth.

The growth of the stolon or reticulum is not by any means
limited to apical growth. Almost any part of the stolon
appears to have the power of sending out bud-like swellings.
The ectoderni under the thick perisare proliferates, forces its
way through the perisarc, and is at first only covered by the

188 ERNEST WARREN.

thinnest cuticle-like layer (text-fig. 3, a,c.). An examina-
tion of a reticulum and its growing apices clearly shows that
the lateral off-shoots are capable of fusing with neighbouring
stolons, the perisare ultimately disappearing at the place of
contact, and the lumen of the lateral shoot becoming con-
tinuous with that of the stolon to which it has joined. In
the annexed figure a piece of the stolon reticulum is
shown. In the case of the shoot A, the outgrowth has bent
round and fused with the parent stolon. In B, a cross-
filament is flattening itself over a neighbouring stolon, pre-
paratory to complete fusion and the disappearance of the
intervening layer of perisarc. C and C are young buds

Trext-Fig. 1.

AA. Branch uniting to parent stolon. c. Young buds covered by thin
membrane. B. Young bud fastening itself to neighbouring stolon.
x 20.

which have recently appeared. It is doubtful whether the
substance of the perisare is the same as that of the chitin of
Arthropods ; but in any case it 1s a peculiarly insoluble sub-
stance, and it might be of interest to endeavour to trace the
mode of absorption of the perisarc in the formation of buds,
and the fusion of lateral shoots, since it is clearly not merely
a mechanical rupture. JI have not, however, had an
opportunity to investigate the matter with any thoroughness.

The stolons may be knarled in places, and the general
contour tends to be somewhat irregular (Pl. XX XIII, fig. 2, %).
The stems bearing the hydranths are irregularly rmged or
spirally ridged to a variable extent. ‘he ridging (r.) may
extend throughout the whole length ; but it is generally con-
fined to the basal region, where the hydrocaulus springs

ON PARAWRIGHTIA ROBUSTA. 189

from the stolon, and the greatest strain has to be borne.
The older portions of the stolon tend to be dark brown,
while the younger portions are pale brown or whitish. The
whole surface of the perisare is generally clothed with
particles of mud and very fine sand (Pl. XX XIII, fig. 3, ex. L.,
and XXXIV, fig. 6, o. l.).

The base of the hydranth is cup-shaped, and the perisare
extends over it forming a kind of calyx (Pl. XX XIII, fig. 3) ;
but the crown of the hydranth is not perceptibly retractile
into it, as in the case of Hinck’s genus Atractylis. There
are thirteen to eighteen tentacles which spring in a single
whorl around the edge of the cup-shaped base. The tentacles
are sometimes held alternately elevated and - depressed
(Pl. XXXIII, fig. 2, p. el.), as in the genera Perigonimus,
Bimeria, etc. The hypostome can scarcely be described
as conical, it is intermediate im shape between the conical
hypostome of a Perigonimus and the widely expanded,
trumpet-shaped hypostome of an Eudendrium.

The hydranth can be found in three conditions (Pl. XX XIII,
fig. 2): (1) the ordinary contracted condition (c. p.) with
tentacles placed more or less vertically, covering the hypos-
tome; height about 0°90 mm.; (2) hypostome and cup
elongated (p.el.) and tentacles alternately elevated and
depressed; height about 1:10 mm.; (3) flattened condition
(ev. p.) forming a star, where the hypostome and cup con-
stitute the disc, about 0°90 mm. in diameter, and the tentacles
form the rays. The thick perisare of the cup must be
highly elastic in order to allow such very complete lateral
expansion.

The endoderm of the hydranth is red, the other parts of
the hydroid are translucent and white.

The sexes are separate, and fixed gonophores are formed.
Both male and female gonophores have only been found
springing singly from the stems carrying the hydranths
(Pl. XXXITI, fig. 2., g, y.g.), they have not been seen
originating from the stolon.

(2) Histology.—Perisare.—The perisare of the stolon is

190 ERNEST WARREN.

about 18°7 u in thickness, and is obviously laminated in
structure. The stem carrying the hydranth has a perisarc
consisting of two layers: (1) An inner laminated layer con-
tinuous with the perisarc of the stolon (Pl. XX XIII, fig. 3, ch.’),
and (2), an outer vertically striated layer (ch.'!). This layer
alone forms the cup of the hydranth, and its upper edge is
inserted in the shallow perisarc-groove just below the verticil
of tentacles. The outer layer gradually thins out towards the
base of the hydranth-bearing stem, while the laminated layer
thins out and disappears at the base of the hydranth. The
thick perisarc around the gonophore consists of the laminated
layer only (Pl. XXXIII, figs. 3 and 5, ch. 2; text-fig. 3,
B, ch.*). The whole of the perisarc is apparently covered
with a gelatinous or sticky substance, as there is normally a
very conspicuous coating of small particles of mud and sand.
In the developing hydranth the whole of the ectoderm of
the cup-like base appears to take part in the secretion of
perisarc. The perisarc-groove, where the perisare terminates,
is a more obvious structure than in the mature hydranth
(text-fie. 2, B, p.g). It would appear that the outer
vertically striated layer (ch.') is only secreted by the
hydranth, and that as the hydranth elongates and forms a
stalk, this layer of perisarc is, so to speak, left behind, and is
then strengthened by the secretion of the inner laminated
layer by the general ectoderm of the stem. It may be noticed
that in the majority of the gymnoblastic hydroids the groove
or “collar” occurs at the top of the stem bearimg the polyp,
where the perisare terminates ; while in the present species it
occurs just below the verticil of tentacles of the hydranth.
EHctoderm.—The histology of the ectoderm is typical.
The nematocysts are of one kind only in the hydranth, and
are somewhat small, measuring about 5:0 win length and 2°9 pu
in breadth. I have not found them occurring in the endo-
derm. They are scattered somewhat sparsely throughout the
ectoderm of the tentacles and general ccenosarc. In the
gonophores, nematocysts are abundant, and they tend to be
somewhat larger than those on the tentacles (5°6 « in length) ;

ON PARAWRIGHTIA ROBUSTA. 19]

they occur in the thickening of the umbrella ectoderm
around the opening of the umbrella (Pl. XXXIII, figs. 3
and 5,7; text-fig. 6, .). They also occur plentifully in the
midst of the spermatic mass (Pl. XXXIV, fig. 6, 7).

It seems fairly clear that nematocysts can have no function
imbedded in the spermatic mass, or even in the general
coenosarc, since they are shut off from the exterior by a thick
perisare. It would appear that the ectoderm, being endowed
with the power of forming nematocysts, produces these
structures whether or not they can become functionally active.
It is conceivable that the presence of nematocysts throughout
the entire substance of the ectoderm would render the
hydroid distasteful to be eaten as a whole; but I do not
consider that this is the correct view. The abundance of
nematocysts around what would be the margin of the
umbrella is an interesting case of the persistence of a
structure after the loss of its function. ‘he fact that the
nematocysts are somewhat larger in the gonophore than in
the hydranth is also of some interest.

Kndoderm.—The endoderm of the hypostome consists of
narrow columnar cells, with small deeply-staining nuclei
(Pl. XXXITII, fig. 4, en.h.). The small vacuolated cells, which
stain intensely, and are frequently seen in the endoderm of
the hypostome of various hydroids, do not appear to occur in
this species. In transverse section the endoderm of the
hypostome is seen to be raised into six to eight prominent
ridges.

At the base of the hypostome the endoderm abruptly
changes its character, and the cells are large, vacuolated,
and richly granular. These endoderm cells frequently
contain globules of yolk (Pl. XXXIII, fig. 3, y. gl.), more
especially towards the base of the polyp. Special glandular
cells (gl. ¢.) with large granules occur in the upper portion of
the digestive cavity of the hydranth, and they are also found
irregularly scattered in the endoderm of the stem and stolon.

(3) The development of the hydranth.—A lateral
branch of the stolon, a hydranth, or a gonophore, originate

VOL. 1, PaRT 2-

192 ERNEST WARREN.

from the parent coenosare in a closely similar manner. A
bud is formed by an outgrowth of the ccenosarc, which pushes
its way through the perisarc, partly by mechanical pressure,
and partly, I think, by absorption (text-fig. 3,4). The bud
is at first covered merely by a very thin membrane (c.). In

TEXT-FIG. 2.

a

A.

A. Adult hydranth of Eudendrium sp. bs. Developing hydranth of
Parawrightia robusta.

c.en. Central endoderm. jf.en. Fan-shaped endoderm. 7.e. Reflexed
ectoderm. ¢. Tentacle. p. Perisarc, consisting of the vertically
striated layer chiefly. p.g. Perisarc groove. d.g. Digestive cavity.
x 140.

the case of a lateral branch of the stolon the bud can readily
mould itself so as to fit into any irregularities of surface of
the foreign body which supports the hydroid. In this way
the colony becomes very firmly attached (Pl. XXXIII,
fig 2) clsr:).

ON PARAWRIGHTIA ROBUSTA. 1938

The developing hydranth shows a very marked resem-
blance to the adult structure of an undescribed Hudendrium-
like species, which was dredged off Bird Island last year by
the museum collector.

In the annexed illustration are shown, side by side, vertical
sections of the adult polyp of the Hudendrium (a) and of the
developing polyp of Parawrightia (8).

The striking similarity in structure in the two cases is
remarkable. At this stage in the development there is no
hypostome, and the elongated endoderm is spread out in a
fan-shaped manner so as to be widely exposed to the exterior.
The reflexed ectoderm (7. e.), the arrangement of the endo-
derm over it (f. en.), and the origin of the tentacles (¢.) may
be directly compared with the condition seen in the adult
polyp of the supposed Kudendrium. The central portion of
endoderm (c. en.), consisting of very elongated cells, is
essentially alike in the two cases. The digestive cavities
(d. g.) are both exceptionally narrow. In the Hudendrium
the perisare is less strongly developed; but it does extend
over the base of the polyp (p.) to form a kind of calyx, as in
Parawrightia.

It is, of course, debatable how far such a resemblance is to
be regarded as arising through genetic relationships ; but the
similarity of structure in the two cases is so close that the
matter appeared worthy of record ; and it is not unreasonable
to assume the possibility that the mode of development of the
hydranth may give some hints of phylogenetic significance.

(4) Gonosome.—tThe sexes are separate, ale and female
gonophores being formed on different colonies. ‘The gono-
phores, whether male or female, arise on the stems carrying
the hydranths. As a rule only one gonophore is produced on
a stem, and it is generally situated not far from the polyp.
The gonophore, like a lateral polyp, arises by proliferation of
a patch of ectoderm and endoderm, which grows through the
perisare in the form of a swelling, and is at first only covered
by a very thin membrane.

Ultimately the gonophore becomes surrounded by a uni-

194, ERNEST WARREN.

form layer of perisare continuous with the inner laminated
layer of the stem (text-fig. 3, 8B). An umbrella-cavity (w. ¢.)
appears at an early stage with ectoderm on ectotheca (e. e.)
and endotheca (g. e.) Later on the umbrella-cavity becomes
relatively small and is mostly confined to the apical region of
the gonophore (text-fig. 4, a and B). Below the apically-
situated cavity the endotheca, consisting of the generative
epithelium, is continuous with the ectotheca; but there are
gradually developed four canals (Pl. XX XIII, fig. 4, sp., and
text-fig. 4, sp.) which are continuous with the umbrella-
cavity in the apical region, and are situated in such a manner

TEXT-FIG. 3.

A. Polyp-bud. x 100. B. Young gonophore. x. 70.
c. Thin membrane covering bud. ch}. Outer striated layer of perisare.
ch. Inner laminated layer of perisarc. g.e. Germinal epithelium.
u.c. Umbrella cavity. e.c. Ectoderm of Ectotheca.

as to alternate with the endodermal radial canals (7. ¢.) An
endodermal lamella (en. /.) is present, connecting together the
radial canals.

The male and female gonophores differ in shape. ‘The
male is a somewhat elongated ovoid, about 1°20 mm. in
length, and 0°70 mm. in breadth, and is so placed that it hes
almost parallel to the parent stem. The female gonophore
is more nearly spherical, about 1:10 mm. in length, and
0°99 mm. in breadth, and it tends to stand out nearly at right
angles to the stem (Pl. XXXIII, figs. 3 and 5).

The umbrella-cavity (w. c.) of the mature female gonophore
has an apical opening (0. uv.) surrounded by nematocysts (1.) ;
but it does not communicate with the exterior on account of

ON PARAWRIGHTIA ROBUSTA. 195

the presence of the thick perisarc (about 25 uw thick). In
the case of the male gonophore I have not actually seen such
an opening developed, but its position is clearly indicated by
the dipping in of the ectoderm and the cluster of nematocysts
(Pl. XXXII, fig. 5; text-fig. 4, a.) The ectodermal epi-

thelium of the ectotheca lining the umbrella-cavity is seen in

TEext-Fic. 4.

XS
= SS
MASS AN | Ee
: SG aS
a wd + : F ZS
yey, <b

‘ 3S
Sp.
/ SZ ‘ ><
he ‘
ont Noe
ch. 2

A. B.

A. Female gonophore, longitudinal section. 8. Male gonophore,
transverse section.

c.s. Capsule of Schizophyte. n. Nematocysts. wu.c. Umbrella-cavity.
en.l. Endodermal lamella. o. Ovum. jl. Follicle cells. 7.c. Radial
canal. ch®. Laminated perisare. sp. Canals opening into umbrella-
cavity. x 140.

Pl. XXXIV, fig. 6, n. 0. e., and of the endotheca, as distinct
from the spermatic tissue, at e.c. The space marked w. s. is
one of the channels which opens into the umbrella-cavity at
the apical region of the gonophore. The spaces marked ch.
occur irregularly through the spermatic mass. <A distinct
circular canal is not formed.

196 ERNEST WARREN.

It is far from clear how fertilisation takes place, since the
gonophore is surrounded by a thick perisarc, and the eggs,
after they have become charged with yolk-globules, segment
and ultimately form planule while still inside the parent
gonophore (Pl. XX XIII, fig. 3, Pl.). The planule are solid,
and consist of an outer columnar epithelium, within which is
a granular mass densely crowded with yolk. During matura-
tion the eggs are surrounded by follicle cells (Pl. XXXIII,
fig. 4, fl.; and text-fig. 4, a, fl.). On the formation of the planule
the follicle cells disappear, but a layer of ectoderm cells
persists on the ectotheca and also covering the spadix
(Pls XCar aioe oanect..\e

(5) Systematic position.—As it has been seen, the re-
productive bodies are fixed gonophores, and consequently
the present species is at once marked off from Perigonimus,
where free medusz are produced. It agrees with Bimeria
and Garveiain having fixed gonophores. It differs, how-
ever, from Bimeria in habit, in not having the perisarce
continued over the base of the hypostome and tentacles, and
in having the spadix unbranched. It also differs from
Garveia in habit and in the structure of the gonophore.

In the gonosome the species agrees with Hinck’s definition
of Atractylis arenosa (Allman’s Wrightia arenosa),
in which a special point is made that although closely allied
to Perigonimus it differs in having fixed gonophores.

The polyp of this species is, however, very little like that
of Wrightia arenosa. The average number of tentacles is
about tifteen instead of about nine; the tentacles are not
muricated, and the polyp is scarcely at all retractile into the
upper cup-like portion of the stem. Also, the extrusion of
the eggs from the gonophore into a gelatinous sac, which
Hinck’s describes in Atractylis, has not been observed in
the present species; and in fact it probably does. not occur,
since fully-developed planule are found in the unruptured
gonophore. ‘The mode of fertilisation is problematical.

The general appearance of a colony is like that of Perigo-
nimus; the endoderm of the hydranth is red, as in several

ON PARAWRIGHTIA ROBUSTA. 197

species of Perigonimus, while in Wrightia it is white.
On the whole it appears advisable to separate it from
Wrightia; and the name Parawrightia robusta is pro-
posed to indicate the relationship.

i:

The structures about to be described, and shown in situ in
Pl. XXXIII, fig. 3, p.b., Pl. XXXIV, fig. 6, y.s., have only been
found in the male and female gonophores. I have searched
carefully in the substance of the general tissues of the hydroid,
both in the ectoderm and endoderm of the polyp, stem, and
stolon. Also, they have not been seen in any cavities,
such as the digestive cavities of the polyp or ccenosarc, or
in the irregular cavities between the coenosare and the
perisarc. ‘The structures occur in the umbrella-cavity, in the
generative epithelium, and in one case embedded in the
ectoderm filling up the openmg of the umbrella-cavity.
‘hey have not been seen in the endodermal radial canals
of the gonophore. It will be noticed that the structures
develop in temporary spaces, and that they must ultimately
pass out to the exterior by the dehiscence of the gonophore.

In no sense do the bodies appear to exercise a malignant
effect on the host; they may increase in numbers in the
gonophore to a considerable extent, but the development of
the planule and sperm continues, to all appearance, in a
perfectly normal manner.

From the amount of material available it is difficult to
obtain much idea of their frequency of occurrence. I have
collected six colonies ; three of these possessed no gonophores;
of the remaining three, two from one locality (Park Rymie)
were female, and one from another locality (Isipingo) was
male. About fifteen male and female gonophores in various
stages of development were present on these three colonies,
and the structures were present in ten. They occurred in
all three colonies, but not in every gonophore. The evidence,
as far as it goes, indicates a general occurrence of the struc-

198 * ERNEST. WARREN.

tures in. the gonophores. The organisms have not been
found outside the hydroid, either adhering to the surface of
the perisarc, or to the sea weed or other foreign body to which
it is attached. ‘The structures occur. especially in the four
canals of the umbrella cavity left between the ectotheca and
endotheca, and alternating in position with the radial endo-
dermal canals (text-fig. 4, 4 ands). They may also be found
in cavities that occur in the spermatic mass (Pl. XXXIV,
fig. 6, ch.), and between the ova or developing planule of a
female gonophore.

The youngest stage observed is seen in Pl. XXXIV, fig. 7
(1). It is a rounded body, about 2°24 in. diameter, and
staining with hematoxylin or aniline dyes with great diffi-
culty. The body is refringent, and appears homogeneous
with Zeiss’s DD objective, but with Zeiss’s apochromatie +y
it is seen to be shehtly granular with one or two minute
globules. In this stage it is capable of division. I have
found such bodies embedded in the spermatic tissue, and in
one case in the ectoderm of the ectotheca. The bodies
apparently work their way out into any spaces, such as the
umbrella-cavity and channels, and any cavities in the sper-
matic mass (Pl. XXXIV, fig. 6, b) or between the ova. No
distinct, separable cell-wall can be observed.

When in a cavity the body soon secretes a transparent and
refringent envelope of one or two layers (fig. 7 (2, 3, 4,
and 5)). This envelope, especially the outermost and older
layer, is capable of staining deeply with hematoxylin (PI.
XXIV, fig. 8 (4); 0.en=).

In this condition (fig. 7 (4) ) it is capable of fission, both
the central body (ec. b.) and the envelope dividing.

The envelope, forming a definite capsule, increases in size
and in the number of laminations, and the central body
becomes smaller and more refringent. The central body, which
in the young stage is rounded and about 2°24 in diameter,
varies in size in the fully-grown. capsule from about 1:10"
to 0°48 w (figs. 9 and 10, c.b. and d.b.). The central body can
divide by fission (d.b.) inside the capsule. Previous to fission

ON PARAWRIGHTIA ROBUSTA. 199

the body becomes rod-shaped, and is about 1:10 w. in length,
and 0°50 win width (fig. 10,¢.b.) On the division of the central
body, one of three events nay take place :

(1) The envelope divides, and fission or budding takes
place as mentioned above. This may continue a number of
times until a cluster of individuals is produced (fig. 8 (2, 4, 4)).
Such occurs only when the capsules are small and consist of
but few layers.

(2) The division products of the central body may each
form a centre from which fresh laminated layers originate.
In this way a number of separate centres of lamination may
be formed inside the original system (fig. 8 (7) and fig. 11).

(8) The division-products of the central bodies may not
start fresh centres of lamination, but may remain stranded, so
to speak, in the laminated system of the parent body
(figs. 9, 10, 11, s.b., and text-fig. 5 (7) s.b.). In an old
capsule such bodies are seen irregularly scattered through
the laminations.

In the case of the female gonophore the capsules may grow
to a very considerable size, and may reach a length of 70 4 or
more. The appearance of these structures varies according
to the number of centres of lamination. The central bodies,
and the bodies stranded in the substance of the capsule, are
highly refringent, while the substance of the capsule is_per-
fectly clear like spun glass, and is considerably refringent.

The structures may be spherical with a single central body,
or they may be elongated with a row of central bodies along
the long axis (Pl. XXXIV, fig 10), or they may be lobed
with various centres of lamination (fig. 11). In the case of
fig. 9 the structure was bilobed with two centres of lamination,

In the accompanying illustration (text-fig. 5), the arrange-
ment of the laminations and the bodies is shown on a large
scale, and consequently more clearly than is possible in the
plate. The cone-shaped figures are wedges taken from
various capsules; the centre of the capsule being situated
towards the right hand side of the page.

Figures 1 to 5 and 7 show the early. stages of development,

Text-Fic. 5.

e.b. Central body or supposed Schizophyte. d.b. Dividing body. d.i.c.
Disintegrating inner capsule. f.d. Faintly defined layer of capsule.
o.c. Outer capsule. o.c.l. Outer layer of inner capsule. s.b. Stranded
body. s.d. Sharply defined layer. sp. Spore. sp,f. Spore formation.
x 2000.

ON PARAWRIGHTA ROBUSTA. 201

and have already been sufficiently explained. Fig. 6, and Plate
XXXIV, fig. 8 (8, 9, 10), represent capsules which frequently
occur in the male gonophore, and they will be mentioned below.

In text-fig. 5 (8) are seen two central bodies, 1°34 and
0-6 « in length, which have clearly been formed by fission.
The outermost layer (0. ¢.) of the capsule is about 3°94 im
thickness, it is apparently very compact and dense, and the
lamination, if visible at all, is very close. On the inner side
of this dense envelope are two thin layers separated by a
narrow clear space (0. c. /.). The outer capsule and these two
thin layers stain more intensely than any other part of the
capsule. Passing inwards we find concentric layers arranged
with great regularity, often there are alternating layers of
sharply and less sharply defined strata. In these layers and
between them can be seen refringent bodies irregularly
placed. They may be clustered on one side of the capsule
only (Pl. XXXIV, figs. 9, 10, 11, s.b.). They vary in size
from about Il u to about 0°44, and were presumably formed
by the division of the central body. The layers of capsule
immediately surrounding the central body or bodies are less
pellucid than the others (Pl. XXXIV, figs. 10 and I], 7. /.).

In fig. 9 there are four central bodies, obviously formed by
the division of two. The capsule shows the usual outer com-
pact layer with little trace of lamination, then the sharply
defined double line. Within this there are twelve or thirteen
closely apphed layers (m. c.), following which are two
sharply defined layers, and then a considerable number of
thick layers (s. d.), each consisting of two or three faintly
defined layers (f. d.).

In fig. 70 the central body has broken up into a cluster of
smaller bodies, and it is supposed that it is in such a way
that the bodies (s. b.) scattered through the substance of the
capsule originate. Most of the bodies of the cluster appear
to become included in the new laminations, while one or more
remain as central bodies producing fresh laminations.

In fig. 17 the central bodies, of which there were clearly six,
have each broken up into clusters of small bodies.

202 ERNEST WARREN.

When the life of the gonophore is drawing to a close, and
the generative products are nearly ripe, a change takes place
in the central bodies. These divide rapidly and form an
irregular collection in the central region of the laminated
system ; at the same time the innermost layers of lamination
become indistinct, and then disappear. ‘The bodies, derived
from the splitting up of the central bodies, increase consider-
ably in size; they are rounded in shape, and some of them
become conspicuously large (Pl. XXXIV, figs. 11 and 12, d.b.;
and text-fig. 5 (12), sp. f.).

Ultimately the whole of the laminated layers of the capsule,
except the outer dense layer, and the double sharply defined
layer immediately on the inside, disintegrate, and more or
less completely disappear (text-fig. 5 (13), d.i.c.). Most of
the bodies derived from the splitting up of the central body
or bodies also disappear ; but a few persist, ten to twelve for
a good-sized capsule, having grown into oval bodies of very
fairly constant size and shape. The average size is 3°44 in
length and 2°54 in breadth. These bodies may probably
be regarded as “spores ;” they are characterised by readily
becoming intensely stained with hematoxylin, while the
ordinary central bodies of the capsules do not stain very
readily. The “spore” is surrounded by a membrane which
appears to be firm, since a constant shape 1s maintained.

In Pl. XXXIV, fig. 12, the process of ‘‘ spore-formation ” is
in progress, and the inner layers are shrinking (see also text-
fig. 5 (13) ). Ultimately the capsules become hollow, and
include nothing but a few oval “spores” and perhaps the
shrunken residue of some of the inner layers (Pl. XXXIV,
figs. 13 and 14).

Under the Inghest magnification I could detect no obvious
structure in the “spore” except an outer membrane and a
very finely granular interior.

It may be noticed that in Pl. XXXIV, fig. 11, a number of
different systems are enclosed in the compact outer envelope.
In two of these spore-formation is beginning (d.b.), while in
the other two centres vegetative growth is still continuing.

ON PARAWRIGHTIA ROBUSTA. 203

In the case of the male gonophores, I have never found
such large capsules as occur in the female gonophores. Their
final growth appears to be much reduced, and they do not as
a rule exceed 10 to 12 in diameter. In some of these
smaller forms the concentric lamination, especially towards
the centre, is replaced by a radial striation (Pl. XXXIV,
fig. 8 (8, 9, 10), and text-fig. 5 (6) ). In figs. 9 and 10 of the
plate the central body has split up (d. b.) preparatory to the
formation of spore-like bodies. In correlation with the small
size of the capsules in the male gonophore it may be noticed
that there is but little space for development, as the umbrella-

TEXT-FIG. 6.

u.c. Umbrella-cavity. ¢.b. Central body, supposed Schizophyte. end.l.
Endodermal lamella. 2. Nematocysts. sh.sp. Shrinkage space be-
tween capsule and hydroid. x 750.

cavity is very restricted. It is interesting to note this
adaptation to the environment.

In one female gonophore a capsule was found embedded in
the ectoderm at the future opening of the umbrella-cavity.
It differed from the capsules growing free in the umbrella-
cavity in the almost complete absence of the outer compact
layer, and in the layers of the capsule being apparently more
separable from one another.

The real nature of the central bodies and capsules is
problematical. The central bodies are, as we have seen,
highly refringent, and stain in section only with difficulty.
The structures in a young condition can be embedded and cut
with a microtome fairly well, but in an older condition, when

204 ERNEST WARREN.

the capsule is fully formed, it has not been found possible to
cut satisfactory sections. The circumstance is, however, of
less importance as the capsules are exceedingly transparent.
It is probable that the difficulty in cutting is due to the
impermeability of the capsule to the paraffin-wax.

The capsule itself is refringent, but it stains readily with
Delafield’s hematoxylin.

I have treated some fragmentary sections with a series of
reagents in order to endeavour to test the nature of the
capsule. I have had no opportunity of testing fresh material!
in a similar manner,

Cellulose tests.—lIt is coloured faintly yellow by liquor
iodi; it is slightly disorganised by concentrated sulphuric
acid ; it is not coloured blue with iodine and sulphuric acid ;
it is not appreciably swollen or dissolved by an ammomniacal
solution of cupric hydrate.

The capsule accordingly does not give a characteristic
cellulose reaction.

Proteid tests.—Concentrated nitric acid causes a faint
yellow tinge both in the capsule (especially in the outermost
layer) and in the perisare and mesoglea of the hydroid. On
the addition of 50 per cent. caustic potash or ammonia the
yellow tinge becomes somewhat intensified. ‘here would
thus appear from this test, the xanthoproteic reaction, that there
is some proteid matter both in the capsule and in the perisarc.

Neither 50 per cent. caustic potash nor concentrated
ammonia have any appreciable swelling action on the capsule.

Concentrated hydrochloric acid and aqua regia do not
obviously affect the capsule.

Picric acid has no marked staining effect.

Delafield’s hematoxylin stains the capsule and the mesoglea
intensely, while the perisare is somewhat less strongly acted
upon.

Methyl blue stains the capsule deep blue.

Methylene blue is an excellent stain for the capsule and

1'The hydroid had been fixed with a half-saturated solution of corrosive
sublimate in 80 per cent. spirit with 1} per cent. acetic acid.

ON PARAWRIGHTIA ROBUSTA. 205

also for the central bodies. By means of this stain the outer
envelope of the capsule, which generally appears homo-
geneous, 1s Shown to be finely laminated.

Methyl eosin does not stain the capsule or the perisare of
the hydroid to any extent.

Congo red tends to stain the inner envelopes of the capsule
more than the outer sheath. All the tissues of the hydroid,
including the perisarc, are stained. On faintly acidulatine
the preparation the perisare becomes blue, while the capsules,
the mesoglea, and the rest of the tissues of the hydroid
remain red.

Mr. Arnold Cooper has kindly tested for me the effect of
the capsules on polarised light. He finds that they sometimes
possess shght depolarising power. The innermost layer of
the capsule immediately surrounding the central bodies ap-
pears generally to possess this power to the greatest extent.
The central bodies themselves remain dark. In some capsules
areas of hght of different sizes appear in various parts of the
laminations. Perhaps this is due to stresses and strains set
up in the substance of the capsule during its growth. The
perisarc, especially the innermost layer, and also the mesoglea
may at times possess a certain amount of depolarising power.

In all the tests that have been applied, the capsules, the
mesoglea, and the perisare react very similarly, and it is
possible that they may all consist of somewhat analogous
substances; but the capsules are certainly harder and more
brittle than either the mesoglea or the perisare.

It would appear probable that the organism gains entrance
into the gonophore when the latter is in the form of a bud,
for at this stage the young gonophore is practically naked,
as the coverimg membrane is excessively thin.

The central body is possibly a Schizophyte, and the lami-
nated capsules recall Leuconostc; but in the present case
the capsules are hard and not gelatinous.

The formation of a limited number of somewhat large
spore-lhke bodies, with the disappearance of the middle
portion of the capsule and the greater number of the cocci,

206 ERNEST WARREN.

are also somewhat analogous to the phenomena which occur
in Leuconoste.

The central bodies are too small, and apparently structure-
less to be regarded as unicellular alow of the nature of Gleo-
capsa, and besides the character of the capsule is not the
same. 3

Certain bacteria form capsules, and perhaps in the present
case in extracting their nourishment from the fluids of the
hydroid the central bodies form the capsules as a by-product,
which is not very dissimilar in nature from the mesoglea.

On the other hand, it should be noticed that in the forma-
tion of the so-called spores the whole of the inside of the
capsule is dissolved, and only the outer compact envelopes
remain. It would, therefore, be more reasonable to regard
the inner substance of the capsule as reserve food-material
which is utilised in the reproductive stage when the spores
are formed. It is scarcely conceivable that the relatively
huge capsule should be formed merely in connection with the
vegetative growth of the minute central bodies. This hypo-
thesis is rendered somewhat more probable in that the
reproductive processes take place when the gonophores are
dehiscing and losing their vital activity.

EXPLANATION OF PLATES XXXIIT AND XXXIV,

Illustrating Dr. Ernest Warren’s paper “On Parawrightia
robusta gen. et sp. nov., a Hydroid from the Natal Coast;
and also an Account of a supposed Schizophyte occurring
in the Gonophores.”

Fie. 1.—Natural size. Colony growing over sponge and sea-weed.

Fie. 2.— x 15 diameters. A small piece of a female colony attached
to sea-weed. The hydranths are shown in different conditions of
expansion and contraction (ex. p.; p.el.; c. p.). The attachment of the
stolon to its support is assisted by clinging branches (el. 7.). The
planule have escaped from the empty gonophore (em. q.).

Fie. 5.—x 60. Longitudinal section through hydranth and female

ON PARAWRIGHTIA ROBUSTA. 207

gonophore. The umbrella cavity (w. ¢c.) of the gonophore has an opening
(o. u.). Two planule nearly ready to escape are shown (Pl.). ect.
Ketodermal epithelium lining spadix and ectotheca.

Fia. 4.— x 60. Transverse section of a young female gonophore.
Alternating with the endodermal canals are seen four spaces (sp.),
which are ectodermal canals opening into umbrella-cavity.

Fie. 5.—~x 60. Longitudinal section through a male gonophore.
The umbrella-cavity (wv. ¢.) is closed, although the position of an opening
is indicated at n. e. ep. is the ectodermal epithelium lining spadix and
ectotheca.

Fia. 6—x 750 diameters. Small piece of transverse section of male
gonophore with capsuled Schizophyte in ectodermal canal and in
channels (ch.) in the generative epithelium.

Fra. 7.— x 1300 diameters. (/) Youngest stage; (2) and (3) show the
beginning of the formation of a capsule; (4) shows the fission of central
body and of capsule.

Fie. 8.—x 1300 diameters. Older stages. (2) (4) and (5) are
examples of budding; (6) central body (d. b.) dividing; (7) several
central bodies are present, and dense outer capsular layer ; in (8) (9) (70)
the laminated structure is replaced by a radiating structure, except for
the outer layer (0. en.). Specimens were taken from a male gonophore.
In (9) and (10) the central body has divided considerably (d. b.).

Fie. 9.— x 1000 diameters. Specimen taken from a female gono-
phore, showing central bodies (c. b.); capsule with outer compact
envelope, and bodies (s. b.) embedded in the substance of the capsule.

Fie. 10.— x 1000. Specimen from female gonophore with central
bodies (c. b.), and dividing bodies (d. b.). Two outer envelopes or
sheaths to the capsule are seen. The innermost layer (7. /.) is less
pellucid than the other layers of the capsule.

Fie. 11.—x 1000 diameters. Lobed specimen with several centres
of lamination. At d. b. the central bodies are splitting up preparatory
to the formation of spores.

Fre. 12.—x 1000. Older stage with dividing bodies (d. b.) and the
contracting inner capsule (I. sh.) with cavity (e. cy.). The two outer
envelopes, en,. and en,. are shown.

Fie. 13.— x 600. Capsule showing spores (spo.) and cavity (e. cy.).

Fie. 14.—x 1500. Specimen with disintegrating inner capsule (Jn. sh.)
and spores (spo.).

Fie. 15.— x 2000. Spores.

VOL. 1, PART 2. 15

208 ERNEST WARREN.

EXPLANATORY REFERENCES.

b. Youngest stage of Schizophyte. c. b. Central body. c¢. p. Con-
tracted polyp. ch,. Outer vertically striated layer of perisare. chy.
Inner laminated layer of perisare. ch. Channel in spermatic tissue.
cl. ry, Clasping branch of stolon. d.b. Dividing body. d. gq. Digestive
cavity. e. Eetoderm of umbrella. ect. Ectoderm lining spadix and
ectotheca. e. cy. Empty cavity of capsule. e.c¢, Ectodermal nuclei in
generative epithelium. e./. Endodermal lamella. e. ep. Ectodermal
epithelium of umbrella cavity. en,. Outer envelope of capsule. eng.
Outermost envelope of inner capsule. en. h. Endoderm of hypostome.
em. g. Empty gonophore. ew. 1. Outer layer of perisare with adhering
débris. ea. p. Expanded polyp. jl. Follicle cells. g. Female gono-
phore. gl. c. Glandular cell of endoderm. 7. J. Innermost layer of
capsule, less pellucid than the other layers. J. sh. and In. sh. Inner
sheath or capsule. k&. Knarled stolon. 1. s. Laminated structure. 2.
Nematocyst. n. 0. e. Nuclei of ectodermal epithelium of ectotheca
lining umbrella cavity. 0. Ovum. o. /. Outer coating of mud and sand.
o. en. Outer envelope of capsule. o. wu. Opening of umbrella cavity.
Pl. Planula. p. b. Capsules of supposed Schizophyte. p. el. Polyp
elongated, with tentacles alternately elevated and depressed. 7. Ringed
portion of stem. +. ¢. Radial endodermal canal. s.b. Stranded bodies
in the substance of the capsule. s./. Laminated perisare layer. s. p.
Kctodermal canals continuous with umbrella cavity. spo. Spore. uw. ¢.
Umbrella-cavity. w.s. Umbrella space. y.g. Young gonophore appear-
ing asa bud on the stem. y.s. Young stages of Schizophyte. y. gl.
Yolk-globules.

Ann. Natal G. Mus. Vol.1.

)
4
\

2)

at.size.

soc

rani

ane Oe a

Tip ete OO, TMU

jini

ee

London.

uth, Lith®

—
a

7
ak

ven, de

ar

We

f

.
&
7 .
-
'
ao

val

Anam Natal G. Mars. Vel).

Q.

VII x 130

[x 1g00

VIL

Xx 1000

~

~
_——
ee:
1a ;

ye
‘Tsh

Ne
oS

IIx 1000

XIV x 1800

uth, Lith” London

Warren, del.

HALOCORDYLE COOPERI. 209

Note on the Variation in the Arrangement of
the Capitate Tentacles in the Hydroid, Halo-
cordyle cooperi Warren.

By
Ernest Warren, D.Sc.Lond.,
Director of the Natal Government Museum.

In the description of this new species in the ‘ Annals of the
Natal Government Museum,’ vol. i, part 1, it was pointed out
that the arrangement of the capitate tentacles exhibited
considerable variation,

According to Allman,! the genus Pennaria is charac-
terised by the hydranth having twenty to thirty capitate
tentacles scattered irregularly over the sides of the body,
while the genus Halocordyle has about twelve capitate
tentacles arranged in two yerticils of about six in each.
Typically H. cooperi possesses two alternating verticils of
four tentacles, but the species exhibits marked variations in
the arrangement, and it was thought that a statistical investi.
gation into the matter might prove of interest.

It is unfortunate that the hydroid is comparatively rare.
The coast of Natal has been searched diligently at different
places on a number of occasions, but only forty-two colonies,
and some of these were in poor condition, have been found
altogether. As it is improbable that any considerable number
will now be obtained, I will give the results, although the
amount of material is not sufficient for a strict analysis.

‘*A Monograph of the Gymnoblastic or Tubularian Hydroids, Part
II, pp. 364, 367.

210 ERNEST WARREN.

| | es | H
Stell @ 3 p | Bg - |
© P| H n| B Rn SD
AS) Locality. Sex. mee me az Remarks.
a) | Sse) Te aS
Oe 6$Ss| 6a) 299
ns 283/483] gee
2 | Scottsburg | 9? 4. 4. 8 Regularly alternating.
4 : OF a 5 9 =
7 3 f°) — — |12—13 Irregularly scattered.
hes! i Pp | 4 5 | 49 | iss
10 - Pp) | 4: 4, 8 Regularly alternating.
11 | F Ba eas " ,
| 12 vs [ees ia eeu liens e .
13 - [eee ec A | 38 MH 5
14. 99 ? 4 4 8 ” ”
15 op P 4. 6 10 —
16 a5 iy 4, ey |) ale =
17 | 9 ? 4 4 ) § | Regularly alternating.
19 | 33 ? 4 4 | 8 > 3)
20 3 2 4 ce 3 » |
21 . eo | Se |) 5; 3 |
26 A Oy eae ited 8 : |
28 | - iy 4 | 4 8 ; -
29 | : tie mm ee 5 9 =
30 - [eae eee eica IE 8 Regularly alternating.
ail * Bi a |) 8 * ”
41 5 ? = — | 9—10 | Irregularly scattered.
5 | Park Rynie| ? 4 TC — |
6 S P| i) 9 | = |
9 fi 2 4 | 5 9 | _
32 | is 2 4 | 5 9 =
33 | a | Ar A a. 8 Regularly alternating.
35 | e ) P| 4) 4] 8 | - - |
36 2 a 8 : 3 |
37 op P| 4 4. 8 ” »
38 | a Pn ca 3 :
4.0 . P| — — |10—11, Irregularly scattered.
42 S P| 4 4 8 — Regularly alternating.
43 S P 4. 4. 8 : 5
1 Isipingo P 4. 4 8 ; +
22 * | 3 4 | 4 ee | ‘s >
23 < Ga eis 8s ii - s
25 | : Dee | 2 Sti . 7
4 | : nn enfin 8 | P 2
45 | 3 P| «64 4 8 | 90 0
45 Kosi Bay ? 4, 4, 8 | $5 -
46 Z ? Aye de | SIS P| : .

|

In the accompanying table is given the average arrange-
ment of the capitate tentacles in the hydranths of the forty-
two colonies. It should be mentioned that the arrangement

HALOCORDYLE COOPERI. PAA

of the tentacles in the hydranths of one and the same colony
is not absolutely invariable, although the greater number of
the polyps conform to the same plan; but aberrations do
occasionally occur. ‘The variability of the polyps in this respect
in the same colony has not been investigated, but only the
general average arrangement in the colony has been recorded.

Thus the average arrangement of the tentacles in colony
“92” Scottsburg, is for four capitate tentacles around the
mouth to alternate with four at a lower level; in colony “3”
the lower verticil in the majority of the polyps consisted of
six or seven tentacles.

The sex of the colonies could only be determined when
gonophores were present. When a distinct oral verticil could
be distinguished, it will be noticed that in no case did it
normally consist of more or less than four tentacles. The
lower or basal whorl is the variable one; typically it is
composed of four tentacles alternating with the four of the
oral verticil; but it may consist of five, six, or even seven
tentacles. In three cases out of the forty-two no definite
verticils could be distinguished, and the average numbers of
the tentacles in these three colonies, “7,” “41,” “40,” were
twelve to thirteen, nine to ten, and ten to eleven respectively.

To show in the form of a diagram the distribution of the
symmetry of the arrangement of the tentacles in the population
it will be necessary to arrange a scale of symmetry which, of
course, must be empirical.

Let (“0”) be the condition of having thirty-two to twenty
sapitate tentacles irregularly scattered. This has not been
observed in H. cooperi, but it occurs in Pennaria and H.
tiarella.

Let (“1”) be the reduced number of thirteen to nine
tentacles irregularly scattered. Observed in three colonies
(about 7 per cent.).

(“2”) Colonies with the average arrangement of four in
an oral verticil, and six to seven tentacles below, which may
be arranged in more or less of a verticil. Observed four
times (about 9°5 per cent.).

212 ERNEST WARREN.

(“3”) Colonies with the average arrangement of four in
an oral verticil, and five in a basal whorl. Observed six
times (about 14 per cent.).

(“4”) Colonies with a vertical of four oral tentacles alter-
nating with a basal whorl of four tentacles. This is the
perfectly symmetrical condition. Observed twenty-nine times
(about 69 per cent.).

TErxtT-Fic. 7.

80

60
50

40

“SOIMO[OD Jo a.dRqUAd1Eg

20
10

0
Degree of symmetry 0 1 2 3 4
Diagram showing distribution of symmetry in forty-two colonies.

It must be remembered that these degrees of symmetry
may not be of equal value, although they are thus represented
in the diagram. For example, there may be a step where
the basal verticil has been reduced to four tentacles, but
these have not acquired the regular alternating position with
the oral tentacles. This cannot be observed with any great
certainty under the microscope.

The results of this investigation may be summed up in the
following paragraphs :

(1) The amount of material available is not sufficient to be
certain as to whether locality or sex have any effect on the
arrangement of the tentacles.

HALOCORDYLE COOPERIT. O13

(2) The oral verticil of four capitate tentacles tends to be
constant, except in the few cases where all the tentacles were
irregularly scattered.

(8) The variations in symmetry show how easily H.
cooperi could have descended from a Pennarian ancestor,
where the capitate tentacles are present in a considerable
number, and are quite irregularly scattered.

(4) An important pomt to consider is whether the dis-
tribution of symmetry ina population throws any light on
the steps by means of which this symmetry was acquired.

It may be assumed that symmetry of this nature could not
easily be acquired by imperceptible steps, it would more
readily be acquired by larger steps, or in other words by
“ discontinuous” variations.

On account of the tendency which the observations have
to form a symmetrical curve, it may be considered possible
that the scale of symmetry suggested has a natural meaning,
and that the steps by which the symmetrical arrangement
has been acquired correspond to the degrees of symmetry
detailed above. For, it may be observed, that in no case did
I find other combinations, such as three tentacles in the upper
whorl and six in the lower, or three in the upper and five in
the lower, or five in the upper and three in the lower, etc.

The variations appear to be sharp and definite, so that an
unit amount of variation, so to speak, is a discrete and quite
perceptible quantity.

It is unfortunate that the scarcity of material prevents a
more thorough and exhaustive enquiry.

ane
1 = . ‘ ~ _
~ iy ee
» 7 P : a
: -
- =i » 4
= 7 "
7 5
E = ;
=_ i
- i‘
S : : «
= é
= ‘ ri q
% i
> ad —
- I
ke
~ a :
ie , i
- A 7
i i - 1 ~
)
a r i
: :
7
= : - f
P ‘
= —
e 4 a
‘ ¢ t a “0 @

LARVA OF A FLY IN THE HUMAN INTESTINE. 25

Note on the Larva of a Fly (Sarcophaga sp.)
occurring in the Human Intestine.

By
Ernest Warren, D.Sc.Lond.,
Director of the Natal Government Museum.

In June of 1903 the Rev. A. T. Bryant, of Natal, sent to
me for identification some larvee in spirit expelled from the
intestine of a Kafir girl. ‘his gentleman informed me that
such larvee frequently occur among the Kafirs.

An examination of the larvee showed that they were the
orubs of a fly.

In January of 1904 Dr. Campbell Watt, of Maritzburg,
brought to me two living larve, about $ in. in length, which
had been passed by a female patient (Huropean) having the
ordinary symptoms of worms. I was informed that dozens
of these larvee had been passed from time to time. ‘The diet
of the patient had been extremely abstemious, and included
but little flesh food of any kind. The grubs were passed _
more or less intermittingly in batches.

The larve were obviously fly larve, and resembled in every
. way, except in their small size, those sent to me by Mr.
Bryant.

The two grubs were placed in an incubator at 90° F. in a
glass-capsule and supplied with excrement. One of the
specimens, unfortunately, escaped from the capsule and died.
In four days the second specimen was full grown, and had
increased in length from 4 in. to about ?in. The larva now
became extremely restless, crawling round and round the

VOL. 1, Part 2, 16

216 ERNEST WARREN.

capsule with the greatest uneasiness. It was therefore re-
moved from the incubator, and on the following day it
pupated. After twelve days the fly emerged, the pupa having
been kept on slightly damp mould at the ordinary tempera-
ture of the laboratory.

The fly closely resembles the common flesh-fly of the genus
Sarcophaga.

In February of the same year Dr. Watt sent to me in spirit
asimilar grub passed by a white male patient.

A few months ago I forwarded the bred fly to Mr. Ernest
KE. Austen, of the British Museum, aud he has very kindly sent
to me a technical description of it, and also some interesting
observations on the parasitic habits of Sarcophaga.

Unfortunately the specimen is a female, and it appears
that it is not possible to determine the species with any
certainty from a solitary specimen of the female of a Sar-
cophaga.

The following is Mr. Austen’s description of the specimen:

“Sarcophaga sp., bred from larva from human intes-
tine. 92. Length, about 12mm. Grey, with dark markings.
Head: face and sides of anterior portion of front silvery ;
frontal stripe deep black; palpi black, with paler tips.
Thorax with three sharply-defined black longitudinal stripes,
of which the median one extends nearly to the posterior
margin of the scutellum. Abdomen marked with shimmer-
ing patches, which appear lighter or darker according to the
direction from which the light falls on the insect, and
marked with a median black longitudinal stripe and a stripe
on each side of this midway between the median stripe and
the lateral margin ; the admedian stripes are interrupted on
the hind margins of the segments, and do not reach the tip
of the abdomen. Apical segment orange.”

Mr. Austen further writes that he has ‘two species of
Sarcophaga bred from larve from a sore on a girl’s foot
in British Guiana. In Russia an allied species frequently
infests the nasal sinuses of human beings and domestic
animals. There are cases on record in which larvee of species

LARVA OF A FLY IN THE HUMAN INTESTINE. 217

of Sarcophaga have been expelled from the human intes-
tinein Europe, and it has been stated that they are introduced
into the stomach with raw meat. In the case of Kafirs the
eating of raw meat may well afford the explanation.”

The common Sarcophaga is viviparous, and the newly-
born larvee on being placed on animal matter instantly
penetrate beneath the surface, and so disappear from sight.
Itis probable in the present case that a number of minute
newly-born larve were swallowed with food, and that they
lived in the stomach or intestine. It is clearly impossible
that dozens of large (4 in. long) larvee should have been
inadvertently swallowed.

The great restlessness of the larvee when fully grown would
probably assist them in being expelled from the body, so that
pupation could occur outside of the host.

In the present case it is certain that raw meat was not
eaten; but I have found that a female of a Sarcophaga,
enclosed in a box without food of any kind, nevertheless
produced large numbers of larvae, which could be observed
escaping from the body of the parent without any apparent
violition on the part of the fly.

From this observation it 1s seen that a female, enclosed in a
receptacle containing bread or any other food, would readily
let fall considerable numbers of minute white grubs, which
on adhering to bread or many other articles of food would
be practically invisible.

These observations are of value in showing the importance
of keeping flies away from all food and kitchen utensils.

It is interesting to observe that the fly belongs to the
family Muscide, and not to the modified parasitic family
the Oestride or Bot flies. It is a quite typical flesh-fly,
identical or closely similar to a species very common in this
district.

The larva possesses no special organs for adhering to the
mucous membrane of the alimentary canal, and it is very
possible that the parasitic habit is purely accidental.

It is to be surmised, either that the species in question has

218 ERNEST WARREN.

a general tendency to become parasitic, and it would then
be a matter for conjecture as to the modifications which
would thereby be ultimately induced in the species, or that
the larve have an extraordinary power of adapting them-
selves to various environments, so that they are indifferent
whether they live in the alimentary canal of a mammal or
on ordinary decomposing animal matter.

Vol. I, part 3, issued May 15th, 1908.

FRESH-WATER FISHES, BATRACHIANS, AND REPTILES. 219

On a Collection of Fresh-water Fishes, Batra-
chians, and Reptiles from Natal and Zululand,
with descriptions of New Species.

By
G@. A. Boulenger, F.R.S.

With Plates XXXV and XXXVI.

THe collection dealt with was forwarded to me from the
Natal Museum for identification, and the following faunistic
lists possess considerable interest.

Descriptions of the new species are given after the lists.
The types of the new species are being preserved in the British
Museum.

FRESH-WATER FISHES.

Characinide.

Alestes natalensis Blgr.

Pools at Elcheleselwane, Zululand.

Cyprinide.

Labeo darlingi Blgr.
Mkusi River, Zululand.

Barbus gibbosus Peters.
Pools at Indukuduku, Zululand.

voL. 1, PART 3. 7

220 G. A. BOULENGER.
Barbus decipiens Blqr.
Pools at Elcheleselwane, Zululand; Mkusi R., Zululand. -
Neobola brevianalis sp. n.
Mkusi River, Zululand.
Siluride.

Clarias capensis. 0. & V.

Pietermaritzburg, Natal.
Clarias gariepinus (Burch.).
Richmond, Natal.
Anguillide.

Anguilla mossambica Peters.
= A. bengalensis Ham.-Buch.

Dorp Spruit, Pietermaritzburg, Natal.
Anguilla virescens Pefers.
Usilonda Lake, near Kosi Bay, Zululand.
Cyprinodontide.

Haplochilus johnston Gthr.
Pools at Indukuduku, Zululand.

Haplochilus myapose sp. n.

Myaposa River, Zululand.

Cichlide.

Haplochromis moffatti (Casteln.).

The dentition varies as in the closely allied H. desfon-
taines1 (Lacep.) and specimens have in consequence been

FRESH-WATER FISHES, BATRACHIANS, AND REPTILES. PPA k

referred to two genera, Paratilapia (P. moffatti Casteln.)

and Tilapia (T. philander M. Weber). As a rule the outer

teeth are conical in the males, and bicuspid in the females.
Pools at Indukuduku, Zululand, Lake Sibayi, Zululund.

Tilapia sparrmani 4. Smith.

Indukuduku, Zululand.

Tilapia melanopleura A. Dum.

Lake Sibayi and Mkusi River, Zululand.

Tilapia natalensis (M. Weber).
Mkusi R., Indukuduku and Lake Sibayi, Zululand.

Tilapia mossambica (Peters).

Lake Sibayi, Zululand.

Gobiide.

Gobius ewneo-fuscus Peters.

Mseleni River, Zululand; Umsindusi, Pietermaritzburg.

BATRACHIA.

Dactylethride.

Xenopus levis (Daud.).
Mseluzi R. and Myaposa R., Zululand; Impendhla, Natal.

Bufonide.

Bufo regularis Reuss.

Mseleni, pools at Indukuduku, Zululand.

Bufo carens (A. Smith).

Botanic Gardens, Pietermaritzbure.

i)
bo
bo

G. A. BOULENGER.

Engystomatide.

Cacosternum boettgeri (Blgr.).
Natal.

Breviceps verrucosus Rapp.
Natal.

Breviceps mossambicus Peters.

Mseleni, Hlabisa, Zululand.

Hemisus guttatum (Rapp.).
Indukuduku, Zululand.
Ranide.

Rana adspersa (D. & B.).

Mseleni, Zululand.

Rana natalensis (4. Smith).

Pietermaritzburg, Natal.

Rana mascareniensis D. & B.

Mseleni, Zululand.

Rana queketti Blgr.
Natal.

Rana oxyrhynchus 4A. Smith.
Kosi Bay, Zululand; Natal.

Rana fasciata A. Smith.
Natal.

Phrynobatrachus natalensis A. Smith.

Mseleni, Black Umfolosi and Hlabisa, Zululand ; Natal.

FRESH-WATER FISHES, BATRACHIANS, AND REPTILES.

bo
be
oo

Arthroleptis wahlbergi dA. Smith.

Hlabisa, Zululand; Richmond, Natal.

Rappia concolor (Hallow.).

Lower Umluluzi, Zululand.

Rappia undulata Blgr.

Pietermaritzburg, Natal.

Rappia cinetiventris (Cope).
1

Elcheleselwane and Indukuduku, Zululand; Pietermaritz-
burg, Natal.

LACERTILIA.

Geckonide.

Lygodactylus capensis (A. Smith).

Indukuduku, Zululand; Natal.

Hemidactylus mabouia (Mor.).

Mseleni, Zululand.

Homopholis wahlbergii (dA. Smith).

Mselemi, Zululand.

Pachydactylus capensis (A. Smith).

Junction of the two Umfolosi Rivers, Kosi Bay, Zululand ;
Bergville, Natal.

Pachydactylus maculatus Gray.

Junction of the two Umfolosi Rivers, Entendweni, Zululand;
Thornybush, Natal.

224: G. A. BOULENGER.

Agamice.,

Agama armata Peters.

Eshowe, Entendwene, Kwambonambi and Indukuduku,
Zululand; Natal.

Agama atricollis A. Smith.

Indukuduku and Mseleni, Zululand; Natal.

Zonuride.

ZAonurus warreni Sp. 1.

Ubombo, Zululand.

Zonurus vittifer Reichen.

Junction of the two Umfolosi Rivers and Ubombo, Zulu-
land.

Pseudocordylus microlepidotus (4. Snith).

Balgowan, Natal.

Platysaurus guttatus A. Sith.

Ubombo, Zululand.

Agrees with thé specimen from Mashonaland, noticed in
‘Proc. Zool. Soc, 1802 3n, pol:

Chamesaura enea (Wiegm.).

Natal.

Chamesaura anguina (L.).
Natal.
Chamesaura macrolepis (Cope).

Indukuduku and Mseleni, Zululand ; Pietermaritzburg,
Natal.

FRESH-WATER FISHES, BATRACHIANS, AND REPTILES. 225

Varanide.

Varanus albigularis (Daud.).

Isitasa and Ubombo, Zululand.

Varanus niloticus (Z.).

Pietermaritzburg, Natal.

Amphisbenide.

Amphisbena violacea Peters.
Kosi Bay, Zululand.

Lacertide.

Nucras tessellata (A. Smith).

Junction of the two Umfolosi Rivers, Zululand.

‘wo specimens agreeing with the variety ornata Gray
(holubi Stdr., camerani Bedr.) in the short foot, not longer
than the skull in the adult, and in the colouration ; but with
smaller scales, 58-60 across the middle of the body. ‘The
larger specimen measures 95mm. from snout to vent.

Nucras delalandii (M.-Hdw.).
Zwaartkop, Natal; Drakensberg at altitude of 6000 ft.,
Natal.
Ichnotropis capensis (4. Smith).

Mselem, Zululand.
Ichnotropis squamulosa Peters.
Kosi Bay, Zululand.
Gerrhosauride,

Gerrhosaurus grandis sp. 1.

Ubombo, Zululand.

226 G. A. BOULENGER.

Gerrhosaurus flavigularis Wiegm.

Mseleni, Zululand ; Pietermaritzburg, Natal.

Tetradactylus africanus (Gray).
Melmoth, Zululand.

Scincide.

Mabuia homalocephala (Wiegm.).
Indukuduku and Mseleni, Zululand.

Mabuia quinqueteniata (Licht.).
Ubombo, Zululand.

Mabuia varia (Peters).

Junction of the two Umfolosi Rivers, Zululand ; Thorny-
bush, Natal.

Mabuia striata (Peters).

Indukuduku, Mseleni, and Ubombo, Zululand ; Town Bush,
Pietermaritzburg, and Estcourt, Natal.

Ablepharus wahlbergii (A. Smith).
Unmntolosi Drift, Kosi Bay, and Ubombo, Zululand.
A specimen from Ubombo is abnormal in having the pre-
frontals in contact in the middle line, separating the fronto-
nasal from the frontal.

Scelotes bipes (L.).
Indukuduku, Zululand.

Scelotes guentheri Blgr.

Junction of the two Umfolosi Rivers and Ubombo, Zulu-
and; Drakensberg at altitude of 6000 ft., Natal.

Scelotes inornatus (A. Smith).
Kosi Bay, Zululand.

FRESH-WATER FISHES, BATRACHIANS, AND REPTILES. 227

Herpetosaura aren icola Peters.

Mseleni, Zululand.

Acontias plumbeus Biane.
Kosi Bay, Zululand.

Anelytropide.
Typhlosaurus aurantiacus Peters.
Mselemi, Zululand,
RHIPTOGLOSSA.
Chameleontide.

Chameleon dilepis Leach.

Indukuduku, Zululand.

Chameleon teniobronchus A. Smith.

Town Hill, Pietermaritzburg, Natal.

Chameleon damaranus Bblgr.

South Africa.

OPHIDIA.
Typhlopide.
Typhlops mossambicus (Peters).
Mseleni, Zululand.
Typhlops bibronii (A. Smith).

Bulwer, Pietermaritzburg, Hilton Road, Natal.

Glauconiide.

Glauconia distanti Blgqr.
Junction of the two Umfolosi Rivers, Kosi Bay, Zululand ;
Natal.

228 G. A. BOULENGER.

Glauconia conjuncta (Jan.).
Mseleni, Zululand; Natal.

Colubride.

Tropidonotus levissimus (Gthr.).
Natal. :
The habitat of this rare snake was previously unknown.

Ablabophis rufulus (Licht.).
Natal.

Boodon infernalis Gthr.

Zwaartkop, Natal.

Boodon lineatus D. & B.
Mseleni, Zululand ; Natal.

Lycophidium semiannulus Peters.
Kosi Bay, Zululand.
Differs from the type in the absence of dark cross-bars on
the body.
Lycophidinm capense (A. Smith).

Mseleni, Zululand ; Pietermaritzburg, Natal.
Simocephalus capensis (4. Smith).
Durban, Natal.
d
Pseudaspis cana (L.).

Springvale, Pietermaritzburg, Natal.

Chlorophis hoplogaster (Gthr.).
Mseleni, Ubombo, Zululand; Natal.

Chlorophis natalensis (A. Snuth).
Dargle Road, Natal.

FRESH-WATER FISHES, BATRACHIANS, AND REPTILES.

Philothamnus semivariegatus A. Smith.

Umfolosi Drift, Zululand.

Prosymna ambigua Bocage.
Ubombo, Kosi Bay, Zululand.

Prosymna jani brane.
Kosi Bay, Zululand.
Homalosoma lutrix (L.).

Melmoth, Zululand; Hilton Road, Natal.

Homalosoma variegatum Peters.

Mseleni, Zululand.

Dasypeltis scabra (L.).

Mseleni, Ubombo, Zululand; Pietermaritzburg, Natal.

Leptodira hotambeia (Lauwr.).
Mseleni, Zululand ; Natal.

Trimerorhinus rhombeatus (L.).

Pietermaritzburg, Natal.

Trimerorhinus triteniatus (Gthr.).

Cedara, Natal.

Psammophis sibilans (L.).

DOQ

a

Kosi Bay, Mseleni, Zululand; Greenwood Park, Pieter-

maritzburg, Natal.

Psammophis crucifer (Daud.).

Hilton Road, Natal.

Thelotornis kirtlandii (Hallow.).
Kosi Bay, Ubombo, Mseleni and Hlabisa, Zululand.

230 G. A. BOULENGER.
Dispholidus typus (4A. Smith).

Kosi Bay, Zululand; Pietermaritzburg, Natal.

Amblyodipsas microphthalma (Biane.).
Kosi Bay, Zululand.

Calamelaps warreni sp. n.
Kosi Bay, Zululand.
Macrelaps microlepidotus (Gthr.).
Natal.
Aparallactus capensis A. Smith.

Junction of the two Umfolosi Rivers, Kosi Bay, Zululand.

Klapechis decosteri (Blgr.).
Kosi Bay, Zululand,

Elapechis sundevallii (4. Smith).
Natal.
Naia haie (Z.).
Mseleni, Zululand.
Naia nigricollis Reinh.
Pietermaritzburg, Natal.
Sepedon hemachates (Lacep.).

Natal.

Homorelaps lacteus (L.).
Natal.

Dendraspis angusticeps (4. Smith).
Mount Edgecombe, Natal.

Viperide.

Bitis arietans (Gray).
Zululand ; Natal.

FRESH-WATER FISHES, BATRACHIANS, AND REPTILES. 23]

Atractaspis bibronii A. Smith.

Junction of the two Umfolosi Rivers, Entendweni, Zulu-
) )
land.

DESCRIPTIONS OF THE TWO NEW FRESH-WATER FISHES.

Neobola brevianalis sp.n. (Text-fig. 1.)

Depth of body 43 times in total length, length of head 3}
times. Snout obtuse, not projecting beyond the mouth, shorter
than the diameter of the eye, which is 31 times in length of

TExT-FIG. 1.

nts sy

ee y
ne
ae

head; mouth extending to below anterior third of eye. Dorsal
with II 7 rays, its origin slightly in advance of that of the
anal; its distance from the end of the snout twice and + its
distance from the caudal ; first branched ray longest, about 2
length of head. Anal II 12. Pectoral acutely pointed, as long
as head, reaching root of ventral. Caudal peduncle once and
2 as long as deep. Caudal deeply forked. Scales 52 7 1
between lateral line and root of ventral, 16 round caudal
peduncle. Yellowish, with a silvery lateral band ; fins white.

Total length 37 millm.

A single specimen from the Mkuzi River, Zululand. Three
species of Neobola were known: N. bottegi Vincig. (Lake
Rudolf, Gallaland, Somaliland), N. argentea Pellegr. (Lake
Victoria),and N.minuta Blgr. (Lake Tanganyika). N. brevi-
analis differs from its congeners in the lower number of anal
rays (14 instead of 18 to 20).

Io, G. A. BOULENGER.

Haplochilus myapose sp. n. (Text-fig. 2.)

Depth of body 4 to 44 times in total length, length of head
32 to 4times. Snout very short, truncate, with the lower Jaw
projecting ; eye longer than the snout, as long as or a little
shorter than postocular part of head ; interorbital space about
half length of head. Dorsal with 10 rays, originating above
anterior third of anal, nearer to root of caudal than to occiput ;
median rays longest, about half length of head. Anal with 14
or 15 rays, median longest, the fin rounded, like the dorsal.
Pectoral reaching a little beyond base of ventral. Candal
rounded, as long as head. Caudal peduncle once and + as long
as deep. Scales 27—28 in a longitudinal series, 16 round body in

TEXT-FIG. 2.

front of ventrals. No lateral line pits. Pale olive, with darker
edges to the scales ; fins greyish.

Total length 28 millim.

Four specimens from the Myaposa River, Zululand.

This little Cyprinodontis is so closely allied to H. pumilus,
from Lakes Tanganyika and Victoria, that I have hesitated
before describing it as distinct. It differs, however, in having
the eve as long as, or but httle shorter, than the postocular
part of the head and in having the dorsal and anal fins more
rounded, the median rays being the longest.

DESCRIPTION OF THE THREE NEW REPTILES.

Zonurus warreni spn. (Pl. XXXV.)

Head longer than broad, strongly depressed. Head-shields
rugose ; frontonasal as long as broad, forming a narrow suture
with the rostral, separating the nasals; latter scarcely swollen ;
nostrils in the posterior part of the nasal; prefrontals in con-
tact with their inner angles, or forming a short suture ; frontal

FRESH-WATER FISHES, BATRACHIANS, AND REPTILES. 233

hexagonal, shehtly widened anteriorly ; frontoparietals broader
than long ; four equal parietals with a small hexagonal inter-
parietal between them; a row of six short occipital spines ;
temporals large, keeled; four or five small temporal spines ;
four supraoculars ; three or four supraciliaries ; lower eyelid
opaque; loreal and preocular large; three suborbitals;
rostral nearly three times as broad as deep; seven upper
labials. Symphysial forming a very open angle posteriorly ;
six lower labials, bordered below by five large shields; small
irregular chin-shields ; gular scales small, obtusely keeled ;
larger keeled scales under the neck; sides of neck with small
erect spines. Dorsal scales forming regular transverse series,
obtusely keeled on the back, spinose on the sides; about 40
transverse series of about 20 shields between the well-marked
lateral folds. Ventrals quadrangular, mostly broader than
long,

pair of feebly enlarged preanal plates, with smaller ones in

outer keeled, others smooth; 14 longitudinal series. A

front and on the sides. Limbs above with large spinose
imbricate keeled scales; 12 femoral pores on each side, with
3 or 4 rows of callose scales in front of them. Tail with
whorls of large, strongly keeled, spinose scales, separated
from each other by whorls of smaller scales. Dark brown
above, with small yellow black-edged spots forming more or
less regular transverse series on the body; lower parts pale
brown, lower lip and throat spotted or marbled with darker.

Total length ; . 270 millim.
Head : ; 1 oe
Width of head. : a ae
Body : ; : ae
Fore limb : eo aa
Hind limb , , 2 100s Se
Tail ; . 160

3)
Two male specimens from Ubombo, Zululand.

Gerrhosaurus grandis sp.n. (Pl. XXXVI.)
Form stout, head and body much depressed. Head a little
longer than broad. Head-shields rugose ; frontonasal as lone

234 G. A. BOULENGER.

as broad, narrowly in contact with the rostral; prefrontals
forming a long median suture ; frontal as long as its distance
from the rostral or the frontoparietals and interparietal
together ; a narrow, band-like shield on the anterior border of
the ear-opening ; three or four upper labials anterior to the
subocular. Dorsal shields rugose, strongly keeled, in 14
longitudinal and 32 transverse series; ventrals in 10 longi-
tudinal series. 12 or 15 femoral pores on each side. Tail
once and a half the length of head and body, cylindrical.
Head and body pale brown with black spots, which become
confluent posteriorly, the upper surface of the body being
black with elongate yellowish spots corresponding with the
shields ; a yellowish lateral band from above the ear, becoming
gradually indistinct by being broken up into spots on the
posterior part of the body ; chin and throat white; belly and
lower surface of limbs and tail brown.

Total length . 475 millim.
Head ; : : 4 AON
Width of head. ec ee
Body) = : : satOO)) aa
Fore limb , Ne oper.
Hind limb : eR Sane
Tail : 285

A single specimen from Ubombo, Zululand.

This fine lizard agrees in size with G. validus A. Smith, or
G. major A. Dum. It differs from both in the number of
longitudinal series of dorsal shields, and from the former by
the number of longitudinal series of ventral shields, the
separation of the nasal shields by the rostral and fronto-
nasal, and the low number of femoral pores. In coloration it
resembles the former.

Calamelaps warreni sp.n. (Text-fig. 3.)

Rostral large, a little broader than deep, the portion visible
from above as long as its distance from the frontal; inter-
nasals much broader than long, half as long as the prefrontals;

FRESH-WATER FISHES, BATRACHIANS AND REPTILES. 23%

frontal hexagonal, once and a half as long as broad, longer
than its distance from the end of the snout, shorter than the
parietals ; nasal entire; supraocular small; a very small post-
ocular; a single temporal; six upper labials, second and third
in contact with the prefrontal, third and fourth entering the
eye, fifth very large and forming a long suture with the
parietal ; four lower labials in contact with the anterior chin-
shields, fourth very large and narrowly separated from its

TEXT-FIG. 3.

fellow by the chin-shields. Scales in 19 rows. Ventrals 161 ;
anal divided ; sub-caudals 26. Uniform plumbeous grey.

Total length 235 millim.; tail 25.

A single female specimen from Kosi Bay, Zululand.

In the number of rows of scales this species is intermediate
between C. unicolor Reinh. and C. polylepis Bocage ; it
differs from both in having the nasal shield entire. C.mironi
Mocquard, recently described from Natal, is stated to have
the nasal divided, the scales in 17 rows, and only 133 ventral
shields.

Vol. | PART.o: 18

hd

i a b fall ae Bane se | me _ i

1
: = = 4 ‘ae
= =f)
ty i .
I
Eas * .: ue
~— » = af
TV =e a ¢ 7
i . a
y nd
7 ¥ f
b
- 7 - @ Pa -
ir : = ow
= f :
os ha — / : i
a
1 ~
=
. -
J nt ; oi
> : ‘a *
— a sod Z ‘ 7
o '
5 ay :
: 2 i 4
| 7 ¥ 7
- ao '
i
.
#
AY, = =
. ee. : j =A
J
Ko
H
, l '
- *
’ ™ ~~
*
0 ! 7
ae
a iz
a EB a
: a. Tas 4 ae
2 : vm
i ia
1 \ %
1
& :
7 *
-
UI
- - 7 gr Dt
7 7
. ee a
‘
= a
“ I
”
a
i -
if
. 1

ra
4
=
—
)
7,
ro

imp.

J.Green

WARRENI.

CN
2)

BSA

7

e)

IN

7

3

A
AMT

er
a a on ae

2

os)
NI

NOTE ON CLARIAS CAPENSIS.

Note on Clarias capensis C. &

By
G. A. Boulenger, F.R.S.

AtrnoucH ‘Barbels,’ as the Clarias are called by the
colonists, are common over a considerable portion of South
Africa, little attention has been paid to their characters. A
species described in 1840 in Cuvier and Valenciennes’s great
work, Histoire des Poissons, vol. xv, p. 377, from a
single stuffed specimen labelled as from the ‘Cape of Good
Hope,’ a term which in those days was often taken to mean
South Africa, has until lately been a puzzle to ichthyologists.
This Clarias capensis was regarded by Giinther as a
doubtful synonym of Clarias gariepinus (Burchell), 1822,
a specimen of which had been described and figured by
Andrew Smith under the former name. Some years ago, I
had an opportunity of taking notes on the type specimen of
Cl. capensis, preserved in the Paris Museum, and I
expressed the opinion (Poiss. Bass. Congo, p. 255, 1901)
that it constitutés a species distinct from Cl. gariepinus,
the common ‘Barbel, the range of which extends from
Angola and the Zambesi to the Orange River and Natal.

The specimen of Clarias capensis remained unique
until a few months ago I received from Dr. Warren another
590 millim. (about 21 in.) long, procured from the pond in the
Botanic Gardens at Pietermaritzburg, which answers in
every important pomt to the long-sought-for Clarias
capensis. It is distinguished at a glance from Cl. garie-
pinus, of which I have also received specimens from Dr.
Warren, in having the caudal part of the body (behind the

238 G. A. BOULENGER.

ventral fins) more elongate,| and the space between the
pointed occipital process and the origin of the dorsal fin
greater (more than one fourth of the length of the head
neasured to the extremity of the occipital process). These
two characters ought I think to enable anyone to identify
further examples, which I hope may soon turn up. In the
meantime I here give a description of the specimen for which
the British Museum is indebted to Dr. Warren:

Depth of body 73 times in the total length, length of head
A times. Head once and } as long as broad, its upper surface
coarsely granulate ; occipital process angular; frontal fonta-
nelle nearly 4 times as long as broad, $ the length of the
head ; occipital fontanelle very small, well in advance of the
occipital process; eye very small, its diameter 4 times in the
length of the snout, 7 times in the interorbital width, which
equals the width of the mouth and 3 the length of the head ;
band of premaxillary teeth 6 times as long as broad; band of
vomerine teeth a little narrower than the premaxillary band,
rather widely interrupted in the middle, composed of small
partly pointed, partly granular teeth. Nasal barbel 4 the
leneth of the head; maxillary barbel as long as the head,
reaching middle of pectoral spine ; outer mandibular barbel ¢
the length of the head, inner about 3. Gnll-rakers on first
arch long and closely set, 55 in number. Clavicles hidden
under the skin. Dorsal fin with 65 rays, its distance from
the occipital process 2? the length of the head, its distance
from the caudal fin 4 times the diameter of the eye. Anal
fin with 50 rays, narrowly separated from the caudal.
Pectoral fin } the length of the head, the spine feebly serrated
on the outer border, + the length of the fin. Ventral fins
once and } as distant from the root of the caudal fin as from
the end of the snout. Caudal fin } the length of the head.
Dark olive brown above, whitish beneath.

The type specimen, which measures only 480 millim., has
a rather larger eye,—its diameter 3 times in the length of

‘In Cl. gariepinus the ventral fins are nearly equally distant from
the end of the snout and from the root of the caudal fin.

NOTE ON CLARIAS CAPENSIS. 239

the snout, 5 times in the interorbital width; the frontal
fontanelle is 3} times as long as broad and 4 times in the
length of the head. Dorsal fin with 72 rays, anal with 52 ;
the distance between the occipital process and the dorsal is 4
the length of the head; the distance between the dorsal and
the caudal equals the diameter of the eye. The pectoral
spine measures only 3 the length of the fin.

Perhaps Cl. capensis grows to the same large size as
Cl. gariepinus, of which | have seen a skull measuring 260
millimetres, brought home from the Tugela by Capt. J. W. H.
Seppings, R. Dublin Fusiliers.

. = a.
= —- a a
nm

a
7
cf 5 :
~ _
|
|
= ,

Pep ree we )

"a

=
2
}
4
- ¢
*
7
ined
7
if

-
’
ee
—

=
a
ant
@

i
=
i
te
: a
,
“
\
aaa
<
‘

FISHES FROM NATAL, ZULULAND, AND CAPE COLONY. 24]

A Collection of Fishes from the Coasts of Natal,
Zululand, and Cape Colony.

By
Cc. Tate Began, WI.A.

With Plates XXX VIJ—XLII.

A coLLEcTION of marine fishes from Natal, Zululand, and
Cape Colony, received from Dr. Warren, is of considerable
interest. The specimens were obtained at five localities, viz.
Kosi Bay, Zululand; Scottburgh, Durban Bay, and Congella,
near Durban, Natal; and sixteen miles N.E. of Bird Island,
Algoa Bay.

A complete list of the collection is first given, and after-
wards the new species are described.

I.—SELACHII.

Fam. CArcHARID®.
Mustelus vulgaris M. § H. . Bird Island.

Fam. ScyLIORHINID®.

Scylorhinus africanus (L.) ; . Bird Island.
4 edwardsii (Cuwv.) 2
7 variegatus (A. Smith) -
a natalensis (Rgn.)

Fam. SQuALIDA.
Squalus acutipinnis sp.n. . : . Bird Island.
Pliotrema warreni Rgn.

3)

242 CG. TATE REGAN.

Fam. SQUATINID.

Squatina africana sp. 2.

Fam. 'ORPEDINIDA.

Durban Bay.

Torpedo marmorata Risso... Bird Island, Congella, and

Algoa Bay.

Astrape capensis (L.)

Fam. R#AINOBATIDA.
Rhinobatus columne M. & H.

Bird Island.

Bird Island.

* blochii M. & H. Bird Island and Durban Bay.

Fam. Rapa.
Raia marginata Lacep.
» ocellifera Rgn.
» Vvhizacanthus Rgn.

Fam. DasyBatip#.
Dasybatis uarnak (Forsk.) .
Myliobatis aquila (L.)

Fam. CHimarip®.
Callorhynchus antarcticus (Lacep.)
I1.—TELEOSTOMI.
Fam. CLUPEIDE.
Clupea durbanensis Rgn. .

Spratelloides delicatulus (Benn.)

Fam. SILurip2®.
Galeichthys feliceps C. & V.

Bird Island.

Durban Bay.

Bird Island.

Bird Island.

Durban Bay.

Kosi Bay.

Bird Island.

FISHES FROM NATAL, ZULULAND, AND CAPH

Fam. ANGUILLIDA.

Murenesox cinereus (forsk.)

Ophichthys unicolor sp. n.

ea kirkii Gthr.
Fam. Mura@nip™.

Murena polyophthalmus Blkr.

- macrurus Blkr.

Fam. SCOMBRESOCIDA.

Hemirhamphus dussumieri C. & V.
Belone robusta Gthr.

Fam. MuaGILip#.

Mugil robustus Gthr.
ceylonensis Gthr.

>)
if smithii Gthr.
_ conustantize C. & V.

Fam. ATHERINIDA.

Atherina pinguis Lacep.

Fam. GADID#.

Merluccius capensis Casteln.

Fam. PLEURONECTID#.

Platophrys pantherinus (Ripp.)
Pseudorhombus russell (Gray)
Paralichthodes algensis Gilchr.
Synaptura ciliata Gilchr.

aS pectoralis Kaup.
Plagusia marmorata Blkr.

COLONY. 243

Durban Bay.
Bird Island.
Kosi Bay.

Durban Bay.
Kosi Bay.

Kosi Bay.

”

Kosi Bay.
*

”

Durban Bay.

Kosi Bay.

Bird Island.

Kosi Bay.
Durban Bay.

d

Bird Island.

Kosi Bay.

DAA C. TATE REGAN.

Fam. SERRANID®.

Kpinephelus sonnerati (C. & V.)

i andersoni Blqr.
r tauvina (Forsk.)
- hemistictus (Ripp.).
. maculatus (Bl.)

Therapon servus (Bl.)
Parascorpis typus Blkr.
Apogon warren sp. 1. :
Cirrhitichthys maculatus (Lacep.)
Pomatomus saltator (L.)
Lutianus gembra (C.&V.).

a jyohnii (B1.)

x marginatus (C. & JV.)

Fam. PoMADASID®.

Pomadasys teniophorus Sp. 1.

e multimaculatus (Playf.)

- hasta (Bl.) : :

. opercularis (Gthr. & Playf.)
Diagramma griseum C. & JV.

ms affine Gthr.

f crassispinum Riipp.

Fam. SPaRID#.

Cantharus emarginatus C. & J.
Dentex argyrozona C. & J.
a undulosus sp. n.
Lethrinus nebulosus (Forsk.)
Sargus cervinus (Lowe)

" capensis A. Smith
ve holubi Stdr. . He
2 nigrofasciatus sp. 2.

Pagrus laniarius C. & JV.

Durban Bay.
3)

Congella.

Kosi Bay.

3)
Durban Bay.

Bird Island.
Durban Bay.
3)

Bird Island.
Durban Bay.
+)

i)

Kosi Bay.
Durban Bay.
3)

Kosi Bay.
Durban Bay.
Kosi Bay.

Bird Island.
Kosi Ray.

+)

Kosi Bay and Durban Bay.

3)

Bird Island.

3)

FISHES FROM NATAL, ZULULAND, AND CAPE COLONY. 245

Chrysophrys gibbicepsC.&V. . . Bird Island.
5 hasta (Bl. Schin.) ; . dXosi Bay.
7s bifasciatus (Forsk.) . *
~ sarba (Forsk.) . Durban Bay.
Pagellus lithognathus C. & V. : Bird Island.
. mormyrus (i) . ; . Kosi Bay.

Fam. MuULiipaé.

Mulloides flavolineatus (Lacep.) . . Kosi Bay.

Fam, LioGNarHip»,
Gerres longirostris Ripp. . Durban Bay and Kosi Bay.
a lineolatus Gthr. ; : . Kosi Bay.
Equula edentula (Bl.) . Durban Bay.

Fam. ScL=NIDz. :
Otolithus equidens C. & V. : . Bird Island.
Umbrina capensis Puppé.

res
Scizena aquila Lacep. : %
- margaritifera Haly. . Durban Edy
nl \ ~ 7 > a
Fam. SILLAGINIDE.
Sillago sihama (Forsk.) . Kosi Bay.
ue chondropus Blkr. . : . Durban Bay.
Fam. PEMPHERID.
Pempheris molucca C. & V. , . Kosi Bay.
Fam. Scorprpip®.
Psettus falciformis (Lacep.) , . Kosi Bay.
Fam. CypHosip®.
Cyphosus fuscus (Lacep.) . : - Kosi Bay.
Fam. CH#TODONTID#.
Chetodon vagabundus FL. . ; . Kosi Bay.

- setifer Bl.

3)

2.4.6 C. TATE REGAN.

Fam. Drepanip#.
Drepane punctata (L.) ; . Durban Bay.

Fam. Trururpips.
Teuthis oramin (Bl. Schn.) . ; . Kosi Bay.

Fam. ACANTHURIDE.

Acanthurus triostegus (L.) . Kosi Bay.
S strigosus Benn. ; : 4:

Fam. PoMAcENTRID®.
Glyphidodon sordidus (Forsk.) — . . Kosi Bay.
< celestinusC.&V.. : =

Fam. LABRID®.

Julis umbrostigma Riipp. . . Kosi Bay.
oe elttuemnieytarse (lya) Se Be
Platyglossus scapularis (Benn.) . : :

Fam. ScARIDA.

Pseudoscarus maculosus (Lacep.) . Kosi Bay.

Fam. CARANGID2.

Trachurus trachurus (Z.) . ; . Bird Island.
Caranx carangus (Bl.) . Durban Bay.
" rottleri (BL) : ; 5
A; ciliaris (Bl.) : : : .
*. melampygus C. é& J. : . Kosi Bay.
oe speciosus (forsk.) . : : .

- hippos (L.) : : .
Trachynotus ovatus (Z.) . Kosi Bay and Durban Bay.
Chorinemus sancti-petri C.& V. . Durban Bay.

Fam. T'RICHIURID.

Lepidopus caudatus (Huphras.) . . Bird Island.

FISHES FROM NATAL, ZULULAND, AND CAPE COLONY. 247

Fam. Goprip®.

Kleotris ophiocephalus Ce

Fam. BLENNIID»®.
Blennius bifilum Gthr.
eS punctifer sp. n.
Salarias quadricornis Grae WA
ne rivulatus Riipp.

kosiensis sp. 2.

be)

Fam. OpHIDIID®.

Genypterus capensis (A. Smith)

Fam. ScorpeNID».
Scorpena natalensis Rgn.
ss rosea Day.
re haplodactylus Blkr.
Pterois miles (Benn.)
3 volitans (L.)
Agriopus spinifer A. Smith

Fam. TRIGLID#.

Trigla capensis C. & V.

Fam. PLATYCEPHALID2.
Platycephalus tentaculatus Riipp.
55 insidiator (Forsk.)

Fam. BatistTipa.

Balistes aculeatus LD.

Fam. OsrRacionrip».

Ostracion cubicus JL.

Fam. Terropontrips.
Tetrodon honckenii Bl.
7 immaculatus Bl. Schn.

Kosi Bay.

Kosi Bay.
”
Au

29

Bird Island.

Bird Island.
Durban Bay.
Kosi Bay.
Scottburgh.
Kosi Bay.

93

Bird Island.

Kosi Bay.

3)

Kosi Bay.

Kosi Bay.

Kosi Bay.

Durban Bay.

248 C. TATE REGAN.

DESCRIPTIONS OF THE NINE NEw SPECIES.
Squalus acutipinnis sp.n. (Pl. XXXVILI.)

Acanthias blainvillii (part.) Gunth. Cat. Fish. vii,
p. 419 (1870).

Snout pointed; nasal flaps bilobed; distance from nostrils
to end of snout 3—% that from mouth to nostrils. Base of
second dorsal (without the spine) } of its distance from the
upper caudal lobe and }$ that of the first (without the spine),
which is less than its height and about 2 of its distance from
the second dorsal; spires without ridges or grooves, that of
the second dorsal not quite so high as the fin, in great part
exposed. Pectoral extending well beyond the end of the
base of first dorsal and at least 3 of the distance from last
gill-opening to origin of ventral, with the free edge nearly
straight, the posterior angle nearly a right angle and the
anterior angle much more acute than in 8. blainvillii;
ventrals not nearly reaching the second dorsal. Lower
caudal lobe without posterior notch, its lower edge con-
tinuous with the posterior edge of the upper lobe. Grayish
or brownish above, pale below.

South Africa; Mauritius.

Four specimens, a stuffed one from Mauritius (Robillard),
one of 560 mm. from Natal, presented by Dr. E. Warren, one
of 540 mm. from Table Bay, presented by Dr. J. D. F. Gil-
christ, and one of 190 mm. from the Cape-of Good Hope,
from Sir Andrew Smith’s collection.

Squatina africana sp.n. (Pl. XXXVIII.)

Folds at sides of head not produced into lobes. Outer nasal
flap with entire edges; inner flap with two nearly simple pro-
longations, the outer of which has a fringed lobe at its base.
Distance between the spiracles a little less than the inter-
ocular width. Outer angle of pectoral nearly a right angle ;

FISHES FROM NATAL, ZULULAND, AND CAPE COLONY. 249

distance from anterior angle to posterior end of base of
pectoral 2 the extreme length of the fin. Ventral not
reaching the vertical from origin of first dorsal. Width
of tail (at the base) about 4} of its length. Base of first
dorsal a little more than } its height, which is a little more
than its distance from the second; second dorsal a_ little
shorter, but scarcely lower than the first; interspace between
the dorsals a little less than the distance from second dorsal
to caudal, much less than the distance from base of tail to
origin of first dorsal. Posterior edge of caudal fin notched,
the upper lobe vertically truncate above, the edge becoming
oblique before its junction with the lower lobe, which is
obliquely truncate. Upper surface with small pointed
denticles, each with 5 keels; no median series of enlarged
denticles ; small imbricated denticles at outer edges of paired
fins, extending on to their lower surface and, on the pectoral,
forming an inferior marginal strip about ¢ as wide as the fin ;
denticles on lower surface of tail not extending forward to its
base; lower surface of head and abdomen naked. Brownish,
with numerous pale spots covered with brown reticulations.

A single specimen (¢), 800mm. in total length, from
Durban Bay, Natal.

This species is nearest to Sq. californica Ayres, which
differs im markings and in having the greater part of the
abdomen and of the lower surface of the paired fins covered
with denticles inthe adult. From the Japanese Sq. nebulosa
Regan it differs especially in coloration, in having the folds at
the sides of the head not produced into lobes and in the form
and dimensions of the fins.

The species of Squatina may be arranged thus ;

I. A mid-dorsal series of enlarged denticles present in the
adult.
(A) Distance between the spiracles greater than the inter-
ocular width. 1. japonica Bleek.
(B) Distance between the spiracles not greater than the
interocular width.

950 C. TATE REGAN.

Distance from anterior angle to posterior end of base of
pectoral much more than 4 the extreme length of the fin.
2. armata Philippi.
Distance from anterior angle to posterior end of base of pec-
toral a little more than } the extreme length of the fin.
3. aculeata Cuv.

II. No mid-dorsal series of enlarged denticles in the adult.
(A) Dermal denticles not carinate.
Folds at sides of head produced into an angular lobe on each
side. 4. angelus Dum.
Folds at sides of head not produced into lobes.
5. australis Regan.
(8) Dermal denticles tricarinate.
1. Folds at sides of head not produced into lobes.
Abdomen and lower surface of paired fins, in the adult, in
great part covered by dermal denticles.
6. californica Ayres.
Abdomen naked; lower surface of paired fins with marginal
strips of denticles. 7. africana Regan.
2. Folds at sides of head produced into two convex
lobes on each side. 8. nebulosa Regan.

TEXT-FIG, 1.

ZZ

Ophichthys unicolor sp.n. (Text-fig. 1.)
Teeth pointed, subequal, in a double series in both jaws and

on the vomer. Leneth of head 2 the distanee from g@ill-
¢c to)

0

FISHES FROM NATAL, ZULULAND, AND CAPE COLONY. 201

opening to vent; tail nearly twice as long as the rest of the
fish. Snout nearly twice as long as eye, projecting beyond
the mouth; cleft of mouth about + the length of head,
extending to below the posterior edge of eye. Origin
of dorsal a little behind the end of the pectoral, which is
2 as long as the head. Uniformly brownish.

Sixteen miles N.E. of Bird Island at a depth of 40 fathoms ;
bottom mud.

A single specimen, 260 mm. in total length.

Apogon warreni sp. n. (Pl. XLII.)

Depth of body equal to the length of head, 2% in the length
of the fish. Snout 2 as long as eye, the diameter of which is
34 in the leneth of head; interorbital width 5 in the length
of head. Maxillary extending slightly beyond the vertical
from posterior edge of eye; lower Jaw shorter than the upper.
Scales 25 2 Dorsal VI, I 9; second spine much stronger
and a little longer than the third, nearly } the length of head ;
first branched ray the highest, 3 the length of head ; free edge
of soft dorsal straight. Anal IT 8. Caudal notched. Pectoral
3 the length of head ; ventrals extending to the anal. A dark
band from eye to base of pectoral; a large dark longitudi-
nally expanded spot on the caudal peduncle.

Kosi Bay, Zululand.

A single specimen, 50 mm. in total length.

Pomadasys teniophorus sp. n. (Pl. XXXIX.)

Depth of body 2} to 24 in the length, length of head 34 to
33. Snout shorter than eye, the diameter of which is 35 in
the length of head and nearly equal to the interorbital width.
Maxillary extending to below anterior 4 of eye; depth of
preorbital nearly equal to the diameter of eye; 12 gill-rakers
on the lower part of the anterior arch. Scales 51-54 54.
Dorsal XII 14-15, commencing above the opercular cleft ;
fourth spine the longest, a little more than } the length of
head ; soft dorsal highest anteriorly, with straight or slightly

von. 1, parr 3. 19

yy) C. TATE REGAN.

convex free edge, scaly at the base and with a series of scales
behind each ray ; longest rays less than 4 the length of head.
Anal III 7; second spine the longest, 3 to % the length of
head. Caudal truncate or shghtly notched. Pectoral longer
than the head, extending to above the origin of anal; ventrals
reaching the vent. Five or six pairs of dark longitudinal
stripes on each side of the body, the stripes of each pair
confluent posteriorly ; vertical fins dusky.

Kosi Bay, Zululand.

Two specimens, 260 mm. in total length.

Allied to P. furcatus Bl. Schn., which has III 8-10 anal
rays, and still more closely to P.anas Val., described and
figured by Sauvage in his work on the fishes of Madagascar.
The latter has the snout more produced, the maxillary not
extending beyond the vertical from the anterior edge of the
eye, and the dorsal commencing above the axil of the pectoral,
which is shorter than the head.

Dentex undulosus sp. n. (Pl. XL.)

Dentex rupestris (non Cuv. & Val.) Casteln. Poiss. Afr.
Austral., p. 28 (1861).

Depth of body 2? to 3 in the length, length of head
32 to 32. Snout 14 to 2 as long as eye, the diameter of
which is 34 to 54 in the length of head; interorbital width
3 to 34 in the length of head. Jaws equal anteriorly ;
maxillary nearly reaching the vertical from the anterior
edge of eye; 4 canines in the upper jaw, moderately strong,
the inner pair scarcely smaller than the outer; 6 canines
in the lower jaw, the innermost pair small. Depth of pre-
orbital equal to the diameter of eye (adult) or less (young) ;
cheek with 9 or 10 series of scales; preoperculum scaly ;
14 to 16 gill-rakers on the lower part of the anterior arch.
Scales 57-60 23°. Dorsal XII 10; origin above axil of
pectoral ; spines of moderate strength, the third to the fifth
the longest, 2 ora little more than 2 the length of head ; soft

FISHES FROM NATAL, ZULULAND, AND CAPE COLONY. 2538

rays 4 or nearly 3 the length of head. Anal III 9; second
and third spines subequal, 4 or nearly 4 the length of head.
Caudal forked. Pectoral longer than the head, extending to
above the origin of anal. Upper half of body with from 3 to
6 undulating longitudinal dark stripes with pale edges ; a large
blackish spot on the side above the middle of the pectoral.

Sixteen miles N.E. of Bird Island, Natal, and Table Bay,
Cape Colony (Gilchrist).

Dentex argyrozona Cur. & Val. is nearest to this species,
but has a larger mouth, stronger lateral canines, 6 to 8 scales
in a transverse series above the lateral line, III 8 anal rays
and a different system of coloration.

Sargus nigrofasciatus sp. n. (Pl. XLI.)

Depth of body 2 to 24 in the length, length of head 3 to
34. Snout with nearly vertical profile, longer than eye, the
diameter of which is 44 in the length of head; interorbital
region very convex, its width 22 in the length of head.
Maxillary extending to below anterior 1 of eye; depth of
preorbital equal to diameter of eye; cheek with 5 series of
scales; 10 short gill-rakers on the lower part of the anterior
arch. Incisors moderately broad, implanted vertically in the
upper jaw, obliquely in the lower; 4 or 5 series of molars in
the upper jaw, 3 in the lower. Scales 55-57 54. Dorsal
XI 12, commencing above the axil of pectoral; fourth spine
the longest, } to 2 the length of head; last spine 2 to + the

ra
oO

length of head and as long as the soft rays. Anal III 11;
second and third spines subequal, as long as the last of the
dorsal. Caudal forked. Pectoral longer than the head,
extending beyond the origin of anal; ventrals reaching the
vent. Body with 6 or 7 blackish vertical bars; thoracic
region and upper part and sides of head blackish.

Sixteen miles N.E. of Bird Island, Natal, at a depth of 40
fathoms.

Two specimens, 360 mm. in total length.

Allied to 8. holubi Stdr. which has the preorbital deeper,

254 C. TATE REGAN.

the spinous dorsal lower, the soft dorsal usually of thirteen
rays, and the coloration uniform except for a small dark spot
at the root of the pectoral.

Blennius punctifer, sp.n. (Pl. XLII.)

Depth of body 32 to 4 in the length, length of head 44-42.
Snout obtuse, with nearly vertical anterior profile. Diameter
of eye 35 to 4 in the length of head and twice the width of
the flat interorbital region. Maxillary extending to below
posterior edge of eye; very small canines in the lower Jaw.
On each side a short fringed nasal tentacle and a similar
supra-orbital tentacle ; occiput with a median series of simple
filaments. Dorsal XII 14-15, with a very shght notch, com-
mencing above the edge of preeoperculum, ending just before
the caudal; spinous part as high as or a little lower than the
moderately elevated soft-rayed part. Anal 17-19. Caudal sub-
truncate. Pectoral extending to above origin of anal. Back
with 6 more or less distinct dark cross-bars; numerous very
small dark spots on head, body, dorsal fin and base of pectoral
fin; series of larger spots on caudal, anal and distal part of
pectoral ; a more or less distinct dark spot or ocellus behind
the first dorsal spine.

Kosi Bay, Zululand and Port Natal (Ayres).

Three specimens, measuring up to LOO mm. in total length.

Blennius cristatus Linn. is described by Cuvier &
Valenciennes from the Island of Ascension as having the
supra-orbital tentacles very small and simple, the anal fin with
16 rays, and the markings somewhat different.

Salarias kosiensis sp.n. (Pl. XLII)

Depth of body equal to length of head, 32 in the length of
the fish. Snout obtuse, with nearly vertical anterior profile.
Diameter of eye 5 in the length of head and equal to the
width of the somewhat concave interorbital region. Maxillary
extending to below posterior edge of eye; canines present in
the lower jaw. On each side a short fringed nasal tentacle

FISHES FROM NATAL, ZULULAND, AND CAPE COLONY. 255

and along simple supra-orbital tentacle ; a transverse series of
short filaments across the nape. Dorsal XI 12, deeply notched,
ending just before the caudal; spinous part lower than the
rather elevated soft-rayed part. Anal 15. Caudal rounded.
Pectoral extending beyond the origin of anal. Blackish-grey,
with some white spots and markings.

Kosi Bay, Zululand.

A single specimen, 185 mm. in total length.

This species is very close to 8. variolosus C. & V., which
has the supra-orbital tentacles small and fringed.

j Hi

7 >
Pe eK,

ei

: s ue Fai iy fi en ae 7 ui) _ Cat * es, - Wt, 7] ay qin .
os aay a [PSG hia) ana rd hy ae
ih i I Teak ta irae i. wir T aa) eaten Mr Th
= < oe A wo Voy noe To ‘oa = j Sang ae ai = np
- Wee he SIN = (Pan rie ge tM deh is Spe ios ieee
P. A ; cae ea vk ee one T 7
ee, | ot a esate Hiss vee bn a
7 ae De 1 ae eee Atay haa Ti mo
SS a AR. i p= hae ices aren A Mw a
Ces . 4 alt na seeuive wert ge pbgiruirey Wile) pale
7 a i \ > 7 ca aero a
ee io 7 ra a | : 7 : a : 7
~s : i : 7 7 .

STMMIa ih ow “Ss CaIveRiOsS
Cutt weurmoyyqs9M, UML 9819p PreMpoom W'D

Te el. I TPA STIL OD Teen UUV

Peek Vile

Ann. Natal G. Mus..Vol. I.

West,Newman imp

G.M.Woodward del.et lith.

SQUATINA AFRICANA.

‘SNUOHGOINYL SASVAVWOd =
‘dur weurmen‘qsem y, “UNL 1818p PASMpoom: W'D

“XIXXX Id Toh soy 4) [eos cay

ff

SMSOTMNGN iD Soe NewiGr

‘Cult WEULM9 KT SOA

“YATy qa [ep PIEMPOONM W Z9)

TWX Td ‘TOA “Sy, Deven Uy

‘
ees
4
Wear Le
i 5
: ‘ '
am “ es
: : i
~ 72 y
i a ‘ 7 2 . + ie P
" ‘ ‘ - i . a2 =
- \ rs 7 ’ x = 1" e J *
a * ~ = s
at y
a + oad ‘>= a ‘
; 1 &
= . *
$ - .
4 » : e
4 se a 4
on ‘= . “ +
J = as is ~
Ps =
. : ; ms
Y a 7 : a = 7
=? ie "
“ ii : © 2 ‘
— - ’ a? “
= cE - 7
r ' re. Fs
; :* => a3
- H
> ’ : I
- 7
iy : :
. =

ee

5,

Sl Vi viSiViiLO aoOUN will ays

‘duit wewMmel Se ‘UTIL 12° [PP PAeEMpPoom W'D

ee Te ‘T TOA ‘SU. D [een Vay

"‘SISN HI S'OM SVIRV Tans
THthboONOa SMINNETG 6 INHESV AM NODOdY ‘7

‘dwt weurme py 9sayy UT 19-[9P PLPEMpPOOM WD

SUX Te ‘T TOA SNYL'D 1[ezyeN Uuy

a

A SHORT STUDY ON ZULU MUSIC. 297

A Short Study on Zulu Music.

By

Rev. Father Franz Mayr.

With Plates XLIII and XLEV.

THe Zulus have a great liking and a certain natural ability
for music, which rejoices the hearts of old and young alike
ot both sexes.

In spite, however, of the good musical ear which most
Zulus possess, and their great fondness for playing musical
instruments and for singing, it cannot be said that they have
reached any proficiency in either instrumental or vocal
music,

At the end of this article there will be given some eight
examples of Zulu songs. These have been taken down with
care, and may be regarded as fairly accurate transcriptions ;
they will clearly show the absence of art, or at least what
Europeans would call art. Nevertheless, the study of native
music should prove of interest, and it discloses a considerable
variety of strange airs and rhythms, especially in the direc-
tion of dances. It is certainly high time for such a study, as
European music is rapidly penetrating into every part of the
‘country, and harmonicas, concertinas, etc., are taking the
place of the original primitive instruments.

I. InstrumentaL Music.
The native musical instruments will now be described.
(1) Umqangala, or stringed bow (Pl. XLIII, figs. 1-3).
It is made of a bent stick or reed with a string of ox-tendon
stretched tightly across. The bows vary considerably in size,

258 REV. FATHER FRANZ MAYR.

and sometimes they may be ornamented; in the specimen
shown in fig. 3 the surface of the reed is covered with engrav-
ings. In using the instrument, one end of the bow is held by
the lips of the player, and the other end with the left hand.
The string is twanged with the thumb of the right hand, and
notes of different pitch are produced by means of the fingers of
the left hand pressing on the string. It would appear that the
mouth of the performer acts as a resonator (vide Pl. XLIV).

(2) Ugubu, or ugumbu, isa stringed bow with a calabash
attached towards one end (fig. 10). A small portion of the
calabash is cut off square at the free end. The bow is held
vertically, and the opening of the calabash is pressed against
the chest with the left hand, while with the right hand the
string is struck with a small stick. The pitch is altered by
the fingers of the left hand pressing on the string, while the
tone is varied by the varying pressure of the calabash against
the chest.

(3) Uqwabe isa stringed bow with the string tied down
at the middle towards the bow, and at this place a large
calabash is fixed (fig. 11). The opening of the calabash is
shehtly pressed against the chest of the player as in the
ugubu, but the bow is sometimes held horizontally instead
of upright. With a small stick the player strikes the string
alternately on each side of the calabash, and the pitch is
changed by pressing the string with the first finger of the
lett hand.

(4) Ugwala is in form and size hke the umqangala,
except that at one end the string of ox-tendon is attached to
the split quill of a feather. The other end of the quill is
either bound down to the end of the bow by a thin strip of
skin, with a piece of quill projecting freely beyond, or it is
passed through a hole in the wood, in which it is tightly
wedged by a peg of wood. Figs. 4 and 5 give side and front
view of this instrument ; the spht quill is seen at the bottom
of the string.

The instrument appears to be a difficult one to use, and
women are the chief performers. The mouth is placed over

A SHORT STUDY ON ZULU MUSIC. 959

the split quill and a whisthng sound is produced by the
breath; the pitch is varied by pressing the string at the
opposite end with the fingers of the left hand.

(5) Isitontolo. This instrument has been adopted from
the Basutos. It is illustrated in figs. 6 and 7. It often con-
sists of a reed (fig. 7) through which a flexible stick is passed,
and a string of ox-tendon is stretched tightly across. The
string is tied down in the middle towards the bow. The reed
may be replaced by a curved stick, into the ends of which are
fixed short flexible sticks (fig. 6). In playing the instrument
one end of the middle thicker portion is placed against the
mouth, as in the case of the umqangala, while the string is
twanged with the right hand. The mouth acts asa resonator,
while the pitch can be altered by the fingers of the left hand
pressing on the string.

(6) Umtshingo, or reed-pipe whistle. The end placed to
the mouth is cut obliquely (fig. 9), the other end is cut trans-
versely. The lower end is more or less closed with a finger,
and the pitch can be regulated. Asa rule, two whistles are
played together by two players, one taking the lead and the
other responding.

(7) Igemfe is another form of whistle. It is made of a
large reed fitting over one of smaller diameter (fig. 8). It is
played like the umtshingo.

(8) Isigubu, or drum (fig. 12). A piece of the trunk of a
tree, about eighteen inches long and one to two feet im dia-
meter, is hollowed out into a cylinder. The Umsenge, or
Cabbage Tree (Cussonia spicata), or the Umhlonhlo
(Kuphorbia grandidens) are often selected on account of
the softness of the wood. Calf or goat skin is stretched
across the two ends and tied tightly together by strips of
skin or tendon. The drum is beaten with small drum-
sticks.

These eight instruments are still in use to a small extent ;
but the music elicited from them by the untutored Zulu could
seldom please a Huropean, for in most cases Zulu instrumental
music is extremely monotonous, and with very little value in

260 REV. FATHER FRANZ MAYR.

melody or rhythm. Except in the case of the drum, the
volume of sound produced is very small, and practically the
performer himself is the only person who derives any enjoy-
ment from the music.

II. Vocat Music.

Zulu songs may be either of a public or private character.
Among the natives anyone may invent a song, text and air ;
and most of them have their own private songs, made at some
important moment of their lives, or after some event.
Children when playing invent nursery rhymes and songs ;_ so
also do boys when herding their father’s goats or cattle, and
girls when occupied in their homes or at field work, or when
sitting round the fire in the evening hours.

Special songs are composed when young people reach
puberty, and particularly when marriage arrangements begin.

A Zulu will invent a mournful song in remembrance of the
death of a near relative. A witch-doctor has his or her own
lamentations to the spirits of the dead—amadhlozi.

The arrival of a European neighbour, the opening of a
railway, a war, famine, a plague of locusts, a disease, etc., etc.,
may become subjects for semi-public songs, which may attain
a circulation, more or less wide, among the people.

Songs of a specific public character are those which are
used at the public functions of chiefs (e. g. at the feast of
the first frnits—ukwetshwama, or at royal marriages), war
songs and the tribal songs which are possessed by every chief
and tribe.

At marriages and other public ceremonies it is a Zulu
custom for not only the songs of the living chief to be ren-
dered, but also those of his father and grandfather. It is for
this reason that sones used at the time of Tshaka and Dingane
are known by the present generation.

Songs among the Zulus are composed more or less in the
following manner: Anyone who feels able and inclined to
compose a song invents one or more sentences appropriate to

A SHORT STUDY ON ZULU MUSIC. 261

some event or feeling which occupies his mind and heart. He
continually hums the sentences to himself, and changes and
improves the air until it pleases him. Soon after, on meeting
a friend, he may inform him as to his composition, who in his
turn may suggest some alteration in the air, or he may add
another sentence. In this way the song travels from one to
another, and is passed on at beer-drinks or dances, and _ ulti-
mately it may become the property of the tribe, while the
originator is in most cases forgotten.

Their method of rendering their songs is very lax. One
and the same song may be rendered in quite different ways,
both as regards the repetition of words and the sequence of
the musical sentences. Great freedom is allowed, and thus
scope is given for the individual feeling or the genius of the
singer. Even the same person will make considerable altera-
tions in singing the same song at different times; but the
general meaning of the text and the main notes of the air are:
retained.

The time is very much “tempo rubato.” When there is
only one singer the text is sung with or without action, and in
a feigned or loud voice. If there are several singers, one will
take the lead, and the others will accompany in different
parts, or the text may be divided among the singers and sung
im turn.

The Zulu chants are endless, with a constant repetition of
the same text and air.

Rhythm is marked by action, such as stamping the feet,
clapping hands, brandishing a dancing-stick, or by other
movements of the body.

In singing a war-song—igama lempi—the men stand in
a single row, or, if numerous, in many rows, one behind the
other, and the chief stands in the centre of the front row.
On both sides stand the women and children, who keep time
to the chant by clapping their hands. The strong, deep
voices of the men cause a roar like distant thunder, and the
stamping of the feet makes the earth to resound. All enter
thoroughly into the spirit of the song, and the whole is grand

262 REV. FATHER FRANZ MAYR.

on account of the great noise and the weird gesticulations of
the performers.

At marriages the grown-up girls, with the bride hidden
among them, sing the first songs on their arrival at the kraal
of the bridegroom. These introductory songs and dances,
performed by the bride’s party, are called isingeniso,
umcanguzo, and inkondhlo. The action in these dances
consists of gradual slow movements forward and backward
without clapping the hands. Then the bridegroom’s party
(iketo) follows, and the dances become more and more
excited, and after a time complete confusion reigns, and
everyone, both male and female, is trying to make the greatest
possible noise.

The regulation time for Zulu marriages is from about
1 or 2 p.m. to sunset, when the eating and drinking begins.
Late in the evening another noisy dance—umkahlelo—is
performed by the young people, accompanied by the beating
of a drum—isigubu. The personal friends of the bride and
bridegroom are not satisfied with one day’s feasting, and they
may remain for a second or even third day. The dances on
these days are more private in character, and the bride mostly
takes the lead—isimekezo. By the way of taking leave
from her parting friends, the bride distributes small presents
of bead-work among those of her own age.

Returning to our subject, Music, it must be said that the
texts of Zulu songs are mostly without much meaning,
and of no poetical value. Like the official court-praisers—
izimbongi—the Zulu poets are fond of exaggeration ; thus
they may speak of a small chief as the conqueror of heaven
and earth, who has destroyed great tribes, he is hke heaven
itself, he is king of kings.

The melodies have, as a rule, a descending tendency, each
musical sentence beginning at a high pitch and descending
towards its end. Fourths are intervals very frequently used,
also minor keys and mournful cadences, which are strange,
difficult and barbarous to ears accustomed to modern music.
The harmony of the native tunes, in correspondence with the

A SHORT STUDY ON ZULU MUSIC. 963

melody, is equally mournful. Without effort the Zulus fall
into a second or third yocal part for accompanying the tune,
and the absence of discords 1s notable.

I will now give some specimens of Zulu songs; and in
adapting them to modern musical notation I had to resist the
temptation of doctoring the native music, lest it should appear
more artistic than it really is. Dr. Alan Miller has very
kindly rendered me much assistance in this matter, and my
hearty thanks are due to him for his valuable aid.

1. Hayiza ma Pondo, Shout ye Pondos!
Helele ma Pondo, Alas, Pondos !
Vumani ma Pondo, Reply Pondos!
Ayeza ma Pondo, They come the Pondos,
Vumani ma Pondo. Reply Pondos!

LS

Hayiya ma Pondo, Helele ma Pondo Vumani ma Pondo
: = =
ay

Ayexya ma Pondo, Vumani ma Pondo

This song was sung by two native girls in a spirited manner.
It is a children’s ditty—indhlamu—and was probably com-
posed by a young Zulu man. It has been taken up by
children in their play, and refers to fights with the Pondos.

2. Anongilondolozani, Keep me safe ye, [heroes,
Uye watint’ a-o-Nqakamatshe,* He went and attacked the
Anongilondolozani, Will you protect me,
Zinyane lendhloyu, Young one of the elephant,

Zinyane lendhlangamandhla.t Young of the great heroes.

* Name of one of Cetshwayo’s regiments.
+ Praise-name for chiefs.

264: REY. FATHER FRANZ MAYR.

td hs |
-nongi-londolo- ya- - ni ztnyan’ lendhlangamandhla
Zinya n elendhlo vu

This song comes from Cetshwayo’s time, and is widely used
as the isingeniso, or first song at a marriage, when the
bride makes her first appearance with her friends at the
place for dancing—isicawu. The bride takes the lead, and
the whole song and dance are rendered slowly.

3. Yek uMajoz katandwa ndawo, Poor Majozi is not loved,

Wayiwa le, He was rejected far away,
Uyimbule, He is a lion,

Wayiwa le, He was rejected far away,
SigudhVunde lolo, We pass this long range,*
Wayiwa le. He was rejected far away.

Majozi is the surname of chief Ngoza, and the song may
be called iketo lenkosi, chief’s song. It is sung by men

* Drakensberg Mountains.

A SHORT STUDY ON ZULU MUSIC. 265

only, and sounds impressive on account of the rude, powerful
rendering and the strange action.

The composers of chiefs’ songs are invariably men, not
women; in fact, very few songs originate from women.
4. Aihlom’ Imidhleke,* Let the Imidhleke prepare for war,

He ldume, Let the thunder of war roll,
LadhVamadina, The dinner has been eaten,
Lidume. Let the thunder of war roll.

. Se
Ai-hlomImidhleke He oe Ladhl’amadina Lidume.At--

hlom’ Imidhleke He lidume, Ladhl’amadina Lidume.

This is anihubo lempi, war-song, of Hennhemu, father

oO?

of Mvel, the present chief of the Mafunzi tribe, in the
Zwaartkop location, near Pietermaritzburg

5. Inga buqili See the crafty hiding themselves,
Siyausilanda, We will fetch their cattle,
Nang’u Zulu. Here we are (we Zulus).

War song used in some parts of Zululand, which was com-
posed after a fight with the Swazis at the beginning of
Pande’s reign.

6. Hlasela le, Prepare for war,
Wayiwa le, Far, far away,
Ikowinkosi yezwe, There is the chief of the land,
Sibuz’inkosi ezaukuta. We ask which chief is going to
die.

* The name of a regiment.

266 REV. FATHER FRANZ MAYR.

ye swe Sibuyinkosi eyau-hufaHla-se - la-le.

Chief Mvel’s song—ihubo lenkosi. Note the transition
from one key to another without the least warning. Absence:
of modulation is very marked, and this is the main reason
why native songs sound so barbarous, strange, and harsh to
our ears.

7. Badabula kwa Hlongwa, They went through the Hlongwa

Yi-ya yi-yo ye maye, Alas, alas ! [ district,
Badabula ezizweni, They went through many tribes,
Yi-ya yi-yo, ye maye. Alas, alas!

Ga eS = SSeS

| Bada bula kwa Hlongwa, Yeyayiyo ye maye, ye maye,

ae

Badabula extisweni,

Yiyayiyo, ye maye, ye maye

Hlongwa was the name of an Ama Lalatribein Alexandra
county at Dingane’s time. The name of the chief was Joli.
The above song is used at marriages in Zululand.

8. Ye he ubaba wangikolisa, Well my father gave me satis-

faction,
Kodwa au yek’amadhlozi But, alas, the spirits (of the
Angibulala. Kall me. [ dead)
O, kodwa ngiyazisa, And yet I praise them,
Amadhlozi angilaya, The spirits inform me,

Yek ubaba epansi, Oh! my father is dead,

A SHORT STUDY ON ZULU MUSIC. 267

Wo zintandane zakwetu, Poor orphan children,
Ning’azise ukuti nakolwa. ‘Tell me do you believe (that
the spirits inform me).

35 eS

Ye he ubaba wangtikolisa,wangikolisa, Kodwa au yek’ama-

8 Sl, ee Wee Ged
eres 1 8 8 ee +S ee \ |

eo BS Pesta e

yakwe ti, ningayi se ukutt nea

This is an isililo, or lamentation of a witch-doctor. It is
sung without rhythm or action.

The literal translation gives some idea of the general signi-
ficance of the text; but a great deal of explanation would be
required to make the meaning clearer, and this would carry
us beyond the limits of the present article. My intention was
to give a few samples of Zulu songs of different kinds, in
order to illustrate the general character of native music. It
may be added that with the kind assistance of Dr. Alan
Miller the songs were carefully taken down, partly from the
lips of the singers, and partly from phonograph records.

A study of the specimens given will show that the Zulu is
not able to attain to much art without outside assistance,
although he has great natural ability for music, and can very
readily be trained in this direction.

OL. 1, PART 3. 20

ut - ais Te a

yao! SRE” Cees
‘ & : San’ / Orie: oe . _
¢ 7 = ‘ a:

OS 2) tee Oe ; : r
* a i
6 a
’ . hs 7
. s -
* : na ‘sa -

Pe : i”
2 =. _ ay wey > in
7 a

f a

: ¢ . nn et
; aye ae a | A. Pee ii §
7 _ t 7 ie

wt J
_ a Da y Hi
cs a) “s e iN ea
a : =~ iy rae ta :
, - = ng
a2 ‘ ] 2 7 _ - xe ‘
. = a.
oa - », 7 . ‘
a 7 = os is
i 7 : Cea 7 7
= i 7 a 1 as Ah ae ear
y “y 2 ; Pay a. on
ko - i f F of 4
- Pa » Shi A, : _
a - fi = “ - seeath
: : i ’ «
: ~
* re 7
— tT

7 ;
f *)
; wt 4 ¥
y eee Lo
A ory 5 in : — 2 Sen
, ou a = Ss | | A oe = 7 :
i > an . : = :
ese a er
—— 3 : “ j r nr* i
e 7 i =
i ‘ ita!
| ” &
- ‘ a ;
= - .
i. - . | ;
: A ' i _~ } 4
. . |
S y -
' . ’
fe , . r ' :
f
’
| ’ i
7

- ; .
7
7 »
a
+
- f
i . i
: Uy » bre A i
vis - wer ive : ;
37 e
| a
~=
i
'
,
; 4
1
'¢@
1
.

Ann. Navat G. Mus., Vou. I, Parr III. Pr x ie

—~ «tin By “ae ses) ——
Dept Adlard &

ZuLtu MusicaL INSTRUMENTS

ry,

Photo. by Natal Railway

1—3. Umaqangala. 4. Side view of Ugwala, with split quill at the botton 5. Front view of Ugwala.
6—7. Isitontolo. 8. Igemfe. 9. Umtshingo. 10. Ugubu or Ucumbu. 11. Uqwabe. 12. Isigubu.

(A yard-measure was placed below to mdicate size

Ann. Natat G. Mos., Vot. I, Parr III. Pr She

Adlard & Son, [nipr.

NATIVE MUSICIANS.

Names of instruments: back row, beginning on the left, Uqwabe, Umqangala, Igemte ;
front row, Uugub, Isitontolo, Ugwala.

{(

A COLLECTION OF HYDROIDS. 269

On a Collection of Hydroids, mostly from the
Natal Coast.

By

Ernest Warren, D.Sc.Lond.

With Plates XLV—XLVIII.

ily

Durie the last few years a collection of thirty-one species
of hydroids has been made while shore-collecting on the Natal
and Zululand coasts. In addition, some four species have
been dredged from a depth of 40 fathoms in the neighbour-
hood of Bird Island, near Algoa Bay.

In the following list the species are enumerated under their
various families, and the localities are also given.

Except when otherwise stated, the specimens were obtained
from the rock-pools at low-tide. They were generally fixed
with a half-saturated solution of corrosive sublimate in 30 per
cent. spirit, with 1} per cent. acetic acid. The best results
were obtained when the solution was applied fairly hot.

For staining the hydroids whole, Mayer’s paracarmine was
found satisfactory, and for sections, Delafield’s hematoxylin,
or iron-hematoxylin followed by orange, were employed.

Fam. Bimeriide.

(1) Parawrightia robusta Warren.
Loc.: Isipingo, Scottburgh, Park Rynie.

ERNEST WARREN.

Fam. Eudendriide.

Eudendrium parvum sp. n.
Loc.: Park Rynie.
Eudendrium angustum sp. n.
Loc.: Ata depth of 40 fathoms, sixteen miles N.E.
Bird Island, near Algoa Bay.

Fam. Clavatellide:

Clavatella multitentaculata sp. n.
Loc.: Isipingo.

Fam. Tubulariide.

Tubularia solitaria Warren.
Loc.: Tongaat, Isipingo, Park Rynie.
Tubularia betheris sp. n.
Loc.: Coast Quarry, between Alexandra Junction
and Park Rynie.

Fam. Pennariide.
Pennaria australis Bale, var. cooperi Warren.
Loc.: Kosi Bay, Zululand; Isipingo, Scottburgh,
Park Rynie.
Fam. Cladocorynide.

Cladocoryne floccosa Rotch.
Loc.: Isipingo, Scottburgh.

Fam. Syncorynide.

Asyncoryne ryniensis g.e. sp. n.

Loc.: Park Rynie.

Fam. Corynide.

(10) Coryne pusilla Gartner.

Loc.: Isipingo, Park Rynie.

(11)
(12)

(13)

bo
sJ
—

A COLLECTION OF HYDROIDS.

Fam. Sertulariide.
Sertularella polyzonias (Lin.).
Loc.: Isipingo, Scottburgh, Park Rynie.
Sertularella fusiformis Hincks.
Loc.: Park Rynie.
Sertularella tumida sp. n.
Loc.: At a depth of 40 fathoms, sixteen miles N.E.
Bird Island, near Algoa Bay.

(14) Sertularella campanulata sp. n.

(15)
(16)

(17)

(22)

(25)

(26)

Loc.: Scottburgh, Park Rynie.
Sertularia acanthostoma Bale.
Loc.: Park Rynie.

Sertularia operculata Lin. LS%
f~
Loc.: Isipingo, Alexandra Junction. f8a./5
Sertularia loculosa Busk. Ls

Loc.: Isipgo, Scottburgh, Park Rymie. — —)
\~

Sertularia linealis sp. n. \Fe
= . Ne ? —
Loc.: Kosi Bay, Zululand. He
Sertularia bidens Bale.

Loc.: Park Rynie.

Pasythea quadridentata (Hillis & Sol.).
Loc.: Scottburgh, Park Rynie.

Thuiaria tubuliformis (M.— Turneretscher).
Loc.: Isipingo, Scottburgh, Park Rynie.

Fam. Plumulariide.

Plumularia tenuis sp.n.
Loc.: Kosi Bay, Zululand; Durban, Park Rynie.
Antennella natalensis sp. 7.
Loc.: Isipingo:
Plumularia spinulosa Bale.
Loc.: Park Rynie.
Kirchenpaueria mirabilis (Allman).
Loc.: Scottburgh.
Paragattya intermedia g.e. sp.n.

Loc.: Park Rynie.

27 2, ERNEST WARREN.

(27) Halicornaria segmentata sp. 1.
Loc.: Coast Quarry, between Park Rynie and Alex-
andra Junction.
(28) Aglaophenia chalarocarpa Allman.
Loc.: Kosi Bay, Zululand; Durban, Isipingo, Scott-
burgh.
(29) Aglaophenia parasitica sp.n.
Loc.: Scottburgh.

Fam. Campanulariide.
(30) Campanularia tincta Hincks.
Loc.: Park Rynie.
(31) Campanularia caliculata Huncks.
Loc.: Isipingo.
(32) Clytia elongata sp.n.
Loc.: Ata depth of 40 fathoms, sixteen miles N.E.
Bird Island, Algoa Bay.
(33) Lafcea scandens Bale.
Loc.: Isipingo.
(34) Lafcea magna sp.n.
Loc.: Scottburgh, Park Rynie, Natal; Bird Island,
Algoa Bay.
(35) Thyroscyphus equalis sp. n.
Loc.: At a depth of 40 fathoms, sixteen miles N.E.
Bird Island, Algoa Bay.

The species will now be described in order.

(1) Parawrightia robusta Warren.
PARAWRIGHTIA ROBUSTA Warren, ‘ Ann. Nat. Gov. Mus.,’ vol, i, 1907,
p. 187.
This species has a certain affinity with Wrightia
arenosa! (Alder), of the British Coasts; both hydroids
possess fixed gonophores.

(2) Eudendrium parvum sp.n. (Pl. XLV, figs. 1-4.)

At first this hydroid was considered to be a diminutive
form of E.capillare Alder; but without a more detailed

' Hincks, T.,* A History of the British Hydroid Zoophytes,’ 1868, p. 88.

A COLLECTION OF HYDROIDS. 7s

account of the structure and variability of Alder’s hydroid, it
is not possible to refer it to this species with any certainty.
The hydroid grows on the surface of sea-weed, and may
extend over a very considerable area. In all the specimens
obtained the hydrocaulus has exhibited very little branching,
and the total height has not exceeded about } in.
The polyps are of a very pale horn colour.

TEXT-FIG. 1.

ee MALE GONOFHORES.

Eudendrium parvum, sp. 7.

TroeHosomE.—The hy drorhiza or creeping stolon branches
in an irregular manner, and may form a more or less com-
plicated network. The diameter is about 0°15 mm., and the
thickness of the perisare 4°4 pu.

Hydrocaulus.—The stems bearing the hydranths are
given off irregularly from the hydrorhiza, and very frequently
they are unbranched. When branching occurs, it is irregular,
asin text-fig. 1, 4. Groups of irregular annulations are present
at the base of the stems, and at the points of origin of the

274 ERNEST WARREN.

branches, and also at intermediate places. The diameter of
the hydrocaulus is about 0°10 mm., and the thickness of the
perisare 4°4

Hydranth.—tThe polyp has the typical eudendrium-shape,
and there are 15-18 tentacles in a single verticil.

In preserved specimens the trumpet-shaped hypostome
measures about 0°09 mm. in height and 0°25 mm. in diameter.
The body of the hydranth, measured from the verticil of
tentacles to the base, has a height of about 0°39 mm., and its
greatest breadth is about 0°32 mm.

Gonosomy.—The male only has been found. The gono-
phores are carried in dense clusters on very short annulated
stems, which spring direct from the hydrorhiza (text-fig. 1, B).
Kach cluster may be regarded as arising from an abortive polyp.

The gonophores are three-chambered (Pl. XLV, fig. 4); the
proximal chamber is generally very small. There is some-
times a tendency for the development of a slight terminal
tubercle, but no nematocysts have been seen in such.

The length of a gonophore is about 0°57 mm., and the
diameter of the distal chamber 0°19 mm.

General histology.—There is a shallow groove running
round the hydranth at about one third of the height of the
polyp above the base (Pl. XLV, fig. 1, ¢.pr.). A thin continua-
tion of the perisare of the hydrocaulus is present over the
base of the hydranth, and it terminates in this shallow groove
(figs. 2 and 3, t. pr.). The ectoderm below the groove differs
somewhat in character from the ectoderm above, and it ends
in a line of especially large cells (c.c.), which have large
nuclei and stain deeply. The perisare of the hydrocaulus
consists of an inner and outer layer (fig. 2, pg and p;), and it
is the latter (p,) which is continued upwards over the base of
the hydranth. Allman states that in Eudendrium vagi-
natum Allman “the body as far as the origin of the
tentacles” is “ enveloped in a loose corrugated membranous
sheath, which loses itself below on the hydrocaulus.”! The

' Allman, G. J.,‘A Monograph of the Gymnoblastic or Tubularian
Hydroids,’ Part IT, p. 339.

A COLLECTION OF HYDROIDS. al

structure described in E. parvum is, doubtless, strictly homo-
logous with the sheath of vaginatum, only in the former
species it does not envelop the body to such an extent.

The nematocysts of the tentacles are somewhat small, and
have a length of 4°8 uw, and breadth 271 uw.

The endoderm at the base of the hypostome consists of
elongated cells with large granules (Pl. XLV, fig. 2, gl.c.).
The endoderm of the vertical sides of the polyp is very thin,
and it contains some flattened granular cells. Towards the
base of the polyp the endoderm cells are conspicuously
vacuolated (v.c.).

Systematic Posrrron.—From the published descriptions of
KH. capillare Alder, it is not possible to identify the Natal
hydroid with this species. The characters in which it differs
are: the three-chambered condition of the male gonophore,
the absence of a well-defined terminal tubercle to the gono-
phore, and the extension of the perisare over the base of the
polyp. In the last character, an approach 1s made to the con-
dition seen in E. vaginatum. In Hudendrium insigne
Hincks a shallow circular groove near the base of the hydranth
is mentioned and figured by Allman, and doubtless it has the
same structure as that described in E. parvum.

(3) Eudendriumangustum sp.n. (Pl. XLV, figs. 5, 6.)

This hydroid was dredged from a depth of 40 fathoms in
the neighbourhood of Bird Island, off Algoa Bay. The speci-
men was torn from its attachment, and the hydrorhiza is
unknown. It was about 3 inches in height, and of a dark
brown colour.

TropHosome.—H ydrocaulus.—Monosiphonic, aborescent,
irregularly branching; the stems which immediately carry
the branches bearing the hydranths have occasional groups of
annulations, and the branches are annulated at their origins,
and have irregularly scattered groups of 2-10 rings. The
perisare is dark brown over the older portions of the colony,
and paler over the younger parts.

276 ERNEST WARREN.

The main stems are nearly black ; they may attain a dia-
meter of 1 mm. or more.

The secondary stems, which carry the hydranth-bearing
branches, measure about 0°38 mm. in diameter, thickness of
perisare 264 p.

The diameter of the branches carrying the hydranths is
about 0°19 mm., thickness of perisare 6°1 wu.

Hydranth.—lIt has a superficial resemblance to a typical

TEXT-FIG. 2

i

A(naé. Ste)

EKudendrium angustuwm sp. un.

eudendrium polyp (Pl. XLV, fig. 5); there are 25-30 tentacles
i a closely-set verticil.

In the preserved specimen the height of the hydranth from
the base to the level of the verticil is about 0°31 mm., the
breadth at the widest part 0°29 mm., and the height of the
hypostome 0°16 mm.

Gonosome.—Unknown.

General histology.—The general cell-structure of the
hydranth is interesting and peculiar; it differs widely from
that seen in E. capillare (cf. figs. 2 and 6, Pl. XLV).

A COLLECTION OF HYDROIDS. 2070

The perisarc, terminating in a groove at the base of the
polyp, is generally shehtly inverted into a shallow cup (figs.
5 and 6, p.cp.), which is doubtless comparable to the sheath
seen in E. vaginatum.

The ectoderm is provided with two kinds of nematocysts ;
the large measure about 23°34 im length, and 10°44 in
breadth; the small variety 5 and 2°2m respectively. The
small nematocysts occur chiefly on the tentacles, while the
large are found on the body of the hydranth, and more
especially around the edge of the reflexed ectoderm of the
hypostome (fig. 6, 7.).

The character of the endoderm is remarkable; there is
practically no digestive cavity, and the mouth is blocked up
by a plug of elongated endoderm cells, continuous with the
digestive endoderm at the base of the polyp.

The upper surface of the hypostome consists of endoderm
of two kinds—the central plug of digestive cells above
mentioned (fig. 6, e.p.), and the outer reflexed epithelium (7.e.)
continuous with the skeletal endoderm which runs up the
tentacles. Between the elongated endoderm cells are wedged
glandular cells (v.c.), which stain intensely with hematoxylin ;
the protoplasm is densely crowded with small, rounded
vacuoles.

The plug of endoderm can directly ingest food; 7.f., in
fig. 6, represents a cluster of copepod eggs. The exposed
surface of endoderm is probably sticky during life, since small
copepods and other organisms may frequently be found
adhering to it.

This throwing out of digestive endoderm through the
mouth, for the ingestion of food, is, as far as I am aware,
umque among the hydroids.

Systematic Posrrion.—The general character and branching
of the hydrocaulus is distinctly that of an Eudendrium.
The hypostome is also trumpet-shaped, although it has special
pecuharities. It is remarkable that these peculiarities are
present in the developing hydranth of Parawrightia
robusta Warren (‘ Ann. Nat. Gov. Mus., vol. i, p. 192).

278 ERNEST WARREN.

In the absence of the gonosome, however, the present
hydroid can only provisionally be referred to the genus:
EKudendrium. ‘The species is named angustum with
reference to the narrow or almost obliterated digestive cavity.

(4) Clavatella multitentaculata sp.n. (Pl. XLV, figs.
7-9.)

A small clump of this species was found embedded in a
white siliceous sponge at the edge of a small rock-pool full of
corallines, and situated at some distance above the low-tide
line (figs. 7 and 8).

This delicate little hydroid, reaching a height of about
+ ich, has a circle of numerous capitate tentacles, and the
hypostome is richly coloured. Around the mouth there is an
area of chalky white, below this there is a band of lemon
yellow, and just above the circle of tentacles an irregular
band of bright red.

TropHosome.—The hydrorhiza consists of a thin tube,
which ramifies through the substance of the sponge. Diameter
about 0°15 mm.; the perisarc is yellow and fairly stout, bemg
about 17°8 nu.

Hydrocaulus.—Upright stems spring from the hydrorhiza
and grow through the sponge to the upper surface. In com-
parison with the size of the hydranth, the hydrocaulus is
slender, and it appears to be highly extensile.

Foreign bodies, suchas minute particles of sand and diatoms,
readily cling to the perisarc, and they may form a quite dis-
tinct coat.

Diameter of embedded portion about 0°15 mm., thickness of
perisare 15°12; on emerging through the sponge the hydro-
caulus has a greater diameter, being about 0°31 mm., while
the perisarc becomes very thin and transparent, bemg about
4-4 uw thick.

Hydranth.—The polyp is relatively large (fig. 8), and
looked at from above it is star-shaped. The hypostome is
swollen and conical, with a small apical mouth (mo.), and is

A COLLECTION OF HYDROIDS. 279

brightly coloured as above described. The tentacles are all
capitate, and are arranged in one series at the base of the
hypostome ; they arise close together from the body of the
hydranth, and they are too numerous to form a single verticil.
There are some 28-35.

The following are the measurements of a preserved typical
specimen. ‘Total height of polyp 0°84mm.; diameter, imme-
diately above the verticil, 0°58mm.; diameter of the capi-
tulum of a tentacle 0°15 mm.

GonosomME.—The reproductive bodies (fig. 8, p/.) are pro-
duced in clusters on short outgrowths from the body of the
hydranth just above the circle of tentacles. These bodies
are shown in section in fig. 9.

In none of the specimens collected were the reproductive
bodies mature, and consequently it is not possible to state
whether free meduse are formed.

General histology.—The perisarc of the hydrocaulus
dips into a groove at the base of the hydranth, and then
extends for a short distance upwards over a modified band of
ectoderm (fig. 9, p.). The capitulum of a tentacle possesses
both large and small nematocysts; the former measure 19 1
in length and 11min breadth, and the latter 9u and 44
respectively (.).

The endoderm of the hypostome (e.iy.) is thick and com-
pact, with numerous small nuclei, which have a_ special
affinity for stains. Below the hypostome there are pendulous
lobes of large cells (gl.c.) with large pigmented granules,
and there are also a small number of oval cells (ov.c.) which
stain readily and appear to consist of somewhat dense and
homogeneous protoplasm. The usual vacuolated cells (v.c.)
occur at the base of the hydranth, and skeletal septate endo-
derm runs up the tentacles.

Systematic Posrrion.—The present species is clearly closely
alhed to Clavatella prolifera Hincks! of the British
coasts; it agrees with it in the tentacles being all capitate

' Hincks, T., ‘A History of the British Hydroid Zoophytes,’ 1868,
p. 73.

280 ERNEST WARREN.

and arranged at the base of the hypostome in a single series.
It also agrees m the exceptionally delicate nature of the
hydrocaulus. It differs from prolifera in possessing 28-35
tentacles instead of 8, and also in the place of origin of the
reproductive bodies. In prolifera free meduse are budded
from short processes which spring from the base of the
hydrocaulus, while in the present species such processes are
budded by the hydranth itself. It is, of course, very probable
that free medusze are also formed by multitentaculata.

(5) Tubularia solitaria Warren.

TUBULARIA SOLITARIA Warren, ‘Ann. Nat. Gov. Mus.,’ vol. i, 1906,

29

p. 83

This species, which occurs embedded in siliceous sponges,
has been fully described. It is a solitary form, and in the
arrangement of the endodermal canals and in other ways it
is allied to Corymorpha; but the occurrence of an actinula
in development has caused it to be referred to Tubularia.

(6) Tubularia betheris sp. n. (Pl. XLV, figs. 10 and 11;
Pl Vis ther 23)

This graceful hydroid was found in a sheltered rock-pool
near the Coast Quarry, between Park Rynie and Alexandra
Junction. It was attached to the rock by a small hydrorhiza.
The colony consisted of 4 individuals, and its height was
about 14 inches (fig. 10).

The hydranths are translucent, and the hydrorhiza and
hydrocaulus are a very pale brown.

TropHosoME.—The hydrorhiza is a_ branching stolon
creeping over the surface of rocks or worm-tubes.

Diameter 0°541mm., thickness of perisare 20 «1.

Hydrocaulus.—It consists of upright stems springing
irregularly from the hydrorhiza and bearing terminal polyps.
The perisarc is irregularly ringed, especially towards the base.

Diameter 0°30 mm., thickness of perisare 14 yu.

Hydranth.—The general shape is rather elongated and

A COLLECTION OF HYDROIDS. 281

slender (fig. 11). At the base of the polyp there is a charac-
teristic dilatation which is covered by a continuation of the
perisare of the hydrocaulus (fig. 12)... The proximal or basal
tentacles are of considerable length and are about 17 in
number ; the distal or oral tentacles are short and are about
14 in number.

The total height of the polyp, measured to the base of the
dilatation, is about 2°51 mm., the greatest breadth, measured at
the level of the basal tentacles, is about 1°2lmm. The height
of the dilatation is about 0°55 mm., and its width 0°61 mm.

Gonosome.—The gonophores are budded in irregular clusters
on semi-erect peduncles (pd.) or blastostyles, which spring from
the hydranth just above the basal verticil of tentacles. The
gonophores reach a high state of development ; marginal ten-
tacles, radial and circular canals, and a velum (v.) are present.

The gonophore figured (fig. 12) is a young male, but owing
to the lack of material older stages have not been seen. A
female gonophore has not been found, and consequently the
nature of the actinula is undetermined.

General histology.—A vertical section of a hydranth is
represented in fig. 12, Pl. XLVI. At the upper edge of the
basal dilatation the ectoderm is conspicuously thickened, and
on its under side there is a shallow groove (p.g.). The peri-
sarc of the hydrocaulus is contimued up into the groove, and
then fades away on the rounded surface of the thickened
ectoderm. The perisarc of the hydrocaulus consists of an
inner and outer layer, the later being continued over the base
of the hydranth.

The nematocysts in the ectoderm are mostly small and
inconspicuous ; they measure about 5'l a im length and 4”
in breadth (s.2.). In the endoderm there is a larger variety
(J...) measuring 8°3 «and 7°9 jw respectively.

The endoderm around the mouth contains, as usual, nume-
rous deeply staining nuclei; the oral or distal tentacles are
provided with septate skeletal endoderm. The vertical sides
of the hydranth have a very thin endoderm with a few

d

vacuolated cells (v.c.), and also large nematocysts. The base

282 ERNEST WARREN.

of the hydranth is internally constricted by a large mass of
cellular skeletal endoderm (b.M.), which constitutes the basal
support of the proximal tentacles (p.t.). These are provided
with cellular skeletal endoderm in place of the more usual
septate structure, and it is contimuous with the basal mass.
Over the skeletal endoderm there is an epithelium (e.ep.) of
ordinary endoderm with numerous vacuolated cells (v.c.).

The cavity of the basal dilatation (b.d.) is lined by elongated
endoderm cells with large granules. This chamber is partially
separated off from the rest of the digestive cavity of the
hydranth by the projecting skeletal endoderm (b.M/.), and the
communication which is left tends to be more or less bridged
across by a thin sheet of cells (s.e.) continuous with the endo-
dermal epithelium.

Systematic Posirion.—The present species is characterised
by the presence of the large dilatation at the base of the
hydranth. In general aspect and in the erect position of the
peduncles it resembles Tubularia attenuata Allman! and
Tubularia humilis Allman! from the British and Irish
coasts.

In a number of species of Tubularia the hydranth is
supported on an annular expansion of the hydrocaulus. The
dilatation at the base of the hydranth of betheris would
naturally be directly compared with this annular expansion ;
but in betheris, at any rate, there is a distinct differentiation
of the endoderm lining it (fig. 12), and doubtless it accom-
plishes some special physiological function. Without a
detailed knowledge of the mode of development of the
hydranth it is not possible to decide whether it should be
regarded as a basal differentiated portion of the hydranth, or
as an expansion of the upper end of the hydrocaulus.

(7) Pennaria australis Bale, var. cooperi Warren.

PENNARIA AUSTRALIS Bale, ‘Cat. of the Australian Hydroid Zoophytes,’
1884, p. 45.

In the ‘Ann. Nat. Gov. Mus.,’ vol. i, p. 73, the author
1 Allman, G. T., op. cit., pp. 410-411.

A COLLECTION OF HYDROIDS. 288

has described under the name Halocordyle cooperi a
hydroid which occurs along the Zululand and Natal coasts.
It differs from the description of Pennaria australis Bale
in the number, and arrangement in whorls, of the capitate
tentacles. Definite variations in these characters were found
to occur in different colonies of the Natal hydroid, and it is
now seen that a series of gradations, leading to a symmetrical
arrangement in whorls, occur in the different species of
Renmiar a.

The main character which distinguishes the genus Halo-
cordyle from that of Pennaria is the verticillate arrange-
ment in the former of the capitate tentacles ; but an examina-
tion of the following table will show that it is scarcely
necessary to retain the genus Halocordyle.

Filiform Capitate
tentacles. tentacles.
Pennaria gibbosa Agassiz, America. 12 52 scattered.
symmetrica Clarke,* America 14-18 50 f
cavolinii Hhrenberg,! Europe 12 20 A
australis Bale,’ Australia . 7-12 9-14; 4, oral

whorl ; 5—LO

scattered below

var. cooperi, Natal. 8-10, 8-13; typically
typically 8 4 oral and 4

basal tentacles

alternating.
adamsia von Lendenfeld} 24, 4 oral and 4
Australia basal alternating.
Halocordyletiarella Ayres, N. America 12? 6 oral, 6 basal.
australis Bale,’ Australia 8-10 4-5, oral whorl,

rest scattered.

’ Allman, G. J., ‘A Monograph of the Gymnoblastie Hydroids,’ Part
II, pp. 363-370.

? Bale, W. M., “On some New and Rare Hydroida in the Australian
Museum Collection,” ‘Proc. Linn. Soc. N. 8. Wales,’ vol. iii, 1888, p. 747.

* Bale, W. M., ‘Catalogue of the Australian Hydroid Zoophytes,’
1884, p. 45.

* Lendenfeld, R. von, “The Australian Hydromeduse,” Part V,
‘Proc. Linn. Soc. N. S. Wales,’ vol. v, 1884-5, p. 595.

* Bale, W. M., “ Further Notes on Australian Hydroids,” ‘ Proc.
Roy. Soc. Victoria,’ 1893, p. 94.

2

VOL. 1, PART 3. 21

28A ERNEST WARREN.

The variety cooperi has typically two alternating whorls
of four capitate tentacles; but it exhibits marked variability
in this character, and most of the arrangements above men-
tioned, as typical for the different species, occasionally occur
init. The matter has been discussed in the ‘ Ann. Nat. Gov.
Muss vol: a5p) 209:

From these considerations it would appear to be more
satisfactory to regard the Natal hydroid as a variety, or per-
haps as an incipient species, originating from P. australis.

(8) Cladocoryne floccosa Rotch,
CLADOCORYNE FLOcCOSA Rofch. ‘Ann. Nat. Hist.,’ March, 1871.

This unique hydroid with capitate pinnate tentacles was
found originally at Herm, near Guernsey, and has been
described and figured in Allman’s monograph. I am not
aware that the hydroid has been met with elsewhere ; but it
was found fairly abundantly at Isipingo and Scottburgh in
the month of May, 1907. It occurred in rock-pools, and
especially on old polycheet worm-tubes which had become
covered with sea-weeds and polyzoa.

The colony consists of a running stolon which gives off
short upright stems, from one quarter to half an inch high,
which generally carry a single terminal polyp. Occasionally
the stem may give off a short lateral branch.

The following measurements of an average preserved speci-
men are given, in order that a comparison may be made with
the European hydroid.

Hydrorhiza; diameter 0°182 mm, thickness of perisare
15°5 p.

Hydrocaulus; diameter 0°152 mm., thickness of perisarc
8:9 mu.

Hydranth; total height about 0°9 mm., breadth about
O° mm. Diameter of capitulum of tentacle about 0°058 mm.

Nematocysts are at least of two kinds: the large measure
11-1 » in length and 9:0 » in breadth, and the small, 6°7 nu
and 5°53 respectively.

A COLLECTION OF HYDROIDS. 285

The polyp has a chalky-white area around the mouth, and
the body below is red. The perisarc is pale yellow, and in
the hydrorhiza it is smooth, while in the hydrocaulus there
may be a few irregular annulations.

Around the mouth there is a whorl of four to six simple short
capitate tentacles. The number of pinnate tentacles is very
variable, 10-18, and they appear to be irregularly scattered
over the body, although a verticillate tendency is assigned to
them by Allman.’ The number of capitate ramuli on the
tentacles varies from 6-14.

The whole length of the tentacle is occupied by a very
regular septate endoderm.

All the material collected has been carefully searched for
the gonosome, but it is regretted that none has been found,
In the description given by Allman the gonosome is stated to
be unknown.

(9) Asyncoryne ryniensis g.e. sp. n. (Pl. XLVI, figs. 18
—17.)

Only two small colonies of this interesting hydroid have
been found. They were obtained from Park Rynie, as the
specific name implies, and were growing on the surface of
polychet worm-tubes. Although the polyps are large, being
several millimetres in length, yet they were overlooked in
the living condition, and only after fixation with corrosive
sublimate were they detected. It is hence probable that
they are exceedingly transparent when alive (Pl. XLVI, fig.13),

TropHosome.—The hydrorhiza is a creeping stolon which
appears to branch sparingly.

The diameter is about 0°24 mm. The perisarc is frequently
remarkably inflated, being made up of thin layers more or
less widely separated from one another. When not inflated
it has a thickness of about 8°9 wu.

Hydrocaulus.—The hydrorhiza produces at irregular
intervals very short upright stems bearing a single elongated

' Allman, G. J., ‘A Monograph of the Gymnoblastic Hydroids,’ 1872,
Part II, p. 380.

286 ERNEST WARREN.

hydranth. These stems gradually expand into the hydranth
and are only separable from it by reason of the termination
of the perisare (fig. 15, H).

At times a branch of the hydrorhiza may turn upwards, and
its growing apex may develop into a hydranth; and thus the
growth of the branch is terminated. Such has occurred in
the specimen shown in text-fig. 3, d., hy.

Hydranth.—The general shape is that of an elongated
spindle (Pl. XLVI, fig. 14). The hypostome region is not
sharply distinguished on the outside. Around the mouth
there are 4-6 short capitate tentacles; perhaps four is the
typical number. Over the surface of the hydranth there are
scattered some 25 moniliform tentacles. The longest of the
tentacles are those situated at about one third of the length
of the polyp from the mouth; those at the base are very short.

The hydranth in the preserved condition may have a length
of 83mm. or more. The diameter of the capitulum of an oral
tentacle is 0°133 mm., and the diameter of the terminal swell-
ing of a moniliform tentacle may be as much as 0°075 mm.

GonosomeE.—Between the moniliform tentacles, and some-
what below the middle of the body, short outgrowths are
produced, and from these are budded clusters of reproductive
bodies. Itis probable that they are planoblasts, and ultimately
become free (Pl. XLVI, fig. 14, pl.) ; since nm the most mature
of the specimens marginal umbrella-tentacles are fully formed,
and the body is practically a medusa, while the sexual elements
are exceedingly immature.

General histology.—The delicate perisarc of the hydro-
caulus dips into a shallow groove of columnar cells at the base
of the hydranth, and then fades away (Pl. XLVI, fig. 15, p.g.).
The perisare of the hydrorhiza, separated into distinct layers,
is well seen in the figure (p.), also in text-fig. 3, B., p.

The ectoderm is provided with two kinds of nematocysts.
The large occur more especially in the capitula of the oral
tentacles (fig. 17, /.1.), but they are found also in the general
ectoderm of the hydranth and ccenosarc ; they measure 27-04

A COLLECTION OF HYDROIDS. 287

in length and 19°54 in breadth. The small nematocysts,
measuring 10°Ou in length and 8‘Om in breadth, are abundant
on the moniliform tentacles (fig. 16, s.n.).

The moniliform aspect of the tentacles is due to the presence
of thickened rings or patches of ectoderm. The axis of
septate endoderm takes no part in the production of the
swellings, and, in fact, it seems sometimes to be somewhat
constricted by them. ‘The terminal swelling is more or less
spherical, but if is much smaller than the capitulum of an oral

TEXT-FIG. 3.

\

=
y ————
KkEEE tre, ia?

rs

Endodermal canals in As yncoryne ryniensis, sp.n. Ax: 70;

Bx 150] 6X 40D s< 80:

tentacle, and it appears never to possess large nematocysts.
The small nematocysts tend to be crowded on the summits of
the patches and thickened rings.

The endoderm of the hydranth consists of elongated cells
with large vacuoles, and between them are wedged glandular
cells (gl.c.), which stain deeply, and contain granules or
numerous small vacuoles.

Cross-sections of the ccenosare of the hydrorhiza may, in
certain parts, exhibit two, three, or four tubes of endoderm,
surrounded by a common ectoderm. This condition is shown
in text-fig. 3, A, as it occurred in a branch of the hydrorhiza

288 ERNEST WARREN.

which happened to terminate in a hydrocaulus (hy.) and
hydranth (b.h.). The layer shown solid (end.) is the endoderm,
e.c. are the endodermal canals, c.e. the common ectoderm in
which the canals are embedded, b.hy. is a branch coming off
from the hydrorhiza, and it shows a fork in the endodermal
canal at e.c.

Fig. B is a cross-section through the hydrorhiza, and shows
three endodermal canals (e.c.), surrounded by a common
ectoderm which contains three large nematocysts (/.n.). The
perisare (p.), consisting of loose laminee, is well seen.

The endodermal canals do not seem to necessarily arise as
direct tubular outpushings from a main canal, but sometimes,
at any rate, they appear to originate as solid strings of endo-
derm cells, which spht off from the main canal, and in which
a cavity ultimately becomes hollowed out. Occasionally the
perisare dips down into the ectoderm between the forks, so
that for a distance of ; mm. or so there may be two perisare
tubes (fig. C). A cross-section through ab. is shown at fig. D.

Systematic Posrrion.—This hydroid appears to be unique in
the nature of its tentacles. In Pennaria, Cladonema, and
Stauridium the hydranth is provided with both capitate
and filiform tentacles, but the latter are arranged in a definite
basal verticil.

In Asyncoryne the filiform tentacles are scattered, and
they also possess an unusual moniliform structure, and
terminate in a kind of rudimentary capitulum. The structure
of the tentacles recalls that of the marginal tentacles of the
medusa of Cladonema and Syncoryne.

In the nature of the tentacles, Asyncoryne shows affinity
to Cladocoryne. In Cladocoryne there is a verticil of
simple capitate tentacles around the mouth, just as in Asyn-
coryne, while over the body there are arranged a number of
pinnate capitate tentacles. The momiliform tentacle of
Asyncoryne may perhaps be regarded as intermediate in
nature between an ordinary filiform tentacle and a pinnate
capitate one.

In Coryne and Syncoryne we have scattered capitate

A COLLECTION OF HYDROIDS. 289

tentacles; and it would appear probable that both Asyn-
coryne and Cladocoryne are to be regarded as modifica-
tions of the Coryne type.

As the generic name implies the present hydroid is especi-
ally associated with Syncoryne, since the reproductive
bodies are, with great probability, free-swimming meduse.
The general shape of the hydranth also recalls Syncoryne,
although the habit of growth of the colony is simpler.

(10) Coryne pusilla Gartner.

CORYNE PUSILLA Giirtner. Pallas, ‘Spicil Zool,’ fase. x, p. 40; Allman,
G. J.. ‘A Monograph of the Gymno-
blastic Hydroids,’ 1872, p. 266.

The identification of the Natal hydroid with Gdartner’s
pusilla is not made with any certainty; but without a
detailed acquaintance of the variability and general growth-
forms exhibited by pusilla it appears advisable to refer it to
this species, as no obvious specific difference has been deter-
mined.

The hydroid appears to be very variable in size. There is
a dwarf variety, about one quarter of an inch in height, which
stretches over large areas of seaweed (text-fig. 4b) exposed
to the force of the waves, while in more sheltered places it
grows to nearly double the size (A).

TropHosomMe.—H y drorhiza forms an irregular loose mesh-
work on weeds or rocks. The perisarc is smooth, without
annulations.

Diameter 07162 mm., thickness of perisarc 26°34. There is
no marked difference in the size of the hydrorhiza in the
dwarf and large varieties.

Hydrocaulus.—Straggling and irregularly branched ;
serpentine, blind-ending branches occur. The annulation o
the perisarc is variable and irregular; it is generally most
distinct for some distance below the hydranth; there are
about 20 rings to the millimetre.

290 ERNEST WARREN.

Diameter about 0°17 mm., and thickness of perisare 15°1 1 ;
these dimensions are somewhat smaller in the dwarf variety.

Hydranth.—The perisare of the hydrocaulus is not
swollen into a loose sheath at the base of the polyp. The
number of tentacles in the large variety varies from 20-30,
and in the dwarf variety from 10-18.

The average length of the polyp in the large variety is

TEXT-FIG. 4.

BY

ol
a» OD
S
Pha
ry
la 9 a =
<

HD

c
ee)

Q,

ei

s
aaneei
of ft
re@aanl

908,
ates

yy

A.

Coryne pusilla Gdrtner. A. Large form x 15. B. Small
form x -15.

1°31 mm., and breadth 0°363 mm., in the small variety 0°80
mm. and 0:22 mm. respectively. The diameters of the
capitula of the tentacles are about 0°15 mm. and 0:10 mm. in
the two forms.

The nematocysts in the capitula are of two sizes: the large
measure 16°7 « in length and 11:2 « in breadth; the small,
8:9 w and 5°3 w respectively. The sizes of the nematocysts in
the two varieties appear to be about the same.

A COLLECTION OF HYDROIDS. 29]

Gonosome.—In the dwarf variety the reproductive bodies
have not been met with, although they have been searched
for im what appear to be old colonies, since they extended
over considerable areas.

In the large variety the male reproductive bodies only have
been found (text-fig. 4, 4). It is clear that they are fixed
gonophores, and they do not apparently differ from those
described for pusilla.

Small errant polychets and chetognaths appear to form
the staple food of this hydroid.

(11) Sertularella polyzonias (Lim.). (Pl. XLVII, figs.
18-20.)

SERTULARELLA POLYZONIAS (Lin.) Syst; Hincks, T., ‘A History of
the British Hydroid Zoophytes,
1868, p. 235; Bale, W. M.. ‘ Cat.
of the Australian Hydroid Zoo-
phytes,’ 1884, p. 104; Nutting,
C. C., ‘American Hydroids,’ pt.
ii, ‘ Sertularidx,” 1904, p. 90.

It is scarcely possible to define this widely distributed
species with any certainty. It is very variable, and the Natal
hydroid is referred to it, as it is undesirable to found new
species on characters (such as the presence or absence of
internal knobs of chitin at the mouth of the hydrotheca)
which are admittedly highly variable.

The hydroid is found on sea-weeds and worm-tubes, the
colony is pale brown, and the stems may reach a height of
three quarters of an inch.

TropHosome.—Hydrorhiza has an irregular outline, and
branches sparingly.

Diameter, about 0°242 mm., and thickness of perisarc, 23 pu.

Hydrocaulus.—The hydrorhiza produces at irregular
intervals upright stems bearing from 13-20 hydrothece.
The stems are generally unbranched, but occasionally a
lateral branch may be formed (text-fig. 5,4). The stem is
divided into faintly-defined internodes by oblique divisions.

292 ERNEST WARREN.
Asa rule the stem is perfectly straight, and only rarely is it
shehtly zig-zag.

At the origins of the stems and branches there is frequently
an annulation of 5—4 rings.

Diameter of stem, measured a short distance above its
origin from the hydrorhiza, 0°278 mm; thickness of perisare
38 mu.

TEXT-FIG. 5.

A and B. Sertularella polyzonias (Lin.), x 18. Cand D.
Sertularella fusiformis (Hincks). x 18.

Hydrotheca.

Alternate, ventricose below, contracted
above, divergent, expanding mouth. About one half is sunk
in the hydrocaulus. The mouth is quadrangular and provided
with four teeth at the angles, two of which are lateral, one
adeauline, and one abcauline.

» ‘Three internal knobs of chitin, constricting the size of the
mouth, are usually developed near the edge; two are placed
in the middle of the two sides of the square on the adcauline

A COLLECTION OF HYDROIDS. 293

side (P]. XLVI, figs. 18 and 19 /,) and one below the abcau-
line tooth (/,). Sometimes there are four knobs, the abcauline
one being replaced by two in the middle of the abcauline
sides of the square. There is an operculum of four flaps.

Length of hydrotheca, measured parallel to the abcauline
surface, 0°55 mm; greatest breadth, 0°26 mm.

Hydranth.—It has the typical sertularian structure with
the dilatation of the ccelenteron on the outer side, and a sheet
of ectoderm lining the upper portion of the inside of the
hydrotheca. The hypostome is conical, and there is a verticil
of about 25 tentacles. The nematocysts on the tentacles
are small and narrow; they measure about 6°2 « in length
and 1°8 mw in breadth.

Gonosome.—The gonangia are produced from the hydro-
caulus just below a hydrotheca; they are provided with three
apical teeth. The male differs from the female in being
narrower, and the apical teeth tend to be longer and closer
together. The male gonangium (text-fig. 5,4) has 3 or 4
rings around the apical region, while it is smooth below : the
female (B) has 5 or 6 rings.

The male gonangium has an average length of 2°28 mm.
and greatest breadth 0°73 mm. The female measures 2°42
mm. and 0°97 mm. respectively. Thickness of perisarc
about 17 nu.

Sometimes short lateral branches (Pl. XLVII, fig. 19, b)
arise from the inside of a hydrotheca, thus recalling the mode
of origin of the gonangia in the genus Synthecium. In
one case a single hydrotheca, complete with operculum, issued
from the old hydrotheca; this was doubtless formed after the
disintegration of the polyp and during its subsequent renewal.

Gall caused by a Pycnogonum.
Occasionally there may be seen springing from the inside
of a hydrotheca a large fusiform body. This structure, which
may be three times the length of a hydrotheca, is a gall

294, : ERNEST WARREN.

produced by the presence of a parasitic larva of a Pycno-
gonum (Pl. XLVIL fig: 18, g.).

An egg of the Pycnogonum finds its way into the diges-
tive cavity of a hydranth. It is, of course, quite possible that
the parent may actually deposit its eges in the polyp. The
polyp contracts, loses its tentacles and the characteristic
dilatation of the ccoelenteron, and becomes in fact a closed sac
(fig. 18, c.hy.); y.e. is the young developing embryo. The
polyp-sac then elongates and projects out of the mouth of the
hydrotheca (fig. 19, d.g.), and ultimately it grows and produces
a large gall (fig. 18, g.) into which the embryo has passed.

The gall is remarkable for its structure: it is lined by a
layer of ectoderm (fig. 20, e.g.) like a hydrotheca, and it
generally terminates in four marginal teeth and a_ well-
developed operculum (op.). The length of a gall is about
1°50 mm., and its greatest breadth 0°65 mm. ‘Thickness of
perisare 20 uw. The stalk of the gall sometimes shows a
tendency to be annulated.

The outgrowth of the original hydranth has a well-defined
ccelenteron, led by a regular endodermal epithelium, outside
of which is a thin ectoderm. _ Sometimes the structure is
surmounted by a set of well-developed tentacles (figs. 18 and
20, te.) ; but a definite mouth has not been observed. At
other times the structure ends blindly without tentacles.

The embryo, lodged in the ccelenteron, grows, and there
are developed two chelicere (fig. 18, ch.) ending in two little
plates or lappets, which are perhaps of the nature of claws.
In an older embryo two long curved claws may be found in
this position (fig. 18, cl.).

In a_ longitudinal dorso-ventral section of the young
embryo (fig. 20) there may be observed the following
characters :

(1) The embryo surrounded by a cuticle (c.).

(2) The stomodeeum (St.) dipping down towards the arch-
enteron (47.) which is lined by elongated granular endoderm
cells. These cells are placed remarkably separate from one
another, and are frequently branched (end.).

A COLLECTION OF HYDROIDS. 295

(3) The loose mesoderm tissue (m.), and the ova (0.) which
have already appeared on the ventral aspect, and extend into
the base of the chelicere.

(4) The cheliceree (ch.), which are inserted into the endo-
derm of the hydroid.

In an older embryo, represented in fig. 18, 0./., clawed
chelicere (cl.), the oral cone (0.c.), and paired, jointed
appendages (ap.) may be seen.

The larva ultimately breaks out of the gall, and in fig. 19,0.9.,
an empty broken gall is shown.

The peculiar interest of this example of parasitism is the
very definite nature of the response which is made by the host
to the stimulation caused by the presence of the developing
embryo. Many animal galls and cysts are simple, rounded
structures ; but here a structure is formed which is practically
an enormous hydrotheca and hydranth, the former being
provided with marginal teeth and operculum, and the latter
may even possess tentacles. There appears, however, to
be no mouth, and very probably the operculum is never
opened.

In this connection it may be noticed that the hydranth of
this species exhibits a marked capacity for extended growth.
Through some unknown stimulus a hydranth may round itself
off and produce, as we have already seen, a short branch or a
new hydrotheca, which thus springs from the mouth of the
original hydrotheca (fig. 19, 6.).

G. Hodge has already described a somewhat similar case,
wherea larval Pycnogonum forms galls on a campanularian
hydroid. From a figure given by Karl Semper, in his ‘Animal
life,’ p. 332, it would appear, however, that these galls are
simple, irregular structures, and cannot be compared with the
definite structure occurring in Sertularella polyzonias.

(12) Sertularella fusiformis (Hincks). (Text-fig. 5,
Cand =D).

SERTULARIA FUSIFORMIS Hincks, ‘Ann. Mag. N.H.,’ vol. viii, p. 253.

296 ERNEST WARREN.

SERTULARELLA FUSIFORMIS (Hincks), Hincks, T., ‘A History of the
British Hydroid Zoophytes,’
p. 243; Hartlaub, Cl., ‘ Hy-
droiden aus dem Stillen Ocean
Zool. Jahrbicher,’ vol. xiy,
Jena, 1901, p. 372; Nutting,
C. C., ‘ American Hydroids,’
pt. u, “Sertularide,” 1904,
p. 89.

As in the case of the last species, the identification of the
present species with 8. fusiformis can only be regarded as
provisional, until the hmits of variability im fusiformis are
more definitely known.

The Natal hydroid is found on sea-weeds or creeping on
larger hydroids. It is considerably smaller than 8. poly-
zonias, the stems not bemg more than about one quarter of
an inch in height. It is pale-brown in colour.

TropHosomMeE.—H ydrorhiza consists of an irregular, creep-
ing stolon, sparingly branched.

Diameter about 0°15 mm. ; thickness of perisare 0°018 mm.

Hydrocaulus.—The stolon produces upright stems, which
have never been found to branch, and they carry from 6 to 10
hydrothece.

The stem is distinctly zig-zag, and is divided by faint
oblique divisions into internodes, each bearing a hydranth.
The base of each internode is ridged by one, or by one and a
half, spiral turns (text-fig. 5, C).

Diameter, measured close to the base, about 0°130 mm. ;
thickness of the perisare 20 1.

Hydrotheca. — Alternate, ventricose, adnate for about
half its length to the hydrocaulus. It is less divergent,
and the quadrangular mouth is less expanding than in 8.
polyzonias. The adcauline free surface is sometimes shehtly
ridged (7).

There are four teeth, and the operculum consists of four
flaps. ‘There are no internal knobs of chitin constricting the
mouth, as occur in the Natal specimens of 8. polyzonias.

A COLLECTION OF HYDROIDS. 297

Length, measured parallel to abcauline surface, is about
0°420 mm., and greatest breadth 0°250 mm.

Hydranth.—It has the typical structure. Tentacles,
about 24. Nematocysts are small and narrow; they tend to
be especially abundant at the tips of the tentacles. Length
5S'lu, and breadth Iv1 pu.

GonosomE.—Gonangia ovate, but smaller and less elongated
than in S. polyzonias. The whole surface is generally
ridged transversely by 7-8 ruge. Inthe female (text-fig. 5, D)
the opercular surface is wide and nearly flat, and there are
no obvious teeth; in the male the neck is very narrow, and
the teeth, of which there are three, are well developed.

Length of male gonangium about 1°25min., breadth 0'75mm. ;
length of female 1°55 mm., breadth 0°69 mm. ; thickness of
perisare 16 u.

/

(13) Sertularella tumida sp. 1.

A species of hydroid, very closely allied to Sertularella
gayi (Lamouroux), was dredged from a depth of 40 fathoms
about sixteen miles north-east of Bird Island, near Algoa
Bay. In addition to these specimens there was picked up on
the shore near Park Rynie, Natal, a loose stem of a hydroid
which is regarded as the same as that from Bird Island,
although there is a sheht difference in the curvature of the
free adcauline surface of the hydrotheca (text-fig. 6, Band C).
The specimen from Park Rynie, which had doubtless been
thrown up on shore from some depth, was richly supphed
with gonangia, while there were none on the specimen from
Bird Island.

The description given by C. C. Nutting! for 8. gayi will
apply with sheht modification.

TropHosome.—Hydrorhiza unknown.

Hydrocaulus.—Colony straggling in habit, and attaining
a height of about 4 inches. Stem fascicled (f.s.), being made

1 Nutting, C. C.,‘ American Hydroids, Part I, “ The Sertularide,”
1904, p. 78.

298 ERNEST WARREN.

up of intertwining tubes, which in the aggregate form a thick
“woody ” stem, 2-3 mm. in diameter, and bearing no trace of
internodes or regularity of branching. The stems, which bear
hydranths, give off branches more or less alternately with a
pinnate tendency (text-fig. 6, A); but the arrangement may be
very irregular and indefinite. The stems bearmg hydranths,

TEXT-FIG. 6.

MALE.

Sertularella tumida sp.n. A and C. Specimen from Bird
Island. B. Specimen from Park Rynie.

and the branches, are divided by oblique nodes into regular
internodes, each carrying a hydrotheca.

Diameter of terminal branch just below a hydrotheca is
about 0°46 mm., and the thickness of the perisare 35 uw. The
hydrocaulus is generally considerably constricted in the
region of the node.

Hydrotheca.—Ovate, divergent, distal end contracted,
but slightly expanding immediately around the margin ;

A COLLECTION OF HYDROIDS. 299

adnate for one-half to two-thirds of its length to hydrocaulus.
Adcauline free surface smooth. The margin has four small
equidistant teeth, of which the lateral pair are perhaps the
best developed; there is an operculum of four flaps, two
adcauline and two abcauline. Below the margin there are
three internal knobs of chitin, two adcauline (ad.k.) and one
on the opposite surface (ab.k.).

In the specimen from Park Rynie (B) the hydrotheca is
sunk in the hydrocaulus to a less extent than in that from
Bird Island (C).

The adcauline free surface of the hydrotheca is swollen,
especially in the specimen from Bird Island, and it is perfectly
smooth and not transversely ridged as in 8. gayi.

Length 0°59 mm., greatest breadth 0°28 mm.

Hydranth.—Badly preserved, perhaps about 28 tentacles.

Gonosome.—The gonangia are known only from the Park
Rynie specimen (text-fig. 6, B). The male gonangium is ovate,
with a short tubular neck and three short apical teeth. Over
the distal half there are about seven transverse ruge; the
proximal half is smooth.

Total length about 2°08 mm., breadth 0°87 mm. ; thickness
of perisare 28 mu.

Systematic Posrrion.— According to Professor Nutting
S. gayi has not been found in the Pacific Ocean, and it is
now a question whether the differences between the present
hydroid and gayi are sufficient to warrant the formation of a
new species. The only marked differences are: the complete
absence of rug on the hydrotheca, and the amount to which
the hydrotheca is sunk in the hydrocaulus. In 8. gayi not
more than one-half of the hydrotheca is adnate to the hydro-
eaulus, while in the hydroid from Bird Island two-thirds are
generally adnate. This last character, however, is certainly
variable, especially if there is justification in considering the
specimens from Bird Island and Park Rynie as belonging to
the same species (text-fig. 6, cf. B and C).

Only by a series of observations on the variability of
S. gayi, with respect to the rugze on the hydrotheca, would it

voL. 1, part 3. 22

300 ERNEST WARREN.

be possible to arrive at any decision as to whether or not the
Pacific hydroid should be regarded as the same as the Natal
hydroid. In the meantime it has been considered advisable
to give a new name to the present hydroid, and tumida has
been chosen on account of the swollen condition of the
hydrotheca.

(14) Sertularella campanulata sp.n. (Pl. XLVI, figs.
21 and 22.)

This minute but remarkable hydroid occurs creeping on
sea-weeds, often in company with Pasythea quadriden-
tata. It has been found at Scottburgh and Park Rynie, but
is not particularly common. The habit of growth is more hke
that of a Campanularian than a Sertularian, and the species is
clearly allied to Sertularella solitaria Nutting, from the
Bahamas, and Calamphora parvula Allman, Bass Strait,
Austraha.

TropHosome.—Hydrorhiza consists of a creeping, some-
what flattened stolon branching sparingly or forming a loose
reticulum.

Diameter 0°061 mm. On the surface attached to the sea-
weed the perisare is very thin, while on the outer surface it
has a moderate thickness, 8°24 (Pl. XLVII, fig. 22, R.).

Hydrocaulus.—Short stems, with 2-5 spiral turns, arise
from the hydrorhiza and carry a terminal hydrotheca (fig.
22, H.). The stems appear always to arise at an angle of about
45° with the hydrorhiza, and they sometimes come off sloping
alternately to the right and left sides of the hydrorhiza.

The diameter of the hydrocaulus (which is practically the
peduncle of the hydrotheca) is 0°083 mm. and the average
leneth 0°16 mm. The thickness of the perisarc is about 8°7.

Hydrotheca.—Barrel-shaped, set terminally on hydro-
caulus, provided with 5-7 well-marked transverse ruge ; the
proximal portion is sometimes smooth. The mouth is
quadrangular and expanding, and has 4 pointed teeth, which
appear to be arranged in a definite manner with respect to

A COLLECTION OF HYDROIDS. 301

the origin of the hydrocaulus from the hydrorhiza, and the
bilateral character of the perisarc of the hydrotheca; two of
the teeth are lateral, one on the side facing the hydrorhiza,
and one on the opposite side.

There is a four-flapped operculum, forming a four-sided
pyramid,

The cavity of the hydrotheca is partially shut off from the
cavity of the hydrocaulus by a diaphragm (d.), which resembles
that seen in a Campanularian. The opening of the
diaphragm is excentric, being nearer the side facing the
hydrorhiza.

Length 0-40 mm., greatest breadth 0°26 mm. The thick-
ness of the perisare varies considerably on the two sides; on
the side facing the hydrorhiza it is 4 and on the opposite
side 10 u (Pl. XLVII, fig. 22).

Hydranth.—The polyp has the typical sertularian
structure. The hydrotheca is lined by a thin epithelium of
ectoderm (fig. 22, e.s.) which generally joins the ectoderm of
the hydranth near the diaphragm; but it appears not to be
very constant in its place of junction. There is the usual
dilatation of the ccelenteron (C.D.) which is on the side facing
the hydrorhiza. The hypostome is more or less hemispherical.
There is a single verticil of about twenty tentacles. The
nematocysts are few and small; they measure about 3 in
length and 0°7 « in width,

The hypostome has a narrow endodermal epithelium ; the
outer side of the dilatation of the ccelenteron is lined by a
comparatively flat undifferentiated endoderm (w.e.). The rest
of the hydranth is lined by elongated vacuolated cells with
granular cells (gr.c.) wedged between. At the base there
are some clumps of special granular cells (gl.c.).

Gonosome.—Unknown.

Systematic Postrion.—The present hydroid is undoubtedly
closely allied to Calamphora parvula! Allman, and Ser-
tularella solitaria? Nutting. In the former the hydro-

1 Allman, J. G., ‘ Challenger Reports,’ vol. xxxiii, p. 29.

? Nutting, C. C., ‘American Hydroids,’ pt. ii “ Sertularede,” p. 89.

(0s ERNEST WARREN.

theca is practically sessile on a creeping stolon, and its height
is about #4, nmch. It is transversely annulated with about
twelve ridges, and the tetragonal mouth is stated to be
inoperculate. The hydrotheca is rendered bilateral by the
slope of the neck towards one side. It was found at a depth
of thirty-eight fathoms.

In Surtularella solitaria, found in shallow water at the
Bahamas, the hydrotheca is provided with a fairly long stalk
with one or two annulations near the middle. Hydrotheca
radially symmetrical, mouth quadrate, slightly everted, and
four equidistant teeth. Body of hydrotheca annulated with
about eleven transverse ridges. Operculum of four flaps.

In both cases the anatomy of the hydranth is not touched
upon, but it is probable it is sertularian in nature as in 8.
campanulata. It is curious that Calamphora parvula
is stated to be inoperculate ; S. solitaria and campanulata
are both obviously operculate.

It is interesting to note that, independently of the square
neck and mouth, the hydrotheca in both C. parvula and S.
campanulata exhibits a certain bilateral symmetry, thus
still further emphasising the sertularian nature of the hydroids.
In C. parvula the neck of the hydrotheca slopes towards
one side and the mouth is consequently not terminal and
radially symmetrical. In 8S. campanulata the peduncle of
the hydrotheca arises at an angle of 45° to the hydrorhiza,
and on the side facing the hydrorhiza the perisarc of the
hydrotheca is less than one-half the thickness of that on the
opposite side. The diaphragm, also, is excentrically perforated,
the aperture being on the side having the thin perisare.

Allman expressed some doubt as to how his C. parvula
should be classified, whether as a Sertularian or Campanu-
larian. The general habit of growth and apparent radial
symmetry of the three species would seem to place them
among the Campanulariide ; but the anatomy of the hydranth,
the operculated, dentate, quadrate mouth, and the real bilateral
symmetry, which exists under an exterior radial symmetry,
unmistakably point to sertularian affinities.

A COLLECTION OF HYDROIDS. 303

(15) Sertularia acanthostoma Bale. (Pl. XLVI, figs.
23-26.)
SERTULARIA ACANTHOSTOMA Bale, ‘Catalogue of the Australian
Hydroid Zoophytes, 1884, p. 85.
There is probably little doubt that the Natal hydroid is
Bale’s acanthostoma; but, nevertheless, there are certain

differences.

TEXT-FIG. 7.

OS:

Sertularia acanthostoma Bale.

The pinnate shoots reach a height of about 2 in.; the
hydroid is straw-coloured, and grows attached to rocks near
the low-water line. ‘Two small colonies only have been
found.

TropHosomE.—Hydrorhiza consists of a branching, loose
reticulum.

Diameter 0°101 min. ; thickness of perisare 10 p.

Hydrocaulus.—Pinnate shoots, pinne slender at their
origin, opposite. Typically there are three pairs of hydro-

304 ERNEST WARREN.

thecee on the main stem between every two pairs of pinne ;
but sometimes there are only two pairs, and occasionally only
a single pair (text-fig. 7, B).

Diameter of main stem just below the proximal pair of
pinnee, 0°24 mim. ; thickness of perisarc, 40 #.

The specimen shown in text-fig. 7,C is peculiar, in that a
young pinnate shoot is seen springing from the upper surface
of the main stem of an old pinnate shoot (0.s.).

Hydrothece.—Opposite on main stem, subalternate on
the pinne, usually one pair on every internode. Tubular in
shape, expanding upwards; aperture oval with four outer
marginal teeth alternating with four inner (cf. figs. 25 and 24,
Pl. XLVI). Ina vertical section (fig. 25) through the hydro-
theca three projecting thickenings of chitin may be seen, one
in the middle of the inner surface (7.), one near the margin
of the outer surface (plé.), and a curved peg (l.p.) near the
base on the outer surface.

Length of hydrotheca 0°38 mm., and greatest breadth
0-18 mm.

Bale! mentions and figures an oblique fold (intra-thecal
ridge) extending across the hydrotheca from the process (7.)
on the inner surface. Inthe Natal hydroid this fold is scarcely

represented.
Hydrantlh.—Short conical hypostome and a verticil of
23 tentacles (fig. 25). There is the usual dilatation of

the ccelenteron on the outer side (C.D.), and in the con-
tracted condition of the polyp there is also an expansion on
the inner side (¢.p.). The outer dilatation is attached to the
peg of perisare at the base of the outer surface (l.p.) ; the
endodermal epithelium on the outer side is flat and undifferen-
tiated (w.e.). The rest of the endoderm consists of elongated
cells with granular cells wedged between. The inner surface
of the hydrotheca is lined by a thin layer of ectoderm (e.s.)
which joins the ectoderm of the polyp at the extreme base.
At the front outer edge of the hydrotheca there is a concave

' Bale, W. M., ‘ Catalogue of the Australian Hydroid Zoophytes,’
1884, p. 85.

A COLLECTION OF HYDROIDS. 305

depression (plt.) in which lies a thickening of the ectodermal
epithelium (figs. 25, 26 B.n.). This thickening has very much
the structure of a nematophore, it bemg provided with a battery
of large nematocysts which resemble those occurring in the
Plumulariidx; they measure 22°7 « in length and 4°24 in
breadth.

In the specimens at my disposal it was not possible to
decide whether there were ordinary nematocysts on the
tentacles.

Gonosome. — Unknown.

This hydroid was very largely enveloped by a coralline alga
(fig. 26, C.W.), which formed a perfectly regular layer over the
greater part of the outer surface of the colony. It renders the
colony very firm and strong, and in this way it may possibly -
be of some use to the hydroid.

The hydroid is, of course, peculiar among the Sertularians
in the remarkable toothed condition of the margin of the
hydrotheca, and it is further distinguished by possessing no
trace of operculum.

(16) Sertularia operculata Lin.

SERTULARIA OPERCULATA Lin., ‘Syst.,’ Hincks, T., ‘Brit. Hyd. Zooph.,’
p. 263; Bale, W. M., ‘Cat. of Austr.
Zooph.,’ 1884, p. 67; Nutting, C.C.,
‘American Hydroids,’ pt. ii ‘* Ser-
tulariide,” 1904, p. 64.

Found attached to the surface of rocks; greenish-brown in
colour. Height about 3 in.

TropHOsoME.—Hydrorhiza forms a feltwork, very firmly
attached to the surface of the rock. Diameter 0°105 mm;
thickness of perisare 15 «.

Hydrocaulus.—It branches dichotomously; a hydro-
theca at each side of every axil and in contact with each
other.

306 ERNEST WARREN.

The internodes, bearing a pair of hydrothece, are distinguish-
able by the presence of nodes only on the terminal’ branches.
In the older branches and main stems they are not present.

Diameter of main stem at the base 0°24 mm., thickness of
perisare 36 4; terminal branch, diameter 0°20 mm.

Hydrotheca.—In pairs, opposite, not in contact, tubular,
slightly divergent, adnate nearly up to the margin; two
spine-like abcauline teeth, the back one being the larger
and slightly incurved. Operculum two flaps.

Length of hydrotheca, measured along the outer edge, about
0°640 mm., greatest breadth 0°202 mm.

Hydranth.—About 18 tentacles.

(GonosomE.—Gonanegia, long, ovate, aperture operculate with
a shghtly elevated border.

In the male, length 1-414 mm., breath 0°505 mm., thickness
of perisare 7 pu.

The above description only differs from that given by Bale
in connection with colour, absence of nodes in the older stems,
and the mention of an operculum.

It may be noticed that in the figure given by Hincks! the
nodes are not shown.

(17) Sertularia loculosa Busk. (Pl. XLVI, fig. 37.)

SERTULARIA LOCULOSA Busk, ‘ Voy. of Rattlesn.,’ 1852; Bale, W. M.,
‘Cat. of Austr. Zooph.,’ 1884, p. 91.

This is a common hydroid on the Natal coast; it occurs on
coralline and other seaweeds. The colour of the colony is
sometimes almost black, and at other times a pale brown. It
is not certain whether the difference in colour is due to
different physiological and histological conditions which may
occur at different times in the same colony, or whether the
two forms indicate two distinct varieties. There is no obvious
difference in the ordinary specific characters in the two kinds.
The height is about half an inch.

' Hincks, T., ‘ British Hydroid Zoophytes,’ 1868, Pl. LIV, b.

A COLLECTION OF HYDROIDS. 307

TropHosomE.—H ydrorhiza is a creeping stolon, irregularly
branched. Diameter 0°152 mm., thickness of perisarc 18 «.

Hydrocaulus.—Simple upright stems irregularly divided
into hydranth-bearing and non hydranth-bearing internodes
by transverse and oblique nodes. The base of the stem is pro-
vided with two spiral turns. The apex of a stem is capable
of continued growth, and it may produce a stolon-like out-
erowth, indistinguishable from the hydrorhiza, from which
upright stems arise (text-fig. 8, C, O.H.). When the apex of

TEXT-FIG. 8.

SEMALE

¢ A (nat. szze)
BS

Steel

Sertularia loculosa Busk.

a stem comes into contact with a piece of weed or other
object, it appears to be stimulated to grow in this way, as
separate pieces of weed may be firmly bound together by
outgrowths of the ends of the stems.

Length of hydranth-bearmg imternode, measured from a
transverse joint to the opposite pointed extremity of an oblique
articulation 0°50 mi.

Hydrotheca.—Short and broad, opposite; in contact in
front (B), separate behind (1D). Horizontal fold occurs about

308 ERNEST WARREN.

in the middle of the outer surface. Aperture somewhat con-
tracted, directed outwards ; two rounded abcauline teeth.

Length, measured along abcauline edge, 0°293 mm., greatest
breadth 0°172 mm. Distance from mouth to mouth of a pair
of hydrothece 0°570 mm.

Hydranth.—About 12-15 tentacles. Nematocysts, few,
small, and inconspicuous, about 4°14 in length, and 1°24 in
breadth.

GonosomE.—Gonangia ovate, upper surface flat or shghtly
convex ; 7-9 prominent transverse annulations, aperture wide,
operculate.

Length about 1°62 mm., breadth 0°847 mm. The enclosed
blastostyle is branched.

The above description agrees with that given by Bale,
except that the gonangium in the Australian hydroid appears
to have 4—5 ridges in place of 7-9.

Histology.—In the black variety the endoderm is pro-
vided with dark brown, opaque, pear-shaped masses (PI.
XLVIII, fig. 37, ¢.p.). These are, doubtless, cells, but I have
not with certainty detected the nucleus. The colouring
matter nay be very opaque and compact, but sometimes it is
seen in the form of granules.

(18) Sertularia linealis sp.n.

The hydroid is characterised by running in longitudinal
lines over the surface of ribbon-lke sea-weeds. It is of a
dark brown or black colour, and it does not appear to attain
to a height of more than about one fifth of an inch. It was
collected in the rock pools at Kosi Bay, Zululand.

TropHosome.—H y drorhiza runs in longitudinal lines con-
nected together by occasional cross-branches (text-fig. 9,4).

Diameter about 0°23 mm. The outer layers of the perisarc
may become remarkably diffuse and swollen. The inner
compact portion has a thickness of about 18 uw, and the diffuse
outer portion 47 pu.

Hydrocaulus.—lt consists of short upright stems divided

A COLLECTION OF HYDROIDS. 309

by oblique joints into 5-6 regular internodes, each bearing
a pair of opposite hydrothece. At the origin from the
hydrorhiza the stem is spirally ridged by one or two turns.
Length of internode 0°30 mm. Width between the mouths
of an opposite pair of hydrothece 0°48 mim.
Hydrotheca.—Tubular, divergent, mouth facing outwards
and contracted. The hydrothecz are nearly in contact in
front (C), separate behind (B). Teeth lateral and very

TpxT-PiIeG. 9.

( (F7O?L)

Sertularia linealis sp.n.

rounded and blunt, the back one being somewhat the wider.
The mouth is internally constricted by perisare; on the
adcauline surface there is a deep pit on the outside, appearing
as a peg on the inside (k,) ; sometimes the pit may be more
or less filled up with perisare. On the abcauline inner surface
there is a perisarc ridge which may be developed into one or
sometimes two knobs (Kk). Two-flapped operculum.

Length of hydrotheca, measured parallel to abcauline edge,
0-211 mm.; breadth 0°163 mm.

310 ERNEST WARREN.

GonosomME.—Male gonangium subglobular, smooth, with
flat operculum. Blastostyle may develop two gonophores.

Length, including stalk, about 0°84 mm. ; breadth 0°67 mm.
Operculum 0:21 mm. in diameter. The perisare has a thick-
ness of 21 wu.

(19) Sertularia bidens Bale.

SERTULARIA BIDENS Bale. ‘ Cat. of the Australian Hydroid Zoophytes,’
1884, p. 70.

The Natal hydroid does not entirely agree with Bale’s
description of bidens; but the differences appear to be
scarcely sufficient to warrant the formation of a new species.
The colour is a very pale brown, not dark brown, and the
height of the largest specimen found was only about 3 in.

‘'ropHosome.—H ydrorhiza forms a loose feltwork. Dia-
meter 0°172 mm., thickness of perisare 22 wp.

Hydrocaulus.—Main stem bears pinnate stems irregularly
placed (text-fig. 10, 4). Just below the first pimna of the
pinnate stem there is a conspicuous spiral groove running
round the stem about twice (C). The portion of the stem
bearing pinne is about half an inch in length. Pinne
alternate. The stem is zig-zag, but the nodes are indis-
tinguishable ; between two pinne of one side there are three
hydrothecz, one in the axil and two above. On the pinne
the nodes are few, the first internode carrying 4 or 9 pairs
of hydrothece, while the remainder carry from I to 3.

Diameter of a pinna, measured just below the base of a
hydrotheca and about at the middle of its length, 0°152 mm. ;
thickness of perisarc, 5°1 yp.

Hydrotheca.—Subalternate, not in contact with each other,
tubular, upper side horizontal, with the inner margin of the
mouth rather widely separated from the hydrocaulus (text-
fig. 10, D). Mouth horizontal, the outer margin provided with
two pointed teeth (¢.) ; on the inside of the inner margin there
are usually developed two distinct knobs of chitin (ad.k.),
and on the opposite side on the outer edge there are a similar

A COLLECTION OF HYDROIDS. ad

pair (ab.k.). There is also occasionally a little peg of chitin
on the outer wall a little above the base (/.p.).

Total length, measured parallel to the abcauline edge,
0263 mm.; greatest breadth 0-105 mm. Operculum probably
two flaps.

Hydranth.—Tentacles 13-14. The layer of ectoderm lining
the hydrotheca is thickened along the outer margin (fig. D,
n.), and it contains a.row of vertically placed nematocysts,

TrExtT-FIG. 10.

Sertularia bidens Bale.

measuring 6°6 «in length and 1°44 im breadth. Nematocysts
were not distinguished on the tentacles.

There is also a thickening of the hydrothecal ectoderm on
the adeauline surface (D, th.).

GonosoME.—Gonangia long, sub-tubular, with two sharp
angles at the sides of the aperture. Aperture operculate,
margin elevated and shehtly everted. The gonangia originate
from the stem just above the origins of the pinne, and they
are regularly directed right and left.

Sl ERNEST WARREN.

Total length 1°70 mm., length of spines 0°39 mm., greatest
breadth 0°59 mm.

Sysrematic Posrrion.—The description of the Natal hydroid
differs from that given for bidens from Austraha in the
following characters :

(1) The much greater size of the sharp angles of the
ovonangia,

(2) The relative shortness of the hydrotheca.

(3) The presence of internal knobs of perisarc near the
margin of the mouth.

(4) The colony is pale brown.

Taking the characters altogether, the Natal hydroid appears
somewhat intermediate between Bale’s S. maplestone1 and
his S. bidens. The internal knobs of perisare are not men-
tioned; but we know from other species that their presence
or absence cannot be regarded of specific value. It is possible
that the Natal hydroid, 8. bidens and 8S. maplestonei are
varieties of the same species; but a direct comparison of a
series of specimens would be necessary for arriving at a
definite conclusion.

(20) Pasythea quadridentata (Hllis & Sol.).

SERTULARIA QUADRIDENTATA Ellis and Sol. ‘ Zooph., 1786, p. 57.

PASYTHEA QUADRIDENTATA (Ellis and Sol.). Bale, W. M., ‘Cat. of
Australian Hydroid Zoophytes,’ 1884,
p. 112; Nutting, C.C., ‘American
Hydroids, pt. ii, “The Sertularide,”
1904, p. 75.

The hydroid grows on sea-weeds ; it 1s yellowish-brown and
about } inch in height.

TropHosome.—Hydrorhiza forms a wide-meshed reticu-
lum. It is characterised by possessing internal ribs of peri-
sare which project into the cavity from the vertical sides, and
appear from above as pegs which extend internally for about
one quarter to one third of the diameter of the hydrorhiza
(text-fig. 11). The presence of these pegs renders the hydro-

A COLLECTION OF HYDROIDS. ole

rhiza readily distinguishable from the hydrorhize of other
hydroids amongst which it may be living.

Diameter, from above, 07121 mm.; thickness of perisare 9 yu.

Hydrocaulus.—Consists of upright stems with no internal
pegs. Divided by oblique nodes into regular internodes,
bearing a pair of hydrothece, or two or three pairs in a
closely compressed group.

Diameter of base of stem 0:081 mm., thickness of perisare
15 uw, length of internode 0°798 mm.

Hy drotheca.—On the first internode there may be only a
single opposite pair, in contact for about 4—} their height in

Trext-Fic. 11.

Pasythea quadridentata (Ellis and Sol), x 30.

front, the distal part curving to a narrow tridentate mouth
looking outwards. Two lateral teeth (fig. 11, 2,2), one adcau-
line (1). Operculum of two flaps. Length of proximal
hydrotheca, measured parallel to abcauline edge, 0'283 mm.,
greatest breadth 0°145 mm., thickness of perisare 10 p.

Hydranth.—13-15 tentacles.

Gonosome.—Female gonangium borne at the base of the
stem ; ovate, with 5-6 transverse ruge., Very broad aperture
with flat operculum (text-fig, 11, 2).

Length 0°808 mm., greatest breadth 0°412 mm.

The above description is practically the same as that given
by Nutting. It is stated that “this species seems to be
always found growing on floating sea-weed”; but the Natal

314 ERNEST WARREN.

hydroid grows plentifully on the sea-weeds of the rock-pools.
It is also stated to be the rule for the lowest internode to bear
only a single part of hydrothecz ; in the Natal hydroid this
is the exception, as there are generally two pairs.

The presence of the characteristic pegs or ribs of perisare
extending into the hydrorhiza is not mentioned by Nutting,
and it is possible that the Natal hydroid is peculiar in possess-
ing them; and should such be the case the character is sufficient
for founding a variety or even a new species.

(21) Thuiaria tubuliformis (Marktanner-Turneretscher).

DYNAMENA TUBULIFORMIS Marktanner-Turneretscher. *‘ Hydroid desk.
k. natur. H. of Museums,’ 1890, p. 238.

THUIARIA TUBULIFORMIS (Marktanner-Turneretscher). Nutting, C. C.,
‘American Hydroids,’ pt. ii, ‘“ Sertu-
lariide,” 1904, p. 70.

This hydroid is one of the commonest on the Natal coast.
It occurs on worm-tubes, and also attached to the vertical
sides of rocks which are left bare during low-tide. Very
probably, in deeper and quieter water, it has a more luxuriant
habit; but in such positions as it has been found it rarely
attains to more than an inch in height. It is of a pale
yellowish-brown colour. ‘The reproductive bodies are only
rarely found.

The identification of the hydroid with T. tubuliformis
has been made from Nutting’s description.

TropHosomg. — Hydrorhiza, creeping stolon irregularly
branched.

Diameter 0°263 mm., thickness of perisarc 18 pu.

Hydrocaulus.—Stems reach about I inch in height,
ziz-zag, bearing alternating branches, divided into regular
internodes, each bearing a branch and two hydrothece on
one side, and a single hydrotheca on the other (text-fig. 12, C).
Branches are divided into irregular internodes, and are con-
stricted at their origins.

A COLLECTION OF HYDROIDS. 315

Diameter of main stem below the first branch 0:401 mm.,

thickness of perisarc 38 ». Diameter of a branch about in

the middle of its length, and measured just below a hydro-
thecze and at the middle of an internode, 0°282 mi.

Hydrothece.— Sub-opposite, long, tubular, with the

ereater part of their adcauline edge parallel to the branch,

and sunk in it, the upper portion being abruptly bent outwards
and ending in two large, opposite, lateral teeth (text-fig. 12, B).

Taxm-nre, 12:

Thuiaria tubuliformis (M.—Twurneretscher).

Operculum two-valved. There is a distinct tendency for the
hydrothecee to arrange themselves in groups resembling those
seen in Pasythea.

Length, measured parallel to abcauline edge, 0°667 mm.,
greatest breadth 0-182 mm.

Hydranth.—Tentacles about 19. Nematocysts at the tips
of tentacles 5°2 « in length, and 1:2 « in breadth.

GonosomME.—Gonangia originate from main stem or branches
immediately below a hydrotheca. Ovate, with a constricted
curved neck and round terminal aperture (B, 2).

Greatest length 1°550 mm., and greatest breadth 0°646 mm.

vot. 1, part 3. 923

sm me) ERNEST WARREN.

Disrrigution.—There has been detected no character by
which this hydroid can be distinguished from T. tubuli-
formis, described by Nutting. The distribution given is—
“ Dschidda (Dr. Billitzer) ; Bay of Bahia, Brazil (Rathbun) ;
Florida, between Salt Pond and Stock Island (Dr. E. Palmer) ;
Bahama Banks, 3-6 fathoms (Nutting).”
not appear to have been described from elsewhere, and its

The hydroid does

occurrence on the coast of Natal gives to it a rather remark-
able distribution.

(22) Plumularia tenuis sp. n.

This delicate hydroid is common on the Natal coast. It is
exceedingly transparent and inconspicuous during life. It
grows on sea-weeds near the low-water line, and attains a
height of about } inch.

TropHosome.—H y drorhiza formsa reticulum witha strong
tendency for branches to come off at mght angles from the
parent stolon. The perisarc may have an outer diffuse layer,
nearly double the thickness of the inner compact layer.

Diameter of hydrorhiza about 0°162 mm.; thickness of
perisarc, outer diffuse layer 13°31, inner layer 8°9 pu.

Trrecularly placed pegs of various shapes project into the
cavity of the hydrorhiza from the vertical sides.

Hydrocaulus.—Stems divided by transverse joints into
internodes, each of which bears a pinna on a process from its
distal end. Pinne alternate; proximal internode very short,
bearing no structures, the rest of the pmna divided by oblique
nodes into hydrothecate internodes alternating with inter-
nodes bearing a nematophore only. Above and below the
nodes there is an internal annulation.

Nematophores rather large, supra-calycine pair overtopping
the hydrotheca, a mesial nematophore at the base of each
hydrotheca, one at the proximal end of the intermediate
internodes, one in the axil of the pinna and stem, and one at
the basal end of each internode of main stem. Nematocysts

of the nematophores, length 6°6 u, breadth 2°5 pu.

A COLLECTION OF HYDROIDS. on]

Diameter of stem, about at the middle and in the centre of
an internode, 0:116 mm., thickness of perisare 17 4. Diameter
at the middle of a pinna 0:042 mm.

Hydrotheca.—Cup-shaped, shallow, adnate by the whole
adcauline surface.

Height 0°057 mm., breath 0°101 mm.

Hy dranth.—18-20 tentacles. Nematocysts occur especially
at the tips of the tentacles, 2°34 in length, and 0-7 in
breadth.

TExtT-FI4. 13.

Plumularia tenuis sp. n.

Gonosome.—The gonangia arise from the main stem, at the
distal ends of the internodes just below the origins of the
pinne. The male gonangia are very elongated, and when
mature are generally ridged deeply and irregularly about the
middle. The blastostyle does not form definite gonophores,
but the spermatic tissue extends along its whole length (text-
fig. 13, C, bi.).

Length of mature male gonangium about 0°827 mm.,
breadth 0:186 mm.

The female gonangia are more ovate ; when immature they

318 ERNEST WARREN.

are provided with a sharply defined distal ridge and a flat
or slightly convex upper surface (op.). The mature female
gvonangium is more elongated, and is irregularly ridged about
the middle. As in the case of the male gonangium the
blastostyle does not produce well-defined gonophores, but the
ova may be seen in a cluster on one side (text-fig. 13, B, b/.).
Ultimately the eggs or young planule are extruded into a
kind of marsupial case, which appears to be formed by a
secretion produced by the blastostyle at the base of the
depression containing the egg-cluster (b). I am not aware
that such a structure has been previously recorded among
the Plumulariide.

Systematic Posrrron.—This species is certainly very near to
Plumularia setacea (Hillis) of Europe and America. The
gonangia, however, appear to be sufficiently distinct to
warrant the formation of a new species. The male gonangium
in P. tenuis is much larger than the female, while the
reverse is stated to be the case in setacea, also the occur-
rence of the marsupium-like structure in the female is
characteristic.

(23) Antennella natalensis sp. n.

This exceedingly delicate hydroid grows on sea-weeds and
worm-tubes. It does not appear to be very common on the
Natal coast.

TropHosome.—H ydrorhiza, irregular reticulum, diameter
0°09 mm., thickness of perisare 15 m1.

Hydrocaulus.—There is no main stem, the “pinne” or
hydrocladia springing directly from the creeping stolon. The
hydrocladia are regularly divided into internodes, alternate
ones carrying hydrothece. The node immediately below a
hydrotheca is exceedingly oblique, that above is transverse.
Nematophores ; a mesial one below each hydrotheca, a median
nematophore with very thin-walled sarcotheca immediately
above each hydrotheca, two supra-calycine nematophores

A COLLEUTION OF HYDROIDS. 319

borne on long processes from the internode, and two on each
of the intervening internodes (text-fig. 14, C).

Nematocysts of the nematophores measure about 10°5 «4 in
length, and 4°44 in breadth.

Diameter about at the middle of hydrocaulus 0:068 mm.,
thickness of perisarc 5 mu.

Hy drotheca.—Cup-shaped, adnate to the hydrocaulus for

TrxtT-FIG. 14.

Antennella natalensis sp. n.

about one-third of the adcauline edge. Mouth facing out-
wards, its plane cutting the hydrocaulus at an angle of
about 30°.

Height of hydrotheca about 0°177 mm., diameter of mouth
about 0°253.

Hydranth.—About 18 tentacles ; nematocysts very small;
they measure about 2°4 4 in length and 0°8 « in breadth.

320 ERNEST WARREN.

Gonosome.—Both male and female gonangia may occur on
the same stem.

Male gonangia ovate, spring from the internode just on one
side of the mesial nematophore. The gonangium has a short
jointed stalk or peduncle which is provided with two lateral
nematophores (text-fig. 14, C, ¢).

The average size is 0°384 mm. in leneth and 0°253 mm. in
breadth.

In the same figure a larger gonangium is seen, which is
female. It would appear that the blastostyle has produced
a second gonophore, since the operculum has opened, and a
large, round, terminal aperture is present, through which, it is
assumed, the products of the first gonophore have escaped.

Systematic Posrrion, — This hydroid closely resembles
Antennella gracilis Allman! and P. catharina Johnston,
and especially the hydroid regarded by Hincks as a stemless
variety of catharina. It agrees in having male and female
gonangia on the same stem. It is, however, markedly different
from catharina in the extreme obliquity of the nodes. With-
out an examination of a number of specimens of catharina
from various localities, it is not possible to judge whether such
a character should, or should not, be regarded as of specific
value.

(24) Plumularia spinulosa Bale.

PLUMULARIA SPINULOSA, Bale. ‘Cat. of the Australian Hydroid
Zoophytes,’ 1884, p. 139.

This minute species has been found on one or two occasions
at Park Ryne. It grows on sea-weeds or larger hydroids.
About one-sixth of an inch in height.

TropHosome.—Hydrorhiza reticular; numerous pegs of
perisarc project into the cavity from the vertical sides as in
Pasythea.

' Nutting, C. C., ‘ American Hydroids,’ Part I, ‘* Plumularide,’ 1900,
»
joenthee

A COLLECTION OF HYDROIDS. Bal

Diameter 0-081 mm.; thickness of perisare 6 ju.

Hydrocaulus.—Slender transparent stem divided into
short internodes, each carrying a pina on a process near
the middle.

Pinne alternate, bearmg a single hydrotheca; there are
two internodes, a short proximal one, and a longer distal one
carrying the hydrotheca and projecting beyond it as a sharp
spine.

Diameter of stem, about at the middle, 0°040 mm.; thick-
ness of perisare 9°6 4.

Hydrotheca.—Rounded at base, much compressed later-
ally, aperture at nght angles to the pina, margin somewhat
everted. An intra-thecal ridge just below the aperture curves
forwards and downwards nearly to the base of the hydrotheca.
Nematophores bi-thalamic with slender bases, one below each
hydrotheca, one at each side above it, one in each axil, and
one at the lower part of each stem-internode.

Width, measured from aperture along a line parallel to the
pinna, 0°115 mm. ; height, measured at mght angles to width
measurement, 0°137 mm. Nematocysts of the nematophores,
length 6:2 w; breadth 1°5 nu.

Hydranth.—About 13 tentacles; nematocysts not dis-
tinguished.

(;onosoMES.—Gonangia unknown.

Bale’s hydroid was found at Queen’s Chiff, 8. Australia,
and the Natal hydroid agrees in all respects with his
description.

(25) Kirchenpaueria mirabilis (Allman).

DIPLOCHEILUS MIRABILIS, Allman. ‘Challenger Reports.’

KIRCHENPAUERIA MIRABILIS (Allman). Bale, W. M., “ Further
Notes on Australian Hy-
droids,”’ ‘Proc. R.S. Victoria,’
1893, p. 109.

pe ERNEST WARREN.

This interesting hydroid, which in the juxtaposition of the
hydrothecee and in the absence of intervening internodes
approaches the Statoplea, is quite common on the Natal
coast, growing on sea-weeds and worm-tubes. It has not,
however, been found in the reproductive stage.

TropHosome.—Hydrorhiza, creeping stolon irregularly
branched.

Diameter 0°303 mm.; thickness of perisare 26 mu.

Hydrocaulus.—Pinnate stems ; main stem is divided into
internodes by oblique joints. Hach internode carries, as arule,
two pinne on projecting processes, one at the distal and one
at the proximal end. Pinne alternate ; consist of short inter-
nodes with oblique nodes; each internode bears a hydrotheca.
Nematophores, one median above and one median below each
hydrotheca, and one on the main stem in the axil of the pinna
(text-fig. 15, B), and generally one on the front face of the
internodes of the main stem at the distal ends in the middle
line. The sarcotheca is very rudimentary. The sarcostyle above
the hydrotheca is lodged in a depression with a delicate film
of perisarc on each side; below the hydrotheca it hes im a
kind of trough of perisare (figs. B and C).

Nematocysts, length 15°4 yu, breadth 6°8 wu.

Diameter, about at the middle of main stem, 0°225 mm. ;
thickness of perisarc 18 u.

Hydrotheca.—Distal portion curved towards the main
stem; mouth circular, its plane cutting the pinna at an angle
of about 30°. Abcauline wall deeply inflected, forming an
intra-thecal ridge which extends half way across the cavity of
the cell. This ridge is not completely filled m with solid
perisarc as Bale describes in his account of the Australian
specimen (C).

Diameter of mouth, measured in the plane of the long axis
of the pinna, 0:222 mm.; height, measured at right angles to
the plane of the mouth, 0°212 mm.

Hydranth.—About 18 tentacles.

Gonosome.—Not found, but described by Bale for the
Australian specimens in the following terms: “Gonangia

A COLLECTION OF HYDROIDS. 323

large, free, with rounded summit, and irregular, wide, trans-
verse undulations; no digtinct marginal ring or operculum ;
sporosacs, two.”

Sysrematrc Posrrron.—In the unjointed condition of the
sarcothece, and in the absence of intervening internodes
without hydrothece, the present hydroid diverges widely from
the typical Eleutheroplea, and shows affinities with the
Statoplea.

Text-FiG. 15.

de ae
oe

=

AU a
Sth

) 3

a= Dee a eS TR oe rr

Kirchenpaueria mirabilis Allman.

The Natal hydroid can scarcely be separated from Allman’s
Kirchenpaueria mirabilis, although the inflected wall of
the hydrotheca leaves a sinus which is not filled up with
homogeneous perisarc. In the absence of the gonophore,
however, this determination cannot be more than provisional.

(26) Paragattya intermedia g. e. sp. n. (Pl. XLVI,
fie. 27.)

This remarkable hydroid is somewhat common on the Natal

coast, and it occurs on coralline and other sea-weeds. It is

324 ERNEST WARREN.

brown in colour and reaches a height of about one-quarter of
an inch. The hydroid is interesting in a number of ways; it
exhibits characters strikingly intermediate between those of
the Eleutheroplea and the Statoplea, also, the same
stem frequently bears both male and female gonangia, and
when such is the case the male appear to be invariably placed
at the distal end of the series.

TropHosome.—The hydrorhiza may form a very close
meshwork ; sometimes it is so close that there are practically
no meshes and the whole constitutes a sheet composed of vermi-
form tubes more or less coalesced.

Average diameter 0°127 mm., thickness of perisarc 14 yu.

The hydrorhiza is frequently found attached to the same
species of coralline alga as that which is so frequently
attacked by a hydroid about to be described (Aglaophenia
parasitica). This latter species sends suckers into the sub-
stance of the alga; but such is not the case with Paragattya
intermedia, which, nevertheless, can fasten itself firmly to
the smooth, hard surface of the coralline.

Hydrocaulus.—Pinnate stems spring from the hydrorhiza.
The main stem consists of 2-5 basal joimts, and above them it
is divided into regular internodes by oblique nodes, and each
bearsa hydrotheca. It is somewhat zig-zag. The pinne are
alternate, each internode of the main stem bearing one lateral
branch, which springs from the process carrying the supra-
calycine nematophores (Pl. XLVI, fig. 27, b.). The two proxi-
mal internodes of a pinna carry no hydrothece ; the remainder,
of which there are generally only two or three, resemble those
of the main stem.

Nematophores; one above (text-fig. 16, C) and one below (D)
each hydrotheca in the middle-line, and two lateral or supra-
calycine nematophores (£).

The sarcostyle below the hydrotheca is lodged in a short
trough-hke sarcotheca ; the one above the hydrotheca is con-
tained in a fairly stout cup-shaped sarcotheca. The two
lateral nematophores are set on processes from the distal ends
of the stem-internodes, and they are distinctly jointed at their

ape
co

ef iais'

EA Aga
Nig 0 Ofo Ojo
ey 2) i) 7
~ +
7

ye F)

enrae

ere:

- oe

Fr AR sea eC IC!
Fee
==
c 7 >
a 2
S : z
5 WA paar

(=
x
Pi
oS 9 D> asia) 9 oj
—
2)

A COLLECTION OF HYDROIDS.

TExtT-FIG. 16.

Paragattya intermedia g.e. sp. n.

>
ws)

26 ERNEST WARREN.

place of origin (Pl. XLVII, fig. 27, and text-fig. 16 H., J).
‘The pinne or lateral branches spring from these processes.

Diameter of main stem, about at the middle of its length,
0-091 mm. ; thickness of perisare 17 wu.

Diameter of pinna in the middle 0:065 mm.; thickness of
perisare 10 u. Nematocysts, 10°6 w in length, 2°6 uw in breadth.

Hydrothecxw.—They occur on all the distal internodes of
the main stem, and on all the internodes of the pinnex, except
on the short basal pair; there are no intervening internodes
without hydrothece. The hydrothece are so inserted on
the pinne that they face directly forwards (B), and not
upwards as is generally the case in the Eleutheroplea.

The margin of the mouth is deeply notched or toothed. There
are normally four teeth, a median back and front tooth, and
two lateral. In one or two cases two pairs of lateral teeth
have been seen (Pl. XLVII, fig. 27, 2a); but the condition
should probably be regarded as a freak, as there was no
general tendency throughout the shoot for the development of
six teeth.

This indented margin recalls the Statoplea rather than
the Eleutheroplea.

The hydrothece measure 0°192 mm. in height, and 07181
mm. in width.

Hydranth.—About 14 tentacles.

Gonosome.—The gonangia occur on the main stem; they
arise laterally by the side of the mesial nematophore below
the hydrotheca. A mass of germinal cells may generally be
found in the endoderm just beneath a hydrotheca (text-fig.
165(Crand. D926.)

Female gonangium ovate and provided with a large oper-
culum (C, op.) ; the peduncle is transversely ridged and fairly
long. One or more large eges are carried on the side of the
blastostyle and become pushed into an apical position (text-
fig. 16, C, m.o.), and there segmentation takes place, and large
planule are formed. The planula escapes, settles down and
develops the usual rosette-structure (B, young).

Length 0-510 mm., breadth 0°225 mm.

A COLLECTION OF HYDROIDS. O20

Male: the general shape of the gonangium is the same as
that of the female, it is provided with an apical pore (B, po.)
in place of the large operculum; the spermatic tissue is
carried on one side of the blastostyle without the formation
of a well-defined gonophore.

Length 0°485 mm., breadth 0°223 mm.

Asabove remarked, both male and female gonangia arise on
the same main stem; it is interesting to note that at the distal
end of the stem, where the male gonangia occur, the accumu-
lation of nutritive substances common to the colony must be
in less quantity than lower down, where the female gonangia
are found. It is conjectured that the determining cause of
the observed arrangement of the male and female elements is
the relative nutritive supply at the distal and proximal por-
tions of the stem.

Systematic Postrion.—The present hydroid comes fairly
near to Gattya humilis Allman,' with an unknown locality
(possibly from the Cape). The former resembles the latter in
the toothed margin of the hydrotheca, and in the jointed
condition of the lateral nematophores; it differs from it in
having a definite pinnate series of lateral branches, in the
presence of a median nematophore above the hydrotheca, and
in the general shape of the sarcothece. .

Paragattya intermedia exhibits a remarkable mixture
of characters typical of the Hleutheroplea and Statoplea.

The characters which it shares with the Kleutheroplea
are :

(1) Presence of median nematophore above hydrotheca,
seen alsoin Antennella natalensis, and in the aberrant
Kirchenpaueria mirabilis.

(2) Lateral nematophores carried on processes from the
stem and jointed, bi-thalamic.

(5) Mesial nematophore below hydrotheca not adnate to
hydrotheca.

The characters which it shares with the Statoplea are:

* Allman, G. J., “Description of Australian, Cape, and other
Hydroida,” * Linn. Soc. Journ. Zool.,’ vol. xix, 1885, p. 156.

328 ERNEST WARREN.

(1) Absence of intervening internodes not bearing hydro-
thecz ; seen also in the aberrant Kirchenpaueria.

(2) Hydrothece face towards the front, stead of being
placed along the upper edge of the lateral branches.

(3) Toothed margin of the hydrotheca.

(4) The median nematophores above and below the hydro-
theca are fixed ; seen also in Kirchenpaueria.

(27) Halicornaria segmentata sp. n. (Pl. XULVIII,

s, 33-36.)

Eh
gq J

This hydroid grows in an arborescent manner; it is of a
dark brown or black colour, and reaches a height of about
3 inches. Only one specimen has been found; it was attached
to the surface of the rock at the bottom of a deep pool at
some distance above the low-tide line.

TropHosome.—Hy drorhiza forms a felt-work on the sur-
face of the rock, diameter very variable, average about
0°24 mm., thickness of perisarc about 52 yw.

Hydrocaulus.—Main stem shows a tendency to be
fascicled, especially towards the base, where lateral tubes are
given off, which grow downwards in contact with the parent
stem, and on arriving at the base branch out into an ordinary
hydrorhiza ; irregularly branched with pinnate stems. The
pinnate stems are faintly marked into internodes by indistinct
nodes. Pinne alternate, and sharply divided into internodes
by oblique nodes; on the posterior face of the pinne the
internodes are prolonged forwards into a sharp ridge, which
on side view makes the pinna very obviously segmented
(Pl. XLVIII, fig. 33).

Nematophores: The mesial nematophore long and broad,
adnate to the hydrotheca nearly to the marginal spine
(fig. 35, sp7.). Lateral or supra-calycine nematophores large
and cylindrical. The pinnate stems are provided with cup-
shaped nematophores, somewhat irregularly placed; there
appear to be usually three around the base of the pinna, and

A COLLECTION OF HYDROIDS. 329

there are also a certain number on the stem below the
proximal pinne.

Nematocysts from mesial nematophore 45 in length, and
6°2 uw in breadth.

Diameter of main stem 0°50 mm., diameter of branch
0°263 mm., thickness of perisarc 34, diameter in the region
of the pinne 0152 mm., diameter of pinna, front view,
0-071 mm., side view, 0°121 mm.

Hydrotheca.— Laterally compressed (fig. 34), viewed
from the side ventricose (fig. 33). Margin with a broad angular
lobe on each side, and a curved, spine-like tooth in front
(spi.). Intra-thecal ridge situated low down towards the base.

Viewed from the side, height of hydrotheca 0°202 mm.,
width 0°223 mm. Viewed from the front, diameter of mouth
0182 mm., diameter at the middle 0°151 mm,

Hydranth.—About 12-15 tentacles.

Gonosome.—Not known.

Histology.—The mesial nematophore in longitudinal
vertical section is shown in Pl. XLVIII, fig. 35. A well-
developed battery of nematocysts (6.7.) is present. At the
base there is an aperture in the perisare (figs. 33 and 54, 0.s.)
through which a sarcostyle (s.), without a sarcotheca, pro-
trudes. At the base of this sarcostyle there can generally be
found a large cell (c.g.) filled with very large refringent
globules which have the general appearance of yolk. Similar
cells can also be found in the general tissue of the coenosarc,

The hydroid is characterised by its dark colour. The
colour is due to the presence of cells containing numerous
small globules of dark refringent substance (figs. 55 and 36,
c.p.). These cells occur both in the ectoderm and endoderm
(ect., end.). In the case of 8. loculosa, in addition to the
black form, a colourless condition occurs. It is possible that
such is also the case in H. segmentata.

SysTeMATIC Posrrion,—The present hydroid is undoubtedly
closely alhed to Halicornaria mitrata Allman,! of un-

‘ Allman, G. J., “ Description of Australian, Cape, and other
Hydroida,” *‘ Linn. Soc. Journ. Zool.,’ vol. xix, 1885, p. 153.

of:

330 ERNEST WARREN.

known locality, and it is only with hesitation that I have
separated them. The chief difference is the presence in the
former of a forwardly-directed prolongation of the edge of
the internode at the back of the pinna. This, in side view,
gives the strongly marked segmented appearance to which
the hydroid owes its specific name. Also, in segmentata
the median spine at the front margin of the hydrotheca is
relatively shorter than in mitrata, and the outline of the
margin, although similar, does not appear to be identical in
the two hydroids. The colour of mitrata is not stated.

Without, however, studying a series of specimens, it 1s not
possible to judge with any certainty whether these differences
are of specific value.

(28) Aglaophenia chalarocarpa Allman.

AGLAOPHENIA CHALAROCARPA Allman. ‘Description of Australian,
Cape, and other Hydroida,” ‘Linn.
Soc. Journ. Zool., vol. xix, 1885, p.
150.

This is a very common hydroid and has been found in most
of the localities that have been searched. It grows in clumps
on sea-weeds and sponges, etc., and is of a pale straw colour.
It may grow to the height of about one inch, but sometimes
shoots with corbule do not attain a height of more than one-
quarter of an inch.

It appears to agree with the Cape hydroid described by
Allman.

TropHosome.—H y drorhiza, creeping, irregularly branched,
sometimes forming a loose reticulum. Diameter about 0°182
mm., thickness of perisare 35 wu.

Hydrocaulus.—Simple pinnate shoots with stems of
variable length. Sometimes the portion bearing the pinne is
separated from the stalk by a very conspicuous oblique joint.
The portion bearing the pinne is divided into regular inter-

A COLLECTION OF HYDROIDS. 333)
nodes by transverse joints. Pinne alternate, carried on a
process of the stem-internode placed rather above the middle.
Pinne divided into regular internodes by transverse nodes.

Nematophores: mesial nematophore adnate to about three
fourths of the height of the hydrotheca, and then terminating
ina free portion which does not extend beyond the level of the
hydrotheca margin; lateral nematophores strong, cylindrical,
generally reaching the level of the hydrotheca margin. On
the stem at the base of the pinne there are three cup-shaped
nematophores, one on the process, one above, and one below
it. The stem below the pinnate portion may bear an irregular
number of nematophores.

Nematocysts of mesial nematophore about 15°5 w in length
and 2°5 uw in breadth.

Diameter of stem below pinnate portion 07178 mm., thick-
ness of perisarc 24°2 mw, diameter in the middle of the
pinnate portion 0°152 mm., length of stem-internode 0:257
mm.

Diameter of hydrocaulus of pinna from the side 0-071 mm.,
from the front 0°064 mm.; thickness of perisare 8 4; length
of internode of pinna 0°24] mm.

Hydrotheca.—Rather wide ; margin with nine teeth, four
lateral, one median; intra-thecal ridge distinct, situated at
about one-quarter of the height of the hydrotheca from the
base.

Measured from the side, height 0°174 mm., width 0-122
mm.; from the front, breadth at the middle 0°110 mm., dia-
meter of mouth 0°155 mm.

Hydranth.—Tentacles muricated, about 11 or 12.

Gonosome.—Male corbula short, with about eight pairs of
leaflets which are but shehtly adherent to one another. The
leaflets bear cylindrical nematophores along both edges. The
rachis bears two rows of gonangia; the peduncle carries one
hydrotheca.

Length of mature corbula about 1:92 mm., depth about
0-869 mm., width from above 0°687 mm.

vous lL. PART 3: 24,

i

Bon ERNEST WARREN.

(29) Aglaophenia parasitica sp. n. CPL saya
figs. 28-32.)

This hydroid occurs very abundantly at Scottburgh among
the large masses of red coralline sea-weed. It is of a dark
brown colour, and the pinnate stems reach a height of about
J inch.

TropHosome.—Hydrorhiza is a creeping stolon sparingly
branched, it grows on the surface of the coralline weed, into
which it sends suckers.

Diameter about 0°25 mm., thickness of perisarc 30 pu.

Hydrocaulus.—Simple pinnate shoots; the basal portion
without pinne is short, and proximally is transversely
jointed, while distally there may be two or three oblique

‘joints. The portion bearing pinne is divided into regular
short internodes by transverse or shghtly oblique joints, each
joint bearing a pinna on a short thick process. Pinne
alternate, divided into internodes by transverse nodes.

Nematophores: mesial nematophore long and cylindrical,
adnate for about half its length to the hydrotheca (Pl. XLVI,
fig. 28 m.n.), widely separated from the margin of the hydro-
theca, and approximating to a horizontal position; lateral
nematophores rather short, cylindrical, and extending to about
the level of the margin; cauline nematophores cup-shaped,
three in number, situated at the base of the pinne (text-fig.
i/o)

Nematocysts of mesial nematophore about 12°44. in length
and 3°14 in breadth.

Diameter of stem below pinnate portion 0°25 min., thickness
of perisarc 48; diameter in the middle of the pinnate
portion 0°21 mm., length of stem-internode 0'182 mm.

Diameter of hydrocaulus of pinna from the side 0113 mm.,
from the front 0°081 mm., thickness of perisare 16 4, length
of internode of pinna 0°212 mm.

Hydrotheca.—Distal portion and mouth wide, proximal
portion rather narrow. Margin provided with thirteen teeth,
one median and six lateral; the third and fifth point inwards,
and the second and fourth point outwards (fig. 29); the

A COLLECTION OF HYDROIDS. Boe

TrExt-Fic. 17.

i Witty Meh !

ge
O-
X hay Ail
Tin) J faa|\
CQ 2/2

\ [7a Venting

ROA rT

“d Or ( i
{

Shy ( alll iin
Stl 7 Pao Lit) oe
i — u inh
aaa ||
cr, oe
Cc = ( i Ta =~
ce <r |
aN eutt HN yy m
Cc ctu ti _
SUCHER c (a
Gavacs|

Hi eA

SS >
oe > ee,
’ ee Re SZ t SSE ye

: a oa sy

SS » . = 4 5 *y =
: ,
. :

;

Aglaophenia parasitica sp. u.

304 ERNEST WARREN.

median one is nearly vertical. The hydrotheca possesses a
very long conspicuous spine (figs. 28, 29, spz.), which is curved
upwards. The intra-thecal ridge is not very strongly
developed; it occurs rather below the middle of the hydro-
theca.

Measured from the side, height 0°226 mm., breadth
0-166 mm.; from the front, breadth at the middle, 0°125 mm. ;
diameter of mouth 0°253 mm. Length of spine (7.e. length
of the arc) 0°166 mm.

Hydranth.—10-12 tentacles.

GonosomE.—Male and female corbule occur on different
shoots and probably on different colonies; they differ con-
siderably in shape, the male being much longer and thinner.
The female corbula consists of about nine pairs of leaflets
(fig. 31), which fuse together into a continuous sheet by their
inner margins. ‘The outer margins are provided with a close-
set row of cylindrical nematophores. From the base of each
leaflet there arises a lateral leaflet with a row of nematophores
along each edge. The rhachis is divided into regular inter-
nodes. The gonangia are arranged in a single row.

In the male corbula there may be 13-15 leaflets (fig. 30) ;
the lateral leaflets are relatively shorter than in the female,
and almost disappear at the distal end.

Looked at from above the leaflets of the male corbula are
less closely united together by a web than in the case of the
female corbula; and this observation agrees with that of
Torrey and Martin,! to the effect that a sexual dimorphism is
exhibited in Aglaophenia by the male corbula being more
or less open above, owing to the incomplete fusion of con-
tiguous leaflets at their tips.

The peduncles of the corbule carry two hydrothece.

Length of female corbula 3-4 mm., depth 1:11 mm., width
from above 0°82 mm.

Length of male 4-5 mm., depth 0°75 mm., width from above

0:71 mm.

' Torrey, H. B., and Martin, Ann, “Sexual Dimorphism in Aglao-
phenia,” University of California Publications, ‘ Zoology,’ vol. iii, No.
4, 1906, p. 50.

A COLLECTION OF HYDROIDS. 335

Systematic Posrrron.—The present species is rather close to
Aglaophenia gracillima Fewkes, from Martinique (96
fathoms), and described and figured by Professor Nutting."
The general character of the hydrotheca strongly resembles
that of parasitica; also, the corbula is described as being
very long, and having expanded truncated processes springing
from the base of the leaflets with a row of nematophores
situated on the top. These “truncated processes”? are un-
doubtedly homologous with the lateral leaflets above described
in parasitica.

The parasitic habit.—This species is especially interest-
ing on account of its close association with the coralline sea-
weed to which it is attached. The hydrorhiza creeps on the
surface of the alga (text-fig. 17, C), and at some little distance
behind the growing point of the stolon, sucker-like outgrowths
are produced, which grow down into the tissue of the alga.
These suckers are shown in section in fig. D, and also in
Pl. XLVIII, fig. 32. The sucker is composed of a modified
clump of ectoderm cells, which have become exceedingly
elongated. The perisare of the hydrorhiza is divided into an
outer (p,) and inner layer (p.), and neither of these layers
are generally continued down for any distance into the pit
formed by the sucker, although occasionally such may be the
case. Usually there are two rows of suckers on the under
surface of the hydrorhiza (fig. C).

In fig. 32 of the plate it can be seen that the elongated,
modified, ectoderm cells of the sucker are very granular, and
they are shown perforating (p.c.) and entering the cells of the
alea. Generally the sucker stretches down through the cortex
(C) to the medulla (M) of the alga, and here the sucker-cells
appear to retain a direct contact with the lining tissue of the
aloa without a layer of perisarc being formed. It is extremely
probable that the hydroid extracts substances of use to it from
the alga, and that the numerous suckers are not merely for

1 Nutting, C. C., ‘ American Hydroids,’ Part I, “ The Plumulariide,”
1900, p. 103.

336 ERNEST WARREN.

the purpose of firmly fastenmg the hydroid to a support. If
the suckers were for the sake of anchoring the hydroid, the
purpose would be much more effectively acccomplished by the
pits being lined by perisarc, but as a rule there is no perisare
towards the base of the pit.

In the immediate neighbourhood of the sucker, both in the
ectoderm and endoderm, large rounded cells (fig. 32, y.c.) filled
with yolk-lke globules are found. It is possible that these
represent the nutritive substances obtained from the alga.
The tissue of the alga is very hard, being thoroughly impreg-
nated with calcium carbonate, but the sucker can, in some way,
eat into it. The tissue of the alga around the suckers is
curiously stained for some depth, indicating that some sub-
stances are secreted by the hydroid and passed into the plant ;
perhaps some acid for dissolving the calcium carbonate is
formed.

The cross-section of the ccenosarc shows an endodermal
tube (c@.), above which is a crumpled layer of ectoderm (ect.),
from the edges of which a sheet of ectoderm passes round im
contact with the perisare (e.s.) and leaves a distinct cavity
(ec.c.). In the spaces below the crumpled ectoderm (ect.)
numerous loose granules (gr.) may often be seen.

The response of the plant to the stimulus, due to the pene-
tration of the suckers, is an attempt to envelop the hydro-
rhiza and pinnate stems; the cortex of the alga grows up
around them (text-fig. 17, C and D, e.C.), and may ultimately
completely surround the hydrorhiza, as at o.h.

It would appear that the complete envelopment of the
hydrorhiza by the tissues of the alga acts prejudiciously on
the coenosarc, since the ectoderm is found to have dwindled
almost entirely away, and the endoderm tube is reduced to
the smallest dimensions (vide fig. D, o.h.).

There is thus a struggle between the plant and the hydroid ;
the host attempts to smother the parasite by growing over it,
and the spread of the hydroid must be considerably checked
thereby.

A COLLECTION OF HYDROIDS.

Os
(a)
NI

(30) Campanularia tincta Hincks.
CAMPANULARIA TINCTA Hincks. ‘ Ann. Nat. Hist.’ April, 1861; Bale,

W. M., ‘Cat. of the Australian
Hydroid Zoophytes,’ 1884, p. 57.

This hydroid grows on sea-weeds and larger hydroids.
TropHosomeE.—Hydrorhiza forms a reticulum.

TrExtT-FIG. 18.

Diameter, 0°12 mm.; thickness of perisarc, 19 pu.

Hydrocaulus.—Simple upright stems, variable in height,
with single terminal hydrotheca, wavy outline. Small, some-
what compressed spherule immediately below the base of the
hydrotheca with a slhghtly smaller diameter than that of the
stem.

Length of stem varies from about 0-4 mm. to 1°8 mm.;
diameter 0:071 mm., thickness of perisarc 9°7 pu.

Hydrotheca.—Large, tubular, and sometimes shghtly ex-

338 ERNEST WARREN.

panding ; 10-12 crenations around the margin. Diaphragm
close to base.

Length of hydrotheca about 0-47 mm., diameter at mouth
0°30 mm. Distance of upper edge of diaphragm from base
0-035 mm., diameter of opening in diaphragm 0:077 mm.

Hydranth.—About 25 tentacles. Nematocysts on tentacles,
length about 5:2 « and breadth 1°8 pu.

Gonosome.—Gonangia large, very variable in shape, sub-
cordate to elongated and cylindrical; short peduncle.

Upper surface flat or shghtly convex, edge somewhat
everted.

Cylindrical form, length 0°95 mm., breadth 0°42 mm.,
diameter of flat opercular surface 0°52 mm. Subcordate
form, length 0°82 mm., breadth 0°63 mm.

Systematic Posirion.—Compared with the Australan speci-
mens the Natal hydroid tends to be small, and the length of
the hydrotheca is shorter, but taking into account the highly
variable nature of this species there appeared no sufficient
reason to separate it.

(31) Campanularia caliculata Hincks.

CAMPANULARIA CALICULATA Hincks. ‘A History of the British
Hydr. Zoophytes,’ 1868, p.
164; Bale, W. M.,. ° Proc.
Linn. Soc. N. 8S. Wales,’
vol. iii, 1888, p. 755.

Grows on sea-weeds and larger hydroids. As in the case of
C. tincta the Natal specimens are smaller than those described
by Bale from Australia.

TropHosome.—H ydrorhiza forms a reticulum.

Diameter 0°137 mm., thickness of perisare 24m.

Hydrocaulus.—Simple stems of variable length, annulated,
small spherule below the hydrotheca.

Length of stem varies from about O°7 mm. to 2 mm.,
diameter 0°070 min., thickness of perisarc 0°010 mm., diameter
of spherule 0:051 mm.

Hy drotheca.—Cup-shaped and expanding ; perisare much

A COLLECTION OF HYDROIDS. Doe

thickened except along the margin, where an internal layer
projects and is shghtly everted. Diaphragm is separated
somewhat widely from the base.

Length of hydrotheca about 0°251 mm., diameter at mouth
6°239 mm., thickness of perisarc 174. Distance of upper
edge of diaphragm from base 0°049 mm., diameter of opening
in diaphragm 0°036 mm.

Hydranth.—tTentacles 25-28, muricated. Nematocysts,
length about 4, breadth about 1°7 y..

Trxt-FieG. 19.

Q COMDPHORE (th

i

Gonosome.—Gonangium ovate, with a short peduncle and a
wide convex operculum. Male and female similar in shape.
The blastostyle of the female gonangium bears one or two
gonophores (text-fig. 19, 4).

Length of gonangium about 0°77 mm., and width 0°55 mm.

(32) Clytia elongata sp. n.

Only one specimen of this hydroid has been found; it
occurred clinging to a species of Thyroscyphus dredged

340 ERNEST WARREN.

from a depth of forty fathoms in the neighbourhood of Bird
Island, Algoa Bay.

TropHosome.—The hydrorhiza is apparently a creeping
stolon from which simple stems and gonangia arise (text-fig.
20). Inthe specimen the hydrorhiza was not firmly attached
to the Thyroscyphus along its whole length.

Diameter of stolon about 0°10 mm., thickness of perisare
HEM te

TExXT-FIE. 20:

Bxvo.
Q GOWANGIUNN

Clytia elongata sp. n.

Hydrocaulus.—Simple stems of variable height. At their
origin from the stolon they are annulated with 9-12 rings, at
the distal end there are 3-5 rings, and there is a terminal
small spherule which is very variable in size; sometimes it is
exceptionally small (0°05 mm. in diameter).

Height of stem 15-3 mm., diameter 0°092 mm., thickness
of perisarc 8°9 pt.

Hydrotheca.—Long, cylindrical, sometimes expanding ;
margin with about twelve sharp teeth.

A COLLECTION OF HYDROIDS. 341

Length about 1:10 mm., greatest breadth 0°34 mm., height
of teeth about 0-08mm. Distance of upper edge of diaphragm
from base 0°119 mm., diameter of opening in diaphragm
0°017 mm., diameter of diaphragm 0:128 mm.

Hydranth.—About 14-16 tentacles ; nematocysts 4°6 ye in
length and 1-44: in breadth.

GonosomE.—Gonangia long and cylindrical, arising from the
stolon by short peduncles with 2—4 annulations. Opercular
surface flat.

Length of gonangium 1°50 mm., breadth 0°31 mm., diameter
of operculum 0°26 mm., length of peduncle about 0°32 mm.

Free: swimming medusve are undoubtedly formed, and stages
in their development can be observed in the various gonangia
on the colony. It is not, however, possible with the available
material to describe the general character of the medusa.

se

(35) Lafoa scandens Bale.

LAF@A SCANDENS Bale. ‘Proc. Linn. Soc. N. S. Wales,’ vol. i,
1888, p. 758, pl. xiii, figs. 16-19.

This hydroid has been found on two occasions growing on
Thuiaria tubuliformis. The gonangium has not been
seen, but the trophosome of the Natal hydroid does not appear
to be separable from that of scandens from Port Stephens
and Port Jackson, E. Australia, described by Bale.

TropHosomE.—The hydrorhiza is a stolon creeping on the
back of the hydrocaulus of the sertularian. Sometimes the
stolon shows a tendency to be jointed.

Diameter 0°109 mm., thickness of perisarc 3°3 pu.

Hydrocaulus.

As distinct from the creeping stolon, it is
represented merely by the pedicel of the hydrothece. This
pedicel may or may not be jointed into two.

Diameter 0:067 mm., length 0-092 mm.; when jointed,
proximal joint, 0°059 mm., distal jomt 0°054.

Hydrothece.—Tubular, long, straight or shghtly curved,
margin somewhat everted. The hydrothece generally alternate
right and left with the hydrothecz of the sertularian.

342 ERNEST WARREN.

Length 0°48 mm., greatest breadth 0°22 mm. Diaphragm,
diameter 0:075 mm., aperture 0°035 mm.

Hydranth—About 13 tentacles. Nematocysts 4°14 in
leneth, 1:2 4 in breadth.

In the figure given by Bale all the pedicels are jointed,
whereas in the Natal hydroid the general rule is for the
pedicels to be quite unjointed.

Gonosome.—Not found. The description given by Bale for
the Australian hydroid is as follows : “Gonangia about double
the length and diameter of the hydrothece, tapering down-

TExT-FIG. 21.

Lafca scandens Bale. x 40.

wards in the lower half, with more or less distinct transverse
undulations ; margin with three or four shallow emarginations ;
summit of the blastostyle forming a trumpet-shaped expan-
sion; gonophores two, both on the same sides of the blasto-
style.”

(54) Lafcea magna, sp. n.

This hydroid was found on sea-weeds and on larger hydroids.
It occurred on a specimen of Thyroscyphus from Bird
Island, Algoa Bay, on Aglaophenia segmentata from Park
Rynie, also on sea-weeds from other parts of the Natal coast.
The trophosome has some resemblance to that of Lafcea

A COLLECTION OF HYDROIDS. 343

eylindrica von Lendenfeld,' from Bay of Islands, New
Zealand ; but the pedicel is much longer in the Natal hydroid.
TropHosome.—The hydrorhiza is a creeping stolon.
Diameter 0°15 mm., thickness of perisare 12 u.
Hydrocaulus.—Simple annulated stems of variable length
(O°4—1°4 mm.).
Diameter about 0°14 mm., thickness of perisare 18 pu.
Hydrotheca.—Tubular, straight or expanding, strongly

TExtT-FIG. 22.

Lafea magna sp. n.

everted margin. ‘T'wo or three renewed margins may some-
times be seen, marking the same number of regenerations of
the polyp (text-fig. 22, B 1, 2, 3).

Height about 1°5 mm.; diameter, about at the middle,
0°58 mm. ; diameter of mouth, 0°707 mm.; diameter in region
of diaphragm, 0°282 mm. ; diameter of aperture of diaphragm,
0:152 mm.

Hydranth.—About 22 tentacles; nematocysts, 7°04 im
leneth, and 2°34 in breadth.

Gonosome.— Unknown.

1 Lendenfeld, R. von, ‘ Proc. Linn. Soc. N.S. Wales,’ vol. ix, p. 912.

344 ERNEST WARREN.

(35) Thyroscyphus equalis sp.n. (Pl. XLVIII, figs.
38-40.)

This hydroid was dredged from Bird Island, Algoa Bay.

It reaches a height of 4-5 inches. In the preserved condition

it is yellowish white and is very opaque. The gonosome has

not been found.
TEXT-FIG. 23.

on

A(nat. size)

Thyroscyphus equalis sp. x.

TropHosomMeE.—Hydrorhiza unknown since the specimen
was torn from its base.

Hydrocaulus.—Thick woody stem, monosiphonic, zigzag,
divided into regular internodes which bear at their distal ends
alternating hydrothecz on short thick processes (text-fig.
23,B). Lateral branches are given off mght and left in one
plane ; they arise from the internode of the main stem at the
back of a hydrotheca (fig. B, b), and consequently the hydro-
theca comes to lie, not exactly in the axil, but in front of it.

A COLLECTION OF HYDROIDS. 345

The lateral branches resemble the main stem in structure.
Sometimes a joint appears in the middle of an internode
(or Be 7).

Diameter of main stem about 1°3 mm.; diameter of lateral
branch, 0°46 mm.; thickness of perisarc, 0°058 ; length of
internode of branch about 1°8 mm.

Hydrotheca.—Set on a pedicel of 1-3 joints which is
inserted ona blunt process at the distal end of the internode.

Hydrotheca tubular, rather short, expanding, with four
equidistant and low teeth, and a four-flapped operculum, two
of the flaps are adcauline and two abcauline. Diaphragm
massive, situated not far from the base, and excentrically
perforated, the aperture being nearer the abcauline side.

Length of hydrotheca about 1°18 mm. ; breadth at mouth,
0°79 mm. ; diameter at the level of upper edge of diaphragm,
0°28 mm.; diameter of aperture of diaphragm, 0°20 mm.;
length of pedicel, 0°15 mm. ; diameter 0°21 mm.

Hydranth.—About 32 tentacles ; nematocysts on tentacles
‘numerous ; they measure about 7°24 in length and 2°74 in
breadth.

Gonosome.—Unknown; large ova have been seen in the
endoderm at the base of the pedicel in the blunt process of
the internode (fig. 58, O).

Histology.—The hydrotheca is lined by a sheet of ecto-
derm (Pl. XLVIII, fig. 38, e.s.), and on the abcauline and
adcauline sides the sheet is provided with two thickenings,
bearing a series of very large nematocysts (B.n.) arranged
horizontally with great regularity. Enlarged views of these
nematocysts are given in figs. 39 and 40; length 30», and
breadth 9. The cnidoblast (cn.) and nucleus (nw.) are
shown. The nematocysts seem to be always placed so that
the ends from which the thread (fig. 38, Th.) is discharged
are pointed towards the perisare cup. It is consequently not
easy to see how they can be used; perhaps the cnidoblasts
pass up to the edge (fig. 38), in a manner recalling the
succession of teeth of sharks, and are only used at the free
edge just below the operculum.

346 ERNEST WARREN.

The hydrotheca is provided with an internal ridge (rd.)
situated at about one third of the height of the hydrotheca
from the base. Along this ridge the hydranth is attached to
the sheet of ectoderm lhning the hydrotheca.

The endoderm of the hydranth can be everted through the
mouth (fig. 58, e.e.) for a considerable distance, thus recalling
the condition seen in Hudendrium angustum.,

Systematic Posrrron.—The present hydroid is allied to
Thyroscyphus simplex Allman,! from Somerset, Cape
York, Torres Strait, in 8-12 fathoms of water, T. ramosus
Allman from Bahia, and Campanularia tor esii Busk?
from Torres Strait. These three species exhibit a certain

bilateral symmetry in that the hydrotheca is distinctly more
ventricose on the adcauline than on the abcauline surface. In
T. equalis, as the specific name is intended to imply, the
two surfaces are practically equal and symmetrical. Also
the stem is much more zig-zag than in the other species.

In the present species, however, a bilateral symmetry is
still evident by the excentric position of the opening of the
diaphragm, it being situated nearer the abceauline side. From
this it would seem probable, either that equalis is reverting
to a typical radial symmetry, or that it represents a stage in
the development of the bilateral symmetry.

i
Geographical Distribution of the Species.

There have now been described 35 species; 15 of these
appear to be new, including two new genera.

Of the 35 species 32 were found on the Natal Coast and 4
were dredged off Algoa Bay, one occurring in both localities.

In the accompanying table the distribution of the species
is shown as far as can be traced with the available hterature.
This list, notwithstanding its extreme incompleteness, exhibits
certain points of interest and significance.

! Allman, G. J., ‘Challenger Reports’ “ Zool..” V.

2 Bale, W. M., ‘Cat. of the Australian Hydroid Zoophytes,’ 1884, p. 52.

A COLLECTION OF HYDROIDS. 54.

Species.

Australasia.

}
Europe.
America.

Natal.
Cape

Parawrightia robusta Warren
EKudendrium parvum sp. n. : :
augustum sp. iv. ; . xe
Clavatella multitentaculata sp. i.
Tubularia solitaria Warren
betheris sp. n.
Pennaria australis Bale
Cladocoryne floccosa Rotch
Asyncoryne ryniensis g. €. sp. n.
Coryne pusilla Girtner ;
Sertularella polyzonias (Lin.)
fusiformis Hzncks
tumida sp. n.
campanulata sp. 1
acanthostoma Bale
operculata Linn.
loculosa way
linealis sp.
bidens ee
Pasythea quadridentata (Ellis and Sol. )
Thuiaria tubuliformis (M— Turner-
etscher) ,
Plumularia tenuis sp. i.
spinulosa ee
Antennella natalensis Sp.
Kirchenpaueria rabies nan
Paragattya intermedia g. e. sp. n.
Aglaophenia chalarocarpa Allman
93 parasitica sp. n.
Halicornaria segmentata Allman
Campanularia tincta Hincks .
5s caliculata Hineks :
Clytia elongata sp. n. ; ; : ; x
Lafcea scandens Bale.
» magna sp.n. : ; : x
Thyroscy phus equalissp.n. . : : x

x X

PI OX OOO XK XXX KK EX
x

x KKK KK KK KK XK

x X

! Hincks, T., ‘A History of the British Hydroid Zoophytes,’ 1868,
p. 264, * South Africa (Busk).”
? Bird Island, Algoa Bay.

vou. 1, PART 3. 2S)

348 ERNEST WARREN.

Out of the 32 Natal species 13 occur in Australia or New
Zealand, 6 in the Cape, 6 in Europe, and 6 in America.

It is noteworthy that more Cape species have not been
found in Natal. As far as I am aware no hydroid faunistic
lists from the Cape coast have been published, but a consider-
able number of species have been described, and yet out of
the 32 Natal hydroids collected only 6 Cape species have been
found.

There is an obvious affinity between the Natal and Australian
hydroid faune, and the lst points to the view that this
affinity may be stronger than that between the Natal and the
Cape faune.

This apparent difference in the hydroid faunz is perhaps
associated with the course of the ocean currents. The warm
south equatorial current flows westwards, and a part of it is
deflected southwards as the Natal or Mozambique Current.
This flows along the east coast of the Cape until it meets the
cold Antarctic Current at the Agulhas Bank, when the greater
part of it turns eastwards and flows back to Australia and
then passes northwards up the west coast of the continent to
be again caught up in the south Equatorial Current.

It is hence readily understood why there should be a marked
difference between the marine fauna of the Natal coast and
that of the Cape coast, especially the part west of the Angulhas
Bank, since here the influence of the cold Antarctic Current
is more strongly felt.

IEE
Note on the Nematocysts.

After the measurements of the nematocysts in the various
species were made, it was felt that there was some relationship
in the shape of the nematocyst and the family to which the
species belonged; in other words, that species of the same
family tended to have nematocysts of similar shape.

In the accompanying table the second column gives the
approximate length of the hydranth; the third, fifth, and
seventh columns give respectively the length (1) and breadth

A COLLECTION OF HYDROIDS. 349

| a
a Small nemato- Nematocysts in
al coveted Large nematophores |
| OS tentacles. nematocysts. or ectodermal
Species. ie sheet.
HD | L L i
Bg hands 7 100' Land B B 100| T, and B = 100
|
| im, | fe im mn
Parawrightiarobusta . 0:9 |50x29) 58
-Eudendrium parvum 05 48x21 | 44 ( |
angustum . 0°5 | 15022) 44: |238°3«10°4) 44
RG layat alls multitentacu- |
lata! 08 | 90x40) 44 19-0 11-0 57 |
Tubularia solitar lal. lean G-2xoro SO a ele <a ei |
betheris! 2°5 |O 1x45 | 88 } 83x79 | 95
Pennaria australis 0°83 |1L:2x5-4) 48 |38:7x«17-0} 41
Cladocoryne floccosa O09 G7 <o3-79 T9707) Si
Asyneoryne ryniensis 3:0 {10:0 8:0} 80 | 27°0«19°5) 72 |
Coryne pusilla 1:3 /S:0<5r8)| 60) 16° 75c1 12.76% |

Sertularella polyzonias . 05 62x1'8 295 |
fusiformis .|04|/51xI11| 22 | |
992

campanulata | 0°4|3:0x0°'7| 23 \
| 22°7 x42) 18 |

'Sertulariaacanthostoma 0-4 |
- bidens ; = (L033) | | 66x14) 21 |
~Thuiaria tubuliformis 0-7 |\52x1:2| 24 |
Plum ulariatenuis . .| 02 | 2:3 0°7| 30 | 66 x2°5| 38
spinulosa . | 62x15) 24 |
/Antennella natalensis . 02 /2:4x0'8 33 104% 4°4 42 |
Kirchenpaueria mirabilis 0:2) 13-468 51
'Paragattya intermedia . 0:2) | 10°7 X2°6| 24
Aglaophenia chalaro- | |
| carpa )°2 155 x 2°5) 16 |
/Aglaophenia parasitica 0-2 12°4x3:1| 25 |
Halicornaria segmentata (2 | 45°5 x 6°2| 14 |
~Campanularia tincta .| 05 | 52x18] 3 . |
caliculata | 03 40x17 7| 42 | |
Clytia elongata : ; .|11)46x1-4| 30 | | |
Lafea scandens. Od 471 *1:2| 30 [| |
| ps magna. Bi telah fg OL cool inehes | |
Dy.x oseyphus sequalis | 2) 2K 270 | 37 | 30'2 X93) 31

1 Tn Clavatella and Tubularia there were more than two kinds of nemato-
cysts; the size next to the largest was taken for the third column.

350 ERNESLt WARREN.

(8) measurements of the small nematocysts of the tentacles,
the large nematocysts, and the specialised nematocysts from the
nematophores and ectodermal-sheet lining the hydrotheca.

As an indication of the shape of the nematocyst the length-

breadth index was calculated, = x 100.

The following points can be noticed:

(1) A glance at the three columns of the table will show
that among both the Gymnoblastica and Calypto-
blastica a wide variability occurs, the length-breadth
index ranging from 14-96; but that among the species and
genera of well-defined families, e.g. the Hudendriide,
Tubulariidx, Sertulariide, and Campanulariide
there is a distinct tendency for some uniformity. In the
four species of the Sertulariide the index varies from 22—
29, and the mean is 24; and in the six species of the Cam-
panulariide it varies from 30-42 and the mean is 34, It
may thus be said that in certain families, at any rate, the
nematocysts have a characteristic shape, which may be ex-
pressed by the mean length-breadth mdex of the various
species. As the table stands the family-index for the Sertu-
lariide is 24 and for the Campanularide 54; but, of course,
a much longer series of species is required before these figures
can be in any way accepted.

(2) Another feature to notice is that when nematocysts of
two sizes occur in a hydroid the indices of the two sizes do
not, as a rule, diverge from each other very widely, or in other
words, the shapes of the large and small nematocysts tend to
approach one another. The mean index of the small nemato-
cysts, in the case of eight gymnoblastic hydroids, is 66, while
the mean index in the same hydroids of the large nematocysts
is 69.

(3) The absolute size of the nematocyst tends to be related
to the size of the hydranth.

In 10 species with a hydranth not exceeding 0°5 mm. in
length (mean O'4 mm.), the mean length of the small nemato-
cysts is 4°2 4; and in 12 species with a hydranth exceeding

A COLLECTION OF HYDROIDS. SH |

0-5 mm. in length (mean 1°35 mm.), the mean length of the
nematocysts 1s _7°2 min.

The relationship may be expressed mathematically in
accordance with Professor Pearson’s statistical methods ; the
constants calculated from the 22 species are: Mean leneth
of hydranth 0°89 mm., mean length of nematocyst 5°93 wp.
Standard deviation of hydranths 0°686 mm., standard devia-
tion of nematocysts 1565 ph. Coefficient of correlation 0°777 ;
probable error of coefficient 0°057.

It is probable that a similar correlation would be found
between the size of ordinary tissue cells and the general size
of the hydranth.

(4) The nematocysts of the nematophores and of the ecto-
dermal sheets lining the hydrothece tend to be elongated ; the
highest index seen is 51, and this belongs to the very aberrant
genus Kirchenpaueria. The mean index of the 11 species
given in the table is 28.

In correlation with the development of nematophores and
specialised batteries on the sheets of ectoderm, the formation
of ordinary nematocysts on the tentacles tends to be weak or
almost completely absent.

EXPLANATION OF PLATES XLV—XLVIII.

Illustrating Dr. Ernest Warren’s paper “On a Collection of
g I
Hydroids, mostly from the Natal Coast.”

Fie. 1.—x 100. Side view of hydranth of Eudendrium parvum
sp.n., Showing the base enveloped by a thin layer of perisarc, terminat-
ing at t.p.

Fie, 2.— x 260. Median longitudinal section of hydranth of EB,
parvum, showing general histology.

Fie. 3— x 1100. Upper portion of the basal differentiated ectoderm
of hydranth, showing modified terminal cells (¢.c.).

Fra. 4.—x 75. Male gonophore of E. parvum, with three chambers
and slight terminal knob.

Fre. 5.— x 75. Side view of hydranth of Eudendrium aneustum

352 ERNEST WARREN.

sp. ”., Showing the peculiar condition of the hypostome, and the plug
of digestive endoderm (e.p.).

Fie. 6.—~x 150. Median longitudinal section of hydranth of E.
angustum, showing the general histology, the inverted perisare-cup at
base (p.cp.) and ingested food (7,f.) in endodermal plug.

Fie. 7.—Natural size. Small colony of Clavatella multitenta-
culata sp.n. on sponge (sp.).

Fie. 8.—x 35. Side view of C. multitentaculata, showing the
hydrorhiza and the base of hydrocaulus ettbedded in sponge.

Fie. 9—x 140. Vertical median section of the hydranth of C.
multitentaculata, showing the histological structure and the origin
of the planoblasts (pl.).

Fie. 10.—Natural size. Small colony of Tubularia betheris sp. n.

Fie. 11.—x 25. Side view of hydranth of T. betheris, showing the
semi-erect peduncles or blastostyles bearing clusters of gonophores (go.).

Fie. 12x 70. Median longitudinal section of hydranth of T.
hbetheris, showing basal dilatation (b.d.) with differentiated endoderm,
basal mass of skeletal, “cellular” endoderm (b.M.), and blastostyle ( pd.)
with gonophores.

Fie. 13.—Natural size. Small colony of Asyncoryne ryniensis
sp.n. growing on the surface of a polychet tube.

Fie. 14.—x 35. Piece of colony of Asyn. ryniensis with short
upright hydrocaulus (H.) carried by hydrorhiza (f.). The hydranth
bears short peduncles with probable planoblasts (pl.).

Fiag. 15.—x 70. Median longitudinal section of hydranth of Asyn.
ryniensis, showing hydrocaulus (H.), perisare groove (p.g.), and
secondary endodermal canal (8. #. C.) in hydrorhiza.

Fie. 16.—x 225. Longitudinal median section of moniliform tentacle
with the ectodermal swellings and small nematocysts (s.i.).

Fie. 17.—x 225. Longitudinal median section of capitate tentacle
of Asyn. ryniensis with large (l.7.) and small (s.7.) nematocysts.

Fie. 18—x 45. Stem of Sertularella polyzonias (Lin.) with
infected hydranth (¢c.hy.) and with mature galls (g.) of a species of
Pycnogonum.

Fie. 19.—x 45. Stem of S. polyzonias, with a lateral branch (b.)
springing from the inside of a hydrotheca, a developing gall (d.g.), and
an old empty and broken gall (0.9.).

Fie, 20.— x 150. Median longitudinal section through gall and
hydrotheca of S. polyzonias. It shows a longitudinal dorso-ventral
section of enclosed embryo of Pyecnogonum. The gall is lined by an

A COLLECTION OF HYDROIDS. Se

ectodermal sheet (e.g.), it has a definite operculum (op.), and the modi-
fied hydranth has a clump of tentacles (fe.).

Fic. 21.—x 48. Small piece of colony of Sertularella cam panu-
lata sp. n. creeping on weed.

Fra. 22.— x 220. Median longitudinal section through the hydranth,
hydrocaulus, and hydrorhiza of S.campanulata. The opening in the
diaphragm is excentric, the perisare of the hydrotheca is much thinner
on the side facing the hydrorhiza than on the opposite side, and the
hydrocaulus is set at an angle of about 45° to the hydrorhiza.

Fie. 23.—x 180. Front view of hydrotheca of Sertularia acantho-
stoma Bale, showing four outer teeth and four inner teeth on each
side.

Fie. 24.—x 180. Side view of hydrotheca of S. acanthostoma,
showing the lower peg of chitin (/.p.).

Fire. 25.—x 300. Median longitudinal section of hydrotheca and
hydranth of S. acanthostoma, showing sheet of ectoderm lining
hydrotheea, with median thickening and battery of elongated nemato-
cysts (B.n.) lying on special platform (plt.) of perisare.

Fie. 26— x 850. Median longitudinal section through edge of
hydrotheca and ectoderm sheet of S. acanthostoma. The thickening
of the sheet with the battery of elongated nematocysts (B.n.) and the
outer clothing of coralline sea-weed (C.W.) are well seen.

Fia. 27.—x 200. Side view of piece of stem of Paragattya inter-
media sp.n. The hydrothece usually possess one pair of large lateral
teeth (2, 2), but occasionally there may be two pairs (2, 2; 2a, 2a). The
nematophores should be specially noticed; the jointed lateral nemato-
phores (J.), the median nematophore above the hydrotheca (sup.n.), and
the mesial nematophore (7.”.), not adherent to the hydrotheca, which
lies helow.

Fira. 28.—x 80. Side view of small piece of pinna of Aglaophenia
parasitica sp. n.

Fie. 29.—~x 200. Front view of hydrotheca of A. parasitica,
showing the arrangement of teeth.

Fie. 30—x 30. Side view of male corbula of A. parasitica,
showing the lateral prolongations (/./.) of the leaflets.

Fre. 31.— x 30. Side view of female corbula.

Fie. 32.— x 350. Transverse section through the hydrorhiza of A.
parasitica, showing the sucker penetrating into the coralline sea-weed.
The sucker generally passes through the cortex down to the medulla of
the plant. The sucker is formed of modified ectoderm. In the neigh-
bouring ectoderm there are rounded cells (y.c.) filled with globules of
homogeneous substance.

B04 ERNEST WARREN.

Fic. 35.— x 90. Side view of piece of pinna of Halicornaria seg-
mentata sp. n.

Fic. 34.— x 90. Front view of piece of pinna of H. segmentata,
showing hole (0.8.) at base of mesial nematophore (m.n.) for a sarco-
style.

Fie. 35.— x 500. Median longitudinal section of mesial nematophore
of H. segmentata, showing sarcostyle (S.), battery of nematocysts
(B.nv.), and conspicuously large cell filled with globules (c.9.).

Fie. 36.— x 900. Small piece of body-wall of hydranth of H. seg-
mentata, showing cells filled with dark globules (e.p.). They occur
both in the ectoderm and endoderm.

Fic. 37.—x 900. Small piece of body-wall of the hydranth of the
black variety of Sertularia loculosa Busk, showing cells, which are
often pear-shaped, containing dark brown matter (c.p.). They appear
to be confined to the endoderm.

Fie. 38.— x 90. Median longitudinal section through the hydrotheca
and hydranth of Thyroscyphus equalis sp. n. The diaphragm is
excentrically perforated. The hydrotheca is lined by a sheet of ecto-
derm, which, at the upper region, is provided with two large batteries of
elongated nematocysts (B.i.).

Fie. 39.— x 1000. Elongated discharged nematocyst of T. wqualis
in cnidoblast.

Fia. 40.— x 1000. Undischarged nematocyst.

EXPLANATORY REFERENCES FOR PLATES AND TEXT-FIGURES.

a. Aperture. a. b. Line of section for fig. D. ab. k. Abcauline knob
of perisarc. ad. k. Adcauline knobs of perisare. ap. Appendage. Ar.
Archenteron.

b. Branch. b. bl. Branched blastostyle. b.c. Budding corbula. b.d.
Basal dilatation. b.h. Base of hydranth. b. hy. Branch of hydrorhiza.
b. M. Basal mass of skeletal endoderm. B.n. Battery of nematocysts.
bl. Blastostyle.

C. Cortex. c. Cuticle. c.c. Collar cell. C.D. Dilatation of ccelen-
teron. ¢.e. Common ectoderm. c.g. Cell filled with globules of
yolk (?). c. hy. Contracted hydranth. c. p. Cell with “pigment.” c. t.
Capitate tentacles. C. W. Coralline sea-weed. ch. Chelicerz. cl.
Claw of chelicerve. cn. Cnidoblast. co. Corbula. ca. Ccelenteron.

d. Diaphragm. d. g. Developing
tocyst. d.#. Distal tentacles.

gall. d.n. Discharged nema-

5

e. C. Enveloping cortex. e. ¢. Endoderm canal. ec. c. Ectodermal
cavity. e.e. Evertedendoderm. e.ep. Endodermal epithelium. — e. g.

Dh

A COLLECTION OF HYDROIDS. o00

Eetoderm lining gall. e. hy. Endoderm of hypostome. e. p. Endo-
dermal plug. e. s. Ectodermal sheet. ect. Ectoderm. end. Endoderm.
jf. s. Fascicled stem.

G. Gonangium. g. Gall. g. ¢. Germinal cells. gl. ¢. Glandular
endoderm cell. g. 0. Gonophores. gv. Loose granules. gv. c. Granular
cell.

H. Hydrocaulus. h. Hypostome. Hy. Hydrocaulus.

i. f. Ingested food (Copepod eggs). 7. p. Inner expansion of colen-
teron.

J. Joint of lateral or supra-calycine nematophore. j. Joint.
k,. Adcauline knobs of perisare. ky. Abcauline knob.

1.1. Lateral leaflet of corbula. 1. n. Large nematocyst. /. p. Lower
peg of perisarc.

M. Medulla of alga. m. Mesoderm. m. n. Mesial nematophore
below hydrotheca. m. o. Mature ovum. m. ¢t. Moniliform tentacle.
mo. Mouth.

uv. Nematocyst. nu. Nucleus.

O. Ovum. o. c. Oral cone. o. g. Old gall. O. H. Outgrowth of
hydrocaulus. 0. h. Old hydrorhiza. 0. 1. Older larva of Pycnogonum.
o. 8S. Opening for sarcostyle. o.s. Old stem. op. Operculum. ov. c.
Oval cell.

p. Perisare. p,. Outer layer of perisare. p,. Inner layer of perisare.
p.c. Process from sucker-cell entering cell of alga. p. ep. Perisarc-
cup. p.g. Perisare-groove. p.¢. Proximal tentacle. pd. Peduncle or
blastostyle. ped. Peduncle of hydrotheca. p/. Planoblast. plt. Plat-
form for battery of nematocysts. po. Pore.

Rk. Hydrorhiza. +r. Ridge. +r. e. Reflexed endoderm. J. H.
Rounded-off hydranth. rd. Ridge.

S. Sarcostyle. s.e. Sheet of endoderm. S. H. C. Secondary endo-
derm-canal. s. n. Small nematocyst. sp. Sponge. spz. Spine. — sp. ¢.
Spermatic tissue. St. Stomodzeum. sup.n. Superior median nemato-
phore.

t. Tooth. #. p. Thin perisare. ¢. pr. Termination of perisare in
perisarc-groove. te. tentacles. th. Thickening of ectodermal sheet on
adcauline side. Th. Thread of nematocyst. th. p. Thick perisare,

wu. e. Undifferentiated endoderm.

v. Velum. v.c. Vacuolated endoderm cell.

y. c. Yolk-containing cell. y.e. Young embryo of Pycnogonum.
y. l. Young larva.

vot. 1, PART 3. 26

=
5
s
w.
a!
eS
wo
—
=
=
i

Ho Lith? Londen

Ruth, LithY London

i

a
i)
eS.

iF
=
3

a
a

Howh Litht London

Die a a at
ee ss Ss

fa ; ben a 7 <p
5 E —- E cs a .
camry 2, rT a - A - '
1 7 - ‘ : - "

J $e, in, eae

Ann. Natal G.Mus. Vol. I.

Pray . =

7 / =A 4
OB / €/?

= pee
Z Sy — == TY,

= jf

yy

Warver del OG Ses
Huth Lith® London

INDEX OF

MOLLUSCA.

(vO. I.)

SOUTH AFRICAN MARINE

Synonyms are indicated by italics. New species described in this volume are

distinguished by (sp. n.) being placed after the names of the authors.

PAGE

Aecrilla gracilis H. Adams 50
Ampullarina africana Smith 20
Ancilla albozonata Smith 27
Ancilla angustata Sowerby 28
Ancilla contusa (Reeve) . 27
bulloides (Reeve) 27

hasta (Martens) . . 28
ordinaria Smith (sp.n.) 27

reevei Smith 27

africana

Arca (Scapharca)
Sowerby : :
Argobuccinum (Fusitriton)

murrayi (Smith) ‘ :
Astralium andersoni Smith
(Cyclocantha)
christi Sowerby

Astralium gil-

Astralium gilchristi Sowerby
henicus (Watson)

Bullia ancilleformis Smith (sp.n.) +

trifasciata Smith
Bursa (Bufonaria) —lampas
(Lamarck)

Callhostoma bisculptum Simith
(sp. n.)

54:

PAGE

Calliostoma granoliratum
Sowerby
Calliostoma (Lischkeia) grano-
liratum Sowerby :
Calliostoma (Astele) iridescens
Sowerby .
perfragile Sowerby

| Cancellaria producta Sowerby

Cardita (?) minima Smith
pulcherrima Sowerby

Cardium gilchristi, Sowerby

Carditella laticosta Smith

Cassis pirum Lamarck

Cemoria fastigiata A. Adams

Cerithiopsis chapmaniana Smith

(sp. n.) ;
insignis Smith (sp.
n.) ;

trilineata (Philippi)

| Cerithiwm contractum Sowerby

crumena Bayle

_ Cerithium pingue (A. Adams) .

Cerithium teniatum Sowerby
trilineatum Philippi .
Chiton (Hanleya) sykesi Sowerby

| Chlamys fultoni Sowerby

358

PAGE
Chlamys gilchristi Sowerby 59
humilis Sowerby 59
natalensis Smith (sp.n.) 60
tinctus (Reeve) 59, 60
Clathurella crassilirata Smith 26
Clavatula impages (Adams &
Reeve) 24
taxus Chemm 24
tumida Sowerby 24.
turriplana (Sowerby) 24
Clionella bornii Smith 24
confusa Smith (sp.n.) 23
Clionella impages Ad. S R. 24
Clionella rosaria Reeve 23
Colina pingwis A. Adams 44
Colubraria crebrilirata (Sowerby) 41
Columbella filmer Sowerby 37
leptalea Smith 38
mendicaria Sowerby 35
scripta Lamarck 38
versicolor Sowerby . 37
Cominella lagenaria (Lamarck). 34.
Conus augur Hwass 21
Conus bandanus Hwass . 22
eucoronatus Sowerby 22
geographus Linn. 21
gilchristi Sowerby 22
patens Sowerby 22
piperatus Dillwyn 22
punctatus Gmelin 21
punctatus Hwass . 5 all
queketti Smith (sp.n.) . 22
Coralliophila (?) carduus (Brod.) 38
Crassatella abrupta Sowerby 63
africana Sowerby 63
angulata Sowerby 63
gilchristi Sowerby 64
tenuis Sowerby 64.
Cryptodon investigatoris Smith 67
Crenella striatissima Sowerby 61
Cultellus decipiens Smith 66
Cuspidaria (Cardiomya) forti-
costata Sowerby. 69
(Cardiomya) — gil-
christi Sowerby . 69

INDEX OF SOUTH AFRICAN MARINE MOLLUSCA.

PAGE
Cuspidaria nasuta Sowerby 69
optima Sowerby 69
Cyclostrema (Tubiola) semi-
sculptum Martens 53
Cynisca forticostata Smith 53
Cyprea barclayi Reeve 4.2
| fultoni Sowerby . 42
Dentalium africanum Sowerby . 58
belcheri Sowerby 58
exasperatum
Sowerby 58
inflexum Sowerby 58
novemcostatum
Lamarck 58
plurifissuratum
Sowerby 57
pohtum Linn. 58
Dolabella rumphti Cuv. 21
Dolabella scapula (Martyn) 21
Dolium fimbriatum Sowerby, var
natalensis Smith (var.n.) . 41
Donax erythreensis Bertin 65
madagascariensis Wood. 65
spiculum Reeve 66
Drillia albonodulosa Smith 25
albotessellata Smith
(sp. n.) 26
fossata (Sowerby) 25
(Clavus) lignaria
(Sowerby) 24
nivosa Smith 25
pretermissa Smith 25
scitecostata (Sowerby) . 25
subcontracta Smith 25
thetis Smith 25
Elusa natalensis Smith (sp.n.) 51
Engina mendicaria (Lamarck) . 35
Epidromus crebriliratus Sowerby 41
| Ervilia purpurea Deshayes 66
Ervilia scaliola Issel 66
| Ethalia africana Smith . 53
Euchelus natalensis Smith (sp.
n.) 55

INDEX OF SOUTH AFRICAN MARINE MOLLUSCA.

PAGH
Eulima dilecta Smith 50
distincta Smith . 50
Eulimella minor Smith . 52
nivea Smith 52
Euthria pura Martens 35
Fasciolaria rutila Watson 34
Fusivoluta pyrrhostoma
(Watson) 382
Fusus cineulatus Smith 33
Genotia beleformis (Sowerby) 24.
Glyphis fuscocrenulata Smith
(sp. n.) 56
Glyphostoma siren Smith 26
Hanleya sykesi (Sowerby) 57
Hochstetteria Lmoides Smith 61
velaini Smith 61
Kellia macandrewi Fischer 64:
Latiaxis idoleum Jonas . . 89
tortilis H. & A. Adams 39
Latirus burnupi Smith (sp.n.) 34
Lepton fortidentatus Smith 64.
Leptothyra —_ armillata (A.
Adams) 53
Lima perfecta Smith 59
Limopsis pumilio Smith . 62
Liotia bicarinata Martens 53
Littorina ahenea Reeve 46
angulifera Lamarek 46
intermedia Philippi 46
newcombi Reeve 4.6
Littorina scabra (Linn.) 46
Loripes clausus (Philippi) 67
Lucina despecta Smith 67
valida Smith 67
Macoma africana (Sowerby) 68
inclinata (Sowerby) 68
levior (Sowerby) 68
ordinaria (Sowerby) 68
Mangilia africana Sowerby 27

PAGE

Mangilia (Bucythara) africana
Sowerby
Mangilia alfredi Smith
Marginella angustata (Sowerby )
bipheata Krauss
Marginella chrysea Watson
Marginella corusca Reeve
Marginella crassilabrum Sowerby
Mareinella differens Smith
dulcis Smith
labrosa Redfield
munda Smith
multizonata Krauss
Marginella (Volvarina) multizo-
nata Krauss :
Marginella neglecta Sowerby
pseustes Smith
Marginella pura Smith
reevet Krauss .
(Gibberula) reevei
Krauss
Marginella ros Reeve :
shepstonensis Smith
(sp. n.)
zeyheri Krauss 5
Marginella (Gibberula) xzeyheri
Krauss :
Marginella zonata Wiener
zonata MKiener, var.
bilineata Krauss
Minolia (Nacheroplax) congener
Sowerby
undata Sowerby
Mitra dedala Reeve :
Mitra (Turricula) dedala Reeve
paupercula (Linn )
Mitra picta Reeve d
Mitra punctostriata A. Adams
schreeter1 Dillwyn
simplex Dunker
Mitra tessellata Kiener
Modiola tennerima Smith
Montacuta macandrewi (Fischer)
Mormula rissoina A. Adams
Murex axicornis Lamarck, vay. ?

|

360

38 |

PAGE
Murex carduus Broderip
Nassa analogica Sowerby 36
circumtexta Martens 36
Nassa (Amycla) circumtexta
Martens 36
Nassa desmoulioides ae by 36
pecilosticta Smith 36
Nassa trifasciata A. Adams 35
Natica areolata Recluz 48
decipiens Smith 48
forata Reeve 48
napus Snuth 49
pygmea Philippi. 48
sagraiana dOrbigny var. 49
Neptuneopsis pyrrhostoma
(Watson) 32
Niso interrupta Sowerby 51
Nucula irregularis Sowerby 62
sculpturata Sowerby 62
Nuculana belcheri (Hinds) 62
compta Sowerby 62
gemmulata Sowerby . 62
lamellata Sowerby 63
Ocinebra natalensis Snvith (sp.n.) 388
Oncidium burnupi Collinge 20
peronil Cuvier 20
Oniscia macandrewi Sowerby 41
Oxystele impervia Menke 55
variegata Anton 55
Pecten effulgens Reeve 59
pusio Reeve (nec Linn.) 59
textilis Reeve 59
tinctus Reeve 59
Pedicularia sicula Swainson 42
Petricola robusta Sowerby 65
Petricola typica Jonas 65
Phos roseatus Hinds 35
Pinaxia coronata A. Adams 40
Pinna equilatera Martens 60
afra Sowerby 60
Pinna madida Sowerby (nec Reev 5) 60
Pinna natalensis Smith (sp.n.) 60

INDEX OF SOUTH AFRICAN MARINE MOLLUSCA.

PAGE
Pleurotoma (Genotia) belwxformis
Sowerby 24
(Drillia) fossata
Sowerby 25
(Clavus) lignaria
Sowerby 24
Pleurotoma lobata Sowerby 24
| Plewrotoma (Surcula) lobata
Sowerby 24
(Drillia) — scitecos-
tata Sowerby 25
(Clavatula) turri-
plana Sowerby . 24
| Poromya curta Sowerby . 69
| eilchristi Sowerby 69
granosissima Sowerby. 69
striata Sowerby 70
Pseudolivia ancilla Hanley 35
Puncturella fastigiata(A.dAdams) 57
noachina (Linn.) 57
Purpura castanea Kuster 39
Purpura dubia Krauss 34
fenestrata Blainville 40
Purpura texturata Smith 40
Pyramidella 52
Ranella leucostoma Lamarck, vay.
pecilostoma Martens . 4]
Retusa truncatula (Brugwiere) . 21
Ricinula heptagonalis Reeve 39
Rissoa conspecta Smith . 4.7
perspecta Smith . 47
Rissoina alfredi Smith 47
durbanensis Smith
(sp. n.) . 47
shepstonensis Smith
(sp. n.) 48
| Scala bullata Sowerby 49
durbanensis Smith (sp. n.
49, 50
eborea Smith (sp. n.) 50
tenebrosa Sowerby 49
| Scalaria minor Sowerby 50
| Scaphander punctostriatus
Mighels 21

INDEX OF SOUTH AFRICAN MARINE MOLLUSCA.

PAGE
Semele capensis Smith 68
Sepia burnupi Hoyle . 0
Septa leucostoma (Lamarck) var. 41
Sistrum cancellatum (Quoy &
Gaimard ) 40
Sistrum fenestratum (Blainville) 40
heptagonale (Reeve) 39
Sistrum squamosum Pease var. 40
Solariella congener (Sowerby) 55
infundibulum (Wat-
son) 56
levissima Martens 56
persculpta Sowerby 56
undata (Sowerby) 55
Sylvanocochhs ancilla (Hanley) 35
Tapes corrugatus (Gmelin) 65
Tapes obsoleta Chemnitz 65
Tellimya similis Smith 64.
Tellina (Macoma) africana
Sowerby 68
Tellina analogica Sowerby 67
gilchristi Sowerby 67
Tellina (Macoma) inclinata
Sowerby 68
levior Sowerby 68
ordinarva
Sowerby 68
Tellina regularis Smith . 67
vidalensis Sowerby 68
Terebra suspensa Smith . 21
Theora ovalis Smith 68
Trifora cerea Smith (sp. n.) 43
convexa Snvith 43
fuscescens Smith 43

361

PAGE
Trifora fuscomaculata Smith 43
shepstonensis Smith (sp.
15)! c 43
Triton lampas Lamarck 41
Tritonidea carinifera (Kiister) 39
Tritonidea natalensis Snrith 35
Tritonium (Cryotritonium) mur-
rayt (Smith) . 40
Trophon carduus (Broderip) 38
Turritella bacillum Kiener 46
declivis Adams &
Reeve var. Ab
punctulata Sowerby 46
Turbo (Calcar) henicus Watson 53
Turbonilla candida A, Adams 51
Turbonilla decora Smith 51
gemmula Smith 51
hofmani Angas 51
Turbonilla kraussi Clessin dl
lactea Krauss . ol
Urosalpinx contracta (Reeve) 39
Vanikoro cancellata (Lamarck) 48
Venus (Timoclea) arakana
Nevill 65
(Anaitis) intersculpta
Sowerby 5 (6)
Voluta (Ternivoluta) abyssicola
Adams & Reeve Saal
Voluta (Volutocorbis) abyssicola
Adams & Reeve 31
Voluta biannulata Fabricius 31
schretert Dillwyn 38

—— EE LDL

= "

Soph, whit

GHNERAL INDEX.
(VOL. I.)

Synonyms are indicated by italics.

New species described in this volume are

distinguished by (sp. n.) being placed after the names of the authors.

PAGE
Ablabophis rufulus (Licht.) 228
Ablepharus wahlbergii (A.
Smith) 226
Acanthurus strigosus Benn 246
triostegus (L.) 246
Acanthuride 24.6
Acanthias blainvillu Giinth 248
Actinula of Tubularia solitaria,
formation of 91
Acontias plumbeus Biane. 227
Adiaptomus natalensis Cooper
(ge. sp: 1.) 97
Agamidee 224.
Agama armata Peters 224.
atricollis A. Smith 224.
Aglaophenia chalarocarpa All-
man ; 272, 330
gracillima Fewkes 3385
parasitica Warren
(Spin). 272, 3382
Agriopus spinifer A. Sivith 247
Agulhas Bank 348
Alestes natalensis Blg7. 219
Allman, G. J. 283, 285, 301, 3238,
327, 330
Alona glabra G. O. Sars 180

guttata G. O. Sars, var.

PAGE
Alona parvula Kurz. = 8
Alonella clathratula G. O. Sars 181
Alonopsis C . 181
Amadhlozi= Lamentation Song 260
Amaka= Circlet of scentedballs 161
| Ama-Lala Tribe . 266
Amapohlo, amaqanda = large
beads ; 5 lies}
Amatshoba = cattle tails for
arms and legs 5 Al
Amblyodipsas microphthalma
(Biane.) 230
Amphibia 221
Amphisbenide . 225
Amphisbena violacea Peters 225
| Ancilla sarda Reeve 28
Anelytropide 5 a PPA
Anewillidee : ; 220, 243
Anguilla bengalensis Ham.-
Buch. 220
Anguilla mossambica Peters 220
virescens Peters 220
Anophelina ‘ 5) ililg}
characteristics of
the genera of 115
Anopheles aitkeni James 116
biturcatus Linn. 116

364. GENERAL INDEX.
PAGE PAGE
Anopheles christophersi Theo. 119 | Blennius punctifer Rgn. (sp.n.)
Anopheles lindesayi Giles 116 | 247, 254
maculipennis Meigen 116 | Boodon infernalis Gthr. 228
Antarctic current 348 | lineatus D. & B. . 228
Antenne of Anophelina 123 | Boulenger, G. A. 219, 237
Antennella gracilis Allman 320 | Broom, R. 167
natalensis Warren Broteas falcifer Loven 183
(sp. n.) 271,318 | Brady, G. Stewardson . ly}
Aparallactus capensis A. Smith 230 | Breviceps mossambicus Peters. 222
Apogon warreni Rgn.(sp.n.) 244,251 verrucosus Rapp. . 222
Arthroleptis wahlbergii Broteas lamellatus (G. O. Sars)
A. Snuth 223 173, 183
Astrape capensis (Z.) 242 | Bryant, A.T. . ; » BS
Asyncoryne ryniensis Warren Buccinum cataracta Chemnitz. 35
(g. e. sp. n.) 270, 285 | Bufonide , : . 221
Atherinide 243 | Bufo carens (A. Smith) . 221
Atherina pineuis Lacép. 243 | regularis Reuss. . 227
Atractaspis bibronii A. Smith. 2381 | Burnup, H.C. 19
Atractylis arenosa Alder 196 |
Austen, E. E. 216 | Cacosternum boettgeri (Blgr.) 222
Calamelaps mironi Mocquard . 285
Bale, W. M. 283, 308, ees 310, 321, polylepis Bocage 235
322, 341, 342 | unicolor Reinh. 235
Balistidee 247 | warreni Blg7.(sp.n.)
Balistes aculeatus LL. . » 247 || 230, 234
« Barbel” : : 22373) Callorhynchus antarecticus
Barbus decipiens Blgr. . . 220 Lacép. 242
gibbosus Peters 219 Calamphora parvula Allman 300, 302
Batrachia 221 Campanulariide 272

Bead-work ornaments of ie

Zulus : 159
Belone robusta Gthr.. e243
Bertramia asperospora(Fritsch.) 8

kirkmani Warren
(sp. n.) : 7

kirkmani, structure
and growth of . 8

kirkmani, life his-
tory of 5 le
Bimeria . 189, 196
Bimeriide 269
Bitis arietans (Gray) 230
Blenniidee : : . 247
Blennius bifilum Gthr. 247
cristatus D. . . 254

Campanularia caliculata H nets s
272, 338
tineta Hincks 278, :

torresii Busk 346

Cantharus emarginatus C.g V. 244

Carangidee ‘ 24.6

| Caranx carangus (Bl.) . 24.6

ciliaris (Bl.) 246

hippos (Z.) 246

melampygus C. f V. . 246

rottleri Bl. 2 . 246

speciosus ae 246

Carchariidee 241

Cellia albipes Theobald . 7,
jacobi Hill & Haydon

(sp. n.) . 112, 117, 144

GENERAL

PAGE
Cellia pulcherrima Theobald . 117
squamosa Theobald

112, 117, 142

Cephalopoda. : P40)
Ceriodaphnia natalis Brady

(sp. n.) ‘ ; . 180
Cerithiopsis pulchella C. B.

Adams : . 45
Cerithium tricarinatum Pease. 45
Cestracion 2
Cetshwayo 263
Chetodontide 245
Cheetodon setifer Bl. 245

vagabundus L. 245
Chameleontide . 227
Chameleon damaranus Blgr. 227

dilepis Leach 227
teniobronchus A.

Smith . 227

Chamesaura enea (Wiegm.) 224

anguina (L.) 224.

macrolepis (Cope) 224

Characinidie ‘ ‘ 5 PZ)

Chimeride : . 242

Chlorophis Benloger sos (Gthr.) 228

natalensis
(A. Smith) 228
Chorinemus sancti-petri C. 4 V. 246
Chrysophrys bifasciatus
(Forsk.) 245
gibbiceps C. dV. 245
hasta (Bl. Schn.) 245

sarba (Forsk.) . 245

Chydorus barroisi Richard a | iltsts:
gibsoni Brady (sp.n.) 1838

poppei Richard . 183

Cichlid ; . 220

Cirrhitichthys eculatas
(Lacép.) 244:

Cladocera ‘ 3 5 a)
Cladocorynide . : 5 AT)
Cladocoryne floccosa Rotch 270, 284:
Cladonema ; : . 288
Clarias capensis C. \° V. 220, 237

gariepinus (Burch.) 220, 237

INDEX. 365
PAGE
Clavatellidee , A Paik)

Clavatella ie wititentuclata
Warren (sp. n.) 207, 278

prolifera Hincks 5 Age)

| Clothes, Zulu. : 160, 161

Clupeidee : : . 242

Clupza dorsalis ; : 4
durbanensis Rgn. (sp.

n.) : 4, 242

Clytia elongata Warren (sp. n.)
272, 339

Clypeal hairs in Anophelina 124.
Colubride : : . 228
Colunbella mitriformis A.Adams 38
sagena Reeve Od,
Convoluta roscoffensis Graff . 105
branched
ova of 106
Cooper, Arnold W. : 97, 205
Copepoda : : 97, 183
Copeus spicatus (Hudson) ; 8
Corbule of Ag. parasitica, male
and female . : 334
Correlation of size of ee
and size of nematocyst . 350
Corynidee s 3 . 270
Coryne pusilla Gartner 270, 289
Corymorpha : : 98, 280
Cussonia spicata Thb. . 43{8)
_ Cyphosidee , : . 245
Cyphosus fuscus ( Zacép.) . 245
Cypretta sarsi Brady. 173, 174
Cyprinidee : 2 . 219
Cyprinodontide . 220
Cypria armata G. W. Miiller . 174
Cypris Pe Gheuts Brady (sp.
n)). : ‘ 5 173
Cypridopsis newtoni . 176
punctillata Br adi y
Gp) ss
villosa : = LG
Dactylethride . ; . 221
Dalbergia obovata E.M. = Um-
zungulu thorn-bush 5 ANGO)

366

PAGE
Dasybatide 243
Dasybatis uarnak (For sie ) 242
Dasypeltis scabra (L.) . 229
Dendraspis angusticeps (A.
Smith) . 230
Dentex argyrozona C. 4 V. 244, 2538
Dentex rupestris Casteln. 252

Dentex undulosus Rgn. (sp. n.)

244. 252
Diagramma affine Gthr. 244.
crassispinum Rupp. 244:
vriseum C. §° V. 244,
Diaptomus orientalis G. S.
Brady . 97, 1738, 184
Dicynodon ingens Broom (sp.n.) 168
latifrons 168
leoniceps 168
simocephalus 168
Dingane . = 260
Diplocheilus mirabilis Allman 321
Dispholidus typus (A. Smith) . 280
Distribution of species of Hy-
droids 346
T.tubuliformis 316
Drakensberg Mts. 364
Drepanid : 246
Drepane punctata (L.) . 246
Dynamena tubuliformis Mark-
tanner-Turneretscher 314

Ectodermal Sheet lining Hydro-

theca
Elapechis decosteri (Blgr.)

sundevallii (A. Siith)

Eleotris ophiocephalus C. & V.

Eleutheroplea 323, 32
Elusa aclis (A. Adanis)
Endodermal canals of Asyn-

coryne ryniensis
Engystomatide
Epinephelus andersoni Blgr.
hemistictus
(Riipp.)
maculatus CEE ine

sonnerati(C. §° V.)

301, 304, 311,

247
6, 327
52

345
230
230

.

|
|
|
|

GENERAL INDEX.

Epinephelus tauvina (Forsk.) . 244
| Equula edentula ( Bl.) 245
_ Euchelus foveolatus A. Adams 55
| Eudendriide 270
- Eudendrium angustum Warren
(sp. n.) 270, 276
capillare Alder 272, 276
insigne Hincks 276
parvum Warren
(sp.n.) 270, 272
vaginatum Allman 274
| Eulima solida Sow. 50
Euphorbia grandidens Harv. 259
| Fishes 1, 219, 237, 241
Gadidee : 243
_ Galeichthys feliceps C. § Vi 242
Gall caused by a Pycnogonum. 293
Garvela . 196
Gastropoda : 20
| Gattya humilis Allman 327
Gazania longiscapa D. C. 161
| Geckonidee 223
Genypterus capensis a Smith)
6, 247
| Gerres lineolatus Gthr. 245
longirostris Rupp. 245
Gerrhosauride . 225
Gerrhosaurus flavigularis Wiegm. 226
erandis Blgr.
(sp. n.) 225, 233
major 4. Dum. 234.
validus A. Smith 284
Gibson, James 173

Giles, Colonel

Glauconiidee

Glauconia conjuncta (Je an.)
distanti Blgr.

Gleocapsa

_ Glyphidodon eeictaats Cag Vi.

sordidus (Forsk.)
Gobiide .
Gobius eneo-fuscus Peters

221,

Nw WD FE
On WW eH
or) )

ho bo
ASS
~1 DD

GENERAL INDEX. 367
PAGE | PAGE
Halicornaria mitrata Allman . 329 | Thubo lempi= Specimen of war-
segmentata War- song . - 265
ren (sp. n.) 272,328 | Ihubo Ienieost= Ohio Sssong . 266
Halocordyle australis Bale . 283 Ijuba= Blue beads . - 162

Halocordyle cooperi Warren
73, 209, 283
Halocordyle tiarella (Ayres) 79, 283

capitate tentacles
of : . 209

symmetrical ar-

rangement of

capitate tenta-
cles of 212, 283

Haplochromis desfontainesi

(Lacép.)

22
moftatti(Casteln.) 220 |
22

Haplochilus johnstoni Gthr.
myapose Blgr. (sp.

n.) ; 220, 232

punilus. . 232
Haplosporidia . ; oe pala
Harmer, S: FL. : . 109
Haydon, L. G. . : , dala

Hemidactylus mabouia (Mor.). 223
Hemirhamphus dussumieri

C.F V. 248

Hemisus guttatum (Rapp.) . 222
Herpetosaura arenicola Peters. 227
Hill, Ernest F 5 i100
Homalosoma lutrix (ZL. ) . 229
variegatum Peters 229

Homopholis wahlbergii
(A. Smith) 223

Homorelaps lacteus (L.) - 2380 |
Hydroids 73, 83, 187, 209, 269 |
Hydvroid faune, table of 5 Bully

Ibetshu = hind-portion of umu-
tsha : 161
Ichnotropis capensis (A. Smith) 225
squamulosa Peters. 225

Igama lempi= War-song . 261

Igcagcane = Necklet of small
squares “ : 5 Ally}

Igemfe = Reed-whistle . . 259

Ikehla= Man with head-ring . 160
Inidhleke=Name of regiment 265
Imiqubula = Wardress of ox-
tails : 162
| Imingwambi = Witch- sontone s
strips of skin worn over
shoulders. : > Ali!
Incombo= Yellow beads . 162

Inewadi, ubala abuyise = A Zulu
letter d : . 168
| Indhlamu=Children’s ditty . 263
Indondo = Brass ball worn hang-

| ing from the neck in front 160
| Ingeje=Single bead-string . 168
| Inkankane= Dark blue beads . 163
Inostranzewia . 172
Instrumental Music of ane Pane 257
| Intotoviyane=Striped beads . 163
| Isakabuli = Kafir finch . . 160
Isicawu = Place for wedding
| dance : : . 264
, Isicoco or head-ring . 159
| Isidhlodhlo = Head dress of
finch feathers : . 160
| Isidwaba=Skirt of goat or ox
| skin for married women . 161
| Isigubu= Drum : 259, 262
| Isimekezo = Wedding dance of
| 2nd day P . 262

Isinene = Front - aunties of
tmnutsha : =) LoL

| Isingeniso, umcanguzo, in-
kondhlo Bride’s songs 262, 264

Isitimane = Black beads . 162
Isitontolo = Basuto — stringed

bow . : . 209

Itambo = W hite beads . . 162

Izifociya = Grass belt . . 161

James and Liston . ; INS} alas)

| Julis lunaris (L.) . : . 246

368

PAGE
Julis umbrostigma Riipp. 246 |
Kafir sheep, spiral hoofs of 109
Karroo beds of Natal 167
Karyosome ; 9
Kirchenpaueria mirabilis
(Allman) 271, 321
Kirkman, Thos. . il

Labeo darling: Blgr.
Labride .
Lacertiha
Lacertidee

Lafcea cylindrica

bw we bh by
nN
(on)

bw bo
or Ww

Von Lendenfeld 3483
magna Warren (sp. n.)
272, 342
scandens Bale . 272, 341
Larve of Anophelina, duration
of life of 126
Latirus flavidus A. Adams 34
marie Crosse . B4
Lendenfeld, R. von 288, 343
Lendhlangamandhla = Praise-
name of chiefs 263
Lepidopus caudatus (Euphras. ) 24.6
Leptodira hotambeeia (Laur.) . 229
Lethrinus nebulosus (Forsk.) 244,
Leuconostoc 205, 206
Leydigia acanthocercoides Sars . 181
africana Gurney 181
Leydigia propinqua G. O. Sars 181
Liognathidee 245
Lower Beaufort Beds 172
Lutianus gembra (C. & V.) 244.
johnii ( BI.) 244.
marginatus (C. f V.) 244
Lycophidium capense
(A. Smith) 228
semiannulis
Peters 228
Lygodactylus capensis
(A. Smith) 228

Lystrosaurus

GENERAL INDEX.

PAGE
Mabuia homalocephala (Wiegm.) 226
quinqueteniata
(Licht.) © 226
striata (Peters) 226
varia (Peters) . 226
Macrelaps microlepidotus
(Gthr.) 2380
| Mafunzi tribe . . 265
Majozi = Surname of Chief
Negoza : . 265
Marginella pulchella Ki iener 31
Mayr, Father Franz 159, 257
Merluccius capensis Casteln. 4, 243
vulgaris (Linn.) . 5
Miller, Alan ‘ . 263
Mitra cinnamomea A. Ada ns 33
Mollusea 19, 357
Moor, Shirley 109
Mosquitoes 111
Mugilidee : 243
Muegil ceylonensis Gthr. . 248
constantie C. & V. 243
robustus Gthr. . . 248
smithii Gthr. . 243
Miller, G. W. . : ets
Mulhde . . 245
Mulloides flavolineatus Gren ) 245
Murenesox cinereus (Forsk.) . 243
Murenidee : 243
| Murzena macrurus Blkr. . 243
polyophthalnus Bblkr. 243
Musical Instruments of the
Zulus 257
Muscidae 217
Mustelus vulgaris M. 4 H. 241
Myliobatis aquila (L.) . 24.2
Myxobolus : ; x lb
Myxosporidiun . 15
Myzomyia culicifacies (Giles) 116
elegans (James) 116
funesta Giles TL be
112, 116
funesta Giles var.
subumbrosa Theo-
bald 127

INDEX. 369
PAGE
_ Nyssorhynchus annulipes
Walker 117
fuliginosus Giles 117
karwari (James) 117
maculipalpis
Giles 112, 139
maculatus
Theobald 117
pretoriensis
Theobald 112,189
stephensi
Liston 117
theobaldi Giles 117
Oestridee : 217
Oncocypris voeltzkowi
G. W. Miiller 173,179
Ophidia . : F 227
Ophidiidee : : 247
Ophichthys kirkii Gthr. 243
unicolor Rgn.
(sp. n.) 243, 250
Ostracoda 73
Ostracion cubicus L. 247
Ostraciontidee ; 24.7
Otolithus equidens C. 4 V. 245
Pachydactylus capensis
(A Smith) 223
maculatus Gray 223
Pagellus ithognathus C. 4 V.. 245
mormyrus (L.) 245
Pagrus laniarius C. & V. 244.
Pahnate hairs of Anophelina 125
Pande, a Zulu Kine 265
Pansporoblast 15
Paradiaptomus lamellatus
G. O. Sars 183
Paragattya intermedia
Warren (g.e.sp.n.) 271, 323
Paralichthodes algcensis Gilchr. 243
Parascorpis typus Blkr. 244.
Parasitic habit of Ag. para-
sitica : 335
Paratilapia moffatti Casteln 22

GENERAL
PAGE
Myzomyia funesta Giles var.
umbrosa Theobald 127
listoni Liston
rhodesiensis Theo-
bald 112 |
rossi (Giles) 116 |
turkhudi Liston 116 |
Myzorhynchus barbirostris Van |
der Wulp 117, 152 |
mauritianus |
Grandpré 112 |
natalensis Hill |
and Haydon |
(sp.n.) . 112, 152 |
nigerrimus Giles 117 |
paludis Theobald |
112, 146
sinensis Wiede-
mann 117, 146
Natal or Mozambique Current. 348
Naia haie (Z.) 230
nigricollis Reinh. . 230
Nematocysts of Nematophores. 351
Nematocyst, note on the 348
-measurements,
table of . 349
-index of various
Hydroid families 350
Neobola argentea Pellegr. 231
bottegi Vincig. 231
brevianalis Blgr.
(sp. n.) 220, 231
minuta Blgr. 231
Notommatidee 7
Nucras camerani Bedr. 225
Nucras delalandii (M. Edw.) 225
holubi Stdr. 225
tessellata (A. Smith) 225 |
tessellata (A. Smith), |
var. ornata Gray 225 |
Nutting, C. C., 297, 299, 301, 316, 335
a-o-Nqakamatshe, a regiment |
of Cetshwayo 263 |

370

PAGE

Parawrightia robusta
Warren (g.e. sp.n.) 187, 269,
Pasythea quadridentata

272

(Ellis & Sol.) 271, 312
Paterson, P. 172
Pelecypoda 59
Pempheride 245
Pempheris molucca C. § Wo 245
Pennariide 270
Pennaria adamsia
Von Lendenfeld 288
austrahs Bale 283
australis Bale, var
cooperi Warren
270, 282, 283
cavolinii Ehren. 79, 2838
cibbosa Agassiz 79, 283
symmetrica Clarke 283
Perigonimus : 189, 196
Philothamnus semivariegatus
A. Smith 229
Phonograph-records of Zulu
Songs 267
Phrynobatrachus natalensis A
Smith 222
Plagusia marmorata Bl ee 243
Plasmotomy , Ie
Platophrys pantherinus (Riipp. ) 248
Platycephalide . 247
Platycephalus insidiator
(Forsk.) 247
tentaculatus Riipp. 247

Platyglossus scapularis (Benn.) 246
Platysaurus guttatus A. Smith 224
Pleuronectidee 243
Pleuroxus assimilis Brady Ge

n.) : HR
Pliotrema warreni Reyan (sp.

n.) 1, 241
Plumulariide - 2a
Plumularia catharina Johnston 320

setacea (Ellis) 318
spinulosa Bale 271, 320
tenuis Warren (sp.

n.) 271, 316

GENERAL

INDEX.
PAGE
Pomacentridee 246
Pomadasidee 244.
Pomadasys anas Val. 252
furcatus Bl. Schn. . 252
hasta (Bl.) . 244.
multimaculatus
(Playf.) 244
opercularis (Gthr. ¥
Playf.) 244
teniophorus Regan
(sp. n.) 244, 251
Pomatomus saltator (L.) 244.
Pondos 263
Power, H.S8. : 111
| Pristiophorus M. wu. H. 1
cirratus . 2
| Prosymna ambigua Bocage 229
| jani Bianc 229
Proteocypris (?) alebuleider
Brady (sp.n.) 177
reniformis Brady
(sp. n.) 176
Psammophis crucifer (Daud.) . 229
sibilans (L.) 229
| Psettus falciformis (Lacép.) 245
Pseudaspsis cana (L.) 228

_ Pseudocordylus microlepidotus

(A. Smith) .

| Pseudorhombus russellii (Gr a)

| Pseudosecarus maculosus

(Lacép.)

Pterois miles (Benn.)

| Pyramidella aclis

volitans (L.)
A. Adams
Pyretophorus ardensis Power
112,
cinereus Theobald
112,
costalis Loew 112,
jeyporensis Theo-
bald
marshalli
bald
pitchfordi Power
rial

Theo-

112

112

GENERAL
PAGE |
Pyretophorus superpictus Grassi
112
|
Raiidee : 242
Raia marginata Lacép. . 242

ocellifera Regan (sp.n.) 2,
rhizacanthus Regan (sp. n.)

242

3, 242
Rana adspersa (D. & B.) 222
Rana fasciata A. Smith . 222
mascareniensis D. & B. 222
natalensis (A. Smith) 222
oxyrhynchus A. Smith 222
queketti Blgr. 222
Rappia cinctiventris (Cope) 223
concolor (Hallow) 223
undulata Blgr. . 223
Regan, C. Tate 1, 241
Receptacula seminis of Steno-
cypris is
Reptilia . 167, 219
Rhinobatide 24.2
Rhinobatus blochii JZ. f 16l- 242
columne M. 4 H.. 242
Rhiptoglossa 227
Rissoina bicollaris Serawanty AT
fenestrata Schwartz 47
Rousselet, Chas. 7
Salarias kosiensis Rgn. (sp. n.)
247, 254.
quadricornis C. & V. 24.7
rivulatus Riipp. 247
variolosus C. & TV. 255
Sarcocystis 11
Sarcophaga sp. 215
Sargus capensis A. Smith 244.
cervinus (Lowe) . . 244
holubi Stdr. 244, 253
nigrofasciatus Rgn. (sp.
n.) 244, 253
Sars, G. O. life
Scaphopoda 57
Searide : 246
Scelotes bipes (L. ) : 226
guentheri Blgr. 226

INDEX. 57]
PAGE
Scelotes inornatus (4. Smith) . 226

Schizophyte in Parawrightia
robusta 187, 197,
Schizophyte, capsule of
Schizophyte-capsule, chemical
tests of
Schizophyte, fission of .
spore-formation of
Scienidee
Scizna aquila Lacép.
maregaritifera Haly.
Scincidee
Scombresocidee
Scorpenidee ; : ;
Scorpena haplodactylus Blkr. .
natalensis Rgn. (sp.
me) ie 5,
rosea Day
Scorpididee
Scyliorhinide
Seyliorhinus africanus (L.)
edwardsii (Cuv.) .
natalensis (Regan)
variegatus

(A. Smith)
Scymnosaurus ferox Broom
warreni Broom
(sp. n.)

Selachii : .
Sepedon hemachates (dean: ye
Seppings, J. W. H
Serranide
Sertularella Cacinaralere
Warren (sp. n.)
271,
fusiformis (Hincks)

271, 292, 2
gayi (Lamourouae)
297,
polyzonias (Lin.)
ile

sohtaria Nutting
tumida Warren
(sp. n.)

271

Sertulariide

200
198

204:
201
202
245
245
245
226
243
247
247

247
247
245
241
241
241
241

169
241
230
239
24.4

291

372

PAGE
Sertularia acanthostoma Bale
271, 3808
bidens Bale . 271, 310
Sertularia fusiformis Hincks 295

Sertularia linealis Warren

(sp. n.) 271, 308

loeulosa Busk PAR I
306, 3829
maplestonei Bale 312

operculata Lin. 271, 305
Sertularia quadridentata Ellis

and Sol. : 2 ol?
Sexual dimorphism in Aglao-

phenia 334.
Sillaginidee 245
Sillago chondropus Blkr. 245

sihama (Forsk.) . 245
Siluride . 220, 242
Simocephalus capensis (A.Smith) 228

capensis ? G. O.
Sars. > 7s)

Sistrum concatenatum

(Lamarck) 40
elongatum (Blainville) 40
Smith, Edgar A. 19
Sowerby, G. B. 19
Sparide . 244,
Spermatophores of Wataotarn 98
Spiral hoofs of sheep 109

Spore-formation in Bertramia

kirkmani Warren 12
Spratelloides delicatulus

( Benn.) 242
Squalidee 241
Squalus acutipinnis Ppp.

(sp. n.) 241, 248
Squatinidee : 242
Squatina aculeata Cwv. . . 250

africana Rgn. (sp. n.) 242,

24.8, 250
angelus Dum. 250
armata Philippi 250
australis Regan . 250
californica Ayres 249, 250
japonica Bleek 249

GENERAL INDEX.

PAGE

Squatina nebulosa Regan

250

Statoplea 322, 326, 327

Stauridium ‘ . 288
Stenocypris aldabre G. W.

Miiller 178

chevreuxil 178

malcolmsoni 178

sinuata 178

Suckers of Ag. parasitica 335

Swazis 265

Syncoryne 288

Syncorynide 270

Synaptura ciliata Gilenr. 243

pectoralis Kaup. 248

Synthecium 293

Teleostomi 242

| Tetradactylus africanus ( Gray) 226

Tetrodontide . 247

| Tetrodonimmaculatus Bl. Schn. 247

| honckeni B/. . 247

Teuthididee 246

Teuthis oramin (Bl. Schn: ) 246

Thelotornis kirtlandii (Hallow) 229

_ Therapon servus ( BI.) . 244.

_ Therocephalia 5 Sz
Thuiaria tubuliformis (Mark-

| tanner-Turn. ) 271, 314

Thorax of larvee of Anophelina 125

Thyroscyphus ramosus Allman 346
equalis Warren

(sp. n.) 272, 344

| simplex Allman. 346

Tilapia melanopleura A. Dum.. 221

mossambica (Peters) 221

| natalensis (M. Weber) . 221

Tilapia philander M. Weber 221

| Tilapia sparrmani A. Smith 221

Torpedinidee : 242

Torpedo marmorata Risso 242

Torrey, H. B.,and Martin, Ann 334

Trachurus trachurus (L.) 246

_ Trachynotus ovatus (L.) 246

Trichiuridee 246

Tridacna elongata Lemarch 64

GHNERAL
PAGE |
|
Triglidie é ‘ . 247

Trigla capensis C. & V. . 247
Trimerorhinus rhombeatus (L.) 229

triteniatus
(Gthr.) 229
Tropidonotus levissimus (Gthr.) 228
Tshaka 260
Tubulariide : eo
Tubularia attenuata All man . 282 |
bethe Warren (sp. |
n.) . 270, 280, 374 |
humilis Allman . 282 |
solitaria Warren (sp. |
n.) . 88,270 280
Typhlopide : 5 PH
Typhlops bibroni (A. Smith) 5 P40
mossambicus (Peters) 227 |
Typhlosaurusaurantiacus Peters 227 |

Ubande = Necklet of littlesticks 161
Ubendhle = Loin covering for

girls : ; 2 Lo
Ubuhlalu obuluhlaza = Green |

beads : : . 163
Ubuhlalu obumpofu = Pink

beads : , 163
Ugubu, ugumbu= Stringed ne |

with calabash : 258

Uegwala= Bow with split ial 258
Ukutunga = Age for putting on

head-ring. 160
Ukwetshwama = Feast ‘of fir st

fruits 3 : . 260
Ulimi=Necklet . : . 163
Umampapeni = Necklet with |

one square of beads 2 1635 |
Umbrina capensis, Pappé . 245
Umegazi=Red beads. . 162 |
Umhlonhlo = Euphorbia . 259° |

Umkahlelo = Noisy wedding
dance in evening's . 262
Umkosi = Zulu King’s heel

Festival. : . 161
Umlilwana = Transparent brown
beads ; . 163

INDEX. 375

PAGE
Umnaka, ubedu = brass collar 160
Umgangala = Stringed bow 257
Umsenge = Cabbage 'T'ree 259
Umtomboti tree = Excecaria
africana Mull. Arg. 160
Umtshingo = reed-pipe 259
Umutsha = Loin-dress of men 161
Ungiyane = Viscous substance
from scale-insect 160
Uqwabe = Large stringed bow 258
Uzi = Sewing fibre 159
Varanide 225
Varanus albigularis (aut: ) 225
niloticus (L.) . 225
Variations in markiness of palpi
and wings of My-
zomyia funesta
Giles 129
of wing-pattern of
Myzorhynchus
paludis Theo. 149
Viperide 230
Vocal music of Zulus 260
Warren, Ernest 7,73, 838, 105, 109,
187, 209, 215, 269
Watt, Dr. Campbell 215
Wrightia arenosa (Alder)
196, 197, 272
Woodgate, A. 8. 169
Xenopus levis (Daud.) . 221
Zonuride 224.
Zonurus vittifer Reicha: 224.
warreni Blgr. (sp. n.)
224, 232
Zulu clothes : ; 160, 161
dances : 2 257, 262
music 257
ornaments ‘ 159, 160
songs, composition of 260
melody of . 262
rendering of 261
rhythm of . 261
text of 262

374 ERRATA,

ERRATA.

21, 6th line from the top, for “ Scapander” read Scaphander.

33, 8th line from the bottom, omit “‘ Mitra (Costellaria) dedala.”

52, 16th line from the top, for “ E. aclis A. Adams” read E. aclis (A. Adams).

78, 6th line from the bottom, for “ ectoderm” read endoderm,

80, 3rd line from the bottom, for “ Halocodyle” read Halocordyle.
117, 17th line from the bottom, for “ barbirostis” read barbirostris.
152, 5th line from the top, for “ barbirostis” read barbirostris.

189, 10th line from the top, for “ Hinck’s”’ read Hincks’s.

190, 5th line from the top, for “this layer” read the latter layer.
196, 6th line from the bottom, for “ Hinck’s” read Hincks.

196, 16th line from the bottom, for ‘“‘ Hinck’s” read Hincks’s.

205, 5th line from the bottom, for ‘ Leuconoste” read Leuconostoc.
206, 2nd line from the top, for “ Leuconoste” read Leuconostoc.
212, 4th line from the top, for “ vertical” read verticil.

228, 13th line from top, for “semiannulus” read semiannulis.

263, music of Song I, “ brace” to be omitted.

264, music of Song 3, “ braces” to be added.

266, music of Song 7, “‘ brace” to be omitted.

267, music of Song 8, “ braces” to be omitted.

Pl. XLIV, bottom line, for “ Uugub” read Ugubu.

P. 345, Sth line from the top, for “0:058” read 0°058 mm.

P. 348, 14th line from the bottom, for “ Angulhas” read Agulhas.

Pp. 270, 280, 282, 347, 349, 352, for “Tubularia betheris” read Tubularia
beth.

nca-balachelae) tachaychilacd a) lye) Suef laclelac\tackelae] cack ac) Tach 92h)

PRINTED BY ADLARD AND SON, LONDON AND DORKING, ENGLAND.

JUNE, 1906

ANNALS

OF THE

NATAL GOVERNMEN
MUSEUM

EDITED BY

ERNEST WARREN, D.Sc.(Lonp.)., Dtrecror.

PRINTED BY ORDER OF THE TRUSTEES.

LONDON:
ADLARD & SON, BARTHOLOMEW CLOSE.
1906.

Price 1os. net.

ADVERTISEMENT.

Royal octavo. Price 2/6 net.

OBSERVATIONS ON THE FILARIAL EMBRYOS
FOUND IN THE GENERAL CIRCULATION
OF THE EQUIDZ AND BOVIDA,

AND

THEIR PROBABLE PATHOLOGICAL SIGNIFICANCE.

FASCICULUS I,

BURSATI.

[Twelve Plates, Nine Text Illustrations, and Six Charts]

BY

ALFRED LINGARD, M.B., M.S., D.P.H.,

MEMBER OF THE ROYAL COLLEGE OF SURGEONS OF ENGLAND, AND
FELLOW OF THE BOMBAY UNIVERSITY ;
IMPERIAL BACTERIOLOGIST TO THE GOVERNMENT OF INDIA,

LONDON :

ADLARD & SON, BARTHOLOMEW CLOSE.

CONTENTS-

Inrropuctory Note.

of Natal. By C. Tare Regan, B.A. (With Plates
I—V.)

On Bertramia kirkmani sp. nov.; a Myxosporidium
occurring in a South African Rotifer. By. ERNEsT
Warren, D.Sc.Lond., Director of the Natal Govern-
ment Museum. (With Plate VI.)

On South African Marine Mollusca, with Descriptions of
New Species. By Epear A. Smurru, 1.8.0., F.Z.8.
(With Plates VII, VIII.)

On Halocordyle cooperi sp. nov., a Hydroid from
the Natal Coast. By Ernest Warren, D.Sc.Lond.,
Director of the Natal Government Museum. (With
Plate IX.)

On Tubularia solitaria sp. nov.,a Hydroid from the
Natal Coast. By Ernest Warren, D.Se.Lond.,
Director of the Natal Government Museum. (With
Plates X and XI.)

Notes on a New Species of Gymnoplea from Richmond,
Natal, South Africa; Adiaptomus natalensis (gen.
et sp. nov.). By Arnold W. Cooper, F.R.M.S., ete.
(With Plate XI1.)

Note on Convoluta roscoffensis Graff. collected on
the Natal Coast. By Ernest Warren, D.Sc.Lond.,
Director of the Natal Government Museum. (With
Plate XIII.)

Note on the Abnormal Hoofs of a Sheep. By Ernest
Warren, D.Se.Lond., Director of the Natal Govern-
ment Museum. (With Plate XIV.)

ADLARD AND SON, IMPR., LONDON AND DORKING.

Descriptions of new or little-known Fishes from the Coast

Lo)
(

1

83

co
af

105

109

MARCH, 1907

ANNALS

OF THE

NATAL GOVERNMENT
MUSEUM

EDITED BY

ERNEST WARREN, D.Sc.(Lonp.)., Drrec¥o

Ls ae. =
NV &

PRINTED BY ORDER OF THE TRUSTEES.

LODO:
ADLARD & SON, BARTHOLOMEW CLOSE.
1907.

Price tos. net.

ADVERTISEMENT.

Royal octavo. Price 2/6 net.

OBSERVATIONS ON THE FILARIAL EMBRYOS
FOUND IN THE GENERAL CIRCULATION
OF THE EQUIDZ AND BOVIDA,

AND

THEIR PROBABLE PATHOLOGICAL SIGNIFICANCE.

FASCICULUS I,

BURSATI.

[Twelve Plates, Nine Text Illustrations, and Six Charts]

BY
ALFRED LINGARD, M.B., M.S., D.P.H.,

MEMBER OF THE ROYAL COLLEGE OF SURGEONS OF ENGLAND, AND
FELLOW OF THE BOMBAY UNIVERSITY ;
IMPERIAL BACTERIOLOGIST TO THE GOVERNMENT OF INDIA.

LONDON :

ADLARD & SON, BARTHOLOMEW CLOSE.

CONTENTS.

A Contribution to the Study of the Characteristics of ,
Larve of Species of Anophelina in South Africa. By
Ernest Hitz, D.P.H.Camb., and L. G. Haypon, M.B.,
C.M., D.P.H.Aberd. (With Plates XV—XXVL.)

Language of Colours amongst the Zulus expressed by
their Bead-work Ornaments; and some General Notes
on their Personal Adornments and Clothing. By
Rev. Father Franz Mayr. (With Plate XXVIII.)

On Two New Reptiles from the Karroo Beds of Natal.
By R. Broom, M.D., D.Sc., C.M.Z.S., Victoria College,
Stellenbosch. (With Plate XXVIII.) .

On Entomostraca Collected in Natal by Mr. James Gibson.
(Part II.) By G. Srewarpson Brapy, M.D., LL.D.,
D.Se., F.R.S.,C.M.Z.S. (With Plates XXIX—XXXII.)

On Parawrightia robusta gen. et sp. nov., a Hydroid
from the Natal Coast; and also an Account of a
Supposed Schizophyte occurring in the Gonophores.
By Ernest Warren, D.Sc.Lond., Director of the Natal
Government, Museum. (With Plates XXXIII and
XXXIV.) : : :

Note on the Variation in the Arrangement of the Capitate
Tentacles in the Hydroid, Halocordyle cooperi
Warren. By Ernest Warren, D.Sc.Lond., Director
of the Natal Government Museum

Note on the Larva of a Fly (Sarcophaga sp.) occurring in
the Human Intestine. By Ernest Warren, D.Se.Lond.,
Director of the Natal Government Museum

ADLARD AND SON, IMPR., LONDON AND DORKING.

MAY, 1908

ANNALS

OF THE

PRINTED BY ORDER OF THE TRUSTEES.

LONDON:
ADLARD & SON, BARTHOLOMEW CLOSE.
1908.

Price ros. net.

ADVERTISEMENT.

Royal octavo. Price 2/6 net.

OBSERVATIONS ON THE FILARIAL EMBRYOS
FOUND IN THE GENERAL CIRCULATION
OF THE EQUIDZ AND BOVIDA,

AND

THEIR PROBABLE PATHOLOGICAL SIGNIFICANCE.

FASCICULUS I,

BURSATI.

[Twelve Plates, Nine Text Illustrations, and Six Charts]

BY
ALFRED LINGARD, M.B., M.S., D.P.H.,

MEMBER OF THE ROYAL COLLEGE OF SURGEONS OF ENGLAND, AND
FELLOW OF THE BOMBAY UNIVERSITY ;
IMPERIAL BACTERIOLOGIST TO THE GOVERNMENT OF INDIA.

LONDON :

ADLARD & SON, BARTHOLOMEW CLOSE.

CONTENTS.

On a Collection of Fresh-water Fishes, Batrachians, and
Reptiles from Natal and Zululand, with Descriptions
of New Species. By G. A. BovunEeneer, F.R.S.
(With Plates XXXV and XXXVI.)

Note on Clarias capensis C.@V. By G. A. BouLencer,
F.R.S.

A Collection of Fishes from the Coasts of Natal, Zululand,

and Cape Colony. By C. Tate Rreean, M.A. (With
Plates XXX VII—XLIL.)

A Short Study on Zulu Music. By Rev. Father Franz
Mayr. (With Plates XLIII and XLIV.)

On a Collection of Hydroids, mostly from the Natal Coast.

By Ernest Warren, D.Se.Lond. (With Plates
XLV—XLVIII.)

ADLARD AND SON, IMPR., LONDON AND DORKING.

PAGE

JULY, 1908

ANNALS

OF THE

NATAL GOVERNMENT
MUSEUM

EDITED BY

PRINTED BY ORDER OF THE TRUSTEES.

LONDON:
ADLARD & SON, BARTHOLOMEW CLOSE.
1908.

was
an

ny We ne iy a
‘Car’ “yd ie

CONTENTS.

TitLe Pace or Vou. I

Contents oF Vou. I

Inpex or Sourn Arrican Marine Mo 3tiusca

GENERAL INDEX .

ERRATA

ADLARD AND SON, IMPR., LONDON AND DORKING.

=

ae
a a

Sa

a)
me
4)

. aM

}
et
aa

pe

Ara hios ie ience:
ny fei h

‘

BL/WHOI LIBRARY

ee eee es

OP ere ee Pee

FT bears

alngestignsh Rurtereie 700d
ee ee
ere

Masnee
ree nr)
ote eee

. weueiee ae
Peer ir i ry
Are ee ee
+ ced erties _
oa. ausrarecohonstoreimienete are
ie peverne we ves BS
Perr Pe

snore a/ weginiaiets oe
ele na sinaeaieneis 20. 6r*
. erie siete age 8
~ easaiolnce

Te lchadher an eres See tes
O28 C4 REMMI HOO Rew e

alee ane
eer ere ea
irene eee Se Oe
Rrra rrr rt Caen eae
* pPeerernarn rr ore ee ee ee ee
ee en ene tree © the Sinaia \e ereinineiate eet
Poe eer en ee le ee eae
pene a te te ee

oneneee
Ar TC heats

Pa
ed

Pryce wri i ee ee
errno er ee
wrurererr eee ee

19 Ree. 0.870, 07 OR 8 | 814 500 0! FO 2 86 OKe
ones wate

Serer ar ere ere ae
<* eer ee eee re!
ee i ee eee

eer wren eres fe

eserarene
seeenee

seen

sae

eee

er er er er
ee
setae eee nerenens
a ee ee

ate ewe ‘ : essaseidhore
Oe ee rey et een ree
ee oe ee Se eee ee ee ee ee er
ang iedelg.¢-eislaip nave owe
PAR ROP ee Hin

tee a0 ee
20 eet ee
one ee
eer eee

se

7:09 86 ¢0

A AOA ARL OO kn ig
Or Ginhiie we a= wih aim ns
<a 0 in ees 8lO 8 pure

mene

Perrier)
nee ener eee en
prnrirs yr)

ort ee ae
Ae BRO RHO tle etm oleae ®
ore. Pre wee

Lt + Bien awe s

Cee ete eernneee
tinleded OL

ery wit eee ae na
dros eee nedaye

= o8 ttre
muniedeneel Bie sent aresere

Aten nae Sinitinidedy ah gua et,
w eeene te ene. s tgheraasgien cna
oe ees rie
ei ipiat! a al iegaietete : co a

ee ey Se eer
+ 0h BH @ Hae Ot

ie sels wns}
: Seperenarars fer we
270 ese wee ae
Cerri Srasaynainee
9i0)0) 5 mon ehe
.

Ce ee cot eenreners

OO Orn ee reee
Ary rt wer er

oe ewene
wen ee ee
0O 60 © Rg bat pet C8 eer

#24 Mieieie ae

eee
jefe ake ene
ecw mem

aren neee
anne nee

raneati este a ecget ay

enero neers
ere te

Sieh AO Oe Bie

ae
See agen ee ie

shee
e086 shape
ee

emntente

Pepto aC ee Tis
ete dette ee)
Apr-pimtate ae 0.6.0 ye

ener

MAINA Rem e
PA OR EE SC ened gee tue
4/08 08 WS O88 ele" pel aeD

eneenet
wenn.
tee eslere ernie etm . :
sraseterSrerese a ere ree ese 4 eee 29) Cael gone
Por prale Boihe O8 Ba MRI Oe e aren A Ses arp te ei eA oe
PT ot ee ee eee va ecoeg es thefaite ve Fa eelasele eeginor ee ee bepress esis
Fee erer et we ee oaisiaiein masse ee aentithee pee ncenas neveanie .
0 rate ° : re yp malar ae é ;
tay aap rn petyry perme aa ete ansnioie ore SPO Revel see! Bsisiats-einin) aneiala eed beeitad
Serer ae ie ee ee YL . . vont ane |
Aocaie 10-48 US 2 8's ope gp areeen © 6 oe
ee ee ed

Se es
20ers eae

+4 (a a8 acee @ be
esase 7

oon + Pets pean.

0 <3 aaee

en ee ee oes
ty 8 Lege © tele

ome
a eae

ee Ce

ee
weave

9 08 8 me

‘ or
re cee Bot newer . ae Secnieeg mee f

WREte Rene) SCoPE eeU eur tt eee ’ 6 perendi see Hi t-# arts
(o20t Da neme “Ot op e)8 0.8 98 Bee @ H8hr O

meee * ounr

* Seige wiley

peteee mee

A

‘4 wrens

Oe ee
at ’ : as Hdl mien tht rmynenenine
fae heme Spies «epaereave 9B. Gre Repel mee g! eretane g)4 9 0.eh> Ome
are estes parereune ye clerauacara: obekeieinne! Me Ae @ aaiaiaing iz a
: woe . 7 we 2 read
Pit esiagehe gn er Mt ies ad

Rae COR 4 Oni vte ia

ene ete
sisietg inne

. . ey
en ete
*

ane
tone

ee
ow
8 6a) eine
aan —

steers
noon gue ny

re
<8 gf
baa ee

sereer
eae nenecee
rales et mee a ermetrer #60 8
sme aoe eetere

see veeree
ake ew
sere eens

He eeenelng

Here eee ere
rears

oe niahie

0+ he aye! 0 geen e.

as"
ee ee

‘0989

aaa Fae
Aerie eee er ee

Ae ee

weed anny “49
Pe ea ae

any
Hehe eso awe ef ite wee

ty sh we ehy else. je

haha, ein) eee ial

Tiniehgeiniaia's.¢
ee afee @18 Bogie

ane oo
0040 Bk Leen a tee
Te

Ri
Naas rece
hd

oe tated
!, oo

ae
one
SEF Ae 98 the eile ah OR ig ie
aed

Paveyet «®,

ran)
NS teks ROE PH

tileae

9 ye
TF se

SPORE 9 err ate
ae el

oles

epee
Wnewrererene
wneeeene

eee ewe
ores enone

one

=o
tareeenerert
eens

Cab trbste e+
Bopanna

’
aoe
Pie See ere a ee eo
ee

prtealare aig.

- marr
. i thither
Ne Re Rote mi eay, ‘ evahdt theres?

91M per @ aie) 9.0) sare @lalabny
ib ieee

tok pe Op eeat

eee) i
. ee

eee

> HOR eee ate see

fee ee wii at

an 6,

ere ve Ne og Of ®

hie seniaie woe

v3 ye

ene
saetaore*

pe ele le poet ow
nee
veinjecg an) eeniel®
sere ajeleneene *

anne
oem ey

25 ERP wie a

ota

‘ oe eae
0 CF eNe OO BL 1% OLOFe HOE ae o~! o,
Se eieiaainl arial
seve een ne

ayeeniy eee
ochre

seortu
ae
deer

OE aw hee
Pe ee oe
ee eas *

Eeaekd
eee eer ee .

ae
eee eee

*
te

amr sone
6D ab eepenentt 20.2 He eer gms

tear e ©
Cr ee

BePephees
ferrites
we mabe

rae

weed ede
ois algtinw sere)
see

ediadhonsadee
mene Rp seute mP
arte

re ee
. ~ .

ase
Le A Ek POR MOR Hee Mew
oie berets

oe
Oh ery ecely gt 0
waenee

(trove ee
se may
piethbbe dials Tk y

wer eed
nee
a ee a

A yt Othe
hk ee mee

seen
asensertne, 00