$

CAPE OF GOOD HOPE.

DEPARTMENT OF AGRICULTURE.

MARINE INVESTIGATIONS

IN

SOUTH AFRICA.

VOLUME III.

WITH FORTY-FIVE PLATES.

B.8fj6

CAPE TOWN:
CAPE TIMES LIMITED, GOVERNMENT PRINTERS.

1905.

CONTENTS,

PAGES.

Descriptions of New South African Fislies. By
J. D. F. Gilchrist, M.A., D.Sc, Ph.D. With
Plates XIX.-XXXVL . . . . . . 1-16

Marine Annelids (Polychaeta) of South Africa. Part

I. By Professor W. C. Mcintosh, M.D., LL.D.,

F.R.S., etc. With Plates I.-IV. .. .. 17-56

Marine Annelids (Polychseta) of South Africa. Part

II. By Professor W. C. Mcintosh, M.D., LL.D.,
F.R.S., etc. With Plates V.-IX. . . . . 57-92

The Turbinolid Corals of South Africa. By J.

Stanley Gardiner, M.A. With Plates I.-III. . . 93-130

The Development of South African Fishes. Part II.
By J. D. F. Gilchrist, M.A., D.Sc, Ph.D. With
Plates V.-XI. . . . . . . 131-1G7

Echinoderma found off the coast of South Africa.
Part I. Echinoidea. By Professor F. Jeffrey
Bell, M.A. . . . . . . . . 167-176

The Plankton of the South African Seas. By P. T.

Cleve. I., Copepoda. With Plates I. -VI. . . 177-210

The Alcyonaria of the Cape of Good Hope. Part IL
By Professor Sydney J. Hickson, M.A., D.Sc,
F.R.S. With Plates\ni.-lX. .. .. 211-239

Echinoderma found off the Coast of South Africa.
Part II., Asteroidea. By Professor F. Jeffrey
Bell, M.A. . . . . . . . . 241-253

Echinoderma found off the Coast of South Africa.
Part III., Ophiuroidea. By Professor F.
Jeffrey Bell, M.A. With Plate I. .. 255-260

37108

DESdlMI'TIONS OF NEW SOUTH
AFRICAN FISHES.

BV

J. D. F. GILCHRIST. M.A., B.Sc. Ph.D.,
Government Biologist to the Colony of the Cape of Good Hope

The following- is a continuation of the description of new
fishes procured on the South African Coast. It includes one
new genus and seventeen new species, one of which may be a
new genus, as follows : —

1. Serranus knysnaensis, n.sp. (Scrraniddc).

2. Pterois nigripinnis, n.sp. {Triglidae).

3. Histiopterus spinifer, n.sp. (Squamipinnes).

4. Lampris immaculata, n.sp. {Lampridae).

5. Cyttosoma boops, n.g. and sp. (Cyttidae).

6. Cottunculus macrocephalus, n.sp. {Cotiidae).

7. Pseudorhombus natalensis, n.sp. {P leuronectidae).

8. Pseudorhombus ( ?) andersoni, n.sp. {P leuronectidae).

9. Platophrys dimorphus, n.sp. {P leuronectidae).

10. Solea turbynei, n.sp. {P leuronectidae).

11. Cynoglossus attenuatus, n.sp. {P leuronectidae).

12. Cynoglossus brachycephalus, n.sp. {P leuronectidae).
1 5. Arnoglossus macrolepis, n.sp. {P leuronectidae).

14. Synaptura melanoptera, n.sp. {P leuronectidae).

15. Solea fulvo-marginata, n.sp. (P leuronectidae).

16. Synaptura ciliata, n.sp. {Plcuronectidae).

ly. Chlorophthalmus punctatus, n.sp. {Scopclidae).
18. Scopelus argenteus, n.sp. (Scopelidac).

Fam. SERRANID^.
Serranus kynsnaensis, n.sp.

(Plate XIX.)
D. X 14. A. Ill ;. Sc. 83 2^H- L. i. ']2.

Depth of body '^\ times in length ; length of head equals
depth of body. Snout slightly longer than diameter of eye,
which is about \ the length of the head ; interorbital width 55^
m length of head ; lower jaw projecting, with strong canines
on the sides ; maxillary extending to nearly below centre of
eye ; snout naked. Preoperculum and operculum scaly ; pre-
opercula finely serrated on both the posterior and lower border ;
opercular spines strong, the middle, which is equidistant from
the upper and lower, is largest. Gill rakers long, 1 1 on the
lower part of the anterior arch. Dorsal originating slightly
behind extremity of longest opercular spine ; first and second
spine shortest, the rest not very unequal in length, and not
separated from the soft rays by a notch ; fifth spine equal to or
slightly shorter than the longest soft ray, and a little over 2\ in
length of head. Caudal is somewhat imperfect.

The single specimen from which this description is taken
was forwarded by Mr. Jackson, the Resident Magistrate at
Knsyna, where it was found.

Fam. TRIGLID^.
Pterois nigripinnis, n.sp.

Br. VII. D. XIII 9. P. 20. V. I 5. A. II ;. C. 12.

L. r. 35. L. tr. \,

Depth of body 3 times in length without caudal, 4I times
with caudal. Length of head 2\ in body. Diameter of eye a
little less than length of snout, and contained 3 times in length
of head. Interocular space | diameter of eye, deeply concave.
A long tentacle over orbit of eye, longer than breadth of inter-
orbital space, with a black spot at its extremity. The wide
posterior nostril is immediately in front of eye, the anterior
about half way to end of snout, and provided with a long
tentacle, almost as long a5 orbital tentacle, but without black
spot A long tentacle, about twice the diameter of the eye,
hangs down from lower margin of preorbital bordering on the
jn axillary.

ujIlibraryi?

Ihere are 35 rows of scales between temple and ba^L.
caudal ; scales cover the head, including middle line of inter-
orbital space, but not jaws, throat, and top of snout.

The pectorals reach nearly to the base of the caudal the
ventrals to about the 4th ray of the anal. None of the caudal
rays are prolonged.

Colour (m form aim) : Dorsal fin mottled with black, anal
and caudal with scattered black dots^ ventral and pectoral
black, body brownish.

This new species seems nearest to P. macrurus, Alcock.

Localily. Three specimens were found 2^ miles off the
Umhlanga River mouth (Natal). Depth, 22 to 26 fathoms ;
bottom, fine sand. The largest of these was 85 mm. in length
(including caudal).

Fain. SQUAMIPINNES.

Histiopterus spinifer, n.sp.

(Plate XXI.)

D. IV 26. A. Ill 10. P. i;. L. 1. 64. C. 4 + i; + 2.

Height of body a little over 2^ in length, the highest point
being at the last spines of the dorsal. . Head if in height of
body ; snout 2\ in head. Diameter of eye 3f in length of head,
near y equal to tne interorbital space, which is slightlv concave
bmall clusters of spines, which might rather be designated
tubercles, occur (i) in the median frontal region just behind
the head, (2) over the orbit, (3) above the upper angle of the
opercular opening, (4) at the angle of the pre-operculum, and
v5) above the maxillary The scales of the body are small and
ctenoid. On the head they are only present on the cheek and
above the operculum, being absent on the frontal region, snout,
operculum, and apparently the^ lower jaw.

'v7^^^ ^^^^^^ ^^'^ ^™^^^' forming a broad band in both jaws.
1 here are none on the vomer or palatines.

The dorsal fin is very high, its longest ray being longer than

the height of the body in the proportion of 11 to 10. It is

•<:overed at its base by scales. The ventrals extend to beyond

the root of the caudal, and are longer than the depth of body.

The anal extends to about the posterior third of the caudal.

Colour: The body (in spirit) is an uniform dark brown, be-
■coming almost black on the fins.

Only one specimen was procured by trawl.

Locality: Mossel Bay; depth, 30—36 fathoms; bottom, mud.

Measurements of Specimen

Height of body 50 mm

Length.. „ (without caudal) 72 „

., head 30 „

„ snout 12 „

Diameter of eye 8 „

Length of ventral 45 ,.

„ pectoral 24 „

Longest ray of dorsal 55 „.

Fam. LAMPRID.^

Lampris immacuiata, asp

(Plate XXII.)
Br. 1.3. D. 54. A. 36. V. 14.

Depth of body a little under | the length (exluding caudal),
length of head 3 if, diameter of eye about 5? in lengtli of head
and more than h the interocular space.

Pectorals rounded, if in length of head Anal has the first
rays longer than those immediateh' succeeding them Other
fins as in L. liina.

The paired and unpaired fins are of a deep red colour. Snout,,
lips and region anterior to the eyes of the same colour, but less
marked. Edge of pre-operculum, of operculum and posterior
margin of gill opening, tongue and throat, including inner
surface of rather long gill rankers and branchial arches, are
similarly coloured. Body of a dark steel blue colour above,
lighter and greenish below, the whole, including- that part of
the head posterior to the eye, with a suffused rose red colour
There are no white spots on any part of the body

The lateral line is well marked, each scale, with a single
unbranched tube, being of a bright red colour. These spots
are continued forward from the anterior end of the lateral line
ill two directions, one arching upwards and forward to the top
of the head, the other forward over the eye to the end of the
snout. These, about half-a-dozen in each line, were of the
same deep red as the pierced scales of the lateral line, and had a
simple mucous pore in the centre.

The tongue is narrow, free and smooth as Lowe describes it
ill his specimen, and without the asperities noted by other
obser\-ers. No large scale was found before the anal, but the
pits above and below the caudal peduncle were present. The-
pyloric caecae were very numerous.

I have hesitated to describe this form as a new species asit
agrees in many particulars with the Northern form which has
been found to vary considerably. In general proportion of the
body and fin formula it cannot be separated from L. lima, but
the antenor elevation of the anal fin, the entire absence of spots
on the body and the presence of marked lines of mucous pores
over the head, which could hardly escape the notice of other
observers, seems to indicate that it belongs to a different species.
The pectorals axe shorter and more rounded, but this may be
an age characteristic.

The anterior elevation of the anal fin is of interest from a
systematic pomt of view, as connecting the isolated family of
the Lainpridae with the Scombridae.

The specimen described was found on the 2nd August, 1902,
cast up on the beach at jMuizenberg, in False Bay, and was
quite unknown to the fishermen at that place or at the adjacent
Kalk Ba}'. Unfortunatel}-, when procured, it was somewhat
damaged, the right jaw being torn off, the left pectoral and
both eyes gone. It was, however, in a perfectly fresh state.
Wlien placed m formalin tlie brilliant red of the fins rapidly
faded.

There is a stuffed specimen in the South African Museum,
found on the shores of Table Bay in the year 1887. There are
well marked large white spots painted on this fish, but that this
is an error of the taxidermist copied from a figure is very pro-
bable, as a note by Mr. Trimen in the records of the Museum
undoubtedly refers to this fish, viz. : — " A Lampris (red and
without spots) found stranded in Table Ba}, 4th April, 1887.
Length 3 ft. 7 in."^ The elevated anterior rays of the anal are
even more marked than in the specimen described. The pec-
torals are not, however, so rounded.

The following measurements are those of the fresh specimen
from Muizenberg : —

Length of bicdy, excluding tail 805 mm.

„ to end of middle caudal ra}'... 880 „

Depth „ ■ ... 380 „

.Greatest thickness 143 „

. Longest ray of dorsal 185 „

ventral 170 „

„ „ pectoral along curve 200 „

„ .. ,> from tip to base 174 „

Lengtli of head 245 „

Diameter of eye 46 „ ?

Jnterocular space 85 „

f\im. CYTTID^

Cyttosoma, n.g.

Body compresseci and elevate, head not .->o much compressed.
Upper profile concave. Eyes xerry large, scales small, deciduous
and replaced by spiny tubercles on head, nape of the neck (to i st
dorsal), thoracic and abdominal regions and by hard spiny
scales along bases of dorsal and anal fins. Dorsals nearly
separate, being joined by very low membrane, the first having
seven spines and much shorter than the second. There are
three anal spines. Rays of dorsal 29, anal 28, pectoral 22, and
ventral "^ ; pectoral and ventral well separated. Minute teeth in
lower jaw, none in upper jaw, vomer or palatine bones. Bran-
chiostegals 7.

Cyttosoma boops, n.sp.

(Plate XXIII.)
Br. VII. D. \U 29. A. Ill 28. V. I ;. L. 1. roo (.approx.).

The upper jaw 13 very protractible and can be drawn out i|-
times its normal distance from the eye. The greatest depth of
the body is 2^, times the diameter of the eye, and is contained a
little more than i^ in total length which is 158 mm. (mouth
closed). The length of the head is contamed 2^ tmies in the
body. Interorbital space if diameter of the eye The lower
ridges of the mandible serrate. Opercular bones are not serrate.
The superior profile of the body rises abruptly behind the eye
to the origin of the dorsal and continues backwards in the form
of a thin ridge with straight border to within a distance from
the root of the caudal equal to about | of its length. The
ventral profile is somewhat similar, though not so pronounced.
The ridge here begins under the vertical from the commence-
ment of the soft dorsal. It has the same straight outline. The soft
rays of the anal are somewhat longer than those of the dorsal.
The inter-maxillar\- can be received into a liorse shoe shaped
space which extends backwards to over the centre of the orbit.
It is in length i^, times the interocular width. The maxillary is
of the same lengtli.

The nostrils are placed close together in front of an anterior
protuberance of the supra orbital ridge. The pseudobranchiae
are well de\eloped and the gill rakers are stout, short and lamel-
liforni.

There are about 100 scales on the lateral line which extends
on to the caudal The greatest height of its anterior curv^e is
below the commencement of the first dorsal . from this point

it gradually descends tiii under the 15 th soft ra\-, when it con-
tuiues backward in a straight horizontal line.

The scales of the body are thin and deciduous, circular and
with concentric striae only. On the middle region from the
commencement of the dorsal to about the anterior margin of
the orbits they assume the form of rough tubercles. The same
occurs in the ventral region between the beginning of the anal
and the isthmus. One or two enlarged scales occur on the body
under the dorsal spme.

There are about 40 bon}' scales proxided with low spines
along the base of the dorsal and about the same number along
the base of the anal. Two short bony ridges extend backwards
from the posterior region of the eye.

Locality: The single specimen procured was obtained by
trawl, Vasco de Gama Peak (near Cape Point) bearing N. 40*^
E., distance, 13^ miles; depth, 120 fathoms; bottom, rough.

F'ani. COTTID^.
Cottunciilus niacrocephalus. n.sp.

(Plate XXIV.)

D. IV II. A. II ' V. 3. P. 20.

Length of head one-half the length of the bod\'. Depth of
body 3^- in length ; diameter of eye y^, in head, or one-half
depth of body. Snout nearly equal to diameter of eye.

Head and bod)' covered by smooth loose scaleless skin. No
bony tubercles on the head apparent. The posterior extremity
of the maxillary is broad, 3^ in diameter of eye, and falls in the
vertical just in front of the centre of the eye.

The dorsal fin com.mences on a line between the posterior
margins of the opercula. The anterior spines are short, and
scarcely penetrate the skin. Tiie soft rays extend backwards to
near the tail. They fall nearly within the same verticals as the
anal fin. The ventrals with 3 rays scarcely reach the vent,
which is a little behind the \ertical from the posterior edge
of the operculum. The distance of ventral from end of snout
about i length of body.

Colour : Light brown dorsalh', colourless ventrally.

Locality: Lion's Head, bearing S.E. \ S., distant 50 miles;
depth, 230 fathoms ; bottom, green mud ; and South Head,,
bearing E. by S., \ S., distant 25 miles; depth, 190 fathoms;
bottom, green sand and black specks.

8

Fam. PLEURONECTID.^.

Pseudorhombus natalensis, n.sp.

(Plate XX\;)

D. 6;. A. 52. V. 6. L. 1. 62.

• Head 3^ times, depth of body a little over two times in
length of body, without caudal. Diameter of eye i length of
head. Eyes on left .side, the lower slightly in advance of the
upper, close together, divided only by a ridge. The distance of
lower from the end of the snout is li in the diameter of the eye.
The lower jaw^ projects slightly beyond upper. The posterior
end of the maxillary falls in the vertical between the anterior
margin and the centre of the eye. Teeth : about 20 in lower
jaw, a few more in upper, slightly more developed on blind side.

The dorsal hn commences in front of the upper eye, just
behind a small indentation on the snout. The first ray is
partially free, being attached to an inter-radial membrane for
about i of its length, the second to the fifth approach gradually
the form of the remaining rays, viz., with tips free only. All
the rays are simple but the last two, which are small and
bifurcate. The longest is in the posterior third, and is 2f in
length of head. The anal has the last three rays branched.
Left pectoral is if in length of head, the right pectoral a little
more than one-half the left. The pre-anal spine is very slight.
C-audal is w'edge-shaped, with upper and lower border of
posterior margin slightly concave.

Scales : Ctenoid on the left, cycloid on the right side.
Lateral line with strong anterior curve for about ^ of its
length, the height of the curve being a little more than l ^^s
length ; the nucal branch ends about f of the way tow^ards the
8th ray of the dorsal fin ; a branch starting from the same
origin, proceeds to the interorbital region, giving off on its way
another branch towards the pre-operculum.

Colour: Body covered with dark brown ring-shaped spots of
fairly constant arrangement. Heads and fins with smaller
spots.

Length of the largest specimen (including caudal), 132 mm.

Locality: Cape Natal bearing W. by N., distant 6^ miles;
'depth, 54 fathoms ; bottom, fine sand.

Pseudorhombus.

The apparent generic characters of this fish may be described
as follows : — Eyes partly on left side, the upper occupying a
crescentic gap between the dorsal fin and the heiid, and nearly
in the vertical middle line of the head ; mouth wide, the length

of the maxillary more than ^ that of the head ; teeth in a single
series in both jaws, those in front slightly larger ; interorbital
space a mere ridge ; dorsal commencing on a prominence
which projects forward over the upper eye.

This definition approaches so nearly that of the genus
Pseudorhombits m some respects (teeth, length of maxillary,
etc.), that until more specimens are procured, its validity as a
separate genus, and not merely an abnormality, must be received
with some doubt. The peculiar feature which distinguishes it
from Psendorhombiis is the near approach to the original
bilateral symmetry of the head, and such variation, though very
striking, is perhaps to be expected in a character of so recent an
origin philogenetically. It apparently, however, does not
belong to either of the three species of Psendorhombiis already
recorded from this region (P. russclli from the Umbilo River,
Port Natal, recorded by Giinther, ride Catalogue, Vol. IV.,
p. 424 ; and the new species described above). It differs from
these, and all the Indian species recorded by Day, in having the
scales ctenoid on the blind side and it apparently belongs to an
undescribed species.

Pseudorhombus andersoni, n.sp.

(Plate XXVI.)

D. 74. A. 54. L. 1. 85.

Depth i, head 3^ m length (excluding caudal). Eyes close
together separated only by a ridge in diameter a little more
than \ the length of the head, and equal to the distance between
the lower and the end of the snout. The maxillary reaches to
below the hiiid third of the eye. Teeth: about 12 in the lower
jaw on the left side, 8 on the right. The dorsal fin commences
at the tip of the projection which is about i^ times in the dia-
meter of the e}'e. The pectorals are about equal, and one half
the length of the head. The caudal was somewhat damaged,
but appeared to have been wedge shaped, the middle ray
(32 mm.) being apparently the longest.

The scales are ctenoid on both sides ; the anterior curve of
the lateral line is well marked, its depth being about a third of
its length. The dorsal branch of the lateral line goes to the
gth ray of the dorsal. Scales occur on the ra}-s of the dorsal
and anal. Any colour that may have been present has been
entirely removed by the spirit in which the specimen had been
preserved for a long time.

Total length, including caudal, 17G mm.

Locality: The single specimen was found in Durban Harbour
by Mr. A. Anderson to whom we are indebted for this and
other valuable specimens.

lO

Platophrys dimorphus, n.sp.

(Plate XXVII.)
D. 8;. A. 72. V. 6. L. 1. 50.

Depth of bod}- i* in length without caudal; length of head
4| ; longitudinal diameter of eye about 3^ times in head, i^ in
interorbital space (in the male, less than i in the female) and
less than the distance of the lower from the end of snout. The
posterior margin of the lower eye is near the vertical from the
anterior margin of the upper. Lower jaw projects beyond
upper ; the maxillary scarcely reaches to the anterior border of
the lower eye. In the male there is a strong spine in front of
the lower eye projecting beyond the profile of the snout and
a few tubercles on the (morphologically) upper margin of the
orbit of either eye. These are absent in the female. A further
sexual distinction is the presence of cutaneous flaps on the
posterior margin of the eye in the male only. Teeth : a single
row of fine setose teeth on the upper and under jaws.

Rays in anterior half of dorsal the longest, about li^ in head.
The first two or three anterior and posterior rays are to the
right of the median line. Left pectoral, excluding prolonged
ray, a little less than the length of the head ; with prolonged ray,
which is usuall)' thougli not invariably present in the male, it is
i4 in the length of head. The right pectoral is smaller than
the left about \\ in the length of the latter. The right ventral
is in the middle line in advance of the left which is entirely on
the side. The posterior extremity of the left reaches beyond
that of the right. The left is coloured, the right being un-
coloured.

Scales: Ctenoid on the left side, cycloid on the right, the
lateral line has anteriorly a strong cur\'e whose depth is about
half its length.

Colour: Male, dark grey with minute black dots on anterior
border of region between the eye and on ocular flaps, female of
the same colour with the addition of dark spots or blotches
often arranged in rows of 3 or 4 along the middle of the body
and on either side above and below it.

Locality: 2\ miles off L^mhlanga River mouth (Natal);,
depth, 22-26 fathoms ; bottom, fine sand.

Solea turbynei, n.sp.

(Plate XXVIII.)
D. 65. A. 52. L. 1. 94. Vert. 8 - 25.

Depth slightl}- over 2^,, head 4 in length of body (without
caudal). Diameter of eye 5^ in head. Upper in advance of
lower, by about a lialf diameter. Interorbital space \ the

1 1

diameter of the e}-e. Mouth extendi nearl}' to belcAv centre of
lower eye. Lips not as a rule fringed — of lo samples, one had
one papilla on the upper lip, and another had a few on both.
Papillae on left side of head, close up to dilated left nostril,
which is not fringed. The margin of the operculum has also
papillae, and there are a few at the base of the pectoral on the
inner border of the opercular opening.

Dorsal commences on snout above the level of the upper
margin of the upper eye, and extends to near caudal ftn.
Longest ra}- 3 times in head. Length of free portion of caudal
is contained 3 times in its depth. Right pectoral, which is
scaled at the base to about •? of its length, is contained 2^- times
in length of head, left pectoral slightly less.

Scales : Ctenoid on both sides. The lateral line contains
about 94 pierced scales, crossed by about 108 rows of scales.
The dorsal branch of the lateral line rises abruptly towards the
loth ray, and then bends forward in a sharp curve to run
parallel with the dorsal to the snout.

Colour : Light slate colour, with numerous black dots on
head, body and vertical fins. There is a black patch on the
right pectoral and sometimes on the left.

Locality : Numerous specimens have been found in the work
of the Government steamer, under the charge of Captain
Turbyne, after whom the specimen is named. They were
procured at Mossel Ba}^ (18 fms. fine sand), and other localities.

Cynogiossus attenuatus, n.sp.

(Plate XXLX.)

D. 103. A. 90. C. 9. V. 4. L. 1. 84—86.

Height 4I in lengtii without caudal, 4.' v.ith caudal. I^ength
of head il in height of body. Height of head \ of its length.
Diameter of eye ,^,, the length of the head, 4 times in its
distance from the snout. The upper in ad\'ance of the lower,
the lower one situated slightly in front of the middle of the
head ; interorbital Space about -.\ of the diameter of the eye.
The angle of the mouth is behind the posterior margin of the
lower eye, and behind the centre of the head. One nostril
between the eyes, and another in front of the lower, on a level
with its lower margin. One ventral fin scarcely attached to
anal.

Scales : Ctenoid on the coloured, c}'cloid on the blind side.
Lateral lines, two on the coloured side separated by 12 rows of
scales, two on the blind side.

Colour: L'niform brown.

Locality: Tugela River mouth bearing X. b}' W'., distant 4^
miles ; depth, 24 fathoms ; bottom, mud.

12

C.ynoglossus brachycephalus, n.sp.

(Plate XXX.)

D. ic)/. A. 82. \'. 4. C. 9. L. 1. j^. L. tr. 14.

Deptli of body 4 to 4], length of head 55 in length without

• caudal. Diameter of eye 6 to 6^ in length of head. Eyes close
together, orbits almost touching each other, upper slightly in

■ advance of lower, posterior margin of lower in the middle of
the length of the head. Greatest height of head more than its
length. Angle of the mouth falls under the centre of the lower

■ eye. Rostral hook in front of the anterior margin of the upper

• eye. Two nostrils, one on the anterior part of the interocular
region, the other, with rather long tube, above the lip before the
lower e}'e.

There is onl)' one ^•entral fin (the left ?). It is slightly

• attached to the anal.

The scales are ctenoid on both sides. There are two lateral
lines on the left side separated b}' 14 rows of scales; no lateral
line on the right side.

Colour : Body light brown with darker markings. The fins
are strikingly coloured, being light brown with patches of very

■ diark brown at short intervals. A patch of dark brown occurs
on the caudal rays.

Locality: 2\ miles off Umhlanga River mouth (Natal);
> depth, 22-26 fathoms; bottom, fine sand.

Arnoglossus niacrolepis n.sp.

(Plate XXXI.)

D. ;2. A. 50. L. 1. 4;.

Depth 2^, head a little over 3^ in length (without caudal).
IwOngitudinal diameter of eye 45 in head. Eyes on left side,
the upper slightl}' in advance of lower. Interocular space a
mere ridge. Jaws with a single series of small teeth; lower
jaw projecting beyond upper ; maxillary extends to a point be-
tween centre and posterior of lower eye, and is nearly equal one
half the length of the head. An open nostril in front of inter-
orbital ridge with a tubular one anterior and superior to it. Of
the corresponding nostrils of the right side the tubular one is
on the snout in the middle line in front of the upper eye and
the other is on the right side, is large and covered by a flap
which is in a line with the dorsal fin. Gill openings are wide ;
a flap of skin (the continuation of the opercular membrane) is
folded over from tlie left jaw to cover the right.

The dorsal commences on the snout a little to the right of
the median line. The raj-s are divided and gradually increase
in length to the third from the end, which is longest and 2,-'^, in

head. The anal is pretty similar to the dorsal, but the first and:
last rays are longer than the corresponding ones in the dorsal.
The left pectoral is a little over ^ the length of the head ; the
right pectoral slightly less than the left. The caudal is covered
with scales to about half its length, which is about ! of that of.
the body.

Colour: Light brown with a few black patches on body and
on vertical fins. Two black patches on the body at the extrem-
ities of dorsal and anal well marked and seemingly constant.

Locality : Tugela River mouth (Natal) bearing North ; distant-
22 miles ; depth, 63-73 fathoms ; bottom, mud.

Synaptura melanoptera. n.sp.
(Plate XXXII.)
D. 92. A. ;o. L. 1. 128.

Height 3 "4, head 4*3 in body without caudal. Eye 5^- in-
head, 8 times interocular width. Eyes close together, upper in
advance of lower by }, a diameter. Angle of mouth under
centre of lower eye. Gill cleft feebly fringed. Left nostril
dilated, a space free from papillae surrounding it, extending to
mouth. Dorsal commences in front of centre of upper eye aiid
extends to caudal, with which it is slightly confluent at lower'
half of last ray. Longest ray in centre 2.3 in head. Right
pectoral 4 in head the same length as left pectoral. Ventrals
symmetrical equal in length to pectorals. Scales ctenoid on
right imd left sides. Total length 83 mm.

Colour: All fins, except left pectoral which is colourless, are
jet black, the extremity of the right pectoral tipped with white.
The right side is light brown with large patches of darker.

Locality: Lat. 33° 6' 45" S., long. 27° 55' 45'' E. ; depth, 43.
fathoms ; bottom, sand, shells, and mud.

Solea fulvomargrinata, n.sp.
(Plate XXXIII.)

D. 7S-80. A. 61-65. L. 1. IUO-108.

Depth 2.^, head 5I in length of body (without caudal). Longi-
tudinal diameter of eye a little less than 5;^ in head, upper in
advance of lower by about half a diameter. Vertical diameter
about 7 times in head and a little more than interorbital space.
IMouth extends to below centre of lower eve.

14

Dorsal commences on snout at the level of the upper margin
of the upper eye and extends to base of caudal. Longest ray
I the length of the head, the posterior extremities of the dorsal
and anal are situated over the last vertebra and their posterior
margins are almost vertical. The right pectoral, which is de-
void of scales at its base, is 6^ in head, the left pectoral being
slightly longer.

Scales : Ctenoid on both sides. The lateral line is straight
and difficult to make out.

Colour (in fresh state) ; Lemon yellow, most marked on the
margins of all the fins, except the left pectoral, which is colour-
less. Cloudy patches of dark colour occur on the body, some-
times extending on to the bases of the vertical fins, and the
whole is speckled with dark and white dots. The blind side is
strikingly coloured, the dorsal, anal and caudal being of a bright
lemon yellow, which at places extends slightly on to the body.
The rest of the blind side is colourless, including left pectoral
and papillae of head region. (The yellow colour entirely dis-
appears from the fish after being in alcohol a short time.)

Locality : Occasionally procured by seine net in False Bay.

Synaptura ciliata, n.sp.
(Plate XXXIV.)

D. 76. A. 63. V. 3. L. 1. 100.

Height 2" 7, head 5 in body (without caudal). Eye 8' 5 in
head, a little less than double interocular width, upper in
advance of lower; angle of mouth below centre of lower eye.
The lips and border of gill clefts are fringed, and the whole of
the head and body of coloured side is covered with hair-like
processes. A fringed dermal flap at nostnl of blind side.

The dorsal commences on the snout at the level of the
centre of the upper eye. The rays gradually increase in size to
the last, which is confluent with caudal. The anterior are
fringed on the left side. The longest ray 2*3 in head. The
right pectoral is 4' 2 5 in head. The left pectoral is slightly
longer and broader. The ventrals are very small, the right 2f
in pectoral, and a little less developed than left ventral.

Colour: Body dark brown, with darker specks, head almost
black, vertical fins light brown, with dark spots, margin white.
Right pectoral black, with brown tips. Pectoral and ventral
ol left side colourless, as is whole of left side.

Near S. niarginata, Blgr., but has shorter paired fins, origin
<:f dorsal further forward, body covered with short filaments.

Locality: Irmer Harbour, Durban (Natal).

15 /^ z^*^ <i\<p

ujILISRARYUq:

CMorophthalmns punctatus, n.sp.\^\ /^^.,<6,Xrv?

F"am, SCOPELID.^
lialmns puncta

(Plate XXXV.) '^^^^^^t^^L)>

Br. 8. D. 12. A. 9. P. 15. V. 9. L. 1. 56. L. tr. {\,,

Length of head a little less than 4 times m that of body.
.Diameter of eye 2 "6 times in length of head. Interorbital space
a little more than -| the diameter of the eye.

The first rays of the dorsal fin are in front of the ventrals,
and the distance between the adipose fin and the dorsal about
equals the distance between the dorsal and anterior margin of
the eye. The pectorals are also equal to this distance, and they
do not reach the tip of the ventrals, which are nearly f the
length of the head.

Very fine teeth, or asperities, can be distinguished on the
roof of the mouth. The margin of the scales have denticulations

■(2-3)- , ,

Colour (in spirit) : Dark brown patches at nape of neck, an-
terior and posterior of dorsal fin, between this and adipose fin.
Patches of the same colour occur on the side of the body. The
head, body, and fins are covered with minute black spots, which
cause the rows of scales to stand out very markedly, especially
those in a direction descending obliquely backwards.

Locality: The three specimens procured were obtained by
shrimp trawl, Lion's Head bearing N. 63° E., distant 34 miles ;
•depth, 154 fathoms ; bottom, black specks.

Scopelus arg-enteus, n.sp.

(Plate XXXVI.)
D. 14. A. 15. L. 1. 38.

The depth of the body is contained 5 times in its length ; the
length of the head 3! times. The diameter of the eye is con-
tained 4 times in the length of the head. Snout i^- the diame-
ter of the eye. Mouth wide, premaxillary reaching nearly to
the angle of the pre-operculum, the space between the posterior
margin of the eye and the border of the pre-operculum is a
little longer than half the diameter of the eye.

The origin of the dorsal fin is nearly midway between the tip
of the snout and the posterior extremity of the adipose fin ; it is
slightly in front of the point of insertion of the ventral fin. The
pectoral fin reaches to the insertion of the ventrals, not beyond.

Colour: Dark brown along the dorsal region of the body.

i6

reaching- about halt way to the lateral hue, and a ie,33 marked
tinge of brown along the ventral region. In the middle of the
body this colouring matter is nearh' absent, and the scales have
a bright silvery appearance.

The luminous spots are arranged as follows : Caudal 5 (in
this respect differing from all other species of Scopelus and
affording a ready diagnostic character) ; postero-lateral i ; anai
S + 5 extendmg from origin of anal to midway between its
posterior end and beginning of caudal tin, medio-lateral (' supra-
anal " of Liitken) 3 in a line with the last of the ventrals and
lorming an oblique series ; antero-lateral, i between ventral fin
and lateral line ; abdominal 5 in a curved line ; thoracic 5, 3 ki
a line with mandibular and lower opercular, and 2 higher up in
tront of insertion of ventral ; pectoral 3, 2 immediately below
and I above insertion of pectoral fin ; opercular 2, the lower
situated behind the angle of the mouth in a line v/ith mandi-
bular; mandibular 3, antorbital i, somewhat indistinct, on front
margin of the eye a little below nostril.

Besides these luminous organs, there is a patcli of luminous
scales above and below the caudal region, and the snout is
covered by soft glandular tissue, probably luminous in functioo.

Local i^ J': Constable Hill (near Saldanha Bay) bearing E. ^^
S., distant 19^ miles, and Green Point bearing S.E, by E. |- E.,.
distant 36 miles-, also South Head bearing E. by S. | S., dis-
tant 2t; miles.

Published ist March, 1904.

CD

■H

M

^

O

• I—

-O

fd

cB

■ac

o

03

■r-J

1

r^

!^

^,

— 1

D

(D

O

^

LO

^

8^

f— I
W

p

LQ

If]

w

[£)

en

P^

Marine InvestigaLtiOTis
Soiitli Africa..

RsliesPl.XX.

p. m Manns del .

cJ. Greex^ litn..

PTE'R.OIS lTIGRIPnSl"NIS^T-L.sp

lAasPvae Trrve stigations
Soixtli Africa. .

'F^sWes'Pl.XXl

P.McManus del

(J. Green, litli.

HISTIOT^TERUS ST'lNIFER.n .sp .

(V

3;

03

o

? o

<

o

I— I

w

I — I

(1)

S CO

I

w

■H

:fc

-•3

%'i

w

P

o

•H

T3

^ 5.

u^^J

/

, '^-

pj's

' ^v

F ' '

1). C f -1

0)

W

o

i

p
,p

o

p

H
H
O
O

s

^

w

d

(0

g

^ n

r6

-P
t/]
CD

I

0)

■ H

^

1)

^

W

0]

w

<

H
<^

CO

P
ffl

O

o

p
p

m
A.

>:

X

^

CO

(D
/-<

CO

■r-i

C3

lU

O
03

w

O
0

p

LQ

Mamrie Inves ligations
SoutK ATpica.

Kshes Pl.XXVlI.

CiAf^#

1*^311 •'/

■■■•/ ',

»'/•'

(Ill-'

V /

rUV^'^'

*t*,*"

^^^t§^5N#^^Pf^

a

t«siai

'■' .ITWDV

N^r .

^1^

.^i'*

5^ ' k

■'>i90-

:v\'>P' .

<s

p. M"? Ma.rvLLs d^.l.

yJ.GpeerL litK.

PLATOPHRYS DIMORPHUS.n.sp.
(a^ male h. ihtLale.)

t=l

IS)

c5

g

■H 'H
(D <l

I

<

0
w

X

9^

0)

—I

0]

[/]

^

o

:§

ra

c8

O

tu:

V

13

,-a

^

CD

o

CO

a;

0)

I

4J

[f]
o

•H O

--^ -r-H

U] .Pi

CD :j

^ o

I

-§

to

C/]

<
ft

W
O

o
<

pq

CO

D
y}

CO

O

o
o

X

CO
In

WW

6

1^

\

■Si

o

-J— '

CO

"OO

a

(—*

^

-i-1

ca

in

<

>

f"^

f^

'^

OJ

c

CO

ol

\^,\n¥

%v-.^

"""^^\

•v; .

a,

¥

^

i /

^ ^ , /

CO

; ^

1 — 1

Oh

"■fr

rlJ

o

o

::P, •

<

"' t ^

0}

- .^ ,

p

1^ -'"

CO

.%'''y'

w

■ ' %^

o

m/'

H^

'fwC

o

my y

o

, M/-/

%

#1^'

tf

'■¥f

<:

Ah

X

t'j
•1-

fJ]

P-

K

H
O

H
<

WD

O

-ij

rd

f^

O

'OC

•M

■r-l
4-3

m

-<

P

^

>

r^

,^

4^

'r-^

o

O

p: c/Q

■i:-<

r^i

■a

^

w

2

C ■

^

:r«!-

..■':,.^^:-&

;? ■ -ife^.

.^■^-"■f:

^.^, ::,,-..;:

i^iP:,- : '^:i5|:^p-^:'^'

.-sf;*^'"

:t%\^

-\-:i«-
x:^'
S ;-"'
i'^'

Oh
W

I

o

o
o

..-y

0]

f^

0

•r-i

-P

,a

■r^

o

-P

•H

U]

,f^

(D

^

^

^

-P

CD

H

P

5

c/3

I

■1

^

^

>:

CO
(D

•H

[/]

p^

o

■1—1

4-3

'OJD

^

■H

r >

+J

■H

m

^

nj

<

+3

a)

^

H

o

•H

CO

'<-1

a

I — 1

^

in

^

m

<

<
1—1

^ ■

.-J S

P
H

(i)

v<i ■CO'

w

OJ

■H

D

<
H

O
P^

.^

H

s

o

hJ
SI

o

0]

o

•r-l

^ G

'-3 ■'^

C^ o
00

^

s

i

Marine Iirve stigacbioTis
Sou-th. Africa. .

TisKesTl.}(XXVl

P^M9 Man-as del

cT. Green litK.

SCOPE LTJS ARGENTEUS.n.sp.

MARINE ANNELIDS (POLYOHJITA)

()!■■

SOUTH AFRICA.

PART I.

BY

WILLIAM CARMICHAEL McINTOSH,
M.D., LL.D., F.R.S., &c.,

Professor of Natural History in the University of St. Andrews, Director
of the Museum and the Gatty Marine Laboratory in the University.

CONTENTS,

Introduction

Uiiplirosyiic capeiisis, Kinber^

Lepidonoiiis li'alilbergi, Kinberg

Eiiphioiic tiisabetha.', M'Intosh

Ell lion cnpciisis, M'Intosh ...

Polynoc capciisis, M'Intosh ...

Hemilcpidia tiihcrculata, Schmarda...

Stiicmiais boa, Johnston

Eiilalia -,'iridis, O. F. Miiller

Etcoiw sfetsbcrgensis, Mahngreii

Anfolytii.'i prolifer, O. F. Miiller

Pionosyllis Malingreni, M'Intosh

Syllis cornii fa, H. RM^k^: ...

Nereis (Mastigonereis) vaiicgatn, Grube

Lysiilice capeiisis, Grube

Eunice apliroelitois, PMas ...

Marphysa sanguinea, Montagu

Arabella iricolor, Montagu

Liiiiibriconereis fetraiinis, Schmarda

,, cavifroiis, Grube

I'riplionia eapensis, M'Intosh
I'hibcUigcra liicfator, Stimpson

,, Mareiizelleri (an. var. itf/iiii>

S;ii

PAGE.
19
22 •

25

27
28
28
29
32

34
35

35
35
37

37
40

42

44
46

4'^
5'>
50
52
53
35
B

19
INTRODUCTION.

Previous to 1861, little was known of South African Annelids.
During the three centuries which had elapsed since the skilful
Portuguese navigator, Bartholomew Diaz, made Europeans
acquainted with the Cape of Good Hope, or, even during the
somewhat shorter period since his countryman, Vasco de Gama,
rounded the Cape, this group of marine invertebrates remained
in obscurity. It is true that Pallas made known, by the aid of
intelligent ship-captains and others, a few of its Annelids, such
as Pcctiuaria capcnsis and Lcpicloiiofiis ]Valilberifi (which he
included under Lepidonoiiis sqiiauiatiis — his Apjirodifa sqitanuita,
and a form closely allied to Lepidonotiis clava, Montagu). In 1854,
an account of some Annelids, from the east coast at Mozam-
bique, was published by W. C. H. Peters, but the descriptions
are not sufficiently minute for accurate diagnosis. It was not,
however, till the publication of Schmarda's work, in 1861, that
the richness of the region in this group was disclosed. The
Austrian naturalist described no less than about 50 Marine
Annelids (Polych<neta) from the Cape, and everyone of which
he made a new species. About 18 families of Annelids were
represented in this treatise, which marked an era in the literature
of the subject. Unfortunately, the inferiority of his instruments,
or the lack of artistic accuracy in delineating structure, makes it no
easy task to diagnose some of the species, though the majority are
recognizable — either by description or figures. Kinberg
further added to the information concerning the Annelids of
the region, including Port Natal, in his account of those
collected during the voyage of the Swedish frigate " Eugenia."
A few others were described by Grube from the Austrian
"Novara" Expedition. The voyageof the German ExploringShip
'* Gazelle," enabled Grube to extend his list of African Annelids
— mostly from the region of- the Congo and West Africa,
but also from Table Bay at the Cape. Amongst these were
several European forms, such as Sigalioii Edivardsi (Madeira),
Glycera convoluta, Kefertein, and Pista crisiata, O. F. Miiller.
The " Challenger," again, considerably extended the information
on the subject by obtaining 23 species, nine of which were
dredged south of the Cape, in 98 to 150 fathoms, and the rest
were collected between tide-marks at Sea Point, Cape Town,
and at Simon's Bay.

Since the voyage of the "Challenger," Marenzeller has con-
tributed a notice of the Annelids of Angra Pequena-Bucht,
that is, a bay on the western coast of Africa, considerably to
the north of the Cape (Great Namaqua Land). Khleis, some

B 2

20

years later, still further enriched tiie subject hy an account of
Annelids from the East African region — chiefly south of
Zanzibar. Lastly, an important collection has recently been
made at Zanzibar, by a young and able observer, Mr. Cyril
Crossland, and a report on certain groups has been published.*
When the examination of these is completed, much information
on the Annelids in general as well as of the Geographical dis-
tribution of the group will be available.

Dr. Gilchrist's collection was made between tide-marks, and
probably indicates the most conspicuous forms of the region.
In glancing at the series, it is found that the Polynoid^e,
Phyllodocidfi?, Syllidae, Nereidae, and Sepulidie have but few
representatives, and the Hesionidse are absent, yet all of them
usually are found in numbers under stones, between tide-marks,
in a prolonged search. Their number will probably receive
considerable additions in future. The forms frequenting sand,
or sandy mud, are for the most part absent, either because
digging was not resorted to, or because the region was purely
rocky. The Annelids generally found in the media mentioned
are the Nephthydidaj, Glyceridie, Opheliid?e, Scalibregmidai,
Telethusae, Spionidte (including Mageloiia),3.nd Ammocharidae,
whilst in deep water the Ampharetida; are frequent. Doubt-
less, the groups mentioned will be represented when the trawl
and dredge are used. The trawl-net is especially productive —
not only in regard to the debris of old bivalve and univalve
shells in which Annelids lurk, or have their tubes coiled within
them, but from the frequency with which soft Annelids, both
errant and tubicolar (now devoid of a tube) cling to the meshes.
By the flashes of phosphorescence, even small forms like the
Syllida? and Polycirrus may be secured at night, whilst the
bright colours of others, such as the Nemeteans and Hesionidse
readily distinguish them by day. Many Annelids are minute,
and require careful inspection of the apparatus for their detec-
tion. Special inquiry should also be directed to the forms
which bore in calcareous rocks, in or under crusts of Melobcsia,
in shells, or softer media, which become pelagic, like Palolo, at
the reproductive period, or which are useful as bait ; indeed,
there are few which are not available for the latter purpose.
The purely pelagic types such as Alciopa, Halodora, Phalacro-
phorus, Pelagobia, Haliplanes, and Lopadorhynchits, and the
temporarily pelagic, such as Aiitolytus, may also occur in tow-
nets along with Tomoptetis, and the larval and post-larval stages
of the sedentary types.

Much may likewise be accomplished by splitting the rocks
between tide-marks with hammer and chisel, or using a lever
to overturn large stones and separate shelving rocks — in the

*Proceed. Zool. Soc, 1903, Part I., p. 169, Pis. xvi & xvii., and Part II., Aug., 1903.

21

fissures of which many Annelids lurk. Moreover, some frequent
the tubes of other Annehds, hke the British Polyiioe scolopen-
drina in the tubes of Tetehella ucbulosa, or the South African
Polynoc atfcniiafa, occur in the ambulacral areas of certain
starfishes, or on Echini, between mantle and shell of limpet-
like moUusks, or in Balani, m the interior of the basket-like
Hexactinellid sponges, in the tissues of these and of horny
sponges, in the branchial chamber of Ascidians, and a few are
parasitic in the bodies of other Annelids.

Particular care is necessary in the preservation of Annelids,
which should, as a rule, be killed with corrosive sublimate,
carefully washed, and then preserved in spirit, which should be
changed after a few hours, and again within 24 hours. If the
Annelids are large, the change should be repeated in a few days.
Those intended for section may be treated with Flemming's, or
with Bles's fluid. In any case, it is unsatisfactory to place
many in a single bottle, or jar, of spirit.

When living, a special note should be made of their coloration,
the presence of crustacean ectoparasites on their surface, or
gills, the colour of their circulatory fluid (red or green), the
presence of eggs, in sacs, or otherwise, the issue of sperms, and
their habits and movements, both in nature and in confinement.

One of the most interesting features in Dr. Gilchrist's collec-
tion is the occurrence of so many European annelids at the
Cape, and of others which can scarcely be distinguished from
well-known European representatives. Moreover, some of these
stretch far eastwards to the Indian and Pacific seas on the one
hand, and on the other are found to the westward and the American
shores, for instance, from Canada to Cape Horn. These
Invertebrates seem to be independent of the laws which govern
the distribution of such forms as the food-fishes, for instance,
the common Gadoids,which prefer the colder waters of the north,
yet there is a remarkable coincidence in the complete identity —
even to their crustacean parasites — of annelids frequenting the
widely distant littoral belts of Britain and South Africa. Nor
is this community confined to one group of Invertebrates, for,
amongst others, Mr. Stebbing* has noticed similar features in
regard to the Crustacea. These and other facts lead to the
belief that in the sandy bays of South Africa the hardy plaice
of the North Sea might live and flourish in a noteworthy
manner ; nor would there be more difficulty in its transport
than in the case of Australia. The Turbot, Conger, Whiting
and Sole might also be useful additions. In anv case, and as
might be expected, it would appear that the laws which govern
the distribution of land animals do not hold for all of the
marine. The contrast, for instance, between the mammals and

* South African Crustacea, Part II., p. i., 1902.

birds of South Africa and those of Britain is sufficiently pro-
nounced, and even the fish-fauna of the two is divergent, yet
in the group of the AnneHds not a few are common to both
regions, and as knowledge extends such mav even be found
more general than is supposed.

In the following report, a few collected by the "Challenger" are
indicated in their proper places — occasionally with additional
information. The references under the various species have
been selected as those which, in all probability, may be most
useful.

EuPHROSYNE Capexsis, Kinberg, 1808.

1858. Euphrosyiic capcusis, Kinberg, Ofversigt K. Vet.-Akad.

Forhaudl, d. 14 Januari, 1857. Stockh, 1858.
„ Eiiphrosyiic capcnsis, Id., Fregatten Eugenics Resa. Taf.

12. fig. 14.
1861. Eitplirosync polybranchui, Schmarda, Neue wirb. Thiere,

I. ij., p. 136, Taf. 32, figs. 264-287.
1885. Euphivsyiic capensis, M'Intosh, x\nn. " Challenger," p.

I., PI. II. fig. 5., PI. lA. figs. 1-3.

This widely distributed species, which ranges from the Cape
to St. Paul, is characterised, when contrasted with the British,
by the proportionally longer, more finely spinose and more
flattened body. The dorsal surface is convex, the ventral flat-
tened. The head bears a caruncle, consisting of an elevated
keel and a flat band, which reaches to the eighth segment. In
front of it is a short and somewhat conical tentacle with an eye
on each side of its base. A pair of more minute and in some
almost confluent eyes occur on the ventral surface of the
cephalic ridge. In front of the puckered opening of the mouth
are two kidney-shaped pads, separated by a deep median fissure,
and which may be homologous with the palpi of other families.
These pads are probably of some importance as pivots during
the action of the buccal apparatus. The arborescent branchiae
are purplish with the elliptical or sub-oval tips pale reddish, and
each complete row generally consists of eleven trunks. The
dorsal bristles are of two kinds — smooth and crenated in the
fissure, whilst all the bifid ventral bristles are smooth. The
size of the largest example agreed with those found by the
" Challenger " under stones between tide marks at Sea Point,
Cape Town, and the British EtipJirosyne foliosa has the same habi-
tat in the Channel Islands, The pigment in both probably
subserves other functions than either ornament or protection.

A typical segment of the body consists of a convex dorsal and
a flattened ventral arch. The former is bare in the middle line^
but has laterally a superior division carrying a dense series of
bristles and a posterior row of branchiae — with a cirrus at the
dorsal margin and another midway between the bristles and the
branchiae. The latter is the longer and has a slight median
constriction. Neither tapers much. The inferior division has
a tuft of bristles, andinferiorly and posteriorly a cirrus, which is
generally more slender distally than either of the foregoing.
All are shorter and thicker than in the British EnpJirosyne foliosa.
The dorsal row of bristles is comparatively short in contrast
with the northern E. borealis, and are even less boldly marked
than in Euplnosync foliosa. The curve of the tip of the bristle
is less pronounced than in the latter, and the disproportion
between the processes more distinct. The notches on the limbs
of the fork are due to transverse grooves. At the dorsal edge
as well as in the centre of the row are many bristles with a
smooth fork, the longer process being much attenuated — so that
they project beyond the serrated kind. The ventral bristles are
considerably larger than the dorsal, and are terminated by a
slightly curved blunt tip, with a rounded spike at the base.
The central canals from the processes coalesce after a short
course downward, and a slight dilatation occurs opposite the
enlargement at the upper end of the shaft.

In Enphrosyne and Cliloeia the bristles are characterised by
their extreme brittleness, their tubular condition, calcareous
nature and the entrance of air into the interior.

The branchiae, which are eleven in number in the typical
segments, form a densely ramose series on the dorsal region of
the segment. Each is dichotomously branched, and the tips
end in elliptical or sub-oval bodies resembling those of EupJiro-
sviic foliosa. So far as observed the branchiae are devoid of a
cavity, and the blood-vessels can be traced from the body-cavity
up to, but not into, the tips of the organs. A complex series of
muscular fibres appear at the base, and the cuticle and hypo-
derm are dense, except distally — where the former becomes very
thm.

In the structure of the body-wall the species offers no
peculiarities. The nerve-cords superiorly have a firm invest-
ment which is continuous from side to side. In the hollow
between them superiorly is a fascicle of muscular fibres and
below them is a central granular structure. Moreover, the cords
are united by a striated isthmus inferiorly.

The anterior part of the buccal apparatus or proboscis is
cylindrical, ensheathed in cuticle, and protrusible. The centre
is occupied by a large muscular and vascular, but chiefly
glandular mass, the inner lining of which is thrown into bold

24

longitudinal rugae. The vast collection of glands would indi-
cate important secreting functions. Behind is a remarkable
lingual organ, which in longitudinal section resembles the
tongvie in mammals. It is attached at the base posteriorly to
a dense glistening muscular mass, separated by a layer of hori-
zontal fibres from the complex series of radiating muscles which
diverge upward to the periphery of the organ. In vertical
longitudinal sections in the middle line, these interwoven fibres
posteriorly are separated, by septa passing downward from the
thick chitinous coat, into spaces which enlarge anteriorly. The
muscles at the tip of the tongue arch backward over the posterior
part of the preceding glandular region (which forms a high
vascular cushion), and pass downward to the front of the dense
basal glistening mass formerly alluded to, the curve of the arch
becoming less and less till the fibres are nearly vertical behind
the cushion, or at the commencement of the smooth and dense
cuticular region of the tongue. The chitinous coat of the organ
is so largely developed at this part that it is as thick as the
hypodermic layer. In this region also are three strong plates of
transverse muscular fibres extending from summit to base. If
the section deviates to the lateral regions, the decussation of
the fibres is very complex. In transverse sections, again, near
the tip, the thickness of the hypoderm on the rug^s of the pre-
lingual mass as well as in the lingual papillae is considerable.
In the centre of the tongue is an inextricable mass of crossed
and interwoven fibres, the base assuming a somewhat stratified
appearance, since the blood-vessels lie at intervals in regular
horizontal rows. The tip of the tongue is highly vascular,
indeed, the organ throughout is richly supplied with blood-
vessels. The dorsum of the free tip is covered by a series of
elevated papillae — truncated at the tip, and with one angle
pointed. They are processes of the hypoderm sheathed in cut-
icle. Nothing can exceed the complexity of the muscular
structure of this organ, which is eminently calculated to sub-
serve important functions. Probably the animal erects its lingual
process and applies it with a rasping action to the surface of a
sponge, the successive pieces being crushed against the firm
rugose wall behind, by the hard posterior surface of the tongue,
and again subjected to the play of the deeply folded or lamellar
region behind it. Before the food reaches the intestinal surface,
it has been partially disintegrated and litted for absorption,
especially soft tissues like those of sponges. Minute shells of
annelids, such las Spirorbis, are of course less affected, though
many are broken into fragments, and all are in a condition to
give up their contents.

Behind the tongue the alimentary canal presents on its
ventral aspect a large and a small fold, the former in longitudi-

25

nal section often having a broad summit applied to the vault of
the canal and a narrow base, so that it resembles a pedicled
mass. This region has numerous transverse folds, so deep in
some cases as to merit the name of lamellae (like those in the
ruminant's psalterium or " manyplies"). The canal then curves
upward Cfrom the posterior pit) and forms a kind of pylorus,
with thick walls and deep furrows, and terminates in the intes-
tine, into which the part freely projects. The thick rugose
intestine, which agrees in structure with that in the ordinary
Annelids, terminates in a dorsal anus, with two cirri at the
sides inferiorly.

Some of the examples were laden with ova which in the pre-
parations were of a pale orange hue and measured '1524 mm.
in diameter. They floated freely in the perivisceral cavity. No
evidence existed as to their mode of escape. Other examples
seemed to have spawned. Schmarda thought there were special
oviducts which opened near the anus, but no trace of these
was observed.

Like the British species at Herm, the South African form
would seem to browse on the patches of Halichondria and
other sponges which abound under stones between tide-marks ;
and there is no wonder therefore that there should be so many
interesting homologies between its buccal apparatus and that of
such Nudibranchs as Doris tuhercnlata and Doris Johnstom,
which follow precisely similar habits, In some examples,
moreover, many fragments of crustaceans, numerous portions
of cells and avicularia of Polyzoa showed that occasionally the
dietary was varied.

The examples were procured between high and low water-
mark at St. James', False Bay.

Lepidoxotus wahlbergi, Kinberg, 1858.

1 861. Polviioc trocJiiscoplioni, Schmarda, Xeue wirb. Thiere

I. II. p. 151, Taf. xxxvi., fig. 310.
1885. Lcpidonotus ivahlbcigi, M'lntosh, "Challenger" p. 66,

PI. xi., fig. I., PI. xviii. fig. 8., PI. xa., figs. 15, 16.
1898. Lcpidoiwtiis waJdbergi, De St. Joseph, Ann. Sc, Vol. 8c.

sec. v. p. 225, PI. xiii., figs. 2-8.
1898. Lcpidoiiofiis ivnJdbcigi, M'Intosh, Ann., Nat. Hist. Sc.

'j. Vol. II. p. loi, PI. ii. figs. II and 12.

A further consideration of this form, afforded by Dr.Gilchrist's
specimens, inclines me to the belief that Schmarda's species is
identical, and it is possible that there is a closer relationship

26

between Stimpson's Lcpidoiiotiis sfinitccta" than is hinted at in
the Challenger volume,! so that this author's title may yet claim
priority. Kinberg's examples came from the Cape as well as
Port Natal, Schmarda's from Table Bay, while Stimpson pro-
cured his from Simon's Bay, and the present examples were got
between tide marks at St, James', in False Bay.

The examples from the "Challenger" were larger than those of
the present series, a male measuring 36 mm.

The head agrees generally with the type in Lcpidoiiotus. The
median tentacle is somewhat longer than the palpi, dilated at
the tip, and with a filiform termination. The lateral tentacles,
though small, correspond in structure. In the males, the ventral
papilhe of the body are longer than in the females, in which
they are tulip-shaped.

Schmarda's figures, presumably from life, though details are
somewhat doubtful on this head, show the middle of the dorsum as
quite bare from the head to the tail, yet he does not indicate
the structure of the head thus exposed, the yellowish hue of
the dorsum passing forward to the base of the tentacles, a con-
dition at variance with nature. On the whole, the structure of
the foot, the scales and the bristles, as shown by this author, do
not offer any objection to the union of the species with
Kinberg's. Schmarda observes that the proboscis is brown and
yellow, and that the intestine (which he figures) has sixteen
pairs of caeca.

The scales are generally covered with sand-grains, mud,
thread-like alga^, Infusoria like Salpiiiga'ca, with, here and there,
an example of Loxosonia.

The species seems to be widely distributed round South
African shores, and takes the place, for example, of such a form
as Lepidouotiis chiva, IMontagu, in Britain. As Kinberg observes,
the South African form is probably included in the Aphrodita
squanuitd of Pallas, who conjoined with it the European
species as figured by Baster.

Baron de St. Joseph, who has done such excellent work
amongst the Annelids of the French coast, was of opinion that
this species was only a variety of Lipidoiiofiis cUiva, Montagu,
but as shown in the Annals of Natural History in 1898,! a con-
siderable margin for variation will be necessary before this can
be established. De St. Joseph's main reasons for the union of
the species are ( i ) that the figure of the scale in the
"Challenger" volume exactly agrees with Lcpidoiiotus clava,
and (2) that m the same work it has been shown that the palpi
are papillose, a feature, also, characteristic of Lcpidoiiotus dura.

'■■ lYodr. Expcd. Ocean. Pacif. Septen., &c. Proceed. Acad. Nat., &c.

Philad., July, 1X55.
t p. 67. X ^t-r. 7, vol. II., "p. loS, V\. ii., ti.us. i I and I2.

27

But in the latter species, only the lirst tour scales have the large
tubercles which are visible under a lens, whereas, in Lcpidonotus
Wahlbcrgi, every scale has them. Yet this is not an impassible
barrier to their union — it in Lcpidonotus Wahlbcrgi prominent
and acute spines had not been present on several of the anterior
scales. The tips of these spines are slightly bent, and are
minutely nodular, so that the surface in some views presents a
slightly areolar or scaled appearance. On the other hand, the
truncated capstan-like tubercle, characteristic of the lirst scale of
Lcpidonotus chiva, has the head minutely and somewhat regularly
nodular, the tip of each process having a spine. When viewed
from above, the rim of the head is thus minutely hispid. The
arrangement and structure of the other papilla.% on the re-
spective scales, also differ. Further, the dorsal bristles of
Lcpidonotus Wahlbcrgi have a stouter tip, with a definite curve,
and is much less elongated and tapered than in Lcpidonotus
clava. The ventral bristles, again, have a more elongated tip,
with a longer row of spinous ridges. From point to point,
therefore, the European and the African forms differ, even the
scales in the latter being distinctly smaller.

Yet it cannot be doubted that a close approach, in many
respects, occurs between the two forms — separated by so many
thousand miles, and in view of the conditions found in other
South African Annelids, it may be that the differences pointed
out in the foregoing paragraph, will suffice only to constitute a
local variety. Chitinous structures of the nature of spmes on
scales, of bristles and the parts of the dental apparatus, are
prone to vary, and intermediate forms of Lcpidonotus clara may
yet be obtained, which would warrant the union of this form
with Lcpidonotus Wahlbcrgi.

ErPHioXE ELiSABETH.^i, McIntosh, 1885.

iHH^,. Euphionc Elisabcthcv, Mcintosh, Ann, "Challenger,"
p. 62, PL ix., fig, 3, PI. xvii., lig. 7, PI. xviii., lig, 10, and PI.
viii. A., figs. 3 — 6,

This striking foim was procured by the dredge south of the
Cape of Good Hope, in 150 fathoms, on a bottom of green sand.
It is about 35 mm, in length, and its breadth at the widest
part (including the bristles) is 23 mm. The chief features
are the correspondence of the head with that in Lcpidonotus,
only the eyes are almost connate on each side. The smooth
tentacula and dorsal cirri have an enlargement at the tip, and a
constriction beneath it. Scales — 13 pairs, as in Iphiouc, flexible,
leathery, the first being rounded, the rest somewhat rhomboidal,
the anterior edge, however, being incurved, the posterior

28

convex. Their surface is studded on the inner region with
rotate papilla?, some small and adpressed, others large, and
elevated on a pedicle. Externally are softer elongated spinose
papill?e, which, in shape, resemble a prickh^ pear ; margin of
the scale ciliated ; dorsal bristles, pale, silky, with a very
attenuate tip — hispid with opposite spines ; ventral bristles, dull
yello^ ', with simple, slightly hooked tips.

So far as known, this form has hitherto occurred only in
the region mentioned, and is one of the most characteristic
Annelids of the Cape.

EUNOA CAPENSIS, McIntosh, 1885.

1.S85. Eiiiioa capeusis, Mcintosh, Ann. "Challenger," p. 72,
PI. XV., f. 5 ; PI. xi A., figs, i — 3.

A fragmentary form, procured between tide-marks at Sea
Point, Cape Town, along with Sabcllaiia, Tcrebella, and Sabclla,
Unfortunately, the specimen had no scales, but it will be re-
cognized by the following features : — The head is characterised
by the large size of the base of the median tentacle ; the lateral
tentacles are short, with a filiform tip passing abruptly from the
nearly cylindrical organ. Palpi of moderate length, and
papillose. A pair of small eyes occur at the outer and posterior
border of the head, whilst a pair lie behind the anterior border,
and lateral in position. The tentacular cirri have a slight
enlargement below the filiform tip, and the cylindrical column
is papillose. The dorsal bristles end in a somewhat blunt
rounded process, as in Einion liispaiiica* and an allied bristle
is seen in Kinberg's Antiiioc pukhclla,] from the mouth of the
La Plata. The ventral bristles have comparatively short tips,
with a simple terminal hook.

POLYNOii CAPENSivS, M'Intosh, 1885.

1885. Polynoc Capeusis, M'Intosh, Ann. '•Challenger," p. 114
PI. iv. lig. 4, PI. XV. fig. I, PI. xix. fig. 4, PI. ixa. figs. 4, 5,

Procured by the " Challenger " in the dredge south of the
Cape in 98 fathoms.

Head elongated from before backwards. Two rather small
eyes at the posterior border and two largei- in front of the lateral
prominences. The tentacular and dorsal cirri have a tendency
to assume a fusiform condition, and all are smooth. The ven-

* Trans. Zool. Soc. Lond. ix., p. 396.

+ F'reg. Eugen. Resa., p. 20, Tab. vi., I. 29, Gs.

tral cirri are long and large. Body moderately elongated.
Scales 15, ovoid, blackish grey, and smooth, with the exception
of a triangular group of blunt tubercles near the anterior notch.
Foot somewhat short with feebly developed pale dorsal bristles
having many thecate infusoria parasitic on them. They have a
short, peculiarly curved tip with line rows of spikes. The
ventral bristles have a well-marked hook at the tip and a
secondary process beneath, besides a spinous edge. The
bristles on the whole approach those of Halosydna, whilst the
head somewhat resembles that of Lepidonotus.

Hemilepidia TUBERCULArA, Schmarda, 1861.

1861. Heniilepidia tiiberculata, Schmarda, Neue wirb. Thiere,
I., II., p. 149, Taf. xxxvii., fig. 317, and woodcut.

1885? Polynoc atiennata, M'Intosh, Ann. "Challenger" p.
120, PI. XV., fig. 2., PI. XX., fig. 9. PI. xia. figs. 8 and 9.

This species was first described by Schmarda as having been
procured in Table Bay in shallow water under stones. The
present example was obtained between tide marks at St James',
in False Bay, 5th October, 1902.

It measures about 45 mm. in length, while its widest part
anteriorly, inclusive of the bristles, is 7 mm. A considerable
amount of coloration still remains along the centre of the dorsum,
from the head to the tail, and it is somewhat symmetrically
arranged throughout, especially in the posterior region which is
devoid of scales. The sides of the body anteriorly covered by
the scales, are pale as likewise are the feet, the bases of the
tapering dorsal cirri, however, being set in a ring of dark pig-
ment with a crescent attached in front, whilst the organ itself is
coloured about half its length. The bare posterior region again
is prettily marked with light and dark touches, making a surface
almost like chain-work. Along the centre is a row of papilla
elongated from before backward and dark coloured, a bar of
pigment tapering from each side and passing between the two
nearest lateral papillae, since they are alternately arranged with
the median. Their tips are pale, whilst a triangle of dark pig-
ment passes inward from each, the apex entering the pale
region in front and rear of the central papilla, and thus behind
the middle of each segment. Externally a dark patch intervenes
between each papilla and surrounds the base of the dorsal
cirrus, a little pigment also appearing on the basal part of the
column of the organ. Pigment of an iridescent dark brownish
hue tints the ventral surface for nearly two thirds of its posterior
moiety, each segment having two long lozenges on each side,
the posterior pair, which have their broad bases internally, being

30

specially distinct. The neural region is distinguished by its
darker hue in front where no lateral pigment is, and hy its pallor
in the middle of the darker posterior region. A touch of pig-
ment marks the base of each ventral cirrus at the anterior part
of the coloured region, then various bars of dark pigment occur
on the under surface of each foot, the whole having a character-
istic arrangement.

The head is slightly leaden in hue, with a deep notch filled
by the brown base (ceratophore) of the median tentacle in front.
Two eyes of moderate size lie posteriorly near the nuchal
border, whilst two less distinct occur towards the anterior
angle, though they were so hidden by parasitic Loxosomixi that
doubt remains as to their precise position. The median ten-
tacle is comparatively short, little tapered, and ends in a slender
filament. The lateral are typical in length and also have
slender terminal processes. The palpi are marked by longi-
tudinal cirri closely arranged, especially ventrally. The tentacu-
lar cirri have the same shape as the median tentacle.

The body is elongate, slightlv narrowed in front, remains for
some distance of nearly equal diameter and then gently dimin-
ishes throughout more than the posterior half towards the tail,
which has two short cirri. Yentrallv a median lozenge of pigment
passes to the mouth. The anterior third of the body has
numerous Loxosoma? between the feet and at the tentacular
cirri. The bristles project on each side as pale yellow fasciculi.

The first pair of scales are paler than the succeeding, but
have a similar pattern, viz., an ovoid pale patch at the scar, with
a dark band traversing it. The broad crescent of pigment
along the inner border, however, is less marked both in super-
ficial extent and tone than in the succeeding scales. They
adhere with considerable tenacity. In outline, the front pair
are more or less circular with a projection opposite the pale
region. The others are somewhat ovoid, the pale region being
at one end, and the bar by-and-by becomes a spot. Though
apparently smooth to the naked eye, minute and somewhat
clavate papilla? are scattered over the surface, here and there
appearing beyond the margin, as near the white spot (where
they are specially numerous), and other parts.

The pale outer region ha^' numerous small clavate papilla?,
some of which extend beyond the border, and a few short
papillae also project at intervrtls from the inner border.

The typical foot resembles that in Poh'iioc scolopciidriiia,
Savigny, 3^et Schmarda gives "pinna unica," as one of the
specific characters, though he probably means apparently
single. It really has two divisions. The foot forms a short
cone with a bevelled tip (Plate I. fig. i). The dorsal cirrus is
comparatively short and slightly tapered, with a slender filiform
tip. The surface has numerous short clavate papilla? which are

3^

longest towards the tip of the organ. The dorsal division forms
a separate elevation with the spine-papilla (pierced by the spine)
at its outer and lower border. It carries a group of transparent
slightly curved spinous bristles {Plate II., tig. 2, one of
the shorter), which Schmarda figures as bilid terminally, and,
which at first sight have this appearance, but the condition
is only apparent, as more careful examination shows that
the tips have been fractured, and, that in many, the split
parts give a pseudo-bifid aspect. The shorter forms at the
inner border of the fan, which have been shielded by the
longer, have simple tips tapering to a point. The latter is often
coated with a finely granular investment. These bristles are
more slender and elongate posteriorly. They are evidently
brittle, and, if the species, like Polyiioe scolopcndrina, inhabited
a tube, and was commensal, the fracture of the tips would he
explained.

The mass of the foot is formed by the ventral division, and
the great yellow spine is more powerful than the superior,
whilst its papilla is posterior. In front, another and longer
conical papilla projects outward. The first or upper bristle is
much stronger than the others, of a deep yellow colour, and
hastate (Plate II., fig. 3, from the 40th foot), yet there are
indications that it is only a modified form of the series which
follows, for traces of rows of spines are present — especially in
the anterior and posterior parts of the body, and even the tip
occasionally shows a minute spike, or secondary process. At
the loth foot, for instance (Plate II., fig. 4), it is only a some-
what larger bristle, with a long spinous simple tip. At the 30th
and 40th feet, on the other hand, it is truly hastate or javelin-
shaped, and nearly symmetrical, yet a trace of the spinous
rows occurs on the left. These are barely visible at the 50th
foot, but, again re-appear in the bristles of the caudal region.
All the bristles below the larger conform to the same type,
having straight translucent and slightly yellowish shafts, the
terminal region being enlarged ^t its commencement, dimin-
ishing distally till it reaches the bifid tip, where a slight
dilatation again occurs (Plate II., fig. 5). Rows of spikes
appear on the upper edge of the tip. These bristles have longer
tips in the loth foot, and they diminish in length and increase
in strength for some distance backward, and remain short in
the caudal region.

The ventral cirrus is short and subulate, the tip reaching
only a little beyond the bases of the nearest bristles. Its
surface, with the exception of the filiform tip, has clavate papillae.

In all probability, the habit of this species is akin to that of
such forms as Polvnoc scolopeiidriiw, being a commensal with
another Annelid, mollusc or echinuderm. It is doubtful, if

32

Schmarda's Hcmilepidia eiythrotcenia* is not a variety of tlie
same species.

The example is a female with ova far advanced.

The relationship of this species to the Polvnoc attenuata of
the "Challenger" expedition,! is so close, that it may be
necessary to unite them. The " Challenger " form was obtained
between tide-marks at Sea Point, near Cape Town, along with
Sabellaria, Tcrebella, and Sabella, and from its great length and
general appearance, it resembles a commensalistic form,
probably living in the tube of a Tcrebella. The two forms
agree in general outline of the body, and in the number of the
scales, those in the specimen from the " Challenger," however,
being more distinctly papillose, especially at the margin, yet, it
is only a question of degree. In the specimen from the
" Challenger," no large bristles, such as are figured in this
paper, were present, and the fact was commented on at the
time. Such may be a variation, or an exceptional condition,
for the other bristles closely correspond.

Sthenelais boa, Johnston, 1833.

1833. Sigalion boa, Johnston. Loud. Mag. Nat. Hist.vi,, 332,

f. 42.
iS^g. Sigalion boa, Id. Ann., Nat. Hist, ij.,439, PI. xxiii. figs. 6-15,
1843. Sigalion Idnncv, H. Rathke. Nova Act, Acad. Caesar.

Nat. Cur. xx., p. 150, Tab. ix., f. i — 8.
1 85 1. Sigalion Iduncv, Grube. Fam. Ann. p. 38.
1851. Sigalion 6ofl, Williams. Rept. Brit. Assoc. 1851, p. 201.
1851. Sigalion Idnnw, Sars. Nyt. Mag. Natur. xi. 3, p. 264.
1861. Sigalion Iditnce, Id. Vid. Selsk. Forh., 1861 (sep,

copy), p. I.
1865. Sigalion Iduncv, Malmgren. Nord. Hafs.-Ann. p. 86.
1865. Sthenelais Idnnoe and S. boa, De Quatrefages. Ann. i.,

p. 276,
1865. Sigalion boa, Johnston. Cat. Brit. Mus., p. 124, pi. 13, f.6.

1873. Sthenelais Idnnoe, Sars. Bid. Christ. Faun., p. 14.

1874. Stiienelais boa, Mcintosh. Invert, and Fishes St. And.,
p. 118.

1875. Sthenelais Idnnoe, Ehlers. Ann. " Porcupine," 1869,
ibid., p. 18.

1876. Sthenelais boa, Mcintosh. Trans. Zool. Soc. ix.. p. 390.
1876. Sigalion Idnna, Tauber. Ann. Danic. 83.

1880. Sthenelais Iduncv, Langerhans. Zeit. f.w. Zool. xxxiii.,

p. 276, Taf. xiv., f. 6.
1883. Sigalion Idnnoe, Levinsen. Nord. Annul., p. 199.

* Op. cit. p. 150, Taf. xxxvii., fig. 318.
t Op. cit. p. 120, PI. XV., fig. 2 ; PI. XX., fig. 9 ; PL \\(i., figs. 8 and 0-

33

1884. Sigdiion Ichimv, V. Cams. Faun. Medit., p. 205.

1888. Sfheiiclais Idiiiia, De St. Joseph. Ann. d. sc. nat. (7),

v., p. 187, PI. viii., f. 55.
1890. Signlioii IdiiiKv, Malaquin. Ann. Boulon., p. 23.
1900. Stiieiiclais boa, Mcintosh. Brit. Ann. II. (Ray Soc),

p. 409, Pis. xxvi., f. 7 and 8 ; xxvi<7. f. 21 and 22 ;

xxix., f. I ; xxxi., f. 5 ; xxxiii., f. 16 ; xli., f. 19 — 23.

Procured between tide marks at St. James' and False Bay,
19th July, 1901, and 25th June, 1902, the former being the
larger. A few about 60-90 mm. in length have a breadth,
over the bristles at the widest part, of 5-7 mm.

The head is somewhat shield-shaped, broad in front and
narrower behind, with two eyes in front, one on each side of
the ceratophore of the median tentacle. Another pair lie in
front on the anterior border, and thus are not readily seen from
the dorsum. The median tentacle is comparatively short and
subulate in outline. A ctenidium occurs at each side of its
base. At the posterior border of the head is on each side a
truncated papilla. The structure of the first feet, which occupy
the usual position in front, and that of the palpi is typical, a
large ctenoidal flap ensheathing the inner base of the latter —
which are long tapering organs.

The body is a little diminished in front, remains for some
distance of nearly equal breadth, and then diminishes gradually
towards the tail which bears two slender cirri of moderate
length. The dorsum is mottled with the touches of pigment on
the numerous scales, which have, in those best marked anteriorly,
a madder-brown crescent behind the scar, joined by a process
which runs into another crescent just within the upper and
posterior border of the scale. In passing backward, however,
various additional pigment-touches appear, so that the pattern
becomes more complex, some presenting three spurs from the
anterior to the posterior crescent, besides a broader spur beneath.

In structure they appear to agree in the closest manner with
Sthendais boa, Johnston, in regard to outline, papilla? and cilia.

The typical foot likewise corresponds with the European
form in every minute detail.

It is noteworthy to find a variety of the well-known British
species frequenting the shores of South Africa, and which does
not in any essential feature differ from it. The wide range of
the species, however, makes this less remarkable, for it not only
frequents the shores of Norway but stretches far southward
along the western shores of Europe and into the African section
of the Atlantic at Madeira.''' A closely allied, if not identical
form occurs at Bermuda.!

■■' Lan.uerhans, Zeit. f.w. Zool., Bel. xxxiii., p. 276, tiu". 6.
t Trans. Conn. Acad, x., p. 666, ujoo.

C

34
EULALIA viKiDis, O, F. Miillcr, var. capensis, Schmarda.

1800. Die i^ri'inc Xeyeide, O. F. Miiller. Naturges. cin warm

arten, p. 162, Tab. 11, fti^, i.
1820. Nereis viridis, Savigny, Syst. Ann. p. 45.
1834. Pliyllodoee elavigent, And. and Ed. Annel. p. 226, PI.

va. rigs. 9-13.
1840. Phvllodoce viridis, Johnston, Ann. Nat. Hist, iv., p. 228,

PI. vi., figs. 1 1 -1 5.
1843. Eulalia viridis, CErsted, Ann. Dan. Consp. p. 27, hgs.

22, 30, 35, 86, and 88.
1849. Pl'yllodoce viridis, Leuckart, Arch, f. Naturges, 1849, p.

202.
1851. „ „ Grube, Fam. Ann. pp. 56 and 129.

1865. „ „ Johnston, Cat. B.M. p. 178, PI. xvi.,

figs. 11-15.
1861. Eulalia capeiisi<,, Schmarda, Neue, ^virb. Thiere, I. ij.,

p. 86, PI. xxix., fig. 231.
1867. Eidalia viridis, Malmgren, Nord. Hafs- Ann. p. 98, Tab.

^v., fig. 39.
1874. „ „ M'Intosh, Invert, and Fishes, St. And.,

p. 120.
1885. „ capensis, Id. Ann. "Challenger" p. 168, PL

xxvii., fig. 7, PL xxxii., fig. 10, PL xiva.,
figs. 16 and 17.
1888. „ viridis, De St. Joseph, Ann. sc. nat. (7) v. p.

283, PI. xii., fig. 154.

Obtained between tide marks at St. James', in False Bay, 5th
August, 1902.

A specimen of average length — 55 mm. — and of the usual
proportions.

The chief features are the size and general signs of luxuriant
growth, as shown in the larger dorsal and ventral lanielLe, and
the larger bristles. Yet in all essential characters of proboscis,
feet and bristles, though the proportions slightly vary, as indi-
cated minutely in the account of the " Challenger " Annelids,
the distinctions are so slight as to warrant its being placed as a
variety of the European form.

This species aboinids not onlv ui European waters, but
extends to Madeira (Langerhans*) and to Angra Pequena —
Bucht (Marenzellert).

* Zeitsch, f.w. Zool. Bd. x.wiii. p.309. t Zoolot^isch. Jahrb. Bd. III. p. 5 (sep. abdr.).

35

ETEOXli SPETSBEKGEXSIS, Millmgicn, 1 867.

1867. Eteonc spdshcrgciisis, Malmgren, Nord. Hat's.- Ann.
p. 102., Tab. XV. fig. 38.

Found between tide-marks at St. James, False Bay, 2nd Sep-
tember, 1902.

This form extends from Europe to the shores of America,
and again to those of South Africa.

The specimen is hunger than the finest example from St.
Andrews, or Lochmaddy, representing the eastern and western
shores of Britain. This free growth affected the size of the
lamellae and the foot generally — without altering the characters
of the parts or the bristles. It is chieflv a dweller in sand.

AUTOLYTUS PROLIFER, O. F. Miiller, 1788.

1788, Nereis pwlifcni, O, F. Miiller, Zool, Danic II, p. 15,

Tab. 52, f. 5—7.
1862. Polybc^fricliiis Mi'iUeri, Keferstein. Zeit. f. w. Zool. xii.,

p. 113, 'i\ab. xi., f. I — 6. (male)
1855. Saccoiiciris JiclgoUindicn, Max Miiller. Arch. f. Anat, u.

Physiol., 1855, p. 18, Tab. II. (female)

It occurred amongst the debris of other forms from St. James',
False Bay, 8th July, 1902.

The example was minute, about 4 mm. in length, yet it had
all the specific characteristics of head, eyes, and cirri. The
compound bristles, also, were essentially the same as those from
Britain and other European localities.

The range of this form is, therefore, very wide — stretching
from the shores of Norway, where the distinguished author of the
"Zoologia Danica " found it budding, to those of Madeira and
South Africa on the one hand, and from the Channel to the
American coast on the other.

PlONOSYLLIS MALMGREXI, McIntosh, 1 869.

1869. Pioiiosyllis Mdhiigiriti, IMcIntosh. Trans. K. S, E. xxv.,
p. 414, PI. xvi., f. ID.

A small Syllidean, measuring about half an mch, oi- a little
more, in length, with rather long cirri, which aie distinctly
articulated.

Procured between tide-marks, St. James, in False Bay, 8th
July, 1902.

v2

36

The head is typical in regard to palpi, and the position of the
eyes, the posterior pair of which are nearer each other than
the anterior, and thus lie obliquely forward ;uid outward from
the posterior pair. The median tentacle is somewhat shorter
than the longest cirri in the example. The body is proportion-
ately stout, tapering posteriorly towards the tail, which ends in
two long cirri. The sides are lianked by the curved cirri, which
cling firmly to their bases, thus differing from the ordinary
forms, in which these organs are readily detached. They are
long, gently tapered, and distinctly articulated from base to
apex, the number of segments ranging from below 20, to 29 or
30. The setigerous region is somewhat elongate, with a bifid
tip, a small papilla at the end superiorly, and a group of stout
bristles, which are pale, slightly curved, dilated, bevelled at
the tip of the shaft, and furnished with a moderately elongated
terminal piece, which has a minutely bifid tip, after the manner
of Pionosyllis, and a spinous edge beneath. None of the
swimming bristles were present.

The ventral cirrus extends fully as far as the tip of the
setigerous region.

So far as observed, this form agrees with the British species,
though, perhaps, the bristles are less numerous.

It is interestmg that a form first found at Lochmaddy, North
Uist, should stretch to the shores of South Africa. Langerhans,
however, procured no less than four species of the genus at
Madeira.

Schmarda found at least nine representatives of the Syllids.^
at the Cape, but it is not possible to identify any of them with
this form. His Syllis cJostcrobvaiicliia has jointed bristles with
an elongated bifid terminal piece, but the dorsal cirri are thick
and short. No other form described by him approaches it,
though several of his species had compound bristles with
bidentate tips.

A single Syllidean {Syllis capcnsis) was obtained by the
"Challenger," at Station 141, south of the Cape of Good Hope,
in 98 fathoms, on a bottom of green sand. Though devoid of
the long capillary bristles Malmgren associated with the genus,
and having moniliform cirri, it would seem to approach
Pionosyllis very closely, indeed, in the proofs it was so termed.
In the structure of the compound bristles, it approaches the
present form, as it also does in the general structure of the
head, and in the absence of long simple bristles, but it diverges,
in so far, as its dorsal cirri have only eight or nine segments,
whereas, Pionosyllis malnigreiii has as many as 20 or 30. How
far age affects the number of these organs is, however, an open
question ; certainly the forms are nearly allied.

h7

Syllis cokntta, H. Rathke, 1843.

1843. Syllif. conintci, H. Rathke (Beitr. z. P^aiina Norw.),
Xova Act. Acad. C.L.C. Nat. cur. xx. I. p. 164, Tab. vii.

fifT T '>

1867. Syllis conintd, Malmgren, Ann. Polych. p. 4^, Tab. vii.

1869. Syllis coniiifd, M'Intosh, Trans. R.S.E., xxv. p. 415, PI.

xvi. fig. 14.
1869. Syllis coniutd, M'lntosh, Trans. R.S.E., xxv. p. 415,

PL xvi. fig. 15.
1885. Syllis coniiifti, ]. V. Cams. Faun. Medit., 1. p. 228.

Found in the debris of specimens of Ddsvbi-diulnis from St.
James'. False Bay, between tide-marks, 2nd September, 1902.

The head agrees with the tvpical Svllidean, and the body is
rather narrow and elongated, though the softened specimen is
not in a satisfactory condition.

A comparison with the European form indicates that in the
small specimen (about 10 or 11 mm. in length) from South
Africa, the dorsal cirri varied in a similar manner in length,
being characteristically moniliform and with 18 to 28 segments
in the longer forms. The setigerous region is similar, but,
apparently from difference in size, bears fewer bristles than the
European, though their structure corresponds. The ventral
lobe has the same proportions.

No form described by Schmarda can be identified with this
species. His Syllis gracilis approaches it in regard to the dorsal
cirri, but the tips of the bristles are long and tapering. His
Giialliosyllis diplodonta would seem to be near it in external
characters, but its dental apparatus is peculiar.

Sylliscoriinta ranges from Finmark to Norwav and Spitsbergen,
to the British shores, as well as occurs in the Mediterranean
and at Madeira.

Nereis (Mastigoxereis) variegata, Grube, 1858.

1858. Xercis vahcgdhi, Grube, Annulat, ffirsted., Vid. Meddel,

Nat. Foren. p. 164.
1861. Masfigo lie iris potlocirni, Schmarda, Neue wirb. Thiere,

I., ii. p. 108, Taf. xxxi. fig. 247, and woodcuts.
1867. Nereis (Nereilepas) Sliinpsonis, Grube, Ann. Novara.

Exped. p. 18, Taf. i, fig. 8.
1901. Xereis7'arieg(if(t, Ehlers, Polych. Mag. u. Chilen, p. 112,

Taf. xiv. figs. 1-21.

Obtained between tide marks at various dates, at St. |ames',
False Bay.

3«

The examples ranged from i^ to 5 inches and in life the
latter probably reached 7 or 8 inches.

The dull bluish iridescent head is shield-shaped, with the
broad end posterior. Two eyes occur on each side posteriorly,
the anterior being wider apart. A small cuticular lens is in the
middle of each. The tentacles are stout and subulate. The
palpi are massive organs with bulbous terminal processes. The
dull brownish tentacular cirri are of moderate length, and the
dorsal of the second pair is longest as in ordinary Nereids.

The bodv is slightly tapered in front, then dilates to its widest
region and again diminishes gradually to the tail which ends in
two cirri of moderate length below the anus. The dorsum in
the preparations has a dull bluish iridescence, which Schmarda
states (presumably in life) is greenish blue, and the ventral sur-
face is yellowish brown. The sides are flanked by the feet with
their amber-coloured bristles.

The proboscis (Plate I, figs. 6 and 7) in the majority of the
examples is extruded, having distally a pair of powerful maxilla?
which, in the old examples, have their teeth abraded, but in the
younger forms show four or five distinct teeth in each. Dorsally
(Plate I. fig. 6) a group of paragnathi lies a little behind each
maxilla, and they are arranged in short lines, the longest being
posterior. In the median line between these is a single larger
tooth. On the ventral aspect of the maxilla? on each side is a
group of paragnathi similar in si/,e to the dorsal, but arranged
m four oblique posterior rows, and an irregular anterior series
(Plate I. fig. 7). A little posterior to these in the middle line is
another group consisting of three curved rows and one or two
points in front. A double belt of larger paragnathi occurs
ventrally at the base of the proboscis (in extension), ending at
each side in a single row, while dorsally the arrangement is
continued by a transversely elongated and bluntly conical tooth
on each side, and a single isolated tooth in the centre. The
organ is symmetrically cut by lines into various areas so that it
has a '' quilted " aspect.

The first foot is typical of the Nereids, having a somewhat
blunt and subulate dorsal cirrus, an ovate superior lobe, a trun-
cated setigerous region with two black spines beneath, a smaller
ovate ventral lobe extending beyond the former, and a lanceolate
ventral cirrus. The setigerous region is supported by a single
black spine which in lateral view separates the dark amber-
coloured bristles into two groups. The upper consists of about
three with long but stout tapering tips having spmous edges and
two with the shorter falcate tip. The inferior group conforms
to the latter type.

In its progress backward a tendency to a lamellar expansion
of the base of the dorsal cirrus appears. Thus at the 20th foot

39

it forms a distinct ridge at the base of the cirrus dorsally, whilst
the homologue of the dorsal lamella of the first foot
has coalesced with the ventral base of the cirrus and is continu-
ous with the dorsal ridge just mentioned (Plate I. fig. 8). More-
over, a new process, which first appears at the third foot in front
of the spinigerous lobe, has at the 20th foot placed itself between
the dorsal and ventral divisions of the foot, that is, has the dorsal
tuft above it and the ventral beneath it. Two black spines are
present, one to each division. The upper tuft consists of
bristles with more slender shafts and longer spinous tips than
any in the tuft beneath. At the 30th foot the upper division,
viz., the fiattened lamella with the dorsal cirrus at its tip, is now
larger in proportion to the rest of the foot beneath, a condition
still more evident at the 40th foot (Plate I. fig. 9), in which the
vertical diameter of the lamella is equal to the rest of the foot
beneath it. At the 60th foot the large somewhat rectangular
lamella exceeds in vertical diameter the rest of the foot, and in
some the dorsal cirrus springs from the upper angle and not
from the middle of the upper edge, but this appears to be due
to irregularly from injury or abnormality. The vascularity of
the large flattened lamella is now very distinct, and it has a
greenish hue. The various lobes of the feet are more dis-
tinctly separated from each other, as is also the ventral cirrus
from the lobe adjoining. At the 70th foot the lamella stands
nearly erect, the inner or dorsal edge being convex, the outer
slightly concave (Plate I. fig. 10). There is little change till the
tip of the tail is reached — when the flattened process diminishes.

The upper bristles (Plate II. fig. 12) have moderately elon-
gated tips which are finely and closely serrated, the spikes
coming off with a curve (the concavity being distal) at the
base, but towards the tip they are straight and somewhat more
prominent — if less regular.

Tlie falcate bristles are generally devoid of terminal pieces
and it is only by examining small examples of the developing
bristle m the foot that a perfect one is obtained. This
(Plate II. fig. 11) presents a short brownish tip slightly hooked,
and as yet devoid of serrations, but these soon appear as in the
figure. The bevelled tip of the shaft is also brownish, and from
it the longitudinallyand transversely striated central region passes
backwa'-d. With the exception of the tip, the shaft is pale. In
the older examples the entire bristle is deep brownish.

The largest example is a female with well developed ova
flowing freely in the coelomic space.

The food in the alimentary canal consisted — in one — of a
peculiar alga with minute stalked globular processes.

Schmarda's woodcuts are recognizable, though he shows a
papilla to the inner side of the ventral cirrus which is not visible

40

in these examples. He states that the maxillae hav^e only two
teeth (fig. K.), but the specimen was probably adult, when
considerable changes occur. He terms the dorsal cirri and
their lamelliform base — gills. His examples were procured in
Table Bay.

This species is very similar to the Nereis lieieropoda of Chamisso
and Eysenhardt,* procured on the shores of Unalaschca.

It also approaches Nereis (Lipephile) uiacropos, Claparede.t
which again is closely allied to A^. Marioiiii Aud and Ed., and
the same may be said of De St. Joseph's Neanthes PerrieriX
allowance being made for variations. The Nereis vexillos{i of
Grube § is another form the relationships of which require re-
investigation : there is much in the description and figures
which agree with N . Mariouii. It is also clear that the Nereis
(Nercilepas) Siimpsoui (an vor. vdriegata Grube-Kroyer ?) is
the same form, procured at the Cape during the Austrian
" Novara " Expedition.

Kinberg || found another species of the genus, which he
procured at San Francisco, California ; whilst Michaelsen's If
descriptions and figures of the proboscis of Nereis (Mastigonereis)
longicirra, Schmarda, from Ceylon, show a very close approach to
the present species, indeed the main difference is in the basal belt
of paragnathi in the extruded proboscis, which is distinctly
double in the South Afiican species. Ehlers, therefore, has
good grounds for uniting them. This author,** indeed, has
placed Schmarda's Masiigouereis podocirra, M. longicirra, Kin-
berg's Parallel eis elegaiis, Grube's Nereis Sfimpsonis, and four of
Hansen's species under the Nereis variegata of Grube.

The distribution of this species, indicates what may yet be
accomplished by a careful revision of specimens from various
quarters of the globe.

Accompanying the foregoing was a young example (8th
July, iqo2), the feet of which approach those of such forms as
Nereis Dniiierilii, with glandular masses on the dorsum.

Lysidice CAPENSIS, Grube, 1867.

Not uncommon between tide-marks at St. James, in False
Bay.

1867. Lysidice capeiisis, Gvuhe. Ann. "Novara" Exped., p. 12,
Taf. 1., fig. 4.

* Dc anim.quih. c Classc Vcnniiim, &c. F"ascic. II. p. 349, Tah. xxiv., 1". 2. 1819.

t Ann. Chet, Nap. Suppl. p. iSo, PI. viii. fi.t;. i.

X Ann. Sc. Nat. (8) v. p. 288, PI. xv. fifjs. 69-77.

§ Middendorf's Keise, .^nnulaten, p. 4, Taf. II. tigs. 1, 5, and 6.

II Ann. Nova. Olvers. af K. Vet. Akad. 1865, No. 2, p. 173.
5f Jahrh. Hamburg, Wiss. Anat. ix., p. 0. tigs. 8-10.

'-'"'' Polychajt. d. Magellanisch, u. Ctiilcnisch Strande, Berlin, 1901, p. J12, Tai".
xiv. figs. 1-21.

41

A form reaching the length of 150 mm., or more, and having
a breadth of 6 — 7 mm. at its widest part over the bristles.

The head is deeply bilobed, and carries three rather short
stumpy tentacles, with a pale patch at the tip, the rest being of
the brownish hue of the head. The median is the longer, and
in the preparations its tip just touches the central fissure in
front. External to the base of the outer tentacle, is an eye on
each side, with a well-marked corneal thickening. The nuchal
collar covers the back part of the head, concealing a median
band, passing to the first segment, and a lateral junction at each
outer tentacle, the margin of the central curving outward to the
lateral.

The body is very little tapered in front, and the diminution
towards the tail is also gentle. The latter terminates in two
short subulate styles beneath the vent. It is pale brownish
anteriorly, minutely dotted with pale specks, but in the pre-
parations these disappear from more than the posterior half.
Some of the anterior segments ventrally show the same specks
on the brownish ground-colour, but the rest of the surface is of
a uniform pale brown, variegated only by the reddish brown
pad beneath the ventral cirrus.

The proboscis has maxilhv with rather short blunt blades in
front, the posterior process narrowing to a notch, dilating, and
again narrowing to form a lozenge-shaped appendage po,s-
teriorly. The maxilke have three strong, blunt teeth in front.
The notches on theazygos plate are indistinct, and there are few
(about two) on the left anterior plate — beyond which is a small
•accessory plate. The right anterior curved plate has a bare
edge internally, and four teeth externally, and at its outer edge
is an accessory process.

The mandibles are boldly wedge-shaped, the anterior ventral
plates being large and with curved lines. Internally, they are
symmetrically striated, and are curved externallv in front, a
dense black longitudinal plate indicating the curvature. Thev
taper posteriorly.

The apparatus is tinted blackish, and, on the whole, cor-
responds to the somewhat crude outline of the parts, in Lvsidicc
(ifra, given by Schmarda. The posterior appendages are less
elongated, and more definitely bevelled, than in Lvs/'dicc
Niiiiita of European waters, but it may yet be a question, how
far such variations hold as specific distinctions, especiallv, as
the size of the African examples is so much greater.

The tenth and 20th feet differ from the 30th and those fol-
lowing, by the absence of the long black ventral hook, otiier-
wise, their structure is similar.

The 30th foot (Plate III., fig. 13) presents the single powerful
black spine, which pierces the surface between the dorsal and

I

42

ventral bristles. Dorsally is the somewhat short subulate
dorsal cirrus, the tip of which falls short of the setigerous
region, which forms a thick truncated lobe, separated by a brief
interval from the short and blunt ventral cirrus. The wliole
foot is richly vascular. The setigerous region bears above the
spine a tuft of bristles with tapering tips, which have but faint
indication of wings. A few brush-shaped forms are also present.
Beneath the spine is a dense group of compound bristles with
straight shafts, which are bent backwards and dilated at the tip,
as well as bevelled for the terminal piece — which is somewhat
short, biiid at the tip, and with wings.

The intestine contained sandv mud with fragments of sponge-
spicules, echinoderms, and small calcareous fragments.

The specimen named bv Grube Lysidicc capc'iisis* seems to
have been verv small, onlv 21 mm. in length, but its general
characters a< well as the structure of the foot correspond.

Except in the coloration, Schmarda'^ Lysidicc atra from the
Cape agrees with this form. When the feet of the latter are
contrasted with those '^f Lxsidicc Xiuctia the African has black
spines whilst the other has yellow spines, and the tips of the
compound bristles appear to be slightly shorter in the African.

Some of the examples had free ova of a greyish colour in the
perivisceral chamber, so that the breeding season would be at
the time of capture.

As Langerhans found Lysidicc Xiuctia as far south as Madeira
and the Canaries, it may yet be possible to reduce the number
of species of this genus, and it may be that the form from the
Cape will be placed as a variety of the European. The litera-
ture of this genus is much in need of critical revision.

It is curious that no example of the genus was procured
during the voyage of H.M. Ship "Challenger."

ErNiCE APHKODiTOis, Pallas, i78«.

lyHtS. Xciris aplirodiiois, Pallas, Nova Acta Acad. Sc. Imper.

Petropol. II. 4, p. 229, Tab. v. fig. i.
1820. I.codicc oigaiitiui, Savigny, Syst. Ann. p. 49.
1861. Eunice uiacrohraiichia, Schmarda, Neue wirb. Th. I. ij.,

p. 130, PI, xxxii. fig. 258.

1864. EripJivlc cdpciisis, Kinberg, Ofvers. af K. Vet. Akad.
Fiirh, p. 385 (fide Marenzeller).

1865. Eunice gigaiiiea, De Quatrefages, Ann. p. 311.
1878. ,, apliroditois, Grube, Ann. Semper, p. 147.
18H5. ,, (ipJirodiiois, M'Intosh, Ann. "Challenger" p. 282,

PI. xxxviii. figs. 16, 17, PI. xxa. figs. 8-10.
1889. Eriphylc aipcnsis, ATarenzeller, Zool. Jahrb. Bd. III. p.
7, Taf. 1. tig. 3.

* Anncl. Xovara Expcd. p. 12, Taf. I. ti.ii'. 4. 1S67.

43

Oi-)t;iinecl between tide marks, St. James', in False Bay.
A lai\Lie Eunice, apparently common at the Cape, stretching
to 1 80 mm. in spirit, and in life probably considerably beyond
that measurement.

In dealing with spirit-specimens it is difficult to make out
resemblances in colour, but in the examples from the Cape the
tints of most are recognizable. The cupreous or brownish red
and somewhat dappled hue, and the white bands anteriorly
I'eadilv distinguish it.

The head, tentacles, bristles, branchiae, and teeth all conform
to the type. The simple bristles, which are modified winged
bristles, are marked under a moderately high power by minute
and slightly oblique lines (probably extending from the wings),
and thus the surface is file-like, a condition figured by Schmarda
in his Eunice niacrohniiichid, which is here considered to be
the same species. The wing is very narrow, but the lines from
the serrations on the edge pass inwards as described. These
bristles have long tapering tips. The compound bristles with
bifid tips were few in the examples procured b}' the " Challenger,"
and consequently an average example from the anterior region
of the body is given in Plate 11. fig. 14. The brush-shaped
bristles have rather wide teeth with slender filaments distally,
and a long filament at one end (Plate II. fig. 15). The posterior
hooks correspond with those figured in the "Challenger."

The dental apparatus also coincides with the type, the great
dental plates each having 5 teeth ; the azygos 6 ; the left anterior
lateral 4 teeth ; and the right 7 teeth. Each of the two
accessory pieces attached to the latter has a horny dental
edge. The posterior appendages and the mandibles are typical.
The branchiae commence in these examples on the sixth foot
as a minute simple filament, and the organs become pectinate
on the nth foot. The pinna3 reach the number of 16 or 17,
but they probably vary, and may increase with age. They are
continued almost to the tip of the tail, which has two cirri
beneath the vent.

The food in the alimentary tract consisted of portions of a
Polyzoon with long vibracula, spicules of sponges, numerous
diatoms, fragments of alg^e and spores. In others, fragments of
a larger sea-weed occurred, besides smaller branched forms en-
crusted with large diatoms and mingled with mud rich in Fora-
minifera, bristles of annelids, crustacean hairs, sponge-spicules
and radiolarians.

Though Schmarda's description and figures of his Eunice
uiacrobyauchia leave much to be desired, yet, it is evident he
refers to this species, and Ehlers is inclined to the same view.
Thus he describes, under the specific characters, the semi-
cylindrical body as cupreous, or reddish brown, and, further.

44

notes that it is 200 mm. long. His figures and description of
the dental apparatus, however, are erroneous, and he has,
apparently, joined the mandibles to the posterior appendages
of the maxill?e. He observes that the first 12 segments are
devoid of branchiae, but he may have overlooked the simple
branchial processes commencing on the 6th foot. He recog-
nized the striated condition of the simple tapering bristles, and,
though inaccurate, his figure of the compound form is recog-
nizable. He gives the last segment of the body four cirri, but
only two are present in those sent by Dr. Gilchrist.

The relationship of this form to the well-known Eunice
iiphroditois, Pallas, is interesting. In many respects they agi'ce,
and though the condition of the branchiae on the anterior feet
calls for notice, yet, it is easy for a simple branchial filament to
become branched as age advances, or the environment alters.
The wide distribution of Eunice aphroditois gives room for
great variation, stretching, as it does, over the Indian and
Pacific oceans to South Africa. Thus, specimens measuring
fully two feet in spirit, from Viti Island, Samoa, show certain
variations from others procured off Port Jackson, Australia, and
between tide-marks at Samboangan. In the large examples,
even the bristles are affected, for many of the compound forms
have lost their tips, and, besides, are considerably abraded.

As Ehlers* has also pointed out, it is doubtful if Schmarda's
hliiuicc nigricaus, and his Eunice scJu^uiacepluda, are other than
varieties of the same form (Eunice npJiroditois). Kinberg'sf
figures of Eunice capensis would also appear to refer to this
species, as likewise Marenzeller's account of the same form
from Angra Pequena-Bucht.+

]\Iakphysa sAXGnxEA, Montagu, TCS15.

iSj^. Xereis .'^(inguiiu'u, Montagu. Trans. Lin. Soc. xi., p. 20,

^ PL iii., fig. I.
186H. Mdiphysei scini>niiu'u, Ehlers. Borstenw. ii., p. 360, Taf.

xvi., fig. 8 -II.

I^rocured between high and low water marks at St. James, in
False Bay.

The head agrees with that of MarpJiysti sdUiiiiinen, from the
Channel Islands.

The body in all is incomplete, the longest — with a short re-
generated posterior region, measures 85 mm., but the perfect
form is probably more than double the length. It is rounded

* Borstenw. ii., p. 306.

t Op- ^it-^ Tah. w/fiii. i6.

I Zool. Jahrh. Bd. lii., p. 7 (sc-p. abdr.).

45

in front, flattened throughout the rest of its exterior, and pro-
portionally broad. In life, it must have been finely coloured,
for in spirit the anterior region is dappled with reddish brown,
and iridescent.

The branchiae commence on the 19th foot, or in others on
20th and 2ist feet, as a simple filament, thus corresponding
in origin with those of Marpliysa sanguinea from the Channel
Islands and southern Europe, and, as noted subsequently, their
condition on corresponding feet is similar. No example was
complete, but, as far as could be observed, their distribution
posteriorly was similar.

At the loth foot, the dorsal cirrus is proportionally shorter
and stouter than in Marpliysa saiigninca, and there are live
black spines in the setigerous region, instead of three. The
simple dorsal bristles are similar in both, and no appreciable
distinction betw^een the ventral bristles, except in size, in the
respective forms is noticeable. The ventral cirrus is slightly
more prominent in the British form. The foot remains the
same till the appearance of the branchiae — often posterior to
the 20th foot. At the 30th foot, whilst Marphxsa sanguinea has
two long slender divisions of nearly equal length to the
branchia, and the dorsal cirrus is slender, the African form has the
gill much shorter — the longest not being more than twice
the length of the dorsal cirrus, the shorter about the same
length. Four divisions of the branchia occur in the 50th foot
of the British form, and they are nearly of equal length — two
arismg distally, and two beneath (externally) ; the African has
also four divisions — two terminal, and two external. The pro-
portions of the other part of the foot are the same. In each,
four divisions of the branchia occur in the 70th foot.

So far as can be observed, both from the external appearance
and the minute structure of feet and bristles, as well as of the
dental apparatus, the British and tlie South African forms are
the same. The species stretches to the American shores of the
Atlantic, having been found on the coast of Rhode Island and
New Jersey by Leidy,* by Websterf further south on the Vir-
ginian shores, and he states that it is common, especially near
high water-mark. He points out that De Quatrefages made a
new species {Marpliysa Leidii) of Leidy's examples, having, by
mistake, changed the 16 of the American authors' description
into 60. Alarenzeller,! again, finds the same species at Angra
Pequena, on the south-eastern coast of Africa.

* Jour. Acad. Nat. Sc. Philad. ii., Vol. iii., p. 147.
t Trans. Albany Inst, ix., p. 36 (sep. copy).
X Zool. Jahrb. iii., p. 11 (sep. abdr.).

46

Arabella (Akacoda) ikicolok, Montagu, 1802, var
c.4^:KrLa^A, Sclimarda.

1802. Xereis iricolor, Montagu, Linn. Trans, vii. p. 82.

1 86 1. Aiacodd avnthva, Schmarda, Neue wirb. Thiere. I. ij.

p. 115, Taf. xxxii. fig. 253.
1865. Liiinbricoiicn'is fricoloi, Johnston, Cat. B.M. p. 141.
1885. Notocirnis capensis, M'Intosh, "Challenger" p. 236, PI.

xxxvii. figs. 3, 4; PL xviiia. fig. 15.
1888. Maclovia <ligiintc(i, De St. Joseph, Ann. sc. nat. <S:c.,

ser. p. 230, PI. ix. figs. 92-95.
1901. Anicoda uienihuu Ehlers, Polych, Magellan, u. Chilen.,

p. 143, Taf. xix. figs. 1-6.

Between tide marks at St. fames', False Bay, appartntlv in
considerable numbers.

In general aspect this form closely resembles the Anihclhi
iricolor of Montagu.

The head forms a blunt cone which varies in outline accord-
ing to the specimen, in some having a tendency to an ovoid
outline, whilst in others it is more elongate. In lateral view it
is distinctly conical. At the base dorsally are four black eyes
arranged in a transverse line. The central pair are usually
most distinct, and young examples show the eyes best. The under
surface of the snout is often marked by a central hollow. The
first (peristomial) and the succeeding segments resemble those
of A. iricolor, and the same may be said for the rest of the body,
including the feet.

The dental apparatus (Plate iv. fig, 16,) presents the same
arrangement, three toothed plates being in front of the great
dental plate, but the size of the individual teeth in these plates
is considerably larger in the British forms. This is very noti-
ceable in the first — a single fang, and in the second. There is
thus a tendency to diminution. The great dental plates have a
more distinct, pointed process externally and anteriorlv, and
they are somewhat shorter than in the British type.

The maxillae are also rather shorter, and their bases are trun-
cated transversely. About 6 teeth are visible at the edge of the
broad basal region. These teeth are more numerous in those
from the Channel Islands, 7 or 8 at least being present. The
posterior appendages articulate with the maxilhe by a broad
triangular process which obliquely slopes to the long appendage
on each side, whereas in the British this is longer and narrower.
Moreover, three appendages instead of two occur posteriorly,
the additional process being somewhat broader, sh(^rter and
median.

The mandibles (Plate iv., fig. 17) ;ilso show certain differences,
being broader, stouter and shorter in the African, narrower and

47

more elongate in the British, but essentially they are the same.
TJnfortunately the injured condition of the teeth in examples
procured by the ''Challenger" led to ambiguity as to the rela-
tions of the species.

In the shape of the feet there is little to discriminate between
the forms, both having the setigerous lobe with its group of pale
spines anteriorly, whilst the slightly curved conical lobe passes
backward beneath it. Nor are the differences of the bristles
noteworthv. In both some of the bristles are longer, have a
marked curve at the commencement of the tapering tip and
well-marked wings, but do not show marked serrations, where-
as others have shorter tips with broader wings and bold serra-
tions, three of which at least have ridges running in from the
edge, in the British form this expanded region with the
serrations appears to be more differentiated, the wings being
narrower beyond it, so that the tip is more slender, but such
distinctions are variable.

The body terminates posteriorly as in the European forms,
and regeneration of the tail occurs. In some, a slight tube of
mucus with sand-grains surrounds the body. On the whole
the forms approach each other very closely, and if even the
variations of the European examples are considered, e.g., as
shown by Ehlers and by the British specimens, then there is little
ground for specific distinction. Thus Ehlers, for instance,
figures* the same arrangement of the posterior appendages of
the maxillae as seen in the African forms.

Further, the genus Anicoda of Schmardaf is clearly founded
on this species, which he procured not only at the Cape but on
the coast of Chili, and therefore lapses, since Grube's title has
priority.

There is little in Schmarda's description of the genus, except
that there are no eyes or dorsal cirri, bristles simple, 8 or lo
maxillae, of which the inferior alone are calcareous. He
noticed the serrated edges of the bristles with wide tips. His
specific characters are : — Body^ cylindrical, bluish ; Cephalic
lobe longer than the first two segments ; ten superior niaxilhe.
Feet conical ; bristles winged, bent towards the tip. He gives
a length of 240 mm. and 500 segments. His figures of the
dental apparatus (woodcuts, p. 115) are fairly good, and so with
that of the serrated bristle.

Schmarda's coloured figure shows the anterior end trilobed
and tinted brown, but such a condition could only have been
produced by the partial extrusion of the proboscis — a feature
occasionally seen in the present examples. This author
followed no strict law in grouping his Annelids, for his second

* Horslemviinner, II. Plate xvii. tij;-. u).

f Xeiic Wirix Tliifi'c, I. ii. p. i 15, Tal". \x\ii. ti.c. 2^t,.

4^

species of Aracocia** (A. heteroclueta) is apparently a Liinihri-
conereis with a different dental apparatus.

It is probable that this species occurs at the Canaries ("Langer-
hans§^, so that the step to South Africa is broken. The Arabella
opaliiia, of VerrillJ is a very closely allied, if not identical
species, and this has also been found at Porto Rico, by Tread-
well. ft The Arabella maculosa, n.s., of Verrill,$t from Flatts
Inlet, Bermudas, presents no feature other than what might
occur in a variety of this almost cosmopolitan form.

Further, the Aracoda cacnilea, Schmarda, as described by
Ehlers, would not seem to differ in any essential particular
from the British representatives, for the presence of a third
narrow appendage to the maxillae is of comparatively little
moment. This third appendage occurs in the same species
from Japan (Kada Bay). The view here suggested is the more
likely, from the fact of the very wide distribution of the species.
A form which stretches from Britain to Japan on the one hand,
and from South Africa to Juan Fernandez on the other, should
not be too rigidly judged, according to artificial standards of
distinction.

LUMBRICONEREIS TETRAURUS, Schmarda, 1861.

1861. Notocirnis teiraurus, Schmarda, Neue wirb. Th., I. ij.,
p. 117.

1885. LiiDibricoiicicis Pcittgreivi, Mcintosh, Annel. "Chal-
lenger," p. 239, PI. xxxvi., figs. 7, 8, 9, PI. xviirt., figs. 11
—14.

1901. LiDiibricoiiL'iris tctranra, Ehlers. Polych. Magell. u.
Chilen., p. 137, Taf. xvii., figs, i — 10.

Found between tide-marks at St. James, False Bay.

The head of this species forms a blunt cone with one or two
longitudinal grooves dorsally, and a close series ventrally.
Posteriorly is a median band, with a depression on each side,
where it joins the first segment.

An elongated species of the usual shape, and the tail is
terminated by four short caudal cirri, the inferior being the
longer. The colour is brownish, with fine bluish iridescence —
especially anteriorly.

** Grubef adopted Schmarda's genus Aracoda for those which had the bases
of the maxilke toothed, and Ehlers* followed. Considerable confusion,
however, exists as to the toothed condition of the maxilke, for the teeth
are often hidden — from the nature of the edge of the maxilhe.
Ann. Semper., p. 175.
Op. cit., p. III.

§ Langerhans, Nova Acta. Bd. xlii., No. 3, p. 112, Taf. li.. lig. iS,
II Repoit Invert. Vineyard Sound, 1S74, p. 594, figs. 6g and 70.
ft Bullet. U.S. Fish Com. for 1900, p. 199.
XX Trans. Conn. Acad. Arts and Sc. \., 1900, p. 651.

49

The dental apparatus comes near that of Schmarda's species,
the long pointed ends of the posterior appendages being diag-
nostic (Plate III., fig. 1 8), and resembling those of Kinberg's
Liimhriconcrcis Jacksoiii.^' The maxillae are strongly curved,
and articulate posteriorly with a dilated region, which is again
constricted, and ends in two slender tapering points (one on
each side). In lateral view, the maxillae show a considerable
curve. The great dental plates have four distinct teeth on each
side, and the movement of these plates is such that the dental
edge can be placed either horizontally or nearly vertically with
the teeth pointing dorsally. A considerable portion of the edge
of each posteriorly is smooth. In front of these is a small
plate, which also appears on the ventral surface with one or two
teeth, then a larger plate similarly placed in front, with a single
tooth at its posterior (dorsal) edge ; whilst externally, is a flat
horny plate, and after an interval, a thin curved horny process,
which passes backward to a point opposite the middle of the
maxilla.'.

The whole apparatus is of a blackish-brown colour.

The mandibles are elongate, wedge-shaped as in Luiiibri-
conereis Pettigreivi, only shorter.

The feet are normal in outline, a conical lobe, or process,
slanting backward from the setigerous process. This is
probably what Schmarda calls " hranchia (cirrus dorsalis) basi
angitstior," a statement which has given rise to misapprehensions.

The bristles (Plate II., fig. i^a) anteriorly are winged with
the usual curve at the base of the tip, and with finely serrated
edges. Towards the 50th foot, winged hooks take the place of
the inferior groups of winged bristles, and, by-and-by, winged
hooks alone occupy the feet to the tip of the tail (Plate II.,
figs. 19 and 20). The wings of these hooks are somewhat broad,
and the tip ends in a main fang, and several smaller points
superiorly. In most of the larger forms these are abraded,
being best seen in young specimens. The edges of the wings
are serrated. The spines are yellowish.

The constricted condition of the posterior lamella of the foot
(Schmarda's branchia) may have been connected with preserva-
tion, as this was not a feature observed in the present specimens.

Schmarda obtained it both at the Cape and on the coast of
Chili, and Ehlers has recently described it both from Magellan
and Chili. The distribution of this species is, therefore, very
wide. The Luuibriconereis Pettigreivi of the "Challenger " would
seem to be the same form.

* Freg. Eugen. Kesa, cSic, Taf. xviii., tiLi 34.

D

50

LrMBKICOXEREIS

Another form dredged by the " Chiillenger," :it Station 141
(off the Cape of Good Hope), differs from any in the present
collection. It is distinguished by the presence of a single long-
winged hook (" Challenger " Annel., PI. xvii<7., fig. 15) in the
anterior feet. An increase in the length of the tips of the second
group of superior bristles anteriorly, and yellowish spines are
other features. In the dental apparatus, the maxilLie had trans-
lucent hard tips, and the basal ridges did not correspond. The
left great dental plate had four, and the right five teeth. The
outer lateral plate in front is larger than in Luiiibi icoucreis
tetraiinis, and bears two teeth. The inner has a short tooth.

As the only example was injured, it is well to draw attention
to this species.

LUMBRICOXEREIS CAVIFROXS, Grube, 1 867.

1867. f^nmlmcoiiereis Cdvifroiis, Grube. Ann. "Novara"
Exped., p. 13, Taf. i, fig. 5 ?

1868. Lmubriconcreis gracilis, Ehlers. Borstenw. ii., p. 393,
Taf. xvii., figs. 6 — 10.

Procured at St. James, False Bay, between tide-marks.

A small species about if ins. long.

The head forms a smoothly rounded blunt lobe in front, and
is slightly constricted at the posterior border. A minute median
band connects it dorsally with the first segment. In lateral
view, the snout is nearly of equal depth throughout, so that the
tip is characteristically blunt, but such is probably due to the
degree of contraction, for in another example, with a deep
median sulcus on the ventral surface of the snout, the latter
was less blunt in profile. The dorsal groove mentioned by
Grube in Lmnbricoiicreis cavifrons, was not present in these.

The body is very little narrowed anteriorly, remains for some
distance of nearly uniform diameter, and then tapers gentlv
towards the tail — which, in the examples, had four short cirri
(like papillcc ), the two upper being the longer. The segments
are very distinctly marked. The general colour of the body is
pale brown, wilh only a little iridescence on the curves of the
segments, and thus is in marked contrast with Lnnihricoiiciris
fiiniiinis. The cuticle is also somewhat delicate.

The dental apparatus (Plate iii. Hg 21) presents a pair of
strongly curved sharp maxilla? which pass backward to converge
at a narrow pedicle connecting Ihem wilh the posterior ap-
pendages. After a constricted region the latter again dilates, but
the specimen is incomplete, though apparentlv they terminate in

51

the narrow plates. The great dental plates are pale translucent
brown, and each has live prominent teeth of great strength.
The plate in front has two teeth, and the anterior plate one tooth.
Two prominent papillae occur on the anterior edge of the lower
lip, but no mandibles were present in either specimen.

The granular condition of the borders of the anterior horny
plates, so well seen in British examples, is likewise present,
though less distinct both in these and in the opaque areas
exterior to the great dental plates.

The feet are not so prominent as in the preceding species,
the posterior lobe being less conspicuous. The anterior feet
have winged bristles dorsally and jointed hooks interiorly. The
former have the tip curved on the shaft — with broad wings and a
short tapering tip. The hooks (Plate iv. fig. 22) have the long
shaft slightlv curved backward at its distal end and bevelled for
the basal process of the terminal piece — which tapers to a
crown with a main fang aud several minute teeth above it, and
with guards or wings which come from the end of the shaft.

The first foot has a winged bristle or two, a long simple hook
and two jointed hooks. At the fifth and sixth foot only one
jointed hook occurs, but they evidently vary as in the 7th and
(Sth two winged bristles with tapering tips occur, then three
jointed hooks, and ventrally a single smaller winged bristle.
The latter disappears in the 12th, 13th and subsequent feet, the
14th having only a single simple hook.. The tapering tip of
the bristle becomes shorter and the wings broader at the base.

The bristles by-and-by disappear, and the hooks remain
simple (Plate IV. fig. 23) the shaft having a prominent curve at
its stoutest or distal region, then narrowing to a crown which
has a main fang and a series of small teeth above it. Moreover
two small teeth occur on the distal edge of the great fang.

The description of Liniibriconcrcis cavifrotis by Grube is
somewhat imperfect, but it would seem to refer to the same
form, though none of the present examples showed a dorsal
groove on the head, which it may be remarked, also occurs in
certain specimens of Liiinbi icoiifivis gracilis, Ehlers. The latter
seems to be a closely allied if not identical form — first found
by Ehlers at F'iume in the Gulf of Quarnero in the Adriatic,
but which is now known to occur at Madeira and on British
shores, so thai its distribution is very wide and ma}^ yet embrace
South Africa. Grube's examples came from Chalk Bav.

Young Examples.

Certain young forms procured on various dates, show the
foil owing differences from the ioregoing{LiimbricoiiiTcis!^rin His).
Thev occurred in the tissue of a sponge inside a fiagmeiilary
Troiiiiis.

I) 2

52

The head has a similar shape, but it has two distinct eye-
specks near the anterior border.

The body (Plate iv. fig. 24) has from 15 bristled segments in
the smallest to 28 in the largest, and the tail terminates
in four flat papilla?. The proportions of the first two body-
segments are the same. In that with 28 bristled segments, the
first twelve have bristles with tapering winged tips, these as well
as all behind having also long simple (unjointed) winged hooks,
which appear to agree in structure with Liniibricoiifiris gi-acilis
beside which these were found.

So far as could be observed in transparent preparations
the dental apparatus corresponded in general structure with
that of Liiiubriconeveis gracilis, though the posterior appendages
of the maxilhu were less developed. The great dental plates
had 5 teeth, and the anterior plates respectively two teeth and
one tooth. No mandibles were present. A horny structure in the
gut posteriorly was either a portion of the dental apparatus which
had been shed, or a structure which had been swallowed as food,
the former being the more likely. These may be the young stages
of the previous species, if it were possible that the jointed
hooks anteriorly were only developed in the later stages.

Obtained between tide-marks at St. James, in False Bay.

Tkophonia capensis, Mcintosh, 1885.

1885. Trophoniii capciisis, Mcintosh. Annel. " Challenger," p.
363, PI. xliv., figs. 7, 8, PI. xxxiiifl, figs, i — 3.

The examples procured by Dr. Gilchrist are considerably
longer than that obtained by the " Challenger " between tide-
marks at Sea Point, Cape Town, in December, 1873, for they
measured up to 90 mm., whereas, the " Challenger " example
was only 60 mm.

The outline anteriorly is somewhat fusiform, and then the
body gradually tapers to the tail which, however, is by no
means slender. It terminates in the vent — with a slightly crenate
margin. The segments were more numerous (about 90) than
in that described. The skm had numerous sand-grains and
siliceous and calcareous particles attached to it, the dorsal
surface being coated with larger semi-translucent (pale) frag-
ments— especially on the anterior region. The whole forms a
tough investment of hardened mucus, fragments of sponge-
spicules and light sand-grains, which can be peeled from the
body — carrying many of the bristles with it. Thus exposed,
the surface is definitely ringed and smooth — with papilla?
arranged at regular intervals in each segment. Anteriorly,
within the rows of bristles, are two rows of papillae ; between

the bristles and the hooks laterally are two rows ; whilst,
ventrally, there are four nearly equi-distant rows anteriorly ; at
the end of the anterior third are six rows, and these continue to
the posterior region. These papillae are readily recognized by
their darker colour, the general surface of the body being pale
brown. When the investment is removed, the body becomes
flaccid — demonstrating that its comparative rigidity is due to
this coat. The uniform arrangement of the small grains on the
ventral surface and sides and their general effect, together with
the addition of the larger translucent grains to the dorsal
surface (to which, indeed, they were confined), showed dis-
crimination and care on the part of the Annelid.

The snout bears two grooved tentacles, and a dense tuft of
branchiae superiorly on each side. The latter are more
numerous and more slender than in the common European
species {Trophonia plitiuosn), and they arise from the ventral
aspect of a tongue-shaped triangular process, which projects
upward and forward on each side from the upper lip. Each is
marked, in extrusion, on the dorsal surface by a line of dark
pigment just within a pale margin. The processes appear to
fuse along the middle line, though a deep groove exists on the
ventral surface. Ventrally, the branchiae are closely arranged,
and in somewhat regular rows. Each has an afferent and art
efferent vessel. In the central line in front of the mouth is a'
longer papillae, which may be bifid, each division being slender.

One example had the body-cavity tilled with dark green ova,
as in the case of the " Challenger " example, which was obtained
in December. Another appeared to have masses of sperms in
the same cavity.

The genus is probably identical with Stylnividcs, Delia Chiaje.

Flabelligera luctatok, Stimpson, 1856.

1856. Tecturella litctator, Stimpson. Proc. Nat. Sc. Philad.

vii., p. 391.
1861. Plicnisa tcinLgoua, Schmarda. Xeue wirb. Thiere, i.,

li., p. 20, Taf. XX., lig. 168.
1889. Flabelligera litctator, Marenzeller. Zool. Jahrb. iii., p.

15 (sep. abdr.).

Found between tide marks at St. |ames's, in False Bay.

A form somewhat fusiform in outline and ranging from 30
to over 50 mm. in length, and with a breadth at its widest part
of 5-6 mm. It is of a firmer consistence than the ordinary
Flabclligerit alj'iiiis, and thus its habitat probably differs.

The tentacles are of considerable len<fth and frilled m a
characteristic manner. As a transparent object a series of club-

54

shaped i^raiuilar papilla? occur in rows at the sides of the organ.
The brancliiic form a dense group of thread-like structures on
eacli side dorsally. Schmarda shows them of a bluish tint.

The foregoing organs are enclosed by a fringe of golden
bristles, which with a dense series of elongated papilla? between
them form a web continuous at the sides, but with a gap
dorsally and ventrally. These bristles are comparatively short
in contrast with Troplioiiui, but their structure is similar, and
their curve is adapted to the contour of the wide fringe or pali-
sade which they form around the central organs.

The long papilla? have slender stems with a fusiform dilata-
tion near the tip — which ends in a bulbous enlargement.

The body is fusiform in outline, though by no means finely
tapered, indeed the ends are blunt. The skin is dark brown
and does not appear to have the abundant gelatinous tissue so
characteristic of Flabdligcm affiiiis. On each side of the dor-
sum is a series of bristle-tufts, the bristles having a similar
structure to those in front, and at each tuft is a dense group of
the long papilla?. The colour is dull greenish.

Ventrally, again, each segment has two long and strongly
curved hooks (Plate iv. fig. 25) with slightly curved shafts
which have a dilatation below the sharp hook at the tip, and
another less marked at the commencement of the transverse
lines at the end of the shaft. The number of the lateral bristle-
tufts ranges from 41 to 54, Schmarda giving 37 body segments.
His specimens, however, were smaller. The surface of the
body has a considerable quantity of adherent sand-grains,
which are found both dorsally and ventrally, but they do not
form a continuous coat, much of the body being bare. The
rows of ventral hooks have a narrower space between
them than the dorsal bristles, the ventral surface of the tetrahedral
body being thus much less than the convex dorsal region.

In regard to the sand-particles and their attachment it could
not be said that each was fixed to the tip of a papilla. The
grains seemed to adhere to mucus in which were sponge-spicules,
diatoms, and here and there groups of papilla?, but the latter
appeared to have no special connection with the particles
though they may have with the mucus. The sand-grains con-
tained manv calcareous fragments amongst the siliceous.

The same sand with its coarser and finer particles occurred
in the alimentary canal.

The general aspect of this form would seem to indicate that
it leans to Flabelligera, though no example had the membrane
between the hooks and bristles, and in this view it is satisfactory
to have the support of so able an investigator as Dr. E. Marenzel-

55

ler, who found the species on the eastern coast of Africa, at
AnjLjra Pequena.* There are sHght differences in the figure of
the hook, to which the Austrian author gives transverse striae
in the distal region. Such, however, may be due to age or
local variation.

Flabelligera marenzelleri, n.s. (an var. affinis.).

Collected between tide-marks at St. James, False Bay, South
Africa.

Procured along with Flabclligcui Itictator, Stimpson, from
which it is readily distinguished by its smooth body, the
presence of a membrane between the hooks, and the flattening
of the bristles to the sides of the body, apparently by a similar
gelatinous membrane. On the whole, however, it has a less
gelatinous aspect than such as Flabelligera affinis.

The frontal bristles, in general aspect, resemble those of
Flabelligeta affinis, though minute examination shows that the
former are darker yellow, and that the segments are shorter.
No appreciable distinction could be drawn between the
elongated filaments, with their characteristic terminal enlarge-
ments. The tentacles and branchiae are similar.

The fusiform body is more consistent than that of Flabelligera
affinis, and retains its straight outline, the convex dorsum being
covered by a glistening cuticle. The dorsal area is large, since
the lateral rows of bristles and hooks are carried far down-
wards, thus encroaching on the ventral area.

The dorsal bristles are shorter, stouter, and duskier in hue
than those of Flabelligera affinis, and the transverse bars seem
to be rather closer. In the British form, these bristles taper to
a more slender elongated tip^followed by the long papillae.
The latter appeared to agree.

The long golden hooks are arranged, for the most part,
singly, along the sides of the ventral surface, and have, about
the commencement of the distal third, a curve backward.
Distally, the shaft joins the flattened sickle-shaped tip (Plate
111., tig. 26), which dilates at its commencement, and ends in
a sharp strongly curved brownish claw. The internal striations
nearly correspond with those in the British species, indeed, it
is hard to draw a distinction between the two. The South
African has, apparently, a more brittle nature, since a separation
often occurred at the transverse segments, is darker (as a trans-

* Op. cit.

56

parent object), and the terminal claw is usually more erect, a
condition seen in certain varieties of the European form, such
as those from the Clyde.

The body-wall of the South African form has thicker
muscular layers, and the papillai seem to be more elongated
terminally than in the British, but preparation may be
responsible for this.

The question, therefore, as to specific distinction is a delicate
one, and separation rests on the shortness and strength of the
bristles, with their slightly narrower transverse lines, the curves
and colour of the hooks, and the more consistent body. Future
investigation may show that the Mediterranean Siphotiostonia
diplochaUos of Otto, has relations both to the northern forms
and to those from the south.

[Pubh.shed 5th February, 1904.]

-o-

EXPLANATION OF PLATES.

-o-

Pl.ATK I.

Fu;.

I. Aiilcii<ir loot ol' Ihiiiilc'pid/ii liihciiiihilti, Schmarda. 13 diain.

(). Dorsal view of Ihc liead and extruded proboscis ot A'tVc/.s {^Miistii^t<iit'i\is)
Vdriegaki, Gnibe. Enlarji'ed under a lens.

7. Ventral aspect of the prohf)Scis of the same form. l-2iilai'gcd imdef a leii^.

iS. Twentieth foot of tlie same. 13 diam.

y. P^)rtieth foot of the foreii'oin.U. 12 diam.

10. Seventieth foot of the same. 17 diam.

Marine Invest! gautions
South Africa..

Annelids .PI. I.

W. CM. del.

MmternBros .iitK.

HEMILEPIDIA &TN1EREI S

]>],.\TK 11.

Fk;.

2. l)i>rs;il bristle of Honilcfidid tiihtii'iiliita, Schin;u'd;i. X240 diam.

3. L;ir.L;c upper bristle of the \entral series of the foi'tieth foot. Similarly

nia,Li;"iiitied.

4. l-5ristle from the upper series of the ventral of the tenth fo<it. Similarlx

maLi'nihed.

5. Hilid bi istle from the ventral series of the same form. Similarly maynilied.

11. P'alcate bristle from the anterior region of Xciris (Masti,i^(>i/crc/sj vnrifi^tiiti

Gruhe. X240 diam.

12. Jointed d(3rsal bristle (with tapering tip) from the same region. Siniilarh

magnified.

14. Falcate bristle from the \entral series of an anterior foot nf Einiice upliro-

(l/hv's, Pallas. Magnified as before.

15. P>rush-shaped bristle fi-om the dorsal -series of the same form. .\s liefore.
iSa.'Pwo winged bristles of Liniihricoiicir/s titriiitnis, Sehmarda. 100 diam.

19. Winged hook of the same form in lateral view. X240diam.

20. Head of another winged bristle \ie\\ed from the face (front). Siniilarh'

magnitied.

i/ia^rine investiga-tions
South Airica.

5

Annelids .PI .11.

- i

14. '

*'\ i

H-

^Mk

15,

V/.C.-M.ael.

liIintem.Bros.lith.

HETVilLEPIDlA, :EUTSiIClD^.

Plate HI.

Fig.

13. Anterior toot of Lysiditc < (//'<//n/>, Griihe. 24 diam.

i<S. Dental apparatus of Liiniliriioufrcis tiirimnis, Schmarda. The second tooth
is soniewiiat imaginary as its hitid coiiditi(Mi is not seen in this position.
15 diam.

21. Dental apparatus of Liiiiihricoiuri/s ciivifroiis, Grube. The dotted line con-

tinued from the broken posterior appendages is imatiinary. 15 diams.

22. I^ateral \iew of winged tapering bristle of the same form. Zeiss oc. 2. obj.

D. + .1 in. draw-tube.

26. Hook of Fliibfllii>,ni Miutiizelleri (an. \ar. nffinis. Sars). r40 diam.

Marine Investigations
South AfncOL.

Annelids, PI. HI

W. CM. del.

Mirvtsrn Bros-lith.

EUNICIB^ &c CHLOR^^MTD.^.

Plate IV.

Fic.

i(). Dental apparatus of Arabella iricolor, Montagu. 2i diam.

17. Mandibles of the foregoing". 15 diam.

22. f< tinted anterior hook of Lii iiibricoin-reis cavifrons, Gruhe. Zeiss oc. 2. ohj.

D. + 3 in. draw-tube.

23. Winged simple hook of the same species. Similarly magnified.

24. Young examples of tiie same form with eyes, and the dental apparatus in

si III. 50 diam.

2^. Flook of Flahelli,iierii Mcyciizellcn (an var. atiiiiis sars). too diam.

¥a.rme Inves tig actions.
South Africa.

Annelids. PI. IV.

16.

22.

r^

r-

W. CM. del.

TjfinternB-fos .lith .

EUTN'ICID^ <?^ CHLCT^yEMID^.

MARINE ANNELIDS (POLYCHiETA)

OF

SOUTH AFRICA.

PART II.

BY

WILLIAM CARMICHAEL McINTOSH,

M.D., LL.D., F.R.S., &C.,

Professor of Xatural Histoiy in the University of St. Andrews, Director
of the Museum and the Gatty Marine Laboratory in tlie University.

CONTENTS,

Introduction

Poly HOC {MaceUiccphala) iiiim bills ...

Polyeu}W({ hvvis ...

Chaioptcnis vdriopcdalits ...

Theodisca {Aiilhostoina) hcxaphyllum

Young of Theodisca \Atilhostoma) hexaphyllum

Cirratulus cirraiiis

Do. Icutaciilatus
Dasybninchus cadnciis ... "
Nicoviache Ininbricalis

Do. Mcintosh a
PraxUla pnvtcnnissa, var. capensis ...
Sahcllaiia capensis
Pecti)iaria capensis
Schmardanella plcwchaiii ...

Thelepiis ?

Polamilla reniforuiis

Bispira voliitacornis

Sabelliphilus (?) bispinv (crustacean parasite)

I'AGK.

59

59
60

61

63
66

67

68

70

7i
72
73

74
76

79
80

81

»3
86

B

59
INTRODUCTION.

In this as in the previous part attention may again he drawn
to the comparatively large number of European forms in South
African waters, and to the very wide distribution of others
which occur there. Thus the purple Polynoid, Polynoe (Macel-
liccphala) inirabilis, ranges from the north-west of the North
Island of New Zealand to the Cape on the one hand, and to the
Kara Sea (Levinsen) on the other; whilst Polyeunoa Icevis
stretches from the Strait of Magellan to Prince Edward's Island
and is now brought within 25 miles of the Cape. Indeed
almost everyone of the species in this section is remarkable for
its cosmopolitan tendencies, and, as critical examinations extend
in future, it is probable that one or two still doubtful may be
included in the same category. The range of such forms as
Chaioptcnis variopcdatiis, Cimifiiliis cirnifiis and feiifacnlatus,
DasvbraiicJiiis caduciis, Nicouiachc Imuhricalis, Potamillu reiii-
fonnis, Bispira 7'oliitaconiis and even the crustacean parasites of
Annelids is remarkable, and show how much has yet to be done
in the distribution of marine animals, and moreover illustrates
the bold contrast they present to the distribution of land
animals.

Fam. POLYNOID^.
Polynoe (Macellicephala) AIirabilis, Mcintosh, 1885.

1885. Polynoe {Macelliccphahi) niiiahilis, Mcintosh, Ann.
"Challenger" p. 121, PI. xvi. fig. 1, PI. xii. figs. 9-1 1.

1886. OligoUpis violacea, Levinsen, Kara-Havets, Ledorme,
p. 4, PI. XXV. figs. 1-4.

This purple polynoid was procured in the trawl at 470 fathoms,
25 miles off Cape Point Lighthouse, bottom, green sand and
hard ground. The specimens had been put in formalin and
their colour was fairly preserved. The rich purple or deep mauve
tinted the whole of the dorsum where it was most intense, the
proboscis and the feet, the tips of which were less deep in
colour. The ventral surface was also somewhat pale — marked
along the ventral line by the nerve-cords ; and a pale line
passed from each segmental papilla to the middle line — nearly
transversely in front, but obliquely forward and inward behind.

This form, a single example of which was first dredged
during the cruise of H.M. Ship "Challenger" at Station 169
(off the north-west corner of the North Island of New Zealand;
in 700 fathoms, is evidently an inhabitant of deep water. The
South African examples, from their mode of preservation are

B 2

)(J

much more richly tnitccl than the original one. They agree,
however, in all essential characters with the type-specimen.
They are somewhat larger, the body of the most complete
measuring 28 mm. in length (exclusive of the appendages) and
of similar breadth to that procured by the "Challenger."

No scales existed on any of the examples (three), so that they
would seem to be very slightly attached. Whether the delicate
scale with a group of small papillae — in the bottle beside the
specimens — pertained to this form is unknown. The caudal
cirri were likewise absent, the bristles of the last pair of feet
being directed backward and greatly diminished in length.

In the description of the specimen from the " Challenger " *
it is stated tiiat the dorsal surface of the partially extended
proboscis shows two papilhe, one beneath each flattened frontal
lobe, but such probably refers to the two marked lateral fillets,
which when the organ is fully extended occur opposite the gap
between the dorsal and ventral series of terminal papilla;. The
basal region is somewhat constricted and minutely corrugated
and divided by a furrow into two fillets. Beyond the latter the
organ dilates and forms a dorsal and a ventral crescent, whilst
the lateral region passes forwards as a peak. The dorsal and
ventral series of papilla; at the tip of the organ in extension are
9 in number, and they readily fall off in the preparations. The
horny jaws follow the ordinary arrangement, the right of the
dorsal pair passing between the ventral.

In the structure of the feet and of the bristles — both dorsal
and ventral — they agree with the example from New Zealand.
Opposite the base of each foot is a deep dimple causing the foot
from the dorsal surface to have a bilobed base. Attached to
the tip of one or two segmental papillae — by a stalk from the
lumen — were little whitish masses of sperms — all the specimens
apparently being males. It comes to be a question as to
whether such a type is not more or less pelagic at this period,
a feature perhaps not unconnected with its wide distribution.
The alimentary canal posteriorly formed a small empty tube,
the bodv-cavity being filled with masses of sperms.

The OligoU'pis viohucci of Levinsen, from the Kara Sea,
published the following year (i886j appears to be the same
form.f

POLYEUXOA L.'EVIS, MclntOsh, 1885.

188^. Polxcinioa Ln'vis, Mcintosh, Ann. "Challenger," p. 76,
^Pl. xii., fig. 2, and PI. XX., fig. 8.

A single example devoid of scales was obtained in the trawl
by Dr. Gilchrist in 470 fathoms, 25 miles off Cape Point Light-

* Op. cit. p. 122. f Kani-Havets Ledoniie, Kjohenhavn, 1886.

6i

house'; bottom — green sand and hard ground. Several loos(
scales' however, were in the bottle beside it -dnd Poly iiol' (Macc'l

Several loose

7-

Jiccplidhi) ininibilis, the other form procured at the same time.

The occurrence of this species off the Cape is interesting —
showing a nearer approach to shore than in those dredged by
the " Challenger " off Prince Edward Island in 310 fathoms.
The species likewise stretches fo the Strait of Magellan. Whilst
two of the loose scales in the bottle agreed with those of the
examples from the "Challenger," a third had a group of small
papillae, but its connection with the rest was uncertain. The
example was a female with ripe eggs.

Fam. CHiETOPTERID^.

ChcetOPTERUS variopedatus, Renier, 1804.

1804. Triccvtid vdiiopcdnta, Renier, Osservazioni postume di

Zool. Adriatica, p. 35. -
,, ,, Meneghini, ibid. p. 38.

1844. Chivtoptcrns pergauiciitacciis, Will, Archiv. f. Xaturg.

t. X., p. 328.
,, Leuckart, ibid. t. xv. p. 340.

Leiickaiiii, De Quatref. Annel. II., p. 216.
pcrganiciifiis, Kowalewsky, Entwick. ' d.

Rippenqualler, p. 6.
Capeiisis, Stimpson, U.S. Surveying Exped.

Proceed. i\cad. Nat. Sc, No. 33.
lidiiuifiis, Schmarda, Neue Wirb. Thiere.

ii., p. i6,Taf,xix.,f. 166.
I'tiriopL'ddtiis, Claparede, Ann. Nap., p. 338.
,, Cams, Faun. Medit. i., p. 257.

variopcdatiis, Joyeux-LufHne, Arch. Zool.
Exper. {2) viii., p. 245,
Pis. xv.-xx.
„^ De St. Joseph, Ann. Sc. Nat.

8"^^ Ser., xvii., p. 147, Pi. vii.-
f. 189-199.
„ Ehlers, d. Hamburg. Magal-

haen. Sanmielreise, p. 109.

Between tide-marks, at St. James', in False Bay.

Ten bristled segments occur in the first region, the last joining
the long wings, as in the example of the same species from the
Channel Islands, England and Scotland. The head dorsally bears
the long tentacles, at the bases of which externally are the eyes —
a linear series of isolated pigment-spots which in the prepara-
tion are covered by the free flap of the collar. One or two

if

!■>

))

i«55.

t»

1861.

!f

1868.
1885.
1890.

>>

1894.

<}

1897.

Jt

62

specks, however, are visible externally at the base of the collai
on each side. The great setre occur on the 4th segment and
are modifications of ordinary bristles for special work (Plate V,
fig. 51). They are of great size, widened at the tip and
specially thickened by the chitinoiis secretion there, a feature
well seen in forms which have not yet been worn. These, as in
the figure, have the tip on one side acute, and, moreover, the
edge is minutely serrated, as if to show its connection with the
ordinary forms. These serrations disappear in those subjected
to wear. Considerable variation in size occurs in representatives
of the species from different quarters, thus in some Neapolitan
examples these bristles are considerably smaller (Plate V, fig 52 j,
but they retain their essential characters.

In the sickle-shaped anterior feet, again, the bristles present
a characteristic gradation from the slender terminal forms with
narrow wings (Plate V, fig. 53) to the spathulate kinds which
succeed, and which in varying proportions occur throughout
the greater part of the edge of the foot ( Plate V., fig. 54)
to the greatly developed series towards the base (Plate V.^
fig. 55) which not onlv have a much larger spathuUte
tip, but a thicker shaft, the figures being drawn to the
same scale. The flattened and tapered tips of the spathulate
bristles have probably special functions in progression, and in
the various movements in the manufacture of the tube, just as
the powerful bristles of the fourth segment have in these and
other respects.

The hooks of the anterior region (Plate V. fig. 56) and those
of the posterior agree with those from European waters, though
the figures just quoted may not correspond in all cases. Thus
in the able memoir of Joyeux-Laffine these hooks (Arch. Zool.
Exp6r. 2d Ser. viii., PL XV., f. 4) are represented by an example
which has the processes for the tendons at each end too large, as
IS also the tendon itself, and either an unusuallv long hook or
one not fully on its flat has been selected for illustration. The
French author has correctly indicated a double series of
striations on these hooks. In the South African examples the
most superficial striations slant downwards and forwards from
the posterior border of the hook, whilst by deeper focussing a
liner series pass from the teeth obliquely- — with a slight inclina-
tion downwards and backwards.

A former comparison of examples from the various British
areas in which it occurs with those from Naples led to the view
expressed by M. Joyeux-Lafline,* viz., that so far as these go
one species alone is concerned. The same conclusion is
reached when the form from the Cape is included, for in all
essential particulars it is identical with the European examples.
This opinion is shared by Mr. Cyril Crossland, who is doing

* Arch. Zool. Exper. 2d. Ser. viii. pp. 346-50.

63

such good work amongst the annehds collected by himself at
Zanzibar, and to whom the specimens had been shown before
then- final examination. The present examples are compara-
tively small — much less than those from Herm and Guernsey,
or even from Shetland.

So far as can be gleaned from Stimpson's account, the same
form is referred to, though he places the great bristles on the
3rd. foot. The general description and the number of teeth in
the uncini agree. He found his examples at Simon's Bay, Cape
of Good Hope. Schmarda procured his Chcetoptenis hauiatus
(op. cit.) from the same bay, and though his description is lax,
there is little doubt, taking his figures also into consideration,
that he refers to the present species. He mentions that the
tube is constructed externally of fine sand-grains, whilst in-
ternally it is smooth. His figure of the animal in its tube gives
a good idea of the relative proportions.

Cluvtopicnis variopcdatits is one of the most beautifully lumin-
ous annelids, bright flashes being emitted from the posterior
feet, but the most vivid phosphorescence is at a point on the
dorsum between the lateral wings of the tenth segment. Here
the copious mucus exuded by the animal can be drawn out as
bluish purple fire of great intensity, which besides, now and
then gleams along the edges of the wing-like processes and
illuminates the surrounding water. A very characteristic odour,
somewhat resembling that of phosphorus in combustion, is
given out by the animal during such experiments, and in this
connection it may be observed that Quoy and Gaimard mention
that an odour similar to that around an electric machine is
produced by luminous marine annelids. An elaborate account
of this phenomenon is given by Panceri,* who concludes that
the luminosity arises in special epithelial cells.

The species of Clia'topfci iis stand much in need of revision,
and when this is carried out it will be found in all probability
that the species here mentioned (C. vai iopedaius) has a very wide
distribution and has been described under various names as
new species ; nor is this surprising in an annelid which frequents
the shores of at least three great continents, viz. : — Europe,
Africa, and America.

Fam. ARICIIDJE.

Theodisca (Anthostoma) hexaphyllum, Schmarda, 1861.

1861. Antliostoiiui //d'avi'/>//\7//n;/, Schmarda, Neue wirb.Thiere,
I., II., p. 61, Taf. xxvii., fig. 217.

Obtained between tide marks at St. James', False Bay.

Atti d. Accad. d. Scien/e. Xapoli. 1875.

64

The head forms a small conical flattened process devoid of
eyes or appendages.

Body of variable length, ranging from 135 to 200 m.m.,
somewhat tetrahedral in form, slightly tapered in front, and more
gently posteriorly, but the tail is by no means slender, and it
ends in an oblique funnel directed ventrally, with a somewhat
frilled margin and a ventral process with three papillae. So
frequent is this organ in process of reproduction that it is
difficult to say what its precise condition is. In many it is
bordered externally by a frhige of the developing and rudimentary
feet which resemble papilhe. The latter, indeed, seems to be
the normal condition, the frilled lips of the anal funnel being
sometimes separated by a smooth margin due to the papillose
continuations of the feet. All that can be said of the adult
specimens is that the tail ends in an anal funnel, the folds of
mucous membrane internally forming six or seven symmetrical
plaits on each side, whilst its rim is smooth. The dorsal surface
is somewhat hollow, the ventral convex.

The first segment is acha^tous, and has the mouth on the
ventral surface, out of which protrudes the foliaceous mass
which gives origin to the name of the genus, and which is
is the proboscis, as in allied forms. In partial protrusion three
lobes are visible, whilst in full extension there are six or more.
The organ forms a complex series of folded lobes the grooves
of which converge to the central one. It can be entirely retracted.

The second segment has dorsally two processes, an inner
flattened lamella with a blunt tip, and a tuft of long, slender
and nearly straight bristles with closely arranged serrations,
springing from its inner side. Externally is a shorter bluntly
conical process arising behind a dense row of shorter and
stouter bristles, th^ longer showing serrations, the shorter with
blunt tips and'only a trace of the serrations, the former lamella
pertains to the dorsal region, the latter to the ventral division of
the foot.

At the sixth bristled segment another conical process
(branchiae) appears on the inner side of the longer lanceolate
leaf and is continued backwards. Whilst the former process
remains more or less lanceolate, the outer lobe at (his foot
presents a broad lamellar base, which by and by extends
downwards as a ridge behind the bristles.

The arrangement just noted reaches its maximum at the 25th
bristled foot ( Plate V., fig. 27 ), which has dorsally and ventrally
a somewhat enlarged and tapering inner gill, then the lanceolate
lobe and its tuft of long serrated bristles, and, after an interval,
a short conical process and the long ridge behind the bristles.
The latter consist of a long row of arcirate blunt forms, of
others with slender serrated tips, and, ventrally, of elongated
serrated intermediate forms. The cjndition resembles that in
such forms as Avicia Ciivicri.

65

Behind the 25th foot the inner gill rapidly increases in size
and inchnes inwards over the dorsum, the tips turning forwards
when they meet in the middle line. Coincident with the change
in the inner gill, the dense row of ventral bristles diminishes in
size and strength (the 28th foot terminating the larger anterior
series) so that at the 45th foot (Plate V., lig. 29), the ventral
division has assumed the shape of a short broad leaf with
a few short blunt bristles in front of it- ventrally a short
tuft of linelv tapered serrated bristles, and dorsally a longer series
having the same structure, Dorsally is the longer lanceolate
process, which like the ventral has its convex base external, and
the long tuft of tapering serrated bristles. Internally is the
vascular and richly ciliated gill, the tip of which is tapered and
•devoid of cilia.

At the 70th bristled foot the change is small, viz., a slight
diminution of the ventral lobe which also has lost its external
convexity at the base.

Little alteration occurs towards the tip of the tail except the
increase of the dorsal lanceolate lobe which is carried erect, and
the slight diminution of the branchiae on the more limited
surface, for they are continued to the anal funnel. This region
of the bodv is readily reproduced, but its precise condition in
the adult as contrasted with the young is still in need of eluci-
dation.

A whitish encapsulated body occurred on the ventral edge of
an anterior foot. The capsule was filled with somewhat coarse
granules.

A small Isopod was found on the surface of an example, but
its presence may have been accidental. The head had tw(3
black eyes, and was followed by 7 segments and the
telson.

This species is at once distinguished from Aricici Luvici-i. by
the condition of the dorsum anteriorly, and the absence of the
pectinate rows of papilhe in the same region laterally. It is,
however, in all respects an Aricia.

Schmarda's genus Aiifliosfoiiia, which he constituted for the
reception of the species from the Cape and one from Jamaica,
had little that was definite in it, the characters being as follows : —
Tentacles and tentacular cirri absent. Segments differing. A
lobate proboscis. Branchiae (cirri) with the exception of the
anterior segments, three in number, and so with the bristle-tufts.
Capillary and aciculate bristles. Fritz Miiller appears to have
included the same type in his genus Thcodiscd — afterwards more
fully exemplified by Claparede, whose diagnosis is : — Ariciidie
with strap-like branchi.'e ; distichous feet the inferior division
bilabiate. Proboscis extensible as a digitate membrane. Head
devoid of tentacles.

66

The relationship of the Mediterninean forms, viz., Tlicodisca
anserina, and Th. liriostouui to the South African is un-
certain, but the latter (7. liriosfoiiia) comes very near it.
Kinberg* records another species {A. dendriticiiui) from
Vancouver.

The Anthostoma robiistiiiii of Verrillf seems to have a very-
similar structure, and the same may be said of the Aricia
Icevigaia (Grube) of De St. Joseph,^ from the Mediterranean.
The species described by Treadwell || from the Antilles are very
similar, as also is the Aricici setosa of Verrill,§ from Flatts bilet
beach, Bermuda — in shell-sand at low tide. In the latter form
the branchii« also begin on the sixth setigerous segment and
continue nearly to the end of the body.

A small form about 5 or 6 mm. in length presents certain
differences from the foregoing.

Head forming a small blunt cone — larger in proportion than
in the adult.

Body very little tapered in front — more distinctly diminished
towards the tail — where it ends in 4 short cirri (Plate V., fig. 28.)

Each segment has similar appendages to those of the adult,
viz., an inner ciliated branchial process, a dorsal lamella with
the tuft of tapering finely serrated longer bristles, a conical
ventral lobe with a few spines, the tips of which have an f-shaped
curv'e, and a group of shorter serrated bristles as in the adult.
The f-shape of the tips of two of the spines is noteworthy at
this stage.

The presence of 4 short subulate anal cirri — two dorsal and
two ventral, if this be the same species (and it appears to be) is
therefore a feature of moment, and shows that a change occurs
during the development of the anal cup. At this stage, there-
fore, if the above interpretation is right, the posterior end
agrees with that of Tlicodisca as described by ClaparedelF and
Cunningham and Ramage,** but Claparede does not mention
bilid bristles or hooks, whereas Cunningham and Ramage do, so
that it is dcubtful if the latter pertains to the same genus. At
any rate the young form alluded to in the foregoing sentences
agrees with the Tlicodisca of Claparede ; and in all probability
the adults have the same appendages.

* Olvfi^. K. Vetensk-Akad. Fork. iS60. No. g, p. 337.
t Op. cit. p. 597, PI. xiv., fii;. 76.

X Ann. Sc. Nat. 8e., Ser. v., p. 360., I'l. wi, 1". 168-175.
II Bullet. U.S. Fish Com. lyoo, p. 203.
§ Trans. Conn. Acad, x., p. 651. ic;oo.

* Glanurcs Zoolom. p. 44, PI. iv., iigf), and .Vnn. Chcl. Xap. p. 310. Pl.x.\iv.,fig. 3.
** Trans. K.S.E., vol. xxxiii., p. 642, PL xl., tig. 8.

67

Fam. CIRRATULIDJE.

CiRRATULUS CIRRATUS, O. F. Mitller, 1776,
( = capensis, Schmarda).

1776. Lunibyicus ciyratiis, O. ¥. Miiller, Zool. Dan. Prodr., p.

214.
1843. Cirratnliis horcahs, CErsted, Ann. Dan. Consp., p. 67.
1843. „ „ Id. Groenl. An. Dors., p. 54.

1856. ,, aitstralis, Stimpson, Proc. Acad. Nat. Sc,

Philad., p. 392.
1861. Chraiulns capensis, Schmarda, Neue wirb. Th. I. ij., p.

56, Taf. xxvii., fig. 213.
1865. Cirmtiiliis aiisfmlis, Johnston, Cat. B. M. 210.
1867. ,, ,, Malmgren, Ann. Polych., p. 205.

1885. Cirratnliis capensis, Mcintosh, Ann. " Challenger," p.

383, PI. xxivA., figs. 9 and 10.

The two Cape species form a parallel to the two common
British forms, viz., C. tcnfacnlatns and C. cirrafns.

This form takes the place of the Cirratnliis cirratits of the
British area, and it is very closely allied to it, so closely that
it has now been thought unnecessary to separate them.

The head is less horse-shoe shaped and more pointed, and it
is devoid of eves — so conspicuous in the British form. The
buccal segment is of considerable breadth, and is followed by
another segment free from bristles. The third segment bears
hooks and bristles, as well as a branchiae dorsally, close to the
bristle tuft. The fourth has a similar arrangement. At the
fifth is the dense group of branchiae on each side, but the fila-
ments spring from one or two of the succeeding segments.
For forty of the following seganents, the branchiae arise close
above the dorsal bristles, but they afterwards, and to the tip of
the tail, have a considerable interval between them an.d the
dorsal bristles, in this respect agreeing with the British examples.
As in the British form, the size of these branchiae is in contrast
with that of the other species from the Cape.

The first ventral tufts are slender and bristle-like, but soon
two or three dark brown hooks are found in each segment
(Plate vi., fig. 30), the shape being f-like, with a forward bend
at the point, and considerablv stouter than in the next form.
As a rule, only two are found in the posterior region of the
body. Dorsally is a tuft composed of three long and finely
tapered simple bristles — alternating with three slightly curved
and somewhat longer spines than in the ventral series.

68

Posteriorly, the ventral hooks increase in size and strength, but
the dorsal tuft has the same arrangement as in front.

In the intestinal canal is muddv debris, containing fragments
of Crustacea, sponge-spicules, diatoms, and other structures.

The branchiae are, on the whole, more numerous than in the
British examples.

Stimpson first found this form in the " circumlittoral zone,"
at False Bay, Cape of Good Hope, and he describes it as
greenish or reddish brown, nine inches in length, and having a
breadth ol -45 inch. He points out that the ulterior bristles
(except anteriorly) are short, stout, and arranged in groups of
three. Schmarda gives the examples he procured at Table Bay
a violet colour. The specimens obtained by the " Challenger "
came from Sea Point, near Cape Town. Langerhans* met
A\ith this species, three cm. long, in the Canaries, whilst
Marenzellerf includes it in his account of the Annelids from
Angra Pequena. He notes that Stimpson's form is the same,
and, therefore, that his title should have priority to Schmarda's,
but this is now of less consequence, since the form is identical
with O. F. Miiller's.

CiRRATiLUS TENTACiLATrs, Montagu, 1808.

1808. TcrcbcUa icntaculatd, Montagu, Linn. Trans, ix., p. no.
1834. Cirniinhi^i Laiiuiirki/, Audouin and Edwards, Annel.

p. 271, PI, VII., figs. 1-4.
1865;. Ciiratiiliis icntaculatus, Johnston, Cat. B.M., p. 209.
1868. Aiiiloiiinid filiilcra, Claparede, Ann. Nap. p. 267, PI.

XXIII., fig. 3.
1889. Cirnifiihis fciiinciihdiis, var. iiicriilioiidlis, Marenzeller,

Zool. jahrb. iii., p. 16 (sep. abdr.) Taf. I., hg. 7A.

This form takes the place of the British C. tciiliuiilatiis dwd is
larger than the preceding.

The head is bluntly conical and devoid of eyes. The peristom-
nial and the succeeding segments have similar proportional
breadth.

The body is elongate, measuring from three to five inches in
length, tapered at either extremity, and furnished with numerous
long, slender cirri. The first cirrus is attached above the fourth
bristles, ;uid another is over the fifth and sixth respectively ; then a
dense group occurs in a tran.sverse series to the inner side of
the latter, a short gap in the middle line separating the two

* Nova Acta. Bd. xlii., Xo. 3, p. 115.
t Zool. Jalirb. iii., p. H) (sep. adr.).

69

sides. This barrier indicates the division between the wider
anterior and the narrower segments which follow. The cirri
continue along the sides backward to the tail, on every segment
and close to the dorsal bristles. They are on the whole longer
and more slender than those of the preceding species, just as
occurs in the British forms.

The anus is a longitudinal dorsal slit at the tip of the tail.

The bristles in this species are longer than in the former,
both ventral and dorsal of the hrst bristled segment being
slender, long, and finely tapered, and the tips are densely
covered with a slender filamentous algoid. For a long distance
backward both dorsal and ventral bristles remain capillary, and
the middle of the body is passed before the ventral hooks are
well developed (Plate VI., fig. 31). They are proportionally
longer and more slender as well as more numerous than in the
former species, with a slight bend of the shaft and a marked
curve at the tip. A slender bristle or two occurs amongst them.
The dorsal bristles retain the characters they have in front to-
the tail, so that in this respect alone there is a marked difference
between the species.

The majority of the specimens were loaded with ripe ova.

This species resembles the British Cirratulus tentaciilatiis in
several respects, but differs in the number and shape of the
ventral hooks, which are shorter and fewer in the British form,
in the segments of the anterior region and other particulars..
In both the cirri spring close to the dorsal bristles.

The parallelism between the two species of Cirratulus at the
Cape and the two common British forms is of great interest,
and, allowing for the variations due to the divergent environ-
ment, it has been thought proper to unite the respective species^
This opinion to some extent is that of Marenzeller, who makes
the examples from Angra Pequena only varieties of the European,
species {I'ar. uici idioiialis, Marenzeller). Moreover, if Aitdoui-
nia filigera, Delle Chiaje, is synonymous then this common
form ranges to the Straits of Magellan.*

Fam. HALELMINTHID^.

The Hcdclniiiiihidcv or Cupilcllidivwere, represented by several
examples, one or two of which were devoid of branchiae,
whilst all the rest had these organs. As, however, both had the
same number of capillary bristles (13 pairs) and segments
("thoracic") in front, and both agreed in the arrangement of

Ehlers, Polych:t;t. Ma^yalhaen, Hamburj?, i^<97, p. iio.

70

the rows of hooks (toii), in all probability all the specimens
pertain to the genus Dasybranchus, those devoid of branchiae
having either retracted or lost them.

Dasybranchus caducus, Grube, 1846.

1846. Diisxnuilliis caducus, Grube, Archiv f. Naturges, 1846,
p. 166, Tab. V. figs. 3, 4.

1885. Dasybiriiiclnts, sp. (caducus \ Mcintosh, Ann. "Challen-
ger," p. 290, PL XLV., figs. 13, 14, PI. XXI Va, fig. 17.

1887. DasyhiancJiHS caducus, Eisig, Capitell, Nap., p. 823, et
ubique, Taf. 16-23.

The examples vary from 90-170 mm. in length, and the largest
has a diameter at its widest part of 7 mm., a size considerably
exceeding that of the European specimens, though not much
larger than the Japanese form found by the naturalists of the
" Challenger."

The conical snout resembles that of Xotoniastus, and the
number of distinctly bristled segments of the anterior region is
13. The bristles are long and slender, with a filmy margin
indicating the presence of narrow wings.

The dense rows of minute hooks are similar in structure to
those of Notoiuastus, having a main fang and several small
points above it, but the shaft appears to have a more distinct
dilatation, and the larger wings are united distally so as to form
a kind of hood.

The branchia' occur only on the posterior part of the bod}' —
as small tufts of simple or slightly branched filaments arising
on each side of the ventral pads, and thus are comparatively
near each other. The ease with which the branchia* are retracted
or lost accounts for their varied appearance. Those newly
reproduced are short and simple, the older forms are more or
less branched and larger.

There is little to distmguish this form from the European.
In all essential respects it corresponds both generically, and so
far as can be at present observed specifically also with Dasy-
branchus caducus, Grube, a form which has been found at
Madeira and other parts of the Atlantic, in the Mediterranean,
in the Indian and Chinese seas and in the Pacific.

The HyboscoU'x lougiscta of Schmarda* is probably the same
form — procured at Table Bay.

* Neue wirb. Thicre, I. ii., p. 54, Taf. XXVII. fig. 211.

71
Fam. MALDANID^

NICOMACHE Li'MBRiCALis, (O. Fabriciiis), var. CapensTSj-

(Mclntosh).

1780. Sahclla luinbricalis, O, Fabricius, Faun. Groenl, p. 374,

1856. Clymene liiinbi icalis, Sars, Faun. lit. Norveg, II., p. 16.
Tab. 2., figs. 23-26.

1885. Nicouiache capcnsis, Mcintosh, Ann. " Challenger," p.
399, PI. XLVL, fig. 4., PI. XXIVa., figs. 18, 19, and PI.
XXXVI lA., fig. 2.

A species of considerable size, probably reaching 130 mm. or
more in length, and having a diameter of 5 mm.

The cephalic segment is enlarged, suddenly narrowed into a
blunt ridge dorsally and running downward to the thickened
oral border. On the ventral face the oral margin projects as
a frilled ring, more or less dilated according to its condition.

Though on a comparatively gigantic scale in contrast with
the British forms it is remarkable how closely the outline of the
snout, its touches of reddish pigment, the form of the anterior
segments, and the general conformation of bristles and hooks
correspond.

The first four sets of bristles are stout, simple, tapering,
slightly winged forms, the fourth having also a few very
slender, almost capillary bristles. The fifth group has a few
slender forms amongst the stronger with very minute spikes,
and in the sixth and seventh these increase in distinctness till the
condition shown in the figure of the "Challenger" is attained.
Beneath each of the first three is a powerful spme, or occasion-
ally two in the last. Under the fourth is a short row of the
hooks characteristic of the species, and in the succeeding feet
these increase considerably in number so that the rows are
longer.

In comparing the loth bristles and hooks of the British and
South African forms little distinction is apparent except in size.
The types are the same. It is true a larger number of spikes
occur on the crowns of the hooks of the South African form,
but this is due to their size.

The funnel (Plate VI., fig. 32) in the South African has from
14 to 30 papilla' round its border, whereas in the small British
examples the number ranges from 14 to 20.

In a South African example there were 20 bristled segments,
the entire animal being about 180 mm. in length. The smaller
British examples appear to have the same number of segments
hough none were complete.

72

Tlie tube is composed of whitish or translucent grains of
coarse sand with frafjments of shells and coral bound together
by the secretion. Though comparatively thick it is readily
broken.

It is difticult to make out the species described from the Cape
by Schmarda,* under the names of Clymene mioocephala, and
Clvuii'iic Ixroccpluihi. The former may have lost its head, and
the latter might be Xicomachc viewed from the ventral surface
with the proboscis partially protruded.

The wide distribution of this form and its varieties is note-
worthy, for it frequents both shores of the Atlantic, that is
Europe and Africa on the one hand, and America on the other.

NicOMACHE McIntoshii, Maren/.eller, 1889.

1889. (?) Nicomache McIntoshii, Marenzeller, Zool. Jahrb. III.,
p. 19, Taf. i., fig. 8.

A fragment of the posterior end apparently of this form
comes from St. James', False Bay, between tide-marks.

The body ends obliquely in a large ovoid expansion (Plate
VI., fig. 33) which is fixed to the dorsal (?) surface, but extends
elsewhere all round as a broad rim, with a slight notch in the
mid-ventral line. The anus forms a prominent cone with 13
crenations on the margin and situated slightly nearer the ventral
than the dorsal edge. A considerable area posterior to the
vent is dotted with low flat papilUv like variola?.

After the bristle-papilla and row of hooks on each side the
last segment narrows to the segment-junction. The segment
in front is short, with a peculiar dorsal flap or process which
is somewhat shield-shaped — the median curve posteriorly being
close to the segment-junction, though elevated above it, and
a lateral curve on each side running forward to the bristle-
papilla — which is nearly in the centre of the short segment.

The segment in front is also short, and has a similar modified
dorsal shield with a central prominent edge just behind the
bristle-papilhv ; and a similar, though less distinct shield occurs
on the third from the end.

The fourth bristle tuft from the posterior end shows rather
stout but finely tapered winged bristles.

The hooks (which are probably posterior) have a broad and
strong head with a powerful chief fang, and four prominent sharp
processes with the points directed forward above it, but of course
other hooks are present in the crown, these being apparent
distally in lateral view (Plate VI., fig. 34). A filament passes

* Op. cit., p. 15, Taf. XIX., tigs. 163 and 164.

73

from the neck of the hook and curves over the tip. The shaft
diminishes from the crown to the shoulder near the anterior
third, and then narrows with a curve to the base.

Marenzeller had complete examples of this Maldanid from
Angra Pequena, somewhat further northward on the eastern
coast of xlfrica, but they seem to have been much smaller,
only from 40 to 45 mm. long and 2*5 mm. in breadth, whereas
the fragment from the Cape measures 48 mm. and is about
3 mm. in diameter, so that in all probability the entire animal
would not be less than 60 mm.

The head appears to be less developed than in N. liimbricalis,
though having similar characters. The first six bristle-bearing
segments are shorter than the three following. The first three seg-
ments have simple hooks after the plan of those in the common
species, the rest have the characteristic hooks, and Marenzeller's
figure (fig. 8 E) of the anterior hook difi^ers only in comparative
length, being slightly shorter than that from the posterior
fragment from the Cape. He also describes the anal plate as
granular, but apparently the small size of his examples did not
exhibit the variolar condition so conspicuous in the larger
specimen from the Cape. No tube is mentioned by this author.

The posterior end of Grube's Muldanc glcbifcx* is bevelled
somewhat in the same manner, but it appears to be on a much
smaller scale proportionately. The posterior end of Mnldanc
cloiigcita, Verrill,t has an obliquely truncated posterior border
surrounding the base of the large anal process, which is
obliquely placed, foliaceous, obovate, with the posterior edge
broadly rounded, the upper surface concave, and the margin
entire.

The posterior end of the Makhiiic dinpliiglxpta of Ehlers,|
from Magellan, has likewise a free flap, but of a different char-
acter from the South African form.

Unfortunately no figure is given, so that it is difficult to
make a minute comparison. The American form was got
at low water mark near New Haven in thick tubes of fine mud.

Pkaxilla PR.^TErmissa, Malmgren var. CAPENSIS, Mcintosh.

1885. Pi'^i'^illa Capeiisis, Mcintosh, "Challenger" Annel., p.

404, PI. XXXV A., fig. 8.
1867. Praxillapra'tcrinissa, Malmgren, Nord. Hafs-Ann.p. 191.
1867 „ „ Ibid., Ann. Polych., p. 100, Taf.

xi., f. 62.

A form about 250 mm. in length and proportionately large.
The head has a frontal shield like that in the European Praxilhi

* Archiv. f. Naturges, i860, p. 92, Taf. IV., 1. 4,

t Op. cit., p. 609.

X Polychaet. Magalhaen, Sammelreise, Hamb. 1897, p. 119, Taf. VIII. f. iSj-ii;^.

c

74

pnvtcuuissa, though on a much larger scale than in British
forms. The lateral folds of the cephalic region are very
distinctly marked, though not long. On each side of the
median in front is a well marked division, followed by three
larger lobes which are notched in the centre. Behind are about
six crenations before the median dorsal furrow, with indications
of a tendency to form larger lobes. The median and two
lateral ridges are similar to those of P.pnvicnuissa. The buccal
and the following segments, and to some extent the vent are
marked by creases like hard grained morocco leather.

The body has 21 segments besides the buccal segment and
the funnel. It is rare to find a complete example in Britain,
and only one occurs in the South African collection. The
funnel has from 22 to 26 teeth.

The bristles, spines and hooks appear to be very closely
similar to those of the British forms though all are larger.

The tubes are formed of fine and coarse sand grains and
fragments of shells — lined by secretion, and are friable whitish
structures.

The preparations from the " Challenger " material were
labelled Pmxilla praicniiissn var. capciisis and it was the connate
condition of the dorsal longitudinal muscles that led to
separation. It is doubtful if this is sufficient to sepiU'ate forms
so closely resembling each other.

The distribution of the species would seem to be very wide.

Fam. HERMELLIDJE.
Sabellaria (Pallasia) Capensis, Schmarda, 1861.

1861. Hemiella capciisis, Schmarda, Neue wirb. Thiere I. ii.,

p. 23, Taf. XX., fig. 171.
1885. Sabellaria (Pallasia) Capciisis, Mcintosh, Ann.

" Challenger," p. 418, PI. xxv.A, figs. 24, 23, PI. xxvi.A.,
figs. II, 12.

The specimens were procured at St. James', False Bay,
between tide-marks. Those obtained by the " Challenger "
were collected between tide-marks at Sea Point, near Cape
Town, where it was first found by Schmarda, and subsequently
by Kin berg.

The largest example, including the tail, measures 105 mm.
in length, and it is evidently a richly coloured form. Schmarda
states that his were olivaceous, but in many parts were purplish
brown.

The species is allied to the ordinary Sabellaria^, differing
from the common southern form — Sabcllaiia alvcolaia — in

75

having only two rows of paleae, the shape of the inner being
siicli that it compensates for the absence of the second inner
row. Though Schmarda described and figured the palea^ as
notched at the tip, the condition differs, since the thin chitinous
spathulate tip has a fold on the under surface the end of which
projects dorsally in the form of a spur. When viewed in
certain positions, as obhquely from above, the appearance of a
notch is simulated by the arrangement. On the other hand a
lateral view gives a hatchet-shape to the blade, and the terminal
spur is very prominent. The palea3 are crossed by curious
transverse lines, which, especially along the convex edge,
assume a wavy direction, and, moreover, a scaly aspect, appar-
ently from friction, occurs along the same edge near the tip.
Whilst it is possible to recognize these paleiv in Schmarda's
description and figure, it is otherwise with the inner palea-. for
the contour of the latter, according to him, corresponds with a
lateral view of the outer form — as a comparison of his figures
with those of the " Challenger " will show.

The tip of the inner pale;i3 is bluntly pointed, then it gradually
expands into a large thick heel which projects beyond the
somewhat slender shaft, so that the latter nearly forms the apex
of a triangle, and the serrated upper or anterior edge the base.
These are evidently modifications of the same type. In this
case the outline resembles an attenuated leg with a long tapering
foot furnished with a huge heel, the latter having the dorsal
surface crenated. In one of the examples (a large one) two
blades of a broad Alga were attached by their stems to the inner
paleiv, as in the case of that mentioned in the " Challenger "
volume — and which had a single blade which overhung the
crown like an operculum. A little below the bases of the
outer palete of the crown a closely set series of lobate cirri with
narrow stems occurs They are tinted of the deep purplish
hue of the region, though in some specimens much of it is
abraded. The cephalic branchia? have about i6 stems on each
side of the fork. The mouth has a long cirrus, or occasionally
two, at the outer and posterior border of each palpus.

In the richly tinted thoracic region are the three typical sets
of bristles. The oar-shaped forms of the dorsal row have a
smooth blade. A high power shows only a few minute spines
at the point of the most perfect examples, few of which are
symmetrical distally. All have marked stria- at intervals. The
ventral bristles are shaped somewhat like the hairs of Otnitho-
rliviichus, the shaft dilating into a fusiform tip which is pointed.
The hooks have a double row of teeth and an indistinct process.

The branchia- are continued — from the first segment of the
long posterior region of the body — throughout the greater
part of its length, the posterior organs, however, being rudi-
mentary. They cease at the seventh or eighth from the end— to
which is attached the long anal tube.

C 2

76

No tube accompanies these examples, but in those brought
home by the '' Challenger," the structure was as follows : — The
tube was dense, composed of entire small shells, coarse fragments
of shells, sand-grains and other structures cemented by the tough
secretion which also forms a lining to the interior. This lining is
in many parts tinted of a dull purple, yet this does not prevent it
from being semi-translucent. As in many other tubes in the
collection the colour is pale, from the semi-translucent grains. It
is thus in contrast with huge and heavy masses formed by a Saln-l-
hiria brought from St. Vincent by Dr. R. M. Gunn, composed of
agglutinated tubes of coarse yet smoothly worn minute pebbles.
The majority are white, but dotted here and there are many
minute black pebbles which seem to be less rounded than the
white. Such a mass illustrates well the power of Sahellavia io
form structures capable of resisting the heaviest surf-waves.

As indicated in the "Challenger" volume,* an intricate
central muscular region occurs in the anterior division of the
body, e.g., behind the mouth, and it is especially regular and
beautiful in connection with the two median and two lateral
channels in the dorsal region. The nerve-cords are internal
and abut on the perivisceral chamber. Behind the foregoing
the circular muscular coat is well-developed. The dorsal longi-
tudinal muscles are much more massive than the ventral and
are continuous over the vascular channel in the median line.
The separate nerve-cords lie on each side of the median line.
In the posterior region the dorsal and ventral muscles are
proportionally small. The neur.il canal lies at the inner and
inferior region of the nerve-cord. The great central area of
the body is occupied by the distended alimentary canal, with
the reproductive organs at the sides.

The contents of the alimentary canal consisted of sandy
mud, numerous sponge-spicules, a few diatoms and organic
fragments, besides many Gregarime.

The same species was found by Marenzeller t in the collec-
tion from Angra Pequena on the West Coast.

Fam. AMPHICTENID^.

Pectinaria Capensis, Pallas, 1766.

1766. Nereis cyliiii/riiyid, Pallas, jNIiscell. Zool. p. 117, Tab..
IX, f. 1-2.

A large species measuring in spirit about 150 mm. in^length,
and having a diameter behind the pale?e of fully 12 mm.

* Op cit., p. 420. j'/a)u\. Jahrb. III., p. 21 (scp. :ibdr.).

77

Anteriorly the paluliv are fourteen on each side, set in a sliglitly
oblique row, and with a dorsal curve. Each group forms a
kind of stiff fan with the longest palea' in the centre, the outer
especially being shorter. A typical palea measures about 15
mm., is flattened and of a rich golden colour, iridescent under
the microscope, and marked with longitudinal and transverse
stria'. The tip is usually abraded — forming a transverse dark
edge. A wrinkled area slopes obliquely downwards, and is
bounded by a membrane which commences at the subulate
tentacle a little in front of the outer edge of each fan, and has
its free edge cut into processes, small at first and then triangular
and larger. Below the palea is a broad membranous fold with
long filaments at its free edge, which encircles the two dense
groups of tentacular cirri.

The body is broad in front and tapers after a short distance
gently to the tail, which ends in a sausage-like appendage,
having a broad furrow with thick folded edges dorsally, and
terminating in a tongue-like flap with a wide crenated margin.
Beneath the process is marked by fine creases (like leather) and
has a median groove and three lateral furrows which slope
outwards and backwards.

Anteriorly the dorsum has an elevated glandular whitish
thickening from which a tapering tentacle extends forward. A
ridge cut into flaps passes from the elevated mass and joms the
pre-oral fold. Another smaller whitish pad occurs behind the
former to the inner side of the first branchia. The latter is the
larger, and arises from the ventral ridge behind the mouth, is
fixed for half its length, the other or outer part passing as a free
fold dorsally. From the basal stem the branchial tissue splits
into fine lamella, so that aeration is readily carried out. The
second gill is similar to the first, but arises from the succeeding
glandular ridge further out, and its free stem extends further on
the dorsum. On the ventral surface in this region are four small
central areas or cushions connected with conspicuous bands
passing dorsally. The first lies in the ridge behind the mouth
and has a frilled (glandular ?) band in front of it, connecting
the inner (ventral) ends of the first branchia. The second has
a stift" glandular ridge passing from its anterior border to the
second branchia. The thuxl is the least, and from its outer
border a prominent fillet passes to and beyond the first bristle-
bundle on each side. The fourth is shield-shaped, and from its
anterior edge a flattened fillet extends to the side of the body,
but doe.s not reach the second bristle-bundle.

Sixteen pairs of bristle-bundles extend along each side, thus
the number differs from that found in Pcctiiun in, where it is 17,
the same niuiiber occurring in Amphktouc Cisfciiidt's and Pisfa,
whilst Lngis has only 15. This form therefore occupies an
intermediate position in this respect, though most nearly ap-
proaching Av////^//7V/ (Plate VII). The bristles (fig. 33 a and h)

7«

are somewhat stout, with tapering, winged tips ( a ) which often
show a spHt {b) in the wings from friction, thus giving the tips
a peculiar appearance, especially as a change seems to take place
thereafter. The last bristle-bundles are small, but show similar
structure, except that the shaft dilates towards the tip, which is
often split as in front.

The hooks (Plate VH tig. 36) approach those of Pcctiiuir'ni,
in so far as below the eight upper and boldly curved teeth is a
short portion of the margin armed with minute teeth above the
truncated inferior angle. As the tip of the last large tooth
curves over these they are often indistinct.

The alimentary canal wa.'^ distended with white sand — like
coral sand, much of which disappeared on the addition of acid.

The beautiful, straight tube formed by the animal was almost
entirely composed of spicules of sponges in short lengths
placed transversely and fixed by secretions so as to form a per-
fectlv roimd tube 115 mm. in length gently tapered from the
wide to the narrow end, the former having a diameter of 15
mm., the latter of seven or eight mm. The spicules appeared
to be of the same size throughout the tube, which presented no
special lining, the inner surface being as smoothly and neatly
formed as the outer, though under the microscope the minute
reticulations or cells of the cement often stretched over the
component spicules.

The labour involved in selecting ^.'^c\ fitting with such mar-
vellous skill the sponge-spicules comp-jsi-ig a tube so large
must have been both continuous and considerable. Moreover,,
the smaller end of the tube had a diameter of seven or eight mm.,
so that in view of P. Fauvel's statement that but one tube suffices
for the life of the animal, we are left in doubt as to how it
fared in its earlv condition. In the preparation about an inch
of the tail projected beyond the small end of the tube.

As stated, the intestine showed that the annelid frequented
calcareous sand, and, therefore, specially selected the sponge-
spicules for the formation of its tube. As this example was
procured between tide marks, the conditions are wholly
different from similar tubes formed by the deep-sea representa-
tives of the family.

The account of the genus Pectinaria by Malmgren differs
from the condition in this example in so far as the marginal
lamina of the area on the ventral aspect of the paluhv is not entire,
but has at each side a series of acutelv conical smaller limbria',
followed by larger and somewhat triangular processes of con-
siderable breadth, which occupy the central region, lliere are
16 instead of 17 bundles of bristles, but only 13 rows of uncini^
beginning at the fourth bristle-bundle. The minute structure
of the hooks corresponds with that of examples of Pectinaria
hclilica from various regions, and it may be that variation

79

occasionally occurs. In certain examples of Pcctiiuirin helj^icd
also the ventral lamina near the paleae is frilled and slightly
scalloped.

This large and fine species was first described and figured as
Teredo cJirvsodoit by Bergius,* and subsequently noticed by the
sagacious Pallas, who examined many specimens procured at
the Cape of Good Hope by Vosmaer, and he mentions that
Rumphius discriminated the same form as a third spc'^ies —
" Penicilli marini," Pallas noticed the structure of the tube,,
which he experimented with in various ways, but did not make
out its nature. His familiarity with the European species
{Pcctinaria hcloicii, his varictas Rclgiai) makes his description
of considerable value, and he would seem to have observed the
difference between the smooth dorsal edge behind the paleje in
the European and the slightly scalloped one of the South
African. The general description which he gives, indeed,
does credit to his acuteness and accuracy, and his figures of
the annelid and its tube are recognizable, the former being
shown of its natural size.

Ehlers includes Pcciiiuir'ui hclgica in his series from Magellan^
so that much has yet to be done in the distribution of this,
species.

Fam. TEREBELLID^.
SCHMARDANELLA PTEROCH^TA, Schmarda, 1 86 1.

1861, Tcvcbclhi picrocluvtd, Schmarda, Neue wirb. Thiere I.,

II., p. 43.
li^S^. ScJiinanldiiflld ptcrocJuvta, Mcintosh, Annel. " Chal-
^lenger,"p. 449, PI. Llll. fig. i, PI. XXVII. A. figs. 24-26.
1889. Lcpren ptcrocluvia, Marenzeller, Zool. Jahrb. III., p. 21

(sep. abdr.j

Obtained between tide marks at St. James', False Bay.
Schmarda simply states he found it at the Cape. It is probable
that he procured it in the same way as the naturalists of the
" Challenger," who collected it between tide-marks at Sea Point,
Cape Town. The specimens were scarcely so fine as those
procured by the " Challenger," but traces of the greyish or
olive-green pigment occurred on the dorsum.

As shown in the " Challenger " Annelids this form agrees
with Nicolcd in having two branchiae, but instead of 15 pairs of
bristle-bundles it has 33. The bristles, moreover, are diagnostic,

* Abhand. d. schwed. Akad. deutsch v. Kastner XXVII. p. 235, Taf. IX.
(tide Grube).

8o

for whilst the lower part of the tip is winged as usual, a special
terminal modification in the shape of a tapering pectinated
region occurs {vide "Challenger" Plates XXVII.A., figs. 24 and
25). The tips diminish in length from above downwards, and
the wings become broader, the pectinated region at the tip
being longer and more evident. The pectinations, indeed, are
visible in the developing bristle in the interior of the setigerous
lobe. The hooks had three or four small teeth above the
great fang, while the posterior margin forms an almost con-
tinuous and uniform curve with the ventral or basal margin, a
small mucro only indicating the separation. The posterior
hooks present very little difference in form.

The examples were females laden with apparently ripe
ova.

Marenzeller* considers that this form should fall under the
genus Lcpira, Malmgren, as shown by him in his careful classi-
fication of the group, and there is no objection to this arrange-
ment in the meantime. He procured various examples from
Angra Pequena-Bucht.

Thelepus

The examples of this form are much softened so that an
accurate description of the exterior is not possible. In general
aspect, however, they agree with Scottish forms of Thclcpits
cincimuitiis

The hooks (Plate VII., fig. 37) have a well-marked dorsal
hollow, and the inferior margin terminates in a rounded process
which projects almost as far as the mucro above it. Only a
single hook occurs above the great fang. Posteriorly a
pointed process projects from the angle as in Grynuva, but there
is no tapering process beneath the mucro in front for the
attachment of the ligament as in Stirblosoiiia (Gryiiuva), which
also has two teeth above the great fang. A further noteworthy
feature is the tendency to the elongation of the stem of the
mucro ; and a slight tilting forward of the process as in Eiithclcpus
st'iiibdlt'iisis.'t The hooks of Tliclcpus sctosiis, De Quatrefages, as
figured by De St. Joseph $ in his well-known *' Annelides Poly-
chetes des Cotes de Dinard," approach the South African form
very closely. The chief differences are the presence of two
hooks above the great fang, the less graceful curve of the base
and the more erect mucro in the French form.

* Sitzunjish. d. k. Akad. d. Wiss. i Abth. Jhnji, 1S.S4, p. iX, etc.
t Ann. "Challenger," p. 465, PI. xxviii.A, tig. 13.
+ Ann.Sc. Nat. 8e Ser., xvii., p. 230, PI. lo. i. 260.

8i

Fam. SABELLID^.
POTAMILLA KEXIFORMIS, O. F. Milller.

1800. Die iiicrciifonnigcn Ainpliifrifc, O. V. ]\Juller, Xaturges.

Wurmarten, p. 194, Tab. xvi., tigs, i, 2 and 3.
1849. Sabclhi iriiifonnis, Leuckart, Arch. f. Naturg. xv, i,

p. 183, Tab. 3, f. 8.
1865. Sahelhi saxicavci, De Quatrefages, Anneles, II., p. 4^7,

PI XV., f. 1-7.

1867. Potaiuilhi iriiifoniils, Malmgren, Ann. Polychiet., p.
114, Tab. xiii., f. 77.

1868. Sabclla sax/awa, Mcintosh, Ann. Nat. Hist., 4th Ser.
' II., p. 286, PI. XX., f. 5-8.

1894. Potamilla rciiifoniiis, De St. Joseph, Ann. Sc. Xat. 8e
Ser. XVII., p. 292, PI. xi., fig. 296-298.

A species which shows a considerable amount of purplish
brown pigment on the anterior region of the body and on the
bases of the branchiie even in spirit.

The collar anteriorly has a median furrow on the dorsal
surface, with a slightly frilled triangular lobe on each side.
The rim slopes obliquely downward and outward, making" a
shallow notch on the dorsum before reaching the line of the
bristle-tufts. It then curves to the ventral surface, the rim on
each side being separated by a median fissure. Dorsallv, a
median furrow passes backwards a short distance, not beyond
the third bristle-tuft. The anterior region of the body consists
of eight segments, marked by the bristle-tufts and pads. The
ventral median groove goes forward to the second segment
behind the anterior region, then slopes to the right towards the
furrow between the anterior and the following region of the
body, but is not continued on the dorsum.

The length of the largest example was 80 mm. hut the
posterior region was imperfect.

The branchiae are about fourteen in number, appear to be pro-
portionally short, and the filaments proceed quite to the tip.
On viewing the fans from the dorsum, the rachis in several has
three distinct pigment-specks, in others one or two, so that the
branchiae have a characteristic appearance. The purplish
brown specks project from the line of pigment on the anus as
globular bodies. Their number is variable, but none liad so
many as those described by Langerhans,* (his var. polyopJitJinl-
inos with six or seven eye-spots), from Madeira. The branchii\3
in the British examples are often speckled green and white.
The tentacles are somewhat lanceolate and pointed processes.

* Zcit. f.w. Zool. Bd. XL., p. 266.

82

Tlic first biistle-tuft occurs immediately behind the collar
and is unaccompanied by hooks. The bristles of the anterior
region form vertical rows, the upper eight or ten bristles have
tapered tips with wings (Plate VII., hg. 38). They are followed
by two rows with spathulate tips (Plate YII., fig. 39) the upper
row having longer tips often with a central filament, and in
some indications of transverse rows of spikes, the lower having
broader tips (fig. 40).

The anterior hooks (Plate VIL, fig. 41) have a prominent
fang and a crest or crown with fine serrations. The anterior
curve is bold, and the posterior process long.

Posteriorly the tips of all the bristles are produced as long
processes, the wings being rather abruptly widened above the
slightlv constricted end of the shaft (Plate VII., fig. 42). The
posterior hooks are smaller, have a rounded anterior curve and
a shorter posterior process.

The tubes inhabited by this form are often grouped together
in longitudinal bundles, and are composed of tough hardened
secretion coated with sand. They resemble those of the
British ''Sabclhi siixicaiui," though they are larger. In the
same way the tips of the horny tubes are sometimes closed by
being bent. In the larger tubes the grains of sand are coarser,
and fragments of shells are added. In all cases the horny
secretion is characteristically firm and niore or less brittle. In
one the tubes projected from a whitish mass of a compound
ascidian, the sandy tube being in the centre and forming the
axis on which the ascidian was supported. De St. Joseph finds
that the tubes occur in a similar way between Cyiithici gloiiicnifd
and the granite rocks. In the Channel Islands as well as on
the southern coast of England this species is very common.
In the former it is abundant in oysters, Pcctcii, Aiioiiiid and
other dead and living shells, in Balani covering the sides of the
Gouliot caves at Sark, the tube is coiled beneath the Balani and
pierces the latter to reach the surface. The species likewise
perforates CcUcpora, and even bores quite through the valve of
a living Pi'cfcii />//.->/o. It often occurs in the same oyster shell
with Giisirocluvita, Polvdora and Clioiic, and it sometimes-
places its tubes in groups in convenient fissures of the shell
without boring, so that they can be dislodged cii masse, as has
been the case in a series from South Africa. Another site is
under emptv limpet-shells amongst muddy debris, part of each
tube being inserted into a perforation in the shell ; while again
the cracks and fissures of the rocks near low-water mark afford
a verv favourite habitat, and the tubes often project through
incrusting sponges and ascidians both simple and compound.
As a rule the tcnigh hornv tube has grains of sand protecting
the exposed portion, whilst the part immersed is hyaline and
more delicate. The tunnels in shell, limestone and chalk are

83

circular, and are easily distinguished from those of Dodcaiccrid
or Polvdoni. PoiauiiUa iriiiforiiiis not only frequents north
European shores, but extends to Iceland, Greenland, America,
ISIadeira, and the Canaries, so that its occurrence at the Cape is
less remar-kable.

BiSPiKA VOLUTACORNIS, Montagu, 1808.

1808. Aiiipliitritc roliitdconiis, Montagu, Lin. Trans. VII.,,

tab. vii., f. 10.
1865. Disivlia voliiicicoi iiis, De Quatrefages, Anneles, II., p.

421,. PI. 20, f. 5-7.
189-}.. Bispini 7'ohifiicoriiis, De St. Joseph, Ann. Sc. Xat. 8e.

Sen XVII., p. 286, PI. xi., f. 289-295.

A gigantic form (Plate viii., fig. 50) which in spirit measures
fullv 130 mm. with a breadth at its widest part of 17 mm.

Anteriorly a wide gap exists between the edges of the collar
which are close to the first bristle-bundle. The collar expands
in the middle line ventrally and is reflected, the left overlapping
the right flap. Dorsally an elevated fold on each side of the
middle hne is sometimes formed between the edges of the
collar — apparently bv contraction. De St. Joseph describes
the collar as of a deep violet with a border of white in the
French examples.

The anterior region consists of eight segments. On the ventral
surface of tlie third bristled segment on each side of the middle
line are two elongated slit-like marks, and they are distinct as
far back as the eleventh segment.

De St. Joseph found considerable variability in the thoracic
segments, the number ranging from six to eleven, and, more-
over, the sides are sometimes unequal, e.g., seven on the left and
eight on the right.

The dorsal tentacles (/'^///>c'.s De St. Joseph) are long, flattened
(with a thicker and a thinner edge) tapering processes, and
from the tuiter edge of each the web at the base of the branchiae
l^egins. The branchiae are rolled in a spiral of great complexity
on each side. The outer whorls have longer filaments, the
central have shorter, and in addition to the ordinary filaments
a series of almost pectinate folds occurs in some at the dorsal
edge of the whorl where the rim joins the tentacle. The central
whorls are very short. In all there are not less than 260
filaments with their pinna> which are rather short. The whole
system is beautifully variegated. Thus a purplish belt marks
the web, each filament having several touches of browaiish
purple alternating with pale regions, whilst the tips of many are

84

dark brownish purple. De St. Joseph mentions that some have
entirely white branchia'. The same author describes the basal
branchial cartilages as forming only two spiral twists, whereas in
spirit they appear to have three or four. Their vessels, he says,
contain green blood. He observed an example with one of
these organs in process of reproduction and represented by
twelve white filaments. Both sides are spirally rolled, but the
filaments are much shorter than in Spiroi>r(ipliis Sptilhnizdiii
from the Mediterranean.

Along the bases of the whorls an elevated collar or crest of
mucous membrane winds, and probably has important functions
in directing the streams of water and perhaps secreting mucus.

The arrangement of the spiral in Bispira as shown in this
example is such that in a horizontal section only portions of
three rows of branchite appear, for example near the base, the
outer being the most complete, the next within (a continuation
•of the former) being less, and the third very short. The whole
system really consists of a single lamina with its filaments
■spirally rolled on a firm axis which is thickest inferiorly and
tapers superiorlv. The upper central filaments are thus shortest,
the outer and inferior, the web of which adjoins the collar oi
the annelid, are the longest. The spiral attachment thus winds
round the axis from above downwards.

The outer whorl commences on each side at the tentacle, the
web of attachment on the left having about two short tapering
filaments as supports — before joining the first long branchial
filament. On the right the branchial stems are disconnected,
the membranous lamina attached to the base of the tentacle
abutting inferiorlv on a groove.

The body is massive, somewhat fusiform in outline — ni so far
as it is slightly tapered in front, and still more tapered to a blunt
point posteriorly. The bristled segments are about 172 or more.
It is rounded on the dorsal surface, flattened on the ventral, and
terminates in the anus. A groove commences on the ventral
surface at the anus, travels forwards in the middle line to the
second post-thoracic segment, bends to the right towards the
front of it, and passes obliquely through the segment in front
to the space between the anterior and posterior regions, and is
there lost, though in some old examples a shallow groove runs
forward dorsally a little above the bristle-tufts and is lost at the
collar. In front dorsally are the tentacles, and close to their
bases ventrallv is a frilled ridge on each side which amalgamates
with the inner edge of the massive lips. In all probabilitv the
muddy water is conveyed along these to the mouth from the
branchial spiral, so that both digestion and respiratitMi are
subserved. The lips form two prominent fleshv lobes project-
ing forward — d.u-ker in hue distally where slightly enlarged-
paler below. Their inner edges are flattened, mdeed, somewhat

85

hollowed below the vertical slit forming the mouth, and the
ventral edge of which trends to the fissure between the lobes of
the collar.

The species seems to be brightly coloured probably of a rich
brownish purple, which in some uniformly tints the branchiae.
Each foot in the anterior region has a dark pigment-speck just
in front of the papilla at its anterior and dorsal edge, and
though the papilla is less distinct in the posterior region a
similar speck of pigment continues to the tip of the body.
Moreover, at the dorsal end of each hook-row in this region a
dark speck is present.

The thoracic bristles are in eight bundles and differ from
those in SpirograpJils Spallanzani by the greater length of the
wmged tips of the dorsal and the narrower wings of the ventral
series (Plate VIII., ligs. 43 and 44, the former form the upper
series, the latter form the middle). In S". Spnlldiizniii, the short
broad spear-tips of the ventral are diagnostic, one being
sketched by way of comparison in Plate VIII., fig. 45. The
foregoing distinctions are still more pronounced in the pos-
terior bristles, which have much more elongated tapering
tips and narrower wings than in SpirograpJiis SpaUniizani — the
respective forms being shown in Plate VIII., figs. 46 and 47
for the former and 48 for the latter. Moreover, the somewhat
abrupt narrowing of the tip in Spirograpliis Spallanzani is
diagnostic.

The anterior hooks have a similar arrangement to those in
Sahclla, and in minute structure (Plate VIII, fig 49) are closely
allied to those of the Italian form, with which it has just been
contrasted, except that the crown above the great fang is more
distinctly serrated. The same features are shown in the
posterior hooks.

The tube is composed of tough secretion having at the wider
or lower end a few fragments of shells and coarse sand, the
greater part, however, being bare. Above the middle this
horny coat has a thick investment of muddy sand with here and
there a shell-fragment. This leads to the upper region which
is composed for the most part of muddy sand with only a thin
lining of secretion. On the French shores De St. Joseph* has
found tufts of Aniathia Icndigcia growing on the posterior end, a
feature observed in the tubes of other forms, such as CJiaioptcrns
and Thclcpns, which are frequently feathered with graceful
zoophytes.

This beautiful Sabellid, originally found on the beach at
South Devon by the indefatigable Montagu, appears according
to De St. Joseph, to live in colonies on the French shores, and
probably also in South Africa, the young attaching their tubes

*Bollet, d. Soc. Adriat. d. Sc. Nat. viii., p. 296 [pik De St. Jiiscph) also List
Zcit. f. w. Zool. xlix'., p. 248-286, Taf. iv. ancl v.

86

to those of the adults (De St. josepli). Even the crustuceiiii
parasites on the body in the regions thus widely separated have
a parallelism as shown in the account of the South African
form which follows. Both males and females occur on the
branchifC of large specimens as well as on the body,

De St. Joseph found the sexual elements developed at the
•end of September, the eggs being ciomm. in diameter, and the
sperms very minute. He states that these elements escape by
the pores of the segmental organs at the ventral base of the
setigerous lobe. A young specimen of seven mm, had nine
thoracic and nineteen abdominal. Its branchiae (few in
number) formed spirals.*

SaBELLIPHILUS (?) BISPIR^.

Crustacean parasites of annelids have been known to
zoologists since Kroyerj in 1837 described Scliiis Jrilobiis as
frequenting Lcpidoiiotiis sqiuniiatiis, L. The same author next
year found another on Hctnuotlio'c iiiibriaifa, L. which lie pro-
visionally termed Silciiimii Polyiioi-s, but delay in publishing a
description caused Steenstrup and Liitken, with Kroyer's con-
sent, to substitute the title Hcrpyllobius arciicus for this crus-
tacean. Sars| next contributed a description of four new
species, viz., Te rebel I i coin, irplaiis on Terebclla deb His, Malmgren,
Sabellacheres giacilis, on Myxicola Steeiislrnpi, Kr., Sabelliphiliis
eloiii^dliisonSdbella pei7'oiiiiiei,'dnc\ ChoHiep/iiliiscl/spar on Euchoiie
papulosa, Sars. Keferstein in i863§ gave an account of Nereieola
ovata occurring on Nereis ciillrifeia, Grube, and the author and
Grube likewise noticed the same form which is not uncommon
in the Channel Islands. In 18(34 Nordman || found a new
copepod on Xicoiiiaclie liimbricalis, and he termed it Doiiiitsa
iiyiiieiiicola. A few years later Hesse H in his account of new
crustaceans from the shores of Fr3.nce described Clielodinifoi mi
iypiciis from the lob-worm (Aieiiicola marina, L. ). Claparede *
the following year gave a resume of the literature of the subject
lip to date, as a preface to his description of a new species,
SabellipJiilus Sarsi on Spirograpliis Spallanzaiii, Viviani. Three
additional species were described by the veteran naturalist,tt Sars,
the same year, viz., MeliiiiiacJieres ergasilo'ides on Mel in 11 a crisfaia,
Sars; Hepyllobins crassirostris on Evartie inipar, Johnston; and

*Op. cit., p. 291.

t Xaturhist. Tidsskrift, iste K. I. Kjobenhavn.

X Forhandl. Vidcnsk. Selsk. I. Christiania, 1861, p. 46 (sep. copy).

JjZeit. f. w. Zool. XII. Bd.
• II Bullet. Acad. St. Petersb. 1864.
4 Ann. Sc. Nat. 5th Ser. XI. p. 275. }869.

** Ann. Sc. Nat. 50 Ser. .xiii. Note p. i, PI. 7.

It Xyt. Mag. Nat. Bd. xvii. 1870.

87

Euiysileninni tiiiiicatiim on Haniiothoe iuibricnfa, L. In 1877
Kurz* published an account of Eimicicola Chnisii as a parasite
on Eunice Claparcdii, De Quatrefages. The same year an
important contribution by Levinsenf cleared up the ambiguity
connected with HcrpyUobiiis aniiciis, Steenstrup and Li'itken,
and, besides, added four new genera and species to the list, viz.,
Sclloiilt's Bolbroci on HaiiiioiJio'c iiiibiicata, and a curious variety
on Gattviiiia ciirosa, Pallas, Rliodiiiicola cloii^aici on Rhodiiic
Lovcni, Malmgren, Bmdophihi pyonKva on Bnidii rillosd, H.
Rathke, Saccopsis TcrcbcUidis on TercbcUidcs Sfniiiii, Sars, and
Crypsidomiis Tcirbclhv on Aiiipliitrifc cirrata, O. F Miiller. In
the Annelids of the "Challenger " is recorded a new form attached
to the foot of Lcaiiim airohita, McIntosh,| from 345 fathoms,
south of Yedo, Japan, viz., Leaiiiricola roiiindata. List§ in
1890 gave a minute account of two forms, one of which —
Gdstrodclplixs Chnisii — had apparently been first observed on
the branchina of Bispini volutacornis, Montagu, by De St.
Joseph, II the other, (uistiodclpliys Myxicohv, from the branchiae
of Mxxicola iiifiiiidibiihiiii, Grube. The outline of (iasfivdclpJiys
Chnisii differs considerably from the form subsequently
described from South Africa, the latter having the appearance
of an Ergasihis, and closely approaching the Sabcllipliiliis Scnsii
of Claparede, the crustacean parasite of Spirognipliis Spalhnizani
in tlie Mediterranean. Various examples adhered to the largest,
an.d probably oldest, specimen of the annelid both dorsally and
ventrally, the anterior end being fixed in the fissure between
two segments, and they also frequented the branchiae, but they
were not confined to the dorsal edge of the branchior, most,
indeed, occurring on the pinnoe or close to their origin.

The crustacean, which may provisionally be termed Sahclli-
philiis bispinv, parasitic on this annelid dift'ers from the (rcistiv-
dclplivs Chnisii as described by List,1[ and which he found on the
branchiae of Bispini volufdconns of Montagu from the Adriatic.
De St. Joseph had apparently obtained the same form on the
French coast some years previously — fixed by the rostrum and
ventral plate to the dorsal edge of the branchiii?, its long axis
being parallel to that of the rachis. He had provisionally
named it BispiropJiilus tciiax, but, as he himself observes, the
name of List has precedence, and he appears to have satisfied
himself that List was dealing with the same species.

In outline the South African ectoparasite leans more to tl^e
typical Ergasilidae than GastrodclpJiys, which in the female is
narrow in front and broad behind, whereas the present form is

* Sit/Aing-ber. Akad. Wiss. Wien, 1877.

t Videiiskab. Meddcl, Nat. Forcii. Kjobenhavn.

+ Ann. " Challenger," p. 133- Woodcut, Fig. 2.

§ Zeit. f. w. Zool. xlix. Bd. p. 71 Taf. iv.-viii.

jl Ann. Sc. Nat. 8e Ser. xvii. p. 292, 1894.

\ Viik Dr. Thos. Scott, 20th Kept. S. F. B., Part III., p. 288.

88

broad anteriorly and narrow posteriorly. In the male, more-
over, the resemblance between the species is slight, the elongated
body gently tapered from front to rear having little resemblance
to the male of that from South Africa, It may be that two
parasitic crustaceans occur in Europe, and that the present
.species agrees with one of them, but no evidence on the subject
has been found. These crustacean parasites of Annelids-
approach in a close manner those of various fishes, e.g., such as
Boiiiolocliiis,-dve of the Ergasilidae, parasitic on the sole and the
ling.

The female (Plate IX fig. 56) is about 1.2 mm. in length and
in general outline resembles an Eri^asilns, often seen on Doris
tiihcfciil/itd, hcixmg an ovate body (cephalothorax, Claparede)
to which is appended the tapering tail. The anterior end is
shaped like a hoof and the first segment is distinguished from
the rest by a lateral peak on each side. The second segment is
somewhat less in transverse diameter than the body at the peaks ;.
in antero-posterior diameter it is wider than the two following,
and the last segment is still less. A lozenge-shaped segment
follows with a few hairs at each lateral peak ; and this is
succeeded by a larger ovigerous segment from which the bulky
ovisacs extend. Four progressively diminishing caudal seg-
ments occur posteriorly, and to the last are attached the two-
caudal processes with long spines. The first of these segments
in the male especially appears to be more or less fused with the
genital segment. A lozenge-shaped greenish or brownish opacity
occurs in the centre of the carapace in some and is probably
due to the food.

The antennules (Plate IX fig. 53) are of average length and
have seven segments, the second being the longest and the last
minute. They have similar hairs to Claparede'sS<^//)r////>//////5Sc/r,s/7'*
from SpirogiripJiis. The antenniv are shorter than in the latter
species, and have only two terminal hooks, and these do not
seem to be so closely parallel as in Claparede's form. The buccal
appendages appear to be similar to those of the latter. Dr.
Thomas Scott, who, along with Prof. G. S. Brady has done so
much good work in the group, kindly examined this form, "which
has a general resemblance to BoiiioIoLiiiis, though it belongs to
the closely allied family Lichomolgidti). The appendages of
the mouth in the latter are very well described and figured in
the 3rd volume of Prof. G. S. Brady's Monograph, and if the
drawings be compared with the corresponding appendages of
Boiiiolocliiis given in my paper in last year's Report of the
Fisliery Board for Scotland, a marked difference between them
will be noticed, especially in the structure and armature of the
mandibles and maxilhi'. On the other hand there is among the
genera composing the Lichomolgida? a certain similarity in

* Op. cit.

89

the structure of the mouth-organs. In Part III. of the 12th
Report of the Fishery Board is a note setting forth the more
prominent of these differences — SabcUiphiliis inchided."

The first four pairs of thoracic feet are biramous, the stronger
showing externally three joints, the distal segment being some-
what ovoid, and furnished with strong spines. The next segment
also has a long spine — the other has an elongated terminal
segment. The fifth pair appears to be rudimentary — probably
represented by the lateral processes of the narrow segment in
front of the genital segment (the first abdominal of some e.g.
Claparede).* This segment has the enlarged vulva^ to which
in the sketch the ovisacs adhere (Plate IX., fig. 56). These are
distended with moderately large eggs. In some the ova were
still in the ovaries, though ready for deposition, and in one
filmy tissue adhered to the vulvae — either for the purpose of
receiving the ova on discharge, or which had been left after the
escape of the eggs from the ovisacs. Moreover, in several a
spermatophore was attached to the pigmented tissue of the vulva
on one side, but was detached during examination (Plate IX.,

The male is less than the female, measuring "7239 mm., and is
proportionately more elongated (Plate IX., figs. 51 and 52), all the
segments of the body being narrower. The antennules (Plate
IX., fig. 53) agree in regard to the proportions of their segments
with those in the female and so with the antennie (Fig. ibidem).
The other appendages of the region also correspond. In lateral
view (Plate IX., fig. 52 ) the four anterior thoracic feet are more
distinctly observed, and the shape of the rostrum is more clearly
defined. The genital and caudal segments, however, do not
readily take this posture, so that they are generally seen in a
horizontal position — not on edge — probably because the genital
segment, which is flattened from side to side, maintains this
position. The condition of the genital segment at once dis-
tinguishes this sex, for it is broadly ovate, convex in front, but
somewhat concave posteriorly. It contains an ovoid spermato-
phore on each side, and the apertures appear to be posterior, each
debouching from a papilla. On extension the spermatophore is
spindle-shaped (Plate IX., fig. 54). It however assumes a different
aspect after the sperms have been utilized (Plate IX., fig. 55) for
then the elasticity of its capsule gives it an elongated slipper-
shape. The chitinous investment of the genital segment seems
to lend itself more readily to its varying contents than that of
the other parts — a feature probably due to its delicacy. As in
the female, four caudal segments succeed the genital — each
being slightly narrower than the one in front, and the last being
somewhat longer. It bears the two elongated processes from
which the spines extend.

* Ans. Sc. Xat. 5th Se. XIII p. 14.

90

The spermatophorcs would seem to he discharged hy the
male at the vulwc of the female, one or more becoming
attached thereto and fertilizing the ova as they pass, or other-
wise. Whether the male secretes new spermatophores and
repeats the process of fertilization could not he determined.

It is interesting that the South African form so closely agrees
with the Sdbc'lh'pln'his .Sc//.s// of Claparede from the Mediterranean
Spirognipliis SpuUcuizani. The proportions of the antenules
and of the antennae differ, as \\ell as the number of the ter-
minal hooks in the latter. Moreover, in his figure he shows
only three caudal segments behind the genital (the first having
apparently been joined to the genital).

The following note by Mr. Andrew Scott on the Family, and
the remarks of Dr. Thomas Scott, after consideration of all the
features of the case, are of special interest in connection with
this ectoparasite.

The Family ScippliiriiiidiV, which nicludes the genera Pscii-
daitthc'ssius, Licliojiiolgiis, HcdnianiicUa, and SabellipJii/.is, and
the Family Ergasilidcv, containing such genera as Ergasilus,
Bomoloclins, etc., are founded chiefly on the structure of the
cephalic appendages, especially the mandible.

In the Sappliiriiiidiv the mandible is in the form of a thin
plate, with a more or less distinctly serrate edge. The mandible
in the Ergasilidiv is jointed, and terminates in a strong tooth,
with a serrate edge.

The genera Psciiddiif/u'ssiiis, Lichoiuolgiis, and Hedmaiinclla
are distinguished by the structure of the inner branch of the
fourth pair of swimming feet. In Pscndaiithcssius the inner
branch is one-jointed ; in LicJiomolgiis, two-jointed ; and in
Hcdiiiaiinclla, three-jointed. The other appendages are very
much alike in the three genera.

Sabellipliilns is ^ep'ciY^xted on account of the structure of the
second maxilliped of the female. In Pscudanthessins, Licho-
fnolgus, and Hcduuinuclla the female second maxilliped ter-
minates in a short, almost obsolete claw. In SabcUiphilits the
maxilliped is well developed, and terminates in a long, strong
claw.

Dr. Scott observes that this crustacean certainly belongs to
the Sapphirinidce (= the LicJioiiiolgida- of Brady and others).
It does not agree with SabcUiphiliis in the structure of the
antennule, and to some extent in the structure of the first and
second maxillipedes, and especially in the structure of the
endopodites of the fourth pair of the thoracic legs. In Sabclli-
pliilus the first two joints of the antennules are considerably
dilated, and the endopodite of the fourth pair of legs are three-
not two-jointed. On the other hand the specimens from South
Africa — in the structure of the antennules, and especially in the
structure of the endopodite of the fourth pair of legs- -agree

91

very closely with LicJiOfiioloi>iis, but differ very much from that
genus in the structure of the second maxillipedes. In Licho-
nioloi^iis this pair of maxillipedes in the female are stout, and
terminate in an almost obsolete tooth-like claw. Indeed, he
(Dr. Scott) does not remember any described genus to which
these specimens can be satisfactorily ascribed. If the specimen
dissected had been a male mstead of a female it would have
agreed fairly well with Lichoniologtis, as the male in that genus
has the second maxillipedes usually furnished with a moderately
long claw, but as it is a female it exhibits in its appendages a
relationship to both Lichoiiiologiis and Stihc'llipliiliis.

Fam. SERPULIDJE.

PrOTI'LA Capensis, Mcintosh, an var. tiihiihirid, Montagu.

1803. Protnla fuhuhina, Montagu, Test, Brit. p. 513, and

Suppt. p. 171.
1885. Protnla capensis, Mcintosh. Ann. " Challenger," p.
\so9. PI. LI v., hg. 2 ; PI. XXXl.A, figs. 12-13.

The species is large, attaining (in spirit) a length of 100 mm.,
and a breadth between the bases of the anterior bristles of. 14
mm. Of the total length, about 18 mm. pertain to the
branchiae, 27 mm. to the anterior (thoracic) region, and 55 mm.
to the posterior region.

The dorsal surface is grooved along the middle line
anteriorly, but is convex posteriorly. The ventral surface, on
the other hand, is deeply grooved throughout, the groove, how-
ever, being interrupted by the deep membranous apron at the
posterior part of the thoracic region.

At the truncated anterior end the mouth opens dorsally as a
deep pit with a smooth papilla projecting out of it. Similar
grooves to those in Bispiirh lead from the spirals of the
branchiae to the mouth, and probably subserve the same pur-
pose, viz., the conveyance of food and mucus.

The branchiae form two dense masses on each side, attached
by a surface shaped somewhat like a trefoil to the bodv. More-
over, the left mass is composed of three spirals which are
mounted on a basal region composed of a single spiral. Such
may be an abnormality.

The right side has but a single complex spire considerably
longer than the compound one on the left, the last whorl
terminating dorsally and sending from its edge a lamina
towards the mouth, apparently for directing currents thereto.

The filaments have a dense series of somewhat short pinnae
which proceed to the tip, the latter only being distinguished by

D 2

92

its larger bulbous end. As transparent objects the filaments
(radioles) present an interrupted dark line in the centre, pro-
bably from a blood vessel, and the pinna? are richly ciliated.

The question as to whether the complex spirals in this large
form (double the size of that procured by the "Challenger")
are the product of age is interesting, and the fact that each side
differs from the other in the character of the spirals shows that
considerable variation is possible.

The collar commences by a wide fan-like flap at each side
dorsal Iv, and extends round the ventral border to the other
side The first or lateral flap is partially separated by a fissure
from the rest of the collar, which otherwise is continuous. The
lateral flap of the collar is continuous again dorsally with the
loosely folded membrane which flanks the anterior or thoracic
region, and which passes between each of the bristle-bundles
to be attached as a folded or frilled band to the ventral surface,
embracing on each side the soft pad for the rows of hooks.
Dorsally, the whole forms a continuous membrane, for it is
attached on the dorsal aspect of each bristle-bundle, and it is
highly vascular, every part of it being permeated by the finely
branched vessels with their dark blood. Posteriorly the ventral
edges of the membrane are joined by a thicker fold (already
alluded to), and which probably contains glandular elements
like the area in front.

The anterior regions has seven pairs of golden bristle-bundles,
each consisting of a dense group of fine bristles with delicately
tapered tips and narrow wings. The great number of these
slender bristles in a tuft is a feature of moment. No spine is
present. The bristle-bundles are nearly of the same size from
the first to the last. They appear to correspond in minute
structure with those of Protitla aipciisis.

In the same way the anterior hooks appear to dift'er onlv in
size, for the curves are the same. These rem;irks also applv to
the posterior hooks.

The species occupies a large white, calcareous tube (which
was absent in the "Challenger" collections), sinuous in
outline, and having an average diameter of 13-14 mm. It is
smooth internallv, but roughened externally bv various growths
such as tubes of Filigraiin, patches of Cellcpora (Lepnilia ),
Bala III, coarse sand and Mclobesla.

The many close structural resemblances between this and the
European species {Proliila hibulaiia, Montagu) raises some
doubt as to the specific distinction on grounds so slender as
the slight modification of hooks and other parts. Its much
larger size m South African waters will account for some of
these variations. On the whole it mav be considered as a
variety of the European species.

[PUBl.lSHKD 25TH Jl NK, ig04.]

-o-

EXPLANATION OF PLATES.

Plate V.
Fig.

27. Twenty-fifth foot ot Tluodisca [Aiiiliostoiiiii] lic.wipliylhi. Sclimai'da. X 2\

diam.

28. Posterior end lil' a youi-i.Lj example of the fore.t;oin.s;' (?) with four anal cirn.

X yo diam.

2y. Foot of tiie foret;'oing (adultlfrom the middle ret^ion of the body. X 21 cham.

51. Tip of .threat bristle fnim the 4th se.iiment of CJuilof>tcntsviiriopt(i,iliis. x 100

diam.

52. Smaller bristle nt the same kind from an example from Naples. X loodiam.

53. Terminal win.^ed bristle at the tip of the sickle-shaped .anterior feet. X

Zeiss oe. 2. obj. D.

54. Spathulate bristle fi^om the middle of the same foot. Similarly ma^nihed.

55. Lar.ye spathulate bristle from the inner end of the series. Similarly matjniticd.

56. .Anterior hook of the same species. X Zeiss oc. 2. obj. I). + 3 in. dr.iw-tube.

Marine Invest igaitioTis
Soutli Africa.

Annelicis.Pl.V.

W.C.l/r del.

MirvterTi. Bros .lith.

ARICIID/E &c CH^ETOPTERID^

Plate VI.

30. Anterior hooks and bristles of Cimtlnliis cinutns. var. caf>crisis, x 100

diameters.

31. Anterior hooks and bristles (drawn to the same scale) of Cirratulus cirratus

from Britain. An ovum is on the left. X 100 diam.

32. Anal funnel of Nicoiiuiclic lu)7thytialrs var. capfnsis. X 20 diam.

33. P'oliate anal expansion of Maldane Mc/iitoshii, Marenzeller. (Under a lens.)

34. A posterior hook of the foregoing. X 150 diam.

Ma.ririe Investiga^tions.
SoLitli Africa.

Annelids. PI .VI

■W.C.M.deL

CIRRATULlDyE &i1vTALDAT1iD.4l

Mi lit em Br OS .litli..

E

Plate VII.

Fio,

^^a. rip of bristle of Pcctiiuniii ctifcii^is, Pallas. X 150 diam.

35b. Tip of an injured bristle fiom the same region. X 150 diam.

36. Hook of tlic same species in lateral view. X Zeiss oc. 2, obj. D. + 3 in draw-

tube.

37. Hoolc of TIiclcpiis—? X As in tlic fore.ijoing.

3S. I'piper bristle of Polniiiillii rciiif'oniiis, O. F. M. X As before.

3g. Tip of upper spalliulate bristle of the same species, x As in tlie forecjojn.tj.

40. Tip of a lower spathulate bristle of the same form. X As before.

41. Hook of the same form. X As in the previous example.

42. Tip of a bristle from the posterior region of the same species. Similarly

niagnilied,

Ma.rme IrLvestigSLtions.
South Africa.

Annelids. PI .VU.,

'3 5 a.

.m

A';

35b.

41

4.2.

Oto.

ie". . Mii^tsrn Bros.litli.

AMPHICTENID^ , TEREBELLID^ &c Sy\BELLID^E .

Plate VIII.

Fig.

41. Upper winged bristle from the re.iiion of Bispira voliitdconiis. xZeissoc.2.
Obj. D. '

44. Winged bristle from the middle of tiie tuft (tlioracic region). Similarly

magnified.

45. Winged bristle from tlie same region in Spiivi^rnpliis Spallatizniii. Similarly

magnified.

46. Upper liristle with narrow wings from the posterior region of nixpini voliitii-

coniis. X As before.

47. Inferior bristle of the same tuft. Similarly magnified.

4S. Winged bristle fron the same region in Spiivi^inpiiis SpciHiiii-aiii. x As
before.

40. Hook of Bispira I'oliifnconiis. x Zeiss oc. 2. Obj. D. + 3 in draw-tube,

50. Lateral view of Bi-ipim "i'ohitiicoiiiis. About natural si/e,

Ma-rme mveSoigSLtioTiS.
SoutK Africa..

Annelids. Pl.Vni.

W.C.M.del.

Miiater-n. Bros .liih .

SABELLID^.

Plate IX.

Fig.

51. Mulv ui Siilnilif'liiliis ( ?) bispiKf, ivom [he (.hni,um. x 100 diam.

5:;. Another example partly in piolile. The t^enital seyment and the eaudal
region, however, are seen from the dorsnin. X As l3ef(jre.

^^. Anterior region of the carapace of a male with the antennules and anlenn:f,
the two powerful hooks of the latter lieini; conspicuous. X Zeiss oc. 2.
Olij. D.

54. Spermatophore when lust extruded. X loo diam.

55. Spermatophore as detached from the region of the vulva with its elastic cap-

sule more or less contracted. X 100 diam.

56. An ovigerous female SabcUipliihts { .^ ) bispii a-. Magnified to the same scale

as the males.

16 I:ivesti?5a.tions.

Annelids. PI. iX.

^^

/

>^

X:<- ■

^•

t^'

^^^v.

:1

i

^'

-^

.:^

nn

i

^N

v

/•

V'.';.i,i

W'iin.ter-n. t

THE
TURBINOLID CORALS OF SOUTH AFRICA

WITH NOTES ON THEIR ANATOMY AND

VARIATION,

BY

J. STANLEY GARDINER, M.A.,

Fellow of Gonville and Caius College, and Demonstrator of Animal Morphology

in the University of Cambridge.

CONTENTS.

I. Introduction.
II. Systematic List.

III. The Anatomy of the Turbinolidae.

1. General Observations.

2. Trochocyathus rawsonii.

3. Heterocyalhus aequicostatus.

4. Caryofhyllia epithecata, with a note on C. Smilhi.

5. Duncani.1 capensis.

B

95
I. INTRODUCTION.

The present paper, containing an account of the Turbinolid
corals of the Cape, has entailed the careful examination of up-
wards of 2,000 specimens from the present collection, and in the
British Museum I investigated also over 1,000 specimens of the
genus Flabellum, the account of which it did not seem advis-
able to defer until the whole family should have been finishea
(see " Marine Investigations in South Africa" Vol. II., pp. 1 17-
154, 1902). In addition, it was necessary to carefully study the
genera not represented in the collection, a work which clearly
showed Duncan's classification* to rest on no scientific basis.
However, I have provisionally adopted that arrangement of
genera, preferring to leave the revision until I shall have worked
through all the different families.

I would like here to express my indebtedness to the Govern-
ment Biologist of Cape Colony for entrusting me with the ex-
amination of the collection. All the specimens of the different
corals that came up in the dredge appear to have been taken and
carefully preserved. Their species may seem to be few, but
the large numbers of several have allowed a p;oper
examination to be made of their variabihty. The latter
is shown to be very great in solitary corals. It may be divided
into three classes:- — (i) that due to environment, (2) continuous,
and (3) discontinuous, or in other words, (i) vegetative, (2) nor-
mal, and (3) specific. The presence of all three is quite clear
m Heterocyathus aequicostatus, as it was also in Flabellum
riibrum,% (i) the general characteristics and sizes from different
depths and locahties ; (2) to some extent the variation within the
types, but in particular the meristic variations ; and (3) the exist-
ence of definite types. Indeed, the collection has enabled us
to get an insight into the variability of corals, such as must be
carefully considered in any future systematic work on the
Madreporaria.

The anatomy of the polyps has also been studied, and as
far as it goes, confirms the diagnosis of the species from their

* Jour. Linn. Soc, XVIII., pp. 1-204 (1^84).

§ Vide also " Variatimi in V. rubrum," Proc. PJiil. Soc, Ciinih., Vo].
XL, pp. 463-71-

96

coralla. After the full account given of F. riibrum, only a few
notes are required on those subsequently examined, although
several have been investigated almost as carefully as that species.
The uniformity in the minute anatomy of the different forms is
surprismg, the only new point developed being in connection
with the endodermal nematocysts of the tops and sides of the
septa, which have been shown to be of wide distribution in the
different species (pp. 122-127).

II. SYSTEMATIC LIST.

Genus DESMOPHYLLUM.
Desmophyllum, Ed. and H., Cor. II., p. 76 (1857).

I. DESMOPHYLLUM CAPENSE, n. sp.

There are five specimens, with eight separate corallites of a
gigantic species, which is very closely related to D. ingens.
Moseley (Challenger Reports, Vol. II., p. 161, 1881). It differs
from that species in that (a) the tops of its corallites are always
straight, never expanded, nor trumpet-shaped ; {b) in the largest
specimens there are six cyles of septa, instead of five, and {c)
there is no columella, though the septal edges may be thickened
and fused together for i cm. or so above the base of the axial
fossa.

Beyond the above, the species requires no description, in its
other characters generally agreeing with D. ingens, from which,
nevertheless, it appears to be a distinct species. Three of the
larger corallites are respectively 45x25 mm., 40x26 mm., and
41 X ig mm. across their upper ends, and have 162, 144, and 152
septa, of which 24, themselves divided into two series, are con-
siderably broader, thicker, and more exsert (5 mm.) than the
rest. The corallites vary up to 12 cm. in length, and are through-
out rather flattened. The edge-zone of the polyps extends
down the outer sides of the corallites for a varying distance, in
the specimens up to 5 cm. ; and this part is ridged with rather
irregular bluntly lobate. or ridge -like costae, decreasing in size
from the calicular openings. Below the edge-zone there is no
trace of any epitheca, the surface of the corallites being simply
covered by incrusting organisms or bare. Two, three, or more

97

corallites frequently grow together, but, as in D. ingens, there is
no trace of any connection, either former or present, between
their polyps. It is in certain cases possible, however, that the
lateral polyps may have arisen by budding at the edge-zone of
the central polyp, but there is in some cases a little ridge, where
presumably the edge-zones of the two neighbouring polyps met
one another, while in other cases there may have been a fusion
of the edge-zones, as commonly occurs in other corals.

The deposition of exotheca, shown by Moseley in D. ingens
(erroneously termed epitheca, see PI. IV., fig. 4a), has been car-
ried to a much greater extent in the present form, at the base of
one corallite where the edge-zone spreads out, a thickness of
over I cm. being visible. It is to some extent due to this deposi
tion that the corallites have not got that trumpet appearance
found in D. ingens. At the base of one of the type specimens
ol D. ingens in the British Museum endothecal dissepiments art^
distinctly recognisable. This, too, is the case in D. capense,
where they are ratlier irregular (2 — 5 mm. apart), straight or
more or less curving, more closely set near the theca. They
may be almost absent, and in no case are visible from the sur-
face ; they never cross the axial fossa, and only deep down (2 cr
3 cm.) extend out for more than a milHmetre or two from the
theca.

The septa have entire edges, a little thickened at the base of
the axial fossa, and are smooth-sided, with little or no trace of
ridges.

Eight corallites, forming five specimens, from Cape Hangklip,
S.E. \ E., 71^ miles. Depth, 44 fms. Bottom, fine sand and mud.
(Refer. No., 15549.)

Genus FLABELLUM (Lesson).

F labellmn, Gardiner, Marine Investigations in South Africa,
Vol. II., p. 117 (1902).

I have already reported on the specimens of this genus, but
two corallites which I at first referred to Desinophylluni, evi-
dently belong to it as well. The essential difference between
the two genera lies in the fact that the wall of the calicle is
formed by epitheca in Flahelhini, but by theca in Desmo-
■phyllum. Delicate wavy transverse growth-lines are character-
istic of Flabellum, and belong to the epithecal corallum. A
true columella is present in neither genus, but either may have
the axial fossa filled in by the fusion of the septal edges or by
trabeculae from the same.

98

2. FLABELLUM SIBOGAE.

Desnwphylluin alabastriini, Alcock, Siboga Deep Sea Madre-
poraria, p. 28, pi. IV., figs. 5, 27, 27a (1902).

There are two specimens of this very dehcate species, both
freed from their bases of attachment and somewhat chipped
They correspond in every particular with the form described by
Alcock from the East Indies as Desmofhylluni alabastrum,
which I have had to rename as the specific name, " alabaslrum,"
was preoccupied by Moseley in the " Challenger Report." The
specimens measure 24 and 22 '5 mm. in height, and 9x7 and 8' 5
X 8' 5 mm. in diameter, across the mouths of the calicles.

The form is certainly quite distinct from the other Cape of
Good Hope species, and so far as I can at present see, from all
other described species of the genus. In Duncan's system of
classification it would be referred to the genus Rhizotrochus.

Locality. — Buffalo River, north of, 15 mi. Depth, 310 fms.
Nature of bottom, coral and mud. Date, April 24th, 1901.
(Two specimens.)

3. FLABELLUM PAVONINUM (LESSON).

Flabelliini pavoninum, Gardiner, Marine Investigations in
South Africa, Vol. II., p. 123, pi. IV., figs. 18-21 (1902).

4. FLABELLUM RUBRUM (Q. et G.)

F labellinn rubruni, Gardiner, Marine Investigations in South
Africa, Vol. II., p. 125, pis. I. -IV., figs. 1-17, 22-31 (1902).

tl-enus SPHENOTROCHUS Ed. and H.

Sfhenotrochus, Ed. and H., Cor. II., p. 65 (1857).

The genus is one of the best defined and most isolated among
the TurbinoKdae. The polyp completely covers the whole of
the corallum, so that there can be no epitheca present.

5. SPHENOTROCHUS GILCHRISTI, n. sp. (PI. I.,

fig- I. a— g).

Corallum free, greatly compressed, rounded at the base with
no mark of any former attachment.

The septa are 24 in number, and form two series. Cycles I.
a^id II. are equal in size, '75 — r5 mm. exsert in accordance with
the size of the whole corallum. The septa of cycle III. form
the second series, and are only about half as exsert as those of

99

I. and II. The columella is a plate, one-third to one-half the
length of the calicle, its upper edge intermediate in height be-
tween the upper edges of the septa of the two series ; it is
joined by processes from the 12 larger septa, about i'^ mm.
below its edge.

The theca has the appearance of having been formed by
thickenings of the septal edges. It has no definite
costae, but the whole of the exterior is covered by
a series of minute, linearly extended granules, forming lines,
which commonly converge a little round the base of the coral-
lum. They start somewhat irregularly as elongated paired gran-
ules, each pair forming a V, on the exsert portions of the septa,
but below, on the widest portion of the corallum, 8 or 9 Hnes
correspond to two neighbouring septa.

Height of the specimens by the greatest and least diameters
of their calicles in mms., 13 x 1 1 x 6, 1 1-5 x 9-5 x 5-5, 9-5 x 8"5
X 5, 11x8x3-5, 10 X 8" 5 X 5, io'5 X 8 X4-5..

Locality. — Near Kowie, Lat. 33° 45' 20" S., Long. 26° 44' 20"
E. Depth, 44 — 43 fms. Date, November 19th, 1898. (Seven
specimens.)

The species most closely approaches to 5. granulosum, Ed.
and H., from the Eocene. It differs from the four recent species
described by various authors, in the fact that it has no costae
and no wings on either side. Two specimens have the columella
slightly notched in the centre, and one has two pairs of thick-
enings on the sides of the same, opposite the tertiary septa on
either side of the large central pair.

Genus TROCHOCYATHUS, Edw. and H., Duncan.

The genus appears to be fairly well defined, but I do not see
any reason in the characters that Duncan gives for the retention
of Thecocyathus as a subgenus, since they are due to purely
environmental causes. The hollow root-like growths from the
" interseptal chambers," described by Pourtales in Th. cylin-
draceiis and laevigatus {III. Cat. Mus. Comp. Zool., No. IV.,
1871, pp. 13, 14), may provide a character of value, but this de-
scription must be re-read in view of the now known ectodermic
origin of the corallum. Similar roots are also found in T.
cmcticulatus below, a close ally. If they have really any con-
nection with the interseptal chambers, they can only be formed
ill the youngest stage of growth of the polyp, unless it be as-
sumed that the polyp redissolves its theca in places so as to push
them out. I have, however, been quite unable to see any such
connection, and if there be none, the innermost wall alone can
be regarded as a thecal formation, and all the external walls
must be epithecal.

lOO

5. TROCHOCYATHUS RAWSONII (Pourtales) (PI. i, fig.
2, a, b, and PI. II., figs. A-K).

■ . rawsonii, Pourtales, Cat. Mus. Comp. Zool, Harvard,
Vol. IV., p. 35, pi. VI., figs. 7-IO (1871), and Bull. Mus.
Comp. ZooL, Harvard, Vol. V., p. 199 (1878).

I refer no less than 38 dried specimens to this species, though
their general type differs from the description of the species in
the fact that the shape is conical, with base varying in breadth
up to one-third the diameter of the calicle. This, however, is
probably purely a matter of environment, and Nos. 31 and 32
of the table are in this character typical members of T. raw-
sonii.

The species characteristically seems to have four cycles of
septa complete, with a fifth incomplete, decreasing in breadth
and height from the lower to the higher cycles, and varying
up to 1.5 mm. in exsertness, edges smooth, sides covered with
low, granular ridges radiating from the theca. The costae
equal the septa in number, and are subequal in size, very low
but distinct and granular. The intercostal valleys are generally
filled in below with epitheca, almost to the level of the top of
the costae, this deposition extending to within 2 or 3 mm. of the
thecal edge. Pali are present in front of septa of cycles I. to
III, before cycle I. round and before III. rough, flattened pillars
I mm. or so broad ; those before III. are more distant from the
centre of the cahcle, and with the pali of cycle II., typically form
chevrons before the secondary septa. Columella about i mm.
lower than the upper ends of the tertiary pali, formed by about
6 rods, which are with difficulty to be distinguished from the
pali of cycle I.

lOI

ti

c

.*H W)

o-c

-• 1J

O u,

ZQ

Septa
in Cj'cles.

rt

c

Pali
in

ti .

Cycles.

p c

■ss

Wop J

K b

Greatest

by Least

Diameter

in mm

Remarks.

II

III.
IV.

V,

VI.

VII.

I 6, 6, 12, 22

2

3

4
5

6

7
8

9
lo
II

12

13
14

15
i6

17
i8

19

20

21
22
23

24

25

26

27

28
29

30

31

32

33

34

35
36
37

38

6, 6, 12, 22
6, 6, 12. 24

6, 6, 12, 24
6, 6, 12, 24

6, 6, 12, 24

6, 6, 12, 24, 2

6, 6, 12, 24

6, 6, 12, 24
6, 6, 12. 24
6, 6, 12. 24

6, 6, 12, 24
6, 6, 12, 24, 4
6, 6, 12, 24, 6

6, 6, 12, 24
6, 6, 12, 24, 8

6, 6, 12, 24
6, 6, 12, 24, 10
6, 6, 12, 24

6, 6, 12. 24, 16

5, 5, 10, 20, 4

7, 7, 14. 28, 4

6, 6, 12, 24, 6

6, 6, 12, 24, 8

6,6, 12,24

6, 6, 12, 24, 10
6, 6, 12, 24, 12

6, 6, 12, 24, 14
6,6,12.24, 14
6, 6, 12, 24

6, 6, 12, 24, 6
6, 6, 12, 18
6, 6, 12, 24, 14

6, 6, 12, 24, 16

6, 6, 12. 24
6, 6, 12, 24
6, 6, 12, 22

6, 6, 12, 16

6,6

6.6,1
6,6

5,6,12
6, 6, 12

6, 6, 12

6, 6, 12

5,6,12

?
?

6, 6, 12

?

6,5,4
6, 4. 12

6. 6, 12
5,4,12

6, 6, 12

?

5, 5, 10
7, 7, 14
6,6, /

6,6,7

6, 6, 12

6, 6, 12
6, 6, 12

6, 6, 12

6, 4, 10

]

6, 6, 12

6, 6, 12

6, 6, 12

6,6
6,6 ?
6,6

?

5
6

15
10

7

8

?

?

12

?

3
II

5
10

6

11/

9

1

6

15

'?

12

8
9

10

2

?

19

9

5
2

6

5

5
5
4

2
3
5

3

3
4

5
6

4
4
6

5

4
8
6

16

28
13+

33
23+

35

32

26+

30
22

18+

15+

15

26

28
23

23+
17+
29
28

26
55
14

43

33

37
35

35
41
15

17

14

23

10 xio

12x12
9X9'5

16x10
13x13

14x14

15x14
14x12-5

16x16
13x12
13 X 13

10x10

9x9
17x16

15x15

13 X 13

14 X 13
13x12

15x15
16x14

11 XII

18x17
14x13

16x16
17x17

17x17
16x15

19x17
17x16

IO-5X9-5

I

11x10-5

lox 9'5
17x16

Broken base, 3-5 X 3-5
mm., with about 14 septa.

Bent through 90 degrees.

Broken base, 3x3 mm.,
with 12 septa.

Broken base, 4-5x4 mm.,

with 12 septa.
Broad encrusting base, 14

mm. across.
Rods of pali and columella

very thick.
Broken base, 5-5x5 mm.,

16 septa.
Dead when obtained.

,, ,, ,,

„ ,. „ Pall

im certain.
Broken end,4mm.,i2 septa.
A very delicate specimen.
Some of pali II. fused with

III.

Some pali II. fused with

III.
Broken end,6mm.,i4 septa.

„ ., 6mra.,20 „
Dead.
Dead. Rods much fused

together.

thick. Pali

17+: 15x14

14
15
H-5

II

9X 8-5
9-5x9
7x7

6X 5-5

Constricted. 9 mm. below

mouth.
Rods very

fused.
Pali I. and II. pushed into

columella.

Rods remarkably thin and

deep down.
Rods medium,very shallow
2 pali II. fused with III.
Calicle constricted, pali

delicate and twisted.

Calicle much broken.

Remarkably shallow ca-
licle.

Broken end, 6 mm., 24
septa.

Pali III. not formed.
„ „ forming.

Pali I. scarcely separate
from columella.

Pali I. scarcely separate
from columella.

I02

Localities. — L, 22 specimens, Vasco da Gama, N., 71° E.,
i8i miles, stones. II., 6, Vasco da Gama, S., 75° E., 13! miles,
black specks. III., i, Cape Point Lighthouse, E., 26|- miles,
sand, black specks, and coral. IV., i, Buffalo River, N., 15
miles, coral and mud. V., 2, lat. 30° 3' S., long 27° 57' E.,
sand, shell, and rock. VI., 2, Cape St. Blaize, N.W. \ N., 30
miles, rock. VII., 4, Cape Vidal, N.N.E. \ N., (^\ miles, rock.

The nature of the adult pali and columella cannot be clearly
understood without a knowledge of their gradual development
(PI. II., figs. A-E). The earliest stage could be seen only in
sections, where there were 12 septa, 6 pali, and one rod in the
columella. The smallest corallum. No. 38, is in a slightly later
stage ; the pali are somewhat indefinite, and there are several
rods in the columella. The other coralla of VII. and Nos. i, 2,
and 3 have 12 rather oval-shaped pali, situated before the septa
of cycles I .and II. Those before cycle I. are nearer to the
centre of the axial fossa, and in the largest specimens merge
into the rods which form the columella, not being clearly dis-
tinguishable from them, though their separate origin is clearly
shown in sections.

In the specimens from station I. every stage between the
last could be followed to the fully adult, with well-formed, more-
flattened tertiary pali. Although scarcely typical, in the largest
specimens there seemed to be a tendency for the tertiary pali to
push the secondary pali into the centre of the fossa, where they
simply appeared as rather more flattened rods at the sides of
the columella, opposite the pomts of the chevron then formed
by the tertiary paii alone. Occasionally the secondary pali be-
come fused with the tertiary.

The variation in the septa may best be seen by reference to
the table. It appears to be mainly due to growth, only two
specimens showing radial, i.e., meristic variation, one having 5
and the other 7 systems.. The specimens, in addition, differed
materially from one another in the number of the rods of the
columella. In the 22 specimens of I. the number varied from
3 to 15, but the difference, while to some extent perhaps due
to age and the size of the axial fossa, was mainly brought about
by the variations in size of the separate rods of the pali and
columella. Sections of the narrow base have a simple rod in
the columella, and as fresh rods arise on the trabeculae which
join the septa to this rod, there should be typically a gradual
increase, with increase in size of the calicles. In No. 30 there
was a tendency to twisting in the rods, but this was also visible
in Nos. 12, 27, and 38. Again, the depth of the rods in the axial
fossa varied from "5 to 6 mm. below the edge of the theca.

The epitheca differed greatly, in some specimens giving their
external surface a glazed appearance, and in others allowing the
costae to be plainly visible almost to their bases of attachment.

I03

Many specimens were broken off above the base. In perfect
specimens the attached part was usually broadened by an epithe-
cal deposit, varying up to 3 mm. in thickness. Some specimens
were much eroded or dissolved away on the outside, and in
these it was noticeable that the epitheca was attached before
the septa, the costal edges of which were frequently brought
into prominence by the epitheca — and sometimes theca — in be-
tween them being removed. Other dead specimens showed
intermediate characters to one, in which nearly all the corallum
had been dissolved, leaving a cast of the intersepted and other
spaces in calcite.

It will be quite clear from a glance at the table and figures
that the number of growth lines is not a character of any im-
portance, nor probably due to seasonal changes. The lines con-
sist of circular elevations around the corallites, and appear to
represent pauses in their growths, which would seem to have
restarted with slight constrictions of the calicles.

It is possible that the specimens 35 to 38 do not belong to
this species They cannot be accurately compared with sec-
tions of the larger coralla at the same size, since the spaces in
the latter are filled in by secondary deposits. They approach
the smallest specimens of the other dredgings, but are evidently
much younger.

7. TROCHOCYATHUS CINCTICULATUS (Alcock),

(PI. II., fig. 2).

T. cincticiilatus, Alcock, Investigator Deep-sea Madreforaria,
p. 17, pi. II., figs 5, 5a (1898).

I refer four specimens to this species, which is presumably
founded on a single corallum, and that an extreme one, perhaps
produced by unfavourable circumstances. The epitheca is a
difficult character, though probably not a very important one.
It is in my specimens a deposit formed as the coral grows by
the retreating edge-zone of the polyp, i.e., that part which lies
external to the calicle. It covers over the costae, but shows
their position as ridges, and fills in the intercostal valleys. It
is somewhat obscured by boring organisms and the action of
erosion, which seems to have attacked the epitheca first. Al-
CGck's specimen in this respect doubtless shows the more typical
gro\Ath of the corallum.

Three cf my specimens closely resemble one another, and
show what I consider to be the typical central form of the
species, the description of which would be as follows : —

Corallum broadly or narrowly attached, subcylindrical or
conical, straight or curved. Epitheca covering the whole
externally to within i mm. of the top of the theca, filling in
the intercostal valleys, but showing ridges over the costae.

104

Calicle almost circular set round with four cycles of septa.
Septa entire, all about ' 5 mm. exsert, ending perpendicularly
against the axial fossa. Primaries and secondaries rather
thicker and broader, inner edges slightly thickened and
twisted.

Pali two series, those in front of the primary septa blunt
rods, intermediate in size between the rods of the columella
and those of the second series in front of the secondary
septa, which are so greatly broadened that they extend al-
most to "the tertiaries on either side.

Columella forming a fifth to a quarter of the diameter
of the calicle, situated about '75 mm. below the ends of the
pali, which again lie i '5 to 2 mm. below the upper edge of
the theca, essential, consisting of fine rods (fasciculate type),
varying with growth up to about 30 in number.

Diameter of calicle, 6-9 mm. ; height of corallite, 10-20

mm.

Locality. — Lat. 33° 3' S., long. 27° 57' E., 32 fms. ; sand,
shell, and rock.

The septa and pali do not vary from the above description
in the three type specimens, but the fourth is constricted round
the mouth of the calicle and in its pali, septa, columella, and
epitheca closely approaches Alcock's specimen. The columella
has five rods only, is equally deeply situated, as in the other
specimens, but its rods are not so sharply marked off from the
pali.

Genus HETEROCYATHUS, Ed. and H-

The genus is characterised by a low, broad, simple corallum,
having an As-pidosi-phon living in its base. The coral larva
fixes itself on a small Gastropod shell, in the cavity of which the
worm takes up its abode. The shell is generally of such a size
that the coral in its growth is able to completely surround it,
but the Aspidosiphon growing at the same time stimulates the
coral to broaden its base, in which it retains a spiral cavity open-
ing to the exterior, in which it lives. Where the coral does not
so affix itself, its shape is conical, base pointed. Only a single
species of the genus appears to be known. A number of so-
called species have been described on single or few specimens.
All are represented in the series from the Cape or in the col-
lection of the British Museum, and it is obvious that they re-
present merely the descriptions of isolated specimens drawn by
the dredge, etc., at random out of series of the same species.

105

S. HETEROCYATHUS AEQUICOSTATUS (Ed. and H.).

(PI. III., figs. 1-43-)

H. aequicostatus, Ed. and H., Cor. II., p. 51 (1857).

H. phillipinensis, Semper, Zeit. f. wiss. ZooL, XXIL, p. 254
(1872).

//. parasiticus, Semper, Zeit. f. wiss. ZooL, XXIL, p. 255 (1872).
(1872).

H. pulchellus, Rehberg, Abh. Ver. Hamb., XII., p. 8 (1892).
H. oblongatus, Rehberg, Abh. Ver. Hamb., XII., p. 9 (1892).

The collection contains upwards of 900 specimens of this
species, made up as follows : —

Number of
Specimens.

Depth in
Fatlioms.

Locality. Character of Bottom.

I .

About 700

54

Cape Natal W. by N. b\

Fine sand and algae.

miles

II.

14

47

Cape Natal W. b.y N. 41
miles

Sand and shell.

III.

45

56

Umhloti R. Mouth N. by
W. i W. 151 miles

Sand and shell.

IV.

4

39-40

Off Umhloti K. Mouth

Sand and shell ;
ground.

hard

V.

14

36

Tongaati K. Mouth, N.W.
by N. \ N., si miles

Sand and shell ;
ground.

hard

VI.

About 100

90

O'Neil Peak, N.W. \ W.
9^ miles

Broken shells.

VII.

I

45

Lat. 32° 53' S., Long. 28°
12' E.

Coralline stuff.

The collection contains so large a number of specimens that
I have been enabled to make a study of its variation. Com-
paring the specimens from each locality together, and with the
collections from different localities in the British Museum, it
was clear that each locality had a size-mode of its own. This
was apparently due partially to the particular species of shell on
which the coral had affixed itself in each locality, but mainly
to the environment, tropical forms, for instance, being usually
larger and coarser than Cape forms. At the same time, the
characters of the costae, septa, theca, pali, and columella were
not affected; and plenty of intermediates existed, so that all
clearly belonged to a single species.

io6

^

.

-*-3

1— <

a

^~'

"sisuadKO

o

PI

1)

r>

'~^^

O

X

X

O X

'^

KIUEOUnQ

<

t— 1

rt

5J

^•

o

r-

r»-.

^-4

1^

D

•iiniuis
•iniiAqdoA-iKQ

+

+

-1-

■n

-t

i

•X

o

X

5 '^^

■3 If

ii

<n

P)

J

^

a

>-a

H-t

«

rt

D

s

-^-*

a

Pi

uinAL(do.(.iB3

1— I

1)
a

a

CI

I

i"

X

X

I^

.

CS

g

-4-*

a

^

u

•sn;sooinb3T3

72

« ■+

C/J

ri

^

c3

o

O X

— ■

snq;B.Coo.i3pH

o
U

a

K^

£

S

X

^

_j

-J

r^

pH

ft

•sniEA.it:d

r^

11

72

-+

, ;

snqjBXoBJBj

lO

"H

ri

CI

o

o

O '^•

<

CO

a

a

rt

n

g

"a
aj

■CJO..

•mos.wBJ

•-f

72

-t

^

~

o

1) ,j

-f

snq^EXooqooJX

+

D

r;

Cl

i'

£

X

^

'O

>

ri

rt

;l

P<

ft

,11

fti^

u

u

-^ ,

D O

•uinuiuoAKd

■n

^ 7)

00

^

i^ C'

uinipqKid

o

- 1)

•+

<

o

o

o

rt

rt

ij

ft

Ph

u

.

.

i^

•um.iqnj

'fi

72

XI

<-i—

1>

^

uinqaqHiji

o

a

•*

a

I

o

o

S X

<u
Q

<

,

,

1

.

.

.

,

1)

•

;

;

o

;

"

*

\

>

c

r- '

_o

P3

D5

72

11

3J

•n

X

.5

^

p

\^,

-^

'fj

.

K*.

■J

cj

^/3

'J

c :

:_/.

o

2

0

"rt

rt

x: p

u

-^

Cm

g

>

-4-»

c

72

'I

13
o
u
o

•J

"aj
o
'J

'Si

g
11

X

1)

•-.«

U

o

'^

c

u

c

o

T*

■—

y.

u<

•-

(—"

T*

o

•J

Z3

11

^

a

a

1)

'~)

3 c

■zti

-c

:z

1)
72

u

~ft

(73

'X'

a

En do
Aeon

D ~fj

<

o
g

H

X
H

O

72

o

X

o

I07

The specimens from locality I. were particularly interesting,
as three types were distinctly visible in them. The descrip-
tions of each of these are given in the accompanying table. No.
3 were obviously young forms leading to Nos. i and 2, and
showing plenty of intermediates. Nos. i and 2, however, re-
present two distinct modes, between which odd specimens
existed, showing intermediate conditions in each of the charac-
ters which serve to separate them. Only two specimens were
at all doubtful in their position, and in one of these, of which
one side had apparently been broken down and regenerated,
half the corallum belonged to type i, and half to type 2. On
each of 450 of these specimens, besides a number from other
dredgings, I made 1 1 measurements, in an attempt to compare
them together, and to find out with some precision the charac-
ters of the species. The results were exceedingly disappoint-
ing, partially no doubt owing to the peculiar conditions existing
between the coral and its commensal Aspidosiphon, but mainly
owing, I believe, to the variation of these forms being purely
vegetative in their measurable characters, the direct result of
environment, ana hence of quite a different nature to that usually
included under the term " variation."

My first attempts were to correlate (i) the length by (2)
breadth of calicle, (3) breadth of base containing the Aspido-
siphon, (4) height of whole corallum, and (5) weight of the dried
corallum. Of these, i to 4 are dependent on the age, size, and
rate of growth of the two organisms, as well as on the size and
form of the original shell, on which the coral affixed itself, and
also on the growth of the Aspidosiphon. No. (i) by No. (5) at
fiist gave better results, but the weight of the corallum is largely
influenced by the size of the gastropod shell and the weight of
sand in the body, or in the tube of the individual Aspidosiphons,
a variation ranging through 1 5 per cent, of the total weight. A
real meristic variation was found in seven specimens, i'2 per
cent., in the number of cycles of septa. The number is so small
that they can only be chronicled with the remark that in this,
as in all its other specific characters, the form is extremely
highly specialised, and more or less definitely fixed. Yet it is
particularly interesting to note that four of these were in the
much less numerous specimens of the 2nd type. In many of
the specimens some septa of a fifth order were present. They
formed 2\ per cent, of type i, and over 50 per cent, of type 2,
while the per cent, in young forms was intermediate ; it was
noticeable that in only one case where 6 systems were present
were there less than 48 septa.

io8

Table showing the measurements, etc., of H. ncquicostatus
from dredging I. : —

Plates III

(I) Type I.

(2) Type 2

(3) Young

X IclLV'O Jl I. i •

Figs. i-ii.

Figs 12-21.

Figs. 22-32

Number of Speci-

mens

200

50

200

Average Length of

Calicle

g* I" mm.

9-33 mm.

6-3 mm.

Average Breadth of

Cahcle

8-21 „

9-47 n

5-84 „

Average Height of

Coral

7-03 „

9'7 „

4-28 „

Average Weight of

Coral

•640 grms.

•588 grms.

•277 grms.

Spiral right to left

193

50

198

„ left to right

7

0

2

Spiral, I opening...

4

6

II

„ 1 + 1 openings

136

29

104

» 1+2

43

14

66

„ 1+3

17

I

19

Colour, All Septa...

37

} 50

181

No Colour

54

CyclesI&II-I>II

16

0

0

„ „ 1=11

77

0

0

„ „ &III

16

0

19

Septa, 6, 6, 12, 24...

194

22

185

Irregular ...

I. (6, 6, 12, 14)

I. (6, 5, 11,22)

2. (6, 6, 12, 24, 2)

,, ...

I. (6, 6, 12, 24, 2)

I. (6, 6, II, 20)

7. (6, 6, 12, 24, 4)

I)

I. (6, 6, 12, 24, 4)

2. (6, 6, 12, 22)

I. (6, 6, 12, 24, 6)

1) •••

2. (6, 6, 12, 24, 6)

7. (6, 6, 12, 24, 2)

2. (6, 6, 12, 24, 8)

. • •

I. (6, 6, 12, 22, 4)

I. (6, 6, 12, 24,^.16)

• > •

8. (6, 6, 12, 24, 4)

>•■

• • ■

3. (6, 6, 12, 24, 6)

■ ■•

2. (6, 6, 12, 24, 10)

>••

MeristicVari-

ations

(7, 7, 14, 28)

I. (8, 8, 16, 32)

I. (5, 5, 10, 20)

I. (5, 5, 10, 20)

I. (7, 7, 14, 23)

I. (5, 5, 10, 20, 2)

109

The results are valuable, because they so definitely point to
the existence of two perfectly distinct modes of growth, almost
two varieties, in the species as well as the ordinary continuous
and vegetative vanations. Type i also shows the presence in a
single habitat of small though perfectly discontinuous variations
in the colouration of its septa, variations not found in type 2 nor
in the young. Some are colourless, others have the septa of
cycles I. and II. black, and yet others have all the septa black.
In some the septa of cycle I. are blacker than those of cycle
II., and in yet others cycle IV. septa have not as yet attained
their colouration. When the caHcle is looked at from the
surface, the spiral chamber in which the Aspidosipkon lives
runs from right to left in most cases, but in a certain number
from left to right. In 50 specimens that I have ground down
a shell is present in each. In one of a left to right spiral, the
difference was due to the shell being sinistral, but in two others
examined I could not determine with certainty their character,
though both seemed to be dextral. The Aspidosipkon cham-
ber commonly has, in addition to its large opening to the ex-
terior, a second very minute op>ening on the under side of the
corallum, close to the mouth of the shell it originally inhabited.
This is, I believe, normal, but in 30 or 40 per cent, of the speci-
mens there are 2 or more of the small openings, the additional
ones lying along the spiral tube of the commensal.

in addition to the specimens "already referred to, four speci-
mens, all belonging to type 2, show fission of the calicle, or
possibly in two cases budding from within the calicle (see PI.
III., figs. 35-38). A fifth specimen, belonging to type I, had
become killed over the greater part of its calicle, but the re-
mainder budded out a fresh caUcle, which now stands up on a
broad base, situated on the parent cup. The dead specimens,
of which over 100 were obtained in dredging i, do not differ in
size from types i and 2 as described, none belonging to the
third or young series ; they bear the same proportion, nearly 4
to I, that these types have to one another.

C

no

G-'

ti:

en
C
<v

s

'o

CI,

O — '

tuO

5f;

o

M— 1

(1)

C)

•n-;

Id

C

B
o

M

U(

M-l

T3

C

a

M

ly)

<u

&

>-

u

"♦-J
O

u

o

en

d

rt <u ^ p ^

c
o

cu

a,

St'o

~ oj t-.

n c ra .1)

i« >.

-^•^■u

■^ ^-1 rt ^ c

§ >^l '3 •- S

p o & o « ^

-1 cfi rt ;-. c/) H-»
C/2

ex,

T3

5
■-^

« aJ

o o
o o

^c

O

x: u

«

-^

n

XI

x:

^

u

rn

"-,

rt

■r>

o

u

l_

w 1) C O

O tu
'Jo .^

O

t- i; ?

c
o
1)

^ -2

O ■«-,

U-t- X

g<3

•—I •," •

1) —

1) tn

[«■--■ O

^ O Q.

- s

1)

<u ■:; s

P c? ^

O p

>" 3 t5
I— , o u

^ "^ h
3

" =^ ^'o

„ ^ M

(u :: o

GOP

O

*j 1) c

l^^

o o

rt

-50 5^

it; « "o

(U I— . u

c n CO

c — g
D >-< o

en u

i5-3-
c ?►. >

u '-J ^■
=" 2 ^

rt rt C
Xi a!

1- SJ^

1>

w p

rt

' S •- ^

X3S •-

c :: rt

<u X ^

■-« "r- U

c ■" l-i

o::

° u u

'-' <U OJ

3 O

4- o I-

O V-, X

a

O .
CJ .ti

G O

X u
o «

o V 'Q

o

cs

0 ■> x" '

(/5 .^ r

o-2ii

rt

3 °

O

>x a,

— - -J5

O 'U u

'J

1) -x

./^ J) ^ f*

.5 f^-

^xi=g

X qj
X X .

« 05

0)

Wio ><

•C XI t^

<U rt 0)

3 S-"

& C G

G U ii
01 "w o:

•nap

^ qj i-<

3 ^.-^ ,
'> '-' '^ '

G c "^'^ X

00
•*

(J ^

10 -1

^^;

tJ3

' 3
>

■1 £1. u

^"Ig

OJ

'J

o

r-. . '—I

O ,-^

CJ X.

3

_o

"o .•

CJ K^

OJ u

G tiO

in O

tJO G 5 x

t3 c^ „ rt
OJ 3"."G V)

-S S 3 y

p

u -- 0 x!

f— «

C/) .r ■-> G

(-^

as
-last
and
alii

a.

u ■) t; G

3j r- aj C/D

, .

X :n T3

G o G t"
G ^ G aj

2 ^ P'G

(U ^ aT ^

P O .^ ul

•3 ~ 'J/5 iJO

o

CO

4)
tj

en

c

5-S
O tfi

o o
G ^

c

x:

> io

p ^<

^ G

5 >-•

E aJ n

c " ;^ :

G >^ S ■
.o«

aj i;^
tfi ^

rt G r

' .^ T- i-^

I z: t; -p

G 5^.

o

•Si

u aj ■

aj Tj

aj

c

a)

0

>

CJ

0

(4-,

c

0

g"

a
0

X

-4->

-*-»

aj

(1)

>

In/l

0

Tl

X

aj

c«

aj G

-gen'

.:; aj

X CJ
-►-' ct

rt t

3 ■"
O

ji -c :: ~JD^^

tn G

:^«=^^

aj -p

aj S — '— '

G -o ^ ;i; B

-0 u .

'J CJ

1-' rt • rt

rt

3 p:: r:£
^x .5

S P-,3

^-G ^

OJ . _; :n; ^

:So^

fa y rt

G ''•X
G CJ ^

aj tn

^ 6 fa

> S °

aj
rt
-4-*
tn
O

rt

u

0

CO

u

rt

x:

CQ

H

u

aj

Ill

J}i I^^V'S. ^

to
O

CSS

<3

en

C

(U

a

-»-H

u

<v

" "^

c
3

o

t**

;>i

O

<.)

(U

^

s— '

M-<

>— (

n

tj.)

'CJ

r^

r-"

• fH

rC

tJO

Tj

ri

OJ

-C T3

r-i

rt

g

M

o

'.f)

M-H

cu

Oh

>-.

u
<u

o

OS

u

-4—1

C

■^
o

I

Oh

(X,

o^

3 -1-' 3

I- c ;^ i^ - S «

f— !

• — rt

a,

•i^-w

Ml

■n 0

•^ L>

3
0

it ^

>-

> c

, .

>> .

i~o

i3o

>

^ = 1^

= ° >■
.2 5 •

^ si

C

5J l^ OJ p
(U 3 r- P3

"= G

c S
5 -^'

O r
O 1-1

S W

^^

SB
3
O

-a "-

U en
O

tn ■

«^
en C

S >'

0) c3

rt.2

en

rt

C r- IJ P

Sen

il

o

o
o

o

^ en

;^ .5 u

a en ,Q

£ H-- ^•- «

"S "!! 1^ ^

w-^ en
ire rt i5

C3

-^ 1> li U

U en E ."

gj U 0; ^

c

i[5^i^

1

U y en "tn
Xj P IJ 0

W"^ "S P

h ^ P-.—

^ en CJ rt

rt

O"^^

en .3 0 nJ"

0

2 0 rt 0

rt C en en

rt
0

t£^ (U'-S

Lj

c >-,x:

^

5 ^ -" OJ

j_;

OJ

0 jj n 0

^'Xj JZ u i

>

u o

!=-0

r-, ,.^
en ^

3 ^^

^

^-/.
c

^•?

rt C

en i^
"u P

OJ

3

>

f—

tj

OJ

rt

>-

tr,

d;

0

t;i

en

en

u

r^

0

..-4

-a

OJ

(1)

"rt
0

^a

>. <u
rt ■"

.2 °

o -£
>;

ti o

0

en « X3 E

c I-; o c;

£ o

OJ >,

en '—
3 ^
^*- rt

>l

una
^' rt ?e

O -►-, -"
p rt -4-.

P ^ S

(J en ?^

p"- ^
tn -ti 'J

(U en

N -a

■« rt

C &
— O

eu

W
en

C
(U
T3

rn

0

-c

.•-;

0

_^j

•^

JZ

*-*-<

-.1

n

Jj

rn

^

en

rt -n

S

rt

.2

5 £
o to

1) o
3 "O

CU

9
"o
O

112

Of the other dredgings, the specimens of II., Ill, and VI.
closely resemble tnose of I. No. II. has 7 specimens of type i,
3 of type 2, and 4 young ; while III. has 20 of i, 2 of 2, and 22
young. No. VI. has only type i and young in the proportion
to one another of 4 to i, but the tendency of all its specimens
is to be smaller than in I., II., and III. The calicle of the single
specimen of VII. belongs to type 2, but the specimen is unique
in that it is conical, and has no Aspidosiphon nor any trace of a
gastropod or other shell (PI. III., figs. 33 and 34).

The specimens from Nos. IV. and V. differ from the rest
in that {a) their costae are very distinct ridges, rounded, but
with deeper intercostal spaces than in type i ; {b) their septa
tend to have a few of cycle V., and are intermediate in most
characters between types i and 2 ; {c) their pali rather belong
to type 2, the specimen shown in fig. 39 being one of the nearer
to type I ; and {d) their columella is generally represented by a
series of very fine rods. The average size of the specimens is
larger than from other dredgings, 16 corals (4 of IV. and 12 of
V.) giving average length of calicle io"5 mm. by breadth g'dg
mm. Size almost precludes any except one of those shown in
fig. 40 from being regarded as young, but some almost resemble
young forms from I. greatly magnified, noticeably those shown
hi figs. 41 and 43.

At first sight, the specimens from Nos. IV. and V. almost ap-
pear to represent a different type to the rest, and in this con-
nection it is noticeable that they come from similar depths and
localities, considerably differing from the other localities. Some
specimens, however, approximate towards each of the types —
IV. are closer to type i — and in each character there is a series
between. In seven of the specimens from these dredgings the
corals have settled on relatively large, conical shells, over which
they are attempting to spread themselves (figs. 40, 41, and 43).
In one where there is no trace of an Aspidosiphon the shell is
flattened and still larger (fig. 42). One has its base, in which
no shell can be seen, much distorted by the presence of two
worms, having two external apertures and two spiral cavities.

Genus CARYOPHYLLIA (LAMARCK).
9. CARYOPHYLLIA RERTERIANA (Duchassaing).

C. berteriana, Ed. and H., Cor. II., p. 19, pi. D I., fig. i.

C. bertcrinna et formosa, Pourtales, Cat. Mjts. Comp. Zool., IV.,

p. 7 (1871); Mnn. Mus. Comp. Zool., IV., p. 34, pi. VI.,

figs. I and 2 (1874); Bull. Mus. Comp. Zool., VI., p. 99

(1879)-

There are two specimens in the collection of this typical West

Indian coral, which was also obtained by the Challenger Expedi-

113

tion at the Canary Islands. Both were dead when obtained, but
had preserved their structure, which approaches most closely to
the form fibred in the Hasslar Expedition Report. One is
round, 22 mm. in diameter across the calicle, while the other is
slightly compressed, i/xiS'S mm.

Localities. — Smaller, 90 fathoms, O'Niel Peak, N.W. \ W., 9}
nules, broken shells ; and larger, 30 fathoms, Buffalo River, N.,
15 miles, coral and mud.

TO. CARYOPHYLLIA CAPENSIS, n. sp. (PL I., figs. 4 a-d).

Corallum conical, closely set on the outside with low, equal,
finely-granular ridges, corresponding in number to the septa
and formed by an epithecal deposit, covering over the costae
and to some extent filling in the valleys between to within 2 mm.
or less of the edge of the theca

Calicle rather longer than broad, with theca sHghtly higher
against its long axis. Septa five cycles, the last generally in-
complete, none more than I mm. exsert, with very little differ-
ence between the orders. Primaries and secondaries sub-equal,
considerably broader than the rest, tertiaries being intermediate
']:ese three orders alone connected to the columella and with

itire (not toothed) edges. Quaternaries much thinner, but

ojecting again in front of the quinaries, and, where the latter
are present, generally with from 2 to 4 blunt teeth.

Pali present in front of the tertiary septa alone, varying from
rather thick, round pillars of rough appearance to flattened
plates of upwards of rs mm. in iDreadth. Columella rough
pillars, varying in size with number present, about 4 mm. below
the edge of the theca, almost level with the tops of the pali.

Locality. — 32 fathoms, lat. 33° 3' S., long. 27° 57' E., sand,
shell, and rock.

The size and variation may best be seen in the accompanying
table of the specimens : —

No.

I

2
3

4
5
6

Septa in cycles.

6, 6, 12, 24, :iA
6, 6, 12, 24, 34
6. 6, 12, 24, 26

6, 6, 12, 24, 12
6, 6, 12, 24, 8
6, 6. 12, 16

Pali.

Rods to
Colum-
ella.

HeiL'ht
in mm.

12

1=^

21

II

6

20

II

I.

4

22
15 "5

12

q
8

13-5
6

Diameter of
Calicle in mm.

14-5x12
15 X 14
II -5 x 10-5

9X8-5
8x8

5-5x5

114

In the smallest three specimens pali have not as yet been
properly formed, and the tertiary septa, except in a few systems,
have not become connected with the columella No. 4 has
distinct rods arising before five of the secondary septa, and
forming part of the columella. No. 5 has three (? four) true
pali before tertiary septa, but as also in No. 6, the other spines
are less definitely connected with particular septa than in No.
4. Nos. I, 4, and 5 are conical, with narrow bases of attach-
ment ; but No. I has secondarily deposited an epitheca, covering
over the upper sides of two barnacles that have become at-
tached to the outer side of its corallum. Nos. 2 and 4 have
large and spreading bases, while No. 3 is intermediate in this
character.

The most important characters of the species appear to be
five cycles of almost equally exsert septa, low costae corre-
sponding in number to the same and equal in size, and 12
relatively narrow pali, scarcely higher than the deep fasciculate
columella.

The rule in the species, most typical of Caryofhyllia, appears
to be that the number of pali is equal to one quarter of the
total number of septa present, whereas in the fully adult speci-
mens of the species under consideration they would seem to be
equal to one eighth. This character alone serves to separate
it from most other described forms, but it allies it to those of
which accounts have been given under the names Bathycyathus
indicns, Ed. and H. {Ann. des. Sc. nat., 3e ser., pi. TX., fig.
4) and B. inacjilaius (Pourtales, Mem. M?is. Coinp. Zoo I., vol.
IV., p. 34, pi. VI., figs. 5, 6), though it will be obvious that
neither of these can be the form now described.*

II. CARYOPHYLLIA EPITHECATA, n. sp. (PI. I,

figs. 3 a-c).

(?) C. clavus, var. efithecata, Duncan, Trans. Zoo. Soc, VIII.,
p. 312, pi. XLVIIL. figs. 13-16 (1871).

I refer 30 specimens to a new species which I believe to be
identical with a form that Duncan deemed to be a variety of
C. clavus. While I cordially recognise the valuable work that
Duncan did in showing the range of variation m this species of
coral, I fail to see that the greater number of his varieties really
represent types of growth — or modes of discontinuous series —
to which the term " variety " should properly be restricted. In-
deed, all the varieties of this coral seem to me to fall within
continuous series with the exception of var. epithecata, the

The genus Btitliycviiflnis was lijihtly absorbed by Duncan into CarvopUyllia.

115

meagre description and more particularly the figures of which
lead me to doubt whether it belongs to the species at all.

Caryophyllia clavus, according to Duncan and all other authors,
has five cycles of septa, whereas var. epilhecata in the figures,
supposed to be typical of it, is represented with a roimd calicle
and only four cycles. It is not a matter of size, since five of
Duncan's figures of C. clavus have their calicles of the same size
or smaller than the type, and yet have pali and septa of the
fifth cycle. Furthermore, I have examined a large number of
specimens of Duncan's var. smithi of 5 to 12 mm. in diameter
from English, Norwegian, and Mediterranean seas, and all with-
out exception possess a considerable number of septa of the
fifth cycle (quinary septa).

The description of C. efithecata would be as follows : —

" Corallum typically conical, often bent to one side and slight-
ly compressed, sometimes attached by its small end, sometimes
free owing to its peduncle having been dissolved away or
broken. Epitheca present, extending over the outside of the
corallum almost up to the edge of the calicle, and filling in the
intercostal spaces. Sub-equal costae present, showing through
the epitheca to the base of the corallite (if the corallum has not
been dissolved away) as low, broad, sub-equal ridges separated
by narrow valleys.

" Septa forming four cycles, those of I. and II. sub-equal,
slightly broader and more exsert than those of III., which
again are broader and more exsert than those of IV. Pali only
before the tertiary septa, relatively thick, much-flattened rods,
of about 1*5 mm. in breadth. Columella a quarter to a fifth of
the diameter of the calicle in breadth, consisting of 2 to 9
twisted rods filling up the axial fossa."

I append a table of the specimens, from which it will be seen
that 22 specimens have the typical number of septa, 4 specimens
have the fourth cycle incomplete by two or four septa, while 4
others show meristic variation, two having five, one seven, and
one eight systems. Eighteen of the specimens were, so far as
I could see, free, some showing scars of former adhesion and 12
attached, all except two, by quite narrow peduncles. Some
specimens show on their sides transverse ridges of growth, but
these are much obscured in many by the deposited epitheca and
in certain by both theca and epitheca having been dissolved
away, leaving the edges of the septa exposed with the inter-
septal spaces filled in, apparently by deposited corallum (stereo-
plasma).

The size of the processes of the columella vary with the size
of the axial fossa, but more particularly with their own num-
ber ; in all cases they nearly fill in the fossa. The summit of
the columella likewise varies in depth, in some specimens being

ii6

almost level with the top of the theca, and in others up to 3*5
mm. below. In all cases the summits of the pali are about inter-
mediate between the height of the columella and the theca.
The septa vary up to i " 5 mm. in exsertness, in all cases those ot
cycles I .and II. standing up about twice as high as those of
III. and IV. ; they show the usual ridges found in Caryophyllia,
radiating out from the theca, passing at places into spines, which
are characteristic of the pali.

The species approaches C. arcuata, Duncan {non Ed. and H.)
and C. antillarinn, Pourtales, the descriptions of neither of
which can, in my opinion, be at all exact.

Table of C. epithecata.

Locality

No. of Septa in

No. of

Rods of
Colum-
ella.

Height

Diameter of

and No.

Cyies.

Pali.

in mm.

Calicle in mm.

I. I

6, 6, 12, 24

12

6

29

14-5 X 13

2

6, 6, 12, 24

12

2

22

12 X II

3

6, 6, 12, 24

12

9

22

13 X II -5

II. 4

6, 6, 12, 24

12

3

21

II X 10

5

6, 6, 12, 24

12

4

21

11-5 X 10-5

6

6, 6, 12, 24

12

4

19

10-5 X 10

7

6, 6, 12, 24

12

6

i6-5

10 X 9

8

6, 6, 12, 24

12

6

18

II X 10

9

6, 6, 12, 20

II

4

12

9x8

10

6,. 6, 12, 22

12

5

II

8x7

III. II

6, 6, 12, 24

12

4

18

II X 10

12

6, 6, 12, 24

12

4

26

12 X 10-5

IV. 13

6, 6, 12, 24

12

2

16

8-5 X 8

14

6, 6, 12, 24

12

4

27

11-5 X 10-5

15

6, 6, 12, 24

12

3

18

9 X 8-5

16

6, 6, 12, 24

12

4

17

8 X 8-5

17

6, 6, 12, 22

12

4

13 '5

8 X 8-5

18

6, 6, 12, 24

12

3

17

8-5 X 7-5

19

6, 6, 12, 24

12

5

17

io-=? X 9

20

6, 6, 12, 24

12

3

15

8-5 X 8

21

6, 6, 12, 22

II

5

16

9-5 X 9

22

6, 6, 12, 24

12

5

17

10 X 9

23

6, 6, 12, 24

12

3

25

9 X ?

V. 24

6. 6, 12, 24

12

3

23

13-5 X 12

25

7, 7, 14, 28

14

3

24

13 X II -5

VI. 26

6, 6, 12, 24

12

16

1 ^3

9-5 X 8-5

VII. 27

5, 5, 10, 20

10

4

14-5

9 X 8-5

28

5, 5, 10, 20

ID

3

19

10 X 9

VIII. 29

6, 6, 12, 24

12

4

2.T

12 X 10

30

8, 8, 16, 30

16

6

17

16 X 13

Localities. — I., 124 fathoms, lat. 34' 26'' S., long. 25° 42' 45"
E., 3 specimens, rock. II., 8 5 -go fathoms, Cape St. Blaize, N.
by E. \ E. 65 miles, 7 specimens, rough. Ill, igo fathoms,
Eion's Head, S. 72° E., 47 miles, 2 specimens, green sand and
black specks. IV., 230 fathoms, Vasco da Gama, N. 71° E.,
\%\ miles, II specimens, stones. V., 210 fathoms^ Cape Point

117

Lighthouse, E., 26^ miles, 2 specimens, sand and coral. VI.,
80-100 fathoms. Cape Vidal, N.N.E. i N., 9^ miles, i specimen,
rock. VII., 185-200 fathoms, Cape Natal, W. by N. | N., 11
miles, 2 specimens, sand and mud. VIII., 310 fathoms, Buffalo
River, N., 15 miles, 2 specimens, coral and mud.

12. CARYOPHYLLIA EPHYALA (Alcock).

C. epkyala, Alcock, Investigator, Deep-sea Madreporaria, p. 13,
pi. I., figs. 4, 4a (1898).

C. scyllaeoniorpha, Alcock, idem., pi. I., figs 3, 3a.

C. scobinosa, Alcock, Siboga, Deep-sea Madreporaria, p. 8, pi. I.,
figs. 2, 2a (1902).

I have referred five specimens to this species though they ai^e
to some extent divergent from it, approaching both to C.
scobinosa and also to C. scyllaeomorpha. Their variation may
best be seen in the accompanying table : —

H—

0 d

<f> ~

No.

0

Septa in Cycles.

Pali.

CO c
CO C

'3

Diameter of
Calicle.

J

O,^

X

I

I.

11, II, 22

II

4

12

9x8

2

I.

6, 6, 12, 22

12

5

II

8x7

3

I.

6, 6, 12

6

2

4

3-5X3-5

4

II.

6, 6, 12, 24

12

6

13

9-5X8-5

5

III.

8, 8, 16, 30

16

6

17

16x13

Nos. 2, 4, and i — the latter an obvious variation from hexa-
symmetry — approach closely to C. ephyala in shape, but are
rather more constricted at the base than the type of that
species. All further agree with that species in the general
characters of costae, septa, pali, and columella, but all parts in
the Cape specimens are more delicate and thinner.

Considering the whole question, if Duncan's account of varia-
tion in C. clavus {Trans. Zoo. Soc, VIII., p. 310) and my account
in Flabellum rubrum (Marine Investigations in South Africa,
vol. II., p. \\J et seq) be correct, I cannot consider that C.
scyllaeomorpha and C. scobinosa are beyond the limits of varia-
tion of C. ephyala. The same remark applies to several other
species of the same and other authors, but my specimens do not
approach any of them closely enough for me to deal with them
here.

ii8

Lastly, No. 3 is interesting, as showing the mode of growth
of the species. Six pali are present before the secondary septa,
while in the larger forms pali are only found before the tertiary
septa. This change must be brought about by the first six pali
fusing wdth their septa, and a fresh series forming opposite to
the tertiary septa,* a modification of the same condition ap-
parently producing the variation found in No. I.

No. I approaches C. scobinosa and scyllaeomorpha in shape
and in the character of its costae, which are overlaid and filled
ir. by epitheca. They are distinct near the margin of the
calicle, primaries more prominent, but below in some parts are
not visible, as in scyllaeornorpha, and in others present char-
acters described for scobinosa. In one system the quaternary
septa may be more exsert than the tertiary as in scobinosa, but
in another the reverse is the case. Granular striae on the
septal sides, radiating from the theca and running perhaps into
Imes of spines, or small ridges, are common, so far as I have
seen, to the septa of all Caryophyllia, and represent the centres
of growth or deposition of the septa. The axial fossa in all
my specimens is relatively shallow, but varies in every individual
specimen.

The arrangement of septa in Caryophyllia is essentially
hexaradiate. I have not enough specimens to see definitely
whether the species has typically four or five cycles of septa.
If, as I believe, four cycles is the adult number, No. 5 is a true
variation, but if five cycles, possibly only a stage of growth ;
the same remark applies also to scyllaeomorpha.

Localities. — I., 85-90 fathoms, Cape St. Blaize, N. by E. \ N.,
65 miles, rock. II., 80-100 fathoms, Cape Vidal, N.N.E. i N.,
g\ miles, rock. III., 310 fathoms, Buffalo River, N., 15 miles,
coral and mud.

Genus CERATOTROCHUS.

13. CERATOTROCHUS JOHNSONI (Duncan. PI. I.,
figs. 5 a-c, and PI. II., fig. M.)

C. johnsoni, Duncan, Proc. Zool. Soc, 1882, p. 217, pi. VIII.,
figs. 5-8.

Locality. — I., six specimens, from 54 fathoms, Cape Natal, W.
by N., 6j miles, fine sand and algae. II., one specimen, from
45 fathoms, lat. 32° 53' S., long. 28° 18' E., coralline stuff.

Seven specimens in the collection agree with Duncan's de-
scription in the absence of pali and any determinate epitheca, in

* The same method of jjrowth is indicated in Duncan's account of C. cyathiis
(Proc. Zoo. Soc, 1882, p. 214).

the character of the costae, in having wide interseptal locuii,
in the character and ornamentation of the septa, and in the
nature of the columella.

The specimens, however, are larger than the type, i.e., have
a wider calicle ; the six from No. 1. above, varying from 1 2 x
io*5 mm. to 9 X 8 mm., and averaging 10 x 9-2 mm., while the
specimen from No. II. is obviously young. They are very
slightly curved, and their p-rowth appears to have been vigorous.
All are relatively thin and dehcate compared to the types. In
the six apparently adult specimens, the full number of septa
of four cycles are present, and in one specimen there are two
septa of a fifth cycle. The septa of cycles I. and II. are sub-
equal, varying up to 2 mm. in exsertness, rounded above, but
ending almost perpendicularly against the axial fossa ; deep
down in the calicle — scarcely visible from above — they meet
the columella. The tertiary septa are a little larger than the
quaternary, but much more delicate, and only about half as
broad and exsert as the first twelve. A little below the top of
the columella they extend in almost horizontally to join it, com-
monly rising a httle, this part generally ending above in a
v/avy edge. The columella consists of a series of slightly
twisted, thin, anastomosing, plate-like lamellae, the whole in the
largest specimens forming about a quarter of the diameter of
the calicle, but in the smaller decreasing down to a sixth or
seventh.

The series of arched thickenings on the septal sides, parallel
to their upper inner edges, described by Duncan, are most
noticeable and peculiar. I do not find any record of their oc-
currence in any other member of the Turbinolidae, though I
have myself found them in a fragment of a coral apparently be-
longing to the genus Diincania from the Cape. Superimposed
on them transversely, and continued into the valleys between,
are the minute ridges, which, corresponding to the so-called
" lines of growth," seem to be a constant feature in the family.

None of the specimens show any definite external growth
lines, though one had evidently been partially killed at one edge
of its calicle, but grew out again, the polyp subsequently en-
closmsf the whole corallum as before. None likewise show any
definite marks of attachment, but it is quite evident that all
reallv were derived from fixed stocks. One specimen ends
basally in three points. One of these appears properly to be
the point, the other two having arisen in the regeneration which
succeeded some injury.

The specimen from No. II. is a young form, with only 14 of
the quaternary septa developed, though the rest are complete.
The characteristic appearance of the axial fossa, so largely
broug-ht about by the continuations of the tertiary septa into
the axial fossa, is not as yet developed, though clearly indi-
cated.

I20

Genus DUNCAN I A, Pourtales.*

There is only a single perfect specimen of this genus in the
collection. It was broken off its narrow attached base by the
trawl, but before this its cavity near the base had become quite
solid, filled in by corallum.

In the British Museum Collection, I found a single specimen
of D. barbadensis, Pourtales, which evidently had been sent
by Pourtales hmiself to Duncan, or Moseley. It was not in
good condition, but showed signs of the regular hexameral ar-
rangement. Neither theca nor epitheca were distinguishable,
distinct from one another. There was a simple wall to the
cahcle, outside which the tissues did not extend. The tissue,
which forms a theca, arises by an uprising of that depositing
the basal plate, while that building up an epitheca is the side
wall of the polyp. The wall therefore of the calicle of Dun-
cania is an epitheca. The interseptal chambers or the interior
of a cahcle may be filled in by solid corallum (stereoplasma),
deposited by the tissues, but any deposition without the forma-
tion of tabulae or dissepiments is not a character of any import-
ance in the TurbinoHdae, nor, so far as I am at present aware, in
Madreporaria.

One of the characters then on which Pourtales founded the
genus Duncania, and both the characters which Duncan gave
for his alliance Haplophylloida, I can only regard as quite
worthless in the genus under consideration. Whether it be
really distinct from certain other genera in the Turbinoloida,
with which it should be placed, I cannot at present express an
opinion. It is almost certainly identical with Haplophyllia,
which was also described by Pourtales. The only difference
lies in the columella, which in the latter genus is described as
being " styliform, strong, very thick at the base," while the
Duncania type has a " multipillared columella," and the present
q:>ecimen is intermediate.

14. DUNCANIA CAPENSIS, n. sp. (PI. I., figs. 6 a-c.)

Corallum cylindrical, with widely open, rounded calicle. Wall
a thick epitheca, with a tendency to shghtly constrict at the
mouth, worn away below showing fine longitudinal lines, de-
pressions opposite the attached edges of the septa.

Septa 48, in four cycles, I. to III. meeting in the centre of the
calicle. Primaries and secondaries sub-equal, with their upper
edges rising i to 2 mm. above the upper edge of the epitheca,
rounded above, but with smooth edges, save for a single fine
paliform tooth to each where it runs in to join the columella,

* Zool. Results of Hasslar Exped., 1874, P- 44-

121

from surface view scarcely distinguishable from the pillars of the
same. Tertiaries not so exsert and narrower, generally with
three paliform lobes to each, decreasing in breadth but in-
creasing in height and sharpness towards the centre of the
calicle. Quaternaries very narrow, but quite distinct.

Columella shallow, three or four fine-pointed spines, quite
similar to and possibly formed by the inner pali of the larger
septa.

Height, 17 mm.; diameter of calicle, 15*5 mm.; of broken
base, 3' 5 mm.

o

E., i8i

Locality. — 230 fathoms, Vasco da Gama, N.
miles, stones.

I have described the single specimen somewhat minutely, be-
cause I cannot be sure what are really its specific characters. It
is just possible that it may be the same as D. barhadensis,
Pourtales, if one supposes that the specimens, on which that
species was founded, were of slow growth, and so came from
an unfavourable environment, while the Cape specimen came
from a peculiarly suitable one.

Genus CYATHOCERAS Moselev.

The genus differs only from the last in the fact that the
columella is formed by twisted processes, not straight rods, and
that there are no paliform lobes.

15. CYATHOCERAS CORNU (Moseley).

C. cormi, Moseley, Challenger Report, p. 156, pi. IV., figs. 7, 7a
(1881).

Three specimens (1-3), 310 fathoms, Buffalo River, N., 15
miles, coral and mud ; and one (4), 330 fathoms, Vasco da
Gama, N. 71° E., i2\ miles, stones.

en i5

m ij

Height

Diameter

Septa.

in mm.

in mm.

Sliape.

I

6-{-6+i2+4

3

II

5

Bent conical attached
to a piece of.

2

6-f-6-M2-(-i6

4

19

7

Bent conical, two lines
of regrowth.

3

6-1-6+ 12+24

4

15 "5

3-5

Straight conical.'

4

64-6+12+24

->
0

14-5

10

Conical broken base,
mouth a little over-
grown.

122

The specimens are smaller, but otherwise closely correspond
in every particular with the types in the British Museum. All
have the inner edges of the septa against the columella sinuous
as described, but there is rather more marked difference in size
between the primary and secondary septa, which are sub-
equal, and the tertiary. One specimen shows growth lines
clearly. Nos. II. and IV. have an appearance as if formerly
broken ; it is, however, due to the epitheca overgrowmg the
edge of the calicle and the subsequent regrowth of the polyps.

III. THE ANATOMY OF THE TURBINOLIDAE.
I. GENERAL OBSERVATIONS.

I have below appended some remarks on the anatomy of
such of the Turbinolidae as I have been able to examine. The
results, so far as the classiftcatory characters of the different
genera are concerned, may be seen by a glance at the accom-
panying table, in which I have included Flabellum rubrum and
pavoninum, of which I previously gave an account, Para-
cyathus farvulus, a reef species with commensal algae from
my own Maldive collection, and Caryophyllia smithi from
Norway.

The investigation assists but little in any attempt to find out
the relationship of the different genera to one another. I have,
however, employed the examination of the polyps wherever
possible to checK the species, which, in the first place, I deter-
mined by their coralla alone. Having already seen that the
polyp anatomy did not differ in Flabellum rubrum, in spite of
iis considerable variation in growth form, I examined polyps of
the most widely-divergent growths, of Heterocyathus aequi-
co status (lo), Paracyathus farvulus (4), Trochocyathus raw-
sonii (4), and Caryophyllia epithecata (2). Differences due
t(j age, the forms of the coralla, states of contraction of the
polyps, conditions of feeding, etc., were common, but I found in
no case any in the gross anatomies of the different specimens
of the same species. A regular difference lay in the number of
the mesenteries, which sometimes increase with age, entailing a
corresponding increase in the number of the septa. The mesen-
teries reaching the stomodoeum and possibly the tentacles
acquire their adult number at a very early age, and remain
constant. Throughout the collection either half or all the
septa were entocoelic, and, although it may have been due only
to age, it was a rule that generative organs were present in
the pairs of mesenteries on either sides of all the septa in each
system, save those of the highest cycle, and in them alone.

123

When a polyp extends down the outside of its cup, the
mesentenes and niter-mesenterial spaces continue into its edge-
zone over the top of the corallum, dividing it up into a series
of spaces lying over the costae. The tentacles first appear
over the top edges of the primary septa, and continue to in-
crease with the growth of the other septa, until they have
attained their adult number. Thin-threaded nematocysts in
batteries cover all except in Paracyathiis parvulus, which has
the whole epithelium of the tentacles evenly set with nemato-
cysts.

The Stomodoeum in all forms shows ridges over the
attachments of the mesenteries, the filaments of which merge
into them, and evidently belong likewise to the ectoderm. No
definite groove is present, the whole epithelium being usually
ciliated. The lower ends of the filaments commonly form
coiled masses, the acontia ; that they are ever capable of
being ripped off the mesenteries and shot out is doubtful, but
Icops can and do frequently project into the stomodoeum. The
mesenterial nematocysts, when present, are always of the thick-
threaded type found in Coenopsammia {Willeys Zoo.
Results, p. 357 et seq.). They perhaps kill or paralyse the prey
when it is taken into the stomodoeum, but their size and
enormous number suggest that they must have, or must have
had, some other function.

The body layers in all parts of the polyps have so
nearly the same structure as in F labelluni rubruni
that their description in that form, applies to all. They
only vary in their minute anatomy in the presence or absence
and distribution of the gland cells and nematocysts. Small
thread-cells of the thin and thick-threaded tentacular and
filamentar types are found anywhere in the ectoderms of the
edge-zone, tentacles, peristome, stomodoeum, and filaments,
though the large ones of each kind are restricted to their proper
positions. But nematocysts also occur in the endoderm towards
the tops of the septa. In CaryophyUia epithecata they are not
so degenerate as in other forms, and clearly belong to the usual
filamentar type. They are evidently not now functional in any
way, though their arrangement and regularity is such that one
is led to consider that they must have been so at one time. In
this connection one may recall their abundance in the young
post-larval stages of Flabellum rubrum. Indeed, it seems not
unlikely that the method of development described in that
form may be in reality the primitive one ; the stomodoeum then
would be of quite secondary formation (p. 151, iialics).

Of course, I am well aware that the view set forth above is
absolutely opposed to nil preconceived ideas of the origin of the
stomodoeum and the phylogeny of the Actiniaria. It, however,
serves to explain their anatomy — endodermal nematocysts,
ectodermal mesenterial filaments, massed filaments and true

124

acontia, filamentar nematocysts, ridged stomodoea, etc. — better
than the view derived from the ontogeny of the few types that
have been descnbed. In effect, it seems to me that this is an
excellent instance where natural selection may have acted on
the developing polyp, in which the whole of the cavity of the
body was open to the water, the nematocysts serving for its
protection. With the formation of the stomodoeum in the adult
the main function of the endodermal and filamentar nematocysts
would be gone. It would then obviously be of no small advan-
tage for the formation of the stomodoeum to be hurried on as
early as possible, so that all trace of the original condition would
only remain in the degenerate endodermal nematocysts and in
the filamentar ones, which I cannot consider to have at the pre-
sent time any adequate function to perform.

2. TROCHOCYATHUS RAWSONII.

The polyp extends down outside its cup for about a third of
its height. It has 24 tentacles, the bases of which are confluent,
and each crossed by a pair of mesenteries. The longitudinal
muscles extend into them, and in contraction of the polyp com-
pletely introvert them from the base.

There are 48 mesenteries, septa IV. being exocoelic ; 24
of these reach the stomodoeum, which is ridged over their attach-
ments precisely as in FlabelLum. Their filaments are of the
usual form, but have no massed loops, corresponding to the
acontia ; their lower ends though are free, and may be capable
of some degree of extension. Nematocysts much larger than
in Coenopsammia, and with about 16 coils are present, parti-
cularly in the lower halves of the filaments, which are often
almost choked with them.

The endoderm forms pads under the mesenterial filaments,
but is not especially thickened over the generative organs. Scat-
tered throughout it e\^erywhere, but more numerous over the
muscular bands, are a series of deeply staining bodies. A fe\*
of these, round in shape, deeply and evenly stained, are evidently
fat globules. Others, more granular in appearance, and less
regular in shape, appear to be simply food patches in the
Plasmodium, which forms the endoderm. Nematocysts are
present in patches on the septal sides ; they are of about the
same size as the filamentar nematocysts, and almost in the same
condition as in Flabellum, but with the thread less distinct.
Testes occur in all the mesenteries. They are only one follicle
thick, but in the larger mesenteries form great masses, deep
down in the calicle, almost completely filling up the interseptaJ
spaces.

125

In one of the specimens examined, two pelagic fish ova were
found, one tightly clasped by the mesenterial filaments, the
other in the lower half of the stomodoeum.

3. HETEROCYATHUS AEQUICOSTATUS.

The polyp completely covers the whole corallum, only stopping
short around the aperture of the Aspidosiphon chamber. Canals
from the intermesenterial chambers extend downwards over the
costae, only becoming obliterated towards the centre of the base.
The tentacles number 24, and are connected at their base,
each with a pair of mesenteries, the muscles of which serve in
contraction to introvert them. In the ectoderm mucous cells
are very common, and there is also a deposit of brown pigment,
showing that the polyps were probably coloured in the living
state.

The mesenteries number 48, of which 12 reach the stomo-
doeum. Their lines of attachment are marked by ridges, which
are similar to the filaments in structure, merging into them be-
low. The latter end in coiled masses, the acontia, which, how-
ever, scarcely look as if they were capable of being extruded.
Nematocysts of the usual type occur sparingly, except in the
massed loops, where they are very numerous.

The endoderm forms large, rather granular pads at the
base of the filaments, which in the contracted specimens almost
smk into them. The gastro-vascular cavity of one specimen
was crowded with small pieces of green weed, many pieces of
which could be seen in these pads, apparently ingested whole.
There was no trace of any endodermal nematocysts. Genera-
tive organs occur in all the mesenteries. In the larger
ones, instead of forming a single layer of sperm masses or ova,
they may be three or more follicles or ova thick between the
two endoderm layers of the mesenteries. There are also in
each ovary several ova from the free edge towards the attached
base of the mesentery, not a single line, as is common in other
genera. The specimens of dredging I. of type i (6 examined)
were all female, and of type 2 (4 examined) all male. The
young are either male or female, and specimens from other
dredgings, independent of their type, are male or female.

4. CARYOPHYLLIA EPITHECATA.

The polyp only extends down the outside of the corallum
for 1-3 mm. It has four cycles of tentacles over the septa,
decreasing in size from I. to IV. Over septa IV. they are very

D

126

small, some mere little raised areas of the wall, packed with bat-
teries of nematocysts of the usual type. When the polyp con-
tracts, none of the tentacles seem to be capable of being intro-
verted, though I., 11., and III. are a little pushed m round tlieir
bases. Their protection is rather afforded by the folding in-
wards of the edge of the disc of the polyp, brought about by
the contraction of a sphincter muscle. The anatomy, too,
shows that there can scarcely be any introversion. Indeed, it
would seem impossible, as the structureless lamella both of the
tentacles and disc is remarkably thick Emd firm. The mesen-
teries, too, are attached between the tentacles, and do not seem
to send up muscular fibres into them.

There are 48 mesenteries, septa IV. being exocoelic*
I'welve are attached to the base of the stomodoeum, into the
wall of which their filaments pass, continuing up as ridges for
about a half of its length ; 1 2 also extend down from its mouth
for about a third of its depth, but have no corresponding thick-
enings. Filaments are found on all the mesenteries, the edges
of which are looped, but with no acontial portions. Nemato-
cysts only occur sparingly, save in the lower halves of the fila-
ments. In none is the black thread, which commonly occurs,
distinct, and few have any eversible basal portion at all con-
spicuous. Their appearance may be due to the coral having
recently been feeding, but is rather, I think, caused by a real
degeneracy in them as stinging cells.

The endoderm is thin, but presents a smooth, even surface,
as the inequalities of the corallum are filled in by thickenings
in the structureless lamella Food masses are found through-
out it, but are more conspicuous in the pads at the bases of the
filaments. The sides of the septa are in places packed with
nematocysts, lying parallel to their surfaces, and pointing to-
wards the mouth of the calicle. They are similar to those of
the filaments, but have no trace of any eversible bases and are
evidently degenerate. The testes are of the usual type, their
follicles presenting all stages of development to be flaggellated
spermatoza.

The corallum is much bored into, especially below the pali
and columella, by the filaments of Achy la.

Note on CARYOPHYLLIA SMITHL— I am indebted to
Mr. R. C. Punnett for some well-preserved specimens of this
form from 80- 1 00 fathoms, off Bergen, Norway. As compared
with C. cfithecata the polyp extends much further down the
outside of the corallum. The tentacles number 24, alternately
larger and smaller, but where septa V. are present additional
ones occur over IV. The structureless lamella is half to a

* In one specimen there were 50 mesenteries, only 23 septa of IV. being
exocoelic.

127

quarter as thick as in the last species. There is no sphincter
muscle and the tentacles are introversible from the bone.

There are such a number of mesenteries (62 in one specimen)
that the highest septa in each system are exocoelic. There are
24 mesenteries reaching the stomodoeum, which has a similar
number of ridges over their attachments, passing below into
their filaments. The nematocysts are arranged in the latter as
in the last species, but are quite well-developed.

The endoderm is similar to the last species. No definite
nematocysts exist, but there are a number of granular, oval
masses in patches on the septa sides, which appear to represent
them in a still further reduced state. Boring algae are not
present.

5. DUNCANIA CAPENSIS.

The polyp is seated on the cup, and does not extend in any
way outside its wall. When expanded, it evidently stands up for
some considerable distance above its wall, but in contraction this
part is by means of a few sphincter muscular fibres drawn over
the peristome and tentacles. These latter number 12, the
bases of each of which are crossed by a pair of mesenteries.
Underneath their epithelium the structureless lamella is rela-
tively thick and rather vacuolated, and in correspondence the
tentacles only appear to be introversible at their bases.

The single specimen examined has 68 mesenteries, 12 pairs
on either side of septa I. and II., equal in size and alone joined
to the stomodoeum, which is thickened over their attachments
for the whole of its length, 12 pairs on either side of septa III.
and 10 pairs enclosing the 10 septa of cycle IV. ; the septa hence
are all entocoelic. Filaments occur on all except one rather
rudimentary pair of cycle IV. The lower ends of all are much
convoluted, forming acontia, and crowded with nematocysts of
the usual type.

The endoderm and testes are the same as in other species,
but the former has no nematocysts. The testes are formed by
a single thickness of follicles, and are situated behind and above
the acontia They occur in the mesenteries of cycles I., II. and
III., but only on the largest of the latter. This distribution may
be due to age — the coral was only 1 1 " 5 mm. across the calicle —
but some of the mesenteries of cycle IV. are larger than those
of III., which have testes, and it seems not improbable that the
mesenteries of cycle IV. never have any generative organs.

128

EXPLANATION OF PLATES.

Plate I.

Fig. I. Spliciiotroclins i^ilclin'sti.
((() View of calicle X 3.

(/)) Edge of the calicle, showing the exsert ends of tlie Septa,
(^c) The ridges on the sides of the corallum.
(</) Side view of corallum in (rr) X I2.
(e — g) Side views of other coralla X i-

Fig. 2. Trochocyathus niwsonii.
(d) View of calicle X 2.
(6) Side view of corallum of (d) X i.

P'iG. 3. Cai'vophyllia cpithccata.
(a) View of calicle X 2.
(fo) Side view of corallum of [a] X i-
(c and d) Side views of ether coralla X i-

Fig. 4. Caryophyllia capensis.

(a') View of calicle X 2^.

(ft) Side view of corallum of (u) X i

(c) View of another calicle X 2.

(d) Side view of corallum of (c) X i-

Fig. 5. Ccnitotivclins Joliiisoiii.

((7) View of calicle X 2.

(/)) Side view of corallum of (a) X i.

(c) Side view of another corallum X i-

ViG. 6. Dn Ilea II in Ctipcnsis.

(tr) View of calicle X 2.

(ft) Side view of corallum of (a) X i.

(c) Side view of upper edges of three septa of cycles I., III. and IV.

129

Plate II.

Figs. A — E. Diagrams to show the development of the septa, pali and
coUunella in TrocJiocyatliiis raivsonii. The coralla are represented all of the
same size, whereas they are very different, the figures showing the changes with
increase in size. The rods of the columella are hollow throughout, the pali of
cycles I. and II. shaded, and of III. black. The cycles of septa are marked
with Arabic numerals.

Figs. F — K. Side views of various specimens of Trochocyathits rawsonii
(XI.) to show the variation in growth and appearance of the coralla. The lines
of growth show well in K.

Fig. L — Diagram of the calicle of Trochocyathiis ciiidiciilafits.

Fig. M — Diagram of the calicle of Ceratrodms johnsoni.

Plate III.

Coralla of Hetcrocyathus aequicostatus,

1—8. Type i from dredging I. X i^.

9 — II. Type i from dredging I. X i.
12 — 19. Type 2 from dredging I. X li-
20—21. Type 2 from dredging I. X i.
22 — 29. Young from dredging I. X if-
30 — 32. Young from dredging I. X i-
33 — Specimen from dredging VII. X li-
34 — The same X i-

35 — 38. Specimen from dredging I. showing fission or possibly budding.
39 — Specimen from dredging V. X 15.
40 — 43. Other specimens from dredging V. seated on various shells X i.

(Published 20th Sept., 1904.)

E

Marine Inveobigabions of Soubh Africa.

Plabel.

■W\aV\\vV\\

^\,,^

lb

m

A mA /hi

llW\V">>^\.K

//"/lU\^^f^

\

i ■' ■'S3 v* f. f-

3a

,j.i-fei-<^-

STANLEY GARDINER- TURBINOLIDAE

E. Wilson del.et iith.,Cam'bridqe.

Marine Investigations of South Africa.

Plate II.

F.

STANLEY GARDINER-TURBINOLIDAE.

Marine Investigations of South Africa.

Plate III.

STANLEY GARDINER-TURBINOLIDAE.

THE DEVELOPMENT

OP

SOUTH AFRICAN FISHES.

PART 11.

BY
J. D. F. GILCHRIST, M.A., B.Sc, Ph.D., C.M.Z.S.

Government Biologist to the Colony of the Cape of Good Hope.

The following- observations on the development of South
African fishes are in continuation of those published at an
earlier date (Jllarine Investigations in South Africa, Vol. 11,
page i8i).

The material was procured by the Government steamer,
chiefly while engaged in deep waters off Cape Point, and in
occasional tow nettings in False Bay. Whenever possible the
eggs were brought in alive, and their development observed.
Many eggs and larvae were simply preserv^ed in formalin, and
an attempt was made to sort these out into what might be con-
sidered to belong to different species. In most cases, however,
this turned out to be unsatisfactory, and to avoid confusion the
observations made are heldMn reserve until living specimens
can be procured. A few well-characterised eggs, however, are
described from preserved specimens.

In Part I. of these observations the ^^g of a fish named pro-
\ isionally " Species I." was described, along with the larva
hatched from it. It is one of two kinds of eggs that were found
attached to shells and stones. This now appears to be the egg-
of a species of fish known to fishermen as " KHp-zuiger," or
" Sucker-fish," a name applied to fish having a well-developed
sucker, by means of which they can adhere firmly to rocks, etc.
They are small and of no commercial value, so that whatever
injury may be done to the eo-g by nets or trawls is not likely to
have any direct effect on the fishing industry. This identifica-
tion was made possible by the finding of several young fish in

132

different stages of development in a tow-net attached to the
beam trawl in about 20 fathoms in False Bay, off the Roman
Kock, 13th October, 1898.

The smallest of these closely resembled the larvae of Species
I., and the largest showed the widely separated ventrals with
adhesive apparatus between them and the short dorsal fui situated
on the tail, characteristic of the family Gobiesocidae. The
ethers represented intermediate stages.

The only member of this family recorded from the Cape is
Chorisochisnuis dcntcx, but as the posterior half of the ventral
disc in these young forms has a free margin, they cannot be re-
garded as belonging to this fish, and they are not sufficiently far
advanced to allow of more than a reference to the family with
certainty. They have a gill cover free from the isthmus, and
therefore do not belong to a species (apparently new) of Lepado-
gaster recently found in False Bay. In view of the characteristic
oval eggs of this last-named genus,* a number of eggs were re-
examined. In Part I. they were described as about one milli-
metre in diameter. A number of measurements show that none
of them are perfectly circular, though some are ver}' nearly so,
C'ne being ro6 x "98 mm., while others varied from about this
to I '37 X -gj. In most of these eggs there was also one
oil globule from '\y to "3 mm., and the space between the eggs
was less than in the case of those at first examined. In the
fresh egg there was no evidence of a filamentous fringe round
the basal part of the egg capsule, though some preserved in
formalin showed radial striae on an irregular border.

The identification of the other demersal egg found (Species
11.) has not yet been possible, but several speciniens were
again procured in dredging on rough ground in False Bay in the
month of November. That they belong to a fish of small
dmiensions seems probable, as they have been on more than
one occasion found inside an empty barnacle shell, the opening
of which was small. Plate V, fig. 34, represents such a shell,
(natural size) with one side cut away to show the blue eggs
attached to its inner surface. Fig. 35 represents one of the
eggs enlarged to show the numerous oil globules which occur
towards one side of the egg.

In the following account, the various species dealt with are
arranged according to the number of the oil globules and
diameter of the eggs, as this artificial arrangement has proved
very convenient. For this reason also a key (p. 150) to the e.^gs
and larvae similar to that given in Part I., has been drawn up,
including not only the species here dealt with, but those already
mentioned in Part I. The items taken from the first table are
printed in italics for convenient reference.

* See Holt, Trans. K. Dublin Soc, S. II., IV., iSyi.

In n

The following species are dealt with here : —

(//) Spt'cit's with one oil globule'.

Species XL Arnoglossiis capciisis, Blgr.
Species XII.

Pagnis laiiiariiis, C. & V. (The " Panga ") ...
Mcicninis fas,ciatii^, Giinth —

,, pdinllcliis, Giinth.
Species XIII.
Species XIV.
Species XV.
Species XVI.

(/)) Specie's 'ivitJi several oil globules.
Strouiateus iiiicivchinis, Bonap.

(c) Species with iiKiiiy oil globules.

Syugiuithiis acus. ...

Species XVII.

Cataetyx uiessievi, Giinth. ...

((/) Species ivith no oil globules.

Species XVI 11.

Agriopus spinifer, Smith

Species XIX. ... ... ...

Sconibyesoxsau rus, W'd\b ...
Species XX.
Species XXI.

^age

131

f>

133

}>

134

134
136

136

136

137
138

>>

138

ft

139

>»

139

>>

140

>»

142

>>

143

>>

143

M

144

»l

147

>i

147

(a) Species unth One Oil Globule.

SPECIES XI.

Arnog'lossus capensis, Blgr.

A single i^^-g was procured by tow-net in Buffel's Bay, on the
14th November, 1903. It measured '79 mm. in diameter, and
liad a single oil globule ' 1 1 mm. in diameter. The embryo was
fairly advanced, being about half the circumference of the &%g.
There was then no colouring matter on ^g'g or embryo, but on the
following day, when the embryo was about three-quarters the
circumference of the <i^g, a faint reddish tinge could be seen,
and this, when examined with a higher power, proved to be a
number of small pigment cells scattered over the body from
licad to tail.

134

Hatching took place on the following day, and the larva
showed several distinctive features (^cide Plate V, fig. 36). The
colouring matter, a bright brick red, was arranged in the follow-
ing manner : Patches on snout, crown, orbit, otocyst, middle of
the margin of dorsal fin, along the superior margin of
body to near the tail, in the middle of the bod}', yolk (most
marked behind oil globule), between yolk and rectum, and on
inferior margin of tiie body. The most distinct were the
patches above and below the body and extending on to dorsal
and ventral fins. These were situated between the yolk and
the extremity of the tail, but nearer the former.

Other characteristic features of the larva, were the tuberculate
appearance of the dorsal and ventral fins and their margins.
Shght markings, apparently of the same nature, appeared on
the posterior region of the yolk, the substance of which was
homogeneous. The oil globule, which was situated posteriorly,
had a purplish tinge on its periphery. The rectum was anterior.
The notochord was mostly unicolumnar.

The relatively small size of the ^^^ and the characteristics of
the larva agree, on the whole, so well with the description and
figure of Arjio gloss US'-' that I have little hesitation in provision-
ally identifying the parent as Arnoglossus capcnsis, Blgr., a flat
fish recently described from False Bay.t

On 24th November, 1903, an egg was procured in False Bay
•7 mm. in diameter and having seven small oil globules in a
group and about six branching cells on the yolk near embryo.
On the 27th this produced a larva apparently identical with the
above. The seven oil globules had fused into one, " 1 1 mm. in
diameter. The larva was 2'8 mm. in length, had a few black
spots on body and some black branching cells on yolk (not
observed in the first specimen). The oil globule had a yellowish
tinge, and the general colouring of the body and fins was the same
as in the first specimen, witli the same general arrangement.
The fins were tuberculate in the same way.

SPECIES XII.

On several occasions fish eggs were procured in False Bay
in the month of November, having a diameter of '89 mm. with
an oil globule of "15 mm. This agrees so closely with the eggs
of Chrysophj-ys globiceps {vide Marine Investigations in South

* See Holt : Recherches sur la Reproduction des Poissons osseaux, Ann.
Mus. d'Hist. Nat. de Marseille Zoologie, Tome V., 1899.

t The Flat Fishes of Cape Colony, by G. A. Boulenger, F.R.S., "Marine
Investigations in South Africa," Vol. I.

135

Africa, \o\. II, p. 187) that they were at first supposed to belong
to this fish. The larvae also were very similar in form, but were
invariably devoid of all colouring matter. As the eggs were pro-
cured in abundance on several occasions, and in no case did the
larvae show any trace of pigment, it seems advisable for the
present to regard it as belonging to a different species.

Pagrus laniarius, G. and V. (Panga)

' Although this fish occurs in abundance in the trawl, it is some-
what remarkable that only two ripe females and no ripe males
have been as yet found among the many hundreds examined.
These two specimens were procured on the Qth and i6th March,
1903, about 7 miles off Cape Infanta. The eggs (in formalin)
measured '93 to ro2 mm. in diameter, and the old globule 'ig
mm.

A number of ripe eggs of this fish were procured (6th June,
1904) by one of the Kalk Bay fishermen, who now show consider-
able interest in these investigations. They were larger than the
above, being from ro6 to i"i mm., with an oil globule of '2 mm.

Macrurus fasciatus, Guntli.

Many specimens of Macrurits fasciatus have on several occa-
sions been procured in trawling in abovit 100 fathoms, but of
these there was only one perfectly ripe female. This was on
the 28th October, 14 miles off Cape Point, when 684 specimens
were brought up in the trawl. None of these were ripe males,
and the eggs could not therefore be fertilized.

The eggs (PI. VI, fig. ^'j) when procured from the female were
clear, homogeneous, and floated freely on the surface of the water
in which they were placed. They were of a fairly uniform size,
ranging from ri5 to roo mm. in diameter, the oil globule, which
was yellow or red in colour, being from '29 to '27 mm. The
vitelline membrane was thick (about '02 mm.), and marked in
a very distinctive manner. In some cases the markings appeared
to be separate dots onh', but these were seen in man}' to be
connected by fine hyaline lines, so as to form polygonal mark-
ings on the surface of the egg. The dots themselves appeared
on a side view to be small connecting pillars between the
outer and inner surface of the vitelline membrane (fig. 38).

During the time the " Pieter Faure " was engaged in deep-
v.ater work off Cape Point none of the surface tow-nettings
contained any eggs at all corresponding to these, but on four

136

occasions one or two eggs procured in the tow-net attached to
the beam-trawl exhibited all the characteristics described above.
The polygonal markings were very distinct, as well as the con-
necting columns. One egg contained an embryo fairly well
advanced, and this, when dissected out, showed many black
branching cells over the body from head to tail. The rectum
was apparently anterior.

The eggs were procured at the following times and places : —

(i) August 20th, 1903. Cape Point, N. 81° E., 32 miles, 460
fathoms. One procured. Refer. No. 17385.

(2) August 27th, 1903. Cape Point, N.E. by E. | E., 8 miles,

91 fathoms. One procured. Refer. No. 17555.

(3) September 27th, 1903. Cape Point, N.E. ^ N., 46 miles,

760 fathoms. One procured. Refer. No. 181 34.

(4) September 27th, 1903. Cape Point, N.E. i N., 47 miles,

700-1,000 fathoms. Two procured. Refer. No. 18097.
"(5) July 8th, 1903. Cape Point, N.E. by E., 36 miles. 650-700
fathoms. One procured. Refer. No. 16799.

On these occasions a moderately fine net was used on the
surface, while a coarse net and a moderately fine net were
attached to the beam of the shrnnp trawl, the eggs being found
072 /v in the last.

The measurements of these eggs were as follows : — •
(^1) Diameter of egg, ri5 mm.; of oil globule, "3 mm.

\- y » ») i> » » '» " »

\}i) )> V 121 „ ,, ,, 25 ,,

(contained embryo).

(4) .. » I '07 .. » .' '3 mm-

(contained embryo).

The manner m which they were procured seems, on the whole,
to point to the fact that their natural place of occurrence is at
or near the bottom of the sea. As against this supposition,
however, we have to bear in mind that the eggs when procured
from the ripe female floated, and that the bottom net was not
a closing one. Neither of these considerations are conclusive
however, as the fish from which the eggs were procured had been
brought up from a depth at which there must have been great
pressure, and the eggs under these new conditions of diminished
pressure might float. Again, as against the second objection
we have to consider that on none of the four occasions on which
the eggs were procured were any found in the surface tow-net,
which was in use at the same time. The question can only be
satisfactorily settled by the use of a closing net.

Contained 2t->-'?o Small Oil Globules, but otherwise like the others.

137
Macrurus parallelus, Gunth.

Of 145 samples of this fish procured on the 17th September,
from 310-560 fathoms, 39 miles off Cape Point, one ripe female
was found. On all other occasions when this fish was procured
none proved to be ripe. The eggs were clear, not vesiculated,
and floated freely at the surface of the water in which they were
placed.

They vary from r32 to I "3 mnx in diameter, the single oil
globule measuring from "31 to '26 mm. The surface of the
egg shows the same polygonal markings, with pillars at the
angles, observed in the case of 71/. fasciatus, but it is readily
distinguished by its larger size.

None of these eggs were found in tow-nettings.

SPECIES XIII.

On the gth July, 1903, a few eggs were procured in a surface
tow-net, Cape Point. N.E. by E., 36 miles. They measured
i"j mm. in diameter, and had a single oil globule of '},ii mm.
The yolk was vesiculated.

These produced transparent colourless larva, difficult to see
even in a tube of clear water. They were about 4' 67 mm. in
length. The notochord was multicolumnar, the rectum posterior,
and oil globule posterior. The yolk had well-marked vesicula-
tions. The newly-hatched larva had a very marked downward
curvature of the body anteriorly. The dorsal and anal fin were
comparatively low, and the former commenced behind the middle
of the yolk (Plate VII, fig. 43).

SPECIES XIV.

Seven eggs were found in a surface tow-netting taken on the
Qth July, 1903, Cape Pomt, bearing N. by E., 36 miles. They
were all \"j mm. in diameter, and had a single oil globule of "38
mm. The volk was vesiculated. The more advanced showed a
well-developed embryo, with dark pigment cells. On the 12th,
all the eggs had sunk to the bottom of the jar, and on the fol-
following day some had hatched out (temp. 60° "5 Fahr.).

The lar\'a proved to be very elongate (6' 7 mm.) and provided
with characteristic black tufts, which did not disappear in for-
malin after being kept for some months. This characteristic

138

may be useful for identification purposes. There was no yellow
pigment present, and the black pigment (Plate VII, fig. 41) was
arranged as follows: Branching cells on snout, crown, behind
eye, over part of yolk sac and the body adjacent to it ; dense
tufts, three above and two below the body posterior to the
rectum. The oil globule was covered by a network of black
cells. The notochord was multicolumnar and the anus anterior.

SPECIES XV.

Eight ova were procured in a surface tow-net on the 2nd July,
1903, Cape Pohit, bearing N.E. by E., 38 miles. The}- were
I '87 mm. in diameter, with an oil globule from "38 to '42 mm.
They contained embryos fairly advanced, with tail slightly free,
and were about half the circumference of the egg. The yolk
was transparent and homogeneous, except for some clear dots
over its surface, which may not however be constant. The
embryo at this stage showed the following pigment patches of
a greenish yellow colour : a patch behind each eye, a fainter
tract along each side of the body, on the caudal region and
oil globule. On the 26th, all the eggs had sunk to the bottom
of the jar. They lived till the 31st, but were not in a healthy
condition, and did not hatch out. Another lot, about 60 in
number, was procured on the 17th September, Cape Point, bear-
ing N.E. (approx.) 40 miles, and hatched out satisfactorily. The
lar\^a (Plate VII, fig. 42) is long (4' 8 mm.), the rectum anterior,
and the notochord multicolumnar. The pigment is arranged
as follows : Yellow patches round eyes, on hinder part of yolk
sac and oil globule, yellow tufts above and below body and
extending on to fins. The arrangement of these tufts varied
somewhat in the different larvae. The following were constant :
One above the body a little behind the rectum, one below body
about as far again behind it, another between the second and the
end of the tail, and two at the end of the body. The variation
occurred in the addition of another yellow tuft, sometimes above
and sometimes below the body, near the third above mentioned.
The eggs were procured in surface tow nettings as follows : —

Kef. \o.

Localitv.

Procured.

Date.

17161 ! Cape Point, N.E. by E. 38 miles. S July 23rd, 1903.

17230 : Cape Point, E. i N. 25 miles. 5 July 30th. 1903.

18053 I Cape Point, X.E. appro.x. 40 miles i 60 Sept. i/tli, 1903.

17414 j Cape P'jint, E. i N. 34.I miles. t 4 .\u.u. 20th, 1903.

139
SPECIES XVIr

This species is very distinctly characterised by having a thick
envelope of a clear gelatinous substance, with polygonal facets on
its surface. This substance is apparently homogeneous, and is
transparent, so that the egg proper appears as a dark sphere
surrounded by a sort of halo, having however occasionally a
few dark spots outside its periphery. When viewed by trans-
mitted light (Plate VI, fig. 39), the surface is seen to be divided
up into polygonal markings, usually six-sided. When viewed
from the side near the periphery of the egg, the sides of the poly-
gonal facets were seen to be cuirved and raised into ridges having
a smooth, sharp, and clearly-defined edge. This becomes very
evident when the egg is removed from the water and viewed as
an opaque object (fig. 40).

The total diameter, including envelope, varies from rgi to i"7
mm., the egg proper from i to ro6 mm. A single oil globule
is present, varying from "25 to '21 mm. in diameter, and is
usually greenish in colour, but sometimes clear and colourless.

Some of the eggs showed the early stages of embryos, but
none far enough advanced to show any specific characteristic.
All were examined and measured in 2 per cent, formalin.

The eggs were procured as follows :■ —

Ref.
No.

No.
Pro-
cured.

Locality.

Means ui
Capture.

Depth.

Date.

17555

18

17863

4

17838

S5

Cape Point, N.E. by
E. I E. 8 miles.

Cape Point, N. 50°
E. 34A miles.

Cape Point, E. 'n N.
20 miles.

Moderately tine
net on beam.

Coarse net on
beam .

Modei-ately tine
net on surface.

380-2
3^0-475

Any. 27th, 1903.
Sept. nth, 1903.

Note. — C<3mpare somewhat similar eggs described by Hensen and Kattaele.

(d) Species with Several Oil Globules.
Stromateus microchirus, Bonap.

A ripe male and female of this fish were procured in Mossel
Bay, igth February, 1903. Fertilized eggs were secured and
kept till hatching took place, 34 hours later No facilities for
microscopic examination were available, and the following notes
are from material preserved in formalin.

The eggs vary from '81 to '85 mm. in diameter. There are
several oil gloVjules, one to five of which are of a dark colour,

140

with indefinite outline, the others being smaller globules of the
normal appearance. The former seem to be oil globules with
dark pigment cells, as they could be dissolved, leaving branching
pigment cells. The substance of other globules entirely dis-
appeared in dissolving. The larger globule generally occupied
a posterior or ventral position in the yolk.

The larva (Plate \TI, fig. 44) at the time of hatching is short
and stumpy, being about i\ milhmetres from snout to tail and
about one-half this in greatest depth. The posterior border of
the yolk is near the middle of the total length.

(c) Species with Many Oil Globules.

Syng-nathus acus, Linn.

A male of this fish was found in False Bay on the "th August.

In the pouch were rows of eggs of a }'ellovvish red colour.
They measured about \'2^ mm. in diameter, and contained ma'Y
small oil globules, var}'ing considerably in size.

SPECIES XVII.

A single Qgg (Plate VIII, fig. 45) was procured on 24th ISlovem-
ber, in False Bay, in a surface tow-netting, and measured I'S/
mm. in diameter. It contained many small oil globules, which
occurred in groups and singly throughout its substance. On
the following day the embryo was a little less than half the
circumference of the &gg and showed very characteristic pig-
m.ent, quite visible to the naked eye as an orange red patch on
the lower side of the floating ^^^. This was seen under the
microscope to be a mass of pigment cells, dark red by trans-
mitted light and light orange by reflected light. The whole
embryo was covered by this pigment, which extended on to the
yolk, gradually becoming fainter till only some scattered branch-
ing cells appeared on the surface of the yolk furthest from the
embrvo. These were of a dark and fainter colour.

Hatching took place on the 27th November.

In the larva the rectum is anterior and the notochord appeared
to be multicolumnar. though this could not be made out with
certainty owing to the presence of the colouring matter.

141

Cataetyx messieri, Gunth.

On the 27th August, 1903, a large specimen of Cataetyx mes-
sieri, Gi'inth., was found, Cape Point, bearing N. 80° E., 32 miles.
The depth was 460 fathoms with a bottom of green mud. It
was relatively large for a deep sea fish, being about 2 feet in
length, and much larger than the type and only specimen hitherto
found. The latter, obtained by the " Challenger " in Messier
Straits, was 8 inches in length. A conspicuous feature in this fish
not seen in the smaller and probably immature " Challenger "
specimen was an anal depression covered by a flap, on the inner
side of which was situated a papilla, the whole being ap-
parently a copulatory organ of some kind.

A little later, on the 17th September, Cape Point N.E., 40
mile (approx.), in 560-700 fathoms, bottom green mud, we were
fortunate in procuring another specimen about the same size,
but a female. The ovaries were distended with ova, which were
readily discharged on applying a little pressure. They were of
a warm reddish colour, which at once suggested the peculiar
red characteristic of many deep sea animals, and which has been
seen several times in the groups of the Alcyonaria, Nudibranchs
and Crustacea in deep waters in this same region. The egg
(PI. Vni, fig. 46) contained many small oil globules, some of a
bright red colour, others more of an amber tint, which was also
the colour of a single relatively large oil globule which was
present in each egg. In some of them when fresh a small patch
of what appeared to be protoplasmic matter was observed along-
side of this large globule, and this was more apparent when' the
colouring matter in course of time became quite bleached out
by the light, but there was no indication of segmentation or
embiyos in any of the several hundred eggs examined. The
two large ovaries were each about 1 15 mm. in length and 35 mm.
in diameter. They were enclosed in a tough capsule fully dis-
tended with ova of the same bright colour as those which had
been pressed out. The ova ^vere produced from a series of
transverse leaf-like expansions hanging from the roof of the
cavity, their free margins extending nearly to its floor. They
were closely packed in between these expansions and filled the
remaining space under them where they were more loosely
packed. From a rough calculation, there must have been over
30,000 in each ovary. It appeared at first that as the eggs were
evidently so well developed and flowed so readily from the fish
on pressure, they were the ordinary unfertilised eggs
found in most fishes, only that they had the colouring and other
characteristics of some demersal eggs. A more careful search,
however, amongst the ova revealed eight larvae in a fairly ad-
vanced stage of development. They were not very difficult to
see, though enveloped in a mass of eggs, as the black pigment

142

of the eyes formed a marked contrast. One was found in the
right ovary and the others in the left, and they all occurred on
the peripheral region of the mass of eggs on its ventral aspect.
They were entirely removed from the ova-producing lamella, and
had no connection with these or with the walls of the ovary.
1 he eggs in their immediate neighbourhood were embedded in
a sort of mucous substance, which coagulated in formalin, and
which contained scattered throughout it isolated oil globules.
In nearly every case the larvae were found coiled up in a peculiar
manner, with a strong flexure near the middle of the body and
another in the same direction towards the end of the tail. In
one instance, in breaking up the mass of eggs surrounding in
embryo, that is carefully detaching eggs and embryo from the
surrounding mucous, there was an escape of numerous oil
globules, which were also found in the open mouth and throat
of the embryo, thoui^h none of the e.^.^'s or larvae seemed in-
jured. In another case (PI. xi., fig. 58) the embryo was coiled
round a white substance, which at first seemed to be a mass of
mucous and oil globules, except that it had numerous black pig-
ment spots. Further examination showed that one end of this
mass was enveloped in the mouth of the larva, and this on being
withdrawn (Fig. 58a) appeared to be the tail of another and
smaller larva. On some of the surrounding mucous being re-
moved this proved to be the case. Fig. 58 represents the whole
in situ after the tail had been withdrawn from the mouth. The
rest of the body of tiie smaller embryo was coiled up in the form
of a figure 8, the abdomenal and head region being somewhat
broken up.

The largest larva was found lying on the floor of the ovarian
wall, that is between the mass of eggs and the tough capsule, and
quite free from ova and mucous. " It was 10 mm. in length. The
pectorals were well developed, and a few rudimentary ra}'S ap-
peared in theni, and in the dorsal fin at its commencement over the
head (see PI. XI, fig. 57). The body and fins were covered with
black branching cells These were most marked in the head re-
gion, being somewhat more sparsely scattered on the body. They
formed a border to the dorsal fin and a less marked one to the
anal ; a group of pigment cells occurred near the posterior ex-
tremity above and below the caudal portion of the bod}'. The
dorsal fin commenced well forward on the crown of the head (in
the adult it is much further back), and continues backwards as
a fairly wide border. The ventral fin behind the vent was also
well developed, and tliere was a slight pre-anal fin. The vent
was approximately medium, but somewhat nearer the head than
the tail, much more so than in the adult. The rectum was well
developed, and the intestine was continued forwaro, and after
making one or two coils over a large liver could be traced to the
oesophagus. The notochord appeared to be multicolumnar.
There were four well-developed branchial arches and a clavicle.

143

1 hope to make out more details by sectioning the material,
which seems sufficiently well preserved for this purpose.

This is not the ftrst time that evidence has been secured that
some deep sea hshes are viviparous. Several species have been
found with anal papillae, presumabh' male sexual organs, and
Alcock has definitely established the fact in his important discov-
ery of embryos in the eggs of Saccogastcr. He has also suggested
that the embr}'os may, when hatched out, procure in some way
means of subsistence from the surrounding ova. The present case
of comparatively large larvae, which from their size and develop-
ment have apparently existed some considerable time in the
ovarian sac after hatching, is a further step in the solution of the
problem, and the fact that in those advanced larvae the yolk
sac was completely absorbed, the intestinal tract vv'as fairly well
developed, and oil globules were found in their mouth and
oesophagus, seems to indicate some confirmation of Alcock's
hypothesis, and, further, that there is no special organ for ab-
sorption of these eggs, but that nourishment is taken in by the
mouth. Before finding the small larva in the coils of another
of much larger size, apparently in the act of devouring it, I was
inclined to believe that the egg capsule, though tough, was in
some way broken down, allowing the contents to escape. This
case, however, seems to indicate that the larger larvae live
on the smaller or those just hatched, the scattered oil globules
found being tnose from the ruptured yolk sac of the younger
larvae. This is quite in keeping witli the fact of there being so
few larvae in proportion to the number of the eggs, and it would
be easy to construct a probable life history of an -embryo from
the time of hatching to birth should it be one of the few that
only can survive to that period.

(d) Species zvith No Oil Globules.

SPECIES XVIIL

On the 14th November, 1903, seven eggs, characterised by their
small size (•76-72 mm.) were procured in a surface tow-net in
False Bay. No oil globules were present. There were embryos
in each, fairly well developed, being a little over half the circum-
ference of the egg. Distinct black dots occurred along the body,
but no yellow pigment, either on embryo or yolk. Kiippffer's
vesicle was present at this stage. The embryos hatched out on
the following da>-. There were a few black dots along the back
of the body, and faint yellow pigment over the body, fins, and
yolk. Clear dots occurred abundantly on fins, especially towards
their margin. The notochord was unicolumnar, the rectum
anterior, and the total length 1-74 mm. Fig. 47 is a drawing

144

representing this larva on the day following (the i6th). and
^S 48 (from photograph) on the i8th, when the pigment
appeared somewhat differently arranged, viz., a diffused \-ellow
(m which the hidividual pigment cells were indistinct) occurring
on head and body to a point half-way between the rectum and
the end of the tail. Here the yellow pigment expanded into a
band, traversing the dorsal and ventral fins. Yellow pigment
also occurred on the anterior part of the yolk and on the rectum
down to anus. Black branching cells appeared extending from
the margin of the body above and below into the adjacent fins.
■^^1 ^SS' which seems to be the same, was found in a surface
tow-net on the 17th September, 25 miles off Cape Point. It was
about '/6 mm. in diameter, and had no oil globule.

Agriopiis spinifer, Smith.

Some time ago the eggs of Agriopus were procured (znde
J\Iarinc Investigations, Vol. II., p. 189), but fertilization was not
effected. ]\Iore recently a ripe male and female were found in
the same haul, and fertilized eggs were secured (Buftel's Bay,
25th November, 1903). The eggs varied from r83 to i"J^ mm.
in diameter."* Seven hours after fertilization the germinal disc
showed about 30 to 40 cells, and on the following morning, or
24 hours after fertilization, the blastopore was about three-
quarters over the e^^ (Plate IX, fig. 49). The corrugated
appearance noted in the unfertilized eggs formerly presen'ed
were readily seen at this and all subsequent stages, and seems to
be a characteristic of the &^g. Development seemed to be pro-
cteding quite normally, and on the 27th the length of the embryo
was about half the circumference of the o.^^, at which stage
faint branching pigment cells of a dark colour began to appear
all over the yolk. On the following day the embryo had
increased in length, and the heart, otocyst, and rudiments of the
pectoral fin were evident. On the 29th, however, all the embryos
v.ere found to be dead or dying. One was dissected out, and is
represented in fig. 50. Dark branching pigment cells occurred
over the yolk, along the head and body, and on the pectoral fin.
1 lie notochord was multicolumnar and the rectum anterior.

SPECIES XIX.

On one or two occasions a few opaque spheres were found in
tow-nettings preserved in formahn. They were of a dull white
colour, minutely vesiculated, but without any signs of oil globule
or embryos, and there was therefore some doubt as to their being

* .\s there was some doubt as to the maturity of the eggs in tiie first case
these dimensions are substituted in the key on page 152 \ox those fir.st given.

•145

iish-eggs. About half a dozen of these were found, however,
in a fresh surface tow-netting, on the iith September. 1903,
from about 34I miles off Cape Point (Cape Point bearmg N.
5c'-' E.). They measured from rS/ to i'/2 mm. in diameter, and
some contained an embryo a little less than half the circumfer-
ence of the egg. Yellow pigment spots could be seen on the
embryo from head to tail. Unfortunately, these did not hatch
out, in spite of special care. Similar eggs were again procured
on the 27th September, 46 miles off Cape Point, but the}' again
did not survive till hatching.

Some time afterwards living larvae were procured in a surface
tow-net 13 miles off Cape Point, on the 30th October, 1903, with
the yolk opaque and vesiculated as in the above eggs, but with-
out any trace of pigment on the embr}'o. The yolk was drawn
out and somewhat rectangular in shape, and the rectum was
anterior. The total length was 2*04 mm.

Scombresox saurus, Walb.

The diameter of 12 eggs varied from 2"76 to 2'i mm. There
was no oil globule, and the yolk was clear and not vesiculate. All
showed a sprinkling of minute dark dots over the surface. The
eggs seem to be of a particularly hardy nature, as, although some
sank to the bottom of the jar owing probably to the collection of
particles on them, and frequently became coated with a white
substance apparently of fungoid growth, yet after clearing with a
camel's hair brush they floated fairly well, and ultimately
developed as the others. One procured on the iith Sep-
tember, Cape Point bearing N. 50° E., distant 34^ miles,
showed the embryo fairly well developed, but the tail not free.
On the 1 6th, the embryo was further advanced, being about one-
half the circumference of the egg and with the tail partially
free. There was little colour in the embryo, there being only
a few dark dots scattered sparsely over the body and a few
stellate (mostly 3 -rayed) pigment spots on the surface of the yolk,
immediately adjacent to the body of the embryo, as shown in
Plate X, fig. 53 (from an egg preserved in formalin). At this
stage an active circulation, a characteristic feature of this
pelagic egg, appeared. The heart beat 1 1 2 in a minute. A single
afferent vessel could be perceived running along the ventral
surface of the egg. The course of this vessel was in a straight
line from the caudal region to the heart. By focussing deeper
for the low-er side of the egg till the dark spots on embryo and
yolk become visible, the following efferent vessels become
visible : One on each side, starting from the region of the rudi-
ments of the pectoral fin and proceeding at right angles to the
body over the yolk till about half-way to the periphery, when
they each turn abruptly to the head region and return on either

146

side to the heart. Another efferent vessel may be seen, though
less distinctly, proceeding along the body to the caudal region,
where it apparently joins the afferent vessel first mentioned.
On the day following, the pectoral fins were very distinctly
visible behind the lateral efferent vessels, and the day after that
the circulating system was much more developed, the whole sur-
face of the yolk showing a network of vessels through which the
blood corpuscles could be seen coursing rapidly.

The embryo remained without much further change for fourteen
days. About two days before hatching a change was noted in the
pigment. The orbits become of a dark blue and a group of cells
of the same colour appears on the body near the pectoral fins,
which at this stage are about half the length of the head. In
addition to this, small blue specks appear along the whole
length of the body. Very little movement is observed in the
embryo before hatching, with the exception of the free pectoral
fin. (that not appressed against the egg capsule), which is in
almost constant motion. In the only case in which the actual
hatching was observed there was no rupture of the shell, but a
sort of scaling-off of part of the capsule. At the time of hatch-
ing, the embryo was about one and a quarter the circumference
of the egg.

The newly-hatched larva (Plate X, fig. 54) is as much charac-
terised by the presence of distinctive colouring as it had been
previously by its absence. With the exception of the yolk and
fins, it is of a deep blue, which on examination with a low power
may be resolved into dark blue stellate cells, closely packed on
the dorsal aspect, but separated on the ventral region in front
cf the rectum and on a small part of the yolk adjacent to the
body (that part which at an early stage was characterised by
the presence of black stellate cells). The mouth is well deve-
loped and the lower jaw protrudes slightly beyond the upper.
The embryonic ventral fins are well developed ; that anterior to
the anus slightly overlaps the posterior one. The dorsal is not
so deep and begins further back, about half-way between the
tip of the snout and the end of the caudal rays. The rectum is
median, being about half-way between the posterior margin of
the yolk and the end of the body.

The larva is long (one measured 8' 5 mm.), and swims with a
rather slow undulation of the whole body.

After hatching, the absorption of the yolk proceeds at a rapid
rate, and three days later it has disappeared (Plate X, fig. 55).
A marked change has meanwhile occurred in the pigment. It
becomes differentiated into a dark blue band along the back,
extending on the side for about a third of the depth of the body.
The remainder of the body below this is of a silvery colour,
tmged with blue. At this stage, the rays of the caudal which
were present in the egg stage are more marked, but none have
as yet appeared in the other vertical fins.

H7

One larva surx'ived for a few days longer, the chief change
benig the more definite demarcation of the blue band of pig-
ment along" the dorsal aspect, the appearance of rays in the dorsa)
and anal fins, and the more marked protrusion of the lower jaw
(Plate X, fig. 56).

The behaviour of this larva, as observed on the animal when
in the jar of water, differed considerably from that of other
larvae reared. It appeared to seek its food entirely from the
surface of the water, and kept up a ceaseless movement, the tail
being well submerged, while the mouth-part skimmed along the
surface. Small copepods were put into the water, but received
nc attention ; whereas small pieces of bread put on the sur-
face were again and again visited, the larva making repeated
darts, perhaps at the minute forms which collected round the
floating particles.

There is little doubt that this fish is Scoiuhresox saunis,
specimens of which in various stages of growth were found in the
tow-nets about the same time and place. These varied from
about 2 mm. longer than the form reared from the egg, but similar
to it in every other respect, to fonns about 70 mm. in length,
and adults have frequently been found.* (I am aware that
Haeckel, KoHiker, and Ryder have stated that the eggs of
Sconibresox are provided with long filaments, but not having
access to all the literature at present I can only refer to them
now.)

The eggs were procured at the following localities, and on
the dates mentioned : —

Ret . X. ).

Procur

17,70.^

I

17,^35

2

17,161

I

18,021

12

17,078

I

17,888

I

16,798

I

17,861

T

i7,^\v^

I

16,735

I

16,732

2

16,734

5

16,733

5

I7,92,s

1

1 6, 860 A

1

17,861

I

17,108

I

17,969

5

17,730

5

18,048

10

Locality.

Cape Point, X. 81 E. 32 miles

Cape Point, X. 64 E. 37 miles

Cape Point, X.E. by E. 38 miles

Cape Point (approx.), X.E. f X. 39 miles

Cape Point, X.E. by E. 34 mile's

Cape Point, X. 49° E. 38 miles

Cape Pomt, X.E. hy E. 36 miles

Cape Point, X. 50 E. 34^ miles

Cape Point, E. \ X. 20 miles

Cape Pdint Lighthouse, X.E. 28 miles

Cape Point, X.E. by E. \ E. 384 miles

Cape Point Lighthouse, X.E." 32 miles

Cape Pi lint Liiilithouse, X.E. 12 miles

Cape Point hv D. R., X.E. \ E'40 miles

Care Point, X.E. by E. 36 miles

Cape Point, X. ^6 E. ^4.^ miles

Cape Point, E.X.E. 36^ miles

Cape Point. X.E. (appro.x.) ^C) miles ...

Cape Point, E. % X. 38 miles

Cape Point (appmx.) X.F^. 40 miles ...

Date.

Sept. 9th, T903
Sept. loth, 1903
July 23rd, 1903'
Sept. 17th, 1903
July 2 1st, 1903
Sept. nth, 1903
July 8th, 1903
Sept. nth, 1903
Sept. nth, 1903
June 24th, 1903
June 23rd, 1903
June 24th, 1903
"tune 24th, [903
Sept. 15th, i9o"3
July 9th, 1903
Sept. nth, 1903
[uly 22nd, 1903
Sept. i6th, 1903
Sept. 9th, 1903
Sept. 17th, 1903

" Mr. G. A. Boulenyer, F.K'.S., has been yi.isd ennuyh t^ ewntirm thi=
idcntitiealiun.

i4cS

SPECIES XX.

On several occasions an egg was procured of comparatively
large size and with a large peri vitelline space. It ranged from
2 '97 to 2 "64 mm., the perivitelline space being a little over
one-fifth of this total diameter. Only the yolk of the egg is
visible in the water to the naked eye, so that they may readily
be mistaken for much smaller eggs. The yolk was characterised
in nearly all cases by vesiculations, showing on its surface as
polygonal honeycomb-like markings.

Development was comparatively slow. Some eggs which
were procured on August 20th, with embryos about half the
cnxumference of the yolk, hatched out on the 28th.

The embryo appears round the yolk as a series of globular-
looking myotomes. Shortly afterwards, with the growth of the
embryo, the yolk is drawn out into a lenticular-shaped mass,
touching the outer envelope at its ends. The movements of the
embryo were somewhat peculiar. After a long period of rest,
it would suddenly make a spasmodic wriggle, which often com-
pletely reversed its position in the longitudinal direction.

The larva when hatched out proved to be well characterised.
There was an entire absence of pigment ; the yolk was drawn
out so that it extended over more than half the total length of
the animal. The anus was situated midway between the yolk
and the extremity of the tail, and the notochord was multi-
columnar (Plate IX, fig. 51).

The eggs were procured at some distance off Cape Point, as
follows :- — Cape Point, N. 50^ E., 34I miles — by townet at sur-
face ; and Cape Point, E. \ N., 34^ miles — by townet at surface.

In specimens preserv^ed in formalin, the perivitelline space
sometimes becomes opaque and the }'olk disintegrated, but
the eggs are readily recognised.

SPECIES XXI.

This egg is readily recognised on account of its large size
(.1."2— 4 mm.). It has a clear and glassy appearance in the water.
Under a low power of tlie microscope it is translucent, and
showed a number of polygonal markings. No oil globule is
present.

Development was remarkably slow. One procured on the
20th August, and showing no trace of an embryo, showed the
first traces of the embryo on the 24th. On the 28th, the
embryo was less than quarter the circumference of the egg. The
egg did not float ; perhaps, however, on account of small particles
of foreign matter adhering to it. On the 4th September the

149

embryo was about three-quarters the circumference of the egg,
and showed occasional movements. On the Sth it was
well developed. As a more advanced embryo had died at this
stage and this one seemed to be feeble, it was removed from
the egg, and the following characteristics observed (Plate IX,
fig. 52): there was little colouring matter, only a few black
branching cells being observed on head region and over yolk.
The rectum was median (somewhat nearer the yolk than the
end of the tail). The dorsal fin was low, and the tail came to a
point without any caudal expansion. The ventral fin was
deeper than the dorsal and the notochord was multicolumnar.
The most characteristic feature was a low triangular prominence
on the dorsal aspect over the centre of the yolk.

On three occasions the eggs were procured alive, but the
larvae did not on any of these hatch out naturally.

The eggs were procured as follows :—

Kef. No.

Xo.
Procured.

Locality.

Date.

17.724

I

Cape Point, K. -^ N. 38 miles

Sept. 9th, ic;o3.

i6,S()0

■J

Cape Point, X.E. hv E. j E. 40 miles...

-Tuly 14th, 1903.

17.754

I

Cape Point, E. bv N. 35 miles

Sept. 9th, 1903.

16,690

1

Cape Point Lighthouse, E.X.E. 30 miles

June 22nd, 1903.

17,761

Cape Point, X.E. bv E. 38 miles

July 23rd, 1903.

17,078

1

Cape Point, X.E. bv E. 14 miles

July 21st, 1903.

i6,q8o

1

Cape Point, X.E. hv E. 36 miles

July 15th, 1903.

16.826

1

Cape Point, X.E. bv E. f E. 38,1 miles

Julv Sth, 1903.

16,798

I

Cape Point, X.E. by E. 36 miles

n n

17,284

1

Cape Point, E. i X. 46 miles (D. K.) ...

.\ug. 18th, 1903.

18,048

1

Cape Point (approx.), X.E. 40 miles ...

Sept. 17th, 1903.

i^o

X

O

en

o

(XI

o

>

<

p

o
o
w

o

cr

''t''

t— (

y:

1— *

^4

<

u

ii ■

<

-5- =:

r.

-y,

■=/.

y.

■« .^ c

7i f. •_■ :; /

A y,y, y

■r. r. X-

.— 1^'

-/

^

9--^

■<^r=

■;

-■^

r. ^

^. ^

^^

X 0

-r

c/.

<s y. y-. f. r. x x

X

_r o - _x

"/• > *^ >

55 -n c/5 y y-^

-r, I- y, 03>-

<: "-:

^T/.

5 o

o ■

X S i

:/;

X

■J".

0

- u --^

c/. :

, t

a;

_jj

Q

1^

3J

oi

_ c

s

^0

-^ . X

_S 'x -

?>:■<=— ii

r: u rt

' u -1;

>— ~

--: >^

0 0

— f^x

■X, —

-J-. ^i I /-,

IT) 'V) f^* O

1^

^1 r'*.

1-^ i^'y.

I^I

i- ^'

0 ;;;

^

W ■:;

™ ■->

y

X

1* -^

•5 r;

—

•X

o

If.

C '-

>

y

^

.^ =-^

---■■- ^ ^3

\

>< X

= >:'

<-, ^

r;

r- Cn

'■^. ~ '< ^

v

Z' >-.

V.

-y:

E

- -^

■5c5-6^

■,■;;

■>

.—

:;

V — ' '■_ —

-^

^ "i£

-~

_

-^

-<^ -t-, >.

<^

S- ^

•^

X

y:

-r. J-. r.

J-.

r. 'u

-r.

h. 1. >,■

•€•

'-f. _

'-*

0

"^ S:i

•^

y.

— p

j

-f

§=s|

5

y.

I

'E. ~ c—

? - ;-^ ."^^

-1 5

> — - '■'■-

§■5 5^

<L

0

_^ ?■

^ -^ -^ • '^

- 5 ^ i;

^ r- '~ ,L1.

S:

__

>

■^ .^

a

<f, ffi

0

;_ :^

•„

:^

J-'

^ "~"

Z^

i;^

--^

'•^

;:::

H S

E

'i:. '

J

•■^

^

'■-' .'-'

'--

s^

^ ^

s

Ss

c

i-.

■-

r^

'" , S

■:^

\-—

7;

• —

'l-

i^

"

'^ ' ■

0

1^

^ :-*

--^

~~^

—

<5

<

!-r C _£

: : :

X ._ ;

-T C TZ

" w

r|

■X

^ '^^^

*-

i^

•— ^

^

.^

•J

■f.

T^ •■'.

■— ""■ 0

r? S

w J* —

:

Is

-^ ^ V

:

^ s

s 5

^ r u
"X '-"

•^ '■— ^

ijC

I'Ji i

"S ^t

'i.

>,■ C!>-

:i a

a

-y.

r- 31

x

0 ":

;

c

'-' T" — ~

^ ^

^

^

■" ^ -^ '^

—

~-

"~~ ■*" '-~'~.

- ~--

■^ .5"

■~-

.

— 5

>

i^ Q

■^

>

-

^ ■

u J

t

?0£

j .

7.
11

n ■— c

10

--!:

■

1

5^0

^

-<

•"I

-

^-

5 -x

?^

i^

'

'-> -/.

\

r; .

10

-M

•+

M

1^

^

5 -

30

■^

r"

fr;

ir, t^

1^

^-..^--^.

^ ^ M

►s

^ w

♦-..

'-<

I.

.

— p

"S

•

^-^ C

j;

r b

>

rS

s^f

u

^

^

c«

«i

15-

<3

CO

a:

2

<

O

K

<

X
P

o

en

o

CO

O
>

as
<

Q

CO

o
o

w

.-;^

s

cr'

.2

^

C-

X

X

x

A

y.

X

_t: ""-^

X

X

<

A

y.

Q.

;^.

V,

rj_

X

Y.

"

X

X.

O

CO

O

'r*

'"*- ."^

.X

'^

,"^

r-

o

' O

Cj

■<

■"/.

^

-< O

\j

x"

y.

, '-^^

'^

T.

VTO!

CO

JT

=

"^

y.

D

-$■

cj

^

^

—

-

^

"

-^

■^

c

>i

^

\j

!>

X

'^

r-

^

",

. -f<

^

^

s

^

_^

.' ^

C-

ij

::

3

g

1j

;r

^

X

■ —

'-J

~

>,

X

?

t ■

c

5 -^

5

'^r.iz

-:^

h— 1

X

<-

ct

^o

^

—

c

r*

&Lh

Q

T*

>*

■^^

'^

Q

b

b

rt

"^

C^

"^

■^

~

r-

r^

^

_^

i;

"1

^

O

—

• ^

""

.* .

.>-.

~

^

"^

'«

"]S

0

"

■

O

*"

c

^

;^

-^

A

>

-

1

P

c -;

•55 0-5

:l5l i

|c|

Q -

■c/.

'— "^ —

^ ;

. — X

-^- :

I ^^

~ — —

s '^' 3

0 >

X

• 0

"5 ^"> c

0

0 ^

0

■r; u d

■^ a. -

;j

X

^ — —

•- ^ ' — *->

•— X -^

Ol

X

_,

;+

1

7

n

rt

-1-

■^1 0
0 0

1

30

■X.

0
i^

1

^ ■-.

"

'-H

<^i PI

M

^

CO

c

X
r.

PLATES AND EXPLANATIONS OF PLATES.

Fl.ATK V.

Fi.^'. 34. Dead barnacle shell laid npen lu slmw blue eijLjs (.)t Species II.
(nat. si/e.)

.. ,^5. Single e.U.Lj <it" Species II. eiilar.ued and ^lidwiny luimenuis oil
I/lobules.

36. Larva liiewlv hatched) i>l" Species XI. \.l niOi^l^'SSiis (ir/^iz/s/x, \>\-J,\ .
Xat. -i/e J' Sw mill.

>

CD

v:
• ^-t

o

-P

>

CD
Pi

OT

• i-* 'IT

to

\

1

L.

//

fe

Plate VI.

Fi.u". 37. K'4'^ i>i Miuriinis tdsi-iiitiis. N'al. si/e, i"i mm.

,, 3^. Sick- \ie\v ot part (»f ciL;,U capsule ni Miuninis /t/.M /i////>. --hi Avini;
poh'tional mailcinti's and pitlais.

,, 311. K,^^ ui Species X\']. bv transmitlcd li.L;lit. Xat. si/e, 1 'N unu.

,, 40. ,, ., „ bv iL-llectcd li.irlit. „ „ „

Pl.AlK VII.

Fi.U. 41. Larva (newly liatchedl <>\ Species XI \'. Xat. si/.e, b'J nun.
.. 42. ,, .. .. X\'. .. .. -I'i-^ ..

.. 4.V . .. xni 4-('- „

.. 44. Sfroiiidft IIS mil ri'iiiiriis. Hiiuiip. Xat.si/e,

1 'S mm.

a.

(D

■ ^

Cm
O

P.
o

>

CD

Q

i

pq

Pl.ATK VI 11.

Fi.ij. 45. Et,'y III Species XVII. Xat. si/e, 1 57 mm.

„ 46. „ ,, Ciitiirfvx iiuss/iT/, (luntli. \at. si/e. 1 'j mm.

„ 47. Larv.'i Uic\\l\' liatcliedi of Species X\'ll 1. Xat. si/e, |-74inm.

„ 4'^. .. 1. „ „ Oua dav older.

a.

in

03

w
o

CD

o

I 1

>

(U

CO
•a

m

a;

t'l.ATK iX.

Fiiu. 4(y. Iv^.u 1)1' Af^ni'l'iis spniihT, 24 li(iui\-- after iL-rtili/ati' in. Xat. si/.e.
I 'S 111111.

,, 5". Larva ot" .li^nof^ns sf^/uilcr taken frmn euu 4 days alter lertiii/atimi..

„ 51. ,, (ne\\l\- liatelieclml' Speeies XX. Nat. size, 6-() mm.

., ^2. „ oi Specie> XXI. taken Iroui an eyy abmit ig cla>s alter

fcrtili/atidn.

X

f/0

c
o

>

CD

Z-^'/

y/

;«' -t .:•

\

;^

re
ft

I

J 0

I

vJ V

^

Plate X.

Fig. 53. Egg of Sconihresox saiiriis, Walb. Nat. size. 26 mm.

,, 54. Larva (newly hatched) of the same. Islat. size, 8'5 mm.
„ 55- )i (three days after hatchins:;) of the same.
,, 56. ,, (about six days after hatching) of the same.

X

I

pq

0)

cs

o

s

*:■

•If- Mr. ;!'

Plate XI.

Fig. 57. Larvae of Cataetyx messieri. Nat. size, 10 mm.

58. Laiva of Cataetyx messieri surrounded by ova [h) and mucous (c).
The part (a) of the smaller larva was extracted from the
mouth of the larger.

^.

CD

to

■ r— I

o

QJ
P

^s

0)

o

CO

01

I

ON THE

ECHINODERMA

FOUND OFF TIIK COAST OF SOL'TII AFRICA

PART I. ECHINOIDEA.

BY

Professor F. lEFFREY BELL. M.A.

■ Dr. Gilclinst has fi-uin lime to time submitted to me Ins coL
lections of Actinogonidiate Echinoderms ; these are now quite
numerous, and I cannot hope to have a report on them all ready-
in the present year.

As there is no confusion to be feared from reporting
separately on the several classes, I purpose to publish first the
obser\'ations which I have made on the Echinoidea.

The collection is not of interest as containing many new
forms, but rather as widening (or confirming the widening found
by the " Challenger ") of the area of distribution of known forms.
Here it is to be observed that the care of a really extensive col-
lection of Echinoderms has imbued me with a strong, perhaps
too strong, conviction that Echinoderms are very variable
animals ; 1 am not, in other words, prone to raise what may be
individual, and are often slight, peculiarities into specific charac-
ters ; still less do I think that any single character should he
made the basis of a classification, or that a distance of even
hundreds of miles of sea-bottom is sufficient evidence of specific
distinctness. I own that T recognise that T do not belong to the
most modern school of Echinologists, Init I am not so old that
1 do not hope to see the pendulum incline again to my point of
view. At least, I am still in good company.

The discovery of Pdlacolaiiipas m a living state is of great
interest.

io8

DESMOSTIOHA.

ClDARlDAE.

Cidaris sp.

Single Examples of this laiuily were collected as follows

Local it\'.

Dak

]\<>\\ I'idciii ed. DlpIIi.

Rcl.
No.

East Luudon, \.\V. i N., | April i5tli, 1901 .By vSliiimp Trawl 250 t<i 303 b!i
I J miles. i I I'nis.

I'lnhlan^'akulii River, X.W'.. March i4tli, 1901 H\ Dredge ...; 50 ,, 9J

by X., 7.1 miles.

i do not recognise them as belonging to any described species,
and as the classification of this family has lately been thrown
into the melting pot by Dr. Mortensen, whose rearrangement of
the genera is, in his own words, " a provisional classification,
which can scarcely be correct throughout," I propose to wait till
something further* can be said about South African Cidarids.
The only species as yet recorded from the Cape is Cidaris
iiictularia.

DlADEMATIDA^.

Diadema saxatile.

lii/iiii/is saxa/ilis, Linn., Syst. Nat. x. (1758), p. 664.
Diad cilia scfcsniii. Gray, Ann. Phil. x. (1825), p. 426.
Diad cilia saxatile, Loven, Bih. Sv. Vet. Ak. Hdlgr., xm.» 4,
No. 5 (1882), p. 135, ihiquc citata.

A single )'oung specimen of tliis widely distributed species
was collected as follows : —

Ldcalih',

Cape Mot.gan, \. \ W.,
10.1 miles.

Date.

July 20tli, 1901

HdW I'meiiied.

Hv Dredge

Depth.

//

tms.

Ret.

Xo.

I I I

Astropyg-a radiata.

Cidaris radiaia, Leske a])ud Klein (^i/"78), p. 52.

As/roprga radial a, Gray, Ann. Phil. x. (1825), p. 426; A. Ag.,

Rev. Echin. (1872). ]). 94, ihiquc citata.

* vSinee writins:; the above I have received Dr. Meijere's repoi 1 on the
Kchiiioids of the " Siboga." where I see that the Danish Xaturalist's relinements
itre iKit iicceptcd.

I (K)

Four young specimens procured as follows : —

I,<pc;ilit\'.

Dak

How 1*1 iicurcd

Depth.

Ret".

X.'>.

Tiifcjela River Mouth, X.

I'\-b, bth,

1001...

Bv Shrimp Traw 1

-1^^

55 I'llT^.

X... 57

bv W., i8 miles.

Cape Xatal, W. hv X.. ()',

Dei_-. i7lh

, lc)00

I'y Lar,L;e Traw 1

4'^

iins....

., 100

miles.

Cape Xatal. W. bv X.. 6.',

Dee. 14th

, 1000

Bv Lai>ie Diecl.i;e

=^4

l"ms....

„ IDS

miles.

Tu-,'ela River Mouth, N.W.

P\'b. hth.

1001...

B^• Slirimp Trawl

Vi

t(i 42

.. 17.1

by X. f X., I5i miles.

i

Ims.

(F.F.
'"7491

This species was long- since collected at the Cape by Dr.
Krauss, although it does not seem to have been mentioned in any

echinological list.

ECHINOTHURIIDAE.

For, I think, the first time Echinothuriae have been found
south of the Cape ; the family is, in this collection represented
by two specimens of Plwnnosoina, allied respectively to
P. tenite A. Ag. and P. Inirsariinn A. Ag. ; it is useless to at-
tempt to diagnose from isolated specimens, and the present con-
dition of the family does not warrant any addition to it that need
not be made ; it is to be hoped that the large collection made by
Professor Agassi/, will allow us to gain a definite idea of the
range and character of the variation exhibited by species, of
which as yet too few examples are known. It is on the basis
o^' such knowledge ratlier than on the distinction of minute
differences that satisfaclor)' systematics is bascnl. | Dr. Cilclirisl
has lately been so good as to send me a large series, which I shall
take an early opportunity of studying.]

The specimens were obtained as follows : —

Locality.

Date.

How Procured.

Depth.

Ref.
Xo.

Cape Xatal, X. hv K. (.Vp-

proximate) 24 miles.
Cape Point, X. Si' K.32

miles.

.\pril. 4th, TQOi Bv Shrimp Trawl 440 I'ms....
.\u.U. 20th, 1003 By Shrimp Trawl 460 l"ms....

1 V)

ECHINIDAE.

Echinus ang'ulosus.

Cidaris angii/flsn, Leske, Addit. (1778), p. 92.
Echinus au^i^ii/osns, A. Ag., Rev. Echm (1872), p. 122, ihiqite
citata.

170

This species was obtained as follows : —

L.H-alitv.

Date.

H.

i\\ I'l'ocured.

Depth.

Ret.

No.

Simnu's Tdwii

17th Sept.

iS()7

Co

llected at low
tide

121

Three Anchor Bay

26th April,

1898

n '1

'^5

Somerset West

2M-c\ Oct.,

1897

)i '•

128

,,

28th April

1898

- -

130

St. lames'

22nd April

1 898

1,

-

13'

Lat. 34 12' 54" \

31st Au.t;-.,

1 897

P>v Trawl

23-37 'ms.

135

Lon.L,'. jX. 36' 9" )

East London

17th Dec,

] 898

Co

llected at low
watei'

^

I3'>

Between Mossel Bav

Pier

()lh June,

1898

n " n

—

36 D

andSt.BlaizcLi.uh'tl

oiise

Knysna sliore between

17th Aug.,

1898

}' >'

—-

318 A

jetty and villu^'e.

Between River and Sebas-

15111 July,

1900

1! •!

—

9098

tian BliitY, nearer

•the

Imnier.

The spines of this species appear to be pretty constantly
either red or purple ; the difference is so marked and seems to
be so equall}' divided ainono- the specimens that it appears to
me that it would be worth while to make observations on fresh
specimens to see if it is due to differences of sex. I do not
thini< llii-, ditTercr cp can 1)'.- regarded as specihc, notwithstandinp'-
Dr. Mortensen's staten:ent that colour is an " excellent gruide "
for dislinouislnng the sj^ecies of Eclnnus.''

Echinus gilchristi.

A number of specimens were obtained of an Ecliinus whidi
a])pear to be unclescribed.

Test tending" to be conical, with rather small apical area ;
above the ambitus there is, ordinarily, but not always, a large
primary tubercle on each alternate interambulacral plate ; simi-
larly below the ambitus there is one on each plate ; the secon-
daries are of moderate size, and are rather closely packed. On
the ambulacral plates above the ambitus there is more irregu-
larity in the disposition of the primaries, but below it there is
nearly always one on each plate; these and the secondaries are
somewhat smaller in size than on the interambulacrals. The
primar)' spines are white, moderately long, giving the uninjured
creatures the general appearance of the northern E. ac.it tiis ;
the secondary -spines are also white. The denuded test is of a
stone-grey colour ; young tests are flattened rather than conical.

This species may be named after its discoverer, the Govern-
ivient Biologi.st at tiie Cape.

* Danish Ingolf Exped. Echinoidea (1903) p. :

V'

lyi

Measurements of three denuded specimens : —

Test.
Diam.

Height.

Interambul-
acra.

Ambulacra.
Breadtli at Ambitus.

Mouth
Diam.

Vent.

55

385

23

12

19

6

45

36

20

10

175

5-5

38

21

14

8-5

11

4-5

This new species was obtained as follows

LocaUty.

Date.

How Procured.

Depth.

Kef.

No.

Cape St. BUii/e, N. by E.

Oct. 15th, 1899

By large trawl

35 fms.

34

1 E., 6^ miles.

Cape Pt. Light Ho , N. 16°

Sept. 5th, 1902

From shrimp

85 fms.

70

E., 10 miles.

trawl

Cape Pt. Light Ho., S.W.

Dec. 5th, 1902

By shrimp trawl

33 fms.

83

bv W. \ W., T, mi., trawl

hauled up Cape Pt Light

House S.W. bv \V., 4.}

miles.

False Bav

Oct. 4tli, 1898

Bv large trawl

—

85

Lat. 34 iq' 28" S.

Long. 18 :^Ti' 50" E.

Cape Seal, N. a E., 38 miles

Feb. 2oth, 1902

Bv dredge

80 fms.

138

Cape St. Blaize, X. by E.

Oct. 15th, 1899

By large trawl

35 f"is.

145

J E., 6} miles.

Cape Point, N. 77' E. ...

July 29th, 1903

By shrimp trawl

660-700*

fms.

200

Echinus juv.

The young- of what I take to be an undescribed species ot
Echmus, and which I have reserved for future study, were taken
as follows : —

Locality.

Date.

H

ow Procured.

Depth.

Rcf.

No.

Cape Pt. Light H

J., S.

1
■1

Tune 8th,

1900

B\

• large dredge

14 fms.

I/O

W., ij miles.

Zwartklip, N.E. ^,

E.,

li-

Nov.

26th,

1902

By dredge

II fms.

180

miles.

Zwaitklip,N.E.iN.

, I mi

les

Nov.

17th,

1902

Bv dredge

10 fms.

182

Kocky Bank, False

Bay

Oct

8th,

1902

By dredge

17 Ims.

183

All small llatlened specimens.

172

PETALOSTICHA.

NUCLEOLIDAE.

Palaeolampas crassa.

Palacolauipas crassa, Bell, P. Z. S. (1880), p. 48.
Obtained at lollowing place : — ■

Locality.

Date.

How Procured.

Depth.

Ref.

No.

Kromhout, N.N. E.,2 miles

11 11

Dec. 12th, 1902

11 11

By dredge

11

14 fins.

n

16326
16326.4

It was a great and unexpected pleasure to me to find that Dr.
Gilchrist had dredged living specimens of this interesting form,
to our knowledge of which there has been no addition since I
described it in 1880. It appears to be fairly abundant, and
examples of va.rious sizes have been obtained. In the fresh
state the test is uniformly covered with short yellowish green
spines, longer below than above the ambitus.

The only criticism that has been passed on this genus that
I know of is by M. de Loriol,* but, as he compares with
P. crassa, " EcJiinolamfas rangi," wliich has petaloid ambulacra,
I can only suppose that I did not make my meaning clear to him.

SPATANGIDAE.

Pourtalesia carinati.
Pnurtalcsm carhiata, A. Ag., Rep. Chall. Ech. (1881), p. 133.

A rather small specimen of what appears to be this species
was obtained as follows: — Locality: Cape Point, N.E. by E.
.:|- E. 4() mi. From shrimp trawl. Depth, 900 fms. Date,
July 2 1st, 1903. Reference, No. 140.

It is a matter for congratulation that the resources of the
Cape of Good Hope naturalist enable him to dredge in deep
waters, for it was in the sea between Marion and Crozet Islands
(at Stats. 146 and 147) that the " Challenger" made two of her
most prolific hauls.

Mem. Sue. Plus, (iencne. xxvii., p. SS.

173

Ureohinus naresianiis.

Urecktmis nnrcsiaii/is, A. Ag"., Chall. Rep. Ech. (iS8i), p. 146.

This IS, I believe, the first time that this interesting and very
variable species has been dredged since the " Challenger " ex-
]oedition. The student may be reminded that valuable observa-
tions supplementary to the original description were made by
Loven in 1883 from duplicates obtained from the British
Museum (see Kongl. Sv. Vet. Akad. Hand, xix., No. 7, pp. 90
el seq.).

Spatangus raschi.

Spafdiigiis raschi, Loven, OEfvers. Vet. Akad. Stockholm,
1869, p. 733 ; Bell, Cat. Brit. Ech. (1902), p. 167.

Localitv.

Date.

Cape Seal, X. ;,' K., 3S miles Feb. jolli, H)02.

Lovenia clong'ata.

Spatniigiis eloiigntiis, Grav m Eyre's Australia i. (1845) p. 436.

Lovenia eiongata id. Ann. and Mag. vii. (1851), p. 181 ; A. Ag.,
Rev. Ech. (1872) p. 139.

Locality.

Date.

How Procured.

Depth.

Rel.

Xo.

Cape Natal, W. ■{ X.,

12.',

Dec. i/tli, 1000

Bv larj^e dredge

H5 fms.

10.785

miles.

I'mhlot Kiver Moulh,

X.

Dec. ii)tli. i^oo

mo fms.

10.1)16

\V. 0; \V., \yh miles.

rmhloti, River Mouth

. X.

*• ,j

M M

100 inis.

I0016K

\V. '. \V., 15.1 miles.

174

Echinocardiiim australe.

lichiiuHnrdiuiu australe. Gray, Ann. and Mag. vii. (185 1), ]x
131 ; A. Ag-., Rev. Echin. 1872), p. ing.

Lncalitw

Date

How Procured.

Depth.

Ref.

Xo.

False Rav, Kumaii Rock,

fuiie 7th,

1 900

Hv lart:e drcdije

22 I'ms.

30H2

X. hv W, 1 ' miles.

Cape Han- Klip, \. K. \

Feb. 26th,

1()02

Rv shrimp trawl

los t'ms.

14449

E. 27.'. miles.

False Hav, Hakkov en Rock,

June 7tli,

I()00

Hv larj^'e dredge

22 fms.

30()8 C

W. ;, \. ,' miles.

Constable Hill, K.S.K. 15

March 13th

, 1902

Rv shrimp trawl

100 fms.

75

miles.

Lion's Head Fast, iS miles

March 22nd, 1900

Rv large dredge

104 fms.

93

Lat. 34^ 43' 15" S. (
Lonjf. 18° 30' oo"F. )'

Oct. 1 8th,

1898

By shrimp trawl

123 fms.

98

Algoa Bay, near Roman

Nov. 10th,

1898

Rv dredge

—

107

Rock. Lat. ^4 00" S.

Long. 25° 4S' F.

Mossel Rav

.lime i^th,

1898

Rv shiimp trawl

18 fms.

10(>

Faiilsber,!4-, W.X.W. 1 miles

June Hth,

1900

Rv largu dredge

24 fms.

17.S

Paternoster Point, S.F. \.

March 17th

, J 902

Rv dredge

80 fms.

181

S. I) miles

Echinocardium flavescens.

Spatagiis ftavesccns, O. F, Miill, Prod. Zool. Dan. ( ij'jd),
p. 236.

Echinoc<ir(liiii>i Jlcircsci'iis, A. Ag., Rev. Ecli. (1872), p. lio;
Ik'll, Cat. l-5rit. Ech. (1892), p. 171 ihiquc citdla.

Localitv.

Date.

How Procured.

Depth.

Ref.

Sebastian RkilY, X.W bv
W. \, W. 8^ miles.

.Iul\- 5tli, 1900

Ii\- shiimji ti .iwl

34 fms.

7050

Cove Rock, X.W. :; W.
134 miles.

July 30th, 1901

Py dredge

80- 130 fins

102

,\Igoa Bay, neighbourhood
of Roman Rock, Lat. 34
2' S. Long., 25° 45' 30" E.

Xov. 10th, 189X

By dredge

—

I 12

L.at. 33°' 54' 15". I'Ong. 25
53' .3f>" •

Xov. 7th, 189S

By dredge

31 fms.

i«5

^75

LESKIIDAE.

Brissopsis lyrifera.

Btissiis lyrifer, P'orbcs, Jirit. Starlish ( 1841), p. 187.
Brissopsis Ivn/eni, Ag. & Des., Ann. Sci. Nat. viii. (1847), P- ^S !
lidl, Cat. Brit. Ech. (1902), p. 172.

Locality. Date How Procured. Dcptli. ^^ "

Struys Point, X.W., 15 inilc-s July 9th. 1902 By shrimp tiawl 48 fins. 15,30/
Cape Hanj^' Klip, X.E. \ Keh. 26tli, 1002 ,, ,, 105 fins. 14,440

E., 27^ miles.
Stiuvs Point, \. by W. !, Auj,;. 2Sth, 1002 ,, ., : 42 fms. I5.34('

\V., ji miles.
Lion's Head, S. 72K., 47 INLir. idlli, 1900 By large dredge j 190 fms. 2.152

mile^. j I

Cape Point Li'jht House, Sept. 8tli, 1902 Bv shrimp trawl i 8=; fins. 74

N.E. liy N., 7J miles. '

Lat. 34" 34' S., Long. 18" Oct. uSth, i.S(),S Bv dredge 100 fms. 103

32'45"'E.
Table Mountain, S.K. by .\p1il4th, 1902 By shrimp trawl 120 Inis. 197

S. f S., 45.i miles.

Schizaster fragilis.

Brissus fragilis, Diib. and Kor., Vet. Ak., Hoclg 1844 (1846),

p. 280.
Schisastcr fragilis, Ag. and Des. ; see Bell, Cat. Brit. Ech.

(1892), p. 164, ibiqnc citata.

Locality: — In the neighbourhood of Bird Island, Lat. 33°
55' S., Long. 26° 11' 45" E. Date: — 16th March, 1899, by
dredge. Depth, 29 fms. Reference, No. 29.

These two last species attain a much greater size than in
Northern waters. The mean monthh^ temperature of the water
ui this locality for 1902 was f)4'4 deg. Fahr., the maximum 74,
the minimum 56.

[Published 5TH October, 1904.'

THE PLANKTON

OF

THE SOUTH AFRICAN SEAS.

MY

P. T. CLEVE.

I.— COPEPODA.

INTRODUCTION.

Some time ago Dr. Gilchrist suggested that I should examine
the samples of plankton collected in the South African Seas
during the Government's survey of that region. I accepted the
proposal with pleasure as this region has a certain bearing on
some oceanographical questions in which I am particularly
interested. The first set of samples had, with one exception,
been obtained by means of a comparatively coarse tow-net
fastened on the beam of the trawl. They contained almost
exclusively animals, schizopoda, amphipoda, sagittae, tomop-
teris and a large amount of copepoda. I intend to give in the
following a report on the latter, and hope later on to be able to
give an account of the halocyprida and the chaetognatha.

The region of South Africa is of the greatest interest in its
planktonological aspect, as it represents a battlefield of two
mighty currents of different origin, the Agulhas Current and
the Westwind Drift. The former carries the warm water of the
Indian Equatorial Current towards the south, the latter the
water of the Southern Pacific Ocean through the space between
Fuegia and the Antarctic Continent.

The east coast of South Africa has a steep slope so that
depths of 900 to 3000 metres (500 to 1800 fathoms) will be found
at a short distance from the land. On the south side, from
about Algoa Bay to the Cape of Good Hope, extends the
triangular Agulhas Bank, its southern point being about 2
degrees of latitude south of Cape Infanta. The depths in this
region are comparatively small, as a rule 70 to 100 metres (4a
to 60 fathoms).

B

178

The warm Agulhas Current flows along the east coast of
South Africa and east of the Agulhas Bank with a variable
velocity, strongest in the summer months, especially in Feb-
ruary. As a rule the average velocity amounts to about 2 knots ;
in the winter, especially in July, the velocity diminishes con-
siderably, and the cold water from the Westwind Drift
repels and permeates the current.

South-east of the Agulhas Bank the waters of the Agulhas
Current become deflected and mixed with the water of the
Westwind Drift, which moves eastwards, in the wintei months
and April with a velocity of about i-i"25 knots.

The mixed water of the Agulhas Current and ihe Westwind
Drift continues towards the east with a velocity of i"5-2 knots.
During the passage it sends a branch along the west coast of
Australia, and a very mighty one, when it meets the south end
of America, or the north going Humboldt Current. This may
explam the fact that I found in the samples collected east of
Natal a very considerable number of the copepoda discovered
by Giesbrecht in the depths west of South America.

The currents over the Agulhas Bank are as a rule weak and
variable. Between Cape Town and Cape Agulhas a current
sets in in an E.S.E. direction with a velocity of one mile per
hour. Along the south coast from Cape Agulhas to about
27° E. an eastward running current has been observed. Thus
the water south of the Cape Colony is derived, at least to a great
extent, from the west, that is from the origin of the northwards
running Benguela Current. The latter is to be considered as
a branch of the Westwind Drift, more or less mixed with wateis
from the Brazil Current, the continuation of which it touches
during the passage from the South of America.

It might be expected that the waters on the Agulhac Bank
would possess a kind of plankton different from that of the
waters East of South Africa. This will be proved by the
followmg account. It will be found that of the copepoda
found south and west of the Cape Colony a considerable num-
ber also occur in the northern hemisphere, north of a line
traced from the Newfoundland Banks to the Azores and the
Cape Verde Islands, some so far to the north that they pass
through the Faroe Channel and reach the coasts of Scandi-
navia. There is thus a strong probability that the hvpothesis 1
have enunciated, that the waters of the temperate Atlantic in the
northern hemisphere originate not from the Gulf Stream, but
from the Benguela Current, which is supposed to pass as an
under-current below the waters of the tropical Atlantic.

In order that the reader may judge of this question I have
compiled from the literature an account of the geographical
distribution of all copepoda found in the South African Seas.

I will in the following treat first of the copepoda of the west
coast, then of the south coast and, finally, of the east coast.

179

Copepoda of the AVest Coast.

The samples examined were the following.*

1. (No. 23) Table Mountain, N. 79° E., 40 miles, 450 metres

(250 fathoms) iHth April, 1900, coarse net. See
Marine Investigations of South Africa, I., p. 193.

2. (No. 30) Lion's Head, N. 54° E., 38 miles, 350 metres

(195-204 fathoms), i6th lApril, 1900, moderately fine
net.

3. (No. 17) Lion's Head, N. 76° E., 28 miles, 250 metres

(140 fathoms), i6th April, 1900, half-coarse net.

4. (No. 8) Lion's Head, N. 63° E., 34 miles, 277 metres

(15 fathoms), 19th April, 1900, moderately coarse net.
The result of the examination of these four samples was 43
species, a considerable number, considering that all samples were
collected in one and the same month, and chiefly along the bottom.
The following table indicates the species found in every
sample ; the signs here, as in the following, indicate the appar-
ent frequency, viz. : cc very common, c common, -|- not rare,
r rare, rr very rare.

It is evident that the list cannot be a complete one of the
forms occurring in the origin of the Benguela Current, but,
nevertheless, it may be of interest to examine the distribution
of the species in the Atlantic and Mediterranean.

I consider as tropical such forms as occur in the Equatorial
and Florida Currents, as temperate such as occur in the space
north of a line from the Newfoundland Banks to the Azores and
Cape Verde. I count in this category also such forms as have
been found in the Gulf of Guinea, but not in the Florida Current.
Of the species noted west of South Africa

32 per cent, have been found in the tropical Atlantic,

42 ,, ,, „ „ temperate Atlantic,

71 ,, ,, ,, ,, Mediterranean,

35 „ reach the Faroe Channel,

21 ,, have been found north of the Faroe Channel.

Copepoda of the Agulhas Bank.

The samples examined were as follows :

1. (No. 26) False Bay, off Roman Rock, 37 metres (i9'5-22

fath.) 13. X. 1898, fine net.

2. (No. 18) Cape Infanta, N.N.W., 4 miles, 62 metres (35

fath.) 6.vii.i90o, moderately fine net.
(No. 4) Cape Infanta, N.N.W., 4 miles, 67 metres, (38

fath.) 6.vii.i9oo, coarse net.
(No. 2) Cape Infanta, N.JW., 6 miles, 70 metres (40

fath.) 6.vii.i900, moderately fine net.
(No. 5) Cape Infanta, N.^W., 13^ miles, 74 metres (42

fath.) 13.vii.1900, coarse net.
(No. 20) Cape Infanta, N.^W., 13^ miles, 74 metres (42

fath.^ 13.vii.1900, moderately fine net.

* The numbers in brackets are the Reference numbers.

i8o

3. (No. 16) Cape Infanta, N.E.fN,, 4^ miles, 65 metres

(37 fath.) 29.vi.1900, moderately fine net.

4. (No. i) Sebastian Bluff, N.W. by N.JN., 4 miles, 60

metres (34 fath.) 12.iv.1900, moderately fine net.

5. (No. 12) Sebastian Bluff, N.W. by N.JN., 4 miles, 60

metres (34 fath.) 22.vi.1900, coarse net.

6. (No. 21) Sebastian Bluff, N. by W.fW., 9I miles, 71

metres ^o fath.) 11.vii.1900, moderately fine net.
(No. 29) Sebastian Bluff, N.N.W., 7^ miles, 69 metres
(39 fath.) 11.vii.1900, moderately fine net.

7. (No. 27) Mossel Bay, 37 metres (21 fath.) 24.vi.1898.
(No. 28) Cape St. Blaize, N.W. by 3°, 5^ miles, 58 metres

(33 fath.) 29.vi.1898.

8. (No. 25) Cape St. Blaize, N.E.FN., 8^ miles, 63 metres

(36-37 fath.) 23.x. 1900, coarse net.

9. (No. 19) Buffalo Bay, Walker's Point, N.W. by W.^W.,

i^ miles, 40 metres (23 fath.) 12.X.1900.

10. (No. i^) Buffalo Bay, 53 metres (30 fath.) 26.iv.1900.

11. (No. 9) S.E. Coast, 33°6'46'S., 27°56'45"E., 71 metres

(40 fath.) 30.xii.1898, coarse net.

In spite uf the comparatively large number of samples the
number of species did not amount to more than 27, but on
the other hand most species were represented by a very large
number of individuals. Of these copepoda only 63 per
cent were found off the west coast, which may be accounted
for by the fact that the latter were collected along the bottom
at a considerable depth.

Of the species noted above the Agulhas Bank

78 per cent, have been found in the Mediterranean.
41 ,, ,, ,, ,, tropical Atlantic.

63 ,, ,, ,, ,, temperate Atlantic.

48 „ ,, ,, „ north of the Faroe Channel,

in the North Sea and along the coasts of Scandinavia.

Table I. Copepoda, collected in April, 1900, west of

South Africa.

No. of the sample (page 179)

Depth in metres

Acartia dan;e

Aethidius i:;ieshrechti C:e. ...

Calanus brevicornis

tinmarchicus

minor

■ tenuicornis

Calocalanus pavo

Candacia chirura C:e. n.sp...

pachydaetyla

Clausocahimis arcuicornis ...

furcatus ...

Corycaius daiue...

furcifer ...

ovalis

robustus... ... ... ... ... rr

1

450

2

350
rr

3

250
rr

4

277
rr
rr

r

e

...

rr

r

+
r

r

r

rr

rr

rr

rr

rr

+
r

r

+

rr
rr

rr
r

i8i

Euchieta marina
Gaetanus armiger
Heteroihabdus abyssalis

papilliger

spinitrons

Lubbockia aculeata
Lucicutia Havicornis
Metridia luceiis
Oithona plumifera
Onciea conifeia

media

mediterranea

venusta ...

Paracalanus parvus
Pleuromamma abdominalis

gracilis ...

— robusta ...

xiphias ...

Pseudodiaptomus serricaudatus
Rhinocalanus nasutus
Sapphirina angusta

salp;e

sinuicaiida

Scolecithricella minor

• neptimi C:e n.sp.

Scolecithrix dan:e
Xanthocalanus fragilis

rr

...

...

...

rr

...

rr

• • •

rr

rr

rr

...

• • •

* . .

rr

. . •

+

+

c

+

n

r

+

r

rr

...

r

r

r

r
r

...

...

r

+

r

r

+

c

r

+

c

rr

r

...

r

+

rr

+

rr

rr

...

• . .

rr

rr

rr

rr

rr

rr

. . .

r

rr

rr

rr

Table II. Copepoda, collected south of the Cape Colony.

No.

I

2

3

4

5

6

7

8

9

10

II

Depth in metres

37

62-74

65

60

60

69-71

37- 5«

63

40

■53

71

i

13

6-13

29

12

22

II

24-29

23

12

26

30

Date ... •

X

VII

VI

IV

VI

VII

VI

X

X

IV

XII

1

[898

1900

1900

1900

1900

1900

1898

1900

1900

1900

1898

Acartia dan;e

r

Aethidius armatus ...

+

r

r

...

Calanusbrevicornis...

+

r

r

c

— ftnmarchicus

+

+

r

r

r

c

cc

+

+

cc

c

c

c

cc

+

c

c

+

— IllliU'i ...

Calocalanus pavo ...

rr

Candacia bipinnata...

r

+

r

+

c

r

c

c

r

r

Centropages chierchia;

+

r

r

r

+

typiciis

c

r

...

r

...

...

+

r

Claasocalanus arcui-

cornis...

+

^' .

r

+

r

t. .

■ ■■

Coryc:eus ovalis

r

. . .

Eucalanus crassus ...

r

..,

Euch:eta spinosa ...

rr

Metridia lucens

+

+

r

r

Oithona nana

r

+

plumifera

r
r

r
r

...

...

...

— silllilib ...

Onaea conifera

'V-

mcH if (-'i*!'"i n fi

1*

— IllCLULCl 1 tlilCcl

subtilis

+

1

+

. • t

1-

Paracalanus aculeatus

r

+

1

r

• • .

cc

r

...

cc

Pleuromamma gracilis

...

...

...

...

...

r

...

Pseudodiaptomus ser-

ricaudatus

c

r

c

c

...

...

Rhinocalanus nasutus

+

c

r

c

r

r

r

r

r

+

Xanthocalanus fragilis

rr

...

...

...

...

...

...

...

...

l82

Copepoda of the Agulhas Current.

The samples examined were the following : —

1. (No. ii) Cape Natal W. by N. 6^ miles, 97 m. (57 fath.)

14.xii.1900, moderately fine net.

2. (No. 13) Port Shepstone N.W. by W.fW. 12 miles,

530 m. (300 fath.) 5.xi.i90i, moderately fine net.

3. (No. 71) Cape Natal N. by E. appr. 27 miles, 780 m.

(440 fath.) 4.iv.i90i, coarse net.

4. (No. 22) Port Shepstone N.W. by W.fN. 12 miles,

900 m. (500 fath.) 5.xi.i90i, moderately fine net.

5. Morewood Cove (Natal) N.W. by N.|N. 3 miles,

surface 17.xii.1900 (contained abundantly Xoctiluca
miliar is).

6. (No. 24) Port St. John between Anchorage and Bluft'

Point 29 m. 8.iv.i90i, moderately fine net.

Table III. Copepoda, collected east of South Africa.

No. ...

Depth in metre ...

Date ...

Acartia dan:e
Acrocalanus gibber

ijracilis ...

Aegisthus aculeatus

mucronatiis

Aethidius {^iesbrechti
Augaptilus palumboi
Calanus brevicornis

darwini

finmarchicus

minor

tenuicornis

• vulgaris

Calocalanus pavo ...
Candacia catiila ...
curta

inermis C : e. n.sp.

pachydactyla

tenuimana

truncata ...

varicans ...

Centropages chiercliia-

furcatus

typicus

Chiridius poppei ...
Chirundina streetsii
Clausocalanus arciiicornis
furcatus

I

2

.•)

4

D

f)

'>/

5.^iO

7S0

900

0

29

17

5

4

5

17

8

xn

XI

i\-

XI

XII

IV

900

r

I goo
rr
rr
rr

rr
rr

1 90 1

190 1
rr

rr
ri
rr

T900
c

1 901

r
r

+

rr

r

rr

r

+

1"

rr

rr

rr

rr

rr

rr

I'r

rr

r

rr
rr

rr

rr

rr

r
r

rr

rr

rr

rr

r

r

+

i83

Conaea rapax
Coryc^Eus danie ...

furcifer

robustus ...

— speciosus ...

Eucalanus attenuatus

crassus

elongatus ...

monachus ...

— mucronatus

pileatus

subcrassus

subtenuis ...

Eiichaeta acuta

affinis C: e, n.sp.

longicornis

marina

media

tonsa

Euchirella messinensis

— venusta

Gaetanus armiger...

miles

Heterorhabdus abyssalis

austrinus ...

papilliger ...

spinifrons ...

tanneri

Lucicutia aurita C: e, n.sp.

■ bradyana C: e, n.sp.

clausii

flavicornis

Labidoceia acuta ...
Metridia brevicauda

prmceps ...

venusta

Oithona plumifera

rigida

Oncaea conifera ...

media

mediteiranea

venusta

Pachysoma tuberosum
Phaenna spinifera
Phyllopus bidentatus
Pleuromamma abdominalis

gracilis

• robusta

xiphias

Pontella securifer ...
Pontellina plumifera
Rhinocalanus cornutus

nasutus

Sapphirina gemma

nigromaculata

. opalina

sinuicauda

Scolecithrix danae

• persecans ...

securifrons

Temoia discaudata

stylifera ...

Undechaita major
minor

I

2

3

4

rr

r

rr
rr

r

r

r

+

+

r

r

r

+

+

+

r

...

...

r

r

rr

...

r

r

r

r

r

r

r

...

rr
rr

rr

rr

r

r
rr

rr

r

rr
rr

rr

rr

rr

r

rr

V

rr

rr

...

r

r

r

r

r

rr
rr
rr

rr

rr

rr

...

rr

...

...

rr

rr

rr

rr

rr

rr

'•

r

r
r

r

r

r

r

r

rr
rr
rr

c

+

+

r

+

+

+
r

r

r

r
rr

rr

rr

rr

r

+

+

r

r

rr

r

rr
rr

r

rr

+

+

+
rr

r

< • >

rr

rr

rr

rr
rr

+

rr

rr

1 84

The number of species found east of South Africa amounts
thus to 92. The geographical distribution will be found as
follows : —

In the western Pacific, east of South

America, have been found ... 80 per cent.

In the Indian Ocean ... ... 64 „ ,,

In the temperate Atlantic ... ••• 55 >; »

\n the Mediterranean ... ... 50 ,, „

In the tropical Atlantic ... ... 19 ,, ,,

North of the Faroe Channel ... 8 „ „

There is thus, as concerns the copepod-fauna, a very con-
siderable resemblance between the region west of South America
and the Agulhas Current.

List of Species of Copepoda Found Round South
Africa and their Geographical Distribution.

In the following I have not indicated the literature, because
the reader will find in most cases sufhcient references in the
works of Giesbrecht : " Copepoda, Gymnoplea " in " Das
Thierreich," Berlin, 1898, and " Systematik und Faunistik der
pelagischen Copepoden des Golfes von Neapel," Berlin, 1892. —
In some few cases only references have been necessary.

Abbreviations for authorities are : —

B. Brady. A. S. A. Scott.

C:e. Cleve. A. S. & T. A. Scott and

Cls. Claus. I. C. Thompson.

G.O.S. Sars.
D. Dana. T. S. T. Scott.

G. Giesbrecht. T. I. C. Thompson.

L. Lubbock. a.o. and others.

N. Norman.

Acartia dance, Giesb. — Sparingly E., S. and W. of South
Africa.

G.D.-^Pacific i8°N.— 3°S. lyi^E.- 8o°W. {G.).— Atlantic :
Benguela Current, 30°S. ii°E. (C:e). Cape Verde Region
(G.), 34°— 39°N.i2°— 66'^W. (C:e). West Mediterranean (C:e).

Acrocalanus g-ibber, Giesb. — Agulhas Current.

G.D.— Hongkong (G.). S.E. of New Guinea (A.S.). Malay
Archip. (C:e). Ceylon (A.S. & T.). Arabian Sea, Red
Sea rC:e, G., a.o.;. From ii°N.84°E. to i6°S.42°E. (T.).
Along the east coast of Africa from the Gulf of Suez to 30°S.

(J-h

Acrocalanus gracilis, Giesb. — Agulhas Current.

G.D.— Pacific 20°N— 4°S. (G,). Indian Oc, Malay Archip.
(C : e). Ceylon to Socotra (A.S & T.), Arabian Sea and Red
Sea(C:e), 2°S. 84°E. fC:e).

i85

Aegisthus aculeatus, Giesb.— Agulhas Current.

G.D.—Pacific 3°S. 99°W. (G.). Atlantic, 52°N. i6«W., hcUil
from 1410 m. (T.).

Aegisthus mucronatus, Giesb. — Agulhas Current.

G.D.—Pacific 3°S, 99°W. (G.) Atlantic, Gulf of Guinea
(A. longirostiis T.S. partim), 52°N. 12°— i6°W., hauls from 2264
and 664 m. (T.).

Aethidius armatus, Boeck. — G. O. Sars : Crustacea of
Norway, Calanoida, p. 25 P. XIII., XI Y. A. a. Brady, partim,
z'idc G.O. Sars. — Sparingly south of the Cape Colony.

G.Z).— Shetland, Orkney (T.S.). Shagerak. (C : e). Off south
and west Norway from Christiania to 7i°N. (G.O.S.) — Othei
localities doubtful, because this species has been confounded
with the following. The South African specimens agreed per-
fectly with the northern. They were only a little smaller or
1,6 millim. in length (1,8 millim. according to G.O.S.)

Aethidius giesbrechti, C:e — nomen nov. (A. armatus
Giesb. F. Neap. p. 213). — Very rare east and west of South
Africa.

G.Z).— 3°S-ii°N. 99°-i24°W. (G.) Gibraltar (G.)

Augaptilus palumboi, Giesb. — One specimen, Agulhas current.

G.D. -Pacific 3°S. 99°W,, 1800 m. (G). North Atlantic,
52°N. i6°W., haul from 1905 m. ( T.).

Calanus brevic'omis, Lubbock. — Common south and west
of the Cape Colony, sparingly east of Natal.

G.D.— 4i°S. 45"E., S5°S. i8°E. and i°S. i°W. (L.) 2o°S.
38°\V. (G.), i4°-i8°N., i8°W. (C:e). Off Gibraltar (G.).

Calanus darwini (Lubbock). — Agulhas current.

G.D.— Pacific o°-i4°N. 91°— 97°W., and the tropical part
(G.). Malay Arcliip. (C:e). N.W. Indian Ocean, Arabian
Sea and Red Sea (various authors). East of Africa 8°N. (T.).
Atlantic 37°N. 46°W. (B.), i°S. i°W. (L.), 26°N. 34°W. (B.).

Calanus finmarchicus (Gunnerus).— Common south of the
Cape Colony, rare east and west.

G.D. — Port Townsend (T.). Along the west coast of South
America, from 8°N to " Jungfernkap " (G.), 37°S. i33°W. (B.),
Hongkong (G.) — Arctic Sea and North Atlantic to 39°N. in
the west and 29°N. in the east CC:e). Off Gibraltar, west
Mediterranean (G., C:e). Adriatic (Grobben).

i80

Calanus minor, Cls. — Very common south of the Cape Colony,
rarer west thereof and extremely rare in the east.

G.D.— Pacific 34°-o°N. 9i°-i32° W. (G). West of South
America, from Caldera to the Equator (G. )• North and west
of Australia (B.). S.E. of New Guinea (A.S.), Malay Archip.
(C:e). Indian Ocean, I4°-I3°-N. 6o°-54°E. (G.). Ceylon and
the north west Indian Ocean, Arabian Sea, Red Sea (T., A.S.,
G:e.). Atlantic : Benguela Current to Cape Verde (G., C:e).
North Equatorial and Antilles currents, Florida current to
48°N. 29°W. (C:e). Mediterranean (G.)

Calanus tenuicornis, Dana.— Sparingly east and west of South
Africa.

G.D.— Pacific. 40°N. I37°W. (D.), between 40°N. and 3°S.
(G.). .4//^n///c-, 22°-34°N.'53°-28°W. (C:e). Gibraltar (GO—
Mediterranean (G., C:e). 52°N. i5°W., haul from 1078 m. (T.).

Calanus vulgaris ( Dana). — Agulhas Current.

G.D.— Pacific 20°N.-8°S. (G.). Sandwich Islands, Fiji Islands,
N. and E. of Australia, New Guinea, Philippines (B.), Samoa,
Banka Straits (D ), Hong Kong (G.), Malay Archip. (C:e). —
N.W. Indian Ocean (C:e, A.S. and T.), Red Sea (various authors),
East Africa from Socotra to 28°S. (T.). — Atlantic, Gulf of
Guinea (T.S., C:e), Brazil Current (C:e), North Equatorial,
Antilles and Florida Currents to 46°N. 38°W., the Azores and
Cape V^rde Islands (C:e). — Mediterranean (various authors ;.

Calocalanus pavo, Dana — Sparingly round South Africa.

(;./).— East Pacific i2°N.-3°S. (G.) S.E. of New Guinea
(A.S.), Malay Archip. (C:e). N.W. Indian Ocean and Arabian
Sea (A.S. & T., C:e), Red Sea (A.S., C:e), East of Africa
28°S. (T.). — Atlantic, Gulf of Guinea (T.S., C:e), Equatorial
Antilles and Florida Currents to 56°N. I2°-26°W., Azores and
the Canaries (C:e). — Mediterranean (various authors).

Candacia bipinnata, Giesb. — Common south of the Cape
Colony.

G.D.— East Pacific 24°N.-3°S. (G.), West of Australia (G.).

Candacia chirura, C:e, n.sp. — West of South Africa, 250-350
m., sparingly.

Candacia catula, Giesb. — Agulhas Current.

G.D. — Pacific, Panama and 3°N. ii5°W. (C), Malay Archip.
(C:e). Indian Ocean from Ceylon to Socotra (A.S. & T.),
Arabian Sea (C:e), Red Sea (G., A.S., C:e).

Candacia curta, Dana. — Agulhas Current.

G.D.— Pacific Ocean, o°N. 88°W. and Caldera. (G.)— Indian
Ocean, Ceylon to Socotra (A.S. & T.), Arabian Sea (C:e),
Red Sea (G., C:e), 5o°S. 8i°W. (D.), 30°S. 38°W. (G.), Gulf
of Guinea (T.S.).

i87

Candacia inermis, C:e n.sp. — Agulhas Current, 530 m.

Candacia pachydactyla, Dana. — East and west of the Cape
Colony.

G.D.— Pacific o°-i2°N. 87°W.-iio°E. (G.). East of Aus-
tralia, Fiji Islands, Philippines (B.), South China Sea (D.)
Malay Archip. ^C:e). — Indian Ocean, from Ceylon to Socotra
(A.S. & T.), Arabian Sea (C:e), 5°S. 83°E. (Cx).— Atlantic,
South Equatorial Current (B.), North Equatorial and P'lorida
Currents (C:e).

Candacia tenuimana, Giesb. — Agulhas Current.
G.D.— Pacific, i6°N. i66°E. (G.), Mediterranean (G.).

Candacia truncata, Dana. — Agulhas Current.

G.D.— Pacific, 9°-i5°N. ii9°-i38°W. (G.), Samoa and Kmgs-
mill Islands, Sulu Archip., Ellice Islands' (D.). — Indian Ocean,
Ceylon to Socotra (A.S. & T.), Red Se'i (C:e) ; from o°S.
64°E. to 2o°S. 39°E. (T.), East of Africa 2o°-28°S. (T. , 3°-5°S.
SfE. (C:e). Mediterranean (T.).

Candacia varicans, Giesb. — Agulhas Current, 900 m. rr.
G.D. — Western Mediterranean (G.).

Centropages chierchiae, Giesb. — Round South Africa.

G.D. — Indian Ocean, Ceylon and Minikoi (A.S. & T.). —
Atlantic, 38°-48°N. io°-25°W. (C.:e), 52°N. ii°W. (T.),
Gibraltar ('6., a.o.). West Mediterranean (C:e).

Centropages furcatus, Dana. — Agulhas Current.

Q^U^ — West of America io°N.-io°S. (G.), east and north of
Australia (B.), S.E of New Guinea (A.S.), Philippines (B.),
Malay Archip. (C:e), Banka Straits (D.), from the Bay of
Bengal to Mozambique (T.j, Ceylon to Socotra (A.S. & T.),
Arabian and Red Sea (C:e, A.S., T.). — Atlantic, Gulf of Guinea
(T.S., C:e), Brazil Current (C:e), North Equatorial Current
(C:e), Mediterranean (T.). ^

Centropages typicus, Kroyer. — Agulhas Current and Agulhas
Bank.

G.D. — Atlantic : Gulf of Guinea, W. of Africa and Europe to
64°N. off the coast of Norway. Azores to the Newfoundland
Banks and 38°N. 74°W. in the south ; north of Iceland (C:e).
Mediterranean (various authors). Adriatic (Car.).

Chiridius poppei, Giesb. — Agulhas Current.

G.D. — West Mediterranean (G.).

Chirundina streetsi, Giesb. — Agulhas Current 780-900 m. rr,

G.D.— Pacific Ocean : 35°N.i25°W\ haul from 530 m. (G.).

i88

Clausocalanus arcuicornis, Dana. — Round South Africa.

G.D. — Pacific : West of South America to 53°S., off CaUfornia
(G.), 20°N.-26°S. (G.).— Indian Ocean : 6°N.-36°S., 92°-89°E.
(C:e). Malay Archip. (C:e). Bay of Bengal (T.). N.W.
Indian Ocean, Arabian Sea, Red Sea (C:e, a.o.). i6°S. 43°E.
(T.). — Atlantic : Benguela Current, Gulf of Guinea, Cape Verde,
area between Madeira, Gibraltar, the English Channel, 56°N.
31'^W., the east of New York and the Azores (C:e). — Mediter-
ranean (various authors). Adriatic (Car.).

Clausocalanus furcatus, Brady. — Sparingly east and west of
South Africa.

G.D.— Pacific : California Current 34°N. i3i°W. (G.). 2o°N.-
26°S. until i75°W. (G.)— S.E. of New Guinea (A.S.). Malay
Aichip. (C:e)— Indian Ocean : 8°-6°N. 72°-95°E., Arabian Sea,
Red Sea (C:e). Aden to 7°N. 77°E. (A.S.).— .4//^?////c : Brazil
Current, Equatorial Currents, Antilles and Florida Currents,
Sargasso Sea and Canary Current ; northern limit 40°-46°N.
(C:e). West Mediterranean (G).

Consea rapax, Giesb. — Agulhas Current, 900 m., rr.

G./J.— East Pacific if^.-fS. 99°-i32°W. (G.)— S.E. of
New Guinea (A.S.)— Atlantic : 52°N. i5°-i6°W\, haul from 270
and 900m. (T.), Faroe Channel 1060m. (N.)

Corycaeus danos, Giesb. — Sparingly east and west.

G.D.— Pacific : 34°N. i3i°W. and i9°X.-3°S. 88°-i75°W.
(G). — Malay Archip. (C:e). — Indian Ocean : Bay of Bengal
(T.), Ceylon to Socotra (A.S. & T.), Arabian Sea (C:e), Red
Sea (C:e., a.o). Mediterranean (T.).

Corycaeus furcifer, Cls. — Agulhas Current and West of
South Africa.

G.D.-^Pacific : ii°N-3°S. 99°-i24°\V. (G.).— Malay Archip.
{C:e).— Indian Ocean : Bav of Bengal (T.). Cevlon and N.W.,
Indian Ocean (A.S. & t.), east"^ of Africa" i6°-2o^S (T.).
Mediterranean (Cls., a.o.).

Corycaeus ovalis, Cls. — Agulhas Bank //., W. of South
Africa rr.

G.D. -Malay Archip. (C:e). Ceylon and N.W. Indian
Ocean (A S. & T). Arabian Sea (C:e) Red Sea (C:e, a.o.).
East of Africa i2°S. (T.)

Atlantic: Gulf of Guinea to Cape Verde, Brazil Current to
i6°S 34°W., North Equatorial, Antilles, Florida Currents to
45°N. i6°W., Azores (C:e). Mediterranean (Cls., a.o.).

Corycaeus robustus, Giesb. — Agulhas Current //'.

G.D.—East Pacific Ocean: 20°N.-7°S. (G.). S.E. of New
Guinea (A.S.), Malay Archip. (C:e). Indian Ocean, 6°N. 86-
95°E., Arabian Sea, Red Sea (C:e, A.S. & T.).

i8g

Corycseus speciosus, Dana.- -Agulhas Current.

G.D.— East Pacific 5°N.-25°S. until 137 E. (G.)— S.E. of
New Guinea (A.S.). — Indian Ocean : Bay of Bengal (T.).
From Cevlon to Socotra (A.S. & T.), Arabian Sea (C:e), Red
Sea (A.S:^, C:e), East of Africa 8°N.-i7°S. {T.).- Atlantic :
from St. Helena to the Gulf of Guinea and Cape Verde, Equa-
torial, Antilles, and Florida Currents until 42°N. 62°-5o°W.,
Azores and Canaries, exceptionally until 5i°N. 20°W. (C:e),
West of Ireland (T.), Brazil Current until 33°S. i5°-i6°W.
(C:e), Mediterranean (Q.).

Eucalanus attenuatus (Dana.) — Agulhas Current.

G.D.— East P^?f//7C i5°N.-i5°S. (G.), California Current 25°N.
no W. (G.), Kingsmill Island, China Sea. (D.) Malay
Archip. {C\e).— Indian Ocean from 7°N. 82°E. to 20°S. 39°E.
(T.), Ceylon and the N.W. Indian Ocean (A.S. & T.),' Red
Sea (C:e), 5°S. 83°E. (C:e), East of South Africa i2-i7°S(T.).
— Atlantic: Gulf of Guinea (T.S.), Mediterranean (various
authors), Bay of Biscay (L.), 52°N. i2°-i6°W., hauls from 900
and 1095 m. (T.), 56°N. i9°-i2°W. (T.), -N.N.W. of Ireland
(N.), Faroe Channel (N.), 39°N. 66°-64°\V. (C:e).

Eucalanus crassus, Giesb. — Agulhas Current. False Bay.

G.D. — Pacific: i9°-20°N. i38°E-i75°W., west of South
America i4°-26°S. (G.), Malay Archip. (C:e). — Indian Ocean :
Ceylon (A.S. & T.), Red Sea (A.S. & T.). — Atlantic :
Brazil Cunent (G.), Gulf of Guinea (T.S.), Mediterranean (G.),
Faroe Channel (Wolfenden), N. and N.E. of Scotland (T.S.),
Skagerak, exceptionally (C:e).

Eucalanus elongatus, Dana. — Agulhas Current.

G.D. — Along the west coast of South America from Valpa-
raiso northwards 6°-35°N. 8i°-i25°W., east Pacific i4°N.-3°S.
until i32°W. (G.), east of Kamtschatka (G.)-Sulu Archip., N. of
Celebes {D.).— Atlantic : Gulf of Guinea (T.S.), 33°N. 32°VV.
(C:e), W. of Gibraltar, Mediterranean (G.), 52°Nyi5°\V. haul
from 1006 m. (T.), 6o°N. 7°W. (T.S.), N. and N.E. of Scotland,
Moray Firth (T.S.), Skagerak 160 m. (Aurivillius).

Eucalanus monachus, Giesb. — Agulhas Current.

G.D.— Malay Archip. (C:e). Ceylon (A.S. & T.). Gulf of
Aden (C:e). Off Gibraltar, Mediterranean (G.).

Eucalanus mucronatus, Giesb. — Agulhas Current.

G.D.— Pacific i6^-i9°N. i75°W.-i66°E. (G.). Malay Archip.,
Arabian Sea (C:e).

Eucalanus pileatus, Giesb. — Agulhas Current.

G./J>.— West of South America io°N.-3°S. (G.). Ceylon,
Maldives and the Red Sea. (A.S. & T.).

190

Eucalanus subcrassus, Giesb. — Agulhas Current.

G.D.— West of South America io°N.-3°S., Amoy, Hong-
kong (G.)— Malay Archip. (C:e).— Ceylon, Minikoi. (A.S.&T.)
Red Sea (G., A. S. & T.) — Atlantic, Gulf of Guinea and off
Cape Verde (C:e).

Eucalanus subtenuis, Giesb. — Agulhas Current.

(;./;._East P^?c-/77V, i5°N.-26°S., until i38°W. (G.)- Malay
Archip. (C:e) — Indian Ocean, Ceylon (A,S. &T.). Arabian Sea
<C:e). Red Sea (A.S.). 8°N.74°E. (A.S.) -Atlantic, 6°-5°S.
8°E-i7°W. (C:e). Off Cape Verde (G., C:e).

Euchseta acuta, Giesb. — Agulhas Current.

G.D.— Mediterranean (G., A.S. & T.) Atlantic, 52°N.i6°W.,
haul from 2264m. (T.).

Euchaeta affinis, C:e, n.sp. — Agulhas Current.

Euchaeta longicornis, Giesb. — Agulhas Current.

G.D.— West of South America 3°-6°N. (G.). Malay Archip.
(C:e).

Euchaeta marina, Prestand. — East and West of South Africa.

G.D.— West of South America 20°N.-26°S until iio°E.—
South of Australia, Crozet Island (B.). Malay Archip (C:e).
Ceylon to Socotra (A.S. & T.). 3°S. 83°E. (C:e). Arabian
Sea, Red Sea (C:e). From Calcutta to Madagascar (T.).
Atlantic, 32°N.-38°S. (B.). South Equatorial and Brazil
Currents, North Equatorial, Antilles, and Florida Currents,
Sargasso Sea, Azores and the Canaries, rarely until 5i°N.-20°W,
(C:e), 49°-56°N. i9°-68°W\ (T.) Mediterranean (various
authors).

Euchaeta media, Giesb. — Agulhas Current.

G.D.— Pacific i6°N. i63-i66°E., 1500m. (G.)

Euchaeta spinosa, Giesb. — S.E. coast 33°S. 28°E. 71 m.
G.D.- Mediterranean (G.)

Euchaeta tonsa, Giesb. — Agulhas Current, 530 m. ;r.
G.D.-^Pacific, 35°N. i25°W., haul from 530 m. (G.).

Euchirella messinensis, (Cls.) — Agulhas Current.

G.D.— Indian Ocean, Minikoi (A.S. & T.). 23°S. 6i°E. (C:e).

? Gulf of Guinea (T.S.) — Mediterranean. (Cls., G.).

Euchirella venusta, Giesb. — Agulhas Current, 900 m. (length
of the anterior part of the body 4 millim., of the posterior part
I millim.).

G.D. — West of South America at the Equator (G.).

191

Gaetanus armiger, Giesb. — Agulhas Current and off Table
Mountain,

G.D.—Pacific : 5°N.-3°S. 99°-iis°W. {G.).—Athuiiic : Gulf
of Guinea (T.S.). 52°N. i6°W., haul from 1420 m. (T.).
Faroe Channel 1060 m. (N.)

Gaetanus miles, Giesb. — Agulhas Current 530 and 900 m.

G.D.— Pacific : 35°N. i25°W., fS. 99°W. (G.).— Atlantic :
52°N. i2°-i6°W. hauls from 900 and 1095 m. (T.). Faroe
Channel 1060 m. (N.).

Heterorhabdus abyssalis, Giesb. — Agulhas Current, 530-900
m., IT. Off Table Mountain, 450 m., rr.

G.D.—Pacific, i4°N. i32°W. (G.)— N.W. Indian Ocean (A.S.
& T.). Atlantic, 52°N. i6°W. haul from 1655 m. (T.).

Heterorhabdus austrinus, Giesb. — (Expedition Antarctique
Beige, Copepoden, p. 28, P. VI., 1902). — Agulhas Current, 530
to 780 m., n:

G.D. — Antarctic Ocean, yo^-'S. 84°W. haul from 250 m. (G.).

Heterorhabdus papilliger (Cls.) — Agulhas Current and off the
West Coast, rr.

G.D.—Pacific, i9^N.-3'^S. 175-99'^ W. (G.). — N.W. Indian
Ocean (A.S. & T.)— Gulf of Aden (C:e).— Mediterranean (Cls.,
G.).— Atlantic, 52''N. 16" W., haul from 2580 m. (T.).

Heterorhabdus spinifrons (Cls.).— East and west of the Cape
Colony, rr.

G.D.—Pacific, 3o°S. 9o''W. and i4''N. i6o^E. (G.).— Minikoi
and N.W. Indian Ocean (A.S. & T.)— Atlantic, 52''N. i5°-i6«
W., hauls from 664 and 1710 m. 55^-56"N. 32°-44°W. (T.).—
Mediterranean (Cls., G. ).

Heterorhabdus tanneri, Giesb. — Agulhas Current, rr.

G.D.—Pacific, 35''N. i25?W., haul from 530 m. (G.).

Labidocera acuta (Dana). — Agulhas Current.

G.D.—Pacific, 26°N. iio^W. and between io°N. and io°S.
(G.). Hongkong (G.). Port Jackson (B.).— Malay Archip.
(C:e). Indian Ocean, 8"-i3°N. 54''-75^E. (G.)— Bay of Bengal
(T.). From Ceylon to Socotra (A.S. & T.). Arabian Sea (C:e).
Red Sea (various authors). East of Africa, 9°S. (T.), ^7°S
49°E. (C:e).— Gulf of Guinea (T.S.).— Isle of Man (T.).

Lubbockia aculeata, Giesb.— West side of the Cape Colony,
277 m., rr.

G.D.—Pacific, o°N, 108° W. (G.). Red Sea (C:e). Mediter-
ranean (G.).

192

Lucicutia aurita, C:e, n.sp. — Agulhas Current, 900 m. n:
(C:e).

Lucicutia bradyana, C:e, n.sp. (? L. i^rainfis, Giesb.) — Agul-
has Current, 900m., it.

QD^ — L. gfandis has been found by Giesb. in the Pacific,

i°N. 83°W.

Lucicutia clausii, Giesb. — Agulhas Current, 530 and 900 m.

G./j).__ Pacific, 3°S. 99°W., 1800 m. (G.). Malay Archip.
(C:e). Mediterranean (G.)— Atlantic, 52°N. i5°W., haul from
918m. (T.).

Lucicutia flavicornis (Cls.). — Agulhas Current, 530 and 900 m.,
IT. West of the Cape Colony, 250 m. n\

G.D.— Pacific, i2°N.-3°S. 87°-i28°W. (G.)— Indian Ocean,
from Ceylon to Socotra (A.S. & T.). Red Sea (A.S., C:e)—
Atlantic : Antilles and Florida Currents until 40°N. 56° W.,
Sargasso Sea, Azores and 34°N. io°W., (C:e) — Mediterranean
(various authors).

Metridia brevicauda, Giesb.— Agulhas Current, 900 m. rr.

G.D.- -Pacific, i6°N.-3°S. 99°W.-i66°E. 1000 to 4000 m. (G.).

Metridia lucens, Boeck. — Abundant south and west of the
Cape Colony.

Qj)^ — Mediterranean (A.S. & T.). Atlantic, from Nova
Scotia to Ireland and the Faroe Channel, North Sea, Skagerak
and along the coast of Norway to 7i°N. (C:e). Iceland (T.).

Metridia princeps, Giesb. — Agulhas Current, 780 and 900 m.
11:

G.D.—Pacific, 3°S. 99°W. 1800 m. (G.)— Atlantic, Gulf of
Gascogne (r/V/f Giesbrecht), ^2°N. i6°W., haul from 2580 m.
(T.). 62°N. 56°W., (N.).

Metridia venusta, Giesb. — Agulhas Current, 530 m.

G.D.— Pacific, i6°-5°N. ii5°-i66°E. 450-1000 m. (G).
Atlantic, 52°N. i6°W., haul from 2825m. (T.).

Oithona nana, Giesb. — (Van Beemen : Publications de cir-
constance du Conseil intern, pour I'exploration de la mer
Nr. 7, 1903). — South of the Cape Colony.

G./).— S.E., of New Guinea (A.S.)— Malav Archip. (C:e).
Ceylon (A.S. & T.). Arabian Sea (C:e). Red Sea. (G.)—
Mediterranean (G.). Atlantic, 36°N. 6°W., (C:e), S.W. of
Ireland (T.), English Channel, coasts of Holland, North of
Jutland, West of Sweden (C:e).

193

Oithona plumifera, Baird. — E. and W,, of the Cape Colony.
False Bay.

G.D.— Pacific : 20°N.-3°S. 99°W.-i66°E., o°N. 9i°W. (G.).
New Guinea. (A.S.). Malay Archip. (C:e). Bay of Bengal to
20°S. 39°E. (T.). Ceylon to Socotra (A.S. & T.). Arabian
Sea (C:e). Red Sea(C:e., a.o.). Atlantic : 33°S-4o°N., North
thereof sparingly, North Sea, Skagerak, off the coast of Norway
into the Murman Sea (C:e).

Oithona rigida, Giesb. -C:e Kongl. Siv. Vet. Akad. Handl.
XXXV., n. 5 p. 45. P. v., f.7-18, 1901— Agulhas Current
97 m. n:

G.D.— Malay Archip. (C:e). Ceylon and Minikoi (A.S.&T.).
Gulf of Aden (C:e). Red Sea (G.).

Oithona similis, Cls. — -South of the Cape Colony, sparingly
(probably because the nets used were too coarse),

G.D. — Cosmopolitan, also in the Antarctic Seas, 7o°-7i°S.
8i°-92°W. (G.), and in the Polar Basin, off the New Siberian
Islands (G.O.S). Off Franz Josef Land (C:e).

Oncaea conifera, Giesb. — Sparingly round the Cape Colony.

G.Z).— Pacific, i2°N.-3°S. 87°-i32°W. down to 4000 m. (G.),
S.E. of New Guinea (A.S.), Malay Archip. (C:e), N.W. Indian
Ocean (A.S. & T.), Arabian Sea (C:e), Red Sea (A.S., C:e),
East of Africa, I2°-I7°S. (T.)— Antarctic Ocean 7o°-7i°S..
83°-9o°W. (G.).

'North Siberian Sea, 78°N. I36°E. (G.O.S.), West of Green-
land (Vanhoeffen), St. Lawrence Gulf (T.), North Atlantic,
7i°N. 2i°W. to 62°N. i°E. in 50 and more metres (C:e), Faroe
Channel, 900 m. (N.), south coast of Norway ( G.O.S. ),
Skagerak exceptionally fC:e), Azores rarely (C:e), Mediter-
ranean (G.).

Oncaea media, Giesb. — Agulhas Current and off Table Moun-
tain.

G.D.— Pacific, 5°N.-3°S. 99°-ii5°W. (G.), Malay Archip.
(C:e), from Ceylon to Socotra (A.S. & T.), Arabian Sea (C:e).
Red Sea (A.S., C:e).

From the east coast of the Cape Colony tc the Cape Verde
Islands, east of Rio Janeiro, Antilles and Florida Currents to
47'N. 4i^W., the Azores and the Canaries (C:e), Mediter-
ranean (G., a.o.).

Oncaea mediterranea (Cls). — Round the Cape Colony^
sparingly.

(7. Z>.— Pacific, i2°N.-3°S. 99°-i32°W., 34°N. i3i°-i32°W.
(G.), Malay Archip. (C:e), Bay of Bengal (t.), from Ceylon to
Socotra (A.S. & T.), Arabian Sea (C:e), Red Sea (C:e, A.S.),
East of Africa 8°N. (T.). — Atlantic : South Equatorial and
Florida Currents to 40°N. 22°W., Sargasso Sea (C:e), Medi-
terranean (Cls, a. o.), 52°N. i6°W., haul from 900 m. (T.).

C

1 94

Oncaea subtilis, Giesb. — South of the Cape Colony.

(?.£>.— Benguela Current 32*^8. i3°E. ; 39°-49°N. 6°-24"W.
(C:e), Mediterranean (G. C:e), Faroe Channel (Wolfenden,
vide G. ), Skagerak, once sparingly (C:e).

Oncaea venusta, Phil. — Round the Cape Colony.

G.D. — Pacific, 2o°N.-4°S. 87°W.-i37°E. (G.). S.E. of
New Guinea (A.S.), Malay Archip. (C:e). From 19'^N.
87°E. to 2o«S. 39«E. (T.), from Ceylon to Aden (A.S.).
Arabian Sea (C:e). Red Sea (C:e., A.S. )• East of Africa
i2°-28°S. (T.).

Wes^t of the Cape Colony (C:e). Brazil Current (G.;, until
37°S. 36°W. (C:e). North Equatorial and Florida Currents
to 4i°N. 57°-58°W., Azores and the Sargasso Sea. CC:e).
Mediterranean (various authors). 56°N. i8°-24°W. (T.).

Pachysoma tuberosum, Giesb. — Agulhas Current, 900 m, rt.

G.D.— Gulf of Panama ^G.)—3°3o°S. 83°3o°E., surface (C:e).

Paracalanus aculeatus, Giesb. — South of the Cape Colony,
sparingly.

G.D.—E. Pacific io°N.-io°S., until ii9°W. (G.). Hongkong
<G).— S.E., of New Guinea (A.S.. Indian Ocean, 6°N. 92°E.
(C:e). Arabian Sea (C:e). Red Sea (various authors) —
Atlantic : from Ascension to Cape Verde, North Equatorial,
Antilles, and Florida Currents to 4i'=N. 56°W. (C:e). West
Mediterranean (C:e.).

Paracalanus parvus, Cls.— South of the Cape Colony.

G.D.— West of South America, 55"-io°S. (G.)— East of
Australia (G.)— o°N. io8°W. (G.)- Hongkong (G.)— S.E., of
New Guinea (A.S.). From 9°N. 7i°E. to Aden (A.S.). Red
Sea (A.S.).

Atlantic : West of the Cape of Good Hope, Gulf of Guinea,
Cape Verde, Azores, Mediterranean, English Channel, Faroe
Channel, North Sea to 6i°N., Skagerak, south of Norway,
Kattegatt (C:e). West Baltic (Mo^bius.).

Brazil Current (C:e). North Equatorial and Florida
Currents to the Newfoundland Banks (C:e).

Phaenna spinifera, Cls. — Agulhas Current, 900 m. it.

G.D.— Pacific, 3°S. 99°W. and I4°N. i6o°E., in 500 m. (G.).
N.W. Indian Ocean (A.S. & 1\). Gulf of Aden (C:e). Red Sea
(A.S. & T.). From 3°S. 54°E., to 20°S. 39°E. (T.)— Medi-
terranean (G.).

Phyllopus bidentatus, Brady. — Agulhas Current, 900 m. rr.

G.D.— Pacific, 3°S. 99°W., 1800 m. (G.)— South Atlantic,
37°S. 46°W. (B.), Gulf of Guinea (T.S.), 52°N. i5°W., haul
from 1060 m. (T.), Faroe Channel, 1060 m. (N).

195

Pleuromamma abdominalis, Lubbock. — East and West of the
Cape Colony.

G.D.— Pacific : 36°N.i25°W., i6°N.-26°S. (G). Malay Archip.
(C:e). From Calcutta to Mozambique (T.). 23°S. 6i°E.
(C:e). Along the east coast of Africa i2°-2o°S. (T.) Ceylon
and the N.W. Indian Ocean (A.S. & T.). Red Sea (various
authors).

Atlantic : Gulf of Guinea (T.S., L.), Cape Verde Islands
(C:e). Mediterranean (Cls., G.). 40°-5o°N. 56=-3i°W. (C:e).
52°N. i2°-i6°W. (T.). From the north of Ireland to St.
Lawrence Gulf (T.). Faroe Channel (T.).

Pleuromamma gracilis, Cls. — Round the Cape Colony.

G.D.— Pacific, i6°N.-26°S. (G.)— Malay Archip. (C:e).
Indian Ocean, i3"-i4°N. 54°-6o°E. (G), from Ceylon to So-
cotra (A.S. & T.).— Gulf of Aden (C:e).— Atlantic : South of the
Azores to 40°-5o^N. i6°-64°W., (C:e). 52°N. i6°W., haul from
2825 m. (T.). — Mediterranean (G.).

Pleuromamma robusta, Dahl. — G. O. Sars. Crust, of Norway,
Calanoida, p. 115 PI. LXXVIIL, LXXIX, 1903.— East and
west of the Cape Colony.

G.D. — Arabian Sea, Gulf of Aden, Red Sea (C:e).— Atlantic :
North of Ireland (N.), Faroe Channel (Wolfenden), North of
the Faroe Channel, 200 to 400 m. (G.O.S.), West Norway, off
Aalesund (G.O.S.).

Pleuromamma xiphias, Giesb. — Agulhas Current, 530 to 900 m.
off Table Mountain, 450 m.

G.D.— East Pacific, 20^N.-3°S. 99°W.-i6o°E. (G.)— ? Off
Gibraltar (A.S. & T.).

Pontella securifer, Brady. — Agulhas Current, one single
specimen,

G.D.— Port Townsend (T.)— Pacific, 3S^-25°N. i38"-iio°W.,
i6°N.-2o°S. io9°W.-i65°E. (G.). Indian Ocean, 8°N. 72°E.
(G.). Ceylon and Minikoi (A.S. «&T.), 2°-3'*S. 83'^-84°E. (C:e).
Atlantic : Gulf of Guinea (T.S.), 5«-i8"S. 24°-4o'^W. (G.).

Pontellina plumata, Dana. — Agulhas Current.

G.D.— Pacific, 34°-i3"N. 97°-i3i°W., i5°N.-9°S. 8o°W.-i37°
E. (G.). KingsmiU Islands and i°-3°N. i73°E. (D.). Between
Port Jackson, Fiji Islands and the Philippines (B.). Malay
Archip. (C:e). From Ceylon to Socotra CA.S. & T.). Arabian
Sea (C:e). Red Sea (A.S., C:e), 2°S. 84- E. (C:e). East of
Africa, 25°S. (T.).- Atlantic : off the Cape of Good Hope (D.),
Gulf of Guinea (T.S.,a.o.), along the west coast of Africa to
Cape Verde, the Azores (C:e, a.o.). Mediterranean (various
authors). Brazil Current, 5°N.-25^S. 24°-44''W. (G.).

196

Pseudodiaptomus serricaudatus, T. Scott. — South of the Cape
Colony, as a rule common.

G.D.— Ceylon and the N.W. Indian Ocean (A.S. & T.).
Arabian Sea (C:e). Aden (A.S.). Tropical western Africa
(T.S.).

Rhinocalanus cornutus, Dana. — Agulhas Current, not rare.

G.D.— East Pacific, i5°N.-7°S., 138W. (G.). Suln Sea
(D.), Philippines (B.), Malay Archip. (C:e). Bay of Bengal
(T.). Ceylon and the N.W. Indian Ocean (A.S. & T.).
Arabian Sea (C:e). Red Sea (A.S.), East of Africa, 3°N.-20°S.
(T.).— Gulf of Guinea (T.S.). West Africa, i°-27°N., Canaries
(D,, L., T.), Mediterranean (G., a.o.). Antilles, and Florida
Currents to 420-43°N. 42°-45°W. (C:e), 52"N. i5°-i6°W.,
hauls from 1006 and 1438m. (T.). Faroe Channel (T.).

Rhinocalanus nasutus, Giesb. — East of Natal rare, south of
the Cape Colony abundant in 37-74 m.

G.D. — West of South America from Magellan Straits to
6°N. ; 2o°N. i73°E. (G.). Ceylon to Socotra (A.S. & T.).
Arabian Sea, rare (C:e), 2o°S. 39°E. (T.). Afuiiific, from
6°N. 22°W. to the Azores (C:e), Gibraltar, Mediterranean (G.);
from 46°N. 34°W. to 4o°N. 56°W. (C:e) ; 52°N. i6°W., haul
from 1438 m. (T.); N.N.W. of Ireland, 389 m. (N.) ; 65°N.
io°W. 250-400 m. (G.O.S.) ; North of Scotland (Mcebius) ;
Moray Firth (T.S.) ; North Sea, between Scotland and Norway
(G.O.S.) ; Skagerak in 120 and more m. (C:e).

Sapphirina angusta, Dana. — Off the west coast in 277 m. rr.

G.D.— 30°N. i34°W., i8°N. i45°W., Coquimbo (G.). Arabian
Sea (C:e), 43°-35°S. 78°-23^E. (B., D.). 27°N. 2o°W. (L.j.
West Mediterranean (G.).

Sapphirina gemma, Dana.— Agulhas Current, rr.

G.D.— East Pacific, 8°N.-33°S. 35°-27°N. i3i°-iii°W. (G.).
Arabian Sea (C:e). N.E. of New Zealand and south of the
Cape of (jood Hope (D.). 7°-ii°N., i7°-2o°W. (C:e).
Mediterranean (G., a.o.).

Sapphirina nigromaculata, Cls. — Agulhas Current, rr.

G./J.— Pacific, 2o''N.-4°S. (G.), Malay Archip. (C:e), Bay
of Bengal (T.). Ceylon, Minikoi, Socotra (A.S. & T.), Arabian
Sea (C:e), Red Sea (G., a.o.), o°-6°S., 64°-5i°E. (T.).-Gulf of
Guinea, Florida Current to 49°N. 23°W., Azores (C:e).

Sapphirina opalina, Dana. — Agulhas Current, rr.

G.D.--20°N. i73°E., Gulf of Panama (G.), Philippines (B.),
Red Sea (Steuer).— Gulf of Guinea (T.S.). i°S. i°W. (L.),
i"-4°N. I7"-25°W. (D.;, Mediterranean (G.,a.o.), Caribbean Sea
(C:e).

197

Sapphirina salpae, Cls. — Off Table Mountain 450 m. ;/-.

G.D.— 38°S. i8°W. (G.), Bay of Bengal (T.), Ceylon and
the N.W: Indian. Ocean (A.S. &T.), 5°S. 83°E. (C:e), Mediter-
ranean (G., a.o.)/.52°N. i6°W., haul from 1438m. (T.).

Sapphirina scarlata, Giesb. — Off the west coast, 350 m. /r.

G.D.— o°-6°N., 8o°-88°W. (G.), 3°S. 83°E., (C:e), Red Sea
(C:e.).

Sapphirina sinuicauda, Brady. — Off Natal, surface ; off Table
Mountain 450 m. n:

G.Z).— 6°N.-4°S. 8o°-89°W. (G.), 2°S. 84°E. (C:e), lo'^N.
i37°E. (G.), Philippines (B.), Ceylon (A.S. & t.), Red Sea
('Steuer), Gulf of Guinea (T.S.), Azores (C:e).

Scolecithricella minor (Brady) — G.O. Sars, Crust, of Norway,
Calan. p. 55, P. XXXVII, XXXVIII, i902.^0ff the west coast,
250-277 m. very rare. The South African specimens were only
1,2 millim. in length ; those from the Northern Atlantic
1,4 millim.

G.D.—lnd'mn Ocean— 47°N. 5o°E. (G.), Gulf of Guinea
(T.S.), 56^N. 28°-23°W. (T.), 67°N. 3^W. haul from 400 m.
(C:e), south and west of Norway to 68°N. (G.O.S.), Skagerak
(C:e).

Scolecithricella neptuni, C:e n.sp. — Off the west coast, very
sparingly in 250-350 m. (C:e).

Scolecithrix danse, Lubbock. — Off the east and west coasts.

G.D.—P'dcif\c 20°N.-26°S. (G.), East of japan (G.j, North
and East of Australia (B.), Malay Archip. {C:t), Bay of
Bengal to o°S. 64°E. (T.), Ceylon to Socotra (A.S. &"T.),
Arabian Sea (C:e).

32°S.-26^N. i3°-34'' (B.), Gulf of Guinea (T.S., C:e) to Cape
Verde Islands (C:e), i2°-37°N. i4--25°W. (L.), 34"N. 38°W.
(C:e), Canaries (T.), Mediterranean (G.), 52°N. i6°W., haul
from 2825 m. (T.), 52''-56°N. 22°-44PW. (T.).

Scolecithrix persecans, Giesb. — Agulhas Current in 900 m. n:

G.D.— 35''N. 125'^W., haul from 530 m. (G.), 52°N. i6°W.,
haul from 2580 m. (T.).

Solecithrix securifrons, T Scott. — Agulhas Current, 530-
900 m., n:

G.D.-^Gulf of Guinea (T.S.), 52°N. i5°-i6°W., haul from
918 and 1710 m. (T.), N.W. of Ireland (N.), Faroe Channel
down to 1060 m. (N.).

198

Temora discaudata, Giesb. — Agulhas Current, it.

G.D. — Entire Pacific between 2o'N. and 9°S. and 26°-35°N.
iio°-i3i°W. (G.). S.E. of New Guinea (A.S.). Malay Archip.
(C:e). Ceylon and the N.W. Indian Ocean (A.S. & T.).
Arabian Sea (C:e). Red Sea (G., a.o.). Mediterranean
(A.S. & T.).

Temora stylifera, Dana. — Off Natal in the surface, rr.

G.D.— Malay Archip. (C:e). From i9°N. %fK. to 2o°S. 39°E.
(T.). Ceylon, N.W. Indian Ocean (A.S. & T.). Arabian Sea
(C:e). Red Sea (A.S., C:e). From Ascension to the Azores,
off Morocco. Mediterranean. Antilles and Florida Currents
to 42°N. 42° W. Brazil Current (C:e).

Undechaeta major, Giesb. — Agulhas Current, 900 m., rr.

G.Z).— Pacific, 36°N. i25°W. and 2o°N. i73°E. (G.). Indian
Ocean, i9°N. 87°E. Gulf of Suez (T.).

Undechaeta minor, Giesb. — Agulhas Current, ;-.

G.D.— Pacific, 35°N. i25°W. and i6°-20°N. i66°-i73°E. (G.).
Minikoi and the N.W. Indian Ocean (A.S. & T.). Off Gibral-
tar (A.S. & T.).

Xanthocalanus fragilis, Aurivillius (Kongl. Sv. Vetensk.
Akadem. Handlingar XXX nr 3 p. 32, 1898. — A'. bojcalisG. O.
Sars in "The Norwegian North Polar Expedition, 1893-1896"
V Crustacea, p. 49, P. xi, 1900 ; Crustac. of Norway, Calan. p.
45, P. XXXI, XXXI 1, 1902.— T. Scott : 20th Report of the Fishery
Board for Scotland, III., p. 452, P. xxii, f. 8, 9, 1902. Compare
the systematic notes in the following.) — The west and south
Coast, rr.

G.D.— The north Siberian Sea, 78''N. I36^E. 50 m. ; off the
west coast of Norway, 59°N., 710 m. (G.O.S.). West Sweden,
50-120 m. (Aurivillius). Shetland (T.S.).

SYSTEMATIC NOTES.

Candacia cheirura, n.sp.

Diagn. of the female. — Terminal seta? of the third leg with
outwards turned apex and as long as the distal part of the
Re 3. The proximal part of the anterior antennae 7-jointed.
Genital segment with large projecting opening. Second joint
of the abdomen with a protuberance on the under side.
Proximal part of the setae of the furca broad. First pair of legs
with Si on Bi. Fifth pair of legs with tricuspidate apex and
with two spines on the exterior side. Length 2,7 mm.

199

Male. — Terminal seta of the Re of the third leg as in the
female. Proximal part of the anterior antennae 7-jointed. On
the prehensile antenna the 15th joint has distally on the upper
side a small spine, the 19th joint carries on its entire margin
close small ridges, the 17th and i8th joints not uniting, the
i8th carries strong transverse ridges, joints 15 + 16 ^ of the
joints 17 + 18. Fifth pair of legs: the left with forceps and
shorter than the right. Length 2, 3 mm.

Description of the female. — Colour. — The following parts are
brown : Re 2 and 3 of the first pair of legs, interior side of
Re 2 and the entire Re 3 of the second pair, Re 3 of the third
and fourth pairs, the terminal spines of the last thoracic
segment.

Length of the anterior part, 2 millim. ; of the posterior,
o, 7 millim. Greatest breadth i millim., in the middle of the
anterior part. First thoracic segment as long as the four others
together. Last segment symmetrical, ends in two spines, and
has on the sides one or two small bristles. Abdomen 3-jointed ;
genital segment as long as the two following segments together,
ventrally with a prominent large genital opening. The second
segment, nearly twice as long as the anal segment, ventrally
with a prominent rounded protuberance. Furcal branches
slightly longer than broad, with a short spine on the exterior
corner ; Se broad, slightly shorter than the four St, which
are broad and of nearly equal length. All very sparingly
hairy. Si thin, arises from the mterior distal end.

Anterior anteiince reach to the furca. Proximal part 7-jointed;
distal part 17-jointed. Length of the joints in o, 01 m.m. :

Aa ... 2 3 4 5 6 7 8 9 ID II 12 13 14 15 16

o, 01 m.m. 12 7 7 8 8 5 2 5 5 5 8 8 9 9 13

17 18 19 20 21 22 23 24-5

17 13 16 17 10 6 7 II

Posterior antenna'. — Re-j| as long as Ri ; Ri i & B2 2^ times
longer than broad. Re cylindrical, carries at its end i small
and 6 longer S. Ri carries on the exterior lobe 6 long plumose
Sa and one small Si, the mterior lobe 5 long plumose setae, and
three very thin bristles.

Mandible, as in C. varicans (see Giesb., F. Neap., P. XXI, f. 24).

Maxilla, similar to the maxilla of C. longiniana (see Giesb.
P. XXII, f. 7). Re 2 & 3, with i + 2 + 2 S.

Anteiior niaxilliped. — Li with one long and one very small S ;
L2 with 2 small S, L3 and L4 each with one spine and a small
thin S. B2 with 2 almost equal strong spines inserted at some
distance from each other and slightly shorter than the 3 falci-
form terminal spines.

200

Postetior MaxilUped. — Bias long as B2 and Ri together;
B2 as long as Ri. S as in the other species. S2 of B2 a little
shorter than Si and S3.

Natafoiy legs, 1-4 pair. — Bi of the first pair with Si. Distal
part of Re 3 of the ist pair longer than the proximal, — St of the
3rd pair as long as the distal part of the joint. Serration of
the edge of Re fine.

Fijtii p:ur of legs. — B2 with small Se, Ri without interior
setfe ends in three apices, of which the median is the largest ;
besides two spines on the exterior margin.

Description of the male. — Colour. — The ends of the last
thoracic segment, the comb on the prehensile antennae and the
Si, St and, partly, the third joint of Re of the natatory legs have
brown chitin.

Length : — anterior part 1,8, posterior part o, 6 millim. ; 2^
times as long as the greatest breadth, in front of the middle.

Anterior part 5-jointed. Last thoracic segment asymmetrical,
the left corner being prolonged. It carries on the side some
small bristles. Of the five abdominal segments the three first
are of about the same length and twice as long as the fourth
and the anal segments. The first joint carries on the right side
a hornlike apophysis (as in C. cnrta and C. pectinata). Furca
short and broad. Se and the four Si of about the same length,
all of ordinary form and plumose. Si half as long and thin.
On the distal exterior corner is a small spine.

Prehensile antenna very similar to that of E. arniata and C.
varicans ; its proximal part 7-jointed. Joints 15 + 16 | as
long as 17 + 18. The joint 15 has distally on the upper side
a small spine. The joints 17 and 18 do not unite. The joint
18 carries a comb of large and coarse ridges, the joint 19 along
its entire length very fine and close ridges.

The stli pair of legs closely resembling that of C. varicans.

Candacia inermis, n.sp.

Diagn. of the female. — Terminal setae of the third pair of legs
much shorter than the distal part of Re 3. Proximal part of
the antennae 7-jointed. Last thoracic segment rounded. Fifth
pair of legs ends in three teeth of which the distal is the
longest ; there is, besides, somewhat proximal from the middle,
another tooth. Length 3,3 millim.

Male unknown.

20I

Description of the female. — Body, anterior pari 2,5 millim. in
length, I millim. in breadth, posterior part 0,75. The last
thoracic segment with rounded ends. Abdomen symmetrical.
Genital segment symmetrical as long as broad and longer than
the second and third together. Furcal branches somewhat
longer than broad. Se and the four terminal setae of equal
length, three times longer than the Si.

Anterior allien iicB do not reRch to the end of the furca, 24-
jointed, its proximal part 7-jointed. Length of the joints in
0,01 millim.

Aa ... 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16
o, 01 m.m. 27 9 9 9 9 9 6 8 7 8 11 12 17 19 18

17 18 19 20 21 22 23 24-5

17 17 14 14 10 7 II 18

Posterior antenna. Ri li times longer than Re + Bi, three
times longer than bioad.

Mandible as in the other species, the masticatory apparatus
M-ith one strong exterior tooth, by a gap separate from two
smaller ones.

Maxilla. Li 2 reaches only slightly beyond the Ri, which
carries 1 + 2 + 2 S. Re with 6 longer plumose setae.

Anteiior niaxilliped. Bi li times longer than B2. Li with
3S. L3 and 4 each with 2S, one strong and one small. B2
with 2 falciform strong spines of about equal length and
breadth inserted at some distance from each other. The falci-
form spine of Ri only slightly longer than those of B2.

Natatory legs. Bi of the ist pair with a short Si, B2 with a
Si as long as the Ri ; distal part of Re 3 twice as long as the
proximal. Re 3 of the third pair i^ times longer than Re i +2,
its distal part half as long as the joint, its edge coarsely denticu-
lated proximally, but finely serrate distally. Terminal seta
falciform, not half as long as the distal part of the joint. Fourth
pair, as usually, without Si.

Fifth pair of legs. Bi + B2 as long as Ri. The latter with a
tooth in the middle of the exterior edge and 3 teeth at the end,
the distal one the longest. There are no Si.

Euchaeta affinis, C:e, n.sp.

The only female of this species, which belongs to the group
of E. norvegica, measured in length 6,3 millim. (anterior part
4,5, posterior part 1,8 millim.). The frontal organs occurred
on a moderate elevation not protracted into a conical pro-
jection. The corners of the last thoracic segment carried a
tuft of hairs, but were not provided with an apiculus. The
abdomen was naked, and its first segment carried in the
middle the strongly projecting genital opening, enclosed

202

between dissimilar plates. There was no elevated ridge in
front of the genital opening. The ist pair of natatory legs
had a two-jointed Re, Re i and 2 uniting into one single joint
with concave exterior margin, which carried in its distal end a
Se as long as Re 3. The sinus between the Se 2 and Se 3 of
the Re 3 of the 2nd pair of legs was larger and deeper than
that between Se i and Se 2 of the same joint. The Se of Re 2
of the 2nd pair reached to the basis of Se i of Re 3, was much
larger than Se i of Re 3, and as long as Se 3 of that joint.

Lucicutia aurita, C:e, n.sp.

Diagn. of the Female. Length of the body 8 millim. The
head broad, on each side with a short triangular spine.
Anterior antennae reach to the end of the furca. St of the ist
pair of legs as long as Re 3. St of the 5th pair of legs much
shorter than Re 3. Si 2 of the same pair of ordinary form, not
thicker than the ether.

Description of the Female. — Body. Anterior part 4,5, posterior
part 3,5 millim. in length. Its form is oblong, only slightly
broader in the middle than at the ends, 3^ as long as broad.
Head broad in the front, on each side wdth a short but broad
spine. Abdomen 4-jointed, its genital segment as long as the
two following together, which are of equal length. Anal
segment as long as the preceding. Furca as long as the three
first abdominal segments together. Se nearly in the middle.

Antoior antcnnxB reach to the end of the furca, 25-jointed.
Length of the joints in o, 005 millim.

Aa 2 3 4 5 6

7 8 9 10 II 12 13

14

15 16

6 6 10 6 6

6 7 7 II 12,512,520

25

27.5 27,5

17 18

19 20 21 22 23 24

25

26 27

30 25 21 21 20 21

9

Posterior antciuioB.

B I with one Si as lone

[ as

B2. B2

twice as long as broad, with one Si. Re as long as Ri,
8-jointed. Re 8 with one plumose Si, inserted at the basis of
the 3 terminal ones and much shorter than them. Ri i nearly
4 times longer than broad ; its two Si inserted nearer to the
distal than the proximal end ; the exterior margin of the joint
hairy of stiff, small bristles. Ri 2 + 3, half as long as Ri i ; Le
with 5 long plumose S and i Si. Li with 5 long and 3 short S.
The Le carries some small bristles.

Mandible elongate ; B with 3 plumose Si, reaching to Ri.
Re somewhat longer than Ri, 3 times longer than broad. Ri,
3-jointed ; joints of equal length. Ri i and 3 outside with
some small stiff bristles. Re carries 6 plumose S and Ri i four,
Ri 3 2 + 8 long plumose S. The masticatory apparatus was not
examined.

:o3

Maxilla. Similar to that of L. jiavicornis (Giesb. F. Neap,
P. XIX., f. 9) ; Le I with 5 long plumose Si ; Re twice as long
as broad and with 5 thinner and 6 coarser, plumose S. — Ri with
5+4, plumose S. B2 with 3 S. — Li 3 cylindrical with a
small spine on its exterior end and 3 plumose long S. Li 2
also cylindrical and with 3 plumose S. Li i twice as long as
broad, with strong, almost straight, sparingly hairy spines.

First max illiped simihi- io that of L. fiaviconiis (Giesb. F.
Neap. P. XIX., f. 11).

Posterior maxilliped. B2, li times longer than Bi and as
long as Ri, with 5 plumose S, of which S4 is somewhat shorter
than Si— 3 and S5, which latter is the longest, but shorter
than the joint. Of the 5 joints of the Ri the 2nd is the longest,
all joints naked on the inside.

Natatory legs of the ist-4tli pairs. Re and Ri of all pairs 3-
jointed. Bi with straight margins, of the 2nd-4th pairs with
a short plumose Si. B2 with convex interior margins ; it
carries on the ist pair a short cylindrical processus, from
which issues a Si, and at its basis some stiff hairs. Ri of the
ist pair at a higher level than Re ; but on the other pairs at the
same level. Ri of all pairs reach to the distal margin of Re 2.

Proportional length of the 3 joints of Re and St.

ip.

2p.

3P.

4P.

Re I

10

10

10

10

Re 2

7

9

10

9

Re 3

12

25

25

25

St.

15

15

15

12,5

Se of Re 1,1,2 on the first pair, 1,1,3 on the other, longer on
the ist pair than on the other. St of all pairs with finely
serrate edges (about 18 teeth in o,oimillim.). Si 1,1,4 0'"i ^he
first 1,1,5 o" the other pairs. Re i and Re 2 hairy on the
inside. The Se of the 2nd-4th pairs are enclosed between a
pair of spinules.

The three joints of Ri of equal length. Ri 2 ends on all
pairs in a small spinule. Se 0,0,1, Si 1,2,4 o^"^ the first 1,2,5 on
the other pairs. Exterior margin of Ri i and Ri 2 of the 2nd-
4th pairs hairy.

Fifth pair of legs. — Bi without Si. Re and Ri 3 jointed.
Re more than twice as long as Ri ; Se 1,1,2 ; Si shorter than
Re 3 ; Si 0,1,3, ^^ equal thickness. The 3 joints of Ri of equa,
ength ; Se 0,0,1 ; Si 1,1,4. Interior margin of Re i and Re 2
nd the exterior margin of Ri i and Ri 2 hairy.

204

Lucicutia bradyana, C:e, n.sp.

Diagn. of the female. — Anterior antennie reach to the end of
the furca. Ri of the ist pair of legs 3-jointed. St of the ist
pair of legs nearly as long as Re 3, St of Re of the fifth pair of
legs shorter than Re 3. Si of the Re 2 of the 5th pair of legs a
stout spine as long as Re 3. Length 5,7 millim,

Diagn. of the male.— St of Re 3 of the ist pair of legs nearly
as long as Re 3. Ri 2 & 3 of the right leg of the fifth pair loaf-
like, with 6S. B2 of the left leg on the interior margin produced
into a short spine. Length 5,5 millim.

Brady (Rep. Challenger VIII. i. p. 50 P. XV., f. 1-9, 16
1883) has identified a copepod, found by the Challenger
Expedition, with Leuchartia flavicornis of Claus. This, how-
ever, cannot be correct as the differences, especially as
concerns the size, are too considerable. One female which I
found in the gathering off Port vShepstone, 900m., seems to me
to agree sufficiently with Brady's species. I found in the same
sample one specimen of a male of about the same size as the
female, and believe they belong to the same species. There are
between the male of my form and of Brady's serious
differences, which make the identification somewhat doubtful.
The right leg of the fifth pair carries on the Ri 4 seta? in
Brady's form, but 6 in my specimen. This may perhaps be
accounted for by the supposition that Brady's specimen had
lost two setaj. The other difference is in the B2 of the left
legs of the fifth pair. It was on my specimen produced into a
spine, but in Brady's specimen into a denticulated processus.
The difference depends perhaps on the age.

The Lucicuiia grandis, Giesbrecht, is doubtless very closely
allied to both Brady's and my form. The male only is known
and differs principally in the B2 of the fifth pair of legs.

The variation in the forms of the group of L. clausii is,
according to Giesbrecht, considerable, and it therefore seems
very probable that Lcitckaiiia fiaviconiis, Brady (von Claus),
Lucicutia i>rniidis, Giesb., and Lucicutia bradyana C:e represent
only variations in one and the same species.

Description of the female. Body in length 5,7 millim
(anterior part 3,5, posterior part 2,2 millim), 3^ times longer
than broad (greatest breadth in the middle of the anterior
part). Head broad, on each side with a slight angular pro-
minence.— Abdomen 4-jointed, genital segment as long as the
two following together, and as long as the inflated abdominal
segment. Furca at least 5 times longer than broad, and as long
as the two last segments together. The furca has two Se, one
very small near the anal segment and another in the middle,
reaching to the end of the furca.

205

A II f L'rior niifcmice reach, to the end of the furca, 25-johited.
Length of the joints in 0,005 millim.

Aa 2
6

3
7

45678
6 6,5 6,5 6 8

9

7

TO

7

II

9

12
9

^3
15

14
18

15
18

16 17 18 19 20
20,5 20,5 23,5 23,5 18

21

18

22

15

23
18

24
18

25
9

Posterior ajitennce. Bi with i Si ; B2 twice as long as broad
with iSi. Re as long as Ri and 8-jointed ; Ri i three times longer
than broad, naked on the outside. Re 2+3 | as long as Re i.
The outside of Le carries some small bristles. Else as in L.
aiiritn.

Mandible. Re i| times longer than Ri, Ri 3 carries in its
top a row of small stiff hairs. Else as in L. auriia. The
masticatory apparatus carries four strong, almost straight teeth
and towards the interior side five smaller ones.

Maxilla similar to the maxilla of L. aurita.
Anterior niaxilliped as in L. aurita.

Posterior maxiUiped. B2 with 4 Si, of which Si i and 3 are
slightly shorter than the other. Interior margin of B2 with a
row of small stiff hairs. Ri 1,2 och 3 with a row of stiff hairs
at the basis of the setaa. Else as in L. aurita.

Natatory legs of the i-4th pairs. Ri on a slightly higher level
than Re, B i and 2 of the first pair hairy on the interior
margin. Proportional length of the joints of Re and St.

I p. 2 p. 3 p. 4 p.

Re I .. 10 10 10 10

Re

2

...

7

7,5

9

7,5

Re

3

...

17

20

25

20

St

...

15

10

10

as

in L.

aur,

it a.

>

Fifth pair of legs. — Re and Ri 3-jointed. Re more than twice
as long as Ri. St shorter than Re 3, Se 1,1,2. Si 0,1,3 J
Si 2 a strong thick spine, as long as Re 3. Of the three joints
of Ri the Ri 3 is the shortest. Ri 2 ends in a small spine. Se
0,0,1. Si 1,1,4. Outside margin of Ri i and Ri 2 hairy.

Description of the male.— Body in length 5,2 millim. (anterior
part 3, posterior part 2,2 millim.), nearly 4 times longer than
broad. Abdomen 5-jointed ; anal segment li times longer
than the preceding. Furca about six times longer than broad
and as long as the three last segments together, with a small
Se near the anal segment and a longer one in the middle.

\

I p.

2 p.

3P-

4P

10

10

10

10

7

7

10

10

15

18

20

20

13

8

10

8

206

Anterior a iitennce T:G3,ch to the end of the fiirca (the distal
part of the prehensile antennae was lost in my specimen ). The
right antenna 25-jointed. Length of the joints in 0,005 millim.
2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 18 19 20
3>5 4>5 7;5 7.5 6 6 6 6 6 9 9 14 14 18 18 18 18 20 16
21 22 23 24 25
16 15 15 15 9

Posterior antennce as in the female.

Mandibtc. Ri 2 and 3 uniting, else as in the female.

Maxilla as in the female.

Fi)st niaxilliped as in the female.

Second maxilliped. — Bi somewhat shorter than B2. Ri as
long as B2. Else as in the female.

Natatory legs of the i-^th paiis. — Proportional length of the
three joints of Re and Si : —

Re I
Re 2
Re 3

St
Else as in the female.

Fifth pair of legs similar to that of L. grandis and L. clausii.
Ri of the right leg loaf-like, with 6 sets. The margin and the
flat side with patches of small stiff hairs. B2 of the left leg
protracted mside into a spine-like processus.

Scolecithricella neptuni, C:e, n.sp.

Diagn. of the female. Head without crista, anterior antenna
22-jointed. Teeth of the masticatory apparatus of the mandible
with close long and stiff hairs. Spinules on the natatory legs few
and delicate. Fifth pair of legs 3-jointed, the third joint ending
in three triangular spines, two articulating and with some few
spinules, one hairy. Length 1,4 millim.

Diagn. of the male. Head without crista. Anterior antennae
23-jointed. Masticatory apparatus and natatory legs as in the
female. Fifth pair of legs : left and right leg of equal length,
so also the Bi ; B2 of both legs of about the same thickness
throughout. The right leg with rudimentary Ri and a falci-
form Re I as long as B2. The left leg with a long curved Ri
and a short 3-jointed Re. Length 1,2 millim.

Description of the female. — Body in length 1,4 millim.
(anterior part 1,1, posterior part 0,3 millim.), clumsy, its
anterior part being highly vaulted, 5-jointed (length of the
joints 0,5 ; 0,18 ; 0,15 ; 0,12 ; 0,13 millim.). Abdomen short,
the genital segment as long as the three following together.
Furca slightly longer than broad.

207

Anterior antennce reach to the last thoracic segment, 22-
jointed, Aa 8-10 and 24-25 uniting. Length of the joints in
0,005 milHm.

Aa. 234567 8-10 II 12 13 14 15 16 17 18 19 20 21
3 I I I I I 4,5 II 1,5 1,5 1,5 1,5 1,5 1,5 I I I

22 23 24-25

1,5 2 4,5

Posterior antennce. B 2 with i S ; Re twice as long as Ri
and thicker than it. Re 2 twice as long as Re i and Re i -|- 2
as long as the last joint. Ri i twice as long as Ri 2+3 and
three times longer than broad. The Si reaches to the end of
Ri 2+3. The Ri 24-3 twice as long as broad. Le with 54-1
seta?, Li with 54-3 seta?. Exterior margin of the joint hairy.

M audible e\ongAiit\ Re inserted in the middle of B2, which
carries i Si reaching to the end of Ri, Re longer and broader
than Ri. Teeth of the masticatory apparatus, especially the
interior ones, covered with dense long and stiff hairs, giving it
the appearance of a brush.

First maxilUped stout with the lobes crowded distally. Li
with 3 long spiniferous and i somewhat shorter hairy seta.
L2 and 3 with 2 longer and i shorter seta. L4 with i stout
falciform and i thinner seta. B2 carries i very strong falci-
form seta, I thinner but of nearly the same length, and a third
somewhat shorter and much thinner seta. The setas of Ri are
all soft and flexible.

Second niaxiUipcd. Bi and B2 of about equal length ; Ri ^
as long as B2. Bi with 14-24-14-2 + 3 settu, B2 with 2)+ 2
setaj ; of th:; first group the distal is the longest and the median
the shortest. Ri 5-jointed, Ri 2 the longest and Ri 5 the
shortest. Ri i carries 4 seta3.

Natatory legs. In my specimen the 3rd and 4th pair were
lost. B2 of the ist pair with i plumose Si. Re 1,2 and 3 of
about equal length, Se 1,1,1 longer than the joints, narrow and
with stiff hairs. St similar to the Si. Si 0,1,3, jointed in the
middle. Ri reaching to the middle of Re 2, with a lobus on
the exterior side and 2-1-3 Si, the 3 latter being jointed in the
middle. Bi of the 2nd pair with i plumose Si. Ri reaches to
the distal margin of Re 2. Proportional length of the three
joints of Re and St 10:12:18:18. Se 1,1,3. ^€ i comparatively
short, the other as long as the joints are broad. Si 1,1,4. Ri 2
nearly three times as long as Ri i ; Se 0,1 ; Si 1,4. The fiat
side of Ri carries near the Si a group of some few very delicate
spinules.

Fifth pair of legs symmetrical, 3-jointed ; the third joint,
three times longer than the other together, carries in its top
3 strong triangular teeth, the two exterior not articulating and
with some few spinules, the third articulating and hairy.

208

Description of the male. Body in length 1,2 millim. (anterior
part 0,8, posterior part 0,35 millim.). Anterior part 7-jointed.
Abdomen 5-jointed ; the anal segment very short, the other of
about equal length and breadth. Furca scarcely longer than
broad. The longest furcal setae as long as the abdomen.

Aiiicriot antenna reach slightly beyond the last thoracic
segment, 23-jointed or 21-jointed, the Aa 8-10 uniting in
younger specimens.

Posterior antciDUV and tlic mandible as in the female.

Maxilla. Le i with i shorter and 6 longer plumose sette.
Li as prominent as Ri, with about 7 nearly straight hairy
coarser set^e and several finer ones. Li 2 with 2 strong and
longer seta^, i shorter seta. Li 3 with 1 +4 setae. Re is a slight
protuberance with 4 comparatively short setaj. Ri 3-jointed,
with 2,2, 4+ I setae.

First and second tnaxilliped as in the female.

Natatory legs of the i-4th pair. Re 3-jointed ; Ri of the
ist p. i-jointed, of the 2nd 2-jointed, of the 3rd and 4th
3-jointed. B i of the ist pair without Si, of the 2nd-4th with a
short Si. B 2 of the ist pair with i Si. Ri of the ist and 2nd
pair reach to the distal margin of Re 2, beyond it in the 3rd
and 4th pairs. Se of Re of the ist pair 1,1,1, more than twice
as long as the joints, on the other pairs 1,1 ;3, short and
triangular. St of the ist pair bristle-like, of the other pairs
straight and knife-like, in the 2nd and 3rd pairs longer, in the
4th pair shorter than Re 3 ; their edge rather coarsely serrate
(teeth about 4 in 0,01 millim.). Si of the ist pair 0,1,3 ; of the
others 1,1,4. Ri of the ist pair with 5 setae. Se of the 2nd
pair 0,1, of the 3rd and 4th 0,0,1, but their Ri 2 ends in a
small apiculus. Si of the 2nd pair 1,4, of the 3rd and 4th
pairs 1,1,4. Small, delicate spinules on the fiat side of the legs
occur in the 2nd pair on Ri 2 (a set of 3-4 near the distal and
proximal part) and on the distal part of Re 3, in the 3rd pair
in the distal part of Ri 3 (a row of 6 small bristles) and on the
distal part of Re 3, on the 4th pair in Ri 3 and at the distal
margin of Re 2.

Fifth pair of legs nem-\y ^s long ds the anterior part of the
body, the right and left leg of equal length. Bi of both legs of
equal length. B2 of the left leg slightly shorter than of the
right one, both of about equal breadth throughout. The B2
of the right leg carries near its distal end a small tubercle
(rudimentary Ri) and in the end a falciform Re, | as long as
B2. The B2 of the left leg carries a curved Ri, longer than
B2, and a 3-jointed Re, half as long as B2.

209

The species in question differs in so many important
cliaracteristics, especially in the 5th pairs of legs and the
masticatory apparatus, from the type-species that I am
inclined to form for it a new genus, for which I should like to
propose the name Psc'iuloH'olfciflirix.

Xanthocalanus fragilis, Aurivilhus.

Gicsbicclif has described from the Mediterranean two species
of this genus, viz., A', (igilis and A', minor, which, according to
G. 0. Sars, however do not represent more than one species,
A', minor being founded on young specimens. Aurivilliiis
added in 1898 two other species, X. fi{ii>ilis and X. simpUw. In
1900 G. 0. Sars described from the Polar Basin a form found
by Xanscn and named A', boicalis. .So far I can see there is no
essential difference between X. fiagilis and A', boreal is. The same
form was found by T. Scoff at Shetland, and the specimens I
found in the South African Seas are evidently of the same kind,
their fifth pair of legs being 2-jointed. Later Professor G. 0.
Sc//'5 describes more fully in the Crustacea of Norway Calanoida,
A', borealis, which, in full-grown specimens, has 3-jointed 5th
pair of legs with 4, not 3 terminal spines. Besides, the right leg
of the 5th pairs of the male is rudimentary or wanting. In the
species of Aurivilhus the right leg is nearly as long as the left
one. It seems to me possible that two nearly allied species have
been confounded. In all cases the specimens from South
Africa did not differ essentially from the description of
Aurivilhus. All my specimens were females, in length 2,9
millim. (Norwegian specimens 3,50 millim, from the Polar
Basin 4 millim, according to G. O. Sars ; Aurivilhus gives no
measures for the female, but for the male 2,5 millim).

I. C.Thompson has described under the name of A', gies-
brechfi (Ann. Mag. Nat. Hist. [7] XII. p. 22, P. IV., f. 1-9,
1903) another apparently very nearly allied form, found
west of Ireland in a haul from 2580 m. In case the illustration
be correct one cannot identify it with X. fragilis as the anterior
antenna reach beyond the furca, but in X. fragilis scarcely
beyond the genital segment.

There were in the collections I have received still some few
new species, in parts belonging to new genera, but the rare
specimens I have hitherto found were too incomplete for a
full description.

2IO
EXPLANATION OF PLATES.

Plate L

Fig. I. Cainiacia chinira, Cleve, 9 abdomen lateral (X 40).
„ 2. ,, ,, „ $ furca (two of the exterior setse broken-

In all other specimens the three remaining setce were also broken,
which gave the furca the appearance of a hand. Therefore the
name), (x 85).
,, 3. Camiacia cliinini, Cleve, $ abdomen dorsal (x 70).

4. „ „ „ $ furca (X 170).

,, 5. „ ,, „ $ prehensile antenna (distal pai't) (X 170).

6. „ „ „ ^ 5th pair of legs (X 170).

Plate 1 1.

„ 7. Canclacia chlrura, Cleve, $ leg of the 5th pair (x 170).

8. „ „ „ 9 leg of the 3rd pair (X 85).

9. „ „ „ 9 mandible (X 85).

,, 10. „ ,, ,, 9 maxilliped of the 1st pair (X 85).

,, II. Caudacia iuenitis, Cleve, 9 abdomen ventral (X 40).

„ 12. ,, „ „ maxilliped of the 1st pair (X 40).

,, 13. ,, „ ,, masticatory apparatus ( X 170).

Plate III.

,, 14. Candacia incniiis, Cleve, proximal part of the ist antenna (X 85).
„ 15- „ n ., leg of the 5th pair (X 85).

„ 16. „ „ „ legof the 3rd pair (X 85).

,, 17. Eiicluvhi affiiiis, Cleve, 9 genital segment (X 40),
„ 18. „ „ ,. /?t' 3 of the 2nd pair of natatory legs (X 85).

19. „ „ „ leg of the ist pair (X 85).

Plate IV.

,, 20. Scohcitlincclla ncptuiii, Cleve, $ lateral (X 40).

21. „ „ „ 9 antenna of the 1st pair (X 170).

22. „ „ ,, 9 antenna of the 2nd pair (X 170).

23. „ „ „ 9 mandible (X 170).
„ 24. „ „ „ $ maxilla (X 300).

( Two set:c of Li 3 broken.)

Plate V.

„ 25. Scolt'ciilniciila neptiini, Cleve, 9 ist maxilliped (X 300).

„ 26. „ „ ,. 9 2nd maxilliped (X 300).

„ 27. „ „ „ 9 leg of the ist pair (X 170).

„ 28. „ „ „ 9 legof the 2nd pair (X 170).

„ 29. „ „ „ ,? legof the 3rd pair (X 170).

PL.-iTE VI.

„ 30. Scolccitliriciila iifptiiiii, Cleve, 9 leg of the 5th pair (X 500).

„ 31- H " " ^ 5th pair of legs (X 500).

„ 32. XiVitIioca!niiiisfnt!^ilis,Am-i\; 9 leg of the 5th pair (X 170).

„ 33. Litcictitia bradyana, Cleve, 9 ^<^ - of the 5th pair of legs (X 85).

„ 34- .. " " ^ 5tli pair of legs (X 40).

Marine Investigations,
South Africa.

Plankton. PI. I.

Marine Investigations,
South Africa.

Plankton. PI. II.

8.

Marine Investigations,
South Africa.

Plankton. PI. III.

n.

Marine Investigations,
South Africa.

Plankton. PI. IV.

Marine Investigations,
South Africa.

Plankton. PL V.

2 9.

^5.

Marine Investigations,
Soiith Africa.

Plankton. PL VI.

53.

THE ALCYONARIA
OF THE CAPE OF GOOD HOPE

PART II.

BY

SYDNEY J. HICKSON, ALA., D.Sc, F.R.S.,
Beyer Professor of Zoology in the Victoria I'niversity of Manchester.

Two parcels of Alcyonaria were sent to me in the year 1900.
With the exception of a dried specimen of Gorgonia albicans^
all the specimens were, as before, carefully preserved and in
excellent condition for histological examination.

I regret that a heavy pressure of work has prevented me from
completing my anatomical description of the species, but rather
than delay the statement as to the species occurring in Cape
waters, I have put together in this preliminary paper the notes
I have made upon the species.

In these two consignments there are twelve species of
Alcyonaria and one of Stylasterina. Very striking pecu-
liarities in the structure of two Gorgonacea have necessitated
the foundation of two new genera — Malacogorgia and Tricho-
gorgia — and, in addition, four species of old-established genera
must be regarded as new to science, namely, Alcyonium pur-
pureum, Ceratoisis ramosa, Eugorgia gilchristi, and Juncell.^
spiralis. Of the remaining seven species, three have not
hitherto been known to occur in Cape waters, and one, of which
a small specimen was incorrectly described in my last paper as
Primnoisis, is now referred to the species Wrightella coccinea,
which is also new to Cape waters. One specimen of a Gor-
gonia, probably belonging to a new species, is too incomplete
to describe adequately.

The following is a complete list of the Alcyonaria which have
at present been found in Cape waters: —

212

Sub-Order STOLONIFERA

Family Clavulariidae.
Anthelia capensis, Studer (15). 33*^ S. 17*^ W. 50 Fathoms.

Sub-Order ALCYONACEA

Family Xeniidae
Heteroxenia capensis, Hickson (6). False Bay. 20 Fathoms.

Family Alcyoniidae.

Alcyonium pachyclados, Klz (8). Off Cape St. Blaize, etc.
15-18 Fathoms.

Alcyonium antarcticum, W. and S. (23). Off East London.
45 Fathoms.

Alcyonium purpureum, n. sp. Mossel Bay, etc. Shore.

Acrophytum claviger, Hickson (6). Algoa Bay. 26 Fathoms.

Sarcophytum trochiforme, Hickson (6). Off East London.
45 Fathoms.

Anthomastus grandiflorus, Verrill (18). Off C. Recife. 256
Fathoms.

Family Nephthyidae
Eunephthya thyrsoides, Verrill (20). False Bay. 20 Fathoms.

Sub-Order GORGONACEA

Family Briareidae

Spongioderma verrucosum, Mobius (12). Several localities.
Shallow water.

Family Melitodidae.

Melitodes dichotoma, Pall. (13). Several localities. 41
Fathoms.

„^ . , , ,, . ^ / N f Mossel Bay. Shore.

Wnghtella coccmea, Gray (4). | ^^ j^^^^^ ^^^ ^^ Fathoms.

213

Family Dasygforgiidae.

Trichogorgia flexilis, N. G. et. sp. Off Cape Recife. 56
Fathoms.

Family Isidae.

Primnoisis capensis, Studer (15). 33° S. 17'' W. 50 Fathoms.
Ceratoisis ramosa, N, sp. Off Vasco de Gama Peak. 230

Fathoms.

Family Muriceidae.

V^illogorgia mauritiensis, Ridley (14)- Off East London. 85
Fathoms,

Acanthogorgia armata, Verrill. Vasco de Gama Peak. 230
Fathoms.

Family Plexauridae.

Eumcella papillosa, Esper (3). { ?;,\,f/Bay. 50 Fathoms!
Euplexaura capensis, Verrill (20).

New Family Malacogorgiidae.
Malacogorgia capensis, N. G. et sp. Algoa Bay. 25 Fathoms.

Family Gorg-oniidae.

Gorgonia flammea, E. et S. (2). Common in Shallow water.

Gorgonia capensis, Hickson (6). Off Cape St. Blaize. 40
Fathoms.

Gorgonia albicans, Kolliker (9). Several localities. 17-20
Fathoms.

Eugorgia gilchristi, N. sp. 34^^ S. 25^^ E. 52 Fathoms.

Gorgonia Sp. (?). 34*^ S. 25" E. 52 Fathoms.

Gorgonia (?) crista, Mobius (12). ? ?

Family Gorgonellidae.

Gorgonella stricta, Lamarck. Mossel Bay. 30 Fathoms. Sent
to me by Air. Strugwell.

Juncella elongata, Pall. (13). Off Algoa Bay. 25 Fathoms.

Juncella spiralis, N. sp. Off Cape Morgan. 36 Fathoms.

214

Sub-Order PENNATULACEA.

Family Virgulariidae.

Virgularia reinwardti, Herklots. St. Francis Bay. 30 Fathoms.

Family Anthoptilidae.

Anthoptilum grandiflorum, Verrill (19). Off Lion's Head. 136
Fathoms.

Family Cavernulariidae.

Cavernularia elegans, Herklots. False Bay. 25 Fathoms.
Cavernularia obesa, Val. Port Alfred. 43 Fathoms.

Specimens of all these species except Anthelia capensis,
Eunephythya thyrsoides, Pnmnoisis capensis, Gorgonia (Lopho-
gorgia) crista, Euplexaura capensis and Gorgonella stricta
have been collected by Dr. Gilchrist and sent to me for
examination.

The specimen I have provisionally marked Gorgonia ? may
be of the same species as Verrill's Euplexaura capensis, but the
spicules of our specmien are smaller and more uniform in shape.

The present paper contains an account of the following
species : —

Alcyonium purpureum. N. sp.
Anthomastus grandiflorus. Verrill.
Wrightella coccinea. Gray.
Trichogorgia flexilis. N. G. et Sp.
Ceratoisis ramosa. N. sp.
Acanthogorgia armata. Verrill.
Gorgonia albicans. Kolliker.
Gorgonia. ?

Eugorgia gilchristi. N. sp.
Malacogorgia capensis. N. G. et Sp.
Juncella spiralis. N. sp.
Anthoptilum grandiflorum. Verrill.

Geographical Distribution. — The general features of the
Alcyonarian fauna of Cape waters are particularly interestmg.
Affinities with the Atlantic fauna are seen in the occurrence of
the genera Anthomastus, Acanthogorgia, Ceratoisis, An-
thoptilum, and possibly Eugorgia. Affinities with the Indian
Ocean fauna in the occurrence of the genera Wrightella^
Juncella, .Sarcophytum and Cavernularia.

215

The genera Acrophytum, Spongioderma, IVIalacogorgia and
Trichogorgia appear to be peculiar to the district, and among
the pecuhar and characteristic species of the district may be
mentioned Alcyonimn purpureum, Sarcophytum trochiforme,
Gorgonia flammea, Gorgonia capensis, Eugorgia gilchristi and
Juncella spirahs.

We have, therefore, assembled together in this region repre-
sentatives of the Indian Ocean, of the Atlantic Ocean, and
possibly in Alcyonium antarcticum a representative of the
Southern Ocean, but, at the same time, we find a considerable
number of species v/hich appear to be peculiar to the district.

It may be of interest, for comparison with this statement, to
quote the following sentence from Agassiz's report on the
"Challenger" Echinoidea (p. 263).

" The assemblage of species at the Cape of Good Hope is
most peculiar, it is the meeting of the western boundaries of the
African Indian Pacific and of the Indo-African, the southern
boundary of the Atlantic and the northern extreniities of tiie
Southern Ocean faunae, and it has no species characteristic of
its own in the continental or abyssal range."

Family Alcyoniidae.

Alcyonium purpureum, n.sp,

Plate VII., fig. I. Plate IX., fig. 18.

Locality : Jetty at Mossel Bay.

Some specimens of this Alcyonium growing upon Tunicate
tests and worm tubes were tal^en from the piles of the jetty at
Mossel Bay. Dr. Gilchrist states that he has found the same
species in Saldanha Bay. With some hesitation I have decided
to constitute for them a new species. They are clearly distinct
from the A. pachyclados described in my last contribution (6),
and they are also distinct from A. antarcticum, which they re-
semble superficially in some respects. They appear to differ
from all species of the genus about which we have adequate in-
formation in the fact that, when alive, they are of a brilliant
purple colour. Dr. Gilchrist tried all sorts of expedients, in
vain, to get the specimens to retain their colour. Sometimes
he seemed to have succeeded, but when the colony was apparently
quite dead, the pigment would pour out into the preservative
fluid in streams. I do not call to mind any description of a
species of Alcyonium in which a soluble colour of this kind is
mentioned, and I have never noticed anything of the kind in
the living specimens I have examined. In the British speci-
mens of Alcyonium digitatum there is frequently noticed a pale
pink colour, which gradually fades when the colonies are pre-

2l6

served. It is by no means constant, however, some specimens,,
particularly those in very shallow water, being devoid of it.
The " brilliant purple " colour of this species is particularly in-
teresting, and is deserving of further investigation.

The colour character by itself is not one which would justify
us in constituting a new species for the specimens.

The Mossel Bay Alcyonium, however, differs from other
species in a variety of other characters.

It resembles A. antarcticum in the hemispherial shape of some
of the colonies and in the general transparency of the
coenenchym. It differs from it in the general character and
distribution of the spicules and in the absence of a stalk. It
comes nearest, perhaps, to Alcyonium sphaerophorum Ehrb., as
described by Klunzinger (8), but differs from it in the greater
size of the lobes, the entire absence of a stalk, the absence of
smooth calcareous corpuscles, and in the shape and size of the
spicules. Ehrenberg described the colour of his species as
" pallida, polypis fuscis." 1 his is a good description of the
colour of A. purpureum after the purple colour has been
washed out.

In all the specimens the zooids are partially expanded. I
have little doubt that if they had been carelessly killed and pre-
served, their resemblance to Alcyonium sphaerophorum would
have been closer than it is.

The largest colony sent to me is attached to a Tunicate test.
It is roughly hemispherical in shape, 40 mm. in diameter, 20 mm.
ni height. It springs from a flattened incrusting base bearing
polyps, there being no stalk (sterile Fiiss), as in A. fulvum
Forsk. There are fifteen lobes, varying from 15 mm. — 8 mm.
in diameter. The zooids are large, being in the specimens nearly
I. mm. in diameter at the base of the anthocodiae. The other
colonies are much smaller, and consist of only one or two lobes.

The six ventral mesenteries are well developed. The figure
given by Kiikenthal (11), in a recent paper, of Alcyonium fulvum
Forsk. has some resemblance to our new species, but in the
size and character of the spicules it is clearly distinct.

The general coenenchym and the body walls of the antho-
codiae are singularly transparent. At the surface of the
coenenchym the spicules are closely crowded together, but a
little way below the surface they are isolated and scattered. At
the base of the tentacles there are eight triangular shields,
pointed above, of scattered spindle-shaped spicules. The ten-
tacles themselves are dark brown in colour and entirely devoid
of spicules. The spicules of the coenenchym vary considerably
in shape, but are mostly double clubs or double balls with a nar-
row neck. Their average size is about o' i mm. The spindles and
clubs of the anthocodial body- wall are about 0T4 mm. in length.
All the spicules are provided with numerous protuberant
tubercles (fig. 18.)

217

\The sex of the specimens were not determined, but the
gonads are not mature.

Alcyoniuin purpureum. — Colony impregnated with a soluble
purple pigment. Without a stalk (sterile fuss). Lobes hemi-
spherical up to 15 mm. in diameter. Spicules of the coenenchym
double clubs and double spheres about o" i mm. in diameter.

Anthomastus G-randiflorus, Venill.

Plate VII., fig. 2.

Verrill (18). Amer. J. Sci. XVI., 1878, p. 376.

Locality : Off Cape Recife, 34° 27' S. 25° 42' E. 256 fathoms.

One specimen.

The single specimen obtained clearly belongs to Verrill's
species A. grandiflorus, of which many specimens have been ob-
tained in deep water 500-1,000 fms. off the East Coast of North
America and the West Indies. It has close affinities, too, with
Studer's (17) species A. agaricus, which was found in 1,267 rnetres
in 46°04' 40" N., 4g°02' W., but differs from it in the fact that
the autozooids of A. agaricus are smaller, more numerous, and
more rigid.

The measurements given are as follows :- — ■

A.

Length of autozooid anthocodiae
Diameter of ,, „

Length of tentacles
Height of Colony
Diameter of Colony

Verrill (21) had a large number of colonies to examine, and
the specmien he measured he described as " a well-grown speci-
men, but not the largest."

In the Cape specimen there are five autozooids with their
anthocodiae apparently fully expanded. In the specimens ob-
tained by Verrill the number of autozooids varied from two in
the smallest specimen to twenty or more in the larger ones. In
Studer's specimen of A. agaricus there were ten autozooids.

The anthocodiae of the autozooids are nearly transparent, the
body wall being provided with only a few scattered spicules ;
the tentacles are also transparent, but bear rather more spicules.
It is very probable that they are capable of complete retraction.

The siphonozooids are minute and very numerous. It is
difficult to give in figures any measurement of these siphono-

agaricus.

A. grandi-

A. grandi'

florus.

florus.

Verrill.

mihi.

mm.

mm.

mm.

9

36

22

5

7-9

9

5

12

10

20

about 40

40

0-38

50-82

14-18

2l8

zcoids that is satisfactory for comparison, as their size must de-
pend very much on the degree of contraction of the general
coenenchym. Between two points of the compasses fixed at
5 mm. apart from 7-9 siphonozooids may be counted. In A.
agaricus the siphonozooids are o"/ mm. apart.

The spicules are of two kinds. At the surface of the
coenenchym they are very irregular in form and very abundant.
They may be technically called clubs or double clubs or spindles,
but with their form masked by the long spiny tubercles. The
majority of them are about " 1 5 mm. in length. Below the im-
mediate surface of the coenenchym there are many long needles
irregularly spined and warted. They may be swollen in the
middle or at one extremity, but more usually are fairly uniform
m diameter. These needles vary very much in length, but some
I have measured are over o"/ mm. in length, with a maximum
diameter of 0*04 mm. Similar needles have been described in
A canadensis (o'5) and A. steenstrupii (o'5) by Wright and
Studer, and in A agaricus by Studer, but they do not appear to
reach quite such a length as they do in the Cape specimen of
A. grandiflorus.

The spicules of the body wall of the anthocodiae are spiny
sclerites o-o5-o-o7 mm, in length, and of the tentacles tubercu-
lated rods o'l x -03 mm. The colour of all the spicules is red.

The occurrence of this genus and species in Cape waters is a
fact of considerable interest. It is the first recorded instance
of the genus occurring elsewhere than in the Atlantic Ocean,
north of the Equator (except A. steenstrupii W. and S. (23),
off Japan). This is just the region from which the genus Sar-
cophytum, so common and abundant in the shallow waters of
the Indian and Pacific Oceans, is absent.

In my last paper I recorded the existence in 32*^53 S., 28*^12
E., 45 fathoms, of a species which I placed in the genus Sar-
cophytum under the name S. trochiforme. At the time I hesi-
tated whether S. trochiforme should be placed in the genus
Sarcophytum or the genus Anthomastus. The small size of the
autozooids and the relatively large size of the siphonozooids, how-
ever, appeared to me to point to its proper place being in the genus
Sarcophytum. I see no reason to dissent from that view now.
At the same time, there are some respects in which S. trochi-
forme does approach the genus Anthomastus. and it is clearly
an intermediate form.

It is an interesting fact that, whereas Sarcophytum is essen-
tially a shallow water genus (only two species occurring in water
as deep as 1 8 fathoms), and Anthomastus is a deep-water genus,
occurring in water from 200 to over 1,000 fathoms in depth;
Sarcophytum trochiforme, the intermediate form, was found in
45 fathoms.

219

Family Melitodidae.

Melitodes dichotoma, Pallas (13).

Localities: 1. Vasco de Gama Peak, N.W. f N., 8 miles. 41

fathoms. Rock. April 27, 1900.

2. East of C'ape Agulhas (Sebastian Bluff), W. by
N. I W., 6 miles. 26 fathoms. Mud.

In the first dredging two specimens were obtained, one red and
one yellow. It is interesting to find the two varieties of the
species in the same haul of the dredge in this locality. In my
last communication (6) I pointed out that the two varieties occur
in the same locality in False Bay. Beyond the fact that the
yellow variety in this new locality is rather more orange coloured
than pale yellow, as the specimens of the yellow variety were
in False Bay, no important differences were observed. In the
second dredging a small piece of (probably) the same species
was found growing on a shell of Terebratula rosea.

Wrightella Coccinea, Gray.

Gray (4), Catalogue Lithophyt., B. Mus., p. 32.

Ridley (14), Zool. of Alert, p. 581.

Locality : Fairly abundant on the piles of the jetty at Mossel
Bay.

The specimens from Mossel Bay are very closely related to the
specimens described by Gray as Wrightella coccinea from the
Seychelles. Unfortunately, the genus Wrightella is not very
well known, and is not very well defined from its neighbours.

The specimens in the collection agree with the description of
the genus in having foliaceous clubs in the cortex ; in the fact
that the nodes are not perforated by canals, and that the
branches arise from the nodes.

The foliaceous clubs (Blattkeule) are not very numerous, but
are very variable m form. Ridley says that the genus is dis-
tinguished by the " very massive form of the Blattkeule and the
swelhng out of their ' Blatt ' into rounded bodies with scarcely
perceptible edges." This description is perhaps a little difficult
to understand. It appears to me that the Blattkeule are formed
by the exaggerated growth of the tubercles at one end of a
tuberculated club. They may become expanded and anasto-
mose, forming leaf-like processes of irregular shape, or, as in the
case under description, the tubercles of nearly two-thirds of the
thicker end of many of the clubs may be prolonged into long
rod-like process with rounded ends. There is so much variety
m the spicules of this species, however, that it is impossible to

220

describe any one form as typical. Ridley does not mention the
size of the spicules. Klunzmger (8), in his description of
Mopsea erythraea from the Red Sea, which Ridley places in this
genus, says the spicules are from 0'i6-'02 mm. in length and
■03-0"o6 in breadth. The spicules I have measured in these
specimens vary in size up to about o"i5 mm. in length.

An examination of a fragment of the type specimen of
Wrightella coccinea in the British Museum shows that the
spicules are much more crowded in the superficial coenenchym
than they are in the Cape specimens and the foliaceous clubs
larger ('2 mm. in length), and much more numerous. The dif-
ferences in this respect are so great that I was tempted to con-
stitute a new species for the Cape specimens, but when I exam-
ined a considerable number of specimens, and found that there
is a great variability in respect of the number and crowding
together of the spicules in the coenenchym and in the relative
number of clubs and foliaceous clubs, it appeared to me wiser to
include them in the species named by Gray.

There is a difference of 30° latitude between Mossel Bay and
the Seychelles, and it is not surprising that the specimens of
Wrightella from the former locality should differ in this respect
from those found in the latter. There are other species of
Alcyonarians in which the tropical varieties exhibit larger and
more numerous spicules than the temperate varieties.

The genus is characterised by its dwarf size. Wrightella
chrysanthus is 2j inches in height, and Wrightella (Mopsea)
erythraea is said to be 40-60 mm. in height by Klunzinger (8).
The average height of the Cape specimens is about 40 mm.

The colour of the majority of specimens is red, but some are
orange or an intermediate colour between red and orange and
some pink. Ridley says that the verrucae of W. coccinea are
yellow. In this respect the Cape specimens differ from the
type, all the verrucae being red. The anthocodiae are white and
transparent, as described by Klunzinger for W. erythraea from
the Red Sea.

I may here call attention to an erratum in my last paper on
the Cape Alcyonaria Wlien I examined Wrightella coccinea
I was strongly reminded of the specimen I described in that
paper as Primnoisis capensis (Studer), and I was led to a re-
examination of its structure. I am now convinced that it belongs
to this species. The pink colour and the somewhat stronger
habit may be associated with the fact that it was found in 20
fathoms of water, i.e., much deeper than the Mossel Bay speci-
mens, but the comparison of the spicule preparations revealed
so many general similarities of form that there can be no longer
doubt of their identity. The Blattkeule, which I overlooked in
my first examination, do occur, although they are not numerous.

221

I have not had the opportunity of examining Studer's speci ■
men from 50 fathoms in latitude 33° 59' S. and longitude 17° 52'
E., which he named first Isidella and subsequently Primnoisis
capensis, but although this nominally belongs to another family,.
the Isidae, I am convinced that it must be closely related to, if
not identical with, Wrightella coccinea.

Family Dasygorgiidae, W. and S.

The family Dasygorgiidae of Wright and Studer (23) was
defined as follows : —

Colony consisting of a simple or branched axis. Main axis :
calcareous at its base, which latter is either flattened and disc-
like, or ramifying into numerous root-like processes ; the fibrous
portions of the stems and branches with calcareous particles
intermixed; often brilliantly iridescent. Coenenchyma; for
the most part thin, sometimes without spicules, at other times
with nujnerous transparent glassy, fusiform often spiny spicules,
or with irregular scale-like spicules ; sometimes the spicules are
in two layers. Polyps ; large, prominent, inserted on the axis
either at right angles or obliquely; covered with spicules
variously arranged ; tentacles retractile, sometimes only imper-
fectly so."

In his recent important paper on the Gorgoniidae of the Siboga
Expedition, Dr. Versluys (22) has revised the genera and species
included in this family, and has reinstituted Verrill's original
name Chrysogorgiidae for the family, and given a more restricted
definition of it. The specimen I am about to describe should
be clearly included in the family as described by Wright and
Studer, but in the fact that the zooids are situated on all sides
and not on one line only of the branches it would not be included
in the more restricted family as defined by Versluys. The same
difficulty arises v/ith regard to the species Chrysogorgia con-
stricta Hiles (7), in which the zooids are situated on at least two
sides of the branches.

Two courses are open therefore. Either to constitute a new
family for Hiles' Chrysogorgia constricta and this Cape specimen,,
or to aaopt VVnght and Studer's definition of the group and in-
clude them in the family Dasygorgiidae.

The latter course appears to me the more reasonable of the
two. The new genus, which I propose to call Trichogorgia, is
clearly related closely to many species of the genus Dcisygorgia
(W. and b.) or Chrysogorgia (Versluys), in the mode of origin
of the branches, in the character and disposition of the spicules,
in the arrangement of the calcareous deposits in the axis and in
the mode of retraction of the tentacles, etc. To constitute a new

222

family for a form so closely related to a well known genus would
be clearly inconvenient. I have re-examined the type specimen
of Chrysogorgia constricta (Hiles), and I am convinced that its
affinities with the genus Chrysogorgia as defined by Wright and
Studer are pronounced. In neither case is there sufficient justi-
fication for the estabhshment of a new family. The specimen
from which Miss Hiles described her new species was fragment-
ary, and the question whether a new genus should be constituted
for it may be postponed until complete specimens are forth-
coming.

Trichogorgia Flexilis N. gen et spec.
Plate VIII., fig. 13. Plate IX., figs. 16 and 17.
Locality : Off Cape Recife, 34° 7' S., 25° 43' E. 56 fathoms.

This species was obtained not very far from the station where
Malacogorgia capensis was found but in deeper water. The
colony is attached by a short main unbranched stem 8 mm. long,
2 mm. in diameter, expanding into an irregular disc 6 mm. in
greatest diameter. Branching begins by a long but narrow
branch 8 mm. from the base, and this is soon followed by what
appears to be dichotomy, and then a third branch is given off
at a considerable distance from the right hand division of this
dichotomy, and finally a fourth.

Diagram of the Branching of Trichogorgia. — I., Main axis; II.,
primary branches ; III., secondary branches ; IV., tertiary branches,ij;

If we consider that, in the position in which the specimen is
drawn, the branch to the right is in all cases the main axis, then

2 2"?

there is a close similarity to the branching of such a species as
Dasygorgia (Chrysogorgia) flexilis (W. and S.) all the primary
branches being given off along a definite line on one side of the
axis. This line is, however, not spirally twisted, and the colony
is consequently branched in one plane only, and does not form,,
therefore, a bushy or shrubby colony, as is usually the case with
the branched species of the family. Secondary, tertiary and in
some cases quaternary branches are given off from the primary
branches. The terminal branches may be 20 mm. in length, but
it is characteristic of the species that the axes of the terminal
and subterminal branches are very fine and hair-like 0.1-0.02-
mm. in diameter. The axis is throughout fibrous and horny, but
it contains a considerable quantity of calcium carbonate in the
form of fine amorphous granules.

The coenenchym is missing on the unbranched stem and very
thin on the larger branches, but it becomes thicker proportion-
ately on the secondary branches, and reaches its greatest relative
thickness at the end of the terminal branches. At a distance of
2.5 mm. from the extremity of a branch I found that the total
diameter of the branch was .15 mm. and of the axis .025 mm.

The anthocodiae are all partially retracted into long cylindrical
calicles i mm. in length. These calicles are set at an acute
angle, and are directed towards the terminal extremity of the
branch. The calices are very numerous, crowded on all sides' of
the terminal branches and closely set on the sub-terminal
branches. Both calices and coenenchym are protected by a great
number of calcareous overlapping scale-like spicules, arranged in
a single layer.

The characteristic feature of the species is that nearly all
these spicules are of the same general shape, i.e., a very thin, flat
double disc or double paddle. The size varies from O'l x o"04
(greatest breadth or O'Oj breadth across the neck)
to o"i5 X •035 ("03 across the neck). A few simple
spindles of the shape shown in fig. 13c. may be found,,
and these and all other varieties are rare or very rare. The
spicules are very nearly smooth, but in some cases a fine pattern
and a minute serration of the edges may be seen (fig. 1 3d.).

The colony is female. Each zooid contains not more than
one ripe ovum ; but young zooids and adult with no ova occur on
the same branch. (Plate IX., fig. 17.)

The genus may be characterised as follows : —

Colony branching in one plane. Axis becoming very thin in
the terminal branches. Calices numerous and situated on all
sides of the terminal and subterminal branches. Spicules very
thin double discs or double paddles, numerous, overlapping, in
one layer on calices and coenenchym.

224

Family Isidae.
Ceratoisis ramosa, n. sp.
Plate VI L, figs. 3 and 4. Plate VIII., fig 12.

Locality : Vasco de Gama Peak, N. 71 E., i8i miles. 230
fathoms. Stones. May 4, 1900.

This specimen is a good deal broken, but considering its
•extremely fragile character it is sufficiently well preserved to
merit congratulation on its arrival.

The base is missing, and the terminal branches have in many
places been lost, but I estimate its total height to have been
about 100 mm. The main stem is relatively thick (3 mm. in
diameter) and made up of calcareous intern odes 3-5 mm. in
length, alternating with homy nodes i mm. in length. The cal-
careous internodes give rise to the calcareous intemodes of the
branches, aiid the calcareous intemodes of the branches give
rise in a similar manner to the intemodes of the secondary
branches. The bronze colour of the nodes contrasts finely with
the ivory-white calcareous internodes as described by Verrill in C.
ornata. The branches arise very irregularly in all planes, and
the whole colony has the appearance of a low, dwarfed, finely
branched shrub. The smaller and terminal branches are very
slender, and have very long internodes with relatively short
nodes. The terminal branches are about 0.2 mm. in diameter,
with internodes 5 mm. long and pad-like nodes 0.4 mm. thick.

Throughout the whole colony the coenenchym is extremely
thin. The main stem has the appearance of being quite naked,
but an extremely thin membrane bearing a few isolated zooids
can be seen on closer examination. The zooids are irregularly
scattered on the main stem, but are arranged on the branches at
intervals of 2 to 4 mm. apart. In some places on the terminal
branches three or four zooids may be close together, but they
are never so crowded as they are represented to be in Ceratoisis
grayii (4). The zooids with the tentacles folded over the crown
are about i mm. in height. The surface of the zooids and the
coenenchym is covered by plate-like spicules which in dried
specimens overlap in places, giving the surface a primnoid
appearance.

The aboral side of the bases of the tentacles are usually pro-
tected by spicules of the shape given in fig. 4b., and remind one
of the opercular spicules of Primnoidae. Surrounding the neck
of the zooid there are curved spindle-shaped flat spicules about
9*4 mm. in length (Fig. 4c.). On other parts of the zooids and on
the general coenenchym the spicules are very irregular in size
and shape, but long flat spindles up to 0.6 mm. in length pre-
dominate. Irregular plates such as those drawn in Fig. 4a.
and Fig. 4d. are also found on the general coenenchym.

225

This species differs from all the other known species of the
genus in its profuse ramification, but the justification for referring
it to the genus lies in the fact that the branches arise solely
from the calcareous intemodes, and that the coenenchym is
throughout extremely thin. In its profuse ramification it
resemoles the genus Acanella belonging to the same family, the
general character of the spicules recalls that of many species
■of the Dasygorgiidae, whilst the large flat plates covering the
bases of the tentacles indicate some affinities with Primnoidae.

The distribution of the species of Ceratoisis is interesting.
The original type specimen of C. grayii (Wright 24) was ob-
tained in 400 fms. off Setubal, in Portugal. Verrill (21a) records

C. ornata as occurring in 250 fathoms off Nova Scotia. C.
palmae was obtained by the Challenger in 1,125 fathoms off the

Canary Islands. The other species are Pacific Ocean forms.

C. Phillippinensis was found off the Phillippines in 82 fathoms.

C. grandiflora in 210-255 fathoms, and C. nuda on the reefs off

Fiji. C. paucispinosa in 345 fathoms on the Hyalonema grounds

off Japan (23).

Family Muriceidae.

Acanthogorgia armata, Verrill.

Verrill : Am J. Sci XVI., 1878, p. 376.

„ Bull. Mus. Comp. Zool., 1883, XL, p. 31.

Locality : Vasco de Gama Peak, N. 71 E., i8i miles. 230
fathoms, stones.

This species is fully described by Vernll (1883). It was first
brought by the Gloucester fishermen from the fishing banks of
Nova Scotia and Newfoundland, and it was also obtained by the
U.S. Fish Commission in 1882 off Martha's Vineyard. It is
essentially a deep water form, occurring in 220 fathoms off
George's Bank down to 640 fathoms off Martha's Vineyard. It
is closely allied to and perhaps identical with specimens obtained
in deep water by the " Blake " off the West Indian Islands, and
with A. hirsuta Gray obtained in deep water off Madeira. The
genus Acanthogorgia is principally found in the Atlantic waters,
A. longiflora from 700 fms. off the Phillippines, Acanthogorgia
muricata off Funafuti (7a), and Acanthogorgia spinosa off New
Britain (7), being the only species which up to the present time
have been described as occurring outside this area. The species
have been found off the Eastern shores of America, from New-
foundland to Patagonia, and the species A. ramosissima has been
found in the Southern Ocean off Prince Edward's Island. Studer
(17) records the genus from the Gulf of Gascony and the Azores,
and the species A. hirsuta was originally described from
Madeira.

226

The specimens obtained by Dr. Gilchrist are probably frag-
mentary. One specimen is 52 mm. and another 72 mm. in
length. They correspond with the description of the type
specimens in that the calices are from 5-7 mm. in height and
2 mm. in diameter, and the large bent spicules are about 0.8 mm.
m length, and correspond in shape and variation with the descrip-
tion of the type spicules. The species that comes nearest to this
geographically is A. ramosissma (23) from Prince Edward's-
Island, but this species has smaller spicules and a more regular
arrangement of the calices on the stem.

New Family Malacogorgiidae.

This family is constituted for the reception of the remarkable-
specimen described below. Its characters are : Colony branched
and upnght. Axis slender, horny. Spicules and all other forms,
of calcareous skeleton absent.

The position of this family in our system must for the present
remain uncertain. In general form the single specimen
resembles Trichogorgia flexilis, which is undoubtedly a member
of the family Dasygorgiidae, but as there is no calcareous deposit
in the axis and the method of branching differs from that of the
Dasygorgiidae in some respects, I think it is best, as a tempor-
ary measure, to place it next to the family Gorgoniidae.

Malacogorgia Capensis, N. Gen. et Spec
Plates VII., fig. 5, and VIII., figs. 10 and 11.

Locality : Algoa Bay. Lat. 35° 40 S., Long. 35° 56 E. 25.
fathoms.

This remarkable specimen shows absolutely no calcareous
structures. As it was possible that the preservative might have
contained sufficient acid to dissolve the spicules, sections were
prepared and carefully examined with the high power of the
microscope, but they revealed no spaces from which the spicules
could have been dissolved out. The spicules of none of the
other specimens collected at the same time show signs of
corrosion by acid, and it is extremely difficult to believe that any
acid fluid acting for a short time could have so completely dis-
solved away every trace of lime. The only conclusion that can
be reached, with reason, is that this is a Gorgonian devoid of
spicules.

Whether we should be justified in including this genus in the
family Gorgoniidae, or whether it should be made the type of a

227

hew iamily are questions \\hich the future must decide. Jror
the present 1 think it is better to place it in a new family. The
single specimen consists of a cyUndrical stem 15 mm. long by
1.75 mm. in diameter attached to a stone by a broad base. It
gives rise to four main branches which after a course of io-2u
mm. again divide into 12 secondary branches. The secondar)-
branches are very slender, about 70 mm. in length, and alone
bear the zooids. The zooids are fully expanded 0.65 mm. in
length from the base of the anthocodia to the base of the ten-
tacles and about 0.3 mm. in diameter. The tentacles are 0.4 mm.
ui length, and bear from 12-14 pinnae in a single row on each
side.

The stomodaeum of the zooids is long and bears a very well-
marked siphonoglyph.

The zooids increase in numbers partly by additions at the base
of the secondary branches (Fig. 10) where there may be seen
two rows, one on each side of the branch, of ver)- young zooids,
the youngest being nearest to the primaiy branch. In the lower
two-thirds of the secondary branches the bilateral arrangement
;s preserved, but in the upper third other zooids are added, and
tlien they appear to be distributed on all sides of the branch in
a dense cluster. The colony was female, a few of the polyps
bearing a single immature ovum 0.05 mm. in diameter. The
genus may be described as follows : —

Genus Malacogorgia. — Colony slightly branched. Axis horny
with no trace of lime. No spicules in emy part of tlie colon)'.
Polyps arranged bilaterally in the plane of branching at the basal
two-thirds of the secondary branches and on all sides of the
terminal one-third of the secondary branches.

M. capensis willi the characters of the genus. Colour in
spirit white.

Family Gorgfoniidae.

Tliere are man}- unsatisfactory features about the present day
definition of this family, and a thorough revision of the genera is
ver)' necessary.

The species m Cape waters which is clearly i^latygorgia
albicans of Studer (16) and Eunicella palma of Verrill, and con-
sidered by these authors to be the same as Gorgonia albicans
of Kolliker (9), Gorgonia palma var. alba of Esper (3) and
l>oplu)gorgia palma of E. and H. exhibits the peculiar torch-
shaped spicules of the Plexaurid Eunicella, but it differs from the
Plexauridae in having a \'er)' thin and not a thick fleshy
roenenchym on the main axis. The remarkable flattening of
the axis is similar to that of Gorgonia flammea, in which species
there may also be noticed a very remarkable .scantiness of
coenenchym on the main branches. Until the anatomy of these

c

22S

corals IS more thoroughly, mvestig-ated, 1 am inclined' to niclude
Gorgonia albicans in the family Gorgoniidae notwithstanding
the remarkable difference between il and Gorgi^nia flammea in
the character of the spicules.

Studer (i6) in his original description of the genus Plalygorgia
stated that the spicules, although similar to the spicules of
h^unicella in form, are characterised by the fact that the clubs
attain the same length as the spindles. " Die spiculen sind
characteristisch in dem die Keulen hier die Lange der Spmdeln
erreichen." This is not confirmed by my specimens. 1 have
been able to compare carefully the spicules of my preparations
ot Platygorgia (Gorgonia albicans) with those of Eunicella
papillosa, and I have found them remarkably alike. The torches
(clubs) of both are on an average o. i mm. m length and the
spindles 0.14 mm. The torches are fairly uniform in size when
full grown, the s])indles are much more ^■ariable. The most
striking point, however, of the comparison is the extraordinary
resemblance in detailed structure between the torches of the
Eunicella and of the Gorgonia albicans. The examination of
my preparations of spicules alone would lead an}- one to the con-
clusion that they were taken from the same species. Vet the
specimen of Eunicella papillosa obtained on Rij Bank 25 fathoms
has a cylindrical axis and prominent verrucae, whilst the speci-
men of Gorgonia albicans picked up on the beach at Port Alfred
has a very much flattened axis, thin coenencliym and incon
spicuous verrucae.

These facts seem to point to the conclusion that Eunicella
papillosa and Gorgonia albicans should be included in the same
genus and family. Further in\estigation of the anatomy of the
two forms is necessary to justify the suggestion tliat Eunicella
papillosa should be jilaced in the Genus Gorgonia, but I think
we should be justified in transferring Eunicella papillosa to the
family Gorgoniidae.

G-orgonia, Albicans, Kollilier.

Gorgonia allvcaiis. Kolliker, Icones liist.

Eunicella paliiin, Venill.

Platyfiorgiii alhicaiu Stiidcr, Arcliiv. Naturg. 53, i, p. 60.

? Goriionia paliim.. Tar. alba. Esper., Pflan/.entliieff.

Localities: (1.) Port Alfred. Beach dried.

(2.) Off Gape St. Blai/c, S.W. Seal Island, W. '. S.
17 fathoms. Sand.

The- colony of the lirst-naincd specinuMi is uprigiit, 300
iiiai. in height, branching in one plane, tlu- branclu-s
free, not anastomosing. Tiie stem and tin- inferior j^arts of the
branches are ver\- iiuicli flattened in the ])]ane of branching.

2 2Q

1 he diameters of the main stem are 5 mm. and 22 mm., the axis
l^emg bare, of the lower branches are about 4 mm. and 10 mm.,
including tjic coenenchym. The terminal branches are almost
cylindrical in shape. The coenenchym is thin. The axis is
liorny, with very little or no lime. The \'errucae are small o'5
mm. m diameter and scattered unevenl)- on all sides of the
branches.

The surface of the coenenchym presents a series of peculiar
torch-like spicules similar to those of Eunicella (Hickson 6 Plate
v.c). They are about o. i mm. in length. The other spicules are
spindles with very pronounced and regularly arranged tubercles,
varying considerably in si/.o, l)ul frequently reaching 0.15 mm. in
length.

The second specimen, which is well preserved, is a good deal
smaller, being not more than 70 mm. in height. The coenen-
chym of the base is very largely destroyed, and the horny axis
if covered with a varied fauna of zoophytes, of which a species
of Hydrozoon allied to Bimeria is a predominant feature. The
most important fact about this specimen is that the stem and
branches are not flattened but nearly cylindrical throughout, and
the branching is not strictly in one plane. The coenenchym
moreover, is in many parts of the branches swollen to form oval
knobs, or knuckledike nodes. These two features are noteworthy
because, to a superficial observer, they might seem sufficient to
separate the specimen from a species to which the generic name
Platygorgia has been applied. I am very strongly of opinion
that the flattening of the stem and branches so frequently noticed
in this species and in Gorgonia (Lophogorgia) flammea (Ellis) is
a character produced by local circumstances, and should be used
very cautiously for diagnostic purposes. For this reason I have
adopted Kolliker's view that the species should be retained in
the genus Gorgonia for the present. The colour of both speci-
mens is pale cream to white.

The. second specimen is, in external features, very similar to the
specimen that is described next (Crorgonia species?), but it differs
from it decidedly in the character of the spicules. In Gorgonia
albicans, moreover, the elongated slit-shaped openings of the
calices are irregularly arranged, not parallel with the axis as
tb.ey are in Gorgonia liitkeni, G. capensis aaid the unnamed
specimen in the collection.

Grorg'onia, Sp. (?).
Locality: Lat. 34° 5' 20" S., Long. 25° 43' E. 52 fathoms. Rock.

Svii. Eii/^liwiiiirii tiipcusis (.*) Verrill, 20.

In the bottle containing Eugorgia gilchristi (p. 230) there was a
fiagment of what may have been a large Gorgonian consisting

230

01 an axis 120 mm. lon^ witli six lateral branches, themselves
supporting a few small branchlets. The branches are 4 mm. x

2 mm. in diameter, the axis being decidedly flattened, and they
terminate distally in distinctly swollen extremities. The zooids
are few in nimiber and confined to the lateral areas of the
flattened branches. The coenenchym is thick and fleshy on the
terminal branches, but very thin on the principal branch or axis.
The spicules are warted spindles about . i mm. in length and
fairly uniform in size. Some longer pointed spindles with fewer
tubercles occur in the calices. The surface of the branches is
smooth, the position of the calices being represented by slit-
shaped pores parallel with the axis and in some cases with a
slight mound or convexity round them. The specimen is pro-
bably related to Gorgonia liitkeni (W. and S.) and Gorgonia
capensis (Hickson), but differs from both in some particulars.

I he terminal branches are a good deal more fleshy than in both,
and in colour it is white not yellow or yellowish red. Tn tliese
particulars the specimen approaches Verrill's Enplexaura
capensis. The axis is devoid of calcium carbonate.

Eugorgia Gilchristi, n, sp.

Plate IX., figs. 15 and iq.

Localiiy : Lat. 34" 5' 20" S., Long. 25° 43' E. 52 fathoms. Rock.

Considerable difficulty was found in determining the proper
])Osition of this form. In its mode of growth the character of the
superficial coenenchym, the character of the verrucae, it has a
close resemblance to the genus Verrucella, and at flrst it seemed
to correspond witli the Verrucella granifera, Koll, obtained by
Mobius off the coast of Africa (sic !) But it differs from Verru-
cella and the family Gorgonellidae to which Verrucella belongs
ii' the fact that the axis is not calcified. I have decided, there-
fore, with some hesitation, to place it in the genus Eugorgia
(Verrill), with which it agrees in the horny character of the axis,
the general character of the spicules, and the " finely granulous "
nature of the surface. It seems to approach most nearly the
description of Eugorgia Daniana (\"errill) from Panama and
Pearl Islands, but I think it may justifiably be regarded as a
distinct species.

It is a large flexible coral sj^ringing from a single horny stem
_j.5 mm. in diameter and reaching a total height of 240 mm. It
branches profusely, pinnately or irregularly with occasional
anastomoses. ihe terminal branches are of variable length, and
about I mm. in diameter. The axis is horny throughout and
verv slender indeed in the terminal branches. The verrucae are
dome-shaped, quite distinct, but not very prominent, and they are

231

evenly distributed on all sides of the branches. The coenenchym
IS very thin on the thicker branches, but relatively thick on the
smaller ones. The surface is finely granular.

The spicules are ver}' numerous, and consist of spindles with
prominent tubercles of various sizes up to .12 mm. in length,
double wheels up to "05 mm. in length, Maltese crosses, fig. 19 c,
rough stars or warted spheres .03 mm. in diameter. In the
calices there are finely pomted spindles with relativel)' few
tubercles and some which are distinctly flattened, fig. iqc, as
in V'erruceila granifera.

The specimen is almost white.

Family Gorgonellidae.

Juncella spiralis, n. sp.
Plate VIII., figs. 6, 7, 8 and 9.

Locality : Off Cape xMorgan, 32° 45' 45'' 8., 28" 26' 15" E. 36
fathoms. Stones, 29 specimens. January 12,
1900.

Twenty fine specimens of this remarkable, new, spirallv twisted
juncella were sent to me, with the remark, " There were no
e:pecimens showing the tendril-like stock twisting round an\-
support."

The longest specimen was 220 mm. in lengtli when slightly
stretciied, but this was imperfect, the base having broken off.
The longest perfect specimen was 150 mm. Two small perfect
specimens which are not spirally twisted, are 50 and 55 mm. in
lengtli. The greatest breadth is 4-5 mm. in some large frag-
ments. The smaller specimens are only 2-3 mm. in breadth.
The stock is unbranched in all the specimens. The axis is pale
brown in colour, and has rings of lime imbedded in the horny
matrix. The verruca; are all prominent, and arranged irregularly
on t\\ o-thirds of the circumference of the stock, leaving a bare
track on one side free from verrucae for the whole length of the
stock. In a large specimen the bare track is r75 mm. wide, and
the verruc;e project about 1-5 mm. The tentacles are in all cases
retracted. 1 n colour, the bare track is orange red, and the verruca-
are a pale yellow. A specimen that I dried became \-ellow all
over, and I have noticed that the spirit in which the specimens
are preserved has a j^ink colour. The red colour may therefore
fade away in the course of time. The bare track and verrucae are
covered with a dense armature of spicules, and it is difficult to
believe that the verrucae can be ever retracted. The spicules
covering the surface'of the verrucae are irregularlv tuberculated
plates about "i mm, in length and rather less in breadth, but

2'?2

iiiixed with these are a nuiuber of clubs (Fig. (S) and s]Dindles with
thick coarse tubercles about i mu]. x -05 mm. in size. On the
coenench\'ni the surface is armed ])rincipally with these clubs
and spindles, but a few more flattened forms ma\- be found among
tliem. It is mi].iossible to describe the man) \ a nations in form
and character oi' the spicules. Many of them are like the types
described, l)ut otners are quite irregular. When a fragment of
the hard crust which covers the surface is detaclied with needles
iind examined with the nucroscoj^e it is seen that the spicules arc;
tightly iainmed together to form an impeiielrabie armour. The
surface of the verrncse has a distinctly sqnamate appearance, the
plate like spicules slightly overla]:)])ing (Fig. 9b.). There are
also some long needles o'q mm. in length (Fig. ga) in the deeper
parts of the zooids. There can be little doubt 1 think that the
species is right!)' placed in the genus bnicella, and it appears to
be most closely related to Studer's (15) Juncclla tlexilis from
between Flat Island and Mauritius in 25 fathoms. It differs from
this, however, in its smaller size (the specimen of J. flexilis was
20 cm. in length), in its spiral form, m the more irregular arrange-
ment of the ven-ucap on the stock, in the presence of a bare track
on one side, and in some other peculiarities.

This bare tr;i( k is seen in some other species oi Juncella. In
the description of juncella juncea from tlie Isle of Bourbon
Milne-Edwards and Haime state thai the calices leave some
trace of a median coenenchymatous space. Ridle)' (14) also
states that there is a distinct median grooxc in tlie s])ecimen of
juncella juncea obtained b)' the "Alert."

The squamate armature of the veniiCcV shows some aftinities
with the characters of the Primnoidae, but, as the plate-like
s]:)icules are so small and there are no definite opercular ]:)lates,
its affinities with juncella are closer. It is noteworth)-, how-
ever, that in the Primnoine genus Cal)q3terinus the calices do not
occur on one side of the stock. The track which is free from
the calices in Calvpterinns, however, is covered hv the over-
lapping scales of the lateral calices so as to form a tube. These
bare tracks on one side of the stock in Juncella spiralis and
Calypterinus allmani have a certain resemblance to the bare
tracks on one side of the smaller branches of some forms of
Solenocaulon, and suggest the presence of symbiotic crustacea.
There is no evidence in support of this at present, but it would
he worth the trouble for an)- naturalist who has the opportunity
of dredging in these v.aters to note the character of any
Alpheidae or other animals that might possibl}' live with this
luncella. Dr. Gilchrist's note that nothing was found around
which the stock twisted is of the nature of a support for the sug-
gestion that the spiral form of the larger stocks is associated
with the presence of some epizoic animal. We may for the
present regard the spiral form and the bare track as characters
oi the species, but if they prove to be mere adaptations to an

233

epizoic animal their importance must be consiclerabl)' discounted.

The species may be described as follows :—

C"olon\- unl)ranclied ; in llic larger forms sjMrally Iwisled.
X'errucse prominent, situated irregularly and closely on three-
fourths of the circumference of the stock, leaxiiii^- a bare track on
one side free from vernicse. Spicules of the surlace ol verruca'
a.nd coenenchym densel}- ])<\cked. irregular m form or clubs or
spindles with prominent blunt tubercles, all about o" i mui. in
lenj^th. A fc\\- lonq- needles 0"4 mm. in the more deep-seated
tissues.

Colour ot the cociiciich\-in DranL^c I'cd, <:il the \crruca> vell.)w.

XUTIi UN .irXCELLA liI,0.\G.\T.V (I\VIJ..\S).

.Since the publication of my last paper on the Cape Alcyonaria
1 liave had the opportunity of examining a large number of pre-
parations of spicules of specimens of the genus Juncella from
the Indian Ocean and elsewhere. In none of them can be found
a single spicule of the kind which 1 called " triple stars," These
peculiar spicules are \'ery abundant in the preparations made
from the Cape Juncella elongata, and might be regarded as a
character sufficient to distinguish the Cape specimens as a dis-
tinct species. Other characters, however, that they exhibit are
so strikingly similar to those of Juncella elongata that I do. not
]-.ropose to 'take this step at present. They ma>' be regarded,
hovxever, as constituting a distinct \ariety, and should be called
bincclla elongata (Pall.), var. capensis.

Sub=()rder PENNATULACEA.
Family Anthoptilidae. KOll.

Anthoptiluni grandifl-jrum. X'onill.

]'n\i;iilani-i i^rdinlitlor,!. Verrill ( i<)).
Aiilhoptiliiin llioiiisoiii . Ivijlliker (liO.

Plate VI 11., tig. 14.

Locality: Lion's Head. N. 6;° E. 25 miles. Depth : 131-136
fathoms. Bottom : Black sjoecks on the lead ; no deposit ob-
t.nned b)' the dredge

The following notes were kindl\- sent by Dr. Gilchrist : —
" The colour when fresh was ;i uniform i)right brick red.
" Some difficult\- was experienced 111 pre\entiug the polyps from
' bemu' washed off b\' the motion of the ])reser\ati\e fluid owin

'fc>

fc>

'■ to tlie roll of the vessel.

234

" They seem to occur in great abundance in particular places,
" probably formnig mmiature forests at the bottom of the sea.
" They are not straight when fresh, but more or less bent, with
three gentle curvatures."

lift)- four specimens were obtained on i\Iarch 28, 1900, of
w inch twelve were sent to me.

There can be no doubt of the identity of this .species with that
obtained by the " Challenger " in the iVtlantic Ocean, S. of
Buenos Ayres, depth 600 fathoms ; and described by KolliUei-
as Anthoptilum thomsoni ; but it is difficult to distinguish tJie
species from Yerrill's \'irgularia grandiBora described m 1879.

The largest unbroken* specimen had the following measure-
ments : —

Cape Specimen. " Challenger."

Length of Polypidom lojo mm. 560 mm.

Length of stalK ibo mm. 87 mm.

Breadth of stalk bulb 30 mm. 18 mm.

Length of polyps 16 mm. I7'5 nun.

Length of tentacles 6 mm. 6'8 mm.

In the second column are given the measurements of the
largest specimen of the species obtained by the " Challenger."

The other specimens sent from the Cape were a little smaller
than the one mentioned above, but 1 have not measured them
accurately, as there is no sufficient reason for supposing that they
l)elong to different species. The mam point of interest is that
the specimens found on the Eastern side of the Atlantic arc
larger than those obtained by the " Challenger " on the Western
side; but whether this is to be associated with the fact that the
' Challenger " specimens were found in deeper water (Ouu
I'athonis, acfamst 135 fathoms) or that they were obtained in 20
degrees of latitude further South, or whether the difference in
size is simply a matter of geographical isolation, is a question
v.hich it is premature at present to discuss. It is noteworth}',
liowever, that notwithstanding the difference in si/e of the
polypidoms, the auto/.ooids and t/ie tentacles of the autozooids
of specimens from the Eastern and Western waters are approxi-
mately of the same size. Accurate measurements haAe not been
made, and, as it is impossil)le to pre\ent a certain amount of
shrinkage in the preservation, cannot be made ; but the differ
ence in the measurements of the autozooids and their tentacles
given b)' Kolliker and those of m\' specimens is so slight that it
is probable that when alive in their nati\e habitats tlie\' were
exactly the same.

* Some ut the specimen? tli;it wei^e cut into twn m- tlnxH: iiicces lor tr;iii?il
were probably larger ttian tliis.

235

The specimens agree with the general descnption given by
Kolliker in most respects, but there are a few minor differences
which are worthy of record.

KolUker remarks that very often the lowest parts of two or
three or more polyps of a row are united so as to produce the
a]jpearance of very small pinnules, but in no place are all the
polyps of one row united in such a manner. This fusion of the
bases of the autozooids is much more pronounced in the Cape
specimens. Most of the autozooids of a row are fused at their
bases for a distance of 3-5 mm. in the middle rows, but there are
usually one or two gaps irregular in position in each row where
the fusion of the bases is reduced to a mere trace. In the lower
rows the fusion of the bases is scarcely noticeable. (Fig. 14.)

The siphonozooids are, as in Kolhker's specimens, very
numerous, but, unlike Kolliker's specimens, they exhibit an al-
most contmuous dorsal* series unuiterrupted at the bases of the
rows of autozooids. I have also found no siphonozooids between
the individuals of the rows of autozooids.

I have searched for the calcareous corpuscles found in the
muscular layers at the base of the stalk without success. I can-
not find any calcareous corpuscles in any part of the colony.
This may be an oversight on my part, but it is practically im-
possible in seapens of this gigantic size to search all parts of the
fleshy substance with equal care.

* The dorsal side of this pciinatuUd was called '-ventral " by Kolliker. For a
full discussion of this point, the reader is referred to the important memoir by
H. F. E. Jungersen on the •'Pennatulida of the Danish liiyolf Expedition, 1904,"
a copy of which was kindly sent to mc by the author, and came to liand as I was
passing the proofs of this paper f(^r llie press.— S. J. M.

236

LITERATURE REFERRED TO IN THE TEXT.

1. Ehrenberg. Die Korallenthiere- des Roten Meeres. Ber-

lin. 1834.

2. J. Ellis and D. Solander. Natural History of Zoophytes.

i;86.

3. Esper. Die Pflanzenthiere. Nurnberg-. 1791-1797.

4. Gray, J. E. Catalooue of the Lithophyta. Brit. Mus.

'1870.

5. Gray, J. E. Description of a new genus of Gorgonidae.

Proc. Zool. vSoc. 1835.

6. Hickson, .S. J. Marine Investigations in South Africa.

The Alcyonaria and Hydrocorallinae. 1900.

7. Hiles, I. The Gorgonacea collected by Dr. Wille)'.

Willey's Zool. Results. 1899.

7a. Hiles, I. Report on Gorgonacean Corals. Proc. Zool.
Soc. 1 899.

8. Klunzinger. Die Korallthiere des Roten Meeres. 1877.

9. Kolliker, A. Icones Histologicae. Leipzig. 1865.

in. Kolliker, A. Pennatuiid.-i of llie "Challenger" Expedi-
tion, \'o]. I.

I I Kiikentlial, W. Ueber einige Korallentliiere des Roten
IMeeres. Festschrift von L. Haeckel. 1904.

12. Mobius, K. Neue Gorgoniden des Natur-hist. Museums in
Hanilnirg. Nov. Act. Acad. Leop. XXIX. 1862.

[3. Pjillas. Elenchus /.oopli}-torum. The Hague. 1766.

14. Ridley, S. O. Report on the Zoological Collections made

in H.M.S. " .Mcrt." 1884.

15. .Studcr, Th. Anthozoa Alcyonaria. .S.M.S. "Gazelle."

Monats. K. Acad. Wiss. 'Berlin. 1878.

16 Studcr, Th. X'ersuch eincs Systems der Alcyonaria.
Archiv. Naturges Vol, 53. 1887

^37

I/. Studer, Tli. Alcyonaires proveuaiit des Caiupagnes dc
rHirondelle, Monaco. 1901.

18. \'ernll, A. E. Recent additions to Llie marine fauna of the
eastern coast of North America. Am. J. Sci. XVI.
1878.

uj. Verrill, A. E. American |ournal of Science. Vol. XVII.
1 8/9-

20. X'crrill, A. E. Classificalion of Polvps. Proc. Essex

Instit. Vl.

21. Verrill, A. E. Report on the Anthozoa. IJull. Mus. (onip.

Zool. \'oI. n. 1883.

2 1 a. Verrill, A. E. Bull. Mus. Comp. Zool. XI.

22. Versluys, J. Die Gorgoniden. Siboga Expeditie. Mono-

graphic. XIII.

23. Wright, E. P., and Studer, Th. The Alcyonaria of H.M.S.

" Challenger."

24. Wright, E. P. Ann. and Mag. Nat. Hist. 1868 ii. and

1869 iii.

E

:23^

DESCRIPTION OF THE PLATES.

Platk \T1.

Fig. I. Alcyonium purpureum. New species. The largest
colony obtained, attached to a Tunicate test.
Natural size.

Fig-. 2. Anthomastus grandiflorus. Verrill. Natural size. Aa
Autozooids. Si. siphonozooids.

hig. 3. Ceratoisis ramosa. New species. The specimen was
a good deal broken, but it will be noticed from the
drawing that the branching is profuse and not con-
fined to any one plane. Twice natural size. See
also PI. VIII., fig. 12.

Fig. 4. Spicules of Ceratoisis ramosa. N. sp. a. and d. Ir-
regular plates found on the surface of the general
coenenchym ; b. triangular plate found on the
aboral side of the base of the tentacles ; c flat
spindle-shaped spicule from the neck of the an-
thocodia.

Fig. 5. jMalacogorgia capensis. New genus and species.
This species has no spicules. See also Pi. VIll.,
hgs. 10 and 11.

Fi..\TK Vili.

iMg. 6. juiicella spiralis. New si^ecies. A large but incom-
plete specimen of the species. All the larger
specimens are spirally twisted. Natural size.

I'l^- /■ Jiincella spiralis. N. sp. A smnll portion enlarged
showing the " bare track " (b.t.) on one side of the
stem and the prominent club-shaped \errucae (ca.)
on tlie other.

Fig. <S. Club-shaped spicule of Juncclla spiralis from the sur-
face of the general coenenchym.

239

Fig. 9. Spicules from the verriicfie ol Juncella spiralis, a.
Needle ; b. plate from surface of the verruca.

Fig-. 10. Malacogorgia capensis. N. sp. A portion of the
colon}' at the origin of the primary branches to
show the origin of the young zooids.

Fig. II. Malacogorgia capensis. N. sp. A portion of a ter-
muial branch showing the expanded zooids.

Fig. 12. Ceratoisis ramosa. N. sp. A portion of a terminal
branch, much enlarged.

Fig. 13. .Spicules of Trichogorgia flexilis. New genus and
species, a., b., c. different torms of the flat scale-
like spicules which cover the anthocodiae and
general coenenchym ; d. a portion of one of these
spicules, very much enlarged.

Fig. 14. Anthoptilum graiidiflorum. Verrill. A portion of
the rachis is figured to show the connection be-
tween the bases of the autozooids (au) to form
rudimentary leaves. In nearly all the rows a gap
(g) is seen. Si. the scattered siphonozooids.

Platk IX.

Fig. 15. A small ])ortion of the flabellum of Eugorgia gil-
christi (N. sp.), showing at a, a. the occasional
anastomoses of the branches. Natural size.

Fig. 16. Trichogorgia flexilis. New genus and species. A
small portion of a colony. Natural size.

Fig. 17. Trichogorgia flexilis. New genus- and species. A
fragment of a branch highly magnified, showing
at a. a young zopid, b, b. two large zooids each con-
taining a single ovun'i, c. a bud, d. a c\'lindrical
zooid not containing an ovum.

I'lg. i(S. Spicules of Al(:}'onium [airpureum (n. sp.), a. a typical
tuberculated spicule of the coenenchym, b. a club.

Fig. 19. Spicules of Eugorgia gilchristi (n. sp.), a. b. two of
the many varieties of spicules of the coenenchym,
c. one of the very minute Maltese crosses, d. and e.
spicules of the anthocodiae.

(Publibhed I5tli December, 1904.)

Marine Investigations of South Africa.

Alcyonaria Plate VII.

1 ALCYONIUM PURPUREIJM.

2 AN T H C MAS T U S GR ANDIFL ORUS .

3,4 CERA.TOISIS RAMOSA.
5 MALACOGORGIA CAPENSIS.

E. Wilson, Car.^bridge.

Marine Invesbigabions of South Africa.

Alcyonaria Plate VIII.

au-

JkMmh

ill /?/

6J,8,9 JUNCELLA SPIRALIS. 12 CERATOISIS RAMOSA.
10,11 MALACOGORGIACAPENSIS. 13 TRICHOGORGIA FLEXILIS.
14 ANTHOPTILUM GRANDIFLORUM .

E. Wilson, Cambridge .

Marine Invesbigations of South Africa

Alcyonaria Plabe IX.

15,19 EUGORGIA GILCHRISTI. 16,17 TRICHOGORGIA FLEXILIS.
18 ALCYONIUM PURPUREUM.

E .Wilson , Cambridge .

ox THE

ECHINODERMA

FOUND OFF THE COAST 'OF SOUTH AFRICA.
PART II. ASTEROIDEA.

F. JEFFREY BELL, M.A.,

Enierilus I'ldlessm" in King's College, London.

For the convenience of others, I propose to adopt the order in
which the genera are arranged in the " Challenger " report, but
the careful student will remember that an improved disposition
of the genera is to be found in Bronn's Thierreich.

A very large number of specimens has been sent me, and the
investigation has been laborious, but instructive, for the only
new species has been made on a single specimen ! The very
long series of Astropcctcn pontoporcens forms an interesting
object-lesson in variation ; I have kept for the collection of the
British Museum a most interesting and instructive series of what
I took at first to be several new species of Pentagonasier, but
which are now seen to be different stages in the growth of the
till now rare and little known Pentagonasier tubereitlatns of
Gray.

Not only do the Cape Starfishes show an alliance with those
of the North Atlantic, but there are also indications of the
]~resence of species best known as yet from the Indian Ocean ;
this is indeed only to be expected when we examine the trend
of the currents round the southern peninsula of the Old World.

It is unfortunate that a large part of the collection was put
{ I do not say preserved) in formol, with the result that some of
the specimens had to be destroyed at once. I have already*
called attention to the dangers attendant on the use of this

" " Collections of the ' Soul hern Ci'oss ' " (1902I, p. 214,

244

— /

^

Localitw

Means ot
Capture.

•

Xature of
bottom.

Bird Islaiid Li^lUhousc, X. ',

By shrimp trawl

57' 59

Mud and dark

1 4025

W., ,S^ niik-s.

sand.

Capo Xalal, \V. hy X.. 4^ miles

By large dredge

47

Sand and shells

10039

Cape luiaiUa, X. f E.. 12 miles

By small dredge

44

Sand and shells

^P54

About 16 miles AWslnl East

By large dredge

37 4-2

7574

London, Lat. 33^' 13' 30"

S., Long. 27 39' 40" E.

Bird Island, Cape I'oint Light-

By dredge

22

Fine sand

13005

house, W. '■', X.. 7'J miles.

False Hav, Cape I'tniit Light-

By dredge

45

K(.iugh biittdUi

3»-'5''

house, X.W. by \v. ',\y.,

and a small

J'l miles.

quantityol mud

Ott Mossel L>av, Cape St.

By large trawl

31

Mud

Ulo

Blai/.e, X.W." ', X., 3 miles

Cape St. Blai/e, X.E. ', E., 27

By large liawl

45 4''i

Fine sand

1-J59

miles (Fositiiin by D.R.)

Cape St. Blai/.e, S.W. by \V. J

By di edge

15-18

Stones

215a

W., about 6^ miles.

Buffalo Bav (near Knvsna),

By shrimp Ir.iw 1

14 to

Fine s.md

io2!^-

WalkerFoint, X.W. hv W.

33'

^ W., 1.] miles.

Along Coast, between Cape St.

By dredge

4

Sand

107

Blaize and Anchorage,

Mossel Bay.

Lat. 35 54' 15', Long. 25 53' 30"

P>v di edge

.11

Fine dark sand

537

Frcim niiuith i>\ XalnMm Ui\er

vShurecnl lectin;^"

.

123

4. Astropecten capensis.

Aslropcctcn capensis, Studer, Abh. Akad., Bcrl. 1S84, ii, p. 44.

Dredged off Cape Natal, W. \ N. \2\ miles; depth, 85 fms.
Bottom: Sand and shells. Reference No. i/l

245

5- Psilaster acuminatus.

Psilater acuminatus, Sladen, Chall. Report, Aster. (1887),

p. 225.

Obtained as follows:- —

Locality.

Means of
Capture.

Depth in
fathoms.

Xature of
bottom.

OIT Lions Head, S.K. 21

1
miles By large dredge

95

I^(jugh

2049a

Cape Point Liglithouse,

X.E. Bvslirinip trawl

8=;

Fine green sand

1 543''^

h\ N. 7| miles

Cape Point, X.E. by E.

' E. 8 Bv shrimp traui

9^

Sand andspeeks

17523

miles

Cape Point, X.E. by E. ■_

• E. 8

Bv shrimp trawl

91

Sand andspeeks

^1$^^

miles

A young foi-ni was dredged off Vasco De Gama, X. 40'' E.
13! miles ; depth, 120 fathoms ; bottom rough.

There are small and doubtful examples of Astropecten from
the following- localities which I cannot determine specifically.
Most are probably varieties of Astropecten pontoporceits : —
(i) Bluff Light House, S.W. q miles (Refer. No. 12528) ; (2) Lat.
34° I/' 54", Long. ii;0 .c;' 6" (Refer. No. n3); (3) Sebastian
Bluff, N.W. 2.^ miles (Refer. No. 6069) ; (4) East side of Riv
Bank, off Port Elizabeth (Refer. No. 1089A).

6. Luidia africana.
Liiidia africana, .Sladen, Chall. Report, Aster. (1887), p. 256.
Obtained at the following localities : —

Locality.

Means of
Capture.

Cape St. Blaizc, X. by E. { E.,
65 miles.

Vasco De Gama, X,, 10 E.,

13 miles.

False Bay, Buff els Bay

Cape Point Lighthouse X., 16
E., 10 miles.

By large dredge

By lish Iraw 1

P.y trawl

By shrimp trawl

Xature of
bottom.

N5 00 I^cnigli

iS5 (ireeii sand

85 ' Gi'een mud .and
I hard ground.

1887

^739

409
15362

246

In connection with the apparent affinities of the Cape species
with tliose of the North Atlantic it should be noted that this
species is said by Mr. Sladen to be " nearly allied to Liiuiia sarsii
of the North Atlantic," and that, further, he recognised it from
a locality 35° north of the Equator. A series of forms from
intermediate localities may show that the species should be
united."*

PENTAGONASTERIDJE.

7. Pentagronaster tuberculatiis.

Astro goniv.m tiiberailatum, Gray, Proc. Zool. Soc. i<S47, p. 79;
id., Synopsis Starfish. (1866), p. 10. pi. I., fig-. 2.

Pentagonaster titberculahis, Perrier, Arch. Zool. exper. v. (1876),
p. 37; Studer, Abh. Ak., Berl. 1884, 11., p. 33.

1 ill now this species has been regarded as very rare ; the
specimen described by Gray in 1847 from Port Natal had no
companion till 1898, when the Government Biologist sent one
specimen from False Bay ; one specimen was collected by Pro-
fessor Studer in 117 fathoms south of the Cape of Good Hope.
The species is now well represented, and is found to grow to a
good size.

The species was obtained in the following localities :- — ■

Locality.

Lnt. T,\ 4V 15" S. I.on«. iS'
30' 00" E.

Means of
Capture.

0.5

Nature of
Bottom.

By shrimp trawl 1 23

Cape St. Blai/.e, X.K. \ E. 27 I By laryt- trawl 44 46 Fine sand
miles. '

Cape St. Blaize, N. 42 E. ma«. By ba.u tangles

11 miles and 25 miles S.W.
^ W. Irom Cape St. Blaize.

Cape St. Blai/.e. X. 10 W. 33
miles.

on bar of
large trawl

Bv otter trawl

44 46 ; Fine sand

"J

4^\S

i:!.s;^

60 and
73:'

201

* It may be noted that the author of the above quoted sentence .goes on
to say "of which it is unquestionably the representative." If by this the
taxonomic representative is meant. I submit it is tautologous, and so an offence
against style ; if bionomic representative is meant, it is to be regretted that
evidence was not given in place of the always suspicious-looking word
' unquestionably."

247

— Tl

1 0

Lncahty.

.Means iil
Capture.

Xature of
Bottom.

52

'i'i

2> miles S.W. irdiii Cape St.

Bv small shrini)i

44

246

Hiai/e.

trawl.

Cape St. Bhiize, X. 3O miles ...

By shrimp trawl

.S4

Sand and shells

l.\S-'

Lion's Head, East, iS miles ...

By large dredge

1(14

Blackish specks
and rocks.

2212

Vasco Ue Gama, S. 75 E., 13.I

By large dredge

i5(i

Black specks

^5'^4

miles.

Vasco De Gama, X. 10" East,

Bv lish trawl ...

'^5

Dark green sand

-74.)

13 miles.

Same Locality, etc., as Xo. 2743

- -

—

-V44

Table Mountain, S.E. bv S.,

B\- shrimp trawl

120

Green sand

IS-'O^,

f S. 45i miles.

False Bay

-

—

<'-!. Calliaster baccatus.

Calliastcr haccalits, Sladen, Chall. Report, Aster. (i.SH/). p. 280.

Reasons for separating- Calliastcr from ]\ii/a;^<iiias/cr are
definitely given by Professor Perrier on page 3 v" "^ '''^ report on
the Starfishes of the " Tahsnian "-" Travailleur " expeditions.

The species was obtained as follows : —

Locality.

Means of
Capture.

Depth in
fathoms.

Nature nf
liottoni.

'11

'■J

Rocky Bank, False Bay

By dredge

1

]

'17-27

Rocks

15611

Cape St. Blai/.e, X. by E.,
6~1 miles.

J E.,

By large Iraw 1

.i.S

Afiid and s.ind

1710

Cove Rock, X.W., i W.,

134

By dredge

80 130

Coral and ri ick

13201

miles.

1

(Magnetic) S.W. , \ \X.
miles from Cape St. Bl

10

li/e.

Large otter traw

1 —

1
t

28

248

PENTACEROTID^.

9. Culcita veneris.

Ciilcita veneris, Perrier, Arch. Zool., expcr. viii. (1879), p. 48.

Of this species there is only one specimen known ; it has not
hitherto been represented in the National collection. The
specimen was obtained as follows: — Lat. -54° 12' 54", I^ong.
18° 36' 9". Depth, 23-3; fms. Station V.'

ASTERINIDJE.
10. Asterina coccinea.

Astcrisais coccineiis, M. Tr., Syst. Ast. (1842), p. 43.

Asterina coccinea, Perrier, Arch. Zool., exper. v. (1876), p. 234.
Was obtained as follows : —

Locality,

Seal Island, False Bay, N.W,,
ri X., 6i miles.

Between River and Sebastian
Bluff, nearer the I'oimcr.

\ c

C/f

(-"

r*

C 1

.**>

0 1

1)

jr^

Q

^

Means ol | £ c 1 Xature of \§ "^
Capture. i 'o^s: \ bottom. "S ^

u

By drcdtje ' 22 Broken shells 16211

Collected at low — 1 10004

tidu

Mossel Bay I Collected at low — ! i6jo

tide.

II. Palmipes novemradiatus.

Although there is but a single specimen of this, I think I am
entitled to regard it as new, since no Palmipes is known with
more than five rays.

Rays nine, disc very thin, thicker and carinate along the
dorsal line of each ambulacrum, cream- white in alcohol, yellower
along the ventral line of each ambulacrum ; the free part of eacli
ray obtusely conical. Dorsal ossicles small, set in ver\' regular
rows.

R. = 28. r=2i.

Cape Natal, 48 fathoms, fine sand.*

* Since writing the above I have found two very much injured specimens
that are twice as large as my type, from off the Tugela, and one mucli smaller
off the Itongazi River, N.W. i W. 3 miles ; depth, 25 fathoms ; bottom, rocks
and stones." I have not altered my diagnosis, with which they agree.— 26th
May, 1904.

249

STICHASTERID^.

12. Stichaster felipes.

Stichaster felipes, Sladen, op. cit., p. 433.

This species was obtained as follows : —

Locality.

Means of
Capture.

'^ E
us

18

Nature of
bottom.

. 0

Mossel Bay

By shrimp trawl

Fine sand

70

Cape Point Lighthouse, X.E.

By shrimp trawl

85

Fine green sand

15434

by N. 7J miles.

Lion's Head, 22 miles

By large dredge

25

Rough.

2049B

SOLASTERID^.

13. Solaster penicillatus.

Crossaster pcnicillalus, Sladen, op. cit., p. 446.

A well-marked and quite common form. Obtained from the
following localities : —

Locality.

Means of
Capture.

Q4:

Xature of
bottom.

Cape Point, N.E. by E., | E.,
46 miles.

By net on beam

—

17055

Lion's Head, E. by S., | S.,

By large dredge

171

Oreen sand and

^J57

37 miles.

rocks.

Lion's Head, S., 72' E.. 47
miles.

By large dredge

IQO

Green sand and
black specks.

-''44

Table Mountain, X., 71)' E., 40

By shrimp trawl

250

Green sand

2462

miles.

Sebastian Bluff, X., \ W .. 6.\

By shrimi"> trawl

."i<>

Mud

(>037

miles.

Table Mountain, S.E. by S., \
S., 51 miles.

By shrimp trawl

196

Green sand and
black specks.

151X1

250

PTERASTERID^.

14. Retaster capensis.

I'lcras/cr capensis, Gray, P. Z. S. (1847), p. 82; Perner, o[). cit.,

p. 302.

Rc/(istcr capensis, Sladen, op. cit.,. p. 47

/•

From Vasco di Gama, Cape St. Biaize, Lion's Head. Youn^
examples of the genus were procured as follows: —

Locality.

Means dI
Capture.

57-"

.\ a tine til
bdtlDm.

5

Bullels Bay, False Bay. fTlic

l>v hui^e trawl

;v>i

specimen was injured by
forniol.)

-

Table Mountain, South Head,
E. by S., i S., :>5 miles.

V>\ shiinii) trau 1

igo

Gieeii sand and
black specks.

15140

Constable Hill, E. I S., 19.^

By shrimp trawl

U5

Green sand and

i47''4

miles.

black specks.

iiiieljes Bay, _'4 i .' 07

SliDiL' (.ullceting

—

. i7«

15. Hymenaster sp.

A small injured s|iecimcn from ("ape Point, X. 86° E. 43 miles.
Depth, 900-1,000 fathoms. Bottom, greyish nmd. Procured by
shrimp trawl.

ECHINASTERID^.
16. Henricia ornata.

lichinaslcr {Cribclla) oniatns, Perrier, Ann. Sci. Nat. xii. (1869).

p. 251.

Cribrclla oniafa, id., Arch. Zool. expcr. iv. (1875), P- ,v7-
C ribrella simplex, Sladen, op. cit., j). 547.

Reasons for using Gray's generic name will be found in Ann.
and Mag. Nat. Hist., vi. (1890), p. 473. I can see no reason for
distinguishing C. simplex from C. ornata ; Mr. Sladen can
never have thought of comparing it with the widelv distributed

'■Si

species described by Peirier ; there are very many specimens,
and I think 1 have arrived at a right conchision.

The specimens were obtained at the following localities : —

Iv( icalitw

Cleans of
Capture.

Xature of
Bottom.

Q .i:

u

C.ipc I'diiil, X. 50' E., ]!S.\ miles By shrimp liaw 1

180

Green sand and

"4535

black specks.

Seal Island, S.W. !, S., ■ miks

P)y diedj^e

1 1

Rocks.

1 5907

Linn's Head, East, iS miles ...

-

By lai j.;e dredge

104

Black specks
and rocks.

2219

Hoetjes P>a\

Shore collecting

—

122

Lat. 34 17' 54", Lon.-i. 15 35' 6"

By trawl

.^^-37

- —

1 33

From BuHalo Bay, 7 10 98 ...

31

i.U

(Jape I'()int, E. by X., 29 miles

By shrimp trawl

250 to
300

Green sand and
mud.

'757''

False Bav, Seal Island. X.W. bv

By dredge

17

Broken shells

16320

W.. "i W. 7i miles.

Mossel Bay

By otter trawl

■ 12

Iktween Mossel liay Pier and

On shore

—

3''.i

St. Blai/e Li.uhthuuse.

Mossel Bay

On shore

--

202

\'asco De (iama. S. 73 E..

B\- large dredge

160

Black specks

-\SS4

13^ miles.

Table Moimlain, S.IC. by S. i S.,

By shrimp trawl

]<■)()

Green sand and

15182

51 miles.

black specks.

Table Mountain, S.E. by S, J S.,

By shrimp trawl

120

Green sand

15204

45J miles

ASTERIID^.
[7. Asterias calamaria.

Astciidi. i(il(ini<ii-ln, (ir;iy, Ann. and Mdg. N. 11., \ i. (1840),

p. 1 79.

This species, which is coniiuon off New Zealand, has alread)"
been recorded from Mauritius ; it would appear to be rare off
the Cape. The specimen was obtained at low tide in the rock
pools, Three Anchor Bay.

2^2

iS. Asterias capensis.

As/cnas cnpcnsis, Perrier, Arch .Zool., exper., iv. (1875), p. 338.

Dredged off Cape St. Blaize, N. b)- E. \ E. 65 miles. Depth,
89-90 fathoms. Bottom, rough.

19. Asterias giacialis.

Ashrias giacialis, Einn., Syst. Nat., x! (1758), p. 661 ; Bell,
Catal. Brit. Echinod. (1902), p. 98.

In 1882 I pointed out (Zool. An/.eig. v., p. 282) the great vax'ia-
tions t'xhii'tited bv A. gincinlis, but, if 1 have correctly identified
No. 18 as A. ai'pciisis, mv implied sugs^estion that it and A.
giacialis were one is not correct.

The specimens were procured as follows : —

Locality.

Means ol
Capture.

Depth ill
fathoms.

X.ilLiie of
bottom.

s

W.S.W., oH Bird Island, Lat.
33' 51' S., Long. 26" 11'
45" K.

Knysna, low tide along shore,
from jetty towards village.

By dredge
Shore Lollee

iiig

4^

S

nitl cUicl mud

1117
314

-

Cape St. Blaize. N.W. by X. By large tiaw 1 3O Mud 1435

i X., 7 miles. " 1

20. Asterias volsellata.

Asterias (S/olas/crias) volsellata, Sladen, op. cit., p. 584.

I have ascribed some remarkable fragments to this species,
which far more than some species of Mr. Sladen's deserves a
special genus for its reception.

Dredged off Great Fish Ponil Eight House. N. In W . 'l W.
17 miles. Depth, 100 fathoms. Bottom, sand, shells, and stones.

El addition to the twenty forms here enumerated, I note the
presence of two examples of EcJiinastcr which closely resemble
a specimen from Port Natal which has been many years in the

^ F- -,

Museum, and which I was unable to determine when I went over
the whole of the iviuseum series ; the two specimens in the
present collection are so similar to it that they do not help to
link it on to any of the many described species of the genus; T
cannot help thinking- that Pcrknaster is only an aberrant
Echinaster. Dr. Gilchrist's examples are from the shore at
Somerset West, and from Simon's Bay by trawl.

From Mossel Bay and off Cape Point N.E. by E. \ E,, distant
46 miles, come small specimens of an Archasterid which I am
enable to identify more closely.

A few Asteroids were so injured by the formol in which they
were sent that they had to be destroyed.

[Published qth May, 1905.]

ON THE

ECHINODERMA

FOUND OFP^ THE COAST OF vSOUTH AFRICA.
PART III. OPHIUROIDEA.

BY

F. JEFFREY BELL, M.A.,

Emeritus Professor in King's College, London.

The number of species of Ophiuroids in tiic collection is not
large. What there are are not very interesting, with the excep-
tion of the new species of OpJiiozono and Ophiura, and the
addition of a good series of the hitherto rare Ophiothaniniis
reniotus, which was dredged by the " Challenger " in the neigh-
bourhood of the Cape. Though Mr. Lyman has described a
large number of Ophiuroids from the .Southern sea-area, it would,
no doubt, be possible to find among the small specimens in the
present collection several that might well appear to be the " types
of. new species." But the study of Ophiuroids has suffered
enough from the description of isolated " species " based on one
or a few specimens ; this is notably the case with Ophiothrix, of
which a revision based on a long series ought to be made before
ever another species of it is described.

I. ZYGOPHIUR^.
1. Ophioderma, wahlberg-i.

O phioderma wahlbergi, M. Tr., Syst. Ast. (1842), p. 86.

This species, which was originally described from Port Natal,
appears to be widely distributed, as there are examples in the
Museum from the Red Sea and from Puerto Cabello.

256

2. Ophiozona capensis. (PI. I., Figs, i and 2).

No tentacle-scales to most of the ossicles, three or four
minute spines, and moderately sized, irregularly shaped radial
shields.

This diagnosis will show that Ihe Cape Ophiozona is distinct
from those enumerated by Mr. Lyman in his " Challenger "
report, while the species lately described by Professor Koehler
has two tentacle-scales along the whole length of the arm.
That which intrigued me most was the disposition of the
tentacular pores on the arms ; they are limited to the three
proximal joints, are quite large, and are protected by several
scales each ; this is a very abnormal arrangement among
Ophiuroids, and is, so far as I know, only known in the appar-
ently distant genus Ophiolipus ; in spirit preserved specimens
there are no signs of podia ; in his definition of Ophiolipus Mr.
Lyman says merely " tentacle pores only at the basal under arm
plates," and makes no reference to the presence or absence of
podia.

The comparatively large size of the pores in this species and
the apparent absence of podia may, in general terms, be trans-
lated into the statement that the respiratory apparatus is not
diffused along the whole arm as in most Ophiuroids, but
localised ; if this is true, it would follow that we should expect
to find haemoglobin or some other oxygen-bearing compound ;
haemoglobin was, of course, detected several years since by
Foettinger in Ophiactis virens. The resolution of this question
is not possible for me with the preserved specimens which I have,
but it could easily be settled by naturalists on the spot ; and it
1? one which I should very much like to see investigated.

I may be allowed to remark that the typical arrangement of
the podia on the arms, as described in text books, is probably
very often altered ; we know now of two genera, not apparently
closely allied, in which the proximal pores are alone retained ;
scattered through the writings of the describers of brittle-stars
there are indications of variations, which, had the writers been
morphologists, rather than systematists, they would have dilated
on.

Put in general terms, it may be said that a comparative in-
vestigation of the disposition of " tentacle-pores " and " tentacle-
scales " promises to be of great interest.

2^7

The specimens were

obtained as follows :-

—

Locality.

How Procured.

Depth in
fathoms.

Xature of
bottom.

Refer-
ence No.

Cape Point, N.E. by E., \ E., By shrimp trawl

800-900

Green mud

16928a

40 miles.

Cape St. Blai/e, X. h\ K., 73 Ry lar.i;c diedye
miles.

1^5

Sand anil shells
Oough.)

1871

Cape St. Blaizc, X. hy E. \ E., By larye diedge
65 miles.

85 90

Rough.

1886

Lion's Head, S.E., 22 miles By large dredge

25

Rough

2049

Lion's Head, East 18 miles

By large (hedge

104

Black specks and
rocks.

2211

Lion's Head, N., jH-" E, 12
miles.

By shrimp trawl

60

Sand and small
black specks.

2252

Cape St. Francis, N.E., 29
miles (D.R.)

By dredge

75

Sand, shells and
rocks.

14237

3. Ophiura trimeni. (PI. I., Figs. 3 and 4).

No arm comb, the radial shields of moderate size, separate and
diverging ; scales of disc small, but not minute ; three arm
spines, which break off very easily ; the uppermost the longest,
and about as long as an arm joint ; pores at base of arm
large and guarded by about six scales ; more distally both pores
and scales seem to disappear altogether ; the side arm plates are
of great size, and cause the almost total disappearance of the
lowermost plates. Colour in alconol whitish.

Diameter of disc, 8 ; 7 ; 9.

^Length of arm 21 (ca) ; 28 ; 25.

Breadth of arm at base, ■2'2; 2*5 ; 2"5.

The specimens were obtained as follows : —

— 'ji

_■

Localitv.

How Procured.

.Xaliuc 111

3^

oT^

bottom.

Xi

Q.5

<u

Lion's Head, S.E \ E., 42 miles

By large dredge

156

Dark green sand

2085

South Head, Table Mountain,

By shrimp trawl

190

Green sand and

15109

E. by S., 25 miles.

black specks.

Lion's Head, S.E., j E., 50

By shrimp trawl

230

Green sand

15034

miles.

Lion's Head, S.E. \ E.47 miles

By shrimp trawl

175

Green sand

14985

* Xo specimen had an unbroken arm.

258

I have named this species in remembrance of Mr. Roland
Trimen, F.R.S., who was for many years the leader of South
African Zoologists.

It will be observed that the pores on tlie arms present very
much the same characters as in Ophiozona capansis.

4. Amphiiira incana.
Ampliiiira incana, Lyman, Bull. Mus. C Z. v. (1878),

This species was obtained off the South Head, Tugela River,
N. by W., 4^ miles. Depth, 25 fathoms, black mud. Refer.
No. 11480A. 1 am not very confident of the accuracy of this
determination.

Other species of the genus were collected off Lion's
Head, S.E. \ S., 10^ miles. Depth, 69 fathoms. Bottom,
red and green sand. Refer. No. 2038. But the specimens were
not in good condition, and T cannot assign them to any known
species.

5. Ophiothamnus remotus.

Ophiothmmtus remotus. Toyman, Chall. Rep., p. 212.

The " type " of this species was a single specimen taken off
the Agulhas Bank ; it is well, therefore, that a good set was ob-
tauied from Cape Natal, N.W. \ W., 5^ mi., Refer. No. 18/ A, and
a few from off Algoa Bay, close to Riy Bank, Refer. No. 97.

6. Ophiothrix aristulata.

Ophtolltrlx ansi'.tlahu L\-man, Bull. Mus., ('. Z., vi. (1879), p. 50.

Obtained as follows : —

(i) Mossel Bay fishing ground, Cape St. Blaize, N.W. \ N.
Lat. 34-12-45 S. Long. 22-15-00 E. 35 fathoms. Mud.
Refer. No. 11 80.

(2) Great Fish Point Lighthouse, W. by N., 5 miles. By
dredge. Depth, 22 fathoms. Bottom, rocks, stony Polyzoa
and coral. Refer. No. 13619A.

7. Ophiothrix roseo-coerulans.

Ophiotlirjx rosco-cocnilans, Grube, JB. Schles. Gesell. (1868),

p. 45.

This species (from St. Helena) was so briefly described that
one cannot be certain of recognising it.

Obtained on Rocky Bank. False Ba}-. Depth, 17-27 fathoms.
Bottom, rocks. Refer. No. 15612.

259

8. Ophiothrix trig-lochis.
Ophiothrix triglochis, M. Tr., Syst. Aster. (1842), p. 1 14.

Obtained off Rocky Bank, False Bay. Depth, 17 fathoms.
Bottom, rock and coral. Refer. No. 15599.

Specimens of this genus, which I cannot specifically determine,
were obtained as follows ; those from the two last localities in the
table appear to be of the same species : —

Locality.

Means of
Capture.

Q>5

Xature of
Bottom.

'zy.

Tiigela River Mouth, X.W. hv Bv lar,i;e clred<ie
\V., 3i miles.

f4

Kocks

113S7

Lion's Head, S.E. \ E., 47
miles.

By shrimp trawl

17.5

Green sand

14984

South Head, Table Mountain,
E. by S., \ S., 25 miles.

By shrimp trawl

190

Green sand and
black specks.

15110

STREPTOPHIUR^.
9. Ophioscolex dentatus.

Ophiascolex dentatus, J.yman, Hull. Mus. C. Z., v. (1878), p. 157.

This species was taken by the " Challenger " off the Agulhas
Bank ; I know of no other record of its capture.

The specimens were taken by the " Pieter Faure " off Buffalo,
N.N.E., 17 miles, by shrimp trawl (Refer. No. 12836), at a depth
of 195 fathoms, or 45 fathoms deeper than tlie " Challenger "
haul.

CLADOPHIUR^.

10. Gorgonocephalus pourtalesi.

Astrophyton poitrtalesii, Lyman, 111. Cat., Mus. C. Z., viii. (1875),

p. 28.

Obtained off Vasco de Gama, S. 75° E., i^\ miles, b)' large
dredge. Depth, 166 fathoms. Bottom, black specks. Refer,
No. 2565.

26o

II. Gorg-onocephalus verrucosus.

Euryale verrucosuni, Lamk., An. s. Vert. ii. (1816), p. 537.
Astrophyton verrucas iiui, M. Tr., Syst. Ast. (1842), p. 121.
Gorgonocephalus verrucosus, Lyman, Chall. Rep. (1882).

Obtained as follows : —

(i) Off Buff els Bay by shrimp trawl. Depth, 30 fathoms.
Bottom, sand, with stony polyzoa. Refer. No. 2647.

(2) Off Sebastian Bluff, N.W. by W. \ W., U miles. By
shrimp trawl. Depth, 34 fathoms. Bottom, mud. Refer. No.
7050.

DESCRIPTION OF PLATE 1.

Fig. I., Ophiozona capensis, from above, to show the scaling
of the disc, and the form of the upper arm plates ; fig. 2, the
same from below to show the mouth-parts the under arm-plates
and the size of the spines.

Fig. 3 and 4, Ophiura Iriuicni, to show the same points.

[Published 8th May, 1905.]

Mar. Inv. S. A.

Echinodeima, PI. I,

C. lierjeau.

J-Jutterwortli,

I, 2. OPHIOZONA CAPENSIS x -h
3, 4. OPHIURA TRIMENI x 5.

INDEX

TO

FISHES, COPEPODA, MARINE ANNELIDS, ECHINOIDEA,
ASTEROIDEA, TURBINOLID CORALS, ALCYONARIA.

(a) Generic and Specific Names. (h) Localities.

riSHES.

(n) Generic and Specific Names.

133. 134.

I u

14

/
1 1,

Achirus, 152

acus (Syngnathus), 140, 151

Agriopus, 144, 152

andersoni (Pseudorhombus), i, 9

aquila (Sciaena), 150

argenteus (Scopelus), i, 15, 16

Arnoglossus, i, 12, 13, 133, 134

attenuatus (Cynoglossus), i, 11

boops (Cyttosoma), i, 6

brach3'cephalus (Cynoglossus), i, 12

capensis (Achirus), 152

capensis (Arnoglossus),

ISO
Cataetyx, 141, 142, 143,
Chlorophthalmus, i, 15
Chrysophrys, 134, 150
ciliata (Synaptura), i,
Cottidae, i, 7
Cottunculus, I ,
Cynoglossus, i,
Cyttidae, i, 6
Cyttosoma, i, 6
dentex (Chorisocliismus), 132
diniorphus (Platophrys), i, 10
fasciatus (Macrurus), 135, 136,

150
fulvo-marginata (Solea), i, 13,
gibbiceps ( (Chrysophrys), 150
globiceps (Chrysophrys), 1 34,
Gobiesocidae, 132, 151
Histiopterus, i, 3, 4
immaculata (Lampris),
knysnaensis (Serranus),
Kumu (Trigla), 150
Lampridae, i, 4, 5
Lampris, i, 4, 5
laniarius (Pagrus), 135,
Lepadogaster, 132
luna (Lampris), 5
macrocephalus (Cottunculus),
macrolepis (Arnoglossus), i, i
Macrurus, 135, 136, 137, 150

1^7.

14

ISO

I- 4. 5

ISO

13

macrurus (Pterois), 3
marginata (Synaptura), 14
melanoptera (Synaptera), i, 13
messieri (Cataetyx), 141, 142, 143,

151
microchirus (Stromateus), 139, 151

mormyrus (Pagellus), 150

natalensis (Pseudorhombus), i, 8

nigripinnis (Pterois), i, 2, 3

Pagellus, ISO

Pagrus, 13s, ISO

Panga, 135

parallelus (Macrurus), 137, 150

pectoralis (Synaptura), isi

Platophrj^s, i, 10

Pleuronectidje, i, 8, 9, 10, 11, 12,

13. 14
Pseudorhombus, i, 8, 9

Pterois, i, 2, 3

punctatus (Chlorophthalmus), 1,15

russeli (Pseudorhombus), 9.

Saccogaster, 143

saurus (Scombresox), 14s, 146, i47.

152
Sciaena, 150

Scombresox, 145, 146, 147, 152
Scombridae, $
Scopelidae, 1, IS. 16
Scopelus, I, 15, 16
Serranidae, i , 2
Serranus, i , 2
Solea, I, ID, II

spinifer (Agriopus), 144, 14S, IS^
spinifer (Histiopterus), i, 3
Squamipinnes, i, 3
Stromateus, 139, 151
Synaptura, i, 14, 151
Syngnathus, 140, 151
Triglids, i, 2, 3, iso
Trigla, 1 50
turbynei (Solea), i, 10, 11

262

Fishes — confiinied.

(b) Localities.

Bu ft el's Bay, 133
Cape Natal, 8

Cape Point, 131, 135, 136, 137, 138,
139, 141, 144, 145, 147, 148, 149
Constable Hill, ir>
Durban Harbour, 9, 14
False Bay, 4, 5, 131, 132, 131, \ \n
Kalk Bay, 5, 135
Knysna, 2
Lion's Head, 7, 13
Mossel Bay, 3, 11, 1 30

Muizenberg, 5
Natal, 2, 182
Port Natal, 9
Port St. John, 182
Roman Rock, 132
Saldanha Bay, 16
Table Bay, 5
Table Mountain,
Tugela River, 1 1
Umhlanga River,
Vasco da Gama

103

13

10, T .^

Peak, 7

COPEPODA.

(a) Generic and Specific Names.

abilominalis (Plenromamnia), 181,

i83, 195
abvssalis (Heterorhabdus), 181, 183,

191
Acartia, 180, 181, 182, 184
acuta (Labidocera), 183, 191
acuta (Euchaeta), 183, 190
Acrocalanus, 182, 184
aculeata (Lubbockia), 181, 191
aculeatus (.\egisthus), 182, 185
aculeatus (Paracalanus), 181, 194
Aegisthus, 182, 185
.\ethidius, iSo, 181, 182
affinis (Euch.neta), 183, 190, 201,

202, JIG

Angusta (Sapphirina), 181, 196

Augaptilus, 182, 185

arcuicornis (Clausocalanus), 180,

181, 182, 188
armatus (Aethidius), 181, 185
armiger (Gaetanus), 181, 183, 191
attenuatus (Eucalanus), 183, 189
aurita (Lucicutia), 183, 192, 202,

203, 205

austrinus (Heterorhabdus), 183, 191
bidentatus (Phyllopus), 183, 194
bipinnata (Candacia), 181, 186
borealis (Xanthocalanus), 209
bradyana (Lucicutia), 183, 192, 204,

205, 206, 210
brevicauda (Metridia), 183, 192
brevicornis (Calanus), 180, 181. 182,

185
Calanus, 180, 181, 182, 185, 186
Calocalanus, 180, 181, 182, 186
Candacia, 180, 181, 182, 186, 198, 210
catula (Candacia), 182, 186
Centropages, 181, 182, 187
cheirura (Candacia), 180, 186, 198,

210
chierchiae (Centropages), 181,182,187
Chiridius, 182, 187

Chirundina, 182, 187
Chorisochismus, 132
clausii (Lucicutia), 183, 192, 206
Clausocalanus, 180, 181, 182, 188
Conaea, 183, 188
Conifera (Oncaea), 181, 183, 193
cornutus (Rhinocalanus), 183, 196
Corycaeus, 180, 181, 183, 188, 189
crassus (Eucalanus), 181, 183, 189
curta (Candacia), 186, 198
danse (Acartia), 180, 181, 182, 184
danae (Corycaeus), 180, 183, 188
danae (Scolecithrix), 181, 183, 197
darwini (Calanus), 182, 185
discaudata (Teraora), 183, 198
elongatus (Eucalanus), 183, 189
Eucalanus, 181, 183, 189, 190
Euchaeta, 181, 183, 190, 201, 202, 210
Euchirella, 183, 190
finmarchicus (Calanus), 180, 181,

1S2, 185
flavicornis (Lucicutia), 181, 183,

192, 204
fragilis (Xanthocalanus), 181, 198,

209, 210
furcifer (Corycaeus), 180, 183, 188
furcatus (Centropages), 182, 187
furcatus (Clausocalanus), 180, 182,

188
Gaetanus, 181, 183, 191
gemma (Sapphirina), 183, 196
gibber (Acrocalanus), 182, 184
giesbrechti (Aethidius), 180, 182, 185
gracilis (Acrocalanus), 182, 184
gracilis (Pleuromamma), 181, 182,

183, 184, 185
grandis (Lucicutia), 204, 206
Heterorhabdus, 181, 183, 191
inermis (Candacia), 182, 187, 200,

201, 210
Labidocera, 183, 191
Leuckartia, 204

263

Copepoda — continued.

longicornis (Eucheeta), 183, 190
longimana (Candacia), 198
Lubbockia, 181, 191
lucens (Metridia), 181, 192
Lucicutia, 181, 183, 192, 202, 203,

204, 205, 206, 210
major (Undechjeta), 183, 198
marina (Euchaeta), 181, 183, 190
media (Euchaeta), 183, 190
media (Oncaea), 181, 183, 193
mediterrasnea (Oncaea), 181, 183
messinensis (Euchirella), 183, 190
Metridia, 181, 183, 192
miles (Gaetanus), 183, 191
minor (Calanus), i8o, 181, 182, 186
minor (Scolecithricella), 181, 197
minor (Undechaeta), 183, 198
monachus (Eucalanus), 183, 189
mucronatus (Aegisthus), 182, 185
mucronatus (Eucalanus), 183, 189
nana (Oithona), 181, 192
nasutus ( Rhinocalanus), 181,183,196
neptuni (Scolecithricella), 181, 197,

206, 207, 208, 209, 210
nigromaculata(Sapphirina), 183, 196
Oithona, 181, 183, 192, 193
Oncaea, 181, 183, 193
opalina (Sapphirina), 183, 196
ovalis (Corycaeus), 180, 181, 1S8
pachydactyla (Candacia), 180, 182,

187
Pachysoma, 183, 194
palumboi (Augaptilus), 182, 185
papilliger (Heterorhabdus), 181,

183. 191
Paracalanus, 181, 194
parvus (Paracalanus), 181, 194
pavo (Calocalanus), 181, 182, 186
pectinata (Candacia), 198
persecans (Scolecithrix), 183, ly/
Phacnna, 183, 194
Phyllopus, 183, 194
pilcatus (Eucalanus), 183, i8y^
Pleuromamma, 181, 183, 195
plumifera (Oithona), 181, 183, lyj
plumifera (Pontellina), 1S3, 195
Pontella, 183, 195
Pontellina, 183, 195
poppei (Chiridius), 182, 187

princeps (Metridia), 183, 192
Pseudodiaptomus, 181, 196
Pseudoscolecithrix, 209
rapax (Conaea), 183, 188
Rhinocalanus, 181, 183, 196
rigida (Oithona). 183, 193
robusta (Pleuromamma), 181, 183,

19s
robustus (Corycaeus), 180, 183, 188
salpae (Sapphirina), 181, 197
Sapphirina, 181, 183, 196, 197
scarlata (Sapphirina), 197
Scolecithricella, 181, 197, 206, 207,

20S, 209, 210
Scolecithrix, 181, 183, 197
securifer (Pontella), 183, 195
securifrons (Scolecithrix), 183, 197
serricaudatus (Pseudodiaptomus),

196
similis (Oithona), 181, 193
simplex (Xanthocalanus), 209
sinuicauda (Sapphirina), 181,183,19/
speciosus (Corycaeus), 183, 189
spinifera (Phatnna), 183, 194
spinifrons (Heterorhabdus), 181,

183, 191
spinosa (Euchaeta), 181, 190
streetsii (Chirundina), 182, 187
stylifera (Temora), 183, 19.8
subcrassus (Eucalanus), 183, 190
subtenuis (Eucalanus), 183, 190, 193
subtilis (Oncaea), 181, 194
tanneri (Heterorhabdus), 183, 191
Temora, 183, 198
tenuicornis (Calanus), 180, 182, 1S6
tenuimana (Candacia), 182, 187
tonsa (Euchaeta), 183, 190
truncate (Candacia), 182, 187
tuberosum (Pachysoma), 183, 194
typicus (Centropages), 181, 182, 187
Undechaeta, 183, 198
varicans(Candacia),i82,i87,i98, igy
venusta (Euchirella), 183, 190
venusta (Metridia), 183, 192
venusta (Oncaea), 181, 183, 194
vulgaris (Calanus), 182, 186
Xanthocalanus, 181, 198. 209, 210
xiphias (Pleuromamma), 181, 183,

195

{b) Localities.

Agulhas, 177, 178, 179, 184,
186, 187, 188, 1 89, 190,
192, 194, 196, 197, 198

Bluff Point, 182

Buffalo Bay, 180

Cape Infanta, 179, 180

Cape Natal, 182

Cape St. Blaize, 180

185, False Bay. 179

191, Lion's Head, 179

Morewood Cove, 182
Mossel Bay, 180
Port Shepstone, 182
Sebastian Bluff, 180
Table Mountain, 179

264

MARINE ANNELIDS.

(a) Generic and Specific Names.

Amphictenidie, 76
Amphitrite volutacornis, 83
Aphrodite squamata, 19
Arabella (Aracoda) iricolor, 46
Aracoda coerulea, 46
Ariciidse, 63
Audouinia fiiigera, 68
Autolytus prolifer, 35
Bispira volutacornis, 59, 83
Capitellida;, 69

Cha'topterus variopedatus, 59, Oi
Cirratulidae, 67
Cirratulus australis, 67
Cirratulus borealis, 67
Cirratulus capensis, 6y
Cirratulus cirratus, 59, 67
Cirratulus Lamarckii, 68
Cirratulus tentaculatus, 59, 68
Clymene lumbricalis, 71
Dasybranchus caducus, 59, 70
Dasyniallus caducus, 70
Distylia volutacornis, 83
Ergasilidae, 90
Eteone spitsbergensis, 35
Eriphyle capensis, 42
Eulalia capensis, 34
Eulalia viridis, 34
Eunice aphroditois, 42
Eunice gigantea, 42
Eunoa capensis, 28
Euphione Elisabethae, 27
Euphrosyne capensis, 22
Flabelligera affinis, 55
Flabelligera luctator, 53
Flabelligera Marenzelleri, 55
Glycera convoluta, 19
Gastrodelphys Clausii, 87
Haleminthidae, 69
Hemilepidia tuberculata, 29
Hermella capensis, 74
Hermellidas, 74
Leodice gigantea, 42
Lepidonotus clava, 19
Lepidonotus squamatus, 19
Lepidonotus wahlbergi, 19, 25
Leprsea pterochseta, 79
Lichomologus, 90
Lumbriconereis, 50
Lumbriconereis cavifrons, 50
Lumbriconereis gracilis, 50
Lumbriconereis Pettigrewi, 48
Lumbriconereis tetraurus, 48
Lumbriconereis tricolor, 46
Lumbricus cirratulus, 67
Lysidice capensis, 40
Macloria gigantea, 40
Maldane amphiglypta, ^^

Maldane elongata, ys
Maldane glebifex, 75
Maldanida\ 71
Marphysa Leidii, 45
Marphysa sanguinea, 44
Mastigonereis podocirra, ^7
Nereis cylindrica, 76
Nereis iricolor, 46
Nereis prolifera, 35
Nereis (Nereilepas ) Stmipsoui, 37
Nereis (Mastigonereis) variegata, ^^y
Nicomache capensis, 7 1
Nicomache lumbricalis, 59, 71
Nicomache Mclntoshii, 72
Notocirrus capensis, 46
Oligolepis violaceus, 59
Pectinaria belgica, 79
Pectinaria capensis, 19, 76
Pherusa tetragona, 53
Pionosyllis Mahngreni (probably

var. P. prolifera, Krohn), 35
Pista cristata, 19
Polyeunoa isevis, 59, 60
Polynoe attenuata, 29
Polynoe capensis, 28
Polynoe (Macellicephala) mirabilis,

Polynoe trochiscophora, 25
Potamilla reniformis, 59, 81
Praxilla capensis, 73
Praxilla pra^termissa, var. capensis,

73-
Protula capensis, 91

Protula tubularia, 91

Sabella lumbricatis, 71

Sabella renformis, 81

Sabella saxicara, 81

Sabellidse, 81

Sabellaria (Pallasia) capensis, 74

Sabelliphilus bispirae, 86

Sapphirinida;, 90

Schmardanella pterochaeta, 79

Sigalion Edwardsi, 19

Sigalion Idunae, 32

Sthenelais boa, 32

Stylaroides, 53

Syllis capensis, 36

Syllis cornuta, 37

Tecturella luctator, 53

Terebella pterochseta, 79

Terebellidse, 79

Thelepus, 80

Theodisca (Anthostomata) hexa-

phyllum, 63
Theodisca liriostoma, 66
Lrophonia capensis, 52
I rophonia plumosa, 53

265

ECHINOIDEA.

{) Generic and Specific Names.

acutus (Echinus), 170
australe (Echicodardium), 174
Brissopsis, 175
Brissus, 175

carinata (Pourtalesia), 172
crassa (Palaeolampas), 172
Echinocardium, 174
Echinolampas, 172
Echinus, 170, 171
elongata (Lovenia), 17 j
elongatus (Spatangus), 175
flavescens (Echinocardium), 174
flavescens (Spatangus), 174
fragilis (Brissus), 175

fragilis (Schizaster),
gilchristi (Echinus),

17s
170

Lovenia, 173

lyrifera (Brissopsis), 175

lyrifer (Brissus), 175

naresianus (Urechinus), 173

Palaeolampas, 172

Pourtalesia, 172

rangi (Echinolampas), 172

raschi (Spatangus), 173

Schizaster, 175

Spatangus, 173, 174

Urechinus, 173

(6) Localities.

Algoa Bay, 174

Bird Island, 175

Cape Hang Klip, 174, 17 s

Cape Morgan, 168

Cape Natal, 169, 172

Cape Point, 169, 171, 17^

Cape Seal, 171, 173

Cape St. Blaize, 170, 171

Cove Rock, 174

East London, 168, i7<j

False Bay, 171, 1 74

Knysna, 170

Kromhout, 172

Lion's Head, 174, 175

Mossel Bay, 170, 174

175

, 170,
170
170

Nahoon River, 244
Paternoster Point, 174
Paulsberg, 1 74
Sebastian Bluff, 170, 174
Simon's Town,
Somerset West
St. James, 170
Struys Point, 175
Table Mountain, 175
Three Anchor Bay, 170
Tugela River, 169
Umhlangukulu River, 168
Umhloti River, 173
Zwartklip, 171

ASTEROIDEA.

(11) Generic and Specific Names.

acuminatus (Psilaster), 245
africana (Luidia), 245
Archasteridae, 242
Asterias, 251, 25:
Asteriidae, 251
Asterma, 248
Asterinidae, 24S
Asteriscus, 248
Astrogonium, 246
Astropecten, 241, 243
'\stropectinidae, 242
baccatus (Calliaster), 247
calamaria (Asterias), 251
Calliaster, 247
capensis (Asterias), 252
capensis (Astropecten), 244
capensis (Pteraster), 250
capensis (Retaster), 250
coccmea (Asterina), 248

coccineus (Asteriscus). 24S

Cribrella, 250

Crossaster, 249

Culcita, 248

Echinaster, 250, 252, 253

Echinasteridte, 250

felipes (Stichaster), 249

glacialis (Asterias), 252

Henricia, 250

Hymenaster, 250

k< rguelensis (Leptoptychaster), 242

Leptoptychaster, 242

Luidia, 245

novemradiatus (Palmipes), 248

ornata (Cribrella), 250

ornatus (Echinaster), 250

ornata (Henricia^ 250

Palmipes, 248

penicillatus (Crossaster), 249

266

Asteroidea — coiitimicd.

penicillatus (Solaster), 249
Pentacerotidac, 248
Pentagonaster, 241, 246
Perknaster, 253
pontoporaeus (Astropecteii)

243
Pseudarchaster, 242
Psilaster, 245
Pteraster, 250
Pterasteridse, 250
Retaster, 250
simplex (Cribrella), 250

Solaster, 249
Solasteridae, 249
Stichaster, 249
! Stichasterida-, 249
241, Stolasterias, 252

tesselatus (Pseudarchaster), 242
tuberculatum (Astrogonium), 246
tuberculatus (Pentagonaster), 241,

246
veneris (Culcita), 248
volsellata (Asterias), 252

(6) Localities.

Buffalo Bay, 244

Buffels Bay, 245, 250, 251

Bird Island, 244, 252

Cape Infanta, 244

Cape Natal, 244, 248

Cape Point, 242, 243, 245, 249, 250,

251, 253
Cape St. Blaize, 243, 244, 246, 247,

251, 252
Constable Hill, 250
Cove Rock, 247
East London, 243
False Bay, 243, 244, 245, 247, 248,

250
Great Fish Point, 252

251

243, 24?,

'■47.
253

Hoetjes Bay, 250,
Knysna, 244, 252
Lion's Head, 242,

249, 251
Mossel Bay, 244, 248, 249, 251,
Rockland Point, 243
Rocky Bank, 247
Seal Island, 248, 251
Sebastian Bluff, 243, 249
Simon's Bay, 253
Somerset West, 253
Table Mountain, 242, 247, 249, 250,

251
Three Anchor Bay, 251

OPHIUROIDEA.

((?) CiENERIC A-ND SPEtlEIC NaMES.

Amphiura, 258
aristulata (Ophiothrix), 258
Astrophyton, 259, 260
capensis (Ophiozona), 256, 258
Cladophiura;, 259, 260
dentatus (Ophioscolex), 259
Euryale, 260

Gorgonocephalus, 259, 260
■ ncana (Amphiura), 258
Ophioderma , 255
Ophioscolex, 259
.Ophiothrix, 258
Ophiothamuus, 258

Ophiozona, 256, 257, 258
Ophiura, 257

pourtalesii (Astrophyton), 239
pourtalesi (Gorgonocephalus), 259
remotus (Ophiothamnus), 258
roseo-coerulans (Ophiothrix), 258
Streptophiura', 259
triglochis (Ophiothrix), 259
trimeni (Ophiura), 257
verrucosum (Euryale), 260
verrucosus (Gorgonocephalus), 260
wahlbergi (Ophioderma), 255
Zygophiurae, 255

(6) Localities.

Algoa Bay, 258
Buffalo River, 259
Buffels Bay, 260
Cape Natal, 258
Cape Point, 257
Cape St. Blaize, 257
Cape St. Francis, 257
False Bay, 258, 259
Great Fish Point, 258

Lion's Head, 257, 258, 259
Mossel Bay, 258
Port Natal, 255
Ri}' Bank, 258
Sebastian Bluff, 260
Table Mountain, 257, 259
Tugela River, 258, 259
Vasco da Gama, 259

267

TURBINOLID CORALS.

(a) Generic and Specific Names.

aequicostatus (Heterocyathus), 105,

125
berteriana (Caryophyllia), 112
capense (Desmophyllum), 96
capensis (Caryophyllia), 113
capensis (Duncauia), 120
capensis (Duncania), Anatomy of,

127
Caryophyllia, 112
Ceratotrochus, 118
cincticulatus (Trochyocyathus), 103
Corals (Turbinolid), 95
Corals (Turbinolid), Anatomy of,

122
Corals (Turbinolid), Comparison of,

io5
Corals (Turbinolid), Variation in,

95, loi, 102, 107, 108, no, 116
cornu (Cyathoceras), t2i
Cyathoceras, 1 2 1

Desmophyllum, 96

Duncania, 120

epithecata (Caryophyllia), 114

epithecata (Caryophyllia), Anatomy

of, 125
ephyala (Caryophyllia), 117
Flabellum, 97

gilchristi (Sphenotrochus), 98
Heterocyathus, 104
johnsoni (Ceratotrochus), 118
pavoninum (Flabellum), 98
rawsonii (Trochoc^-athus), 100
rawsonii (Trochocyathus), Anatomy

of, 124
rubrum (Flabellum), 98
smithi (Caryophyllia), Anatomy of,

126
Sphenotrochus, 98
sibogae (Flabellum), 98
Trochocyathus, 99

(/;) Localities.

it:

Buffalo River, 98, 102, ii

Cape Hangklip, 97

Cape Natal, 105, 117

Cape St. Blaize, 102, 116, iiS

Cape Point, 102, 1 17

Cape Vidal, 102, 117, 118

118 Kowie, 99

Lion's Head, 1 16
O'Neil Peak, 105, 1 13
Tongaati River, 105
Umhloti River, 105
Vasco da Gama. 102, 1 16,

121

ALCYONARIA.

((() Generic and Specific Names.

Acanella, 225

Acanthogorgia armata, 213, 225
Acanthogorgia hirsuta, 225
Acanthogorgia longiflora, 225
Acanthogorgia ramosissima, 225,

226
Acanthogorgia spinosa, 225
Acrophytum claviger, 212
agaricus (Anthomastus), 217
albicans (Gorgonia), 213, 228
albicans (Platygorgia), 228
Alcyonacea, 212
Alcyoniidae, 2 1 2

Alcyonium antarcticum, 212, 216
Alcyonium digitatum, 215
Alcyonium fulvum, 216
Alcyonium pachyclados, 2 1 2
Alcyonium purpureum, 212, 215,

217
Alcyonium sphaerophorum, 216
antarcticum (Alcj-onium), 212, 215
Anthelia capensis, 212, 214
Anthomastus, 214, 218
Anthomastus agaricus, 217
Antliomastus grandifiorus, 212, 217

Anthomastus steenstrupii, 218
Anthoptilum grandiflorum, 214, 233
Anthoptilum thomsoni, 233
armata (Acanthogorgia), 213, 225
Briareidae, 212
Calypterinus, 232
capensis (Anthelia), 212
capensis (Euplexaura), 213, 214,

229, 230
capensis (Gorgonia), 213, 229, 230
capensis (Heteroxenia), 212
capensis (Malacogorgia), 213, 214,

226, 227
capensis (Primnoisis), 213, 220
Cavernularia elegans, 214
Cavernularia obesa, 214
Cavernulariidae, 214
Ceratoisis grandiflora, 225
Ceratoisis grayii, 224, 225
Ceratoisis nuda, 225
Ceratoisis paucispinosa, 225
Ceratoisis Phillippinensis, 225
Ceratoisis ramosa, 214, 224
chrj'santhus (Wrightella), 220
Chrysogorgia, 221, 222, 223

268

Alcyonaria — continued.

claviger (Acrophytnm), 212

Clavulariidae, 212

coccinea (Wrightella), 212, 219, 220,

221
crista (Gorgonia), 213, 215, 229
Daniana (Eugorgia), 230
Dasygorgiidae, 213, 221, 226
dichotoma (Melitodes), 212, 219
elongata (Juncella), 213, 233
Eugorgia (Daniana), 230
Eugorgia gilchristi, 213, 214, 215,

229, 230
Eunephthya thj'rsoides, 21:?
Eunicella palma, 227, 228
Eunicella papillosa, 213, 228
Euplexaura capensis, 213, 214, 229,

230
erythraea (Wrightella), 220
flammea (Gorgonia), 213, 215, 227,

228, 229
flexilis (Juncella), 232
flexilis (Trichogorgia), 213, 222, 223
fulvum (Alcj^onium), 216
gilchristi (Eugorgia) 213, 229, 230
Gorgonacea, 212
Gorgonella stricta, 213, 214
Gorgon ellidae, 230, 231
Gorgonia albicans, 213, 228
Gorgonia capensis, 213, 215, 229
Gorgonia crista, 213, 214
Gorgonia flammea, 213, 215, 227,

228, 229
Gorgonia liitkeni, 229, 230
Gorgonia palma, 227, 228
Gorgoniidae, 213, 226, 227
grandiflora (Ceratoisis) 225
grandiflora (Virgularia), 233
grandiflorum (Anthoptilum), 233
grandiflorus (Anthomastus), J12,

217
grayii (Ceratoisis), 224, 225
Heteroxenia capensis, 212
Isidae, 213, 224
juncea (Juncella), 232
Juncella elongata, 213, 233
Juncella liexilis, 232
Juncella juncea, 232
Juncella spiralis, 213, 214, 231
liitkeni (Gorgonia) 229, 230

Malacogorgia capensis, 213, 214, 226
Malacogorgiidae, 213, 226
mauritiensis (Villogorgia), 213
Melitodes dichotoma, 212, 219
Melitodidae, 212, 219
Muriceidae, 213, 225
Nephthyidae, 212
nuda (Ceratoisis) 225
pachyclados (Alc^'onium) 212
palma (Eunicella), 227, 228
palma (Gorgonia), 227, 228
papillosa (Eunicella) 213, 228
paucispinosa (Ceratoisis) 225
Pennatulacea, 214, 233
Phillippinensis (Ceratoisis), 225
Platygorgia albicans, 228
Plexauridae, 213, 227
Primnoidae, 224, 232
Primnoisis capensis, 213, 220
purpureum (Alcyonium), 212

217
ramosa (Ceratoisis), 213, 214,
ramosissima (Acanthogorgia),

226
reinwardti (Virgularia), 214
Sarcophytum trochiforme, 212,
spinosa (Acanthogorgia), 225
spiralis (Juncella), 213, 214, 231
Spongioderma verrucosum, 212
steenstrupii (Anthomastus), 218
Stolonifera, 212
stricta (Gorgonella), 2 1 3, 214
thomsoni (Anthoptilum), 233
thyrsoides (Eunephthya), 212
Trichogorgia flexilis, 214, 222, 223
trochiforme (Sarcophytum), 212,

215, 218
Verrucella, 230, 231
verrucosum (Spongioderma), 212
Villogorgia mauritiensis, 213
Virgularia grandiflora, 233
Virgularia reinwardti, 214
Virgulariidae, 214
Wrightella, 213
Wrightella chrysanthus, 220
Wrightella coccinea, 212, 219, 22c,

221
Wrightella erythraea, 220
Xeniidae, 212

215

224
225,

218

{b) Localities.

Algoa Bay, 212, 213, 226
Cape Agulhas, 219
Cape Morgan, 213, 231
Cape Recife, 212, 213, 217, 22
Cape St. Blaize, 212, 213, 22S
East London, 213
False Bay, 212, 219

Lion's Head, 214, 233
Mossel Bay, 212, 213, 215, 216, 219
Port Alfred, 214
Riy Bank, 213
St. Francis Bay, 214
Vasco de Gama Peak, 213, 219, 224.
22s

269

CORRIGENDA.

For kynsaensis read knysnaensis . . . . . . , . . . page 2

For Knsyna read Knysna . . . . . . . . . . . . pa.oe 2

• 'jKk»

I

MBL/WHOI LIBRARY

H 1AF7 ^

^

. ?»