Ill

m

3r

A MEMOIR

ON

THE ECHINODERMATA

OF THE

AECTIC SEA TO THE WEST OF GREENLAND,

BY
P. MARTIN DUNCAN, M.B. (LOND.), F.R.S., F.G.S., F.L.S.,

HON. FELLOW KINO'S COLLEGE, LONDON, CORRESPONDENT OP THE ACADEMY OF NATURAL SCIENCES of

PHILADELPHIA, TJ. S. ;

AND

W. PERCY SLADEN, F.G.S., F.L.S., &c.

LONDON:
JOHN VAN VOORST, 1 PATERNOSTER ROW, E.G.

MDCCCLXXXI.

ALKHK

FLAJIMAM.

PRINTED BY TAYLOR AND FRANCIS,
11ED LION COURT, FLEET STREET.

CONTENTS.

Pago

Preface v

List of the Species noticed and described vii

The Holothuroidea 1

The Echinoidea 19

The Asteroidea 23

The Ophiuroidea and Astrophytidse 54

The Crinoidea (Comatulse) 73

General Conclusions 79

Index 81

Six Plates, and their Descriptions.

PREFACE.

THE Echinodermata which form the subject of this memoir were collected during the
voyage of H.M.SS. 'Alert' and 'Discovery,' under the command of Sir George S. Nares,
E.N., K.C.B., F.E.S., to the Polar Sea, by the Naturalists attached to the Expedition,
Capt. (now Major) H. W. Feilden and Mr. Hart. The greater number were obtained
from 79° 20' N. lat. to 82° 27' N. lat. inclusive— that is, from Franklin-Pierce Bay to
Floeberg Beach. Some forms were captured on the outward voyage, in 65° N. lat.;
and others were dredged in the Expedition of H.M.S. ' Valorous,' between 66° 56'
and 70° 30' N. lat., by Dr. J. Gwyn Jeffreys. F.R.S. One specimen was taken by
Mr. A. C. Horner (who accompanied Sir Allen Young in the 'Pandora'), in Smith's
Sound, 78° 19' N. lat. The Arctic Circle has been regarded as our southern limit.

The collections were made under great difficulties, and especially those from the
highest latitudes : dredging, when the tangles froze on coming out of the sea, could not
be attempted frequently. Nevertheless the specimens collected were numerous. They
were carefully cleaned ; and some were preserved in spirit, and others were permitted
to dry.

After being deposited at the Royal Society in 1877, the collection was forwarded
to the British Museum. Dr. Giinther, F.R.S., confided the specimens to the authors
of this Memoir ; and we wrote a brief description of them in the ' Annals & Magazine
of Natural History,' 1877, vol. xx. pp. 449-470. Subsequently we contributed a
short notice of these interesting objects to the admirable book written by Sir George

VI

Nares, F.R.S., entitled 'A Narrative of a Voyage to the Polar Sea during 1875-1876,'
published in 1878. In that short contribution we stated that the collection was so
interesting and the specimens were so variable, that we should publish their descrip-
tion in a separate Monograph.

The Government Grant Committee of the Royal Society voted a sum of money in
order to assist us in the production of this work.

We beg to thank Dr. Giinther, F.R.S., in whose charge the specimens now are,
Dr. J. Gwyn Jeffreys, F.R.S., the Rev. A. M. Norman, FX.S., Dr. Carpenter, C.B.,
F.R.S., and Major Feilden especially for their kind assistance.

LIST OF THE SPECIES NOTICED AND DESCRIBED.

HOLOTHUROIDEA.

CUCUMARIA FRONDOSA (Gunner), Forbes.
CUCUMARIA CALCIGERA (Stimpson), Selenka.
ORCULA BARTHII, Troschel.
PSOLUS PIIANTAPUS (Strusscnfeldt), Jaeger.
PSOLUS FABRICII (Duben 8f Koren), Lutken.
CHIRODOTA L^VIS (Fabricius), Grube.
MYRIOTROCHUS KINKII, Steenstrup.

ECHINOIDEA.

STRONGTLOCENTROTUS DROBACHIENSIS (Muller), A. Agassiz.

ASTEROIDEA.

ASTERACANTHION POLARE, Muller & Troschel.

ASTERACANTHION GRCENLANDicuM (Steenstrup), LutJcen.
STICHASTER ALBULUS (Stimpsori), Verrill.
CRIBRELLA OCULATA (Linck), Forbes.
PEDICELLASTER PAL^EOCRYSTALLUS, Sladen.
CROSSASTER PAPPOSUS (Linck), Muller & Troschel.
SOLASTER ENDECA (Gmelin), Forbes.
LOPHASTER FURCIFER (Duben & Koreri), Verrill.
PTERASTER MILITARIS (0. F. Muller), Muller & Troschel.
CTEXODISCUS CORNICULATUS (Linck), Perrier.

OPHIUROIDEA.

OPHIOPLEURA ARCTICA, Duncan.
OPHIOGLYPHA SARSII, Lutken.
OPIIIOGLYPIIA ROBUSTA, At/res.
OPHIOGLYPHA STUWITZII, Lutken.
OPHIOCTEN SERICEUM, Forbes.
OPHIOPIIOLIS BELLIS, Linck.
AMPIIIURA HOLBCELLI, Lutken.
OPHIACANTHA SPINULOSA, Muller & Troschel.

ASTROPHYTID5).

ASTROPHYTON AoASSizn, Stimpson.

CRINOIDEA.

ANTEDON ESCIIRICHTII (Muller), Verrill.
ANTEDON CELTICA (Barrett), Norman.
ANTEDON PROLIXA, Sladen.

A MEMOIR

ON THE

ECHINODERMATA OF THE ARCTIC SEA

TO TIIE

WEST OF GREENLAND,

HOLOTHTTROIDEA.

JUDGING from the results of such dredging as has hitherto been conducted in high
latitudes, the Holothuroidea as a group do not appear to extend in the North-Atlantic
area to a very great distance within the Arctic Circle. Exception must be made,
however, in the case of Myriotrochus Rinkii, Stp., which was obtained by the Polar
Expedition of 1875-76 in Discovery Bay, lat. 81° 41' N., the most northern locality at
which any Holothuroid has yet been found. The other members of the order have not
been recorded in this area beyond the 70th parallel ; but as they occur in higher lati-
tudes elsewhere, it has been deemed expedient to include those which extend beyond
the Arctic Circle (66° 30' N.) in the present description.

The following is a list of the Holothuroids in question : —

1. Cucumaria frondosa (Gunner), Forbes.

2. Cucumaria calcigera (Stimp.), Selenka.

3. Orcula Barthii, Troschel.

4. Psolus phantapus (Strussenf.), Jager.

5. Psolus Fabridi (D. & K.), Liitken.
C. Chirodota Icevis (Fabr.), Grube.

7. Myriotrochus Rinkii, Steenstrup.

B

2 ON THE ECHINODERMATA OF THE

CUCUMARIA FBONDOSA (Gunn.), Forbes. Plate I, Figs. 1 & 2.

1767. Holothuria frondosa, Gunnerus, Act. Holm. 1767, p. 115, pi. iv. figs. 1, 2.
1776. Holothuria frondosa, 0. F. Miiller, Zool. Dan. Prodr. no. 2802.
1776. Holothuria pentactes, 0. F. Miiller, Zool. Dan. Prodr. no. 2806.
1780. Holothuria pentactes, Fabricius, Fauna Groenlandica, p. 352. no. 343.
1780. Holothuria frondosa, Fabricius, Fauna Grcenlandica, p. 353. no. 344.
1780. ? Holoihuria minuta, Fabricius, Fauna Groenlandica, p. 354. no. 346.
1788. Holothuria pentacta, Gmelin (pars?), Syst. Nat. Linn. ed. xiii. p. 3139.
1788. Holothuria frondosa, Gmolin, Syst. Nat. Linn. ed. xiii. p. 3138.

1788. Holothuria pentactes, 0. F. Miiller, Zool. Dan. vol. i. p. 36, pi. xxxi. fig. 8 (juv.).

1789. Holothuria pentactes, Abilgaard, Zool. Dan. vol. iii. p. 45, pi. cviii. figs. 1-4.

1806. Holothuria pentactes, Vahl et Rathke, Zool. Dan. vol. iv. pp. 3-7, pis. cxxiii.-cxxvii.
1816. Holothuria frondosa, Lamarck, Anim. s. Vert. ed. i. vol. iii. p. 73.

1833. Pentacta frondosa, Ja'ger, De Holothuriis, p. 12.

1834. Cuvieria frondosa, Blainvillc, Manuel d'Actinologie, p. 192.

1835. ? Dactylota minuta, Brandt, Prodr. descrip. anim. ab Mertensio obs., fasc. i. p. 45.
1835. ? Cladodactijla pentactes, Brandt, Prodr. descrip. anim. ab Mertensio obs., fasc. i. p. 45.
1839. Holothuria grandis, Forbes & Goodsir, Athenaeum, no. 618, p. 647.

1839. Cucumaria fucicola, Forbes & Goodsir, Athenaeum, no. 618, p. 647.

1841. Cucumaria frondosa, Forbes, Hist. British Starfishes, p. 209.

1841. Cucumaria fucicola, Forbes, Hist. British Starfishes, p. 227.

1841, ? Cladodactyla pentactes, Gould, Invert, of Massachusetts, p. 345.

1844. Cucumaria frondosa, Diiben & Keren, K. Vet.-Akad. Handl. 1844, p. 293.

1852. Bothryodactyla grandis, Ayres, Proceed. Boston Soc. Nat. Hist. vol. iv. p. 52.

1852. Bothryodactyla affinis, Ayres, Proceed. Boston Soc. Nat. Hist. vol. iv. p. 145.

1852. ? Cucumaria fucicola, Forbes, Sutherland's Journ. of a Voyage &c,, vol. ii. Append, p. ccxiv.

1853. Pentacta frondosa, Stimpson, Syn. Mar. Invert. Grand Manan, p. 16.

1857. Cucumaria frondosa, Liitken, Vid. Meddel. N. Forening i Kjobenhavn, 1857, p. 2.

1857. Cucumaria frondosa, M'Andrew & Barrett, Ann. & Mag. Nat. Hist. sor. 2, vol. xx. pp. 43, 45,

1861. Cucumaria frondosa, Sars, Oversigt af Norges Echinodermer, p. 100.

1863. Pentacta frondosa, Stimpson, Proc. Acad. N. Sci. Philad. 1863, p. 142.

1866. Pentacta frondosa, Verrill, Proceed. Boston Soc. Nat. Hist. vol. x. pp. 352, 357.

1866. ? Pentacta minuta, Verrill, Proceed. Boston Soc. Nat. Hist. vol. x. p. 353.

1867. Cucumaria frondosa, Selenka, Zeitsch. f. wiss. Zool. Bd. xvii. p. 347.

1868. Cucumaria frondosa, Semper, lleisen im Archipel der Philippinen, Holothurien, pp. 52, 234.

1869. Cucumaria frondosa, Pourtales, Bull, Mus. Comp. Zool. Harvard, 1869, p. 359.

Body subcylindrical or subpentagonal in young stages, ovate in form when at rest
and contracted, but capable of considerable elongation ; uniformly rounded at both
extremities.

The ambulacral sucker-feet are arranged in five longitudinal series, each being a
double row in which the tube-feet alternate ; in old specimens, however, a quadruple
disposition in the middle of a series is not unfrequent, consequent on the action of
growth-crowding, and a few additional irregularly-placed suckers also occur on the
dorsal interradial areas, all the suckers being capable of entire retraction.

The skin is very thick, tough, and smooth, although specimens preserved in spirit
are generally much wrinkled and puckered up, owing to the great contraction which
the animal is capable of exerting prior to death through the medium of the powerful
and largely developed muscular system which it possesses.

ARCTIC SEA TO THE WEST OF GREENLAND. 3

The tentacles are ten in number, frondose, equal-sized, and very robust in habit.
The oral armature is, as a whole, large, although the calcareous mouth-ring in some
examples is but very imperfectly calcified. The radial elements are somewhat of
the form of an inverted Y, the muscular bands being attached to the extremity of the
upward odd prolongation ; the interradial pieces are smaller, and often in the form of
a simple arch. The present writers have been unable to detect in any specimens they
have examined the secondary ring mentioned by Selenka * ; but in a single small
example there seem to be traces of an incipient calcareous prolongation proceeding
from the interradial piece, and giving it the " Y " shape similar to that of the radial
elements. Perhaps such a development may be a stage towards the structure described
by our learned contemporary.

In a young individual the Polian vesicle is single, and consists of a simple elongated
sac, of moderate capacity, and about one third the length of the body. The respiratory
organs, the intestine, and the generative organs are largely developed.

The muscular system is very powerful, the longitudinal bands being of such a
breadth as to partially overlay the ampulla? of the sucker-feet ; at the oral extremity
they taper off rapidly, and their continuation forms the double band of the extensor
muscles of the mouth-apparatus. The TO. retractores are very strong and stout, the
band being attached to the mouth-ring between the two fascicules of the extensor
muscle. The series of transverse muscles are numerous and closely placed.

Very conflicting statements occur respecting the calcareous spicules in the cuticle,
not only as to the form, but even as to the presence at all of such bodies in C.frondosa.
Diiben and Koren described the integument as covered with calcareous granules which
are irregular, deformed, and never perforated. Selenka states that he has found no
plates whatever in the general body-skin, but only very minute arragonite needles
•006 millim. in length ; whilst both Semper and Liitken specify that large and
perforated plates occur in this Holothuroid, but are only to be met with in the neigh-
bourhood of the feet. Our observations tend to confirm the latter naturalists, although
there would appear to be very considerable variation in the quantity of spicules present
in different specimens. In some examples, for instance, the skin seems entirely devoid
of any calcareous deposits whatever, whilst in others spicules are more or less numerous
throughout the body ; and this has been the case in moderately large examples. Gene-
rally, however, it must be said that spicules are most frequent in the neighbourhood
of the ambulacral rows ; but it would seem that no fixed rule of any kind can be
definitely laid down as to their occurrence in this species, whilst the same remark would
even hold good as to their form.

Size. — This species attains perhaps the largest size of any of the Cucumarice,
occasionally measuring a foot from end to end, large specimens being able to extend
themselves to twice or three times this length.

Colour. — In colour C. frondosa is of a dark purple mingled with grey, shading off
to ashy grey or white on the underside. Young examples are lighter and also of a
more uniform shade, generally pinkish or flesh-coloured.

* " Beitriige zur Anatomie und Systematik der Holothuricn," Zeitsch. f, wiss. Zool. Bd. xvii. p. 347.

B2

4 ON THE ECHINODERMATA OF THE

Premature Form. — In early stages of growth the body is much more pentagonal
in fonn and usually lighter in colour, which not unfrequently approaches a pinkish or
flesh tint. The ambulacral sucker-feet are well spaced and arranged in single almost
straight lines. In a small specimen 9 millims. in length the two dorsal ambulacral
series are not so fully developed as the three ventral ; they contain fewer suckers ; and
these are arranged in an almost straight line, except at the extremities, where the zigzag
alternating character of the series is clearly manifest. The oral tentacles, although
only partially extended in the specimen under notice, are already thick, frondose, and
many times divided, whilst the body-skin is filled with regularly-spaced calcareous
spicules roundish in form and punctured with holes, the solid interspaces being broader
than the apertures. Comparing this individual with another somewhat larger, 20
millims. in length, it will be noted that the body is proportionately more elongate, the
ambulacral feet more numerous, and now arranged distinctly in double rows of alter-
nating suckers. The tentacular plume is slightly fuller, but still exactly the same in
general character as in the earlier stage ; in fact the changes above noted are the only
conspicuous accompaniments of increased size perceptible to the naked eye. On micro-
scopic examination, however, of the larger specimen it is found that no spicules are
present in the skin ; and this is a feature which at once constitutes the most striking
difference between the two stages of growth. Such a circumstance is very remarkable ;
and although we are unable to say, with the limited amount of material at our disposal,
whether this is a state of things which always obtains, at least two other individuals, of
succeeding and still premature stages of growth, are equally wanting in spicules — a
character which, as previously observed, is not unfrequent in the adult form of
C. frondosa.

It is noteworthy that the features presented by the young specimen of 9 millims.
accord exactly with those given by Liitken as characterizing C. minuta, Fabr. (excepting,
we imagine, the tentacles); but on this point Liitken does not say much, as these organs
were only partially extended in his specimen. If the view, therefore, which is here
taken be correct, it would lead naturally to the deduction that C. minuta is nothing
more than the young of C. frondosa. Before this can be definitely asserted, however,
the examination of the growth-phases in a greater number of examples of the present
species would be desirable ; still we feel bound to say, after a very careful study of all
the available material, that we are unable to separate the young form above described
from the series of undoubted C. frondosa, the only feature in which it differs altogether
from the older stages being the presence of the calcareous bodies in the integument.

From the description given, it would seem that the Holothuroid named by Forbes
and Goodsir C. fucicola is a young form of the present species, a determination now
generally concurred in by most naturalists.

Variations. — Amongst the list of synonyma will be found Rotliryodactyla grandis,
Ayres, which we have been led to include rather from a diffidence against dissenting from
the opinion of so many eminent writers upon Holothuroids than from personal conviction.
Indeed there would seem to be great doubt about the absolute identity of this form ;
for, although the main characters recited in the cursory description accord well enough

ARCTIC SEA TO THE WEST OF GREENLAND. 5

with those of C. frondosa, they are by themselves quite insufficient for comparative
determination ; whilst, on the other hand, it is definitely stated that the calcareous sup-
ports, though few, " are in the form of slender, perforated, crested spicula, similar to
those found in Thyone and Thyonidium " (1. c. p. 53). Such crested spicules certainly
do not occur in any specimens of C. frondosa from the North-European seas ; whilst
from the fact that they are specially noted by Mr. Ayres (and it is a feature much
more readily passed over than not) we are inclined to regard Bothryodactyla grandis, if
not an independent species, certainly with but little doubt as a well-marked variety.
In further support of this opinion it may be noted that Sars examined specimens of the
American C. frondosa both from Massachusetts and Fundy Bay, and specially asserts
their identity with the European type.

Distribution.

a. Greenland : Godhavn, lat. 69° 14' N. (Stimpson, Ltitken, ' Valorous' Exped.).

b. North of American Continent: Assistance Bay, about lat. 74° N. (Penny's
Exped.}, the most northern locality on record ; Labrador ( Verrill) ; St. George's Bank,
30 fms. (Verrill); Grand Manan, low water, stony bottom (Stimpson); Massachusetts
(Gould); Eastport, 20 fms., stony bottom (Verrill); Gulf of Georgia (Selenka);
^ Florida Reef, 118 fms. (Pourtales) ; San Francisco (Ayres).

c. North of European Continent : Spitzbergen (LutJcen); Scandinavian coasts, Ice-
land, Faeroe Islands, Shetlands, British Islands.

Description of the Illustrations of this Species on Plate I.

Fig. 1. A medium-sized specimen: natural size.

2. A young specimen of the same species : natural size.

CUCUMAKIA CALCIGEKA (Stimp.), Selenka. Plate I, Figs. 3-8.

1851. Pentacta calcigera, Stimpson, Proceed. Boston Soc. Nat. Hist. vol. iv. p. 67.

1852. ? Cucumaria Hyndmanni, Forbes, Sutherland's Journal of a Voyage &c. vol. ii. Append, p. ccxiv.
1857. Cucumaria Korenii, Liitken, Vid. Heddel. N. Forening i Kjobenhavn, 1857, p. 4.

1866. Pentacta calcigera, Verrill, Proc. Boston Soc. Nat. Hist. vol. x. p. 352.

1867. Cucumaria Korenii, Selenka, Zeitsch. f. wiss. Zool. Bd. xvii. p. 350.

1867. Cucumaria calcigera, Selenka, Zeitsch. f. wiss. Zool. Bd. xvii. p. 351.

1868. Cucumaria calcigera, Semper, Ecisen im Archipel der Philippinen, Holothurien, p. 53.
1868. Cucumaria Hyndmanni, Semper (pars), ibid. p. 269.

Body elongate and cylindrical, the posterior third tapering gradually to a fine
extremity ; anterior end more rounded, and the median thickness of the body main-
tained till very much nearer the oral region. The body is generally more or less curved
to one side, sometimes through more than half a circle, which gives to this species a
crescent- or even horseshoe-shaped contour. The ambulacral feet are arranged in single
pairs, and form five narrow, equally-developed bands. The suckers are placed very
closely together ; consequently the number extending from tip to tip is proportionally
great, whilst sometimes, in the middle part of a band, the pairs become reduplicated by

6 ON THE ECHINODERMATA OF THE

reason of extreme crowding. Owing to the densely tessellated and compact nature of
the integument, the sucker-feet are incapable of being perfectly retracted ; and this
circumstance, together with the size and general habit of the animal, affords a striking
superficial feature, which readily characterizes the species amongst its Arctic con-
geners.

The cuticle, although thin, is very strong, being indurated with calcareous plates,
of which two distinct layers are present ; in each of these the plates are quite different
both in size and form, and are so closely packed as to imbricate upon one another,
whilst the whole Holothuroid is rough and scabrous to the touch, in consequence of
a small spiculate elevation which rises from the centre of each of the plates that form
the superficial layer.

The tentacles are ten in number, two being much smaller than the rest. They are
of delicate habit, and redivided into many branches, which are fine and thin.

The mouth-ring is elongate, and resembles in a marked degree the form found in
Thyone. It is composed of ten plates, each of which is made up of two elementary
pieces, as may be well seen in the mouth-apparatus of a young specimen (Plate I, Fig. 4).
The radial elements are produced upward into a fine wedge-shaped peak, and down-
ward into two long, thin prolongations, which extend to the base of the apparatus, the
cleft formed between them being wide and well arched above. The alternating inter-
radial elements are somewhat wedge-shaped, equal in size to the upper portion of the
radial processes, and extend upward to the same height as these. The angle of
the plates, which abuts against the neighbouring radial member, is somewhat trun-
cate, and the suture between the two primary pieces slightly hollowed out in the
middle.

The Polian vesicle of a young individual is represented in Fig. 4, in which
it is seen as a very distended sac, emanating midway upon a fine short tube.

According to Dr. Liitken (I. c.) the respiratory organs in the adult animal cleave
immediately into four rays, one small and one larger on each side. They are but feebly
branched ; and although becoming generally thicker towards their extremities, it is only
here and there that any thing like a vesicle can be detected.

Eespecting the anatomy of the alimentary canal, it may be noted that the oesophagus
is moderately developed, with its outer wall villate or densely covered with minute
papilla?, and that the intestine is large and much convoluted.

The muscular system is somewhat small for a Cucumaria. The m. longitudinales
are thin and very narrow, and the m. transversales similarly slight. The m. retractores,
however, are strong and by far the most powerful bands of muscle in the animal ; they
are attached to the central part of the wedge-shaped upper portion of the radial member
of the calcareous ring, and join the longitudinal band of muscles midway between
the extremities. When the mouth-apparatus is retracted and these muscles are relieved
from tension, they appear very broad in comparison with the other bands, and are,
besides, much swollen out in their middle portion. The m. extensores are very fine
indeed and double, and are affixed to the radial pieces of the ring a little above the
attachment of the retractor muscles.

ARCTIC SEA TO THE WEST OF GEEENLAND. 7

The reproductive organs consist of simple tubes. A figure is given of their condi-
tion in an early stage of growth (Plate I, Fig. 5).

The calcareous plates and spicules which indurate the integument are highly cha-
racteristic. As previously stated, they are so closely placed as to imbricate upon one
another; and it is consequently very difficult to determine the form of individual pieces
until separated by treatment in caustic potash. The plates which form the outermost
layer vary in shape from roundish to oblong, and even irregularly stellate forms, their
greatest diameter ranging from 0'28 to 0'35 millim. (Liitken). They are perforated with
closely-placed holes, whose distance apart is less than their diameter, no plan being
discernible in their arrangement. From the centre of each rises a subconical elevation,
made up of fine calcareous rods, about one third as high as the diameter of the plate,
and which terminates with two or three spikelets. It is to these spicular plates that
the rough texture of the skin is due, and the granular appearance it presents when
examined with a hand-magnifier of low power.

Underlying the plates just described is another layer of oblong and much narrower
plates, also closely packed together and overlapping, and which are present in greater
number over a given area in consequence of their smaller size. These pieces are generally
punctured with two (or sometimes three) rows of holes, which become smaller in
diameter towards the extremities of the plate.

The skin of the sucker-feet is comparatively as densely plated as that of the body,
the spicules at the base being similar in character to those above described, but smaller
in size ; towards the extremity, however, they become much narrower, bear fewer holes,
and are somewhat more curved, whilst mixed up with them may be found a number
of quite rudimentary plates. When the great number of these scales and the method
of their arrangement are borne in mind, the reason of the suckers being incapable of
retraction becomes at once self-evident, as well as that of the small size of the internal
ampullae, which follows as a correlative result. The calcareous disk at the extremity of
the sucker is well developed and very characteristic. It is a delicately fine circular
network, having the immediate centre occupied with a number of small round holes, the
rest of the piece being characterized by large subhexagonal apertures, which diminish
in size as they approach the margin.

Respecting the affinities of C. calcigera, it may be said that although the form bears
undoubted resemblance to Cucumaria Hyndmanni, Thomson, this is in external appear-
ance only, as has been already pointed out by Diiben and Koren*; whilst Liitken f, in
his description of C. Korenii, specified in detail the differences which exist between the
two species. As a fact, the calcareous plates are altogether different — in C. Ilyndmanni
being alike and constant in form, whilst in C. Korenii (—C. calcigera) there are not
only two layers of quite distinct kinds, but the plates themselves vary very considerably.
The plates also of the sucker-feet are broader, perforated with more holes, and are less
regular and less uniform than in C. Hyndmanni ; and, lastly, the fully-developed disk at
the extremity of the sucker is characteristic of the present species.

* Kongl. Vetensk. Akad. Handl. ar 1844, p. 299.

t Videnskabelige Meddelelser f. d. Naturh. Forening i Kjobenhavn 1857 p. 6.

8 ON THE ECH1NODERMATA OF THE

With this evidence in view, it is difficult to account for the grounds on which
Semper*, after the examination of specimens, grouped the two forms as identical,
merging C. Jiorenii, Liitken, into C. Hyndmanni. It consequently follows that his adopted
association of P. calcigera, Stimpson, with C. Hyndmanni is equally erroneous.

Distribution.

a. Greenland: Godhavn, lat. 69° N. ('Valorous' Exped.); Fiskernsesset (Liitken);
Arksut, 15-25 fms., mud bottom (Barrett).

b. North of American Continent : Assistance Bay, about lat. 74° N., 7-10 fms.,
muddy bottom (Penny's Exped.), the most northern locality on record ; Labrador, 15 fms.,
sandy bottom (Packard, fide Verrill) ; Massachusetts ( Verrill).

Description of the Illustrations of this Species on Plate I.

Fig. 3. Cwntmaria calcigera : natural size.

4. Sketch of the mouth-ring and adjacent parts of a young individual :

magnified.

5. Generative tubes at an early stage of growth : magnified.

6. Spicules of the superficial layer in situ : magnified.

7. Small spicule in profile : magnified.

8. Disk at the extremity of the sucker-foot : magnified.

OECULA BAETHII, Troschel.

1846. Orcula Bartliii, Troschel, Wiegm. Archiv f. Naturgesch. Jahrg. xii. p. 63.

1857. Orcula Barthii, Liitken, Vid. Meddel. N. Forening i Kjobenhavn, 1857, p. 9.

1867. Orcula Barthii, Selenka, Zeitschr. f. wiss. Zool. Bd. xvii. p. 352.

1868. Orcula Barthii, Semper, Eeisen im Archipel der Philippinen, Holothurien, pp. 68, 274.

Dr. Liitken, in describing some Greenland examples of this Holothurian, states f
that they present a regular sausage form, with a length (in one which seemed to
have best preserved its natural proportions) of 3 inches and a thickness of over f inch.
The body-skin is brownish in colour, but almost concealed by the numerous bluish or
whitish sucker-feet with which the whole body is closely crowded. Neither in the
skin, which is thick and tough, nor in the feet is there any trace of solid calcareous
formations. A thick layer of transverse muscles lies under the outer skin ; and also
five strong longitudinal bands, each of which gives off, at a distance from the anterior
extremity equal to about one third the entire length of the animal, a short thick
muscular bundle, which is attached to one of the radial elements of the ocsophageal
ring, whilst the main bands themselves are continued up to the anterior extremity of
the body, and are then inclined backwards again, running as five thin narrow bands
along the outer wall of the inverted oesophagus until they reach the points of the

* Eeisen im Archipcl der Philippinen, Holothurien, p. 237.

t The present writers having unfortunately had no opportunity of examining this Holothuroid, have
accordingly availed themselves of giving the ahove translation of the observations made upon the form by
Dr. Chr. Liitken, the eminent and accurate Danish zoologist, in the memoir ahove quoted.

ARCTIC SEA TO THE WEST OF GREENLAND. 9

same pieces of the mouth-ring to which the muscular bundles above mentioned are
also attached. These thick transverse bundles are the musculi retractor es of the
oesophagus and tentacles, whilst the finer longitudinal bands outside the gullet function
as m. eoctensores. As these are much weaker than the former, it is natural to find that
the tentacles are, in all preserved examples, so thoroughly retracted that they can
only be seen by cutting open the oesophagus. The tentacles, 15 in number (ten large
and five very small, which alternate with them), are all finely branched as in Cucumaria.
Immediately below their point of origin is situated the deep cartilaginous mouth-ring,
which is composed of ten pieces merged together, five broad and five narrow, all forked
below, the five narrow pieces terminating upward with one point, and the broader with
twp ; and it is to these broader pieces that the above-mentioned muscular bands
(musculi retractores) are attached. The Polian vesicle is long and thin, with
moderately strong walls. The cloaca is clothed internally with a white sinewy skin ;
and from the outside pass fine muscular filaments by which it is attached to the body-
wall of the animal. From the cloaca proceed the long thin-walled intestine and the
respiratory apparatus, the latter presenting the usual character of thickly branched
tubes, the branches being fine and dilated at the extremities into vesicles. On the
side of the animal opposite to these organs, and, in the contracted condition, almost
equidistant from either end, lies the reproductive apparatus, having the appearance of
a bundle of fine, dilated, and here and there spool-shaped filaments.

Distribution.

a. Greenland: Holsteinborg, lat. 66° 56' N. ('Valorous' Exped. : this species or
new, fide Norman, loc. cit.), the most northern locality on record; Egedesminde, Ju-
lianshaab and Fiskernsesset (Lutken).

b. North of American Continent : Labrador (Troschel).

PSOLUS PHANTAPUS (Strussen/eldt), Jdger.

1765. Holoikuria phantapus, Strussenfeldt, Act. Holm. 1765, p. 265, Tat'. 10.

1766. Holothuria phantapus, Linne", Syst. Nat. ed. xii. p. 1089.

1776. Holothuria phantapus, 0. F. Miiller, Zool. Dan. Prodr.no. 2803.

1777. Ascidia rustica, Pennant, British Zoology, vol. iv. p. 48, pi. xxiii. fig. 35.

1789. Holothuria phantapus (Miiller), Abildgaard, Zool. Dan. vol. iii. p. 54, tab. cxii., oxiii.
1816. Holothuria phantapus, Lamarck, Anim. s. Vert. ed. 1, vol. iii. p. 73.
1828. Cuvieria phantapus, Fleming, Hist. British Animals, p. 483.

1833. Psolus phantapus, Jiiger, De Holothuriis, p. 21.

1834. Cuvieria phantapus, Blainville, Manuel d'Actinologie, p. 191, pi. 13. fig. 1 .

1835. Psolus phantapus, Brandt, Prodr. Descrip. anim. ah Mertensio obs., fasc. i. p. 47.

1836. Cuviera phantapus, Johnston, London's Mag. Nat. Hist. vol. ix. p. 472, fig. 68.
1841. Psolus phantapus, Forbes, Hist. British Starfishes, p. 203.

1844. Cuvieria phantapus, Diiben & Koren, K. Vet.-Akad. Handl. 1844, p. 313.
1851. Psolus Icevigatus, Ajres, Proceed. Boston Soc. Nat. Hist. vol. iv. pp. 25, 36.
1853. Psolus phantapus, Stimpson, Syn. Mar. Invert. Grand Manan, p. 16.
1857. Psolus phantapus, Liitken, Vid. Meddel. N. Forening i Kjobenhavn, 1857, p. 12.

C

10 ON THE ECHINODERMATA OF THE

1861. Psolus 2>hantapus, Sars, Oversigt af Norges Echinodermer, p. 112.

1867. Psolus phantapus, Seleuka, Zeitsch. f. wiss. Zool. Bd. xvii. p. 342.

1868. Psolus phantapus, Semper, Reisen im Archipel der Philippinen, Holothurien, pp. 62, 272.

The Rev. A. M. Norman, by whom the Echinodermata of the ' Valorous ' Expe-
dition were determined *, has kindly informed us that specimens of this Holothuroid
were taken at Holsteinborg and Godhavn. The writers, not having had the oppor-
tunity of examining any Arctic examples of this well-known form, content themselves
with simple reference to the principal literature and distribution of the species.

Distribution.

a. Greenland: Godhavn, lat. 69° 14' ('•Valorous' Exped.), the most northern
locality on record ; Holsteinborg Harbour (' Valorous ' Exped.) ; Fiskernsesset (Lutken) ;
Arksut (Barrett}.

b. North of American Continent : Grand Manan, at low water and in 40 fms.
(Verrill); Maine (Stimpson).

•c. North of European Continent : Iceland, Fseroe Islands, Finmark, Scandinavian
coasts, British Isles.

PSOLUS FABEICII (D. & K.), Lutken. Plate I, Figs. 9-13.

1780. Holoihuria squamata, Fabricius, Fauna Grcenlandica, p. 356. no. 348 (non 0. F. M., nee D. & K).

1841. Holoihuria squamata, Gould, Invert. Anim. Massachusetts, p. 345.

1844. Cuvieria Fabricii, Diiben & Koren, Kongl. Yet.-Akad. Handl. 1844, p. 316.

1851. Cuvieria Fabricii, Ayres, Proceed. Boston Soc. Nat. Hist. vol. iv. pp. 35-37.

1853. Cuvieria Fabricii, Stimpson, Syn. Mar. Invert. Grand Manan, p. 16.

1857. Psolus Fabricii, Lutken, Vid. Meddcl. N. Forening i Kjb'benhavn, 1857, p. 13.

1866. Lophothuria Fabricii, Verrill, Proceed. Boston Soc. Nat. Hist. vol. x. p. 354.

1867. Cuvieria Fabricii, Selenka, Zeitschr. f. wiss. Zool. Bd. xvii. p. 343.

1868. Psolus Fabricii, Semper, Eeisen im Archipel der Philippinen, Holothurien, pp. 62, 272.

1877. Psolus Fabricii, Marenzeller, Coelent. Echin. u. Wiirmer d. ost.-ung. Nordpol.-Exped. p. 32
(Denksch. d. k. Akad. d. Wissensch. Wien, Bd. xxxv.).

Body somewhat limpet- or chiton-shaped, arched above and flat beneath ; dorsal
and ventral surfaces separated by a perfectly sharp angle, the contour of the periphery
being oval. Both oral and anal apertures are situated upon the dorsal area, one at
either extremity, and rather nearer to the margin than midway from the centre in the
longitudinal axis of the body. The dorsal surface is covered with very large plate-like
scales, which imbricate upon one another, their free margins being round in contour
and directed centripetally ; those in the neighbourhood of the apertures are somewhat
smaller, and are grouped round the opening itself as the centre of their arrangement,
whilst a number of the scales that immediately surround the orifice become developed
into thick calcareous wart-like tubercles. Bordering the extreme outer edge of the
dorsal area are about two rows of very much smaller scales, not more than one tenth

* In the " Preliminary Report of the Biological Results of a Cruise in H.M.S. ' Valorous ' to Davis Strait
in 1875,'' Proceed. Roy. Soc. vol. xxv. pp. 202-215.

ARCTIC SEA TO THE WEST OF GREENLAND. 11

the size of the others. The scales are densely covered with very large granules, but
which in old specimens are not unfrequently abraded in places.

The integument of the ventral area, although thin, is tough and leathery, and
presents no asperities to the touch. The ambulacral feet are entirely confined to the
ventral area, and are limited to its extreme outer margin, round which the suckers
form a broad border about four or five rows deep, very closely crowded, and having
no perceptible order in their arrangement except in the outermost series, which are
smaller than the rest, and form a line parallel with the margin. At the anterior and
posterior ends of this oval marginal series may be found a few additional suckers
extending along the median line : these are the rudiments of the aborted middle
^ambulacral series ; and it is rare that more than four or five rows of four suckers each
are present.

The mouth-ring resembles that of P. squamatus, D. & K., the elements having
more or less the form of an inverted Y, and the five interradial pieces being smaller
than the radials.

The Polian vesicle is large and single, the generative tubes extremely numerous
and extensive, and the muscular system much more strongly developed than in
P. squamatus (Lu'tken).

Respecting the spicules which occur in the cuticle of this Holothuroid, those of
the ventral skin are indefinite in shape, and are more correctly described as calcareous
reticulations having large circular apertures than as plates punctured with holes.
Short prolongations frequently rise at right angles from the decussations or midway on
the intercalary portions ; and occasionally secondary branchings are formed from these.
Here and there small cup-shaped spicules are to be met with, probably undeveloped
stages of the larger spicules. The tubes of the sucker-feet are overlaid with spicules
somewhat similar in shape and character to those of the ventral cuticle; and the
extremity is provided with a large circular calcareous disk, likewise of reticulated
structure, with the holes very large, close together, and nearly equal in size.

Colour and Size. — Psohis Falridi is, when alive, of a dark red colour, which
in spirit becomes changed to brown. Ayres speaks of his American specimens as
being of a bright brick-red when alive (loc. cit. p. 36). One of the largest examples
measured 3^ inches in length and over 2 inches in breadth (Lutken).

Premature Form. — In a young state the dorsal surface is more depressed, and the
orifices of the alimentary canal are somewhat nearer the margin than in the adult
animal. The ambulacral feet are limited to a double row of suckers surrounding the
margin of the ventral area ; and no trace is visible of the median series. It is of
interest to note that at this stage the young of the present species are almost undis-
tinguishable from those of Ps. pliantapus of a corresponding period of growth, although
the latter is a form whose habit differs most strikingly from that of the present species
when adult examples are compared. The same remark holds good for the young stages
of Ps. squamatus as well, the superficial differences at an early phase being nothing
more than slight variations in the character of the spicules that indurate the ventral
cuticle.

c2

12 ON THE ECHINODERMATA OF THE

Variations. — The preceding remarks respecting the resemblances that occur in
the premature phases of three allied species (two of which at least are associated in
the same waters) are of themselves sufficient to suggest the existence of ancestral
affinities ; and, indeed, so near is the relationship in one case that, although the im-
practicability of uniting the whole group is clearly apparent, the propriety of merging
two of them, Ps. Falricii and Ps. squamatus, is, in our opinion, far from questionable ;
for that Ps. Fabricii is in reality the outcome of locational variation from the
squamatus type there would seem to be but little doubt. At the same time it is
interesting to note the wide area over which the forms are found distributed and their
divergence maintained — both Ps. phantapus (teste Sars) and Ps. Fabricii (teste Ayres)
being found on the E. coast of North America, and having associated with them two
allied forms, viz. Ps. l&vigatus, Ayr., and Ps. granulatus, Ayr., both of which may,
without much difficulty, be regarded as derivative forms when the whole group is
passed in review.

Distribution.

a. Greenland : Holsteinborg, lat. 66° 56' N. (' Valorous ' Exped.) ; Jakobshavn,
Julianshaab (Lutkeri).

b. North of American Continent : Newfoundland (Lutken) ; Grand Manan, 2-8
fms., adhering to rocks ( Verrill) ; Massachusetts Bay.

c. ? North of European Continent : Lat. 74° 48' N., long. 69° 26' E. (Weyprecht &
Payer's Exped.), the most northern locality on record.

Description of the Illustrations of this Species on Plate I.

Fig. 9. Dorsal aspect of the animal : natural size.

10. Ventral aspect of the same specimen: natural size.

11. Ventral aspect of a young specimen : natural size.

12. One of the small cup-like spicules of the ventral integument: magnified.

13. Spicules of the ventral integument : magnified.

CHIRODOTA L^VIS (Fabricius), Grube, Plate I, Figs. 14-19.

1780. Holoihuria Icevis, Fabricius, Fauna Grcenlandica, p. 353. no. 345.

1806. ? Holoihuria pellucida (Vahl), Ratbke, Zool. Dan. vol. iv. p. 17, tab. cxxxv. fig. 1.

1829. Chirodota discolor, Eschscholtz, Zool. Atlas, Heft ii. p. 13, tab. x. fig. 2.

1833. Pentacta Icevis, Jager, De Holothuriis, p. 13.

1833. Pentacta pellucida, id. ibid.

1834. Cucumaria Icevis, Blainville, Manuel d'Actinologie, p. 195.

1834. Cucumaria pellucida, id. ibid.

1835. Dactylota Icevis, Brandt, Prodr. Descrip. anim. ab Hertensio obs., fasc. i. p. 45.
1835. Dactylota pellucida, id. ibid.

1835. Chirodota discolor, Brandt, loc. cit. p. 59.

1837. Pentacta Icevis, Lamarck, Anim. e. Vert. ed. 3, vol. i. p. 553.

1837. Pentacta pellucida, id. ibid.

1851. ? Thyonidium pellucidum, Sars, Reise i Lofoten og Finmarken, p. 44.

1851. ? Synapta coreacea, Agassiz, Proc. Amer. Acad. vol. ii. p. 269.

1851. ? Synapta rotifera, Pourtales, Proc. Amer. Assoc. Adv. Sci. 1851, p. 15.

ARCTIC SEA TO THE WEST OF GREENLAND. 13

1851. Chiridota lewis, Grube, Middendorff's Sibirische Reiso, Bd. ii. p. 41.

1851. ? Chiridota discolor, Grube, Middendorff's Sibirische Reise, Bd. ii. p. 35.

1852. Trochinus pallidus, Ayres, Proc. Boston Soc. Nat. Hist. vol. iv. p. 243.

1853. Chirodota Icevis, Stimpson, Mar. Invert. Grand Manan, p. 17.

1857. Chiridota lave, Liitken, Vid. Mcddel. N. Forening i Kjobenhavn, 1857, p. 16.

1861. ? Chirodota pellucida, Sars, Oversigt af Norges Echinodermor, p. 124, pis. 14-16.

1862. Chirodota Icevis, Dujardin & Hupe, Hist. Nat. Zooph. ISchinodormes, p. 616.

1866. Chirodota Iceve, Verrill, Proceed. Boston Soc. Nat. Hist. vol. x. p. 354.

1867. Chirodota pellucida, Selenka, Zoitsch. f. wiss. Zool. Bd. xvii. p. 366.
1867. ? Chirodota tiyillum, Selenka, ibid.

1867. ? Chirodota typica, Selonka, ibid.

1868. Chirodota Icevis, Semper, Holothurien, Reise im Archip. d. Phil. pp. 23, 267.

1868. Chirodota pellucida, Semper, Holothurien, Eeise im Archip. d. Phil. pp. 23, 267, pi. v. fig. 2.

Body elongate and cylindrical, to which the absence of sucker-feet and of all
hooklets or asperities imparts a strikingly vermiform appearance. Specimens preserved
in spirit, however, are extremely contracted and generally contorted out of all natural
shape. The skin is smooth, whitish, and semitransparent in spirit preparations ; and
the longitudinal muscles form five dense white bands extending from end to end and
prominently marking out the interradial spaces. Within these areas are found a
number of milk-white spots or sac-like papillae, ranging up to a millimetre in diameter,
and arranged in three of the interradia (which may be regarded as the dorsal area) in a
fairly continuous line of about 20 to 30, whilst in the remaining two interradia, which
form the ventral surface, there are not more than from 3 to 12, and these confined
generally to the extremities, principally the anterior one. The white excrescences or
sacculi are not all of uniform size ; for a smaller one frequently alternates with a larger ;
and they contain a collection of the elegant wheel-shaped spicules which charac-
terize this genus ; and of these there may be as many as 90 or even more in the large
spots. The wheels measure O'l millim. in diameter, and are uniformly six-rayed; the
edge of the rim is bent over inwards and finely denticulated, a character which is only
perceptible under certain methods of illumination ; and this, as well as the manner in
which the " spokes " are attached to it, seems to have been frequently misunderstood,
and to have given rise in consequence to much diversity of opinion, since the differences
were considered to be of specific importance.

The tentacles are twelve in number, and somewhat hand-shaped ; and the digita-
tions, of which there are 10-12, are capable of being closed in upon the " palm." The
tentacles contain a few small spicules, which in the " fingers " are little more than
simple elongate bodies ; but nearer the base their extremities are enlarged and frequently
either cleft or crenulate.

The mouth-ring is very compact, the elements being in such firm adherence as to
convey the idea of a solid annul us ; they are subquadrate in form, having the lower
margin incurved, and on the upper a slight prominence with a small hollowing-out of
the margin on either side of it ; five of the pieces (alternately placed) are punctured for
the nerve. On the unpunctured plates there is an elevation or crest-like prominence,
somewhat in the form of an inverted Y.

14 ON THE ECHINODERMATA OF THE

The retractor muscles are thick and powerful, and attached along their length by
a muscular membrane to their corresponding longitudinal band. The whole muscular
system is very strongly developed, the longitudinal bands being remarkably thick and
robust, whilst the transverse series are numerous and closely placed.

The Polian vesicles are long and thin, 10-15 in number, and of unequal length, the
longer ones being twice (or even more) the length of the shorter.

The alimentary canal is nearly three times the length of the body, and is bent twice
upon itself. Passing from the mouth it runs four fifths the length of the body ; its
course is then reversed, and the canal proceeds as far as the anterior third of the body,
when it is again sharply bent backwards and passes to the anal extremity — the three
lengths formed by the convolutions being held in their places by mesenteries.

The mesenteries attached to the two lower portions are accompanied along the line
of their attachment to the body-wall by a longitudinal series of small pyriform bodies
of peculiar shape, attached to a common cord which passes nearly up to the oral extre-
mity— the " infundibular " organs or " Wimpertrichter." Their form is represented in
Fig. 17; and it will be noted that they accord very nearly with those given by Sars of
C. pellucida, and in like manner with those of a Greenland specimen of C. Icevis, which
are figured for comparison.

The generative organs consist of two series of long and extensive dichotomosing
tubes connected together by a slender branch.

Size. — The largest example of this species has been recorded by Sars, and measured
100 millims. in length ; generally, however, they range from 20 to 40 millims.

Premature Form. — In young stages the spots or sacculi which occur in the interradial
areas are smaller in number and proportionally larger in size than in the adult animal.
The calcareous wheels contained in the sacculi have both the spokes and the rim con-
siderably broader in the old than in the young form. Young individuals have also
fewer " fingers " on the tentacles, those of a small specimen 5 millims. in length having
only six digitations to each (Sars).

Variations. — Chirodota Icevis may unquestionably be regarded as a circumpolar
species; and the modifications which it presents are comparatively slight. We are
unable to consider the forms from Finmark and Lofoten, so carefully described by Sars
under the name of C. pellucida (Vahl), as other than the representatives of C. Icevis, the
Greenland type of Fabricius and Lvitken, the modifications which led Sars to place
them as distinct species being frequently found much less pronounced, in both forms
mutually, than in the specimens he examined. Amongst the specimens of this species
procured by the ' Valorous ' dredgings there are examples in which the form of the
wheels and of the tubular infundibular organs accord perhaps more nearly with Sars's
figures of C. pellucida than with those which he gives for comparison of the same
structures from the specimen of C. Icevis which he had dissected. Upon both of these
points he placed great importance.

C. discolor, of Grube, from Behring's Straits, although presenting greater divergence
than the above, seems indubitably to belong to the same type. The differences noted
in the form of the wheels and the structure of the infundibular organs, as well as the

ARCTIC SEA TO THE WEST OF GREENLAND. 15

greater length of the alimentary canal and the longer and more attenuated generative
sacs, are results which may not unnaturally be expected to arise when the distribution
of the forms compared is kept in mind.

Distribution.

a. Greenland : Godhavn, lat. 69° 14' N. (' Valorous ' Exped. & Hayes s Exped.),
the most northern locality on record.

b. North of American Continent: Labrador, 10 fms., sandy bottom (Packard, fide
Verrill) ; Grand Manan (Stimpson) ; Eastport, under stones at low water (Verrill).

? Sitcha (under the name of C. discolor, Eschscholtz).

c. North of European Continent'. Finmark and Lofoten, 1-2 fms., sand (Sars).

d. North of Asiatic Continent : Ochotsk Sea 1 (under the name of C. discolor,
Grube, Middendorff's Exped.).

Description of the Illustrations of this Species on Plate I.

Fig. 14. The animal, much contracted : natural size.

15. Generative tubes : magnified.

16. Portion of the mouth-ring : magnified.

17. The pyriform infundibular organs or " Wimpertrichter : " magnified.

18. One of the wheel-like spicules : magnified.

19. One of the sacculi, to show the aggregation of the spicules : magnified.

MYRIOTROCHUS RINKII, Steenstrup. Plate I, Figs. 20-24.

1851. Myriotrochus Binkii, Steenstrup, Vid. Meddel. N. Forening i Kjobonhavn, 1851, p. 55, pi. iii.

figs. 7-10.

1852. Chiridota brevis, Huxley, Sutherland's ' Journal of a Voyage,' Append, vol. ii. p. ccxi.
1857. Myriotrochus BinJcii, Liitken, Vid. Meddel. N. Forening i Kjobenhavn, 1857, p. 22.
1863. Myriotrochus Binkii, Stimpson, Proc. Acad. Nat. Sci. Philad. 1863, p. 142.

1867. Myriotrochus BinJcii, Selenka, Zeitsch. f. wiss. Zool. Bd. xvii. p. 367.

1868. Myriotrochus Binkii, Semper, Reisen im Archipel der Philippinen, Holothurien, p. 24.

1877. Myriotrochus Binkii (?pars), Theel, Nova Acta Keg. Soc. Sci. Upsal. ser. 3, 1877, no. xvii. p. 3.

Body cylindrical, and tapering only slightly towards the posterior extremity, in
some specimens the middle portion being somewhat swollen out, which causes the Holo-
thuroid to present a slightly arched profile. Anterior extremity very broad and
incapable of being retracted within the body, the habit of the animal being compara-
tively short and thick, the length not more than four or five times the thickness. The
skin is smooth and whitish grey, and in some examples semitransparent, in which case
the five longitudinal muscular bands as well as some of the internal organs are visible
through the membrane. The surface of the body is overstrewn with large calcareous
wheel-shaped spicules of characteristic form, attached to the integument by means of a
short peduncle or prolongation of the skin, which is affixed to the central portion or
" boss." This, however, is not easy to be seen in all the spicules, as some are closely
appressed to the body, in consequence no doubt of muscular contraction. The spicules,
as a rule, are more numerous upon the dorsal than upon the ventral surface ; in some

16 ON THE ECHINODERMATA OF THE

specimens, however, they seem to be wanting altogether on the ventral interradia, and
are confined to a few isolated spicules on the anterior and posterior extremities of the
dorsal area.

The spicules are wheel-shaped and slightly concave outwardly, varying from '2 to
•25 of a millim. in diameter. The spokes or radii, which are straight and slender, and
are united into a central boss, average about nineteen in number ; but they may vary
by three or four more or less.

The circumference or portion which represents the outer rim is divided by sutures
midway between each of the radial spokes ; and prolongations given off from the margin,
and bent sharply over at an acute angle, are directed inward towards the centre, and
form a continuous series of equal, regular, elongate denticles. These denticles are
about one third of a radius in length, generally somewhat three-sided in form ; and their
number frequently exceeds that of the spokes of the wheel.

The manner in which the spicules are developed is interesting, and is easily followed
in some examples. The central boss is the first commencement, the margin of which
becomes crenulated ; and then denticles develop, which are gradually prolonged in the
form of spokes until the full size of the wheel is attained ; processes then form at the
extremities at right angles to the spoke, and these eventually meet and coalesce with
the corresponding processes from the neighbouring rays or spokes, whereby a con-
tinuous rim is formed equal in thickness to the rays ; this rim next increases some-
what in breadth, and the inwardly directed denticles begin to develop from the margin.

The tentacles are twelve in number, and, judging from their structure, are somewhat
hand-shaped, and would resemble when extended the form which is found in Chirodota
(cf. Steenstrup, loc. cit.) ; when retracted, however, the upper portion is drawn within
the basal part or stem of the tentacle itself, the membrane forming a kind of hood
which arches over from the inward side, but leaves the extremities of the digitate pro-
longations exposed and visible on the outer side.

Within the broad circlet of the tentacles stretches a muscular membrane, in the
centre of which the mouth-aperture is situated. The alimentary canal passes directly
from this to the posterior portion of the animal, where it is bent upon itself, then two
or three times sharply convoluted, and prolonged as far as the anterior third of the
body, when its course is again reversed, and it finally passes direct to the anal
aperture, this lower or intestinal portion being several times narrower than the rest of
the canal (see fig. 22). The various convolutions are held in position by fine reticulated
mesenteries.

The calcareous mouth-ring is very large and strong, and composed of ten ossicles ;
eight of these bear on their upperside a long, three-sided, wedge-shaped prolongation,
whilst the remaining two, which are situated on either side of the median dorsal ossicle
of the ring, have each two prolongations. Thus the prolongations correspond in number
with the tentacles, the bases of the latter being situated between them, whilst a fine
muscular band, which would seem to function as their retractor, stretches between the
prolongations halfway between the base and tip. A foramen for the passage of the
radial nerve is punctured through the prolongation, somewhat above its base.

ARCTIC SEA TO THE WEST OF GREENLAND. 17

The water-vascular ring is broad and conspicuous ; and although only one Polian
vesicle is developed, which is large and heart-shaped, a number of dilatations occur at
intervals on the ring, which are suggestive of incipient or aborted vesicles (cf. fig. 24).

The generative organs consist of two groups of dichotomosing tubes which unite in
a single duct, that makes its exit outside the tentacular ring. The sexes are distinct
and readily distinguishable : in the female the tubes are shorter but very much broader,
and the large-sized ova can generally be seen within ; in the male, on the other hand,
the tubes are longer and thinner as well as more numerous. In neither sex do these
tubes, as a rule, extend beyond the anterior third of the body.

Premature Form. — We are unable to speak from personal observation of the
growth-phases of Myriotrochus, as all the specimens at our disposal have been more or
less perfectly developed. Dr. Hj. Theel* mentions an individual 10 millims. in length,
presumably referable to this species, as the smallest he had examined. He states that
" the skin contained a great quantity of the wheels in different stages of development,
some even perfectly formed, although amongst the whole it was scarcely possible to
find two that were identical. Many of the spicules had not more than six or ten radii,
but bore twice that number, or even more, of the denticles. This example, which was
a female, contained, notwithstanding its small size, fully developed eggs " (loc. tit.).

Variations. — On the whole, Myriotrochus Rinkii is a species that seems to
undergo only very slight variation in form; the main difference which we have
remarked occurs in the number of spicules that are present on the cuticle. For
instance, in some examples the whole of the three dorsal interradia are closely filled
with spicules, and a fair sprinkling is distributed over the ventral area as well ; whilst,
on the other hand, specimens are not rare in which these are confined to the median
dorsal interradial area, with, perhaps, just a few on the edge of the adjoining areas,
the ventral portion being quite bare, except three or four isolated wheels at the anterior
extremity ; and there are yet others even more sparingly provided, which have only a
few isolated spicules at the extremities of the dorsal surface. Respecting the above-
named variable character, we are bound to confess that we are at present unable to
assign even a locational value to these modifications, since both extremes occur together
in the material received from the most northern collecting-station recorded.

With these facts before us, it would seem that the characters upon which Sars
founded his genus Oligotrochus are all to be found in Myriotrochits ; and consequently
no grounds exist on which that genus can be longer maintained. Indeed we have for
some time been of opinion that even the species 0. vitreus, Sars, could only be regarded
as a variety comprehended within the form of Myriotrochus Rinkii; but a recent paper
of Dr. Theel's, and a careful comparison of Sars's description and figures with the
material we have had before us, lead to the conclusion that Oligotrochus vitreus may
probably rank as a species of Myriotrochus allied to M. EinMi ; and we are also further
inclined to think, from the description given, that Dr. Theel's specimens from Novaya
Zemlya and the sea of Kara would accord more nearly with Myriotrochus vitreus than

» Nova Acta Keg. Soc. Sc. Upsal. ser. iii. 1877, no. xvii. p. 6.

18 ON THE ECHINODERMATA OF THE

with the Greenland form of Steenstrup's species. In one characteristic especially do
they correspond with, and by the same differ from, the ordinary arctic forms of M. EinJcii.
In this latter the spicules are attached to the skin by a kind of stalk or prolongation
of the cuticular membrane, whilst in 0. vitreus and Dr. Theel's specimens of Myrio-
trochus it is particularly mentioned that these are embedded in the skin itself. So
marked is the feature, that amongst all his material Dr. Theel has never seen an indi-
cation of the " stalk ;" and he is hence led to question the accuracy of Steenstrup's
description in this respect, and to take, at the same time (quite erroneously, it would
seem to us) certain words used by Prof. Huxley when describing Chirodota brevis as
supporting the same conclusion (Theel, I. <?. p. 4 ; Huxley, I. c. p. ccxii). Now it is
very remarkable that, on the other hand, every specimen which we have examined,
whether from Greenland or the extreme north, has the spicules attached by membranous
stalks, and that this is a character which is at once apparent on examining them with
no greater magnifying-power than a simple hand-glass. It is, in all probability, true
that the spicules could be drawn more or less tightly down to the surface, as Steenstrup
has already suggested, from the nature of their attachment (Steenstrup, I. c. p. 58) ;
and in some rare cases, where the spicules have been abraded, there seem to be little
cavities left in their place in the membrane ; but we are disposed to believe that this
results rather from the change which has taken place in the tissue around the closely
appressed spicule whilst in spirit, than from a natural depression existing during life.

Distribution.

a. Northward of Smith Sound : Discovery Bay, lat. 81° 41' N. (Naress Exped.),
the most northern locality on record ; Port Foulke, lat. 78^° N. (Stimpson) ; lat. 69° 31',
100 fms. (' Valorous' Exped.).

In Greenland : Godhavn ; Omenak ; Arksut, 10 fms., mud (Llitken) ; Germania
Harbour in E. Greenland, 2 fms. (Moelius).

b. North of American Continent : Assistance Bay (Penny's Exped.) ; Labrador
(Packard, fide Verrill).

c. North of European Continent: Spitzbergen (Lutken); Novaya Zemlya,
2-70 fms., clay ; Vaigatch ; Kara Sea, 11-60 fms., sandy clay ; Sea of Murrnan (Theel).

Description of the Illustrations of this Species on Plate I.

Fig. 20. The animal : natural size.

21. One of the spicules : magnified.

22. Sketch of the internal organs, showing the relative proportions of the alimen-

tary canal, the Polian vesicle, and the generative tubes.

23. Portion of the mouth-ring : magnified.

24. Underside of mouth-ring, with parts attached, seen from below, showing the

dilatations of the oral water-vessel.

ARCTIC SEA TO THE WEST OF GREENLAND, 19

ECHINOIDEA.

ONE solitary species of Echinus is the sole representative of this large class in the
Smith-Sound area ; no other Echinoid passes in this direction from the North Atlantic
up Davis Straits beyond the Arctic Circle. The form in question, Strongylocentrotus
drobachiensis, would, indeed, seem to hold undisputed sway in the boreal regions, being
essentially and unequivocally circumpolar in its distribution, as will be seen from the
list of localities appended below. Considerable variation is manifest, as might naturally
be expected in so wide a range ; but the stages are so gradual and the intermediate
forms so constantly maintained, that no characters of specific value are recognizable
amongst them ; and we are, in this instance, presented with an interesting example of
a species modified under the influence of various conditions of existence (habitat) alone,
without the cooperation or hostile action of any closely allied competitive forms.

STRONGYLOCENTKOTUS DROBACHIENSIS (Miiller), A. Agassiz, Plate II. Figs. 1-3.

1776. Echinus drobachiensis, 0. F. Miiller, Zool. Dan. Prodr. p. 235. no. 2846.

1780. Echinus saxatilis, Fabricius (non Miiller), Fauna Groenlandica, p. 372. no. 368.

1788. Echinus drobachiensis, Gmelin, Syst. Nat. linn. ed. xiii. p. 3169.

1816. Echinus neglectus, Lamarck, Anim. s. Vort. ed. i. vol. iii. p. 49.

1826. Echinus granularis, Say (non Lamarck), Proc. Acad. Nat. Sci. Philad. 1826, p. 225.

1828. Echinus subangularis, Fleming (non Leske), Hist. British Animals, p. 479.

1835. Strongylocentrotus chlorocentrotus, Brandt, Prod. Descript. anim. ab Mertensio obs., faso. i. p. 264,

1836. Echinus neglectus, Agassiz, Mem. Soc. So. Nat. Neufchatel, i. p. 190.

1837. Echinus chlorocentrotus, Des Moulins, Etudes sur les Eehinides, p. 282,

1840. Echinus granulatus, Gould, Invert, of Massachusetts, p. 344.

1841. Echinus neglectus, Forbes, Hist. British Starfishes, p. 172.

1844. Echinus neglectus, Diiben & Keren, K. Vet.-Akad. Handl. 1844, p. 277.
1846. Toxopneustes drobachiensis, Agassiz, Cat. rais., Ann. Sci. Nat. vi. p. 367.
1846. Toxopneustes neglectus, Agassiz, Cat. rais., Ann. Sci. Nat. vi. p. 367.
1846. Toxopneustes granulates, Agassiz, Cat. rais., Ann. Sci. Nat. vi. p. 368,
1846. Toxopneustes Dubenii, Agassiz, Cat. rais., Ann. Sci. Nat. vi. p. 368.
1848. Echinometra drobachiensis, Gray, British Eadiata of the Brit. Museum, p. 4.

1851. Echinus neglectus, Brandt, Middendorff's Sibirische Eeise, Bd, ii. p. 34.

1852. Echinus neglectus, Forbes, Sutherland's Journal of a Voyage &c. vol. ii. Append, p. ccxiv.

1853. Echinus granulatus, Stimpson, Syn. Mar. Invert. Grand Manan, p. 15.

1857. Toxopneustes drobachiensis, Liitken, Vid. Meddel. N. Forening i Kjobenhavn, 1857, p 24.
1857. Toxopneustes drobachiensis, Sars, Bidr. Middelhavete Littoral-Fauna, p. 115.

1857. Echinus chloroticus, Stimpson, Journ. Bost. Soc. Nat. Hist. vi. p. 86.

1858. Toxopneustes neglectus, Desor, Syn. des Echin. foss. p. 135.

1863. Toxopneustes drobachiensis, A. Agassiz, Proceed. Boston Soc. Nat. Hist. 1863, p. 191.

1863. Toxopneustes carnosus, Barnard, in Agassiz, Proc. A. N. S. Philad. 1863, p. 357.

1863. Toxopneustes chlorocentrotus, Liitken, Vid. Meddel. N. Forening i Kjobenhavn, 1863, p. 144.

1863. Toxopneustes granulatus, Liitken, Vid. Meddel. N. Forening i Kjobenhavn, 1863, pp. 144, 148.

1866. Eury echinus drobachiensis, Verrill, Proceed. Boston Soc. Nat. Hist. vol. x. pp. 341, 352.

D2

20 ON THE ECHINODERMATA OF THE

1866. Euryethiniis granulatus, Terrill, Proceed. Boston Soc. Nat. Hist. vol. x. pp. 340, 352.

1868. Toxopneustes pictus, Norman, Report Brit. Assoc. Adv. Sci. 1868, p. 314.

1871. Toxopneustes pallidas, G. 0. Sara, Vidensk. Selsk. Forhandl., Nye Echinodermer, p. 25.

1871. Toxopneustes drobachiensis, Loven, Ofvers. K. Vet.-Akad. Forhandl. 1871, no. 8.

1872. Strongylocentrotus drobachiensis, A. Agassiz, Revision of the Echini, pp. 162, 277.

1877. Strongylocentrotus dr'obachiensis, Marenzellor, Ccelent. Echin. u. Wiirmer d. ost.-ung. Nordpol-
Exped. pp. 3, 29 (Denksch. d. k. Akad. d. Wissensch. Wien, Bd. xxxv.).

Test depressed, the height being proportional to the diameter as 1 : 2 approxi-
mately. Ambital outline almost circular, and only slightly subpentagonal ; contour of
the profile somewhat conoid, rounded at the apex. Test well rounded beneath, and
incurved towards the interior at the margin of the actinostome. Actinostome large,
and occupying three eighths or more of the diameter. Branchial notches very slight.
Buccal membrane thin, and with comparatively few small, granulated, oblong scales
and ten large buccal plates. Coronal plates comparatively high, each bearing one
large primary tubercle, imperforate and uncrenulate. The series of these tubercles form
two prominent vertical lines, which extend from the apex to the actinostome, both in
the ambulacral and in the interambulacral areas. In the interambulacral areas the
plates above the ambitus bear a number of miliary tubercles, which are disposed in an
irregular manner around the primary tubercle, and form not unfrequently in small spe-
cimens a more or less clearly defined scrobicular ring ; below the ambitus a secondary
tubercle is present on each side of the primary, and, in fair-sized specimens of the
Echinus, the external ones (which stand nearest to the poriferous zone) may be traced
upon the abactinal surface of the test, extending in very rapidly diminishing series for
some distance above the ambitus. In the ambulacral areas there is a small secondary
tubercle on each plate on the inner side of the primary tubercle, the series diminishing in
size as they approach the apex, in the neighbourhood of which the secondary tubercles
become altogether undistinguishable from the few other miliaries which are present
on the plate. Near the ambitus the miliaries are fairly numerous and irregular in size,
the most conspicuous amongst them frequently forming, on the pore side of the plate,
an arched row of three or four, which stands between the arc of pores and the tubercle,
the adoral miliaries being largest. The pores are arranged in arcs of 5-6, those above
the ambitus diverging only slightly from the vertical. The apical system is large,
and also the anal ring, into which two ocular plates enter. The madreporiform and its
companion anterior genital plate considerably exceed the other genital plates in size,
which are often not much larger than the two entering ocular plates. The genital
foramina are large. The spines are comparatively fine and delicate, and vary consider-
ably in length, which ranges from 8 to 13 millims.

Colour. — The colour of the test is a varying shade of purplish brown, that of the
spines greenish grey, the test being sometimes green in tint also.

Size. — The northern specimens are comparatively smaller than individuals from a
more southern habitat. The largest example obtained during Sir George Nares's expe-
dition was taken at Cape Napoleon, and measures 43 millims. in diameter, 21 millims.
in height, and has 20 primary interambulacral tubercles.

Diameter.

Height. Anal system.

Actinostome.

A

. . 39

20

6

15-5

B

42

19

5

15-1

ARCTIC SEA TO THE WEST OF GREENLAND. 21

The following are the measurements in millims. of

A, a specimen from Discovery Bay ;

B, a specimen from the Drobach Fjord.

Interamb. tubercles.

17
15

Variations. — Owing to the extensive range of this boreal Echinoid, the variations
to which it is subject are so great that there are perhaps few other species which
include in their list of synonyma so large a number of specific determinations. This
arises from the fact that distant observers, depending too confidently upon the stability
of " local forms," have founded upon these characters numerous so-called new species,
all of which, however, have hitherto proved untenable when comparison has come to
be made with a large series of specimens.

Thus a range of examples of the northern varieties, known as S. granulatus (Say),
Gould, and S. chlorocentrotus, Brandt, fail to present any characters of sufficient
importance to warrant their separation from the drobachiensis group, although, when
isolated and extreme examples are compared, the differences at first sight appear very
marked.

When compared with what may be called a typical specimen, the test of the
Arctic forms of S. drobachiensis is relatively high, with the contour slightly more arched,
and having a larger actinostome and apical disk. The greatest superficial difference,
however, is presented by the spinulation. In some forms the spines of the abactinal
surface are so small and delicate, and so widely spaced, that the Echinus has almost a
naked appearance when placed beside others which are more uniformly clothed ; and it
will hardly be necessary to add that a corresponding modification exists in the size and
uniformity of the tubercles upon which the spines are borne. The more densely-spined
and more uniformly tuberculated urchin represents well the form known as granulating ;
and a glance at two such extreme individuals as those just indicated will easily account
for much of the confusion which has arisen in this species.

Both of these forms (the uniformly and the sparsely spined) occur together at
several stations in Smith Sound, and extend even to Discovery Bay. At present we
are unable to offer any definite explanation of the difference : possibly the variation
may be due to stunted growth ; or perhaps it may indicate a sexual character.

Distribution.

a. Northward up Smith Sound : Discovery Bay, lat. 81° 41' N., 15-20 fms., muddy
bottom (Nares's Exped.}, the most northern locality on record ; Richardson Bay, 70 fms. ;
Franklin-Pierce Bay, 13-15 fms., stony, bottom temperature 29°' 5 Fahr. ; Cape
Napoleon ; Hayes Point, 35 fms., bottom temperature 29°'5 Fahr. Greenland.

b. North of American Continent : Assistance Bay (Penny's Exped.). All the east

22 ON THE ECHINODERMATA OF THE

coast of N. America from Labrador to New Jersey, and probably even as far south as
Cape Florida ( Wiirdemann, fide A. Agassiz) ; west coast of N. America to Vancouver.

c. North of European Continent : Spitzbergen, Novaya Zemlya, Iceland, Faeroe
Islands, Scandinavian coasts, Britain.

d. North of Asiatic Continent : Behring's Straits, Kamtschatka, Ochotsk Sea
(Brandt}.

Description of the Illustrations of this Species on Plate II.

Fig. 1. An example of the thinly-spined form. Abactinal aspect : natural size.

2. The same specimen seen in profile : natural size.

3. Portion of the test near the ambitus, showing the ambulacral and inter-

ambulacral areas : magnified.

ARCTIC SEA TO THE WEST OF GREENLAND.

ASTEROIDEA.

ALL the Asteroids of the Smith-Sound fauna belong to well-known forms, with the
single exception of Pedicellaster palceocrystallus from Discovery Bay and Cape Frazer.
Several of the species are here recorded from much higher latitudes than they have
previously been known to inhabit ; and one of them, Lophaster furcifer, is found for
the first time within the entreme boreal regions.

The following is a list of the Asteroids herein described from the Smith-Sound
area, including all which have been found in the confluent portion of the N. Atlantic,
north of the Arctic circle (66° 30') : —

1 . Asteracanthion polare, Miiller & Troschel.

2. Asteracanthion grcenlandicum (Steenstp.), Liitken.

3. Stichaster albuhis (Stimps.), Verrill.

4. Cribrella oculata (Linck), Forbes.

5. Pedicellaster palceocrystallus, Sladen.

6. Crossaster papposus (Linck), M. & T.

7. Solaster endeca (Gmel.), Forbes.

8. Lophaster furcifer (D. & K.), Verrill.

9. Pteraster militaris (O. F. M.), M. & T.
10. Ctenodiscus corniculatus (Linck), Perrier.

ASTERACANTHION POLARE, Miiller & Troschel. Plate II. Figs. 4-8.

1780. Asterias rubens, Fabricius (non Linnd), Fauna Gromlandica, p. 367, no. 362.

1780. Asterias minuta, Fabricius, Fauna Groenlandica, p. 370. no. 365 (juv.).

1821. ? Asterias violacea, Sabine (non Miiller), Append. Parry's Voyage,_1819-20 in ' Hecla ' and ' Griper,'

p. ccxxiii.

1842. Asteracanthion polaris, Miiller & Troschel, System der Asteriden, p. 16.
1851. ? Asteracanthium ochotense, Brandt, Middendorffs Keise in den aussersten Norden und Osten

Sibericus, Bd. ii. Th. 1, p. 28.

1857. Asteracanthion polaris, Liitkon, Vid. Meddel. N. Forening i Kjobenhavn, 1857, p. 28.
1862. Asteracanthion rubens (pars), Dujardin & Hupe, Hist. Nat. des Zooph. £chinodermes, p. 331.
1866. Asterias polaris, Verrill, Proceed. Boston Soc. Nat. Hist. vol. x. p. 356.
1869. Asteracanthion polaris ?, Perrier, Recherches sur les Pedicellaires et les Ambulacres des Asteries et

des Oursins, p. 33.
1875. Aslerias polaris, Perrier, Revision des Stellerides, p. 58.

A Starfish of large robust habit, readily distinguishable from the other species of
Asteracanthion which occur in the Arctic regions. Rays generally six in number,
rounded, swollen at the sides, and tapering only at the extremity. Disk-radius to arm-
radius as 1 : 4'4-4'75 approximately. Each adambulacral plate bears two ambulacral
spines, which radiate apart from one another. In some specimens every pair throughout

24 ON THE ECHINODEEMATA OF THE

the ray stands at the same angle to the plate, and thus two regular rows are formed on
either side of the furrow. In others, however, each alternate pair of spinelets radiate
at a rather smaller angle to the floor of the ray, the internal spine being directed
further over the furrow, and the external one standing opposed to the interspace between
the pairs of spinelets of the neighbouring plates ; by this means an arrangement of
four rows of spinelets on each side of the furrow is brought about, which, together
with the swollen character of the sides, gives an appearance of considerable breadth to
the ventral portion of the rays. The ambulacral spines are long, cylindrical, and
rounded at the tip, not tapering or pointed in the slightest ; and the innermost spine is
somewhat thinner than the others. Pedicellarise forficiformes are very numerous
amongst the rows of ambulacral spinelets, and are densely crowded at the base of the
rays in large specimens. The lateral spines, which are long, cylindrical, and somewhat
crumpled or nipped together at the tip, form three longitudinal rows extending along
the ray. In addition to these there is a short secondary row situate between the most
ventral of the lateral series and the ambulacral spines ; but, in consequence of the
rapidity with which the spinelets diminish in size as they proceed outwards, this series
extends only along the inner half of the ray.

In some large specimens there is often a short row of 6-8 supplementary spines at
the base of the ray between the ventral and middle series of lateral spines ; or a
tendency to reduplication may be shown even in each of the rows at the extreme base
of the ray. Occasionally the lateral spines are more or less tapering, those of the
dorsal and median rows being also slightly curved upward.

The calcareous network of the abactinal surface is very robust, the membranous
meshes or interspaces being in proportion comparatively small. The papula3 are, as a
rule, single in young specimens; but in those of larger growth two or three occur
together. The dorsal spinelets are large and cylindrical, not tapering, but nipped
together at the tip, or even sometimes subclavate as in A. violaceum. They do not
assume any lineal or regular arrangement whatever upon the abactinal surface. In
some examples the dorsal spinelets are closely surrounded by a number of rather
smaller subclavate spinelets, which give the appearance of compact groups of spines
crowded together upon the abactinal surface of disk and rays alike.

Pedicellarise forcipiformes, although very numerous upon the abactinal membrane,
cannot be strictly said to form regular wreaths at the base of the dorsal spines ; for
whilst in some places they are much crowded, in others only three or four are present.
In the lateral series, however, the spines of the upper or dorsal row are surrounded at
their bases by a true wreath of pedicellaria?, whilst in the ventral or lower row the
pedicellariae are principally grouped on the dorsal or upper side of the base.

The pedicellarise forficiformes, which are moderately numerous upon the abactinal
surface, averaging perhaps one or two to a mesh, are densely crowded amongst the rows
of ambulacral spines, especially on the innermost portion of the furrow. These pedi-
cellariaB (p. forficiformes) are in the present species comparatively small, only a little
longer than broad, and dumpy in form ; the p. forcipiformes, on the other hand, are
large, very gibbous, and truncate at the apex of the jaws.

ARCTIC SEA TO THE WEST OF GREENLAND. 25

The madreporiform body is situated in the interradius, about midway between the
centre and the margin of the disk, but generally nearer the centre in fully developed
specimens.

Size. — Large examples measure 230 millims. in diameter ; and the species is known
to attain even greater dimensions.

Colour. — Dr. Liitken, recording the observations made by Inspector Olrik upon
the colour of this Starfish when alive, states that it is generally purple-red, and some-
times greyish olive-green.

Habitat. — According to the above-mentioned Danish naturalist, A. polare is to be
found at low tides in May and June, but later in the year in deeper water (for instance,
in August at 40 fathoms). This species was taken at Arksut by Mr. Barrett, from the
shore down to 10 fathoms depth, on rocky bottom.

Premature Form. — A young specimen 28 millims. in diameter (the relative
proportions of the disk and radii being 1 : 3'5 approximately) presents the following
characters. The rays are broad and stout, tapering near the tip, and have their
junction with the disk very frequently marked off by a sharp constriction in examples
of this early age. The ossicles of the abactinal network are remarkably stout and
robust, the interspaces being small and restricted in consequence ; these are usually
round in outline, with the membrane punctured by a single papula only, although two
may occur occasionally. The dorsal spinelets are large and truncate, many quite sub-
clavate, and are widely spaced over disk and rays alike without any regularity in their
arrangement; their bases are closely surrounded by pedicellariae forcipiformes, the
large size of which is very striking in this young stage of the starfish, whilst similar
pedicellarise also accompany the secondary or intermediate spinelets, which occur with
moderate frequency upon the imbricating ossicles. In the small specimen under notice,
no pedicellariae forficiformes have been detected upon the abactinal surface either of
the disk or rays ; there are, however, a few isolated ones on the margins of the furrows,
but comparatively insignificant in size, and scarcely, if at all, larger than the p. forci-
piformes of the abactinal surface.

The madreporiform body, which is large and deeply fissured, lies nearer the margin
than in the adult forms, being situated about two thirds of the distance from the centre
of the disk to the arm-angle.

Two longitudinal rows of lateral spines are present, one on the dorsal and the other
on the ventral margin, the sides of the ray being vertical. The spinelets of both series
are large, stout, cylindrical, and rounded at the tips ; they are isolated and well spaced,
the members of the upper series standing immediately over their ventral companions.
In dried specimens the dorso-lateral row forms a conspicuous fringe along the margins
of the ray, and the bases of the spinelets are surrounded by a wreath of pedicellariae
forcipiformes. These are also present at the bases of the ventro-lateral series, though
not so numerous, and are there confined principally to the upper side of the spinelet.
Between the ventro-lateral row and the ambulacral spines a short series of secondary
spinelets extends along the inner half of the ray only ; these are smaller, even at the
base of the ray, than the other spinelets, and diminish rapidly in size as they proceed

26 ON THE ECHINODEKMATA OF THE

outwards. The ambulacral spines are stout and cylindrical, two to each ambulacral
plate ; and every pair radiating at the same angle, a regular double row is formed on
both sides of the furrow.

Variations. — The number of rays is very constant at six, although examples having
only five are occasionally met with.

A considerable amount of variability occurs in the spinulation of the dorsal surface.
In some specimens the spinelets are densely grouped, in others well spaced and with
the encircling pedicellarise more conspicuously displayed. The spines themselves are
either cylindrical, with tips sharply pinched together, or fairly clavate and resembling
in character the spinelets of A. violaceum, O. F. Miiller. It would seem that these
differences exist irrespective altogether of age or growth.

Similarly variable in character appears the arrangement of the ambulacral spinelets,
as indicated by the deviation from the regular double row on either side the furrow to
the fourfold series produced by the greater divergence of the alternating pairs. In fact,
we are disposed to believe that the two modes of arrangement may be present on a single
individual, and that it was probably such a specimen that M. Perrier referred with doubt
to the species under notice, giving at the same time a most admirable and lucid
description, which leaves no doubt as to the correctness of his determination*.

It may be inferred, from the description given in the ' Fauna Grcenlandica,' that
A. polare and possibly also A. grcenlandicum were merged together by Fabricius under
the designation of Asterias rubens ; and it also seems highly probable that the Starfish
there named A. minuta is nothing more than a young stage of the present species — an
opinion which Dr. Lu'tken has already expressed f.

The description given by Brandt of A. ochotense, in MiddendorfFs ' Reise '$, indicates
a range of variation which leads to the assumption that more than one form may be
there included. If such a view be correct, a part of the description there given would
answer well for young forms of A. polare.

Miiller and Troschel's type specimens of this species were also unquestionably
young examples.

Distribution.

a. Greenland : W. coast of Davis Strait, about lat. 70° N. (Sabine) ; Upper Torske
Bank, lat. 67° 50' N., long. 55° 27' W. ('Valorous' Exped.) ; lat. 66° 59' N., long.
55° 27' W., 57 fms. (' Valorous ' Exped.) ; W. coast of Greenland, lat. 65° N. (Nares's
Exped.) ; Arksut, lat. 61° 10' N., long 48° 15' W. (Barrett).

b. North of American Continent : Labrador ( Verrill) ; St. George's Bank,
35-40 fms.

d. North of Asiatic Continent : Ochotsk Bay 1 (under the name of A. ochotense,
Brandt).

* Perrier, Recherches sur les Pedicellaires et les Ambulacres des Asteries et des Oursins, 1869, p. 33.

t Vidensk. Meddel. Naturh. Forening i Kjobenhavn for 1857, p. 29.

J Middendorffs ' Eeise in den aussersten Nordon und Osten Siberiens,' Bd. ii. Th. 1, p. 28.

ARCTIC SEA TO THE WEST OF GREENLAND. 27

Description of 'the Illustrations of this Species on Plate II.

Fig. 4. Abactiual aspect of the animal : reduced one half.

5. Actinal aspect of the same specimen : reduced one half.

6. Portion near the middle of a ray, actinal aspect: magnified.

7. Portion near the middle of a ray, abactinal aspect : magnified.

8. A similar portion from the more compactly spinulated variety of this

species : magnified.

GKffiNLANDicuM (Steenstrup), iMken. Plate II, Figs. 9-12.

1852. ? Uraster violacea, Forbes (non Miiller), Sutherland's ' Journal of a Voyage,' vol. ii. Append, p. ccxiv.

1853. Asterias Miilleri, Stimpson, Syn. Mar. Invert. Grand Manan, p. 14.

1854. Asteracanthion Miilleri, Bars ?, var., Steenstrup, Vid. Meddel. N. Forening i Kjobenhavn, 1854, p. 240.
1857. Asteracanthion grcenlandicus, Liitken, Vid. Meddel. N. Forening i Kjobenhavn, 1857, p. 29.
1863. Asterias grcenlandica, Stimpson, Proc. Acad. N. Sci. Philad. 1863, p. 142.

1866. Asterias grcenlandica, Verrill, Proc. Boston Soc. Nat. Hist. vol. x. p. 357.
1866. Asterias grcenlandicus, Gray, Synop. Spec. Starfish Brit. Mue. p. 2.

This species maintains a small habit, with five moderately thick arms, the propor-
tion of disk-radius to arm-radius being 1 : 4>5 or 5. The ambulacral spines are rather
long and cylindrical, and stand, in very irregular alternation, two and one to each adam-
bulacral plate, the single spinelets having a position vertical to the floor of the furrow,
whilst the pairs, on the other hand, radiate apart from one another in opposite direc^
tions. Except in young individuals, or near the extremity of the ray, the pairs are more
numerous than the isolated spinelets, and are generally borne by two or three plates in
succession, some rays even being regular for nearly half their length. The lateral spines,
which occupy the sides of the ray, form, according to age, either two or three longitu-
dinal series of isolated spinelets, not quite so long as the ambulacral series, and tapering
slightly at their tips — the middle series, when present, being smaller than the others,
and placed midway upon the lateral imbricating pieces. At the base of each of these
lateral spines are a number of pedicellaria? forcipiformes, sometimes grouped and some-
times forming a more or less perfect encircling wreath, the latter generally upon the
lower series.

The ossicles of the abactinal network that lie in the longitudinal direction of the
ray are much shorter than those that occupy a transverse position ; and being of delicate
habit, the membranous interspaces are consequently large, as well as much broader than
long. The dorsal spinelets are finer and shorter than those of the ventro-lateral series,
and are placed in groups of two or three together at the intersections, whilst isolated
spinelets occupy the intermediate intercalary ossicles, by which means, in consequence
of the above-noted proportions of the elements of the calcareous framework, the cha-
racter (although somewhat an irregular one) of a transverse arrangement of spinelets
across the ray is produced. Upon the disk the spinelets are more closely placed ; and
this, in specimens preserved in spirit, gives quite a distinct appearance to that portion of

28 ON THE ECHINODERMATA OF THE

the Starfish ; and, further, in some examples the disk-spinelets are also rather longer
than those which are found upon the rays.

The papulae are single. The pedicellariae are not crowded, and generally either
isolated or only two or three together. Pedicellarise forcipiformes are as a rule not
very numerous upon the dorsal surface of the rays, except towards the extremities, where
they occur somewhat more abundantly. Those which accompany the lateral spines are
situate near the middle of the shaft. Pedicellariee forficiformes form a fringe along
the sides of the ambulacral furrow. The madreporiform body is situated near the
margin, and is frequently almost concealed by the spinelets that surround it.

Size. — This comparatively small species measures from 35 to 50 millims. in diameter,
none of larger size having been recorded.

Colour. — According to Inspector Olrik's observations, made during life, small
examples of A. grcenlandicum are described as of a greyish colour on a violet ground, and
larger ones of a grey olive-green. All the specimens preserved in spirit which we have
examined are entirely bleached, whilst the dried ones are of a dirty yellowish grey.

Habitat. — Mr. Barrett has taken this Starfish upon rocks near the shore ; Insp. Olrik
reports that he has dredged it at very various depths ; and Capt. Feilden obtained the
species from a depth of 80 fathoms, off" Cape Frazer, during the British Arctic Expedi-
tion of 1875-76, under Capt. Sir George Nares.

Variations. — In this species considerable variation occurs in the general form and
habit of the Starfish, as well as in the spinulation borne by the calcareous network of
the test. There may, in fact, be set apart a long, thin-armed variety with the rays taper-
ing but slightly, and another form in which the rays are broad, short, and sharply
tapering at the extremity. The difference of character presented by the spinulation is
perhaps even more conspicuous. In some specimens, for example, the spinelets of the
abactinal surface of the rays are confined mainly to the groups that stand at the intersec-
tions of the network, the intermediate portions of the ossicles bearing only isolated spine-
lets, few and far between ; whilst in others a very marked divergence is produced by
the occurrence of well-developed and comparatively clustered groups of spinelets upon
the intercalary pieces, as well as an increase in the number of those composing the main
series. The effect of this tendency to develop an increased number of spines is perhaps
most striking in its result on the lateral series, where, in consequence of the intercala-
ting ossicle bearing a supplementary spinelet, an additional and intermediate longitudinal
series is formed between the dorsal and ventral rows of lateral spines, extending in some
cases almost to the extremity of the ray ; and besides this, it not unfrequently happens
that the development of extra spines is carried to such an extent that a regular redu-
plication of several of the lateral spinelets at the base of the rays takes place.

It must be remarked, however, that, beyond the simple record of the facts, the
amount of material at our disposal is insufficient to warrant the expression of any decided
opinion as to the significance which these modifications may bear ; for, although at first
sight it would seem probable that these were only the conditions of age and growth, a
careful study of the specimens which we have examined, together with the consideration
of the influence of habitat, lead us to favour the belief that the structural changes above

ARCTIC SEA TO THE WEST OF GREENLAND. 29

noted may mark the effect of conditions of environment, and thus represent the steps of
an actual locational divergence in the species under notice. A much larger series is
needed, however, from various localities, before a definite statement can be made upon
this subject.

Distribution.

a. Northward of Smith Sound : Discovery Bay, lat. 81° 41' N., 25 fms. (Nares's
Exped.), the most northern locality on record ; Port Foulke (Stimpson).

b. North of American Continent : Assistance Bay (under the name of Uraster
violaceus, Forbes, Penny's Exped.); Labrador (Packard); Newfoundland (Lutkeri);
Gulf of St. Lawrence, near Anticosti ( Verrill) ; Grand Manan (under name of A.
Mullen, Stimpson).

c. North of European Continent: Off S.W. coast of Novaya Zemlya, lat. 71° 6' N.,
long. 50° E., 62 fms. ('Willem Parents' Exped.).

Description of the Illustrations of this Species on Plate II.

Fig. 9. Abactinal aspect of the animal : natural size.

10. Actinal aspect of the same specimen : natural size.

11. Portion near the middle of a ray, actinal aspect; magnified.

12. Portion near the middle of a ray, abactinal aspect : magnified.

STICHASTER ALBULUS (Stimps.), Verrill. Plate II, Figs. 13-17.

1842. Asteracanthion roseus (pars), Miiller and Troschel, System der Asteriden, p. 17.
1853. Asteracanthion albulus, Stimpson, Syn. Mar. Invert. Grand Manan, p. 14.
1855. Asteracanthion prollema, Steenstrup, Vid. Meddel. N. Forening i Kjobenhavn, 1854, p. 240.
1857. Asteracanthion problema, Liitken, Vid. Meddel. N. Forening i Kjobenhavn, 1857, p. 30.
1863. Asterias albula, Stimpson, Proc. Acad. N. Sci. Philad. 1863, p. 142.
1866. Stichaster albulus, Verrill, Proceed. Boston Soc. Nat. Hist. vol. x. p. 351.

1875. Stichaster albulus, Perrier, Stellcrides du Museum, p. 82 ; Arch, de Zool. exp. et gen. vol. iv.
p. 346.

A small Starfish with narrow disk and rounded or somewhat arched rays, the num-
ber of which is almost invariably six, three rays on one side being as a rule very much
shorter than those on the other. Proportion of the diameter of the disk to that of
the arms 1 : 5 or rather more. The ambulacral furrows are wide, with suckers arranged
in zigzag series, forming two or four rows, according to age and size. Each adambu-
lacral plate bears two "ambulacral" spines which radiate slightly right and left, and
form regular rows ; occasionally in larger specimens an additional spine accompanying
the ten or twenty innermost pairs of the ray. Closely succeeding to these, there fol-
lows on the side of the arm a slightly oblique series of three (or, in large specimens,
four) similar spinelets, representing ventro-lateral spines, but not always a series opposite

30 ON THE ECHINODERMATA OF THE

to each adambulacral plate, owing to the plates which bear the spines being more widely
spaced. The abactinal body-frame is formed of transverse bands of ossicles ; and these
are packed closely together in longitudinal series along the ray : the plates are very
uniform in size ; and as each generally stands opposed to a corresponding plate in the
neighbouring transverse row, a fairly regular rectilinear arrangement is the result. The
median ossicles, which are somewhat larger than the rest, imbricate upon one another
and form a distinct line along the ray. The membranous interspaces are of moderate size
and punctured with one or two papulae. Each ossicle bears a small subquadrate group
of from three to five short dorsal spines, amongst them being placed two or three of the
pedicellarise forcipiformes, but which are somewhat more numerous towards the sides of
the ray. The spine-groups, in consequence of the regularity of the network, are disposed
in longitudinal and transverse lines, those of the middle row being rather larger and more
densely packed than the others, and thereby forming a more or less distinct medinn
line down each ray. From the apices of the spines, which are broader than the bases,
proceed three or four small denticles ; and the spinelets being all of equal length and
closely set, give a smooth velvet appearance to the Starfish. Pedicellarise forficiformes
are found along the margins of the ambulacral furrow at intervals of every two or three
plates, but are comparatively small in size, short, and somewhat rounded or obtuse at the
points. Two or three larger ones are frequently met with in the neighbourhood of the
actinostome, often quite gigantic ; and a few isolated individuals also occur on the dorsal
surface, but seldom beyond the base of the rays ; these, it need scarcely be said, are
considerably larger than the p. forcipiformes which accompany the spine-groups above
mentioned. The p. forcipiformes that occur on the lateral portions of the rays are, as
a rule, situate on the aboral or outward side of the spines. The madreporiform body
is situated near the margin, and is large, simple (having but few stria?), and surrounded
by a circlet of the dorsal spines.

Size. — This species seldom measures more than from one inch to an inch and a
half in diameter, and is generally less. A variety, noted below, is recorded from
Eastport 4 inches in diameter.

Colour. — In a living state Stichaster albulus is described as of a faint red or
cream-colour. When dried, or in spirit, it is usually light yellow, varying to more or
less warm shades of light brown.

Habitat. — On stony ground and Laminaria in 5-20 fathoms, and seldom in greater
depths (OlriJc). It has been taken by Barrett in from 3-25 fathoms, also on stony
ground (LiitJcen, 1. c. p. 30). Verrill similarly records its occurrence on the coast of
Maine and Grand Manan, in 10-20 fathoms, on rocky bottoms and among Nullipores,
and frequent also at low water of spring-tides among rocks (1. c. p. 351). Stimpson
states that at Grand Manan the species occurred most frequently among branching
Nullipores in 4 or 5 fathoms on the east side of the island.

Variations. — Although the number of rays in this species is almost invariably six,
about two per cent, of the specimens collected have only five, according to Stimpson's
observations at Grand Manan; and examples with seven occur occasionally. The same
authority also records that four out of five had three of the rays much larger than the
others.

ARCTIC SEA TO THE WEST OF GREENLAND. 31

Verrill has described * a variety under the name of nitida, taken at Eastport, Maine,
at low water, but which, in his opinion, presents no peculiarities which may not
be considered due to increased age. The specimen is remarkable for its large size
(4 inches in diameter) and regular form. The rays are equal, and the median row of
plates quite distinct. The lateral spines are four or five in number, in a transverse
row, those nearest the ambulacra being longest. The adambulacral plates bear two,
three, or sometimes four long, tapering, and rather slender spines. The dorsal plates
are crowned by eight to twelve small, somewhat radiating spines, thicker and more
obtuse than those of the ordinary variety. The papulae are not very numerous, rather
large, and occur mostly in pairs. Suckers numerous and much crowded, in four
rows.

Judging from the description alone of a single specimen, it would be difficult to
say whether the modifications above noted possess any greater significance than mere
conditions of growth. Augmentation in the number of ambulacral and lateral spines,
as well as of those upon the ossicles of the disk, greater development of the median
row, increase in the number of papulae, and crowding and increase of the ambulacral
suckers, are, it is true, changes that accompany structural growth ; still, if we are
only to give these characters such a value in the present case, it seems strange that the
form in question is not of more frequent occurrence in localities where Stichaster
albulus may be taken in the utmost abundance.

A single large example which may be referred to this variety was dredged at
Proven during the outward journey of Capt. Nares's Arctic Expedition, the diameters
of the rays and disk being respectively 30 millims. and 6 millims.

Comparing specimens from the coast of Maine with those from Franklin-Pierce
Bay (lat. 79° 25' N.), we have detected no remarkable difference ; and Mr. Verrill records
the same conclusions in regard to specimens from Greenland which he had studied.

Distribution.

a. Northward of Smith Sound : Franklin-Pierce Bay, lat. 79° 25' N., 15 fms.
(Nares's Exped.), the most northern locality on record.

In Davis Strait : Godhavn Harbour and Holsteinborg (' Valorous ' Exped.) ;
Proven, 13 fms. (Nares's Exped.}.

b. North of American Continent : Grand Manan (Stimpson) ; Eastport, Maine
(Verrill).

c. North of European Continent: Spitzbergen (Liltken) ; Iceland, Ofjord (Lutken).

Description of the Illustrations of this Species on Plate II.

Fig. 13. Abactinal aspect of the animal : natural size.

14. A larger and almost equal-rayed variety from Proven.

* Proceed. Boston Soc. Nat. Hist. vol. x. p. 351.

32 ON THE ECHINODERMATA OF THE

Fig. 15. Actinal aspect of the same specimen : natural size.

16. Portion near the middle of a ray, actinal aspect : magnified.

17. Portion near the middle of a ray, abactinal aspect : magnified.

CRIBRELLA OCULATA (Linck), Forbes. Plate II, Figs. 18-21.

1733. Pentadactylosaster oculatus, Linck, De Stellis marinis, p. 35, pi. xxxvi. No. 62.
1776. Asterias sanffuinolenta, 0. F. Miiller, Zool. Dan. Prodr. p. 234, No. 2836.

1776. Asterias pertusa, 0. F. Miiller, Zool. Dan. Prodr. p. 235, No. 2839.

1777. Asterias oculata, Pennant, British Zoology, vol. iv. p. 61, pi. xxx. fig. 56.
1780. Asterias spongiosa, Fabricius, Fauna Groenlandica, p. 368, No. 363.
1805. Asterias seposita, Retzius, Dissert, sist. spec. cog. Asteriarum, p. 21.

1823. Asterias pertusa, Fabricius, K. Danske Vid. Selsk. Skrifter, vol. ii. p. 41, pi. iv. fig. 2.
1828. Asterias oculata, Fleming, Hist. British Animals, p. 487.

1839. Linckia oculata, Forbes, Mem. "Wern. Soc. vol. viii. p. 120.

1840. Henricia oculata, Gray, Ann. <fe Mag. Nat. Hist. vol. vi. p. 184.

1841. Asterias spongiosa, Gould, Invert, of Massachusetts, p. 345.

1841. Cribella oculata, Forbes, Hist. British Starfishes, p. 100.

1842. Echinaster oculatus, Miiller & Troschel, System der Asteriden, pp. 24, 127.
1842. Echinaster Eschrichtii, Miiller & Troschel, System der Asteriden, p. 25.

1844. Echinaster sanguinolentus, Sars, Wiegm. Arch. f. Naturgeschichte, vol. x. p. 16.

1844. Echinaster Sarsii, Miiller & Troschel, Wiegm. Arch. f. Naturgeschichte, vol. x. p. 179.

1844. Echinaster oculatus, Diiben & Koren, K. Yet. Akad. Handl. 1844, p. 241.

1848. Asterias spongiosa, Desor, Proceed. Boston Soc. Nat. Hist. vol. iii. p. 67.

1851. Echinaster Eschrichtii, Brandt, MiddendorfFs Sibirische Eeise, vol. ii. p. 32.

1853. Linckia. oculata, Stimpson, Invert. Grand Manan, p. 14.

1853. Lin ckia pertusa, Stimpson, ibid.

1857. Cribrella sanguinolenta, Lutken, Vid. Meddel. N. Forening i Kjohenhavn, 1857, p. 31.

1862. Cribella sanguinolenta, Dujardin & Hupe, Hist. Nat. Zooph. Echinodermes, p. 349.

1862. Cribella Eschrichtii, id. ibid.

1865. Cribrella sanguinolenta, Norman, Ann. &'Mag. Nat. Hist. ser. 3, vol. xv. p. 124.

1866. Henricia oculata, Gray, Synop. Spec. Starf. Brit. Mus. p. 5.

1866. Cribrella sanguinolenta, Verrill, Proceed. Boston Soc. Nat. Hist. vol. x. p. 345.

1866. Echinaster oculatus, von Martens, Troschel's Archiv f. Naturgesch. Jg. 32, i. p. 84.

1869. Echinaster oculatus, Perrier, Rech. sur les Pe'dicell. et les Ambul. p. 57.

1875. Cribrella oculata, Perrier, Stellerides du Museum, p. 109 ; Arch. d. Zool. exp. et gen. iv. p. 373.

Rays normally five in number, although examples with six or seven are occasionally
found. Proportion of disk- to arm-radius 1 : 4 or 5. The arms are round, excepting a
slight flattening on the under surface, and long, slender, and more or less tapering
towards the extremity. In some specimens there is a considerable swelling-out of the
ray at the base, followed by a constriction at the junction with the disk — a feature
which is generally most developed in female specimens in egg, and must probably be
regarded in a great measure as simply a character dependent on sex.

The abactinal network is composed of very short stout ossicula, irregularly and very
closely reticulated, the interspaces being small in area, in correspondence with the
shortness of the calcareous elements. The ossicula bear more or less compact groups
of fine spinelets, which vary in number, size, and habit, the spinelets themselves being

ARCTIC SEA TO THE WEST OF GREENLAND. 33

3-5-laminate and correspondingly denticulate at the apex. The membrane of the
interspaces is punctured for papulae, of which, in young stages of growth, seldom more
than one is present in each ; but in specimens of larger size one to four or more may
be found. The ventro-lateral plates form two or three series, and their spinelet-groups
are arranged at right angles, or sometimes slightly diagonally, to the median line of
the ray. The foot-papillae are somewhat larger and stouter than the rest of the spinu-
lation ; those which form the immediate margin of the ambulacral furrow consist of
one or two still stouter spinelets, which at an early age are generally arranged in single
file upon the adambulacral plate, and are followed by 2-4 pairs of spinelets merging
gradually into the size of those of the ventro-lateral series. At a more advanced stage
of growth the innermost foot-papillae have not unfrequently become arranged as a pair of
spinelets, very obliquely placed upon the plate, and followed by one or two of similar
size and character, these in turn being succeeded by the series of intermediate spinelets
forming the transition into the ventro-lateral spinulation. Occasionally the innermost
spinelets form an oblique comb-like arrangement on the plate. Each adambulacral
plate also bears another spiuelet, very small in size, situated high up in the furrow,
and generally invisible to superficial examination. The anal orifice is central and
conspicuous. The madreporiform body is large and irregular in form, and situated
midway between the margin and the centre ; in young stages it is covered with spinelets,
but in older specimens these very frequently disappear, leaving visible the more or less
irregular arrangement of curved and dichotomosing striae that intersect the organ.

Size. — The largest Greenland specimen is cited by Lutken as 140 millims., whilst
Sars mentions one 165 millims. in diameter from Esmark in Christiania Fjord. The
usual size is from 50 to 75 millims.

Colour. — The Greenland specimens are described as brick-red when alive (Lutken).
Those on our own coasts are generally dark red or deep purple above and straw-coloured
beneath, whilst occasionally bright vermilion is met with (Forbes). A bright saffron-
yellow variety occurs at Bergen (D. & K.) and Shetlands (Norman), which is a thick-
armed deep-water form ; and it is also on record that the same form was taken off
Fife by Henry Goodsir. In the Asiatic specimens brought home by Von Middendorffs
Expedition, bluish or purple tints seem to have prevailed on the dorsal side, the
underside varying from vermilion to flesh-colour (Brandt, Midd. Seise, p. 34).

Habitat. — From the shore to 175 fathoms depth — a specimen being taken at this
last-named depth off Hare Island, Waigat Strait, during the ' Valorous ' Expedition.

Premature Form. — In young examples the length of the ray is much less in pro-
portion to the disk than in the adult Starfish. The arms are also relatively wider at
the base, and taper more rapidly to the extremity. The spinulation of the dorsal sur-
face is very delicate, in fact almost microscopic, and the ossicles bear fewer spinelets,
which are associated in more compact groups. The ventro-lateral series are more
distinctly defined from the rest of the spinulation than at a greater age, two to four
lateral rows, in addition to the foot-papillae series, being clearly discernible. Seldom
more than a single papula is present in the interspaces of the dorsal network, and the
madreporiform body is more prominently covered with long spines.

F

34 ON THE ECHINODERMATA OF THE

Variations. — Much variation occurs in this species, as might naturally be expected
in an organism enjoying such a wide geographical range ; indeed some of the divergen-
cies are so well marked that they have from time to time been taken by several writers
as types of distinct species. Thus Echinaster Sarsii, M. & T., is in reality nothing more
than the young phase of our Greenland form, whilst its adult stage has served as the
type of Echinaster Eschrichtii, M. & T. (= Asterias pertusa, Miiller). Similarly Linckia
(Cribella) oculata, Forbes (the Asterias spongiosa of Gould), is the shorter-armed and
more densely spinulated variety of the species, and which inhabits, as a rule, the
southern portion of the area of distribution. All these forms, however, when studied
in series, are found to fall, by imperceptible gradations, into simple modifications of the
Starfish named by O. F. Miiller Asterias sanguinolenta, and earlier known to Linck as
Pentadactylosaster oculatus.

Brandt, as well as Diiben and Koren, have already indicated the chief variations
that occur in this Starfish — the former classifying them under the terms of varietas
macrodiscus and v. microdiscus, and the latter authors as forma major and forma minor
(vulgaris), each mentioning the main superficial characters that accompany the
lengthening and shortening of the ray. Brandt records the short-armed variety from
the west coast of the White Sea, and the long-armed one from the Ochhotsk Sea, near
Chantar Island.

Distribution.

a. Greenland: Hare Island, Waigat Strait, lat. 70° 30' N., 175 fms. (' Valorous'
Exped.), the most northern locality on record ; Egedesminde, 50 fms. (Lutkeri).

b. North of American Continent : Grand Manan (Stimpson) ; Massachusetts,
Maine, and east coast of N. America.

c. North of European Continent : Spitzbergen (Lutken) ; Iceland, Fa3roe Islands,
Scandinavia, and British coasts as far south as the English Channel ; White Sea
(Brandt).

d. North of Asiatic Continent. — Sea of Ochhotsk (Srandt).

Description of the Illustrations of this Species on Plate II.

Fig. 18. Abactinal aspect of the animal : natural size.

19. Actinal aspect of the same specimen: natural size.

20. Portion near the middle of a ray, actinal aspect : magnified.

21. Portion near the middle of a ray, abactinal aspect : magnified.

PEDICELLASTEB PAL^OCETSTALLUS, Sladen. Plate II, Figs. 22-26.

1877. Asteracanthionpalceocrystdttus, Sladen, Ann. & Mag. Nat. Hist. ser. 4, vol. xx. p. 455.
1880. Pedicellaster paleeocrystullus, Sladen, Ann. & Mag. Nat. Hist. ser. 5, vol. v. p. 216.

In general appearance this Starfish bears strong resemblance to a small Cribrella, the
rays, five in number, being round and tumid ; they are long, somewhat swollen out on

ARCTIC SEA TO THE WEST OF GREENLAND. 35

the inner third, and taper considerably towards the extremity. The disk is small, its
radius being proportional to an arm-radius as 1 : 5 '5, or rather less. The ambulacral
pores are well spaced, and form two simple rows of sucker-feet. Each adambulacral
plate bears two very slender spines, one radiating towards the furrow and the other
towards the margin, forming two regular rows of so-called " ambulacral " spines. The
spines upon the sides of the arms are much shorter than the ambulacral spines, and com-
paratively more robust, being of the same size and character as the spinelets of the
dorsal surface. The abactinal network is arranged more quadrilaterally than in Asterctr
canthion, a regular median line of ossicles passing down each ray, to which the other
ossicles run parallel and transverse, with more or less regularity. A single spinelet is
borne at each decussation, and an occasional one frequently on the intercalary ossicle.
The spinelets are consequently widely spaced, and assume (although somewhat irregu-
larly) a fairly rectilineal arrangement. These dorsal spinelets, which are all of the same
shape and structure, resemble those of Stichaster albulus. They are deeply grooved,
being formed, in fact, of three or more longitudinal lamellae, which radiate from a
common median axis. The extremity of the shaft becomes slightly expanded and trun-
cate, and delicate denticles, developed from the free angles of the lamellae, are present
at the tip to the number of 3-5. In the ambulacral spinelets the outer margins of the
lamellse are serrate or denticulate, whereby, in these appendages, a thorny appearance
is imparted to the shaft as well as to the tip.

The pedicellariae are remarkably large and numerous, and thoroughly characteristic
of the genus both in structure and position. One kind only is present, viz. the pedicellariae
forcipiformes, which are of extraordinary size, and but very little shorter than the
dorsal spinelets ; they stand isolated and alone upon the membranous tissue that covers
the interspaces of the abactinal network, about three or four being present in each mesh.
No traces of papulae have been detected in the membrane of the dorsal interspaces of
the examples under notice. The pedicellariae increase in size in the neighbourhood of
the ventral portion of the ray, the length of the calcareous skeleton of many of them
measuring 0-35-0-375 of a millim.* The contour of the jaws of the pedicellarise is con-
siderably swollen out about the upper third, and then tapers rapidly towards the extre-
mity, which is somewhat truncate. Several large curved denticles, usually three or
four, occur upon the median portion of the inner margins. The tail-parts of the jaw-
pieces are moderately long, and taper towards the extremity ; and, when the pedicellaria
is examined from above, the interlocking lips of the jaws are seen to be broad, rounded,
and finely denticulate. Every appendage of the body, pedicellariae and spinelets alike,
is invested with a thick, semitransparent, cuticular membrane, to which is due the papil-
late appearance observable in specimens of the Starfish that are preserved in spirit. The
spines are somewhat more crowded upon the disk than upon the rays ; and the " eye "-
spines at the extremities form a robust terminal fringe. The madreporiform plate,
which is obscure and situated near the margin, bears only two or three striae.

* The pedicellarise on a young Asteracanthion glacialis, three or four times as large as the present speci-
men, do not measure more than O19-O22 of a millim.

F2

36 ON THE ECHINODERMATA OF THE

Size, Locality, and Colour. — The largest specimen we have examined measures
30 millims. in its greatest diameter, and 5'5 millims. across the disk. It was collected
by Capt. Feilden in Discovery Bay (lat. 81° 41' N.) ; depth 25 fathoms, hard bottom.
Another example was dredged off Cape Frazer (lat. 79° 44.' N.), in 80 fathoms depth,
which measures only 10 millims. in diameter, and is evidently a young specimen. The
former of these Starfishes is dry, and in that state is of a dirty yellow or grey colour ;
the smaller one has been preserved in spirit, and is of a fawn-colour or light-brown
shade.

Remarks. — The present species is undoubtedly a near relative of Sars's typical form,
P. typicus, but clearly differs from it in general size, proportions, and habitus, as well as
in the form of the pedicellarise and spinelets. P. palceocrystallus is of larger size, and
the length of the arm-radius, in proportion to that of the disk (about 5 : 1), is less than
in P. typicus, in the largest examples of which it is 6^ or 6 : 1. The contour of the
arms is also different in our form, being more tumid on the inner third, and much more
attenuated on the remaining outward portion of the ray. The dorsal spinelets are
decidedly radio-laminate and somewhat expanded at the tip, instead of being conical,
as described in P. typicus, and the shaft of the ambulacral spinelet is also denticulate or
serrate. The pedicellariaB in the present species are even of relatively larger size, and
differ in having the contour of the jaws considerably swollen out about the outer third,
and then tapering rapidly towards the extremity, which is somewhat truncate. Indeed
the general facies of the appendage is unmistakably distinct from that of the more
southern form. (Compare PI. II, Fig. 26, with figure given by Sars, I. c. pi. ix.
figs. 15-17.)

Description of the Illustrations of this Species on Plate II.

Fig. 22. Abactinal aspect of the animal : natural size.

23. Actinal aspect of the same specimen: natural size.

24. Portion near the middle of a ray, actinal aspect : magnified.

25. Portion near the middle of a ray, abactinal aspect : magnified.

26. One of the pedicellarise : magnified.

CROSSASTER PAPPOSUS (Linck), Mutter & Troschel. Plate III, Figs. 1-4.

1733. Triskaidecaetis papposa, Linck, De Stellis marinis, p. 43, tab. xxxii. no. 52, tab. xxxiv. no. 54.

1733. Dodecactis reticulata in dorso, id. 1. c. p. 41, tab. xvii. no. 28.

1777. Asterias helianfhemoides, Pennant, British Zoology, vol. iv. p. 66. no. 72.

1780. Asterias papposa, Fabricius, Fauna Groenlandiea, p. 369. no. 364.

1783. Asterias papposa, Eetzius, K. Vet.-Akad. Handl. Stockholm, vol. iv. p. 230. no. 4.

1788. Asterias papposa, Gmelin, Syst. Nat. Linn. ed. xiii. p. 3160.

1816. Asterias papposa, Lamarck, Anim. s. Vert. ed. 1, vol. ii. p. 559. no. 22.

1821. Asterias papposa, Sabine, Parry's Journ. of a Voyage for the Discovery of a N.W. Passage &c., in

1819-20, Append, p. ccxxii.

1828. Asterias papposa, lloss, Parry's Narrative of Attempt to reach the North Pole in 1827, p. 202.
1828. Asterias papposa, Fleming, Hist. British Animals, p. 487.

ARCTIC SEA TO THE WEST OF GREENLAND. 37

1834. Asterias (Solasterias) papposa, Blainville, Manuel d'Actinologie, p. 241.

1834. Asterias afflnis, Brandt, Act. Acad. St. Petersb. 1834, p. 271, et Prodr. Dcscr. anim. ab Mertensio
obs., fasc. i. p. 71.

1834. ? Asterias alboverrucosa, id. ibid.

1835. Stellonia papposa, Agassiz, Prodr. Monog. Rad., Mem. Soc. Sci. Nat. Neufchatel, t. i. p. 192.

1836. Asterias papposa, Johnston, London's Mag. Nat. Hist. vol. ix. p. 474, fig. 69.

1839. Solaster papposa, Forbes, Ast. Irish Sea, Mom. Wern. Soc. vol. viii. p. 121.

1840. Solaster (Polyaster) papposa, Gray, Ann. & Mag. N. Hist. vol. vi.p. 183.

1840. Crossaster papposits, Miillor & Troschel, Wiegmann's Archiv, iv. pt. 1, p. 183.

1841. Solaster papposa, Forbes, Hist. British Starfishes, p. 112.

1842. Solaster papposus, Miiller & Troschol, System der Asteriden, p. 26.

1852. Solaster papposa, Forbes, Sutherland's Journ. of a Voyage, vol. ii. Append, p. ccxi

1853. Solaster papposus, Stimpson, Invert. Grand Manan, p. 15.

1857. Solaster papposus, Liitken, Vid. Meddel. N. Forening i Kjb'benhavn, 1857, p. 40.
1862. Solaster papposus, Dujardin & Hupe, Hist. Nat. des Zooph. £chinodermes, p. 353.

1865. Solaster papposus, Norman, Ann. & Mag. Nat. Hist. ser. 3, vol. xv. p. 122.

1866. Solaster (Polyaster) papposus, Gray, Synop. Spec. Starf. Brit. Mus. p. 5.
1866. Crossaster papposus, Verrill, Proc. Boston Soc. Nat. Hist. vol. x. p. 345.

1871. Solaster papposus, Hodge, Nat. Hist. Trans. Northumb. & Durham, vol. iv. p. 134.

1875. Solaster papposus, Perrier, Stellerides du Museum, p. 94 ; Arch, de Zool. exp. et gen. vol. iv. p. 358.

1877. Crossaster papposus, A. Agassi/, North-American Starfishes, pp. 98, 112.

1878. Solaster papposus, Viguier, Squelette des Stellerides, Arch, de Zool. exp. et gen. vol. vii. p. 134.

A Starfish of depressed habit, having 10-15 rays, which are shorter than, or only
equal to, the diameter of the disk, and taper uniformly to the extremities. The calca-
reous network of the abactinal surface is very widely spaced, and composed of a great
number of small ossicles that overlap or imbricate upon one another ; the intermediate
meshes are consequently large, and the membranous skin which covers them frequently
bears in the centre one or more isolated ossicles, round which the papulae are grouped,
not unfrequently 20-30 in number. Paxillae, composed of a brush-like group of fine
spinelets articulated on a rounded pedicle, are situated at each of the intersections, one
being also occasionally present upon the line of plates that lies between, as well as
upon, the isolated ossicles ; their length is about equal to their distance apart, and in
an adult specimen 20-30 spinelets occur in each fasciculus. The paxillse stand mode-
rately well spaced in consequence of the open character of the network, and, although
there is no regularity whatever apparent in their arrangement upon the disk, a certain
lineal disposition can be more or less clearly traced upon the rays. 10-12 paxillse
may be counted in a line drawn from the centre of the disk to an arm-angle, 15-20
from the tip of a ray to the base, and not more than 4 or 5 in an oblique row running
from the median line of a ray up to the series of large marginal paxillge. These lateral
paxillse, of which there is only a single series, are large and compressed, being two or
three times the breadth of the dorsal paxillse; they are widely spaced and placed obliquely,
or even at right angles, to the median line, their direction being outwards and at an
angle somewhat greater than 45° to the contour of the ray ; midway between each
of the large paxillaB is a small secondary paxilla, less than those of the dorsal surface.
There are about 16 to 17 lateral paxillae from the tip of the ray to the arm-angle.

Each adambulacral plate bears two series of spines: one upon the inner side,

38 ON THE ECHINODERMATA OF THE

running parallel with the furrow, and another, placed at right angles to this, forming
a radiating comb, which holds a transverse position in relation to the direction of the
ray. In the innermost or furrow series there are 3 to 4 spinelets to each plate, forming
a line frequently more or less oblique and curved in its position ; and their length at the
base of the ray is fully equal to two thirds of the breadth of the furrow at its widest
part. In the transverse series, the spinelets, which are 5-7 in number, are connected
at their base by a membranous skin ; they are stout, and the innermost spines are some-
what larger than those of the longitudinal series.

The interbrachial spaces are very narrow, and either entirely naked or bearing
only a few small spine-groups resembling rudimentary papillae, irregularly disposed here
and there.

The mouth-plates are broad and subtriangular, having a marginal fringe of large
robust mouth-spines upon their outer edges; the four or five innermost, which are directed
towards the actinostome, diverge only slightly from one another ; whilst the following
five or six, which are somewhat smaller, radiate sideways, and interlock more or less
deeply with those of the neighbouring mouth-plate. From the surface of each mouth-
plate, and standing perpendicularly thereto, midway between the outer edge and the
median suture, is a secondary series of two or three large robust spines.

Size. — Large examples are on record measuring 280 millims. in diameter ; but the
northern specimens are usually very much smaller, seldom exceeding 100-150 millims.,
the largest individual from Discovery Bay being only 93 millims.

Colour. — According to Liitken the Greenland specimens are generally of a deep
fiery red ; but in the more southern portion of its area of distribution the species is
frequently more or less mottled with lighter shades, whilst the general tint is often
inclined to shades of purple. Small examples are white or straw-coloured. All colour
is usually bleached in specimens preserved in spirit.

Habitat. — C. papposus frequents a hard-ground locality, its bathymetrical range
extending from low water to great depths, as indicated in the citations of occurrence
given below.

Premature Form. — In a young stage the ossicles of the abactinal network are more
widely spaced, and consequently the paxillse, which at this age are only found at the
intersections, stand further apart and are fewer in number ; whilst the paxillse, although
composed of a smaller number of spinelets, are proportionately larger in size when
compared with those on the adult Starfish. The large compressed lateral paxillse are
fewer in number, as are also the papulae. Consequent on the smaller number of ray-
elements there are fewer transverse ambulacral combs ; and these themselves are made
up of only 3-5 spinelets (the number varying according to age), whilst not more than
two spines of the longitudinal furrow-series are present. The mouth-spines are, in like
manner, less striking, both in character and number.

Variations. — The variations which are the most strikingly conspicuous in this
species are those that affect the relative proportions of the rays to the disk and the
character of the dorsal paxillse. Dr. Liitken * has already remarked on the existence

* Vid. Mcddol. N. Forening i Kjobcnhavn, 1857, p. 42.

ARCTIC SEA TO THE WEST OF GREENLAND. 39

of a long-armed variety, in which the rays are flatter and more pointed, and the paxillae
smaller but more numerous. Examples of a similar form occurred in the collections
made during the British Arctic Expedition of 1875-76 ; and there may also be noted in
these specimens an increase in the number of lateral paxillary combs concurrent with
the greater length of the ray.

The number of rays is also subject to variation, the presence of ten especially
seeming to bear a locational significance. Only one example from Greenland is cited
by Liitken ; but amongst the specimens collected in the extreme north by the naturalists
of Captain Nares's Expedition (mentioned above) this number of rays was the general
rule.

It would seem probable that the two species established by Brandt under the
names of Asterias affinis and A. alboverrucosa should be ranked in the present category,
since the characters upon which the specific individuality of these forms is based arc,
excepting the number of radii, only those which accompany early phases of growth ;
and this fact, together with the small size of the recorded type (disk 1 inch in diam.),
would go far to warrant the conclusion that (in the case of affinis at least) Brandt's
species is nothing more than the young stage of the northern ten-armed form of
C. papposus. The description given in the ' Prodromus Descrip. animalium &c.' is itself
so brief, that it furnishes no features upon which a contrary opinion can be maintained;
but at the same time the statement should not be lost sight of that the diagnosis was
founded not upon an actual specimen, but only upon a drawing by another hand.
Obviously, therefore, no great reliance upon the determination can be accorded by, or
even be expected of, posterity in such a case.

In the recently published memoir by MM. Danielssen and Koren on the Echino-
dermata of the Norwegian North- Atlantic Expedition*, two specimens of Crossaster are
referred to this obscure species of Brandt's (A. affinis) and carefully described (1. c. p. 57).
It would seem to us, however, that the details there cited may, with perhaps only a
single exception, be shown to occur in forms which can be traced through all the stages
of variation up to the undoubted typical C. papposus, and not unfrequently even in
British examples of the species.

One very marked peculiarity, however, is noted in the description above referred
to, and which consists in the great number of spinelets present in the series on the
furrow side of each adambulacral plate, viz. seven (or even eight rarely) ; whilst in large
specimens of C. papposus it is very seldom that more than four occur, and similarly in
the northern ten-armed specimens from Discovery Bay. Such a divergence is very
remarkable. The number of spinelets in the transverse comb is also greater (8-10)
than generally obtains in C. papposus of similar size.

Balancing the whole evidence, however, it would appear very doubtful whether the
divergence is greater than might be expected in a locational or deep-sea variety, and
which we should greatly question the propriety of separating as an independent
species from Crossaster papposus. The specimens were dredged in lat. 64° 35' N.,long.
10° 20' W., 290 fathoms.

* Nyt Mag. f. Naturvidensk. 1877, Bd. xxiii. 3, p. 45.

40 ON THE ECHINODERMATA OF THE

Sir Wyville Thomson likewise records the presence of a ten-armed variety in the
Faeroe channel at a depth of 610 fathoms, but gives no details of its characters.

A variation of secondary importance occurs in the character of the paxillse, by the
length of the spinelets of the crown being sometimes reduced to extreme shortness, and
thereby imparting to the paxilla3 the appearance of small granulate cones. This
condition, however, is in a great measure the result of abrasion, and is dependent
consequently upon the special nature of the locality ; thus a Starfish inhabiting the
comparative calms of deep water would be subject to much less friction than one
frequenting a littoral district or amongst pebbly shingle.

The interbrachial spaces are also liable to change, being sometimes quite naked,
whilst in other cases they are more or less filled with paxillary spinulation.

Distribution.

a. Northward up Smith Sound : Discovery Bay, lat. 81° 41' N., 25 fms., hard
bottom (Naress Exped.), the most northern locality on record ; Cape Frazer, 80 fms. ;
Franklin-Pierce Bay, 15 fms., bottom temperature 29°-5 Fahr.

b. North of American Continent: Assistance Bay (Penny's Exped.), Newfoundland
(Verrill), Grand Manan (Stimpsori), Massachusetts (Verrill, Gould, &c.\

c. North of European Continent: Spitzbergen (LutJcen) ; Barents Sea, lat. 76° 58'
N., long. 45° 40' E., 110 fms. (' Willem Barents' Exped.); Finmark, Scandinavian coasts,
Iceland, Faroe Islands (LutJcen) ; British coasts ; French coast as far as the dept. of
Finisterre (LutJcen).

d. North of Asiatic Continent : Behring's Straits 1 (under the name of A. affinis,
Brandt).

Description of the Illustrations of this Species on Plate III.

Fig. 1. Abactinal aspect of a young specimen.

2. Actinal aspect of the same specimen : natural size.

3. Portion near the middle of a ray, actinal aspect : magnified.

4. Portion near the middle of a ray, abactinal aspect: magnified.

SOLASTEE ENDECA (Gmel.\ Forles. Plate III, Figs. 5-8.

1776. Asterias aspera, 0. F. Miiller, Zool. Dan. Prodr. p. 234. no. 2833.

1788. Asterias endeca, Gmelin, Syst. Nat. linn. ed. xiii. p. 3162.

1816. Asterias endeca, Lamarck, Anim. s. Vert. vol. ii. p. 560. no. 23.

1828. Asterias endica, Fleming, Hist. British Animals, p. 487.

1834. Asterias (Solasterias) endeca, Blainvillo, Manuel d'Actinologie, p. 241.

1835. Asterias endeca, Johnston, Mag. Nat. Hist. vol. ix. p. 300, fig. 44.
1835. Stellonia endeca, Agassiz, Mem. Soc. Sci. Nat. Neufchatel, vol. i. p. 192.

1839. Solaster endeca, Forbes, Mem. "Wern. Soc. vol. viii. p. 121.

1840. Solaster (Endeca) endeca, Gray, Ann. & Mag. Nat. Hist. vol. vi. p. 183.

1841. Solaster endeca, Forbes, Hist. British Starfishes, p. 109.

1842. Solaster endeca, Muller & Troschel, System dcr Asteriden, p. 26.

ARCTIC SEA TO THE WEST OF GREENLAND. 41

1853. Solaster endeca, Stimpson, Syn. Mar. Invert. Grand Manan, p. 14.

1857. Solaster endeca, Liitkcn, Vid. Meddel. N. Forening i Kjobenhavn, 1857, p. 35.

1861. Solaster endeca, Sars, Ovorsigt af Norges Echinodormer, p. 75.

1862. Solaster endeca, Dujardin & Hupe, Hist. Nat. Zooph. Echinodennes, p. 354.

1865. Solaster endeca, Norman, Ann. & Mag. Nat. Hist. ser. 3, vol. xv. p. 122.

1866. Solaster (Endeca) endeca, Gray, Synop. Spec. Starf. Brit. Mus. p. 5.

1866. Solaster endeca, Vorrill, Proceed. Boston Soc. Nat. Hist. vol. x. pp. 345, 350.

1871. Solaster endeca, Hodge, Nat. Hist. Trans. Northumb. & Durham, vol. iv. p. 135.

1875. Solaster endeca, Porricr, Stellerides du Museum, p. 95 ; Arch. d. Zool. oxp. et gdn. vol. iv. p. 359.

Rays 8-11 in number, in length equal to, or only a little longer than, the diameter
of the disk. The disk is high and considerably arched, the rays being consequently
very deep at their base, with sides almost vertical, which gives an appearance of lateral
compression ; they taper rapidly towards the tip, however, and become round or sub-
cylindrical in shape. In the median interradial line, at the junction of neighbouring
rays, there is often a slight depression or furrow, which imparts a very graceful appear-
ance .to the Starfish. The abactinal calcareous network is very compact, the interspaces
being small and constricted. In large-sized specimens, each of the meshes upon the
disk and upper portion of the rays usually contain two (and occasionally three) papulae,
but seldom more than one is present in small examples, and towards the extre-
mities of the rays. A paxilla or small tuft of spinelets (5 to 10 in number) springs
from each intersection, and these form regular parallel lines running obliquely from the
margin to the median line of the ray, in consequence of the symmetrical character of
the calcareous network and the diagonal arrangement of the ossicles across the ray>
10-12 paxillse being counted in such a row at the extreme base of the ray.

The periproct is central and prominent, and the madreporiform tubercle is situated
in an interradium midway between the centre and the arm-angle. There are two rows
of ventro-marginal paxillse, those of the upper series, although larger than the general
dorsal paxillse, being very much smaller than those of the companion ventral series.

The paxillse of the lower series are large and compressed in form ; the lateral
expansion of the pedicle exceeds the thickness by seven or eight times, and the longer
axis is placed at right angles to the median line of the ray. 40-50 of these large
paxillae extend between the tip of the ray and the arm-angle ; the series is situate quite
over on the ventral surface, and runs with a curve round the margin of the arm-angle,
forming a continuous series with those from the neighbouring ray. The paxilla3 which
bound the interradial area are very much smaller than those which occur upon the rays
proper.

Each adambulacral plate bears two sets of spines : — One a small series, placed very
high up in the furrow and running parallel with it, which consists of two or three small
spines, the aboral being largest ; the other series forms a fan-like comb of six or seven
spines, connected by a membranous web, and is borne upon an elevated keel at right
angles to the furrow, almost concealing the small marginal series, which is placed deep
within the groove. The spinelets of the transverse series gradually increase in size as
they approach the furrow.

The interbrachial space, which is elongate, narrow, and sagittiform in contour, is

42 ON THE ECHINODERMATA OF THE

occupied by a number of small paxillse, less than those of the dorsal surface, and seldom,
except in largely grown specimens, bearing more than five or six spinelets, and all of a
compressed, flat, comb-like form. The paxillse are regularly arranged and closely placed
•within the area, but only extend for a short distance along the ray, gradually becoming
rudimentary as the space between the " ambulacral " transverse combs and the marginal
series of paxillae diminishes.

The mouth-plates are large and somewhat spatuliform, having an armature arranged
after the following manner : — The two innermost mouth-spines are long, stout, and
directed towards the centre of the peristome ; then follow a series of about six or seven
spines placed at regular intervals apart, which radiate in the plane of the plate, one after
the other, at an increasing angle from the median line. The two most adoral mouth-
spines are much longer proportionally than the rest, which gradually diminish in size
as they approach the radial series, and being likewise curved somewhat upward from
the plane of the mouth-plate, cross and interlock with the corresponding spines of the
neighbouring mouth-angle. Between the outer margin of the mouth-plate and the side
that falls in the median line of the ray runs a more or less definite keel, upon which is
borne a secondary series of seven or eight spinelets, forming a modified comb that stands
almost perpendicular to the plane of the mouth-plate, and runs parallel with the
marginal series or mouth-spines proper, above described.

Size. — Sars mentions an unusually large example measuring 330 millims. ; but the
ordinary size seldom exceeds 200 millims., whilst the northern forms are usually much
less; thus at Tromso 80 millims., and at Floro 100 millims. (Sars).

Colour. — Shades of red varying to purple on the upper surface and cream-colour
beneath. Occasionally in preserved specimens a trace is left of the natural red colour.
According to Verrill, this species is, in the Bay of Fundy, usually deep purple above
and orange below ; one large example occurred entirely orange.

Habitat. — On stony or sandy bottoms, from low-water mark down to 90 fms.
(Lutkeri).

Premature Form. — In young ten-armed examples from Greenland, 10 millims. in
diameter, described by Liitken, the arms are remarkably short (2^ millims.), and the
dorsal surface is pretty closely covered with groups of 4-6 spinelets, these immature
paxillse being proportionally large and few in number when compared with those on the
mature animal. Bordering the margin of the ray there are about eight paxillae that
separate the dorsal and ventral surfaces, but do not, as in the case of older stages, lay
entirely on the ventral area. Along each side of the furrow is a series of about 14
of the transverse " ambulacral " combs, each of four spinelets. The ventral interbrachial
areas are still quite small, and have only very few spinelets. Each mouth-plate bears
four spines, of which the innermost is longest and is directed horizontally towards the
centre of the mouth-opening. The papula? and the spinelets composing the paxilte
increase in number with age.

We have been enabled to confirm the tenour of these observations by a small
individual dredged by Captain Feilden, during the outward voyage of the British Arctic
Expedition, in lat. 65° N., 26 miles from the coast of Greenland, at a depth of 30 fms.,

ARCTIC SEA TO THE WEST OF GREENLAND. 43

bottom rocky with rounded pebbles. This example is nine-armed, and measures 14
millims. across its greatest diameter, R=7 millims., r=3 millims., the length of the
arm from the tip to the arm-angle being rather less than 4 millims. The ventro-lateral
series of paxillse are about twelve in number on each side of the ray and quite marginal,
instead of being entirely on the ventral surface as in older forms. There are about
twenty of the transverse " ambulacral " combs between the extremity of the ray and the
mouth-plate, each composed of four spines. Each mouth-plate carries four spines on
its margin, which gives eight to the mouth-angle, and there are about two or three of
the inner secondary series present. The dorsal paxillae upon the disk are composed of
four to six spinelets, but those on the rays have seldom more than three.

Variations. — This species on the whole would seem to be very fairly constant in
character, although considerable variation is manifest in the number of rays. Brandt *
notes a ten-armed variety from Sitcha ; and seven-, eight-, and nine-armed forms are cited
by Liitken. The last-named observer also records a variety, taken in Iceland, Floroen,
and Greenland f, in which the rays are blunter and coarser, as well as more uniform in
thickness throughout their length, and in which the paxillae of the ventral interbrachial
spaces are more compressed and carry a greater number of spinelets (10-20). With our
present information, however, it is difficult to say what claim these modifications
have upon our recognition.

Distribution.

a. Greenland: up tolat. 70° N., Omenak (Lutkeri), being the most northern locality
on record ; Davis Straits, lat. 67° -50' N., long. 55° 27' W., 20 fms., bottom of broken
barnacles and shells (' Valorous ' Exped.) ; lat. 65° N., 30 fms., rocky bottom (Nares's
Exped.).

b. North of American Continent : Newfoundland (Liitken) ; St. George's Bank,
25 fms. (Verrill) ; Grand Manan (Stimpson); Fundy Bay and Maine, low-water mark
down to 20 fms. (Verrill).

c. North of European Continent: Finmark, Tromso (Liltken); Iceland (Liitken);
Fa?roe Islands ; British Isles to S. of Ireland and the Channel.

d. Sitcha (Brandt).

Description of the Illustrations of this Species on Plate III.

Fig. 5. Abactinal aspect of the animal : natural size.

6. Actinal aspect of the same specimen: natural size.

7. Portion near the middle of a ray, actinal aspect : magnified.

8. Portion near the middle of a ray, abactinal aspect : magnified.

LOPIIASTEB PUECIPER (Diiben & Koren), Verrill. Plate III, Figs. 9-12.

1844. Chcetaster borealis, Diiben, (Efv. Kongl. Vet.-Akad. Forhandl. 1844, p. 113.

1844. Solaater furcifer, Diihen & Koren, Kongl. Vet.-Akad. Handl. 1S44, p. 243, pi. vi. figs. 7-10.

1862. Solaster furcifer, Dujardiu & Hupe, Hist. Nat. Zooph. Eohinodernies, p. 355.

1878. Lophaster furcifer, Verrill, Amer. Journ. Sci. & Arts, ser. 3, vol. xvi. p. 214.

* Prodr. Descrip. animalium &c. p. 71. t Vidensk. Jleddelelser for 1857, p. 38.

44 ON THE ECHINODERMATA OF THE

A Starfish of rather depressed form, having five broad flat arras ; the proportion
of the greater to the lesser radius being about 3 : 1 approximately. The calcareous
skeleton of the abactinal surface forms a very regular network, and the large con-
spicuous paxillae that spring from the intersections form longitudinal series which run
parallel to the median line of the ray ; consequently only two or three of the middle
rows reach to the tip, although from fourteen to sixteen may be counted at the base of
the arm. The paxillae, which are very compact, have a remarkably large stout pedicle
nearly twice as wide as high, and bear a crown of spinelets, numbering from fifteen
to twenty, or even more in large specimens, in length about equal to the diameter
of the base. The spinelets are flat in young examples, and from the angles of
their apex, which is as broad or broader than the base, proceed two small denticles,
giving to the spinelet the appearance of a two-pronged fork; with increasing age,
however, the spinelets become multilaminate, and the number of apical denticles
corresponds with that of the laminae, which radiate from the median axis of the
spinelet.

The papulae are in groups of from four to ten or twelve in number, varying according
to age and size. They are most numerous upon the disk, the interspaces between the
ossicles of the calcareous network being smaller further out upon the rays, where the
papulae are consequently fewer and the paxillee closer together.

The sides of the rays are deep and perpendicular, and are bounded by a dorsal and
a ventral series of marginal paxillae ; they stand wide apart from one another, and are
about twenty (17-22) in number from the arm-angle to the extremity of the ray. The
paxillae of the dorso-lateral series are larger than the general paxillae of the dorsal area,
and are directed somewhat upwards. The ventro-lateral series, which are much larger
and very prominent, alternate with these, standing opposite to the interspaces, and are
directed at an angle downwards ; their pedicles are nearly twice as broad as the fore-
going and oval in section, whilst the spinelets attached to them are somewhat shorter
than those on the dorsal paxillae. There are no paxillae in the space between the
dorsal and ventral series above named, the sides of the rays being bare.

Each adambulacral plate bears two series of spinelets: one which runs parallel to the
furrow, and is composed of three to five equal-sized spinelets ; and a transverse series of
three or four placed obliquely or even at right angles to the furrow-series, a thick
membrane uniting the spinelets and forming a webbed comb.

The ventral interbrachial areas are large, and bear a number of paxillae arranged
in rows parallel with the furrow, two, three, or sometimes even more of such rows
being present. In the largest example we have examined the innermost row has
eleven or twelve paxilte extending from the median interbrachial line out along the
ray ; the next outermost numbers 5, the third 3, the angle thus left being filled up by
a solitary paxilla. The paxillaj are large and all covered with a thick membrane ; their
bases or pedicles are very short, and carry six to eight spinelets arranged somewhat like
a double fan, all radiating and directed outwards. The main group of paxillae which
stand upon the area are nearly equal in size, but the five or six most aboral ones of the
innermost row diminish in size as they recede from the mouth ; they do not extend

ARCTIC SEA TO THE WEST OF GREENLAND. 45

beyond one third of the length of the ray, and the most outward ones are made up of
only three or four small spinelets, not sufficient to form a paxilla proper.

The mouth-plates are large and broad, and the marginal spines interlock with those
of the neighbouring mouth-angle.

The madreporiform tubercle is small and often difficult to find, being almost hidden
by paxillae ; it is situated a little nearer to the centre than midway between that point
and the arm-angle, and in some specimens it stands at about one third the distance.

Size. — The greater radius of the three specimens known to Diiben and Koren
measured respectively 30, 16-18, and 11 millims. ; the specimens from Cape Frazer
are somewhat larger than this, whilst those from Discovery Bay measure 82 and 53
millims. in their greater radius. The largest of the Cape-Frazer specimens is about
32 millims. in the greater radius, 65 millims. in its greatest diameter, and 21 millims.
across the disk, the arms at their base being 13 millims. broad.

Colour. — According to Diiben and Koren, the colour of the living animal is brick-
red above and white beneath ; the eye-spots bright red.

Habitat. — Although no actual record has been preserved of the nature of the
ground inhabited by this Starfish in the far north, a certain amount of incidental
evidence is furnished by the fact that the smaller of the two specimens from Discovery
Bay has its stomach filled with the remains of Antedon, and upon which it had
evidently taken its last meal ! In all probability they were fellow residents on the
same sea-bed.

Variations. — Under this head should be recorded a specimen obtained by Mr. Hart
in Discovery Bay during the British Arctic Expedition of 1875-76, and which we have
included within the present species only after much hesitation. The greater radius of
this example measures 82 millims., the lesser 30 millims. The arm-angles are much
more rounded, and the rays comparatively much broader at the base and more rapidly
tapering (being remarkably attenuated towards the extremity) than in smaller forms
with which we have previously been acquainted. By this means an extremely broad
character is imparted to the disk, and a contour altogether different from that given by
Diiben and Koren and by Sir Wyville Thomson in figures of Solaster furcifer. The
present specimen is also noteworthy from the great number of papulse that are present ;
in fact so numerous are they, that the dorsal paxillre have the appearance of
springing from quite a forest of these tubelets, whilst the degree to which every
appendage of the body is invested with membrane, and all the ventral spinelets webbed
together, tends to produce a character which, although superficial, is remarkably
striking and conspicuous.

At first sight it would seem that these modifications should be considered marks of
specific distinction ; and such per se the writers would have been disposed to regard them,
had not a second (and smaller) specimen, likewise obtained in Discovery Bay, furnished
a phase of gradation between this apparently independent form and examples of the same
species dredged off Cape Frazer by Capt. Feilden — the larger of these latter presenting
a stage which diverges from the ordinary form (as diagnosed by Diiben and Koren) in
the direction of the above-named specimen.

46 ON THE ECHINODERMATA OF THE

We are still much in ignorance as to the changes undergone by L.furdfer during
growth, whilst the records of its occurrence and structure are so few, that we prefer for
the present to regard the specimens from Discovery Bay as largely developed examples
of the species under notice, rather than to place them as independent forms, from such
scanty material.

Distribution.

a. Northward up Smith Sound : Discovery Bay, lat. 81° 41' N. (Nares's Exped.),
the most northern locality on record ; Cape Frazer, 80 fms.

b. North of American Continent : George's Bank, Gulf of Maine, 150 fms. ( Verrill).

c. North of European Continent: Bergen, 30-50 fms. (Sars, Dliben and Koren).

Description of the Illustrations of this Species on Plate III.

Fig. 9. Abactinal aspect of the animal: natural size.

10. Actinal aspect of the same specimen : natural size.

11. Portion near the middle of a ray, actinal aspect: magnified.

12. Portion near the middle of a ray, abactinal aspect : magnified.

PTERASTER MILITARIS (0. F. Mutter], M. & T. Plate III, Figs. 13-16.

1776. Asterias militaris, 0. F. Miiller, Zool. Dan. Prodr. p. 234. no. 2828.

1788. Asterias militaris, Gmelin, Syst. Nat. Linn. ed. xiii. p. 3100.

1806. Asterias militaris, Eathke, Zool. Dan. vol. iv. p. 13, tab. 131.

1842. Asteriscus militaris, Miiller & Troschel, System der Asteriden, p. 44.

1842. Pteraster militaris, Miiller & Troschel, System der Asteriden, p. 128, tab. vi. fig. 1.

1844. Pteraster militaris, Diiben & Koren, Kongl. Vet.-Akad. Handl. 1844, jx 246, tab. vii. figs. 11-13.

1850. Pteraster militaris, Sars, Nyt Mag. f. Naturv. vol. vi. p. 161.

1853. Pteraster militaris, Stimpson, Invert. Grand Mauan, p. 15.

1856. Pteraster militaris, Sars, Koren, & Danielssen, Fauna litt. Norv. Heft 2, p. 55, tab. vii. figs. 1-8.

1857. Pteraster militaris, Liitken, Vid. Meddel. N. Forening i Kjobenhavn, 1857, p. 43.

1861. Pteraster militaris, Sars, Oversigt af Norges Echinodermer, p. 48.

1862. Pteraster militaris, Dujardin & Hupe, Hist. Nat. Zooph. Echinoderm.es, p. 434.
1866. Pteraster militaris, Verrill, Proceed. Boston Soc. Nat. Hist. vol. x. p. 355.
1871. Pteraster militaris, Verrill, Trans. Connect. Acad. A. & S. vol. i. p. 569.

1875. Pteraster militaris, Perrier, Stellerides du Museum, p. 381 ; Arch. d. Zool. exp. et gen. vol. v. p. 301.

This Starfish is readily distinguished by the singular fin-like margin surrounding
the rays, and by the membranous skin which is supported over the upper surface of
the test, as well as by the largely-developed series of webbed spines that resemble fans
placed on the adambulacral plates at right angles to the direction of the furrow.

The contour of the present species is pentagonal, the body being high and arched
and the underside flat. Proportion of disk-radius to arm-radius 1 : 2. The ossicles
of the abactinal skeleton are subcruciform, and suggest fancifully the idea of a
St. Andrew's cross. The major diameter of these plates lies in the direction of the
ray, and their prolongations imbricate upon corresponding parts of neighbouring

ARCTIC SEA TO THE WEST OF GREENLAND. 47

ossicles, whereby it follows that the intermediate spaces are suboval or diamond-
shaped, and an apparent diagonal arrangement of calcareous plates is produced. Each
of the cruciform ossicles bears a single paxilla situated on the margin of the angle
that lies nearest to the centre of the disk. The pedicle of the paxilla is short, usually
little more than a tubercle, and bears three or sometimes four comparatively long
tapering spinelets ; these diverge at a small angle from one another, and are main-
tained in this position by numerous extremely fine muscular fibres, which are attached
near the tips, and pass from spine to spine in every direction over the entire area. By
this means the foundation is laid for a membranous tissue, which is supported over
the body like a tent-cloth by the spinelets, and by which the whole dorsal surface of
the animal is covered and concealed. A hollow infradermal cavity is thus formed,
wherein the development of the ova and embryos takes place*. Each of the inter-
mediate spaces or meshes in the calcareous network of the skeleton has a single
puncture in its membrane near the base of the paxilla, which gives passage to a papula
of very peculiar form. This remarkable organ, instead of being cylindrical with a
simple conical tip, as is usually the case, has its extremity split up into a number of
short blind cseca or knob-like branchlets, and is attached to the base of the paxilla,
being in all probability non-retractile.

Each adambulacral plate bears a transverse series of five or six long spines, which
are connected together by a membrane, and form a webbed fan that stands at a right
angle to the ambulacral furrow. The outermost spine of each comb is double the
length and thickness of any of the others, but does not always stand in a line with the
rest of the fan-like series, being sometimes nearly midway between its own series and
the next succeeding ; the other spines of the " ambulacral " comb are nearly equal in
size, the middle ones being slightly longer. The long external spines above named
extend about half their length beyond the edge of the ray, and are united to one
another by a connecting-tissue, which forms the fan-like fringe that surrounds the
entire Starfish ; and this also is the boundary of the secondary or supradermal covering
of the dorsal surface before described. The mouth-plates are subquadrate in form ; and
the mouth-spines, which are 5 or 6 in number, are webbed together and directed at an
angle downwards, the innermost spinelet being the longest and stoutest of the series.
From the centre of each mouth-plate, and standing perpendicular to its plane, is a
very large stout spine, much thicker and larger than any of the others ; this is articu-
lated on a small rudimentary tubercle, and has the tip, which is abruptly pointed,
transparent and glass-like.

The anal aperture is situated at the centre of the inner or true dorsal surface of
the animal ; and the madreporiform body, which is circular and more or less arched and
granulate or tubercular in appearance, lies in a median interradial line near to the
centre, and not more than its own diameter away from the periproct. A moderately
wide orifice, surrounded by a circlet of enlarged paxillae, is situated in the centre of

* For details consult Koren arid Danielssen, Fauna litt. Norv., Heft 2, p. 58; Saxs, Overeigt af Jforges
Echinodcrmer, p. 58.

48 ON THE ECHINODERMATA OF THE

the dorsal area of the upper investing membrane, and this forms the common oscular
entrance to the cavity, and stands immediately over the anal aperture. Other openings
into the infradermal cavity occur upon the abactinal surface between each of the long
lateral spines, the apertures being guarded (and can also probably be closed) by a
modified spinelet or scale-papilla articulated on the body-frame. There are also a
number of small circular punctures distributed over the membrane, which Sars referred
to under the style of spiracula, and according to whose observations they are surrounded
with a sphincter muscle that would, in all probability, enable them to be opened and
closed at will. The superficial dermal covering contains a number of very fine
calcareous spicules, and the projecting tips of the supporting paxillary spinelets produce
fine asperities over its surface.

Size. — The northern specimens would appear, according to Sars's observations, to
be larger than those occurring in the southern portion of the area of distribution.
Those taken at Finmark reach the size of 3 inches, whilst specimens from the neigh-
bourhood of Bergen seldom exceed 1J-2 inches ; the largest size on record is 4 inches
(Lilt ken).

Colour. — Sars remarks that the colour of this species is tolerably variable, the
dorsal surface being snow-white, yellowish white, or pale reddish flesh-colour, with a
round blood-red eye-spot at the tip of the arm, and the ventral surface of the animal
always white. The same author cites a small example from Finmark which was
straw-coloured, the tips of the rays orange, a few small scattered spots of the same colour
upon the dorsal surface, and the oscular aperture in the centre being also margined
with orange. Larger examples from the same coast presented the variations above
noted in comparison with specimens from Bergen.

According to the ' Zoologia Danica,' the animal is more of a brick-red, which
colour Liitken also states to have been retained in a large specimen sent to the Copen-
hagen Museum by Capt. Holboll.

Habitat. — This Starfish occurs in 40-60 fms. depth at Bergen, and 20-100 fms.
off the coast of Finmark ; in 35 fms. off Grand Manan, shelly bottom (Stimpson).

Premature Stages. — We are unfortunately not able to furnish any information
upon the characters presented by this species after it has passed the embryonic stage,
and prior to the assumption of the adult features. Sars, and also Koren and Danielssen *,
have, however, made very valuable observations upon the earlier phases, the two last-
named having traced the development of the egg, whilst Sarsf gives very admirable
descriptions and figures of the later stages of the larval form. Four of these were
found in the infradermal cavity of a specimen •£ inch in diameter from Tromso ; and
twelve young, along with two or three eggs, in one 3 inches in diameter from Vadso.
In the young individual of l^—lf millim. in diameter, the pentagonal form is already
apparent, and three pairs of sucker-feet, together with the odd terminal tentacle, are
present in each of the radii. In the place of the actinostome there is a large hemi-
spherical protuberance or knob-shaped body, equal to nearly one third of the animal's

* Fauna littoralis NorvegisD, Heft 2. t Oyersigt af Norges Echinodermer, pp. 58-62, pi. vi.

ARCTIC SEA TO THE WEST OF GREENLAND. 49

diameter, the still unresorbed remains of the larva-body ; and there is yet no trace of
the future mouth-opening. Upon the dorsal surface, and surrounding the centre, are
five large tubercular prominences or wart-like knobs of the body-skin, which Sars
regarded as the rudiments of the perianal paxillae ; and two similar bodies occur at the
margin of each radius, one on either side of the median line, double the size of those
just mentioned and oval in outline, which would seem to be primitive paxillae, or
rather the skin-like sheaths in which these are ultimately formed. No opening was
to be found on the centre of the dorsal surface, nor any trace of the calcareous forma-
tions of the future Starfish.

Distribution.

a. Northward up Smith Sound : Dobbin Bay, lat. 79° 40' N., 30 fms. (Naress
Exped.), the most northern locality on record.

In Davis Strait : lat. 67° 50' N., long. 55° 27' W., 20 fms. (< Valorous ' Exped.).

b. North of American Continent : Grand Manan (Stimpson) ; Bay of Fundy,
35 fms., shelly bottom (Verrill).

c. North of European Continent : Spitzbergen (LutJcen) ; Finmark (Sars) ; Har-
danger Fjord in Norway, the southern limit known to Sars.

Description of the Illustrations of this Species on Plate III.

Fig. 13. Abactinal aspect of the animal : natural size.

14. Actinal aspect of the same specimen : natural size.

15. Portion near the middle of a ray, actinal aspect : magnified.

16. Portion near the middle of a ray, abactinal aspect : magnified.

CTENODISCUS COKNICULATUS (Linck), Perrier. Plate III, Figs. 17-20.

1733. Astropecten corniculatus, Linck, De Stellis marinis, p. 27, tab. xxxvi. no. 63.

1805. Asterias crispata, Retzius, Dissert, sist. species cognitas Asteriarum, p. 17.

1821. Asterias polaris, Sabine, Parry's Journ. of a Voyage for the Discovery of a N.W. Passage &c. in

1819-20, Append, p. ccxxiii, pi. i. figs. 2, 3.

1834. Asterias aurmiciaea, Dewhurst, Nat. Hist, of the Order Cetacea, p. 283.

1840. Astropecten polaris, Gray, Ann. & Mag. Nat. Hist. vol. vi. p. 180.

1842. Ctenodiscus polaris, Miiller & Troschel, System der Asteriden, pp. 76 et 129.

1842. Ctenodiscus pygmceus, id. ibid. p. 76.

1844. Ctenodiscus crispatus, Dtiben & Koren, K. Vet.-Akad. Hand!. 1844, p. 253.

1852. Ctenodiscus polaris, Forbes, Sutherland's Journ. of a Voyage, vol. ii., Append, p. ccxiv.

1853. Ctenodiscus crispatus, Stimpson, Syn. Mar. Invert. Grand Manan, p. 15.

1857. Ctenodiscus crispatus, Liitken, Vid. Meddel. N. Forening i Kjobenhavn, 1857, p. 45.

1861. Ctenodiscus crispatus, Sars, Oversigt af Norges Echinodermer, p. 26.

1862. Ctenodiscus crispatus, Dujardin & Hupe, Hist. Nat. Xooph. £chinodennes, p. 431.
1866. Astropecten polaris, Gray, Synop. Spec. Starf. Brit. Mus. p. 3.

1866. Ctenodiscus crispatus, Verrill, Proceed. Boston Soc. Nat. Hist. vol. x. pp. 345 and 356.
1809. Anodiscus (= Ctenodiscus} rispatus, Perrier, Eech. sur les Pedicell. et les Ambul. p. 106.

H

50 ON THE ECHINODERMATA OF THE

1875. Ctenodisciis corniculatus, Perrier, Stellerides du Museum, p. 380; Arch, de Zool. oxp. et gen.

p. 300.
1879. Ctenodiscus corniculatus, Viguier, Squelette des Stell., Arch, de Zool. exp. et gen. t. vii. p. 226.

Body depressed and goniodiscoid in outline ; radii five in number, with the arm-
angles well rounded ; proportion of greater to lesser diameter 2 : 1. The calcareous
elements of the abactinal surface form a compact network, similar to Astropecten, in
which the interspaces are but very small. A great number of small closely-placed
paxillas are borne upon this framework, each carrying 5-10 round, blunt spinelets, the
whole so densely crowded together that the spinelets are normally directed upward
from the pedicle. The abactinal surface is frequently puffed up and more or less
convex in profile (owing probably to the quantity of sand or clay with which this Star-
fish fills its stomach) ; whilst a small peak-like protuberance rises from the centre,
around which the paxilla? rapidly diminish in size. The sides of the disk are perpen-
dicular, and formed of two series of marginal plates — one ventral, the other dorsal. Each
dorsal plate is ankylosed to a corresponding ventral plate — the pair thus formed being
separated from the neighbouring pair on either hand by a deep furrow, which follows the
lateral suture of the plates, the margins being fringed with a series of fine, compressed,
cilia-like spinelets, which arch over the furrow. Each of the dorso-lateral plates bears
a small compressed, but pointed, spinelet, which stands erect on its upper margin ; and
the lower or ventro-lateral series likewise carry a similar spinelet, which is placed near
the junction of the ventral and dorsal plates, and projects at right angles to the side
walls of the test. The last or terminal dorso-lateral plates of each side of a ray are
ankylosed together, and form a large arched or tubercular plate, indented on its outer
margin, and bearing three more or less prominent tubercles — the rudiments of dorso-
marginal spines. The furrows between the marginal plates are continued onto the
actinal surface of the animal and extend to the ambulacral furrow, cutting up the
ventral interradial areas into band-like spaces, each of which is tessellated with irregular,
subquadrate, scale-like plates that imbricate upon one another, and form normally, in
large adult examples, a double alternating series behind each adambulacral plate. The
innermost band, however, of each area comprises two adambulacral plates ; and the
trapezoid tessellating scales, which here always form a regular double alternating series,
are, in consequence, twice as large in the neighbourhood of the furrow as the scales in
the other bands. All these plates bear on the margin that opens on the sutural furrow
a series of papilla? that form a continuation with the papillae above-noted on the sides
of the lateral plates, from which they differ only in being not flattened, nor are they at
the same time so regular and closely placed. The adambulacral plate presents a wedge-
shaped projection into the furrow, and carries five or six papillae, three only of which
usually stand on the margin of the ambulacral furrow, the remaining two or three
(which are generally much smaller) being situated on the aboral margin opening on the
sutural furrow of the interbrachial area ; not unfrequently, however, one of them is as
large as the ambulacral spinelets, and is placed somewhat inward upon the plate, away
from the sutural fringe and behind the ambulacral series. Towards the extremity of

ARCTIC SEA TO THE WEST OF GREENLAND. 61

the ray the adambulacral plates stand next to the ventro-marginal plates, and are not
separated from them by the trapezoid imbricating scales above described.

Each pair of mouth-plates forms an ovoid mass, the inner or apposed margins of
the plates being elevated into a prominent keel., The innermost pair of mouth-papillae
are very large and thick, and taper to a point — the remainder, from 7 to 9 in number,
being considerably smaller and arranged round the free margin of the plate. Along,
or near to, the median keel of the mouth-plate are 3-5 coarse spinelets, the innermost
being large and thick, and are much less pointed than the marginal series. The
madreporiform body is frequently not more than its own diameter distant from the
margin, and is generally oval in outline and covered with elongate striae running in
the direction of the major diameter.

The entire body and all its appendages are covered with an investing leathery
skin.

Size. — Ordinary specimens are about 30-40 millims. in diameter, the largest
recorded by Sars, from Tromso, being 65 millims.

Colour. — The colour is recorded as brown-red ; specimens preserved in spirit are
either black, greenish, or various shades of drab.

Habitat. — Ctenodiscus corniculatus is found in mud or soft clay bottoms at very
various depths, being dredged by Sars, at Finmark, in 40-200 fathoms depth ; and
further south, at Christiansund, in 40-80 fms. It has also been taken in 25 fms. at
Arksut, by Barrett ; and at Igaliko, in 60 fms., by Insp. Moller (Lutkeri).

Premature Form. — The young form of this species was described by Miiller and
Troschel under the name of Ct. pgmceus. Small individuals, of about half an inch in
diameter, are characterized by the natter test, the comparatively greater prominence of
the latero-dorsal spinelets, and the three large, conspicuously-developed spinelets which
are present on the terminal plate of the ray. The upper margin of this plate, which
lies towards the centre of the disk, is fringed with a series of papilla? similar to those
on the sides of the lateral plates ; and these papillae, as well as the spinelets, appear
to be subject to a greater or lesser degree of obliteration (or resorption) during the
progress of the growth of the Starfish ; in fact in old specimens the spinelets become
reduced to mere tubercles.

According to Liitken, the apical prominence in the centre of the disk is more pro-
minent and characteristically developed in young forms ; but in the specimens which we
have examined (from Novaya Zemlya) it would appear to be quite the reverse, for we
have been unable to detect any difference, except a proportional diminution in size,
from the condition presented by the mature animal. The adambulacral plates in these
specimens bear their papillae or " ambulacral spines " on the furrow-margin, with one
large one placed thumb-like behind them on the aboral side,

At the extremity of the ray there are only two of the ambulacral spinelets ; and
the thumb-like spinelet is larger than either of them, and is persistently present on every
plate. The mouth-plates are small and simple, having only three or four mouth-
papillae on the margin of each plate, and two only on the median ridge (or at most
three), the innermost of these being very large and prominent, and standing perpen-

52 ON THE ECHINODERMATA OF THE

dicular to the plane of the plate, midway between the extremities. When the young
Starfish is examined under the microscope from above, it will be seen that in the lateral
sutural furrows, which open on the dorsal surface, there are, in addition to the marginal
fringe of compressed spinelets, an inner series of fine, pointed, cilia-like spinelets, at
least at the upper portion.

In the early stages of this species, the lower or ventral series of lateral plates lies
much more upon the actinal or ventral surface of the Starfish than it generally does
in the fully grown form. In the examples above mentioned the row of single or
true lateral spinelets, borne by the ventro-lateral plates, stands almost at the angle
formed by the vertical side of the test and the actinal surface, the plates themselves
arching sharply under, onto the actinal surface. The spinelets also seem to occupy a
position relatively nearer to the centre of the plate than at a later stage. In large
specimens no portion of the lateral plates curve onto the ventral area, and the spinelets
are situated much nearer to the upper extremity of their respective plates.

Variations. — Dr. Liitken records examples from a station off Norway, which pro-
bably lies near the southern limit of the area of distribution, that differ strikingly from
the normal short-armed form by their longer and more pointed rays, whereby the
contour approaches that of Archaster Parelii; but in other respects no differences were
noticed. We have observed a similar variation in the radial proportions amongst a
series of specimens from Barents Sea, but not developed to such a marked degree as in
those examined by our learned contemporary. The relative proportions of the greater
to the lesser radius in two specimens are respectively 17 millims. to 9'75 millims.
in the one, and 17 millims. to 7 millims. in the other; two smaller examples
measured similarly 10 millims. to 6 millims., and 10 millims. to 5 millims. — the
difference in character presented by the wide and gentle curve of the arm-angle
of the one in comparison to the more acute and angular outline of the other being
much more striking to the eye than the figures which indicate the actual proportions
would seem to imply. From the occurrence of both these forms together, as well as
• the identity of their general structure, it is, perhaps, not improbable that we have here
nothing more than a sexual character.

Upon the whole, this species would seem to be remarkably constant. After a
careful study, however, of specimens from Greenland, Novaya Zemlya, and North
America, we are inclined to believe that a certain amount of variation does occur
(probably of locational permanence) in the features of the ventro-lateral plates and of the
ambulacral spines, after the manner indicated whilst treating of the phases of growth ;
and although this would seem rather like a confusion of the stages characteristic of
growth with the features presented by circumstantial variation, the evidence has been
such as to lead to the inference that certain characters of early growth-phases are,
in some localities, retained until a much later period of growth, — perhaps even
becoming a permanency through life — a state of things which is perfectly explicable on
the not improbable assumption that the exigencies of arctic existence have acted in

ARCTIC SEA TO THE WEST OF GREENLAND. 53

retarding the progress of growth-characters and in the maintenance of the youthful or
more simple form. The spinulation of the paxillac is similarly subject to variation.

Distribution.

a. Greenland : Hare Island, Waigat Strait, lat. 70° 30' N., 175 fms. (' Valorous '
Exped.).

b. North of American Continent : Melville Island, about lat. 74° 47' N., long.
110° 48' W. (Parry's Exped.) ; Assistance Bay, 7-15 fms. (Penny's Exped.) ; New-
foundland (Sars) ; Bay of Fundy, 50-60 fms. (Stimpson) ; Maine.

c. North of European Continent : Spitzbergen (LutJcen) • Barents Sea, lat. 76°
58' N., long. 45° 40' E., 110 fms. (' Willem Barents ' Exped.), the most northern locality
on record ; Finmark ; Scandinavian coast.

Description of the Illustrations of this Species on Plate III.

Fig. 17. Abactinal aspect of the animal: natural size.

18. Actinal aspect of the same specimen : natural size.

19. Portion near the middle of a ray, actinal aspect : magnified.

20. Portion near the middle of a ray, abactinal aspect : magnified.

54 ON THE ECHINODERMATA OF THE

OPHIUROIDEA.

THE Ophiuroidea collected in Smith's Sound, in Discovery Bay, near Cape Frazer,
Hayes Point, and Franklin-Pierce Bay, and in some other localities in those high lati-
tudes, were numerous in individuals, but the number of genera and species was restricted.
Some specimens were collected in Baffin's Bay, but so far to the south (lat. 65° N.) that
their consideration hardly comes within the scope of this monograph ; they will,
however, be noticed briefly.

All the Ophiuroidea of Smith's Sound and to the north belong, with one exception,
to well-known genera and species ; and it is interesting that a comparatively new genus
should be represented by a well-marked species and large individuals.

List of the Species.

1. Ophiopleura arctica, Duncan. Discovery Bay, 25 fathoms.

2. Ophioglypha Sarsii, Liitken. Floeberg Beach, 10 fathoms ; Discovery Bay,

25 fathoms; Hayes Point, 35 fathoms.

3. Ophioglypha robusta, Ayres. Discovery Bay, 25 fathoms ; Richardson Bay,

70 fathoms; Hayes Point, 35 fathoms; Franklin-Pierce Bay, 15 fathoms.

4. Ophiocten sericeum, Forbes. The same, with the omission of Richardson Bay.

5. Amphiura Holbolli, Liitken. Franklin-Pierce Bay, 15 fathoms.

6. Ophiacantha spimdosa, Miiller & Troschel. Discovery Bay, 25 fathoms ; Cape

Frazer, 80 fathoms ; Franklin-Pierce Bay, 15 fathoms.

7. Astrophyton Agassizii, Stimpson. Latitude 78° 19' N. ; depth 600 fathoms.

Species found to the South.

8. Ophioglypha Stuwitzii, Liitken. Latitude 65° N., 26 miles off the coast of

Greenland ; depth 60 fathoms.

9. Ophiopholis bellis, Linck. In the same place.

The temperature of the water whence the six species from the higher latitudes
were dredged was 29°-5 Fahr.

Genus OPHIOPLEURA, Danielssen & Koren.

LiltJcenia, Duncan.

This genus was determined by MM. Danielssen and Koren in 1867, and the
diagnosis was published in the ' Nyt Magazin for Naturvidenskaberne,' Christiania.
The specimens were obtained in the Norske Nordhavsexpedition, whose Echinodermata
were placed in the hands of those distinguished naturalists. Before this description
came to hand, the genus Lutkenia had been described, in order to include a species

ARCTIC SEA TO THE WEST OF GREENLAND. 55

founded on two fine specimens, in August 1878 *. But it became evident, on the recep-
tion of Danielssen and Keren's communication, that they had forestalled the genus
Lutkenia, which is, of course, a useless synonym of Oplriopleura.

The following is the generic diagnosis of Ophiopleura : —

"The skin of the disk is hard, naked, smooth, and covers a rich plating of scales.
The arms are in like manner overlaid with a smooth, thin skin, which extends up on the
side spines and lets the arm-plates appear. The back of the disk has ten elevated ribs.
Mouth-papillae occupy both sides of the mouth-slits, and are flat, with rounded,
smooth margins. The teeth form irregular rows, are compressed, and have a broad
basis. Foot-papillae. Two genital clefts in each interbrachial space."

The diagnosis published in the ' Annals and Magazine of Natural History,' August
1878, page 188, under the title of Liitkenia, is as follows: —

" Disk notched, covered with very small scales. Radial shields small, widely
separate. Mouth-papillae numerous. Tooth-papillse. Teeth resembling tooth-papilla;
in double series, with accessory knobs. Generative slits small, midway between
mouth-shields and margin. Accessory scales to tentacular openings ; tentacle-scales
numerous ; on mid arm two. Spines small, distant, irregular. Lower arm-plates very
broad and short within the disk, and small and triangular without. Side arm-plates
meeting below throughout, but not above. Upper arm-plates broad and keeled near
the disk."

OPHIOPLEURA ARCTICA, Duncan^. Plate IV, Figs. 1-2 c.

The disk is large, subcircular in outline, tumid above and at the sides, flat below,
and is notched over the arms (1-j^ inch in diameter).

The arms are twice and a half as long as the disk is broad, come well within it, are
very broad within the disk, and considerably so until the second third of their length.
They are flat beneath, convex and almost keeled above near the disk, and less so distally,
tall at the sides, and generally triangular in outline. The arm-spines are very small and
few in number. The colour is white with a little brown.

The upper surface of the disk and the interbrachial spaces, to the aboral edge of
the mouth-shields, and except the naked radial shields, have a stout flaccid derm, covered
with excessively minute scales. The radial shields are small, pear-shaped, narrow, and
angular within, where they slightly overlap, and broad and curved without, with a free
edge there ; they are very distant, and bound the incision for the arm on either side
(-!% inch long). Many small scales, some elongate and others extremely small and oval,
are situated between the radial shields and the arm. There are no radial scales with
spines ; and the generative plates are hidden.

The mouth-shields are small (-£$ inch long), about as broad as long, somewhat
pentagonal, broadest without, angular within, the aboral edge being nearly straight. The

* ' Annals & Magazine of Natural History,' ser. 5, vol. ii. p. 188. P. Martin Duncan on " Liltkenia."
t Ibid. pp. 188 & 266.

56 ON THE ECHINODERMATA OF THE

sides of the shields are rather straight and are at right angles to the aboral margin for
some space, and then they slope inwards to the oral point. An accessory plate exists,
in some, between the oral angle of the mouth-shield and the side mouth-shields. The
madreporic plate is cribriform.

The side mouth-shields are rather large, do not unite closely within, are long and
rectangular, being widest at the side angle of the mouth-shield, where they are more or
less pointed, curved, and rounded off.

The generative slits are short and linear ; the edges are close together, and have on
them small, flat, rounded spinules, sixteen or more on each ; they are distant from the
mouth-shields and from the margin of the disk ; and a series of fine scales passes out-
wards from their distal end, by the side of the arm, to the margin. Other minute scales
are in a patch on the oral side of the slits.

The jaws are rather long, stout, tumid and bossed, and form rather a sharp angle ;
and the lower edge of the jaw-plate is broad and stout ; the angular spaces are wide and
large ; and the tentacles are very well developed.

There are mouth-papilla? and tooth-papillae ; and the teeth are in a double vertical
series with some accessory knobs, so that they resemble large tooth-papillae.

The mouth-papillae are numerous, fourteen or fifteen to each angle, small, much
joined together at their bases, irregular in size, shape, and number, short, and never very
broad. At the apex of the angle, Avithin the distinct jaw-plate, there are three principal,
and one or more smaller, lowest tooth-papilla;. No satisfactory distinction, except that
of position, can be made between these tooth-papillae and mouth-papillae. Above the
three or more lowest tooth-papilla;, the others are in a crowded vertical series. They are
most numerous and small on either side ; and there are six or seven pairs of large, long,
pointed and irregular-shaped ones in the midst, and reaching up the jaw-plate to its upper
end, occupying the position of the teeth. The side of the jaws, close to the jaw-plate,
is occasionally covered with small and close papillae ; and there is a small accessory
papilla close to and at the side of the uppermost large ones. The upper part of
the jaw beneath the stomach is stout and tumid, and the jaw-plate is large and well
developed.

There are two rounded knobs on the side of the jaw, above the attachment of the
mouth-papillae, which are in relation with the upper tentacle ; and the lower tentacle
of the angle has five or six short unequal-sized tentacle-scales, forming, with several
accessory scales, an obliquely placed curved wedge-shaped mass within the first lower
arm-plate and on either side of its oral margin.

The lower arm-plates, there being six or seven within the disk, are mostly very broad
and very short; further out they are small and triangular, with an aboral projection.
They form but a small portion of the lower surface of the arm. The side arm-plates
meet below, from the first to the last, giving a broad and comparatively flat under
surface.

The first lower arm-plate is unlike the others in shape, and it is elliptical in
outline and much broader than long ; the second, longer and very much broader than
the first, is somewhat rectangular ; its sides are slightly incurved ; and there is a

ARCTIC SEA TO THE WEST OF GREENLAND. 57

central angular process on the distal and proximal edges, from which there is a reen-
tering curve on either side to the lateral angles of the sides of the plate.

The third lower arm-plate is very broad, extending across the arm ; it is short, the
relation of length to breadth being one to three ; the sides are incurved for the
tentacle, and are slightly convex towards their distal angle. There is an angular
process or cusp on the broad oral margin, and a smaller one on the aboral ; and there
is a reentering curve on both sides of the processes, giving a very elegant outline.
The fourth lower arm-plate is as broad as the third ; but it is shorter, and the proximal
angular process is more decided than that on the distal edge. The next plate has the
general shape, but is shorter, and the proximal angle is more pronounced. From this
plate to the end of the arm the others narrow more and more, become angular at the
sides and more or less triangular as a whole, and are broader without than within.
There is a projection in the median line on the aboral margin, and a reentering curve on
either side to the lateral angles ; and the proximal angular process has faintly reentering
curves on either side of it. Far out and towards the tip of the arm, the lower plates
become more quadrangular or diamond-shaped ; they are small, broader than long ; and
there is an angular process without and within. At the tip the minute lower arm-plates
have the distal edge curved, and they are angular orally.

The first lower arm-plate is separated from the side mouth-shields by several scales
or plates, which are continuous with the base of the tentacle-scales already mentioned
as being within and at the sides of the plate.

The second lower arm-plate has two or three small tentacle-scales on it. In
some arms they are fused into one ; or there may be several minute accessory scales
present.

The third plate has the same number and accessory arrangement ; and they are
seen, more or less modified, in the fourth and fifth. The other lower arm-plates have
no tentacle-scales.

The side arm-plates form much of the lower surface of the arms, and also the
greater part of the sides in mid arm and towards the tip. All unite with their fellows
along the median line below ; and all are convex from side to side. The first, just
touching its fellow in the median line, is oval in outline, broader than long; it supports
five, short, close tentacle-scales on a curved base ; and the tentacular opening is large
and circular, having a rim of membrane. The second is broader than the first,
touches its fellow, and has four tentacle-scales, and sometimes a fifth or a small
accessory one.

The third side arm-plate, still broader than long, and not much longer than the
second, touches its fellow by a longitudinal short and straight line. The margin without
is curved boldly, and within very slightly. Like the others it is flat below ; and it has
three tentacle-scales, the outer one being subspiniform. The fourth, still broader, is
not longer ; and its outer end is large and supports three tentacle-scales, one of which is
sometimes wanting. The fifth plate the broadest, is short, and is narrow towards the
median line, where it has two small tentacle-scales close together ; and there is a spinule
external to them, and sometimes a second.

I

58 ON THE ECHINODERMATA OF THE

Between the tentacle-scales of these first five or six tentacles and the generative slit,
there are occasionally one or two spinules.

Towards the mid arm, the side arm-plates are tumid at the sides, nearly flat below,
broader than long, and shortest where they are joined longitudinally. They have a
sharp bend to reach the side of the arm, and terminate above in an angular edge by
joining the outer edges of two upper arm-plates. Their distal margin, at the side of the
arm, is thick, and supports two very small, distant, irregular, sharp, short spines and
two tentacle-scales, the inner of which is small and scale-like ; and the outer is usually,
but not invariably, a minute spine longer than the scale and the other spines. Some-
times the two tentacle-scales are equal ; and further out the largest spine becomes
independent of the tentacle. There are often no spines, while some plates have several
very minute ones. All are very ill-developed and small.

The side arm-plates form the bulk of the tip of the arm ; but, although convex at
the sides and swollen above, they do not separate the small and somewhat elongated
hexagonal upper arm-plates there.

The upper arm-plates, within the notch in the disk, are four in number, and are
broad, short, and curved to form a convex roof-like surface. Until far out on the arm,
all are much broader than long, and have slanting straight sides and very faintly curved
distal and proximal margins ; the plates are convex and angular longitudinally, and they
form the upper and much of the side arm. A little beyond the mid arm, the upper arm-
plates are smaller, not much broader than long, broadest without, where they are curved ;
and further out they are longer than broad, narrow proximally, with sides reenteringly
curved and the distal margin boldly curved without. Towards the tip the elongated
hexagonal form is assumed, the distal edge, however, being curved irregularly.

Remarks. — Two specimens of this fine Ophiuroid were collected by Mr. Hart : one
is in spirit, and the other is dry, in the British Museum ; and they both have the same
anatomical details.

The minutely scaled disk, the widely separated radial shields and their free aboral
edge, the position and ornamentation of the generative slit, the presence of mouth- and
tooth-papilla?, the absence of true teeth in the ordinary sense, the papillose nature of
the jaws, the accessory scales to the tentacular openings, the shape of the lower arm-
plates, the numerous tentacle-scales and few spines on the side arm-plates, and the
angular roof-shaped upper arm-plates, whilst they partly suggest Amphiuran and Ophio-
glyphan affinities as a whole, are very characteristic. The absence of spined generative
and radial scales, and the presence of tooth-papillae, separate the new forms from Ophio-
glypha ; and the nature of the dental apparatus and tentacle-scales prevents their being
placed in any hitherto known arctic genus, except OpMopleura.

The specimens were dredged in Discovery Bay at a depth of 25 fathoms, the
temperature being 29°'5. The species has since been found in high latitudes on the
eastern side of Greenland.

OpMopleura borealis, Danielssen & Koren, was not found very far north, and only
in latitude 63° 5' N., and to the east of Greenland ; but the specimens were from
considerable depths, 510 to 570 fathoms, and the bottom temperature was above

ARCTIC SEA TO THE WEST OF GREENLAND. 59

freezing-point, 10>3 Centigrade. It is distinguished from Ophiopleura arctwa by the
presence of 10 very definite rib-like elevations of the upper surface of the disk, and by
its large disk-scaling, and general greater robustness and dimensions of the whole body.
The species Ophiopleura arctica has a more pentagonal outline ; the upper arm-plates
are more convex and medianly more pointed ; and in shape the second and third
lower arm-plates differ entirely from those of Ophiopleura borealis. The other lower
arm-plates of this last species have the breadth, but the few within the disk have not the
aboral point of the arctic form. The jaws differ in shape : the accessory pieces are
not seen in Ophiopleura borealis. And its tentacle-scales are differently arranged and
are more numerous in mid arm than in Ophiopleura arctica. The arctic forms have
round tentacle-spaces, and the others have them elongate, at the root of the arms.
The mass of tentacle-scales and accessory pieces at their base, in relation to the
tentacle at the side of the first lower arm-plate, are very strongly marked in the form
from Smith's Sound, but not so in that described from the sea to the east of Greenland.
Both are very fine forms and large ; and the slight increase of dimensions in the boreal
type is not sufficient to explain the structural differences.

In estimating the importance of the remarkable condition of the true teeth in
Ophiopleura arctica it is necessary to remember the great variations of the teeth of
such a form as Ophioglypha texturata, Lamarck, sp. Hardly any specimens of this
species can be found with entire teeth, so constantly are they divided and conical.
The same remark holds good for Ophioglypha Sarsii, Liitken, sp.

Description of the Illustrations of this Species on Plate IV.

Fig. 1. The upper part of the disk and arms : natural size.
2. The under surface of the disk : magnified.
2 a. The teeth, on the jaw-plate, oblique view : magnified.
2 b. The teeth from above : magnified.
2 c. The teeth approaching : magnified.

Genus OPHIOGLYPHA, Lyman.

OpliioglypTia, Lyman, Proc. Boat. Soc. N. H. vii. 1860.
Ophiura, Forbes.

Disk covered with unequal, crowded, naked, more or less distorted scales, some of
which are swollen. Radial shields naked and swollen. Teeth : no tooth-papillae.
Mouth-papilla? long within, but small and short near the outer edge of the mouth-
tentacles. Arm-spines few, arranged along the outer edge of the side arm-plates.
Tentacle-scales numerous ; the innermost pair of tentacle-pores shaped like slits,
surrounded by numerous scales, and opening diagonally into the mouth-slits. Side
arm-plates meeting nearly or quite below, but not above. A notch edged with
papillse over the arm in the disk. The genital slits starting from the mouth-shields.

l2

60 ON THE ECHINODERMATA OF THE

OPUIOGLYPHA SAKSII, LutJeen, sp. Plate IV, Figs. 3, 4.

Ophioglypha Sarsii, Liitken, Videnskabeligo Meddelelser fra den Naturhistoriske Forening i Kjobenhavn,
November 1854.

An Ophioglyplia with mouth-shields shield-shaped, longer than broad ; length less
than, or only equal to, their distance from the margin of the disk. Papillae of the
disk-incision about fifteen and rather broad. Under arm-plates widely separate, of a
very broad, short triangle-shape. Two tentacle-scales. No infrabrachial indentations.
Spines rather long, equal in length to the side arm-plates.

This is the most northerly Echinoderm brought home by the Expedition, a fine
specimen, with a disk-diameter of 26 millims., having been taken by Capt. Feilden at
the winter-quarters of H.M.S. 'Alert,' in N. lat. 82° 27', in 10 fathoms. Other
examples of this species were obtained at Discovery Bay in 25 fathoms, and among
them one which is provided with remarkably long arm-spines, being in relative
proportion fully twice the length of the spines generally occurring in Ophioglypha
Sarsii. In this individual the three spines of the sixth joint measure respectively
2'45 millims., 2-25 millims., 1'4 millim., the under arm-plate being 0-7 millim. long,
the arm-joint 1 millim., and the disk-diameter 15 millims. The remaining features of
the specimen agree too closely with the characters of Ophioglypha Sarsii (Liitken, sp.),
in our opinion, to warrant its removal from that species, even as a provisional variety.
Specimens were also found oft' Hayes Point.

In some cases great irregularity is exhibited in the mouth-papillae, one abnormal
example being particularly worthy of notice. In the Ophioglyphce the innermost
mouth-papilla generally stands immediately under the teeth, and might be easily
mistaken for a tooth, being, in fact, affixed to the tooth-plate and not to the lateral
plates. In 0. Sarsii, as well as in other members of the genus, two additional
papillae are generally associated with it, one on either hand, and are in like manner
borne by the ossicle upon which the teeth are placed.

In consequence of this arrangement it has long seemed probable to us that these
subdental papillae should be regarded as tooth-papillae (of which they are in truth the
homologues) rather than as mouth-papillae so-called, along with which they are
commonly counted. One of the specimens taken in Discovery Bay throws considerable
light upon this question.

In this individual the dental armature consists of four teeth regularly superposed,
following upon which, and occupying the same breadth as a tooth, are three ossicles
which fit to one another, wedgewise, with sloping sides. Then come two which fit
together and correspond in their shape with the irregularities of the upper and under
tier, which latter consists of from three to five compact, close-fitting papillae ; and these
again are succeeded by three or four (in some rays five) moderately long, round-tipped,
smaller papillae, the whole forming a compact mass suggestive, in the highest degree,
of ordinary tooth-papillae, such as occur, for instance, in Ophiothrix ; and yet in every
detail, even to measurements, the specimen conforms to the diagnosis of Ophioglypha
Sarsii. This individual has a disk-diameter of 22 millimetres.

Bearing in mind the tendency towards vertical reduplication of the mouth-
papillae in some genera, this cannot fail to be regarded as suggestive of the manner in

ABCTIC SEA TO THE WEST OF GREENLAND. 61

which primitive tooth-papillae may have been developed ; nor is such an assumption
by any means extravagant when the great irregularity of these parts amongst Arctic
forms is taken into consideration.

The specimen of Ophioglypha Sarsii, the teeth of which have been alluded to as
being abnormal, is dry and of a dirty white colour. The disk has a pentagonal outline
above, but is circular below ; the notches for the arms in the upper part of the disk
are distinct ; and the scaling is small. There is a small circular scale centrally ; and
the radial shields are embedded, as it were, except at their outer extremity. Several
scales separate the radial shields, some being larger than the others. The radial shields
are separated aborally, by a linear groove, from distinct radial scales. These bear the
majority of the flat, broad, closely-set spinules, which are continued on the interbrachial
space below, along the generative slit, and finally touch the sides of the mouth-shield.
A corresponding series of much smaller, adpressed and blunt spinules is on the side
arms close to the others ; and it can be traced, as a ragged dentate border, to the generative
slit, close to the arm on the under surface of the disk. Three or four small upper arm-
plates exist before the full-sized fifth is seen ; and they are more or less within the
notch ; and their aboral edges are very faintly incised. Below, the mouth-shields have
the usual shape of those of the species, and the side mouth-shields are small, and their
outer and aboral part is in contact with the generative slit. About six close, short,
blunt scales are on each, and are opposed to a smaller number of slightly larger
tentacle-scales on the arm ; and the tentacular opening (the 1st visible but the
2nd really) is long. The jaws are very short, stout, and lumpy inferiorly ; and the jaw-
plate is small. The mouth-papilla?, usually so symmetrical, are placed obliquely, are
irregular; and the usual central one is replaced by from four to six smaller ones,
which have been noticed before. The underpart of the side arm-plates near the
disk somewhat resembles that of Ophioglypha lacertosa, Forbes, sp., being almost
perforated where the oral angle of the under arm-plate nearly or quite touches the
aboral edge of its predecessor. Union of the side arm-plates below, occurs at about
the sixth or seventh. The three arm-spines are small, irregular in shape, are adpressed,
rounded, and rather flat ; sometimes a fourth is seen ; and all are slightly constricted
above their origin. Two tentacle-scales exist to the end of the arm ; but in some
joints there are three, or even four, small scales. In mid arm the tentacle-scales are
on the side arm-plates only ; one is short, broad, flat, and narrowed at its origin, and
is internal to the other. Each either resembles a small conico-cylindrical spine, or is
more like the first. In the neighbourhood of the disk the narrow peduncled and
broad, short, blunted tentacle-scales, normally two on the side arm-plate over the
tentacular opening, have a third, placed flat against the under arm-plate, from which
it springs. Nearer the oral opening, the number of tentacular processes increases as
usual. Where only one tentacle-spine exists, it is broad, and appears to be double with
a common base.

Description of the Illustrations of this Species on Plate IV.

Fig. 3. Ophioglypha Sarsii, Liitken, the top : natural size.

4. Ophioglypha Sarsii, Liitken, the underpart of the disk.

62 ON THE ECHINODERMATA OF THE

OPHIOGLYPHA ROBUSTA, Ayres, sp. Plate IV, Figs. 5-7.

Ophiolepis robusta, Ayres, Proc. Best. Soc. N. H. iv. 1851.
Ofhlura squamosa, Liitken, Vid. Moddel., Nov. 1854.

An Ophioglypha with arms very finely tapering, and disk with regularly arranged
scales of nearly equal size. Mouth-shields ovate shield-shaped ; length less than, or
at most only equal to, their breadth ; length much less than the distance from the
margin of the disk. Papillae of the disk-incision very short and stout, often grouped.
Under arm-plates broadly heart-shaped ; one tentacle-scale.

This species was obtained at various stations, as indicated in the list of localities ;
and though neither the abundance nor the size of the specimens was remarkable,
several good series were collected. The characters which have been regarded as
" specific " are remarkably constant ; and no essential difference can be traced between
these Arctic forms and specimens taken from the coast of Maine, U. S., with which
they have been compared, excepting that in the northern Ophiurans the arm-spines are
longer and somewhat more delicate, and that the outer margin of the under arm-plates
is more arched, and the reentering angle is far less developed, being even altogether
untraceable in certain specimens. In some large examples the upper arm-plates are
very markedly hexagonal. Although this deviation is very constant, the foundation of
a " variety " on the strength of such characters alone, is hardly justifiable.

The arm-spines are moderately stout and tapering, the upper one being flattened
and much larger than the others.

In most of the specimens under present consideration the under arm-plates are
well separated from one another by the side arm-plates, and do not overlap, although in
one individual from Discovery Bay the first ten impinge distinctly, in consequence of
their side arm-plates not meeting. This feature at the basal portion of the arm has
been noted by Dr. Liitken as occurring in large specimens from Greenland, whilst he
remarks at the same time that in none of the Danish examples examined by him do
the under arm-plates touch.

Coll. Feilden. — Discovery Bay, 25 fins., hard bottom ; Eichardson Bay, 70 fms. ;
Hayes Point, 35 fms., bottom temperature 290-5, and also at 25 fms. ; Franklin-Pierce
Bay, 15 fms., bottom temperature 25°'5.

Coll. Hart. — " Winter quarters," Discovery Bay ; Franklin-Pierce Bay, 13-15 fms.,
bottom stony.

The largest specimen was taken by Capt. Feilden in Franklin-Pierce Bay, the
diameter of the disk (dried) being 10 millimetres.

The scaling of the upper part of the disk is very variable in its pattern, relative
size, and colour. In one specimen the usual smaller scaling is in the centre, and the
larger around ; but the central scale is almost circular in outline, and is surrounded
by smaller ones, some of which are triangular, with the apex broadly rounded and the
intermediate one rhombic. This produces a curious rosette, eminently crinoidal in
its configuration, but flat. The ornamentation of grains of calcareous matter on these
scales is very beautiful ; it radiates in numbers of separate and close lines of granules
from the centre of the central scale to the outside of the bluntly triangular scales ; and

ARCTIC SEA TO THE WEST OF GREENLAND. 63

the rhombic scales and the others around them, which are not very definite in shape,
have their ornamentation radiating from their own centres.

In another specimen the scaling is perfectly irregular, no sign of a rosette appears,
and the large radial shields are separated by several small scales in some radii and not
at all in others, when one large overlapping scale is placed between, or rather overlaps
their oral and diverging median line. In other specimens where there is a rosette the
oral separation of the radial shields is by three large plates, one being flanked by others,
or by two plates, one large and one small, the latter being aboral to the other. The
ornamentation is not so much by grains as by cells which are not yet opaque with
carbonate of lime. The radial shields are convex, large, and are broadly convex where
more or less free aborally ; they are separated aborally by small upper arm-plates, in
some instances two, and in others three in number. The shields, thus separated aborally,
and, as has been mentioned, orally, are united in some instances along a very short
line. With regard to the upper arm-plates within the notch, the first is often more
or less triangular in shape, with rounded edges, and the apex is oral ; and when this
shape prevails, this plate is usually larger than the second, which is broad, short, and
sometimes notched distally. The third is larger, and shaped more or less like the
second.

The spinules, which are stout, distinct, and more or less grouped in two rows, are
not often more than six or eight in number. They are beyond the radial shields, and
appear to be on the derm between them and the upper arm-plates. It is evident that
in some specimens there is a spine, or even two (but not in all radii of one specimen),
on either side of the first and second arm-plates, and even on the third. This distribu-
tion of a few well-developed, short, stout, blunt spines is very characteristic. In some
specimens, and in the largest, the first upper arm-plate is in two pieces.

The coloration of the upper part of the disk varies from a perfect white to a
spotted state, with indefinite blue, black, and white, and brown and white. Blotches
appear on the centres of the radial shields and larger plates, so that the whole assumes
a maculated appearance. The slate-blue colour, very light iu its intensity, may prevail
so much on the disk that the white parts come out very decidedly ; and usually that is
the colour of the third row of the rosette, of the scales external to the radial shields,
and of some scales in the interbrachial spaces near the edge.

The upper arm is banded in some specimens with the disk-colour and white ; and
usually the colour is lost towards the termination of the arms.

The upper arm-plates are much more convex in their distal curve in some specimens
than in others ; but in all they are much broader than long near the disk, and in
mid-arm take on the kite-shape with the angle oral and the convex curve distal.
Gradually they become longer than broad ; and towards the end of the arms the side
arm-plates meet above.

There is some variation in the shape and size of the mouth-shields ; but their acute
oral angle and broad aboral part are invariable. The side mouth-shields are small,
narrow, much curved, and just reach to the end of the generative slit. Owing to the
angular nature of the proximal part of the mouth-shields, the side mouth-shields, with

64 ON THE ECHINODERMATA OF THE

their very distinct oral reentering curve, produce as a series a beautiful festooning,
the well-developed front lower arm-plate entering into the circle ; for the side mouth-
shields come up close to its flanks.

The jaws are small, short, narrow, and gibbous inferiorly, and are fringed with
mouth-papillae, one of which, or the homologue of a tooth-papilla, is large and
solitary. This is larger than the others, is often spear-headed ; and there are usually
three smaller ones on either side of the jaws. As a rule, all these mouth-papillse vary
much in size and shape, and project obliquely downwards. The true teeth are angular
and broadly spear-headed in outline ; and they are in a single series of at least five.

The first lower arm-plate is either unlike the others in shape and is perfectly or
irregularly elliptical, being broader than long from the oral to the aboral edge, or it is
heart-shaped and broadest distally and truncated orally. The second and lower arm-
plates are triangular, broad distally, where there is a bold curve, and angular orally,
the sides sloping, with scarcely any reentering curve ; the length and breadth are about
equal. The next and following plates retain their oral angle and broad distal curve,
which is sometimes notched in the median line ; but the sides near the distal corners
are at first parallel and form a space for the tentacle, and then they slope in with more
or less of a reentering curve to the angle. Far out they become smaller and smaller
and fan-shaped, with reentering curves on the sides, and often have a slight notch.

The side arm-plates unite under the arm, and become longer and longer until they
occupy most of the under surface. After the fourth or fifth tentacle-opening there
is one small scale, or, rather, conico-cylindrical and blunt small spine. The first
visible tentacle-opening has four, and rarely five, scales on either side of it ; the next
has some (three) rudimentary scales on the lower arm-plate, and one larger and flat
scale on the side arm-plate ; under the next lower arm-plate there is often a rudimentary
scale, besides the fully developed one on the side arm-plate.

The arm-spines, three in number on mid arm, are slender ; and the upper one is
the longest. There is a row of blunt and distinct small spinules (much smaller than
those in relation to the radial shields) on one side of the generative slit ; but they are
large in relation to those of other species.

Lyman, after noticing that Ophioglypha robusta differs from the other species of
the genus in its short, broad mouth-shields and finely tapering arms, states that it has
been found even in the high latitude of Wellington Channel. He quotes A. E.
Verrill, who states that its range generally is from low-water mark to 18 fathoms.
Lyman's description of the typical specimen and of some varieties may be read with
great advantage.

Description of the Illustrations of this Species on Plate IV.

Fig. 5. The species, natural size, from above.

6. The under surface of the disk : magnified
7. The upper surface of the disk : magnified.

ARCTIC SEA TO THE WEST OF GREENLAND. 65

Genus OPHIOCTEN, Lutken.

OpUocten, Lutken, Vid. Meddel. 1854.

Disk very flat, with margin forming a sharp angle; covered with imbricating
scales and a superficial squamo-granular layer, through which only portions of the
radial shields and primary plates are visible. No disk-incisions, the disk forming a
little arch over the base of the arms. A row of papillae edges the genital slit, and
passes over the arm along the disk-margin, continuous with the series from the other
side. The mouth-shields are scutiform, and are prolonged into the interbrachial space.
The first group of tentacle-scales pass inwards, close to the angle of the jaws. Three
arm-spines, arranged along the outer edge of the side arm-plate, the two upper spines
being much the largest.

OPHIOCTEN SEBICEUM, Forbes, sp. Plate IV, Figs. 8-10, 14..

Ophiocten Kraeyeri, Lutken, Vid. Meddel. 1854, p. 8.

Ophiura sericea, Forbes, Sutherland's Journ. Voyage Baffin's Bay, Appendix.

A typical specimen should have the outline of the disk, seen from above, perfectly
circular, without arm-notches and without straight lines to the margin of the interradial
areas. The disk is slightly tumid above ; and the sharp margin merges into a more
largely scaled and flatter under-disk. Disk tessellated with circular plates, one in the
centre, five around at some distance, and a few others, but all separate and ornamented
with radiating lines of cells, all being surrounded and overlapped by a derm of minute
circular scales cellular in aspect. The radial shields, which are wide apart, are larger
than the other plates, narrowest aborally and broadest within ; obliquely placed, their
distal end being arched and free ; and between them, under the derm, are some rudi-
mentary upper arm-plates. There are distinct radial scales projecting beyond the radial
shields ; and on their distal edge from 5 to 7 spinules with narrow bases, swollen cylindro-
conical trunks, and rather angular and sharp terminations are visible. These are in
a row ; and there are others in some instances projecting from the cells of the derm
between the radial shields, and, deeper still, from the rudimentary upper arm-plates.

The upper arm-plates, broad, short, and convex near the disk, form much of the
arm ; and usually the first five or six large ones are covered with a row of smaller
spinules than those on the edge of the disk, which project along the line of, and close
to, the arm, extending, in some, along the whole distal curve. After the fourth they
diminish in number ; and one or two only are seen on the six arm-plates further out.

The upper arm-plates increase gradually in length, and only diminish in breadth
far out, and at last become more or less triangular in outline, whilst quite at the top they
are long, narrow, triangular, and separated by the side arm-plate. In some specimens
there are no spinules on the upper arm-plates.

Beneath, the scaling is larger and overlapping in the interbrachial spaces, and is

£

66 ON THE ECHINODERMATA OF THE

often large near the mouth-shields ; and the junction of the two kinds of scaling at the
margin is sharp.

A row of spinules is often (but not always) seen along the generative slit, and is
continuous with that seen above the arm.

The mouth-shields are distant from the edge, and are boldly curved and almost
circular, except orally, where they are produced into an angle ; and there are many
variations of this shape. The side mouth-shields are small, swollen and enlarged
where they impinge against the first side arm-plate ; and they end between the first two
tentacle-openings visible under the arm. They are united orally in some instances ; but
as a rule the union is not perfect ; and each carries a broad short scale which protects
the large first arm-tentacle, and which may be divided into two or even three parts.
The jaws are often slightly separated in front of the side arm-plates ; and they are
swollen just within the large and often triangular-looking jaw-plate. There are four
or five mouth-papillae on each jaw-margin, the aboral being broad and short, and the
others more or less spiniform or spear-headed. A large papilla is below the true
teeth on each jaw-angle ; and it and one on either side are to be seen on the jaw-
plate.

The tentacular openings are very large ; and the first under the arm opens, as it
were, along the side of the jaw. The tentacle-scales are few, and like broad plates-
there being not often more than two on the side arm-plate, and two or three on the
side mouth-shield.

The special characters of Opliiocten sericeum, as given by Lyman, are : — " Arms three
to four times the length of the diameter of the disk ; two upper spines much the
longest ; papillae of arm evenly continuous, 1-4 upper arm-plates bearing papillae."

The main variation which we have noted in the Arctic specimens of this species
consists in the greater length of the arm-spines as compared with those of more southern
examples. In a specimen 9-2 millims. in disk-diameter the length of the upper arm-
spine of the sixth joint was T85 millim. (in one case 2' 3 millims.) ; in another, with a
diameter of disk of 8'5 millims., the same spine was 1*8 millim. long, three arm-joints
in this individual being exactly 2 millims. In addition to the above, variations occur in
the contour of the mouth-shields ; and in the larger examples considerable irregularity
is also found in the number and position of the mouth-papillae. Amongst this collection
are several specimens having a very decidedly pentagonal form of disk. In our opinion
none of the above variations can be regarded as of greater morphological significance
than growth-phases, or, at most, than individual variations. The largest specimen
obtained was 1] millims. in disk-diameter.

Numerous specimens were collected by both of the naturalists of the Arctic
Expedition ; and one set was collected by Capt. Feilden in Discovery Bay, 25 fms., hard
bottom, Cape Frazer, 80 fms., and Hayes Point, 35 fms., bottom temperature 29°-5.
The others were in Mr. Hart's collection, from Discovery Bay, 15-20 fms., muddy
bottom, also at 11 fms. ; Franklin-Pierce Bay, 13-15 fms.

ARCTIC SEA TO THE WEST OF GREENLAND. 67

Description of the Illustrations of this Species on Plate IV.

Fig. 8. The upper surface : slightly magnified.

9. The scaling of the upper part of the disk and part of the arm : magnified.
10. Part of the underpart of the disk: magnified.
14. Part of the disk-scaling : magnified.

Genus AMPHIURA, Forbes.

AmpJiiura, Forbes, Linn. Trans, vol. xix. 1842.

Disk small and delicate, covered with naked overlapping scales, and furnished with
uncovered radial shields. Teeth: no tooth-papilla;; six (rarely eight) mouth-papillae
to each angle of the mouth. Arms slender, even, more or less flattened. Arm-spines
short and regular along the sides of the arm-plates. Two genital slits.

AMPHIURA HOLBCELLI, Liitken. Plate IV, Figs. 15-17.

Ampliiura Holbcelli, Liitken, Vid. Medd., Nov. 1854, p. 98.

The mouth-papillae are six to each angle of the mouth (the two innermost stoutest
and pointing to the centre of the mouth), and are oblong and running upwards towards
the teeth. The innermost papilla is attached to the side mouth-shield ; it is small, flat,
with a curved cutting-edge, and is long and sharp like a canine tooth. Teeth four :
two uppermost large, flat, longer than broad, squarish, rather thick, presenting a broad
free surface. Mouth-shields rounded, truncated, and narrower without ; and madreporic
shield the largest, and with pores on its edge. Side mouth-shields large, broad, and
triangular, with all their sides reenteringly curved, meeting within, and also their
neighbours, between the first and second under arm-plates.

Only a single specimen ofAmphiura was taken in high latitudes ; and this, although
it differs slightly from the type form in the relative measurements of certain points of
detail, we have little hesitation in assigning to Dr. Liitken's species, the variations, in
our opinion, not being of greater importance than such as we should regard as dependent
on locality and conditions of life.

The solitary specimen was found in Franklin-Pierce Bay in 15 fathoms, the tempe-
rature being 290-5 Fahr., and presents the following characters :—

An Amphiura with disk lobed ; radial shields long and narrow ; mouth-shields
rounded ; side mouth-shields large, subtriangular, with the sides reentering and angles
rounded. Three pairs of mouth-papillae, the middle ones placed higher than the others.
Under arm-plates pentagonal. One tentacle-scale, rounded. Arm-spines three or four.

The arms are less broad, and take their origin in a more deeply reentering curve of
the disk-margin ; the radial shields are narrower ; and the breadth of upper arm-plates,
in proportion to their length, is less than in the type forms, as the following measure-
ments will indicate: — Diameter of disk 8 millims. ; radial shield, length 1-3 millim.,
breadth 0'35 millim.; sixth upper arm-plate, length 0'6 millim., breadth 0-9 millim.
The spines are hollow cylinders, stout, blunt, and but slightly tapering; the upper spine on

K2

68 ON THE ECHINODERMATA OF THE

each side-plate tapers most. The first fifteen arm-joints bear four spines, the succeeding
joints three only.

An interesting feature connected with this specimen is worthy of record, and is
one which does not appear to have been noted by previous observers. The central spines
are more or less flattened throughout their whole length ; and at the tip, compression
has been carried to such a degree as to form a thin and somewhat expanded head — a
peculiarity which is at once suggestive of a characteristic spine-appendage possessed by
A. Jiliformis ; and although, in the specimen under notice, this structural feature is by
no means so fully developed as in that Ophiuran, it is still sufficiently marked to impress
upon the mind the near relationship of the two species, and the community of their
descent — an hypothesis which is also further strengthened by the association of both the
forms in more southern waters.

Description of the Illustrations of this Species on Plate IV.

Fig. 15. Part of the underpart of the disk : magnified.

16. Radial shields : magnified.

17. Side view of the arm : magnified.

Genus OPHIACANTHA, Mutter & Troschel.

Ophiacantha, Hiiller & Troschel, Syst. Ast. 1842.

Disk with little thorny spines. Radial shields covered. Teeth ; mouth-papillae ;
no tooth-papillae. Spines numerous, slender, minutely thorny, arranged on the sides of
the side arm-plate. Side arm-plates meeting nearly, or quite, above and below. Two
genital slits, beginning outside the mouth-shields. (Lyman, op. cit.)

OPHIACANTHA SPINULOSA, Miiller & Troschel. Plate IV, Figs. 11-13.

OpMacantha spimdosa, Miiller & Troschel, Syst. Ast. 1842, p. 107.

An Ophiacantha with disk covered with small round scales, each bearing a small
short spinelet. Radial shields very obscure, sometimes quite covered. No disk-incision ;
and the dorsal membrane is prolonged over the base of the rays. Mouth-shields twice
as broad as long, irregular ovate. Side mouth-shields long, narrow, arched, and meeting
within. Under arm-plates heptagonal or sub-heptagonal ; breadth equal to length.
Dorsal arm-plates triangular. Side arm-plates meeting above and below. Spines
7 or 8, long, thin, and denticulate, placed on a keel.

A greater number of this Ophiuran were brought home by the British Arctic
Expedition of 1875-6 than of any other Echinoderm. The specimens range in size
from those having a disk-diameter of 15 millims. to the young form of only 3 millims.,
and consequently furnish a most instructive series.

The variations dependent on growth "are very considerable, so much so that isolated
specimens taken from different stages in the series might easily be regarded as affording
the types of distinct species. Conclusive proof has been furnished by the material which
we have had at our disposal that the 0. grcenlandica, M. & T., and the 0. arctica,
M. & T., are untenable species, as Dr. Liitken has already pointed out ; and, further,

ARCTIC SEA TO THE WEST OF GREENLAND. 69

that the characters which had hitherto been looked upon as of specific value, are not,
as that eminent authority seems to infer, even variations such as can be regarded as
dependent on distribution, but must be considered simply the phases incidental to age,
together with ordinary individual variation.

Amongst the specimens procured by the naturalists of H.M.SS. ' Alert ' and
' Discovery ' there are many presenting features developed in a manner which might be
regarded as " ultraspecific " when compared with the previously recognized modifications
of this " form." In the present state of knowledge, however, it seems preferable to
comprehend them under 0. spinulosa of Miiller and Troschel, rather than to burden
further the nomenclature with novel designations.

The mouth-shields and the under arm-plates in this species are subject to very
considerable changes and variations, both in contour and in their relative proportions of
length to breadth. In large and adult specimens the number and arrangement of the
mouth-papillae is also irregular, and not only is there a frequent increase in number in
the ordinal horizontal series, but there is also a great tendency towards reduplication of
certain papillae in the vertical axis of the Ophiuran. This seems to arise from the
longitudinal cleavage of preexisting papillae.

In young individuals the spinelets of the disk are proportionally long, five or six
times their own diameter, and present all the appearances of ordinary embryonic spines.
During the progress of growth, however, increase is made in thickness only ; so that
when maturity is attained, and the spinelets, along with the disk, are invested with the
semitransparent leathery membrane of the body, the appearance is more that of short
stumpy prominences than of actual spines — a deception which at first sight gives a totally
different character to the Ophiuran.

Coll. Feilden : Discovery Bay, 25 fms., hard bottom ; Cape Frazer, 80 fins. ;
Franklin-Pierce Bay, 15 fms., temperature 290>5 Fahr. Coll. Hart: Franklin-Pierce
Bay, 13-15 fms., bottom stony.

Description of the Illustrations of this Species on Plate IV.

Fig. 11. The upper surface : natural size.

12. Part of the under surface : magnified.

13. The spines on the disk: magnified.

ASTROPHYTID^E.

*

ASTROPHYTON AoASSizn, Stimpson. Plate V, Figs. 1-6 a.

1819. Gorgonocephalw arcticus, Leach, Ross's Voyage of Discovery in H.M.SS. ' Isabella ' and 'Alexander,

vol. ii. Append. No. iv. p. 178.

1841. Euryale scutatum, Gould (non Blainville), Invert, of Massachusetts, p. 345.
1853. Astrophyton Agassizii, Stimpson, Syn. Mar. Invert. Grand Manan, p. 12.

1865. Astrophyton Ayassizii, Lyman, 111. Cat. Mus. Comp. Zool. Harvard, i. p. 186.

1866. Astrophyton Agassizii, Verrill, Proceed. Boston Soc. Nat. Hist. vol. x. p. 344.
1866. Astrophyton Agassizii, Ljungman, Ofvers. K.. Vet.-Akad. Fordhandl. 1866, p. 335.
1869. Astrophyton Agassizii, Liitkcn, Yidensk. Selsk. Skr. Usekke 5, Bd. 8, ii. p. 66.

70 ON THE ECHINODERMATA OF THE

1876. AstropTiyton Agassizii, Norman, Proc. Roy. Soc. vol. xxv. p. 208.

1877. AstropTiyton Agassizii, Lyman, Proceed. Boston Soc. Nat. Hist. vol. xix. p. 102.
1877. AstropTiyton arctlcum, Smith, MS.

Disk with prominent radial ribs, which taper slightly towards their inner
extremities, and reach nearly to the centre, — their outer extremities forming a bold
protuberance at the margin of the disk, above the junction of the rays. The ribs
bear a number of irregularly disposed, short, conical spinelets, or spiniform tubercles.
The disk is covered with a leathery skin ; and upon the interradial spaces, which are
smooth and well shield-shaped, a few granules are present. Mouth-shields either
quite insignificant or aborted. The side mouth-plates, which are large and irregularly
pyriform in outline, meet along the greater portion of their apposed sides. The jaws
(scutella oralia) are nearly as large as the side mouth-plates, subtriangular in form,
and touch along their apposed sides, except for a short distance at the aboral extremity
of the same. In spirit it is difficult and often impossible to distinguish the various
parts, in consequence of the thick investing skin. Mouth-papillae and tooth-papillse,
which are spiniform and slightly tapering, are undistinguishable in form ; 6-10, which
may, by their position, be classed as the former, are disposed on the margin of the
jaw-plate, whilst an irregular group stands in the place of tooth-papillse. The
interbrachial spaces on the actinal surface contain a few calcareous plates or
scales ; and at the extreme margin a closely aggregated number of these plates form
a band which divides the upper and lower surfaces of the disk. Under arm-plates
variable and divided ; side arm-plates broad and band-like. The tentacle-scales, which
are spiniform and not flattened, are normally four in each series, this number being
diminished by one or two on the portion of the ray below the first forking. Bays
forked dichotomously, the portions between the forks being comparatively long and
unequal in length. Upper surface of the rays granulated, each joint being marked
off by a suture-like depression, and the granules being gathered into a saddle-like
point on the sides of each joint. A similar sutural depression or furrow traverses the
dorsal median line of the ray. Towards the extremity the granules form a regular
double series round the joint, the space between neighbouring series being several
times the breadth of the band. These granules have claw-like booklets articulated
upon them, which are attached to a right- and left-hand knob alternately, and are
present in greatest numbers towards the terminal portion of the rays. The ray up
to the first forking has but very few granules, the stem next the disk being covered
with a thick investing membrane.

Colour. — According to Stimpson, the radial ribs are yellow, and the interbrachial
spaces brown, in the living animal. In dried examples the ribs are a yellowish brown,
the rays being somewhat lighter in shade, and the interbrachial spaces a rich dark
brown. In spirit preparations this difference is less marked, and frequently destroyed
altogether.

Remarks. — In the preliminary Eeport on the Echinodermata collected during
the Arctic Expedition of 1875-6 (Ann. & Mag. Nat. Hist. ser. 4, vol. xx. p. 468), an

ARCTIC SEA TO THE WEST OF GREENLAND. 71

example belonging to this species was cited under the name of A. arcticum, Leach^
the writers, as there stated, not having had at the time an opportunity of seeing the
specimen. Upon subsequent examination the Astrophyton in question, which had
been dredged off West Greenland by Mr. A. C. Horner (who accompanied Sir Allan
Young in the ' Pandora '), at a depth of 600 fathoms in Smith's Sound, lat. 78° 19' N.,
long. 74° 30' W., was found to be identical with the species which had been described
by Stimpson as A. Agassizii, a form not uncommon in the Arctic Seas of Greenland
and North America. Careful and thorough search was then made for Leach's type
specimen of Gorgonocephalus arcticus (obtained by Sir John Ross), which was
deposited in the British Museum. Prof. F. Jeffrey Bell very kindly took much trouble
on our account, and rendered every assistance in his power by carefully looking up
all the specimens in the national collection. Unfortunately it is impossible to say
with certainty that the object of our quest was found. An old dried specimen, however,
is still extant which is supposed to be Leach's type, and which, from its fragile nature,
has never been moved since it was originally set out by him. This is at least the
traditional (!) representative of Gorgonocephalus arcticus ; for it is without ticket of any
kind. Although this absolute evidence is wanting, there appears to be little doubt about
the correctness of the supposition. This old specimen would seem to be unquestionably
referable to Stimpson's species, and also to accord most closely with the example which
forms the subject of the present remarks. If, therefore, the specimen above
mentioned be really the Gorgonocephalus arcticus of Leach, that name and Astrophyton
Agassizii of Stimpson will stand as synonymous terms.

We quote in full the original diagnosis given by Dr. Leach ; as a further evidence
in support of our views, and also as a testimony to the careful study and acumen of
Mr. Edgar A. Smith, who was the first to refer to this hitherto overlooked and almost
forgotten description : —

" Genus GORGONOCEPHALUS, Leach (1815). Euryale, Lamarck (1816).
" Species Arcticus. Corpore supra radiatim costato : costis tuberculatis, radiis
longissimis, tenuibus, supra granulatis ; articulis (apicalibus prsesertim)
distinctissimis."*

From the brevity and ambiguity of this description, together with the element
of uncertainty which is now inseparable from the type, and which can never be
cleared away, we do not feel justified in restoring Dr. Leach's name in the face of the
present universal adoption of Dr. Stimpson's ; although, if all points were equal, the
former would undoubtedly claim priority.

Distribution.
a. Northward up Smith's Sound: Lat. 78° 19' N., long. 74° 30' W., 600 fins.

* ' A Voyage of Discovery in H.M.SS. ' Isabella ' and ' Alexander ' for the purpose of exploring Baffin's
Bay and enquiring into the probability of a North-west Passage.' By John Boss (1819), Yol. ii. [Append.
No. iv. by W. E. Leach], p. 178.

72 ON THE ECHINODERMATA OF THE

(Homer in ' Pandora '), the most northern locality on record. Lat. 73° 37' N.,
long. 77° 25' W., 800 fms. (Sir John Boss). OS Hare Island, Waigat Strait, lat. 70°
30' N., 175 fms. ('Valorous' Evped.).

b. North of American Continent: Gulf of St. Lawrence (Lyman, Verrill). Grand
Manan, 30 fms. (Verrill). Eastport, at low water of spring tides amongst rocks
(mostly small specimens), and abundantly (all sizes) in 15-20 fms., shelly and stony
bottom (Verrill). Cape Cod (Atwood, fide Lyman).

c. North of European Continent : Vadso, Finmark (Lutkeri).

Description of the Illustrations of this Species on Plate V.

Fig. 1. Abactinal aspect of the animal : reduced about one third.

2. One mouth-angle of a much smaller specimen : magnified.

3. Portion of a ray, abactinal aspect : magnified.

4. Portion of a ray below the first forking, actinal aspect : magnified.

5. Portion of a ray, side view : magnified.

6. Portion of a ray near the extremity : magnified.
6c. One of the booklets : magnified.

ARCTIC SEA TO THE WEST OF GREENLAND. 73

CRINOIDEA.

THREE species of Comatulse were obtained in high latitudes during the British Arctic
Expedition of 1875-76, under Capt. Sir G. S. Nares. These were Antedon Eschrichtii
(the widely distributed northern form), A. celtica, and a new species dredged in Dis-
covery Bay. Respecting the general occurrence of the Crinoidea within the Arctic
Circle, or even within the area treated of in the present memoir, we shall refrain from
speaking, as, unfortunately, this group has hitherto been much neglected by naturalists,
either from the difficulty of obtaining material in a satisfactory state of preservation,
or from the complexity of the subject itself. From this cause little confidence can be
placed in many of the determinations.

Mr. P. Herbert Carpenter is at present engaged in a critical investigation of the
Comatulse ; and having had the opportunity of examining very extensive series of spe-
cimens (amongst others those of the ' Challenger ' Expedition, of which he has already
published a preliminary report*), we feel that the result of his labours, when completed,
will embrace an amount of actual knowledge which, in comparison with an analysis of
the confused determinations of previous writers, will prove to be so much more trust-
worthy and acceptable to biologists at large, that no further remark upon our reticence
in that direction will be necessary.

ANTEDON ESCHRICETII (Muller), Verrill. Plate VI, figs. 1-4.

1841. Alccto Eschrichtii, Hiiller, Archiv f. Naturgesch., Jahrg. vii. vol. i. p. 142.
1853. Alecto Eschrichtii, Stimpson, Mar. Invert. Grand Manan, p. 12.
1857. Alecto Eschrichtii, Liitken, Vid. Meddel. N. Forening i Kjobenhavn, 1857, p. 55.
1862. Comatula Eschrichtii, Dujardin & Hupe, Hist. Nat. Zooph. £chinodermes, p. 199.
1866. Antedon Eschrichtii, Verrill, Proceed. Boston Soc. Nat. Hist. vol. x. p. 343..

Dorsocentral semiglobular, closely covered with a great number of cirri, a hun-
dred pits being present in large specimens. Frequently, in old examples, the imme-
diate apex is smooth and not pitted ; and sometimes this portion of the plate is slightly
indented, forming a little hollow or depression, surrounded on all sides by cirri.

The first radial is invisible (or with only the smallest trace to be seen at the sides
in small examples), the second radial appears on the outer surface only as a very short,
thin plate at the margin of the dorsocentral, and is seen to thicken at the sides as it
recedes inwards ; the third (or axillary) radial is a large, conspicuous plate, rhomboid or
somewhat diamond-shaped in external contour, length greater than the breadth ; the
sloping sides, which diverge from the distal angle on the margins of the articular facet
being somewhat incurved and slightly longer than the lower sides, which join to form
the proximal angle, these latter sides being incurved for their distal half and then grace-
fully bent outwards to form the slightly rounded proximal angle. Considerable variation

* Troc. Hoy. Soc. no. 194, 1879, p. 392.

74 ON THE ECHINODERMATA OF THE

is to be found in different specimens in this lower angle ; and when very obtuse a much
smaller relative proportion of the length of the plate lies below a horizontal line drawn
through the lateral angles, than is the case when the proximal angle is smaller.

The first brachial joint is short and cuneiform ; the second often as long as, or even
longer than, broad ; the proximal margin forming a fairly acute angle in the profile
contour of the plate. This joint has also the appearance of being very considerably
twisted round to the front ; so that the afore-mentioned proximal angle is found opposite
the middle part of the margin of the horizontal suture of the first brachial, when seen
from the front. The third brachial is a comparatively short joint, of nearly equal
length at either side, and bears the first syzygy. The succeeding joints of the ray have
their longer side shorter than their breadth, and taper wedge-form to the other side, —
their profile, when seen from the exact median line of the ray, being regularly triangular.
On the lower portion of the ray, the angle formed by the longer side of the arm-joints
and the proximal suture is prolonged into a prominent peak, which is also brought
somewhat forwards. The neighbouring joint has a little corresponding projection or
lip, upon which this articulates ; and owing to this structure, a semitubercular or knobby
character is given to the lower portion of the ray, which is highly characteristic of the
present species. The rays are very robust, and maintain this character throughout.
Towards the extremities the joints become very short, and their breadth equal to
several times the length of their longer side. Syzygies occur on the 3rd, 8th, 12th,
15th, 18th brachial joint, and so on — the first two being very constant, the third some-
times varying onto the llth or 13th joint; in the latter case the next syzygy does
not occur till the 18th brachial is reached.

The pinnules are very robust, 100 being present on the one side of an arm of a
moderate-sized specimen ; and the joints are more or less compressed. The first five or
six upon the ray have the flattened dorsal prominences which give the peculiar saw-like
character mentioned by Miiller and others. As a rule, the lowest pinnules in this species
are of fairly uniform length, increasing slightly as they proceed along the ray ; some-
times, however, the first pinnule is somewhat longer than its immediate successors ;
but there is no great disparity in the second and third, such as occurs in the two next
species.

In a first pinnule, measuring 20 millims., were 45 joints ; in another, a fraction
longer, 48 joints ; and the third and fifth pinnules measured, within a fraction, the same
length. A pinnule taken midway along the ray was 25 millims., and had 36 joints — in
this position the joints being relatively longer and more cylindrical, and their distal
margin finely denticulate. In all the pinnules after the first eight or ten, the two
lowest joints are very much larger than the rest, and they maintain this disparity
throughout the ray ; their form, also, is noteworthy, the first or articulatory joint being
very compressed and somewhat halfmoon-shaped, whilst the proximal margin of the
second joint is also considerably incurved, so that a large vacant space is left between
the two joints, which is occupied by muscle or ligament, — these first two joints being
the only ones that preserve along the ray the slightest trace of the dorsal carina?,
which have been mentioned above as characteristic of the joints of the entire

ARCTIC SEA TO THE WEST OF GREENLAND. 75

pinnule at the commencement of the series. The ovarial sacs attached to the pinnules
are large and thick — a circumstance which, in conjunction with the closeness of the
pinnules to one another, adds to the compact and dense character of the plume as a
whole.

The dorsal cirri are long and robust; and the joints are slightly compressed. Con-
siderable variation occurs in the length and in the number of the joints. A long
example measuring 60 millims., contained 42 joints ; and another, on the same specimen,
about half that length, had 32. The longest joints, which commence at about the
12th from the base, are twice as long as broad, and are somewhat dicebox-shaped,
the proportion of length to breadth diminishing rapidly towards the extremity. The
terminal joint consists of a robust claw as long as the preceding joint ; the penultimate
joint bears no secondary claw, but simply has the margin thickened and produced into
a sharp angle — a development which also occurs to a certain extent on several of the
preceding joints.

Locality. — Franklin-Pierce Bay (Feilden).

Description of the Illustrations of this Species on Plate VI.

Fig. 1. Antedon Eschrichtii : natural size.

2. Diagrammatic sketch of the radial plates.

3. Extremity of one of the dorsal cirri : magnified.

4. A first pinnule : magnified.

ANTEDON CELTICA (Barrett], Norman. Plate VI, figs. 5 & 6.

1857. Comatula Woodwardii, Barrett, Ann. & Mag. Nat. Hist. ser. 2, vol. xix. p. 32, pi. vii. f. 1 (non

Forbes).

1857. Comatula celtka, Barrett, Ann. & Mag. Nat. Hist. ser. 2, vol. xx. p. 44.
1865. Antedon celtieus, Norman, Ann. & Mag. Nat. Hist. ser. 3, vol. xv. p. 104.
1877. Antedon celtkus, Marenzeller, Denksch. k. Akad. Wissensch. Wien, Bd. xxxv. p. 24 (separate copy).

Dorsocentral semiglobular or hemispherical, and somewhat flattened*. Sixty
to seventy cirrus-pits may be traced on the specimens under notice. A cirrus
measuring 40 millims. in length has 38 joints: these are somewhat flattened; and the
longest occurs at about the 9th or 10th, which is twice as long as broad — this propor-
tion gradually decreasing as they approach the extremity, where the breadth is greater
than the length. The terminal joint is a robust claw, longer than the preceding joint,
which has the opposed margin angulated and sharpened, but not enough to form a
secondary claw. In the smaller cirri, however, this sharpened joint is more con-
spicuous. In the shorter cirri, which are scarcely half the length of those above
mentioned, the average number of joints is about 25, and the distal extremity of each
joint is broader than the proximal which next succeeds, in a more marked degree than
in the older cirri.

* Mr. P. Herbert Carpenter has figured and described this piece as elongate and conoid in young spe-
cimens (Trans. Linn. Soe. Zool. ser. 2, vol. ii. p. 61, pi. iv. fig. 8). No trace of this shape remains in our
examples, which measure about 135 millims. in length.

L2

76 ON THE ECHINODEKMATA OF THE

The three radial plates are visible; the first is a very short band-like plate at the
margin of the centrodorsal ; the second is comparatively long, and diminishes with a
somewhat rapid but graceful curve into a neck-like form at the distal extremity upon
which the axillary radial is articulated ; and there occurs in the median dorsal line a
slight prominence or lip opposed to the proximal angle of the axillary. This latter,
which is the third radial plate, is quadrately diamond-shaped in its outer contour,
and somewhat broader than long, the distal angle being more obtuse and the margins
of the distal facets much less incurved than in the preceding species. The first
brachials are also relatively longer and more conspicuous than in A. Eschrichtii ; and
the second brachials are subtriangular or even subcrescentiform in profile (as seen from
outside), and are broader than long, their apex being brought forward very considerably
to the front. The third brachial, which bears the first syzygy, is nearly as long as
broad. At some distance from the disk the joints of the ray are almost as long at their
deepest part (i. e. on their longer side) as they are broad, the length of the shorter side
being, perhaps, rather more than one third of the breadth on the lower third of the
ray. The sutures of the joints are not nearly so much inclined as in A. Eschrichtii', and
the joints diminish in length as they approach the extremity of the ray. Syzygies
occur normally on the 3rd, 8th, 12th, 15th brachial, and so on — although the second is
sometimes found on the 9th, the third on the loth or 14th, and the fourth on the 16th
or 17th brachial joint.

80-90 pinnules occur along one side of a ray. The arms, as a whole, are long,
tapering, and decidedly delicate in character, although so numerously pinnulated ;
whilst the narrowness of the calyx, the angle of ray-attachment, and the comparative
shortness of the pinnules give a very long and narrow appearance to the closed plume.

The first and second pinnules are equal in length, measuring 15 millims., and
contain 36 joints each. The third pinnule is very much smaller, being scarcely
11 millims. long, and with only 23 joints. One from the middle of the ray has like-
wise 23 joints, and measures 15 to 16 millims. long. In another specimen the first
pinnule had 37 joints, and the third 18. The joints on the proximal third of the first
three pinnules have the dorsal side produced into flattened processes, similar to those
developed much more prominently in the last species. The pinnules beyond the first
three have very much longer joints, as will be seen by reference to the measures
above given ; and these are also more cylindrical in shape. In these pinnules, further-
more, the first two joints are much flatter, broader, and shorter than the rest, but do not
present the conspicuous character noted in A. Eschrichtii. The ovarial sacs are thin,
delicate, and semitransparent, presenting a very marked difference in comparison with
those of the latter species.

Remarks. — It is not without considerable hesitation that we have referred the form
under notice to Barrett's species ; for the original diagnosis is so meagre that little can
be made out of it. Dr. Marenzeller, however, has given a perfectly intelligible and
more comprehensive sketch of (what is in our opinion) the same species, in the
description which accompanies his determination of examples obtained by the Austro-
Hungarian Arctic Expedition, with which specimens ours would seem to accord in all
essential particulars.

ARCTIC SEA TO THE WEST OF GREENLAND. 77

Mr. P. Herbert Carpenter (loc. cit.) has entered with great care and detail into
certain parts of the anatomical structure of a Comatula referred to under this name ;
but no other writer, excepting those mentioned above, appears to have spoken of
A. celtica from a systematist's point of view*.

Locality, &c. — Discovery Bay, lat. 81° 41' N., 25 fathoms, hard bottom (Feilden) ;
Franklin-Pierce Bay.

Description of the Illustrations of this Species on Plate VI.
Fig. 5. Antedon celtica : natural size.

6. Diagrammatic sketch of the radial plates.

ANTEDON PEOLIXA, Sladen, sp. nov. Plate VI, Figs. 7-10.

Dorsocentral very conoid, as long as, or longer than, broad, with four or five tiers
of cirri arranged one above the other, about 60 to 70 all together being present ;
the apex having no pits, and presenting the appearance of a sort of tubercular boss.
The cirri are remarkably long and delicate — one, with the extremity missing, measuring
58 millims. Another, 47 millims. long, has forty joints; and this appears to be the
average length and number. One of the small cirri (which is 21 millims.) has twenty-
nine joints. The joints of the cirri are very long and cylindrical, but narrower in the
middle than at the extremities; the eighth from the basal joint measures 1*76 millim.
long and '43 millim. broad, or four times as long as broad ; the terminal joint is
modified into a delicate claw ; and no actual secondary claw occurs at the extremity of
the penultimate joint, only a slightly produced triangular peak.

The first radial is not visible in front, only at the extreme sides, where it may be
seen rising up towards the interior. The second radial is of moderate length, nearly as
long at the sides as broad, and is not much (if at all) constricted at the distal extremity.
The third or axillary radial is very regularly quadriform, placed diamond-wise (the
length being about equal to the breadth), and with all the sides incurved. The first
brachial is comparatively long ; the second longer than broad, and subtriangular in
profile from the outside, the proximal angle remaining in a much more lateral position
than in either of the preceding species ; the third brachial, which bears the first syzygy,
is as long as, or longer than, broad; and the succeeding joints of the ray are all
comparatively long, and increase in relative length towards the extremity of the ray.
Syzygies occur normally on the 3rd, 8th, 12th, and every third joint beyond.

On the whole it may be said that the radials stand very high in this species, and
that the rays spring off from them well separated into pairs — the rays being comparatively
thin and delicate in habit, and the pinnules being seemingly placed wide apart, in
consequence of the length of the joints.

* Since -writing the above, we have had the opportunity of examining specimens which we owe to the
kindness of Sir Wyville Thomson, and which he, in common with other British naturalists, has been in the
habit of regarding as the representatives of A. celtica. They are altogether different from the Comatula here
described, and resemble to a certain degree the form that we have named A. prolira, but from both of which
they arc, in our estimation, a perfectly distinct species .

78 ON THE ECHINODERMATA OF THE

41 pinnules are present on one side of the longest unbroken ray, which measures
86 or 87 millims., — a little portion of the extremity being wanting. From the axillary
radial to the third syzygy measures 14-5 millims. The first pinnule measures 10 millims.
long, with 26 or 27 joints ; the second is very short, being only 3 -25 millims., and has
11 joints ; the third is 5 millims., with 11 joints ; the tenth 8 millims., with 12 joints ;
the twentieth about 14 millims., with 19 or 20 joints. In another example the first
three pinnules have 22, 15, and 14 joints respectively.

The joints of the pinnules are cylindrical and very long, excepting the first six of
the first pinnule, which are little longer than broad, and slightly compressed ; and the
three or four lowest joints in the second and third, which are also short and somewhat
flattened, and diminish rapidly in breadth at their distal extremities. The second joint
of the pinnules along the ray is broad and robust in comparison with the rest ; and the
proximal extremity is truncated on either side, so as to form apparently two sloping and
converging facets for the reception of ligament : owing to this circumstance, and also
to the mode of their attachment, the pinnules have the semblance of being placed
transversely upon the first joints ; and this presents a striking character about the
middle of the ray. Ovarial sacs fine, delicate, and chiefly developed on the proximal
third of the pinnule.

Remarks. — The present species more nearly resembles A. phalangium, Miiller, of
the Mediterranean, than any other Comatula with which the writer is acquainted.
That it is distinct, however, from that form will readily be seen from the preceding
description. It is altogether different from its Arctic congeners.

Locality &c. — All the specimens of this species were dredged by Capt. H. W. Feilden
during the British Arctic Expedition of 1875-76, in Discovery Bay, lat. 81° 41' N.,
depth 25 fathoms, hard bottom. Antedon celtica was also obtained in company with
them.

Description of the Illustrations of this Species on Plate VI.

Fig. 7. Antedon prolixa : natural size.

8. Diagrammatic sketch of the radial plates.

9. Extremity of one of the dorsal cirri : magnified.

10. Joints with pinnules attached, from near the middle of a ray : magnified.

ARCTIC SEA TO THE WEST OF GREENLAND. 79

GENEEAL CONCLUSIONS EEGAED1NG DISTRIBUTION.

OP the Holothuroidea, Myriotrochus Einkii is the most polar in its distribution ; for
it has been found in Discovery Bay and to the south as far as Labrador ; and its
habitat to the east is in the seas of Spitzbergen to Novaya Zemlya &c. Chirodota
lasvis has also a wide Arctic distribution, but is found further to the south than
the last-mentioned : Godhavn, in Greenland, and Finmark and the Lofoten Islands
are its limits in one direction, and Sitcha and Ochotsk in another. Then Cucumaria
frondosa is distributed from Godhavn to Florida, from Spitzbergen to the British Isles,
and to San Francisco. Of the species of Psolus, one is found from Godhavn to
Maine and across to Iceland and the British Isles ; whilst the other extends far to the
east, but appears to be restricted to the polar waters. The most restricted species was
Orcula Eartliii, which was found at Holsteinborg and has been found off Labrador.

The only species of Echinoidea, Strongylocentrotus drobachiensis, has a vast
distribution, as far north as Discovery Bay, and south to Florida, from Iceland to
Spitzbergen, and Novaya Zemlya to the British Isles. It has been found in Behring's
Straits, Kamtschatka, and on the American coast to Vancouver. Fed upon by
Asteroids, and caring little for a bottom temperature of 29° Fahr., this species is very
typical of the collection under consideration. It is essentially a polar species, migra-
ting now and then to the south ; and it forms part of a true polar fauna.

The Asteracanthia have one species as far north as Discovery Bay ; and it
extends southward to Newfoundland and across to Novaya Zemlya; the other lives
from 70° N. lat. to Labrador, but it has a polar distribution, as it extends to the
Ochotsk sea. Stichaster albulus is found at Franklin-Pierce Bay to Godhavn and
Grand Manan ; the eastern habitat is from Iceland to Spitzbergen. Cribrella
oculata is distributed from 70° N. lat. to Massachusetts, Iceland, Spitzbergen,
Scandinavia, and White Sea, and is found in the English Channel. Moreover this
vast area is enlarged by the discovery of the form in the Sea of Ochotsk.

The new species of Pedicellaster is esentially a dweller in the Palaeocrystalline
sea of Nares.

As might be supposed, Crossaster papposus has a vast distribution. Nearly cir-
cumpolar, it extends along the European and eastern American coasts, along the

80 ECHINODEKMATA OF THE ARCTIC SEA.

north of the European and Asiatic continents, and is found in Behring's Straits.
Finisterre and Massachusetts are its most southern habitats.

Solaster endeca has been found in Smith's Sound, Newfoundland, Iceland, and
in the English Channel ; and Brandt obtained it at Sitcha. The solitary species of
Lophaster is very Arctic, being found at Discovery Bay ; and it extends south in the
Gulf of Maine and at Bergen.

Pteraster militaris has almost a corresponding, but slightly intermediate range ;
and the Ctenodiscus extends far over the north of the European continent, comes
only as far south as Maine in America, and reaches to lat. N. 70° 30' off Greenland.

It is evident that the grouping of these Asteroids is as if they were part of a
great Polar fauna, with no very great southern distribution.

Of the Ophiuroidea, the new species Ophiopleura arctica belongs to a genus
characteristically Arctic. The Ophioglyphm, with the exception of 0. Stuwitzii, are
found very far to the north ; and their fellow Ophiuroidea at Discovery Bay and
Franklin-Pierce Bay are Ophiocten sericeum, Amphiura Holbcelli, and Ophiacantha
spinulosa. The range of these resembles that of the Asteroids to a certain extent ;
and they have curious structural affinities with even the forms from the Korean
Sea. The forms with a smaller northern area are the Ophioglyphce just alluded to
and Ophiopholis bellis.

The Astrophyton ranges from 78° N. lat. up Smith's Sound to Cape Cod, and has
been found at Vadso, in Finmark.

Antedon Eschrichtii and A. celtica are widely distributed northern forms ; and the
new species Antedon prolixa is from Discovery Bay.

When these details are carefully considered, it becomes evident that each one of the
great groups of Echinodermata tells the same story regarding distribution. The fauna,
as a whole, is not an extension northwards of species from more temperate climates,
but is essentially circumpolar. The modifications in the character of the species are
probably due to variation produced by the changes of condition which necessarily occur
in different parts of any great area.

INDEX.

[Synonyma arc printed in italics.]

Pago

Alecto Eschrichtii 73

Amphiura 67

- HolbooUi 67

Anodiscw crispatus 49

Antedon celtica 75

Eschrichtii 73

prolixa 77

Ascidia rustica 9

Asteracanthion grcenlandicum 27

- Mulleri 27

ocliotense 23, 26

polare 23

rulens 23

Asteracanthion albulus 29

palceocrystallus 34

proHema 29

roseus 29

Asterias ajjinis 37, 39

alloverrucosa 37, 39

albula 29

aspera 40

auranciaca 49

crispata 49

endeca 40

grcenlandica 27

helianthemoides 36

militaris 46

minuta 23, 26

Miilleri 27

oculata 32

papposa 36

pertusa 32

- polaris 23, 49

rubens 23

sanguinolenta 32

seposita 32

Page

Asterias spongiosa 32, 33

violacea 23

Asteriscus militaris . . 46

Asteroidea 23

Astropecten corniculatus 49

polaris 49

AstrophytidaB 69

Astrophyton Agassizii 69

arcticum 70

Bothryodacti/la affinis 2

yrandis 2, 4

CJicetaster borealis 43

Chiridota brevis 15, 18

Chirodota discolor 12, 14

laevis 12

pellucida 13, 14

tigillum 13

typica 13

Cladodactyla pentactes 2

Comatula celtica 75

EschricJitii 73

Woodwardii 75

Cribrella Eschrichtii 32

oculata 32

sanguinolenta 32

G'rinoidea 73

Crossaster papposus 36

Ctenodiseus corniculatus 49

crispatus 49

polaris 49

pygmceus 49, 51

Cucumaria calcigera 5

frondosa 2

fucicola 2

Hyndmanni 5, 7

Korenil 5, 7

H

82

INDEX.

Page

Cucumaria minula 4

Cuaimaria lavis 12

pellucida 12

Cuvieria Fabridi 10

frondosa 2

phantapus 9

Dactylota leevis 12

minuta 2

pellucida 12

Dodecactis reticidata in dorso 36

Echinaster Eschrichtii 32, 34

oculatus 32

sanguinolentus 32

Sanii 32, 34

Echinoidea 19

Echinometra drobachiensis 19

Echinus Morocentrotus 19

Moroticus 19

drobachiensis 19

granularis 19

granulatus 19, 21

neylectus 19

saxatilis 19

subangularis 19

Euryale scutatum 69

Euryechinus drobachiensis 19

granulatus 20

Oorgonocephalus 71

arcticus 69

Henricia oculata 32

Holoihuria frondosa 2

grandis 2

Icevis 12

minuta 2

pellucida 12

pentactes 2

phantapus 9

squamata 10

Holothuroidea 1

Linckia oculata 32, 34

pertusa 32

Lophaster furcifer 43

Lophothuria Fabridi 10

LiitJcenia 54

Myriotrochus Einkii 15

Oligotrochus vitreus 17

Ophiacantha 68

Ophiacantha arctica 68

yrcenlandica 68

spinulosa 68

Ophiocten 65

Page

Ophiocten Krceyeri 65

sericeum 65

Ophioglypha 59

robusta 62

Sarsii 60

Stuwitzii 54

OpTiiolepis robusta 62

Ophiopholis bellis 54

Ophiopleura 54

arctica 55

borealis 58

Ophiura serkea 65

squaniosa. 62

Ophiuroidea 54

Orcula Barthii 8

Pedicellaster palseocrystallus 34

typicus 36

Pentacta calcigera 5

frondosa 2

Icevis 12

minuta 2

pellucida 12

Pentadactylosaster oculatus 32

Psolus Fabricii 10

Icevigatus 9

phantapus 9

Pteraster militaris 46

Solaster endeca 40

furcifer 43

papposa 37

Stellonia endeca 40

papposa 37

Stichaster albulus 29

• , var. nitida 30

Strongylocentrotus clilorocentrotus 19, 21

drobachiensis 19

Synapta coreacea 12

rotifera 12

Thyonidivm pellucidum 12

Toxopneustes carnosus 19

chlorocentrotus 19

drobachiensis 19

Dubenii 19

c/ranulatus 19

neglectus 19

pallidus 20

pictws 20

TrisJcaidecactis papposa 36

Trochinus pallidus 13

Uraster violacea 27

PLATE I.

HOLOTHUROIDEA.

Figure 1 . CUCUMARIA FRONDOSA (Gunn.), Forbes. Natural size.

2. A young specimen of the same, natural size.

3. CUCUMARIA CALCIGERA (Stimp.), Seleuka. Natural size.

4. Sketch of the mouth-ring and adjacent parts of a young individual, magnified.

5. Generative tubes at an early stage of growth, magnified.

6. Spicules of the superficial layer in situ, magnified.

7. Small spicule in profile, magnified.

8. Disk at the extremity of the sucker-foot, magnified.

9. PSOLUS FABRICII (D. & K.), Liitken. Dorsal aspect, natural size.

10. The same specimen. Ventral aspect, natural size.

11. A young specimen. Ventral aspect, natural size.

12. One of the small cup-like spicules of the ventral integument, magnified.

13. Spicules of the ventral integument, magnified.

14. CHIRODOTA L.EVIS (Fabr.), Grube. Natural size.

15. Generative tubes, magnified.

16. Portion of the mouth-ring, magnified.

17. The pyriform infundibular organs or " Wimpertrichter," magnified.

18. One of the wheel-like spicules, magnified.

19. One of the sacculi, to show the aggregation of the spicules, magnified.

20. MYRIOTROCHUS RINKII, Steenstrup. Natural size.

21. One of the spicules, magnified.

22. Sketch of the internal organs, showing the relative proportions of the alimentary canal,

the Polian vesicle, and the generative tubes.

23. Portion of the mouth-ring, magnified. (In this drawing the upward prolongations ought

to taper more rapidly, and the foramina ought to be further up from the base.)

24. Underside of the mouth-ring, with parts attached, seen from below, showing the dila-

tations of the oral water-vessel.

Plat, 1

4.

GRDeWJcU.lith WPercy Sladen.dir

Arrfio Holothuroideau.

Mintern Bros imp .

PLATE II.

ECHINOIDEA and ASTEROIDEA.

Figure 1. STRONGYLOCENTROTUS DROBACHIENSIS (O. F. M.), A. Ag. Abactinal aspect, natural size.

2. The same specimen, seen in profile, natural size.

3. Portion of test near the ambitus, showing the ambulacral and interambulacral areas,

magnified.

4. ASTERACANTHION POLARE, M. & T. Abactinal aspect, reduced one half.

5. Actinal aspect of the same specimen, reduced one half.

6. Portion near the middle of a ray, actinal aspect, magnified.

7. Portion near the middle of a ray, abactinal aspect, magnified.

8. A similar portion from the more compactly spinulated variety of this species, magnified.

9. ASTERACANTHION GR<ENLANDICUM (Steenstrup), Lutken. Abactinal aspect, natural size.

10. Actinal aspect of the same specimen, natural size.

11. Portion near the middle of a ray, actinal aspect, magnified.

12. Portion near the middle of a ray, abactinal aspect, magnified.

13. STICHASTER ALBULUS (Stimps.), Verrill. Abactinal aspect, natural size.

14. A larger and almost equal-rayed variety from Proven.

15. Actinal aspect of the same specimen, natural size.

16. Portion near the middle of a ray, actinal aspect, magnified.

17. Portion near the middle of a ray, abactinal aspect, magnified.

18. CRIBRELLA OCULATA (Linck), Forbes. Abactinal aspect, natural size.

19. Actinal aspect of the same specimen, natural size.

20. Portion near the middle of a ray, actinal aspect, magnified.

21. Portion near the middle of a ray, abactinal aspect, magnified.

22. PEDICELLASTER PAL^EOCRYSTALLUS, Sladen. Abactinal aspect, natural size.

23. Actinal aspect of the same specimen, natural size.

24. Portion near the middle of a ray, actinal aspect, magnified.

25. Portion near the middle of a ray, abactinal aspect, magnified.

26. One of the pedicellarise, magnified.

Plate Jl

% * . V »

;*:

w. V .'; •> •

•IH

ASFoord.lith. W.Pepey Sladen,dir.

Mm tern Bros . imp.

& Asteroidea.' and Eckuwidea.*.

PLATE III.

ASTEROIDEA.

Figure 1. CROSSASTER PAPPOSUS (Linck), M. & T. Abactinal aspect. Young specimen.

2. Actinal aspect of the same specimen, natural size.

3. Portion near the middle of a ray, actinal aspect, magnified.

4. Portion near the middle of a ray, abactinal aspect, magnified.

5. SOLASTER ENDECA (Gmel.), Forbes. Abactinal aspect, natural size.

6. Actinal aspect of the same specimen, natural size.

7. Portion near the middle of a ray, actinal aspect, magnified.

8. Portion near the middle of a ray, abactinal aspect, magnified.

9. LOPHASTER FURCIFER (D. & K.), Verrill. Abactinal aspect, natural size.

10. Actinal aspect of the same specimen, natural size.

1 1 . Portion near the middle of a ray, actinal aspect, magnified.

12. Portion near the middle of a ray, abactinal aspect, magnified.

13. PTERASTER MILITARIS (O. F. M.), M. & T. Abactinal aspect, natural size.

14. Actinal aspect of the same specimen, natural size.

15. Portion near the middle of a ray, actinal aspect, magnified.

16. Portion near the middle of a ray, abactinal aspect, magnified.

17. CTENODISCUS CORNICULATUS (Linck), Perrier. Abactinal aspect, natural size.

18. Actinal aspect of the same specimen, natural size.

19. Portion near the middle of a ray, actinal aspect, magnified.

20. Portion near the middle of a ray, abactinal aspect, magnified.

ri.,i,- in

A3 Foord.lith. W P»rcy Sl»den,dir-

Arctic Asten'idfau

PLATE IV.

OPHIUROIDEA.

Figure 1. OPHIOPLEURA ARCTICA, Duncan. The upper part of the disk and arms, natural size.

2. The under surface of the disk, magnified.

2 a. The teeth on the jaw-plate, oblique view, magnified.
2 b. The teeth, from above, magnified.
2 c. The teeth, approaching, magnified.

3. OPHIOGLYPHA SARSII (Liitken). From above, natural size.

4. The underpart of the disk, magnified.

5. OPHIOGLYPHA ROBUSTA (Ayres). From above, natural size.

6. The underpart of the disk and an arm, magnified.

7. The upper surface of the disk, magnified.

8. OPHIOCTEN SERICEUM (Forbes). The upper surface, slightly magnified.

9. The scaling of the upper part of the disk and part of an arm, magnified.

10. Part of the underpart of the disk, magnified.

11. OPHIACANTHA SPINULOSA. The upper surface, natural size.

12. Part of the under surface, magnified.

13. The spines on the disk, magnified.

14. OPHIOCTEN SERICEUM (Forbes). Part of the disk-scaling, magnified.

15. AMPHIURA HOLB(ELLI. Part of the underpart of the disk, magnified.

16. Radial shields, magnified.

17. Side view of the arm, magnified.

Plate IV.

A S Foord.del etlith.P.M Ihincan dir.

Arctic

Mintem Br- c

PLATE V.

ASTKOPHYTID.2E.

Figure 1. ASTROPHYTON AGASSIZII, Stimpson. Abactinal aspect, reduced about one third.

2. One mouth-angle of a much smaller specimen, magnified.

3. Portion of a ray, abactinal aspect, magnified.

4. Portion of a ray below the first forking, actinal aspect, magnified.

5. Portion of a ray, side view, magnified.

6. Portion of a ray near the extremity, magnified.
6 a. One of the hooklets, magnified.

5.

AS Foord.litli. W Percy

Arctic* Astrophytorv.

Mintern Bros i

PLATE VI.
CRINOIDEA.

Figure 1. ANTEDON ESCHRICHTII (Miiller), Verrill. Natural size.

2. Diagrammatic sketch of the radial plates.

3. Extremity of one of the dorsal cirri, magnified.

4. A first pinnule, magnified.

5. ANTEDON CELTICA (Barrett), Normau. Natural size.

6. Diagrammatic sketch of the radial plates.

7. ANTEDON PROLIXA, Sladen. Natural size.

8. Diagrammatic sketch of the radial plates.

9. Extremity of one of the dorsal cirri, magnified.

10. Joints with pinnules attached, from near the middle of a ray, magnified.

Pisa* VI.

. ;ff

I I

'mm

• "crd,lltHrW Percy Sladen, dir.

Arctic Comatida..

Miniorn Bros ;n.';

J

USTJUL 15 193ft

g

u

fi!

I

o

SI "81

OD

hi
0)i
-P!

Q)i

i

®i

o

University of Toronto
Library

DO NOT

REMOVE

THE

CARD

FROM

THIS

POCKET

Acme Library Card Pocket

Under Pat. "Ref. Index File"
Made by LIBRARY BUREAU

[ a •--