^1 -m^ym

'. "^ .' VT''^ ■ . '. ■

■•^ ':"m

^^

-V.
'■••■*

,,flBl

fr^-^^^h

'\^'

-*-.

fli

'.•'»:.-.•■ ^f•.'^.■''

^1^"

1 ■• -'■■• V ;- " • ■

-f^H;'

• '-"* ' ^

■■, *■■*-,■■

V*^^ '

** '. I

,••■ J,- .':; '•'-. .

'-■ 4- , ,

;'%-^ ,,■;

<' > f

..-'■ ' '^:■,

.■.I .

•i" ■■■ ■>•'

■A.: %■■

. .■ .•>'., ■ <;■*/< \

'■-!' -T^iFiijVV

■• •^:.t,'r«.<";

..J/,

THE

VOYAGE OF H.M.S. CHALLENGER.

ZOOLOGY-VOL. XI,

^

REPORT c,

ON THE

SCIENTIFIC RESULTS

OF THE

VOYAGE OF H.M.S. CHALLENGER

DURING THE YEARS 187 3-7 6

UNDER THE COMMAND OF

Captain GEORGE S. NARES, R.N., F.R.S.

AND THE LATE

Captain FRANK TOURLE THOMSON, R.N.

PREPARED UNDER THE SUPERINTENDENCE OF

THE LATE

Sir C. WYVILLE THOMSON, Knt., F.R.S., &c.

REGIUS PROFESSOR OF NATURAL HISTORY IN THE UNIVERSITY OF EDINBURGH
DIRECTOR OF THE CIVILIAN SCIENTIFIC STAFF ON BOARD

AND NOW OF

JOHN MURRAY

ONE OF THE NATURALISTS OF THE EXPEDITION

Zoology— Vol. XL

Publisijeti bp <!5rDer of lt)er JWajestp's ©oliernment

PRINTED FOR HER MAJESTY'S STATIONERY OFFICE

AND SOLD BY

LONDON :— LONGMANS & CO.; JOHN MURRAY; MACMILLAN & CO.; SIMPKIN, MARSHALL, & CO.
TRUBNER & CO. ; E. STANFORD ; J. D. POTTER ; and KEGAN PAUL, TRENCH, & CO.

EDINBURGH :— ADAM & CHARLES BLACK and DOUGLAS & FOULIS. /\^

DUBLIN :— A. THOM & CO. and HODGES, FIGGIS, & CO. /

1884

Price Fifty Shillings.

PRINTED BY NEILL AND COMPANY, EDINBURGH,

FOR HER MAJESTY'S STATIONERY OFFICE.

CONTENTS.

I. — Report on the Keratosa collected by H.M.S. Challenger, during the years

1873-1876.

By N. PoLEJAEFF, M.A., of the University of Odessa.

(Recewed 16th April 1884.)

II. — Report on the Crinoidea collected during the Voyage of H.M.S. Challenger,
during the years 1873-1876. — The Stalked Crinoids.

By P. Herbert Carpenter, D.Sc, Assistant Master at Eton College.

{First instalment of Manuscript received 18th March 1884 ; the last 13th December 1884.)

III. — Report on the Isopoda collected by H.M.S. Challenger, during the years

1873-1876.— The Genus Serolis.

By Frank Evers Beddard, M.A. (Oxon), F.R.S.E., F.R.M.S., F.Z.S., M.B.O.U.,
Prosector to the Zoological Society of London.

(Received 6th September 1884.)

78083

EDITOKIAL NOTES.

This volume contains Parts XXXL, XXXII., and XXXIII. of the

Zoological Series of Reports on the Scientific Results of the Expedition.

Part XXXI. — The Report on the Calcarea, by Dr. N. Polejaeff, was
published at the end of last year as Part XXIV. of the Zoological Series of
Rej)orts, and the present Report on the Keratosa is by the same author, and
will be welcomed by all Spongiologists.

The Manuscript was received on 16th April 1884.

Part XXXII. — The Stalked Crinoids, both on account of their rarity and
their palasontological relations, are perhaps the most interesting and remarkable
of deep-sea animals, and have been in a special manner associated with the
Challenger Expedition.

The joint work of the late Sir C. Wyville Thomson and Dr. W. B.
Carpenter, C.B., first on Comatula and afterwards on Pentacrmiis, together with
the discovery by Professor G. 0. Sars of Rhizocrinus off the Lofoten Islands in
1864, led directly to the expeditions of the "Lightning" and "Porcupine"
in 1868 and the following years, and was thus indirectly concerned in the
despatch of the Challenger Expedition in 1872. Sir Wyville Thomson
himself proposed to draw up the Report on this group of animals, to which he
had devoted much attention during the cruise. The circumstances which
led to Dr. P. Herbert Carpenter undertaking the preparation of the Report
after Sir Wyville Thomson's death in 1882 are referred to in detail in the
Preface to the Report.

viii THE VOYAGE OF H.M.S. CHALLENGER.

The Manuscript was received by me in batches between 18th March 1884
and 13th December 1884.

Part XXXIII. — This Monograph of the genus Serolis is the first instal-
ment of a Report on the Isopoda collected during the Expedition, by Frank
Evers Beddard, Esq., M.A. (Oxon), Prosector to the Zoological Society of
London.

Mr. Beddard was for nearly two years associated with me in the work of
this office, and I embrace this opportunity of acknowledging the very valuable
assistance he rendered in connection with the translation and editing of many
of the Reports.

The Manuscript was received on 6th September 1884.

John Murray.

Challenger Office, 32 Queen Street,.
Edinburgh, .30/7; December 1884.

THE

VOYAGE OF H.M.S. CHALLENGER

ZOOLOGY.

REPORT on the Keratosa collected by H.M.S. Challenger during the years
1873-76. By N. Polejaeff, M.A. of the University of Odessa.

INTRODUCTION.

The Keratose Sponges do not belong to the deep-sea fauna ; it is therefore not
surprising that the number of forms brought home by tlie Challenger Expedition does
not exceed the comparatively small number of thirty-seven (twenty-one new, three
undeterminable). But if not numerous, the collection is still very interesting, embracing
as it does almost all the genera of the horny sponges hitherto distinguished, and most of
the specimens being in a very good state of preservation.

Entering upon their classification and description, I feel myself in a position rather
different from that which I occupied some months ago when writing about the Challenger
Calcarea. For iu this latter case I had to deal with but one elaborate system (that of
Haeckel), and my task consisted simply in the reconciliation of Haeckel's systematic
arrangement with the modern state of spongiological knowledge. In the group
of Keratose Sponges the classifier meets with many detailed systems, constructed in many
instances upon radically different principles. My first task is consequently to show
which of the existing systems is to be most recommended, and with what modifications.
There are two ways of doing this. One way might be called " historical," and would
consist of a critical discussion in chronological order of all the systems of the horny
sponges we possess ; the second might be called " morphological," and would consist of
a critical discussion of the comparative systematic value of their different organs. I
purpose choosing this latter mode, for the following reasons. Firstly, because it ensures the
avoidance of superfluous repetitions, the existing arrangements of the Keratosa being con-

(ZOOL. OHALL. EXP. — PART XXXI. — 1884.) Hh 1

2 THE VOYAGE OF H.M.S. CHALLENGEE.

flicting in some points and harmonious in others. Secondly, because it promises more
logically fitting opportunities, whilst discussing the discoveries of other spongiologists,
of communicatincf whatever observations of general interest I have made whilst examin-
ing the Challenger material, so that the descriptive part of this memoir may be
for the most part exclusively devoted to systematic purposes. The following pages
accordingly consist of a discussion of the organisation of the Keratosa and of the
principles of their classification.

L-ORGANISATION AND CLASSIFICATION OF THE KERATOSA.

One might perhaps feel inclined to say that this title promises but very little ; that a
classifier has to search for systematic characters not only into the organisation of the
animals in question, viz., into their Anatomy and Histology, but also into other regions
of Biology, and, in the first instance, into Embryology and Palaeontology. Unluckily
this is impossible so far as the horny sponges are concerned. Some fossils have been
described which may possibly be referred to the Keratosa, but this cannot be regarded
as scientifically proved, nor is the number of such forms sufficieut to permit any further
conclusions.^ The possibility of a future application of Palaeontology to phylogenetic
purposes respecting the Keratose Sponges is not entirely excluded, though there are
reasons to believe that this group is a very recent one, but up to the present time the
application above mentioned is impossible. Again, with respect to embryological data
even such a modest hope cannot be assumed. Of course our knowledge is stUl very
fragmentary, but what we know only confirms the opinion that the ontogeny of the
horny sponges is very monotonous, and that therefore its further profound study would
probably be of consequence only for the solution of certain embryological problems (in
the strict sense of the word), but not of much service in augmenting the number of
systematically important characters. The classifier is thus thrown on the resources of
Anatomy and Histology alone ; chiefly on those of Anatomy, since it is only in exceptional
cases, as in lanthella or Cacospongia vesicidifera,^ that histological characters can be
applied to systematic purposes. But, nevertheless, this would be of no further consequence
were the anatomical characters of, so to speak, unconditional value. Yet even this is not
the case, and this is just what renders the classification of the Keratosa so very difiicult,
and makes the danger of " describing individuals instead of genera and species " (0.
Schmidt) greater in this group than elsewhere. For Comparative Anatomy can only state
this or that difi"ereuce in organisation, but is very often quite powerless, at least in the
Keratosa, to decide the question whether this or that anatomical peculiarity is constant or
merely accidental. It is therefore obvious that the systems of the Keratosa we are now
so diligently elaborating will prove, with the progress of the Comparative Physiology, to

1 Zittel, Zur Stammesgeschichte der Spougien, Mimchen, 1878, p. 9. - Page 58 of this Report.

4 THE VOYAGE OF H.M.S. CHALLENGER.

be in many points quite artificial ; but we must console ourselves with the impossibility
of altering the matter, and although wdth respect to this group we have to deal with
anatomical cho,racters almost exclusively, we must ground upon them our systematic
arrangements, proceeding, however, with all possible prudence, and bearing in mind the
necessity of a critical attitude towards our own conclusions.

These general remarks will now be followed by more special observations as to the
systematic value of various characters of the sponges in question. I begin with the
properties of the skeleton. Its high systematic significance has always been recognised ;
what is more, it has been exaggerated. The systems of Duchassaing de Fonbressin and
Michelotti,^ of Gray,^ Hyatt,^ Carter,* are founded simply on its properties. In the diag-
noses of the last-named naturalist, indeed, the " sarcode " is also very often spoken of ; but
these particulars might in almost all cases be omitted ; and it is precisely owing to the
circumstance that the former spongiologists were inclined to pay to the properties of the
skeleton an exclusive attention, that its modifications for a long time past have been
submitted to a careful study, and considerable difi'erences in its structure discovered.
It has been stated that while in some of the Keratose Sponges the horny fibres show no
differentiation in their central and peripheral parts, the fibres of the skeleton of many
others admit of a distinction into a central pith-substance (" Marksubstanz " of German
authors) and of a horny laminar envelope ; and while the homogeneous fibres are almost
always more or less cored with foreign bodies, so that the horny substance shows in
many instances a tendency to disappear entirely, the heterogeneous skeletal fibres, on the
contrary, are in most cases quite free from any foreign enclosures. Finally, it has been
stated that in certain sponges [lanthella) the horny envelope of their skeletal fibres is
charged with true ceUs (Flemming,® Carter ''). To the first of these characters the greatest
systematic significance has been repeatedly ascribed, and the two spongiologists to whom
we owe the most elaborate systems of Keratosa (Carter and Hyatt) have made use of it
in order to subdivide the grouji into two main divisions (Ceratina and Psammonemata,
Carter ; Aplysinte and Sponginse, Hyatt), which proceeding deserves a more detailed
critical study, since we have recently learned from the spongiological writings of F. E.
Schulze that the above-mentioned differences in the properties of the skeleton present a
certain antagonism with regard to the internal structure of the soft parts, that both in
Ceratina or Aplysinse and Psammonemata or Sponginse two types, or at least modifica-
tions, of the canal system, are to be seen. Schulze ascertained, in fact, that while an
Aplysina, and on the other hand a Euspoiigia or Cacospongia, are characterised, in
the organisation of their canal system, by comparatively small, round, or pear-shaped

1 Spongiaires de la mer Caraibe, Harlem 1864. ^ Ptoc. Zool. Soc. Lond., 1867, pp. 503, 508.

• Revisiou of the North American Porifers, Boston Soc. Nat. Hist., 1875 and 1877.

* Ann. and Mag. Nat. Hist., ser. 4, vol. xvi. pp. 132, 134-140, 1875.

6 WUrzhurger Verhandl., N. F., Bd. ii. • Ann. and Mag. Nat. Hist., ser. 5, voL viii. p. 112, 1881.

REPORT ON THE KERATOSA. 5

flagellated chambers, each possessing its own narrow inhalent and exhalent canaliculi,
and while the ground-mass surrounding these flagellated chambers is always opaque
owing to the presence of small granules, the forms like Spongelia and Aijlysilla possess
no special cameral canaliculi, their large pouch-shaped flagellated chambers receiving the
water from the subdermal cavities directly by means of the pores in their walls, and
expelling it also immediately, without the help of any intermediate narrow canals,
into large exhalent cavities, the diameter of these latter being usually far larger than
that of the exhalent opening of the corresponding flagellated chambers ; and that in
these latter instances the parenchyma in the zone of the flagellated chambers is devoid
of any granules, being lucid and transparent.

It remains to be decided what position the modern systematist should take with

respect to the antagonism in question. F. E. Schulze himself gives in his papers

no answer to it. He assigns^ to Aplysilla the place of a new genus in his family

Aplysinidse, while he is inclined to regard the genus Spongelia, an analogue of

Aplysilla in the Keratosa with homogeneous skeletal fibres, as the representative

of an independent family.^ Again, his distinguished pupil Dr. Vosmaer, stands also

perfectly neutral, dividing, in his interesting paper on Velinea gracilis,^ all the

Keratosa directly into families, according to the properties both of the soft parts and of

the skeleton. But though neutral as far as his actions are concerned. Dr. Vosmaer

expresses very clearly his opinion on the matter. He does not ascribe any special

importance to the diff"erence between the fibres, whether homogeneous or heterogeneous.

On the contrary, another pupil of Prof. Schulze, Dr. v. Lendenfeld, seems inclined to

accept the opinions of Hyatt and Carter. At least, in his memoir on the Aplysinidae of

the South Sea, he characterises^ the family of Apl3'sinid£B by their heterogeneous

skeletal fibres, splitting it into two subfamilies, that of Aplysininae and Aplysillinae,

according to the type of the canal system. And, indeed, such a proceeding appears at

first sight very logical and natural. Of course, on the whole, the canal system is of

greater significance for the sponge organism than ■ the skeleton. No sponge can be

imagined without canal system, be it represented as in Asconidaj by the undifferentiated

central cavity without any trace of separate flagellated chambers, or, as in Ap>lysina or

Corticium, by a very complicated system of subdermal cavities, inhalent cameral

canaliculi, flagellated chambers, exhalent canaliculi, &c., while there are sponges, like

Halisarca, Oscarella, or Chondrosia without any supporting apparatus. It should

not, however, be overlooked that whatever importance may be ascribed to the canal

system, this importance is of a pronounced physiological character. On the contrary,

so far as the properties of the skeleton are concerned, — all this holds true within

the group Keratosa, and the last-mentioned point with regard to its internal

1 Zeitschr.f. iviss. Zool., Bd. xxx. p. 404. ° Zeitschr.f. wiss. Zool., Bd. xxxii. p. 117.

' Mittheilungen a. d. zool. Station zu Neapel, 1883, p. 444. * Zeitschr.f. loiss. Zool, Bd. xxxviii. p. 309.

6 THE VOYAGE OF H.M.S. CHALLENGER.

structure, — they appear to be of a more trustworthy morphological character. The function
of the skeleton is to support the soft parts. The heterogeneous and homogeneous skeletal
fibres are equally fit for this function ; and when once a sponge has adopted the
heterogeneous fibres, it would but conserve them. One might logically compare the
heteroeeneous and homogeneous skeletal fibres with calcareous and siliceous skeletal
spicules, these latter propping the sponge-sarcode equally well in both cases ; and since
it is necessary to separate systematically the Calcarea and Silicea, it must be equally
necessary to separate the Ceratina and the Psammonemata.

To sum up, the procedure of the above-mentioned systematists may be regarded as
very logical. Yet each question admits of numerous answers equally logical and
undeniable ; and it is not to be forgotten that what we think to-day to be thoroughly
logical, we may perhaps regard to-morrow as quite impossible, that the really logical is
that which alone corresponds with the reality. The ascribing of such systematic
importance to the properties of the skeleton in question cannot be reconciled with the
reality, and in no case can these properties serve as the basis for the primary subdivision
of the group Keratosa. Such a subdivision would express that the Keratosa with
homogeneous skeletal fibres form one phylogenetic branch of the group, the Keratosa
with heterogeneous fibres another, phylogenetictilly equivalent to the first, the
intermediate connecting links having died out. This, however, is not the case, the
differences in the structure of horny fibres, as already pointed out by Vosmaer,
being but of a quantitative nature. There are amongst the Keratosa, forms, the horny
envelope of whose heterogeneous skeletal fibres is very thin {Apliisina, Aijlysilla,
Darwinella), so that the pith-substance forms the main part of the fibre ; there are
again other forms ( Verongia, lAiffaria), the horny walls of whose fibres are far thicker,
and, at least in some representatives of the genus Luffaria, there are to be found
amongst fine fibres, fibres quite similar to those of Euspongia, or Spongelia, viz., fibres
apparently entirely devoid of any central diff"erentiation (PL IX. fig. 6). I say
apparently, for, thanks to F. E. Schulze, we know that each normal horny fibre does
possess what he calls an " Achsenstrang." He has been able to discern it in the fibres of
Euspongia ^ and Spongeliaj'^ and that certain homogeneous fibres show no differentiation,
even under high microscopic power, seems to be due to the fact that, hand in hand
with a more voluminous development of the pith-substance, there is a variation
in its chemical and optical properties. In the fibre of an Aplysinu or Aplysilla the core
and the surrounding horny laminae can be readily distinguished under very low
magnifying power, and the designation " hollow-fibred," which has been adopted by
many spongiologists with respect to forms like Aplysilla or Verongia, owes its
existence to the fact that even when simply dried out, the skeletal fibres of the above
forms show no more trace of the pith-substance.

1 Zeitschr. f. loiss. Zool, Bd. xxxii. p. 633. ^ Ibid., pi. \i. figs. 6 and 7.

EEPORT ON THE KERATOSA. 7

On tlie other hand, in Luffaria the horny laminae and the central differentiation of
its fibres resemble each other optically in a far greater degree ; higher powers of the
microscope are necessary in order to show this internal differentiation ; when dried, the
fibres stUl possess their core, and only after treatment with caustic alkali or ammonia do
the fibres become hollow. Certain differences are also to be found in the structure of the
pith-substance. While in Aplysina or Verongia the core is represented by a fine and
irregularly twined network,^ in Luffaria the structure of the core of its fibres appears
similar to that of its horny envelope, being however represented not by continuous
laminae, but by minute horny splints, still disposed parallel to one another, as well as
with regard to the surrounding lamina, just as Schulze has figured ^ the central canal
of the fibres of Spongelia pallescens. But though the core of the skeletal fibres of Luff-
ana appears to be structurally quite equivalent to that of Euspongia, and differs con-
siderably from that of typical Aplysinidae, it would still be premature to assume that
Luffaria is but a specifically modified Spongid, and not a link connecting the Aply-
sinidse with the Spongid^, in consequence of the identical manner in which both kinds
of fibres, the heterogeneous as well as homogeneous, develop.

It has indeed been stated — I allude to Dr. v. Lendenfeld's ^ observations — that the
development of the heterogeneous fibres only slightly resembles that of the homogeneous
ones ; that while, according to Schulze,^ the skeletal fibres of a Cacospongia or EusjJongia
grow by reason of the activity of spongohlasts exclusively, the growth of a hetero-
geneous fibre is dependent on the function both of spongoblasts and spongohlasts.
Like F. E. Schulze, Dr. v. Lendenfeld distinguishes two kinds of spongoblasts — those of
elongated, and those of polygonally massive, form ; the first are to be found along the
developing fibre, the second on its summit. He thinks, however, that the function of the
last-named is not to secrete the pith-substance of the central canal, but to sink down
into the interior of the developing fibre, in order to transform into pith-substance the
original horny mass, secreted by the elongated spongoblasts (Ich nehme an, dass die
ZeUen in den Kuppeln, gleich den Osteoklasten der Wirbelthiere, die harte Einde der
Skelettheile auflosen und in Marksubstanz verwandeln). This statement Dr. v.
Lendenfeld accompanies by an illustration, and recalls on this occasion the statements
of Flemming as to the structure of the skeletal fibres of lanthella, which he (Dr.
Lendenfeld) supposes to be very nearly allied to his Dendrilla, and in whose fibres the
presence of true cells has been proved. Of course there can be no doubt that the
skeletal fibres of lanthella are charged with true cells. The statements of Flemming have
been corroborated by Carter, and, for my own part, I can only confirm their observations
Yet these cells have been found not in the pith-substance, but between the surrounding

' Comp. F.E.,Scliulze's statements on tliis point in Zeitschr. f. wiss. Zool, Bd. xxx. p. 401, and my drawing, PI. X. fig. 3.
" Zeitschr. f. ^mss. Zool., vol. xxxii., pi. vi. iig. 6 ; comp. my drawing, PI. IX. fig. 6.
3 Zeitschr. f. wiss. Zool., vol. xxxviii. p. 291. * Ibid., vol. xxxii. p 635.

8 THE VOYAGE OF H.M.S, CHALLENGER.

horny lamina, and Flemming/ like Carter,^ emphatically denies their occurrence in the
core of the fibres. My own endeavours to discern them there have been equall}^
unsuccessful, and this not only vnih regard to lanthella, but also with regard to Verongia,
Aplysina, Darwinelkt, and Aplysilla, in which connection I should lay stress on the fact
that the specimens of Ajylysilla sulpharea, kindly placed at my disposal by Prof. F. E.
Schulze, have been examined both in the living state as well as preserved by the methods
recommended by Dr. v. Lendenfeld. In no single case have I been able to discern in
the central canal of the horny fibres anything that might be regarded as spongoklasts.

Again, Dr. Vosmaer states the same with respect to his Velinea gracilis.^ And far
from supposing that Dr. v. Lendenfeld has described what he did not see, I feel
convinced that he has misinterpreted what he did see. It may be said that what
does not occur in lanthella or Aplysilla may be characteristic of Dendrilla. Such a
possibility indeed is not excluded, but it is extremely improbable. For, firstly, Dendrilla
seems to be so closely allied to Aplysilla that both these genera wiU probably in time
be united into one ; and secondly, because with an instrument like the microscope one sees
very often precisely what one desu-es to see, and that Dr. v. Lendenfeld has been desirous
to find out his spongoklasts is beyond any doubt, the spongoklasts having been for him
a logical necessity. He refers in his above-mentioned memoir to the statements of F. E.
Schulze as to the fact that, though on the whole the old larger skeletal fibres of Aplysina
aerophoba must be called thick-walled, the young ones of small diameter on the contrary
thin-walled, still the diameter of the central pith-substance in larger fibres is comparatively
greater than that of the small ones. He tells us further that the same can also be said
with respect to the Aplysillidfe which he had for examination, and he deduces from this
the conclusion that hand in hand with the growth of the fibre its pith-substance increases
also.* If all this be so, indeed, without the theory of spongoklasts, the phenomenon would
be quite inconceivable. I must, however, deny the reliability of the statements in question.
The fibres of the representatives of the genera lanthella,^ Verongia, and Luffaria have
been submitted by me to the most careful examination, and I must state that in aU the
above specimens I found very often fibres of the same size but with difi"erent diameters of
the central canal ; and since I find also that this is the case not only as regards the fuUy
developed fibres but also those in embryonic condition, I believe that the phenomenon we
are speaking of is easily explained without any reference to spongoklasts, if we assume that
the pith-substance is a product of the p>olygonal spongoblasts, and the laminar horny sub-
stance the product of those of elongated shape, and that the differences in diameter of the cen-
tral canals are dependent upon their having been deposited broad or narrow. The second

1 Loc. cif., p. 4. - Loc. cit., p. 115.

3 Loc. cit., p. 441. ■• Loc. cit, p. 290-293.

5 I should call attention to the fibres constituting the skeleton of the stem of this sponge ; some of them are far
thicker than the piimary fibres of its leaf-like part, but, nevertheless, with the diameter of the central canal not only
considerably smaller than that of the primary fibres just mentioned, but occasionally not larger than that of a Spongeiia.

REPORT ON THE KERATOSA. 9

conclusion, as stated before, is founded on actual facts ; as to the first, I have no decisive
proofs for it, but I believe it will yet be adopted, at least provisionally, as a hypothesis
of comparatively great probability. That it is so, when compared with that of Dr. v.
Lendenfeld, is evident, since it does not, as does his theory, contradict the actual facts ;
that, again, the intussusception theory of Prof. Schulze ^ is not plausible has been indicated
by Dr. v. Lendenfeld ^ himself.

The foregoing remarks have had two distinct aims : first, to give a plausible explana-
tion of the phenomenon of the formation of the horny fibres ; and second, to show that
this phenomenon is the same with regard both to the homogeneous and heterogeneous
fibres. Whether I have succeeded in my first task will be shown by later investigations ;
at any rate we must assume that the elements forming the homogeneous horny skeletal
fibres are just the same as those secreting the heterogeneous fibres ; and since in both
cases no special functional transformations of any of these elements take place, we must
come to the conclusion that, in thorough harmony with the fact that homogeneous and
heterogeneous horny fibres difier from one another only quantitatively, the development
of both kinds of fibres admits also of only relative distinction ; and that, accordingly, the
subdivision of the Keratosa into two groups, the one characterised by homogeneous, the
other by heterogeneous, skeletal fibres, would be thoroughly artificial. Whether such a
subdivision may be made according to the structure of the canal system will be discussed
later on. I proceed for the present to treat of the modifications of the skeleton, and now
pass on to those influenced by the tendency of most of the Keratosa to take up foreign
bodies into their skeletal fibres.

As is well known, this tendency is characteristic only of Keratosa with homogeneous
skeletal fibres. Beginning with forms like most Coscinodermata and Hipjjospongioe,
whose fibres contain foreign bodies only exceptionally, here and there a sand-grain or frag-
ment of a spicule, going on to forms like many representatives of the genera Euspongia
and Ccicospongia, whose primary fibres are full of foreign enclosures, but the secondary
ones in most cases quite free from them, and, further, passing by forms like Psa'nwiocle7na
vosmaeri or Spongelia avara, both kinds of fibres of which are overcharged with foreign
enclosures, we come to the genus Psammopemma, characterised by an entire absence of any
fibres, the supporting skeleton consisting of sand-grains, portions of Foraminiferal shells,
fragments of spicules, &c., all lying separately, the secretion of the horny substance being
reduced to the formation of a thin horny envelope around each foreign body.

To this tendency, again, a high systematic importance has been ascribed. Gray ^ and
Marshall * characterise their family of Dysideidse mainly by the richness of their fibres
in foreign enclosures. The systematic application of this character plays also a great

1 Zeitschr. f. iviss. Zool., vol. xxx. p. 403. ^ Loc. cit, p. 291-292.

2 Froc. Zool. Soc. Land., 1867, p. 503. ■• Zeitschr. f. wiss. Zool., vol. xxxv. p. 92.
(zooL. CHALL. EXP. — PART X.XXI. — 1884.) Hh 2

10 THE VOYAGE OE H.M.S. CHAIXENGEE.

part ill the system of Mr. Hyatt. ^ Again, Mr. Carter makes an exclusive use of it as to
tlie special subdivisions of his order of Psammonemata, following the principle of
" beginning with horny fibre sparingly cored with foreign bodies, in order to go to that in
which the core is more general, and finally to end with that in which the horny element
is scarcely visible, and the core of foreign bodies only held together by a minimum of
sarcode, like the spicules in the Holorhaphidota."'" That, as a matter of fact, all the
naturalists in question have been wrong in this proceeding is clear to every one who is at
all acquainted with the recent progress of spougiology ; but as to Mr. Hyatt, I must still
add that logically he has had the best grounds for the division of the genus Dysidea
(Sjjongelia) into two independent families. His dermal-membrane theory of the formation
of skeletal fibres is false ; the dermal membrane, as we know now, stands in no connection
with this formation. Furthermore, it is improbable even theoretically, and indeed more
diflicult to understand than the phenomenon itself, but having once adopted the idea
that in different sponges the secondary fibres are of quite difierent origin (those of his
Sjwngelia owing their formation to his " mesoderm," those of his Dysidea to his
"ectoderm"), he was certainly right in ascribing to this difi"erence the significance of a
family character. For this character would be an absolute character, while Carter and
Marshall have been sure of the contrary. Of course, the proceeding of Mr. Carter is
still comprehensible, since his system was devised before the important investigations
of F. E. Schulze were published, but the proceeding of Dr. Marshall is to me quite incon-
ceivable. He makes use of a quantitative distinction in order to characterise a family.^
I am very well aware that the systematic definitions we give to the species, genera,
and accordingly to the families, particularly when young groujjs of animals are con-
cerned, must be according to circumstances more or less conditional. But this is the
privilege of natural arrangements. Is that of Dr. Marshall's Dysideidas a natural one ?
Surely not. Among his Dysideidje we find sponges with quite different internal organis-
ation. We find * there Oligoceras coUectrix, F. E. Schulze, a sponge whose canal system
follows the type of that of Spongidse; we fiud"^ there some representatives of the genus
Dysidea, whose canal system presents, according to Marshall, quite difierent characters —
those of a vesicular type; we find" there also sponges with a canal system arrantrcd
according to the so-called dendroid type, which has no more real existence than the
vesicular type; finally, we find^ there sponges whose canal system could not have been
made out, the specimens having been very badly preserved. Dr. Marshall calls F. E.
Schulze the most eminent spongiologist of the present time; he calls his sj)ongiological
papers briUiant; but the chief merit of F. E. Schulze consists precisely in having made

1 Revision, &c., jiart ii. p. 482. - Ann. and Mag. Nat. Hist., ser 4, vol. xvi. p. 135, 1875.

2 " Die Dysideiden siud Homscliwdinme bei deuen die auch alien iibrigen Horuschwammen in bolierem oder werin-
gercm Masse innewohneude Fahigkeit das eigene Skelett dm-ch aufgenommene Frerndkorper zu verstarkeu, den
hochsten Grad erreicht hat." — Loc. cit., p. 92.

■• Loc. cit., p. 92. ' Loc. cit, p. 99. <= Loc. cit., p. 105. ' Loc. cit., p. 98.

REPORT ON THE KERATOSA. 11

out the internal organisation of the Porifera and applied it to systematic purposes.
Prof. Schulze ^ characterises his family of Spongelidse primarily by the structure of the
soft parts ; in Dr. Marshall's diagnosis of Dysideidse not a single word is devoted to this
character. Does he regard it as devoid of any systematic importance 1 This seems to be
the case, but the grounds for this opinion are not to be found in Dr. Marshall's paper,
and without this the paper in question is of very doubtful profit for systematic know-
ledge, and this is the case with every paper on the Keratosa based on specimens insuf-
Jiciently preserved for anatomical exa.mination. Such papers, if from the pen of authors
of talent, may still contain something of general interest, and this is in a high degree
the case with regard to Dr. Marshall's memoir on Dysideidte and Phoriospongise, but just on
that account it is the more to be regretted that from a systematic point of view this memoir
only renders the systematic chaos relating to the group Keratosa still more impenetrable.
In the Bemerkungen liber die neue Gray'sche Hornschwammgattung lanthella"
by Dr. Flemming, spongiological literature has been enriched with a new and very valu-
able contribution to our knowledge of the properties of the horny skeleton. Dr. Flemming
has stated — and, as I before remarked, I can but confirm his statements — that the fibres
of lanthella flabelliformis and lanthella basta contain true cells between the laminos
of their walls. The specimens of both (?) the sj)ecies obtained and examined by Dr.
Flemming were dried, and thus he has not been able to decide the question as to whether
lanthella is really to be referred to Porifera. He believes, however, that, if so, a very
isolated place in the group is to be assigned to this sponge (" Es scheint mir zuniichst
nicht viel iibrig zu bleiben, als sie bei den Spongien, aber auf einem sehr isolirten Platze
einstweilen stehen zu lassen ").^ Dr. Flemming's suggestion is to be regarded as quite
natural, if we remember that his paper on lanthella appeared in the year 1871, i.e.,
five years before the important discovery of spongoblasts by F. E. Schulze ; but at
present one may perhaps form another opinion as to the systematic importance of the
peculiarity in question. The gist of the matter consists in the fact that each horny fibre
is the product of cellular elements; and whether the spongoblasts, after having accomplished
one part of the work, recede before the developing fibre in order to partake in its further
growth, or remain in their former places in order to be buried in the horny substance
secreted by younger spongoblasts, seems to me to be of secondary significance.
Dr. Flemming demands for lanthella a quite isolated place amongst the Keratosa ;
again, Gray, Carter, and Hyatt, on the ground of characters of undoubtedly less
value, of characters to which in other instances we should apply only a generic if
not a specific significance, are inclined to regard it as the representative of an inde-
pendent family. For my own part, I must confess frankly that the systematic importance
of the peculiarity we are speaking of is rather ambiguous. For though, as before

1 Zeitschr.f. wiss. Zool, vol. xxxii. ■^. 153. - Wiinhmj Verhandl., N. F., Bd. ii.

' Loc. cit., p. 7 ; comp. also p. 6.

12 THE VOYAGE OF H.M.S. CHALLENGER.

stated, from a theoretical point of view it appears to be but of a subordinate character,
the possibility is not excluded that the property in question might have been assumed
very early (in a pala3ontological sense), and thus if we should in time find lanthellidse, i.e.,
horny sponges whose skeletal fibres are charged with true cells, of thoroughly different
internal organisation, we should be obliged to elevate the character in question to the rank
of that of a subfamily or even family. But hitherto this has not been the case ; all the lan-
thellidge known up to this time are only representatives of the same genus, and in order
to avoid a superfluous overburdening of spongiological nomenclature, sufiiciently overbur-
dened already. I should propose to regard temporarily the genus lanthella as merely a
genus of the family Darwinellidse (Aplysillidte), the more so as to a similar variation
(in the reverse direction, however,) in the Vertebrata^ no systematic consequence has been
ascribed.

This would be the place to discuss the systematic value of the organisms known by
the name of " Fibrillen " or " filaments," which for a long time have been considered a
constituent part of the horny skeleton. The inducement to this has been given by the
statements of Oscar Schmidt, who thought ^ he had seen filaments in immediate connec-
tion with true horny fibres. On the ground of this false supposition he adopted the
genus Hircinia, Nardo, and characterising it primarily by the presence of filaments,
subdivided it into two subgenera; Gray'* in the year 1867, relying on Oscar Schmidt's
statements, established a new family " Hircinidse," characterised by the possession of a,
so to speak, double skeleton. There has been however, on the part of other spongio-
logists, some doubt as to the reliability of Schmidt's statements. KoUiker* proclaimed
the filaments to be parasites, and this opinion has been warmly defended also by Mr.
Carter,' apart from the point that, according to Kolliker, the filaments are fungi, accord-
ing to Carter, algfe. In the meantime, Prof. Schmidt" altered his opinion ; further and
more careful examination led him to the conclusion that the filaments were in no
connection with the true skeleton, and this suggestion has been confirmed also bj^
Schulze in his paper on " die Gattung Hircinia, Nardo und Oligoceras, g. n." Agreeing so
far as the real facts are concerned, both the investigators just named difier, however, with
regard to further conclusions. In his report on the Spongien der Kiiste von Algier
Prof. Schmidt believes^ that if it were once proved that the filaments are nothing but
parasites, the genus Hircinia ought to be dissolved as an independent genus, and incor-
porated in the genus Cacospongia. Prof. Schulze comes to quite a different conclusion ;
while inclined, as he has been, to regard the filaments as independent organisms, he
ascribes* to their presence in certain Keratosa a generic and even family character, which

1 I allude to the osteoblastless skin-bones of Cceciliae (Leydig) as well as to the equally osteoblastless bones of
Teleostei supporting their fins (Kolliker).

2 Spong. d. Adriat. Meeres, pi. iii. fig. 10. 3 p^gg^ Zool. Soc. Land., 1867, p. 503.

* Icones histologicce, Abth. i. p. 49. ^ Ann. and Mag. Nat. Hist., ser. 4, vol. viii. p. 330,

^ Zeitschr. f. v:iss. Zool., Bd. x.xxi. p. 661. ' 1871.

' Zeitschr. f. vnss. ZooL, Bd. xxxiii. p. 34. ^ Loc. cit., p. 36.

REPORT ON THE KERATOSA. 13

opinion, in spite of objections made by Dr. Marshall,^ has also been adopted by
Dr. Vosmaer." Prof. F. E. Scliulze is indeed the most eminent spongiologist of the
present time ; Dr. Vosmaer is his pupil, and since I have entirely different ideas on the
matter, I must submit it to a most careful examination. There are three questions to
■ be answered, namely, first, whether the filaments form a constituent part of the organism
of certain sponges or are independent organisms ? second, if they are independent organisms
are they to be regarded as parasites ? and third, if so, is their presence to be used as a
character of systematic consequence ?

Do the filaments stand in an intimate connection with the sponge organism, or are
they independent beings ? We have seen that this question has been answered in
the negative by numerous investigators. This negative answer has, however, had either
no foundation in fact, or only an ambiguous one, till F. E. Schulze's paper on Hircinia
appeared. What we read in the Icones histologicse of Kolliker (p. 49) is to be regarded
merely as a simple supposition ; in the papers of Carter and Hyatt we have indeed
to deal with a conviction, but this conviction is far from being contagious. Hyatt states'
nothing more than that the examination of filaments by Dr. Farlow led to no definite
results as to their nature. Carter wishes to prove their parasitic nature by the fact
that he did find filaments in many non-Keratosa, and again missed them in notorious
Hircinida3 such as Hircinia campana. It may be said, however, that, as to the latter
argument, the sponges in question have accordingly not been Hircinidae, and as to
the former one, that it is also of no decisive nature, the reliability of the observations
upon the point being still questionable. This has been pointed out by F. E. Schuke in
his above-mentioned paper (p. 33), and there has been no answer on the part of Mr.
Carter. The discovery of 0. Schmidt that the filaments, which by their shape vividly
recall skipping-ropes, are quite free at both extremities, proved that they had nothing
to do with the skeleton, but did not prove their independence of the sponge organism
in general. This latter has been made obvious liy Schulze, who made out the struc-
ture of sponges characterised by the presence of filaments, and found that anatomically
and histologically they do not differ from sponges which like Euspongia have never
been found with filaments. To this statement I ascribe the highest importance. If
the filaments have nothing to do with the skeleton, and if again there exist no
deviations in histological structure of the corresponding sponges, which deviations,
according to the law of correlation, ought to be expected, provided that filaments
form a constituent part of their body, there are no grounds to consider them to be
part of it. Whether they are algfe or fungi still remains questionable, but their
nature as independent organisms is, I think, now clearly established. Schulze himself is
also of this opinion, although, with his usual prudence, he states it rather conditionally.
Notwithstanding, he is still inclined to ascribe to the presence of filaments a high sys-
tematic significance ; he appeals to an analogous instance in the vegetable kingdom. He

' Zeitschr.f. loiss. Zool, BJ. xxxv. p. 112. - On Velinea gracilis, p. 445. ^ Revision, &c., part ii. p. 546.

14 THE VOYAGE OF H.M.S. CHALLENGER.

says the lichens are also nothing but fungi and algfe together ; still, they form a systemati-
cally independent subdivision, owing precisely to their double nature. I do not know
whether botanists are right in separating systematically the lichens, l)ut this question is of
no consequence to us ; I do not, however, believe the comparison just mentioned to be
fortunate. The lichens represent an extreme instance of that kind of symbiosis which
v. Beneden caEs " mutualismus." The symbiosis of filaments with sponges is even not so
much a commensalism as an indubitable parasitism. In the case of mutualism, the beings
constituting the whole undergo such modifications in their structure that a separate
existence becomes impossible. There are to be found no deviations in the organisation of
sponges attacked by filaments, as compared mth that of closely allied forms devoid of
them. Microscopic preparations of Hircinia variahilis and Cacospongia scalans show no
differences apart from the presence and absence of filaments. There are, accordingly,
absolutely no grounds for supposing that Hircinia, having got rid in some manner or
other of filaments, would be no longer able to live. It is also quite obvious that the
phenomenon under consideration is not commensalism, but parasitism. Beginning with
forms like the Challenger specimen of Cacosjwngia coUectrix, where filaments are repre-
sented but very scantily, we come through all possible intermediate stages to the forms
which, as in the case of the Challenger specimen of Cacospongia irregularis, are so
overloaded with them that the parenchyma is almost entirely replaced by these
curious organisms ; all this, now that the independence of the filaments is no longer
doubtful, seems to me a decisive proof in favour of the opinion held for many years
by Carter, that in the filaments we have really to do with nothing but parasites.
This conclusion alters the matter. There are instances in which the diagnosis of
parasites involves the mention of the host they inhalnt. This does not occur, however,
except in connection with certain modifications in the structure of the parasite,
rendering it impossible for it to inhal)it anotlier host ; but nobody would characterise
the host systematically by reference to its parasites, and should we adojjt the family
of Hircinidse as a fiimily represented by forms witli filaments in the parenchyma, we
should be equally obliged to subdivide the species Homo sapiens, according to the
presence or al^sence of Tcenia solium, into two systematic groups. Hence I propose the
dissolution of the family Hirciuidse, and the location of its representatives in other
groups, according to more positive characters.

The question as to the nature of the filaments, whether j)lauts or animals, is therefore
beyond the domain of spongiology ; but since it is of great general interest, I
venture to communicate here what I was able to make out in this direction durina: the
examination of the Challenger material. As to their structure, I must refer the reader to
the statements of F. E. Schulze. I was able to discern all he has seen, but was unable
• — in spite of the excellent homogeneous immersion system (^) of E. Winkel — to make
out anything more as to their organisation. The comparative size and shajie of the heads

REPORT ON THE KERvVTOSA. 15

are shown in PI. VIII. figs. 4 and 5. What I have to communicate concerns their origin,
provided 1 am right in identifying the dumb-bell shaped bodies whose description will be
given later on along with the filaments. Bowerbank, Schmidt, and Kolliker found in
many of the specimens characterised by the presence of filaments certain round bodies,
and it has been supposed by the last named naturalist that these bodies give rise to the
filaments. F. E. Schulze, on the contrary, maintains that there can be no question as to
any such connection ; he finds these bodies too, but is inclined to consider them to be
unicellular algffi, the more so as he has observed them in phases of division more or less
complete. I am uncertain whether the bodies I am about to describe, which are
represented on PL VIII. fig 2, are identical with those of F. E. Schulze and Kolliker. I
believe them, however, to be so, since the corresponding description of Schulze is thoroughly
applicable to them, and also because I found some of them on the point of dividing into
two halves ; but even if identical, I am yet more inclined to adopt the opinion of Kolliker,
and to ascribe to them a certain connection with the filaments. I observed these round
bodies in my Cacospongia dendroides, and comparing them with the heads of its
filaments, was struck by their mutual resemblance. Not in every case, however ; for,
beginning with forms characterised by the thickness of their walls and indistinct central
difi'erentiation through numerous intermediate stages, I came to forms with walls far
thinner, and, in their central contents, recalling very much the drawings Schulze gives of
his conjectural alg^ in Hircinia (loc. cit, pi. iv. fig. 15). An attentive examination
of their walls shows that they are provided with a small thickening, which appears to
correspond exactly with the spot where the thread of the filament separates from its
head. In a couple of cases, I think, I have also seen these bodies, when grown thin-walled,
6tni in connection with the thread; but even if this were not an optical illusion, it seems
to occur pretty seldom, and we have also to suppose that the further difi'erentiation of the
filamental heads — provided that they are identical with our round bodies — takes place
only after the head has separated from the thread. The final phase of this development
consists in the bursting of the wall so that the internal contents escape. It is represented
by corpuscles which, when out of the capsule, present a great variety of size, some so
small that they appear to be merely points, others so large that they are readily to be
discerned l)y a magnifying power no higher than 200-250. In some cases these corpuscles
appear to have linear or even fusiform outlines. This, however, is very rare; they almost
always have a very definite dumb-beU shape, so that if such a body were to be imagined
as growing in length a typical filament would result. But if this reaUy occur, it apparently
does not take place immediately, the corpuscles having the faculty of multiplying previously
by division; sometimes, but not often, a cross division has been observed, sometimes a
star-like one, sometimes a longitudinal one. As the result of this division, a new
generation of corpuscles of the same form may again occur. As to their internal structure,
I can state that, as iii the filamental heads, their walls consist of concentric layers ; a

16 THE VOYAGE OF H.M.S. CHALLENGEE.

certain clifFerentiation in their centre is also to be discerned, but I am not sure whether
this is not the result of an imperfect penetration of the staining fluid into their bodies.
Their whole appearance is a " bacterian " one. So much I can state from actual
observation. Induced by these observations, I was desirous of pursuing their conjec-
tural transmutation into true filaments. I was, however, unable to accomplish this, and
thus corroborate with decisive proof the statement as to the mutual connection of the
round alg£e-like bodies, the dumb-bell shaped corpuscles and the filaments. Can this be
supposed nevertheless ? When deliberating over the phenomenon theoretically, I was
inclined to answer this question in the negative ; for, as remarked before, the dumb-bell
shaped corpuscles, having reached a certain size, show in most cases a tendency to
multiply, which again cannot be denied with respect to the round bodies themselves. All
this would be very strange, if we should identify the former with filamental heads and
the latter with filaments themselves ; and again, though I have seen single dumb-beU
shaped corpuscles of comparatively very large size (0'06 mm.), 1 never saw them so large
that one might regard them really as young filaments ; while if they really undergo
transformation into filaments, one would expect to find all possible intermediate stages.
On the other hand, when examining the corresponding preparations, and comparing the
round bodies with the filamental heads and the corpuscles wath true filaments, I can
give to the above question but one answer — a decided " Yes." But whatever be the fact,
I consider it my duty to communicate in a most detailed manner what 1 have observed,
and what may stand in connection with the origin of the filaments, hoping that my
statements may be of help to any naturalists who may at some future time enter upon a
special research into the nature of these enigmatical formations.

I turn now to a detailed discussion of the systematic value of the properties of the
canal system. The diagnoses of the two types of canal system characteristic of Keratosa
have been already given on pages 4, 5. It has been also stated that the properties of the
canal system are in a certain sense antagonistic to those of the skeleton with regard to
its composition either of homogeneous or heterogeneous fibres ; and again, that according
to this latter character the horny sponges do not admit of their subdivision' into two
main groups, since such a proceeding would have a certain phylogenetic signification
inconsistent with the circumstance that the difi"erences in the histological structure of the
skeletal fibres of an Aplysina and Euspongia, and on the other hand of an Aj^Iysilla and
a Sponyclia, are of a quantitative and not a qualitative nature. But, owing to the fact
that Aplysina, through Veroiujia and Liiffariu, is connected with true Spongidse, and again
that the genus Aplysilla is connected, as Vosmaer has lately shown, through Velinea
with Sjyongelia, the canal system of the sponges first mentioned being constructed upon
one type, that of Aplysilla, Velinea, and Spongelia, including other allied genera upon
another, one would question whether the Keratosa can be subdivided precisely according
to the structure of their canal system. This question is again to be answered in the

REPORT ON THE KERATOBA. If

negative, and on grounds of exactly the same nature as those forbidding the subdivision
of the Keratosa into Ceratina and Psammonemata, Carter, or Aplysinae and Sponginas,
Hyatt. There is a striking difference between the canal system of Aplysilla and Aiolysina.
The flagellated chambers of the representatives of the genera Aplysina and Verongia
(PL X. fig. 7) are small, pear-shaped, or rather hemispherical, each provided with one (1)^
inhalent and one exhalent narrow canaliculus; and again, the surrounding ground-mass
is so very rich in granules that the outlines of the cellular elements in the neighbourhood
of the flagellated chambers are scarcely distinguishable. On the other hand, the flagel-
lated chambers of an Aplysilla or Icmthella (PI. II. figs. 4 and 5) are large and either of
regularly elongated form (pouch-shaped) or of quite irregular outline ; no special cameral
canaliculi are to be discerned; the flagellated chambers receive the water from the sub-
dermal cavities by means of numerous pores in. their walls, and expel it by means of
a large exhalent aperture ; the surrounding ground-mass is clear and transparent. There
are, however, amongst the horny sponges forms uniting these two extreme differences in
every direction. As to the size of the flagellated chambers, in Aplysina or Verongia it is
0"02 mm. on an average, in Euspongia or Cacospongia 0"026 mm., in Fhyllospongia 0'037
mm., in Carteriospongia 0"05 mm., in Spongelia 0'08 mm.; finally, in Aplysilla or
lanthella the flagellated chambers are still larger, reaching occasionally 0'15 mm. in length
by 0*05 mm. and more in width ; and it must be noticed — and this is very important — that
in some Spongelidaa (comp. PI. III. fig. 6) the fltigellated chambers are again very small,
their dimensions not exceeding those of the flagellated chambers of a typical Euspongia.
Further, as to their form, we have a thoroughly similar series of connecting links. In
the Aplysinidge they are either pear-shaped or rather hemispherical, in the Spongidse
typically hemispherical, in the Spongellidse more or less roundish, in the Darwinellidse
elongated. The same is also the case with respect to the presence or absence of special
cameral canaliculi. While in Aplysinidse each flagellated chamber possesses but one
exhalent, and probably also but one inhalent, canaliculus, these canaliculi being compara-
tively long and narrow, in the Spongidse they are short and broad, the inhalent system
of each flagellated chamber being besides represented not by one but by three, four, or
five canals, which sometimes are so very short that in many cases they can scarcely be
properly regarded as special differentiations of the corresponding subdermal cavities.
I refer the jeader in this connection to the drawing of F. E. Schulze^ and to my own
drawing on PL V. fig. 3, and wish to add that in many, indeed exceptional but still
numerous, instances I found in true Spongidai the flagellated chambers devoid of
any special exhalent canals, but just as is the case with the genus Carteiiospongia

1 This question F. E. Schulze {Zeitschr. f. iviss. ZooL, Bd. xxx. p. 398) leaves undecided. I also was unable to come
to a decisive result with respect to Aplysina and Verongia owing to the inconvenience of these forms for certain manipu-
lations, but so far in this respect as analogous forms like Corticium, Chondrosia, and Chondrilla are concerned, there can be
no doubt that each iiagellated chamber possesses but one inhalent canaliculus.

2 Zeitschr. f. loiss. ZooL, Bd. xxxii. pi. .xxxvi. figs. 11, 12.

(ZOOL. CHALL. EXP. — PART Xi'XI. — 1884.) Hh 3

18 THE VOYAGE OF H.M.S CHALLENGER.

(PI. V. fio-. 8), communicating with the subjacent exhaleut cavities by means of a \avge
exhalent opening. Finally, as to the question whether the ground-mass surrounding the
flagellated chambers is full of granules or devoid of them ; the more or less clear manifesta-
tion of this character appears to be in such a high degree dependent on other properties
of the canal system that it may be said beforehand that we shall have in this respect a
series of intermediate stages between Spongelia and Aplysina similar to that noticed with
reo-ard to the form and size of the flagellated chambers. As is well known, it has been
stated by F. E. Schulze that (except in Oscarella {Halisarca) lobularis, which is, however,
not a typical representative of the sponges with the canal system after Dr. Vosmaer's*
4th type) the ground-mass around the flagellated chambers has been found to be granular
whenever the canal system shows a high degree of development in the direction of
enlarging the surface washed by the water. But Schulze has also stated that, whenever
the development of the canal system has reached its highest point {Aiylysina, Chondrosia,
Corticium), the ground-mass is literally over-loaded with granules ; that further, whenever
the canal system possesses a transition character (Spongidse and particularly Plakinidas),
the granules are by no means so numerous ; and, finally, that in the representatives of the
genera Spongelia and Aplysilla the granules are not to be found at all. For my own
part, I have been fortunate enough to investigate some forms which are, in this respect, a
connecting link between the genera Euspongia and Spongelia. In Phyllospongia — whose
flagellated chambers are, as before stated, approximately one and a half times as large as
those of Euspongia or Cacospongia — I could observe the presence of the granules in
question only with the help of the system No. 7 of R. Winkel;^ in Cartcriospongia — and
in this form the flagellated chambers, although of the same shape as those of Euspongia,
are comparatively much larger and at least in most cases devoid of special inhalent and
exhalent canaliculi — the granules proved to be extremely few in number, and their presence
could have been placed beyond doubt only by the system No. 10 of Winkel.^ To sum
up, we have, with regard to the structure of the canal sj^stem, a series of transitions quite
analogous to those concerning the skeletal fibres, whether homogeneous or heterogeneous.
Accordingly, the final conclusions must be also analogous one to another.

I may now summarise the foregoing observations. There have been discussed five
systematic characters, and we came to the conclusion that two of them are absolute
characters, the remaining three being only relative. We came also to the result that one
of the first-mentioned, namely, the presence of filaments, is capable of no systematic
application, and again, that so far as the second absolute character — I mean the presence
of true cells in the walls of the skeletal fibres — is concerned, its systematic value cannot,
in the present state of our knowledge, be defined.

Of the three relative characters, one, the tendency to take in foreign bodies, only

' Ueber LeucanJra aspera, Tijdschr. d. Ned. Dierk. Vereen., B<1. v. p. 163.

' The objective No. 7 magnifies 275-700 diameters, No. 10, 500-1000 diameters, according to the eye-piece emploj-ed.

REPORT ON THE KERATOSA.

10

concerns tlie sponges with homogeneous skeletal fibres, and it has been stated that this
character is of a very conditional nature, since the tendency in question is common to
sponges of entirely different internal organisation. The properties of this latter, as well as
those of the more detailed structure of the skeletal fibres, constitute the two other relative
characters, and we have seen that, apart from the point that they are relative, they are
also antagonistic to one another in a certain sense of the word. Such is the case with
respect to the two characters to which, in the Keratosa, the highest systematic value has
been repeatedly ascribed. The fact of their being relative does not permit us to make
use of them in order to divide the group in question into two subdivisions, while
this is demanded by their mutual antagonism. And yet these characters are un-
doubtedly the most important, the only characters according to which any main sub-
divisions may be realised, for they concern the structure of both the constituent parts
of the organism of a horny sponge, and as to other systematic characters of the Keratosa,
we shall soon see that they are unfit even for good generic distinctions.

What kind of arrangement can be adopted under such conditions 1 Dr. Vosmaer, as if
in order to reconcile the contradictions in question, proposes ^ a subdivision of the group
directly into families, characterising them by the properties both of the skeleton and of the
soft parts. His arrangement will be adopted in the descriptive part of this paper, as no
better arrangement seems at present possible; it is, however, an artificial one. By our
systems we have to express the phylogenetic affinities of the corresponding animals, and the
arrangement of Dr. Vosmaer does not express them. He subdivides the Keratosa into five
families directly. How is this to be understood from a phylogenetic point of view? Are
these five families divergent branches from the same spot of the general genealogical tree
as represented by the diagram A 1 or do they form together the figure represented by the

diagram B ? There are no answers to these questions, and this is not the fault of Dr.
Vosmaer, whose arrangement is at any rate the best of all others, for it pays attention to
the characters both of the soft parts and of the skeleton ; the matter itself is of a very

1 On Vdinca gracilis, p. 444.

20 THE VOYAGE OF H.M.S. CHALLENGEE.

ambiguous nature, since systematically the horny sponges present a kind of circtdus
vitiosus. Through Luffaria and Verongia, Euspongia is connected with Aplysina, but,
on the other hand, it is also, through Carteriospongia, connected with Spongelia; now
Spongelia through Velinea is connected with Ajjlysilla, and again A2)lysiUa through
lanthella is connected with Aplysina. It is evident that under such circumstances the
families of Dr. Vosmaer are by no means natural systematic groups. With regard to the
mutual relations of the genera of different families, such a circulus vitiosus is inconceivable,
and proves only that such families are not natural. As I remarked before, I adopt the
arrangement proposed by Dr. Vosmaer ; to attempt a better one would, I believe, be at
present premature ; but I adopt it only as an arrangement of a provisional character.

The foregoing observations were begun in order to decide the question — Which
of the existing systems of the Keratosa is the most to be recommended ? The unexpected
result at which we arrived is of such high importance that its further foundation becomes
desirable, and as this latter demands a detailed discussion of characters used by classifiers
as systematic distinctions between different representatives of our conjectural families, I
now pass on to those of generic value.

IL-CPiITICISM OF THE GENEEA.

There are in Science two directly opposite opinions as to the question of what a genus
ought to be. Some naturalists demand for the genus the presence of an absolute,
quahtative distinction ; others are satisfied with a distinction of a quantitative nature, and
while, e.g., Nageli summarises his opinions as follows : —

" Fur die generisehe Trennung scheinen mir folgende Bedingungen als unerlasslich
bezeichnet werden zu mtissen :

1. Uebereinstimmung im ganzen Verhalten, also innigere natiirliche Verwandtschaft
unter den Arten der einen und unter den Arten der anderen Gruppe, — und Differenz im
ganzen Verhalten, also geringere natiirliche Verwandtschaft zwischen den beiden Gruppen.

2. Mangel von Zwischenformen zwischen den beiden Gruppen ; — die Arten der gleichen
Gruppe konnen durch Uebergangsglieder verbunden sein oder nicht.

3. Vorhandensein von absoluten (nicht bloss von relativen) Unterscheiduugsmerk-
malen."^

In F. E. Schulze's paper on the Spongidae^ we find the following sentence: —
" Besonders schwierig erscheint oft die Entscheidung der Frage, ob eine Anzahl
verwandter Formen als Arten einer Gattung oder als Varietaten einer Art hinzustellen
sind." Under such circumstances it is but obvious that the systematic literature of
Keratosa, like that of all other groups, must be of a rather chaotic character. It
is so. In the following passages I hope to clear up this chaos so far as the genera
are concerned, and so far as this is necessary for the descriptive and some other more
general purposes of this paper.

Family Daewinellid^.
It contains the following genera : —

Darwinella.

This genus was established by Fritz Mllller^ and characterised by him primarily
by its horny spicules. There are no data in the literature of the subject as to the
internal structure of the soft parts of the representatives of this genus ; both F. E. Schulze

1 Sitzwvjsb. (I. h. baier. Akad. d. Wiss., 1867, Bd. i. p. 168. 2 zdtachr.f. iviss. Zool, B;l. xxxii. p. 612.

'^ Archivf. mikrosk. Anat, 1865, Bd. i. p. 344.

22 THE VOYAGE OF H.M.S. CHALLENGER.

and Vosmaer, thougli inclined to place it systematically near Aplysilla, do so with great
reserve, " provided that its internal organisation does not differ from that of other
Aplysillidse." I am in a position to authorise this proceeding entirely. Some time after
having published his paper on this form, Prof. Schulze visited a Dalmatian Island, Lesina,
and there dredged up specimens of a sponge which does not admit of even a specific
distinction from Darivinella aurea, Fritz Miiller. Prof. Schulze has been kind enough to
hand me over these specimens for examination, and I can state that the internal organisa-
tion of Darivinella aurea follows on the whole the type, anatomical as well as histological,
characteristic of Aplysilla. Accordingly, and in opposition to Dr. Vosmaer,^ since the
family name of Darwinellidae was established by Merejkowsky ^ in the year 1878, that
of Aplysillin^, v. Lendenfeld' and Aplysillidas, Vosmaer,^ only in the year, 1883,
I propose to return to the former name of the family in question. The genus Darwinclla
is a good one, its horny spicules differentiating it sharply from all other Keratosa.

A'plysilla, F. E. Schulze^ {Simplicella, Merejkowsky ").

Lendenfeld' characterises this genus by its crust-like form and by numerous inde-
pendent tree-like skeletal fibres, — a character common also to Darivinella. The
distinction consists in the absence of spicules. This genus would be a good one only
if united with the next under consideration.

Dench'illa.

Here v. Lendenfeld® finds the generic character in the property that the skeletal
fibres do not stand separately as in Aplysilla, the whole skeleton presenting a single
tree with numerous secondary, tertiary, and other branches. This distinction is, how-
ever, scarcely of generic consequence. In the Spongelidse F. E. Schulze " has
described a form [Spongelia S2nnifera), which differs from other representatives
of the genus in the same direction, but even more than Dendrilla from A2Jlysilla;
yet he merely placed it as a new species in his genus Spongelia. At any rate Ap)lysilla
and Dendrilla show much closer affinities with one another than with Darwinella or
lanthella.

lanthella.

This genus was established in the year 1869 by Gray,'" as the representative of an
independent family of horny sponges, and excited among naturalists a great doubt

' On Velinea gracilis, p. 444. ^ jVfg,^ j^, I' Acad. d. Sci. de St. P^tersb., tome ,\xvi., No. 7, p. 44.

3 Zcitschr. f. wiss. Zool., Bel. xxxviii. p. 235. i Loc. cit., p. 44.3.

5 Zeitschr.f. vdss. Zool., Bd. xxx. p. 404. « Loc. cit., p. 43. ' Loc. cit, p. 309.

8 Ibid., p. 270. " Zeitschr.f. wiss. Zool, Bd. xsxii. p. 152. i" Proc. Zool. $oc. Lond., 1869, p. 49.

REPORT ON THE KERATOSA. 23

as to whether it is really a sponge. Flemming^ and F. E. Schulze ^ have expressed such
doubts ; and Marshall^ feels more inclined to regard it as an alga than a sponge. Amongst
the Challenger specimens I find two forms which are well-marked representatives of the
species lanthella jlahellifortnis, and which are also distinct sponges with the canal system
after the type of that of Aplysilla or Dai-winella. The genus lanthella, like Darwinella,
is indeed a very good one, its chief generic character, viz., the presence of true cells
between the horny laminae of the skeletal fibres, being of an absolute nature, but it must
be added that with respect to the conjectural family Darwinellidse, this genus narrows the
diagnosis of the family as given by Vosmaer (On Velinea gracilis, p. 477), the skeleton of
its representatives being represented by fibres anastomosing with one another.

Family Spongelid.b.

In the family Spongelidse (I retain this name since the only one having the priority
over it is derived from the still doubtful genus Dysidea) six genera have been hitherto
distinguished, namely : —

Velinea.

We owe this curious genus to Dr. Vosmaer,* who places it in his family of Sponge-
lidoe, notwithstanding that its skeletal fibres are, at least in many instances, obviously
heterogeneous, but in accordance with its skeleton being represented by a network of
anastomosing fibres ; his Aplysillidje, on the other hand, being characterised by a
skeleton of tree-like shape. We have, however, learned that this latter character is
not admissible. The skeleton of lanthella is composed of anastomosing fibres, yet these
latter are heterogeneous, and the whole internal organisation constructed upon the type
of true Darwinellida3. The fact is that Velinea is a connecting link between this latter
family and Spongelidse, and as such it is indeed a very interesting form, but, as an inde-
pendent genus, belongs to those which from a systematic point of view are very ambiguous.

S'pongelia, Dysidea, Psammascus.

In the year 1834 Nardo^ subdivided his previous genus Aplysina into two subgenera,
" AplysincB velatice" and" AplysimB spongelioi." In the year 1842 Johnston^ created
the genus Dysidea for his Dysidea fragilis, which, according both to 0. Schmidt ' and

' hoc. cit., p. 6. 2 Zeitschr.f. iviss. Zool, Bd. xxx. p. 381. ^ Jahresb. d. zool. Stat. Neapel, 1881, p. 161.

* Mittheil. zool. Stat. Neapel, B.l. iv. Heft. 4, p. 437. ^ /jij.^ 1334, p. 714.

" British Sponges, &e., p. 185. ^ Spoiig.-Fauiia des atlantiscli. Gebietes, p. 27.

24 THE VOYAGE OF H.M.S. CHALLENGER.

Marshall/ cannot be separated even specifically from Spongelia pcdlescem, Nardo. Not-
withstanding, in England, the name Dysidea has not been given up, and in order to
reconcile English and Continental naturalists, Marshall proposed to retain both these
names in order to apply them to different forms. He groups Johnston's Dysidea
fixigilis in the genus Spongelia as restricted by F. E. Schulze, and, on the other hand,
Bowerbank's species Spongelia coriacea and Hyatt's Spongelia fragilis in the genus
Dysidea, giving it a new and detailed diagnosis, and in harmony with Hyatt — who
four years before also retained both these genera, and even grouped them in different
families on the ground of a thoroughly false supposition as to the manner of grou-th of
the horny skeletal fibres — and, I repeat, in harmony with Hyatt, demands a more
pronounced separation of both these genera. This latter view is expressed in a report on
his own memoir on Dysideidse and Phoriospougise,^ while in the memoir itself (foe. cit.,
p. 91) he seems to be of a rather different opinion, writing as follows : "Ganz scharf zu
trennen siud diese genera freilich nicht, so wenig wie Euspongia und Cacospongia, diese
und Spongelia." This is, however, of little consecpience indeed, but unfortunately, on
the whole, instead of having simplified the matter Dr. Marshall complicated it still more.
The series of his Dysideidse in the paper above mentioned he opens by the newly created
genus Psammascus, characterising it by a great number of peculiarities, each of which,
however, must be regarded as almost devoid of any systematic importance. This is also
but of little consequence ; it is not for the first time that in the Keratosa bad genera
have been established and bad generic definitions given, but Dr. Marshall adds to the
above definition the following remark : " Of all Dysideidaj this genus (Psam^nasciis)
shows the closest affinities to the genus Spongelia, being however to be readily
distinguished from it by the presence of foreign enclosures also in the soft parts. "'^ If
now the reader will compare the definition Dr. Marshall gives of his genus Psammascus
{loc. cit, p. 92) with that by which he describes the genus Dysidea [loc. cit., p. 98), he
will find that this latter genus — apart from the character consisting in the presence of
a skin containing numerous foreign enclosures and to be easily di'awn off (a character of
a very doubtful systematic consequence,* but for which nevertheless Dr. Marshall evinces
the greatest predilection)— differs from Psammascus only in the presence of foreign
bodies in the parenchyma ; and, when I add, in harmony with F. E. Schulze,^ that
Dr. Marshall has been quite wTong in supposing the true representatives of the genus
Spongelia, in the sense of F. E. Schulze, to be devoid of any foreign enclosures
both in the parenchyma and in the dermal membrane ; when I further mention that
neither in Dr. Marshall's paper on Dysideida) nor in the highly detailed and precise
memoir on the genus Spongelia by F. E. Schulze are any statements to be

1 Zeitschr. f. wiss. Zool., Bd. xxxv. p. 91. 2 Jahresb. d. zool. Stat. Neapel, 1880, p. 178.

2 Zeitschr. f. iviss. Zool., Bd. xxxv. p. 92.

* Coiiip. F. E. Schulze's discussion of the question in Zeitschr./. vdss. Zool., Bd. xxxiii. p. 14.
^ Zeitschr./. wiss. Zool., Bd. xxxii. ^t. 1.36.

EEPOET ON THE KERATOSA. 25

found as to the question of whether the upper skin of the representatives of this latter
genus is to be drawn oflf easily or not, and that this is due both to the state of preserva-
tion and to the number of foreign bodies in the dermal membrane ; and when I finally
lay stress upon the fact that Dr. Marshall makes no positive statements as to the internal
structure of his genera Psammascus and Dysidea, so that these forms, like the species
Dysidea coriacea of Bowerbank and Dysidea fragilis, Hyatt, perhaps all belong to the
family of Spongidae, or are to be distributed in different families of Spongidse and
Spongehdse ; — then the reader will see that after Dr. Marshall's paper the matter becomes
comparable with the Gordian knot, which can no longer be disentangled but only cut
asunder. This I perform in the following manner. I unite the genera Psammascus,
Dysidea, and Spongelia into a single genus Spongelia, characterising it by large flagellated
chambers — of course, as in the whole family, devoid of any special cameral canaliculi —
and by the tendency to form " conuli."

Psammoclema.

This genus of Dr. Marshall ^ I adopt for the forms with small flagellated chambers
and wdth the external surface smooth and devoid of any projections of the skin owing
to those of the skeleton. All this is only of a temporary character ; indeed, the
possibility is not excluded that many good species, still undoubtedly allied amongst
themselves, are all alike in the possession, for instance, of the character of forming
cylindrical tubes with a weU-developed central cavity, so that my species Psammoclema
vosmaeri would have to be transferred into another new genus ; but it must be remem-
bered that in the group Keratosa all the present arrangements are but of a provisional
character.

Psa7nmoj)emma.

This genus, established by Dr. Marshall,^ admits of a very sharp diagnosis. Among
its representatives no horny skeleton is to be found at all, the propjjiug apparatus of
the sponge being represented by foreign enclosures exclusively. Moreover, the genus is
perhaps really a good one, since, as suggested by Dr. Marshall, the possibility is not
excluded that the sponges in question receive the foreign bodies in a way different from
that in which they are obtained by other Spongelidaj (comp. Dr. Marshall's above-
mentioned paper, p. 121). At any rate, this genus is to be referred to the Keratosa,^ and
according to the internal organisation of the soft parts to the Spongelidae.

> Loc. cit, p. 109. - Loc. cit, ji. 113. ^ For grounds, see p. 46.

(ZOOL. CHALL. ESP. — PART XXXI. — ISSi.l Hh i

26 THE VOYAGE OF H.M.S. CHALLENGER.

Family Spongid^.

The family of SpongidfB possesses a larger number of genera than any other family
among the Keratosa, but it is only in Aplysinidae that we meet with genera of such con-
ditional characters.

Euspongia, Hijypospongia, Caxiosiwngia, Stelospongos.

The genus Euspongia has its own history. Established in the year 1859 by Bronn,
it has been adopted by 0. Schmidt, although not immediately. It was not adopted by
Alpheus Hyatt, who returned to the old name of Spongia, auctorum, but has been again
recalled by F. E. Schulze, although with a certain reserve. Prof Schulze, though adopting
this name as a generic one, still lays stress upon the fact that the genera Euspongia
(which he characterises by fine skeletal fibres forming very small meshes) and Caco-
spongia (characterised by him as well as by 0. Schmidt,^ from whom it originates, by
comparatively thick skeletal fibres and large meshes) are very closely allied to one
another, and that with respect to some intermediate forms the question of whether the
classifier has to do with a Cacospongia or Euspongia is to be decided only according
to his individual opinion.^ One might say that the matter is not so very complicated ;
one would believe that the Cacospongice and Euspongice are divergent branches of the
general genealogical tree (in the sense that the skeleton presenting a supporting apparatus
for the soft parts, one group of Spongidaj have adopted thick fibres and large meshes, the
other fine fibres and small meshes, both kinds of skeleton being thus mechanically
perhaps of equal strength), connected by the presence of all possible intermediate stages;
that in some thousands of years, when the latter have died out, spongiology wUl have to
deal with two very good genera. The matter is, however, by no means so simple, owing
to the fact that each of these genera shows other special modifications, and the genus
Euspongia, namely, in the direction which leads us to the genus Hipp)ospongia, F. E.
Schulze ^ ; a Cacospongia in order to become a Stelospongos, 0. Schmidt.* F. E. Schulze
characterises his genus Hippospongia by the presence of a well-developed system of
canals permeating the body of the sponge, often in such a manner that between them
only comparatively thin partition -walls can be found, and further, by the absence of
primary fibres which are directed in Euspongia and Cacospongia perpendicularly to the
external surface, the latter character being in causal connection with the peculiarity first
mentioned; the genus Stelospongos was established in 1870, and characterised by the
differentiation of the skeletal fibres in separated columns du-ected in a more or less
regular manner radially from the basis of the sponge towards the external surfaces,
and consisting each of a compact network of vertical primary and horizontal

1 Spong. d. adriat. Meer., p. 26. 2 Zeitschr.f. wiss. ZooL, Bd. xxxii. p. 612.

3 Zeitschr. f. iviss. ZooL, Bd. xxsii. p. 614. * SpoQg. des atlant. Gebiet, p. 29.

REPORT ON THE KERATOSA. 27

secondary fibres, these latter also occasionally uniting tlie columns with one another,
the fibres themselves being thick and hard. To the diagnosis given by F. E. Schulze
of his Hippospongia I can add the following point, allowing an easy distinction
to be made between the representatives of this conjectural genus, and concerning the
character of the outer surface, though admitting of no examination of the outermost
ends of the canals perforating the body of the sponge, but still betraying their presence
by an alternation of thick and massive portions of the sarcode with spots where only
very thin membrane covering the subjacent cavities is to be found. The drawing of
my Hippospongia anomala given on PI. VII. will illustrate my idea. This character,
however, is only of practical importance. To sum up, both in Stelospongos and Hippo-
spongia we have to deal with porous forms, the skeletal fibres of the first genus,
however, admitting of a distinction into thick primary and finer secondary ones, those
of the genus Hippospongia being on an average all of the same size and thickness ;
a good distinction indeed, and further, a typical Stelospongos has a quite different
shape from that of a typical Hippospongia. But as stated before, the matter is by no
means so very simple. As to the typical Hippospongice, I found many of them (aU
belonging to the species Hippospongia equina) in the collection of Prof. F. E. Schulze at
the Zoological Institute of Graz, and I make use of this opportunity in order to express
my great thankfulness to Prof. F. E. Schulze for his liberality ; as to the typical
Stelospongos, Prof. Steenstrup of Copenhagen has been kind enough to send me three
specimens of it determined as Stelospongos by 0. Schmidt himself. But together with
these three specimens, Prof. Steenstrup sent me some other horny sponges (different
varieties of Hyatt's Spongia agaricina, subsp. dura) distinguished also by radial
columns as described before, the fibres constituting them being, however, all of the
same dimensions ; and again, in the Challenger Collection I find one specimen also of the
same character, but with fibres thick and rigid, while those of Prof. Steenstrup's specimens
just mentioned are fine and elastic. The skeleton of a typical Hippospongia has a
rather difierent appearance from that of a typical Stelospongos,^ but it is obvious that
if the canals perforating the body of a Hippospongia were to assume a more regular
disposition, we should have a skeleton in the form of numerous columns standing separately,
which is so very characteristic of the genus Stelospongos. This is, as we have seen, the
case both with regard to Hyatt's Spongia agaricina, subsp. dura, and again, atl east
in a certain degree, with regard to the sponge of the Challenger Collection above alluded to.
The first mentioned can indeed be still regarded as Hipposp)ongia, the last mentioned,
however, only if we enlarge the diagnosis of the genus Hippospongia in order to group
into it forms with thick skeletal fibres. Neither is it Stelospongos, since its fibres do
not admit of the distinction into primary and secondary ones. I ask, to which character

1 Comp. Scliulze's paper on Spongklfe, Zeitschr. f. tviss. ZooL, vol. xxxii. pi. xxsv. fig. 14, and Schmidt's Spongien
des atlantischen Gebietes, pi. iii. fig. 13, and also my drawing PI. VI. fig. 2.

'28 THE VOYAGE OF H.M.S. CHALLENGER.

have we to give the preference, to the cliflferentiation of the fibres into primary and
secondary ones, or to their thickness and rigidity ? Through my Hippos2)ongia anomala,
and indeed many other still unknown forms, the genus Eusjyongia is very closely allied
to the genus Hippospongia ; through Cacospongia moUior, 0. Schmidt, it is not less
closely allied to the true CacospongicB ; both Cacospongia and its special modification
the genus Stelospongos being connected with typical Hippospongice by means of forms
similar on the one hand to my Cacospongia intermedia, and on the other to Stelosiwiigos
friahilis and Sjwngia agaricina, subsp. dura, Hyatt. The reader sees that in these
genera we meet the same circulus vitiosits as in speaking of the mutual afiinities of
difi"erent families of the group Keratosa, and that the classifier in numerous cases has
no other guidance than his own individual opinion. A quite analogous phenomenon we
find also with regard to the next genus.

Coscinoderma.

This genus was created in the year 1883, and defined by Mr. Carter,^ by many
characters of which, however, only one can claim the designation of a generic one, namely,
the uniformity of the skeletal fibres as in Hip>pospongia , these fibres not admitting of
the distinction into primary and secondary ones, being all of the same thickness, and not
forming polygonal meshes but such as may be compared with wool-whorls. Of course the
system of internal canals, so very characteristic of Hippospongia, is not to be found here.
Mr. Carter established his genus for only one species, Coscinoderma laniiginosum, and
characterised it, inter alia, by a specially differentiated dermal membrane full of foreign
bodies, the fibres of the skeleton being almost free from any enclosures, and by the
evenness of the external surface. In the Challenger Collection I find a specimen with the
dermal membrane, like that of Coscinoderma lamiginosum, full of foreign enclosures,
and in general, apart from the colour of the skeletal fibres, just of the same properties as
the above-mentioned species, with very fine skeletal fibres, and forming no polygonal
meshes. But the outer surface of this specimen proved to be uneven, owing to the
sharp-pointed dentieulations of the skeleton. Further, I find a specimen whose dermal
membrane cannot be easily drawn ofi", whose skeleton meshes are polygonal, but which
shows on the surface of its skeleton the same dentieulations, corresponding with sharp-
pointed networks of the skeletal fibres, precisely as in the specimen I have just spoken
of, but whose fibres are all of the same thickness, their colour — of a paler shade in the
specimen before mentioned — being, as in Coscinoderma lanuginosum, rather brownish, and
almost entirely devoid of any foreign enclosures. And finally, I find a specimen
quite difi'erent from those before mentioned in its external shape, with fibres cored
with foreign bodies, but still all of the same thickness. Are all these forms really so
closely allied to one another as to be united into one genus ?

' Ann. and Mag. Xat. Hist., ser. 5, vol. xii. p. 309, 1883.

REPOET ON THE KERATOSA. 29

Phyllospongia, Cartenospongia.

The first of these genera was created by Ehlers/ the second by Hyatt ; ^ both have
been united by the last-named naturalist into a special family Phyllospongiadse, char-
acterised by the leaf-like shape of its representatives, those of the genus Phyllospongia
being lamelliform and with the external surface quite smooth, those of the genus
Carteriospongia, though still lamelliform, very often indeed provided with numerous
lateral outgrowths, but far thicker, and probably in physiological hannony with this
latter peculiarity, with the outer surface presenting on its whole extension an alternation
of more or less deep elongated hollows, and more or less high, also elongated, tubercles.
I also think that both these genera are closely allied to one another, but this is only
my individual opinion, for while the skeletal fibres of Phyllospongm, thin and elastic as
they are, recall those of Euspongia, the skeletal fibres of Carteriospongia, far thicker than
the preceding and overcharged with foreign enclosures, resemble those of Cacospongia, the
possibility is not excluded that the PhyllospongicB are modified Euspongice, the Carterio-
spongice, on the contrary, modified Cacospongia. It may be further stated tliat the natural
systematic place of Carteriospongia is among the Spongelidse (comp. p. 17), while Phyllo-
spongia as regards its internal organisation belongs to the Spongidae. On the other hand,
the chief point concerning their external shape seems to be also of a rather ambiguous
nature, owing to the great variability of the form of the body in the group Keratosa, and
particularly of the true Spongidse. And it is very possible that though 0. Schmidt^ united
SpongioneUa, Bowerbank, with his Cacospongia, the species Spongionella pidchella,
Bowerbank, instead of being a British variety of Cacospongia scalaris as Schmidt sup-
poses, is merely a link connecting the true Spongidae with Phyllospongia. At least the
figure of Spongionella pidchella given by Bowerbank in his Monograph* recalls very much
that of Spongia [Phyllospongia) papyracea in Esper's Pflanzenthiere, the only distinction
consistino- in the comparative thickness of specimens which are leaf-like in both cases ;
and, on the other hand, it must be noticed that as to the second species of Spongionella
described by Bowerbank [Spongionella holdsworthii) , Carter^ identifies it directly with.
Spongia papyracea, Esper. To sum up, — the afiinities of both the genera I am speaking
of are surrounded by no less uncertainty than those of the genera Euspongia, Cacospongia,
&c., the generic distinctions being of the same conditional character.

Oligoceras, Hircinia, Ceratella.

The genus Oligoceras, established by F. E. Schulze " for some specimens from Lesina,
is indeed one of the worst genera. Marshall ^ classes it under his Dysideidse, and this

1 Die Esper'schen Spongien, p. 23. ^ Revision, &c., vol. ii. p. 540.

3 Spong. d. adriat. Meer., Bd. ii., Suppl., p. 9. * Vol. iii., pi. Ixv. fig. 5.

5 Ann. and Mag. Nat. Hist., ser 4, vol. xvi. p. 193. " Zeitschr.f. iciss. ZooL, Bd. sxxiii. p. 34. '' Ibid., Ed. x.xxv. p. 92

30 THE VOYAGE OF H.M.S. CHALLENGEE.

proceeding would have been as natural and comprehensible bad it taken place before the
classical investigations of F. E. Schiilze were published, as it is strange'now, Dr. Marshall's
paper in question having been issued in the year 1881. Dr. Vosmaer^ places this genus
in his and F. E. Schulze's family Hircinidse, another proceeding which would be quite
inexplicable, since the species of Oligoceras hitherto described is entirely devoid of any
filaments, had it not a very simple explanation, that of an erratum. Schulze himself lays
great stress on the close affinity of Oligoceras with Cacospongia, and indeed, since in its
somewhat absolute character (I speak of that of the canal system) the genus thoroughly
agrees with other Spongidse, and differs from Cacosjpongia only by secondary and therefore
extremely conditional characters, its natural systematic place is near Cacospongia. The
differentiating characters above mentioned are the foUov/ing : (l) the tendency to take
foreign bodies into the parenchyma, and particularly on to the external surface ; (2) the
want of a proper network of horny fibres, the horny substance being developed so scantily
that portions of the body as large as peas are completely devoid of any skeletal fibres ;
(3) the structure of the skeleton, its fibres being overcharged with foreign enclosures,
and the skeleton on the whole being represented by isolated fibres which have only rare
anastomoses and ramify widely like the antlers of a stag. Do these characters together
justify the estaljlishment of a genus, even from the naturalist's point of view, not demand-
ing for generic distinctions differences of an absolute nature 1 I see, logically at least,
no grounds for answering this question in the negative, since one may regard the
Oligoceratidse as a group of forms with a tendency to lose the horny skeleton entirely
in order to become Myxospongida3, or at least analogues of Psammopemma among the
Spongidse. Oligoceras has accordingly the same right to exist as a genus as Cacospongia
or Hippospongia, each personifying a new principle, all being connected amongst them-
selves by all possible intermediate stages. From this point of view I should be obliged
to adopt F. E. Schulze's genus in question ; I cannot, however, do so on account of the
unusually conditional nature of the characters distinguishing the Oligoceratidse, apart
from the point that the transformation of a true Cacospongia into a not less typical
Oligoceras appears to be very easily realisable (eomp. p. 84). The characters separating
Cacospongia and Euspongia from one another are also conditional, but in this latter case
at least a conventional boundary is admissible. We can, if necessary, group in Euspongia
forms with fibres not thicker, and with meshes not larger, than a given dimension, the
forms with larger meshes and thicker fibres being grouped in the genus Cacospongia.
But even a similar, quite artificial boundary is not applicable to the distinctions between
Cacospongia and Oligoceras. All Spongidje take foreign bodies into the parenchyma as
well as into the skeletal fibres, and F. E. Schulze " himself warns us not to ascribe to this
character too great a significance. But apart from this, even did the taking in of foreign
bodies represent the manifestation of an " unknown intellectual power," and were their

' On Velinea gracilis, p. 445. • Zeitschr.f. wiss. Zool., Bd. xxxiii. p. 14.

REPORT ON THE KERATOSA. 31

number and quality tlius characteristic of every sponge and not dependent on the
peculiarities of the surrounding soil, it would still be practically impossible to
distinguish Cacospongia from Oligoceras by these characters alone. And the
third additional character of the latter would be also of very little assistance.
The skeleton of Cacospongia is represented by a continuous network of fibres ; that
of Oligoceras is interrupted by spaces of parenchyma devoid of any skeleton, so
that, on the whole, its chief fibres show a tendency to free themselves, to separate
from one another in order as is the case with Aplysilla, and to form tree-like forma-
tions. But this character, or rather tendency, is common to all Spongidre with
skeletal fibres overcharged with foreign enclosures ; in my Cacospiongia spinifera and
Cacospongia tuberculata, amongst fibres forming obviously a continuous network, I find
fibres whose secondary ramifications do not reach the neighbouring primary fibres, so that
a small tree is actually formed. In Spongelia spinifera F. E. Schulze^ found a form with
a still more pronounced tendency in its skeletal fibres to ramify, to lose connection with
one another in order to form small tree-like structures. He did not, however, create
for it a special genus. Again, as before alluded to, the conjectural generic characters
of Oligoceras seem to be of a very unstable nature, and this is the second ground why
the genus Oligoceras should not be adopted even provisionally.

The genus Halisp)ongia, Bowerbank, being according to 0. Schmidt identical with
his Cacospongia, the genus Ditela, 0. Schmidt, having been given up by Schmidt himself,
who pointed also to the necessity of the same proceeding with respect to the genus
Auliscia, Bowerbank, there remain only the genera Hircinia and Ceratella to be
mentioned, since the names Stematumenia, Bowerbank, Polytherses, Fonbressin and
Michelotti, and Filifera, Lieberkiihn, are synonyms of Hircinia. The subgenus
Sarcotragus, established by 0. Schmidt in the year 18G2, was abandoned by him in the
following year.

As to the genus Hircinia, the reasons why I cannot adopt the family Hircinidse
in the sense of F. E. Schulze and Vosmaer have been stated in the foregoing pages
(pp. 12-14), and indeed the grounds above mentioned which forbid us to use the
presence of parasites in order to characterise the family, also forbid the use of this
character for purposes of generic distinction. It is not without interest that amongst
the Challenger specimens I have forms attached by filaments, some of which, according
to their other properties, I must group in the genus Stelospongos, others in the genus
Oligoceras, had it been retained, and others in the genus Cacosp)ongia. Should we
adopt the name Hircinia for forms with very large meshes and with fibres overcharged
with foreign enclosures as Carter and Hyatt have done ? I think this would be a very
doubtful proceeding ; the above characters are also common to Oligoceras, and we have

1 Zdtschr.f. wiss. Zool., Bd. xxxii. p. 152.

32 THE VOYAGE OF H.M.S. CHALLENGER.

seen that this genus cannot be adopted. The best thing that could be done would
consist in a total disuse both of Oligoceras and Hircmia.

Finally, as to the Ceratelladse, I regard their nature as sponges insufficiently proved to
speak here of their place in the system. The late Dr. Gray/ to whom we owe the first
description of these interesting beings, though classing them among the sponges, did so
with the greatest reserve, " until their organisation and growth is known. At the same
time," he adds, "I know no group of sponges with which they can be compared" {loc.
cit, p. 576). In his paper on the "Transformation of an Entire Shell into Chitinous
Structure," " Carter comes to the conclusion that the Ceratelladse have absolutely nothing
to do with sponges. He adopts both the genera distinguished by Gray in his family of
CeratelladEe, but he groups them in that of Hydractinise. Hyatt ^ is again of quite a
difierent opinion on the matter. He considers the forms in question to be indubitable
sponges, and, rejecting Gray's genus Deliitella, unites all the Ceratellada^ known to him in
the single genus Ceratella. Now, so far as my own information goes — and like that
of the naturalist before mentioned it is limited to the properties of the skeleton
exclusively — I must confess that I can only agree with Mr. Hyatt. There is really in
the skeleton of Ceratella nothing hydroid or coral-like ; " on the contrary, the whole is
thoroughly sponge-like." But this opinion can only be scientifically endorsed after an
investigation of the structure of the soft parts of the animals in question, and, until we
are acquainted -ndth their internal organisation, any discussion as to their systematic place
would be premature.

Family Aplysinid^.

Apart from the genus Dendrosjjongia, Hyatt, which, if really to be adopted as an
independent genus, may belong to the Darwinellidae, the family in question contains the
genera Luffaria, Verongia, and Aplysina.

Lvffaria.

This generic name originated with Duchassaing de Fonbressin and Michelotti,^ who
have described under this designation numerous sponges from the Caribbean Sea. How
far these sponges really belong to Luffaria, and not to Verongia or even Aj^li/sina, is
very difiicult to say, the descriptions of these authors being very superficial. As a matter
of fact, it must be stated that they regard the Spongia Jistnlaris, Linue- — illustrated in
Esper's Pflanzenthiere (Bd. ii. pis. xx., xxi., xxIa.), and the specimen represented on
pi. XX. is still in the Museum of Erlaugen — as belonging to their newly-created genus

' Proc. Zool. Soc. Lond., 1868, p. 575. ^ Ann. and Mag. Nat. Hist., ser. 4, vol. xi. p. 1.

3 Eeyision, &c., ii. p. 550. < Spongiaii'es, &c., p. 59.

REPORT OX THE KERATOSA. ' S3

Luffaria. Now, since for forms like Spongia fistularis the name of Verongia had been
proposed by Bowerbank many years before, Hyatt, in the year 1875, rejected^ the name
Ltiffaria as identical with Verongia. It may be so with respect to the Lttffaria of de
Fonbressin and Michelotti,but Mr. Hyatt was wrong in not paying attention to the proceeding
of 0. Schmidt,^ who in the year 1870 adopted Lx(ffaria, and furnished the genus with a
tolerably distinct diagnosis. He states that Luffarice are sponges with skeletal fibres
resembling those of a Cacosi-yongia but still differing from them chiefly in three respects :
— by a glass-like ("glasig") character of the fibres, by their faculty of splitting easily,
and by the possession of a narrow central canal, not identical, however, with that of the
true Aplysinidse. This latter statement Schmidt illustrates by a drawing representing a
fibre of a Liiffaria highly magnified. To these three characters 0. Schmidt appends a
fourth concerning the structure of the network formed by the fibres. He ascer-
tains that the fibres of Luffarice are approximately all of the same size, i.e.,
comparatively thick, and that the irregular network formed by them terminates peri-
pherally not with an even surface but with a surface roughened by many prominent
fil^res, so that a portion of the skeleton if devoid of soft parts would represent
something comparable to a brush. In the Challenger Collection I find two specimens,
the properties of whose skeletal fibres agree with those of Luffarice as described and
illustrated by 0. Schmidt. I found, however, that the network of their skeletal fibres,
though ending peripherally in projecting fibres (PI. IX. figs. 2 and 4), is composed not
of fibres of one kind, as suggested by Schmidt, but of two kinds, of larger fibres all
approximately of the same thickness and of smaller ones originating from the first men-
tioned, but differing from them not only in diameter but also in their histological
structure, the central canal of many of these smaller fibres having been found to
be inconspicuous. Whether my specimens are to be regarded as deviating forms, or
whether the statement of 0. Schmidt was based on an insufficient study of the skeleton
of the Liffarice which he had for examination, I am not prepared to say. I must add,
however, that this last supposition seems to me to be very plausible, not only because the
later spongiological papers of 0. Schmidt do not show that attention to practical
details so characteristic of his " Spongien des adriatischen Meeres," but also because a
portion of the skeleton at least of the Challenger Luffarice, when washed and dried,
demands a microscopic examination in order to prove the presence of finer fibres, these
latter being not only of a paler colour than the larger ones, and covered by them almost
throughout, but also very scantily developed near the external surface. At any rate,
I see no grounds for rejecting the genus in question. Thanks to the great amiability of
Prof. Selenka of Erlangen, the type specimen of Spongia ( Verongia) fistularis, Esper,
has been placed at my disposal, and thus I am able to state that the difference between
Luffaria and Verongia — which latter genus differs from Aplysina only by the compara-

1 Eevision, &c., vol. i. p. 401. ^ Spong. d. atlaiit. Gebiet., p. 30.

(zooL. CHALL. Exr. — PAKT s?aci. — 1884.) Hh .5

34 THE VOYAGE OF H.M.S. CHALLENGER.

tive thickness of the horny walls of its fibres — is greater than that between Verongia
and A'plysina.

Verongia.

On page 7 I have stated the grounds which make it, though not very probable,
yet not impossible, that the Lvffari(S have perhaps nothing to do with AplysinidfB,
being only specifically modified Cacospongm. At any rate — though bearing in mind
the necessity for certain concessions to the present state of our knowledge, we must
necessarily place the genus in the family Aplysinid^ — the genus, owing to the fact
that many of the smaller fibres of the skeleton of some of its representatives must
be called homogeneous, is so very closely connected with true Cacospo7igi(B that its
diagnosis can only be of a very conditional character. The same must be said with
respect to the mutual relations of the genera Verongia, Bowerbank,^ and Aijlysina.
Hyatt characterises his Aplysinidse " by the regular net-like anastomosis of the fibres,
the tendency of this to occur in the same plane, the flatness of the fibres, and the thinness
of their walls." ^ He characterises his Dendrospongiadse, of which the genus Verongia
is a representative, " by the irregular anastomosis of the fibres of the skeleton, by their
rotund form, and by the thickness of the horny walls." ^ As to the regularity or irregu-
larity of the network of the skeletal fibres, the conditional nature of this character is but
too evident ; — it is well known what approximate expression geometrical outlines find
in organised beings ; again, as to the flatness of the fibres of Hyatt's Aplysinidre, this
character seems even more doubtful. F. E. Schulze has not only ascertained that the
fibres of Aplysina acroplioha are all more or less mathematically cylindrical and their
transverse section circular, but has also made it probable that the above-mentioned state-
ment of Hyatt is due to the circumstance that the specimens he had for examination were
dried, in which state all thin-walled round tubes filled with a fluid mass shrink, and form
compressed tubes with transverse sections of elliptical outline.* Finally, so far as the
comparative thickness of the fibre-walls is concerned, many of the foregoing lines have
been written precisely to show that there are thick-walled fibres which show a tendency
to become thin-walled, and again, in the Aplysince, fil:)res with the contrary tendency ;
we see therefore that the three characters in cpiestion — provided that Hyatt is correct as
to the conjectural flatness of the skeletal fibres of the Aplysinidce — are of a thoroughly
conditional nature.

Ap)hjsina.

The diagnosis of this genus having been already given when speaking of the j^receding
genus, all existing genera of the Keratosa have been presented to the reader, and I

' Monogr. ot Brit. Spong., vol. i. p. 209 ^ Revision, &c., vol. i. p. 404.

2 Revision, &c., vol. i. p. 400, * Zeitschr. f. vdss. Zool., Bd. xxx. p. 39!).

REPORT ON THE KERATOSA. 35

have only to summarise the foregoing observations. I do it in the following words : —
ivith the exception of the genera Darwinella, lanthella, and Psammopemma, all genera
are devoid of any properties separating them absolutely from one another.

The further conclusions deducible from this statement will be given in the last
part of this Report, for they will only be instructive when we shall have learned the
properties used at the present time in order to distinguish and characterise the species.
An abstract discussion of all such properties would lead me too far ; it is besides unneces-
sary, since the following chapter, devoted to the description of the forms brought home by
the Challenger Expedition, may serve as a better illustration of them than any however
detailed but abstract discussions. Some remarks with respect to the external character
of the following descriptions : as in my memoir on the Challeilger Calcarea, the reader
will not find any specific diagnoses ; I must confess I regard them in most cases as a loss
of space and time, and altogether superfluous, especially as nearly every species in the
Challenger Collection of Keratosa is represented by a single specimen only, so that the
tendency to vary could not be made out. As to this question in general, I refer the
reader to the extremely instructive paper of Heincke on the varieties of the herring ^ ; on
the other hand, I invite him to peruse the diagnoses of Calcarea given by Haeekel in
his " Kalkschwamme " for every species. Would he receive any idea of the animal from
a similar diagnostic description ? I think not. And this with respect to Sponges, the
geometrical properties of whose spicules present far more tenable systematic distinctions
than is the case with regard to the Keratosa. As to those described in this paper, I
regard their entire descriptions as diagnoses, and end this chapter with the following
observations of a practical nature.

All the specimens in the collection not devoid of soft parts have been examined with
regard to their canal system and skeleton. The skeletal fibres have been examined
both in spirit and mounted in Canada balsam, after previous treatment with ammonia,
in order to remove the soft parts. These latter have been examined in sections stained
in different ways by different staining fluids ; for it must be stated that while for
the Calcarea no other staining fluid but picro-carmine is to be recommended, the matter
is quite difterent with regard to the Keratosa, so that in each case the investigator must
proceed experimentally.

1 Die Varietaten ties Herings, Berlin, 1877.

III.-DESCRIPTION OF THE SPECIES.

Group KER ATOSA, Grant, 1861.

Ceraospongice, 0. Schmidt, 1862.

EuspongicB, Duchassaing de Fonbressin and Michelotti, 1864.

Ceratina and Psammonemata (e. p.), Carter, 1875.

Porifera with horny skeleton devoid of proper spicules.

Family Darwinellid^, Merejkowsky, 1878.

AplysinidcB {e. p.), F. E. Schulze, 1878.
Aphjsillince, v. Lendenfeld, 1883.
Aplysillidce, Vosmaer, 1884.

Keratosa with large pouch-shaped flagellated chambers communicating by means of
numerous pores on their walls with inhalent cavities, by means of one wide mouth with
exhalent cavities ; ground-mass without granules, transparent ; axis of fibres thick.

lanthella, Gray.

Veroiigia (e. p.), Ehlers.

Darwinellidoe, the fibres of whose skeleton contain true cells.
lanthella flciheU if or mis, Pallas, sp. (PI. I.; PI. II. figs. 1-7).

Spongia jiahelliformis, Pallas, Elenchus Zoophytorum, p. 380.
lanthella flalelliformis, Gray, Proo. Zool. Soc. Lond., p. 49, 1869.
Verongia flalelliformis, Eblers, Die Esper'schen Spongien, p. 11.

This species is represented in the Challenger Collection by two specimens both from
the Eastern Coast of Australia ; one of them may be seen drawn of natural size on PL I.
Their shape, in accordance with the specific name of the animal, is that of a thin leaf
terminating in a short stem ; which latter in both cases is of irregularly cylindrical outhne
and about 25 mm. long by 12 mm. broad.

The Skeleton. — The constituent elements of the leaf-like part of the skeleton admit

38 THE VOYAGE OF H.M.S. CHALLENGER.

of a distinction into primary fibres, taking their origin from the upper end of the stem
in order to assume a radial direction, as can be seen on pL xiii. of Esper's Pflan-
zenthiere, ii., and into secondary fibres uniting the primary ones with one another both
perpendicularly and parallel to the surfaces of the sponge. Both the surfaces are uneven,
and this is due to tlie circumstance that the secondary fibres, which are nothing but
lateral branches of the primary ones, interlacing amongst themselves, form prominent
irregular networks. The last-named are most fully developed near the upper end of the
stem, growing gradually lower towards the edge of the leaf -like part of the animal. The
external surface of this stem is comparatively smooth ; its skeleton admits of no distinction
into primary and secondary fibres, the whole presenting a very compact and solid network,
with meshes of smaller diameter than those of the leaf-like extension (PL II. fig. 3).
The fibres themselves, as is weU known, are heterogeneous and, on the whole, to be classed
as thick walled. The microscopical structure of the central pith-substance docs not differ
from that of Aphjsina, as described by F. E. Schulze ^ in the case of Aj^lysina aerojihoha,
nor do the walls surrounding this central differentiation, and presenting, as is always the
case, many layers of concentrically disposed horny laminae, with the sole distinction that
between these latter true cells are to be found in abundance. As to the histological
properties of these interlamiuar elements, I refer the reader to Dr. Flemming's ° paper on
lanthella, for I can confirm all his statements except as to the ends of the fibres being, as
he suggests, devoid of cells. For my own part, I could always discern them even on the
youngest fibres, viz., on the fibres with only a very thin horny envelope, and am able
to state that they are absent only at the ends of developing fibres, represented by
pith-substance surrounded immediately by spongoblasts. That the ceUs enclosed in the
walls are nothing but transformed spongoblasts is beyond doubt, but of course this will
be actually proved only when direct observations on the actual process of development of
the fibres in question have been made. For myself, amongst the elongated sj)ongoblasts, as
I have drawn them on PI. II. fig. 5, I have very often seen cells of a more massive and also
compressed form, and I believe these to be intermediate stages between the typical
spongoblasts and the interlamiuar cells. But I have not figured them, because their
deviating appearance may yet be ascribed to the influence of the preserving fluid.
Around the "mantle of spongoblasts" of young fibres I always found aggregations of
mesodermic cells, though by no means always disposed parallel to the developing fibre
as drawn and described by v. Lendenfeld ^ in his Bendrilla rosea and Dcnclrilla aerophoha.
Around old fibres I have missed them completely. According to Flemming, the colour of
the fibres of lanthella hasta, so fer as the horny substance of then." laminae is concerned, is
yellow, that of the enclosed cells deep violet. I have found in most cases the colour
of the laminae also to be violet. I am, however, inclined to ascribe this merely to the

' Zeitschr.f. wiss. Zool, Bd. xxx. p. 401, pi. .\xii. tig. 11. ^ Wiirshurg Verhandl., N. F., Bd. ii.

3 Zeitschr. f. wiss. Zool, Bd. xxxviii. p. 286 ; pi. xiii. figs. 25, 28, 29.

REPORT ON THE KERATOSA. 39

influence of preservation in alcohol. I think that the colouring substance, by which the
sj)ongol)lasts (if not all the cells of lanthella) are coloured in the living state, is soluble
in spirit, and might thus when dissolved have coloured the true horny substance also ;
for, while I found most of the fibres of a violet colour, and containing violet cells,
I found also fibres whose cellular elements were of a deeper colour than usual, and that of
the horny laminee considerably paler.

The Soft Parts. — As to the anatomy of its soft parts, the species differs but little
from AplysiUa, DendriUa, and HaUsarca ditjardini. Both the surfaces of the sponge
when examined with the naked eye show a great number of groups of openings, as
represented ou PL II. fig. 1, under a low magnifying power. These are the oscula, each
b eing usually represented by four or five apertures. The parts of the membrane
between them, when seen under the microscope, show themselves as usual to
be provided with numerous minute pores. The water entering through these porz
dermales reaches the cavities under the covering membrane, and passes from these latter
by means of pori camerales into the flagellated chambers, in order to be expelled
through a large mouth into exhalent canals, finishing with the oscula above mentioned.
Thus each osculum, with its pores and its subdermal or inhalent cavities, flagellated
chambers, and exhalent cavities, presents an independent whole, which may be compared
with a state in a federal republic. The inhalent and exhalent canals are of very irregular
outline ; they may be very large and short, very narrow and long, &c.

The form of the flagellated chambers is also very variable. Its mathematical mean
shape is expressed by the two flagellated chambers on PI. II. fig. 5. But though such
regular pouch-shaped flagellated chambers may be found without difficulty, the outlines
of most others are quite irregular (comp. PI. II. fig. 4). Sometimes they are cylindrical
and elongated, sometimes irregularly roundish, often provided with secondary ramifica-
tions ; their size is inconstant.

Now so far as the histological structure of the present species is concerned, it is
constant to the general type prevailing among the Keratosa, presenting, however, a few
new peculiarities. The ectodermic pavement-epithelium could be discerned only on the
surfaces of the inhalent canal system. My endeavours to make it out on the external
surfaces were unsuccessful, owing probaljly l^oth to the state of preservation and to
the fact that the external surfaces above mentioned are covered by a thin cuticle such
as Schulze^ has described in Cacospongia cavernosa. No distinction can be detected
between the ectodermic and endodcrmic pavement-cells. In all cases they are flat, irre-
gularly polygonal, with a comparatively small nucleus, and showing the protoplasmic
granules only around the nucleus. As to other representatives of the endoderm, viz.,
flagellated cells, they seem to agree as regards their form with those of Ap>lysiUa^- but it
must be noticed that the flagellated cells are very sensitive to every method of preservation,

1 Zeitschr.f. loiss. Zool, Bd. xxxii. p. 654. 2 ji,i£^ Bd. xxx., pi. xxiii. fig. 26.

40 THE VOYAGE OF H.M.S. CHALLENGER.

and that tlieir properties can be studied -^-ith success only in the living state. The
constituent parts of the mesoderm are as usual represented by a transparent ground-
mass, by stellate or fusiform cells and cellular elements of amoeboid character. The
amoeboid cells differ from the first-mentioned apart from their general shape by their
nucleus being comparatively larger. It is generally accepted that the amoeboid cells
in the sponges give origin to the generative products ; both the specimens of
lanthella proved, however, to be sterile ; in one of them, indeed, I found here and there
large egg-shaped bodies, but since I did not succeed in discerning in them anything like
a nucleus, I am far from being sure whether they were really ova or something else.
The stellate mesodermic cells, as well as those of fusiform shape, do not differ from those
of Sycon rciphanus as described by F. E. Schulze,^ but it must be said that whUe their
fusiform modification is very seldom found in Sycon raphamis and the Calcarea generally,
it is far more common in lantheUa than the stellate form. Particularly near the outer
surfaces these fusiform, probably contractile, cells are very numerous, surrounding in rows
the pores and oscula (PI. II. fig. 6). I have spoken of them in their special modification
as spongoblasts before, and it only remains for me to mention the interesting hypodermic
elements, without entering upon the discussion of the question as to whether they are
really modified stellate and not amoeboid ceUs. At any rate they are larger than both
the stellate and the amoeboid ones, and their protoplasm is far richer in granules. It is
indeed difficult, when seeing these elements Ipng separately amid fusiform muscle-ceUs
(as drawn on PL II. fig. 6), to resist the idea that these elements are of a nervous nature,
and their histological properties, so far as they could have been studied from the material
preserved in alcohol, agree tolerably well with what we regard as tjrpical nerve-ceUs. I must
say, however, that I was unable to discern any connection between them and the fusiform
ceUs, and on the whole consider theu" nervous nature to be as doubtful as that of certain
mesodermic cells described by SoUas - in Tlienea muricata, as well as that of the anastomos-
ing " Strange " discovered by F. E. Schulze ^ in the Spongidse. I believe these cells to be
equivalent to the gland-ceUs stated by v. Lendenfeld * to be present in his South Sea
Aplysinidce, and five years before by Merejkowsky* in his Halisarca (Oscarellaf) scJmlzei.
I have found similar cells in Ajilysilla sulj)hurea and Darwinella aurea, and I am the
more inclined to compare the enigmatic elements of lanthella with these gland-cells,
since, as I remarked before, the external surfiice of this sponge is covered by a thin
cuticle. Of course they arc larger than common spongoblasts, whUe the gland-cells of v.
Lendenfeld agree mth these latter both in shape and size, but this difference seems to me
to be of no great importance. The best methods for rendering these, as well as hypo-
dermic fusiform cells, visible are eosine and gold, for which latter the alcohol must be
previously extracted.

I Zeitschr.f. wiss. Zool., Bd. sxv., Suppl., p. 253. ^ /(,),;. and Mag. Nat. Hist, ser. 5, vol. ix. p. 44G.

^ Zeitschr.f. wiss. Zool., Bd. xxxji. p. 629. ■* Ibid., Bd. xxxviii. p. 278.

' Mem. Acad. Sci. St. Petersk, vol. xxvi. Xo. 7, pi. ii. fig. 9.

EEPORT ON THE KERATOSA. 41

Colour. — Violet.

HaUtat.—^taiioxi 188, September 10, 1874, lat. 9° 59' S., long. 139° 42' E. ; depth
28 fathoms ;, green mud.

Family Spoxgelid^, F. E. Schulze, 1878.

Dijsideidm, Gray, 1867, and Marsliall, 1881.

Arenosa, Carter, 1875.

Spmgiadce {e. p.), and Hireiniadce (e. p.), Hyatt, 1877.

Keratosa with flagellated chambers of more or less regularly roundish outlines,
communicating by means of numerous pores in their walls with inhalent, by means of
one wide mouth Tsdth exhalent, cavities ; ground-mass transparent, without granules.
Axis of fibres thin ; fibres cored in most cases with foreifrn enclosures.

Spongelia, Nardo.

Dysidea, Johnston.

Sponge! ia, Dysidea, and Psammasais, MarsliaU.

Spongelidse with large flagellated chambers ; outer surface proYided with conuli.

Spongelia spinifera, F. E. Schulze.

Spongelia spinifera, F. E. Schulze, Zeitsckr. f. wiss. Zool, Bd. xxxii. p. 152, 1878.

This species, established by F. E. Schulze in the year 1878 for some forms from the
Adriatic, is characterised by him as follows : — " The conuli, 5 to 8 mm. high, situated at
equal distances from one another, run out in simple thorn-like processes ; among the
simple ramified primary fibres no communicating secondary fibres are to be found, so that
accordingly the skeleton forms no network." The chief character, since the size of the
conuli is in other representatives of the genus variable, is thus the dendroid ramification of
the skeletal fibres which do not anastomose with one another. In this chief character both
the Challenger specimens, which I have determined as Spongelia spinifera, agree ; but
while one, diff"ering from the Adriatic forms as regards its lower conuli, agrees with them
in its mode of growth, being found like them in the form of a crust, the second specimen
difi'ers from them even in this latter point, presenting a laterally compressed leaf 25 mm.
high, 20 mm. broad, and 5 mm. thick in the middle, and rather thinner near the border.
The conuli of both the specimens do not exceed 4 mm. in height, and in accordance with
this peculiarity I propose to establish for them an independent variety — p>arviconulata,
while the designation magniconulata might be used for forms like the Adriatic Spongelia
spinifera. No peculiarities of consequence were noticed with regard to the structure of

(zool. CHALL. EXP. PAKT XXXI. — 1881.) Eh 6

42 THE T^OYAGE OF H.M.S. CHALLENGEE.

the soft parts ; the osciila, scanty in the incrusting specimen, proved to be thoroughly
deficient in that of leaf-like shape ; one of the specimens proved to be sterile ; the other
was provided with numerous sperm-balls. The habitat of Spongelia sjnnifera, hitherto
found only in the Adriatic, is now extended to the east coast of Australia.

Colour. — Pale greyish and dirty yellowish.

Habitat. — Off Port Jackson, depth 7 fathoms.

Spongelia pallescens, 0. Schmidt (PI. III. fig. 1).

This species, whose detailed definition we owe also to F. E. Schulze,^ has been divided
by him into two subspecies, and each subspecies in its turn split into two varieties. The
numerous Challenger specimens of Spongelia pallescens are all to be determined as sub-
speciesfragilis var. ramosa, but I must add that while the diagnosis of this variety as
given by F. E. Schulze is thoroughly applicable to the Challenger specimens, the meshes
of their skeletons proved to be of rather larger diameter than those of any of the
Spongelidse with anastomosing fibres examined by F. E. Schulze, and whose skeletons,
through his kindness, have all been placed at my disposal. Whether this diSerence is of
individual or varietal or even specific value, I am unable to decide, and therefore limit
myself to merely mentioning it.

Colour. — Pale yellow and lilac.

Habitat. — Bahia, shallow- water.

Spongelia horrida, Selenka (PI. III. fig. 2).

Spongelia horrida, Seleuka, Zeitschr. f. wiss. ZooL, Bd. xvii, p. 566, pL xxxv. figs. 1-4, 1867.

This species, placed by Selenka in the genus Spongelia, has also been adopted by
F. E. Schulze,^ but apparently only on account of its form and the properties of the
skeleton. I am pleased to be able to sanction this proceeding, as the external shape and
structure of the skeleton of the Challenger specimens agree closely with those of the
forms described by Selenka, and their internal organisation is that of a typical Spongelia.
The identification of the forms in question is, on the other hand, confirmed by the fact
that both the Challenger and Selenka's specimens were from Australia,

The Challenger Collection possesses three specimens of Spongelia horrida; two are
in the form of crusts 12 to 18 mm. high, the third being of a more massive appear-
ance. But all three specimens are equally constant to the main character of the
species, the meshes of their skeleton being far larger than in any other Spongelia, as may
be seen on PI. III. fig. 2. The foreign enclosures of the fibres proved to be chiefly

I Zeitschr. f. wiss. ZooL, Bd. xxxii. p. 154, 1879. ^ Ihid., Bd. xxxii. p. 122.

REPORT ON THE KERATOSA. 43

fragments of spicules and small sand-grains ; tlie capsules with cellular elements found
within the fibres by Selenka i^oc. cit., p. 566, pi. xxxv. figs. 3-4) were not to be seen.

Colour. — Pale yellowish.

Habitat.— Station 186, September 8, 1874, lat. 10° 30' S., long. 142' 18' E.; depth
8 fathoms ; coral mud. ,,

Psammoclema, Marshall.

Spongelidas with small flagellated chambers ; external surface smooth.

Psammoclema ramosum, Marshall (PI. III. fig. 8 ; PI. IV. fig. 1).

Psammoclema ramosum, Marshall, Zeitschr. f. wiss. Zool., Bd. xxxv. pi. vii. fig. 12, 1880.

It is sufficient to compare the drawing of this form given by Marshall with that
given by me on PI. IV. fig. 1, in order to become persuaded of the fact that both the
Challenger and Dr. Marshall's specimens belong to the same sj)ecies. There are, however,
in the description which Dr. Marshall gives of his specimens, two points at variance with
my observations, but I am inclined to explain one of them by a mistake on the part of Dr.
Marshall, and the other by a real difi"erence in organisation but of no essential nature.

It has been stated by Marshall that the skeleton of his Psammoclema ramosxim
is tree-like, the main fibres sending lateral branches which ramify in their turn but form no
anastomoses. I can state the same with respect to the xipper ends of the branches of the
animal, but so far as other parts of its body are concerned, I discerned the anastomoses
clearly (PI. III. fig. 8). Again, Dr. Marshall believes the canal system of this species
to be of special interest. He says : " Die Munddfi"nungen befinden sich alle auf einer
Seite, wie bei manchen Formen von Halichondria oculata, Grant, und bei Veluspa
polymorpha var. digitata, Miclucho. Es scheint dies dafiir zu sprechen, dass der Schwamm
nicht aufrecht wachst, sondern wie viele Pflanzen im Meere, horizontal. Ein aufrechtes
Wachsthum diirfte wohl auch schon bei der Schwere des Schwammes seiner geringen
Festigkeit und Widerstandsfahigkeit gegeniiber unmoglich sein."' Indeed, on some
branches of the Challenger specimen the oscula were all found on the same side, but on
others they were scattered everywhere, and having examined the external surface of the
specimen, I found all its parts equally smooth and clean. I must therefore difier from
Dr. Marshall with regard to his conclusion as to the kind of o-rowth of the animal. I think
it grows vertically, and am inclined to believe that the peculiarity in the disposition of the
oscula on Dr. Marshall's specimens is characteristic only of the individual. For a more
detailed description of the animal, I refer the reader to the paper of Dr. Marshall above
mentioned, and will only add that its flagellated chambers agree closely as regards form
and size with those of my Psammoclema vosTuaeri (PI. III. fig. 6), and that, on the

1 Loc. cit., p. 112.

44 THE VOYAGE OF H.M.S. CHALLENGER.

whole, its internal organisation does not differ from that of the representatives of the
genus Spongelia, except in the size of the flagellated chambers. What Dr. Marshall
means by the words, "Das Gastrovascular-System ist hier nach dem astformigen Typus
ano-eordnet" {loc. cit, p. 113), is of course to be regarded as an echo of corresponding
erroneous statements of Prof. Haeckel. For any fine histological examinations the
Challeno-er specimen proved to be insufficiently preserved. The Alga found by Marshall
in all the specimens he had for examination was found to be characteristic also of the
single specimen collected by the Challenger.

Colour. — Grey, skeletal fibres white.

Habitat. — Station 162, April 2, 1874, off East Moncoeur Island, Bass Strait; depth
38 fathoms ; sand and shells. Dr. Marshall's specimens were also obtained from Bass
Strait.

Psammoclema vosmaeri, n. sp. (PL III. figs. 5 and 6).

This interesting species has been found in one colonial specimen, which is depicted of
natural size on PL III. fig. 5. Some individuals of the colony are mouthless, the oscula
of others are very small, and others again present cylindrical tubes, the diameter of the
oscula being thus the same as that of their body in its different parts.

The surfaces of the sponge though entirely devoid of any conuli, are still rather
uneven and rough, but whether this is due to the state of preservation or is characteristic
of the species I cannot judge. The skeleton, when seen from the outer surface, admits
of no distinction into primary and secondary fibres, the network of the fibres being-
very compact, i.e., the meshes of a very small diameter, and it is only on the inner
surface that the vertically directed primary fibres, 0"16 mm. thick on an average, are easily
distinguishable ; most of the fibres, and particularly the larger ones, are charged with
foreign enclosures to such a degree that the enveloping horny substance is very thin
and the surface of such fibres very rough. When dried the skeleton becomes hard
and of a greyish colour. The type of the canal system and the histological structure
do not differ from those, for example, of SiMugelia pallescens, except that the flagellated
chambers are of smaller dimensions and comparatively more constant in their roundish
form, and that the internal cavity in some individuals is very large. Through the pores
of the outer surface the water reaches the system of inhalent canals, flagellated chambers
and exhalent lacunse opening by comparatively small holes into the internal cavity.

From all other representatives of the genus hitherto known this species can be readily
distinguished both by its external shape, which is more like that of a calcareous than a
horny sponge, and by the compact network of its skeletal fibres, also perhaps by the
tendency to form colonies.

I dedicate the species to Dr. G. C. J. Vosmaer of Naples.

REPORT ON THE KERATOSA. 45

Colour. — Pale greyish-yellow ; skeletal fibres dirty greyish.

i/'a^^to^.— Station 18G, September 8, 1874, lat. 10° 30' S., long. 142° 18' E. ; depth
8 fathoms ; coral mud.

Psammoclema foliaceum, n. sp. (PL III. fig. 7).

Like Psammoclema ramosum, the species in question was brought home in fragments,
all, however, in both cases belonging to one specimen. When put together, the fragments
of the Challenger Psammoclema foliaceum form a whole which may be compared with my
drawing of Carteriospongia radiata, but which is approximately four times as large, with
walls from two to four times as thick in different parts of the body, and with surfaces
of a rather shagreen-like appearance and not distinguished by the system of longitudinal
ribs. The resemblance in the form ; the fact that, as regards the anatomical structure of
the canal system, Psammoclema is closely allied to Carteriosjjongia ; and again, that most
of the representatives of this latter genus also possess fibres full of foreign enclosures ; all
render the form I am now describing extremely interesting from a phylogenetic point of
view. This has led me to make greater endeavours to find out its histological structure
in the hope of discovering in it the anastomosing "Strange" of F. E. Schulze, the function
of which is of course doubtful, but which a.re so very characteristic of Carteriospongia
and of all the Spongidre generally. The specimen proved, however, to be so very badly
preserved that I had to be content with making out its anatomical organisation, and thus
assigning to it a definite systematic position.

A portion of its skeleton is represented on PL III. fig. 7 ; like that of Psammoclema
vosmaeri, it is of a dirty-greyish colour, due to the abundance of foreign enclosures in its
fibres, but the meshes are larger, and when seen against the light the primary fibres
become very clearly visible. I characterise the species exceptionally by its external
shape, but of course I am uncertain whether I am right in this proceeding. Later
investigations wiU decide the question.

Colour. — Black ; skeleton grey.

Habitat. — Station 162, April 2, 1874, ofi" East Moncceur Island, Bass Strait; depth
38 fathoms ; sand and shells.

Psammopemma, Marshall.

Spongelidse without any differentiated skeletal fibres, the supporting skeleton being
represented by foreign enclosures lying separately in the parenchyma, and the secretion
of the horny substance having been reduced to the formation of only a thin envelope
around the enclosed foreign bodies.

46 THE VOYAGE OF H.M.S. CHALLENGEK.

Psammopemma densiim, Marshall (PI. III. figs. 3, 4).

Psammopemma densum, Marsliall, Zeitschr. f. wiss. Zool., Bd. xxxv. p. 116, 1880.

When characterising this species, Dr. Marshall did not feel quite certain whether he
was really describing a sponge and not a worm-tube or something of that kind. Of
course, I have scarcely the right to express an opinion on this delicate question, still I
believe it to be a sponge; at any rate I can state wdth the greatest certainty that such sponges,
i.e., sponges ^dthout any differentiated skeletal fibres but still secreting horny substance,
do really exist. Sanctioning thus the establishment of the genus Psaimnojxmma, I
adopt also Dr. Marshall's species Psammopemma densum, since its specific designation
is very characteristic of two specimens in the Challenger Collection, and especially of that
from Port Jackson.

This last specimen is represented in PL III. fig. 3, and it wdl be obvious from this
drawing that the original is probably not the whole animal but only a fragment of it.
Now even if this fragment give no jjrecise idea as to the shape of the whole animal, it
must still be assumed that this latter was of a plate-Hke compressed form, supposing
the fragment in question was not merely an outgrowth ; while the external shape of
the second Challenger specimen, represented also only by a fragment, must have
been either crust-like or massive, the plane of the fracture being parallel to its outer
surface.

So far as this latter is concerned, it is in both specimens uneven and throughout its
whole extent rough, owing to prominent sand-grains, often 1 mm. in diameter. In the
Australian specimen, at one point on its surface, I found a shallow dej)ression, which I am
inclined to regard as the osculum ; but I refrained from trying to prove it by immediate
dissection lest I should destroy the specimen in vain, the sponge presenting such a com-
pact aggregation of sand-grains that only very thick sections could have been obtained
from it by the microtome. Besides, the question is of little consequence.

In contrast to the specimens which Dr. Marshall had for examination, both the Chal-
lenger specimens proved to be free from any parasitic inhabitants, as well as from any
skeletal fibres, either foreign or produced by the sponge itself. Nor can I say that the
foreign enclosures are held together by protoplasm as stated in Marshall's definition of the
genus (loo. cit, p. 113); each of them, though surrounded by a thin horny envelope, having
been found lying separately in the parench^ona. That the envelope just mentioned is of
a horny nature I judge from the fact that it does not differ from the envelopes to be
found around foreign enclosures in other Spongelidse, lying free in the " sarcode," these
latter in their turn not differing from the envelope of true skeletal fibres overcharged
with foreign bodies. Contrarily to Marshall {loc. cit., p. 114), I find this envelope to be

EEPORT ON THE KERATOSA. 47

quite devoid of any enclosed corpuscles. As to the foreign enclosures themselves, they
are represented in both the Challenger specimens almost exclusively by sand-grains, some
of Dr. Marshall's specimens having proved to be also very rich in them, while others, on the
contrary, were poor in sand and rich in fragments of mussel-shells. I agree entirely, how-
ever, with Dr. Marshall as to the impossibility of paying any systematic attention to such
differences. Of course the faculty has been ascribed to sponges of choosing from the avail-
able foreign bodies those which they need. Haeckel adopts it with respect to his Physe-
maria;'^ and Carter, though on a different occasion,'' speaks also of " that developmental
intelligent power whose existence in every organised product is only known to us by its
manifestations." However, the contrary opinion, held by F. E. Schulze and Marshall, is
supported by more valid arguments, and there is absolutely no necessity for introducing
into our scientific calculations a new thoroughly unknown factor, while the phenomenon
admits of a very simple and plausible mechanical explanation.

So far now as the anatomy of the form in question is concerned, Marshall believes the
Psammopemmata to be of very low organisation, and if I understand him aright, he sees
it in their lipostomy and lipogastry. These two peculiarities, provided that the lipostomy
be really characteristic of the genus, are, however, of very subordinate significance ; a
lipostomic and lipogastric Leuconid is yet more highly organised than a Sycon provided
with the broadest central cavity and with an osculum fringed with the most elegant
spicules, — such questions, without knowledge of the structure of the canal system, not
being at all capable of solution. That of the Psammopemmata agrees in its features
closely with that, for instance, of Sp>ongelia pallescens, as described by F. E. Schulze.
The flagellated chambers are large, of more or less regularly roundish outlines, and devoid
of any special cameral canaliculi ; the ground-mass surrounding them is transparent and
without granules (PI. III. fig. 4). There is also a close resemblance to the representatives
of Spongelia in the histological properties, the only difference being that such aggrega-
tions of fusiform cells as Schulze ^ describes for Spongelia avara are not to be found
in the two Challenger specimens of Psammopemma. In this statement I differ from
Marshall, who lays stress on their constant occurrence in (under ?) the covering dermis.
I find these fusiform cells scattered everywhere and also under the dermal membrane,
but in most cases lying isolated, and never in such mutually parallel disposition as in
Marshall's illustration {loc. cit., pi. viii. fig. 10). Nor can I agree with him as to this
dermal membrane being homogeneous {loc. cit., p. 113). I found it to contain nuclei
tlisposed at approximately equal distances from one another, and on the ground of
numerous analogies I am inclined to regard it as a common pavement-epithelial layer,
the boundaries of its separate ceUs having disappeared owing to the preservation in
alcohol.

1 Biolog. Studien, Heft. ii. p. 213.

2 I refer to his discussion of the process of the horuy skeletal fibres taking in and enclosing the foreign bodies,
Ann. and Mag. Nat. Hist., ser 5, vol. viii. p. 113. ' Zeitschr.f. wiss. Zool., Bd. xxxii. p. 136.

48 THE VOYAGE OF H.M.S. CHALLENGER.

The reader sees that there are many points in which I disagree with Dr. Marshall
Most of the contradictions are, however, insignificant, and the remaining ones, I feel
certain, are to be explained by the fact that Dr. Marshall's specimens were either in a
dried state or very badly preserved. At any rate, I see no reason against adopting both
the genus and the species.

Colour. — Sandstone-yellowish.

Habitat.— ^t&tion. 49, May 20, 1873, lat. 43° 3' N., long. 63° 39' W.; depth 85
fathoms ; gravel and stones. Ofi" Port Jackson, 7 fathoms.

Psammopemma porosum, n. sp.

In his paper on systematic rules Strickland suggests that generic and specific
designations should be given which allude to the main generic or specific character of the
given form ; — very good counsel indeed, yet it occurs but too often that whilst following
it the classifier creates very unfortunate systematic designations, and so far as the
species I am going to describe is concerned, I ieel by no means certain that — just as we
find AncorincB without any anchor-like spicules — specimens of Psammopemma porosum
will be found of no less density than that of the Challenger specimen of Psammopemma
densum from Port Jackson. Still I do not take refuge in naming the species in question
according to its second and undoubtedly more important character, for this latter character
may prove to be of a subgeneric consequence. The species is represented in the collection
by numerous small fragments, which, formless as they are, show distinctly that the
external shape of the specimen — or, perhaps, specimens — dredged by the Challenger must
have been of a rather stout, massive appearance. Apart from this character as well as
from the porosity, which latter peculiarity may stand in connection with the massive
shape of the animal, the fragments in question recall vividly as regards their form the
Challenger specimens of Psamviopemma densum, and especially that from Nova Scotia.
Their colour is a dirty sandstone greyish, and the presence in the parenchyma of foreign
enclosures (chiefly sand-grains, and, only in the upper skin, fragments of spicules) is well
marked in the roughness of the surfaces.

As I remarked before, one of the characters of the species may be its porosity. I have
but to add that this peculiarity must be really regarded as characteristic at least of the
specimen, for there is no reason to suggest that the parenchymal cavities in question are
due to worms as in the specimens of Psammopemma densum examined by Dr. Marshall.
Except a Hydroid {Agl(eop>henia, sp. ?) rooting in one of the fragments and the doubtful
round bodies to which I shall refer later, I miss any parasitic inhabitants entirely.

The second peculiarity characterising this form is the minuteness of its flagellated
cells, which are smaller than in any horny sponge hitherto described. It is only in some
Corticatte (StcUetta gruhei, Geodia gigas, Ancorina aaptos, Tethya lyiicurium) that I

REPORT ON THE KERATOSA. 49

have seen these elements so strikingly minute. This peculiarity renders the sections of
them readily distinguishahle under the microscope from those of any other Porifera, and my
first idea when examining the preparations of Psammopemma porosum was that this form
might represent an offspring of the family above mentioned. Yet an attentive examination
showed that the species has even more right to be referred to the Keratosa than Psammo-
pemma densiim ; for while in this latter form the horny substance enveloping the foreign
enclosures has been found in the form of an extremely thin layer, so that its presence
beyond doubt could only be proved in most instances after treatment with hydro-fluoric
acid, here in Psammopemma porosum it proved to he far more fully developed, and
occasionally with very conspicuous outgrowths.

On p. 17 1 have shown that on the whole, as the canal-system reaches a higher
degree of development, the size of the flagellated chambers gradually diminishes. Of
course the flagellated chambers of Psammoclema vosmaeri are not larger than those, for
instance, of Euspongia officinalis, and yet they are devoid of special cameral canaliculi,
but every rule has its exceptions. On the ivhole, the above dependence is still to be
observed, and it is not without interest that as flagellated chambers become smaller
and smaller the flagellated cells in their turn grow gradually more and more minute.
The flagellated cells of Apli/silla sulphurea are larger than those of Spongelia elegans,
which again are larger than the flagellated cells of Cacosp>ongia scalar is or Aplysina
aerophoba. But, as in the preceding case, exceptions are not wanting here also ; lanthella
flabelliformis does not deviate in the structure of its soft parts from the type
characterising Aplysilla sulphurea, and yet possesses flagellated cells of far smaller
dimensions than those of the form just mentioned. Again in Euplectella aspergillum,
in spite of its large, radial tube-like flagellated chambers, these cells are very minute.
Psammopemma porosum presents, in this respect, the most interesting exception. As
stated before, its flagellated cells are very small ; while for instance in Cacospongia
scalaris the diameter of their transverse section is 0"0025 mm., on an average in
Psammopemma porosum it does not exceed O'OOl mm. ; and yet its flagellated chambers,
although smaller than those of Psammopemma densum, are rather larger than those of
Cacospongia scalaris or Euspongia officinalis. The explanation of this curious deviation
must be of course left to later investigations, but there can be scarcely any doubt that,
if not even of a subgeneric importance, at any rate it necessitates the establishment of a
new species, although on the whole the internal organisation of Psammop>emma porosum
agrees closely with that of Psammopemma densum. As a probably accidental
peculiarity I can notify the presence in the parenchyma of numerous round bodies, with
an average diameter of 0'012 mm., which at first sight recall vividly the spermospores of
Calcarea or Aplysinidae (p. 72). Under high microscopic j)ower these bodies proved
however to be devoid of any covering cell, and though their contents are apparently
cellular elements and their size approximately the same as that of the spermospores of

(ZOOL. CHALL. EXP. — PART XXXL — 1884.) Hh 7

50 THE VOYAGE OF H.M.S. CHALLENGEE.

Verongia hirsuta{V) or Verongia tenuissiina {Vj , I yet incline to the belief that they
stand in no connexion with the sponge organism.

Coheir. — Sandstone yellowish-grey.

Habitat. — Bahia, shallow water.

In addition to the well-marked Spongelidse just described, there are in the Challenger
collection three specimens probably belonging also to the family in question, but so very
badly preserved that there are no soft parts on their skeleton. I believe it advisable
to abstain entirely from their detailed description. I think that an undescribed form is
of greater profit to science than a form described insufiiciently for subsequent recognition.
I believe the sponges just mentioned belong to the Spougelidse, but the question
whether a sponge belongs to this or that family in the Keratosa cannot be decided from
the properties of the skeleton alone. Again I cannot describe them only as species ; the
external shape plays a great part in the specific description, luit the exterior of the
sponge is influenced by the properties of its soft parts, and these as mentioned before are
entirely absent. An entire abstinence from any description seems to me therefore to
be the best plan.

Family Spongid^e.

Spongiadm and Hirciniadce, Gray, 1867.

Bibulida and Hircinida (e.;;.), Carter, 1875.

Spongiadm (e.p.), Hirciniadce {e.p.), and Phyllospongiadm, Hyatt, 1877.

Spongidm and Hircinidce, F. E. Schuize, 1879.

Keratosa with small hemispherical flagellated chambers, communicating by means of
numerous pores with inhalent, by means of special canals with exhalent, cavities.
Axis of fibres thin ; ground-mass in the neighbourhood of the flagellated chambers
granulated.

Coscinoderma, Carter.

Spongidae with skeletal fibres admitting of no distinction into primar}^ and secondary
ones.

Coscinoderma confragosum, n. sp.

" Battledore-shaped, covered with a white continuous cribriform incrustation ; surface
even, with fibres almost uniformly alike in size and colour, viz., very small and fine, very
long, scantily branched, curled up together in little whorls, of a deep sponge colour."
In these words Mr. Carter ^ characterises his genus Coscinoderyna, and together with it
his species lanuginosum. The species I am about to describe agrees very well with that
description, but presents the following difi'erences. When bisected longitudinally the

1 Ann. and Mag. Nat. Hist., ser. 5., vol. xii. p. 309, 1883.

EEPOET ON THE KERATOSA.

51

form of the section may be compared with a battledore, though the stem must be
called very thick, and besides ends with a basal extension ; but only as regards the plane
of the section, the whole not being compressed, but of a massive form and of rather
roundish outline. The colour of the incrustation is not white but pale greyish ; and the
colour of the skeletal fibres is only straw-yellow. Again, I cannot call the outer surface
of my specimen unconditionally even ; at anyrate the surface of the skeleton devoid
of soft parts is denticulated; these denticulations are not, however, the terminal points of
single fibres, but whole networks in the form of sharp-pointed projections. All these
peculiarities together, and particularly the formation of sharp-pointed projections on

Fig. 1. — Skeletal fibres of Coscinoderma cmifra<josum.

the surface of the skeleton, quite absent in Coscinoderma lanuginosum, Carter, necessitate
the establishment of a new species. In the accompanying diagram a portion of the
skeleton is represented in order to show the manner in which the fibres interlace with
one another. The fibres themselves are entirely devoid of any foreign enclosures, and
their average thickness is 0 '0 1 5 mm. The species is represented in the collection by a
single specimen, dredged by H.M.S. " Porcupine."

Colour. — Outer surface greyish, parenchyma pale grey-yellowish, skeletal fibres
straw-yellow.

Habitat. — H.M.S. " Porcupine," Station 13, 1870, ofi" the coast of Portugal; depth 220
fathoms.

Coscinoderma denticidatimi, n. sp. (PI. VI. fig. 4).

This species, also represented by a single specimen, agrees with Coscinoderma con-'
fragosum as regards the formation of sharp-pointed projections of the skeleton, though

52 THE VOYAGE OF H.M.S. CHALLENGER.

here they are visible also when covered with soft parts, so that the outer surface of the
sponge is distinguished throughout by small conuli 0"2 to 0'33 mm. high, the average
distance between them being 1 mm.; but it differs both from Coscinoderma lanuginosum
and Coscinoderma confragosum with respect to the character of its surface, which has no
special superficial incrustation, the skin not differing from that of Euspongia or
Cacosjyongia , and the manner in which the fibres form meshes. As in most other
SpongidsB, the meshes are here polygonal, although still formed by fibres all of the same
size — on an average 0"042 mm., i.e.. almost three times as thick as those of Coscino-
derma confragosum — a new character distinguishing this species both from Coscinoderma
lanuginosum and Coscinoderma confragosum. The average size of the meshes is 0"25 mm.
In external shape the specimen representing this new species in the Challenger collection
is in the form of a crust, but with local thickenings 10 mm. in height. In contrast
to the Challenger specimen of Coscinoderma confragosum, which shows a very well
developed osculum, no oscula can be discerned on the surface of this specimen.

Colour. — Soft parts pale greyish, skeletal fibres brownish-yellow.

Habitat. — On the reefs Honoruru, Sandwich Islands ; depth 40 fathoms.

Coscinoderma ahum, n. sp. (PL V. figs. 4-6).

This species is a good species also, provided of course that the genus Coscinoderma
be good. The meshes of the skeleton, as in the species just described, are of polygonal
outline, but on an average twice as large, the fibres themselves being one and a half
times as thick. Unlike all other representatives of the genus hitherto known, the fibres
are abundantly charged with foreign enclosures, so that they appear of a rather dirty-
white colour. Again, the external appearance of the sponge itself is novel, and is perhaps
not devoid of systematic importance. While in the two preceding species we have had to
deal with forms either incrusting or massive, Coscinoderma altum is found in the form
of a long cylindrical stalk with rounded upper end 230 mm. high and about 18 mm.
broad, with denticulated- outer surface, the prominences 075 mm. high and 2 mm. apart
being however rather blunt, and the skin itself presents the same appearance as that of
Aplysina aerophoba as drawn by F. E. Schulze.^ The specimen contained numerous
]\torul8e, and I make use of this opportunity in order to turn attention to a peculiarit}^
which, in a stUl higher degree, I have very often observed in Cacospongia and JEuspongia,
as mentioned in my paper "Ueber die Spermatogenese bei Sycandra raphanus," but which,
as far as I am aware, is hitherto uudescribed. I mean the extraordinary growth of the
endothelial cells surrounding the growing embryos. They do not differ in size from
other parenchyma-cells when young, but as they grow they do not become like the pave-
ment-cells, but increase both in area and volume. Again, when surrounding a Morula

1 Zeitsclir. f. tnss. ZooL, Bd. xxx. pi. xxi. fig. 1, 1878.

REPORT ON THE KERATOSA. 53

these cells are readily distinguishable from all other cellular elements not only by their
extraordinary size but also by their containing numerous, and comparatively very large,
granules. That a generative product when it — contrary, e.g., to the spermospores of
Calcarea — increases during maturation, must receive nutritious material from the sur-
rounding parts of the parent body is indeed obvious, but in the Porifera this may be done
in two ways, either by the endothelial cells playing an intermediate part, or, by means of
a certain modification, becoming so to speak staples for the material to be consumed by the
growing product. The first is the case as regards the sperm-balls, their endothelial cells in
all stages of development representing typical pavement-cells, and with further growth
gradually becoming rather flatter. Since a sperm-ball, for instance, of A2:>ly.silla sul-
phurea when quite mature is many times larger than the cell from which it derived its
origin, it must be assumed that its increase is due not only to the nutritious material within
the primitive cell itself, but also to the material absorbed from the surrounding elements.
Now the surrounding elements, viz., the endothelial cells, are neither voluminous nor rich
in nutritious particles ; thus the only possible conclusion is that these endothelial cells
having received nutritious material from other cellular elements do not retain it but give
it up instantly to the generative product. On the contrary, the endothelial cells
surrounding a developing embryo of, e.g., a Cacospongia scalaris retain the nutritious
particles, and it is in this property that we have also a natural explanation of the
striking aggregation of mesodermic cells in the neighbourhood of a developing embryo,
as observed and described by F. E. Schulze.^

Colour. — Outer surface brownish, parenchyma and skeletal fibres pale greyish.

Habitat. — Station 135a, October 16, 1873, off Tristan da Cunha; dep»th, 7^ fathoms;
hard ground, shells, and gravel.

Euspongia, Bronn.

Spongidse with fine skeletal fibres forming a compact network, the meshes being very
small ; primary and secondary fibres readily distinguishable.

Euspongia officinalis (Linnd), var. lohosa, n. var. (PI. VI. fig. 1).

The single specimen on which this variety is founded recalls by its external shape
the drawing which F. E. Schulze ^ gives of Euspongia officinalis var. ttihulosa, with the
distinction that its basal part is not plate-like as in the latter, but massive. As in most
EuspongicB the outer surface is denticulated owing to the prominent primary fibres,
but here the conuli are very low, their height not exceeding 0"3 mm. A portion of the
skeleton is represented on PL VI. fig. 1 ; it is bush-like, and it must be noticed that while

^ Zeitschr. f. wiss. ZooL, Bd. xxxiii. pi. iii. fig. 1, 1880. 2 n^l^ bj_ s^xii. pi. xxxiv. fig. 8, 1879.

54 THE VOYAGE OF H.M.S. CHALLENGER.

in other EuspongicB the i^rimary fibres run usually more or less parallel to one another,
here they show a very pronounced tendency to ramify ; the spaces between them and
their secondary, but still vertically directed, branches lieiug filled with an irregular
network of fibres originating also from the primary ones, but in a more or less horizontal
direction. The average basal diameter of the primary filires is O'l mm., that of the
tertiary only 0"03 mm. The latter, in contrast to the primary fibres, are in most cases
quite free from foreign enclosures. This variety is a connecting link between Euspongia
officinalis and Cacospongia moUior , and maybe classed in the species last mentioned with
the same right as in Euspongia officinalis.

Colour. — Outer surface pale greyish, parenchyma colourless, fibres' straw-yellowish.

Habitat. — Bahia, shallow water.

Hippospyongia, F. E. Schulze.

Spongidse with fine skeletal fibres and small meshes, the fibres admitting of no
distinction into primary and secondary ones ; distinguished by a system of canals
permeating the body in all possible directions.

Hipposp)ongia anomala, n. sp. (PI. YII. ; PI. VI. fig. 2).

This species is represented in the Challenger Collection by a single but very large
specimen 350 mm. broad, 200 mm. high, and 40 mm. thick, drawn of the natural size, but
in a bent position on PL VII. For its characters, common to the whole genus, I refer
to the diagnosis in the foregoing paragraph. The character distinguishing it from other
species of the genus is a property which renders its grouping in the genus Hip)pospongia
rather disputable. I mean the presence of quite distinct primary fibres directed obviously
towards the outer surface, on an average four times as thick as the others, and absent
only in the skeleton supporting the thin membranes covering the lacuna immediately
under the outer surface. It must be said that the terminal skeletal processes of a
Hippospongia equina, when seen with the naked eye, show within them certain
streaks perpendicular to the outer surface ; but these streaks and the fibres in question of
my Hippospongia anomala are quite difi"erent things, the last-mentioned being really
nothing but single fibres thicker than the others, the first-mentioned being composed of
many fibres and difierentiated from the surrounding parts of the skeleton by the fact that
the network formed by the fibres is here more compact than elsewliere. The aj)2iearauce
is therefore only an optical illusion ; at any rate the property in question distinguishes
my species readily from all others of the genus, but does not, however, decide the
question whether it really belongs to the genus Hijjpiospongia. Still I am of this opinion.
As regards the rigidity of the skeleton this specimen is, indeed, allied to Cacospongia
mollior ; but apart from distinctions concerning the presence of the system of canals

EEPORT ON THE KERATOSA. 55

permeating its body, the external shape of the animal and particularly its shagreen-
like surface, its skeleton (more compact than, but not nearly so hard as, that of Caco-
spongia mollior) recalls that of a Euspongia rather than that of a Cacospongia.

If now this species, as regards the property of its skeleton, were a typical Euspongia,
the question as to its systematic place would be still more difficult ; one classifier might
place it in the Euspongice, another, and on grounds of equal validity, among the Hippo-
spongicB. Now the species in question is neither a Euspongia nor a Cacospongia, nor a
typical Hippospongia ; but since one of the main characters of Hlppospongia is very
characteristic of this specimen, the most logical proceeding will be, I think, to group it in
this latter genus. For my own part I do not doubt that the species in question, be it referred
to Hippospongia or Euspongia, will still remain a good species, and I shall be quite content
if my description prove sufficient for future recognition. In order to render it more com-
plete I add the following details. The average diameter of the primary fibres reaches 0"12
mm., that of the secondary fibres being on an average only 0'028 mm.; the latter are
nearly all quite free from any foreign enclosures ; the first-mentioned, however, are so full
of them that their surface is of a rather angular character, owing to the fact that these
foreign bodies lie not only in the central part of the fibres but also in their j^eripheral
portions, projecting outwards and thus rendering the surface of the fibres uneven. It is
in this form that I have found, hand in hand with normal flagellated chambers, chambers
devoid of any special camera! canaliculis, although in other details its internal organisa-
tion presents no deviations from that, for instance, of Euspongia officinalis. The
specimen proved to be sterile.

Colour. — Soft parts and outer surface pale yellowish-white, skeleton straw-yellow.

ifa&itoi.— Station 186, September 8, 1874, lat. 10° 30' S., long. 142° 18' E., depth
8 fathoms, coral mud.

Hippospongia mauritiana (PI. VI. fig. 3).

Spongia lapidescens, Duchassaing de Fonbressin et iliclielotti ; subspecies mauritiana, Hyatt,
Revision, &c., vol. ii. p. 527.

There can be scarcely any doubt that the three specimens of the collection now under
consideration are to be referred to Hyatt's Spongia lapidescens, subspecies mauritiana.
His description is very good and detailed, and the form itself, owing to the density of
its skeleton, admits of a very clear definition. But what appears to me very strange is
that Hyatt still finds it possible to distinguish in its skeleton primary and secondary
fibres. I find them all approximately of the same size ; of course the bundles of parallel
fibres projecting from the outer surface are directed vertically, but seeing that they are
not united one with another by smaller horizontal fibres, and that they do not difl'er in
thickness from other fibres of the skeleton, I see absolutely no grounds for regarding

56 THE VOYAGE OF H.M.S. CHALLENGER.

them as primary ; and it is owing to this character, as well as to the body being per-
meated by a regular system of internal canals, that I must ^ place the form in question
as a new species in the genus Hvppospongia, F. E. Schulze.

Colour. — Soft parts yellowish-pale, skeletal fibres dark yellow.

Habitat— OS A^i, New Hebrides, 18th August, 1874 ; depth 60 to 70 fathoms.

Cacospongia, Oscar Schmidt.

Spongicte with readily distinguishable primary and secondary skeletal fibres of
comparatively thick diameter and forming comparatively large meshes.

Cacospongia levis, n. sp. (PL V. figs. 1-3).

The external surface of the single Challenger specimen is quite smooth except in
some spots where is is rather roughened by the ascending primary fibres. It is in this
property that, bearing in mind the conjectural afiinities of the species, I am inclined to
see its most important systematic character. The species seems to be closely allied to
Cacospongia mollior, but as we learn from F. E. Schidze (compare his ch'a^\ang in the
memoir on the Spongidte), the outer surface of the latter is denticulated throughout.
Again, the meshes of Cacospongia levis are rather smaller than those of Cacospongia
mollior, and its primary fibres are O'OB mm., the secondary ones only 0"04 mm. thick,
the corresponding fibres in Cacospongia mollior being on an average 0"15 mm. and 0"05
mm. in diameter. Approaching as regards these latter characters to my Hippospongia
anomala, the species in question difi"ers both from it and from JEuspongia officinalis var.
lobosa in a sharply pronounced rigidity of its skeleton. Just as in Hippospongia anomala,
the body of my Cacospongia levis proved also to be perforated by numerous canals, but
neither do these canals show, on the whole, any regular arrangement, being C[uite analogous
to those of Cacospongia cavernosa, nor is any membrane to be found above their terminal
points, so that there can be no doubt as to the systematic position of this sponge. The
primary fibres though not prominent may still be readily distinguished (PI. V. fig. 2).
They proved to be cored with foreign bodies, chiefly fragments of spicules lying, however,
only in the central part of the fibre ; in contrast to this, the secondary fibres are quite
devoid of any enclosures. In the soft parts of the specimen in question I found some forma-
tions which though lying free in the parenchyma, i.e., not surrounded by a special endothe-
lial layer, nevertheless presented a great resemblance to sperm-balls. A more careful
examination and the application of high magnifying powers showed, however, that these
formations are scarcely identical with sperm-baUs. Their contents consist of numerous
oval bodies in which no nucleus could be distinguished, but quite homogeneous throughout

' Comp. p. 84.

REPORT ON THE KERATOSA. 57

their whole extent, and, in spite of my endeavours to discover any initial or further
stages in their development, I find these oval bodies enclosed, though differing in size but
always with precisely the same external appearance. I believe they belong to some low
plants.

Colour. — Outer surface black, parenchyma dirty yellowish-grey, skeletal fibres
bro\\Tiish-yellow.

Habitat. — Off Barra Grande, September 10, 1873; depth 400 fathoms, red mud.

Cacospongia amorpha, n. sp. (PI. VI. fig. 5).

The properties of the skeleton of this species recall those of Eusp>ongia officinalis, var-
lobosa, as regards the tendency of the primary fibres to ramify, but the fibres, both primary
and secondary, are comparatively far larger, as are also the meshes formed by them ; one
might illustrate the difi'erence between these two forms by comparing them to a bush and
a tree.

The species is represented by two specimens, one of a massive stout appearance, the
other 100 mm. high by about 20 mm. thick, of roundish outline in the upper part,
and flatly compressed near the rather extended bases. The outer surface is denticulated
by prominent primary fibres forming conuli of 1 mm. on an average, and 3 to 5 mm.
distant from one another. In many instances the membrane between the conuli is not
homogeneous but sieve-like, the pores being very conspicuous. All the primary fibres are
overcharged with foreign bodies, this property rendering them of irregularly angular
outline ; some of the secondary fibres are devoid of any enclosures, others on the contrary
being more or less covered with saud-grains, fragments of spicules, &c., and often a fine
secondary fibre shows an extension in its middle, owing to the presence of a sand-grain
many times larger than the diameter of the fibre itself. It may also be added that, as
in Hircinia variabilis according to F. E. Schulze,^ the primary fibres in my Cacospongia
amorpha show a tendency to form meshes.

Colour. — Outer surface grey, parenchyma butter-like, skeletal fibres pale brownish-
yellow.

Habitat. — Bahia, shallow water.

Cacospongia murrayi (PL IV. fig. 3 ; PI. VI. fig. 8).

It is really very agreeable, after such a bad species as that just described, to pass on to
the description of such a bona species as Cacospongia murrayi. The single speci-
men on which this species is founded is represented on PL IV. fig. 3, the skin of one
part being stripped off. The external surface is smooth throughout, and is represented
by a rather thick dermal membrane, which, owing to the great number of foreign

1 Zeitschr.f. wiss. Zool., Bd. xxxiii. pi. iii. fig. 1, 1880.
(ZOOL. CHALL. EXP. — PART XXXI. — 1884.) Hh 8

58 THE VOrAGE OF H.M.S. CHALLENGER.

bodies enclosed, differs obviously both in colour and consistence from the proper
parenchyma, and can be easily drawn off. This peculiarity is of course of very doubtful
importance. I have very often had occasion to speak of the low systematic significance
of similar properties of the skin, and it is sufficient to peruse the lines which F. E.
Schulze devotes to the matter in his paper on Hircinia ^ in order to become convinced
that species based on such characters are very doubtful. The more the dermal membrane
is charged with foreign bodies the more easily can it be drawn off, and on the other hand
it is obvious that the number of foreign bodies enclosed in the membrane depends on the
nature of the surrounding ground.

Of course it cannot be denied that the tendency to form such a cortex of foreign
bodies may become hereditary, but stUl its systematic application appears to me to be
rather dangerous ; and with respect to my Cacospongia 'riiurrayi, it is the less necessary
as the sponge shows another peculiarity of an undoubtedly more constant character,
concerning the structure of the skeleton. In Cacospongia murrayi the skeleton is,
so to speak, composed of two skeletons. There are vertical primary fibres 0'15 mm.
in diameter, and rather finer secondary fibres united to the first-mentioned in a more
or less regular horizontal direction, and this system of primary and secondary fibres
is enveloped in an irregular network of still finer fibres (0'015 mm. in diameter on an
average), originating both from primary and secondary ones, but forming far smaller
meshes. PL VI. fig. 8, representing the outer surface of a portion of skeleton, when
seen against the light, illustrates this peculiarity sufficiently well. The outermost points
of the primary fibres are either pretty well developed, and accordingly project from the
outer surface into the substance of the cortex, without giving rise, however, to any
unevenness on its outer surface, or are very short, and represented only by low thickenings
over the intersecting poiat of the primary and the uppermost secondary fibres. The
histological character of the cortex presents no deviations from the usual connective
tissue of Keratosa, except that the cellular elements, stellate and fusiform cells, ai-e very
scanty ; the histological and anatomical structure of the proper parenchyma is that of
typical Spongidse.

The species is named in honour of Mr. John Murray of the Challenger Expedition.

Colour. — Outer surface grey, parenchyma pale greyish, skeletal fibres brownish.

Habitat. — Station 163b, June^S, 1874, off Port Jackson; depth 35 fathoms, hard
ground.

Cacospongia vesiculifera, n. sp. (PI. IV. fig. 2 ; PL VI. fig. 9).

This form, like the one just described, isr epresented in the Challenger Collection
by a single specimen, and is in its turn provided with a sj^ecial cortex, but here its

1 Zeitschr. f. toiss. Zool., Bd. xxxiii. p. 14, 1880.

EEPORT ON THE KERATOSA.

59

presence alone justifies the establisliment of a new species ; this mainly owing to its
histological peculiarities. Apart from the foreign enclosures, its constituent parts are
scantily developed transparent ground-mass, and in this latter large vesicular cells
of round or more oval form, 0"02 mm. in diameter, not dissimilar to the renowned
and still debatable " Schleimzellen " of MoUusca, as Dr. Flemming ^ has drawn them,
and thoroughly identical with the vesicular cells of many Desmacidonidse — undescribed
indeed hitherto, but undoubtedly very well known to every spongiologist who has
had to deal with the representatives of the family just mentioned. A portion of the cortex
with such cells is shown in the annexed woodcut.

But the sponge also possesses some other characters which render it readily dis-
tinguishable from other Keratosa, its skeleton recallinar that of
a Spongelia rather than that of Cacospongia, and its external shape
being perhaps also of some systematic value. The sponge has
been found in the form of a thick-walled tube, with very
narrow central cavity, ending in a small osculum. The outer
surface is rough and provided with rounded tubercles corre-
sponding to the prominent secondary fibres. There are in this
sponge three kinds of skeletal fibres — (1) gastric vertically directed
primary fibres, in most cases more or less loaded with foreign
bodies, and on an average 0"2 mm. thick; (2) centrifugal
secondary fibres, originating from the primary ones, and in their direction towards
the outer surface forming with the last mentioned, if from above, a more or less
acute angle, and with an average diameter of 0"1 mm.; and (3) still finer (0"06 mm.)
tertiary fibres uniting the secondary and primary ones ; to sum up, this is a kind of
skeleton very common in Spongelidae, but exceptionally rare in Spongidge. The secondary
and tertiary skeletal fibres proved to be free from any enclosures. Apart from the
vesicular cells there are no other histologically or anatomically deviating peculiarities
to be stated.

Colour. — Outer surface dii'ty greyish, parenchyma pale yellowish- white, skeletal
fibres straw-yellow.

Habitat. — Ofi" Port Jackson, April 1874 ; depth 7 fathoms.

Fio. 2. — Cortex of Cacospongia
vesicnlifera. a, pavement
epithelium of a subdermal
cavity ; b, vesicular cells.

Cacospongia procumbens, n. sp. (PI. VI. fig. 6).

This species — a Sarcotragus in the sense of Oscar Schmidt — possesses a very curious
skeleton. In Cacospongia vesiculifera we have had to deal not only with primary and
secondary, but also with tertiary skeletal fibres ; here there are not even primary and
secondary ones. The sponge has been found in the form of a crust, and its supporting

' Die Bindesubstanz der Mollusken, pi. i.

60 THE VOYAGE OF H.M.S. CHALLENGEE.

apparatus is represented by a large meshed thoroughly irregular basal network of the
skeletal fibres, with numerous ascending branches directed towards the outer surface,
with a denticulated appearance due to conuli, about 075 mm. high and 3 to 4 mm.
distant from one another. All the fibres are approximately of the same thickness
(0'18 mm. on an average), and all cored in their central part, very scantily, however,
with foreign bodies. As before remarked, the species would be referred by 0. Schmidt
to his subgenus Sarcotragus, the filaments filling its parenchyma being very thin
and fine.

In the skin are numerous fragments of siliceous spicules.

Colour. — Parenchyma pale brownish, skeletal fibres deep brown.

Habitat.— UM-^. "Porcupine," Station 13, 1870, ofi" the coast of Portugal; depth
220 fathoms.

Cacospongia dendroides, n. sp. (PI. VIII. figs. 1-3 ; PI. VI. fig. 14).

There are in the Challenger collection three specimens, aU of a rather Spongeloid
shape, and from the same locality, but presenting some distinctions in the structure of
their skeleton, so that a subdivision at least into varieties is necessary. One of the
specimens is represented on PI. VIII. fig. 1, and therefore a further description of
its general form is superfluous. The other specimens all agree as to their external shape
with the one just mentioned, but while the skeleton of this latter (PI. VI. fig. 14) is
composed of an irregular network of comparatively thick and yeUow coloured fibres, the
skeletons of the two other specimens, which do not differ from the former as to the
general character of the network, are composed of fibres twice as thin, forming meshes
approximately twice as large as in the former case, and not of a yellow colour but white.
Is this a character of individual or of varietal, or of specific significance 1 I must confess
I am not prepared to decide the question, and think it advisable to choose the middle course
by establishing two provisional varieties — dura for the specimen with more solid fibres, and
friahilis for the two remaining. In both cases all the skeletal fibres proved to be more
or less cored with foreign enclosures. AU the three specimens have proved to be full
of filaments, and it was in two of them that I found out those dumb-bell -shaped corpuscles
to which I have referred on page 14.

If amongst the skeletal fibres of this form the primary — vertically directed — can be
still distinguished owing to their comparative thickness, it must yet be added that such a
distinction is of a rather conditional character, the skeleton presenting the aspect of a
pretty irregular network.

Colour. — Pale yellowish-white.

Habitat.— Station 208, January 17, 1875, lat. 11° 37' N., long. 123° 31' E. ; depth
18 fathoms, blue mud.

REPORT ON THE KERATOSA. 61

Cacospongia spinifera, n. sp. (PL VI. fig. 12).

Had the genus OUgoceras been retained it would have been necessary to refer the
species I am going to describe to that genus. The outer surface of the single specimen
representing this new species in the collection is indeed a contrast to OUgoceras collectrix,
being quite devoid of any foreign bodies, but they are abundant in the parenchyma,
and on the other hand are so richly represented in the skeletal fibres that these latter
are extremely poor in true horny substance. Again as to the external aspect of its
skeleton, — apart from the point that it is not represented by isolated standing trees as
in OUgoceras collectrix, — it recalls vividly that of the form just mentioned. The large
size of the meshes, the tendency of the fibres to ramify — all this approximates the form
I am now describing to Cacospongia {OUgoceras) collectrix. There is, however, a
character which permits a pretty sharp diagnosis of the species ; I mean the nature of
its outer surface. The latter, in spite of the great friability of the skeleton and its general
tendency to disappear entirely, is still provided with very high conuli (4 to 7 mm.), so that,
if drawn, this specimen would give precisely the same impression as the drawing given by
Hyatt for his Hircinia acuta,^ and this not only with regard to the character of the outer
surface, but also that of the external shape of the sponge in general. The specimen
must have died at the moment when its dermal pores were open, and thus the whole of
its outer surface is of a sieve-like appearance in the proper sense of the word. Apart
from the point that, like the preceding form, this specimen proved to be overloaded with
filaments, their heads being of elongated form and reaching 0"06 mm. in length and
0'04 mm. in breadth on an average, there are no histological nor anatomical peculiarities
of interest to be stated.

Colour. — Outer surface brownish-grey, parenchyma pale greyish, skeletal fibres
black.

Habitat.— Station 177, August 18, 1874, lat. 16° 45' S., long. 168° 7' E. ; depth
130 fathoms, volcanic sand.

Cacospongia tuherculata, n. sp.

So far as the properties of the skeleton are concerned (apart from the fact that the
fibres are not black but pale yellow), this species seems to be closely allied to Cacospongia
spinifera. Indeed, in most cases the primary fibres do not ramify at aU, but ascend
vertically, and are united with one another by horizontal secondary fibres, these latter, as
in Cacospongia spinifera, beginning very often with flat extensions ; yet, here and there
ramifying fibres have been found, their secondary obliquely directed branches instead of
reaching the neighbouring primary fibres ending freely. There is, however, one character

' Revision, &c., vol. ii. pi. xv. fig. 20.

62 THE VOYAGE OF H.M.S. CHALLENGER.

in the two specimens by which the species is represented, admitting probably of a
tolerably good distinction of the form in question from Cacospongia sjnnifcra. It is
the appearance of the outer surface, which is here not spinous but provided with rounded
tubercles. How far this peculiarity permits the estabhshment of a new species is difficult
to say ; at any rate I see at the present time no other course open but to separate the
form in question by the establishment of a new species.

Both the specimens proved to be full of filaments, here, however, with heads of a
rather different shape from that of the filament-heads in my Cacospongia spinifera,
being of a more roundish outline, and with an average diameter of 0'055 mm. One of the
specimens proved to be quite compact, the other, as in Cacospongia cavernosa, was pierced
by numerous large internal channels inhabited by Chsetopoda.

Colour. — Outer surface greyish, parenchyma white, skeletal fibres pale yellow.

Habitat. — Station 162, April 2, 1874, ofi"East MonccBur Island, Bass Strait; depth
38 fathoms, sand and shells.

Cacospongia intermedia, n. sp. (PL VI. fig. 7).

I have already taken occasion to mention (p. 27) this species as presenting to the classifier
many difficulties. The meshes formed by its skeletal fibres being rather large, and the fibres
themselves thick, the form must be referred to Cacospongia ; but the fibres are almost all
of the same diameter (0'35 mm.), and it is only in the prominences of the outer part of
the skeleton that an approximate distinction between primary and secondary fibres is possible.
This character recalls the Coscinoderma of Carter; on the other hand, the body of the sponge
is broken through by numerous canals, the character of the outer surface of the skeleton
is that of Hiiypospongifi, the distinction between primary and secondary fibres, as already
stated, is pronounced only in the tufts ; all this would justify the placing of this sponge
in the genus Hippospongia, provided that one could prove the importance of the char-
acters just mentioned to be greater than that of the skeletal fibres being thick and the
meshes formed by them large. So far as to the general position of the form in question.
Now it must be mentioned that apart from the thickness of the skeletal fibres and the
largeness of their meshes, the sponge recalls vividly Euspongia vermicidata. What
systematic place is to be assigned to it ? Of course merely a provisional one, and accord-
ingly the task of the classifier is reduced to giving a detailed description of it.

The species is represented in the collection by a single specimen of irregularl)-
massive form ; the outer surface is smooth and even, the dermal membrane enveloping
the skeleton with its external outgrowths in the same manner as in Hippospongia anomala.
In many spots the dermal membrane is pierced by larger or smaller (2 to 4 mm.) openings,
sometimes disposed by threes or fours together, sometimes lying isolated ; whether some
of these openings are really oscula is difficult to say without a complete destruction of

REPORT ON THE KERATOSA. 63

the specimen ; but this is certain, that most of them are " pseudo-stomata," leading into
the canals separating the body of the sponge, and the envelope of whose walls is merely
the continued dermal membrane. The properties of the skeleton have been already
discussed, and I hope that with the help of the figure given on PL VI. its characters will
be rendered obvious ; its outer surface recalls that of Hiiypospongia mauritiana (PL VI.
fig. 3), apart from an entire absence of bundles of parallel fibres.

Colour. — Outer surface black, parenchyma greyish, skeletal fibres brown.

Habitat. — Station 177, ofi" Api, New Hebrides, August 18, 1874; depth 130
fathoms, volcanic mud.

Cacospotigia irregularis, n. sp. (PL VI. fig. 10 ; PL VIII. fig. 5).

The chief peculiarity of this curious species consists in this property of its skeleton,
that, in contrast to all other Spongidaj, the bundles of its fibres are so closely and
irregularly intertwined that the distinction of primary and secondary ones is entirely
impossible ; they form a kind of irregular network, its meshes being not represented by
internal channels as in HipposjJOf^gia, but filled by parenchyma. The species is repre-
sented in the collection by a single specimen, apparently of irregularly massive form, in
reality provided with large internal cavities, so that the sponge can be compared to a
casket. The surface is smooth, but not devoid here and there of outgrowths correspond-
ing to the prominent tufts of skeletal fibres, presenting a compact network, and in most
cases ending more or less shar23ly. As to the secondary meshes of the skeleton, they are
quite irregular, and my drawing (PL VI. fig. 10) will give a better explanation of them
than any words. The primary meshes, i.e., meshes formed by the skeletal fibres them-
selves, are in most cases small, but in others very large, and occasionally a mesh is repre-
sented by two bundles forming an acute or obtuse angle and a single usually thick fibre.
The thickness and length of the fibres are very variable, but all agree with one another
as to their tendency to take in foreign bodies. The specimen proved to be overcharged
with filaments, one of which is represented on PL VIII. fig. 5.

Colour. — Outer surface and parenchyma pale dirty yellowish, skeletal fibres pale
grey-yellowish.

Habitat.— Citation 188, September 10, 1874, lat. 9° 59' S., long. 139° 42' E.;
depth 28 fathoms, green mud.

Cacospongia oligoceras, n. sp. (PL VI. fig. 13).

This species is represented only by a small fragment of massive shape, with the outer
surface, where not covered with mussel-shells, stones, &c., provided with low conuli. Its
exterior agrees closely with that of Oligoceras collectrix, Schulze, but there is a great

64 THE VOYAGE OF H.M.S. CHALLENGER.

difference with regard to the structure of the skeleton ; in OUgoceras coUectrix the
skeleton is represented by small isolated horny trees, and there are parts in the body
where no skeleton is to be found at all ; in the species I am describing it is present
everywhere, but though a continuous network, is very often interrupted by large, enclosed,
foreio-n bodies. The skeletal fibres are full of fragments of spicules, sand-grains, &c. ;
their form, whether cylindrical or flat, as well as the size of the meshes formed by
them, is very variable. A portion of the skeleton ( x 2) is represented on PI. VI. fig.
13. The thickness of its fibres, vertical and horizontal indiscriminately, varies from
0-06 to 1 mm.

Colour. — Outer surface greyish, parenchyma butter-yeUowish, skeletal fibres white.

Habitat.— Station 208, January 17, 1875, lat. 11° 37' N., long. 123° 31' E. ;
depth 18 fathoms, blue mud.

Cacospongia compacta, n. sp. (PL VI. fig. 11).

The external shape of the single specimen representing this species in the Challenger
collection recalls that of Euspongia officinalis, var. lobosa, but of course this coincidence
is of no further consequence, the form in question showing the closest affinities to Caco-
spongia oligoceras just described. There are, however, the following distinctions : the
meshes in the former are comparatively large ; in the latter they are so very narrow that
the whole skeleton at first sight appears to be a compact mass (comp. PL VI. fig. 11);
in both cases the horny substance is scautUy developed, the chief foundation of the
skeletal fibres being represented by foreign enclosures, but whde the fibres of Cacospongia
oligoceras and its whole skeleton are friable, those of Cacospongia compacta are rigid and
hard. Again, when a fibre of the former is seen under the microscope, it gives the
impression that the taking in of foreign bodies has been influenced by the kind of
growth of the fibre ; they are disposed in it as if following certain laws of distribution,
and although overloaded with them, the fibres still possess a rather smooth surface. It
is a strained expression to speak of differentiated skeletal fibres with regard to Caco-
spongia compacta; a portion of its skeleton when magnified conveys the' impression
as if there were previously some sand-grains disposed in the parenchjTiia without any
order, and these sand-grains were latterly enveloped in a continuous mass of horny
substance. Here a large sand-grain, and immediately as a continuation of this thickening
a short piece of a typical horny fibre without any enclosures, finishing again with a new
similar thickeninsj.

We are hitherto in complete uncertainty as to the systematic value of the colour" of the
horny substance in these and other cases. But if it be of consequence, this character alone
would justify the separation of both the forms compared ; in Cacospongia oligoceras it is
colourless, w^hile in the species I am characterising it is of an intense yellow colour.

REPORT ON THE KERATOSA. 65

The specimen proved to contain filaments, not being, however, particularly rich in
them.

Colour. — Soft parts dirty white, skeletal-fibres brownish-yellow.
Habitat. — Baliia ; shallow water.

Cacospongia collectrix, F. E. Schulze, sp.

Oligoceras collectrix, F. E. Schulze, Zeitschr. f. wiss. Zool., vol. xxxiii. p. 34.

The two Challenger specimens which I group in this species are so diff"erent in
external shape that, should we classify the Keratosa according to their form, we
should be obliged to refer them, at least, to difierent genera. Yet it is well known that
in most cases the external shape as well as the colour in the Porifera are of extremely
subordinate consequence, and I am the more inclined to regard both the specimens
in question only as individuals of the same species, as in the Dalmatian specimen of
Cacospongia collectrix described by F. E. Schulze we find a form of, so to speak, neutral
characters with respect to the differences above mentioned. These differences concern
the mode of growth and the quantity of foreign bodies enclosed in the parenchyma.
The Dalmatian specimen just alluded to is represented in Prof. Schulze's paper on
pi. ii. fig. 6, and one can see that is not crust-like but of massive shape. On
pi. iii. fig. 7, Prof. Schulze gives also a drawing of the anatomical organisation
of the specimen in question, and this drawing renders obvious that, contrary to the
specific designation of collectrix, its soft parts are comparatively free from foreign
enclosures. This latter character is common to the Challenger specimen of Cacospongia
collectrix from the Philippine Islands, while as to its external shape this specimen has
been found in the form of a crust. On the contrary, the second Challenger specimen (from
Japan) recalls vividly, so far as its exterior is concerned, F. E. Schulze's specimen from
Lesina, but its soft parts proved to be overloaded wdth foreign enclosures, almost
exclusively sand grains often 2 to 3 mm. in diameter. It is therefore plain that the
above differences may be important enough to justify the subdivision of the species into
independent varieties, but in no case of a greater consequence. All this, however, on
the supposition that hotli our specimens are really closely allied to Cacospongia collectrix,
F. E. Schulze. So far as the specimen from the Philippine Islands is concerned this is
beyond doubt. This specimen, together with the mussel-shells which it coats, is also
of a rather lunq>y, massive form ; a more attentive examination shows, however, that
this is due simply to the aggregation of the mussel -shells, and that the sponge itself is of
a well-marked crust-like appearance, sometimes as thin as a leaf of paper. Now, while in
some parts of it no skeleton is to be discovered at all, the others are propped by small
(2 to 7 mm. high and about O'o mm. thick), isolated horny trees very rich in enclosed

(zool, CHiLL. EXP, — PAUT XXXI. — 1884.) Hh 9

66 THE VOYAGE OF H.M.S. CHALLENGER.

forei"ii bodies — all precisely as in the original specimen of the form from Lesina ; and
as its internal oro-anisation also does not deviate from that of Cacospongia collectrix as
described by F. E. Schulze, there is no room to doubt the specific identity of both these
specimens. But the matter is quite different with regard to the Challenger specimen
from Japan. Its internal structure also does not differ from that of a typical Eusjwngia
or Cacospongia, but — and this is the salient point — 1 was unable to make out whether
this specimen really does possess any differentiated horny skeleton. This was owing
to its minuteness, its dimensions being on an average as follows : — length 22 mm.,
height 7 mm., breadth 10 mm. The half of the specimen has been sacrificed to the
decision of the above question ; the result of a lasting treatment with hot water
and ammonia was a heap of sand-grains and a single 2 mm. high horny tree also very
rich in foreign enclosures. Is it to be regarded as produced by the sponge itself 1 Is it
also nothing but a foreign enclosure, like what Marshall states ^ of the specimens of
Psammopemma densum he had for examination ? All this remains an open question.
That our specimen is a horny sponge is quite plain, the foreign bodies enclosed in
its parenchyma being undoubtedly surrounded with obvious horny substance ; that,
being a Keratose sponge, it is also a Spongid is demonstrated by its internal struc-
ture, but whether it is to be classed in the species Cacospongia collectrix, and not
to be regarded as an analogue of Psammopemma in the family of Spongidte, will be
decided but by later investigations. Of course under such circumstances I prefer to
abstain from the creation of a new name.

Both the specimens proved to contain filaments, but while the specimen from Japan
is strikingly poor in them, the specimen from the Philippine Islands on the contrary
is overloaded with them in not a less degree than the Challenger sj)ecimens of Stelospongos
longispinus or Cacospongia irregularis.

Colour. — Specimen from PhUippiue Islands, pale greyish-yellow. Specimen from
Japan, outer surface grey, parenchyma dirty yellowish.

Habitat.— ^ta-tion 203, October 31, 1874, lat. iT 6' N., long. 123° 9' E.; depth,
12 to 20 fathoms; mud. Station 233a, May 19, 1875, lat. 34° 38' N., long. 135° I'E.;
depth, 8 to 50 fathoms ; sand.

Stelospongos, 0. Schmidt.

Spongida3 with comparatively thick skeletal fibres united in separated columns
directed more or less regularly radially from the basis of the sponge towards the outer
surface, and consisting each of a compact network of vertical, primary, and horizontal
secondary fibres.

1 Zeitschr.f. wiss. Zool., Bd. x.\xv. p. 114.

REPOET ON THE KERATOSA. 67

Stelospongos longispinus, Fonbressin et Miclielotti, sp. (PL VI. fig. 15; PL VIIL fig. 4).

Polytherses longispina, Fonbressin et Miclielotti, Spongiaires de la mer Caraibe, p. 71.
Hircinia acuta, var. longupina, Hyatt, Revision, &c., vol. ii. p. 549.

The descriptions of Fonbressin and Michelotti and Hyatt agree closely with
the properties of one specimen of the Challenger collection, and I have no
doubt that I am right in identifying it with Hircinia acuta, var. longispina of Hyatt.
How far the other varieties of this species distinguished hy Hyatt are really to l)e
referred to the genus Stelospongos, I am not prepared to say, but it is obvious that
the conjectural variety we are speaking of is a true Stelospongos. Hyatt himself says,
" In fact, so strong is the resemblance (lietween a Stelospongos and the form in question)
that it was at first classified with that genus," but constant to his dermal-membrane
theory of the formation of the skeleton, he assigned to the form another systematic
position. Now we know that the theory just cited is false, and a different procedure
is necessary. Indeed, when compared with Stelospongos maynardii, or any other true
Stelospongos, our specimen is not a typical representative of the genus. It is but seldom
that one can distinguish in the columns constituting the skeleton — some of which are
represented on PI. VI. fig. 15 — the vertical primary, and more or less horizontal secondary,
fibres, but in some instances this distinction is still evident, and accordingly there can
be scarcely any objections to the proceeding adopted here.

The specimen proved to be full of sperm-balls, and in a far greater degree of
filaments, one of which is represented on PL VIIL fig. 4.

Colour. — Outer surface dark grey, parenchyma dirty white, skeletal-fibres pale
yellow.

Habitat. — Off Barra Grande, September 10, 1873 ; depth, 400 fathoms; red mud.

Carteriospongia, Hyatt.

Spongidae of flabellate, leaf-like, or funnel-shaped form, with skeletal-fibres admitting
of a distinction into larger primary and smaller secondary ones, with flagellated chambers
usually devoid of any inhalent and exhalent canaliculi, with ribbed outer surface.

Carteriospongia radiata, Hyatt (PL IV. fig. 5 ; PL V. figs. 7, 8, 9).
Carteriospongia radiata, Hyatt, Revision, i&c., vol. ii. p. 541.

There are in the Challenger collection four specimens of the genus Carteriosjjongia,
three funnel-like, the fourth presenting a colony of leaf-like individuals, which with regard
to the properties of their skeleton — except that the " veil " of Plyatt proved to be propped
up not by a network of skeletal fibres but by foreign enclosures — and to those of the outer

68 THE VOYAGE OF H.M.S. CHALLENGER.

surface agree so closely with Hyatt's Carteriospongia radiata, var. dulsiana, that I have
but to refer the reader to the description above mentioned, and wish only to elucidate the
anatomical and histological characters of the internal organisation of this interesting
species. So far as these latter are concerned, some details have been already given
(pp. 17, 18), and it now remains to expose them coherently. Through the pores, incon-
spicuous to the naked eye and scattered on both surfaces of the animal, the water reaches,
as usual, more or less developed subdermal cavities ; but these latter, instead of ramifying
liy forming smaller and more numerous canals (as is the case in Aplysina and to a certain
degree also in Euspongia and other allied genera), in most cases communicate immediately
with the flagellated chambers ; the dendroid cliaracter of these ramifications is here cj^uite lost.
The flagellated chambers (all larger than those of true Spongidae, but still of hemisjaherical
shape) having received the water by means of numerous pores in their walls, expel it, not
by means of special canaliculi as is the case in the true Spongidse, but through a large
opening like that in the flagellated chambers of Spongelidse. The water having left the
flagellated chambers enters large exhalent lacunse of variable outline, and many of
these latter uniting together throw it out through the oscula — in the Challenger
specimens all on the internal surface ; in some funnel-shaped specimens in the British
Museum, on the contrary, all on the external surface. It is an interesting fact that when
the sponges are of a leaf-like form, the exhalent orifices are always more or less con-
centrated on one surface only. In lanthella they are indeed to be found on both
surfaces, but still their distribution is not equal — they are numerous on one surface, but
very scanty on the other. The properties of the internal organisation are thus rather
deviating from those of typical Spongidse ; again, as to the histological structure of
the form in question, it difi"ers also from that of true Spongida3, its ground-mass being
almost entirely devoid of granules in the neighbourhood of the flagellated chambers. To
the conjectural systematic significance of these differences many pages have already been
devoted ; we have nevertheless come to the conclusion that at last, provisionally, the
genus must be still grouped in the family Spongidae, owing to the presence of those
enigmatic " Strange " of cells which have been recently described by F. E. Schulze, and
which, in spite of a remark of this naturalist as to their entire absence in some individuals
of Euspongia or Cacospongia, I am yet inclined to regard as very characteristic of the
whole family. In all the Spongidse I have had the opportunity of examining I found
them, and if absent in one region of the body they are still to be found in other 2:)arts of it.

One of the specimens examined proved to be full of sperm-balls, one of which in a
ripe state is represented on PI. Y. fig. 9. I was able to discern also the preceding stages
of their development, but I abstain from their description here, since I shall return to
the matter when describing the spermospores of Verongia.

Colour. — Pale dirty yellowish.

Habitat. — Off Wednesday Island, Cape Yox-k, September 8, 1874 ; shallow water.

REPORT ON THE KERATOSA. 69

Carteriospongia otahitica, Esper, sp. (PI. IV. fig. 4).

Spongia otahitica, Esper, Pflanzenthiere, Bd. ii. p. 270.
Carteriosjjonffia otahitica, Hyatt, Revision, &c., voL ii. p. 541.

This species, represented in tlie collection by numerous specimens from Admiralty
Islands, was established in the year 1794, and there are in spongiological literature
many descriptions of it. Accordingly, I consider it unnecessary to enter here into parti-
culars, and refer the reader to the best of these descriptions, that of Hyatt, in the hope
that with the help of my drawing on PI. IV. the characters of the species may be easily
comprehended.

Colour. — Dirty pale yellowish.

Habitat. — Off Wild Island, Admiralty Islands; reefs; March 1875.

Family APLYSiNiDiE, Vosmacr.

Aplysince (e.p.), Hyatt.
Ceratina (e.p.). Carter.
Aplysinidce (e.p.), F. E. Schulze, v. Lendenfeld.

Keratosa with small either hemispherical or pear-shaped flagellated chambers com-
municating with exhalent and inhalent cavities, each by means of one comparatively long
and narrow inhalent and exhalent canaliculus. Axis of fibres thick ; ground-mass in
the neighbourhood of the flagellated chambers overloaded with granules.

L^iffaria, 0. Schmidt.

Aplysinidse with thick-walled heterogeneous skeletal fibres, their central part but
little difi"erentiated optically from the surrounding horny lamina.

Luffaria variabilis, n. sp. (PI. IX. figs. 1-6).

As I remarked before (p. 33), I am not prepared to say whether all Liiffarice in
addition to thick skeletal fibres also possess fine ones. But should such be the case,
the species I am going to describe may still be readily distinguished from all others of
the genus by the following peculiarities ; its skeletal fibres are not at all so glass-like
and fissile as Schmidt states with regard to the specimens he had for examination.
Again, though on the whole its skeleton admits of comparison with a burst but still
only with a very imperfect one, the distance between the prominent outer fibres reaching
1 to 2 mm. In addition to some small fragments, the species is represented in the
Challenger collection by two sj)ecimens, one of massive shape, the other on the contrary

70 THE VOYAGE OF H.M.S. CHALLENGER.

in the form of three elongated stalks connected together at their bases. I do not think
that this difference in external appearance can justify the establishment of two varieties ;
for in the structure of their soft parts (quite identical -ftdth that of Aplysina), and again
in that of the skeleton, the specimens do not differ at all. Indeed the colour of the
parenchyma in one case (massive specimen) is rather yellow, in the other greyish, but I
am not inclined to ascribe any consequence to this difference. Both the specimens are
represented on PI. IX. I have nothing further to add to this illustration, and as to the
properties of their skeleton in general and to those of their skeletal fibres I refer the
reader to page 7 of this report.

Colour. — Outer surface greyish, parenchyma yellow and dirty greyish-white, skeletal
fibres brownish-yellow and brownish.

Habitat.— ^iAtion 177, August 18, 1874, off Api, New Hebrides; depth 60 to 70
fathoms. Off Tahiti ; reefs; September 1875.

Verongia, Bowerbank.

Aplysinidae, the central axis of whose thick-walled skeletal fibres is readily to be
distinguished optically from the surrounding horny walls.

Verongia hirsiita (?), Hyatt (PI. X. figs. 1-3).

Verongia hirsuta, Hyatt, Eevision, &c., vol. i. p. 403.

I leave to later investigators the decision of the question whether the form I am
going to describe is identical with Verongia Jdrsuta, var. Jistularoides of Hyatt.
His description is very short, and there are no explanatory illustrations ; it agrees,
however, closely with what I can give of the form in cjuestion. I myself indeed
should call the meshes of the skeleton regular, but wath regard to this difference the
utmost caution is advisable. In Verongia Jistularis [Spongia Jishdaris, Linnd) from the
Museum of Erlangen, which both Hyatt and myself were able to obtain for the purpose
of comparison, and which is placed by Hyatt in his family Dendrospongiadae,
characterised inter alia by irregularity of the meshes, I was able to discern meshes of
a geometrically regular polygonal character, and again meshes sometimes of irregularly
round, sometimes of irregularly oval, shape.

As to the anatomy of the form, I am glad to confirm the supposition of Vosmaer ^
that it does not differ from that of Aplysina; the exposition of the canal-system
on PL X. fig. 7 has been made after a preparation of Verongia tenuissima, but it may
be applied also to Verongia hirsuta, the more so as both the specimens of Verongia
hirsuta and Verongia tenuissima, represented in the collection each only by a single

' On Vclinea gracilis, p. 444.

REPORT ON THE KERATOSA. 71

specimen, proved to be full of spermospores. To a more detailed description of these
latter I shall return later, and will now merely call attention to this peculiarity of the
skeletal fibres of my Verongia hirsuta, that they almost all proved to be covered with
small plates of polygonal outline as represented on PI. X. fig. 3 ; I have been unable to
make out their origin. Occasionally, and particularly on young fibres, I found no such
plates, but small drops of yellow substance at a comparatively great distance from one
another. I can but state, and this with the greatest certainty, that these plates are
not modified spongoblasts. I regard them as the last product of the spongoblasts, ready
to lose their spongoblastic properties in order to become common stellated cells.

CoZoztr.— Pale rose-brownish, skeletal fibres dark brown.

Habitat. — Ofi" Bermudas, June 1873; reefs.

Verongia tenuissima, Hyatt (PL X. figs. 4-7).

Verongia tenuissima, Hyatt, Revision, &o., vol. i. p. 403.

Hyatt's work upon the Keratosa is so very poor in explanatory illustrations that
although the short diagnosis he gives to his Verongia tenuissima is entirely applicable to
the form I am going to describe, I do not feel quite certain whether both these forms are
identical, but if not, at any rate they are very closely allied to one another, and to be
probably distinguished merely as diflerent varieties of the same species.

The form is represented by a single specimen. It is of fistular shape, the central
cavity being funnel-like, with a circular upper extension of 22 mm., and the walls 20 mm.
thick in the basal and middle parts of the body, growing rather thinner towards its
upper end. The inner surface, in contrast to that of Verongia hirsuta, is smooth but
undulating, while the outer surface is hilly ; the surface both of hillocks and depressions
being shagreen-like, and studded wdth projecting points of the skeletal fibres. These
latter — in thorough harmony with Hyatt's statement on the point — are far thinner and
more elastic than those of Verongia Jistularis, yet at least one and a half times as thick as
the fibres of Ai^lysina aerophoha; their average diameter is 0"2 mm., and the meshes
formed by them recall vividly in size and shape those of Aplysina. The internal
organisation, both anatomical and histological, agrees so closely with that of Ajilysina
aerophoha that having illustrated it l)y a small drawing (PL X. fig. 7), I can refer the
reader to F. E. Schulze's paper on the Aplysinidse. I must add, however, that I was not
able to discern the bundles of fibrils which he describes ^ and represents on pi. xxii.
fig. 14. of his paper. But I was fortunate enough to discover the male generative product,
the spermospores ; and this both in Verongia hirsuta and Verongia tenuissima, in this
latter form together with ova, scantily scattered in the peripheral parts of the body,
while the spermospores have been found everywhere and in abundance. I call them

1 Zeitschr.f. wiss. Zool, Bd. xxx. p. 397.

72 THE VOYAGE OF H.M.S. CHALLENCxER.

spermospores aud not sperm-balls, as I did in the description of Carteriospongia radiata,
for they are quite difi'erent in appearance and structure. In Carteriospongia radiata, and
in other instances, these formations present an aggregation of cells, if not ripe spermatozoa,
these cells being each equivalent to the other, and the whole lying in a capsule formed
by numerous pavement cells ; in Verongia the analogous capsule is formed by a single
cell, the endothelium is not to be found at all. In one word, the male generative
products are quite equivalent to those of Calcarea as I have described them in Sycon
raphanus} I must confess I can find no explanation of this ; the matter appeared to me
strange even before, when I thought the Calcarea were opposed in this respect to the
ivhole group of Silicea. I was thoroughly startled when, hand in hand with indubitable
sperm-balls in Carteriospongia as well as in some other Keratosa of the collection,
I discerned in my preparations of both species of Verongia the most typical spermo-
spores. I was not able to follow out their development, but when ripe they recall so
vividly the corresponding formations in Sycon raphanus (with the sole distinction that
while in this latter instance the nucleus of the covering cell in quite ripe spermospores is
in most cases indistinct, in Verongia I find quite empty capsules, nevertheless, provided
with it) that I must identify both these formations ; on the other hand, I have now no
doubt that at the development of the sperm-halls no covering cell is formed, that its
description by F. E. Schulze^ agrees closely with the reality. I have repeatedly
examined Oscarella {Halisarca) and Aplysilla, and was able to follow out the develop-
ment of their sperm-balls from tlie first beginning, and I can only affirm the former
statements of F. E. Schulze, that we have in these instances to do with real division of
the sperm cells, and that there is no covering cell to be discerned. On the whole the
enigmatic question requires a special and extensive investigation, the more so that it is
not only of spongiological but also of general interest.

Colour. — External surface black, that of the central cavity and parenchyma grey,
skeletal fibres brownish yellow.

Habitat. — Oft' Barra Grande, September 10, 1874 ; depth, 400 fathoms.

A perusal of the preceding pages shows that of thirty-four determinable forms brought
home by the Challenger Expedition, almost two-thirds (21) have been found to be new,
and that only three species [Psammopemma densum, Luffaria variabilis, and Cacospongia
collectrix) were obtained from more than one locality. Under such conditions any
discussion of the geographical distribution of the Keratosa would be possible only if
extended over all the Keratosa hitherto described. It must be said, however, that on

1 Sitzungsh. Alcad. d. TFiss. JFien, Bd. Ixxxvi., 1882.
- Zeitschr.f. wiss. Zool., Bd. xxviii. p, 24, pi. iii. flg. 19.

REPORT ON THE KERATOSA.

/ .J

the whole, our knowledge of the group is in this direction so very fragmentary that even
in this case the possibility of any deductions and generalisations is entirely excluded.
The annexed Table has, accordingly, no other object than to render conspicuous the
comparative richness in Keratosa of the difierent localities visited by the Challenger, as
well as the depths from which they have been obtained. The depth (in fathoms) is
indicated by the figures in the columns, s.w. = shallow water.

BATHYMETRICAL TABLE.

Brazil.

Eastern Coast
of Australia.

w

lanthella flabdliformis, Pallas,

Spongelia spini/era, F. E. Schuke,
„ pallescens, 0. Schmitlt,
„ horrida, Selenka,

Psammoclenia ramosum, Marsh.,
„ vosmaeri, n. sp.,

,, /oUaceum, n. sp.,

Psammopemma densum. Marsh.,
,, porosum, n. sp., .

Coseinodeniia confragosum^ n. sp.
,, denticidatum, n. sp

,, altum, n. sp.,

Euspongia officinalis, I.inn. , var. luhosa

Uippospongia anomala, n. sp.,

,, mauritiana, Hyatt,

Cacospcmgia levis, n. sp. , . .
,, amorjiha, n. sp., .

,, murrayi, n. sp., .

„ vesictdifera, n. sp.,

,, procuvibens, n. sp.,

,, ^d^ndroides, a. sp.,
,, spinifera, n. sp., .

,, tvherculata, n. sp.,

,, intermedia, n. s^.,

,, irregularis, n. sp.,

„ oligoceras, n. sp. , .

„ compacta, n. sp., .

„ collectrix, F. E. Schulze,

Stelospangos longispin-us, Duch. and Midi

Carteriospongia otahitica, Esper,
„ radiata, Hyatt,

Luffaria variabilis, n. sp., . .

Verongia hirsuta, Hyatt, . .

,, tenuissima, Hyatt,

28

3S-40
38-48

220

60

60-70

400

30-35

7

220

38-40

60-70
60-70

28

18

18
12-20

400

400

60-70

(ZOOL. CHALL. EXP. PART XXXI. — 1884.)

Hh 10

IV.-CONCLUDING REMARKS.

With the last lines of the foregoing chapter the direct and immediate task of this
memoir is accomplished, but it remains still to summarise its more general deductions
with respect to what I regard as its chief purpose. The tendency of this is to clear up
the present state of our knowledge, and in the first place to prove that, in spite of the
acquirement of many new systematically important characters introduced into descriptive
Spongiology by F. E. Schulze, which concern the peculiarities of the internal organisation
of the soft parts, all the arrangements of the Keratosa hitherto proposed are, on
the whole, far from being natural. I hope I have made it clear that the procedure
of subdividing the group either directly into families or preliminarily into suborders may
be adopted merely as a measure of provisional character. I hope I have also proved
that the genera established in the Keratosa are not homogeneous, the characters dis-
tinguishing one part of them being of an absolute, those distinguishing another part of
an extremely conditional, and often very ambiguous, nature. I am, finally, far from any
illusions, and feel certain that the perusal of the descriptive part of this memoir, as well
as of other systematic papers on the Keratosa, must show plainly that in most cases the
classifier has to found his species on characters whose stabiUty is quite unknown to him,
so that whether they are really of specific importance or fit only for distinction of indi-
viduals, remains open to discussion. The reader sees that the state of matters is far from
satisfactory; and it is natural to ask whether it admits at least of some conjectures as to
the way in which our knowledge of the Keratosa may become more perfect ? I believe
this question may be answered in the affirmative ; but since, in cases where decisive,
incontestable proofs are wanting, the probability of the suggestion depends on the
number of arguments, I should like, before I pass on to the recapitulation of the grounds
favourable to my opinion, to increase their number by an argument of no little consequence.
This argument refers to the systematic position of the Keratosa in the whole group of
Porifera, and it appears to me that a detailed discussion of this question will be the
less superfluous, as the corresponding conclusions promise to be not only of theoretical
importance but also of practical applicability. There are mainly in practice two different
methods of systematic procedure. The usual one is to begin with lower systematic
unities, in order to ascend to families, orders, &c. With respect to many other instances

76 THE VOYAGE OF H.M.S. CHALLENGER.

the matter is quite different, and it depends very often exactly on our conceptions as to
whether the given group forms a class or a family whether we subdivide it directly into
genera or preliminarily into orders and suborders. Now we have seen that there exist
in science quite opposite opinions as to what the genus ought to represent, and that
many naturalists find no absolute distinction between genus and species. We have also
seen that in the Keratosa, to use the words of Prof. F. E. Schulze,^ " die Entscheidung
der Frage, ob eine Anzahl verwandter Formen als Arten einer Gattung oder als Varietaten
einer Art hinzustellen sind, oft besonders schwderig erscheint." Be that as it may, the
given forms must be classified and introduced in the system, and the hesitation of the
classifier must come to an end. On what now may his final decision depend ? That it
may dejjend on the inspiration of the moment is undeniable, but I think the importance
of this latter factor must not be exaggerated. Every conscientious investigator will
always search for more positive arguments, and it is plain that if he regard the corre-
sponding group as an order subdivisible into families, he will bring his hesitation to
an end by creating a new genus ; and, on the contrary, if he believe the group to be only
a family, he will describe the series of forms in question as a species with varieties.
Accordingly, and as I remarked before, from this point of view a reliable answer to the
question put on the preceding page is the most desirable.

Numerous and very conflicting opinions have been expressed on the problem of the

affinities of the horny sponges. I begin with that of Oscar Schmidt. This naturalist does

not deny the close relationship of his Ceraospongise with the Monactinellida, and namely

with Chalinidae, moreover, he regards them as forming no larger systematic unity than a

family; but he considers ^ them to present an independent natural group, and is decidedly

against any introduction of true Chalinidas into it. Hyatt goes still further. He regards

the Keratosa as forming an indej^endent order, a very significant statement, since the

naturalist just named, when writing that " the characteristics of the order Keratosa are

more clearly defined than those of any other among the class Porifera," &c. (Revision,

&c., vol. i. p. 399), was of the opinion that the whole group of Porifera form nothing

more than a class of Infusoria.' There are in his valuable memoir on the North American

Porifera no further explanatory observations in this direction, but it is plain that the

words above quoted admit of but one explanation, namely, that the Keratosa are to be

regarded as a group systematically equivalent to the groups Calcarea and Silicea. Gray *

and Bowerbank,^ in harmony with Grant, call the Keratosa also an order, but they

class within it the true horny sponges with sponges producing "proper spicules."

Finally, Carter," agreeing on the whole with Gray and Bowerbank, differs from them in

this point, that he considers the Keratosa to represent two orders, without foiming,

' Zeitschr.f. wiss. Zool., Bd. xxxii. p. 612.

2 Spong. d. adriat. Meeres, II. Suppl., p. 9 ; Spong. d. Kuste v. Algier, p. 36.

' Revision, &c., vol. ii. p. 481. * Proc. Zool. Soc. Land., 1867, p. 503.

^ Monograph Brit. Spong., vol. i. p. 205. " Ann. and May. Nat. Hist, ser. 4, vol. xvi., 1875, p. 13?.

REPORT ON THE KERATOSA. 77

however, an upper group. This latter statement demands an explanation. The fact is
that Carter does not follow the, so to speak, " dendroid " principle of classifying Porifera
recommended by Gray ; he does not adopt his subdivision of Porifera into two chief
groups, that of Calcarea and that of SUicea,^ but subdivides the whole type or subtype
of Porifera into eight equivalent groups (orders), leaving the reader entirely uncertain
how the mutual affinities of these eight orders are to be graphically expressed, — whether
by eight radial branches from the same spot, or by one chief branch with secondary^
tertiary, &c., ramifications. I lay great stress upon this. Our present aims are not only
of purely systematic, but also of phylogenetic, character, and the more the most
experienced spongiologist of our day, Mr. Carter, has the right to express his opinions on
the question, the more one must feel disappointed to find in his system an arrangement
which, neglecting the usual laws of the systematic, and yet not accompanied with
an explanatory genealogical tree, leaves the reader in the mist of uncertainty as to
whether its author regards, e.g., the Calcarea as forming a group systematically equivalent
to that, for instance, of Psammonemata, the Ceratina equivalent to the Hexactinel-
lida, &c., or not. Apart from these more general questions, the opinions of Mr. Carter
as to the affinities of horny sponges are clear ; in harmony with Gray and Bowerbank
he composes his third family of Psammonemata (Pseudohircinida) of groups devoid of
proper spicules and again of groups provided with them. The opinions of difi'erent
spongiologists as to the question I am now discussing are thus very contradictory and
even conflicting, but although most of them are expressed in very decided language, it
would yet be a hopeless task to search into their papers for any grounds in favour of
suggestions upheld by them ; these grounds are shrouded in darkness. Moreover, the
decided language just alluded to is often at variance with other suggestions of the same
author. Thus, for instance, with respect to 0. Schmidt. On page 36 of his work on the
sponges of Algeria he lays stress on his Ceraospongiae and Chalineae being quite natural
and independent families; on the following page of the same work he says that he feels
certain that " gewisse Gattungen einer Chalinese zu nennenden Familie unter sich weit
weniger als mit bestimmten Gattungen der Ceraospongise, direct verwandt sind." Under
such circumstances there remains no other way of elucidating the matter but to turn to

' Dr. Vosmaer (Report on the Sponges dredged by the " Willeni Barents," p. 3), agreeing with this proceeding, but
not content with the designation of Silicea, since it embraces forms like true Keratosa or Myxospongije, i.e., forms devoid
of any siliceous spicules, proposes for it the name of " Non-calcarea." I believe this to be scarcely an amelioration. There
are indeed cases when negative designations are very fortunate. On the whole, however, they are not to be recom-
mended, and certainly not with respect to the Non-calcarea of Vosmaer. The name of, e.g., Acrania, Haeckel, being
negative, includes nevertheless a systematically important positive allusion. That of Non-calcarea is devoid of it.
I agree that it is illogical to call SUioea inter alia sponges without arjy siliceous spicules, but though illogical this
designation is in possession of a phylogenetic sense. As the reader will see later, there can be no doubt as to the
fact that Keratosa, like Myxospongias (these latter perhaps not directly, but this does not alter the case), owe their
origin to siliceous sponges, and from this point of view there can be no objection to the grouping of all these closely
allied sponges under the general name "Silicea" — as nobody finds it strange that the order of Diptera includes forms
like Pulex irritans or the type of Arthropoda animals like Rhizocephala or Tardiyrada.

78 THE VOYAGE OF H.M.S. CHALLENGER.

the actual data, viz., to the differences in structure distinguishing the horny sponges from
other Silicea. The usual and natural characteristic of the Keratosa is the following: —
Porifera with horny skeleton devoid of proper (siliceous) spicules. This diagnosis alludes
to this, that there must exist amongst Silicea, sponges although provided with a true
horny skeleton, yet characterised by the possession of spicules produced by the sponge
itself This allusion concerns the groups of Silicea known under the name of Chalinidse,
whose main systematic character consists in the possession of a horny skeleton recalling
as to its external stracture that of true Keratosa, but rich in horny substance as it is, yet
containing within its fibres proper spicules enclosed. Now through the genus Chalinula,
0. Schmidt, the Chalinidse are most closely allied to ty^jical Monactinellida. There are
accordingly between a typical horny sponge and a typical Monactinellid long series of inter-
mediate connecting stages, and their existence proves that the Keratosa and Monactinellida
must have had the same phylogenetic origin. This has never been disputed ; and, on the
whole, it is in thorough harmony ^dth embryological data also. The larvae of Keratosa as
described by Barrois^ [Verongia [AplysiUaf] rosea) and F. E. Schulze [Euspongia
officinalis,^ Spongelia pallescens,^ A'phjsiUa sidphurea*) and those of Chalintda fertilis
and Reniera fiUgrana as described by Keller' and Marshall," as well as their pre^dous
and probably further development, admit of no absolute distinctions. Indeed, while the
usual mode of division of the ovum is equal, that of the ova of Clialinula is, according
to Keller, unequal. But, firstly, this diS'erence is of a very subordinate nature, and,
secondly, it is still questionable whether this statement of Keller is more rehable than
his suggestion as to the sexual dimorphism of the species in question. There can be, I
repeat, no doubt as to the Keratosa and Monactinellida having had the same origin.
But the matter, indisputable as it is, can be interpreted diS"erently. The genealogical tree
accompanying the paper of Prof Schmidt on the sponges of Algeria (Joe. cit., p. 35) shows
that this naturalist considers the Keratosa to be an older group than the Monactinellida,
to represent, namely, a group from which the true Silicea have originated. If this be true,
the systematic proceeding of Hyatt I have spoken of a couple of pages before would
receive a thorough sanction, and the class of Non-calcarea, Vosmaer (for in such a case the
designation of Silicea applied to the group by Gray would be no longer admissible), would
require to be subdivided into two orders, Keratosa and Silicea. This suggestion is,
however, far from being reliable, and a short deliberation renders it obvious. I ask
what appears more easily and naturally realisable, the transformation of a SiHceous into a
Keratose sponge, or vice versa of a Keratose sponge into a Monactinellid. I think there
can be no doubt as to the answer. In the species Chalina limbata, Bk., we have to do
with a sponge whose skeletal fibres are extremely poor in proper spicules ; an insignificant

' Ann. d. Sci. Nat. (Zool), scr. 6, t. iii., 1876, p. 56. ^ Zeitschr. f. wiss. Zool., Bd. xx.\ii. p. 642.

3 Ibid., Bd. xxxii. p. 144. * Ibid., Bd. .\sx. p. 414.

5 Ibid., Bd. xxxiii. p. 317. « Ihid., Bd. xxxvii. p. 221.

REPORT ON THE KERATOSA. 79

qviantity of tliem is, however, still to be found. Are these spicules to be regarded as
physiologically functional, however it may be ? Can we assume that a single row
of siliceous spicules render the horny fibres of Chalina limbata more hard ? Can we
ascribe to them any other than phylogenetic significance ? Of course not ; and, on the other
hand, it is also plain that a sudden apparition of spicules in the horny fibres previously devoid
of them is thoroughly inexplicable. On the contrary, the substitution of a horny skeleton
for a skeleton represented by siliceous spicules is very simply imaginable. In numerous
Silicea we have to deal with forms whose siliceous supporting apparatus forms throughout
a network in the proper sense of the word. In many other Silicea we meet with forms
whose skeleton is represented by spicules lying exclusively isolated ; there are again
amongst them forms whose skeleton is constituted by spicules aggregated in rows. That
a skeleton represented by a compact network is of a firmer build than a skeleton repre-
sented by spicules lying isolated is evident. From this point of view it becomes clear
that had a sponge' once received the faculty of cementing the rows of its spicules with
any connecting substance, this acquirement would have proved to be of a great profit to
it. It is also equally plain that if the cementing substance proved to be equally hard
and steadfast, and at the same time more elastic than the siliceous spicules, it might, in
the course of time, have forced away these latter entirely. I hope this theory gives a
very simple explanation of the phenomenon. It is also in harmony with the facts
concerning the structure of the soft parts of Keratose sponges, and, on the other hand,
of Chalinidae, Renieridse, &c. Of course the literature of the question is very poor ;
some remarks of Dr. Vosmaer^ on the structure of the Renieridse, as well as a couple of
observations on the anatomy of the Chalinidae by Dr. Keller f the observation of this
naturalist^ on the structure of Reniera semitubulosa executed under the influence of
Prof. Haeckel's statements on the non-existing racemose type of the canal-system being-
unreliable ; they are, however, quite sufiicient for our purposes, and with regard to the
Renieridse I can also make use of my own investigations. As to the Chahnidfe, the
drawing by which Dr. Keller illustrates the internal organisation of his Chalinula fertilis
{Joe. cit., pi. xviii. fig. 1) cannot be misinterpreted; the ground-mass being devoid of
any granules, and the flagellated chambers of special cameral canaliculi, it is clear that we
have here to deal with the type of the canal-system characteristic of Spongelidte, and the
size of the flagellated chambers being, according to Keller (loc. cit., p. 327), 0"02 mm. on
an average, and their shape round, with that special modification which distinguishes
my genus Psammoclema. As to the internal structure of Renieridse, I difi"er somewhat
from Dr. Vosmaer. Indeed, his remarks are extremely short. He states, however,
that their anatomical organisation recalls \'ividly that of Eiiplectella aspergillum, as

' Voorloopig berigt omtrent h. onderzoek aan de Nederl. werktafel in h. Zool. Stat, te Napels, Ilaag (?), 1881.
2 Zeitschr. f. wiss. Zool, Bd. xxxiii. p. 326. ^ Ibid., Bd. xxx. p. 579.

80 THE VOYAGE OF H.M.S. CHALLENGER.

described by Prof. F. E. Schulze.^ Now, so far as the delicate construction of Eenieridse
is concerned, I can but confirm Dr. Vosmaer's statements as to the striking resemblance
of these Monactinellida with Euplectella. Prof. Schulze lays stress, however, also on the
fact that in Euplectella aspergillum—andi as he told me in HexactineUida in general —
the flagellated chambers are comparatively very large, and, in appearance, typically
pouch-shaped, recalling the radial tubes of the Sycouida3. In the species of Reniera I
had for examination, Reniera aquceductus, Reniera Jiligrana, Reniera semitubulosa,
Reniera Jibulata, and two or three Eenieridse not determined, I found then' flagellated
chambers to be always of roundish outline, therewith not larger, or at least but little
larger, than those of, e.g., Psammoclema vosmaeri. I would be, however, scarcely right
to lay stress on the contradiction in question, for, firstly, it is but too possible that there
are representatives of the genus Reniera with radial tube-like flagellated chambers,
and, secondly, the difi"erences between flagellated chambers of this kind and those
characterising my genus Psammoclema are of a thoroughly quantitative nature. At any
rate, it is clear that, so far as the Monactinellida, the most closely allied to the Keratosa,
are concerned, they are characterised by an arrangement of the canal-system of a more
primary character than that distinguishing the majority of the Keratosa. Is this not
an evident proof that they are to be regarded as palseontologically older sponges ? ^ I
think aU these circumstances together speak so decidedly for the supposition I am now
asserting that the matter can be regarded as scientifically proved. This deduction is
of great consequence, for under these conditions there are absolutely no grounds for
regarding the group of Keratosa as an order, i.e., a systematically higher unity than the
families Chahnidae, Eenieridse, &c., the more so as it is even impossible to say that
Keratosa are less closely connected with Chalinidse than these latter with the Eenieridse.
Of course the thorough absence of proper spicules in their skeletal fibres admits
of their very sharp diagnosis, while the diagnoses of Chalinidse and Eenieridse are
of a more conditional nature ; but who can warrant that the genus Spongelia is in
closer relationship with Euspongia tlian with Chalina? who can guarantee that
the relative characters distinguishing Spongelia from Euspongia, and concerning the
internal organisation of the soft parts, are of less importance than the equally quanti-
tative distinctions concerning the properties of the skeleton difi"erentiating the

1 Trans. Roy. Soc. Edin., vol. xxxix., 1880, p. 661.

2 The type of canal-system characterised hy an entire absence of special cameral canal iculi, and by clearness of
the ground-mass surroimding the flagellated chambers, can be regarded as chai-acteristic of ^Monactinellida in general.
Apart from the Renieridoe above mentioned, I can state this with respect to the following fonns I had the opportunity
of examining — Suberites domuncnla, N.; Espcria bauriana, O. S.; Myxilla rosacea, Lbn.; Myxilla vcneta, 0. S.; Baspailia
viminalis, O. S.; Acanthella acuta, 0. &.; Axinella polypoides, 0. S. On the contrary, the canal-system of the repre-
sentatives of the genus Papillina, 0. S. {Papillina stiberea, Papillina nigricans), is not less highly developed than
that of Aplysina aerophoba or Corticinm candelabrum ; but it must be added that the genus Papillina, although
undoubtedly closely allied to the genus Suberites, seems also through the genus Osculina, 0. S., to be stiU more closely
connected with the Chondrosidre, and may, together with these latter, represent a family palieontologically not less
recent than that of Ceraospongite.

KEPORT ON THE KERATOSA. 81

Spongelidse from the Chalinidse ? Indeed, the most simple procedure is to adopt the
Keratosa as an independent group ; but it should not be overlooked that this procedure
is nothing more than the concession to our natural wish to have for the groups we
establish the sharpest possible diagnoses.

I should like to summarise my conclusions. We have seen that the subdivision of
the Keratosa into two orders is inadmissible ; we have seen that their subdivision
directly into families gives also as results families of a very ambiguous nature. After the
above deduction we can but say that all this is very comprehensible, since the ivhole
group is nothing more than a simple family. Of course, as is the case with the
suljtype of Acrania, a high systematic subdivision can be represented by a simple
family; and, on the other hand, as is the case with, e.g., Terebellidae, a family can be
subdivided into numerous subfamilies, these latter consisting again of generic unities.
This latter measure finds, however, its application in exceptionally rare cases, and only
then when it is really necessitated by the richness of the forms as well as by the richness
of systematic characters, and, on the other hand, by higher phylogenetic considerations,
while the immediate purpose of my foregoing discussion consisted precisely in the attempt
to prove that it is due exactly to the erroneous opinion that Keratosa forms a systemati-
cally high subdivision, that natviralists split them into orders, suborders, and families.
Of course, it cannot be denied that certain genera established in them are more closely
connected one with another than with the remaining representatives of the group. This
would be, however, only of consequence if all the genera in question were homo-
geneous, while in reality some of them are undoubted genera, the others perhaps but
species. This is the gist of the matter, and I think that the only natural reconciliation of
all these contradictions can be obtained by rendering our genera equivalent one to
another, which can be realised by enlarging the idea of genus, e.g., by uniting forms,
distinguished as Hippospongia, Euspongia, &c. , in the single genus Spongia, which would
be, on the whole, thoroughly equivalent to the genus lanthella or Darwinella. But if the
species constituting the conjectural genera Hippospongia or Stelospongos are yet undoubted
species 1 I answer, prove that they are so, and in that case subdivide the genus Spongia
into corresponding subgenera. As is well known, these latter systematic unities are out
of use ; I regai'd, however, their introduction in systematic practice to be equally profit-
able for systematic purposes in general, as well as with respect to the special case of
classifying the Keratosa in a tolerably natural manner. I opened my " criticism of the
genera " with a comparison of different opinions as to the value of generic distinctions,
and we have seen that in this respect diametrically opposite ideas have been expressed by
different naturalists. The word "diametrically" just used alludes to the impossibility
of their thorough reconciliation; the introduction of subgenera in zoological calcula-
tions would reconcile them at least so far as this is possible, and again it is obvious
that sooner or later this reconciliation must be realised, since neither the opinions of Nageli

(ZOOL. CHALL. EXP. — PART XXXI. — 1884.) Hll 11

82 THE VOYAGE OF H.M.S. CHALLENGER,

nor those of his opponents can be adopted without amendments. As to the suggestion
that generic distinctions must be of an absolute character, of course the acquiescence in
this demand would place the creation of genera beyond the discretion of classifiers, and
thus expel for ever from Science the disputes as to whether this group is to be regarded
as a species or as a genus ; it is however evident that, strictly and exactly prosecuted,
this demand would lead to the most strange and unnatural systematic arrangements.
Following it we should be obliged to unite all the Keratosa, with the exception of the genera
lanthella, Darwinella, and perhaps Psammopemma with all Chalinidse, Renierid^, perhaps
all Monactinellida into a single genus systematically equivalent to that of Danoiniella or
lanthella. For there are no absolute distinctions between Chalina or Reniera and
Spongelia, and there are no absolute distinctions between Spongelia and Etispongia and
Aplysina, and again there exist such distinctions between most of the Keratosa and ]\Ionac-
tinellida on one hand, and lanthella, Darwinella, and Psammopemma on the other. On the
whole, what Prof. Nageli recommends is applicable only to the palgeontologically old groups
of plants and animals. But it is not less evident that an unlimited discretion as to the
creation of genera would render any systematic progress impossible; to those who feel
disinclined to agree with me, I can but recommend an attentive perusal of spongiological
systematic literature. The using of a varietal character, as of generic value, condi-
tions the establishment of new apparently highly interesting and deviating species
out of forms representing nothing more than by no means instructive varieties,
if not individuals of very common species belonging to another genus. It is quite
possible that I have myself committed the same mistake, having adopted F, E. Schulze's
genus Hippospongia, and created a new species H.ipp)ospongia Tnauritiana, while this
conjectural species is very possibly nothing but a variety or subspecies of Spongia
(Euspongia) lapideseens. Both these opinions cannot thus in their entire extension be
adopted. In my paper on the Challenger Calcarea an attempt to reconcile them has
been made. Whilst subdividing this group into genera I called attention to my
intention to execute this task according to the whole of their organisation, by taking
into consideration aU their organs in their mutual correlation ; and with respect to
the group just mentioned the carrying out of this principle met no difficulties,
and I hope that the genera I have there established are really natural and adoptable.
But there are very often cases when such a proceeding is impossilile, when large groups
of forms differ from one another only in a single character, the conjectural specific
distinctions being in one group complete analogues to those in the other. To similar
instances the principle I have followed in my above named paper is not applicable, and for
my own part I see no other issue but to recommend for such cases the adoption of the scheme
of Nageli, and this in order to protect the establishment of genera so far as possible from
the vagaries of classifiers, so that generic unity might serve as a firm basis, which has
been wanting in descriptive zoology since the mutability of species was actually proved.

EEPORT ON THE KERATOSA. 83

As I remarked before, the realisation of this demand necessitates the introduction of
subgenera ; and I am the more in favour of this because otherwise I see absolutely no
possibility of arranging the Keratosa in a manner not conspicuously artificial, while in the
contrary case we should obtain a very natural family of Ceraospongise with the following
quite homogeneous genera: — lanthella, Dartvinella, Simiilicella (with two subgenera
Ai^lysilla and Dendrilla), SiMngia (embracing forms classed now in the conjectural
genera Euspongia, Hipi^o&pongia, Cacospongia, Stelospongos, and Coscinoderma),
Phyllospongos (with subgenera Phyllospongia and Carteriospongia, provided that these
subgenera are really in a close relationship), Spongelia (with subgenera Dijsidea and
Psammoclevia), Psammopemma (supposing that this conjectural genus be not connected
with Spongelia by intermediate stages), and Velaria (including forms referred at the
present time to Ap>lijsina, Verongia, and Luffaria, which may perhaps all be elevated
to the rank of subgenera); and it is clear that each of these genera either admits of a very
sharp diagnosis or evidences its right to be regarded as a genus by series of characters
concerning the whole of the organisation. If now the question be asked, why instead of
following, in the descriptive part of this memoir, the scheme just exposed, I have yet
followed the arrangement of Dr. Vosmaer, the answer will be because the above scheme
only partly decides the problem of a natural arrangement of horny sponges, and I myself
am of opinion that when new systems are proposed they must be well established in all
their parts. The reader will remember that in the foregoing three chapters we met with
a large number of contradictions issuing from the present mode of classifying the Keratosa.
The scheme in question reconciles most of them ; it does not do so, or at least but partly,
with respect to what I called circulus vitiosus, characterising the mutual affinities of the
genera of the group. This circulus vitiosus is striking as concerning the genera belonging
to different families ; it has, however, equally little right to exist as concerning the mutual
relationships of the subgenera. It would have been very easy to proclaim the Keratosa
as forming but a single family, and to classify according to this the specimens of the
Challenger Collection ; but by this proceeding we should not have got rid of difficulties
concerning questions of course of a more subordinate nature but nevertheless of vital
importance ; this proceeding would not have decided the problems as to whether sub-
generic value can be really ascribed to the characters distinguishing the HippospongicB or
Coscinodermata, &c. This is the second, and the most difficult part of the task, and
without the help of Palseontology and Comparative Physiology it will be scarcely decided.
As must be evident from the above discussions, by the term " subgeneric character " I
understand a character containing in it a new principle of organisation, the corresponding
representatives of the group being connected by intermediate stages. That Euspongia,
Cacospongia, Hippospongia, &c., present each in their organisation what may be called a
new principle is clear ; but it is by no means evident whether this, their conjecturally
main character, be really constant. Should we feel certain that the fossils described as

84 THE VOYAGE OF H.M.S. CHALLENGER.

horny sponges do really belong to the Keratosa, that, e.g., Dysidea antiqua. Carter,' from
the Carboniferous system, is really a horny sponge and not a worm-tube or something of
that kind, in a word, should we feel certain that Keratosa is a palseontologically old group,
of course we should regard Hippospongia, Euspongia, &c., as representing each a
subgenus, and pay but little attention to the existence of forms like Velinea gracilis, Vos-
maer, or my Cacospongia intermedia conditioning the circulus vitiosus above mentioned.
But this is precisely the question to be answered; it is but too possible that Keratosa is a
very recent group, and in this case the many-sidedness of their affinities may be explained
by their very high variability, in which case only specific and varietal importance must
be ascribed to the characters we regard now as of subgeneric or even generic value.
Prof. F. E. Schulze established a new genus OUgoceras ; he told me that all specimens of
his OUgoceras collectrix were found between stones in a position which renders the
presence of a special supporting skeleton superfluous ; it is to be asked whether the loss
of this latter is immediately realisable or not. It is, in one word, necessary to expi*ess
more or less approximatively the proportion between the stability of these and other
characters and their mutability, i.e., their faculty of conforming to existing influences.
His genus Hippospongia, F. E.. Schulze characterises by the presence of numerous
channels, i.e., cavities breaking through the body of its representatives in diS'erent
directions. That this property is due to the necessity of enlarging the outer
surface is perfectly clear ; it is again to be asked whether this character can be
adopted in a short space of time. Mr. Carter established a genus Coscinoderma, but with
the diagnosis he gave ^ to it in his last paper the genus is not adoptable : " Sieve-like
incrustation, composed of foreign bodies uniformly foraminated and continuously spread
over the surface, whose evenness is not disturbed by the usual polygonal projection of the
subdermal fibre. Fibre fine, woolly." Should we follow it we should have this genus
represented but by the single species lanuginosum. I widened the diagnosis and de-
scribed as Coscinoderma also Coscinoderma altum, characterised by comparatively thick
skeletal fibres, in most cases cored with foreign enclosures. The type-specimen of
Coscinoderma possesses very fine fibres, all of the same diameter ; the skeletal fibres of
all CacospongicB are thick, and usually overloaded with foreign bodies ; they admit,
however, of the distinction of primary and secondary ones. Coscinoderma altum has
thick fibres cored with foreign bodies, but all its fibres are of the same size. Ought I
to class the form in question in the genus Coscinoderma or in that of Cacosp)ongia ?
Ought we to ascribe to the difierentiation of the fibres into primary and secondary ones a
higher systematic consequence than to their equal size ? The reader who has perused
my description of the Challenger specimens will find there such alternatives at every
step. And to sum up, so long as we possess no statements as to the stability of the
characters of the horny sjDonges we shaU have no natural arrangement of them. There

I Ann. and Mag. Nat. Hist., ser. 5, vol. i., 1878, p. 139. ^ Ibid., February 1884, p. 129.

REPORT ON THE KERATOSA. 85

was a time in Zoology when the external shape of the animals was considered quite
sufficient for descriptive purposes ; this period was followed Ijy the period of, so to
speak, anatomical character, naturalists discovered a new source of systematic characters
in Comparative Anatomy, and believed its help to be sufficient for deciding the most
complicated phylogenetic questions ; the apparent success was so great that this conviction
finds its echo in papers of very recent date — as for instance in v. Jhering's Monograph
on the Mollusea. The time came, however, when zoologists became aware that their
anatomical hopes were but illusions, and the word Comparative Embryology became
the watchword of the day. That this department of Zoology also has not justified the
hopes based upon it is but too well known, and the modern watchword is Comparative
Physiology. Whether this study, at the present time in its infancy, wQl justify what
it promises, will be seen later ; on the whole, it is clear that in most cases not this
alone but all branches of Zoology together will give us the answers we require ; but in
instances like that concerning the Keratosa, where almost all other lines of research are
of no avail, this new science may be particularly welcome. The possibility of a
disappointment is of course not excluded, and therefore further purely systematic papers
on the Keratosa are of course very desirable. It would he, however, still more desirable
that Science, hand in hand -^dth this, would follow up also another way, that, namely, of
Comparative Physiology. This is the immediate task to be executed, and in the case of
spougiologists residing near the sea-shore it is very easily realisable.

CONTENTS.

page

Introduction, . . . , . . . . .

1,2

I. Organisation and Classification of the Keratosa, .....

2-20

Prefatory Eemarks, .........

2

Systematic Value of Skeletal Fibres, whether Homogeneous or Heterogeneous, .

3

Systematic Value of Skeletal Fibres, whether Cored with or Devoid of Foreign Bodies, .

9

Systematic Value of the Peculiarity of Skeletal Fibres characteristic of lanthella,

11

Systematic Value of the Presence of Filaments, .....

12

Systematic Value of the Properties of the Canal System,

16

Corollaries, ........

18

II. Criticism op the Genera, ......

. 21-35

Darwinella, . . . .

21

Aplysilla, ........

22

Dendrilla, ........

22

lanthella, ........

22

Velinea, ........

23

Spongelia, Dysidea, Psammasciis, .....

23

Psammoclema, ... ."^ ... .

25

Psammopemma, .......

25

Euspongia, Hippospongia, Cacospongia, Stelospongos,

26

Cosdnoderma, ........

28

PJij/Uospongia, Carteriospongia, .....

29

Oligoceras, Hireinia, Ceratella, ...

29

Liiffaria, .... ...

32

Verongia, ........

34

Aplysina, .

34

III. Description of the Species, ... ...

37-72

Darwinellidse,

lanthella jldbelliformis, .....•••

37

Spongelidse.

Spongelia spinifera, .....-•..

41

Spongelia pallescens, .......

.

42

88

THE VOYAGE OE H.M.S. CHALLENGER.

Spongelia horrida,
Psammoelema ramosum,
Psamnioclema vosmaeri,
Psammoelema foliaceum,
Psammcrpemma densum,
Psainmopemma porosum,

Spongidae.

Cosdnnderma confragosum,
Cosdnoderma dentlndaium,
Gosdnoderma altum,
Euspongia officinalis, var. lohosa,
HipiMspongia anomala, .
Hippospongia mauritiana,
Cacospongia levis,
Cacospongia amorpha,
Cacospongia murrayi,
Cacospongia vesicidifera,
Cacospongia procAimbens,
Cacospongia dendroides,
Cacospongia spinifera, .
Cacospongia tubercidata,
Cacospongia intermedia, .
Cacospongia irregularis,
Cacospongia oligoceras, .
Cacospongia compaeta, .
Cacospongia collectrix, .
Stelospongos longispimts,
Carteriospongia radiata,
Carteriospongia otahitica,

Aplysinidae.

lAiffaria variabilis,
Verongia hirsuta (?),
Verongia tenuissima {1), .

PAGE

42
43
44
45
46
48

50
51
52
53
54
55
56
57
57
58
59
60
61
61
62
63
63
64
65
67
67
69

69
70
71

Bathtmetrical Table, .

73

IV. Concluding Ekmarks, ......

Short Eecapitulation of the Deductions of the Foregoing Chapters,
Affinities of the Keratosa and their place in the General System,
Definition of the ideas " Genus" and " Subgenus,"
Scheme of a Conjei-turally Natural Arrangement of the Keratosa,

75-85

75
76
81
83

Problems to be Solved La Order to Bender an Undoubtedly Natural Arrangement of them realisable, 83

PLATE I.

(ZOOL. CHALL. EXP. — PAfiT XXXL — 1884.) — Hh.

PLATE I.

lanthella fiahelliformis, Pallas; natural size.

n-ieVoya^e of H.M.S, .Challenge

- .-oJ-^v^^V ?^.' vr^tv ,

PLATE IT.

PLATE II.

lantlielJa flaheUiformis, Pallas.

Fig. 1 . Pt)rtiou of the outer surface, . . . . , x 6

Fig. 2. Portion of the .skeleton of the leaf-like extension ; natural size.

Fig. 3. Portion of the skeleton of the stem ; natural size.

Fig. 4. Portion of a section showing the arrangement of the canal system, . x 200

Fig. 5. Portion of a section with two flagellated chambers, showing histologi-
cal elements of the connective tissue, — Am, amoeboid cells ;
St, stellate and fusiform cells, . . . . . x 600

Fig. 6. Portion of the outer surface (from above) showing the subdermal

aggregations of fusiform cells and conjectural gland-cells, . x 600

Fig. 7. Portion of a longitudinal section through the horny walls of a

skeletal fibre, ....... x 700

The Voyage of H.M.S..>Challenger.

iio^a.Fl.n.

J.

r M

Si.

•:?

\"

'4W

:: ic

••

^ ^ri<,

••

»

«»

•».■

•i*

'

J}:

/

•/.

^^.^^^^

^swcv

5 I

fr^,5

? t ■■•

■-v^

g...

.'/

--^^

-'■s.^*^\

V '

"^i>

\y

^ r'

.'•:)

I

\

(/-

PLATE IJI.

(ZOOL. CHALU EXP. PART XXXI. — 1884.) HI:.

PLATE III.

Fig. 1. Portion of the skeleton of Spongelia pallescens, 0. Schmidt; natural size.

Fig. 2. Skeleton of Spongelia liomda, Selenka ; natural size.

Fig. 3. Fragment (?) of a specimen of Psammopemma densiim, Marshall ; natural size.

Fig. 4. Portion of a section through its body showing the form and disposition of tlie
flagellated chambers ; x 300.

Fig. 5. A colonial .specimen of Psammoclcnia vosmaeri, n. sp.; natural size.

Fig. 6. Portion of a section through one of its individuals ; x 300.

Fig. 7. Portion of the skeleton of Psammoclema foliaceum, n. sp.; natural size.

Fig. 8. The same of Psammoclema ramosnm, Marshall ; natural size.

The Voyage of H.M.S.„Challenger:

Keratosa.Pl.m.

V

/Jk, ^^%

^»:^

u

Im. iKiirVtiner yV\Kn, rianiSnt'/K

PLATE IV.

PLATE lY.

Fiff. 1. Fmminoclema ramosum, ]\Iar,sliall ; uatural «ize.
Fig. '2. Cacospongia vesiculifera, u. s}).; x 2
Fig. 3. Caeosponyia mvrrca/i, n. sp.; natural size.
Fig. 4. Cartenospongia otahitica, Esper ; natural size.
Fig. 5. Carteriospongia radiata, Hyatt ; uatural size.

TheVoyage of H.M. S.,. Challenger,"

Keratosa. Pi.IV.

,.-ji 'Wi/Vi.'

mm,, m

Pirsck v\.

lliriiT.SlTV.V::

PUTE V.

(ZOOL. UHALL- EXP. — PART XXXI. — 1884.) — HL.

PLATE V.

Figs. 1-3. Cacospongia levis, n. sp.
Fig. 1 . Portion of the skeleton ; natural size.

Fig. 2. A longitudinal section through it, . . . . . x 4

Fig. 3. Portion of a section through parenchyma, . . . . x 300

Figs. 4-6. Coscinoderma altimi, n. sp.
Fig. 4. Portion of the skeleton ; natural size.
Fig. 5. A transverse section of it, . . . . . . x 4

Fig. 6. Portion of a section through parenchyma, . . . . x 130

Figs. 7-9. Carteriospongia radiata, Hyatt.

Fig. 7. A peripheral segment of the skeleton, . . . . x 4

Fig. 8. Portion of a section through parenchyma, . . . . x 300

Fig. 9. A sperm-ball and some mesodermic cellular elements, . . x GOO

The Voyage of H.M.S.„Challengcr."

Keralosa.Pl.V.

)i.

/^^' /:<P^

/^

^. i^:^

K

"::%^

% ,

Krohse et aut del

PLATE YI.

PLATE VI.

Fig. 1. Portion of the skeleton oi Etispongia officinalis, var. lohosa, no v.; natural size.

Fig. 2. The same oi Hippospongia anomala, n. sp.; natural size.

Fig. 3. The same of Hippospongia mauritiana, Hyatt (a, from above ; b, from the
plane of section) ; natural size.

Fig. 4. The same of Cosciiioderma denticidatum, n. sp.; x 5.

Fig. 5. The same of Cacospongia amorpha, n. sp.; natural size.

Fig. 6. The same of Cacospongia procumbens, n. sp.; x 2.

Fig. 7. The same of Cacospongia intermedia, u. sp.; natural size.

Fig. 8. The same of Cacospongia nwrraiji, n. sp.; natural size.

Fig. 9. The same of Cacospongia vesiculi/era, n. sp.; natural size.

Fig. 10. The same of Cacospongia irregidaris, n. sp.; natural size.

Fig. 11. The same of Cacospongia compacta, n. sp.; natural size.

Fig. 12. The same of Cacospongia spinifera, n. sp.; natural size.

Fig. 13. The same of Cacospongia oligoceras, n. sp. ; x 2.

Fig. 14. The same of Cacospongia dendroides, n. s]").; natural size.

Fig. 15. T\iQ &&va.Q oi Stelospongoslongispinus,T)\\(:h..&n({Mich..-, natural size.

The Voyage of H,M.S.„Cliallenger:

Keralosa Pl.Al.

/-*.

12

8.

//

Kiohse del

Iirh AnsIvWu-nolc'WimKFrarJifiiH'iM

PLATE YII.

(ZOOL. CHALL. EXP. — PART XXXI. — 1884.) — Hll.

PLATE VII.
Hippospongia anomala, n. sp.; natural size.

The Voyage ol H. M. S. .Challenger.

Kerciiosa. PM.

Pirsc>. M

^■A Ir.st T^Wfrnnr 4Wmla.?ralikh;t W

PLATE VIII.

PLATE Vlll.

Fig. 1. A specimcu of Cacuspongia dcndroidea, var. dura ; utitural size.

Fig. 2. Portion uf a .section through the peripheral part of the body of Caco-
spongia dendroides, var. friahllitf, showing numerous algoid
roTnid l)odies and dumb-l)cll shaped corpuscles, . . . x 700

Fig. ."). A duml)-l)ell shaped corpuscle; highly magnified.

Fig. 4. A filament of Stehspongos longispinns, . . . . x 400

Fig. 5. A filament of Cacosioongia irregidarin, . . . . x 400

The Voyage of H.M.S.„Challenger."

Keratosa.H.VDI.

Pusch ct aui Cfl

, ,'r.st vWtintr V(tf.r.:u Fr.r.'n^l^.

PLATE IX.

(ZOOL. CHALL. EXP. — PART XXSI. 1884.) — Hh.

PLATE IX.

Liiffaria vanahilis, n. sp.

Fig. 1. A colony of three specimons of elongated shape ; natural size.

Fig. 2. Portion of the skeleton of one of them ; somewhat magnified.

Fig. 3. A specimen of massive shape ; natural size.

Fio-. 4. A horizontal section through the skeleton of one of its outorowths ; natural size.

Fio". 5. Portion of this latter section ; x 14.

Fio-. G. Some of the finest skeletal fibres ; x 100.

The Voyage of H.M. S.„Challen^

cr.

Keratosa.Pl.K.

.5.

^^M

,0

r^

.^^.^--^^'^:

j'^H

i-^Wl

4

.^"'

X^ftfcistvVt:.

PLATE X.

PLATE X.

Figs. 1-3. Verongia hivsuta (?), Hyatt.

Fio-. 1. An entire specimen (the central cavity is opened by means of a longi-
tudinal incision) ; natural size.

Fig. 2. Portion of its skeleton ; natural size.

Fig. 3. Portion of a skeletal fibre, . . . . . . x 280

Figs. 4-7. Verongia teniiissima (?), Hyatt.

Fig. 4. An entire specimen ; natural size.

Fig. 5. Portion of its skeleton ; natural size.

Fig. 6. Pavement-epithelium of the outer surface, . . . . x 240

Fig. 7. Portion of a section through parenchyma, showing the arrangement
of the canal system as well as the histological jiroperties of
the species, — S, spermospores ; R, " Eeservenahrungs-Material "
formations of F. E. Schulze, . . . . . x GOO

'hoVoyagc of H.M. S.Xhallcn^er

Kcralosa. PIX.

,i*

6.

*:■

->iit ad.

THE

VOYAGE OF H.M.S. CHALLENOEE.

ZOOLOGY.

REPORT upon the Crinoidea collected during the Voyage of H.M.S.
Challenger during the Years 1873-76. By P. Herbert Carpenter,

D.Sc., Assistant Master at Eton College.

PART I.-GENERAL MORPHOLOGY, WITH DESCRIPTIONS
OF THE STALKED CRINOIDS.

PREFACE.

The circumstances under which I have come to be charged with the duty of reporting
upon the entire collection of Crinoidea made during the Challenger Expedition are as
follows. — The researches of my father, Dr. Carpenter, C.B., F.R.S., early led me to take a
special interest in Comatida and its allies. Some of his statements respecting the anatom}'
of the arms having been called in question, I was led to reinvestigate the matter towards
the end of the year 1875, by methods which were almost unknown during the progress
of his researches nearly fifteen years before ; and I had the pleasure of verifying all those
points in his descriptions of the arms of the European Comatulse which other observers
had disputed. I was then working in the zoological laboratory of Prof C. Semper
at Wiirzburg, who most kindly placed at my disposal the arms of some tropical Comatulge
which he had obtained in the Philippines. It soon appeared that the minute structure of
Crinoids oflferod a promising field for investigation, and Prof Semper therefore generously
put into my hands several specimens of a remarkable tropical Actinometra as material
for a monograph of the type. This work occupied me during the whole of the year 1876

(ZOOL. CHALL. ESP. — PART XXXII. 1884.) a

ii THE VOYAGE OF H.M.S. CHALLENGER.

and the earlier months of the following year, the greater part of which time was spent
at Wurzburg, where I had the constant advantage of Prof Semper's criticism and
advice. I also received much valuable help from my father, who freely placed at my
disposal all the material which he had accumulated some years before for his investigation
of the structure of Antedon, Pentacrimis, and Rhizocrinus. A portion of his observations
were communicated to the Eoyal Society in his now classical memoir on the skeleton of
Antedon rosacea, and in a later paper on the anatomy of the disk and arms. But he has
still a large amount of unpublished material ; and of this I have always been permitted to
avail myself as fully as I wished. How important this help has been to me will be
apparent to every subsequent worker at Crinoid mor^jhology, my own researches having
followed very closely on the lines which he had laid down. The results of my study of
Actinometra and various other Crinoids were communicated to the Ijinnean Society in the
summer of 1877, and shortly afterwards Sir Wy\'ille Thomson offered to entrust me with
the preparation of the Report upon the Comatulse dredged by the Challenger. The
collection was sent to me in January 1878 ; and for the next four years the time which I
could spare from my professional duties was devoted pretty continuously to the examina-
tion and description of some hundred and fifty new species. Eighteen plates had been
drawn and nearly all the specific diagnoses written out, when on Sir Wyville's untimely
death in March 1882 I was requested by Mr. John Murray to include the Stalked
Crinoids in my Report.

During the cruise of H.M.S. Challenger, and also for some years before it, Sir
Wyville had devoted much attention to the Stalked Crinoids, and he proposed himself
to investigate the collection of this group of animals which was made during the Expedi-
tion. He also arranged with Prof. Alexander Agassiz that he should embody the
descriptions of the Stalked Crinoids dredged in the Caribbean Sea by the U.S. Coast
Survey steamer " Blake " in his Report on the Challenger collection, so that it might
assume the form of a monograph of all the species known to science. He was able
to do but little with the " Blake" collection, however; and with the concurrence of Prof.
Agassiz it was sent to me by Mr. Murray along with the Challenger collection, proofs
of plates, drawings, preparations, and some notes, in the spring of 1882.

Sir Wyville had not made much progress with the preparation of his Rej)ort. Twenty-
eight plates illustrating the structure of Holopus and of the more remarkable types
dredged by the Challenger had been drawn and lithographed at Edinburgh under his
superintendence by Messrs. George AVest and W. S. Black, but he was never able to
draw up any specific diagnoses ; and he left no manuscript behind him of any kind,
except one or two generic and specific names which he had written upon the proofs of
some of these plates. Descriptions of Hyocrinus, Bathycrrmts, and of Pentacrimis
maclearcmus had, however, already been published in his popular work on the Voyage of
the Challenger — The Atlantic.

REPORT ON THE CTJNOIDEA. iu

Upon examining the two collections of the "Blake" and the Challenger, I found that
they contained some half dozen species of which no figures had been drawn at all ; so
that several additional plates would be necessary in order to illustrate them adequately;
while diagnoses were wanted of nearly thirty specific types, most of them being
undescribed, and some representing new genera.

The preliminary study of these various forms in their zoological aspect alone occupied
a good deal of time ; but having a strong conviction that they could only be properly
understood by the help of a detailed knowledge of their fossil representatives, I was led
to take up the study of the fossil Crinoids, more especially those of the Secondary and
Tertiary periods. This naturally brought up the question of the relation between the
Neocrinoids and the Palseocrinoids, which had already been occupying my thoughts for
some time past.

Besides these palseontological questions, many others presented themselves of a
morphological nature ; and I have endeavoured to obtain such a knowledge of the
morphology of recent Crinoids as would be of substantial aid in the interpretation of the
many puzzling forms among their fossil representatives, and also bring out their relation
to the other Echinodermata. Many others have been at work upon the same lines, with
varying opportunities and difterent degrees of success. Very much, however, still
remains to be done, more especially with regard to the finer details of microscopic
anatomy, which can only be thoroughly investigated in individuals specially prepared for
the purpose when quite fresh. In this way alone is it possible to obtain series of sections
of the requisite thinness and perfection of histological detail ; and I would therefore ask
those who may be disposed to criticise my figures to rememl^er that the preparations
were mostly made before the introduction of the Jung or CaldweE microtomes, from
material which had been in spirit for periods varying from two to fifteen years.

Several morphological and one or two systematic works upon the Crinoids have
appeared during the passage of this Eeport through the press ; and I have therefore added
an Appendix which contains a series of notes discussing the various questions to which
those works relate. Reference is given to these notes in the text, so that this first part
of the Crinoid Report may be regarded as presenting an epitome of our knowledge of
Crinoid structure up to the end of the present year.

The second part of the Report, containing descriptions of the Comatulae collected
by the Challenger and " Porcupine," will be published as soon as ever the necessary
plates can be drawn, and the manuscript revised, parts of the latter being now more than
five years old.

Of the sixty-nine plates accompanying Part I., twenty-eight were drawn at Edin-
burgh under the direction of Sir Wyville Thomson. Five others illustrating the
structure of Pentacrinus wyville-thomsoni were prepared by ]\Ir. George West to
accompany a memoir upon this type, commenced some years ago by my father, who has

iv THE VOYAGE OF H.M.S. CHALLENGER.

kindly permitted me to make use of them here. They are distinguished by the initials
" W. B. C"; while thirty-one of the remaining thirty-six which are marked " P. H. C,"
have been drawn under my own superintendence by Messrs. C. Berjeau, F.L.S., and
P. Highley. I am very greatly indebted to both these gentlemen for the care and skill
with which they have performed a difficult and laborious task, one, moreover, involving
a considerable acquaintance with minute structural details. I am under similar obliga-
tions to Mr. George West, who lithographed three anatomical plates, and also gave me
valuable information respecting the specimens from which he had drawn several plates
for Sir Wyville Thomson. I have likewise to thank Mr. Black for similar information
concerning those plates which had been drawn by himself; and also Mr. W. E. Hoyle, M.A.,
of the Challenger Office, for kindly working out the magnification of some of the figures
ujaon Mr. Black's plates.

I have further gratefully to acknowledge much valuable bibliographical assistance from
my friends Prof F. J. Bell, F.Z.S., and Mr. W. P. Sladen, F.L.S.; while my thanks
are also due to Prof. H. N. Moseley, Dr. P. P. C. Hoek, Mr. H. B. and Dr. G. S. Brady,
and to Mr. Edgar Smith, F.Z.S., for kindly identifying various organisms which I had
found attached to difi'erent parts of the Crinoids.

Prof. Carl Zittel of Munich and Prof. G. Meneghiui of Pisa have been good enough to
send me several fossil specimens oi Rhizocrinns for examination, and I take this opportunity
of thanking them. But I find it difficult to express my indebtedness to Mr. Charles
Wachsmuth of Burlington, Iowa, U.S., who is so well known as the leading authority
upon the Palseocrinoidea. Not only has he frequently sent me for examination very rare
and valuable specimens illustrating the morphology of the Pateocrinoids and Blastoids,
but he has repeatedly answered my inc[uiries in the fullest and most complete manner
possible. The relations of the Neocrinoids and the Palseocrinoids have been the subject
of a prolonged discussion between us, extending over more than two years ; and it is a
matter of sincere regret to me that we have been obliged to agree to differ. Time will
show how far our respective views will need modification. We have approached the
subject from different sides ; but upon one point we are in comjilete accordance, viz., the
desire to find out the truth, whether or not it agree with our own ideas upon the
subject. Mr. Wachsmuth will shortly ^lublish an explanation of his own position, to
which I would refer those who are interested in the matter.

In conclusion, I must express my sincere thanks to Mr. John Murray and to
Prof. Alexander Agassiz for having entrusted me with the working out of the Crinoid
collections in their charge, and for their readiness to afford me every possible assistance
in doing so ; while I would pay a heavy debt of gratitude to the memory of the late Sir
Wyville Thomson. It was my privilege to accompany him and my father in the short
but eventful cruise of H.M.S. " Lightning " in the year 1868 and in the "Porcupine"
expedition of 1869; and to his kindness and encouragement both then and sub-

REPORT ON THE CRINOIDEA. V

sequently is due much of such success in zoological work as I may subsequently have
attained.

To my father, as will be readily understood, I owe far more than can well be put into
words ; and I will only express the hope that the following pages may be regarded as a
not unworthy sequel to his many contributions to Invertebrate Zoology.

Eton College, Decemher 1884.

TABLE OF CONTENTS.

MORPHOLOGY.

I. — The Skeleton generally, with the Modeh of Union op its Component Joints.
Syzygies, .
Articulations,

II. — The Stem and its Appendages,

A. Pentacrinidce, .

Tenaination of the Stem below,

B. BourgueticrinidaB,

Morphology of the Stem,
C Hyocrinirise,

111. — The Calyx,

A. The Basals,

B. The Eadials, .

C. The Interradials,

Anal Series,

IV. — The Eays,

A. The Ray-Divisions and Arras,

B. The Pinnules, .

Pinnules of Palseocrinoids,

v. — The Visceral Mass,

A. The Oral Plates,

B. The Perisomatic Skeleton,
C The Visceral Skeleton,

VI. — The Minute Anatomy of the Disk and Arms,

A. The Geography of the Disk,

B. The Digestive Tube, .

C. The Water- Vascular System,

D. The Blood- Vascular System,

Labial Plexus,
Plexiform Gland,
Chambered Organ,

E. The Genital Glands,

F. The Nervous System,

Ambulacral Nerve,
Axial Cords,
Parambulacral Network,
Circular Commissure,

G. The Sacculi, and the Colouring Matters,

PAGE
1

3
5

12
12
18
23
28
31

33
24
36
38
41

47
47
58
61

67
70
73

85

88

88

88

92

96

97

100

106

108

111

112

114

123

125

127

vm

THE VOYAGE OF H.M.S. CHALLENGEE.

VII. — On tub Habits op recent Crinoids, and their Parasites,

Modes of Attachment, ......

Myzostoma, .......

VIII. — The Geographical and Bathymetrical Distribution ov the Crinoids,

A. Geographical Eange, ......

B. Bathymetrical Eauge, ......

C. The Association of the Genera at Particular Stations, .

IX. — On the Eelation between the Eecbnt and the Fossil Neocrinoids,

X. — On the Eelations of the Neocrinoids to the Pal/Eochinoids,
The Mullerian Classification ; Articiilata and Tessellata
General Characters of the Neocrinoids,

Structure of the Calyx,

Characters of the Arms,

The Disk and Oral Plates,

The Vault of the PaliEocrinoids,
Vault of Haplocrinm, AUaQeainus, and Cocrocrinn.
Vault of Cyathocrinns,

Subtegminal Ambulacra,
Vault of Actinocrinida3,

Homologies of the Calyx and Dome Plates,

Eadial Dome Plates, .
Vault of Platyorinidse,
Vault of Ichthyocrinidse, .
Vault of Glyptocrintis, Reteocrinas, and Xenocrinus.

XI. — Classification, ....

Definition of the Pelmatozoa,

Definition of the Crinoidea,

Other Systems of Classification,
Characters of the Pelmatozoa,

Definition of the Neocriuoidea,

XII. — Description of the Specimens,

Family Holopidas and Genus Holopus,

A. General Account of the Type,

Holopus rangi, d'Orbigny,

The Calyx,

The Axillary Eadials, .

The Arms,

The Disk and Ambulacra,

Anatomy of the Arms, .

B. On the Systematic Position of Holopus,

Cyathidium, .

Cofylecrinus, .

Eiidesicrinus, .

Gymnocrinus and Micropocriniis,

Edriocrinus,

p.\oe
130
131
134

136
136
138
139

142

145
145
149
149
154
155
156
157
163
164
167
168
173
175
181
183

186
186
187
189
193
196

197
197
197
199
199
204
206
208
209
211
211
213
214
216
217

REPORT ON THE CRINOIDEA.

IX

Family Hyocriiiidic, and Defiuition of Hyocrinus,

A. General Account of the Type, Hyocrinus lethellianm, Wyville Thomson,

B. On the Systematic Position of Hyocrinus, .
Family Bourgueticrinidse, ....

Definition of genus Bathycrinus, .

Structure of the Calyx,

Distribution of the Axial Cords,

Trifascial Articulations of the Outer Radials and Arm-Joints,

The Disk, ....

Axial Cords of tlie Arms,

Comparison of Bathycrinus and Rhizocrinus,

Bathycrinus campbellianus, n. sp.,

aldrichiaiius, Wyville Thomson,
gracilis, WyvOle Thomson,
Definition of Genus RMzucrinus,

Characters of Conocrinus,

Composition of the Calyx in Rhizocrlnus,

Views of Pourtales, Sars, and Ludwig,

Distribution of the Axial Cords,

Syzygies in the Arms,

Affinities of Rhixocrinus,

Rhizocrinus lofotensis, ]\I. Sars,
rawsoni, Pourtales,

Variations in the Shape of the Calyx, .

The genus Deniocrinus, Perrier,
Definition of the family Pentacrinidse,

Synopsis of the Genera,

Genus Pentacrimis,

A. Characters of the Geuus,

Characters of Extracrinus,
Cenocrinus and Neocrimis,
Cainocrinus, ....
Variations in the Size of the Basals in Pentacr
Arms of Pentacrimis, .
Characters of the Stem,

B. On the Characters of Young Pentacrinidre,

C. The Calyx and its Contents,

D. The Geological History of Pentacrimis,
Synopsis of the recent Species, .

Pentacrinus asterius, Linn. sp.

miilleri. Oersted, .

maclearanns, Wyville Thomson,

wyville-thomsoni, Jeffreys, .
Termination of the Stem below, and the Mode of
Pentacrinus alternicirrus, n. sp.,

naresianus, n. sp.,

Makei, P. H. Carpenter, .

decorus, Wyville Thomson,
Variability of its Character.*, .
Pentacrinus mollis, n. sp.,

(ZOOL. CHALL. EXP. PART XXXII. 1884.)

Attachment of the Pentacrinidie,

PAGE
217

218

222

225

225

226

228

231

234

236

237

238

241

243

245

247

248

250

252

254

255

259

262

264

268

270

271

272

273

274

279

281

282

284

287

289

291

294

299

300

306

312

.•?13

315

321

324

328

330

335

338

THE VOYAGE OF H.M.S. CHALLENGEE.

XIII.

Definition of genus Metummis,

Synopsis of the Species,
Metacrinus amjulatus, n. sp.,

cingulatus, n. sp.,

murrai/i, n. .s]i.,

nobilig, n. sp.,

varians, n. sp.,

moseleyi, n. sp.,

loyvillii, n. sp.,

contains, n. sp.,

nodosus, n. sp.,

interruptus, n. sp.

iuherosHS, n. sp.,
Family Comatnliilw, and Definition of genus Thaumatocrinux,
Tluiumatocrinus renovatus, P. H. Carpenter,

-Bathymetrical Distribution and Station List.s,

Station List of the Stalked Criuoids which have been obtained by the various liritis

Expeditions for Deep-Sea Exploration between the- Years 1868 and 1880,
Station List of the Stalked Crinoids which have been obtained by the various American

Expeditions for Deep-Sea Exploration (mostly under the direction of Mr. Alexander

Agassiz) between the Years 1867 and 1880,

A List of the known Living Species of St

Geographical Distribution,
Analysis of this List,
Bathymetrical Tables,
Summary,
Crinoids of the Abyssal Zone,

alked Crinoids, showing their Bathymetrical and

page
339
344
345
347
349
351
353
355
358
360
364
367
369
370
372

374
374

379

38.5
388
388
390
391

APPENDIX.

Note A. — On the Homologies of the Criuoidal Calyx in the other Echinoderms,

Apical System of Urchins and Starfishes,

Origin of the Odontophores of Starfishe.?,

The Mutual Relations of Crinoids, Asterids, and Echinids,
Note B. — On the Basals of Fossil Comatulse, .....
Note C. — On the Excentric Position of the INIouth in Actinometru,
Note D. — On the Supposed Communication of the Chambered Organ and Labial Plexus w
Note E. — On the Intervisceral Blood- Vessels, .....
Note F. — On the Eelation of the Vascular Systems of a Crinoid to those of the other
Note G. — The Nervous System of the Crinoidea, ....

The Nervous Nature of the Axial Cords,

Their Morphology and Development,

Nervous System of the Blastoids,

The third Nervous System of Jickeli,

Bibliography of the Neocrinoidea,
Index to Authors quoted,
General Index, ....

Explanation of the Plates, .

PAGE

393

393

398

400

402

403

vith the Exterior

404

.

405

Echinoderms,

405

407

407

410

413

415

.

417

429

431

441

INDEX TO WOODCUTS.

PAGE

Fig. 1. Calyx oi Promachocriwis hvrgueleuiiitt, . . .... 37

Fig. 2. Diagram showing the course of the Digestive Tube iu an endocjclic Crinoid (Aufcduii, Penta-

crinus, &c.), as seen from the ventral side, ...... 89

Fig. 3. Diagram showing the course of the Digestive Tube in an Adinometra with Interradial Mouth,

as seen from the ventral side, ........ 92

Diagrammatic transverse section of an iingrooved ovarian pinnule of Adinometra jiarvicirra, . 113

Diagrammatic transverse section through the end of a grooved pinnule of Adinometra

paroinrra, . . . . . . . . . . .121

Longitudinal section of a pinnule joint of ^rf//(omrf)v( /(«>//•«, ..... 121

Longitudinal section of tlie ventral perisome in a pinnule of Adinometra nigra, , .122

Diagrammatic transverse section of an ambulacrum on the disk of Anfedo7i esdiridifi, . . 123

Interradial plates of ^|)W(;)TKM»' rowsy«?H(s, . . . , . . .183

View of the upper part of the calyx-tube of Hohpus rnngi, on its lower or bivial side, . . 203

Diagram of a horizontal section through the lowest portion of the basal ring of Bathycrinus

aldridiiaims, ......>... 226

Fig. 12. Diagram of a horizontal section through the calyx of Bathycrimis aldridiianns at the level

of the upper part of the Lasiradial suture, ....... 228

Fig. 13. Diagrammatic vertical section through tlie calyx of Bathycrinus aldridiiamis, . . . 229

Fig. 14. Plan of the distribution of the axial cords in the calyx of Bathycrinus cddrichianus, . . 229

Fig. 1-5. Bathycrimis camphelUanns, n. sp., ........ 239

Fig. 16. Bathycrimis gracilis, y^j\\\\&T\\oms.ow, ....... 244

Fig. 17. Diagrammatic vertical section through the calyx and disk of iiA/.W(;?Vm« Zo/ofen«6-, . . 251

Fig. 18. Diagram of the distribution of the axial cords in the calyx of Rhizoerinus lofotensis, . . 253

Fig. 19. The calyx and arm-bases of two specimens of Rhizoerinus raiosoni from Panama, . . 267

Fig. 20. Diagram showing the arrangement of the axial cords in the calyx of a Pmtremites, supposing

it to be the same as in the calyx of a Crinoid, . . . . . .413

Fig. 21. Tiari'chinus princeps, Laube, inverted so as to show the resemblance of its apical system to the

calyx of a Crinoid, . . • • • • • • ^'^''^

Fig-

4.

Fig.

5.

Fig.

6.

Fig.

7.

Fig.

8.

Fig.

9.

Fig

10.

Fig.

11.

INDEX TO THE TABULAR STATEMETNS IN CHAPTERS I.-X.

PAGE

Table I. — Water-pores of Hijocriiius, ......... 95

Table II. — Showing tlie Geographical aud Bathymetrical Kange of all the Genera of Eecent CrinoiJs, 137

Table III. — Showing the Frequency of Occurrence of all tlie Genera of Recent Crinoiils at deptlis

below 250 fathoms, . . . . . . . . . .138

Table IV. — Showing the number of times that Isolated Species of Stalked Crinoids have been dredged, 139

Table V. — Showing the Association of Crinoid Genera at different Stations occupied by the " Porcupine,"

Challenger, and "Talisman," together with the number of Species obtained at each Station, . 140

Table VI. — Showing the Association of Crinoid Genera in the Caribbean Sea, together with the number

of times that two or more Species of Peniacrinus were found at the same Station, . . 141

Table VII. — Showing the Mutual Homologies of the Principal Plates in the Actinal and Abactiual

Systems of Echinodermata, . . . . . . , • • .169

MORPHOLOGY.

L— THE SKELETON GENERALLY, WITH THE MODES OF UNION OF

ITS COMPONENT JOINTS.

The organisation of a Crinoicl is broadly divisible into two well-marked portions, to
which the general names ambulacral and antiambulacral may be given. They correspond,
on the whole, to the left and right larval antimers respectively, though probably not
exactly so. The first is the visceral mass or " disk," in which is situated the whole of
the digestive tube, with both its terminal openings. It likewise contains the central
ends of the radial water-vessels and blood-vessels, wliich converge towards their
respective circum-oral rings, and also the corresponding portions of the ambulacral
nervous system (PI. LXIL).

Both the disk and its extensions in the perisome clothing the ventral surface of the
arms and pinnules are usually more or less covered by calcareous plates, the arrangement
of which will be described subsequently (PI. VI. fig. 4; PL XVII. figs. 6-10; PI. XXYI.
figs. 1, 2; PL XXXIIL figs. 6, 7; PL XXXIX. fig. 2; PL XLL figs. 4, 12-14; PL XLHL
fig. 3 ; PL XLVII. figs. 10-13 ; PL L. fig. 2 ; Pis. LIV., LY.). They represent a portion
of that element of the Crinoid skeleton to which the name " perisomatic " was given by
Sir Wyville Thomson^; for they are all originally developed from simple cribriform films
of limestone, such as appear in all young Echinoderms, and thicken by continual repetition
of the same formation.

The antiambulacral portion of a Crinoid consists of the stem and its appendages,
the calyx, and the skeleton of the rays, arms, and pinnules. This constitutes the radial
skeleton, as the term is understood by Dr. Carpenter,^ A^iz., that which is perforated b)^ a
central canal lodging an extension of the fibrillar envelope around the chambered organ
(PL Yllb. fig. 2; PL XXIY. figs. 2-6, ca; figs. 7-9, ar; PL LYIIL figs. 1-3, ar ;
PL LXIL).

Sir Wyville Thomson was unaware that the primary interradial cords proceeding from
the chambered organ perforate the basal plates (PL Yllb. fig. 2 ; PL XXIY. fig. 7 ;

1 On the Embryogeny of Antedon rosaceus, Linck, Phil. Trans., 1865, pp. 540, 541.

2 Researches on the Structure, Physiology, and DevL-lopnieut of Antedon rosaceus, part i., Phil. Trans., 1866,
p. V42.

(ZOOL. CH.VLL, EXP. — PART XXSII. — 188-1.) K 1

2 THE VOYAGE OF H.M.S. CHALLENGER.

1*1. LVIII. iii;s. 1, 3 — cd; PI. LXII.) ; and he therefore considered them as belonging to
the perisomatic system, on the ground of their not being thickened by the peculiar fascicu-
lated tissue of parallel rods such as is found in the growing stem- and arm-joints. He
described this fasciculated tissue as appearing in the first radials. But according to
Dr. Carpenter " there is no distinction in texture between, the endogenous additions by
which the first radials and the basals are respectively thickened ; so that we cannot place
them in separate categories on this score. But further, we have seen that in the stage
now described, the basals as well as the radials are perforated to give passage to the
radiating extensions of the sarcodic axis of the stem, which only reach the radials through
the basals ; so that this ground of distinction also fails to separate them."

The basals and radials form part of the apical or abactinal system of the Crinoid, and
are represented by the genital and ocular plates of the Urchins, and their homologues in
the other Echinoderms.^ The oral plates of the Crinoid correspond in like manner with
the mouth-shields of Ophiurids, the " odontophores " of Asterids, the oral plates of the
Psolidse, and possibly also with the five actinal plates of Palceostoma mirabilis among the
Urchins.^ As regards the rest of the Crinoid skeleton, however, nothing more than a
very general homology can be established with the skeleton of the other Echinoderms.

The perisomatic skeleton consists essentially of numerous minute plates which are
usually more or less isolated, but sometimes slightly connected by fibrils of connective
tissue (PL VI. fig. 4 ; PI. XVII. figs. 6-10 ; PL XXVI. figs. 1, 2 ; PL XXXIII. figs. 6, 7 ;
PL XLL figs. 4,^2-14; PL XLIII. fig. 3; PL XLVIL figs. 10-13; PL L. figs. 1, 2;
Pis. LI v., LV. ; PL LVII. fig. 3, an; PL LXIL). The radial skeleton, however, consists
of successive joints and rods which are developed in a longitudinal direction, and are
united to one another by articulation or suture. In either case the ultimate imion of the
two joints is efl!ected by means of connective tissue fibres, which pass from the nucleated
and pigmented organic basis of the one joint into that of the other (PL Vllb.
figs. 1, 8, li, Id. PL XXIV. figs. 6, 7 ; PL LVIII. figs. 1-3—^, Ih, L). These fibres are
sometimes quite short, and their ends are surrounded by the denser layers of calcareous
reticulation on the apposed surfaces of the two joints, which are thus closely and
immovably fitted together, though they can be separated by the action of alkalies.
This mode of union is called a " suture," or better, a " synostosis." ^

The first i-adials of the Comatulse are connected in this manner both with one another
and with the centro-dorsal. The same mode of union also occurs between the radials of
the Pentacrinidse and the basals on which they rest, as well as between the five individual

' On the Oral and Apical Systems of the Echinoderms, part i., Qxiart. Joum. Micr. Sci., vol. xviii., N. S., pp.
367-382. Some disputed points in Echinoderm Morphology, Ibid., vol. xx. pp. 322-329. On the Apical System of the
Ophiurids, Ihid., vol. xxiv. pp. 1-22. Vide, Note A.

2 Oral and Apical Systems, part ii., Ihid., vol. xLx., N. S., pp. 191-193.

8 See Sirarotli, Zdfschr. f. wiss. Zool., Bd. xxvii. jx 435 ; and also P. H. Carpenter, On the genus Actinometra,
with a morphological account of a new species from tlie Pliilippine Islands, Tmns. Linn. Soc. Land. (Zool.), ser. 2,
vol. ii. pp. 55, 66, pi. iii. fig. 4.

EEPOET ON THE CRINOIDEA. 3

plates whicli make up each of these rings (PI. XII. figs. 1, 2, 11-16, 22-25 ; PL XVIII.
figs. 4-7 ; PI. XX. figs. 1-3, 6-9 ; PI. XXI. figs. 6, 7 ; PL XXIV. fig. 6, lb ; fig. 7, 1, lb, L ;
PL LVIII. fig. 1, /, lb; fig. 2, L). The radials of Bathycrinus are likewise united in this
way (PL Vllb. fig. 4, 1) ; but the sutures between the basals seem to be of a still closer
nature. They are visilile externally in young individuals but disappear in the adult, so
that the "base " seems externally to consist of but one single piece^ (PL Vila. figs. 12-14).
Sections through a decalcified specimen show, however, that it is really composed of five
parts like the base of Pentacrimis (PL Vllb. fig. 2, B). These five parts are, nevertheless,
very closely anchylosed. No parallel fibres of connective tissue pass between them, such
as unite the five first radials together (PL Vllb. fig. 4, I). Pnit the organic basis of the
skeleton is much less close, if not absent altogether, along five lines which radiate out-
wards from the chambered organ and indicate the position of the sutures. They stain
less deeply with hsematoxylin than the surrounding tissue, but do not reach the exterior
of the section (PL Vllb. fig. 2). Neither are they visible in sections through the top
and bottom of the basal piece ; and but for the knowledge obtained in this way, the
basal piece would probably have been described as an uppermost stem-joint, as has
actually happened in the case of Rhizocrinus.

In some individuals of Rhizocrinus lofotensis (PL IX. figs. 1, 2) there are no
indications of suture, even in the adult ; though in others the sutures are visible
externally (PL X. fig. 2). In the former case the nuclear connective tissue network is
continuous through the whole section, which exhibits no unstained radiating lines, as do
similar sections of Bathycrinus. There is a marked diflerence between the two genera,
however, as regards the lateral union of the radials. Those of Bathycrimis are united by
synosteal fibres (PL Vllb. fig. 4, /), just as in Pentacrinus (PL XXIV. figs. 7-9 ;
PL LVIII. fig. 2, L) and Comatiila. But those of Rhizocrinus are much more closely
connected, just in fact in the same way as the basals of Bathycrinus are. No ligamentous
fibres are visible in horizontal section, but only five (or six) radiating lines where the'
nuclear network is incomplete (PL Villa, fig. 6). The radials in this type, and the
basals in Bathycrimis, seem to be united by a limestone deposit which has difl'erent
characters from that usually found in Echiuoderms ; and I think it will be advantageous
to denote this by the term " anchylosis," reserving " synostosis " for cases in which
ligamentous fibres are concerned in tlie union of the joints ; though both of them, like
the " syzygy," would be considered as sutures.

Smooth sutural unions like those between the basals and radials, though somewhat
less close in their character, occur at intervals in the stem of the Pentacriuidfe, every
nodal or cirrus-bearing joint being united in this way to the joint below it (PL XIX.
figs. 3, 4; PL XXIL figs. 16, 20, 21; PL XXVL figs. 12-16; PL XXXL fig. 3;

> This must be cirefully distiuguished from the "article basal" of the ApiocrinUiu, which is the uppermost
stem-joint that supports the basals.

4 THE VOYAGE OF H.M.S. CHALLENGER.

PI. XXXII. figs. 1, 2 ; PI. XXXVII. figs. 5-8, 19, 21 ; PL XXXIX. figs. 3, 5, 6 ; PI. XLI.
figs. 5, 15 ; PI. XLV. figs. 3, 5, 6 ; PI. XLVII. figs. 1-3 ; PI. L. figs. 4, 20, 21 ; PI. LI.
figs. 5-8).

The external line of separation frequently disappears altogether, or is only traceable
with great difiiculty, and the two joints, primitively separate, become practically fused
into one (PI. XLII. fig. 1 ; PI. L. fig. 3). I cannot find any mention of this peculiarity
in the classical memoir of Johannes MllUer,' who spoke of the successive unions of
the stem-joints indifferently as " Nathe oder Gelenke." So far as I can make out, it
was first noticed by Sir Wyville Thomson in Pentacrinus decorus ;^ and Liitken'
subsequently described it in more detail in Pentacrinus asteria and Pentacrinus mUlleri.
Quenstedf* also noticed it in the fossil species Pentacrinus scalaris and Pentacrinus
jurensis.

An essentially similar mode of union between certain of the arm-joints was spoken
of by Miiller as a " syzygy," and described as an immovable sutural union. The name
syzygy has since been applied to the sutural union of the nodal stem-joints with those
next below them ; and Miiller's terms " hypozygal " and " epizygal " for the two arm-
joints which are united by syzygy (PL XII. figs. 7, 10, 18, 21 ; PL XXXa. figs. 9, 10, 12 ;
PL XXXII. figs. 4, 9, 15, 18) may be conveniently applied in the case of the stem-joints
also.

In all the recent ComatulsB the apposed faces of the two portions of a syzygial joint
are marked by a series of slightly elevated ridges with alternating furrows, which radiate
from the opening of the central canal towards the dorsal margin of the joint. In
Actinometra typica these ridges are frequently not perfectly continuous ; but they are
broken up into a row of little elevations, squarish or oblong in shape, and arranged with
their longer axes radiating outwards from the central canal. On some joints these are
not very numerous, and as their terminal faces are marked by median vertical lines, they
have been wrongly described as surfaces effecting a ligamentous articulation of the
bifascial type, such as will be described immediately.'

The radiating arrangement is usually much less marked in the Pentacrinidse than in
the Comatulse, the striation being frequently only visible at the extreme marginal
portions of the syzygial surfaces (PL XII. figs. 7, 10, 18, 21 ; PL XXI. figs. Id, 2d,
5a; PL XXX. figs. 20, 21) as figured by Miiller in Pentacrinus asteria,^ while in
some cases it appears to be absent altogether, the apposed faces being perfectly smooth

1 Ueber den Bau des Pentacrinus caput-Mediiste, Ahlmndl. d. k. Akad. d. IViss. Berlin, 1843.

2 Sea Lilies, The Intellectual Observer, No. 31, August 1864, p. 7.

' Om Vestindiens Pentacriner, med nogle Bemaerkninger om Pentacriner og Sdlilier i Almindelighed, Vidensk.
Meddel.f. d. nat. Foren. i Kj^henhavn, 1864, Nos. 13-16, pp. 198, 199.

^ Petrefactenkunde Deutschlands, Bd. iv., Asteriden und Encriuiden, pp. 196, 230, Taf. 98, figs. 2, 3, 107.

^ See Lov^n, Phanogenia, ett liittills okiindt slagte af Ma Crinoideer, Ofversigt k. Vetensk.-Akad. Forhaiidl.,
Arg. xxiii.. No. 9, p. 230, fig. c; and also P. H. Carpenter, The Comatulre of the Leyden Museum, Notes from the Leyden
Mtiseum, vol. iii. pp. 197-199. ^ Op. cit., Taf. ii. fig. 4.

REPORT ON THE CRINOIDEA. 5

(PI. XXV l. figs. 5, 8 ; PL XXXVII. figs. 3, 4 ; PI. L. figs. 6, 7, 12, 13) like the apposed
surfaces of the radials and basals respectively (PI. XX. figs. 2, 3, 6, 9). This is also the
case in Rhizocrinus, which presents another peculiarity as well. Near the dorsal edge of
the upper face of the hypozygal there is a more or less well-marked pit (PI. X. figs. 1, 6,
8, 18), and a corresponding peg-like process projects from the under face of the epizygal
(PI. X. figs. 17, 19), so that closeness of union is effected in this way instead of by the
usual radiating ridges.

Another arrangement which effects a somewhat closer syzygial union than usual,
presents itself in the arms of Pentacrinus naresianus (PL XXX. figs. 20, 21, 23 ;
PL XXXa. figs. 9, 10, 12) and oi Pentacrinus Uakei (PL XXXII. figs. 4, 5, 7, 9, 12, 14).
The apposed surfaces are not flat or slightly curved, but the proximal face of the epizygal
rises to a sharp crest which is interrupted by the central canal, and fits into a correspond-
ing re-entering angle on the distal face of the hy|Jozygal, so that the two joints interlock
very closely. This peculiarity is very apparent in a side view of the arm (PL XXX.
fig. 23) ; but when seen from the dorsal side, the distal edge of the hypozygal appears to
be very convex and to project strongly forward into the epizygal (PL XXX. fig. 1 ;
PI. XXXI. fig. 2).

Very different modes of articulation occur in the Crinoid skeleton, so that the amount
of play between two successive joints varies considerably. It is probably at a minimum
in the stem of Hyocrvrms (PL VI. fig. 2). We do not know the nature of the basal part of
the stem ; but the lower joints of the fragment, 170 mm. in length, which was obtained,
are cylindrical, with their terminal faces devoid of any fossaj, but marked by a radiating
pattern of grooves and ridges. This, however, is sometimes absent, as shown in
PL Vc. fig. 4. Sir Wyville Thomson spoke of the joints as united by a close syzygial
suture ; ^ but the ligamentous fibres which effect their union (PL Vc. fig. 5, Is) are
longer than I have ever seen them in any real syzygy, and rather resemble those which
unite the successive pinnule joints of other Crinoids w^hen muscles are absent. They are
all of the same length and not longer in the centre than at the periphery, as they are
between the deeply hollowed stem-joints of the Bourgueticrinidse (PL Vila. figs. 8-11 ;
PL X. figs. 11-14). But I should hardly describe this mode of union as a syzygy
(or suture) ; for there must have been some amount of play between the successive
joints, and a syzygy was described by Miiller as an immovable sutural union of two
joints.^

Next to Hijocrinus, the Pentacrinidai have the most closely united stem-joints. In
this family each internode of the stem contains five oval bands of elastic fibres (PL XXIV.
figs. 1, 3-5, Is). They run through all the joints between the hypozygal of one syzygy
and the epizygal of the next below it, which is the true nodal joint. Miiller^ and Wyville

1 Notice of new living Crinoids belonging to the Apiocrinidii', Journ. Linn. Soc. Land. (Zool.), vol. xiii. p. 52, 1S76,

2 Oji. cit., p. 39. ^ Op. cif., p. 17.

6 THE VOYAGE OF H.M.S. CHALLENGEE.

Thomsou^ described them as extending from end to end of the stem ; but this is not
strictly true. At the toj) of the stem where the young nodal joints are very close
together and of no great thickness (PI. XIII. fig. 1 ; PI. XV. fig. 1 ; PI. XXXI. fig. 2 ;
PL XXXVII. figs. 1, 2 ; PI. XXXIX. fig. 1 ; PL XLIII. fig. 2 ; PI XLIX. fig. 2), these
bands of ligament are, no doubt, directly continuous from their attachment to the basals
above, through more or fewer of the last-formed nodes and internodes. But this is
certainly not the case in the lower parts of the stem. For after it has been decalcified
the nodal and infra-nodal joints which form the syzygies are just as readily separable
from one another as they are after the animal matter has been x'emoved by the action of
hot alkalies. This would obviously be impossible if the fibrous bands passed right through
the joints " from one end of the stem to the other ;" while, as a matter of fact, their
terminations in the substance of the nodal or infra-nodal joints may be readily traced by
microscopic examination. They correspond to the five radiating petaloid figures which
are so well known on the surfaces of the stem-joints of the Pentacrinidse (PL XV. fig. 5 ;
PL XXVI. figs. 17, 18 ; PL XXX. figs. 28-30 ; PL XXXa. fig. 7 ; PL XXXIX. figs. 4-10;
PL XLL figs. 2, 3, 6, 7, 16, 17; PL XLVII. figs. 3-5, 7-9); and they are characterised
by a somewhat looser calcareous reticulation than exists in the remaining portions of the
joint (PL XXIII. fig. 3). The apposed faces are more or less cut up into ridges wdth
intervening furrows ; and the ridges on the lower face of the one joint correspond to the
furrows on the upper face of that l^elow it. (PL XV. figs. 1, 2; PL XIX. figs. 2-5 ;
PL XXVII. fig. 1 ; PL XXXVL ; PL XLL figs. 1,5; PL XLVIL figs. 1, 2, 6 ; PL XLIX.
fig. 3). Hence, while the composition of the stem out of a large number of discoidal
joints gives it a certain amount of motion, that motion is very limited ; and it is probably
only of a passive character, due to currents in the water, &c. , and independent of the will
of the animal. In this respect it differs from the rays and their subdivisions, the joints
of which are united by pairs of muscular bundles (PL Vc. fig. 2, m. ; PL Vllb. figs. 1, 5 ;
PL Villa, fig. 7, rm.; PL XXXIX. fig. 13; PL XLL fig. 11); and the contractions
of these bundles are governed by an influence proceeding outwards from the fibrillar
envelope round the chambered organ in the calyx (PL Vllb. figs. 1, 2 ; PL XXIV.
figs. 6-8 ; PL LVIII. figs. 1, 3 — ch) along the axial cords of the rays and arms
(PL Vc. fig. 2. PL Vllb. figs. 1, 5-8 ; PL Villa, figs. 5, 7, 8 ; PL XXIV. fig. 9—^.
PL LXIL).

Although there are no true articular surfaces on the stem-joints of the Pentacrinidae
in the sense in which that term is employed in anatomy, yet this is by no means the
case in the Bourgueticrinidje. In all the members of this family there are true
articulations between the successive stem-joints, of the same nature as those between the
cirrus-joints of all Crinoids, and between the two outer radials of most Comatulas and of
some species of Pentacriniis. But they are efiected only through the agency of

1 Sea Lilies, p. 3.

EErORT ON THE CRINOIDEA. 7

ligamentous bundles, muscles being confined exclusively to the rays and their sub-
divisions.

This ligamentous articulation of the successive cirrus-joints and of the stem-joints of
the Bourgueticrinidse is of a very simple nature. There is an articular surface around
the opening of the central canal with a more or less marked fossa on either side of it
(PL Vila. figs. 7-11). The fossae lodge the elastic iuter-artieular ligaments, which are of
precisely the same nature as the tendinous cords in the stem of Pentacrinus ; but they
differ from these cords in being confined to the intervals between the elongated joints
without traversing their substance (PL Vila. fig. 1, Is).

The two ligaments may be of unequal size, as in the middle and outer portions of the
cirri, whereby the prehensile power of these organs is much increased ; or they may be of
equal bulk, as in the lower portions of the cirri and between the stem-joints generall)'.
In the former, however, the long axes of the terminal faces of any joint lie in the same
plane ; but in the stem-joints of the Bourgueticrinidae this is by no means the case, for
the long axes of the terminal faces cross one another at various angles (PL Villa, figs. 2, 3 ;
PL IX. fig. 3). The outline of the interarticular ligaments of the middle and lower
parts of the stem, as seen in optical or in longitudinal section, has a diamond shape with
somewhat extended lateral angles. This is, of course, due to the fact that the fibres are
longest in the centre of the stem, immediately around the vascular axis, and shortest
towards the periphery where the foss33 are shallowest (PI. VII. figs. 12, 13; PL Vila,
figs. 7-11 ; PL X. figs. 11-14).

But there are no fibres at all in the middle half of each joint, which is composed of
the usual limestone network. Towards the upper part, however, the fibres (PL Vila.
fig. 1) become relatively slxorter in the centre but longer and longer towards the
periphery, and their ends approach more and more nearly to the middle planes of the two
joints which they unite. By the time the joints become cylindrical and then discoidal
immediately below the cup, the filjres are continuous through their whole length
(PL Vllb.), as is the case in the upper stem-joints of Pentacrinus (PL LVIII.
fig. 3, Is).

The articular surface which surrounds the opening of the central canal varies
considerably in its character. In many cirrus-joints it is merely an expansion of the
thickened rim of the opening, but does not extend across the joint face. Quenstedt
figures a stem-joint of this kind from the M^stricht chalk. ^ It is nominally referred to
Bourgueticrinus ellipticus, but I have seen no recent ones like it. In other cirrus-joints
and in the stem-joints of the Bourgueticrinidae the articular surface takes the form of a
more or less well-marked ridge, which lies either across or in the direction of the long
axis of the oval joint-face, and is pierced in the middle by the opening of the central

i Encriiiiden, Tab. 104, fig. 70.

8 THE VOYAGE OF H.M.S. CHALLENGER.

canal ; while the fossas for the attachment of the ligaments are on either side of it
(PL Vila. figs. 7-11 ; PL X. figs. 11-14).

A similar mode of articulation occurs between the two outer radials of most Comatulaj
and of a few species of Pentacrinus (PL XXX. figs. 11, 12 ; PL XXXII. figs. 16, 17 ;
PL XXXIV. figs. 3, 6), as well as in the fossil Extracrvmis, Apiocrinus, &\idi Millericrinus.
It is likewise very common between the first two joints after the radial and every other
successive axillary, in those species which have branching arms ; and also between some
of the lowest pinnule joints. It has often been incorrectly described as a syzygy or a
modified syzygy, though clearly distinguished therefrom by Miiller^ and by Dr.
Carpenter.^ Each of the two apposed faces is divided into two lateral halves by a
vertical ridge ^iierced by the opening of the central canal, around which it is more promi-
nent than at its ends. These fossse lodge the strong interarticular ligaments, and no
muscular bundles are interposed between the two joints. They are only capable of
lateral movement upon one another, and cannot take part in any movements of flexion or
extension, in which they act as a single segment only.

A peculiar modification of this bifascial articulation, as it may be called, occurs in
Bathycrinus. It is naturally best seen between the two outer radials (PL Vila. fig. 16)
and the two lowest brachials, on account of their larger size ; but it may be traced all
through the arms (PL Vila. figs. 20, 22). It was wrongly described as a syzygy by
Sir Wyville Thomson in Bathycrinus aldricliianns,^ and also by Danielssen and Koren
in Bathycrinus carpenteri.* The vertical articular ridge is relatively large, and the two
fossae at its sides small in proportion (PL Vila. figs. 16, 20, 22, //'). But at its lower
end is a small though tolerably deep pit {Id'), which lodges a bundle of closely set liga-
ment-fibres corresponding to those forming the dorsal ligament in an ordinary muscular
joint (PL Vllb. fig. 5, Id). This bundle probably enables the two joints to take a
larger share in the movements of flexion and extension than is possible in the bifascial
articulations of the other Crinoids. The three lis;aments, viz., the two lateral ones
(//) and the median one on the dorsal side (Id) are all seen in section in PL Vllb.
fig. 8.

A side view of a decalcified arm shows that there is a greater length of ligament
between the two joints united in this way than there is between two joints which are
united by syzygy in the much larger arm of a Pentacrinus ; and though its length is but
little greater than that of the fibres forming the syzygies in the arms of Rhizocrinus
rawsoni, yet the latter are less numerous than in the trifascial articulation of Bathycrinus,
especially on the upper (ventral) side of the central canal.

In all the four species of Bathycrinus which are considered in this Report, a trifascial
articulation, like that between the two outer radials, occurs between the first and second,

1 Op. cit., pp. 26, 30. 2 Oj}. a<.,pp. 715, 716. ' Journ. Linn. Soc. Land. (Zool.), vol. xiii. p. 50, 1876.

* Ilyorinus carpenterii, Nyt Magazin for Naturvidenslcaherne, Bd. xxiii. pp. 6-8 (of separate copj").

REPORT ON THE CRINOIDEA. 9

the fourth and fifth, and the seventh and eighth brachials (PI. VII. fig. 2). The third
brachial is articulated by muscles and ligaments to those before and behind it, as arc
also the sixth and ninth brachials. Beyond this point trifascial and muscular articula-
tions alternate with one another throughout the arm. But those brachials which are
united to their successors trifascially bear no pinnules as the remaining joints do ; and
in this respect they lose their morphological value as arm-joints, just as the hypozygal
of a syzygium does. From this point of view, therefore, the description of the trifascial
articulations as syzygia is perfectly correct. But they do not correspond to Miiller's
definition of a syzygy as an immovable sutural union of two joints. They occupy a
curiously intermediate position between a bifascial articulation and a syzygy jaroper ; for
they resemble the former in the movement of the joints upon one another, and the latter
in their occurring throughout the whole length of the arms, and in the absence of a
pinnule on the lower joint of every pair so united. They correspond exactly in their
distribution to the syzygies of Rliizocrinus, which come nearest to them in character,
being perfectly plain and simple, and not marked with radiating ridges as in the
Comatulse and some Pentacrinidse. But the trifascial articulation must not be
confounded with the peg and socket form of syzygy which is met with in Rliizocrinus
(PL X. figs. 1, 6, 8, 17, 18). In both cases there is a pit near the dorsal edge of
one of the apposed faces ; but in Bathycrinus this lodges a ligament (PL Vllb. fig. 8, Id)
which is attached in a corresponding pit in the other face (PL Vila. figs. 16, 22, Id') ;
while in Rliizocrinus this other face bears a peg-like process (PL X. fig. 17) which fits
into the pit, and thus checks rather than facilitates motion.

It is noteworthy that there seems to have been a trifascial articulation between the
two outer radials of the fossil Apiocrinus insignis, d'Orbigny, for the articular face of
the second radial is described by de Loriol ^ as presenting " un bourrelet vertical large,
epais et bifurqud pres du bord externe." The fork of this ridge at its dorsal end gives
the joint-face an altogether different appearance from the corresponding part in
Apiocrinus parhinsoni, and it is diflicult to see what can have been lodged in the
fossa between the two limbs of the fork, except a third ligamentous bundle such as
occurs in Bathycrinus.

In all the Neocrinoidea muscular articulations occur between the first and second
radials, between every axillary and the two joints which it bears, and between most
of the following arm-joints (PL III.; PL Vila. figs. 15, 17, 18, 19, 21,23; PL X.
figs. 1-4; PL XII. figs. 3-G, 8, 9, 12, 19, 20, 23; PL XXI. figs, la, ih, 2a, 2b,
Sb, 46, 4c, 5c, 6d, &c.). When the arms divide and the axillaries are simple, they may
be united by muscles to the preceding joints in Pentacrinus and Metacrinus (PL XII.
fig. 3); though this is never the case in the Comatulse. But if the axillaries are
syzygial joints, there is always a muscular articulation below the hypozygal. No
1 Paleontologie Frangaise. Terrain Jurassique, t. xi. Crinoides, p. 309, pi. Ivi. fig. 2c.

(ZOOL. CHALL. EXP. — PART XXXII. — 1884.) " 2

10 THE VOYAGE OF H.M.S. CHALLENGER.

muscles are ever found between the second and axillary radials of any Neocrinoid except
Eudesicrinus (p. 215), nor (except in Metacrinus) between the first and second brachials,
unless the latter be a syzygial joint. In the aberrant Metacrinus, however, the second
radial is a syzygial joint (PL XII. figs. 7-10) and the axillary is usually either the fourth or
the sixth radial (PI. XII. figs. 3, 4; PL XXXIX. fig. 1 ; Pis. XLIL, XLIIL, XLV., XLVL,
XLVIII.-LL); and there is usually a syzygy in the third joint after each successive axillary,
just as in the free arms of most ComatulEe. But the first two joints are united by muscles
instead of by ligaments, as in the Comatulse ; and I have not met with any instances of
bifascial articulation in the arms of this genus, which have nothing but syzygies and
muscular unions like the arms of Actinometra Solaris or Actinometra typica.

Each pinnule has a muscular attachment to the arm-joint which bears it (PI. Vc.
fig. 2, m; PI. Vila. fig. 21; PL XVII. fig. 1; PL XXXa. figs. lOa, 12«; PL XLI.
fig. 11 ; PL XLVII. figs. 11, 12) ; while in Metacrinus and some tropical Comatulse the
lower joints of the pinnules are united by muscles instead of by ligaments or suture
only (PL XIII. fig. 12).

The articular face of a joint which is connected with its successor by muscular
bundles presents a variety of fossai separated by intervening ridges, that are frequently
somewhat ill defined (PL Vila. figs. 15, 17, 18, 19, 21, 23; PL X. figs. 1-4; PL XII.
figs. 3-6, 8, 9, 12, 15, 19, 20, 23; PL XXL figs. la. ih, 2a, 2b, Sb, Ab, 5c, Qd, &c.).
Beneath its upper and lateral edges lie the two fossae in which the muscular bundles are
attached (PL Vila. fig. \^,,rm'). They are sometimes separated by a vertical ridge as
on the first radials of Antedon lyhalangium, Antedon eschrichti, and of most species of
Antedon and Prmnacliocrinus ; and sometimes by a groove which may extend down-
wards to the articular rim around the opening of the central canal, as in the first radials
of Antedon rosacea and most species of Actinometra and Pentacrinus (PL XXI. fig. Qd).
In these last types the fossae are comparatively small ; but in Promachocrinus and in
most species of Antedon they occupy nearly or quite half of the articular face, as in
Bathycrinus aldrichianus (PL Vila. fig. 15, rm' ; PL Vllb. fig. 5, rm). Below the
muscular fossae, and separated from them by more or less distinct ridges, are those
lodging the interarticular ligaments (PL Vila. fig. 15, W ; PL Vllb. fig. 5, and
PL Villa, fig. 7, li). They are bounded below by the transverse articular ridge, which is
pierced by the opening of the central canal. This is usually surrounded by a more or
less prominent rim, from which the ridges start that separate the muscular fossae from
the ligamentous ones and from one another. All these characters arc much less distinct
in the Pentacrinidse than in the ComatulfB.

Beneath the transverse ridge, and bounded below by the curved edge of the articular
face is a large fossa extending right across the face and lodging the dorsal elastic
ligament (PL Vila. fig. 15, Id'; PL Vllb. fig. 5, and PL Villa, fig. 7, Id). As a
general rule this ligament is chiefly concentrated in a small, more or less oval j^it situated

REPORT ON THE CRINOIDEA. 11

immediately beneath the articular ridge (PI. X. figs. 1-4 ; PI. XX. fig. 7 ; PI. XXXa.
fig. 11a; PL XXXII. figs. 6, 8-11, 15), and corresponding to the single pit which
lodo-es the third ligamentous bundle in the trifascial articulations of Bathycrinas
(PI. Vila. figs. 16, 20, 22, Id'). In Millericrinus, and more especially in Apiocriniis, the
portion of the dorsal fossa outside this central pit is enormously expanded ; and the true
articular surface becomes relatively insignificant.

The pairs of fossse are symmetrical on the two sides of each articular face in the
first and second radials (PI. Vila. fig. 15 ; PI. XII. figs. 20, 23 ; PI. XXI. figs. 5c, Qd);
but in the ordinary arm-joints, which have more or less oblique terminal faces and bear
pinnules at their sides (PL Vila. fig. 21 ; PL XXI. figs, ih, 4c), there is a disturbance
of the symmetrical arrangement of ridges and fossae which is so very evident in the
calyx of all Neocrinoids (PL Vllb. fig. 5 ; PL Villa, fig. 7; PL X. figs. 1, 4 ; PL XII.
fig. 15; PL XVIII. fig. 4 ; PL XXX. fig. 3 ; PL XXXIV. fig. 7), with the exception of
HolopiLS and its allies (PL III. figs. 1, 2 ; PL V. fig. 1).

12 THE VOYAGE OF H.M.S. CHALLENGER.

II.— THE STEM AND ITS APPENDAGES.

We have already seen that three types of stem occur in the recent Crinoids, which
are characteristic respectively of the families Pentacrinidas, Boui-gueticrinidse, and
Hyocrinidfe. Broadly speaking, it may be said that these embrace all the varieties of
stem which are met with in the Neocrinoidea. For the mode of union of the joints in
Apiocrinus, Millencrinus, and Eugeniacrinus must have been very much what it is in
Hyocrhms ; though the " root " of the first-named genus is somewhat peculiar in its
nature.

A. PENTAOPaNID^.

In all the genera of this family the stem consists of discoidal joints which are never
(in the adult) higher than wide, and have the characteristic petaloid markings on their
terminal faces (Pis. XL, XIV.; Pis. XV. figs. 4, 5 ; Pis. XIX., XXII ; PI. XXVI.
figs. 12-18; Pis. XXVIIL, XXIX; PL XXX. figs. 25-30; PL XXXa. figs. 1-7;
PL XXXVn. figs. 10-22; PL XXXIX. figs. 3-11; PL XLL figs. 1-5, 6-8, 15-17;
PL XLVII. figs. 1-9). Certain of these joints, separated from each other by intervals of
variable length, bear whorls of cirri, and they have consequently received from Sir
Wyville Thomson^ the very appropriate name of " nodal joints." There ai-e usually five
cirri at each node, situated in the direction of the rays (Pis. XL, XVIII. , XIX., XXXIV.-
XXXVL, XL., XLII.-XLVL, XLVIII.-LIL). Sometimes, however, one cirrus is
deficient, as shown in the left hand figure on PL XXXVI. ; while in Pentacrinus
alternicirrus there are only three cirri at one node, and two at each of those above and
below it, the positions of these two corresponding to the two gaps at the node of three
cirri (PL XXV.; PL XXVI. figs. 13, 14; PI XXVII. figs. 1-3). The cirri of two
successive nodes therefore alternate in position as the leaves do in the stem of a Labiate
plant, the two faces of the stem which bear no cirri at one node being the only ones which
have cirri at the nodes above and below it.

The function of the cirri, which vary considerably both in length and in stoutness, is
described by A. Agassiz ^ as follows : — " These they move more rapidly than the arms, and
use them as hooks to catch hold of neighbouring objects, and on account of their sharp

' Sea Lilies, p. V.

- Letter No. 3, on the dredging operations of tlie U. S. Coast Survey, Sr. " Blake," from December 1878 to March
10, 1879, Bull. Mus. Comp. Zoijl. vol. v. No. 14, p. 296.

REPORT ON THE CRINOIDEA. 13

extremities they are well adapted to retain their hold. The stem itself passes slowly
from a rigid vertical attitude to a curved or even drooping position."

The shape of the nodal joints is markedly dilfereut from that of the internodal joints
which separate them. They are not only somewhat higher (PI. XIX. figs. 3, 4 ;
PI. XXVI. fig. 12 ; PI. XXVII. fig. 1 ; Pis. XXXV., XXXVI. ; PI. XXXIX. fig. 3 ;
PI. XLI. figs. 1, 5, 15 ; PL XLIL; PI. XLV. fig. 6 ; PL XLVII. figs. 1, 2), but the outline
of the upper non-syzygial face is different from that of an internodal joint. (Compare
PL XXXIX. figs. 4, S ; PL XLI. figs. 2, 3, 6, 7, 16, 17 ; PL XLV. figs. 2, 4 ; PL XLVII.
figs. 4, 5, 8, 9 ; PL LI. figs. 9, 10 ; PL LIII. figs. 3, 4.) In the recent Pentacrinus and
Metacrinus every nodal joint is united by syzygy to the top joint of the next internode
below. I propose to give the name "infra-nodal" to this joint, which is really the
hypozygal of the syzygial pair (PL XXII. figs. 19-22 ; PL XXVI. figs. 12-16 ; PL XXX.
figs. 26-29 ; PL XXXVII. figs. 5-8, 19-22 ; PL XXXIX. figs. 3-7 ; PL XLV. figs. 3-5 ;
PL XLVII. figs. 3, 7 ; PL L. figs. 19-22 ; PL LIII. figs. 3, 5). The apposed faces of
these two joints are much more distinctly stellate than are those of the remaining stem-
joints, the re-entering angles of the star being the points of attachment of the cirri,
and the syzygial surfaces of the mature joints are almost smooth and devoid of any
markings whatever.

The syzygial union of two stem-joints is efi'ected, just as it is in the rays and arms,
by short fibrils of connective tissue, very numerous and closely set. They form a kind
of " cement substance," as it was formerly called, which is connected at its ends with the
organic basis interpenetrating the calcareous network of the stem -joints, just as the
" cement-substance " of the arm-syzygies is connected with the organic basis of the
brachials. But these fibrils are absolutely distinct from those of the five long ligamentous
bundles which occupy the internodes. The latter are often spoken of as tendons, and
have been wi'ongly described as extending throughout the whole length of the stem.
Were this really the case, it is difiicult to see how the stem could break across at the
syzygies so easily as it does ; for there would be no reason why the five tendons should
be weaker at these points than at anywhere else in the internodes, while the loose ends
of the tendons should appear at the surfaces of fracture, just as they do where an inter-
node is forcibly broken across. But this is not the case : when the stem is decalcified the
joints separate very readily along the lines ^of syzygy, and it is then apparent that the
five tendons run from the lower portion of each infra-nodal joint down into the upper
portion of the next nodal joint below it. They end within these two joints, just as do
the ligaments which connect two brachials, terminating either in looped extremities or
else passing into the connective tissue plexus which forms the substance of all the joints
whether of arm or stem.

The various internodal joints are, as it were, strung upon these tendons, which are
thus not continuous, but divided up into lengths, each corresponding to an internode ;

14 THE VOYAGE OF H.M.S. CHALLENGER.

and there is uo more direct communication between tlie fibres forming one of the tendons
that ends in a nodal joint and the fibres of the syzygy between that joint and the
infra-nodal, than there is between the ligaments uniting the second radials of Pentacrinus
wyville-thomsoni to the first (PI. XXIV. figs. 8, 9, Id), and the fibrils of the syzygy
between the second and third radials. In both cases the two sets of fibres are separated
by the organic basis of a joint of the skeleton ; in the one of the ray, and in the other of
the stem.

The amount of increase in the size of the nodal joint varies considerably in the
difi'erent species. They are not specially prominent in Pentacrinus asteria (PI. XIII.
figs. 4, 8), Pentacrinus miilleri (PI. XV. fig. 4), Pentacrinus hlalcei (PL XXXI. fig. 3),
or in Pentacrinus naresianus (PL XXVIII. fig. 2 ; PL XXXa. fig. 6). In Pe7itacrinus
decor us, on the other hand (PL XXXVI. ), the joint expands considerably from its upper
edge down to the top of each cirrus -socket, and then narrows again ; while in Pentacrinus
ivyville-tliomsoni (PL XIX. figs. 3, 4) the sockets are very jM'ominent, and the joint is
widest near its lower edge. In the genus Metacrinus, however, the cirrus-sockets are by
no means specially prominent (PL XXXIX. fig. 3 ; PL XLI. figs. 1, 5, 15 ; PL XLVII.
figs. 1, 2 ; PL XLIX. fig. 3 ; PL LI. figs. 6-8 ; PL LII. fig. 2 ; PL LIII. fig. 6).

In Pentacrinus asteria (PL XIII. figs. 4, 8) and Pentacrinus miilleri (PL XV. fig. 4),
the sockets are usually more or less transversely oval in shape and weU defined below, so
as to be almost or entirely limited to the nodal joint. In Pentacrinus decorus, however,
the articular surface occupies the broad end of a pear-shaped depression, which is con-
tinued down on to the infra-nodal joint (PL XXXVI. ), and thus gives it a distinctly
stellate outline even when seen from beneath (PL XXXVII. figs. 10, 20); while the
ordinary internodal joint has a rounded or pentagonal outline. The lowest internodal or
" supra-nodal" joint of this species is in no way different from the other internodal joints
above it ; while the infra-nodal is hollowed laterally by the downward extensions of the
cirrus-sockets(PL XXXVI. ; PL XXXA^II. figs. 8, 19). In Pentacrinus ivyville-thomsoni,
however, the lower edge of the socket proj ects outwards beyond the level of the infra-
nodal (PI, XIX. figs. 3, 4). This joint is but little difi'erent from those below it, except
just at its upper edge where it meets the enlarged surface of the nodal joint that rests on
it (PL XXII. figs. 21, 22). The supra-nodal joint, on the other hand (PL XXII. fig. 17),
is not so round as the remaining stem-joints (fig. 23); for it is slightl)?^ hollowed hj the
upper portions of the cirrus-sockets, and thus more nearly approaches the shape of the
upper face of the nodal joint on which it rests (fig. 18).

This condition becomes still more marked in Metacrinus, which has relatively low nodal
joints with wide and comparatively inconspicuous sockets (PL XXXIX. fig. 3 ; PL XLI.
figs. 1, 5, 15 ; PL XLVII. figs. 1, 2 ; PL LI. figs. 6-8 ; PL LIL fig. 2 ; PL LIII. fig. 6),
just as in Pentacrinus asteria and Pentacrinus miilleri (PL XIII. figs. 4, 8 ; PL XV.
fig. 4) ; but both supra- and infra-nodal joints share in the formation of the cirrus-socket,

REPORT ON THE CRINOIDEA. 15

the re-entering angles of the former being deeper than those of an ordinary internodal
joint. (Compare PL XXXIX. figs. 8, 9 ; PL LIIL figs, 2, 4).

The occurrence of syzygies in the stem of the Pentacrinidse has long been familiar
to palaeontologists. This is due to the fact that the stem breaks most easily at these
points, so that stem fragments are not uncommon in the fossil state with syzygial faces
at one or both ends. This has been noticed both by Quenstedt^ and by de LorioP ; but
these fragments have sometimes been figured in an inverted position, the syzygy being
represented as at the upper surface of the nodal joint, whereas it is naturally at the
under surface.

Within each nodal joint the vascular axis of the stem expands into a miniature
edition of the chambered organ in the calyx (PL XXIV. figs. 3, 4, ch.n.) ; and from each
chamber is given off one cirrus- vessel (fig. 4, cv). This is sheathed in a very delicate
extension of the fibrillar envelope of the vascular axis [ca), and passes outwards into the
central canal of the cirrus.

The number of internodal joints varies very considerably in the different species of
Pentacrinidae. There may be only one or two as in Pentacrinus madearanus
(PL XVI.) ; or the number may reach forty-five, as in the lowest parts of the stem of
Pentacnniis wyville-thomsoni (PL XIX. fig. 1), or any lesser figure (Pis. XL, XIV.,
XXVIIL, XXXIV.-XXXVL, XXXVIIL, XL., XLIL, XLIV., XLV., XLVIIL-LIII.)
These internodal joints are sometimes smooth externally (PL XL ; PL XIII. fig. 8 ; PL XIX.
figs. 1-5 ; PL XXVII. fig. 1 ; PL XXVIIL fig. 2 ; PL XXXa. fig. 6 ; PL XXXI. fig. 3 ;
PL XXXV. fig. 2; PL XLL figs. 5, 15; PL XLIL-XLIV. ; PL XLVII. fig. 6); or
they may be more or less ornamented with ridges and tubercles (PL XIII. fig. 7 ;
PL XXXV. fig. 1 ; PL XXXVI. ; PL XXXVIIL; PL XXXIX. figs. 3, 8-11 ; PL XL. ;
PL XLL fig. 1; PL XLV. figs. 1, G ; PL XLVIL figs. 1, 2 ; PL LI. fig. 8; PL Lll. fig. 2;
PL LIIL figs. 2, 4-6). The young joints formed at the top of the stem are steUate or
nearly pentagonal (PL XIIL fig. 9; PL XXII. figs. 1-12; PL XXIII; PL XXVI.
fig. 18 ; PL XXXIV. fig. 9 ; PL XXXVII. figs. 13-16), so that the upper part of the stem
is marked by five more or less prominent interradial ridges (PL XL ; PL XIIL fig. 1 ;
PL XV. figs. 1, 2 ; PL XVIII. figs. 1, 2 ; PL XIX. figs. 1, 6, 7 ; PL XXXI. ; Pis. XXXV.-
XXXVII. ; PL XXXIX. fig. 1 ; PL XLIIL). In some species {oi Metaainus especially)
this condition is retained throughout the whole length of the stem (PL XXXVIIL ;
PL XXXIX. figs. 3-11 ; PL XL. ; PL XLIX. figs. 1-3) ; but in others the joints gradually
become more rounded, or at any rate pentagonal, as new ones are formed successively
above them (PL XL ; PI XIIL fig. 11 ; PL XIX. figs. 1, 4, 5 ; PL XXII. figs. 13, 14,
23-26 ; PL XXVIIL fig. 2 ; PL XXX. figs. 25-30 ; PL XXXVI. ; PL XLL figs. 3, 7 ;
PL XLA'II. figs. 4, 8). A continual production of new joints goes on at the top of

1 Encriniden, pp. 196, 230, Tab. 98, figs. 2, .3, Wi.

- Monographie des Crinoides fossilesde la Suisse, Mifm. Soc. Pal. Suisse, 1877-79, pp. 122, 144.

16 THE VOYAGE OF H.M.S. CHALLENGER.

the stem immediately beneath the calyx, what appears to be the top stem-joint in a side
view having one or two more smaller joints resting in its upper surface but not reaching
the exterior as shown in PI. XXXIV. fig. 9.

The top stem-joint — for the time being — of a specimen of Pentacrinus ivyville-
thomsoni is shown in PI. XXII. fig. 2. Resting on its upper si;rface is a smaller stellate
plate without any markings whatever, which in its turn would appear as the top stem-
joint until it was replaced by the development of another above it. As these joints are
relatively carried downwards from the calyx in succession by the appearance of younger
ones above them, they also become separated from one another by the intercalation of
new joints between them. Various stages of this process are shown in PI. XXII.
figs; 9-12, while fig. 4 shows an isolated young joint, and fig. 5 the depression in the
nest joint, which lodged it. Similar intercalated joints are shown in Pis. XXII. fig. 3,
and PL XXIII. figs. 1, 2. The result of this process is that the growing part of the stem
appears to consist of thick and thin joints alternating with one another (PI. XIII. fig. 1 ;
PI. XIV.; PL XV. figs. 1, 2; PL XVIII. fig. 1, 2; PL XIX. fig. 2, 6, 7 ; Pis. XXV.,
XXVIII. -XXX.; XXXIV.-XXXVIL; PL XXXIX. fig. 1; PL XLIII. fig. 2; PL XLVIII.
fig. 2). The former are the older, the latter being subsequent additions. These
intercalated joints are always internodal, and the process goes on until the number of
joints between any two nodes reaches a certain average, which is constant for each
individual species.

There are many indications that the increase in length of the stem of the fossil
Pentacrinidae was due to the same process. Quenstedt ^ gives an excellent figure of a
young concealed joint superposed upon an older and larger one, very much as shown in
PL XXII. figs. 9-12; and others of his figures upon the same plate illustrate the
difierent stages of growth on the stem of Extracrinus subangularis, as many as three or
four concealed joints being sometimes found intercalated between two of the larger ones.
In this species, too, with a stem which may reach 50 or 70 feet in length, the production
of nodal joints at the top of the stem must have been very rapid. According to
Quenstedt' forty or more succeed one another without any internodal joints being visible
externally, though the presence of concealed intercalated joints is revealed by the
examination of longitudinal sections of the stem. The final result of their growth was to
enormously increase the total number of internodal joints.

Stem -fragments have been found by Quenstedt consisting of as many as eighty joints,
all internodal, and it is impossible to say how many more there may have been ; while
he states that he finds traces of small intercalated joints in almost every part of the stem.

In most cases the new joints which have been intercalated between two older ones
eventually reach the same size as their predecessors, so that it is difficult to tell the
older from the younger joints in any mature stem. But in some species there appears to
1 Encriniden, p. 298, Tab. 101, figs. 24a, 246. ' Hid., p. 297, T;i1_). 101, figs. 16-19.

REPORT ON THE CRINOIDEA. 17

be a permanent inequality, large and small joints alternating all down, the stem. This is
the case, for example, in the fossil Pentacrinus jaccardi and Pentacrinus nicoleti figured
by de Loriol,^ and also to some extent in the recent Metacrinus mnseleyi (Pis. XLV.,
XL VI.).

There is very little difference between the upper and the under faces of the young-
nodal joints, or between either of them and the ordinary internodal joint (PI. XXII.
figs. 1, 6-8, 15; PL XXXa. figs. 2, 3; PL XXXVIL figs. 14-16; PL LL figs. 2-5).
But as the joints become older and more pentagonal their difterences are more apparent
(PL XIIL figs. 2, 3, 5, 6, 10; PL XXIL figs. 16-18; PL XXXVIL figs. 11-13, 17, 18);
while in the lower part of the stem the simple syzygial nature of the under face of the
nodal joint and its loss of the denticulate petaloid markings become very distinct
(PL XV. fig. 6 ; PL XXIL figs. 19, 20 ; PL XXVI. figs. 13, 14 ; PL XXVIL figs. 2, 3 ;
PL XXX. figs. 26, 27 ; PL XXXVIL figs. 21, 22 ; PL XXXIX. figs. 4, 5 ; PL XLV.
figs. 4, 5 ; PL L. figs. 21, 22).

In some species of Pentacrinus, e.g., Pentacrinus ivi/viUe-thomsoni, the lowest and
therefore the oldest stem-joints gradually lose the more or less prominent ridges which
appear on the faces of those higher up the stem, and become much more smooth and
simple in their character (PL XXIL figs. 23-26).

A similar change seems to take place in Pentacrinus asteria (PL XIIL figs. 10, 11),
and also in other species, though I have not been able to trace it so distinctly as in
Pentacrinus ivyville-thomsoni. It is manifested externally by the gradual disappearance
of the crenulation of the interarticular lines, which is so very prominent in the upper and
midcUe parts of the stem (PL XIIL figs. 7, 8 ; PL XV. figs. 1, 2, 4 ; PL XIX. figs. 2-5 ;
PL XXV. ; PL XXVIL fig. 1 ; PL XXXI. fig. 8 ; Pis. XXXV. -XXXVIL ; PL XLI.
figs. 1, 5, 15; PL XLIILfig. 1).

The amount of crenulation varies considerably in diff"erent species, according to the
position of the large teeth bordering the' outer ends of the petaloid spaces. When these
start from near the edge of the joint, as in Metacrinus cingulatus, Metacrinus nohilis, or
Metacrinus costatus (PL XLI. figs. 1-3, 5-7 ; PL XLIX. figs. 3-5), the interarticular line
is well crenulated. But there is sometimes a sort of rim outside the ends of the teeth, as
in Metacrinus murrayi and Metacrinus varians (PL XLI. figs. 15, 17 ; PL XLVII. figs.
6-9), and the external crenulation is then less marked. This outer rim is only formed
comparatively late, the teeth of a young joint starting directly from its edge, as is well
shown in Pis. XXIL, XXIIL, and XXXVIL

The increase in the length of the internodes only takes place gradually, and i^ciri
passu with the continual formation of new joints just below the calyx. Hence, in the
upper part of the stem, there is a variable number of premature internodes, those nearest
the summit being the shortest, and consisting of the smallest number of joints. The

1 Swiss Crinoids, pp. 130, 140, pL xv. figs. 13, 36.
(ZOOL. CHALL. EXP. — PART XXXTI. — 1884.) li 3

18 THE VOYAGE OF H.M.S. CHALLENGER.

upper part of the stem is also distinguished from the more fully grown portion below it
by the presence of the " interarticular pores." These are small pore-like openings between
the successive joints which are situated in the re-entering angles between the interradial
ridges, and are therefore radial in position (PL XL ; PL XIII. fig. 7 ; PL XV. figs. 1,2;
PL XIX. figs. 2, 3 ; PL XXV. ; Pis. XXVIIL, XXIX., XXXIV.-XXXVIIL, XLIV.,
XLVIII. ; PL XLIX. fig. 2 ; PL L. figs. 1, 3). They are produced by the apposition of
two faint grooves radiating outwards from the centres of the contiguous joints, which are
largest at their central ends and shallowest towards the periphery (PL XIII. fig. 9 ; PL
XXII. figs. 8, 10, 12 ; PL XXXVII. figs. 10-18, 20, 22). They do not reach the central
canal of the stem ; and so there is no communication eff"ected by the agency of these
pores between the internal vascular axis and the exterior. Similar openings occur in
many Comatulaj, leading into spaces between the upper surface of the centro-dorsal and
the under surfaces of the radials which rest upon it ;^ but they efi"ect no communication
between the body-cavity and the external medium.

Until comparatively lately but little has been known respecting the termination of
the lower part of the Pentacrinus stem. Nearly all the specimens obtained had a
fractured stem, from which no conclusions could be drawn. From what we know of the
development of the Comatulaj we may fau-ly assume that the young Pentacrinus
commences life attached to some foreign body by means of a terminal plate (the dorso-
central) at the base of its stem, and a calcareous expansion subsequently developed around
it. Mature individuals have been found attached in this way to telegraph cables.
Capt. E. Cole of the telegraph steamer " Investigator " has reported to Prof. Agassiz
" that he has frequently brought up the West India telegraph cable on which Pentacrini
were attached, and that they are fixed, the basal extremity of the stem spreading slightly,
somewhat after the manner of Holopus, so that it requires considerable strength to
detach them."^ The condition of the lower part of the stem fragment of Pentacrinus
naresianus, represented in PL XXXa. fig. 4, is perhaps due to this mode of attachment.

At the same time there appears to be ample evidence that a Pentacrinite may lead
the same sort of free life that a Comatula does, attaching itself temporarily by its cirri.
Sir Wyville Thomson long ago pointed out, in the ease of Pentacrimis decorus,^ "that the
animal seems to have had the power of detaching itself" at any of the syzygies of the
stem in the same sort of way as the arms are thrown off during life, or break up after
death. He described an individual in which the stem terminated below in a worn and
rounded nodal joint, and he supposed it " to have finally parted from its attachment and
to have led a free Life." He stated some years afterwards that this was the case in aU the
complete specimens which he had seen, " showing that the animal must have been for
long free from any attachment to the ground."* He then went on to describe the same

■ The Genus Actinometra, loc. cit, pp. 8S-90, pi. viii. figs. 5, 7.

- Bull. Mm. Comp. Zibl., vol. v. Ko. 14, p. 296. ^ Sea Lilies, p. T.

' Tlie Depths of the Sea, pf. 442-44^1.

REPORT ON THE CRINOIDEA. 19

condition as it occurs in Pentacriniis loyville-thomsoni. " All the steins of mature
examples of this species end inferiorly in a nodal joint surrounded by its whorl of cirri,
which curve downwards into a kind of grappling root. The lower surface of the terminal
joint is in all smoothed and rounded, evidently by absorption, showing that the animal
had for Ions been free. I have no doubt whatever that this character is constant in the
present species, and that the animal lives loosely rooted in the soft mud, and can
change its place at pleasure by swimming wdth its pinnated arms ; that it is, in fact,
intermediate in this respect between the free genus Antedon and the permanently fixed
Crinoids."

Many other species of Pentaerinus and some of Metacrimis exhibit the same condition.
It is best seen in Pentaerinus ivyville-thomsoni, in which the nodal joint sometimes loses
its ordinary characters altogether, becoming much enlarged and rounded below so as to
be almost hemispherical in appearance (PI. XXII. fig. 17)} In other cases, however, it
retains its petaloid form and more or less of the small amount of sculpture which is
usually found upon its lower face ; and a small rounded tubercle appears in the centre of
the latter closing up the opening of its central canal. This is the usual condition of other
species of Pentaerinus {e.g., Pentaerinus asteria, PI. XL) and of Metacrimis ; and
the analogy between it and the condition of a young Comatula just detached from
its stem is very striking, as was pointed out by Sir Wyville Thomson."^ In both
cases the severance takes place between a nodal joint and the top joint of the internode
below it.

The relations of the two longest stems that I have met with in this condition are
shown as follows : —

Pentaerinus decorus, stem 48 cm. long, rounded ofi"at the thirtieth node.
Metacrinus angulatus, stem 3 8 '5 cm. long, rounded ofi" at the thirty-fifth
node.

With one exception {Pentaerinus maclearanus, PL XVI.) the three shortest of these
semi-free stems that I have examined all belong to Pentaerinus alternieirrus. A, 47 mm.
long, ends at the eleventh node ; B, 49 mm., ends at the eleventh node ; and C, 55 mm., ends
at the twelfth node. On the other hand, the smallest number of nodes in a semi-free stem
occurs in Pentaerinus wyville-thomsoni; one individual having a stem 90 mm. long, which

1 The unusual enlargement of the lowest nodal joint in this individual suggests the idea that the structures which
have been described by Hall under the nameof ^7ic!/rocnnMs(Fifteenth Annual Report, New York State Cabinet of Natural
History, 1862, pp. 89, 90) may be the detached stems of a Palseocrinoid in the semi-free condition. According to HaU they
"have the form of a bulb or thickened column, with lateral ascending processes and a central ascending column of greater
or less length ; " and he suggests that they "indicate the existence of a free floating Crinoid with the thickened bulb below
serving as a balance for the column and body above. The articulating scar on the lower extremity of the smaller ones
indicates that the animal was fixed in its yoimg state." The four lateral spine-like processes may very well have been
cirri, the jointed structure of which has become obliterated by a calcareous overgrowth, just as in the lower part of the
tetramerous stem.

» Sea Lilies, p 10.

20 THE VOYAGE OF H.M.S. CHALLENGEE.

ends at the sixth node. This is due to the great length of the internodes in this species
(PI. XVIII. fig. 3 ; PI. XIX. fig. 1).

Pentacrinus maclearanus presents exactly the opposite type of structure. There are
only twelve nodes in the stem of the solitary individual obtained (PL XVI.). But these
all occur in a stem barely 40 mm. long, as there are never more than two, and generally
only one internodal joint ; while the cirri cluster thickly round the stem, so that it has
an appearance more like that of Exti-acrinus hriareus than is commonly met with in the
Pentacrinidse. It is noteworthy that in the last-named sjjecies the stem does not seem
to have reached any great length, and that it sometimes tapered downwards.^

This peculiarity was also noticed by MM. Eudes-Deslongchamps in some stems
belonging to a large colony of Pentacrinites which they discovered in the Great Oolite
of Soliers, near Caen ; ' while it is very characteristic of Millericrinus lyratti from the
same horizon in Gloucestershire ; ' and also of the Carboniferous Woodocrinus, certain
Blastoids, and of the Silurian Glyptocystites, Pleurocystites, and other forms. The most
remarkable instance of this in a fossil Crinoid, however, is that of the Lower Silurian
Gly2)tocrinus schqfferi of S. A. Miller/ for which he has recently established the new
genus Fycnocrinus. In one specimen found by Miller the lower jiart of the stem was
" wound around a Crinoid column of a distinct species, almost as neat as a thread can be
wound upon a spooL The column gradually tapers as it coils, until it becomes so smaU
as to be scarcely visible to the naked eye, the larger plates of the column which, give it
that banded appearance, or make it resemble a string of small spools, gradually diminish,
and before the column terminates it becomes as smooth as a silken thread."

Two other species from the same locality at Cincinnati, Lichenocrinus dubius and
Dendrocrinus navigiolum, were also found by Miller to have tapering stems. In the
case of the former he infers that " the column was free and used to direct and guide
the course of the animal through the water, and perform such other functions as were
performed by the columns of other floating Crinoids, except that it was never used
for purposes of attachment." One must not, however, conclude at once, from the
tapering condition of the stem in a fossil Crinoid, that the animal was free in its habits.
In a young Eucalyptocrinus crassus, for example, which is figured by HaU,' the stem
tapers downwards very considerably, but is attached below liy a spreading root.

I have found a tapering stem in certain individuals belonging to six species of
recent Pentacrinidge, but it appears to ha the exception rather than the rule, and is
therefore entirely devoid of any systematic value.

1 Encriniden, p. 271.

- Etudes sur les I'tages Jurassiques i:ifi-rieurs de la Normandie, Paris, 1864, p. 232.

■' On some new or little kno^ra Jurassic Crinoids, Quart. Jov/m. Geol. Soc, vol. xxxviii. pp. 31-33, pi. i. figs. 6-8,
10-14.

* Description of i'oiir new Species and a new variety of Silurian Fossils, and remarks upon others, Journ. Ciiicinn.
Soc. Nat. Hint., vol. iii., 1880, pi. vii. fig. 3, p. 2 (of separate copy).

5 Tweuty-eiglitli Annual Report of tlie New York State Museum of Natural History, Albany, 1879, pi. xvii. fig. 5.

REPORT ON THE CRINOIDEA. 21

In one example of Pentacrinus cdternicirrus, which has a stem only 55 mm. iu
length, and rounded off at the twelfth node, the width diminishes gradually from above
downwards, though not to any very great extent. The same is the case in another
specimen with a stem 47 mm. long and rounded off at the eleventh node. In other
examples of the same type, however, the width of the stem remains uniform or even
increases slightly from above downwards. In a large specimen of Pentacrinus naresianus
the width of the stem, which is 5 mm. at the eighth node, is reduced to 3 "5 mm. by the
thirtieth node, which is rounded off below.

In the young Pentacrinus ivyville-thomsoni represented in PI. XVIII. fig. 3, the stem
is 65 mm. long ; but it is not half as wide at the lowest (sixth) node as it is beneath the
calyx. It was described by Sir Wyville Thomson ^ as follows : — " The stem is
broken off in the middle of the eighth internode from the head. The lowest complete
internode consists of fourteen joints, the next of eighteen, the next of twenty,
and the next of twenty-six joints. There are eight joints in the cirri of the
lowest whorl, ten in those of the second, twelve in those of the third, and fourteen
in those of the fourth. This is the reverse of the condition in adult specimens,
in all of which the numbers of joints in the internodes, and of joints in the cirri, decrease
regularly from below upwards. The Ijroken internode in the young example and the
three internodes above it are all atrophied and undeveloped ; and suddenly at the third
node from the head the stem increases in thiclaiess and looks as if it were fully
nourished. There can be no doubt that in early life the Crinoid is attached, and that
it becomes disengaged by the withering of the lower part of the stem." ^

The diminution in the size of the stem is rather more gradual than is imphed in
the above description, for it commences at the head and extends regularly down to
the third node, where there is a more sudden change, as there is again at the fourth,
below which the diameter decreases but slowly.

But it is in a very young individual of Pentacrinus decorus (PI. XXXV.), in fact
the youngest Pentacrinite that I have seen, that this downward tapering of the stem
and gradual diminution in the size of the cirri are most marked. The stem is a trifle
over 60 mm. in length, and has ten distinct cirrus-whorls apart from the very small

1 On the Crinoids of the " PorcxTpiiie " Deep-Sea Dredging Expedition, Proc. Boy. Soc. Edin., vol. %'ii. p. 768.
Also in the Depths of the Sea, p. 445.

2 The above description requires a little correction. The stem iu its present condition, as represented in PI. XVIII.
tig. 3, is broken at the top of the seventh internode, ivhich is certainly the one referred to by Sir Wyville as the eighth.
But its length is at least 5 mm. greater than as stated by him ; and the number of joints in the two lowest inteniodes
should be given as fifteen and seventeen, not fourteen and eighteen. They are di-awn correctly in the figure,
but the joints at the top of the third internode are not properly represented, as is also the case with some of the
cirri. The figures of this and of the other plates drawn for Sir WyviUe Thomson had been on stone for so long
when the stalked Grinoids came into my hands after his death, that I thought it better to let any errors remain
unaltered rather than to risk spoiling the plate by correctmg them. The description given of the cirri is also only
roughly accurate. Neither on the fourth whorl from the bottom nor anywhere else on the stem are there any
cirri with fourteen joints, even where the terminal claw is included ; though the regular decrease in the size of the
cirri from above downwards is very striking, as pointed out by Sir WyvUle.

22 THE VOYAGE OF H.M.S. CHALLENGER.

ones just beneath the calyx. Throughout its whole length it tapers downwards from
the calyx, rapidly at first, then more slowly and afterwards somewhat rapidly again,
till the joints are but little larger than those of the most developed cirri on its middle
part. The number of internodal joints also diminishes in this lower part of the stem ;
for while it i-eaches eleven or twelve in the middle of the stem, the numbers in the four
lowest internodes are respectively nine, seven, five, five ; and the remains of the cirri borne
at the intervening nodes show a corresponding diminution in size. The inferior termi-
nation of the stem is not known, as it is broken at the syzygy beneath the lowest whorl
of cirri.

The free mode of life appears to be attained in these individuals, not by actual
fracture of the stem at a node so as to shorten it more or less, but by the lower and
therefore older part remaining undeveloped, while new joints appear in succession above
it, each growing to a larger size than those previously formed. The stem thus becomes
slender and tapering, and but ill adapted for attaching itself below ; but its length
is not diminished so much as if it were broken at a node.

The downward tapering of the stem in some of the fossil Pelmatozoa has been
already noticed ; and it is evidently a character of more general occurrence than
was suspected by Sir Wyville Thomson. Quenstedt^ contrasts the comparatively
short tapering stems of Extracrinus briareus with the gigantic ones of Extracrinus
subangidaris, which may reach the length of 50 or even of 70 feet ; and he
suggests that the former type and its allies " konnten gleichsam als eine Comatula
betrachtet werden, deren Knopf zu grosserer Lange in einer Zeit heranwuchs, wo es
noch keine eigentlichen Comateln gab." De LorioP in like manner regards it as
probable — " qu'ils avaient, h I'etat adulte, une tige court, libre, et qua I'aide de leurs
cirrhes tres nombreux et tres longs ils pouvaient nager facilement et se transporter,
rapidement peut-etre, d'un lieu h, un autre ; ils avaient aussi la faculte de se fixer a
quelque objet, lorsqu'ils en avaient le desir, au moyen des crochets dont est munie
Fextremit^ de leurs cirrhes."

I suspect, however, that the swimming was efi"ected rather with the arms than with
the cirri, which are not used for that purpose by the Comatulte, and would have to
be moved with considerable power in order to efi"ect the locomotion of the animal.
The condition of so many recent species is a strong argument in favour of the views
formerly expressed by Buckland ^ and others regarding the possible locomotive powers
of the Liassic Pentacrinidae, though they have been somewhat discredited of late. Now
too that their recent representatives have been found so abundantly in depths of less
than 100 fathoms, instead of being exclusively abyssal ty|)es as was once supposed, the

1 Encriniden, p. 271.

2 Notice sur le Pentacrinus de Sennecey-le-Grand, Clialon-sur-Saone, 1878, p. 12.
' Geology and Mineialogy, vol. i. p. 437.

REPORT ON THE CRINOIDEA. 23

possibility of their becoming attached to floating timber does not seem so very distant after
all. They may have been attached above by a slightly spreading base as on the modern
telegraph cables ; or, on the other hand, they may have been drifted in large numbers
by the currents after detaching themselves from their original base of attachment.

Occupying the innermost part of the stem of a Pentacrinite, and lodged within its
central canal, is its internal vascular axis (PL XXIV. figs. 1-5 ; PI. LXIL). This
consists of five peripheral vessels arranged around a central one. The former
PL XXIV. figs. 2-5 ; PL LVIII. fig. 3 — ch') are downward extensions from the chambers
of the quinquelocular organ within the calyx (PL XXIV. figs. 6-8 ; PL LVIII.
figs. 1-3— c/i; PL LXIL) ; while the latter (PL XXIV. figs. 2-5, v.) is similarly
connected with the axial vessel or vessels of the chambered organ (PL XXIV. fig. 6 ;
PL LVIII. figs. 1-3 — V. ; PL LXIL). This central vessel does not increase in size at
the nodes, where the peripheral vessels expand considerably, so as to form a miniature
chambered organ (PL XXIV. figs. 3, 4, ch.n.) ; and each chamber gives off one cirrus-
vessel (fig. 4, cv.). The fibrillar sheath around the chambered organ in the calyx
(PL XXIV. figs. 6, 7 ; PL LVIII. fig. l—ca; PL LXIL) is continued down the stem,
around its vascular axis (PL XXIV. figs. 1-5, ca; PL LVIII. fig. 3; PL LXIL).
It is sometimes closely surrounded by a more or less complete ring of pigment masses,
similar to those which occur in the surrounding tissue (PL XXIV. figs. 2, 5, p.) ; but
in other parts of the same stem these are absent in the immediate neighbourhood of the
central axis (figs. 3, 4). Radiating extensions of the latter are frequently to be seen
(fig. 1. fig. 2, ca'). They proceed outwards into the organic basis of the skeleton, and then
become lost, though they probably reach the epidermis, like the similar branches from the
axial cords of the arms and pinnules, to which a nervous nature has long been attributed.

B. BOURGUETICKINID.E.

The type of stem which occurs in this family differs in many respects from that
characteristic of the Pentacrinidse. The joints are very variable in their relative propor-
tions, instead of being uniformly discoidal ; they never form syzygial unions, liut are
freely movable upon one another, and are connected by successive pairs of hgamentous
bundles instead of being strung, as it were, upon five tendons of variable length.

In the only two recent genera which belong to this family, Bathycrinus and
RMzocrinus, one or more of the young upper stem-joints are simple circular disks, with
little or no markings of any kind upon their terminal faces (PL VII. figs. 1-3, 11 ;
PL Vila. fig. 3-6 ; PL Villa, fig. 1 ; PL IX. figs. 1-3 ; PL X. figs. 2, 9, 10 ; PL LIII.
figs. 7,8). But lower down the stem the joints become first cuboidal and then elongated,
so that their length may be two or three times their diameter (PL VII. figs. 1, 10;
PL Villa, figs. 2, 3 ; PL IX. figs, 1, 3 ; PL LIII. figs. 7, 8). The younger of these
elongated joints are simply cylindrical ; but the older ones are more dice-box shaped mth

24 THE VOYAGE OF H.M.S. CHALLENGER.

expanded ends, while the articular ridges in the long axes of the terminal faces cross one
anotlier at various angles.

Tliis mode of articulation is common to all the Bourgueticrinidie, though the stem-
joints are not always so long as in Rhizocrimis and Bathycrinus. It occurs also in the
curious senus TJiioHiericrimts and in the stem of the larval Comatida. Both Thiollieri-
crinus and Bourgueticrinus occur in the Jurassic rocks ; while the same kind of column as
occurs in these genera existed also in the Carboniferous Platycrinus, and according to
Messrs. Wachsmuth and Springer^ "forms one of the most characteristic features
of the genus."

There is a considerable amount of variation amono; the diiferent members of the
Bourgueticrinidge in the characters of the terminal faces of the stem-joints. In the
Jurassic genus TliiolUericrinus, in Bourgueticrinus (Jurassic and Cretaceous), and in the
Cretaceous Mesocrinus the articular ridge is narrow and linear, expanding somewhat
around the opening of the central canal to form the real articular surface. ^

In all these genera a median groove extends along each half of the ridge, from the
central opening towards the margin of the joint face ; and short shallow branches proceed
from it on either side so as to cut out the upper portion of the ridge into a double row of
small teeth.

In TliiolUericrinus, Mesocrinus, and Bourgueticrinus ellipticus the ligament-fossse at
the sides of the articular ridge are either uniformly shallow throughout their whole extent,
or they are deepest in the immediate neighbourhood of the central canal. But they are
completely separated from one another by the articular ridge, which is continuous from
end to end of the elliptical surface. Very much the same is the case in the upper and
middle stem-joints of Bathycrinus (PI. Vila. figs. 8, 9), except that the articular ridge
is relatively larger and is destitute of teeth. But in the lowest stem -joints of this genus
(PL VII. figs. 12, 13; PL Vila. fig. 11), and in all parts of the stem of Rhizocrimis
(PL X. figs, 11-14), the articular surface is incomplete, and instead of surrounding the
central canal, is actually divided by it into two trihedral portions, the upper edges of
which are toothed just like the corresponding parts of the complete ridge in Bourgueti-
crinus ellipticus or Mesocrinus. The two ligament-fosspe communicate with one another
around the opening of the central canal, which thus appears to lie at the bottom of a
deep depression. Quenstedt ^ figures some stem-joints of this kind from the white chalk
of Eiigen under the name of Apiocrinus constrictxis.

'^ Eevision of the Palreocrinoidea, part ii., Proc. Acad. Nat. Sd. Philad., 1881, p. 69 (243).

2 In a stem-joint from the Msestricht Chalk, which is figured by Quenstedt as Apiocrinus {Bowguetierinus)
ellipticus (Encriniden, Tub. 104, fig. 70), there is no articular ridge at all, but merely an oval articular siirfoce
around the opening of the central canal. Unless this be the result of an accMental removal of the ends of the
articular ridge, it is a somewhat striking peculiarity which tends to approach the condition of the middle stem-
joints in Bathycrinus, and has a still closer resemblance to a form of articular siu'face which is especially characteristic
of the cirrus-joints {ante, pp. 7, 8).

3 Encriniden, Tab. 104, tigs. 64-66.

REPORT ON THE CRINOIDEA. 25

While the general characters of the stem are identical in RMzocrinus and Batliy-
crinus, there is a good deal of variation in its details, and especially in the mode of
growth.

In all cases the new joints are added at the top of the stem, immediately beneath
the cup ; but the rate at which they increase in length is very diflferent in the diflferent
species. It appears from Sars's figures,^ and from my own observations, that the pro-
duction of new joints in the stem of RMzocrinus lofoteiisis is slow compared to their
subsequent increase in length. For there are very rarely more than three joints beneath
the cup which arc wider than high, and even these have an appreciable thickness
(PI. IX. figs. 1, 2 ; PI. X. fig. 1). Sars remarks of the uppermost one that it is
"annulaire et a 2 ou 3 et meme souvent 5 ou 6 fois plus de largeur que de hauteur;"
while there are usually not more than eight cylindrical joints beneath it. Below this
limit the joints have the well-known dice-box shape, with the characteristic terminal
faces, the peculiarities of which begin to appear very few joints below the cup.

Most individuals of Rhizocrinus rawsoni seem to be generally similar to RMzocrinus
lofotensis in these characters (PI. IX. fig. 3 ; PI. LIII. figs. 7, 8). But in one example
I found five joints beneath the cup which were wider than high. The second and third
are mere circular disks with perfectly plain faces like those of the fourth (PL X. fig. 10) ;
and the faces of the newly formed joints of Rhizocrinus lofotensis which are figured by Sars ^
are of the same nature. But the uppermost joint of all is of a difierent character
altogether (PL X. fig. 9). It has a pentagonal outline, and its surface, which rises
gradually from the circumference towards the centre, is divided by five radiating ridges
into an equal number of trapezoidal fossae that receive the lower ends of the elongated
basals (PL X. figs. 3, 5). Here, therefore, we find the top stem-joint presenting the
same characters that it does in Apiocrinm ^ and Millerioinus,^ and entering to some
extent into the composition of the cup, while the new joints are probably intercalated
below it. Quenstedt^ speaks of this uppermost stem-joint in the Apiocrinida^ indif-
ferently as "Endstiick, Endsaulengiied, or Fiinfrippenglied." De LorioP has named it
" article basal ;" while Zittel ' speaks of it as the " Centro-dorsal," and remarks " Dasselbe
scheint, wie aus der Andeutung von Nahten hervorgeht, aus 5 ursprunghch getrennten
Stucken entstanden zu sein und entspricht wahrscheinlich den 5 Infrabasalplattchen
bei Encrinus." It is perhaps a little inexpedient to employ the term " centro-dorsal " for
a joint which bears no cirri, as its similarly named homologue does in the Comatulae.

1 Memoires pour servir h, la connaissance des Crinoides vivants. 1. Du Bhizocrinm lofotensis, tab. L, ii.

2 Op. dt., tab. ii., figs. 20-22.

3 D'Orbignj, Histoire Katurelle, generale et particuli^re des Crinoides vivans et fossiles., pL ii. fig. 3, pi. iii.
fig. 4, pi. V. fig. 4.

^ Ibid., pi. xiv. figs. 15, 23, 24 ; and Quart. Jouni. Geol. Soc, vol. xxxviiL p. 33, pi. l.

5 Encriniden, pp. 314, 315.

* Swiss Crinoids, p. 4 ; Paleont. Frang., loc. cit., p. 19.

' Handbuch der Pakeontologie. Palseozoologie, Bd. i., pp. 388-390.

(ZOOL. CHALL. EXP. PART XXXII. 1884.) •" *

26 THE VOYAGE OF H.M.S. CHALLENGER.

But I should prefer doing this to using a name, as de Loriol has done, which is so very
similar to that universally employed to designate some of the calyx plates, although
there is no sort of homology between the two structures. This latter point is recognised
by de Loriol, who makes it clear that he regards the " article basal " as a stem-joint.

Whatever may be the case in Apiocrinus, this plate is single in Rhizocrinus rawsoni
(PI. X. fig. 9), and I cannot help suspecting that Zittel has been misled by the appearance
of sutures into regarding it as probably consisting of five coalesced under-basals. I
have noted a somewhat similar condition in Millericrinus pratti}

Owing to the larger number of discoidal joints in the stem of Rhizocrinus rawsoni
than in that of Rhizocrinus lofotensis, the gradual development of their articular faces
is more easily made out in the former species. As the joints become longer than wide,
shallow fossaj appear to the right and left of the opening of the central canal, which
thus seems to be the deep middle portion of an oval depression occupying the shorter
axis of the elliptical face (PI. X. fig. 12). These fossae gradually increase in relative
size, and encroach more and more upon the original plane surface of the joint face,
stm, however, remaining in connection with one another around the central canal
(PI. X. figs. 11, 13, 14). They reach nearly to the margin of the joint face, so as
to leave a small articular rim outside them ; but they do not reach so far in the direction
of the long axis, at the ends of which the original surface of the joint remains to form
the so-called " articular ridge." This is, however, by no means continuous across the
opening of the central canal, as is implied by its name ; for it is merely represented by
two triangular surfaces which encroach upon the original oval depression so that it
assumes the form of a i-ather short-handled dumb-bell (PL X. figs. 11, 13, 14). The
two ends slope rapidly downwards towards the centre, where the opening of the axial
canal is situated. It thus establishes a connection between the two fossae and interrupts
the continuity of the articular ridge.

This ty[5e of joint-face also occurs in the lowest part of the stem of Bathycrinus
(PI. Vila. figs. 10, 11), but it is the result of a different mode of growth altogether.
There are several thin discoidal joints at the top of the stem (PI. VII. figs. 1-3, 11 ;
PL Villa, fig. 1), and the uppermost one, on which the basals rest, has its surface
marked by a ten-rayed depression which extends outwards from the five-lobed opening
of the central canal (PL Vila. fig. 3). The corresponding face of the basal ring is
marked in the same manner (figs. 13, 14); and the fossae lodge the five horse-shoe
shaped bgamentous bundles which unite the basals to the stem-joints below them.
On the upper face of the second joint, however, the opening of the central canal
is surrounded by a raised articular rim, stiU showing traces of fossse like those on the
top joint, and this rim is more marked on the next few joints (figs. 4, 5). As the
joints below become thicker and their terminal faces more oval, the articular rim also

1 Ou some new or little known Jurassic CrinoiJs, Quart. Journ. Gcol. Soc, vol. xxsviii. p. 34.

REPORT ON THE CRINOIDEA. 27

assumes an oval form, and the ten markings on its edge first become arranged into two
groups of five each (fig. 5), and then finally disappear (fig. 6). This is due to the five
ligamentous bundles mentioned above as beiag attached to the basals, becoming
gradually replaced by the two larger and somewhat crescentic bundles which unite
the joints lower down the stem. The articular rim eventually becomes restricted
to a broad ridge which occupies the short axis of the oval oblong face, and is pierced
in the centre by the oval opening of the central canal (figs. 7, 8). In the slender joints
of the upper third of the stem (fig. 7) this ridge takes up the greater part of the
articular surface, and there is merely a shallow fossa on either side of it. As the
joints increase in stoutness the ridge becomes relatively narrower, and the lateral fossae
proportionately larger (figs. 8, 9). They are shallowest at the edge, and gradually
deepen as they approach the ridge. When the joints begin to shorten again but
continue to increase in stoutness, their terminal faces become more circular (fig. 10),
though the planes of the articular ridges at the two ends of each joint stUl continue
to cross one another. The ridges themselves still diminish in relative width, and
become somewhat constricted in the middle, until there is only a very narrow rim
around the opening of the central canal (figs. 9, 10). Lower down the stem this rim
disappears altogether (PI. VII. figs. 12, 13; PL Vila. fig. 11), so that the two lateral
fossse communicate around the central opening just as in Rhizocrinus (PL X.
figs. 11-14).

The joint-faces also become oval again, and the articular ridges now occupy their
longer axis (PL Vila. fig. 11) instead of the shorter ones, as is the case at the top
of the stem (figs. 7-9). The two halves of each ridge which are separated by the
opening of the central canal are of an elongated triangular shape, and relatively narrower
than in Rhizocrinus raivsoni, owing to the greater size of the lateral fossse. Each is
denticulate along its median line, as in the other Bourgueticrinidse.

The distinction of the lower stem-joints of Rhizocrinus and Bathycrinus thus becomes
a matter of some difiiculty, though those of the upper and middle parts of the stem are
very different in their characters.

In the young individuals of both genera most of the stem-joints are simple, more
or less elongated cylinders ; and the characteristic dice-box shape is only visible in
a few joints immediately above the root (PL Villa, figs. 2, 3 ; PL LIII. figs. 7, 8 ;
woodcut, fig. 16, p. 244).

The lower part of the stem in the Bourgueticrinidse may bear a large number of
irregularly branched radicular cirri, two or more proceeding from each joint near the end
of the long axis of one face (PL IX. fig. 1 ; PL X. figs. 13, 15) ; whde the main axis may
eventually break up into a similar set of branching rootlets. These two forms of roots
may coexist in the same individual, or either may occur alone. In some specimens of
Rhizocrinus lofotensis the radicular cirri are exceedingly abundant. Thus in an indivi-

28 THE VOYAGE OF H.M.S. CHALLENGER.

dual fio-ured by Sars' they occur on the lowest thirty-three joints of a stem with fifty-
nine joints altogether. In another case five out of eighteen joints are cirriferous ; while
the individual represented in PL IX. fig. 1 has only nine cirriferous joints in a stem of
over forty. A similar variation occurs in Khizocrimis raivsoni. Every joint in the
lowest part of the stem may bear cirri at one or both ends of the long axis of its upper
face. But I have in no case found more than fifteen joints in this condition, and they
are sometimes not consecutive, a cirrus-less joint being occasionally interposed between
two others which bear cirri (PL X. fig. 15). On the other hand, in the only individual
with a complete stem which was obtained by the Challenger, and also in the young
specimens dredged by the " Porcupine," there are no radicular cirri at all, but only a
spreading root formed by subdivision of the main axis of the stem (PL LIII. fig. 7) ; and
this appears to be a constant condition in Bathycrimis (PL VII. figs. 1, 9 ; PL Villa,
fig. 3).

Below the last of the regular and dice-box shaped joints, which may or may not bear
cirri, there come one or more others of irregular shape and variable size. Spreading
rootlets proceed outwards from these, as a rule more abundantly in Rhizocrinus than in
Bathycrinus. In Rhizocrinus lofotensis this inferior joint usually bears several slender
root filaments disposed around a central one ; while one or two stronger and branching
rootlets sometimes come ofi" between it and the regular stem-joints. This is more
especially the case in Rhizocrinus rawsoni ; but in Bathycrinus the inferior joint, or " root-
joint " as it has been called, is quite short, and gives ofi" two or rarely three chief roots,
which themselves subdivide into smaller ones (PL VII. figs. 1, 9 ; PL Villa, fig. 3).

Both these rootlets of the stem-axis itself and the radicular cirri are composed of a
series of gradually diminishing joints closely united by ligaments. They attach them-
selves to foreign bodies by calcareous expansions round their ends or beneath the sides on
which they happen to rest (PL IX. fig. 1 ; PL X. fig. 15). Anything serves for this
purpose which may improve the anchorage of the Crinoid in the soft mud, which is nearly
universal at great depths, e.g., fragments of shell, grains of sand, sponge-spicules, foramin-
iferal tests, &c. Hence, whatever be the case in the Pentacrinidse, Rhizocrinus and
Bathycrinus must remain permanently fixed in one place throughout life.

In a specimen of Rhizocrinus raivsoni which was dredged by the " Travailleur," and
was described as a new genus Democrinus by Perrier,^ the diameter of the stem is lessened
at the origin of two groups of rootlets, and regains its former size lower down. Perrier
suggests the question " si la partie qui se prolonge au delk des racines n'est pas destinee a
devenir un second pedoncule surmonte d'un second calice. Si cette induction se verifie,
les Democrinus constitueront le premier exemple actuel d'Echiuodermes vivant en

• Op. cit., tab. i. fig. 1.

' Sur un nouveau Crmoide fix^, le Democrinus Parfaiti, provenant des dragages du " Travailleur," Comptes rendus,
t. scvi,, No. 7, pp. 450, 451:

EEPORT ON THE CRINOIDEA, 29

colonies et ramifies." I can, however, see no probability whatever in this supposition,
having met with no facts confirmatory of it either in any of the numerous roots of
Ehizocrinus and Bathycrinus which I have examined, or in the descriptions of these
genera by Sars and Danielssen. Neither is it supported in the slightest degree by what
we know of the embryology of the Crinoids.

Prof. Perrier has suggested some further views of the morphology of the stem which
I find myself unable to accept. He begins by saying " De tons les Crinoides fixes actuels,
les Democrinus sont ceux chez qui les dimensions trausversales du cahce sont le plus
foibles jiar rapport au diametre du pddoncule."

I am indebted to his kindness for the following measurements of his specimens —

Length of the calyx from the terminal furrow to the top stem-joint, . . 9 mm.

Maximum diameter of the calyx, . . . . . . . 2 „

Diameter of the stem-joints, . . . . . ■ . . 1 »

That is to say, the diameter of the calyx does not exceed twice that of the stem-joints.
But at the time Prof. Perrier made the statement quoted above he had before him the
following measurements of the calyx and stem in three varieties of Rliizocrinus rawsoni —

Basal tube. Stem-joints diameter.

1. Blake, . . . . 5-5 x 2-50 mm. 2-25 mm.

2. Challenger, . . . S'O x 2-00 „ 2-00 „

3. Porcupine, . . . 3'0 x 1-75 „ 1-25 „

Hence the maximum width of the calyx in the Challenger specimens of Rliizocrinus
rawsoni is the same as that of the largest stem-joints ; while the proportion is as 2 : 1 in
the so-called Democrinus, and this is not attained by the calyx, either of the " Blake " or
of the " Porcupine " specimens. It seems to me that the calyx of Democrinus is larger
relatively to the stem than in any of the Crinoids most nearly allied to it (instead of
being narrow as stated by Prof Perrier), who proceeds as follows : — "Si Ton songe que, chez
les Echinodermes libres actuels, le corps tout entier ne represente que le calice des
Crinoides fixes surmonte de ses bras, on est etonne de voir une partie qui est absolument
nuUe chez les representants des autres groupes prendre chez les Democi'inus un
developpement tel qu'elle represente cinq k six fois au moins le volume du corps propre-
ment dit. Ce fait seul nous avertit que le pedoncule doit Stre pris en grande consideration
pour la determination de la forme fondamentale des Echinodermes. Chez les Demo-
crinus, il produit un appareil radiculaii-e forme de rameaux articules ramifies ayant la
meme structure que lui-meme et presentant des dimensions superieures a celles des bras ;
cet appareil ne saurait etre davantage neglige au point de \mq morphologique, et Ton est
conduit t, considerer ses diverses branches comme ayant la meme valeur que le pedoncule
lui-meme dont elles ont la structure."

A far better instance than Democrinus of disproportion between stem and head is

30 THE VOYAGE OF H.M.S. CHALLENGER.

furnished by the Liassic Uxtracrinus, the stem of Extracrinus siibangularis reaching a
leno-th of 50 to 70 feet. But even as regards Democrinus I cannot admit that the
dimensions of the stem are so much greater than those of the arms. This may indeed be
the case in Perrier's three specimens, of which " deux sont totalement ddpourvus de bras ;
le troisieme n'en preseute c{ue des restes tres courts, d'apres lesquels il est aise de voir
que les bras devaient etre extremement peu developpes." But in the Carribbean examples
oi Rhizocrinus rawsoni the longest stem (180 mm.) contains sixty-eight joints above the
root, while there are five arms, each consisting of about eighty joints. Nearly half of
these bear pinnules, so that even if the radicular part of the stem is taken into account,
the superior dimensions would seem to be on the side of the arms rather than on that of
the stem, which Perrier considers to represent five or six times the volume of the calyx
and arms together.

It is likely enough that this may have been true in his three specimens of Demo-
crinus, which had lost the whole or greater part of their arms, owing to fracture at the
syzygies, as is only too often the case with both species of Rhizocrinus. But when a
tolerably perfect individual is obtained the arms are found to be considerably more than
" extremement peu developpes," as was so easily inferred by Perrier upon totally
insufiicient evidence. He goes on to say, " Alors meme cpi'ils ne vivraient pas en colonic,
le volume considerable de leurs racines ramifiees, la ressemblance de ces racines avec les
bras qui surmontent le calice et dont elles sont probablement homologues, sufiisent
a ddmontrer cpie la disposition arborescente des parties, preface en cpielque sorte de la
symmetric radiaire, n'est pas plus ^trangere au type des Echinodermes cp:'au type des
Coelenterds."

The relationship of the Echiuoderms to the Coelenterates need not be discussed here ;
but the resemblance and "probable homology" which Prof Perrier sees between the
arms and the root of a Crinoid appear to me to be forced in the extreme. The arms
are merely extensions of the body, containing the same nerves, vessels, and body-cavity
as are found in the calyx, together with the fully developed genital glands which are
usually sterile in the body. But the branches of the root have the same structure as
the stem, as remarked by Perrier himself ; and this is very difiereut from that of the
cup and arms. It is true that the rootlets, like the arms, are traversed by axial cords
which are connected with the fibrillar envelope of the chambered organ ; but there the
resemblance ends. They support no soft parts as the joints of the arms and pinnules do;
and being formed entirely on the right antimer are totally devoid of any of the ambu-
lacral structures which are so important in the morphology of the arms. If the term
" homologous " is to be employed for a mere superficial resemblance of this kind, a new
word must be introduced to denote community of origin and morphological similarity.
One might almost as reasonably say that the quills on the back of a porcupine are
homologous with its limbs.

REPORT ON THE CRINOIDEA, 31

Both in Bathycrinus and in Rhizocrinus the organic and calcareous networks, which
interpenetrate one another, are not always perfectly continuous throughout the
substance of the stem-joints. For both in optical and in transverse sections of de-
calcified stems empty spaces are often visible, especially in the immediate neighbourhood
of the central axis. Five of these spaces appear in some of my transverse sections of
the lower stem-joints of Bathycrinus (PI. Vila. fig. 2, rs). They are radially disposed,
being situated immediately outside the five peripheral vessels of the central axis {ch'),
which they may not greatly exceed in size. Both spaces and vessels may contain
larger or smaller masses of pigment granules [p). The walls of the former are less
well-defined than those of the latter, but are distinctly sharper than the outlines of
the spaces in the organic plexus forming the remainder of the stem-joint.

It sometimes happens in the lower part of the stem that one of these spaces may
increase very greatly in size, and so displace the central axis from its median position ;
or the space may approach quite near to the surface of the joint. They are much
more irregular in the lower part of the stem than in the uppermost elongated joints,
where they are also of larger relative size (PL Vila. fig. 1, rs) ; and they appear to
communicate with one another from joint to joint, through the elongated oval opening
of the central canal (PL Vila. fig. 7). But there is nothing to be seen of them in the
discoidal joints at the upper part of the stem, which are continuously traversed by
closely set ligamentous fibres (PL Vllb. fig. ] ). The best preparations that I have
been able to obtain illustrative of this point have been from the stem of Bathycrinus;
but I have also seen these spaces, though not so well, in Rhizocrinus. The former
genus (or at any rate Bathycrinus aldrichianus) is remarkable for the abundance of
delicate fibrils which may be seen proceeding outwards from the central fibrillar axis
of the stem, the nervous nature of which is gradually coming to be recognised. Some
of the larger of these fibrils are shown in optical section in PL Vila. fig. 1, ca'. But
a much larger number of smaller ones may be seen with a high power. They leave
the central axis in a more or less transverse direction, and form an open plexus, im-
mediately beneath the external surface of the stem. This is most distinctly seen over
the radial spaces, where the decalcified stem is, of course, more transparent than else-
where. It is noteworthy that the arms of Bathycrinus aldrichianus, like the stem, are
also remarkable for the large numl)er of branches which proceed outwards from their
axial cords (PL Vllb. figs. 6, 7 ; PL Villa, figs. 4, 5 — a'). The corresponding branches
in the stem of Pentacrinus have been akeady noticed {a^ite, p. 23).

C. Hyocrikid^.

The stem of Hyocrinus is one of considerable interest, because it is the only recent
Crinoid in which the terminal faces of the stem-joints are of the same nature as those of

32 THE VOYAGE OF H.M.S. CHALLENGER.

the Apiocrinidte and of so many of the Palfeocrinoids. The applied surfaces of the cylin-
diical joints, forming the lowest portion of the stem -with which we are acquainted, were
described by Sir Wyville Thomson ^ as " being marked with a pattern of radiating grooves
and rido-es." There appears, however, to be a good deal of variation in this respect ; for,
while some of the joint-faces have the radiating pattern very well developed, others are
perfectly plain (PL Vc. fig. 4), and others have only slight indications of the striation.
But the radiating stripe never " resemble minute pores jienetrating the walls," as stated
by Wachsmuth and Springer," of the similarly marked joint-faces in the PaliBoerinoids.

The stem of Hyocrinus is much more rigid than that of the BourgueticrinidsB. The
short cylindrical joints are united by uniformly disposed ligaments (PI. Vc. fig. 5, Is), the
fibres of which are all of equal length and not longest in the centre as in the oldest parts
of the stem of Bathycrinus and Rhizocrinus. The ligamentous fibres at each end of the
joint extend into its substance for about one-fifth of its length, so that the calcareous
tissue is closer towards the ends than in the median parts of the joints. These contain
radial spaces (PI. Vc. fig. 5, rs) of the same nature as those just described in the
Bourgueticrinidse (PL Vila. figs. 1, 2, rs).

So far as I can make out from the only two fragments of stem which have reached
me, measuring 70 and 85 mm. respectively, there are a large number of discoidal joints
at the top of the stem (PL VL figs. 1-3). Theii- thickness gradually increases from
above downwards, until they are about half as long again as wide. The length then
diminishes again and the width increases, rising in one example from 1 mm. to nearly
1'5 mm. -ttithin twelve joints. The lower joints thus become more discoidal again, like
those some little way below the calyx. What they were in the stem-fragment, 170 mm.
long, which is mentioned by Sir Wyville Thomson, I have no means of knowing.
Neither are we acquainted with the nature of the actual base of attachment.

' Journ. Linn. Soc. Loud. (Zool.), vol. xiii. p. 52, 1878. ^ Revision, part i. p. 14.

REPORT ON THE CRINOIDEA. 33

III.— THE CALYX.

In all the recent Crinoids, with the single exception of Tliaumatocrinus (PI. LVI.
figs. 1-4), the calyx proper is formed of but two series of plates, viz., the basals, which rest
upon the top stem-joints, and the radials, which are supported by and alternate with the
basals. The plates of these two series are suturally united to their fellows and to one
another ; and they enclose a central cavity which is sometimes large enough to contain
the whole visceral mass, as in Holopus (Pis. I.-IV.) and Hyocrinus (PI. VI.). In
other cases, however, the internal cavity of the calyx is reduced to a minimum, and it
lodges nothing but the chambered organ with the plexiform gland which rises from it ;
while the entire visceral mass is situated altogether above it, and is very easily detached
from it, as in many Comatulfe, so that there can hardly be said to be any cup at all.

Intermediate conditions between these two extremes are met with in the Bouro-ueti-
crinidse and Pentacrinidse. In the last named family the cup which is formed by the
united upper surfaces of the radials is wider than in Antedon, but somewhat more
hollowed than in Actinometra (PL XII. figs. 1,2, 15, 16 ; PL XVIII. figs. 4, 5 ; PL XX.
figs. 5, 8; PL XXVI. fig. 11 ; PI. XXX. figs. 3, 4; PL XXXIII. fig. 5; PL XXXIV.
figs. 7, 8 ; PL L. figs. 5, 6) ; and the lowest portion of the visceral mass rests in this
concave upper surface of the radial pentagon ; while the greater part of it lies in the cup
which is formed by the outer radials and lowest distichals (PL XVII. fig. 5 ; PL L. fig. 1 ;
PL LXIL). But the central funnel of the calyx, which is between the inner faces of
the radials, only contains the plexiform gland ascending from tlie chambered organ
(PL LVIII. fig. 3, x).

This is also the case in Rhizocrinus and Bathycrinus (PL Vllb. figs. 1, 4, 5, a;). In
the former genus the upper surface of the calyx has a deep hollow which lodges a portion
of the intestinal coil (PL X. figs. 1, 4, 6, 7, 8); while the upper part of the visceral
mass is entirely supported by the first four brachials in two syzygial pairs (PL X.
figs. 2, 20). In Bathycrinus, however, the gut hardly descends to the level of the first
radials ; and though the visceral mass is very largely supported by the large wing-like
processes of the axiUaries, it is practically quite free from the lowest brachials (PL VII.
fig. 3; PL Vllb. figs. 1, 7, 8).

Concealed in the lower portion of the calyx is the chambered organ, the position of
which, relatively to the calyx-plates, varies considerably. In aU the Comatulse it is lodged
within the cavity of the centro-dorsal, and it is therefore entirely on the dorsal side both
of the radials and of the basals, whether the latter have been metamorphosed into a
rosette or not. But in the stalked Crinoids there is no enlargement of the central
canal of the stem within its uppermost joint, and the vascular axis passes up into the
calyx before expanding to form the chambered organ. In Rhizocrinus, Bathycrinus, and

(ZOOL. CHALL. EXP. — PART XXXII. — 1884.) li 5

34 THE VOYAGE OF H.M.S. CHALLENGER.

Pentacrinus this expansion takes place near the top of, but entirely within the basal ring,
the raclials having no share in the protection of the chambered organ, though they
surround the lowest portion of the plexiform gland which rises out of it (PI. Vllb.
figs. 1, 4, 5; PI. Villa, figs G, 7; PL XXIV. figs. 7-9; PL LVIII. figs. 1-3;
PL LXIL). Within this central funnel of the radials, and closely enveloping the
2)lcxiform gland, is a kind of plug formed by numerous irregular limestone bars which are
dcve]o])ed from the inner faces of the radials. In Bathycrinus it is practically little
more than calcified connective tissue (PL Vllb. figs. 1, 4, 5, c), Init it becomes very solid
in Rhizocnnus, and has been wrongly described as a basal rosette (PL Vila. fig. 7 ;
PL X. figs. 1, 4, hr.). It is also well developed in Pentacrinus {2\. XXIV. figs. 8, 9, rp.),
and is lodged in a small depression upon the upjjcr surfixce of the basal ring, which is
formed by the truncation of the inner ends of its component pieces (PL XX. figs. 2-G, 9).
Among the Apiocrinida^ the cavity which lodged the chambered organ is bounded in
varying proportions by the basals and radials. In some species, such as Millericrinus
milleri, this structure must have lain altogether upon the ventral aspect of the basals.
These form a complete ring, just as in the aberrant Comatula;, Atelecrinus and
lliaumatocrimis ; but in both these types the basal ring is on the ventral side of the
chambered organ, which is precisely the opposite condition to that of Millericrimts
milleri

A. The Basals.

The basals of the Neocrinoidea vary considerably in the extent to ^\'llich they are
developed. In all tlic Pala3ocrinoids the radials are separated from the top stem-joint by
one complete ring of plates, to which a second is often added. But in tlie Neocrinoids
no basals may be visible at all upon the exterior of the calyx, as in most Comatula? ; or
there may be a single complete ring of high plates as in Rldzocrinus (PL IX. figs. 1-3 ;
PL X. figs. 2, 3 ; Pi. LIII. figs. 7, 8) ; or there may be two rows of plates of variable
size as in Encrinus, Extracrinus, and Marsupites.

The absence of external basals in most recent Comatula3 is due to their having
undergone metamorphosis into the well-known rosette, which is concealed between the
centro-dorsal and the radials. But the occurrence of this condition in a stalked Crinoid
would appear somewhat improbable. Personal examination luis convinced me that in
two cases * at any rate the supposed absence of basals in fossil Pentacrinidse is merely
the result of defective observations ; but this may not be invariably true. No basals are
visible externally in the Jurassic Isocnnus pendxdus, Meyer,^ nor in the Pentacnnus
pentagonalis personatus from the Brown Jura, which is figured by Quenstedt' without

' These are (1) Pcnfacji'jiws /i.s7i.«ri, Forbes, from the Kimmeridge Chiy of Weymouth ; aiul (2) a fine speeinieu
from the Chalk, which is figured in Dixon's Geology of Sussex (1S78 edition, pi. xix. 22).

' Isocrinua und Chelot-rinus, Museum Senckenbergiaiiuni, Frankfurt, 1637, Taf. xvi. figs. 1, 2.
" Eueriiiiden, Tab. 98, iig. 137.

REPORT ON THE CRINOIDEA 35

any notice of its peculiarities. They likewise appear to be absent in the Forest-Marble
specimen from Farley in Wiltshire, which was described l)y Goldfuss as Pentacrinus
scnlaris ; ^ while they are certainly absent externally in one of the two known specimens of
Metacrinus costatus, though fully developed in the other (PI. XLIX. figs. 1, 2). This is
a most curious anomaly ; but as the specimen cannot be sacrificed to investigation, it is
impossible to ascertain whether the basals are absent entirely, or whether they have been
metamorphosed into a rosette.

It is possible that they arc so greatly reduced in size as to fail to appear externally,
as occasionally happens in Encrinus and in the fossil Comatulaj, which retain their
embryonic basals in an unmctamorphosed condition.^ They are sometimes quite small
and insitrnificant in comparison with the radials, as in the Liassic Pentacrinus tuherculatus,
and in some varieties of the recent Pentacrinus decorus (Pis. XXXIV.-XXXVL). In
these and similar forms they appear at the lower angles of the calyx as minute rounded
plates, between which the lower edges of the radials rest directly upon the top stem-
joints. The basals are therefore only in contact with one another by their inner ends
(PI. XXXIV. fig. 8). But in other species, both recent and fossil, they are considerably
larger, and their outer ends separate the radials more completely from the top stem -joint
(PI. XIII. fig. 1 ; PI. XV. fig. 2) ; while the union of their inner ends is more extensive
PI. XII. fig. IG ; PI. XXVI. fig. 11). In fact all degrees of union may be traced (both
in different species and in different individuals of the same species) from the condition of
Pentacrinus hlahei a.m\ Pentacrinus decorus (Pis. XXXI., XXXV.) to that oi Pe^itacrinus
tvyvllle-thomsoni and Pentacrinus maclearanus, in which the radials are separated from
the top stem-joint by a ring of large and closely-united basals (Pis, XVI., XIX. figs. 1,
G, 7). The genus Cainocrinus of Edward Forbes has been lately revived by de Loriol^
for a few fossil species which possess a closed basal ring, but are not otherwise different
from Pentacrinus. The condition of the recent Pcntacrinida), however, is such as to entirely
preclude the possibility of employing this very variable character as a generic distinction.

A similar series of gradations is to be met with among the fossil Comatulfe, in
which group there appears to be much more individual variation than among the
PentacrinidjB. In some few species no basals are visible externally at all. In others,
the outer ends of small prismatic rods may appear at some angles of the calyx but
not at others, while their inner ends do not meet at all or only very slightly so. In
some species again, the outer ends of the basal rods are smaller than the inner ends,
which meet together and entirely separate the median portion of the radial pentagon
from the centro-dorsal beneath. Lastly, in a chalk Comatula mentioned by Schluter*

1 Petrefacta Germaaije, vol. i. pi. Ix. fig. 10.

2 On the genus Solanoorinus, Goldfuss, and its Relations to recent Comatulse, Journ Linn. Soc. Land. ZooL, vol.
XV. pp. 211, 212.

^ Swiss Fossil Crinoids, pp. Ill, 112.
Ueber einige astylide Crinoiden, Zeitschr. d. deutsch. geol. Gesellsck, 1878, p. 66.

36 THE VOYAGE OF H.M.S. CHALLENGER.

there is a complete ring of united basals as in the recent genera Atelecrinus and
Thaumatocrinus (PL LVI. figs. 1-4). In the four other genera of recent Comatulidse,
however, the primary embryonic basals undergo transformation into the well-known
rosette, which is really a secondary structure. It lies over the chambered organ, between
the under surface of the radial pentagon and the^ upper face of the centro-dorsal ; but
it is entirely concealed, and does not appear at all upon the exterior of the calyx.

In many Comatula^ more or less prismatic rods proceed outwards from the inter-
radial angles of the rosette, and their ends are often visible on the exterior of the
calyx. They occupy exactly the same positions as the rod-like basals of many fossil
Comatulse, but do not represent them morphologically, for they are not developed from
the primary embryonic basals. These last become transformed into the rosette, which
is a secondary structure as shown by Dr. Carpenter ; ' while the tertiary basals which
are connected with its interradial angles are merely ossifications in the connective
tissue of the synostosis between the radials and the centro-dorsal, and are somewhat
variable in their development.^ They do not occur in Antedon rosacea, nor in the
species on both sides of the Atlantic which are most nearly allied to it, viz., Antedon
phakmgium, Antedon dentata, Antedon hageni, &c. ; and they seem to be absent in
Eiidiocrinus, at any rate in Eudiocrinus semper i. But they reach a relatively large
size in many tropical species both of Antedon and of Actinometra, in which latter
genus I have never found them to be absent, and they also occur in Promachocrinus
(see fig. 1, A on p. 37). [See Note B.]

B. The Eadials.

There is considerable variation in the degree of lateral union between the individual
basals and radials of stalked Crinoids. Those of Holopus (Pis. I. -IV.) are so intimately
fused that the sutures are entirely invisible on the exterior of the tubular calyx. In
Hyocrinus (PI. VI.) the interradial sutures are quite evident, but those between the
basals are very obscure. In Rhizocrinus, and especially in Rhizocrinus Iqfotensis, the
sutures between the basals are usually entirely invisible, not only externally but also
in transverse sections of the decalcified calyx ; while the basiradial and interradial
sutures are merely indicated by very faint lines on the surface of the cup (PI. IX.
figs. 1,2); and strong measures are necessary before the individual joints will se23arate
from one another. In the Pentacrinidae and Comatulidas the union between the radials
is less close and the sutures well defined, while the condition of the basals varies accord-
ing to circumstances (PI. XIII. fig. 1 ; PL XV. figs. 1, 2 ; PL XVI. ; PL XX. ; PL XXXIII.
figs. 8-10 ; Pis. XXXV.-XXXVII. ; PI XXXIX. ; PL XLIIL). The union of the

' Phil. Trans., 1866, pp. 744, 745.

'' The genus Actinometra, Trans. Linn. Soc. Lond. (Zool.), vol. ii. pp. 93-105 ; and the genus Solanocriuus, Journ.
Linn. Soc. Lond. (Zool.), vol. xv., 1880, pp. 212-214.

REPORT ON THE CRINOIDEA.

37

basals of Bathycrimis is extremely close, and in old individuals no sutures are visible
externally (PI. VII. figs. 1, 2, 11 ; PI. Vila. figs. 12-14), though they appear in
transverse section (PI. Vllb. fig. 2) and also in young examples. I have never suc-
ceeded, however, in separating the plates from one another by the usual methods.

The radials of Bathycrimis, on the other hand, are much less closely united. The}'
are thin plates in contact with one another by quite narrow sides (PL VII. figs. G, 6a).
Those of Bathycrimis aldrichianus were described l^y Sir Wyville Thomson as being
" often free ; but in old examples they also are frequently anchylosed into a funnel-shaped
piece."* All the specimens which I have examined arc in the latter condition, though
the plates are readily separable. But I do not think it possible that they could ever
be perfectly free as the other two radials are; and I have always found them to be
closely united by ligaments up to the level of the circular commissure (PI. Vllb. fig. 4, /),
though they become much more free near the top of the calyx (PI. Vllb. fig. 5). The

Fig. l.—Promachocrinus kcrguelensis. Calyx, x 6.

A. Side view, sliowing the alternation of the five primary radials with the five others, which are separated from the ceutro-

dorsal by the rays of the basal star. B. Upper view, showing the interior of the central funnel formed by the radials.

difference between the " free " and the " anchylosed " conditions is probably only due to
variations in the extent to which limestone is deposited around the fibres of the above
mentioned interradial ligaments.

The difference between the basal and radial rings in the amount of lateral union
{i.e., in the distinctness of the sutures) between their component joints, which is more
or less evident in Hyocrinus, Bathycrimis, and Rhizocrinus, appears also in some fossil
Crinoids. Beyrich has pointed out ^ that in young individuals of Encrimis gracilis the
sutures between the basals are invisible, although those between the radials are distinct
enough ; and the same character has been noticed by Mr. E. Etheridge, jun., and myself
as occurring in the Palaeozoic Allagecrimis atistini.^

' Journ. Linn. Soc. Land. (Zool)., vol. xvi. p. 50.
■■^ Ueber die Crinoiden des Muschelkalks, Berlin, 1857, p. 44.

■' On Allagecrinu.s, the representative of a new family from the Carboniferous Limestone Series of Scotland, Ann.
and Mag. Nat. Hist, ser. 5, 1881, vol. vii. pp. 285-288.

38 THE VOYAGE OF H.M.S. CHALLENGER'.

The second brachials of Rhizocrinus (PI. Villa, fig. 8, B^, and the second radials of
Baihycrinus (PI. Vllb. fig. 6, R^ are in contact by their lower edges, but soon become
entirely free from one another (PI. VII.; PI. VIII. figs. 1, 2; PI. IX. figs. 1-3; PL X.
fio-s. 1, 2, 6-8) ; while the corresponding plates in Hyocrinus (PI. VI.) are absolutely
free. But in many Crinoids the second radials are often very closely united by
lio-amentous bundles. These are lodged in fossre at the sides of the proximal face
which is not quite so wide as the distal one (PI. XII. figs. 9, 10 ; PI. XXI. figs. 5a, 5c).
The first joints beyond every axillary are more or less closely united in the same way
(PI. XXI. figs. 36, 46). But the second joints and the axillaries themselves are free,
though often in very close contact with their neighbours, so that their apposed sides are
more or less flattened (PI. XXI. figs, lb, 2b) ; and in some cases the first four or five
of the free brachials have their sides flattened in this way, where they come in contact
with one another and with the corresponding joints of adjacent rays (PI. XV. fig. 2 ;
PI. XVI. fig. 1 ; PI. XXXa. fig. 8).

In all the Pentacrinida3 there are invariably five rays. I have never met with any
exception to this rule, and all the specimens of Bathycrinus that I have seen conform to
it also. The original specimen of Holopus is tetramerous ; while I have seen four
Comatulse with the same peculiarity, and one with six rays. Four and six rays are more
common in Rhizocrinus (PI. Villa, figs. 6, 7), and in very rare cases there may be seven.
In the Comatulid genus Promachocrhms, however, ten is the normal number, but the
basals are pentamerous. Only five rays extend outwards from the central rosette to
appear externally beneath five of the radials (fig. 1 , a), and they must therefore be regarded
as representing the primary interradii of the type. Hence those radial pieces which are
not separated from the centro-dorsal by basal rays are the original embryonic radials,
homologous with those of other Crinoids and of the five-rayed Star-fishes. The five
others may perhaps be compared to the additional radials developed in many-armed
Star-fishes, in which, however, the positions of the five primary rays are not indicated in
the adult as they are in Promachoc7inus.

C. The Intereadials.

In almost all adult Neocrinoids the first radials meet one another all round the calyx
so as to form a complete ring ; and until lately this character has been regarded as one
specially distinctive of the group. For in a large number of PaljBocrinoids an anal jjlate
retains its primitive embryonic position and rests upon a basal, thus separating two of
the radials and destroying the complete pentamerous symmetry of the calyx. A good
instance of this type is the Carboniferous genus Belemnocrinus, which has a calyx very
similar to that of the recent Rhizocrinus except for the presence of the intercalated anal

EEPOKT ON THE CRINOIDEA. 39

plate. Hexacrhms, Dichocrinus, and their allies present a similar condition. In a few
genera of the Rliodocrinidse, which have been grouped together into a section Ehodocrinites
by Wachsmuth and Springer, the radials are not contiguous laterally ; but between every
two there is an interradial plate which rests on a basal below.

This character, which occurs in no stalked Neocrinoid, either recent or fossil, reappears
in the singular Comatulid Thaumatocrinus^ (PI. LVI. figs. 1-4). But with this excep-
tion all the primary radials of every adult Neocrinoid, recent or fossil, stalked or free,
form a complete ring.

Calyx-interradials are very usually present in the Palseocrinoids, helping, together with
the hio-her orders of radials, to increase the size of the cup, and strengthen its walls.

According to Wachsmuth and Springer^ "The first interradial is always larger than
any of the rest, and is situated between the upper sloping margins of the adjoining first
radials, except in some species of the Rhodocrinidse, in which it rests directly upon the
basals, separating the ring completely. There are generally two plates in the second
series, and two or three in each succeeding one."

In the Mesozoic genus Guettardicrinus, and in some species of Apiocrinus [Ajnocrinns
martini, Apiocrimts roissyanus), there are calyx-interradials essentially similar to those
of the Palseocrinoids. Each series commences with a single plate resting upon the upper
angles of two first radials which are truncated for its reception (see fig. 9, on p. 183).
It is followed by several others, more or less irregularly arranged ; and these, together
with the two outer radials, and sometimes also the two lower brachials, form the im-
movable wall of a large cup just as in the Palseocrinoids. No recent Crinoid presents
this condition, at any rate in the adult state ; though it occurs in many Ophiurids, as
pointed out elsewhere.^ But in all the Pentacrinidse, recent and fossil,* the interradials,
if present, are not calyx-plates at all, but merely small and more or less irregular
plates developed in the perisome which unites the rays and their subdivisions (PI. XIII.
fig. 1 ; PI. XXXI. fig. 2 ; PL XXXIV. fig. 1 ; PI. XXXV. fig. 2 ; PI. XXXVII. fig. 1;
PI. L. fig. 1). It would seem, however, that regular calyx-interradials may appear in
the early larval stages of Antedon rosacea. At any rate this is the way in which I
should interpret the following statement by Sir Wyville Thomson.' "In one or two
cases, however, I have observed about the time of the first appearance of the anal plate,
a series of five minute rounded plates developed interradially between the lower edges of
the oral plates and the upper edges of the basals." These plates therefore, separate
the radials from one another all round the calyx. Their ultimate fate is uncertain. Sir

1 On a new Crinoid from tlie Southern Sea, Phil. Trans., 1883, pp. 919-926, pi. 71.
^ Revision, part ii. p. 12.

3 On the Apical System of Ophiurids, Quart. Journ. Micr. Sci., vol. xxiv., X.S., January 1884, p. 12.

4 See, for example, Quenstedt's Encriniden, Tab. 101, figs. 23, 39a; Austin's Crinoidea, pi. xiii. fig. Ic ; and
Buckland's Geology (Bridgewater Treatise), vol. ii., pi. liii- tig- 2.

' Phil. Trans., 1865, p. 540.

40 THE VOYAGE OF H.M.S. CHALLENGER.

Wyville made no further reference to them except to say that they sometimes remained
permanently in the adult Antedon, usually in groups of three or five. These last,
however, like those already noticed in Pentacrinus (PI. XIII. fig. 1), are merely periso-
matic plates developed in the tissue uniting the second radials. Good figures of them
were given by Dr. Carpenter ;* and it is by no means certain that they are a further
develoj^ment of the primitively single plates which appear between the orals and basals,
for the direct continuity of the two structures has never been definitely traced.

I am inclined to believe that where they do appear in ordinary Comatulse these
primary calyx-interradials eventually undergo resorption like the orals and the anal
plate. But they are permanent in Thaumatocrinus (PI. LVI. figs 1-4), as are also
the orals (fig. 5). This remarkable genus has five calyx-interradials which rest on
basals and separate the radials just as in certain Ojahiurids and in some types of the
Palaeozoic Ehodocrinidse. It is much to be regretted that this extraordinarily interesting
form is only represented by a single individual. For the study of the distribution of
the axial cords within the calyx would have been of some importance.

If Thaumatocrinus resembles the ordinary Comatulse and Pentacrinidse, the circular
commissure (PI. XXIV. fig. 9, c.co., i.co.) is formed by both interradial and intra-
radial commissures which connect the paired branches of the five jjrimary interradial
cords (PL XXIV. figs. 7-9 ; PI. LVIII. figs. 1-3 — ar) ; and the interradial commissures
must traverse, or at any rate, lie upon the inner surfaces of the interradial plates. But
the general embryonic characters of Thaumatocrinus lead me to think that the arrange-
ment of its axial cords must be more like that which occurs in BathycHnus. In this
genus the primary cords do not fork within the basal ring, but pass upwards through it
and enter more or less complete canals which are formed by the apposition of two
grooves, one on each of the contiguous lateral faces of adjacent radials (PL VII. fig. 6a).
When they reach about half the height of the radials they fork, and the resulting
branches themselves form the interradial commissures, entering the radials by the
apertures in their lateral faces.

I cannot help suspecting that the same condition may occur in Thaumatocrinus, i.e.,
that the primary interradial cords run upwards through the basals into the interradials,
and there fork, one branch entering each of the two radials which are separated by the
interradial lodging the primary cord. This simple condition would correspond very
well with the general embryological characters of the type, as revealed in other ways.
But owing to the want of material there is unfortunately but little chance of the above
hypothesis being verified or disproved.

While resembling the Rhodocrinites in having five large plates separating the radials,
Thaunmtocrinus differs from most Pateocrinoids, with the exception of the Platycrinida?,
in the absence of any higher series of interradial plates. Except on the anal side these

' Phil. Trans., 1866, pi. xxxiii. fig. 7 a, b.

REFORT ON THE CRINOIDEA. 41

primary interradials of Thaumatocrimis all end simply in a free rounded edge at the
margin of the disk (PL LVI. figs. 1-3, 5), which is doubtless partly due to the simplicity
of the arms. For these become free at once, and are not connected laterally by perisome,
in which higher orders of radials could be supported. The interradial of the anal side,
however, bears a small tapering appendage of four or five gradually decreasing joints,
which terminates in a blunt point without any connection whatever with the anal tul)c
near it (PI. LVI. figs. 2, 4, 5). It appears to me to be of the same nature as the so-called
proboscis of Taxocrinus, Gnorimocrinus, Onyclwciimis, &c. The anal plates of these
genera do not support a huge " ventral sac," such as occurs in the Cyathocrinida), but are
of an altogether different nature. Good figures of them are given by Schultze,' Augelin,''
and by Meek and Worthen.^ They may be advantageously compared with figs. 2 and
4 on PI. LVI.

According to Wachsmuth and Springer* the first anal plate of Taxocrinus " has a
truncated upper side, and is succeeded by from two to six similar, narrow, quadrangular
plates, longitudinally arranged. The plates diminish in size upwards, and form the dorsal
side of a short and slender lateral proboscis, whose ventral parts, as well as the wall
supporting them, have never been found preserved, and evidently consisted of more fragile
material." A few pages farther on they describe Onychocrinus as follows : — " In the anal
area there is a series of from three to five very narrow, quadrangular plates, which rests
upon the truncated or slightly excavated upper side of the basal, and forms a small
lateral proboscis as in Taxocrinus. Interradials three to twenty, perhaps more in some
species ; the first one large, resting between the first and second radials, the succeeding
ones smaller, rapidly decreasing in size and thickness upward, and having an inward
curvature. They are followed by very minute irregular polygonal plates, which form
the interradial portion of the vault." Meek and Worthen ^ described this anal series
as resting upon the larger truncated basal, " much as the arms of Platycrinus rest upon
the first radials, and really looking very much like a diminutive arm rising from the anal
area. This arm-like range of small pieces seems never to consist of more than from four
to six or seven pieces, which are so small and narrow as to leave a wide open space
between them and the posterior rays on each side." Subsequently, however, they met
with a specimen showing " the space between the little arm-like range of anal pieces, and
the radials and vault to be occupied by very numerous minute pieces."^ These last occur
in each interradius, and are directly continuous with those forming the so-called " vault "
or ventral disk, just in the same way as the perisomic plates between the rays of recent
Crinoids (PI. XIII. fig. 1 ; PI. XXXIV. figs. 1,2; PI. L. figs. 1, 2), of Extracrinus, and

1 Monographie der Echmodernien des Eifler Kalkes, Denkschr. d. k. Akad. d. IFiss. IFiai., Bd. xxvi., 1866, Taf. iv.
tigs. 2, 26, 3, 46.

2 Iconographia Crinoideorum, Stockholm, 1878, Tab. xvii. fig. 8 ; Tab. xx. figs. 9, 13, 16 ; Tab. xxiii. fig. 5.

3 Paleontology of Illinois, vol. v. pi. xiv. fig. 4.

4 Revision, part ii. p. 46. « PaL-eontology of Illinois, vol. ii. p. 243. « Ibid., vol. iii. ]). 494.

(^OOL. CIIALL. EXP. — P.^RT XSXII. — 1884.) Ii (J

42 THE VOYAGE OF H.M.S. CHALLENGER.

of Apiocrinus, are continued upwards on to the surface of the disk between the
ambuhxcra. Wachsmuth and Springer described this ventral disk of Onychocrinus and of
other Ichthyocrinidas as " composed of a more or less soft or scaly integument, }delding
to motion in the body and arms ; " ^ and they regard it as homologous with the more
solid vault of Platycrinus and Actinocrinus. I believe, however, as I shaU explain
more fully further on, that this was the real ventral surface of the Crinoid and not
a " vault " at all ; while the so-called proboscis or anal tube with the small interradials
round it is just in the condition which the anal appendage of Thaumatocrinus would
assume, did it exist in a larger Crinoid such as Pentacrinus asteria (PL XIII. fig. l),
with a well-plated perisome between the rays. This plating may be continued up on to
the disk and to the summit of the anal tube (PI. VI. fig. 4 ; PL XVII. fig. 6 ; PL XXVI.
figs. 1,2; PL XXXIV. fig. 2 ; PL XXXIX. fig. 2 ; PL LV.).

It unites the lower arm divisions closely together ; and any additional appendage
in the anal interradius would naturally be bound in with it, just as the four to seven joints
of the anal appendage in Onychocrinus are bound in with the numerous minute pieces
between the rays. But I see no reason for supposing that such an appendage would
form part of the tube up to its opening, and be in any way grooved on its inner
side. For it seems to taper away rapidly and to become merged into the general
plating of the anal interradius in the flexible vault, or disk as I should call it. The
passages quoted above both from Meek and Worthen, and from Wachsmuth and
Springer, would admit of this interpretation ; and in the first case at any rate, it seems
(from the context) to be the one which was intended. But Wachsmuth and Springer
also speak of the anus of the Ichthyocrinidas as " unknown except in Taxocrinus and
Onychocrinus, which have a small lateral tube." '^ This observation refers to the small
appendage already mentioned ; but it must not be understood to imply (as it well
might) that this appendage is hollow and pierced by the rectum.

Some older Crinoids, however, than Taxocrinus and Onychocrinus seem to
have had an anal appendage like that of Thatimatocrinus, which was sometimes
surrounded by numerous minute interradial pieces, so as to form a support to the anal
side of the disk between the rays. I mean the genera Hetcrocrinus, Hall ; ReLcocrinus,
Billings ; and Xenocrinus, Miller, all from the Lower Silurian of America.^

According to Wachsmuth and Springer's definition of Reteocrinus, the posterior

1 Revision, part i. p. 31. ^ Revision, part i. p. 31.

3 As regards the second of these, I shall speak of it in the sense in which it is used by Wachsmuth and Springer,
Wetherby, and others (Revision, part ii. p. 191 ; Amer. Journ. Sci. and Arts, April 1883, p. 256). I should say, however,
that S. A. Miller differs from his fellow-workers in America upon this subject, and refers the species gTouped under
Keteocriwus hy Wachsmuth and Springer to, at least, three genera {ATner. Journ. Sci. and ^rts, August 1883, p. 105 ; and
Journ. Cine. Soc. Nat. Hist, vol. vi., December 1883, pp. 217-230). As, however, all the species referred by them to this
genus appear to possess an anal appendage like that of Thaumatocrinus, Onychocrinus, and Taxocrinus, it is obviously
more convenient to consider them all as congeneric, as I am also inclined to do for other reasons (see Phil. Trans.,
1883, pp. 923-933).

PtErOET ON THE CEINOIDEA. 43

interraclial area is wider than the other four " with a conspicuous row of decidedly
larger and more prominent pieces along the median part," Billings ^ gave a good
fio-ure of this in Reteocnnus stellaris, and spoke of it as follows : — " If this series of
joints constitute a true arm there must be six arms in this species." Miller, who has
examined the original specimen of Rcteocrinus stellaris, describes it thus — "Azygous
interraclial area covered by a large number of plates, probably one hundred or more,
very unequal in size, the middle row being decidedly larger and more prominent than
the others, so as to form a ridge up the middle. The plates in this row, however,
do not rapidly diminish in size and fade out in their distinctive character before reaching
the top of the vault ; on the contrary, they are longer than the primary radials, four of
them reach nearly as high as the secondary radials, and while the specimen is not
preserved above this, enough is disclosed to the palseoutologist to show that this series
continued up the face of a proboscis that extended, may be as far, or farther, than the
arms and the pinnules."

In default, however, of further evidence I prefer to believe that the middle row of
plates in the anal area of Reteocrinus stellaris was of the same nature as, though perhaps
on a larger scale than, that of Reteocrinus nealli, which Miller describes as follows : —
" Azygous interradial area covered by fifty or sixty plates, very unequal in size, the
middle row being decidedly larger and more prominent than the others, so as to form a
ridge up the middle, while the other smaller and less prominent ones are crowded in,
irregularly, on each side. The plates in this middle row, however, have no uniformity in
size or shape ; the first one is large and elongated, the fourth is small and subquadrate ;
and the row has become almost obsolete at the sixth plate, where all are nearly of the
same size and scarcely distinguishable from the minute pieces which cover the flattened
vault, and with which they unite." The figures of this type in the Palaeontology of
Ohio (vol. i. pi. ii. figs. 3b, 3c) illustrate this description admirably, the original
specimens having doubtless been seen by Miller; while the figure of Reteocrinus
subglohosus on the same plate (fig. 2c) shows the incorporation into the body of a
pinnule borne by one of the secondary radials. This pinnule is closely surrounded by
the minute interradial plates, but may be distinguished from them at its origin just as
the anal appendage is. This condition is still better shown va. Reteocrinus richardsoni,
Wetherby, which has two " fixed pinnules " in the anal interradius, one on either side of
the median appendage. All three are " soldered " together by the minute irregular plates
which pass insensibly upwards into those of the so-called "vault;" and the ordinary
pinnules on the lower parts of the arms after the last axillary are united in just the same
way.' This condition recurs constantly in the Liassic Extracrinus and in the recent

1 Decades of the Geological Survey of Canaila, vol. iv. p. 64, pi. ix. fig. 4a.

2 Descriptions of New CrinoiJs from the Cincinnati Group of the Lower SUiirian and the Subcarboniferous of
Kentuek}-, Journ. Cine. Soc. Nat. Hist., vol. ii. pi. x^•i. ligs. 1, la.

44 THE VOYAGE OF H.M.S. CHALLENGER.

Pentacrinidse and Comatulse ; and I see no reason to believe that the minute interradials
of Reteocrinus are in any way different from those of the Neocrinoids. But I regard
them as perisomic phxtes continuous with those of the disk above, which was in no sense
a " vault " like that of the Actinocrinidse. Meek's figure of Reteocrinus nealli '
illustrates this point admirably, and after examining disks like those of Pentacrinus
ivyvillc-thomsoni, Pentacrinus alternicirrus, Pentacrinus naresianus, and Pentacrinus
mollis, together with similar disks in numerous Comatulas (PI. XVII. fig. 6 ; PI. XXVI.
figs. 1, 2; PL XXX. fig. 2; PI. XXXIII. fig. 7; PL LIV. fig. 10; PL LV.), and also
(thanks to the kindness of Mr. AVachsmuth) that of Reteocrinus nealli, I find it difiicult
to believe that the so-called vault of Reteocrinus was anything but the true oral surface
of the animal.

Miller's genus Xenocrinus " is in this respect essentially similar to Reteocrinus.
" The azygos area is remarkably large and covered in the central part by a vertical series
of plates having about the same size as the regular radial series; and upon each side of the
vertical series there is a depressed area covered by small plates having a tubercle in the
central part, as in the regular interradial areas. There are seven plates, each having a
length about twice as great as its width, in the vertical series, from the basal plate, upon
which the series rests to the top of the vault. This vertical series is continued to the
top of the proboscis, and contains in its entire length more than fourteen plates. It has
such strong resemblance to the radial series, except as to the branching at the secondary
radials, that the general appearance of the body is that of a species having six radial
series." Miller figures the specimen with fourteen plates in the vertical series, and
remarks that we learn from it that "the proboscis extends as high as, and probably
beyond the extremity of the arms." He also says that the small plates between the rays
and their subdivisions " continue over the margin of the vault, and undoubtedly cover it,
and also more or less of the long proboscis." I do not see, however, that this so-called
but unknown proboscis is anything more than an anal tube covered by j^erisomic plates,
us in Extracrinus and so many other Neocrinoids. I also doubt whether Miller is right in
stating that the vertical series is continued to the top of the proboscis, for (to judge from
his figures) he does not seem ever to have met with a specimen perfect enough to show
the top as it is shown in Meek's figure of Reteocrinus nealli.'^ But I think it quite
possible that, considering the size of this vertical series, it may have become free at the
top of the calyx as the anal appendage of Thaicmatocrinus does (PL LVI. figs. 4, 5),
instead of tapering away quickly and ending on the lower part of the anal tube as in
Reteocrinus nealli.

The ventral sac or proboscis of Cyatliocrinus is usually much larger and more

' Palseoutology of Ohio, vol. i. pi. ii. tig. 3c.

'^ Description of some New and Remarkable Crinoids and other Fossils of the Hudson River Group, Journ. Cine. Soc.
Nat. Hist, vol. iv. pp. 72, 73, pi. i. fig. 3 ; pi. iv. fig. 6. ' Palaeontology of Ohio, vol. i., pi. ii. tig. 3c.

REPORT ON THE CRINOIDEA. 45

complex in nature than that of Reteocrinus and Xenocnmis. But a transition between
the two appears to be presented by some forms of Dendrocrinus, Heterocrinus, and
Tocrinus. A little specimen figured by Meek, and referred to the aberrant type
Dendrocrinus casei, shows the anal side very well.^ Meek's description runs as follows :
— " Anal series with the first piece resting directly upon the upper truncated side of the
heptagonal posterior subradial {i.e., basal) hexagonal in form, and bearing in direct
succession above a series of hexagonal pieces gradually diminishing in size ; while
alternating with these similar small hexagonal pieces can be seen on each side of the
mesial series, for some distance above the body between the free rays, and connecting
with those of the ventral part." His figure is a curious one, and does not quite agree
with his description; for there seems to be a single large and pentagonal anal plate which
separates two radials and rests in the angle formed by the upper edges of two basals
(subradials, Meek). Upon this plate rests a series of seven gradually diminishing
hexagonal pieces which stand out prominently from the smaller plates at their sides,
just like the middle row of plates in the anal area of Reteocrinus necdli with which
they seem to be comparable. If they supported a ventral sac like that of the typical
Dendrocrinus, it was relatively much larger than that of Reteocrinus necdli, so that
the vertical series of plates would end much farther from its summit than in that
species.

Thus then in Onychocrinus, Taxocrimis, Reteocrinus, Xenocrinus, and even in
Dendrocrinus casei the anal side shows this regular vertical series of plates which rests
on a basal below and gradually diminishes in size. The only essential difference between
it and the anal appendage of TJmumatocrinus is that it forms part of the body, being
bound in with the rays by minute interradial plates which are not present in the simpler
Thaumatocrinus. But this is often the fate of the lower pinnules in the Neocrinoids ;
and it would assuredly also be the fate of an anal appendage in a Crinoid with the same
calyx-characters as Thaumatocrinus, but standing in the same relation to it as an
extensively plated and multiradiate Comatula does to the naked and ten-armed Antedon
rosacea.

In the Cyathocrinoid genus Heterocrinus there appears to have been an anal append-
age like that of Onychocrinus and Reteocrinus ; but it rested on the upper sloping sides
of two adjacent radials instead of on a basal.

In this ty^e, as in the Cyathocrinidse generally, the capacity of the cup is compara-
tively small, and the visceral cavity within the disk is almost entirely limited to its anal
interradius, which is enormously enlarged, and forms the structure known as the " ventral
sac." In Cyathocrinus itself this is a heavily plated tube, that commences at the upper
edge of the " special anal " plate, above which its characteristic porous structure appears
at once. But in Heterocrinus the ventral sac appears to be less robust, while the anal

• Paleontology of Ohio, vol. i. pi. iii. his. fig. 2c, p. 29.

46 THE VOYAGE OF H.M.S. CHALLENGER.

series consists of " a single row of plates, longitudinally arranged, the outer side rounded
and forming a prominent ridge, which gives the appearance of an arm." ^

The physiological condition of this type appears to me to be very similar to that of
Reteocrinus nealli, i.e., the anal interradius is supported by a special row of plates, the
lowest of which are more or less incorporated into the calyx, while the upper ones are
closely surrounded by perisomic plates.

In Poteriocrinus and Eupachycrinus this anal appendage is practically reduced to one
plate, the so-called third anal plate, or the first plate of the ventral tube ; but it rests on
the first anal or azygous plate, which is itself supported by a basal. This azygous plate
is absent in Ceriocrinus, and the first plate of the tube (anal appendage) comes dowai to
rest directly upon a basal.

Another curious modification is presented by locrinus. The posterior side of the
calj'x is occupied by a large plate that rests between two basals and is in line with the
radials. It has been variously described as a radial and as an azygous piece.^ On its
upper surface is a triangular plate which " supports on its right sloping side the usual
number of brachials, and on the left a row of quadi'angular plates, vertically arranged,
extending to the tips of the arms, and forming the posterior wall of a large ventral tube.
In external appearance these plates resemble the brachials and arm plates, only they are
somewhat higher and not quite as wide ; they are gibbous and form an elevated ridge,
which causes this appendage to resemble an arm or a branch of the ray." This median
ridge extends to the full length of the ventral sac, and it is bordered, just as is the case
in Reteocrinus nealli, liy a number of more delicate perisomic plates. I have little
doubt that it served the same purpose in both cases, supporting the anal interradius,
though in no way specially grooved for the reception of the hind gut. It is well figured
by Meek in the Palaeontology of Ohio, vol. i. pi. i. fig. 9b.

1 Revision, part L p. 69.

- For further information upon this subject, see Wachsrauth and Springer, Revision, part i. pp. 65, 71 ; and Ariie7:
Jovrn. Sci. and Arts, voL xxvi. 1883, pp. 370, 376 ; also P. H. Carpenter, Quart. Journ. Geol. Soc, vol. xx.wiii., 1832,
lip. 306, 307.

REPORT ON THE CRINOIDEA. 47

IV.— THE RAYS.

A. The Eay-divisions and Arms.

The arms of a NeocrinoicI, viewed in a strictly morphological aspect, must be regarded
as commencing with the first joints beyond the primary radials. The same is the case in
many Palseocrinoids, of which Schultze says, " Die Arme (brachia) beginnen unveritn-
derlich da, wo eine deutliche Gelenkfacette eines festen Kelchstiickes ihren Ursprung
anzeigt."^ In the Platycriuidas and other Palaeocrinoids, and in all Neocrinoids (except-
ing perhaps Guettardicrinus), this articular face is on the first radial. In the five-armed
Eudiocrinus indivisus the next joints beyond the radials are syzygial, with pinnules on
the epizygals,^ which clearly shows that they must be considered as arm-joints and not as
belonging to the calyx, although they undoubtedly represent the so-called second and
third radials of a ten-armed Crinoid. The other species of Eudiocrinus have these two
primitively separate joints not united by syzygy but articulated, just as in Hicmmatocrinus
(PI. LVI. figs. 1-4). The second one bears a pinnule both in Thamnatocrinus and in
Eudiocrinus varians ; but in Eudiocrinus semper i n,ni\. Eudiocrinus jajwnictis the &Tst
pinnule is on the fourth joint after the radial. This would correspond to the second
brachial of a ten-armed Crinoid, but it is really the fourth brachial in Eudiocnnus.
Lastly, in Perrier's Eudiocrinus atlanticus ^ the first pinnule is on the fifth brachial, which
corresponds to the third brachial of an Antedon.

The well-known genus Rhizocrinus resembles Eudiocrinus indivisus in the syzygial
union of the first two joints beyond the primary radials (PL X. figs. 1, 2, 6-8, 20).
They have generally been called the second and third radials ; and there is some ground
for this in the case of Rhizocrinus lofotensis, as they are considerably broader than aU the
joints which follow them except the first (PI. IX. figs. 1,2). But in Rhizocrinus rawsoni
(PI. IX. fig. 3 ; PI. LIII. fig. 7) they are not much larger than the four following joints, which
contribute with them to support the visceral mass (PI. X. fig. 20) ; while the first pinnule
is on the last of these, i.e., on the sixth joint above the cal}-x (PI. IX. fig. 3 ; PI. X. fig. 20).
Considering the evidence afi'orded by Eudiocrinus, I think, however, that it will be more
consistent to describe Rhizocrinus as having only one radial; while the first pinnule
would then be on the sixth {Rhizocrinus raivsoni) or on the eighth brachial {Rhizocrinus
lofotensis). The terms second and third radials would then be used only in those cases
where there are ten or more arms, owing to the thii-d radials and more or fewer of the

1 Op. cit, p. 5. See also the genus Actinometra, Trans. Linn. Soc. Land. (Zool.), ser. 2, vol. ii. pp. 20-25, 1883 ; Zittel's
Palc-eontologie, t. i. p. 339; de Loriol, Paleont. Frang. Terr. Jurass., t. xi. p. 15; Wacbsmuth, Revision, part ii. pp. 9, Irt.

2 Eudiocrinus and Atelecrinus, Journ. Linn. Soc. (Zool.), vol. xvi. p. 495, 1882.

3 Sur des Eudiocrinus de I'Atlantique et sur la nature de la faune des grandes prol'ondeurs, Covvjtes Rendus, t
xcvi. No. 11, p. 726.

48 THE VOYAGE OF H.M.S. CHALLENGER.

following joints being axillaries. This is in accordance with the nomenclature employed
by Zittel, who speaks of the joint in Cui^ressocrinus, which is caUed " articulare " by
Schultze, and "second radial" by Eoemer, as a "first brachial ;" while he onlj' describes one
series of radials in the five-armed Pisocrinus.^ The developmental history of the plates
also indicates clearly that the second and following radials are really ann-joints. For
they commence as imperfect rings, which soon become filled up with lengthening
fasciculated tissue, just as is the case with the stem-joints and later brachials. But the
first radials, like the basals and orals, commence as expanded cribriform films ; while the
endogenous additions by which they are subsequently thickened are cribriform like those
of the basals, and not fasciculated like those of the two outer radials and the following
arm-joints. Messrs. Wachsmuth and Springer' have been led by their study of the
Palseocrinoids to the same conclusion, i.e., that " the arms fundamentally commence with
the second radials ;" although they find in practice that for purposes of description " it is
more convenient to regard the arms as commencing with the first free plate beyond the
calyx." In very many Neocrinoids with ten or more arms this would be the second
radial ; and in the multiradiate Metacrinus (PL XXXVIIL; PI. XLIII. fig. 2 ; PI. XLVL;
Pis. XLVIII.-LII.) this is actually a syzygial joint with a pinnule on the epizygal just
as in the simpler Eudiocrinus indivisus, but an axillary appears a few joints farther on,
and the rays begin to divide. In the other Pentacrinidse, however, in Bathycrinus,
Iloloims, and in most Comatulse, as well as in the fossil Encrimis and Apiocrinidse, the
second joints above the primary radials are axillaries, and it is not tiU the second (or
rarely the first) joints beyond these that pinnules appear. In all these t}^es the axillary
and the joint immediately below it are of the same width as the primary radials in the
calyx. But in Marsupites and in many Palseocrinoids [Platycrinus, Cyathocriniis, &c.)
they are very much smaller than the primary radials, just as the homologous joints are
in Hyocrinus (PL VI.).

The primary radials which form the upper part of the calyx are generally distin-
guished as i)\Q first radials ; while the following joints, as far as the first axillary inclusive,
are called the second, third radials, &c., though they are really only arm -joints as is shown
by their bearing pinnules in Metacrinus (PI. XII. figs. 6, 8 ; PL XXXVIIL ; PL XXXIX.
fig. 1 ; PL XLIII. fig. 2 ; PL XLV. fig. 1 ; PL XLVL ; PL XLVIII. fig. 1 ; PL XLIX.
figs. 1, 2 ; PL L. figs. 1, 8, 10, 14, 16 ; PL LI. fig. 1 ; PL LII. fig. 1). Since, too, it is
very convenient for descriptive purposes to use different names for the diflferent regions
of the arms, I see no reason for altering the names by which these plates are generally
known, provided that their real nature is not lost sight of.

The conventional use of the term " radials " for the joints between the calyx and the

1 PalcBontologie, pp. 348, 349.

2 Phil. Tram., 1865, p. 541, pi. xxvii. figs. 1, 3 ; Ibid., 1866, pp. 729, 742, pi. sli. fig. 1.

3 Kevision, part ii. p. 10.

REPORT ON THE CRINOIDEA. 49

first axillary (inclusive) is particularly advisable in the case of the Neocrinoids. For
among all the members of this sub-class which have ten or more arms/ Metacrinus is the
only genus besides Plicatocrinus in which the second joint beyond the primary radials is
not an axillary, a character which has elsewhere been pointed out as distinguishing the
Neocrinoids from the Palaeocrinoids.^ In the various tyj^es of Comatulce and in some
species of Pentacrinus there is a similar constancy in the number of joints which intervene
between the successive axillaries of the dividing arms. I have therefore found it
convenient ' to give special names to the joints composing the primary and secondary
arms respectively, and to restrict the term " brachials " to the joints composing those
portions of the arms which undergo no further division.

The joints of the primary arms may be called " distichals," a term no longer used with
the precise meaning which Miiller attributed to it;* while the joints of the secondary
arms (if there be any) may be termed "palmars." These names are of much use in
descriptions of Comatulaj ; for in this family the number and character of the segments
between the successive divisions of the arms exhibit variations which are, to a great
extent, constant in different species, and thus give us the means of classifying them into
larger or smaller groups.^

The Pentacrinidse, however, exhibit a much greater irregularity in this respect ; and
they also present more exceptions to the following rule, which holds good in almost all
the Neocrinoids. The first two joints beyond every axillary of the dividing rays are
united to one another in the same manner, either syzygy or bifascial articulation, as the
second and third radials are. Thus, for example, there is a syzygy between the two outer
radials of Encrinus, and another between the two lowest brachials. In Aiyiocrimis and
MiUericrimis the corresponding joints are respectively united by bifascial articulations.

This rule holds good in all the ten-armed Comatulse, whether the joints are articulated
{Antedon rosacea) or united by syzygies {Actinometra Solaris) ; and it is equally true in
all the many-armed species with the exception of two groups of Actinometrse, together with
a few unusually aberrant types. In one group, which is represented by Actinometra.
multiradiata, the two outer radials and also the first two of the three distichals are
articulated by ligaments only; but in all the subsequent arm divisions there is a
muscular joint between the first two segments after each axillary, and the second one is
traversed by a syzygy, whether it be itself an axiUary or a free brachial, while the first
bears a pinnule. Another variation occurs in Actinometra typica, Actinometra novcB-
guinece, and their allies. These forms have three distichals in the primary arms, the
first two of which are articulated, while the axillary is a syzygy ; but the two outer

> This passage does not refer to Promachocrinus, in wliich genus there are ten primary raxjs.

2 On Allageorinus, &c., Ann. and Maij. Nnt. Hist., ser. 5, vol. vii., 1881, p. 296.

3 Actinometra, Trans. Linn. Soc. Land., (ZooL), ser. 2, vol. ii. p. 24.
* Bail des Pentacrinus, p. 31.

^ Classification of Comatiihe, Proc. Zoul. Soc. Land., December 1882, pp. 746, 747.

(ZOOL. CilALL. EXP. — PART XXXII. — 1884.) Ii 7

50 THE VOYAGE OF H.M.S. CHALLENGER.

radials are united by syzygy, as are tlie first two joints beyond tlie distichal and all
subsequent axillaries. Except in these and a few other cases, however, there is a very
great uniformity throughout the arm-divisions of the Comatulse.

In five of the eight recent species of Pentacrinus the two outer radials form a
syzygy, and in correspondence with this the lowest distichals and brachials are similarly
united in paii-s (PI. XII. figs. 18, 21 ; PI. XV. figs. 1, 2; PL XVI. fig. 1 ; PI. XVIII.
figs. 1-3, 8, 11 ; PL XIX. figs. 1, 6, 7 ; PL XXI. figs. Id, 2d, 5a ; PL XXV.; PL XXVI.
figs. 4, 5, 8). On the other hand, the ten-armed Pentaa-inus naresiamis has a bifascial
articulation between the two outer radials, and also between the two lowest brachials,
just as in Antedon rosacea (PL XXX. figs. 1, 11, 12, 16, 17). But in Pentacrinus
decorus and Pentacrinus hIaJcei the rays divide twice or thrice ; and though the two
first joints beyond the lowest axillaries resemble the outer radials in being articulated by
ligaments, yet there is a muscular joint between the two lower brachials of the ultimate
arms, the second of which is usually a syzygy (PL XXXI. figs. 1, 2 ; PL XXXII.
figs. 16-18 ; PL XXXIV. figs. 3, 6 ; Pis. XXXV.-XXXVIL).

The syzygial union of two arm-joints is of a somewhat peculiar character. For
the hypozygal entirely loses its individuality as a separate segment of the arm, and
bears no pinnule as the epizygal and the remaining brachials do (PL XII. fig. 9;
PL XV. fig. 3; PL XXX. figs. 1, 19, 20, 23; PL XXXa. figs. 10«, 106, 12a, 126;
PL XXXII. fig. 4 ; PL L. figs. 6-16). Thus, for example, in very nearly aU Comatulse
the original third and fourth joints of the growing arm differ from those which ultimately
appear beyond them. For " whilst the majority of these gradually come to possess the
true articulations, and to be separated by the intervention of muscles and ligaments, a
certain small proportion become more intimately united on a simpler plan, which admits
of no motion between them." ^ The double or syzygial joints thus formed resemble the
ordinary brachials in bearing but one pinnule, and they are therefore best considered as
single joints. In Antedon rosacea, for example, the third and fourth, the ninth and
tenth, and the fourteenth and fifteenth joints of the growing arm are respectively united
in pairs by syzygy ; but the arm is best described as having syzygies in the third, eighth,
and twelfth joints. So again in the numerous Comatulaj, such as Actinometrcc parvicirra,
which have axUlaries on some or all of the primary arms. Counted from the third radial,
the distichal axillary is primitively the fomth joint. The first, as is almost invariably
the case, bears no pinnule, while the second has a pinnule, but the thuxl not, for it is
united to the following (axillary) joint by a syzygy. The first ray-division would therefore
be described as consisting of three distichal joints, the second bearing a pinnule, and the
third (axillary) a syzygy.

The same arrangement occurs in the genus Metacrinus, which is distinguished from
Pentacrinus and from all other Neocrinoids by having, not three radials only, but

1 Phil. Tram., 18C6, p. 721.

REPORT ON THE CRINOIDEA. 51

primitively five or eight. Two instances in which Pentacrinus varies in the direction of
Mctacrinus have come under my notice. One is in Pentacrinus mulleri (PL XV. fig. 2).
The second and third radials are articulated, but the latter is an axillary with a syzygy,
so that there are primitively four radials. In the other case {Pentacrinus decorus) there
are seven primitive joints in the ray. The first two above the primary radials are
united by a bifascial articulation, while the axUlary is a syzygy. I have elsewhere de-
scribed two specimens of Millericrinus pratti in which there are four radials, in one case
on two out of the five rays (Quart. Jouru. Geol. Soc, vol. xxxviii. p. 35, pi. i. fig. 23).
Similar variations occur among the Comatulse. In one Antedon that I have examined,
one of the rays consists of five joints, the axillary being a syzygy ; while in individuals
of two other species, the axillary rests directly upon its first radial, the second radial
having remained undeveloped. When there are five radials in Metacrinus, as in Meta-
crinus angidatus, the third and fourth bear pinnules ; but the second does not, for it is
united by syzygy to the third, and has lost its individuality as a separate joint (PI. XII.
figs. 5-10 ; PI. XXXTX. fig. 1). The radials of Metacrinus angulatus, therefore, are
practically four in number, the second of which is a syzj'^gy and bears a pinnule like the
third.

In Metacrimts nodosus, on the other hand, there are primitively eight radials, but
besides the syzygy between the second and third, there is another between the sixth and
seventh; so that there are really only six joints, all of which except the first and last
(axillary) bear pinnules, while the second and fourth have syzygies, and are <as much
single arm-joints as the third brachial or any other syzygial joint in the arm of Antedon
rosacea (PI. L. figs. 1, 6-16 ; PI. LI. fig. 1).

In Encrinus, Extracrinas, and in most recent species of Pentacrinus, as also in a few
Comatulge (Actinometra Solaris, Actinometra typica, &c.), the two outer radials and the
first two joints beyond them are respectively united by syzygy. On the principle ex-
plained above, each pair would therefore be considered as forming a single joint, so that
the true third brachial (itself a syzygial pair) would come to be the second. This would
involve our describing these forms as having but two radials, the axillary with a syzygy,
and syzygies both in the first and in the second brachials. I think, however, that this
would be misleading, and make the difi"erence between this type and that of Antedon
rosacea and Pentacrinus naresianus appear much greater than it really is.

The presence of three radials is such an absolutely constant character in all the five-
rayed Neocrinoids excepting Metacrinus and Plicatocrinus,^ that the fact of the outer
ones being united by syzygy and not articulated seems to me to be of minor importance;
and I do not assign to it the same morphological value as the syzygial union of the third
and fourth primitive brachials, in which the former loses its pinnule. No Crinoid with

1 Zittel has described a six-rayed example of this genus, in which the first joint above the cup (called by him the
first brachial) is axillary as in many Pateocrmoids [Sitzmujsh. d. II. CI. k. baier. Akad. d. TViss., 1882, Bd. i. p. 105).

52 THE VOYAGE OF H.M.S. CHALLENGER.

three radials ever has a pinnule on the second one; and when this becomes the hypozygal
of a syzygy, it does not therefore lose its individuality, as is the case with the hypozygals
of the ordinary brachial syzygies. Almost the same may be said respecting the first two
brachials. Most Comatulse, like Pentacrimis naresianus (PL XXXa. figs. 1, 10a, 106,
12a, 126), have a syzygy in the thiixl brachial with a bifascial articulation between the
two preceding joints, of which only the second bears a pinnule. Hence, when these
two are united by syzygy, as in Actinometra Solaris, Actinometra typica, &c., the lowest
or hypozygal loses no individuality as an arm-joint. They are, therefore, better
described as the first and second brachials, and not as a first brachial which "is a syzygy."
This method has the advantage of retaining the third brachial as a syzygial joint as a
condition which is common to by far the larger number of Comatulse. For it is only in
a very few species like Actinometra Jimbriata and Actinometra multirad lata that there is
a syzygy in the second brachial and a pinnule on the first, as is often the case in
Metacrinus. This is an entirely different type, and arises from the coalescence of the
primitive second and third joints of the growing arm.

Syzygial unions of two primitively separate arm-joints occur with great regularity
throughout the arms of the Comatulse. In the two principal genera Antedon and
Actinometra, there are large groups of species typified by Antedon eschrichti and
Actinometra parvicirra respectively, in which syzygies occur at tolerably regular inter-
vals of three joints. It is rare, however, to find a perfectly regular arm, especially in.
the latter species, in which the "syzygial interval" may vary from 0 to 10 joints.^ In
other species the interval may be as much as twenty joints or more ; while it is occasion-
ally two, as in Antedon rosacea, and in rare cases one joint only, as in Rhizocrinus. But
it is generally possible to find a considerable amount of regularity in the number of joints
which form the syzygial interval in any given species, and this is often of some value for
systematic purposes.

Among the Pentacrinidge, however, this is only the case to a very slight extent. The
syzygial interval is perhaps most regular in Pentacrinus naresianus (PI. XXVIII.) ; but
it is long as in many tropical Comatula^, and in other Pentacrinidce the brachial syzygies
are usually " few and far between."

In Rhizocrinus and Hyocrinus, on the other hand, the syzygial union of the primitive
brachials is carried on to a very great extent. In the former genus syzygial and muscular
iinions alternate with one another continuously from the calyx to the arm-cuds (PL IX.;
PL X. fig. 20 ; PL LIII. fig. 7). In Hyocrinus (PL VI. figs. 1, 2), as was well described
by Sir Wyville Thomson,^ the five arms "consist of long cylindrical joints deeply grooved
within, and intersected by syzygial junctions. The first three joints in each arm consist
each of two parts separated by a syzygy ; the third joint bears at its distal end an
articulating facet from which a pinnule springs. The fourth arm-joint is intersected by

1 Actinometra, loc. cit, p. 49. 2 Journ. Linn. Soc. Lond. (Zool.), vol. xiii. p. 52.

REPORT ON THE CRINOIDEA.

53

two syzygies, and thus consists of three parts, and so do all the succeeding joints ; and
each joint gives off a pinnule from its distal end, the pinnules arising from either side of
the arm alternately." In this type, therefore, two-thirds of the arm-joints lose their
individuality altogether. They bear no pinnules and take no part in the movements of
the arms. In Rliizocrinxis half the bi-achials are in the same condition ; while more than
half are devoid of pinnules, as the lowest pinnule-bearing joint is the sixth or sometimes
even the eighth primitive brachial (PI. IX.).

It is worth notice that the modes of arrangement of the arm-joints which are
characteristic of Hyocrinus and Rhizocrinus respectively, are precisely paralleled by the
condition of certain species of the Palaeozoic Heterocriniis. Thus in Heterocrinus con-
strictus, Hall, the pinnules are borne alternately on opposite sides of the arm by every
third joint ; and I have little doubt, from the figures of the arms which are given both by
HalP and by Meek,^ that each group of three joints is intersected by two syzygies just
as in Hyocrinus (PL VI. fig. 1).

On the other hand, the alternation of syzygies and muscular joints, which is so
characteristic of Rhizocrinus, also occurs in Heterocrinus simplex ; and Meek's figures ^
show that the opposed syzygial surfaces were striated as in Apiocrinus and Comatula, and
not plain as in Pentacrinus and Rhizocrinus.

It has been pointed out already * that the suj^posed syzygies in the arms of Bathy-
crinus (PI. VII. fig. 2 ; PL VIII. figs. 1, 2 ; PL Villa, fig. 1) are really articulations of a
peculiar type, though the fossae and vertical ridge are barely visible in the outer parts of
the arms, and would probably have escaped notice altogether, but for the very marked
differences from ordinary syzygial surfaces which are presented by the apposed faces of
the two outer radials, or of two of the paired lower brachials (PL Vila. figs. 16, 19, 20, 22).
Nevertheless, the proximal joint of a pair so united resembles the hypozygal of a syzygy
in the non-development of its pinnule ; and it might therefore be urged that every pair
so united should be properly considered as a single joint, just as in the case of a syzygial
pair which only bears a pinnule on the epizygal. It must be remembered, however, that
the syzygial union is an immovable one, which is far from being the case with any arti-
culation, whether bifascial or trifascial ; and the reasons given above for retaining the
individuality of the two outer radials and of the first two joints beyond any axillary, even
when they are united by syzygy, apply equally well in the case of Bathycrinus. For the
hypozygal joints of syzygial pairs are not the only ones which never bear pinnules. The
lower joint of every pair forming a bifascial articulation is distinguished in the same way,
e.g., the first joints of the various arm-divisions in most Comatulse, and the first brachials
of Pentacrinus naresianus (PL XXXa. fig. 12&). The same is also true in the many-
armed Pentacrinidae, when there are many joints in an arm-division and the axillary is

1 Twenty-fourth Annual Report on the New York State Museum of Natural History, Albany, 1372, pi. v. fiys. 13, 1-1.

2 Palceontology of Ohio, vol. i. pi. i. fig. 10. ^ Ihid; pi. i. fig. 7. * Ante, pp. 8, 9.

54 THE VOYAGE OF H.M.S. CHALLENGER.

united to the preceding joint by a bifascial articulation instead of by syzygy. In fact it
is a general rule in all Crinoids that pinnules are only borne by those joints which are
united to their successors by paired muscular bundles.

The hypozygal in the brachial syzygies may be fairly considered as losing its indi-
viduality. Not only does it bear no pinnule, but it takes no part in the movements of
the arm. But when two joints are united by ligamentous bundles on either side of a
vertical ridge, they are able to share in the lateral movements of the arm, though not in
those of flexion and extension ; and it therefore seems unreasonable to consider a pair so
united as equivalent to one joint only.

Sir Wyville Thomson was accustomed to regard the stem and its appendages as
constituting the " vegetative system " of the Crinoid, as distinguished from the more
strictly animal portions, viz., the cup and arms. In describing Rhizocrinus for example,^
he specially alluded to the great " preponderance in bulk of the vegetative over the more
specially animal parts of the organism;" and he subsequently pointed out that in
Hyocrinus and Bathycrimts,- as in Rhizocrinus, there is "a comparatively excessive
development of the vegetative system." This was generally the case throughout the
Bourgueticrinidse and Apiocrinidse, none of which have any very great number of arm-
joints, though the "body" may be considerably enlarged with the help of the upper
part of the stem. Thus, for example, d'Orbigny^ describes two twenty-armed species
of Millericrinus, each reaching a total length of one metre, out of which the calyx and
arms together only take up 86 and 94 millimetres respectively, less than one-tenth of
the whole ; while in one ten-armed species the calyx and arms together only measure
29 out of 920 millimetres.

Among the Pateocrinoids there is considerable variation in the relative development
of the stem as compared with the body and arms. The latter are often absent altogether,
as in the Blastoids* and many Cystids ; while they are few in number and poorly
developed in Haplocrinus, Pisocrimis, Symhathocrinus, &c. On the other hand, the
body and arms, so enormously developed in Croialocrinus, are quite extensive in
many Cyathocrinidse and Actinocrinidse ; but the stem is often large and complicated
at the same time, as in Barycrinus and Megistocrinus.^

In the Liassic Extracrinidaj the stem, immensely developed as it may be, still falls
considerably short of the body and arms in the complication of its structure. Extracnnus
hriareus has a comparatively short stem ; but in Extracrinus subangularis it may
exceed 50 or even 70 feet," with but few cirri except near the calyx, and those

1 " Porciipine " Crinoids, Proc. Hoij. F!nc. Edin., vol. vii. p. 771.

2 Journ. Linn. Soc. Land. (Zool.), vol. xiii. p. 48.
2 Hist. Kat. des Crinoides, pp. 39, 41, 44.

* The so-called " pinnules " of tlie Blastoids cannot Ije properl)' compared to those of the Crinoids, for they do not
seem to have contained the genital glands.

^ Eevision of the PaL-eocrinoidea, vol. i. pp. 14, 15.
<^ Encriniden, pp. 271, 291.

REPORT ON THE CRINOIDEA. 55

only small ones. The arms, however, reach an enormous development, and Quenstedt
calculates the total " Krone " to contain not less than five million pieces.'

In the recent PentacrinidaD the arms are generally well developed in proportion to
the stem. This proportion is of course least in young individuals, as shown in
Pis. XXXV. and LI., while it is greatest in forms like Pentacrinus mulleri, Pentacriniis
maclearanus, Pentacrinus tvyville-fhomnoni, and Pentacrinus alternicirrus (Pis. XIV.,
XVI., XIX., XXV.). These lead a semi-free existence, owing to the fracture of the stem
at a node, as was probably also the ease in Extracrinus hriareus with its large
" Krone."

Among the Comatulidse the vegetative system is reduced to a minimum, as they have
no stem in the adult condition. The arms, however, are often very extensively developed,
f;ir more so than in any recent Pentacrinidas. The ultimate arms of a Pentacrinus or
Metacrinus do not often exceed forty in number ; and they rarely consist of more than
one hundred joints, though twenty or thirty more may intervene between the last
axiUary and the calyx (Pis. XIV., XVI.; PI. XVIII. fig. 1 ; PI. XIX. fig. 1; PI. XXV.;
PI XXVIII. fig. 1; PL XXXI. fig. 1; PL XXXIV. fig. 1; Pis. XXXVIII., XL., XLIL;
PL XLIII. fig. 4 ; PL XLIV. fig. 2 ; PL XLV. fig. 1 ; Pis. XLVL, XL VIII. ; PL XLlX.
fig. 1 ; PL LII. fig. 1). On the other hand, although there are quantities of ten-armed
Comatulse, very many species, especially of Actinometra, have from forty to sixty arms;
some, like Actinometra bennetti, and Actinometra schlegeli, eighty or more ; and in a few
gigantic types like Actinometra nobilis- there may be over one hundred arms. Further, the
number of arm-joints is generally from one hundred and twenty to one hundred and fifty,
apart from the syzygies ; while in a large Antedon eschrichti or Actinometra bennetti and
in other multiradiate species of the latter genus there may be over two hundred arm-joints.
Nearly all of them bear pinnules, which are often very long until quite near the arm ends.
But in Metacrinus and also in Pentacrinus, though to a considerably less extent, the
development of pinnules stops short some little way from the extremity of the arm ; and its
outermost segments bear little stumps of two or three joints only, or may even show no
signs of pinnules at all (Pis. XXV., XXVIIL, XXXI., XXXIV., XXXVIIL, XL., XLIL;
PL XLIII. fig. 4; Pis. XLIV., XLVL, XLVIIL, XLIX., LL, LIL).

The same peculiarity is repeated on a smaller scale in the ambulacral plates of the
pinnules. Those of Comatulse (when present) are continued almost to the end of the
pinnule (PL LIV. fig. 6). But in the Pentacrinidse the last few pinnule joints, some-
times even four or six, are totally devoid of any ambulacral plating (PL XV. figs. 7-9 ;
PL XVL fig. 2; PL XLL fig. 9; PL XLVIL fig. 10; PL XLIX. fig. 7; PL LII.
fig. 5). The same is the case with the extremities of the arms. In fact, both in the
persistence of the stalk and of the external basals, and also in the nature of the arms,

1 Encriniden, p. 292.

2 0

'' The specific formula of this type is— a . 3 . - . 3 . 3 . — •

56 THE VOYAGE OF H.M.S. CHALLENGEE.

the Pentacrinidse show very clearly that they are rightly regarded as permanent larval
forms of the Comatulas.

Rhizocrinus and Bathycrinus, with their relatively large vegetative system, manifest
the same character in another way, viz., the absence of pinnules from the arm-bases ;
though the ambulacral plating is continued to the end of both arms and pinnules
(PL VII. figs. 2, 7 ; PI. VIII. figs. 1,3, 5 ; PI. Villa, fig. 1). There is good reason to
believe that the late appearance of the basal pinnules (excepting on the second brachial)
is a marked developmental character among the Comatulse ; ^ and in one genus, Atele-
crinus, the first pinnule is as far out as the twelfth brachial, the lower pinnules not
developing at all. In Rhizocrinus raivsoni it is on the epizygal of the third syzygial pair,
or the sixth primitive joint, and in Rhizocrinus lofotensis on the epizygal of the fourth
pair (PL IX. figs. 1-3) ; while in Bathycrinus it may be as many as eleven joints from
the radial axillary, though occasionally only eight or nine (PL VIII. figs. 1, 3).

In all the genera of living Crinoids, with one singular exception, the mouth is situated
at or near the centre of the disk (PL III. fig. 2 ; PL VI. fig. 4 ; PL VII. fig. 3 ; PL XVII.
figs. 6, 10 ; PL XXVI. figs. 1, 2 ; PL XXXIV. fig. 2 ; PL XXXIX. fig. 2 ; PL LV. figs. 3-7 ;
PL LVI. fig. 6), and the arms are about equally developed on all the five rays. But in
the large Comatulid genus Actinometra the mouth is excentric or even marginal (PL LV.
fio-s. 1,2; PL LVI. figs. 7, 8) ; and there is frequently a considerable amount of difi"erence
in the development of the oral or anterior, and the aboral or posterior arms.

Even when all the arms are provided with food-grooves on the ventral surface as in
other Crinoids, those which come ofi" round the mouth are usually longer, sometimes
considerably so, than those which spring from the hinder part of the disk ; while in other
species the anterior and posterior arms are all grooved and all equal in length, but the
distribution of their syzygies is quite difi"erent.

A great many species of Actinometra, however, are characterised by a still more
striking difi"erence between the anterior and posterior arms. The former have a wide
food-groove of the usual character and a well-developed tentacular apparatus at its sides,
while they always appear to end in a " growing point." The posterior arms, on the other
hand, have an ungrooved and convex ventral surface, which is without any respiratory
tentacles at all (PL LVI. fig. 7). They are only about half the length of the grooved
anterior arms, and, therefore, taper much more rapidly, while they terminate in an
axillary segment which bears two pinnules of the ordinary character. The genital glands
which they contain are usually far more developed than those of the anterior arms. Not
only are there more fertile pinnules, though the total number of pinnules may not be
much more than half that of an anterior arm ; but the portions of the glands within these
pinnules also attain a greater size than in the oral arms, the basal and median pinnules
of the latter being usually less swollen than the corresponding pinnules of a posterior arm.

* Preliminary Report on the ComatulcC of the Caribbean Sea, Bull. Mus. Comp. Zool., vol. ix., No. 4, pp. 14, 15.

REPORT ON THE CRINOIDEA. 57

A similar inequality in the development of the genital glands has been noticed by Prof.
A. Awassiz as oceurriuo; in the Echini/

This frequent difference in length between the anterior and the posterior arms of
Actinometra, accompanied by the difference in the character of their terminal pinnules,
seems to be to some extent dependent upon the condition of the respiratory apparatus
occupying their ventral surface. When this is well developed the arm seems to have the
power of indefinite growth. For in a great many individuals of various species which
have all the arms grooved and tentaculate like those of Antedon, there is no very
appreciable variation in their length or in the development of their genital glands.

There appears to be no rule of any kind respecting the condition of the arms in any
given species of Actinometra. In the case of Actinometra parvicirra, for example, I have
seen individuals with thirty -three arms, all of which were grooved and tentaculate ; while
in another wdth thirty-one arms as many as nineteen were grooveless and unprovided
with tentacles. All sorts of gradations between these two extremes will be found in any
large collection of Actinometrce.' Half the species of this genus which were dredged by
the Challenger have more or fewer ungrooved and less developed arms. They may
occasionally be found upon the anterior rays ; while in Actinometra nohilis and
Actinometra magnijica,^ which have one hundred arms or more, several of those on each
ray are short and less developed, with neither food-groove nor tentacles on their ventral
surface (PL LVI. fig. 7).

Even in the normal grooved arms of Actinometra the lower pinnules are frequently
grooveless and non-tentaculate, just as the hinder arms may be (PL LXI. fig. 3). Some-
times only three or four, sometimes as many as forty, are in this condition, being more
or less swollen by the development of the genital glands within them ; but they do not
receive any branches from the brachial ambulacrum, which is itself often but imperfectly
developed (see woodcut, fig. 4, p. 113). This ungrooved condition of the lower pinnules
may also occur on all the arms of some species of Antedon ; and it is especially remarkable
in types like Antedon accda and Antedon angusticalyx^ which have a strongly plated
ventral perisome. The ambulacral grooves of all the arms and of tke later pinnules are
well protected by plates (PL LIV. figs. 4, 7, 8, 9) ; but they do not extend on to (about)
the first twenty pinnules which contain the large genital glands, though the latter are
protected by a very close and regular pavement of anambulacral plates (PL LIV. figs.
1-3, 5). In other species, however, which have equally plated pinnules, such as Antedon
incefta,^ the ambulacra extend over their ventral surface in the usual way (PL LIV. fig. 6).

1 Revision of the EcWni, part iv. pp. 680, 681. - Actinometra, loc. cil., pp. 31-41.

3 The specific formula of this type is — a .3.2.3.3. -^ ■

* The following are the specific formula- of these types : A ntedon accela,~A . 10 . — ; ^ ntedon angusticalyx,—A . 3 . ^ .
5 The specific formula of Antedon incerta is — A. 10. — .

(ZOOL. CHALL. EXP. — PART XSXII. — 1884.) K 8

^8 THE VOYAGE OF H.M.S. CHALLENGE!;.

B. The Pinnules.

The pinnules are repetitions of the arms on a small scale, and are especially adapted for
the protection of the genital glands (PL Vc. figs. 7, 8, 10, t ; PI. VII. fig. 7 ; PI. X. fig. 20).

In no case is a pinnule developed earlier than the second joint above the first radials
of the calyx. This condition occurs in the two five-armed genera Thaumatocrinus
(PL LVI. figs. 1, 2) and Eudiocrinus. One species of the latter [Eudiocrinus varians)
has this second brachial free and capable of lateral movement, while in another
{Eudiocrinus indivisus) it is the epizygal of a syzygy. The corresponding radial joint
of Metacrinus is of the same character, and there are pinnules on each of the following
radials as far as the axillary (PL XXXVIII. ; PL XXXIX. fig. 1 ; PL XLIL; PL XLIII.
fig. 2 ; PL XLIV.; PL XLV. fig. 1 ; PL XLVL; PL XLVIII. fig. 1 ; PL XLIX. fig. 1 ;
Pis. L.-LII. fig. 1). But in the majority of Neocrinoids which have the third radial
an axillary, the preceding joint bears no pinnule, while it sometimes contributes to the
enlargement of the cup.

Pinnules are always absent from every axillary joint, from the hypozygal of every
syzygy (PL XXXa. fig. lOh ; PL XXXII. figs. 4, 6, 13, 14; PL L. figs. 11, 12), and
also from the lower one of every pair of joints which are united by a ligamentous articu-
lation ; so that in the great majority of Comatulse, as in some sj^ecies of Pentacrinus,
the first joint after each axillary bears no pinnule. In the former group too the pinnules
on the third and the four or five following brachials which form the arm-bases, do not
appear till after those of the eighth and following joints, though the pinnule of the
second brachial is developed comparatively early ; while in Atelecrinus, Rhizocrinus, and
Bathycrinus more or fewer of the lowest brachials are permanently devoid of pinnules.

The lowest pinnules of the Comatulse, and in a less degree those of the Pentacrinidas
also, usually ditfer somewhat from their successors ; and they may present a variety of
characters, which are of considerable value in the discrimination of species, owing to
the comparative constancy of their occurrence. They are frequently distinguished by
the presence of spurs or keels upon their basal joints, as in Actinometra Solaris ; or they
may be long, slender, flexible, and flageUiform, as in Antedon rosacea; or they may be
stifi", straight, and spine-like, as in Antedon 2>i"otecta; or they may have large prismatic
basal joints, as in Metacrinus (PL XXXVIIL; PL XXXIX. fig. 1 ; PL XLIII. figs. 2, 4 ;
PL XLIV. fig. 2; PL XLVL; PL XLIX. figs. 1, 2; PL L. figs. 1, 2 ; PL LIL fig. 1) ;
or the dorsal surfaces of their joints may have forward projecting keels, as in Pentacrinus
asterius (PL XIII. figs. 1, 14).

Dr. Carpenter^ has observed that the first pinnules of the ten-armed Antedon rosacea,
which habitually arch over the disk and are much longer than their successors, are

' On the Structure, Physiology, and Development of Antedon (Comatula) rosaceus, Proc. Roy. Soc, vol. xxiv., 1876,
p. 226.

REPORT ON THE CRINOIDEA. 59

extremely susceptible of irritation. When they are touched in the living animal, the whole
circlet of arms is suddenly and simultaneously coiled up over the disk ; while irritation of
one of the ordinary pinnules is simply followed by the flexion of the arm which bears it.
The structure of these " oral pinnules," which in Antedon rosacea and allied species
are borne by the second brachials, differs very considerably from that of the pinnules
on the other arm-joints. For not only are they sterile, but they have neither
tentacular apparatus nor ambulacral groove. Their ventral surface is slightly
convex, instead of being concave, as in the ordinary arms and pinnules ; while the
ciliated ambulacral epithelium, together with the subjacent nerve and radial blood-vessel,
are also absent. This appears to be the case with the oral pinnules of almost all
Comatulag ; while in some species of Actinometra whole arms, with all the pinnules which
they bear, are in the same condition.

In the Pentacrinidas, however, the lowest pinnules of the rays are usually all grooved
like their successors, and not devoid of the ambulacra with aU their accessory structures, as
in the Comatulse (PI. XXXIV. fig. 2). In fact the pinnule-ambulacra of Metacnnus
often start directly from the peristome or from the five primary groove-trunks of the
disk, instead of from the particular branches corresponding to the arms which bear the
pinnules (PL XXXIX. fig. 2 ; PI. XLIII. fig. 3 ; PI. L. fig. 2).

The fact that the pinnules are only arms in miniature is very well shown by the
process of their development at the terminal growing points of the young arms. The
first indication of a pinnule is the formation of a fork at the growing point, the two
limbs of which are at first almost equal (PI. XXXV. fig. 1). "One of these rami,
however, grows faster than the other, and soon takes a line continuous with that of
the axis of the arm, from which the other diverges at an acute angle, so that the former
comes to be the proper extension of the arm, while the latter soon takes on the
characters of a pinnule. Ere long, however, the growing point of the arm again
subdivides ; two branches are formed as previously ; and whilst one of these becomes a
continuation of the arm, the other is soon to be distinguished as a pinnule given ofi' from
it on the side opposite to that of the first formed pinnule." ^

In all the Neocrinoidea, with the exception of Hyocrinus, the pinnule borne by any
joint is small in comparison with the arm of which that joint is a part. But in this
aberrant genus (PI. VI. figs. 1, 2) the pinnule-bearing joints have rather the appearance
of axillaries. For the pinnules are large in proportion to the arms, and are nearly equal
in length to the parts of the arms which are beyond the joints where they originate, so
that the outer ends of all the pinnules, as well as those of the arms, meet nearly on a level.
Hence the pinnules of Hyocrinus have some resemblance to the "armlets" or small and
undivided but pinnule-bearing arms which come off" from the inner faces of the axillaries
of Extracrinus. But their function as pinnules is shown by the way in which they are

• Phil. Trans., 1866, p. 734. See also Trans. Linn. Soc. Land. (ZooL), eer. 2, vol. ii., 1877, p. 40, pi. ii. figs. 4, 6.

60 THE VOYAGE OF H.M.S. CHALLENGER.

enlarged to receive the fertile portions of the genital gknds, no part of these appearing
in the arms (PI. Vc. figs. 7, 8, 10, t; PI. VI. fig. 1); while they have no appendages of
their own as the armlets have in Extracrinus.

The peculiar pinnule arrangement of Hyocrinus helps us to understand why there
are no pinnules upon the axillaries of multiradiate Crinoids. These may be considered
as ordinary pinnule-bearing joints, so modified that the pinnule and the continuation of
the arm which bears it are equal in size or nearly so. As mentioned above, this is in
fact the mode of formation of the pinnules at the growing points of the arms, as is well
shown in a very young individual of Pentacrinus decorus (PI. XXXV. fig. 1). The
joint which bears the last formed pinnule is an axillary with two nearly equal distal
faces ; and the pinnule can only be distinguished from the continuation of the arm by
the greater length of its component joints. Furthermore, in the short posterior arms of
Actinometra, the only ones in which the normal mode of termination has been observed,'
the last joint is an axillary which bears two pinnules of the ordinary character.

In Rhizocrinus (PL IX. figs. 4, 5) as in Hyocrinus (PI. VI. figs. 1, 2) the pinnule-
bearing joints have very much the appearance of axiUaries with unequal distal faces; and
a similar inequality is shown by the axillaries of Extracrinus, each of which bears an
" armlet " on one face and the continuation of the main arm -trunk on the other.

Numerous instances of reparation after injury also indicate the close similarity of
arms and pinnules. A very common one, sometimes to be met with in Antedon rosacea,
is as follows : — The epizygal of the third brachial is broken away, carrying with it all the
outer part of the arm, as well as the pinnule which it bears. But it is replaced by an
axillary with two distal faces, from each of which an arm eventually grows out, one or
other of them perhaps dividing again, as in the specimen of Pentacrinus decorus shown
on PL XXXVI. On the other hand, in an abnormal individual of Metacnnus angidatus,
the eighth distichal is not an axillary, as is usually the case. But it is somewhat swollen
and has a slightly larger pinnule than the preceding joint, so that it resembles an axillary
with unequal faces. In the specimen of Actinometm strota which is represented on
PL LV. fig. 2, one of the second brachials of the right posterior ray bears two fully
developed pinnules instead of an arm and its own proper pinnule, so that it looks like an
axillary. There is no disk-ambulacrum corresponding to this undeveloped arm.

Considering therefore the fundamental identity of arms and pinnules, one w^ould
scarcely expect that an axillary joint which gives rise to two arms (often unequal in size)
should bear a pinnule as well (see pp. 347, 358).

The pinnule arrangement of Hyocrinus is totally unlike that of any other Neocrinoid,
although, according to Sir Wj^irille Thomson," we have something very close to it in some
species of the Palseozoic genera Poteriocrinus and Cyathocrinus. These names were
probably employed by Sir Wy ville in the ■wide sense, and not with the restricted meaning

' Actinometra, loc. cit., p. 40, pi. ii. fig. 6. ^ Journ. Linn. Soc. Lend. (ZooL), vol. xiii., 1876, p. 52.

■REPORT ON THE CEINOIDEA. 61

which they now possess. There arc no true pinnules in Cyathocrinus proper, but only
repeatedly branching arms, which must therefore have contained the genital glands ; and
each of the branches borne by one of the lower axillaries may fork again several times, so
that they cannot be compared to the long, undivided pinnules of Ilyocrinus (PI. VI.
figs. 1,2). In Poteriocrinus, on the other hand, the successive arm -joints bear pinnules
which are not specially different from those of Neocrinoids ; while the mode of branching
of the arms resembles that characteristic of Pentacrinus miillerl, and more especially of
Extracrinus, the axillaries being generally limited to the outer arms of the ray, and
having unequal distal faces. ^

It is in the curious genus Barycrinus, which was separated from Cyathocrimis by Mr-
Wachsmuth, that we find the nearest approach to the pinnule-arrangement of Ilyocrinus.
According to Wachsmuth and Springer" "all the main arms, instead of Ijifurcating,
give off" at regular intervals, alternately on opposite sides, and from the inner margins of
the plates, short, rounded, simple armlets, which in most species throw ofi" secondary
branches as in Botryocrinus, and these armlets here as there, probably performed the
office of pinnulje." Barycrinus hoveyi, var herculeus, M. and W., is one of the excep-
tional species in which the armlets are simple and without secondary branches. The
excellent figure of it which is given by Meek and "Worthen^ shows these armlets
to come off alternately on opposite sides just as in Ilyocrinus, but from every second
joint, instead of from every third (PI. VI. figs. 1, 2). They have unfortunately never
been found in a perfect condition ; and we cannot tell therefore whether they reached
to the level of the top of the arms proper as in Ilyocrinus, though Meek and AVorthen's
description seems to indicate that such is the case.

The difference between these armlets of Barycrinus herculeus and those of Extra-
crinus is that the former seem to bear no pinnules as the latter do, and must therefore
have contained the genital glands ; while they come ofi" alternately from opposite sides of
the main arm-trunk, and not from its inner one only as in Extracrinus.

We have seen that the pinnule of a Neocrinoid is practically a reduced copy of an
arm, but modified by the great development of the fertile portion of the genital gland
which it contains, that part of the gland which is confined to the arm being usually
sterile and known as the " genital cord " or " rachis " (PL Vc. fig. 1 ; PI. Villa, figs. 4, 5 ;
PI. LX. fig. 6—rjc.).

Although it would seem improbable that the pinnules of Palseocrinoids are essentially
different in nature from those of the Neocrinoids, Messrs. Wachsmuth and Springer
have attempted to show that the small alternating plates covering the brachial ambu-
lacra of Cyathocrinus are homologous with the pinnules of the Actiuocrinida3 and
Platycrinidje. These are long, comparatively slender, and closely arranged side by side

1 See p. 277. ' Revision, part i. p. 101.

^ Palu'ontology of Illinois, vol. v. pi. xiii. fig. 2.

62 THE VOYAGE OP H.M.S. CHALLENGER.

just as on the feathery arms of Antedon eschrichti and hosts of other ComatulaB ; and
they are obviously of the same nature as the pinnules of Neocrinoids generally. " When
the arms are closed, the two series of pinnulfe of one arm are laid upon each other so
neatly, that the arm-furrow must have been thereby perfectly shut off from the surround-
ing water. No additional covering has yet been observed in these genera, and it was
evidently unnecessary. All this seems to point to the conclusion that the pinnulse had
the same functions, partly at least, as the alternate plates in Cijathocriniis, &c., and as
both have the same position, and evidently could be opened and closed by the animal,
we do not hesitate to consider the latter as the homologue of the former, or in fact
as rudimentary pinnulse."^

The first sentence of the above passage concerning the pinnules of Actinocrinus and
Platycrinus would apply equally well to any Comatula or Pentacrinus, whether the
ambulacra be plated or not (PL XIII. fig. 13 ; PI. XIV.; PL XVII. fig. 1 ; PL XXVII.
fig. 13; PL XXXIII. fig. 3; PL XLI. fig. 11; PL XLVII. fig. 12; PL LIV. fig. 1).
As regards the last paragraph, I cannot help thinking that it affords an instance in
which analogy has been mistaken for homology.

The overlapping of the pinnules so as to cover in the ambulacra may occur in all
recent Crinoids ; while the grooves of the pinnules themselves, like those of the arms
and disk, are often bordered by two more or less distinct rows of minute movable
alternating plates, the " covering plates." These may themselves be supported on
" side plates," thus making four rows in all', which are sometimes very fully developed
as in Hyocrinus, together with many Pentacrinidse and Comatulse (PL Vc. figs. 9, 10;
PL Villa, fig. b—cp. PL XIII. figs. 15, 16; PL XVII. fig. 8; PL XXVII. figs. 4-6, 11-13;
PL XXXIII. figs. 1-4; PL XLI. figs. 4, 11-13; PL XLVII. figs. 10-12; PL XLIX.
figs. 6, 7 ; PL LI. figs. 11, 12 ; PL LII. figs. 5, 6 ; PL LIV. figs. 4, 6-9). Similar plates
occur on the pinnules of Actinocrinidaj and Platycrinidse, their grooves being " covered
by a double series of very minute pieces, though, owdng to defective preservation, this
covering is rarely observed."'" These pinnules "fit together so neatly and cover the
arm -furrow so perfectly that additional plates were scarcely needed."'

A teleological argument of this kind is, however, no proof that the brachial ambulacra
are unprovided with plates in Actinocrinus and Platycrinus, when there is a double
series on the pinnules which they bear ; and, as a matter of fact, the evidence afforded
by the Neocrinoids is all against this view. Hyocrinus, Bathycrinus, and Rhizocrinus
all have covering plates on the arms as well as on the pinnules (PL Vc. figs. 8-10 ;
PL VI. figs. 1-4 ; PL VII. figs. 2, 7, 8 ; PL VIII. figs. 3, 5 ; PL Villa, fig. 1 ; PL IX.
fio's. 1-4 ; PL X. fig. 20). The first of these has side plates on the pinnules (PL Vc.
figs. 9, 10, sp.), but they are not distinctly differentiated on the arms; and the same is
the case in the Pentacrinidse and Comatulaj. But except in Pentacrinus maclearanus

' Revioion, part i. p. 25. - nid., part ii. p. 25. = Ibid., p. 24.

EEPORT ON THE CTJNOIDEA. 03

(PI. XVI. figs. 2, 3 ; PI. XVII. fig. 1), with its extraordinarily narrow arm-groove, the
ambulacra! plating of the pinnule always arises from a similar hut less defined skeleton
on the brachial ambulacrum (PI. XVII. figs. 7-9 ; PI. XXXIII. figs. 3, 4 ; PL XLI.
figs. 4, 13; PI. LIV. fig. 7). This is itself directly continuous with the ambulacral
plates of the disk; while the perisomic plates, which may appear at its sides (PI. XXVII.
figs. 6, 13 ; PI. XLI. figs. 4, 13), are in like manner connected witli the anambulacral
system over the arm-bases (PI. XXVI. figs. 1,2; PI. L. fig. 2). Wachsmuth's own
beautiful observations have demonstrated the existence both of anambulacral and of
ambulacral covering plates on the upper surface of the body, beneath the vault of
Actinocrinns ; and since the latter also appear on the pinnules, it seems unreasonable
to doubt their presence on the arms.

But if, as I firmly believe, brachial covering plates occurred in Actinocrinus as in
Cyathocrinus, what becomes of the supposed homology between these covering plates in
the latter genus and the pinnules of the former type ?

Mr. Wachsmuth appears tt) me to have been much nearer the truth when he suggested
that the many little branches of the bifurcating arms in Cyathocrimts performed the
functions of pinnules,^ though he gave no explanation as to what these functions were.
In recent Crinoids, and most probably therefore in the fossil ones also, the functions of the
pinnules are threefold, viz. , ( 1 ) the protection of the fertile portions of the genital glands,
which are all connected together by the sterile rachis in the arm ; (2) respiration ;
(3) alimentation.

Dr. Carpenter" has pointed out that the Crinoids are very closely dependent for the
maintenance of their life upon pure, weU aerated water. He alludes to the importance
of the pinnated arms in bearing a vast aggregate of tubular tentacles by which
respiration is eflected ; and regards it as probable "that the ordinary pinnules are
specially related to the function of respiration, in virtue alike of their proper branchial
canals, and of the ambulacral canals and the tubular tentacula with which they are
furnished."

This process of respiration was doubtless efi"ected just as well by the tentacles
connected with the water-vessels in the many-branched arms of Cyathocrinus, as by
those on the pinnules of Actinocrinus or Comatula ; and there is no reason why the
genital glands should not have l)een contained in these pinnule-less arms, for they
frequently extend from the pinnules down into the arms both in Holoims (PI. Vc.
figs. 1, 2, ov.), in many Comatula3 (PI. LXI. fig. 3), and even in Pentacrinidaj ; so that
they often appear in section as taking the place of the sterile genital cord, which unites
the more fertile portions of the gland that are contained in the successive pinnules.

The third great function of the pinnules of a Crinoid arm is to present as large an

^ Notes on the Internal and External Structure of Palssozoic Crinoids, ^mcr. Journ. Sci. and Arts, vol. xiv. p. 120.
2 Phil. Trans., 1866, pp. 701, 702.

64 THE VOYAGE OF H.M.S. CHALLENGER.

area as possible of the ciliated ambulacral grooves on tlieir ventral surface in order
to catch the minute organisms in the surrounding water which might serve as food, and
send them down the ambulacra of the arms towards the central mouth. For this purpose,
as for that of respiration, the repeated branching of the long arms of Ci/athocrinus would
be as eflfective as the development of pinnules on the successive joints of less divided arms
in other Crinoids. The three great functions of these pinnules would thus have been
performed without difficulty by the branching arms of Cjjathocriniis. Biit for which of
them are the covering plates of the arm -grooves at all adapted, and how far can these
plates be considered as repetitious of the arms on a small scale? To each of these
questions only a negative answer is possible.

The covering plates of recent Crinoids may be found closed down over the food-groove
after death (PI. XIII. fig. 16 ; PL XVII. fig. 7 ; PI. XXVII. fig. 12 ; PI. XXXIX.
fig. 12 ; PI. XLVII. fig. 10 ; PL LI. fig. 12 ; PL LIL fig. 6; PL LIV. figs. 4, 6); but they
are just as often met with in a more or less erect position, thereby opening the food-
groove to the exterior (PL Vc. figs. 8-10; PL Villa, fig. 5— cp. PL XVII. figs. 2, 8, 9;
PL XLVII. figs. 4, 13; PL LI. fig. 11 ; PL LIV. figs. 7-9). Just in the same way the arms
are frequently closed round the disk in the dead animal (Pis. XVIII. , XIX., XXV., XXVIII. ,
XXXI., XLV., XLIX., LIL) ; while in other cases they are more or less expanded, as
they were during life (Pis. XXXIV. , XL. . XLIL). Messrs. Wachsmuth and Springer argue,
however, (l) because the arm-groove of the fossil Cijathocrirms is closed by covering
plates which could be opened and closed by the animal (as it is in the dried arm-fragment
of Pentacrinus asterius which is shown in PL XVII. fig. 7), and (2) because the arm-
groove of Actinocrinus must have been perfectly shut ofi" from the surrounding water
by the apposition of the pinnules ivhen the arms loere closed; therefore the covering
plates of Cijathocrinus are homologous with the pinnules of Actinocrinus.

But what advantage is it to the animal to have its arm-grooves closed up, whether
by covering plates or by pinnules, and so shut ofi' from the surrounding water? It
could not breathe properly in this condition, neither could it get its food. None of the
food particles which one finds so frequently in the alimentary canal of a Crinoid,
e.g., Eadiolarians, Foraminifera, Diatoms, &c., could enter the food-grooves of the arms
if they were closed by covering plates or by the apposition of the pinnules over them.
The habitual expansion of the arms is essential to the whole fife of a Crinoid, and
Prof. L. Agassiz has well described their movements in the living Rkizocrinus. " We
had the Crinoid alive for ten or twelve hours. When contracted the pinnules are
pressed against the arms, and the arms themselves shut against one another, so that
the whole looks like a brush made of a few long coarse twines. When the animal opens,
the arms at first separate without bending outside, so that the whole looks like an
inverted pentapod ; but gradually the tip of the arms bends outward as the arms
diverge more and more, and when fully expanded the crown has the appearance of

REPORT ON THE CRINOIDEA. 65

a lily of the Lilium Martagou type, in which each petal is curved upon itself, the
pinnules of the arms spreading laterally more and more as the crown is more fully
open. . . . When disturbed, the pinnules of the arms first contract, the arms straighten
themselves out, and the whole gradually and slowly closes up." ^

Taking all these facts into consideration, I cannot but feel that a homology is of
no real value when it is based upon the physiological condition of the arm-grooves in
the dead animal, and still more in the fossil forms, closed up as they are in every
possible way, especially when this condition is one which the living animal only assumes
when disturbed, and cannot long maintain without the risk of being both starved and
suffocated. The whole point of AVachsmuth and Springer's argument, however, is
based upon this closure of the arm-grooves by pinnules and covering plates respectivel}- ;
and they attempt to support the proposed homology by certain morphological considera-
tions, which must now be discussed.

On each side of the brachial ambulacra of Ci/athocrinus iowensis there are, accordiuo-
to Wachsmuth,'^ two rows of minute alternating plates, six to each arm-joint. A
similar structure is shown in one of Angelin's figiu'es of an arm-fragment of Gissocrinus
pimctuosus,^ though in another figure only one row of plates is visible at the side of the
ambulacrum instead of two, while the explanation of the figures simply says, " Dioiti
cum pinnulis magnitudine aucta." A somewhat diff"erent structure appears in Cyatlio-
crinus longimanns figured on the preceding plate.'* In this species, according to
Wachsmuth and Springer,^ " there are in place of only two, a series of five successive
plates from each side, alternately arranged. The plates of each side taper toward the
end and enfold over the furrow, covering it as perfectly and in the same manner as in
the two former cases {i.e., Cyathocrinus ioioensis and Gissocrinus punctuosus). Angelin
gives no description, but in his tal)le of contents he calls the successive plates ' pinnulse.' "
Although, however, Angelin may haA^e used the word " pinnulje " for these lateral plates,
I doubt how far he meant to imply any correspondence with the true pinnules of
Actinocrinus and Platycrinus and other types in which they occur. For in his definition
of Crotalocrinus he gives the same name to the lateral processes of the arms by which
tliey are united into the well-known complex network ; and he then continues, " Perisoma
ventrale totum assulis variantibus tectum ; assulfe ambulacrales minutae, biseriatfe ali
imis brachiis usque ad extremes digitos radiatim exeunt, quarumque numerus prout
digitorum numerus magis magisque per repetitam dicbotomiam increscit."'' The magni-
ficent figure which he gives of the ventral surface of an expanded Crotalocrinus pulcher '
shows that the minute ambulacral plates on the arms are identical with the covering

1 Quoted l-iy Pourtales, On ,1 New Species of Rhixocrhnis from Barliadoes, Mem. Mus. Comp. ZoiiL, vol. iv.,
No. 8, p. 29.

2 Amcr. Journ. Sci., vol. xiv. p. 121. ^ Iconograpliia Crinoideoruni, Tab. xxvii. fig. 1/
< Ibid., Tab. xxvi. figs. 4, 5. * Revision, part i. pp. 24, 25.

8 Iconographia Crinoideonim, p. 26. ' J^'i''- Tali, ■i-iii. fig. 6 ; see also Tab. xxv. tigs. 15, 17.

(ZOOL. CHALL. EXP. — PART XXXII. — 1884.) li 9

6(5 THE A'^OYAGE OF H.M.S. CHALLENGER.

plates ou the arms of Gissocrinus, which he described in his generic diagnosis^ as
" pinnulse brevissimse verrucseformes." They do not correspond to the " assulse ambula-
crales " of Gissocrinus ; for these, as pointed out by Wachsmuth and Sjjringer,^ are the
joints of the arm-bases which are recumbent upon the vault, being raised above the
general level, and " covered by small alternating plates like the fi'ce arms."

Thus, then, the expression " pinnulse " was used by Angelin for three diflerent types of
structure in Hahrocrinus, Crotalocrinus, and Gissocrinus respectively. The first of these
represent tlie true pinnules of recent Crinoids ; while, as was first recognised by Zittel,"'
the plates on the arms of Gissocrinus represent the " Saumplattchen " which Miiller had
described on the ambulacra of Pentacrinus. He takes the same view of the plates which
Angelin called "rudimenta, pinnularum semiglobosa" in Cyathocrimis ramosus. The
corresponding plates in Cijathocriims longimanus have indistinct cross markings ; and
Messrs. Wachsmuth and Springer have interjareted this as indicative of segmentation into
pinnule-joints ;* for they speak positively of " two rows of five successive plates each, one
row being given off' from the right, the other from the left side of the furrow, and perfectly
covering it." The arms of Cyathocrimis, however, are only single-jointed ; and they
state elsewhere ' that " in doul)le-jointed arms every joint at each side bears a pinnule,
while in those with single joints the pinnules are found only on alternate sides." But
Angelin's figures show that on five arm-joints there are twelve of these rudimentary
pinnules, six on each side, while in Gissocrinus there are from four to six ou each arm-
joint. In spite of these facts, however, Wachsmuth and Springer give it as their opinion
that these plates which cover in the ambulacral groove in the arms are the homologues of
the pinnules, though too rudimentary to be ranked as such.^ [See p. 84.]

The chief point of their position (and they make ithe most of it) is the apparent
segmentation of the " rudimentary pinnules " into five joints, as shown in Angelin's
figures of Cyathocrinus longimanus. It would be very desirable if this could be verified
on the actual specimens. The presence of two rows on each side of the ambulacrum in
Oyathocrinus, Gissocrinus, &c. , is nothing unusual ; for they represent the side plates and
covering plates of Neocrinoids, of which there may be from five to seven to one arm-joint.
Precisely similar plates occur on the pinnules as well as on the arms ; and they also form
part of the skeleton round the radiating tubes beneath the vault of Actinocrinus " whicli
connect with the ambulacral furroM's in the arms." '^ But I think that Wachsmuth and
Springer would hesitate before considering the two upper rows of plates in these radiating
tubes as "rudimentary pinnulfe."

^ Iconograpliia Crinoirleorum, p. in. - Revision, part ii. p. 91. ^ PaUt'ontologiL', pp. 338, 354.

* Revision, part i. pp. 24, 82. ^ V'id., part ii. p. 24. " Ihid., part i. p. 82. ' Ihirl, part ii. p. 28.

REPORT ON THE CRINOIDEA. 67

v.— THE VISCERAL MASS.

The body-cavity enclosed between the dorsal skeleton and the ventral perisome of a
Crinoid consists of two principal parts, which have been conveniently designated by
Ludwig ^ as the " intcrvisceral " and the " circumvisceral " ccelom respectively. In some
Comatulse. such as Antedon rosacea and Actinometra strota,^ these two divisions of the
coelom are very distinctly separated ; while in other types, such as Antedon eschrichti,
Actinometra 2>cirvicirra, and also in tlie stalked Crinoids, it is difficult to fix a definite
boundary between them. In the former case the coiled digestive tube, wliich is covered
in above by the ventral perisome, is protected below by a continuous sheet of connective
tissue. This forms a definite membrane enclosing the lower part of the visceral mass or
disk, and has been spoken of as the visceral layer of the peritoneum.^ In Antedon rosacea,
Actinometra strota, and similar forms, this visceral layer is only loosely attached to the
parietal layer which lines the interior of the cup formed by the rays and arm-bases. The
result is that a comparatively slight amount of violence is sufficient to separate the
visceral mass from the calyx and to tear the ambulacra across at the arm-bases. The
whole visceral mass, including the digestive tube and plexiform gland, together with the
circumoral rings of the blood-vascular, water-vascular, and ambulacral nervous systems,
is very apt to be turned out of the calyx, which is then left to swim about on its own
account.

The so-called " recent Cystidean," Hyponome sarsii of Loven,* is, in fact, nothing
more than the much-plated visceral mass of an Antedon common at Cape York (PI. LV.
figs. 3, 4) ; and the same thing may happen to the disks of the equally abundant
Actinometra strota and Actinometra julcesi. Several of these isolated disks were dredged
by the Challenger, together vvdtli a number of entire individuals and some eviscerated
calices ; and Sir Wyville Thomson informed me that he had observed them perform slow
movements of locomotion over a flat surface ; while we know from Dr. Carpenter's
experiments that the eviscerated but arm-bearing calyx of Antedon rosacea will execute
the usual graceful movements of swimming as perfectly as the entire animal had
previously done.*

The characters of the perisome covering the ventral surface of a Criuoid vary con-
siderably. It may be more or less completely covered by plates (PI. YI. figs. 1-4
PI. XYII. figs. 6, 10; PI. XXYI. figs. 1, 2; PL XXX. fig. 2; PI. XXXIII. fig. 7
PI. XXXIY. fig. 2 ; PL XXXIX. fig. 2 ; PI. XLIIL fig. 3 ; PL L. fig. 2 ; PL LY.

1 Beitriige zva Anatomie tier Crinoideen, ZcUschr. f. u-iss. ZooL, Bd. xxviii. pp. 306-308.

2 The specific formula of Actinometra strota is — a . R . 10 . — . —

' Proc. Roy. Soc. Land., vol. xxiv. pp. 213-215.

* On Hyponome Sarsii, a recent Cystidean, Canadian Naturalist, N. S., vol. iv., 1869, pp. 265-268.

* Proc. Eoy. Soc. Lond., vol. xxiv. p. 453.

68 THE VOYAGE OF H.M.S. CHALLENGER.

PI. LVI. fig. 5), or it may be entirely bare to the naked eye, although microscopic
examination usually reveals the presence of calcareous spicules at the sides of the
ambiUacra. Its thickness and consistency vary greatly, more particularly in the arms.
In Thaiimatocrinns (PI. LYI. figs. 2, 4), Atelecrimis, Promachocrinus, some species of
Eudiocrinus, in many of the ten-armed Antedons of the temperate and Arctic Seas, and
also in those from the greatest depths (1000 to 2900 fathoms), the ventral perisome of the
arms is quite thin and delicate, except in the median line, where it is occupied by the
ambulacra. These are consequently brought close down upon and between the muscular
bundles, which are distinctly visible through the thin layer of perisome that covers them.
In Actino7netra, however (PI. LV. figs. 1, 2 ; PI. LVI. fig. 7), and also in the larger
tropical Aiitedons, the ventral perisome of the arms is relatively thick and substantial,
and no muscular bundles show through it at the sides of the ambulacra. These pass over
the arm-bases on to the disk, where they are usually raised, sometimes considerably so,
above the general level of its ventral surface. They converge towards the radial centre,
where they unite into a smooth flattened space, the peristome ; and somewhere in this
space, though not necessarily in its centre, is the opening of the mouth.

In all recent Crinoids, with the single exception of the large Comatulid genus
Actinometra,^ the peristome is situated at or near the centre of the disk (PI. III. fig. 2 ;
PI. VI. fig. 4 ; PL VII. fig. 3 ; PI. XVII. figs. 6, 10 ; PI. XXVI. figs. 1,2; PL XXXIV.
fig. 2; PL XXXIX. fig. 2; PL XLIII. fig. 3; PL LV. ; PL LVL figs. 5, 6).
The same was doubtless the case, both in the fossil species of these genera, and also in
the allied but extinct genera of Neocrinoids, e.g., Cotylecrinus, Eugeniacrinus, Apio-
crinus, Marsupltes, &c.; while the Blastoids and most, if not all, of the Palseocrinoids
present the same character.

In all the Comatulee which have a central mouth, five primary ambulacra diverge
from the peristome and fork more or less frequently according to the number of arms
which they have to supply (PL LV. figs. 3-7). This may also be the case in the Penta-
crinidse, more especially in those species which have a small number of arms (PL XVII.
fig. 6 ; PL XXX. fig. 2 ; PL XXXIII. fig 7 ; PL XXXIV. fig. 2 ; PL L. fig. 2). But
in the multiradiate species the amijulacra of the disk are often quite irregular in their
arrangement (PL XVII. fig. 10 ; PL XXVI. figs. 1,2); and in the case of Mctacrinus
the ambulacra of the large lower pinnules sometimes start directly from the peristome
or from one of the large primary groove-trunks, instead of from one of the subdivisions
of the latter (PL XXXIX. fig. 2 ; PL XLIII. fig. 3 ; PL L. fig. 2).

The brachial ambulacra of the regular Crinoids, which have no very great develop-
ment of limestone in their ventral perisome, are usually well defined and bounded
laterally by elevated folds of the perisome. The edge of each fold is cut out into a
series of minute marginal leaflets ; while at the base of each of these, and to some extent

' See Appendix, Note C.

REPORT ON THE CRIXOIDEA. 09

protected by it, is a group of three tentacles arising from tlie radial water-vessel. These
groups of tentacles alternate on the opposite sides of the ambulacrum along the whole
length of the arm, and they accompany its lateral branches on to the pinnules.

In the irregular genus Actinometra the mouth is not central or sub-central, but
excentric or even marginal. It is sometimes radial in position, as in Actinometra strota
and Actinometra jukesi (PI. LV. figs. 1, 2),' and sometimes interradial, as in Actino-
metra magnijica (PI. LVI. fig. 7) ; while the number of ambulacra diverging from the
l^eristome may vary, especially in the multiradiate species, from four to ten or even more.
In most cases, however, the large intorpalmar area which contains the central or sub-
central anal tube is bounded by two large aboral groove-trunks, which form a horse-shoe
shaped curve, and give off the ambulacra of the lateral and posterior arms. If the mouth
be radial these grooves are often tolerably equal and symmetrically arranged, as in
Actinometra strota and Actinometra jukesi (PI. LV. figs. 1, 2). But in ty|3es witli an
interradial mouth like Actinometra magnijica or Actinometra stelligera (PL LVI.
figs. 7, 8),'^ the right or western limb of the curve is much larger than its fellow. [See
p. 88.] The latter supplies the arms of the left lateral ray only ; while the larger right
limb represents the posterior ambi;lacrum combined with part or the whole of the right
lateral ambulacrum as well.

In many forms of Actinometra more or fewer of the arms which arise from the
posterior and postero-lateral portions of the disk are in the ungrooved and non-tentacu-
late condition, as has been already described ; and the disk ambulacra corresponding to
them are less developed than those of the normal grooved arms. In fact they may be
altogether absent in some of the large multiradiate species such as Actinometra magnijica.
and Actinometra nohilis, both of which are remarkable for the presence of ungrooved
arms on each ray. In Actinometra magnijica for example, which has an interradial
mouth (PL LVI. fig. 7), not only are all the seventeen arms of the hinder ray entii-ely
unprovided with food-grooves, but more or fewer of the closely crowded arms on each of
the other rays are in the same condition, as they do not receive branches of the dividing
groove-trunks which supply their fellows.

Faint lines are sometimes visible, indicating the directions which these grooves should
have taken, i.e., the positions of the simple water-vessels which are distributed to the
ungrooved arms, but give oft' no tentacular branches. This is particularly well seen in the
case of the posterior ray and parts of the hinder divisions of the two lateral rays of Actino-
metra magnijica (PL LVI. fig. 7). The primary groove-trunk, which leaves the right lower
corner of the peristome to supply tlie right and posterior rays, divides into two princij^al
branches before reaching the equator of the disk in which the anal tube is situated.

1 The specific formula of ^c(inome«ra j«/im is— a. R. -^ — ; ;,'°j-^°^]""i I-

2 The specific formula of Actinometra stelligera is— a. 2.2.— .

70 THE VOYAGE OF H.M.S CHALLENGEE.

One of these supplies most of the arms in the anterior division of the right ray, and tlie
other most of those in its posterior division. But some of the arms in both divisions
receive no groove at all ; and there is only a faint line on the arm-bases indicating the
course of the water-vessel which proceeds to the hindermost arms of this ray, while
there is a similar line on the eastern side, indicating the course of the water-vessel
suppljdng the hinder arms of the left ray. [Compare fig. 3 on p. 92.]

The posterior ray, however, has no separate ambulacrum of its own upon the disk ;
and the right limb of the usual horse-shoe shaped curve is therefore incomplete. At the
point where the large and widely open primary groove-trunk forks into the two divisions
of the right lateral ambulacrum, a faint line may he seen starting from its inner side.
This is all that there is to represent the posterior ambulacrum. It curves backwards
round the margin of the disk to its hinder edge, gradually becoming less and less distinct
the farther it goes, until its ultimate branches to the arm-bases are only traceable with
the greatest difiiculty. All the arms of this ray are therefore unprovided with ambulacra,
and with most of the structures which are correlated with their presence (PL LXI. fig. 3).
They have neither blood-vessel nor ventral nerve, and their water-vessels are simj)le
tubes like the integumentary water- vessels of the Molpadidse.

These variations in the development of the ambulacra on the disk and arms of
Actinometra have somewhat important morphological bearings, as will be explained in a
later chapter. Curious malformations of the disk are occasionally to be met with among
the Crinoids. Thus in the disk of Metacrinus angulatus, which is represented on
PI. XXXIX., there are two anal tubes of unequal size, between which I found a Myzoatoma
reposing. Three monstrosities of Comatula disks have also come under my notice. In
one case, an Antedon, there are two anal tubes as in Metacrinus angulatus, but of more
equal size ; while in another, Actinometra stelligera, there are not only two anal tubes,
but also two mouths (PI. LVI. fig. 8). The true mouth is interradial, and there is a
large sub-central anal tube immediately behind it, as in all species of this genus. But
there is also a second anal tube occupying the interval between the posterior ambulacrum
which curves round the western portion of the disk, and the branches of the lateral
ambulacrum on its eastern side. This is very irregular in its character, and expands at
one portion of its course into a second peristome in which is a small mouth. Lastly, in
a specimen of Antedon rosacea from Milford, only four ambulacra reach the central mouth,
while the fifth (tlie left antero-lateral) has a second mouth all to itself (PI. LVI. fig. 6).

A. The Okal Plates.

The well-known oral plates which are so characteristic a feature in the Pentacrinoid
larva of Antedon and Actinometra are resorbed before maturity is reached. But they
persist through life in Holopus, Hyocrinus, Rhizocrinus, and Thaumatocrinus, imme-

REPORT ON THE CRINOIDEA. 71

cliately surrounding the peristome and covering it more or less completely (PI. III.
fig. 2 ; PI. Vc. fig. 6, 0; PL VI. figs. 3, 4 ; PI. X. figs. 7, 20 ; PI. LVI. fig. 5).
Their rudiments appear in the free-swimming larva simultaneously with those of the
basals, which are developed spirally round the right peritoneal tube ; while the orals
appear in a similar spiral around the left one.

The skeleton is at first limited entirely to these two rings of plates, the edges of
which meet around the equator of the growing cup, though they ultimately become
separated by the appearance of the radials between them.

At the base of the closed pyramid formed by the oral plates is the upper portion of
the larval body, in the centre of which the opening of the mouth is formed. The rest of
the space above the circular lip and beneath the oral pyramid is occupied by the
tentacular vestibule. This, according to Goette,' is derived from the left peritoneal tulje,
and contains the fifteen first formed tentacles which are borne on the water-vasculai-
ring. At a certain period of development the five valves of this oi'al pyramid graduall}-
separate so as to open the mouth to the exterior and allow of the protrusion of the
tentacles ; while the floor of the original tentacular vestibule, with the mouth in its
centre, becomes the peristome of the growing Crinoid. Five of the tentacles correspond
to the intervals between the oral valves ; and a conical projection, the commencement of
a ray, appears at the base of each of them. The growing rays are supported by the first
radial plates, which appear in the rapidly expanding equatorial portion of the body, i.e.,
the band of perisome between the upper edges of the basals and the lower edges of
the orals. As the rays grow the second radials appear between the bases of the orals,
and the equatorial band continues to increase in diameter. But the orals maintain their
original position round the mouth, so that they become completely separated from the
basals and radials by the equatorial perisome and are relatively carried inwards, while the
second radials project somewhat outwards. The diameter of the oral circlet continuall}'
decreases in proportion to that of the disk, which enlarges rapidly as new arm-joints are
added in succession. The orals are thus left as a circlet of five separate plates protecting
the peristome in the centre of the upper surface of the disk ; and the ambulacral grooves
extend outwards between the bases of the orals, as the growing rays carry the fii'st
formed tentacles away from the water-vascular ring.

In all the Pentacrinidte, and also in the Comatulaj, with the single exception of
Thawnatocrinus (PL LVI. fig. 5), the orals eventually undergo a process of resorption,
which commences in the hitter case before the young Comatula detaches itself from the
larval stem, so that no traces of the orals are to Ije found in the adult. Neither are
there any in the adult Bathycrinus aldrichiamis (PL VII. fig. 3), nor even in the young
Bathycrinus gracilis (PL Villa, fig. 1) ; though according to the observations of
Dauielssen and Koren they would seem to be present in Bathycrinus carjyenteri, liut in ;i

1 Vergleichende Entwickehingsgeschichte der Comatula Mediterraiiea, Archir f. mikroal: Awil., V,d. xii. y. G21

72 THE VOYAGE OF H.M.S. CHALLENGER.

reduced aud partially resorbed conditiou.' In Bhizocrinus, Hyocrinus, Holopus, and
Thaumatocrimis they persist through life, and in each case present a different stage of
development.

The orals of Holojms retain their embryonic position, and are scarcely separated at all
from the first radials, coming into close relation with the inner faces of these plates, while
the arms are altogether above and outside them (PL III. fig. 2). But in Hyocrimis
(PL Vc. fig. 6, O; PL VI. figs. 1-4) and also in Thaumatocnnm (PL LVI. fig. 5),
though still relatively large, they are separated from the edges of the radials by a
marginal zone of perisome which is paved with closely-set plates, and occupies about one-
fifth of the total diameter of the disk.

The orals appear to be unequally developed in the two living species of Rhizocrinus.
In Rhizocrinus lofotensis they would seem either to undergo some amount of resorption,
or else to remain in a comparatively undeveloped condition. For they are figured and
described by Sars ^ as minute valvule-like plates which occupy the central ends of the
triangular interpalmar fields on the disk ; so that there is a comparatively large amount
of perisome between their bases and the edge of the disk, just as there is in many young
Comatulje after separation from the stem. In Rhizocrinus raivsoni, however, they are
relatively larger, and their bases approach more closely to the lower brachials, from which
they are only separated by a narrow band of perisome (PL X. figs. 7, 20).

Under these circumstances, therefore, it is hardly to be expected that the orals
should 1)0 preserved in the fossil species of Rhizocrinus ; for as they are only united
to the calyx by membrane, they would naturally become separated from it when the soft
parts were destroyed. In all the recent Comatulse, with the exception of the archaic
tj-pe Tliaumatocrinus (PL LVI. fig. 5), they are ]-esorbed before maturity is reached ;
and if this was not the case in the fossil species, they probably persisted in somewhat
the same form as in Rhizocrinus. Even in Holopus there is nfi very close connection
between the orals and the tubular cup (PI. III. figs. 1,2); and the type is so rare in the
fossil state, that specimens with the orals preserved are not likely to be found.

In the Palseocrinoids, however, the orals reached a greater development than in the
later Neocrinoids, resembling rather the solid plates of Holopus and Hyociinus (PL III.
fig. 2 ; PL Vc. fig. () ; PL VI. figs. 1-5) than the mere films of delicate limestone
network which represent them in Rhizocrinus and in the Comatulfe. It will, however, be
more advantageous to postpone the discussion of the nature and position of the oral
plates in the Palasocrinoids until the chapter which deals with the relation of these older
forms to the Neocrinoids.

1 Fra den norske Nordhavs-E.xiitditioii, Echinodei-mer, Nyt Mag. f. Naturvid., Bd. xxiii. ]\ '.).
- Crinoides vivnnts, p. 17, 1'gs. 40, 41, 85, s(i. >^9-91— ».

REPORT ON THE CRINOIDEA. 73

B. The Perisomatic Skeleton.

This name was given by Sir Wyville Thomson^ to " the basal and oral plates, the
anal plate, the iuterradial plates, and any other plates or spieula which may be developed
iu the perisome of the cup or disk." He pointed out that the plates of this system are
" essentially variable in number and arrangement ; most of the minor structural modi-
fications throughout the group depend upon the multiplication or suppression of plates
of this series. Even in the same species they are by no means constant," e.g., Antedon
rosacea. The nature of the basals and orals has been already discussed; and very little
need be said about the anal plate. For although this forms an essential part of the cup
of the Pentacrinoid larva of Comatula, and is of extreme importance in its palasontological
relations, yet it disappears soon after the termination of Pentacrinoid life, undero^oino-
exactly the same process of resorption as the orals have previously done. It is curious,
however, that there should be no special anal plate in Hijocrinus, which has such large
orals (PL VI. figs. 1-5), while it is also absent in the adult Rhizocrinus, and is perhaps
never developed at all ; for Sars figures a young individual only 25 mm. long in which
the first brachials are comparatively large and form a sort of pyi-amid, while the second
brachials are undeveloped, and he makes no mention whatever of an anal plate. ^ Whereas
in Antedon rosacea the anal plate appears soon after the second radials (which represent
the first brachials of Rhizocrinus) ; and it is relatively quite large by the time that the first
brachials are developed, forming a nearly complete circle together with the first radials,
between two of which it is intercalated.

The interradial plates are those minute disks or granules which occur in the substance
of the perisome uniting the rays and their subdivisions, and are sometimes difficult to
distinguish from the lowest joints of the pinnules. They were first detected in Antedon
milleri by J. S. Miller,^ who figured them as forming one " intercostal " between every
two second radials. This was probably due, as remarked by Dr. Carpenter,* to his having
only employed a low magnifying power in his examination of them. Miiller ' described
them as occurring in Pentacriniis asteria (PL XIII. fig. 1), and noticed their difi'erence from
the plates on the ventral surface of the disk which are pierced by the water-pores
(PL XVII. figs. 6, 10). They are very abundant in some species of the Comatulidte and
Pentacrinidse, uniting the rays and their lowest divisions very closely together ; while in
other t}'pes they may be whoUy or entii'ely absent in some individuals, and more or less
well developed in others. In fact, the same individual may have them in one or two of

1 Phil Tram., 1865, pp. 540, 541.

2 Crinoides vivants, p. 27. Tab. iv. fig. 95.

3 This is the Comatula fimhriata of Miller, which occurs in Milford Haven. See his Natural History of the Crinoidea,
Bristol, 1821, Frontispiece, fig. 2, G.

* Phil. Trans., 1866, p. 716. ' Bau des Pentacrinus, loc. cit., p. 49.

(ZOOL. CHALL. EXP. — PAET XXXII. — 1884.) 11 10

74 THE VOYAGE OF H.M.S. CHALLENGER.

the interradial angles but not in the others. They are, however, very well developed in
many of the fossil Neocrinoids, e.g., Ilarsupites, Ajnocrinus, and Extracrinus ; and
they often have a very solid and substantial appearance. Like the anal plate they are
most important in their palseontological relations ; as is also the single calyx-interradial of
Tliaumatocrinus {J*l. LVI. figs. 1-5), which has been already described {ante, pp. 39-41).

The margins of the ambulacra of the disk, anns, and pinnules, and the interpaknar
areas of the disk, are rai'ely, if ever, perfectly free from any traces of calcareous structures.
Those of the ambidacra may take the form of simple short spicules which are almost
entirely limited to the marginal leaflets ; or they may be forked and branching spicules,
or rounded cribriform plates of variable size, which are movable and can either be
erected or closed down over the grooves. They are well developed in Hyocrinus,
Bathycrinus, and Rhizocrinus (PI. Vc. figs. 8-10, cp; PI. VI. figs. 1, 6 ; PL VII.
fig. 7 ; PL VIIL figs. 3, 5 ; PL Villa, fig. 1 ; PL IX. fig. 4 ; PL X. fig. 20). In
the first named genus they are sometimes separated from the dorsal skeleton by other
plates which will be noticed later (PL Vc. figs. 9, 10, sp) ; and the same is often the case
in the Pentacrinidaj and ComatuHdiB (PL XXXIII. fig. 1; PL XLVII. figs. 11, 13;
PL XLIX. figs. 6, 7 ; PL LI. figs. 11, 12 ; PL LII. figs. 5, 6 ; PL LIV. figs. 4, 6-9).

These covering plates of the ambulacra of Pentacrinus were termed " Saumpliittchen "
by MliUer, to distinguish them from the uncalcified marginal leaflets of the ambulacra or
" Saumliippchen," which correspond to them in most Comatulas ;^ while Sars,^ doubting
the mobility of these parts in Pentacrinus and Comatula, gave the name " lamelles du
sillon " to the large, oval, and movable plates which border the ambulacra of Rhizocrinus.
All three structures, however, are of essentially the same nature. The covering plates of
the stalked Crinoids are abundantly represented in many of the Comatulje, always,
however, resting upon a more or less developed " side plate " as in the Pentacriuidce ;
and the fact that they are merely an extensive development of the limestone rods and
networks in the perisome bordering the ambulacra is evident when all the intermediate
stages are examined. All the Pentacrinidte have plated ambulacra ; but this is by no
means the case in the Comatulse, especially in Actinometra ; though individual species of
Antedon have a relatively larger and more substantial ambulacral skeleton than any
Pentacrinus or Metacrinus (compare PL XXVII. figs. 4, 11, 12; PL XXXIII. fig. 1;
PL XLVII. figs. 11, 13 ; PL LI. figs. 11, 12 ; PL LIL figs. 5, 6 ; PL LIV. figs. 4, 6-9).

The covering plates which border the ambulacra on the arms and pinnules of
Pentacrinus asteria were briefly mentioned by JMliUer.' But he described them as
resting upon the joints of the skeleton, which is not always the case ; and in fact, two
pages further on he spoke of the plates which are situated on the j)erisome at the sides
of the arms and pinnules, just as on that of the disk ; while he does not seem to have

' Ueber deu Bau der Echinodermen, Ahluindl. d. Berlin A':ad., Jakrg. 1853, p. 57 (of separate copy).
^ Crinoides vivants, p. 24. ^ Bau dea Pentacriuus, p. 46.

KEPORT ON THE CRINOIDEA. 75

noticed the great differences of form, size, and regularity of aiTangement between the
ambiilacral plates of the arms and pinnules respectively. But he pointed out that the
covering plates of the disk-ambulacra rest upon other plates which he called " side plates,"
and that both are distinguished from the general anambulacral plating of the disk by the
absence of water-pores.^ It is difficult to individualise these plates when looking at the
disk from above, as they are so irregularly arranged (PI. XYII. fig. 6 ; PI. XXXIX. fig. 2 ;
PL XLIII. fig. 3 ; PI. L. fig. 2) ; but they are more easily distinguished in a cross section
of an ambulacrum (PI. LIV. fig. 11; PI. LVII. fig. 3, sp.).

Miiller further mentioned a series of median subambulacral plates as lying beneath
the food-groove and water-vessel, which he believed to rest in a furrow along their upper
surface ; * and he described a series of ambulacral pores between the median row and the
side plates, which might be related to the tentacles, and possibly served for the passage
of vessels connecting these organs with ampullae. He had previously figured some plates
as underlying the sides of the food-groove, with j)ores in or between them, which he spoke
of as "Oefiuungen des Tentakelcanals in die Tentakeln der TentakeLrinne." ' But it is
difficult to make out whether they are identical with those which he subsequently
described and figured as ambulacral pores. "*

In reality, however, there are no pores of this kind beneath the ambulacra of the
disk ; and there are no large ampuUse connected with the tentacles as there are with the
tube-feet of the Stellerids. But there is often a large amount of calcareous tissue beneath
the water- vessels of both disk and arms, which takes the form of more or less regular
plates (PI. LIV. fig. 11 ; PI. LVII. fig. 4, sub; PI. LXIL). They have no definite arrange-
ment, however, and are practically only a portion of the general limestone plating beneath
the upper surface of the disk. Although therefore, owing to their subambulacral position,
they are generally equivalent to the rotulte of the Urchins, the lancet-plates of the Blastoids,
and the radial pieces in the oral ring of Holothurians, I do not think that they quite deserve
the morphological importance which was attributed to them by Miiller. It is possible
that the series of plates which were discovered by Prof. Huxley and described by
Billinss ^ as formins: an elongated arch beneath the subtegminal ambulacra of Actino-
criims rugosus may be true subamljulacral plates. But from the descriptions of them
which are given by Meek and Worthen," and also by Wachsmuth and Springer,' I
am rather inclined to think that they may be the adambulacral or side plates (PL LVII.
fig. 3, sp.).

Besides going somewhat fully into the nature of the ambulacral skeleton in Penta-
crinus asteria, Miiller drew attention, as his predecessors had done, to the plates on the

1 Bau der Echuiodernien, p. 58. " Ibid-, pp. 57, 58, Taf. vi. figs. 7, 9, d.

3 Bau des Pentacrinus, p. 70, Taf. ii. fig. 14. * Bau der Echinodermen, pp. 58, 6.3, Taf. vi. fig. 7, e.

5 On the Cystidea- of the Lower Silurian Bocks of Canada, Geol. Surv. of Canada, Decade iii. p. 27.

^ Xotes on the Structure and Habits of the Paleozoic Criuoidea, Palitontology of Illinois, voL v. p. 331.

' Revision, ^rt. ii. p. 28.

7G THE VOYAGE OF H.M.S. CHALLENGER.

iuteqialmar areas of the yentral perisome. They are coutiuuous over the edge of the
disk with the perisomatic plates uniting the rays ; and he came to the conchxsion that
while many of the ventral plates are perforated by water-pores which lead downwards
into the body-cavity, these openings are never found in the iuterradii at the sides of the
disk.^ He termed them " anambulacral " to distinguish them from the " ambulacralen
Kelchporen fUr Fiisschen ; " ^ and this name has been conveniently extended both to the
plates which they pierce, and also to the remaining imperforate plates of the interpalmar
areas. Owing to the large size of the oral plates in Hyocrinus, which are themselves
pierced by water-pores (PI. Vc. fig. 6, u'p), the number of these anambulacral plates on
the disk is very small. But in a large Pentacriniis or Comatula they may be very
extensively developed, and the pores are occasionally to be found on the sides of the disk
between the rays (Ph XVIT. figs. 6, 10; PI. XXVI. figs. 1, 2; PL XXXIII. fig. 7;
PL XXXIV. fig.;2 ; PL XXXIX. fig. 2. PL LVII. figs. 1, 3, 4 ; PL LIX. figs. 2, 4, Q,—v:p ;
PL LXIL).

At the edge of the disk the anambulacral plates of its upper surface pass gradually
downwards into the interradials, which are developed in the perisome uniting the rays ;
so that in some species both of Comatula and Pentaci'inus the visceral mass is every-
where protected by a continuous armour of plates. Many of the fossil Pentacrinidse and
also some species of Apiocrinus show signs of the same structure. It is especially well-
marked in the Liassic genus Extracrlnus, which had a very large and thickly plated
" ventral sac." In fact the disk of these Crinoids seems to have borne stouter plates than
that of many of the Palaeozoic Ichthyocrinidas ; and I do not understand how the ventral
disk of this famUy, which is described by Wachsmuth' "as composed of a more or less
soft or scaly integument, jdelding to motion in the body and arms," can be compared
to anything else than the oral surface of a recent Crinoid, with which, however,
Wachsmuth says that it " cannot in the remotest degree be homologised.* "

I have not seen any good disk of Pentacrinus asteria; but, judging from the
condition of its peripheral part in the specimen figured by Mullcr, I imagine it to have
been covered with a continuous pavement of tolerably large plates. This is also the
case in Pentacrmus wyville-tliomsoni (PL XVII. fig. 6). The interpalmar areas are
covered with a very closely-fitting pavement of polygonal plates, the largest of which
may be pierced by four or five water-j)ores. The anal tube, which is plated almost up to
its summit, occupies the greater part of the corresponding interradius ; but the anambu-
lacral plates which are between it and the mouth (in the specimen figured) are smaller
than elsewhere, and less distinctly defined. In fact they look as if they had fused into
two irregularly- shaped plates which abut directly on the peristome. A similar fusion of
small plates appears to have taken place on the anal tube of the Mctacrinus nodosus

' Bail des Pentacrinus, p. 49. • Bau der Echinodernien, p. 63.

^ Kevrision, pt.i. p. 31. ' Amer. Journ. Sci, vol. xiv. p. 190.

REPORT ON THE CRINOIDEA. 77

represented in PI. L. fig. 2. It seems at first sight to be perfectly bare, but careful
examination proves it to be covered by very closely set small plates with ill defined
boundaries.

The ambulacra of Pentacrinus u'ljvilh-thomsoni form rather prominent ridges, which
are composed of four irregular rows of plates. The plates of the two inner rows arc
somewhat elongated transversely and generally closed down over the grooves, representing
the covering plates of the pinnule ambulacra.

A well plated disk also occurs in Pentacrinus alternicirrus (PI. XXVI. figs. 1,2);
but the ambulacra are less heavily plated than in Pentacrimis wyville-tJwmsoni, and are
therefore not so readily distinguished from the anambulacral plates. These are
sometimes larger than in the Atlantic species, and are pierced by abundant water-pores
which are not shown in the figure ; but they do not always fit quite closely together, so
that gaps of bare perisome are visible here and there. As in Pentacrinus wyville-
thomsoni, the plates are generally larger in the anal interradius than elsewhere.

A disk of Pentacrinus naresianus was drawn for Sir Wyville Thomson by Mr. Black
(PI. XXX. fig. 2) ; but it seems to have been mislaid or else cut into sections, for it has
not come into my hands. So far as can be judged from the figure, the anambulacral
plates were small ; while the ambulacra appear to be well-defined ridges and to come
into close union around a very small peristome, which is thus entirely concealed by the
apposition of their large covering plates.

I have not seen a disk either of Pentacrinus maclearanus or of Pentacrinus hlakei ;
but in Pentacrinus miilleri and Pentacrinus decorus it is far from being as completely
plated as in the species already noticed (PI. XVII. fig. 10 ; PI. XXXIV. fig. 2). For the
anambulacral plates are generally isolated and not in contact with their fellows. They
are small and numerous in Pentacrinus miilleri; but in Pentacrimis decorus they are
fewer in number and comparatively large, some of them containing as many as twenty
water-pores.^

There are about four irregular rows of plates on the ambulacra, the inner ones
being elongated, and sometimes standing up rather prominently at the edges of the
grooves.

The relation of the food-groove to the arm-joints varies greatly in the difierent species
of Pentacrinus, so that mere fragments of the arms can be identified by the characters
of their ambulacra, quite apart from any peculiarities of their arm- and pinnule-
joints.

The middle line of the upper surface on each joint of the brachial skeleton is occupied

l)y a groove of variable depth and width, to which MiiUer gave the name " ai-m-groove "

(PI. XVII. figs. 1, 4, 7, 8, 9). It is bordered on each side by the more or less prominent

muscle plates of the successive joints ; and the vascular structures which are partially

' These are omitted in the figure, and the plates are dra\\Ti too close together.

78 THE VOYAGE OF H.M.S. CHALLENGEK.

contained in it are continued from one joint to another between the two large muscular
bundles that unite them. In Antedon eschriehti and in many other Comatulte, more
especially those belonging to the genus Actinometra, this arm-groove merely lodges the
lowest part of the coeliac canal ; while the genital cord, with the water- vascular and
blood -vascular trunks and the ambulacral epithelium, are all situated above the arm-
groove, and separated from it by a variable amount of intervening perisome, so that little
more than half the vertical height of the arm is due to its dorsal skeleton. The lower
parts of the arms in Metacrinus murrayi present a somewhat similar condition
(PL XLI. fig. 13).

In other Comatulse, however, and in Pentacrinus a great part, sometimes even the
whole, of the soft parts of the arm are lodged within the groove on the upper surface
of the skeleton (PI. XVII. figs. 1,4; PI. XXVII. fig. 6) ; and there is no substantial
ventral perisome in the ordinary sense of the word, or it is reduced to a mere film,
sometimes thinly plated, which covers up the muscular bundles. In many species, and
especially in the small deep-sea Comatulse, this layer of perisome is excessively thin and
transparent, so that the food-groove appears to rest upon and between the muscular
bundles. In some of the tropical Antedons, however, it bears a continuation of the
anambulacral plates of the disk, and this is also the case in Pentacrinus unjville-tliomsoni,
Pentacrinus cdternicirrus, Pentacrinus naresianus, and Pentacrinus hlakei (PL XVIL
fig. 4; PL XXVIL figs. 6, 13; PL XXXIII. fig. 3). The third of these, Pentacrinus
naresianus, has the greatest development of this plated perisome on the arms (PL XXVIL
fig. 13). It is continuous from one pinnule socket to the next on the same side, so as
to cover in both the muscular bundles and also the upper surface of the intervening
arm-joint ; and the ambulacra are thus distinctly above and outside the arm-groove.
They are bordered by large oval covering plates which overlap alternately from opposite
sides, and are continued on to the pinnules (PL XXVIL figs. 11, 12). These plates do
not rest directly upon the pinnule-joints, but are separated from them by a thin
limestone band which is a continuation of the lateral plating of the arm. It does not,
however, exhibit any differentiation into side plates, though its edges are cut out into
alternate teeth and notches (PL XXVIL fig. 11). The latter are occupied by the
tentacles, but can be closed, or nearly so, by the covering plates which rest on the
intervening teeth.

In the arms of Pentacrinus hlakei (PL XXXIII. fig. 3) the sides of the joints bend
inwards towards the middle line more than they do in Pentacrinus naresianus, so that
the arm-groove is narrower, and the ambulacrum practically coincides with it instead
of lying above it. It is bordered by long plates which are really the covering plates
fused -n-ith the side plates. When they pass on to the j^innules the former become
more difi"erentiated, but the latter lose their individuality and become parts of a con-
tinuous denticulated band just as in Pentacrinihs naresianus (PL XXXIII. fig. 1).

REPORT ON THE CRINOIDEA. 79

A somewhat different type of ambulacrum is presented by two other Caribbean
species, Pentacrlnus asteria and Pentacrinus decorus. The arm-groove of the former
is rehitivcly wide and the proximal j)arts of the ambulacra are distinctly above it,
though they gradually sink do^\^a into it as they get farther from the disk ; as long as
the rays continue to divide their ambulacra and those of their pinnules are covered
by an UTegular double row of large plates (PI. XIII. fig. IG ; PI. XVII. fig. 7). After
the last bifurcation these plates become smaller and more regularly arranged, so that
they take the form of oblong covering plates with rounded ends which stand up at the
sides of the groove (PL XVII. fig. 8). They do not, however, extend uninterruptedly
along each side of the groove, but are arranged in a series of linear groups between the
successive pinnules of either side, so that they alternate in position on the two sides
of the arms successively. They are largest and best developed at the base of a pinnule,
where its ambulacrum comes off from that of the arm, and from this point they diminish
gradually in size towards the disk until the base of the next pinnule is reached, when
a fi'csh set appears upon the proximal edge of its ambulacrum.

Thus, then, the covering plates which pass on to the pinnule-ambulacrum from that
of the arm are at first limited to its proximal or outer side only. But a second
set soon appears on the inner side of the ambulacrum ^ {i.e., that next the arm), and their
outer ends gradually become more and more rounded until they present the appearance
shown in PL XIII. fig. 15. Their bases are aU fused into a narrow band of limestone
which rests on the pinnide-joint and represents the side plates that are better developed
in other species ; while the rounded outer portions represent the covering plates
proper, which alternate with one another from opposite sides, so as to leave a series of
openings through which the tentacles are extended.

The lower portions of the ambulacra of Pentacrinus decorus are essentially like those
of Pentacrinus asteria, except that they sink more deeply into the arm-groove, while the
plates covering them are smaller and far less regularly arranged (PL XXXIII. fig. 6).
But the muscle plates of the successive arm-joints fit less closely together than in most
other species of Pentacrinus, so that the muscular bundles are long and also visible
externally ; for they are not covered in by plated perisome as in the allied Pentacrinus
hlakei (PL XXXIII. fig. 3). In the middle and outer parts of the arms the ambulacra
are generally like those of Pentacrinus asteria, though not so open (PL XXXIII. fig. 4) ;
for the groups of plates which protect the bases of the pinnule-ambulacra overlap
somewhat closely from opposite sides, while theii- parts are more distinctly differentiated.
Farther out on the pinnules the segmentation of the lateral limestone band is sometimes
carried so far that the side plates can be distinctly indi\T.dualised ; but there is a good
deal of variation in this respect (PL XXXVII. figs. 23, 24).

A third type of arm, with a very narrow median groove to which the ambulacrum is

1 Compare PI. XVII. fig. 3 ; PI. XLVII. fig. 11.

80 THE VOYAGE OF H.M.S. CHALLENGER.

almost entirely limited, is presented by the four species PentcLcrinus miilleri, Pentacrinus
maclearanus, Pentacrinus wyville-thoTnsoni, and Pentacrinus alter nicirr us. The nearest
approach to Pentacnnus asteria is to be found in Pentacrinus miilleri (PI. XVII. fig. 9),
as might be expected for various reasons. The arm-groove is narrower, but the covering
jilates which rest on its edges pass up on to the pinnules alternately from opposite sides
very much as in Pentacrinus asteria ; though the successive groups do not overlap one
another so much as in that type, and there is more differentiation of the side plates upon
the pinnules (PL XV. figs. 7, 8). In Pentacnnus wyville-thomsoni the arm-groove is still
narrower, and the ambulacrum almost entirely withdrawn into it (PI. XVII. fig. 4). The
plates bordering it are smaller and more irregular than in Pentacrinus mulleri, and more
distinctly limited to the pinnule-bearing side of the arm ; while the intervals between
the joints are larger and covered by small irregular plates as in Pentacrinus naresiamis
and Pentacrinus hlalcei. The plating of the pinnules is limited at first to their outer
sides (PL XVII. fig. 3) ; but it eventually appears on the inner sides as weU, and
becomes difi'erentiated into covering plates resting on a limestone band which is sometimes
imperfectly separable into side plates (PL XVII. fig. 2).

A further reduction in the width of the arm-groove and in the size of the plates at its
edges appears in Pentacrinus alternicirrus (PL XXVII. fig. 6). The intervals between
successive joints which are occuj)ied by the muscular bundles are larger than in Penta-
crinus tvyville-thomsoni, and are more distinctly pLited. The rudimentary covering
plates are limited to the origins of the pinnule-ambulacra, and a short distance behind
them ; so that between every two pinnule-ambulacra of one side there is a short space of
unprotected arm-groove. As in Pentacrinus loyville'thomsoni, the bases of the pinnule-
amluilacra are plated on the outer side only, and in their distal portions the lateral band
on which the covering plates rest is not divided into side plates (PL XXVII. fig. 5).

Lastly, in Pentacrinus maclearanus the arm-groove is extraordinarily narrow, and
bounded by little else than the broad plate-like upper surfaces of the component joints
(PL XVII. fig. 1), while the covering plates are almost entirely limited to the pinnules
(PL XVI. figs. 2, 3). They are relatively small, and the lateral band supporting them,
though broad at first, soon narrows away considerably.

The disk of Metacrinus presents much the same variations in the extent to which it
is plated as that of Pentacrinus does. In Metacrinus nohilis (PL XLIII. fig. 3) there is
a tolerably continuous pavement with well defined ambulacral ridges. These are bounded
by about four rows of plates, those of the two inner rows being transversely elongated,
and alternating with one another. In other types the anambulacral plates are more
isolated as in Pentacrinus decorus, being more closely set, however, along the sides of the
ambulacra, which are covered by longish plates. This is the case in Metacr^inus angulatus
(PL XXXIX. fig. 2), Metacrinus cingulatus, and Metacrinus nodosus (PL L. fig. 2).
The scattered arrangement of the anambulacral plates is not well represented in the last

REPORT ON THE CRINOIDEA. 81

mentioned figure ; and it is too small to show the very numerous water-pores piercing
the plates, some of which have as many as twelve or fifteen openings. They are much
less abundant in Metacrinus cingulatus and Metacrinus angulatus (PI. XXXIX. fig. 2),
while the plates- are also smaller.

The disk of Metacrinus differs from that of Pentac^Tnus in the greater irregularity of
its ambulacra, the branches of which proceeding to the large lower pinnules often come
off directly from the primary groove-trunks, or even from the peristome itself This is
especially well shown in Metacrinus angulatus and Metacrinus nobilis (PI. XXXIX.
fig. 2 ; PI. XLIII. fig. 3).

Another point of difference is the relatively larger size of the anal tube in Metacrinus,
which is well shown in Metacrinus nodosus (PI. L. fig. 2). It may occupy the whole of
the interpaknar area in which it lies, and is often considerably inflated, so as to be a
somewhat prominent object on the surface of the disk. It is erroneously represented as
perfectly bare in the figure of Metacrinus nodosus, and this actually seems to be the case at
first sight. Closer examination shows, however, that its apparent bareness is really due
to the smoothness and very close approximation of the plates which cover it.' They are
thinner than the corresponding plates in the other interpalmar areas, and form a smooth
continuous pavement over the whole of the lower part of the tube, becoming more
nodular and. irregular towards the top. The whole appearance of the anal tube in this
species forcibly recalls Buckland's well known figure of the " abdominal integument " of
Extracrimis briareus.' There are indications of this close pavement on the anal tube of
Metacrinus nobilis (PI. XLIII. fig. 3) ; and it is better sho'^vn in a curious specimen of
Metaxirinus angulatus (PL XXXIX. fig. 2), which has a smaller supplementary anal tube
by the side of the larger one.

The plates of the pinnule-ambulacra in Metacrinus are better differentiated on the
whole than those of Pentacrinus. For in the outer parts of the pinnules, at an}- rate, the
covering plates rest upon a row of distinct side plates (PL XLVII. fig. 11; PL LI.
figs. 11, 12 ; PL LII. figs. 5, 6), and not upon an almost undivided band of limestone as
in most species oi Pentacrinus (PL XIII. fig. 15 ; PL XXXVII. fig. 23).

In the lower parts of the rays and arms the anambulacral plating of the disk extends
outwards at the sides of the ambulacra, in which the arrangement of plates is confused and
indefinite (PL XLI. fig. 13). Farther out, however, where the zig-zag course of the
ambulacrum (still distinctly above the arm-groove) is more marked, and the ambulacra!
plates less abundant, the elongated shape of the plates immediately bordering the groove
is more distinctly visible (PL XLI. fig. 4). In most species their extremities gradually
become bifid, as is well shown in Metacrinus angulatus and Metacrinus murrayi
(PL XXXIX. fig. 13 ; PL XLI. fig. 14). Both of these, especially the former, have the

' See the remarks on the disk of Pentacrinus icyville-thomsoni, ante, p. 76.
2 Geology and llineralogy, voh i. p. 439 ; voh ii. pi. 51. fig. 2.

(ZOOL. CHALL. EXP. PART XXXIT. 1884.) li 11

82 THE VOYAGE OF H.M.S. CHALLENGER.

ambulacra well above the level of the arm-grooves ; and there are uo other plates
on the arm than the covering plates which really belong to the pinnule-bases, while
the muscular bundles are freely visible at the sides of the ambulacra. But in other species,
such as Metacrinus nohilis (PI. XLI. fig. 11), the food-groove is more concealed within
the arm-groove, and the forked covering plates are less abundant at the pinnule-bases.

Farther out on the pinnule, the pro.ximal half of the fork becomes gradually less and
less prominent ; and it is eventually absorbed into the basal part of the plate, which thus
represents a side plate ; while the distal half of the fork, becoming larger and better
defined, separates itself off as a rounded covering plate (PI. XLI. fig. 12; PI. XLVII. fig. 11).

The branches of the ambulacra which pass on to the massive basal joints of the
prismatic lower pinnules are usually but little plated, as is the case in Pentacnnns asteria
(PL XIII. fig. 16; PL XLI. figs. 4, 12, 13 ; PL XLVIL fig. 13 ; PL LL fig. 12). But
beyond the first two joints the plating reappears ; and the four rows of plates become
gradually developed from the irregular plates at the sides of the groove, which come to
assume a definite form and break up into covering plates and iU defined side plates.

The gradual differentiation of side and covering plates upon the pinnules from the
single forked plates at the sides of the brachial ambulacra takes place in this way in most
species of Metacrinus; but the four rows are never so distinctly separable as in the
Comatulse (PL LIV. figs. 4, 6-9).

A slight variation of this process occurs in Metacrinus costatus (PL XLVII. fig. 13 ;
PL XLIX. figs. 6, 7) ; while Metacrinus murrayi and Metacrinus nodosits (PL XLI.
fig. 12 ; PL LI. fig. 12) are the intermediate links between this species and the other types of
Metacrinus. The bases of the pinnule-ambulacra just beyond the wide lower joints are
bordered by a series of rounded plates, which are deeply hollowed in the centre so that
their edges stand up rather prominently. The first eight or ten of these are attached to
the pinnule-joints on each side by a continuous band of limestone. This gradually becomes
absorbed into the raised proximal edges of the rounded plates so as to form the side plates ;
while the distal halves eventually separate themselves off as the covering plates (PL XLVII.
fig. 13). The side plates only become properly differentiated in the outer parts of the
lower pinnules, and in the later pinnules on the arms (PL XLIX. figs. 6, 7) ; but they
retain a more or less prominent backward process, which is the remains of the raised
hinder edge of the rounded plates on the j^roximal parts of the ambulacra.

Although there are no side plates on the arms and pinnule-bases of Hijocrinus, yet
they are large and well developed on the enlarged portions of the pinnules which contain
the genital glands (PL Vc. fig. 10, sp). The proximal ones, taking the place of
numerous small anambulacral plates, are smaller than their successors, which considerably
increase the depth of the body-cavity within the pinnule. Distally, the side plates
gradually diminish in size and finally disapj)ear altogether, so that the covering plates
come to rest directly on the edges of the pinnule-joints (PL Vc. figs. 8, 9).

REPORT ON THE CRIXOIDEA. 83

In man}' of the tropical Comatulse the pinnule-ambulacra are fully as well or even
better plated than those of the Pcntacrinidai. Antedon acoela and Antedon iiuecfialiii,^
both from Stations where Pentacrinida) are abundant, are good instances of this (PL LIY.
figs. 4, 6-9). The differentiation of side plates and covering plates is more complete ; and
the plates themselves are not only larger relatively to the pinnule-joints, but also absolutely
so. When the covering plates are erected and the groove opened, as shown in Antedon
incequcdis (PI. LIV. fig. 8), the tentacles are extended between them. But the tentacles
can be completely retracted and the plates closed down so as to convert the grooves into
tunnels, as shown in Antedon acoela (PL LIV. fig. 4).

In the distal edge of each of the side plates is a small rounded notch, so that there
is a series of gaps along each side of the ambulacrum, one between Qwavy two plates.
These lodge the problematical " sacculi " which are so characteristic of the g&mx?, Antedon.

In those Comatulas with plated pinnules which have short and rounded genital
glands, instead of the long fusiform structures characteristic of Antedon eschrichti, Hyo-
crinus, and the Bourgueticrinidaj, there is sometimes a curious modification of the
perisomatic skeleton on the genital pinnules. The enlarged part of the pinnule is
protected by a very strongly developed anambulacral plating, which is much more
regular and closely set than that of the disk and arm-bases, especially iu the case of
Antedon acoela (PL LIV. figs. 1-3). Resting upon the four or five middle joints of the
short pinnules there is a double row of large plates, which are rectangular at the base
but somewhat more irregular in shape at their upper ends. There are generaUy five or
six plates in each row ; but those of the two sides have no fixed relative positions,
sometimes corresponding exactly, and sometimes alternating as exactly. They have
the same protective function and very much the same appearance as the large side plates
of Hyocnmcs (PL Vc. figs. 9, 10, sij), but difier from them in two ways. In other
species of Antedon, such as Antedon angusticalyx and Antedon incerta (PL 1A\.
figs. 5, 6), these protecting plates are smaller and more irregular than in Antedon acoela;
while in Antedon incerta the two rows are separated by the ambulacrum with its well
developed side and covering plates. But in Antedon angusticalyx and iu Antedon acoela
these swollen lower pinnules receive no branches from the brachial food-groove, just as in
many species of Actinometra ; and the anambulacral plates covering the genital glands
consequently meet one another in the medio-ventral line of the pinnule above the gland
within. The sacculi, however, which lie at the sides of the ambulacra may extend on
to these grooveless pinnules, and occupy small holes between the large protecting plates ;
while in the outer joints of the pinnules, beyond the glands, the sacculi occupy the
median gTOOvc on the upper surface of the skeleton, as is well shown iu Antedon angusti-
calyx (PL LIV. fig. 5).

a''
1 The specific formula of this type is— A . 3(1 . or 2)) . ) ^ .

84 THE VOYAGE OF H.M.S, CHALLENGER.

It is curious that these protecting anambulacral plates ou the genital pinnules of
Antedon acoela should be so largely developed/ while those which cover the inter-
palmar areas of the disk are comparatively small and irregular in character (PL LV.
fig. 5).

In many Comatulce, however, the disk is very closely plated, both in the ambulacral
and in the interambulacral areas. The plates of the latter are mostly small, and rarely
pierced by more than one water-pore ; while the ambulacra are generally marked by an
irregular double row of transversely oblong plates, as in Antedon angusticalyx (PI. LV.
fig. 6). But these are sometimes barely distinguishable from the anambidacral plates,
and the whole set encroach very much upon the peristome, so that it is scarcely visible in
the dry state, as shown in Antedon basicurva^ (PL LV. fig. 7). This is still more
marked in Antedon acoela (PL LV. fig. 5). Both the ambulacral and the anambulacral
plates are palisade-like in form, as the former are in Pentacrinus decorus (PL XXXIV.
fig. 2); and they are very much crowded, so that the course of the food-grooves can only
be made out with difliculty even in spirit specimens, while the peristome is frequently
entirely invisible.

The disks of the three Antedon species just mentioned are very much incised between
the ambulacra, so that they are markedly stellate in form. The arrangement of the
coiled digestive tube is consequently much less complex than in large disks Like those of
Antedon eschrichti or of Pentacrinus, w^hich have the rays united by perisome so as to
increase the capacity of the cup ; while the interradial spaces are filled up with connec-
tive tissue which supports extensions of the digestive tube.

In some species of Antedon with an incised disk the anambulacral plates are somewhat
squamous, with a tendency to overlap one another. This is the case, for example, on the
disks represented on PL LV. figs. 3, 4. They probably belong to Antedon nmltlradiata^'
having been dredged at Cape York in an isolated condition, together with entire indivi-
duals of this species. The edges of the interpalmar areas rise rather sharply towards the
ambulacra, which are marked by strong ridges with indications of a median groove visible
upon their upper surface. The food-groove beneath is really comparatively deep, with its
edges plated somewhat regularly and turned in towards one another. This is very
marked in the immediate neighbourhood of the peristome, which is thus comj)letely closed.

1 The position of the plates which protect the genital glands as regards the pinnule-joints, and the very regular
appearance which they frequently present, have led me to think that the so-called " rudimentary pinnules " of
Cijathocrinus loiic/imatms, Angelin, may possibly he of the same character, and not ambulacral side plates more largely
developed than usual (see chap. iv. pp. 62-66). A comparison of Angelin's figures of these structures as seen from the
side and from above (Tab. xsvi. figs. 4b, 4c), with the corresponding tows of the protecting plates in Antedon acoda
(PI. LIV. figs. 2, 3), shows a remarkable similarity in their number and general aiTangement. In the recent forni,
however, these plates are on the jiinnules, while those which they ajjpear to resemble in the Pala?ozoic Cyathocrinus
are on the arms. But as these bear no pinnules, and must therefore have themselves contained the genital glands, the
dift'erence between the two structures is not so great as it would seem to be at first sight.

2 The specific formula of this type is — A . 10 . — 3 ^Jie specific formula of this type is — A . R . 3 . 3 . — .

REPOllT ON THE CRINOIDEA. 85

It is CQncealed in the smaller specimen by the large and prominent anal tube which pro-
jects forwards over it. The original of Ilypononie sarsii, the " recent Cystidean," was
the disk of a plated Antedon, very probably of this species, Antedon midtiradiata.

Allusion has already been made to the frequency with which these disks are met with
in an isolated condition ; and their resemblance to the curious Paleeozoic forms Agela-
crinus, Lepidodiscus, Sec, is very striking. I know that Sir Wyville Thomson had a
suspicion whether these problematical organisms may not have been the separated
disks of some one or other of the numerous PalaBOcrinoids, as suggested by Loven and
Lutken.^

Deep ambulacral grooves with strongly plated sides are also met with on the disk of
Actinometra strota. This species is very common at Cape York, and its disk, which was
also obtained in an isolated condition, may be nearly bare, or plated very completely, as
is shown in PL LIV. fig. 10, and PI. LV. fig. 2. The whole of the large interpabnar area
in which the anal tube is situated is covered with more or less scaly plates, which become
stouter and more granular in the neighbourhood of the subcentral anal tube. The sides
of the deep ambulacra are bounded by numerous smaller plates without any definite
arrangement. But they are strictly limited to the disk, not extending on to the arms.
The large size of this armature, relatively to the tentacles and the ambulacral groove proper,
is well shown in the cross-section represented in PI. LIV. fig. 11. Much of it extends
beneath the water-vessels, and corresponds to what MiiUcr called the subambulacral
plates of Pentacrinus^ (PI. LXIL).

Actinometra juJcesi is another species which is common at Cape York. The large anal
area is often beset with numerous irregular plates, many of which bear nodules of variable
size (PI. LV. fig. 1). They are smaller on the base and sides of the anal tube; and
there are few or none in the small interpalmar spaces between the edge of the disk and
the circumferential ambulacra, which are themselves devoid of supporting plates,
though deep like those of Actinometra strota.

Some species, both of Antedon and Actinometra, have the ventral perisome of
disk and arms entirely devoid of any continuous plating ; though this may be strongly
developed between the lower divisons of the rays, sometimes extending up to the level of
the third axillary.

C. The Visceral Skeleton.

I use the term " visceral skeleton" to denote the numerous spicules and networks of
limestone w'hich occur more or less plentifully in the bands of connective tissue that
traverse the visceral mass of the Comatulse. It also includes the more or less regular
plates, often quite well defined, which occur loithin the disk of Pentacrinus. They are

1 Canad. Nat., 1869, p. 268. ^ Bau der Echinodermen, pp. 57, 58, Taf. \\. figs. 7, 9, rf.

f^() THE VOYAGE OF H.M.S. CHALLENGER.

formed, like the anambulacral plates, of a calcareous network interpenetrated l)y an
organic basis, Avliich is of the same nature as in the joints of the rays and arms (Pis.
LVIL, LXIL).

The simple spicules and thin networks of limestone which occur in the less heavily
plated disks are especially abundant in the visceral layer of the peritoneum. This is well
seen in those Crinoids, such as Antedon rosacea^ and Actinometra strata, in which there is
but little connection between this visceral layer and the parietal one lining the interior of
the cup, so that the entire visceral mass is readily detached from the calyx.

Unlike that which lines the cup, the oral perisome is usually very closely adherent
to the visceral mass, and cannot be separated from it without some trouble. The
peritoneal covering of the latter also contains limestone deposits, so that sections
through the upper part of the disk show two layers of plates and spicules. The
upper one is in the perisome itself, and belongs to the anamlDulacral system ; while
the lower belongs to the upper surface of the visceral mass. These lower plates were
described and figured liy Mliller.^ Together with those of the sides and lower
parts of the visceral mass they seem to be the modern representatives of the
so-called " intervisceral plexus " which lined the cup of the Actinocrinidse. Wachsmuth ^
has pointed out that in some members of this family " almost thi' entire test is lined
with a delicate calcareous plexus or network. This lining is not in contact with the test
directly, Init connected with it by small partitions, producing innumerable little chambers,
which communicate with each other and with the visceral cavity." It rarely extends
lielow the level of the second radials, and passes gradually upwards into the plates,
coating the interpalmar areas on the ujjper surface of the disk below the vault. These,
which extend right up to the edge of the peristome, are of course anambulacral in
character, and it is not easy, any more than it is in Pentacrimis, to say where the one
set begins and the other ends. But so fcir as the lower part of the cup is concerned,
the intervisceral plexus of the Actinocrinidse is merely a greater development of the
limestone deposits in the visceral layer of the peritoneum of recent Crinoids.

All of these have more or less abundant plates and spicules in the connectiA-e tissue
which lies beneath the peristome and supports the lip, and also in that which unites
the coils of the digestive tube (PL LXIL). In the regular endocyclic Crinoids this
organ makes rather more than a single round turn upon itself (fig. 2 on p. 89) ; and it
is the aggxegation of limestone deposits upon its central side which forms the so-called
" columella," once regarded as a sand canal.

This supporting skeleton of the digestive tulse, lUce that enclosing the visceral mass,
was much better developed in the Actinocrinid?e than in recent forms. Occupying

' Liulwif,' (ojj. cit., p. 3311, Taf. xix.) has criven an excellent diagrammatic section of this type, in whicli this
point is well illustrated.

- Ban der Echinodermen, p. 58, Tat. vi. fi;;. 9,/. ^ Revision, part ii. p. 2().

REPORT ON THE CRINOIDEA. 87

the greater part of the interior of the cup, but not reaching down to the basals, is a
laro-e convoluted organ, which lias a general resemblance to the shell of a Bulla. It is
open at both ends, and its longer axis nearly coincides with that of the body of the
Crinoid. Wachsmuth and Springer' describe its wall as being " simple in all cases, very
delicate, and constructed of an extremely fine filigree work, which generally in the fossil
became thickly incrusted with siliceous matter on both sides." Hall, who was the first to
notice this orwan, made no suggestions respecting its nature. Meek and Worthen supposed
it to be a kind of framework supporting the coiled digestive tube ;^ while Wachsmuth and
Sprino-er suo-gest that it might be " an extensive plexus of blood-vessels surx'oundiug the
ambulacral (1) canal ; " and desiring that it " should receive a more appropriate name
than any yet given," they propose to call it the " oesophageal network."'

That it supported some of the intervisceral blood-vessels I have very little doubt;
but there is no reason to suppose that it actually represents the vessels themselves, which
would have passed through the meshes of its network (compare PI. LVII. fig. 5). As
a similar though less developed structure occurs in recent Crinoids, I see no reason to
douljt the correctness of Meek and Worthen's determination.

Neither do I think Wachsmuth and Springer's name a good one, for it implies that
the structure in question was connected with the oesophagus, and not with the rest of the
digestive tube. But as it is so large, relatively to the interior of the calyx, the remainder
of the digestive apparatus must in that case have been quite small, which is improbable
for many reasons.

According to Meek and Worthen,* " its slightly dilated upper end seems to stand
with its middle almost, but apparently not exactly, under the middle of the nearly
central proboscis of the vault ; while at the anterior side of its upper margin, and a httle
out from under the proboscis, it shows remains of a kind of thickened coUar, which we
found to l^e composed of minute calcareous pieces. From this there radiate five
ambulacra, composed of the same kind of minute pieces as the collar itself."

The thick collar was the edge of the peristome with its more or less regular supporting
plates as in any recent Crinoid. The mouth was placed within this peristomial space, and
the greater part of the convoluted organ would thus have lain altogether behind it. The
direction of its spiral is exactly the same as that of the digestive tube in Antedon or
Pentacrinus, as may be seen by comparing Dr. Carpenter's figure of the latter ^ (viewed
from above) and the " inferior end view " of the convoluted organ given by Meek and
Worthen." I believe that the gullet ran downwards and backwards as it does in
Pentacrinus ; and that the intestine, after following the convolutions of its support,
turned upwards again to end in the long anal tube, the so-called " proboscis."

1 Eevision, part ii. p. 35. '" Paleontology of Illinois, vol. v. p. 329.

3 Re\-ision, part ii. p. 35. * Patoontology of Illinois, vol. v. p. 331.

5 Proc. Boy. Soc. LoncL, vol. xxiv. pi. viii. fig. 1. ° Palaeontology of Illinois, vol. v. pi. ix. fig. 125.

88 THE YOYAGE OF H.M.S. CHALLENGER.

VI.— THE MINUTE ANATOMY OF THE DISK AND ARMS.

I do not propose to treat the subject of this chapter as fully as I have clone the
comparative morphology of the Crinoid skeleton. Much has been written about it lately,
and a general resume of this recent work, together with some independent observations
of my own, was published in the Quart. Journ. Micr. Sci. for April 1881.

Since that time I have been able largely to increase the range of my observations on
the anatomy of the Comatulae, and have also extended them to Pentacrinus, Bathycrinus,
and Rhizocrinus. The result has been that I am able to confirm in almost every respect
the admirable investigations of Ludwig on the minute anatomy of Antedon rosacea.^

On the other hand, there are some points, notably in the relations of the axial cords
of the skeleton, which were entii'ely overlooked by him; while he also omitted to describe
some remarkable peculiarities in the structure of the plexus of blood-vessels which is
situated in the lip. and is connected with the oral blood-vascular ring.

It is only fair to state, however, that my new observations upon the nervous and
vascular systems of this type owe their origin, in great measure, to my' having been aljle
to examine other species in which the peculiarities in question are much more developed
than thev are in Antedon rosacea.

A. The Geography of the Disk.

I find that it is most convenient, on the whole, to use the terms right and lift
precisely as in human anatomy. When the ventral surface is upwards, with the mouth
north, or pointing away from the observer, and the anus posterior, the right side of the
disk would be west in a map, and the left side east (figs. 2, 3). On the other hand,
when the dorsal surface of the skeleton is upwards, the anal area being, of course,
posterior, the eastern rays are those of the right side, and the left rays are in the west.
The same m^ethod applies to the arms, a. pinnule on the right side being east in a dorsal
view, and west in a ventral one.

B. The Digestive Tube.

Little need be said about the alimentary canal, the general course of which alters
but slightl}^ in the endocyclic Crinoids {Antedon, Pentacnnus, Rhizocrinus, &c.), though
it varies a good deal in the complexity of its cavity. Both in Antedon rosacea and in
Antedon eschrichti the gullet runs downwards and backwards, trending slightly to the

1 Op. cit., Zeitschr.f. uiss. Zool. 1877, Bd. xxviii. pp. 255-353, Taf. xii.-xix.

KEPORT ON THE ClUNOIDEA. 89

left. "When it has reached the bottom of the visceral mass the intestine turns off to
the ridit, and coils round its anterior side. It follows the watch-hand, until it has
reached the hinder part of the disk, beliind the commencement of the first coil (fig. 2).
Here it turns upwards and slightly forwards, to end in the anal tube. The spiral form of
the whole organ is thus almost identical with that of the so-called digestive organ in
the Palajocrinoids, which I believe to be nothing but the more or less calcified connective
tissue that supported the intestinal wall, as explained in the previous chapter.

In simple forms, like Rhizocrimis and Bathycrinus, more especially the former, the
development of the gut is but slightly more advanced than it is in the Pentacrinoid.
Horizontal sections through the lowest part of the cup of the larva are remarkal)l3-
similar to corresponding sections of Rhizocrinus and Bathycrimis, such as are represented
in PI. Vllb. figs. 6, 7, and PI. Villa, fig. 8.

W

E

Fig. 2.— Diagram showing the course of the Digestive Tube in an endocyclic Crinoiil (Antedon, I'entacrimui, &c.),

as seen from the ventral side.

A, B, C, D, E, the five ambulacra of the disk ; m, mouth ; «, anus.

The lower part of the cup between the second radials is occupied in Batkyerinns
and in the Pentacrinoid larva by a large expansion of the lowest portion of the coiled
gut, just as described in Rhizocrinus by Ludwig.' It is somewhat kidney-shaped in
section, and the concavity is occupied by the plexiform gland, which is always inter-
radial in position where it comes out of the calyx (PL Vllb. fig. 6, x).

At the level of the third radials of Bathycrinus, or the second brachials of Rhizocrinus,
the circular course of the intestine is more apparent, and the plexiform gland is separated
from the body-wall by the rectum, as shown in PI. Vllb. fig. 7, and PI. Villa, fig. 8.
In both of these figures the x indicates the plexiform gland, which is here situated just
below {i.e., south of) the lower end of the fore-gut, where it passes into tlie mid-gut or
intestine generally.

1 Zitr Anatoiiiie dcs Rliizocrinu.s lofoteii.si.'*, M. Sar?, Zdhchr. f. wiss. Zool, 1877, Bd. xxix. p. 64.

(ZOOL. CHALL. EXP. — PART XXXIT. — 1884.) li 1'-

90 THE VOYAGE OF H.M.S. CHALLENGER.

The further course of the rectum is shown in PI. Vllb. fig. 8, which represents a
section through the upper part of the disk at the level of the articulation between the
first and second brachials ; and the last trace of the plexiform gland (x) is seen between
the fore-gut and the final coil of the intestine.

Both Rhizocrinus and Bathycrinus exhibit a peculiarity in the relations of the
digestive tube and visceral mass which does not appear in the Pentacrinoid. The large
crests at the sides of the arm-groove in the second brachials of Rhizocrinus, which were
well figured by Sars,^ have always been a puzzle to me, for they are something more
than unusual developments of the muscle-plates at the distal end of the joint. Horizontal
sections of the cup show, however, that they support the sides of comparatively large
interradial diverticula from the intestinal coil, the outer ends of which are overlapped to
some extent by the broad dorsal surfaces of the brachials (PL Villa, fig. 8).

The first traces of these extensions of the digestive cavity appear in Rhizocrinus lofo-
tensis at the level of the muscle-plates of the first brachials, by which they are in a great
measure sujiported. They become more independent of the skeleton at the syzygy -with
the second brachials ; but enter into close relations with these joints at their distal ends,
where the crests on the ventral surface become more prominent. They are much more
marked in some individuals than in others, but the outline of the \dsceral mass is
always distinctly pentagonal, even if its angles be not produced outwards, so as to give
it a stellate appearance. The same peculiarity appears in Bathycrinus. The crests on
the radial axillaries are large, wing-like processes, altogether distinct from the muscle-
plates to which the first brachials are attached, as is shown in PL VII. fig. 4a, and PL Vila,
fig. 1 7 ; and they afi"ord a large amount of sujiport to the interradial diverticula of the gut
(PI. Vllb. fig. 7). These develop themselves gradually from below upwards, no trace
of them appearing below the articulation of the second and the axillary radials ; but
indications of them are stiU visible at the upper part of the first brachials, as shown in
PL Vllb. fig. 8.

In this figure, too, may be seen the earlier and simpler stages of that plication of the
inner or adcentral wall of the gut which is so much more marked in Pentacrinus, and
still more so in the Comatulae, especially in Anteclon eschnchti, Promachocrinus, and
allied forms (PL LVIII. figs. 4-6).

The disk of the Comatulse is generally flatter than that of the stalked Crinoids, and is
almost entirely independent of any lateral support from the second radials. These lie
beneath more or less of its dorsal surface, but do not protect its sides.

The descending portion of the fore-gut is therefore comparatively short, and the
coiled intestine into which it passes lies spread out on the upper surface of the radials ; so
that there is no general dilatation of the gut at the bottom of the cup as in Rhizocrinus,
Bathycrinus, and the Pentacrinoid. Some of the Pentacrinidse approach Anteclon, while

1 Op. cit., figs. 55-57.

REPORT ON THE CRINOIDEA. 01

otliers rather resemble Rhizocrinus and Bathycnnus in the arrangement of the digestive
tube. Pentacrinus decorus is one of the hitter. Even at the level of the radial axillaries
the gut appears in section as a simple, but spacious cavity, with slight extensions at two
points round the connective, or rather calcareous, tissue in which the plexifonn gland is
embedded. But it could hardly be called kidney-shaped, as it is in Rhizocnnus and
Bathycriniis.

Some sections of a Pentacrinus disk, that were made for Sir Wyvillc Thomson by
Dr. Stirling, show the indentation of one wall of the gut at the level of the radial
axillaries by the plexiform gland and its surroundings to be considerably more marked
than in Pentacrinus decorus (PL LVIII. fig. 4).^ I am unfortunately unable to determine
the species, as the sections were not properly labelled, and the series is not sufiiciently
complete for the purpose.

In Pentacnnns wyville-thomsoni, however, the condition of the gut is much more like
that found in Antedon, which it further resembles in its disk being somewhat more inde-
pendent of the skeleton than in other Pentacrinidae. Even at the level of the second
radials, a horizontal section of the disk shows a strong concavity at one side of the gut,
which is thus almost crescentic in outline. The plications of the inner wall, however,
are by no means so well developed as they are in many Comatulse.

In the remarkable genus Actinometra, the radial centre of the water-vascular system
does not correspond with that of the dorsal skeleton ; and the curious duality of the
Crinoid organisation is thus seen more distinctly in this type than in any other. The
mouth is not sub-central but excentric, or even marginal (PI. LXI. fig. 2), and there is
no regular symmetry in the distribution of the amliulacra (PL LV. figs. 1,2; PL LVI.
fig. 7). The mouth may be radial, as in aU endocycHc Crinoids, and such species of
Actinometra as Actinometra Solaris, Actinometra pidchella, and Actinometra jukesi
(PL LV. figs. 1, 2); or it may be interradial, as in Actinometra magnijica (PL LYI.
fig. 7) ; while in some types its exact position is difficult to determine. This is,
however, immaterial as regards the course of the digestive tube, which proceeds directly
downwards to a point somewhat behind and on the left {e(tst) of the centre of the disk,
and then commences to wind.

Its direction, just as in the endocyclic Crinoids, follows the watch-hand when seen
from the ventral side ; but there are four coils instead of one. This is shown in fig. 3,
where the -f at the end of the first coil marks the termination of that part of the gut
which represents the whole digestive tube in the endocyclic forms.

This first coil occupies the extreme edge of the lowest part of the disk, and consequently
passes in front of the mouth, so as to appear beneath it in longitudinal section (PL LX].
fig. 2). The second coil passes immediately behind it, and is followed by two more in an
ever narrowing Ijut ascending spiral, which terminates in the more or less central anal

' This fi,:;ure nearly corresponds to the southea?t comer of PI. \\\\\ tig. 7.

92 THE VOYAGE OF H.M.S. CHALLENC4ER.

tube (PL LXI. fig. 1). This is often some little distance in front of the point where the
lowest part of the long gullet turns off westwards to enter the great outside coil. The
walls of this long digestive tube are tolerably simple and but slightly plicated. For an

E

D

Fig. 3.— Diagram showing the course of the Digestive Tube in an Actinomclro with Iiiterrailial

Mouth, as seen from the ventral side.

Letters as in fig. 2. Tlie + on the hinder portion of the outer coil indicates the limit of that part of the gut which

corresponds to the entire digestive tube (excepting the rectum) of an endocyclic Crinoid.

extensive secreting surface is amply provided, without the necessity of this further
complication, which is so largely developed in the simple spire of the gut in Promachu-
crinus, Antedon eschrichti, and Antedon antarctica.

C. The Water- Vascular System.

The water- vascular system of a Crinoid consists, like that of the Stellerids, of an oral
ring and radial vessels, the former being connected indirectly with the exterior through
the intervention of the water-tubes, water-pores, and the body-cavity. Neither the oral
ring nor the radial vessels have any ampuUge connected with them ; though, as suggested
by Ludwig,^ these are perhaps represented by the small lateral pouches of the radial vessels
which are opposite to the tentacular branches, and are crossed hj muscle-threads'"
(PL LX. fig. 6).

The presence of these tentacular branches is invariably correlated with that of the
food-groove. Where this is well marked, and lined by ciliated epithelium with the
subjacent ambulacral nerve and blood-vessel, the water- vessels beneath the latter give ofi"
their branches to the tentacular groups in the usual regular way (PL Vc. fig. 7.
PL LVIL figs. 1, 3, 4 ; PL LIX. fig.s. 1,5; PL LX. figs. 1, 2, 6—tb ; PL LXI. figs. 4, 6).

' Crinoideen, Ice. cif., p. .337. - Sec also fig. 5 on p. 121, m'.

REPORT ON THE CRINOIDEA. 03

But if the food-groove on the ventral surface of the arm or pinnule remains undeveloped
(PI. LXI. fig. 3), not only are the amhulacral epithelium, nerve, and blood-vessel absent
altogether, but the water-vessels are simple tubes like the integumentary water-vessels of
the Molpadidse, and have no lateral extensions, as tentacles are absent (fig. 4 on p.
113,21;).

This condition may occur in a majority of the arms and even on the disk of
Actinometra (PI. LVI. fig. 7) ; on more or fewer of the lower pinnules of Antedon acoela
and Antedon angusticahjx (PL LIV. figs. 1-3, 5) ; and on the proximal pinnules of
Antedon eschrichti and Antedon rosacea, which receive no branches from the brachial
ambulacra.

In Metacrinus, on the other hand, the ambulacra, and ^\-ith them the water-vessels,
of the large basal pinnules may start dii-ectly from the primary ambulacra of the
disk, or even from the peristome itself (PI. XXXIX. fig. 2 ; PI. XLIII. fig. 3 ; PL L.
fig. 2).

The radial water-vessels which underlie the disk-ambulacra of the Comatulte pass off
from the angles of the somewhat pentagonal water- vascular ring as single trunks, situated
beneath the median lines of the ambulacra. But in Pentacrinus, at any rate in Pentacrinus
decorus and Pentacrinus wyville-thomsoni, there is a radial extension of the labial Ijlood-
vascular plexus in this position (PL LVII. figs. 1, 3, 4, Ir); and the two trunks which
ultimately unite into the single water- vessel of the ambulacrum are thus kept separate
from one another to a considerable distance, 1"5 mm. or more, from the edge of the
peristome ; that is to say, the angles of the water- vascular ring are produced in the
direction of the rays, so that its outline is that of a short-armed Asterias rather than the
more regularly pentagonal figure of a Goniaster.

The ciliated water-tubes (" Steincaniile," Ludwig) by which the w^ater- vascular system
communicates with the body-cavity, and thence with the exterior, vary very greatly in
their development. The early Pentacrinoid has but one, situated in the same interradius
as the fore-gut. In the later stages of Pentacrinoid life and in the young Comatula just
free there are five, one in each interradius; and the same is the case in RhLocvinus
lofotensis. They are multiple in Bathycrimis, though not abundant ; whUe in the adult
Antedon rosacea there are about thirty in each interradius ; and in Antedon eschrichti
and in Pentacrinus the number becomes still larger (PL LVII. figs. 1, 3, 4 ; PL LIX.
fig. 5 — ivt).

The radial water- vessels of Comatula commence as single trunks arising from the
water-vascular ring at the edge of the peristome ; and in a large Comatula like Antedon
eschrichti the w^ater-tubes may be found depending from the bases of the radial vessels
beneath the middle line of the ambulacrum in the first two or three sections beyond the
edge of the peristome.

In Pentacrinus, however, the middle line of the commencing ambulacrum is occupied

94 THE VOYAGE OF H.M.S. CHALLENGER.

!)}• the blood- vascular plexus (PL LYII. figs. 1, 4, Ir), which has a water-vascular trunk
ou each side of it ; and the water-tubes extend outwards in a radial direction as long as
the Avatcr-vessels remain double. This is very evident in some horizontal sections
through the tapper part of the disk of Pentacrinus naresianus, which were made for
Sir Wyvdle Thomson by Dr. Stirling. The double row of water-tubes may be seen
extending along the sides of the ambulacra beneath the line of tentacles, to a distance of
3 "5 mm. from the peristome, and then it is not complete.

The position of the water-tubes beneath the primary ambulacra is well shown in
PI. LVII. figs. 1, 3, 4. They are seen, as it were, coiling round the subambulacral
plates to open below into the upper part of the circumvisceral coelom. When, however,
the radial extensions of the labial plexus gradually thin out and the two lateral water-
vessels unite into a single median trunk, the water-tubes become less numerous, and are
only found in the first two or three sections beyond the point of union. They are thus
really in the same position as in the Comatulse, if we consider the double water- vessels as
expressing extensions of the angles of the water-vascular ring in the direction of the
rays.

The water-pores which pierce the ventral perisome, whether it be plated or not, are in
a close functional relation with the water-tubes. They are the openings of minute canals
which are lined by columnar epithelium, and expand almost immediately into enlarge-
ments where the epithelium is ciliated (PI. LVII. figs. 1, 3, 4 ; PI. LIX. figs. 2, 4, 6 — zf^j).
The inner end of the canal beyond the enlargement is lined by pavement epithelium, and
opens into the body-cavity. According to Perrier ^ the primary water-tubes of the early
larval Comatute are directly continuous with the inner ends of the water-pores, without
the intervention of the body-cavity.

He has not yet figured this connection, however ; and after the careful observations of
Ludwig upon the subject, with which my own are in complete accordance, I have
considerable hesitation in accepting Perrier's statements, especially as he admits that the
water-pores of mature Comatulte do establish communication between the body-ca\dty and
the exterior, just as described by Ludwig in the Cystid phase. This subject is discussed
more fully elsewhere.^ [See Appendix, Note D.]

In the smaller and simpler types of Crinoids the water- pores correspond in number to
the water-tubes. The young Antedon has one in each interradius ; and the same is the
case in Rhizocrhms, the single water-pore piercing the oral plate. In Hyocriniis,
however, the number of pores is larger. In both the sjiecimens which I have examined
there are two pores in the oral jjlate of the anal interradius, and there are no others in
any of the anambulacral plates which lie between it and the edges of the radials. In the
other interradii the disposition of the water-j)ores is as follows : —

1 Sur le cUveloppement des Coraatules, Comptes reiidns, t. xcviii., 1884, pp. 444-446.

2 On some points in the Anatomy of Larval ComatiiI;p, Quart. Journ. Micr. Sci., N. S., vol. xxiv., 1884, p. 320.

EEPOET ON THE CRINOIDEA.
Table I. — Water-jJores of Hyocrhms.

95

Interradius.

Oral Plates. Anambulacral Plates.

A.

B.

A.

B.

No. P(anal), .
„ 2 . . . .
„ 3 . . . .
„ 4 . . . .
„ 5 . . . .

2
0
0
0
1

2

?
1
1

1

0
2

3
3
2

0
4
9
7
7

Thus, therefore, one specimen has all the oral plates perforated by water-pores, with
one possible exception ; while in the other it is only in the anal interradius and in the
nest one to it that the oral plate is pierced by a water-pore. It is significant that this
interradius (No. 5) is that in which the primary water-pore ajDpears in larval Asterids,
Ophiurids, and Crinoids ; and also that in liotli the specimens of Hyocrinus the water-
pore is situated a little beyond the middle line of the large oral plate (PI. Vc.
ficr. 6, top), just as is the case with the single water-pore of the Pentacrinoid.

The disk of Holopus is practically entirely made up of the oral plates, and there is
no ventral perisome at all. In the figure of these plates which was drawn for Sir Wyville
Thomson by Mr. Black (PI. III. fig. 2), each plate seems to be pierced by several pores ;
but as the specimen was further dissected after being drawn, and the oral plates were not
preserved, I have been unable to satisfy myself upon this point.

The number of water-pores which may occur in any ordinary Comatida or Pentacrinns
varies to a considerable extent. Ludwig estimates that there are 1500 in the ordinary
adult Antedon rosacea, and there must be even more in Antedon cschrichti. Except in
the immediate neighbourhood of the mouth they are less abundant in the anal interradius
than elsewhere ; and though they sometimes occur on the sides of the disk, i.e., over the
edge of the ventral surface, they never extend far down towards the dorsal side. When
the disk is plated, the pores may be scattered singly in individual plates (PI. XVII.
fig. 6), or they may be grouped together on one plate, sometimes even to the number of
twenty, as in Pentacrinns decorus.

In the exocylic type Actinomctra, the water-pores are generally situated in the
immediate neighbourhood of the ambulacra ; and the greater part of the disk which is
occupied by the large anal interradius is almost entirely free from them. They are not
necessarily limited to the disk, for they may also be found on the lower parts of the

1 These numbers correspond to those which are employed by Ludwig in Asterid morphology.

V

90 THE VOYAGE OF H.M.S. CHALLENGER.

arms and also on the proximal pinnules. In both cases they open into that section of
tlie Iwdy-cavity which surrounds the generative apparatus, and is known as the genital
canal (PL LXI. fig. 5). In a few cases too, I have found water-pores on the middle and
later pinnules of the arms. They open into the genital canal of the pinnule, close to
the point where it arises from that of the arm.

D. The Blood- Vascular System.

The foundation of almost all our accurate knowledge of the blood-vascular apparatus
of the Crinoids is due to the researches of Ludwig. The blood-vessels form a highly
complex system, parts of which are entirely unrepresented in the unstalked Echinoderms
(EcJiinozoa) ; while other parts of it, such as the oral ring and its radial extensions abo^'e
the water-vessels, conform to the ordinary Echinoderm type.

The radial vessels (PI. Villa, figs. 4, 5 ; PI. LIX. figs. 1,5; PI. LX. figs. 1, 4, 6—h)
vary considerably in size, and are often invisible if the section be at all oblique.
Tliey are large in Antedon eschrichti and in Actmometra nigra, and may frequently be
found to contain yellow pigment-masses or coagula. In the latter ty^e they are some-
times triangular in section, with the apex pointing downwards so as to be received into
a strongly marked concavity in the upper edge of the water-vessel (PI. LXI. fig. 6) ;
but in Antedon eschrichti, Pentacrinus decorus, and in most other types their section is
more or less lenticular. Sometimes, however, it is triangular with the apex projecting
upwards towards the epithelial layer above, and so rendering the ambulacral nerve much
thinner in the middle line than in its more lateral portions (PL LX. figs. 1, 6, b).

Its cellular lining is much more delicate than that of the intervisceral ])lood- vessels,
and is not easy to make out. Ludwig ^ found that this radial blood-vessel in the arms
of Antedon eschrichti is sometimes divided into two parts by a vertical septum, which
has a distinctly cellular covering ; and I have not unfrequently found in the disk of the
same species that the lumen of the vessel may be crossed in various directions by
delicate threads with nuclei upon them. These resemble the nucleated muscle-threads
in the water-vessels, but are much finer and less refractile.

The oral ring which connects the radial blood-vessels, and resembles them in structure,
extends somewhat beyond the inner edge of the water- vascular ring (PL LXIL);
and its wall thus projects into that part of the body-cavity which is contained within
the dense mesh-work of connective tissue supporting the lip (PL LX. fig. 4, c). Ludwig
has described in Antedon rosacea a number of more or less branching tubules which are
connected with the oral blood- vascular ring, and have a somewhat better defined epithelial
lining than the intervisceral blood-vessels.^ He thought it possible that they might be in
connection with the ramifications of the upper (ventral) end of the plexiform gland,

1 Crinriideen, loc. cil., p. 2G7. - Ilnd., p. 328.

REPORT ON THE CRINOIDEA. 97

but he was unable to satisfy himself ujDon this point. The Naples variety of Antedon
rosacea, upon which he worked, is about the worst type that could have been chosen
for delicate observations of this kind, the quantity and character of the pigment-
granules which the lip contains almost totally obscuring the other tissues.

This pigment-substance is altogether different from the yellow and brown pigment
masses so common in all Echinoderms. It is apparently related to tlie calcareous
spicules which are so abundant in the under part of the perisome and in the intervisceral
connective tisstie. When these have been removed by acid their outlines are seen
to be well defined by streaks of pigment-granules, which appear black by transmitted,
and yellowish-white by reflected, light. They are especially abundant in the lip, and
in some individuals almost entirely conceal the blood- vascular plexus which it contains ; '
while in specimens from other localities, and in Antedon eschrichti, it is scarcely developed
at all, and I have found no difficulty in tracing the connection of the plexiform gland
with the oral ring. This is very evident in all the other Comatulee, including Antedon,
Actinometra, and Promachocrinus, which I have examined, as well as in Pentacrinus
(PL LVII. fig. 3 ; PL LXIL).

The branching tubules depending from the oral ring which much resemble the visceral
blood-vessels, open into a dense plexus of more glandular looking tubules that is
supported by connective tissue, and extends right round the lip (PI. LVII. figs. 1, 3, 4 ;
PL LIX. fig. 5 ; PL LX. figs. 1, 2, i—lp).

It is connected with (l) the ventral branches of the plexiform gland (PL LVII.
fig. 8, xv) ; (2) with the genital vessels of the rays (PL LX. figs. 1, 2, gv) ; and (3) with
some of the intervisceral vessels (PL LVII. figs. 1, 3, 4 ; PL LX. figs. 2, 3, 5, ih ; PL LXIL).
These last form an extensive network over the coiled digestive tube, and are also directly
connected with the plexiform gland (PL LVII. figs. 2, 5 ; PL LVIII. fig. 6, ih). This
labial plexus is most abundantly developed beneath the south and south-east portions of
the peristome, i.e., in the neighbourhood of the left posterior ambulacrum. It lies
between the hind-gut forming the last coil of the digestive tube, and that part of the
capacious fore-gut which lies between the two lateral ambulacra of the right side. It is
always pretty sharply defined from the surrounding connective tissue, and is usually a some-
what prominent object in a well-stained section which is examined with the naked eye or a
simple lens. This is partly owing to the relatively thick walls of its component tubules, and
partly to the delicacy of the connective tissue holding them together. Its relations are easily
made out by the study of series of tranverse and longitudinal sections through the disk.

It is not very specially developed beneath the origins of the anterior and antero-
lateral ambulacra, the jilexiform genital vessels of which may be traced into it ; though
it is somewhat denser on the left (east) than on the right side, where it is connected

1 I Lave found the same equally impracticable pigment in the disk of Adhiomctra peclinata from Singapore, and also
in some individuals of Actinometra parvicirra from Bohol, although others from the same locality are totally devoid of it.

(ZOOL. CHALL. EXP. PART XXSII. — 1884.) li 13

98 THE VOYAGE OF H.M.S. CHALLENGEE.

below with the plexiform ghmd rising alongside the fore-gut. But it is very marked
indeed in transverse sections behind the mouth (i.e., between it and the anus), as is well
shown in the case of Pentacrinus loyville-thomsoni (PI. LVII. fig. 1, Ip), Antedon
eschrichti (PI. LX. fig. 5), or of any Actinometra. It gradually diminishes in size as
the distance from the mouth increases, occupying an intermediate position between the
two posterior ambulacra. It is usually rather nearer to the left one (C), but is sometimes
closer to the other {D). The genital vessels of these ambulacra are derived from it
(PI. LVII. fig. 3 ; PI. LX. figs. 1, 2 — gv), and it finally passes insensibly into the inter-
visceral vesssels of the hinder part of the disk. In the specimen of Antedon carinata,
one section of which is represented in PI. LX. fig. 2, the labial plexus lies, as usual,
rather nearer the left posterior ambulacrum, continuing very close and compact until just
in front of the anal tube, where its meshes open out, and it passes into the ordinary net-
work of intervisceral vessels.

In like manner the examination of a series of longitudinal sections shows that the
labial plexus is denser, and extends farther from the peristome on the left side than it
does on the right. In the specimen of Antedon rosacea, a section of which is figured in
PI. LIX. fig. 5, the plexus is much more developed, both anteriorly and posteriorly, at
the left edge of the mouth-slit than at the right edge, or even than in the median plane
which traverse? the anterior ambulacrum. The section figured (PL LIX. fig. 5) passes
through the left angle of the peristome, from which the two lateral amljulacra diverge ;
and the labial plexus is seen as a broad band {Ip), which lies between the water-tubes
depending from the water-vascular ring {ivt), and the fore-gut {/(j) ; it diminishes in size
as the distance from the mouth increases, and loses its individuality when the two
ambulacra become differentiated.

Unfortunately I do not know the locality of this specimen ; but it is singularly
devoid of the dichroie pigment which so unpleasantly increases the difliculty of accurate
observation in the Naples variety of Antedon rosacea.

In Promachocrinus herguelensis, in Antedon eschrichti and its allied species Antedon
quadrata and Antedon antarctica, a portion of the labial plexus between the mouth and
the anal tube differs very considerably in structure from the rest of this organ. The
limits of this portion are so well defined, and it difiers so much from the remainder of
the labial plexus, that I propose to designate it by the name "spongy organ." Its
relations to the rest of the labial plexus in Antedon eschrichti are shown in PI. LX.
figs. 3, 5, so. The former represents a longitudinal, and the latter a transverse section
of it ; while in PI. LIX. fig. 8, a portion of the spongy organ of Promachocrinus
hergiielensis is shown more highly magnified. It lies between the mouth and anus on
the left or eastern side of the gullet, and therefore is slightly nearer to the left posterior
than to the right posterior ambulacrum.

In its most fully developed condition, only found between the mouth and anus, the

REPORT ON THE CEINOIDEA. 99

spongy organ is a somewhat egg-sliaped mass, consisting almost entirely of a delicate net-
work of connective tissue with more or less open meshes.

The latter are rather wider in Promachocrimis than in any other Crinoid which I
have examined. The trabeculse forming its outer portion are much more delicate than
those nearer the surface ; and I have been unable to make out that they possess
any epithelial covering. The surface of the organ, however, is more compact, with larger
trabeculse and generally smaller meshes. These are lined by epithelial cells, beino' in
fact the ends of those blood-vessels forming the labial plexus which are connected with
the spongy organ, mostly, if not entirely, on its ventral side (PI. LX. fig. 5).

Some of these vessels are seen in more or less oblique section at the rio-ht end of the
spongy organ of Fromachocrinus kerguelensis (PI. LIX. fig. 8), while the relation of the
spongy organ to the labial plexus generally is well shown in the transverse and longi-
tudinal sections through the disk of Antedon eschrichti (PL LX. figs. 3, 5). The spongy
organ of this species is more compact than that of Promachocrimis. It is similarly
situated in the space left by the incomplete adhesion of the visceral and parietal layers
of the peritoneum ; and it is suspended in this space by threads of connective tissue. It
is practically the direct backward continuation of the labial plexus at the eastern angle
of the mouth, where it is much more largely developed than on the opposite side. The
relatively thick epithelial wall of the vessels gradually disappears as they enter the
spongy organ ; while the latter in its turn passes insensibly backwards into the plexus of
vessels on the upper surface of the visceral mass, from which are derived both the inter-
^dsceral vessels and the genital vessels of the two posterior ambulacra. The absence of
an epithelial lining in the spaces of the spongy organ is very marked, although the
epithelium is quite distinct in the blood-vessels which terminate therein. But, on the other
hand, I have found, both in this species and in Antedon quadrata, that the nuclei of the
connective tissue forming the trabeculiB stain very prominently, much more so than I
could get them to do in Promachocrinus kerguelensis.

Although the simple reticular structure of the spongy organ in this latter tyj^e is
limited, as described above, to that part of the labial plexus which is situated between
the mouth and anus, yet the distinction between it and the remainder of the labial plexus
is far less sharp than in Antedon eschrichti. For the vessels forming the labial plexus of
this type are much more closely grouped, and have a less definite epithelial wall than is
the case in Antedon eschrichti, so that it assumes a decidedly reticular character.
This is also the case with the upper cud of the plexiform gland, which retains its
individuality till quite close to the mouth (PI. LIX. fig. 9, xr), for it remains large and
lobulated, instead of breaking up as in Antedon ; and the connection of its reticidar portion
with the labial plexus at the north-east angle of the mouth is quite distinct ; while at the
western angle the labial plexus, though somewhat reticular in structure, is but poorly
developed.

100 THE VOYAGE OF H.M.S. CHALLENGER.

Towards the hinder edge of the mouth the trabecular structures in the left division of
the labial plexus become more and more delicate, their cellular covering thins out, and
the true spongy organ appears, with the characters described above. It is formed almost
entirely from the network on the left side of the mouth in which the plexiform gland ends
(PI. LIX. fig. 9, xi'). It remains throughout nearer to the left than to the right of the
two posterior ambulacra, gradually becoming more compact again, and finally passes into
the plexus of intervisceral and genital vessels.

The spongy organ of the type which is generally called Antedon rosacea varies very
much in its structure. In some cases it is almost as reticular as in Antedon eschrichti,
whUe in others it is hardly differentiated from the rest of the labial plexus, and consists
of a mass of twisted tubules, which have well defined epithelial walls. It is possible
that these variations may be due to specific difi'erences, but upon this point I can, as yet,
ofi"er no opinion.

The labial plexus of Pentacrinus, at any rate of the two species which I have studied
(^Pentacriniis decorus and Pentacrinus wyville-thomsoni), is much more highly developed
than in the ComatulEe, but it contains nothing like the spongy organ of Antedon eschrichti
and its allies. It extends outwards for some little distance from the peristome, both
beneath and between the ambulacra. In the former case it is connected with the radial
blood-vessels, beneath the middle line of the groove, thus keeping apart the converging
water- vascular trunks at its sides as already described (PI. LVII. figs. 1, 3, 4, Ir).

Its histological condition in the individuals of both the species which I have examined
is not such as to facilitate the observation of minute structural details, but from what I
have seen of the better preserved portions of it, I have no reason to think that it difiers
essentially from the corresponding organ in Antedon. The epithelial lining of its cavities
is often fairly distinct, as sho'WTi in PI. LVII. fig. 4.

As the ambulacra recede from the peristome and thus diverge more and more, the
vascular plexus underlying the interpalmar area which separates them gradually thins
out, untU it is only represented by the uppermost inter\'isceral vessels and the genital
vessels of the rays, both of which originate in it. The former belong to the circumvisceral
layer of the peritoneum, and the latter to the parietal layer ; but the two systems are in
free communication with one another (PL LVII. fig. 3, gv, ih).

The tubules depending from the radial blood-vessel and entering the subambulacral
plexus gradually become less and less aliundant, and finally disaj^pear altogether, shortly
before the limit of the water-tubes is reached. In the anal interradius the labial plexus
is specially developed, as is well shown in transverse section in Pentacrimis uyville-
thomsoni (PI. LVII. fig. 1, Ijj), lateral extensions of it being connected with the radial
blood-vessels by the tubules above mentioned. [See Appendix, Note D.]

The subdivision of the upper end of the plexiform gland into numerous branches
which terminate in the labial plexus, can generally be made out without any difliculty

EEPOET ON THE CRINOIDEA. lOl

in vertical sections of a disk that is moderately free from pigment. In none of the
species of Pentctcrinus or Comatida which I have examined, have I failed to observe
this connection satisfactorily ; and in some fortunate sections the vascular structure may
be traced right down from the oral ring into the lower and more compact portion of the
plexiform gland (PL LXIL). This is generally more or less lobulated in form, and in
this respect very striking differences exist among the various tjrpes of Crinoids.

Both in Rhizocrinus (PI. Villa, figs. 7, 8, x), and in Bathycrinus (PI. Vllb.
figs. 1, 4-8, x) it is almost as simple as in the Pentacrinoid. It is oval-oblong in section
and has but slightly developed irregularities of outline. In Antedon rosacea, and in the
lower part of the disk of Pentacrinus, untd it commences to subdivide, it is more irrefnilar
in form (PI. LVIII. fig. 4, x) ; and it shows indications of its lobular structui-e, as is well
seen in Lud'^'ig's admirable illustrations ' of the former type. I have made no horizontal
sections of the disk of Antedon eschrichti, but imagine that its plexiform gland must be
not very unlike that of PromachocHnus kerguelensis. Figs. 5 and 6 on PI. LVIII. represent
horizontal sections through the gland in the lower half of the disk of this type, fig. 6
being that nearer the calyx. The u'regularity in the form of the organ is very striking.
The upper part of the same disk was cut into vertical sections. They show the extreme
subdivision of the ventral end of the plexiform gland, and the termination of its branches
in the well defined labial plexus, the spongy part of which is represented in PL LIX.
fig. 9, XV.

The minute structure of the plexiform gland of the Crinoids is as yet unknown.
According to Perrier ■ it is identical with that of the same organ in other Echiuotlerms.
Fresh specimens, and not spirit-material, are absolutely essential for its elucidation ; but
as Perrier and Koehler, who have both studied the plexiform gland of the Urchius, give
accounts of its structure which difier from one another, and from Apostolid^s' description
of the same organ in Ophiurids, there is a difificulty in determining from analog}^ and
stdl more so from observation, what its real nature is in Crinoids.^ Certain peculiarities
that I have noticed in the appearance of its lower portion in Pentacrinus decorus lead
me to think that Koehler's account of it in the Urchins * is probably more correct than
those of his fellow-workers.

Upon one point, however, I am quite satisfied. Although the condition of my material
entirely precluded any minute observations on the structure of the plexiform gland, I
have had no difficulty in tracing its connection with the intervisceral blood-vessels of the
lower part of the disk (PI. LVII. fig. 2, PI. LVIII. fig. 6). This point was first noticed by
Ludwig in Antedon rosacea, and he illustrated it with an excellent figure.^ I have

' Crinoideeu, loc. ciL, Taf. xviii. figs. 57-59.

2 Sur rorganisation des Crinoides, Compfes rendm, t. xcvii., No. 3, p. 18.

' For a discussion of this subject see Quaii. Jouni. Micr. ScL, N. S., vols, xxii.-.xjciv., 1882-1884.

* Recherches sur les i;chimdes des Cotes de Provence. Ann. du Mus. d'Hist. Nat. de Marseille, t. i. pp. 73-77, 95-99.

^ Crinoideen, loc. cit., Taf. xviii. fig. 59.

102 THE VOYAGE OF H.M.S. CHALLENGER.

myself described and figured the same connection in Actinometra parvicirra ;' and I
have since seen it so frequently in the different species which I have studied, that I
read the following statement of Perrier's with some little surprise. Speaking of the
plexiform gland of Antedon rosacea, he says/ " Les vaisseaux qui paraissent en partir ne
sont autre chose que les ramifications de la glande, se terminant d'ordinaire par des ren-
flements ayant I'aspect de culs-de-sac. Ces ramifications courent au milieu des innom-
brables trabecules du tissue conjonctif de la cavitd gdndrale, qui peuvent eux memes
^larfois prendre I'apparence des vaisseaux." I cannot gather from this passage whether
Perrier means to deny the existence of intervisceral vessels altogether, or merely the
connection of this system with the plexiform gland. I have good reason to believe, as
pointed out elsewhere/ that his statements refer solely to Antedon rosacea ; but even in
this unfavourable type I have had no difficulty in confirming Ludwig's observations
respecting the relations of this organ to the blood-vessels. If the latter be ramifications
of the gland, as Perrier asserts, one would expect that they should have the same structure
as it ; whereas their nature is the same as that of the intervisceral blood-vessels, which
are lined by a layer of pavement-epithelium (PI. LVII. fig. 5); while their apparent
blindness is simply due to the impossibility of any single section showing more than a
very small portion of their winding course. A careful study of a good dissection, or of
a moderately thick transparent section, especially with a binocular, or an accurate
plotting out on paper of a series of thin sections by means of a camera, will reveal much
that is totally unrecognisable in other ways. [See Appendix, Note E.]

I have studied the intervisceral blood-vessels principally in Antedon eschrichti,
Pentacrinus decoriis, and Actinometra parvicirra. In the Pentacrinus, with its body-
cavity reduced by the presence of much calcareous tissue, the visceral blood-vessels are
less abundant within the simple coils of the digestive tube than on its outer surface,
where they may be excellently seen in tangential sections of the visceral mass, as shown
in PL Lvil. fig. 5, ib.

But in Antedon eschrichti and Actinometra, which have a more complex digestive tube,
the intervisceral blood-vessels are more largely develoj)ed. The connection of one of them
with the plexiform gland of Promachocrinus kerguelensis is shown in horizontal section
in PI. LVIII. fig. 6, while PL LVII. fig. 2 represents a portion of one of several vertical
sections of Pentacrinus decorus in which the same point is visible, the plexiform gland
itself unfortunately having no recognisable structure.

The blood-vessels may be readily distinguished from connective tissue, especiall}- in
Pentacrinus ; for nearly all the visceral connective tissue of this type is regularly calcified,

1 The Mimite Anatomy of tlie Brachiate Echinoderms, Quart. Journ. Micr. Sci., N. S., vol. xxi., 1881, p. 185, pi. xii.
fig?. 14, 15.

" Comptes rendus, t. xcvii. p. 188.

3 Notes OH Ecbinoderm Morphology, No. V^I. ; On the Anatomical Relations of the Vascular System, Qruxrt. Journ.
Micr. ScL, 1883, vol. xxiii. p. 610.

REPORT ON THE CRINOIDEA. 103

and therefore presents tlie usual reticular structure after the limestone has been removed ;
and even those bands of ordinary connective tissue (both in Comatula3 and in Pentacrinus)
which contain irregular limestone deposits have a totally different facies from the blood-
vessels (PI. LVII. fig. 5). They meet at different angles, and have no internal cavity
which is lined by epithelium and frequently contains coagulum.

The relation of the plexiform gland to the intervisceral blood-vessels is especially well
seen in Actinometra.

Owing to the excentric position of the mouth, the characters of the plexiform gland are
considerably different from those of the same organ in the regular Crinoids. I have
studied it more particularly in the two species Actinometra j^arvicirra and Actinometra
'pulchella, the latter having the mouth much nearer to the centre of the disk than the
former. It is also radial in position, while that of Actinometra parvicirra is interradial.
In both types the labial plexus is most developed behind the mouth, and especially
towards the left or eastern angle, where it contains an imperfectly differentiated spongy
orsan ; and it also extends outwards between the ambulacra farther than on the
right side.

It is the right anterior portion of the labial plexus, however, which is more especially
connected with the plexiform gland, while the spongy organ at the left posterior angle of
the mouth passes backwards into the intervisceral vessels.

In Actinometra pulchella with its radial, and but slightly excentric mouth (PL LXI.
fig. 1), the relations of the parts are, as might be expected, much more like those of Antedon
than is the case in Actinometra jMrvicirra, with its nearly marginal and interradial mouth
(PI. LXI. fig. 2). In the former type the plexiform gland rising out of the calyx ascends
nearly in the vertical axis of the disk for some little distance and then divides into two
principal portions. The left hand one is little more than a bundle of vessels which runs
forwards, upwards, and a little outwards till it comes to lie above the gullet imme-
diately in front of the base of the anal tube, and terminates in the spongy part of the
labial plexus at the left posterior angle of the mouth. The right division, which is connected
with the intervisceral vessels of the anterior half of the disk, retains its glandular character,
and passes upwards between the coiled gut and the central rectum to join that part of
the labial plexus which lies beneath the origins of the ambulacra of the right side.

The fore-gut of Actinometra parvicirra passes more directly backwards and down-
wards than that of Actinometra p>ulchella, until it comes to lie immediately above
the base of the posterior ray, rather to the left of the centre of the disk (I'l. LXI. fig. 2).
The spongy organ at the left posterior angle of the mouth is continued downwards and
backwards slightly above the gullet, between it and the three inner coils of the gut, as a
gradually diminishing bundle of vessels, with which the intervisceral vessels of the left half
of the disk are connected. This corresponds to the left division of the plexiform gland in
Actinometra pulchella, though not quite in the same position as regards the digestive

104 THE VOYAGE OF H.M.S. CHALLENGER.

tube. It is connected witli the plexiform gland proper by numerous intervisceral
vessels, which pass both above and below the fore-gut so as to completely surround it.

The plexiform gland proper benda directly forwards as soon as it enters the visceral
mass, lying at first a little way from the right end of the gullet, but ultimately comes to
be close to its anterior end.

This is more nearly in the median longitudinal plane of the disk, so that the
plexiform gland lies between the gullet and the first or outside coil of the gut ; and as in
Actinometra pulchella, the glandular structure passes directly into the labial plexus at the
right angle of the mouth.^ It does not therefore, as in the more regular forms, directly
give ofi" the cESophageal bundle, being only in connection with it by the intervisceral
vessels which encircle the gullet ; but it is the centre for all the vessels of the right and
anterior sides of the disk, just as the corresponding structure is in Actinometra lyulchella.

The plexiform gland diminishes gradually in size when it has passed out of the visceral
mass into the central funnel within the radials ; and it begins to lose its glandular, more
or less lobulated appearance (PL XXIV. fig. 9, X). This is retained longer in forms
like Bathycrinus and Ehizocrinus, which have relatively high radials (PL Vllb. figs. 3-5 ;
PL Villa, fig. 7- — x). Vascular cavities begin to appear in it and group themselves into
an inner set surrounded by a ring of five vessels, which are radially disposed. These
expand within the basals into the chambers of the chambered organ (PL VIIlj. fig. 2 ;
PL XXIV. figs. 6-8 ; PL LVIII. figs. 1-3 — ch). Ludwig has given an excellent account
of this connection in Antedon rosacea,'^ which applies equally well to Actinometra,
Rhizocrinus, Bathycrimis, and Pentacrinus.

According to Perrier,' however, " Le corps ovoide s'implante chez la Comatule adulte
sur I'un des planch ers horizontaux de I'organe cloisonne." In this, as in the former case,
I am at a loss to make out Perrier's real meaning. For he admits in a later communica-
tion* that the ovoid gland or axial organ terminates below " en un tube conique qui
penetre, en s'amincissant toujours, dans I'axe de I'organe cloisonne;" and that this tube
is the upward continuation of the cellular " cordon " in the centre of the larval stem.

If, however, he intends to deny the communication of the cavities of the chambers
with those of the plexiform, or, as he calls it, the " ovoid " gland, I must totally disagree
with him. For I have the most satisfactory evidence of this connection in my series
both of horizontal and of vertical sections through these structures in Pentacrinus decorus,
and also in other t}'pes.

The upper part of the stem contains the usual six vessels, five peripheral and one
central (PL XXIV. figs. 2, 5 ; PL LVIII. fig. 3). The single axial vessel which represents
Perrier's " cordon central " in the larval stem divides when it enters the chambered

1 Two figures, illustrating these points in Acti7iometra parvicirra, will be found in the Quart. Journ. Micr. Sci.,
N. S., voh xxi., 1881, pi. xii. flgs. 14, 15.

2 Crinoideen, loc. cit., p. 317. ^ Gomptes rendus, t. xcviL p. 188.
* Comptes rendus, t. xcviii. pp. 445, 446.

REPORT ON THE CRINOIDEA. 105

organ, first into two (PI. XXIV. fig. 6, v), and tlien into four or more; while the
peripheral vessels around it expand into the chambers (PI. XXIV. figs, 6, 8 ; PI. LVIII,
figs. 1-3 — ch), just as they do on a smaller scale within the nodal stem-joints (PI. XXIV.
fig. 4, clin).

As the ascending vascular axis of Pentacrinus decorus passes out of the basal ring, and
enters the central plug within the radial pentagon, the chambers become elongated and pear-
shaped in the direction of the rays ; and at the same time their cavities are traversed by
trabeculae, which break them up into numerous smaller cavities that contain masses of
yellowish-brown pigment-granules (PI. LVIII. fig. 2, ch). Some of these spaces have a
definite epithelial lining, like the chambers with which they are connected, while others
are deficient in this respect. Before reaching the level of the circular commissure,
however, these radial extensions of the outer vessels of the plexiform gland terminate
somewhat abruptl}' ; and the gland is reduced to a small but compact bundle of vessels
in the middle of the central plug.^ As it rises it becomes extended laterally, and its
shape when it enters the visceral mass (as seen in section) is that of a more or less
irregular L ; while this soon passes into a lobular form of the usual variable
character.

The appearance of these radial extensions of the lowest part of the plexiform gland in
Pentacrinus decorus forcibly recalls Koehler's description of the minute structure of the
ventral end of the ovoid (plexiform) glaud in the Urchins. This is connected with the
oral ring by a single vessel, which is called by Koehler the glandular canal. The other
end of the gland is said by the French anatomists to open externally through the
madreporite in all the Echinozoa, together with the water-tube ; though both in Asterids
and in Ophiurids Ludwig has described it as connected with an aboral vascular ring in
which the genital vessels arise, and there is strong reason to l^elieve that the same is the
case in the Urchins too.

It wiU be a matter of no little difficulty to 'determine exactly the portions of the
blood- vascular system of a Crinoid which are represented in a Starfish or Urchin ; and the
question will probably only be satisfactorily settled by careful studies in organogeny.
There is no doubt about the oral ring upon the one side of the disk, and the chambered
organ on the other. But where is the line to be drawn between the two ?

It seems to me not unlikely, as I have suggested elsewhere,^ that the labial plexus of
Crinoids may represent the aboral ring of the Echinozoa, the plexiform glaud being much
shortened, but expanded laterally instead. Both the intervisceral and the genital vessels
are in communication witli it, just as they are -with the aboral ring of Asterids, according
to Ludwig ; and in this class too the plexiform gland is continued upwards beyond the

1 The upward passage of the chambers into the peripheral cavities of the lower part of the gland is better showTi
in the section next to that represented in PI. LVIII. fig. 2.

2 Q«arf. Joum. Micr. Scl, N. S., vol. xxi., 1881, p. 185 ; and also Notes on EchinodermMorphology, Xo. V. ;^Oii
the Homologies of the Apical System, with some Remarks upon the Blood-vessels, Ibid., N. S., vol. xxii., 1882, p. 375.

(ZOOL. CHALL. EXP. PART XSXII. ISSl.) 11 1 4

106 THE VOYAGE OF H.M.S. CHALLENGER.

aboral ring to terminate in the perisome of the central part of the disk. It would be
very interesting to determine the relation of this dorsal extension of the plexiform gland
in those Asterids such as Zoroaster fulgens, which retain well-developed basal and
radial plates in the centre of the disk, so as to resemble the calyx of a Crinoid.^

It will be remembered that Prof. Perrier has noted the identity in structure between
the axial organ of a Crinoid and the so-called ovoid gland of the Echinozoa. This organ
is considered by Ludwig and mj'self to be in a close relation with the blood-vascular
system, and intimately united to the oral blood-vascular ring, just as the axial organ is
in the Crinoids. But Perrier believes it to be an excretory gland, unconnected with the
blood-vascular system and opening to the exterior through the madreporite. Koehler's
observations on the Urchins, however, tend to disprove this theory, as I have shown
elsewhere.^ Perrier has recently asserted that the axial organ of a Crinoid communicates
with the exterior (see Appendix, Note D) ; but although he has described its structure
as identical with that of the ovoid gland of Starfishes and Urchins, he nevertheless
compares it with, the madreporic or stone-canal of these types.^ He thus considers an
organ which is related to the blood-vascular system of a Crinoid to be represented by a
part of the water-vascular system of other Echinoderms ; and he denies that the latter
corresponds to the water-tubes and water-pores of a Crinoid, as is generally supposed.

The only reasons which he brings forward for this conclusion are that the walls of the
axial organ in the young Crinoid are folded like those of the stone-canal in Starfishes ;
while it has the same position with regard to the digestive apparatus as the stone-canal
of Urchins. The first reason appears to me to be of very Uttle value, as I have pointed
out elsewhere;* while Perrier seems to have overlooked the fact that the second one is
equally applicable to the doctrine of a general homology between the axial organ of a
Crinoid and the so-called heart of an Urchin. For this structure lies in the immediate
neighbourhood of the stone-caual, and Perrier himself admits that it is identical in
structure mth the axial organ of a Crinoid, which is certainly not the case with the
stone-canal.

It is difficult to see what rational grounds Perrier has for his suggestion that a part
of the water- vascular system of a Starfish or Urchin is represented by an organ which
belongs to the blood-vascular system of a Crinoid, as his own observations show, though
he nowhere admits that such is the case. (See Appendix, Note F.)

The chambered organ of Comatula is contained within the cavity of the centro-dorsal,
and is covered in above by the rosette of metamorphosed basals ; it is a larger structure,
both relatively and absolutely, tlum that of a stalked Crinoid, owing to the concentration
of the cirri at the toji of the larval stem.

1 See Sladen, On the Homologies of the Primary Larval Plates in the test of Brachiate Echinoderms, Quaii.
Joiirn. Micr. Scl, 1884, vol. xxiv., N. S., pp. 32-34, pi. i. t^R. 16.

- Quart. Journ. Micr. ScL, 1883, vol. xxiii., J^. S., pp. 599-609. ^ Comptes rendas, t. xcviii. pp. 445, 446.

* Quart. Journ. Micr. Sci., 1884, vol. xxiv., X. S., p. 323.

REPORT ON THE CRINOIDEA. 107

lu Bathijcnnuff, Rhizocrinus, and Pcntacrinus the central vascular axis of the stem
consists of five peripheral vessels surroimding a core of smaller ones (PI. Vila. fig. 2 ;
PI. XXIV. figs. 2, 5 ; PL LVIII. fig. 3— c/t') ; and from these peripheral vessels are
derived the central vessels of the cirri borue hj the stem. In PentacHnus these cirri
are borne by special nodal segments which occur at more or less regular intervals all
down the stem. The five large peripheral vessels expand slightly in each nodal joint,
and each gives ofi" one cirrus-vessel (PL XXIV. figs. 3, 4, dm, cv; PL LXIL). Hence,
every nodal joint contains, as it were, a small edition of the chambered oi-o-an situated
in the calyx (PL XXIV. figs. 6, 8, ch ; PL LXIL).

In Comatula, however, the centro-dorsal represents physiologically "a coalesced
series of the nodal stem-joints in the stalked Crinoids,"^ and the downward prolonga-
tions of the chambers into the stem are ruptured when this organ is discarded. A
minute opening in the floor of each chamber close to the central axis remains to indicate
their former existence, whde a small aperture in the middle of the peripheral wall of
the chamber leads into a cirrus-vessel. The vessels of the remaining cin-i are derived
from those forming the central axis of the chambered organ. They pass outwards
horizontally beneath the chambers in five groups which are thus radial in position, as are
the earlier cirri and those on the stem of Pentacrinus (PL XXIV. fig. 4, cv ; PL LXIL).

In Actmometra imrvicirra, the only species of the genus in which I have made
horizontal sections through the calyx, the central axis of the chambered organ contains
only two vessels, instead of the larger number present in Antedon rosacea; and there
are fewer cirrus- verticils beneath the chambers. This is only what might have been
expected, from the reduced size of the centro-dorsal in this type and the small number
of cirri which it bears. In the Pentacriuoid again, with an undeveloped centro-dorsal
bearing only five cirri, the vessels of these organs are derived directly from the cavities
of the chambers, just as in the nodal joints of the stem of Pentacrinus.

Perrier has described the cirrus-rudiments as originatiua; from the " cordon central " of
the larval stem, and as alternating with the rays, i.e., as interradial in position.^ He has
given no figures in support of his statements, which are far from being in accordance
with the observations of M. Sars, Dr. Carpenter, and myself, as I have explained else-
where.^

The smaller size of the cavities of the chambered organ in the stalked Crinoids than
in the Comatulse, and the greater simplicity of its central axis, are obviously related to
the absence of a cirrus-bearing centro-dorsal.

Both in Rhizocrinus and Bathycrinus, so far as my experience goes, the axis of the
chambered organ is formed throughout of a single vessel (PL Vllb. fig. 2, V). In Penta-
crt'mw there is only a single vessel in the upper part of the stem (PL XXIV. figs. 2-5, T'^.

• Wyville Thomson, Phil. Trans., 18G5, p. 536. * Comptes rendus, t. xcviii. pp. 445, 446.

2 Quart. Journ. Micr. Sci., 1884, vol. xxiv., N. S., pp. 325, 326.

108 THE VOYAGE OF H.M.S. CHALLENGER.

This divides into two or more which pass upwards, like the chambers themselves into the
glandular organ above (PL XXIV. fig. 6 ; PI. LVIII. fig. 3—V; PI. LXIL). But a sieve-
like axis, such as occurs in Antedon, is absent in the stalked Crinoids, which have no
cirri just beneath the calyx that require to be supj^lied with blood.

E. The Genital Glands.

Since the discovery, made independently by Dr. Carpenter and Prof. Semper,^
that the so-called nerve of Miiller, which is situated between the skeleton and the water-
vessel of the Crinoid arm, is really a part of the generative system, our knowledge of its
minute anatomy has been largely increased by Ludwig.

He discovered that the actual genital tube, the epithelial cells of which develo]) into
ova (PL Vb. fig. 1 ; PL Vc. fig. 1 ; PL Villa, figs. 4, 5 ; PL LX. fig. &—gc.), is suspended
by fusiform or branched cells within a blood-vessel ; and he believed this vessel to arise
from the vascular plexus underlying the subtentacular canals of the disk.

Dr. Carpenter had previously come to the conclusion, as the result of his dissections,
that the genital cord or so-called rachis of the arm is a radial extension of this plexus.
He further believed the latter to be in connection with the ventral end of the plexiform
gland, having seen the division of this organ in the Pentacrinoid into five branches, one
of which passes to each ray, an observation which may be easily confirmed by examina-
tion of optical sections of a Pentacrinoid, soon after the appearance of its first whorl of
cirri. According to Perrier the body-cavity of a mature Pentacrinoid or recently
liberated Comatula contains no structures wliieh could be considered as blood-vessels ;
although he finds within the meshwork of connective tissue that occupies the body-cavity
"un petit nombre de cordons ceUulaires pleius qui se rendent manifestement aux bras"^
I have little doubt that these are a further development of the branches from the axial
organ of which one pas-ses to each ray, as described by Dr. Carpenter ; and that they
eventually give rise to the subambulacral plexus of genital vessels (PL LVII. fig. 3;
PL LX. figs. 1, 2—gv).

The genital tube of the arm is regarded by Ludwig as a sterile portion of a complex
genital orgau,^ the epithelial lining of which only develops into ova or spermatozoa in
particular places, usually within the pinnules. The ceUs lining it are relatively large, and
project into its interior, thereby reducing its cavity very considerably.

Dr. Carpenter, who was unacquainted with the blood-vessel ensheathing the true
genital cord, spoke of the latter as having a tubular wall and granular contents ;* and he
stated that in the plexus beneath the subtentacular canals of the disk of Antedon rosacea

' Kurze anatomische Bemerkungen iiber Comatula, Arb. zool.-zootom. Inst. Wiirzhurg, Bd. i. p. 260 ; translated,
with -MX Addendum by Dr. Carpenter in the Ann. and Mag. Nat. Hist., ser. 4, vol. xvi., 1875, pp. 202-209.
^ Comptes rendiis, t. xcviii. p. 445. ' Crinoideen, loc. cit., p. 293.

* Proc. Hoy. Soc. Land., 1876, pp. 220, 221.

REPORT ON THE CRINOIDEA. 109

the cavity with its granuhir contents bears a smaller proportion to the thickness of the
tubular wall. I suspect that this varies somewhat according to the sexual maturity of the
individual ; but I have generally noticed that the vessels of this plexus in Antcdon eschricJiti
do not show such a clear section as the visceral blood-vessels, their lumen being occupied
by cellular structures ; while in some disks of this species I have found as distinct a
genital tube within the vessels of this subambulacral plexus as is to be met with in the
arms between the bases of two successive pinnules.

Further, in one example of Antedon rosacea, I found a small but well-developed ovaiy
occupying the position of the genital plexus beneath the left posterior ambulacinam of the
disk. The first traces of it appear in the sections which pass through the hinder part of
the spongy organ ; and it extends outwards to the point where the primary radial groove
divides into the two which proceed to the arms. It contains the nuclei of half a dozen
ova in various stages of development, some with a germinal spot, and some without.

A still larger and more fully developed ovary occurs in the disk of one of the three
examples of Actinomctra lyuldiella which I have cut into sections. It commences close
to the peristome, and extends outwards beneath the left anterior ambulacrum nearly to
its bifurcation, lying close down upon the upper surface of the intestine, and moulded to the
plications of its Y\-all.

In Antedon carinata 1 have not only found a distinct genital tube wdthin some of the
vessels forming the plexus beneath the disk ambulacra, but I have also met with detached
portions of ovaries containing more or less fully developed ova in various parts of the
body-cavity, e.g., in the spaces of the connective-tissue netw^ork forming the lip ; in the
intervisceral portion of the body-cavity, between the two parts of the coiled gut ; and in
the subtentacular canals between the genital plexus and the water-vessels.

There can, therefore, I think, be hardly any question as to the relation betw^een the
genital glands and portions of the blood-vascular system ; while the occasional develop-
ment of rudimentary ovaries within the disk of recent Crinoids is of considerable
importance. For it shows that there is no morphological improbability in the theory which
supposes the genital glands of extinct armless types, like the Blastoids and Cystids, to have
been situated within the body, rather than in the so-called pinnules, even when these are
present, which is by no means always the case.

The fertile intra-pinnular portions of the genital glands vary considerably in shape.
In most of the British varieties of Antedon rosacea, in Antedon angitsticali/x, Antedon
acoela, and Eudiocrinus japonicus, they are short, thick, and rounded. They some-
times terminate in rounded ends, and are sometimes continued onw^ards as slender cords
through two or three pinnule-joints. But in the Antedon rosacea from Naples, and in
the group of species aUied to Antedon eschrichti, they are long and fusiform, extending
over several pinnule-joints. The same is the case in Hyocrinus (PL Vc. figs. 8, 10, t),
Bathjcrinus (PL YII. fig. 7; PL YIII. fig. 5), and Rhizocnnus (PL X. fig. 20), though to

110 THE VOYAGE OF H.M.S. CHALLENGER.

a less extent. The ovaries of the Peutacrinidas are likewise long and fusiform, some of
them a2)pearing to present somewhat anomalous characters. For in some sections which
were made for Sir Wyville Thomson by Dr. Stirling, the ovary appears in the arm,
occupying the usual position between the subtentacular and the cceliac canals where the
sterile genital cord is normally found. This is also the case in the lower parts of the
arms of Holopus (PI. Vc. fig. 2, ov), but I have not yet succeeded in discovering which
species of Pentacrinus or Metacrinus is distinguished by this peculiarity; for the sections
above mentioned were not labelled with any name or reference number. I have cut
sections of the arms of all the more common Pentacrinida^, but in none of them have I
found any such departure from the type of the ordinary Antedon as is presented by the
ovaries of this unknown species.

Many years since it was discovered by Prof Semper,^ during his residence in the
Philippine Islands, that the ovaries of Actinometra parvicirra [ = Actinometra armata,
Semper, M. S.) are not attached to the genital cord by their ends in the usual way. For
a backw^ard process is given off at the point where the short branch of the sterile genital
cord expands into the fertile poi'tion of the gland ; and it lies within the ventral perisome
of the arm on the proximal side of the pinnule attachment. This is as fertile as the
rest of the gland which is actually within the pinnule, so that the whole structure
apjjears to be attached to the genital cord at some little distance from its end ; and it
comes right down into the arm at the sides of the subtentacular and cceliac canals, being
attached almost directly to the genital cord (the so-called rachis), the lateral branches of
whicli are quite short. In most sections of the arms, therefore, an ovary is to be seen on
either side of the central genital cord (PI. LXI. fig. 3).

This condition also occurs in Metacrinus angidatus, and in other Philippine
Comatulse, e.g., Actinometra nobilis and Actinometra dissimilis;' and so far as one can
judge from the appearance of the ventral perisome, without cutting sections, I suspect
that it is tolerably common in the larger tropical Comatulae.

Although I have examined the ovarian pinnules of a large number of species, I have
never met with definite openings for the discharge of the ova ; and I must therefore,
like Ludw^ig, leave undecided the question of the origin of the relatively large openings
which occur on the inner side of the pinnules of Antedon rosacea at the time of
sexual maturity. On the other hand, I have found male indi^dduals in which the
testicular openings are evident enough (PI. LIV. fig. 3). In Antedon acoela and Antedon
angusticalyx, for example, the fertile part of the gland is short, thick, and rounded. It
only extends over four or five of the pinnule segments, and is protected by the tolerably
regular pavement of plates already described. At about the middle of its length one or
two small conical projections rise from it towards the ventral surface of the pinnule, and

1 Arh. zool.-zootom. Inst. Tfiirzhirg., loc. cit., fig. 1, p. 261.

2 a

2 The specific formula of this type is — a . 3 . ^ . 3 . 3 . y

REPORT ON THE CRINOIDEA. Ill

may be seen in an optical section of the decalcified pinnule to end in a small rounded
opening. One of these openings is shown in PI. LIV. fig. 3, and I do not think that, as
a rule, there is more than one to each pinnule.

In nearly aU the Crinoids which I have examined the structure of the genital gland
is the same as described by Ludwig in Antedon eschrichti. The epithelial lining of the
genital tube is continuous with that of the ovary. Individual cells of this lining enlarge
at the expense of their fellows, and are gradually enclosed in follicles which are derived
from the original ovarian epithelium. These follicles project freely into the lumen of the
gland ; but there is practically nothing which could be called an ovarian stroma.

In Actinometra nigra, however, there is a highly organised stroma separating the
follicles ; and young ova may be found in it at all stages of growth, the smallest being
no laro-er than the nucleolus of a mature ovum. But I have been unable to make out
the derivation of these germs from the epithelial cells as clearly as may be seen in Antedon
eschrichti and similar forms.

In Actinometra fxirvicirra, on the other hand, I have failed to discover any follicular
structure at all. The ova are smaller than in other types, and project slightly from the
surface of a dense brown stroma in which no organised structure can be made out
(fig. 4 on p. 113 ; PL LXI. figs. 3, 4). The genital cord of both these species (PL LXI.
figs. 5, 6) is very difi'erent from that of the other Crinoids, varying greatly in shape and
appearance. But the description and illustration of its peculiarities must be deferred to
another occasion.

In two other multiradiate species of Actinometra from the Philippine Islands,
however, Actinometra dissimilis and Actinometra nohilis, the genital cord of the arms
is much less complex, and closely resembles that of Antedon eschrichti, except that it
frequently contains a quantity of ill defined pigment masses. The ovaries have the same
structure as those of Antedon eschrichti and Antedon rosacea; and the ova which they
contain, exhibit exactly the same relation between yolk and yolk-sac as have been
described by Ludwig in the ova of the latter type.^ The whole ovum appears to be
sm-rounded with a network formed of dark threads and clear meshes. The latter,
however, are larger, and the former smaller and more delicate than in the ova of Antedon
rosacea, as described by Ludwig.

F. The Nervous System.

That portion of the organisation of a Crinoid which corresponds to the ambulacral
nervous system of other Echinoderms is of a somewhat limited character. So far as is
yet known, it is confined to the ciliated floor of the food-grooves and to its mmute
lateral extensions in the direction of the tentacles. "Andere Zweige als die schon

1 DiL- Bililung del- EiliiiUe bei Antedon rosacea, Zool. Anzeiger, Jahrg. iii., 1880, pp. 470, 471.

112 THE VOYAGE OF H.M.S. CHALLENGER.

erwithnten zu den Tentakelu sah icli nirgends von dem Nervcnstamm des Arms oder der
Pinnula abtreten." ^ Thus speaks Ludwig, who has given us a careful description of
tlie ambulacral nerves of Antedon eschrichtt, the type in which they are more completely
differentiated from the ectoderm than in any other Crinoid.

Immediately beneath the closely packed ciliated epithelium which lines the food-
groove is the band of nerve-fibrils, which thins away somewhat in the middle line
above the radial blood-vessel. It is covered by a very delicate sheet of connective
tissue on which the epithelial layer rests, and it is traversed vertically by delicate
threads of a similar nature which break up the whole nerve-band into bundles of fibrils
with numerous minute cells intercalated among them (PI. LX. figs. 4, 6, n). Ludwdg
was uualile to find this connective tissue sheet in Antedon rosacea, and I have not met
with it in any other Crinoid but Antedon eschrichti. All the other tyj)es that I have
examined have a much less defined nerve-band than this species (PI. Villa, figs. 4, 5 ;
PI. LVII. figs. 3, 4 ; PI. LIX. figs. 1, 5 ; PI. LX. figs. 1, 2— n) ; and the vertical fibres
which cross it are continuous with the extended lower ends of some of the epithelial
cells in the layer above. In fact, both Ludwig and myself have observed this absence
of a basement membrane and the connection of the epithelium with the vertical fibres
in some individuals of Antedon eschrichti, other sections of which j)reseut the appearance
described above.

Judging from Hamann's observations on the Asterids and Holothurians,^ and also
from those of Koehler^ on the Urchins, we may consider it certain that among these
" StutzzeUen " there are likewise sense-cells or neuro-epithelial cells, the inferior ends
of which are connected with nerve-fibrils.

Not having w^orked with sufficiently well preserved material, I have never seen them,
and they escaped the notice of Ludwig both in Asterids and in Criuoids ; but I have no
doubt whatever as to their presence.

As regards the ceUular elements of the Echinoderm nervous system, it is becoming
gradually recognised that nerve-cells of the usual well defined type are either altogether
absent, or confined to certain specially sensitive parts of the body. Neither Hamann nor
Senon have found anything but bipolar cells in the nervous system of Holothurians.
Similar cells have been described and figured by Koehler in the Urchins ; while, according
to Eomanes and Ewart*the cell protoplasm " is generally seen to project in two, or
sometimes in three directions." The radial nerves of the Asterids, according to Hamann,
contain numerous fusiform or bipolar cells, among which are a few with more than two
processes ; while larger cells, both bipolar and multij)olar, occur in the neighbourhood of
the terminal tentacle. Considering the reduced condition of the ambulacral nerve in the

^ Crinoideen, loc. cit., p. 264.

^ Beitrage ziu' Histologie der Echinodermen, Mitth. I., II., Zeitschr.f. kiss. ZooL, Bd. xxxix. pp. 146, 309.

3 Op. cit, pp. 50-54, pi. vi, fig. 47.

* Observations on the Locomotor System of Echinodermata, Phil. Trans., 1881, p. 836.

REPORT ON THE CRINOIDEA.

113

s.tc.

Criuoids, one is not surprised to find Ludwig limiting liimself to the following statement —
" Die winzigeu Zellen, die sich zwischen den Nervenfasern finden, sind vielleicht aucb nur
die Kerne von Zellen, die in den Verlauf der Fasern eingeschaltet sind ; " ' and my own
observations have not enabled me to add anything to this suggestion.

Hamann's observations on the Asterids have led him to discover that the so-called
ambulacral nerves are only specially developed parts of a general subepithelial plexu.3
containing ganglion-cells. This surrounds the whole body, and is traversed, though very
sparingly, by vertical fibres extending downwards from some of the "Stutzzellen"; whUe
sense-cells are intercalated among these at the bases of the respiratory caeca and elsewhere.
Hamann has found a similar ectodermic plexus, distinct from the radial nerve, in
Holothurians ; and the presence of a corresponding structure in Coelenterates, Nemertines,
Turbellarians, Chsetognatha, &c., is now thoroughly established. Thus, then, there
would seem to be no reasonable doubt respecting the existence of an ectodermic nerve-
plexus in the Crinoids. But even if the existence of this plexus be admitted, it is
difficult to suppose that the nervous system of a Crinoid with its innumerable pairs of
muscular bundles is limited to the subepithelial band in the floor of the food-grooves, its
tentacular branches, and the as yet undemonstrated
plexus. Leaving the latter out of consideration for the
moment. Where, I would ask, are the nerves of the pin-
nules of Antedon acoela and Antedon angusticalyx
which are shown in PI. LIV. figs. 1-3, 5 % Where is the
nervous system of the large number of arms on the two
posterior rays of the gigantic Actinometra magnijica
represented on PI. LVI. fig. 7, or that of the numerous
ungrooved arms on the three other rays 1

It isdiflicult to conceive that these ungrooved arms and
pinnules (PL LXI. fig. 3), which form so large a part of
the entire organisation of this animal, are entirely unpro-
vided with a nervous system, and yet the ambulacral nerve
is altogether absent (fig. 4). Even when this is present,
Ludwig himself admits that he has seen no other branches
proceeding from it than those which supply the tentacles.

Now the muscular system of a Crinoid is unusually
well developed. From the first radials to the ends of
the arms every two joints are united by a pair of

muscular bundles (PL LXL figs. 4, 5), except in the case of the syzygies and the rare liga-
mentous articulations. The pinnules are always united to the arms by muscles, and in many
forms there are muscles between several pairs of the pinnule-joints (fig. 6 on p. 121, m).

Fio. 4. — Diagrammatic transverse section of
au ungrooved ovarian pinnule of -Ic/mw-
tiuitra parvicirra, x 50. a, axial cord ;
a\ the branches proceeding from it ; c.c,
cceliao canal; ci.c, ciliated cup; c.s.,
connective tissue spaces in the perisonio ;
g.d., genital canal; g.v., genital vessel;
j, skeleton of the pinnule-joint; I.e.,
lateral canal connecting the cteliac and
subtentacular canals ; ov, ovary ; s.t.c,
subtcntacular canal ; w., radial water-
vessel.

' Crinoiileen, he. cit., p. 264.

(ZOOL. CHALL. EXP. — PAUT XXXII. 1S84.)

li 15

114 THE VOYAGE OF H.M.S. CHALLENGER.

Some species have as many as two hundred arms, each consisting of from
one hundred and twenty to two hundred joints ; while I have counted three
hundred joints in the arms of a large Antedon eschrichti. The regular and
graceful mode in which a ten-armed Crinoid swims is well known. The simul-
taneous flexions of the five right and of the five left arms alternately involve the
co-ordinated contraction of several hundred pairs of well defined muscular bundles, and
yet these are performed in entire independence of the ambulacral nervous system, with
which the muscles are in no direct connection. In fact the eviscerated skeleton which
has lost its disk and oral nerve-ring will swim as well as the entire animal. Whence
does it get this power ?

According to E. Hertwig ^ " Zunachst ist fiir mich die schon oben vertheidigte Grund-
anschauting maasgebend dass die Lcbenserscheinungen der Ctenophoren nicht gut ohne
die Annahme eines mesodermalen Nervensystems verstandlich sein mochten. Seitdem
durch die neueren Untersuchuugen mit sicherheit Nerven bei den Medusen und Actinien
nachgewiesen worden sind, ist kein Fall im Thierreich bekannt, in welchem complicirtere
und raschere Muskelbeweguns-en ohne sleichzeitige Anwesenheit von Nerven zu Stiinde
kamen. Soil ten die Ctenophoren in dieser Hinsicht eine Ausnahme machen ?"

Substitute for " Ctenophoren " the name " Crinoideen " in the above quotation, and
the question arises. Where is the co-ordinating centre of the muscular movements of a
Crinoid ?

A centre of this kind, if it exists in such a highly organised type as a Crinoid, cannot
but be regarded as belonging to a nervous system ; whereas a denial of its existence
brings us face to face with a physiological problem of much complexity. As a matter
of fact, however, there is both physiological and anatomical evidence for the existence of
such a centre, though the morphological difficulties which its presence involves are of
the most perplexing character.

The well-known experiments of Dr. Carpenter* have shown conclusively that the
fibrillar envelope of the chambered organ is the governing centre on which all the
muscular movements of the animal depend, and that the movements of each individual
arm depend upon the integrity of the axial cord of that arm. For they stop directly
it is injured, just in the same way as injury to the chambered organ causes all the arms
to be rigidly stretched out by the action of the dorsal elastic ligaments. The fibrillar
envelope of the chambered organ, therefore, is the centre of a nervous system, the peri-
pheral portion of which consists of tbe axial cords of the rays, arms, and pinnules, and of
the numerous branches proceeding from these cords.

The occurrence of this fibrillar tissue in the stem and cirri (PI. XXIV. figs. 1-5, ca.),
and also in the ventral perisome, whether bare or plated (PI. LIX. figs. 2-4, 6, 7;

' Ueber deu Ban der Ctenophoren, Jenaisclie Zeitschi:, Bd, xiv. p. 437.
2 Proc. Boy. Soc. Loud., 1876, p. 453.

REPORT ON THE CRINOIDEA. 115

PL LX. fig. 2—a.d.; fig. G, a'. Figs. 4, 5, 7, «'; fig. 8, a.d.), where muscle fibres are absent,
indicates that if it be nervous, it must be not only of a motor but also of a sensory nature.
This conclusion also follows from the fact that stimulation of one of the oral pinnules of
Antedon rosacea causes the flexion of all the ten arms. This result is not simply due to
general irritability ; for if so, it would follow when any pinnule was stimulated ; whereas
stimulation of one of the ordinary pinnules is only followed by flexion of the arm which
bears it. This experiment therefore is evidence of a reflex action of a somewhat complex
nature ; and the axial cords must be the paths of both afferent and efi"erent impulses.
For there is no ambulacral nerve in these oral pinnules, which resemble those on the
hinder arms of Actinometra (PL LVL fig. 7 ; PL LXI. fig. 3) in being ungrooved, and
devoid of tentacles, blood-vessels, and ventral nerve (fig. 4). The latter is normally
connected with the tentacles, and possibly also with a general subepidermic plexus ;
but it has nothing whatever to do with the bodily movements of the animal, though
perhaps influencing those of the tentacles and of the marginal leaflets or covering plates.

Here then we have evidence in the Crinoid of a mesodermic nervous system analogous
to that which has been discovered of late years in the Ccelenterates, Worms, and
Chsetognatha. As regards the latter group, 0. Hertwig is inclined to think that " bei
den Chsetognathen sensibles und motorisches Nervensystem von einander vollstiindig
gesondert sein, ersteres wiire ektodermal, letzteres gleich den Muskeln mesodermal." ^

Considering that nervous tissues are well develo^jed in the mesoderm of Ccelenterates,
one would certainly exj^ect to find them in that of Echinoderms. The nervous system of
a Holothurian only remains in connection with the epidermis at the distal ends of the
entacles and tube-feet ; w^hile the radial nerves of Ophiurids and Urchins are separated
from the exterior by limestone plates, though coming into connection with the epidermis
on the tube-feet. Besides the subepidermic plexus on the outside of the shell of an
Urchin which sends fibrils to the muscles of the pedicellarise, there is another which is
formed by filaments that are given oS from the lateral branches of the radial nerves, the
connection of which with the subepidermic plexus has not been definitely traced.
Eomanes and Ewart have further discovered that the general co-ordination of the spines
for the purpose of locomotion depends on the integrity of an internal nervous plexus
which is " everywhere in intimate connection with the external, apparently through the
calcareous substance of the shell." ^

There is, therefore, no very great difiiculty involved in the belief that a mesodermic
nervous system is present in the Crinoids. The morphological difliculties resulting from its
anti- ambulacral position are, however, considerable. But they are of precisely the same
character as we have to face, when describing the chambered organ and the vascular axis
of the stem as a part of the circulatory system of a Crinoid. Ludwig appears to hiwc

^ Die Chcetognathen, Jenaische Zeitschr., BJ. xiv. p. 234.
2 Phil. Trans., 1881, p. S74.

116 THE VOYAGE OF H.M.S. CHALLENGER.

found no difficulty in this, but despite the physiological improbability of the want of
innervation to the muscular bundles, he declines to accept the fibrillar envelope of the
chambered organ and its extensions into the stem and arms as belonging to the nervous
system of a Ciinoid. One ground for his objection lies in the presence of this fibrillar
tissue in the axis of the stem and of the cirri that it bears, which contain no muscular
tissue. This point, which supports Dr. Carpenter's view rather than opposing it, will be
considered later.

The only explanation of the presence of this fibrillar tissue within the skeleton which
Ludwig can suggest is as follows : — " Die Faserstriinge sind zu betrachten als unverkalkt
gebliebene Theile der bindegewebigen Grundlage der Kalkglieder, deren Aufgabe es ist,
aus dem Blutgefiissystem, genauer aixs den funf Kammern, die eruiihrende Flussigkeit
aufzunehmen und den Arm- und Pinuulagliedern zuzuftihreu." ^

I have no doubt whatever that the axial cords are permeated by a nutritive fluid,
which finds its way from the chambered organ into the substance of the organic basis of
the skeleton. For I have frequently found coagulum, both in this last and in the axial
cords themselves ; but I should hesitate to speak of either of these as connective tissue.
The closely set fibrils forming the axial cords are quite distinct from the nuclear network
which interpenetrated the calcareous substance ; and both are as diflerent as possible
from the connective tissue fibres of the articular ligaments, or the general connective
tissue of the ventral perisome (see fig. 6 on p. 121). Simroth has given an excellent
description and figures of the nucleated reticulum forming the organic basis of the skeleton
in Ophiurids,' and nearly everything which he says is equally apj^licable to the Crinoids.
The nature of this nuclear tissue is well shown in PL Vb. fig. 1, PI. XXIV., and PI. LVIII.
figs. 2, 3 ; and its distinctness from the close fibrillar structure of the stem-axis is very
aj)parent. The general aspect of the axial cords in thin sections is identical with that of the
Ophiurid nerves as represented by Simroth, who finds the nerve-fil )rils to be connected
with bipolar cells. Fig. 32 on his pi. xxxiv., which shows the relation of the radial nerve
to the organic substance of the under arm-plates beneath it, would serve, with very little
alteration, for a part of a longitudinal section of the arm or pinnule of a Crinoid. The
fibrillation of the axial cords and their marked difierences from the organic basis of the
skeletal plates is well seen in sections through the calyx of the Pentacrinoid. At this
early stage the cords lie on the ventral surface of- the flat calyx plates outside the organic
basis of the skeleton altogether ; and it is by an endogenous thickening of the calcareous
substance of these plates that the cords come to lie in grooves which are subsequently
closed into canals ; while by a further continuance of the same process these canals are
eventually so surrounded by calcareous tissue that they come to occupy the centre of the
successive joints of the skeleton.

Ludwig's view of the nature of the axial cords is therefore not altogether in accordance

' Crinoideen, loc. cit., p. 340. ^ Op. cit., pp. 433, 434.

REPORT ON THE CRINOIDEA. 117

with what we know of their origiu ; and it likewise fails to account for their relations in
the adult Crinoid. If this fibrillar tissue were limited to the skeleton, there might be
some reason in Ludwig's suggestion. But it gives no rational explanation whatever for
the extension of branches from the axial cords of the skeleton through the perisome of
the disk and arms, up to the bases of the tentacles at the sides of the food-groove
(Figs. 4-7, a'. Fig. 8 ; PL LIX. figs. 2-4, 6, 7 ; PL LX. fig. 1—ad ; PL LX. fig. 6, a'), and
even as asserted by Perrier, into the tactile hairs l^orne by these tentacles.^

Ludwig's theory too entirely fails to account for the elaborate arrangement of
commissures which one finds in Comatula and Pentacrinus (PL XXIV. figs. 7-9 ;
PL LVIII. figs. 1-3), and in a less degree in Bathycrinus (PL Vllb. fig. 4, ceo),
Rhizocrinus (PL Villa, fig. 6, ceo, ico), and Encrinus. Why should the first
radials and the axillaries be in such special need of nutrition that the former should
possess both interradial and intraradial commissures, and the latter no less than
fovir cords, to say nothing of the transverse commissure ? Five radial cords starting
directly from the envelope of the chamljered organ would surely serve aU the necessary
purposes of nutrition. As it is, however, each ray and indeed each arm is supplied by
fibres from two of the primary interradial trunks. This complex arrangement receives
no explanation whatever on Ludwig's theory, though it is easily understood if we suppose
that the axial cords are the means by which co-ordinated impulses reach the muscles
from a governing centre.

Their anatomical structure also favours this view. In a paper which was published
some years before the discovery of ambulacral nerves in tlie Crinoids, Baudelot quoted
MiiUer's description of the so-called arm-nerve {i.e., the genital cord), and apparently
adopted it as correct.^ But he also stated that he could not help being struck with the
resemblance " qui existe entre la structure du cordon fibreux central des bras et la cordon
nerveux des autres Echinodermes." He described its relations pretty accurately, and
then proceeded to say " Ainsi done chez les Comatules il existe des parties qui e\ddemment
n'appartienuent point au systeme nerveux, et qui dans leur disposition aussi bien que
leur structure ofirent une analogic presque complete avec les cordons nerveux des autres
Echinodermes."

I do not know what reason Baudelot may have had for his conviction that the axial
cords are evidently not of a nervous nature, unless he had implicitly accepted Miiller's
account of the nervous system of a Crinoid. A very little troul)le, however, would have
convinced him that this was totally incorrect. In fact Dr. Carpenter had referred to
MiiUer's error four years before the publication of Baudelot's observations, and had also
mentioned that he had reasons for regarding the branching fibres proceeding from the
axial cords to the muscles as probably having the function of nerves. Had Baudelot

' Comptus rendus, t. xcvii. p. 188.

2 Contribution k rhistoire du systJjme nerveux des :6chinoJerme3, Archives d. Zool. exp^r., t. i. p. 211.

118 THE VOYAGE OF H.M.S. CHALLENGER.

been acquainted v/itli these facts, the non-nervous nature of the axial cords would
probably have been somewhat less " evident " to him.

According to Ludwig the axial cords consist of " fciiicn Fasern, zwischen welchen
man, namentlich an der Peripherie der ganzen J\Iasse, Zellen odcr doch zellenahnliche
Gebilde (Zellkerne ?) findet." * As a matter of fact I can find no difference between the
appearance of the fibrils forming the axial cords and those of the ambulacral nerve,
either in transverse or in longitudinal section ; and I wonder that Ludwig was not struck
by the resemblance of the two, especially in Antcdon eschrichti, in some specimens of
which, at any rate, it is very marked. There are the same delicate fibrils with inter-
calated cells as in the ambulaci'al nerve, and in some individuals the two have exactly
the same appearance in cross-section ; though the axial cords more usually are some-
what of a yellowish tinge, which renders it easy to recognise their branches that extend
outwards from the skeleton into the connective tissue of the genei'al jDerisome (figs. 4-8 ;
PI. Vllb. figs. 6, 7 ; PI. LIX. figs. 2-4, 6, 7 ; PL LX. figs. 2, 6).

Dr. Carpenter's theory of the nervous nature of the axial cords of the arm was
originally suggested by his discovery that they give ofi" branches which extend over the
ends of the muscular bimdles. This is well seen in moderately thick transverse sections
of an arm which are viewed as opaque objects. But the study of thin transparent
sections shows that these branches to the muscles are only portions of a largely developed
network which originates in the axial cords and extends both to the dorsal and to the
ventral surface of the arm or pinnule. Ludwig states that he had been unable to
convince himself of the existence of the muscular branches described by Dr. Carpenter.
This may well have been the case in the small arms of Antedon rosacea; though I have
had no difiiculty in finding them in this species, and Perrier has been equally successfuL
But I cannot comprehend his not having seen some of these lateral extensions of the
axial cords in the arms and pinnules of Antedon eschrichti. They are not limited to the
skeleton, for I have hardly a section that does not show a part of one or other of the two
main trunks which extend up into the ventral perisome at the sides of the food-groove,
as represented in PI. LX. fig. 6, a'. Pinnule sections too may be obtained without
difficulty, in which the whole course of one of these branches may be seen from its origin
in the axial cord right up into the substance of one of the respiratory leaflets bordering
the food-groove. In Ludwig's figures of sections through the arms and pinnules,
however, the axial cord is represented as a mere dark circle without any trace of lateral
extensions.

The doctrine of the nervous nature of these cords has recently received support from
a quarter in which it was formerly denied ; for Prof. Perrier has reinvestigated the
subject and has brought forward additional evidence of much value. He has seen the
branches of the axial cords in Antedon rosacea, and states, like Baudelot and Teuscher,

1 Crinoideen, loc. cil., p. 316.

REPORT ON THE CRINOIDEA. 119

that tliey have all the appearance of being true nerves.^ He further describes how their
ultimate subdivisions " aboutissent h, dcs cellules ^toil^es, dont chacune sc prolonge en
un fibre musculaire. Des ramifications de ce genre sont egalement en rapport avec les
fibres que contiennent les tentacules ambulacraires et dont un grande nombre so tiennent
dans les papilles sensitives de ces tentacules que Ludwig considere h. tort comme creuses."
Elsewhere he states that the cords are invested with a sheath of stellate cells which are
themselves related to connective-tissue corpuscles, and through these with the ectodermal
cells of the arm. These statements of Perrier's are of consideruljle importance, and
should his observations be confirmed the nervous nature of the axial cords will, I think,
be at last admitted, even by those whose scheme of Echinoderm morphology is founded
upon the archetype of a Stellerid or Urchin. These, however, arc formed almost entirely
upon the left larval antimer, whereas the chambered organ of a Crinoid and its downward
extension into the stem are formed in the right peritoneal tube. (See Appendix, Note G.)

The branches from the axial cords of the rays and arms, to which allusion has so
frequently been made, vary considerably in their development and distribution. Among all
the numerous Crinoids, stalked and free, that I have examined, Rhizocrinus is the only one
in which these branches have not been visible. I see no reason to doubt their existence,
however ; but the genus is one of small size, and is also permanently fixed through life, so
that one would not expect to find large muscular branches proceeding from the axial
cords, as in the Comatulse and Pentacrinidse which are free or semi-free, and can use their
arms for the purpose of swimming ; whereas, according to Agassiz,^ the movements of
extension and fiexion of Rhizocrinus are but slow and gradual. The branches are also
poorly developed in the massive and sessile Holopus. But in Bathycrinus, in the
Pentacrinidse, and in the Comatulas they are very largely developed, occurring not only
in the arms and rays but also in the stem and cirri. They vary considerably in their
extent, some portions of the stem showing them al)undantly( PL XXIV. fig. 2, ca'),
while in others they are less numerous. An optical section of two decalcified stem-joints
oi Bathycrinus aldrichianus is shown in PL Vila. fig. 1. The larger branches of the
axial cord {ca') are seen with a low power where the radial spaces render the stem-
substance more transparent than elsewhere, Init this gives no idea of the minuteness and
complexity of their subdivision, which only reveals itself by the use of a high power.

In Pentacrinus wynille-thonisoni again I have found these branches to be abundantly
developed in some stem-joints and almost entirely absent in others (PL XXIV. figs. 1-5).
The fibrillar envelope surrounding the vascular axis is sometimes in immediate contact
with the reticular tissue which forms the organic basis of the skeleton (figs. 2, 3). But
in other joints it is closely surrounded by a layer of large pigment bodies like those
which occur scattered in more or less abundance through the skeletal tissue (fig. 5, jj).
In other sections again, fibrillar extensions of the central axis pass outwards from it

1 Comptes rendus, t. xcvii. p. 188. - HI- Cat. Mus. Comp. ZoOL, No. viii. p. 29.

120 THE VOYAGE OF H.M.S. CHALLENGER.

between the surrounding pigment bodies, and ramify in the small spaces left between
the petaloid areas that are occupied by the five interradial ligaments (PI. XXIV. fig. 1 ;
fig. 2, ca'). These extensions towards the surface of the stem eventually become so fine
that I have been unable to trace them in the somewhat thick sections with which I have
been obliged to content myself, owing to the large size and the toughness of the stem
ligaments. But from what I have seen in the arms, and more especially in the pinnules,
I have very little doubt that the ultimate subdivisions of these branches are in connection
with a subepidermic plexus. I have found similar branches in the cirri.

The occurrence of this tissue around the axial vessels of the stem and cirri has been
employed by Ludwig as an argument against its nervous nature on account of the
absence of muscles in these organs.^ " Wozu also ein dieselben in ihrer ganzen Lange
durchziehender (motorischer) Nervenstrang?" The same argument might be employed
with respect to the extension of branches from the axial cords within the calyx up into
the anambulacral plates on the sides and ventral surface of the disk of Pentacrimis
decorus (PL LIX. figs. 2-4, ad). But their presence is readily understood if we consider
them as sensory nerves establishing a communication between an ectodermic plexus and
the axial cords of the rays, which all commence in the envelope of the chambered organ
situated within the calyx. It is however in the arms and pinnules that the lateral
extensions of the axial cords are most evident (PI. LXI. fig. 6), and the inadequacy of
Ludwig's theory as to their natiire is strikingly manifest.

The species in which I have found these branches best develo^^ed are Bathycmius
aldrichianus, Pentacrinus decorus, Antedon eschrichti, Actihometra ixtrvicirra, and
Actinometra nigra. But they may also be met with in less abundance in the small
arms and pinnules of Antedon rosacea, and I cannot understand how they escaped the
notice of Greefi", Teuscher, and Ludwig.

The lateral branches from the central fibrillar axis in the stem of Bathjcrinus
aldrichianus have been already mentioned (PI. Vila, fig. 1, ca'). The axial cords within
the rays and arms have a similar extensive distribution. Numerous branches proceed
outwards from them into the calcareous substance of the successive joints, as is shown in
the second and axilliary radials, which form an important part of the cup enclosing the
visceral mass (PL Vllb. figs. 6, 7, a'). The arms become free higher up, however, and
the deep median groove in the ventral surface of the skeleton receives the ambulacrum
with its armature of covering plates (PL VII. fig. 8). The axial cord gives off a large
branch on each side which proceeds upwards, subdividing freely as it goes. The ultimate
branches, many of which have bipolar cells intercalated in their course, extend right up
to near the top of the side-walls of the arm-groove, where they become so small that I
have been unable to trace them further (PL Villa, figs. 4, 5, a'). I have occasionally
seen a cell with three processes instead of two, but these are rare. The ambulacral nerve

1 Crinoideen, loc. cit., p. 335.

EEPORT ON THE CRINOIDEA.

121

varies considerably iu appearance, being relatively thick in some parts and almost totally
absent in others ; for I have never found it to extend right across the food-groove as it
does in other species (PI. Villa, figs. 4, 5, /;).

I have cut sections of the arms and pinnules of some half dozen species of Actinometra
f nd have found abundant extensions from their axial cords in all cases. The fibrillar
tissue generally has a strong yellowish tinge, which renders it easy to follow when it leaves
the substance of the skeleton and enters the ventral perisome. Actinometra parvicirra
and Actinometra nigra, both from the Philippines, have, however, given me the best results.
The axial cord enlarges slightly in the centre of each joint of the ray, arm, or pinnule.

J a'

Fig. 5. — Diagrammatic transverse section through the Fig. 6. — Longitudinal section of a pinnule-joint of

end of a grooved pinnule of Actinometra parvi- Actinometra nigra, x 50. «, axial cord ; a', the

cirra, x 70. a, axial cord ; a', the branches pro- branches proceeding from it ; c.s., connective tis.'^ue

ceeding from it ; a.e., ambulacral epithelium ; 6, spaces in the perisome ; e, epidermis ; j, skeleton

radial blood-vessel ; c.c, cceliac canal ; ci.c, ciliated of the pinnule-joint; W, dorsal ligament; li,

cup ; C.S., connective tissue spaces in the perisome ; interarticular ligament ; m, muscle.

j, skeleton of the pinnule-joint ; in', transverse
muscle-threads in the water-vessel ; n, radial trunk
of the ambulacral nervous system; s.^.c, subten-
tacular canal; w, radial water- vessel ; t.b., its
tentacular branch.

and gives ofi" four branches, or occasionally more (PI. LXI. fig. 6 ; woodcuts, figs. 4, 5)
One pair of these runs towards the dorsal surface, and breaks up into successive sub-
divisions, the last of which are exceedingly fine and can be traced no further (woodcut,
fig. 6, a'). The other pair extends towards the ventral side of the skeleton, and passes
out of it into the perisome at the sides of the genital glands, where they are continued
upwards towards the ventral surface of the arm or pinnule, which may or may not bear
a food-groove, according to the part of the body from which it comes.

In both these species the perisome contains a number of more or less regularly
arranged spaces in the connective tissue (woodcuts, figs. 4-7, cs.), and the branches of the

(ZOOL. CHALL. EXP. — PAET XXXII. — 1884.) I' 1*^

122

THE VOYAGE OF H.M.S. CHALLENGER.

axial cords run in tlie partitions between them. These branches have no regular mode of
subdivision, no two pinnules being exactly alike ; while they arc not symmetrical on the
two sides of the same pinnule. Longitudinal sections of the pinnules of Actinometra
nigra ' show that these branches which come up to the ventral side in successive segments
of the pinnules (woodcuts, figs. 4-7, a') are united by continuous trunks that run along
the upper surface of the pinnule right and left of the ambulacrum (woodcut, fig. 7, It);
they send branches upwards between the connective tissue spaces, of the same kind as
those which appear in transverse section.^ The nervous structures in these pinnules thus
consist of four principal trunks, three of which are intimately united Ijy a network of
fibrils, while the fourth and smallest is the ambulacral nerve. This is not yet known
to be connected with any other structures, though I strongly suspect that it is continuous
at the sides of the food-groove with a subepidermic plexus covering the pinnule and
communicating with the numerous branches of the axial cord.

Fig. 7. — Longitudinal section of the ventral peiisome in a pinnule of Actinometra nirjra, x 60.
a'. Ventral ascending branch of the axial cord ; c.s, connective tissue spaces in the perisome ; ct, connective tissue above

the ovary ; e, epidermis ; Ic, lateral canal, connecting the cceliac and subtentacular canals ; It, one of the lateral trunks

which connects the ascending branches of the axial cord (a').
N.B. — This section passes to one side of the medio-ventral line of the pinnule.

Sections of the arms and pinnules of Antedon eschrichti give much the same results.
The ventral branches of the axial cords in the arms extend upwards along the sides of
the coeliac canal, curve round the outer walls of the subtentacular canal, and pass on into
the elevated folds of tissue bounding the food-groove, as shown in PI. LX. fig. 6, a'.
They do not seem to subdivide so freely as in the tropical Actinometra, but in both
genera I have traced their smaller fibrils into the little respiratory leaflets along the edges
of the food-groove. Pcrrier has seen the same thing in Antedon rosacea, and believes

1 Theel's figure of a dorsal nerve-trunk in Elpidia glacialis with its muscular branches {K. Svensk. Vetemk. Akad_
Handi, Bd. 14, No. 8, Taf. iv. fig. 18) has a wonderful resemblance to a longitudinal section of the axial cord in a
Crinoid pinnule.

2 Woodcuts figs. 4 and 5 are composite figures made up from tlie stud}' of some half dozen sections through the
central part of a pinnule-joint and tlie overlying ventral perisome. Woodcuts figs. 6 and 7, however, are diagrammatic
representations of single sections, and I have plenty more of the same character.

REPORT ON THE CRINOIDEA.

123

liimself to have traced the connection of the ultimate fi1)rils with those in tlie tactile
paj^ilJae of the tentacles.'

The discovery of this extensive perisomic or paramhulacral network, derived from the
axial cords of the arras and pinnules in various species both of Antedon and of
Actinometra, \iii\ nie to suspect its presence at the sides of the disk-amljulacra ; and
after several unsuccessful attempts, chiefly due to the poor state of preservation of my
material, I met with one disk of Antedon eschrichti which j'ielded the most satisfactory
results. Portions of two sections are shown in PI. LIX. figs. G and 7 ; while woodcut
fig. 8 embodies the result of my studies of a few successive sections in the same series.

g-v.

'-^^

Fig. 8. — Diagrammatic transverse section of an ambnlacruni on the disk of Antedon eschrichti, x 70.
a.d., The parambulacral nervous networli — this is filled in from a few successive sections, only isohated portious of it being
visible in any single one; a.c, ambulacral epithelium; b, radial blood-vessel; g.v., genital vessel; n, radial or
ambulacral nerve, the subepithelial baud; sa., sacculi ; Stc, subtentacular canal; t.b., tentacular branch of w, the
radial water-vessel ; u'./'., water-pores.

There appears to be a good deal of individual variation ; l)ut in this one species, at
any rate, the elevated folds of perisome which bear the ambulacra contain a wonderfully
rich network of delicate fibrils of precisely the same nature as those which occur at the
sides of the brachial amliulacra (PI. LX. fig. 6, a,') ; and the brachial plexus may be
followed down on to the disk at the sides of the food-groove (woodcut, fig. 8, a.d.).
I have very little doubt that it is joined by branches which proceed upwards into the
ventral perisome from the axial cords within the radials and lower ])rachials. But as

1 Comptes rendus, t. xcvii. p. 188.

124 THE VOYAGE OF H.M.S. CHALLENGER.

I have never cut through an entire Antedon eschrichti, I am unable to say positively that
this is the case, though it certainly is so in Pentacrinus decorus.

This parambulacral network extends right down along the food-grooves, being especially
developed among the sacculi (woodcut, fig. 8, a.d.); and it forms an annular plexus in
the connective tissue of the lip, but of course farther from the mouth than the ambulacral
nerve-ring. Bij^olar, and occasionally multipolar cells arc in connection with its finer
fibrils, which can be followed very close to the superficial epithelium. Hardly any traces
of it are visible in the interpalmar areas between the ambulacra, which are chiefly occupied
by the water-pores, though it is extensive enough at their sides. I have seen it more or
less satisfactorily in various other disks of Antedon eschrichti, in Antedon rosacea, and
in Antedon carinata (PI. LX. fig. 2, ad) ; and I have no doubt that the action of suitable
reagents upon fresh material would give very valuable results. [See Appendix, Note G.]
In the mean time I would draw attention to Hamann's figures of the " zu der
Epidermis abgehenden Nervenziige, die man kurzweg als Hautnerveu bezeichnen kann "
in Synapta digitata} If the structures described above as forming the parambulacral net-
work in Antedon eschrichti, Actinometra parvicirra, Actinometra nigra, &c. (PL LIX.
figs. 6, 7 ; woodcuts, figs. 4, 5, 7, 8), be not " Hautnerven," I am entirely at a loss to
understand their nature.

The same filjrillar threads appear in the disk of Pentacrimis, not only in the plates of

its ventral surface, but also in those which are developed on the perisome uniting the rays

and support it below. Many of these plates, including those on the anal tube, are

produced into small Ijlunt spines, and these " Hautnerven " extend from plate to plate,

sending delicate fibrils up into the spines, as shown in PI. LIX. figs. 2-4, ad. I have

many sections which contain these filires in the plates on the sides of the disk, and they

are evidently derived from the axial cords of the rays and arms, which give off numerous

branches. Even in the basal plates I have found branches extending from the axial cords

towards the surface of the skeleton, as shown in the diagrammatic figure on PI. LXII. ;

and a curious modification of this occurs in one of the basals of the Pentacrinus wyville-

^/io?nson{ which was devoted to anatomical research. The two secondary cords (PL XXIV.

fig. 7, ar) which result from the bifurcation of the primary interradial trunk {cii) and

eventually enter diff"erent radials, are united to one another within the substance of the

basal plate by a commissure. This reminds one at once of the horizontal commissure

discovered by Ludwig in the radial axillary, by which the individual cords of the two

arms borne by the axillary are united immediately beyond their point of separation.

The arrangement of the axial cords within the radials of Pentacrinus or Metacrinus
is exactly the same as in the Comatulfe. The primary trunks which proceed from the
angles of the chambered organ (PL XXIV. fig. 7; PL LVIII. figs. 1, 3, ai) fork within
the basals ; and the two secondary cords which result from the bifurcation pass out from

> €p. cit., Zeitschr. f. wiss. Zool, Bd. sxxix. p. 322, Taf. x.xii. figs. 43-45.

REPORT ON THE CRINOIDEA. 125

the basals and enter two adjacent radials (PL XXIV. figs. 7-9; PL LVIIL figs. 2, .3, ar)
There are thus two apertures on the upper surface of each basaL and two on the under
surface of each radial (PL XIL figs. 11, 14, 22, 25; PL XVIII. figs. 5, 7; PL XX. figs. 2,
3, 6, 9 ; PL XXI. figs. 6a, 6b, 6c, 7a, 7h ; PL XXX. fig. 5, 7, 8 ; PL L. fig. 5). As the
two cords which enter each radial converge towards its distal surface, each of thena is
joined laterally by a commissure to its fellow in the next radial which springs from the
same primary trunk as itself These lateral branches form the interradial, and by far the
larger portions of a circular commissure which unites the five pairs of cords within the
radials (PL XXIV. fig. 9, c.co ; PL LXII.) ; while the two converging cords within the
substance of each radial are also united by a very short intraradial commissure (PL XXIV.
fig. 9, i.co).

This circular commissure occupies a canal which traverses the radials from side to
side, lying in the Comatulse very close to their distal face, but more centrally in the
Pentacrinidse. Its openings on the lateral faces of the radials are shown in PL XIL
figs. 11, 22 ; PL XX. fig. 6 ; PL XXL figs. 6a, 6h, 6c ; PL XXX. figs. 5, 8.

The two secondary cords within the basals and the proximal parts of the radials of
Pcntacrinus are more widely separated than the corresponding structures in the Comatulse
(PL XXIV. figs. 7, 9 ; PL LVIIL figs. 1, 2 — ar), so that the intraradial commissure is
better defined (PL XXIV. fig. 9, i.co). Beyond the circular commissure the two axial
cords of each ray lie very close together, though still distinct, just as in the Comatula3.
Each of them forks in the axillary, and there is the same horizontal commissure as was
described by Ludwig in Antedon rosacea.

So far as the fossil Neocrinoids are concerned, Encrinus and Apiocrinus seem to have
had an arrangement of the axial cords essentially similar to that prevailing in the
Pentacrinidae and Comatulse. In the former genus, as already pointed out by Ludwig,
no canals have been described either for the intraradial commissure or for the horizontal
commissure in the axillary ; while the secondary radial cords remain distinct as far as
the axillaries, even to the extent of being lodged within separate canals in the second as
well as in the first radials. The double canals are continued through the whole length
of the arms ; and in some species these consist, after the base, of a double row of joints,
each of which is pierced at its inner end by the two canals. The absence of any canals
which could lodge an intraradial commissure is very singular ; and it is also remarkable
that the interradial portions of the circular canal should lie so completely in the distal
parts of the radials. They only join the axial canals where these open on the distal
faces of the radials ; so that if an intraradial commissure were present at all, it must have
lain just at the edge of the first radials, almost among the ligamentous bundles uniting
them to the following joints.

The axial cords of Encrinus were lodged in canals throughout their whole length,
those within the basals occupying grooves within the substance of the plates which were

126 THE VOYAGE OF H.M.S. CHALLENGER.

covered and converted into canals by what Beyricli calls "an inner epiphysis."' This was
of no great thickness, and was therefore easily worn so as to expose the grooves beneath.
In Ajnocrinus, however, the basals were simply grooved for the reception of the
bifurcating interradial cords, though the radials were pierced by canals as usual.
Beyrich ^ speaks of the arrangement of the canals being the same as in E^icrirms, and
de Loriol refers to the circular canal ; ^ but I have been unable to make out definitely
whether any intraradial commissure were present or not. At any rate the tj^je resembles
Pentacrinus rather than Encrinus ; for there is only one opening on the distal face of each
radial instead of two, and in correspondence with this only a single series of arm -joints.

We know nothing respecting the distribution of the canals in the calyx of the
Bourgueticrinidse, but the course of the axial cords in Rhizocrinvs is somewhat different
from that of Apiocrimis, and this is still more the case in Bathycrinus. The basals of
RMzocrinus are of considerable height (PI. IX. figs. 1-3 ; PL X. figs. 2, 3 ; PL LIII.
figs. 7, 8), and the primary interradial cords of greater length than usual ; but they are
completely enclosed in canals, and bifurcate immediately beneath the synosteal surface
on which the first radials rest. The two limbs of each fork are very widely separated,
turn off horizontally, and form themselves the interradial portion of the circular
commissure, instead of proceeding directly onwards through the radials, as their fellows
do in Pentacrinus (PL XXIV. figs. 8, 9, ar). The two secondary cords which enter the
small radial are immediately united just within its inner face by an intraradial
commissure (PL Villa, fig. 6, c.co), and then proceed onwards towards the single opening
on the distal face (PL X. figs. 1-4). The above description differs in some points from
that given by Ludwig, who took an entirely erroneous view with respect to the basals of
this type, and failed to find the intraradial commissure. The subject is discussed more
fully in the anatomical account of the genus (pp. 249-252).

A stUl simpler condition than that of Rhizocrimis is presented by the aberrant genus
Bathycrinus. The basals are low and the radials high (PL VII. fig. 2 ; PL Villa, fig. 1),
exactly the reverse of what we meet with in Rhizocrinus (PL IX. figs. 1-3 ; PL X.
figs. 2, 3 ; PL LIII. figs. 7, 8). The primary interradial cords (PL Vllb. figs. 2, 3, m)
do not fork within the basals, as is usually the case ; but they pass upwards between every
two radials, the sides of which are grooved for their reception (PL VII. fig. 6a). At the
level of about half the height of the radials the primary cords divide, and the two branches
of each pass off right and left into the radials, where they form an interradial commissure,
from the angles of ^^■hich the axial cords of the rays proceed (PL Vllb. fig. 4, c.co) ; while
the two converging portions of the interradial commissure within each radial are united by
an intraradial commissure just as in Rhizocrinus. The principal difference between the two
ty|)es is that the primary interradial cords of Rhizocrinus fork within the basals, whUe
those of Bathycrinus pass upwards between the radials, and then turn off laterally witliin

1 Op. cit., p. 22. 2 02). cit., p. 21. ^ Paleoiit. Frang., op. cit., p. 313.

REPORT ON THE CRINOIDEA. 127

these plates. It appears to me verj'- probable that this may also be the case iu
Thaumatocrinus (PL LVI. figs. 1-4), i.e., that the primary cords pass right up out of the
basals into the interradials and then divide, so that the secondary corc^s would enter the
sides of the radials as in Bathycrinus, instead oftheir inner ends as in Pentacrinus and
Comatula.

G. The Sacctjli, and the Colouring Matters.

The nature and functions of the sacculi are as much a puzzle to me now as they were
when I first began to study the Criuoids in 1875 ; and I have nothing to add to the
observations of Wyville Thomson and Perrier on their appearance in the li\ing animal,
both larval and adult. Colourless during life, they become strongly tinged after death
by the pigment set free from the perisome. Their occurrence in the wall of the digestive
tube in Antedon rosacea was first noticed by Ludwig ; and I have found them in the
very lowest part of the cup of a larva with five cirrus-stumps, just above the chambered
organ. But this is the only species known to me which presents this peculiarity. In all
other types in which the sacculi occur at all, they are invariably limited to the immediate
neighbourhood of the water-vessels. Abundant in most species of Antedon, they
never occur in Actinometra, and I suspect that Ludwig's reference to their presence in
this genus is due to an oversight.^ At any rate I have not been able to find them in
Actinometra trachygaster and Actinometra bennetti, the two species w^hich had come
under his observation.

They vary considerably in distribution among the other genera of Comatula. I
have not succeeded in finding them in Thaumatocrinus, while they are but scantily
developed in the three species of Atelecrinus. Eudiocrinus indivisus and Eudiocrinus
atlanticus have them in al)undance ; while there are few in Eudiocrinus varians and none
in Eudiocrinus semperi or in Eudiocrinus japonicus, so far as I have been able to make
out. Promachocrinus kerguelensis has them on the pinnules, but they are very scanty or
absent elsewhere. Neither have I found any in a small series of sections through a
Holopus-arm ; and though structures of the same nature occur sparingly in Pentacrinus,
Rhizocrinus, and Bathycrinus, they are but poorly developed and irregular in their
occurrence.

In some species of Actinometra individual vesicular bodies resembliDg the elements of
the sacculi are scattered through the ventral perisome ; but there is no regular arrange-
ment of them into groups at the sides of the ambulacra as in the endocyclic Crinoids.
When the ambulacra are plated, as in many tropical Antedons, the sacculi are lodged
between the successive side plates, the front edges of which are notched for their reception
(PL LIY. figs. 4, 6-9) ; while they occupy little pits in the large plates which cover the

1 Zeitschr. f. xviss. Zool, BJ. xxix., 1877, p. 59.

123 THE VOYAGE OF H.M.S. CHALLENGER.

ovarian pinnules of Antedon acoela and Antedon angustiealyx, and also in the substance
of tlie small later joints of the pinnules (PI. LIV. fig. 5).

As regards the colouring matters in the body of a Crinoid I can do uo more than
refer to the important observations of Prof. Moseley, who employed the spectroscope on
five different occasions when Pentacrinidse were dredged in the Pacific (Stations 170, 192,
209, 210, 214). Metacrinus was obtained at all these Stations, but Pentacrinus only at
the first and the last two.^

The colouring matter yielded by most of the specimens is distinguished by very well-
defined absorption spectra, and has been termed " Pentacrinin " by Prof. Moseley. It is
freely soluble when fresh in slightly acidified alcohol, and gives a solution which is of an
intense pink when viewed by transmitted light, but acquires a bluish-green colour when
rendered alkaline by the addition of ammonia. " The fresh colouring matter is soluble in
fresh water, but remains partly suspended, forming a slightly opaque dark purple solution,
which gives, when quite fresh, a mixed acid and alkahne spectrum." All the species,
both of Metacrinus and Pentacrinus, which were obtained off the Kermadec Islands at
Station 170 were of a uniform dusky colour when fresh, being evidently coloured by acid
pentacrinin. The three species of Metacrinus dredged off the Ki Islands at Station 192
" when in the fresh condition, had their stems almost white, and their crowns of a light
yellow or fight reddish -orange, showing no purple coloration at all ; and those dredged
offthePanglao and Signij or Islands (Station 210) were almost colourless; nevertheless,
when placed in alcohol, they yielded a solution which was deeply coloured of a sap green,
and which when acidified, became of the usual deep pink of pentacrinin. The pentacrinin
was thus in these examples, though present in great adundance, entirely masked."

I am unfortunately unable to name with certainty the species which was obtained at
Station 210, no specimens having reached me with the label of this Station; though,
according to Sir Wyville Thomson's MS. notes, four specimens of two species of Penta-
crinid^ were obtained here. One of these was perhaps the single individual of Metacrinus
murrayi, which reached me mthout any record of its locality ; while the other must
have been either Pentaci^inus alternicirrus or Pentacrinus naresianus. Three out of
seven individuals of the latter species came into my hands without any indication of
locality; whUe a few of the fourteen specimens of Pentacrinus alternicirrus were in the
same condition. Both were obtained at Station 214, off the Meangis Islands.

There is, however, one difiiculty respecting the occurrence of Pentacrinus naresianus
at Station 210. For whde the individuals of this species dredged at Stations 170 and
214 had the dusky purple colour of acid pentacrinin, those supposed to have been
obtained at Station 210 were almost colourless, and the pentacrinin, though present,
entirely masked.

1 On tlie Colouring Matters of various Animals, and especially of Deep-sea Forms diedged by H.M.S. Challenger,
Quart. Journ. Micr. Set. 1877, vol. xvii., N. S., pp. 5-10.

REPORT ON THE CRINOIDEA. 129

The other species, besides these two which were dredged at Station 214, off tlie
Meangis Islands, were Metacrinus costatus, Metacrinus moseleyi, Metacrinus varians, and
Metacrinus ivyvillii. As this last also occurred at Station 170, off the Kermadec Islands,
it was probably one of those which were coloured dark purple, owing to the presence of
acid pentacrinin. Prof. Moseley can give me no clue, however, as to the anomalous
form which did not contain pentacrinin at all, but an entirely different colouring matter.
It was of a light pink colour when fresh, and when treated with absolute alcohol gave a
simple, light red solution, with an absorption spectrum totally different from that of
pentacrinin.

All Prof. Moseley 's spectroscopic observations of the colouring matter of the
Pentacrinidse were made in the Pacific ; and it therefore became a matter of some import-
ance to determine the presence of pentacrinin in the Atlantic species of the family. For
this purpose Prof. Moseley has been kind enough to examine some of the deeply
coloured spirit in which there had been preserved several individuals of Pentacriniis
ivyville-ilwinsoni that were dredged by the " Porcupine " in 1870, off the coast of Portugal.
As might have been expected, the colouring matter proved to be pentacrinin.

A still more interesting point is the presence of this substance in Holopus. Immersion
of a dry, greenish-black Hohpus in alcohol for a few. minutes yielded a diehroic solution,
which gave the characteristic spectrum of pentacrinin. It is more than probable, there-
fore, that this substance will be found to be present in the four Caribbean species of
Pentacrinus.

Most species of the Comatulse " appear to be either usually of a rose colour or of an
orange or yellow, running into a yellow-brown or of a dark purple. Both the rose or red
and yellow colouring matters are freely soluble in alcohol, and usually in fresh water."
The colouring matter of most of them, like that of Antcdon rosacea, has no characteristic
absorption spectrum showing bands. But a dark purple species common at Cape York
yielded a special colouring matter which Prof Moseley has termed "Antedonin." It is
freely soluble in weak spirit, and gives an intense fuchsin -coloured solution, which yields
spectrum consisting of three well defined absorption bands.

I have little doubt that the species in question was Actinometra strata. Two species
of Antedon which were also obtained at Cape York, but in less abundance, seem to have
been tinged with the same colouring matter, as their appearance is the same as that of
the Actinometra. The most remarkable point about this colouring matter is its presence
in a slightly modified form in various deep-sea Holothuriaus.

(ZOOL. CUALL. EXP. — PART XXXII. — 1884.) li 17

130 THE VOYAGE OF H.M.S. CHALLENGER.

VII.— ON THE HABITS OF EECENT CRINOIDS, AND THEIR PARASITES.

The Crinoids, like the other Echinoderms, are very gregarious in their habits. This
is especially the case with those living near the shore and in depths down to 150 fathoms.
In Torbay, at Roscoff, in the Bay of Naples, and in many other places large quantities
of Antedon rosacea may be readily obtained. Ten thousand individuals of Antedon den-
tata are estimated by the officials of the U. S. Fish Commission to have come up at one
haul of the dredge off the coast of New England.^ The " Porcupine " dredged Antedon
phcdangium in abundance in 50 to 100 fathoms off the Tunis coast ; while the Challenger
found Antedon carinata to be very plentiful off the coast of Bahia. Numerous examples
of Actinometra jukesi and Actinometra strata were dredged at Cape York, and large
quantities of both genera have been obtained in the shallow water round the Phili^ipines
by Semper, Meyer, and the Challenger. Station 192, near the Ki Islands (140 fathoms),
and Station 214, off the Meangis Islands (500 fathoms), proved to be very prohfic, both
in Comatulse (12 and 7 species), and in Pentacrinidae (4 and 6 species) ; and the
dredgings of the " Blake " in the Caribbean Sea have yielded similar results. A few
stations where Comatute were especially abundant (off St. Vincent, Grenada, and
Montserrat, all 120 fathoms or less) also yielded several examples of Pentacjnnus. Prof.
Agassiz records that on one occasion no less than one hundred and twenty-four indivi-
duals were obtained at a single haul of the bar and tangles.^

On the other hand, the score of individuals of Pentacrinus ivyville-tliomsoni dredged
by the "Porcupine" in 1870 in 1095 fathoms off the coast of Portugal were unaccom-
panied by Comatulas. This is perhaps due to the depth being one at which these Crinoids
are comparatively rare ; as two species were obtained, together with fragments of
Pentacrinus, within a few miles of this Station, where the depth was 350 fathoms less.
Some thirty individuals were taken by the "Talisman" in 1883 at a depth of 1480 metres
(800 fathoms); but no information respecting the presence or absence of Comatulceat the
same Station has yet been published.

The Bourgueticrinida3, like the Pentacrinidte, are generally found living in large
numbers together. Both Sars and Verrill have obtained numbers of specimens of
Rhizocrimis at a single haul of the dredge, and Agassiz records that " on one occasion
off Sand Key the dredge must have passed through a forest of them, judging at least
from the number of stems and heads of all sizes which it contained."^ Bathycrinus
aldrichianus was found by the Challenger to be equally plentiful in the Southern Ocean
(Stations 146, 147); and it is evident that in certain localities Crinoids of the more
common genera may be obtained in almost any quantity.

1 Amer. Journ. ^ci. and Arts, vol. xxiii. p. 136. - Bull. Mus. Comp. Zool, 1879, vol. v. p. 296.

3 Bidl. Mus. Comp. Zool, 1879, No. 1, vol. v. p. 5.

REPOET ON THE CRINOIDEA. 131

It is well known that the Crinoids were equally plentiful in several former geological
periods. In the British area, for example, there are the remains of enormous forests of
Crinoids both iu the Silurian and in the Carboniferous rocks. The marvellous abundance
of these animals in beds of the same age in America is well known. In a less degree also
the Silurian of Sweden, the Devonian of the Eifel, and the Carboniferous of Belgium and
Russia were characterised by a great development of Crinoid life. This terminated,
however, with the close of the Paleozoic epoch ; but in the Lower and Middle Lias, both
of Britain and of the Continent, there were enormous colonies of Extracrinus, slabs of
which are so well known in every museum. Although the limestone bands which are
made up of the fragments of the skeleton of Extracrinus are by no means so thick as
the Palaeozoic Crinoidal limestones, yet the association in one place of a large number of
individuals must have been, for the time at least, as considerable as in the case of the
Pateocrinoids. A similar band, 10 to 20 centimetres thick, which was discovered by
M. Eudes-Deslongschamps in the Great Oolite at SoUers near Caen,^ is evidence of a
singularly localised colony or "Station" of Pentacrinus (Extracrinus?). For no trace
of a similar bed occurs in other sections of the Great Oolite in the neighbourhood.
Another horizon at the top of the Great Oolite, near Sennecey-le-Grand," is marked by
the very great abundance of a species of Extracrinus which is also found in corresponding
beds elsewhere (in the Department de la Meurthe) ; while the Forest Marble of Gloucester-
shire contains numerous remains of Pentacriuidos which occur associated in slabs much
like those formed by Extracrinus hriareus, though somewhat less extensive.

Although the Middle and Upper Jurassic rocks of this country and of the Continent
have been found to contain numerous species oi Pentacrinus, I do not know that any large
forests of them have been met with, like those of the Lias, Great Oolite, and the Recent
Seas ; and the same may be said of the Cretaceous and Tertiary beds.

As regards the Apiocrinidse, the abundance of Apiocrinus parhinsoni in the Bradford
clay is well known, and the Sequanien (Coral Rag) of the Continent is exceedingly rich
in Millericrinus. The same is true of Eugeniacrinus in the White Jura of Wurtemberg,
though it does not occur in Britain at aU. The coral bed at Nattheim is famous for the
number of Coina^wZa-remains which it contains ; though as these, like the Eiigeniacrmus-
calyces, are all more or less rolled and fragmentary, we do not meet with evidence of
gregarious habits such as is represented by either of the colonies of Lyme Regis, SoUers,
or Sennecey-le-Grand.

The different modes of attachment which occur among the Crinoids have been
discussed in Chapter II. In all the Bourgueticrinidas there is a spreading root of variable
extent, the subdivisions of which attach themselves by calcareous expansions to foreign
bodies. Holopus is a permanently fixed type like the Bourgueticrinidae. But the

1 Etudes sur les etages jurassiques inferieurs de la Nonnandie, Paris, 1864, pp. 229, 235.

2 See de Loriol, Notice sur le Pentacrinus de Sennecey-le-Grand, op. cit., pp. 11-13.

132 THE VOYAGE OF H.M.S. CHALLENGER.

Pentacrinidse are very variable in tliis respect, some individuals remaining fixed through-
out life ; while others become detached and are henceforward more or less free like
the Comatulse, attaching themselves temporarily by means of their cirri.

Some Comatulse, however, lose their cirri altogether when mature. The older ones
gradually drop off without being replaced, while their sockets become obliterated until
nothing remains of the centro-dorsal but a flat plate in the centre of the radial pentagon.
This is the case with the Actinometra jtikesi, Actinometra stellata, and the fossil Actino-
metra loveni. Some specimens of Millericrinus pratti reach the same condition, nothing
remaining of the stem except a pentagonal disk in the centre of the ring of basals ; so that
in the absence of other stalked individuals these would naturally be taken for Comatulse.
This suggests the question whether the single plate in the centre of the calyx of Mar-
siipites and Uintacrinus may not be really a separated top stem-joint, and not a dorso-
central plate homologous ■wdth that of Urchins and Stellerids as is generally supposed.

It is noteworthy that certain Blastoids, e.g., Eleutherocrinus and Astrocrinus, were
stemless and free just like Marsupites ; and it is possible that the same was the case with
some species of the Palaeozoic genus Agassizocrinus, at any rate in later life. Other
Pala30crinoids, together with some Blastoids and Cystids, must have been almost equally
free, as has been already explained in Chapter II. For though a stem was present, it
was often quite short, and almost or entirely devoid of cirri ; though it was sometimes
fixed by coiling itself round other Crinoid stems and similar bodies.

The apparently perfect freedom of many of these forms is very singular and difficult to
understand. Much would be learnt about them, no doubt, if the habits of a living Acti-
nometra jukesi were carefully watched, for it is well established that Comatulse which
have once anchored themselves by their cirri remain so fixed for a considerable time,
except perhaps at the period of sexual activity; and it would therefore be very interesting
to know how far the cirrus-less forms remain permanently fixed. Seaweeds, Polyzoan
colonies. Corals, and Zoophytes often serve as the anchorage of Comatulse. Thus in one
case that I have met with the cirri were coiled round a DendrophyUia, and in another
round the stem and branches of a Gorgonia; while it sometimes happens that the cirri
of a Comatula are fastened round the still larger cirri of a Metacrinus or Pentacrimis.

The food of a Crinoid is considerably varied in its nature according to the character
of the sea-bottom on which it lives. The horny casings of Entomostraca and the larvse
of larger Crustacea are frequently to be found in the digestive tube, together with the
frustules of Diatoms, spores of AlgSB, &c. Dr. Carpenter mentions Peridinium (Ceratium)
tripos, Ehr., as a principal article of food of the Arran Comatulse ; while in sections of
Bathycrinus, Rhizocrinvs, and Pentacrinus from deeper water the sihceous shells of
Kadiolarians may be found in considerable abundance and variety. Foraminifera too
form a staple article of food for these deep-sea species. I have frequently found Glohi-
gerina, Biloculina, and other types beneath the covering plates of the food-grooves on

EEPORT ON THE CRINOIDEA. 133

the arms and pinnules ; while remains of their sarcodc bodies occur in the intestine of
decalcified specimens. Dr. Carpenter mentions the frequent presence in the alimentary
canal of Antedon rosacea, so as almost completely to choke it, of the body of a suctorial
Crustacean with its egg masses ; ^ and he supposes " that it has been introduced either as
an egg or as a larva, and has undergone its development parasiticaUy where it is found."
The same explanation will probably account for the frequent presence in the anal tube
of Actinometra jukesi and Actinometra strata from Cape York of an Isopod [Anilocra)
nearly half an inch long (PI. LV. fig. 1). Either as an egg or as a larva it must have
been caught in one of the ciliary currents converging on the mouth from the arms, and
have then been carried through the digestive tube to the rectum where it remained.

A third form of parasitic Crustacean is one which I have found encysted in the
ventral perisome of the disk of some individuals of A ntedon escJirichti which have been cut
into sections ; but though one or two accomplished zoologists have examined its remains,
I have not been able to learn anjrthing about its affinities. Another equally obscure
internal parasite of the Crinoids is a peculiar worm which I first found in some sections of
Actinometra jta^rvicirra that were cut some years ago in the zoological laboratory of the
University of Wiirzburg. The Crinoid had been obtained in the Philippine Islands by
Prof. Semper, and I found three individuals acting as hosts to this singular and entirely
unknown creature, which I have not met with in any other Comatulse from the same
locality. It was first noticed in the coeliac canal of the arms, which it often almost filled,
so as to suggest the idea that the egg had been introduced into the body-cavity and had
developed in that part of it (PI. LXL fig. 4). I subsequently found it in the visceral
mass of two other individuals, occupying some of the meshes in the connective tissue net-
work which fills up the intervisceral coelom.

The external parasites of the Crinoids ai-e many and various ; though it may be
doubted whether some of them can be considered as real parasites, i.e., as living at the
expense of the Crinoid. Besides the well known Mijzostovia, of which I will speak later,
Willemoes Suhm found four other parasites on one Comatula, all resembling it in
coloration.^ " Es waren das erstens auf dem Kelch sitzende Ophiuriden, zweitens kleinere
Aphroditaceen, drittens Amphipoden,^ die sich in den Magensack eingebohrt hatten und
viertens ein Alpheus. Mit Myzostomum also fiinf Parasiten auf dieser allerdings sehr
grossen Comatida!"

I have frequently found Ophiurids entangled in the cirri, which is probably merely
accidental ; while small bivalves, Sertularian Hydroids, Polyzoa, tube-worms, and
corals (PL LI. fig. 8) may be attached to the stem, not for any special nutritive
purposes, but simply because the larvae had to find a resting-place somewhere.* Various

1 Phil. Trans., 1866, p. 701. ^ Zeitschr. f. wiss. Zool., 1876, Bd. xxvi. p. Ixxix.

3 Probably the same as the Isopod above mentioned.

* The same may be said of an Ophiurid larva, which was attached by its long Pluteus-arms to the soUtaiy stem-
fragment of Metacrinus tuberosus from near the Ki Islands (Station 192).

134 THE VOYAGE OF H.M.S. CHALLENGER.

species of Cirripedes are also frequently met with ou the stems aud cirri of Pentacriuidse
(PI. LII. fig. l), both in the Caribbean Sea and in the East Indian Archipelago. Thus,
for example, Scalpellum album is common on the stem and ciiTi of many individuals of
3Ietacrinus dredged off the Ki Islands (Station 192), while Scalpellum halanoides and also
Veri^uca nitida occur attached both to Pcntacrinus and Metacrinus hora. off the Meangis
Islands (Station 214). At this last Station too, an obscure larval Cirripede occurred,
attached to a cirrus of Metacnnus varians by its ventral margin, while a minute Avicula
was anchored by a few threads to a cirrus of Metacrinus interruptus at Station 209.
Yon. Graff has described a small Stylina} as parasitic on the anal tube or pinnules oiAntedon
rosacea, and found that holes remained after its removal. Rhizocrbms lofotensis in like
manner is often infested with two or three small shells of Stylifer which bore comparatively
large holes in its calyx. Pourtales has desci-ibed some examples of this species from the
Florida Channel as having the calyx and part of the stem coated with an encrusting
Hydroid polype ; and some small Ehizopods are shown in PL X. fig. 16 on the stem
of a Rhizocrinus rawsoni from the Azores. I have found Truncatidina lobatula to
be abundant on the cirri of the Comatulaj which were obtained by the Dutch Arctic
Expedition in the Barents and Kara Seas ; while Poli/trema miniaceum is common on
the stem and cirri of the Pentacrinidse dredged at Stations 192 and 214.

The especial parasite of the Crinoids, however, is the well-known Myzostoma," which
is sometimes found infesting them in great numbers. I have myself removed five
moderately large specimens from a single individual of a Bathycrinus aldrichianus, and
as this had been some years 'in spirit before reaching me, it may very likely have served
as host to a still larger 'number; while in the Mediterranean twenty-seven have been found
infesting a single Comatula. They attach themselves to the stem, disk, and arms, either
on the ventral or on the dorsal surface. I am not aware that they have ever been met
with on Rhizocrinus, which often supports boring Stylifers ; but v. Willemoes Suhm found
them on Hyocrinus, Bathycrinus, and on many Comatulse, though he did not meet -nith
free Myzostomida on any of the Challenger Pentacrinidse. A closer examination,
however, has revealed their presence in a few cases. The abnormal specimen of
Metacrinus angulatus from the Ki Islands (Station 192), which is figured on PL XXXIX.
fig. 2, had a Myzostoma ivyville-thomsoni resting between the ordinary anal tube and
the second smaller one at its side. In many cases the Myzostoma, instead of living in the
free state, causes an abnormal growth of the calcareous tissue of the arm so as to form a cyst
in w^hich two or three individuals live. At two Stations in the South, and at another in
the North Pacific, these cysts proved to be tolerably common. At the first named
(Nos. 170 and 174) the cysts were limited to the arms of four Comatulse ; while at

1 Stylina comatulicola, ein neuer Schmarotzer der Comatula mediterranea, Zeiischr.f. iciss. Zool, 1875, Bd. sxv.,
Suppl., pp. 124-126.

* For a fuU account of the Myzostomida, see Prof. L. v. Graff, Das Genus Myzostomum, and Zool. Cball. Exp.,
jjart xxvii., 1884.

EEPORT ON THE CRINOIDEA. 135

the last (Station 214, off Meangis Islands) the principal hosts were Antedon
angusticalyx and Pentacrinus alter nicirr us, several examples of the latter being
remarkable for the number of cysts on their arms. Two views of one of these cysts are
shown in PI. XXVII. figs. 7, 8, in the former of which the edge of the parasite is just
visible at the opening of the cj^st. This cyst is principally formed round a jiinnule; but
the same kind of thing is sometimes formed in the substance of the arm itself, as shown
in PI. XXVII. figs. 9 aud 10, the inhabitant of the cyst being again visible through its
mouth.

Thus then, while Stylifer bores holes into the calyx of a Crinoid, Myzostoma produces
swellings and inequalities of growth in the arms and pinnules. I have never met with
any distortion of the stem which could be considered as resulting from the action of a
parasite ; and it is therefore curious that abnormal growths in the stems of fossil Crinoids
should have attracted the attention of so many palaeontologists. Eofe showed, for
example, that one cause of the enlargement of the stems of Carboniferous Crinoids arose
from the attachment of a parasitic coral, and the subsequent endeavours of the Crinoid to
envelop the latter by an undue secretion of calcareous matter.' It has also been shown
by Mr. E. Etheridge, jun.,'^ that a similar distortion may be due to the adherence of
certain Brachiopods (Productus or Chonetes). These grew less quickly than their hosts,
and so became gradually surrounded and enveloped by the calcareous deposit secreted by
the latter. The attachment of Polyzoa, again, may also give rise to enlargement, and
even the accidental approximation of the stems of two individuals seems to have
sometimes resulted in a complete but irregular union between them. Enlargement and
irregularities of growi;h seem to be very common in the stems of Ainocrinus and
Millericrinus, though not in Pentacrinus ; and they have often been regarded as the
results of injury. But their exact nature and causes have not yet been determined as
satisfactorily as in the case of the Palteocrinoids.

There are, however, some stems of Millericrinus figured by de Loriol,^ from the
Jurassic rocks of France and Switzerland, which present characters of the same nature as
those shown on the arms and pinnules oi Pentacrinus alternicirrus (PL XXVII. figs. 7-10),
i.e., cystiform enlargements, each with an external opening. Both Prof. L. von Graff
and myself are inclined to regard these as due to the action of Myzostomida or of similar
parasites. But it is singular that they should be developed on the stem ; for I have
never found a Myzostoma-cyst on the stem of any recent Crinoid, though at some
Stations (170, 214) they were abundant upon the arms, both of Comatulse and of
Pentacrinidse.

1 Note on the Cause and Nature of the Enlarf^ement of some Crinoidal Columns, Geol. Mag., vol. vi. p. 351.

2 Observations on the Swollen Condition of Carboniferous Crinoid Stems, Proc. Nat. Hist. Soc, Glasgow, vol. iv.,
1879, pp. 19-36, pis. i., ii.

3 Swiss Crinoids, pi. xi. figs. 18, 36-38. French Jurassic Crinoids, pi. 65, figs. 4-6, 8 ; pi. 80, figs. 2, 2a, 13 ; pi. 99,
figs. 5a, 55, 5d.

136

THE VOYAGE OF H.M.S. CHALLENGER.

Till. —THE GEOGRAPHICAL AND BATHYMETRICAL DISTRI-
BUTION OF THE CPJNOIDS.

A. Geographical Eange.

The geographical distribution of the Crinoids is fully as extensive as that of the other
Echinoderms. Comatulae range between the two parallels of 81° N. and 52° S., while
the corresponding limits for the stalked Crinoids are 68° N. and 46° S. Some genera, -
such as Actinometra and Pentacrinus, occur in shallow water everywhere, or almost
everywhere, within the tropics ; though Pentacrinus has not yet been obtained in the
Pacific east of long. 170° W., nor between the meridians of 5° W. and 120° E., its
nearest approach to the Indian Ocean being the Moluccas. Metacrinus, on the other
hand, has been found in the neighbourhood of Singaj)ore ; but is otherwise almost
entirely limited to the tropical regions of the West Pacific, ranging, however, as far
north as Japan ; while Pentacrinus ivyville-thomsoni occurs in the Atlantic as far north
as lat. 45°. HoJopus, however, bas never been obtained out of the Caribbean Sea.

Of the stalked Crinoids Rhizocrhms has the farthest northern range (68° N.), but it
has not been met with more than once (Station 122), or possibly twice (Station 323),
south of the equator, and is limited to the Atlantic and the Caribbean Ocean. Bathycrinus,
while rangino- throuoh 1 10° of latitude in the Atlantic and Southern Oceans, does not occur
in the Pacific at all. Antedon, on the other hand, is cosmopolitan, five species inhabiting
the Arctic Ocean, while the genus is also well represented in the Southern Ocean. It is
curious, however, that so far as I am aware, there are no museum specimens, nor any
records of Comatulae from New Zealand.

As a general rule, the individual species of Crinoids are much limited in their range,
though there are a few well marked exceptional cases. Both species of Ehizocrinus occur
in the east as well as in the west Atlantic, Ehizocrinus lofotensis ranging 'from the
Lofoten Islands to 9° S. or perhaps to 35° S.; while three species of Pentacrinidaj occur
in the South Pacific near the Kermadecs, and also among the Philippines, as the following
list shows : —

Species.

Near the Ker-
madec Islands.

Near the
Fiji Islands.

Panglao.

Meangis Islands.

Pentacrinus naresianun,
Pentacrimis alter nieirrus, .
Metacrinui loyvillii.

170
170

171
171

175

210?
210?

214

214
214

EEPORT ON THE CRINOIDEA.

137

The four Caribbean species of Pentacrinus are widely distributed among the West India
Iskmds ; and it is possible (or rather probable) that systematic dredgings among the
islands of the Pacific, like those of the " Blake " in the Caribbean Sea, would largely
increase the specific I'auge both of Pentacrinus and of Metacrinus.

Among the Comatulse the well known Antedon escJmchti is universally distributed
in the Arctic Ocean, and on both sides of the Atlantic as far south as lat. 43° N.; while
Antedon dentata of the Barents Sea, the north-east Atlantic, and the Scandinavian
coasts is abundant on the banks ofi" New England. Antedon carinata is another widely
distributed species, occurring both on the Atlantic and on the Pacific coasts of South
America, at Java, Mauritius, the Seychelles, and elsewhere; while Actinometra pulchella of
the Caribbean Sea has also been dredged in moderately deep water off the coast of Morocco.

The greater number of the Comatulse, however, like the four species of Bathycrinus

Table II. — Showing the Geograjjhical and Bathymetrical Range of all the

Genera of Recent Crinoids.

Genus.

Depth.

Tempera-
ture.

Latitude.

Oceans.

Remarks.

Pentacrinus, .

Fathoms.
42-1350

°C.
1-8-20

46° N.-30° S.

A.Cl'.B.'

Metacrinus,

68- 630

4-21 -7

35° Jf.-30° S.

P.E.

Shizocrinus, .

73-1900

0-21 -5

68° 15' ^.-35° 39' S.

A.C.

A .«tem-{ragnient only at St.
323, 1900 fath. Lat."35° 39' S.
(/rfeC.W.T.)

Bathycrinus, .

1050-2435

O-S-2-5

65° 55' N.-46° 46' S.

A.S.

Uyocriims,

1600-2325

0-8-1 -8

5° 31' N.-46° 16' S.

A.S.E. ?

A 3'oung specimen, perhaps of
Byocrinus, from St. 223, in
N.W. Pacific, 2325 fath. [fide
C.W.T.).

ffolajms,

About 100

■ 10°-20°X.

C.

...

Antedon,

2-2900

0-2-25

81° 41' N.-52° 5' S.

A.C.S.LP.E.

Actinometra, .

2- 533

8-5-25

35° ]Sr.-35° S.

A.C.LP.E.

Temperature at greatest depth,
533 fath., not known.

Eudiocrinus, .

30-1050

2-4-4-2

45° N".-37° 34' S.

A. P.E.

Highest temperature iu sliallow
water among the Philippines
not known ^Eudiocrinus indi-
• visus).

Promachocrinus,

28-1800

0-3-5-3

4° 33' N.-50° 1' S.

S.P.

One specimen at St. 214, in N.W.
Pacific.

Atelecrinus, .

291- 610

3-7-8-5

24°8'N.-19°10'S.

A.C.P.

Lowest limit (St. 174) uncertain.
Possibly not below 450 fath.

Thaumatocrinus,

1800

0-3

50° -1 S.

S.

1 A. Atlantic. C. Caribbean Sea. S. Southei-n Ocean. I. Indian Ocean. P. Pacific. E. The more or less enclosed
seas of the East Indian Archipelago, bet-^'een Cape York and Singapore. Within this limit are comprised Stations
201 to 212, in the neighbourhood of the Pliilippine group, but not Station 214, off the Meangis Islands.

(ZOOL. CHALL. EXP. PART. XXXU. — 1S6L) li IS

138

THE VOYAGE OF H.M.S. CHALLENGER.

;iad most of the Pcntacrinidse, have as yet been obtained at one or two localities only,
and those not very widely separated.

T.ible IT. on the preceding page gives the bathymetrical and geographical ranges, so far
as yet known, of all the genera of recent Crinoids ; their distribution in the various oceans
beii3g indicated by the capital letters in the fourth column. A few cases of uncertainty,
owing to deficiency of information and other causes, are noted under t!:e head of remarks.

B. Bathymetrical Eange.

Tlie frequency of occurrence of all the genera of recent Crinoids at depths below 250
fathoms, so far as has yet been made known, is shown in Table III. One or two curious

Table III. — Showing the Frequency of Occurrence of all the Genera of Recent Crinoids

at dejyths heloiv 250 fathoms.

Genus.

PcTitacrinus, .

Metacrinus,
Ehizoeriims, .

' ithycrinus, .
flyocrimis,

Holopus,

Antedon,
Adinomeira, .

Etuliocrinus, .

Promachocrinus,

Atelecrinus,

Tlmumatocrinus,

200-500 f. 500-7001

1

10

56
5

1?

700-1200 f. 1200-2000 f.

2000-2500 f.

1

1?

2500-2900 f.

Remarks.

Only four of the seven dreilgings
noted in the first column were at
a gieater depth than 250 fath.

A stem-fragment only at St. 323;
1900 fath. Lat. 35° 39' S. {fido
C.W.T.).

Possibly a young specimen from St.
223, in N.W. Pacific, 2325 fath.
ijidc C.W.T.).

{ Ai tlireo of the Stations recorded in
\i!c first column (for each genus)
the depth was between 200 and
230 fath. Stations 174b, c, d
{•2iZ. SIO ind 210 fath.) are not
i rs;- r either genus,

I 0. ■■■',! information.

Lowest limit (station 174C; 610
lath.) uncertain. Possibly not
below 450 fath.

contrasts are noteworthy. While Bathycrimis and Hyociinus are essentially abyssal
forms, Actinometra is not known with certainty to occur below 533 fathoms, and has
not been obtained more than seven times below 200 fathoms.^ Pentacrinus, having a

1 Comatulse were dredged at Stations 174b, 174c, 174d, the deirths being resi^eutiyely 235, 610, and 210 fathoms ;
but there is no record of the particular haul which yielded the Comatula'.

REPORT ON THE CRINOIEEA.

130

much wider geographical range than Metacrinus, also extends to a great -r depth ; while
the same contrast, including also thermal range, may be noticed between Antedon and
the stalked Crinoids generally. It is somewhat singular that this widely distributed
genus, which was dredged at 2900 fathoms in the Pacific and at 2600 fathoms in the
Southern Sea, should not have been met with between the latter dojith and IGOO
fathoms.

O. The Association of the Genera at Particular Stations.

The mode of grouping of the various genera of Crinoids, both stalked and free, whicli
have been obtained at different Stations and by the exploring expeditions of varii)u,s
countries, presents some points of interest. The Challenger dredged Comatulse at fortylive
Stations, and stalked Crinoids at sixteen. If the different North Atlantic expeditions of
the British and Norwegian Governments 1ie also reckoned, this last number rises to 28.

A single species of stalked Crinoid w;is found at several Stations, unaccompanied by
any Comatulae, as shown in the following Taljle.

Table IV. — Showing the numher of times that Isolated Species cf Stalked Crinoids

have been Dredged.

Rhkocrinus.

Batliijcrinus.

Hi/ocrimis.

Pentacriniis.

Metacrinus.

Cliallenger —

3

1

1
(jomg 1 fide C.W.T.)

1

North Atlantic Expe-
ditions—

1. British.

7

1

1

2. Norwegian.

4

...

At eleven Stations during the Challenger and "Porcupine" Expeditions which yielded
stalked Crinoids, there were either more than one species, or if only one was obtained it
was associated with one or more species of Comatulae (see Table V.). Pentacriniis has
been found associated with all the Comatulid genera except Thaumatocrinus, which was
only obtained once in the Southern Sea, considerably beyond the most southern limit
of Pentacrinus.

The " Blake," however, dredged Pentacrinus in shallower water than was worked by the
Challenger ; and found it associated with Actinometra at nine out of the fifty-four Stations
at which stalked Criiioids were obtained (see Table VI.). As in the case of the Challenger
and " Porcupine," the two species of Rhizocrinus were never found at the same Station.

140

THE VOYAGE OF H.M.S. CHALLENGER.

Table Y. — Showing the Association of Crinoid Genera at different Stations occupied hy
the "Porcupine," Cliallenger, and "Talisman," together with the number of Species
obtained at each Station.

Eemaeks.

Stations.

'A

1

a

Bathyerinus,

■i
g

s

.a

•J

i

g

s

32

§
s

Separate hauls at very different
depths.

Metacrinus muvrayi and Pentn-
crinus naresianvs referred
here conjecturally (see p. 1 28).

Separate hauls at slightly differ-
ent deptlis.

Hijocrinns, stem only {fide
C.W.T.).

No information ahout Comatulas.

" Porcupine,"
1870.

Challenger

Stations.

210

175

235

170

122

171

214
192
106

147
158

f "Talisman," ■

1882.
, (A.

1883. -

I (b.

1
1?

1

1
1
1

2
2

1

1?

2

4
4

...

1

1

1

...

1
1

1

1?

1

2
2

3

6

6
12

3

1

1

1

1

1

1

1

Table VI. shows the association of Pentacrinus, Holopus, Rhizocrinus, and the three
Comatuhd genera in the Caribbean Sea, together with the number of times that two or
more species of Pentacrinus were found at the same Station. Comatulse were obtained
some eighty times ; but as I have not yet been able to determine them critically, I have
found it impossible to give the number of species occurring at each Station^ as has been
done for the Challenger dredgings.

REPORT ON THE CRINOIDEA.

141

Table VI. — Showiiig the Association of Crinoid Genera in the Caribbean Sea, together
ivith the number of times that two or more Species of Pentacrinus were found at
the same Station.

Number of
Stations.

Pentacrinitn.

Rlihocriniis.

Holopus.

Antedon.

Aetinometm.

Atelecrinus.

13

1 species

12

X

1

^ jj

X

1

■*■ ))

X

X

X

3

^ 5J

X

3

^ )I

X

X

1

^ IJ

X

4

0

1

2

X

X

1

0

X

X

1

0

X

2

X

X

1

4 :,

X

X

X

2

X

1

X

X

4

X

X

3

X

X

X

1

X

X

1

X

X

X

Stalked Crin-
oids at 56
Stations.

Pentacrinus

at 33

Stations.

Rliizocrinus

at 26

Stations.

Holopus at

3 Stations

(two isolated).

Comatulae at 82 Stations altogether.

The above Table embraces — (l) all the dredgings of the U. S. Coast Survey Steamers
iu the Caribbean Sea and Gulf Stream; (2) those of Captain Cole of the Telegraph
Steamer " Investigator ;" (3) those carried on at Barbadoes by the collectors for
Sir Rawson Eawson ; (4) those of a few isolated collectors who have sent specimens to
the Museum of Comparative Zoology. The whole series will be found recorded in the
Station List at the end of the Report. No attempt has been made to classify any
observations anterior to those of Pourtales in 1867-

142 THE VOYAGE OF H.M.S. CHALLENGER.

IX.— ON THE RELATION BETWEEN THE RECENT AND THE
FOSSIL NEOCRINOIDS.

Onr knowledge of the Crinoids of the Secondary rocks is largely due to the labours of
d'Orbigny, Quenstedt, and de Loriol, the last of whom, having completed an elaborate
Monogi-aph on the Fossil Crinoids of Switzerland, is now publishing a still more extensive
one devoted to those occurring in the Jurassic rocks of France.

He groups the Neocrinoids into ten families —

1. Marsupitidse.

2. Uintacriuidje.

3. Encrinidie.

4. Eugeniacrinidse.

5. Plicatocrinid^.

G. Apiocrinid^.

7. Bourgueticrinidie.

8. Holopodidge.

9. Pentacrinidae.
10. Comatvdidse.

The first two of these, each based on a single genus, are placed provisionally among
the Neocrinoids by de Loriol, who has transferred them from the Palseocrinoids ( = TesseUata)
vrith which they were ranked by ZitteL In this step, and also in the establishment of the
new family Bourgueticrinid^ for the reception of Bovrgueticrinus, Rhizocrinus, and
allied genera, I entirely agree with de Loriol ; but I am not disposed to follow him and
Zittel in the association of Hyocrinus and PUcatocrinus into one family, and prefer to
consider the former genus as the type of a new family " Hyocrinidse." It has not yet
been discovered in the fossil state ; and of the ten families enumerated by de Loriol, the
first six in the above list died out at or before the close of the Secondary period ; while
all the four others have living representatives.

Little need be said about the extinct Neocrinoids, except that the association of the
Cretaceous Marsiqntes and Uintacrimis mth the Palseocrinoids, the so-called TesseUata,
appears to me to be based on a misconception ; and that Encrhms, as might be expected
from its stratigraphical position, finds its nearest allies in genera of the Carboniferous
and not of the Jurassic epoch, as will be pointed out later.

The characters of Flicatocrinvs are entirely difi"erent from those of any recent Crinoid,
though its calyx has a singular resemblance to those of young Pentacrinidae. The
Eugeniacrinidse are a less aberrant group ; but though the symmetrical forms of JEitgenia-
crinKS appear to have a considerable resemblance to Rhizocrinus and Bathycrimis in the
structure of the lower part of the cup and in the distribution of the canal system, jct
such types as Phyllocrinus and the distorted Torynocrinus arc altogether different from
any recent Crinoid.

The Apiocrinidie, represented by some doubtful species in the Lias, flourished exten-

REPORT ON THE CRINOIDEA. li?,

sively througli the Jurassic period, aad died out in the Lower Neocomian. Their phice
was taken by the Bourgueticrinidse, which were formerly associated with them into one
family. We have no certain evidence of the occurrence of the typical genus Bourgueti-
crmus in other than Cretaceous rocks ; though stem-joints which have been referred to
this genus occur both in Jurassic and in Eocene deposits. It is not unlikely, however,
that they belong to Thiolliericrinus or to Rliizocrinus respectively. The latter genus,
which is represented by two living species, is probably exclusively confined to the
Tertiary rocks, except perhaps for the so-called Boiirgueticrinus alabamensis from the
Cretaceous formation of Alabama, U.S.; and some of the stem-joints hitherto referred to
Rhizocrinus or to Bourgueticrinus may possibly belong to Bathyoinus, no calyx of
which has yet been found in the fossil state.

As regards the fossil Holopodid^, there are Micropocrinus of the Italian Miocene,
Cyathidium in the Chalk of Faxoe, the singular Gymnocrinus in the Oxfordien of
France and Switzerland, and lastly in the Middle Lias Cofylecrinus and Eudesicrinus ;
while Edriocrinus from the Upper Silurian and Devonian, a tyjje much resembling
Holopus in character, is a proof of the great antiquity of these sessile Crinoids.

The Pentacrinidse are remarkable for their long geological history. The type genus
Pentacrinus first appears in the Trias, together with the short-lived Encrmus. It per-
sisted through the whole of the Secondary and Tertiary periods, and is represented by
eight species at the present time. On the other hand, Extracrinus with its characteristic
stem, dic}clic base, and peculiar arm-divisions is confined to the Lias and Lower Oolites,
unless, as I am disposed to think, the Pentacrinus asteriscus which has been found in the
Western Territories of the United States associated with Alpine Trias fossils and Spirifera,
is also to be refexTcd to this genus. Balanocrinus, kuown as yet only by stem-joints,
ranges from the Middle Lias to the Lower Neocomian. The remaining genus Metacrinua
is confined to Oceania and the shallower parts of the Pacific, and is at present unknown
in the fossil state.

The general character of the fossil Pentacrinidse is essentially the same as that of
their recent representatives, except that they often had much longer stems which
sometimes reached as much as 50 or even 70 feet ; while the number of arms was
frequently limited to ten, which is not the case in any recent species but Pentacrinus
naresianus {Pis. XXVIII. -XXXa.). Some of them also appear to have been devoid of
external basals. But in two cases, at any rate, Pentacrinus dixoni and Pentacrinus
fishen, this is due to en-or ; for the basals have been overlooked and therefore described
as absent.^

Three genera of the Comatulidse, Antedon, Actinometra, and Eudiocrinus are known
in the fossil state. The last named is at present limited to the Lower Neocomian of

1 Oil the supposed Absence of Basals iu the Eugeniacrinidae and in certain tther Neocriuoids, Ann. and Mag. Nut.
Hist, 1883, ser. 5, voL xi. pp. 327-334.

144 THE VOYAGE OF H.M.S. CHALLENGER.

Switzerland ; thougli both the other two genera range back as far as the inferior Oolite,
Antedon being much the more common. But the distinctions between them, so far as
can be made out in the calyx alone, are much less sharp than in recent Comatulse. Many
of the Jurassic species combine in a singular degree various characters which are of
considerable value for the generic determination of recent Comatulae.

Besides their tendency to combine the characters of recent generic types, the
Jurassic Comatulse are remarkable for their large size, as are also the Cretaceous species.
The centro-dorsal may reach from 9 to 13 mm. in diameter, which is greater than that of
nearly every recent species except Antedon eschrichti ; while this type and Actinometra
robusta are almost the only living Comatulae with arm-bases anything like as massive as
those of the fossil species. Some of the Cretaceous forms must have been very large.
Thus the united centro-dorsal and radials of Antedon campichei {rom the Neocomian of
Switzerland may reach 15 mm. in height and over 20 mm. wide; while several
centro-dorsals of Antedon from the Upper Chalk are almost equally gigantic.

The Eocene fossils are of moderate size ; but the Miocene Antedon rhodanicus has a
centro-dorsal 13 mm. in diameter, while the three species described by Forbes from the
Coralline Crag of Sutton are all considerably smaller.

Of the remaining Comatulidse neither Promachocrinus nor Tliaumatocrmus has yet
been found fossil ; and though specimens have been described from the Maestricht Chalk
with a complete basal ring,^ I should hesitate at present to refer them to Atelecrinus.

1 Zeitschr. d. deutsch. yeol. Gesellsclo., Jahrg. 1878, p. 66.

REPORT ON THE CRINOIDEA. 145

X.— ON THE RELATIONS OF THE NEOCRINOIDS TO THE

PAL^OCRINOIDS.

The term " Neocrinoidea," which was first introduced by the writer in 1881/ embraces
all the " Crinoidea Articulata " of Miller and Miiller, together with Marsupites and Holopus,
both of which were placed elsewhere by the German zoologist ; while the Palaeozoic
Crinoids generally were referred to by him as the " Crinoidea Tessellata," ^ this group
includins the Semiarticulata and Inarticulata of Miller. His definitions of these two
great groups, however, were meagre in the extreme. The Articulata comprised the genera
Pentacrinus, Apiocrinus, Encrinus, and Comatula in the wide sense, i.e., types in which
the radii are free down to the base of the calyx ; while his only distinct reference to the
Tessellata is that they are Crinoids " deren Kelch ganz aus Tafeln zusammengesetzt ist." ^
From his numerous references to individual genera, however, it is possible to obtain a
tolerably clear notion of the ideas which led Miiller to establish these two principal
divisions of Crinoids ; and various palaeontologists have in consequence attempted, with
more or less success, to formulate characters which should distinguish them from one
another.

The most satisfactory of the earlier attempts in this direction was that which appeared
in Bronn's Thierreich (vol. ii. p. 228). Besides the supposed sutural union of the calyx-
plates and the presence of a subtegminal mouth in the Tessellata, reference is also made
to the asymmetry of the calyx in this group, the more frequent presence of a dicyclic
base, and the greater rarity of stemless forms than in the Articulata.

On the other hand, Liitkcn'' and others have pointed out the weakness of these
definitions.

By far the best of the numerous diagnoses which have been drawn up since the time
of Miiller are those which we owe to Zittel.^ But the freedom of the rays in the Articulata,
on which Miiller laid stress, is omitted Ijy him, while an important error runs through his
as through all the earlier definitions. The calyx plates of the Tessellata are " unbeweglich
durch einfache Nahte verbunden ; " while those of the Articulata are " meist sehr dick,
durch gelenkartig ausgehcihlte und gewolbte oder ebene Nahtflachen verbunden." Now
the lowest articulation to be found in the calyx of an articulate Crinoid is that which
unites the first and second radials (PL LXIL). The former are suturally united both to
one another and to the basals ; while, when underbasals are present (Marsujntes,
Extracrinus), the union between them and the basals is of the same kind. In both
groups the interradials (when present) are suturally united to the radials and to one
another ; so that the name-giving difference between the Articulata and the Tessellata is

1 Ann. and. Mag. Nat. Hist., 1881, ser. 5, vol. vii. p. 296. ^ Pentacrinus, he. cit., p. 30. ^ md., p. 29.

* Op. cit, pp. 219-222. ^ PaliBontologie, pp. 335, 342, 345.

(ZOOL. CHALL. EXP. — PART XXXII. 1884.) Ii 19

146 THE VOYAGE OF H.M.S. CHALLENGEE.

reduced to a supposed difference in the mode of union of the first radials with the joints
which they bear.

In all the Mesozoic and later Crinoids this is an articulation effected by a pair of
muscles and three ligaments, as described above (PL Vila. fig. 15 ; PL Vllb. figs. 1, 5 ;
PL Villa, fig. 7 ; PL X. figs. 1-4 ; PL XII. figs. 9, 12, 15, 20, 23 ; PL XX. figs. 7, 8 ;
PL XXI. figs. 5c, 6d ; PL LXIL). It presents itself even in forms which have such decided
Palaeozoic afiinities as Hyocrinus, Marsupites, and Thaumatocrinus, but it appears, accord-
ing to de Loriol,^ to be absent in Guettardicrinus, in which type the reduced articular
facet found in Ajyiocrinus seems to have disappeared altogether, not only from the distal
faces of the first radials, but also from those of the axdlaries.

In this respect, therefore, as also in the presence of numerous interradials, and in the
close lateral union of the lower parts of the rays, Guettardicrinus is a true Tessellate, as
has been already pointed out by Liitken.^

On the other hand, the first radials of many of the so-called Tessellates have just as
weU-marked articular facets as those of the Pentacrinidge, and the joints belonging to the
other orders of radials are united to one another in just the same way as the distichals
and palmars of Pentacrinida^ and Comatulse. Marsiqntcs, Allagecrinus, and Platycrinus
are notable examples of the presence of a muscular joint between the first and second
radials ; and I feel confident that it would be found to recur in numerous other forms
with perforated first radials, such as Cyathocrinus, Foteriocrinus, and Myrtillocrinus, if
properly looked for.

Thus then, so far as regards the presence or absence of a true articulation between
the first and second radials, no hard and fast line can be drawn between the older and
the younger Crinoids. Many of the former certainly had a muscular joint in this
position ; while it was absent in Guettardicrinus, though not from an arrest of
development, as in the older Crinoids.

The want of a distinct articular facet on the first radials of many Tessellata is only
another indication of their being in the condition of permanent larval forms, as is so
clearly shown by many of their other characters.^ The fossse for muscles and ligaments,
the central canal, and the transverse articular ridge of the radials and arm-joints of aU
the post-Palseozoic Crinoids (except Guettardicrinus) only make their appearance at a
comparatively late developmental stage. The axial cords of the young joints, whether of
rays, arms, or pinnules, Lie at first upon their upper surfaces, eventually being received
more deeply into the substance of the plates. They lie for some time in open canals,
which are only closed up and received into the substance of the skeleton at a later
period. Consequently the transverse section, which at first resembles a horse-shoe,
finally comes to be a ring, with muscles and ligaments attached round it. Now in many

1 Paleont. Frang., oj). cit., pp. 216, 218. ' Op. cit., p. 221.

^ See Wachsmuth, Amer. Journ. Sci. and Arts, vol. xiv. p. 190.

REPORT OJT THE CRINOIDEA. 147

Palasocrinoids the distal faces of tlie radials remain permanently in the horse-shoe
condition, and the ligaments and muscles must therefore have remained small and poorly
developed, just as they are in recent Crinoids until the central canal is completely closed
in. The gradual development of complete articular facets, commencing before the horse-
shoe stage, has been traced in the radials of the Palaeozoic Allagecrinus, just as in the
Comatulfe ■} and I see therefore no reason to doubt that many Palseocrinoids had an
imperfect articulation and not a suture between the horse-shoe facets of the first two
radials. This may perhaps be correlated with the small development of the arms of the
Palaeocrinoids, relatively to that of the calyx. The integrity of long arms with two
hundred or three hundred joints, like those of many Comatulae, would be much more
perfectly preserved if the bundles of muscles and ligaments were large and well developed
than if they remained small, as miist necessarily be the case on an imperfect terminal
facet of a semicircular or horse-shoe shape.

Believing then that in a very large number of the Palseocrinoids the second radials
were at least as movable on the first as in Ajyiocrinus, and in some cases a good deal
more so, I cannot regard the " difi"erentia " of the Palaeozoic and the later Crinoids on
which MllUer and his followers laid so much stress, as a point of great systematic
importance.

Wachsmuth^ omits all reference to the mode of union of the plates in his diagnosis
of the Palaeocrinoidea ; and had it not been revived by Zittel as a means of
distinguishing the two great groups, Muller's name would long ago have fallen into
disuse.

The name Palaeocrinoidea was proposed by Wachsmuth^ in 1877 to denote "all true
Crinoids in which the actinal side is closed ; " but it was not actually defined by him
until two years later, nearly simultaneously with the appearance of Zittel's classification.
He regards the group as of sub-ordinal value, and as specially distinguished by two
characters — (l) the interradials constitute important elements of the test ; (2) the
absence of external food-grooves or oral aperture. He proposed incidentally to group the
later Crinoids together under the name " Stomatocrinoidea " ; but he did not attempt to
define the group ; and so far as I am aware, this name has not been adopted by
systematic zoologists, while Wachsmuth himself is now inclined to abandon it. Various
reasons, which will be explained more fully subsequently, have induced the writer to
propose the name of " Neocrinoidea " for the Mesozoic and later Crinoids. This has been
adopted by Prof. Zittel, and also by de Loriol in his work on the French Jurassic
Crinoids, and it will be used throughout these Eeports.

Although Marsupites is ranked among the Tessellata by Miiller, and also, together
with Uintacrinus, by Schliiter and Zittel, I can see no reason for excluding these two

1 Ann. and. Mag. Nat. Hist., 1881, ser. 5, vol. vii. pp. 283-287. « Revision, part i. p. 30.

' Amer. Journ. Sci. and Arts, vol. xiv. p. 190.

148 THE VOYAGE OF H.M.S. CHALLENGER.

types from the group of tlie Neocrinoids, to which they have been already provisionally
referred by de Loriol.

Both have uniserial arms, a symmetrical calyx, and no anal side ; while in
Marsupites, at any rate, the first radials were perforated by canals, and united to the
second by muscular joints. Interradial plates, however, are well developed in both
genera, especially in Uintacrinus. But the upper series of so-called interradial and
interdistichal or interaxillaiy plates are really parts of the radial system, and correspond
to the pinnules of free arms, as was pointed out by Meek. At the same time he noted
their unusual size, and the fact that they are united with each other and with the main
divisions of the rays for some distance up, so as to constitute a part of the walls of
the body.^

Schliiter,^ to whom Meek's remarks seem to have been unknown, speaks of this as a
possibility, but rejects it on account of the absence of a central canal in the suj^posed
pinnule-joints, and other less important reasons. I cannot help suspecting, however,
that the canal will be found, and that the plates in question are really the basal joints of
the large lower pinnules. He describes how these plates group themselves together in
double rows, the lowest of which " geht aus von dem zweiten Distichalgliede. Sie
besteht vieUeicht aus 9 Stiicken jederseits. Die folgende Doppelreihe, aus kleineren
Tiifelchen zusammengesetzt, nimmt ihren Anfang vom ftinften Stiicke liber dem
Axillare." The first pinnule being on the second brachial, the next on the same side
would Ije on the fourth ; but since the third is a syzygial or double joint, the fourth
brachial is primitively the fifth above the axillary ; while Schluter's figures ^ show that
the double row of interdistichal pieces which " nehmen ihren Anfang vom vierten
Distichale aus " are really the pinnules on the epizygals of the two thii'd brachials. There
are many species of recent Crinoids (Pentacrinus, Metacrinus, Actinometra) which have
large lower pinnules with the basal joints closely fitted together just as in Uintacrinus
(PL XXXVIII. ; PL XXXIX. fig. 1 ; PL XLIII. fig. 2 ; PL LII. fig. 1) ; so that the
supposed resemblance in this respect between Uintacrinus and the Palseocrinoids goes
for nothing. Apart from these two genera there are no Secondary Crinoids which could
by any possil)ility be referred to the Tessellata ; and this is still more em^jhatically the
case with the Tertiary and recent forms. It is true that the most striking characters of
the recent Thaumatocriaus (PL LVI. figs. 1-4) indicate an affinity to early Palaeozoic
types {ante, pp. 39-46) ; but, considering that Thaumatocrinus is a Comatula, it is
more than probable that this resemblance is not due to any genetic connection.

Thus then I regard the Neocrinoids as constituting a group or subclass which is
distinctly marked ofl' from its Pala30zoic predecessors. These became extinct with the
Palseozoic epoch, like the Blastoids, Cystids, and Palsechinoids. The latter in fact

1 Grinnel, Note on the genus Uintacrinus, Bull. U.S. Geol. and Geog. Siu'vey of tlie Territories, vol. ii., No. 4, p. 377.
' Op. cit, p. 58. 3 Qp_ ^i(_^ xaf. iv. figs. 1, 3.

REPORT ON THE CRINOIDEA. 149

constitute a parallel group to tlie Palaeocrinoids, as Urchins are abundant in the
Secondary and Tertiary rocks and also at the present time. But the Blastoids and
Cystids shared the fate of the Trilobites, and did not persist into the Triassic Seas.

It "ftall now be advisable to discuss the various characters which distinguish the
Neocrinoids from the Pateocrinoids. Some of them have been alluded to elsewhere, but
no attempt has yet been made to group them together.

In the first place the calyx of a Neocrinoid is very symmetrical in its composition,
though it may undergo more or less distortion, as in the Eugeniacrinidse and Holopidse
(Pis. I. -IV.).

Under-basals are rarely present {Encrimis, Extracrinus, and Marsupites) ; while by
far the greater number of genera have five equal and similar basals, with five equal and
similar radials resting upon them. Hyocrinus it is true has only three basals (PI. VI.) ;
while the radials of Holopus (PI. III. fig. l) and Eudesicrinus are not symmetrical ; Ijut
their want of symmetry is not due to the intercalation of any anal plate as in nearly all
Pala)Ocrinoids, so that one side of the calyx becomes especially distinguished as the anal
side. In all Neocrinoids, with the single exception of TJiaumatocrinus, these primary
radials are in contact with one another l:)y the entire length of their sides ; or more
rarely, as in Guettardicrinus, Uintacrinus, and Ajyiocrinus^roissyamis,^ their distal angles
are cut away, so as toreceive the lower part of the first interradial. This feature, which
is common enough in the Palaeocrinoids, is rare in the Neocrinoids ; for the second radials
are usually wider than the first, and more or less completely united to their fellows by
their lower angles (PI. XV. figs. 1, 2 ; PL XXIX. fig. 1 ; PL XXXVII. figs. 1,2;
PL XXXIX. fig. 1 ; PL XLIII. fig. 2 ; PL XLIX. figs. 1, 2).

In like manner there is only one known genus and species of Neocrinoids
{Thcmmatocrinus renovatus, PL LVI. figs. 1-4) which has the rays completely separated
by primary interradials that rest on the basals ; though there are several genera of
Palaeocrinoids distinguished by this peculiarity, which is characteristic of the large and
important family Ehodoerinidse (Wachsmuth and Springer).

. In most Palaeocrinoids a certain numljer of the plates above the primary radials
become closely united to one another and to the interradials so as to form the walls of a
relatively large and substantial calyx. This is especially the case in the Actinocrinidas ;
"while in the Platycrinidee comparatively few plates are enclosed within the calyx, many
of them, which in the Actiuocrinidse form a conspicuous part of the body, being here
found in the lateral appendages. In the Platycrinidae the calyx proper is constructed
almost exclusively of basals and first radials, all higher orders of radials either forming a
part of the brachial appendages, or, when partially incorporated with the calyx, being
insignificant compared with the other parts." ^

In the lehthyocrinidse, however, while the secondary and tertiary radials form a part
1 See woodcut, fig. 9 on p. 183. " Wachsmuth, Eevision, part ii. p. 55.

150 THE VOYAGE OF H.M.S. CHALLENGER.

(if the body, they are often counected by a pavement of minute irregular plates, which
may commence as low down as the level of the second radials, and are thought Ijy
Wachsmuth and Spriuger to indicate the presence of a flexible perisome. This is
especially the case in some species of Taxocrinus, the rays of which must have been at
least as free as those of many Pentacrinidse and Comatulse, and much more so than those
of Apiocrinus and Guettardicrinus. In these two genera, as in Uintacrinus, the calyx
interradials are not only present but well developed, so as to increase the size and com-
plexity of the cup. In fact the rays of Guettardicrinus are immovably united as far as
the second brachial, either directly, or by the intervention of interradial plates ; whUe
some species of Apiocrinus (Ajnocrinus -parhinsoni) have the second and third radials in
close lateral contact with their fellows. Other species, however, with the arms free from
the radial axUlaries, have a well defined pavement of interradial plates, the lowest of which
are large and regular and rest on the upper angles of the first radials, as in Ap>ioc7'inus
roissyanus ; ' while the upper ones are smaller and more irregular, and pass gradually
upwards into those of the ventral side. The same is the case in Marsupites.

Many Pentacrinidse and Comatiilse have wide rays which are in close lateral contact
just as in Ajnocrinus parhinsoni (PL XV. fig. 2 ; Pis. XVIII., XIX.; PI. XXV.;
Pis. XXVIII.-XXX.), while others have the rays more separated from one another, but
united by flexible perisome in which the joints of the lower pinnules and numerous small
irregular plates are imbedded (PL XIII. fig. 1 ; PL XXXI. ; PL XXXVII. fig. 1 ;
PL XXXIX. fig. 1 ; PL XLIX. figs. 1,2; PL L. fig. 1). These may cease at the level
of the thii'd axillaries, or pass up into the plating of the ventral side as in Apiocrinus
roissyanus, Marsupites, and the Liassic species oi Extracrimis. But they are never so large
as in these fossils, and more nearly resemble the small ii-regular plating of the Ichthyocrinida3.

Thus then there are many Neocrinoids with no interradial plates in the calyx ; and
when these plates are present and well defined, as in Apiocrinus, Guettardicrinus, and
Marsupites, or Uintacrinus, they are not limited to any special side of the calyx, but
are equally distributed all round it ; so that there is no distinction of the anal side,
Thaumatocrinus of course excepted.

In the Palasocrinoids, however, the pentamerous symmetry of the calyx is almost
always disturbed by a greater or less modification of the plates on the anal side. The
difl'erence may be very slight, as in Phimocrinus and Cupressocrinus, which have the
anal opening separating the muscle-plates of two adjacent radials. But even this
character apjiears to be absent in the remarkable genus Erisocrinus from the Upj)er
Carboniferous of America, which has a calyx unusually Uke that of Encrimts; and also in
Steminatocrinus from the Eussian Carboniferous, which is still more like Encrinus in the
structure of the arms.

Some forms have a special anal plate between two of the primary radials. This is the

1 See woodcut, fig. 9 on p. 183.

REPOET ON THE CRINOIDEA. 151

case in Belemnocrinus, and has been thought to afford the only point of difference between it
and the cup of the recent Rkizocrhius ; though the structure of the stem is entirely different
in the two types. More commonly, however, there is a considerable development of plates
in the anal interradius, and the corresponding basal plate, together with the two radials
that rest upon it, differ from their fellows in size and shape. There is, however, one
Neocrinoid, the aberrant Thaumatocrinus (PI. LVI. figs. 1-5), which has the symmetry
of its calyx disturbed by the presence in the anal interradius of a jointed appendage,
somewhat similar to that of Reteocrinus from the Trenton grouji of America, and of
Taxocrinus, Onychocrinus, &c. But there is no other modification of the plates, the five
basals, the lowest interradials, and the radials being all equal and similar to their fellows.
Even in Palseocrinoids which have the radials closely united all round, and the general
contour of the calyx perfectly regular, a want of symmetry is indicated by the inequality
in the numbers of basal and radial plates. This is the case, for example, in Eucalypto-
crinus, which has five radials but only four basals.

In all Neocrinoids (with the possible exception of the doubtful Comaster) the basals
are pierced by interradial canals or grooves, which lodge the cords proceeding from the
angles of the chambered organ. These cords fork and pass on into the radials, where they
occupy the more or less double axial canals (PL Vllb. fig. 2, ai ; PL XXIV. figs. 7-9 ;
PL LVIII. figs. 1-3 — ai, ar). These axial canals open on the distal faces of the radials in
the centre of the transverse articular ridge, and are thence continued into the rays and
arms (PL III. figs. 1, 3-15 ; PL Vb. fig. 1, A; PL Vc. fig. 2, A, figs. 3, 7, 8, 10, a;
PL Vila. figs. 15-23; PL Vllb. figs. 1, 5-8, A; PL Villa, figs. 5, 7, 8, A ; PL X.
figs. 1-4, 6-8, 17, 18 ; Pis. XII., XXL, LXTL). Chapman, who has recently attempted a
classification of Crinoids,^ appears to imagine that Marstipites and Uintacrinus have
imperforate radials ; for he places them both in his Division 1, " Emedullata," which he
characterises as follows, " Calyx and arm-plates without internal canals." I know nothing
about the calyx-plates of Uintacrinus; but the central canal of the third brachial is
shown in Schlliter's figure ; ^ while the perforation of the radials of Marsupites was
mentioned by Miller,^ and has been figured by many later writers.

The presence of a ti'ansverse ridge on the articular faces of the radials indicates that
they were united to the second radials by muscles and ligaments, just as the successive
arm -joints are. This is invariably the case in all Neocrinoids except Guettardicrinus,
which has the plates suturally united, and possibly also in Uintacrinus. I have pointed
out already that there was a fully developed muscular joint between the two lowest
radials of many Palseocrinoids such as Platycrinus and some, if not all, species of Cyatho-
crimis; and though the arms of Platycrinus are pierced by axial canals, it is placed
among the " Emedullata " by Chapman.

1 A Classification of Crinoids, ruaJ before the Eoyal Society of Canada, May 26, 1882.

2 Op. cit; Taf iv. fig. 4. " 3 Qp. cit., p. 138.

152 THE VOYAGE OF H.M.S. CHALLENGER.

None of the Neocrinoids, however, have permanently imperforate radials as so many
Pateocrinoids have, the latter group remaining in anembryonic condition as stated already.
Tn by far the larger number of Neocrinoids which have divided rays, the axillary is
the third of the primary radials. The only exceptions are Metacrimis and Plicatocrinus.
In the former genus (PI. XXXIX. fig. 1 ; PI. XLVI.) the first and the axillary radials
are primitively separated by from three to six joints, some of which afterwards become
united by syzygy ; while in Plicatocrinus there appear to be only two radials altogether,
the first and the axillary. Zittel ^ describes three, it is true, or rather two radials and
an axillary brachial ; but he speaks of the " innig verschmolzenen Plattchen der unteren
Zone ' as quite small, and I am strongly inclined to suspect that they represent basals
rather than first radials. For what he calls the second radials seem to me to be the
first or calyx radials. They are the large trapezoidal plates forming the greater part of
the calyx, and rmited to the lower series by suture ; and such a mode of union of the two
lower radials occurs in no other Neocrinoid except Guettdrdicrinus.

The position of the axillary joint in those Palseocrinoids which have divided rays is
by no means so fixed as in the younger types, for the first radials themselves may be
axilliary as in AUagecrinus ; while in Poteriocrinus radiattis the axillary is the sixth joint
beyond the first radial, as in some species of Metacnnus (Pis. XLIV., XLVL, XLVII.-L.) ;
and in other genera its position may be anywhere between these two extremes. This is
in fact the only important character which distinguishes the Palaeozoic Erisocrinus,
Philocrinus, and Stemmatocrinus from the well kuowai Triassic genus .Encrimis.
Erisocrinus has distinct under-basals like those of Encrinus, though relatively larger ;
but in Stemmatocrinus, according to Wachsmuth and Springer," these plates are
represented by a flat disk, which is undivided, regularly pentagonal, and extends
considerably beyond the periphery of the column. TrautschokP appears to take the same
view of Stemmatocrinus. Tempting as it may be, owing to the way in which it would
increase the resemblance between these types and Encrinus, I feel somewhat loth to accept
it. For the plate in question appears to me to be much more truly represented by
the central pentagonal piece on which the basals of Citjyn'ssocrinus rest ; this is larger
than the stem-joints beneath it, and is obviously what Schultze^ calls it, viz., "Eine
fiinfseitige, aus der Erweiterung des obersten Siiuleugliedes gebildete Platte." Unfortun-
ately we are not acquainted with the mode of development of the under-basals, as they
occur in no recent Crinoid ; but the analogy of the development of the other calyx-
plates indicates that they are primitively five separate plates, like their homologues in the
apical system of Ophiurids and Stai-fishes ; and a theory which would homologise them
with a plate that first appears as a simple ring, seems to me to run counter to all true

' Ueber Plicatocrinus, he. cit., pp. 107, 108. ^ Revision, part i. p. 141.

3 Einige Crinoideen und andere Thierreste des jiingeren Bergkalks im Gouvernment Moskan, Bull. Soc. Imp. des
Nat. Moscov, 1867, p. 28.

* Op. cit, p. 15, Taf. ii. fig.'^. 1 a, 6 a.

REPORT ON THE CRINOIDEA. 153

notions of morphology. I freely admit the functional analogy of the under-basals of
Encrinus, JErisocrinus, &c., with the central plate of Cupressocrinus or Stemviatocrinus ;
but until the apparently simple nature of the latter shall have been proved to be really
due to the disappearance of sutures, as in the basal ring of Bathycrinus, Rhizocrinus, and
Agassizocrinus, I think that we must regard it as a top stem-joint, corresponding to what
de Loriol calls the " article basal " in Aioiocrinus and Millericrinus.

Encrinus is remarkable as being the only Neocrinoid with ten (or twenty) arms of
biserial joints, which increases its resemblance to Stemmatocrinus. There are, however,
some species (Encrinus gracilis) with ten uniserial arms, as in the other Neocrinoids and
in Erisocrinus so far as yet known. This is also the case in de Koninck's genus
Philocrinus from the Carboniferous strata of the Punjaub.^ But the basals seem to be
much higher and the cup generally deeper than in either Erisocrinus or Stemmatocrinus.
The structure of the lower part of the cup was unfortunately obscured in de Koninck's
specimen, so that the presence of under-basals is still doubtful.

Wachsmuth and Springer point out that the absence of any anal plates in Erisocrinus
and Stemmatocrinus, and the want of any knowledge of their ventral side render it
doubtful " whether they belong to the Cyathocrinidae, or even to the Pateocrinoidea ; and
if it had not been for their marked resemblance to Eupachycrinus, in which a ventral
tube has been observed, and that both were representatives of the same geological age,
living under the very same conditions, we should have felt strongly disposed to place the
whole genus with Encrinus, with which it has, indeed, both in body and arms, the
closest affinities." ^ They think the number of radials to be not of material, or, at most,
" only of generic importance ; but in Encrinus the aboral side of the body, or the plates
which in all Cyathocrinidse constitute the calyx, form almost a flat disk — at least do
not extend beyond the basal plane — and this is the only distinction which can be
discovered between the two forms in the fossil state. This, however, may involve
important structural modification in the internal anatomy of the animal, and probably
shut out Encrinus entirely from the Pateocrinoidea."

Our knowledge of the anatomy of recent Crinoids, however, does not favour this
supposition. There is very much less difi'erence between the calices of Encrinus and
Erisocrinus than between that of Antedon eschrichti with high radials and a narrow but
deep central funnel, and the low flattened calyx of any Actinometra. But the only
difference exhibited by the ventral sides of these two types is that the mouth is central
in the Antedon and excentric in Actinometra. I can therefore see no reason for
supposing that Erisocrinus had a solid inflexible vault built up of the so-called oral plates,
like that of the CyathocrinidES, with which family it, as well as Stemmatocrinus, is placed

1 Description of some Fossils from India, discovered by Dr. A. Fleming of Edinburgh, Quart. Journ. Geol. Soc,
vol. xix. p. 4, pi. ii. fig. 5.

2 Revision, part i. p. 142.

(ZOOL. CHALL. EXP. — PART XX.XII. — 1884.) ^^ -^

154 THE VOYAGE OF H.M.S. CHALLENGER.

by Waclismutli and Springer on account of its resemblance to Eupacliyciinus. No sucli
vault has been found in this type, but only a small " ventral tube " which rests on a series
of three anal plates and has been traced to the height of the fourth or fifth arm-plate, where
it is composed of small, very delicate, hexagonal plates. But these anal plates are absent
in Ensocrinus as in Encrinus, and since a ventral tube or sac like that of Ci/athocrinus
is always found associated with a system of anal plates, the lowest of which is inter-
calated between two radials, it seems rash to postulate its presence in the symmetrical
Erisocrimis.

It should be remembered too that the Liassic Extracrinushas a symmetrical calyx with
a dicyclic base, i.e., of the same composition as that of Encrinus and Erisocrinus ; while
its summit or ventral side was in no way different from that of a recent Pentacrinus.
I have a strong suspicion that this is also true of many Pateocrinoids, and do not therefore
believe that Erisocrinus must have had a closed vault because it was a Palseocrinoid.
Together with Encrinus, Philocri7ius, and Stemmatoci'inus it certainly aifords the best
transition yet known between the Neocrinoids and Palseocrinoids. For the only point of
difierence about which we are entitled to speak with certainty is the absence of a second
radial in the three older forms, and the constancy of its presence in Encrinus, as in most
other Neocrinoids. The occurrence of Encrinus-like forms in the Carboniferous strata of
India, America, Russia, and also Spain (according to C. Barrois) is therefore very interesting.

Some species of Taxocrinus and Heterocrinus bave been thought to bear a superficial
resemblance to Pentacrinus, owing to the freedom of their rays. But both genera have
an asymmetrical calyx with a well marked anal side, and also a variable number of radials,
peculiarities which, when occurring together, are very characteristic of the Palseocrinoidea.

Wachsmuth and Springer^ have pointed out that " another very characteristic distinc-
tion between ancient and recent Crinoids is to be found in the comparatively large size
and massive body plates in the fossil, contrasted with the diminutive body and very long
and highly developed arms of recent types ; and the same is even more strikingly true as
to Blastoids and Cystideans."

But it is a mistake to suppose, as they do, that while the arms are in progress of
growth in the Palseocrinoids, those of the Pentacrinidas are fully developed ; for it is
among the Comatulse that the greatest development of the arms is to be found. Very
few Pentacrinidse, except Extracrinus, Pentacrinus asterius, and three or four species of
Metacrinus have more than fifty arms, a number which is never reached by Apiocrinus
and Bourgueticrinus. In certain sjoecies of Actinometra, however, the rays may branch
six or eight times, and the number of arms exceed one hundred ; while in many species
both of Antedon and Actinometra, the number of joints in a single arm is over two
hundred, and in rare cases reaches three hundi-ed. Nearly all of these bear functional
pinnules, the last of which are sometimes longer than their predecessors.

1 Revision, part i. p. 6.

REPORT ON THE CRINOIDEA. 155

In the recent Pentacrinidse, however, there are never as many as two hundred arm-
joints, and several of the outermost are entirely devoid of functional pinnules, having
nothing but mere stumps in their place, without any ambulacral plates at the edges of
their ventral grooves (Pis. XXXVIIL, XL., XLII.-XLIV., XLVIII., XLIX.). Nearly
one-third of the length of the arm of a Pentacrinus or Metacrinus may be in this un-
developed condition, which is never met mth among the Comatulse.

In almost aU Neocrinoids the articular facets on the first radials occupy the whole
width of their distal faces, so that the lowest parts of the rays, whether divided or not,
are of nearly the same width as the radial plates which bear them. In many Palseo-
crinoids, however, such as Platycrinus, the articular facet of the first radial simply
occupies the middle of its distal edge, so that the lowest parts of the rays are quite small
compared with the calyx. This is the case among the Neocrinoids in Hyocrinus
(PI. VI.), Plicatocrinus, and to a less extent also in Marsupites; while it is very
characteristic of the young stages of the Pentacrinoid larva of Comatula. But the
occurrence of this feature is far more general among the Palseocrinoids than in the later
forms.

Except in some species of Encrmus, the arms of a Neoerinoid are invariably uniserial,
i.e., composed of a single series of joints which are placed end to end, and bear pinnules
alternately on opposite sides. The arms of the earlier PalsBocrinoids were also composed
of single joints ; but in all the three principal divisions of the group the composition of
the arms changed from a single to a double row in the Upper Silurian period. If this is
to be considered as an advance in development, then all the Post-Triassic Crinoids are in
this respect permanent larval forms. According to Wachsmuth and Springer's descrip-
tion of Mariacrinus, the double joint arrangement is brought about by the coalescence of
two contiguous arms, an approach to which may perhaps be found in the flattening of
the sides of the lower parts of the arms in many Pentacrinidse and Comatulse, and in
Holopus (PI. Va, fig. 3 ; PL XV. fig. 2 ; PI. XVI. fig. 1 ; PL XXX. fig. 1). But this is
merely superficial, and the alternate arrangement of the pinnules is unchanged, which is
far from being the case in Palseocrinoids with biserial arms. Other Palseocrinoids, such
as the Ichthyocrinidse and some Cyathocrinidse, seem to have had no pinnules at all,
though the arms branched freely.

It will be apparent from what has been said above that except in one or two points,
e.g., the symmetry of the calyx, the diflferences between the so-called TesseUata and
Articulata are not so great as has been sometimes imagined. But there is one other
structural character of great importance, to which attention has been especially drawn of
late by Messrs. "Wachsmuth and Springer as distinguishing the two groups. I refer to the
condition of the mouth and of the oral surface generally.

The American palaeontologists^ define the cup of a Palseocrinoid as " closed on the

1 Revision, part i. p. 30.

156 THE VOYAGE OF H.M.S. CHALLENGER.

ventral or oral side by a more or less solid integument, without external food-grooves or
oral aperture ; " though they imply that the mouth may not have been internal in some
cases. ^ In recent Crinoids, however, the mouth and food-grooves are external, though
capable of being closed by plates, and the name " Stomatocrinoidea " was consequently
suggested for them by Wachsmuth and Springer.

Eeference has already been made to the gradual removal of the orals which surround
the tentacular vestibule of the larva, from the radial plates ; and also to the separation of
these orals from one another so as to oj)en the tentacular vestibule to the exterior and
expose the mouth in the centre of its floor. In Holopus, Hyocrinus, and Tliaumatocrinus
the orals persist as large triangular plates which cover up and protect the peristome
(PI. III. fig. 2 ; PI. VI. figs. 1-4 ; PL LVI. fig. 5). They are only removed to a short
distance from the radials, scarcely at all in fact, in Holopus. In Rliizocrinus they are
relatively much smaller ; while they disappear altogether in the Pentacrinidse and
remaining Comatulse (except Tliaumatoa'inus), so that the mouth is directly exposed to
the exterior (PL LV.).

In all the recent Crinoids the food-grooves of the disk are perfectly open, like those
of the arms, i.e., they are never closed in any other way than by the folding down over
them of the plates at their sides (PL XVII. fig. 6 ; PL XXVI. figs. 1,2; PL XXX. fig. 2 ;
PL XXXIX. fig. 2 ] PL LV.). But in many Palseocrinoids such as Actinocrinus, these
food-grooves themselves were concealed beneath a vault or dome of rigid heavy plates ;
so that the mouth towards which they converged was truly subtegminal. The nature of
this dome is a point of very considerable importance with reference to the relationship) of
the Neocrinoids and Pal^ocrinoids.

Wachsmuth'^ thinks that it " cannot in the remotest degree be homologised " with the ,
ventral side of the Neocrinoids. " The solid dome forms, as I think I have proved, a
continuation of the radial and interradial series of the dorsal side, and serves merely as a
covering and protection for the organs underneath. It is in every sense of the word
aboral, and forms a part of the abactinal system^ [while the actinal system], which being
already reduced in the Pentacrinidse and Comatulse to a narrow tentacle furrow, recedes
in Palaeozoic Crinoids one step farther and disappears within the solid walls of the body.
The actinal system here consists externally only of the arm furrows, whence it continues
underneath the vault. These Crinoids, therefore, are evidently of lower development and
belong to an inferior type. . . . The Palaeozoic Crinoids, embracing therein all true
Crinoids in which the actinal side is closed, represent the young stage of growth of the
living types." Elsewhere Wachsmuth* speaks of the ventral covering of Actinocrinus

1 Revision, part i. p. 6. ^ Amer. Journ. Sci. and Arts, vol. xiv. p. 190.

3 The words enclosed in brackets were unfortunately omitted in the original, thereby confusing the author's
meaning not a little.

* Revision, part ii. p. 14.

REPORT ON THE CRINOIDEA. 157

and its allies as forming "a free arch which braces the entire oral side of the body
without the aid of oral plates."

The use of the term " aboral" as applied to the vault is somewhat unfortunate, for it
actually does cover in the mouth ; while the j^lates of the opposite or dorsal side of the
body have generally been called aboral by writers on Echinoderm morphology. Wach-
smuth himself applies this expression to the plates of the cup up to the level of the arms.
In like manner he gives the name " apical dome plates " ^ to " a system of plates in the
vault which occupy a position analogous to that of the apical plates of the calyx ; " viz., a
central plate, with a proximal ring of interradial, and a distal ring of radial plates disposed
regularly around it. These must be carefully distinguished from the dorsocentral, basals,
and radials, which are the apical plates of the calyx ; and, as mentioned above, have definite
homologues in the apical system of Urchins and Stellerids. [See Appendix, Note A. j

The suggestion of Wachsmuth and Goette that the Palaeocrinoids represent a
comparatively early stage in Crinoid ontogeny, before the opening of the tentacular
vestibule to the exterior, has been very generally accepted. But it must be borne in
mind that though the Palaeocrinoids may be considered as permanent larval forms with
respect to the closure of the actinal side, yet that in other respects they have developed
to a far greater extent than any Neocrinoid. The solid vault of an Actinocrinus is a
structure sui generis, unless, as I believe, its proximal ring of interradial plates is
represented by the orals of a Neocrinoid. The extraordinary development of arms or of
other appendages which we find in forms like CaUicrinus, Pterotocrinus, Ollacrinus,
Eucalyptocnnus, Crotalocrinus, &c., is entirely without a parallel among the more
regular and symmetrical Neocrinoidea. We must be careful therefore not to make too
much of the one or two embryonic characters presented by the Pala^ocrinoids, as com-
pared mth the facts of their great complexity of structure and immense variety of form.

The simplest type of summit to be met with in any Pala30crinoid is that presented by
the Devonian genus Haplocrinus, which remains permanently in the condition of a very
early larva. For the orals, together with certain upward processes of the radials on
which they rest, form a closed pyramid just as in an early Pentacrinoid. There are five
openings which lead in Ijeneath this oral pyramid and correspond to the points of
attachment of the arms ; but its apex is completely closed so that there is no external
mouth. The lines of suture between its component plates are generally marked by deep
grooves which descend from the closed apex to end below at the radial openings.
Wachsmuth and Springer call them compartments for the reception of the arms, while
they have been described by Zittel ^ as " nach unten geschlossene, nach oben offene
Ambulacralf urchen. "

This appears to me to be a mistake, and I do not see any reason for supposing that
these orals were ever covered in by plates as he elsewhere suggests.

1 Revision, part i. p. 28. ^ Palaeontologie, p. 347.

158 THE VOYAGE OF H.M.S. CHALLENGER.

If, as Zittel supposes, the ambulacra of the arms were continued past the radial
openings and up the steep sides of the oral pyramid, which is closed at the apex, how did
they reach the peristome underneath ? Did not food-grooves and the subjacent
ambulacral structures pass in beneath the oral pyramid at the radial openings, so as to
reach the mouth and the vascular and nervous rings around it, just as they must have
done beneath the dome of Actinocrinus ? It appears to me that Haplocrinus is
permanently in the condition of a Pentacrinoid larva with a closed tentacular vestibule ;
though in other respects, such as the attachment of the brachial plates by muscles ^ and
ligaments above a perforated articular ridge, it j)resents an advance upon that condition.
There is an important feature in the structure of the oral pyramid of Hcqilocrinus which
has long escaped notice. Goldfuss^ long ago described the furrows which mark the
sutures between the oral plates, and went on to say "wo diese im Scheitelpuncte
zusammenstossen, steht ein rundes Knopfchen." This is quite clearly shown in his figures,
but has attracted no attention from subsequent workers. Lately, however, it has been
discovered by Mr. Wachsmuth that there is really a small but distinct plate in this
position, occupying the central portion of the summit of the oral pyramid. I am indebted
to him both for the opportunity of verifying this observation by examination of his
specimen of Haplocrimis mespiliformis, and for permission to make use of it in this
chapter. This plate is one of considerable importance in its morphological relations. In
accordance with the views which I have expressed elsewhere,^ I believe it to be the
representative on the actinal side, or left larval antimer, of the dorsocentral plate which
is developed in the centre of the right antimer or abactinal side of Urchins, Stellericls, and
Crinoids. I have reason to think that this belief has been adopted by Mr. Wachsmuth
not only so far as regards Haplocrinus and Symbathocrinus, but also for the Platycrinidse
and Actinocrinidse as well. The orals, as shown by Goette,* are the actinal repre-
sentatives of the basals, being developed spirally around the left vaso-peritoneal tube ;
while the basals, like the so-called genitals of Urchins and Starfishes, appear in a spiral
around the right peritoneal tube. They are disposed interradially, and rest in most cases
directly against the dorsocentral ; whUe the orals of Haplocrinus, also interradial, rest
against the single plate discovered by Wachsmuth, which occupies a central position in
the summit, immediately above the peristome, as in so many other Palseocrinoids. I
propose to call it the " orocentral."

In the remarkable series of specimens of AUagecrinus^ which have been obtained by
the collectors of the Geological Survey of Scotland, the smallest and least developed
individuals are in the same morjihological condition as Haplocrinus, or perhaps even in
a more primitive one. The calyx is covered by a round dome of oral plates, which rests

1 Schultze, Echinodennen des Eifler Kalkes, p. 105.

2 Petrefacta Germanire, Tlieil. i. p. 214, tab. Ixiv. figs. 6i, 6c

3 Quart. Jonrn. Micr. Sci., 1879, vol. xix., N. S., pp. 179-182 ; 1884, vol. xxiv. p. 14.

* Loc. cit, p. 621. '" Ann. and Mag. Nat. Hist., 1881, ser. 5, vol. ^'ii., pp. 282-289.

REPORT ON THE CRINOIDEA. 159

on the summit of the radial pentagon. Although I have examined a large number of
individuals with especial reference to this point, I have not succeeded in finding any
trace of an orocentral plate. At the centre of the upper edge of each radial is a minute
opening which leads inwards beneath the dome ; but there is no arm-facet corresponding
to this opening, still less a first brachial resting on the radial as is often found in
Haplocrinus. The dome is rounded and smooth, and not marked by any radiating
furrows like that of Haplocrinus, so that the ambulacra must have passed in beneath it
over the upper edges of the radials. Why then may we not suppose them to have done
the same in Haplocrinus ?

In the larger specimens of Allagecrinus the orals are smaller relatively to the radials,
the upper edges of which have minute semicircular arm-facets ; while the ambulacral
openings above these facets are relatively larger. In the next stage of development the
orals are stiU more reduced relatively to the radials, which bear distinct articular facets
for the attachment of the brachials by means of muscles and ligaments around a perforated
transverse ridge, just as in any recent Crinoid. Even in these individuals, however,
which must have had fairly well developed arms, the relatively smaU oral pyramid is still
closed, just as in the early Pentacrinoid and in Haplocrinus.

Another form which remained permanently in the same condition, but had even better
developed arms, was Symbathocrinus. For the so-caUed "apical dome plates"^ (which I
regard as orals) rest directly upon the upjier edges of the articular faces of the radials ;
and they form a closed pyramid or dome with five radial or ambulacral openings, one
between every two orals. While, however, the orals of Allagecrinus form the whole
dome, its centre is occupied in Symbathocrinus by a single orocentral plate, around
which the orals are grouped, just as in Haplocrinus. For the knowledge of this import-
ant fact and permission to make use of it here, I am again indebted to Mr. Wachsmuth,
who will shortly describe it more fully. He is, I beheve, disposed to agree with me in
considering the central plate as an orocentral, and the circle of apical dome plates around
it as orals, homologous with those of Haplocrinus, Allagecrinus, and the Pentacrinoid.

This dome of oral plates in Symbathocrinus is only very rarely found preserved ; but
its discovery by Wachsmuth is of extreme importance in many ways ; while it indicates
that although no dome has been met with in the two largest specimens of Allagecrinus,
its absence may be only accidental and not natural.

On the other hand, there is the possibility that the dome of oral plates in Allagecrinus
became separated from the radials by the growth of intervening perisome, just as the
orals are in aU recent Crinoids except Holopus ; though whether they also separated from
one another so as to open the mouth to the exterior, must remain undecided for the
present. It is of course possible that they may have separated from one another without
being removed from close proximity to the radials, just as is the case in Holopus ; and

1 Revision, part ii. pp. 17, 67.

160 THE VOYAGE OF H.M.S. CHALLENGER.

tlicy miglit then liave been lost after the decay of the perisome uniting them. Those of
Holojnis were retained in the dry specimen which was figured by Pourtalfes,^ and
subsequently by Sir Wyville Thomson ; and the condition of the Palasozoic Coccocrinus
seems to me to be entirely explained by that of the recent Holopus.

Wachsmuth and Springer^ describe it as follows : — " In well preserved specimens of
Coccocrinus, the vault is constructed of five large oral plates, which rest upon five
interradial pieces. The oral plates are not in contact laterally, but leave five slits, which
in the fossil have no floor nor covering, and leave an open space in the centre." They
are strikingly similar to the orals of the recent Hyocrinus (PI. VI. figs. 1-4), as has been
pointed out by Zittel ; and the resemblance to the orals of Holopus (PI. III. fig. 2) is still
greater, as the latter rest directly against the calyx plates, which is not the case in
Hyocnnus. In both the recent forms and also in Thaumatocrinus (PL LVI. fig. 5) the
clefts between the triangular oral plates are open and uncovered, as in Coccocrinus.
Schultze^ follows Eoemer in thinking that these slits do not penetrate into the cavity of
the calyx ; Ijut that they were hollows for the reception of the arm bases, as in
Eucalyptocrinus. But Wachsmuth, having examined Schultze's specimens, states
distinctly that these grooves have no floor. He says in the Revision (part ii. p. 17) that
'■' the similarity to Hyocrinus is probably merely superficial, as the lateral grooves in
Coccocrinus were evidently {why ?) closed by additional plates as in other Platycrinidse,
while they are open in Hyocrinus." Again "it is evident that the central space and
open furrows were covered in the animal as in similar genera." The oral plates " do not
join laterally nor in the centre, but leave a median space and lateral shts, which in perfect
specimens were doubtless closed, the one by the apical dome plates and the slits by small
marginal pieces." . ..." In Coccocrinus a covering of the ambulacral groove has not
yet been observed, but, judging from the fissure between the oral plates, it probably rested
just upon their edges, and formed an intermediate link between the vault structure of
the Cyathocrinidse and Platycrinidse."*

When Wachsmuth wrote the passages which have been quoted above, he held, like
Zittel and myself, that the five large triangular plates which rest on the primary inter-
radials of the calyx are homologous with the orals of recent Crinoids. He has since,
however, come to the conclusion that " Coccocrinus had externally no oral plates, its so-
called orals are secondary interradials, and mouth and food-grooves were covered by
supra-oral plates " (Extract from Letter). I must confess that I greatly doubt the
existence of this additional covering in Coccocrinus, which seems to Wachsmuth so
evident ; for I find it difficult to believe that the " Scheitelstiicke," as Schultze called
them, are not oral plates like those of the Neocrinoids. It is of course possible that their
resemblance to the orals of Holoinis, Hyocrinus, and Hiaumatocrinus is simply an

1 Hassler Crinoids, pi. x. fig. 9. ° Revision, part ii. pp. 17, 58.

5 Op. cit., p. 89. * Revision, part ii. pp. 17, 30, 58, 59.

REPORT ON THE CRINOIDEA. 161

accidental one, and that I am inclined to attach too much importance to it ; and in that
case I should certainly agree mth Wachsmuth in calling them secondary interradials.
One point, however, is worth notice. The orals of Tliaumatocrinus, like those of
Hyocrinus (PI. VI. figs. 1-4 ; PI. LVI. fig. 5), are only separated from the calyx plates
by the thin band of perisome round the edge of the disk, which develops in the equatorial
zone of the larva, concurrently with the increase in the diameter of the cup. These
genera are thus in the condition of the advanced Pentacrinoid of Antedon rosacea, the
oral circlet of which, owing to the increase in the size of the visceral mass, becomes
detached from the summit of the primary radials ; and the space between the two series
is filled by the perisome of the equatorial zone, which separates the right and left larval
antimers.

In the early stages of Hyocrinus, before this process had taken place, the orals would
rest directly against the upper edges of the radials, as they do in Holopus (PL III. fig. 2),
but in Tliaumatocrinus they must rest against the primary interradial plates. They
would thus be in precisely the same relative position as the " Scheitelstiicke " of Cocco-
crinus ; but I do not think that when a young Tliaumatocrinus is found in this condition
Wachsmuth will call the oral plates "secondary interradials"; though he uses this term
for the plates of Coccocrinus which I, like AUman and Zittel, consider as representing
the orals of Neocrinoids, including Tliaumatocrinus.

If then the "Scheitelstiicke" of Coccocrinus be oral plates, there are strong
morphological reasons against the supposition that the ambulacra which pass in between
them were closed by small marginal pieces, i.e., by covering plates like those which occur
so frequently on the arms and pinnules of many Neocrinoids (PI. Vc. figs. 8-10 ;
PI. XIII. figs. 15, 16 ; PI. XVII. figs. 2-4, 7-9) and Palseocrinoids, and on the summit
of Cyathocrinus. Why should we invoke the existence of a "vault" over the orals of
Coccocrinus, simply because it is a Palseocrinoid l There is nothing of the kind in the
recent Holopus which dates back to the Chalk ; while the allied genera Cotylecrinus,
Eudesicrinus, and Eugeniacrinus are almost as old as the Mesozoic period. In no recent
Crinoid, nor in the Pentacrinoid larvae of those species which have covering plates to the
ambulacra do these plates rest upon the edges of the orals. They commence at the
margin of the peristome which is covered in by the orals, but there are none bordering
the edges of the grooves between these plates.

It is partly, I believe, owing to his feeling the force of this objection that Mr.
Wachsmuth prefers to consider the " Scheitelstiicke " of Coccocrinus as secondary inter-
radials, rather than as orals. When he has found a specimen showing the small pieces at
their edges, I shall probably agree with him. But for the present I am inclined to lay
more stress on the resemblance of the " Scheitelstiicke " of Coccocrinus to the orals of
Neocrinoids ; although, according to Wachsmuth,' " this resemblance is probably merely

1 Revision, part ii. p. 17.
(ZOOL. CHALL, EXP. — PART li^xn. — 1884.) Ii 21

162 THE VOYAGE OF H.M.S. CHALLENGER.

superficial, as the lateral grooves in Coccocrinus were evidently closed by additional
plates as in other Platycrinidse, while they are open " in Neocrinoids. But the existence
of these additional plates does not seem to me so evident as it does to him. He attempts
to support his position by reference to a " close resemblance " between the summit
of Coccocrinus and that of the Cyathocrinidse, which he describes as follows :^ — " Calyx
surmounted by five large oral plates, with a central opening between them, and forming
at their sutures five shallow ambulacral grooves converging toward the centre. Central
opening covered hj the apical dome plates, and the five grooves arched over by two rows
of small immovable pieces alternately arranged."

The so-called oral plates of this description are those which Wachsmuth had previously
called " consolidating plates," thinking them to be homologous with the plates of the
same name in Cupressocrimis. He subsequently came to the conclusion, as did Zittel
about the same time, that they are homologous with the orals of the Pentacrinoid ; and
he therefore also spoke of them by this name. He likewise regarded the deltoid pieces of
the Blastoids in the same way. I must plead guilty to having also adopted this view,
which had much to recommend it at first sight. I did not do so, however, without con-
siderable hesitation, on account of one morphological difiiculty which it involved. For
the ambulacra would then pass over and not hetween the edges of the oral plates, which
would bear a double row of marginal pieces or covering plates continuous with those on
the arm-grooves. This, as pointed out above, is contrary to the nature of the oral plates
of Neocrinoids ; and the result of the correspondence on the subject between Mr. Wachs-
muth and myself is that we can no longer regard either the deltoid pieces of the
Blastoids or "the principal vault pieces" of Cyathocrinus (as Meek and AVorthen call
them) as representing the orals of Neocrinoids. I am now disposed to consider that
" Cyathocrimis and the Blastoids have but one interradial plate, which reaches up to
the peristome." Mr. Wachsmuth's first criticism of this statement was that " it expresses
exactly my views."

I believe, however, that his views have since undergone some further modifications,
as he will himself explain in the forthcoming part (iii.) of his Ee vision.

But the question of the structural resemblance between Coccocrinus and Cyathocrinus
is independent of the nomenclature of the plates. It is one of some importance in its
bearing on the belief of Zittel and myself on the condition of the mouth in the former
type and its relation to that of Neocrinoids. In the first place, as pointed out by Wachs-
muth and Springer," Cyathocrinus has only one plate resting " against the incurved
upper margins of the radials ; " while in Coccocrinus the primary interradial bears a
triangular plate, which they now consider as a second interradial, though formerly
regarding it as an oral plate, as do Zittel and myself. The central opening in
Cyathocrimis is much larger than in Coccocrinus, owing to the truncation of the

' Reviaiou, part i. p. 68. ^ ]]ji±^ yatI ii. p. 17.

REPORT ON THE CRINOIDEA. 163

principal vault pieces or interradials, as they were also called by Meek and Worthen/
The grooves which converge upon it are not like those of Coccocrinus " which have
no floor," ^ but they are formed at the sutures of the interradials, i.e., the apposed edges
of these plates are bevelled away so as to form a groove with the suture in the middle of
its floor.

According to Wachsmuth and Springer,^ the central space " in perfect specimens is
completely covered by the apical dome plates. The food-groove and ambulacral canal
are also arched over solidly by two rows of alternate plates which connect with the
movable covering of the arm furrow." The existence of these marginal alternating
plates has yet to be proved in Coccocrinus, and until this has been done, the resemblance
between this type and Cyathocrinus does not seem to me to be very " close," for while
the arm-grooves of CyatJioc7'inus are continued towards the peristome over the united
edges of the interradials, this is by no means the case in Coccocrinus ; and there is no
groove on the ventral disk at all, any more than there is in Holo^Jus, or in any young
Crinoid before the separation of the orals and radials by the expansion of the equatorial
zone. The fact that the principal vault pieces of Coccocrinus are not united laterally, as
is the case in Cyathocnnus, seems to me to he one of very considerable morphological
importance. It must of course be remembered, as Wachsmuth has pointed out in other
cases, that the absence of a covering to the central space and its radial clefts in the fossil
Coccocrinus is no proof that it was not present during life as in Cyathocrinus. But the
two genera are not in the same morphological condition, and all that we hnoiv about
Coccocnnus goes to indicate its resemblance to Holopus with open slits between the
orals and an uncovered mouth. Wachsmuth,* however, states that he has " yet to
discover a single palaeozoic genus in which a special oral aperture has been identified, or
in which the existence of a solid vault has been disproved, or cannot be traced by
analogy." I believe, on the contrary, that the sjjecial oral aperture is to be found in
Coccocrinus as in Holopus, and that it is pushing analogy too far to assert the existence
of an as yet undiscovered vault in this genus.

If then, as I believe, there was an unobstructed mouth in Coccocrinus as in Holopus,
I cannot agree with Zittel's association of this type with Haplocrinus, which had a closed
oral pyramid.

Wachsmuth and Springer ^ place Coccocrinus near Platycrinus. "The two genera
are identical in the construction of the calyx, and the summit really forms the only
distinction between them." To this point I shall return. Coccocrinus, hke the recent
Holopus, seems to me to be permanently in the condition of a Crinoid larva in which the
orals have not yet moved away from the radials, though separated from one another.

Haplocrinus and Symbathocrinus are permanently in the condition of an unopened

^ Palaeontology of Illinois, vol. v. pi. ix. fig. 13. ^ Revision, part ii. p. 58.

8 Revision, part i. p. 84. ■• Amcr. Journ. Sci. and Arts, vol. xiv. p. 190. ' Revision, part ii. p. 58.

164 THE VOYAGE OF H.M.S. CHALLENGER.

Pentacrinoid ; while the apical dome plates in the Cyathocrinidse, to which reference has
been made already, as covering the central opening of the summit, do not reach anything
like either the size or the regularity of arrangement that is to be found in the plates
which have been described by Wachsmuth under the same name in other Palseocrinoids.
In the Blastoids, with the exception of Eleacrinus, they are generally small and irregular
as in CyathoGrinus ; but in the Platycrinidse, Actinocrinidse, and Rhodocrinidfe they form
a group of seven plates in the centre of the vault. The central plate is surrounded by
six others, four of which are of equal size, while the remaining two are smaller. They
are separated by the anal tube, and correspond to a single plate, just as Eleacrinus has
the deltoid piece on the anal side divided into two parts, which are separated by the anal
opening and its supporting plate.

Professor AUman ^ pointed out long ago that many Palseocrinoids have a group of
plates in the centre of the vault, which is nothing but a more or less extensive develop-
ment of the simple oral system of the young Comatula. This pregnant suggestion refers
to the seven apical dome plates of Platycrinns and Actinocrinus which correspond
as I believe to the orocentral and five orals of Haplocrinus and Symhathocrinus, having
been developed like them on the left larval antimer ; while the rest of the vault in
Actinocrinus is a further extension of this oral system, which is unrepresented in the
Neocrinoids. Thus then, though Platycrinus and Actinocrinus are in the condition of
having the tentacular vestibule and peristome permanently closed, just as in Haplo-
crinus, yet they have undergone an immense development upon this condition as a basis.
The tentacular vestibule in the Pentacrinoid larva is merely the peristome concealed
beneath the oral pyramid ; but in Actinocrinus it is greatly enlarged so as to take in the
whole surface of the disk ; and the ambulacra passed over this surface towards the
central mouth from the periphery of the disk, where they entered the dome from the
arms through the well known arm-openings or ambulacral openings. These gave passage
not only to the ambulacra proper or food-grooves, but also to extensions of the body-
cavity, and to the radiating trunks of the nervous, blood-vascular, and water-vascular
systems. All these last lay between the body-cavity and the food-grooves, and converged
towards their respective circum-oral centres. The upper surface of internal casts of the
vault of Actinocrinus is marked by bifurcating ridges which indicate the position of the
food-grooves radiating from a central peristome, just as in the disk of a recent Antedon,
as has been pointed out by Wachsmuth.^ Traces of these ambulacra are often found in
the interior of the vault. In many cases they were covered in by a double row of
alternating plates just like those of the arms, with which they were continuous at the
arm-openings. They were floored by a double row of plates, and so formed tunnels beneath
the vault, but closed independently of it by the covering plates on their upper surface.

1 On a Prebraehial stage in the development of Comatula, Trans. Roy. Sue. Edin., 1863, \-ol. xxiii. pp. 245-251.

2 Amer. Journ. Sci. and Arts, vol. xiv. p. 119.

REPORT ON THE CRINOIDEA. 165

I suspect that tliese tunnels lodged not only the food-grooves, but also the nerves, blood-
vessels, and water-vessels ; for no recent Crinoid has any calcareous structure between
the epithelium of the food-groove and the under surface of the water-vessel. But there
is often a tolerably definite plating which extends inwards from the side plates toward the
median line of the ambulacrum, beneath the water-vessel (PI. LIV. fig 1 1 ; PI. LVII.
fifT. 4, suh ; PI. LXII.) ; and I suspect therefore that the lower row of plates flooring the
ambulacral tunnels may be of this nature. Possibly, however, they should be regarded as
subambulacral plates better defined in character than those of recent Crinoids, though I
am inclined to doubt this, owing to their alternate arrangement. Wachsmuth has dis-
covered that the proximal ends of these ambulacral tunnels are connected by a circular
vessel which encloses more or less of the upper part of the convoluted digestive organ.
The lower part of this structure, with its floor of minute interlocking plates, was obviously
the water-vascular ring ; and the five interradial openings in its floor were referred to by
myself in 1879^ as indicating the position of the water- tubes which depend from the
water-vascular ring into the ccelom, and serve to admit water into the ambulacral system,
a view which has been adopted by Zittel. Rhizocrinus lofotensis has only one water-
tube in each interradius, which seems also to lie the case in Actinocrinus verneuilianus.

Besides the ambulacral skeleton, we also find in the Palteocrinoids representatives of
the anambulacral plates of Pentacrinus and Comattda. In describing some natural casts
of the visceral mass of Actinocrinus, Messrs. Wachsmuth and Springer ^ say " the inter-
palmar fields are composed of a soft skin, but although this is more or less encrusted with
limestone particles, which sometimes almost look like vault pieces, they have no affinities
with the plates of the vault;" while in some specimens of the Actinocrinidae ^ "almost
the entire test is lined with a delicate calcareous plexus or network. This lining is not
in contact with the test directly, but connected with it by small partitions, producing
innumerable little chambers, which communicate with each other and with the visceral

cavity The structure extends but little below the region of the second radials,

leaves passages at the arm-openings, and toward the vault " reaches to a place near the
median portion of the ray, leaving at the centre an open space in the test which is
occupied by the central vault piece." This open space, lying beneath the central one of
the apical dome plates and uniting the five ambulacra, was evidently the peristomial area
like that of recent Crinoids (PL XVII. fig. 10 ; PL XXVI. figs. 1,2; PL XXXIX. fig. 2 ;
PL L. fig. 2 ; PL LV.) ; while the calcareous network within the vault is divided by the
ambulacra into five interpalmar fields. It corresponds to the limestone particles on the
surface of the internal casts, and represents the anambulacral plates developed in the
perisome of recent Crinoids.

These important observations go to show the complete resemblance between the

■ Quart. Journ. Micr. Sci., 1879, vol. six., N. S., p. 185.

2 Revision, part ii. p. 31. ^ Revision, part ii. p. 26.

166 THE VOYAGE OF H.M.S. CHALLENGER.

ventral perisome of a recent Crinoid and the upper surface of the body beneath the vault
of an Actinocrimis. Both had plated ambulacra and anambulacral plates. The admitted
resemblance of the interpalmar anambulacral plates of the Actinocrinite to " vault pieces "
is another point in favour of the view which I take of the so-called vault in the Ichthyo-
crinidfB. I believe this to be a true " ventral disk " similar to that of Pentacrimis, and
not in any way homologous to the soHd vault of the Actinocrinidaj.

In describing the Palseocrinoids, Wachsmuth uses " vault," " summit," and " ventral
disk " as synonymous terms.^ This is somewhat unfortunate, as tending to lead to con-
fusion.

The expression " oral disk " or " ventral disk " is universally used to denote the upper
surface of the visceral mass of a Crinoid, i.e., that in which the mouth is placed, with the
food-grooves radiating outwards from the peristomial area around it. Wachsmuth speaks
of this surface as the " soft or ventral peristome " (perisome 1), and says that it cannot in
the remotest degree be homologised with the solid vault of the PaliBocrinoids. Hence his
occasional reference to this vault as a ventral disk is a little confusing ; and his use of the
word " interpalmar " is equally so.

He sometimes employs it^ to denote the interradial spaces between the ambulacral ridges
on the upper surface of the casts of Actinocrimis. This surface corresponds to the ventral
disk of Pe«tacrmMs; and " interpalmar " is here used by Wachsmuth in the same sense
as it was by Mtiller, i.e., for the " interradialen Felder zwischen den Tentakelrinnen."
When therefore he employs " interpalmar " to denote the interradial plates of the vault,^
its meaning is entirely different ; for the vault was a dome of solid plates, completely
concealing the mouth, food-grooves, and interpalmar areas on the ventral disk.

The vault and ventral disk are, to my mind, entirely distinct structm-es. The former
is necessarily formed of closely-fitting solid plates ; while the latter, lying beneath it,
may be bare as in many Comatulse, or more or less completely plated as in Actinocrinus
and the Pentacrinidse. But no recent Crinoid, not even Tliaumatocrinus, has anj'thing
like a dome or vault rising above a ventral disk. Numerous specimens are known with
the covering plates at the sides of the food-grooves closed over them so as to convert them
into tunnels (PL XVH. fig. 6 ; PL LV. figs. 3-7). But this was also the case beneath the
vault of Actinocrinus. Sometimes, indeed, the plates of the disk may be so closely set that
the opening of the mouth, which may be large in other specimens, is nearly or quite con-
cealed, as in the so-called recent Cystidean Hyponome (PL LIV. fig. 10 ; PL LV.
figs. 4, 5, 7). But this is in no way comparable to the embryonic closure of the mouth
before the separation of the valves of the oral pyramid. It is only in this and similar
cases that I admit the presence of a vault, dome, or tegmen calycis. This structure
reaches its fullest development in the Actinocrinidte and Platycrinidte ; though Haplo-

1 Amer. Journ. Sci. and Arts, vol. xiv. pp. 181, 185. Revision, part i. pp. 5, 32, 54 ; part ii. p. 53.
- Revision, part ii. pp. 26, 27, 31. ' Revision, part ii. pp. 64, 107.

REPORT ON THE CRINOIDEA. 167

crinns and Symhathocriniis had a more or less rudimentary one, consisting of an
orocentral and five oral plates only, without any radial extensions.

The vault of the Actinocrinidae is much more complex than in the Platycrinidse and
Ehodocrinidje ; but its relations to the internal organs are so much better understood
than in these families that it may well be considered first. This is more especially the
case, as certain parts of Wachsmuth's generalised description of the vault of these three
families apply to the Actinocrinidse only, and occasionally even to some genera only.

It has been stated above that the vault of the Actinocrinidse formed a solid roof or
dome over the oral surface of the visceral mass, which was covered with ambulacral and
anambulacral plates like the disk of a Pentacnniis. Wachsmuth has discovered that
whatever the size and extent of the vault the apical dome plates are invariably present.
" They consist of a central piece, occupying a position directly above the oral centre,
which in this family is quite rmiformly the centre of the disk. It is surrounded by six
proximal plates, interradial in position, of which four are large and equal, and two smaller.
The four large plates are placed above the four regular interradial spaces respectively ;
the two smaller ones, which are equivalent to and take the place of one large plate, are
directed posteriorly, being separated from each other by anal plates or the proboscis. . . .
There are other vault pieces occupying a radial position which are either in contact with
those just described, or, as is more frequently the case, separated from them by a belt of
small pieces. Their number varies considerably among species', and depends upon the
number of primary arms, without reference to the number of bifurcations after they
become free. They increase in proportion to the number of primary arms, in the same
manner and on the same principle as the plates of the calyx, each order of radials has its
corresponding plates in the vault. . . . There are also interradial plates represented
in the summit, occupying intermediate spaces between the radials, but then- arrangement
is very irregular and their number variable."' The number of these interradial plates
depends greatly upon the age of the individual, and is therefore very uncertain ; but as a
general rule the number of summit plates increases regularly with that of the primary
arms, just as that of the calyx plates does. The above statements, though true enough
for the Actinocrinidse, are scarcely so accurate as regards the Platycrinidse.

This discovery of Wachsmuth's respecting the distribution of the radial dome plates
in the Actinocrinidse is of enormous importance for the proper comprehension of the true
nature of the vault of a Palaeocrinoid ; and, taken in connection with the embryological
work of Goette, throws much light on the morphology of the Crinoids, and indeed of
Echinoderms generally.

By far the most interesting of these summit plates are the " apical dome plates " in the
centre, which Wachsmuth says are relatively larger in young specimens. This and other
considerations led him to point out their resemblance to the apical plates of the aboral

1 Revision, part ii. pp. 14, 15.

\

168 THE VOYAGE OF H.M.S. CHALLENGER.

side, which are of equally fuudameutal importance in the calyx.^ " The six proximal
plates surrounding the central piece represent the basals or genitals, and the radial dome
plates the radials or oculars. The centre piece may perhaps be compared with the
underbasals, or the subanal jalate of the Echini." In an earlier statement of these
homologies^ no "perhaps" was used respecting the nature of the central actinal plate ;
and I have pointed out that while accepting Wachsmuth's^ comparison to the full extent,
as regards the radial and interradial plates in the centre of the dome, I cannot follow him
in his recognition of the dorsocentral, nor of its fellow, the orocentral, as homologous with
under-basals. * His views, as expressed in the Eevision, are essentially those of Lovfen,
formerly also held by Agassiz, when allowance is made for the different systems of
nomenclature used by him and by them. I have endeavoured to show elsewhere that
there cannot be a true homology between a dorsocentral plate which is single from the
first, and the five under-basals of a dicyclic Crinoid. These are by no means universally
j)resent, as one would expect them to be, did they correspond to the dorsocentral of the
Eehinozoa, which is such a prominent object in the larva of an Urchin, Starfish, or
Ophiurid ; while representatives of the under-basals of Crinoids are actually present, to-
gether with a dorsocentral plate, in some Starfishes and Ophiurids.* The dorsocentral
is developed at the distal end of the right peritoneal tube ; and as there is a plate
occupying the same position in the Crinoid larva, viz., the future terminal plate at the
base of the stem, it is only natural to regard the two as homologous, as pointed out by
myself in 1878. I am glad to find that this view has commended itself to Dr. Ltitken'
and also to Sladen ; and I understand from Mr. Wachsmuth that he is now in accordance
with me respecting the homologies of the central vault piece, considering it as representing
the dorsocentral of Echinoderms generally, and not the under-basals of those Crinoids in
which these plates occur. [See Appendix, Note A.] His view of the proximal inter-
radial dome plates of the Actinocrinidte, however, is entirely difi"erent from that here
advocated, and will shortly be published in extenso by himself; while he has also
abandoned his suggestion that the interradial dome plates in the Actinocrinidee, Platy-
crinidae, and Rhodocrinidse are " the homologues of the oral plates, which are here broken
up, and represented by five plates instead of one." This relieves me from the necessity
of discussing it here, as I had otherwise intended to do.

My own idea of the homologies of the calyx and dome plates of Crinoids is expressed
in Table VII.

1 Revision, part ii. pp. 15, 16. ^ Amer. Journ. Sci. and Arts, vol. xiv. \>. 189.

2 Quart. Journ. Mia: Sci., 1878, N. S., vol. xvui. pp. 369-371 ; 1879, vol. xix. pp. 181, 182 ; 1882, vol. xxii.
pp. 377, 378.

* See Sladen, On the Homologies of the Primary L;u-val Plates in the Test of Brachiate Echinoderms, Quart.
Journ. Micr. Sci., 1884, vol. xxiv., N. S., pp. 32-36 ; also. The Apical System of Ophiurids, ibid., pp. 3-15.
' Dyreriget, en Haand- og Lferebog til Brug ved h0iere Lajreanstalter, Kjpbenhavn, 1882, p. 597.
" Eevision, part ii. p. 17.

REPORT ON THE CRINOIDEA.

169

I regard the inner ring of interradial dome plates (proximals) in the Actinocrinidae
and PLatycrinidse as representing the orals of a Neocrinoid, or of Symbathocrinus,
Haplocrinus, and the young Allagecrinus. This view was first suggested in 1879/ and
all that has since been written on the subject by Wachsmuth and Springer, who have
not discussed it in any way, has only served to make me more confident of its truth.

In the first place, Wachsmuth himself has recognised from palseontological evidence
the close similarity between the proximal interradial plates of the actinal and abactinal
systems respectively. Those of the abactinal system are the basals, which in Actino-
crinus and Cyathocrinus " develop very early in the young, and attain almost their full

Table VII. — Shoiving the Mutual Homologies of the Princiiml Plates m the Actinal
and Abactinal Systems of Echinodermata.

Urchins.

Opliiurids.

Crinoids.

Actinal.

Abactinal.

Abactinal.

Actinal.

Abactinal.

Symbathocrimts.

Actinocrinus.

1.

Central plate.

Dorsocentral.

Dorsocentral.

Terminal plate
at base of
larval stem.

Orocentral.

Orocentral.

2.

First series,
radial.

Under-basals,
(variable).

Under-basals,
(variable).

3.

Second series,
interradial.

Genitals.

Basals.

Mouth-shields.

Basals.

Orals.

Proximal dome
plates.

4.

Third series,
radial.

Oculars.

Radials.

Primary calyx
radials.

Orders of calyx
radials.

Calyx interra-
dials.

Primary dome
radials.

Orders of dome
radials.

Dome interra-
dials.

No representatives of the under-basals have been detected as yet in the apical system of an
Urchin, nor in the oral (or actinal) system of any Echinoderm.

size when even the first radials are comparatively much smaller." * This is also true of
the basals of recent Comatulae, and of the orals as well ; while in like manner the
proximal dome plates of Actinocrinus are relatively largest in young individuals.

Goette's observations show the complete homology between the basals and orals, both
being developed spirally, the former round the right, and the latter round the left
peritoneal tube. The basals are primitively next to the abactinal centre in Urchins and
Stellerids, and are only removed from it in the Crinoid by the growing stem ; while the
orals are next the actinal centre, no plate being developed there, however, in the recent

1 Quart. Joimi. Micr. Sci., 1879, vol. xix., N. S., p. 179.
(ZOOL. CHALL. EXP. — PART XXXH. — 1884.)

2 Revision, part i. p. 20.
1122

170 THE VOYAGE OF H.M.S. CHALLENGER.

Criuoid. Did it appear, it would only be in the way, and have to undergo resorption to
a greater or less extent, just as the dorsocentral of many Urchins is more or less com-
pletely resorbed after the appearance of the anus.

As regards the recent Crinoids, therefore, the embryological evidence clearly indicates
that the basals of the abactinal system are represented in the actinal system by the orals.
The former are within the ring of radials and next to the dorsocentral ; and it seems
therefore only natural to regard the six proximal interradial plates surrounding the
central piece (orocentral) in the vault of a Palaeocrinoid as representing oral plates.

"Wachsmuth admits that Zittel was right in regarding the interradial pla.tes which
form the dome of Hcqjlocrinus as representing the orals of Neocrinoids ; and he takes
the same view of the interradial plates in the dome of Symhathocrimis. I should have
thought therefore that he would have given a similar interpretation of the summit plates
in Platycrinus and Cidicocrinus. Those of the latter genus were described by Miiller as
follows : — " Der Scheitel besteiit aus 5 Tafeln, welche dicht an einander schliessen durch
Niihte und anderseits bis an die Arme und die Interradialia reichen, sie stehen iibrigens
interradial, so dass jedes Interradiale ^u einem der funf Scheitelstiicke stimmt. Der
Mund {cf. anus) befindet sich seitlich in einem Interradius zwischen dem Interradiale
und entsprechenden Scheitelstuck. Auf jeder der 5 Scheiteltafeln ea-hebt sich ein
kurzer Dorn."^ Wachsmuth calls these " Scheitelstiicke " of MilUer's the proximal vault
pieces, and suggests that " probably his largest plate includes four plates, the spiniferous
central vault piece, the two proximal vault pieces, and a small anal plate between them.
The four large proximal vault pieces, each crowned with a spine, are no doubt correctly
represented. Those few plates occupy the greater part of the summit, leaving Ijut little
space for the radial dome plates, which as yet are unknown." ^

I think myself that this suggestion is totally unnecessary ; for the five summit plates
appear to me to form a closed oral pyramid :in which the plate on the anal side is somewhat
wider than its fellows, but not divided into two as in the Actinocrinidas and Platycrinidse.

CuUcocrinus has the proximal dome plates resting against the calyx interradials, just
as the orals do in Coccocrinus. But no distal ring of radial dome plates is known in
this genus. This advance in complexity is presented by certain forms of Platycrinus from
the Carboniferous limestone, both at Burlington and at Bolland. Miiller gave some
figures of the latter under the name of Platycrinus ventrieosus, Goldfuss, which is
certainly wrong ;^ and I have examined several specimens in the British Museum collec-
tion which are in this comparatively primitive condition. The proximal dome plates rest
directly against the calyx interradials,* that on the posterior side being represented by
two small plates with the anus between them ; while there is a more or less tubercular

^ Verlil. d. Naturhist. Verein. d. preuss. Eheinl., Bd. xii., Jahrg. 1855, p. 23.

' Revision, part ii. p. 62. ^ Bau des Pentacrinus, loc. cit., Taf. vi. figs. 1«, 16.

* It is only just to Mr. AVaclismuth for me to state here that he was the tir.st to direct my attention to the fact that
Platycrinidie e.\ist with tlie proximal dome plates resting directly against the calyx interradials, two individuals in this

REPORT ON THE CRINOIDEA. 171

riug of radial dome plates outside them. These proximal dome plates thus correspond
exactly to the orals of Symhathocrinus and Hcq)locrinus, covering in the peristome and
resting against the calyx plates, which in the Platycrinus are the interradials, and not
the upper edges of the radials, as in the simpler forms.

The disk underneath was therefore larger, owing to the greater width of the cup, and
an additional ring of plates, the radial dome plates, was necessary to protect it.
Wachsmuth admits the homology of these with the calyx radials, which are separated
from the dorsocentral by the basals ; and I cannot therefore see what other view can be
taken of the proximal dome plates which immediately surround the orocentral, than to
regard them as orals, i.e., as the actinal representatives of the basals, like the correspond-
ing plates in SymhathocHnus. K this be admitted, it follows that the proximal dome
plates of all Platycrinidae, Actinocrinidse, and Ehodocrinidse are also homologous with
the orals of Neocrinoids.

In most Platycrinidas the primary radial dome plates are succeeded by others of a
different character, which do not precisely correspond to the various orders of calyx
radials, as will be mentioned later ; but in the Actinocrinidse the correspondence between
the actinal and abactinal sides is very complete. Thus in Stelidiocrinus capitulum ^ the
orocentral and six proximals (orals, mihi) take up almost three quarters of the summit,
the remainder of which is occupied by a single anal, and three radials to each ray,
the two o-uter ones being very small. All gradations can be traced from this condition
into the complex vault of a Strotocrinus, which is described, as follows by Messrs.
Wachsmuth and Springer : —

"The disk is paved by many hundreds of small minute pieces which decrease in size
towards the arms, and which at the outer points of the rays become almost microscopic.
The apical plates are larger, and are separated from each other, but not otherwise
distinguished from the other plates, and hence are not easily identified." ^ In young
individuals, however, there are comparatively fewer summit plates than in the adult.
" The apical and principal radial pieces are larger than the intervening interradial plates
which, exceptionally in this genus, attain by age the same size as the apical and
radial pieces.'"^ The same kind of complexity and irregularity as in Strotocrinus,
though not quite so well marked, is characteristic of the summit of Mariacrinus,
Periechocrinus, and Physetocrinus, and also of Marsiqyiocrinus, which is placed next to
Platycrinus by AVachsmuth and Springer. In Mariacrinus " the vault is composed of
very minute irregular pieces without definite arrangement, even the apical dome plates
are obscure."* In Physetocrinus only the proximal dome plates are distinguishable ;

condition having been found by liim in the Burlington limestone. He %vas kind enough to send me one of these for
examination ; and it was only after seeing it that I was struck by Miiller's figures, to which I had previously not paid
very much attention. I have since examined several similar specimens from the Bolland limestone.

1 Revision, part ii. p. 99. ^ IhicL, part ii. p. 159.

s Amer. Journ. Sci. and Arts, vol. xiv. p. 188. * Revision, partii. p. 116.

172 THE VOYAGE OF H.M.S. CHALLENGEE.

while in Periechocrinus the vault is composed " of small irregularly arranged smooth
pieces, among which the apical dome plates are indistinctly represented."^

All these genera, therefore, resemble Strotocrinus, and from the evidence of that genus
we are entitled to believe that the apical dome plates would be relatively more prominent
at earlier stages of development. Now this is exactly the case with the orals of recent
Crinoids, which at first surround the whole of the upper surface of the larval body ; but
eventually may be altogether resorbed or gradually reduced in comparison with their
fellows, the basals, as in Rhizocrinus (PI. X. figs. 2, 3, 7).

Thus then I believe the oral or actinal system forming the vault of Actinocrinus to
have been developed on the left larval antimer, in exactly the same way as the apical or
abactinal system is developed on the right ; but the oral system, instead of being limited
to five oral plates as in Neocrinoids, reached a very extensive development, so that in its
completest form it presents such a parallel to the apical or abactinal system as is to be
met with in no other Crinoid, much less in any other Echinoderm.

The greater variability in the development of the proximal plates, and theii-
occasional separation by smaller intercalated pieces, resembles the extreme irregularity
of the apical system of an Ophiurid, as compared with that of a Crinoid or Urchin.^ The
interradial plates of the former (or basals) have important relations to the chambered
organ. Those of the Urchin (genitals) are often connected with the genital ducts ; and
the radial plates in the same way have important functions in both groups. In the disk
of an Ophiurid, however, neither interradial nor radial plates are of any functional
importance ; and we find accordingly that their state of development difi'ers very
considerably even in allied species. Much the same is the case in the dome of the
ActinocrinidaB, where the plates are not in any way specially related to internal organs,
though serving to protect them.

Wachsmuth totally denies that there is any homology between the solid vault of a
Palseocrinoid and the ventral or oral disk of a recent form. So far as the Actinocrinidse
are concerned, I entirely concur in this opinion, except as regards those Neocrinoids,
such as Hyocrinus, Thaumatocrinus, and Holopus, which have persistent orals. For I
believe that representatives of these plates exist in the vault of all PalEeocrinoids,
whether simple or complex, although they are sometimes very greatly reduced ; and I
cannot therefore agree Ydth. ZitteP in considering them as entirely absent in the vault of
Actinocrinidse, Platycrinida^, Melocrinida3, Rhodocrinid^, &c.

The Cyathocrinidge and the Blastoids seem to resemble Strotocrinus and Periecho-
crmus in the small size and want of definite arrangement of the apical dome plates.
Neither in Wachsmuth's famous specimen of Cyathocrinus malvaceus, nor in any of the
Cyathocrinidae figured in Angelin's Iconographia can any definite arrangement of the

1 Revision, part ii. p. 131. ^ Quart. Journ. Micr. Sci., 18S4, vol. xxiv., N. S., pp. 4-14.

^ Palseontologie, p. 331.

REPORT ON THE CRINOIDEA. 173

ummit plates be made out. One of the figures^ of the "perisoma ventrale " in Cyatlio-
crinus alutaceus shows a central plate with seven others round it ; while in the other,
three large plates meet in the centre of the vault, but no one of them could be considered
as an orocentral. The same may be said of the vault of Cyathocrinus IcBvis represented
in tab. xxvi. fig. 36.

As regards the Blastoids I have never been able to trace any definite grouping of
the summit plates, although I have examined a very large number of picked specimens
of Pentremites, Granatocrinus, Schizoblastus, and Orophocrinus.

In Eleacrinus, however, the case is difi"ereut. There are comparatively fewer plates
over the peristome, and they certainly often do have a definite grouping, five surrounding
a central one as was first described by Roemer.*

No more need be said about the Blastoids, as their ambulacra are in many ways
abnormal, though they have strong points of resemblance to those of the Cyathocrinidse.
This family is one of special interest, for Wachsmuth ^ says that in the structure of their
vault they " bear closer resemblance to the recent Crinoids than almost any other group,
and seem to hold an intermediate position between modern and Palaeozoic types. If the
alternating plates, covering the furrows, could be turned back at the vault by the
animal, as the Saumplatten of the arms, then the food-groove of these Crinoids was open
throughout, as in recent forms." It is possible therefore that although the mouth and
peristome were subtegminal, i.e., covered in by the apical dome plates, yet the food-
grooves of the body may have been just as much external as those of the arms, and in no
way difi"erent from those on the disk of a Pentacrinus. As regards the Cyathocrinidfe,
therefore, one of the characters on w^hich Wachsmuth relies as separating the
Palseocrinoids from the Neocrinoids would then have no existence, i.e., the absence of
external food-grooves. I am not prepared to assert, however, nor indeed is Wachsmuth,
that these alternating plates in the radial areas of the vault of Cyathocrinus were
movable, like the covering plates of the disk in recent Crinoids. For it seems to me
quite possible that the closure of the peristome may have been continued outwards on to
the very short calyx-ambulacra, which would then first become open to the exterior at
the bases of the arms. But I have no question as to the homology between the radial
vault pieces in Cyathocrinus, and the covering plates of recent Crinoids, each set passing
continuously into the covering plates of the brachial ambulacra.

According to Wachsmuth's descriptions of the vault of the Actinocrinidse the arrange-
ment of the radial dome plates is the same as that of the radial calyx plates ; and he is
obliged to admit* that the alternating radial dome plates which he finds in Cyathocrinus
are " not so readily distinguished as in the PlatycrinidjB and forms with free rays, in

1 Op. cit, tab. xxiii. figs. 10 J, 11.

2 Monographie derfossilen Crinoideenfamile der Blastoideen undder Gattung Pentatreinites im besondern, Archiv
f. Naturgesch., Jahrg. xvii., Bd. i. pp. 377, 378, Taf. v. figs. \h, \c.

3 Amer. Journ. Sci. and Arts, vol. xiv. p. 184. * Revision, part ii. p. 30.

174 THE VOYAGE OF H.M.S. CHALLENGER.

wMcli they are well marked in tke extended parts. In the recent Crinoids the alternate
plates are represented by the ' Saumj)lattchen,' which, however, instead of forming
a part of a solid vault, are movable, and line the lateral margins of the tentacle furrows."

Now these " Saumpliittchen " or covering plates of recent Crinoids are imbedded in
the ventral perisome, and Wachsmuth admits that this is represented in the Actino-
crinid^ by the interpalmar areas on the upper surface of his internal casts, and in the
calcareous network which lines the interior of the vault. He describes how the ambu-
lacra pass outwards from the peristome within the body, and communicate directly with
the arm-grooves.

I believe myself that the tubular skeleton beneath the vault, which has thus far been
observed only in the Actinocrinidse,^ represents the covering of the disk-ambulacra of the
recent Crinoids, passing at the arm-openings directly into the ambulacral skeleton of the
arms. The following passage^ seems to imply that Wachsmuth is of the same opinion : —
"It is now generally conceded that the tubular canals beneath the vault contain the same
organs which in modern Crinoids are exposed on the ventral disk, and like them embrace the
food passages and certain other vessels connected with the ambulacral system." If then the
tubular skeleton beneath the vault correspond to the covering plates of the disk in recent
Crinoids, how can these last be represented by the alternating plates in the dome of the
Actinocriuidse, which, as Wachsmuth himself admits, are not readily distinguishable ?

Another difficulty also presents itself in connection with Wachsmuth's views respect-
ing these alternating plates of the ambulacra. Those on the calyx he considers as vault
pieces ; those on the arms as representing the covering plates of recent Crinoids. But in
another place he tries to prove that they are rudimentary pinnules, a question which has
been already discussed.^

Now it is obvious that the plates covering one end of an ambulacrum cannot be vault
pieces, while those at the other end are covering plates or rudimentary pinnules — they
cannot be both. It appears to me tolerably certain that the whole series of regular
alternating plates, calicular and brachial, represent the covering plates on the disk and
arms of recent Crinoids; but I will not venture to assert that they were invariably
movable on the vault and free rays, so as to expose the food-grooves to the exterior.

The condition of Gissocrinus seems to me to confirm this view very strongly, and
also to emphasise the difference between Cyathocrinus and Coccocrinus, to which I have
alluded above.

The composition of the calyx in Gissocrinus is the same as in Cyatlwcrinus, but the
lowest arm-joints instead of resting on the outer edges of the radials, lie upon their
ventral surface, and extend downwards towards the peristome over the sutures between

1 Mr Waclismvith tells me that he Las lately found " tubular canals beneath the vault " both in Platijcrinus and in
some of the Rhodocrinidce (August, 1884).

^ Revision, part ii. p. 29. ^ Ante, pp. 61-66.

EEPORT ON THE CEINOIDEA. 175

the interradials. Wachsmuth^ expresses this by saying that " the arms are reciuabent
upon the vault. There are five oral {cf. interradial) plates, upon the sutures between
which, and raised above the general level, the arm -joints are imbedded, being covered by
small alternating plates like the free arms." Here then we have a further development
of the abnormal condition presented by Cyatliocrinus and the Blastoids. For not only
ambulacral plates, but arm-joints themselves, extend over the sutures between the inter-
radials towards the opening at the centre of the summit. The two rows of alternating
plates which cover in the furrows clearly represent the plates arching over the grooves
between the interradials of Cyathocrinus ; l^ut it is equally clear that they are the cover-
ing plates of ambulacra which are borne by the arm-joints. This is very evident in some
of Angelin's figures" of the brachial ambulacra, which may be advantageously compared
with those of the ambulacra in the Comatulidee and Pentacrinidse (PI. XIII. fig. 16 ;
PI. XVII. figs. 2, 6, 7 ; PI. XXVII. figs. 4, 5, 11, 12 ; PL XLVII. figs. 10-13 ; PL LIV.
figs. 4, 7, 8 ; PL LV. figs. 3-7).

The two rows of alternating plates in the dome of the Platycrinidse have a
close resemblance to those on the vault of Cyathocrinus, and I have a strong suspicion
that they are of the same character, and not radial dome plates homologous wth the
calyx radials, like those in the Actinocrinidas. Wachsmuth, appears to have been in
much doubt about their nature, and to have had considerable difiiculty in making up his
mind. For he Jias described them in very difierent terms at different times. The
following general description was written by him as applying to both Platycrinidae and
Actinocrinidse, as well as to the Ehodocrinidse. Speaking of the radial dome plates,^ he
says "as a general rule, the summit plates increase in proportion to the number of
primary arms of a species in the same manner and on the same principle as the plates of
the dorsal side. Every radial from the third radial upward has a corresponding plate on
the ventral side, and additional interbraehial plates between corresponding brachial
plates above the arms."

This description, although true of most of the Actinocrinidee, does not appear to hold
good for any typical Platycrinoid, as far as can be judged from Wachsmuth's accounts of
the vault structure in the diff'erent genera of the fiimily. The vault of Coccocrinus has
been sufficiently discussed already. Tliat of Coixlylocrinus is not known. The radial
dome plates of Culicocrinus are as yet unknowTi, and but little room is left for them, as
the apical dome plates occupy the greater part of the summit. In the next genus,
Marsupiocrinus, however, the condition of the vault is entirely diff'erent. Wachsmuth
and Springer* describe it as follows: "vault low, hemispherical, composed of a larger
number of plates than usually found in this family. These are generally formed into

• Revision, part i. p. 91. - Op. cit., tab. xxvii. figs, le-lg ; tab. xxix. figs. 75rf, 76a.

3 Artier. Joiirn. Sci. and Arts, vol. xiv. p. 187. This passage appears again witb a slight alteration in the Revision,
part ii. p. 15.

* Revision, part ii. p. 64.

176 THE VOYAGE OF H.M.S. CHALLENGER.

narrow ridges, which bifurcate fwice within the body.^ Interradial dome plates larger than
the radial, apical plates not prominent and identified with difiiculty, interpalmar spaces
paved with small pieces." Figures of the vault of Marsupiocrinus radiatus and
Marsupiocrinus depressus are given by Angelin.^ The former shows a few larger plates
in the centre which may be the apical dome plates ; but in the other figure none of the
jjlates in the centre are specially large ; so that it is possible that Marsupiocrinus may
resemble many recent Crinoids in the total resorption of the orals, causing the proximal
ends of the interpalmar areas to be thickly studded with plates which tend to obscure
the position of the mouth (PI. XVII. fig. 6 ; PL LIV. fig. 10 ; PI. LV. figs. 4, 5, 7).
From the numerous bifurcating ridges formed by the radial dome plates, I cannot help
suspecting that these plates are not true vault pieces as in the Actinocrinidse, but the
covering plates of closed ambulacral tunnels. They have a very diflerent arrangement
from the various series of radials on the abactinal side, which should not be the case if
they belong to the vault, i.e., to the oral system. It will be remembered that
Wachsmuth has compared them to covering plates, while regarding them as true vault
pieces ; and he speaks of the interradial areas between them as " interpalmar," a term
which is inapplicable to true vault pieces, though I think he has used it correctly in the
case of Marsupioc7'inus. For I have a very strong impression that the so-called vault of
this genus is really the strongly plated ventral perisome, in the centre of which the
remains of the orals (apical dome plates) are perhaps to be found. I cannot see any
such essential diff'erence between it and the plated disk of Pentacrinus ivyville-thomsoni
or of many Antedons (PL XVII. fig. 6 ; PL LV.) as would lead to the supposition that
the homologue of the latter is to be sought for beneath the vault of Marsupiocrinus. At
the same time I have no intention of asserting the presence of an external mouth and
open food-grooves on the calyx of this genus. For although these are present in
the apparently similar disks of the recent forms, I think it not impossible that the
tentacular vestibule over the peristome of Marsupiocrinus may never have opened to the
exterior, and that the covering plates of the food-grooves proceeding from it may have
been immovably closed down over them. They were thus converted into tunnels, but
were still " external," in the sense of not being covered by a " tegmen," as those were which
formed the tubular skeleton beneath the vault of the Actiuocrinidte. In the recent Hyo-
crinus, which has many Palneocrinoid afiinities, the food-grooves pass from the oral to
the ambulacral system in the body before they reach the arms (PL VI. figs. 1-4) ; and
I see no reason why they should not have done the same in some of the Platycrinidse,
the family which is supposed by Wachsmuth to represent an incompletely developed
condition of the Actinocrinidse.

In Platycrinus, Ilexacrinus, and Talaivcrinus the structure of the vault is

1 They bifurcate considerably more than " twice " in Marsupiocrinus radiatus.

2 Op. cii., tab. x. tigs. IG, 21.

REPOET ON THE CRINOIDEA. 177

very different from that of Marsiqjiocrinus, and is more like that of the Actinocrinoid ;
but the apical dome plates (orocentral and orals) take up the greater part of the summit,
so that the radial dome plates are but little developed, and do not follow the arrangement
of the calyx plates, as described by Wachsmuth in Strotocrinus and other Actino-
crinidse. Those of Platycrinus extend on to the free rays as a double row of alternating
plates. Wachsmuth says nothing about the covering of the arms ; but there are two
rows of alternating plates on the pinnules, and these are obviously covering plates. It
is difficult to believe that the food-grooves of the arms were unprotected, when those of
the pinnules borne by them were bordered by covering plates ; and if these were present
on the arms, what was their relation to the two alternating rows of radial dome plates
upon the free rays ? Should not the alternating plates of the free rays, arms, and
pinnules be considered as parts of one system, just as the small covering plates of the
pinnules of Pentacrinidae are traceable into the larger ones on the arms, and through
them into those of the disk (PI. XVII. fig. 7 ; PL XXXIII. figs. 3, 4, 6 ; PI. XLI.
figs. 4, 13 ; PL LIV. fig. 7) ?

The free rays of Platycrinus find a parallel in many recent Crinoids. In Platycrinus
hurlingtonensis the first division borne on the radial axillary consists of two joints only,
of which the second is axillary. " One face of this bears an arm directly, while .the other
supports two pieces in direct succession, the second of which is an axillary piece
and gives origin to two arms." ^ This arrano-ement would be described in a Pentacritms
or Comatula as consisting of two distichal and two palmar joints, the latter only
occurring on the inner pair of every two secondary arms ; so that the arm formula
would be 1, 2, 2, 1, instead of 2, 1, 1, 2 as in Pentacrinus alternicirrus, &c.
(PL XXV.). Many Comatulidaj and Pentacrinidse have more than two joints in
the distichal and palmar series, and these do not belong to the arms proper,
but support radiating extensions of the disk, which is often much incised and
heavily plated (PL L. fig. 2 ; PL LV. figs. 3-7). In the dry state the plates
remain attached to the distichal and palmar joints, and cover them in by a kind of vault
(PL XXXIII. fig. 6), just as the radial dome plates cover in the free rays of Platycrinus,
though on a much smaller scale. As in the case of Cyathocrinus and Marsupiocrinus,
therefore, I believe that these alternating radial dome plates of Platycrinus are really
covering plates of the amliulacra, though permanently closed down, and terminating
against the primary radial plates of the dome outside the ring of proximals.

These primary radials are sometimes well developed, as in Talarocrinus and
Pterotocrinus. In the former^ " the first radial vault-piece is spiniferous in most species,
the succeeding plates small and nodose, arranged longitudinally in rows, forming together
regular arches over the ambulacral passages within the body." The vault of Pterotocrinus
seems to have had a closer resemblance to that of Actinocrinus than is the case in most

1 Pali-eontology of Illinois, vol. v. p. 453, pi. iii. fig. 6. ^ Bevision, part ii. p. 86.

(ZOOL. CHALL. EXP. PART XXXII. — 1884.) K 23

178 THE VOYAGE OF H.M.S. CHALLENGER.

Platycrinidse, for it had radial dome plates of the first, second, and even occasionally of
the third order.

Apart from this aberrant type, however, the radial regions in the vault of the
PlatycrinidfB seem to have consisted of a double row of small, more or less alternating
plates. Their arrangement does not by any means correspond regularly to that of the
radial calyx plates, as would appear from Wachsmuth's generalised description of the
vault ^ in the Sphaeroidocrinidse, i.e., Platycrinidse, Actinocrinidaj, and Ehodocrinidse.
In fact the greater part of this description holds good for the Actinocrinidae only.

The peripheral portion of the vault of Platycrinus, i.e., the zone betweenthe proximal
dome plates in the centre and the calyx interradials, is comparatively small ; and its
interradial spaces are " occupied by three — rarely five — plates, smaller than the central
dome plates, and less nodose, but yet comparatively large, and resting upon the inter-
radial of the calyx. "^ This series of four or six interradials, taken all together, doubtless
corresponds generally to the single large interradial of Cyathocrinus, as was supposed by
Wachsmuth when he considered the latter as an oral. I do not mean that the one plate
is homologous to the larger number ; but only that they all belong to the same system
of interradial plates. The position of the alternating dome plates in Cyathocrinus and
Platycrinus would then be very much the same. They rest in the one case between, and
in the other upon the interradials, and terminate against the apical dome plates. Wachs-
muth says, for example, " in Platycrinus the interradial plates thus take exactly the
same position as the exposed parts of the oral plates in Cyathocrinus, while the covered
parts are unrepresented."* In this type too the calyx interradials enter into the com-
position of the summit, just as is the case in Cyathocrinus. Thus Wachsmuth says
that "the first interradial, which exceptionally in this group is placed almost within the
dome regions, is identical with the outer («.e., primary) interradial plate of Coccocrinus,"*
in which I entirely agree.

He further says, " the vault of the Platycrinidse difi'ers in several particulars from
that of the other Sphseroidocrinidse, and in these same characters it approaches the
Cyathocrinidse."^ I do not myself think that the vault of a Platycrinite was exactly of
the same nature as that of an Actiuocrinite, i.e., that it covered in the whole of the visceral
mass and the ambulacra on its upper surface. For if the alternating dome plates represent
the covering plates of recent Crinoids, as Wachsmuth suggests, then all the periphery of
the dome, outside the apical dome plates (orocentral and orals), must be the real ventral
surface of the body, and not a tegmen calycis as in Actinocrinus. Wachsmuth himself
admits that the alternating plates in the dome of Platycrinus, like those of Cyathocrinus,
are represented by the covering plates of recent Crinoids ; and also that no tubular
skeleton has been discovered beneath the vault of any Palseocrinoid except an Actiuocrinite.

• Amer. Journ. Sci. and Arts, vol. xiv. p. 187 ; and Revision, part ii. p. 15. ^ Revision, part ii. pp. 17, 30, 69.

3 Ibid., part ii. p. 30. ■• Hid., part ii. p. 18. Ibid., part ii. p. 16.

REPORT ON THE CRTNOIDEA. 179

If his first admission be correct, as I believe it to be, one would not of course expect to
find a subtegminal ambulacral skeleton in Platycrinus}

There is some point on the actinal side of every Crinoid where the food-grooves leave
the oral system covering up the peristome in which they originate, and are only closed
by the covering plates at their sides.

In the recent Hyocrinus this closure takes place on the disk, where the food-grooves
come out from under the oral pyramid (PL VI. figs. 1-4). In Symhathocrinus the oral
pyramid was closed by the orocentral plate, and not open as in Hyocrinus ; while the disk
must have been even smaller than in that type, as the orals rest directly on the radials,
so that the ambulacra of the arms which were protected by covering plates, commence
directly from the sides of the oral pp-amid almost as in T/iaumatocrimis (PL LVI. fig. 5).
The BoUand Platycrinus which was figured by Miiller is essentially in the same condition
as Symhathocrinus, the orals (apical dome plates) resting directly against the plates of
the abactinal side, in this case the calyx-interradials ; while the oral or actinal system
is increased by the development of the radial dome plates corresponding to those in the
calyx, which rest directly over the arm-openings and are followed by the ambulacral
plates of the free ra5^s.

If the orals of Thaumatocrinus formed a closed pyramid resting directly on the inter-
radials, as it must in earlier stages of growth ; and if this pyramid were further extended
at its base by the development of radial plates in the actinal system, then the ambulacra
would start from the periphery of these plates just as the alternating plates of the free
rays do in Platycrinus.

In other Platycrinidse the oral system seems to have been still larger, having
secondary and tertiary dome-radials ; but sooner or later it came into contact with the
alternating series of plates which I take to be the skeleton of closed ambulacra, that
perhaps only opened to the exterior at the origins of the arms from the free rays. There
was a membranous disk, the radial regions of which were traversed by the ciliated food-
grooves beneath the ambulacral skeleton above ; while the interpalmar regions supported
the interradial plates of the vault. Both the ambulacral skeleton and the interradial

^ By this I mean distinct covering plates such as are found beneath the radial dome plates of the Actinocrinidse.
There must, of course, have been a " tubular passage beneath the vault," the presence of which is indicated on the
natural casts from a cherty bed in the Upper Burlington limestone, which have been recently examined by Mr. Wachs-
muth. From what he has told me about these internal casts of Platycrinus I imagine that they show very much what
he has already described in similar siliceous casts of y4c<inom?M(«, viz., "elevated rounded ridges, almost like strings
overlying the surface " ; and his remarks upon these last {Amer. Journ. Sci. and Arts, vol. xiv. p. 120) seems to me to be
equally applicable to the Platycrintcs-casts. He says : " The position of the string- like ridges (in case they represent
passages as I can hardly doubt) is analogous with that of the open food-grooves of recent Crinoids." In Adinocrinus, how-
ever, he not only found this evidence of passages beneath the vault which lodged the food-grooves ; but he also discovered
in some specimens preserved in a different way, that these passages were protected by a distinct ambulacral skeleton, itself
below the radial dome-plates. I imagine that this subtegminal skeleton, which corresponds to the ambulacral skeleton
on the disk of Pentacrinus (PL XVII. fig. 6) does not exist in Platycrinus. For the ambulacral skeleton of this type
was largely developed and e.xternal, forming the " alternate plates of the dome " (August 1884).

180 THE VOYAGE OF H.M.S. CHALLENGER.

plates were more massive than in recent Crinoids, just as the former were in the Cyatho-
crinidse ; and I do not think therefore that there was anything within the vault of
Platycrinidae like the tubular skeleton and the network of anambulacral plates that
occur in Actinocrinus. The greater part of the above argument appears to me to be a
mere logical deduction from Wachsmuth's very suggestive remark^ that the alternate
plates in the dome of Platycrinidge are represented in the recent Crinoids " by the
' Saumpliittchen,' which, however, instead of forming a part of a solid vault, are movable,
and line the lateral margins of the tentacle furrows."

Although believing that the vault of a Platycrinoid corresponds collectively to the orals,
interradials, ambulacral, and anambulacral plates of Neocrinoids, I do not wish to assert
that the Platycrinidse either had an external mouth or open ambulacra on the disk. For
I imagine that both were closed as in Cyatlwcrimts, the whole system of plates being
much more substantial than in Neocrinoids, and forming part of a solid covering, but not
a true vault or tegmen calycis.

In the Actinocrinidae, on the other hand, not only the food-grooves themselves, but
also their skeleton of alternating plates, were subtegminal, together of course, with the
plated interpalmar areas of the disk. The oral or actinal system of plates does not
consist merely of an orocentral with one or two rings of plates round it, which cover in
the peristome and the origins of the ambulacra from it. It is so greatly developed as to
cover in and conceal the whole ventral surface of the body, i.e., the disk proper. The
subtegminal food-grooves passed outwards from the peristome over this upper surface of
the disk, and were continued on to the arms through the ambulacral openings round the
dome. A primary dome-radial is always present beyond the orals, and may be followed
by secondary, tertiary plates, &c. Stelidiocrinus has very few dome-radials, but in other
types the number becomes very large, in correspondence with the development of
different orders of radials in the calyx. Sooner or later, however, the subtegminal food-
grooves reached the arm-openings, and the minute plates protecting them were continuous
with the skeleton of the brachial ambulacra.

There is one form which is placed by Wachsmuth and Springer among the Actino-
crinidaj, but has a vault of very different construction from that of the other members
of this family. In fact it somewhat resembles that of Marsupiocrvmis. I mean
Carpocrinns ornatus {Habrocrimis, Angelin). Wachsmuth describes its radial portions
as " covered by two rows of low transversed pieces ; interpalmar fields paved by somewhat
larger and elongate plates."" As pointed out above, the word " interpalmar " denotes
the areas between the radiating food-grooves ; and its use by Wachsmuth is therefore
significant. I cannot resist the suspicion that the double row of low transversed pieces
indicates the position of a food-groove ; and that the covering plates may have been
permanently closed down so as to convert the grooves into tunnels, without the additional

1 Revision, part ii. p. 30. ^ Ibid., part ii. p. 106.

REPORT 0:S THE CRINOIDEA. 181

help of a tegmeu, such as occurs in the Actinocrinidse. The proximal ends of these
tunnels would open into the closed peristomial space beneath the pyramid of apical dome
plates or orals. In fact, Carpocrinus appears to me to be in the condition of a Hyocviiius,
with an oral pyramid composed of somewhat smaller plates, but permanently closed, like
the ambulacra of the disk. While therefore I am in complete-accordance with Wachsmuth
respecting the closure of the peristome and calyx- ambulacra in the Cyathocrinidse,
Platycrinidge, and Actinocrinidae, I cannot altogether agree with him in denying all
homology between the solid vault of a Palseocrinoid and the soft disk of a recent form.
For I believe that both in the Cyathocrinidse and in the Platycrinidse the plates which
form the vault are unusually massive representatives of the ambulacral, anambulacral,
and interradial plates which are developed in the perisome of a Pentacrinus or
Comatula.

The Ichthyocrinidae and some of the doubtful Silurian forms, such as Reteocrirms and
Xenocrinus, appear to me to occupy an intermediate position between the heavily vaulted
Platycrinidse and the more thinly plated recent forms. Some of the Mesozoic species,
such as Extracrinus, Ajnocrinus, Guettardicrimis, and Marsupites, which have com-
paratively thick plates on the sides and surface of the disk, also help to fill up the gap.

The only genus of Ichthyocrinidas in which the summit is known at all satisfactorily
is Onychocrinus. Wachsmuth and Springer describe it as follows ■} — " Interradials three
to twenty, perhaps more in some species ; the first one large, resting between the first
and second radials ; the succeeding ones smaller, rapidly decreasing in size and thickness
upward, and having an inward curvature. They are followed by very minute, irregular
polygonal plates, which form the interradial portion of the vault. The radial summit
areas consist of two rows of somewhat larger plates, alternately arranged, which extend
to the ventral covering of the free rays, and probably throughout their fuU length. In
the median portion of the vault there are six rather thin but large apical dome plates."
I understand, however, from Mr. Wachsmuth that he is now less inclined to believe in
the presence of apical dome plates in the Ichthyocrinidse ; and I will not therefore take
their presence as established. If they exist I should call them oral plates, and compare
the vault to the disk of a Hyocrimis with a closed oral pyramid. But in their absence
the vault appears to me so closely to resemble the disk of Pentacrinus and Comatula,
that I cannot question the identity of the two for the merely a priori reason of the
Ichthyocrinidas being Palseocrinoids.

The two rows of alternating plates which radiate outwards over the "squamous
integument," and extend on to the free rays {i.e., distichal and palmar series), are surely
nothing more than the covering plates of the ambulacra, which were perhaps permanently
closed as in the Platycrinidae, or only temporarily so, as in the Neocrinoids ; while the
small irregular plates which form the interradial portion of the vault, correspond to the

• Revision, part i. pp. 53, 54.

182 THE VOYAGE OF H.M.S. CHALLENGER.

anambulacral plates of recent Crinoids (PI. XVII. fig. 6 ; PI. LV.). They pass down-
wards into the interradials at the sides of the calyx, just as in the recent species and in
the Liassic Extracrinus.

Wachsmuth stated in 1877^ that although he had not found the summit of any
Ichthyocrinoid jserfectly preserved, he felt convinced from what he had observed " that it
did not consist of a soft skin." Subsequently, however, he described the ventral disk of
the Ichthyocrinidaj " as " rarely presei'ved ; composed of a more or less soft or scaly
integument, yielding to motion in the body and arms. . . . The interradial areas are
sometimes found depressed and in other cases distended, showing that there had been
some expansion or contraction of the body-walls due to the moliility of the radial parts,
and indicating likewise flexibility in the vault."

Under these circumstances I find it difiicult to believe that the ventral disk of the
Ichthyocrinidse did not correspond to the similarly named structure in recent Crinoids,
but represents the solid vault of Actinocrinus. Were this the case, there must have been
another flexible skin inside the " pliant scaly integument," with the food-grooves passing
over its upper surface as they do over that of an internal cast of Actinocrinus. It is of
course impossible that a proof of the existence of such a structure can ever be obtained.
But why should its existence be postulated at all, simply because Ichthyocrinus is a
Palseocrinoid ?

According to Wachsmuth and Springer " this fjimily might very appropriately be
called the Articulates of the Palaeozoic Crinoids, being especially distinguished in most of
its species by what seems to be an articulate structure in the whole skeleton.'" I cannot
but believe that they present a similar approximation to Neocrinoids in the structure of
their vault, ventral disk, or whatever else it be called.

Any Crinoid with a well-plated disk (PI. XIII. fig. 1 ; PI. XVII. fig. 6 ; PI. XXVI.
figs. 1, 2 ; PI. L. figs. 1, 2 ; PL LV.) appears to me to be a recent Onychocrinus. If the
summit of this genus was soft, pliant, and flexible, it must have consisted like the ventral
disk of a Pentacrinus of a perisome formed of connective tissue, with the numerous
irregular interradial plates imbedded in it ; and I cannot bring myself to believe that the
flexible summit was really the " tegmen " overlying another disk, which itself represented
the plated ventral perisome of Pentacrinus. It is admittedly a direct continuation of
the " squamous integument " uniting the rays on the dorsal side. This is " composed of
very minute irregular polygonal plates, or by distinct interradial and axillary plates, the
former varying in number from one to thirty or more.*

Thus Onychocrinus has a large first interradial which rests between the first and
second radials. " The succeeding ones are smaller, decrease rapidly in size and thickness,
and pass gradually into the very minute irregular plates which form the interradial

' Amir. Jmim. Sci. and Arts, vol. xiv. p. 185 ^ Revision, part i. p. 31.

^ Ibid., part i. p. 31. * Ibid., part i. p. 31.

EEPORT ON THE CRINOIDEA. 183

portion of the vault." ^ I have seen a similar arrangement in well preserved specimens of
Marswpites, while both d'Orbigny ^ and de Loriol have figured and described the same
thing in Apiocrinus roissyanus. After stating that there is a considerable amount of
variation in the interradial areas, even in the same individual, de Loriol says — " Presque
toujours la serie commence par une piece unique, hexagone on
heptagone, qui est la plus grande, et se trouve encastree entre les ^j i^'

premieres et les secondes radiales, de chaque cote, reposant sur les '~\

troncatures des premieres radiales. Au-dessus il y a deux, trois, b, .ifPl

et meme quatre pieces plus petites, irregulieres, polygonales, qui - _ ';V

arrivent au niveau des facettes articulaires des troisiemes radiales, B;' ^ '' , Vl
une troisieme et une quatrieme rang^e comprennent encore chacune ^
trois ou quatre pieces polygonales plus petites, et occupent I'espace
entre les premiers articles brachiaux ; elles sont suivies par d'autres
rangees de pieces, plus petites encore, qui paraissent concourir a la

I, .- T, A, 1 •. T • 1 " 3 "n T • l> FiG- 9- — Interradial plates of

formation dune voute sur la cavite calicmale. De Loriol s Apiocrinus roissyanus

enlarged representation of this structure is reproduced in fig 9. first and second brachials-

, 1 „ Ti • Til Rl, R2, K3, first, second,

Here, surely, there was a "dome as solid as m any Ichthy- and third radiais; pi,

■^ _ . basal joints of lowest

ocrinoid ; but it will scarcely be contended that this dome pinnule; /, caiyx-inter-

•^ _ _ radial, resting on the

rem'esents the heavy rigid vault of Actinocrinus, rather than the "pp^^ angles of two first

-T ^ D ' radials (Rl) ; i, smaller

plated ventral perisome of recent Pentacrinidse and Comatulidse. interradiais, but probably

r J. only pensonuc plates.

The former range back to the earliest Mesozoic times, long anterior

to Apiocrinus ; and the Liassic Extraainus had a vault essentially similar to that of the
Apiocrinidse. But there was no regular calyx-interradial resting upon the upper angles
of the first radials. Its place was taken by a number of movable irregular perisomic
plates, lilie those which occur in the same position in Pentacrinus asterius (PI. XITL
fig. l). They are represented by Miller,* Austin, and also by Quenstedt;^ and were
continued upwards into those " which cover the dome-like integument over the abdominal
pouch,'' just exactly as is the ease with their fellows in recent forms.

The Silurian genera Glyptocrinus, Reteocrinus, and Xenocrinus appear to me to have
been in the same condition ; though I will not go so far as to say that the mouth was
open to the exterior. For the peristome may well have been closed by the more or less
well defined apical dome plates, which covered the central part of the disk, just as in the
Ichthyocrinidae. The vault of Glyptocrinus is best known in Glyptocrinus decadactylus.
According to Miller' "The regular interradial areas have one plate resting upon the
primary radials, two in the second range, three in the third, two or three in the fourth,

1 Revision, part i. p. 54. " Eist. Nat. des Crino'ides, p. 21, pi. ill. figs. 1, 3.

3 Paleont Frang., loc. cit., pp. 272, 273. * Op. cit, pp. 57, 59.

» Encriniden, Tab. 101, fig. 39a. * Austin, op. cit., p. 104, pi. xiii. figs, la, Ic, Ih

' Glyptocrinv^ redefined and restricted, Gaurocrinus, Pycnocrinus, and Compsocrinus establislied, and two new
species described, Joum. Cincinn. Soc. Nat. Hist., vol. vi. pp. 220, 221.

184 THE VOYAGE OF H.M.S. CHALLENGER.

and above these fifteen or twenty small plates in eacli depressed intertertiary area-
Intcrsecondary radial areas liave one rather large plate in each axil, and a dozen or more
smaller ones filling the depressions between the tertiaries. Intertertiary areas have in
like manner one plate in each axil, and several smaller ones above. . . . The vault is
somewhat convex in the central part, and undulates towards each intertertiary area. It
is composed of numerous polygonal plates. Those in the central part are the larger ones,
and each of these bears a central tubercle, which is sometimes prolonged so as to be
designated a spine toward the margin, or rather following the undulations toward the
intertertiary areas, the plates are smaller and possessed of slight convexity. They unite
in the depressions in the intertertiary areas with the plates of the calyx, or rather the
interprimary radials graduate through the intersecondaries and intertertiaries to the plates
of the vault without any line of separation. The plates become smaller as they approach
the inner face of the arms, over the swelling undulations of the vault, and continuing to
decrease in size, form a somewhat granular continuous integument, that covers the
ambulacral furrows."

Except as regards the larger central plates (orocentral and orals), this description
would apply equally well to the disk of many Pentacrinidfe and Comatulida^. The vault
of Glyptocrinus would appear to have been more or less flexible as in the Ichthyocrinidse ;
and the so-called " continuation of the vault up the inner side of the arms " seems to me
to be nothing but the extension on to the arms of the ambulacral skeleton, together
perhaps with some of the anambulacral plates at its sides, just as in Pentacrinus asterius,
Pentacrinus alternicirrus, Pentaco'inus naresianus, and Metacrinus murrayi (PI. XVII.
fig. 7 ; PI. XXVI. figs. 1, 2 ; PL XXVII. fig. 13 ; PL XXX. fig. 2 ; PL XLI. fig. 13).

The vault of Retocrinus nealli is thus described by Meek' " Interradial areas occupied
by numerous (70 to more than 100) small pieces of very irregular size and form, and
without any definite arrangement. . . . Axillary areas each occupied by about fifty to
sixty very small, irregularly arranged, unequal pieces. Vault composed of numerous
minute pieces, generally of hexagonal form ; highest on the anterior side, with a ridge
radiating to each arm-base, and a corresponding sulcus between ; opening minute,
penetrating a small tubercle situated behind the middle, and directed backward."

Wachsmuth says that the plates in the median part, which probably include the
apical plates, are somewhat larger than the rest ; while he further states that " the
peculiar depressed state of the interradial and interaxillary arese, the irregularity with
which their plates are arranged, suggests the possibility that they were adapted to
expansion by the animal."^ Here again, then, he admits the possibility of the "vault"
having been pliant and flexible. Thanks to his kindness, I have had the opportunity of
examining specimens of all three genera, Glyi^tocrinus, Reteocrinus, and Xenocrinus,
while Meek gives an excellent figure of the summit in Reteocrinus nealli.^ I am sorry,

1 Palieontology of 01ui:>, vol. i. p. 35. - Revision, part ii. p. 192. ^ Palreontologj of Ohio, vol. i., pi. ii. fig. 3c.

REPORT ON THE CRINOIDEA. 185

however, that I cannot agree with Wachsmuth respecting its nature. For, like that of
Ichthyocrmus, it appears to me to represent the plated ventral perisome of the
Neocrinoids ; while the ridges which radiate to the arm-bases seem to me to consist of the
covering plates of the ambulacra. I do not deny that they may have been closed down,
either temporarily as in the recent Pentacrinus ivyviUe-thomsoni (PI. XVII. fig. 6), or
permanently as in Platycrinus. But I cannot imagine that they represent parts of a solid
vault like that of Actinocrinus.

I would say the same of Xenocrinus, of which Miller^ speaks as follows : " Interradial and
intersecondary radial spaces. . . . These long, narrow, depressed areas are covered with
small plates, having a tubercle or short spine in the central part of each. There are
more than seventy-five plates in each interradial area, and twenty-five or more in each
intersecondary radial area before reaching the top of the cup, but the small plates
continue over the margin of the vault, and undoubtedly cover it, and also more or less of
the long proboscis, which is extended from the anterior or azygous side."

Wachsmuth denies that any Palseocrinoid is known in which the existence of a solid
vault has been disproved or cannot be traced by analogy ; and also that there can be
any homology between this solid vault and the ventral perisome (whether soft or plated)
of a Neocrinoid. He has since admitted, however, that the radial pieces in the vault of
Cyathocrintis and Platycrinus correspond to the ambulacral skeleton on the external
surface of the body of recent Crinoids ; and I venture to think that in the case of Glypto-
cnnus, Reteocrmus, and Xenocrinus, and also of the Ichthyocrinidse, the resemblance to
the Pentacrinidse, Apiocrinidse, and Comatulae is such as to leave no reasonable doubt that
the so-called vault of these Palseocrinoids is homologous with the ventral surface of the
body in the Neocrinoids. Except as regards Coccocrinus, however, I am not prepared to
deny that the mouth was subtegminal, i.e., concealed beneath the apical dome plates,
which I regard as representing a permanently closed oral pyramid. When the presence of
these plates has been demonstrated in Coccocrinus, I will admit that the " Scheitelstiicke "
which AUman, Wachsmuth, Zittel, and myself have all considered as orals, belong to the
interradial system, and do not surround an open mouth as the orals of Holojms,
Hyocrinus, and Thaumatocrinus do.

1 Journ. Cincinn. Soc. Nat. Hist, vol. iv. p. 72.

(ZOOL. CHALL. EXP. — PART X3LSII. — 1884.) li 24

186

THE VOYAGE OF H.M.S. CHALLENGEE.

XL— CLASSIFICATION.

The following is the classification of the Stalked Echinoderms which is adopted in
these Eeports : —

Phylum, ECHINODEEIilATA.

Branch, PELMATOZOA.

Class L Ck I NO IDE A.
Class 2. C Y S T I D E A.

Class 3. Blastoidea.

Pelmatozoa, Leuckart,
1848.

Synonymy and other Classifications.

Bracliiata.

Grinoidea, Burmeister,
18.56.

Cnwoiiea,Eoemer, 1856. -

Tentaculata, Kay-

Lankester, 1877.

Crinoidea, Zittel, 1880;
and de Loriol, 1882.

Antliodiata.

An order of the
EcMnodermata,

Three suborder.s.

A branch of the Phylum
Echinodermata,

Three classes.

A class of the Stamm
Echinodermata.

Three orders.

1. Crinoidea, sensn. str.
( 2. Cystidea.
( 3. Blastoidea.

1. Actinoidea =
Crinoidea, sensu. str.

2. Cystidea.

3. Blastoidea.

\ I. Crinoidea.
\ II. Blastoidea.
[ III. Cystidea.

1. Eucrinoidea =
Crinoidea, t-ensu.str.

2. Cystoidea.

3. Blastoidea.

PELMATOZOA.

Definition. — Echinoderms which are fixed either permanently or temporarily by the
middle of the aboral surface. A jointed stem containing a neuro-vascular axis is usually
present, butma}^ be lost when maturity is reached ; or, in the case of a few sessile forms,

REPORT ON THE CRINOIDEA. 187

remain altogetlier undeveloped. The apical system consists of a dors ocentral plate,
basals, and radials, with the frequent addition of under-basals and interradials. These
plates form a cup, which either simply supports, or more or less completely encloses the
visceral mass, and often bears jointed appendages, the arms and pinnules.

An oral system, consisting of a central plate (orocentral) and five orals is developed
above the peristome of the larva to a very variable extent, and may be either altogether
resorbed, or reach a high degree of importance by the appearance of additional plates so
as to form a vault or " tegmen calycis." The anus is situated on the oral surface, which
may be bare, or more or less covered by calcareous plates. The water-vascular ring does
not communicate directly with the exterior, and the lateral branches of the radial vessels
(when present) are respiratory, but not locomotor in function.

Class 1. Ckinoidea, Miller, 1821.

Crinoidea, Auctoriuu.

Stilasteritm, Goldfuss, 182G-1835.

Ast&rencrinidea, de Blainville, 1834.

Pinnigrada, Eorbes, 1841.

Pinnastella, Austin, 1842.

Brachiata, Burmeister, 1856.

Actinoidea, Roemer, 1856 (Date, of Preface).

Encrinidm, Quenstedt, 1876 (Date of Preface).

Euainoidea, Zittel, 1880 (Date of Preface).

Definition. — Pelmatozoa, in which the radial plates of the calyx bear more or less
branching arms. These consist of segments which are articulated by means of muscles
and ligaments, and in most cases bear similar jointed appendages, the pinnules. The
nervous system consists (1) of a central organ situated in the calyx, and fibres
extending from it through the skeleton of the stem, arms, and pinnules ; (2) of a
circum-oral ring and radial extensions which are in close relation with the ciliated
epithelium of the ambulacral grooves. These are more or less extensively distributed
on the ventral surface of the disk, arms, and pinnules ; and are bordered by groups of
tentacles which alternate on opposite sides. When they are absent, the radial water-
vessels give ofi" no tentacular branches. The water- vascular ring opens by five or more
water-tubes into the body-cavity, which itself communicates with the exterior by a
(corresponding number of water-pores. The mouth is central, except in one genus, and
the anus subcentral or excentric. The genital glands are lodged in the lower parts of
the arms, but are usually fertile only in the pinnules.

Remarks. — Various writers have attempted at difli"erent times to separate the Crinoids
and their allies from the remaining classes of the Echinoderms by somewhat more definite
characters than those which distinguish these various classes intei- se. Very little was

188 THE VOYAGE OF H.M.S. CHALLENGER.

known, Iiowevci', respecting the anatomical relations of the soft parts of a Crinoid ;
while the morphology of the Echinoderms generally was far from being properly
understood ; and in most cases the gronp w\as regarded as consisting of four or five
divisions of equal value, which were considered as classes or orders according to the
position assigned to the Echinoderms generally in the animal kingdom. The Ophiurids
and Asterids have sometimes been united into one group, the Stellerids, which, though
convenient for purposes of reference, appears to me to be somewhat too comprehensive.
Considering the totally dissimilar modes of development, and the great difference
in anatomical structure between a Brittle-star and a Starfish, I find it difficult to do
otherwise than regard the Ophiuroidea and Asteroidea as independent classes of
Echinoderms, of equal systematic value with Echinoidea and Holothuroidea.

In like manner, I should rank the Blastoidea and Cystidea as classes of the Echino-
derms rather than as sub-classes or orders of the Crinoidea. The essentia] difference
between them and the true Crinoids (Eucrinoidea of Zittel and de Loriol) is the
presence in the latter of branching articulated arms developed above the radial plates of
the calyx ; while the symmetrical composition of the calyx in the Blastoids, the
complexity of their ambulacra, and the regular arrangement of their hydrospires, sharply
distinguish them from most of the Cystids, the morphological characters of which are, as
it were, more plastic and less crystalline.

These three groups, however, the true Crinoids, the Blastoids, and the Cystids, are
distinguished from the remainder of the Echinoderms by certain very definite peculi-
arities, i.e., the more or less permanent attachment of the aboral surface of the body,
and the absence of any locomotor organs in connection with the ambulacral system.
For the lateral branches of the water- vessels, when present, are simple respiratory
tentacles of an altogether difi"erent nature from the tube-feet of an Echinid or
Stellerid.

Important as these differences are, few zoologists seem to have recognised them as of
any greater systematic value than those between an Urchin and a Starfish, Crinoidea,
Asteroidea, and Echinoidea having been generally regarded as equivalent divisions of
Echinoderms. As long ago as 1848, however, Leuckart separated the stalked
Echinoderms from the remainder of the group under the name " Pelmatozoa." This
suggestion seems to have been adopted by Bronn in 1860,^ while Leuckart classed the
Echinoderms in accordance with it in his annual "Berichte ;" but little notice was taken
of it in this country except by Sir Wyville Thomson, as will be seen further on. In
1869 Prof. Huxley^ recognised the same principle when he wrote "The Crinoids are so
difi'erent from the other living Echinodermata that they will probably have to form
a distinct primary division or sub- class of the class ; and this may possibly be the

1 Die Klassen nnd Ordnungen des Thier-Reichs, Bd. ii., Aktinozoen, 1860, pp. 3, 421.

2 An Introduction to the Classification of Animals, London, 1869, p. 130.

REPORT ON THE CRINOIDEA. 189

case with some of the following extiuct forms," viz., Cystidea, Edriasterida, and
Blastoidea.

The difference in the functions of the water- vascular system between the stalked and
the unstalked Echinoderms respectively was applied by Prof. Ray Lankester for
systematic purposes in his division of the Echinoderms into AmhulacraUa and Tenta-
culata, the latter including Crinoids, Cystids, and Blastoids.^

The name " Tentaculata " is unfortunately open to the objection that even in recent
Crinoids some of the radial water- vessels may be totally unprovided with tentacles at theii-
sides ; while if, as I believe, the water- vessels of the Blastoids occupied the subambu-
lacral canals within the lancet-pieces,^ they must certainly have been non-tentaculate.
To one division (class) of the group, therefore, the name " Tentaculata " would not be at
all applicable. Neither do I like the extension of the term " Crinoidea" to the Blastoids
and Cystids, and the consequent limitation of the brachiate forms by the name
Eucrinoidea, which we owe to Zittel so far as I have been able to trace it ; though it has
recently been adopted by de Loriol.

In Miller's original definition of the Crinoidea^ he described them as having "a cup-
like body containing the viscera, from whose upper rim proceed five articulated arms,
divided into tentaculated fingers more or less numerous." The presence of these arms is
essential to the idea of a " lily-shaped animal." The very characteristic appearance of
the Crinoid typ)e is lost if the arms be not attached to the calyx ; while morphologically
they are of the utmost importance.

On the other hand, jointed appendages of this kind, attached to the rim of the cup,
and containing radial extensions of the nervous axis of the stem, as well as of all the
ambulacral structures which surround the peristome, together with the genital glands, are
entirely absent both in the Blastoids and in the Cystids. In the former group, it is
true, there were jointed appendages at the sides of the ambulacra ; but although the
latter are very often spoken of as "recumbent arms," they are not composed of articulated
pieces, and only a very general homology can be traced between them and the branching
arms of a Crinoid. In the Cystids, however, segmented arms somewhat like those of
Crinoids seem to have been occasionally present, and even grooved by the ambulacra.
But they were mostly attached somewhat irregularly in the neighbourhood of the mouth,
and not to the radial portions of the cup as in the Crinoids ; and I much doubt whether
their component segments were regularly articulated together.

Neither the Blastoids nor the Cystids, therefore, can properly be classed as Crinoidea,
in the sense of Miller's definition of the group ; though this has been very frequently
done during the last forty years, more especially hj continental naturalists. Von

1 Notes on Embryology and Classification, Quart. Journ. Micr. Sci, 1877, vol. xvii., N. S., p. 444.

2 On Certain Points in the Morphology of the Blastoidea, Ann. and Mag. Nat. Hist., 1881, ser. 5, vol. viii., p. 420;
Ibid., 1882, vol. ix. p. 218.

5 Op. cit., p. 7.

190 THE VOYAGE OF H.M.S. CHALLENGER.

Buch, however, clearl)^ distinguished the essential differences between Crinoids and Cystids.^
The same was the case with Edward Forbes, who having given the Crinoids ordinal rank
in 1841^ under the remarkable name " Pinnigrada," assigned the same position to the
Cystids in 1848.' Von Buch seems to have considered the Blastoids as a third group of
equal value with these two. Roemer, on the other hand, degraded the Cystidea and
Blastoidea to the level of families or sections of the order Crinoidea, separating off the
brachiate forms of the latter as true Crinoidea.* A few years later he proposed to call
these by the name " Actinoidea," and to rank them together with Blastoids and Cystids
as suborders of the Crinoidea.'' This term was thus employed, not in the strict sense of
Miller's original definition, Ijut as co-extensive with the name " Pehnatozoa," which had
been proposed Ijy Leuckart four or five years previously ; though Roemer appears to
have been unacquainted with it. This was unfortunate, as the use of Leuckart's excellent
name in the Lethaea Geognostica would have avoided much subsequent confusion.

In the second volume of Bronn's "Thier-Eeich " the Echinoderms aret hrowoi tooether
with the Coelenterates into the comprehensive " Kreis " of Strahlenthiere or Aktinozoa.
Four classes of Coelenterates are first considered, and then the Blastoidea and Crinoidea,
for which the cumbersome names " Blastactinota " and " Crinactinota " are proposed.
Fortunately, however, they have not come into general use. The Cystids are thrown
back among the Crinoids, for Bronn did not consider them as difi"ering from the brachiate
Crinoids to the same degree as the Blastoids. This was altogether in opposition to the
views of Von Buch and Edward Forbes, and also to those of Roemer," to whom the
peculiarities of the Blastoids and Cystids appeared so marked, " dass sie als gleichwerthige,
wenn auch nicht gleich umfangreiche Scctionen oder Unterordnuugen den iichten
Crinoiden entgegen zu setzeu sind." Viewed by the light of later knowledge, Bronn's
classification was of a distinctly retrogressive nature.

Besides the Cystids he recognised two other divisions of the Crinoidea, viz., the
Brachiata or the Crinoidea proper, and the Costata, Miiller, the latter including the
problematical Saccosoma.

The terminology employed by Bronn for the diflerent groups of the stalked Echino-
derms is extremely difficult to understand, and appears to contain many errors. Thus
on pp. 193 and 421 {op. cit.), the name "Actinoidea" for the true Crinoids is attributed
to Miiller, though it is really Roemer's, as explained above ; while on pp. 207 and 210 the
true Crinoids are referred to as " Anthodiata," in contradistinction to the other division

1 Ueber Cystideen, Ahhandl. d. k. Akad. d. IViss. Berlin, 1845, pp. 12, 13, 17, 27.

^ A History of British Starfishes and other Animals of the class Ecliinodei-mata, London, 1841, p. xiv.

3 On the Cystidese of the Silurian Rocks of the British Islands, Mem. of ilw Geological Survey of Great Britain, and
of the Mmeam of Practical Geology, 1848, vol. ii. part 2, pp. 526, 527.

* Monographie der fossilen Crinoiden-familie der Blastoideen, und der Gattung Pentatrematites im Besondern,
Archiv f. Naturgesch., Jahrg. xvii., Band i. pp. 387, 388.

■■' Lethaea Geognostica, Bd. i., Theil 2, p. 224.

" Op. cit., pp. 180, 193. ^ Lethaea Geognostica, Bd. i. Theil 2, p. 224.

EEPORT ON THE CRINOIDEA. 191

of the " Crinactinota," viz., the Cystids. On the other hand, according to the scheme
on p. 421, " Anthodiata" was the name proposed by Burmeister for the Blastoids, while
the term "Brachiata" was also his, and included the true Crinoids and the Cystids. On
pp. 3, 227, and 230, however, the term " Brachiata" is used by Bronn to denote the true
Crinoids only, and it is attributed to Miiller. Zittel has followed Bronn in this respect,
and, as I believe, erroneously. For I have searched Mliller's writings on Crinoids
repeatedly without finding this expression, though frequent reference is made to the
" Crinoidea Tessellata mit Arm en."

After various unsuccessful attempts to discover where Burmeister's nomenclature was
published, I applied to Prof. F. J. Bell, who was kind enough to make a search in the
library of the Zoological Department at the Museum of Natural History, with the
following result. In his Zoonomische Briefe, published at Leipzig in 1856 (vol. i.
p. 243), Burmeister gives the following " Systematische Uebersicht der Crinoideen."

I. Crinoidea anthodiata.

1. Cystideen. 2. Blastoideen.

IT. Crinoidea hruchiata.

3. Tessellaten. 4. Articulaten. 5. Gesippten {Crinoidea costata). 6.
Holopus.

This classification of Burmeister's deserved more attention than it has hitherto
received ; for it was the first which clearly brought out the difference between the true
Crinoids with segmented arms attached to the radials and the " Anthodiata " or Blastoids
and Cystids, in which there are either no arms at all or structures of an entirely
different nature from those of the true Crinoids. In this, as in other respects, the
Blastoids and Cystids at once difi"er from the Crinoids and resemble each other. In fact
they are so closely linked together that it is extremely difficult to refer forms like
Hybocystites and Cystohlastus to one group rather than to the other.^

The term Crinoidea should, I think, be limited to the strictly brachiate forms for
which it was proposed by Miller ; and it is much less applicable to the stalked Echino-
derms generally than Leuckart's name " Pelmatozoa." But except as regards this
question of nomenclature Burmeister's classification agrees far better with our present
knowledge than many of those published before or since his time, e.g., that of d'Orbigny,
Pictet, or of Dujardin and Hupe.

Low as the Cystids had fallen in Broun's classification from the ordinal position to

1 Quenstedt has solved the difficulty respecting the systematic position of Cystohlastus by describing it twice over.
On p. 684 of his " Encriniden " it appears among the Cystids, and is figured on Tab. 113, fig. 89 ; but on p. 724 it is
described as a Blastoid, and it is figured on Tab. 114, fig. 98, under the name of Cijclohlasttis.

192 THE VOYAGE OF H.M.S. CHALLENGER.

which they had been raised by Forbes, they had been, and were subsequently, still further
degraded. For d'Orbigny^ took an entii'ely different view of the characters of the
various types of the Pelmatozoa from those held by some of his predecessors ; and he not
only threw the Cystids and Blastoids back among the Crinoids, but he considered these
two groups merely as families. He divided the order Crinoidea into twelve families,
among which are the Comatulidse, Pentremitidae, Cystidse, and lastly the Pentacrinidae ;
and Pictet ^ subsequently reduced this number to nine, but without making any change
in the four above mentioned.

Dujardin and Hupe '' also adopted this singular arrangement, according to which the
differences between a Pentacrinus and a Pentremites, Pchinosjyhcentes or Actinocrmus,
are of no greater systematic value than those between Pentacrinus and Comatida. In
this country, however, thanks mainly to the teaching of Prof. Huxley,* Crinoids, Cystids,
and Blastoids have always been regarded as independent but equivalent divisions,
formerly orders, but now' classes of the Echinodermata. To these Huxley * has since
added another, as to the necessity for which there has been a considerable difference of
opinion, viz., the Edriasterida.

This group, which includes the curious sessile forms Agelacrinus, Edrioaster, and
their allies, has been generally placed among the Cystids ; but it has been re-established
quite lately under the name of Agelacrinoidea by S. A. Miller, in ignorance of Prof.
Huxley's classification of fifteen years ago.

I am inclined to think myself that if these forms be anything more than the
isolated disks of Palseocrinoids, as was thought possible by Sir Wy\^ille Thomson [ante,
p. 85), theii- proper place is among the Cystids.

Two other new orders (i.e., classes) of the class [i.e., subkingdom) Echinodermata have
recently been proposed by S. A. Miller.* These are the Lichenocrinoidea and the
]\Iyelodactyloidea. But I cannot regard them as of equal value with the Crinoids, Cystids,
and Blastoids. Our knowledge of the structm'e of Lichenocrinus is of the most limited
character ; and it is therefore totally insufficient for the basis of a class definition. The
same may be said of Cyclocystoides, which together with the so-called Myelodactylus is
placed by Miller in a new order that he proposes to call Myelodactyloidea. Wliatever
be the nature of Cyclocystoides, there can, I think, l)e little doubt that Salter,
Charlesworth, and more recently Nicholson and Etheridge ^ were right in regarding the

1 Oours elimeutaire de Paleontologie et de Geologie stratigraphique, Paris, 1852, t. ii. fasc. i. p. 134.
- Traite de Paleontologie, t. iv. p. 282.

^ Histoire Naturelle des Zoophytes, Echinodermes, Paris, 1862, pp. 55-58.
■* Lectures on General Natural History, Medical Times and Gazette, November 1856, p. 463.
'" An Introduction to the Classification of Animals, London, 1869, p. 130.

^ Description of three New Orders and four New Families, in the class Echinodermata, and eight New Species
from the Silurian and Devonian Formations, Journ. Cincinn. Soc. Nat. Hist., vol. v. jjp. 221-223.

^ A Monograph of the Silurian Fossils of the Girvan District in Ayrshire, Edinburgh, 1880, pp. 330-334.

REPORT ON THE CRINOIDEA. 193

supposed arms and pinuules which were described by Hall as Myelodactylus ' as a coiled
up stem of peculiar structure. It may perhaps belong to some Crinoid of which the
head is not yet known ; but until Salter's statements ^ have been satisfactorily refuted
by Hall or JMiller, I cannot admit the Myelodactyloidea as a class of Echinoderms
equivalent to the Crinoidea, Ophiuroidea, or Blastoidea.

The Echinoderms which have no tube-feet in their ambulacra, and are more or less
permanently attached by their aboral surface, seem to me therefore to fall very naturally
into three classes, Crinoidea, Cystidea, and Blastoidea. They have several characters
in common which sharply distinguish them from the other Echinoderms, and serve to
define the branch or division Pelmatozoa, Leuckart, which is of course synonymous
with Crinoidea in the widest sense.

I am indebted to my friend Prof. F. Jeffrey Bell for the reference to Leuckart's
original definition of the group. I heard the name first from Sir Wyville Thomson, who
was greatly struck with its appropriateness, and introduced it into the syllabus of his
class lectures. He could, however, give me no reference to it ; but Prof. Bell was
fortunately able to find it in Leuckart's Bericht liber die wissenschaftlichen Leistungen
in der Naturgeschichte der niederen Thiere for 1864-65, where the Echinoderms are
divided into Pelmatozoa, Echinozoa, and Scytodermata (Holothurians). Working back
from this year Prof. Bell eventually succeeded in tracing back this classification of
Leuckart's to a morphological essay published in 1848, where, however, the familiar name
AcTiNOZOA is used to denote the Urchins and Starfishes together. After alluding to the
essential characters of the Pelmatozoa, i.e., the presence of a stem either temporarily or
permanently, Leuckart referred to the two orders of this class, the Cystids and the true
Crinoids.^ The latter is distinguished by the fact that " An dem obern peripherischen
Eande des Kelches noch besondere zahkeich gegliederte Arme sich vorfinden, deren
Skeletstiicke immer dem Perisom angehoren und stets von dem dorsalen Pole ihren
Ursprung nehmen." In this description of the Crinoids, as w^ell as in the prominence
given to the presence or absence of a stalk in the morphology of the Echinodenns,
Leuckart seems to me to have been peculiarly fortunate. The only point to which one
might be disposed to take exception, and it is in reality more a verbal one than anything
else, is his description of the arm-skeleton as belonging to the perisome ; for the term
" perisomatic" skeleton is now somewhat limited in its meaning {ante, p. 73).

The Pelmatozoa therefore differ altogether from other Echinoderms in the presence
of a stem, and in the consequent departure from the ordinary habits of an Urchin, Star-
fish, or Holothurian. Whether sessile, or provided with a stem, the Ciinoid lies on its
aboral surface, instead of creeping about mouth downwards in search of food. The lateral

1 Palfeontology of New York, 1S52, vol. ii. p. 191, pi. xlii. figs. 5, 6.
- Catalog. Camb. Silur. Foss. VVoocItv. Mus., Cambridge, p. 118.

5 Ueber die Morphologie und die Verwandtscliaftsverhaltnisse der wirbellosen Thiere, Braunschweig, 1848, p. 42.
(ZOOL. CHALL. EXP. — PART XXXII. — 1884.) Ii 25

194 THE VOYAGE OF H.M.S. CHALLENGER.

brandies of the water-vessels are therefore either absent altogether as in the Blastoids
and in some of the arms in many Actinometrce ; or they supply delicate papillate tubules,
the tentacles, the chief functions of which are probably those of respiration and sensation.
Ludwig's researches have demonstrated that the primary water-pore of the Antedon-lavva
is homologous with the commencing madreporite of an Echinozoon. But in no Starfish or
Urchin is the communication between the water- vascular system and the exterior effected
through the intervention of the body-cavity as it is in the adult forms of recent Crinoids,
whatever be the condition of the larva. There is good reason to believe that the
hydrospires of the Blastoids and Cystids were neither connected with the water-
vascular ring nor with the body-cavity, so that the absence of a continuous madreporic
canal may be regarded as eminently characteristic of the Pelmatozoa, though it occurs in
some Holothurians and jiossibly also in certain Ophiurids.

The limitation of the functions of the water-vascular system in the Pelmatozoa,
although a natural consequence of the presence of the stem, is in no way connected with
this organ morphologically. With the blood-vascular and nervous systems, however, the
case is different.

There can be little doubt that the remarkable neuro-vascular axis which occupies
the central canal of the stem of a Neocrinoid, occupied a similar position in the Palseo-
crinoids. Even in the case of sessile forms like Edriocrinns there must have been a
chambered organ, from the fibrillar envelope of which were derived the axial cords of the
rays and arms, just as in the recent Holopus. The Blastoids aU have basal and radial
plates, and, with the exception of the Astrocrinidse, a perforated stem, so that one can
hardly be going too far in assuming that its central canal lodged a neuro-vascular axis,
as in recent Crinoids, The same is probably true of the pedunculate Cystids ; though I
much doubt whether the Agelacrinidee had a chambered organ. This peculiar structure,
which is so important a part both of the blood-vascular and of the nervous systems, is
essentially anti-ambulacral, being developed in the right larval antimer. With its
connections it is as important a part of the organisation of a Pelmatozoon as the cerebro-
spinal axis is in a vertebrate animal ; and there is in many respects a striking analogy
between the two.

Although in such intimate relation with the basal and radial plates, the chambered
organ cannot be correlated -ftdth their presence ; for true homologues of these plates occur
in the Echinozoa, in none of which have any traces of an aboral neuro-vascular centre yet
been discovered.

There are two characters, however, by which some or all of the Pelmatozoa are
especially distinguished, viz., (1) the presence of a stem, and (2) the development of arms
upon the primary radials, with muscular articulations between their component joints. The
majority of the Pelmatozoa have both stem and arms, but the Astrocrinidae seem to
have had neither, though possibly stalked when young ; while the remaining Blastoids

REPORT ON THE CRINOIDEA. 195

and the Cystids had no such arms, and the Holopodidse and Edriocrinus were stemless.
None of the typical Pelmatozoa, however, are devoid of both stem and arms, with one
or both of which the chambered organ seems to be correlated. A so-called pedunculate
Starfish has already been described by Prof. Perrier,^ and it was with much disappoint-
ment that I learnt from Mr. Sladen" that Caulaster is far from being the interesting type
which it was at first supposed to be. I am not without hopes, however, that future
morphological work upon Urchins and Starfishes may throw more light upon this question ;
and there is very much to be done by those who will go into the study of the Palaeozoic
Starfishes equipped with a knowledge of the morphology of recent Echinoderms, and
will not be content with merely compiling empirical descriptions of new sj^ecies.

Miiller's original classification of the true or brachiate Crinoids divided them into
three groups, Articulata, Tessellata, and Costata, the last including the problematical
genus Saccosoma, which may perhaps eventually turn out to be an Ophiurid. Eeference
has already been made {ante, pp. 145-147) to the unsuitability of the Mullerian names
Articulata and Tessellata for the two other principal divisions of the brachiate Crinoids.
The latter is practically co-extensive with the Palseocrinoidea of Messrs. Wachsmuth and
Springer.^ These authors have gone further than Zittel and de Loriol, and have pro-
posed to divide up the Crinoidea (understood in the widest sense) into the following
orders: — (1) Blastoidea ; (2) Cystoidea ; (3) Palseocrinoidea ; (4) Stomatocrinoidea ;
(5 ?) Costata. This classification, however, has been by no means generally accepted.
Difi"erent as are many of the Palseocrinoids, e.g., Eucalyptocrinus, from a Pentacrinite,
others, such as the Ichthyocrinidse, have many of the characters of a recent Crinoid ; and
an arrangement which elevates the difference between Pentacrinus and Ichthyocrinus to
the same importance as those between Pentacrinus, Pentremites, and Echmosphcerites
respectively, appears to me to be founded on a misconception of the value of morpho-
logical characters.

On the other hand, although the definition of the Palseocrinoidea which has been given
by Wachsmuth and Springer is capaljle of improvement in one or two respects, it is far
more correct and is based upon sounder morphological principles than any definitions of
the Tessellata which have been di-awn up by Miiller and his followers. But I cannot
regard the two groups Palseocrinoidea and Stomatocrinoidea (Articulata or Neocrinoidea)
as equivalent to the Blastoids and Cystids ; so that while keeping the Palseocrinoids at the
level of an order, I should rank the Blastoidea and Cystidea as classes, in accordance
with the practice generally adopted in this country.

Prof. Chapman* has proposed a classification of the brachiate Crinoids which "is

1 Snr une Ast^rie des grandes profondeurs de I'Atlantiq^ue, pourvue d'un piidoncule dorsal, Comptcs rendus,
t. xcv. pp. 1379-1381.

2 The Asteroidea of H.II.S. Challenger Expedition, part ii., Journ. Linn. Soc. Land. (Zool.), vol. xvii. p. 217.
5 Revision, part ii. p. 3.

* A Classification of Crinoids. Trans. Eoy. Soc. Canada, vol. i., 1883. Section iv., 1882, pp. 113-116.

196 THE VOYAGE OF H.M.S. CHALLENGER.

based essentially on the presence or absence of a canaliculated structure in the calyx and
arm plates." The last of his three leading divisions, the Canaliculata, corresponds to the
Articulata, Miiller ; but Chapman's name does not appear to me to be so greatly superior
to Mtiller's that it could be adopted without hesitation. Platycrinus, Marsupites, and
Uintacrinus have perforated first radials. They are, however, placed l^y Chapman
among the Emedullata, in which "the calyx plates are without internal canals."

The absence of canals in the calyx plates is a character of very general occurrence
among the Palseocrinoids, and one which does not present itself in any adult Neocrinoid.
But there are several Palseocrinoids in which it does not occur at all, and it cannot there-
fore be used as the principal " differentia " for separating the older (Tessellate) from the
younger (Articulate) Crinoids.

Eeasons have already been given for preferring the name " Neocrinoidea " for the
latter group to either Stomatocrinoidea or Articulata. The second of these names is
certainly founded upon a misconception, and I believe the same to be the case with the
first one ; while neither of them was ever properly defined by its author. The name
Neocrinoidea is coming gradually into use ; and as the essential differences between this
order and that of the Paleeocrinoidea have already been discussed in Chapter X., it is not
necessary for me to go into them again, though they may be conveniently put in the
form of a definition.

Order Neocrinoidea.

Crinoids with a regularly pentamerous calyx, which is generally without primary
interradial plates, and except in one genus has no anal or azygous side. The calyx-
radials are perforated, and are generally united to the succeeding plates by a muscular
articulation. The rays may remain simple, or divide from one to eight times, the first
axillary being usually the second joint after the primary radials of the calyx. Orocentral
plate probably never developed ; orals, when present, may be limited to larval existence,
or remain through life partially covering the peristome, luit capable of being separated
so as to open the mouth to the exterior. The oral surface of the visceral mass,
with the ambulacra traversing it, may be more or less paved by plates, but is not in
any way covered up and shut off from the exterior by a vault.

REPORT ON THE CRINOIDEA. 197

XII.— DESCRIPTION OF THE SPECIMENS.
Class CEINOIDEA.

Order Neocrinoidea.

Family HolopiDoE, Roemer, 1856 ; emend. Zittel, 1S79, and P. H. Carpenter, 1884.

Genus, Holopus, d'Orbigny, 1837 ; emend. P. H. Carpenter, 1884.

A. General Account of the Type.

Definition. — Basals and radials completely anchylosed into an asjonmetrieal tube-
like calyx, -wliich is fixed by an irregularly expanded base. On the upper edge of the
cup are five unequal articular surfaces for the attachment of the second radials. Arms
ten, massive, and closely inroUed. Disk relatively small, with a central mouth pro-
tected by five oral plates, between which and the edge of the cup is a very narrow irregular
pavement of smaller plates. Anus probablj^ present, but not yet observed.

Remarks. — The description of this genus, which was established by d'Orbigny in
1837, attracted much attention on account both of the novelty and of the rarity of this
singular type. The specimen described by d'Orbigny, which was for a long time unique,
was brought from Martinique by M. Sander Rang. It is tetra- instead of penta-radiate
as most Crinoids are, and this seems to have caused some misconception respecting its
true nature, for it receives no mention whatever in Pictet's Paleontologie ; while
Dujardin and Hupd^ expressed their opinion in 1862 that Holopus is not a Crinoid at
all, but an altogether different type, probably a Cirripede.^

A few years later, however, a normal pentaradiate example was brought up on a
fisherman's line from deep water off Barl^ados. It was fortunately acquired by the
governor, Sir Rawson Eawson, and was placed by him in the hands of Prof. Louis Agassiz
during the stay of the " Hassler " at Barbados in 1872. Prof. Agassiz intended to have
published a full description of the specimen, but was prevented from doing so by failing
health, and after his death the figures wliicli he had prepared were published by
Mr. Alexander Agassiz, together with a short descriptive note by Count Pourtales.^

1 Memnire sur une seconde ei3pece vivante de la famille dea Crinoides ou Encrines, servant de type au nouveau
genre Holope (Holopus), Gue'rin, Mag. de Zool., 7nie annee, Classe x., 8 pp., pi. 3, Paris, 1837.
^ Op. cit., p. 218.
3 Description of a Specimen of Holopus rangi,i from Barbados, Mem. Mus. Cowp. ZooL, vol. iv. No. 8, p. 51.

198 THE VOYAGE OF H.M.S. CHALLENGEE.

This specimen was subsequently entrusted hj Sir Eawson Eawson to Sir Wjrvdlle
Thomson, together mth two others which he had obtained in 1876, after the publication of
Pourtal^s' notice of the first one. Sir Wy ville described the three as follows : ^ " One
is very complete in all important points, wanting only the two ' bivial ' arms, but retain-
ing the mouth- valves. The second is a little larger ; it wants the mouth-valves, and
again the bivial arms ; and with Sir Eawson Eawson's sanction I boiled this specimen
down to figure and describe the separate parts. The third specimen is quite perfect,
the arms closely curled in in their normal position when contracted; but it is very young,
only about 8 mm. in height. Besides the four examples mentioned, I am aware of only
another which I have not yet seen ; it was shown at the Philadelphia Exhibition, and
was afterwards bought by the Museum of Comparative Zoology at Cambridge, Mass."

The second of these seems to have been the original specimen described by Pourtales,
from which the oral plates or mouth- valves had dropped away ; and as it was gradually
falling to pieces from natural decay. Sir Eawson Eawson allowed it to be dissected.
The figures on PL III., with the exception of fig. 2, and figs. 1-4 on PI, V., show the
results of this process. Fig. 2 on PI. III. is a slightly idealised view of the interior of
the cup, so as to show the oral plates of the large specimen represented in PI. 11.^ This
was supposed by Mr. Murray to belong to Sir Eawson Eawson, and as it corresponds
to No. 1 of Sir Wyville's list, I quite imagined this to be the case ; but Sir Eawson
Eawson does not recognise it as his, and I conclude therefore that it is the mutilated
dry specimen which Prof Agassiz informs me was sent by him to Sir Wyville with
permission to 'cut it up for details. In like manner Sir Eawson Eawson thinks it
possible that the original of PI. IV. may be his young specimen mentioned by Sir
WyvUle as only about 8 mm. in height, but as Prof Agassiz tells me that he also sent
Sir Wyville a small individual, I fear that two specimens have somehow been mislaid.
The one which was shown at the Philadelphia Exhibition, and subsequently bought by
the Museum of Comparative Zoology, is the original of PI. I.

It was obtained by Mr. AVilderboer, the collector for Sir Eawson Eawson, after the
latter gentleman had left Barbados, and having come into the hands of Prof Agassiz,
it was sent by him to Sir Wyville Thomson, together with the Holopus material obtained
during the dredging expeditions of the " Blake." This consisted of (l) the very youuo-
individual shown in PI. V. figs. 9, 10; Cruise of 1877-78; Station 22, 100 fathoms;
ofi" Bahia Honda, lat. 23° 1' N., long. 83° 14' W. ; temperature 71° F. (2) The single
ray shown in PI. Vb. fig. 4. This was preserved in spirit, and the greater part of it
was subsequently cut into sections. Cruise of 1878-79; Station 157, ofi" Montserrat,
120 fathoms.

1 On the Structure and Relations of the genus Holopus, Proc. Eoy. Soc. Edin., 1877, p. 407.

- I did not find this out until too late to alter the notice of the oral plates of Holopus, which appears on p. 95.
See p. 208.

REPORT ON THE CRINOIDEA. 190

Holopxis rangi, d'Orbigny (PLs. I.-Vb. ; PL Vc. figs. 1-3).

Diraensions.

Total height, ....

Greatest height of cup on trivial side,

Least height on bivial side, .

Greatest diameter of upper edge of calyx,

Greatest width of trivial axQlary (composite),

Greatest width of arm.

Diameter of smallest specimen (PI. V.),

Height of smallest specimen, .

A. (PL I.) B. (PL IV.)
40-00 mm. 8-50 mm.

15-00 „ 4-25 „

5-25 „ 1-75 „

17-00 „ 5-00 „

13-75 ,, 3-75 „

9-75 „ 2-00 „

3-25 „

1-00 „

encnistiug calcareous expansion

The tubular calyx which is attached by an irregular
of variable extent, is thick walled, inversely conical, and slightly bent to one side
(Pis. I., II., IV. ; PI. III. fig. 1). A more or less distinctly marked constriction separates
the spreading base from the actual cup, the greatest height of which, measured on the
convex side, is 15 mm. Its cavity narrows very rapidly from above downwards, so that
the thickness of its walls, which is everjTvhere considerable, is greatest at its lower
extremity (PI. V. figs. 1-4) ; and it is probable that the cup is completely closed below
by the spreading base, if not some little way above it.

The analogy of other Crinoids leads one to believe that the cup is composed of radial
plates above, and of basals below ; but it is difl&cult to define the limits of either. The
radials, however, may be traced downwards some little way, owing to the difi"erences of
texture in the limestone network. Sections were made for Sir Wyville Thomson of the
least perfect of Sir Rawson Rawson's specimens. The articular faces round the upper edge
of the calyx are shown in PI. V. fig. 1 (compare also PI. III. figs. 1, 2). They were
described as follows by Sir Wjrv'ille Thomson ^ — " Each facet is traversed by a transverse
articulating ridge, a little in front of which there is the mouth of the tube which lodges
the sarcode axis of the joints, and a little behind its centre there is a somewhat longer
aperture which appears to lead into the cancellated structure of the outer part of the
wall. There are two large shallow muscular impressions on the surface of the facet on
the proximal aspect of the transverse ridge." The larger of these two apertures is not
the opening of a canal, like the smaller and inner one ; but it leads into a deep pit which
lodges the dorsal ligament connecting the radials with the joints above them. It
reappears upon the proximal faces of these joints, and upon the articular surfaces of all
the arm-joints (PI. III. figs. 3-15). In most Crinoids this pit is merely the deepest part
of a large fossa lodging the dorsal hgament (PI. Vila. fig. 15, Id'. PL Vllb, fig. 5 ;
PL Villa, fig. 7— Id ■ PL X. fig. 4 ; PL XX. fig 7) ; and there is an approach to this
condition in the later arm-joints of Holopus, which have a large portion of the articular

1 Proc. Roy. Soc. Edin., 1876-77, p. 407.

200 THE VOYAGE OF H.M.S. CHALLENGEE.

face on the dorsal side of the pit (PI. III. figs. 14, 15). But on the lower arm-joints
and on the radials there is practically nothing of this kind, and the pit lies immediatel}'
next to the dorsal edge of the articular face. But the dorsal surface of the joint is
strongly convex and produced far below this edge, as is well shown in PL III. figs. 5-13.
This is also the case with the distal portions of the first radials, as may be learnt from a
comparison of figs 1 and 2 on PL V. The latter represents a horizontal section of the
radials which passes 2 mm. below theii" edge on the concave (bivial) side, and 7 '5 mm.
below it on the convex (trivial) side.

Around the opening of the central funnel, which is narrower than at the top of the
calyx, is an irregularly shaped pentagonal figure. This is formed by the Hnes of the
transverse articidar ridges, which in Holopus, as in all other Crinoids, are formed
of a much closer and denser limestone reticulation than the remainder of the skeleton.
Immediately within these lines are the indications of the small openings of the central
canals of the radials ; and just outside them are the ends of the pits lodging the dorsal
ligaments. The texture of the limestone network forming the inner faces of the radials
and the fossae for the attachment of the muscles and the interarticular lio-aments
is remarkably difi"erent from that of the outer portion of the cup. The two are separated
by the lines of the transverse articular ridge, as is shown in fig. 2 on PL V. and more
distinctly in fig. 7, where the dark line indicates the position of the articular ridge. The
substance of the radials inside this line is formed of an irregularly open network, the
meshes of which reach 0'08 or 0'09 mm. in diameter, though many of them are much
less, sometimes not a quarter that width.

The peripheral portion of the cup, however, is formed of a much more regular net-
work. This consists of concentric and radiating rods which enclose circular or elliptical
meshes from O'OIS to 0"035 mm. in diameter, and disposed in regular rows with then- long-
axes tangential. Here and there, as shown in PL V. fig. 7, the lines of the meshwork
are a little irregular, but its general character is very uniform. The difference between
the two types of network is most marked, as much in the regularity as in the size of the
meshes, as is w-eU shown in the inner and outer portions of fig. 7, and also in the
enlarged portions of limited areas which are represented in figs. 5 and 6. Fig. 8, on the
same plate, is an ideal diagram, constructed by ]\Ir. Black, showing the regular disposi-
tion of this peripheral reticulation. The difference of the two textures is obvious enough
to the naked eye ; but it becomes more apparent with the help of a lens which brings out
the regularly striated aspect of the outer part of the cup. This is well shown in PL \.
fig. 2, and less clearly in fig. 4, which represents a section taken about 5 mm. above the
basal expansion, and corresponding to the upper face of the vertical section shown
in ficr. 3.

O

The central funnel is here much narrowed, and the lines of the ai-ticular ridges are
seen at a distance of about 1 mm. from its opening. Outside the pentagonal figure

EEPORT ON THE CEINOIDEA. 201

bounded by these Hues, tlie whole of the skeleton is formed by the regular, small-meshed
network ; while the inner, less dense portion which has a whiter look, narrows gradually
downwards as shown in fig. 3, until it entirely disappears from the wall of the diminish-
ing central funnel. There is no trace of it round the greatly reduced opening on the
under surface of the segment of the cup which is represented in figs. 3 and 4. In this
specimen, as in that represented in PL I. and also in d'Orbigny's original, the lowest
portion of the calyx-tube immediately above the spreading base, is partially covered with
a chitinous-looking layer of variable height, and marked by roughly concentric lines
which somewhat obscure the calcareous network beneath. It is merely a thin superficial
skin, however, and is evidently of no special importance, or it would be universally
present.

The pentagonal figure indicating the position of the articular ridges on the radials is
still visible in the section shown in fig. 4, which corresponds to the upper surface of the
fragment represented in fig. 3 ; and the openings of the central canals are also traceable.
This would indicate that the greater part of the calyx-tube is composed of elongated
radials. One of these canals is seen in nearly longitudinal section in the portion of the
lower half of the cup which has been removed to expose the view given in fig. 3. But it
is not traceable beyond the limit of the whiter, less dense portion of the skeleton. I
strongly suspect, therefore, that this indicates the position of the lower surface of the
radials ; and the analogy of all other Crinoids would lead to the conclusion that the
small portion of the calyx-tube between this and the spreading base consists of closely
anchylosed basal plates, the presence of which was taken for granted by Sir Wyville
Thomson.^ There must certainly be a chambered organ, from the fibrillar envelope of
which the axial cords of the rays and arms originate (PI. Vc. fig. 2, A) ; and one would
naturally expect it to be situated at the lowest part of the calyx-tube. This narrows
rapidly downwards, and its interior is marked by five vertical ridges corresponding with
the radials in position. They are fairly distinct at the level of the section shown in
PL V. fig. 4 ; but they become less marked as they proceed downwards, and, being
composed of the whiter, less dense network, disappear together with it. They extend
upwards to the edge of the cup at the intermuscular notches ; though they are much less
distinct on some of the radials than on the others. They thus occupy the position of the
ventral radial furrows which are often so marked on the interior of the central funnel of
the calyx in other Crinoids (PL X. figs. 1, 4, vrf; PL XII. fig. 15 ; PL XX. fig. 8 ;
PL XXX. fig. 3).

From the facts detailed above, we may, I think, assume with tolerable certainty that
the tubular body-chamber of Holopus is not composed of a " piece centro-dorsale sessile "
as stated by de Loriol ;" but that it consists of basals and radials like the calyx of any
other Crinoid. I cannot quite make out whether de Loriol employs the word " centro-

1 hoc. cit., p. 407. - Palcont, Frang., he. cit., p. 188.

(ZOOL. OHALL. EXP. PART XXXIl . 1884.) 11 26

202 THE VOYAGE OF H.M.S. CHALLENGEE.

dorsal" in the sense in which it is usually understood, i.e., as the cirrus-bearing top stem-
joint of the Comatulae. The tubular body-chamber of Holopus is, however, distinctly
not of this nature ; so that the use of the name " centro-dorsal " is apt to lead to confusion.
In a subsequent passage ^ de Loriol expresses another view of the composition of this cup
or "cu23ule." He speaks of the axillaries which are articulated to its upper edge as
" radiales uniques," resting as in Cyathidium, " sur les angles de la cu2iule, qui pourrait
done etre evisagee comme etant composee de cinq pieces basales interradiales." This
would be a most singular morphological condition, and one without a parallel in any
(ither Crinoid. Primary radials would be in contact wdth each other, but not united, and
rest on articular surfaces each of which would be formed by the upj)er edges of two
basals.

The union between basals and radials is invariably a simple synostosis such as I have
described above (pp. 2, 3), and never a muscular joint like that between the upper edge
of the calyx-tube of Holopus and the compound axillaries. The evidence afibrded by
sections of the cup, however, indicates clearly that it is principally composed of closely
united first radials which, as will be pointed out subsequently, have a remarkable
similarity to the radials of the Liassic genus Eudesicrimis.

Sir AVyville Thomson thought it probable that second radials are also present in the
cup. If so, they must be imited to the first by synostosis, which would be a most
unusual condition in any Neocrinoid ; and the close resemblance of Holopus to Eudesi-
crinus seems to negative this idea altogether ; while, as pointed out by Sir Wyville,' there
would be a true muscular joint between the second radials and the radial axillaries, which
is not the ease in any other recent Crinoid.

Considering then the articular surfaces at the edge of the calyx tube as those of
first radials, we find that they difi"er considerably in size. According to Sir Wyville
Thomson' " the upper border of the cup, bearing the facets, is very irregular in thickness ;
and in all the specimens which I have seen, including d'Orbigny's, one side of the border
is much thicker and considerably higher than the other side, and the three arms articu-
lated to it are much larger than those articulated to the opposite side. There is thus
a very marked division into " bivium " and " trivium," and consequently a bilateral
symmetry underlies the radiated arrangement of the antimeres." This is shown in PL
V. fig. 1, and also, though less clearly, in PL IlL fig. 1. Besides this again the individual
facets, both of bi\dum and trivium, are of difi"erent sizes and shapes. The articular ridge
which crosses the central facet of the trivium is considerablj' longer than that of either of
the two remaining facets, and these are longer than the ridges on both the bivial facets.
The adjacent muscular plates of these two last are fused into a short tongue-shaped
process which stands up prominently in the angle of the bivium. It is essentially of the
same nature as the "clavicular piece" which projects in the middle of the distal

1 Paleont. Frang., he. cit., p. 191. - Loc. at., pp. 407, 408. ^ Loc. cit., p. 408.

REPORT ON THE CRINOIDEA. 203

face of every axillary, radial or otherwise (PL III. fig. 4; PL XXI. fig. Ic). A similar l)ut
larger process is formed by the fusion of the remaining muscular plate of one of the
bivial facets with its fellow on the adjacent facet of the trivium (woodcut, fig. 10).
This is seen in the lower part of PL V. fig. 1 ; and it is also
visible projecting into the cavity of the calyx in the corre-
sponding part of fig. 2. The other projection seen to the west
of it in the same figure is formed by the united muscle-plates
of two of the trivial facets, which extend inwards in a more ^ .

horizontal direction than the larger processes already described.' ' ^

These two large projections are also seen in the north-east por- Fk. lo.— view of the upper pan of

.^, -,-,-,- 1 ■ 1 Ti • 1 1 1 j_i the calyx-tube of Holujiiis rangi

tion of fig. 1 on PL 111., which likewise shows verj'' clearly the on its lower or uviai side. The

. . two bivial facets are well shown,

separation of the two bivial facets by a liointed upward together with the interrarUaipro-

■*■ . ''^ss separating them. The left

extension of the outer surface of the calyx. This is very ii-™<i one >s separated from the

•' •' adjacent trivial facet by a still

evident in PL I. fig. 1, and in the right hand figure on larger process ; but there is only

O ' o o a small one on the right side.

PL II. The latter on its left hand side shows traces of the

same condition between the other angle of the bivium and the trivial facet next it. This
is also visible on the left of PL III. fig. 2. But it is much less distinct in the large
specimen represented on PL I. In the young individual shown in PI. IV. this character
is fairly well marked, except at the two angles of the trivium ; while in the still younger
and very remarkable specimen obtained by the "Blake" (PL V. figs. 9, 10), the shallow
calyx is much more symmetrical, and its outer surface sends a pointed extension upwards
between every two facets.

This surface is marked by an irregular row of scattered tubercles, though none are
visible in the other young specimen (PL IV.). They are replaced, however, by tolerably
well defined ridges which occupy the middle line of the radials, and extend downwards
from their upper border to a little distance from the spreading base. They diminish as
they go, and finally disappear altogether at a level which probably marks the downward
limit of the radials. They are naturally more distinct on the trivial than on the
bivial side, and are better marked on the united second and axillary radials,
where they bifurcate and are continued outwards on to the arms as well defined
medio-clorsal ridges. The lower joints, especially of the trivial arms, also bear one or two
small tubercular elevations on either side of the median ridge. These median ridges like-
wise appear on the second and third radials of the youngest specimen, in which, however,
they have more the appearance of a partially disconnected line of tubercles (PL V.
figs. 9, 10). A row of ill defined tubercles is also visible immediately inside each lateral
edge of the second radials. There is a good deal of difference in the external ornamenta-
tion on the calyx-tube of the two adult individuals. The large American specimen

1 Both these figures, as well as the remaining ones on the plate, and in fact all those Jraflii for Sir Wyville
Thomson, are reversed, having been drawn upon the stone in the natural position of the speciiueus.

20-1: THE VOYAGE OF H.M.S. CHALLENGER.

(PI. I.) is almost bare, though faint elevations are visible along the middle lines of the
radials, and indistinct, scattered tubercles appear between them, except in the centre
of the trivium. In the smaller specimen, however (PI. II.), there are three, fairly distinct
double rows of blunt tubercles which correspond to the trivial radials. But on the
l)ivial side there is more indication of median ridges; while in the fragment shown in
woodcut (fig. 10) the double row of tubercles is tolerably distinct all round the cup, except
on one of the bivial radials. Other tubercles are scattered about between these rows,
though without any definite arrangement; while they are abundant on the dorsal surfaces
of the two outer radials and of the large lower arm-joints (PL III. figs. 3-9), disappear-
ing, however, as the joints become more and more compressed laterally.

All the entire specimens of Holopus which are known to science have been preserved
in the dry state, and have a blackish-green tint which is due to pentacrinin, as stated above
{ante, p. 129). It is darkest in the older individuals, and contributes to the shagreen-like
appearance that is so characteristic of the type. An isolated ray (PI. Va. fig. 3 ;
PI. Vb. figs. 4, 5) was, however, obtained by the " Blake " ofi" Montserrat, and preserved
in spirit. In this condition the skeleton has a dead white appearance.

In the figure of d'Orbigny's original specimen the first radials are shown to bear large,
pentagonal axillaries which appear to be all in one piece. They were so described by
Pourtales,^ and also by Sir WyviUe Thomson,^ who did not, however, exclude the
possibility that they might be formed of the second and third radials coalesced, with the
syzygy between them obliterated. The very young individual dredged by the " Blake,"
and the somewhat older form, shown in PL IV., throw much light upon this question.
The calyx-tube of the former (PL V. figs. 9, 10) is wide and shallow, while the second
radials which it supports are widely hexagonal and only partially in contact laterally.
Eesting upon their distal edges are the smaller triangular plates to which I have referred
as the axiUaries. Agassiz, in his brief description of this remarkable form,' states that the
larger hexagonal plates are "the radial axiUaries of Sir AVyville Thomson, but the smaller
triangular ones seem to become fused with them in the adult." I think, however, that
there can be little doubt that the larger plates are second radials, and the triangular ones
the third or axillary radials. They are all equal and similar, and meet one another all
round so as to completely close the cavity of the calyx. It might be suggested that
these are the combined second and axillary radials, while the hexagonal plates are the
first radials, and no others are present. I do not think, however, that this can be the
case ; partly on account of the very marked manner in which the hexagonal plates are
separated from one another and from the shaUow cup below them ; and partly because
there is no indication whatever of their sending upward extensions between the

1 Mem. Mm. Comp. ZoUl, vol. iv. No. 8, p. 62, 1878.

" Loc. cit., p. 408.

^ Description of a young Holopus, Bull. Mus. Comp. Zoiil, vol. v. p. 213, 1879.

EEPOET ON THE CRINOIDEA. 205

axillaries, like those visible iu the next youngest specimen (PI. IV.) ; whereas the hexa-
gonal plates themselves are separated in this manner.

Further, in nearly all Neocrinoids which have ten or more arms there are three
radials. This is true of all the recent Crinoids except Metacrinus, which has a larger
number, four or six ; and the only fossil genus which has two radials is the aberrant
form Plicatocrinus.

In all the Neocrinoids, except de Loriol's recently established genus Eudesicrinus,
there is either a syzygy or a ligamentous articulation between the two outer radials ; and
the existence of a syzygy in Holopus is therefore nothing unusual, though there is less
evidence of its presence in the adult condition than is usually the case. But this is
scarcely surprising when we remember the excessively intimate union of the first radials,
of which no indication whatever is visible on the exterior of the calyx. Some individuals,
however, exhibit distinct traces of a sutural line dividing the large axillary into two parts.
Such a line is visible in the young specimen (PI. IV.) on all the axillaries of the trivium,
crossing them at the point where the medio-dorsal ridge bifurcates as described above ;^
but it is less distinct in the two bivial axillaries. On the other hand, the three trivial
axillaries of the large American specimen present no indications whatever of being
composite joints, and have a regular, broadly pentagonal shape. This is well shown in
PI. I. fig. 2 ; but the bivial axillaries represented in fig. 1 are of an entirely different
character, each of them being distinctly in two parts, which look as if they were
articulated rather than suturally united, while they do not present the symmetrical
appearance characteristic of the corresponding parts in other Crinoids.

In the one case there is a large and wedge-shaped second radial which has all the
appearance of an ordinary brachial. It supports a triangular axillary, but the apposed
faces of the two do not correspond exactly. The axillary extends beyond the narrower
end of the second radial, and so comes in contact with the upward extension of the first
radial akeady described. This is shown in PI. I. fig. 1. The broader end of the second
radial, however, extends considerably beyond the axillary, and meets not only the com-
posite axillary of the adjacent trivial ray, but also the first brachial of its own ray as well
as that of the next.

The second radial of the other bivial ray, which is shown in the middle of PI. I. fig. 1,
is more oblong than its fellow. Like it, however, it is wider than the roughly triangular
axillary, and supports a considerable portion of the large first brachial. But it is not
overlapped by the axillary at the other end, and completely cuts it ofi" from the first
radial below.

None of the four remaining axillaries of the specimen figured in PI. II. show any
distinct traces of their being of a composite character ;'" though there are some lines upon

* These lines are not clearly seen in tlie positions of tlie specimen which are represented on PI. IV.
2 The lower angle of one of these exhibits an accidental fracture.

206 THE VOYAGE OF H.M.S. CHALLENGER.

one or two of them whicli may perhaps admit of this interpretation. But not even this
can be said of the originals of figs. 3 and 4 on PL III. I think, however, that the evidence
detailed above is sufficient to bear out the statement that Holojnis has three radials, of
which the two outer ones are united by syzygy. We should accordingly expect to find
a similar syzygial union between the first and second brachials ; bi;t of this there is no
evidence whatever. The distal face of the first and the proximal face of the second
brachial (PI. III. figs. 3, 7) present the ordinary characters of a muscular joint. There
are, indeed, in the small specimen shown in PL IV. some traces of lines crossing the first
brachials, which might be taken as indicating a syzygial union of two primitive joints ;
but they are nothing like as distinct as those in the radials. I think, therefore, that for
the present, at any rate, we must regard Holopus as an exception to the general rule
which holds good in other Crinoids, as to the similarity between the modes of union of the
two outer radials and the two lower brachials respectively {ante, p. 49). It is further
remarkable from the fact that there seem to be no syzygies between any of the other
arm-joints.

The outer surfaces of the composite radial axillaries were described by Sir Wyville
Thomson^ as "very gibbous, thrown out into almost hemispherical projections, studded
with low tubercles" (PL III. figs. 3-5). They are produced dorsally a considerable
distance beyond the edges of the articular faces, as is the case with all the lower arm-
joints (PL III. figs. 6-13); and they fit very closely against their fellows, their sides
being flattened and more or less marked by ridges and furrows, which interlock with
those on the adjacent axillaries. These furrows are also apparent on the sides of the
lower arm-joints (PL II. ; PL III. figs. 6-12 ; PL Va. fig. 3 ; PL Vb. fig. 4). The
muscle-plates of the axillaries, and in a less degree also those of the arm-joints, are greatly
thickened, and their upper edges are cut out into coarse teeth. This is well shown in the
right-hand figure on PL II. and in the upper part of PL III. fig. 2, where some of the
adjacent axillaries are seen interlocking wdth each other.

In all the specimens of Holopus yet known, including the fresh fragment dredged by
the " Blake " ofi" Montserrat, the arms are strongly recurved, and by their close mutual
apposition conceal the disk entii-ely (Pis. I., II., IV.; PL Va. fig. 3 ; PL Vb. figs. 4, 5).
Obviously, however, this cannot be the natural condition of the living animal. There is a
large food-groove on the upper surface of each arm and pinnule (PL Va. fig. 1 ; PL Vb.
figs. 1, 4, 5) ; and there is every reason to think that the living animal, when undisturbed,
spreads out its arms with the ventral surface upwards just as other Crinoids do for the
purpose of obtaining food. The large size of the paired flexor muscles uniting the joints
(PL VI3. fig. 1, m) would seem to give the power of rolling in the arms very rapidly
and completely, so as to aftbrd the utmost protection to the soft parts contained
within the cup ; while the small, but very close and compact bundles of elastic

1 Loc. cit., p. 408.

REPORT ON THE ORINOIDEA. 207

ligaments on the dorsal side of the articular ridges would help in the extension of the
arms again.

It has already been mentioned that the trivial arms are larger and better developed
than those of the bivium ; but in both cases a variable number of the lower joints
(PI. III. figs. 6-13) are considerably larger than those which follow them (figs. 14, 15),
and the passage from one to the other is usually somewhat sudden. On the trivial arms
there are generally from 8 to 10 of these large, massive joints ; but on the bivium there
are only about seven, six, or even less. The difference between the two is very well
shown in the small specimen represented in PI. IV. The shape of these lower arm-joints
is rather variable. They may be roughly oblong as is the case with the first two or three,
or their edges may be more oblique so as to give them a truncated wedge-like form.
The more wedge-shaped these joints are owing to the obliquity of their terminal faces,
the greater is the inequality in the size of the muscle-plates on the two sides of the
median groove. This inequality is visible in the joints represented in PL III. figs. 10 to 12,
though it is sometimes still more distinct. The pinnule-socket of such a joint is on the
thickened upper edge of the higher muscle-plate. The general character of these lower
arm-joints is much less regular and symmetrical than is the case in other Crinoids, so
that many of them are more or less of a monstrous nature. In some few cases, indeed,
the joint is smaller than usual and triangular, not extending completely across the arm,
so that the joints above and below it come into contact with one another. This is shown
in various parts of both figures on PI. I.; and it is comparable to the condition of other
parts of the same specimen, viz., the way in which the first brachials may partly rest on
the second radials, or the axillaries on the first radials, as has been already described.

Sometimes again, a first brachial becomes unusually large, as is shown on two of the
bivial arms in PL I. fig. 1. The inner one of the two bears a small, triangular, second
brachial, and consequently comes into contact with a similarly large, third brachial along
its outer edge ; but the outer edge of the other second brachial sends a long process
forward by the side of the next three joints, which are much smaller than their feUows
of the adjacent arm.

Other irregularities of growth appear in the same individual, but they are by no
means so marked in that shown in PL II. This, moreover, shows very well the rather
sudden diminution in the size of the arm-joints which lose their tubercles and gradually
become laterally compressed, so that their medio-dorsal edge is tolerably sharp. This
form of joint is figured in PL III. figs. 14, 15, and PL Vc. fig. 2. The longest arms
seem to have about eighteen of them, raising the total number of brachials to between
twenty-five and thirty.

The larger, outer sides of all the brachials bear the pinnules (Pis. II., Va., Vb.).
That of the first brachial is comparatively small, and is attached close to the distal edge
of the joint ; the next pinnule is invisible in aU the specimens, but those of the third

208 THE VOYAGE OF H.M.S. CHALLENGER.

and following brachials are much larger and have broad lower joints that gradually come
to take up more and more of the whole surface of the arm-joints to which they are
attached. In fact the bases of the pinnules of alternate joints that are borne upon the
same side of the arm are only just separated from one another by the narrow ends of
the intervening joints, which have their pinnules on the opposite side of the arm. This
is well shown in the right-hand figure on PL IL, and also in PI. Va. fig. 3. The pinnules
are rolled in upon themselves (PI. III. fig. 16) exactly in the same way that the arms
are (PI. Va. figs. 1, 2). The four or five lower joints are very broad, but the rest of the
pinnule tapers away rather rapidly. The joints are united by paired muscular bundles
(PI. Vc. fig, 2, m), which is a somewhat unusual condition.

The disk of Holopus is unfortunately still but very imperfectly known, and I have
only been able to examine it in one specimen. The central mouth is protected by five
large and triangular oral plates which are opposite to the clavicular pieces of the united
radials (PL III. fig. 2). The lateral edges of each of these plates are thickened and some-
times more or less cut into false teeth ; while the raised central portion is pierced by from
fifteen to twenty minute holes, the water-pores. The bases of the orals seem sometimes
to rest directly against the edge of the radials ; while they are sometimes separated from
this edge by an irregular row of small triangular plates. It is not unlikely that an
anal tube is concealed somewhere or other among these plates, as in the case of Hyocrinus
(PL VI. figs. 3, 4) ; but I have seen no certain traces of it in the dry specimen. The
same would probably be the case with Hyocrinus under similar conditions.

The food-grooves which come away from the mouth between every two of the oral
plates are continued out on to the axillaries and from thence on to the arms. They
occupy the deep channel between the large muscular processes at the sides of the joints,
and in the dry specimen appear to be bordered by small, irregular plates. These, how-
ever, do not seem to correspond either to the side plates or to the covering plates of other
Crinoids (PL Vc. figs. 9, 10; PL XLIX. figs. 6, 7; PL LI. figs. 11, 12; PL LIV.
figs. 4, 6-9); for an examination of spirit specimens shows that these small plates really
belong to the tentacles, which are relatively large and stout (PL Va. figs. 1, 2. PL Vb,
fig. 2 ; PL Vc. figs. 1-3 — T). The bases of these tentacles are protected by scale-like
plates formed of the usual calcareous reticulation (PL Vb. figs. 2,3). They are not easily
made out at the side of the arm-groove, but on the lower parts of the pinnules there seem
to be from two to three tentacles on either side of each joint. It is diflicult to get a correct
estimate of their absolute size ; but after careful comparison with an eyepiece micrometer
I should judge them to be nearly twice the size of the largest that I could find in any
preparations of Antedon eschrichti. The general arrangement of the tentacles is the same
as in other Crinoids ; but the epithelial layer covering them is, if anything, thinner than
in Antedon eschrichti, though thrown into much stronger corrugations at the outer ends
of the tentacles.

REPORT ON THE CRINOIDEA. 209

Shortly before his death Sir Wyville Thomson phiced in my hands a portion of the
ray represented in PI. Vb., with the request that I would cut it into sections for him. I
found this to be an exceedingly difficult task, j)artly because of the roUed-up condition
of the arms, and partly because the calcareous substance of the skeleton is so much
denser than that of other Crinoids; so that the organic basis which is interpenetrated by it
and remains behind after decalcification, has nothing like the consistency that we meet
with in the corresponding parts of the Comatulse or of Bathycrinus. The presence of
large bundles of muscles and ligaments without any helping syzygies also increases the
difficulty of all attempts to obtain thin sections. But although I was not so successful
as I could have wished, I was able to determine satisfactorily that the anatomy of a
Holojnis-aviai is similar in all essential respects to that of an ordinary Criuoid (PI. Vb,
fig. 1 ; PI. Vc. figs. 1, 2). The axial cord traversing the central canal of the skeleton
gives oiFits pinnule branches in the usual way, i.e., alternately on opposite sides. These
branches have a long distance to go before they reach the pinnules, owing to the
attachment of the latter on the upper edges of the large muscle-plates. As long as the
branch remains in the substance of the arm-joint it does not take a straight course as is
the case in the other Crinoids, but is thrown into a series of loops in a dorsoventral
direction (PI. Vc. fig. 2, a), and after it enters the pinnule its course is still somewhat
sinuous (PI. Vc. fig. 3, a).

These branches, like the main arm-trunk, are relatively of very small size, which is
perhaps to be accounted for by the fixed position of the animal. No swimming-
movements are of course possible, but only those of flexion and extension are performed
by the arms. All the ambulacral structures of the Holopus-Sirm. are lodged in the deep
median gi'oove of its skeleton, and are usually small in comparison with the great
transverse diameter of the joints. The coeliac canal is situated, as usual, between the two
large muscular bundles, with a small genital canal separating it from the single
subtentacular canal above (PI. Vb. fig. 1).

The epithelial lining is very much the same in character in all these canals, consisting
of low flattened cells. According to Ludwig ^ this is also the case in Ayitedon eschrichti,
but this statement is not borne out by his figures. In one figure ' he represents a well
marked cellular lining to the coeliac canal and subtentacular canal, but leaves the genital
canal without any ; though in a more magnified representation ^ the wall of the genital
canal bears an excessively delicate layer of much flattened cells, which consist of little
more than nuclei. This is more in accordance with my own observations, for I have
always found that the epithelial cells in the genital canal are much flatter and less easy
to see than those in the coeliac and subtentacular canals. In Holopus, however, the
difierence is much less marked. The genital cord is of essentially the same nature as in
other Crinoids ; though it is of a much less branching character in the axillary radial

1 Crinoiden, he. cit., p. 29. - Ibid., pi. xii. fig. 8. ^ Ibid., pi. xiii. fig. 13.

(ZOOL. CHALL. EXP. PAET XXXII. — 1884.) li 27

210 THE VOYAGE OF H.M.S. CHALLENGEE.

than is usuallj^ the case so near the disk. It is connected with ovaries alternately on
opposite sides of the arm, from about the first to the fifteenth brachial (PL Vc. fig. l,gc.).
The ovaries are short and stout, and confined to the pinnule-bases in the broader, lower
parts of the arms; but where the joints are smaller the ovaries appear immediately beneath
the water-vessel, and the boundaries between the three arm-canals cannot be traced
(PL Vc. fig. 2, ov). The ova, of which all stages are visible, are more like those of
Antedon eschrichti than is the case in many Comatulse, but they are somewhat larger,
reaching a diameter of 0'22 mm.; while 0"1737 mm. is the size of the largest ovum of
Antedon eschrichti which was measured by Ludwig.

I have unfortunately been unable to make out anything definite wdth regard to the
presence of a radial nerve and l)lood-vessel, which are ordinarily found between the
water-vessel and the ciliated epithelium of the food-groove, but this is so often the case in
other Crinoids, except in sections of more than average goodness, that I have no doubt
whatever respecting the existence of these structures in Ilolojnis ; and I see no good
reason to believe that in any essential point of its visceral anatomy there is any important
difi'erence between it and other Crinoids.

All the specimens of Holojnts which have been preserved in the dry state are of a
dull dark green tint, sometimes verging on black. But Mr. Agassiz records that on one
occasion, ofi" Montserrat, the " Blake " dredged an imperfect whitish specimen. This
consisted of a detached axillary joint and the two arms belonging to it, as shown in
Pis. Va. fig. 3, and Vb. fig. 4. The green colour assumed by the dry specimens is
possibly due to post-mortem changes, as seems also to be the case in the Pentacrinidee.
Prof. Moseley informs me that many of the individuals dredged by the Challenger
were white when captured, although tinged with pentacriniu, owing to the colouring
matter being in some way masked during life, and only manifesting itself after death.
During a visit which I paid him recently at Oxford, one of the dry specimens of Holopus
was treated with spirit, and yielded a dull green solution with a red fluorescence.

Prof. Moseley examined it with the spectroscope, and found the colouring matter to
be identical with the pentacrinin which he had discovered in the various species of
Pentacrinus and Metacrinus that were dredged by the Challenger in the Pacific and
East Indian Archipelago.

Holopus has not yet been met with outside the Caribbean Sea. D'Orbigny's original
specimen was obtained at Martinique, while, thanks to Sir Rawson Raw^son, others have
been discovered in the neighbourhood of Barbados. The young individual figured
in PL V. was dredged by the "Blake" in 100 fathoms off" Bahia Honda; while
the white fragment, already mentioned, was found at a depth of 120 fathoms off"
Montserrat.

During the stay of the Challenger at Bermuda,^ Sir Wyville Thomson obtained from

1 The Atlantic, vol. i. p. 321.

REPORT ON THE CRINOIDEA. 211

a local collector " a small worn and rounded fossil, which seemed to bo the cup of a
Crinoid allied to Holopus."

Prof. Moseley tells me that he tbinks it wfis a recent specimen in the dry state ; but
since it has unfortunately been lost, I am unable to say anything as to its nature.

B. On the Systematic Position of Holopus.

For some time after the publication of d'Orbigny's original descrijition of Holopus
the real nature of this remarkable type was more' or less misunderstood, partly, perhaps,
because the original specimen was tetramerous and not pentamerous like most Crinoids.
Eventually, however, Eoemer ^ made the genus the type of a new family, Holopocrinidse ;
though he did not characterise it more closely. This proceeding was objected to by
Quenstedt^ partly on account of the imperfection of our knowledge of the type, and
partly because he considered it possible that Holoinis might be merely a larval form,
destined eventually to become detached and to imdergo further transformations. He
added " Die Kiirze der Saule, die keilformigen Armglieder mit einfachen Pinnulen sprechen
an meisten flir die Comatulafamilie." The latter character, however, is absolutely
worthless as a generic distinction, many Comatulse having discoidal or saucer-shaped arm-
joints like those of Pentacrinus and Ajnocriims, while all the Neocrinoids have simple
pinnules. The first peculiarity mentioned hj Quenstedt is founded on a misapprehension,
for he considered the calyx to be formed of the axillary radials only, regarding the tube-
like body-chamber as a stem. It exhibits no transverse segmentation, however, and has
five articular facets on its upper edge, while it encloses the viscera ; and all these characters
are totally foreign to the stem of a larval Crinoid, or indeed of any Crinoid whatever.

In the year 1847 a remarkable new type of fossil Crinoid was described under the
name of Cyatliidiwn by Steenstrup,' who spoke of it as like Eugeniacrinus, but without a
stem. In Theil ii. of the Lethsea Geognostica, Eoemer made it the type of a sejjarate family
Cyathidiocrinidse, which he placed next to the Holopocrinidse ; but in Theil v. he refers
to it as belonging to the Poteriocrinidse, together with Eugeniacrinus and Taxocrinus.

Between Holopus and Steenstrup's Cyatliidiwm from the Faxoe Chalk there is
certainly a very close analogy, though there are a few well marked difi'erences. Apart
from the bud-like peculiarities of growth presented by Cyathidium, it has a more open
cup, with relatively thinner walls than that of Holopus. Its appearance varies consider-
ably in different individuals, being sometimes low and shallow, and in other cases longer
and more tapering.

The articular facets on its upper edge are much smaller than in Holopus, and their
downward slope faces inwards instead of outwards, as in the recent form (PI. III. fig. 1),

1 Lethaea Geognostica, Theil. ii. pp. 226, 227. 2 Encriniden, p. 186.

^ Anit Bericht ii. d- 24 Vtrsamml. deutsch. Kuturf. und Aerzte in Kiel, 1846, puLli.sbeJ 1847, p. 15.

212 THE VOYAGE OF H.M.S. CHALLENGER.

In fact the articular ridge and not tlie edges of the muscle-plates forms the immediate
boundary of the central cavity; and the fossa for the dorsal ligament is still more
reduced than in Holopus, where it is represented by a median pit that is scarcely to be
traced at all in Cyathidnim. The presence of these articular faces seems to have
escaped the notice of de Loriol, which is doubtless due to his not having been able to
examine sufficiently good specimens. For he describes the calyx of Cyathidnim ^ as
composed " d'une piece centro-dorsale cupuliforme portant, sur son bord superieur, cinq
facettes syzy gales, sur lesquels reposaient, sans doute, des pieces radiales dont le nombre
est inconnu." These supposed syzygial facets are, however, the articular faces of first
radials, which are by no means so unknown as de Loriol suj^poses, though their inferior
boundary is still uncertain.

The inner face of each side of the more or less pentagonal cup formed by these
radials is marked by a median furrow corresponding to the ventral radial furrow of other
Crinoids (PI. X. figs. 1, 4, vrf; PL XX. fig. 8), and immediately external to its upper
end is the opening of the central canal. These features indicate that in Steenstrup's
Cyathidium the sides of the pentagonal cup correspond to the radials, just as they do in
Holopus (PL V. figs. 1, 2, 4). To this same genus Cyathidium, Schliiter has doubtfully
referred a minute Crinoidal calyx discovered by him in the Eocene of Monte Spilecco
near Venice.^ It only reaches 9 mm. in height, but has the same general form as the
cup of the recent Holopus, being attached by a spreading base, between which and the
cup proper there is a more or less well marked constriction. This does not appear to be
generally the case in the Faxoe specimens. In one or two cases there seem to be traces
of basiradial and interradial sutures, and the position of the former, if real, would indicate
that the basals are relatively much higher than they can possibly be in Holopus.

The great peculiarity of Cyathidium spileccense, however, lies in the relative position
of the radials in the pentagonal cup. Their articular surfaces correspond with the angles
of the pentagon, and not -nath its sides, as is the case both in Holopus and in Steenstrup's
Cyathidium, while the middle of each side is raised into a slight ridge which separates
the articular facets of two contiguous radials. The result of this is that the five openings
of the radial canals are situated at the angles of the calyx. They occupy about the middle
of the rim, the outer edge of which is slightly truncated, and shows traces of a fossa for an
elastic ligament. The peculiar semilunar shape of these articular facets is considered by
Zittel as one of the generic characters of Cyathidium. But it certainly does not appear
in any of the Faxoe specimens which were lent to Sir Wyville Thomson by Prof.
Loven. "We know too little about both of these species to make them types of
different genera ; and in default of further information it appears undesirable to unite
them with Holopus. Both forms resemble and differ from it in various points, and are

^ Pal^ont. Fran^., Inc. cit., p. 187.

2 AstyliJe Crinoideii, loc. cit., pp. 50-54, Taf. iii. figs. 11-15.

REPORT ON THE CRINOIDEA. 213

unqnestionably very closely allied to it, but I am inclined to think that it would he
premature to consider them all as congeneric.

Two other fossil genera of Neocriuoids, Cotijlecnnus and Eudedcrinus, both confined
to the Lias, are nearly related to Holopiis and Cyathidium, and should in my ojjinion be
placed in the same family. This has been generally done Avith Cotylecrinus, which is
perhaps better known by its older name of Cotylederma, Quenstedt. But Schliiter denied
its relationship to Cyathidium} which had been previously pointed out by Roemer^ and
Deslongschamps,^ on the ground that there are no perforated plates in Cotylecrinus. It
is true that Quenstedt's original specimens had no radials attached, and were therefore
imperforate, as were most of those figured by Deslongschamps ; but the latter author also
described and figured a fine specimen of Cotylecrinus docens, showing the large articular
surfaces of the radials, and the openings of their central canals, just as in Cyathidium.
This must surely have been overlooked by Schliiter, or he could scarcely have questioned
the relationship of the two t}^es.

Both in Cotylecrinus and in de Loriol's new genus Eudesicrinus, the radials rest upon
a more or less tubular structure which is slightly expanded below and has been variously
described. In the former genus it has been called a stem by Quenstedt and by
MM. Deslongschamps,* and a top stem-joint by Schliiter.^ Zittel" suggested that t^he
upper part of it, which is marked by crests with intervening fossae for the reception of
the radials, should be considered as composed of anchylosed basals, and that the lower
part is a centro-dorsal ? De Loriol,^ however, considers the whole as a centro-dorsal piece,
not having been able to find any trace of sutures separating the upper part from the
lower. It is sometimes found in an isolated condition, while in other cases the radials
still remain in connection with it, and form a perfectly symmetrical whole, no one of
them preponderating in size over the others. They have only been seen in Cotylecrinus
docens, in which they were first figured by Deslongschamps.

As in Holopus and in Steenstrup's Cyathidium, they correspond to the sides
of the pentagon, and the dorsal fossa is greatly reduced. With regard to the so-called
centro-dorsal of Cotylecrinus, I am decidedly of opinion that the upper portion on which
the radials rest represents the basals. The absence of sutures noted by de Loriol is no
proof to the contrary, as we know from the condition of the Palaeozoic Allageci'inus and
Agassizocrinus, and of the recent Rhizocrinus and Bathyciinus.^ This is in fact tacitly
admitted by de Loriol himself in the suggestion that the whole of the calyx tube in
Holopus and Cyathidium consists of anchylosed basals.^

If then the so-called cupule of Cotylecrinus, instead of being a centro-dorsal as its

1 Loc. cit, p. 53. ^ Neues .lalirtucli fiir Mineralogie, 1S57, p. 817.

3 Memoire sur la Couclie a Leptaena, Bull. Soc. Linn, de Normandie, t. iii. p. 181, pi. v. figs. 5, 6.

■* Op. cit., pp. 174, 179. * Loc cit, p. 53.

" Palseontologie, p. 386. ' ' Paleont. Fran{., loc. cit, p. 190.

8 See Ann. and Mag. Nat. Hist, 1883, ser. 5, vol. xi. p. 329. ^ Paleont. Frang., loc. cit., p. 191.

214 THE VOYAGE OF H.M.S. CHALLENGER.

uame implies, really consists, either wholly or partially, of united basals, the composition
of the calyx is the same as in Holopus and Cijathidium, in fact as in most Neocrinoids,
few of which are dicyclic. In all cases of which we have full knowledge, the basals rest
upon something representing a stem, the special organ which is so characteristic of the
Crinoids. It may perhaps be only a central abactinal plate, which becomes transformed
into an expanded disk of attachment, as in the Pentacrinoid larva of Comatula. But I
strongly suspect on general morj^hological grounds that the basal element in the cup of a
Crinoid does not come into dii-ect contact with the supporting surface ; and I have an
equally strong suspicion that it is never entirely absent. The radials of all Crinoids,
excepting the ordinary Comatulse, rest in fossae which are separated by interradial ridges
that mark the median lines of the united basals (PL XX. figs. 2, 3). Such ridges occur
in Cotylecrinus, the raised angles of the upjjer edge of the cupule being interradial in
position ; ^ and I am therefore disposed to agree with Zittel in regarding them as belong-
ing to basals, the lower limits of which are as yet unknown.

There is another character, besides the symmetrical radials, in which Cotylecrinus
resembles Cyathidium. This is the association of two or more individuals in a manner
suggestive of a process of budding, both internal and external. Steenstrup noticed this
pepuliarity in Cyathidium, and de Loriol has described it in Cotylecrinus miliaris} This
last species is further remarkable from the fact that the outer surface both of the cup and
of the second radials found associated with it, " est couverte de petites pustules, tantot un
peu ecartees, tantot, au contraire, tres serrdes." These at once retail the blunt tubercles
of Holopus (Pis. L, IL).

Cotylecrinus, Cyathidimn, and Holopus are evidently very closely alhed, though the
two former differ from the latter in the symmetry of their radials.

In the remarkable form, which after having been referred to Plicatocrinus and also
to Eugeniacrinus, has been made the type of a new genus Eudesicriniis by de Loriol,
there are, however, five asymmetrical radials. These rest " sur uue base large, assez
elevee, adherente par un epatement anx corps soumarins. Cette base n'est point une
piece centro-dorsale semblable a celle des Cotylecrinus, renfermant les parties molles de
I'animal, c'est un simple support." The radials of this type are much higher than those
of Cotylecrinus, and enclose a less extensive space in the centre of the funnel which they
form by their apposition. It narrows considerably below, however, and I question very
much whether it contained more than quite a small portion, if any, of the digestive
canal. But according to de LorioP " c'est done dans la cavite formee j)ar les pieces radiales
que se trouvaient logees les parties molles de I'animal." I suspect, however, that the greater
part of the visceral mass lay above the surface of the cup, and was protected by the two
outer radials and lower arm-joints, just as in Rhizocrinus and Bathyco-inus (PI. VII.
figs. 2, 3 ; PL IX. figs. 1, 3 ; PL X. fig. 20).

1 Palcout. Franj., loc. cit., yl 20, figs. 1, 2, 5, G. - Ihid., p. 204, pi. 19, tigs. 15, 17. ^ jjjy.^ p. gg.

EEPORT ON THE CRINOIDEA. 215

With regard to the support beneath the radial pentagon of Eudesicrinus, I see no,
reason for doubting that the upper part consists of united basals. In fact, one of de
Loriol's figures* shows a distinct horizontal suturalline crossing the middle of the support,
and separating the infra-radial portion with interradial crests on its upper surface from
the more spreading, attached part below. In another specimen this suture seems to be
indicated by an external circular ridge ; but the upper face of the support is marked by-
five petaloid depressions, one of them considerably larger than the rest, which surround
a central pit. De Loriol, and, I think, rightly so, regards these depressions as corresponding
to the cavities of the chambered organ, the largest being that of the large radial in the
trivium. Similar but more regular depressions appear round the middle of the upper
surface of the large basal pentagon in Apiocrinus milleri, Quenstedt ; and they are also
shown in de Loriol's figure af Ajnocrinus roissyanus,^ while it will be remembered that
the chambered organ is invariably in close relation with the basals (PL Vllb. figs. 1,2;
PL XXIV. figs. 6, 7 ; PL LVIII. figs. 1, 3—ch; PL LXL). There can then, I think, be
no doubt as to the presence of basals in Eud-esicrinus, so that the so-called support does
not in reality difi"er essentially from the centro-dorsal of Cotylecrinus. The radials, how-
ever, are very different in the two cases. Those of Cotylecrinus are equal and similar ;
but in Eudesicrinus they are thus described by de Loriol,' " Ces pieces sont fort inegales ;
I'une est notablement plus longue et plus large que les autres, convexe et un pen arqude
en dehors, mais d'une maniere uniforme ; les deux qui la touchent, bien plus courtes et plus
etroites qu'elle-meme, ne sont guere plus larges, mais plus longues que les deux autres ;
ces dernieres, qui sont les plus courtes et placees vis-k-vis de la plus large, s'arquent en dedans
et se coudent un peu vers leur bord superieur. Dans les echantillons frais la surface externe
est couverte de granules epars, ecartes, plus ou moins gros et plus ou moins serrds."

Here then we have a type which bears an unusually close resemblance to Holoims.
The calyx is higher on oiie side than on the other o-ndng to the inequality of the radials,
the central one of the trivium being the largest, while the outer surface, not only of the
radials, but also of the arm-j oints, is coarsely granular or tul^ercular. Eudesicrinus, however,
difi"ers from Holopus in one or two minor points. The fossse on the radials which lodged
the dorsal ligaments are larger ; and there seems to have been a true muscular joint be-
tween the second and the axillary radials, a character presented by no recent Crinoid, and
also, so far as I know, by no other fossil species. In Eugeniacrinus may alls from the
same horizon the two joints are united by syzygy, just as I believe to be the case in
Holopus, while the calyx is less coarsely granular. These two species are quite small
relatively to Holopus, the radials of Eudesicrinus not reaching a height of more than
1 1 mm. Associated with them in the Xe29tonct-bed (Middle to Upper Lias) of Calvados, in
Normandy, are some wedgershaped arm-j oints with the muscle-plates on their broad outer
sides produced into strong upward processes.* These joints have very much the appear-

i Paleont. Frang., loc. cit., pi. 29, fig. 7. - Hid., pL 44, fig. 2, h. ^ Ibid, p. 78. » Ihid, pi. 11, figs. 1-4.

216 THE VOYAGE OP H.M.S. CHALLENGER.

^uce of the more wcdge-sliaped brachials of a liolojnis-avm (PL III. figs. 10-12), but differ
in having the pinnule-socket at the base of the lateral process instead of on its upper edo-e.
Before the discovery of the support below the radials de Loriol considered Eudesi-
crinus to be a species of Eugeniacrinus ; and he still regards it as a member of the family
Eugeniacrinidse, to which he has also thought of transferring Cotylecrinus, though he
has never actually done so. This is partly due to his having been led to regard the
calyx-tube of IIolopus and Cyathidium as possibly composed of the five basal pieces
only/ though there are very serious objections to this view. We know also that the
Eugeniacrinidse, i.e., Eugeniacrinus, Phyllocrinus, and Tetracrinus, have a jointed stem,
which is not the case either in Eudesicrinus or in Cotylecrinus. Both these genera seem
to me to find their proper place in the family Holopidse, which I should characterise as
follows — Basals and radials closely united into a more or less tubular calyx of variable
depth. It is sessile and attached by a somewhat spreading base, the foundation of which
is probably formed by a dorsocentral plate, like that of Marsupites. Ten simple arms,
composed of a small number of massive joints.

A. Radials high but asymmetrical, exhibiting a difference of bivium and trivium.

a. Radials fused together with basals into a tubular body-chamber lodging the -viscera.

A syzygy between the two outer radials, . . . . .1. Holoptts.

/3. Visceral mass was probably lodged above the radials, which are mostly found
separated from the subjacent basals and the spreading base of attaehmeut.
A muscular joint between the two outer radials, . . . .2. Eudesicrinus.

B. Radials apparently all alike. Two or more calyces sometimes associated as if budding.

a. Radials and basals fused into a tubular body-chamber, . . . .3. Cyathidium.

j3. Radials low, and readily separated from the basals and disk of attachment, . 4. Cotylecrinus.

The remarkable Jurassic fossil, described by de Loriol as Gymnocrinus,^ is still too
imperfectly known to be placed in this family ; but I cannot help suspecting that it is
only a portion of the cup of a larger Crinoid. On the other hand, Micvo])ocrinus
gastaldii, described by Michelin^ from the Miocene of Superga near Turin, seems to be
closely allied to Holopus. Michelin's diagnosis runs as follows : " Radix expansa, non
ramosa, adhaerens, sublsevis ; corpus breve crassum, rotundatum, subpentagonale, exterius
granulosum, interius profundum, irregulariter vacuum ; margine revoluto in decern
segmentis acutis subdiviso." I am somewhat puzzled as to the identity of the ten
marginal segments. I do not think that they can represent the individual muscle-plates,
of which there would be ten in a decalcified calyx ; nor does it seem likely that
Micropocrinus is a ten-rayed type like Promachocrinus {ante, pp. 37, 38). The real
nature of this Crinoid must therefore remain undecided for the present.

On the other hand, the Palaeozoic Edriocrinus, which has been described by Hall

1 Paleont. Frang., he. cit., p. 191. ^ Hid., p. 209,

2 Description d'un nouveau genre de la Faniille de Crinoides, Eev. et Mag. ZooL, ser. 2, t. iii. p. 93.

REPORT ON THE CRINOIDEA. 217

from the Upper Silurian of North America, appears to be very closely allied, to the recent
Holopus and to Cotylecrinus. According to Meek and Worthen/ it seems to differ from
this last type " only in having an anal piece on the same range with the first radials, the
relations between the two groups being exactly the same as between Hexacrinus and
Platijcnnus." These two genera, however, are both Palaeocrinoids ; but Belemnocrinus
and RMzocrimis, a Palseocrinoid and a Neocrinoid respectively, are related in precisely
the same way. The former, like Edriocri7Xus, has an anal plate in line with the radials ;
while there is no such structure in Rliizocrinus nor iu Cotylecrimis. The arms of
Edriocrmus are more numerous than those of Holopus, as there are &«econdary axillaries
beyond those in the radial series ; but they were rolled in on one another very much after
the manner of the Holopus-Sixms; and. this was also the case in the Dewomsji. Lecanocrinus
roemeri, Schultze. The latter type has a stem ; but this organ appears to have been
altogether absent in Edriocrinus, which is thus described by Hall:- — "These Crinoids are
sessile in the young state, adhering singly or in groups to other substances until fully
developed, when they are separated from the foreign bodies, and gradually secreting
calcareous matter to cover the cicatrix or point of adhesion, become finally smooth
rounded bases." Elsewhere ^ again he described the radial plates as proceeding from this
" short pedicle " as from the summit of a column. According to Wachsmuth and
Springer* this pedicle really consists of five closely anchylosed, basal plates, with the
sutures between them obliterated by a secondary calcareous deposit, W'hich eventually
removed all traces of the scar denoting the previous attached condition of the individual.
The Mesozoic and recent Holopidse do not seem ever to have passed into the "free"
condition characteristic of Ag(issizocrinus and Edriocrinus, so that there is no scar of
attachment to be obliterated. But I strongly suspect that the subradial portion of the
body, centro-dorsal, support, cupule, or whatever it be called, consists either wholly or
(more probably) in great part of anchylosed basals, just as it does in the Palaeozoic
Edriocrinus.^ For I find it diflicult to believe in the existence of a family of Crinoids
which are normally devoid of any basal plates, as these are of fundamental importance
both in the morphology of the Crinoids, and in that of Echinoderms generally.

Family Hyocrinid^, P. H. Carpenter, 1884.

Genus Hyocrinus,^ Wyville Thomson, 1876.

Definition.— Caljx high, and composed of basals and radials which are nearly equal in
length. The former narrow gradually downw^ards, while the latter are broad and spade-like,
each bearing a small undivided arm in the middle of its upper edge. Arm -joints united

1 Palaeontology of Illinois, vol. iii. p. 371. - Natural History of New York, Pateontology, vol. iii. p."120.

3 Ibid., p. 143. * Eevision, part i. p. 21. ^ Ann. and Mag. Nat. Hist., ser. 5, vol. xi., 1883, pp. 327-334

^ Named after Hog Island, one of the Crozets.
(ZOOU CHALL. EXP. — PART SXXII. — 1884.) li 28

218

THE VOYAGE OF H.M.S. CHALLENGER.

by syzygy into groups of two or three, only the terminal joints of which bear pinnules.
Lowest pinnules the longest, and the following ones proportionately shorter, so that they
all terminate on the same level as the arm-ends. JMouth protected by five large oral
plates. Stem composed of short, cylindrical joints with simple or slightly striated faces.
Mode of attachment unknown.

A. General Account of the Type.

Hyocrinus bethellianus, Wyville Thomson, 1876 (PI. Vc. figs. 4-10; PI. VL).

Hyocrimtshethellianus, Wyv. Thorns., Journ. Linn. Soc. Load. (Zool.) (1876), 1878, vol. xiii. p. 51.
Hyocrimts hethellianus, Wyv. Thorns., The Atlantic, 1877, vol. ii. pp. 96-99.

Dimensions.

Total length of calyx and arms {fide C. W. T. ),

Total height of calyx.

Total diameter of calyx,

Height of radial,

Width of radial,

Height of oral plates,

Length of first pinnule.

Diameter of arm-joints.

Greatest height of stem-joints.

Diameter of stem-joints,

60-00 mm.

7-25

6-00

4-00

3-00

2-75
31-00

1-00

1-15

1-25

The stem is rigid, and consists of short, cylindrical joints, usually a trifle higher than
wide, and closely united by thin disks of ligamentous fibres (PI. Vc. fig. 5, Is). The
terminal faces of the joints (PI. Vc. fig. 4) are slightly hollowed, and either plain or
marked with indistinct radiating striae. The opening of the central canal is more or less
definitely stellate, and in the substance of each joint itself there is a considerable space
(PI. Vc. fig. 5, rs) around the central axis [ca). The longest portion of the stem obtained
was about 170 mm. in length, but its mode of attachment is not known. Towards the
upper end the joints become much shorter, and in the uppermost 5 mm. they are mere
disks with a slightly increased width (PI. VI. fig. 3). The cup, which enlarges gradually
upwards, consists of two tiers of very thin plates, the basals and radials, the latter being
rather the higher of the two. The basiradial and the five interradial sutures are fairly
distinct, but neither in the specimen represented in PI. VI. nor in a fragment from the
same locality, can I make out more than three interbasal sutures. Were it not that
this seems to be the case in both specimens, I should be inclined to regard it as unim-
portant; but under the circumstances I think we must consider that the lower part of
the cup consists of two larger pieces and one smaller one, as in certain Palaeocrinoids. The
small single basal is the one immediately to the right of the anus, i.e., in the interradius
beyond the anus to a watch-hand, when the disk is placed upwai'ds. The remainder of

REPORT ON THE CRINOIDEA. 219

the cup and the arms cannot be better described than in the words of Sir Wyville
Thomson/ "The second tier consists of five radials, which are thin, broad, and spade-
shaped, with a slight blunt ridge running up the centre and ending in a narrow
articulating surface for an almost cylindrical first brachial.

"The arms are five in number, they consist of long cylindrical joints deeply grooved
within, and intersected by syzygial junctions. The first three joints in each arm consist
each of two parts separated by a syzygy; the third joint bears at its distal end an arti-
culating facet from which a pinnule springs. The fourth arm-joint is intersected by two
syzygies, and thus consists of three parts; and so do all the succeeding joints; and each
joint gives ofi" a pinnule from its distal end, the pinnules arising from either side of the
arm alternately. The proximal pinnules are very long, running on nearly to the end of
the arm; and the succeeding pinnules are gradually shorter, all of them, however, running
out nearly to the end of the arm, so that distally the ends of the five arms and the ends
of all the pinnules meet nearly on a level." In all cases the first pinnule is on the left
side of the arm. I can say nothing as to the total number of pinnules, the longest arm
remaining having six of these appendages on eacli side. Owing to the large size of the
pinnules in comparison with the arms, the epizygal joints to which they are articulated
have the appearance rather of axillaries than of ordinary brachials. This is also the case
in Khizocrinus, but to a less extent (PL IX. figs. 4, 5). But as these appendages are
simple and contain the genital glands like the pinnules of other Crinoids, they are
undoubtedly of that nature, and must not be regarded as branches of the arms.

The mouth is protected by a very perfect, five-sided pyramid of triangular oral plates,
the outer surfaces of which are deeply hollowed along the median line (PI. VI, figs. 1-4),
while the inner surface slopes away rapidly on either side from a strong central keel
(PL VI. fig. 5). Sir Wyville Thomson described it as marked with deep impressions for
the insertion of muscles ; but I believe him to have been mistaken in this point. There
is no trace whatever of any such muscles l)eing attached to the inner surface of the oral
plates in the mutilated specimen represented in fig. 5 ; while the orals of Rhizocrinus and
of the Pentacrinoid larva of Comatula are certainly not so provided with muscles, and
there are no a priori reasons whatever for invoking their presence in Hyocrinus.

About half the diameter of the disk is occupied by the oral pyi'amid which covers up
the central mouth. Between its base and the edge of the cup there is a pavement of
closely set, thin plates belonging to the anambulacral system, which have no regularity either
of form or of arrangement. Some of these extend upwards on to the anal tube, which is
situated near the edge of the disk in one of the interradial spaces. As in Rhizocrinus
the oral plates are pierced by the ciliated water-pores which lead downwards into the
body-cavity (PL Vc. fig. 6, ivp). But the pores are more numerous than in Rhizocnnus,
which has only one in each oral plate. In both the specimens of Hyocrinus which I have

^Journ. Linn. Soc Lond. (ZooL), vol. xiii. p. 52.

220 THE VOYAGE OF H.M.S. CHALLENGER.

oxamincd tlicre arc two pores in the oral plate of the anal intorradius, and there are no
others in any of tlio anfimbnlaoral plates which lie between it and the edges of the radials.
The remaining oral plates, however, are not invariably pierced by the water-pores, as
pointed out in Chapter VI. {ante, p. 95).

Tlic five ambulacra which radiate outwards from thu inoutli are protected as soon as
llicy Iiave passed through the angles of the oral pyramid by a very complete armour of
calcareous plates (PI. VI. fig. 8). This seems, as in most recent Crinoid.s, to be less
coniph'tcly differentiated on tlie arms than on the j)innulc;s. In the wider, basal and
middle portions of tlie pinnules which contain the fusiform genital glands, every
jiinnule-joint supports two or three quadrate side plates upon each side (PI. Vc.
figs. 9, 10, sp). Upon each of the side plates rests one of the covering plates (cp), which
overlap (me another alternately from opposite sides. There are no large side plates,
however, in the narrow, proximal portion of the jiinnule before the genital cord swells
out into the fusiform genital gland ; ])ut the covering plates are separated from 'the
pinnule-joints by a number of small irregular plates which belong to the anambulacrai
system (PI. Vc. fig. 10). Towards the distal end of the pinnule, on the other hand, the
covering plates Test directly upon the edges of the pinnule-joints (PI. Vc. figs. 8, 9 ;
PI. VI. fig. (')), as is the case throughout the entire length of the ambulacrum in
Bathicrinus and Rhizocrinm (PI. Vfl. fig. 7; PI. VIII. figs. 4, 5; PI. IX. figs. 2, 4).
They are of a slightly oval shape, and may be as much as 0"6 mm. in diameter.

The genital glands are long and fusiform, and give a swollen appearance to the lower
portions of the pinnules (PI. VI. fig. l). This lasts for about six or seven of the
elongated joints, after which the pinnules taper away slowly to their extremities, while
the glands themselves are continued onwards for some little distance as delicate cords
^^llil■ll (ifYen have a somewhat imdulating course, and gradually diminish in size until
they are no longer traceal)le (PI. Vc. fig. 8, i). The axial <xird of the .skeleton (a) is also
thrown more or less into curves. The specimen obtained was a male, and the testes have
lost all trace of histological structure, as seems to be not unfrequcntly the case with these
glands in other species of Crinoids. They fill up the cavity of the pinnxde almost entirely
(I'l. \c. fig. 7, t). A reduced ca>liac canal (of) with its ciliated cups [cic) being visible
below the gland in transverse section; while a small subtentacular canal {stc) intervenes
between it and the water-vessel. Tlie amliulacral nerve and blood-vessel, however, could
not be detected, though there can of course be no possible doubt as to their presence.

In one of the fragments ' which was obtained, the interior of the oral pp'amid is
ex]>osed (PI. VI. fig. 5). There is a ring of tentacles around the mouth, and, so far as can
be judged from the condition of the specimen, there .seem to be four of these on either
side of the strong lui'dinn keel of eaeb oral plate, so that there would be forty in flU.

' Mr. Blnek did not st'o the original of this figiuv, luit siiuidy copied the woodcut, drawn in fho first instance by Mr.
Wild, which iipiH'arod in thv Joxim. Linn. Soc. Land. Zool.) vol. xiii. p. 04, and subsequently in The Atlantic vol. ii. p. 96.

REPORT ON THE CRITSTOIDEA. 221

The mouth leads into a funnel-shaped gullet, the internal epithelial lining of which is
raised into strong glandular ridges, as is the case all iilong the first part of the digestive
tract. There is no stomachic dilatation, but the tubular lower portion of the oesophageal
funnel is continued into a narrow intestine which forms one simple loop, and turns
upwards again to end in tlic anal opening, as shown in the riglit hand portion of
PL VI. fig. 5. The body-cavity is occupied by the usual loose network of connective
tissue, with dark brown granules dispersed through it in abundance. It is not
strengthened, however, by any of the calcareous rods and plates which are so often found
in a similar position in other Crinoids. Careful search also reveals the presence of
visceral blood-vessels interpenetrating its meshes ; but I have not been able to discover
satisfactory evidence of any water-tubes depending from the oral ring into the body-
cavity, although these organs must certainly be present, and are probably numerous, like
the water-pores on the dislc. The colour of the spirit-specimen is a light yellawish-white.

Localities.— Station 106. August 25, 1873 ; lat. 1° 47' N., long. 24° 26 'W. ; 1850
fathoms; Globigerina ooze; bottom tem2:)erature, 36°'6 F. {l°'8 C). Stem-fragments only
{Jide C. W. T.).

Station 147. December 30, 1873; west of the Crozets ; lat. 46° 16' S., long. 48° 27' E.;
1600 fathoms; Diatom ooze; bottom -temperature, 34°'2 F. (0°"8 C). "One or two
complete specimens and several fragmentary portions" (Jide C. W. T.).

The stem-fragments from Station 106 seem to have been mislaid; as neither Mr.
Murray nor I have been able to find them in the collection of Crinoids which was in
Sir Wyville's hands at the time of his death, and was subsequently sent on to me. But
the characters of the stem are unmistakable, being utterly difierent from those of either
Bathycrinus or Rhizocrinus ; and unless the fragments in question belonged to a new
generic type altogether, which seems improbable, I see no reason for doubting Sir
Wyville's identification of them with the Hyoco'inus which he di-edged four months later
at Station 147. The point is one of some interest as regards distribution, for Station
106 is in the Mid Atlantic, just north of the Equator, while No. 147 is in the Southern
Ocean, 30 miles to the westward of the Crozet group.

The mention of " one or two complete specimens " said to have been obtained at this
Station is unfortunately somewhat ambiguous. The original of figs. 1-4 on PI. VI.
seems to have been obtained in a fiiirly complete condition, lacking, however, the
attached portion of the stem ; but I am sorry to say that the stem and the head have
since parted company. Another stem-fragment was sent me with portions of the thin
basals still in connection with its upper end, and also two fragments of the disk, one with
four of the oral plates (PI. VI. fig. 5), and the other with the remaining plate (PL Vc.
fig. G). These were mounted in balsam by the late Dr. von WiUemoes Suhm, and doubtless
belong to the stem-fragment' just mentioned, but the greater part of the cup is wanting.

Su" W}^'ille concluded his first account of Hyocrinus by a reference to a specimen

222 THE VOYAGE OF H.M.S. CHALLENGER.

which he provisionally named Hyocrinus hethellianus? with the following remarks : —
" The last is a beautiful little thing which we dredged from a depth of 2325 fathoms at
Station 223, in lat. 5' 31' N., long. 145° 13' E., in the east Pacific, with a bottom of
Globigerina ooze, and a bottom-temperature of 1°*2 C. It certainly is in many respects
very unlike the adult Hyocrinus hethellianus; but it may possibly turn out to be the
young of that species. There was only one specimen." ^ No reference whatever was
made to this type in the description of Hyocrinus which was subsequently published in
The Atlantic, and is substantially the same as that which appeared in the Journal of the
Linnean Society. One would be inclined to conclude from this that the specimen in
question was not a young Hyocrinus after all ; for even though it was obtained in the
Pacific, reference would probably have been made to it in Sir W}"^'ille's later account of
this very interesting genus. But as the specimen has totally disaj^peared, and has eluded
all Mr. Murray's anxious search, I am naturally unable to say anything about it.

B. On the Systematic Position of Hyocrinus.

Hyocrinus was established by Sir W3^viUe Thomson in the year 1876,^ with the
remark that " it presents certain general resemblances and even certain correspondences
in structure which seem to associate it also with Rhizocrinus. There seems little doubt
that Rhizocrinus finds its nearest known ally in the Chalk and Tertiary Bouryueticrinus,
and that it must be referred to the neighbourhood of the Apiocrinidse. Were it not that
Bathycrinus and Hyocrinus are so evidently related to Rhizocrinus, the characters of the
Apiocrinidse are so obscure in the two first-named genera that one would certainly have
scarcely been inclined to associate them with that group." Bathycrinus, though an
aberrant form, is far more closely related to Rhizocrinus than Hyocrinus is. It has the
same form of stem-joint and the same absence of pinnules from the arm-bases ; while the
arm-joints themselves are united in pairs in a very nearly similar manner in both genera.
But except in this last point, there is no resemblance between Rhizocrinus and Hyocrinus.
The only knouTi species of the latter genus was said by Sir WyviUe Thomson to have
" much the appearance, and in some prominent particulars it seems to have very much
the structure, of the Palseozoic genus Platycnnus, or its subgenus Dichocrinus."^ In
fact, Sir Wjrville seems to have had considerable hesitation in referring Hyocrinus to the
Apiocrinidse ; and it was eventually associated by Zittel along with PlicaAocrinus, in a
family Plicatocrinidse. But the definition which he gave of the family was far from being
a satisfactory one, as it stated that basals were absent, which is by no means the case in
Hyocrinus, and also that there are long, forked arms. Since then, however, he has found ■
that there is an axillary second radial (first brachial, Zittel) in PlicaAocrimis, which thus

1 Journ. Linn. Soc. Land. (Zool.), vol. xiii. p. 55. ^ Ihid., p. 48. ^ Ihid., p. 51.

REPORT ON THE CRINOIDEA,. 223

has twelve (ten) arms, instead of five, as in Ui/oorinus. These arms are composed of
short, stiff joints in which no syzygial unions occur; while they bear short pinnules,
all of which, except the first four, consist merely of one elongated joint. Zittel further
says,^ " Aus dem Vorhergesagtcn geht hervor, dass Hyocrinns in Bezug auf den
Bau der Arme einen differenzirteren Typus darstellt als PUcatocrimis. Immerhin aber
stimmen beide Gattungen hinsichtlich ihres Kelchbaues besser mit einander liberein, als
mit irgend einer anderen bis jetzt bekannten Crinoideen-Genus und durften darum wohl
derselben Familie zugetheilt bleiben." It appears to me, however, that this supposed
resemblance between Hyocrinus and Plicatocrinus is really very superficial ; and that it
consists essentially in the condition of the thin and somewhat flattened calyx-plates.
This is also the case with the radials of Bathycrinus, while the calyces of young Penta-
crinidse have a very considerable similarity to that of Plicatocrinus. On the other hand,
and apart from the question of basals, the arms of Plicatocrinus, as discovered and
described by Zittel himself, are utterly and entirely different from those of Hyocrinus ;
and although de Loriol says, " Les analogies tendrent h, montrer que les deux genres sont
dela meme familie,"^ he concludes as follows, " il faudra peut-etre etablir une familie pour
chacun de ces genres." This I propose to do in the case of Hyocrinus, the definition of
the family Hyocrinid^e being for the present the same as that given above for the genus.

While resembling Apiocrinus and also many Palseocrinoids in the nature of the stem-
joints, Hyocrinus differs in several respects from the other Neocrinoids. In the first
place the apparent presence of only three basals and the small size of the articular facets
as compared with the great breadth of the radials, give it a strong resemblance to some
of the Pateocrinoids, and more especially to the Platycrinidfe. Although Hyocrinus
resembles Platycrinus in having a symmetrical, tripartite base, the position of the dorsal
axis'" which divides the base symmetrically is not the same in the two genera. If a
Platycrinus be " orientirt " with the anal interradius posterior, the dorsal axis runs
from the right anterior interradius to the left posterior radius ; whereas that of
Hyocrinus (in the only specimen examined) runs from the left anterior radius to the
right posterior interradius. But the general form of the calyx, as seen from the side
(PL VI. fig. 3), is very like that of the Carboniferous Dichocrinus intermedivs, figured
by de Koninck.* Its composition, however, is different, as Dichocrinus only has two
symmetrical basals.

The persistence of the large oral plates is a noteworthy feature of Hyocrinus, but it
finds a parallel in the Comatulid genus Thaumatocrinus (PI. LVL fig. 5), and also to a
certain extent in Rhizocrinus.

1 Ueber Plicatocrinus, Sitzungsb. d. II. CI. k. baier. Akad. d. JViss., 1882, Bd. i. p. 112.

2 Pal^ont. Frang., loc. cit, p. 63.

^ See Beyi'icli, Ueber die Basis der Crinoidea bracliiata, Monatsber. d. k. preuss. Akad. d. IViss. Berlin, 1871,
p. 42.

^ Kecherclies sur les Crinoides du terrain Carboniffere de la Belgique, Bruxelles, 1854, pi. iv. fig. 9.

224 THE VOYAGE OF H.M.S. CHALLENGER.

It is in the character of the arms and pinnules, however, that Hyocrinus is most
remarkable. The syzygial union of successive pairs of arm-joints is characteristic of
Rhizocrinus ; but in Hyoci'inus the third and following joints are triple and not double
only. A similar difference between the arms of Heterocrinus simplex and Heterocrinus
C07istrictus has been already noticed.^

The arrangement of the pinnules of Hyocrinus was described by Sir Wyville Thomson
as " hitherto entirely unknown in recent Crinoids, although we have something very close
to it in some species of the Palaeozoic genera Poteriocrinus and Cyatliocrinus"^ I do
not think, however, that this resemblance is such a very close one after all. For the
lateral appendages of the arms of Hyocrimis, long as they may be, are true pinnules.
Cyathocrinus, on the other hand, has no pinnules whatever, but long branching arms,
each branch bifurcating several times. It is true that the terminations of all the branches
are about on the same level, as is the case with the aims and pinnules of Hyocrinus.
But in the one genus a bifurcation gives rise to two equal arms which divide again,
and in the other there is no bifurcation at all, but the arm-joints bear a series of
pinnules which i-emain perfectly simple throughout their whole length, great though
this may be. It has been ah-eady pointed out that the nearest approach to the
pinnule arrangement of Hyocrimis is to be found in Barycrinus herculeiis from
the Carboniferous series of Indiana, United States {ante, p. 61). The so-called
armlets of this type alternate with one another upon opposite sides of the main
arm-trunk and l^ear no pinnules, so that they seem to correspond somewhat closely
with the pinnules of Hyocriiiits.

The closest approximation among the Neocrinoids to the arrangement of the pinnules
which occurs in Hyocrinus, though stiU differing from it in important points, seems to
me to be found in the Liassic genus Uxtracnnus. In this curious type each arm
consists of a principal trunk bearing pinnules as usual, and giving off at intervals from
its inner side a series of smaller armlets which also bear pinnules. The lowest of these
are as long as the remaining portion of the arm -trunk from which they spring; and
the following ones are of successively diminishing lengths, so that the ends of the
orio-inal arm-trunk and of its numerous armlets are all on about the same level. In this
respect the armlets of Extrcicrinus are comparable to the pinnules of Hyocrinus ; but
they bear pinnules themselves, and only come off from one side of the main arm-trunk,
instead of alternating from opposite sides..

There is, therefore, no exact parallel to the condition of the arms of Hyocrinus
to be found in any Neocrinoid; and remembering this, as well as the peculiarities
of the calyx, we cannot say that Hjocrinus is specially related to any of the
other Neocrinoidea, while it presents important characters which connect it with the
Palaeocrinoids.

' Ante, p. 53. ^ Journ. Linn. Soc. Loud. (Zool.), vol. xiii. jr. 52.

REPORT ON THE CRINOIDEA. 225

Family Bourgueticrinid/E, de Loriol, 1882.

Genus Bathycrinus, Wyville Thomson, 1872; emend. P. H. Carpenter, 1884.
Ilycrinus, Danielssen and Koren, Nyt Mag. f. Naturvidensk., 1877, Bd. xxiii. p. 4.

Definition. — Stem consisting of dicebox-sliaped joints and attached by a branching root,
the joints above which bear no cirri. The upper part of the stem, immediately beneath the
cup, is formed of a large number of thin, discoidal joints. Calyx expanding upwards from
the basals, which are closely united into a thickened, discoidal piece without any visible
sutures, and but slightly wider than the upper stem-joints. First radials trapezoidal, and
united to form a rapidly expanding cup. Second and third (axillary) radials united
by trifascial articulation; the muscle-plates of the axillaries produced upwards into
strong, wing-like processes. Arm-joints (with the exception of the third, sixth, and
ninth) united in pairs by trifascial articulations, only the distal joint of each pair bearing
a pinnule, and there are no pinnules on the first few pairs. Interradial areas of the disk
naked, paved with loose anambulacral plates, or supported by a single oral plate. Am-
bulacra have covering plates, but no side plates.

RemavTcs. — This genus was established by Sir Wyville Thomson in 1872 for a small
immature individual which was dredged at the mouth of the Bay of Biscay by the
"Porcupine " in 1869 from a depth of 2435 fathoms.^ But since the discovery by the
Challenger of adult examples of two much larger species in the Atlantic and Southern
Oceans, the original description of the genus requires modification. One of the Challenger
species {Bathycrinus aldrichianus) was described by Sir Wyville in the Journal of the
Linnean Society for 1876 ; but in the meantime a fourth species was discovered in the
North Atlantic by the Norwegian North Sea Expedition, and it was made the type of a
new genus Ilycrinus by Danielssen and Koren. ^ For it appeared to them to difi"er
chiefly in size and in the presence of pinnules from Bathycrinus, as described by Sir
Wyville from the immature " Porcupine " specimen ; and his amended account of the
genus, founded on the examples dredged by the Challenger, had not reached them in time
for reference.

The nearest ally of Bathycrinus is undoubtedly Rhizocrinus. In fact, without an
acquaintance with this genus, one would hesitate to place Bathycrinus in the neighbour-
hood of the Apiocrinidse at all. There is but a very slight upward expansion of the stem
below the head and even in the ring of basals which rests upon it ; while the characters
of the radials are very difi'erent from those of Rhizocrinus and Bourgueticrinus.

The general character of the dicebox-shaped stem -joints and of the branching root
is essentially the same as in Rhizocrinus. But the modes in which these joints are

'■ Proc. Eoy. Soc. Edin., 1872, vol. vii. p. 772 ; see also The Depths of the Sea, p. 450.
2 Nyt Mag. f. Naturvidensk., Bd. isiii. p. 10.

(ZOOL. OHALL. ESP. PART XXXII. 1884.) li 29

226

THE VOYAGE OF H.M.S. CHALLENGER.

developed from the thin disks that successively appear immediately beneath the calyx
are different in the two genera {ante, pp. 26, 27). There is alwaj^s a large number of
these thin joints at the top of the stem of JBathycrinus (PI. VII. figs. 1-3, 11; PI. Villa,
fig. 1), whereas in Rhizocrinus (PL IX. figs. 1-3 ; PL LIII. figs. 7, 8) there are very
few, often not more than one or two, and these by no means so thin as in Bathycrinus.

An entire stem, or the upper and middle part of one, could therefore be referred
without difiiculty to its proper genus. But the lower and middle joints are so much
alike in the two genera that the pi'oper identification of a fragment or of isolated joints,
either recent or fossil, would become a matter of uncertainty, if not of impossibility.

The genus Bathycrinus was never formally defined by Sir Wyville Thomson ; but in
his first account of it ^ he said that, like Rhizocrinus, it " must also be referred to the
Apiocriuidse, since the lower portion of the head consists of a gradually exjDanding funnel-
shaped piece, which seems to be composed of coalesced upper stem-joints;" and he
nowhere mentioned the presence of any calycular plates below the radials. Subsequently,
however, he stated,^ after examining Bathycrinus aldrichianus, that the stem of this
genus " barely enlarges at its junction with the cup ; " and he described the lower portion
of the latter as consisting of a series of basals which are soldered together into a small
ring, scarcely to be distinguished from the upper stem-joint (PL VII. figs. 1, 2, 11 ;
PL Vila. figs. 12-14 ; PL Vllb. figs. 1, 2).

The existence of basals in Ilycriniis {Bathycrinus) car-
penteri was also recognised by Danielssen and Koren,^ who
were fortunately able to see the interbasal sutures in young
individuals, though these entirely disappear in the adult.

Although invisible on the upper and lower surfaces of the
basal ring of Bathycrinus aldrichianus, as well as externally
(PL Vila. figs. 12-14), the sutures are clearly seen in sections
through its middle portion (PL Vllb. fig. 2). It expands very
slightly from below upwards, and its somewhat hollowed under
surface is marked by ten fossje radiating outwards from the
centre and separated by intervening ridges (PL Vila. fig. 14).
They correspond to similar fossae on the upper face of the thin
top stem-joint (PL Vila. fig. 3), and lodge five strong but short
interradial ligamentous bundles, each having somewhat the
form of a horseshoe or V with thick limbs (woodcut, fig. 11, bl).
These, as already described, unite the basals to the thin, u^aper
stem-joints, and are gradually replaced as the joints become thicker by the two larger bundles
which form cushion-like pads between every two of them (ante, p, 27; PL Vila. figs. 4-6).

Fig. 11. — Diagram of a horizontal
sectiou through the lowest portion
of the basal ring of Bathycrinus
aldrichianus ; x 70. bl, ligaments
uniting the basals to the top
stem-joint ; ch', the outer vessels
in the vascular axis, wliich are
continued downwards from the
chambers of the chambered organ ;
if, interradial portions of the fib-
rillar sheath round the vascular
axis which are separated by rs,
the radial spaces in the upper part
of the stem ; v, central vessel of
the vascular axis.

1 The Depths of the Sea, p. 450.

3 Nyt Mag. f. Naturvidensk., Bd. xxiii. pp. 4, 5.

2 Journ. Linn. Soc. Land. (ZooL), vol. xiii. (1876) 1878, x^p. 48, 50.

EEPORT ON THE CRINOIDEA. 227

Although in sections through the lower part of the basal ring the limits of its
component joints may be traced by the grouping of the five ligaments referred to above,
yet the interbasal sutures do not become clearly visible till the level of the lowest part of
the chambered organ is reached. Here they appear as actual gaps in the otherwise
continuous network of nucleated connective tissue which forms the organic basis of the
skeleton, so that in a stained preparation they are shown as five radiating white lines on
a coloured ground (PI. Vllb. fig. 2). They do not, however, reach the outer edge of the
section where the connective tissue network forms a complete ring, and this accounts for
the absence of any sutural lines upon the exterior of the composite basal piece (PL Vila.
fig. 13). The sutural union between this piece and the stem -joints below it appears to
be closer than that between the basals and radials, so that the head has a considerable
tendency to break away from the stem at the basiradial suture. This was unfortunately
the case with one of the two individuals of Bathycrinus gracilis which were met with by
the "Porcupine's" dredge in 1869, and the head was consequently not brought to the
surface. On the other hand, only the head of Bathycrinus camphellianus is now known
(PL VIII. figs. 1, 2. Woodcut, fig. 15 on p. 239), the stem with the basals having separated
from it ; while Danielssen and Keren figure an isolated head of Bathycrinns carpenteri
which has lost its basals.^ But the most remarkable case of this kind was met with at
Station 146, in the Southern Ocean, where the dredge must have passed over a small forest
of Bathycrinus aldrichianus.

About a dozen tolerably perfect individuals were obtained, together with a consider-
able number of stems retaining the basal ring at their upper ends. This fact is one of no
little importance from the light which it throws on the supposed composition of the
calyx in the fossil genus Eugeniacrinus and its allies Phyllocrinus and Tetracrinus.
These genera are very common in the Jurassic and Lower Cretaceous rocks, especially of
the Continent ; but by far the greater number of calyces which are met with consist of
the radials alone, just like that of Bathycmius camphellianus {¥\. VIII. figs. 1, 2), and
the family has accordingly been described as distinguished by the absence of basals.
De Loriol ^ says, for example, " Le calice est formd de pieces radiales seulement sans pieces
basales." Occasionally, however, a calyx is met with still retaining a portion of the stem
attached to it. But no sign of sutures is visible in what appears to be its uppermost
joint immediately beneath the radials. This joint, with a portion of the stem attached
to it below, is also sometimes met with separate from the radials, as in the case of Bathy-
crinus aldrichianus and Bathycrinus gracilis. But the absence of sutures, as shown by
the condition of an adult Batlujcrinus, is no proof that the piece in question does not
consist of a ring of closely united basals, a point as to which I have no doubt whatever.^

The fibres which efi"ect the synostosis of the basals with the radials above them are

1 Nyt 3Iag.f. Nccitirvidenslc, Bd. xxiii. Tab. i. fig. 6. - Paleont. Franj., loc. cit., p. 74.

3 Ann. and Mag. Nat. Hist., 1883, ser. 5, vol. xi. pp. 327-334.

228 THE VOYAGE OF H.M.S. CHALLENGEE.

rather more uniformly distributed over the apposed surfaces of the joints than those
uniting the basals to the top of the stem. The upper surface of the basal ring presents
five smooth and single sutural fossse for the attachment of the radials. They slope
downwards and outwards from the edge of the central funnel, and each is marked near
its outer edge by a single crescentic opening, or by two smaller ones in close proximity
(PI. Vila. figs. 12, 13). At first sight this more or less double opening would naturally
be taken for the termination of the converging right and left forks of two adjacent
interradial canals within the basal ring. This apparently obvious explanation is, however,
very far from being the true one. The under faces of the radials which rest in these
fossae on the upper surface of the basal ring are marked in the same way by more or less
double openings ; but these are not the openings of the central canals, as the apparently
similar openings are on the under faces of the radials of Pentacrinus and Metacrinus
(PI. XII. figs. 11, 22 ; PL XX. fig. 9 ; PI. L. fig. 5). They are usuaUy quite smaU and
inconspicuous, and not nearly so well defined as the openings of the central canals on the
distal faces (PI. Vila. fig. 15), with which indeed they have no communication, for they
are merely small pits into which portions of the basiradial ligament are inserted ; and the

same is the case with the corresjDondiug openings in the
fossae on the upper surface of the basal ring (PL Vila,
figs. 12, 13). Although the fibres of the basiradial
ligament are generally distriljuted over the whole
synosteal surface, they are more especially concentrated
in ten bundles which are lodged in ten corresponding pits
on each of the apposed surfaces of the basal and radial
circlets. Owing to the curvature of these surfaces,
these more defined bundles are not shown in the section
represented in PL Vllb. fig. 3, which passes rather

f 10. 12. — DiaOTam of a horizontal section , , i • i i j_i i j_i i i r , i

tiirough tiie calyx of Bathycrinus aidric abovc their Icvel turougli the general plane 01 the

hianus, at the level of the upper part of the • i • i ■ i i i

basiradial suture ; x 45. ai, primary inter- S\T10StOSlS ; DUt lU tUC UCXt SCCtlOU tUCy are Clcaiiy
radial cords ; I, parts of the basiradial •' iivi i ■ ^

ligament ; L, interradial ligament uniting ftl, visible, CUt SOmCwhat obliquch^ aS shoWU lU the WOodcUt
the first radials ; x, plexiform gland.

(fig. 12, 1). Both in the woodcut, and in PL Vllb. fig. 3,
the axial cords {ai) are seen to be situated interradially. They retain this position
until they reach about half the height of the radial pentagon, where they fork for the
first time ; and the branches enter the radials by the openings in their lateral faces
(PL VII. fig. Q>a). The right branch of one fork and the left branch of its neighbour
in the adjoining basal occupy converging canals in the intervening radial, which meet
almost directly, so that there is only one opening on the distal face (PL Vila. fig. 15).

The circular commissure of Bathycrinus is thus mainly formed by the actual branches
of the primary interradial cords, and not by special interradial commissures uniting these
branches as in the Comatulae and Pentacrinidse (PL XXIV. fig. 9, ceo). This is also the

REPORT ON THE CRINOIDEA.

229

case in Rlnzocrinus, as pointed out by Ludwig ; ^ but Bathycrinus occupies an inter-
mediate position between RMzocrinus and the types just mentioned, for the two
converging cords within each radial are united by an intraradial commissure just as in
Pentacnnus and Rhizocrinus (PL Villa, fig. 6; PI. XXIV. fig. 9— fco). This does not
appear in the section represented in PI. Vllb. fig. 4, though it is phiinly visible in the
next one.

The distribution of the axial cords proceeding from the chambered organ oi Bathycrinus
is thus of a very singular character (woodcuts, figs. 13, 14). Each of the five primary

Fig. 13. — Diagrammatic vertical section througli the Fio. 14. — Plan of the distribution of the

c.aly.x oi Bathycrinus aldrichianus ; x 35. It is inter- a.xial cords in the caly.x of Bathycrinus

radial on the right side, passing between two of the aldrichianus. B, basals ; ijl Ii2 first

chambers of the chambered organ, and through the and second radials.

sjTiostosis of two radials (which is shaded darker) so
as to show the primary iuterradial cord (ai) ; while on
the left side it passes along the middle line of a ray,
and shows the axial cord (.-1) in the distal part of the
radial (i?'). B, ring of anchylosed basals ; ca, fibrillar
sheath round vascular axis of stem ; ch, one of the
chambers of the quinquelocular organ ; ch', its down-
ward extension into the stem ; c.co, iuterradial portion
of circular commissure ; I, basiradial ligament ; Id,
dorsal ligament ; li, interarticular ligament ; m,
muscle ; v, central vessel of the stem ; x, plexiform
glaud.

iuterradial cords proceeds upwards close to the central axis of the basal ring (woodcut, fig.
13, ai). They pass out of this ring just inside the edge of its central funnel, at the inner
ends of the ridges which separate the fossje lodging the radials and marking the median
lines of the basals. For the remainder of their course each of them is contained in the
more or less complete canal which is formed by the apposition of two grooves, one on each
of two contiguous lateral faces of the radials. These grooves, which run downwards from

^ 0}). cit., Zcitschr.f. wiss. ZooL, 1877, Bd. xxix. p. 72.

230 THE VOYAGE OF H.M.S. CHALLENGER.

the openings in the lateral faces of the radials, but close to their inner edges, are well
shown in PI. VII. fig. 6a.

As in the Comatulse, therefore, the circular commissure of Bathycrinus (PI. Vllb.
fig. 4, ceo) is in the innermost part of the radial pentagon, i.e., quite near its centre.
There is but a thin layer of limestone between it and the central space, while almost the
whole of the fibres formina: the interradial lisfaments are outside it. The length of these
primary interradial cords and their reception in grooves on the apposed surfaces of
contiguous radials is very anomalous ; and although I detected the true nature of these
grooves at first sight, it was nevertheless very long before I could get rid of the notion
that the radial openings on the tojj of the basal ring (PL Vila. figs. 12, 13) were those
of the converging branches of the forked interradial canals, as in other Crinoids ; and it
was not until after some time that I was able to reconcile the apparently conflicting
evidence afforded by the study of series of transverse sections on the one hand and of the
dissected calyx on the other.

From the facts detailed above, it will be seen that Bathycrhms occupies a some-
what anomalous position among Neocrinoids. In Comatula, Pentacrinus, Apiocrinus,
and Encrinus the primary interradial cords fork within the basals ; and the adjacent
branches of neighbouring forks enter the radials by more or less distinctly double
openings on their inner or under faces ; but in Bathycrinus not only do the cords not
fork within the basal ring, but they rise through half the height of the radial pentagon
before doing so (woodcuts, fig. 13, ai ; fig. 14). The nearest approach to this condition is
presented by Rhizocrinus, though the relative proportions of the plates are exactly the
reverse of what we meet with in Bathycrinus. In fact, if we make allowance for this
difi"erence the condition of Bathycrinus, except for the presence of the intraradial com-
missure, is almost exactly that which was described by Ludwig in Rhizocrinus; though,
as pointed out already, the real condition of this genus is slightly more normal, i.e., the
primary cords fork within the basals, and their converging branches enter the inner ends
of the radials as in other Crinoids.

The first radials, which form by their apposition a rapidly expanding cup, have an
elongated, trapezoidal outline and rounded outer surface. According to Sir Wyville
Thomson,^ those of Bathycrinus aldrichianus are " often free ; but in old examples they
also are frequently anchylosed into a funnel-shaped piece." In all specimens of this type
which I have seen, however, the radials are united laterally, just as in other Crinoids ;
though they separate more readily than usual when treated with hot alkalies. The
ligaments uniting them are close and well defined in the lower part of the funnel (PL Vllb.
fig. 4, I); but in the upper part, i.e., just below the level of the articular surface, there is
no interradial ligament (PL Vllb. fig. 5), which probably explains the description that
has just been quoted from Sir Wyville Thomson. The distal articular faces of the radials

' Journ. Linn. Soc. Lond. (Zool.), 1876, vol. xiii. p. 50.

REPORT ON THE CRINOIDEA. 231

are relatively large, but otherwise of the usual character, with paired fossae for the
muscles and interarticular ligaments (PL Vila. fig. 15, rm' and W), and the single one
beneath the articular ridge for the reception of the dorsal ligament (Id'). The second
radials (PI. VII. figs. 5, 5a) are broad, flattened, and somewhat quadrate in form, with
a more or less distinct medio-dorsal convexity (PI. Vllb. fig. 6, R2) and a well-marked
furrow in the middle line of the ventral surface (PI. VII. fig. 5a). The proximal face
resembles that of the first radials, which is high relatively to its width (PI. Vila. fig. 15),
while the articular surface at the distal end is low and much extended laterally. The
axiUaries are more or less pentagonal, with a forking median ridge on the flattened dorsal
surface. The ventral surface is flattened like that of the second radial, with which the
axiUaries articulate without the intervention of muscles. But the lateral margins of the
median ventral furrow rise gradually from the distal to the proximal end of the joint,
where they are produced into more or less expanded, wing-like processes which project
forwards over the first brachials (PI. VII. figs. 4, 4a; PI. Vila. fig. 17). The bases of
these serve for the attachment of the muscles and ligaments which unite the axillaries
and first brachials, while their upper portions support the iuterradial diverticula of the
gut (PL Vllb. fig. 7, E3).

According to Sir Wyville Thomson, the two outer radials of Bathycrinus gracilis and
JBathycrinus aldrichianus are united by syzygy ; while Danielssen and Koren make the
same statement respecting Bathycrinus carpenteri} This is not really the case, how-
ever, and as a matter of fact there are no true syzygies in Bathycrinus aldrichianus
at all ; nor, as I believe, in any species of the genus. The distal face of the second
radial is shown in PL Vila. fig. 16. Apart from its external form, it has a general resem-
blance to the corres ponding face of the second radial in Pentacrinus decorus, Penta-
crinus naresianus (PL XXX. fig. 1 ; PL XXXIV. fig. 6) and the Comatulse, i.e., there
is a vertical articular ridge which separates the two fossae lodging large bundles of
ligament [W). But in Bathycrinus a third and smaller bundle of ligament is inserted
into a deep pit (Id') at the lower or dorsal end of the vertical articular ridge. The
proximal face of the third radial is of the same character ; and in reality the union of
these two joints, instead of being an immovable syzygy, is a modification of the bifascial
articulation permitting lateral movement only, which is so common in the Comatulse, and
is also characteristic of four recent species of Pentacrinus. Externally this form of
articulation looks very much like a syzygy, as the joints are brought into closer connec-
tion than when they are united by a pair of muscular bundles ; but a glance at their
apposed faces is sufiicient to show that the plainness of the syzygies in Pentacrinus or
Rhizocrinus, and the striation so common in the Comatulse, are altogether absent,
being replaced by distinct ridges and fossae. In describing Bathycrinus gracilis, Sir
Wyville Thomson^ pointed out that " the first brachial is united to the second by a

1 Nyt Mag.f. NaturvidensL, BJ. xxiii. p. 6. ^ Proc. Roy. Soc. Edin., 1872, vol. vii. p. 773.

232 THE VOYAGE OF H.M.S. CHALLENGER.

syzygial joint, but after that tlie syzygies are not repeated, so that there is only one of
these peculiar junctions in each arm. . . . And the alternate syzygies in the arms, which
form so remarkable a character in Rhizocrinus, are absent." He subsequently stated that
in Bathycrinus aldrichianus " the first and second, and the fourth and fifth brachials are
united by sj'zygies ; and after that the syzygies occur sparingly and at irregular intervals
alons the arms."^ In like manner Danielssen and Koren'^ described the first two brachials
of Bathycrinus carjpenteri as united by syzygy ; a true articulation between the second
and third, and also between the third and fourth brachials ; while the fourth and the fifth
are united by syzygy, the sixth articulated both to the fifth and to the seventh, and the
ninth both to the eighth and tenth. After this point muscular articulations and syzygies
alternate all along the arms. Owing to the kindness of Dr. Danielssen I have been able
to satisfy myself that these " syzj^gial unions " in the arms of Bathycrinus carpenteri are
really trifascial articulations like that between the two outer radials of Bathycrinus
aldrichianus. If, however, this term be substituted for syzygy in the descrijation by
Danielssen and Koren, their statements respecting the grouping of the brachials would be
perfectly correct, i.e., in the nine lowest brachials there are alternations of a pair of joints
united by trifascial articulation and a single joint with muscular articulations at both ends.
Beyond the ninth brachial the two forms of articulation alternate with great regularity.
Apart from the question of nomenclature, therefore, the Norwegian naturahsts were the
first who correctly described the grouping of the joints in the arms of Bathycrinus ; for
I find that their description of Bathycrinus carpenteri applies both to the little Bathy-
crinus gracilis dredged by the " Porcupine " and to the two Challenger species, Bathy-
crinus aldrichianus (PL VII. figs. 1, 2) and Bathycrinus campbelUanus (PL VIII.
figs. 1, 3). The non-syzygial nature of the paired unions in the arms of the two last has
been determined by actual investigation of the joint faces ; while careful microscopic
examination of the small individual of Bathycrinus gracilis has convinced me that it
resembles the ^ther three species in this respect.

The proximal face of the fifth brachial of Bathycrinus aldrichianus is shown in
PL Vila. fig. 19 ; and that of a later joint in fig. 20. In both cases the three fossae are
visible around the opening of the central canal ; while in PL Vllb. fig. 8, the three
ligamentous bundles uniting the first and second brachials are seen in section (//, /(/).
The same three fossae {li', Id') appear in PL Vila. fig. 22, which represents a trifascial
articular face of Bathycrinus camphellianus; while an ordinary muscular joint-face at the
proximal end of a brachial is seen in fig. 23.

The presence of this trifascial articulation, and its peculiar grouping, may therefore be
considered as distinctive of Bathycrinus. Hence the alternation which was referred to by
Sir Wyville as so characteristic of Rhizocrinus is repeated in Bathycrinus, though w^th
two points of difference. In the latter genus the syzygies of Rhizocrinus are replaced by

^ Journ. Linn. Soc. Land. (Zool.), 1876, vol. xiii. p. 50. ' jVj/i Mag.f. Natiirvidensk., Bd. xxiii. pp. 6, 7. ,

REPORT ON THE CRINOIDEA. 233

trifascial articulations; while the third, sixth, and ninth brachials are usually free, i.e.,
united by muscles to the joints before and behind them, as the joints of a Crinoid arm
gendi-ally are. In Rhizocrinus, however, the two joints of each successive pair are united by
syzygy, so that muscular articulations and syzygies alternate regularly all along the arm.
But as regards the last point Bathycrinus approaches Rhizocrinus much more closely than
was supposed by Sir Wyville Thomson. Instead of there being one syzygy (trifascial
articulation) only, or two at the base and others scattered sparingly at irregular intervals,
there is just as much regular alternation after the ninth brachial as there is throughout
the whole arm of Rhizocrinus. In fact, in some irregular arms of Bathycrinus the third
brachial is the only joint which has muscles attached at both ends, the alternation which
would ordinarily commence with the tenth appearing on the fourth and following joints.

Except in Bathycrinus aldrichianus the bases of the arms are but little wider than
the succeeding portions. The flattening of the dorsal surface at the sides of the median
ridge, which commences on the two outer radials, is continued on to the first two or
three brachials and then disappears (PI. VIII). figs. 7, 8). The same is the case with the
sharp lateral edge which is so marked on the second and axillary radials. In Bathycrinus
aldrichianus it is continued out on to the first eight arm-joints (PI. VII. fig. 2), and
marks the line of junction of the curved dorsal surface and the side faces, which slope
upwards and inwards towards the medio-ventral line (PL VII. fig. 8). An arm of this
species, 30 mm. long, consists of fifty joints ; but the first seven or more bear no pinnules.
In most cases the first pinnule is borne on the eleventh joint, when this, as is normally
the case, has a muscular articulation at its distal end. The ninth brachial is usually a
joint of this kind, and in Bathycrinus cddrichianus and Bathycrinus campbelUanus some-
times bears the first pinnule, while the eighth may do so. But the joint which bears a
pinnule is invariably united to its successor by muscles, so that the pinnules only occur
upon every alternate joint through the whole length of the arm, exactly as in Rhizocrinus.
The only difference is that the joints which do not bear pinnules are united to those
which do by syzygies in Rhizocrinus, and by trifascial articulation in Bathycrinus. The
distal face of a pinnule-bearing joint of Bathycrinus aldrichianus is shown in PL Vila,
fig. 21; and a corresponding joint-face of Bathycrinus campbellicmus in fig. 23. In the
latter type the pinnide-socket is more at the side of the joint and less at the end than in
Bathycrinus aldrichianus.

The pinnules of Bathycrinus (PL VII. fig. 7 ; PL VIII. fig. 5) are moderately short
and slender, and composed of relatively few joints. The lateral edges of these joints,
especially the outer ones, are produced upwards at the sides of the medio-ventral furrow
so as to protect it very completely. This is most marked in Bathycrinus camiiibellianus
(PL VIII. fig. 5) and to a less degree in Bathycrinus aldrichianus (PL VII. figs. 2, 7).
It is also figured by Danielsson and Koren^ in Bathycrinus carpenteri. The first six or

1 Nyt May. f. Naturvidemk., Bd. xxiiL, Tab. L fig. 14.

(ZOOL. CHALL. EXP. — PART XXXII. — 1884.) li 30

234 THE VOYAGE OF H.M.S. CHALLENGER.

eight pinnules which contain the short fusiform genital glands show this feature most
distinctly. Somewhat the same character presents itself in the enlarged lower portions
of the long pinnules of Hyocrinus. But in this case the plate-like sides of the piunule-
joints diverge considerably, so that the median groove is widened instead of narrowed as
in Bathycrinus. It is therefore partially roofed in by side plates (PI. Vc. figs. 9, 10),
of which there is no trace in Bathycrinus. But the plate-like sides of the joints bound a
narrow ventral furrow, just as in the arms, and the covering plates rest directly upon
their edges (PI. VII. figs. 7, 8 ; PI. VIII. figs. 3, 5), as is the case in Rhizocrinus and in
the outer parts of the pinnules of Hyocrinus (PL Vc. fig. 9 ; PL IX. fig. 4 ; PL X. fig. 20).
They are continued down the sides of the arm-grooves on to the disk, and in Bathycrinus
cddrichianus, which has no orals, they stand up all round the edge of the peristome, as is
represented, though badly, in PL VII. fig. 3. These covering plates are scarcely so
substantial as the corresponding plates in Hyocrinus, and are also narrower, though nearly
as long.

The two lower brachials and part of the third assist in the protection of the visceral
mass, the upper surface of which is more or less strongly convex (PL VII. fig. 3). There
is a very large, funnel-shaped peristomial opening, at the bottom of which is the mouth,
and the anus is on a low papilla in one of the five interradial areas. In Bathycrinus
carpenteri these are supported, according to Danielssen and Koren, by large retiform
calcareous plates a little sunk in the perisome, which are obviously persistent orals,
though they seem to be entirely absent in the three other species of the genus. In his
preliminary description' of Bathycrinus aldrichianus Sii- Wyville Thomson said " the
disk is membranous, with scattered calcareous granules. The mouth is subcentral ; there
are no regular oral plates ; but there seems to be a determination of calcareous matter to
five interradial points round the mouth, v/here it forms little ii-regular calcareous bosses."
This description was accompanied by a woodcut which was definitely stated to represent
the Bathycrinus from Station lOG (in Mid- Atlantic), and not one of the numerous
specimens obtained at Stations 146 and 147 in the Southern Ocean. The whole set were at
first regarded by Sir Wyville Thomson as belonging to one and the same species, to which
he gave the name Bathycrinus aldrichianus ; but he subsequently limited this name to
the individuals from the Southern Ocean. They are figured on PL VII., which was lettered
and printed off" before his death. The disk of one of them is shown on PL VII. fig. 3, and
obviously bears neither scattered calcareous granules nor interradial bosses round
the mouth. This individual has not come into my hands ; but the disk of another which
I have examined is perfectly naked, except for the covering-plates at the sides of the
ambulacra, and the peristome is like that shown in fig. 3. The specimen which was
obtained at Station 106, and had been already figured in the Journal of the Linnean
Society as Bathycrinus aldrichianus, was drawn by Mr. Black for PL VIII. ; but on one
1 Journ. Linn. Soc. Land. (ZooL), 1876, vol. xiii. pp. 50, 51 ; also in The Atlantic, 1877, vol. ii. pp. 92-95, fig. 23.

REPOET ON THE CEINOIDEA.

23i

of the copies of tlais plate which were sent to me I found the MS. name Bathycrinus
camjihellianus in Sir Wyville's handwriting. He had evidently therefore come to regard
this individual as specifically distinct from those subsequently obtained in the Southern
Ocean, to which he eventually limited the name Bathycrinus aldrichianus ; and the disk
of this individual {Bathycrinus Campbell iamis) corresponds in most respects with his
description quoted above.

The whole of each interradial area is covered with closely set, calcareous plates, not
scattered granules ; but I cannot make out that they are more abundant in the immediate
neighbourhood of the mouth than elsewhere. The covering plates of the ambulacra are
smaller and less regular than in Bathycrinus aldrichianus, and their course can only be
followed with difficulty. In the solitary young specimen of Bathycrinus gracilis
(PL Villa, fig. 1) there are neither orals nor anambulacral plates, so that in these
respects it resembles the southern Bathycrinus aldrichianus rather than the northern
Bathycrinus carpenteri.

In order that I might be able to compare the anatomy of Bathycrinus with that of
Rhizocrinus, two of the least perfect examples were cut into thin sections with a Leyser's
microtome. Unfortunately, however, the circum-oral portions of the one used for the
vertical sections were somewhat mutilated, and the remainder was found to be contracted,
and, as it were, coagulated by the action of the spirit, so that not even its general anatomy
could be made out, much less any histological detail. The arms, on the other hand,
yielded more satisfactory results.

In correspondence with the different size of the basals, the chambered organ (PL Vllb.
figs. 1,2, ch) is more flattened than that of Rhizocrinus. It is continued below into the vas-
cular axis of the stem, which is thrown into numerous folds as it passes through the closely
packed discoidal joints at the top of the stem. As in Rhizocrinus there is only one central
vessel (PL Vila. fig. 2; PL Vllb. fig. 2 — v), and not two or more as in some species of
Pentacrinus and in the Comatulse. The plexiform gland rising from the chambered organ
expands considerably within the central funnel of the calyx (PL Vllb. figs. 1, 3-5, x), and
soon becomes lobulate. It appears to be made up of numerous small cells, which can be
traced downwards into the flattened epithelial lining of the vessels in its lower part, and in
the chambers of the chambered organ. But it has nothing like the glandular appearance
of the corresponding organ in Antedon carinata, the numerous lobules of which consist
of distinct tubules lined by columnar cells, like those of the urinary tubules.

On its way up through the disk the plexiform gland becomes narrower again, probably
on account of its giving off extensions which form the intervisceral vessels, as in other
Crinoids. It is interradial in position, as usual, and may be followed at the sides of the
fore-gut both in transverse and in longitudinal section (PL Vllb. figs. 6-8, x) to a little
way below the peristome, where its further course cannot be traced. I have little doubt,
however, that its connections are essentially the same as in other Crinoids. But owing to

236 THE VOYAGE OF H.M.S. CHALLENGER.

the imperfect condition of the top of the disk in the specimens which were cut, I have
been quite unable to make out many details of structure. One point, however, is of
interest, and that is that there are more than five water-tubes ; for there seem to be
three in each interradius, and not one only as in Rhizocrinus. As in this genus too, there
are strongly marked interradial diverticula of the gut (PI. Vllb. fig. 7), which are
supported by the expanding processes attached to the inner faces of the third radials
(PI. VII. fig. 4a; PL Vila. fig. 17).

The arms of Bathycrinus present no essential anatomical differences from those of
other Crinoids. The food-groove which is sunk within the ventral furrow of the skeleton
(PI. VII. fig. 8 ; PL Villa, figs. 4, 5), instead of being some distance above it (PL LXI.
figs. 4-6), is narrow relatively to the width of the arm, and protected by covering plates,
as already described. The radial blood-vessel (PL Villa, figs. 4, 5, h) and ambulacral
nerve («) could be clearly distinguished in sections, the latter being exceedingly thin, or
apparently sometimes even absent beneath the middle line of the ambulacrum.

Except at the arm-bases the -water-vessel (ty) is relatively small, being much flattened
from above downwards ; but the tentacles are large and bear numerous papillae. Beneath
the water-vessel, and projecting into the subtentacular canal, so as almost to divide it
into two parts, is a more or less continuous band of closely nucleated connective tissue,
which perhaps represents the structure marked x by Semper^ in A^tinometra parvicirra
(Actinometra armata, Semper, MS.). At the bases of the arms the subtentacular canals
are hardly traceable, their places being occupied by a complicated network of genital
vessels, which are doubtless connected in the disk with the upper end of the plexiform
gland, as in other Crinoids. But this plexus soon passes into a simple genital cord, as
represented in PL Villa, figs. 4, 5, gc. It sometimes nearly fills up the small genital
canal in which it lies, while there is a large and triangular coeliac canal beneath it (cc).

The axial cords of the rays and arms of Bathycrinus, like that within the stem, are
remarkable for the extensive subdivisions of the branches which proceed from them.
Like those within the pinnules of Holopus and Hyocrinus (PL Vc. figs. 2, 3, 8, a), they
take a somewhat wavy course within the radials, as is seen in PL Vllb. fig. 1, A ; while
the branches which come off from them in the second and third radials are shown in
figs. 6, 7, a'. Owing to the small height of these joints, the two dorsal bi'anches which are
usually so well defined in the Comatulse (PL LXI. fig. G) extend themselves laterally in
the plane of the transverse articular ridge, while they are scarcely visible at all in the
arms. On the other hand, the branches which extend upwards towards the ventral
surface of the arm subdivide again and again, giving rise to a number of exceedingly
fine fibrils, in the course of which bipolar, and occasionally multipolar, cells are clearly to
be distinguished (PL Villa, figs. 4, 5, a'). This character is better shown in Bathycrinus
than in any other Crinoid which I have yet examined.

1 Arleiten aus dem zool.-zootom. Instihit in Wiirzbiirg, 1874, Bd. i. p, 261.

REPORT ON THE CRINOIDEA, 237

The nearest, and in fact the only ally of Bathyci'inus is Rhizocrimis ; but the
differences between the two genera are much greater than their resemblances. The latter
may be summarised as follows : — 1. The bifascial union of the stem-joints. This is
common to Bourguetici'inus and Mesocriims, to the Pentacrinoid larva of Comatula, and
the Palseozoic Platycriniis. 2. The presence of large processes on the second joints above
the radials, which support the disk. 3. The absence of pinnules from the lower parts of
the arms, and the union of the arm-joints in pairs, with a pinnule on the second joint of
each pair only.

The differences between the two types are shown in the following Scheme.

Rh kocrinus. Bathycrinus.

a. May have radicular ciiri. a. No radicular cirri.

b. Only one or two discoidal joints at the top of b. Many thin discoidal joints at the top of the

the stem, and those not very thin. stem.

c. Basals long; radials short, and very closely c. Basals short; radials long, and comparatively

united. free.

d. Primary interradial cords fork within the d. Primary intenvadial cords fork within the

basals. sutures between the radials.

e. Five arms. e. Ten arms.

/. All the joints above the first radials are united /. The fifth, eighth, and eleventh joints above the
in successive pairs by syzygy. primary radials have a muscular articulation

at each end ; the remainder are united in pairs
by trifascial articulations.
g. The first pinnule may be either on the g. The first pinnule not lower than the eleventh
primitive sixth or eighth joint above the first joint above the first radial,

radial, i.e., on the epizygal of the third or
fourth syzygial pau-.

It has been already pointed out that Bathycrinus ranges through a greater number
of degrees of latitude than any other Stalked Crinoid, even Rhizocrinus ; and it is only
surpassed in this respect by the ubiquitous Antedon. Bathycrin.ts carpenteri was
found by the Norwegian North Sea Expedition as far north as 65° 55' N. lat. ;
w]iU.e Bathyc7'inus aldrichimius was twice met with by the Challenger in the Southern
Ocean beyond the parallel of 46° S. lat. In the intervening Atlantic Ocean have been
found Bathycrinus gracilis (Bay of Biscay) and Bathycrinus campbellianus (just north
of the equator) ; while other examples of the genus, the specific characters of which are
as yet unknown, were dredged by the "Talisman" in the Atlantic (1883) at a depth of from
2000 to 2380 metres (1200 fathoms).^ It is distinctly an abyssal type, ranging from
1050 to 2435 fathoms. The only Crinoids which have been found at greater depths than
the latter are two species of Antedon.

We have no certain knowledge of the occurrence of Bathycrinus in the fossil state ;
though it is by no means impossible that some of the stem- joints so common in the

• Rapport priliminau'e sur I'expedition du " Talisman " dans I'Ocean Atlantique, Comptes rendus, t. xcvii.
p. 1392.

238

THE VOYAGE OF H.M.S. CHALLENGER.

Tertiary beds of the Continent may really belong to this genus. Meneghini has indeed
suggested that two forms of joint found in the Italian Tertiaries are those of Bathycrinus:^
but there is no evidence either for or against this idea. For the lower stem-joints of
Bathycrinus are indistinguishable from the non-cirriferous joints of Rhizocrinus, though
the differences between the immature joints of the upper part of the stem may be readily
recognised in the two genera.

Key to the Species of Bathycrinus herein Described.

I. The lower part of the radial funnel much constricted, ....

II. The radial funnel slopes uniformly downwards from the upper to the lower edge.

a. Calyx constricted at the basiradial suture. Basal ring scarcely wider

above than below. Arm-joints smooth, ....

h. The slope of the radials is continued on to the basal ring, which is wider

above than below. Arm-joints overlap, ....

1. Campbellianus.

2. Aldrichianus.

3. Gracilis.

1. Bathycrinus campbellianus, n. sp. (PI. Vila. figs. 22, 23; PI. VIII. ; woodcut, fig. 15).

Bathycrinus aldricManus, Wyv. Thoms. (pars), Journ. Linn. Soc. Lond. (Zool.), (1876), vol. xiii.,
1878, pp. 47-51, fig. 1 ; The Atlantic, 1877, vol. ii. pp. 92-95, fig. 23.

Dimensions.

Total length of specimen, without stem, ....

32-00 mi

Greatest height of radial funnel, .....

1-80 ,

Greatest diameter, ......

3-00 ,

Least diameter, .......

0-70 ,

Length of second radial, ......

1-85 ,

Length of third radial, ......

MO ,

Stem and basals unknown.

Description of an Individual. — The radial funnel widens slightly from below upwards
to just beneath its equator, where it expands considerably, owing to the dorsal surface of
the radials suddenly becoming much more convex. The rim of the funnel is thus drawn
out into five curved edges in which the second radials rest. They are trapezoidal in
form, widen from below upwards, and have a strong medio-dorsal convexity which starts
from the whole width of the lower edge and narrows rapidly till just beneath the dorsal
edge, whence it is continued on the axiUary. The lateral portions of the surfaces of both
joints are flattened. The axillaries are shorter than the second radials, but wide and
barely pentagonal in form, with a medio-dorsal ridge which forks at its proximal end and
is continued on to the arm-bases, where it soon disappears. The flattened lateral portions

' Processi Verbali, Soc. Tosc. di Sci. Nat., 7 Luglio 1878, p. xxxii.

REPORT ON THE CRINOIDEA.

239

of the axillaries have sharp edges, and these are continued along the sides of the first
three or four brachials, after which the joints become more cylindrical in form. The two
lowest are squarish, and both, but especially the first, are wider than their successors,
which are longer than wide, and overlap rather
sharply lioth at the muscular and at the trifas-
cial articulations, but more so at the former.

The first pinnule is almost always on the
ninth brachial, and the pinnules are attached
some little way behind the distal edges of the
joints which bear them, so that the socket is quite
distinct from the articular face.

The joints of the six or eight lower pinnules
which are enlarged to hold the genital glands
have a sharp dorsal edge and broad thin sides
which are much produced upwards, but the later
pinnules are more slender. The disk is paved
with closely set plates.

Colour, in spirit, white.

Locality. — Station 106. August 25, 1873 ;
lat. 1° 47' N., long. 24° 26' W. ; 1850 fathoms ;
Globigerina ooze ; bottom temperature, 36°'6 F.
(1°'8 C). One specimen, now without stem or
basal rincr.

Remarks. — This species may be readily dis-
tinguished from the other three by the shape of
the funnel formed by the united first radials, and
the overlap of the arm-joints. As pointed out
already [ante, j). 234), it was not at first dif-
ferentiated by Sir Wyville Thomson from the
larger form obtained in the Southern Ocean, to
which he ultimately limited the name Bathy-
crinus aldrichianus. In fact it seems to be the
type from which the description of Bathycrinus
aldrichianus was mainly drawn up. Although
an entire specimen was obtained, the stem
appears to have separated from the head and to have been eventually lost ; for
otherwise we may take it for granted that the stem would have been drawn under
Sir Wyville's direction, together with the head belonging to it. In fact the upper
part of the stem was drawn, together with the head, for the woodcut (fig. 15) which

Fig. 15."

Bathycrinu scampbelHanus, n. sp. ; three times
the natural size.

240 THE VOYAGE OF H.M.S. CHALLENGER.

appeared iu the Journal of tlie Linnean Society, aud also in The Atlantic. There
can be no doubt that the head there represented is that of Bathycrinus campbel-
lianus, owing to the strongly marked serration of the arm -joints, and the character of
the pinnules. In fact the figure given in The Atlantic is definitely stated to represent
the specimen from Station 106 ; while the fragment of a stem which it shows is entirely
difierent from the upper part of the stem of Bathycrinus aldiichianus represented in
PI. VII. fig. 11. The numerous thin joints immediately beneath the cup, which are so
characteristic of the genus, are not properly represented in the woodcut, and the joints
just below where these ought to be are considerably longer than one would expect to
find so near the cup. It may be assumed that Mr. Wild's drawing was photographic in
its accuracy, so far as he could make out the structure of the small specimen ; but errors
may have crept in during its reproduction on wood, and the cut was published during
Sir Wyville's absence from England, so that he had no opportunity of revising it.
Under these circumstances, therefore, it appeared preferable to say nothing about the
stem in the specific diagnosis given above, rather than to attempt to describe it from
a probably incorrect woodcut.

Two irregularities in the arrangement of the pinnules appear in this specimen. In
one case the sixth brachial is not free as it normally is, but is united to the seventh ;
whUe the eighth is free instead of the ninth, and bears the first pinnule. In the other
case the ninth brachial is not free as usual, but is united to the tenth, which bears a
pinnule, so that there are only two free joints, the third and the sixth.

In his first account of the Challenger species of Bathycrinus, Sir WyviUe Thomson ^
stated that " as the stalked Crinoids are perhaps the most remarkable of all the deep-sea
groups, both on account of their extreme rarity and of the special interest of their
palseontological relations," he meant " to associate the names of those naval officers
who have been chiefly concerned in carrying out the sounding, dredging, and trawling
operations with the new species whose discovery is due to the patience and ability with
which they have performed their task." In accordance with this idea Sir WyviUe
established the new species Hyocrinus hethelUanus and Bathycrinus aldrichianus ; and
he left the MS. names Bathycrinus campbellianus and Pentacrinus naresianus upon
the proofs of two of his plates. These names I have of course adopted. They refer to
Lieutenant Lord George Campbell, and to Admiral Sir George Nares, who commanded
the Challenger during the earlier part of the cruise.

> Journ. Linn. Soc. Land. (ZooL), 1876, vol. siii. p. 47.

EEPUKT ON THE C'lliNOlDEA. 241

Bathycrinus aldrlchianus, Wyville Thomson, 1876 (PI. VIJ.; Tl. Vila. figs. 1-21 ;
PI. Vllb. ; PI. Villa, figs. 4, 5).

Batliyci-lumt uhlra-hlaniis, Wyv. Tlioins. (pais), .Juiiiii. Linn. Soc. Loud. (Zool.) (l^l76) 1S78,
vol. xiii., pp. 47-51 ; Tlie Atlantic, vol. ii. pp. 'Ji-'d'i, 1877.

Dimensioi

is.

Greatest length of stem (Jiile C. W. T.), ..... i'OU-250 mm.

Longest stem-joints, ....... 3'50 ,,

Greatest width of lower joints, . . . . . . 2 -00 „

Total length of largest head, , . . . . . . 30-00 ,,

Height of basal ring, . . . . . . 0'80 „

Greatest height of radial funnel, ...... 2-10 ,,

Greatest diameter of radial funnel, . ..... 4"25 ,,

Least diameter of radial funnel, ...... 1'50 ,,

Length of second radial, . . . . . . . 3'00 ,,

Length of third radial, . . . . . . . 2-75 ,,

Stem composed of about one hundred joints, of vvliieli the first twenty or twenty-five
are wider than high, those immediately beneath the cup being mere circular disks, and
slightly wider than the thicker ones on which they rest. The next joints below increase
slowly in width and more rapidly iu length. The ends are slightly expanded and the
terminal faces oval-oblong in form. Their shorter diameter is occupied by a strong and
continuous transverse ridge, the directions of the ridges on the two faces of any joint
being nearly at right angles to one another. About the middle of the stem the diameter
begins to increase more rapidly, and the ends of the joints appear less expanded while the
terminal faces are circular ; and near the Ijase the diameter increases considerably,
while the length remains the same or even sliglitly diminishes. The ends of the joints
are considerably expanded and their faces oval, with their longer axes occupied by
an articular ridge which is interrupted at the opening of the centi'al canal. The stem
ends below in a small branching root.

The basals are closely united into a low, smooth ring, which is scarcely wider above
than below, where it is of the same diameter as the thin uppermost stem-joints. On its
upper surface rests the funnel-shaped radial pentagon which expands uniformly upwards
to its distal edge, so that the calyx has the appearance of being much constricted at the
basiradial suture. The surface of each radial is strongly convex in the middle but falls
away at the sides, so that the rim of the funnel is drawn out into five curved edges in
which the second radials rest. These are trapezoidal in form and convex at their
proximal ends. This convexity is continued onwards as a ridge of tolerably uniform
width, so that there is a flattened surface on either side of it, which increases in size
towards the distal edge. This feature is continued on to the axillarics, which are wider
than the second radials, and barely pentagonal iu form. The medio-dcrsal ridge enlarges

(zool. ch.vltj. f.xp. — I'ART XXXII. — 1884.) Ii 31

•242 THE VOYAGE OF H.M.8. CHAL7.ENGER.

;iliniir the centre of the joint, and is eontinuocl on to eaeh of the arm-l)ases, M'here it disappears
after about tlie first eight joints. These have sharp lateral edges where their flattened
sides meet the dorsal surface. The first thi'ee or four are wider than long, their successors
gradually liecoming more cylindrical, but never much longer than wide and not over-
lapping, so that the dorsal surface is smooth. The third joint is sometimes the only free
one. all its successors being united in pairs ; while in another case tlie fourth is free as
AvcU us the third, and also the seventh and tenth, the intermediate pairs being united.
In a third variety three pairs after the third brachial are united, and the tenth is free
again, while the remainder of the arm is normal. Tliere are fiftv joints in an arm
30 mm. long {fde C. W. T.).

The first pinnule is generally on the ninth, l)ut sometimes not till tlie eleventh
brachial. The pinnule attachment is near the end of the joint, so that the socket enters
into the composition of the terminal face (PI. Vila. fig. 21). The joints of the lower
pinnules are not greatly larger than those of tlieir successors. The disk is unprotected
by i:)lates.

Colour, in spirit, white.

About a dozen heads and several stems, some retaining the basals.

Localities.— Station 146. December 29, 1873 ; lat. 46° 46' S., long. 45° 31' E.;
1375 fathoms ; Globigerina ooze ; l)ottom temperature 36°'6 F.

Station 147. December 30, 1873 ; lat. 40° 16' S., long. 48° 27' E.; 1600 fathoms;
Diatom ooze; bottom temperature 34°"2 F.

I have no record respecting the number of individuals obtained at each Station.

According to Sir Wyville Thomson' this species " seems to be widely distributed ; we
have detected fragments of it at at least six or seven Stations in the Atlantic and the
Southern Sea."

Remarks. — The head of this species is readily distinguished from that of Bathycfiuvs
camphelliamis l)y the calyx lieing constricted at the basiradial suture, and not in the
radial funnel above it ; Ijy the greater width of the arm-bases, and by the smoother
dorsal surfjice of the arms. In the first of these characters it resembles Bathycrinvs
carpenteyl, so far as can be made out from Danielssen and Koren's figure; but it is much
more robust than the northern species, in which the axillaries are considerably shorter
than the second radials, while the lower stem-joints are relatively longer, and have much
more expanded ends than those of Bathijcrimis aldrtchianiis, in which the dice-box shape
is less apparent.

When the collection of Stalked Crinoids came into my hands, T found one bottle
lalielled " Pterocirnus and Hijocrinus, Stations 140 and 147." As I had never heard of
the foi-mer genus, my curiosity was much excited, and it was with considerable
disappointment that I found the contents of the bottle, besides Hyoerinv.^, to consist

' .Tniiru. Liini. Snr. Liniil. (Zool.), vol. xiii. p. "il (ISTG) 1S78,

REPORT ON THE CRINOIDEA. 243

pniK'ipally of headless stems ul" Ba(/i//rrini(s (i/drichidims, suiiic witli ami smnc without
the basal ring at the .summit.

■Mr. Murray tells me that one specimen of " Ptcmcj'invs" was presented to the
University Museum at Sydney ]>y Sir Wyville Thomson, whieh indicates that there must
have been several duplicates of the type. The only explanation of these focts which 1
can think of is, that Sir Wyville had not then (1874) made up his mind that the
relatively large individuals dredged at Stations 106, 14(), ami 147 were generically
identical with the small and immature specimen whieh lie had described two years
previously as the type of the new genus Bathycrinus ; so that he was led to give them the
MS. name Pterocriiuis, which he afterwards abandoned (1876) in favour oi Batlnjcriuus.

Bailii/ci'iitus tjr((c(/is, Wyville Thomson, 1872 (PL Mlla. figs. 1-3; woodcut, fig. 16).

1872. Bathyci'inus (jracillti, Wyv. Thorns., Proc. Vm\. Soc. Edin., vol. vii. p. 772; The Depth.s
ufthe Sea, pp. 450-454, tig. 73, 1873.

Dimensions.

Greatest lengtli of stem (about forty joints), . . . 57'00 mm.^

Longest stem-joints, ....... 2'50 ,,

Length of head, ........ 8'00 ,,

Description of an Individual. — Stem composed of about forty joints, of which the
first ten or twelve are \\dder than high, increasing in thickness from above downwards.
The following joints, at first cylindrical, increase rapidly in length, and acquire slightly
expanded ends, wdiicli become more and more marked in the lower joints, till the widtli
of the end is 1 mm., twice the width of the shaft. The joints diminish in length towards
the base of the stem, which ends in a small branching root.

The basals are closely united into a smooth ring, the lower part of which is of the
same size as the upper stem-joints, but expands gradually upwards." The radial pentagon
aboA'e it expands more rapidly, so that there is a constriction at the basiradial suture,
though not so marked as in Baflnjcrinus cddriddanus. The radials are strongly convex
in the medio-dorsal line, but more flattened at the sides. Second radials nearlj' oblong,
with a sharp median ridge and a slight hollow on either side of it. Axillarics shorter
than the second radials and nearly oblong, and also marked 1)}' a sharp medio-dorsal
ridge which forks al)out the middle of the joint, and is continued on to the three lowest
brachials. There are only alxiut twelve brachials in the arm, all of which are distinctly
longer than wide, while the first two or three are scarcely wider than the rest, and have
straight lateral edges like the radials. The later joints overla}) rather sharply, so that
the dorsal edge of the arm is serrate. The disk is unprotected by plates.

Colour, in spirit, brownish-white.

Zoca?%.— H.M.S. " Porcupine," 1SG9. Station .37. Lat. 47' 38' N., long. 12° 8' W. ;

1 Sir Wyville gave this measurement as 90 mm., which was possibly a misprint for (>(>.
-' This expansion is not quite marked enough in the figure.

244

THE VOYAGE OF H.M.S. CHALLENGER.

2435 fathoms; Globigerina ooze; Itnttom temperature 3G°'5 F. One nearly complete
specimen, and one stem with the basal ring attached, but wanting the rest of the calyx.

Remarks. — The shortness of the arms
and the absence of well grooved pinnules
indicate that the single specimen of this
elegant little species is not fully developed.
Sir Wpalle Thomson' said that " there is no
trace of pinnules, and the arms resemble in
character the pinnules of Rhizocrinvs."
There are, however, little .stumps on the
terminal joints of one or two of the arms
which give them the appearance of bifurca-
tion, just as at the growing point of the
arms of a young Comatula or Pentacrinns;
and I see no reason to doulit that these
stumps are the commencing pinnules.

T have already pointed out that the arm-
ioints of Bathi/cri)uis gmcili.'! are united in
pairs exactly like those oi Bathi/crinus aJdri-
chianus and Rhizocrimis, excej^t that in the
latter genus the union is a syzygial one ; and
the supposed resemblance of the arms of
Batliijc rinus gracilis to the pinnules of Rhizo-
crimis thus disappears.

As might be expected from its geogTa-
phical position, Bathi/crimis grcwilis comes
nearer to Bathycrinus carpcnteri than to
either of the Challenger species. The lower
stem-joints are strongly constricted with
markedly expanded ends as in that type ; 1 >ut
the absence of orals, the well defined medio-
dorsal keel on the radials and arm-bases, and
the serrate condition of the arms are sufficient
to distinguish it without difficulty.

During the expedition of the " Talisman "
in 1883 ■■ a few much mutilated specimens
both of Bathycrinus gracilis ViiiA of Rhizocrinus raivsoni {Democri mis jxii fait i, Perrier)

' The Depths of the Sea, p. 452, 1873.

" Comptes reiidus, t. xcvii. p. 1392. See also H. Filhol, Voyage du " Talisman " in La Xafurc, No. 572, 17th Jlay
1884, p. 391.

Fig. 16. — Lalhycrinas gracilis, WyviWe Tbomsoa ; twice
the natural size.

REPOr.T ON THE f'TtTNOTDEA. 24. J

were obtained off tlio Morocco coast at a deptli of from. 2000 to 2300 metres (1200
fathoms), "par le travers d\\ cap Ghir et du cap Noun, a 120 milles environ dc la cote."
This locality (in al)out hit. ,30° N.) is the only one at which these two genera have
been found associated. It is some 18° farther south than the position at which BnfJn/-
crinus gracilis was first discovered by the " Porcupine."

Genus Rhizocririvs, M. Sars, 1804 ; emend., V. IT. fVirpcntcr, 1884.

Eugeniacrinites "]
Apiocrinitrs i
Asterias [' Auctniiim.

Goniaster j

184G. nourgueticrinus, d'Archi.ac, jMiJiii. Soc. giiol. de P"i'ance, sdr. 2, t. ii. p. 200.

1847. Conocrinus, d'Orbigny, Couvs lili'mcnt. de I'aler.nlol. et de Geol. .slratigT.,^ t. ii. Faso. 1, I'aris,

1852, p. 147.'
1850. B<nirfjiietkrinn>i, d'Archiac, jMem. Soc. gool. de France, .ser. 2, t. iii. p. 418.
1850. Conocrinus, d'Orbignj', Prodrome de Paleontologio stratigraphique luiiver.scllp de.s Aiiiiiiaux Mollnsques

et Eayonnc-es, t ii., Paris, 1850, p. .332.
1852. BoiintKi'ticrhnis, Forbes, Monograph of tlie Ecliinodermata of the Jlriti-sh Tertiaries, p. 30.
18G4. Rliizorrhnis, ]\I. Sars, Forhandl, Videusk. Selsk., p. 127.
1868. Boiirijuetifrimis, Ponrtales, Hull. Mus. Corap. Zoiil., vol. i.. No. 7, p. 128.
1868. liki::ocrimi>i, ]\I. Sars, ]\Iemoires pour .servir a la counaissance des Crinoules vivants, p. 38.
18G8. Rhkcorrinus, W. 1!. Carpenter, Proc. Roy. Soc., p. 173.
1872. llhi::ocrinm, Wyv. Thorns., Proc. Roy. Soc. Edin., vol. xvii. p. 770.
1874. Rhizocrimis, Pourtales, 111. Cat. Mus. Conip. Zoo!., No. 8, p. 27.
1874. Rhizocrlnm, iManzoni, Eollett. d. R. Cora. Geolog., p. 158.

1874. Rhizocrinw, Beyrich, jNIonatsber. d. k. prouss. Akad. d. AViss. Berlin, p. GCG.

1875. Conocrinus, Meneghini, Atti doll. Soc. To.so. di Sci. Nat., vol. ii. p. 46.
1875. Rhizocrimis, Meneghini, Ibid., p. 57.

1878. Conocrinus, de Loriol, ^lonogi'aphie des Crinoides fossiles de la Suisse, p. 190.
1880. Conocrinus, Zittel, llandbuch der PaUoontoIogie. Paheozoologie, Bd. i. p. 391.

1882. Rhizocrinus, P. H. Carpenter, Bull. Mus. Comp. Zoiil., vol. ix., Xo. 4, ji. 173.

1883. Democrinus, Perrier, Conii)tes rendus, t. xcvi. p. 450.

1883. Rhizocrimis, P. H. Carpenter, Ann. and Mag. Nat. Hist, ser. 5, vol. xi. p. 33G, 1883.

1884. R/iizocriniis, P. H. Carpenter, Proc. Roy. Soc. Edin., vol. xii. p. 356.

The stem is composed of dice-liox sliaped joint.s, and terminates below in a spreading
root, or bears a number rif liranchino- radicular cirri which come off irreoularlv, luit not
in definite verticils.

Only a very few of the ujiper joints are thin and discoidal, and slightlv wider tlmn
those lielow them.

■ The second volume of tlio Cours elementaire bcar.s the date 18J2 ou tlio title page ; l>ut the genus is referred to
as "Conocrinus, d'Orb., 1847." This also appears in tlie rrndrome de Paleontologie published in 18£0, in wliich work
according to de Loriol, the genus was first established ; and I suspect that the date 1847. which is claimed by itOrbigny
for his genus, merely has reference to an unpublished MS.

246 THE VOYAGE OF H.M.S. CHALLENGEr>.

The cal} X i« higher than wide, geuerally much su. It consists of five Ixisals and
five radials, of which the former (in the recent species at least) are much the hmger.
Arms fi'S'e, the joints united in pairs by syzygies, with pinnules on the epizygals only.
The first pinnule on the epizygal of the third or fourth pair. The epizygal of the first
pair has the sides of its ventral furrow produced upwards into strong processes which
stipport the disk. The peristome is protected by five oral plates of variable size, but the
remaining portions of the interpalmar areas are not plated. The ambulacra have covering
j^lates, but no side plates.

Remarha. — The name lihizoci'iiius WixsHrst employed in 1864 by the late M. Sars ^
to designate a singular new Crinoid which had been discovered by G. 0. Sars in his
dredgings among the Lofoten Islands ; and it was the extreme interest of this type as
a sort of degraded Apiocrinite which, through the intervention of Dr. Carpenter and the
late Sii' Wyville Thomson, led to the di'edging cruise of H.M.S. "Lightning" in 1868.
The results of this cruise, Rhizocrinns among them, were so remarkable that it was
followed l:)y the "Porcupine" Expedition of 1869-70, and eventually by the voyage of
the Challenger. It Avas the discovery of Rhizocrinus, therefore, and the interesting-
speculations to which this discovery gave rise, that led this country to take a foremost
place in the work of deep-sea exploration. Meanwhile, however, Rhizocfiuits had been
rediscovered by the late Count Pourtales during the ch-edgings carried on by the U.S.
Coast Survey in connection with the regular exploration of the GJulf Stream.'-^ At that
time (May 1868) Sars's elaborate memoir on the tyjjc had not yet appeared, and tlie
specimens dredged by Pourtales were described by him as " undoubtedly belonging to
the genus Bowgueticrimis, as defined by d'Orbigny," a remark in which I entirely
concur. He gave the provisional specific name " Hotessieri" to his specimens, thinking
that they might possibly be identical with Bourguetio'lnus Hotessieri, d'Orb., stem-joints
of which had been discovered in a recent Ijreccia at Guadeloupe. After the appearance
of Sars's memoir, however, he recognised the identity of the Gulf Stream and of the
Lofoten examples of the type, for which he adoj)ted Sars's name Rkizocrinus lc>f'otensis.'

But he took a totally diflerent view of the composition of the cup from that proposed
by Sars, and in this respect, as will be pointed out immediately, he was decidedly in the
right. For the large subradial portion of the summit, which was considered by Sars
as an enlarged upper stem-joint, was shown by Pourtales to consist of five closely
anchylosed basals.

In the year 1875 the name Rhizocrinus was doubtfully given by Meneghini^ to some

1 Forhandl. Vidensl: Selsk., Chiistiauia, 1864, p. 127. "Den UJe. Octuber, Hr. Sar.s lioklt et Foiedrag om
Piiiitacrinoide tilstanden af Comatula mrsii og mu eu uy levendu Crinoide Bhkocrinus lofotcnsis."

- Contributions to the Fauna of the Gulf Stream at Great Depths, Hull. Mus. Comp. ZooL, vol. i., No. 7, p. 128.

" List of the Crinoids obtained on the coasts of Florida and Cuba by the U.S. Coast Survey Gulf Stream E.xpedi-
tions, in 1867, 1868, 1869, Ihid., Ko. 11, p. 357.

■* I Crinoidi Terziarii, Atti dell. Soc. Tosc. di Sd. Nat., vol. ii. \>f. 46, 50.

•S'

REPORT ON THE CRINOIDEA. 247

stem- and arm-joluts from the Italian Tertiaries, while he revived d'Orbigny's name
Conocrinus for d'Archiae's Bourgueticnnns thorenti, after an interval of nearly twenty-
five years. During this period, so far as I am aware, no palaeontologist had taken any
notice of d'Orbigny's attempt to differentiate Conocrinus from Bourgueticrinus. The
first description' which he gave of the former genus (1847 Vj ran as follows: " C'est un
Boiirgueticrinus ayant la tige comprimee, mais avec une seule serie de pieces brachiales,
sans pieces basales ; " and he referred to it one unnamed species from the Suessonien
(Lower Eocene). Throe years later (1850) he spoke of Conocrinus as "genre voisin des
Bourgueticrinus, mais sans pieces basales, eomme les Engeniacnnns;" and he mentioned
Bourgueticrinus thorenti of d'Archiac as belonging to this generic ty|ie.- jMeneghini has
sho-mi, however, that two species were described under this name by d'Archiac. One
is a much elongated type, first described in 1846, and probably that referred to by
d'Orbigny in the following year; while the other that was not noticed till 1850, the year
in which the second (first ?) description of Conocrinus appeared, is tlie Eugeniacrimn
pyriformis of Miinster. This species was not referred l)y d'Orbigny to his new genus
Conocrinus, though undoubtedly belonging to it, as pointed out by Meneghini. But it
was retained by him in Eugeniacrinus, so that the only species of Conocrinus recognised
by d'Orl:)igny was the elongated Bourgueticrinus thorenti of d'Archiac. The figures and
descriptions of this type given by the latter author are somewhat incomplete. He had
very few specimens, and was exceedingly doubtful about the jiositioia of th(^ sutural lines,
while they are certainly placed wrongly in his figure," according io which the second
radials rest upon the sutures of the first. There is likewise no indication of an enlarge-
ment of the uppermost stem-joints so as to form a "summit," which is sd characteristic
of Bourgueticrinus ; while the presence of basals below the radials or " pieces superieures "
of Bourgueticrinus thorenti was distinctly described by d'Archiac, though he probably
figured them incorrectly. Nevertheless, Conocrinus is a Bourgueticrinus without basals,
and d'Archiae's species in which basals are present is made the type of the genus !

Neither does it help in the difterentiation of the two genera to speak of Conocrinus as
a Bourgueticrinvs with a compressed stem, when the stem of Bourgueticrinus itself is
descriljed as being compressed.'* I find very considerable difficulty in comprehending
what d'Orbigny really meant by Conocrinus. If it be "voisin de Bourgueticrinus," but
also resembles Eugeniacrinus in the absence of basals, why was it omitted in his scheme
of classification of the Apiocrinida^, published in 1858, from the fourth section comprising
Eugenicicrinus alone,^ and distinguished l^y having only " une serie de pieces au sommet"?
On p. 95 he pointed out that no Tertiary species of Bourgueticrinus were then known,
from which one may inft>r that the Tertiary fossils previously referred to this genus

' Conrs elcmt-nt. de PnU'ontol. pt tie Gool. stratif;!'., t. ii., 1852, p. 147.

2 Prodrome de Paleoutologie stratigraphiqne universelle des Aiiimaux ilollusques et Raj'onnues, t. ii. p. 322.

^ Mem. Soc. gM. de France, .ser. 2, t. ii. p. 2(»0, pi. v. lig. 20.

' Conrs (?l('mciit. de Paleontol. et de llenl, stratigr., t. ii., 18.j2, p. 117. '• Hist. Nat. des Crinoides, p. 2.

248 THE VOYAGE OF TT.M.S. CIIAJ^LENGER.

really belonged to Conocrinux. But as he did not place the latter type among the
Apiocriiiid;e together with Bourgaeticrinus and EugeniacriiiuK, it would seem that he had
cither abandoned it altogether, or else entirely misunderstood its real character and
affinities ; and in the absence of figures or original specimens his account of it would be
absolutely unintelligible.

Rldzoci'inus was first described by Sars in 1864/ and more fully in 1868"; and
tliough he was led to consider the anchylosed basals as a top stem-joint, this error was
corrected by Pourtales and myself before a fresh diagnosis of Conocrinus was given Ijy
de Loriol/ This indeed was only provisional, in default of better knowledge, and owing
to Meneghini's failure to find the interbasal sutures in a section through the lower part of
the calyx,^ just as in a recent Rhizocrinus or Bathycrinus (PI. Vila. fig. 13), de Loriol
was led to consider it probable that the basals " n'existent pas et sont intimement soudees,
de maniere a former comme une seule piece centro-dorsale." He thus fell into exactly
the same error as had Ijeen made Ijy Sars and Ludwig respecting the recent Rhizocrinus
lofotensis. Zittel,° however, who had satisfied himself regarding the presence of inter-
basal sutures in Conocrinus pijriformis, recognised the identity of this genus with Rhizo-
crinus, but did not adopt the latter name on the ground that " Nach den Regeln der
Prioritlit gel)uhrt dem Namen Conocrinus, d'Orb. die Prioritiit, weini gleich die Gattungs-
diagnose d'Orbigny's unvollstaudig und theilweise unrichtig ist."

It seems to me, however, that this is stretching the rules of priority to the widest
possiljle limit, or even beyond it ; and that definitions which are incorrect, meaningless,
and altogether incomplete have no claim whatever to recognition. Lutken remarked in
1864 that the distinction of Conocrinus from Bourgueticrinus was still a matter of
uncertainty ; while d'Orbigny's own countrymen Hebert and Munier-Chalmas did not
adopt his generic name for the new type which they described as Bourgiieticrinus s^tessi ;
and although it was subsequently referred by de Loriol to Conocrinus, and carefully
described, the genus Rhizocrinus had meanwhile become thoroughly well established and
universally recognised by zoologists.

Both Sars and de Loriol were in error as to the composition of the calyx in this type ;
and a correct definition of Conocrinus was not given until the publication of Zittel's
Palajontology in 1879 ; while even as early as 1868,'' and subsequently more fully in
1874," Pourtales had correctly jjointed out the characteristic features of Sars's genus
Rhizocrinus, especially as regards the presence of basals, which had Ijcen supposed to be
either absent altogether, or else modified into a kind of rosette. According to Sars**
" Ce qui est remarquable et characteristiquc pour la tigc du Rhizocrinus, c'est son sonimet

1 Forlmndl. Vulensk Selsk, p. 127. ^ Crinoides vivants, \)\\ 38, 39.

3 Swiss Crinoiils, p. 191. ' Loc. cit., p. 50.

'"' Paluiontolcigie, p. 392. " Bull Mus. Comji. ZuuL, vol. i., No. 7, ji. 129.

' III. Cat. Mus. C'oni}). ZouL, Xo. 8, pp. 27, 23. " Crinoitlus vivants, p. 4.

REPORT ON THE CRINOIDEA. 249

qui forme un grand evasement obeonique ou caliciforme, ce qui domic ^ notre Crinoitde
de la ressemblance avec Ic genres fossiles Ajnocrinus et Bourgueticrinus. Cettc extremity
n'est done pas autre chose que le sommet de la tige qui s'elargit successivement." Sars
considered the summit therefore as an enlarged uppermost stem-joint analogous to the
centro-dorsal of the Comatulse. He was unable to resolve it into any component parts
by treating it with alkalies ; and it is consequently not surprising that he was led to
regard the whole subradial portion of the calyx as composed of but one single i^iece. In
fact he was never able to separate this piece from the radials or the radials from one
another, the existence of these plates being only indicated on the outside of the calyx by
very faint sutural lines, occasionally rendered more distinct by slight furrows.

It never seems to have struck him, however, that there might be basal plates below
the radials which were similarly, but more closely anchylosed ; and he was led to consider
the basals as fused, like those of Comatulse, into a kind of rosette. This circular plate
(PI. Villa, figs. 6, 7 ; PL X. figs. 1, 4 — hr) "est situee a I'interieur dans I'espace central
laisse en dedans de I'anneau forme par I'adhesion des premiers radiales;"^ but Sars was
unable to isolate it, owing to its very close adherence to the first radials.

It has been pointed out that Rhizocrinus was dredged by Pourtales in 1868 before he
had heard of its discovery by Sars ; and the condition of his specimens was fortunately
such that he was able to describe the calyx as " composed of a cycle of elongated basal
(pelvic) pieces, followed by the much shorter first radials (costals) alternating with them.
These j)ieces are all so intimately connected with each other that the sutures are seen
with difficulty."^ This account seems to have escaped the notice of Sir Wyville Thomson;
for in his description of the " Porcupine " Crinoids' he stated that " in Rhizocrinus the basal
series of plates of the cup are not distinguishable. They are masked in a closed ring at
the top of the stem." He did not, however, entirely accept Sars's view of the composi-
tion of the calyx ; for he went on to say that " whether the ring be composed of the
fused basals alone, or of an upper stem-joint with the basals within it forming a rosette
as in the calyx of Antedon, is a question which can only be solved by a careful tracing
of successiye stages of development." The relatively large specimens which were dredged
by the "Porcupine" in 862 fathoms ofi" Cape Clear, show the interbasal sutures very
clearly ; and though they were referred at the time to Rhizocrinus lofotensis, they really
belong to the Caribbean species Rhizocrinus raivsoni, larger specimens of which were
obtained by the "Hassler" ofi" Barbados in 1871, and described by Pourtales in 1874.
These, like the " Porcupine " specimens and the Gulf Stream variety of Rhizocrinus
lofotensis, also showed distinct interbasal sutures. Pourtales was therefore led to
dissent from Sars's description of the calyx in this genus, and to repeat more emphatically
his own previous statements concerning the existence of long but closely united basals.*

1 Crinoides vivants, p. 12. ^ Bull. Mus. Comp. Zool, vol. i. p. 129.

3 Proc. Boy. Soc. Edin., vol. vii., 1872, p. 770. * Mem. Mus. Comp. Zool., No. 8, pp. 28, 29.

(zOOL. CHALL. EXP. — PART XiXII.- — 1884.) li 32

250 THE VOYAGE OF H.M.S. CHALLENGER.

He found all the sutural lines in tlie calyx of both species, including those between the
basals, to be " always plainly visible with a lens of moderate power ; " and he was there-
fore naturally surprised at the description of the calyx which was given by Sars as the
result of his study of the Norwegian specimens without interbasal sutures. In order to
verify the truth of his analysis of the calyx, Pourtales " forced a needle through the
central hole of the calice of a Rhizocrinus lofotensis until it split. The fractures followed
the joints between contiguous basals and between the latter and the first radials."

These important observations were entirely overlooked by Ludwig,^ who followed Sars
in describing the subradial part of the calyx as an enlarged uppermost stem-joint. He took,
however, another and more correct view of the circular plate which Sars had called the
basal rosette (PI. VHIa. figs. 6, 7 ; PI. X. figs. 1, 4 — br) ; for he regarded it as an unusual
development of the calcareous network which occupies the central portion of the radial
funnel in all Crinoids (compare PL XX. figs. 4, 6, 8) and surrounds the plexiform gland
ascending from the chambered organ (PI. XXIV. figs. 8, 9 ; PL LVIII. figs. 2, 3 — rp).

I have long ago expressed my acceptance'^ of this modification of Sars's views which we
owe to Ludwig ; but I could never quite reconcile myself to believe in the account which
he gives of the basals of Rhizocrinus. Having disestablished the rosette of Sars, he found
it necessary to seek elsewhere for the missing basals of this type ; and here he fell into
error, probably, I think, from a too exclusive reliance on his interpretation of sections
through the decalcified calyx, without properly considering the characters of an isolated
calyx minus its muscles and ligaments, such as was excellently figured by Sars in his
Tab. ii. fig. 43.

A horizontal section through the upper part of the calyx (PL VHIa. fig. 7), or a view^
of the calyx from above (PL X. figs. 1, 4), like those given by Sars, shows five (or six)
apparently interradial pieces (mp) surrounding the so-called basal rosette (br), and
occupying the interval between it and the large muscular and ligament-fossae on the
sloping distal faces. Sars considered these pieces, and rightly so, as integral parts of the
fii-st radials. Ludwig's sections, however, led him to believe that these pieces
(PL Villa, fig. 7, mp) " nieht radiiir, wie es nach der Sars' schen Autfassung sein miisste,
sondern interradiar liegen, so niimlich, dass stets die Mittellinie eines jeden Stiickes B in
die Trennungsebene zweier aueinanderstossender Radialieu fallt." ^ The interradial position
of the pieces in question is, however, shown just as well in Sars's figures as in those of
Ludwig, who had no real grounds for stating that Sars had indicated their boundaries
incorrectly. He considered that their interradial position precluded their being connected
with the radials, and was therefore led to regard them as " nach innen verschobene und

-^o

1 Zeitschr. f. wiss. ZooL. 1877, Bd. xxix. pp. 66-68.

2 On some points in the Anatomy of Pentacrinus and Rhizocrinus, Journ. Anat. and Physiol, 1877, vol. xii. p. 50.
^ The apparently interradial position of these pieces (mp) is more marked in Ludwig's figures, where they are lettered

B, owing to the semidiagrammatic character of these figures and the omission of the interarticular and dorsal ligaments.

REPORT ON THE CRINOIDEA.

251

in ihrer urspriinglichen Gestalt veranclerte Basalia." He supposed these basals to be in

a more embryonic condition tlian those forming the rosette of Comatula, and still

remaining independent of one another as five

isolated j^lates. But if this be the case one

would surely expect to find the basals

below the radials, and separating them from

the top stem-joint as in the early stages of

the Pentacrinoid larva of Comatulge and in t t Vj^gg^M ''tMS^^ nMmms^-f>i

the Stalked Crinoids generally.

Ludwig, however, represents the supposed
basals both in horizontal and in vertical
sections as occupying a position above the
radials, and forming the floor of the body- ^

cavity some way above the chambered organ,

with which the basals of Crinoids are always ■ . ^mi^-

closely related. The position assigned by
him to the basals of Hhizocrinus is one which
I have no hesitation in describing as utterly
impossible for these plates to occupy ; and
the dotted line inserted in his longitudinal
section to mark the junction of the supposed
basals with the first radials below them is
purely imaginary. The same may be said
of the lines that indicate an enveloping of
the radials by upward processes of the
supposed top stem-segment below them,
which would cut off the radials altogether

frnm tlip pytprinr of -f-hp palvsr ^pninmrp wnnil- ^"'- 17.— Diagrammatic vertical section through the calyx and
irom tne exterior OI ine CaiyX ^^COmpare WOOa- ^-^^^ ^j Rhlzocrinus h/otensis ,- x 75. Altered from Ludwig.

+ fl 1 "tN T1io+ v>or.-f /-vf +V>q /^oIttxt /-.-P PTiiV/i The section is represented as passing through the middle line of

cut, ng. Li ). 1 nat part 01 tne CaiyX 01 HIUM- ^ ray on the left hand side, and as almost (but not quite) iuter-

1-1 • n 1 1 T 1 • J. radial on the right. A, axial cord of ray ; at, primary inter-

CrinUS which was considered by Ludwig to radial cord ; b; basal tube ; *„ first brachial ; br, central

1 ^ -111 1 • calcareous plug, the basal rosette of Sans ; ca, fibrillar sheath

represent tne embryonic basals altered in round vascular axis of stem ; c/i, chambers of the qmnque-

. , .... . nil locular organ ; ch' their downward extension into the stem;

position, though not m nature, really belongs c.co., portion of circular commissure formed by the secondary

./ o interradial cords ; fg, fore-gut ; G, mid-gut ; I, basiradial liga-

tO the radials as was figured by SarS,^ who ment ; m, muscle ; nr, oral ring of the ambulacral nervous

° *' system ; 0, oral plate ; Ri, radial ; re, rectum ; xl, top stcm-

Spoke of these plates as follows: "etaut jilnt; T, tentacle ; i., central vessel of stem ;»/. nadial water-

^ ■■- vessel ; v)r, water-vascular ring ; x, plcxiform glaua.

tronquees dans leur partie interieure, elles

forment reunies un large anneau en laissant entre elles au milieu un esj)ace arrondi
ou un peu pentagone, remph par la plaque ' en rosette ' (fig. 42, 43, r) mentionnee plus
haut, reste probable des basales. Au 2?oint oti cette plaque rencontre le bout tronque

1 Crinoides vivants, p. 14.

!>52 THE VOYA.GE OF H.M.S. CHALLENGER.

interieur ou amiuci des radiales, on remarque a chacun d'elles im petit trou rond,
d'oti sort un sillon droit et lineaire (fig. 43, s), qui longe la ligne mediane de la face
superieure ou ventrale de chaque article et continue son parcours en remontant aussi de
long du milieu de la face ventrale des radiales suivantes." This interior truncated
portion of the radials is really their ventral face ; while the openings at its central
portion are the ends of the axial radial furrows descending to the lower part of the
calyx, and the furrows j)roceeding outwards from them are the ventral radial furrows
(PI. Villa, fig. 7 ; PI. X. figs. 1, 4 — vrf) as described and figured by Sars, though
Ludwig took them for interbasal sutures. Sars's fig. 42 is particularly instructive in
this respect, as four out of the five first brachials are in situ, and their ventral furrows
are seen to be continued downwards on to the radials.

The ventral interradial furrows which are so marked on the upper aspect of the calyx
of many Comatulse are absent or but slightly indicated in Rhizocrinus. Traces of them
may be seen, however, in fig. 42 on Tab. ii. of Sars's memoir. But the adjacent muscle-
plates of every two contiguous radials are intimately fused and also slightly everted.
Each is separated from its fellow on the same radial by a well marked, ventral radial furrow;
and the united halves of the inner faces of adjacent radials thus assume somewhat the
appearance of isolated interradial plates resting within and against the outer faces of
the radials. Ludwig was thus led to consider them as basals, and so to fall into exactly
the same kind of error wdth regard to their genetic relations as he attributed to Sars.

I have nothing to add to his account of the chambered organ ; but his description of
the cords which proceed from its fibrillar envelope needs a little modification. He has
pointed out that they are interradial and not radial as descrijjed by Sars ; but he says
that they " verbinden sich dann in den untersten Radialien durch Commissuren, ohne dass
vorher eine Gabelung stattgefunden hiltte." Were this really the case, Rhizocrinus would
be a much more anomalous form than it actually is. For in all other Crinoids, recent or
fossil, in which this point has been worked out, with the exception of Batliycrinus, the
primary interradial cords fork within the basals, and there are two openings either on the
inner (Comatulse) or on the under face (Pentacrinus) of each first radial (PI. XII.
figs. 11, 22 ; PI. XX. fig. 9). Ludwig, however, figures these cords ' in Rhizocrinus as
single so long as they remain within the basals (top stem-joint, Ludwig) ; and he believes
them to fork in the suture between two radials, so that their branches would not enter
the radials through their inner or under faces, but at then- lower lateral angles.

This is not quite the case, however, and it is probably to be explained by Ludwig's
having used the section-method only, without attempting to separate the pieces of the
calyx. This operation is one of no little difficulty, and some of the radials are sure to be
fractured in the process ; but others separate from the basals along the sutural lines, and
the arrangement of the canals can then be seen. The radials are comparatively low

1 Zeitsclir. f. wiss. Zool, Bd. xxix. p. 72, 1877, Taf. vL fij;. 18

EEPORT ON THE CRINOIDEA. 253

externally, and their upper and under faces approacli one anotlier rather sharply, owing
to the convexity of the upper surface of the basals, so that their inner ends are very thin.
The centre of the funnel which they form is occupied by the compact plug of close cal-
careous network which was mistaken by Sars for a basal rosette (PI. Villa, fig. 7 ; PL X.
figs. 1, 4 — hr). This is attached firmly to the inner ends of the radials, and comes away
with them. It lies on the top of the convex upper surface of the basals, the sloping sides
of which are divided into fossse for the reception of the radials. Close to the
inner end of each fossa is an elongated opening ; and there is a similar one at the
centre of the inner end of each radial ; but it is not exposed
until after the removal of the central plug. Ludwig's view,
however, requires that there should be an opening at the inner
lateral angle of each radial and none in the centre, as is reaUy
the case. The inference from this fact is that the primary
interradial cords actually fork in the basals, and that the left
branch of one fork, and the right branch of the next one pass
out together through one of the elongated openings on the
upper surface of the basal ring, which crosses the line of the fm. is.— Diagram of the distribution

111 °^ "'^ axial cords in the calyx of

interbasal suture. I find this to be actually the case. RUzocrinus lo/otensis. b, basais ;

R, radials ; i', first brachials.

Sections through the uppermost part of the basal ring, above

the level of the chambered organ, clearly show the forking of the primary cords,
though no interradial lines of suture are visible at these points, as should be the
case on Ludwig's theory of the composition of the calyx. At the same time I can
readily understand how the use of the section-method alone led him to faU into this
error. For owing to the convexity of the upper surface of the basals, a horizontal
section through their central upper part might pass through the lowest and outer edge of
the radial pentagon, and thus show interradial sutures, although the central portion of
the section with the forking cords really belonged to the basals. The question is not a
very important one ; and but for my having been able to examine a dissected calyx, the
exclusive use of the section-method would have led me to follow Ludwig's description, ex-
cept in one point. He does not seem to have noticed the presence of the intraradial commis-
sures first described by himself in Antedon ; for he neither mentions them, nor introduces
them in his diagram and figures. The special interradial commissures are of course absent,
being unctionally replaced by the secondary interradial cords; but every two of these which
converge within the substance of a radial are united by an intraradial commissure, just as
they are in the Comatulse. This is clearly distinguishable in both the series of horizontal
sections which I have cut through the calyx of Rhizocrinus lofotensis (PI. Villa, fig. 6, ico).
The distribution of the axial cords in the calyx of Rhizocrinus is thus somewhat
difi"erent from that described by Ludwig, as will be seen if the accompanying woodcut
(fig. 18) be compared with fig. 18 on Taf. vi. of his memoir.

254 THE VOYAGE OF H.M.S. CHALLENGER.

There is another point in the anatomy of Rhizocrinus which is not brought out at all
in the semidiagrammatic figure given by Ludwig (Taf. v. fig. 7). The second brachials
(third radials of his nomenclature) are relatively much too small ; but whether he pur-
posely neglected the appearance presented by them, or whether they were really small
and undeveloped in his specimens from reparation after injury, I cannot say. The whole
of the visceral mass is occupied by the winding gut (PL Villa, fig. 8, G) ; so that the
body-cavity is reduced to a minimum. But the outline of the visceral mass is not
circular as represented by Ludwig, for a large diverticulum of the gut extends outwards
between every two brachials. These approach one another over its outer end, so as to
protect it, and it is supported on either side by one of the large processes bordering the
ventral furrows of the brachials which were described and figured by Sars. This is well
shown in the left hand portion of PL Villa, fig. 8 ; while the right hand side shows the
second brachials almost meeting one another over the interradial diverticulum of the
rectum. The visceral mass and third radials of Bathycrinus present the same characters
as seen in PL VII. fig. 4a, and PL Vllb. fig. 7 , the first of which shows the great pro-
cesses on the ventral face of the axillary radial. It would be interesting to determine
whether the axillaries of Bourgueticrinus present similar j)rocesses.

The syzygies of Rhizocrinus, at any rate in the lower parts of the arms, are slightly
different in character from those of other Crinoids. The apposed faces are not completel)^
striated as in Comatula, or even partially so as m Pentacrinus (PL XII. figs. 7, 10, 18, 21 ;
PL XXI. figs. 1 c?, 2(i, 5 a ; PL XXX. figs. 20, 21), as Sars has already pointed out in
the case of Rhizocrinus lofotensis} Neither, however, are they perfectly simple, as is
sometimes the case in Pentacrinus (PL XXVI. figs. 5, 8 ; PL XXXVII. figs. 3, 4 ;
PL L. figs. 6, 7, 12, 13). For there is an indistinct vertical ridge around the opening of
the central canal of the hypozygal, resembling that of a bifascial articulation ; and this
forks at its lower end so as to enclose a somewhat triangular pit into which there fits a
corresponding process of the epizygal. The hypozygal faces which I have found to show
this character most clearly are those of the first brachials of a Rhizocrinus rawsoni from
the Azores (PL X. fig. 8). It is less visible in the corresponding joints of the Havana
specimen (PL X. fig. 6). It likewise appears, though less distinctly, on the first brachials
of a specimen of Rhizocrinus lofotensis from Havana (PL X. fig. 1). Sars makes no
reference to it in his description of this species, but the pit on the distal face of the first
brachial is clearly shown in his Tab. iii. fig. 53, and also in a dorsal view (fig. 54). The
backward projection on the second brachial of Rhizocrinus rawsoni is represented in
PL X. fig. 19 ; while figs. 17 and 18 show the aj^posed faces of a syzygial union farther
out on the arms, the backward process of the epizygal and the corresponding pit on the
hypozygal being very distinct.

A curious peculiarity which is presented by one of the Azores specimens of Rhizocrinus

' Criuoides vivauts, pp. 15, 22.

REPORT ON THE CRINOIDEA. 255

rawsoni is shown in PL X. fig. 8. The adjacent edges of two contiguous first brachials
send out short processes which meet one another and rest upon the fused muscle-plates of
the two radials beneath, so as to cut oft' a tubular space from the body-cavity. I have
found nothing like this in either of the other specimens, which have lost all the visceral
mass together with the second and following brachials (PI. X. fig. 6). The structure in
question may be only accidental, or on the other hand it may perhaps have some relation
to the anal tube.

The presence of the syzygy between the first and second brachials of RMzocrinus
renders it a matter of some difficulty to obtain entire specimens, especially of RMzocrinus
rawsoni. For the arms break away at this point, carrying with them the entire visceral
mass. This is the condition of the individuals represented in PI. X. figs. 1, 6, 8 ; while
fig. 20 shows the arm-bases from the second brachial onwards, and the visceral mass which
they enclose. Two of them have been removed so as to expose the simple digestive
apparatus, with the oral plates around the mouth and a relatively long anal tube. The
position of the genital glands in the pinnules borne by the epizygals of the third syzygial
pair is also well shown.

I am strongly inclined to believe that the loss of the visceral mass and arm-bases may
occur and be made good during life. This appears to have taken place in the Havana
specimen represented in PI. X. fig. 7. Small oral plates are visible in the centre at a
point much below their usual level, which is seen in fig. 20 ; while the second brachials
are so much smaller than the first that the difi'erence is evidently due to fracture and
reparation, just as so often occurs on the arms of Pentacriniis and Comatula. But in
this case all the arms broke away together, carrying the visceral mass with them, so that
a new one had to be developed within the ring of regenerated arms. Not much is to be
seen of it, however, except the oral plates, the replacement of which is a point of
considerable interest. It has long been suspected that eviscerated Comatulae have the
power of restoring their disks ; ^ but this is the first instance of the kind that I have met
with in a Stalked Crinoid.

Owing to the fact that the ring of anchylosed basals in Rhizocriniis was at first
regarded as an enlarged top stem-joint, the affinity between this genus and Bourgueti-
crinus has been supposed to be much closer than can be admitted in the light of our
present knowledge. For in some respects Rkizocrimis stands farther from Bourgneti-
crinus than even Bathycrinus does. The fossil genus Mesocrinus,^ although still but
imperfectly known, is probably nearer Bourgueticrinus than either of these two genera.
But it appears to diff'er from them in the smaller size of the upper stem- joints, a point in
which it resembles both RMzocrinus and Bathycrinus. In the two best known species

1 See Marshall, Quart. Journ. Micr. Sei., 1884, vol. xxiv., N. S., pp. 525, 526.

2 On Two New Crinoids fi-om the Upper Chalk of Southern Sweden, Quart. Journ. Geol. Soc, vol. xxxvii. p. 130,
pi. vi. figs. 1-7.

256 THE VOYAGE OF H.M.S. CHALLENGER.

of Bourgueticrinus (ellipticus and cequalis) the uppermost part of the stem is formed of
joints of an altogether different character from those which are to be found lower do-wii.
The latter are wider than high, with articular faces of a pointed oval shape, the two
diameters of the oval being very unequal in length. Towards the upper part of the stem
this inequality disappears, and the joints become more discoidal, though never thin and
lamellar, as in the highest part of the stem of Rhizocrinus and Bathycrinus (PI. VII.
figs. 2, 1 1 ; PI. Villa, fig. 1 ; PL IX. figs. 1-3 ; PL X. fig. 2). Above these cUscoidal
joints, however, the stem gradually enlarges, and from two to four of the uppermost
joints on which the calyx rests are of great relative size, both height and diameter being
considerable. Sometimes the top joint is highest and sometimes the one below it ; but
at any rate one or more of these large upper joints remain united with the calyx to form
the so-called summit. The upper stem-joints of Apiocrinus are not as a rule much
higher than those below them ; but the diameter often increases considerably from a
point in the stem a little way below the calyx, so that a tolerably large number of joints
enters into the composition of the summit. In species like Apiocrinus crassus,
Ainocrinus magnijicus, and Ajnocrimis onurchisonianus, however, there is scarcely any
enlargement of the stem below the calyx, the uppermost joints, except the highest one
on which the basals rest, being but little if at all larger than those below them. The
same variations appear in Millericrinus. Millericrinus nodotianush.as high upper joints,
while those of Millericrinus simjilex are thin and discoidal externally ; though the upper-
most joint has a large synosteal surface for the reception of the basals which rest upon it.

It seems to me very probable, therefore, that the existence of similar variations wiU
have to be admitted in Bourgueticrinus ; though on the other hand a revision of the
genus may result in the transfer of all the species without enlarged upper stem-joints to
Rhizocrinus or Mesocrinus, especially if the articular faces of their radials are well
developed and not reduced to a minimum as in Bourgueticrinus cequalis. But I am
quite prepared to have to abandon Mesocrinus as a distinct generic type, and to modify
the descriptions of Bourgueticrinus which are given by palaeontologists so as to include
in this genus the two species Mesocrinus Jisheri and Mesocrinus suedicus, on which the
genus Mesocrinus was based.

Even then, however, Apiocrinus and Bourgueticrinus would difi'er from Rhizocrinus
and Bathycrinus in the character of the upper stem-joint. In the two genera last
mentioned, as in Pentacrinus, this joint is the youngest and smallest of the whole stem,
being merely a delicate film of calcareous reticulation which is received into and concealed
by the curved under surface of the basals. In Apiocrinus, however, and in Bourgueti-
crinus this upper joint, though perhaps smaU externally, is large internally and supports
the basals in five large fossse on its surface.

As it closes the calyx below and really belongs thereto, de Loriol ^ has called it the

1 Swiss Crinoids, p. 6; Paleont. Fran?., loc. cit., p. 19.

REPORT ON THE CRINOIDEA. 257

" article basal." It is entirely undeveloped in Rhizocrinus, though this type shows a
slight decrease in the width of the stem a little below the cup, before the commencement
of the gradual downward enlargement, just as is so much more marked in Bourgueticrinus.

There are certain other points in which the two recent species of Rhizocrinus, together
with some fossil ones, differ very markedly from Bourgueticrinus ; and although these
differences hold good for som'^, of the fossil species of Rhizocrinus, it is difficult to say how
far this may be the case with others, owing to their imperfect state of preservation.

In the first place, the basals are of great relative height, often five or six times that
of the radials, and they occupy much the larger part of the exterior of the calyx ;
while the lower stem-joints are usually longer than wide. If they bear radicular cirri
these come off somewhat irregularly from near the ends of the joints, but their sockets
are not verticillate, nor are they ever formed by portions of two apposed joints, as is often
the case in Bourgiieticrimis and Mesocrinus. Rhizocrinus lofotensis and Rhizocrinus
rawsoni show these characters very well (PI. IX. fig. 1 ; PI. X. fig. 15). The latter has
the longer basals, but its stem-joints, though longer than wide, are not so markedly so as
in Rhizocrinus lofotensis. The same is the case with the stem-joints of the so-called
Bourgueticrimis londinensis, which is really a well-defined Rhizocrinus ; while in those of
Conocrinus [Rhizocrinus) suessi and Conocrinus pyriformis the width of the articular
faces is more nearly equal to the length of the joint. In all these species the basals are
longer than the radials, though not greatly so ; but in Bourgueticrinus {Rhizocrinus)
thorenti they are very long, as in the recent Rhizocrinus rawsoni, while the stem-joints
resemble those of Rhizocrinus lofotensis in their proportions.

Owing to the shape of the basals, the calyx of Rhizocrinus is usually cylindrical or
obconical, and though it expands gradually upwards it is nowhere very greatly wider than
the stem, as is the case in Bourgueticnnus, In Conocrinus suessi and in Conocrinus
pyriformis, and perhaps also in Conocrinus seguenzai, it takes on a more ovoid form ; while
in Rhizocrinus raivsoni and Rhizocrinus thorenti it may be very considerably elongated.

So far as I am aware, no true Rhizocrinus has been obtained from any formation
lower than the Eocene. Quenstedt ^ figures some moderately elongated stem-joints of
Apiocrinus constrictus from the White Chalk of Eiigen. But in the absence of a calyx it
is almost impossible to determine these generically, owing to the rarity of the association
of calyces and stem-joints at the same spot. The same is the case with regard to the
Jurassic species of Bourgueticrinus, e.g., Bourgueticrinus ooliticus from the Bradford
Clay, which is perhaps referable to Thiolliericrinus as suggested by de Loriol. The
distinguished Swiss palaeontologist has also described a fossil from the Cretaceous of
Alabama, U.S., as Bourgueticrinus alabamensis.^ It consists only of the basal cone
which " supports the calyx, and which is composed of several enlarging segments of the

1 Encriniden, Tab. 104, figs. 64-66.

2 Description of a New Species of Bourgueticrinus, Journ. Cincinn. Soc Nat. Hist., vol. v. p. 118, pi. v. fig. ].
(ZOOL. CHALL. EXP. PAKT XXill. — 1884.) li 33

258 THE VOYAGE OF H.M.S. CHALLENGER.

columa surmounted by the basal plates." His figure shows five of these uppermost stem-
joints, which are all low and discoidal ; and it consequently appears to me that the fossil
should be referred to Rhizocrinus rather than to Bourgueticrinus. If this be the case,
and its horizon really Cretaceous, this species is of interest as being the only known
instance of a Cretaceous Rhizocrinus.

On the other hand, Bourgueticrinus although abundant in Cretaceous deposits, is not
certainly known to occur in any Tertiary formation. Some of the types described under
this generic name from the Italian Tertiaries have been referred to Conocrinus b)^
Meneghini and others. Among these is the Aioiocrinus cornutus of Schafhautl, which
was doubtfully referred to Bourgueticrinus by Meneghini;^ though Zittel,^ while describ-
ing its calyx as "niedrig schiisselformig," spoke of it as Conocrinus cornutus. I have
been enabled by the kindness of Prof Zittel to examine the calyx of this species for
myself ; and I was interested in finding its shape to be very like that of a singular bowl-
shaped calyx from the London Clay which is preserved in the Natural History Museum.
This has relatively large radials and low basals. I do not see how it can possibly be
placed in the same genus as Conocrinus thorenti or Rhizocrinus rawsoni with their
elongated calyces mainly formed by the long basals ; and I think that it will be necessary
to establish a new genus for the reception of these two sjjecies, to which others will
probably be added when the calyces are found corresponding to some of the other Tertiary
stem-joints that are now referred to Bourgueticrinus in default of further evidence, e.g.,
Bourgueticrinus didymus, Schaur.

Rhizocrinus was supposed by Pourtales to have a considerable resemblance to the
genus Belemnocrinus from the Burlington limestone of Iowa and Illinois. Wachsmuth
and Springer ' have spoken of this resemblance as being very close and interesting, and
stated that " the most important diff'erence, and indeed the only essential distinction
between these genera in their external structure, is found in the solid proboscis and
covered dome of Belemnocrinus." It appears to me, however, that the American
authors lay too much stress on the fact that the calyx is formed in both genera of five
long and narrow basals, and that they have overlooked other and more important
structural characters. In the first place the stems of the two types are totally
difl^erent. That of Belemnocrinus is pentagonal, consisting of short joints with crenu-
lated feces ; while the stem-joints of Rhizocrinus are elongated and more or less dice-
box shaped, with the well known, enlarged and elliptical ends. Stem-joints articulated
like those of Rhizocrinus do indeed occur in the Palaeozoic Platycrinus, and under these
circumstances we may fairly expect that any genetic relationship between Belemnocrinus
and Rhizocrinus would have manifested itself in this character. But the stem of
Belemnocrinus, at any rate of Belemnocrinus jlorifer, seems to have borne successive

' Atti della Soc. Tosc. di Sci. Nat., vol. ii. p. 53. ^ PalKoiitologie, Bd. i. p. 392.

' Revision of the genus Belemnocrinus, and description of two new Species, Anier. Journ. Sci. and Arts, 1877,
vol. cxiii. p. 255.

REPORT ON THE CRINOIDEA. 259

verticils of five cirri, just like that of a recent Pentacrinus ; and this is a much more
important morphological resemblance than the length of the basal plates.

Then again the radials of Rhizocrinus are all in close lateral contact, while Belemno-
crinus has an anal plate intervening between two of the radials and resting on a basal.
It supports a heavy proboscis on its upper foce, while Rhizocrinus has a disk protected by
five insignificant oral plates. The arms of Belemnocrinus are ten in number, while
Rhizocrinus has only five. This, however, is unimportant ; but the arms of Belemno-
crinus bear two rows of pinnules alternately, while in some species these primary pinnules
bear alternating secondary ones, a condition totally unknown in any Neocrinoid.
Altogether, therefore, the structure of the arms of Belemnocrinus is very difi"erent from
that of Rhizocrinus, in which every joint has a syzygial surface at one of its ends ; though
it must be admitted that syzygies are plentiful in the arms of Belemnocrinus, as there
are some species, e.g., Belemnocrinus pourtalesi, in which " throughout the greater portion
of the arms every alternate joint is a syzygium." This character, however, and the
length of the basals are of minor importance compared to the intercalation of the anal
plate in the calyx and the nature of the articulation between the stem-joints, so that I
cannot in any way regard Belemnocrinus as an ancestral form of Rhizocrinus.

1. Rhizocrinus lofotensis, M. Sars, 1864 (PL Villa, figs. 6-8; PI. IX. figs. 1, 2;
PI. X. figs. 1, 2).

1864. Rhizocrinus lofotensis, M. Sars, Forhandl. Vidensk. Selsk., p. 127.

1868. Rhizocrinus lofotensis, M. Sars, Memoires pour sorvir a la connaissance des Crinoides vivants,

p. 38.
1868. Bourgueticrinus Hotessieri, Pourtalfes, BuU. Mus. Comp. Zool., vol. i., No. 7, p. 128.
1872. Rhizocrinus lofotensis, Wyv. Thorns, (pars), Proc. Roy. Soc. Edin., vol. vii. p. 770; The Depths of

the Sea, 1873, pp. 447, 450.
1874. Rhizocrinus lofotensis, Pourtalfes, 111. Cat. Mus. Comp. Zool., vol. iv., No. 8, p. 28.
1882. Rhizocrinus lofotensis, P. H. Carpenter, P>ull. Mus. Comp. Zool., vol. x., No. 4, p. 173.
1884. Rhizocrinus lofotensis, P. H. Carpenter, Proc. Roy. Soc. Edin., vol. xii. p. 356.

Dimensions.

Greatest length of stem (Pourtalfes), ...... 5 inches.

Greatest length of stem (Sars), sixty-seven joints {f) . . . .70 mm.

Greatest length of entire specimen (Sars), . . . . . 80 „

Greatest length of arm, about thirty-five joints (Sars), . . . .11,,

Stem slender, bearing branching radicular cirri on. its lower part, and ending below in
a more or less spreading root. The cirri come ofi" near the terminal faces of the lower
joints at the ends of their longer axes. The joints are markedly dicebox-shaped, and
nearly three times as long as wide.

The calyx is smooth and obconical, of somewhat variable proportions. Basals two or

260 THE VOYAGE OF H.M.S. CHALLENGER.

three times the height of the radials, and closely fused, so that their sutures are rarely
visible. Eadials short and quadrate, with concave upper and convex lower edges ; usually
five in number, but sometimes four, six, or seven. Arms of from thirty to forty joints,
which are imited in pairs by syzygies. First brachials flattened, and quite free laterally,
square or slightly longer than broad. The second similar, but somewhat shorter. The
third still shorter and trapezoidal, so that the arm narrows considerably at this point.
The remaining joints as long or a little longer than wide, and more convex than the
preceding ones. The epizygals which bear the pinnules are rather longer than the
hypozygals and irregularly pentagonal, as the pinnule facets are large relatively to the
length of the joints. The first pinnule is on the eighth brachial or fourth epizygal. The
two lowest pinnule-joints are somewhat trapezoidal in shape, and in contact by their broader
ends. The remaining joints have a medio-dorsal keel and flattened sides, with the edge
of the ventral furrow produced upwards into broad thin plates, especially in the thiixl
and following joints. The peristome is at the level of the second brachial, and protected
by five linguiform oral plates which occupy the central ends of the interpalmar areas.

Colour, in spirit, brownish-white or greyish-white.

Localities. — The Lofoten Islands, 80 to 300 fathoms ; the Josephine Bank.

H.M.S " Lightning," 1868. Station 12. Lat. 59° 36' N., long. 7° 20' W. ; 530 fathoms;
Globigerina ooze; bottom temperature, 4 7° "3 F. Three small specimens without arms.

Station 16. Lat. 61° 2' N., long. 12° 4' W.; 650 fathoms; Globigerina ooze. Two
small specimens without arms.

Stem-fragments were occasionally found in the " cold area " during the cruises of the
" Lightning " and " Porcupine."

H.M.S. Challenger. Station 24a. March 25, 1873, ofi" Culebra Island; lat. 18° 43' N.,
long. 65° 5' W. ; 625 fathoms; Pteropod ooze. Two specimens.

Station 122c. September 10, 1873, off'Barra Grande ; lat. 9° 10' S., long. 34° 49' W. ;
400 fathoms ; red mud. Two specimens.

Station 323. February 28, 1876 ; lat. 35° 39' S., long. 50° 47' W. ; 1900 fathoms;
blue mud; bottom temperature 33°'l F. The occurrence of a stem-fragment here is
recorded in the Station-book, but it has not come into my hands, and I am therefore
unable to speak positively about it.

H.M.S. "Knight Errant," 1880. Station 5. Lat. 59° 26' N., long. 7° 19' W.; 515
fathoms ; mud ; bottom temperature, 45°"4 F. Two young specimens without arms.

Station 6. Lat. 59° 37' N., long. 7° 19' W. ; 530 fathoms; grey mud; bottom
temperature, 46°*5 F. A fragment only.

Rhizocrinus lofotensis has also been dredged several times by the surveying ships of
the U. S. Coast Survey, as recorded in the following list.

SS. "Bibb," May 4, 1868, off the Samboes; 237 fathoms. May 11, 1868, off Sand Key; 248 and 306
fathoms. March 4, 1869, off Cojima, near Havana; 450 fathoms; and several times at lesser depths.

EEPORT ON THE CEINOIDEA. 261

SS. "Blake," 1877-78. No. 29, lat. 24° 36' N., long. 84° 5' W. ; 955 fathoms; bottom temperature, 39^'.
No. 35, lat. 23° 5' 46" N., long. 88° 58' W. ; 804 fathoms; bottom temperature, 40r. No. 43,
lat. 24° 8' N., long. 82° 51' W. ; 339 fathoms; bottom temperature, 40°. No. 44, lat."25° 33' N.,
long. 84° 35' "VV. ; 539 fathoms ; bottom temperature, 39^°. No. 5G, off Havana, lat. 22° 9' N., long.
82° 21' 30" W. ; 175 fathoms.

1878-79. No. 238, ofif Grenadines, 127 fathoms; fine coral sand; bottom temperature, 56°. No. 248, off
Grenada; 161 fathoms; fine grey ooze; temperature, 53J°. No. 259, off Grenada; 159 fathoms; bottom
temperature, 53i°. No. 274, ofi' Barbados ; 209 fathoms ; fine sand and ooze ; bottom temperature, 53|°.

1880. No. 306, lat. 41° 32' 50" N., long. 65° 55' W. ; 524 fathoms.

U. S. Pish Commission, 1882. No. 1124, S.S.E. off Nantucket; 640 fathoms.

Remarks. — An elaborate account of this well-known species lias already been given
by Sars, and I have little to say about it except on one or two points, (l) The sub-
radial portion of the summit is not formed by the top stem -joint as supposed by him, but
it consists of anchylosed basals, as was originally described by Pourtales.^ (2) I have also
followed Pourtales in considering the two joints immediately above the radials as the
two lowest brachials, and not as the second and third (axillary) radials, as they have been
called by Sars, Ludwig, and Wyville Thomson. No matter what the number of arms of
a Crinoid, the so-called second and third radials are morphologically brachials, as I
have already pointed out [ante, pp. 47, 48) ; and though it is convenient for descriptive
purposes to speak of the successive divisions of the rays as radials, distichals, and
palmars, I see no advantage whatever in calling the two lowest arm-joints of a five-armed
Crinoid the second and axillary radials. That they are homologous with the second and
third radials of Comatula and Pentacrinus is undisputed, but these are fundamentally
brachials ; and as it is the distinctive character of an axUlary joint that it should bear
two arms (or arm-divisions) on its distal face, the use of the term " axillary " in the case
of a five-armed Crinoid is misleading and unnecessary.

There is another point in Sars's description that I would just notice. On p. 23 he
says distinctly that the covering plates of the ambulacra occur " sur le disque, aussi bien
que dans toute la longueur des bras et des pinnules." But there is no sign of them in
either of his figures of the disk (Tab. iv. figs. 85, 86, 89). In fact, in fig. 85 no covering
plates are represented at all, though the food-groove is showTi as far as the distal part of
the second brachial ; while in the other two figures the first covering plate is shown
resting on the distal part of the second, or the lower part of the third brachial.

The form of the calyx in this species varies very considerably ; for it is nearly
hemispherical in some specimens and much elongated in others. These last have the
best developed arms; and to some extent, therefore, the forms wath a low and wide cup
must be regarded as premature. But differences of development will not entii'ely account
for the variation, as the calyx of a young specimen found by Sars^ is distinctly higher
(longer) than broad.

' Crinoides vivants. Tab. iv. fig. 95.

2G2 THE VOYAGE OF H.M.S. CHALLENGER.

2. Rhizocrimis rawsoni, Pourtal^s, 1874 (PI. IX. figs. 3-5; PI. X. figs. 3-20;
PI. LIII. figs. 7, 8 ; woodcut, fig. 19).

1872. Rhizocrinus lofotensis, Wyv. Thorns, (pars), Proc. Eoy. Soc. Edin., vol. vii. p. 770; The Depths of

the Sea, 1873, p. 450.
1874. Rhizocrimis Rawsonii, Pourtales, IIL Cat. Mus. Comp. Zool., vol. iv., No. 8, p. 27.

1882. Rhizocrinus rawsoni, P. H. Carpenter, Bull. Mus. Comp. Zodl., vol. x.. No. 4, p. 173.

1883. Democrimts Parfaiii, Perrier, Comptes rendus, t. xcvi., No. 7, p. 450.

1883. Rhizocrimis rawsoni, P. H. Carpenter, Ann, and Mag. Nat. Hist., ser. 5, vol. xi. p. 335.

1884. Rhizocrimis rawsoni, P. H. Carpenter, Proc. Roy. Soc. Edin., voL xii. p. 357.

Dimensions:

Greatest length of entire specimen (Captain Cole),

190-0 mm.

180-0 „

6-0 „

2-5 „

75-0 „

Greatest length of stem, sixty-eight joints ("Blake" specimen),
Greatest length of calyx (" Blake "), ....
Greatest diameter of same calyx, ....
Greatest length of arm, sixty double joints (Captain Colo),

Stem robust, bearing few radicular cirri on its lower part, but ending below in a long
spreading root which attaches itself at intervals. The joints are from once and a half
to twice (rare) as long as broad, and tolerably cylindrical or barrel-shaped in outline.
The planes of the articular ridges at their ends cross one another, but the ends are not
much expanded, so that the dice-box shape is but little marked. The calyx is very
variable in form, sometimes long and slender, sometimes short and broad. The expansion
from below upwards is rarely quite uniform, and there is often a more or less defined
constriction about the level of the basiradial suture. The basals are separated by
distinct sutures, and are generally four or more times the height of the radials, which are
five in number, and more or less distinctly pentagonal.

The arms may have one hundred and twenty joints united in paii's by syzygy. The
first brachials are flattened, quite free laterally, and wider than long. The second are
more nearly square, and the next four shorter but of about the same width, the last one
(or the epizygal of the third syzygial pair) often bearing the first pinnule. The following
hypozygal joints are obliquely oblong ; while the epizygals are more irregular in shape,
and sometimes almost triangular, so as to look like axillaries. The first pinnule is
generally on the sixth or eighth brachial, but sometimes not till the fourteenth. The
two lowest joints of the basal pinnules are broader than their successors, which are
elongated and of gradually diminishing width.

The peristome is about at the level of the sixth brachial, and is protected by small
oral plates.

Colour, in spirit, brownish-white or greyish-white.

Localities.~RM.^. " Porcupine," 1869. Station 42. Ofi" Cape Clear, lat. 49° 12' N.,

' Some other measurements of particular details will be found on p. 265, where also reference is made to the
unusually elongated calyx of the individuals dredged by the " Travailleur."

EEPOllT ON THE CRINOIDEA. 263

long. 12° 52' W. ; 862 fathoms; bottom temperature, 39°7 F. ; ooze with sand and
shells. Two armless specimens.

Station 43. Lat. 50° 1' N., long. 12° 26' W. ; 1207 fathoms ; bottom temperature,
37'''7 F. ; Globigeriua ooze. Two young specimens, one without arms.

H.M.S. Challenger. Station 76. July 3, 1873; south of Terceira (Azores),
lat. 38° 11' N., long. 27° 9' W. ; 900 fathoms ; Pteropod ooze; bottom temperature,
40° F. Three specimens without arms.

Rhizocrinus rawsoni has also been dredged by the surveying ships of the U. S. Coast
Survey, as recorded in the following list.

SS. " Hassler," December 29-30, 1871. Off Sandy Bay, Barbados ; 100 fathoms.

SS. "Blake," 1877-78. No. 32, off Havana; lat. 23° 32' N., long. 88° 5' W. ; 95 to 175 fathoms.

1878-79. No. 155, off Montserrat; 88 fathoms; bottom temperature, 69° F. No. 166, off Guadeloupe; 15C
fathoms; bottom temperature, 59|°. No. 177, off Dominica; 118 fathoms; bottom temjjerature, 65°
fine sand and broken shells. No. 211, off Martinique; 357 fathoms; fine yellow sand and broken
shells. No. 273, off Barbados ; 103 fathoms; bottom temperature, 59|°; coral and broken shells, yellow.
No. 277, off Barbados; 106 fathoms; bottom temperature, 58°; hard rocky bottom. No. 290, off
Barbados; 73 fathoms; bottom temperature, 70J° ; coarse coral sand and broken shells. No. 296, off
Barbados; 84 fathoms ; bottom temperature, 61 i° ; hard bottom. No. 297, off Barbados ; 123 fathoms;
bottom temperature, 56 1° ; calcareous stones.

Telegraph steamer " Investigator," Captain E. Cole. Saba Bank ; 200 fathoms. Fifteen miles N. by E. from
Panama ; 300 fathoms.

The French steamer " Travailleur," 1882, i " 1900m. de profondeur sur les cotes du Maroc, par le travers du cap
Blanc."

Also the "Talisman," 1883,^ " Par le travers du cap Ghir et du cap Noun, k 120 mdles environ de la cote,"
2000 to 2300 metres.

Remarks. — The first examples of this type which were actually obtained were those
dredged by the "Porcupine " in 1869, at depths of 862 and 1207 fathoms off Cape Clear
(Stations 42 and 43). They were, however, considered by Sir Wyville Thomson merely
as unusually large specimens of Rhizocrinus lofotensis ; and the correctness of this
identification would perhaps not have been doubted, but for the discovery in 1871 by the
U, S. Coast Survey steamer " Hassler" of some fine individuals, which Mr. Pourtales
recognised as specifically distinct from Rhizocrinus lofotensis. Two specimens which
were obtained by the Challenger in 900 fathoms among the Azores (Station 76) were at
first referred to Rhizocrinus lofotensis ; but having compared them with the Rhizocrinus
rawsoni of the Caribbean Sea, I find that they likewise belong to that species. It is
generally larger and more robust than Rhizocrinus lofotensis, and the calyx, instead of
being regularly obconical, is extremely variable in form, as will be seen subsequently.

All the specimens that I have seen have been regularly pentamerous, while in
Rhizocrinus lofotensis the number of radials is not unfrequently four or six (PI. Villa,
fig. 7), and may reach seven. Those of Rhizocrinus rawsoni are generally shorter

1 Gomptes rendus, t. xcvi. p. 459. - End., t. xcvii. p. 1392.

264 THE VOYAGE OF H.M.S. CHALLENGER.

relatively to their width, than in Rhizocrinus lofotensis, while the third brachial is oblong
and not trapezoidal, so that there is no sudden narrowing of the arm at the syzygy
between the third and fourth brachials. The shape of the arm-joints too, especially of
those which bear pinnules, is not the same in the two species ; while the pinnules them-
selves, and more particularly those on the proximal parts of the arms, differ very
considerably in appearance, those of Rhizocrinus rawsoni having broad lower joints.

The visceral mass of Rhizocrinus lofotensis is relatively lower than that of Rhizocrinus
rawsoni, in which it is supported by the first six brachials (PL X. fig. 20) ; while in
Rhizocrinus lofotensis the ambulacra leave the peristome at the level of the second
brachials, on to which they pass.

The stem-joints of Rhizocrinus rawsoni are relatively shorter and thicker than those
of Rhizocrinus lofotensis, in which the length is three times the width, and the radicular
cirri at the base of the stem are much more numerous in this species than in Rhizocrinus
raivsoni. In fact there are no radicular cirri whatever in one of the Challenger
specimens of Rhizocrinus rawsoni, and only two, borne upon the first joint above the
root, in one of those dredged by the " Porcupine." In correspondence with this, the root
of Rhizocrinus rawsoni is more like that of Baihycrinus, the lowest stem-joint giving
rise to three or more stout branches, which themselves eventually subdivide and bear
radicular cirri (PL X. fig. 15). This condition appears to be comparatively rare in Rhizo-
crinus lofotensis, in which the lowest stem-joint is often provided with from five to nine
slender cirri, but does not give attachment to a spreading root.

The longest cup yet known in Rhizocrinus rawsoni was found in some individuals
which were dredged by the " Travailleur " in 1882, at a depth of 1900 metres (1000
fathoms), off Cape Blanc, on the coast of Morocco, and were referred by Prof. Perrier
to a new genus Democrinus} The cup is singularly elongated in form, as will be seen
by comparing the measurements kindly furnished me by Prof. Perrier, with those
given below for the Challenger, " Blake," and " Porcupine " specimens.

Democrinus Parfaiti, Terriei = Rhizocrinus rawsoni, Pourtales.

Dimensions.

Length of the calyx from the terminal furrow to the first stem-joint, .
Maximum diameter of the calyx, .....

Height of the radials, ......

Diameter of the stem-joints, ......

Length of the stem-joints, ......

The basals of this type thus form a cup 9 mm. high from its lower extremity to the
constriction at the level of the basiradial suture ; while its maximum diameter is not
more than 2 mm., twice that of the stem-joints.

^ Comptes rendus, t. xcvi., 1883, p. 450.

9-0

mm

2-0

))

0-2

tj

10

jj

2-0

)>

REPOKT ON THE CRINOIDEA.

265

The following table shows the great amount of variation in the proportions of the
hasal tube in other individuals of Rhizocrinus rawsoni, together with its diameter as
compared with that of the stem-joints.

Species.

Rhizocrinus rawsoni.

Rhizocrinus lofotensis.

How
obtained.

"Blake"

Challenger.

Capt. Cole,
"Investigator."

"Porcupine."

Sars.

Deptli.

fathoms.
175

900

300

862
300

Basal tube.

Height.

5-5
5-0

3-5

3-0

2-0

Width.

2-50
2-00

3-00

1-75
1-50

Length

Stem-joints

of Stem.

Number.

Length.

Width.

mm.

180

68

3-50

2-25

150

53

3-00

2 00

145

45

3-50

2-25

50

30

2-25

1-25

70

67

1-50

0-50

N.B. — Pourtal6s described his largest specimen of Rhizucrinus lofotensis as having a stem nearly 130 mm.
long and composed of fifty-nine joints, the length of which averages three times their diameter.

It will be seen from the above table that in absolute size, as well as in the proportions
of the basals and of the stem-joints, the "Porcupine" examples of this type are those
which approach Rhizocrinus lofotensis most nearly ; though the stem is slightly more robust
than in Perrier's specimens which have such an extraordinarily elongated calyx. Both are
smaller than those from the Azores, which are themselves smaller than the Caribbean
specimens (though not always so in the length of the cup), a fact which is doubtless due
to variations of temperature. The difference in size between the largest individuals of
Rhizocrinus lofotensis found by Sars and Pourtales respectively is likewise probably the
result of the difference between the temperature of the Gulf Stream in the Florida Straits
and that of the north-east Atlantic.

The youngest specimens of Rhizocrinus rawsoni which I have seen are those dredged
by the "Porcupine" in 1869 at a depth of 1207 fathoms off Cape Clear (PL LIII.
figs. 7, 8). Each has twenty-eight joints in the stem from the calyx to the root ; but
its length, which is only 20 mm. in the smaller, is 24 '5 mm. in the larger individual.
The uppermost joints are decidedly wider than those below them, the majority of whicli
are elongated and cylindrical, only a few at the base of the stem (more in the larger than
in the smaller individual) having the characteristic dicebox shape, with expanded ends
(PL LIII. fig. 7). The length of the calyx is almost the same in both sjaecimens,
1'8 mm., though its diameter across the radials is greater in that which has the longer
stem. It is mainly formed by the basals, which are 1'2 mm. in height. In the smaller
individual (PL LIII. fig. 7) they expand very slowly upwards to the level of the lowest

(ZOOL. CHALL. EXr. — PAKT XiXU. 1884.)

li 34

266 THE VOYAGE OF H.M.S. CHALLENGER.

angle of the basiradial suture, where a radial rests on two basals ; and then the surface
of the radials slopes outwards rather more rapidly than that of the basal tube below it,
so as to considerably increase the diameter of the cup. All the sutures, interradial,
basiradial, and interbasal, are perfectly distinct ; and there is a very faint circular
constriction of the basal tube rather below its middle, analogous to that described by
Sars in certain individuals of Rhizocrinus lofotensis} This smaller individual has the
lowest portions of the arms preserved, the longest of which has small pinnule-stumps on
the fourteenth and sixteenth brachials, i.e., on the epizygals of the seventh and eighth
syzygial pairs. In the larger individual, however, all the arms are broken away at the
syzygy on the distal faces of the first brachials, which are a little higher and more
trapezoidal in form than those of the smaller example (PI. LIII. fig. 8). The calyx is
also slightly difi'erent in outline. The expansion of the basal tube from below upwards is
a trifle more rapid than in the smaller form, so that its outline is less cylindrical ; while
the radials are bent outwards a little at about one fourth of their height from their lower
angles. This causes the calyx to appear slightly constricted at the highest level of the
basiradial suture, a feature which is very marked in some varieties of the adult form.

As compared with equal sized specimens of Rhizocrimi.^ lofotensis, these young indi-
viduals are distinguished by the relatively great height of the calyx, especially in the
basal tube, and the expansion at the basiradial suture. The cup of Rhizocrinus lofotensis
is not usually so high in projjortion to its width ; and it expands uniformly upwards,
from the stem to the upper margin of the radials, so that its shape is pretty regularly
obconical {PI. IX. figs. 1,2; PI. X. fig. 2).

In Rhizocrinus rawsoni, however, the shape of the calyx varies in a most remarkable
degree. It is elongated (exceeding 9 mm.) and relatively very narrow in Prof. Perrier's
specimens ; while in those lately dredged by Captain Cole ofi" Panama,^ the diameter
varies between 80 and 90 per cent, of the length, which is not more than 4 mm. (wood-
cut, fig. 19). But as a general rule, the form of the calyx may be described as elongated
and subcylindrical. In a few individuals (woodcut, fig. 19, B) it expands uniformly from
below upwards throughout its whole length, as is generally the case in Rhizocrinus lofotensis
(PI. X. fig. 2). Sometimes the width increases very slowly and sometimes more rapidly,
but there is no indication of constriction at any point in the basals or radials. In other
specimens the basals expand slowly but uniformly, and the radials slope outwards more
strongly, as in PL LIII. fig. 7. Sometimes again, the basals widen a little, but then
narrow slightly till they join the radials, which slope outwards so as to again increase
the diameter of the cup.

1 Crinoides vivants, p. 5, Tab i. figs. 35, 39.

2 I am indebted to Prof. F. J. Bell, F.Z.S., for calling my attention to ttose four remarkable specimens which
are divided between the Zoological and Geological Departments at the Museum of Natural History. The keepers of
these departments. Dr. A. Giinther, F.R.S., and Dr. H. Woodward, F.R.S., kindly permitted me to examine them, and
the former gentleman was good enough to allow the accompanying figures to be made of the two abnormal individuals
under his charge.

REPORT ON THE CRINOIDEA.

267

Fig. 19. — The calyx and arm-bases of two specimens of
Rhizocrinus rawsoni from Panama ; x 4. In both cases
the cup is unusually wide in proportion to its height ;
and iu tlie riglit hand specimen (B) the grouping of
the lower brachials is very irregular.

In a considerable number of individuals from different localities on both sides of the

Atlantic, I have found the radials to be marked by a circular furrow of variable depth.

It crosses the body of the radial at the level of

the upper angles of the basals between which

the radial rests, as is well shown in the young

specimen represented on PI. LIII. fig. 8.

In the more mature individual from the

Caribbean Sea, shown in PI. IX. fig. 3, the

furrow crosses the radials at about half their

height ; while in one of the specimens from the

neighbourhood of the Azores (Station 76) the

basals are much less angular at the top, so that

the radials are more nearly oblong and almost

entirely above a rather strong constriction at

the level of the basiradial suture (PI. X. fig. 3).

Other individuals, however, are entirely desti-
tute of any indication of such a constriction.

This is the case, for example, with those

dredged by Captain Cole ofi" Panama, which

are further remarkable, not only for the great

relative width of the basal cup as shown in the woodcut (fig. 19), but also for the extreme

variation in the position of the first pinnule.

In one individual this is on the epizygal of the seventh brachial in two arms, on that
of the fifth in two others, and on that of the fourth in the remaining one. In four arms
of another specimen the epizygals bearing the first pinnule are respectively those of the
third, fourth, fourth, and fifth brachials ; while in a third individual three arms are normal,
with a pinnule on the third epizygal, the two others not bearing a pinnule till the next
(fourth) joint. Lastly, in a fourth individual every one of the arms is developed
abnormally. Three of them are shown in woodcut, fig. 19, B. I will not attempt to
ofi"er an opinion upon the grouj)ing of the syzygies and muscular articulations in this
specimen. But in one arm at least there appear to be two syzygies in succession ; so
that the composite brachial is in three parts instead of in two only. This is a variation
of some interest, as it is normal in the arms of Hyocrinus (PL VI. fig. 1). Both types
of brachial, that with one and that with two syzygies, occur in different species of the
Palaeozoic Heterocrinus, as pointed out already (ante, p. 53).

Although the occurrence of Rhizocrinus rawsoni in the East Atlantic and its more
striking peculiarities, especially the length of the basals, were noticed in the first Eeport
upon the Caribbean Crinoids ^ which Prof. Perrier quotes, he was led to describe the

' Bull. Mus. Comp. Zo'ol., vol. x., No. 4, p. 174.

268 THE VOYAGE OF H.M.S. CHALLENGER.

" Travailleur " examples of this species as a new generic type Democrimis^ which is
really, however, only a synonym of RMzocrinus. His account of Democrinus was as
follows : — ■" Le Democrinus se distingue immediatement de tous les autres genres par la
composition de son calice forme de cinq longues basales constituant h. elles seules un
calice en entonnoir ; un sillon circulaire separe ces cinq basales de cinq radiales rudi-
mentaires, en forme de croissant, alternant avec elles et surmontees elles-memes de cinq
radiales axillaires libres, rectangulaires, mobiles, sur lesquels se fixent respectivement
cinq bras, beaucoup moins larges que les radiales. Ces bras se brisent tres facilement au
niveau de leur articulation avec les radiales axillaires qui se rabattent alors sur la votite
du calice." He further adds that in Rhizocrinus " les basales sont confondues et le calice
forme en partie de radiales." The basals of Rkizocrinus, hoTvever, are very far from being
"confondues," but are large and independent, as was pointed out by Pourtales in 1868
and 1874, and by myself in 1877 and 1882. But Perrier, unaware of this fact, was
unfortunately misled by the erroneous descriptions of the basals as internal and concealed
which were given by Sars and Ludwig {cinte, pp. 249-251); so that when he found
a Rhizocrinus-\\kQ form with long and well defined basals, he naturally (though
erroneously) considered it as new to science.

Although, however, the radials of Democrinus may be small and rudimentary
externally, there is no reason why the calyx should consequently be considered as formed
by the basals alone ; though Perrier regards this as another character distinguishing
Democrimis from Rhizocrinus. In one of the specimens of Rhizocrinus lofotensis which
was figured by Sars ^ the radials are quite small externally ; but they have large distal
faces for the attachment of muscles and ligaments, the inner surfaces of which form the
funnel lodging the lower part of the coelom. The same is undoubtedly the case with
the radials of Democrinus, to which the movable first brachials (axillaries, Perrier) are
attached just as in Rhizocrinus.

It is difiicult to understand why the radials of this type should be considered as
forming part of the calyx, while those of Democrinus are excluded from it on account of
their smaller size. On the same principle one would have to describe the cup of those
species of Antedon in which the first radials do not appear externally, as formed by the
centro-dorsal only !

Prof. Perrier describes the rudimentary radials of Democrimis as separated by a
circular furrow from the basals below them. But a drawing of the type which he has
kindly sent me, shows that while the basiradial suture is marked by five strong
elevations with intervening depressions in which the radials rest, the furrow crosses the
radials at the level of the highest angles of the basals. This furrow is more or less
distinct in various specimens of Rhizocrinus raivsoni (PI. IX. fig. 3 ; PL LIII. fig. 8), as
has been pointed out already. But it can hardly be said to " separate" the radials from

^ Comptes rendus, t. xcvi. p. 450. * Crinoides vivants, Tab. ii. fig. 44.

REPORT ON THE CRINOIDEA. 269

the basals, as it crosses the former at a variable distance from their lower angles. In a
few cases, however, the basiradial suture is more uniformly horizontal, and not marked
by alternate elevations and depressions (PI. X. fig. 3) ; so that the furrow really does
indicate the line of separation between the basals and radials. But this is far from being
the case in Prof. Perrier's drawing of the Democrinus calyx.

The fragmentary condition or absence of the arms in his specimens is nothing unusual.
Only one-third of all the individuals of Rhizocrinus rawsoni which I have examined have
any arms at all, including the young form represented in PI. LIII. fig. 7. There may,
however, be as many as one hundred and twenty joints, or rather sixty syzygial pairs,
with pinnules on all but the first three or four. But they are very apt to break away at
the syzygy in the first brachial, which Perrier speaks of as an articulation between a
radial axillary and the lowest arm-joint. This had happened in two of his three speci-
mens of Democrinus, which are " totalement depoux'vus de bras; le troisieme n'en presente
que des restes tres courts, d'apres lesquels il est aise de voir que les bras devaient ^tre
extremement peu developpes." The drawing of this individual which he has sent me
shows that its longest arm-fragment consists of only five joints, i.e., two composite
brachials and the hypozygal of a third. This fully accounts for the absence of pinnules,
which never appear below the third epizygal in any Rhizocrinus ; and I have little doubt
that further research will prove the existence of properly developed, pinnule-bearing arms
in the so-called Democrinus. But I do not suppose that they are quite as fully developed
as those of the Caribbean variety of Rhizocrinus rawsoni. This has a stem more than
twice the width of that of Democrinus; and it is generally more robust, though the
calyx is distinctly shorter and broader than in Perrier's type.

The " Travailleur " specimens are of interest, both on account of their aberrant form,
and because they give another locality for Rhizocrinus rawsoni in the East Atlantic in
addition to the two discovered by the "Porcupine" in 1869; while the "Talisman"
met with another locality of the type during the dredgings of 1883.^

It is remarkable for its close resemblance to the Rhizocrinus londinensis from the
London Clay, isolated stem-j oints of which were referred by Forbes ^ to Bourguetic7-inus.
But a well preserved and very characteristic calyx has since been discovered, and is now
to be seen in the Natural History Museum at South Kensington.

1 Dertiocrinus dies hard. Perrier's mistake about the condition of tlie basals in Rhizocrinus was pointed out in the
Ann. and Mag. Nat. Hist, ser. 5, voL xi., 1883, p. 334. Under these circumstances the character on which he relied as
distinguishing Democrinus from Rhizocrinus became non-existent ; and I therefore expressed my conviction that
Democrinus Parfaiti and Rhizocrinus rawsoni were identical. Perrier, however, appears to be of a different opinion,
for in the Preliminary Report of Mans. A. Milne-Edwards, the President of the "Talisman" Commission of 1883,
Democrinus is specially nientioned as one of the captures [Goinptes rendus, t. xcvii. p. 1392) ; while in the semi-official
account of the collection published in La Nature (No. 572, p. 391) by Mons. H. Filhol, also a member of the
Commission, particidar reference is made to Democrinus Parfaiti. As the addition of a new generic type to the family
BourgueticrinidiB is of considerable importance in many ways, Prof. Perrier's revised account of its characters will be
awaited with interest, both by zoologists and by pala;outologist3.

2 British Tertiary Echinoderms, p. 36.

270 THE VOYAGE OF H.M.S. CHALLENGER.

Family Pextacrinid^, d'Orbigny, 1852.

Calyx small relatively to the stem and arms, composed of five basals and five radials,
with under-basals in one genus. The rays divide from one to eight or ten times. The
stem bears verticils of cirri at intervals. Two joints are united by syzyg}- at each node,
to the upper one of which the cirri are articulated. The internodes are traversed by five
ligamentous bundles which are interradially disposed and give rise to a more or less
petaloid figure on the joint-faces. No root nor radicular cirri.

Remarks. — Although definitions of this family have already been given by d'Orbignv,
de Loriol, and Zittel, they have been based almost entirely upon palseontological knowledge,
and have not therefore given sufficient prominence to the syzygies between certain of the
stem-joiats, and to the ligamentous structures which produce the well known petaloid
markings on their faces. The regular verticillate arrangement of the cirri along the whole
length of the stem is especially characteristic of the Pentacrinidae among the Xeocrinoids,
though there are a few Palaeocrinoids, e.g., Belemnocrinus Jlorifer, in which this peculiarity
presents itself. But it does not necessarily follow that the nodal joints in the stems of
these older forms are the epizygals of syzygies, as is the case in the Pentacrinidse.

The same may be said of the so-called Encrinus heyrichi, in wliich Picard has described
a verticillate arrangement of the cirri on the stem, without mentioning the presence of
any syzygies at the nodes. ^

In Apiocrinus and Bourgueticnnus the upper part or even the whole of the stem is
entirely free from cirri ; and even when they do occur in verticils, it is only by two at a
time instead of by fives, as in all the recent Pentacrinidse except the one species Penta-
crinus alternicirrus (PI. X5V. ; PL XXVII. figs. 1-3). Further, there is nothing like a
syz}"gy between the two joints forming a node in a Bou)yueticrinus-stem, which are
articulated to one another in the usual way.

It is in fact the characters of the stem, much more than those of the calyx, which
constitute the special distinctive mark of the Pentacrinidse. For although Extracrinus
is known by its under-basals, the composition of the calyx is identical in Millencrinxjis
and Pe Iliac rin us, and also in Metacrinus, if we foUow strict morphology and consider
the second radials as really arm-joints.

The calyx of Balanocrinus is unfortunately not yet known. The genus was founded
by Agassiz for a fragment in the Basle Museum, which he supposed to be a peculiar form
of calyx. But de LorioP has shown that " ce pretendu calice n'est qu'un fragment de
tige attaqu^ et deforme par un parasite." Agassiz had, however, referred the stems
associated with it to Pentacrinus suhteres; and de Loriol, finding that the stem-joints of

' Ueber erne neue Crinoiden-Art aus dem MuschelkaU: der Hainleite bei Sondershaufen, Zeitschr. d. deutsch. geoL
GeselUck., Jahig. 18S3, p. 201.
* Swiss Crinoids, p. 163.

REPORT ON THE CRIN'OIDEA. 271

this type really are different from those of other Pentacrinidae, has re-established the genus
Balanocrinus upon them. The five sectors of each more or less circular face have no
ridses or denticulations along their sides, those beinor limited to the outer marmn of the
joint-face. They are usually therefore of greater size than the corresponding parts on
the stem-joints of Extracrinus and Pentacrinus, which are sometimes much constricted
by the development of ridges at their sides. The stem-joints of Balanocrinus, therefore,
are somewhat like those of MUlericrinus ; though in the latter type the whole joint-
face is uniform in character, and not marked out into sectors as is the case in the Penta-
crinidae. Many PaUeocrinoids have joints somewhat like those of Balanocrinus, i.e.,
crenulated round the edge, but nothing more. The genus commenced with Pentacrinus
in the Trias, and survived to the Lower Xeocomian, no remains of it having vet been found
at any higher horizon ; while I have not met with this simple form of stem-joint in any
recent species.

Owing to the deficiency of our knowledge respecting the nature of the calj-x of
Balanocrinus, I have found it necessary to use the characters of the stem as the basis of
the classification of the family. Balanocrinus may have under-basals Kke Extracrinus,
or more than three radials like Metacrinus ; but until we know more about its calyx a
classification of the Pentacrinidae must depend primarily upon the varying features of the
stem.

I. Five to eight large teeth at the eides of each petaloid sector, most of which start from the outer edge
of the joint-face, while the remainder meet their fellows in the interpalmar spaces.

1. Three radials, ......... Pentacrinus.

2. More than three radials, ....... Meiaerinug.

IL Sectors linear with delicately crenulated edges. Under-basals. The first radials much prolonged

downwards. Secondary arm-trunks each bear a succession of armlets on the same side, Extraeriniu.
IIL Joint-faces crenulated round the edge only, not along the sides of the sectors, . Balanocrinus.

Three other supposed genera have also been referred to this family. One is Isocrinus,
von Meyer,^ of which the stem is scarcelv known : while it is probable that von Meyer's
description of the primary rays as consisting of but two joints, basals being likewise
wanting, is also somewhat incorrect. Tlie mode of division of the rays, on which von
Meyer laid considerable stress, is perfectly normal. I prefer therefore to refer the tjrpe,
temporarily at any rate, to the genus PerUacrinus, as has been done by Bronn and
others, though I will not attempt to foUow them into specific details.

Another unrecofirnised crenus of the Pentacrinidae is the Chladocrinus of L. Agassiz.*
After defining the stem of Pentacrinus as "portant de distance en distance des rayons
simples verticilles," he continued, "on pourra designer sous le nom de Chladocrinus les
especes dont les rayons accessoires forment des verticilles plus ou moins distans."

' Jsoerintis und Chehcrinus, Mnseiun Senctenhergianum, Frankfort, 1837, p. 251.

* Prodrome d'une Monographie des Baxliaiies ou Echinodermes, Mem. de la Soc des Sci. Xai. dc Xeuchaifl, t. i.,
1835, p. 194.

272 THE VOYAGE OF H.M.S. CHALLENGER. .

No further definitiou of Chladocrinus was ever given by Prof. L. Agassiz ; and it is
not surprising therefore that the genus has never been accepted by naturalists. The
remaining type which is supposed to be generically distinct from PentacHnus, is the
Cainocrinus of Forbes.^ It has recently been revived by de Loriol ; " but since it was
based on a misconception, and its only distinctive character depends upon a feature
which is very variable among the recent species, viz., the presence or absence of a closed
ring of basals, I see no good in retaining it (see pp. 281-283). Practically, therefore,
owing to the well marked characters of Extracrinus and our want of knowledge of
Balanocrinus, a definition of Pentacrinus for the study of recent forrtis need only
emphasise those points in which it diifers from Metacrinus. I have, however, referred
to one or two characters in which the genus differs from Extracrinus.

Genus Pentacrinus, Miller, 1821.^

176L Palmier marln, Guettard, Memoires de Mathematique et de Physique tir^s des Eegistres derAcademie

Eoyale des Sciences, de I'annee MDCCLV., Paris, 1761, p. 225.

1762. Encrinus, Ellis, Phil. Trans., vol. lii. pt. i. for the year 1761, London, 1762, p. 35S.

1766. Isis, Linnaeus, Systenia Natural, ed. xii., Holmiffi, 1766, t. i. p. 1288.

1816. Encrinus, Lamarck, Histoire Naturelle des Animaux sans Vertebras, t. ii., Paris, 1816, p. 432.

1820. Pentacrinites, von Sehlotheim, Die Petrefactenkunde, Gotha, 1820, p. 327.

1821. Pentacrinus, Miller, A Natural History of the Crinoidea, Bristol, 1821, p. 45.
1832. Pentacrinites, Goldfuss, Petrefacta Germaniae, Dusseldorf, 1832, t. i. p. 168.
1832. Solanocrinites, Goldfuss, Ibid., p. 168.

1834. Pentacrinus, de BlainvOle, Manuel d'Actinologie, Paris, 1834, p. 257.

1834. Encrinus, de Blainville, Ibid., p. 254.

1835. Pentacrinus, Agassiz, M^m. de la Soc. d. Sci. Nat. de Neuchatel, t. i. p. 194.

1835. Chladocrinus, Agassiz, Ibid., p. 195.

1836. Pentacrinus, Buckland, Geology and Mineralogy, London, 1836, vol i. p. 432.

1837. Isocrinus, von Meyer, Museum Senckeubergianum, Frankfurt, ii. p. 251.

1843. Pentacrinus, MliUer, Abhandl. d. k. Akad. d. Wiss. Berlin, Jahrg. 1841, p. 177.

1845. Pentacrinus, Austin, A Monograph on Eecent and Fossil Crinoidea, Bristol, 1843-45, p. 110.

1845. Pentacrinus, Desor, Bull Soc. d. Sci. Nat. de Neuchatel, vol. i. pp. 213, 214.

1845. Isocrinus, Desor {non von Meyer), Ibid., p. 213.

1845. Balanocrinus, Agassiz {non de Loriol), in Desor, Ibid., p. 214.

1847. Pentacrinus, d'Orbigny, Cours element, de Paliiontol. et de Geol. stratigr., t. ii. Fasc. 1, Paris, 1852,

p. 149.
1852. Isocrinus, d'Orbigny, Ibid., p. 149.

' British Tertiary Echinoderms, p. 33. ° Swiss Crinoids, pp. Ill, 112.

5 The above Ust contains, I believe, all the most important references to the recent Pentacrinus since the time of
Guettard, together with notices of the chief pateontological works in which this type and its fossil representatives
are mentioned. But it makes no pretence whatever of recording all the various names which have been bestowed at
dififerent times upon fragments of fossil Pentacrinidae. A task of this kind is scarcely worth undertaking, as the result
would be totally incommensurate with the labour involved.

REPORT ON THE CRINOIDEA. 273

1852. Pentacrinus, Forbes, Monograph of the Echinodermata of the British Tertiaries, p. 33.

1852. Cainocrinus, Forbes, Ihid., p. 34.

1857. Pentacrinus, Pictet, Trait6 de PaMontologie, 2°" 6d., Paris, 1857, t. iv. p. 342.

1857. Isocrinus, Pictet, Ibid., p. 344.

1857. Comatida (pars), Pictet, Ibid:, p. 288.

1862. Pentacrinus, Dujardiu and Hup(5, Hist. Nat. des Zoophytes, fchinodermes, Paris, 1862, p. 179.

1864. Cenocrinus, Wyv. Thorns., The Intellectual Observer, August 1864, p. 3.

1864. Neocrinus, Wyv. Thorns., Ibid., p. 7.

1864. Pentacrinus, Liitken, Videnskabelige Meddelelser fra den natuihistoriske Forening i Kjobenhavn, 1864,

Nr. 13-16, p. 207.
1872. Pentacrinus, Wyv. Thorns., Proo. Roy. Soc. Ediu., vol. vii. p. 765; and The Dej)ths of the Sea,

p. 435.

1875. Picteticrinus, do Loriol, Monographie Paleontologique et G^ologique des Etages Superieurs de la forma-

tion Jurassique des Environs de Boulogne-sur-Mer, 2"'° partie, p. 297.

1876. Pentacrinus, Quenstedt, Petrefactenkunde Deutschlands, Bd. iv. ; Asteriden und Encriniden, p. 186.
1879. Pentacrinus, de Loriol, Monographie des Crinoides fossiles de la Suisse, p. 114.

1879. Cainocrinus, de Loriol, Ibid., p. 111.

1879. Pentacrinus, Zittel, Handbuch der Palseontologie, PaliBozoologie, Miinchen und Leipzig, 1876-1880,

Bd. i. p. 393.

1880. Pentacrinus, P. H. Carpenter, Journ. Linn. Soc. Lond. (Zool.), vol xv. p. 210.
1882. Pentacrinus, P. H. Carpenter, Bull. Mus. Comp. Zool., vol. x. p. 167.

1884. Pentacrinus, P. H. Carpenter, Proc. Roy. Soc. Ediu., vol. xii. p. 355.

A. CJiaracters of the Gemis.-

The petaloid sectors on the faces of the stem-joints are bordered by a few large
ridges, of which the smaller proximal ones meet those of adjacent sectors in the inter-
petaloid spaces, while the large distal ridges reach the outer edge of the joint. The
supranodal joints are scarcely modified for the cirrus-sockets, and the articular facets
rarely reach the upper edges of the nodal joints.^ The cirri consist of from twenty to
fifty joints, and vary considerably both in length and in appearance.

The basals may be very small knobs, or form a complete ring, and have but a slight
tendency to downward extensions of their lower angles. There are never more than
three radials, none of which bear pinnules.

Tlie rays may divide five times, but rarely more than thrice ; and their subdivisions
are equal in value or nearly so. The basal joints of the lower pinnules are usually
rather broad and flattened laterally, with sharp dorsal edges.

Remarks. — The genus Pentacrinus is generally, and with good reason, attributed to
Miller. But a few authors have associated with it the name of von Schlotheim." It is
true that this able palaeontologist used Pentacrinites as a generic name a year before the
publication of Miller's classical monograph ; but he made no attempt to define it as MiUer
did, nor did he give diagnoses of any of the species which he referred to the genus. In

' In several fossil species, however, the ciiTus-facets take up the whole height of the nodal joints.
2 Die Petrefactenkunde, Gotha, 1820, p. 327.
(zool. chall. EXP. — PART xsxu. — 1884.) li 35

274 THE VOYAGE OF H.M.S. CHALLENGER.

fact, the first of these, Pentacrinites vulgaris, designates a fossil type to which he also
referred (under the name of Encrinus caput-Meduscp) the recent specimens described by
Guettard and Ellis. These were subsequently referred by Miller to his Pentaci'inus
caput-Medusce { = Pentacrinus asterius, Linn, sp.) ; and the type was eventually rendered
classical by the researches of Miiller. Under these circumstances I see no reason for
departing from the practice of d'Orbigny, Forbes, Pictet, de Loriol, and Zittel, and have
therefore attributed the genus to Miller, with the date 1821. When establishing it,
he simply converted into a generic designation the name which had long been commonly
employed for fragments of stems with the characteristic petaloid markings on their
terminal faces. Miller's generic diagnosis of this type, like those of the numerous other
Crinoids described by him, corresponds to the definition of a family, when considered by
the help of our present knowledge.

Five species were established by Miller' — (1) the recent Pentacrinus caput-Medusce
from the West Indies ; (2) the two fossil species from the Lias, Pentacrinus hriareus
and Pentacrimis subangidaris ; and (3) two other fossils which need not be considered
here. Although apparently taking the single recent species then known as the type of
the genus, he gave a generic diagnosis which represents, although imperfectly, the
dissected calyx of one of the two Liassic species. These have the radials prolonged
downwards over the upper stem-joints between and below the outer ends of the basals ;
and the Messrs. Austin consequently proposed to establish the new genus Extracrinus
for their reception, while restricting Pentacrinus to species having the general character
of the recent Pentacrinus caput-Medusce { = Pejitacrinus asterius, Linn, sp.).

Miller described the " jielvis " of the fossil Pentacrinus hriareus and Pentacrinus
suhangularis as similar in character to that of the recent Pentacrinus asterius, namely, as
consisting of five small and nearly cuneiform basals in contact by their central ends.
The Messrs. Austin, in accordance with their peculiar method of nomenclature, gave the
name " dorsocentral plate" to the pelvis of Miller (basals, Miiller) ; and they described
that of Pentacrinus asterius as " resembling an enlarged and thickened supracolumnar
joint," without divisions, the salient angles of which alternate with the five first
radials, or, as they called them, the first series of perisomic pieces. The pelvis of the three
fossil species Pentctcrinus johnsoni, Pentacrinus tuherculatus, and Pentacrinus milleri,
was described as closely resembling that of Pentacrinus asterius ; but in their diagnoses
of Extracrinus hriareus and Extracrinus siihangularis they differed considerably from
Miller and Goldfuss. They gave the name dorsocentral plate, not as usual to the pelvis
of Miller as in Pentacrinus asterius, but to five small and nearly concealed pieces which
are placed beneath the true pelvis, and were unknown to Miller. They are radial in
position, and not interradial like the pieces described by Miller and Goldfuss as composing
the pelvis of these two Liassic species. These, the true basals, which thus alternate with
' A Monograph on Recent and Fossil Crinoidea, Bristol, 1843-45, p. 95.

REPORT ON THE CRINOIDEA. 275

the elements of the so-called dorsocentral plate, were described by the Messrs. Austin as
" solid pointed pieces, whose points pass outward, and rest on the salient angle of the
pentagonal column," precisely, in fact, like the rays of the stellate dorsocentral plate or
pelvis of Pentacrinus asterius, with which Miller had rightly regarded them as
homologous. The Messrs. Austin, however, supposed Miller to have been in error upon
this point, as they believed the so-called dorsocentral plate of Extracrinus to be
homologous with that of Pentacrinus, not paying any attention whatever to its position
with regard to the radial symmetry of the animal. Neither did they notice that in
Pentacrinus the five elements composing the dorsocentral plate are perforated by bifur-
cating canals, which occur in Extracrinus, not in the component pieces of the dorso-
central plate, but in those of the next series, the pelvis of Miller ; and they were conse-
quently led to regard these last, the basals of Miillcr, as representing the first series of
perisomic plates (or the first radials, Miiller) of Pentacrinus ; while the first radials of
Extracrinus, alternating in position with the basals, were described by them as a second
series of lateral or perisomic pieces which are unrepresented in Pentacrinus. The fact is,
however, that it is the dorsocentral plate of the former genus which is not represented
in Pentacrinus, the pelvis of which represents the so-called first series of perisomic plates
in Ext7'aci'inus, as was rightly supposed by Miller. The two sets of plates are precisely
similar in their position relatively to the radial symmetry of the animal and in being
perforated in the same way by bifurcating canals. These two important points, which
were entirely left out of consideration by the Messrs. Austin, demonstrate the homology
of the first series of perisomic plates in Extracrinus, not only with the pelvis or dorso-
central plate of Pentacrimis, but also with the outer circlet of basals in Encrinus ; while
the five small, nearly concealed pieces forming the so-called dorsocentral plate of
Extracrinus, which alternate with the true or outer basals, obviously represent the inner
circlet of basals of Encrinus. The Messrs. Austin were fortunate enough to obtain
a specimen of Extracrinus hriareus showing the interior of the calyx, the centre of the
floor of which is occupied by the five small, radially placed elements of the so-called
dorsocentral plate ; and the resemblance in every respect between these and the inner
circlet of basals in Encrinus is so close as to leave little doubt that they are homologous
with one another.

It is therefore somewhat striking to find that the extensive downward prolongation
of the first radials over the upper stem-joints, which is the chief characteristic of
Extracrinus, as defined by the Messrs. Austin, is also found, though to a smaller extent,
in Encrinus, as pointed out by Beyrich. Encrinus, like Extracrinus, has a dicycHc base ;
but the same tendency to downward prolongation of the radials is seen in Pentacrinus
naresiamis (PI. XXX. fig. l) and more distinctly in Metacrinus (PI. XXXIX. fig. 1;
PL XLIX. fig. 2).

The Austins' genus Extracrinus has not met with the attention which it deserved,

276 THE VOYAGE OF H.M.S. CHALLENGER.

possibly on account of the clumsy nomenclature adopted by these authors. Eamsay,
Phillips, and Etheridge have used it in this country ; but the well known German
palaeontologist Quenstedt/ who has probably seen it in greater abundance than any other
naturalist, speaks of it in a decidedly scoffing tone. This, however, is only to be expected
from one who does not separate Bourgueticrinus from Apiocrimis. Zittel mentions
Austin's name, but without committing himself to an acceptance of it ; and I have reason
to believe that M. de Loriol is prepared to accept the genus after it has been redefined
in a manner which is consistent with the present state of our knowledge.

The genus Extracrinus, which includes the two groups " Briariden " and " Subangu-
laren " of Quenstedt, is thus distinguished by the presence of under-basals (dorsocentral
plate, Austin) and the downward prolongation and jointing of the radials. But it also
diflPers very markedly from the Post-Liassic and recent Pentacrinites in the characters of
its stem and arm-divisions.

The precise structure of the stem of Extracrinus, i.e., the relations of its nodal and
internodal joints, has yet to be worked out. Some fragments of stem, probably from
near the top, seem to consist entirely of nodal joints ; while in others there are several
joints between any two whorls of cirri, just as in the ordinary Pentacrinites. The joint-
faces of Extracrinus, however, are very different from those of Pentacrimis. In the
latter genus, the five petaloid figures indicating the position of the stem ligaments are
more or less oval in shape, pointed at one or both ends, and bounded by strong ridge-
like processes, with alternating furrows (PI. XIII. figs. 2-6, 9-11 ; PL XXII.
figs. 13, 14, 22; PI. XXX. figs. 25-30; PL XXXa. fig. 7; PL XXXII. fig. 3;
PL XXXVII. fig. 22). There may be only about three of these on each side of the
petal, as in most recent forms, or there may be from six to twelve of a smaller size.
But all of these ridges, except those most centrally placed, slant inwards from the edge
of the joint, where their outer ends cause the denticulation which is so marked in all but
the oldest specimens (PL XIX. figs. 2-5 ; PL XXVII. fig. 1 ; PL XXXa. fig. 6 ;
PL XXXI. fig. 3 ; Pis. XXXV., XXXVL).

In Extracrinus, however, the five interradial petals are quite narrow, and much less
distinctly oval than in Pentacrinus, sometimes being linear with rounded outer ends.
Even when the joints are stellate the petals do not occupy more than the central portion
of each ray of the star, instead of the whole of it as in Pentacrinus ; and when the joints
are pentagonal or circular there is a more or less triangular space between every two
petals, which is plain and devoid of sculpture. Then again the markings at the sides of
the petals are much more delicate than in Pentacrinus, having more the character of
stria3 or crenulation than of coarse ridges. They are also much more numerous than in
Pentacrinus, and are strictly limited to the sides of the petals, not reaching the outer
edge of the joint. These characters are well shown in several of the figures published

1 Encrmiden, p. 270.

REPORT ON THE CRINOIDEA. 277

by different palaeontologists, e.g., on pi. 53 of Buckland's Geology and Mineralogy,
figs. 9-13 ; on Tab. 101 of Quenstedt's Encriniden, especially figs. 24, 27, 33, and 37;
and also on pi. 12 of the Messrs. Aiistin's Monograph, figs, d,f, k, n, o, r. None of these
authors, however, seem to have noticed the distinction of this type of joint from that of
the ordinary Pentacrinidse, either recent or fossil ; and attention was first drawn to it by
de Loriol, as will be pointed out subsequently.

In the ordinary PentacrinidEe, as in the multiradiate Comatulse, there is no special
regularity in the mode of division of the secondary and tertiary arms. The twenty
secondary arms borne upon the distichal axillaries may either remain simple, or fork
again, once or oftener. But in most cases the forking is very irregular. Secondary
(palmar) axillaries may appear upon any of the four secondary arms; and the two
tertiary arms borne by them are of equal size, and have equal power of forking again,
though as a rule they do not all do so. A good instance of this is shown in Miller's
figure ^ of Pentacrinus asterius {caput- Medusce) , which is represented more diagrammati-
cally in Quenstedt's Tab. 97, fig. 3. A similar arrangement has been described by
Liitken in Pentacrinus miilleri,^ while it also occurs in Pentacrinus maclearcmits,
Pentacrinus ivyville-tliomsoni, and Pentacrinus alternicirrus (PI. XIV. ; PI. XV. fig. 1 ;
PI. XIX. fig. 2 ; PI. XXV. ; PI. XXVI. fig. 4). In all these cases the secondary (and
tertiary) axillaries are limited to the outer arms of each successive pair, so that the arrange-
ment of the arms on the ray is 2, 1 ; 1, 2 ; or 2, 1, 1 ; 1, 1, 2.^ But the two (or four)
inner arms are equivalent to the outer ones in all respects, neither of them dividing again.

"While the arm-division in Extracrinus proceeds to a much greater extent than in
Pentacri7ius, it is confined as a rule to the sides of the ray, only the outer arm of each
successive pair bearing axillaries, just as in the secondary and tertiary arms of
Pentacrinus asterius, Pentacrinus miilJeri, &e. The four tertiary arms which spring
from each pair of palmar axiUaries are rarely of equal size, and never absolutely equi-
valent. The two inner ones are usually rather the smaller, and except in some forms of
Extracrinus suhangidaris do not divide again. Each of the larger outer arms, however,
divides again after a few joints, but the division is unequal. The smaller inner face of
the axillary, i.e., that turned towards the other axillary, bears a slender armlet; while the
main arm-trunk is continued directly onwards without change of direction. It gradually
diminishes in size, and gives off" at short intervals a series of slender armlets from its
inner side, but it never really forks. But for the pinnules borne by it and its
subordinate armlets, one would be almost inclined to say that the distichal axillary bears
two secondary arms which have long slender pinnules placed at intervals upon their
adjacent inner faces, but none whatever upon their outer sides. These organs are real

1 Op. cit., p. 48, pi. i. ^ Om Vestindiens Pentacriner, loc. cit, pp. 203, 204.

' The Caribbean Antedon spinifera often shows exactly the reverse condition to this. Palmar axillaries are frequently
only developed on the inner pair of the four secondary arms, so that the grouping on each ray is 1, 2 ; 2, 1.

278 THE VOYAGE OF H.M.S. CHALLENGER.

arms, however, for they bear the pinnules as their larger fellows do ; and in some forms
the lowest of them (tertiary arms) have a series of unequal axillaries and bear armlets on
their inner faces just as the large outer arms do. In such individuals these inner tertiary
arms are more equal in size to the outer pair. The two extreme types are well represented
on Tab. 97 of the Encriniden, figs. 5 and 6, by Quenstedt, who partially founds upon
them the distinction between the " Briariden " and the " Subangularen." In the former
group the inner tertiary arms are undivided armlets like those which come off farther out
on the ray ; while the inner tertiary arms of the Subangularen, though smaller than the
outer ones, bear armlets on their inner faces, which correspond to those on the inner faces
of the outer arms that spring from the same axillaries.

Owing to the presence of these armlets on the inner tertiary arms, the Subangularen
generally have the " finger-reichsten Krone " as pointed out by Quenstedt. This is not
always the case, however, for in a specimen from the Posidonia-heds of Holzmaden, which
is figured by him,^ the inner tertiary arms are undivided, and their successors are more
equal to the outer arms of the ray than in some forms of Extracrimis hriareus.

Under the name of Pentacrinus briareus imnutus, Quenstedt * has figured a curious
little species in which the division of the arms seems to be somewhat irregular, and the
distinction of arms and armlets less marked than is usually apparent in ExtrcLcrinus.
But I do not think that this variation, even if it be established, need have much effect
upon the stability of Extracrinus as a generic type.

The differences between the Liassic and the recent Pentacrinidse on which the genus
was founded by the Messrs. Austin were at first regarded by the late Sir Wyville
Thomson as of merely subgeneric value. Believing that Pentacrinus hriareus " seems to
have a just claim to be recognised as the type of the genus Pentacrinus," he introduced
the name Cenocrinus for the Pentacrinus caput-Medusce of Miiller, and one or two fossils
which closely resemble it.^ He subsequently abandoned this name, however, and referred
the type to Pentacrinus as all later writers have done, some recognising Extracrinus as
a separate genus and some not. The Messrs. Austin^ pointed out that Miller "in his
arrangement of the Crinoidea has taken the PeMacrirnis caput-MeduscB for the typical
species, while at the same time his generic plate represents the dissected skeleton of quite
a different Crinoid. In the hope to remedy this intermingling of genera, we propose to
retain Miller's genus Pentacrinus, and to continue the Pentacrinus caput-Medusce as the
type of the genus;" while the name ^xiracnnws was proposed for the Liassic Pentacrinus
hriareus and Pentacrinus suhangidaris. This arrangement seems decidedly preferable
to that proposed by Sir Wyville, who eventually gave up Cenocrinus as a subgenus ;
though I cannot learn that he ever formally adopted Extracrinus.

A second subgenus of Pentacrinus besides Cenocrinus was also proposed by Sir

1 Encriniden, Tab. 101, fig. 1. - Ibid., Tab. 99, fig. 177.

2 Sea Lilies, The Intellectual Observer, August 18G4, p. 3. * Monograph, p.

REPORT ON THE CRINOIDEA. 279

Wyville in 1864,^ the type being a new West Indian species which he designated as
Pentacrinus (Neocrinus) decorus. The differences between this type and Pentacrinus
asterius {caput- Mechisce) are undoubtedly considerable, as I have expressed by separating
the two as far as possible in my arrangement of the species (see p. 299). But they are rather
physiological than morphological in chai'acter, and one or two errors of observation have
caused them to appear greater than they really are. There are syzygies below the nodal
joints of Pentacrinus asterius (PL XIII. figs. 3, 5, 8), just as in Pentacrinus decorus
(PI. XXXVI.; PL XXXVIL figs. 5-8, 12, 19, 21) ; so that although the stem of the
former species is very strong and rigid, it can hardly be said to be distinguished by tke
" absence of all provision for its rupture," such as Sir Wyville^ described in the more
slender stem of Pentacrinus decorus.

The supposed difference between the disks of Penta^ririus asterius and Pentacrinus
decorus, which was founded on a belief in the presence of tooth-like oral plates in the
former genus, is also due to error, owing to the unsatisfactory condition of Michelin's
specimen, which was said to possess these embryonic structures. Sir Wyville recognised
this subsequently when he obtained a spirit specimen of Pentacrinus asterius, the disk
of which he described as follows : ^ — " The perisome of the disk is covered with irregular
calcareous plates, and at the free inner angles of the interradial spaces these plates become
closer, and form a solid kind of boss ; but there are no distinct oral plates." On the
other hand, the disk of Pentojcrinus decorus* (PL XXXIV. fig. 2) could hardly be
called " comparatively unprotected " as distinguished from that of Pentacrinus asterius,
which Sir Wyville described on the previous page as " uniformly defended and plated
with calcareous pavement."

In the nature of the arms, however, there is a considerable difference between the two
types, as was well described by Sir Wyville. Those of Pentacrinus asterius are " greatly
multiplied, large and strong. No syzygies, save those at the base, which can be used on an
emergency, tend to diminish their strength, an arrangement essential to the full supply
of food in their fixed condition." On the other hand, in Pentacrinus decorus the number
of arms is "greatly less, and the arms are provided throughout with syzygies, an
arrangement apparently suitable to its greater liability to trivial accidents in its free
condition." He went on to say, " At first I had some doubt as to the propriety of making
this species the type of a new subgenus, and any one of the above characters would
certainly not have afforded sufficient grounds ; but all these characters taken together
form a remarkably compact assemblage, which places Neocrinus in a directly intermediate
position between Cenocrinus and Comatula." Two of the principal points of difference
between Cenocrinus and Neocrinus have, however, no foundation in fact, while a third
is, at most, one of specific value ; and the fourth, the supposed difference in the mode of

I Sea Lilies, Tlu: Intellectual Observer, August 1864, p. 7. ' Ibid., p. 10.

3 Proc. Boy. Soc. Edin., vol. vii., 1872, p. 760. * Sea Lilies, The Intelleciual Observer, August 1864, p. U.

280 THE VOYAGE OF H.M.S. CHALLENGER.

life, is valueless. For individuals of Pentacrinus decorus have been found attached to
teleoraph cables by a spreading base ; and one specimen of Pentacrinus asterius at any
rate, which I have seen, had the stem broken at a nodal joint, which was worn and rounded
below, its central canal being closed up by a small median tubercle ; while this condition
is common to several other Pentaerinidae, as I have pointed out already {ante, pp. 18-22).

Apart from the length of the internodes and the characters of the stem-joints, cirri,
and arms, all of which are merely of specific value, the chief difference between
Pentacrinus asterius and Pentacrinus decorus is in the mode of union of the two outer
radials. In the latter type, as shown in PL XXXIV. figs. 3 and 5 (wdiich were drawn
under Sir Wyville's own direction), these joints are united by a bifascial articu-
lation. But in Pentacrinus asterius (PL XII. figs. 18 and 21), and also in Pentacrinus
miilleri and Pentacrinus wyville-thomsoni (PL XVIII. figs. 8, 11), there is a syzygy in
this position. This difi"erence, however, is one which occurs continually among the
numerous species of the Comatulid genera. Antedon rosacea and Actinometra meridion-
alis are types of many species having the bifascial articulation ; while Antedon Jluctuans^
and Actinometra Solaris represent a smaller number of species which have the syzygy. I
see no reason, therefore, for considering this difi'erence as one of subgeneric value among
the Pentacrinidge, so as to separate Pentacrinus decorus, together with Pentacrinus hlakei
and Pentacrinus naresianus under a separate name, Neocrinus, from the other five species
which have a syzygy between the two outer radials. Four of these, and probably
Pentacrinus asterius as well, become free at a certain period of their life, just as Sir
Wyville discovered to be the case in Penta.cHnus decorus; so that one of the physiological
characters on which he relied as giving Neocrinus an intermediate position between Penta-
crinus asterius and the Comatulse is of much more general occurrence than he supposed.

The separation of Pentacrinus asterius and Pentacnnus decorus as types of sub-
genera appears to have been abandoned by Sir Wyville within a year after he had proposed
the name Cenocrinus for the former species. For in his well known memoir On the
Embryogeny oi Antedon rosaceus, published in the Philosophical Transactions for 1865,
frequent reference is made to Pentacrinus (Neocrinus) asterias as well as to Pentacrinus
{Neocrinus) decorus ; while Oersted's species Pentacrinus miilleri was also referred to the
subgenus Neocrinus. Sir Wyville seems, therefore, still to have regarded Pentacrinus
hriareus as having the first claim to the generic name Pentacrinus, although the Messrs.
Austin had expressed an opposite opinion. He appears, however, to have eventually
adopted their view, as all later writers have done. For in The Depths of the Sea
reference is made to two AVest Indian species only, viz., Pentacrinus asterius and Penta-
crinus miilleri ; ^ and neither Neocrinus nor Cenocrinus is mentioned, while Pentacrinus
decorus is confused with Pentacrinus miilleri. Subsequently also, when describing new

1 The specific formula of this type is— A.R. 3.2.2.^.

2 The Depths of the Sea, pp. 436, 442, 1873.

REPORT ON THE CRINOIDEA. 281

"Porcupine" and C-hallenger species, Sir WyvUle named them Fentacrinus wyville-ihomsoni
and Pentacrinus maclearanus ; and the plate which was drawn under his supervision
was lettered Pentacrinus asteria (PI. XL).

We may therefore feel tolerably certain that Sir Wyville had recognised the inex-
pediency of limiting the name Pentacrinus to the Liassic species only, though their
generic differences from the recent Pentacrinidse had been noticed by him.

We have seen that the name Cenocrinus, which was applied by Sir Wyville in 1864
to the classical species Pentacrinus caput-Medusce of Miller and Mitller, was afterwards
dropped by him ; but I cannot make out whether or not this arose from his becoming
acquainted with the genus Cainocrinus which had been established twelve years pre-
viously by Forbes.^ The essential difference between this type and Pentacrinus, as
defined by Forbes, is that the pelvis or basal ring of Pentacrinus is " composed of a single
piece formed out of five anchylosed plates," while that of Cainocrinus is "formed out
of five free plates." These are seen in Forbes's figure to compose a closed basal ring which
separates the radials from the top stem-joint; and this is not the case with the basals
either of Extracrinus or of Pentacrinus asterius, the only recent species known to
Forbes.

What Sir Wyville thought of Forbes's genus I cannot say. He never referred to it,
and the fact of his having himself proposed Cenocrinus as a subgeneric type looks rather
as if he had not been previously acquainted with Cainocrinus. In any case, however,
whether he knew it or not, he still referred to the same genus Pentacrinus, the species
which was dredged by the "Porcupin-e" in 1870, and was named after himself by his
colleague Dr. Gwyn Jeffreys,^ F.R.S. ; and this is in all respects a true Cainocrinus with
a closed basal ring (PI. XIX. figs. 6, 7 ; PI. XX. figs. 1-3). Quenstedt^ was unable to
see any essential difference between Cainocrinus and Pentaerinus ; but de Loriol,* writing
about the same time, took a different view. Unaware of Forbes's genus, he proposed
to establish a new genus Picteticrinus for a fossil species of Pentacrinus presenting the
then unusual character of a closed basal ring. But he subsequently discovered this to be
a feature of the type described by Forbes as Cainocrinus, which he adopted as a generic
name instead of Picteticrinus ;^ and he referred to- this type a species that had been
originally supposed by Desor" to belong to von Meyer's genus Isocrinus, which has been
discussed above [ante, p. 271). Cainocrinus was regarded by de Loriol as establishing
a transition between Millericrinus and Pentacnnus. He defines Pentacnnus as difier-

1 Monograph of the Echinodermata of the British Tertiaries, pp. .33, 34.

2 Proc. Roy. Soc. Edin., vol. vii., 1872, p. 767 ; and also The Depths of the Sea, p. 444.

3 Encriniden, p. 269.

* Monographie Paleontologique et Gi5o!ogique des iltages Superieurs de la formation Jurassique des Environs de
Boulogne-sur-Mer, S""" partie, p. 297.
5 Swiss Crinoids, p. 111.
" Notice sur les Crinoldes suisses, Bull. Soc. d. Sci. Nat. de Neuchatel, vol. i. p. 213.

(ZOOL. CHALL. EXP. — PART XXXII. 1884.) li 3G

282 THE VOYAGE OF H.M.S. CHALLENGER.

ino- from Millericrinus in the verticillate arrangement of the cirri, and in having very
small basals, which do not meet externally. Thus he says " pieces basales fort petites, en
general arrondies, non contigues et reposant sur les cinq angles de la tige." On the other
hand, Cainocrinus has a complete ring of basals like Millericrinus, but a stem with
verticils of cirri like Pentacrinus. I cannot, however, regard this classification as
satisfactory ; for even in those species of Pentacrinus which have an incomplete basal
ring there is a great amount of variation in the extent to which the central ends of the
basals are joined, and in the size of their outer ends which appear between the radials and
the top stem-joint (PL XL; PL XII. fig. 16; PL XIII. fig. 1; PL XIV; PL XV.
figs. 1,2; Pis. XXVIIL, XXIX. ; PL XXX. figs. 1,4; PL XXXI. figs. 1, 2 ; PL XXXIV.
figs. 1,8; Pis. XXXV., XXXVI. ; PL XXXVII. figs. 1,2). I have elsewhere stated my
reasons for not adopting Cainocrinus ae, a genus distinct from Pe7itacrinus ;^ and the
results of my examination of the large series of Pentacrinidse dredged by tlie Challenger
and the " Blake " Expeditions has only served to strengthen my opinion. It must be
remembered that the question of the more or less perfect closure of the basal ring simply
has reference to its appearance on the exterior of the calj^x. The inner ends of the basals
always meet one another around the neurovascular axis. But they are sometimes not in
contact by the whole length of their sides, so that their outer ends appear to be separated
by the radials (compare PL XII. figs. 1, 2, 16 ; PL XIII. fig. 1 ; PL XIX. figs. 6, 7;
PL XX. figs. 1-3; PL XXVI. fig. 11 ; PL XXX. figs. 1, 4 ; PL XXXIV. figs. 1, 8 ;
PL XXXVI. ).

The closure of the basal ring, therefore, is so extremely variable within specific
limits that it scarcely afi"ords characters of specific, much less of generic value.

So far as the fossil species are concerned, however, it is quite possible that the more
or less perfect closure of the basal ring on the exterior of the calyx may afi'ord characters
of some systematic value. But I strongly suspect that the examination of a large series
of specimens would indicate a very great variability in the size of the basals, just as in
the recent types.

The unique specimen o? Pentacrinus maclearanus (PL XVI.) has a closed basal ring,
and would therefore be called a Cainocrinus by de Loriol. In Pentacrinus wyville-
thomsoni there is sometimes a very close union between the pentagonal basals as in
PL XIX. figs. 6, 7, and PL XX. figs. 1-3 ; while in other specimens the basals are more
triangular in shape and less closely united, as is shown, with a little exaggeration, in
PL XVIII. figs. 1,2. It occasionally happens that one or two of the basals fail to meet
their fellows, but the ring is always more or less complete.

Variations of a similar kind, though greater in degree, are presented by Pentacrinus
alternicirrus. Some individuals have prominent, rhomboidal basals not meeting laterally,
like those of Pentacrinus asterius (PL XIII. fig. 1 ) ; while in others they are much less

^ Jonrn. Linn. Soc. Land. (Zool.), vol. xv. p. 210; and Bull. Mm. Comp. ZooL, vol. x., No. 4, p. 168.

REPOET ON THE CRINOIDEA. 283

prominent and triangular or pentagonal in shape (PL XXY.). The same difference
appears in Pentacrinns iniilleri. The specimens in the Copenhagen Museum, one of
which is figured l)y Liitken/ have a closed ring of pentagonal basals. Some of those
dredged by the " Blake " are in the same condition ; but this is far from being the case
in the three individuals figured in PI. XIV. and in PL XV. figs. 1, 2. Sir Rawson
Eawson's specimen (PL XV. fig. 1) has the smallest basals that I have j&t seen in this
species ; while I have met with all intermediate stages between this condition and that
of Oersted's types at Copenhagen.

There is also a certain amount of variation in Pentacrinus hlahei, though I have not
seen a sufficient number of specimens to be able to say much about it. In Pentacnnus
naresianus, again, some individuals have pentagonal basals forming a closed ring ; while
in others the basals are triangular and barely meet their fellows. But as a rule their
outer ends are comparatively small and separated by the radials, which are sometimes
prolonged slightly downwards over the upper stem-joints (PL XXVIIL fig. 1 ; PL XXX.
fig. 1). A few specimens exhibit both conditions, some of the basals meeting their
fellows, while the rest are separated by the downward projecting radials.

But the most remarkable variations in the development of the pieces of the basal
ring occur in Pentacrinus decorus. They are sometimes smaller than those of Penta-
crinus asterius, and scarcely more conspicuous than the interradial ridges on the stem
beneath them (PL XXXIV. figs. 1, 8 ; PL XXXV.; PL XXXVI. fig. 3); or they may be
larger rhomboidal knobs standing out prominently from the general plane of the calyx,
and meeting one another by their extended lower angles (PL XXXVI. fig. 1 ;
PL XXXVII. figs. 1,2); or they may present any intermediate condition between these
two. To some extent these differences are perhaps due to age, both the individuals
figured on PL XXXV. being very young. But those represented in PL XXXVI. are
apparently of about the same age, so far as can be judged from the characters of the
stem, while their basals are at the two extreme stages of development ; and the original
of PL XXXIV. fig. 1 is very far from being a young individual. In the young
Pentacrinus tvyviUe-thomsoni , again, the basals are of about the same relative size as they
are in the adult (PL XVIII. figs. 1-3). I do not, therefore, see any reason for regarding
the variations in the development of the basals as of any more importance than the
differences in the number of arm-divisions. In Actinomctra parvicirra the number of
arms may vary from thirteen to thirty-nine, and much the same is the case in some
species of Pentacrinus and Metacriniis. But these differences are rarely of specific, and
much less of generic value ; and in the same way I find it impossible to consider the
presence of a closed basal ring as a valid generic character separating Cainocrinus from
Pentacrinus. There is no recent Pentacrinus in which the basals do not appear upon
the exterior of the calyx, so as to separate the radials either wholly or partially from the

1 Om Vestindiens Pentacriner, loc. cit., Tab. iv., v., figs. 1, 2.

284 THE VOYAGE OF H.M.S. CHALLENGER.

top stem-joint. A few fossil species, however, have been figured or described as not
possessing any external basals. Two of these are from British rocks, Pentacrimis fisheri
and Pentacrinus dixoni. The former was described by Baily,' who was unable to detect
the presence of interradial basals, and was led therefore to regard the first radial s as
basals. True basals are really present, however, and may be seen in the original
specimen in the Dorchester Museum, or in another found subsequently and now in the
possession of Mr. Damon of Weymouth.

The same is the case with the specimen of Pentacrinus dixoni in Mr. Willett's
collection at Brighton, which was figured in Dixon's Geology of Sussex (1878 edition,
pi. xix. fig. 22). In both these species, therefore, the supposed absence of basals is
really due to error ; and I think it likely that the same may be true both of Isocrimis
2)endulus, von Meyer, and of the Forest Marble specimen from Farley in Wiltshire, which
was described by Goldfuss^ as Pentacrinvs scalaris ; and also of Pentacrinus penta-
gonalis 'personatus from the Brown Jura. According to Quenstedt ^ three pieces rest on
the top of the stem " womit jedes der 5 Hauptradiale beginnt." But neither then nor
in the Eucriniden did he make any comment on the absence of basals, though he
must have noticed it. They may perhaps be small and only just in contact by their
central ends, so that they are concealed beneath the radials, as sometimes hapjaens in
Encrinus and in the fossil Comatulae.* But it ap23ears to me improbable that the
embryonic basals of any Pentacrinus should have undergone transformation into a
rosette, as those of many Comatulse do. A St.alked Crinoid with a rosette would be a
novelty indeed.

One would greatly like to know the real condition of the anomalous specimen of
Metacrinus costatus represented in PI. XLIX. fig. 2, wdiicli has no basals visible externally.
They are generally so very well developed in this genus that their abssnce altogether
seems unlikely ; and I suspect therefore that they are quite small and concealed between
the top stem-joint and the radial pentagon, as in the case of Encrinus and the fossil
Comatulse.

It is a curious fact that there are so very few species of Pentacrinidaj with only one
ray-division, i.e., with only ten arms; while at the same time the number of arms rarely
reaches the large total of one hundred or more, as it does in some of the giant species of
Actinometra from the Philippines. 1\\ Pentacrinus maclearanus (PL XVI.), Pentacrinus
ivyville-thomsoni (Pis. XVIIL, XIX.), Pentacrinus alternicirrus (PI. XXV.), Pentacrinus
hlakei (PI. XXXI.), and Pentacrinus decorus (Pis. XXXIV.-XXXVIL), the rays may
divide three times, i.e., there may be distichal and palmar axillaries above the radials.

' Description of a new Pentacrinite from the Kimnieridge, rf. O.xfortl Chy of "Weymoutli, Dorsetshire, Ann. arid
Mag. Nat. Hist., ser. 3, vol. vi. pp. 25-28, pi. i.

2 Petrefacta Germania", vol. i. pi. Ix. fig. 10.

3 Der Jura, 1858, pp. 363, 364.

* Jourii. Linn. Soc. Land. (Zool.), vol. xv. p. 195, pi. ix. fig. 6.

KEPOET ON THE CRINOTDEA. 285

This would give the maximum number of arms as forty ; but although the ten distichal
axillaries may all be present, the palmar axillaries are frequently only developed on the
two outermost of every set of four secondary arms, as in the unique specimen of Penta-
G7-inus maclearanus (PI. XVI.), and the individual of Pentacvinus alternicirrus shown in
PI. XXV. This would give six arms on each ray, making thirty in all. Sometimes,
however, there are no palmars at all, or only one or two series of them (PI. XVIII. fig. 2 ;
PI. XIX. figs. 1, 6, 7), so that the number of arms varies between twenty and thirty.

The ray-divisions of Pentacrinus decorus present almost as much variation as the
basals do. Some young specimens have only one or two distichal series (PI. XXXV.);
while one individual has but ten arms like Pentacrinmi naresianus. In others, again, the
arms are more numerous, though palmars are rare (PI. XXXVI. fig. l). I am inclined
to suspect, from an examination of several young specimens, that the many-armed
condition is to some extent a secondary one. Thus if none of the arms were broken
and subsequently repaired, the original of PL XXXV. fig. 1 would have grown up with
no more tlian eleven arms. When, however, an arm is broken oft" at a syzygy, and a new
one developed in its place, an axillary is nearly always formed in this new one sooner
or later, whether there were ojie present on the original arm or not. An instance of
this kind is shown in PL XXXVI. fig. 2; and it is not uncommon to meet with indi-
viduals of ten-armed species of Comatula which have replaced some of their arms after
fracture and have developed axillaries in the reparation, so that tlie number of arms may
reach eleven or twelve. This increase in the number of arms after reparation seems to take
place largely in Pentacrimis decorus ; for it is rare to meet with a specimen which does
not show these signs of reparation, certain axillaries and the arms which they bear being
distinctly smaller than their fellows.

In Pentacrinus asterius and Pentacrinus milUeri there are always twenty arms or
more, all the primary arms ending in distichal axillaries. Most of the secondary arms bear
palmar axillaries, and there are sometimes even one or two more beyond these, so that
the rays may divide five times in all. There is no special regularity of division in
Pentacrinus asterius, though the number of arms is large, exceeding one hundred,
according to Sir Wyville Thomson.' But in Pentacrinus miilleri there are usually not
more than four raj'-divisions, the (palmar) axillaries being limited to the outer arms as in
Pentacrinus maclearanus and Pentacrinus alter nicir rus ; while the fourth and fifth
axillaries, if present, occupy a similar position, so that there are six, eight, or ten arms to
the ray, as 2,1; 1,2—2,1,1; 1,1,2— or 2,1,1,1; 1,1,1,2.

The arms oi. Metacrinus branch as a rule more freely than those oi Pentacrinus, except
Pentacrinus asterius, all of the species having two, and most of them four axillaries
l)eyond the radials ; but there is no special regularity about the grou^jing of the arm-
divisions,

' Sea Lilies, Tlie Intellectual Observer, August 18G4, p. 5.

286 THE VOYAGE OF H.M.S. CHALLENGER.

Among the recent Pentacrinidse there is but one solitary example of a uniformly ten-
armed type, which embraces a majority of the species of Antedon. This is Pentacrinits
naresianus, represented in Pis. XXVIII.-XXX.

Owing to the fragmentary condition in which many of the fossil Peutacrinidse occur,
it is impossible to say much about the nature of their arm- divisions. But in Perita-
crinus heaugrandi, de Loriol,' s])., the remains of the primary arms bear no axillaries up
to the twelfth joint from the radials ; while eleven simple joints are still preserved in the
specimen from the Lias of Vaihingen, which is referred by Quenstedt to Pentacrinus
tubercidatus.^

Eeference has already been made to the low state of development of the arms of
recent Pentacrinidee as compared with those of Comatulas (ante, p. 55). They are fewer
in number (i.e., when multiradiate forms are compared), and have both pinnules and
ambulacral plating less developed towards their ends ; while the number of joints
separating successive axillaries is far more variable within specific limits, and does not
seem to have become tolerably fixed as is the case in the Comatulse. Singularly enough,
the two species Pentacrinus wi/ville-thomsoni and Pentacrinus aUernicirrus, in which the
distichal and palmar series are most uniform, are the very ones which most resemble the
Comatul^e in their mode of life {ante, p. 19).

It is curious that in the Pentacrinidas and Apioerinidse the external appearance of the
arm-joints should be so much more constant than it is among the Comatulae. In the
latter family the arm-joints may be saucer-shaped, more or less sharply wedge-shaped,
&c. , and it is in many cases easy to identify a species from detached portions of the arms,
especially as there is also very considerable variation in the characters of the pinnules.
But both in Pentacrinus and in Metacrinus there is a very great sameness, not only in
the form of the arm-joints as seen from their dorsal side, but also in the appearance of
the pinnules which they bear. The tubercular projections on the pinnule-joints of
Pentacrinus asterius (PI. XIII. figs. 1, 14), and the indications of carination on the
pinnules of a few species of Metacrinus, are almost the only variations in the character of
the pinnules through all the recent species of the family. It is true that the features of
the lower pinnules of Metacrinus are such as to afford a character of some generic value
for separating it from Pentacnnus. But with the exceptions above mentioned the
pinnules of all the difi"erent species of Metacrinus are very much alike. Both in tliis
genus and in Pentacrinus the arm-joints are almost invariably of the transversely oblong
type (Pis. Xr., XIV. ; PL XV. figs. 2, 3 ; PL XVL ; PL XVIII. figs. 1^3 ; PL XIX.
figs. 1, 6, 7; Pis. XXV., XXXVL, XXXVIIL, XLIL, XLIIL-XLVL, XLVIIL,
XLIX., LIL); and the same is the case in most, if not all, of the fossil species. It is
therefore at first sight by no means easy to identify the species to which isolated

^ Monographie des Stages Jurassit^ues Supijrieurs de Bovilogne-sur-Mer, 2"" partie, p. 298, pi. xxxvi. fig. 23, a.
a Encriniden, Tab. 97, fig. 39.

REPORT ON THE CRINOIDEA, 287

arm-fragments belong. But as I have pointed out above, the peculiarities of the arm-
grooves upon the ventral surface of the skeleton, and their relation to the ambulacra,
afford characters of considerable systematic value. The joints of Pentacrinus naresianus,
however (Pis. XXVIIL, XXIX.; PL XXX. figs. 1, 23), show a distinct indication of the
more oblique shape which is common among the Comatulge; while both in this species
and in Pentacrinus hlakei the peculiar nature of the syzygial union renders the arms
very readily distinguishable (PI. XXX. fig. 23; PI. XXXa. figs. 9-12; PI. XXXI.
figs. 1, 2 ; PL XXXII. figs. 5, 7, 12, 14).

The characters of the stem of the Pentacrinidse have already been fully discussed
[ante, pp. 12-23). The fact that during growth it undergoes rather considerable modifi-
cations in its appearance has led to a very general belief in the impossibility of identifying
species by means of stem-fragments only. This is more especially the case as regards
the fossil species, which are often based upon the stem-characters alone, since calyces
are but rarely met with ; and I think it not improbable that two or even more
types of stem from the same horizon, to which different specific names have been given,
may sometimes be only different stages of growth of one and the same species. Thus,
for example, four species might be made out of the joints represented in PL XXII.
figs. 13, 14, 23, and 26, wdiich are, however, merely different stages of growth in the
stem of Pentacrinus loyville-thomsoni. But on the other hand, the difference between
the stem-fragments figured on PL XIII. fig. 8 [Pentacrinus asterius) and PL XIX. fig. 4
[Pentacrinus wyville-thomsoni) are obviously not due to any developmental causes ;
and the two would be universally recognised as belonging to different species, even
if nothing whatever were known about the calyces and arms belonging to them.
The same remark holds good in the case of Metacrinus, isolated stem-fragments of
Metacrinus alternatus (PL XXXIX. fig. 3), Metacrinus cingxdatus (PL XLI. fig. 1), and
Metacrinus nohilis (PL XLI. fig. 5), to say notliing of other species, being very
readily recognisable.

What has been written above refers simply to the general appearance of the stem-
fragments ; but when the number of internodal joints is taken into consideration, and also
the markings on their terminal faces, the characters of the stem as a whole must be
regarded as of very considerable systematic value. The stem-joints of Balanocrinus and
Extracrinus are very readily identified by the sculpture on their faces ; but as far as the
internodal joints are concerned, I am unable to find any constant difference in this respect
between Pentacrinus and Metacrinus. In most (recent) species of both genera there are
from three to six strongly marked ridges along the sides of each petaloid figure. The
proximal ridges join their fellows in each interpetaloid space, while the outer ones reach
the exterior and cause the crenulation along the Line of union between every two joints,
the ridges of each face alternating in position with those of the face opposite to it
(PL XIIL figs. 10, 11 ; PL XXII. figs. 22-24 ; PL XXVI. fig. 17 ; PL XXX. figs. 25-30;

288 THE VOYAGE OF H.M.S. CHALLENGER.

PL XXXVI.; PI. XXXVII. %. 22; PL XXXIX. figs. 4, 7-11 ; PL XLI. figs. 1-3, 5-8,
15-17 ; PL XLV. figs. 2, 4 ; PL XLVII. figs. 1-9 ; PL XLIX. figs. 3-5).

The same characters may be seen in the numerous Pentacrinus joints figured in
Tabs. 97-99 of the Encriniden, and on pis. xiv.-xvii. of de Loriol's Swiss Crinoids.
In some of these fossil joints the number of ridges at the sides of the petaloid figures
may sometimes reach eight or ten ; but the inner ones always meet their fellows in the
interpetaloid spaces, while the outer ones appear externally. This is not the case in
Extracrinus, which has a more extensive but smaller crenulation at the sides of the linear
petals, as pointed out already {ante, p. 276).

In the recent species of Pentacrinus the cirrus-socket may extend downwards below
the articular surface so as to encroach very considerably on the internodal joint beneath
it, as in Pentacrinus naresianns and Pentaa'inus deconis (PL XXXa. fig. 6 ; PL XXXYL),
and in a less degree also in Pentaoinus miiUeri and Pentacrinus hlakei (PL XXXI. fig. 3);
or the joint above the node may be slightly incised to receive the upper part of the
socket, as in Pentacrinus alternicirrus (PL XXVII. fig. 1), Pentacrinus wyville-thomsoni
(PL XIX. figs. 3, 4), and occasionally also in Pentacrinus asterius. But in these cases
the alteration of the supra-nodal joint is not very considerable. Among the Pentacrinus
species it is most marked in Pentacrinus ivyviUe-thomsoni (PL XXII. fig. 17). With the
exception of those on the lowest nodal joint the cirri of this type are often found directed
upwards (PL XIX. fig. 1), while in others like Pentacrinus miiUeri, Pentacrinus
maclearanus, Pentacrinus naresianus, Pentacrinus hlakei, and Pentacrinus decorus they
are directed downwards (Pis. XIV., XVL, XXVIIL, XXIX., XXXL, XXXIV.), and their
bases are received in the hollowed sides of the infi'a-nodal joints.

On the other hand, the tendency of the cirri of Metacrinus is to take an upward
direction, the supra-nodal joints being slightly incised to receive their bases. This is well
shown in Metacrinus angulatus (PL XXXVIII. ; PI XXXIX. fig. 9), Metacrinus
cingidatus (PL XL.), Metacrinus wyvillii (PL XL VIII.), Metacrinus costatus (PL XLIX.
figs. 1-3), Metacrinus interruptus (PL LIL), and Metacrinus tuberosus (PL LIII.
figs. 1, 2). In some species, such as- Metacrinus varians (PL XLIV.), this character is not
very prominent ; but it can be traced with more or less distinctness in all the species of
the genus that I have seen, and is therefore (as far as it goes) of considerable value in
the separation of Metacrinus from Pentacrinus, as in the case of Metacrinus tuberosus
(PL LIII. figs. 1-6) and Metacrinus stewarti,^ of which only the stems are known. But
I have been unable to apply it to the determination of any fossil species, as this point is
naturally but rarely illustrated in sufficient detail in the figures of Quenstedt and
de Loriol.

' On Three New Species of Metacrinus, Trans. Linn. Soc. Land. (Zool.), ser. 2, vol. ii. p. 443.

EEPORT ON THE CRINOIDEA. 289

B. On the Characters of Young Pentacrinidce.

Young individuals of Pentacrinus are naturally rare, as is only to be expected. It
is related, however, that on one occasion a large number of them of all ages and sizes
were thrown up on the shore at Barbados after a gale ; but unfortunately for science
no one on the spot had knowledge enough to recognise the value of this extraordinary
event, and a great opportunity, such as may never occur again, was therefore lost.

Nevertheless the discovery that recent Pentacrinidse flourish in great numbers on certain
parts of the sea-bed, like their predecessors in the Liassic and Jurassic Seas, has brought
about a considerable increase in our knowledge of their premature stages of growth. The
dredgings of the " Porcupine," Challenger, and " Blake " have yielded several young
individuals of three Pentacrinus species and of Metacriniis nodosus, some of which are
figured on Pis. XVIIL, XXXa., XXXY., and LI.

Like the young Comatula, they are all distinguished by the relatively great height of
the first radials as compared with those of the adult, which are wider than high, often
considerably so (PL XIX. figs. 1, 6, 7 ; PI. XXX. fig. 1 ; PI. XXXVII. figs. 1,2;
PL L. figs. 1, 5), while the radials of the young individuals are spade-like, to use an
expressive term introduced by Sir Wyville Thomson. This is naturally most marked in
the youngest specimen Tvith a total length of 80 mm. (compare Pis. XXXV. and XXXVII.) ;
and the cup with its small basals presents a singular resemblance to that of Plicatocrinus
and Bathycrinus (PL VII. figs, 1-3, 6 ; PL Villa, fig. 1). The little we know of the
former, however, shows that it is a totally aberrant type, and the resemblance must
therefore be considered as in great measure accidental and not as indicating any genetic
relationship. But Bathycrinus is a decidedly embryonic form, as is shown by the length
of all its three radials and th-e absence of pinnules from the arm-bases. Hyocrinus
(PL VI.) is another type with high spade-like radials; but the basals are of the same
character, and not small and inconspicuous as in young Pentacrinidae and in Bathycrinus,
while the arms are totally dilferent.

Another character of the incompletely developed Crinoid, which is very marked in
Rhizocnnus and Bathycrinus, and still more so in the aberrant Plicatocrinus and
Hyocrinus, is the comparative freedom of the second radials. In many Comatulse they
are closely united laterally ; while in most of the Palseocrinoids, as in Apiocrimis
and Guettardicrinus, they are practically immovable, and enter into the composition of
the body. The second radials of Pentacrinus, however, rarely show any traces of the
lateral pits lodging interradial ligaments such as occur in many multiradiate Comatulaj.
But they are often in very close apposition, while in young individuals of the same
species they are comparatively free (Pis. XVIIL, XIX., XXIX., XXXa., XXXV.,
XXXVIL). The arm-joiuts of most young Crinoids, as well as those in the outermost
and growing parts of the arms of more mature individuals, are always distinguished by

(ZOOL. CHALL. EXP. — PART XXXII. — 1884.) li 37

290 THE VOYAGE OF H.M.S. CHALLENGER.

the relatively great proportion of length to width, whatever may be the shape of the full-
grown joint. In some Comatulse this condition is permanently retained, but in others
and in the Pentacrinidse the arm-joints of mature individuals are usually much wider than
long. In accordance with this, we find that even in young individuals of Pentacrimis
decorus with quite high radials and arms only 12 mm. long, .consisting of about a
dozen properly formed joints, the great relative width of the latter is already very
distinctly indicated (PI. XXXV.) ; while the Jast few joints are much smaller than then-
predecessors, and have rather the appearance of a pinnule than of the continuation of the
arm, the preceding joint of which looks like an axillary bearing two small arm-stumps.

This mode of development is less marked in a somewhat older individual of
Pentacrinus decorus with al^out forty arm-joints, the later ones of which gradually
decrease in size instead of . becoming abruptly smaller ; and the imperfect state of
development of the later pinnules which is so characteristic of the Pentacriniclje is
very well shown. This would seem to indicate that the mode of growth of the arm-bases
proceeds on a different plan from that of their middle and outer portions. In still larger
individuals, as in the youngest specimens of Pentacrimis iri/ville-thomsoni' (PI. XVIII.
fig. 3), Pentacrinus naresiauus^ (PL XXXa. fig. 1), and Metacrinus nodosus (PI. LI.
fig. 1), the terminal arm-joints are distinctly longer than wide, although the lower ones
have almost assumed their adult form ; and in all of them, but especially in Metacrinus
nodosus, there is the characteristic reduction in the size of the later pinnules. Nothing is
to be learnt regarding the order of the pinnule-succession in the Pentacrinidae from any of
these young individuals ; for the smallest of them are larger than some young Comatulse
already detached from the stem, but without pinnules on the ai'm-bases, and they all have
their proper complement of pinnules. The stem-joints of the immature Pentacrinidse,
like the later joints of growing arms, are relatively high in proportion to their width.
This is exactly the reverse condition to that of the young joints formed immediately
beneath the calyx (PL XXXIV. fig. 9). The same distinction appears in the very different
type of stem characteristic of the Conuttula-larva, and of the Bourgueticrinidse (PL VII.
fig. 11 ; PL Villa, fig. 1 ; PL LIII. figs. 7, 8). The young joints are at first discoidal,
then lengthen out, and finally the width increases relatively to the length so as
sometimes even to exceed it considerably. The two types of stem are so very different
that it is perhaps a little rash to reason about the one on the basis of the other. The
intercalation of new joints, which is so characteristic of the Pentacrinidse, seems never to
occur in the Bourgueticrinidse, new joints being only formed beneath the calyx. In this
last respect, however, the mode of growth in the youngest Pentacrinidse with very
slender stems appears to be very much what it is in the equally slender Rhizocrinus and
Bathycrinu-s. But as the diameter of the stem increases to 3 or 4 millimetres the

• The arm-joints of this species are more like those of the Comatulae than is the case in any other Pentacrinus.
Instead of being nearly oblong, they have somewhat oblique ends, especially in the lower parts of the arms
(Pis. XXVIII.-XXX.).

REPORT ON THE CRINOIDEA. 291

length of the joints does not increase in the same proportion ; and I am therefore not
surprised at having met with but one individual in which the stem-joints are longer than
wide, as shown in PL XXXV. fig. 1,

A comparison of the figures on this plate, which represent the two youngest
individuals dredged by the "Blake," with those of the older, but still immature forms
figured in PL XXXVL, will- show, however, that the great relative height of the stem-
joints is a very characteristic feature in the development of Pentacrinus decorus. The
same feature is apparent in Pentaorinus wyville-thomsoni (Pis. XVIIL, XIX.),
Pentacrinus naresianus (PL XXVIIL ; PL XXXa. fig. l), and Metacrinus nodosus
(PL L. fig. 3 ; PI LL fig.. 1).

The sculpture on the terminal faces of these internodal joints of young individuals
(PL XXXa. figs. 2, 3 ; PL XXXVIL fig. 9 ; PL LI. figs. 2-5), as in the case of those
intercalated between the pre-existing joints in other stems (PL XXII. figs. 3-12;
PL XXXVIL figs. 10, 13, 14), is more or less obscure; but the petaloid markings are
evident from the first, as might be expected.

The external appearance of the nodal joints not unfrequently alters considerably
during growth. Thus, for example, in Pentacrinus decorus the characteristic expansion
down to the upper edge of the cirrus-socket (PL XXXVL); scarcely appears at all in the
young nodal joint (PL XXXVIL fig. 5) ; while in Pentacrinus naresianus there is a trace
of this expansion in the young stem, though it entirely disappears in the adult. In
Pentaci-iims wyville-thomsoni, again, the overlap of the lower edge of the cirrus-socket
above the infra-nodal joint is far less evident in the young stem (Pi. XVIIL fig. 3) than
it is in the adult (PL XIX. figs. 3, 4). In Metacrinus nodosus, however, the characteristic
enlargement of the nodal joints between their cirrus-sockets is very evident in a young-
stem with relatively high joints (PL LI. figs. 6, 7).

O. Tlie Calyx and its Contents.

The calyx, i.e., the combined ring of basals and radials, is constituted in the same way
in both the living genera of Pentacrinidaj, viz., Pentacrinus and Metacrinus, as is evi-
dent from a glance at PL XII. The former may therefore be taken as the t}^e ; for it
is the better known genus, and has a long geological history, while Metacrinus is not
yet known to occur in the fossil state. The calyx of Pentacrinus, while more like that
of ordinary Comatulse, i.e., Antedon, Actinometra, and Eudiocnnus, than the corre-
sponding part is in Rhizocrinus and Bathycrinus, nevertheless difi'ers from it in a very
important character. This is the presence of unmetamorphosed basals, such as are only
found in Atelecrinus and Tliaumatocrinus (PL LVI. figs. 1-4) among Comatulse ; for
the embryonic basals of the ordinary types undergo the well known transformation into
a rosette. This structure covers in the chambered organ which is lodged within the

292 THE VOYAGE OF H.M.S. CHALLENGER.

cavity of the centro-dorsal (PI. LXI. fig. 2), and the same is the case with the flattenetl
hasals of Atelecrinus.

The chambered organ of Pentacrinus is both relatively and absolutely smaller than
that of Comatula, and the reason for this is obvious. The Comatulid centro-dorsal
represents, as it were, a coalesced series of the nodal stem-joints of Pentacrinus, and the
five cavities of the chambered organ lodged in its upper part ai-e each in connection
with one cirrus-vessel only. But the remaining cirri which are borne by the centro-
dorsal in greater or less abundance are supplied by a number of vessels that come off"
below the chambers. Within the dorsal portion of the chambered organ, " lying at the
bottom of the centro-dorsal basin, there is a succession of verticils of five triangular
leaflets, increasing in size from below upwards, from the extremities of some of the upper
of which leaflets issue groups of three diverging cords that proceed to the cirri. I can
scarcely doubt that these verticils mark the origins of the earlier cirral cords from the
Crinoidal axis ; and this obviously suggests that the five-chambered organ is itself only
another and larger verticil, which has come by the formation of ventricular cavities in
its substance (analogous to the lateral ventricles of the brain), to occupy the whole
cavity of the enlarged centro-dorsal basin." ^ In Pentacrinus, however, the cirri all
come off" from the nodal joints of the stem, where the five downward prolongations of
the cavities of the chambered organ in the calyx enlarge and each gives off" a cirrus-
vessel (PI. XXIV. figs. 3, 4, chn). No cirrus-vessels come off" from the enlargement
of the vascular axis within the calyx, which represents the chambered organ of Comatula
without the verticils of cirrus-vessels below it. The chambers, however (PI. XXIV.
figs. 6-8, ch), are scarcely larger than the nodal cavities in the stem from which the
cirrus-vessels arise,^ and are much smaller than the corresponding chambers within the
upper part of the centro-dorsal of Comatula (PL LXI. fig. 2), which give off" the vessels
to the younger cirri. In fact they are sometimes almost equalled in size by the central
vessels within the ring of chambers, as seen in PL XXIV. fig. 6. They are not closed
below as is practically the case in the Comatulse, nor do they present as sudden enlarge-
ments of the stem- vessels as in Rhizocrinus and Bathycriniis ; but these vessels are, as it
were, permanently enlarged in tjie upjDer part of the stem, owing to the closeness of
the nodal joints which are successively formed beneath the calyx, and are only gradually
separated by the intercalation of internodal joints between thenx The chambers of
Pentacrinus therefore taper very gradually downwards into the stem-vessels, and it is
difficult to say where the latter begin and the former end (PL LVIII. fig. 3).

The primary Y-shaped interradial cords which come off" from the fibrillar envelope
surrounding the chambered organ of Pentacrinus (PL XXIV. fig. 7 ; PL LVIII.
figs. 1, 3 — a.i.) sometimes bifurcate before entering the basals. On the inner face of the

' See AV. B. Carpenter, Proc. Roy. Soc, vol. xxiv., lS7fi, p. 219.
- Compare PI. XXIV. figs. 3, 4, 6, which are all equally magnified.

EEPORT ON THE CRINOIDEA. 203

basals of Penfacrinus decor us tliere is but one central opening, so that the primary cords
fork within the basals (PL LVIII. fig. l) as they did in Ajnocrinus. But in PentacrinuH
hlakei this opening is widely extended laterally, while in Fentacrinus naresianus, Penta-
cnnus alter nicirrus,^ and especially in Pentacrinus wyville-thomsoni (PI. XXI. fig. 7a),
it is more or less completely divided into two by a vertical partition. In Pentacrinus
nsterius and Metacrinus angulatus, lioth of which, and especially the latter, have large
basals with a great development of calcareous network on the upper ends of their inner
faces, the opening of the central canal is pretty distinctly double.

On the upper surface of the basals the openings are, of course, perfectly separate.
They are situated on either side of the median ridge, and correspond to similar openings
on the under surface of the two contiguous radials which partly rest upon each single
basal (PL XIL figs. 11, 14, 22, 25; PL XVIII. figs. 5, 7; PL XX. figs. 2, 3, 6, 9:
PL XXI. figs. 6, a, h, c, and 7, h; PL XXVI. fig. 9; PL XXX. figs. 5, 7, 8; PL XXXIII.
fig. 5 ; Pi. L. fig. 5). In most specimens of Pentacrinus unjville-thomsoni, as in other
Pentacrinidae, removal of the basals exposes the apertures on the lower surface of the
radials where the secondary axial cords enter them, together with the lowest portion of
tlie central plug (PL XX. fig. 9). But in one individual of Pentacrinus wyville-thomsoni
Dr. Carpenter found this plug to be somewhat unusually developed. Its low^er surface
forms a tolerably well defined pentagonal plate, the angles of which are interradial in
position (PL XX. figs. 4, 5, 6, 8). It lies between the basals and radials, and is pierced
in the centre for the passage of the plexiform gland rising out of the chambered organ,
together with openings through which the secondary axial cords passed on their way
from the basals to the radials, the apertures in the latter plates being entirely concealed
by it.

The radials of Pentacrinus thus difier very considerably from those of Comatula, for
the proximal openings of their central canals are really on their dorsal faces, viz., at the
central ends of wliat I have called the inner dorsal surface, i.e., that part which rests
on the basals (PL XII. figs. 11, 22 ; PL XX. fig. 9 ; PL XXI. figs. Ga, 66, 6c; PL L.
fig. 5) ; whereas in the Comatulse, as I have shown elsewhere,^ these openings are on the
inner faces of the radials, i.e., those which form the walls of the radial funnel that con-
tains the plexiform gland rising from the chambered organ. They are also less closely
approximated in Pentacrinus than in Comatula, one being situated on either side of the
strong crest which divides the inner dorsal surface into two lateral halves, instead of
licing only separated by a very delicate calcareous bar. The converging axial cords
which enter these openings on the radials of Pentacrinus (PL XXIV. figs. 7-9 ;
PI. LVIII. figs. 1-3 — ar) run upwards and outwards for some little way before they
are united by the interradial and intraradial cords to form the circular commissure

^ I have only seen single specimens of the dissected calyx in these three species.
2 Trans. Linn. Soc. Land. (Zool.), ser. 2, vol. ii. p. 78, pis. iv.-vii., 1877.

294 THE VOYAGE OF H.M.S. CHALLENGER.

(PI. XXIV. fig. 9, ico, ceo). The canal in which this is lodged is never close down to
the proximal openings in the inner or dorsal face, as it is in all Comatulse, even Atele-
crinns. But its position varies somewhat in different sjjecies. Thus, for example, it
comes nearer to the edge of the central funnel in Pentacrinus decorus (PI. XXXIII.
fig. 5) than in Pentacrinus wyviUe-thomsoni (PI. XX. fig. 6). The doulile axial cords
of the rays which proceed outwards from it resemble those of Comatulse in their very
close approximation. Small portions of them, cut very obliquely, are seen in PI. XXIV.
fig. 9, A. They are lodged close together in the same canal as far as the axillary radial,
not being so widely separated as in Encrinus ; and the arrangement of the commissures
in the axillary is just the same ais was discovered by Ludwig in the Comatulse.

D. Tlie Geological History of Pentacrinus.

Excepting for some doubtful forms from the Eifel, the earliest known Pentacrinidaj
occur in the " Wellenkalk " of the Jura, at an horizon somewhat lower than the well-
known limestone in which Encrinus liliiformis occurs. According to Quenstedt, both
generic tj'pes occur together in the Wellenkalk of Wiirtemberg ; and he refers all the
Pentacrinida3 to one species, Pentacrinus dubius, though they have received various
other names, both generic and specific, from earlier writers. Nothing being known of
them but fragments and isolated joints of the stem, any detailed classification of them
is hardly possible. But the similarity of the stem-fragment from Waltershausen^ with
ten ciiTus- whorls at intervals of eight or ten joints, to the stems of recent Pentacrinidae,
is very striking. This resemblance was noticed by von Schlotheim,^ who described the
fossil as Pentacrinus vulgaris, and referred to the same type the recent specimens of
Gruettard and Ellis. Some j^ears later Quenstedt ' gave an excellent figure of it ; but in
the absence of an associated calyx lie hesitatedi to refer it to Pentacrinus as von
Schlotheim and Goldfuss had done, and so described it as Encrinites dubius. Beyrich
and later writers, however, have generally regai-ded it as a Pentacrinus, as Quenstedt
himself has done in the Encriniden, and the reference of the fossil to the Pentacriuidai
will now be scarcely disputed. Another very similar stem from the Muschelkalk of
Silesia was described by von Meyer* as Chelocrinus acutangidus. This genus was
established to receive certain forms with more than ten arms, owing to the presence
of distichal and palmar series, which had been generally referred to Encrinus. It
has been abundantly proved, however, by the researches of von Strombeck" and others,

' Encriniden, p. 198, Tab. 97, iigs. 14-22.
- Die Petrefectenliunde, p. .327.

2 Uebev die Encriniten des Muschelkalks, Archiv f. Katurgesch., Jalirg. i. Bd. ii., 1835, p. 225, Taf. iv. fig. 2.
■* Fische, Crustaceen, Echinodernien und andere Versteinerungen aus dem Musclielkall^ Oberschlesiens, Palajonto-
grapliica, Bd. i., 1851, p. 272.

'^ Ueber Missbildungen von Encrinus liliiformis. Lam., Pala;ontograpliica, Bd. iv., 185G, p. 17G.

EEPORT ON THE CRINOIDEA. 295

that the character on which von Meyer established CJielocrinus is one of no systematic
value whatever. One of the two species on which he founded the genus was the
Encriniis pentactinus of Bronn.^ This name was proposed for a single specimen which
presented a very remarkable combination of characters. The cup appears to be that of a
tjqDical Encrinus with some twenty biserial arms. The stem has a rounded pentagonal
outline, and consists of about twenty -five joints, some of which bear verticils of five
cirri. Beyrich appears to consider these characters as of altogether subordinate value,
for he says^ — "die otwas abweichende Form des fiinfseitigen Stengels mit deutlich
eingesenkten Seiten kann eben so wenig fiir ein uuterscheidendes Ai'tnierkmal gelten, wie
die von Bronn beobachteten kurzen Cirren, welche anscheinend individuell hier und
da bei verschiedenen Encrinus-avten vorkommen konnen." There is a considerable
difference, however, between a regular verticillate arrangement of cirri and their occasional
presence here and there upon the stem, about which Beyrich does not speak very
confidently, so far as Enarinus is concerned. I cannot make out, indeed, that any stem
of an Encrinus has ever been described as bearing isolated cirri in the manner referred
to by Beyrich ; while, on the other hand, the verticillate arrangement has been discovered
by Picard * in an imperfectly preserved fragment from the Muschelkalk which he has
described as Encrinus beyriclii—" Dev Stengel ist von seinem untersten Theile an mit
Ranken versehen. Dieselben treten von unten herauf CTst sparlicher und in grosseren
Abstanden auf; am oberen Tlieile ist je das achte Sauleuglied ein Cirrentrager und
erscheint horizontal etwas verdickt. Die Anhaftestelle fiir die Cirre befindet sich
regelmassig auf der concaven Fliiche zwischen je zwei der fiinf Kanten, so dass ein
Wirtel von 5 Cirren den Stiel umgiebt." He continues — "Als Pentac.rinus wage ich
meinen Fund nicht auszusprechen. Denn obwohl die Fiille an Cirren und der ganze
Habitus der Saule sehr zu dieser Auffassung drangt, hat mich doch die Zusammensetzung
der Krone, soweit sie klar ist, bestimmt, bei der Gattung Encrinus zu bleiben." There
would seem therefore to be no doubt as to the occurrence of cirrus-verticils on the stems
of other Neocrinoids besides the Pentacrinidse ; while there are certain Palaeocrinoids
both in this country and in America which present the same character. Although,
however, this fact diminishes the value of that peculiarity of the Pentacrinidse which is
most evident at first sight, it does not follow that the verticillate arrangement of the
cirri in these other types is also associated with the presence of syzjj'gies at the nodes and
the mode of union of the internodal joints, which are such especial features of all the
Pentacriuidge. The Pentacrinus duhius of Goldfuss and Quenstedt shows these points
very well, the five petals being visible on the faces of many joints found in the Muschel-
kalk at different localities ; while the stem of von Schlotheim's specimen (6 inches long)

' Ueber die Krinoideen-Reste im Muschelkalk, Neties Jahrh.f. Mineralogie, Jalirg. 1837, p. 32, Taf. ii.

- Crinoideen des Muschelkalks, loc. cit., p. 36.'

3 Zeitschr. d. deutsch. geol. Gesellsch., Jalirg. 1883, p. 201.

296 THE VOYAGE OF H.M.S. CHALLENGER.

terminates below iu a nodal joint owing to its separation at the syzygy, as is the case
with so many fossil as well as recent Pentacrinidse. Although therefore it seems
tolerably certain that cirrus-verticils may occur in some species of Encrinus, the other
characters of many of the stem-fragments from the Muschelkalk render it equally certain
that Pentacrinus or an allied genus existed in the same seas as Encrinus. This has long-
been admitted in the case of the St. Cassian beds, which are considerably above the
horizon of the Muschelkalk ; and they have yielded to Laube ^ five varieties of Penta-
crmws-stems, which he has temporarily referred to as many species. Associated with
these he found one calyx which appears to belong to Pentacrinus; but it was too
imperfectly preserved for a definite opinion to be foi-med upon this point.

All these eai'lier forms seem to belong to the section of the genus which was desig-
nated " Basaltiformes " by Quenstedt, after one of Miller's species with a pentagonal stem.
Essentially similar stems are found all through the remaining secondary rocks from the
Lias upwards, in the Nummulitic Limestones of Biarritz and Kreissenberg, in the London
Clay, and in the Italian Tertiaries. The calyces associated with them have a monocyclic
base, and bear simple or dichotomous arms, the two limbs of each fork being of equal
value, just as in the recent forms. Quenstedt's other two groups, the Briariden and the
Subangularen, are those to which the generic name Extracrimos was applied by the
Messrs. Austin. The type is limited in Europe to the Lias and Lower Oolites. Extra-
crinus briareus appears to occur in the lowest " quick ledge " at Lyme Regis, which
corresponds to Quenstedt's lowest bed a; but the t3rpe does not reach into the Middle
Lias, nor indeed into the upper beds of the Lower Lias. In the Jura, however, Extra-
crinus briareus has not been found below the Marlstone (Lias 7 and S); but it ranges up
through the Posidonia-heds, and is said to occur in the Lower Oolites.

The range of the Subangularen is more limited. They do not occur in the Lower
Lias of either England or Germany, nor do they reach the top of the Posidonia-heds.
Many varieties have been made of lioth species, especially by Quenstedt ; Ijut it is note-
worthy that neither species occurs in all the three divisions of the Lias, Extracrinus
briareus being limited to one only, at any rate in this country.

Those who have recognised Extracrinus as a genus, as, for example, Mr. E.
Etheridge,^ F.R.S., have usually regarded it as limited to the Lias. The Messrs. Austin,
however, speak of Extracrinus briareus as occurring in the Cornbrash at Kingscote Turn-
pike. They are certainly wrong about the species, l)ut I am very strongly inclined to
believe that they are correct as regards the genus, and also that the occurrence of the
Briariden in the Brown Jura (Inferior Oolite), as mentioned by Quenstedt,^ is not
unlikely.

> Die Fauna cler Schichten von St. Cassian, Denlcschr. d. k. Akad. d. TFiss. Wicn., Bd. xx., xxiv., pp. 267-278.
- See his Presidential Address to the Geological Society, Quart. Journ. Geol. Soc, 1882, p. 147 (of separate cop}-).
^ Encriniden, p. 270.

REPORT ON THE CRINOIDEA. 297

Tn the year 187G a Inrge Pentacrinus colony was discovered at Sennecey-le-Grand.
Numerous very perfect individuals were obtained and carefully described by de Loriol.'
He found a verticil of cirri on every joint, and described their faces as follows : —
" La surface articulaire est plane ; les petales de la rosette articulaire sont fort
etroits, et ont I'apparence de cinq petites rigoles aboutissant aux cinq angles du
pentagone et limitdes par deux petits bourrelets paralleles, plus eiev^s sur I'une des
faces de 1' article que sur I'autre, et converts de tres fines crenelures." The figure which
he gives (pi. ii. fig. 10) shows the complete correspondence of these joints with those
of the Liassic Extracrinus ; and he found this correspondence also in the other characters
of the type. The radials extend slightly downwards over the top stem-joints, and the
characters of the arm-divisions are almost exactly as in Extracrinus, except that the
main arms and the armlets which they bear are more equal in size than in the typical
species.

De Loriol identified this species with Pentacrimis dargniesi of Terquem and Jourdy,
though he considered it as belonging to the same group as Extracnnus hriareus ; but
he hesitated to adopt Austin's genus, and he subsequently stated that there was no reason
to do so.*

He also pointed out that the characters of the stem-joints and cirri of Quenstedt's
two species Pentacrimts hriareus zollerianus and Pentacrinus hriareus achalmianus, both
from the Brown Jura (Inferior Oolite), indicate their affinity to this grouj). In the same
memoir he described and figured some other stem-joints presenting all the Extracrinus-
characters from the same formation (Bajocien) of Langres, and he subsequently found
both these types at corresponding horizons in Switzerland. He referred at the same time
to the Pentacrinus nodotianus, d'Orbigny, whicli was descril^ed l^y its founder as being
" voisine du Pentacrinus hriareus." Very similar stem-joints, each bearing five
cirri, and having crenulated linear petals, occur in the Coralline Oolite of various
parts of Switzerland, and are described as Pentacrinus buchsgauensis by de Loriol,
who notes their resemblance to those of Pentacrinus hriareus as a point of special
interest.'

We may therefore, I think, consider it certain that Extracrinus extends up above
the Lias into the Lower Jurassic rocks of the Continent, and the same is undoubtedly the
case in England. The Great Oolite of Minchinhampton contains stem-joints ■s\'ith the
same linear, crenulated petals as those of the Liassic Extracrinus hriareus ; while similar
joints, together with arm-fragments showing the characteristic inequality of division,
abound in the Forest Marble at Malmesbury.

To the genus Extracrinus I would also refer the Pentacrinus asteriscus from tlie

' Notice sur le Pentacrinus de Sennecey-le-Granil, p. 7. Both in this work anrl in the Swiss Criuoicls this type
is called Pentacrinua dargniesi by de Loriol. But the plates are lettered Pentacrinus chahasi, P. de Loriol.
2 Swiss Crinoids, p. 116. 3 xiUfi^ pp 153^ 154,

(ZOOL. CHALL. EXP. — PAIIT XXXII. — 1884.) Ti 38

298 THE VOYAGE OF H.M.S. CHALLENGER.

American Jura-Trias (Jura only 1), which was first discovered by Meek and Hayden.^ It
was subsequently found in Utah by Wheeler's Survey, and owing to the constancy of its
characters, even in examples collected at widely separate localities, it has been relied on
with considerable confidence in the identification of Jurassic strata. According to Meek
and Hayden there are five large, oval, petaloid areas, bounded on either side by " rather
narrow, slightly elevated, transversely crenulated margins,"^ and the figure which was
published later by White^ clearly shows that the affinities of this species are with Extra-
crinus rather than with Pentacrinus, of which it has been hitherto regarded as the only
species yet recognised in the American Jurassic rocks. Hayden's Survey found it in
1877 extending through a considerable thickness of Jurassic limestones in Idaho and
Wyoming.* The supposed Triassic beds of Dun Glen and the Pah Ute range contain a
slightly larger form of joint, which differs from the Jurassic specimens. It was found
associated with what are regarded as unmistakable Triassic fossils and also a Spirifera.
Mr. Emmons^ speaks of it as follows: — " It should be stated also that these disks of Penta-
crinus found in the Dun Glen limestone vary somewhat from the type specimens, and
are all of larger size, reaching one-fourth of an inch in diameter, while those of Jurassic
age scarcely reach one-fifth of an inch. Prof Whitfield suggests that the Dun Glen
variety may possibly be a new species."

Elsewhere, too, it is stated by HaU and Whitfield® that this Triassic form from Dun
Glen differs from the Jurassic specimens " in the more obtuse points of the star, and
the filling up of the angles between the points, and also in the broader form of the
elliptical figures on the articulating surfaces of the disks." These are just the difi"erences
between the stem-joints of Extracrinus and those of Pentacrinus, as explained above ;
and I am therefore disposed to think that this Triassic type with broader petals may be
a true Pentacrinus, more especially as we have no knowledge of any European Extra-
crinus below the Lias.

It must be remembered that nearly all of these identifications are based upon the
characters of the stem-fragments only, the calyx and arm- bases being but rarely preserved.
Ever}'^ one who has examined moderately long pieces of stem, wdiether recent or fossil,
has noted the variation of characters which they present in different parts ; and there
has therefore been a general disposition not to jilace too much reliance on sj^ecies of
which only the stems are known. I have been surj^rised to find, however, in the case of

^ Palaaontology of the Upper Missouri, Invertebrates, pari i. p. 67, pi. iii. fig. 2.

2 Proc. Acai. Nat. Sci. Philad., March 1858, p. 49.

2 Report upon Geographical and Geological Explorations and Surveys west of the One Hundredth Meridian, in
charge of First Lieut. Geo. M. Wheeler, vol. iv. part i., Pahvontology, p. 162, pi. xiii. fig. 6,a.

' Eleventh Annual Report of the U. S. Geological and Geographical Survey of tlie Territories, embracing Idaho
and Wyoming, 1879, pp. 626, 627.

^ Report of the Fortieth Parallel Survey, Descriptive Geology, vol. ii. p. 711.

6 Hid., vol. iv. p. 280, pi. vi. fig. 16.

EEPORT ON THE CRINOIDEA. 299

recent types that when allowance is made for the different stac^es of trpowth the
characters of the stem are of much use in the distinction of species. This is especially
the case if the number of internodal joints can be ascertained, which is not often possible,
however, with fossil stems. But I am more disposed than I formerly was to accept
the numerous species described by de Loriol and Quenstedt on the characters of the stem
alone. Under these circumstances it is possible that more of the Continental species
may eventually be found to occur in Britain, the number at present known beino- very
small.

The geological range of individual species, both of Pentacrinus and of Extracrinus,
seems to be rather limited, no species occurring in all three divisions of the Lias. Out of
fifteen species of Pentacrinus which are found in the Middle and Upper Lias of Britain,
only two are common to the two horizons. Neither of the two Lower Lias species reach
the Middle Lias, though Pentacrinus hasaltiformis does so on the Continent. This and
Pentacrinus psilonoti are the only Continental species which have yet been recognised in
the British Lias. One species of the Kimmeridgve Clay, Pentacrinus sigmaringensis, also
occurs in the White Jura of the Continent.

This agrees very well with the geographical distribution of the Pentacrinidfe in the
recent seas, the four West Indian species, though abundant in the Caribbean Sea, not
occurring elsewdiere ; while the remaining tw^o Atlantic species have respectively been found
at one and at three stations only. Pentacrinus naresianus was obtained in the West
Pacific in lat. 29° 55' S., and also in lat. 4° 33' N. (or possibly 9° 26' K), and has the
widest geographical range of any Pentacrinus, recent or fossil, that I am acquainted with.
Like the Silurian Crinoids, therefore, the species of Pentacrinus, both recent and fossil,
seem to be somewhat limited in their geographical range.

The genus has, however, a wider geographical range than Metacrinus, wliich is
confined to the Pacific Ocean and the East Indian Archipelago. On the other hand,
while fourteen species of Metacrinus are known, Pentacrinus is only represented by eiglit,
together with the doubtful form which I have called Pentacrinus (?) mollis (PL XXXIIL
figs. 7-10). The mutual relationships of these eight species are shown in the
following table ; and it will be seen that they fall into two very natural groups,
which have the four Caribbean and the two Pacific species equally divided between
them.

300

THE VOYAGE OF H.M.S. CHALLENGER.

The two
outer rad-
ials and
the first
two joints
beyond
each axil-
lary united
by syzygies

Five

cirri at

each

node.

The two
outer rad-
ials and
the first
two joints
beyond
them uni-
ted by bi-
fascial arti-
culations.

L

Cirri large and
stout, of
thirty-five to
fifty joints.

Rays may

divide five
times..

Fifteen to

twenty - five
cirrus -joints.
Eays only
divide tliree
times. Each
division of
two joints
united by

syzygy.

Genus Pentacrinus, Miller.

Usually fifteen to eighteen inter-
nodal joints. The hypozygal
does not share in the cirrus-
sockets. Generally more than
three joints between the irre-
gular arm-divisions. Pinnule
joints have forward projecting
processes.

• L asterius, Linn.

Six to eight internodal joints. The
hypozygal shares in the cirrus-
sockets. Only two or three j oints
between the regular arm-divi-

2. miilleri, Oersted.

i

One or two internodal joints,

3. ?ftac?eara>M<s, Wyv. Thorns.

Thirty-five to forty-five internodal 1 , .„ ,, „ • t„«'„„^„

. .■' •' > 4. wyviUe-thornsoni, Jetlreys.

Two and three cirri alternately at successive nodes. Rays '\

divide three times ; each division of two joints united by |- 5.^ altemiciri-us, n. sp.
.syzygy. Four internodal joints. j

Ten arms only. The third brachial a syzygy with angular \
syzygial faces. First pinnule on the second brachial. ;■ 6.
Eight to seventeen internodal joints. )

iiaresianm, n. sp.

Rays may divide three
times. The second
free brachial a syzygy.
Pinnule on the first
brachial.

Five to seven internodal joints. ]
Nodal joints not enlarged, and j
the hypozygal sharing but little J- 7. hlaJcei, n. sp.
in the cirrus-sockets. Syzygial |
faces angular. J

Eleven or twelve internodal joints. "^
Nodal joints enlarged and pro-
jecting, and the hypozygal
grooved to receive the cirrus-
bases. Syzygial faces nearly ,
plane. J

8. decortis, Wyv. Thorns.

1. Pentacrinus asterius, Linn., sp. (PI. XL; PL XIL figs. 15-25; PL XIII.;
PL XVII. figs. 7, 8).

1761. Palmier niarin, Guettard, Memoires de Mathematique et de Physique tires des Registres de TAcademie

Royale des Sciences, de I'ann^e MDCCLV., Paris, 1761, p. 225.

1762. Encrinns, capite stelhdo, ramnso dichotomo. SfipHe lymtagono-equisetiformi, Ellis, Phil. Trans.,

vol. Hi. pt. i. for the year 1761, London, 1762, p. 358.
1766. Ms asteria, Linnaeus, Systema Naturse, ed. xii., Holmise, 1766, t. i. p. 1288.

REPORT ON THE CRINOIDEA.

301

1775. Isis asterias, Miiller, Linne's Natursystem, nacli der zwulflen lateinisclien Ausgahe, Niirnberg, 1775,

Bd. ii. p. 742.
1788. Isis asteria, Linnfeus, Systema Naturte, ed. xiii., cura, J. F. Gmelin, Lipsia;, 1788, t. i. pars vi. p. 3794.
1816. Encrimis capiit-Mediisw, Lamarck, Histoire Naturelle des Animaux sans Vertobres, t. ii., Paris, 181G,

p. 435.

1820. Pentacrinites vulgaris, von Sclilotheim, Die Petrefactenkunde, Gotlia, 1820, p. 327.

1821. Pentacrinus Caput-Medusm, Miller (pars), A Natural History of the Crinoidea, Bristol, 1821, p. 48, pi. i.
1834. Encrinus caput-Medusm, de Blainville, Manuel dActinologie, Paris, 1834, p. 254.

1836. Pentacrinus Caput-MedusK, Buckland, Geology and Mineralogy, London, 1836, p. 432, pi. 52, fig. 1.
1843. Pentacrinus caput- Afedtisce, Miiller (pars), Abhandl. d. k. Akad. d- "Wiss. Berlin, Jahrg. 1841,

p. 9, Taf. 1.
1845. Pentacrinus Caput-iMedusa-, Austin i (pars), A Monograph on Recent and Fossil Crinoidea, Bristol,

1843-45, p. Ill, pi. 14.
1856. Pentacrimis caput-Medusm, Oersted, Forhandl. Skand. Naturf., 7^' Mode i Christiania, 1856, p. 202.
1862. Pentacrinus caput-Medusce, Dujardin and Hup^, Hist. Nat. des Zoophytes, Echinodernies, Paris, 1862,

p. 181.
1864. Cenoeriyius Caput-Medusa;, Wyville Thomson, The Intellectual Observer, August 1864, p. 3.

1864. Pentacrinus asteria^ Liitken, Vidensk. Meddel. f. d. nat. Foren. i Kj0beuhavn, 1864, p. 207.

1865. Pentacrimis {Neocrinits) asterias, Wyville Thomson, PhiL Trans., vol. civ., 1865, p. 542.

1872. Pentacrimis asteria, Wyville Thomson, Proc. Eoy. See. Edin., vol. vii. p. 765; and The Depths of the

Sea, p. 435.
1877. Pentacrinus asteria, Wyville Thomson, The Atlantic, London, 1877, vol. ii. pp. 123-126.
1882. Pentacrinus asteria, P. H. Carpenter, Bull. Mus. Comp. Zool., vol. x. p. 168.

Dimensions.

Length of stem to twentieth node.

Greatest diameter of stem.

Longest cirrus (fifty joints),

Diameter of calyx across first radials,

Diameter of disk,

Length of arm (one hundred joints).

Length of distichal pinnule (twenty-five joints),

Length of first pinnule after tertiary axQIary (tb'

rteen joints),

48

cm.

7

mm

70

)»

11

))

17

»

100

»?

28

i»

12

1,

Stem long, robust, and generally smooth. Outline pentagonal, with slightly rounded
angles, but more circular in the lower part. luternodal joints thirteen to twenty-one
(usually fifteen to eighteen) in number, with but slightly crenulated edges even in the
upper part of the stem. Nodal joints marked by large, transversely oval cirrus-sockets,
which occupy almost their whole height. The sockets have sharp, well defined rims, and
are entirely filled by the articular facets. Cirri composed of thirty-five to fifty stout,
smooth, and tolerably equal joints, with a small terminal claw and no opposing spine ;
though the ventral surface of the later joints is often marked by two or three blunt points.

Infra-nodal joints rarely grooved to receive the cirrus-bases, and then but slightly so.
Lowest limit of the interartieular pores between the ninth and twelfth nodes.

1 The Messrs. Austin and also Dujardin and Hupe give an undated reference to Farm's Hist. Nat. Havannw
p. 191, pi. 70, in which this type seems to have been mentioned by the name Palma animal. But I have been unable
to consult the work, and can therefore do no more than make this reference to it.

302 THE VOYAGE OF H.M.S. CHALLENGER.

Basals triangular or rhomboidal, not meeting laterally on the exterior of the calyx.
The two outer radials united by syzygy. The grouping of the ray-divisions and the
number of their component joints are somewhat irregular. Primary arms of two to six
(usually three or four) distichal joints ; secondaries of two to eighteen (usually seven to
nine) palmar joints ; tertiaries of six to fourteen joints, and occasionally another division
after eight or ten joints more. The tertiary arms which divide are often only the two
outer ones in each group of four, so that the distichal axillary bears six arms, viz.,
2, 1 ; 1,2; but there are sometimes " more than a hundred arms." ^ The first two joints
beyond each axillary united by syzygy, with a pinnule on the epizygal. No other syzygies
on the arms, which consist of about a hundred smooth, oblong joints, the outermost ones
overlapping slightly. Distichal and palmar pinnules very large and stout, with thick
lower joints, those of the later arm-divisions and of the free arms being much smaller.
The joints of these large lower pinnules have the distal edge raised into a strongly marked
keel which projects forward over the base of the next joint ; and this feature recurs on
all the pinnules of the arms, though it is less distinct in their middle and outer portions.

The perisome uniting the rays up to the level of the distichal axillaries is paved by
large, closely set plates. Similar plates cover the ventral surface of the disk and arm-
bases. Arm-groove moderately wide, and protected by numerous irregular plates.
Pinnule-ambulacra have covering plates, but no definite side plates. Colour in spirit or
dry, a light yellowish-brown.

Localities. — Various parts of the Caribbean Sea — Nevis, Martinique, Barbados,
Guadeloupe (Dr. Schramm) ; off Saba Island, 320 fathoms (Captain Cole). This species
was only once dredged by the U. S, Coast Survey steamer "Blake," cruise of 1878-79,
No. 157, ofi" Montserrat, 120 fathoms.

Remarks. — The specific name caput- Medusce, which was originally bestowed by
Lamarck on the West Indian Pentacrinus described by Guettard, was used by Miller,
Goldfuss, Mtiller, and all later writers till the year 1864. In August of this year the
late Sir WyviUe Thomson* published a popular article on Sea Lilies, in which he' ex-
pressed the opinion that the fossil Pentacrinus hriareus, a widely different species from the
recent Pentacrimis caput-Medusce, " seems, however, to have a just claim to be recognised
as the type of the genus Pentacrinus ; " and he therefore proposed to give the generic
name Cenocrimis to the West Indian species, for which he retained the original title
Caput-Medusce. Lixtken," however, reverted to the original specific designation employed
by Linnaeus,'* and called the type Pentacrinus asteria. In the following year*^ Thomson
referred to it as Pentacrinus {Neocrinus) asterias, Linn. ; but he subsequently" gave up

1 See "Wyville Thomson, The Intellectual Observer, August 1864, p. 5.

2 The Intellectual Observer, August 1864, p. 3. ^ Om Vestindiena Pentacriner, loc. cit., p. 207.
* Systema Nuturw, ed. xii., Holmia', 1766, t. i. p. 1288. ^ Phil. Trans., 1865, p. 542.

« Froc. Roy. Soc. Edin., vol. vii., 1872, p. 765 ; aud also The Depths of the Sea, p. 435.

REPORT ON THE ORINOIDEA. 303

both the generic names Cenocrimis and Neocrinus and described the type as Pentacrinus
aster ia, Linn., as Liitken had previously done.

It was clearly right to return to tlie specific name employed by Linuasus, although he
was utterly at fault as regards the generic position of the type. But every writer on the
subject, myself included, has used a wrong termination to the specific name. Linnaeus
wrote Isis asteria, which appears as Isis asterias in Miiller's edition of the Systema
Naturae (Bd. ii. p. 742), published at Nuremberg in 1775; and this has been quoted
by de Blainville and others. But when the sj^ecies came to be referred to Pentacrinus, and
the Linnean specific name was restored in place of caput- Medtt see, it should have been
written Pentacrinus asterius, the expression Pentacrinus asteria, used by Liitken, Thom-
son, and myself being a false cancord ; for it is evident that the etymology of Linnaeus's
name Isis asteria is the adjective darepLo?, starry, and not the noun Asteria, cat's eye.

I am indebted for this tardy correction to the critical acumen of my friend Prof F.
Jeffrey Bell. But as it did not reach me till all the plates illustrating the type and most
of the morphological section of the Eeport had been printed off, I have been unable to
avail myself of it as fully as I should like to have done.

It is somewhat curious that this species, which for the greater part of a century was
the only known living representative of the genus, should be comparatively so little
known at the present time. But one specimen of it was ever dredged by the " Blake,"
whereas Pentacrinus decorus was obtained by the hundred ; and even stem-fragments
were very rarely met with. One specimen was taken by Caj^tain Cole of the telegraj^h
steamer " Investigator," in 320 fathoms off Saba Island ; and it is now in. the zoological
collection of the Natural History Museum. The agents of Mr. Damon of Weymouth
have been successful in procuring several excellent specimens, which have been bought by
different museums, but I have not been able to examine more than a very few of them.

The preceding description is based upon the characters presented Ijy the following
examples of the type : — A. Miller's original specimen from Nevis, now in the geological
department of the Natural History Museum. B. One dry specimen and two others in
spirit, all in the zoological de^jartment of the same museum. C. One dry specimen in
the Hunterian collection of the Royal College of Surgeons. D. Two dry specimens
obtained by Dr. Carpenter and Sir Wyville Thomson from Mr, Damon of Wejinouth. I
have not made a personal inspection of Guettard's original specimen, but when Mr. Percy
Sladen was in Paris for the purpose of investigating the collection of Asterids in the
museum, he was permitted by Prof. Perrier to examine it on my belialf ; and from the
notes of its characters which he was kind enough to give me, together with the original
figures of Guettard, I have no doubt that it is a fairly normal specimen of the tj^e.

Pentacrinus asterius is much more robust than any of the other recent species of the
genus, none of which have such wide stem-joints, though these are not so large as in some
fossil species. The stem also seems to grow to a greater length than that of any other recent

304 THE VOYAGE OF H.M.S. CHALLENGER.

type but the slender Pentacrinus decorus. But as in so many other species it was liable
to fracture just below a node, so that the individual led a semi-independent existence; for
I have one stem-fragment in which the lowest joint is decidedly worn and its central canal
closed up by a small round boss which projects above the remaining surface of the joint.

The length of the internodes in Pentacrinus asterius distinguishes it at once from
Pentacrinus milUeri, its nearest ally, which has similar long and stout cirri ; while the
infra-nodal joint is usually somewhat grooved to receive the cirrus-bases. This is but
rarely the case in Pentacrinus asterius, and then only to the slightest possible extent, so
that the cirrus-socket is practically limited to the articular facet, without any extension
either upwards or downwards. The stem of Pentacrinus asterius is thus very readily
identified, and the same may be said of the arms, which is rarely the case with the other
species of the genus, unless the ventral groove be examined. For the peculiar features of
the pinnules are very characteristic. They are well shown in Miller's figure' and likewise
in those given by Miiller,^ who specially referred to the projections from the ends of the
pinnule-joints. The great thickness of the basal joints in the distichal and palmar
pinnules, especially the former, is a somewhat unusual character in a Pentacrinus. A trace
of it may be seen in some forms of Pentacrinus mulleri ; but in most species of the genus
the lower joints of the first pinnules are laterally compressed, and lie close against the
arm. They thus present a great contrast to the almo.st cubical or prismatic basal joints
of the first pinnules in Metacrinus (PI. XLIIl. figs. 2, 4). The preceding description
differs in one important respect from those given by Miiller and Lutken. The former
author spoke of the union between the second and third radials as an articulation, but
was somewhat obscure about its nature. He was not able to sef)arate the two joints, but
seems to have inferred that they were united by a bifascial articulation such as he found
m Antedon rosacea.^ While, however, he stated expressly that the latter type has no
muscles between the second and axillary radials, he described and figured muscles as
existing in this position in Pentacrinus caput- Med usee.* This led Lutken to state ** that
the existence of an articulation between the two outer radials of Pentacrinus asterius was
an important character separating it from Pentacrinus tniilleri, which has these joints united
by syzygy. Neither he nor Miiller, however, had ever actually separated the joints,
specimens being then too valuable, or the real state of the case would have become evident
at once. This was first discovered by Sir Wyville Thomson," who pointed out that there
is really a syzygy between the two outer radials of Pentacrin us asterius. In this respect,
therefore, this species resembles Pentacrinus mulleri, instead of differing from it as was
supposed by Liitken. But unfortunately Sir Wyville totally misconceived the real
character of Oersted's species, and confounded it with the one previously described by

' Op. cit., p. 51, pi. ii. fig. 5. - Bau des Penfcicriims, p. 43, Taf. ii., iii.

^ Ibid., p. 26, Taf. ii. fig. 12. ■> Ibid., p. 30, Taf. ii. fig. 8.

^ Om Vestindiens Pentacriner, loc. cit, p. 202. " Proc. Roy. Soc. Edin., vol. vii., 1872, p. "66.

REPORT ON THE CRINOIDEA. 305

himself under the name of Pentacrinus decorus. He was thus led to state that while the
two outer radials of Pentacrinus asterius are united by sj'zygy, those of Pentacrinus
miiUeri are articulated, a mistake almost the opposite of that made by Liitken. It is not
difficult to understand how Miiller's error arose ; for the line of syzygy is not dotted as it
is in the Comatulae, where the apposed faces are marked by radiating ridges. There is
little or nothing of this striation on the syzygial surface of Pentacrinus asterius (PI. XII.
figs. 18, 21) ; and the muscular unions of the remaining calyx- and arm-joints are so very
close that there is hardly any external character by which the syzygy between the two
outer radials may be detected. Its presence is evident enough between the two joints
following each axillary, as they are shorter than their successors. But nothing of this
kind appears in the case of the radials, and as they are very apt to become slightly
separated at the edges I have found it almost impossible, especially in dry specimens, to
determine the real nature of the union without separating the joints. This was done
by Sir Wyville Thomson, whose preparations are figured in PI. XII. figs. 15-25.

The basals of Pentacrinus asterius are of very variable size, like those of Pentacrinus
decorus, though not to the same extent. They are sometimes small, rounded knobs, not
meeting one another upon the exterior of the calyx ; or larger and more prominent,
meeting their feUows in the re-entering angles beneath the radials ,' but they never form
a completely closed ring of triangular or pentagonal plates flush with the radials, as they
do in Pentacrinus wyviUe-thomsoni, Pentacrinus cdternicirrus, and some other species
(PL XVIII. figs. 1-3 ; PI. XIX. figs. 1, 6, 7 ; PL XX. figs. 2, 3; PI. XXV.).

Liitken, Thomson, and others have endeavoured to establisli as one of the characters
of Pentacrinus asterius that the nodal joints of the stem are low and simple, while those
of Pentacrinus miilleri and Pentacrinus decorus are thick and double, consisting of two
parts united by syzygy. This is not really the case, however. In all recent Pentacrinidse
the nodal joints proper, i.e., those which are pierced by the canals lodging the cirrus-
vessels, are always single and united by syzygy to those next below them. But the line
of junction is sometimes so nearly obliterated as to be only visible with difficulty; while
in other cases, such as Pentacrinus asterius (PI. XIII. fig. 8), it is hardly distinguishable
from the ordinary junction lines between the internodal joints, which are crenulated
much less than usual.

The following may therefore be regarded as the special marks of Pentacrinus asterius: —
a robust stem with long internodes and wide cirrus-sockets ; the ray-divisions rather
irregular in number and grouping ; the projections of the pinnule-joints.

Little is known about the range of Pentacrinus asterius, either in depth or in space.
Examples have been obtained ofi" Barbados, Guadeloupe, Martinique, Montserrat, Nevis,
and Saba Island ; while the only two cases in which the depth is known with certainty
are respectively 120 and 320 fathoms. Like the three other West Indian species, it is
not known to occur out of the Caribbean Sea.

(ZOOL. CHALL. EXP. PAET XXXU. — 1884.) 39

306

THE VOYAGE OF H.M.S. CHALLENGER.

2. Pentacnnus mulleri, Oersted, 1856 (Pis. XIV., XV.; PI. XVII. figs. 9, 10).

1821. Pentacrinus capuf-Medusie, MiUer (pars), A Natural History of the Crinoidea, p. 46.
1813. Pentacrums capiit-Medusce, Miiller (pars), Abhandl. d. k. Akad. d. Wiss. Berlin, 1843, p. 185.
1845. Pentacnnus Capuf-Medusce, Austin (pars), A IMonograph of Eecent and Fossil Crinoidea, p. 111.
1856. Pentacrinus Mulleri, Oersted, Forhandl. Skand. Naturf., 1^' Mode i Christiania, 1856, p. 202.

1864. Pentacrinus Miilleri, Liitken, Yidensk. Meddel. f. d. nat. Foren. i Kjobenhavn, 1864, Nr. 13-16, p. 207.

1865. Pentacrinus (Neocriniis) Miilleri, Wyville Thomson, Phil. Trans., 1865, vol. civ. p. 542.
1882. Pentacrinus miilleri, P. H. Carpenter, Bull. INIus. Comp. Zobl., vol. x.. No. 4, p. 170.

non Pentacrimis Miilleri, WyviUe Thomson, in Proc. Koy. See. Edin., vol. vii. p. 776 ; and in The

Depths of the Sea, p. 442.
non Pentacrinus Miilleri, Agassiz and Pourtales, in Bull. Mus. Comp. Zoijl., vols, v., vi.

32

cm.

185

mm

117

»>

6

n

56

J>

11

?»

14

)»

160

»>

18

J?

20

11

Dimensions.

Total lengtli of largest specimen,^

Greatest length of entire stem, rounded off at sixteenth node,

Shortest stem, rounded off at twelfth node,

Diameter of stem, ....

Longest cirrus (forty-three joints),

Diameter of calyx, ....

Diameter of disk, ....

Length of arm (one hundred joints),

Length of pinnule on first free brachial (fifteen joints).

Length of pinnule from middle of arm (twenty-one joints).

Stem robust, but of no great length. Outline pentagonal, with rounded angles and
smooth surface. Internodal joints four to eleven (usually six to eight) in number, with
but slightly crenulated edges, even in the upper part of the stem. Cirrus-sockets trans-
versely oval and not reaching the upper edges of the nodal joints, but extending more or
less downwards on to the infra-nodals, which are grooved to receive the cirrus-bases.

Cirri composed of thirty to forty-five stout, smooth, and tolerabl}^ equal joints, the
later ones of which may have a couple of small, blunt projections on the ventral side.
Terminal claw small and without an opposing spine. Lowest limit of the interarticular
pores between the fourth and eighth nodes.

Basals variable ; sometimes pentagonal, forming a closed ring ; sometimes triangular
or rhomboidal, barely in contact })y their lower angles ; and sometimes c^uite small, not
meeting at all upon the exterior of the calyx. Pays and their subdivisions not separated
by perisome, but in close lateral contact, the joints as far as the lowest free brachials
beyond the tertiary axillaries having their sides more or less flattened, often very much
so. The two outer radials united by syzygy. There are usually six or eight arms on

1 The total length of this individual, which was obtained by Captain Cole, and is now in the Natural History Museum,
is slightly greater than that of the largest specimen dredged by the " Blake." The stem, which is 19 cm. long, is broken
just below the twenty-first node. But in the same bottle there is a fragment which appears to be the bottom part of
this stem, and has the lowest nodal joint closed in the usual way.

REPORT ON THE CRINOIDEA, 807

each ray in the following order — 2,1; 1,2; or 2,1,1; 1,1,2, the axillaries being limited
to the outer arm of each successive pair. Primary arms of one to three (usually two)
distichal joints, which are united by syzygy. Secondary arms of two to seven palmar
joints, usually three. There may be one, or more rarely two, further divisions, which
are usually of three joints each, but may have five or six.

The first two joints after each axillary are united by syzygy, the epiz}'gal, when not
itself an axillary, bearing the first pinnule. No other syzygies on the arms, which
consist of about one hundi'ed smooth, oblong joints, the lowest of which have flattened
sides, and are sometimes more or less tubercular, while the outer ones overlap slightly.
The first pinnules are somewhat longer and stouter than the following ones, after which
the size again increases. The pinnules have a smooth, sharp dorsal edge, and are generally
composed of flattened joints, the lowest of which are slightly wider than their successors,
and are sometimes markedly prismatic.

The disk bears numerous scattered calcareous granules, but no regular pavement of
plates, and there is no plated perisome between the rays. Arm-groove moderately wide,
and closely covered by numerous, small, irregular plates. Pinnule-ambulacra have
covering plates and Ul defined side plates.

Colour in spirit, white or brownish-white, somewhat darker on the pinnules.

Localities. — Guadeloupe, St. Croix, St. Thomas, Barbados ; between Saba and
Eustatius Islands, 531 fathoms (Captain Cole).

The dredgings of the U. S. Coast Survey steamer "Blake," cruise of 1877-78, oft"
Havana; 175 fathoms.

Cruise of 1878-79. No. 100, ofi" Morro Light ; 250 to 400 fathoms. No. 101, off"
Morro Light ; 175 to 200 fathoms. No. 157, ofi"Montserrat; 120 fathoms. No. 171, oft"
Guadeloupe; 183 fathoms; bottom temperature, 5 5|^° F. No. 193, off" Martinique ; 169
fathoms; fine sand, dark mud, and shells; bottom temperature, 51° F. No. 218, off
St. Lucia; 164 fathoms; bottom temperature, 56° F. No. 269, off St. Vincent ; 124
fathoms ; bottom temperature, 57^° F. No. 274, off Barbados ; 209 fathoms ; fine sand
and ooze ; bottom temperature, 53^° F. No. 280, off St. Charles Lighthouse, Barbados ;
221 fathoms; Globigerina sand; bottom temperature, 50i° F. No. 283, off Barbados;
237 fathoms; hard bottom; bottom temperature, 49° F. No. 291, off Barbados; 200
fathoms ; flat calcareous stones ; bottom temperature, 49f ° F. No. 295, off" Barbados ;
180 fathoms ; hard bottom ; bottom temperature, 50|°F. No. 296, off Barbados; hard
bottom ; 84 fathoms ; bottom temperature, 61^° F. No. Y. (Bartlctt), Santiago de Cuba ;
288 fathoms.

Remarks. — This species was discovered by the Danish naturalist Oersted, who brought
an example to Europe from the Danish possessions in the West Indies. It was exhibited
in the year 1856 at the meeting of Scandinavian naturaHsts in Christiania,^ and a brief

I Forhandl. Skand. Naturf., 7'^' Mode i Christiania, 18CG, p. 202.

308 THE VOYAGE OF H.M.S. CHALLENGER.

preliminary diagnosis of it, together with a similar notice of Pentacrinus caput-MeduscB
{asterius), was published in the report of the meeting. Such was the rarity of these
Crinoids that Oersted's discovery of a new species attracted but little attention, although
an example of it from Guadeloupe had long been contained in the collection of the
Geological Society of London, and had been referred by Miiller to Pentacrinus caput-
MedttscB. These facts seem to have escaped the notice of Sir Wyville Thomson, who
himself described a new species [Pentacrinus decorus) in 1864, and spoke of it and Penia-
crinus caput-Medusce as the only two known living species of the Stalked Crinoids.'

Early in the next year, however, an elaborate memoir on the West Indian
Pentacrinidse was published by Dr. Liitken, which has served as the basis of most of the
subsequent work on the genus. Not only did he make a careful examination of Oersted's
original specimen of Pentacrinus miilleri, but he found that two other individuals in the
Copenhagen Museum were identical with it ; he was thus able both to discover some
more important points of difference between Pentacrinios miilleri and the Linnean type,
and also to work out some of the individual variations in the characters of Pentacrinus
mitlleri as defined by Oersted,

In his preliminary diagnoses of Pentacrinus asterius and Pentacrinus miilleri, Oersted
had already indicated the differences in the numbers of joints composing the arm-divisions
of the two species. This character was still further investigated by Liitken,^ who pointed
out its influence upon the external appearance of the animal. Belying chiefly uj)on the
figures of Pentacrinus asterius which were given by Miller and Miiller, he showed that
the numbers of joints in the successive arm-divisions were respectively 5 or 6, 9
or 10, and 12. In Pentacrinus miilleri, on the other hand, these numbers are 2;
2-4 ; 3 ; and 3-5 ; and it is almost always only the two outer arms on the ray
which divide at all, so that the arms of any ray with secondary axillaries would be
represented by the expression 2,1; 1,2; and by 2,1,1; 1,1,2, if tertiary axillaries be
present. This is a sort of indication of the inequality of the arm-divisions of Extra-
crinus, and is tolerably constant in Pentacrinus miilleri, though not limited to that
species, for it is visible in Pentacrinus asterius, as detected by Quenstedt ^ in Miller's
figure.

After the publication of Lutken's Memoir, Pentacrinus miilleri. Oersted, came to be
recognised as a ly^^ distinct from the old Pentacrinus asterius. It was referred to by
Sir Wyville Thomson,* together with Pentacrinus asterius and Pentacrinus decorus, so
that he evidently regarded it as distinct from both of them. Later on, however, he
seems to have come to the conclusion that his Pentacrinus decorus was identical with
Oersted's species. For having previously said that Pentacrinus asterius and Pentacj-imis
decorus were the only two known living species of the genus, he made nearly the same

' Sea Lilies, The Intellectual Observer, August 1864, p. 1. - Om Vestindieiis Pentacriiier, lot: cit., p. 203.

3 Encriniden, p. 190, Tab. 97, fig. 3. •• Fhil. Trans., vol. civ., 1865, p. 542.

REPORT ON THE CRINOIDEA. 309

statement concerning Pentacrinus aster ins and Pentacrinus miilleri} He likewise
repeated most of his original description of Pentacrinus decorus as a diagnosis of Penta-
crinus iniilleri, witli a reference under the latter name to the specimen which he had
before him when describing Pentacrinus decorus. He stated that the two outer radials
of Pentacrinus astei'ius were united by syzygy, and further added that " the arrange-
ment of the joints and the syzygies in the cup is the same in Pentacrinus miilleri as in
Pentacyinus asteria, only the syzygy between the second radial and the radial axillary is
not so complete." This passage obviously refers to a ligamentous articulation as distin-
guished from a true syzygy on the one hand, and from a must;ular joint on the other ;
and it is by no means in accordance with Liitken's very positive statements as to the
presence of a true syzygy between the two outer radials of Pentacrinus miilleri. Neither
does Sir Wj^dlle's description of the nodes as occurring about every twelfth joint agree
with Liitken's diagnosis, which records only four to ten internodal joints in Pentacrinus
miilleri. As a matter of fact there are eleven or twelve internodal joints in Pentacrinus
decorus, and there is no syzygy at all between the two outer radials, but only a bifascial
articulation such as occurs in the majority of the Neocrinoidea, and has often been
wrongly spoken of as a syzygy, though clearly distinguished from it by Miiller. This is
shown in figs. 3 and 6 on PL XXXIV., a copy of which was lettered " Pentacrinus miilleri,
Oersted," by Sir- WyvUle Thomson. I cannot Ijut think, however, that if he had lived to
w^ork out the " Blake " collection more fully than he was able to do before his health gave
way, he would have retained his original views as to the distinctness of his Pentacrinus
decorus from Pentaerinus miilleri, Oersted. The result of tliis confusion was that the
numerous specimens of Pentacrinus decorus wlaich were dredged by the "Bibb" and the
" Blake " in the Gulf Stream and in the Caribbean Sea were referred to Pentacrinus
miilleri by Pourtales and Agassiz,^ The two species have really no sort of resemblance
to one another, differing in all the characters of the stem, the cirri, the calyx, and the
arms.

The foregoing description is based upon an examination of _^four specimens from the
"Blake" collection, two purchased by Sir Wyville Thomson from Mr. Damon, one in the
collection of Sir Eawson Rawson, and lastly that in the Museum of the Geological
Society of London, which is mentioned by both Miller and Muller as Pentacrinus caput-
Medusce.

Pentacrinus miilleri is readily distinguished from Pentacrinus asterius, which is its
nearest ally, by the shortness of the internodes and the modification of the hj^ozygal
joints, which, however, is far less marked than in Pentacrinus decorus. The basals
generally form a complete ring ; while the lu-anching of the arms is mut-li more regular

1 Proc. Boy. Soc. Edin., vol. vii. p. TG6 ; and The Depths of the Sea, pp. 434, 435 ; see .also The Atlantic, vol. ii.
p. 126.

2 Bull. Mits. Comp. Zool, vol. i. p. 357 ; Ibid., yol. v. pp. 56 and 214 ; Ibid., vol. vi. p. 296.

310 THE VOYAGE OF H.M.S. CHALLENGER.

than in Pentacrinus asterius. Thei-e are fewer joints between the successive axillaries,
and the characters of the pinnules are altogether different. In these latter points Penta-
crinus miilleTi closely resembles Pentacrinus macleaixmus (PI. XVI.), Pentacrinus
wyviUe-thomsoni (PI. XVIII. fig. l), and Pentacrinus alternicirrus (PI. XXV.). But
the first named has only one or two internodal joints in the stem, while Pentacrinus
wyville-thomsoni has from thirty to forty-five, so that they are both readily distinguish-
al)le from Pentacrinus mulleri ; while the grouping of the cirri on the stem of
Pentacrinus alternicirrus is sufficient to distinguish this species at once.

So far as I can judge from the material at my disposal, Pentacrinus mulleri is
certainly the most variable of the Pentacrinidje with the exception of Pentacrimts
decorus. The stem does not seem to reach the length which it attains both in the latter
species and in Pentacrinus asterius. In one instance it is rounded ofi" at the twelfth node,
only 135 mm. from the calyx, and Rhizopocls are attached to the under surface of the
lowest nodal joint. Another stem tapers gradually downwards from a width of 5 mm. at
the calyx to 3 mm. at the sixteenth node, where it is rounded ofi" 185 mm. from the calyx.

The length of the internodes varies a good deal in different individuals, though as a
rule it is tolerably constant in any given stem. The component joints are usually thick
and thin alternately. This is very marked in the specimen represented in PL XIV.,
though not well shown in the figure ; while in other cases the joints are more equal in
height, as shown in PL XV. fig. 4. This figure should be compared with the correspond-
ing one of Pentacrinus asterius (PL XIII. fig. 8), in which the cirrus-sockets are not so
deeply hollowed as they are in Pentacrinus mulleri. Their shape, too, is somewhat
variable in the latter type. In some stems (PL XV. fig. 4) they are transversely oval as
in Pentacrinus asterius (PL XIII. figs. 4, 8), though not reaching so near the top of the
nodal joint. But they always extend slightly downwards on to the hj-pozygal, which is
not the case in that species. In other examples, however, the hypozygal is deeply
grooved to receive the bases of the cirri, and the sockets thus become more circular in
form ; so that it appears as if the cirri were borne conjointly by the two syzygial joints.
This has been described as an important difi"erence between Pentacrinus mulleri and
Pentacrinus asterius, but erroneously so ; for the whole of the articular surface is always
on the nodal joint, which is the oul}- one pierced by the canals lodging the cirrus- vessels.

The cirri, though always stout, are considerably shorter in some forms than they are
in others ; and Avliile some of them have quite smooth terminal joints, those of other
individuals bear small blunt processes which never, however, reach to the size of a spine.

The "Blake" collection includes a curious fragment of a stem which had broken
between a nodal joiut and the first joint of the internode above it. The upper part of
the stem and the calyx are missing ; but six irregularly shaped joints have been added
above the node. One would like to know whether this reparation would ever have
resulted in the formation of a new calyx and arms. Such an extensive reparation

REPORT ON THE CEINOIDEA. 311

seems scarcely possible, though the development of a new visceral mass inside the calyx
is not uncommon.

As in Fentacnnus asterius, there is a considerable variation in the development of the
basals. In the Copenhagen specimens described by Lutken, and in some of those which
I have examined, they are pentagonal in outline and form a closed ring separating the
radials from the top stem -joint. In other forms they are more prominent and
rliomboidal or triangular in shape, but only just meeting one another in the re-entering
angles of the calyx (PL XIV.; PI. XV. fig. 2) ; while in Sir Rawson Rawson's specimen
they are small and inconspicuous (PL XV. fig. l), as in some varieties of Pentacrinus
decorus (PL XXXVI.).

The number of arms may vary from thirty-five to forty-five, some individuals occa-
sionally having ten arms to the ray. Generally, though not invariably, the axillaries are
limited to the outer arm of each pair in the manner already described ; Ijut I have not
seen any individual in which the six- or eight-armed arrangement is constant on every ray.

As there are very few joints separating the axUlaries, there is comparatively little room
for the arms, the liases of which are therefore more or less flattened laterally, both
in the outer and in the inner parts of the rays. In fact, wherever an axillary occurs the
two arms which it bears have their inner faces flattened, while the outer sides of the rays
are flattened continuously from the second radials to as far as some six or eight joints
beyond the fifth axillary.

This feature is especially marked in two fine specimens from the " Blake" collection,
which are also distinguished by the shape of their lower pinnule-joints. One is from
Martinique and the other from Barbados. The calyx and arm of the former are repre-
sented on PL XV. figs. 2, 3. The outer edges of the joints from the second radial
onward are produced somewhat sharply upwards, and fit closely against those of adjacent
joints. This is less prominent in the Barbados specimen, which shows an occasional
tendency towards carination of the arm-bases. The other form is remarkable for the
abnormal condition of one of its rays, as shown in PL XV. fig. 2. The third radial is
articulated to the second instead of being united to it by syzygy. But it is itself a
syzygial joint; so that there are primitively four radials, a character wliieli indicates a
tendency to variation in the direction of Metacrinus with its five or eight primitive radials
(PL XXXIX. fig. 1 ; PL XL VI.). The pinnules of these two individuals are also different
from those of other examples of the type. They are generally composed of moderately
broad, flattened joints, the lowest of which are somewhat stouter than their successors.
But in the two " Blake " specimens the pinnules are less flattened than usual, and the
lower joints markedly trihedral in form, recalling, though in a less degree, the prismatic
shape of the pinnules, which is characteristic of Metacrinus (PL XXXIX. fig. 1 ;
PL XLIII. fig. 4).

The plating of the disk of Pentacrinm mulleri (PL XVII. fig. 10), like that of

312

THE VOYAGE OF H.M.S. CHALLENGER.

Pentacrinus decorvs (PL XXXIV. fig. 2), is far less complete than in Pentacrinus
asterius. There is no trace of the large polygonal plates on the perisome between the
rays which we find in the latter species (PL XIII. fig. 1), and those on the ventral sur-
face of the disk are small and scattered, often being mere granules. The ambulacral
skeleton too is imperfectly developed. The covering plates of the pinnules do not rest
upon distinct side plates, but only upon an almost undiff"erentiated limestone band
(PI. XV. figs. 8, 9 ; PL XVII. fig. 9).

3. Pentacrinus maclearanus, WpHille Thomson, 1877 (PL XVI. ; PL XVII. fig. 1).
1877. Pentacrimis madearanits, Wyville Thomson, The Atlantic, vol. ii. pp. 123-126.

Di<me7\sions.

Total length {fide C. W. T.),

Length of stem, rounded off at twelfth node,

Diameter of stem,

Longest cirrus (twenty-five joints).

Diameter of calyx, .

Length of arm (sixty-eight joints).

Length of pinnule on first free brachial (ten joints),

Length of pinnule from middle of arm (fifteen joints),

13-00 cm.
34-00 mm.

5-25 „
28-00 „

9-00 „
80-00 „

7-00 „
14-00 „

Description of an Individual. — Stem short and pentagonal, with rounded angles,
terminating below at the twelfth, node. The internodes consist of only one or two
comparatively thin joints. Nodal joints thicker, with enlarged and j^rominent angles ;
the cirrus-sockets, occupying almost their whole height, are cii-cular or slightly oval in form,
with a well-defined rim which extends dowTiwards on to the infra-nodal for a variable
distance. The cirri consist of twenty to twenty-five stout joints of tolerably equal size,
with a small terminal claw and no opposing spine. luterarticular pores scarcely visible.

Basals rhomboidal, just in contact laterally, and extending slightly downwards over
the ujipermost stem-joints. Rays and their subdivisions in close lateral contact, the
joints as far as the tenth or twelfth brachial having flattened sides. The two outer
radials united by syzygy. Primary and secondary arms each of two joints also united by
syzygy. Total number of arms thirty-one, i.e., usually six to each ray in the foUowing order
— 2, 1 ; 1,2; palmar axillaries being generally developed only on the two outermost of the
four secondary arms. A tertiary axillary in one ray. The two lower brachials united by
syzyg}^, the epizygal bearing a pinnide. No other syzygies on the arms, which consist of
about seventy short and wide, oblong joints, overlapping very sHghtly at the base.

The lowest j)innules are much shorter than their successors, and have only ten or
twelve joints, the basal ones being trihedral and the outer ones flattened. The middle
pinnules much longer, with about fifteen more rounded joints.

REPORT ON THE CRINOIDEA.

313

The disk is ouly partially visible, but seems to have a tolerably regular pavement of
small plates. Arm-groove exceedingly narrow, the ambulacral plates being limited to the
pinnule-bases. The pinnules have well defined covering plates with occasional faint
indications of side plates.

Colour in spirit, browTiish- white.

Xoca%.— Station 122, August 10, 1873; lat. 9° 5' N., long. 34° 50' AV. ; 350 fatjioms;
red mud. One specimen.

Remarks. — The leading characters of this elegant species were well described by Sir
Wyville Thomson in The Atlantic. Its nearest allies are Pentacrinus miilleri (PL XIV.),
Pentacrinus ivyville-thomsoni (PL XVIII. fig. 1), and Pentacrinus alternicirrus
(PL XXV.), all of which it resembles very closely in the regularity and grouping of the
arm-divisions. The stem, however, is totally different from that of Pentacrinus wyville-
thomsoni with its long internodes, and the number of arms is also greater than is usually
the case in this species. Pentacrinus miilleri and Pentacrinus alternicirrus approach it
most closely, having relatively short internodes ; but the arrangement of the cirri on the
stem of the latter species distinguishes it at once. Pentacrinus maclearanus closely
resembles some specimens of Pentacrinus miilleri in the shape of the nodal joints ; but,
apart from the characters of the internodes, it is readily identified by the unusual
shortness of its lowest pinnules and the more rounded form of the joints in the later ones.

The close approximation of the nodal joints, and the downward curvature of the
cirri, together with the slight downward extension of the basals and the grouping of the
arms, are of interest as recalling the Liassic Extracrinus hriareus.

Only four cirri are present at one of the nodes, the fifth socket being totally absent.
This indicates a variation in the direction of Pentacrinus alternicirrus, which has only
two or three cirri at each node (PL XXVI. figs. 1-3).

4. Pentacrinus wyville-thomsoni, Jeffreys (PL XVII. figs. 2-6 ; Pis. XVIII. -XXIV. ;
PL LVII. fig. 1).

1870. PeMacrinus Wyville-Thomsoni, Jeffreys, Proc. Roy. Soc, 1870, vol. xix. p. 157.

1872. Pentacrinus icyville-thmnsoni, Wyville Thomson, Proc. Roy. Soc. Ediii., vol. vii. p. 767; and The
Depths of the Sea, p. 444.

Dimensions.

Total length, ......... 24-00 cm.

Longest stem, rounded off at seventh node.

Shortest stem, lounded off at fifth node,

Diameter of stem,

Longest cirrus (nineteen joints),

Diameter of calyx,

Diameter of disk.

Length of arm (seventy joints),

Length of pinnule on first free brachial (ten joints),

Length of pinnule from middle of arm (twenty-one joints),

(ZOOL. CHALL. EXP. — PART XXXII. 1884.)

. 155-00

mm

. 80-00

3-50

2400

7-20

9-75

, 100-00

. 10-00

. 24-00

Ii40

314 THE VOYAGE OF H.M.S. CHALLENGER.

N.B. — Examples of this species have been distributed to several museums, and only
a few have come into my hands ; so that the first three of the measurements given above
must not be taken as indicating the whole range of variation throughout all the indi\dduals
which were dredged by the " Porcupine."

Stem smooth and moderately robust, but of no great length. The upper part is
pentagonal with rounded angles ; but the lower portion of an old stem is almost perfectly
circular. Thirty to forty, and occasionally more, internodal joints with crenulated edges
which are less distinct below. Nodal joints enlarged, with sharp angular ridges which
stand out prominently between the cirrus-sockets. These are transversely oval, and
occupy the whole height of the nodal joint, which projects outwards over the upper edge
of the infra-nodal, while the supra-nodal is slightly grooved to receive the bases of the
stout cirri. These consist of about eighteen tolerably ec[ual, smooth, and thick joints, the
lowest of which are broader than their successors, especially in mature indi\4duals.
Terminal claw small, without an o^aposing spine. The lowest limit of the interarticular
pores is a little on either side of the fourth node.

Basals pentagonal, but sometimes approaching the triangular, ^^'ider than high, and
forming a closed ring. The rays and their subdivisions in close lateral contact, the first
five or six joints after the distichal axillary having flattened sides. The two outer radials
united by s}^zygy. Fourteen to twenty-two arms, distichal axillaries being often absent,
and palmars very rare. Primary and secondary arms each of two joints united by syzygy ;
the first two brachials united in the same way, the epizygal bearing a pinnule. The arms
of about seventy smooth, oblong joints in which syzygies are very rare.

The first pinnides c[uite short, consisting only of nine or ten joints, the lowest of
which are broad and flat, the later ones longer and more slender. The following pinnules
increase rapidly in size, and soon become long and tapering, consisting of about twenty
smooth, elongated joints, the lowest of which are slightly flattened. Disk completely
covered with a j)avement of small plates, as is the brachial perisome above the muscular
bundles. Arm-groove moderately wide, and bordered by a discontinuous series of
ambulacral plates. The pinnule-ambulacra have covering plates, but very ill defined
side plates.

Colour grass-green, becoming white in spirit, which acquires a purplish-red tinge.

Locality.— RM.^. "Porcupine," 1870. Station 17 ; lat. 39° 42' N., long. 9° 43' W.;
1095 fathoms; ooze; bottom tem^ierature, 39°'7 F. About twenty specimens.

Also the "Talisman," 1883; off" the Morocco Coast: and again off Rochefort ;
lat. 45° 59' 30" N., long. 6° 29' 30" W. of Paris ; 1500 metres (800 fathoms).

Remarhs. — This fine species was first obtained by Dr. Gwpi Jeff"reys during the
" Porcupine " expedition of 1870 ; and it was dedicated Viy him to his friend and colleague
Sir AVyville Thomson in the general account of the voyage which was published in the

REPORT ON THE CRINOIDEA. 315

Proceedings of the Royal Society. It was not described, however, till the year 1872,
when Sir Wy^dlle contributed a notice of the " Porcupine " Crinoids to the Royal Society
of Edinburgh ; and in the following year he reproduced this description in The Depths
of the Sea, together with a woodcut which gives a very fair idea of the princij)al char-
acters of the type. All the entire specimens obtained were dredged at Station 17; but
a few fragments of stem and arms were also met with at Station 17a (740 fathoms),
together with ten specimens of Antedon lusitanica. Thirty specimens were recently
dredged by the "Talisman" in 1500 metres off Rochefort. Dr. Gywn Jeffreys^ records
that " portions of the arms occurred in several other of the ' Porcupine ' dredgings on
the Lusitanian coasts ; and joints of apparently the same species have been found
by Prof. Seguenza in the Zanclean formation or older Pliocene near Messina." The latter
point, however, can hardly be properly decided without a careful study of both types.

In the structure of the ray-divisions and arms, Pentacrinus wyville-thomsoni is closely
related to Pentacrinus miilleri, Pentacrinus maclearanus, and Pentacrimis alter nicirr us,
especially the latter ; but it is at once distinguished from them all by the shape of the
nodal joints, the short stout cirri which they bear, and the great length of the internodes
which separate them. It is also remarkable for the manner in which the stem ends below
in a nodal joint which is closed up beneath and rounded off, as shown in PL XXII. fig. 27.
According to Sir Wyville Thomson^ "all the stems of mature individuals of this species
(which were dredged by the ' Porcupine ') end uniformly in a nodal joint, surrounded
with its whorl of cirri, which curve downwards into a kind of grap^iling root (PI. XIX.
fig. 1). The lower surface of the terminal joint is in aU smoothed and rounded, evidently
by absorption, showdng that the animal had for long been free " (PL XXII. fig. 27). The
positions of this terminal nodal joint and the corresponding length of stem in three
individuals which I have examined are as follows : — stem 80 mm. long, terminating at
the fifth node ; stem 90 mm. long, terminating at the sixth node; stem 155 mm. long,
terminating at the seventh node.

The zoologists of the " Talisman " claim to have proved, however, that Sir W}mlle
Thomson was wrong in his belief that the individuals dredged by the " Porcupine " were
leading a semi-free existence, loosely rooted in the soft mud. In one of a series of
popular articles by Mons. H. Filhol,^ a member of the " Talisman " expedition, it is stated
that Sir AVyville came to this conclusion after having examined one of the "Porcupine"
specimens ; and a free translation is given of the last sentence of the paragraph just
quoted, from which, however, the words "in all " are entirely omitted. It is thus made
to appear as if Sir- Wy\Tlle had drawn his conclusions from the condition of only one
example of Pentacrinus uyville-thomsoni, which is very far from being the case ; while
he also stated in the next paragraph to that cjuoted by Filhol that he had remarked

' Proc. Roy. Soc, 1870, vol. xix. p. 157.

2 Proc. Boy. Soc. Edin., vol. vii. p. 767 ; The Depths of the Sea, p. 444.

3 Explorations sous-mai-iiies. Voyage tlu " Talisman," La Nature, No. 568, April 19, 1884.

316 THE VOYAGE OF H.M.S. CHALLENGER.

the same character "as occurring in some specimens of Pentacnnus mulleri" i.e., the
type now known as Pentacrinus decorus.

Filhol continues, however, " nous avons constate, apres avoir remonte des debris de
roches, que ces animaux vivaient, contrairement h ce que Ton avait cru pouvoir supposer,
completement fixds par des cirres recourbes se detachant de I'articuLation terminale de la
tige. Ces sortes de crochets se soudent en quelque sorte avec le fond sur lequel ils
reposent et il faut les briser pour les detacher. Par consequent les Pentacrinus Wyville-
Thompsoni {sic), que Ton a recontres libres, avaient d<\ etre arraches a la suite de quelque
accident du fond sur lequel ils vivaient, car il parait l^ien difficile d'admettre que les
niemes animaux en des points divers de I'ocean aient des modes d'esistence differents."

The observations here recorded are undoubtedly of great value ; but the conclusions
drawn from them by Filhol appear to me to be somewhat rash. The " Talisman" speci-
mens of Pentacrinus wyville'thomsoni seem to have been living on a stony or rocky
bottom ; and in fact Prof. Perrier ^ records that " plusieurs ont ete ramenees avec les
cailloux sur lesquels Us sont fixes." There can be no question therefore that Pentacrinus
wyville-thomsoni lives in a permanently fixed condition on a hard bottom. But the
"accident" which is supposed by Filhol to have liberated some fifteen fixed individuals
must have been of a rather extensive character ; and it must further have taken place at
a sufficiently long time l:)efore they were dredged by the " Porcupine " for the lowest nodal
joint of one of them to have lost its natural appearance (PI. XXII. fig. 20) and have
become enlarged and rounded as shown on PI. XXII. fig. 27. But in other specimens the
lowest nodal joint is far less modified. It retains its pentagonal shape and the thickened
rim of the syzygial face, in the centre of which there is a small rounded tubercle covering
the opening of the central canal. If aU these specimens had been detached by one
general " accident " anterior to the arrival of the " Porcupine's " dredge and tangles
among them, their lowest nodal joints should have been in the same condition and not
in difi"erent stages of modification. The same " accident " must have happened to the
Pentacrinus decorus of the Caribbean Sea and to the Pentacrinus maclearanus of the
Challenger dredgings, lioth of which w^ere described by Sir Wyville as having the stems
closed up at a modified nodal joint ; but Filhol makes no reference whatever to these two
types. He does not appear to dispute the fact that the " Porcupine " individuals of
Pentacrinus wyville-thomsoni were free when captured ; but he regards the observations
of the "Talisman" as proving that this condition was not a natural one. Sir Wyville^
believed that although the Pentacrinus, like the young Comatula, " was doubtless attached
in its early days, it appears to have finally parted from its attachment, and to have led a
free life ; " and he pointed out that the syzygial union of the stem-joints at the nodes
facilitated the rupture of the stem, just as is the case with the syzygies in the arms. His

' L'Expedition du Talisman, Revue Scientifique, No. 24, December 15, 1883, p. 741.
^ Sea Lilies, The Intellectual Observer, August 1864, p. 7.

REPORT ON THE CRINOIDEA. 317

theory accounts for the varying conditions of the lowest nodal joint which are presented
by different individuals; while the "accident" hypothesis of Filhol's does not explain
this fact at all, unless he means that some individuals had Ijecome detached at one time
and some at another. But this is precisely what Sir Wyville believed ; only he regarded
it as a natural event correspondmg to the separation of the centro-dorsal of a ComatuJa
from the rest of the stem below it, which appears to me to be a more rational explanation
of the facts than that suggested by FilhoL

It is possible that the " accident " referred to by the latter author may not mean
the fracture of the stem at a syzygy beneath a nodal joint as I have supposed above,
but a separation of the stem from its anchorage by the lowest whorl of cirri. This would
account for the modification of the nodal joint in the " Porcupine " specimens ; for in the
figure which FUhol gives to illustrate the mode of attachment of those dredged by the
" Talisman," the lowest nodal joint is represented as not in contact with the rock beneath, to
which its cirri are soldered ; so that it might very well have been thickened and rounded
in the manner described above. But this supposition, while removing one difficulty, only
introduces another. The " Porcupine " specimens were living on a bottom of ooze at a
depth of 1095 fathoms, considerably greater than the 1500 metres (800 fathoms) at which
the " Talisman " examples were discovered on a rocky or stony bottom. Now in the
first place, the cirri could not solder themselves to this ooze with such firmness that they
would break rather than loose their hold ; and secondly, there is no trace of such a
connection in the cirri borne by the modified lowest nodal joint of any of the " Porcupine "
specimens. For in some individuals (PL XIX. fig. l) they are as perfect as those at the
nodes above them, and not broken as they were in the detached specimens dredged by
the " Talisman." In the case of Rhizocrinus, however, the attachment of the radicular
cirri to stones and shells by slight calcareous expansions is well known (PI. X. fig. 15).

All the Challenger Pentacrinidae, with one exception, were dredged from mud or ooze,
and though the stems of several of them terminate below in a rounded nodal joint as in
the "Porcupine" examples of Pentacrinus wyville-thomsoni, I have seen no traces of
their being attached by the cirri of the lowest whorl soldering themselves to the bottom,
as described by the French zoologists.

The condition of four species of Pentactinus which I have carefully examined vAih.
reference to this point appears to me to show conclusively that these observations have
by no means the general application which is claimed for them. The internodes oi
Pentacrinus ivyville-thomsoni are very long, while the cirri are short (PI. XIX. fig. 1), so
that only those of the lowest whorl can come in contact with anythmg beneath the lowest
nodal joint. But the case is far difi'erent in many other species, among which I select
four for special consideration, as they are represented by individuals in which the lowest
whorls of cirri are better preserved than usual. In the present condition of many of the
specimens several of the cirri on the stem are more or less broken ; and tliough in the

318 THE VOYAGE OF H.M.S. CHALLENGER.

case of the lowest whorls this may be due to the cmi having been soldered to the bottom
by calcareous expansions, I have never seen any ti'aces of such a condition.

In the first place, there is a stem-fragment of Pentacrinus asterius, consisting of a few
iuternodal joints with a nodal joint beneath them which is somewhat worn and has its
central canal closed up by a low rounded tubercle ; so that it must have been detached
for some time from the infra-nodal joint which completed the syzygy. But all the cirri
borne by this modified nodal joint are perfect throughout their whole length, nearly
70 mm., and were most certainly not soldered to the bottom at the time the animal was
captured.

Then again, in a fine Pentacnnus millleri from Martinique with a stem 120 mm.
long, which is closed below at the thirteenth node, all the cirri of this node are perfect
from the base to the terminal claw. They reach 50 mm. in length, and are spread out
in different directions, two being curved sharply upwards, while the others are more or
less horizontal. Their general appearance is very similar to that of the long cirri of
Antedon phalangium. But not one of them shows any trace of having been soldered to
the bottom. The cirri of the next four whorls above are all long enough to have touched
the bottom had the animal been attached like the individuals of Pentacrinus ivyville-
thomsoni dredged by the "Talisman." Eight of these twenty cii'ri, four in the first
whorl, two in the next, and two in the highest one, are now more or less broken ; but
this is clearly due to accident, and not to the fracture of an attachment. The same may
be said of a specimen of Pentacrinus millleri in the Natural History Museum, which has
a stem 135 mm. long, with several of the lower cirri remaining unbroken; while the
under surface of the twelfth nodal joint at which the stem ends is worn and somewhat
rounded, and bears two or three attached Foraminifera. The animal cannot, therefore,
have been attached by the base of the stem, though the cirri may have been soldered to
the bottom ; but their appearance is against this supposition.

The same remarks apply to Pentacrinus alter nicirrus, in which the cirri reach 50 mm.
in length. Those of the four or five lowest whorls turn more or less directly downwards,
and all reach below the level of the terminal nodal joint, which may be anywhere between
47 and 155 mm. from the calyx. But none of these cirri in any of the twelve individuals
which I have examined show any signs of having been soldered to the bottom. Many of
them are now broken ; Ijut others are cpite perfect, though they must have been in
contact with the bottom, had the animals been permanently anchored like the " Talisman "
examples of Pentacrinus wyville-thomsoni. The single Pentacrinus dredged by the
Challenger on hard ground belonged to this species ; and as all the cirri of the lowest
whorls are more or less broken, it is quite possible that they may have been torn away
from a permanent attachment to the bottom. Another conclusive argument against the
general application of the " Talisman " observations is afforded by the condition of the
single specimen of Pentacrinus maclearanus which was dredged by the Challenger from a

EEPOET OX THE CPJNOIDEA. 319

bottom of red mud (PL XVI. fig. l). The cirri, though short, are very closely set, and those
of the five lowest whorls bend downwards underneath the last nodal joint and interlace
with one another so as to form a kind of basket-work just like that beneath the centro-
dorsal of a Comatula with many cirri, such as Anteclon eschrichti. This led Sir Wyville
Thomson ^ to remark that " from the attitude of the cirri and from the appearance of the
end of the stem there can be no doubt that this specimen is complete, that it is mature,
and that it was li\'ing in an unattached condition." I do not well see how this statement
can be disputed. Neither do I understand the difficulty of admitting that the mode of
life of a Pentacrinus may vary in different localities. The Comatulse are fixed when youno-,
and semi-free when mature, attaching themselves by their cirri to various objects ; but
some species [Actinometrajukesi and Actinometra stellata, &e.) eventually lose their cirri
altogether, and must then live an absolutely free life. The Palaeozoic Agassizocrinus and
Edriocrinus were attached when young, but subsequently became perfectly free. Con-
sidering that these great changes take place during the life of a single individual, I fail to
see the difficulty of admitting that a particular species of Pentacrinus can adapt itself to
the conditions of its existence, some young individuals fixing themselves permanently when
they have the opportunity ; while others living ou soft ooze in deeper water separate
themselves from their original anchorage and lead a partially free existence, being only
attached temporarily, just as a Comatula is. A precisely similar case to that oi Pentacrinus
wyville-thomsoni is presented by Pentacrinus deco7'us. Some individuals are firmly
fixed to telegraph cables by the spreading base of their stem, while others have been
found in the semi-free condition.

Circumstances alter cases ; and the question of the natural freedom of the uidi\'idual
represented in PL XIX. fig. 1, which has five perfect cirri on a rounded nodal joint at the
base of the stem, is by no means negatived, because the " Talisman " found several others
attached by calcareous growths round the cirri of the lowest whorl. The French
zoologists, however, appear to consider that this observation proves Sir AVj^'Ule to have
been WTong ; whereas, on the contrary, the dredgings of the Challenger and the " Blake "
have confirmed his views in the most satisfactory manner.^

Except at the lowest nodal joint the cirri of Pentacrinus wyville-thomsoni appear to
be usually directed upwards (PL XVIII. figs. 1,3; PL XIX. fig. l) ; and the supra-nodal
joint is accordingly slightly grooved for the reception of the cirrus-bases (PL XIX.
figs, 3, 4 ; PL XXII. fig. 17) instead of the infra-nodal joint as is so markedly the case
in Pentacrinus hlalcei and Pentacrinus decorus (PL XXXI. figs. 1, 3 ; PI. XXXIV. fig. 1;
PL XXXVI.), in which the cirri are usually directed downwards. In this respect, there-
fore, Pentacrinus wyville-thomsoni presents an approach to the genus Metacrinus, ia

1 The Atlantic, vol. ii. p. 126.

^ Mucli of what has been written above would have appeared more suitably in Chapter II. pp. 18-22, where the
mode of life of the PentaerinidK is discussed. But as Filhol's article did not appear till after this chapter had gone to
the printers, and did not come under my notice till five months later, I have been obliged to take up the subject again.

820 THE VOYAGE OF H.M.S. CHALLENGER.

wliicli the upward direction of the cirri and the groo-sdng of the supra-nodal joint ae
tolerably constant characters (PI. XXXVIII. ; PI. XXXIX. figs. 3, 9 ; Pis. XL., XLIL,
XLV. ; PI. XLVII. figs. 1, 2 ; PL XLVIIL).

Another small point of resemblance between the European Pentacrinus and the
Pacific Metacrinus is the slight tendency sometimes showTi by the basals of the former to
send median downward extensions over the interradial ridges at the top of the stem
(PI. XVIII. fig. 2), for this character is very generally distinctive of Metacrinus
(PL XXXIX. fig. 1 ; PL XLIII. figs. 1, 3 ; PL XLVIIL). The basals of Pentacrinus
wyville-thomsoni are almost always markedly pentagonal, the height being decidedly
greater in the midtUe than at the sides, where, however, it is usually distinctly appre-
cialile (PL XVIII. fig. 3 ; PL XIX. figs. 1, G, 7 ; PL XX. fig. 3). Sometimes, however,
they become almost triangular in outline (PL XVIII. figs. 1, 2), and one or more of them
occasionally fail to meet their feUows, a variation which is more frequently met wdth in
Pentacrinus mUlleri (PL XV. figs. 1, 2) and Pentacrinus naresianus (Pis. XXVIIL,
XXIX.).

The number of arms in Pentacrinus wyville-thomsoni is comparatively smaU, being
sometimes as low as fourteen ; for two or even three of the rays may have no axillary
but the third radial, as is sometimes the case in Pentacrinus decorus (PL XXXV.); and the
distichal axillaries, when present, rarel}^ occur all round the cup (PL XVIII. fig. 3). The
examples figured in PL XVIII. figs. 1, 2, and PL XIX. figs. 1, 6, 7, are some of those
with the greatest number of arms, a tertiary (palmar) axillary being occasionally present
beyond the distichal ; but I do not know of any specimen in which the number of arms
exceeds twenty-two.

The disk (PL XVII. fig. 6) is closely covered by a pavement of anambulacral plates,
several of which are pierced by water-pores. These are almost entirely absent in the anal
interradius, in the proximal part of which the plates are closer set than usual, and arranged
into two lateral groups. At first sight these look like large single plates, and are
suggestive of orals, but they become resolved on further examination into small and very
closely set plates. The ambulacra of the disk are protected by irregular plates which
cover them in completely in the dry state. They are more regularly arranged on the
arms, but are discontinuous on alternate sides between the pinnule-bases (PL XVII. fig. 4) ;
and the perisome covering the muscular bundles is likewise plated, as in the allied
Pentacrinus alternicirrus (PL XXVII. fig. 6). The covering plates of the pinnule-
ambulacra are not very distinctly marked ofi" from the lateral calcareous band, especially
at the bases of the pinnules ; and this band itself is but very imperfectly difi"erentiated
into side plates (PL XVII. figs. 2, 3).

Unlike many of the Pacific Pentacrinidse, which were white when fresh, living
examples of Pentacrinus wyville-thomsoni have a beautiful grass-green colour. This
becomes duller in spirit, which acquires a purplish-red tinge. Prof Moseley kiucUy

REPORT ON THE CRINOIDEA.

321

examined some of it with the spectroscope, and found the usual Ijands of pentacrinin.
A few specimens which have not been kept in the dark, but have been more or less
exposed to light, have bleached completely white.

195

mm.

113

47

5

52

9

U

100

10

21

5. Pentacrinus alternicirrvs, n. sp. (Pis. XXV., XXVI.; PL XXVII. figs. 1-10).

DimensioTis.

Total length, ....

Longest stem, rounded off at sixteenth node.

Shortest stem, rounded off at eleventh node.

Diameter of stem.

Longest cirrus (thirty joints), .

Diameter of calyx,

Diameter of disk.

Length of arm (eighty joints), .

Length of pinnule on first free brachial (eleven joints).

Length of pinnule from middle of arm (twenty-one joints),

Stem smooth, short, and pentagonal, with rounded angles. Four to nine (usually
five or six) internodal joints with crenulated edges. The nodal joints bear two and
three cirri alternately, those at one node corresponding to the positions of the absent
cirri at the nodes next above and below. Cirrus-sockets nearly circular, occupying the
whole height of the nodal joint, and extending upwards on to the supra-nodals. Infra-
nodals scarcely modified at all. The sockets are deeply hollowed and have prominent
lateral rims, owing to the angles of the joint between them being produced outwards
and rounded. Cirri composed of about thirty stout joints, the lowest of which, after
the first four, are somewhat longer than their successors. The terminal claw is
moderately large, and has no opposing spine, but the ventral surface of the later joints
is a little uneven. The lowest limit of the interarticular pores is between the fifth and
eighth nodes.

Basals rhomboidal, extended slightly downwards, and produced laterally so as to
meet their fellows in the re-entering angles of the calyx. The rays and their subdivisions
in close lateral contact ; the joints as far as the sixth or eighth brachial having flattened
sides. The two outer radials united by syzygy. About thirty (twenty -five to thirty-two)
arms, usually six to each ray, the axillaries being limited to the outer di^asions. Primary,
secondary, and tertiary arms (the latter very rare) each of two joints united by s}-zyoy.
The two lowest brachials united in the same way, the epizygal bearing the first pinnule.
Arms of about eighty smooth, oblong joints with syzygies at intervals of three to eight
(usually five or six) joints. The first one between the ninth and twenty-sixth brachials.
First pinnules quite short, consisting only often or eleven joints, the lowest of which are
broad and flat, and the later ones long and slender. This inequality gradually disappears
as the pinnules increase in length towards the middle of the arm, where they are tapering

(ZOOL. CHALL. EXP. — PART XXXII. — 1SS4.)

Ii41

322 THE VOYAGE OF H.M.S. CHALLENGER.

and styliform, consisting of about twenty smooth, flattened joints with sharp dorsal
edges. The disk is completely covered with a pavement of small plates, as is the brachial
perisome above the muscular bundles. Arm-groove moderately wide, and bordered by a,
discontinuous series of ambulacral plates. The pinnule-aml^ulacra have covering plates,
and sometimes moderately distinct side plates.

Colour in spirit, yellowish- white, sometimes retaining a rosy tinge.

Localities. — Station 171, July 15, 1874 ; near the Kermadec Islands; lat. 28° 33' S.,
long. 177° 50' W. ; GOO fathoms; hard ground; bottom temperature, 39°'5 F. One
specimen.

Station 214, February 10, 1875; off the Meangis Islands ; lat. 4° 33' N., long. 127° G' E. ;
500 fathoms ; blue mud ; bottom temperature, 41°"8 F, Several specimens.

Uncertain — Station 210, January 25, 1875 ; off the Panglao and Siquijor Islands; lat.
9° 2G' N., long. 123° 45' E. ; 375 fathoms ; blue mud ; bottom temperature, 54°! F.

Some of the fifteen specimens sent to me were without labels ; and I strongly suspect
that this species, together with Pentacrinits naresianus, three examples of which are
without labels, are those referred to by Sir Wyville, who recorded in his journal that
four specimens of two species of Pentacrinidse were dredged at Station 210; for the
collection contains no specimens at all with the label of this Station.

Remarhs. — Pentacrimis alternicirrus, like Pentacrinus wyville-thomsoni, apjaears to
be pre-eminently one of those which lives in a semi-free condition, the stem having been
broken at a nodal joint, the syzygial face of which becomes worn and more or less rounded,
and has its central canal closed up. The following list shows the position of this terminal
nodal joint and the coi'respouding length of the stem in twelve specimens.

Stem, 47 mm. long and terminating at tlie 11th node.

„ 49 „ „ 11th

„ 63 „ „ 11th

„ C4 „ „ 11th

„ 65 „ „ 11th

„ 55 „ „ 12th

„ 66 „ „ 12th

„ 69 „ „ 12th

„ 70 „ „ 12th

„ 65 „ „ 14Lh

„ 91 „ „ 14th

„ 113 „ „ 16th

The remarkable arrangement of the cirri in this species distinguishes it at once from
all the other recent Pentacrinidte. Except that the symmetry is pentamerous instead of
tetramerous, the arrangement recalls that of the leaves on the stem of a Labiate plant.
In one case only have I found any irregularity. The eighth whorl has its two regular
cirri like the sixth, together with an additional one which therefore comes to be on the

REPOET ON THE CEINOIDEA. 323

same side of the stem as the third cirrus in the seventh whorl. But the ninth whorl has
only two cirri instead of three, and is the beginning of a new cycle ; for the tenth whorl
is not like the sixth with only two cirri, but resembles the irregular eighth one with
three. In the same way the eleventh whorl is like the ninth and not the seventh,
and so on.

The absence of cirri at some of the nodes of Pentacrinus alternicirnis is the more
striking as there are regularly five cirri at each node in all the Pcntacrinidse, both recent
and fossil, with three exceptions. These are Pentacrinus hronnii, Hagenow,^ from the
White Chalk of Riigen, and Pentacrinus didactylus, d'Orbigny,^ from the Eocene of
Biarritz, both of which have only two cirri at a node ; while under the name of Penta-
crinus tridactylus, Quenstedt ^ has described another Tertiary stem-fragment from Le Vit
in the south of France, which has a verticil of three cirri only. It is just possible that
if longer pieces of these stems were known they might show the same regular alternation
in the positions of the successive cirrus-whorls which is so striking iu Pentacrinus alterni-
cirrus. But whether this be the case or not, the departure from the pentamerous
arrangement of cirri which is so characteristic of the Pentacrinidai is not a little remark-
able. For verticils of two cirri alternating with one another in position sometimes occur iu
both Bourgueticrinus and Mesocrinus ; though the structure of the stem in these genera
is totally different from that of Pentacrinus, as has been fully explained in Chapter II.

In consequence of the absence of two or three cirrus-sockets, the nodal joints of
Pentacrinus alter iiicirr us depart considerably from the symmetrical form presented by
those of other Pcntacrinidse, as is shown in PI. XXVI. figs. 13, 14, and PL XXVII.
figs. 2, 3. The last two represent syz3^gial faces of two successive nodes in their relative
positions, the two empty sides in fig. 2 being occupied by sockets in fig. 3.

Apart from the arrangement of the cirri, Pentacrinus alternicirrus resembles Penta-
crinus maclearanus and Pentacrinus miilleri in the shortness of the internodes, while it
agrees with both these species and also with Pentacrinus loyville-thomsoni in the
regularity and the grouping of the arm-divisions. The general arrangement of the crown
of arms (PI. XXV.) is most like that of Pentacrinus ivyville-tliomsoni (PI. XIX. fig. 1);
and the long middle pinnules of the two species are very similar, while the characters of
the perisomatic skeleton are almost identical (compare PI. XVII. figs. 2-4, and PI.
XXVII. figs. 4-6).

The leading characters of Pentacrinus alternicirnis appear to be very constant, the
South Pacific specimen from near the Kermadecs being in no way distinguishable from
those dredged ofi* the Meangis Islands. This is a striking contrast to the variations of

1 Monograpliie der Eugenscheu Kreide-Versteinerungen, Neiies Jahrh.f. Mineralogie, Jahrg. 1840, p. 663, Taf. ix.
fig. 9.

- See d'Archiac, Description des fossiles recueillis par M. Tliorent, dans les couclies h uummulines des enviro n
de Bayonne, Mem. Soc. (jM. de France, 2"'" ser., t. ii. 1« partie, 1846, p. 200, pi. v. figs. 16«, 17a.

3 Encriniden, p. 268, Tab. 99, fig. 170.

324 THE VOYAGE OF H.M.S. CHALLENGER.

Pentacrimis decoriis and Pentacrinus miilleri in the Caribbean Sea. There are ncarlj-
always five or six internodal joints in the stem, and I have only noticed two cases of an
arm-division consisting of more than two joints united by syzygy. One has three joints, of
which the first two form a syzygy, and in the other there are four, those of each pair being-
united by syzygy. In the arms, however, the position of the syzygies, after that at the
base, is exceedingly variable. But this is always the case in the few species of Penta-
crinus which have syzygies in the arms, the contrast between them and the Comatulse
being very striking in this respect.

Pentacrinus alternicirrus appears to inhabit moderately deep water, the depths at the
two Stations from which it is recorded being respectively 500 and 600 fathoms ; while
at the doubtful Station 210 the depth was 375 fathoms. Five of the individuals dredged
at Station 214 (Meangis Islands) were infested with encysted Myzostomas, as were also
many of the Comatulse. In one specimen no less than eight arms bore more or less
perfect cysts of Myzostoma pentacrini, von Graff, two of them having two cysts a short
distance apart. In other cases the cysts were principally formed in the skeleton of the
pinnules by Myzostoma deformator, von Grafi",^ as sho-s\Ti in PL XXVII. figs. 7 and 8 ;
while figs. 9 and 1 0 represent cysts formed in the substance of the arm.

6. Pentacrimis naresianus, n. sp. (PI. XXVII. figs. 11-13 ; Pis. XXVIII.-XXX.).

Dimerisions.

Total length of largest specimen, stem broken at thirtieth node,

Length of this stem, ....

Diameter of stem, .....

Longest cirrus (thirty-five joints).

Diameter of calyx, .....

Length of arm (eighty joints),

Length of first pinnule (twelve joints).

Length of pinnule from middle of arm (twenty-two joints),

54-00 cm.

38-00 „

5-00 mm.

25-00 „

7-75 „

150-00 „

8-.^0 „

24-00 „

Stem long and smooth, of a rounded pentagonal or circular form. Eight to eighteen
(usually al>out ten or twelve) internodal joints wiih but slightly crenulated edges.
Nodal joints high, not projecting outwards at the angles, but deeply hoUowed by the
cirrus-sockets, which have nearly cu-cular facets and terminate far below the upper edges
of the nodal joints. lufra-nodals deeply grooved to receive the cirrus-bases, so that
the sockets appear to have pyriform do-wnward extensions. Cirri moderately slender,
of thirty to thirty -five tolerably uniform joints, all but the lowest of which have one
or two blunt projections on the dorsal edge. Lowest limit of the interarticular pores
between the fifth and eidith nodes.

o

Basals small, triangular or pentagonal, sometimes meeting laterally and sometimes

1 Zool. Chall. Exp., part xxvii. pp. C2-C6, 1884.

REPORT ON THE CRINOIDEA. 325

not, owing to the radials lieing prolonged slightly downwards over the upper stem-
joints. Ten arms only, but the rays are in close lateral contact, the second radials l)eing
united all round, while the axillaries and the first two brachials have flattened outer
sides. Second and third radials and the first two lirachials respectively united by
bifascial articulation. Arms of about ninety joints,^ the lower ones bluntly wedo-e-
shapcd and the later joints nearly oblong or squarish, with raised distal edges so as to
overlap slightly. A syzygy in the third brachial, another between the sixth and
fifteenth joints, and others at intervals of four to twelve (usually seven to nine) joints.
The proximal face of the epizygal forms a sharp angle projecting backwards into the
retreating distal face of the hypozygal, the dorsal surface of which projects foi-wards
into that of the epizygal. The first pinnules arc quite short, consisting oidy of about
twelve joints, the lowest half of which are broad and flattened and the later ones quite
small. This inequality gradually disappears as the pinnules increase in length, the
lower joints becoming less broad and the later ones more elongated. Those on the
middle of the arm are long and styliform, consisting of about twenty joints, the first of
which are much broader than the rest.

Disk covered wdth numerous, closely set, irregular plates. These extend on to the
arms at the sides of the ambulacra., which are altogether al)0ve the arm-groove, and
are protected by a continuous series of tolerably regular covering plates. There are
no definite side plates on the pinnules, but only a narrow liand of limestone with its
edges cut into teeth which bear the covering plates.

Colour in spirit, pinkish-white.

Localities. — Station 170, July 14, 1874 ; near the Kermadec Islands ; lat. 29°55'S.,
long. 178° 14' W. ; 520 fathoms ; volcanic mud ; bottom temperature, 43° F. Two
quite young specimens.

Station 170a, July 14, 1874; near the Kermadec Islands; lat. 29° 45' S., long.
178° 11' W. ; 630 fathoms ; volcanic mud; bottom temperature, 39°'5 F. A stem-
fragment only.

Station 171, July 15, 1874; near the Kermadec Islands; lat. 28° 33' S., long.
177° 50' W. ; 600 fathoms ; hard ground ; bottom temperature, 39°-5 F. A stem-
fragment and some broken arms.

Stationl75, August 12, 1874; near Fiji; lat. 19° 2' S., long. 177°10'E.; 1350 fathoms;
Globigerina ooze ; bottom temperature, 36° F. A calyx with a fragment of a stem.

Station 214, February 10, 1875; off" the Meangis Islands; lat. 4° 33' N., long.
127° 6' E. ; 500 fathoms ; blue mud ; bottom temperature, 4r-8 F. Six specimens.

Uncertain — Station 210, January 25, 1875 ; off" the Panglao and Siquijor Islands;
lat. 9° 26' N., long. 123° 45' E.; 375 fathoms ; blue mud ; liottom temperature, 54°-l F.

Three specimens reached me without any lal)els. Some or all of them were probably

' I have seen no arm with more than eighty joints ; but the extremities of the arms are broken in all the specimens.

326 THE VOYAGE OF H.M.S. CHALLENGEE.

obtained at Station 210, where four individuals of two species of Pentacriniis were
dredged (see \). 128) ; but, as in the case of Pcntacrinus alternicirrus, I can only infer
this from the fact that these two species, together with the single specimen of
Metacrinus murrayl, were the only ones which came into my hands without any record
of locality ; while there were none with the label of Station 210.

Remarks. — Pentacriniis naresianus is remarkable as being the only recent species of
the genus in which there are not more than ten arms ; while the shape of the arm-joints,
especially the lower ones (PL XXVIII. fig. 1; PL XXIX; PL XXX. fig. 1), is also
more oblique than is usually the case in the Pentacrinidse, so that in both respects it
approaches the Comatulse. Apart from these characters and the well plated ambulacra
(PL XXVII. fig. 13), the arms are readily distinguished by the peculiar form of the
syzygial unions. AVhen seen from the dorsal side (PL XXIX. fig. 1 ; PL XXX. fig. 1),
the distal edge of the hypozygal appears to be very convex and to project strongly
forwards into the epizygal ; while in a side view (PL XXX. fig. 23) the epizygal shows
a sharp backward projection into the hypozygal. When the syzygial faces are exposed
(PL XXX. figs. 20, 21), a sharp angle appears across the middle of the proximal face of
the epizygal ; and the distal face of the hypozygal has a corresponding re-entering angle,
so that the two joints interlock very closely. Essentially the same form of syzygy
recurs in Pentacrimis blcd-ei of the Caribbean Sea (PL XXXII. figs. 5, 7, 12, 14) ; but
the other characters of this species are entirely difi"erent from those of Pentacriniis
naresianus, as it has twenty to thirty arms and a more slender stem with shorter
internodes (PL XXXI. ).

These two species, together with Pcntacrinus decorus, are the only three recent
Pentacrinidge in which the two outer radials and the two first joints beyond them are
united by bifascial articulation. The articular faces of these joints in Pentacrirms
naresianus are shown in PL XXX. figs. 11, 12, 16, 17. It further resembles Pcnta-
crinus decorus in the pyriform downward prolongation of the cirrus-sockets over the
infra-nodal joints, the upper faces of which are markedly stellate in consequence
(PL XXX. fig. 29). The nodal joints, however, are not produced outwards at the angles
between the cirrus-sockets, nor do they slope outwards from their upper edge to the top
of the sockets ; so that the general outline of the stem is very even (PL XXVIII. fig. 2).
But in Pcntacrinus decorus this enlargement of the nodal joints is very perceptible in
mature individuals (PL XXXVI. ; PL XXXVII. figs. 1, 2), though not in the youngest
(PL XXXA^.). On the other hand, it appears in the youngest specimens of Pentacrinus
naresianus (PL XXXa. fig. 5), though disappearing some time before maturity is reached.

Pentacriims naresianus also presents a considerable variation in the size and shape
of the basals, which is so very remarkable a character of Pentacrinus decorus. In aU
the four figures of the calyx which were drawn for Sir '\Vy\'ille Thomson (Pis. XXVIII.-
XXX.), the basals are represented as separated by small downward extensions of the

REPORT ON THE CRINOIDEA. 327

radials ; but this is not quite the case in the original of fig. 2 on PI. XXIX., for the
basals form a compact ring entirely separating the radials from the top of the stem.
In other specimens, again, some of the basals meet their fellows, while the remainder
are more or less completely separated by the downward extending radials.

Pentacrinus naresianus does not appear to be one of those which live in a semi-
free condition like the three last described, while the stem grows to a greater length
than in most of these forms. It is broken below in all the specimens obtained, and
though this has sometimes taken place at a node, the fracture is evidently a recent
one, the syzygial surface not being worn and more or less rounded, as in Pentacrinus
toyville-iliomsovi and the other semi-free types.

The young individuals of Pentacrinus naresianus, besides exhibiting the usual
characters common to all young Pentacrinidse {ante, pp. 289-291), have one or two
peculiarities of their own. The second radials are less closely united, only meeting one
another for half the length of their sides (PI. XXXa. fig. 1 ) ; while the sides of the
axillaries and of the two following joints are not so much flattened as in the adult, but
the edges where the ventral and dorsal surfaces meet are sharp and straight.

The characters of the nrm-syzygies are also slightly diff'erent from those which
appear in the adult. The backward projection of the epizygal is much nearer the edge
of the joint than in the adult arm, in which the crest of the ridge on the syzygial face
crosses the axial canal. This gives an entirely different appearance to the joints when
seen in profile, as will be evident upon a comparison of figs. 9-12 on PI. XXXa., wdiich
represent a young and an old syzygial pair, as seen from the side and from above
respectively.

The difference in the sculpture on the young and on the older stem-joints is also
shown in PL XXXa. figs. 2, 3, 7. In the young individual figured on the same plate
the head is but 55 mm. long, and there are only about fifty joints in the arms. The
diameter of the stem is 2 mm. Its internodes are exceptionally long, seventeen or
eighteen joints ; and there are only two cirri at one of the nodes (the fifth),^ just as is
apparently the case through the whole stem of Pentacrinus didactylus.

Two stem-fragments from this Station (170), one of which (and possibly both)
belong to this same individual, exhibit some remarkable peculiarities of growth. In the
upper one (PL XXXa. fig. 5) two of the nodal joints are slightly enlarged as described
above. But seven joints lower down a kind of calcareous sheath appears on the outside
of the stem, wdiich is segmented in the same way as the stem, and is continued do\vn-
wards over the next node. This is of an altogether abnormal character. The outer
crust shows various irregular lines, and seems to have filled up the downward extensions
of the cii-rus-sockets on to the infra-nodal joints, so that no tra(;e of them is visible.

1 The absent cirri at this node were erroneously inserted 1iy the artist, when restoring the broken ones elsewhere ;
and I did not notiee the fact till it was unfortunately too late to remedy it.

828 THE VOYAGE OF H.M.S. CHALLENGER.

This couclition recurs over a length of three iutemodes ou the remaining stem-fragment,
the sockets gradually becoming more and more obscured ; and at the lowest node the
incrustation seems to have completely overgrown the bases of the cirri, nothing appear-
ing to indicate their presence but minute irregular stumps. Fifty joints lower down
the stem terminates in a flattened expansion by which it was probably attached. The
abnormal condition of this stem is interesting from its resemblance to that presented by
a specimen of the fossil Millericrinus pratti, which I have described elsewhere.^ In
this case, however, the secondary deposit of limestone which is outside the uppermost
stem-joints is divided up into segments not corresponding \\T.th those enclosed by it.

7. Petitacrmus hlakei, P. H. Carpenter (Pis. XXXI., XXXII. ; PI. XXXIII.

figs. 1-3).

1882. Fentaunnus hlakei, P. H. Carpenter, Bull Mus. Comp. Zool., vol. x. p. 172.

Dimensions.

Total length, ......

Longest stem, rounded off at twenty-fourth node,
Diameter of stem, .....
Longest cirrus (twenty-nine joints),
Diameter of calyx, .....
Length of arm (one hundred and tive joints), .
Length of pinnule on first distichal (eighteen joints), .
Length of pinnide from middle of arm (twentj' joints).

27-5 cm.

160-0 mm

-I'O

17-0

6-0

120-0

9-0

12-0

Stem slender and smooth, with a rounded pentagonal outline ; five to seven inter-
nodal joints with distinctly crenulated edges. Nodal joints high, not projecting out-
wards between the cirrus-facets, which are comparatively small and cii'cular, and do
not nearly reach their upper edge. Infra-nodals more or less grooved to receive the
cirrus-bases, so that the sockets appear to have pyriform downward extensions of
variable size. Cirri small and slender, of twenty -five to thirty joints, the first six of
which are quite short, and the remainder squarish ^^ith a tolerably smooth dorsal edge.
Terminal claw small, with no opposing spine. Lowest limit of the interarticular pores
between the sixth and tenth nodes.

Basals small, triangular, and more or less rounded, well separated laterally Ijy the
lowest parts of the rather high radials. The rays and their subdivisions moderately close,
without any iutei'vening perisome, but scarcely flattened at all except just on the lower
brachials. The two outer radials and lowest distichals respectively united by bifaseial
articulation. The third radials and the second distichals project backwards into the middle
of the preceding joints, their proximal surfaces being somewhat deeply hollowed from side
to side, and slanting downwards and backwards. About twenty arms, usually only

1 Qimrt. Joarn, Col Soc, vol. x.\xviii. p. 33, pi. i. tig. 21.

REPORT ON THE CRINOIDEA. 329

four, but oceasioually three or five to each ray. Primary arms of one to four distichal
joints, of which the first two are united by a ligamentous articulation. The second one
bears a pinnule, and the next may be either a simple axillary or consist of two syzj^gial
joints. Secondary arms (rare) of two palmar joints united by syzygy. Anns of about
ninety slightly overlapping joints, nearly oblong at first, but afterwards becoming more
unequal sided. The first free brachial bears a pinnule, and the second is usually a
syzygial joint, while other syzygies occur on the arms at intervals of two to twenty
joints. The proximal face of the epizygal forms a sharp angle which projects backwards
into the retreating distal face of the hypozygal.

The lowest brachial pinnules consist of seventeen or eighteen joints, the first six of
which are rather broad, and the later ones more slender. Farther out on the arm this
inequality gradually disappears, all the joints but the first two being moderately long and
narrow. The disk, so far as visible, is covered with small, closely set plates which extend
on to the arms, covering the muscular bundles at the sides of the narrow aiTu-groove.
The covering plates of the pinnule-ambulacra rest upon a denticulated band of limestone
which is not fully difi'erentiated into side plates.

Colour in spirit, yellowish-white.

Localities. — All in the Caribbean Sea. From the dredgings of the U.S. Coast
Survey steamer " Blake." Cruise of 1878-79. No. 157, ofi" Montserrat ; 120 fathoms.
No. 209, ofi" Martinique; 189 fathoms; hard bottom; bottom temperature, 49|° F.
No. 291, ofi" Barbados ; 200 fathoms ; flat calcareous stones; bottom temperature, 49|° F.
No. 295, off Barbados ; 180 fathoms ; hard bottom ; bottom temperature, 50|° F.

Remarhs. — This interesting species, like Pentacrinus asterms, appears to be some-
what rare, having only been dredged four times by the " Blake"; and it has never, so far
as I know, been obtained by Mr. Damon's agents, as the other three West Indian .species
have been. I have only been able to examine a mutilated specimen from Montserrat,
and three nearly perfect ones from Barbados.

The slender cirri and theii- close approximation on the stem give this type a
more elegant appearance than Pentacrinus decorzis, which is its nearest ally, these two
being the only species with more th-an ten arms which have a bifascial articulation
between the radials. The third species with this character (Pentacrinus naresianns)
is only ten-armed ; but it has a singular resemblance to Pentacrinus hlahei in the form
of the syzygies on the arms. The stem of Pentacrinus hlaJcei is at once distinguished from
that of Pentacrinus decorus by the shortness of the internodes and the absence of any
enlargement at the nodes (PI. XXXI. fig. 3). The ciiTUS-sockets, too, are generally pro-
longed dowTiwards to a less extent than they are in Pentacrimis decorus (V\. XXXYL);
though Pentacrinus blakei shows a considerable amount of variability in this respect, some
sockets having almost no downward extensions at all, while others are more like those of
Pentacrimis decorus. There is, however, remarkably little variation in the length of the

(ZOOL. CHALL. EXP. PART XXXII. 188-1.) ^' ^'-^

830 THE VOYAGE OF H.M.S. CHALLENGER.

internodes, especially as compared with Pentacrinus mulleri, in which the stem reaches
about the same length. There are also more arms in Pentacrinus decorus than in Penta-
crimis hlahei, in which palmar series are very rarely present, so that tjie total number of
arms would not exceed twenty. But the great distinguishing character of Pentacrinus
hlahei is the nature of its bifascial articulations and syzygies. Seen from the dorsal side
or in profile (PL XXXI. figs. 1, 2 ; PL XXXII. fig. 15), the third radial shows a strong
backward projection into the second. But its proximal face not only is concave from
side to side, but also slopes strongly downwards and backwards ; and the upper ventral
edge of the second radial is bent forward so as to fit into the gap thus formed
(PL XXXII. fig. 18). There is a very slight indication of this in Pentacrinus
naresiamis (PL XXX. figs. 1, 11, 12) ; but the two species resemble on,e another much
more closely in the curious angular form of the syzygial faces. Those of Pentacrinus
naresianus (PL XXX. figs. 20, 21, 23) have been alre.ady described, and those of
Pentacrinus hlakei are shown in PL XXXII. Whether it be a brachial syzygy
(figs. 4, 5, 7) or one in the distichal axillary (figs. 9, 12, 14) the form is just the same.
The proximal face of the epizygal rises to a sharp crest, which is interrupted by the
central canal, and fits into a corresponding re-entering angle on the distal face of the
hypozygal, so that the muscle-plates of its proximal face are bent strongly forwards,
just as they are in the bifascial articulation of the second radials with the axillaries
(PL XXXII. figs. 15, 18). The general appearance of the syzygies in the side view of
an arm is well shown in PL XXXIII. fig. 2, which should be compared with the
corresponding figure of Pentacrinus naresianus (PL XXX. fig. 23). The flattened
shape of the lower joints is also well shown in the former figure. Judging from the
torn fragment of the disk which came away from this arm-base, we may suppose that
its anambulacral plating was tolerably well developed. This j)lating extends out on to
the arms, covering in the muscular bundles at the sides of the narrow arm-groove,
though to a less extent than in the four preceding species. The pinnule-ambulacra
(PL XXXIII. fig. 1) are much in the same condition as those of Pentacrinus naresianus
(PL XXVII. fig. 11), the covering plates resting upon the toothed edge of a coijtinuous
calcareous band which is not perfectly differentiated into side plates.

8. Pentacrinus decorus, WyviUe Thomson, 1864 (PL XXXIII. figs. 4-6 ; Pis.
XXXIV.-XXXVIL; PL LVIL figs. 2-5; PI. LVIIl. figs. 1^3; PL LIX. figs. 1-4;
PL LXIL).

1864. Pentacrinus {Neocrinus) decorus, WyviUe Thomson, The Intellectual Observer, August 1864, p. 7.

1864. Peiitacrinus decorus, Liitken, Vidensk. Meddel. f. d. nat. Foren. i Kjfibenhavn, 1864, Nr. 13-16, p. 208.

1865. Pentacrinus {Neocrimis) decorus, Wyville Thomson, Phil. Trans., 1865, vol. civ. p. 542.
1869. Pentacrinus Miilleri, Pourtalfes, Bull. Mus. Comp. Zocil., vol. i.. No. 11, p. 357.

1872. Pentacrimis MiUleri, Wyville Thomson, Proc. Eoy. Soc. Edin., vol. vii. p. 766; and The Depths of the
Sea, p. 442.

REPORT ON THE CRINOIDEA.

331

1878. Pentacrinns MiiUeri, Agassiz, Bull. Mns. Comp. Zool., vol. v., No. 6, p. 56.

1879. Pentacrinus Mulleri, Pourtalis, Bull. Mus. Comp. Zool., vol. v., No. 9, p. 214.
1879. Pentacrinus Mulleri, Agassiz, Bull. Mus. Comp. Zocil., vol. v., No. 14, p. 296.
1882. Pentacrinus decorus, P. H. Carpenter, Bull. Mus. Comp. Zool., vol. .\., No. 4, p. 171.

Dimensions.

Greatest lengtli of stem, to fiftietli nodo,

Diameter of stem.

Longest cirrus (thirty-two joints).

Diameter of calyx, .

Diameter of disk.

Length of arm (one hundred joints).

Length of distichal pinnule (thirteen joints).

Length of pinnule from lower third of- arm (twenty joints),

80 '00 cm.

4-75 mm.
27-00

8-00
12-30
80-00
13 00
16-00

Stem moderately slender aud usually smooth, with a rounded pentagonal outline.
Seven to sixteen (usually eleven or twelve) internodal joints, with much erenulated
edo-es. Nodal joints somewhat enlarged, expanding from above downwards to the top
of the cuTus -sockets. These are deeply hollowed, and have a pear-shaped form, being
continued- downwards as well marked grooves on to the infra-nodiils. Cirri small and
slender, rarely composed of more than thirty joints, the first six of which are Cjuite short
and the remainder squarish or slightly elongated^ with a moderately smooth dorsal edge.
Terminal claw small, with no opposing spine. Lowest limit of the interarticular pores
between the ninth and eleventh nodes.

Basals very variable, sometimes quite small, inconspicuous, and isolated ; but
presenting all gxadations from this condition up to that of large triangular knobs
standing out prominently from the general plane of the calyx, and meeting one another
by their extended lateral angles.

The two outer radials united by bifascial articulation. The rays and their sub-
divisions sometimes separated by plated perisome, and sometimes in close apposition ;
but the sides of the lower arm-joints are scarcely flattened at all. Ten to twenty -five
arms. Primaries of two to seven distichal joints. When three or more distichals are
present the first two are united l>y l)ifascial articulation, th^e second bearing a pinnule,
while the third or sometimes the axillary is a syzygy. But if there be only two
distichals, and the axillary be a syzygy, the first bears a pinnule and is united to the
second by muscles. Lastly, if the axillary be not a syzygy, there is a bifascial articula-
tion between it and the first distichal. Secondary arms (when present) of one to nine
palmar joints, which vary in character as in the primary arms, though to an even
greater extent. If the arms l^ecome free at the radial axillary the first syzygy is usually
on the third brachial ; but if the primary arms divide there is generally a syzj-gj^ in the
second free brachial, though not unfrequently it occurs between the first and second. The

332 THE VOYAGE OF H.M.S. CHALLENGER.

next syzygy is from the eighth to the thirtieth brachial, after which there is an interval of
five to eleven joints between successive syzygies in the free arms. These consist of from
ninety to one hundred slightly overlapping joints, nearly oblong in shape in the lower
parts, but gradually becoming squarer, and finally elongated.

The first pinnule is on the first distichal, if it be united to the second by muscles.
In other cases the second distichal bears the first pinnule ; and the same is the case in
the free arms. The lowest pinnules consist of about a dozen flattened joints, all of
which except the first are longer than broad. They have sharp dorsal edges, and are
much flattened laterally, especially those near the base of the pinnule, the terminal
ones being more slender, so that the pinnule tapers considerably. Farther out on the
arm the joints become more equal, the three lowest being the only ones that are much
flattened, but the pinnules do not increase greatly in length.

The disk bears numerous plates of variable size, rather closely set, but not forming
a perfectly continuous pavement ; and several of them support small blunt spinelets.
The ambulacra of the disk and arm-bases are protected by somewhat narrow and spine-
like plates irregularly disposed. Arm-groove moderately wide and continuously plated.
Pinnule-ambulacra variable, sometimes having moderately distinct side plates and some-
times hardly any trace of them.

Colour when fresh, white, purple, or yellow (Sigsbee), bleaching nearly white in spirit.

Localities. — 'Abundant in the Gulf Stream and in the Caribbean Sea, but not known
elsewhere. A few specimens have been obtained by the collectors for Sir Eawson Eawson
and Mr. Damon ; also by W. Stimpson ofl" Cuba, and by Captain E. Cole of the telegraph
steamer " Investigator " at the following localities — south side of Porto Rico, 667 fathoms ;
ofi^ Saba Island, 320 fathoms (with Pentacrinus asterius) ; between Saba and Eustatius
Islands, 531 fathoms (with. Pentacrinus miilleri). But great numbers have been dredged by
the surveying ships of the U.S. Coast Survey, as recorded in the following list : —

S.S. "Corwin," 1867. 1-6 miles from Chorrera, Cuba, 270 fathoms.

S.S. "Bibb," 1869. Off Double-headed Shot Keys ; 315 and 471 (?) fathoms.

S.S. " Blake," Cruise of 1877-78. No. 21, off Bahia Honda; lat. 23° 2' N., long. 83° 13' W. ; 287 fathoms. Off

Havana (Sigsbee); 150 to 200 fathoms. No. 56, off Havana; lat. 22° 9' 15" N., long. 82° 21' W.;

175 fathoms. No. 57, off Havana; lat. 22° 9' K, long. 82° 21' W.; 177 fathoms.
Cruise of 1878-79. No. 100, off Morro Light; 250 to 400 fathoms. No. 101, off Morro Light; 175 to 250

fathoms. Nos. 155 and 156, off Montserrat; 88 fathoms; bottom temperature, 69° F. No. 157, off

Montserrat ; 120 fathoms; bottom temperature, 69° F. No. 233, off St. Vincent; 174 fathoms; rooky

bottom; bottom temperature, 49J°F. No. 269, off St. Vincent; 124 fathoms; bottom temperature, 5 7 J" F.

No. 296, off Barbados; 84 fathoms; bottom temperature, 61^° F. No. 298, off Barbados ; 120 fathoms;

bottom temperature, 61° F.
Cruise of 1879-80 (Bartlett). Off Santiago de Cuba. Off Kingston (Jamaica); 100 fathoms.

Captain Sigsbee ^ describes the specimens dredged by him in the following terms : —
" Some of them came up on the tangles, some on the dredge. They were as brittle as

1 Bull. Mus. Comp. Zobl, vol. v.. No. 6, p. 60.

o

REPORT ON THE CRINOIDEA. 333

glass. The heads soon curled over and showed a decided disposition to drop off. At
a haul made soon after we got more, and being afraid to })ut so many of them in the
tank together I tried to delude the animals into the idea that they were in their native
temperature by putting them into ice-water. This worked well, although some of them
became exasperated, and shed some of their- arms. They lived in the ice-water two
hours, until I transferred them to the tank. They moved their arms one at a time.
Some of the lilies were white, some purple, some yellow ; the latter was the colour of
the smaller and more delicate ones." Mr. Alexander Agassiz^ records that " our collection
of Pentacrini is quite extensive ; we found them at Montserrat, St. Vincent, Grenada,
Guadeloupe, and Barbados, in several places, in such numbers that on one occasion we
T)rouo-ht up no less than one hundred and twenty -four at a single haul of the bar and
the tangles. We must, of course, have swept over actual forests of Pentacrini crowded
too-ether, much as we find the fossil Pentacrini on slabs. I have nothing to add to the
o-eneral description of then- movements given by Captain Sigsbee, with the exception of
their use of the cirri placed along the stem. These they move more rapidly than the
arms, and use them as hooks to catcli hold of neighbouring objects, and, on account of
their sharp extremities, they are well adapted to retain their hold. The stem itself
passes slowly from a rigid vertical attitude to a curved or even drooping position."

Although the dredgings of the " Blake " have shown that Pentacrinus decorus is
extremely abundant in the neighbourhood of several of the West Indian Islands, it does
not appear to have been discovered till a century after Pentacrinus asterius. Its
distinctness from that type was first recognised by Mr. Damon of Weymouth, who
procured an example of it from the seas of the Outer Antilles. Its occurrence was
recorded liy the late Sir AVj^ille Thomson in a popular article on Sea Lilies, which
appeared in the Intellectual Observer for August 1864, Imt he published no further
description of it before his death. AVhen he first noted its discovery he seems to have
been unacquainted with the description of Pentacrinus milUeri by Oersted, published
six years previously; for he spoke of Pentacrinus asterius and Pentacrinus decorus as
the only two known living species of Stalked Crinoids. But in the following year^ he
referred to Pentacrinus mulleri as well, Liitken's Memoii- having appeared in the
interval ; so that he evidently regarded Pentacrinus decorus and Pentacrinus miilleri
as distinct species.

Later on, however, as I have described above, he came to the conclusion that his
Pentacrinus decorus was identical with Oersted's type,^ and he seems to have held this
view till his death. For he wrote " Pentacmius mulleri, Oersted," on a copy of
PI. XXXIV. This represents a specimen which he had obtained from Sir- Eawson Rawson,
and it is totally different from Pentacrinus miilleri, as is evident from a glance at Liitken's

1 Bull. Mus. Comp. Zobl, vol. v., No. 14, p. 296. - Phil. Trans., vol. civ., 18G5, p. 542.

3 Proc. Roy. Soc. Edin., vol. vii. p. 76(5 ; ami The Depths of the Sea, p. 442.

334 THE VOYAGE OF H.M.S. CHALLENGER.

figure of that type. But it corresponds in every respect with the two individuals in the
national collection which Sir Wyville himself described in 1864 as Pentacrinus
[Neocrinus) decorvs. A large number of examples, obviously of the same iyjie, were
dredged by the " Blake " ; and I have no hesitation in considering Pentacrinus decorus
as a good species ; though for reasons given above I do not regard it as a type of
subgeneric value. In fact Sir Wyville himself seems to have recognised this sub-
sequently ; for while still confounding Pentacrinus decorus with Pentacrinus millleri,
he dropped the names Cenocrinus and Neocrinus altogether, and simply spoke of
Pentacrinus asterius and Pentacrinus millleri.

Pentacrinus decorus differs from Pentacrinus hlakei and Pentctcrinus naresianus
in the flatness of the syzygial faces on the arm-joints (PL XXXVII. figs. 3, 4), both
these species having strongly angular syzygial faces (PL XXXa. figs. 9, 10 ; PL XXXII.
figs. 4, 5, 7, 9, 12, 14). Pentacrinus naresianus has only ten arms, while the primary
arms of Pentacrinus decorus, like those of Pentacrinus hlahei, may divide once or
twice. The second di^^sion is, however, more common in Pentacrinus decorus than in
Pentacrinus blaJcei, in which palmar series are rare (PL XXXI.) ; though distichals
generally occur wdth considerable regularity aU round the cup, which is by no means
always the case in Pentacrinus decorus (Pis. XXXV., XXXVI. ). The general
characters of the pimiules and of their ambvdacral plating are much the same in the
two species ; but the two sets of ambulacral plates are on the whole much better
difterentiated in Pentacrinus hlakei than in Pentacrinus decorus (PL XXXIII. fig. 1 ;
PL XXXVII. figs. 23, 24). In the latter species (PL XXXIII. fig. 4) the arm-groove
itself is more completely covered in by the bases of the pinnule-ambulacra, which over-
lap one another alternately from opposite sides much more perfectly than in Pentacrinus
hlakei (PL XXXIII. fig. 3). But the perisome covering the muscular bundles in the
intervals between the ventral edges of the arm -joints is not plated in Pentacrinus
decorus (PL XXXIII. figs. 4, 6) as it is in Pentacrinus hlakei (fig. 3), and also in some
other Pentacrini previously described, together with some species of Metacrinus. But
the chief and most obvious difference between Pentacrinus decorus and Pentacrinus
hlakei, apart from the peculiarities of the brachial syzygies in the latter species, lies in
the characters of the stem. The internodes in most stems of Pentacrinus decorus are
considerably longer than those of Pentacrinus hlakei, as is evident upon comparison of
Pis. XXXIV. and XXXVII. with PL XXXI. ; and the nodal joints are luarkedly difi'erent
in the two species. Those of Pentacrinus decorus are considerably enlarged above the
deeply hollowed cirrus-sockets (PL XXXVI. ), so that the outline of the stem is not
uniform as it is in Pentacrinus hlakei (PL XXXI. fig. 3); while the contour of the
nodal joints as seen from beneath is less rounded in Pentacrimis decorus (PL XXXVII.
fig. 21) than in Pentacrimis hlakei (PL XXXII. fig. 1). The general aj^pearauce of
the infra-nodal joints (PL XXXII. fig. 2 ; PL XXXVII. fig. 19) and also of the ordinary

REPORT ON THE CRINOIDEA. 335

joint-faces (PI. XXXII. fig. 3 ; PI. XXXYII. fig. 22) is, however, very much the same
ill the two species.

The minor characters of Pentacrinus decorus, i.e., those wliich ai-e of least import-
ance for systematic purposes, present a very remarkable amount of variation. The
number of internodal joints may vary almost as much as in Pentacrinus asterius,
some of the indi\'iduals having the internodes as short as those of Pentacrinus hlahei
(seven joints), while in others they may consist of sixteen joints as in Penta-
crinus asterius ; and this character sometimes runs through the whole stem, so that
at first sight two individuals will look as if they belonged to entirely distinct species,
especially if the development of the basals and arm-divisions be also different in the
two eases.

The internodal joints are generally quite smooth externally ; but they occasionally
bear groups of iiiterradial tubercles at more or less regular intervals, and these tubercles
sometimes appear on the nodal joints, thus increasing the prominence of their angles
between the cii'rus-sockets (PI. XXXV. fig. 1 ; PL XXXVI. figs. 1, 2). Two individuals
are remarkable for the absence of some of the cirri on the stem. Thus in a specimen repre-
sented in PL XXXVI. fig. 1, one of the cirri is missing at the fourth node, no socket
having been developed at all ; while in another shown in PL XXXVII. fig. 2, there are
no cirrus-sockets along one face of the stem to as far down as the twelfth node ; and at
the eleventh node another socket is absent, so that there are only three cirri at this
node, the empty faces of the stem intervening between the cirrus -bearing ones exactly
as they do in those nodal joints of Pentacrinus alternicirrus which bear three cirri
(PL XXV.; PL XXVI. figs. 13, 14; PL XXVII. fig. 2).

The stem of Pentacrinus decorus^ though more slender than that of Pentacrinus
asterius, seems like it to grow to a considerable length (compare Pis. XL and XXXIV.).
The longest which I have seen, consisting of fifty internodes, measures 80 cm. Sii" WyviUe
Thomson mentioned one which was about two feet in length ;^ and this seems to have
been the original of a drawing which was made for him by Mr. AVild. He spoke of the
final joint, which is the epizygal at about the forty-second node, as being woni and
rounded ; and having subsequently found several other examples in the same condition,
he expressed his belief that disengagement at a syzygy is habitual. This is doubtless
often the case as in Pentacrinus wyville-thomsoni and other species [ante, pp. 18—20),
though I have not myself met with any specimens in this condition. JMoreover, it
appears certain that this species may be sometimes permanently fixed. Captain Cole's
observation that they may be attached to telegraph cables by the basal extremity of the
stem spreading slightly has been noticed already ; and the individual mentioned above
as having a stem 80 cm. long (which is now in the British Museum) was found by him
attached in a slightly different way. The stem is detached from its basal portion at the

1 Sea Lilies, The Intellectual Observer, August 1S64, p. 7.

336 THE VOYAGE OF H.M.S. CHALLENGER.

syzygy below the fiftieth node ; but it lies along the cable by the next eleven internodes,
and is more or less attached to it by means of the cirri.

The variation in the development of the basals in Pentacrinus decorus is most
extraordinary, and shows what little reliance is to be placed on the characters of this part
of the calyx for systematic purposes. In young specimens they are generally quite small
and inconspicuous, appearing only as little triangular knobs or rhomboidal pieces which
are sometimes just in contact by tlieir lateral angles (PI. XXXV. fig. 1). But they do
not stand out at all from the general plane of the calyx, and simjily appear as continua-
tions of the interradial ridges of the top of the stem upon which they rest. In older
and even in adult individuals they may retain this condition (PI. XXXIV. figs. 1,8;
PL XXXVI. fig. 3) ; while in others which are still premature they are a little more pro-
minent (PI. XXXV. fig. 2) ; and when a large series of specimens is examined all stages
are traceable between this condition and that of the individuals represented in PI. XXXVI.
fig. 1, and PI. XXXVII. figs. 1, 2. These have large rhomboidal basals, meeting one
another laterally in the re-entering angles of the calyx and standing out as prominent
knobs which sometimes extend beyond the interradial ridges of the stem.-' A somewhat
similar though less extensive variation in the development of the basals occurs in the
fossil Antedon scrohiculata from the Brown Jura.

The ray-divisions of Pentacrinus decorus are exceedingly variable. I have met
with one case in which the number of three radials, so constant throughout the majority
of the Neocrinoids, is consideral )ly exceeded. The radial axillary is primitively the
seventh joint aliove the basal ring, i.e., it is itself a syzygial joint and there are five
below it, inchuling the primary or first radials. The second and third of these five
joints are united by bifascial articulation, just like the ordinary second radials and
axillaries of this species ; and in this respect therefore this aberrant form difi"ers from
the type represented by Metacrinus, in which the second and third joints of the ray are
always united by syzygy, whether the axiUary be the fifth or eighth in the series of
primitive radials (PI. XXXIX. fig. 1 ; PL XLVL).

One young specimen of Pentacrinus decorus which was brought by Sir Rawson
Rawson from Barbados has only ten arms, as is the case in Pentacrinus naresianus ;
while in another dredged by the " Blake " there are only eleven, one of the primary arms
dividing on the fifth joint (PL XXXV. fig. 1). A thii'd individual (PL XXXV. fig. 2)
has two arms on each of four rays ; but on the fifth ray one of the primary arms divides
and one of the secondary arms so formed divides again, so that there are twelve in all.
Other examples again may have distichal series developed with considerable regularity
all round the calyx, and occasionally also one or more palmar series as weU (PL XXXVI. ;
PL XXXVII. fig. 2). But the total number of arms rarely exceeds twenty -five in all ;

' In the specimen represented in PI. XXXVI. some of the first radials are considerably distorted, apparently as
the result of injury and subsequent regeneration.

REPORT ON THE CRINOIDEA.

337

though Sir Wy\'ille Thomson described a specimen with thirty arms.^ The irregularity
in the number of joints between successive axillaries is very striking, especially as
compared with the very constant character of the ray-divisions in the Comatulse.
Taking for example the genus Actinometra, we find in Pentacrinus decorics the follow-
ing series of distichals and palmars which are specially characteristic of different groups
of the species of that genus.

A. Ten arms only ; the first two brachials united bifascially, and

criiius narcsiamis,

B. Two distichals united by a syzygy, .

C. Two distichals, the axillary not a syzygy,

D. Three distichals, the axillary a syzygy,

E. Two palmars united by a syzygy,

F. Two palmars, the axillary not a syzygy,

G. Two palmars, the axillary a syzygy, .
H. Three palmars, the axillary a syzygy.

the third brachial a syzygy as in Fenta-
Group of Actinometra meridionalis.
„ Actinometra JiJcesi.
„ Actinometra pulchella.
„ Actinometra paruieirra.
„ Actinometra tijpica.
„ Actinometra stelligera.
,, Actinometra multiradiata.
,, Actinometra hennetti.

In addition to these, Pentacrinus decorus may show numerous combinations of distichal
and palmar series such as are characteristic of other groups of Comatulse, and also certain
conditions such as two distichals with the axillary a syzygy, which I have not as yet
met with in any Comatula at all. These facts well illustrate what has been said above
{ante, p. 55) respecting the difference between the arms of Pentacrinidee and Comatulse.

The rays of Pentacrinus decorum and their subdivisions are sometimes in pretty close
contact, though rarely flattened laterally ; while in other cases they are more or less
separated by perisome (Pis. XXXV. -XXXVII.). This perisome is sometimes nearly
bare and sometimes plated pretty continuously ; and a similar variation is apparent on the
upper surface of the disk. This is sometimes covered tolerably closely by rather large
plates (PI. XXXIV. fig. 2) ; but the plating, is not cpite so continuous as in Pentacrinus
wyville-thomsoni and Pentacrinus alternicirrus (PI. XVII. fig. 6 ; PL XXVI. figs. 1, 2).
On the other hand, the gaps between the plates, though sometimes comparatively large,
are not so extensive as in Pentacrinus miilleri (PI. XVII. fig. 10). The plates sometimes
bear small blunt spinelets which are possibly tactile in function, as they contain branches
from the antiambulacral nerves which extend upwards on to the disk fi'om the envelope
of the chambered organ in the calyx (PI. LIX. figs. 2-4, ad). The plates bordering the
ambulacra of the disk are narrow and spine-like, often forming a kind of palisade, which
is more distinct than in any other type of the Pentacrinidaj (PI. XXXIV. fig. 2).
They become somewhat irregular on the arm-bases (PI. XXXIII. fig. 6) ; but further
out (PI. XXXIII. fig. 4) they begin to show signs of a more or less perfect
diflerentiation into the side and covering plates of the pinnule-ambulacra (PI. XXXVII.
figs. 23, 24).

■ Sea Lilies, Tlie Intellectual Observer, August 1864, p.

(ZOOL. CHALL. EXP. — PART XXXII. 1884.)

114.3

338 THE VOYAGE OF H.M.S. CHALLENGER.

9. Pentacrimis mollis, n. sp. (PL XXXIII. figs. 7-10).

Dimensions.

Greatest height of specimen (fig. 9), . . , . . . lO'OO mm.

Least height (fig. 10), ....... 8-00 „

Diameter of calyx across the first radials, ..... 6'50 „

Diameter of disk, ........ 4r'75 ,,

This is a very singular type, and it is witli much hesitation that I have referred it to
Pentacrinus ; for it differs considerably from this genus in the relative proportions of the
lower parts of the calyx, and its stem is entii-ely unknown. But on the other hand it pre-
sents no character which seems to me of sufficient importance to justify a generic separation.

Its most obvious peculiarity is expressed in the specific name which I have applied to
it. The cup, radials, and lower arm-joints, which alone remain, are not formed of the
usual hard limestone found in most other Echinoderms. But, externally at any rate,
they are of a somewhat leathery texture, and yield readily to pressure, though there
appears to be a rather harder core within ; while some of the joints show traces of the
usual calcareous skeleton, and the disk is extensively plated (PL XXXIII. fig. 7).

Nothing remains of the stem except three or four thin stellate joints immediately
beneath the calyx. The rays of the star are bent upwards considerably, and enclose in a
sort of cup the lower j)ortion of the basal plates with which they correspond in position
(PL XXXIII. figs. 8-10). There are no traces of broken ligament-fibres on the under
face of the lowest remaining stem-joint, and this would seem therefore to have been
the upper joint of a syzygial pair with its cirri as yet undeveloped.

The basals which form an entirely closed ring are unusually high for a Pentacrinus,
but also of considerable width, so as to support the expanding calyx above them. Their
outline, so far as it can be seen, is irregularly quadi'ate, the two lateral edges approaching
one another somewhat rapidly until they disappear beneath the upturned stem-joints ;
while the upper edges meet at a very open angle. The basi-radial suture therefore has
five well marked depressions in which the radials rest, and five intervening elevations
which mark the middle lines of the basals. The radials are short and widely pentagonal,
sloping upwards and outwards at a considerable angle. They have a somewhat strongly
marked median ridge which starts from near the basi-radial suture, but disappears before
reaching their distal edge, as is well shown in fig. 8 ; and the distal edge itself is more or
less everted, while the dorsal ligament uniting the first and second radials is somewhat
prominent. The remaining radials and the lower arm-joints are only properly visible in
the anterior and left antero-lateral rays as shown in PL XXXIII. fig. 9 ; for the joints
of the other three rays are so shrivelled and contorted as to make their recognition a
matter of considerable difficulty. Their general aspect is somewhat diagrammatically
represented in PL XXXIII. fig. 10.

REPORT ON THE CRINOIDEA. 339

The second and third radials appear to be united by syzygy. The former are slightly
trapezoidal in shape, meeting one another more or less extensively by their lower angles
and then diverging. The axillaries are pentagonal, and scarcely Tvider than the second
radials, so that a considerable gap is left between the rays. This, however, is much wider
on the front of the calj^x than elsewhere, as will be evident from a comparison of
figs. 9 and 10 on PI. XXXIII. It is closed higher up by the approximation of tlie
first brachials of adjacent rays. Where they are properly visible they appear to have
the usual somewhat wedge-shaped form ; and the next joint was perhaps an axillary.
But the condition of the specimen renders the determination of the real nature of the
lower arm-joints entirely uncertain. From the mode of division of the ambulacra of the
disk, however, it would appear that there were twenty arms (PI. XXXIII. fig. 7).

The colour is dirty brown with occasional patches of white, indicating the presence of
calcareous tissue.

ioe«/;iy.— Station 235, June 4, 1875 ; lat. 34° 7' N., long. 138° 0' E. ; 565 fathoms;
green mud ; bottom temperature, 38°'l F. One mutilated specimen.

Three much mutilated individuals of Eudiocrinus japonicus were obtained at the
same Station ; but they exhibit no trace of the extraordinary deficiency of limestone in
the skeleton which disting-uishes Pentacrinus mollis. The heisfht of the basals and the
pecuHar way in which they are received into a sort of cup formed by the uppermost
stem-joints distinguish this type very markedly from all the other Pentacrinidee. For
the stem-joints of this family usually decrease rapidly in size towards the top of the
stem, the upper ones being concealed within the concavity formed by the lower faces of
the basals, as is well show^n in PI. XXXIII. fig. 5. But it is of course possible that this
may also be the case even in the doubtfid Pentacrinus mollis, though on a smaller scale.

Genus Metacrinus, n. gen.

Characters of the Genus.

The petaloid sectors of the faces of the stem-joints are bordered by a few large ridges,
of which the smaller proximal ones meet those of adjacent sectors in the interpetaloid
spaces, while the larger distal ridges reach the outer edge of the joint. The internodes
of six to thirteen joints. Nodals fully occupied by the cirrus-sockets which reach their
upper edge ; supra-nodals incised to receive the bases of the upturned cirri. These are
long, consisting of forty or fifty uniform joints, and vary but little in appearance. The
lower cirri often smaller than those about the twelfth node.

Basals large, rhomboidal, or hexagonal, and in close contact. Then- lower angles are
generally distinctly produced downwards. Four to six radials, the second a syzygy and
bearing a pinnule, as do the remaining ones till just before the palmar axillary. The

340 THE VOYAGE OF H.M.S. CHALLENGEE.

fourth radial is sometimes a syzygy. The rays generally divide four and sometimes five
times ; and their subdivisions are equal in value or nearly so. The first joint after each
axillary bears a pinnule, and the third is usually a syzygy.

The terminal parts of the arms have a lai'ge number of undeveloped pinnules. Those
on the lowest parts of the rays have large, thick joints, the basal ones cuboidal and the
rest oblong. Those borne by the radials and lower distichals receive their ambulacra
direct from the peristome, or from the primary ambulacra of the disk.

Remarks. — The name Metacrinus for this very w^ell characterised genus originated
with Sir Wyville Thomson ; but he drew up no diagnosis of it. Indeed it was only by
finding the name in his handwriting upon a proof of PI. XLVII. that I was informed of his
recognition of the type as distinct from Pentacrimis, under which name it had been
mentioned in the Station Book, and in various publications that dealt with the work of
the Challenger in the Pacific. I had known of the existence of a second livino; eenus
of Pentacrinidse for some months before the Challenger and "Blake" Crinoids came into
my hands. For my friend Mr. Charles Stewart, F.L.S., the present Curator of the
Hunterian Museum at the Eoyal College of Surgeons, had shown me early in 1881 a very
fine dry specimen which had been obtained in the neighbourhood of Singapore by one of
the ships belonging to the Eastern Telegraph Company. It was accompanied by a stem-
fragment of another species which I now know to belong to the same genus. Thanks to
his kindness, I have been able to describe them recently,^ together with yet another
species which had been dredged in the Japanese Seas by Dr. Doderlein, the Conservator
of the Natural History Museum at Strassburg, who courteously placed it in my hands for
this purpose. Eleven species were dredged by the Challenger, and there is another
which I have not examined personally. It was dredged by the famous " Vega " in
Yeddo Bay, during her stay in Japan in October 1879; but no description of it has
yet been published. It is at present in the hands of Prof. S, Loveu of Stockholm ;
and with his characteristic kindness he not only sent me some fragments of the stem,
but also allowed Mr. Percy Sladen, who was examining the Starfish collection in the
Stockholm Museum, to draw up a description of it for me. So carefully was this done,
that I have been enabled to recognise the afijnities of the type, and to assign it a
place in the classification of the genus which appears on p. 344.

Both the calyx and the stem of Metacrinus, but especially the former, present very
striking diS"erences from the corresponding parts in Pentacrinus. In the latter genus
it is quite the exception for the number of radials to exceed three, which is so constant in
Ainocrinus, Millericrinus, Encrinus, and Comatula; and even when there are more,
none of them bear any pinnules. Thus there is no pinnule on the third joint of the
abnormal ray of Pentacrinus mulleri represented on PI. XV. fig. 2, in which the axillary
is the fourth joint above the basals ; nor on any of the five joints below the radial axillary

■ On Three New Species of Metacrinus, Trans. Linn. Soc. Land. (Zool.), ser. 2, vol. ii. pp. 435-447, pis. l.-lii.

REPORT ON THE CRINOIDEA. 341

(itself a syzygy) in a singular specimen of Pentacrinus decoriis, whicli was clredgccl by
the " Blake" (ante, p. 336).

In Metacrinus, however, the number of primitive radials is typically either five or
eight. In both cases the second and third of these eventually become united by syzygy,
and the latter, which corresponds exactly to the third radial in the abnormal Pentacrinus
miilleri already mentioned (PI. XV. fig. 2), bears a pinnule, as does also the next joint,
which is sometimes followed by the axillary (PI. XXXIX. fig. 1 ; PI. XLIII. fig. 2).
When there are eight primitive radials the original fifth and sixth joints become united
by syzygy in the same way ; so that the ray is described as consisting of six joints, of
which the second and fourth are syzygial, both of these as well as the third and fifth
bearing pinnules (PI. XLVI. ; PI. L. fig. 1).

These pinnules also afibrd a very good distinguishing character of Metacrinus. Their
lower and middle joints are very large and massive, having a considerable vertical height
in addition to their thickness from side to side ; while one or two of the basal joints are
nearly cubical in appearance (PL XXXIX. fig. 1 ; PL XLIII. figs. 2, 4; PL XLVL; PL
XLVIII. fig. 1 ; PL L. fig. 1). The later joints, while remaining long and high, gradually
diminish in breadth so that they come to be prismatic, with a sharp dorsal edge. The
distichal and palmar pinnules have the same characters as those on the radials, though
in a less prominent degree.

The shape and large size of their joints are well shown in PL XXXIX. fig. 2, and in PL
XLIII. fig. 2. There is no regularity as to the side of the ray on whicli the first pinnule
occurs ; so that the same interradial space may be occupied by the first pinnules of two
rays as is shown on the right of PL XXXIX. fig 1 ; while on the left of the same figure
the second pinnules of two rays also occur in the same interradius. This not unfrequently
causes a certain amount of crowding ; and it is not uncommon for one of the later radial
or lower distichal pinnules to be very much flattened laterally and almost knife-like in
appearance, while the corresponding pinnules on the other rays of the same individual
are large and massive.

About the level of the distichal axillary, or rather farther out if the number of arms
be large, the stout lower pinnules begin to lose their distinguishing characters and to
pass gradually into the ordinary pinnules of the free arms. The transition is very
difficult to explain without a long series of figures illustrating its successive stages, and it
varies a good deal in difierent types; but it may be generally described as a flattening of
the pinnule-joints (as it were) against the arm and against the succeeding pinnules.
They thus retain a considerable height in a vertical (dorsoventral) direction and have a
sharp dorsal edge, with the exception of the lowest joints which are set in a different
plane from their successors and are often of a much greater width. This is the case, for
example, in the palmar and lower lirachial pinnules of Metacrinus moseleyi as seen on
the left hand ray of the specimen represented in PL XLVI. ; while in Metacrimis

342 THE VOYAGE OF H.M.S. CHALLENGER.

murrayi, Metacrhms costatus, Metacrinus nodosus, and Metacrinus wyvillii the disticlial
pinuules are flattened as well as their successors, and all have wide basal joints, as shown
in PI. XLVIII.

The lower pinnules of Pentacrinus do not present anything like the characters of
those of Metacrinus, the only approach to this type being the relatively large size of
their lower joints in Pentacrinus asterius (PI. XIII. fig. 1) and in some forms of
Pentacrinus millleri (PL XV. fig. 3) ; while the distiehal pinnules never receive their
ambulacra direct from the peristome or from one of the five primary ambulacra of the
disk, as do the radial and lower distiehal pinnules of Metacrinus (PI. XXXIX. fig. 2 ;
PL XLIII. fig. 3 ; PL L. fig. 3). This character i& at once sufiicient to separate the disks
of the two genera ; but it does not produce any important efi'ect upon their external
appearance in the same way as do the large number of radials and the great size of
the pinuules borne by them. There are other very striking features, however, which
considerably aflect the general facies of the species of Metacrinus. There is always a
much greater length of the terminal portions of the arms which have undeveloped pinnules
than in any species of Pentacrinus; and this gives a curious rat-tailed appearance to the
general plume of arms. Compare Pentacrinus (Pis. XL, XV., XVIIL, XIX., XXV.,
XXVIIL, XXXL, XXXIV.), Metacrinus (Pis. XXXVIIL, XL., XLII.^XLIV., XLVIIL,
XLIX., LL, LIL).

An examination of these Plates will also bring out the more striking difi"erences in the
characters of the stems of the two types. The cirri of Pentacrinus vary considerably in
length and in appearance, being long and stout in Pentacrinus asterius and Pentacrinus
miilleri (Pis. XL, XIV.) ; shorter in Pentacrinus ivyville-thomsoni, though stiU stout
(PL XIX.) ; and slender in Pentacrinus naresianus, Pentacrinus decorus, and Penta-
crinus hlahci (Pis. XXVIIL, XXXL, and XXXIV.), the number of joints varying from
twenty to fifty. In Metacrinus, however, the mature cirri nearly always have over forty
joints, which are generally moderately stout ; though of course they are not absolutely so
large in the less robust types like Metactinus costatus (PL XLIX.) and Metacrinus
nodosus (PL L.), as in Metacrinus angidatus (PL XXXVIIL), Metacrinus cingidatus
(PL XL.), and Metacrinus murrayi (PL XLIL). In the three types last mentioned, and
also in Metacrinus inierrujAus (PL LIL), the cirri about the tenth or twelfth node are
larger than those below them ; but in other species this difi'erence is not so manifest.
All the species of the genus, however, have the older cirri more or less directed upwards,
as is especially well shown in Metacrinus angulatus (PL XXXVIIL), Metacrinus wyvillii
(PI. XLVIIL), Metacrinus interruptus (PL LIL), and Metacrinus tuherosus (PL LIII.) ;
though it is less marked in Metacrinus varians (PL XLIV.). This character occurs in
no Pentacrinus excepting Pentacrinus unjville-thomsoni (PL XIX. fig. 1), and is by no
means constant in that type. But the result of it is that the supra-nodal joint takes a
considerable share in the formation of the cirrus-socket, being more or less deeply incised

REPORT ON THE CRINOIDEA. 343

to receive the wide basal joints of the cirri ; so that its re-entering angles are deeper than
those of an ordinary internodal joint (compare PI. XXXIX. figs. 8, 9 ; and PI. LIII.
figs. 2, 4). The infra-nodal joints, however, are less deeply incised than those of Penta-
crinus, and are sometimes not modified at all, as in Metacrinus interruptus and
Metacrmus rotundus. This is the case in Pentacrinus wyville-thomsoni (PI. XIX.
figs. 3, 4), in which the lower edge of the cirrus-socket of the nodal joint projects out-
wards beyond the level of the infra-nodal joint below it, whde the supra-nodal is only
slightly modified to receive the bases of the upward projecting cirri. Traces of a similar
rim at the lower edge of the cirrus-socket are to be seen in Metacrinus interruptus
(PI. LII. fig. 2).

I know of no characters by which isolated internodal joints of Metacrinus can be
distinguished from those of Pentacrinus, the relations of the teeth to the petaloid sectors
being the same in both types. But the number of internodals is much more constant
in Metacrinus than in Pentacrinus. In the latter genus there may be only one or two,
as in Pentacrinus maclearanus (PI. XVI. fig. 1) ; or about sixteen in Pentao-inus
asterius (PI. XL), and about forty in Pentacrinus wyville-thomsoni; whereas in Meta-
crinus the Kmits of variation are from six or seven in Metacrinus ivyvillii (PI. XLVIII.)
to thirteen in Metacrinus murrayi (PL XLII.) ; so that this character is of less value in
the classification of the species than it is in Pentacrinus.

The geographical range of Metacrinus is much more limited than that of Pentacrinus.
Neither genus has been found in the Pacific east of long. 170° W.; but there are no
Atlantic species of Metacrinus. Pentacrinus was obtained without Metacrinus at
Stations 171 and 175 in the South Pacific ; while Metacrinus occurred alone at Station
209 (among the Philippines), and also in considerable abundance and variety at Station
192, in the Ai-afura Sea. It extends as far west as Singapore, and is represented by two
species in Japanese waters, where no Pentacrinus is known, excelling the doubtful
Pentacrinus mollis (PI. XXXIII. figs. 7-10). On the other hand, Metacrinus has not
been met with below 630 fathoms, while Pentacrinus has been dredged at 1095 fathoms
in the Atlantic, and 1375 fathoms in the Pacific Oceans.

The mutual relations of the thirteen species of Metacrinus in which the calyx is
known, are shown in the following Table. The Challenger species are numbered 1 to
11, and the letters P. H. C. are appended to those which I have recently described
in the Transactions of the Linnean Society. The " Vega " specimen, which concludes the
list, has not yet been described. There are two other species of the genus (one obtained
by the Challenger) which are known as yet only by the characters of their stems ; and I
have therefore been obliged to place them in a separate list, as the number of radials in
the cup is still an unknown quantity.

344

THE VOYAGE OF H.M.S. CHALLENGER.

Genus Metacrinus (Wyville Tliomsou, MS.), P. H. Carpenter, 1884.

Fourradials;
tlie second
a syzygy.

Eight to twelve inter-
nodal joints witli hori-
zontal ridses.

Ten to
thir-
teen
inter-
nodal
jO'ints.

Lower parts of

rays smooth.
Forty-five cu'-
rus joints.

Ridges incomplete,

Ridges continuous.

Thirteen internodal joints with in-
curved sides. Interarticular
pores to thirteenth node.

Ten or twelve internodal joints V
with iiat sides. Interarticular >
pores to tenth node. j

Lower parts of rays have raised distal edges.
Sixty-five cirrus-joints. Interarticular pores to
eighteenth node.

Six to ten internodal joints with nearly flat sides,
articular pores to about the sixth node.

Inter-

1. angulatus, n. sp.

2. cingulatus, n. sp.

3. murrayi, n. sp.

4. nobilis, n. sp.

swperhus, P. H. C.

5. varians, n. sp.

Eadials irre-
gular; usu-
aUy five
or six ; the
second and
also the
fourth
often a

syzygy.

Six radials ;
the second
and fourth
syzygies.

Seven to nine internodal joints; the larger ones with a strong
median girdle.

Eleven or twelve smooth internodal joints,

' Three ray -divisions. Cirrus-facets as high as
the nodal joints. Five to eight internodals with
horizontal ridges.

Six to
nine
inter- -i
nodal
joints.

Four ray-divi-
sions. Nodal
joints higher -
than cirrus-
facets.

Stem with strong interradialiidges \
and nearly smooth, slightly (_
hollowed sides. Seven or eight i
internodals. )

I

Eight or nine internodals, ■n'ith ^
flat tubercular sides. Angles of
nodals enlarged.

I

6. moseleyi, n. sp.
rotundas, P. H. C.

7. iDyviUii, n. sp.

8. costatus, n. sp.

9. nodosus, n. sp.

Tenor

eleven
inter-
nodal

joints.

Stem pentagonal, with a tubercle at each angle \

and ridges on the sides, but not forming a \ 10. intevnqjtiis, n. sp.
complete girdle. Usually eight distichals. j

Rounded stem without ridges. Usually ten dis-
tichals.

The "Vega" specimen.

Seven internodals with median tubercles; Infra-nodals not
incised.

Angles of
stem sharp
and pro- \

ducedout- ! Eleven internodals with shght horizontal ridges
wards. [ nodals not incised.

}

\ 11. tuberosus, n. sp.

Infra-

:'a- (

stewarti, P. H. C.

38-50

cm.

5-75

mm

53-00

8-25

20-00

90-00

25-00

10-00

REPORT ON THE CRINOIDEA. 345

1. Metacrinus angulatus, u. sp. (Pis. XXXVIII., XXXIX.).

Dimensions.

Greatest length of stem, terminating at the tlurt3'-fifth node,
Diameter of stem, ......

Longest cirrus (fifty-seven joints), ....

Diameter of calyx, ......

Diameter of disk, ......

Length of arm (one hunJred and ten joints),

Length of large distichal pinnule (twenty joints),

Length of first pinnule after tertiary axillary (twelve joints),

Stem robiLst, with a sharply pentagonal outline. Eight to twelve internodal joints
with but slightly crenulated edges. Their faces are distinctly stellate, while the ano-les
are sharpened and slightly produced outwards ; and their sides bear strong horizontal
ridges, the edges of which are coarsely denticulate. Nodal joints more distinctly stellate
in outline and without horizontal ridges. The cirrus-facets reach closer to their upper
than to their lower edges, while the supra-nodal joints are slightly incised to receive the
bases of the large cirri, which are mostly directed upwards, and are longest at the twelfth
node. They usually consist of forty-five to fifty -five stout joints, which are very uniform
in appearance except at the two ends. The lower joints are c^uite short, and the tlistal
edges are longer than the proximal ones, so that the joints seem to have slight dorsal
projections at their distal ends. The interarticular pores terminate between the tenth
and twelfth nodes.

Basals widely hexagonal, united into a complete ring, and rather prominent on the
exterior of the calyx. They are much higher in the middle line than at the lateral edges,
and the lower angle is often produced considerably downwards over the upper stem-
joints. Four radials, the first comparatively short and the second a syzygy with a
pinnule on the epizygal, the third also bearing a pinnule and the last axillary. Rarely
more than three divisions of the rays, which arc in close apposition, being only separated
by the bases of the large lower pinnules. Forty to fifty arms, generally consisting of about
one hundred .slightl)^ overlapping joints beyond the last axillary. Primary arms of six
to ten distichals (usually eight or nine), of which the second or third is a syzygy.
Secondary arms of ten to fourteen palmars, sometimes as many as twenty, the third of
which is usually a syzygy. Tertiary arms rare ; consisting when present of fourteen to
thirty joints, the third of which is a syzygy. There is generally a syzygy in the third
brachial of the free arm. Another somewhere between the tenth and thutieth brachials,
and others at intervals of six to twelve joints.

A pinnule on the epizygal of the second radial, and on the first joint after each
axillary. The first pinnules, as far as the beginning of the secondary arms, are much
larger than their successors, consisting of about twenty joints, the lowest of which are

(ZOOL. CH.VLL. EXP. PART XXXII. — lSS-1.) K -44

346 THE VOYAGE OF H.M.S. CHALLENGER.

very massive, and generally somewhat cuboidal ; while the next few are narrower with
flattened sides, but still of great thickness in a dorsoventral direction. The thickness
gradually diminishes, and the outer part of the pinnule consists of moderately long,
somewhat flattened joints, with the dorsal edges sharpened and projecting slightly forward
over the bases of their successors. The distichal pinnules on the outer sides of the ray
are longer and have somewhat larger joints than those borne by the radials. Beyond the
distichal axillary, the size of the pinnules gradually decreases, the lower joints becoming
at first prismatic and then flattened, but remaining distinctly larger than their successors
for some little distance beyond the palmar axillaries. The later pinnules are short and
.styhform.

The disk bears numerous small scaly plates, which are more thickly grouped on the
anal tube than elsewhere. Disk-ambulacra strongly but irregularly plated ; those of the
arms distinctly above the arm-groove, and supported by regular bifid plates which become
diflerentiated on the pinnules into covering plates and ill defined side plates.

Colour when fresh — the stems almost white, and the crowns light yellow or light
reddish-orange (Moseley) ; in spirit, white or whitish-brown.

Locality. — Station 192, September 26, 1874 ; in the iVrafura Sea, off' the Ki Islands ;
lat. 5° 49' S., long. 132° 14' E.; 140 fathoms; blue mud. Seven specimens, and possibly
more.

Remarks. — This species is readily distinguished from its nearest ally (Metacrinus
cingulatus) by the characters of its stem-joints (PL XXXIX. figs. 3-11). They are
much more sharply stellate than in that t^-pe (PL XLL figs. 1-3), having deejier
re-entering angles ; while the horizontal ridges on the sides of the internodal joints are
generally not continuous, but interrupted at the angles, which are somewhat produced
outwards (PL XXXIX. fig. 3). One specimen presents a curious variation in this
respect. The horizontal ridges on the thicker joints are enlarged so as to have a some-
what diamond shaped aspect, with more or less produced lateral angles (PL XXXIX.
fig. 11) ; and when this ridge is large it shows itself very plainly in a terminal view of
the joint-face, outside the line of teeth (compare PL XXXIX. figs. 8 and 11). In this
specimen too the downward extension of the basals over the upper stem-joints is
especially well marked, and the supra-nodal joint is rather more hollowed to receive
the cirrus-bases than it is in the type. The stems of five specimens aU terminate below
in a nodal joint. In two cases there appears to have been an attached portion of stem
beneath ; for the surface of this lowest nodal joint is comparatively fresh and its central
canal visible ; but in the other three stems this surface is somewhat worn, and I cannot
make out the opening of the central canal, which appears to have been closed up, the
animal living in a semi-free condition like Pentacrinus ivyville-thomsoni, Pentacriiius
maclearanus, or Pentacrinus alternicirrus. The respective lengths of these stems are
as follows : — (l) 3 8 "5 em. long, closed at the thirty-fifth node ; (2) 23 "5 cm. long, closed

REPORT ON THE CRINOIDEA.

347

at the twenty-fourth node; (3) 22-5 cm. long, closed at the twenty-fifth node. The
individuals of this species present various forms of monstrosity, some of which have
already been mentioned : — (I) The presence of six joints in one ray, of which the second
and the axillary are syzygies, thus approaching the other group of Metacrinvs species.
(2) The presence of eleven radials on one ray, which recalls the irregularities of the
Palaeocrinoids. (3) The eighth joint of one of the primary arms is not a distichal axillary
as usual, but it is swollen and bears a slightly larger pinnule than the preceding joint,
so that it resembles an axillary with unequal faces. Fourteen joints further on there is
an axillary which corresponds to that on the secondary arms of other rays. (4) The
presence of a second smaller aual tube at the side of the ordinary one (PI. XXXIX.
fig. 2), an individual of Myzostoma u'l/ville-thomsoni, von Grafi", having attached itself
between the two. (5) The two arm-trunks borne by one of the palmar axillaries each have
a syzygy in the third brachial, and then unite again into a single trunk. The fourth
joint of the right-hand arm has the shape of an axillary reversed, i.e., it has two proximal
faces, one of which meets the epizygal of the third brachial, and the other the fourth
brachial of the left-hand arm. It bears a pinnule on its left side, and as it is the fourth
joint of one arm-trunk, and the fifth of the other, the ordinary sequence of the pinnules
is uninterrupted, and continues normal throughout the rest of the arm which is borne by
this reversed axillary.

2. Metacrmus'cingulatus, n. sp. (PI. XL., PL XLI. figs. 1-4).

Dimensions.

Length of stem to thirty-third node,

Diameter of stem,

Longest cirrus (fifty joints), .

Diameter of calyx, .

Diameter of disk.

Length of arm (one hundred joints),

Length of large distichal pinnule (twenty-five joints).

Length of first pinnule after tertiary axillary (fifteen joints).

35-00 cm.

5-50 mm.

51-00 „

8-00 „

18-00 „

135-00 „

28-00 „

10-00 „

Stem robust, with a sharply pentagonal outline. Six to fifteen (usually eight to
twelve) internodal joints, with moderately crenulated edges. Their faces are bluntly
stellate, with shallow re-entering angles ; while their sides bear strong horizontal ridges
which are slightly more prominent at tlie angles of the stem, and are often marked by
irregular indentations. The nodal joints are more distinctly stellate, and have no ridges
except at the angles and where the cirrus-facets are absent. These reach closer to th-.'
lower than to the upper edges of the nodal joints, but the incision of the supra-nodals
is fau:ly distinct. The cirri have about forty to fifty tolerably uniform joints, and arc

348 THE VOYAGE OF H.M.S. CHALLENGER.

lono-est between the eleventh and twelfth nodes. The interarticular pores end between
the eiarhth and tenth nodes.

Basals prominent, more or less extended downwards. Radials four, the second a
syzygy. ' Earely more than four divisions of the rays, giving forty or fifty arms, which
consist of about one hundred joints beyond the last axillary, the basal ones slightly over-
lapping. Primary arms of six to eight, or sometimes ten distichals, of w^hich the second
or third is a syzygy. Secondaries of ten to fifteen (usually twelve to fourteen) palmars,
the third generally a syzygy. Tertiary arms rare, consisting of fourteen to twenty-six
(usually eighteen to twenty) joints, the third of which is a syzygy. In rare cases there
is another axillary after twenty joints more. There is generally a syzygy in the third
brachial of the free arm, another betw^een the seventh and twenty-sixth, and others at
intervals of six to eighteen joints. The distichal pinnules have large outer joints, and
are therefore larger on the whole than those on the radials, though the basal joints are
generally less massive than in the radial pinnules. All the pinnules, and especially the
lower ones, have a serrate dorsal edge. The disk bears numerous small plates, which are
not very closely set, except in the anal interradius. The ambulacra of the disk and arm-
bases are supported by irregular elongated plates, the latter being distinctly above the
arm-groove, with a few ambulacral plates at their sides. The brachial ambulacra pro-
tected by smaller bifid plates, which become differentiated on the pinnules into covering
and side plates.

Colour when fresh — the stems almost white, and the crowns light yellow or light
reddish-orange (Moseley) ; in spirit, white.

Localitij. — Station 192, September 26, 1874 ; in the Arafura Sea, ofi" the Ki Islands;
lat. 5° 49' S., long. 132° 14' E. ; 140 fathoms ; blue mud. Two specimens.

RemarJcs. — This species is at first sight not unlike Metacrinus angidatus (PI.
XXXVIIL), having about the same number of internodal joints in the stem, and a nearly
identical arrangement of the arm-divisions. The stem-joints, how-ever, are very diff"erent
in the two types. The horizontal ridges, which are interrupted at the angles of the stem
in Metacrinus angidatus (PI. XXXIX. figs. 3, 11), are usually continued right round the
joints in Metacrimis cingidatus (PI. XLI. figs. 1,3); and they appear also as enlargements
of the angles of the nodal joints (PL XLI. fig. 2), which are much less sharp than in
Metacrinus angidatus (PI. XXXIX. fig. 4). As a rule too there are generally slightly
fewer joints between the successive axillaries of the dividing rays than in the latter species,
but the character of the arms and of the pinnules which they bear is very much the same
in both.

The two specimens of Metacrinus cingulatus which were obtained by the Challenger
difier somewhat in their characters, and each exhibits a certain amount of variation.
In the smaller individual there is an irregularity in one of the rays. The fourth or
axillary radial is not articulated to the preceding joint, as is usually the case, but the two

REPORT ON THE CRINOIDEA.

349

are united by syzygy. The third radial therefore, being a hypozygal, bears no pinnule.
But the normal number of pinnules, one on each side of the ray, is still maintained ; for
the second radial, itself a syzygial joint, bears a pinnule on each side.

The stem of this specimen, which is broken at the bottom, tapers slightly downwards,
and has tolerably regular internodes, with five cirri at each node in the usual way.
The stem of the other individual, however, does not decrease in size at its lower end,
which terminates in a nodal joint with a freshly exposed under surface. The inter-
nodes of this stem are very irregular in length, and generally consist of one or two joints
more than in the smaller specimen, while the development of cirri at the nodes is
extremely irregular. No less than ten of the thirty-three remaining nodes have one or
more cirri deficient, as shown in the following Table : —

Node.
5th

Missing
2

Cirri.

Node.
20th

Mi.ssing Cirri
2

8th

1

24th

2

9th

3

27th

4

10th

1

30th

4

13th

1

33rd

■2

In the lower part of the stem there is no trace of the absent sockets, which are replaced
by ridges like those on the internodal joints ; but in the middle and upper parts the
socket is present, though imperfectly developed, especially at the fifth node. At the
thirteenth, the place of the socket is entirely taken up by a group of sessile Cirripedes
( Verruca).

3. Aletacrinus murrayi, n. sp. (PI. XLI. figs. 12-17 ; PI. XLIL).

Dimensions.

Length of stem to eighteenth node, ..... 184'00 mm.

Diameter of stem, . ....... 6'75 ,,

Longest cirrus (forty-eight joints), ...... 43-00 ,,

Diameter of oalyx, . . . . . . . . 1 1 '00 „

Diameter of disk, . . . . . . . . 21-00 „

Length of arm (one hundred and fifteen joints above tertiary axillary), . 100-00 „

Length of pinnule on second radial (fifteen joints), .... 18-00 „

Length of first palmar pinmile (seventeen joints), . . . .23-00 „

Length of first pinnule after tertiary axillary (eighteen joints), . . 14-00 „

Description of an Individual. — Stem robust and of a rounded pentagonal form, wdth
very shallow re-entering angles. Internodes composed of thirteen joints, wdth smooth
sides and faintly crenulated edges. The supra-nodal joints are slightly incised, but the

350 THE VOYAGE OF H.M.S. CHALLENGER.

articular surfaces of the wide and deep cirrus-sockets are limited to tlie nodal joints,
which are markedly stellate in form, though their angles are not produced outwards.
The infra-nodals are also deeply notched by the downward extensions of the cirrus-
sockets. Cirri composed of about forty-five joints, almost all of which, and especially
the basal ones, are wider than long. Interarticular pores disappear at the thir-
teenth node.

The basals appear externally as rhomboidal knobs, but they extend laterally to meet
their fellows in the re-entering angles of the calyx. Four radials, of which the first is
relatively short and wide, and the second a syzygy. The rays divide four, and
occasionally five, times, giving about ninety arms. These consist of about one hundred
and ten joints above the tertiary axillaries, and, like the rays, are quite smooth at the
base, only becoming serrate towards the extremities. Four or rarely six distichals in the
primary arms ; eight or ten palmars in the secondaries ; and the tertiaries of eight to
eighteen (usually twelve or fourteen) joints. The next division (when present) occurs
after about fourteen or sixteen (ten to twenty-four) joints more. The third joint after
each axillary is usually a syzygy. Another between the tenth and thirtieth brachials,
and then an interval of five to thirteen joints between successive syzygies.

The two radial pinnules, and also those on the lower distichals, have one or two
massive basal joints ; but the following joints, though long and moderately thick, are
very much flattened laterally, so that the dorsal surface is reduced to a mere edge. The
longest pinnules are those immediately above and below the distichal axillary, and are
less compressed than their predecessors, so that the joints are more uniform in appear-
ance, though the lower ones are relatively large and cuboidal. The palmar pinnules are
all long ; but the size begins to diminish beyond the axillaries, rapidly at first, and after-
wards somewhat slowly.

The disk is thickly covered with plates which are small and more closely set upon the
anal tube than elsewhere. Brachial ambulacra not much above the arm-groove, and
supported by bifid plates which are diflerentiated into side and covering plates about the
middle of the pinnules.

Colour in spii-it — calyx and arm-bases grey ; arms and stem nearly white, but the tips
of the arms light brown.

RemarJcs. — The fine specimen which forms the sul:)ject of the above description has
unfortunately lost most of its arms in the usual way, viz., by fracture at one or other of
the lower syzygies. In the frequency of its ray-divisions, in the constant presence of
supra-palmars, and in the diameter of its stem, it ranks among the largest types of
recent Pentacrinid*, and I have much pleasure in associating it with the well known
name of Mr. John Murray.

The species which it most nearly resembles is Metacrinus nobilis, from Station 192,
near the Ki Islands ; though the two forms difi"er considerably in the characters of the

REPORT ON THE CRINOIDEA.

351

stem, as may be seen by comparing figs. 5-7 and figs. 15-17 on PI. XLI. Tlic
whole stem of Metacrinus murrayi is slightly grooved along the interradial lines
(fig. 15), so that the pentagonal outline of the internodal joints has shallow re-entering
angles (fig. 17), while those of the nodal joints are sharper than in Metacrinus nohilis
(figs. 6, 16). The modification of the supra-nodals is about the same in both types, but
the infra-nodals of Metacrinus nohilis are more cut away to receive the cirrus-bases than
are those o{ Metacrinus murrayi (PI. XLI. figs. 5, 15). The internodes of the stem in
the latter species are slightly longer, and the interarticular pores reach farther from the
cup than in Metacrinus nohilis, but there are fewer joints in the primary arms, and the
dorsal edges of the pinnules are less serrate (PI. XLI. figs. 9, 10 ; PI. XLII. figs. 2, 3). Of
the two remaining species with smooth stems and four radials, Metacrinus superhus is
readily distinguished from Metacrinus mu7-rayi by the great size of the cirri and the
roughness of the arm-bases ; while the stem of the smaller Metacrinus varians has
considerably shorter internodes and nearly flat sides.

The habitat of Metacrinus mvrrayi is unfortunately unknown, no record having been
kept of the Station at which it was dredged. It may perhaps be one of the two species
which are noted in Sir Wyville Thomson's diary as having been obtained at Station 210
(Panglao), no Crinoid having reached me with the label of this Station. On the other
hand, it may have been found at Station 192 (Ki Islands) or Station 214 (Meangis
Islands). A Cirripede attached to one of the cirri was identified by Dr. Hoek as
Scalpellum halanoides. This species is abundant at Station 192; and Dr. Hoek had
seen individuals from no other locality. Under these circumstances one would almost
have been justified in assigning Metacrinus murrayi to that Station ; but Dr. Hoek has
also recognised Scalpellum halanoides on a cirrus of Metacrinus varians which I sent
him from Station 214. Either of these Stations therefore, to say nothing of Station 210,
might have been the habitat of Metacrinus murrayi.

4. Metacrinus nohilis, n. sp. (PI. XLI. figs. 5-11 ; PL XL HI.).

Dimensions.

Length of stem to twenty-fifth node,

Diameter of stem, .

Longest cirrus (forty-eight joints).

Diameter of calyx, ....

Diameter of disk, ....

Length of arm (one hundred and twentv joints above tertiary axillary).

Length of large distichal pinnule (twenty-three joints).

Length of first pinnule after tertiary axillary (thirteen joints),

28-00 cm.

7 '00 mm.
46-00 „

9-00 „

21-00 .,

125-00 „

27-00 „

14-00 „

Stem robust, A^dth a simple pentagonal outline and smooth, flat sides. LTsually ten or
twelve internodal joints, with slightly crenulated edges. Nodal joints bluntly stellate, with

352 THE VOYAGE OF H.M.S. CHALLENGER.

wide sockets which take up their whole height and encroach considerably both ou infra-
nodal and on supra-nodal joints. The cirri have about forty-five tolerably uniform joints,
and are longest between the twelfth and fourteenth nodes. The interarticular pores end
at the tenth node.

Basals prominent, with slight downward extensions. Eadials four, rather strongly
convex, the second a syzygy. Generally four, and sometimes five divisions of the rays,
giving seventy arms or more. These have from one hundred to one hundred and twenty
joints beyond the last axillary, and are smooth at the base ; but their middle and outer
portions are markedly serrate in the medio-dorsal line. Primary arms usually of five
distichal joints, one or sometimes two of which are syzygial. Eight or nine palmars in
the secondary arms, the second or third of which is a syzygy. Tertiaries of twelve to
twenty joints (usually about fifteen), with the third a syzygy. In a few cases there
is another division after about twenty joints more. There is generally a syzygy in the
third brachial of the free arm ; another between the twelfth and thirty-seventh brachials,
and others at intervals of four to thirteen joints.

The pinnules on the radials and lower distichals are all very long and much com-
pressed aljove the enlarged basal joints, wlide their terminal portions have a serrate
dorsal edge. The following pinnules, as far as the tertiary axillaries, have wide
and somewhat prismatic basal joints like those lower down on the rays, but with more
curved sides, and consisting of more uniform joints, the dorsal edges of which project
forwards.

Disk rather closely plated, especially in the anal interradius and along the ambulacra.
Brachial ambulacra partially withdrawn into the arm-groove, and supported by
irregularly shaped plates. Side plates not diff'erentiated tUl near the ends of the
pinnules.

Colour when fresh — the stems almost white, and the crowns light yellow or light
reddish-orange (Moseley) ; in spirit, white, with traces of light l^roM'n.

Localitij. — Station 192, September 26, 1874 ; in the Arafura Sea, off the Ki Islands ;
lat. 5° 49' S., long. 132° 14' E. ; 140 fathoms ; blue mud. Two large specimens, one of
which has lost all its arms, and one smaller varietal form.

Remarhs. — This fine species is readily distinguished from Metacrinus murrayi by its
flat ungrooved stem (PL XLI. fig. 5), with shorter internodes and more markedly incised
infra-nodal joints (PI. XLIII fig. 1). The primary arms are generally longer than in
that type, and the extremities of the arms and pinnules more serrate. Metacrinus
varians, which resemljles Metacrinus nohilis in having a flat ungrooved stem
(PI. XLVII. figs. 6, 8), is altogether a smaller type with shorter internodes and no
axillaries after the palmars, so that the number of arms does not exceed forty ; while
the large Metaxrinus superbus has many more cirrus-joints and its arm -bases uneven,
owing to the thickness of the proximal and distal edges of the joints.

REPORT ON THE CRINOIDEA.

3.53

Togetlier with the two hirge specimens of Metacrimis nohiUs which agree very well
in their general characters, there was also obtained a fragment of stem which appears to
belong to this species. The iuternodes are much shorter than in the type, and contain a
number of thin joints which are evidently newly formed ; while there are traces of inter-
articular pores in its upper portion. As, however, the cirri are all directed downwards it
is possible that the individual was in a semi-free condition, with a comparatively short
stem like Pentacrinus maclearanus (PI. XVI.) or Pentacrinus alternicirrus. At one of
the nodes a cirrus-socket is totally undeveloped ; and the same appears to be the case at
another node farther down the stem, where a large Verruca covers the whole of one side
of the nodal joint, also extending on to those above and below it.

Besides the two large specimens mentioned above there was also obtained at Station
192 a smaller individual which agrees so closely with the type of this species that I find
some difficulty in separating it, although the two forms appear at first sight to be totally
different. The stem, which is much more rounded than in the ty[3e (PI. XLI. figs. 7, 8),
is barely 5 mm. in diameter ; and although the internodal joints are generally smooth,
their angles are sometimes slightly produced outwards, while indications of horizontal
ridges appear here and there below the twelfth node. The grouping of the arm-divisions
is essentially the same as in the larger type, so far as can be made out in the fragmentary
condition of the dried specimen ; and in default of further information respecting
its characters, I do not see how to classify it otherwise than as an immature or smaller
variety of Metacrinus nobilis.

5. Metacrinus varians, n. sp. (PI. XLIV. ; PI. XLVII. figs. 6-12).

Dimensions.

Length of stem to thirty-second node,

Diameter of stem, ....

Longest cirrus (fifty joints), .

Diameter of calyx, ....

Length of arm (ninety joints beyond palmar axillary

Length of distichal pinnule (eighteen joints).

Length of first pinnule after palmar axillary (nineteen joints).

34-00 cm.

5 •2.5 mm.
40-00 „

8-50 „
85-00 „
16-00 „
1200 „

Stem moderately robust, with a simple pentagonal outline, its sides being nearly flat,
and scarcely grooved at all. Six to ten internodal joints (usually eight or nine), ■with
smooth sides and but faintly crenulated edges. Nodal joints bluntly stellate, with
moderately wide sockets which take wp their whole height, extending well downwards on
to the infra-nodals and also upwards on to the supra-nodals. Cirri of forty-five to fifty
very uniform joints, the lowest of Avhich are but little wider than their successors ; the

(ZOOL. CHALL. EXP. — PART XXXII. — 1884.) li 45

35-1 THE VOYAGE OF H.M.S. CHALLENGER.

lower cirri not specially shorter than the upper. Interarticular pores end between the
fifth and seventh nodes.

Basals variable, sometimes flattened and almost ol^long, with wide upper angles and
no downward extensions ; sometimes very prominent and sharply pointed below.
Eadials usually four, with rather flattened surfaces, the second a sjzjgj. But there are
several variations from this type. The rays divide three, or rarely four, times, giving
about forty arms. These have about ninety joints beyond the last axillary, and are
moderately smooth at the base ; but their terminal portions become strongly serrate in
the medio-dorsal line. Primary arms of six to ten (usually six or eight) distichals, the
second or third of which is a syzygy. Secondary arms of ten to twenty-five (usually
twelve or fourteen) palmars, the third of which is generally a syzygy. On the outer side
of the ray there is sometimes, but rarely, another axillary after from fourteen to twenty-
six joints, of which the second or third is a syzygy. There is generally a syzygy in the
third brachial of the free arm ; another between the seventh and eleventh brachials, and
others at intervals of two to seventeen (usually four to eight) joints.

Some of the radial pinnules are large, with massive and cuboidal lower joints ; but the
four following pinnules are smaller, and the basal joints, though broad and somewhat
flattened, are not as a rule specially massive, becoming comparatively inconspicuous
after the palmar axillary. Their dorsal edge is sharp but not strongly serrate.

Disk well plated, both at the sides of and between the ambulacra. Brachial ambulacra
but little above the arm-groove, and supported by large plates which soon pass into dis-
tinct side and covering plates on the pinnules.

Colour in spirit, greyish-white ; " of a uniform dusky purple when fresh " (Moseley).

Locality. — Station 170a, July 14, 1874; near the Kermadec Islands; lat. 29° 45'
S., long. 178° 11' W. ; 630 fathoms ; volcanic mud; bottom temperature, 39°-5 F. One
specimen, but doubtful.

Station 214, February 10, 1875; ofi" the Meangis Islands; lat. 4° 33' N., long.
127° 6' E.; 500 fathoms; blue mud; bottom temperature, 41"-8 F. Three (four?)
specimens.

Remarks. — The individual figured on PI. XLIV., which is very well preserved, was
obtained at Station 214, off the Meangis Islands, together with two other examples which
have the arms broken off" at the syzygy in the second radials. A fourth mutilated
individual, which had lost the disk and two rays, reached me in a bottle which also con-
tained one specimen of Metacrinus unjvillii, and the label of Station 170a (near the
Kermadecs). On the other hand, the figured specimen of Metacrinus wyvilUi and also
that of Metacrinus varians were together in a jar with the label of Station 214
(Meangis Islands). There is no question from other evidence that Metacrinus ^vjJvillii,
like Pentacrinus naresianus, occurs at both Stations ; but I am a little uncertain al^out
the single specimen of Metacrinus varians. For the two species, although distinctly

REPORT ON THE CRINOIDEA. 355

differentiated by Sir Wyville Thomson, and drawn on separate plates, do not seem to have
been kept apart by him ; and it is just possible that one individual of each may have been
accidentally interchanged. An additional reason for believing that this doubtful example
of Metacrinus varians really belongs to Station 214, and not to Station 170a, is that both
the Cirripedes [Scalpellum hcdanoides and Verruca nitida) which are attached to the
stem and cirri are recorded from Station 214, but not from Station 170a.

The four individuals of Metacrinus varians which the collection contains present a
considerable amount of variation in several points of structure. One of them has a cirrus
missing at one of the nodes on the stem, while two more are undeveloped at another node.
The radials vary greatly, and present the following modifications, the second being a
syzygy in every case : —

Three radiala, the axillary a syzygy.
Four radials, the axillary simple (type).
Four radials, the axillary a syzygy.
Five radials, the axillary a syzygy.

In correspondence with this there is a great amount of variation in the size of the
basal joints of the lowest pinnules. For there may be sometimes only one, and some-
times two of these appendages in each interradial space ; and in the latter case the basal
joints of one or both of them are much smaller than usual. The size of the distichal
pinnules is also to some extent affected in the same way, the first one being frequently
much smaller at the base than its successors.

Metacrinus varians is a species that stands very much by itself among those forms of
Metacrinus which have normally but four radials. It is smaller and less robust than all
the species previously described. From Meta^crinus angulatus and Metacrinus cingu-
latus it differs in the absence of the markings on the sides of the stem-joints
(PL XXXIX. figs. 3, 11 ; PI. XLI. fig. 1 ; PI. XLVII. fig. 6), and in the slightly shorter
internodes ; while in the types with smooth stems [Metacrinus nohilis, Metacrinus
murrayi, and Metacrinus superbus) the internodes are considerably longer than in the
more slender Metacrinus varians, and the grouping of the arm-divisions is different.

6. Metacrinus moseleyi, n. sp. (Pis. XLV., XLVL),

Dimensions.

Length of tapering stem of young individual to twenty-third node.
Diameter of lower part of stem, ....

Longest cirrus (forty-five joints), ....

Diameter of calyx, ......

Length of the first radial pinnule (fifteen joints).

Length of first pinnule after palmar axillary (thirteen joints),

138'00

mm

3-00

j>

35-00

)i

7-00

»)

12-50

}j

950

))

356 THE VOYAGE OF H.M.S. CHALLENGER.

Stem slender and of a rounded pentagonal form, without lateral grooves. Usually
seven internodal joints with scarcely crenulated edges and more or less marked horizontal
ridges, which are especially prominent on the larger joints. The nodal joints are deeply
hollowed by the cirrus-sockets, and have their angles slightly produced, so as to be lobate
in form. The sockets extend but very slightly either on to the supra- or on to the infra-
nodals. Cirri composed of forty to forty-five joints, the basal ones of which are not
specially broad, while their successors are often a trifle longer than wide. Interarticular
pores end at the sixth node.

Basals relatively very large and convex, pentagonal or almost oblong in form. {N.B.,
the preceding description is based upon one individual only.) Radials very variable,
sometimes only three or four, but usually five or even six, with the second and sometimes
also the fourth a syzygy. Most of the rays divide three times, giving from thirty to
forty arms, the longest unbroken ones of which consist of about fifty joints beyond the
palmar axillary.

The dorsal surface of the skeleton is rather uneven, owing to the overlapping of the
joints and the elevation of their distal edges, especially between the radials and the
palmar axdlaries. Primary arms of six to twelve (generally eight or ten) joints.
Secondaries of seven to sixteen (usually ten or twelve) palmars. The third joint, but not
unfrequently the fourth or fifth, after each axillary is generally a syzygy. The next
syzygy in the free arms may be anywhere between the fifth and the thirtieth brachials,
and others follow at very irregular intervals.

The radial pinnules vary in appearance according to their position, but the two lower
joints are larger and more cuboidal than the rest, which are flattened laterallj'. The
distichal and palmar pinnules are shorter than those on the radials, and gradually decrease
in size. Their dorsal edge is sharpened, but the two lower joints are broad and
expanded, as is also the case, though in a less degree, with the first pinnules on the free
arms.

The disk is thickly covered with comparatively large plates, and the ambulacral
skeleton is well developed. The brachial ambulacra are withdrawn into the narrow arm-
groove, and but little plated independently of those of the pinnules, which have scpiarish
side plates and large, rounded, covering plates.

Colour — a uniform dusky purple when fresh (Moseley), light grey when dry, almost
white in spirit.

Locality. — Station 214, February 10 ,1875 ; off the Meangis Islands ; lat. 4° 33' N.,
long. 127° 6' W. ; 500 fathoms ; blue mud ; bottom temperature, 41°"8 F. Two specimens.

Remarks — I have much pleasure in associating this pretty little species \dih the
name of Prof. H. N. Moseley, F.R.S. Two specimens of it were obtained by the
Challenger. One of them, represented on PI. XLV., is evidently immature, as shown
by the relatively large size of its basals, the length of its lower and middle cirrus-

REPORT ON THE CRINOIDEA. 357

joints, the frequent alternation of thick and thin joints in the stem, and the small size of
its arms.

The larger specimen, figured on PI. XL VI., appears to have met wdth an accident since
it was drawTi, for it came into my hands in the dry state, having lost its stem and basal
ring. The characters of the rays and arms, however, are so essentially similar to those of
the smaller individual that I have no hesitation in regarding the two as identical. The
characters of the stem as a whole come nearest to those of Metacrinus cingulatus
(compare PI. XLI. figs. 1-3, and PI. XLV. figs. 2-6). In both cases there is a continu-
ous horizontal ridge round each of the mature internodal joints, but the articular faces
are more lobate in Metacrimts cingulatus (PI. XLI. fig. 3) than in Metacrinus moseleyi
(PI. XLV. fig. 2). The same is the case with the nodal joints which have more produced
angles and consequently deeper cirrus-sockets in Metacrinus moseleyi than in the larger
species. The characters of the cup, however, are quite difi"erent in the two types, that of
Metacrinus cingulatus being extremely regular in the number of its radials, while in
Metacrinus moseleyi there may be as few as three or as many as six. The only other
species which resembles it in this respect is the large Metacrinus rotundus ^ from Japan,
which has a smooth stem, with much longer internodes. The ten rays of the two indi-
viduals of Metacrinus moseleyi are constructed as follows : —

* One of three joints, the second and the axillary both syzygies.

* One of four joints, the second and the axillary both syzygies.
Three of five joints, the second a syzygy.

Two of five joints, the second and fourth syzygies.

One of six joints, the second a syzygy.

Two of six joints, the second and fourth syzygies.

The number of primitive joints iu the ray, therefore, before the union of one or more
pairs by syzygy, varies from five to eight, just as in other species of Metacrinus. The
irregularity which distinguishes Metacrinus moseleyi thus lies rather in the mode of
union of the primitive joints to form syzygial pairs than in any excess or defect of their
numljer ; though as a general rule there are either five {Metacrinus angulatus) or eight
{Metacrinus ivyviUii), and not both t}"pes in the same individual.

Two very anomalous instances which occur in the dry specimen are marked with an
asterisk iu the above list. In the first case the five primitive radials have become
reduced to three, owing to the union of the last four into two syzygial pairs. In
Metacrinus cmgulatus (PI. XXXIX. fig. l) the fourth primitive joint remains distinct
from the axillary to which it is united by muscles, and bears the second pinnule. But
on this abnormal ray of Metacrinus moseleyi these two joints are united by syzygy, and
as the hjrpozygal of a syzygy never bears a pinnule, the natural condition would have

' See Trans. Linn. Soc. Lond. (ZooL), ser. 2, vol. ii. p. 437.

S58

THE VOYAGE OF H.M.S. CHALLENGER.

been the suppression of tlie second radial pinnule. As a matter of fact, however, it is
present on the epizygal of the second radial, which thus bears two pinnules, one on each
side. A similar instance of two pinnules on one joint also occurs in Metacrinus
cingulatus {ante, p. 349), and recalls the condition of certain Palseocrinoids.

In the next ray to this one the fourth radial, which is a syzygial joint, is also the
axillary; and although the first and second radial pinnules are present as usual, the
axiUary epizygal bears a pinnule in addition to the two primary arms. This pinnule,
however, really belongs to one of the first distichal joints, which is much smaller than its
fellow^ and does not reach the outer edge of the arm at all. The consequence is that the
first pinnule of the primary arm is borne by its second and not by its first joint, as is
usually the case.

Apart from these irregularities, the larger dry specimen of Metacrinus moseleyi
appears to be tolerably normal in its character. But the smaller individual is different.
The stem tapers downw^ards, being only 3 mm. Avide at the twenty-third node, but
rather over 4 mm. at the twelfth node. The arms too are much malformed by com-
mencing Myzostoma-cysts, either in the arm itself, or in the base of a pinnule, as shown
in PI. XLV. fig. 7.

. 27-00

cm.

4-50

mm

37-00

7-25

80-00

13-00

10-00

7. Metacrinus wyvillii, n. sp. (PL XL VII. figs. 1-5 ; PL XLVIIL).

Dimensions.

Length of stem to thirty- fourth node,

Diameter of stem, ......

Longest cirrus (forty-five joints), ....

Diameter of calyx, ......

Length of arm (ninety joints beyond palmar axiUary),

Length of first distichal pinnule (fifteen joints),

Length of first pinnule after palman axillary (sixteen joints), .

Stem pentagonal and moderately robust. Five to eight (generally seven) internodal
joints, with more or less well defined horizontal ridges and slightly crenulated edges.
Nodal joints bluntly stellate, wdth large fiicets of variable shape which take up their
whole height, while the sockets extend on to both supra- and infra-nodal joints, especially
the latter. Cirri of forty to forty-five very uniform joints, the lowest of which are but
little wider than their successors. The lower cirri not specially shorter than the
upper. Interarticular pores extend down to the sixth or seventh node.

Basals pentagonal, large and prominent, more or less sharply pointed lielow, some-
times being almost hexagonal. Radials usually six, with syzygies in the second and
fourth. The rays, which are somewhat closely set, divide three or very rarely four times,
giving nearly forty arms. These have about ninety joints above the palmar axillaries,
and are almost quite smooth dorsally, except in the terminal third. Primary arms of

EEPORT ON THE CRINOIDEA. 359

four to ten (usually eight or nine) disticlials ; twelve to twenty-six, but usually not more
than sixteen, palmars in the secondary arms. Sometimes, but rarely, there is another '
axillary after some thirteen to twenty joints more. The third joint after each axillary is
generally a syzygy. The following syzygies are distributed very irregularly in the free
arms. The second is between the ninth and thirtieth brachials, usually about the
twelfth or fifteenth ; and the others at intervals of three to twenty (generally ten or
twelve) joints.

The second and following radials bear pinnules of about a dozen moderately stout
joints, the lowest of which are more or less massive and cuboidal, but vary considerably
in size according to circumstances. The first distichal pinnule is nearly similar to those
on the radials, but the following ones consist of more flattened joints, the lowest of which
are much broader than their successors. This inequality is very marked in the pinnules
which are borne on the palmars and the lower parts of the free arms, and also in a less
degree in the smaller terminal pinnules. Brachial ambulacra but little above the
narrow arm-groove, and only slightly plated between the origins of the pinnule-
ambulacra, which soon begin to show well defined side plates.

Colour in spirit, gre5dsh-white ; a uniform dusky purple when fresh (Moseley).

Localities. — Station 170a, July 14, 1874 ; near the Kermadec Islands ; lat. 29° 45' S.,
long. 178° 11' W.; 630 fathoms; volcanic mud; bottom temperature, 39°'5 F. Two
specimens, one rather young.

Station 214, February 10, 1875; off" the Meangis Islands; lat. 4° 33' K, long.
127° 6' W.; 500 fathoms ; blue mud ; bottom temperature, 41°'8 F. One good specimen ;
one mutilated individual, and two stem-fragments, one of which has the calyx and a
portion of the arm-bases remaining.

RemarTcs. — This species is the only Metacrinus which is known with certainty to
occur in the South as well as in the North Pacific. Although nearly resembling
Metacrinus moseleyi, Metacrinus costatus, and Metacrinus nodosus in the length of the
internodes of the stem and in the number of the distichal joints, it diff'ers from them all
in the other characters of the stem. The two last mentioned types have a somewhat
sharply pentagonal stem, and the cirrus-facets are not so high as the nodal joints, the
angles of which are much produced ; while the sides of the joints are smooth or slightly
tubercular (PI. XLIX. fig. 3 ; PL LI. fig. 8). Metacrinus ivyvillii, however, has a more
rounded stem with horizontal ridges on the internodal joints ; while the cirrus-facets
occupy the whole height of the nodal joints (PI. XL VII. figs. 1, 2). It is altogether a
larger species than Metacrinus moseleyi, and has entirely diff"erent stem-joints, as is
immediately evident upon comparison of the figures on Pis. XLV. and XLVII. respectively.

Another good character of Metacrinus wyvillii, which is more or less visible, however,
in the species mentioned above, is the peculiar enlargement and flattening against the
arm of the basal joints of the pinnules immediately above the radials, and the persistence

560

THE VOYAGE OF H.M.S. CHALLENGER.

of this character to far out on the arms. It is well shown in some parts of fig. 1 on
PI. XLVIII.

It was on a proof copy of this plate that I found the MS. name Metacrinus in Sir
"\Vy\dlle's own handwriting ; and I have therefore taken the opportunity of associating
this elegant species with his memory. It is one of some importance on account of its
occurrence at two such widely separated localities as the Kermadec Islands (Station 170a)
in lat. 29° 45' S., and the Meangis Islands (Station 214) in lat. 4° 33' N. The two
specimens obtained at the former Station are both of very considerable interest. The
stem of the larger one, which is 25 cm. long, is broken at the thirty-sixth node, the inter-
nodes being shorter than in the more northern individuals, as seven joints are the
exception and not the rule. As in so many other cases, this stem tapers gradually down-
wards, being almost 5 mm. wide immediately below the calyx, and less than 3 mm. at
the lowest node, just above which the diminution in size is most rapid.

The other specimen obtained at Station 170a is a young individual, with a head not
more than 55 mm. long, and less than sixty joints in the free arms. The attached portion
of the stem, which is 185 mm. long, is a trifle over 3 mm. wide just below the basals,
which are relatively very large and prominent. Its diameter decreases slightly to about
the fifth node, and then slowly increases again till the sixteenth node, where the stem
widens rather suddenly. It reaches 4 mm. at the last joint immediately below the
twenty-seventh node. The ornamentation of this stem is less well defined than in the
older specimen from the same locahty. In fact there is a good deal of individual variation,
some of the stems from Station 214 ha\'ing very well defined horizontal ridges (PL XLVII.
figs. 1, 2), while they are rather inconspicuous on the figured specimen (PI. XLVIII. fig. 2).

One individual presents a curious amount of variation in the character of the rays.
Three are normal, consisting of six joints, of which the second and fourth are syzygies ;
but on one of these the usual pinnule on the epizygal of the fourth brachial is
missing, though that on the next joint is present. Another ray consists of seven
joints, the fifth and sixth of which bear pinnules on the same side ; while the fifth ray
has eight joints with no pinnule on the sixth.

8. Metacrinus costatus, n. sp. (PL XLVII. fig. 13 ; PL XLIX.).

Dimensions.

Length of stem to the twenty-fourth node, .

Diameter of stem, ....

Longest cirrus (forty-five joints).

Diameter of calyx, ....

Length of arm (one hundred joints beyond the palmar axillary),

Length of pinnule on first distichal (seventeen joints),

Length of first pinnule after palmar axiUary (twenty joints),

. 20-00

cm.

3-75

mm

32-00

j»

6-50

J

70-00

J

U-50

»

11-00

>

REPORT ON THE CRINOIDEA. 361

Stem slender and pentagonal in outline, with .slight re-entering angles. Generally
seven or eight interuodal joints with distinctly crenulated edges. Their relatively high
sides are somewhat hollowed, and smooth or marked with foint ridges ; while their angles
are sharpened and a little produced outwards, so that the whole stem is traversed by five
well defined interradial ridges.

The supra-nodals are but slightly incised, and the wide cirrus-facets do not reach the
upper edges of the nodal joints. They likemse barely reach the lower edges, coming
much nearer to them in some specimens than in others ; and the infra-nodals are scarcely
grooved, so that their re-entering angles are but little more marked than those of the
other internodal joints.

Cirri composed of about forty joints, the first two of which are short and wide, while
the eighth and a few following ones are sometimes a little longer than wide. The lower
cirri do not seem to be specially shorter than the upper. Interarticular pores not
visible below the tenth node.

The basals (in the only specimen possessing them) appear as small rhomboidal knobs
with their pointed lower extremities resting on the interradial ridges at the top of the
stem ; but they extend laterally and meet their fellows in the re-entering angles between
the first radials. The rays consist of six rounded joints, of which the second and fourth
are sj^zygies, and are well separated laterally above the hypozygals of the second radials.
They all divide three times, and there are generally additional axillaries on the two outer-
most of each set of four tertiary arms thus produced, so that the total number of arms
reaches about sixty. They consist of about one hundred joints above the palmar
axillaries, and are almost smooth in the medio-dorsal line till near the ends, which are
slightly serrate.

The distichals, palmars, and lower brachials present a peculiarity which is much more
marked in the baseless individual than in the more normal one. The pinnule-bearing
side of each joint is slightly bent outwards above the pinnule-socket, and its edge is cut
into several small teeth or spines. In addition to this the front edge of each joint and
the coiTesponding part of the hinder edge of its successor are slightly raised on one or
both sides, and are also more or less spinose. These characters are perhaps most distinctly
marked upon the palmars, not being fully developed upon the distichals, and disappearing
a little beyond the level of the tertiary axillaries. Six or eight joints in the primary
arms ; secondaries of eight to fourteen (usiially ten or tAvelve) palmars. The next division
(when present) may be from eight to twenty (generally twelve to sixteen) joints, and in
one case there is another axillary after sixteen joints more. The third joint after each
axillary is usually a syzygy. The next syzygy in the free arms may be anywhere
between the sixth and thirtieth brachials, after which an interval of three to eighteen
joints occurs between successive syzygies.

The pinnules on the radials and first distichals are large and massive, consisting of

(ZOOL. CHALL. ESP. — PAET XXXII. — 1884.) li 46

362 THE VOYAGE OF H.M.S. CHALLENGER.

eighteen or twenty joints, the lowest of which are cuboidal. The next few, though still
very thick, are much flattened laterally and gradually diminish in width, with the lateral
edges of the dorsal surface raised and thickened. The terminal part of the pinnule
consists of much smaller joints, and is more or less distinctly serrate. In the pinnules of
the third and the next following distichals the dorsal surface of the two or three thick
basal joints is rather broad, but the following joints diminish rapidly both in breadth and
in thickness. Beyond the distichal axillaries all the pinnule-joints are longer than wide,
with the exception of the first two, which are much expanded, and this character is very
marked in all the following pinnules till about the level of the fourth axillaries. Beyond
this point the second and third jiinnule-joints more nearly resemble their successors,
though traces of the expansion of the first joint are visible for some distance farther.

The disk (so far as it is visible) is well protected by plates, both on its ventral surface
and on its sides, right down to the h}^ozygal of the second brachial. The brachial
ambulacra are but little above the narrow arm-groove, and are protected like those of the
large lower pinnules by very irregular plates. The terminal pinnules have well defined
and rather jjointed side plates.

Colour — a uniform dusky purple when fresh (Moseley) ; in spirit, light brownish-
white.

Zoca^%— Station 214, February 10, 1875; off the Meangis Islands; lat. 4° 33' N.,
long. 127° 6'E. ; 500 fathoms ; blue mud ; bottom temperature, 41°'8 F. Two specimens,
with Myzostoma wyville-thomsoni, von Graff.

Remarks. — This elegant little species is the smallest Metacrinus which I have yet
seen, with the exception of Metacrinus nodosus ; and it has many points of resemblance
with that type, as will be explained later. Although the stem and cup are much less
robust than in Metacrinus wyvilUi, the number of arms is nearly half as large again as in
that species, which rarely has an axillary beyond the palmars ; while in Metacrinus
costatus this is generally the case on four out of the eight tertiary arms, and there may
be another axillary beyond the supra-palmar. The number of internodal joints is nearly
the same in the two species, being rarely less than seven in Metacrinus costatus, though
sometimes falling to five in Metacrinus ivyvillii. But they are totally difi"erent in their
form and external markings, as will be evident from a comparison of PL XLVII.
figs. 1-4, and PI. XLIX. figs. 3, 4. The nodal joints are also quite difi"erent in the two
species. The cirrus-sockets of Metacrinus wyvillii (PI. XLVII. figs. 1, 2) extend both
upwards and downwards on to the supra- and infra-nodal joints beyond the articular
facets, which occupy the whole height of the nodal joints. But this is far from being the
case in Metacrinus costatus ; and the nodal joints therefore are less deeply incised than
in Metacrinus wyvillii, while their angles are much sharper and more produced outwards
(PI. XLIX. figs. 3, 5). There is also a good deal of difi'erenee between the pinnules of
the two t}^es. Both those on the radials and distichals of Metacrinus wyvillii and those

REPORT ON THE CRINOIDEA. 363

farther out on the arms are smaller than tlie corresponding pinnules of Metacrimis
costatus ; while the lowest pinnules are smaller, smoother, and have more rounded joints
than the more massive but flattened pinnules of Mctacrinus costatus with their serrate
ends. On the other hand, the expansion of the two basal joints in the pinnules beyond
the palmar axillary is more conspicuous in Metacrinus wyvUlii than in Mctacrinus
costatus, and the remaining pinnule-joints are distinctly stouter than in that species.

There is a great amount of difference between the two specimens described above
except in the characters of the stem, which are extremely constant, the number of
internodal joints being almost invariably seven or eight. In the first place the basal
plates of one individual are entirely absent ; and each of the radials, which are slightly
higher than in the other example, has a small downward projection in the middle of its
base which rests directly on the top of one of the five ridges of the stem, as shown in
PI. XLIX. fig. 2. The basals of the other example are smaller than is usually the case
in the genus ; but their total absence, at any rate on the exterior of the calyx, is a most
singular anomaly. One result of it is that the position of the cirri is interradial and not
radial, as is generally the case ; and presumably therefore the peripheral vessels from
which the cirrus-vessels are supplied have a similar position. But these peripheral
vessels are continuous above with the chaml)ers of the chambered organ, which are
normally set in the direction of the rays (PL XXIV. figs. 5-8 ; PL LVIII. figs. 1, 2 ;
PL LXII. — ch) ; while the primary axial cords of the rays start from the interradial angles
of the chambered organ (PL XXIV. fig. 7 ; PL LVIII. figs. 1, -3 ; PL LXU.—al). If the
mutual relation of these organs in this anomalous specimen were only kuowTi it would
very probably throw much Hght upon the structure of the lower j)art of the calyx in
those Palseocrinoids which have interradial cirri, such as Heterocrinus, locrinus, Bary-
crinus, and Belemnocrimis Jlorifer. It is of course possible that the basals may be
internal and concealed as in most Comatulse and in some varieties of Encrinus; but I
cannot help thinking that if they were really present at all the cirri would be placed
radially as they usually are, and not interradially as is actually the case.

This baseless specimen presents the only irregularity in the number of the radials
which occurs in the two individuals. The second I'adial is not traversed by a syzj^gy, as
is invariably the case in all the other rays, though the second syzygy is in its normal
position between the fifth and sixth joints of the primitive ray as in the ordinary t}"pe ;
but there is no additional joint between this syzygy and the axillary, so that the ray
consists of six joints with the fifth a syzygy.

All the ten primary arms of this indi\ddual, however, consist of six joints, of which
the third is traversed by a syzygy ; while in the other specimen with a more normal
calyx there is only one distichal series of this character, together \vith one of eight
joints, of which the second is a syzygy ; and the remaining six also consist of eight joints,
but have a syzygy in the thiixl. The later arm-divisions of the baseless specimen are

364 THE VOYAGE OF H.M.S. CHALLENGER.

exceedingly regular, as it almost invariably has six arms above each clistichal axillary
arranged in the following order 2, 1 ; 1, 2. There are only two instances of an axillarj'-
occurring on one of the inner pair of the four tertiary arms ; while it is never absent on
one of the outer pair. This regular arrangement recalls that already noticed in some
species of Pentacrinus. But it is much less constant than in the more uormal example
wth the regular calyx, and it is in this individual that the solitary instance of a fifth
axillary occurs; while the peculiar development of spines on the lower arm-divisions is
much more marked in the one without a base.

The pinnule-ambulacra vary considerably in their appearance according to their posi-
tion, as will be seen by a comparison of PI. XLVII. fig. 13 and PI. XLIX. figs. 6, 7 ; and
it is only on the later pinnules of the arm that the side and covering plates come to be
distinctly difi'erentiated. The peculiar mode of development of the side plates on some
of the larger pinnules has been already described {ante, p. 82). The same character is also
visible in Metacrinus tvyvillii and Metacrinus nodosus (PI. LI. fig. 12), though to a less
extent.

The occasional enlargement of some of the arm -joints appears to indicate that an
encysting Myzostoma had commenced operations. This is probably the Myzostoma
pentacrini of von Graff". ^ It " does not produce real cysts upon the arms of its host, but
only swellings of several (three to six) joints, which gradually disappear." It was found
in abundance on Pentacrinus alternicirrus from the same Station (214) as Metacrinus
costatus ; whereas the free Myzostoma ivyville-thomsoni which was attached to the disk of
the latter species occupied a similar position on the disk of Metacrinus angulatus from
Station 192, off the Ki Islands,

9. Metacrinus nodosus, n. sp. (Pis. L., LI,).

Dimensions.

Length of stem to seventeentli node, ....

Diameter of stem, .......

Longest cirrus (forty-five joints), .....

Diameter of calyx, .......

Diameter of disk, .......

Length of pinnule on fourth radial (fourteen joints),

Length of pinnule on first joint after pabnar axillary (thirteen joints),

Stem slender and sharply pentagonal. Eight or nine internodal joints with distinctly
crenulated edges and a faint tubercle in the middle of each side. Nodals, especially in
the young stem, produced outwards at the angles between the wide cirrus-facets which
do not reach either edge. Supra- and infra-nodal joints both somewhat modified. Cirri

1 Zool. Chall. Exp., part xxvii. p. 63, 1884.

. U-00

cm.

3-00

mm.

. 35 00

))

6-00

7

. 12-50

>

. 13-00

J

9-00

»

REPORT ON THE CRINOIDEA. 365

of about forty joiuts, some of which (after the short and wide ones at the base) are
longer than broad. The dorsal edge of the middle joints Ls slightly serrate, but the later
ones become almost smooth. Interarticular pores end about the ninth node.

Basals pentagonal and forming a closed ring, but very convex in the centre, so as to
appear like rhomboidal knobs resting on the interradial ridges of the stem. Radials six,
very convex, the second and fourth syzygies. Rays well separated laterally, and divide
four times. Arms twelve or more to the ray, slightly serrate in the medio-dorsal line,
especially in the lower divisions. Primaries of six or eight (rarely ten) joints ; second-
aries of eight to fourteen palmars ; tertiaries of sixteen to twenty-eight joints. The
rliird joint after each axillary is a syzygy, and the next syzygy in the free arms is
anywhere between the tenth and twenty-fourth joints ; after which the syzygies follow at
intervals of five to fourteen joints.

The radial pinnules, especially the first one, mostly have wide and massive basal
joints with thickened edges ; but the middle joints are more compressed and the terminal
ones slender. The distichal pinnules are also stout and composed of large joints ; but
the following ones consist of compressed and more elongated joints, the first two of which
are considerably wider than the rest until some little way beyond the last axillary.

The disk is covered with numerous small plates, which are mucli smaller and more
closely set in the anal interradius than elsewhere. Brachial amljulacra above the arm-
grooves, and protected by irregular plates from which the covering plates and large
pointed side plates are developed on the pinnules.

Colour — of a uniform dusky purple when fresh (Moseley); greyish-white in
spirit.

Locality. — Station 170a, July 14, 1874 ; near the Kermadec Islands; lat. 29° 45' S.,
long. 178° 11' W.; 630 fathoms; volcanic mud; bottom temperature, 39°"5 F.

Remarks. — Three examples of this elegant little species seem to have been obtained
by the Challenger. One of them is cj^uite young (PL LI. fig. 1), while the other two were
in a much mutilated condition. The dissected calyx figured on PI. L. figs. 5-18 appears
to belong to the fragment represented on fig. 1 of the same Plate. It is certainly not
the same as that of which the disk is shown in fig. 2. This last and a few detached arms
have served as the basis of the foregoing description. The only species with which
Metacrinus nodosus is liable to l)e confused is the little Metacrinusi costatus from ofl" the
Meangis Islands (Station 214). For Metacrinus loyvilUi, which also has six radials and
about the same number of internodal joints, is altogether much larger and more robust.
But it has a smaller number of arms than either Metacrinus costatus or Metacrimis
nodostts. The difi"erence between these two last lies chiefly in the characters of the stem,
both of them having twelve or more arms on the ray, o-ndng to the presence of axdlaries
on the outermost of each set of four tertiary arms, as well as on some of the inner ones
occasionally. The pinnules are generally similar in the two types, and the peculiar

866 THE VOYAGE OF H.M.S. CHALLENGER.

modificatiou of the lower arm-joints above the pinnule-sockets, which was noticed
in Metacrinus costatiis, is also visible in Metacrinus nodosus, though to a less
extent.

There is a good deal of difference, however, between the stems of the two species.
While the normal number of internodal joints in Metacrinus costatus is seven or eight,
that of Metacrinus nodosus is eight or nine, and they are very regularly marked by a
faint tubercle in the middle of each side, which is flat and scarcely hollowed at all (PL LI.
fig. 8). In Metacrinus costatus, on the other hand, the sides of this stem are almost
smooth, or only marked by a few occasional horizontal ridges, while they are distinctly
hollowed between the prominent angles (PL XLIX. figs. 3, 4), so that the stem appears
to be traversed along its whole length by five rather sharp interradial ridges. In the
uppermost and growing part of the stem these ridges are much more prominent on the
closely set nodal joints than on the thin internodals which separate them. But as the
latter increase in thickness their interradial angles are also enlarged, so that those on the
nodal joints are not specially prominent. This character is also visible in Metacrinus
nodosus, both in the stem-joints of young individuals (PL LI. figs. 6, 7) and in the
growing part of the stem of the more mature specimen. But there are no strong ridges
developed at the angles of the internodes, as is the case in Metacrinus costatus, so that
those of the nodal joints are always more or less prominent (PL L. fig. 4 ; PL LI. fig. 8).

The side plates on the pinnule-ambulacra of Metacrinus nodosus are relatively large
and pointed (PL LI. figs. 11, 12). They are developed in the same style as those of
Metacrinus costatus (PL XL VII. fig. 13) from the somewhat irregular plates of the
brachial ambulacra, which are not so bifid as in Metacrinus angidatus (PL XXXIX.
fig. 13) or in Metacrinus varians (PL XL VII. figs. 11, 12).

The young individual of Metacrinus nodosus, which is represented in PL LI. fig. 1,
has a slightly tapering stem containing sixteen nodes, at the eighth of which one cutus-
socket is undeveloped. The characters of the young stem-joints, which are shown in
figs. 2-7, have been noticed already (ante, p. 291).

Of the four rays remaining in this specimen only one is normal, i.e., composed of six
joints, of which the second and fourth are syzygies. In one case the fifth joint is the
axillary, and in another the fourth, which is at the same time a syzygial joint like the
second ; whde in another ray the fourth radial is not a syzygy, though the fifth is
axdlary and united to it somewhat closely, so as to give almost the ajipearance of a
syzygy.^ But the presence of a pinnule on the fourth joint shows conclusively that it
cannot be the h}^3ozygal of a syzygy, as its homologue is in the next ray.

This specimen is so young that the palmar axUlaries are with difficulty distinguished
from the ordinary joints of the secondary arms ; and in some cases at any rate the}'
seem to have been farther from the distichal axUlaries than is usual in the larger indi-

' A little too mucli has been made of this resemblance to a syzygy in the right hand side of the figure.

REPORT ON THE CRINOIDEA. 367

vidual. Attached to a stem-fragment which was brought up wath these individuals
were a small Ophiuran and a young Turbinolian coral (PI. LI. fig. 8).

10. Metacrimis interruptus, n. sp. (PI. LIL).

Dimensions.

Length of Kteni to nineteenth node, . . . . . . 2 1 '00 cm.

Diameter of stem, . . . . . . . .4-25 mm.

Longest cirrus (forty-five joints), ...... 43'00 „

Diameter of calyx, ........ 8'00 ,,

Length of arm (one hundred and thirty joints heyond the pahuar axillary), . 105 '00 „

Length of pinnule on second radial (twenty joints), .... 22-00 „

Length of pinnule on third distichal (eighteen joints), . . . 16'00 ,,

Length of pinnule on first joint heyond the palmaraxillary (eighteen joints), . ll'OO „

Description of an Individual. — Stem moderately slender, with a sharply pentagonal
outline. Ten or eleven internodal joints with but slightly crenulated edges. Their sides
are somewhat hollowed, and marked by tolerably distinct horizontal ridges. These are
interrupted at the angles which generally bear very faint tubercles. The supra-nodal
joints are slightly incised, but the cirrus-sockets terminate in thickened rims distiuctlj'
above the lower edge of the nodal joint. Its syzygial face, like that of the infra-nodal, is
thus regularly pentagonal, as is its outline when seen from above, although the upper
surface is distinctly lobate. Cirri composed of forty to forty-five very uniform joints,
the lowest but little wider than their successors, few or none of which are longer than
wide. The cirri are longest between the ninth and twelfth nodes ; and the inter-
articular pores end at the eleventh node,

Basals widely pentagonal, but not specially prominent. Eadials six, with syzygies
in the second and fourth. The joints are somewhat sharply rounded and relatively
narrow, so that the rays are widely separated above the hypozygal of the second joint.
They all divide three times ; but there is a fourth axillary in two cases, so that the total
number of arms is forty-two. These have about one hundred and thirty joints above the
palmar axillary, the basal ones almost c^uite smooth, and the later joints only with very
slightly raised distal edges. Primary arms of eight (rarely ten) distichals, and secondaries
of twelve to sixteen or eighteen 2:)almars. In two cases there is a third di^asion after ten
and twelve joints respectively. The third joint after each axillary is a syzygy. The
next syzygy in the free arms may be anywhere between the ninth and sixtieth brachials ;
after which an interval of six to fifteen joints occurs between successive syzygies.

The radial pinnules are very large and massive, the first one esjiecially so. It consists
of twenty or twenty-two joints, the first six of which are very stout and almost cubical
in appearance, .the second being the largest, and the terminal joints slightly serrate.

368 THE VOYAGE OF H.M.S. CHALLENGER.

The following pinnules gradually decrease in size and their joints become more flattened,
the two lowest retaining a certain amount of preponderance, being often much broader
than their successors. Beyond the distichal axillary, however, this is almost entirely
lost, the pinnules tapering gradually and symmetrically from the basal joints, which are
not sjjecially distinguished in any way.

The disk (so far as it is visible) is covered with numerous small plates which are not,
however, set perfectly close to one another. Brachial ambulacra but httle above the
arm-grooves, and bordered by somewhat forked plates from which the large side plates
(if the pinnule-ambulacra are soon developed.

Colour in spirit, greyish-white, with a tinge of brown at the tips of the cirri, arms,
and pinnules.

Locality.— Station 209, January 22, 1875 ; lat. 10" 14' N., long. 123° 54' E. ;
95 fathoms ; blue mud ; bottom temperature, 71° F. One specimen.

Remarhs. — This species is readily distinguished from the three preceding ones in the
same group by the greater length of the internodes in the stem. Their component
joints (PI. LII. fig. 3) are altogether ditferent from the lobate joints of Metacrinus costatus
(PL XLIX. fig. 4) and of Metacrinus nodosus (PL LL fig. 10), having the same
pentagonal form and horizontal ridges as Metacrinus wyvlllii (PL XLVIL fig. 4).
The nodal and infra-nodal joints, however, are entirely difi"erent from those of this
type, in which the infra-nodals are distinctly incised by the cirrus-sockets, so that
their syzygial surface is lobate (PL XLVIL fig. 3) and not pentagonal as in Metacrinus
intermptus. Another character in which this species differs very markedly from the three
previously considered is the small size of the basal joints on the palmar and lower
brachial pinnules.

The type to which on the whole Metacrinus interruptns appears to be most closely
alhed is the as yet undescribed specimen dredged by the "Vega" at a depth of G5
fathoms in the Bay of Yedo. By the kind permission of Prof S. Lovdn, who was good
enough to send me some fragments of its stem, and also to allow my friend Mr. W. Percy
Sladen to examine it on my behalf, I am able to say that it appears to be totally different
from Metacrinus interrujitus. The stem-joints that I have seen have a smaller diameter
and a greater height both relatively and absolutely than those of that species ; and they
are not provided with horizontal ridges, but only with faiut tubercles at the angles, and
still less distinct ones at the sides. In the character of the nodal joints, however, and in
the absence of any extension of the cirrus-socket down on to the infra-nodals, the two
types are very closely similar, as they are in the length of the internodes. There seem
to be several other points of difference between Metacrinus interrujitus and the " Vega "
specimen, such, for example, as the length of the primary arms and the characters of the
pinnules. These will doubtless be explained more fully when the "Vega" Crinoids are
described • and I have therefore done no more than assign to the Japanese form a place

EEPORT ON THE CRINOIDEA. 369

in the tabular key to the species which embraces all the types of this genus that have yet
been discovered.

The individual described above is the only specimen of any kind of Crinoitl which was
obtained at Station 209 ; and it has fortunately suffered much less injury than many of
the larger types dredged by the Challenger. I did not succeed in finding any Mijzostoma
upon it ; but a Scalpellum is attached to its stem, and several individuals of Vei-r-uca to
the cirri, on one of which a small Avicida was fixed by its byssus.

11. Jletacrinus tuberosus, n. sp. (PI. LIII. figs. 1-G).

Three species [Metacrinus angulatus,Metacrinus nohilis, Metacrinus cingulatus), repre-
sented by " about a dozen individuals,"' were dredged by the Challenger at Station 192, off"
the Ki Islands. Together with these there came up a fragment of a stem which I cannot
refer to either of these species, nor to any other Metacrinus yet known. It presents a
curious combination of certain features which are characteristic of the stems of Metacrinus
angulatus, Metacrinus costatus, and Metacrinus nohilis respectively, three very distinct
species from widely separated localities on diff'erent sides of the equator (Ki, Meangis,
and Kermadec Islands). I have no hesitation in regarding it as belonging to another
species of the genus, although the characters of its calyx are as yet unknown.

It cannot belong to a Pentacrinus on account of the upward extension of the cirrus-
sockets on to the supra-nodal joints (PL LIII. fig. 2), a character which is eminently
distinctive of Metacrinus.

The fragment consisted of five complete internodes, one of which was sacrificed to
anatomical purposes, so that only four are shown in the figure (PL LIII. fig. 1). Each
internode consists of seven slightly crenulated joints. The five middle ones are as usual
different from those immediately above and below the nodes. The}' are sharply
pentagonal in form, with a somewhat prominent tubercle in the middle of each side
(PL nil. figs. 4, 6) ; while the angles are sharp and slightly produced outwards beyond
the ends of the petaloid areas, as is to some extent the case in Metacrinus angulatus and
3£etacrinus costatus (PL XXXIX. figs. 3, 8, 10, 11 ; PL XLIX. figs. 3, 4). This is still
more evident in the nodal joints (PL LIII. fig. 3) as it also is in the other two species
(PL XXXIX. figs. 3-5 ; PL XLIX. fig. 5), and more distinctly in Metacrinus nodosus,
the internodes of which are not specially produced at the angles (PL LI. figs. 8-10).
The consequence is that the stem of Metacrinus tuberosus, Like that of Metacrinus
angulatus and Metacrinus costatus, is traversed by prominent interradial ridges
(PL XXXVIIL; PL XXXIX. figs. 3, 11; PL XLIX. figs. 1, 3; PL LIIL figs. 1, G).
The nodal joints have somewhat deeply hollowed cirrus-sockets which have relatively
wide facets, and encroach lioth on the supra- (PL LIII. fig. 2) and on the infra-nodal

1 See R. V. Willemoes Suhm, Briefs von der Challenger Expedition, No. iv., Zdtschr. /. vdss. Zool, Bd. xxvi. p. liii.,
1876.

(zool. chall. EXP.— part xxxil — 1884.) li 47

370 THE VOYAGE OF H.M.S. CHALLENGER.

joints (PL LIII. fig. 5) ; so that their re-entering angles are deeper than those of the
remaining internodal joints (PL LIII. fig. 4). On Loth of these, but especially on the
infra-nodal (fig. 5), the single tubercles at the sides of the joint are more or less double,
and enlarged into a horizontal ridge. A peculiarity of somewhat the same kind occurs
in Metacrinus costatus (PL XLIX. fig. 3).

The cirri are about 35 mm. long, and closely resemble those of other species of the
genus. They consist of some forty uniformly squarish joints, the basal ones of which are
not much \vdder than their successors, though projecting a little beyond them on the
dorsal side. There is no trace of interarticular pores, so that this fragment cannot have
come from near the top of the stem.

Thus, then, the peculiarities of this stem-fragment are sufiiciently characteristic to
indicate that it belongs to a difi"erent species of Metacrinus from any of those described
above. In the prominence of the angles of the nodal joints, and in the presence of
tubercles along the sides of the internodes, it resembles Mctacrmus nodosns (PL LI. fig. 8).
But in the sharpness of the ridges formed by the angles of the internodes it approaches
Metacrinus angulatus (PL XXXIX. figs. 3, 11) and Metacrinus costatus (PL XLIX.
figs. 1,2). As the composition of its calyx is unknown, no place can be assigned to
Metacrinus tuberosus in the tabular scheme of the genus. If there be four radials it
would come near Metacrinus angulatus, though the internodes are shorter than in this
type ; but if the number of radials be six, its place would be next to Metacrinus costatus
and Metacrinus nodosus, both of which it resembles in the length of its internodes.

Clinging to this stem by its long arms was the pluteus-larva of an Ophiurid, with
three arm plates beyond the disk.

Locality. — Station 192, September 26, 1874 ; in the Arafura Sea, near the Ki Islands;
lat. 5° 49' S., long. 132° 14' E. ; 140 fathoms ; blue mud. A stem-fragment only.

Family Comatulijd^, d'Orbigny, 1852.
Genus Thaumatocrinus, P. H. Carpenter, 1883.

TliaumatocriHus, P. H. Carpenter, Pliil. Trans., part iii., 1883, p. 919, pi. 71.

Definition. — -Calyx composed of a centro-dorsal, basals, radials, and primary inter-
radials, the latter resting on the basals and so separating the radials laterally. That
on the anal side bears a short jointed appendage. Mouth central, and protected
by five large oral plates which occupy the greater part of the disk, and are separated
from the calyx interradials by two or three rows of small irregular plates. Five arms
only.

Remarks. — Although this very singular genus is a true Comatula, i.e., provided with a
centro-dorsal plate or cirrus-bearing top stem-joint which separates it from the remainder

REPORT ON THE CRINOIDEA. 371

of the larval stem, I have preferred to describe it in the same part of the Criuoid Report
as the Stalked Crinoids. For it is only among certain of the Palseocrinoidea that
we meet with characters which are at all like the more striking pecularities of
Thaumatocrinus.

There can, I think, be no doubt that the large and comparatively dense oral plates are
not in a state of resorption as they are in other Comatulse of the same size ; for they
have all the appearance of being permanent structures. Thaumatocriiius is therefore the
only Comatula yet known in which the oral plates of the larva persist through life as in
Hyocrinus and Rhizocrinus.

Another striking peculiarity is presented by the closed ring of relatively large basals
which have remained in their primitive position upon the exterior of the calyx and have
not undergone transformation into a rosette, as is the case in most other Comatulse. The
only other recent type in which the basals remain visible on the exterior of the calyx is
the curious genus Atelecrinus ; ^ and here they are very small in proportion to the ra dials.
This is probably also the case in the Cretaceous species which is mentioned by Schliiter -
as provided with a closed basal ring.

Both the persistence of the basals and the considerable development of the orals are
characters which, either singly or combined, would cause the type to be regarded as one
of no little interest ; but they are altogether cast into the shade by the other peculiarities
of the calyx, viz., the complete separation of the radials by relatively large interradial
plates and the presence of the anal appendage. It has been sho'UTi elsewhere ^ that in
the separation of its radials laterally Tliaumatocrimis is permanently in the condition of
a Crinoid larva at a very early period of Pentacrinoid life, and that this condition is
characteristic of certain Palseocrinoids belonging to the family Rhodocrinidas. Some genera,
such as the Lower Silurian Reteocrinus and Xenocrinus, have the radials separated by
what Messrs. Wachsmuth and Springer* describe as an "interradial series resting directly
upon the basals, consisting of a very large number of minute pieces of irregular form, and
without definite arrangement." A similar development of small irregular plates between
the rays occur in many Neocrinoids, both stalked and free, but the interradial series always
commence at the level of the second or third radials, and are completely separated from
the basals hy the ring of united first radials. This is well seen in Pentacrinus asterins
(PL XIIL fig. 1) and in the fossU Extixicrimis.

In other genera of the Rhodocrinidse such as Rhodocrmus itself, Tliylacocnnus, and
others forming the section Rhodocrinites, the first radials are separated not by small and
irregular plates as in Reteocrinus, but by large plates, one resting on a basal in each
interradius ; and this is the condition of Thaumatocrinus (PI. LVI. figs. 1—1). AMide

1 Bull. Mus. Comp. Zoiil, vol. ix., No. 4, 1881, p. 16, pi. i. fig.^. 1-7.

2 Zeitschr. d. deutsch. geol. Gesellsch., Jalirg. 1878, p. 66.

3 Phil. Trans., 1883, part iii. pp. 923-925 ; and ante, pp. 39, 40. * Revision, part ii. p. 192.

372 THE VOYAGE OF H.M.S. CHALLENGER.

resembling the Rhodocrinites in having five large plates separating the radials,
Thauviatocrinus differs from them and from most Palseocrinoids in the absence of any
higher series of calicular interradial plates resting upon the first series which separate
the radials.

Except on the anal side the primary^interradial plates of Thaumatocrimts end simply in
a free rounded edge at the margin of the disk (PI. LVI. figs. 1-3, 5), which is doubtless due
to the simplicity of the arms ; for these become free almost at once, and are not connected
laterally by much perisome in which higher orders of radials could be supported. But in
the presence of the anal appendage on the azygous interradial (PI. LVI. figs. 2, 4, 5)
Thaumatocrinus bears a remarkable resemblance to Reteocrimis as understood by
Wachsmuth and Springer, and to the Xenocrinus of S. A. Miller ; while the appendage
has an even closer resemblance to the so called " anal series " of Omjchocrinus and
Taxocrinus, the lowest plate of which rests, not on a basal, but on the upper angles of the
two first radials.

There can, I think, be no reasonable doubt that the anal appendage of Thaumato-
crinus, although free laterally, is homologous with the vertical series of plates in the anal
interradius of Reteocrinus and Xenocrinus, Onychocrinus and Taxocrinus. But owing
to the small size of Thaumatocrinus and the simplicity of its rays the anal appendage is
free ; whereas in the Palgeocrinoids it is united to the more or less branching rays by the
general series of minute irregular plates which occupy the anal interradius and pass
gradually upwards into those of the so called vault.

It is difficult to consider the existence of interradials and of the anal appendage of
TJiaumatocrinus as instances of atavism, for no known Neocrinoid presents any similar
characters, and it is a long w\ay back from a recent Comatula to a Palseozoic Crinoid.
The reappearance of these characters in such a specialised t}'pe as a Comatula is conse-
quently not a little surprising. Associated with them we find the distinctly embryonic
characters of persistent basal and oral plates, the latter occurring in no other Comatula,
together with the simplicity of the undivided arms.

Thaumatocrinus renovatus, P. H. Carpenter, 1883 (PI. LVI. figs. 1-5).

Description of an Individual. — The total width of the calyx across the disk is barely
2 mm. ; and the height of the centro-dorsal and radials together is about the same.
The former (PI. LVI. figs. 1-4) is rounded below, with its central canal completely
closed up, so that it must have been detached for some little time from the remainder of
the stem. The bases of half a dozen cirri are attached to it, and there are pits for the
reception of two or three more. In the largest stump which is preserved (PL LVI.
fig^- 1)3) the first two joints are quite short, as is usually the case in aU cirri ; but the
third reaches a length of r5 mm., so that the cirri must have been very like those of

REPORT ON THE CRINOIDEA. 373

some species of Eudiocrinus which have a succession of very long joints following the
short basal ones.

The basals are almost trapezoidal, much wider below than above, and in contact with
one another by their truncated lower angles (PI. LVI. figs. 1, 2). The midtUe of the
lower edge of each is slightly tubercular. On their narrow upper edges rest the interradials,
which are oblong and a httle higher than wide. Four of them terminate in a free edo-e at
the margin of the disk where they are in contact with the lowest anambulacral plates.
But that on the anal side bears a small tapering appendage of four or five joints, the last
of which seems to end freely (PI. LVI. figs. 2, 4, 5). The radials are larger than the
interradials, and somewhat strongly arched. There is a muscular articulation between
them and the first brachials ; but the union of these to the next joints appears to be by
ligament only. The arm-joints are long, slender, and cylindrical. One arm seems to be
broken at the syzygy in the sixth brachial ; while another has a syzygy in the fourth
and again in the eighth brachial. The second brachial bears the first pinnule, which is
on the right side in three arms and on the left in the other two. The pinnules are very
delicate, and composed of long slender joints.

The central portion of the disk is occupied by five relatively large oral plates which
stand up around the peristome (PL LVI. fig. 5) ; while between them and the margin are
two or three irregular rows of small anambulacral plates, some of them extending up on
to the lower part of the long anal tube. The brachial ambulacra are not plated, however,
and lie in the arm-grooves, close down between the muscles, but with no traces of
sacculi.

Colour in spirit, dirty white.

Zoca%.— Station 158, March 7, 1874 ; lat. 50° 1' S., long. 123° 4' E. ; 1800 fathoms ;
Globigerina ooze ; bottom temperature, 33°'5 F. One specimen much mutilated and
probably young.

374 THE VOYAGE OF H.M.S. CHALLENGER.

XIII.— BATHYMETEICAL DISTRIBUTION AND STATION LISTS.

Station List of the Stalked Crinoids which have been obtained by the various
British Expeditions for Deep-Sea Exploration between the Years 1868 and
1880.

This list also contains the names of the rarer Comatulse, when dredged at a Station
where Stalked Crinoids occurred. The presence of the more common genera [Antedon
and Actinometra) is also indicated, though without any attempt to go into specific
details.

H.M.S. "Lightning," 1868.

Station 12. Lat. 59° 36' N., long. 7° 20' W.; 530 fathoms; Globigerina ooze ; bottom

temperature, 47°'3 F.

Rhizocrinus lofotensis.

Station 16. Lat. 61° 2' N., long. 12° 4' W.; 650 fathoms ; Globigerina ooze.

Rhizocrinns lofotensis. Also at Station 12; and the "Knight
Errant," 1880, Stations 5 and 6 ; the Challenger,
Stations 24a, 122c, and 323 {fide C. W. T.). Also
at several Stations in the Gulf Stream, the Carib-
bean Sea, and off the New England coast.

H.M.S. "Porcupine," 1869.

Station 37. Lat. 47° 38' N., long. 12° 8' W.; 2435 fathoms ; Globigerina ooze ; bottom

temperature, 36° '5 F.

Bathycrinus gracilis. Also the "Talisman" (1883), off the
Morocco coast — " Par le travers du cap Ghir et
du cap Noun h. 120 milles environ de la cote,"
2000-2300 metres (about 1200 fathoms).

Station 42. Lat. 49° 12' N., long. 12° 52' W.; 862 fathoms ; ooze with sand and shells ;

bottom temj)erature, 39° '7 F.

Rhizocrinus rawsoni.

REPORT ON THE CRINOIDEA. 375

Station 43. Lat. 50° 1' N., loug. 12° 26' W.; 1207 fathoms ; Globigcrina ooze ; bottom

temperature, 37°7 F.

Rhizocrinus rawsoni. Also at Station 42 ; the Challenger, Station
76 ; and at various Stations in the Caribbean Sea.
Also the " Travailleur " (1882)—" 1900m. de pro-
fondeur sur les cotes du Maroc, par le travers du
cap Blanc;" and the "Talisman" (1883)— " Par
le travers du cap Ghir et du cap Noun h, 120
milles environ de la cote," 2000-2300 metres
(about 1200 fathoms).

H.M.S. "Porcupine," 1870.

Station 17. Lat. 39° 42' N., long. 9° 43' W. ; 1095 fathoms; ooze; bottom temperature,

39°-7 F.

Pentacrinus ivyviUe-tliomsoni. Also the " Talisman " (1883); oflF
the Morocco coast: and ao'ain off Rochefort; lat.
45° 59' K, long. 6° 29' W. of Paris; 1500 metres
(800 fathoms).

H.M.S. "Knight Errant," 1880.

Station 5. Lat. 59° 26' K, long. 7° 19' W. ; 515 fathoms; mud; bottom temperature,

45°-4 F.

Rhizocrinus lofotensis.

Station 6. Lat. 59° 37' N., long. 7° 19° W. ; 530 fathoms; grey mud; bottom

temperature, 46°'.5 F.

Rhizocrinus lofotensis.

H.M.S. Challenger, 1873-76.

Station 24a. March 25, 1873; off Culebra Island; lat. 18° 43' N., long. Q>b" 5' W.; 625

fathoms ; Pteropod ooze.

Rhizocrinus lofotensis. Also Stations 122c and 323 {fde C. W. T.) ;
the "Lightning" (1868), Stations 12 and 16;
and the "Knight Errant" (1880), Stations 5 and
6. Several Stations in the Gulf Stream, the
Caribbean Sea, and off the New England coast.

376 THE VOYAGE OF H.M.S. CHALLENGER.

Station 7G. July 3, 1873 ; south of Terceira (Azores) ; lat. 38° 11' N.,long. 27° 9' W. ;

900 fathoms; Pteropod ooze; bottom temperature, 40° F.

Rhizocrinus raivsoni. Also the "Porcupine" (1869), Stations 42
and 43. Also the " Travailleur " (1882), the
"Talisman" (1883), and at several Stations in
the Caribbean Sea.

Station lOG. August 25, 1873; lat. 1° 47' N., long. 24° 26' W. ; 1850 fathoms;

Globigerina ooze ; bottom temperature, 36°'6 F.

Bathycrimis campbeUicmus.

Hyocrinus hethelUanus (stem-fragments, Jide C. W. T.). Also
at Station 147 ; and Station 223 ?

Station 122. September 10, 1873 ; off Barra Grande ; lat. 9° 5' S., long. 34° 50' W. ;

350 fathoms ; red mud.

Pentacnnus maclearanus.

Atelecrinus balanoides. Also at several Stations in the Caribbean

Sea.

Station 122c. September 10, 1873 ; off Barra Grande; lat. 9° 10' S., long. 34° 49' W. ;

400 fathoms ; red mud.

Rhizocrinus lofotensis. Also Stations 24a and 323 {Jide C. W. T.) ;
the "Lightning" (1868), Stations 12 and 16 ; and
the "Knight Errant" (1880), Stations 5 and 6.
Several Stations in the Gulf Stream, the Carib-
bean Sea, and off the New England coast.

Station 146. December 29, 1873; lat. 46° 46' S., long. 45° 31' E. ; 1375 fathoms;

Globigerina ooze ; bottom temperature, 36°'6 F.
Bathycrimis cddrichianiis. Also at Station 147 ; and "at at least
sis or seven Stations in the Atlantic and the
Southern Ocean."

Station 147. December 30, 1873; lat. 46° 16' S., long. 48° 27' E. ; 1600 fathoms;

Diatom ooze ; bottom temperature, 34° '2 F.
Bathycrimis aldrichianus. Also at Station 146.
Hyocrinus bethellianns. Also at Stations 106 and 223 [fide

C. W. T.).
Antedon (three species).
Pronachocrinus abyssortim. Also at Station 158.

REPORT ON THE CRINOIDEA. 377

Station 158. March 7, 1874; lat. 50° 1' S., long. 123° 4' E. ; 1800 fathoms; Globi-

gerina ooze ; bottom temperature, 33°"5 F.

Thaumatocnnus renovatus.

Promachocrinus ahyssorum. Also at Station 147.

Station 170a. July 14, 1874; near the Kermadec Islands; lat. 29° 45' S., long.

178° 11' W. ; 630 fathoms; volcanic mud; bottom
temperature, 39°'5 F.

Pentacrinus naresianus. Also at Stations 171, 175, 210 (?), 214.

Metacrinus nodosus

Metacrinus wyvillii. Also at Station 214.

Antedon (six species).

Station 171. July 15, 1874; oflf Eaoul Island; lat. 28° 33' S., long. 177° 50' W. ;

600 fathoms ; hard ground ; bottom temperature, 39°'5 F.

Pentacrinus alter nicirrus. Also at Stations 210 (?), 214.
Pentacrinus naresianus. Also at Stations 170a, 175, 210 (?), 214.

Station 175. August 12, 1874 ; near Kandavu, Fiji; lat. 19° 2' S., long. 177° 10' E. ;

1350 fathoms; Globigerina ooze; bottom temperature,
36° F.

Pentacrinus naresianus. Also at Stations 170a, 171, 210 (?), 214.
Antedon (three species).

Station 192. September 26, 1874; near the Ki Islands; lat. 5° 49' S., long.

132° 14' E.; 140 fathoms; blue mud.

Metacri7ius angidatus.
Metacrinus cingulatus.
Metacrinus murrayi (?).
Metacrinus nohilis.
Metacrinus tuberosus.
Antedon (twelve species).
Actinometra (one species).

Station 209. January 22, 1875; lat. 10° 14' N., long. 123° 54' K; 95 fathoms; blue

mud; bottom temperature, 71° F.

Metacrinus interruptus.

(ZOOL. CHALL. EXP.— PART XXSH. — 1884.) li 48

378 THE VOYAGE OF H.M.S. CHALLENGER.

Station 210. January 25, 1875 ; off the Panglao and Siquijor Islands; lat. 9° 26' N.,

long. 123° 45' E. ; 375 fathoms; blue mud; bottom
temperature, 54°"1 F.

Pentacrinus alternicirrus (T). Also at Stations 171 and 214.

Pentacrinus naresianus{Vj. Also at Stations 170a, 171,175, and 214.

Metacrinus murrayi (?).

Antedon (two species).

Station 214. February 10, 1875; off the Meangis Islands; lat. 4° 33' N., long.

127° 6' E. ; 500 fathoms ; blue mud; bottom temperature,
41°-8 F.
Pentacrinus alternicirrus. Also at Stations 171 and 210 (?).
Pentacrinus naresianus. Also at Stations 170a, 171, 175, and

210 0).
Aletacrinus costatus.
, Metacrinus moseleyi.
Metacrinus murrayi (?).
Metacrinus varians.

Metacrinus wyvillii. Also at Station 170a.
Antedon (six species).
PromacJiocrinus iiaresi.

Station 223. March 19, 1875; lat. 5° 31' N., long. 145° 13' E. ; 2325 fathoms;

Globigerina ooze ; bottom temperature, 35°"5 F.
Hyocrinus hethellianus, young {fide C. W. T.). Also at Stations
106 (C. W. T.) and 147.

Station 235. June 4, 1875 ; lat. 34° 7' K, long. 138° 0' E.; 565 fathoms ; green mud ;

bottom temperature, 38° "1 F,

Pentacrinus (?) mollis.
Eudiocrinus japonicus.

Station 323. February 28, 1876 ; lat. 35° 39' S., long. 50° 47' W.; 1900 fathoms ; blue

mud ; bottom temperature, 33°*1 F.
RMzocrinus lofotensis {fide C. W. T.). Also the "Lightning"
(1868), Stations 12 and 16 ; the " Knight Errant"
(1880), Stations 5 and 6 ; and the Challenger,
Stations 24a and 122c. Also at several Stations
in the Gulf Stream, the Caribbean Sea, and off
the New England coast.

REPORT ON THE CRINOIDEA. 379

Station List op the Stalked Cricoids which have been obtained by the various
American Expeditions for Deep-Sea Exploration (mostly under the direction
OF Mr. Alexander Agassiz) between the Years 1867 and 1880.

The presence of Comatulse is noticed as in the previous list ; but the occurrence of a
species at several Stations is not, as a rule, recorded more than once.

SS. " CoRWiN," 1867.

No. 2. May 24, 1867 ; 1'6 miles from Chorrera, Cuba ; 270 fathoms.

Pentacrinus decorus (stem only ?).

SS. "Bibb," 1868-69.

May 9, 1868 ; off the Samboes; 237 fathoms.

May 11, 1868 ; off Sand Key; 248 and 306 fathoms.

Rhizocrinus lofotensis. Also off Havana, Barbados, Grenada, and
Grenadines. Dredged by the " Lightning," the
Challenger, and the " Knight Errant."

March 4, 1869 ; off Cojima, near Havana ; 450 fathoms.

Rhizocrinus lofotensis. •
Atelecrinus cuhensis.
Antedon.

March 10, 1869 ; off Double-headed Shot Keys ; 315 and 471 (?) fathoms.

Pentaxirinus decorus (stem only X).

SS. "Hassler," 1871-72.

December 30, 1871. Off Sandy Bay, Barbados ; 100 fathoms.

Rhizocrinus rawsoni. Also off Havana, Montserrat, Guadeloupe,
Dominica, Martinique, and Panama. Dredged in
the East Atlantic by the " Porcupine," Challenger,
"Talisman," and " TravaiUeur."

SS. "Blake," 1878-80.

Cruise of 1877-78.

Station 21. Off Bahia Honda, Cuba ; lat. 23° 2' N., long. 83° 13' W. ; 287 fathoms.

Pentacrinus decorus. Also off Havana, Montserrat, St. "Vincent
Barbados, Kingston, and Saba Island.

380 THE VOYAGE OF H.M.S. CHALLENGER.

Station 22. Off Bahia Honda, Cuba; lat. 23° 1' N., long. 83° 14' W. ; 100 fathoms;

bottom temperature, 71° F.

Holopus rangi. Also off Barbados (Rawson) and Montserrat.

Station 29. Lat. 24° 36' N., long. 84° 5' W. ; 955 fathoms ; bottom temperature,

391° F.

Rhizocrinus lofotensis.

Station 32. Lat. 23° 32' N., long. 88° 5' W. ; 95 fathoms.

Rhizocrinus rawsoni.
Antedon.

Station 35. Lat. 23° 52' N., long. 88° 58' W. ; 804 fathoms ; bottom temperature,

401° F.

Rhizocrinus lofotensis.

Station 43. Lat 24° 8' N., long. 82° 51' W. ; 339 fathoms.

Rhizocrinus lofotensis.
Atelecrinus halanoides.

Station 44. Lat. 25° 33' N., long. 84° 35' W. ; 539 fathoms; bottom temperature,

391° F.

Rhizocrinus lofotensis.

Station 56. Off Havana; lat. 22° 9' N., long. 82° 21' 30" W. ; 175 fathoms.

Rhizocrinus lofotensis.

Pentac7'inus decorus.

Pentacrinus mulleri. Also off Santiago de Cuba, Barbados,

Guadeloupe, Martinique, St. Lucia, St. Vincent,

Montserrat, Saba Island.
Antedon.

Station 57. Off Havana; lat. 22° 9' 15" N., long. 82° 21' W. ; 177 fathoms.
Pentacrinus decorus.
Antedon.

Cruise of 1878-79.

Station 100. Off Morro Light; 250 and 400 fathoms. [

Pentacrinus decorus.
Pentacrinus miilleri.

REPORT ON THE CRINOIDEA. 381

Station 101. Off Morro Light; 175 and 250 fathoms.

Pentacrinus decorus.
Pentacrinus mulleri.
Antedon.

Station 155. Off Montserrat ; 88 fathoms; bottom temperature, 69° F.

Rhizocrinus raivsoni.
Pentacrinus decorus.
Antedon.
Actinometra.

Station 156. Off Montserrat ; 88 fathoms; bottom temperature, 69° F.

Pentacrinus decorus.
Antedon.
Actinometra.

Station 157. Off Montserrat ; 120 fathoms.

Holopus rangi. Also off Barbados (Rawson) and Bahia Honda(Cuba).

Pentacrinus asterius. Also off Saba Island.

Pentacrinus hlakei. Also off Barbados and Martinique.

Pentacrinus decorus.

Pentacrinus miilleri.

Antedon.

Actinometra.

Station 166. Off Guadeloupe ; 150 fathoms; hard bottom; bottom temperature, 5 9f° F.

Rhizoci'inus raivsoni.

Station 171. Off Guadeloupe ; 183 fathoms; bottom temperature, 55^° F.

Pentacrinus miilleri,
Actinometra.

Station 177. Off Dominica; 118 fathoms; fine sand and broken shells; bottom

temperature, 65° F.
Rhizocrinus raivsoni.
Actinometra.

Station 193. Off Martinique ; 169 fathoms; fine sand, dark mud, and shells; bottom

temperature, 51° F.
Pentacrinus mulleri.
Antedon.
Actinometra.

382 THE Voyage of h.m.s. challenger.

Station 209. Off Martinique ; 189 fathoms; bottom temperature, 49f° F.

Pentacrinus blaJcei. Also off Barbados and Montseri;at.

Station 211. Off Martinique ; 357 fathoms ; fine yellow sand and broken shells.
Rhizocrinus rawsoni.

Station 218. Off St. Lucia; 164 fathoms; bottom temperature, 56° F.
Pentacrinus miilleri.

Station 233. Off Milligan's Key, St. Vincent; 174 fathoms; rocky bottom; bottom

temperature, 49|° F.

Pentacrinus decorus.
Antedon.

Station 238. Off Grenadines ; 127 fathoms ; fine coral sand ; bottom temperature, 56° F.

Rhizocrinus lofotensis.

Station 248. Off Grenada; 161 fathoms; fine grey ooze ; bottom temperature, 53^° F.

Rhizocrinus lofotensis.
Actinometra.

Station 259. Off Grenada; 159 fathoms; bottom temperature, 53^° F.

Rhizocrinus lofotensis.
Antedon.
Actinometra.

Station 269. Off St. Vincent; 124 fathoms; bottom temperature, 57|° F.
Pentacrinus decorus.
Pentacrinus miilleri.
Antedon.
Actinometra.

Station 273. Off Barbados ; 103 fathoms ; coral and broken shell, yeUow ; bottom

temperature, 59^° F.
Rhizocrinus rawsoni.
Actinometra.

Station 274. Off Barbados ; 209 fathoms ; fine sand and ooze ; bottom temperature,

53|° F.

Rhizocrinus lofotensis.
Pentacrinus miilleri.

REPORT ON THE CRINOIDEA. 383

Station 277. Off Barbados; 106 fathoms; hard rocky bottom; bottom temperature,

58° F.
Rhizociinus rawsoni.
Actinometra.

Station 280. Off Barbados; 221 fathoms; Globigerina sand; bottom temperature,

50^° F.
Pentacrinus miilleri.

Station 283. Off Barbados ; 237 fathoms; hard bottom; bottom temperature, 49° F.

Pentacrinus miilleri.
Antedon.

Station 290. Off Barbados ; 73 fathoms ; coarse coral sand and broken shells ; bottom

temperature, 70f° F.

Rliizocrinus rawsoni.

A^itedon.

Actinometra.

Station 291. Off Barbados ; 200 fathoms ; flat calcareous stones ; bottom temperature,

491° F.
Pentacrinus hlahei. Also off Martinique and Montserrat.
Pentacrinus miilleri.

Station 295. Off Barbados ; 180 fathoms ; hard bottom ; bottom temperature, 50|' F.

Pentacrinus blakei.
Pentacrinus miilleri.

Station 296. Off Barbados ; 84 fathoms ; hard bottom ; bottom temperature, 61^° F.

Rhizocrinus rawsoni.
Pentacrinus decorus.
Pentacrinus miilleri.
Actinometra.

Station 297. Off Barbados; 123 fathoms; calcareous stones; bottom temperature,

561° F.

Rliizocrinus rawsoni.

Antedon.

Actinometra.

384 THE VOYAGE OF H.M.S, CHALLENGEE.

Station 298. Off Barbados; 120 fathoms; broken shells and coral; bottom tempera-
ture, 61° F.
Pentacrinus decorus.
Antedon.
Actinometra.

Station V. Santiago de Cuba ; 388 fathoms.
Pentacrinus decorus.

Cruise of 1879-80.
Off Kingston (Jamaica) ; 100 fathoms.

Pentacrinus decorus.

Station 306. Lat. 41° 32' N., long. 65° 55' W. ; 524 fathoms.
Rhizocrinus Iqfotensis.

Other Localities in the Caribbean Sea, the Gulf of Mexico, and in

THE Atlantic Ocean.

U. S. Fish Commission, 1882. No. 1124, S.S.E. off Nantucket; 640 fathoms.

Rhizocrinus lofotensis.

SS. " Investigator," Captain E. Cole.

Saba Bank ; 200 fathoms.

Rhizocrinus rawsoni.

Saba Island ; 320 fathoms.

Pentacrinus asterius.
Pentacrinus decorus.

Between Saba and Eustatius Islands; 531 fathoms.
Pentacrinus mulleri.
Pentacrinus decorus.

Fifteen miles N.E. from Panama ; 300 fathoms.
Rhizocrinus rawsoni.

South side of Porto Rico ; 667 fathoms.

Pentacrinus decorus.

REPORT ON THE CRINOIDEA. 385

Dr. Schramm, Guadeloupe.

Fentacrinus asterius.

W. Stimpson, Cuba.

Pentacrinus decorus.

Sir Rawson Rawson, Barbados.

Pentacrinus milUeri.
Pentacmms decorus.
A ntedon.
Actinometra.
B. Holopus rangi.

A List of the known Living Species of Stalked Crinoids, showing their
Bathymetrical and Geographical Distribution.

Explanation of the Letters used.

A. Species discovered by the " Blake."

B. Previously known species collected by the " Blake " and the U. S. Coast Survey

Expeditions.

C. Species discovered by the Challenger.

D. Previously known species collected by the Challenger.

E. Species discovered by the ships of the Eastern Telegraph Company.

F. Species discovered by Dr. L. Doderlein.

I. Previously known species collected by the telegraph steamer " Investigator,"

Captain E. Cole.
J. Previously known species collected by the " Josephine " (Swedish).
N. Species discovered by the " Voringen " (Norwegian).
P. Species discovered by the " Porcupine."
Q. Previously known species collected by the " Lightning," " Porcupine," and " Knight

Errant."
R. Previously known species collected for Sir Rawson Rawson.

T. Previously known species collected by the " Talisman " and " Travailleur " (French).
V. Species discovered by the "Vega" (Swedish).
X. Species discovered by Mr. Damon's collectors.
Y. Previously known species obtained by Mr. Damon's collectors.

(ZOOL. CHALL. EXP. — PART XSXII. — 1884.) li ^^

386

THE VOYAGE OF H.M.S. CHALLENGER.

How

obtained.

Range in
Depth.

Principal Localities.

HoLOPiD.!:, Koemer,

Fathoms.

...

Holopus, d'Orb.,

...

...

H. rang!, d'Orb.,

B. R.

100-120

Caribbean Islands.

IlYOCRiNiDiE, Carpenter,

...

...

Hyocrinus, Wyv. Thoms., .

...

H. hetheUianus, Wyv. Thorns., .

C.

1600-1850
2325!

Mid- Atlantic, Station 106 (stem-fragments);
Southern Ocean, Station 147; East Pacific,
Station 223 (young!).

BocRGUETiCRiNiD^, de Loriol,

...

Bathycrinus, Wyv. Thorns.,

...

...

B. aldrichianus, Wyv. Thoms., .

C.

1375-1600

Southern Ocean, Stations 146, 147.

B. camphellianus. Carp., .

C.

1850

Mid-Atlantic, Station 106.

B. carpenter! , Dan. & Kor.,

K

1050-1495

North-East Atlantic.

B. (jracilis, Wyv. Thoms.,

P. T.

1280-2435

East Atlantic.

Rhizocn'nus, Sars,

...

...

a. lofotensis, Sars, .

B. D. J. Q.

80-955
1900!

Lofoten Islands; Faroe Channel; North-East
Atlantic ; Josephine Bank. Atlantic Coast
of U.S. (Northern States); Straits of Florida;
Caribbean Islands; South- West Atlantic,
Station 122c and Station 323 (?).

R. rawsoni, Pourt., .

B. D. I.
P.T.

73-1280

East Atlantic, Station 76 ; Straits of Florida ;
Caribbean Islands ; Yucatan Bank ; off
Panama.

Pentacrinidjj, d'Orb.,

...

Pentachnus, ^liller.

...

P. alternicirrus, Carp.,

C.

500-600
375?

Pacific — near the Kermadecs, Station 171; and
near the Meangis Islands, Station 214; ofl' the
Panglao and Siquijor Islands, Station 210 (?).

P. aster ius, Linn., .

B. I. Y.

80-320

Caribbean Islands.

P. Uakei, Carp.,

A.

120-200

Caribbean Islands.

P. deeorus, Wyv. Tlioms.,

B. I. X.

84-667

Caribbean Islands ; Straits of Florida.

REPORT ON THE CRINOIDEA.

387

How

obtained.

Range in
Depth.

Principal Localities

p. madearanus, "VVyv. Thorns.,

C.

Fathoms.
350

South-West Atlantic, Station 122.

P. mollis, Carp.,

C.

565

North- West Pacific.

P. mulleri, Oerst., .

B. I. R. Y.

84-531

Caribbean Islands ; Straits of Florida.

P. iiaresianus, Carp.,

C.

500-1350
375?

Pacific — neartheKermadecs, Stationsl70, 170a,
171; near Fiji, Station 175; and near the
Meangis Islands, Station 214; off the Panglao
and Siquijor Islands, Station 210 (?).

P. zvyville-thomso7u, JefTr.,

P. T.

740-1095

East Atlantic.

Metacrinus, Carp., .

...

...

M. angulatus. Carp.,

C.

140

Ki Islands, Station 192.

M. cingulatus, Carp.,

C.

140

Ki Islands, Station 192.

M. costatus, Carp.,

C.

500

Pacific, near the Meangis Islands, Station 294.

M. interru-ptus, Carp.,

C.

95

Philippine Islands, Station 209.

M. moseleiji, Carp., .

C.

500

Pacific, near the Meangis Islands, Station 214.

M. murrmji, Carp., .

C.

?

Western Pacific, or East Indian Archipelago.

M. nolilis, Cai'p.,

C.

140

Ki Islands, Station 192.

M. 7iodosus, Carp., .

C.

630

Pacific, near the Kermadecs, Station 170a.

M. rotund us, Carp.,

F.

70

Japan.

M. Stewart i, Carp., .

E.

?

Singapore.

M. superhus, Carp., .

E.

1

Singapore.

M. tuberosm, Carp.,

C.

140

Ki Islands, Station 192.

M. varians, Carp., .

c.

500

Pacific, near the Meangis Islands, Station 214.

M. wymllii. Carp., .

c.

500-630

Pacific — near the Kermadecs, Station 170a;
near the Meangis Islands, Station 214.

Metacrinus, sp., .

V.

65

Japan.

388

THE VOYAGE OF H.M.S. CHALLENGER.
Analysis of the above List.

Genus.

Number of
Living
Species.

" Porcupine."

Challenger.

"Blake."

New
Species.

Species

previously

known.

New
Species.

Species

previously

known.

New
Species.

Species

previously

known.

Holoptis,
Hyocrinus, .
Bathycrinus, .
Kliizocrinus, .
Pentacrinus, .
Metacrinus, .

1
1
4
2

9
15

1
1
1

11

1

2

4
11

2

1

1

2
3

Total known species,

32

3

1

18

2

1

6

It thus appears that (1) the cruises of the " Porcupme " in 1869-70 added one genus
and three species to the three genera and five species previously known. (2) More tlian
half of the number (thirty-two) of knowTi species, including two new genera, were discovered
liy the Challenger. (3) More than 65 per cent, of the known species have been dis-
covered by H.M.SS. "Porcupine" and Challenger; while the small number of genera
known before 1869 has been doubled by the explorations of these two ships.

BATHYMETRICAL TABLES.

A Roman numeral opposite the name of a species shows that it also occurs in one
of the other Tables.

Table I. — Species found at depths down to 100 fathoms.

Pentacrinus milllen, II., III., IV.
Metacrinus interruptus.

rotundus.
Metacrinus sp. (Vega).

Holopus rangi, II.

Rhizocrinus lofotensis, II., III., IV. V., VI.(?)

rawsoni, II., III., V., VI.
Pentacrinus asterius,^ II., III.

decorus, II., III., IV.

^ Dredged by the "Lightning" in 1868, if not by the "Porcupine" in 1869. See Proc. Roy. Soc. Edin., vol. xii.
p. 356, 1884.

- Two specimens in the Bristol Museum are believed to have been taken at a depth of 50 to 80 fathoms (Austin) ;
and the earliest known examples must certainly have been obtained at depths above 100 fathoms.

REPORT ON THE CRINOIDEA.

389

Table II. — Species found at depths between 100 and 250 fathoms.

Pentacrinus mulleri, I., III., IV.
Metacrinus angulatus.
cingulatus.

Holopus rangi, I.

Rhizocrinus lofotensis, I., III., IV., V., VI. (?)

rawsoni, I., III., V., VI.
Pentacrinus astenus, I., III.

blakei.

decorus, I., III., IV.

mnnxvyi [V)}

nobilis.

tuherosus.

Table III. — Species found at depths between 250 and 500 fathoms.

Rhizocrinus lofotensis, I., II., IV., V^., VI. (?)

rawsoni, I., II., V., VI.
Pentacrinus alternicirrus, IV.

asterius, I., II.

decorus, I., II., IV.

viacleai^anus.

miilleri, I., II., IV.

Pentacrinus naresianus, VI.
Metacrinus costatus.

moseleyi.

murrayi (?).^

varians.

wyvillii, IV.

Table IV. — Species found at depths between 500 and 700 fathoms.

Rhizocrinus lofotensis, I., II., III., V., VI. (?)
Pentacrinus alternicirrus. III.

decorus, L, II., III.

mollis.

Pentacrinus miilleri, I., II., III.
Metacrinus nodosus.

loyvillii, III.

Table V. — Species found at depths between 700 and 1200 fathoms.

Rhizocrinus lofotensis, I., II., III., IV., VI. (?)
rawsoni, I., II., III.

Bathycrinus carpenteri, VI.
Pentacrinus tvyville-thomsoni.

Table VI. — Species found at depths between 1200 and 2000 fathoms.

Rhizocrinus lofotensis (?),' I., II., III., IV., V. Bathycrinus carpenteri, V.

rawsoni, I., II., III., V. gracilis, VII.

Bathycrinus aldrichianus.
Campbell icmus.

Hyocrinus hethelliamis, VII.
Pentacrinus naresianus, III.

Table VIL— Sptcies found at depths between 2000 and 2500 fathoms.
Bathycrinus gracilis, VI. | Hyocrinus hethellianus (?),* VI.

1 If at Station 192 ; 140 fathoms.
' If at Station 323 ; 1900 fathoms.

2 If at Stations 210 or 214 ; 375 and 500 fathoms
* If at Station 223 ; 2325 fathoms.

390

THE VOYAGE OF H.M.S. CHALLENGER.

B.

C.

Summary.

I. 9 species found at depths down to 100 fathoms.

11. 11 (or 12) species found between 100 and 250 fathoms.

III. 13 (or 12) species found between 250 and 500 fathoms.

IV. 7 species found between 500 and 700 fathoms.
V. 4 species found between 700 and 1200 fathoms.

VI. 8 sjiecies found between- 1200 and 2000 fathoms.

VII. 2 species found between 2000 and 2500 fathoms.

1 species descending from Table

I
2
2
2
1
1
1

9 species found at depths down to 100 fiithoms.
14 (or 15) „ „ 250

22 „ „ 500

to

II.

to

III.

to

IV.

to

VI.

III.

to

IV.

III.

to

VI.

V.

to

VI.

VI.

to VII.

To these we may fairly add the two species of Metacrinus found near Singapore by
one of the ships of the Eastern Telegraph Company, as it is improbable that they were
living at a greater depth than 500 fathoms.

It has been the custom of late years, more especially since the discovery of Penta-
crinus wyville-thomsoni and Bathycrinus gracilis by the "Porcupine," at 1095 and 2435
fathoms respectively, to regard the Stalked Crinoids as pre-eminently abyssal types, and
as probably forming " rather an important element in the abyssal fauna." ^ It will be
apparent, however, from the statistics given above, that subsequent researches have not
altogether confirmed these views. If, as Mr. Agassiz has done in his Eeport on the
Echinoidea,^ we take the limit of the continental line at 500 fathoms, it appears from
Summary C, that twenty -four of the thirty- two recent species of Stalked Crinoic's, or
75 per cent., occur within this limit ; while nine of these, living at depths less than 100
fathoms, may be called littoral.

Although numerous and extensive dredgings have been carried on in the abyssal zone
{i.e., at depths below 500 fathoms) during the last fifteen years, Stalked Crinoids have

' The Depths of the Sea, p. 455. - Zool. Chalh Exp., part ix. p. 222.

REPORT ON THE CRINOIDEA.

391

only been found at thirty -four Stations. These have yielded thirteen species, representing
five different genera, as shown in the following lists.

Table shoiving the Number of Times that Stalked Crinoicls have been Dredged

in the Abyssal Zone.

Rhizocrinus alone,
Pentacrinus alone,
Bathycrinus alone,
Hyocrinus alone,
Bathycrinus and Hyocrinus,
Pentacrinus and Metacrinus,
Rhizocrinus and Bathycrinus,
N.B. — All the doubtful cases are included in this list.

15
8
6

1 V =34
2
1
1

Table showing the Species of Stalked Crinoids which occur in the Abyssal Zone.

Genus.

Species confined to the
Abyssal Zone.

Continental Species

occurring in the

Abyssal Zone.

Littoral Species

occiu'ring in the

Abyssal Zone.

Bathycrinus, ■

Hyocrinus,
Pentacrinus, .
Metacrinus, .
Rhizocrinus, ■

-

' aldricldanus.
camphellianus.
carpenteri.
gracilis.

bethelliamcs.

' mollis.

[ wyville-ihomsoni.

nodosus.

i altemicirrus.
{ naresianm

wyvillii.

( lofotensis.
( rawsoni.

8

3

2

392 THE VOYAGE OF H.M.S. CHALLENGEK.

Thus, then, thirteen species have been found in the abyssal zone, two of which are also
littoral, while three are continental. The two former both belong to the genus Rliizociinus,
of which no exclusively abyssal species are known ; though it has been met with at sixteen
out of the thirty-four Stations in the abyssal zone. It is well represented in the Lower
Tertiaries, and perhaps ranges back to the Cretaceous period, when its larger ally
Bourgueticrinus was so abundant.

Pentacrinus has been found at nine Stations where the depth exceeded 500 fathoms ;
and two of its four abyssal species are also continental. But on the other hand,
Bathycrinus, which occurs in the Atlantic at nine abyssal Stations between lat. 65° N.
and 46° S., has never been found at a less depth than 1050 fathoms ; while it embraces
four out of the eight species which are peculiar to the abyssal zone.

No fossil Bathycrinus is known, hoR^ever, and the genus has no special affinities
except with Rhizocrinus, of which it may almost be said to be the "benthal" ^ representative.
Of the four remaining abyssal species, one is the sole representative of the remarkable
genus Hyocrinus, and has only been met with at 1600 fathoms and still greater depths.
Like the Comatulid genus Thauinatocrinus, which occurs at 1800 fathoms in the Southern
Ocean, it has certain strong points of resemblance to the Palseocrinoids.

Pentacrinus ranges back to the Trias and Rhizocrinus to the Eocene or Upper
Cretaceous. But they are both abundant at depths of less than 100 fathoms, Pentacrinus
occurring in the Pacific and in the East Indian Archipelago, as weU as in the Atlantic and
among the Caribbean Islands ; while Rhizocrinus, though limited to the eastern
hemisphere, ranges through over 100° of latitude.

In spite, therefore, of the existence of a few characteristic abyssal types, it is some-
what of an exaggeration to speak of the Stalked Crinoids as a group " on the verge
of extinction," of which a few survivors may occasionally be discovered in the deeper
parts of the great ocean basins.

^ Dr. Gwyn Jeffreys has suggested that this word be employed to denote depths exceeding 1000 fathoms. See
hisadcbess to the Biological Section at the Plymouth Meeting of the British Association, 1877, p. 79.

APPENDIX.

NOTE A.

(Pages 2, 157, and 168.)

On the Homologies of the Cpjnoidal Calyx m the other Echinoderms.

The relations of the plates which enter into the composition of the calyx of a Crinoid
to those constituting the apical system of an Urchin or a Starfish are still the subject of
much discussion.

The apical system of an Urchin, as described by Loven,^ consists typically of a dorso-
central plate enclosed ■svdthin two rings of five plates each. The plates of the proximal
ring (genitals, Auct.) are interradial in position, while those of the outer ring (oculars) are
radially situated. Believing that the dorsocentral is represented in the Pentacrinoid
larva of a Comatula by the terminal plate at the base of the stem,"^ I have suggested that
the basals of the Pentacrinoid are homologous with the so called genitals of an Urchin ;
while the ocular plates of the latter correspond to the radials of the Crinoid. So far as the
basals and radials are concerned this view is fundamentally the same as that previously
expressed by Loven and Agassiz, allowance being made for the difference between their
terminology and that employed by myself; so that the homology between the basals and
radials of a Crinoid and the two rows of plates surrounding the dorsocentral of an
Urchin has been pretty generally accepted.

Dr. R. Hoerues,^ however, objects to this homology (which he associates with my
name) on the ground that the resemblance should be most marked in the geologically
oldest tj^ies, " was keineswegs der Fall ist, da gerade der ScheitelapjDarat der Pala3chinoiden
durchaus nicht an die Basis-Entwickelung der Crinoiden-Kelches sich anschliesst."
This objection may be answered in two ways. 1. Loven* says that the general formula
of the apical system of Echinoidea " has remained, more or less altered, but always recognis-
able, from Palaeozoic to recent time." 2. The positive fact that the basals and radials of

1 Etudes sur les Ecliinoiclees, K. Sve7isk. Vetensh. Akad. Handl., Bd. xi., 1874, No. 7, p. 65.

2 Quart. Joum. Micr. Set., 1878, vol. xviii., N. S., p. 374.

3 Elemente der Palteontologie (Palceozoologie), Leipzig, 1884, p. 130.

* On Pourtalesia, a genus of Echinoidea, K. Svendc. Vetcnsk. Akad. Handl., Bd. xix., 1883, No. 7, p. (i2.
(ZOOL. CH.\LL. ESP. — PART SXSII. — 1884.) li 50

394 THE VOYAGE OF H.M.S. CHALLENGER.

the Crinoid larva have precisely the same relation to the vaso-peritoneal system as the
corresponding plates (genitals and oculars) of an Urchin is a strong reason for not laying too
much stress upon the negative evidence of a confessedly imperfect palseontological record.^

In one respect it is somewhat unfortunate that the Urchins should have been selected
as affording the tj^ical apical system of the Echinozoa, wdth which that of a Crinoid
could be compared. For their apical system is primitively a comparatively simple one ;
whereas many Crinoids have a ring of plates immediately beneath the basals which are
unrepresented in the Urchins, though present in many Ophiurids and Asterids.

In his endeavour to find an early Crinoidal form with a calyx of the same simple
description as the apical system of an Urchin, Professor Loven* was led to select the
genus Cyathocrinus (Poteriocrinus) ; and he proposed the following homologies between
the two types : —

L Dorsocentral of Urchin = tlie five under-basals of Cijathocrinus (the basals of J. Miiller).

2. Genitals of Urchin = the five basals of Cyatliocrimis (the parabasals of J. Miiller).

3. Ocular plates of Urchin = radials of Cyathocrinus.

The two last of these propositions have been generally, but not universally, accepted.
As regards the first, however, I am sorry 'to say that I have found myself unable to
agree with Professor Loven.

1 pointed out six years ago^ that the under-basals of Cyathocrinus constitute an
element in the calyx which is by no means so constant in its occurrence as it should be,
were it a fundamental part of the apical system and homologous with the dorsocentral of
an Urchin or Starfish. Under-basals are present in Encrinus, Extracrinus, and Marsu-
pites among the Neocrinoids, and in Cyathocrinus, Poteriocrinus, Rhodocrinus, and a
large number of allied genera among the Palteocrinoids ; while they are absent in Apio-
crinus, Pentacrinus, Actinocriniis, Platycrinus, and in many other less known genera.
When present, there are generally five distinct plates, resting on the upper stem-joint ;
and this fact, together with the "want of constancy in their occurrence, caused me to
suspect that they could not be collectively homologous to the primitively single dorso-
central plate of an Urchin or Starfish, as supposed by Loven. I was therefore led to
seek for the homologue of this last in the terminal plate at the end of the stem of the
Pentacrinoid larva, which occupies the same j)osition wath regard to the right peritoneal
tube as the dorsocentral of a larval Urchin or Starfish. This suggestion has been
accepted by Liitken and by Sladen, as I have pointed out above (p. 168), though it is
altogether ignored by Loven. But no serious arguments have been yet brought forward
against it by other authors who have discussed the question ; w^hUe, on the other hand,

' Dr. Hoenies does not appear to have gone into the sulyect very deeply. I have nowhere suggested that the
radials of an Urchin are homologous with the basals of a Crinoid ; nor that the madrejiorite of Clypeaster is compar-
able to the centro-dorsal of Comatula and to the central plate in the ealj'x of Manupites. Nevertheless Hoernes thinks
fit to express his dissent from these views, which have originated with no one but himself, and he entirely misses the
real point at issue.

2 Etudes, loc. at, p. 80. 3 Quart. Journ. Mkr. Sd., 1878, vol. xviii., N. S., pp. 358-361.

EEPORT ON THE CRINOIDEA. 395

several recent researches have supplied further cogent reasons for rejecting the homologies
which Loven seeks to establish between the dorsocentral of an Urchin or Starfish and the
uuder-basals of a dicychc Crinoid.

Six years ago the numerous modifications of tile apical system which arc presented
by Asterids and Ojjhiurids had received comparatively little attention ; and I was
therefore led to regard the under-basals of Encrinus, Extracrinus, and of the Palaeozoic
Crinoids as " additional elements which occur in the apical system of some Crinoids,
while they are unrepresented in other members of the order and in the other Echino-
derms."^ Four years later, however,^ I was able to show that the apical system of
the young Amphiura squamata, which had been recently described by Ludwig,'
corresponded precisely with that of Marsupites, the type which was first selected
by Loven for comparison with Salenia. Both in Amphiura and in Marsupites there
is a central abactinal plate representing the dorsocentral of an Urchin. Next to this
come, not the interradial plates corresponding to the genitals of an Urchin and the
basals of Cyathocrinns, as Loven formerly supposed,* but a ring of radially situated
plates which correspond to the under-basals of Cyathocrinus, but are not represented
at aU in the apical system of an Urchin, as at present known. Outside these come the
interradial basals (genitals) and then the radials (oculars). Ludwig discovered that
the latter remain on the disk of Amphiura, and are not carried away from it by the
growing arms as had been generally supposed.

Having discovered, as I believed, the homologues of the under-basals of a Crinoid
in a larval Ophiurid, I naturally began to seek for them in the adult members of
the class ; and it soon appeared that they were represented in the rosette of primary
plates which occupies the centre of the disk in certain species of Ophioglypha,
Ophioceramis, OpMomusium, and Ophiozona.^ At the same time two important
discoveries bearing on this question were made by Sladen.'' (1) The radial plates
of the larval Asterid remain on the disk, like those of the Ophiurid, and are not
carried outwards by the growing arms, as was formerly supposed. (2) In the late larvae
of Zoroaster fulgens, Asterina gihhosa, Asterias ruhens, Asterias glacialis, and other
species, the so called genital plates ( = basals of a Crinoid) are separated from the
dorsocentral by a ring of radial plates which occupy exactly the same position as the
under-basals of Marsupites, and the corresponding plates in the Ophiurids mentioned

I Qmrt. Journ. Mier. Sci., 1878, vol. xviii., N. S., p. 374. 2 ji^d^ i882, vol. xxii. p. 380.

3 Ziir Entwicklungsgeschichte des Ophiureiiskelettes, Zeitschr.f. iciss. Zool.,BA. xxxvi. 1882, pp. 181-200, Tafn. x., .\i.

* Lovin appears to have been so far influenced by my criticisms on his comparison of the radially placed
imder-basals of Marsupites with the interradial genitals of Salenia that he makes no further reference to the former
type, although in his earlier " i^tudes " he laid great stress upon its resemblance to Salenia. This is unfortunate,
because the presence of a dorsocentral in Marsupites, as well as of under-basals homologous with those of Cyathocrinm,
proves conclusively that the latter cannot represent the dorsocentral of Marsupites, and therefore of Salenia, as Lov^n
formerly supposed.

5 Quart. Journ. Micr. Sci., 1884, vol. xxiv., N. S., p. 11. " Ibid., pp. 29-34.

396 THE VOYAGE OF H.M.S. CHALLENGER.

above. Further, these plates "are large and well-developed in the adult forms in
Pentagonaster semilunatus, Gymnasteria carinifera, in various species of Pentaceros,
and a large number of the Goniasteridee."

It thus appears that a small nuniber of Ophiurids and a larger number of Starfishes
have two rings of plates between the radials and the dorsocentral, just as is the case in
Marsupitcs ; though in the majority of the members of both groups there is only one
ring, the elements of which are interradial, as is the case in all the Urchins, so far as
is yet known. The homology of these interradial plates with the basals of the Penta-
crinoid larva and of Marsupites is a part of Loven's theory ; while he long ago pointed
out the homology between the central abactinal plate of the Starfish larva and the dorso-
central of Marsupites} I do not see therefore, how he can do otherwise than accept
the views of Sladen and myself respecting the homology of the radial plates immediately
surrounding the dorsocentral of Ophiomusium, Amphmra, Asterias, and Zoroaster with
the under-basals of Marsupites. In both cases these radial plates separate the dorso-
central from an interradial series which are called basals in a Crinoid and genitals in
an Urchin or Starfish, i.e., the plates for which Loven proposes the general name of
"costals." I do not imagine that he will deny (1) that the radial plates between the
costals and dorsocentral of Marsupites are homologous with the radial plates between
the costals and dorsocentral of the young Asterias; nor (2) that these radial plates are
homologous with the under-basals of Cyathocrinus, which are immediately within the
basals, or as he prefers to call them, costals. But, according to his present view, these
under-basals of Cyathocrinus represent the dorsocentral of the young Asterias. There
is, however, no reason to seek for the homologue of the five under-basals of a dicyclic
Crinoid in the single dorsocentral of a larval Starfish ; for this dorsocentral is
surrounded by five plates which correspond exactly in their relative positions with
the under-basals of Marsiqntes, and therefore of other dicyclic Crinoids, including
Cyathocrimis.

If I may be permitted to use here an expression employed by Prof. Loven with respect
to another homology which he establishes, I would say that " to anyone believing in the
consistency of Nature's ways, there is no reason whatever for doubting " that the apical
systems of some Asterids and Ophiurids contain plates which are truly homologous with the
under-basals of a Crinoid. It is unfortunate that their presence was not discovered in
time to be noticed by Prof. Loven in his recent discussion of the question; for I am
sanguine enough to believe that it would have led him to reconsider his views respecting
the homology of the five under-basals of a Crinoid with the primitively single dorso-
central of a Starfish.

The interradial plates in the apical system of a Starfish or Urchin are usually known
as the genitals; but this term, " besides being expressive of incidental relations peculiar

' Etudes, loc. cit., pp. 72, 86.

REPORT ON THE CRINOIDEA. 397

to the Ecliinoidea aud partly to the Asteriadea, cannot by any means be applied to the
homologous parts in the Crinoid," i.e., to the basal plates. Loven,' however, objects to the
extension of the latter name to the so called genitals of Urchins and Starfishes, because
the position of their apical plates, " while basal in the Crinoidea, is culminating in the
Echinoidea and the Asteriadea, and consequently any appellation involving the notion of
a basal position must be avoided." ' There is undoubtedly some force in this objection ;
and I have long been endeavouring to find some general expression that would conveniently
describe the interradial plates in the apical system of all Echinoderms. Not having
succeeded in this quest, T have been obliged to fallback upon the word " basals." The
interradial position of these plates in the calyx of a Crinoid, as defined by Miiller,^ is
now universally recognised, and the use of the word is not likely therefore to lead to any
confusion respecting the position of the plates with regard to the general symmetry of
the Echinoderm type ; while it has the further advantage of avoiding the multiplication
of " terms already too numerous." As Loven objects to the use of " basals " for the
reasons given above, he seeks to avoid the introduction of new names by reviving the old
term " costals " of Miller. To this there could be no possible objection were it only
employed in the sense in which it was generally u-sed by Miller, but this is unfortunately
not the case. In seven out of the nine genera which were described by Miller as
having costal plates, this term was used for the radial plates of the calyx ; while it was
only in describing some species (not all) of Cyathocrinus that he employed the term costals,
for the interradial plates of the calyx or basals, and in the case of Marsujntes he gave this
name to the under-basals. Loven admits this inconsistency,* but adds that " it has
always been considered allowable to suggest the use in a strict sense of a term elsewhere
vaguely applied." This is of course quite true, but the term should surely be limited to
that sense in which it was most generally used by its author. This is not the case,
however, with Lovdn's revival of the term " costals," for he employs it to designate the
basal plates which were only called costals by Miller in four out of the many species
described by him ; while he applied this name to the radials in all the other cases in
which he used it at all, except in that of Marsupites. I cannot help feeling, therefore,
that this revival of a name which has been disused for half a century is somewhat
inexpedient, and is likely to lead to a confusion between the radial and interradial
plates of the apical system which it is very desirable to avoid. Every one knows that the
iuteiTadial abactinal plates of an Urchin or Starfish are not situated at the base of its
body as they are in a Crinoid ; and this reservation being made I do not see that there
can be any objection to calling them basal plates. This would avoid all possibility of
any confusion with respect to their position as regards the general symmetry of the

1 On Pourtalesia, he. cit, p. 63.

2 It may lie noted with respect to tLis point thatwlien Loven inverts an Urchin for a lietter comparison of its calyx
■n-ith that of a Crinoid, his " costal " plates (the genitals) really do become " basal " in position (see p. 414).

3 Bau des Pentacrinus, loc. cit, p. 25. ' * On Pourtalesia, he. cit., pp. 63, 64.

398 THE VOYAGE OF H.M.S. CHALLENGER.

Echinoderm, which seems to be inseparable from Lov^n's use of the term. Should Prof.
Loven ever write anything more upon the apical system of Echinoderms, he cannot avoid
referring to the radial plates between the dorsocentral and the so called genitals (costals
or basals) of an Asterid. He cannot speak of them by the name which was given them
by Sladeu who discovered them, viz., " under-basals," for this term would be meaning-
less and confusing unless the plates outside them (the costals of Loven) were also called
basals ; and he would therefore have to invent a new name for them, a proceeding to
which he objects, or else adopt the terminology of Sladen and myself. I am sanguine
enough to hope, not only that this will be the case, but also that the presence in Asterids
and Ophiurids of plates homologous with the under-basals of Crinoids will lead him to
abandon his theory of the homology of these under-basals with the dorsocentral of an
Urchin or Starfish. I little expected six years ago to get so complete a confirmation of
the views I then expressed as the presence of a " dicyclic base " in several Asterids and
Ophiurids as well as in Cyathocrinus and Marsupites. The similarity in structure of
the apical system in all the groups of brachiate Echinoderms thus becomes exceedingly
striking ; and it affords a further proof (if such were needed) of the homology between
the apical systems of the Echinozoa and the Pelmatozoa respectively.

Two authors, however, have been led to an entirely different conclusion respecting
the interradial abactinal plates of the Starfish larva from that of Loven, Agassiz, Sladen,
and myself. Ludwig regards them as homologous with the orals of a Crinoid, because
one of the latter is pierced by the primitive water-tube ; ^ while the madreporite of an
adult Starfish is in relation with one of the so called genital plates (costals or basals).
The morphological diSiculties inseparable from this inversion of the relations between a
Starfish and a Crinoid, as ordinarily conceived, have been discussed by Sladen" and
myself.' With the exception of Studer,* whose errors have been discussed elsewhere,® no
other writer has alluded to the subject ; though it has recently made its appearance in a
somewhat modified form. Perrier stated two years ago " that the primary interradial plates
around the dorsocentral of the young Brisinga develop into the so called odontophores
of the adult. The former are the plates which are usually known as the genitals (costals
of Loven ; basals of Sladeu and myself) ; and if Perrier's statement be correct, the views
of Loven, Agassiz, Sladen, and myself respecting the homology of the apical plates
through the whole group of Echinoderms are no longer tenable. No proofs of it have yet
been offered, however, though in a later note by Perrier " the following passage appears,
"Les jeunes Asteries, les jeunes Brisinga presentent aussi, comme Loven et nous-meme

1 Zeitschr.f. wiss. Zool, Bd. xxxiv. p. 318, 1880.

2 Quart. Journ. Micr. Sci., 1884, vol. xxiv., N. S., pp. 35-40.

3 Ibid., 1880, vol. XX., N. S., pp. 322-329.

* Uebersicht iiber die Ophiurideii welche -wiihrend der Eeise S.M.S. " Gazelle " iim die Erde, 1874-76, gesammelt
\nirden, Ablmndl. d. k. Akad. d. Wiss. Berlin, aus dem Jahre 1882, Phys. Kl. Abh., i. p. 10.
^ Quart. Journ. Micr. Sci., 1884, vol. xxiv., N. S., pp. 15-18.
« Note sur les Brisinga, Comptes rendus, t. scv., 1882, p. 63. ^ Ibid., p. 1381.

REPORT ON THE CRINOIDEA. 399

I'avons etabli, des plaques dorsales disposees au debut, comme celles du calice des
Crinoides ; nous avons demontre que, chez les Brisinga, les plaques de la premiere rang^e
deviennent les odoutopliores." Even yet, however, no figures of the various develop-
mental stages of Brisinga have been published in " demonstration " of Perrier's
statements, which were summarised as follows, " Ainsi les odontophores sont les rcstes
des pieces du premier rang du disque primitif de la Brisinga. L'identitd dvidente du
plan d'organisation des Brisinga et des Asteries proprement dites rend la meme conclu-
sion probable pour les autres Etoiles de mer." It is undoubtedly probable that what is
true of Brisinga also applies to all the other Asterids ; and it is therefore the more
desirable that some proof should be offered of the very definite statements made by
Perrier. They have recently been disputed by Sladen ^ on the ground that in all the
Starfishes of which the embryonic stages are sufficiently known, the basals and odonto-
phores are " separate and distinct, and co-exist independently from their first formation ; "
while he further expresses his belief, based on sound morphological arguments, that the
origin assigned by Perrier to the odontophoi'es of an Asterid is theoretically impossible.

Perrier has recently repeated his statements in somewhat greater detail," and having
compared Loven's figures of the young Asterias glacialis with his young specimens of
Brisinga, he says that he has no doubt whatever, "que les choses se passent de la mSme
facon dans les deux genres, et nous pouvons, dfes lors, affirmer que les pieces radiales {sic)
des tres jeunes AsteriadfB deviennent dans cette ordre de Stellerides les odontophores."

Unfortunately for his theory, however, these interradial abactiual plates of the
young Asterias develop in other Starfishes into relatively large plates which remain in
more or less close relation with the dorsocentral, and are the very plates described as
basals by Sladen not only in the larval Asteiias, but also in the following genera —
Zoroaster,'^ Pentagonaster, Tosia, Astrogonium, Stellaster, Nectria, Ferdina, Pentaceros,
Gymnasteria, and others.

As there is an odontophore on the ventral side in each of these types, it is perfectly

' Q%mrt. Joum. Micr. Sci, 1884, vol. xxiv., N. S., p. 39.

2 Memoire sur les Etoiles de Mer recueillies clans la mer des Antilles et le Golfe du Mexique durant les expeditions
de dragage faites sous la direction de M. Alexandre Agassiz, Nouv. Arcliiv. du Mus. cVHist. Nat, S"" ser., 1884, t. vi. p. 159.

3 Several months before the appearance of Perrier's Eeport upon the AVest Indian Starfishes, Sladen figured the
apical system of Zoroaster fidgens, and described it in the following terms : "Surrounding a dorsocentral and five small
radially placed plates are five large plates iutemulial in position ; and outside and alternating with these are five
similar but rather smaller radially placed plates. ... It will be noted that these plates represent in a remarkable manner
the dorsocentral, the under-basals, the basals, and the radials respectively of the Crinoid calyx " (Asteroidea dredged in
the Faroe Channel during the cruise of H.M.S. " Triton " in August 1882, Trans. Roy. Soc. Edin., vol. xxxii. p. 160,
figs. 9, 11). Precisely the same arrangement appears in the apical system of Zoroaster acMeyi, so far as one can judge
from Perrier's figure of an entire specimen (Nouv. Archiv. du il/its. d'Hist. Nat., 2™" ser., 1884, t. vi. pi. iii. fig. 1); but he
makes no mention of Zoroaster fulyens. Even if he had not seen Sladen's reference to it, one would have thought that
he would have been struck liy the Crinoidal aspect of Zoroaster ackleyi, though he does not refer to it at all, and he
gives no detailed description of the plates. It would be interesting to know his reasons for believing that the large
interradial plates in the immediate neighljourhood of the dorsocentral are not the " plaques de la premiere rang(5e "
of the larva, which occupy exactly the same position with reference to the dorsocentral, and are believed by Perrier
to become the odontophores in all Starfishes except Caulaster.

400 THE VOYAGE OF H.M.S. CHALLENGER.

clear that this structure cannot have been developed from the primary iuterradial plates
in the abactinal system of the larva ; for these last remain in the apical system, just as
they do in the Urchins. Sladen's observations, to say nothing of those of Loven, render
Perrier's views respecting the development of the odontophores of an Asterias from the
primary iuterradial plates round the dorsocentral of the larva, absolutely untenable,
and" one is therefore the less disposed to accept his statements concerning Brisinga, of
which no proof has yet been offered to his fellow-workers.

In connection with this subject he has recently advanced some theories respecting
the mutual relations of a Crinoid and an Urchin which are altogether at variance with
those of most other naturalists, except perhaps Ludwig. He thinks that in comparing
the apical system of an Urchin with the calyx of a Crinoid, Loven " a attribue a I'Oursin
une position exactement inverse de sa position normale." ' He regards an Urchin as a
Crinoid with a lai'ge visceral mass to which the arms are fixed, as for example in
Eucahjptocrinus;'^ while "la bouche serait situee au point d'insertion du disque sur
la tige." Under these circumstances the nervous system and ambulacral canals of an
Urchin would have " exactement les memes rapports c|ue ceux qui nous sout offerts par
la Comatule. 11 est a remarquer que prdcisement, en ce point, le calice de nombreux
Crinoides pedoncules s'invagine, et prdsente des plaques qui ne sont pas sans analogic
avec celles qui constituent la lanterne d'Aristote des Oursins et plus particulierement
des Clypeastres." This idea has since been further developed.^ The arms of a Crinoid
grow at their free end, while the new ambulacral plates of an Urchin are formed round
the periproct. The base of the ambulacra is thus in the peristome. "Mais alors les
pieces homologues des plaques calicinales des Crinoides sont non pas les dix plaques du
periprocte, mais bien les pieces constitutives de la lanterne d'Aristote. Quelque bardie

que paraisse cette interpretation nous sommes persuade que tout esprit non

prevenu sera frappd de I'etroite ressemblance d'un Oursin regulier avec des Crinoides tels
que le Callicriims et surtout les Eucalyptocrinus." I fear that in this matter I
cannot be said to have an " esprit non prevenu " ; but it certainly appears to me
somewhat rash to attempt to overthrow the generally accepted ideas respecting the
mutual relations of an Urchin and a Crinoid by reference to such very highly
specialised t}'pes as Eucalyptocrinus and Callicrinus. Both in this respect and in the
comparison of the Crinoidal calyx to the lantern of Aristotle, I cannot help feeling
that Prof. Perrier has altogether lost sight of the embryological arguments by
which questions of homology are generally decided. The calyx of a Crinoid and the
apical system of an Urchin or Starfish have precisely the same relations to the vaso-
peritoneal apparatus of the larval Echinoderm ; and until some better reason can be
adduced for disregarding this relation than a more or less uncertain resemblance between

' Nouv. Archiv. du Mus. (THist. Nat, 2""= ser., 1884, t. vi. p. 161. - Comptes rendus, t. xcviii. p. 1450.

3 XoHV. Archiv. da Mus. d'Hist. Nat., 2°>« ser., 1884, t. vi. p. 161.

REPORT ON THE CRINOIDEA. 401

a stemless Urchin and a highly specialised Palseocrinoid, I think that most naturalists
will be inclined to regard the mouth of an Urchin as representing that of a Crinoid, and
not the point of attachment between the stem and the body.

If this last view be correct, it follows, as Perrier points out, that " Ic dos des Astdries
correspondrait a la region buccale des Oursins et non a leur region anale," which not
even Ludwig would assert. Two years ago Perrier described a small Starfish which
had been dredged l)y the " Travailleur," and was distinguished by the possession of
a small dorsal a])pendage comparable to the stem of a Crinoid.^ He stated that
" quelques caracteres des Asteries dont nous avons k parler ici paraisseut indiquer
que I'appendice dorsal dont elles sont munies est bien reellement I'homologue du
pedoncule des Crinoides." He named the type Caulaster, and added that it is allied to
CtenodisGUS. " II existe chez ces derniers un leger tubercule qui nous paralt homologue
de I'appendice dorsal des Caulaster, et peut-^tre en pourrait-on rapprocher un boutou
saillant qui, chez les Agtropecten, occupe la j)lace oh se trouve I'anus chez les autres
Etoiles de mer." Sladen subsequently pointed out that a central epiproctal prominence
of this kind is very general in the family Astropectinidae.^ It is " frequently developed
into an elongate tubular prolongation " in the subfamily Porcellanasteridse. He doubted
the affinity of Caulaster with Ctenodiscus, and was inchned to regard it as a young
Porcellanaster. More recently Danielssen and Koren ^ have described a new genus
Ilyaster, in which a disk of 30 mm. diameter bears an epiproctal process 8 mm. long
and covered with paxilte, as in the Astropectinidae described by Sladen. They agree
with Perrier in regarding it as homolooous with the stem of a Crinoid; and it would
appear that Agassiz is of the same opinion.* It may be that this view of the case is the
right one ; but it could only be satisfactorily proved to be so by the demonstration that
the cavity of the epiproctal prolongation is derived from the right vaso-peritoneal tube.
For it is a diverticulum of this division of the primitive body-cavity of Comatula which
extends backwards and has the joints of the larval stem developed in its walls. Future
observations upon the early larval stages of the Astropectinidae would throw much light
upon this question. Perrier's Caulaster appears to be the youngest known form pos-
sessing this curious appendage, and some of the plates of the primitive calycular system
are still visible. " A la base de I'appendice dorsal, se trouvent en efi'et quatre grandes
plaques calcaires, disposees en croix et portant chacune un petit piquant ; ces plaques
sont t, peu pres orientees dans la direction des hras ; une cinquieme plaque, alterne
avec deux d'entre elles et opposee a la plaque madreporique, fait evidemment partie du
meme cycle ; cinq autres plaques plus petites viennent se placer dans les angles laissds
libres par les cinq plaques de la premiere rangee. On ne pent manquer d'etre frappe

^ Comptes rendus, t. xcv. p. 1379. = Journ. Linn. Soc. Land. (Zool.),vol. xvii. p. 214.

' Den Norske Nordliavs-Expeditioii, xi.; Zoologie. Asteroidea, p. 101, pi. vii. lig. 16, 1884.

4 Reports on the Results of Dredgins by the U.S. Coast Survey steamer -'Blake" ; Report on the Echini, Mem.
Mils. Comp. Zool, vol. x., 1883, No. 7, p. 17.

(ZOOL. CHALL. EXP. —PART XXXII. 1884.) IlOl

402 THE VOYAGE OF H.M.S. CHALLENGER.

de la ressemblanee absolue de ces dix plaques alternes avec eelles qui forment le periprocte
des Oursins et que Loven a rapprocli^es, a leur tour, de eelles qui constituent le calice des
Crinoides, opinion que nous aurons prochainement occasion de discuter. L'identite de dis-
position des plaques dorsales des Caulaster avec eelles du calice de Crinoides est evidente."
It would appear, however, from tlie foregoing description that Perrier's comparison of
the plates round the dorsal appendage of Caulaster with those forming the periproct of
an Ui'chin cannot be followed out in detail. The first row of plates in Caulaster, if
radially situated as Perrier states, cannot correspond to the first or inner row of the
apical system of an Urchin ; for these last are the genitals, and are situated interradially.
Their homologues are the plates in the second row of Caulaster, which alternate with
those of the first ; while the second ring of plates in the Urchins, the oculars or true
radials, appear not to be represented in Caulaster. If Perrier's description of the positions
of the plates in this type be correct, its apical system consists, not of genitals and oculars
(basals and radials) as in an Urchin, but of under-basals and basals ; and the true radials
must be so small as to have escaped his notice.

NOTE B.

(Page 36.)

On the Basals of Fossil Comatul.^.

In certain fossil Comatulse the ends of the basals are visible on the exterior of the
calyx between the radials and the centro-dorsal. They are sometimes quite small, as in
some forms ot Pentacrimis decorus (PI. XXXIV. fig. 1 ; PL XXXV. figs. 1,2; PL XXXVI.
fig. 3) ; while in other species, such as Solanocrinus scrobiculatus, Miinster, they may
reach a considerable size. As long as basals were supposed to be absent in the calyx of
the recent Comatulse, their presence in fossil forms appeared to be a character of generic
value. But after the discovery by Dr. Carpenter and Sir Wyville Thomson that the
Pentacrinoid larva has true basal plates which eventually become metamorphosed into
the concealed rosette, this distinction between the recent and fossil Comatulse no longer
holds good. Schluter^ recognised this fact in 1878, and pointed out that Solanocrinus
was merely a synonym of de Freminville's name Antedon. He referred to this genus
both Solanocrimis costatus, Goldfuss, and Solanocrinus scrobicidatus, Miinster, together
with two other fossil species, in both of which the basals appear on the exterior of the
calyx. Zittel,^ however, regarded Solanocrinus as a subgenus of Antedon, distinguished
from it l)y the presence of external basals.

In the following year^ the examination of a considerable number of fossil Comatulse

' Zeitschr. d. dcutsch. geol. Geselkch., Jahrg. 1878, pp. 36, 40. ^ Pateontologie, vol. i. p. .396.

2 On some undescribed Comatulai from the British Secondary Rocks, Quart. Journ. Gcol. Soc, vol. xxxvi.
pp. 36-46, 1880.

REPORT ON THE CRINOIDEA. 403

led me to adopt Schliiter's view, as de Loriol * had jireviously done. For I not unfre-
•juently met with calyces in which basals might appear externally at some of the angles,
but not at others ; while in other fossU types no basals were visible at all. In both cases,
however, the basals were present as more or less prismatic rods extending outwards from
the centre of the under surface of the radial pentagon somewhat as in the Peiitacrinus
decorus represented in PI. XXXIV. fig. 8. But they were not always long enough to
reach the edge of the radial pentagon and appear externally between it and the centro-
dorsal ; so that one side of a calyx would be that of an Antedon and the other that of a
Solanocrinus. Under these circumstances it would seem that Schliiter was undoubtedly
right in uniting Solanocrinus with Antedon. But in a Manual of Palaeontology recently
published by Hoernes,^ Zittel's classification is still adopted, and Solanocrimis is placed
as a subgenus of Antedon, differing from it in the presence of basals on the exterior of
the calyx ; while it is also described as represented by a living species and not by fossil
ones only. This apparently refers to the doubtful genus Comaster, Agassiz, which is
only known from the descrijition given of it by Goldfuss.' AVhatever be the nature of
Comaster, however, the supposed difference between Solanocrinus and Antedon cannot
any longer be regarded as of generic value.

NOTE C.

(Page 68.)

On the Excenteic Position of the Mouth in Actinometra.

The genus Actinometra comprises quite two-fifths, if not more, of all the species of
li\'ing Crinoids. The character by which it is most readily distinguished at first sight is
the excentric position of the mouth, as was pointed out in 1877 * and again in 1879 ;^
while its generic position is recognised by Glaus in the last edition of his Grundzuge der
Zoologie with the character "Mund excentrisch" (PL LV. figs. 1,2; PI. LVI. figs. 7, 8;
PL LXI. fig. 2 ; see also fig. 3 on p. 92).

In spite of these facts, however, Hoernes stated in his Palseontology (p. 131) that in
recent Crinoids the mouth, is always (stets) in the centre of the disk, which is very far from
being the case, as explained above. This error was avoided by Zittel, whom Hoernes
usually follows very closely ; though the generic position of Actinometra was not fuUy
recognised by the former author, who placed it along with Solanocrinus and Promacho-
crinus as a subgenus of Antedon. But all subsequent writers, Hoernes excepted, have
recognised that Antedon and Actinometra are totally distinct generic t}^es.

1 Swiss Crinoids, p. 254. • 2 Elemente der Palseontologie, p. 149.

3 Petrefacta GermaniiP, vol. i. p. 202 ; see also Journ. Linn. Soc. Loud. (Zool.), vol. xiii. p. 454, 1877.

* Journ. Linn. Soc. Lond. (Zool.), vol. xiii. p. 441, 1877.

* The Genus Actinometra, T/Tois. Linn. Soc. Loud. (Zool.), ser. 2, vol. ii. p. 18.

404 • THE VOYAGE OF H.M.S. CHALLENGER.

NOTE D.

(Pages 100, 106.)

On the Supposed Communication of the Chambered Organ and
Labial Plexus with the Exterior.

Perrier's statements respecting the direct continuity of the water-tubes depending
from the oral ring of the larva with the inner ends of the water-pores of the disk have
recently been extended to the adult Antedon. He further asserts not only that some of
the water-pores open into the more or less glandular tubules of the labial plexus, but
also that the canals forming the inner ends of the water-pores on the lower j)art of the
disk open into the cavities of the chambered organ. ^

I will not go so far as to deny the truth of these statements ; but can only say that
the results which Prof. Perrier believes himself to have obtained by " 1' etude minutieuse
de plus de deux cents coupes " are far from being in accordance with those of Ludwig,
Greeff, Teuscher, or myself. It seems to me unlikely that the complex relations of the
(■anals forming the inner ends of the water-pores which Perrier describes should have entirely
escaped the notice of all of us. I freely admit that I may have overlooked the connection
of the water-pores with the water-tubes and with the labial plexus ; for the state of preser-
vation of my material has not been such as to yield sections of one-fortieth of a millimeter
thick. But, on the other hand, I have carefully studied many more than two hundred
sections, nearer two thousand in fact, of several different types ; and I believe it to be
impossible that I could have avoided seeing such a connection between the water-pores
and the chambered organ as is described in the follomng sentence, " leur plexus se
continue jusqu'jt I'organe cloisonne dans les cliambres duquel s'ouvrent encore chez
Y Antedon rosaceus les canaux issus des entonnoirs inferieurs du disque."

The chambered organ of a Comatida is lodged within the cavity of the centro-dorsal
basin, covered up by the rosette, and surrounded by the ring of united first radials
(PI. LXr. fig. 2). It is therefore a perfect mystery to me how any of these canals which
lead inwards from the ciliated water-pores and traverse the perisome of the disk can
possibly open into its chambers.

Perrier describes himself as having been the first since the time of Miiller to cbaw

O

attention to these ciliated water-pores ; - and he gives the date of his having done so as
1872.^ In making this claim, however, he entirely ignores the fact that on the 21st
of March 1871 Grimm had communicated a description of them with illustrative figures
to the St. Petersburg Academy.^ His description and figures were published in 1872, and

^ Anatomie des !l6chinodermes ; siir I'organisation des Comatules adultes, Comptes rendus, t. xcviii., No. 23,
1884, p. 1449.

2 Comptes rendus, t. xcviii. p. 1449.

^ Recherches sur I'anatomie et la r^g(5n6ration des bras de la Comatula rosacea. Archives d. Zool. exper., vol. ii. p. 42.

* Zum feineren Bau der Crinoiden, Bull. Acad. Sci. St. Petersb., t. x\-ii., 1872, col. 3-9, Mit elner Tafel.

REPORT ON THE CRINOIDEA. 405

Perrier cannot well be unaware of the fact at the present time, as it is noticed in the text of
Ludwig's work on the Crinoids, and the paper is properly quoted in his bibliography. It
is difficult, therefore, to see on what grounds Perrier bases his claim respecting the re-
discovery of these ciliated funnels, " sur lesquels, dejjuis Johannes Mtiller, j'ai le premier
attire I'attention (1872)." When he published the paper here referred to, he had only
made a superficial examination of these openings, each of which he described as leading
into a small cul-de-sac. It would seem indeed as if he were then unaware, not only of
Grimm's more correct observations of the previous year, which may be readily understood,
but also of Miiller's description of these pores 'published nearly thirty years before ;^ for
he never mentioned Miiller at all, and suggested that the pores might be special sense
organs ! Now, however, he tells us that they have a threefold connection — (l) -with the
water-vascular ring; (2) with the plexus of glandular tubules round the gullet ; (3) with
the ca\'ities of the chambered organ. No other observers have noticed these points, and
Prof. Perrier's proofs of his statements will be awaited with much interest.

NOTE E.

(Page 102.)

On the Intervisceral Blood- Vessels.

Perrier admits in his latest note " that diverticula of the cavity of the axial organ
extend through the coelom, and that while some of them " apparaissent sur les coupes
comme termines en culs-de-sac, d'autres se plongent manifestement en canaux. Quelques-
uns de ces canaux coureut parmi les trabecules innombrables de la cavite generale ; il en
est qui se rendent vers les bras."

Some of these canals, which are regarded by Ludwig and myself as the intervisceral
blood-vessels, are represented in PI. LVII. figs. 2-5 and PL LX. figs. 3, 5 — ib. If they
are merely parts of a " vaste systeme aquifere," as Perrier believes, it is difficult to
understand their existence ; for the body-cavity through which they ramify already
contains water which has entered it by the water-pores of the disk. What then is the
object of a special set of aquiferous tubes distributed over the coils of the digestive canal
and not communicating with the ambulacral system, but with the axial organ and the
labial plexus ; and w^hy is it that their lumen is so frequently filled up with eoagulum ?

NOTE F.

(Page 106.)

On the Relation of the Vascular Systems of a Crinoid to those

OF the other Echinoderms.
Perrier's latest views respecting the vascular system of a Orinoid are expressed in the

1 Durch diese capillaren Poren kann das Wasser bis in die Nahe des ini Kelcli liegenden Eingeweidesacks
eindiingen (Ban des Pentacrinus, loc. cit., p. 49 ; see also the Bau der E<'hiuoderiuen, loc. cit., p. G3).
- Coinptcs renins, t. xcviii. p. 1449.

406 THE VOYAGE OF H.M.S. CHALLENGEE.

following passage :^ — " Ainsi le plus grande uombre des entonnoirs cilies, Vorgane
spongieux, Vorgane axial, les chambres de Vorgane cloisonne, ne forment qu'iin seul et
meme systeme, a la fois I'analogue et I'liomologue du systeme formd chez les Oursins,
les Asteries et les Ophiures par la lylaque madreporique, le canal hydrophore ou canal
du sable et la glande ovoide qui lui est constamment annexee."

The above statement harmonises admirably with the theory which Perrier has so long
l)een advocating respecting the fundamental unity of what are generally kno\\Ti as the
water-vascular and the blood-vascular systems of Echinoderms. This theory is by no
means new, and appeared to receive confirmation from the results of Perrier's study of
the circulatory apparatus of the Urchins.^ But Koehler's later observations on the same
subject ^ have shown that several important points in the anatomy of the vascular
sj^stem of an Urchin entirely escaped Perrier's notice. Although he adopts Perrier's
views, his observations are capable of an altogether different interpretation, as I have
shown elsewhere ;* while they afford a strong confirmation to Ludwig's description of the
vascular system of the Asterids.' This was founded upon the most careful and elaborate
observations which have yet been published ; and although their correctness has been
called in question by Messrs. Perrier and Poirier,* none of the French zoologists have
published a single figure in proof of their assertion that what is generally called the blood-
vascular system of a Starfish communicates with the exterior through the madreporite.

As regards both Starfishes and Urchins therefore, the latest and most detailed
observations do not tend to support the views of the French school. With respect to
the Crinoids, however, the results which Perrier describes himself as ha\dng obtained,
fall in with his theory in a manner which leaves nothing to be desired for completeness.
In the case of the Urchins, according to Koehler, Perrier saw too little ; while his two
hundred Comatula-sections have revealed more to him than has resulted from all the
observations of Luchvig, Teuscher, Greefi", and myself; and we, not Perrier, have seen
too little. His theory, however, breaks down completely unless he can prove to the
satisfaction of his colleagues that the labial plexus and chambered organ of a Crinoid
are in direct communication with the exterior through the water-pores of the disk.

Unless these points can be properly demonstrated, the doctrine that the water-
vessels and intervisceral blood-vessels of a Crinoid are only parts of a " vaste systeme
aquifere " will have to be abandoned ; while it does not harmonise at all with the present
state of our knowledge of the morphology of the Echinozoa, except in so far as this is
based upon the observations of the French Zoologists.

^ Comptes rendus, t. xcviii. p. 1449.

2 Recherches sur I'Appareil circiilatoire des Oursins, Archivesd. Zool. exp^r.,vol.iv., 1875, pp. 605-643, pis. xxiii., xxiv.

8 Recherches sur les Echinides des Cotes de Provence, loc. cit., pp. 58-79.

« Q^cart. Joimi. Micr. Sci., 1883, vol. xxiii., N. S., pp. 597-609.

^ Beitrage zur Anatomie der Asteriden, Zeitschr.f. wiss. Zool, Bd. xxx. pp. 99-131.

« Sur TAppareil circulatoire des Etoiles de Mer, Comptes rendu!,, 1882, t. xciv. pp. 658-660.

REPORT ON THE CRINOIDEA. 407

There is strong reason to believe that the " ovoid ghmd " of an Urchin or Starfish
does not communicate with the exterior through the madreporite as described by
Perrier, while there can now be no question respecting its direct connection with what
has been generally described as the oral blood-vastailar ring of these ty])es. Koehler's
observations^ have demonstrated this connection in the Urchins, although it was
categorically denied l)y Perrier.' It thus seems probable that although the " ovoid
gland " can no longer be described as a heart or even as a plexus of interlacing vessels,
yet that it is a glandular structure interpolated in the blood-vascular sj^stem, and
possibly one of the factories of the well known respiratory pigment of the Echinodenns.
The following remarks by Welldon ^ are noteworthy in connection with this suliject : —
"It is not too much to say that in every group of Invertebrates in which the vascular
system has been at all carefully investigated, glandular appendages to the vessels have
been found, which can, from their anatomical relations, have no other function than that

of elaborating some of the constituents of the blood In Echinoderms, the

abundance of glandular cells in the cardiac plexus is probably a principal cause of
the whole organ being regarded by many observers as an excretory apparatus."

NOTE G.

(Page 119.)

The Nervous System of the Crinoidea.

Since the section on the nervous system was written {ante, pp. 111-127), the subject
has been still further discussed by various morphologists, most of whom, I am glad to
say, have adopted the \aews advanced therein, and have strengthened them very
considerably. As in so many other cases, it appears that the doctrine of the nervous
nature of the axial cords in the skeleton of a Crinoid is not of such recent growth as has
been supposed. For the foUomng passage from von Schlotheim * would seem to show
that a nervous function was attributed to the contents of the central canal of the skeleton
more than sixty years ago : — " Da die Eneriniten aber sammtbch mit einer durch aUe
Zweige laufenden Nervenrohre versehen sind, und das Thier wenn es gleich mit der
Wurzel angewachsen zu sein scheint, doch mit alien seinen festen Theilen beweglich
bleibt, so gehort er ofFenbar nicht zu der Coralleuarten, und*macht nur ein merk-wiir-
diges Verbindungsgiied zwischen der Classe der Crustaceen uud der Zoophyten aus."

A general, and, on the whole, tolerably accurate account of the morphology of living
Crinoids was published by Weinberg ^ in the course of last year (1883). It is principally

1 Op. cit, p. 65, pi. 3, fig. 13. 2 Archives de Zool. expdr., vol. iv., 1875, p. 613.

' On the Head-Kidney of Bdellostoma, with a suggestion as to the origin of the Suprarenal Bodies, Quart. Joum.
Micr. Sci., vol. xxiv., 1884, N. S., pp. 180, 181.

* Die Petrefaetenkunde, Gotha, 1820, p. 327.

5 Die Morphologie der lebenden Crinoideen mit Beziehung auf die Form Antedon rosacea, Linck, Der Naturhis-
toriker, 5 Jahrg. Jlarz-Juni Heft, 1883, pp. 266-307.

408 THE VOYAGE OF H.M.S. CHALLENGEE.

a summary of the researches of his predecessors, though he states that he has himself
made some observations on Antedon rosacea at the Zoological Station at Trieste. It was
perhaps not to be expected that he should have done otherwise than propagate the
orthodox German view respecting the nervous system. But the account which he gives
of the position in 1883 of the doctrine that the axial cords are nerves, is an extremely
inadequate one. He states (p. 283) that it has been proved to be incorrect by Greeff;
while a few pages fui'ther on {p. 290) he says that attempts have been made to support
it by the supposition {Annahme) that fine branches proceed from the axial cords to the
muscles and arm-segments—" Ludwig und Greeff wiesen jedoch das Unzuliingliche und
Unrichtige der von Carpenter angefiihrten Argumente nach." He then refers to the
experiments performed by Dr. Carpenter, and leaves the question for further investiga-
tion.

Now, in the first place, the only comment which Greeff has made upon the doctrine
that the axial cords are nerves has been a simple denial of its truth, without any attempt
to discuss the subject at all ; ^ and yet this denial is referred to by Weinberg as a proof
of the doctrine being incorrect !

Ludwig, on the other hand, admits that the experimental evidence seems to afford
very considerable support to Dr. Carpenter's views ; " but he declines to accept them on
account of the morphological difficulties which they involve. He has been unable to find
the muscular branches from the axial cords which have been described by Dr. Carpenter
and myself, and more recently by Perrier, Marshall, and Jickeli. But this does not
justify Weinberg in stating that Ludwig has proved the arguments advanced by Dr.
Carpenter and myself to be insufficient and incorrect ; nor that the existence of these
branches is merely a supposition. The fact that they w^ere overlooked, not only by
Teuscher and Greeff, but also by Ludwig and Weinberg, even after I had specially
called attention to them, is no proof of their non-existence. Two figures of arm-
sections, showing these branches, together with a further discussion of the whole
question, were published in my paper '^ On the Minute Anatomy of the Brachiate
Echinoderms, which appeared two j^ears before Weinberg ■^Tote his resume, but is not
referred to by him at all.

Another point of consideralile interest in its bearings on this question is left entii-ely
unnoticed by Weinberg, though it was fully explained in a paper * which he quotes, and
it was illustrated by a diagram which also shows the branches of the axial cords ; although,
according to Weinberg, the existence of these branches is a mere supposition. I refer to
the frequent absence of the ambulacral nerve on more or fewer of the arms of Actino-
metra. Weinberg admits its absence on the oral pinnules of Antedon, for this was

' TJeljcr <len Ban der Crinoideen, Sit::ungsb. d. Gesellsch. z. Beford. d. nesammt. Naturiciss. zu Marburg, Nro. 1.
1876, pp. 21, 22.

- Crinoideen, loc. cit., p. .3.3.i. s Quart. Joum. Micr. Scl, 1881, vol. xxi., N. S., pp. 188-192.

* Remarks on the Anatomy of the Arms of the Crinoids, part ii., Journ. Anat. and Phys., vol. xi., 1876, pp. 90-93.

REPORT ON THE CRINOIDEA. 409

discovered by Ludwig, whose observations he could not very well overlook. But he does
not attempt to discuss the bearings of this fact on Ludwig's doctrine that the axial cords
are the only nerves in the arm of a Crinoid. He likewise describes the regular alternat-
ing movements of the arms of a swimming Comatula, and the muscles by which these
movements are effected ; and he leaves it to be inferred that these muscles are under the
control of the only nervous system of which he admits the existence, although experi-
ments have clearly j)roved that this is not the case ; while Ludwig has admitted that he
could trace no branches proceeding to the muscles from the ambulacra! nerve. As to
this last point, however, Weinberg is altogether silent.

Until this present year no German morphologist, with the exception of Weinberg,
had published any observations upon the Crinoids since the appearance of Ludwig's
important work in 1877 ; and the authors of zoological text-books published in Germany
have confined themselves with remarkable unanimity to reproducing Ludwig's asser-
tions that the nervous system of a Crinoid is essentially similar to that of an Asterid,
and is limited to the fibrillar bands beneath the ambulacra. Dr. Carpenter's views, if
mentioned at all, which was rare, were regarded as untenable from their being altogether
at variance with the established scheme of Echinoderm morphology. Claus, for example,
describes the arrangement of the axial cords in the calyx in some detail, but says not a
word about their functions ; w^hile their presence is not even mentioned by Gegenbaur.
According to these writers, therefore, the nervous system of a complex and highly
specialised type like Pentacriniis is exclusively represented by the subepithelial bands of
the ambulacra and the oral ring which unites them beneath the peristome (PI. LXIL, n,nr).
The extreme insignificance of these structures in comparison with the rest of the organism
cannot fail to strike any one who examines the sections of the ambulacra represented on
PI. LVIL; and yet, according to the orthodox German morphology, they are the only
nerves which a Crinoid possesses.

The theory that the axial cords are nerves has recently been restated by Di*.
Carpenter,' wdth the additional support of a quantity of new facts which had been
discovered since he last wrote on the subject, more than eight years ago. He concluded
by saying, " those who refuse to accept them (my views) are bound, I think, either to
disprove the facts, or to show that my deductions from them are unsound."

Within a very short time after the presentation of this communication to the Eo}'al
Society, two papers were published on the nervous system of the Crinoids, in which
Dr. Carpenter's theory was unreservedly adopted and strengthened by a large body of
additional evidence.

The second of these, by Professor A. Milnes Marshall,^ will be best considered first.
After a short historical sketch of the subject, he describes an elaborate series of experi-

1 On the Nervous System of the Crinoidea, Proc. Roy. Soc. Lond., vol. xxxvii., 1884, pp. 67-76.

^ On the Nervous System of Antedon rosaceus, Quart. Journ. Micr. Sci., vol. xxiv., N. S., 1884, pp. 507-548, iil. xxsv.

(ZOOL. CHALL. EXP. PART XXXII. — 1884.) li 52

410 THE VOYAGE OF H.M.S. CHALLENGER.

ments which have led him to the following conclusions : — " 1. The central capsule and
its prolongations, the axial cords and their branches, constitute the main nervous system
of Antedon. 2. The central capsule is specially connected with the complex co-ordinated
movements of swimming and of righting when inverted. 3. The axial cords act as both
afferent and efferent nerves. 4. The subepithelial bands are probably also nerves, but
their exact function, probably a special and subordinate one in connection with the
ambulacral tentacles and epithelium, is not yet ascertained."

These conclusions are the result of a long series of experimental investigations, in
which Dr. Carpenter's fundamental observations were repeated and largely extended.

Marshall's paper concludes with a valuable discussion of the morphological aspect of
his results. Starting from the generally accepted doctrine that the Asterids are the most
primitive group of the Echinoderms, he shows how this character is well illustrated by
their nervous system. Hamann's observations have demonstrated that this " is in the form
of a continuous nerve-sheath enclosing the whole body, and directly continuous with the
external ejaidermis of which it forms the deepest layer. This nerve-sheath is thickened
at certain places, notably along the ambulacral grooves, where it forms the five radial or
ambulacral nerves." Marshall points out that the analogies of the Coelenterates,
Chsetognatha, and Palseonemertines all tend to show the primitive nature of the Asterid
nervous system.^ There is no difficulty about the relation of the nervous system of the
remaining Echinozoa to that of the Asterids ; but the case is different with the Crinoids,
on account of the antambulacral position of their principal nerve centre and its radiating
extensions. Marshall, however, is inclined to consider them as "descended from forms
which agreed with the recent Asterids in possessing a complete nerve-sheath (though
possibly very unlike Asterids in other respects) ;" and he is therefore " disposed to regard
the antambulacral nervous system of a Crinoid, i.e., the central capsule and axial cords
with their branches, as being derived from the antambulacral part of the primitive nerve-
sheath, and not as an entirely new set of structures possessed by no other Echinoderms."

He endeavours to show that the relations of the axial cords which lie in grooves on
the surface of the radials of the Pentacrinoid larva (a permanent condition in some
Palgeocrinoids) are " very similar to those of the aml)ulacral nerves of an adult Ophiurid
or Echinid, and as the latter have certainly acquired their adult condition by becoming-
detached from the epidermis and shifting inwards, so also may the same process be
supposed to have occurred in the Crinoid." Too much weight must not be laid upon
this point, however, for the supjaosed inward movement of the radial nerves of an
Ophiurid or Echinid would be from the outer or ambulacral surface of the plate towards
the inner one, i.e., that next to the body-cavity.

On the other hand, in the developing Crinoidal calyx the axial cords are at first on the
walls of the body-cavity, which are formed by the inner surfaces of the radials ; but they

1 Compare Chapter VL pp. 113, 115.

KEPORT ON THE CRINOIDEA. 411

gradually come to lie iu canals withiu the radials, and so are relativel}^ (tliougli of course
not absolutely) nearer the epidermis on the exterior of the body, which in this case is
antambulacral.

The ontogenetic change in the relative position of the axial cords of a Crinoid is thus
directly the opposite of that which Marshall supposes to have taken place phylo-
genetically. On his theory the antambulacral portion of the primitive nerve-sheath
should commence by being outside the radials, between them and the epidermis ;
whereas, as he himself admits, the radials are at first in the form of " calcareous plates
between the cords and the integument." His argument is, therefore, only one of analogy,
and the outward movement of the axial cords in the Crinoid larva is not comparable
morphologically to the inward movement, which must have taken place during the develop-
ment of the radial nerve of an Urchin from the primitive nerve-sheath of a Starfish.

It appears to me, however, that there is a possible view of the phylogeny of the
axial cords which would not conflict in this way with their ontogenetic movement.
According to Gotte's observations, the ciliated ambulacral epithelium of the Criuoids is
derived from the cellular lining of the left peritoneal sac ; ' so that the ambulacral nerve,
which is in such close relation with this epithelium, is endodermic in its origin. On the
other hand, the axial cords of a larval Crinoid lie in the walls of the dorsal portion of the
body-cavity, which is lined by an epithelium derived from that of the right peritoneal
sac ; and I would therefore suggest that the embryonic axial cords might have the same
primitive relation to this epithelium as the ambulacral nerves must have to that of the
left peritoneal sac, if Gotte's observations be' correct. The outward movement of the
cords during development, away from the epithelium of the body-cavity, would then be,
so to speak, serially homologous with the supposed inward movement of the ambulacral
nerves of an Urchin. In other words, the axial cords are the nerves of the right
antimer, while the subepithelial bands of the ambulacra are those of the left antimer,
both being derived in the first instance from the epithelium of the enterocoel.

In the Neocrinoids the axial cords eventually come to be some little distance from
this epithelium ; though they must have remained close to it in those Palseocrinoids
W'hich had imperforate radials, just as they are in the Pentacrinoid larva. A variation of
the same nature, though of course altogether difi"erent in degree, presents itself in the
relation of the ambulacral nerve in certain species of Neocrinoids. I refer to the presence
or absence of a delicate connective tissue lamella between the nerve and the overlying
epithelium. This layer is often very conspicuous in Antedon eschrichti, but occasionally
seems to be absent ; while its presence in Antedon rosacea is doubtful. I have never
satisfied myself of its existence in any other species, though Ludwig and Marshall
seem to quote me as having noted its occurrence in Actinometra. Ludwig^ long ago

1 Archiv. f. mikrosk. Anat, Bd. xii. pp. .591-.'59.3.

2 Beitriige zur Anatomie der Crinoideeii, Nachrichf. v. (I. kijl. Gesdlsch. d. {Viss. zu Gottinfjen, No. 5, 1876, p. 108.

412 THE VOYAGE OF H.M.S. CHALLENGER.

suwo-ested -that the great development and calcification of this lamella would bring
the ambulacral nerve into a position corresponding to that of the radial nerve-cords
of Ophinrids and Echinids. Marshall^ has recently put forward a somewhat similar
hjqiothesis, viz., that this lamella "probably represents the earliest stage in the
process by which the nerve becomes detached from the epidermis and shifted inwards."
We know far too little, however, about the ontogeny of the Echinoderm nervous
system to do more than speculate on this subject. According to Selenka and
Ludwig the nervous system of Asterids and Holothurids is of ectodermic origin ; while
Gotte's observations lead to the conclusion that the ambulacral nerves of Crinoids are
derived from the endoderm. Should this really be the case, there can be no difficulty in
taking the same view respecting the axial cords.^ But even then we get no clue to the
morphology of the central capsule, as Marshall has conveniently called the fibrillar
envelope of the chambered organ in which the axial cords originate.

He remarks' that "Dr Carpenter's observations lead to the belief that, at any rate in
its present form, it is connected with the change from the pedunculate to the free-
swimming condition ; and it is worthy of notice that the two actions with which it has
been found to be specially concerned physiologically, i.e., the movements of swimming
and of righting, are ones that the pedunculate form, from the very nature of things, can
never exercise."

I cannot quite share Marshall's belief in the relation between the central capsule and
the change from the attached to the free mode of life. The only difi'erence between the
chambered organ of a Comatula and that of a Stalked Crinoid is the absence of any
cirrus-vessels in connection with the latter ; for these come ofi" from the peripheral vessels
of the stem (PI. XXIV. fig. 4 ; PI. LXII. — cv), which are the downward extensions of the
cavities of the chambered organ. But the central capsule or fibrillar envelope of these
cavities, which in Comatula " is specially connected with the complex co-ordinated move-
ments of swimming and of righting when inverted," is eqiially present in all the Stalked
Crinoids (PL Vllb. fig. 2 ; PL XXIV. figs. 6, 7 ; PL LVIII. figs. 1,3; PL LXIL); and
there can be no doubt that it controls the movements of flexion and extension of the arms.
The latter of these is essential to the proper nutrition of the animal ; and I can quite
believe that the arms may also be used for swimming by those Pentacrinidfe, such as
Pentacrimis maclearaniis, Pentacrinus alternicirrus, and Pentacrinus ivyviUe-thomsoni,
which have short stems terminating below at a nodal joint (PL XVI. fig. 1 ; PL XIX. fig. 1).
In all the Stalked Crinoids the central capsule is continued downwards into the stem as a
sheath around the central vascular axis (PL Vila. figs. 1, 2 ; PL XXIV. figs. 1-6 ;
PL LXII. — ca), and it gives off branches which spread out towards the surface of the stem,

1 Loc. cit., p. 546.

- Whatever be the origin of these cor.ls, they are essentially mesodermic in their distribution ; and it is in tliis sense
that I have spoken of them in the text as constituting a mesodermic nervous syskm (p. 114).
^ Loc. cit, p. 547.

REPORT ON THE CRINOIDEA-.

413

and probably form a subepidermic plexus (PI. Vila. fig. 1 ; PI. XXIV. fig. 2 — ca').
Extensions of it also proceed into the cirri round the cirrus-vessels (PI. LXII., cv), and
give off" similar radiating branches. Jickeli's observations on Antedon rosacea show that
the movements of the cirri, like those of the arms, are dependent upon an influence pro-
ceeding from their axial cords ; ' and it is clear that if the central capsule and axial cords
of Comatula constitute its principal nervous system (as few will now deny), this must be
regarded as also extending throughout the whole stem of a Stalked .Crinoid, even though
it reach 70 feet in length, as in the fossil Extracrinus subangidaris.

It may also be assumed with tolerable certainty that there was a similar neuro-
vascular axis in the stem of all the fossil Pelmatozoa, including the Blastoids. These had
no arms attached to the radials, as a Crinoid has. But if they had the same arrange-
ment of axial cords in the basals and radials as prevail in a Crinoid, which I see no reason
to doubt, it is unlikely that the radial cords, starting from the circular commissure
(fig. 20, A) should have ended in the forks of the radials which receive the distal ends of
the ambulacra. In the Mesozoic Phyllocrinus the radials have almost exactly the same
forked shape as those of a Pentremites ; l^ut the point where the limbs of each fork
separate is marked by an articular facet to which the arm was attached.

In a Blastoid, however, the sinus between the two limbs of the radial is filled up by
the ambulacrum, which terminates in a more or less prominent lip at the same point in
the body of the radial as is occupied by the articular facet in
Phyllocrinus. That the axial cord in the radial of a Pentremites
(fig. 20, A) ended in this lip seems to me improbable ; and I
cannot help suspecting that it may here have become continuous
with the nerve of the ambulacrum. From what we know of the
ambulacral nerves of recent Crinoids, this is not likely per se to
have been related to any other movements but those of the
tentacles, even if these organs were present ; while it may
perhaps have been removed from the superficial epithelium, as
in an Urchin, and lodged within one of the canals in the lancet
plate. This would have brought it deep enough to be continuous
with the axial cord of the radial at the termination of the
ambulacrum. The movements of the so called pinnules of the
Blastoid would thus have been controlled by the central capsule,
instead of by the oral ring, which is not likely to have had
anything to do with them ; for the oral ring of a recent Crinoid
has absolutely no control over the movements of any part of the
calcareous skeleton ; and the jointed pinnules of a Blastoid cannot have remained
permanently closed over the ambulacra, as they are found in the fossils (fig. 20).

' See the r|uotation from Agassiz, ante, p. 333.

Fig. 20. — Diagi-am showiDg the
arrangeinent of tlie axial cords
in the calyx of a Pentremites,
supposing it to be the same as
in the calyx of a Crinoi J. The
pinnules are represented as
preserved on one ainbulacruni,
but not on the other. ^ , axial
cord of the ray; ai, primary
interradial cord" ; «'/■, one of the
converging radial cords which
result from its bifurcation;
B, basal; ceo, circular com-
missure ; /, interradial plate
{deltoid of Koemer) ; Ji, radial.

414 THE VOYAGE OF H.M.S. CHALLENGER.

Loven lias recently pointed out the singularly Crinoidal appearance of the calycinal
system of a Triassic Urchin, Tiarechinus princeps, Laube.^ " The relative magnitude of
the entire system, the prominent share it takes among the constituents of the skeleton,
the forms and proportions of its parts, are such as forcibly to recall the calyx of some
Palseocriuoid, and to justify a desire to turn the Echinoid upside down and to see the

calycinal system in its imaginary original position, when it
formed a part of some remote ancestral type. In this respect
the resemblance becomes still more striking."

Let us compare Lov^n's figure (fig. 21) with that of the
Blastoid (fig. 20). Each has relatively large basals (B) ; l)ut
the radials of the Pentremite (R) are small in Tiarechinus,
their limbs being replaced by the two lateral plates of the
interradius (^) which enclose the ambulacrum. In Tiare-
„„,„.,. , , c/;m7f.f, just as in the Blastoid, however, the ambulacrum ends

Fig. 21. — Tiarechimis prmceps, Laube, ■' '

iTTi^'^McaTsy^stertfthrcalyx o?°a ^gaiust the body of the Corresponding radial ; and its nerve,
^^^^^: .Zr"ls^^:)''V'tJ^\\ lyiiig beneath the ambulacral plates as in recent Echini,
ittrtiakftrS''^*" ' '■""™' ^'o^^ld be in a position where it could be directly continuous

with an axial cord situated within the radial or on its inner
face, if only Tiarechinus were a Crinoid instead of an Urchin. I would not, of course,
be understood as saying that Tiarechinus had a central capsule and axial cords pro-
ceeding from it. My only object has been to jaoiut out that Loven's happy comparison
of this curious type with an inverted Crinoid alFects other systems of organs besides that
of the calycular plates. At the same time, considering the number of Asterids which
have a calyx of relatively large plates, and the fact that there is a continuous nerve
sheath on the dorsal surface, I think it not improbable that indications of a central
capsule and axial cords may eventually be discovered in the Echinozoa.

It remains now to notice the observations of Dr. Jickeli, which, though only published
recently, are nearly four years old.^ Like Marshall he has made an elaborate series of
experiments in extension and confirmation of those originally described by Dr. Carpenter,
whose views respecting the nervous nature of the central capsule and axial cords he
adopts unreservedly. He further believes that the fibrillar bundles uniting the cirrus-
joints and those forming the dorsal and interarticular ligaments (as they are described
above) in the arms (PI. LXIL, Id, li) are muscular in function, though differing in many
points from the fibres which have hitherto been exclusively described as muscles
(PI. LXIL, m); for when the axial cord of a detached cirrus is stimulated " so krlimmt
sich derselbe auf das Heftigste zusammen und gerath selbst in Tetanus." ^

* On Pourtalesia, loc. cit, p. 65.

2 Vorlauflge Mittheilungen iiber den Bau der Echinodermen, ], Ueber das Nervensystem und die Sinnesorgane
der Comatula mediterranea, Zool. Anzeiger, vii. Jahrg., pp. 346-349 and 366-370, 1884.
2 Zool. Anxeiger, vol. vii., 1884, p. 348.

REPORT ON THE CRINOIDEA. 415

This seems to indicate clearly that the iuterarticular fibres of the cirri are muscular
iu function, though not striated like the ventral fibres which unite the arm-joints, and
have hitherto been regarded as the only true muscles of the Crinoid organisation.

Four years ago I pointed out that the appearance of some of my sections had led me
to think that the axial cord of the arm consisted of two lateral fibrillar masses enclosing
a central structure.^ According to Jickeli" this last is a longitudinal septum between
two tubes which contain blood-corpuscles ; while the whole structure is enclosed in
a sheath of nervous tissue. Within the calyx the tubes of the different rays unite
laterally and " gehen durch eben solche das gekammerte Organ theilweise bedeckende
Fortsatze in ein spongioses Geflecht liber, welches dem gekammerten Organ wie eine
Kappe aufsitzt. Von diesem spongiosen Geflecht entspringen die Fortsetzungen in die
Cirrhen. Die Wandung dieses Eohrensystems ist die nervose Substanz, die Masse,
welche dasselbe erfiillt, besteht aus geronnenem Plasma, in welches Blutzellen einge-
baeken sind." It does not appear, however, that these axial blood-tul)es are in any
way connected wdth the cavities of the chambered organ. Jickeli points out that
his observations confirm the statements of Miiller ^ respecting the presence of a Ijlood-
vessel within the central canal of the arms, which subsec[uent workers have generally
considered to be erroneous. He finds that the elements of the nerve sheath surrounding
these blood-vessels are most easily demonstrated in the radial axillaries where the axial
cords of two arms unite, and he describes the presence of ganglionic cells with from two to
six processes, some of which unite with those of other cells. He has also seen the
muscular branches of the axial cords, the existence of which, according to Weinberg, is
only a supposition ; and by the use of polarised light he has traced these into the dorsal
as well as into the ventral musculature.

He doubts the presence of a definite oral ring in the amljulaeral nervous system.
For he finds that the ventral nerves extend down into the fore-gut beneath its
epithelium, which is directly continuous with that lining the food-grooves. The appear-
ance of several of my sections, both of Pentacrinus and of Comatula, has led me to
suspect this fact ; but I have hesitated to say so, as I wished to verify it by making
some sections of individuals which had been properly prepared for histological work.
This having been done by Jickeli, I am glad to be able to confirm his observations.

He believes himself to have discovered yet another nervous structure in the Crinoid
organisation, "es ist noch eindrittes bis dahin nicht bekanntes, im Bindegewebe gelagertes
Nervencentrum vorhanden, welches die Mundoffnung umgibt, und die radialen Wasser-
gefasse jederseits als ein gesonderter, an die Tentakeln in regelmassigen Abstanden
Zweige abgebender Strang begleitet."* The peripheral parts of this system appear to me
to belong to what I have called the parambulacral network, situated iu the ventral

1 Quart. Journ. Micr. Sci., 1881, vol. xxii., N. S.. p. 187. ^ Zool. Aimiger, vol. vii. p. 368, 1884.

' Bail der Pentacrinus, foe. cit, p. 22. * Zool. Anzeiger, vol. vii. p. 370, 1884.

416 THE VOYAGE OF H.M.S. CHALLENGER.

perisome of the arms and disk. Jickeli says, for example, " von diesem dritten Nerven-
centrum gehen auch starke Zweige in die ventrale Korperhaut uud liisen sich dort in feine
nervose Geflechte auf.'" It wUl be strange indeed if these prove to be anything else
than the ramifications of the ventral branches of the axial cords of the arms which I
described long ago as extending to the edges of the food-groove (see fig. 4, p. 113 ; fig. 5,
p. 121 ; and fig. 8, p. 123). I cannot say, however, that I have ever seen the pentagonal
rino- round the mouth W'hich Jickeh mentions, nor even its radial extensions at the sides
of the water- vessels ; unless indeed these last be the lateral trunks which I have described
above in Actinometra iiigi'a, from sections now nearly nine years old (p. 122).

The branches of Jickeli's third nervous system which break up into a plexus in the
ventral perisome appear to me to be identical with those which I described two years ago
as extending along the sides of the ambulacra of Antedon eschrichti from the edge of the
disk to the neighl)ourhood of the mouth.^ A diagrammatic representation of them is
given on p. 123, while illustrations of single sections, both of this species and oi Penta-
crinus decorus, are shown on PI. LIX. figs. 2-7. These branches are unquestionably of
the same nature as those occupying a similar position in the arms (PI. LX. fig. 6, a'), and
belong like them to the system of the central capsule and axial cords, vdth which last
they are connected at the edge of the disk. But at the same time I fully believe that
they are the peripheral portions of the third nervous system described by Jickeli. The
so called papillae of the tentacles have also attracted his attention, and he regards them
as .sense-organs of a somewhat complicated nature, supporting fine sensory hairs. He
thus inclines to Perrier's view of the nature of these papiUge rather than to that of
Ludwig, who regards them as glandular organs. Jickeli, however, describes them as being
innervated by the branches of his third nerve-centre ; while according to Perrier ^ they
receive nerve-fibres from the ventral branches of the axial cords, which form what I have
called the parambulacral network. But if I am right in identifying this with the
peripheral part of Jickeli's third nervous system, his observations are completely in
accordance with those of Perrier.

1 Zool. An::eiger, vol. vii. p. 369, 1884. - QimH. Journ. Mia: Sci., 1883, vol. s-viii., N. S., p. 615.

' Comptes rendus, t. xcvii. p. 188.

BIBLIOGRAPHY OF THE NEOCRINOIDEA.

The followiug list does uot pretend to be a complete Bibliograpliy of tlie Neocrinoidea,
i.e., to give a reference to every description of a fossil Crinoid which has appeared in the
countless publications upon the PaljBontology of the Secondary and Tertiary Periods.
Neither is it meant to include every notice of a recent Crinoid in faunal and other lists.
The numerous controversial references to fossil forms in epistolary communications to
the Neues Jahrbuch by von Buch, von Meyer, and others are also mostly omitted,
and must be sought for in the excellent Bibliography of de Koninck and le Hon.

But I have endeavoured to make the accompanying list complete for the two
following classes of works : — (1) all those containing descriptions of new species of recent
Crinoids ; (2) all those which, though dealing principally with Palseocrinoids or with
other Echinoderms, contain valuable information upon the Morphology of the Crinoids
generally. The literature of the Pelmatozoa has increased at a very rapid rate since the
publication of the admirable Bibliography by de Koninck and le Hon in 1854, which
contained a chronological list of nearly all the earlier works on the subject.' W(> may
hope ere long for a supplement to this as regards the Palseocrinoids from Messrs. Wach-
smuth and Springer ; and I trust in the course of time to be able to complete the
following list by the addition of references to works on general palaeontology which
contain notices of Neocrinoids.

Adams, J., Description of some Marine Animals found on tlie Coast of Wales. Tran^. Linn. Soc. Loud.,

vol. v., 1800, pp. 7-13, Tab. ii.
Agassiz, a., Note on Lovt5n's article on Leskia mirabilis, Gray. Ann. Lijccmn Kat. H/W., vol. ix. pp. 242-

245, New York, 1869.
Eevision of the Echini. ///. Cat. Mus. Coiiip. Zoiil., No. 7, Cambridge (U.S.), 1872-74, xii + 762

pp., 94 pis., 69 cuts.

North American Starfishes. Alrm. Mus. Comp. Zoo!., vol. v., No. 1, 1877, iv + 136 pp., 20 pis.''

Drawing of Young Holopus from Babia Honda, Cuba. Bull. Mus. Comp. Zoiil., vol. v.. No. 9, 1878,

p. 213.
Selections from Embryological Monographs. II. Echinodermata. Mem. Mus. Comp. Zoijl., vol. ix.,

No. 2, 1883, 45 pp., 15 pis. Bibliography in Bull. Mus. Comp. Zobl, vol. x.,'No. 2, 1882, pp. 109-131.

' I have come across a few morphological papers which are not noticed by de Koninck and le Hun, and have
recorded them in the following list.

(ZOOL. CHALL. EXP. — PART XXXII. — 1884.) Ii ^3

418 THE VOYAGE OF H.M.S. CHALLENGER.

Agassiz, a., EeporU on thu ResulU of Drudging uuJur the Supervision of Alexander Agassiz, in the Gulf of

Mexico (1877-78), in the Caribbean Sea (1878-79), and along the Atlantic Coast of the United States

(1880), by the U.S. Coast Survey steamer " Blake," Lieut. -Commander C. D. Sigsbee, U.S.N., and

Commander J. R. Bartlett, U.S.N., commanding — XXIV,, Part I., Report on the Echini. Mem. Mus.

Comp. Zool, vol. X., No. 1, 1883, viii + 94 pp., 32 pis.
Agassiz, L., Prodrome d'une Mouographie des Radiairus ou Echinodermes. Mem. Soc. Sci. Nat. Neiichatel,

vol. i., 1835, pp. 108-199.
Twelve Lectures on Comparative Embryology, delivered before the Lowell Institute in Boston.

Boston, 1849, 104 pp.
Allman, G. J., On a Pre-Brachial Stage in the Development of Comatula, and its importance in relation to

certain Aberrant Forms of Extinct Crinoids. Trans. Roy. Soc. Edin., vol. xxiii., 1864, pp. 241-252,

pi. xiii. Abstract in Rep. Brit. Assoc, 1862, Trans. Geol. Sect., p. 65.
Anontmous, Notes and Observations on Injured or Diseased Crinoids. Proc. Nat. Hist. Sue. Glasgow, voi. iii.,

1876, pp. 91-94.
Austin, T., Proposed Arrangement of the Echinodermata, particularly as regards the Crinoidea, and a subdivision

of tbe class Adelostella (Ecliinidse). Ann. and Mag. Nat. Hist., ser. 1, vol. x., 1842, pp. 100-103.
Observations on the Cystideaof M. von Bucli, and the Crinoidea generally. Journ. Geol. Soc, vol. iv.,

1848, pp. 291-294.
Observations on the connection between the Crinoidea and the Echinodermata generally. Ann. and

Mag. Nat. Hist., ser. 2, vol. viii., 1851, pp. 280-290.
Austin, T., and T. Austin, jun., A Monograph on Recent and Fossil Crinoidea, 128 + xvi pp., 16 pis. (all

published). Bristol, 1843-45.
Baily, W. H., Description of a new Pentacrinite from the Kinimeridge (c/. Oxford) Clay of Weymouth,

Dorsetshire. Ami. and Mag. Nat. Hist., ser. 3, vol. vi., 1800, pp. 25-28, pi. i.
Balfoue, F. M., a Treatise on Comparative Embryology, vol. i., London, 1880. Echinodermata in chap. x.n.

pp. 453-481.
Baudelot, E., Etudes g6n6rales sur le systeme nerveux. Contribution a Thistoire du systime nerveux des

fichinodermes. Archives d. Zool. exper., t. i., 1870, pp. 177-216.
Bell, F. J., An Attempt to apply a Method of Formulation to tlie Species of the Comatulidaj, witli the

description of a new species. Proc. Zool. Soc Loud., 1882, pp. 530-536, pi. xx.

■ Note on a Crinoid from the Straits of Magellan. Proc. Zool. Soc. Land., 1882, pp. 050-652, with cut.

Report on the Zoological Collections made in the Indo-Pacific Ocean during the Voyage of H.M.S.

"Alert," 1881-82. Crinoidea in pp. 153-170, pis. x.-xvii., London, 1884.
Beyrich, E., Ueber die Crinoideen des Muschelkalks. Ahhnndl. d. k. Akad. d. Wiss. Berlin., Phys. KL,

Nr. 1, pp. 1-49, Taf. i., ii.
Eugeuiacrinus. Zeitschr. d. deidsch. geol. Gesellsch., Bd. xxi., 1869, p. 835. ProtocoU der Sitzung

vom 15 Sept. 1869.
Ueber die Basis der Crinoidea brachiata. Monatsher. d. k. preuss. Akad. d. Wiss. Berlin, 1871,

pp. 33-55.
Billing.s, E., On the Cystidere of the Lower Silurian Rocks of Canada. Figures and Descriptions of Canadian

Organic Remains. Decade III., 1858, pp. 9-74, pis. i.-vii.
On the Crinoideie of the Lower Silurian Rocks of Canada. Figures and Descriptions of Canadian

Organic Remains. Decade IV., 1859, pp. 7-66, pis. i.-x.
Note on Leskia mirabilis. Gray, by S. Loven, communicated by C. F. Liitken. Canad. Nat, N. S.,

vol. iii., 1868, pp. 442-445, 5 figs.
— - Note on Ilyponome Sarsi, described by S. Loven, by C. F. Liitken. Canad. Nut., N. S., vol. iv.,

18G9, p. 270.
Notes on the Structure of the Crinoidea, Cystidea, and Blastoidea. Amcr. Journ. Set. and Arts,

vol. xlviii., 1SG9, pp. 09-83 ; vol. xlix., 1870, pp. 51-58; vol. 1., 1870, pp. 225-240. Reprinted in the

Ann. and Mag. Nat. Hist, ser. 4, vol. v., 1870, pp. 251-206, and 409-416; vol. vii., 1871, pp. 142-158.

r.EPORT ON THE CRINOIDEA. 419

r.t.AiNViLLE, n. M. D. DE, MaDUcl d'Actinologie on tie Zoophytologie. Paris, 1834, viii + 694, pp. 100 pis.

CriuoiLs in pp. 247-265 ; p!s. xxvi-xxix.
LaHLscHE, W., Uuber Actiuomctra Eenuettii und eiiie neue Comatula-Art (Aiitedun Dubcnii). Archiv f.

Naiunjesch., Bd. xxxii., 1SC6, pp. 90-92.
BuoNN, H. G., XJebor die Kriiii)iileii-Keste im Muschelkalk. Neues Jahrh. f. Mlneraloffie, Jalirg. 1837,

pp. 30-33, Taf. ii.
Bronx, H. G., and Eoemer, F., Lethroa geognostica, oder Abtildung und Besehreibung der fiir die Gebirgs-

Furmalionen bezeicbnendsten Versteinerungen. Dritte Auflage, Theil v, Bd. ii.-iii., Mesolethsea,

Stuttgart, 1851-52.
Die Klassen und Ordnuiigcn des Thier-Reiclis, Bd. ii. (Aktinozoen). Leipzig and Heidelberg, 1860,

434 pp., 49 pis. and cuts.
BuscH, W., Ueber die Larve der Comatula. Archiv /. Aunt. Plnjnol. u. iriss. Med., Berlin, 1849, pp. 400-402

(Mit. Abbildgn).
Beobacbtungen ueber Anatomie und Entwickelung einige Wirbellosen Seethiere. Berlin, 1851,

143 pp., 17 pis.
Carpenter, P. H., Remarks on the Anatomy of the Arms of the Crinoids. Jonrn. Anat. and Physiol.,

vol. X., 1876, pp. 571-585, 2 cuts.
Remarks on the Anatomy of the Arms of the Crinoids, part. ii. Joiirn. Anat. and Phydol., vol. xi.,

1S76, pp. 87-95, with cut.
On some Points in the Anatomy of Pentacrinus and Rhizocrinus. Joi/rn. Anat. and T'hysiol.,

vol. xii., 1877, pp. 35-53.
On the genus Actinometra, Miill., with a Morphological Account of a new Species (."Vctinonietra poly-

morpha) from the Philippine Islands. Trans. Linn. Soc. Loud. (Zool.), ser. 2, vol. ii., 1879-84, part

1, pp. 1-122, pis. i.-viii. Abstract in Jonrn. Linn. Soc. Land. (Zool.), vol. xiii., 1877, pp. 439-456.
— Preliminary Report upon the Comatulse of the Challenger Expedition. Proc. Boy. Soc, vol. xxviii.,

1879, pp. 383-395.

Notes on Echinoderm Mor^jhology. No. I. On the Oral and Apical Systems of the Euhinoderms,

pt. i. Quart. Journ. Micr. Sci., N. S., vol. xviii., 1878, pp. 351-383, 11 cuts.

Note II. On the Apical and Oral Sy.stems of the Echinodermata, pt. ii. Quart. Jonrn. Micr. &-i..

N. S., vol xix., 1879, pp. 176-206, 7 cuts.
— ■ Note III. Some disputed points in Ecbinoderm Morphology. Quart. Journ. Micr. Sci., N. S.,

vol. XX., 1880, pp. 321-329.
Note IV. The Jlinute Anatomy of the Brachiate Echinoderms. Quart. Journ. Micr. Sci., N. S.,

vol. xxi., 1881, pp. 169-193, 2 pis. and 1 cut.
Note V. On the Homologies of the Apical System, with some Remarks upon the Blood- Vespels.

Quart. Journ. Micr. Sci., N. S., vol. xxii., 1882, pp. 371-386, 4 cuts.
Note VI. On the Anatomical Relations of the Vascular System. Quart. Jonrn. Micr. Sci., N. S.,

vol. xxiii., 1883, pp. 597-616.
Note VII. On the Apical System of the Ophiuriils. Quart. Jcmrn. Micr. Sc-i., N. S., vol. xxiv., 1884,

pp. 1-23, pi. i.
Note VIII. On some points in the Anatomy of Larval ComatiilaB. Quart. Journ. Micr. Sci., N. S.,

vol. xxiv., 1884, pp. 319-327, with cut.

On the Nomenclature of the Plates of the Crinoidal Calyx. Re^h Brit. A.^soc, 1879, pp. 333, 334.

The Nervous Sj^stetn of Comatula. Eep. Brit. Assoc, 1879, pp. 418, 419.

The Chambered Organ of Comatula. Zool. Anzeiger, Jahrg. ii., 1879, pp. 569-571.

Feather-Stars, Recent and Fossil. Popidar Science Review, N. S., vol. iv., 1880, pp. 193-204, pis. v., vi.

On some undescribed Comatulte from the British Secondary Rocks. Quart. Journ. Geol. Soc,

vol. xxxvi., 1880, pp. 36-55, pi. v. and cut.

On some new Cretaceous Comatula^. Quart. Journ. Geol. Soc, vol. xxxvi., 1880, pp. 549-558, pi. xxiii.

■ Oi the Genus Solanocrinus, Goldfuss, and its Relations to recent Comatulae. Journ. Linn. Soc. Land.

(Zool.), vol. XV., 1880, pp. 187-217, pis. ix.-xii.

420 THE VOYAGE OF H.M.S. CHALLENGER.

Carpenter, P. H., On two new Ciinokls from the Upper Chalk of Southern Sweden. Quart. Journ. Geo?.
Soc, vol. xxxvii., 1881, pp. 128-136, pi. vi.

On a new Comatula from the Kelloway Eoek. Abstract in Five. Geol. Soc, No. 407, 1881, p. 98.

On the Characters of the "Lansdown Encrinite" (Millericrinus Prattii, Gray, sp.). Bep. Brit. Assoc,

1881, p. 635.

On some Permanent Larval Forms among the Crinoidea. Rq). Brit. A^soe., 1881, pp. 671, 672.

Note on the European Comatulae. Zool. Anr.eiger, Jahrg. iv., 1881, pp. 520-522.

The Comatula; of the Leyden Museum. Notes from the Leydcn Museum, vol. iii., 1881, pp. 173-217.

Keports on the Results of Dredging under the Supervision of Alexander Agassiz, in the Gulf of Mexico

and in the Caribbean Sea (1877-79), and along the Atlantic Coast of the United States during the summer
of 1880, by the U.S. Coast Survey steamer "Blake," Lieut. -Commander C. D. Sigsbee, U.S.N., and
Commander J. R. Bartlett, U.S.N., commanding — XVI. Preliminary Report on the Comatulas. Bull.
Mas. Camp. Zool., vol. ix., No. 4, 1881, pp. 1-19, pi. i.

On some new or little known Jurassic Crinoids. Quart. Jouru. Gcol. Soc, vol. xxxviii. pp. 29-4.J,

Descriptions of new or little known Comatulae — I. On the Species of Atelecriuus and Eudiocrinus.

IL The Comatula} of the Hamburg ^Museum. Journ. Linn. Soc Land. (Zool.), vol. xvi., 1882, pp. 487-526.

On the Classification of the Comatuke. Proe. Zool. Soc Land., 1882, pp. 731-747.

Reports on the Results of Dredging under the Supervision of Alexander Agassiz, in the Gulf of Mexico

(1877-78) and in the Caribbean Sea (1878-79), by the U.S. Coast Survey steamer "Blake," Lieut.-

Commander C. D. Sigsbee, U.S.N., and Commander J. R. Bartlett, U.S.N., commanding — XVIII. The

.Stalked Crinoids of the Caribbean Sea. Bull. Mus. Camp. Zool., vol. x.. No. 4, 1882, pp. 165-181.
On the Supposed Absence of Basals in the Eugeniacrinidee and in certain other Neocrinoids. Ann. and

Mag. Nat. Hist., ser. 5, vol. xi., 1883, pp. 327-334.

Note on Democriuus Parfaiti. Ann. and Mag. Nat. Hist, ser. 5, vol. xi., 1883, pp. 334-336.

. On a new Crinoid from the Southern Sea. Phil. Trans., part iii., 1883, pp. 919-933, pi. 71. Abstract

in Proc Roy. Soc Lond., vol. xxxv., 1883, pp. 138-140.
On the Crinoidea of the North Atlantic between Gibraltar and the Fseroe Islands (with some notes on

the Myzostomida by Prof L. von Graff). Proc Roy. Soc Edin., vol. xii., 1884, pp. 353-380.
On three new Species of Metacrinus. Trans. Linn. Soc Lond. (Zool.), ser. 2, vol. ii., 1879-84,

part 14, pp. 435-447, pis. l.-lii.
Report on the Coraatulaj dredged in the Barents Sea by the " Willem Barents " during the cruises of

1880 and 1881. Bijdragon tot de Dierkunde. Uitgegeven door het Genootschap iNatura Artis Magistra te

Amsterdam, 10' .\flevering. Onderzoekingstochten van de Willem Barents. (In the Press.)

Report on the Comatulre dredged by the " Varna " in the Kara Sea. Bijdragen tot de Dierkunde.

Uitgegeven door het Genootschap Natura Artis Magistra te Amsterdam, 10' .Atlevering. Onderzoekinj

stochten van de Varna. (In the Press.)
Carpenter, P. H., and Etheridge, R., jun., Contributions to the Study of the British Palaeozoic Crinoids. — No. I.

On Allagecrinus, the Representative of a new Family from the Carboniferous Limestone Series of Scotland.

Ann. and Mag. Nat. Hist, ser. 5, vol vii., 1881, pp. 281-298, pis. xv., xvi.
Carpenter, "W. B., Researches on the Structure, Physiology, and Development of Antedon (Comatula, Lamk.)

rosaceus — Parti. Phil. Trans., 1866, pp. 671-756, pis. xxxi.-xliii.
— Addendum to a translation of Semper's Brief Observations on the Anatomy of Comatula. Ann. and

Mag. Nat Hist, ser. 4, vol. xvi., 1875, pp. 206-208.
On the Structure, Physiology, and Development of Antedon (Comatula, Lamk.) rosaceus. Proc Roy.

Soc Lond., vol. xxiv., 1876, pp. 211-231, pis. 8, 9.

Supplemental Note to the above Paper. Proc Roy. Soc, vol. xxiv., 1876, pp. 451-454.

On the Nervous System of the Crinoidea. Proc Roy. Soc, vol. xxxvii., 1884, pp. 67-76, with cut.

Chabas, F., Notice sur la Deoouverte d'une Couche aboudaute des Crinoides fossiles de TEspece Pentacriuns,
Chalon-sur-Saono, 1877, pp. 1-14, pis. i.-iii

REPORT ON THE CRINOIDEA. 421

Chapman, E. J., A Classification of Crinoids. Trans. Rmj. Soc. Canada, vol. i., 1883; sect, iv., 1882, pp. 113-116.
Dalmkr, C, Die ost-tliiiringisclien Encriniten; mit einem Vorworte von E. E. Schmid. Jcnaische Zeitschr.,

Bd. xi. pp. 382-402, Taf. xsiii.
Danielssen, D. C, and Koren, J., Fra den norske Nordhavsexpedition EcLinodermer, Iljcrimis carpenterii,

Nyt. Mag./. Naiurvid., Bd. xxiii., 1877, pp. 45-56, Tab. i., ii.
Desoe, E., Notice sur les Crinoides Suisses. Bull. Soc. Set. Nat de Neuchatd, t. i., 1845, pp. 211-222.
Notice sur la structure des Eugeniacrinus et de quelques autres fossiles analogues dc I'Oxfordien

calcaire des Laegoru (Argovie). Bull. Soc. Sci. Nat. do Nencltafel, t. iv., 1858, pp. 197-202, PI. D.
DujARDiN, F., Recherches sur la Comatule de la Mediterranee. L'Listitut, No. 119, 1835, p. 268.
Ddjardin, F., and Hdp6, H., Histoire Naturelle des Zoophytes. Echinodermes, Paris, 1862, 625 pp., 10 pis.

Crinoids in pp. 35-218, pis. 1-5.
Duncan, P. M., and Sladen, W. P., A Memoir on tlie Ecliinoderuiata of the Arctic Sea to the "West of

Greenland, vii + 82 pp., 6 pis., London 1881. Crinoidea in pp. 73-78, pi. vi.
Ellis, J., An Account of an Encrinus, or Starfish, with a jointed stem, taken on the coast of Barbados, whicli

explains to what kind of animals those fossils belong, called Starstones, Asterije, and Astropodia, which

have been found in many parts of this kingdom ; in a letter to Mr. Emanuel Mendes da Costa, r/iil.

Trans., vol. lii., part i. fur the year 1761 (1762), pp. 357-362, Tab siii., xiv.
Etheridge, R., jun., Observations on the Swollen Condition of Carboniferous Crinoid Stems. Froc. Nut. I/ist.

Soc. Gtas'joic, vol. iv., 1879, pp. 19-36, pis. i., ii.
Eudes-Desloxgschamps, Mj\r., Memoire sur la couche h Leptsena, intercalee entre le lias moyen et le lias

superieur de Calvados. Bull. Soc. Linn. Normand., vol. iii., 1858. Crinoiils in pp. 170-184, pis. vi., vii.
FoNTANNBS, F., liltudes Stratigraphiques et Pal^oritologiques pour servir a I'histoire de la Periode Tertiaire dans

le Bassin du Rhone, v., 1879. Description de quelques esj^eces nouvelles ou peu conuues. Crinoids in

pp. 50-56, pi. ii. and cut, Lyons and Paris.
Forbes, E., A History of Briti-sh Starfishes and other Animals of the Class Eohinodermata, London, 1841,

XX + 267 pp., several figs.
ilonograph of the Echiuodermata of the British Tertiaries. The Palaeontographical Society, London,

1852, vii + 36 pp., 4 pis.
Feesiinville, Memoire sur un nouveau Genre de Zoojihytes de I'Ordre des Radiaires. Bull. Soc. Philom. Paris,

Bd. ii.,1811, pp. 349, 350.
Gaudry, a., Les Encbainements du Jlonde Animal dans les Temps Geologiques. Fossiles Primaires, Paris,

1883, pp. 1-317, 285 cuts.
GiEBEL, C., Crinoideen im Kreidemergel bei Quedlinburg. Zeitschr. d. ffesammt. N'atimciss., Bd. v., 1855,

pp. 25-34, Taf. iii.
GoETTE, A., Vergleichende Entwickelungsgeschichte der Comatula mediterranea. Arcliiv f. mikroslc. Anat.,

Bd. xii., 1876, pp. 583-648, pis. xxv.-xxviii.
GoLDFUSS, G. A., Petrefacta GermaniiB, Dusseldorf, 1826-35, 3 vols, in fol., cum tab. Crinoidea in vol. i.

pp. 159-205, tab. xlix.-lxii.
Graff, L. von, Stylina comatulicola, ein neuer Sohrairotzer der Comatula Mediterranea. Zeitschr. f. m'ss.

Zool, Supph Band zum Bd. xxv., 1875, pp. 124-126.
Geay, J. E., Notice on the Digestive Organs of the Genus Comatula, and on the Crinoidea of MiUer. Annals

of Philosophi/, N. S., vol. xii., 1826, pp. 392-394.

Description of a new kind of Pear-Encrinite found in England. Phil. Mar/., vol. iv., 1828, pp. 219, 220.

Characters of a new Genus of Radiata (Ganymeda). Proc. Zool. Soc. Land., 1833, part. ii. pp. 15, 16.

Greeff, R., TJeber den Bau der Echinodermen, Vierte Mittheilung — 1. Ueber den Bau der Crinoideen.

Sitzungsh. Gesdlsch. Be/ord. gesammt. Naturiv. Marburg, 1876, No. 1, pp. 1-29, with cut.
Fiinfte Mittheilung — 3. Ueber das Herz der Crinoideen. Sitzungsb. Gesellsch. Be/ord. gesammt.

Natuno. Marburg, 1876, No. 5, pp. 88-95, 4 cuts.
Die erste Mittheilung iiber das funfkammerigo " Herz " der Crinoideen. Sitzungsh. Gesellsch. Be/ord.

gesammt. Naturw. Marburg, 1879, No. 4, pp. 52-54.

422 THE VOYAGE OF H.M.S. CHALLENGER.

GitiMM, 0. VON, Zum feiiiereii ]5au der Crinoiden. Bull. Acad. Sci. St. Petcrsh., t. xvii., 1872, col. 3-9,

with plate.
Gkixnell, G. 1!., On a new Ciinoid from the Cretaceous Formation of tlie West. Anmr. Journ. Sci. and Arts,

ser. 3, vol. xii., 1876, pp. 81-83, pi. iv.
GuuBE, E. [Descriptions of throe new Comatulee.] Jahrb. d. Schles. Gesellsch., 1875, Nat. Hist. Sect.

pp. 54, 55.
GuETTARD, J. E. (?), MtSmoivesurlcs Encrinites et les pierres Etoilees, dans lequel on traitera aussi Jes Entroques,

&c. Memoires de Mathi!'matique et de Plij'sique tir^s des Eegistres de 1' Academic Eoyale des Sciences, do-

I'aim^e mdcclv., Paris, 1761, pp. 224-263, pis. viii.-x.

SeconJe partie. Ibid., jip. 318-354, pis. xiv.-xvi.

GuiLDiNG, Notice of the Discover}' of a recent Encrinus. Zonl. Joum., vol. iv., 1828-29, pp. 173-175.
Hagenow, F. von, Mouographie der Eiigenschen Kreide-Versteinerungeu, Abth. 2, Eadiarien und Annulateii.

Neues Jahrh. f. Mineral, Jahrg. 1840, pp. 631-672, TaC. ix.
Heusingee, C. F., Bemerkungen iilier deti VeiJauungskanal der Comatulen. Archiv f. Anaf. w. Fhi/siol.,

Bd. i., 1826, pp. 317-324, mit 1 Taf.
Anatomische Untersuchungen der Comatvda mediterranea. Zeitschr. f. organ Physih,'S>A. iii., 1829,

pp 367-374, Taf. x., xi.
Jelly, H., The Lansdown Encrinite. Baih and Bristol Mag., vol. ii., 1833, No. 5, pp. 36-47, with plate.
JicKELi, C. F., Vorlaufige Mittheilungen iiber den Ban der Ecliinodermen — 1. Uber dus Nervensystem und

die Sinnesorgane der Comatula mediterranea. Zool. Anzeiger, Jahrg. vii., 1884, pp. 346-349 and

pp 366-370.
. tjber eineu der BegattuDg ahnlichen Vorgang bei Comatula mediterranea. Zool. Anzeiger, Jahrg. vii.,

1884, pp. 448, 449.
KoNiNCK, L. DE, et Le Hon, H., Eecberches sur les Crinoides du Terrain Carbonifere de la Belgique,

Bruxelles, 1854, 208 pp., 7 pis.
Krukenberg, C. Fr. W., Beitrage zur einer Nervenphysiologie der Ecliinodermen. Veigleichend-Pbysio-

logische Studien, Zweite Eeihe, 1882, Abth. 1, pp. 70-82, 2 cuts.
KuHL, VAN Hasselt, Sal. Muller, and Herklots, J. A., fichinodermes peintes d'apres Nature. Bijdragen tot

de Dierkunde. Uitgegeven door het Genootschap Natuia Artis Magistra te Amsterdam. Negende

Aflevering, 1869. Crinoids on pis. ix., x, pp. 10-11.
KuNiscH, H., Ueber den ausgewachsenen Zustand von Encrinus gracilis, Buch. Zeitschr. d. deutsch. geoL

Gesellsch., Jahrg. 1883, pp. 195-198, Taf. viii.
Lamarck, J., Systeme d'Animaux sans Vertebres, Deuxieme Edition, Paris, 1835. Crinoids in t. ii. pii.

649-876 ; and t. iii. pp 204-214.
Laube, G. C, Ueber Encrinus cassianus, Lbe., und dessen Verhiiltnisse zu bekaunten Eneriniten. Wien.

Verhandl. GeoZ., Bd. xiv., 1864, pp. 207, 208.
Die Fauna der Schichtcn von St. Cassian ; ein Beitiag zur Palaontologie der Alpinen Trias, Abth. 1,

Spongiarien, Corallen, Echiniden, und Crinoiden. Denlcsclir.d. k. Akad. d. Wtss. Wien., Bd. xxiv., 1865,

Abth. 2, pp. 223-296 (Crinoids in pp. 267-278), 10 pis.
Leach, W. E., The Zoological Miscellany; being descriptions of new or interesting animals, London, 1815.

Crinoids in vol. ii. pp. Gl-63, with plate.
Leuckart, F. S., Einiges iiber das Asteroiden-Geschlecht Comatula, Lam., uberhaupt, imd Uber Comatula

mediterranea insbesondere. Zeitschr. f. organ Physik, Bd.^iii., 1833, pp. 375-391.
In Beziehung auf den Haarstern (Comatula) und P. europaeus, so wie auf das Schmarotzerthier auf

Comatula. Froriep, Notizen, Bd. 1., No. 1087, 1836, col. 129-131.
LiNCK, J. H, De Stellis Marinis liber singularis, Lipsia?, 1733, xxii + 107 pp., 42 pLs.
Linnaeus, C, Systema Naturre, ed. xiii., cura. J. F. Gmelin, Lipsise, 1793. Crinoids in t. i. pars vi. pp. 3166,

3167, 3794.
Loriol, p. de, Monographie des Crinoides Fossiles de le Suisse, Genfeve, 1877 a 1879, pp. 1-300, pis. i.-xxi.

(Eeprinted from the Mhn. Soc. Pal. Suisse).

REPORT ON THE CRINOIDEA. 423

LoRioL, P. DE, ISTotice sur le Pcutacriuiis de Seunecey-le-Graud, aver, uu Travail sur la couche calcaire qui le

continent, par M. Delafond, et un Pr^ambule, par M. F. Chabas, Chalon-sur-Saone, 1878, viii + 14 pp.,

2 pis.
— - Les Ci'inoides fossiles de la Suisse. Assoc. Frnnr. pour Cavancement des Sciences. Compte rendu

de la S"' Session, Montpellier, 1879, pp. 627-636.

— Description de quatre lichinodermes Xouveaux. Mem. Soc. Pal. Suisse, vo\. vii., Geneve, 1880, 15 pp., 1 pi.

Ifote sur le Genre Apiocrinus. Assoc. Fixitk;. pour Tavancement des Sciences. Compte rendu de la

IP Sos.sion, La Rocbelle, 1882, pp. .334-338.
Description of a new Species of Bourgueticrinus. Juurii. Cincinn. Soc. Nat. Hist., vol. v., 1882, p. 118,

pL V.

Paleontologic Fran^aise, Terrain Jurassique, t. xi., Paris, 1882-84. Crinoide.s, pp. 1-432, pis. 1-107

(all published).
LovifiN, S., Phanogenia, ett hittills okiindt slagte af fria Crinoideer. Ofversigt k. VefensA:-A/Md. Forliandl.,

1866, pp. 223-233, cut.

Om Leskia mirabilis, Gray. Ofversigt It. Vetensk.-Akad, Forhandl., 1867, pp. 431-440.

Note on Hyponome Sarsi, a recent Cystidean. (Reproduced from Forhandl. Skand. Naturf. Christiania,

1868, vol. X. p. liv.) Ann. and Mag. Nat. Hist., ser. 4, vol. iv., 1S69, pp. 1.59, 160.
litudes sur les lilchiunid^es. K. Svensk. Vetensk. Akad. Handl., Bd. xi. part ii., 1874, pp. 1-91,

+ A-H, pis. i.-liii., cuts.
On Pourtalesia, a Genus of Ecliinoidea. A'. Svensk. Vetensk. Akad Hand!., Bd. xix., No. 7, 1883,

pp. 1-95, pis, i.-xxi., cuts.
LuDWiG, H., Zwv ^\iiatomie der Crinoideen. Zeitsckr. f. wiss. ZooL, Bd. xxvi., 1876, pp. 361, 362.
Beitriige zur Anatomic der Crinoideen. Nachricht. v. d. kgl. Gesellsch. d. Wiss. zii Gottingcn, No. 5,

1876, pp. 105-114.
Beitrage zur Anatomie der Crinoideen, liter Artikel. Nachricht. v. d. kgl. Gesellsch. d. Wiss. zu

Giittingen, No. 13, 1876, pp. 1-9,
Zur Atiatomie des Rhizocrinus lofotensis, M. Sars. Nachricht. v. d. kgl. Gesellsch. d. Wiss. zu Gottingen,

No. 23, 1876, pp. 675-680.
Beitrage zur Anatomie der Crinoideen. Zeitsckr. f. wiss. ZooL, BJ. xxviii., 1877, pp. 255-353,

pis. xii.-xi,
Zur Anatomie des Rhizocrinus lofotensis, il. Sars. Zeitschr. f. iciss. ZooL, Bd. xxix., 1877, pp. 47-76,

pis. v., vi.

■ Echinorlermenstudicn, Zool. Anzeiger, Jahrg. ii., 1879, pp. 540-542.

Uber die priraiiren Steinkanal der Crinoideen, nebst vergleichonJ-anatomischen Beraerkungen iiber die

Echinodermen iiberhaupt Zeitschr. f, wiss. ZooL, Bd. xxxiv., 1880, pp. 310-332, pis. xii., xiii.
Ueber einige seltenere Echinodermen des Mittelmeeres. MittheiL d. ZooL Stat, zu Neapel, Bd. ii.,

18S0, pp. 53-71, Taf. iv.

Die Eildung der Eihiille bei Antedon rosacea. Zool. Anzeiger, Jahrg. iii., 1880, pp. 470, 471, 3 cut.s.

LuNDGREN, B., Om en Comaster ocli en Aptychus frSn Kopinge. Ofversigt k. Vetensk.-Akad. ForhandL,

1874, pp. 61-74, Taf. iii.

Anteilon Fischeri. Neices Jahrh. f. Mineral., Jahrg. 1876, pp. 180-182.

LiJTKEN, C. F., Om Vestindiens Pentacriner med nngle Bemaerkninger om Pentacriuer og Solilier i Almin-

delighed. Vidensk. MeddeL f d. nat. Foren. i Kj<;ihenham, 1864, No. 13-16, pp. 195-245, Tab. iv., v.
Communication of Loven's article on Leskia mirabilis, Graj'. Canad. Nat., N. S., vol. iii., 1868,

pp. 437-442.
Hyponome Sarsi, a recent Australian Echinoderm, closely allied to the Palajozoic Cystidea, described by

Prof. Loven ; with some remarks on the mouth and anus in the Criuoidea and Cystidea. Canad. Nat.,

N. S., vol. iv., 1869, pp. 267-270.
Endnu et Par Ord om de gamie Soliliers " Snabel " og ilund, with French resumd. Vidensk. MeddeL

f. d. nat Foren. i KJ^henhavn, 1869, No. 9 13, pp. 160-188, with cuts.

424 THE VOYAGE OF H.M.S. CHALLENGER.

Marenzelleb, E von, Die Ccelenteraten, Ecliinodei-men uud AViirmer dcr k. k. Osterreichisch-Uiigarischeti

Nord Pol. -Expedition. Denkschi: d. It, Akad. d. Wiss. Wien, Math.-Naturwiss. Classe, Bd. xxxv, 1877

pp. 357-398, 4 pis.
Marion, A. F., Draguages au Large de Maiseille, i. Ann. d. Sci. Kut., G' serie, Zool., t. viii., 1879, 48 pp.

pis. XV. -.X viii.
Marshall, A. M., On the Nervous System of Antedon rosaceiis. Qnurf. Jonni. Micr. Sch, N. >S., vol. xxiv.,

1884, pp. 507-548, pi. xxxv.
M'CoY, r., On some new IVIesozoic Eadiata. Ann. and Mag. Kaf. iri,<t., -ser. 2, vol. ii., 1848,

pp. 397-420.
Meckel, J. ¥., Ueber die Oeffnnngen des Speisekanals Lei den Comatnlun. Deutsche^ Archiv f. d. PhytiivL,

Bd. viii., 1823, pp. 470-477.
Meek, F. B., Note on tbo new Genus LTiutacrinus, Grinnell. Bidl. U.S. Oeol. Geogr. Survey Tcrrit., vol. ii.,

No. 4, 1876, pp. 375-378, cuts.
Meek, F. B., and Worthen, A. H., Palajontology of Illinois, vol. v., 1873, pp. 323-619, pis. 1-32.
Meneghini, G., I Crinoidi Torziarii. JHi dellw Soc. Tone. d. Sci. Nat, vol. ii., 1875, pp. 36-59.
Nuove osservazioni sui Crinoidi terziarii. Proe. verb, della Soc. Tosc. d. Sri. Nat, 7 Juglie 1878, xxv.

pp. xsxi, xxxii.
Merlan, p., Beitriige zur Kennlniss des Crinoiden des Juia. Berirld ii. d. Vcrsamml. deidsch. Nuturf. in Basel,

vol. viii., Basel, 1849, pp. 27-29.
Metschnikoff, E., Beitriige zur Entwickelungsgeschichte einigerniederen Thiere. Bull. Acad. Sci. St Petersh.,

t. XV., 1871, pp. 502-509.
Meter, C. E. H. von, Isocrinus und Chelocrinus. ]\ruseum Senkenbergianum, Frankfurt, 1837, Bd. ii.

pp. 249-263, Taf. xvi.
Fiscbe, Crustaceen, Eoliinodermen, und andere Versteinerungen aus dem Muschelkalk Oberschlesiens.

Palmontographica, Bd. i., 1851. Crinoideeu in pp. 260-275.
MiCHELiN, H. , Description d'un nouveau genre de la famille des Crinoides. Rev. ct Mag. ZooL, 2"" ser. t. iii. ,

1851, pp. 93, 94.
Miller, J. S., A Natural History of the Crinoidea, or Lily-shaped Animals ; with observations on the genera

Asteria, Euryale, Comatula, and Marsupites. Bristol, 1821, viii + 150 pp., 50 pis.
Miller, S. A., Description of three new Orders and four new Families, in the class Echinodermata, and eight

new Species from the Silurian and Devonian Formations. Journ. Cindnn. Soc. Nat. Hint., vol. v., 1883,

pp. 221-231, pi. Lx. ■
Mitchill, S. L., a Group of Polypes belonging to the family of Comatula, with an extraordinary form and

configuration, from the Indian Seas. Amer. Journ. Sci. and Arts, vol. v., 1822, pp. 46, 47.
Moore, C, On the Presence of the Genera Plicatocrinus, Cotylederma, and Solanocrinus in British Strata. GeoL

Mag., N. S., Dec. IL vol. ii., 1875, pp. 626, 627.
MoKlfeRE, J., Note sur les Crinoides des Terrains Jurassiques du Calvados. BnU. Soc. Linn. Normand., s6r. 3,

vol. iii., 1879, pp. 323-332.

Crinoides des Terrains Jurassiques du Calvados. Ball. Soc. Linn. Normand., ser. 3, vol. iv., 1880,

pp. 1-22, pis. i., ii.
Deux Genres de Crinoides de la Grande Oolithe. BidL Soc. Linn. Normand., sdr. 3, vol v., 1881,

pp. 78-87, and pi.
MCllbr, J., Ueber den Bau des Pentacrinus caput-^Iedu'ee. Abhandl. d. I,: Akad. d. Wiss. Berlin,

1834, pp. 177-248, 6 pis. Abstract in Monatsher. d. k. preuss. Akad. d. Wins. Berlin, 1840, pp. 88-100.
Ueber die Gattungen und Arten der Comatulen. Monatsher. d. k. preiiss. Akad. d. Wiss. Berlin, 1841,

pp. 179-189.
Neue Beitriige zur Kenntniss der Arten der Comatulen. Archiv f. Nattirgesch., Bd. ix., 1843,

pp. 131-136.
Nachtrag zu der Abhandlung iiber die Comatulen. Monatsher. d. k. preuss. Akad. d. Wiss. Berlin,

1846, pp. 177-179.

REPORT ON THE CRINOIDEA. 425

MuLLEK, J., Ueber die Gattung Comatula, Lam., uml ihrp, Arten. Abhandl. <1. k. Akad. d. Wins. Berlin,

1849, pp. 237-265.
Ueber den Bau tier Ecliinodermen. Ahhandl. d. /,: Alcad. d. Wiss. Berlin, Ed. xliv., 1854,

pp. 123, 219, 9 pis.
Neumayr, M., Morphologische Studien iiberfossile Ecbinodermen. SitzungaJi. matli.-nat. CI. li. Akad. Whs.

Wien, Bd. Ixx.xiv., 1881, pp. 143-176, 2 pis.
Norman, A. M., On the Genera and Species of the British Echinodermata — Part. I. CrinoiJea, ( tphiuroidca,

Asteroidea. Ann. and Ma;/. Nat. Hid., sen 3, vol. xv., 1865, pp. 98-129.
Oersted, A. S., Descriptions of Pentacrinus caput-Medusie and of Pentacrinus miilleri. Fijrhundl. SIcand.

Naturf., 1^^ Mode i Christiania, 1856, p. 202.
d'ORBiGNY, A., Menioiro sur une secoiide espcce vivante de la famille'des Crino'ides on Encrines, servant

de type .au Nouveau genre Holope (Holopus). Magas. de Znnl., 7™*^ ann., Paris, 1837, CI. x. pp. 1-8,

pi. 3.

• Cours elementaire de Paleontologie et de Gtelogie stratigraphiqo, Paris. 1850-52, 3 vols.

■ Prodrome de Paleontologie stratigraphique universelle des Animaux jNIoUusques et Kayonnties, Paris,

1850-51, 3 vols.

Histoire naturelle gonerale et particiiliere des? Crino'ides vivans et fossiles, comprenant la description

Zoologiqne et G(5ologique de ces animaux, Paris, 1858, 97 pp. 18 pis.
Pearce, J. C, On the Locomotive and Non-Locomotive Powers of the family Crinoidea. Proc. Geol. Soc. Lond.,

vol. iv., 1843, pp. 159, 160.
Perkier, E., Eecherches sur I'Anatomie et la Regeneration des Bras ile la Comatula rosacea (Antedon rosaceus,

Linok.). Areliives d. Zool. exper., t. ii., 1873, pp. 29-86, pis. ii.-iv.

Note sur les Brisinga. ComiHe.'; rendus, t. xcv., 1882, pp. 61-63.

Sur une Asterio des graiides profondeurs de I'Atlantique, pourvue d'un pedoncule dorsal. Comptes

rendus, t. xcv., 1882, pp. 1379-1381.
Sur un nouveau Crinoide fixe, le Democrinus Parfaiti, provenant des dragages du "TravaiUeur."

Comptes rendus, t. xcvL, 1883, pjs. 450, 451.
Sur des Eudiocrinus de I'Atlantique et sur la nature de la faune des grandes profondeurs. Comptes

rendus, t. xcvi., 1883, {)p. 725-728.

■ lichinodermes ; sur I'organisation des Crinoides. Comptes rendus, t. xcvii., 1883, pp. 187-189.

Sur le d(Sveloppement des Comatules. Comptes rendus, t. xcviii., 1884, pp. 444-447.

Anatomie des lilchinodermes ; sur I'organisation des Comatules adultes. Comptes rendus, t. xcviii.,

1884, pp. 1448-1450.

Memoire sur les Etoiles de Jler recueillies dans la mer des Antilles et le Golfe du Mexique durant les

expeditions de dragage faites sous la direction de M. Alexandre Agassiz. Nouo. Archiv. du Mus. d'Hist.,

Nat., s^r. 2, t. vi., 1884, pp. 126-276, pis. i.-x.
Philippi, R. a., Alecto alticeps, n. sji., eine tertiare Comatula-Art von Palermo. Neues Jahrh. f. Mineral.,

Jahrg. 1844, pp. 540-542.
PiCAED, K., Ueber eine neue Crinoideen-Art aus dem Muschelkalk der Hainleite bei Sondershausen. ZeitscJir.

d. deutscJi. geol. Gesellsch., Jahrg. 1883, pp. 199-202, Taf. ix.
PicTET, E. J., Traite de Paleontologie, ou Histoire Naturelle des Animaux Fossiles considiSr& dans leurs

Rapports Zoologiques et Geologiques, Seconde Iildition, Paris, 1857. Crinoidea in t. iv. pp. 278-345,

pis. xcix.-cii.
PouKTALfcs, L. F. DE, Contributions to the Fauna of the Gulf Stream at great Depths. Bidl. Mus. Comp.

Zuol, vol. i., Nos. 6, 7, 1867-68, pp. 103-142.
List of the Crinoids obtained on the Coasts of Florida and Cuba by the United States Coast Survey

Gulf Stream Expeditions in 1867, 1868, 1869. Bull. Mus. Comji. Zool, vol. i.. No. 11, 1869,

pp. 355-358.
Zoological Results of the llassler Expedition — Crinoids and Corals. ///. Cat. Mus. Comp. Zool.,

No. 8, 1874, pp. 27-52, pis. v.-.\.

(zool. chall. EXP. — PART xxsii. — 1884.) Ii 54

426 THE VOYAGE OF H.M.S. CHALLENGER.

PouRTALfes, L. F. DE, Eopoi'ts on the results of Dredging, under the Supervision of Alexander Agassiz,
in the Gulf of Mexico, by the United States Coast Survey steamer " Blake," Lieut.-Comniander
C. D. Sigsbee, U.S.N., Commanding — Crinoids. Bull. Mas. Comp. Zu'Ol., vol. v.. No. 9, 1878, pp. 214-
216. Description of a young Holopus, with two figures (pi. ii.), by A. Agassiz, on p. 213.

QuENSTEDT, F. A., Ueber die Encriniten des Muschelkalk.s. Arcliiv f. Naturgesch., Jahrg. i., ISS.'i, Ed. ii.
pp. 223-228, Tab. iv.

Sehwaben's Medusenhaupt — Eine Monographic der Subangularen Pentaoriniten, pji. 1-7.", mit

1 grossen Tableau in 4 Blattern, Tiibingen, 1868.

Petrefactenkunde Deutschlands, Ed. iv. Die Asteriden und Encriniden nebst Cysti- und Elasttiidecn.

Nebst einem Atlas von 2.5 Tafeln, Leipzig, 1876, viii + 742 pp.
Eathbun, E., a List of the Brazilian Echinoderms, with Notes on their l)isti-ibution, &c. Trait--!. Cnniiini.

Acad., vol. v., 1879, pp. 139-158.
Eetzius, a. J., Anmarkningar vid Asterias Genus. Nija K. Svensl: Vefenisl: Alccul. Handh, Ed. iv., 1783,

pp. 234-243.

Dissertatio sistens species cognitas Asteriarum, Lundae, 1805, 37 pp.

Sars, M., Bemaerkninger over Brisinga endecacuemos, Asbjcirnsen, og Comatula Sarsii, Diib. et Kor. Forliamll.

Skand. Naturf., 1^" Mode i Christiania, 1856, pp. 209-211.
Om Comatula Sarsii, Diib. et Kor., i fastsiddende eller Pentacrinus-tilstaud. Forlmndl. Shand. Naturf.,

1'^^ Mode i Christiania, 1856, pp. 212-216.

M^moires pour servir a la coimaissance des Crinoides vivants, Christiania, 1868, 65 pp. 6 pk.

Savigny, J. C, Explication sommaire des Planches d'Echinodermes de I'l^lgypte et de la Syrie, publi^es par

Jules-Cdsar Savigny ; offrant un Expose des caractferes naturels des genres avec la distinction des Especes,

par Victor Audouin. Description del'lilgypte. Seconde lildition, t. xsiii., Paris, 1828. Histoire Naturelle,

Zoologie. Crinoids in pp. 1-5, pi. i.
ScHLiJTEE, C, Ueber einige astylide Crinoiden. Zeitschr. d. deutsch. ijeol. Gesellsch., Jahrg. 1878, pp. 28-66,

Taf. i.-iv.
ScHULTZE, L., Monograpliic der Echinodermen des Eifler Kalkes. Denlisrhr. d. k. Ahad. d. Wiss. Wkn, Mat.-

Naturwiss. Classe, Bd. xxvi., 1866, pp. 113-229, Taf. i.-xiii.
Semper, C., Kurze anatomische Bemerkungen iiber Comatula. Arbeit, aus d. Zool.-Zootom. Institut in WUrz-

burg, Bd. i., 1874, pp. 259-263, with cut.

— Ophiocrinus, eine neue Comatuliden Gattung. Archiv f. Naturgesch., Bd. xxxiv., 1868, pp. 68, 69.

Sladen, W. p.. On the Homologies of the Primary Larval Plates in the Test of Bracliiate Echinoderms.

Quart. Journ. Micr. Set., N. S., vol. xxiv., 1884, pp. 24-42, pi. L
Steenstrup, J. J., Cyathidium holopus. Bericht ii. d. Versamml. deutsch. Naturf. in Kiel, 1846, p. 150.
Stelznee, Ald., Ein Beitrag zur Kenntniss des Versteinerungs-Zustandes der Crinoideenreste. Neues Jahrb.f.

Mineral, Jahrg. 1864, pp. 565-579, Taf. x.
Strombeck, a. von, Ueber Missbildungen von Encrinus liliiformis. Lam. Fata-imtograpJiica, Bd. iv., 1855,

pp. 169-178, Tab. xxxi.
Teuscher, R., Beitrage zur Anatomie der Echinodermen — I. Comatula mediterranea. Jcnaische Zeitschr.

Bd. X., 1876, pp. 243-262, Taf. vii.
TnoMPSON, J. v.. Memoir on the Pentacrinus Europteus : a recent Species discovered in the Cove of Cork,

July 1, 1823. Cork, 1827, 12 pp., 2 pis.
Memoir on the Star Fish of the Genus Comatula, demonstrative of the Pentacrinus europaeus being the

young of our indigenous species. JEd/ii. New Phil. Journ., vol. xx., 1836, pp. 295-300, pi. ii. Abstract

in Proc. Bo;/. Soc, 1835, p. 339.
Thomson, C. W., On a new Palaeozoic group of Echinodermata. Ediji. New Phil. Journ., vol. xiii., 1861,

pp. 106-117, pis. iii., iv.

Sea Lilies. The Jntellectual Observer, vol. vi., No. 31, August 1864, pp. 1-11, with plate.

On the Embryogeny of Antedon rosaceus, Linck (Comatula rosacea of Lamarck). Phil. Trans., vol. civ.

1865 pp. 513-544, pis. xxiii.-xxvii. Abstract in Proc. Roij. Soc, vol. ix., 1857-59, pp. 600-601.

REPORT ON THE CRINOIDEA. 427

Thomson-, C. W., On tlio Structiirf of the PakBOKoic. Ciiiioids. Abstract ia Proc. Roy. Soc. Edin., vol. vii.'

1869-72, pp. 415-418.
On the Crinoids of the "Porcupine" DeepSea Dredging E.^pedition. Proc. Roij. Sue. Edin., vol. vii.,

1869-72, pp. 764-773.
Notice of new Living Crinoids belonging to the Apiocrinida. Jniirn. Linn. Soc.'Limd. (Zool.), vol. xiii.,

1876, pp. 47-55, 5 cuts.

Desciiption of Pentacrinus maclearanus, in The Atlantic. London, 1877, vol. ii. pp. 123-126.

On the Structure and Relations of the Genus Holopus. Abstract in Proc. Roy. Soc. Edin., vol. ix.,

1875-78, pp. 40.5-410.
Trautschold, H., Ueber die Bezeichnung der Kelchplatten der Crinoideeu. Bull. Soc. imp. des Nat. Moscoii,

t. Ivii., 1882, No. 3 (1883), pp. 201-203.
Wachsmuth, C, Notes oa the Internal and External Structure of Pahvozoic Crinoids. Amei: Journ. Sci. and

Arts, vol. xiv., 1877, pp. 115-127, and pp. 181-191.
Wachsmuth, C, and Springer, F., Transition Forms in Crinoids and Description of five new Species. Proc.

Acad. Sri. Nat. Philcul, 1878, pp. 224-266, 2 pis.
Revision of the Palreocrinoidea — Part I. The Families Ichthyocrinidaj and CjathocrinidiT. Proc. Acad.

Sci. Nat. Phi/ad., 1879, pp. 226-378, pis. xv.-xvii.

Revision of the Paleeocrinoidea — Part II. Family Sphceroidocrinidse, with the subfamilies Platy-

crinidae, Rhodocrinidiie, and Actinocrinidaj. Proc. Acad. Sci. Nat. Philad., 1881, pp. 177-411,

pis. xvii.-xix.
Weinberg, A., Die Morphologic der lebenden Crinoideen mit Beziehung auf die Form Antedon rosacea,

Linck. Der Naturhidoriker, Jahrg. v., 1883, pp. 266-307.
Winkler, T. C, Description d'un Crinoide et d'uu poisson du Systeme heersien. Archiv du Mtn<. Tcyler.

Harlem, vol. ii., 1869, pp. 297-307, pi. xxxviii.
Wright, T., On the Occurrence of the Genus Cotylederma in the Middle Lias of Dorsetshire. Geol. Mag.,

N. S., Decade IL, vol. ii., 1875, pp. 505-506.
ZiTTEL, K. A., Handbuch der Palajontologie. Palffiozoologie, Bd. i. Abth. 1, 1876-80. Crinoidea in

pp. 315-437, 96 cuts.
Ueber Plicatocrinus Fraasi aus dem oberen weisseu Jura von Nusplingen in AViirtemberg. Sitzuntjsh.

d. II. CI. k. haier. Akad. d. Wiss. Wien., 1882, pp. 105-113, Taf. i., ii.

The following .short notices of recent Crinoids seem worth quoting in extenso. We
have, unfortunately, no further knowledge of any of the types to which they relate.

" Une espece nouvelle d'Encrine vivante a dt6 d^couverte par le r^v^rend C. PleydeU a Newcastle sur la
riviere Hunter, dans la Nouvelle Hollande ; I'auteur propose de lui donner le nom d'Encrinus australis. Elle
n'a pas de colonne vertebrale, mais le corps de I'animal a environ un cinquierae de pouce de long, et est termini
dans cette direction par une base circulaire. A I'extremite opposee du corps sont attaches cinq appendices
claviculaires." L'Institut, 1845, p. 292.

Is this a Crinoid at all 1

IL

" Professor Sigmund Schultze aus Greifswald zeigte einen neuen Pentacrinus, welcher ihm von Amboina
iibersandt ist, in Abbildungen vor und sprach iiber die drei Arten, welcher cr in dieser seltenen, bisher nur in

428 THE VOYAGE OF HM.F, CHALLENGER.

fiieben Exemplaren bokannten Thiei-gattung unteracheidet ; der Pentacrinus Gueliardi, der Pnutacrimtn caput-
Mediisx mid der rentacrinus, Arndiii." Berieht ii. d. Versamml. deidsch. Naiurf. in Karlsruhe, 1858, p. 293.
Two new species, Pentacrinus gueitardi and Pentacrinus amdtii, are mentioned here, but nothing is said
as to wliich came fromAniLoina, nor as to the nature of the other. The Amboina specimen was probably
a Mitucrinus.

111.

" I learn from a correspondent at Jlclboume, Mr. -J. S. Poore, that during his visit to King George's
Sound, Western Australia, he there dredged up from 8 fathoms a living Encrinite. The stem, which was
attached to a stone, was about 6 inches long ; the arms about 1 1 inch, of a beautiful rose-colour or pink, fading
to white." Sir. E. Owen in Ann. and IJai/. Nat. Hist., ser. 3, voL ix., 18G2, p. 486.

This may perhaps have been a Pentacrinoid larva, but if so, it was of most unusual size. Further iufornia-
tion about it from Australian naturalists would be most valuable.

INDEX TO AUTHORS QUOTED.

Agassiz, A., 12, 18, 57, 130, 197, 204, 210, 30G, 309,
331, 333, 390, 393, 398, 401, 413.

Agassiz, L., G t, 119, 197, 270-272.

Ailmau, a. J., 161, 164, 185.

Angelin, N. P., 41, 65, 66, 84, 172, 175, 17G.

Apostolides, N. C, 101.

(VArchinc, A., 245, 247, 32.3.

Austin, T., 39, 183, 187, 272, 274-278, 296, 301, 306,
388.

Baily, W. H., 284.

Baud^lot, C, 117, 118.

Barrois, C, 154.

Bell, F. J., 30.3.

Beyrich, E., 37, 126, 223, 245, 275, 294.

Billings, E., 43, 75.

BlainviUe, H. M. D. de, 187, 272, 300, 303.

Bronn, H. G., 145, 188, 190, 191, 271, 294.

Buokland, W., 22, 39, 81, 272, 277, 300.

Buoh, L. von, 190.

Burmeister, H., 186, 187, 191.

Carpenter, P. H., 2, 4, 18, 20, 25, 26, 35, 36, 39, 42, 46,
47, 49, 51, 52, 56, 57, 59, 60, 88, 94, 101,
102, 104-107, 143, 158, 16.5, 168, 169, 189,
213, 217, 227, 24.5, 250, 255, 259, 262, 268,
269, 273, 282, 284, 288, 293, 301, 306, 328,
331, 340, 343, .357, 370-372, 393-39.5, 3S8,
402, 40.3, 406, 408, 415, 416.

Carpenter, W. B., 1, 2, 8, 36, 40, 48, 50, 58, 59, 63,
67, 73, 87, 107, 108, 114, 116, 118, 132, 133,
245, 292, 402, 408, 409, 412.

Carpenter and Etheridge, 37, 49, 145, 147, 158.

Chapman, E. J., 151, 195, 196.

Charlesworth, E., 192.

Claus, C, 403, 409.

Cole, E., 18.

Danielssen and Koran, 8, 29, 71, 225-227, 231-234,
242, 401.

Desor, E., 272, 281.

Dixon, F., 34, 284.

Dujardinand Hup6, 191, 192, 197, 273, 301.

Ellis, J., 272, 274, 294, 300.

Emmons, S. F., 298.

Etheridge, E., 276, 296.

Etlieridge, E., jun., 135.

Eudes-Deslongschamps, E., 20, 131, 213.

Filhol, H., 244, 269, 31.5-317, 319.

Forbes, E., 35, 144, 187, 190, 192, 245, 2G9, 272-274,
281.

Gegenbaur, C, 409.

Goette, A., 71, 157, 158, 167, 169, 411.

Goldfuss, G. A., 35, 158, 187, 272, 274, 284, 294, 302,
403.

Graff, L. von, 134, 135, 324, 347, 364.

Greeff, E., 120, 406, 408.

Grimm, 0. von, 404.

Gu6ttard, J. E., 272, 274, 294, 300.

Hagenow, F. von, 323.

Hall, J., 19, 20, 53,87, 193, 217.

Hall and Whitfield, 298.

Hamann, 0., 112, 124, 410.

Hertwig, O., 115.

Hertwig, E., 114.

Hoek, P. P. C, 351.

Hoernes, E., 393, 394, 403.

Huxley, T. H., 75, 188, 192.

.Tetfreys, J. G., 281, 313-31.5, 392.

Jickeli, C. F., 408, 413-416.

Koehler, E,, 101, 10.5, 106, 112, 406, 407.

Koninck, L. G. de, 153, 223.

Lamarck, J., 272, 300.

Lankester, E. Eay, 186, 189.

Laulie, G. C, 296, 414.

Leuckart, E., 186, 188, 190, 191, 193.

Linnaeus, C, 272, 300, 302.

Loriol, P. de, 9, 1.5, 17, 22, 25, 26, 3.5, 47, 126, 131,
1.3.5, 142, 146-148, 153, 183, 186, 188, 189,
201, 202, 212-216, 223, 225, 227, 245, 248,
256, 257, 270, 272-274, 276, 277, 281, 286,
297, 299, 403.

Lovcn, S., 4, 67, 85, 368, 393-400, 402, 414.

480

THE VOYAGE OF H.M.S. CHALLENGER.

l.uJwig, II., G7, 8G, 88, 89, 92-96, 101, 102, 104-106,
108, 111-113, 116-120, 124-127, 209, 210,
229, 230, 248, 250-254, 2G1, 268, 395, 398,
400, 404-406, 408, 409, 411, 412, 416.

Liitken, C. F., 4, 85, 145, 146, 168, 248, 273, 277, 283,
301-306, 308, 309, 311, 330, 333, 394.

Manzoui, A., 245.

Marshall, A. M., 255, 408-412.

Meek, F. B., 45, 53, 148, 184.

Meek and HayJen, 297, 298.

Meek and Worthen, 41, 61, 75, 87, 162, 163, 177,
217.

Meneghini, G., 238, 24.5-248, 258.

Meyer, A. B., 130.

Meyer, H. von, 34, 271, 272, 294.

Michelin, H., 216.

Miller, J. S., 73, 14.5, 151, 183, 189, 191, 272-27.5,

277, 278, 300, 302, 304, 306, 3U8, 397.
Miller, S. A., 20, 42-44, 183, 185, 192.
jVIilue-Edwards, A., 237, 269.

Moseley, H. K, 128, 129, 210, 320, 346, 348, 352,

354, 356, 359, 362, 36.5.
Miiller, .1., 2, 5, 8, 9, 49, 66, 73-77, 8.5, 86, 117, li5,

147, 170, 171, 179, 190, 191, 195, 196, 272,

274, 300, 302, 304, 306, 308, 394, 397, 404,

40.5, 415.
Miiller, P. L. S., 300, 302.
Nicholson and Etheridge, 192.
Oersted, A. S., 301, 306, 307.
d'Orbigny, A, 9, 2.5, 54, 142, 183, 192, 197, 199,

245-249, 270, 272, 274, 370.
Parra, A., 301.
Perrier, E., 28-30, 47, 94, 101, 102, 104, 106-108,

117-119, 122, 127, 195, 244, 245, 262-266,

268, 269, 316, 398-402, 404-408, 416.
Perrier and Poirier, 406.
PliiUips, J., 276.
Picard, K., 270, 294.
Pictet, F. J., 191, 192, 197, 273, 274.
Pourtales, L. F. de, 65, 160, 197, 204, 245, 246, 248-

250, 258, 259, 261-263, 265, 268, 306, 309,

330, 331.
Quenstedt, A., 4, 7, 15, 16, 20, 22, 24, 25, 34, 39, 54,

55, 142, 183, 187, 191, 211, 257, 273, 276-

278, 281, 284, 286, 294, 296, 299, 308, 323.

Ramsay, A., 276.

Roemer, F., 48, 160, 186, 187, 190, 197, 211,
213.

Romanes and Ewart, 112, 115.

Salter, J. W., 192, 193.

Sars, M., 25, 28, 29, 72-74, 90, 107, 130, 245, 246,
248-254, 259, 261-265, 266, 268.

Schlotheim, E. F. von, 272, 273, 294, 300, 407.

Schluter, C, 35, 144, 147, 148, 151,212, 213,371,
402, 403.

Schultze, L., 41, 47, 48, 152, 158, 160.

Selenka, K, 412.

Semper, C, 108, 110, 130, 133, 236.

Sigsbee, C. D., 332.

Simroth, H., 2, 116.

Sladen, W. P., 106, 168, 195, 303, 340, 368, 394-
396, 398-401.

Steenstrup, J. J. S., 211.

Strombeck, A. von, 294.

Stiider, T., 398.

Suhm, E. von W., 133, 134, 369.

Teuscher, R., 118, 120, 404, 406.

TUel,li., 122.

Thomson, C. Wyville, 1, 4-6, 8, 12, 18, 19, 21, 22,
32, 37, 39, 48, 52, 54, 60, 67, 72, 77, 85, 91,
94, 95, 107, 110, 127, 128, 188, 192, 193, 197-
199, 201, 202, 204, 206, 210, 217-219, 221,
222, 224-226, 230-235, 238-245, 249, 259,
261-263, 273, 278-281, 285, 289, 301-306,
308, 309, 312, 313, 315-317, 319, 322, 330,
331, 333-335, 337, 340, 351, 355, 360, 390,
402.

Trautschold, H., 152.

Verrill, A. E., 130.

Wachsmuth, C, 63, 65, 76, 146, 147, 156-182.

Waclismnth and Springer, 24, 32, 39, 41, 42, 46-48,
54, 61, 62, 64-66, 75, 76, 86, 87, 149, 150,
152-157, 159-163, 165-171, 173-17.5, 177,
178, 180-184, 19.5, 217, 258, 371, 372.

Weinberg, A, 407-409, 415.

Welldon, W. F. R., 407.

Zittel, K, 25, 26, 47, 48, 51, 66, 142, 145, 147, 152,
157, 158, 160-163, 165, 170, 185-189, 191,
197, 213, 214, 223, 245, 248, 258, 270, 273,
274, 276, 402, 403.

GENERAL INDEX.

The figures in dark type indicate the page on which the genus or species is hrst described.

Abactinal side, 157, 158, 171.
Abactinal system, 156, 169, 172.
Actinal side, 156, 158, 171, 179.
Actinal system, 156, 169, 172, 178.
ActinocrinidEe, 61, 62, 86, 149, 158, 164-181.
Adinocruim, 62-66, 156-158, 164-166, 169, 172,
177-180, 182, 183, 185, 192, 394.
rugosus, 75.
vemeuilianus, 165.
Actinoidea, 186, 187, 190.

Adinomefra, 33, 36, 59, 68, 74, 78, 85, 93, 95, 97,
102, 104, 111, 115, 121-123, 127,
143, 153, 154, 284, 291, 337, 377,
381-385, 411; arms of, 55-57, 60,
69, 70; digestive tube of, 91, 92;
disk of, 69, 70, 85 ; geographical and
bathymetrical distribution of, 136-
141 ; mouth of, 69, 403 ; plexiform
gland of, 103, 104.
bennetti, 55, 127, 337.
dismnilis, 110, 111.
fimhriata, 52.
jukesi, 67, 69, 85, 91, 130, 132, 133,

319, 337 (FL Iv. fig. 1).
loveni, 132.
magnifica, 57, 69, 91, 113 (PI. hi.

fig. 7).
meridinnalis, 280, 337.
mult ir ad lata, 49, 52, 337.
7ngm, 96, 110, 120-122, 124, 416

(PI. Ixi. fig. 6).
nobill-; 55, 57, 69, 110, 111.
novcB-guinecB, 49.

parviciira, 50, 52, 57, 67, 102, 103,
107, 120, 121, 124, 133, 283, 337,
(PI. hxi. figi 2-5).

Adinometra jndcheUa, 91, 103, 104, 109, 137, 337
(Pi. Ix. fig. 1 ; Pi. Ixi. fig. 1).
rohusta, 144.
srhlegeli, 55.

Solaris, 10, 49, 51, 52, 58, 91, 280.
stellata, 132, 319.

stelligera, 69, 70, 337 (PI. Ivi. fig. 8).
strata, 60, 67, 69, 85, 86, 129, 130, 133

(PI. liv. figs. 10, 11; PI. Iv. fig. 2).
tracliijgaster, 127.
typica, 4, 10, 49, 51, 52, 337.
Actinozoa, 193.,

Agassizocnnus, 132, 153, 213, 217, 319.
Agelacrinidae, 194.
Agelacrinoidea, 192.
Agelacrinus, 85, 192.
Allagecrliius, 146, 147, 152, 158, 159, 169, 213.

austini, 37.
Ambulacra, 56, 57, 68-70, 78-85, 93, 127, 179-181 ;
of Comatuhe, 83-85; of Metacrinus, 81, 82;
of Pentacrinus, 78-80 ; suhtegminal ambulacra
of Palseocrinoids, 164, 179.
Ambulacral epithelium, 59, 93, 112, 113, 415.
Ambulacral groove ; see Ambulacra and Food-groove.
Ambulacral nerve, 59, 70, 93, 96, 111-115, 118, 120,
408-416 ; of Antedon fsdiriddi, 112, 118, 411 ;
of the Blastoidea, 413.
Ambulacral plates, 55, 63, 65, 74-85, 165, 174-181,

184.
Aniphiura, 395, 396.
Anal appendage, 41-46, 371, 372.
Anal jilate, 38, 73.
Anal series, 45, 372.
Anal tube, 69, 70, 81, 89.

Anambulacral plates, 57, 62, 76-78, 80-86,'J65, 180,
184.

432

THE VOYAGE OF H.M.S. CHALLENGER.

Anchylosis, 3, 37.
Ancyrocnnus, 19.

Ant&Jon, 33, 36, 57, 68, 74, 94, 127, 129, 143, 144,
154, 237, 249, 268, 291, 376-38.5, 402-
404 ; ambulacra of, 83-85 ; digestive tube
of, 88, 89 ; disk of, 68, 84, 85 ; geographi-
cal and bathymetrical distribution of, 136-
141; labial plexus of, 97-100, 404-407;
plexiform gland of, 101, 102, 235. '

ncoda, 57, 83, 84, 93, 109, 110, 113, 128
(PL liv. fig.s. 1-4; PL Iv. fig. 5).

angusticahjx, 57, 83, 84, 93, 109, 110, 113,
128, 135 (PL liv. lig. 5 ; PL Iv. fig. 6).

antarctica, 92, 98.

hadcurva, 84 (PL liv. fig. 9 ; PL Iv. fig. 7).

campichei, 144.

carinatu, 98, 109, 124, 130, 137, 235 (PL Ix.
fig. 2;.

ilentata, 36, 130, 137.

eschricliti, 10, 52, 5.5, 62, 67, 78, 83, 88, 90,
92, 93, 9.5-102, 109, HI, 112, 114, 118,
120, 122-124, 133, 137, 144, 153, 208-
210, 319; ambiUacral nerve of, 112, 118,
411 ; parambulacral network of, 123, 124,
416 ; radial blood-vessels of, 96 ; spongy
organ of, 98, 99 (PL lix. figs. G, 7 ; PL Ix.
figs. 3-6).

jiuetuans, 280.

hageni, 36.

inuequalk, 83 (PL liv. Qg. 8).

imerta, 57, 83 (PL liv. figs. 6, 7).

lusitaniea, 315.

miller i, 73.

multiradiata, 84, 85 (PL Iv. figs. 3, 4).

phalani/ium, 10, 36, 130, 318.

proteda, 58.

qiiadrata, 98, 99.

rhodankus, 144.

rosacea, 49, 50, 52, 58, 67, 70, 86, 88, 93,
95-98, 100-102, 104, 107-112, 118, 120,
122, 124, 12.5, 127, 129, 130, 133, 134,
280, 304, 404, 411, 413 ; anal jjlate of, 73;
calyx interradials of, 39 ; genital glands of,
109 ; labial plexus of, 98 ; plexiform gland
of, 102, 104; spongy organ of, 100 (PL Ivi.
fig. 6 ; PL lix. fig. 5).

serohicidata, 336.

spinifera, 277.
Antedouin, 129.
Anthodiata, 186, 190, 191.

Apical dome plates, 157, 167, 168; of ActinocrinidiL',
164, 167; of the Blastoidea, 164, 173; of
Cyathocrinida?, 164, 172; of Ghjptocrinus, 184;
of Ichthyocrinidaj, 181, 183; oi Perieclioerinus,
171, 172; of Platycrinidffi, 164, 17.5-178; of
Ehodocrinida}, 164; oi Strotocriiivs, 171, 172.
Apical sy.stem, 2, 157, 172, 393-402.
Apiocrinida3, 32, 48, 54, 142, 183, 222, 225, 247,

286.
Apiocrinites, 245.

Apiocriuus, 8, 11, 2.5, 42, 49, 53, 68, 76, 125, 126,
135, 14.5-147, 150, 15.3, 154, 181, 183,
211, 223, 230, 249, 256, 270, 276, 289,
293, 340.
constrictus, 24, 257.
cornutus, 258.
crassus, 256.
insiijnis, 9.
marinifieus, 256.
martini, 39.
milleri, 215.
murchisonianus, 256.
par7cinso7ii, 9, 131, 150.
roissyanus, 39, 149, 150, 183, 215.
Arm-groove, 64, 65, 77-83.

Arms, 47-56, 154, 194; oi Adinometra, 55-51 ,(jO,&^,
70 ; of Bathycrinus, 232, 233, 236 ; of Crinoids
and Cystids compared, 189 ; oi Extracrinus, 59-
61, 224, 277, 278 ; of Holopus, 207-209 ; of
Uijocrinus, 52, 53, 219, 224; of Pentacrinidit,
55, 56, 154, 15.5,277, 278,284-287; oi Rliizo-
critius, 52-54, 64, 65, 267.
Article basal of Apiocrinus and Bourgtieticrinus, 3,

25, 15.3, 257.
Articulata, 145, 182, 195, 196.

Articulation, bifascial, 8, 53, 231, 280, 304, 326,
329,330; muscular, 9-11, 146, 194; trifascial,
8, 9, 53, 231-233.
Asterencrinidea, 187.
Asterias, 245, 395, 396, 399, 400.
Astrocrinidae, 194.
Astrocrinus, 132.

Atdecrimis, 34, 36, 56, 58, 68, 127, 137, 138, 140,
141, 144, 291, 292, 294, 371.
halanoides, 376, 380.
mhensis, 379.
Axial cords of rays and arms, 31, 114-127, 146, 407-
416; of Bathycrinus, 31, 119, 236; of Uolo-
pus, 119, 209, 236; of Pentacrinus, 124, 126,
293, 294; of Rlizocrinus, 119, 126, 252, 253.

REPORT ON THE CRINOIDEA.

433

Axillary, 9, 10, 33, -18, 58-61, 90, 125; of Eudesi-

erinus, 215 ; of Holopm, 201-206.
Avicula, 134, 369.
Balanocrimiii, 143, 270-272, 287.
Barycrinus, 54, 61, 363.

hercule'.is, 61, 224.
Basals, 2, 3, 33-39, 71, 73, 106, 124-126, 157, 158,
168-172, 363, 393-399, 402, 403 ; of Batki/-
cn'nus, 37, 226-228; of Coniatulee, 34-36,
251, 284, 402, 403; of Pt-nfacnnus, 34, 35,
282-284, 293, 336; of BMzocrinus, 36,
248-253; of Thaumatocrinm, 34, 36, 371.
Basaltiformes, 296.

Bathycrinus, 38, 48, 54, 56, 58, 62, 91, 104, 117, 127,
143, 213, 214, 222, 223, 225, 226-238,
243, 248, 252, 254-256, 289-292,
386, 388, 391, 392; arms of, 232,
233, 236; axial cords of, 31, 119,
236; basals of, 37, 226-228; chambered
organ of, 34, 107, 235 ; covering plates
of, 74, 234; digestive tube of, 33, 89, 90,
236 ; di.sk of, 234, 235 ; genital glands
of, 109; geograpbical and bathymetrical
range of, 136-140, 391, 392; nerves in
stem of, 31, 119; pinnules of, 233,
234; plexiform gland of, 89, 90, 101,
235 ; primary interradial cords of, 126,
228-230; radials of, 3, 37, 230, 231 ;
stem of, 23, 24, 26-28, 226 ; trif\iscial
articulations of, 8, 9, 53, 231-233;
stem of, 23, 24, 26-28, 226; vascular
axis ill stem of, 107, 235 ; visceral mass
of, 33, 90, 234; water-tubes of, 93, 236.
aldrichianus, 8, 10, 31, 37, 71, 119, 120,
130, 134, 22.5-235, 237-240, 241,

242, 24.3, 244, 376, 386, 389, 391
(PI. vii.; PL viia. figs. 1-21 ; PI. viib. ;
PI. viiia. figs. 4, 5).

rampheUinnm, 227, 232, 233, 235, 237,
238, 239, 240, 242-, 376, 386, 389,
391 (PI. viia. figs, 22, 23; PI. viii.).

carpenteri, 8, 71, 226, 227, 231-234,
237, 244, 386, 389, 391.

gmclUe, 71, 227, 231, 232, 235, 237, 238,

243, 244, 245, 374, 386, 389-391
(PI. viiia. figs. 1-3).

Patbymetrical range of Crinoids, 137, 138.
Bdemnocrinus, 38, 151, 217, 258, 259.

flori/er, 259, 270, 363.

jMurtalesi, 259.

(ZOOL. CUALL. EXP. — P.iET XXXII. 1884.)

Blastactinota, 190.

Blastoidea, 54, 68, 75, 109, 132, 148, 149, 154, 164,

173, 175, 186, 188-195, 41.3, 414.
Blood-vascular ring, 1, 96, 105, 106, 407.
Blood-vascular system, 96-109, 405-407.
Blood-vessel, intervisceral, 87, 97-105, 405, 406 ;

radial, 1, 59, 70, 92, 93, 96, 97, 100.
Botryocrinus, 61.
Bourgueticrinid;B, 142, 225, 269, 290, 386; calyx of,

33 ; stem of, 6, 7, 23-31, 131.
Bourgueticrinus, 24, 142, 143, 154, 222, 225, 237,
245-249, 254-258, 270, 276, 323,
392.
cequalis, 256.
alubamensis, 143, 257.
didymus, 258.
elUpticus, 24, 256.
hotessieri, 246, 259.
londinenais, 257, 2G9.
ooliticus, 257.
suessi, 248, 257.
thorenti, 247, 258.
Brachiata, 186, 187, 191.
; Briariden, 276, 278, 296.
Brisinga, 398-400.
Cainocrinus, 35, 272, 273, 281-283.
CuIHcrimis, 157, 400.

Calyx, 33-46, 124-126, 393, 394; oi Bathycrinus, 37,
■ 226-231 ; of Cofylecriims, 213, 214; of Cyathi-
dium, 211, 212; of Eudesicrinus, 215; of
Extracrinus, 274—276 ; of Holopus, 33, 36,
199-204, 213 ; of Hyocrinus, 33, 36, 218, 223;
of Pentacrinidae, 33, 291-294; of Rlit~.(icrinus,
3, 29, 33, 34, 36, 38, 248-253, 257.
Canaliculata, 196.

Carpocrinus ornatus, vault of, 180, 181.
Caulaster, 195, 399, 401, 402.
Cenocrimis, 273, 278-281, 334.
capid-Medusce, 301.
Central capsule, 410-414, 416.

Centro-dorsal of Apiocrinidffi, 25 ; of Comatulae, 25,
106, 107, 132, 292 ; of Cotylecrinm, 213-215.
Ceriocrlnus, 46.

Chambered organ, 15, 23, 33, 104-108, 114, 119, 194,
404-406; of B,dhycrinus, 34, 107, 235; of
Comatulffl, 106, 107, 292, 404, 412; of Peuta-
erinus, 105, 292, 412.
Cheloerinus, 295.

acutanguliis, 294.
Chladocrinus, 271, 272.

Ii55

434

TliH VOYAGE OF H.M.S. CHALLENGER.

Circular commissure, 40, 117, 413; of Batlujcriiius,
126, 228-230; of Comatula?, 125; of Penta-
crinits, 125; oi Rhizocrimis, 126, 253.

Circuinvisceral coelom, 67.

Cirri, 7, 107, 108, 114, 132, 414 ; of Bnurgueticrinida?,
27, 28; of ComatuLi?, 132, 414; of Pentacri-
nidffi, 12, 288, 342.

Cirrus-vessel, 15, 23, 107, 292, 412.

Clijpeaster, 394.

Gor.cocrlmui, vault of, 160-163, 170, 174, 175, 178, Ic^a.

Comaster, 151, 403.

Oomatula, 3, 4, 6, 8-10, 18, 19, 22, 24, 25, 33-36, 38,
40, 44, 45, 48-53, 55, 56, 58, 59, 62,
63, 68, 70-74, 76, 78, 83-85, 88, 90, 91,
93-95, 97, 106, 107, 110, 117, 119,
124, 125, 130-135, 144-148, 154-156,
164-166, 169, 175, 177, 181,183-185,
192, 230, 249, 252, 254, 255, 261, 273,
277, 279, 285-287, 289-294, 316, 319,
324, 326, 337, 340, 363, 370-372, 393,
401, 412, 413, 415.
Ambulacra of, 68, 156; ambulacral plates
of, 62, 74, 83, 84, 175, 181 ; arms of, 52,
55, 147, 337; axial cords of, 119, 121-
123, 408, 409; basals of, 34-36, 251,
284, 402, 403 ; ceutro-dorsal of, 25, 106,
107, 132, 292; chambered organ of, 106,
107, 292, 404, 412; circular commissure
of, 125 ; cirri of, 132, 414 ; disk of, 68-
70, 84, 85, 90 ; geographical and bathy-
metrical distribution of, 136-141 ; mon-
strosities of, 70; oral plates of, 71, 72,
156; pinnules of, 58, 59, 83; radials of,
147, 289, 293, 294 ; reparation of, 60,
255 ; syzygies of, 4, 53.

ComatulidEe, 142, 192, 370.

Conocrinus, 245, 247, 248.
cornutus, 258.
pyriformis, 248, 257.
seguenzai, 257.
suessi, 257.

Consolidating plates of Cupressocrinus, 162.

Convoluted organ of Actinocriuidae, 87, 165.

Costal plates, 396-399.

Costata, 190, 195.

CotylecHnus, 68, 143, 161; systematic position, 213-
217.
downs, 213.
miliarig, 214.

Cotijlederma, 213.

Covering plates, 62-66, 74-83, 115, 16G, 173-175,

181, 185, 220.
Crinactinota, 190, 191.
Crinoidea, 186, 187, 188-195.
Crotulocrmus, 54, 65, 66, 157.

pinlchci; 65.
C'ulicocrinus, summit plates of, 170, 175.
Cupressocrinus, 48, 150, 152, 153, 162.
Cupule of Cofi/lecri)ui.y; 213, 214.
Cyathidium, 143, 202; systematic position, 211-216.

spileccense, 212.
Cyathidiocrinidse, 211.

Cyathocrinid*, 153, 155, 160, 162, 164, 173, 181.
Cyathor.rlnm, 44, 45, 48, 60-66, 146, 151, 154, 161-
164, 169, 173-175, 178, 18-5, 224,
394-398.
alutaceiis, summit of, 173.
iowensis, ambulacra of, 65.
Icecis, summit of, 173.
longimanus, ambulacra of, 65, 66, 84.
malvaceus, summit of, 172.
ramosus, ambulacra of, 66.
Gydohlastus, 191.
Cydocystoides, 192.
CystidoB, 192.

Cystidea, 54, 109, 132, 148, 149, 154, 186, 188-19.').
CystoUastus, 191.
Cystoidea, 186, 195.
Deltoid pieces of the Blastoidea, 162.
' Democrinus, 28-30, 245, 264, 268, 269.
parfaiti, 262, 264, 269.
Dendrocrinus casei, 45.

navigiolum, 20.
Dichocrinus, 39, 222, 223.

intermedins, 223.
Digestive tube, 1, 84, 86-92.

Disk, 1, 67, 75, 80, 81, 84, 85, 88, 90, 182, 184, 342 ;
of Adinometra, 69, 70, 85 ; of Antedon, 68, 84,
85 ; of Bathycrirms, 234, 235 ; of Hyocrinus,
219; of Metacrinus, 59, 68, 80, 81, 93, 342;
of Pentacrinus, 68, 76, 77, 91, 279.
Distichals, 49, 50, 177.
Dorsocentral, 18, 132, 157, 158, 168, 169, 274, 275,

393-396, 398-400.
Ediinosphoerites, 192, 195.
Echinozoa, 95, 105, 106, 193, 194, 406.
Edrioaster, 192.

j Edriocrinits, 143, 194, 195, 217, 319.
j Eleacrinus, 164, 173.
j Eleutherocrinus, 132.

REPORT ON THE CRINOIDEA.

4:35

EmeJuUata, 151, 196.
Encrinida;, 142, 187.
Eiicrinites duhius, 294.

Encnnus, 25, 34, 35, 48, 49, 51, 117, 125, 126, 142,
143, 145, 149, 150, 152-155, 230, 272,
275, 284, 294-296, 300, 340, 363, 394,
395.
beyrichi, 270, 295.
caput-Mediism, 274, 300.
(jracilis, 37, 153.
liliifurmis, 294.
pentadinus, 295.
Epizygal, 4, 50, 219.
Erisocrinus, 150, 152-154.
Eucalyptocrinus, 151, 157, 160, 195, 400.

crassas, root of, 20.
Eucrinoidea, 186-189.
Eadeskrinus, 10, 143, 149, 161, 202, 205; systematic

position of, 213-216.
Endiomnus, 36, 47, 58, 68, 137, 138, 140, 143, 291,
373.
o.tlanticus, 47, 127.
indivisus, 47, 48, 58, 127, 137.
japonicus, 47, 109, 127, 339, 378.
srmj)eri, 36, 47, 127.
varians, 47, 58, 127.
Eugeniacrinidae, 142, 216.
Eiujeniacrinite.% 245:

Ewjeniacrinus, 68, 131, 142, 161, 211, 214, 227, 247,
248.
mayalis, 215.
pyrifoi~niis, 247.
Eiipachycrinus, 46, 153, 154.

Extmcrinux, 8, 30, 34, 41, 43, 51, 59-61, 76, 131, 143,
145, 149, 150, 154, 181-183,271, 272,
274-278, 281, 288, 296-299, 308, 371,
394, 395; arms of, 59-61, 224, 277,
278 ; caly.x of, 274-276 ; geological
range of, 296-298 ; stem of, 276, 277,
287.
hriareus, 20, 22, 54, 55, 81, 131, 274,

275, 278, 296, 297, 313.
siibangularid, 16, 22, 30, 54, 274, 277, 413.
Food of Crinoids, 132, 133.
Food-grooves, 56, 64, 68, 69, 77-85, 92, 156, 163-

166, 173, 180, 236, 415.
Free rays of Iclithyocrinidse, 181 ; of Platycrinus, 177.
Genital canal, 96, 209.

Genital cord, 61, 63, 108, 110, 111, 209, 210.
Genital glands, 56, 57, 61, 63, 83, 108-111, 220.

Genital plates of Echini, 2, 168, 169, 172, 393-398, 403.
Genital vessels, 97-100, 105, 108, 109.
Geographical range of Crinoiils, 136, 137.
Gissocrinus, 66, 174.

punctuostig, ambulacra of, 65.
Glyptocrinus, 183-185.

decadad!/hi..s, 183.
sdiaffuri, 20.
Glyptocystites, 20.
Goniaster, 245.

Growing point of arms, 56, 59, 60, 400.
Guettardicrimis, 39, 146, 149-152, 181, 289.
Gymnocrinui<, 143, 216.
Habits of Crinoids, 130-132.
Habrocrinus, 66, 180.
Haplocrinus, 54, 157-159, 163, 164, 167, 170, 171.

mespiliformis, 158.
Heterocrinus, 42, 45, 53, 154, 267, 363.
const ridns, 53, 224.
simjilex, 53, 224.
HexacrimM, 39, 176, 217.
Holopidaj, 197, 216, 386.
Holopocrinids, 211.
Holopodidffi, 142, 143, 195, 217.
Holopm, 38, 48, 127, 131, 141, 143, 14.5, 149, 155,
159-161, 172, 18.5, 197, 198-217, 386,
388.
Arm-joints of, 207, 209 ; axial cords of, 119,
209, 236 ; bivium and trivium of, 202,
207; calyx of, 33, 36, 199-204, 213;
colouring matter of, 129, 204, 210 ; genital
glands of, 63, 110, 210; geographical and
bathymetrical distribution of, 133-138,
140, 141 ; oral plates of, 70, 72, 95, 156,
198, 208; pinnules of, 207, 208; second
and axillary radials of, 204-206 ; system-
atic position of, 211-217 ; tentacles of, 208.
ranrji, 199-208, 380, 381, 384, 385, 386,
388, 389 (Pis. i.-vb.; PL vc. tigs. 1-3).
tlyhocystites, 191.

Hydrospires of Blastoidea and Cystidea, 194.
Hyocrinidae, 31, 217, 223, 386.

Ilyocrinus, 48, 54, 60-62, 73, 83, 134, 142, 146, 149,
15.5, 160, 161, 176, 179, 181, 185, 208,
217, 218, 219, 221, 222, 234, 236, 242,
267, 289, 371, 386, 388, 391, 392; arms
of, 52, 53, 219, 224; calyx of, 33, 36, 218,
223 ; covering plates of, 74, 220 ; genital
glands of, 109, 220 ; geograpliical and
bathymetrical distribution of, 137-140;

436

THE VOYAGE OF H.M.S. CHALLENGER.

Hyoerinus continuod —

oral plates of, 70, 72, 94, 95, 156, 219;

pinnules of, 59-61, 219, 224; side plates

of, 82, 220; stem of, 5, 31, 32, 218, 221;

systematic position of, 222-224; testes of,

220; water-pores of, 94, 95, 219, 220,
hethcllimws, 218-222, 240, 376, 378, 386,

389, 391 (PL vc. figs. 4-18; PI. vi.).
Hyponome sarsi, 67, 85, 166.
Hypozygal, 4, 50, 54.

Ichthyocriiiidre, 42, 76, 149, 155, 166, 181-185, 195.
Ichthyocrinus, 185, 195.
Hy aster, 401.
Ilycrimis, 225.
Inarticulata, 145.
Infra-nodal joint, 13.
Interartioular pores, 18.
Internal casts of Actinocrinichv, 165, 174. 179, 182 ; of

Plat ycrinus, 179.
Interpalmar areas of disk, 69, 81, 84, 165, 166, 174, 180.
luterradials of caly.x, 38-46, 74, 162, 178, 371.
Interradials of perisome, 42-44, 73, 74, 76, 18.3, 371.
Intervi-sceral blood-vessels, 87, 97-105, 405, 406.
Intervisceral coelom, 67, 109.
Iiitervisceral ple.xus, 86, 87.

Intra-radial commissure of Comatula and Pentacrinus,
125, 126; of Bathycrinus, 127, 229 ; of PMzo-
crinus, 126, 253.
loerinus, 46, 363.
Ids, 272, 300, 302.
Isocrimis, 271-273, 281.

2)endHlus, 34, 284.
Labial plexus, 93, 97-101, 103-105, 404-406.
Lccanocrinus rcemcri, 217.
Lo/pidodisms, 85.
Lichenocrinoidea, 192. .
Liclienocrinus, 192.

diibiiis, 20.
Ligaments, of arms, 7-9, 414 ; basiradial ligaments of

Bathycrinus, 228 ; of stem, 5-7, 13, 23.
Madreporic canal of Ecbinozoa, 106, 194, 406.
Mariacriiius, 155, 171.
Marsupiocrinus, vault of, 171, 175-177
depressus, 176.
radiatus, 176.
Marsupites, 34, 48, 68, 132, 142, 145, 151, 155, 181,

183, 196, 216, 394-398.
Marsupitidae, 142.
Mi'gistocriniis, 54.
Mesoerinus, 24, 237, 255-257, 323.

Metucrinus, 10, 13, 110, 124, 128, 134, 148, 152, 154,
205, 270-272, 275, 283, 285-288, 291,
300, 304, 311, 320, 334, 336, 339,

340, 341-344, 357, 359, 360, 362, 369,
370, 387, 388-391 ; arms of, 55, 285,
286, 342 ; cirri of, 288, 342 ; disk of,
59, 68, 80, 81, 93, 342; geographical
and bathymetrical distribution of, 136-
142, 343 ; pinnules of, 10, 55, 58, 155,

341, 342; pinnule-ambulacra of, 81,
82; radials of, 48-51, 341, 357; stem
of, 14, 15, 17, 19, 287, 288, 342, 343.

ahf/idatiis, 19, 51, 60, 70, 80, 81, 110,
134, 287, 288, 293, 342, 344, 345,
346, 348, 355, 357, 364, 366, 369,
370, 377, 387, 389 (PI. xii. figs. 1-14;
Pis. xxxviii., xxxix.).

ciiujidatus, 17, 80, 81, 287, 288, 342,
344, 346, 347, 348, 35.5, 357, 358,

369, 377, 387, 389 (PI. xl. ; PI. xli.
figs. 1-4).

costatus, 17, 35, 82, 129, 284, 288, 342,
344, 359, 360-362, 363-366, 368-

370, 378, 387, 389 (PI. xlvii. fig. 13;
PL xlix.).

iletacrinus dredged by tlie " Vega," 340>

343, 344, 368, 387, 388.
interruptus, 134, 288, 342-344, 367,

368, 369, 377, 387, 388 (PL liL).
moseleij!, 17, 129, 341, 344, 355, 356,

357-359, 37.8, 387, 389 (Pis. xlv.,

xlvi.).
murrayi, 17, 78, 81, 82, 128, 140, 184,

326, 342-344, 349, 350, 351, 352,

355, 377, 378, 387, 389 (PL xli.

figs. 12-17; PL xliL).
nohilis, 17, 80-82, 287, 344, 350, 351,

352, 353, 355, 369, 377, 387, 389

(PL -xlL figs. 5-11; PI. xliiL).
nodoms, 51, 76, 80-82, 289-291, 342,

344, 359, 362, 364, 365, 366, 368-
370, 377, 387, 389, 391 (Pis. L, IL).

rotundus, 343,- 344, 357, 387, 388.

stewarti, 288, 344, 387.

superbus, 344, 351, 352, 355, 387.

tuberosus, 134, 288, 342, 344, 369, 370,
377, 387, 389 (PL liii. figs. 1-6).

varians, 17, 129, 134, 288, 342, 344,
351, 352, 353, 354, 355, 366, 378, 387,
389 (PL xliv. ; PL xlviL figs. 6-12).

REPORT OX THE CRINOIDEA.

43:

Mdacrinus mjvilUi, 129, 130, 2SS, 342-344, 354,
357, 358, 359, 360, 362-365, 308,
377, 378, 387, 389, 391 (PI. xlvii.
figs. 1-5; PL xlviii.).

Micropocrinus, 143, 216.

MiUericrimis, 8, 11, 25,49, 54, 131, 135, 153, 256,
270,271,281,-282,340.

nodofkinng, 250.

jn-aitf, 20, 26, 51, 132, 328.

simjihr, 250.

Monstrosities, 70, 347.

Mouth, 56, 68-70, 91, 103, 156, 100, 403.

Moutli-sliields of Ophiurids, 2, 169.

Musdes, 9-11, 11.3-115, 414, 41.'"..

Myelodactyloidea, 192, 193.

Myelodacfyhis, 192, 193.

MyrtiUocrinus, 146.

Mymstoma, 70, 133-135, 334, 358, 304, 369.
dcformator, 324.
pcntacrini, 324, 304.
iryville-thomsoni, 347, 302, 304.

Nuocriiioidea, 9, 38, 49,51, 59, 142, 145, 147-157,
160-162, 170-173, 181, 185, 194, 195, 196,
217, 224, 270, 295, 309, 411.

Neocrinus, 273, 279, 280, 302, 334.

Nerve-fibres of stem, 23, 31, 116, 119, 120.

Nervous syst<'tn, 111-127, 407-416.

Neuro-vascular axis of stem, 194, 413.

Xodal joints in stem of Pentacrinida?, 4, 12-15, 17,
19, 10.5, 107, 291.

Ocular plates of Echini, 2, 168, 109, 393, 394, 402.

Odontophore of Starfishes, 2, 398-400.

Ollacrinus, 157.

Onychommis, 41, 42, 151, 181, 182, 372.

Oral pinnules of Antedon, 59, 115.

Oral plates, 2, 70, 73, 94, 95, 156-164, 167-173, 178,
184, 185, 398 ; of Bathycrinus, 234; of Coma-
Ma, 71, 72, 156; of Uolopus, 70, 72, 95, 156,
198, 208 ; of Hyocrinus, 70, 72, 94, 95, 156,
219; of Rhizocrimis, 70, 72, 156, 219, 255;
of Thaumatocrimis, 70, 72, 160, 161, 371.

Orocentral, 158, 159, 107-172, 178, 184.

Ovary, 109-111, 210.

Ovoid gland, 104-100, 400, 407.

Palaeocrinoidea, 32, 38, 39, 47-49, 54, 01, 08, 72, 132,
135, 147-158, 164-167, 172, 173, 181, 185,
194, 19.5, 217, 218, 224, 270, 271, 295, 358,
363, 370-372, 394.

Palceostoma mirahiUs, 2.

Parambulacral network, 123, 124, 415, 416.

Parasites of Crinoid5=, 133-lS.").
Pahna animal, 301.
Palmars, 49, 177.
Palmier marin, 300.

Pclmatozoa, 186, 187, 188, 190, 191, 19.3-195, 413.
Pcntacrinidse, 50, 71, 83, 110, 119, 125, 132, 142,
143, 140, 150, 154, 156, 100, 183-
18.5, 192, 228, 270, 271, 277, 289-
291, 294, 295, 310, 317, 320, 322,
323, 326, 337, 386.
Ambulacral tkeleton of, 02, 74-82 ;
arms of, 55, 56, 154, 155, 277, 278,
284-287 ; calyx of, 33, 291-294 ;
characters of young, 289-291, 327 ;
cirri of, 12, 288, 342; colouring
matter of, 128, 129, 210 ; colonics of,
130, 131 ; disk interradials of, 73,
70, 183; ovaries of, 110; parasites
of, 134, 135 ; pinnules of, 55, 58, 59,
155, 286; stem of, 3-6, 12-23, 287,
288, 290, 291, 31.5-319, 342, 343 ;
syzygies of, 4, 5, 254, 326.
Pentacrinin, 128, 129 ; 204, 210.
Pentanrinites, 272, 273.

vidyaris, 274, 300.
Pentacrinoid larva of Comatida, 24, 39, 70, 89, 93, 95,
101, 107, 108, 116, 127, 158, 159, 161, 101,
194, 214, 219, 237, 251, 290, 396.
Pmiacrums, 3, 7-10, 13, 17-19, 49, 50, 53, 58, 62,
66, 74, 76-81, 84-88,90, 97, 101, 103,
104, 110, 117, 126, 127, 128-132, 1,34-
141, 143, 145, 148, 155, 165-167, 173,
181, 182, 192, 195, 210, 211, 228-231,
235, 244, 252, 256, 259, 261, 271, 272,
273, 274-300, 302-304, 316-320, 326,
340-343, 364, 309, 386, 388, 391, 392,
394, 409, 415.
Basals of, 34, 35, 282-284, 293, 336;
chambered organ of, 105, 292 ; disk of,
68, 70, 77, 91, 279 ; geographical and
bathymetrical distribution of, 136-141,
300, 343, 391, 392; geological history
of, 294-300 ; intervisceral blood-vessek
of, 101, 102 ; labial plexus of, 93, 100 ;
plexiform gland of, 105; primary inter-
radial cords of, 124-126, 292, 293;
radials of, 289, 293, 294 ; radial plug
of, 34, 293; stem of, 12-23, 287, 288,
31.5-319,342, 343; vascular axis in .stem
of, 23, 107 ; water-vascular ring of, 93.

438

THE VOYAGE OF H.xM.S. CHALLENGER.

Pentacrinus al/eniicirrus, 12, 19, 21, 44, 55, 77,

78, 80, 128, 135, 13G, 177, 184, 277,
282, 284-286, 288, 293, 299, 305,
310, 313, 315, 318, 320, 321, 322,
323, 324, 326, 335, 337, 3tfi, 353,
364, 377, 378, 386, 359, 391, 412
(Pis. XXV., xxvi. ; PI. xxvii. figs.
1-10).

ade)-isciis, 143, 297, 298.

astei-im, 4, 14, 17, 19, 42, 58, 64, 73,
74, 79, 154, 183, 184, 274, 275, 277,
279, 280, 282, 283, 285-288, 293,
299, 300, 301, 302, 303-305, 308-
312, 318, 329, 332-335, 342, 343,
371, 381, 384, 386, 388,3 89 (PI. xi.;
PI. xii. iigs. 1.5-25; PI. xiii. ; Pi. xvii.
figs. 7, 8).

hasaltiformis, 300.

leaugrandi, 286.

blakei, 5, 14, 35, 50, 78, 280, 283,
284, 287, 288, 299, 319, 326, 328,
329, 330, 334, 342, 381-383, 386,
389 (Pis. xxxL, xxxii. ; PL xxxiii.
figs. 1-3).

hnareu.1, 274, 278, 280, 297, 302.

hriareus achalmianus, 297.

hriareus minuhis, 278.

Irriareus zoHerianus, 297.

hronnii, 323.

luclisijauensis, 297.

cupiit-Medime, 274, 278, 281, 300-302,
304, 306, 308, 309.

darrjniesi, 297.

deconis, 14, 18, 21, 3.5, .50, 51, 60, 77.

79, 91, 93, 95, 96, 100-102, 104, 105,
120, 12.3, 231, 279, 280, 283-285, 288,
290, 291, 293, 294, 299, 303-305, 308-
312, 316, 319, 320, 324, 326,329, 330,
331, 332, 333-337, 341, 342, 379-
386, 388, 389, 402, 416 (PI. xxxiii.
figs. 4-6; Pis. xxxiv.-xxxvii.; PI. Ivii.
figs. 2-5; PI. Iviii. tigs. 1-3; PI. lix.
figs. 1-4; PI. Ixii.).

didadylus, 323, 327.
d;xu7ii, 143, 284.
duhius, 294, 295.
fislieri, 34, 143, 284.
jaccardi, 17.
johnsoni, 274.
jurenm, 4.

Pentacrinus madearamis, 15, 19, 20, 35, 55, 62, 80,
277, 281, 282,284, 285,288, 299, 310,
312, 313, 315, 318, 323, 34.3, 346,
353, 376, 387, 389, 412 (PL xvL;
PL xvii. fig. 1).
milleri, 274.
mollis, 44, 300, 338, 339, 343, 378,

387, 389, 391 (PL xxxiii. tigs. 7-10).
miilleri, 14, 51, 55, 61, 77, 80, 277, 280,
285, 288, 299, 304, 305, 306, 307,
308-313, 315, 316, 318, 320, 323, 324,
330, 331-334, 337, 340-342, 380-385,
387-389 (Pis. xiv., xv.; PL xvii. tigs. 9,
10).
naresianus, 5, 14, 18, 21, 44, 50, 52, 53,
77, 78, 94, 128, 136, 140, 143, 184,
231, 240, 27.5, 280, 285-288, 290, 291,
293, 299, 300, 320, 322, 324, 325,
326-330, 334, 336, 337, 342, 354, 377,
378, 387, 389, 391 (PL xxviu figs. 11-
13 ; Pis. xxviii.-xxx.).
nicoleti, 17.
nodotianus, 297.

pentagonalis personatiLS, 34, 284.
psilonoii, 300.
scalaris, 4, 35, 284.
sigmaringensis, 300.
suhangularls, 274, 278.
sulteres, 270.
tridadijlus, 323.
tuberculatu-s, 35, 274, 286.
vulgaris, 294.

wijville-thomsoni, 14-17, 19, 21, 35, 44,
55, 76-78, 80, 91, 93, 98, 100, 119,
124, 129, 130, 136, 176, 18.5, 277,
280-284, 286-288, 290, 291,293, 294,
299, 305, 310, 313, 314, 315-320,
322, 323, 33.5, 337, 342, 343, 346, 37.5,
387, 389-391, 415 (PL xviL figs. 2-6;
Pis. xviii.-xxiv.; PL Ivii. fig. 1).
Pi'idrcmitcs, 192, 195, 413, 414.
Pentremitidse, 192.
Periechoerinus, summit of, 171, 172.
Perisomatic skeleton, 1, 2, 73-85, 182.
Perisome, 67, 68, 97.
Perisomic plates, 39, 40, 41, 124.
Peristome, 68, 71, 84, 87, 164, 179, 181.
Peritoneum, 67, 99, 100.
Philocrinus, 152-154.
Phimocrinus, 1 50.

KEPORT ON THE CRINOIDEA.

439

Phyllocrinus, 142, 21G, 227, 11;?.

Physetocrinus, summit of, 171.

Pigment, 23, 97.

Pideticrinm, 273, 281.

PinnasteUa, 187.

Pinnigrada, 187, 190.

Pinnules, 55-66, 174; of Bathycrinm, 233, 234; of
Comatulte, 58, 59, 83 ; of Holopan, 207, 208 ;
oi Hyoerinus, 59-61, 219, 224; of Mcfacrinm,
10, 55, 58, 155, 341, 342; of Pentacrinus, 55,
58, 59, 155, 286; of Ithizocrivus, 56, GO, 255.

Pisocrinus, 48, 54.

Platycrinidee, 61,62, 149, 158, 160, 1G2, 1G7, 168,
170, 171, 175, 176, 178-181.

Platycrimis, 24, 48, 62, 65, 14G, 151, 15.5, 163, 164,
174, 176-179, 185, 196, 217, 22.3, 237,
258, 394.
hurlimjtonemtis, 177.
ventricosus, 170.

Pleurocystites, 20.

Plexiform gland, 33, 34, 96-106, 108 ; of Aduwmetra,
103, 104; of Antedon, 101, 102, 235; of
Bathycrinus, 89, 90, 101, 235; of Echinozoa,
101, 105, 407 ; of Pentacrinus, 105; of Ilhizo-
crinns, 89, 101, 104.

Plicatocrinidas, 142, 222.

Plicatocrinus, 49, 51, 142, 152, 155, 20.J, 214, 222,
223, 289.

Poteriocrinus, 46, 60, 61, 146, 224, 394.
radiatus, 152.

Primary interradial cords, of Bcdhycrinus, 126, 228-
230; of Encrinvs, 125, 126; of Pentacrinus,
124-126, 228, 230, 292, 293; of Rhizocrinus,
126, 228-230, 252, 253.

Proboscis, 41-44, 87.

Fromachocrinus, 36-38, Q&, 90, 92, 97, 137, 138, 140,
144, 216, 403.
ahyssorurn, 376, 377.
kerguelensis, 28, 99, 101, 102, 127
naresi, 378.

Proximal plates of vault, 168-172.

Psolidae (orals of), 2.

Pterocrinus, 242, 243.

Pterotoorinus, 157, 177.

Radials (first), 2, 3, 33-38, 48, 106, 152, 155, 157,
168-170, 393-395, 399, 402; of Bathycrinm,
3, 37, 230, 231 ; of Comatulie, 147, 289, 293,
294; of Metacrimis, 48-51, 341, 357; of
Pentacrinus, 289, 293, 294 ; of Promachocrinm,
37, 38 ; of Rhizocrinus, 250-253.

Eadials (second), 38, 47, 50 ; of Eudesicrinus, 10, 215 ; of
Ilolofus, 204-206,215; of Metacrinus, 48, 341.

Radial dome plates, 167, 169, 175-178.

Radial jdug, 34, 293.

Radial skeleton, 1, 2.

Radial spaces, 31, 32.

Radicular cirri, 27, 28.

Ray-divisions, 47-51, 337.

Reparation of arms, 60, 255, 285 ; of disk, 255.

Reteocrinue, 42, 151, 181, 183-185, 371, 372.
nealli, 43, 44, 184.
richardsoni, 43.
stellaris, 43.
mhglohosits, 43.

Rhizocrinus, 3, 24, 27-31, 38, 54, 64, 73, 109, 127,
130, 132, 142, 143, 151, 1.53, 213, 217,
222-226, 229-238, 244, 245, 246,
247-259, 268, 269, 289-292, 317,
371, 388, 391, 392; ambulacral plates
of, 62, 261; arms of, 52-54, 64, 65,
267; caly.x of, 33, 34, 36, 38, 248-253,
257; digestive tube of, 88-90, 254;
geographical and bathymetrical range
of, 136-141, 391, .392; intra-radial
commissure of, 126, 253; orals of,
70, 72, 156, 219, 255; parasites of,
134; pinnules of, 56, 60, 255; plexi-
form gland of, 89, 101, 104; primary
interradial cords of, 156, 228-230, 252,
253 ; reparation of disk in, 255 ; second
brachials of, 38, 254; stem of, 23 31,
107, 256; syzygies of, 5, 9, 254, 255;
visceral mass of, 33, 90, 254, 255.
lofotensis, 3, 25-28, 36, 47, 56, 72, 90,
93, 134, 136, 246, 248-251, 253, 254,
257, 259, 260, 261-266, 268, 374-
376, 378-3S0, 382, 384, 386, 388, 389,
391 (PI. viiia. figs. 6-S; PI. ix. ligs. 1, 2;
PL X. figs. 1, 2).
londinensis, 269.

rawsoni, 25, 28-30, 47, 56, 72, 134, 249,
254, 255, 257, 258, 262, 263-269,
374-376, 379-384, 386, 388, 389,391
(PL ix. fig-s. 3-5; PL x. figs. 3-20;
PI. liii. figs. 7, 8).
calyx of, 265-267; pinnules of, 267;
young specimens of, 264, 265.

Rhodocrinidie, 39, 40, 149, 164, 167, 168, 171. 174,
371.

Rhodocrinites, 39, 40, 371.

440

THE VOYAGE OF H.M.S. CHALLENGER.

Jihodocrinus, 371, 394.

Koot-joint of stem, 28.

Kost'tte of Comutulse, U, 3G, IOC, 2.j1, 291, 402: of
RMzocrinvs, 34, 249, 251, 2.i3.

Saccosoma, 190, 195.

Sacculi, 83, 127.

Sale7iia, 395.

Saiimpliittchcn, GG, 74, 173, 174, 180.

ScalpeUum, 3C9.

album, 134.

halanoides, 134, 351, 355.

Semiarticulata, 145.

Side plates, 62, 66, 74, 79-8.3, 165.

Solanocrinites, 272.

Solanoenmis, 402, 403.

Spongy organ, 98-101, 103, 406.

Stelidiocrinus, 180.

capihdum, surmmt ol, 171.

Stem, 5-7, 12-32, 54, 114, 135, 194, 401 ; of Bathy-
crimis, 23, 24, 26-28, 226; of Uxtracrhu/g, 276,
277, 287 ; of Hijocrinus, 5, 31, 32, 156, 218,
221 ; of Metacrmus, 14, 15, 17, 19, 287, 288,
342, 343; of Pentacrinus, 12-33, 287, 288,
315-319; oi Rhizocrinus, 23-31, 107, 256.

Stnmmatocrinus, 150, 152-154.

Stilasteritse, 187.

Stomatoci'inoidea, 147, 156, 195, 196.

Strotocrinus, vault of, 171, 172.

Stylifer, 134.

Stylina, 134.

Subambulacral plates, 75, 85, 165.

Subangularen, 276, 278, 296.

Support of Eudesicrinus, 214-216.

Supra-nodal joint, 14.

Symbathocrinus, 54, 158, 163, 164, 107, 169-171,
179.

Synostosis, 2, 3.

Syzygial interval, 52.

Syzygy, 3-5, 8, 9, 15, 50, 52, 53, 231-233; of Coma-
tula, 4, 53 ; of PentacrinidK, 4, 5, 254, 326 ; of
Rhizocrinus, 5, 9, 254, 255.

Talarocrinus, vault of, 177.

Taxocrinus, 41, 42, 1.50, 151, 154, 211, 372.

Tentacles, 56, 57, 63, 69, 75, 93, 11.5, 117, 189, 194,
416; oillolopus, 208.

Tentaculata, 186, 189.

Tegmen ealycis. See Vault.

Termination of stem in Pentacrinidie, 18-23, 315-319.
Tessellata, 142, 145-148, 101, 195, 196.

Testes of Hyocrinus, 220.
Tetmcrinus, 216, 227.

Thuuinatocrlnus, 33, 34, 36, 39-42, 44, 45, 47, 58,
68, 70-72, 127, 144, 148-151,
156, 160, 166, 172, 179, 18.5, 223,
291, 370, 371, 372, 392; anal
appendage of, 41, 45, 371-372;
basals of, 34, 36, 371 ; calyx
interradials of, 39-41, 371 ; geo-
graphical and bathymetrical dis-
■ tiibution of, 137-140; oral jjlates
of, 70, 72, 160, 161, 371.
renovatus, 149, 372, 373, 377
(PL Ivi. figs. 1-5).
Thiolliericrinus, 24, 143, 257. •
Thylacocriniis, 371.
Tiareeliinus jmiiops, 41 4.
TorynorriHus, 1 42.
Uintacriuida3, 142.

Uintamnm, 132, 142, 147-151, 196.
Under-basals, 152, 153, 168, 169, 394-399, 402.
LTngrooved arms of Acfiiwmetra, 56, 57, 69, 70, 113,

115.
Vascular axis of stem, 15, 23, 104, 107, 235.
Vault, 41, 42, 156, 157, 166-168, 185; of Actino-
crinidie, 164-167, 173, 180, 181 ; of Blastoidea,
173; of Coccocriiius, 160-163; of Cyatho-
crinidae, 173, 181; of Glyptocvinui<, 183, 184;
of IclitliyocrinidBe, 181, 182; of Platycrinidse,
170, 171, 175-181; of Reteocriiuis, 184; of
Xenocrimtii, 185.
Vegetative system, 54-56.
Ventral radial furrow, 252.
Ventral sac, 44, 46, 76.
Verruca, 349, 353, 355, 369.

nitida, 134.
Visceral mas.s, 1, 33, 67-70, 164, 166 ; of Bathycrinus,
33, 90, 234 ; of Rhizocrinus, 33, 90, 254, 255.
Visceral skeleton, 85-87.
Water-pores, 73, 75, 76, 81, 84, 92-96, lOG, 194, 219,

220, 404-406.
Water-tubes, 92-94, 106, 165, 236, 404.
Water-vascular ring, 1, 92-94, 165, 405.
Water-vascular system, -92-96, 106, 194, 406.
Water-vessels (radial), 1, 70, 92-94.
Woodocrinus, 20.

Xenocrinus, 42, 44, 181, 183-185, 371, 372.
Zoroaster acMeyi, 399.

fuJgens, 106, 395, 399.

EXPLANATION OF THE PLATES.

The following letters are em|iloyeil throughout all the Plates :-

A.
a.
a.
at:

ad.

ae.

ai.

an.

ar.

at.

B.

B„B.

h.

hr.

a

c.

ca.

cc.

ceo.

ch.
cli.

chc.

da I.

Axial cord of the ray or arm.

Axial cord of a pinnule.

Branches of the axial cord in the skeleton.

Connective - tissue fibres between the

auauibulacral plates.
The parambulacral extensions of the axial

conls into the ventral perisome.
Ambulacral epithelium.
Primary interradial cords.
Anambulacral plates.
The secondary (radial) cords.
Anal tube.
Basal ring.

First and second brachials.
Eadial blood-vessel.
Ligaments uniting the basals to the top

stem-joint.
Plug of calcareous tissue in the radial

funnel, called a rosette by Sars.
The body-cavity or coelom.
Tlie bands of connective-tissue which

traverse it.
Fibrillar sheath round the vascular axis

of the stem.
Its radiating extensions.
Coeliac canal.

Interradial portion of the circular com-
missure.
Cavities of tlie chambered organ.
Their downward prolongations iuto the

stem.
. Fibres of connective tissue whicli traverse

the fibrillar envelope of the chambered

organ (central capsule).
. Tlie nodal enlargements of the peripheral

vessels of the stem {ch').

(ZOOI- CHALL. EXP. PART XXXH. — 1884.)

ci.

. Cirrus.

cic. .

Ciliated cups in the coeliac canal.

CJK .

. Covering plates.

CS.

Connective-tissue spaces iu the perisome.

'''■-

. Cirrus-vessel.

cvc. .

. Circumvisceral coelom.

A. ^2.

7^3, First, second, and fifth distichals.

e.

External epithelium.

fy. ■

. Fore-gut.

G. .

. Gut.

y- ■

Its epithelial lining.

gc. .

Genital cord.

go'. .

Genital canal

gv. .

. Genital vessel.

ih. .

. Intervisceral blood-vessel.

ico. .

lutraradial portion of the circular com-

missure.

J. .

. Arm-joint.

:/'■

Pinnule-joint.

L. .

Interradial ligament.

I

Basiradial ligament.

Ih. .

Interbasal ligament.

Id. .

. Dorsal ligament between the arm-jointe.

Id'. .

The fossa in which it is lodged.

li. .

Interarticuhir ligament of the arms.

li'. .

. Tlie fossa in which it is lodged.

In. .

Calcareous network iu the jierisonic.

/p. .

Labial plexus.

Ir. .

. Its radial extensions beneath the ambu-

lacra.

U .

luterarticular ligament of stem.

M. .

. JMouth.

1)1.

. Jluscle.

111.

. Transverse muscle-threads in the watei-

vessels.

nij). .

Muscle-plate.

li .56

442

THE VOYAGE OF H.M.S. CHALLENGER.

)l.

Ambulacra! nerve.

sp. .

in:

Its oral ring.

sic. .

(>. .

Oral plate.

sub. .

vh. .

Oral blood- vascular ring.

sy. .

ov.

Ovary.

T. .

p. ov p.

Pigment granules.

t.

Jll. .

Pinnule.

t.h. .

h\, /.*,, A',,

First, second, and thinl radials.

V. or c. .

;■".

Eeotum.

v.r.f.

nu. .

Muscles between the first and

second

W. or /(•. .

radials.

rep. .

I'm . .

The fossiB in wliicli they are lodg

ed.

im: .

riK .

Eadial plug.

id. .

rs.

Eadial space in the stem.

X. or X.

fi.

Spinelets on the disk.

XV. .

X'(.

Sacculi.

«).

Spongy organ.

Side plate.
Subtentacular canal.
Subambulacral plate.

Syzygy.

Tentacle.
Testis.

Tentacular branch of water-vessel.
Central vascular axis of stem.
Ventral radial furrow.
Eadial water-vessel.
"Water-pore.
Water-vascular ring.
Water-tube.
Plexiform gland.

Its ventral end which joins the labial
plexus.

NOTES.

1. In the following lists there is a reference after each figure to a certain page of the Text. In the case of
figures which illustrate entire specimens, the reference given is usually to the page in the systematic part of the
Eeport on which the species is first described. But with those figures that represent structural details,
reference is given to the page containing the explanation which the figure was designed to illustrate. This is
sometimes in the morphological and sometimes in the systematic part of the Eeport. In a few cases there is
no special reference to a figure in the Text ; and the number of the page following its explanation is either that
of the specific diagnosis ; or that of a page containing a description of structural peculiarities which is more or
less illustrated by the figure in question.

2. The magnification given for those figures which were drawn before the collection of Stalked Crinoids
came into my hands is in most cases only approximate (circa). A record was rarely kept of the amount of
enlargement ; and I have therefore estimated it as closely as possible. All the figures for which I am
responsible, however, were drawn according to a definite scale, which varied according to circumstances, but
was carefully noted in each case.

3. All the figures which were drawn for Sir Wyville Thomson, and also those on Plates LVIL, LIX., and
LX. were drawn on the stone in the natural jiositious of the specimens, so that they appear reversed in the
Plates. This should be remembered in future attempts to identify the individuals here described as types ; and
also in reading the description of the anatomy of the disk in Chapter VI.

£ERATA.

Plate Vila. — The figure of the radial axillary in the lower right-hand corner should he lettered " 17 " instead

o/"7."
Plate Vllb.— In the upper part of fig. 6, "a" should be "a'," as in the lower part. On the left of fig. 4, " T

should be " L."
Plates XL, XII., XIIL, and XYll.^ Instead of " Pentacrinns asferia, L.," read " Peniacriiius asierhis, Linn.,

sp." (see p. 303).
Plate XXL— Fig. 1 should be " la."
Plate LX.— Fig. 6, "ad" should be "a'."

PLATE J.

(ZOOI.. (HALL. EXP. — l'.\KT XXXU. — 1S81.)- li.

PLATE I.

HoLOPUS RANGi, d'Orbigny.

The largest specimen, .

Fisf. 1. The bivial side.

Fig. 2. The trivial side.

Diam. Page
X 3 199

5

o

v^,:^-^-

o

<

CO

Q.

O

_J

o

•OO

^
H

PLATE IT.

PLATE 11.
HoLOPUS EANGi, d'Orbigny,

Diam. Page
Views of the trivial and bivial sides of an individual which has lost

three of the four bivial arms, . . . . . x 3 199

o a

1

<

CO

Q.
O

1

PLATE III.

(ZOOL. CHALL. EXP.— PART XXXII. — 1884.)— li.

PLATE III.

HoLOPUS RANGi, d'Orbigny.

Fig. 1. View of the cup from above after removal of two of the com-
posite axillaries, ......

Fig. 2. Interior of the cup of the specimen represented in PL II.,
showing the oral plates. Two axillaries are supposed to
be removed, ......

Fig. 3. Slightly oblicj^ue view of the dorsal surfaces of an axillary and
the attached first brachial, ....

Fig. 4. Dorsal aspect of an axillary, ....

Fig 5. Oblique view of its proximal face,

Fig. 6. Oblique view of the ventral sides of two united brachials,

Fig. 7. Proximal face of the second brachial, .

Figs. 8-13. Terminal faces of some of the large lower brachials.

Figs. 14, 15. Terminal faces of some of the small outer brachials.

Fig. 16. Portions of the pinnules, , . , ,

Diam.
3

Page
203

208

X

3

206

X

3

206

X

3

206

X

3

206

X

3

206

X

3

207

X

3

207

X

4

208

The Voyage of H M S"ChalleRJe:

Crinotdea.Pl III.

1«

" \'SS

^

¥ ^^

1+

14; f^

13

W S BUck del

HOLOPUS RANGM, D'Orbigny.

PLATE IV.

PLATE IV.
HoLOPUS KANGi, cVOrbigiiy.

Diam. Page
Side views of a young specimen, . . • • . x 10 203

>

c

^•:.--

o

O

<

cr

CO

CL

O

_l

o

PLATE V.

(ZOOL. CUALL. EXP. — PART XXXII. 1884.)— II.

PLATE Y.

HoLOPUS RANGi, d'Orbigiiy.

Fig. 1. The articular facets of the radials viewed from above, . x

Fig. 2. Horizontal section of the radials, 2 mm. below the edges of
the two bivial facets. This is an inferior view of the ring
of united radials shown in fig. 1 , . . . x

Fig. 3. Vertical section through the lower part of the cup, . . x

Fig. 4. Horizontal section through the lower part of the cup, about

5 mm. above the basal expansion, . . . x

Fig. 5. The calcareous network forming the peripheral portion of the

cup, . . . . . . .X

Fig. 6. The calcareous network forming the inner faces of the radials, x

Fig. 7. Portion of a horizontal section through the cup, showing the

relations of the two kinds of network, . . . x

Fig. 8. Ideal representation of the more regular network forming the

outside of the cup, . . . . . x

Figs. 9, 10. The youngest individual dredged by the "Blake,". x

Fig. 9. From the side.
Fig. 10. From above.

Diam. Page
3 202

3

9

200
201

200

45

^200

45

^200

12

^ a

g 200

P-,
<

80 J

200

20

204

The Voyage of H M S/'Chanenger."

Cnnoidca PI \^

^.-^Mm.ims.-'ih-

Figs.

Tig 9.

Fig 2.

Fig 6.

Fig 8.

€,;:

tig 7.

t

■-^•^

I .5 t-

HOLOPUS RANGI, D'Orbi^ny.

PLATE v..

PLATE V:i.

HoLOPUS RANGi, d'Oibigny.

Diam. Page

Fig. 1. Ventral aspect of a decalcified arm, showing the ambulacral

groove and one of its lateral branches, . . . x 20 206

Fig. 2. Side view of a decalcified arm, . . . . . x 6 208

Fig. 3. Side view of au entire ray, . . . . . x 6 208

^

r

%

A

,/^

o

CQ

o

o

<

o
o

j^ ^ f ri )^

•00

.J''

Vi !■

"OO

PLATE Yb.

(zooL. ciiAix. t;xi'.--rART xxxii. — 1884.)— li

PLATE Yh.

Tr.e Lettering is the same in all the Figures.
A. Axial conl.
ir. Coeliac canal.
;:/■. Genital cord.
j. Pinniile-jnint.
//. Interarticular ligament.

in. Muscle.

s/c. SiibtHntacuIar cnnal.

T. Tentacle.

V. Radial w.nter-vessel.

HoLopu.s RANGI, rl'()rl)igiiy.

Fig. 1. Transvcr.se section of a lower arra-joi'it,

Fig. 2. The upper portions of three pinnule-joints, showing the arnia
turc covering the bases of the tentacles, .

Fig. 3. The terminal portion of a pinnule showing the tentacular T)latcs
from a dry specimen, ....

Fig. 4. Inner aspect of an entire ray, showing the rolling in of th(
distal arm-joints, .....

Fig. 5. Ventral aspect of an axillary and the lower part of an aim
showing the ambulacral groove,

Diani.
X 80

X 40

X n

X G

Page
209

X GO 208

208

20G

200

TheVoyage of HMS Challenger'

Crmoidea PI Vt

X.

■ h

4

■■^i ^

.4;-

-^

.^f:^""' ,#^-^«^'

4
J

V

it

-w„«f

Geo We',t dti ft iiUi

HOLOPUS RANGI. D'Orb.gny.

G,„ ...„ ■».

PLATE Vc.

I'LATE Vc.

Thu LuttcriuL' is thu same in all the Figures.

A. Axial covd of arm.

Is. Interarticular ligament of stem

a. Axial cord of pimiulc.

711. Muscle.

ae. AmbuliiiTal ciiitheliuiii.

0. Oral plate.

an. Aiiaiiibulacni [ilate.

ov. Ovary.

C. Couloin.

rs. Radial space in stem.

ca. Fibrillar sheath round vascular axis of stem.

sp. Side plate.

cc. Coeliac canal.

stc. Subtentacular canal.

cic. Ciliated cup.
cp. Covcriug I'late.

T. Tentacle.

t. Testis.

i/c. Genital coid.

0. Central vascular axis of stem.

J. Arm-joint.

w. Radial water-vessel.

j. I'innule-joint.

lop. Water-pore.

Id. Dorsal ligament.

Figs. 1-3. HoLOPUs KANGi, cl'Orbigny.

Diam.

Fig. 1. Portion of ;iu oblique sectiou through au arm-joiut aud the

base of a pinnule, . . . . . . x 30

Fig. 2. Transverse sectiou of an arm-joiut with attached pinnule, . x 30

Fig. 3. Portion of a nearly longitudinal section of a pinnule, . . x 35

Page

210
210
209

Figs. 4-10. Hyocrinus bethellianus, Wyv. Thorns.

Fig. 4. Terminal face of a stem-joint, ....

Fig. 5. Optical section of a piece of stem cut longitudinally, .

Fig. 6. Portion of the disk, showing one oral plate and the anambu-
lacral plates lying outside it, .

Fig. 7. Transverse section (jf a pinnule, ....

Fig. 8. Terminal portion of a pinnule, mounted in dammar and seen
in o^Dtical section, ......

Fig. 9. Portion of a dry pinnule about the end of the enlargement to
receive the genital gland, .....

Fig. 10. Lower portion of a pinnule, mounted in dammar and seen in
optical section, ......

X

20

218

X

15

218

X

20

219

X

80

220

X 15

X 12

16

220

220

220

The VoyagG orMS/'Challenger."

'^'C.'h-i'lff'^'v

\-

/

. ...auj'Jtai.io.u^

vv

If

frxrrtjf^-

\. J

A.-

Id

,y

V^-:

±

(^HH^^K

///'■

\r

10

-.-^

hi

ighlcy drl ot.lith

I _3 HOLOPUS RANG

:\V.

4_I0 HYOCRINUS BETHELLIANUS, Spn.

Hanhart imp.

PLATE Yl.

(ZOOL. CHAI.L. EXP.— TAKT XXXII. — llS8t.)— li.

PLATE VI.

Hyocrinus bethellianus, Wyv. Thorns.

Fig. 1. Radial view of the calyx and arms,

Fig. 2. Inteiradial view of the entire specimen,

Fig. 3. The calyx and arm-bases ; anal side, .

Fig. 4. The disk, from above, ....

Fig. 5. Inner view of a fragment of the upper portion of the d

Fig. 6. Terminal joints of a pinnule, mounted in dammar,

X

Diaiti.
4

Page
220

X

2

218

X

5

218

X

10

219

isk,

X

8

220

X

40

220

56 of H M.S. "Challenger.

Crinoidfia.PlVI

HYOCRINUS BETHELLIANUS, Sp .n.

PLATE VII.

PLATE VII.

Bathyceinus aldrichianus, Wyv. Thorns.

Fig. 1. All entire specimen, ....

Fig. 2. The calyx aiicl arms, ....

Fig. 3. Oblique view of the disk.

Figs. 4, 4«. Dorsal and ventral aspects of the radial axillaries.

Figs. 5, 5a. Dorsal and ventral aspects of the second radials.

Figs. G, Go. Outer and inner faces of tlie first radials,

Fig. 7. Terminal portions of two pinnules, mounted in dammar
and seen in optical section,

Fig. 8. The lower part of an arm from aljove,

Fig. 9. Lower stem-joints and root,

Fig. 10. Middle stem-joints,

Fig. 11. Upper stem -joints and basals,

Figs. 12, 13. Terminal faces of lower stem-joints.

X
X

X
X

Diani.
1

2

2

G

G

G

X IG
X 4

X '^J-

Page
241

241

234

231

231

230

233
233

28
27

X
X

* ' S

I '^

oi i S 22G

-'4, o . •^^'J

5 J

27

The Voyage of H.M.S "Chanenger."

; VII

i:;,'"..

4^\ ■

%' ^ <" ''■ J'

#■ .■':'■.

-w-y. V

-' -. '' ^^

5 1 '

-^ ■' <■■' ■'■;_

!■-■<■ '.

■ L/?<

4::^-

T^ -

w-

.. '"? ■

;: ^ , C

V?' ..

"^ ■ .i

.■ "■, .»*.

■^

^», \.,^--*r,.

!■

-v^~>:y

"'- ,'W>

v^'

12

■WSBUdcDii

V, ir AJC JDlmfltaiL.Ii^o^ii:pherd.

BATHYCRINUS A LD R I C H I A N U S, Sp n -

PLATE Vila.

(ZOOU CHAM.. EXP. — TAUT X.X.MI. — ISSt.)— 1 !•

PLATE VILi

The LutteriiiG; is the same in all the Fi'mrt

m. Fibrillar slieatli romi.l vascular axis of stem.

ca'. Its railiatiug uxtensions.

ch'. Downward ]irolonf;ation iuto the stum of the cavities of

the chambered organ.
Id'. Fossa lodging the dorsal ligament.

W. Fossa lodging the iuterarticular ligament.

Is. Interarlicular ligament of stem.

j). Pigment bodies.

/•»(,'. Muscular fosste on the distal face of the lirst radial.

rs. Radial space in stem.

v. Vascular axis o£ stem.

Figs. 1-21. Bathycpjnus aldiiichianu.s, Wyv. Thorns.

Fig. 1. Optical section of a piece of stem, cut longitudinally,

Fig. 2. Transverse section of stem, central portion,

Fig. 3. Upper face of top stem-joint,

Fig. 4. Upper face of tHird joint.

Fig. 5. Upper face of fifth joint,

Fig. 6. Upper face of ninth joint,

Fig. 7. Upper stem-joint.

Figs. 8, 9. Middle stem-joints,

Figs. 10, 11. Lower stem-joints,

Figs. 12-14. Tlie basal ring, .

Fig. 12. From above.
Fig. 13. From the side.
Fia;. 14. From below.

Kg. 15. Distal face of first radial.

Fig. 16. Distal face of second radial, .

Fig. 17.' Distal aspect of axillary.

Figs. 18, 19. Distal and proximal faces of the fifth brachial,

Figs. 20, 21. Proximal and distal faces of an outer brachial,

Diam.

Page

X

66

31

X

300

31

X

26

26

X

26

26

X

2G

27

X

26

27

X

16

27

X

16

27

X

16

27

X

17

226

X

16

231

X

16

231

X

16

231

X

16

232

X

16

232

Figs. 22, 23. Bathycrinus campbellianus, n. sp.
Figs. 22, 23. Proximal and distal faces of an outer brachial, .

1 This figvu-e is wrongly lettered 7 ou the ^date.

30

232

1 :.-*....iv-^ '- . •_-. ■ vit*v

20

"e^

12

18

13

l.i

I*

10

23

16

22

2r

BATHYCRINUS ALDRICH I AN US, Sp.n.
CAMPBELLIANUS, Sp

PLATE Vllb.

PLATE VIII,

The Letterins is the same in all the Figures.

A. Axial cord of ray or ann.

a'. Its branches.

ai. Primary interradial cords.

£, Basal ring.

By First brachials.

C. Cocloni or body-cavity.

c. Network of connective tissue within the coclom.

ceo. Interradial portion of the circular commissure.

ch. Cavities of the chambered organ.

fg. Fore-gut.

O. Gut.

L. Interradial ligament.
I. Basiradial ligament.

Id. Dorsal ligament.

li. Interarticular ligament.
El. First radials.
R,. Second radials.
iij. Tliird radials.

re. Rectum.
■n)i. Muscles uniting the first and second radials.

V. Vascular axis of stem.

X. Plexiform gland.

Bathycrinus aldrichianus, Wyv. Thorns.

Fig. 3. Horizontal section of the calyx at the level of the basiradial

suture,

Diam.

Fig. 1. Vertical section of the calyx, with the lower part of the disk, . x 20
Fig. 2. Horizontal section of the basal ring, . . . . x 45

X 45

Page
235

22G

228

Fig. 4. Horizontal section of the radial pentagon at the level of the

circular commissure, . . . . . x 20 230

Fig. 5. Horizontal section of the calyx, through the articulation of the

first and second radials, . . . . . x 20 230

Fig. 6.' Horizontal section of the calyx, through the upper part of tlie

second radials, . . . . ' . x 20

231

Fig. 7." Horizontal section of the calyx, through tlie middle of the

third radials, . . . . . . x 22

231

Fig. 8. Horizontal section of the disk, through the articulation of the

first and second brachials, . . . . x 20

232

• In the upper part of this fitjnre a should be «', as in the lower part.
- On the U-ft of this fisrure / should be L.

The Voyage of HMS/'Challeng'tr"

Crinoidea Pl.VIIl).

S'?y?5v*«»,-;,.

Berjeau& Highley deleb lith.

PH.C. dir

BATHYCRINUS ALDRI C H I A N U S, Sn ;

Hanliarb imp

PLATE VIII.

(ZOOL. CHALL. EXP. — PART XXXII. — 1884.) — H.

PLATE VIII.

BaTHYCRINUS CAMPBELL IANUS, 11. sp.

Fig. 1. Calyx and arms separated from the stem and basal ring,

Fig. 2. The radials and arm-bases,

Fig. 3. Side view of one arm, .....

Fig. 4. Four of the lower arm-joints, from the side,

Fig. 5. Side view of a pinnule, mounted in dammar, .

N.B. — The magnification given above is only approximate, as it was not recorileil at tlie time the Plate
was drawn, though I have endeavoured to work it out subsequently.

Diam.

Page

X

3

238

X

8

238

X

6

234

X

IG

238

X

20

233

The Vojage or H M.S 'ChaUKiiJir,)

Crinomfla i-i VHI

WS. Black, dsl.

?.Hath,L«Ui'Earf

BATHYCRINUS C A M PBE LLl AN U S , Spn.

PLATE Villa.

PLATE Villa.

The Lettering is the same in all the Figures.

A. Axial cord of arm.

a'. Its branches.

ac. Ambulacral epithelium.

£„. I'econd brachial.
b. Radial blood-vessel.

Jr. Plug of calcareous tissue, called a basal rosette by Sars.

cc. Coeliac canal.
ceo. Interradial portion of circular commissure.

cp. Covering plate.

O. Gut.

gc. Genital cord.
gc'. Genital canal.
ico. Intraradial portion of circular commissure.

/. Arm-joint.

Id. Dorsal ligament.

li. Interarticular ligament.

mj). Muscle -plate.

01. Ventral or ambulacral nerve,

i^i. First radial.

re. Rectum.

rm. Muscles uniting the first and second radials.

ste. Subtentacular canal.

T. Tentacle.

rrf. Ventral radial furrow.

ly. Radial water-vessel.

X. Plexiform gland.

Figs. 1-3. Bathycrinus gracilis, Wyv. Thorns.

Fig. 1. Head and upper part of stem, .

Fig. 2. Middle part of stem, ....

Fig. 3. Lowest part of stem, . . . . *

Fig. 4. Ventral portion of a transverse section of an arm.

Fig. 5. An entire section through the middle of an arm-joint,

Diam.

Page

. X 10

243

. X 10

27

. X 10

27

. X 285

236

. X 165

236

Figs. 6-8. Rhizocrinus lofotensis, Sars.

Fig. 6. Horizontal section through the lower part of the radials, of

which there are six, . . . . . x 35 250

Fig. 7. Horizontal section of the calyx of the same individual at the

level of the articular faces of the radials, . . . x 35 252

Fig. 8. Horizontal section of the disk of a pentamerous specimen at

the level of the second brachials, . . . . x 40 254

TheVoyag-e of H.M.S. Challeng'er"

Crinoidea PI Vllld,,

X.

x

stc

\l

■)

,p-

4i .{

c

/W^ ^^^-r

B2

BerjeaufcHiqhli

KHC dip.
i-3. BATHYCRINUS GR ACI LI S_, '.Vy T:i
4,5. BATHYCRINUS ALD R I C H I A N US, Sp n.
6-8. RHIZOCRINUS LO FOTEN S IS, Sa:^s.

l;,tr, nart irnp.

PI ATE IX.

(ZOOL. CHALt. EXP. — PART SXXII. — 1884.) — li.

PLATE IX.
Figs. 1, 2. Ehizocrinus lofotensis, Sars.

Diam. Page
Fig. 1. An entire specimen, . . . . . . x 3 259

Fig. 2. The calyx and arms, . . . . . . x 6 259

Figs. 3-5. Ehizocrinus rawsuni, Pourtales.

Fig. 3. General view of an entire specimen witli the exception of tlie

lowest part of the stem,^ . . . . . x 2 262

Fig. 4. Portion of an arm, from the side, . . . . x 12 60

Fig. 5. Dorsal aspect of an arm, . . . . . x 12 60

N.B. — The magnification is only approximate, as in PI. VIII.

' This figure was drawn from a dry specimen, parts of which had shown a tendency to separate ; and the gaps had
consequently been filled up with gum. In consequence of this the syzygy between the first two brachials of the middle
ray was wrongly drawn as a muscular joint- like that between the second and third brachials. It should have been
represented as a simple line without any gap between the joints, like the syzygies in the remaining part of the arm.

The V

I' Ti M :' i'ii:riieii','i=r

Ttf 3BUcj£ Del;

RHIZOCRINUS LOFOTENSIS, M. Sars.
RAWSONI, POURTALES.

,i.£r*plioi

PLATE X.

PLATE X.

The same Letters occur in both Figures.

br. The plug of calcareous tissue iu the centre of the radial funnel, which was called a basal rosette by Sars.
vrf. The ventral radial furrows of the calyx.

Figs. 1, 2. Rhizocrinus lofotensis, Sars.
Two views of a calyx with three of the first brachials remaining m situ.

Fig. 1. From above,

■'o

Diam.

Page

X

15

252

X

10

266

X

10

267

X

15

250

X

15

25

Fig. 2. From the side, ......

Figs. 3-20. Rhizocrinus rawsoni, Pourtales.

Fig. 3. Calyx of a specimen from near the Azores ; side view,
Fig. 4. The same from above, ....

Fig. 5. The same from below, ....

Fig. 6. Calyx of a specimen from off Havana, with the first brachials

remaining ?H s^Vh ; from above, . . . . x 10 254

Fig. 7. A similar view of the calyx of another Havana specimen which

has undergone reparation. All the first brachials are

visible and four of the small second brachials ; while the

centre of the cup is occupied by the oral plates, .
Fig. 8. Calyx of another specimen from the Azores with the first

brachials attached ; from above, ....
Fig. 9. Upper face of the top stem-joint of the Azores specimen repre-
sented in figs. 3-5, .....
Fig. 10. Upper face of the fourth joint of the same stem,
Figs. 11,12. Terminal faces of two joints from rather above the middle

of the stem ; fig. 12 representing that nearer the calyx, . x 15 26

Figs. 13, 14. Terminal faces of two joints from the lower part of the

stem ; fig. 13 representing that nearer the root.
Fig. 15. The root of a specimen from Havana,
Fig. 16. Two stem-joints with attached Foraminifera (TnmcatuUna

lohatula), ......

Fig. 17. Proximal face of an epizygal at an arm-syzygy,

Fig. 18. Distal face of the corresponding hypozygal, .

Fig. 1 9. Dorsal view of the second and two following brachials, to show

the backward projection of the proximal face of the second

brachial (an epizygal), .....
Fig. 20. Side view of the upper part of the disk, after removal of one

arm. The arms have separated from the cup at the syzygy

between the first and second brachials,

X

10

255

X

10

255

X

15

25

X

15

25

X

15

26

X

li

257

X

10

134

X

15

254

X

15

254

X

10

254

X

15

255

;iM.S."ChalI. :

Cnnoideji

VPJ

B?rieiau&Hi9Kley ciel.et litVi .

PHC dir

1.2 RHIZOCRINUS LOFOTENSIS, Saps,
3,-20 „ RAWSONI, Pourt,

PLATE XL

(ZOOL. CHALL. EXP. — PART XXXIL — 1884.) — li.

PLATE XL

Pentacrinus astekius, Linn., Sp.
Natural size, p. 300.

PLATE XIJ.

PLATE XII.

Figs. 1-14. Metacrinus angulatus, n. sp.

Figs. 1, 2. Upper and lower views of the basal ring,

Figs. 3, 4. Proximal and distal faces of the fourth (axillary) radial,.

Figs. 5, 6. Proximal and distal faces of the third radial,

Figs. 7, 8. Proximal and distal faces of the epizygal of the second
radial, .......

Figs. 9, 10. Proximal and distal faces of the hypozygal of the second
radial, .......

Fig. 11. Lower or dorsal aspect of a first radial,

Fig. 1 2. Distal articular face of a first radial, ....

Figs. 13, 14. An isolated basal, from below and above,

Diam.

Page

X 3

291

X 3

9

X 3

48

48

3

9

3

293

3

9

3

293

Figs. 15-25. Pentackinus asterius, Linn., sp.

Figs 15, IG. LTppcr and lower views of the calyx,

Fig. 17. Dorsal aspect of the two outer radials with the first distichals
attached, ......

Figs. 18, 19. Proximal and distal faces of the third (axillary) radials,

Figs. 20, 21. Proximal and distal faces of the second radials.

Fig. 22. Lower or dorsal aspect of a first radial.

Fig. 23. Distal articular face of a first radial, .

Figs. 24, 25. An isolated basal, from below and above,

33

X

3

300

X

3

305

X

3

305

X

3

293

X

3

9

X

3

293

ThcTojagi^ of H.M S.'Challen^M-,-

Crinoidea.Pl. Xn.

T\f&'^

CI

13

15

17

s^^

16

W.S-Black del

1-14. METACRINUS ANGULATUS, Sp. n.
15-25. PENTACRINUS ASTERIA, L.

r.K»ai,Ii!!f Ein"

PLATE XI I

(ZOOL. CHAl.L. EXP. — PART X XXII. — ]S84.)— li.

PLATE XIII.

Fig-

Fig.

FisT.

Fi^.

Fio-.

Fig.

Fiff.

Fig.

Fig.

Figs.

Fig-

Fig.

Fig.

Fig.

Fig.

Pentacrinus asterius, Linn., sp.

Diara. Page

1. The calyx and arm -bases, . . . . . x 2 300

Figs. 2-4. A nodal joint from the upper part of the stem.

2. Ujiper face, . . . . . . . x 5 17

3. Lower ftice, . . . . . . . x 5 17

4. Side view, . . . . . . . x 5 14

5. A nodal joint from tlie lower part of tlie stem ; lower face, . x 4 17

6. The same ; upper face, . . . . . x 4 17

7. Portion of an internode from near the top of the stem, . x 2 18

8. Nodal and internodal joints from the lower part of the stem. . x 2 30.5

9. Face of a young internodal joint, . . . . x 5 15

10. 11. Faces of older internodal joints, . . . . x 4 17

12. Terminal face of a pinnule-joint, . . . . x 5 10

13. Termination of an arm, . . . . . x 2 300

14. One of the middle arm-joints and its pinnule. . . x 4 28G

15. The ambulacral skeleton of an outer pinnule, . . x 30 79

IG. Ventral aspect of the hasal joints of a lower pinnule, showing

the development of its ami )ulacral skeleton, . . x 10 79

TheVqyagp :'""." ^' 'Challenger:

Crmoidea.PlXJll.

nerjeau iSoHighley

PENTACRINUS ASTERIA,

II .-i iHidr'r Ultf.'

. PLATE XIV.

PLATE XIV.

Pentacrinus mullert, Oersted.
Natural size, p. 306.

Ine Voyage of B.MS."CluaieiiJ«r"

W SBUi.asl

T ifafl^TiW Sirf

PENTACRINUS MULLERI, Oersted.

PLATE XY.

(ZOOL. CEALL. EXP. — PART XXXII. — 1884.)— li.

PLATE XV.

Pentacrinus mulleri, Oersted.

Fig. 1. The calyx and arm-bases,

Fig. 2. Another specimen with four radials,

Fig. 3. An arm of this specimen (Martinique),

Fig. 4. Portion of the stem, ....

Fig. 5. Face of an internodal joint,

Fig. 6. Lower or syzygial face of a nodal joint.

Figs. 7, 8. The ends of two pinnules, showing the termination of the
ambulacral skeleton ; from the side,

Fig. 9. The same ; from above, . . . . .

X

Diam.

2

Page
311

X

2

311

X

2

311

X

2

310

X

4

12

X

4

17

X

15

80

X

20

312

The Voyage of HMS "Challenger"

Crinoidea.Pl.XV.

Berjeau & HigWey del.et lith

PHC.duv

PENTACRINUS MLJLLERI, Oersted

PLATE XVI.

PLATE XVI.
Pentacuinus macleaeanus, Wyv. Thorns.

Diam. Tagc

Fig. 1. Tli(^ entire specimen, . . . . . . x 2 312

Fig. 2. The end of a pinnule ; from the side, . . . . x 12 55

Fig. 3. Yi'ntral aspect of the basal joints of a piiinnh'. slioAving the

development of its amlndaeral skeleton, . . . x 12 80

The Vova«e of H.M S " Challenger

Crinoid.ea..Pl.XVI.

J.J Wilii CBerieau-deL EH-C dir

WostNe\vm«ii K C° lalli.

PENTACRINUS MACLEARANUS, sp.n.

PLATE XYII.

(ZOOL. CHALL. IS.NP.— PAKT xxxii. — 1664.)— !'•

12

320

12

320

8

320

PLATE XVII.

Fig. 1. Pentacrinus macleaeanus, Wyv. Thorns.

Diam. Page

Fig. 1. Arm-gvoo-ve oi Pentacrinus 7naclearanus, . . . x 8 80

Figs. 2-6. Pentacrinus wyville-thomsoni, Jeffreys.

Fig. 2. Ambulacral skeleton of a pinnule ; from the siJe, , . x

Fig. 3. The same, near the base of the pinnule ; from above, . x

Fig. 4. The arm-groove and In-achial ambulacrum, . . . x

Fig. ."). Vertical section of a decalcified specimen. Much of the alimen-
tary epithelium has fallen away. The lobulated organ to
the right of the centre of the figure is the plexiform gland, x 3 33

Fig. 6. The disk of a dry specimen, . . . . . x 6 320

Figs. 7, 8. Pentacrinus asterius, Linn, sp.

Fig. 7. Ventral aspect of one of the lower arm-divisions, showing tlie

ambulacral skeleton, . . . . . x 8 79

Fig. 8. The arm-groove and In-achial aml)uhierum, . . . x S 79

Figs. 9, 10. Pentacrinus MiJLLERi, Oersted.
Fig. 9. Tlie arm-groove and brachial amliulacrum. . . . x 8 80

Fig. 10. The disk of a spirit-specimen, , . . . x 3 311

The Voyage cP H MS r>i;illfnnf"r"

r.v:no,r]^^ Pl.X.VIl.

Bcrjeau &■ Highiey. dCi.etliLli

I'.H C Dir

PENTACRINUS MACLEARANUS, Pip n 2-6 P WYVILLE THOMSON I, Jeffreys,
7,8. P, ASTER I A. L 9, 10 P MU LLERI, Oersted

Hanhotri imp.

PLATE XVni,

PLATE XVIII.

Pejttacrinus wyville-thomsoni, Jeffreys.

Fig. 1. The head, with the upper part of the stem,

Fig. 2. The calyx and arm-bases,

Fig. 3. A young specimen, ....

Figs. 4, 5. Upper and lower views of the radial pentagon.

Figs. G,7. An isolated basal from below and above.

Figs. 8, 9. Proximal and distal faces of the third (axillary) radial.

Figs. 10, 11. Proximal and distal faces of the second radial,

Diam.

Pago

X

14

313

X

2

320

X

2

21

X

3

33

X

3

293

ial.

X

3

50

X

3

50

ThcToyage oT B.M.S.'Cliallengcr"

Crinoidea PI. XVm.

WS. Black, dtl.

? liitk.I.,(i' ZiW

PENTACRINUS W YV I L L E - T H 0 M S 0 N I, Jeffreys.

PLATE XIX.

(ZOOL. CHALL. EXP. — i'AltT XXXII. — 1884.)--IL

PLATE XIX

Pentacrinus wyville-thomsoni. Jeffreys.

Fig. 1. An entire specimen, .....
Fig. 2. Portion of an internode from near the top of the stem, .
Fig. 3. Nodal and internodal joints from the upper part of the stem.
Fig. 4. Nodal and internodal joints from the lower part of the stem.
Fig. 5. Portion of an internode from the lower part of the stem,
Fig. 6. Interradial view of the calyx and arm-bases.
Fig. 7. Radial view of the same parts in another specimen,

Diain.

Page

nat.

size

313

X

4

18

X

4

14

X

4

14

X

4

17

X

2

320

X

2

320

The Voyage of H M S, "ChaJlenger

Crinoidea.PL.XIX.

Geo. West li*Ji

WB C air.

PENTACRINUS WYVl LLE - TH OM SON I , Jeffrejys.

Tft'est X<nvir.a7i &, C° jTTi^.

PLATE XX.

PLATE XX.

Pentacrinus WYViLLE-THOMSONi, Jeffreys.

Diain. Puge
Kig.'5. l-o. The riug of united basals, . . . . . x 10 282

Fig. 1. From below.

Fig. 2. From above.

Fig. 3. From the side.

Fig. 4. The central plug of limestone network within the ealvx ; from
above, .......

Fig. 5. The same ; from below, .....

Fig. 6. Interior view of the calyx, after removal of two radials and
one basal, .......

Figs. 7-9. The ring of united radials, .....
Fig. 7. From the side.
Fig. 8. From above.
Fig. 9. From below.

X

20

.34

X

10

34

X

10

34

X

10

33

The Voj9L§e of H.M,S "Clmllenger

Crinoidea PI XX

Geo.Wesi litti

W.B C Jir

PENTACRINUS W YVI LLE - TH O MS O N I , Jeffreys.

We si: Ne-^ -man &.€" imp

PLATE XXL

(ZOOU OHALl,. EXP. — PAllT XXXII. 1884.) — 11.

PLATE XXI.

Pentacrinus wyville-thomsoni, Jeffreys.

Fig. 1. A third or axUlary radial,
a. Distal face,
h. Side view,

c. Dorsal view,

d. Proximal face, .

e. Ventral surface.

Fig. 2. A second or axillary distichal,
a. Distal face,
h. Side view,

c. Dorsal view,

d. Proximal face, .

e. Ventral surface.

Fig. 3. A first distichal,

a. Ventral surface,

b. Proximal face, .

Fig. 4. A first brachial,

a. Ventral surface,
h. Proximal face,-.

c. The proximal face of another specimen,

Fig. 5. A second radial,

a. Distal face,

h. Ventral surface,

c. Proximal face, .

Fig. 6. A first radial, .

a. From beneath, .
1). Side view,

c. From above, .

d. Distal face.

Fig. 7. An isolated basal,

a. From beneath, .
h. Side view.

Diam. Page
X 10 313

9
38

50

X 10

X 10

X 10

X 10

X 10

X 10

313
9

38

50

313
38

38

38

313
50

313
125
125
125
9

313
293
125

The Vqya^^e of H.M, S "Challen^w "

Cri-noirlea.P] XX]

Geo West litii

WB.C. dLT.

PENTACRINUS WYVILLE-THOMSONl, Jeffreys.

"YfestKewMLaiL & C° Jinr.

PLATE XXn.

PLATE XXII.

Fig.

1.

Fig.

2.

Fig.

3.

Fig.

4.

Fig.

5.

Fig.

6.

Fig.

7.

Fig.

8.

Pentacrinus wyville-thomsoni, Jeffreys.

Oue of the upper stem-joints, .....

The top stem-joint of a young specimen with a smaller one in-
terpolated above it, .... .
A small interpolated joint, .....
Another interpolated joint, somewhat younger,

An older joint from the upper part of the stem, with an impres-
sion on its upper surface in which rested the young joint
shown in fig. 4, .

The upper face of a young nodal joint,

7. A young iuternodal joint, .....

8. A somewhat older internodal joint, ....

Figs. 9-12. Young internodal joints from the growing part of the stem,
with smaller ones interpolated above them at different stages
of development, ......

Fig. 13. An older, but still immature internodal joint, .

Fig. 14. A young internodal joint from the upper part of the stem,

Fig. 15. Upper face of a young nodal joint from the upper part of the
stem, .......

Fig. 16. Lower or syzygial face of a nodal joint from the middle of the
stem, .......

Fig. 17. Lower face of a supra-nodal joint from the upper part of the
stem, .......

Fig. 18. Upper face of the corresponding nodal joint, .

Figs. 19-24. Mature joints from a little below the middle of the

Fig. 19. Upper face of a nodal joint, .

Fig. 20. Its lower or syzygial face,

Fig. 21. The upper face of an infra-nodal joint,

Fig. 22. Its lower face, ....

Figs. 23, 24. Ordinary internodal joints,

Figs. 25, 26. Old internodal joints from the lowest part of the stem.

Fig. 27. An enlarged nodal joint from the end of the stem with its
lower face rounded and the central canal closed up,

Fig. 28. Portion of the calcareous network forming the substance of the
stem-joints,

Diam.

Page

X

10

17

X

15

16

X

10

16

X

10

16

X

10

16

X

10

17

X

10

17

X

10

17

X

10

16

X

10

15

X

10

15

X

10

17

X

10

17

X

10

14

X

10

14

stem.

X

10

17

X

10

17

X

10

14

X

10

14

X

10

17

X

10

17

19

The Voyage of H.M. S."CKaJleM.|er. "

CriiLOuie,-? PI. XXI!

Geo W=ot lid.

W. B C. dxr

PENTACRINUS, WYVILLE-THOMSON I , Jeffreys

West Newn-..-ui.S:C? iJ'.y.

PLATE XXI] T.

(ZOOI.. CHALL. EXP. — PART XXXII. — 1884.) li.

PLATE XXIII.
Pentacrinus WYViLLE-THO.MSONi, Jeffreys.

Diam. Page
Figs. 1, 2. Young interpolated juiiits from near the top of the stem, . x GO 16

Fig. 3. A somewhat older, but still quite immature joint, . . x GO 17

Tke Voyage of H,M S, " Challenger "

Crinoiaea.Pl XXin

Geo. West Litii

vV. B.C. dir.

PENTACRINUS WYVILLE-THOMSONI , Jeffre_ys.

West -Nevs-jiiiiii & C^ imp

PLATE XXIV.

PLATE XXIV.

The Lettering is the same in all the Figures,

A. Axial cord of the ray. cv. Cirrus-vessel

ai. Primary interradial cords.

ar. The secondary (radial) cords resulting from their bifurca-
tion.

£. Basal.

ca. Fibrillar sheath round vascular axis of stem.

'ca'. Its radiating extensions.

ceo. Interradial portion of the circular commissure.

ch. Cavities of the chambered organ.

ch'. Their downward prolongations into the stem.

cJic. Fibres of connective tissue which traverse the fibrillar

envelope of the chambered organ.
chn. The nodal enlargements of the peripheral vessels of the
stem {ch').

ico. Intra-radial portion of the circular commissure.

i. Interi'adial ligament.
I. Basiradial ligament.

lb. Interbasal ligament.

Id. Dorsal ligament.

Is. Interarticular ligament of stem.

27. Pigment gi'anules.
ill. First radial.

rp. Plug of calcareous tissue occupying the central funnel of
the calyx.

V. Central vascular axis of stem.

X. Plexiform gland.

Diam.

Page

X

15

23

X

90

120

X

90

107

Pentacrinus wyville-thomsoni, Jeffreys.

Fig. 1. Horizontal section of an upper internodal joint.

Fig. 2. The central part of the same section, enlarged,

Fig. 3. Horizontal section through the upper part of a nodal joint,

Fig. 4. Horizontal section of a nodal joint at the origin of the cirrus-
vessels, . . . . . . . X 90 107

Fig. 5. The central part of a horizontal section through an internodal

joint near the top of the stem, . . . . x 90 23

In this figure ch should be cJ/', as in fig. 2.

Figs. G-9. Four out of a series of nearly horizontal sections through the calyx.

Fig. 6. The centre of a section through the lower part of the basal
ring, ......

Fig. 7. Section through the upper part of the basal ring,

Fig. 8. Section through the lower part of the radials, .

Fig. 9. Section through the middle of the radials.

X

90

105

X

12

124

X

12

125

X

12

125

TheVoyaye of H.M.S."ChaIIenger'.'

Crmoidea.Pl.XXIV.

.a«j!'

ch'

Is

\^.

-*•' Hill II i-*^

* di 'i- " „ (^ e .

:*« S

■' Q) ^*^' Xv^"^

ch

lb

.<v::!i.

Ri

• ft ^

'%K-7>i}s^r''

?i f«

^*,-^

chn

J^-ts

f^^

.4

Is

chn

S>tiii&/.>,',,.

M'^'"^m.

.^"

3

Is

vp

ch

Id

Berjeau&Highleydf^ i'\ ^.I'h

P.HC du-

PENTACRINUS WYVILLE-TH 0 MSON 1, Jeffreys

Hanliart imp.

PLATE XXV.

(ZOUL. CUALL. EXP. — PAliT XXXII.- 1884.) — li.

PLATE XXV.

PeNTACKINUS ALTKRNIOIRRITS, n. sp.

Head aiul npi:,^!- part of stem -^'''"^- J'age

X 2 321

The Voyage of H M S "Challenger"

Crinoidei PI XXV

WS BlacVae:

PENTACRINU5 ALTERNICIRRUS.Spn

p Hua.ijii.' T./tr'

PLATE XXYI.

PLATE XXVI.

Pentaceintjs alteenicirrus, n. sp.

Fig. 1. The disk, .......

Fig. 2. The anal interradius, enlarged, ....

Figs. 3, 4. Inner and outer surfaces of the two outer radials with the
distichals and palmars attached, ....

Figs. 5, 6. The third or axillary radial, ....

Fig. 5. Proximal face.
Fig. 6. Distal face.

Figs. 7, 8. The second radial, ......

Fig. 7. Proximal face.
Fig. 8. Distal face.

Fig. 9. The lower face of a first radial, with one basal remaining
attached, ....

Fig. 10. Distal face of a first radial,

Fig. 11. The calyx from below,

Fig. 12. Side view of a nodal and infra-nodal joint in their relative
positions, ....

Figs. 13, 14. Upper and lower faces of a nodal joint,

Figs. 15, 16. Upper and lower faces of the infra-nodal joint,

Fig. 17. Faces of internodal joints.

Fig. 18. Various stages in the development of internodal joints

Diam. Page

X 3 n

X 5 11

X
X

X
X

X

X

X
X

3
3

3

3
3

4
4
4
4
4

50
5

293

9

33

13
323
13
12
12

r IT f.i- '.■ r n

"" "' ^inrien^cr"

Cr:T-r,,i„, PI vx'/l

15

.<;^.

^^

17

vi*^'*^ ;<Qj/NeE'

18

10

11

PENTACRINUS ALTERNlClRRUS,Sp n

PLATE XXVII.

(ZOOL. CUALL. liXP. — PAUT XXXll. — 1884.) — li.

PLATE XXVII.

Figs. 1-10. Pentacrinus alternicirrus, n. sp.

Fig. 1. Portion of stem containing two nodal joints; interradial
view, .......

Fig. 2. Lower or syzygial face of n nodal joint which bears three
cirri, ......

Fig. 3. The corresponding face of a joint bearing only two cirri,
Fig. 4. The ambulacral skeleton of a pinnule ; from above,
Fig. 5. The end of a pinnule ; from the side, .
Fig. 6. The arm-groove and brachial ambulacrum.

Fig. 12. The same ; from above,

Fig. 13. The l)rachial ambulacrum ; from above,

Diam.
3

6

6
15
12
12

Figs. 7, 8. Inner and outer sides of an arm, bearing a pinnule-cyst of

Myzostovia deformator, von Graff, . . . x 3

Figs. 9, 10. Inner and outer sides of an arm, modified by a cyst of

Myzostoma jjentacrini, von Grafi'. . . . x 3

Figs. 11-13. Pentacrinus naresianus, n. sp.

Fig. 11. The ambulacral skeleton of a pinnule; from the side. Some

of the covering plates have been removed, . . x 1 5

Page
270

323

323
323

80

80

324
324

X 12
X 12

78
78

ThfiVoyajje of H M.S. "Challenger"

Crmoidea.Pl.XXVli.

»

12

13

Beijeauifc Highley del.et lith .

P.RC.dir.

■10. PENTACRINUS A LT E R N 1 C 1 R RU S, Sp.n .
■13. „ NARESIANUS, Sp.n.

Hanhart imp.

PLATE XXYin.

PLATE XXVIII.

PeNTACEINUS NAKESIANUS, 11. tip.

Fig. 1. The head,

Fig. 2. The stem

Diam.

Page

X H

324

X H

324

The Voyage of H.M.S "CaalleBgcr:'

Cnnoidei PI XXVllI

m

M

■M

Yf s BUi, ad

PENTACRINUS NARESIANUS, 5p. n

P HoQi4-i<H J*^'

PLATE XXIX.

(ZOOL. ClIALI,. EXP. — PART XXXII. — 1884. ) — li.

PLATE XXIX.

Pentacrinus naresianus, n. sp.

Diam. Page
Fig. 1. The calyx and lower parts of the arms of au adult mdividual.. . x 2 326

Fig. 2. A youug individual, . . . • • • ^ Ij ^^7

Tiie Voyage of H.M S"Cl>alIen|er"

CnDO.aea Pi XXIX

PENTACRINUS NARESIANUS, So r.

PLATE XXX.

PLATE XXX.

Pentacrinus naeesianus, n. sp.

Fig. 1. The calyx and lower parts of tlie arms of au adult individual,

Fig. 2. The disk, ......

Figs. 3, 4. The calyx ; from above and below, .

Fig. 5. The lower face of a first radial, with one basal remaining
attached, ......

Figs. 6, 7. An isolated basal from below and above,

Fig. 8. A first radial, from beneath, ....

Fig. 9. Distal face of a first radial, ....

Figs. 10, 11. Proximal and distal faces of a second radial.

Figs. 12, 1.3. Proximal and distal faces of a third or axillary radial.

Fig. 14. Dorsal aspect of the axillary with the two first brachials
attached, ......

Figs. 15, 16. Proximal and distal faces of a first 1)rachial,

Figs. 17, 18. Proximal and distal faces of a second lirachial.

Figs. 19, 20. Proximal and distal faces of the hypozygal portion of the
third luMchial, .....

Figs. 21, 22. Proximal and distal faces of the corresponding epizygal,

Fig. 23. Portion of an arm containing two syzygies ; side view.

Fig. 24. The arm-groove and brachial ambulacrum ; from above,

Fig. 25. Terminal faces of an internodal joint, .

Figs. 26, 27. Lower and upper faces of a nodal joint, .

Figs. 28, 29. Lower and upper faces of an infra-nodal joint,

Fig. 30. The two faces of the next joint lielow an infra-nodal, .

Diani.

Page

X

2

326

X

3

77

X

4

33

X

4

293

X

4

293

X

4

293

X

4

9

X

4

326

X

4

326

X

4

326

X

4

326

X

4

326

X

4

326

X

4

326

X

3

826

X

5

12

X

5

13

X

5

13

X

5

12

The Voyage of H.M S "Challenger."

Crmoidea PI. XXX

t

10

6

24-

^.^

29

30

19

17

■y^.

15

20

18

16

W, S.BlacV.Jtl.

PENTACRINUS NARESIANU5, S\>-

PLATE XL\a.

;ZOOL. CHALL.

EXP.— PAUT XXXII.— 1884.)— IL

PLATE XXXa.

Pentacrinus naresianus, n. sp.

Fig. 1. The youngest specimeu obtained.

Fig. 2. One of its internodal joints, ....

Fig. .J. One of its nodal joints ; upper face,

Figs. 4, 5. Two fragments, probably from the lower part of the stem
of this specimen, .....

Fig. G. Portion of a full grown stem, ....

Fig. 7. One of its internodal joint.s, ....

Fig. 8. The inner face of an arm from the third to the eighth brachials

Figs. 9, 10. Two syzygial joints from the arm of a young iudiv^idual
as seen from the side (fig. 9) and from above (fig. 10),
a. The epizygal.
h. The hypozygal.

Diam.

Page

X

2

327

X

]3

291

X

13

291

X

327

X

3

12

X

10

327

X

3

32G

X

10

327

i^igs. 11, 12. Similar views of a full grown syzygial pair.

10

32^

[U. '.'-,,-, ^^^ ,,t'lIf./f'Tl-,^il/-,;.

pixxxs-

f]

U J

PHC.dir.
PENTACRINUS NARESIANUS

"diuiart-.Titu

PLATE XXXI.

PLATE XXXI.
Pentacrinus blakei, p. H. Carpenter-

Fig. 1. An entire specimen,
Fi<y. 2. The calyx and arm-bases,

Fio-. 3. Portion of the stem,

Diam.

Page

at. size

328

X 3

330

X 8

329

The Voyage orH.M.S.Uallenger"

Cnnoidea.Pl.XXXi.

Berjeau So HigHey.deletlith .

PENTACRINUS BLAKEI, Sp.n.

PLATE XXXIL

(jiOUL. CHALL. EXP. — PAKT XXXII. — 1884.^ — li.

PLATE XXXII.

Fig.

Fif.

Fig.

Fig.

Figs.

Figs.

Fig.

Fig.

Fig.

Fig.

Fig.

Fig.

Fis.

Fi^.

Fig.

Fig.

Fig.

Fig,

Pentacrinus blakei, p. H. Carpenter.

1. The lower or syzygial face of a nodal joint,

2. The upper face of the corresponding infra-nodal,

3. Face of an internodal joint, ....

4. The two elements of a l^rachial syzygy in their relative

positions ; side view, ....

5. 6. Distal and proximal faces of the hypozygal, .

7 , 8. Proximal and distal faces of the epizygal,

9. The two syzygial elements of a distichal axillary in their
relative positions ; side view,

1 0. Dorsal aspect of the epizygal, .

11. Its distal face,

12. Its proximal face,

13. Proximal face of the hypozygal,

14. Its distal face, . . . .

15. Side view of the two outer ra dials in their relative positions,

16. Proximal face of the axil] ar}-, ....

17. Distal face of the second radial,

18. Upper view of the two outer radials in their relative positions

19. A lower pinnule, .....

20. An outer pinnule, .....

X

Diam.

7

Page
14

X

7

13

X

7

335

X

/

330

X

7

330

X

7

330

X

7

330

X

11

X

7

11

X

/

330

X

7

11

X

7

330

X

7

330

X

7

8

X

7

8

, X

/

330

X

3

328

X

3

328

,r)ru'',Vh-,

GpinoiJeaPiyXKlI

13

18

PHCdip

PENTACRINUS BLAKEl.op

Ha/iharK imp.

PLATE XXXIIT.

Diam.

Page

X 15

78

X 3

330

X 12

78

PLATE XXXIII.

Figs. 1-3. Pentacrinus blakei, P. H. Carpeuter.

Fig. 1. The ambulacral skeleton of a pinnule ; from the side,

Fig. 2. Portion of a ray, consisting of the two outer radials, the dis-
tichals, and the three lowest brachials ; from the side,

Fig. 3. Portion of a brachial ambulacrum ; from above,

Figs. 4-6. Pextacrinus decorus, Wyv. Thoms.

Fig. 4. Portion of a brachial ambulacrum ; from above, . . x 12 337

Fig. 5. Vertical section of the calyx, passing a little to one side of the

middle line; from a preparation made Dr. W. B. Carpenter, x 8 293

Fig. 6. Ventral aspect of one of the lower arm-divisions, showing the

ambulacral skeleton, . . . . . x 7 337

Figs. 7-10. Pentacrinus mollis, n. sp.

Fig. 7. The disk, . . . . . . . x 7 338

Fig. 8. The calyx ; from beneath, . . . . . x 6 338

Fig. 9. The anterior and left antero-lateral rays, . . . x 6 338

Fig. 10. The right postero-lateral ray, . . . . . x 6 338

The Voyage of HM S. "Challenger''

Crinoidea. PI. XXXIII.

Be^jea.ufcHlghle^/.del et lith .

PH.C.a,r

Kanritici .iiip.

1-3. PENTACRINUS BLAKEI, Sp.r.. 4-6. PENTACRINUS DECORUS, Wy.Th.

7-10. PENTACRINUS MOLLIS, Sp.n.

PLATE XXXIV.

(ZOOL. CHALL. EXP. — PART XXXII. 1884.) — li.

PLATE XXXIV.

Pentacrinus decorus, Wyv. Thorns.

Fig. 1. An entire specimen, .....

Fig. 2. The disk, ......

Figs. 3, 4. Proximal and distal faces of the third or axillary radial,

Figs. 5, 6. Proximal and distal faces of the second radial.

Fig. 7. The calyx ; from above, ....

Fig. 8. Inferior aspect of the calyx, ....

Fig. 9. The uppermost stem-joint, ....

Diam.

Page

. nat.

size

330

X

3

337

X

5

309

X

5

309

X

5

33

X

5

336

X

5

16

The Voyage of EMS'CliaTlengsr"

Crinoidea n XXXTV.

■W S.Bl.ck.ael

PENTACRINUS DECORUS, W^. Th.

7,fti(ii.l.lKIirf

PLATE XXXV.

PLATE XXXV.

Pentaceinus decorus, VijY. Thorns.

Diam. Page
Fig. 1. The youngest specimen obtained by the " Blake," . . x 4 21

Fig. 2. A slightly older individual, . . . . . x 4 .336

TheVoydge of H.M,S."Challenger"

Crmoidea.Pl.XXXV.

Berjoaufi;, HitjKley Jelet litli.

PENTACRINUS DECO R U S, Wy.Th.

HanKart imp-

PLATE XXXVT.

(ZOOL. CHALL. EXr. — PAKT XXXII.— ] 884. ) — 1 i

'PLATE XXXVI.

Pentacrinus decoeus, Wjy. Thorns.

Diam. Page
Fig. 1. Calyx and arm-bases of a premature individual with large

basals, . . . . . _ . x 4 336

Fig. 2. The lower jjart of the stem of the same specimen, . . x 4 335

Fig. 3. Calyx and arm-bases of another young individual with small

basals,

X 4 336

TheVoyage of H.M.S "Challenger"

Crmoidea.Pl.XXXVl.

Berjeau &. HigHey del et litK .

THC dir.

PENTACRINUS DECORU S.Wy.Th.

Hanhart imp.

PLATE XXXVil.

PLATE XXXVII.

Pentacrinus decoeus, Wyv. Thorns.

Fig. 1. Tlip calyx aucl arm-bases of a premature individual with large
basals ; interradial view, .....

Fis:. 2. Radial view of a similar specimen with no cirri on one face of
the stem, .

Fify. 3. The syzygial face of one of the distichals,

Fig. 4. The syzygial face of a brachial.

Fig. 5. A nodal joint from a young stem ; from the side,

Fig. G. The lower or syzygial face of the same joint.

Fig. 7. Side view of an infra-nodal joint from a young stem,

Fig. 8. Its upper or syzygial face, .....

Fig. 9. One of the faces of an internodal joint from a young stem, .

Fig. 10. An interpolated joint from the growing part of the stem,

Fig. 11. The upper face of a young nodal joint from the top of
the stem, .......

Fig. 12. Its lower or syzygial face, .....

Fig. 13. A young internodal joint from near the top of the stem,

Fig. 14. The upper face of a very young nodal joint witli a small inter-
polated joint resting upon it, .

Fig. 15. Upper face of a young nodal joint with an interpolated joint
lying on it, and almost concealing it,

Fig. 16. The lower face of the same joint, ....

Figs. 17, 18. The lower and upper faces of a premature nodal joint. .

Fig. 19. The upper or syzygial face of a mature infra-nodal joint,

Fig. 20. Its lower face, ......

Fig. 21. The lower or syzygial face of a. mature nodal joint, .

Fig. 22. Its upper face, ......

Figs. 23, 24. Portions of tlie ambulacral skeleton of two pinnules,

Diam.

rage

X

4

33G

X

4

335

X

10

334

X

10

334

X

10

291

X

10

13

X

10

13

X

10

13

X

10

291

X

6

291

X

7

17

X

7

17

X

t

17

16

X

7

16

X

7

17

X

17

X

6

334

X

G

334

X

334

X

7

335

X

30

337

The Voyage of HMS. 'Challenger"

Cnnoidea.Pl X>

Berjeaufcllighley oleLetlith.

PIIC dir

PENTACRINUS DECORUS.WyTh

Hanharb imp.

PLATE XXXVUI.

(iSOOL. CUAI,L. EXr.— PART XXSII. — 1884.) II.

PLATE XXXVIII.

Metaceinus angulatus, n. sp.
Natural size, p. 345.

ThpVnvafe nf HM.5, Challenger.'

Cnnoidea Pl.XXXVm.

WS Black, del

METACRINUS ANGULATUS, Spn.

vun XXXIX.

PLATE XXXIX.

Metackinus angulatus, n. sp.

Fig. 1. The calyx and arm-bases, . . . . .

The disk ; an abnormal specimen, having a second smaller anal
tube by the side of the ordinary one,

Portion of the stem, ....

5. Upper and lower faces of a nodal joint,

7. Upper and lower faces of an infra-nodal joint.

An ordinary internodal joint,

Lower face of a supra-nodal joint,

Internodal joint of a varietal form of stem, .

Side vievf of a portion of this stem, .

The ambulacral skeleton of a pinnule ; from above,

An arm-groove and brachial ambulacrum.

Fig.

2.

Fig.

3.

Figs.

4,

Figs.

6,

Fig.

8.

Fig. 9,

Fig. 10.

Fig. 11,

Fig. 12,

Fig. 13,

X

Diam.
3

Page
341

X

3

347

X

3

346

X

3

13

X

3

13

X

3

343

X

3

343

X

3

346

X

3

346

X

35

81

X

12

81

•H.M.S. "Challenger"

Crjr.

eydfiletith.

PHC du-
METACRINUS ANGULATUS, Sp n

Idiiliai t imp

PLATE XL.

(ZOOL. CUALL. EXP.— PART XXXII. —1884.)— L.

PLATE XL.

Metacrinus cingulatus, n. sp.
Natural size, p. 347.

Th.8 Voyage of HM S "Challenger,"

Crinoiaea PI XL

V/ S Black, del

METACRINUS ClNGULATUS.Spn

PLATE XLI.

PLATE XLI.

Figs. 1-4. Metaurinus cingulatus. n. sp

Fig. 1. Portion of the stem,

Fig. 2. Upper face of a nodal joint,

Fig. 3. An intei'nodal joint,

Fig. 4. Side view of a brachial ambulacrum,

Diam.

Page

X

4

:M8

X

7

348

X

7

348

X

6

81

Figs. 5-11. Metacrtnus nobilis. n. sp.

Fig. 5. Portion of the stem, . . . . .

Fig. 6. Upper face of a nodal joint, . . . .

Fig. 7. An intemodal joint, . . . . .

Fig. 8. An internodal joint of a small specimen,

Fig. 9. One of the distichal pinnules,

Fig. 10. One of the outer pinnules, . . . .

Fig. 11. The arm-groove and brachial ambulacrum,

X

3

352

X

5

351

X

5

353

X

5

353

X

3

351

X

3

351

X

12

82

Figs. 12-17. Metaceinus mureayi, n. sp.

Fig. 12. The lower part of a pinnule-ambulacrum,
Fig. 13. Side view of a brachial ambulacrum, .
Fig. 1 4. An arm-groove and brachial ambulacrum.
Fig. 15. Portion of the stem, .
Fig. 16. Upper face of a nodal joint.
Fig. 17. An internodal joint, .

X

6

82

X

3

81

X

12

81

X

2

351

X

5

351

X

5

351

The Voyage of HMS."Cha]lenger°

Crinoidea . PI . X L I .

BerjeaufcHighWy.Metllth- PH.C. dir

1-4. METACRINUS CINGU LATUS,Sp ii 5-11, M ETACR I N US NO Bl LIS,Sp n

12-17, METACRINUS M U RRAYI, Spn.

Hanharb jmp.

PLATE XLII.

(ZOOU CnALL. ESP.— PART XXXII. — 1884.) -Ti.

PLATE XLII.

Metaceinus murrayi, n. sp.

Fig. 1. An entire specimen, ....
Fig. 2. A pinnule from some way out on the free arm,
Fig. 3. A pinnule from tlae fourth arm-division,

Diam.

Page

at. size

349

X 3

351

X 3

351

TheVoyage of H.M.S . 'Challenger.'

Crinoidea.PLXLII.

^5^

Serjeau&HigHey cleLetlith.

PH.C. dir

METACRINUS M U R R AY I, Sp.n.

Hanhart imp.

PLATE XLIII.

PLATE XLTII.

Metacrinus nobilis, n. sp.

Fig. 1. The stem,

Fig. 2. The calyx and arm-bases,

Fig. 3. The disk,

Fig. 4. Portion of a rav,

X

Diam.

2

Page
352

X

2

341

X

2

342

X

2

341

The Voyage of HM S"Challen|er"

Crinoidea PlXLIIl.

hr:^' ;^^

l^-.V;

^;

\ku»''ii UC. ,■

(r.:L

Vf S.Bl«*,liil-

METACRINUS NOBILIS, Spn.

?H.lk. Lift" Sic'

PLATE XLIV.

(ZOOL. CUALL. EXP.-PAKT XXXII.— 188i.)-I'-

PLATE XLIV.
Metacrinus varians, n. sp.

Diam. Page
Fig. 1. The stem, . . . . . . . x 2 342

Fig. 2. The calyx and arms, . . , . . . x 2 35.3

The Voyage of H M ? "Challenger"

CrinoiSea/Pl. XI.IY

V/ S Black de!

? )iul)>,ljlk' Sjin'

METACRINUS VARIANS, Sp n

PLATE XLV.

PLATE XLV.

Metaorinus moseleyi, n. sp.

Fig. 1. A young individual, ....
Fig. 2. An internodal joint, ....
Fig. 3. Tile upper or syzygial face of au infra-nodal joint,
Figs. 4, 5. Upper and lower faces of a nodal joint,
Fig. 6. Portion of the stem, ....
Fig. 7. Portion of an arm, with cummencing Myzosfoma-cyst,

Diain.

Page

nat.

size

356

X

12

357

X

]2

357

X

12

357

X

4

357

X

3

358

rhr^/- -.'- 'f'fiM.S.thallen^-er

Cr]noidea.Pl.>:i,Y

Berieau.'

PH.Ciir.
METACRINU5 MOSELEYI. 6p.n.

Ha^ihat't imp

PLATE XLVI.

(ZOOL.

CHALU EXP. — PART XXXII. — 1884.)— Il

PLATE XLVI.

Metaceinus moseleyi, u. sp.
Diameter 2^ (circa), ^j. 357.

The Voya-^e '"';

CrinoideaP: X:"'!

w S BUcl: J-l

METACRINUS MOSELEYI, Spn.

FBua.l'WEBo'

PLATE XLVIL

PLATE XLVII.

Figs. 1-5. Metacrinus wyvillii, n. sp.

Figs. 1,2. Portions of the stem, ....

Figs. 3, 4. Upper and lower faces of an infra-nodal joint,
Fig. 5. Upper face of a nodal joint, ....

Figs. 6-9. Metacrinus varians, n. sp.

Fig. 6. Portion of the stem, commencing with an infra-nodal joint
and inverted, .....

Figs. 7, 8. Lower and upper faces of a supra-nodal joint.

Fig. 9. Upper face of a nodal joint, ....

Fig. 10. The end of a pinnule; from above, .

Fig. 11. The lower part of a pinnule ; from the side, .

Fig. 12. The arm-groove and brachial ambulacrum,

Fig. 13. Metacrinus costatus, n. sp.
Fig. 13. The lower part of the ambulacrum of a middle pinnule,

Diam.

Page

X

4

359

X

9

13

X

9

13

X

3

355

X

8

343

X

8

13

X

14

55

X

15

82

X

10

82

10

364

Thc\bya.!5fM-^'''V -:Tt,.!!...,-V

^M

1-5 METACRINUS WYVILLIl. :'.v.u 6-12 METACRINUS VARIANS.

13 METACRINU5 COSTATUS, :>? ii.

PLATE XLYIII.

(ZOOL. CHALL. EXP. — PAKT XXXII. 1884. )^Ii.

PLATE XLVril.
Metacrinus ^vyvillii, n. sp.

TP- mi Diara. Page

tig. 1. The calyx and arms, . . - . x l^ (circa) 358

Fig. 2. The stem, ^ Ij (circa) 342

Tie Voyage of HM.S "Challenger"

Cnnoiaei Fl XLVIII.

METACRINUS WYVILLII, Spn.

f H»a,LrftfE*.T

PLATE XLTX.

PLATE XLIX.

Metacrinus costatcts, n. sp.

Fig. 1. An entire specimen,

Fig. 2. The calyx of an individual in which basals arc absent,

Fig. 3. Portion of the stem, ....

Fig. 4. An iuternodal joint, ....

Fig. 5. Upper face of a nodal joint,

Fig. 6. The lower portion of one of the outer pinnules ; from the side,

Fig. 7. The ambulacral skeleton of one of the lower pinnules; from
above, .......

Diam.

Page

nat.

size

360

X

2

363

X

5

362

X

4

362

X

4

362

X

15

364

15

364

T>ie Voyage of HM.S'thaJlenger

rr:r-;'l'>a ^]^' 1X

BerjeauiiLHighley del el h*h.

P H C ir

METACRINUS COSTATUS, Sp.n.

Hajih«ri imp.

PLATE L.

(ZOUL. CUALL. EXP. I'Alt'l' X.XXII. ISS-t.) li.

PLATE L.

Fig.

Fig.
Fig.
Fig.

Fig.

MeTACKINUS NODOSUS, 11. sp.

1. The calyx and arm-bases, ....

2. Oblicpe view of the disk from its anterior end,
.3. Portion of the stem, ....

4. Nodal and infra-nodal joint of the stem ; from the side,

5. The first radials, with the hypozygals of the second, as seen 1

from beneath after removal of the basals.

Fig. 6. The same ; from alcove.

Figs,

Figs.

Figs.

Figs.

Figs.

Figs.

Figs.
Fia:s.

7, 8. Proximal and distal faces of the epizygal of the second

J/ciCLlcl'l J* ■ • • • • •

9, 10. Proximal and distal faces of the third radial,

11, 12. Proximal and distal faces of the h5rpozygal of the fourth

radial, .......

I ^
1.3, 14. Proximal and distal faces of the epizygal of the fourth I 5

radial, ......

15, 16. Proximal and distal faces of the fifth radial,
17, 18. Proximal and distal faces of the sixth or axillnry radial, J
19, 20. Lower and upper faces of an infra-nodal joint, . . x

21, 22. Lower and upper faces of a nodal joint, . . x

Diam. Page

X

2 364

X

3 (circa) 81

X

2 364

X

4 366

1

293

51

X

51
51

CO

51

51
51

51

10

13

13

The Vo)ra^e of H M SThallenger"

Crmoidea PI L

«~'

]3

14

12

10

W S BUclc ael.

y Hnth.Lift' Edm'

METACRINUS NODOSUS, Spn

PLATE LL

PLATE LI.

Metacrinus nodosds, u. sp.

Fig. 1. A young specimen, . . . . ■

Figs. 2-4. Faces of young internodal joints, .

Fig. 5. The upper face of a young nodal joint,

Figs. 6, 7. Radial and interradial views of a young nodal joint,

Fig. 8. Portion of the stem, bearing a young coral, .

Fig. 9. Upper face of a nodal joint, . . . •

Fig. 10. Face of an internodal joint,

Fig. 11. Side view of the middle portion of a pinnule,

Fig. 12. The lower part of a pinnule-ambulacrum ; from above,

Diam.

Page

nat,

. size

3G6

X

14

291

X

14

291

X

14

366

X

6

367

X

10

13

X

10

368

X

15

366

X

15

366

TheVoyage of HM.S "Challenger.'

£>erjeau-fcni£mey del el liin

METACRINUS NODOSUS, Spn.

PLATE LII.

(ZOOL. CHALL. EXP. — PART XXXII. — 1884.) li.

PLATE LII.

Metacrinus interruptus, n. sp.

Diam. Page
Fig. ]. An entire specimen, ...... nat. size 367

Fig. 2. Portion of the stem, . . . . . . x 4 343

Fig. 3. Face of an internodal joint, . . . . . x 9 368

A\B. — The sides of this joint are some\Yliat more hollowed than is usually the case.

Fig. 4. Upper face of a nodal joint, . . . . . x 9 368

Fig. 5. The termination of the ambulacrum on one of the small outer

pinnules ; from the side, . . . . . x 1 0 55

Fig. 6. Upper view of the ambulacral skeleton nearer the base of the

pinnule, . . . . , . . x 10 367

TheVoyag-e cf H.M.S."Challeng-er.'

Crinoidea. PI LII.

Berjeau & Highley del.et lith .

PHC dir.

METACRINUS 1 NTE RRUPTUS, Sp.n.

Hanhart

PLATE LTII.

PLATE LIII.

Figs. 1-0. Metagrinus tuberosus, n. sp.

Fig. 1. Portion of the stem, .....

Fig. 2. Lower face of a supra-nodal joint,

Fig. 3. Upper face of a nodal joint, ....

Fig. 4. An ordinary internodal joint, . . ...

Fig. 5. Lower face of an infra-nodal joint.

Fig. 6. Side view of a nodal joint and the two upper joints of the in-
ternode below it, .

X

Diam.
2

Page
369

X

9

370

X

9

369

X

9

369

X

9

370

369

Figs. 7, 8. Rhizoceinus rawsoni, Pourtales.

Fig. 7. A young specimen with the arm-bases slightly displaced at the

syzygy between the first and second brachials, . . x 20

Fig. 8. The upper part of the stem, with the calyx and first l^rachials

of a slightly older individual, . . . . x 20

265

266

I'hoVojrage ot

■[VI T I T I

Berjeau Sc HijS'jiIev del.el.Jith .

P.HC.cLi:

1-6. METACRINUS TUBEROSUS,Spn.
7, 8. RHIZOCRINUS RAWSONI, PourlaJes .

PLATE LIV.

(ZOOL CUALL. EXr. PART XXXII. — 1684.) — li.

PLATE LIV.

Figs. 1-4. Antedon accela, n. sp.

Diam, Page
Fig. 1. Ventral aspect of a portion of an arm from its lower third, . x 7 83

Fig. 2. Side view of one of the genital pinnules, . . . x 10 83

Fig. 3. The ventral surface of another genital pinnule, . . x 10 83

Fig. 4. The ambulacral skeleton of a later pinnule with the covering

plates closed down, . . . . . x 20 83

Fig. 5. Antedon anCtUSTICAlyx, n. sp.
Fig. 5. The ventral aspect of an ungrooved genital pinnule, . . x 13 83

Figs. 6, 7. Antedon incerta, n. sp.

Fig. 6. Side view of a genital pinnule with its ambulacrum, . . x 13 83

Fig. 7. The ambulacral skeleton of an arm and the lower parts of the

pinnules, . . . . . . . x 10 83

Fig. 8. Antedon in^qualis, n. sp.
Fig. 8. The ambulacral skeleton of a pinnule ; from the side, . . x 20 83

Fig. 9. Antedon basicurva., n. sp.

Fig. 9. Side view of the ambulacral skeleton of a pinnule mounted in

dammar, . . . . . . . x 30 83

Figs. 10, 11. Actinometra strota, n. sp.

Fig. 10. The peristome of a dry disk, with the origins of the ambu-
lacra, . . . . . . . X 3 85

Fig. 11. Vertical section of an ambulacrum of a dry disk, . . x 12 85

TheToyage of H,M.S."ChcLllenger"

Crinoidea.Pl.LIV.

;eauS:,ili^li]ey fiel si lilii

PHc a.r

1-4 ANTEDON ACOELA, 5 ANT ANGUSTICALYX ^ 6.7 ANT INCERTA.
8 ANT. INAEQUALIS 9 ANT. BASlCURVA lO.II.ACTINOMETRA STROTA.

Hanhar^ imp

PLATE LV.

PLATE LV.
Disks of Comatul^.

Diam. Page

Fig. 1 Disk oi Actiiiometra jukesi, n. ni^., . . . . x 2 85

A small Isopod is visible within the anal tube (see j). 133).

Fig. 2. Disk oi Actinometra strota, n. s])., . . ■ . x 2 85

One of the posterior arms is replaced by two pinnules (see p. 60).

Fio-s. 3, 4. Disks oi Antedon multiradiata, n. sp.,

Fio-. 5. Disk of Antedon accela, u. sp., ....

Fig. 6. Disk of Antedon angusticcdyx, n. sp., .

Fig. 7. Diiik o£ Antedon hasicurva,n. s])., ...

X

4

84

X

10

84

X

10

84

X

10

84

.-'imlleiigei'

Cnnoidea PI LV.

'^mJ^

B erjeauJb Higiley del . et lith .

PHCdir.

1. ACTINOMETRA JUKESI,Sp.n. 2. ACTI NOMETRA STROTA,^.p.ii

3,4. ANTEDON M ULTI RADIATA, Sp n 5. ANTEDON ACOE LA ,Sp n.

e.ANTEDON ANGUSTICALYX,Sp.n. 7.ANTED0N BASICURVA.Sp.n.

PLATE LVI,

(ZOOL. CHALL. EXP. — PART XXXII. — 1884.) — II

X

15

371

X

15

371

X

15

371

X

15

372

PLATE LVI.

Figs. 1-5. Thaumatocrinus renovatus, P. H. Carpenter.

Diam. Page
Figs. 1, 2. Side views, radial. In fig. 1 the right anterior, and in
fig. 2 the right posterior ray faces the observer,

Fig. 3. Side view, interradial, ....

Fig. 4. The anal side, .....

Fig. 5. The disk, ......

Fig. 6. Antedon rosacea, Linek, sp.
Fig. 6. An abnormal disk with two mouths. . . . . x 5 70

Fig. 7. Actinometra magnifica, n. sp.

Fig. 7. The disk of the only specimen known, . . . uat. size 56

The course of the water-vessel which supplies the arms of the posterior ray is marked by a tine
lino upon the surface of the disk, but it is not accompanied by any food-groove.

Fig. 8. Actinometra stelligera, n. sp.
Fig. 8. An abnormal disk with two mouths and two anal tubes, . x 5 70

.^. ..•'*'■

BerjeauJoEigkley del at liUi

rilC ,(;r

1-5. THAUMATOCRINUS R ENOVATUS, i3p ,n 6.ANTED0N ROSAC E A, ...u.^, . t-
7. ACTINOMETRA M AG N I Fl C A , Sp.n . 8 . ACTI N 0 M ETRA STELLI GE RA.Sp.n.

PLATE LVIJ.

PLATE LVIL

The Lettering is the same in all the Figures.

(ic. Ambulacral epithelium.
(iH. Aiiainbulacral jilate.
I>. KaJial blood-vessel.

p. Pigment granules.
re. Rectum.

Sinnelets on the disk.

Connective tissue lueslnvork in the coelom. I up. Side plate.

fp. Covering plate.

cvc. Circumvisecral eoeloni.

r'g. Koregut.

(J. Epithelial lining of the gut.

!iv. Genital ve.ssi'l.

lb. Intervisceral blood-vessel.

/p. Labial plexus.

ir. Its radial extensions beneath the ambulacra

II. Ambulacral nerve.

sub. Subanibnlacral plate.

T. Tentacle.

III. Tentacular branch of water- vessel.

If. Radial water-vessel.

wp. Water-jjore.

let. Water-tube.

X. Plexiform gland.

./ r. Its ventral end which joins the labial plexns.

Fig. 1. Pentacrinus wyville-thomsoni, Jefireys.

Diam. Page

Fig. 1. Transverse section of the disk in the anal interradius, showing
the labial plexus (Ip) and its extensions beneath tlie am-
bulacra (//■), . . . . . . X 40 100

Figs. 2 -.5. Pentacrinus decorus, Wyv. Thorns.

Fig. 2. Vertical section through the outer j^art of the plexiform gland,

showing its connection with the intervisceral blood-vessels, x 100 101

Fig. o. The upper part of a longitudinal vertical section through the
disk, showing the connection of the labial plexus (?p) with
the genital and intervisceral vessels (gv, ih), and also
with the ventral end of the plexiform gland {.rv). The
section cuts the edge of the anal tube (re), but passes to
the right of the mouth, . . . . . x 30 97

Fig. 4. Transverse section of an amlmlaerum near the peristome.
This shows the two lateral water-vessels (tv), which are
separated by the radial extension of tlie labial plexus {Ir), x 50 93

Fig. 5. Portion of a vertical section through the periphery of the disk,
showing the intervisceral vessels on the outside of the
digestive canal, . . . . . , x 100 102

Tke Voyage of H.M.S. "Qiallenger."

Crinoidea.Pl. LVn.

x^ri"

^'■

...,;^|;...n.

'.'?''

— ' ':■/

ib

¥>

wt

Ip

tb

sp-

Q^'

a\ f"

'y :#^

;l'ii|;./;v-i

.'/>'

^3

??,,"' . ;':^'

3.

"UVp

SPs

.<«W^.v

tb

JW^

i(>

n( Ir

IP

•^

G«o Wirat. JtSona <I*1- l^Oi.vti:

I PENTACRINUS WYVl LLE - T H O MS O N I

2-5. PENTACRINUS DECORUS,

J effreys .
Wyville Tliom.sorL.

PLATE LVm.

(ZOOL. CUALL. EXP. — PART XXXII. 1884). — li.

PLATE LVIII.

The Lettering is the same in all the Figures.

ai. Primary interradial cords.

ar. The secondary (radial) cords.

B. Basal plate.

c. Bands of connective tissue in the body-cavity.

ch. Cavities of the chanibereil organ.

clt'. Their downward prolongations into the stem.

c. External epithelium.

/(/. Fore -gut.

G. Gut.

('6. Intei'visceral blood-vessel.

L. Interradial ligament.

I. Basiradial ligament.

Ih. Intcrbasal ligament.

III. Limestone network in the perisome.

Is. Interarticular ligament of stem.

}). Pigment granules.

R'. First radial.

re. Kectuni.

rp. Plug of calcareous tissue within the radial funnel.

V, Central vascular axis of stem.

X. Plexiform aland.

Figs. 1-3. Pentacrinus decorus, Wyv. Thorns.

Fig. 1. Horizontal section through the lower part of the calyx, showing
the bifurcation of the primary interradial cords {ai) in
the basals, ......

Fig. 2. Horizontal section through the lower part of the radial
pentagon, showing the upward extensions of the cavities
of the chambered organ [ch), ....

Fig. 3. The central part of a median vertical section through the calyx,
showing the chambered organ and its connection with the
vascular axis of the stem, ....

Diam. Page

40

40

40

124

105

104

Fig. 4. Pentacrinus naresianus, n. sp.

Fig. 4. Portion of a horizontal section of the disk, showing the position
of the plexiform gland {X) between the fore-gut and the
rectum. . . . . . . . x 30

101

Figs. 5, G. Promachocrinus kergueleksis, n. sp.

Fig. 5. Portion of a horizontal section of the disk, showing the plexi-
form gland (A'), . . . . . . x 35

Fig. G. A similar section nearer the calyx,

35

101
101

The Vo3raLge of H. M.S." CI

Crmoidea.Pl.LYIlL

*• .•■•■];.■■-■.■ ■-<...

- .' *-* •! -■' V.'-; '^^.'•'i ' '(■'••'W-'-'-:/''-''-' •','•'

'.} ;:;>:.:^*r //■;-;> ell.

^•Iss^

,, ch

"■-lb

•:«;

'-K'

,-^ ■ , >, ^^^* •« t. •« »1 *B t

^^^••

-■*:^;

-'^'vi'^''

\^

■ ■ ..

%\ m

'^^'■IMy^r

\4it

1r

6.

4.

Ber|eau SLHilhley del et lilK

1-3. PEN'

P H e dir-

5 DECORUS,'%v Th. 4. PENTACRINUS NARESIANUS, Sp n.

5,6. PROMACHOCRINUS KERGU ELENSIS.Sp.Tx,

lointem. Sros . ib^.

PLATE LIX.

PLATE LIX.

The Letteriii" is the same in all the Figures.

ac. Connective tissue fibres between the ananibulacral plates.

71.

Anibulacral nerve.

ad. The paranibulacral extensions of tlie axial cords into the

P-

Pigment granules.

ventral perisome.

s.

Spinelets on the ananibulacral plates of the disk

ac. Anibulacral epithelium.

sa.

Sacculi.

b. Radial blood-vessel.

so.

Spongy organ.

c. Bands of connective tissue in the body -cavity.

SjJ.

Side plate.

cp. Covering plate.

sk:

Subteutacular canal

ci'C. Circumvisceral coelom.

T.

Tentacle.

c. External epithelium.

tb.

Tentacular branch of water-vessel.

fg. Fore-gut.

w.

Radial water-vessel.

a. Gut.

wp.

Water-pore.

g. Its epithelial lining.

wl.

Water-tube.

gv. Genital vessel.

X.

Plexiforra gland.

Ij). Labial ple.xus.

XV.

Its ventral end which joius the labial plexus.

Figs. 1-4. Pentaceinus decorus, Wyv. Thorns.

Fig. 1. Transverse section of a brachial ambuLacrum,

Figs. 2-4. Vertical sections of some of the anibulacral plates of the
disk, which bears spinelets and are traversed Ijy fibres of
the parambulacral nerve-plexus {ad),

Diam.
100

50

Page
96

124

Fig. 5. Ante DON rosacea, Linck, sp.
Fio'. 5. Vertical section through the left or eastern ano;le of the

o O O

peristome, at the origin of the two lateral ambulacra,
showing the labial plexus (Ijj) and water-tubes (ai),

40

98

Figs. 6, 7. Antedon eschrichti, MiilL, sp.

Figs. G, 7. Portions of vertical sections of the disk in the neigh-
bourhood of an ambulacrum, showing the parambulacral
nerves (ad), ......

100

123

Figs. 8, 9. Promachocrinus kerguelensis, n. sp.

Fig. 8. Portion of a vertical transverse section of the disk behind the
mouth, showing the spongy organ (so) at the side of the
fore-gut (/(/), . . . . . . X 40 99

Fig. 9. Portion of another section of the same series, but in front of
the mouth, showing the upper end of the plexiform gland
(x), part of which is becoming reticular and spongy (xv), x 30 99

Tke Voya|e of H.M. S . " dialUnger . "

Crinoidea.Pl.UX.

2.

^3?*^^ ■

tp ...

^9

vW'WW'''

-jC

y-

ad'

OLOi

/3

^1

V*'

.-vtti:i,

•;^

« V , -«;^

3^

»*^w»«C^''

;^^

^ "^^

\

a*«>:%%»t A Son* <tal klK.at u»p

1_4, PENTACRINUS DECORUS, TA^ille Thomson. 5. ANTEDON ROSACEA, Linck, sp.
6,7. ANTEDON ESCHRlCHTl, Mull., sp. 8, 9. PROMACHOCRINUS KERGUELENSIS, Sp.n.

PLATE LX.

(ZOOL. CHALL. EXP. — PART XXXIT. — 1884.) — li.

PLATE LX.

The Lettering is the same in

all the Figures.

a'.

Vcutral branch of the axial coid of the arm. |

gi: Genital vessel.

lUl.

The panimbulacral extensions of the axial-cords into the

ib. Intervisceral hlood-vesse

ventral perisonie of the disk.

Ip. Labial plexus.

ac.

Ambulacral epithelium.

11. Ambulacral nerve.

b.

Radial blood-vessel.

nr. Its oral ring.

c.

Bands of connective tissue in the bodj'-cavity.

p. Pigment granules.

tv.

Coeliac canal.

sa, Sacculi.

cvc.

Circumvisceral coelom.

so. Spongy organ.

c.

External epithelium.

stc. Subtentacular canal.

fa-

Fore-gut.

tb. Tentacular braucli of water-vessel

il-

EpitheKum of the gut.

10. Radial water-vessel.

?<•••

Genital cord.

itij). "Water-pore.

</c'.

Genital canal.

■wt. Water-tube.

Fig. 1. AcTiNOMETEA PULCHELLA, Pourtales, sp.

Fig. 1. Transverse section of an ambulacrum behind the mouth, show-
ing the connection of the genital vessels [gv) and labial
plexus {Ip), . . . . . .X

Diam. Patre

50

98

Fig. 2. Antedon carinata, Leach, sp.

Fig. 2. Portion of a vertical transverse section of the disk in the anal
interradius, showing the connection of the genital vessels
(gv) and labial plexus {Ij)), . . . . x

50

9«

Figs. 3-6. Antedon eschrichti, MlilL, sp.

Fig. 3. Portion of a vertical longitudinal section of the disk, which
passes through the peristome, showing the connection of
the intervisceral vessels [ilj) with the spongy organ {so), . x 35 98

Fig. 4. Longitudinal vertical section of the peristome in the direction

of a ray, . . . . . . . x 60 97

Fig. 5. Part of a vertical transverse section of the disk in the anal in-
terradius, showing the spongy organ and labial plexus, . x 40 98

Fig. 6. Transverse section of a brachial ambulacrum, showing one of
the ventral branches of the axial cord {af which extend
into the perisome at the side of the food-groove, . . x 65 122

* This is ■wrongly lettered ad.

TheVoyafp of K.M.S." Challenger.'

Cpinoidea.PL IjX.

0L«

■■■.../

tb

tXCi

tb

Ip

9"

Lp

.H^a"

-■J^-^

ih-

,/ ^<3

it

Ip

1^ ^V, ft:--

ib

/?

«. s-

r

.. fl;^ i^-^

ifc

P H.C. dix

.^.

,as?

w w.

K C V-

r p „ „ ,,,

lj:

oW««t.A3aa« d«l. litlv.at.uBB

I. ACTINOMETRA PULCHELLA, Pour^,.,sp

2. ANTEDON CARINATA, Leich sp
3-6. ANTEDON ESCHRICHTI, Miill.. sp

PLATE LXL

Diam. Pane

PLATE LXI.

Fig. 1. ACTINOMETRA PULCHELLA, Pourtulfes, Sp.

Fiof. 1. A^ertical lono-itudinal section of an isolated disk. The section
passes tLrougli the edge of the mouth, which is seen just
in front of the large anal tube. The lobular structure in
the centre of the lower part of the disk beneath the fore-
gut is the plexiform gland, . . . . x 10 103

Figs. 2-5. ACTIKOMETRA PARVICIREA, Mlill., sp.

Fio;. 2. Vertical longitudinal section of the calvx and disk. The situa-
tion of the chambered organ between the centro-dorsal and
radials is well shown ; as are also the marginal position of
the mouth and the numerous coils of the digestive tube,
from some of which the epithelial wall has fallen away.
The labial plexiis is relatively large in the upper lip {i.e.,
behind the mouth), but is inconspicuous in front of it, . x 7 103

Fis. 3. Transverse section of an unorooved or non-tentaculiferous arm,
close to the base of a pinnule. The ventral surface is
flat, without any food-groove or tentacular apparatus, and
owing to the backward extensions of the ovaries below
their points of attachment to the genital cord, one appears
in section upon each side of the middle line (p. 110), . x 20 113

Fig. 4. The upper part of a transverse section of a normal tentaculi-
ferous arm at the articulation of two joints, showing the
connection of the ovary and genital cord. In the coeliac
canal is a section of a parasitic worm, . . . x 25 133

Fig. 5. Transverse section of the lower part of an arm just beyond the
attachment of a pinnule. On the left side are seen three
funnel-shaped water-pores, the inner ends of which lead
into the genital canal containing the triradiate genital
cord, X 27 9G

Fig. 6. ACTINOMETRA NIGRA, r. sp.

Fig. 6. Transverse section of an arm through the middle of a joint,
showing the branches of its axial cord. The genital canal
is occupied by a relatively large genital cord, and the
radial blood-vessel is well shown beneath the middle line
of the food-groove, between the ambulacral epithelium
and the water-vessel, . . . . . x 30 121

The Voyage of H.M.S. ''Challenger."

Crinoidea. Pl. LXI.

^.Aii.^"

P. H. C. DIR.

AUTOTYPE

1. ACTINOMETRA PULCHELLA. PouRT., Sp. 2-5. ACTINOMETRA PARVICIRRA, Mull.. Sp

6. ACTINOMETRA NIGRA, Sp. N.

PLATE LXII.

(ZOOL. CHALL. EXP. — PART XXXII. — 1884.)— li.

PLATE LXIL

Explanation of the Letters.

A.

a',
ad,

ae.
ai.
an.
B.

b.
c.

ceo.

ch.

ch'.

ohn.

cv.

D,, D^.

fg-

G.

9-
gv.
ib.

Axial cord of the ray.

Its branches in the dorsal skeleton.

Its upward extensions into the plated perisome of

the ventral side.
Ambulacral epithelium.
Primary interradial cords.
Anambulacral plate.
Basal.

Radial blood-vessel.
The bands of more or less calcified connective tissue in

the body-cavity.
Fibrillar sheath round the vascular axis of tlie stem.
Interradial portion of the circular commissure.
Cavities of the chambered organ.
Their downward prolongations into the stem.
The nodal enlargements of these peripheral vessels.
Cirrus-vessel.

Second and fifth distichals.
Fore-gut.
Gut.

Its epithelial lining.
Genital vessel,
lutervisceral blood-vessel.

I. Basiradial ligament.

Id. Dorsal ligament between the ray-joints.

li. luterarticular ligament.

Ip. Labial plexus.

M. Mouth.

in. Muscle.

n. Ambulacral nerve.

nr. Its oral ring.

oh. Oral blood-vascular ring,

ill, -^3- First and third radials. ■

Tc. Rectum.

rp. Plug of calcareous tissue within tlie radial funnel.

s. Spinelets on the anambulacral plates.

stc. Subtentacular canal.

svh. Subambulacral plate.

sij. Syzygy.

V. Central vascular axis of stem.

jr. Kadial water-vessel.

U'p. AVater-pore.

wr. Water-vascular ring.

lot. Water-tube.

.r. riexiform gland.

XV. Its ventral end which joins the labial plexus

Explanation of the Colours.

Black — The nervous system, both ambulacral and autambulacral.

Green — The water-vascular system.

Red — Tlie blood-vascular system and the plexiform gland.

Diagrammatic longitudinal section of a Pentacrinus decorus. On the right side of the
figure the section cuts the anal tube, and on the left side it passes along a ray
and the base of an arm.

The Voyage of H.MS" Challenger."

Cnnoidea. . PI . LXu.

BerjeAU jc HighW dfu ei lita.

? H C dir,

PENTACRINUS DE CO RUS ."^ville Thoinson.

Miru.err- Er(

THE

VOYAGE OF H.M.S. CHALLENGER,

ZOOLOCtY.

KEPORT on the Isopoda collected by H.M.S. Challenger during the Years
1873-76. By Frank Evers Beddard, M.A. (Oxon.), F.R.S.E.,
F.R.M.S., F.Z.S., M.B.O.U., Prosector to the Zoological Society of
London.

PART I.-THE GENUS SEROLIS.

PREFACE.

Among the Lsopoaa collected during the ChaUenger Expedition the genus Serolis
occupies a very prominent position both as regards number of specimens and of new
species; for this renson I have thought it advisable to separate Serolis from the rest of
the group, which will be treated of in the second part of my Report.

As many as sixteen different species were collected during the voyage, of which nine
are new. One of these new species was named Serolis hromlei/aaa, and briefly described
by the late Dr. von Willemoes Suhm in his Preliminary Report on Crustacea observed
during the voyage of H.M.S. Challenger ;' the remaining eight have already been briefly
noticed by myself.^

Into the systematic part of this Report I have also introduced some account of the
remaining species of the genus that were nut obtained by tlie Challenger in order to make
the Report more complete.

The portion which deals with the internal anatomy of Serolis is unfortunately very
incomj)lete ; the specimens were not very favourably preserved for anatomical investiga-
tion, auil 1 dill lint fri'l justified in using a great amount of material.

" Proc. Roy. Soc. Land., vol. xxiv. p. 585, 1876. - Proc. Zool. Soc. Land., pt. iii. p. .330, 1884.

(zooL. CH.vi.r,. Exr. — p.vrt. .xsxiii. — 1884.) Kk 1

2 THE VOYAGE OF H.M.S. CHALLENGER.

A discussion of the systematic position of tbe genus Scrolls within the order Isopoda
will be best postponed until after a more detailed study of the remaining part of the
collection.

With regard to the alleged affinity of Serolis (and of the Isopoda generally) for the
extinct Trilobites, insisted upon by Milne-Edwards,^ I have nothing to add to what has
already been said ; the examination of the Challenger collection of Serolis has brought to
light no facts which tend to show any close resemblances between the two groups.

I have to thank Mr. E. J. Miers of the British Museum for kindly facilitating my study
of the specimens of Serolis preserved in the national collection.

HISTORICAL NOTICE.

The first recorded notice of Serolis is contained in Fabricius's Systema Ento-
mologiae, which was published in 1775. Under the name of Oniscus 'paradoxus is a
short description of a species subsequently named Serolis fabricii by Leach, and which
was obtained at the Straits of Magellan during Captain Cook's second voyage.

A few years later (1767) Fabricius briefly defined this species in his Mantissa
Inseetorum ; in both these works Oniscus is placed in the class Synistata, which forms
the third class of the four into which Fabricius divided what are known now as Arthro-
poda, and which included, besides the Isopoda, the majority of the Insecta now classed
within the orders Hymenoj^tera, Diptera, and Heteroptera.

In his Species Inseetorum, published in 1781, Fabricius gave another description
of this species, still under the name of Oniscus i^'^f'^doxus ; here Fabricius hints at the
Trilobite affinities of the genus " An protypon Entomolithi paradox! ? In multis
certe convenit."

In 1798 appeared Fabricius's Entomologia Systematica, which is a considerable advance
upon his earlier works. His eighth class, Polygonata, includes all the Crustaceans belong-
ing to Latreille's order Isopoda as well as the genus Monoculus;^ the name Oniscus
paradoxus is altered to Cymothoa paradoxa, and a fuller definition of the genus is given.

Fabricius's Cymothoa paradoxa was first recognised as the type of a new genus by the
English naturalist Dr. W. E. Leach ; in the twelfth volume of the Dictionnaire des
Sciences Naturelles, published in the year 1818, is an article by Leach on the Cymothoadae,
where CymotJioa paradoxa is redescribed under the name of Serolis fabricii.

In 1825 Desmarest in his Considerations generales sur les Crustaces recapitulated
Leach's description of Serolis fabricii. Both Desmarest and Leach denied the supposed
affinities of Serolis with the Trilobites.

' In 1833 James Eights described in the Transactions of the Albany Institute, vol. ii.

' Arch. d. Mus., t ii. p. 5 ; Ann. d. Sci. Nat., ser. 6, t. xiL art. No. 3.
^ Monoculus includes aE tlie members of the order Entomostraca.

REPORT ON THE ISOPODA. 8

p. 53, pis. i. and ii., another species of Serolis under the name of Brongniartia trilo-
hitoides, which name was subsequently altered by Audouin and Milne-Edwards into
Serolis trilobitoides, since there was evidently no generic difference from Serolis para-
doxa ; this paper contains merely a description of the new form, accompanied by figures
in which the whole animal as well as the mouth appendages and the two first abdominal
limbs are displayed ; a figure of a Trilobite, Paradoxus holtoni, is given for the purpose of
comparison, but there is no account in the text of any supposed resemblances between
the two forms, except the similarity of the eyes. It seems not unlikely that this species
is identical with Studer's Serolis cornuta, or at most a local variety. This species was
obtained on the coast of Patagonia from the stomach of a fish belonging to the genus
Phycis, and also from the South Shetlands.

The next contribution is a paper by Audouin and Milne-Edwards, which was pub-
lished in the Archives du Museum for 1841. This important memoir on the genus
Serolis contains, after an historical introduction in which all the previous notices regard-
ing the genus are reviewed, a general account of its external characters, and a description
of four species — Serolis paradoxa, Serolis trilobitoides, Serolis orhignyi, and Sei'olis
gaudichaudii , the two last being here described for the first time ; the memoir is illus-
trated by figures of these species and of all the essential parts in their anatomy ; with
regard to the afiiuities of the genus, it is considered as intermediate between the other
Isopoda and the Trilobites. Several of these figures are, repeated in the magnificent
illustrated edition of Cuvier's Eegne Animale, edited by his disciples Audouin, Milne-
Edwards, Latreille, &c., though the text of this work appears only to contain a descrip-
tion of Serolis paradoxa.

The Histoire Naturelle des Crustaces of Milne-Edwards, published in 1840, contains
a description of the genus Serolis and of the four species Serolis fabricii ( = Serolis para-
doxa), Serolis hrongniartiana { = Serolis trilobitoides), Serolis orbigniana, and Serolis
gaudichaudii; in this work Serolis is included in the family Cymothoadse, which is
divided into three tribes — (l) CjTnothoadiens ravisseurs, Serolis; (2) Cymothoadiens
errans, ^ga, Conilera, &c. ; (3) Cymothoadiens parasites, Cymothoa, Nerocila, &c.

In 1852 Dana described a new species of Serolis {Serolis planus) in the Crustacea of
the United States Exploring Expedition.

Six years later (1858) another species was described by Liitken, to which he gave the
name of Serolis schythei.

The next and a most important contribution to our knowledge of Serolis is Gnibe's
Monograph of the genus, published in 1875. Besides the description of a new species
{Serolis tuberculata), a general resume of the external characters of the genus, and some
notes on certain species {Serolis paradoxa and Serolis schythei) that had been previously
described, are to be found in this paper, together with a complete list of the then known
species, eight in number. Grube regards Serolis as being most nearly aUied to Sphceroma.

4 THE VOYAGE OF H.M.S. CHALLENGER.

During the "Transit of Venns " Expedition to Kergnelen in 1875, two species of
Serolis were obtained, which were figured and described by Miers, one of these — Serolis
septemcarinata — being new.

About the same time the German ship " Gazelle " collected a number of species of
Serolis both at Kerguelen and on the shores of South America, which were described by
Studer' in 1879. This paper contains, besides the description of a new species (Serolis'
corimta), a few notes upon the habits of these animals and upon some points in their
anatomy.

A few notes upon certain of the species of Serolis that are described in the present
Report are to be found in v. Willemoes Suhm's Preliminary Report ujjon the Crustacea
collected during the voyage of the Challenger.^

Finally, a recently published part of Bronn's Thierreichs (Bd. v. Abth. 2, Heft i.-x.)
contains a general account of the Isopoda by Prof. A. Gerstaecker, where some description
of the genus Serolis, especially of the geographical distribution, is given ; there are als(j
in this work a number of figures copied from the Memoirs of Grube and Studer ; and a
comparative view of the various schemes of classification of the order Isopoda ; for this
reason I have not in the present Report entered into any account of the systematic
positions which have been assigned to the genus Sei-olis by previous writers.

1 Archivf. Natvrgesch., Jahrg. xlv. Bd. i. p. 104.

2 Proc. Roy. Soc. LoncL, vol. xxiv. p. 585, &c., 1876.

BIBLIOGRAPHY.

The following list contains a reference to most of the treatises and memoirs which
contain any description or figures of Serolis. I have not attempted to include all the
works in which the name of the genus is merely mentioned ; I think, however, that no
paper of importance has been omitted.

1775. Fabricius, Systema Entoniologire, p. 296. Flensburgi et Lipsiai.

1787. Fabricius, Mantissa Insectorum, t. i. p. 240. Hafnice.

1789. Olivier, Article "Aselle" in EnoyclopeJie MethoJiqiie, t. i. p. 252. Paris and Li'-ge.

1793. Fabricius, Entomologia Systematica, t. ii. p. 503. Hafnice.
1818. Lkach, W. E., Dictionnaire des Sciences Naturelles, t. xii. p. 339.

1825. Desmarest, Considerations gonerales sur la Classo des Crustaoes, p. 292. Parix.

1829. Latreille, liegne Animale de Cuvier, t. iv. p. 132. Paris.

GuERiN, Article " Serole," Dictionnaire classique d'histoire naturelle, t. xv. p. 372.
1833. Eights, James, Description of a New Crustaceous Animal found on the Shores of the South Shetland

Islands, with Remarks on their Natural History. Trails.. Albanij Inst., p. 53.

1836. CuviER, G,, Le Regne Animals, 3"° ed., t. ii. p. 213. Bruxdles.

Buckland, Geology and Mineralogy, vol. i. p. 392. London.

1838. Lamarck, Hist. Nat. des Animaux sans Vertobres, 2""^ ed., t. v. p. 291. Paris.

1840. Milne-Edwards, H., Histoire Naturelle des Crustaces, t. iii. p. 228. Paris.
CuviER, G., Le Regne Animale, Les Crustaces avec un altas par. ^I. Milne-Edwards, p. 193,

pi. Ixiv. figs. 2, 3. (No date) Paris.

1841. AuDOUiN ET Milne-Edwards, Description des Crustaces nouveaux ou pen connus, &i;. Arch. d. Mas.

d'Hist. Nat., t. ii. p. 5, pis. i., ii.
1843. White, A., Description of New Species of Insects and other Annulosa. Ann. and Mag. Nat. Hist.,

vol. xii. p. 346.
1852. Dana, J. D., United States Exploring E.\peditiou — Crustacea, pt. ii. p. 794, pi. liii. fig. 1.
1858. LuTKEN, Chr., Beskrivelse af en ny Serolis — Art. Videnslc. Meddid. f. d. nat. Foren. i Kj^henhavn,

1858, p. 98, pi. i.
1866. ScHioDTE, J. C, Krebsdyrenes Sugemund. Kr^ijer, Nat. Hist. Tidsskr., Riikk 3, Bd. iv. p. 1G9, pi. x.
1868. ScHiODTE,- J. C, On the Structure of the Mouth in Sucking Crustacea. Ann. and Mag. Nat. Hist.,

ser. 4, vol. i. p. 1, pi. i.
1868. Cunningham, R. 0., Letter from, to ilr. Speuce Bate. Ann. and Mag. Nat. Hist., ser. 4, vol. i. p. 442.
1871. Grube, E., Bemerkungen ueber Lurabrico-nereis gigantea Qf. 1 und eine neue QJnone (CEnone vitiensis)

und Serolis (Serolis acutangula). Jahrp.^h. scJdes. Gnsellsch., pro xlix., 1871, p. 56. Breslau.

1871. Cunningham, R. 0., Reptdes, Fishes, Mollusca, and Crustacea obtained during the voyage of H.JI.S.

"Nassau" (1865-G9). Trans. Linn. Sor. Loud., vol. xxvii. p. 498, pi. lix. fig. 3.
1S74. WiLLEMOES SuHM, R. VON, Von der Challenger Expedition. Bricfe an C. Th. v. Siobold ; a series of

letters in Zuit-^chr. f. wiss. ZooL, Bd. xxiv., x.xv., xxvi., xxix.

6 THE VOYAGE OF H.M.S. CHALLENGER.

1875. MiERS, E. J., Descriptions of New Species of Crustacea collected at Kerguelen Island by tlie Eev. A. Eaton.

Ann. and Mag. Nat. Hist., ser. 4, vol. xvi. p. 73.
Kidder, J. H., Contributions to the Nat. Hist, of Kerguelen Island — Crustacea, by S. L Smith. Bull.

U. S. Nat. Mits., 1875, p. 63.
Grube, E. a., Beitrag zur Kenutuiss der Gattung Serolis. Archiv f. Naturgesch., .Jahrg. Ixi, Bd. i. p. 208,

pis. Y., vi.

1876. LocKiNGTON, W. N., Remarks on the Crustacea of the Pacific, with De.scriptions of some New Species,

p. 208, pis. v., vi. Froc. Calif. Acad. Nat. Sci., vol. vii. p. 36.
Miers, E. J., Catalogue of New Zealand Crustacea, p. 117, pi. iii. fig. 7. London, 1878.

St0der, Th., Ueber neue Seethiere aus dem Antarktischen meere. Alitth. Naiurf. Gesellsch. Bern,

1876, p. 75.
WiLLEMOES SuHM, R. VON, On Observations made during the earlier part of the Voyage of H.M.S.

Challenger. Froc. Roy. Soc, vol. sxiv. p. 569.
Willemoes Suhm, R. von. Report on Crustacea observed during the cruise of H.M.S. Challenger.

Froc. Roy. Soc, vol. xxiv. p. 585.
ScHiiELZ, Serolis tuberculata, Gr., und Serolis orbignyana, Aud. et Edw. Joarn. Mns. Godeff., part xii.

p. 161.

1877. Lucas. Bull. Eni. Franc., 1877, p. 146.

GiGLioLi, Enhioo, Viaggio intorno al globo della Magenta, p. 956. Milan.

1879. Studeb, Th., Die Fauna von Kerguelen's Land. Arcliiv f. Na,turgesch., Jahrg. xlv. Bd. i. p. 104.

MiEES, E. J., Crustacea, iu An Account of the Petrological, Botanical, and Zoological Collections made in
Kerguelen's Land and Rodiiguez during the Transit of Venus Expedition. Fhil. Trans.,
vol. clxviii., extra vol., p. 200, pi. xi.
Studer, Th., Beitriige zur Kenntniss niederer Thiere von Kerguelen's Land. Archiv f. Natiirgescli.,

Jahrg. xlv. Bd. i. p. 19, pi. iii.
MilneEdwards, a, Compte rendu des recherches sur la faune de I'lle de Kerguelen, faites par lea
membres de I'expedition astronomique. Ann. d. Sci. Nat., ser. 6, t. viLi.

1881. Miers, E, J., Cru.stacea collected during Survey of H.M.S. "Alert." Frnc. Zool. Soc. Lond., 1881, p. 61.

1882. Lutken, Chr., Dyreriget, p. 468, with fig. Kjijhenhavn.
1S83. Bronn's " Thierreich's," Bd. v. Lief 1-10, 1881-83, pi. v. figs. 4-9. Lripsic and Heiddbenj.

Studer, Th., Isopoden gesammelt wiihrend der Reise S.M.S. " Gazelle " urn die Erde 1874-76. Abkandl.
d. I: Al.-ad. d. Wiss. Berlin, 1883, pp. 1-28 with 2 pis.
1881. Beddard, F. E., Preliminary Notice of Isopoda collected during the Voyage of H.M.S. Challenger.
Part I. The Genus Serolis. Froc. Zool. Soe. Lond., 1884, pt. iii. p. 330.

DESCRIPTION OF THE GENUS.

Serolis, Leach.

Oniscus, Fabricius, Mantissa Tnsectoram, p. 240, 1787. In Part.
Asellus, Olivier, Encycl. M6th., t. i. p. 2.52. In Part.

CymotJwa, Fabricius, Entomol. Syst., ii. p. 503, 1793. In Part.
Serolis, Leach, Diet. d. Sci. Nat., t. xvii. p. 340, 1825.
Brongniartia, Eights, Trans. Alb. Inst., ii. p. 53, 1833.

Definition.— Qodj much depressed, round or oval in form, usually furnished with
long sickle-shaped epimera. Cephalic shield broad, terminating anteriorly in a short
median rostrum, uniting with fii'st two segments of thorax to form a cephalothorax.
Terminal segment of thorax rudimentary, its tergum unrepresented ; sternum partially
fused with that of preceding segment ; epimera absent. Abdomen consisting of three free
segments and a caudal shield ; first segment devoid of epimera, and completely enclosed by
the preceding and succeeding segments ; second and third segments with longer or shorter
epimera. Caudal shield usually pentagonal, with one or more longitudinal carinse, rarely
smooth. First pair of antennae subequal in length to second pair, the former with four
basal joints and a multiarticulate flagellum, the latter with five basal joints and a multi-
articulate flagellum. Mandibles stout and strong, with a long three-jointed j^alp. Maxillai
small and delicate. Maxillipedes large and operculiform, entirely covering maxillae.
Ambulatory appendages seven 23airs ; anterior pair (in male second pair also) modified
into a prehensile organ, the terminal joint folding back upon the greatly dilated pen-
ultimate joint ; last thoracic appendage smaller than the others ; first three pairs of
abdominal appendages natatory, with broad basal and two expanded distal joints fringed
with long plumose hairs ; inner joint of second pair prolonged in the male into a
penial filament ; fourth and fifth appendages branchial ; sixth pair natatory, compata-
tively small.

External Characters. — The depressed form of the body which characterises the family
of the Isopoda as contrasted with the Amphipoda is very marked in the genus Serolis ; in
some .species (Serolis schythei, Serolis comuta, &c.) the body is almost flat ; in others
again {Serolis convexa, &c.) the convexity of the dorsal surface is greater, and Serolis

8 THE VOYAGE OF H.M.S. CHALLENGER.

latifrons is conspicuous by its strongly arched dorsal surface, which, together with the
shortness of the epimei-a, causes this species to resemble in general aspect the more
t^'jjical forms of the Cymothoadee.

The outline of the body varies from oval {Serolis tuberculosa) or even pear-shaped
{Serolis convexa) to circular, and in the males of Serolis schyihei and Serolis cornuta the
diameter of the body from side to side is somewhat greater than the length.

In the majority of the Isop)oda the "head" segments become fused with the first
segment of the thorax, and form a cephalic shield which is freely movable upon the
second thoracic segment. In Serolis the disposition of the anterior segments of the body
differs much from other Isopoda. The first two thoracic segments are closely united and
completely fused dorsally, though the sterna of the two remain distinct ; in some species
[Serolis schythei, Serolis necera, &c. ) an incomplete transverse suture upon the first epimera
seems to mark the line of division between the two segments dorsally ; in others again
[Serolis antarctica and all the Australian species) the epimera of the two thoracic
segments are completely united, and show no traces of their original distinctness ; these
epimera are always largely developed, and completely enclose the cephalic shield on both
sides. The cephalic shield is very large and has the form of an heraldic shield ; it is p)ro-
longed in front into a short rostrum ; its shape varies considerably in the different
species, in some [Serolis necera, Serolis paradoxa) the antero-lateral portions are consider-
ably expanded, and the transverse diameter is greatest here and decreases posteriorly ;
in other species [Serolis convexa, Serolis schythei) the cephalic shield is widest at the
level of the eyes. In the majority of species the cef)halic shield is sej)arated from the
two thoracic segments by a continuous suture ; in Serolis longicaudata this suture is
obsolete behind and indicates the commencement of the formation of a cej^halothorax
composed of the head segments and the two first thoracic segments as in Tanais, with
which genus Serolis is considered by some to be closely connected. The five following
thoracic segments are free ; the eighth or terminal segment of the thorax differs from
that of other Isopoda in being only represented ventrally by a short sternum, which
is more or less intimately fused with that of the preceding segment, and is not prolonged
into epimera ; the tergum of this segment is entirely undeveloped, and the rudimentary
condition of the whole segment (including the appendages, which are much smaller than
those of the other thoracic segments) is interesting, inasmuch as in the Isopoda generallj'
the terminal segment of the thorax is only developed very late.

The epimera of the thoracic segments are almost always largely developed in com-
parison with other Isopoda; and in some of the deep-sea species [Serolis bromleyana,
Serolis necera) are extraordinarily elongated, and terminate in sharp spiniform points. As
a general rule the epimera are flat and sickle-shaped, curving back and gradual!}'
increasing in length up to the sixth pair, which are invariably the longest ; sometimes
[Serolis latifrons, Se7vlis longicaudata, &c.) the epimera are very short, and hardly exceed

REPORT ON THE ISOPODA. 9

ill length those of many Cymothoadaa ; where the epimera are only moderately developed
they are in close contact for the greater part of their length, the anterior slightly over-
lapping the posterior ; in Servlis necera, Serolis hromleyana, and Serolis gracilis the
epimera are quite independent for the greater part of their length, and are only in actual
contact for a short space close to the junction of the epimera and tergum ; at this point
the anterior margin of the epimeron projects forwards as a short rounded process which
is received into a "glenoid" cavity between two similar processes, one dorsal and one
ventral, of the epimeron in front ; of these processes only traces exist in Serolis scliythei
and those species in which the epimera are closely applied to each other ; an intermediate
condition is seen in Serolis paradoxa and other species, where each epimeron has two
articular processes, one anterior and one posterior ; the anterior process of each segment
overlaps the posterior process of the segment in front.

The epimera of the first three free thoracic segments are invariably separated from
their terga by a distinct suture ; in some species (Serolis schytliei) the following pair of
epimera are also thus separated ; in Serolis latifrons all the free thoracic segments have
the epimera divided by a suture from the terga.

The abdomen in Serolis consists of three freely movable segments and a terminal
caudal shield which represents the three posterior appendages of the abdomen together
with the telson.

The anterior segment of the abdomen is enclosed between the penultimate thoracic
and the second abdominal segment, and in one species only has any traces of epimera ;
in Serolis latifrons a minute portion is separated off from the median portion on either
side by a suture ; and as this suture is quite continuous with those in front which separate
the terga and sterna of the thoracic segments, it seems that the minute nodule on either side
of the first abdominal segment really represents its epimeron. The two succeeding
segments are always furnished with epimera, which are, however, never separated by a
suture from the terga, and in most cases are short and not prolonged beyond the lateral
margin of the caudal shield ; in some species, notably in Serolis hromleyana, the epimera
of these segments are elongated, and reach midway down the margin of the caudal shield,
or even considerably beyond its termination.

The remaining abdominal segments are fused with the telson to form a caudal shield
which is commonly more or less pentagonal in outline, and is always furnished with a
longitudinal carina and sometimes with a pair of lateral carinse which apjaear to mark the
line of division between the posterior abdominal segment and the telson ; the greater
portion of the caudal shield is occupied by the latter.

Appendages. — The two pairs of antennce are generally subequal in size, sometimes
[e.g., Serolis minuta) the anterior pair, sometimes (e.g., Serolis cornuta) the posterior pair
are the longer.

Tlie anterior antcnua3 consist of four joints and a terminal filament, while the posterior

(ZOOL. CHALL. EXP. — PAET xxxui.— 1884.) Kk 2

10 THE \'OYAGE OF H.M.S. CHALLENGER.

auteuuEe have a stem composed of five separate joints ; tlie terminal filament is usuall)'
shorter than that of the anterior antennae.

Tactile Organs. — The peculiarly modified antennary hairs first described by Leydig/
and believed by him to be sense organs, exist upon the first pair of antennae only in all
the species of Serolis that I have examined. These organs are of two kinds, the most
conspicuous being cylindrical jointed hairs, several of which are displayed in PL I.
fig. 4, b ; each consists of a basal portion somewhat hour-glass shaped, which is attached
to the inner side of the upper extremity of each of the joints which compose the filament
of the antennules ; this is continued into a delicate cylinder with thin walls, which is
always divided into two portions by a transverse septum, and occasionally seems to
consist of three distinctly separated portions ; the distal end of each of these cylinders is
formed by a knob-like thickening of the chitinous wall. Leydig describes and figures
nerve fibres and cells in these olfactory hairs, but the Challenger specimens were not
sufliciently well preserved to show these structures.

As a general rule, only one of these olfactory hairs is found upon each joint of the
antennulary filament, but in Serolis antarctica, Serolis australiensis, and one or two
other species, two such hairs are found upon each joint.

The " tactile " organs described by Leydig in the Memoirs already quoted I have only
seen in Serolis schythei and Serolis necera ; on the antennules of these two species one
tactile hair only is found upon the distal extremity of the terminal point (PL V.
fig. 5, a) ; it is a short slender hair, the upper half being plumose. Similar hairs are also
found over the general body-surfoce and upon the first joint of the ambulatory appendages
(PL III. fig. 11) of many species.

The mandibles are very strong and powerful, and furnished with a long three-jointed
palp which is longer than the mandible itself; the second joint of the palp is longer
than the first joint, and the third, which is very small, is semicii'cular in shape, and
its flat inner margin is furnished with a row of stiff" hairs, of which the two distal
ones are the longest. These hairs are continued for a short way on to the middle joint
of the palp, this part of the joint being of a somewhat greater diameter than the posterior
portion, which is devoid of hairs. The basal portion of the mandible is broadest
proximally, where it articulates with the head, it narrows abruptly into the distal half,
which is not more than one half of its diameter ; the latter is bent at an angle to the basal
part, terminates in the masticatory edge which is of a dark brown colour, and is either
straight or slightly sinuous, or provided with one or two blunt tooth-like projections ; in
adult specimens only are the mandibles thus furnished with a comparatively smooth edge,
Avhich appears simply to be due to wear and tear ; in young specimens of all the species

1 Fr. Leydig, Ueber Geruchs- uiid Gehororgane der Krebse und Insecten, Archivf. Anat. u. Physiol, 1860, pp. 265-
314, Tafn. vii.-ix. See also the same Ueber Amphipoden uud Isopodeu, Zeitschr. f. iciss. Zool, Bd. xxx., Suiiplemeiit,
pp. 225-274, Taf. ix.-xii., 1878.

REPORT ON THE ISOPODA. 11

that I was able to examine the inner edge of the mandibles is strongly toothed. Schiodte,
in a memoir upon the structure of the mouth in the Isopoda,^ has figured and described
the mouth parts of Serolis paradoxa ; the mandibles are distinguished from those of other
Isopoda by their blunt edge, which renders them especially fitted for grinding and not for
tearing, and Schiodte suggests that the food of Serolis probably consists of " such animals as
have firm integuments ; " in all the specimens that I have examined the stomach contained
debris of other Crustacea, frec[uently recognisable fragments of their own species, besides
Diatoms, morsels of the calcareous structure of Echinodermata, &c.; Schiodte is therefore
quite right in his supposition regarding the food of Serolis, and is no doubt justified in
distinguishing this genus from other Isopoda by the characters of its mouth appendages.

The mandibles of Serolis, however, possess additional cutting blades and spiniform
processes which appear to have been overlooked or imperfectly described and figured by
all previous writers with the exception of Schiodte. On PL II. figs. 2, 3 are figured the
distal extremity of both mandibles right and left of Serolis latifrons ; the left mandible
has upon the upper surface a triangular chisel-like blade, and below this a long spine ;
the right mandible has the spine, but the cutting blade is smaller. The left mandible
of other species is quite similar to that of Serolis latifrons, but as a general ruk'
[cf. e.g., fig. of Serolis schythei, PI. II. figs. 12, 13) the cutting blade is not present on the
right mandible, its place being occupied by a large spine-like structure. The presence
of these structures appears to me to fit the mandibles for tearing as well as grinding.

The asymmetry of the mandibles is marked not only by the difi"erences just mentioned
l)ut also by the general shape of the masticatory edge {cf. e.g., PL II. figs. 12, 13) and by
the fact that one mandible, generally the left, is smaller than the other. An asymmetry
of the mandibles appears to be a fairly constant character of the Isopoda.

The maxill(e are small and delicate ; the anterior pair consist of a narrow basal
portion or " cardo," with which is articulated a long flat somewhat curved distal lobe ; the
free edge of this is furnished with a considerable number of strong spines dark yellow in
colour. In some species {e.g., Serolis pallida) the lobe of the maxilla is much more bent
upon itself and proportionately shorter. In many species I have observed an additional
lobe, which is situated below the large masticatory lamina articulating with the cardo
(see PL V. fig. 14); the presence of this was noticed by Audouin and Milne-Edwards,
but subsequently denied by Grube.^

The second pair of maxillse are smaller and more delicate than the anterior }iair, but
like them consist of a basal portion or " cardo " and a distal portion, which is divided into
two lobes, the upper of which is again subdivided ; the two anterior are subequal and
smaller than the posterior lobe, which is separated by a distinct suture ; the free
extremities of all bear a number of hairs more delicate than those on the first maxilla,

' Krpjer, Xat. Hid. TitMcr., ser. 3, vol. iv. (1866) ; translated in A nn. and Mag. Nat. Hist., ser. 4, vol. i. p. 1 et seq.
2 Archivf. Naturgesc.h ., Jahrg. xli. Bd. i. p. 214.

12 THE VOYAGE OF H.M.S. CHALLENGER.

uud pinnate at their extremity. There are generally two upon each of the smaller anterior
lobes and a greater number upon the posterior lobe ; in a specimen of Serolis cornuta
{cf. PI. 1. figs. 9,10) this pair of maxillse, like the mandibles, was asymmetrical on one
side ; each of the two smaller lobes had two long hairs upon its free extremity ; on the
side one of the smaller lobes was markedly larger than the other, and furnished with four
hairs instead of two.

In Serolis convexa the second pair of maxillse differ in that all the three lobes are
approximately of equal size, and all bear a considerable number of hairs (PI. VI. fig. 14).

The maxillipedes are of considerable size and entirely conceal the subjacent maxillse ;
they are closely approximated in the middle line ; each consists of a squarish basal portion
and a palp ; the basal portion is divided by a transverse suture into two, and each of
these is again divided by a longitudinal suture ; the proximal half or cardo articulates
with both the median (sphenoidal plate) and the lateral portion of the sternum. The
outer half of the distal half or "stipes," which is generally, though not always, completely
separated from the inner " lamina" by a suture, is thin and flat ; the lamina is stout, -and
usually furnished on the inner margin with a row of fine hairs ; the upper end invariably
bears two thick spines ; the palp of the maxilla, which consists of three joints, the
middle one the longest, is attached to the lamina about three-quarters of the way down ;
the second and third joints of the palp are furnished with long hairs on the iimer side
in all species ; the other joints of the maxillipede are sometimes smooth, sometimes
furnished with long hairs, according to the species. , ,

The second thoracic appendage {cf. PI. VI. fig.-^^ is modified into a prehensile organ,
the penultimate joint is large and swollen, and the terminal joint is recurved and lies
along its inner margin like the blade of a penknife when closed ; the inner margin of the
penultimate joint has a number of peculiarly formed spines, unlike any that are found
elsewhere on the body. Of these there are two kinds which regularly alternate ; one set
(PI. VIII. fig. 15) consist of a central stem terminating above in a knob,^ the margins are
beset with a number of fine branches which are fused together on each side for the
greater part of their length ; the other kind of spines (PI. VIII. fig. 14) which alternate
wdth these, and are placed slightly more on the ventral side, are long and delicate,
expanding above into two processes, one of which is frequently longer and somewhat
spoon-shaped, between these the central axis of the spine terminates in an oval knob ; the
shape of these hairs varies much in the different species, and will be more fully described
below under the several species.

The fourth joint of this appendage, which is sometimes (Serolis tuhercidata) prolonged
into a forwardly directed triangular process, is always furnished with two spines, and has
in the males of Serolis convexa and Serolis gaudichaudii a tuft of sensory hairs (see p. 17).

The third pair of thoracic appendages in the male (see p. 16) is modified into a

1 Owing to an error in the plate, this is represented as a forked process.

REPORT ON THE ISOPODA. 13

prehensile organ like the preceding limb, but smaller. In the female it is similar to the
succeeding appendages.

The remaining thoracic ajopendages are similar to each other, and are always
furnished with tufts of hairs and spines. The first joint, which is always the longest, is
generally smooth, or provided on the inner surface with fine plumose hairs like the single
sensory hair commonly found upon the terminal point of the filament of the first
antennse. The four following joints are smaller, and generally furnished on the inner and
outer side with tufts of sword-like and of serrated spines. In many species (Serolis
antarctica) these are to a great extent replaced by fine hairs, and in Serolis necera
all the thoracic appendages are clothed with fine plumose hairs similar to those which
in other species are only to be found upon the abdominal appendages. A more detailed
description of the varieties of these hairs and spines will be found under the descri])tion
of the several species.^

The last pair of thoracic appendages (see p. 17) are usually smaller, and very often
in the males {Serolis gracilis) furnished on the inner side with tufts of delicate sensory
hairs like those upon the second thoracic appendage in the males of Serolis convexa.

The abdominal appendages (see PL I. figs. 12-14) are much specialised; the three
anterior pairs form "swimming feet"; each of these consists of a proximal joint, the
protopodite, which is generally triangular in form and attached to the segment which
bears it for about a quarter of the length of tlie base ; the projecting angle bears three
plumose hairs in the first of these appendages and tivo in each of the two following. In
some species (viz., Serolis schythei, Serolis paradoxa, and in all the Australian species), the
protopodite is narrower, less triangular in form, and has no hairs. The inner margin is
usually clothed with a dense covering of fine delicate hairs, which are also found, though
to a less extent, upon the outer border. About half-way between the summit and the
base the endopodite is attached, and the exopodite at the summit of the protopodite ; both
the exopodite and endopodite are delicate semicircular lamellae, the former being larger
than the latter ; the outer convex margin of both is provided with numerous plumose hairs
which, as shown on PI. I. figs. 15, 16, consist of a central hollow stem gradually decreasing
in diameter towards the extremity, and giving off on either side a series of extremely fine
lamellae ; in the interior of the stem I was able to observe, in several cases (PI. I. fig. 16),
a fine thread occasionally looped upon itself and possibly a nerve fibre ; the function of
these hairs may perhaps be to test the quality of the water coming to the gills.

The lower border of the endopodite in the second of these appendages is prolonged in
the male into a penial filament (PI. VII. fig. 2', b). The next two pairs of appendages con-
sist of a short basal joint laterally elongated, with which are articulated two broad lameUse,
au endopodite and exopodite, wdiich are the gills ; the exopodite of the first appendage is

1 I make an arbitrary distinction here and elsewhere between the finer "hairs" and stouter "spiiies" in order to
express more easily the differences between tlie ambukitury appendages of difl'erent species.

14 THE VOYAGE OF H.M.S. CHALLENGER.

much stouter than its endopodite, and forms an " operculum " divided by a transverse
suture, which may be at right angles to the longitudinal axis [Serolis tuberculata, &c.) or
inclined more obliquely to it (Serolis schythei, &c.) ; very commonly {Serolis necera, PL V.
fig. 1 1) the endopodite of this appendage is bifid at the tip, the inner branch being prolonged
beyond the outer ; in most cases, however, the endopodite is entire ; the second pair of
gills is always smaller than the anterior pair, the exopodite and sometimes the endopodite
is divided by a transverse suture which follows the direction of the suture on the exopodite
of the first pair of gills. The terminal appendages of the abdomen or " uropoda " are
always comparatively small ; in Serolis cmtarctica they are extremely minute ; these
appendages consist of a triangular basal joint attached by the apex to a notch on the
lateral margin of the caudal shield, which varies in position in different species, being
sometimes at the anterior end of the lateral margin of the caudal shield, sometimes close
to its posterior extremity ; the inner angle of the protopodite is always longer than the
outer, and the endopodite is longer than the exopodite ; in Serolis latifrons Studer' has
correctly described the elongated and spine-like endopodite which becomes completely
fused with the protopodite, while the exopodite remain very short ; the appendage is
capable of being flexed at right angles to the body, and its sharp spiniform character
doubtless serves as a protection against many foes.

Sexual Characters. — The generative apertures of the male are always situated on the
last thoracic segment, on either side of the median ventral line they are always closely
approximated, but differ slightly in different species; in some {e.g., 'Serolis antarctica)
they are so close together as almost to form a single orifice ; in others {Serolis schythei,
&c.) they are a little further aj^art.

The female generative apertures are more widely separated fi-om each other than the
male generative apertures ; they are situated on the antepenultimate thoracic segment, and
have the form of an oval slit, while the male generative apertures are invariably circular.
The ova are carried about by the female until the young are fully formed ; the
immature females have four ovigerous lamellae (see PI. V. fig. 4) developed on the second to
the fifth thoracic segments inclusive, which are short, oval in shape, and attached close to
the attachment of the thoracic limbs. Studer ^ was the first to point out that these lamellae
become much lai'ger in the mature females with eggs, and overlap each other in the
middle line, those on the right generally covering those on the left ; this disposition of
the ovigerous lamellae, though general, is not, however, universal ; occasionally {e.g., Sei'olis
convexa) the arrangement of the lamellae alternates, the right hand lamellae of the second
segment overlapping that of the left ; in the two succeeding segments the left hand lamellae
overlaps the right, and finally the disposition of the lamellae in the fourth segment is like
that in the second. These cases appear, however, to be irregular, and not to be char-
acteristic of different species. When these brood lamellae are fully developed the sterna

1 Archivf. Naturgesch., 1879, loc. at. 2 ^pj_ ^.j-^.^ p. 20.

REPORT ON THE ISOPODA. 15

of the thoracic segments undergo a retrograde (hn^dopment and ahiiost disappear, and the
animal does not appear to take any nourishment ; in all the mature females that I have
examined the young appear to be actually contained within the body of the mother, the
alimentary canal is pressed up against the dorsal surface of the body, and its cavity is
reduced to a minimum ; a delicate chitinous membrane is all that remains of the
thoracic sterna.

Secondary Sexual Characters. — The males of Scrolls also differ from the females
in a number of secondary sexual characters, which may be divided into two groups —
(1) those which are common to all the species of the genus, (2) certain other characters
which are confined to one or more species.

1. As a general rule the males of Serolis are somewhat larger than the females ;
this is certainly the case with Serolis hromleyana and Serolis newra, in which species
the males are not only longer but also broader, owing chiefly to the greater development
of the spine-like epimera so characteristic of these two species. In Serolis cornuta,
Serolis schythei, and Serolis latifrons the diflference in length between the two sexes does
not appear to be great, but the males are considerably broader than the females; especially
is this so in Serolis schythei, where the proportion between length and breadth is almost
reversed in the two sexes ; in the male the breadth is greater than the length, in the
female, on the contrary, the length is greater than the breadth, and the difference between
these two dimensions is very near to being the same in both sexes. In Serolis antarctica
and Serolis gracilis the males have a greater antero-posterior as well as transverse
diameter than the females.

Among the Australian species, on the contrary, the females appear to be larger than
the males, but since the number of specimens in each case was extremely small, it is
impossible to speak with certainty. Of Serolis australiensis two out of the three speci-
mens are males and smaller than the single female specimen ; of Serolis tuherculata the
Challenger obtained two specimens, one of each sex, and here again the male is the
smaller ; the male specimen also which has been described by Grube ^ is of about the
same size as the male which I have examined ; finally, in the only other Australian species,
Serolis ixdlida, of which more than a single example was obtained by the Challenger,
the female is larger than the male.

In the majority of species of Serolis there is a diflference in the length of the epimera
in the two sexes, and this difference is most strikingly shown in Serolis gracilis; in the
males of this species the epimera gradually increase up to the sixth, the last thoracic
epimera, which are enormously elongated and extend beyond the termination of the
caudal shield for a considerable distance.

In the female the epimera are very much reduced in size ; the sixth pair, instead of
l)eing prolonged beyond the caudal shield, do not reach as far as its extremity.

> Loc. cit., p. 227.

16 THE VOYAGE OF H.M.S. CHALLENGER.

In most other species the same conditions occur, and a comparison in detail of the
posterior epimera, which differ more in length in the two sexes than the epimera of the
anterior segments, will be found under the description of species.

Some few species do not show these differences ; in Serolis pwiudoxa, for example, the
males, on a superficial view, are indistinguishal)le from the females, and the difference in
size between the two sexes is hardly if at all marked.

As a general rule the sterna of the three anterior abdominal segments serve to
distinguish the sex of the individual ; in the females the middle portion is commonly
prolonged into a stout spine, while in the males this structure is not present, and
the posterior margin of the segments is straight or slightly concave. In a great
number of species, however, the two sexes do not differ at aU in this way.

Another marked secondary sexual character, which is quite universal in the form of
the third thoracic appendages ; in the females this pair of appendages is entirely similar
to the succeeding ambulatory limbs ; in the males, on the contrary, the penultimate joint
is swollen and furnished on the inner side with a number of peculiar modified spines,
the terminal joint is recurved, and the appendage thus forms a prehensile organ very
like the second thoracic appendage. Of Serolis tuherculata Grube states (loc. cit.,
p. 230) — "Die Fiisse des 2'™ Fusspaares sind weniger ausgepriigte Greifflisse als bei
andern Arten ; zwar zeichnet sich das 3** und 4'° Glied durch seine Kiirze vor den ent-
sprechenden der folgenden Beine aus, allein das Handglied ist weniger breit als sonst, sein
Innenrand nicht langs der ganzen Schneide mit Zahnen besetzt, und die Klaue scheint
nicht so zum einschlagen geeignet zu sein. Die Ziihne sind ziemlich lang und stachel-
formig mit einer Andeutung von Nebenzacken." The male specimen of this species
which I have examined myself does indeed display such differences from the ordinary
structure of these appendages in aU other Serolis as Grube describes ; both
specimens are, however, evidently immature ; the characters that he mentions exactly
correspond to the appendages of immature males (see infra, p. 27).

These appendages are used by the male during copulation ; the claw is firmly
imbedded in the epimera of the female, so firmly that the individuals can hardly be
separated without injury.^

In aU species of Serolis, as in many other Isopoda, the second pair of abdominal
appendages bear a long penial filament ; these are a continuation of the lower margin
of the endopodite of the limb, and reach in some cases as far back as the end of the
caudal shield ; in other species they are not quite so long. The end of these filaments is
blunt and rounded, and not furnished with any apertm-e ; it seems possible for this reason
that the two are approximated during copulation, and form a groove down which the
spermatophores pass ; the fact of the male generative apertures being placed so closely
together seems to favour this supposition.

' Studer, Isopoden gesamiuelt wahreiid der Keise, &c., loc. cit.

REPORT ON THE ISOPODA. 17

lu the females there is uo trace whatever of these penial filaments.

2. Under this head I may briefly review certain outward differences in male and
female individuals which are peculiar to one or more species, and of which a fuller
description is given below.

In Serolis convexa (PI. VI. fig. 10) and Serolis gaudichaudii the second pair of
thoracic appendages in the male are furnished with a tuft of pinnate hairs upon the
inner side of the antepenultimate joint, and in the male of the former species, at any rate,
the sterna of certain of the thoracic segments are provided with patches of similar hairs.

In Serolis septemcarinata the epiraera of the males bear a row of five or six elongated
ridge-like tentacles on the under surface which are entirely wanting in the female.

The females of Serolis necera are distinguished from the males by the far greater
development of the frontal "sense organ "^ {cf. PI. V. figs. 1, 3), and the epimera them-
selves show certain differences in the two sexes, being considerably wider in the male,
where the anterior margin from the point of junction with the lateral angle of the
cephalic shield slopes gradually backwards and outwards to the termination of the
epimeron ; in the females the anterior margin passes at first directly backwards almost
parallel to the longitudinal axis of the body, and then slopes outwards so that in this
sex the anterior epimera look as if a piece had been cut out.

Another sexual difference is found in Serolis gracilis and Serolis convexa, and is men-
tioned by Audouin and Milne-Edwards as occurring in Serolis gaudichaudii.' In these
species the terminal thoracic appendages of the males are beset with fine hairs, which are
similar in shape to the hairs upon the second thoracic appendages of the males of Serolis
convexa and Serolis gaudichaudii; these same appendages in the females do not differ in
any way from the preceding thoracic appendages.

On PI. II. fig. 6 is figured a single joint from the filament of the second antenna
of a male Serolis schythei ; along the inner margin of the joint are a series of delicate
fan-shaped lamellae which I did not succeed in finding in any female example of this
species; it is possible, however, that on account of their extreme delicacy they may have
been overlooked.

Alimentary Canal. — The alimentary canal of Serolis agrees closely with that of other
Isopoda ; the point to which I may call special attention is the presence of salivary glands,
which I have noticed in Serolis imradoxa and Serolis septemcarinata. On account of

^ Grube was tlie first to point out the presence of a transparent oval area on the first epimera which he imagined
might cover some sense organ ; Dr. Woodward (Geol. Mag., 1883, p. 21) has compared this structure to certain pores
which are to be found in many Trihibites occup}-ing an identical position. I have foimd that this structure exists in
nearly all tlie species of Serolis, generally having the form of a deep and narrow groove surrounded by a, specially
thickened rim ; in Serolis schythei and Serolis cornuta the structure is precisely as described by Grube in Serolis paradoxa,
and as in that species there is a pore on the under surface of the epimeron exactly beneath it. In Serolis antarctica and
others I could discover no trace of it. Concerning the minute anatomy of this " sense organ," I am unfortunately able
to say nothing.

^ Lac. cit., p. 19.

(ZOOL. CHALL. EXP. PART XXXUI. — 1884.) Kk 3

18 THE VOYAGE OF H.M.S. CHALLENGER.

its conveniently small size I selected the latter species for anatomical study by means of
transverse sections, which I found much better than dissection owing to the state of
preservation of the specimens.

On PI. X. fig. 6 is figured a transverse section through the head of Serolis septem-
carinata showing the salivary glands in situ; a is the aperture of the mouth, h the buccal
cavity, and c points to the ventral wall of the body bounding the mouth ; the salivary
glands (d) are arranged in rosette-shaped aggregations of cells surrounding a central
cavity ; these cells are highly granular, and stain deeply with carmine ; each is provided
with a large nucleus which stains rather more deeply than the surrounding protoplasm ;
the glands are imbedded in the cellular connective tissue of the body, some of the cells
of which are shown in the figure (e). Similar structures have been observed by Dohrn'
in Anceus and Faranthura, and are figured and described by him in his account of the
anatomy of these two genera.

The oesophagus, as in other Crustacea, passes upwai'ds into the masticatory stomach,
which is furnished with a complicated series of chitinous plates.

On PL X. fig. 9 is figured the masticatory stomach of Serolis hromlei/ana viewed from
beneath ; the " pyloric " end is directed towards the upper margin of the plate.

There is an upper median azygos cardiac plate (m) ; in front of this and below is a
cylindrical plate {LR, RP) on either side ("Reibe-platte"), the lower surface of which is
traversed by a row of rib-like thickenings. On either side of the median plate are three
lateral plates (I, la, Ip), one median, one anterior, and one posterior ; the median lateral
plate {T) is furnished with numerous fine hair-like processes on the under surface. On
the ventral surface of the stomach is a large ossicle (F) which reaches beyond the
commencement of the dorsal azygos piece (m), and terminates in a bifurcate extremity ;
behind this comes the pyloric portion of the stomach, which has four ossicles on either
side; two small triangular ossicles (t), situated just above the median piece (F), and
behind three pairs of elongated ossicles which are furnished with delicate hairs directed
inwards and form a sieve-like structure ; the outermost ossicle (2) is connected both with
the median ventral ossicle ( F) and the lateral plate (R.P.) of its own side ; s points to the
pyloric aperture.

In Serolis antarctica the masticatory stomach is entirely similar.

In Serolis paradoxa the masticatory stomach apparently differs from that which has
just been described by the very incomplete calcification of the cardiac ossicle, which is
indeed hardly to be recognised as a distinct ossicle ; the ventral median ossicle does
not extend so far backwards as in Serolis hromlcyana, and is less markedly emarginate
at its apex.

The masticatory stomach in the two other species in which I have studied it, viz.,

• Entwickelung und Organisation von Praniza (Anceus) maxillaris, Zeitschr.f. wiss. ZooL, Bd. s.x. p. 55, 1870; and
also Zar Kenntniss d«3 Baues von Parauthura costaua, Ihid., p. 81.

REPORT ON THE ISOPODA. 19

Serolis schjthei and SeroUs latifrons, presents some differences. Fig. 4 is a drawing of
the masticatory stomach of Serolis schythei, which is entirely simihir to that of Serolis
latifrons; the chief difference from Serolis bromleyana is that the lateral ossicle (fig. 4, 1)
is furnished upon the upper surface with short spines in addition to the delicate hairs
which clothe its lower surface. The outermost of the three pairs of ossicles which form
the pyloric half of the stomach is also more extensive, and bears the ribbed lateral plates
{liP, LP) upon the anterior edge.

At the junction of the masticatory stomach with the intestine are four caeca which
are long and coiled in Serolis cornuta (PI. X. fig. 2); in a specimen of Serolis
necera that I dissected there were also four caeca, two situated beneath the gut, and con-
siderably shorter than the other two which lay along the outer margin.

From the masticatory stomach arises the intestine, which is at first wide but gradually
narrows towards the rectum ; the latter commences at about the level of the fifth
thoracic segment, and is separated off from the intestine by an incomplete circular valve ;
the anus is an oval aperture on the ventral surface of the body betw^een the attachments
of the gill plates. The intestine as well as the rectum is provided internally with a
series of longitudinal glandular folds.

In a number of small specimens of Serolis latifrons mounted on slides in Canada
balsam the alimentary canal was distinctly visible ; between the wide anterior portion
of the intestine and the rectum, which is half its diameter, is a narrow portion of the gut,
measuring at its commencement rather less than one half of the diameter of the rectum,
and then becoiuing slightly wider as it approaches the latter.

Nervous System. — The nervous system of Serolis paradoxa is figured in Packard's
Zoology ; ^ Studer has also given a figure and description of the nervous system of
Serolis latifrons ; the former of these two figures appears to represent more strikingly
the concentration of the posterior ganglia into a nervous mass where the commissures
and connectives between the several eranglia are lost.

I have studied the nervous system of the genus in two species — in young examples
of Serolis carinata by means of sections and by simple inspection of the entire animal
mounted in glycerin ; in Serolis necera by dissection.

The nervous system of both these species, as in oth&c Crustacea, shows a relation to
the segmentation of the body ; the fusion of the anterior segments is accompanied by a
fusion of their ganglia, and the same thing has taken place in the posterior region of the
body. On PI. II. fig. 14 is represented the nervous system of Serolis septemcarinata ;
the drawing has been made from a specimen mounted on a slide, but the number of the
ganglia has been checked by comparison with a complete series of longitudinal sections
through an animal of the same size.

The cerebral ganglia are very large, and present the appearance of being composed

I Zoology, Packard, 2nd ed., New York, 1880, p. 307.

20 THE VOYAGE OF H.M.S. CHALLENGER.

of a number of separate lobes ; a pair of connectives unite the cerebral ganglia with
the ventral chain ; the latter is composed of thirteen distinct pairs of ganglia, some of
which bear unmistakable evidence of being the result of a fusion between several pairs
of primitively distinct ganglia ; this is particularly the case with the anterior of the
ganglia, which are seen in longitudinal section to be composed of two fused ganglia.
The eight following ganglia belong apparently to the thorax. Of these the three anterior '
are separated from each other by considerably longer connectives than those which
unite the posterior thoracic ganglia ; the latter get closer and closer together (in
correspondence with the diminished size of the segments to which they belong), and
the last pair of thoracic ganglia are hardly separated from the fused ganglionic mass
belonging to the abdomen ; a careful examination of the latter, especially by the help of
longitudinal sections, shows that it is in reality composed of six pairs of distinct ganglia,
the last of which is considerably larger than the rest, inasmuch as it has to supply the
telson as well as the last abdominal segment. The abdominal ganglia occupy only the
first three segments of the abdomen.

In Serolis necera the fusion of the posterior and of the anterior ganglia is even more
marked.

In the cephalothorax there are three pairs of ganglia, of which the posterior is much
the largest. This is united by a long connective with the next thoracic ganglia, which is
situated at the posterior extremity of the second free thoracic segment ; it is clear, how-
ever, that it belongs to the segment in front, since it was cj^uite easy to trace the nerves
passing forwards to this segment. The fourth thoracic ganglion is closely connected
with the succeeding thoracic and the abdominal oano-lia, the whole formins; an elongated
oval mass lying in the j^osterior thoracic and anterior ten abdominal segments ; the
demarcations between the four anterior pairs of ganglia could be recognised by the stout
nerves given oil" on either side to their respective segments ; of the posterior ganglia,
however, it was impossible to discover how many there were ; a vast number of nerve
filaments take their origin on either side, and afford no indication of the number of
ganglia, two of these passing in a direction parallel to the longitudinal axis of the body
from the hinder end of the ventral chain are especially large ; they supply the telson.

Eye. — With one exception — Serolis antarctica, which is perfectly blind — all the
known species of Serolis are provided with a pair of eyes situated upon each side of
the cephalic shield at about the middle of its antero-posterior diameter ; the eye has
usually a reniform outline, and in general aspect recalls that of the Trilobite, with which
it has been compared ; more rarely it is oval (Serolis hromleyana), and in Serolis minuta
it is almost circular.

In all the shallow-water species the eye is relatively small, l^ut very conspicuous from
the abundant deposition of pigment ; in all the deep-sea forms, with the exception of

' This is not clearly shown in the figure owing to an oversight.

REPORT ON THE ISOPODA. 21

Serolis gracilis, in wliicli species the eye seems to be disappearing (see p. 24), it
is relatively larger but not so conspicuous, owing to the fact that little or no pigment is
present ; these external points of difference between the eyes of the deep-sea and shallow-
water species are accompanied by very striking differences in their minute structure.
Since, however, the shallow-water species agree more closely with other Arthropoda in
the structure of the eye, it will be more convenient to commence with them before
describing in greater detail the deep-sea species.

Among the shallow-water species I have investigated the minute structure of the eye
in the following: — Serolis schythei, Serolis paradoxa, Serolis latifrons, and Serolis cor-
niita. In all there is a fundamental similarity, but there are certain differences in detail.

In Serolis schythei the eye (PI. IX. fig. 2) is facetted externally, and a transverse
section shows that each facet forms a doubly convex lens, the posterior convexity being
more marked than the anterior. Corresponding to each lens is a vitreous body (v) of an
oval conical form, and composed, as in other Isopods, of two halves joined in the middle
line ; as a general rule the vitreous bodies have the even regular form displayed in the
figure, but some are uneven, one half being larger than the other; surrounding the vitreous
body are the remains of the two cells which gave rise to them, and above, between the
vitreous body and the cornea, are two nuclei (" nuclei of Semper ") which belong to
these cells. The whole " vitrella"* is enclosed in a sheath of deep black pigment cells.

Below the vitrella is the retinula, which consists of only four cells ; this is an un-
usually small number ; from Grenacher's work upon the Arthropod eye already quoted,
it appears that five or seven cells is the usual number, and in some cases the retinula
contains a greater number of cells ; Gammarus locusta would seem to be the only known
Crustacean besides Serolis in which the retinula is composed of so few as four elements.
Each of the retinula cells is club-shaped when viewed in front (fig. 2) ; seen laterally they
are hatchet-shaped (figs. 18, 19); at about the middle of the cell is an oval swelling
where the nucleus is situated ; the nucleus is oval in form, the long axis corresponding
to the long axis of the cell, and is contained in a cavity ; in the interior of each nucleus
is a small highly refracting nucleolus. The retinula cells are clothed externally with a
coating of pigment, which is more especially developed at the upper swoUen extremity
and gradually decreases in amount towards the lower end ; this pigment appears to be
contained in long branched connective tissue cells. The retinal cells themselves, however,
do contain intrinsic pigment in addition to this adventitious sheath, as is shown by
transverse sections. Fig. 20 represents a series of sections through a single retinula at
various points which are indicated in the description of plates ; from these sections it
may be seen that the retinal cells contain abundant pigment granules within theii- own

1 The nomenclature used in this description is taken from Grenacher's Untersuchungen liber das Sehoi^an der
Arthropoden, Gottingen, 1879; and from a Jfemoir on the Eyes of Limulus and Scorpio by Prof. Lankester and
Mr. A. G. Bourne (^Quart. Journ. Micr. Sci., N. S., vol. xxiii. p. 177, 1883).

22 THE VOYAUE OF H.M.S. CHALLENGER.

substance, especially towards the circumference. The nuclei of the pigment cells become
conspicuous in teased preparations of the eye depigmented by means of nitric acid.

At the upper extremity of each retinula cell, and lying ujion the inner margin, is a
clear chitinous rod, the " rhabdomere " (fig. 18, r) ; the four rhabdomeres are more or less
closely united to form the rhabdom, which is shown infigs. 2, 5, r ; the lower extremity of the
rhabdom is produced into a fine thread, reaching nearly as far as the pigmented membrane
which bounds the inner surface of the eye. Below this membrane, which is of some
thickness and pigmented only upon its upper surface, the retinal cells are continued into
stout nervous rods which are slightly swollen at the upper end, where they come into
contact with the retinal cells.

The pigment sheath of the retinal cells is continued for a short distance along the
nerve rods ; the latter exhibit transverse markings, and are a little like striated muscular
fibres ; it is very likely that these structures in other Crustacea have contributed to the
erroneous idea that the Arthropod eye jiossesses intrinsic muscular fibres serving as a
focusing apparatus. It has, however, been clearly shown by several investigators that
there are no such muscular fibres present.

So far the eye of Serolis schythei only difi"ers from that of other Isopoda in unim-
portant details. I now proceed to describe another structure which enters into the com-
position of the eye in all the species of Serolis that I have examined, but which has not to
my knowledge been figured or described as occurring in the eye of any other Arthropod ;
this structure consists of two large hyaline bodies situated below the rhabdom, and enclosed
by the upper extremities of the four retinal cells (fig. 2, h). Occasionally only one seems to
be present in a single retinula, and very often the size of the two is unequal, one being
considerably larger than the other (fig. 10). Each of these bodies is clear and transparent,
the substance of which it is composed having very much the appearance of chitin, and I
was at first inclined to think that the whole structure in all probability represented the
rhabdom of other Arthropod eyes. Seeing, however, that a rhabdom is present — though
rather small and inconspicuous — this comparison cannot hold good, and moreover each of
these hyaline bodies shows an irregularly shajjed granular mass, deeply stained by carmine
and other reagents, which would seem to be a nucleus ; it is evidently therefore an indepen-
dent structure and not a product of the retinal cells ; the lower end of the rhabdom is
imbedded in these two cells, and the filiform prolongation of the same appears to pass
through their substance. The large size and transparency of these hyaline cells seems to
indicate that they serve as a dioptric medium. I am not able to say whether these
structures represent highly modified retinula cells or intrusive connective tissue cells,
inasmuch as I have found them already well-developed in the youngest specimens that I
have examined.

In Serolis paradoxa the structure of the eye is in most respects similar, but each of
the cells which compose the retinula secretes in addition to the rhabdomere a small highly

REPORT ON THE ISOPODA. 23

refractive lens-like body or "phaosphere." The sliape of these bodies as well as their
position varies considerably ; sometimes they are quite spherical, occasionally they are
oval, and more rarely bean-shaped ; they are more usually found in front of the nucleus
of the retinula cell, sometimes within its swollen anterior extremity ; often they lie behind
the nucleus, and in a few cases they were observed in close contact to the nucleus and
slightly to one side. The phaospheres of all the four cells of a single retinula in some
instances occupied an approximately similar position, though more generally this was not so.

I only succeeded in making out these structures in Serolis paradoxa ; in Serolis
schythei and Serolis connUa they appear to be absent.

Serolis cornuta. — In this species the cells composing the retinula are longer and
more slender than in Serolis schythei or Serolis i^f^^^'-'-'-loxa, and the pigment upon their'
outer surface is comparatively feebly developed, being chiefly massed round the rhabdora ;
in other respects their structui-e is similar, and very thin longitudinal sections show that
intrinsic pigment is present within the cells themselves in addition to the intrusive
pigmentiferous connective tissue corpuscles which clothe them externall}'. The rhabdom
is a remarkably complicated structure, and differs greatly from the simple rhabdom found
in the two species just described. It is displayed in figs. 3, 9-15, which have been
drawn from preparations teased out in glycerin after having been depigmented by means
of nitric acid and in figs. 7, 8, 16, 17, which represent longitudinal sections of this
portion of the eye undepigmented. In the latter the rhabdom is seen to consist of a
number of coiled threads running in various directions and coated externally with a thick
layer of black pigment. The lower extremity of the rhabdom projects into the " hyaline
cells " and is surrounded by their substance. In fig. 4 is displayed a teased preparation
of the eye of Serolis cornuta undepigmented, and it may be seen that the rhabdom ends
below in blunt rounded prolongations which project into the " hyaline cells " ; the lower
extremity of each of these prolongations — which indicate the composition of the rhabdom
out of four rhabdomeres — is devoid of pigment, so that the rays of light can pass through.
In longitudinal sections of course only one or two of these can be seen, though there are
in reality four.

The shape of the rhabdom can best be made out by teased depigmented preparations ;
fig. 3 represents a single retinula prepared in this way ; the rhabdom is here seen in
profile and resembles an irregular coil of rope projecting below into the hyaline cells
{h) and continued into a thin filamentous prolongation which appears to pass through the
substance of these cells or between them, and extends nearly as far as the lower end of
the retinula cells. Figs. 9-15 represent the rhabdom and the two hyaline cells viewed
in such a position as to bring out the symmetrical character of the former. The form of
the rhabdom, however, differs widely in detail; some of the varieties are displayed in
these figures ; in all it appears to consist of a median unpaired portion which is continu-
ous below with four symmetrically arranged portions which project into the substance

24 THE VOYAGE OF H.M.S. CHALLENGEK.

of the two hyaline cells ; the median portion is usually formed by a sinuous rod of vary-
ing length, which ramifies at its lower extremity into four coils ; the whole structure
reminds one very forcibly of the Malpighian bodies of the kidney ; in some cases the
median rod is double, and more rarely (fig. 13) it is represented by four or five pieces;
it is possible that these varieties correspond to stages in growth, the more complicated
{e.g., figs. 11-13) being the older.

The " hyaline cells," as already mentioned, are two in number in each retinula ; the
nucleus is situated close to the lower extremity ; in sections which have been slightly
depigmented the whole of these hyaline cells becomes tinged of a light mahogany brown,
with the exception of the nucleus, which remains unstained, and is therefore exceedingly
conspicuous ; in, e.g., figs. 16, 17, which are drawn from such sections, it may be seen that
these cells enclose almost the whole of the rhabdom, and this would seem to suggest that
they are concerned in its production and may therefore perhaps be modified retinula cells ;
the cells themselves are quite homogeneous and transparent — apart, of course, from the
nucleus — and in section appear exactly similar to the vitreous body ; they are, however,
more transparent and quite colourless, whereas the vitreous body is always of a chitin-yellow
colour. In a teased preparation of the eye of Serolis cornuta the addition of strong
nitric acid dissolves out the pigment, which is at first absorbed by the hyaline cells,
causing them to assume a very remarkable appearance, represented in PI. X. fig. 1 ; these
(;ells become quite granular with a crenated outer margin and of a dark greyish-black
colour ; wlien the process of depigmentation has gone on a little further the pigment is
dissolved out of the hyaline cells, leaving them perfectly transparent and smooth, or at
most wdth a slight mahogany browm tinge.

Among the deep-sea species of Scrolls the eye is entirely absent in Serolis antarctica ;
its place, however, is occupied by a small tubercle not more than 1 mm. in extent, on
either side of the cephalic shield ; the posterior border of the latter in consequence is
more regular, and the postero-lateral projections which corresj^ond to the eyes are but
slightly marked ; the two tubercles have the same microscopic structure as the rest of
the chitinous integument, and show no traces of facets ; their interior is filled with a plug
of connective tissue, in w^hich no vitreous bodies or retinal elements could be detected.
In Serolis gracilis, another deep-sea form, the eyes present some curious peculiarities.
In two specimens the eyes are well developed, though small, and resemble entirely
the eyes of Serolis hromleyana ; they are conspicuous from their whitish colour,
which contrasts with the surrounding integument ; in one specimen the ocular protu-
berance is only partly occupied by the optic tissues, the rest being exactly similar in
colour to the general integument of the body ; in another specimen there is no trace
whatever of any optic structures, though the oval-shaped elevations of the eye are present
but dark bluish in colour like the rest of the integument. In this species, therefore, the
eye seems to be just on the verge of disappearance.

EEPORT ON THE ISOPODA. 25

Strolis hromleyana. — That portiou of the chitiuous integument that covers the eye
is modified and differs from the rest by its comparative thinness, and by the absence of the
characteristic scale-like sculpturing ; corresponding to each element of the eye is a thicken-
ing of the cornea, which is, however, very slightly developed in comparison witli other
species, and indeed the two surfaces of the cornea appear in transverse section to form two
nearly straight parallel lines, the lower surface alone showing a series of undulations of
unequal extent. PL X. fig. 5, and PI. IX. fig. 2, represent transverse sections through
the cornea of Serolis hromleyana and Serolis schythei respectively ; the conspicuous
corneal lenses of the latter type are iu very marked contrast to the feeble indications
of these structures in Seivlis hromleyana.

The tissues of the eye itself have evidently undergone considerable degeneration, and
this, coupled with the fact that the specimens were by no means well preserved, renders
any satisfactory comparison of their structure with that of the shallow-water species of
Serolis and the Arthropoda generallj^ extremely difiicult. I describe the facts as they
appear to me.

PI. X. fig. 5 is a diagrammatic representation of a transverse section through the eye ;
beneath the cornea are a series of more or less cup-shaped masses of unequal size and of
a granular appearance ; occasionally several of these bodies appeared to have become fused
together at their upper extremity, and in a few instances a short upward prolongation of
the subjacent tissue into the substance of the body gave it the appearance of being
originally formed out of two separate halves ; the granular yellowish coloured matter of
which these structures consist is almost entirely unafi"ected by carmine, which stains deeply
the surrounding tissues, and is only slightly stained by hsematoxylin. In teased pre-
parations of the eye these structures are easily separated, and are seen to have an oval
contour ; from their position they would appear to correspond to the vitreous body, and
in their general characters recall the vitreous bodies of the type of eye termed by
Grenacher " pseudoconous." The compound eyes of the Arthropoda have been arranged
by Grenacher^ into three groups — (l) euconous eyes, (2) aconous eyes, and (3) pseudo-
conous eyes. In the first group the cells lying behind the facets secrete in addition to
it a highly refractive vitreous body or " Kristallkegel," which is composed of as many
segments as there are cells ; in the second group the cells remain unaltered and secrete
no vitreous body ; in the third group these cells secrete " a soft fluid or semifluid
substance " which represents functionally the vitreous body of euconous eyes. It seems
to me very possible that the granular appearance of the vitreous body in the deep-sea
Serolis has been caused by the coagulation (by alcohol) of a semifluid substance.
Pseudoconous eyes, which according to Grenacher are only to be found in the order
Diptera, are further distinguished from euconous eyes by the fact that the nuclei of
the cells of the vitreous body (the so-called " nuclei of Semper ") remain helow the

' Sehorgan der Artliropoden, Gdttingen, 1879.
(zooL. cn.vLL. EXP. — PART xsxiii. — 1884.) Kk 4

26 THE VOYAGE OF H.M.S. CHALLENGEE.

vitreous body itself, instead of being placed above it and in close proximity to the
cornea.

1 was unable to detect with certainty these nuclei of Semper in the eye of Serolis
hromh'i/ana, but in another deep-sea form, Serolis necBra, having an eye of similar
structure they were very conspicuous, and situated above the "vitreous body" between
it and the cornea. Each of the vitreous bodies of Serolis hromleyana is enclosed in a
cup-shaped sheath of tissue (PI. X. fig. 5) in which no trace whatever of any structure could
be made out. This sheath possibly corresponds to the j)igment sheath which encloses the
vitreous body in other Arthropoda, but in Seivlis hromleyana it is entirely free from
pigment. Below this portion of the eye, and corresponding to each of the vitreous
lenses, is a roundish mass of cells which are separated into grou23s by ramifying trabeculse
of pigmented connective tissue ; from the posterior end of this mass of cells a delicate
];)undle of nerve fibres arises which passes back to the ganglion ; it is clothed externally
by a layer of pigmentiferous ramified cells continuous with those in front.

The eyes of Serolis necera resemble in external characters the eyes in the shallow-
water species of Serolis; they are distinctly faceted, and a great deal of pigment is present ;
they are also proportionately larger than in any of the shallow- water species.

The minute structure, however (fig. 3), agrees in the main with that of Serolis hrom-
leyana, but is a little less degenerated ; the cornea, as ah-eady stated, is distinctly faceted ;
beneath each facet are two large and conspicuous nuclei ("nuclei of Semper"), these are
round or oval, sometimes pear-shaped ; they are surrounded by a distinct membrane
showing a double contour, and contain a highly refractive nucleolus. The vitreous
bodies, like those of Serolis hromleyana, are of an irregular oval form and granular opaque
consistency ; the upper half is encircled by a ring of pigment. At the lower end of each
is a roundish mass of small nucleated cells [g), probably nerve cells, and these are con-
nected posteriorly with a nerve bundle, partly sheathed with pigment cells, which extends
into the mass of nerve cells dividing it up into lobules. Fig. 8 is a single element isolated
by teasing in glycerin ; the vitreous body (V) is very distinctly cup-shaped.

The results of my investigations into the minute structure of the eye in Serolis may
be briefly summed up as follows : — (1) The shallow-water s^^ecies invariably possess well-
developed eyes which are fundamentally similar to those of other Isopoda, but differ
in several particulars ; the retiuulge are composed of only four cells ; the rhabdom is
often a highly complicated structure unlike that of other Crustacea. Another element
unrepresented (?) in the eyes of other Crustacea is present, consisting of two large
hyaline nucleated cells placed below the rhabdom and between the retinal cells. (2)
In the deep-sea species the eyes are either altogether absent (Serolis antarctica),
or, if present, show signs of structural degeneration ; no retinula, at least nothing
comparable to the retinula in the eyes of the shallow-water species, is present, but the
vitreous body is represented. The vitreous bodies may be enclosed in a sheath of

REPORT ON THE ISOPODA. 27

pioment {Serolis necera), or there may be no pigment present {Scrolls (jrac'dls, Serolis
hromleyana).

Development. — Several of the species of Serolis contained eggs within the brood
cavity in various stages of development, and although these embryos were not sufficiently
well preserved to admit of any examination by means of sections, I have been able to
note down some developmental facts which have an important Ijearing upon the affinities
of the genus. The development history of Serolis, so far as I have been able to trace it,
is strikingly similar to that of the Cymothoadse, which has recently been worked out in
detail by Mr. J. F. BuUar.^ The Isopoda generally differ from the Amphipoda in that
the embryo has a dorsal flexure in the former group and a ventral flexure in the latter.
Thus Fritz Mliller (Facts for Darwin, p. 71) says — "The curvature of the embryo
upwards instead of downwards was met with by me as well as by Rathke in Idothea
and likewise in Cassidina, Philoscia, Tanais, and the Bopyridse ; indeed I failed to
find it in none of the Isopoda examined by me." And at p. 74 — " The Amphipoda
are distinguished at an early period in the egg by the difierent position of the embryo,
the hinder extremity of which is bent downwards." In the Cymothoadse — at least in
those species which are treated of by Bullar in the paper from which the above quotation
has been made— the embryo is intermediate between the typical Isopoda on the one hand
and the Amphipoda on the other ; it only occupies the ventral surface of the egg, and
does not extend so far towards the dorsal surface as in the Isopoda generally, and the
telson is bent downwards as in the Amphipoda. In Serolis the embryo is exactly similar,
and on PI. X. fig. 7 I have figured a young emluyo of Serolis antarctica; if this
figure be compared with the figures of Ci/mothoa in Mr. BuUar's paper, the close similarity
between the two will be at once apparent.

This fact appears to me to be very strong evidence in favour of Milne-Edwards's view
concerning the zoological position of the Serolidas, which are regarded by him as a division
of the Cymothoadse.^

Post-Embryonic Development. — As in many other Isopoda, the males of Serolis when
first hatched show none of the distinctive secondary sexual characters of males ; the time
at which they assume these characters diflers in difierent species.

In Serolis cornuta the males are at first exactly like the females in general aspect ;
the body is more oval than iu the fully-developed males, being considerably narrower
proportionately; the thoracic appendages of the third pair are, as in the females, similar to
the succeeding thoracic limbs ; the penial filament of the second abdominal appendages is
short, and the sterna of the free abdominal segments resemble those of the female in being

' Phil. Trans., vol. clxix. p. 50"), 1879.
^ Hist. Nat. d. Cnust., 1840, t. iii.

28 THE VOYAGE OF H.M.S. CHALLENGER.

furnished with a median spine. On PI. I. are figures of a female (fig. 2), a fully
developed male (fig. 1), and a young male (fig. 3) of Serolis cornuta, which are drawn
exactly of the natural size ; and judging from the very slight difference in length between
the immature and the fully developed male, it would appear that the male does not
acquire all its secondary sexual characters until the last moult, though the comparatively
greater breadth of the body, which also distinguishes the males of this species from the
females, is recognisable somewhat earlier.

The comparatively small number of specimens (seven) of Seivlis necera makes it
impossible to state with any accuracy the time at which the males assume their proper
sexual characters ; the facts are as follows — -of the four male species two are completely
adult, measuring 42 mm. in length, the remaining two are immature, the largest
measuring 25 mm. in length ; this specimen has the general appearance of a female in its
comparatively short epimera. absence of modification in the third thoracic appendage, and
in all the other secondary characters by which the males of this species differ from the
female, with the exception of the frontal "sense organ," which resembles that found in the
adult males, and is not greatly developed as in the females ; from this it appears that
those secondary sexual characters in which the males of this in common with other
species differ from the female appear comparatively late, while the one sexual character
peculiar to the species is developed early.

I'D. Serolis schythei the males reach maturity much sooner than in Serolis cornuta;
the largest male specimen in the Challenger collection measures 30 mm. in length by
38 mm. in breadth ; the smallest male specimen, with fully developed secondary sexual
characters, is 16 mm. long by 19 mm. broad; another male specimen, in wdiich the-
sex could only be detected by the position of the generative apertures and by the short
penial filament, measures 1 5 mm. in length by 1 7 mm. in breadth ; in this species, therefore,
the secondary sexual characters which distinguish the male appear comparatively early,
and in immature males, which in other respects are outwardly like the females, the greater
proportionate breadth is recognisable ; this character is the first to appear, as also
apparently in Serolis cornuta.

Serolis bromleyana. — In this species the males differ from the females in the
third thoracic and second abdominal appendages, and also in the epimera, which are
shorter in the female and slope downwards at a less angle with the longitudinal axis
of the body. The young males, as in other species, are closely similar to the
females, but acquire the secondary sexual characters peculiar to the males at a
comparatively early period ; the largest male specimen of Serolis hromleyana measures
54 mm. in length, the smallest male, which shows all the secondary characters peculiar to
its sex, measures 25 mm. in length; the largest male specimen, which has liot yet acquired
its proper secondary sexual characters, is 21 mm. in length ; another specimen 22 mm. in
length is nearly adult ; the penultimate joint of the third thoracic appendage is swollen

REPORT ON THE ISOPODA. 29

as in the males, but the terminal claw is not yet reflected back. A single male specimen
of this species was dredged at Station 164a, which presents some interesting peculiarities
in this respect; although only 15 '5 mm. in length, it has all the secondary sexual
characters of an adult male ; the epimera are long and project outwards, the anterior
ones nearly at right angles to- the longitudinal axis of the body, and the tbird thoracic
appendages are fully developed into prehensile organs ; the three middle joints of these
appendages, however, are not provided with the soft sensory hairs which I have noticed
in nearly all the specimens from Station 168 ; the large male specimen from Station 156
is also characterised by the same absence of sensory hairs upon this appendage, and it
appears to me to be just possible that the males of this species are dimorphic, one set
being characterised by the possession of tufts of sensory filaments upon the third thoracic
appendages, and the other set differing by the absence of those structures, and also by the
fact that the individuals acquire their own proper secondary characters at an earlier date.
It is also possible, of course, that this character may be a mere local variation, but in
this case its occurrence in individuals from two such widely separated localities as
Stations 164a and 156 is not a little remarkable, especially when we consider that the
depths which the species inhabit are so different (400 and 1975 fathoms) in these two
localities.

The time at which the females" of Serolis attain to sexual maturity corresponds in
every case to that observed in the case of the males.

In Serolis cornuta only the largest female examples had the ovigerous lamella3 fully
developed, though rudiments of these structures are apparent in many of the smaller
individuals.

In Serolis schythei, corresponding to the early sexual development of the male, the
ChaUeuger collection contains a female specimen only 13 mm. in length, but having
the brood lamellae fully developed and containing eggs.

Oi Serolis hromleyana there is an adult female dredged at Station 168, only 22 mm.
in length ; in this species, as already stated, the males acquire their secondary sexual
characters at a corresponding age.

DESCRIPTION OF SPECIES.

The followdng is a complete list of all tlie species of Serolis tlaat are known to me
either from the specimens contained in the Challenger Collection or from published
descriptions, with the localities and depths at which they occur, and a reference to the
first published descrijition ; they are arranged in order of discovery.

1. Serolis paradoxa (Fabr.) ( = Serolis orhignyi, M.-E.), Mantissa Insectorum ; Patagonia and the Falkland

Islands, 2^ew Zealand (?), Senegal (?), 5 fathoms.

2. Serolis trilohitoides (Eights), Trans. Albany Inst., 1833 ; South Shetlands, Patagonia, a few ffitlioms.

3. Serolis {/audichaudii, And. and M. E., Arch. d. Mus., 1841 ; Valparaiso, a few fathoms.

4. Serolis ^j/fHHW,^ Dana, United States Expl. Exp. — Crustacea, part ii. p. 794, 1852 ; Patagonia, a few

fathoms.

5. Serolis cemvexa, Cunningham, Trans. Linn. Soc. Lend.. 1871 ; Patagonia, a few fathoms.

6. Serolis schythei, Ltk., Vidensk. Meddel. f. d. nat. Foren. i Kjiibenhavn, 1858; Patagonia, 4-55 fathoms.
11. Serolis acutangula, Gr.; Locality?

8. Serolis tuhercnlata, Gr., Archiv f. Naturgesch., 1875 ; Bass Strait.

9. Serolis lati/rons (White), Miers, Ann. and Mag. Nat. Hist., Ser. 4, vol. xvi. ; Kerguelen, Crozets,

Acukland Islands, 1-210 fathoms.

10. Serolis septemcannaf a (Miers) ( = Serolis oralis, St.), Ann. and Mag. Nat. Hist., Ser. 4, vol. xvi. ; Marion

Island, Prince Edwards Island, Kerguelen, 1-150 fathoms.

11. Serolis cornida, Studer, Archiv f. Naturgesch., 1879; Crozets, Kerguelen, 1-120 fathoni.s.

12. Serolis serrei, Lucas, Bull. Ent. Franq.; Patagonia, a few fathoms.

13. Serolis carinata, Lockington, Proc. Calif. Acad. Nat. Sci.; San Diego, California.

14. Serolis Irromleyana, Suhm, Proc. Eoy. Soc. Lond., vol. xxiv. p. 591 ; off east coast of New Zealand,

between New Zealand and Australia, at lat. 37° 53' S., loug. 163° 18' E. (near Antarctic Ice-
Barrier), 410, 700, 1100, and 1975 fathoms.

15. Serolis necera, F. E. B., Proc. Zoo). Soc. Lond., 1884, pt. iii. p. 331 ; off Pdo Janeiro, 600 and 2040

fathoms.

16. Serolis gracilis, F. E. B., loe. rif.; off Pernambuco, 675 fathoms.

17. Sei-olis antarclica, F. E. B., loc. cit.; off Pernambuco, 410 fathoms ; near Crozets, 1375 and IGOO

fathoms.

18. Serolis aiistraliensis, F. E. B., loc. cit.; Australia, 38 fathoms.

19. Serolis elongata, F. E. B., loc. cit; Australia, 30 fathoms.

20. Serolis pallida, F. E. B., loc. cit; Australia, 35-40 fathoms.

21. Serolis longicaudata, F. E. B., loc. cit; Australia, 38 fathoms.

22. Serolis minuta, F. E. B.; Australia, 38 fathoms.

From the above list it will be seen that the new species obtained by the Challenger are
nearly as many as those previously known. Besides the nine new species already briefly

' It seems to me legitimate to alter the termination " v.s " into " a " as done by Grube ; I have accordingly followed
Grube in the te.xt.

32 THE VOYAGE OF H.M.S. CHALLENGER.

noticed by myself and by Dr. v. Willemoes Suhm, there are in the Challenger collection
seven other species which have been more or less fully described by previous writers ;
these are Serolis paradoxa, Serolis schythei, Serolis convexa, Serolis tuherculata, Serolis
latifrons, and Serolis sepiemcarinata ; some notes upon these species will be found below.

Of the remaining six species, Serolis acutanyula is probably identical with some other
form, since Grube, who originally described it in 1871, omits all mention of it in his sub-
sequently published (1875) Monograph of the genus; it is therefore useless to recapitu-
late or make any comments upon his description here. I have quoted the reference to
this description in the list of papers at the beginning of this Eeport. Serolis trilohi-
toides is closely allied to if not actually identical with Studer's Serolis cornuta ; the
figure of this species which is given by Eights is not veiy good,' and the description,
which does not entirely agree with the figure, is not cj^uite complete enough to make it
certain whether this is a good species or not.

Serolis plana certainly comes very near to Serolis convexa, and under the description
of the latter species I have indicated the points of resemblance and difierence between
the two. Serolis gaudichaudii resembles very closely both Serolis plana and Serolis
convexa, but is nevertheless an undoubtedly distinct species ; the points in which it
differs from the two latter are indicated under the description of Serolis convexa. The
two remaining species, Serolis serrei and Serolis carinata, require a fuller description.

The former species is described by Lucas in the following terms : " — " EUe est
plus large que longue, arrondi, testacee et finement tachee de points bruns. Les
six premiers segments thoraciques constituent de tres grandes lames pointues falci-
formes ponctudes, transparentes et non dentelees sur leur hordes comme chez le
Serolis trilohitoides. Le dernier segment abdominal, tricarene non dentele, plus large cjue
long arrondi et legferement sinueux dans le milieu de son bord posterieur presente une
epine assex grande, aigue, de chaque cote de ses bords latero-posterieures. La region
buccale, les pattes ainsi que tout le dessous du corps sont d'un jaune testace. Les
antennes sont testacees maculees de brun." This species, which attains a length of
27 mm. and a breadth of 36 mm., was obtained by the French vessel "Magieienne" in
the Strait of Magellan. Lucas mentions that it is closely allied to Serolis trilohitoides,
but it seems to me, from the above quoted description, that it is probably more nearly
related to Serolis schythei.

It is to be hoped that a fuller description of this species as well as of Serolis carinata
will be published. Serolis carinata, which is one of the most interesting species of the

' Studer {Archivf. Naturgesch., loc. cit., p. 21, note) refers to a supposed error on tlie part of Eights ; the anterior
antennie are said to be represented in Eiglit's figure as if they originated helow instead of above the posterior antenn*.
This is not, however, the case with Eight's figure, but with Audouin and Mihie-Edwards's copy of the figure which is
introduced into their Monograph of tlie genus (loc. cit., pi. ii. fig. 11) ; the absence of shading here does make it appear
as if the anterior antennee originated below the posterior.

' Bull. Soc. Ent. Frang., ser. 5, t. vii. p. cxlv and cxlvi.

REPORT ON THE ISOPODA. 33

genus, on account of its occurrence so far north as California, has been briefly char-
acterised by Lockington ^ as follows : — " Thorax and abdomen conspicuously keeled upon
the centre of every segment ; the first segment slightly waved on its posterior margin,
the curve of the segments increasing rapidly in such a manner that the last entirely
encloses the free abdominal segments on their sides. Caudal segment rounded at the
extremity, with two marginal teeth on each side at a considerable distance from the
extremity, the central carina running the entire length of the segment ; last basal joint
of inner antennae longer than the flagelhmi ; flagellum of outer antenuse much shorter
than either of the two of the preceding basal joints, and last basal joint about equal in
length to the penultimate. Eyes large, reniform, conspicuous. The texture of the
upper surface of a dried specimen under a two-thirds power has a somewhat squamate
appearance. Colour a greyish-brown, diversified with dots and irregular markings of
black ; hands long and slender ; dactylos equal in length to the manus.

"Length, 0-21 inch; width, OIG inch."

Two specimens from San Diego, California.

1. Serolis paradoxa (Fabricius) (PL V. figs. 12-14).

Oniscus paradoxus, Fabricius, Mantissa Insectorum, Hafnise, 1787, t. i. p. 240 ; Systema Ento-

mologiae, Flensburgi et Lipsise, 1775, p. 296.
Cymofhna paradoxa, Fabricius, Entoiiiologia Systematica, Hafniae, 1792^98, Suppl., p. 304,

and t. ii. p. 503.
Serolis fabricii, Leach, Diet. d. Sci. Nat., 1818, t. xii. p. 340.

Serolis orhigmji, Audouin and Milne-Edwards, Arch. d. Mus. d' Hist. Nat., 1841, t. ii. p. 25.
Serolis orhigrmma, Milne-Edwards, Hist. Nat. d. Crust., vol. iiL p. 232.
Serolis paradoxa, Miers, Crustacea collected during Survey of H.M.S. "Alert," Proc. Zool. Soc.

Lend., 1881, p. 61.

Serolis paradoxa is the type species upon which the genus Serolis was founded by
Leach. The animal was first obtained on the shores of Patagonia by Sir Joseph Banks at
the close of the last century, and these specimens were described by Fabricius as Oniscus
jxiradoxus ; in a subsequent work Fabricius altered the name to Cymothoa paradoxa.
In the twelfth volume of the Dictionnaire des Sciences Naturelles, Leach gave some account
of this species which he termed " Serolis fabricii." The first full description of the species
is contained in the memoir of Audouin and Milne-Edwards,' and subsequently Grube^
in his important Monograph gave a fuller account of this same species, but under the
name of Serolis orhigniana. In a short note published in the Journal of the GodeflProy
Museum,* Schmeltz called attention to the identity of Serolis fabricii with another
species described by Audouin and Milne-Edwards under the name of Serolis orbignyi.
More recently Miers, in his account of the Crustacea collected during the voyage of the
" Alert," has also pointed out the identity of these two species.

' Proc. Calif. Acad. Nat. Sci., vol. vii. p. 36. = Loc. cit., p. 25.

3 Loc. cit., p. 225. ' Part .\ii. p. 161.

(zooL. CHALL. EXP. — PART xxxm. — 1884.) Kk 5

34 THE VOYAGE OF H.M.S. CHALLENGER.

I have little doubt that both Schmeltz and Miers are quite right in regai'ding these
two species as identical ; it is impossible, either from the description or the figures con-
tained in the memoir of Audouin and Milne-Edwards, to select any characters by which
the two species Serolis fdbricii and Serolis orhigniana can be definitely distinguished ;
such differences as there are appear to me to arise from the fact that the two " species "
are in reality merely the males and females of the same species, Serolis fahvicii.

With regard to the name I have followed Miers and written Serolis paradoxa ; this
name evidently has the priority.

A number of specimens of Serolis paradoxa were dredged by the Challenger at the
Falkland Islands, and this, together with the adjacent shores of Patagonia, is the only
locality which the species is known wdth certainty to inhabit. According to Leach, Serolis
paradoxa extends to the west coast of Africa ; he makes the following remarks about its
distribution {loc. cit., p. 340) — ■" I have only seen two specimens ; one is in the collection
of Banks and comes from Tierra del Fuego ; the other is iu my possession and was given
me by Dufresne, who tells me that he received it from Senegal." The specimen from
Senegal is now in the British Museum, and is labelled in the handwriting of Leach. The
British Museum contains another specimen of Serolis paradoxa which is labelled " New
Zealand," but I believe that in this case, as in that of Serolis schythei to be mentioned
later, the locality is not thoroughly authenticated. Miers includes Serolis paradoxa in
his list of New Zealand Crustacea apparently on the authority of this same specimen.

The males and females of this species are not distinguishable by any well marked
difference of size ; the males are, however, a little broader proportionately, as is shown by
the following measurements taken from two of the largest specimens obtained by the
Challenger.

Male, .......

Feinale, ......

The form of the abdominal- sterna differs in the two sexes; in the male the outline is
somewhat crescentic, the posterior margin being curved and the anterior margin almost
straight ; in the female each of these segments bears a median spine upon the posterior
margin directed backwards and closely overlapping the succeeding segment. In this
character Serolis paradoxa agrees with Serolis schythei, but the spines in the females
are longer than in that species. The alcove measurements do not express the greatest
size to which this species attains ; there are several specimens in the British Museum
somewhat larger, and White states ^ that individuals have been known to reach the
great length of 6 inches ! This last statement, however, requires confirmation.

Serolis paradoxa is, as Grube remarks,^ more nearly allied to Serolis schythei than to

^ Wliite, Description of New Species of Insects and other Aunulosa, Ann. and May. Nat. Hist., 1843, vol. xii. ]>. 366.
^ Loc. cit., p. 225.

Length.

Breadth.

24-5 mni.

2.5 mm.

27-5 „

26 „

REPORT ON THE ISOPODA. 35

any other species, but it can easily be distinguished by the smaller epimera and by the
shape of the caudal shield, which is more triangular than in that species ; as in Seroh'i:!
schythei, the caudal shield is traversed by three longitudinal carinse, and there is a stout
spine close to its anterior border in the middle line ; another carina runs parallel to
the anterior margin of the caudal shield, of which traces exist in Serolis schythei. The
transverse ridge of the latter is not found in Serolis paradoxa. As in Serolis schythei,
the first four free thoracic segments have their epimera separated from the terga by a
distinct suture.

The antero-Iateral areas of the cephalic shield are more extensively developed in
Serolis paradoxa, so that the breadth of the anterior portion considerably exceeds that of
the posterior portion, while in Serolis schythei the transverse diameter of the cephalic
shield is greatest at about the level of the eyes.

Antennce. — The filament of the second pair of antennae in both the male and female
is furnished with a series of short recurved hooks ; these hooks are not found upon all the
joints of the filament, they commence to be visible (in one specimen) on the seventh,
and extend as far as the sixteenth ; in the last nine or ten of these joints there is a
continuous row on the upper surface not far from the inner margin, commencing close to
the posterior articulation and extending in a straight line up to the anterior articulation,
the row of hooks then bends inwards and follows closely the anterior edge of the
joint, terminating at its outer extremity; in the anterior joints the hooks gradual^
decrease in number until in the seventh joint there is only a short row on the anterior
margin.

Similar structures are found in other species {e.g., Serolis bromleyana, Serolis gracilis,
Serolis septemcarinata, Serolis schythei), though in no case are the hooks so large and
conspicuous as in Serolis paradoxa.

The inner lobe of thejirst maxilhe is nearly half the length of the outer lobe ; it con-
sists (PI. V. fig. 14) of a straight and narrow stem and an oval wider distal portion
furnished with a single spine near the lower border.

The second pair of maxillce has, as in all other species, two smaller anterior lobes and
a larger posterior one ; the anterior lobe is a trifle smaller than the second, and bears
at its free extremity two long spines ; the second lobe has five or six, while the largest
posterior lobe has some twenty or thirty.

The maxillipedes are characterised by the almost square outline of the stipes, which
is quite twice the breadth of the lamina ; the latter is clothed on the inner surface -with
fine hairs; the suture which separates it from the stipes is not complete posteriorly.

The third pair of thoracic appendages in the male (fig. 13) are very much like those
of Serolis necera ; the second, third, and fourth joints are furnished with abundant
plumose hairs arranged in two rows, one above the other ; the upper row spring from
a strong ridge which runs at a short distance from the inner margin of these three

36 THE VOYAGE OF H.M.S. CHALLENGER.

joints, while the second row appears to be attached along the inner margin itself ; the
penultimate joint is swollen and pear-shaped, broader at the base than above ; along the
inner margin are some fifteen pairs of long cylindrical spines, which are of a rather
unusual form ; the distal third bears two rows of short cyhndrical branches ; the centre
of the spine appears to be hollow at the base, and the bounding membrane is continued
above into a solid axial rod, which projects for a short way beyond the anterior extremity
of the spine. This appendage and the peculiar spines upon the penultimate joint are
figured by Grube (loc. cit., pi. v. figs. 3, 3, a), Grube also states that the spines upon
this pair of appendages are alike in both males and females : — " Der Innenrand des 2**°
Paares Greiffusse tragt lange an der Vorderseite gesagte Stacheln, sowohl beim Mannchen
als beim Weibchen." They are in reality quite different in the female ; the fourth joint
bears a row of spines arranged in pairs ; on the distal half of the joint these spines are
strong and serrated along both margins, resembling entirely the serrated spines found
upon the other thoracic limbs ; posteriorly the serrations become less marked, and the
proximal end of the joint has only smooth sword-like spines, the third and fourth joints
have none of the soft plumose hairs that are found in the males ; their place is occupied
by tufts of strong spines, some of which are serrated. In this as in all other species in
which the males have the peculiarity here desci'ibed, the females are invariably different,
the third pair of thoracic appendages being in this respect closely similar to the succeeding
appendages.

The remaining thoracic appendages are not in any way remarkable. The second joint,
which is rather smaller than the first joint, is furnished with a tuft of long hairs upon the
inner surface at the distal extremity, among which are a few serrated spines ; the succeed-
ing joints, which are approximately of this arm length, the fourth being rather the longest,
have a tuft of hairs in the same position, among which are a greater number of serrated
spines ; in the penultimate joint the long delicate hairs are almost entirely absent, their
place being taken by stout spines, many of which are serrated. Along the outer margin
of the joints, with the exception of the first, are tufts of short spines, very few of which are
serrated.

The three anterior abdominal appendages, like those of Serolis schythei, have no
plumose hairs upon the basal joint.

The suture of the operculum is oblique.

The exopodite of the fifth pair of abdominal appendages is bifurcate.

Station 316, February 3, 1876 ; lat. 51° 32' S., long. 58° 6' W.; 4 fathoms; bottom,
mud.

REPOET ON THE ISOPODA. 37

2. Serolis convexa, Cunuingham (PI. VI. figs. 9-15).

Serolis cont'eaa, Cunningham, Crustacea, &c., of H.M.S. " Nassau," Trans. Linn. Soc. Lond.,

vol xxvii. p. 498, 1871.
Serolis cnnoexa, Studer, Isopoden gesammelt, &o., Abhandl. d. k. Akad. d. Wiss. Berlin, 1882, p. 9

(separat Abdriick).

The Challenger obtained only a single specimen of this species, from Port AVilliam,
Falkland Islands ; I have, however, had the opportunity of comparing it with four other
specimens in the British Museum which are the types.

Prof Cunningham's description {Joe. cit, p. 498) is as follows : — " Most nearly allied
in form to Serolis planus, Dana. Body very convex, much more so than in any other
species of the genus with which I am acquainted. A prominent carina along the mesial
line. Two last segments of pereion distinctly narrower than those which precede them.
Last segment of pleon more elongated than in Serolis planus, and possessing three well-
marked ridges — a median, interrupted in the middle, and two lateral, each terminating in
a sharp point. Eyes very prominent, and placed rather near to each other."

Two specimens were obtained by Prof. Cunningham on the north coast of Fuegia.

Prof. Studer has also published some notes upon this species in his account of the
Isopoda collected during the voyage of the " Gazelle," and he has pointed out that the
chief character by which Grube distinguished Serolis convexa from Serolis gaudichaudii,
viz., a tuft of hairs upon the antepenultimate joint of the second thoracic appendages,
is really a secondary sexual character peculiar to the males not only of Serolis convexa
but also of Serolis gaudichaudii ; before Prof Studer's paper came into my hands I had
noted the same fact, and alluded to it briefly in my Preliminary Eeport;^ the speci-
men dredged by the Challenger is a male, and it shows another character which I believe
to be peculiar to the males ; on the sterna of the last three thoracic segments is a small
patch about 2 mm. in length by 1 mm. in breadth, distinguishable by its velvety appear-
ance and darker colour from the surrounding integument ; this on microscopic examina-
tion proved to be covered by a multitude of hairs exactly similar to those found upon the
second thoracic appendages, only rather shorter ; some of the segments in front also,
especially the second, have a small tuft of these hairs.

In other respects the males and females of this species do not difli"er greatly in appear-
ance ; the males are imperceptibly broader in proportion to their length than the females,
as shown by the following measurement taken from the male in the Challenger collection
and the single female specimen in the British Museum: —

Male, .......

Female, .......

The sterna of the abdominal segments present no diiferences in the two sexes.

1 Proc. Zool. Soc. Lond., pt. iii., 1884, p. 330.

Length.

Breadth.

25 mm.

20 mm.

25 „

19 „

38 THE VOYAGE OF H.M.S. CHALLENGER.

The suture dividing the outer lamella of the opercida is accurately at right angles to
the long axis of the body in the female ; it is more oblique, slanting downwards in all the
male specimens that I have seen ; I am inclined, however, to think that this is rather an
individual variation than a mark of sex, inasmuch as nothing of the kind exists in any
other species.

Serolis convexa has been figured by Cunningham ^ and by Studer,^ and as both these
figures, especially the latter, show plainly the characters of the species, I hardly think it
worth while to introduce another into the present Eeport.

Serolis convexa can be easily distinguished from Serolis gaudichaudii by a number
of small characters ; the shape of the body is more oval in Serolis gaudichaudii ; in Serolis
convexa it is more pear-shajaed, owing to the greater length of the caudal shield, which
terminates in a more pointed extremity ; in both species there is a central and two lateral
carinse ; the latter are curved, and follow closely the lateral margin of the caudal shield
but at some distance from it ; in Serolis gaudichaudii these carinse, especially the two
lateral ones, are very faint ; in Serolis convexa the lateral carinse are strongly marked, and
terminate more or less abruptly in a short spiniform projection at the end of the middle
third of the caudal shield ; the median carina is only found in the anterior part of
the caudal shield, posteriorly it becomes obsolete ; another diSerence between the two
species is in the colour, and since this difference is quite constant in all the specimens
examined by me (five of Serolis convexa, four of Serolis gaudichaudii) it is worth noting ;
Serolis gaudicliaudii is distinguished by its dark brown colour, darker in the central
part of the body, and dotted all over with black spots of various sizes ; in Serolis convexa
the colour is of a uniform pale brown, hardly darker in the middle of the body than upon
the epimera. This description of course relates only to specimens preserved in alcohol.
Another species closely allied to Serolis convexa is Dana's Serolis plana, and from a
careful comparison of the figures and descriptions given of these two species I find it
almost impossible satisfactorily to separate them. The general shape of the body, the form
of the epimera, &c., is almost identical in the two. The eyes, however, ot Serolis plana are
stated by Dana to be conical in shape, whereas those of Serolis convexa, as in all other
species, are distinctly reniform. In Serolis plana " the articulation of opercular plates is
more nearly transverse than in Serolis gaudichaudii;" there is also a lateral tooth on the
caudal shield just below termination of carina ; the median carina is obsolete posteriorly,
and the fourth segment has a low jjrominence just inside of the epimeral suture. The
first character is, as I have already shown, of no use in distinguishing the species ; the last
character, namely, the presence of a low prominence to the inside of the epimeral suture
of the fourth segment, is well marked in the Challenger specimen of Serolis convexa ; in
this specimen all the thoracic segments have a slight prominence, almost indistinguishable
in the anterior and posterior ones, but conspicuous in the fourth segment, where it slightly

1 Loc. cit., pi. ILx. fig. 3. 2 Isopoden gesamiuelt, &c., he. cit., Taf. i. figs, la, 16.

REPORT ON THE ISOPODA. 39

overlaps the succeeding segment and recalls Serolis minuta, in which species (see below
p. 77) all those prominences are highly developed and project backwards as flat conical
processes over the following segments.

The presence of a distinct lateral tooth below the termination of the carinse does
seem to distinguish Serolis plana from Serolis convexa where there is no such tooth, but
it would not be safe, I think, for the present to separate the two until a larger number of
specimens of both have been examined.

In all the three species — Serolis gaiidichaudii, Serolis convexa, and Serolis plana —
the epimera of the second, third, and fourth pairs only are separated from the terga by
a distinct suture ; in the first two species the anterior epimera are divided by a
transverse suture as in so many other species ; this suture is, however, not very well
marked, and in the specimen of Serolis convexa obtained by the Challenger I could not
detect an 3^ trace of it, though it was distinctly visible in the British Museum specimens.

Appiendages. — I give here a somewhat full description of the appendages which have
not been described by Cunningham.

Antennce. — The posterior pair of antennte are decidedly longer than the anterior pair,
reaching as far as the end of the first epimera.

Of the anterior pair, the first joint is somewhat broad and quadrangular, the second
joint is narrower and a little longer ; the anterior and posterior surface of both are covered
by a few scattered hairs ; the third joint is a cylindrical rod somewhat bow-shaped, and
only slightly longer than the preceding joint ; the fourth joint is short, about one quarter
as long as the third. The filament appear to consist of about twenty-five joints, and
each joint has two sensory hairs, one situated near the anterior extremity of the joint, the
other near its posterior extremity.

The posterior i^air of antennae consist as usual of five joints and a filament.

The first joint is the shortest, the second a little longer and somewhat hour-glass
shaped. The third, as usual, inserted in a wedge-like fashion between the second and
the fourth ; the fourth joint is the longest of the whole appendage, is somewhat bent, and a
ridge parallel with the curved lower margin runs along its whole length ; the joint narrows
towards its extremity, where it articulates with the terminal joint, which is a little shorter
and more uniformly cylindrical in shape ; the upper margin of the two last joints is
furnished with scattered tufts of hairs. The filament has twenty-one joints, which
increase in length but diminish in thickness towards its extremity.

The mandibles have the cutting edge prolonged posteriorly into two subequal
conical teeth ; on the upper side are two spines situated one behind the other, the
anterior one rather longer and broader.

Thejii'st maxillcB are in no way remarkable.

The second maxillce (PI. VI. fig. 14) differ from those of most other species in that the
two anterior lobes are not much smaller than the posterior, and are furnished on the outer

40 THE VOYAGE OF H.M.S. CHALLENGER.

edge with eight or nine hairs instead of the usual two. In this respect Serolis gaudichaudii
seems to resemhle Serolis convexa{cf. Audouin and Milne-Edwards, loc. cit., pi. i. fig. 11).

The maxilUpedes (fig. 15) are of the usual shape; the lamina is furnished with a
number of long hairs upon the lower surface, and the second joint of the palp has a row
of similar hairs upon the lower surface, some way from the external margin, which, as
usual, has another row of hairs. The third joint of the palp is extremely small, and I was at
first inclined to think that this was a mere accident, and that the terminal joint of the
palp had been recently broken oS", but since Milne-Edwards and Audouin figure ' the same
condition in Serolis gaudichaudii, it appears to be characteristic of these two species.

The second pair of thoracic appendages are figured on PI. VI. fig. 1 0 ; the tuft of
hairs (a) have already been noticed as peculiar to the males ; several of the spines forming
the inner surface of the penultimate joint are more highly magnified in fig. 11 ; the
longer spines, instead of being bifurcate, as is the rule, terminate in a gradually narrowing
extremity, the posterior margin of which is beset with a dense row of fine branches.

The third pair of thoracic appendages of the male are shown in PL VI. fig. 9 ; the
terminal joint is comparatively narrow ; from the inner margin arise some eight pairs
of conical spines, the axis of which projects beyond the termination as a short filamentous
recurved process.

The remaining thoracic appendages were all more or less damaged, with the exception
of the sixth, which is figured on PI. VI. fig. 12 ; the second, third, fourth, and fifth joints
of this appendage are each furnished on the inner side wdth a tuft of fine pinnate hairs
like those on the second pair of appendages in the male ; on the distal border of the fourth
joint is a row of sharp serrated spines ; the terminal claw-like joint of the appendage is
sharply bent upon itself at right angles ; in Serolis gracilis and other species those
appendages in the males are similarly modified.

In the anterior abdominal appendages the basal joint is triangular in shape, and the
projecting (lower) angle is furnished with three plumose hau's in the first of these
appendages, and two in the second and third as in many other species.

The opercida are traversed by a suture at right angles to the long axis.

Port William. Falkland Islands, 5 to 12 fathoms.

3. Serolis schythei, Liitken (PI. II. figs. 5-13).

Sei-olis schythei, Liitken, Vidensk. Meddel. f. d. nat. Foren. i Kjpbenhavn, 1858, p. 98, Tab. i.

figs. 12, 13.
Serolis schythei, Grube, Arcbiv f. Naturgesch., 1875, p. 220, pi. v. fig. 1, pL vi. fig. 1.

This species was originally described from a male specimen by Liitken. Grube in his
Monograph of the genus added some details to Liitken's description, and pointed out the
difi"erences that exist between the males and the females.

' Tab. cit., Am. 12, 13.

REPORT ON THE ISOPODA. 41

The Challencer obtained a great number of specimens of Serolis schythei, ranging
from the Gulf of Peuas on the north, to the Falkland Islands on the south.

The collection of Crustacea in the British Museum contains one specimen which is
labelled "New Zealand," but I believe that it is not quite certain whether this label be
correct. With this possible exception, Serolis schythei is confined to the shores of
Patagonia and the Falkland Islands.

The males of Serolis schythei differ from the females by their greater proportional and
actual breadth ; the largest female specimen in the Challenger collection measures 32 mm.
in length by 32 mm. in greatest breadth, the largest male specimen measures 30 mm. in
length by 38 mm. in breadth. The greater breadth of the males is owing to the greater
leno-th of the epimera. The sixth epimeron in this male specimen measures 17 mm. in
leno-th, and extends for the space of about 5 mm. behind the extremity of the caudal
shield; in the female these epimera only measure 13 mm., and terminate just beyond
the end of the caudal shield. The first abdominal epimera also are longer in the male than
in the female ; the two sexes also difier, like many other species, in the shape of the
abdominal sterna. A third diff'erence is to be found in the antennae, which are described
below.

The colour of Serolis schythei (in alcohol) is pale brown, becoming darker in the
middle of the body, and diversified with dark brown or black spots, which vary considerably
in their number and size in different specimens.

The cephalic shield is not unlike that of Serolis cornuta, but the prominences and
ridges upon its surface are less developed ; there is a slight ridge dividing the antero-
lateral portions from the rest ; to the inside of and behind the eyes are two prominences
somewhat triangular in form, which correspond to those in Serolis cornuta, but are
directed more inwards ; between the two is a triangular area which forms the posterior
boundary, and owing to the comparatively slight development of the lateral tubercles,
lies almost on the same level with them ; the disposition of these three prominences is
also like w^hat occurs in Serolis newra, which in other respects shows considerable resem-
blances to this species. The eyes are very large and prominent.

Thorax. — The thoracic epimera are flat and sickle-shaped ; as in Serolis cornuta the
external half of all the epimera projects freely ; the articular processes upon the epimera
are, however, entirely absent. The first epimera are divided by a transverse ridge.
While in the majority of species only the three anterior of the free thoracic segments are
divided by a distinct suture into a tergal and an epimeral portion, in Serolis schythei there
is also a suture separating the terga and epimera of the fourth segment. The posterior
margin' of the tergura of each segment, with the exception of the first, is curved backwards
in the middle line into a short spine ; these gradually diminish in size from before back-
wards ; the antero-posterior diameter of the first free thoracic segment is a little less
than that of the succeeding segment, which is the largest, and measures 3 mm. in the

(ZOOL. CHALL. EXP. — PART .XiXIII. 188-t.) Kk 6

42 THE VOYAGE OF H.M.S. CHALLENGEE.

female and 2 mm. in the male ; the third segment is a little smaller than the second in
the female and about the same size in the male ; the two last thoracic segments are nearly
the same size, the anterior l^eing a trifle the larger, and measuring about one half the
diameter of the preceding segment.

On the under surface the sterua of the first two thoracic segments are, as in all other
species, divided by sutures into three portions, which resemble in every particular those
of Serolis cornuta ; the posterior thoracic segments are not so fused together as they often
are ; the sixth is quite distinct from the following ones, and is not divided into a central
piece and two wedge-shaped lateral pieces as it is in Serolis necera and other species ; a
deep groove separates the seventh from the eighth, but it does not extend as far as the
suture which marks the boundary between the epimera and sterna.

Abdomen. — The first abdominal epimera are long, and reach beyond the end of the
caudal shield in the male, in the female they do not reach quite so far as the end of the
caudal shield ; the second epimera are short, as in Serolis newra, and only extend for a
very short distance down the lateral margin of the caudal shield ; they are not distinctly
longer in the male than in the female.

The sterna of the abdominal segments differ in the two sexes as already said ; in the
male the posterior margin is straight, and the two ends project backwards as short spines;
the third segment has, in addition, a short median spine ; in the female the median portion
of all three is produced into a short broad spine.

The caudal shield is hexagonal in outline ; the posterior end projects as a short spine ;
there is a distinct median keel and two lateral keels which start from the base of a sti'ong
blunt spine situated in the middle line at the anterior end, and terminate in two short
spines placed some way in front of the attachment of the uropoda ; a transverse ridge,
which is prolonged backwards into thi'ee short spines, of which the median larger one is
upon the longitudinal carina, traverses the caudal shield ; viewed in profile, the caudal
shield appears to cousist of two portions bounded by this transverse ridge ; the anterior
part, which possibly corresponds to the three fused terminal segments of the body, over-
laps the posterior portion or telson ; on the other hand it is possible, as suggested by
Studer, that the two obliquely running longitudinal keels mark the boundary between
the terminal segment of the body and the telson.

The two pairs of antennce are of about the same length. The anterior pair
have a short proximal joint with fine hairs upon the upper surface ; the two succeed-
ing joints are elongated and somewhat curved ; the posterior surface is furnished with
fine hairs ; the fourth joint is short and oval ; the filament is made up of about twenty-
three joints, of which the proximal ones are short, but gradually increase in length towards
the distal extremity. The second pair of antennae consist, as usual, of five joints and a
terminal filament ; the joints increase in length towards the distal end of the antennae,
the last joint being the longest and considerably narrower than the rest, which are about

REPORT ON THE ISOPODA. 43

equal iu diameter; the three last joints are beset with hairs arranged in tufts or scattered
singly over the surface ; the filament is made up of from fifteen to seventeen joints,
which, like those of the anterior antennae, increase in length and diminish in width
towards the free extremity ; the inner side of each joint, just before the attachment of
the joint in front, has a tuft of fine hairs ; the anterior margin of the middle joints has a
series of short curved spines like those already described in Serolis paradoxa (ante, p. 35) ;
the antennae of the males possess a series of delicate lamellar processes (PL II. fig. 6)
arranged in a single line along the inner side of most of the joints, being absent only
from a few of the extreme distal and proximal joints ; these structures, which are very
probably sensory, take the form of oblong lamellae with rounded angles, those at the
anterior part of the joint are larger and somewhat fan shaped ; their surface is marked by
a series of grooves radiating outwards from the base of attachment.

The mandibles (PI. II. figs. 12, 13) are markedly asymmetrical, as in other species.

The maxillcB present no peculiarities.

In the maxillipedes (fig. 10) the stipes and the lamina are not separated by a
suture ; both are covered with scattered hairs, and a row of finer hairs clothes the inner
margin of the latter ; at the summit of the terminal joint of the palp is a small
protuberance carrying three stoutish hairs similar to those upon this and the preceding
joint.

The j^rs^ pair of ambulatory appendages are figured by Grube in his Monograph, but
the hairs on the inner side of the penultimate joint are not quite correctly shown ; the
longer spines (PL II. fig. 7) terminate in two flat lateral expansions, and the axis of the
spine is continued into the posterior and longer of the two.

The second pair of ambulatory a.p)pendages of the male (PL II. fig. 8) are modified
in the usual way {ante,, p. 13).

In the remaining thoracic appendages (fig. 9) the two first joints are subequal in
length ; the third joint is about half the length of the second, the fourth slightly longer,
the fifth shorter than the fourth, and approximately of the same size as the second ; the
terminal joint is more bent and hook-like in the last pair of appendages than in any of the
others. The second joint (in the males only) has a series of about fifteen tubercles close to
the inner margin ; the third, fourth, and fifth joints are furnished with numerous long and
slender hairs as well as stouter sword-like and serrated spines arranged in tufts in the
usual way ; the last pair of appendages in the male are distinguished from the rest by
a greater development of hairs upon the inner surface of the terminal joints ; this is an
approximation to the marked diS"erence that these appendages show in other species
{e.g., Serolis gracilis). In Serolis schythei, however, these hairs are not, as in the above-
mentioned species, in any way diflerent from those upon the rest of the ambulatory
limbs.

The three anterior abdominal appendages are characterised by the absence of

44 THE VOYAGE OF H.M.S. CHALI,ENGER.

branclied hairs upon the basal joint ; in this they agree -n-ith Serolis paradoxa and the
Australian species of the genus.

The opercuJa have the exopodite divided by an oblique suture.

The exopodite of the gill appendage is bifurcate.

The uropoda are attached close to the termination of the caudal shield, and extend
a little way beyond it ; they are comparatively large and foliaceous, and furnished with
numerous branched hairs.

Station 304, December 31, 1875; lat. 46° 53' S., long. 75° 12' "\V.; 45 fathoms
bottom, green sand.

Station 313, January 20, 1876; lat. 52° 20' S., long. 67° 39' W. ; 55 fathoms
bottom, sand.

Station 314, January 21, 1876; lat. 51° 35' S., long. 65° 39' W. ; 70 fathoms
bottom, sand.

Station 315, January 26, 1876; lat. 51° 40' S., long. 57° 50' W. ; 12 fathoms
bottom, sand, gravel.

Station 316, February 3, 1876 ; lat. 51° 32' S., long. 58° 6' W. ; 4 fathoms ; bottom,
mud.

4. Serolis latifrons, White (PI. II. figs. 1-4).

Serolis latifrons, Wiite, List Crust. Brit. Miis., 1847, p. 106.
Serolis latifrons, Miers, E. J., Ann. and Mag. Nat. Hist, 1875, p. 74.
Serolis latifrons, Miers, E. J., Cat. New Zealand Crust., 1876, p. 117.
Serolis latifrons, Smith, Bull. TJ. S. Nat. Mus., 1876, vol iii. p. 63.
Serolis latifrons, Miers, E. J., PhiL Trans., 1879, extra vol., p. 204.
Serolis latifrons, Studer, Th., Archiv f. Naturgescli., 1879, p. 26.

This species was named and described by White from specimens obtained during the
voyage of the "Erebus" and "Terror"; one of these specimens is preserved in the
British Museum, and was dredged at Rendezvous Cove, Auckland Islands. Miers first
described the species from specimens, brought home by the Transit of Venus Expedition
from Kerguelen Island, and a more detailed description of the same species is to be
found in Studer's Beitrage zur Kenntniss niederer Thiere von KerguelensJand.^

A number of specimens were dredged by the Challenger at Christmas Harbour,
Kerguelen, 25 fathoms, and two others which present some differences from the t^'pical
form were dredged in deeper water (210 fathoms) off Possession Island.

Without entering into any systematic description of this species, which has already
been sufficiently done by the above mentioned authors, I may add here a few details.

Studer describes the largest specimens as attaining a length of 40 mm. and a breadth
of 30 mm. ; these are evidently females, which he says are 1 to 2 mm. less than the

' Archiv f. Naturgesch., loc. cit., p. 26.

EEPORT ON THE ISOPODA. 45

males. In the Challenger collection the two largest specimens measure 32 mm. in length
by 24 mm. in breadth, and are females with developed brood lamelte. The males are
proportionately broader than the females ; the length of the largest specimen is 28 mm.,
its breadth 24 mm. ; in one specimen only the breadth was proportionately less than in
all the rest, 28 mm. in length by 22 mm. in breadth ; the sterna of the free abdominal
segments differ but slightly in the two sexes — in the females all have a short, broad,
median spine, in the males the two posterior segments are pro^dded with such a spine, which
is if anything rather smaller than in the female.

The colour of Serolis latifrons (in alcohol) is a dark bluish brown, becoming lighter
brown upon the epimera ; a very characteristic mark of the species is a white spot upon
the anterior part of the cephalic shield, just behind the rostrum ; the fifth pair of epimera
also are generally lighter in colour than the rest of the body.

The six posterior thoracic epimera are separated from the terga by a distinct suture
and the first abdominal segment, which in no other species of Serolis known to me has
any trace of epimera, has distinct though very minute epimera separated from the tergum
by a suture which is continuous with that dividing the epimeron and tergum of the
segment in front.

Appendages. — The second pair of antennae are longer than the first pair ; the latter
have as usual four joints; the fourth joint is extremely short, its length being only one-
third of its breadth ; the filament has sixteen joints ; in the second pair of antennae the
terminal joint (PI. II. fig. 4) is prolonged on the ventral side into a j)rocess which extends
as far as the third joint of the filament ; the filament has twelve joints.

The mandibles are characterised by the great length of the narrow distal half, which
is about half the length of the entire appendage; the masticatory portion of the two
mandibles are shown on PL II. figs. 2, 3 ; they present the usual unsymmetrical disposition
of the laminae and spines upon the lower surface.

The Jirst pair of maxillce resemble the same appendages in other species, but the
second joint of the palp has seven or eight spines upon its extremity instead of two.

The second maxillce present no difi"erences from other species.

Maxillipedes. — The lamina and stipes are separated by a complete suture ; the
outer margin of the stipes and cardo is clothed with fine delicate hairs ; the lamina is
furnished on its under surface with a number of long hairs which extend over the
proximal joint of the palp.

The first ambulatory limb presents no special difierences from that of other species.

Of the second pair in the male, the penultimate joint is rather narrow, only slightly
wider where it articulates with the preceding joint; the distal joint terminates in a
spine-like extremity which is separated by a suture; just behind the articulation of this
spine is a ridge bearing a row of fine hairs. The remaining joints of this Hmb are
almost smooth, and have only a few short hairs.

40 THE VOYAGE OF H.M.S. CHALLENGER.

The remaining thoracic appendages, on the contrary, are abundantly furnished with
long, unbranched hairs, and short, stout, serrated, and smooth spines ; the latter are
especially developed at the distal extremity of each joint just before its articulation with
the succeeding joint.

The proximal joint of the limb is the stoutest, and except in the two penultimate
limbs, the longest, its margin is clothed with fine delicate unbranched hairs ; the
second joint is shorter, and the third shorter still ; the fourth and fifth joints in the
posterior limbs are considerably elongated ; the distal extremity of the terminal joint is
separated ofi"by a suture, and ends in a sharp spine.

The abdominal a2:'pendages present no peculiarities except the terminal uropoda, which
have been well described and figured by Studer ; ^ the endopodite becomes fused with the
protopodite, and is extraordinarily elongated ; the short exopodite is sj)ine-like, and can
be fixed at right angles to the rest of the appendage, forming in this way a defensive
apparatus, which, as Studer suggests, no doubt secures to the animal a greater freedom
from the attacks of birds and other enemies. B'or figures and fuller description of
this structure the above-mentioned Memoir by Studer should be consulted. The
structure of the uropoda in this species recalls that characteristic of the family
Sphaeromidse.

Two specimens of Serolis latifrons dredged ofi" Possession Island in 210 fathoms
present some variations from all the other specimens ; they are both females, with fully
developed brood lamellae, and are approximately of the same size, measuring 28 mm. in
length by 23 mm. in breadth ; the proportions therefore of these specimens are nearly the
same as in the males of the typical Kerguelen form ; the sterna of the abdominal segments,
however, are quite like those in the females. In one specimen the posterior two pairs of
brood lameUse of the left side overlap those of the right, the two anterior pairs are normal.
This variety is quite difi"erent in colour, being of a uniform yellowish brown, with a few
scattered black spots, especially upon the caudal shield. The shape of the caudal shield
is difi'erent from that in the Kerguelen specimens ; it is considerably narrower, and the
lateral margins are almost straight, with only a very slight outward bulging ; the posterior
epimera of the body come therefore to project considerably beyond the margin of the
caudal shield, which is almost completely triangular in shape.

One of the specimens is figured on PI. II. fig. 1, and may be compared with the
figures of Studer and Miers of the normal variety.

Station 148, January 3, 1874; lat. 46° 47' S., long. 51° 37' E. ; 210 fothoms ;
bottom, hard ground, gravel, shells.

Kerguelen Island, January 13, 1874 ; Cascade Bay, 5-40 fathoms. January 17,
1874 ; Royal Sound, 25 fathoms.

1 Archivf. Naturgesch., 1879.

REPORT ON THE ISOPODA. 47

5. Serolis septemcarinata, Miers (Pis., 11. fig. 14, VIII. figs. 3-5).

Serolis qitadricarinata, White, List Crust. Brit. JIus., 18-17, p. 106.
Serolis septemcarinata, Miers, Ann. and Mag. Nat. Hist., 1875, p. 116.
Serolis septemcarinata,'^ Miers, Phil. Trans., 1879 (extra vol.), p. 206.
Serolis ovalis, Studer, Archiv f. Naturgesch., 1879, p. 24.

That Serolis septemcarinata is the same species as Serolis ovalis cannot be doubted ;
the descriiitions and figures given by Miers and Studer agree in every detail.

Both these observers described the species from a single specimen only ; Studer's
description relates to a female specimen from Kerguelen, while Miers's specimen was
dredoed at the Crozets.

The Challenger obtained this species at Kerguelen, and also off Prince Edward's
Island ; the species therefore is common to these three groups of Antarctic Islands, but
has not as yet been obtained elsewhere.

Studer remarks the rarity of this species as compared with Sewlis latifrons, and
suggests that the absence of the spine-like uropoda found in Serolis latifrons may render
it an easier prey to its enemies ; the Challenger, however, collected a large number of
specimens of this species at Kerguelen, very nearly as many as of Serolis latifrons.

The males of this species diflPer from the females by their greater size.

The largest male specimen measures 13 mm. in length by 11 mm. in breadth, the
largest female specimen measures 11 mm. in length by 9 mm. in breadth. All the
specimens obtained by the Challenger, with the exception of a few newly hatched young,
are almost exactly the same size as the two specimens from which the measurements are
taken, and the superiority in size of the males over the females can therefore be very
plainly seen.

The abdominal epimera are similar in both sexes.

A noticeable difference between the two sexes is to be found in the first thoracic
epimera ; the under surface of these epimera in the males is furnished with a row of
ridge-like tubercles, six or seven on either side, which are situated just external to the
attachment of the appendages ; they are arranged in a semicircle, with the convexity
directed forwards and outwards ; in the female there is no trace of any such structure ;
it is probable that these tubercles assist the male in maintaining a firm hold of the female
during copulation.

The general surface of the body, especially the epimera, is sculptured, the latter are
traversed by innumerable ridges anastomosing with each other ; on the cephalic shield
and the tergal portion of the segments the surface is covered by minute scattered pits.
All the epimera, with the exception of the first, are traversed by a crescentic ridge which

1 In my opinion the catalogue name Serolis qaadricarinata ought not to have been changed by Miers in spite of its
being a " nomen ineptum," but since the actual description of the species is under the name of Serolis septemcarinata, it
is necessary to adopt it here.

48 THE VOYAGE OF H.M.S. CHALLENGER.

follows at some distance the anterior margin of the segment ; the first epimera are not
divided by any transverse suture ; the epimera of the three succeeding segments are
separated from their terga by a suture.

The cephalic shield has a crescentic ridge between and a little in front of the eyes
as in Scrolls tuberculata {cf. PL VI. fig. 1) and other species ; its anterior margin is pro-
longed in the median line into a short rostrum, and as in other species a ridge separates
off" the antero-lateral portion from the rest.

The caudal shield is trifid at its extremity, and furnished wdth a central carina which
runs from end to end ; on either side of this are two lateral carinse, which do not extend
along the whole length of the caudal shield, they terminate behind at some distance from
the end of the caudal shield in freely projecting spines ; the inner pair extend farther
back than the outer pair ; another carina runs parallel with the anterior margin of the
caudal shield on either side, and terminates close to the articulation of the uropoda. The
lateral margins of the caudal shield, as well as of the posterior epimera, are denticulate.
A noticeable character of this species is the prolongation of the abdominal epimera a little
way beyond the last pair of thoracic epimera ; in Serolis minuta the same thing occurs,
but in these two species only.

T\iQ first pair of antenncB are a little shorter than the second pair; the filament, which
has about eighteen joints, is rather longer than the basal portion of the antenna, — the
latter is composed of four joints; the first is short, the second about twice its length, and
about equal in length to the succeeding joint, though considerably stouter than it.

In the second pair of antennce the two distal joints of the stem are the longest, and
about equal in length to each other ; they are somewhat bow-shaped, and the anterior
margin is beset with tufts of long hairs, five to each joint in the specimen from which the
description is taken ; the filament has twelve or thirteen joints.

The mandibles show the usual asymmetry; the left mandible has a chisel-like cutting
process, the edge of which is somewhat crenate in outline ; this structure is absent from
the right mandible.

The first pair of rnaxillcB consist as usual of a basal piece with which are articulated
two masticatory lobes ; the posterior of these is delicate and small, and bears one short
spine at its expanded distal extremity as in Serolis par adoxa {cf. PL V. fig. 14).

The second pair ofmaxillcB are comparatively large; the middle lobe is furnished with
four spines upon the cutting margin instead of the usual two.

In the maxillipedes the stipes and lamina are completely separated by a suture ; the
second joint of the palp is not so markedly heart-shaped as in other species.

In the second pair of thoracic appendages the sixth joint is entire ; the fifth joint has
a row of spines upon the inner side, they are as usual of two kinds ; the longer spin es
terminate ia a bifid extremity, of which the anterior bifurcation is the longest ; the axis
of the spine extends between the two branches, and is rather longer than either.

REPORT ON THE ISOPODA. 49

Oue of the third pair of thoracic appendages in the male is displayed in fig. 3 ; the
spines peculiar to this limb are confined to the projecting lower angle of the penultimate
joint ; several of these are shown more highly magnified in fig. 5.

The remaining thoracic limbs present no features of interest.

The third pair of appendages in the male (PI. VIII. fig. 3) is rather different from
the same appendages in other species ; the penultimate joint is comparatively long and
narrow, but widens out in its posterior fourth, where six spines are attached in pairs ; at
the upper end, close to its articulation with the terminal joint, is a single pair of spines ;
the interval between this spine and the six at the hinder end of the joint is beset with a
few scattered hairs ; the posterior spines are broad and conical, and the central filament
projects a short way out from the extremity. The remaining joints of this limb are
smooth and almost devoid of spines. The other thoracic appendages (PI. VIII. fig. 4) are all
similar to each other, save that in the last pair, the third, fourth, and fifth joints are
clothed with a closely set row of fine delicate hairs in addition to the sharp sword-like
spines which are found here as in the other appendages ; in the female these hairs appear
to be absent. The same difference between the two sexes in this appendage has been
already described in Serolis convexa {ante, p. 40), and it exists in other species. The
terminal joints of the thoracic appendages from the third pair onwards are divided by
a suture into a short distal and a long proximal half.

The frst three abdominal appendages, like those of many other species, have the
inner angle of the basal joint prolonged into a process which bears three branched hairs
in the first and two such hairs in the two following pairs of appendages. The suture
upon the opercula is inclined at an angle with the transverse axis.

The uropoda are attached exactly half way down the lateral margin of the caudal
shield, but do not extend as far as its extremity.

Station 145, December 27, 1873 ; lat. 46° 43' S., long. 38° 4' 30" E. ; 140 fathoms ;
bottom, volcanic sand.

Station 149b, January 17, 1874 ; lat. 49° 28' S., long. 70° 30' E. ; 25 fathoms ;
volcanic mud ; off Marion Island, 50 fathoms.

Station 149c, January 19, 1874; lat. 49° 32' S., long. 70° 0' E. ; GO fathoms;
bottom, volcanic mud.

Station 149d, January 20, 1874 ; lat. 49° 28' S., long. 70° 13' E. ; 28 fathoms ; bottom,
volcanic mud.

6. Serolis cornuta, Studer (PI. I. figs. 1-16).

Serolis comiifa, Studer, Beitr. zur Kenut., &c., Arcliiv f. Naturgesch., 1879, p. 19.
This species has been described by Studer in his Memoir on the Fauna of Kergueleu,'
where a figure is given of the male and certain of the appendages. The males and

■ Loc. cit., pp. 21-24.

(ZOOL. CUALL. EXP. — PART XXXIII. — 1884.) ^^^ '

.'0 THE VOYAGE OF H.M.S. CHALLENGER.

females differ from each other in their shape ; as in Serolis schytliei, the male is much
broader in proportion to its length, being nearly circular in outline; in the female
the shape of the body is oval ; this difference depends upon the greater length of the
epimera ; in the male the sixth pair extend nearly so far as the end of the terminal
S2jine of the caudal shield, while the two abdominal epimera extend about half w&y
down the caudal shield, their posterior extremities being situated on a line with the
articulation of the uropoda ; in the female the sixth thoracic epimera only just pass
beyond the articulation of the uropoda, while the abdominal epimera hardly reach as far.

The sterna of the abdominal segments differ in the two sexes ; in the male the two first
are oblong in shape, the posterior margin being slightly concave, the third is prolonged
into a flat triangular spine ; in the female the posterior margin of all the segments is
prolonged into a blunt spine ; the length of these increases progressively from before
backwards.

Studer has given a good general description of this species, comparing it with Serolis
trilobitoides, to which it is most nearly allied. For the sake of completeness I give here a
recapitulation of his description, -udth a few additions, which applies to all the specimens
in the Challenger collection, the variations being inconspicuous and confined to the colour,
which is darker or lighter, and the number of spines upon the carina of the caudal shield,
which vary from three to six.

The length of the largest male specimen is 34 mm., its breadth 31 mm. ; the length of
the largest female specimen is 41 mm., its breadth 35 "5 mm.

The length of the cephalic shield is about one-fifth of the whole length of the body ;
close to the anterior margin is a transverse ridge which runs almost parallel to it ; in
front of the eyes, and occupying the middle of the cephalic shield, is a squarish area
raised and separated off by a groove ; the posterior margin of this is slightly concave,
the two ends being somewhat produced backwards ; the lateral portions of the cephalic
shield in front of the eyes slope gradually downwards towards the side ; behind and to the
inside of the eyes is a large blunt conical tubule, one on either side ; between the two, and
forming the posterior boundary of the cephalic shield, is a flattened triangular area.

Thorax. — The epimera are broad and sickle-shaped, and the terminal portions from
the articular process outwards project freely and do not overlap each other ; the
articular processes are short, and as in many other species, there are only two to
each epimeron ; the posterior articular process of each epimeron overlaps the anterior
articular process of the succeeding epimeron.

The second, third, and fourth epimera are separated from their terga by a distinct
suture. The anterior epimera are separated into two halves by a transverse ridge, and
there is in addition an anterior ridge as in Serolis hromleyana and Serolis gracilis, which
is a continuation of the anterior ridge on the cephalic shield ; it runs at first parallel to
the anterior margin of the epimeron at some little distance from it, and then bends

EEPORT ON THE ISOPODA. 51

abruptly at right angles, terminating a little way in front of the second transverse ridge.
The margins of all the epimera are minutely serrated externally. The terga of the free
thoracic segments have almost the same antero-posterior diameter, the two last being
only a trifle narrower ; the sterna of the first two thoracic segments are as usual divided
by sutures into a median and two lateral sclerites ; the median sclerite of the anterior
segment is keeled, that of the second has a short keel, widening behind and extending
over its anterior half only. ' The posterior margin is raised into a ridge, which is continued
along the lateral sclerites, but is here rather narrower. The sterna of the three last
segments are partly fused as in other species.

Abdomen. — The caudal shield is rather more than one-third of the length of the
whole body ; its shape is pentagonal, and the end is produced into a long spine ; there is
a distinct longitudinal keel, which is furnished with three to six short recurved spines,
the first of which is always the largest ; at the anterior end of the caudal shield are two
short spines, one on either side, and closely approximated to the middle line ; from the
outer margin of each of these runs a fold somewhat curved in its course, which terminates
a little in front of the attachment of the uropoda in a short spine ; this lateral fold is
similar to that found in Serolis paradoxa and Serolis schythei, and perhaps marks the
boundary of the last segment of the abdomen ; the lateral margins from the attachment
of the uropoda to the terminal spine are provided with a number of small teeth.

On PI. I. are figures of a male (fig. 1), female (fig. 2), and an immature male
specimen (fig. 3); the latter is introduced in order to show its similarity in shape to the
female (see p. 27).

The first pair of antennoB are shorter than the second pair by about half their own
length ; the terminal joint of the base is large, being rather more than half as long as the
preceding joint; the filament, which is about as long as the rest of the antenna, has
twenty -five joints.

In the second pair of antennae the basal portion has as usual five joints, of which the
last is the longest, being about equal in length to both the third and fourth joints
together. The filament is nearly but not quite as long as the fifth joint, and is composed
of fifteen or sixteen joints. One of these is shown highly magnified in PI. I. fig. 6 ;
on the inner and lower surface of some of the joints towards the middle of the filament
is a row of short lancet-shaped spines; these decrease in size towards the distal end of the
filament, and in the last three or four joints seem to be entirely absent ; they commence
to be visible on about the third joint of the filament, but arc only present in small
numbers towards the distal end of the joint and close to its articulation with the one in
front. The figure (PI. I. fig. 6) represents the seventh joint ; these spines, which are
present upon the antennse of both sexes, are like those that are found in Serolis j^radoxa,
and which have been described (ante, p. 35).

Mandibles. — The distal extremity of the right and left mandibles of a male specimen

52 THE VOYAGE OF H.M.S. CHALLENGEE.

of Serolis cornuta arc shown on PL I. figs. 7 and 8. The right mandible (fig. 7) has
ujion the upper surface a broad chisel-like process which is absent from the left ; the latter
is compensated by two large bifid spines instead of the single spine which springs from the
lower surface of the right mandible. In another specimen, however, these conditions were
reversed, the chisel-like plate being developed only upon the left mandible, while the
right mandible had two spines.

T\\<i first pair of maxillcB are quite like those of other species.

The second pair (PL I. fig. 9) in one specimen were asymmetrical like the mandible ;
the middle lobe of the left hand maxilla had four spines, while the right hand one (PL I.
fig. 10) had only two; in another specimen, however, both right and left maxillas had only
two spines.

The maxillipedes (PL I. fig. 11) are very much covered with hairs ; the stipes, which
is separated from the lamina by a complete suture, is squarish in outline ; the palp is
broad and flat, the second joint distinctly cordate in outline.

The second ixiir of ambidatory limbs in the male is modified in the usual way ; a
single spine similar to those which are found upon the inner margin of the penultimate
joint of these appendages is also found on the inner side and close to the distal end of
each of the two preceding joints.

The remaining ambulatory limbs are in no way remarkable ; they are more like those
of Serolis schythei than any other species ; the first joint is the longest and broadest ; the
second joint is narrower and slightly shorter ; it increases in width towards the distal
extremity, which is furnished with a tuft of comparatively delicate spines and hairs ; the
third and fifth joints are subequal in size, while the fourth is a trifle longer ; large sword-
like spines mingled with smaller ones, and a very few serrated spines, are arranged in
tufts on the inner and outer sides of the distal extremity of these joints, and a few along
the outer margin.

The three anterior abdominal appendages of the female are displayed in figs. 12-14
of PL I. ; as in Serolis septemcarinata and others the basal joint of the first (fig. 12)
has three branched hairs like those which clothe the outer margin of the exopodite and
endopodite of the same appendages, but smaller ; the second and third of these appendages
(figs. 13, 14) have two hairs instead of three.

The suture upon the exopodite of the fourth pair of abdominal appendages is
oblique.

The uropoda are attached about half way down the caudal shield, and extend as far
as its termination, the exopodite does at least, being almost half again as long as the
endopodite ; the outer margins of both, and the distal extremity of the inner margin are
serrated and beset with branched hairs.

Station 149, January 9, 1874 ; lat. 49° 8' S., long. 70° 12' W. ; 25 fathoms ; bottom,
volcanic mud. January 13, Betsy Cove, Kerguelen Island.

REPORT ON THE ISOPODA. 53

Station 149b, January 17, 1874; kt. 49° 28' S., long. 70° 30' W. ; 25 fathoms;
bottom, volcanic mud.

Station 149D, January 20, 1874; lat. 49° 28' S., long. 70° 13' W.; 28 fatlioms ;
l)ottom, volcanic mud.

Station 149e, January 21, 1874; lat. 49° 37' S., long. 70° iC W. ; 30 fatlioms;
bottom, volcanic mud.

Station 149k, January 29, 1874; lat. 48° 40' S., long. 69° C W.; 45 fathoms;
bottom, volcanic mud.

7. Serolis hromleyana, Suhm (PI. IV.).

Serolin hrojnleijana, v. Willemoes Suhm, Proc. Eoy. Soc. Lond., vol. xxiv. p. 591, 1876.
HeroUs bromleyana, Challenger Briefe, No. II., Zeitschr. f. wiss. Zoo)., Bd. xxiv. p. xix., 1874.

This species, which is considerably the largest of the genus, has been already briefly
characterised by the late Dr. v. Willemoes Suhm from two specimens dredged in 1975
fathoms near the Antarctic Ice -Barrier. The following description is taken from these
specimens : — Length of male 54 mm., greatest breadth (at the level of the third epimera)
56 mm.; length of female 45 mm., greatest breadth 39 mm.

The male differs from the female by its greater size and in the greater length of the
epimera ; in both sexes the length of the thoracic epimera gradually increases up to the
sixth, which are by far the longest, and extend for a considerable distance behind the termi-
nation of the caudal shield. The extreme length of thqse epimera, measured from their
articulation with the epimera of the preceding segment, is 47 mm. in the male and
33 mm. in the female. The epimera of the second and third abdominal segments project
beyond the caudal shield in the male ; in the female the last pair barely reach as far as
its termination. The form of the sterna of the free abdominal segments does not differ
at all in the two sexes.

The outline of the body is more or less oval, and the great length of the epimera,
which become extremely narrow and spiniform towards their end, serves to distinguish
this species from all others, with the exception of Serolis necera and Serolis gracilis.
From both these species, however, it can readily be separated by numerous other points
of difference. The body is covered with scattered hairs, which are especially developed
upon the sides of the epimera. The colour (in alcohol) is violet-grey with whitish j^ellow
patches upon the caudal shield and posterior portion of the thorax ; the colour of the
living animal is described by v. AVillemoes Suhm as being of " a fine blue colour with a
red spot extending over the midst of the body and the eyes."

Cephalic Shield. — The shape of the cephalic shield can be understood by a reference
to PL IV. fig. 1 ; it is longer than broad, owing to the projection of its lateral portion^
for some way in front of the rostrum ; these antero-lateral portions of the cephalic shield

54 THE VOYAGE OF H.M.S. CHALLENGER.

are separated from the rest by a ridge which rims from just below the rostrum to the
lateral margin on either side ; the whole cephalic shield is very distinctly separated from
the thoracic segment by a deep furrow ; the central portion which lies between the eyes
is very strongly convex, and separable into three regions — two round convexities which lie
to the inner side of and behind each eye, and a median X-shaped elevation, at the upper
end of which, on a level with the anterior portion of the eyes, are four tubercles arranged
in a semicircle with the concavity directed forwards ; at the hinder extremity is another
short tubercle ; the whole surface of the cephalic shield, with the exception of ' a flattened
area which extends from the upper end of the eyes to the transverse ridge, is covered with
minute pit-like depressions. The eyes are whitish yellow in colour.

Thorax. — The thoracic segments, like the cephalic shield, are covered with an immense
number of irregular pit-like depressions ; the posterior margin of each segment is furnished
with a minute tubercle. The epimera are of great length, and increase gradually from
the first up to the sixth and last pair.

The epimera of the first thoracic segment are as usual large and expanded, but narrow
rapidly towards the distal extremity, which is narrow and spine-like as in the succeeding
segments. The anterior margin of these epimera bears a short forwardly directed spine at
about 3 mm. distance from the cephalic shield. The epimeron is divided into three pieces
by a Y'Shaped ridge ; the two arms of the Y iovm a very obtuse angle, one is continuous
with the anterior spine akeady referred to, and the other passes outwards towards the
distal end ; the unpau-ed arm seems to correspond to the suture which is often developed
upon the first pair of epimera in other species. The two areas lying respectively in front
of and behind this median ridge are somewhat concave, while that portion which lies to
the outside of the arms is convex.

The epimera of the three succeeding segments are curved and sickle-shaped, and
project outwards at a greater angle with the longitudinal axis of the body than the rest,
which by degrees come to lie almost in the same straight line with this axis. A distinct
suture separates the dorsal portion of the three anterior free thoracic segments from these
epimera. All the epimera of the body project downwards as well as outwards, especiall}-
in the male specimen, so that when the animal is placed upon a flat surface, the body
rests entirely upon the epimera. The second of the free thoracic segments is the widest,
and measures 4"5 mm. in diameter ; the first and third are a trifle smaller, whde the
fourth and fifth are only one-third of the diameter of the second.

The sterna of the two first thoracic segments are divided by sutures into three portions,
a median and two lateral. The median portion of the anterior segment (that which bears
the maxUlipedes) is keeled ; the sterna of the second, third, and fourth segments are
divided by a median suture into two equal halves ; the remaining segments have a rather
peculiar arrangement, which is displayed in PL IV. fig. 2 ; the middle portions of the
three segments are fused together to form a s(«newhat oval plate, divided by a median

EEPORT ON THE ISOPODA. 55

suture, and upon which two transverse furrows mark the boundary between the several
segments of which it is composed. The lateral portions of the sterna of the three segments
are at first fused together, but become separate just before the attachment of the append-
ages ; and a triangular plate with the apex directed posteriorly, lies between the median
and lateral portions of the sternum of the sixth segment.

Abdomen. — The first abdominal segment has as usual no epimera ; its width is about
double that of the preceding thoracic segment ; the second and third abdominal segments
are long and well developed, and terminate in a somewhat bifid extremity, the inner limb
of the bifurcation being considerably longer than the outer (figs. 1, 2). The sterna of the
three free abdominal segments are shown in fig. 2 ; there is a short median backwardly
projecting tubercle, longer in the second and third than in the first ; this is continuous
with a longitudinal ridge which is crossed at right angles by another ridge. The male and
female specimens, as already mentioned, show hardly any difi"erences in the conformation
of these abdominal sterna ; in the female the antero-posterior diameter is somewhat less,
and the median tubercle somewhat more pronounced than in the male.

The caudal shield, which is broader than long, has a somewhat pentagonal contour
with rounded angles ; the uropoda are attached about half way down the side, the
posterior extremity is notched and rather turned up ; there is a median longitudinal keel,
and on either side a short flat spine near the lateral margin, and on a level with the
attachment of the uropoda ; the surface l^etween the keel and these spines is almost flat,
but is strongly bent downwards ; the length of the caudal shield in the male is 1 7 mm. ,
almost half that of the rest of the body, its breadth 21 mm.

Ajyj^endages. — The antennce are almost exactly of the same length. The first pair of
antennas has four joints and a terminal filament made up of thirty-one joints, and equal
in length to the first four joints.

The basal joint is oval, with a truncated distal extremity; the second joint slightly
longer, with a straight outer and convex inner margin ; the third joint is half again as
long as the second and about half its width; it is cylindrical in shape; the terminal joint
is about one-third of the length of the preceding joint, and of the same shape.

The second pair of antennae have six joints and a short terminal filament ; the
sixth joint is the longest, the third is slightly longer than the fourth; the two basal
joints are small, especially the second, which only occupies one side of the antennse,
on the other (inner) side the first joint articulates directly with the third ; on the fourth,
fifth, and sixth joints are a series of ridges running transversely to their long axes,
and bearing bunches of fine hairs.

The mandibles are as usual asymmetrical ; the left hand one has a projecting chisel-
shaped process which is wanting in the right hand one.

The maxillce are like those of other species ; the anterior pair has nine or ten
curved spines upon the cutting edge, the most anterior of which appear to be the largest.

5G THE VOYAGE OF H.M.S. CHALLENGER.

In the maxillipedes {fig. 8) the stipes and lamina are not separated by a complete
suture.

The second pair of thoracic appendages in the male are modified like those of all other
species ; the inner side of the second, third, and fourth joints is clothed with fine delicate
hairs (PI. IV. fig. 6) similar to those of Serolis necera, but more delicate and incon-
spicuous ; these structures I was only able to find in some of the specimens from Stations
168 and 169; in the large specimen from Station 15G ((_/! PI. IV. fig. 5) they are cer-
tainly not present, nor in the single specimen from Station 164a. It does not appear
quite certain whether the presence or absence of these hairs is merely a local variation or
an indication of a dimorphism in the males of this species analogous to that which Fritz
Miiller^ has described in a species of Tanais.

The remaining thoracic appendages are slender, like those oi Scrolls gracilis, and, as in
that species and others, the fourth joint is longer than the third or fifth joints ; the hairs
and spines are also small and delicate ; neither the serrated spines found in many species
nor the pinnate hairs found in Serolis necera and 'Serolis gracilis seem to occur in this
species ; at the distal end of the penultimate joint only are there very long slender hairs,
many of which are considerably longer than the terminal joint of the appendage.

The abdominal ajypendages are like those of Serolis antarctica.

The operculum is divided by a transverse suture at right angles to its longitudinal
axis.

The uropoda are comparatively small, and attached at the commencement of the
posterior third of the caudal shield.

Variations. — The description just given refers to two specimens dredged at Station
156 ; this species was also obtained at three other Stations, viz.. Stations 164c (400 fathoms),
168 (1100 fathoms), and 169 (700 fathoms), and some of the specimens show certain
difi"erences, chiefly in the length of the posterior thoracic epimera and in the conformation
of the two abdominal epimera ; in one specimen (PI. IV. fig. 3) the posterior thoracic
epimera, instead of lying nearly parallel to the long axis of the body, are curved inwards
towards the extremity so as to partly enclose the caudal shield ; in this specimen the last
thoracic epimera are proportionately longer than in the type sjjecimen, measuring 36 mm.
as against 35 mm. length of body, whereas in the type specimen these epimera, as already
stated, measure 47 mm., the length of the body being 54 mm. In the majority of speci-
mens the abdominal epimera terminate in a sharj) point and are not notched at their
extremity; two other specimens, however, had abdominal epimera precisely like those figured
on PL IV. fig. 1, and since both these specimens are small, not measuring more than half
the length of certain other specimens in which the abdominal ej^imera terminate simplj-
in a point, this character cannot be looked upon as peculiar to adults.

In all the specimens obtained at Stations 168 and 169, the ridges upon the cephalic

' Fiir Darwin (Facts for Darwin), Euglish translation, London, 1869, p. 19 ct seq.

REPORT ON THE ISOPODA. 57

shield are somewhat different from those figured and described in the type specimens ; a
strong ridge runs parallel to the outer margin of the epimera ; at its upj^er extremity it
bends backwards and passes for a short distance parallel to the boundary lino of the
cephalic shield, terminating a little below the ridge which crosses the latter ; the transverse
ridge, which indicates the division of the epimeron into two parts corresponding to the
first two thoracic segments, is very slight ; the anterior spine-like process of these epimera
is not present, and the anterior margin is in consequence quite smooth (fig. 3). Many of
these specimens are more darkly coloured.

In a small specimen from Station 164c (400 fathoms), measuring 16 mm. in length,
the outer ridge upon the anterior epimera is not present, but the abdominal epimera have
the same characters as those of the type specimens.

Station 156, February 26, 1874; lat. 62° 26' S., long. 95° 44' E. ; 1975 fathoms;
bottom. Diatom ooze.

Station 164c, June 13, 1874 ; lat. 34" 19' S., long. 151° 31' E. ; 400 fathoms ; bottom,
green mud.

Station 168, July 8, 1874 ; lat. 40° 28' S., long. 177° 43' E. ; 1100 fathoms ; bottom,
blue mud.

Station 169, July 10, 1874 ; lat. 37° 34' S., long. 179° 22' E. ; 700 fathoms ; bottom,
blue mud.

8. Serolis necera, F. E. B. (PL V. figs. 1-11).

ScroUs necera, F. E. Beddard, Proc. Zool. Soc. Lond., 1884, pt. iii. p. 331.

This species almost rivals Serolis hromleyana in size, and resembles it in the great
length of its spiniform epimera.

The largest male measures 42 mm. in length and 48 mm. in breadth ; the largest
female measures 41 mm. in length and about 40 mm. in breadth. The difierence in
the proportion of length to breadth in the two sexes is caused by the greater develop-
ment of the epimera in the male ; the two sexes also differ in the characters of the sterna
of the abdominal segments and in the frontal " sense organ," which is much larger and
more evident in the female {cf. figs. 1 and 3) ; there is not such a marked difference be-
tween the lengths of the abdominal epimera in the two sexes as is often found in the
other species of Serolis.

Cephalic Shield. — The eyes are unusually large and conspicuous, 6 mm. long, bluish
black in colour owing to the comparatively small amount of pigment present; the
diameter of the cephalic shield is greater than its length ; as in Serolis hromleyana, a
transverse ridge passes from the base of the rostrum to the lateral margins of the cephalic
shield, and cuts off a small antero-lateral portion ; the anterior margin of this as well as
of the first epimera is bent upwards. Between the eyes are a number of spiniform pro-

(ZOOL. CHALL. EXP. — PART XXXIII. — 1884.) Kk 8

58 THE VOYAGE OF H.M.S CHALLENGER.

cesses which are hollow and filled up with counective tissue, two smaller median
spines and two outer ones with a bifid extremity ; the posterior margin of the cephalic
shield is indented, and forms three projections, the two outer ones being somewhat
triancfular in shape, and tuberculated on the free margin, while the inner median one is
transversely elongated, and much like the labium in shape, with a slightly convex outer
margin.

Thorax. — Each of the thoracic segments is furnished with a median spine — the first
three are considerably longer than the rest ; the antero-posterior diameter of the thoracic
segments increases from the first to the fourth, the latter measuring 3 "5 mm. in the male;
the two last thoracic segments are comparatively longer than in Serolis bromleyana,
rather more than half as long as the preceding segment. The terga of the three anterior
free thoracic segments are separated from their epimera by a distinct suture. The epimera
of the segments gradually increase in length up to the sixth pair, which are extremely
lono-, measuring 37 mm. in the male, and extend backwards in a direction nearly parallel
to the long axis of the body.

The first epimera are larger in the male than in the female {cf. figs. 1, 3); in the latter
the margin of the epimera passes at first abruptly backwards, and then curves outwards ;
the antero-lateral portion of the cephalic shield projects outwards beyond the commence-
ment of the epimeron for a short space, and ends in a truncated slightly notched extremity;
in the male the outer margin of the first pair of epimera is quite regular. The epimera
are divided into two halves by a longitudinal ridge.

The sterna of the two anterior segments are as usual divided by sutures into three
sclerites ; the anterior segment is similar in shape to that of Serolis bromleyana ; the
median sclerite of the second segment is saddle-shaped as in the last mentioned species,
but instead of being smooth the anterior half is produced downwards into a triangular
process ; the posterior half is raised into two knobs, which are separated from each other
in the median ventral line by a suture. The three succeeding segments are divided by
a median suture which is continuous to the end of the thoracic segments. The sixth,
seventh, and eighth thoracic segments are entirely similar to those of Serolis. bromleyana
above described.

Abdomen. — The epimera of the second abdominal segment extend for a space of
about equal to half its own length beyond the caudal shield ; they are not perceptibly
longer in the male. The epimera of the third segment are very short, and only extend
to about the commencement of the lateral margin of the caudal shield ; the length of these
epimera also does not diff'er in the two sexes. The posterior extremity of the first pair
of these epimera is notched as in Serolis bromleyana. The sterna of the three free
abdominal segments difi"er in the two sexes ; in the female the posterior margin of each
is produced into a short spine, the first being the largest and longest ; in the male the
last of the three segments only has a spine.

REPOET ON THE ISOPODA. 59

The caudal shield is almost hexagonal in outline ; its length in both the male and
female is about 11 mm.; it is therefore smaller in proportion than that in Serolis hromley-
ana; its breadth is about 12 mm. The uropoda are attached at about the commencement
of the posterior fourth. The median portion of the shield is slightly keeled, and bears two
spines one in front of the other, the posterior being the larger of the two. On either
side are two other short flat spines. The caudal shield of this species is very similar to
that of Serolis schythei, but it is not so distinctly carinate, and the anterior median spine
is flatter and not so large as in Serolis schythei, while the posterior median and the
lateral spines are larger than in that species.

Appendages. — The second pair of antennae are a little longer than the first pair.
In the Jirst antennce the two basal joints are of about equal length, the third
joint is twice as long as either of the preceding ones, the fourth joint short, hardly
longer than the first joint of the filament ; the filament is composed of about twenty
joints.

The second pair ofantennce have the third, fourth, and fifth joints as usual covered on
the inferior surface with tufts of hairs in groups of three or four ; the filament has some
fifteen joints.

The mandibles do not appear to diS'er from those of other species.
Tko, first maxillce consist as usual of a stout elongated lobe and small oval lobe, both
of which articulate with the cardo ; the smaller lobe is furnished on th« free cutting edge
with a single spine as it is in Serolis paradoxa {cf. PI. V. fig. 14).

The maxillipedes are figured on PI. V. fig. 6 ; the suture between the lamina and
stipes is complete ; the palp of the mandible is abundantly furnished with setae, and the
second joint has on the inner side the small protuberance characteristic of all the deep-
sea species of the genus (a).

The second pair of thoracic appendages difi"er from those of other species in that the
hairs which clothe the inner margin of the penultimate joint are very similar in form ;
several of these are shown on PL V. fig. 7 ; the longer hairs (a) only difi"er from the
shorter ones (6) by the proportions of their length and breadth.

The tlm-d pair (fig. 8) are as usual modified into a prehensile organ in the male ;
the third, fourth, and fifth joints are covered on the inner side with a dense row
of hairs which are shown magnified in fig. 9, a' ; these hairs are clothed on the
distal half with fine branches, and they closely resemble the hairs that are found on
the second thoracic appendage of the males of Serolis convexa. The fifth joint is oval,
narrowing towards the upper extremity ; it possesses the peculiar spines characteristic
of this appendage ; they are narrow and cylindrical, and terminate in a long thread
which is given oft' from the upper surface of the spine just before its termination
(fig. 9, b'). The sixth joint is divided by a suture into a small distal portion and a long
proximal portion.

GO THE VOYAGE OF H.M.S. CHALLENGEK.

The ambulatory limbs differ from those of other species in the character of the hairs ;
instead of sharp serrated spines, which are generally developed upon these appendages,
the ambulatory limbs of Serolis necera are clothed with long branched hairs simdar to
those found upon the three anterior abdominal appendages, but somewhat shorter. One
of the limbs of the right side is figured on PL V. fig. 10 ; the other thoracic limbs hardly
differ at all, except that the last is considerably shorter as in other species. On the
inner margin of the proximal joint are a number of very fine hairs, the distal half being-
furnished with delicate branches. These hairs are exactly similar to the terminal
hair of the first antennae in this and other species, and are possibly sensory. The remain-
ing joints are provided with bunches of branched hairs, those on the inner side of the limb
being longer than those on the outer side ; the arrangement of these can be seen by an
inspection of the figure (fig. 10) ; the terminal joint forms as usual a claw which is long
and slender.

Abdominal Appendages. — These appendages present no peculiarities ; the basal joint
of the first pair is furnished with three hairs upon the inner posterior angle. The two
following appendages have each two hairs in the same place.

The exopodite of the first pair of gills is divided by an oblique suture, and the
endopodite is bifid at the tip (see fig. 11).

Another specimen of this species which was dredged in 2040 fathoms (Station 318)
is in some respects different, but the differences do not appear to me to be of sufiicient
importance to warrant its separation as a distinct species.

The epimera, instead of being long and spiniform as in all the other specimens
contained in the Challenger collection, are comparatively short, and resemble the epimera
of the typical shallow-water forms {e.g., Serolis scliythei) in being comparatively wide and
flattened; the two posterior thoracic epimera were unfortunately broken off on both
sides of the body, but judging from what remains it seems very likely that the sixth
pair at any rate extend back beyond the termination of the caudal shield. The
anterior pan.' of abdominal epimera terminate a little way in front of the attachment of
the uropoda.^

Station 318, February 11, 1876; lat. 42° 32' S., long. 56° 29' W. ; 2040 fathoms;
bottom, blue mud.

Station 320, February 14, 1876; lat. 37° 17' S., long. 53° 52' W. ; 600 fathoms;
bottom, green sand.

^ This specimen was mounted on a slide in glycerin ; on the same slide, entangled with the appendages of the
Serolis, were several small Nematodes, not sufficiently well preserved to exhibit any distinctive generic character. It is
of course not certain that they came from the same depth as the Crustacean, but, bearing in mind the fact that many
of the marine free-swimming Nematoidea attach themselves in a semiparasitic fashion to other animals {cf. Villot, Arch.
(1. Zool. Exp., t. iv. p. 451, 1875), it is at any rate possible; so little is known respecting the distribution of the free-
swimming Nematoidea that I think it worth while to record this fact, especially as I observed other Nematoidea
among the appendages of one of the specimens of this same species from Station 320.

REPORT ON THE ISOPODA. 61

9. SeroHs gracilis, F. E. B. (PL III. figs. 7-13).

SeroUs gracilis, F. E. Beddard, Pioc. ZooL Soc. Lond., 1884, pt. iii. p. 332.

Five specimens of this species were dredged at Station 120 (675 ftxthoms), three males
and two females.

The males were approximately of the same size, the largest measuring 21 mm. iu
length by 22 mm. in breadth; the females were much smaller, measuring 9 mm. in length,
by 8 mm. in breadth.

The chief difference between the two sexes, apart from that of size and relative
proportions of length and breadth, consists in the greater length of the epimera in the
males ; in the female specimens (PL IIL fig. 8) the last thoracic epimera hardly reach as
far as the end of the caudal shield, while the first abdominal epimera only extend about
haK way down, and the posterior abdominal epimera terminate at about the level of the
end of the anterior third of the caudal shield ; in the male (PL IIL fig. 7) the posterior
thoracic epimera are considerably longer, reaching beyond the caudal shield for a space
of about its own length; the actual length of these epimera is 13 mm.; the first
abdominal epimera extend a short way beyond the end of the caudal shield, and the
posterior pair to about the middle. The difference between the two sexes in the length
of the epimera is more marked in this species than in any other known to me.

The general form of the body is circular, and the dorsal surface is covered with
scattered pits ; the colour (in alcohol) is a dark slate-blue, varying to reddish yellow upon
the terga of the posterior thoracic and abdominal segments.

The cephaliG shield has much the same shape as in Serolis hromleyana ; the portion
l}T^ng between the eyes, which are small and inconspicuous, is strongly convex, while the
antero-lateral areas are flat and depressed, and do not rise above the level of the first
thoracic epimera ; a transverse ridge running from the base of the rostrum, which is very
minute, divides the cephalic shield as in Serolis hromleyana.

TJiorax.- — The first epimera are divided into three portions by two transverse ridges ;
the anterior one is continuous with the ridge that traverses the cephalic shield, it passes
at first across the epimeron and then bends backwards running parallel with the anterior
margin of the epimeron, and joins the distal end of the second ridge ; the continuation
of these two ridges passes along the margin of the epimeron closely applied to it, and
terminates some way in front of the end of the epimeron. The posterior ridge corresponds
to the line of suture between the two fused epimera of the first and second thoracic
segments.

The other epimera are fiat and sickle-shaped, not spinifonn as in Serolis hromleijana
and Serolis necera; they gradually increase in length up to the sixth; the articular
processes, which unite together the succeeding epimera, are placed further than is usual
from the junction between the terga and the epimera, which gives the latter the appear-

62 THE VOYAGE OF H.M.S. CHALLENGER.

ance of beiug shorter than they really are ; the fourth and fifth as well as the three
anterior epimera are separated from the terga by a distinct suture. The antero-posterior
diameter of the three anterior free thoracic segments gradually increases from the first
to the tliird ; the diameter of the fourth and fifth is rather more than half that of the
third. The sterna of the two first thoracic segments are as usual divided by lines of
suture into three portions ; the median sclerite of the first segment is keeled, and that of
the second segment has a median keel which bifurcates posteriorly; the hinder half of this
sclerite is divided hj a median suture continuous with that which separates the right and
left halves of the sterna of the succeeding segments. As in Serolis hromleyana and Serolis
necera, the median portions of the sterna of the last three thoracic appendages are fused,
though the lines of division corresponding to the three segments are more strongly
marked in this species than in the two above mentioned.

Abdomen. — The epimera of the abdominal segments have been already described
{ante, p. 13) ; the sterna do not appear to differ in the two sexes. The coAidal shield is
squarish in outline, 7 mm. in length by 8 mm. in breadth ; its length is therefore not
quite one-third of that of the whole body. The uropoda are attached close to its posterior
margin. There is a slight longitudinal median keel, which is crossed at right angles by
a sinuous ridge with three convexities, one median and two lateral, which correspond
to the spines on the caudal shield of Serolis necera and Serolis schythei ; at about the
end of the anterior third of the caudal shield is a short flat spine in the middle line and
two oblique ridges, one on either side of this spine ; the lateral portions of the caudal
shield are bent down.

The two pairs of antennce are of about equal length. In the anterior pair
the first joint is short and oval ; the second joint is a little longer, and the upper
margin projects a little way beyond the articulation of the succeeding joint, which is
nearly twice its length, narrow, and bent downwards. The filament has fourteen
joints.

The two distal joints of the basal portion in the second antennae are furnished with
hairs arranged in tufts along the upper margin. The filament has twelve joints.

In the maxillipedes (PI. III. fig. 10) the stipes and lamina are not separated by a
complete suture ; the second joint of the palp has a small tubercle neai" the base covered
with hairs as in the other deep-sea species.

The third thoracic appendages in the male resemble those of Serolis necBra in that the
inner margin of the third and fourth joints bears a number of delicate plumose hairs ;
a few are also to be found at the upper extremity of the second joint ; the penultimate
joint is swollen and considerably broader at the base; the inner margin has about a dozen
pairs of conical papilla-like spines which grow longer and narrower towards the distal
extremity, where they come to resemble exactly the plumose hairs upon the anterior
j oints.

REPORT ON THE ISOPODA. 63

The amhulatory limbs are rather slender, and the joints elongated ; the fourth is always
longer than either the preceding or succeeding joint ; the spines upon these appendages
are not very much developed ; the second and third joints in all bear a few pinnate hairs
upon the inner margin close to the articulation with the succeeding joint ; the remainder
of the limb is furnished with delicate sword-like spines arranged in tufts ujjon the inner
side ; mixed with these are a few strong serrated spines which are most abundant upon the
distal extremity of the penultimate joint. The inner side of the third, fourth, and fiftli
joints in the last ambulatory limb (PI. III. fig. 11) is covered with soft pinnate hairs (fig. 12),
like those found ou the second pair of thoracic appendages of the male, and the inner side
of the first ambulatory limb is similarly provided with these hairs, though to a less extent.

The three anterior abdominal appendages are like those of other species ; the liasal
joint is prolonged outwards, and bears two to three branched hairs at the tip.

The suture which divides the exopodite of the opercula is oblique ; the exopodite of
the gill appendage is bifurcate.

Station 120, September 9, 1873; lat. 8° 37' S., long. 34° 28' W.; 675 fathoms;
bottom, red mud.

10. Serolis antarctica, F. E. B. (PI. III. figs. 1-6).

Serolis anfardica, F. E. Beddard, Proc. Zool. Soc. Lond., 1884, pt. iii. p. 333.

The three species above described agree with each other and diS'er from all other known
species of the genus in the great development of the epimera. Serolis antarctica, the
only other deep-sea form, has com^jaratively short, flat epimera as in the shallow-water
species of Serolis. This species may be distinguished by the entire absence of eyes, and
by the strongly marked sculpturing upon the dorsal surface of the body.

The Challenger collection contains eight specimens besides some fragments. The
four male specimens are all approximately of the same length, the largest measuring
33 mm. in length by 31 mm. in greatest breadth. The remaining specimens, which are
females, vary in size; the largest measures 31 mm. in length by 26 mm. in greatest breadth.
The males, therefore, ap^aear to be larger than the females, and also broader in proportion
to their length ; the contour of the body in the males is oval, in the females somewhat
pear-shaped {cf. PI. III. figs. 1, 3); the epimera are proportionately longer in the
males, those of the sixth pair jjroject a little way beyond the end of the caudal shield;
the abdominal epimera are much the same length in the two sexes, and reach about half-
way down the caudal shield ; in the female the sixth thoracic epimera do not project
beyond the caudal shield ; the second pair of abdominal e})imera are a trifle longer than
the anterior pair, and both are very slightly shorter than the abdominal epimera in the
male. The sterna of the abdominal segments do not seem to diff"er in the two sexes.
The colour of this species (in alcohol) is violet-blue, iucliuiug to reddish yellow in the
middle of the body. The body has a nacreous glitter when seen by oblique light.

64 THE VOYAGE OF H.M.S. CHALLENGER.

Cephcdic Shield. — The cephalic shield is very slightly raised above the general
surface of the body ; its posterior surface is furnished with a short blunt tubercle ; a slight
ridge separates the antero-lateral portions from the rest, as in other species. The eyes
are entirely absent, their place being occupied by a small tubercle similar to the rest of
the chitinous integument, and containing no traces of any optic structures.

Tliorax. — The dorsal surface of the thorax, as well as the rest of the body, is very
much sculptured, more so than in any other species of the genus ; upon the epimera the
sculpturing takes the form of a network of ridges enclosing irregularly shaped roundish
pits ; the anterior half of the cephalic shield is sculptured in a similar fashion ; the posterior
part of the cephalic shield and the terga of the thoracic segments (especially upon their
posterior portions) are covered with thicker ridges anastomosing irregularly ; these are
most conspicuous upon the lateral portions of the terga of the three anterior free
thoracic segments. The sculpturing upon the terga of the abdominal segments is very
slight. The sculpturing upon the caudal shield differed in different specimens ; in some
it resembled that of the epimera, in others the central portion between the median and
lateral carinse was occupied by a number of roundish irregularly scattered tubercles.

The posterior margin of all the thoracic as well as the abdominal segments is pro-
longed in the median line into a short blunt tubercle ; these are larger upon the two last
thoracic and the abdominal segments.

The first epimera are not divided into two portions by a suture ; the succeeding
epimera are closely applied to each other, the extreme end alone projecting freely ; they
are rather different in shape from what is customarily found in Serolis; instead of rapidly
narrowang towards the free extremity and terminating in a fine point, the anterior
margin runs parallel to the posterior margin of the previous epimeron until just before
the termination of the latter, when it passes abruptly backwards in a direction nearly
parallel to the long axis of the body, and joins the posterior margin almost at right angles.
The fifth and sixth epimera are more like those of other species.

The epimera of the three anterior free thoracic segments are sej^arated from the terga
hj a suture.

The sterna of the first two thoracic segments have the usual form (PI. III. fig. 4); the
median sclerite of the first segment is keeled, that of the second segment has a short keel
extending from the anterior extremity to about the end of the first third. The sterna
of the three posterior thoracic segments are similar to those of the last described species
(cf. PI. III. fig. 2) ; the male genital pores are very closely approximated in the middle
line.

One of the male specimens jDresents an abnormality in the presence of a jjair of
ovigerous lamellre upon the third free thoracic segment ; the lamella of the right side was
about twice as large as that upon the left side.

Abdomen. — The three anterior abdominal segments have been already described ; the

REPORT ON THE ISOPODA. 6.3

caudal shield measures 12 mm. in length by 12 mm. in greatest breadth in the male speci-
men, and 11 mm. in length by 11 mm. in breadth in the female ; it is therefore more than,
one-third of the length of the animal ; its shape is more or less hexagonal ; the lateral
portions are strongly bent down, and the uropoda come to be attached quite on the under
surface, and are almost invisible from above ; there is a median longitudinal keel which
bifurcates at about the end of the anterior fifth, on either side is a Y-shaped keel inclined
at an oblique angle ; the portion of the caudal shield which lies between the median and
lateral keels is flat, the part which lies outside the inner fork of the lateral keel is bent
downwards ; the posterior end of the caudal shield is slightly bent up. ,

AntenncB. — In the anterior pair the two first joints are short and about equal in size ;
the third joint is narrow and long, twice the length of either of the preceding joints ; the
fourth joint is very short ; the filament has about twenty-eight joints, each of which has
two sensory hairs ; the filament is nearly as long as the basal portion of the antenna.

The second pair of antennae are longer; the filament has about twenty-one joints, and is
almost exactly equal in length to the last joint of the basal half of each of the antennae.

The mandibles are like those of other species in being asymmetrical ; the left bears on
the upper surface a chisel-like process and above a single spine ; the right has two spines.

The upper lobe of the Jirst pair of maxillce has ten spines upon its masticatory edge ;
the lower lobe has the usual form, and has but one slender spine close to the hinder
margin of the cutting edge.

In the maxillipedes, the stipes and lamina are completely separated by suture ; the
margins of both are smooth ; the second joint of the palp is comparatively slender ; its
inner edge is beset with hairs along the distal half; close to the middle of the joint is an
oval prominence as in the other deep-sea species.

One of the second pair of amhidatory limbs in the male is shown on PI. III. fig. 5.
The distal joint is as usual bent back upon the one in front like the blade of a closed
penknife upon its handle ; the penultimate joint is considerably broader at the base, and
bears about fourteen somewhat slender spines, arranged in pairs.

The remaining ambulatory appendages (PL III. fig. 6) are slender, and but scantily
furnished with slender spines, some of which are serrated ; these are only to be found
at the distal extremity of the third, fourth, and fifth joints, and extending for a short
way along the outer margin of the fourth and fifth joints.

The three anterior pairs of abdominal appendages have a tuft of two or three hairs
upon the projecting outer angle of the basal joint.

The suture upon the operculum is at right angles.

The uropoda are extremely small, and attached near the end of tlie lateral margins of
the caudal shield.

Station 122c, September 10, 1873; lat. 9" 10' S., long. 34° 49' W. ; 400 fathoms;
bottom, red mud.

(zooL. cn.u,L. EXP. — PART XXXIII. — 1884.) Kk 9

G6 THE VOYAGE OF H.M.S. CHALLENGER.

Station 146, December 29, 1873; lat. 46" 46' S., bug. 45° 31' E. ; 1375 fathoms;
bottom, Globigerina ooze.

Station 147, December 30, 1873; lat. 46° 16' S., long. 48° 27' E. ; 1600 fathoms;
bottom, Diatom ooze.

The above cited Monograph of Grube contains a description and figures of a species of
Serolis (Serolis tuherculata) which differs from all the species known at that time and
from all those already described in the [iresent Report by the characters of the fifth and
sixth thoracic segments ; the tergum of the fifth segment, wliich is generally narrower
than the preceding ones, is in this species extremely narrow, not measuring more than one-
sixth of the diameter of the segment in front, wliile the tergum of the sixth thoracic
segment has entirely disappeared : the Challenger obtained two specimens of this same
species, Serolis tubercidata, besides examples of four other species which agree witli
Serolis tuherculata in the characters just mentioned ; all these species are inhabitants of
the shallow waters off the southern and eastern coasts of Australia, and form a well-
marked group, agreeing with each other in a number of structural points. These species
I have briefly described in my " Preliminary Report," and named as follows : — Serolis
■pallida, Serolis australiensis, Serolis elongata, Serolis minuta, and Se)-olis longicaudata ;
all these species, with the exception of Serolis minuta, agree with each other and with
Serolis tubercidata in the following points, some of which are peculiar to the group,
wliile others again are not confined to the group, but are also to be found in other species.

They are all of small size ; the females are larger than the males (?). The thoracic
epimera are short and closely applied together for their whole length, while the epimera
of the two abdominal segments are very short and not prolonged beyond the anterior
margin of the caudal shield. The tergum of the fifth thoracic segment is extremely
narrow ; the tergum of the sixth segment is obsolete in the middle line, the suture whicli
separates it from the succeeding first segment of the abdomen passes forwards and
disappears underneath the segment in front {cf. PI. VI. fig. 1) in Serolis tubercidata and
Serolis pallida ; in Serolis australiensis, Serolis elongata, and Serolis longicaudata the
general appearance of the two last thoracic terga is the same, but a careful inspection
shows that the posterior sutures of both segments become obsolete just before the middle
line of the body, so that which apparently is the tergum of the first abdominal segment
in reality includes also the middle part of the terga of the two last thoracic segments.
In Serolis minuta the fifth and sixth thoracic segments are not quite so narrow as in the
other Australian species. The fifth segment is divided off by a sutural line which is
entirely continuous from one side of the body to the other ; the sixth segment, however,
though proportionately somewhat broader, resemljles that of Serolis australiensis, &c., in
being fused mesially with the succeeding abdominal segment. In the other Australian
species the rostrum is long, reaching beyond the first joint of the anterior pair of

REPORT ON THE ISOPODA. 67

autcniue. The sterna of the abdominal segments are prolonged into stout spines in
both sexes. The caudal shield is truncated at its extremity. The proximal joint of the
anterior abdominal appendages is narrow, and, as in Set'olis schythei, the basal triangular
process furnished with branched hairs is aljscnt ; the suture on the operculum is at right
angles to the long axis of the body. The palp of the maxillipede has this peculiarity,
tliat tlie second joint, instead of being cordate in shape, has the two margins curved
inwards and parallel with each other.

This assemblage of characters is very distinctive of all the Australian species with the
sole exception of Serolis mimtta ; this latter appears to be intermediate between the
other Australian species and the more typical species of Serolis such as Sei'olis paradoxa;
in the characters of the last thoracic segments, as above stated, it more closely resembles
the former, while in other respects it comes nearer to the latter; it would be difficult to
classify it definitely with either ; for this reason, and also considering the compactness in
other respects of the genus, it appears to me inadvisable to divide Serolis into two
sejiarate genera or subgenera.

11. Serolis tuhercidata, Grube (PI. VI. figs. 1, 2).

Serolis tuhercidata, Grube, Arcliiv f. Naturgesch., 1875, p. 227.

The Challenger obtained two specimens of a small species of Serolis, which I identify
with Grube's Serolis tuherculata ; one of these specimens is a male, the other a female ;
they were obtained from different localities, the male from Station 161, the female from
Station 162 ; tlie former specimen is unfortunately much damaged.

The female is larger than the male — it measures 19 mm. long by 17 mm. lu-oad ; the
length of the male is 12 mm., its breadth about the same. It appears therefore that the
males of this species, if not smaller than the females, are broader in proportion to their
length, as is generally the case in this genus. The colour of the two specimens is rather
different ; the female is pale yellowish brown, with innumerable black dots ; on the
outside of each epimeron is a larger black spot ; the male is of a uniform Init darker
brown.

I have but little to add to Grube's excellent description of this species ; in my
specimens the epimera are not so closely applied together as he describes, and this is
especially the case in the male, where the points of all the epimera project freely.
On the cephalic shield between and in front of the eyes is a semicircular ridge of the
integument ; the convexity is directed forwards, and the two lateral ends are prolonged
backwards into two short spines; the lateral tubercles on the thoracic segments are eight
or nine in number on either side, and are situated on a low ridge which commences near
the junction of the tergal portion of the segment with its epimeral portion, and terminate
a little before the middle of the segment ; this ridge is longer in the third free thoracic

68 THE VOYxVGE OF H.M.S. CHALLENGER.

segment than in the rest. The two last thoracic segments have only one tubercle, on
either side, which form the outer extremity of a slight ridge.

AntenncB. — The first pair of antennae has a stem composed of four joints, of which
the thii-d is rather more than twice as long as either of the two basal joints ; the filament
consists of some fifty joints ; in the second pair of antennte (PL VI. fig. 2) the third
joint has its posterior border prolonged into a spine as figured byGrube;' the upper
surface of this and the succeeding joint has a strong longitudinally running ridge ; the
upper margin of the fourth and fifth joints, which are as usual the longest, is sinuous, and
.a tuft of hairs springs from the summit of each of the elevations ; the filament has about
twenty joints.

In the maxillipedes the stipes is almost triangular in shape ; the lamina is pro-
portionately very stout and strong, its diameter being equal to that of the stipes. Just
below the articulation of the palp is a curved ridge ; the second joint of the palp, instead
of being heart-shaped, has its two sides almost parallel, the curvature of the outer margin
following that of the convex inner margin.

The third pair of thoracic appendages is quite, as described by Grube, but I am
inclined to think that both his specimen and mine are immature males {ante, p. 16).

The ambulatory limbs of this species are characterised by being mainly provided with
one kind of spines which are of varying length, but always stout and strong. The end is
bent and seems to be of a somewhat softer consistency than the rest. The stout serrated
spines, so characteristic of the ambulatory limljs of other species of Serolis, are only
present in small numbers in Serolis hiherculata.

The three abdominal appendages are hardly difi"erent from those of other species ; the
basal portion is, however, not prolonged into an angle bearing two or three plumose hairs
as in many other species ; in this respect they agree with Serolis australiensis, &c. The
exopodite of the first pair of gills is divided into two by a transverse suture, which is at
right angles to the long axis.

Generally speaking, the appendages of Serolis tuherculata are very closely similar to
those of the next species to be described.

A figure of the female is given on PI. VI. fig. 1.

Station 161, April 1, 1874; off" East Moncoeur Island, Bass Strait; depth, o8
fathoms ; bottom, sand.

Station 162, April 2, 1874 ; off" East Moncoeur Island, Bass Strait; depth, 38 to 40
fathoms ; bottom, sand.

' Loc. cit., 1)1. V. fig. 2.

REl'ORT ON THE ISOrODA. 69

12. Serolis australiensis, F. E. B. (PL VI. figs. 3-8).

Serolis australiendx, F. E. Beddard, Proc. Zool. Soc. Lond., 1884, pt. iii. p. 334.

Of this species three examples were obtained by the Challenger off the coast of South
Australia, and I have been able to compare them with a specimen in the British Museum
lirought from the same locality.

The largest specimen is a female (PL VL fig. 4) and measures 14 mm. in length by
1 1 mm. in greatest breadth; the two remaining specimens are both males (PL VL figs. 3, 7),
and are approximately of the same size, measuring 10 "5 mm. in length and 9 mm. in
breadth.

Apart from the form of the third thoracic and second abdominal appendages, I could
detect no marked differences between the two sexes, unless it be that the females are
really larger than the males in this and in the other Australian species.

The general form of the body of Serolis australiensis is oval, the male a little broader
proportionately ; the distinguishing feature of this species is the immense number of
tubercles which cover the body, and are especially large upon the caudal shield and the
posterior margin of the segments.

The cephalic shield is broadest at the level of the eyes, where it bulges out considerably
on either side. Anteriorly and posteriorly it is narrower ; the anterior margin is prolonged
into a very long rostrum ; there is a transverse ridge which forms the anterior margin of
the caudal shield for a short distance on either side of the rostrum, and then bifurcates,
the outer branch continuing along the anterior margin of the cephalic shield and giving
off a short spine directed forwards at the level of the end of the first joint of the anterior
antennse ; the inner branch follows the margin of the cephalic shield, but at some little
distance from it, and the two unite at the lateral anterior angle, enclosing between them
a somewhat boat-shaped depression. The tubercles on the cephalic shield are arranged
in transverse rows ; there is a larger spine just to the inside of the posterior third of the
eye on either side directed backwards and slightly outwards, and a median spine about
the same size situated near the posterior margin of the cephalic shield.

TJiorax. — The epimera of the first segment are not divided by a suture. The three
following epimera are closely applied to each other along their whole length ; there is a
slight break between the four anterior and the two posterior epimera ; the two latter are
closely applied to each other, but the outer margin of the fifth begins to curve backwards
a little before the outer termination of the fourth epimeron, so that the angle of the latter
projects freely. All the thoracic epimera, with the exception of the first, are separated by
a suture from the tergal portion of the segments.

The terga of the thoracic segments, as well as the epimera, are covered with minute
tubercles, which are distributed in longitudinal rows running from one side of the segment
to the other ; one row, which is constant in all the segments, and is rather more conspi-

70 THE VOYAGE OF H.M.S. CHALLENGER.

(•nous than tlic rest, forms the posterior boundary of the segment, and is continued along
the posterior margin of the epimera. In front of this is another row of tubercles, which
in the middle of the segment lies half way between the anterior and posterior margins ; on
cither side it divides into two rows, each, of wliich run close to the anterior and posterior
margins of the segment respectively ; in the two last thoracic segments this anterior row
of tubercles is not present. All the segments of the body, with the exception of the fifth
and sixth, the former of which, as in the other Australian species, disappears altogether in
the middle line, are furnished with a larger tubercle exactly in the median dorsal line ; these
increase in size from before backwards ; on the fifth segment (in the female specimen at
least) the median tubercle is wide, and flattened and quadrifid at its extremity.

Ahdomen. — -The epimera of the second and third abdominal segments extend for a
very short space beyond the lateral margins of the caudal shield ; beneath, the three
anterior segments are prolonged into a stout siDine of unusual length {cf. fig. 3) ; the
spine of the first segment, which is the longest, completely covers the spine of the
succeeding segment ; the spines in the female apjDcar to be a little stouter than in the
male, for in the male the spines are all rather slender ; the diff"erences, however, are
insignificant and not well marked.

The caudal shield is more or less triangular in outline, distinctly carinate and trun-
cated at its free extremity ; its upper surface is covered with tubercles, which have no
definite arrangement except for a median transverse row which crosses the central keel
at right angles a little below the attachment of the uropoda ; one of these tubercles on
either side, close to the lateral margin, is considerably larger than the rest.

Of the Jirst pair of antennoB the two proximal joints are short, while the third, wliich
is the longest, measures about twice the length of the second. The joints of the filament
are extremely short and numerous ; there are about fifty ; each joint bears two sensor}'
filaments instead of the usual one (PI. VI. fig. 5).

In the second, iKiir of antemuB the third joint has a large backwardly projecting
tubercle just before its articulation with the following joint as in Serolis tubercnlata ;
this joint is also smaller than the preceding one; the fourth and fifth joints are as usual
greatly elongated, and the outer margin is sinuous. A tuft of fine hairs springs from the
surface of each of the four elevations ; the filament consists of fifteen joints, which are
longer and more slender than the joints composing the filament of the first pair of
antennae; the upper surface of the third to the tenth joints has a row of short blunt
tubercles; one or two long fine hairs springs from the inner margin of each joint at its
anterior extremity.

MontJi Appendages. — The mandibles, like those of Serolis jMlIida, have a large and
conspicuous tooth upon the posterior part of the masticatory edge ; behind this is
another smaller tooth; in front the margin .slopes back gradually to meet the outer
margin of the mandilale.

REPORT ON THE ISOPODA. 71

The first maxiUce are sliort and rather beut, and resemble those of Serolis pallida.

Thoracio Appendages. — The maxiUlpedes (PI. VI. fig. 6) closely resemble those of
Serolis pallida and the other Australian species, but are not sculptured like those of the
former; tlie lamina terminates on the inner side in the usual way; on the outer side it is
produced into a rounded prominence, which extends some way beyond the notch on the
inner side; just behind the articulation of the palp is a broad ridge as in Serolis pallida ;
the anterior margin of the stipes is crenate; the second joint of the palp is not so distinctly
cordate in shape as in many species, and resembles in this particular the other allied
Australian forms.

The thoracic appendages are furnished with two kinds of spines — (1) short broad
serrated spines like those of Serolis schi/thei and many other species, (2) longer and more
delicate spines, which are somewhat thicker on the outer edge than on the inner; the
extremities are bent inwards; these spines seem only to be found on the appendages of
the Australian species. The proximal joint of all the ambulatory limbs has a number of
fine branched hairs on the inner surface like those of Serolis necera and other species; the
second joint is a little shorter, and has one long hair on the inner side just before its
articulation with the succeeding joint; the third, fourth, and fifth joints are short, but
increase in length up to the fifth. One of the thoracic appendages is shown in fig. 8 ; the
others are similar except that the three penultimate joints increase in length in the
posterior limbs; in the appendage figured it will be seen that the spines on the outer
side arise from step-like processes, on the three penultimate joints the number of these
steps gradually increases in the posterior appendages. In the male the second thoracic
apjjendage, as in aU other species, is modified into a prehensile organ ; the penultimate
joint is rather more elongated than usual, and its two sides are almost parallel ; on
the inner surface I counted in one specimen nine of the hairs peculiar to this joint.

The first three pairs of abdominal appendages have no hairs upon the basal joint,
and the two sides of this joint are nearly parallel owing to the absence of the inner
projecting angle.

Station 162, April 2, 1874; lat. 39' 10' 30" S., long. 146° 37' E.; 38 fathoms;
bottom, sand and shells.

13. Serolis elongata, F. E. B.

Serolis alungata, F. E. BedJari.1, Proc. Zool Soc. Loud., 1884, pt. iii. p. 335.

Although I have been able to examine only a single female example of this species,
its characters appear to me to be sufticiently diflerent to retain it as a distinct species.

The specimen is a female with fully developed ovigerous lamellae ; its greatest length is
10 mm., its greatest breadth Q'b mm. It is most nearly allied to Serolis australiensis.
from which, however, it difiers in several respects.

72 THE VOYAGE OF H.M.S. CHALLENGEE.

From the above measurements, when compared with those of Serolis anstraliensis
(ante, p. 68), it will appear that Serolis elongata is rather narrower, the length Iseing
proportionately greater than the breadth.

The main difference, however, between the two species is the development of
tubercles upon the dorsal surface of the body ; Serolis australiensis is to be distinguished
from all other species of the genus by the immense number of tubercles upon the
segments of the body, and also upon the caudal shield ; in Serolis elongata the dorsal
surface of the body is by comparison almost smooth ; this cannot be owing to the
difference of age ; the female specimen of Serolis elongata, although actually smaller
than Serolis australiensis, is perfectly mature, with fully-developed ovigerous lamellae ; and
as no other species that I have been able to examine undergoes any change, except mere
increase in size, as soon as it has attained to maturity, there is no reason to suppose that
Serolis elongata is peculiar in that respect. As in Serolis australiensis, each of the
segments of the body is furnished with a curved hook-like spine in the middle line ; a
row of short tubercles occupies the hinder portion of each of the segments, and is
prolonged on to the epimera ; besides these there appear to be a few scattered
tubercles over the rest of the segments and the epimera, which are very inconspicuous,
and contrast with the strongly tuberculated surface of the body in Serolis australiensis.

The caudal shield has the same general shape that it has in the last mentioned species,
with serrated margins and a longitudinal carina. There is also a lateral carina on either
side bearing a short spine, which is situated about half way down the caudal shield, and a
little below the place where the uropoda are attached, and terminating at the end of the
body by becoming fused with the central carina ; as in Serolis australiensis, these lateral
earinse are serrated. Between these and the central carina is a short rido;e running
obliquely towards the margin of the caudal shield from a point a little below and to one
side of the commencement of the central carina : the general surface of the caudal shield
is smooth, and there are only present a few scattered tubercles, especially developed in
the neighbourhood of the two lateral carinse.

Port Jackson, Sydne)^ 30 flxthoms.

14. Serolis longicaudata, F. E. B. (PI. VII. figs. 8-10 ; PI. VIII. figs. 1, 2).
Serolis longicavdata, E. E. Beddard, Proc. ZooL Soc. Lond., 1884, pt. iii. p. 336.

Of this species the Challenger obtained one immature female; its length is 7 mm.,
its breadth 5 mm.

The general form of the body is peculiar, and unlike that of the typical memliers of
the genus; the anterior portion of the body, comprising the head, thorax, and free abdominal
segments is almost completely circular, and the caudal shield projects backwards for a
considerable distance, being proportionately longer than in any other of the species

REPORT ON THE ISOPODA. 73

known to me, while the epimera are all short and abruptly truncated, reaching in no case
beyond the lateral margins of the caudal shield.

The cephalic shield terminates in front in a comparatively long rostrum ; Ijchind, the
suture which separates it from the thoracic segment is incomplete in the middle line, and
for a space on either side of it extending to nearly as far as the level of the eyes ; the
cephalic shield as well as the rest of the body is quite smooth, and free from tubercles.

Thorax. — The epimera of the first thoracic segment are entire and devoid of any trace
of a transverse suture ; the epimera of the remaining thoracic appendages are very short,
and have the appearance of being truncated at their free extremity; the anterior margin
of the epimeron, instead of passing insensibly into the outer lateral margins, meets it almost
at right angles, and the outer margins run backwards in a direction almost parallel to the
long axis of the body ; the epimera of the three anterior free thoracic segments are
separated by a distinct suture from their terga. The epimera are extremely short, as may
be seen in PI. VIII. figs. 1, 2, where this species is figured. The two last thoracic
segments have the same characters that are peculiar to the other Australian species
of the genus ; the tergum of the fifth is extremely narrow, not more than one-fourth
the breadth of the segment in front ; the tergum of the sixth segment is entirely absent.
The sutures which separate both segments from the succeeding and preceding ones
entirely disappear close to the middle line of the body.

The outer margins of all the thoracic epimera are faintly serrated. Ventrally the middle
portion of the segments is elevated into a broad ridge which becomes higher towards the
middle line, and slopes off gradually in the direction of the epimera ; in the sixth
segment this ridge is still more marked, and being developed upon its posterior border
overlaps the terminal thoracic segment. The terminal segment of the thorax bears on
either side, close to and just above the attachment of the first abdominal limb, a circular
aperture which may be the outlet of some gland. I observed a similar pair of orifices
in Serolis pallida and in several other species.

Abdomen. — The epimera of the second and third abdominal segments extend as far
as the edge of the caudal shield ; the outer margin of the second is concave, that of the
third nearly straight. The ventral portion of the first three segments is furnished with
a central triangular keel which projects some way back as a stout conical spine ; the
spine is largest upon the first segment, and its cavity communicates with the exterior by
several large fenestrge upon the lower surface (PL VII. fig. 8).

The caudal shield has a somewhat pentagonal outline, and terminates in an abruptlj"
truncated extremity; the dorsal surface has a middle and two lateral carinse; the portion
which lies beyond the latter is strongly bent down ; the postero-lateral margins are
slightly denticulate.

Appendages. — The antenncB are displayed in figs. 9 and 10 of PI. VII. The anterior
pair (fig. 9) are rather shorter than the second pair ; their terminal filament has thirteen

(ZOOL. CUALL. EXP. — PAET XXXIII. — 1 884.) K.k 10

74 THE VOYAGE OF H.M.S. CHALLENGER.

joints. lu the posterior antennae (fig. 10) the filament has ten joints; the two Last
joints of the basal portion of the antenna are furnished with tufts of fine hairs springing
from shallow depressions on the anterior surface.

The ambulatory limbs are remarkable in that they are only furnished with com-
paratively slender, soft, bluntly terminating spines; no serrated spines appear to be
present. The inner side of the first joint of all these appendages has a row of about a
dozen fine branched hairs entirely like those found in many other species ; the second
joint in all is rather smaller than the first, and has on the inner side just before its
termination a single long slender spine ; the remaining joints are subequal in size and
comparatively short ; the fourth and fifth joints in the penultimate pair of appendages
are longer and narrower than in the preceding limb, and about half as long again as the
third joint.

The last pair of thoracic appendages is as usual small.

The three first abdominal appendages have the basal portion comparatively long
and narrow, and as in other Australian species the inner and lower margins are not
prolonged into a triangular process furnished with two or three branched hairs.

The fourth piair or opercula have the exopodite divided by a suture at right angles to
the longitudinal axis, and terminating exactly at the level of the attachment of the
uropoda ; the latter are attached at about the middle of the caudal shield, and extend
exactly as far as its termination; the endopodite is slightly shorter, and at the same time
slightly broader than the exopodite ; the margins of both are smooth, and only slightly
crenate at the distal end ; they bear no branched hairs.

Station 161, April 1, 1874 ; lat. 38° 22' 30" S., long. 144° 36' 30" W. ; 33 fathoms ;
bottom, sand.

15. Serolis pallida, F. E. B.(Pls. VII. fig. 1 ; VIII. figs. 6-16).

Serolis pallida, F. E. Beddard, Proo. Zool. Soc. Lond., 1884, pt. iii. p. 335.

Of this species two specimens were obtained, a male and a female. The female is the
larger, measuring 16 mm. in length and 13 mm. in breadth; the male measures 9 mm.
in length and 7 mm. in breadth.

The body is oval, somewhat pear-shaped, and recalls Serolis convexa ; as in that
species the epimera are closely applied to each other, and only in the last three thoracic
epimera are the extremities freely projecting. The colour (in alcohol) is a uniform pale
brown, with two darker patches on each side of the third free thoracic segment, the
anterior portion of the caudal shield is whitish grey. The surface of the body is quite
smooth and free from tubercles except for a series, one to each segment, occupying the
median line.

REPORT ON THE ISOPODA. 75

The cephalic shield is almost triangular in shape, from the great development of the
ocular prominences and a large median backwardly projecting spine ; the rostrum is long
and slender, and reaches nearly as far as the distal end of the second joint of the posterior
antennae ; the anterior margin of the cephalic shield is thickened into a ridge, which
bifurcates at the edge, the two branches being directed forwards and enclosing a deep
cup-like space ; shortly before the bifurcation a small spine-like process is given off
anteriorly.

Thorax. — The epimera are short, gradually increasing in length up to the sixth pair ;
the first epimera are entire as in the other Australian species, and not divided by any
transverse suture ; at the u^jper extremity is a short ridge on either side, which extends
for about one-sixth of its length parallel to the anterior margin of the epimeron.

In the three succeeding epimera the curvature of the outer margin is very slight,
which gives them the appearance of being abruptly truncated. The fifth epimera are more
sickle-shaped, since the outer is inclined at a smaller angle to the posterior margin ; the
outer margin of the sixth epimera is concave.

The terga of the thoracic segments, with the exception of the first and the two last,
have a median hooked spine directed upwards and Imckwards ; these increase slightly in
size from before backwards ; the antero-posterior diameter of three anterior free thoracic
segments is about the same ; the fourth is not more than one-fifth as wide as the
preceding segments, and the tergal portion of the last thoracic, as in other Australian
species, has entirely disappeared.

Abdomen. — The epimera of the second and third abdominal segments only reach as
far as the margin of the caudal shield ; they are equal in length. The terga of the three
segments are furnished with a spine like that in the thoracic segments ; these spines,
which are rather smaller than those on the thorax, increase in size from before backwards.
The sterna of these segments are produced into a long median spine, which is larger in
the first segment than in the two succeeding ones.

The caudal shield in the female measures 5 '5 mm. in length, and is therefore rather
more than one-third of the length of the entire animal ; it is irregularly hexagonal in
outline, and keeled ; the extremity is notched ; the uropoda are attached about half way
down ; just below and to the inside of the spine-like process which covers the articulation
of the uropoda is a minute flattened spine on either side.

The two pairs of antennce are approximately of the same length. The basal joint of
the first pair (PL VIII. figs. 7, 8) has three strong tubercles on its upper surface ; the
second joint has a stout tubercle on the posterior margin projecting backwards ; the
filament has twenty-four joints, each of which is furnished with two sensory hairs (fig. 8),
as in Serolis australiensis.

In the second pair of antennae (PI. VIII. fig. 6) the fifth joint is very much enlarged ;
it is not quite so long as the preceding joint, but wider; the filament is short, being

76 THE VOYAGE OF H.M.S. CHALLENGER.

composed of nine joints. The second and third joints of the antennae are raised into
tubercles on the upper surface.

Mouth Appendages. — The mandibles (PI. VIII. figs. 9, 10) as well as the maxillipedes
(fig. 12) are much sculptured, the under surface being raised into a number of ridges
separated by deep depressions as shown in the figure ; the masticatory edge, instead of
being comparatively straight and smooth, as in the majority of species, is denticulate, one
tooth on its posterior boundary being specially prominent. On the upper surface (fig.
10) are two spine-like processes.

The maxilloB do not difi'er much from those of other species ; the anterior pair
(fig. 11) have a short basal joint, and the terminal joint is much bent.

The maxillipedes are shown on fig. 12 of PL VIII. ; the under surface of the basal
portion, instead of being flat and smooth, as is ordinarily the case in Sewlis, is traversed
by a number of ridges, the direction and form of which are displayed in the figure ; the
second joint of the palp has the peculiar shape that it has in Serolis australiensis. The
outer margins, instead of being convex, are concave and almost parallel with the inner
margins ; they are traversed from end to end by a ridge, close to and quite parallel with
the outer margins.

The second pair of thoracic appendages are shown on fig. 13 of PL VIII., and
several of the spines from the inner surface of the penultimate joint on figs. 14, 15 of the
same plate. The latter are very characteristic in shape — the longer spines terminate in
two large oval expansions traversed by longitudinal strife which converge at the proximal
extremity ; the anterior of the two is shorter and broader than the posterior ; but at the
hinder end of the joint they become much smaller and nearly equal in size and similar
in shape ; between the two wing-like processes the axis of the spine is continued into a
thicker cylindrical tapering extremity which reaches a trifle beyond the level of the
posterior wing-like process.

The third pair are in the male modified in the ordinary way (PL VIII. fig. 16).
The penultimate joint is oval, and furnished on the inner surface with six or seven pairs
of cylindrical spines, the axis of which is prolonged into a short filiform process. The
first joint, as in the succeeding ambulatory liml), has a row of fine branched haii-s on the
inner side.

The remaining thoracic appendages are stout and strong, and furnished with serrated
spines, especially upon the outer surface ; the second joint has invariably a single slender
spine springing from al)out the middle of the inner surface; the thml, fourth, and fifth
joints increase progressively in size, the fifth joint being nearly or quite as long as the
second, though narrower.

The three Jirst ahdominal appendages have no inner angle bearing hairs upon the
basal joint.

The suture which traverses the exopodite of the operculum is at right angles

EEPORT ON THE ISOPODA. 77

to the longitudinal axis, and at the level of the end of the first joint of the
uropoda.

The w'Opocla are attached just before the end of the anterior half of the caudal shield;
the exopodites are longer and reach very nearly up to the distal extremity of the caudal
shield ; they are oval in shape, and terminate in a blunt extremity ; the outer margin is
serrated nearly as far as the articulation ; the endopodite is relatively broader, and ends
in a truncated notched extremity, the outer and the posterior margins are serrated ; no
hairs were found upon these appendages, but it is very probable that they may have
become detached.

Station 162, April 2, 1874; lat. 39° 10' 30" S., long. 146° 37' K; 38 fathoms;
bottom, sand and shells.

Station 163, June 3, 1874 ; off Port Jackson ; 30 to 35 fathoms.

16. Serolis minuta, F. E. B. (PI. VII. figs. 2-6).

Serolis minuta, F. E. Beddard, Proo. Zool. Soc. Lond., 1884, pt. iii. p. 337.

The Challenger collection contains only one specimen of this species, which is a male ,
and measures 5 mm. in length and 5 mm. in diameter ; this species, therefore, if
full grown, is the smallest known.

In general shape Serolis minuta resembles very closely Serolis coniuta ; the outline
of the body is almost circular, and the caudal shield projects only a little way beyond the
circle. A conspicuous keel runs down the middle of the body.

The cephalic shield terminates in front in a short and stout rostrum ; the anterior
portion, as in most other species, is traversed by a ridge {cf. PL VII. fig. 2) arising from
the base of the rostrum ; the ridge is continued on to the epimera ; the posterior margin
is provided with three blunt tubercles, of which the middle one is the largest ; each of
the following segments is similarly produced into a blunt tubercle at the median point
of the posterior margin.

Tliorax. — The, first thoracic segment is entire, and not separated into two portions by
a transverse line of division as is the case in so many other species. The dorsal surface
of the remaining segments is prolonged on either side into a triangular process which
slightly overlaps the succeeding segment ; these processes, which are hollow, serve for the
attachment of the muscles moving the limbs ; they are situated on the tergal portion of
the segment close to its junction with the epimeral portion ; while those of the third,
fourth, and sixth segments are the largest and most conspicuous.

The tergal portion of the second, third, and fourth segments is separated by a distinct
suture from the epimera. The sixth segment is partially fused with the first abdominal
segment, the suture which divides them being incomplete for a short space on either
side of the median fine ; its epimera extend about as far as the first third of the caudal

78 THE VOYAGE OF H.M.S. CHALLENGER.

shield ; the outer margins of the epimera iu all the segments are smooth, without
serrations or hairs. On either side of the male generative pores is a slit-like orifice as
in Serolis pallida, &c.

Abdomen. — The second and third segments are provided with epimera which extend
farther than the epimera of the sixth thoracic ; those of the third segment reach nearl)^
as far as the middle of the caudal shield. The ventral portion of the three anterior
abdominal segments is shown on PL VII. fig. 2', together with the appendages belonging
to them ; they are oblong in shape, and each overlaps the succeeding one ; the posterior
margin of the first is almost straight, in the others slightly convex.

The caudal shield is almost triangular in shape, and ends in a blunt prolongation ;
the upper surface slopes gently downward on either side from the central keel ; the
lateral margins commencing from the attachment of the terminal appendages curve
inwards and then slightly outwards, becoming almost parallel just before their termination.

Antennce. — One of the antennae of the first pair is figured on PI. VII. fig. 4 ; it
consists as usual of four joints and a terminal filament ; the proximal joint is round and
entirely free from hairs ; the second joint is the largest, its lower surface is furnished with
a row of short hairs which are continued on to the succeeding joint ; the third joint is
smaller than the second, and the fourth is still smaller. The filament is made
up of ten joints, of which the first eight are subequal ; the penultimate joint is very
small, and the last slender and elongated. The second pair of antennae (PI. VII. fig. 5)
are about one-third longer than the anterior pair ; the basal portion consists of five joints,
of which the last is the longest ; the three last joints are furnished with bundles of hairs
arranged irregularly over the lower surface. The filament is made up of ten joints, and
is almost exactly of the same length as the filament of the anterior antennse, which is an
unusual circumstance.

The mandibles present the usual form, and terminate in a Ijlunt masticatory edge.

Thoracic Appendages. — The large prehensile organs which form the second pair of
thoracic appendages are in general form entirely similar to those of other species, but the
spines developed upon the inner margin are as usual distinctive ; several of these are
shown in PI. VII. fig. 7.

The third pair are modified into a prehensile organ which presents some peculiarities
distinguishing it from the same appendage in other species of Serolis. Fig. 3 represents
the last three joints of the right hand appendage viewed in profile and from beneath.

The terminal joint is furnished with a remarkable tongue-shaped process arising from
the inner side close to the distal end ; along this runs a median crest like the micbib of
a leaf. The penultimate joint has five pairs of conical bent spines which are almost
exactly similar in shape to those of Serolis convexa. The fourth joint has a bundle of
fine closely-set long hairs on its inner surface, as is the case iu Serolis necerxt ; the third
joint has several smaller bundles of precisely similar hairs separated by intervals from each

REPORT ON THE ISOPODA. '79

other ; this recalls the same appendage in Serolis necera, where, however, the third joint
is completely covered on its inner side with a dense bundle of hairs.

The succeeding thoracic appendages diifer from the same appendages in all other
species of Serolis, by the fact that they are provided only with long slender hairs ; the
serrated spines so generally found on the ambulatory limbs are entirely absent. The
first ambulatory limb of the left side (fourth thoracic appendage) is displayed on PI. VII.
fig. 6. As in other species, the proximal joint is the largest and has no hairs ; the second
joint has a few hairs on the inner surface; the three following joints are furnished with
hairs arranged in bundles as shown in the figure, and more abundantly developed upon
the outer than upon the inner edge.

The remaining ambulatory limbs increase slightly in size with the exception of the
last, which, as in other species, is the smallest of the series. In the structure and the
arrangement of the hairs upon the several joints, they present only a few very slight
difierences from the one figured ; for example, the second joint bears a single long hair
upon its outer surface, except in the last pair, where there are three, separated from each
other by equal intervals.

The first three ahdominal appendages present no special characters worthy of remark,
except that the basal joints are entirely unprovided with hairs, and resemble therefore
Serolis schythei.

The terminal pair of abdominal appendages — the uropoda — are attached close to
the upper end of the caudal shield ; both the distal joints are narrow and elongated,
and slightly serrated upon the outer margin ; the inner joint is the shorter of the
two, and its posterior margin is markedly dentate, and bears a number of hairs.

The colour of this species (in spirit) is brown, owing to the very slight development
of pigment, which is chiefly concentrated along the lateral portions of the terga and the
marginal portion of the caudal shield.

The specimen was dredged at Station 161, ofi" the entrance to Port Philip, Australia,
on Aprd 1, 1874, in 38 fathoms.

80 THE VOYAGE OF H.M.S. CHALLENGER.

GEOGRAPHICAL AND BATHYMETRICAL DISTRIBUTION.

The geographical range of the genus Serolis is almost entirely restricted to the
southern hemisphere ; so far as is known at present, this is the case with the abyssal
as well as the shallow- water species ; there is only one exception in Serolis carinata, a
species which has been described by Lockington/ and said to occur so far north as at San
Diego in California.

The shallow-water species of Serolis, with the exception of Serolis caiinata and a
more doubtful exception, Serolis 'paradoxal are confined to the Antarctic area.

AVithin this area there appear to be four centres of distribution, corresponding in fact
to all the land masses that lie within it — (l) the shores of South America as far north as
lat. 30°, the Falkland Islands and the South Shetlands ; (2) Kerguelen, the Crozets,
and Marion Island ; (3) New Zealand (?) ; (4) the shores of Southern and Eastern Australia.

A Kerguelen species, Serolis latifrons, is also known to occur at the Auckland
Islands, off the south of New Zealand, a single sjDecimen having been obtained at
Rendezvous Cove in that island during the voyage of the " Erebus " and " Terror." It
seems very probable also that the genus will eventually be found to inhabit the shores of
New Zealand, though at present there is no certain evidence to that effect. Miers, in his
list of New Zealand Crustacea, includes Serolis paradoxa, apparently on the authority
of a specimen in the British Museum, and the same collection of Crustacea contains a
single example of a species which I hav,e identified with the Patagonian Serolis schythei,
and which is labelled " New Zealand" ; in both these cases, however, I believe that the
locality is not authenticated beyond a doubt. Considering the general similarity between
the Crustacean fauna of the whole " Antarctic region " from Patagonia to New Zealand,
it seems very probable that Serolis is an inhabitant of the shores of New Zealand. I have
called attention later (p. 82) to the fact that Serolis hromleyana occurs ofi" the shores of
New Zealand in cZeej^ water.

From the shores of South America seven species have been described ; these are
Serolis paradoxa, Serolis trilohitoides, Serolis gaudichaudii, Serolis plana, Serolis schythei,
Serolis convexa, and Serolis serrei ; of these species Serolis gaudichaudii extends farther
north than any of the rest; the original specimen was obtained by M. Gaudichaud^
near Valpai'aiso. Cunningham* also mentions that he obtained it at the same locality.
Serolis schythei was dredged during the voyage of the Challenger as far north as the
Gulf of Penas on the west coast ; it also occurs in the Strait of Magellan and at the

^ Loc. cit.

2 This species is stated by Dr. Leach, on the authority of Dufresne, to inhabit the shores of Western Africa,
about the river Senegal, but it has never subsequently been obtained from that locality.
5 Audouin and Slilne-Edwards, Arch. d. Mus. d'Hist. Nat., loc. cit., p. 25.
■■ Loc. cit.

REPOET ON THE ISOPODA. 81

Falkland Islands. Serolis tn'lohitoides is said by Eights to inhabit the coasts of the
South Shetland Islands, as well as the neighbourhood of Cape Horn; the remaining species,
seem to be restricted to the Strait of Magellan and the extreme south of Patagonia.

It is possible that Serolis trilobitoides will eventually prove to be the same species
as Serolis cornuta from Kerguelen ; the two are at least very closely allied.

With this exception the species that inhabit Kerguelen are in every case quite
different from the South American species. Besides Serolis cornuta, two others inhabit
the shores of Kerguelen, viz., Serolis latifrons and Serolis septemcarinata ; the first of
these occurs also in comparatively deep water (210 fathoms) off the Crozets. Serolis
septemcarinata is common to all three gi'oups of Antarctic Islands, Prince Edward and
IMariou Islands, the Crozets, and Kerguelen, — while Serolis conmta was dredged off the
Crozets during the cruise of the " Gazelle."

Finally the shores of the southern and eastern parts of Australia are inhabited by
six species of Serolis, viz., Serolis tuherculata, Serolis australiensis, Serolis longicaudata,
Serolis elongata, Serolis minuta,, and Serolis pallida, ; these species, with the exception of
Serolis minuta, form, as has already been pointed out {ante, pp. 65, QQ), a well marked sub-
division of the genus, differing more from either the Kerguelen or the South American
species than any of these do from each other.

It is rather premature to draw any general conclusions from these facts, even with
regard to the distribution of the genus Serolis alone ; but it may at any rate be pointed
out that the distribution of this genus, as at present known, seems to necessitate the
division of the southern hemisphere into two distributional provinces, (1) an Antarctic,
reaching from South America to New Zealand, and (2) an Australian.

The genus Serolis seems to attain to its greatest development in point of number of
individuals on the shores of Patagonia and at Kerguelen; v. Willemoes Suhm states 'that at
Kerguelen Se7^olis forms no less than 20 per cent, of the Crustacean fauna ; and that " a
large species (Serolis cornuta) is the predominant and most characteristic form of all the
Crustacea in the shallow water of the Antarctic Islands."

It is true that nearly as many species are now kno-wn from Australia as from
Patagonia, but the genus does not appear to form an important element in the fauna of
the first mentioned locabty ; only a very few specimens, not more than two or three, of
any of the six species were obtained by the Challenger ; the fact that no species have
been previously described" either by resident naturalists or from collections made by
exploring vessels, also indicates the rarity of the genus in this region. In Patagonia, on
the other hand, there are not only a comparatively large number of species, but the
number of individuals is also great, though apparently not forming so important an
element in the fauna as at Kerguelen.

' Proc. Roy. Soc, voL xxiv. p. 590.

2 With the sole exception of one specimen of Scrolls tuherculata described by Grube, a>; quoted above.

(ZOOL. CHALL. EXP. — PART XXXIII. — 1884.) Kk 11

82 THE VOYAGE OF H.M.S. CHALLENGER.

The deep-sea species of Serolis have a wider range to the north than the shallow-water
species, though as yet none have been obtained north of the equator; since there are only
four deep-sea species known, and, with the exception of Serolis hromleyana, only a small
number of specimens of each were dredged, it is perhaps rather premature to draw any
deductions from the facts, and the following notes must be accepted for what they are
worth.

In the first place, it must be noted that in no case do any of the shallow-water species
pass the 300 fathom limit ; nor are any of the deep-sea species known to inhabit shallow
water ; the shallow-water are specifically distinct from the deep-sea forms. I may correct
here a misleading statement in Gerstaecker's account of the Isopoda in Bronn's
Thierreichs, from which it would appear that one species is common to " deep " and
"shallow" water; on p. 241 of the above quoted work, Gerstaecker gives a list of the
range in depth of the family Serolidse, and one species, which is my Serolis antarctica, is
stated to occur in 100 fathoms ofi" the Brazilian coast, and again in 1375 and 1600 fathoms
in the neighbourhood of the Crozets; 100 is a misprint for 400, which is the actual depth
at which the species was dredged.

Two out of the four deep-sea species have a comparatively wide horizontal as well as
vertical distribution ; one of these, Serolis antarctica, occurred at Station 120 (675
fathoms), ofi" Pernambuco, and again at Stations 146 and 147 (1375 and 1600 fathoms),
between Prince Edward Islands and the Crozets ; the other, Serolis hromleyana, was
obtained at four localities between Australia and New Zealand, Station 164b (410
fathoms), at Stations 168 and 169, ofi" the east coast of New Zealand, in 1100 and 700
fathoms, and again, considerably to the south, close to the Antarctic Ice-Barrier, at Station
156 (1975 fathoms) ; Gerstaecker, in the work abeady alluded to, calls attention to an
obvious corollary from these facts (which have been already mentioned in v. AVillemoes
Suhm's Preliminary Eeport on the Crustacea observed during the voyage of the
Challenger'), namely, that these two species, as well as others which have a wide
distribution, are found in deeper water passing southwards from the equator to the pole ;
the facts in the distribution of Serolis hromleyana appear to me to furnish grounds for
another deduction, that the size of the individuals increases as they pass southwards and
into deeper water; the specimens dredged at Stations 164b and 169 (410 and 700 fathoms
respectively) are all small; out of the nine or ten specimens dredged at Station 168,
farther south and deeper water (1100 fathoms), five are of considerable size, almost twice
as large as those obtained at the two first-mentioned stations; finally, at Station 156, the
southernmost point at which the species was dredged, two specimens half again as large
as the largest of those from Station 168 were dredged up from 1975 fathoms.

Serolis antarctica also agrees in these respects with Serolis hromleyana, though the
facts in this ease are not so strong, inasmuch as only a single specimen was dredged at

1 Proc. Roy. Soc. Land., vol. xxiv. p. 585, 1876.

REPORT ON THE ISOPODA. 83

the northernmost station in comparatively shallow water (Station 120, 675 fathoms); this
specimen is, however, distinctly smaller than any of those obtained in deeper water at
Stations 146 and 147.

It must be remembered, of course, that Gerstaecker's statements as well as mine
depend after all upon very few facts; it would be extremely rash at present to insist
upon any such generalisation as has been put forward by Gerstaecker in the work already
quoted, but it seems worth while to call the attention of naturalists to the facts such as
they are.

The remaining deep-sea species were dredged close to the east coast of South America ;
Serolis gracilis from a single Station (Station 120), off Pernambuco, in 675 fathoms, and
Serolis necera from two Stations close together and a little farther to the south, off Buenos
Ayres, at Stations 320 and 318, in 600 and 2040 fathoms respectively; 2040 fathoms is
the greatest depth which the genus is known to inhabit.

It appears therefore that the deep-sea forms of the genus, although not absolutely
confined to the neighbourhood of the great continents, attain to their greatest development
both in number of species and individuals in this situation, and are never found at any
distance from some land — continent or oceanic island.

The genus Sei'olis has evidently originated in the southern hemisphere, probably
round the shores of the south polar continent, and has thence spread northwards, its
range being apparently limited by temperature ; accordingly we find that in every case
those species which occur near the equator [Serolis gracilis, Serolis antarctica, Serolis
necBra) occur there in deep water where the conditions, as far as temperature is concerned,
are not so different from the conditions which must obtain on the shores of Kero;uelen and
Patagonia ; the one fact, however, which seems to militate against such an hypothesis is the
occurrence of Serolis carinata in shalloiv water as far north as San Diego in California. It
must be remembered, however, that the temperature of that portion of the Pacific is not so
high as might be expected from its latitude ; a cold current from the Antarctic area sweeps
along the western shores of South America, and the existence of this current has perhaps
rendered it possible for Serolis carinata (or its ancestors) to migrate farther to the north
than would be possible, for example, on the eastern shores of the same continent ; more-
over, a glance at the map of the world will show that here alone is there any du'ect
land communication between the area occupied by the shallow-water species of the genus
Serolis in the southern hemisphere and the more northern regions ; elsewhere tracts of
deep water have possibly aided in preventing their access to the equatorial regions and
the northern hemisphere, though it seems more probable, from what has abeady been
said, that the distribution of the group has been more restrained by conditions of tem-
perature than by any other cause.

Comparing the deep-sea with the shallow- water species of Serolis, it appears (1) that
the genus is pre-eminently a shallow-water genus, the number of deep-sea forms being

84 THE VOYAGE OF H.M.S. CHALLENGER.

comparatively small ; (2) that, as has already been pointed out, the deep-sea are in all
cases distinct from the shallow-water species ; (3) that the deep-sea species show certain
peculiarities, notably in the structure of the eyes, which, as has already been described
(p. 20 et seq.), are either entirely absent (Serolis antarctica) or, if present, show great
evidence of functional degeneration ; none of the deep-sea species possess well-developed
eyes. To compensate for the want of eyes, there is a great development of sensory hairs
on certain of the appendages; the males of Serolis necera, Sei'olis hivmleyana, and Semlifi
gracilis have, upon the third, fourth, and fifth joints of the third thoracic appendages,
tufts of sensory hairs, which have already been described (jjp. 55, 59, 62 ; PL IV. fig. 6 ;
PL V. fig. 8). This structural feature is not, however, peculiar to the deep-sea species,
inasmuch as it is also found in Serolis paradoxa ; but since it occurs in three out of the
four, and Serolis i^aradoxa is the only shallow-water species in which I have noticed it,
it may be considered as characteristic. Serolis antarctica does not agree with the other
deep-sea species in this respect, but the first pair of antennfe are furnished with a larger
number of sensory filaments than is usual — two upon each joint of the filament ; several
shallow-water species, however, Serolis pallida, e.g., and Serolis convexa, present the same
character.

In two of the deep-sea species, Serolis hromleyana and Serolis necera, the genus
attains to its greatest size, and these are indeed among the largest of the Isopoda ;
as a general rule the deep-sea representatives of the Isopoda are not distinguishable
from their shallow-water aUies by their greater size ; there are exceptions to this rule,
notably in the case of Bathynomus, a deep-sea genus recently described by Milne-Edwards,
which is no less than 9 inches long; and the Challenger collection contains a specimen of
another Isopod belonging to the same family Cymothoadse, which is also of considerable size.
The elongated and spine-like epimera of Serolis newra and Serolis hromleyana, and also,
though to a less extent, of Serolis gracilis, are unlike anything that is met with in the
representatives of the genus from shallow water, where the epimera are always moderately
developed in comparison. In all the deep-sea species, without exception, the ambulatory
limbs are furnished with comparatively few spines, which are generally soft and delicate,
The strong sword-like and serrated spines so commonly found in the shaUow-water
members of the group are either completely or partially absent ; a very general character,
inasmuch as it is found in two out of the four deep-sea species (Serolis necera and
Serolis gracilis), is the presence, upon the ambulatory limbs, of plumose hairs similar to
those which are found upon the abdominal appendages; in Serolis necera especially are
these plumose hairs developed in great abundance. They have been more particularly
described above, on pp. 55, 56.

The maxiUipedes in aU the deep-sea species possess a short tubercle on the inner
side of the middle joint of the palp, which may represent some kind of sense organ,
though the hairs wth which it is thickly covered are in no way different from the hairs

REPORT ON THE ISOPODA. 85

which cover the rest of the palp. This is represented in PL HI. tig. 10, a, which is drawn
from the maxillipede of the right side of Serolis gracilis ; with this figure may be com-
pared PI. I. fig. 11, which represents the same appendage in Serolis cornuta; this
character is so trifling, that if it were not regularly present in all the deep-sea species,
and as regularly absent from all the shallow-water species, with which I am acquainted,
it would indeed be hardly worth mentioning. In a young specimen of Serolis antarctica
from the brood cavity of the mother, in which the appendages are still in a. comparatively
undeveloped condition without any hairs and spines, this prominence is conspicuous upon
the palp of the maxillipede, and is armed upon its upper surface with a short pointed
spine ; it is possibly the rudiment of some structure highly developed in the ancestors
of the group.

NOTE.

In the Plates which accompany this Report the figures of the animal.-^ tiiemselves
were in most cases drawn by Mr. Mintern from the actual specimens ; the appendages
were lithographed by him from my drawings. When not otherwise stated the figures are
maofnified from 10 to 30 diameters.

(ZOOK. CHAU. EXP. — PART XXXIII. — 1884.) Kk 12

PLATE I.

(ZOOL. CHALL. EXP.— PAETXXXm.— 1884)— Kk.

PLATE T.

Serolis cornuta.

Fig. 1. Male, natural size.

Fig. 2. Female, natural size.

Fig. 3. Immature male, natural size.

Fig. 4. Terminal joints of one of the anterior antennae to show the sensory hairs (h).

Fig. 5. Terminal joints of one of the posterior antennae.

Fig. 6. Isolated joint towards the middle of the flagellum of posterior antenna.

Fig. 7. Distal half of left mandible, upper surface.

Fig. 8. Distal half of right mandible, upper surface.

Fig. 9. Maxilla of second pair of appendages.

Fig. 10. Same appendage of opposite side to show the occasional asymmetry of the
maxillae.

Fig. 11. Maxillipede.

Fig. 12. First abdominal appendage of right side.

Fig. 13. Second abdominal appendage of right side.

Fig. 14. Third abdominal appendage of right side.

Fig. 15. Several of tlie hairs fringing the endopodite and exopodite of abdominal
appendages.

Fig. 16. A portion of one of the same ; more highlv magnified.

The Voyage of H.M.S" Challenger':

Isopoda. . PI. I.

I-Iintertv Bros . Ut,tv.

A , St-uder .

PLATE IL

PLATE II.

Figs. 1-4. Serolis latifrons.

Fig. 1. Serolis latifrons (variety), female.

Figs. 2, 3. Terminal portions of mandibles, upper surface.

Fig. 4. Terminal joint of second antenna, with commencement of flagellum.

Figs. 5-13. Serolis schythei.
Fig. 5. A single joint from the filament of anterior antenna ; h, sensory hair.
Fig. 6. Three joints from filament of posterior antenna in male.

Fig. 7. Spines from the inner surface of penultimate joint of second thoracic appendage.
Fig. 8. Third thoracic appendage of male.
Fio-. 9. One of the ambulatory limbs.
Fig. 10. Maxillipede.
Fig. 1 1 . Upper lip.

Figs. 12, 13. Distal end of mandibles, upper surface.
Y\". 14. Serolis septemcarinata, nerve cords and ganglia.

The Voyage of H . M. S ^'Challenger"

Isopoda-.Pl.n

1-4.SER0LIS LAT IF RONS, White

SEROLIS ■ Sl.

^OLIS SCH

-vrA,Miers

Minterrv Bros litK.

itken. .

PLATE III.

(ZOOL. CHALL. KXP. — PART XXXIII. — 18S4.) — Kk.

PLATE III.

Fiffs. 1-6. Serolis Antarctica.

o"-

Fig. 1. Male, dorsal surface.

Fig. 2. Male, ventral surface.

Fig. 3. Female, dorsal surface.

Fig. 4. Sclerites and appendages in region of mouth, prepared by Ijoiling in caustic
potash.

Fig. 5. Third thoracic appendage of male.

Fig. 6. One of the ambulatory limbs.

Figs. 7-13. Serolis gracilis.

Fig. 7. Male, dorsal surface.

Fig. 8. Female, dorsal surface.

Fig. 9. Mandible, lower surface.

Fig. 10. Maxillipede, lower surface.

Fig. 11. One of last pair of thoracic appendages of male.

Fig. 12. One of the hairs which fringe inner surface of distnl joints of same appendage;
more highly magnified.

Fig. 13. One of the second pair of antennae.

TheVqya-ge of H .M.S'.'ChalleTiger"

Isopoda. PI. 111.

1-6. SEROLIS ANTARCTICA.Beddard. 7-I2.SEROLIS GRAClLlS.Beddard.

Mintern Bros.litK.

PLATE IV.

PLATE IV.

Serolis bromleyana.
Fig. ]. Male, dorsal surface.
Fig. 2. Male, ventral surface.
Fig. 3. Male (variety), with longer epimera.
Fig. 4. Female, dorsal surface.

Fig. 5. Distal joints of third thoracic appendage of male.

Fig. 6. Distal joints of another individual fringed with numerous sensory hairs (a).
Fig. 7. Second maxilla.
Fig. 8. Maxillipedc.

The Vojyage of H. M.S'.' Challenger "

Isopoda . PI . IV.

SEROLIS BROMLEYANA,Suhm.

Muxtern. Bros.lith

PLATE Y.

(zooi.. uuALL. EXP. — pakt xxxui. — 1884.) — Kk,

PLATE V.
Fio-s. 1-11. Seeolis ne^ra.

o

Fig. 1. Male, dorsal surface.

Fig. 2. Male, ventral surface.

Fig. 3. Female, dorsal surface.

Fig. 4. Female, ventral surface, displaying the eggs within the brood cavity.

Fig. 5. Tip of the filament of anterior antenna to show the presence of two kinds of
sensory hairs («, b).

Fig. 6. Maxillipede.

Fig. 7. Spines fringing inner surface of penultimate joint of second pair of thoracic
appendages.

Fig. 8. One of the third pair of thoracic appendages of male.

Fig. 9. Hairs from penultimate joint of same appendage.

Fig. 10. A portion of one of the ambulatory limbs.

Fig. 11. One of fourth pair of abdominal appendages.

Figs. 12-14. Serolis paradoxa.
Fig. 12. Distal half of one of the third thoracic appendages in female.
Fig. 13. Distal half of one of the third thoracic appendages in male.
Fig. 14. One of the first pair of maxillte.

The Voyage of H.M.S"C}iallenger"

Isopoda-PLV.

Mmtern. Bros . litK

— II.SEROLIS NE^ERAReddard.

12-14. SEROLIS PARADOXA.Fabncius.

PLATE VI.

PLATE VI.

Figs. 1-2. Serolis tuberculata.
Fig. 1. Female ; magnified three diameters.
Fig. 2. One of the second pair of antennae.

Figs. 3-8. Serolis australiensis.

Fig. 3. Male, ventral surface ; magnified three diameters.

Fig. 4. Female, dorsal surface ; magnified three diameters.

Fig. 5. Two joints from filament of first antenna ; h, sensory hairs.

Fig. 6. Maxillipede of left side.

Fig. 7. Male, dorsal surface.

Fig. 8. One of the thoracic limbs.

Figs. 9-15. Serolis convexa.

Fig. 9. Second thoracic appendage of male ; a, tuft of sensory hairs.

Fig. 10. Third thoracic appendage of male.

Fig. 11. Two hairs from inner side of penultimate joint of second thoracic appendage

Fig. 12. Terminal thoracic appendage of male.

Fig. 13. One of the hairs fringing the distal joints of the same appendage.

Fig. 14. One of the second pair of maxillse.

Fig. 15. Maxillipede of right side.

TKe Voyage of H.M.S. Challenger'!

Isopoda . PI .VI.

Minterrx. Bros . litK-

1-2.SER0LIS TUBERCULATA ,Ginite. 3-8.SER0LIS AUSTRALI ENSIS .Beddard.

9-15. SEROLIS CONVEXA, Cunnmghajn.

PLATE VII.

(ZOOL. CHALL. EXP. — PART XXXIII. -- 1884.) — Kk.

PLATE VII.

Fig. 1. Serolis pallida, iemah ; magnified two diameters.

Figs. 2-7. Serolis mindta.

Fig. 2. Male ; maguified about nine diameters.

Fig. 2'. Abdominal sterna and appendages ; b, peuial filament.

Fig. 3. Third thoracic appendage of male, right side.

Fig. 4. Antenna of first pair.

Fig. 5. Antenna of second pair.

Fig. 6. One of the ambulatory limbs.

Fig. 7. Spines fringing the penultimate joint of the second thoracic appendage.

Fis;s. 8-10. Sekolis longicaudata.
Fig. 8. Sterna of last thoracic and three first abdominal segments.
Fig. 9. Antenna of first pair.
Fig. 10. Antenna of second pair.

The Voyage of H.M S'.'CKallenger"

laopoda.Pl.VlI.

SEROLIS PALLIDA.Beddard. 2,-1 . SEROLIS MINUTA

8-10. SEROLIS LONGICAUDATA.Beddard.

Minterrv Bros .lith

PLATE VIII.

PLATE VIII.

Figs. 1, 2. SeROLIS long ICAUD ATA.

Fig. 1. Male, dorsal surface.

Fig. 2. Male, ventral surface ; both figures enlarged about four diameters.

Figs. 3-5. Serolis septemcarinata.
Fig. 3. Third thoracic appendage of male, left side.
Fig. 4. Base of penultimate joint of same appendage.
Fig. 5. One of the ambulatory limbs.

Figs. 6-16. Serolis pallida.

Fig. 6. Antenna of second pair.

Fig. 7. Antenna of first pair.

Fig. 8. Two joints of the filament with sensory hairs ; more highly magnified.

Fig. 9. Mandible of left side, lower surface.

Fig. 10. Terminal portion of left side ; more highly magnified.

Fig. 11. First maxilla of right side.

Fig. 12. Maxillipede of left side, lower surface.

Fig. 13. One of the second pair of thoracic appendages.

Figs. 14, 15. Spines from inner surface of penultimate joint of same appendage; more
highly magnified.

Fig. 16. Third thoracic appendage from right side of male.

The Voyage of H.M.Sl'CKallenger"

Isopoda.Pl.VIII.

MiTvtern. Bros - litK.

-2 SEROLIS LONGlCAUDATA,BeddaJ-d. 3-5.SER0LIS SEPTEM CARI NATA , Miers .

6-16. SEROLiS PALLIDA,Beddapd.

PLATE IX.

(ZOOL. CUALL. EXP. — PART XXXIII. — 1884.) — Kk.

PLATE IX.

Structure of the Eye.

Fig. 1. Surface view of membrane limiting the " ommateum " loelow ; a, perforations for
the nerve fibres arranged in groups of four, corresponding to each retinula ;
n, nuclei.

Fig. 2. Semidiagrammatic section through the eye of Serolis schythei ; c, corneal lenses ;
n, nuclei of Semper ; v, vitreous body ; r, retinula cells ; p, pigmentiferous
connective tissue corpuscles ; h, hyaline cells ; s, rhabdom.

Fig. 3. Single element of the eye of Serolis cormita, depigmented and isolated by
teasing ; r, rhalidom ; r', its posterior filiform prolongation ; h, hyaline cells ;
n, their nuclei.

Fig. 4. Single element of the eye of Serolis cormita to show the pigment sheath sur-
rounding the rhabdom (r).

Fig. 5. Single element of the eye of Serolis schythei ; r, rhabdom ; h, hyaline cell.

Fig. 6. One of the hyaline cells ; n, its nucleus.

Figs. 7, 8. Transverse section through the upper part of the retinula of Serolis schythei;
r, rhabdom ; j:>, pigment.

Figs. 9-15. A series of figures to show the varj-ing form of the rhabdom in Serolis
cormita.

Figs. 16, 17. Transverse section through the upper part of the retinula of Serolis
cormita; r, rhabdom ; j^, its pigment sheath.

Figs. 18, 19. Two isolated retinula cells oi Serolis schythei; r, rhabdomere.

Fig. 20. Series of transverse sections through retinula of Serolis schythei; a, nervous
rods below membrane ; h, lower end of retinula cell just above the pigmented
membrane ; c, retinula cells at the level of the nucleus (n) ; d, four retinula
cells surrounding the hyaline cell (h) ; e, upper extremity of the retinula cells ;
(y), rhabdomere.

The Voyage of H M.S."CliallengeT>':

Isopoda.Pl.lX.

15.

^

20 a..

f\ eM

20 b

20c.

20 d

20 e

■:n Bros . UtK.

STRUCTURE OF EYE

PLATE X.

PLATE X.

Structure of the Eye, &c.

Fig. 1. Single element of the eye of Serolis cornuta, depigmented by nitric acid ;
h, hyaline cell which has absorbed the jjigment.

Fig. 2. Dissection of Serolis cornuta; h, heart; hep, hepatic cseca ; od, oviduct;
r, rectum.

Fig. 3. Diagrammatic transverse section through the eye of Serolis necBra ; n, nuclei of
Semper; /, vitreous body; {g) mass of cells below the vitreous body traversed
by a network of pigment.

Fig. 4. Masticatory stomach of Serolis schytliei ; RP, LP, ribbed lateral plates ; I, la,
Ip, lateral ossicles; 1, 2, 3, pyloric ossicles.

Fig. 5. Diagrammatic transverse section of eye of Serolis hromleyana ; h, vitreous bodies ;
c, cornea ; m, tissue in which vitreous bodies are imbedded.

Fig. 6. Transverse section through the buccal region of Serolis septemcarinata ; a,
entrance to the buccal cavity ; m, mandible ; h, buccal cavity ; c, integument
bounding the lower surface of the body ; d, salivary glands ; e, connective
tissue cells ; n, nerve commissure.

Fig. 7. Embryo of Serolis anta7vtica ; c, telsou.

Fig. 8. Single element of the eye of Serolis necera isolated by teasing in glycerin ; v,
vitreous body with pigment sheath ; g, mass of cells below the vitreous body.

Fig. 9. Masticatory stomach of Serolis hromleyana; m, cardiac ossicle; V, ventral ossicle;
RP, LP, ribbed lateral plates; I, la, Ip, lateral ossicles; 1, 2, 3, pyloric
ossicles ; s, opening of pylorus ; t, triangular ossicle.

The Voyage of H. M.S. "Challenger'

I.sopoda, PI X

ANATOMY OF SEROLIS

Mmtam. Bros . lith .

>^..

>■' {*■■ v>

'/■^

!*• *■ *. i "^

.^■■*-

•oi'-J' *^i!

■■^/'^:^.:

':i , -'.-..■■A"

( "

t^,':.

■/■''i

■■■'■■'■■ ■'.■. . '■ :'^!iM&K\/i

'■•'.1 ■>:';<,'.

. .mi