LIBRARY

UNlVERiWY OF
CALtfORNIA
SANTA CRUZ

A TREATISE ON ZOOLOGY

A

TKEATISE ON ZOOLOGY

EDITED BY

E. BAY LANKESTEK

M.A., LL.D., F.R.S.

HONORARY FELLOW OF EXETER COLLEGE, OXFORD ; CORRESPONDENT OF THE INSTITUTE

OF FRANCE J DIRECTOR OF THE NATURAL HISTORY DEPARTMENTS

OF THE BRITISH MUSEUM

PART III

THE ECHINODERMA

BY

F. A. BATHER, M.A.

ASSISTANT IN THE GEOLOGICAL DEPARTMENT OF THE BRITISH MUSEUM
ASSISTED BY

J. W. GREGORY, D.Sc.

LATE ASSISTANT IN THE GEOLOGICAL DEPARTMENT OF THE BRITISH MUSEUM
PROFESSOR OF GEOLOGY IN THE UNIVERSITY OF MELBOURNE

AND

E. S. GOODRICH, M.A.

ALDRICHIAX DEMONSTRATOR OF ANATOMY IN THE UNIVERSITY OF OXFORD

Reprint A.ASHER & CO. Amsterdam 1964

A

TEEATISE ON ZOOLOGY

EDITED BY

RAYx LANKESTER

M.A., LL.D., F.R.S.

HONORARY FELLOW OF EXETER COLLEGE, OXFORD ; CORRESPONDENT OF THE INSTITUTE

OF FRANCE J DIRECTOR OF THE NATURAL HISTORY DEPARTMENTS

OF THE BRITISH MUSEUM

PART III

THE ECHINODEKMA

BY

F. A. BATHER, M.A.

ASSISTANT IN THE GEOLOGICAL DEPARTMENT OF THE BRITISH MUSEUM
ASSISTED BY

J. W. GREGORY. D.Sc.

LATE ASSISTANT IN THE GEOLOGICAL DEPARTMENT OF THE BRITISH MUSEUM
PROFESSOR OF GEOLOGY IN THE UNIVERSITY OF MELBOURNE

AND

E. S. GOODRICH, M.A.

ALDRICHIAN DEMONSTRATOR OF ANATOMY IN THE UNIVERSITY OF OXFORD

LONDON

ADAM & CHARLES BLACK
1900

Exclusive Agents for U.S.A.
STECHERT-HAFNER SERVICE AGENCY, INC.

31 East 10th Street
New York, New York 10003

Sole agents for India:

Today & Tomorrow1 s Book Agency,

22-B/5, Original Road, Karol Bagh,

New Delhi-5

PEEFACE

THE present volume is the " Third Part " in order of a com-
prehensive treatise on Zoology, which has been for some time
in preparation under my editorship. In this treatise each
of the larger groups of the Animal Kingdom is to be described
by a separate author; whilst, as far as possible, uniformity
in method and scope of treatment is aimed at. The authors
are, for the most part, graduates "of the University of Oxford,
though it may not be possible to maintain this limitation in
future sections of the work.

The general aim of the treatise is to give a systematic
exposition of the characters of the classes and orders of the
Animal Kingdom, with a citation in due place of the families
and chief genera included in the groups discussed. The work
is addressed to the serious student of Zoology. To a large
extent the illustrations are original. A main purpose of the
Editor has been that the work shall be an independent and
trustworthy presentation, by means of the systematic survey,
or taxonomic method, of the main facts and conclusions of
Zoology, or, to speak more precisely, of Animal Morphography.

The treatise will be completed in ten parts of the size of
the present one. It will at once be apparent that this
limitation necessitates brevity in treatment which, however,
will not, it is believed, be found inconsistent with the fulfil-
ment of the scope proposed or with the utility of the work

vi PREFACE

to students. The immediate publication of the following
parts may be expected : —

Part I. Introduction and the Protozoa.

Part II. General Discussion of the Metazoa — The Pori-
fera — The Hydromedusae — The Scypho-
medus^e — The Anthozoa — The Ctenophora.

Part III. The Echinoderma (the present volume).

Part IV. The Mesozoa — The Platyhelmia— The Nemer-
tini.

These parts will be issued, without reference to logical
sequence, as soon as they are ready for the press. In accord-
ance with this procedure, which to some extent evades the
injustice of making an author, whose work is finished, wait
for publication until other more tardy writers have completed
their tasks, the present volume, which is Part III., is the first
to make its appearance.

The following authors have undertaken portions of the
work : — Professor Poulton, F.R.S., M.A. Oxon. ; Professor
Weldon, F.R.S., M.A. Oxon. ; Professor Benham, D.Sc., M.A.
Oxon.; Mr. G. C. Bourne, M.A. Oxon. ; Mr. G. H. Fowler,
M.A. Oxon. ; Professor Minchin, M.A. Oxon. ; Mr. F. A. Bather,
M.A. Oxon. ; Professor J. W. Gregory, D.Sc. ; and Mr. K S.
Goodrich, M.A. Oxon.

E. RAY LANKESTER.

February 1900.

CONTENTS

CHAPTER VIII

GENERAL DESCRIPTION OF THE ECHINODERMA . V- 1

CHAPTER IX

THE PELMATOZOA — CYSTIDEA. . >/. . . 33

CHAPTER X

THE PELMATOZOA — BLASTOIDEA fci&>H .A. ,* ' . 78

CHAPTER XI

THE PELMATOZOA — CRINOIDEA . '. "4,.J . 94

>• , i

CHAPTER XII

THE PELMATOZOA — EDRIOASTEROIDEA . . .205

CHAPTER XIII

THE ELEUTHEROZOA — HOLOTHURIOIDEA . . . 217

CHAPTER XIV

THE ELEDTHEROZOA — STELLEROIDEA . . . . 237

CHAPTER XV

THE ELEUTHEROZOA — ECHINOIDEA . . -« ••» ; . 282

INDEX 333

TABULAE STATEMENT SHOWING THE SYSTEMATIC
POSITION OF THE ECHINODERMA.

ANIMALIA.

GRADE I. PROTOZOA.
„ II. METAZOA.

BRANCH A. BRANCH B.

PARAZOA. ENTEROZOA.

GRADE I. (of the Enterozoa).
ENTEROCOELA = COELENTERA.

GRADE II. (of th« Enterozoa).
COELOMOCOELA = COELOMATA.

PHYLA (of the Coelomocoela).

ECHINODERMA, MOLLUSCA, APPENDICULATA,
PLATYHELMIA, VERTEBKATA, etc.

CHAPTER VIII.

THE ECHINODERMA.1

PHYLUM ECHINODERMA.

GRADE A. PELMATOZOA.

CLASS I. CYSTIDEA.

„ II. BLASTOIDEA.

„ III. CUINOIDEA.

„ IV. EDRIOASTEROIDEA.

GRADE B. ELEUTHEROZOA.
CLASS I. HOLOTHURIOIDEA.

11

II. STELLEROIDEA.

III. ECHINOIDEA.

General Features. — This is one of the best characterised and
most distinct Phyla of the Animal Kingdom. Nearly all the living
animals included in it, such as the sea-urchin (Echinoid), starfish
(Asteroid), brittle-star (Ophiuroid), sea-cucumber (Holothurian),
sea-lily (stalked Crinoid), or feather-star (free Crinoid), can readily
be distinguished through their possession of a radial symmetry, in
which the number five is dominant, of a sub-epidermic skeleton com-
posed of calcium carbonate, with a characteristic micro-structure
resembling trellis-work, and of a system of sacs, canals, and tubes
that carry water through the body, especially by means of five
radial canals from which small branches called podia are given off to
the exterior. The extinct forms known as Blastoidea and Edrio-
asteroidea appear to have had a similar organisation ; and the same
statement maybe made of most of the Cystidea, another extinct class.

It is true that there are recent forms in which the quinqueradial
symmetry or pentamerism is obscure ; but, on the whole, it is so
marked a feature that the early zoologists, and notably Cuvier,
placed the Echinoderma together with the Coelentera in a sub-

1 By F. A. Bather, M.A.
I

2 ECHINODERMA— GENERAL DESCRIPTION

kingdom Radiata. The presence of a gut distinct from the body-
cavity (coelom) is alone enough to mark the superiority of Echino-
derm organisation, as was first insisted on by Leuckart. The
resemblance of certain Holothurians to the Gephyrea is but super-
ficial and secondary ; the above-mentioned characters form sufficient
distinction.

Examples of the Classes. — Within the Echinoderma is great
diversity of organisation. Between the worm-like, semi-gelatinous
Holothurian, Synapta, living in the mud of the shore, and the
stalked Pentacrinus of the depths of the sea, or the brittle-star of
the rock -pools, there might well seem an impassable barrier.
Taking typical examples of the various classes, let us note the
more obvious differences. In an ordinary Holothurian (e.g. Holo-
thuria, Cucumaria, Fig. IV. 4, p. 231) the body is cucumber-shaped,
with a mouth at one end and an anus at the other ; round the
mouth is a ring-canal of the water-vascular system, and from it
are given off five radial canals, running below the surface of the
flexible integument and sending podia to the exterior ; two of the
avenues (ambulacra) of podia run along that surface of the body
which is away from the ground and may be called " dorsal " ; the
" ventral " surface, containing the other three ambulacra, is often
flattened to form a kimj of walking sole. A Holothurian has no
arms or projecting rays, but its mouth is surrounded by a circlet
of tentacles, often branched, retractile at will, and serving to collect
food. A Regular Sea-urchin (e.g. Echinus, Cidaris, Figs. VII., XVII.
pp. 290, 303) resembles a Holothurian in being without projecting
rays; but it is more spherical in shape, with a rigid test, and
moves with its mouth towards the sea-floor, and with its anus at
the opposite pole of the body. In a Heart-urchin (e.g. Spatangus,
Fig. XLV. p. 324), which moves through and swallows mud and
sand, the body has become obliquely elongate, i.e. with the long axis
at an angle of 36° to the position it occupies in a Holothurian ; the
mouth has moved a little forward, and the anus has moved down
from the top of the body to its lower surface, so that both mouth
and anus lie on the under surface, at either end of the long axis.
In Echinoids, the radial water-vessels are beneath the test ("hypo-
thecal ") and stretch from the oral to the anal pole. In a Starfish
the mouth is in the centre of the under surface, while the anus is
almost in the centre of the upper surface, but is absent in a few
forms ; the body, encased in a yielding theca, is either markedly
pentagonal in outline or star-shaped ; in the latter case a central
"disc" may be distinguished from the "arms." The number of
arms varies from five (e.g.Asterias rubens, Figs. L, IV. pp. 240, 242) to
over forty (e.g. Heliaster). The radial water-vessels, one to each arm,
lie in a groove on the oral surface (" epithecal ") and are fringed
by podia, which do not pass on to the aboral surface at all. An

ECHINODERMA— GENERAL DESCRIPTION 3

Ophiuroid (Fig. XIII. p. 261) resembles a starfish in which there is a
sharp distinction between arms and disc ; the mouth is on the
under surface, but there is no anus. Whereas the arms of a star-
fish are merely extensions of the body, containing the generative
glands and processes from the stomach, those of an Ophiuroid con-
tain only blood-vessels, water -vessels, and nerves, and,, being
themselves used as locomotor organs, have a stout internal skeleton
of separate ossicles, worked on one another by well -developed
muscles, but have less developed podia ; they are nearly always
live, and unbranched except in Astrophytidae (Fig. XXXII. p. 277).
As is explained on p. 238, however, no sharp line can be drawn
between Asteroid and Ophiuroid structure. A Crinoid (Fig. III.
p. 98) differs markedly from all the forms just mentioned, in that
the mouth faces upwards, or away from the sea-floor ; the anus
is also on the upper surface. This position is connected with the
fixed habit of the Crinoids, which are attached temporarily or
permanently to the sea -floor by their aboral surface, usually
through a jointed stem. This fixed state of existence is corre-
lated with the development of a jointed process ("arm" or
brachium) from each radius of the rigid theca. The arms are
often forked many times ; they contain extensions of the nervous,
blood-vascular, water-vascular, and generative systems, and have a
ventral groove lined with cilia which sweep currents of water to the
mouth. The Blastoids (Fig. IV. p. 82) may be roughly described
as Crinoids without brachia, but with food-grooves on the oral
surface of the theca, fringed with jointed skeletal processes
(brachiola). The Cystidea, like the Crinoids, are fixed, with mouth
and anus on the upper surface ; the relations of their food-grooves
and water-canals vary greatly ; in some of the older ones (Fig. II.
p. 44) radial symmetry does not seem to have affected even these
organs, still less, therefore, any other organs of the body. The
Edrioasteroidea (Fig. VI. p. 209) are sessile, with upwardly directed
mouth and anus, with five food-grooves radiating from the mouth,
sometimes on to the aboral surface, as in Echinoids, and apparently
with hypothecal water-canals, also as in Echinoids, at any rate
with some portions of the water-vessels penetrating the test along
the ambulacra.

Phylogeny and Ontogeny. — The combined evidence of com-
parative anatomy, embryology, and palaeontology indicates that
the Echinoderma owe most of their obvious characters, such as
radiate symmetry, the ambulacra, and the coil of the gut, to their
having passed through a " pelmatozoic " stage, i.e. a stage in which
the animal was attached by a part of its body wall, in which the
mouth, and to a less extent the other apertures, faced upwards,
while there was a tendency to the radiate (pentamerous) exten-
sion of food-grooves with accompanying organs (see Chapter IX.,

ECHINODERMA— GENERAL DESCRIPTION

" Pelmatozoa "). Setting aside these characters, the origin of which
may be traced in individual development, and selecting those
common to the early stages of all Echinoderms, zoologists have im-

agined a phylogenetic
stage, the two-sided or
Dipleurula stage (Fig.
I.), more or less re-
peated in the Dipleurula
larvae of recent Echino-
derms (Fig. II.). The

M

Tl

0— l.hc:

animal was marine. Its

The

Fio. I.

Diagrammatic reconstruction of the imagined Dipleurula, long axis Was

ancestor. Anterior end on left of drawing ; organs of left nnsrprinr anH

side towards observer, and with stronger outline than those POk

of right side. For description and lettering, see adjoining to the SCa-floor.

mouth (0) was anterior

and ventral ; the anus (As) posterior or postero- ventral. The
two were joined by an uncoiled gut, perhaps with a stomachal
enlargement in the middle. On either side of this lay the coelom,
formed by constriction from the larval stomach or archenteron, in
other words, an " enterocoel " ; it was divided into a right and left
anterior portion (a.c), and a right and left posterior portion (r.p.c
and Lp.c). Each anterior vesicle was connected with the exterior
by a canal, opening at a dorsal pore (M) on each side the median
line, sometimes, perhaps, fusing into one. These canals were in-
directly connected (s.c) with posterior offshoots from the anterior
coelom, the right and left hydrocoels (r.hc and Lhc). Gonads
developed from the coelomic endothelium. The ectodermal
epithelium was probably ciliated, and a portion of it in the " pre-
oral lobe" (p. I) was differentiated as a sense organ, with longer
cilia and underlying nerve-centre (n), from which two gangliated

FIG. II.

Dipleurula larvae seen from right side (partly after Lang). 1, Pluteus of Echinoid ;
2, Auricularia of Holothurian ; 3, Bipinnaria of Asteroid ; 4, Tornaria of Enteropneustan ;
0, mouth ; As, anus ; s, spicule ; np, neural plate with cilia. The strong black lines repre-
sent ciliated bands ; the shaded areas show the course of the gut.

nerves ran back below the ventral surface. In the mesoblastic
connective tissue, derived by the migration of cells, there was a
tendency to the secretion of crystalline calcium carbonate. Except

ECHINODERMA— GENERAL DESCRIPTION

5

for the latter character, the Dipleurula agrees in essentials with
the larva of Enteropneusta, which was described by Joh. Miiller
(1850) as an Echinoderm larva, under the name Tornaria
(Fig. II. 4).

The simplest larval form among recent Echinoderms, that of
the Holothurians, known as Auricularia (Fig. II. 2, and Fig. III.),
differs from the Dipleurula in being bent upon its ventral surface,
so that the mouth lies in the middle of the concavity so formed,
while in front of it is a " preoral lobe," and behind it is a similar
prominence, in the middle of which, on the ventral surface, is the
anus ; the cilia are restricted to a band immediately surrounding
the mouth, and a band that
passes in front of the mouth,
then round the edge of the
ventral concavity, and across
in front of the anus. In
subsequent development the
ciliated ring becomes very
sinuous, and when the Auri-
cularia assumes a barrel
shape, before changing into
the Holothurian, the ring
atrophies in sixteen places,
and the separate pieces unite
in such a way as to form
five rings like hoops round
the barrel. In this stage the
mouth has again passed up
to the anterior pole, and the
anus down to the posterior.
This form is called the Pupa
(Fig. I. 8, p. 219).

The only free larval form
that is known among Pel-
matozoa is that of the highly
specialised free-moving Crinoid, Antedon. It resembles the Holo-
thurian Pupa in general shape (Fig. IV.), and in the possession of
five ciliated bands (cc), probably derived in a similar manner ;
but since the early stages have been pressed out of the develop-
ment, this cannot be considered proven. Here, moreover, there
remains a ventral concavity, through which the definitive mouth
breaks (0) ; there is no anus at this stage. Anteriorly is a tuft
of long cilia.

Most Asteroidea have a larva known as Bipinnaria (Fig. II.
3, and Fig. VI. 10), which passes through an Auricularia stage.
By a meeting of the sinuosities of the ciliated ring anteriorly,

Fi«i. III.

AHrh'ttlaria of Synttpta, the ventral surface ami
left side facing the observer (after Semun). 0,
mouth ; oc, oesophagus ; ft, stomach ; r, rectum ;
hpt hydropore ; A<:, hydrocoel ; r and fyw, right
and left posterior coelom ; it, nervous band ; s?>,
spicule in form of wheel ; c, ciliated bands, viz.
— 1, preoral ; 2, adoral ; 3, circumoral ; and 4,
anal. x25.

ECHINODERMA— GENERAL DESCRIPTION

M

FIG. IV.

Larva of Antedon
(after Bury). Anterior
end, with preoral
lobe (PL), uppermost.
X 95.

there is formed a preoral ring which separates from the rest.

The ventral depression runs up on either side the preoral area,

and eventually surrounds it and its ciliated ring. In the region
of the ciliated rings the body stretches out pro-
cesses, which are symmetrically paired, except the
frontal process, which bears the preoral ring. In
some species this process splits into three branches
and the cilia disappear ; such a form is called
Brachiolaria.

A still further development is the Pluteus
larva (Fig. II. 1, and Fig. V.) of Ophiuroids
and Echinoids, characterised by the decrease of
the preoral area and the increase of the anal area ;
paired processes extend forward, and an unpaired
process stretches backward from the posterior end
of the anal area ; these processes are usually very
long and supported by spicules, but movable.
Two, arising from the posterior and lateral region

of the ciliated ring, are pronounced in Ophiuroids but absent in

Echinoids.

The development of the larval form

from the ovum is effected in much the

same way in all known Echinoderma

(Fig. VI.). The segmentation of the

ovum is total and quite or almost equal

(Fig. VI. 2). A coeloblastula is formed

with a segmentation cavity, and with a

wall of a single cell -layer, thicker in

one region (Fig. VI. 3). This region

is invaginated, forming the archenteron

of a gastrula (Fig. VI. 4). At about

the same time the thickened region,

now the end wall of the archenteron,

proliferates endoderm cells, some of

which wander into the segmentation

cavity, where they may be joined by a

lesser number from other parts of the

gastrula wall (ectoderm), and so form mesenchyme, from which

mesoderm tissue is ultimately developed (Fig. VI. 5). The

archenteron occupies but a small part of the segmentation cavity,

its lumen is usually narrow, and the external opening forms a

small blastopore, which in Antedon soon closes. Both endoderm

and ectoderm are usually ciliated from the beginning ; but in

Antedon cilia appear only on the ectoderm after gastrulation.

The larva becomes bilaterally symmetrical by dorso-ventral

compression and the formation of a ventral concavity. The inner,

-o

Fio. V.

Pluteiis of a Heart-urchin, from
ventral side (after Lang). 0, mouth ;
As, anus.

ECHINODERMA— GENERAL DESCRIPTION

i.e. anterior, end of the archenteron becomes constricted off from
the rest of the archenteron to form the " coelom " (Fig. VI. 6).
The coelom sends backward a process on each side in the dorsal
region ; the hinder parts of these become constricted off as a
" right and left posterior coelom," which almost meet posteriorly ;
the remainder forms the " anterior coelom " (Fig. VI. 7). At the

1 4 7

--v

11

Fio. VI.

Early stages of Echinoderm ontogeny. 1, 2, 3, the Echinoid Echinocyamm (after Theel).
4, 5, 6, the Crinoid Antedon (after Seeliger). 7, 8, 11, the Asteroid Asterina (after
MacBride). 9 (after Bury). 10 (after G. W. Field). 1, ovum in mucilaginous coat ; vitelline
membrane (y) beginning to separate ; s, spermatozoa, one of which is entering the yolk (x!30).
2, segmenting ovum seen from above, 2 hrs. 20 min. after fertilisation. 3, blastula in
longitudinal section, 13 hrs. after fertilisation, c, cilia; sc, segmentation cavity (x200). 4,
gastrula, in section, 16 hrs.; oe, archenteron ; bp, blastopore (x73). 5, the same, 26 hrs., with
mesenchyme (7n,s) developing from endoderm. Letters as before (x88). 6, 48 hrs., with
blastopore closed, and archenteron constricted into oc anterior, mesenteron, and ec posterior
enterocoel (x88). 7, longitudinal section showing extension of right and left posterior
coeloms (rpc, Ipc) from anterior coelom around larval stomach (st), ( x 60). 8, further stage,
showing rpc and Ipc separated from ac, and lobes of the rudimentary left hydrocoel (he), ( x 60).
9, dorsal view of a Bipinnaria,. oe, oesophagus ; hp, hydropore ; y, " blood-vascular space,"
Bury, perhaps rudiment of right hydrocoel. Other letters as before (x50). 10, Bipinnaria
of Asterias, four days old. cc, ciliated band; m, mesenchymatous muscle fibres (x73). 11,
diagram showing relations of stone canal (stc) and pore canal (pc) to left hydrocoel (Ihc) and
anterior coelom (ac).

hinder end of the anterior coelom, on both the right and left side,
there is a small outgrowth, the " right and left hydrocoel " (Fig.
VI. 8) ; that on the left is, as it happens, much more developed,
but the presence of a right hydrocoel has been proved by Metsch-
nikoff (1869) and MacBride (1896), (y in Fig. VI. 9). Near
the median dorsal line, above the hinder end of the anterior
coelom, a perforation arises in a thickening of the ectoderm,

8 ECHINODERMA— GENERAL DESCRIPTION

forming the "hydropore"and the "pore canal"; while, in no connec-
tion with this, a groove arises along the hinder wall of the anterior
coelom, and develops into a canal connecting the left hydrocoel
with the anterior coelom, and called "stone canal," because its homo-
logue in recent adult Echinoderms develops spicules in its walls
(Fig. VI. 11). MacBride (1896) has observed larvae of Asterina
'gibbosa, in which there were a hydropore and stone canal on the
right, and some in which both right and left pores were present.
The latter arrangement occurs also temporarily in the Bipinnaria
of Asterias (Field, 1892, Fig. VI. 10), and is that which we suppose
to have obtained in the Dipleurula.

On the ventral side, at the anterior end of the body, a mouth
is produced by invagination, and leads into the remaining part of
the archenteron, which becomes modified into a larval stomach and
a short rectum curved ventralwards and opening at the blastopore.
The part of the larva in front of the mouth is called " the preoral
lobe," and a portion of it becomes a sense organ, usually ciliated,
with a development of nerve tissue (Fig. VI. 9, 10; Fig. I.).

It must not be supposed that a Dipleurula larva of this simple
type actually exists. In each class it presents some modification,
the outward appearances of which have already been described.
Moreover, the internal structures vary in the order of their
development and in persistence. Enough is common to the
various types to show that the Dipleurula larva is no phantasm, and
to suggest very strongly that it represents an ancestral Dipleurula
stage, differing but slightly if at all from the ancestral Tornaria,
and being one of the lowest of all animals with a coelom. The
hydrocoels and their indirect exterior openings have been compared,
perhaps not very judiciously, with the excretory nephridia of
higher Coelomata. The possible connection of Tornaria with the
ancestral Chordata gives additional interest to the resemblance
between stereom formation and bone formation (see p. 29), and
to the invagination of a primitively superficial nervous system in
the two groups.

Between adult Echinoderms and other groups of the Animal
Kingdom no comparisons are possible. From this stage onward
the Echinoderm follows a path of its own. By a remarkable
metamorphosis, varying in its details but presenting some common
features in the different classes, the almost bilaterally symmetric
larva is transformed into the almost radially symmetric adult.
This metamorphosis undoubtedly represents the changes that took
place in the early history of the classes; and the extraordinary
difficulties of interpretation are due to the enormous compression
of that history, the elimination in some cases of unnecessary
stages, and the unequal acceleration of others. The clue is offered
by the older fossils, which, as explained under Cystidea, forcibly

ECHINODERMA— GENERAL DESCRIPTION

o-

s.c.

l.hc

M

suggest that all Echinoderma are descended from sessile ancestors
(necessarily representing a stage subsequent to the Dipleurula), and
that the oldest among these had not acquired radial symmetry, that
being, it would appear, a consequence of fixation. Fixation was
retained more or less completely by Cystidea, Blastoidea, Crinoidea,
and Edrioasteroidea ; but among the other classes it is only the
Stelleroidea that now preserve traces of it in their ontogeny.

The passage from the Dipleurula to the fixed stage is best
studied in Antedon (Bury, 1888; and Seeliger, 1893); but even
here changes that, in phylogeny, must have succeeded fixation
now precede it, and actually precede the free-swimming stage of the
larva. Fixation takes place by a modified portion of the preoral
lobe (p.l), as also in Stelleroidea.
The phylogenetic result of this was
the passage of the mouth (0) to
the posterior end of the Dipleurula,
which was now directed upwards
(Fig. VII.). With the mouth went
the hydrocoel. The attachment ap-
pears to have been towards the right
side, for thus only can we account for
the fact that the structures on the
left of the Dipleurula increased at the
expense of those on the right. It
was therefore the left hydrocoel (l.hc)
and stone canal (s.c) that moved
upwards with the mouth, while those
on the right disappeared. The
nervous structures of the anterior
end remained there or, possibly,
atrophied. The forward portion of the anterior coelom (a.c) shared
in the construction and elongation of this region ; but its hinder
portion was dragged up along with the hydropore (M ) and formed
the " parietal canal " (par), so called because it lies along the outer
wall of the larva. The left posterior coelom (l.p.c) of the Dipleurula
was caught in between the oesophagus and the stomach, and so
passed upwards, towards what we may now call the oral pole of
the fixed stage ; while the right posterior coelom (r.p.c) was pushed
downwards by the stomach pressing to the right and thus came
to lie nearer the aboral pole. The blastopore is early closed in
the ontogeny of Antedon, but we infer from the position of the
larval rectum that in phylogeny the anus (As) did not move
upwards so rapidly as the mouth. The effect of these changes
was a torsion of all the structures in the upper part of the
body. The gut was thrown not into a simple loop, but into a
dextral coil. The pressure of the oesophagus against the hydrocoel

FIG. VII.

Diagrammatic reconstruction of the
imagined primitive Pelmatozoic ances-
tor. Compare carefully with Fig. I.
ami with adjoining text'.

10

ECH1NODERMA- GENERAL DESCRIPTION

not merely pulled it up, but pressed it into a horseshoe curve,
with the opening directed to the anal side. The left posterior
coelom was curved in like manner. The further elongation of the
fixed aboral end involved lobes of the right posterior coelom
and initiated their downward extension, first on the right side of
the anterior coelom, then gradually curving round it. This torsion
and shifting of internal organs may be compared with the simpler
case of streptoneurous Gastropoda.

Careful study of the two diagrams, representing the Dipleurula
(Fig. I.), and the primitive Pelmatozoan or fixed stage (Fig. VII),

ax

Fio. VIII.

Sections of Antedon larva, semi-diagrammatic (1-4 after Seeliger, x85). (5, 6 after Bury).
1, longitudinal section with preoral lobe turned downwards, the reverse of Fig. IV. ; the anterior
coelom extends into this, as well as upwards into the parietal canal. 2, transverse section in
the neighbourhood of the hydropore, showing the parietal canal leading to it ; two lobes of the
hydrocoel and the lower end of the stomach are seen to be at the same level as right posterior
coelom and lower end of vestibule. 3, part of a longitudinal section in the same direction as
Fig. 1, but passing through the hydropore and the five primitive lobes of the hydrocoel not
yet connected therewith ; between the two is a small extension of left posterior coelom ; adjoining
sections show that the hydrocoel is still horseshoe-shaped, but will eventually close along the
gap between lobes 1 and 5 to form the hydrocircns. 4, transverse section at a slightly more
orad level than in 2, and in an older larva; the vestibule is here closed over; note the
mesenteries between right and left posterior coeloma. 5, transverse section of an older larva,
in which the vestibule has passed right up to the oral pole, and a rectum (r) has formed ;
between this and the stomach is seen the rudiment of the axial organ ; the parietal canal
remains, but the stone canal now opens into it (x!25). 6, median longitudinal section, the
vestibule still open, columnals forming around extensions of right posterior and anterior
coeloms ; left posterior coelom is seen above the stomach as a lobe of the hydrocoel (x!20).

Explanation of letters — ac, anterior coelom ; ax, axial organ ; cc, ciliated bands ; col,
columnals ; he, hydrocoel ; l.pc, left posterior coelom ; nn, nerves ; p, hydropore ; par, parietal
canal ; r, rectum ; r.pc, right posterior coelom ; st, stomach ; st.c, stone canal ; v, vestibule.

will enable the student to appreciate tne peculiar position of the
internal structures in the Antedon larva, of which a few sections
are here given for comparison (Fig. VIII.). The structure is far
more complicated than in Fig. VII., owing to extensions from
the coelomic cavities. In the earlier sections the hydrocoel is still

ECHINODERMA— GENERAL DESCRIPTION u

on the aboral side of the right posterior coelom ; indeed, the mouth
itself is not at the future oral pole, for the larval mouth closed
early, and the place where it was became arched over by lips
of ectoderm, which formed a " vestibule " (v). This vestibule it
is that gradually moves up; a fresh mouth ultimately breaks
through into it, and the lips again unfold at the new oral pole.
The connection of the anterior coelom, through the parietal canal
and the hydropore, with the exterior, persists (cf. Fig. VIII. 1 and
3) ; the hydrocoel opens into the parietal canal by the stone
canal at a later period (Fig. VIII. 5).

There is reason to believe that some of the early Cystidea
(Amphoridea, p. 43) had an internal structure scarcely more
advanced than Fig. VII. But the fixed stage had further effects.
The most notable was the prolongation of ciliated and tentacu-
liferous grooves from the mouth, accompanied by processes . from
the hydrocoel. At first there were three such radial extensions :
anterior, right, and left, since the
presence of anus and hydropore, and
the absence of hydrocoel on the posterior
side prevented extensions in that direc-
tion (Fig. IX.). The five rays, so char- ~*
acteristic of Echinoderma, were produced
by the forking of the right and left
rays. It was only at a later date, when
the hydrocoel had grown into a ring
round the oesophagus, that the five rays
could proceed equally from this ring. Fic IX

The division of the rays into a pair The pentame';.ism ' of Echino.

enclosing anUS and madreporite, arid tlerma contrasted with a regular
, 17.. ill pentamerism. 0, mouth ; As, anus ;

known as the OlWUm, and the three between them is the madreporite.

others (anterior, right ant. and left Jt^&r pifS ^tSSto,"
ant.), known as the trivium, is opposed ^h4ichh* JJ^JJ1*! ftajIjTaS

tO this fundamental Structure. It must the 'five rays of a' Pelmatozoan.

further be noted that this bilateral

symmetry of the rays has nothing to do with the bilateral nature

of the Dipleurula.

While these changes were in progress the formation of stereom
continued. At first there were only spicules deposited in the
mesenchyme (see Fig. II. 1, and Fig. III). These enlarged and
fused into plates, which eventually became so large as to abut
on one another. These plates were arranged in the mesoderm
beneath the ectoderm. An account of their arrangement and
structure will be found under Amphoridea (p. 45). Through
the fixation below and the radiation of the hydrocoel and food-
grooves above, these plates gradually came to lie in definite
positions and to assume a definite number, shape, and size. The

12

ECHINODERMA— GENERAL DESCRIPTION

r.pc

Fio. X.

ontogeny of Antedon suggests their division into two groups
(Fig. X.) : one formed around the upper, oral coelom (l.pc,
i.e. the left posterior coelom of the Di-
pleurula), which gradually encircled the
oesophagus; the other around the lower,
aboral, or apical coelom (r.pc, i.e. the right
posterior coelom of the Dipleurula). The
P former set were affected by radiate sym-
metry before the others, and in Antedon
larya are represented by five large plates,
the " orals " (0). The latter set form the
plates of the aboral side of the adult
Echinoderm. In a Pelmatozoan they form
the dorsal cup (B and IB) and the ossicles
of the stem (col) when that organ is pre-
sent.

We have now traced the history of
the Echinoderma up to a form fixed
aborally, and with rays, normally five in
™mber> Proceeding from the mouth and

hydropore ; fp, fixing plate of underlying the hydrocoel ring or hydro-
stem, " dorso • central "; for . J _» J . / .
other letters, see adjoining text. CITCUS. Ihese ray 8 involved Other 01 the

internal organs, notably portions of the

oral and aboral coelom, and accompanying them was a develop-
ment of epithelial nerves and a circumoral nerve ring. The
dividing wall between the right and left posterior coeloms, the
dorsal mesentery of the Dipleurula, now lies horizontally or
transverse to the long axis. A new vertical mesentery, both
above and below, is formed by the tissue separating the in-
curved ends of the oral and apical coeloms respectively. On
the inner walls of these coeloms, adjoining this mesentery, is a
thickening of the endothelium (ax in Fig. VIII. 5), to form event-
ually a strand passing up to the main axis through the coil of the
gut, and known as the " axial cord." This, in the adult, originates
the gonads, which seem at first to have been expelled through an
aperture in the body wall between mouth and anus, as seen in
Holothurians and some Cystidea. Subsequently this becomes
involved in the radiate symmetry.

The phylogenetic stage thus reconstructed on the evidence of
embryology and palaeontology corresponds on the whole to the
stage imagined by Semon (1888), and named by him Pentactcea
(five-rayed). The question arises : How far does this represent the
ancestor of all Echinoderms ? There can be no doubt that this
actually was a stage in the history of the fixed Echinoderms
(Pelmatozoa) ; that it was also a stage in that of the free Echino-
derms (Eleutherozoa), is coming more and more to be the opinion

ECHINODERMA— GENERAL DESCRIPTION 13

of zoologists. The development/ of Asterina indicates the possible
relation between those two groups. Here MacBride (1896) has
shown that the larva is early attached by the preoral lobe, and
that it bends over on this so as to bring the mouth downwards.
The internal structures do not, however, undergo that complete
translation and torsion which occur in the Crinoid. Traces of it
are seen in the greater development of the left hydrocoel and left
posterior coelom. Fig. XL attempts to show what would happen
in the case of a primitive Pentactcea that bent over in this way ;
while the mouth passed down, the anus and hydropore would tend
to remain on the upper surface, where they could best fulfil their
functions. In the present ontogeny of the Asteroid, the develop-
ment is direct from the Dipleurula to this stage, the intermediate
steps imagined for the phylogeny being omitted as unnecessary.

-M

FIG. XI.

Change from Pentactcea to Stelleroid type. 0, mouth ; As, anus ; M, madreporite ; r.p.c.
and l.p.c, right and left posterior coelom ; l.hc, left hydrocoel ; s.c, stone canal ; pi, preoral
lobe ; ax, axial sinus, remains of anterior coelom.

But to those phylogenetic steps are due the peculiar positions
assumed by the left hydrocoel and left posterior coelom, as well
as the radial folding which they undergo. Study of Fig. XI.
will elucidate the complicated internal arrangement of the develop-
ing Aslerina. Further flexure causes the ends of the curved
hydrocoel to grow around the stalk, which thus deceptively appears
to spring from the oral surface, not from the aboral as in Crinoids.
Subsequently the stalk atrophies, and the young starfish is a free-
moving form, with mouth on the sea-floor, with anus and madre-
porite directed upwards, and with the beginnings of five arms
containing extensions of the left hydrocoel, of the oral coelom
(derived from 1. post, coelom), and of the stomach (Fig. XII.).
During development the larval mouth and anus are closed, and
break through again in their adult positions; this points to a
migration of those openings during phylogeny, which migration
cannot well be repeated in ontogeny.

A vast amount of discussion has taken place over the question
whether the plates of the Crinoid calyx find homologues in other

i4 ECH1NODERMA— GENERAL DESCRIPTION

Echinoderma. The orals on the one side are supposed by some
(e.g. Loven, P. H. Carpenter, Sladen) to be represented by some-
what similarly situated plates in Stelleroidea (" buccal shields " of
Ophiuroids, " odontophores " of Asteroids), and by a circumoral
calcareous ring in Holothurians. The plates of the dorsal cup
have been homologised with plates very similar in shape and
arrangement that are often to be observed in Stelleroidea, notably
on the aboral side of the compact body of Ophiuroids, and with two
circlets of plates at the aboral pole of Echinoids. It has been
supposed that all Echinoderma primitively possessed a definite
calycinal system, thus composed: a central aboral plate ("dorso-
central "), five plates surrounding this (" basals," " genitals " of
Echinoids), five plates following on these and alternating with
them ("radials," "terminals" of Asteroids, "oculars " of Echinoids) ;
these together formed the " apical system," and to them was some-
times added a circlet below, and alternating with, the basals (" infra-
basals ") : five orals, alternated with the radials, and to these P.
H. Carpenter once added an " oro-central," the correlative of the
dorso-central. The oro-central is a discredited myth. The dorso-
central is a plate at the distal end of the Crinoid stem, i.e. in the
preoral lobe (Jp in Fig. X.) ; there is no proof that it ever formed
part of an apical system, and it cannot be considered either homo-
genetic or homoplastic with the aboral central plate sometimes seen
in Stelleroidea. As for the basals and radials
of the Crinoid, they are, as stated above,
formed around the right posterior coelom ;
this also is the position of their supposed
homologues in Asterina (Fig. XII.), and
MacBride's argument that their relations
to the stem are different, does not seem
fatal to the above theory. What is fatal
is the conclusion to which the evidence of
Fm xn fossils forces us — that the free Echinoderms,

Dorsal, i.e.'aborai, view of if they arose from stalked forms at all,
&%rj^Xs%lo£ indubitably did so ages before a calycinal
lobe (p.i) now on oral side ; system had been evolved. Even among
$ stalked forms it appears that regular apical
the systems arose independently in different
- central; M, lines of descent. If, however, it be im-
possible to regard the apical systems of
Echinoidea and Stelleroidea as homogenetic with that of Crinoidea,
there can be no objection to the statement that similar plates are
developed in a similar position with regard to the fundamental
anatomy, under the influence of somewhat similar causes.

The Asteroids were probably the last group to branch off from
the fixed Echinoderms. Hence it is that they retain many features

ECHINODERMA— GENERAL DESCRIPTION

of the Pentactcea, together with epithelial nerves on the floor of
the arm -grooves, as in Pelmatozoa. The Ophiuroids are, fas
explained under Stelleroidea, scarcely to be distinguished from
Asteroids. Whether they branched off at an earlier date or no is
uncertain ; at any rate, they have progressed farther from the
Pentactwa type, in so far as the radial nerves have sunk below
the surface and are covered by " epineural canals," which probably
represent closed food-grooves (Fig. XIII. 1 and 2).

FIG. XIII.

Sections across ambulacra of — 1, Asteroid ; 2, Ophiuroid ; 3, Echinoid ; 4, Holothurian.
amb, ambulacra! ossicle ; amp, ampulla ; fo, radial blood-vessel ; cm, circular muscles ; d.s,
ventral scute ; c, radial epineural canal; Im, longitudinal muscles ; l.s, lateral scute ; m, muscles ;
7i, radial nerve of superficial oral system ; n2, radial nerves of deeper oral system ; p, podium ;
ph, pseudhaemal canal ; v.o, vertebral ossicle ; w, radial canal of the water-vascular system.

The development of Echinoidea has been studied by J. Miiller
(1852), Agassiz (1864), Metschnikoff (1869), Bury (1889), and
many others. The results are summarised by Theel in his admir-
able account of the development of Echinocyamus pusillus (1892).
Up to the stage corresponding to the Dipleurula no important
divergences are manifest. The peculiarities of the ensuing meta-
morphosis appear due to the extreme development of a free-
swimming Pluteus (Fig. XIV.). At an early stage there is an
invagination (am) of the ectoderm on the left side between the
bases of the ventral and dorsal posterior processes of the Pluteus.
The inner end of this sac grows towards the left hydrocoel,
while its opening nearly or quite closes (Fig. XV. 1). The five
primitive lobes of the hydrocoel grow up into the floor of this
sac (Fig. XV. 2), which thus serves as a kind of amnion in which
the young sea-urchin is formed (Fig. XV. 3), until the size of the

i6

ECH1NODERMA— GENERAL DESCRIPTION

FIG. XIV.

Pluteus of Echinocyamus (after Th6el).
About 75 times nat. size.

Explanation of letters to Figs.
XIV. and XV. — o.c, anterior coelom ;
a.d, anterior dorsal arm ; am, am-
niotic invagination ; a.v, anterior
ventral arm ; he, hydrocoel ; hj>,
hydropore ; l.p.c, left posterior
coelom ; 0, mouth ; oe, oesophagus ;
p, podia ; pd, posterior dorsal arm ;
p.v, posterior ventral arm; r.p.c,
right posterior coelom ; », spines of
Echinoid ; sp, spicules of Pluteus ;
sp', the same being absorbed ; st,
stomach.

am

Ip.c,-'

Fio. XV.

Development of Echinocyamus (after Theel). 1, portion of a Pluteus rather more developed
than in Fig. XIV. ; the ectodermic invagination has grown in towards the left coelom, which is
now separating into a posterior portion and a hydrocoel. 2, portion of a Pluteus about twelve
days old, showing the lobes of the hydrocoel growing into the amnion. 3, the same consider-
ably more advanced ; spines begin to develop and the amnion is connected with the exterior.
4, a Pluteus forty-Hve days old, with spines and podia of the young urchin protruding from the
amnion. 5, the young urchin, bearing on its back the remains of the Pluteus spicules and
integument (x 100).

ECHINODERMA— GENERAL DESCRIPTION 17

growing tube feet and spines causes it to break through the outer
wall (Fig. XV. 4). While this takes place the spicular skeleton of
the Pluteus is absorbed, and the body of the Pluteus shrinks up to
a sac on the aboral side of the young Echinoid (Fig. XV. 5). The
hydropore from the first opens on the dorsal surface, which becomes
the aboral side. The right posterior coelom is also under here,
as in Stelleroids. The larval stomach becomes that of the adult,
but a fresh mouth is formed in the centre of the hydrocircus,
while the anus is a fresh formation at the aboral pole.

It is easy to understand that, with this amniotic development
in the body of the larva, most of the traces of the Pentactcea
stage have disappeared. There is, however, evidence of a preoral
lobe, while the coil of the intestine and the radiate structure of
the hydrocoel, nerves, and gonads, bear witness to antecedent
phylogenetic changes. On those changes light is thrown by
palaeontology, which teaches us that the primitive Echinoid had a
spheroidal body, with muscular, flexible walls, in which irregular
plates were developed ; the mouth was at the centre of the lower
surface ; the anus on the upper surface, and near it the madre-
porite (the successor of the hydropore). Combining with the
svidence from fossils that from comparative anatomy, we infer
that the gut had a simple dextral coil ; that the oesophagus was
surrounded by three rings — water-vascular, blood-vascular, and
nervous ; and that from each ring five branches passed up the
inside of the body wall to the aboral pole ; that branches from the
radial water-vessels passed, between the plates in the body wall,
to the exterior, and became suckers assisting locomotion, the
complete structures being ambulacra ; that gonads were five,
unpaired, and interradially disposed in the body cavity. Such a
form had lost the stem of the Pentactcea, and had never possessed
an apical system of plates. It had, however, already developed
food-grooves, with nerves and ambulacral vessels, while there must
have been some radiate arrangement of the gonads. The sinking of
the nerves and closure of the food-grooves forming epineural canals
(Fig. XIII. 3) probably took place as we suppose it to have done
in Ophiuroids. Among Pelmatozoa, the Edrioasteroidea (p. 205)
present a structure removed from that of the primitive Echinoid
in little but the upward position of the mouth and (probably) the
madreporite, and the functional food-grooves ; the notable point about
the latter is the presence of openings between the flooring-plates,
apparently for the passage of processes from the radial water- vessels.

The peculiarities in the structure and development of the
Holothurians may perhaps be ascribed to their having in many
respects regressed from the Pelmatozoic towards the Dipleurula
type (Fig. XVI.). Thus the mouth has again come to lie at one
end of the body, while the anus is at the other. With the mouth

18

ECHINODERMA— GENERAL DESCRIPTION

has gone the genital pore ; and the madreporite, though uncon-
nected with the ex-
terior in the adult of
most now living, must
have moved towards
this end also. But
there is an important
difference between this
and the Dipleurula in
the coil of the gut, and
Fl°- XVI- the vastly altered rela-

Diagrammatic reconstruction of the imagined primitive firvna nf fV,0 pr»plr»rnip

Holothurian type, for comparison with Figs. I., VII., and tlum

XI. 0, mouth ; As, anus ; M, hydropore ; l.hc, left hydro- cavities with their in-
coel ; g, genital opening.

tervemng mesenteries.

The arrangement indicates that the mouth, hydrocoel, madre-
porite, and associated organs of a Pentactcea gradually moved
anteriorly away from the anus, thus coming nearer to the
stem (preoral lobe of larva) and lengthening the gut by another
half-coil. When the fixed existence was given up and the
food -grooves closed in, leaving the external podia from the
water-vessels, then the rays were able to extend equally in all
directions from mouth to anus. The anterior ray and the two
adjoining rays had thus cpme to be on that side of the elongate
body which was directed towards the sea-floor, and to this they
clung, or on it they crawled, by suckers
which developed on the podia, which
thus became " tube-feet." The left and
right posterior rays ran along the upper
surface of the body, forming the bivium Stc
of the Holothurian, homologous with the
bivium of the Pelmatozoa, as shown by
the position of the hydropore.

The view has been held that the
Synaptidae, with their simple structure
and straight antero-posterior gut, repre-
sent the simplest and most ancestral
Echinoderms. But if the above account
be correct, this simplicity is only ap-
parent, and is the result of regressive Semon) T< tentacle8 . ^ ,nadre.

changes. Such is the view that now porite ; stc, stone canal ; st, stomach ;
„ , ° . . mi n . , 7 oc, supposed otocysts ; pb, Polian

finds general favour. The Pentactula vesicle depending from hydrocircus;

stage (Fig. XVII.), in the development i^.^Si'TS" **' "
of Synapta, with five interradial circum-

oral tentacles, slightly curved gut, and aboral anus, is therefore
not the modern ontogenetic representative of the phylogenetic
Pentactcea, as Semon supposed.

ECHINODERMA— GENERAL DESCRIPTION 19

Further, if our present theory be correct, we must suppose that
the larval history of the Holothurians has been exceedingly com-
pressed ; so that, to take but one point, the development of the
straight larval gut in^o the coiled gut of the adult takes place, not
by migration of the mouth and associated organs, but by lengthen-
ing and twisting of the gut itself. It is noteworthy that the two
lateral radii of the trivium with their nerves and muscles and tube-
feet, as well as the oral tentacles to which they eventually give rise,
develop much more slowly than the three other radii. Those are
the three radii which are assumed in the above account to be
homologous with the original three radii of the primitive Pel-
matozoan (cf. Fig. IX.).

It therefore appears that the Holothurian stock branched off
from the Pelmatozoa before complete pentamerous symmetry of
the hydrocoel and associated organs had arisen, before any definite
calycinal system had developed, while the gonads were still a
simple strand opening to the exterior by a single posterior gono-
pore. The diminution of the skeletal elements did not favour
their preservation as fossils. Their spicules indeed are found in
the rocks from at least the Carboniferous downward, but if we
except the problematic Sphaerites, Quenstedt (1852, non Dufts), no
fossil Holothurian is known. The class was perhaps an early
offshoot from the Edrioasteroidea. This theory explains how it is
that the Holothurians are primitive in so many characters, although
the most specialised in others ; they are primitive as regards
Pelmatozoic structure, specialised as regards Eleutherozoic,

Symmetries. — The radial symmetry due to the fixed phylo-
genetic stage is usually pentamerous. Hexamerous symmetry
was independently acquired by some Cystidea. Variation from
pentamerism may arise suddenly (discontinuous meristic variation
of Bateson), producing hexamerous or tetramerous individuals, or
species, or genera, according as the sport becomes fixed. There
may also be a duplication, or further multiplication of radii, as in
the ten-rayed Promachocrinus, or an intercalation during growth, as
in the many-rayed Labidiaster ; this is a different thing from the
branching of a radius, such as occurs in Crinoids, Astrophytidae,
and elsewhere. Again there may be variation by gradual atrophy
of one or more radii, as in Calceocrinidae, and some heart-urchins
and Holothurians. In spite of these variations, it is generally
possible to divide the body of an Echinoderm, by planes passing
through the ambulacra from the long or main axis, into approximately
corresponding portions, " antimeres," normally five. These planes
mark the radii, or better perradii, since the terms ray and radius
have been used loosely. Organs bisected by them are "perradial" ;
such are invariably the main ambulacra! vessels, the arms of
Stelleroidea and brachia of Crinoidea, with their included organs.

20 ECHINODERMA— GENERAL DESCRIPTION

Half-way between the perradial planes are the planes marking the
interradii. Organs bisected by these are " interradial " ; such are
the interambulacral areas of the test, the oral plates of Crinoidea,
the gonads of Echinoidea. Between the perradii and interradii
are adradii, a term little used in practice ; thecal plates adjoin-
ing the ambulacrals are called " adambulacral." In a regular
pentamerous Echinoderm an interradius is opposite to a perradius,
and an adradius opposite to an adradius.

All Echinoderms have a bilateral symmetry. Primitively the
plane of symmetry, the sagittal plane, is determined by the mouth
(anterior), the anus (posterior), and the hydropore (dorsal). But,
in the first place, this sagittal plane, when clearly shown, is not
the same as the sagittal plane of the Dipleurula. In Pelmatozoa
it certainly is not ; in Holothurians it only approximates to it.
Secondly, in Echinoidea and some Cystidea, and in such Asteroidea
as have an anus, the plane passing through the vertical axis and
the madreporite (M plane) is not the same as that passing through
the vertical axis and the anus (anal plane). Thirdly, the rela-
tions of the anal plane to the M plane and to the radii may vary
even within a single class, e.g. Echinoidea and Cystidea. Conse-
quently the selection of any one plane as a plane of orientation
for the different classes is arbitrary. Also it is convenient. We
take then the M plane and note that the hydropore lies in an in-
terradius with a radius opposite to it (Fig. XVIII.). That radius we
denote by A. Then placing the animal with its mouth upwards
and going round the test in the direction of the watch-hand (i.e.
dextrally), we denote the other radii in order, E, C, D, E. The
hydropore lies in interradius CD. In a developing Holothurian,
or in such Holothurians as retain an external madreporite (Fig.
XVIII. 3), the anus and mouth both lie in the M plane, forming
the poles of the long axis, while radius A bisects the ventral
surface ; this therefore is the sagittal plane of bilateral symmetry,
and Cu6not, 1891, calls it the " Holothurian plane." In a Crinoid
(Fig. XVIII. 1), anus, mouth, and aboral pole, all lie in the M
plane, which here also is the sagittal plane ; but the anus, in inter-
radius CD, never marks the aboral pole of the main axis, though
it may usurp the place of the mouth at the upper pole. Many
Cystids, and apparently the Blastoids, have a similar orientation.
Other Cystids differ in that the anus lies to left or to right of
the hydropore, while the relation of the radii to the M plane
is not clearly defined. In Echinocystis (p. 301), which probably
represents the relations in the primitive Echinoid, the symmetry
remains as in Pelmatozoa ; while the mouth is at one pole of the
main axis, the anus lies in or near the M plane, which is therefore
the sagittal plane, but the madreporite is near to the aboral pole.
In later Echinoids the case is altered (Fig. XVIII. 4) ; the first step

ECHINODERMA— GENERAL DESCRIPTION

21

appears to have been the passage of the anus to the aboral pole,
the madreporite remaining eccentric and marking interradius CD ;
whether this CD is identical with CD of Echinocystis, is another
question ; then the anus moved away from the pole in the direc-
tion of radius B, so that the anal plane made an angle of 72° with
the M plane. This new plane (interradius Dfi, radius B) is termed
by Cuenot the "Echinid plane"; Loven (1884) has shown that
the plates of the five interradii in Echinoidea are disposed sym-
metrically with reference to this plane. The sagittal plane of many

Fio. XVIII.

Planes of symmetry in Echinoderma (partly after Cuenot). 1, Crinoid ; 2, aroid ;
3, Holothurian ; 4, Regular Echinoid ; 5, Irregular Echinoid. As, anus ; M, plane passing
through madreporite ; Ech, Echinid plane ; L, Loven's plane. For other letters, see adjoining
text.

other sea-urchins, notably the heart-urchins and their allies (Fig.
XVIII. 5), i.e. the plane passing through mouth, anus, and apical
pole, corresponds with neither the M plane nor the Echinid plane,
but passes through radius D and interradius AB ; Cue"not calls it
" LoveVs plane." The bivium (AB) and trivium ((7, D, E) of a
heart-urchin are therefore in no way identical with those parts in
a Holothurian, a Crinoid, or a Stelleroid. In those starfish that
have an anus (Fig. XVIII. 2), that organ, as shown by Ludwig, is
in interradius BC ; this with the vertical axis marks the " Asterid

22

ECHINODERMA— GENERAL DESCRIPTION

plane " of Cuenot. It is scarcely worth while to describe yet other
divergences of the sagittal plane from the M plane, such as occur
in Cystidea and Blastoidea. All of them are due to the imposi-
tion of a tertiary bilateral symmetry, obscuring the previously
existing secondary bilateralism that had already replaced the
bilateralism of the Diplewula.

Cavities and their Contents. — The cavities into which the
thecal cavity is divided by the ontogenetic changes above described,
are : — (1) Gut and appendages, derived from the archenteron, with
mouth and anus in part produced by invagination. (2) Coelomic
cavities : (a) the ambulacral system, derived from the left hydro-
coel ; (b) the main body-cavity derived chiefly from left posterior
coelom, which in Pelmatozoa, Stelleroidea, and Echinoidea becomes
mainly adoral ; (c) the aboral body -cavity of Pelmatozoa (with

columnal extensions), of Stelleroidea,

,i. and of Echinoidea ; (d) the axial

sinus of the same three classes,
derived from the anterior coelom,
running down into the stem in Pel-
matozoa, indirectly connected with
the hydrocoel through the stone
canal, and containing "the axial
organ" (p. 23); (e) a perioeso-
phageal sinus, sometimes subdivided,
is completely or incompletely sepa-
rated from (b), especially in Holo-
thurioidea, Echinoidea, and Ophi-
uroidea.

The coelomic cavities are lined
by pavement endothelium, usually
ciliated, and sometimes further pro-
vided with special ciliated or flagel-
lated organs which keep the con-
tained fluid in motion (e.g. "urns"
of Synaptidae (Fig. V. 4, p. 233) ;
"ciliated cups" of Crinoid arms,
(after Geddes). i, amoeboid corpuscle especially pinnules ; free flagellate

with granules of brown ferruginous pig- f. • « t.- -j \ mi. ^i • j •
ment, perhaps respiratory. 2, amoeboid Cells of Echinoidea). The fluid IS

SS£ t^£§&& similar to that found in the lacunar

viren, (after Foettingert ; corpuscles of « blood-vascular system " : it is SCa-
various shapes, with red colour supposed J . '

due to haemoglobin. 4 and 5 from the water, perhaps taken in through the

madreporite, containing a variable

amount of. albumen in solution-

especially in the lacunar system,
and sometimes slightly yellowish or reddish. In it float various
bodies, viz. (a) amoebocytes (Fig. XIX. 1, 2, 4) capable of wandering

Pio. XIX.

Corpuscles of the coelomic fluid. 1 and
2 from the Echinoid, Echinus sphaera

ECHINODERMA— GENERAL DESCRIPTION 23

through all the tissues, including the skeletal, and containing re-
fringent granules, proteids, fat, and a yellow pigment called "echino-
chrome" (MacMunn, 1885) ; they seem to be specialised as bearers
of reserve food, as calcigenous cells (p. 28), as phagocytes, and
as bearers to the exterior of waste products often pigmented
(Durham, 1891, St. Hilaire, 1897) ; (b) red corpuscles with haemo-
globin (Foettinger, 1880), non-nucleate, but vacuolate or granular
in water-vessels of Ophiuroids (Fig. XIX. 3), nucleate in various
coelomic cavities of Holothurians (Howell, 1886; Cue"not, 1891,
Fig. XIX. 5); the respiratory nature of these is demonstrated by
their containing haemoglobin.

Whatever may be the homologies of the hydrocoel, there is,
physiologically speaking, no nephridial or other excretory system
in Echinoderma. The function is probably performed by the
wandering cells just mentioned.

The Axial Organ has had many functions ascribed to it, as
shown by its various names : Heart (Tiedemann), Pseudo-heart,
Central Blood-plexus (Ludwig), Glandular or Chromatogen organ
(Hamann), Lymph G-land, Madreporic Gland (Koehler), Collateral
or Plastidogen organ (Perrier), Ovoid Gland (Perrier, Cuenot, and
others), Genital stolon, Plexiform Gland or Dorsal organ (P. H.
Carpenter), Kidney (P. and F. Sarasin). The Sarasins (1888)
give a good account of the literature ; later notes of value are by
Cuenot (1891) and Durham (1891). Generally it is a brownish,
finely lobed, often pear-shaped body, showing under low magnification
a complicated arrangement of tissue strands (Fig. XX. 1). It does
not occur in Holothurioidea. In the other classes it is developed
in the axial sinus by irregular growth of endothelium, which forms
canal-like strands separated from one -another by spaces derived
from the axial sinus; these latter are therefore primitively
connected with the water -vessels and madreporite through the
stone canal. Strands growing out at an early age from the central
plexus become the gonads, but the connection may be lost in later
life. In association with the genital strands are also radiating
" haemal strands," not true blood-vessels, but serving for the
transmission of nutrient cells. Such cells, as well as pigmented,
excretory amoebocytes, are found in quantity in and about the
axial organ. In Pelmatozoa the axial organ, surrounded by the
axia'l sinus and the lobes of the chambered organ, stretches right
down the stem (Fig. XX. 2). The position of the axial sinus with
regard to the gut suggests that nutrient fluid passes by osmosis into
the axial organ, which thus serves as a kind of distributor, but there
is no evidence of pulsatile pump-action, i.e. it is no heart. The
evidence of new cell-formation is too slight to warrant the idea
that the axial organ is a factory of amoebocytes. There is still room
for study of this peculiar body, especially through experiment.

ECH1NODERMA— GENERAL DESCRIPTION

The Genital Organs throw light on the axial organ, since it
exists only in those classes in which the gonads are affected by
radiate symmetry, and not in the Holothurians. It follows that
the axial organ was a secondary development. Ontogenetically
the genital strands bud off from one end of it, where a ring is

ac/

FIG. XX.

The axial organ. 1, longitudinal section through part of the organ in the sea-urchin,
Sphaercchinus yrunularis (after Leipoldt). x 350. pg, pigment masses; lac, peripheral blood
lacunae of the organ ; ire, wandering cells. 2, longitudinal section, showing relations of the
organ to the chambered organ in young Antedon (after Perrier). ax, cells of axial organ ; /,
fibrous membrane, which partly envelopes it ; <•;>, epithelial coat continuous with chambered
organ ; s, septum of chambered organ; n, nerves proceeding from libiillar envelope of same ;
c, cirri, one just budding out ; ac, atrophied axial canal of stem, continuous with axial organ.

formed, and with the extensions of these go also extensions of the
axial sinus (Fig. XXL). The single gonad of the Holothurioidea,
connected with the dorsal mesentery, appears therefore to be the
homologue of the axial sinus and organ rather than of any one of
the interradial gonads of the other classes. The gonads, therefore,

ECHINODERMA— GENERAL DESCRIPTION

are of endothelial origin. There appears, however, as shown by
Hamann, to be a migration of the actual sexual cells ; and the
view of Cuenot that these are primarily amoebocytes derived
from the axial organ suggests their possible mesodermic nature.
Compare the migration of sexual cells derived from the ectoderm
in Hydroids. The growth and minute structure of the ovum have
been described by Crety (1894), of the spermatozoon by Field
(1895); both authors refer to preceding literature. No striking
peculiarity is presented by Echinoderm gonads (cf. Fig. VI. 1).

"bm

FIG. XXI.

Diagrams showing relations of pseudhaemal and water-vascular systems and axial organ in
Asterias rubcns (after Chadwick). "2 is x 50. «J>.h, aboral haemal ring ; ax, axial organ ; ax.p,
axial perihaemal canal ; bm, buccal membrane ; ;/, genital strands ; gh, absorbent haemal strands
leading from the gut-wall ; hr, circumoral haemal ring ; ip, inner perihaemal canal (cf. Fig.
XXII.); I, blood lacunae; Jlf, madreporite ; n, nerve ring; op, outer perihaemal canal; ric,
radial water-vessel ; rh, radial haemal strands; st, stomach with folded wall ; st.c, stone canal ;
wr, circumoral water-ring.

The sexes are nearly always separate, and fertilisation takes place
in the water.

The Haemal Systems of Echinoderma are of two types, which
may coexist, but of which one usually predominates. Neither is a
true vascular system, but each consists of a series of smaller lacunae
(spaces without definite walls) or larger sinuses, sometimes appear-
ing as closed, but probably always having some communication,
however minute or indirect, with the other cavities of the body.
The fluid in these spaces differs from the ordinary coelomic fluid only
in containing more albumen, and has, likewise, no definite circula-
tion. The systems are : (a) Pseudhaemal, consisting of a ring

26 ECHINODERMA— GENERAL DESCRIPTION

placed between the ring and radial nerves of the oral system
above, and the ring and canals of the water-system below. This
system is dominant, perhaps the only one, in Stelleroidea, where
it communicates with the general body-cavity and the axial sinus ;
it is present in Echinoidea and Holothurioidea, in which classes it
is said to be closed ; it is so much reduced in the Crinoidea that
its existence is denied by some authors. In Asteroidea (Figs. XXL,
XXII.) the ring is divided into an outer and an inner ring by an
oblique septum, from each angle of which a vertical septum passes
down each radial canal. Formerly the system was supposed to
develop as a cleft in the mesenchyme, and therefore was called the
" schizocoelic system " ; MacBride (1896) has shown that in Asterina
the inner ring is an outgrowth from the axial sinus, while each of
the five compartments of the outer ring and canals arises separately
as an outgrowth of the coelom, the outgrowth in the madreporic
interradius being derived from the anterior coelom, the rest from

the left posterior coelom. (b) Lacunar,
present in all classes except perhaps
Stelleroidea, and developed as lacunae
or small spaces in the connective tissue,
and therefore mesodermal (Fig. XXL).
It is differentiated into a network in the
wall of the gut, absorbing therefrom the
nutrient fluid, which is carried by a
main trunk on each side of the gut to a
circumoral ring; from this run radial
FIO. xxii. canals, below the pseudhaemal canals

Diagram of the pseudi.aemai when present, and above the water-
system of Asterina (after MacBride). , A , ., ., . , . ,
Seen from above: M marks the M VCSSelS, While it IS Connected With net-
plane; in and ex, inner and outer wnrl,.Q nn tl>A filirfj,«p nf knfV, annnrls nnH
divisions of the perihaemal ring ; a, WOrKS On tne SUHaCC 01 DOtn gOnaQS and

the one arising from the anterior axjai organ. The lacunar system of the

coelom; ox, axial sinus, passing <>,,, ., , ._, . . ,

towards, but not opening into, the btelleroidea diners in the absence of an

absorbent network, and is, says Cuenot,
a derivative of the axial organ, and
therefore endodermal, i.e. it is only the pseudhaemal system greatly
extended.

Respiration takes place through all exposed processes of the
ambulacral system, and through the body wall where thin enough,
as in some Holothurians. Specialised outgrowths or foldings of
the latter are : the " external gills " of Echinoidea, outgrowths of
the circumoesophageal sinus ; the papulae of Asteroids, containing
diverticula of the body cavity; the bursae of Ophiuroids ; the "pectini-
rhombs " of some Cystids ; the " hydrospires " of Blastoids and
some Crinoids. Respiration is also effected by water entering the
alimentary canal, whether through mouth or anus ; in the latter
case it is again expelled. Special structures connected herewith

ECHINODERMA— GENERAL DESCRIPTION

27

are : the " respiratory trees," which occur in some Holothurians
as outgrowths from the cloaca ; the anal tube of Crinoids, which
in some Palaeozoic forms was large and with folded walls, forming
the so-called " ventral sac " ; the " accessory intestine " of Echinoids,
a kind of by-pass, permitting water to flow through without
interfering with the digestive process going on in the main gut.

Lymph-glands. — The amoebocytes are formed in specialised
glandular regions of both haemal and ambulacral systems. Of
the former nature are the radial and-pharyngeal vesicles of Eegular
Echinoids, first described by Prouho (1888) ; the greater part of the

14 9 13

11
FIG. XXIII.

12

14

Echinoderm histology. 1, fundamental fibrous substance, with nuclei and an embryonic
cell, from Echinaster sepositus. 2, stellate embryonic mesenchyine cells of Asterias gladalis.
3, gelatinous connective tissue of Spatangus purpureus ( x 200). 4, elastic fibres of connective
tissue from Asterias glacialis. 5, fibres from stalk of a pedicellaria of same. 6, muscle-fibre
of same. 7, muscle-fibre from a spine attachment of Toxopneustes lividits ( x 200). 8, muscle-
fibre from jaw pyramid of same (x 250). 9, muscle-fibres from gut of Sphaerechinus eseulentus
(x 175). 10, transverse section through a muscle-bundle of Asthenosoma urens; s, sheath of
connective tissue from which proceed septa that limit the smaller divisions. 11, stroma
continued as fibrils across a suture in Spatangus purpureus, the stereom of the ossicles
dissolved away. 12, transversely striate muscle -fibres of Echinus acutus. 13, dorsal ligament
of arm of Antedon ( x 125). 14, interarticular substance of Isocrinus asteria. 1, 2, 4, 5, 6 (after
Cuenot). 3, 7, 8, 9 (after Hoffmann). 10 (after P. and F. Sarasin). 11, 12 (after Hamann).
13 (after W. B. Carpenter). 14 (after Joh. Miiller).

Stelleroid lacunar system, just mentioned; and the " spongy organ"
of Crinoids in the oral ring. Connected with the ambulacral system
are the " Polian vesicles " found in most Echinoderms other than
Crinoids, and the " Tiedemann's bodies" of Asteroidea (p. 243).

The primitive Mesenchyme cells, derived chiefly by migration
from the endoderm, partly from the ectoderm, have a large nucleus
and indistinct, often amoeboid, cytoplasm. From them are developed
connective and muscular fibres, amoebocytes and calcigenous cells,
and intercellular, gelatinous, and fibrous substances. The muscle-fibre
(Fig. XXIII. 6-9) derived from a single cell is smooth and straight,

28

ECHINODERMA— GENERAL DESCRIPTION

clearly defined at the ends, with a lateral nucleus. A few striated
muscle-fibres are known in Echinoidea (Geddes and Beddard, 1881 ;
Hamann, 1887). A semi-muscular, hyaline tissue of wavy, nucleated
fibrils is peculiar to Crinoidea, and is called "ligament tissue."
There are also muscles of endothelial origin. Connective tissue
fibrils are nucleate and vary in length and shape ; there are also
rounded or stellate cells (Fig. XXIII. 4, 2). Intercellular substance,
secreted by mesenchyme cells, often attains great thickness in
the integument ; it may remain a soft jelly, or become tough as
indiarubber, or may split up into interlacing fibrils; it usually
contains amoebocytes and ordinary connective tissue cells ; it forms
also interarticular substance (Fig. XXIII. 14), elastic ligament,

Fia. XXIV.

Stereom formation. 1, from the hinder portion of a Pluteus of Echinus miliaris. s, one of
the large supporting rods of the Pluteus; c, a throe-pronged spicule surrounded by a group of
calcigenous cells, which derive their lime through a meshwork of pseudopodia and cells (a),
from the rods of the Pluteus and from their broken ends, which are seen just below r. 2, 3,
earliest stages of a spicule of Echinocyamu*, surrounded by calcigenous cells. 4, infrabasal of
an Antedon larva forty-eight hours old (x 230). 5, regular stereom from the outer part of the
cup of llolopus ( x 56). 6, portion of horizontal section of Holopus cup, showing relation of
irregular (i) and regular (r) stereom (x 15). 1-3 (after Theel). 4 (after Seeliger). 5, 6 (after
P. H. Carpenter).

and the walls of internal organs. Parallel structures are found in
the cartilage of Vertebrata.

The formation of a calcareous skeleton by the mesoderm was
as pronounced in the oldest known Echinoderms as it is to-day,
indeed, more so. To the prickly skin, so commonly a result of
this, is due the name of the phylum (Ixtvos, a hedgehog; &/o/xa, skin).
Amoeboid cells in the mesenchyme have the power of fusing by
pseudopodia into plasmodia or into reticular tissue (Fig. XXIV.).
Where the pseudopodia meet and fuse, the protoplasm secretes
a small calcareous spicule (intracellular, Theel ; extracellular,
Semon), which gradually increases in size along the lines of the
pseudopodia. Such spicules meet and fuse by their processes,
thus building up a hard tissue ("stereom"), with a structure that
in section appears reticular, but really is more like a beam-and-
rafter-work. As, in the growing Echinoderm, the protoplasmic

ECHINODERMA— GENERAL DESCRIPTION 29

reticulum becomes a more definite stroma, so the pattern of the
stereom acquires definiteness, and varies in the different parts of
an individual, as well as in different species. In the course of
ages the spicules of the adult have themselves come to acquire
definite shapes characteristic of species, and this is notably the
case with the complicated "wheels," "anchors," and "tables" of
Holothurians (Figs. II. 3, 6; V. 6, 7, pp. 222, 233) ; but, as Stewart
and Bell have shown, also applies to the spicules of the thecal
cavity in sea-urchins. The spicules of the theca usually fuse into
plates, those of the appendages (brachia, stem, etc.) into ossicles.
There is no real distinction ; but it is often the case that the
reticulum of the ossicles runs in straighter lines, while that of the
plates is a more open mesh-work; this is due to the definite
arrangement of the connecting fibres of the stroma in the append-
ages, and when this is definite in the theca the result, as we shall
see under Cystidea (p. 42), is the same. Attempts to use this
as an important character in distinguishing brachials from radials
(p. 112), or a dorso- central from columnals, have no secure
foundation. The stereom is absorbed by cells similar in outward
appearance to those which deposit it, and the calcareous salts are
transmitted by communicating pseudopodia from the absorbent to
fresh depositing cells. Thus the spicules of the Echinoid Pluteus
form a reserve for the growing urchin ; thus, too, the anal plate
of the growing Antedon is absorbed, and its material used by the
increasing radials. Theel, to whose observations (1892-96) much
of this knowledge is due, compares the reticular tissue, the
osteoblasts, the osteoclasts, and the " Howship's foveolae " formed
by the latter, of Vertebrata with the similar structures in Echino-
derma. Bone in its first stages, especially that formed in connec-
tive tissue, is marvellously like Echinoderm stereom, and is like-
wise of mesodermic origin. But, whereas bone is an extracellular
formation, it is probable that the spicules of Echinoderms, like
those of Sponges, are intracellular. The otoliths of some
Holothurians and the biscuit spicules of others (see p. 224) are
distinctly intracellular formations. Bone, moreover, is phos-
phate, not carbonate, of lime, and does not retain the markedly
crystalline character always possessed by Echinoderm stereom,
even when highly complicated. Each skeletal element of an
Echinoderm acts as a crystallographic individual, polarising light
and cleaving along the planes characteristic of calcite. In fossils
the cleavage is often emphasised by an infilling of the spaces with
secondary calcite which has axes identical with those of the original
crystal. According to Semon (1887), every skeletal element begins
as a tetrahedron, usually in the form of a trifid spine with branches at
an angle of 120°. The formation of similar spicules in sponges has
been studied by Minchin (1898, see Part II., Chapter on Porifera).

ECHINODERMA— GENERAL DESCRIPTION

The epiblast develops into an ectoderm, ciliated in whole or
part. In the adult this often becomes merged in the mesoderm
so as to be indistinguishable ; in Ophiuroids it is for the most part
calcified by the immigration of calcigenous cells ; in Asteroids and
Echinoids it remains with its cilia ; in Crinoids it remains on the
tegmen of some forms, and in the ventral grooves, where it is
ciliated ; in Holothurioidea it is very variable, being best preserved
in Synaptidae. From the epithelium is derived the superficial
" oral nervous system," composed of the circumoral ring and
radial nerves. In Asteroidea and Crinoidea this remains on the
surface, but in the other classes it sinks below, while the grooves in

cor

Eyes of Echinoderms. 1, the end of a ray in young Asteroid, t, terminal tentacle ; p, podia ;
e, eye-spot. 2, section across the eye-spot in Astcrias, showing seven eye-cups, en, endo-
thelinm of perradial water-vessel ; c/, connective tissue fibres ; nn, nerve below epithelium.
3, section across an eye-cup (after Cuenot). oil, cilia ; cut, transparent cutis, bef«w which
are the pigmented and retinal colls. 4, diagram of eye-spot of Diadcma setosum, modified
from Sarasin. The eye-spot, with its hexagonal elements, is surrounded by the pigmented
integument, composed of a glandular, columnar epithelium (ep) which merges into the cornea
(cor) above, and the eye-cups below ; each cup is coated at its base with pigment (pg), and rests
on a nervous layer (tin), below which is again pigment. Connective tissue fibrils (c/) pass
through this in places.

which the nerves originally lay are closed over them, forming
" epineural canals " (Fig. XIII.).

Two other nervous systems are formed in Echinoderma : (a)
the " deeper oral nervous system " from the mesoderm, underlying
and roughly following the course of the superficial system ; said
to be absent in Crinoids, but is probably represented by their
" sub-epithelial system " ; present in all other Echinoderms except
those Echinoids that have no masticatory apparatus ; it chiefly
'innervates the muscles in the oral side of the body wall ; (b) the
" apical nervous system," most pronounced in Crinoids, and derived
from the endothelium of the axial sinus ; it is believed to occur in
all other classes except Holothurioidea ; it innervates the dorsal
musculature of the test and appendages.

ECHINODERMA— GENERAL DESCRIPTION

Sense-organs are but slightly developed. They are tactile,
visual, and auditory or orienting. Tactile organs are furnished
by the ambulacral appendages, the spines, and the pedicellariae.
The chief among the ambulacral appendages is the " terminal
tentacle," the unpaired end of the perradial water-canal, differ-
entiated only in Stelleroidea and Echinoidea (Fig. XXV. 1). In
Asteroids it is coated with columnar epithelium bearing long cilia
and innervated by the radial nerve ; in Ophiuroids this nerve,
which is sub-epithelial in the arm, becomes epithelial in the
tentacle ; in Echinoidea the terminal tentacle is a sensory papilla
penetrating the pore in the "radial" or "ocular" plate of the
apical system. The remaining ambulacral appendages, the podia,
whether sucking feet, as in Echinoidea, Holothurians, and Asteroids,
or tentacles, as in Ophiuroids and Crinoids, are highly sensitive, and
sometimes have special developments. Thus, on the adoral side
of the oral tentacles of Synaptidae are two rows of papillae, of
which the tip is concave and ciliated ; these are called " sensory
buds," and supposed to be organs of smell or taste. Again, the
podia of Crinoids have small papillose projections, each papilla
armed with three stiff but fine hairs. Similar papillae, sometimes
more developed, occur in some Ophiuroids and Echinoids. Spines
occur chiefly in Echinoids, less pro-
nounced in Stelleroids, and rarely in
Crinoids and Blastoids. Not all spines
are sensory. The smaller spines of Cidar-
oida, surrounding the larger spines and
the main openings of the theca, are
covered with ciliated epithelium, and
bear tactile hairs at the tip ; the minute
spines (clavulae) on the fascicles of Spat-
angidae (see p. 319) likewise have a
ciliated integument, probably with sen-
sory cells. The club-shaped spines of
some Ophiuroids are covered with a
glandular and sensory epithelium. Pedi-
cellariae occur in all Echinoids, some
Asteroids, and a few Ophiuroids; they are
small, forceps -like appendages derived
from spines (see p. 287). All are covered
with a glandular, sensory epithelium,
which in the "glandular pedicellariae"
of some Echinoids develops special tactile
prominences. Visual organs are known
only in all Asteroids, a few Echinoids, and
Synapta, but other (probably all) Echinoderms are sensitive to light,
owing, perhaps, to the action of the pigment-bearing amoebocytes.

FIG. XXVI.

A single cup of Fig. XXV. 4,
reconstructed from the evidence
of Sarasin. Outside is a ciliated
cuticle, covering the transparent
cellular cornea ; below is a refrac-
tive body (re/), possibly a vacuo-
late and multinucleate cell ; the
nuclei (ns) lie in strands of proto-
plasm ; outside the base of the
cup is a layer of anastomosing
pigment cells, which pass up from
the pigment layer below through
the ganglionated (gg) nervous
layer.

32 ECHINODERMA— GENERAL DESCRIPTION

In Asteroids (Fig. XXV. 1) an eye-spot (e) lies at the base of each
terminal tentacle (t) on its aboral side. This spot is a red cushion
in which are many conical cups, each representing an eye (Fig.
XXV. 2). The wall of each cup is formed of pigment cells and
interspersed unpigmented retinal cells (Fig. XXV. 3). The
Echinoid Diadema setosum has a black integument with numerous
spots which, owing to interference of light, appear blue. Each
blue spot, as proved by P. and F. Sarasin (1887), is a com-
pound eye (Fig. XXV. 4). The structure of a single element is
shown in Fig. XXVI. Supposed auditory organs ("Baur's
vesicles" or otocysts) have been described only in some Holo-
thurians, e.g. Synapta (see Fig. XVII. oc; also p. 234 and Fig. V. 5
on p. 233). The sphaeridia of Echinoids (see p. 288) are sup-
posed to be organs of orientation, or of taste and smell (Loven),
or for appreciating chemical changes in the water (Ayers, 1885).
They occur only on the oral side of the theca, and when the
animal is in the natural position they hang down like the clapper
of a bell ; but when the animal is tilted, each sphaeridium presses
against the nerve cushion surrounding its stalk.

Distinctive Characters of Phylum and Classes. — The foregoing
account has introduced the fundamental features of Echinoderm
morphology, laying stress on characters common to the whole
Phylum rather than on those that distinguish the various classes.
It has, however, tended to show the inner meaning of those out-
ward distinctions between the chief types with which the chapter
opened, and the student may perhaps have realised that " the
homologies within the Echinoderm stock " are, as Semon has in-
sisted, often more apparent than real. In drawing up a definition
of the Phylum that shall include the most primitive forms of fixed
Echinoderms known, one cannot utilise most of the characters
usually thus employed in systematic treatises, since they are
secondary, homoplastic acquisitions, often with no true homology.
It is, for instance, not sure that all Echinoderms have a radiate
symmetry, even an obscured one. It is true that all recent
Echinoderms have a lacunar, haemal system ; but that system in
Stelleroidea is not homologous with the one in Crinoidea. It is
highly probable that all animals to which the name " Echinoderm "
could have been applicable since the beginning have had a portion
or portions of the anterior coelom specialised as a hydrocoel ; but
this is different from the questionable assertion that all Echino-
derms have an ambulacral system.

On the other hand, in any attempt to limit the several classes,
respect should be paid to deep-seated structures illustrative of past
history and genetic affinity rather than to the obvious but super-
ficial differentiations that characterise the representatives now
living. We have to make our classificatory partitions run back

ECHINODERMA— GENERAL DESCRIPTION 33

as far as possible. Since the factor determining the lines of
evolution appears to have been position with regard to the sea-
floor, this must no longer be contemned as " mere difference of
habit." The first Echinoderms were not necessarily fixed, but
fixation probably affected all representatives of the Phylum at an
early period and produced gradual changes, the first being the migra-
tion of the mouth and left hydrocoel to the upper pole. Those
forms in which the oral pole remained uppermost, whether actual
fixation by the aboral pole persisted or no, are to be distinguished
from those in which the oral pole again shifted, accompanied by loss
of fixation. Leuckart's term Pelmatozoa (1848), though primarily
connoting the actual or potential possession of a stalk, has come
into general use for the former group. The term Statozoa, pro-
posed by Bell (1891), implies absence of locomotion, and is
therefore not so great an improvement as to compel its adoption.
The term Crinoidea was extended by Roemer (1851) to include
all Pelmatozoa, but such extension does violence to the intentions
of J. S. Miller, the coiner of the name (1821). The forms with
oral pole uppermost may, it is true, be divided into classes ; but,
as maintained by Huxley and Ray Lankester, their genetic con-
nection is so evident that it should be recognised by the establish-
ment of a Sub-phylum, to which we shall continue to apply the
name Pelmatozoa. The included classes, as hereafter explained,
are the Cystidea, Blastoidea, Crinoidea, and Edrioasteroidea. The
remaining classes of Echinoderma have been placed together by
P. H. Carpenter and others as Echinozoa, but may more con-
veniently be spoken of as Eleutherozoa (a term originally used by
Bell in a sense excluding Holothurians). Their genetic connec-
tion, however, is only that due to descent from the Pelmatozoa ;
even if all Eleutherozoa descended from one class of Pelmatozoa,
they did so at widely differing periods. The Holothurians must
have cast loose before the genital organs had been affected by radial
symmetry, and are thus, as well as by the horizontal position of
the oro-anal axis and the retention of the M plane as sagittal,
sharply distinguished from Echinoidea and Stelleroidea. The two
classes last mentioned were with some reason opposed by
Leuckart to Pelmatozoa and Holothurioidea (or Scytodermata,
as he called the latter) as Actinozoa ; but they differ in important
features. If Cuenot's interpretation of the lacunar systems be
correct, it seems as though the Echinoidea branched off before
radial symmetry had greatly affected the coelomic lacunar system
derived from the axial sinus ; similarly the digestive system
retained its coiled and non-radiate arrangement ; moreover, the
sinking of the ambulacral water-vessels and nerves below the test
here diverged further from the Pelmatozoic type than is the case
in Asteroidea. The extension of the ambulacra nearly to the

34 ECHINODERMA-GENERAL DESCRIPTION

aboral pole in Echinoidea and the development of a special
terminal plate at the end of each ray in Stelleroidea afford
features of much diagnostic value, but of less morphological im-
portance. English writers have usually regarded the Asteroidea
and Ophiuroidea as well-defined classes. The normal forms are
in fact markedly separate, but the evidence of ontogeny, as well
as the existence of connecting links now, and the approxima-
tion of the two groups in Palaeozoic time, renders this view
difficult of acceptance, so that they are here combined in a class
Stelleroidea.

Diagnosis of Echinoderma. — Metazoa, coelomata, triptoblastica,
living in salt or brackish water, with a primitive bilateral sym-
metry still manifest in the right and left divisions of the anterior
and posterior coelom ; with a hydrocoel primitively developed from
each half of the anterior coelom, and connected with the exterior
by a water-pore ; with stereom composed of crystalline carbonate
of lime deposited by special amoebocytes in the meshes of a
mesodermal reticulum or stroma, chiefly in the integument (absent
only in the highly modified Pelagothuria, p. 230, and, according to
Koehler, in the Holothurians Stichopus pallens and S. torvus) \ with
gonads derived from the endothelium, apparently of the anterior
coelom ; total segmentation of the ovum produces a coeloblastula
and gastrula by invagination ; mesenchyme is formed in the seg-
mentation cavity by migration of cells, chiefly from the hypoblast.

Known Echinoderma show the following features (imagined to
be due to an ancestral Pelmatozoic stage) : — Increase in the
coelomic cavities of the left side and atrophy of those on the right ;
the dextral coil of the gut, recognisable in all classes, though
often greatly obscured ; an incomplete secondary bilateralism
about the plane including the main axis and the water-pore or its
successor, the madreporite, often obscured by one or other of
various tertiary bilateralisms ; the development of the hydrocoel
into a circumoral, arcuate or ring canal, che hydrocircus ; except
in the small (but increasing) number of known cases in which care
of the brood has secondarily arisen, development is through a
free-swimming, bilaterally symmetrical, ciliated larva, of which
in many cases only a portion is transformed into the adult
Echinoderm.

All living, and most extinct, Echinoderms show the following
features (almost certainly due to an ancestral Pelmatozoic stage) :
— An incomplete radial symmetry, of which five is usually the
dominant number, is superimposed on the secondary bilateralism,
owing to the outgrowth from the peristome of one unpaired and two
paired ciliated grooves ; these have a floor of nervous epithelium,
and are accompanied by subjacent radial canals from the hydro-
circus, giving off lateral podia and thus forming ambulacra, and by

ECHINODERMA- GENERAL DESCRIPTION 35

a pseudhaemal system of canals apparently growing out from
coelomic cavities.

All living Echinoderms have a lacunar, haemal system of
diverse origin. This, the ambulacral system, and the coelomic
cavities contain a fluid holding albumen in solution and carrying
numerous amoebocytes, which are developed in special lymph-
glands and are capable of wandering through all tissues.

The Echinoderma may be divided into seven Classes, the
mutual relations of which are roughly represented in the annexed

table.

' CYSTIDEA EDRIOASTEROIDEA

PELMATOZOAK BLASTOIDBA

| — HOLOTHURIOIDEA ^

— STELLEROIDEA J

GUIDE TO LITERATURE OF ECHINODERMA GENERALLY.

In selecting from the large mass of writings, preference has been given to the
more recent, especially those giving summaries or bibliographies. Works alluded
to in the text may also be found by reference to the Zoological Record and the
Zoologisclicn Jahresberichten (Naples).

1. Agassiz, A. 1882. (Echinodermata . . . Bibliography.) Bull. Mus.

Cornp. Zool. Harvard, vol. x. pp. 109-134.

2. 1883. (Selections from Embryological Monographs — II. Echinoder-
mata.) Mem. Mus. Comp. Zool. Harvard, vol. ix. (2), pp. 1-45, pis. i.-xv.

3. Bell, F. J. 1891. (On the Arrangement and Interrelations of the Classes

of the Echinodermata.) Ann. Mag. Nat. Hist., ser. 6, vol. viii. pp. 206-
215.

4. Buetschli, 0. 1892. (Versuch der Ableitung des Echinoderms aus einer bi-

lateralen Urform.) Zeitsch. wiss. Zool., vol. liii. Suppl. pp. 136-159, pi. ix.

5. Bury, H. 1888. (The Early Stages in the Development of A'ntedon rosacea.)

Phil. Trans. Roy. Soc. London, vol. clxxix. (1888), B., pp. 257-300, pis.
xlii.-xlvii.

6. - - 1889. (Studies in the Embryology of the Echiuoderms. ) Quart.

Journ. Micr. Sci., N.S. vol. xxix. pp. 409-449, pis. xxxvii.-xxxix.

7. 1895. (The Metamorphosis of Echinoderms.) Quart. Journ. Micr.

Sci., N.S. vol. xxxviii. pp. 45-135, pis. iii.-ix.

8. Carpcider, P. H. 1878-85, -87. (Notes on Echiuoderm Morphology — i.)

Quart. Journ. Micr. Sci., N.S. vol. xviii. pp. 351-383 ; (ii.) vol. xix. pp.
176-206; (iii.) vol. xx. pp. 321-329; (iv.) vol. xxi. pp. 169-193, pis. xi.,
xii. ; (v.) vol. xxii. pp. 371-386 ; (vi.) vol. xxiii. pp. 597-616 ; (vii.) vol.
xxiv. pp. 1-23, pi. i. ; (viii.) vol. xxiv. pp. 319-327 ; (ix.) vol. xxv. Suppl.
pp. 139-155 ; (x.) vol. xxvii. pp. 379-391 ; (xi.) vol. xxviii. pp. 303-317.

9. Crety, C. 1894. (Contribuzione alia conoscenza dell' ovo ovarico.) Ric. Lab.

Anat. Roma, vol. iv. pp. 261-281, pi. xiv. [Contains Bibliography.]
10. Guenot, L. 1891. (Etudes morphologiques sur les Echinodermes.) Arch.
Biol., vol. xi. (1891), pp. 313-680, pis. .xxiv.-xxxi. [Gives good Biblio-
graphy.]

36 ECHINODERMA-GENERAL DESCRIPTION

11. Durham, H. E. 1891. (On Wandering Cells in Echinoderms, etc., more

especially with regard to Excretory Functions.) Quart. Journ. Micr. Sci.,
N.S. vol. xxxiii. pp. 81-121, pi. i.

12. Field, G. W. 1892. (The Larva of Asterias vulgaris.) Quart. Journ. Micr.

Sci., N.S. vol. xxxiv. pp. 105-128, pis. xiii.-xv.

18. 1895. (On the Morphology and Physiology of the Echinoderm

Spermatozoon.) Journ. Morph., vol. xi. pp. 235-270, pis. xv.-xvi.

14. Foettinger, A. 1880. (Sur 1'existence de 1'h^moglobine chez les Echino-

dermes.) Arch. Biol., vol. i. pp. 405-413, pi. xvii.

15. Geddes, P. 1880. (Observations sur Ic fluide periviscoral des oursins.)

Arch. Zool. Exper., vol. viii. pp. 483-496, pis. xxxvii.-xxxviii.

16. and Beddard, F. E. 1882. (On the Histology of the Pedicellariae and

the Muscles of Echinus sphaera, Forbes.) Trans. Roy. Soc. Edinb., vol.
xxx. pp. 383-395, pis. xix.-xxi.

17. ffamann, 0. 1883-89. (Beitriige zur Histologie der Echinodermen, Heft

1.) Zeitschr. wiss. Zool., vol. xxxix. pp. 145-190 and 309-333, pis. x.-xii.
and xx.-xxii. (Heft 2), pp. iv. and 126, pis. i.-vii., 8vo, Jena, 1885. (Heft
3), Jena. Zeitschr., vol. xxi. pp. 87-266, pis. vi.-xviii. (Heft 4), Jena.
Zeitschr., vol. xxiii. pp. 233-388, pis. xii.-xxiii.

18. Howell, W. H. 1886, (Note on the Presence of Haemoglobin in the

Echinodemis.) Stud. Biol. Lab. Johns Hopkins, vol. iii. pp. 289-291,
pi. xviii.

19. Lang, A. 1896. Text-book of Comparative Anatomy, translated by H. M.

and M. Bernard, Part II. pp. 284-560, 8vo, London. [Contains a good
classified Bibliography.]

20. Leuckart, C. G. F. R. 1848. Ueber die Morphologic und die Verwandt-

schaftsverhaltnisse der wirbellosen Thiere, etc., 8vo, pp. viii. and 180,
Braunschweig.

21 . Ludwig, H. 1880. (Ueber den primaren Steinkanal der Crinoiden nebst vergl. -

anatom. Bemerkungen ueber die Echinodermen ueberhaupt.) Zeitschr.
wiss. Zool., vol. xxxiv. pp. 310-332, pis. xii., xiii.

22. MacBride, E. W. 1896. (The Development of Asterina gibbosa.) Quart.

Journ. Micr. Soc., N.S. xxxviii. pp. 339-411, pis. xviii. -xxix.

23. MacMunn, C. A. 1885. (On the Chromatology of the Blood of some Inverte-

brates.) Quart. Journ. Micr. Sci., N.S. vol. xxv. pp. 461J-490, pis. xxxiii.,
xxxiv.

24. Metschnikoff, E. 1869. (Studien iiber die Entwickelung der Echinodermen und

Nemertinen.) Mem. Acad. Sci. St. Petersb., ser. 7, vol. xiv. No. 8, 73 pp.
12 pis.

25. 1885. (Ueber die Bildung der Wanderzellen bei Asterien und Echiniden.)

Zeitschr. wiss. Zool., vol. xlii. pp. 656-673, pis. xxv.-xxvi.

25a. Mortcnsen, T. 1898. (Die Echinodermen-larven der Plankton-Expedition
nebst einer systematischen Revision der bisher bekanuten Echinodermen-
larven.) Ergebnisse der Plankton-Expedition der Humboldt-Stiftung, vol.
ii. part J, 120 pp., 9 pis., and 1 map. Also published in Danish, as doctoral
thesis, 1897 ; and in Vidensk.Meddel. for 1898. [Gives agood Bibliography.]

26. Mailer, Joh. 1848. (Ueber die Larven und die Metamorphose der Ophiuren

und Seeigel.) Abh. k. Akad. Wiss. Berlin, Phys. Kl. 1846, pp. 273-312.
7pls.

27. 1849. (Ueb. L. und M. d. Echinodermen), op. tit. 1848, pp. 75-109,

5 pis.

ECHINODERMA— GENERAL DESCRIPTION 37

28. Miiller, Joh. 1850. (Ueb. L. und M. d. Holothurien und Asterien), op.

cit. 1849, pp. 35-72, 7 pis.

29. 1852. (Ueb. L. und M. d. Echinodermen), op. cit. 1850, pp. 37-86, 9 pis.

30. 1852. (Ueber die Ophiurenlarven des Adriatischen Mcercs), op. cit.

1851, pp. 33-62, 8 pis.

31. 1853. (Ueber den allgemeinen Plan in der Entwickelung der Echino-
dermen), op. cit. 1852, pp. 25-65, 8 pis.

32. 1854. (Ueber den Bau der Echinodermen), op. cit. 1853, pp. 123-219,

9 pis.

33. 1855. (Ueber die Gattungen der Seeigellarven), op. cit. 1854, pp. 1-55,

9 pis.

34. 8arasriny P. B. and C. F. 1887. (Die Augen und das Integument der Diadem-

atiden.) Ergebnisse naturw. Forsch. auf Ceylon, Bd. I. Heft 1, 4to,
Wiesbaden.

35. 1888. (Ueber die Anatomic der Echinothuriden und die Phylogenie

der Echinodermen), torn. cit. Heft 3.

36. Secliycr, 0. 1893. (Studien zur Entwicklungsgeschichte der Crinoiden.

Antcdon rosacca.) Zool. Jahrb., Abth. f. Morph., vol. vi. pp. 161-444, pis.
xii.-xx. [Gives Bibliography of Antcdon embryology.]

37. Semon, R. 1888. (Die Entwickelung der Synapta digitata und die

Stammesgeschichte der Ecliinoclermen.) Jena. Zeitschr., vol. xxii., N.F.
xv., pp. 175-309, pis. vi.-xii.

38. 1889. (Die Homologien inner halb des Echinodermenstammes.) Morph.

Jahrb., vol. xv. (2), pp. 253-307.

39. Sladcn, W. P. 1884. (On the Homologies of the Primary Larval Plates in

the test of Brachiate Echinoderms.) Quart. Jouru. Micro. Sci., N.S. vol.
xxiv. pp. 24-42, pi. i.

40. Thee,l,H. 1892. (On the Development of Echinocyamuspusillus[0. F.Miiller]).

Nova Acta R. Soc. Sci. Upsala, Ser. III. pp. 1-57, pis. i.-ix. [Summarises

previous writings. ]
41. 1894. (Notes on the Formation and Absorption of the Skeleton in the

Echinoderms.) Svensk. Vet. Akad. Fbrhandl., 1894 (8), pp. 345-354.
42. 1896. (Remarks on the Activity of Amoeboid Cells in the Echinoderms.)

Festskrift for Lilljeborg, pp. 49-58, pi. iii., Upsala.

See also Nos. 27, 64, 95, 96, in Literature of Pelmatozoa (p. 211) ; Nos. 25,
32, 41, in Literature of Stelleroidea (p. 279) ; and Nos. 36, 39, 62, in Literature
of Echinoidea (p. 328).

CHAPTER IX.

THE PELMATOZOA — CYSTIDEA.1

GRADE A. PELMATOZOA, LEUCKART (1848)
( = CRINOIDEA, sensu lato Auctt.).

CLASS I. CYSTIDEA.

„ II. BLASTOIDEA.

„ III. CRINOIDEA.

„ IV. EDRIOASTEROIDEA.

ECHINODERMA with the viscera enclosed in a calcified and plated
theca, of which the oral surface is uppermost, and which is usually
attached, either temporarily or permanently, by the aboral surface.
Food brought to the mouth by a subvective system of ciliated
grooves, radiating from the mouth either between the plates of the
theca (endothecal), or over the theca (epithecal), or along processes
from the theca (exothecal : arms, pinnules, etc.), or, in part, and
as a secondary development, below the theca (hypothecal). Anus
usually in the upper or oral half of the theca, and never aboral.
An aborally placed motor nerve-centre gives off branches to the
stroma connecting the various plates of the theca and of its
brachial, anal, and columnar extensions, and thus co-ordinates the
movements of the whole skeleton. The circumoesophageal water-
ring communicates indirectly with the exterior ; the podia, when
present, are respiratory, not locomotor, in function.

The origin and meaning of many of these characters have
already been discussed in the general section. The origin of
others will be traced in following the history of the Grade ; and
many of them will be more fully discussed under Crinoidea, in
which class alone are they adequately known.

The classes of Pelmatozoa here adopted are of very unequal

1 By F. A. Bather, M.A. Since the majority of Pelmatozoa, being of extinct
types, present peculiar difficulties, the student unfamiliar with Echinoderm structure
is recommended to begin either with the description of a simple Crinoid (Chapter XL),
or that of a Starfish (Chapter XIV.).

THE CYSTIDEA 39

value ; and to place them either in a line or side by side does not
represent their phylogenetic relations. Such, probably, would be
better shown by placing a primitive class, Amphoridea, at the
base and deducing from it several lines of descent, viz. Edrioas-
teroidea, Anomalocystida, Aporita, Rhombifera, and Diploporita.
From the Edrioasteroid line, we may suppose, there sprang first
Holothurians, then Stelleroidea, then Echinoidea, while the line
itself still survived in more specialised forms to the close of the
Carboniferous period. The Diploporite line ought properly to
include the Blastoidea ; and from it probably there arose, as a fresh
development with a new lease of life, the important class,
Crinoidea. The other lines were unsuccessful and none survived
the Silurian. But to make the classification coincide absolutely
with this history, which after all is not yet proven, would be to
reject names and classes that have held the field for more than
half a century in favour of new and unaccepted terms. Old
names, therefore, have been retained so far as possible. The
diversity of existing opinion, however, may serve as excuse for a
few novelties. Such are the use of Haeckel's Amphoridea, in an
emended sense ; the resuscitation of Edrioasteroidea ; the emenda-
tion of the Rhombifera, Aporita, and Diploporita, and of the
included families, which, when not new, are rarely used in the
sense of the original proposer ; the extension of the Blastoidea,
and their division into Proto- and Eu-blastoidea ; a considerable
revision of the accepted classification of Eublastoidea ; and a
recasting of the classification of Crinoidea.

CLASS I. CYSTIDEA, VON BUCK (1844).

Order 1. Amphoridea.
„ 2. Rhombifera.
„ 3. Aporita.
„ 4. Diploporita.

Pelmatozoa in which radial polymeric symmetry of the theca
is developed either not at all or not in complete correlation with
the radial symmetry of the ambulacra (such as obtains in Blastoidea
and Crinoidea) ; in which extensions of the food -grooves are
exothecal or epithecal or both combined, but neither endothecal
nor pierced by podia (as in Edrioasteroidea).

The earlier and more primitive Cystidea represent the pelma-
tozoic stage through which the Echinoderm race passed, on its
way from the Dipleurula to the various classes. They shed light
not only on the origin of those classes, but on the still more
ancient ancestor of the Phylum. The remarkable adaptability of
the Echinoderm type, the mode of origin of many organs, and the

40 THE CYSTJDEA

biological phenomena of homoplasy and convergence, can also be
studied in this class.

The Cystidea were first separated from other Echinoderms,
under that name, by L. v. Buch in 1844 and 1845. His definition
laid stress on the fixed condition, the irregularity of the thecal
plates, and the absence of arms like those of Crinoidea. Sub-
sequent discoveries of stemless cystids, of cystids with radial
symmetry in the theca, and of arm-like structures in most cystids,
have made the letter of this definition untenable ; but its spirit
holds good. The difficulty that this class has presented to
systematists is chiefly due to these factors : (1) The rarity and ill-
preservation of these old Palaeozoic fossils ; (2) the ancestral
nature of the group and the consequent existence of links between
it and other groups ; (3) the wrongful ascription to the Cystidea
of various genera (e.g. Porocrinus, Stephanocrinus, Hypocrimts,
Echinocystis) ; (4) the extraordinary diversity of structure in the
class, a feature common to most groups at their origin, and pro-
ductive in this case of many lines of development, only a few of
which have become so severed from the rest as to be regarded as
independent classes (e.g. Blastoidea and Crinoidea, distinguished
by all ; Edrioasteroidea, distinguished by a few ; Anomalocystidae,
not distinguished, but quite as separate) ; (5) the rapid develop-
ment of the class, from the exceedingly simple Aristocystis to such
highly specialised forms as Lepadocrinus, Caryocrinus, and Mesocystis.
Hence the diagnosis cannot be elaborate, and must be mainly
negative.

Most of the classifications hitherto proposed have been based
upon one set of characters; thus Zittel's (1879) adaptation from
Johannes Miiller (1854) is according to the structure of the thecal
plates (Aporitidae, Diploporitidae, Rhombiferi) ; Barrande's
division (1887), not intended as taxonomic, is according to the
number of openings in the theca. A far better arrangement is
that initiated by Pictet (1857), extended by Bronn (1859), and
modified by Dujardin and Hupe (1862); this, however, is rather
a key to genera than a classification into orders and families.
Attempts have also been made (e.g. Forbes, 184K ; and.Neumayr,
1889) to determine the lines leading from the Cystidea to other
classes; and on such principles Steinmann (1888) founded his
classification into Eucystoidea, Cystechinoidea, Cystasteroidea, and
Cystocrinoidea. A classification on true phylogenetic principles
was first published by Haeckel (1896), who only failed from want
of acquaintance with the facts of Cystid structure. The classifi-
cation in this text-book attempts to express the 'actual lines of
descent as inferred from an independent study of the fossils.

The main lines of descent are these. The starting-point is a
simple, many-plated, sac-like form (e.g. Aristocystis, Fig. II. p. 44),

THE CYSTWEA 41

in which neither ciliated food-grooves, though perhaps present,
nor radial ambulacral vessels, have left any trace on the skeleton,
in which the porous structure of the stereom is indefinite, and in
which no stem is differentiated. Modifications of this soon
appeared in many directions. In one direction arose an antero-
posterior flattening of the theca and the extension of food-grooves
along two lateral articulated spines, with a peculiar and character-
istic arrangement of stereom ; this was accompanied by develop-
ment of a stem (Anomalocystidae, pp. 49, 52, Figs. XI.-XIIL). In
another direction was an extension of the theca downwards to
form a stem, and upwards from the mouth to form a single jointed
process for the support of a ciliated groove (Dendrocystidae,
p. 47, Fig. IX.). Neither Anomalocystidae nor Dendrocystidae
proceeded very far, and they may conveniently be grouped with
Aristocystidae and a few other primitive forms into an order,
AMPHORIDEA, distinguished from the re^st chiefly by absence of
radial symmetry in food-grooves and ambulacra.

A very different modification was that which produced a theca
flattened horizontally, with five ciliated grooves passing from the
mouth between its plates (" endothecal1"), and protected by distinct
covering-plates ; ambulacral vessels lay beneath or within the
grooves, and podia from them passed between the adjacent thecal
plates. So different is this type from those of other Echinoderma,
that such forms have here been separated as a class, EDRIO-
ASTEROIDEA (Chapter XII.).

Returning to the primitive Amphoridea, we find a difficulty in
distinguishing some of them from their immediate descendants,
owing to the very slight traces left on the theca by the originating
extensions of the food-grooves. Those forms in which such traces
are perceptible may almost from the outset be grouped under two
heads. One group includes those in which the grooves wander
outwards from the mouth over the thecal plates, which gradually
become arranged regularly on either side of the grooves, while still
further extensions ascend from the " epithecal " grooves on small
"exothecal" processes called " brachioles." In the other group
the grooves do not tend so much to stretch over the theca as to be
raised away from it on relatively larger brachioles, arising in
the immediate neighbourhood of the mouth.

At the same time, a difference manifests itself in the structure
qf the thecal plates. From the indefinite relations of stereom and
stroma noticed in earlier Amphoridea arise two types of structure
(Fig. I.). The canals traversing the stereom, more or less per-
pendicularly to the thecal surface, either cease to be simple
(" haplopores ") and become connected in pairs (" diplopores ") still
perpendicular to the surface ; or they come to lie parallel to the sur-
face and at right angles to the sutures. In the latter case we may

42

THE CYSTIDEA

suppose that the canals represent stroma strands continuous
across the sutures ; those crossing any one suture come to occupy
a rhombic area bisected by the suture-line, and, since, in weathered
plates, there appear to be pores at the ends of these canals, the
areas have been called " pore-rhombs " (Poren-rauten, see Fig. XV.,
Echinosphaera, and XVI., Orocystis). There also takes place a gradual

3

Fio. L

Structure of the test in Aristocystidae. 1, plate of Aristocystis boJiemicvs showing haplopores,
some of which are connected by a horse-shoe canal, x4 ; 2, portion of same further enlarged ; 8,
portions of surface of other specimens, gradually leading up to such a structure as in 4, a plate
of Calix Sedgvricki, with diplopores ; 5, a diplopore of rather different form ; 6, section of a
diplopore ; the hypostereom is shown, but the euistereom, if there were any, is removed ; 7, a
natural replacement (by infiltration of mineral matter) of the original stroma-strands and
sutures in plates of Calix (?), the stereom dissolved away ; 8, plates of Aristocystis in similar
condition snowing vertical strands (= haplopores) in the middle, aud radiating strands at the
sutures. All figures enlarged. (1, 2, 3, and 8 after Barrande ; 4 and 7 after Rouault.)

increase in the area, and a decrease in the number, of the thecal
plates relative to the size of the theca; perhaps the folds that
often radiate from the urn bo of each plate are connected with this,
for they must have strengthened the plates, like the folds in cor-
rugated iron or pasteboard. These folds may coexist with diplo-
pores or with pore-rhombs ; but they are clearly more adapted to
the latter structure, and often seem to merge with it and accentuate

THE CYSTIDEA 43

it. Thus is evolved a highly specialised type of stereom-folding
known as a " pectini-rhomb." Now, although it is difficult to
separate all the forms at the first parting of the ways, it is soon
seen that diplopores are almost confined to the genera with
epithecal extensions of the subvective system, while those with only
exothecal extensions are characterised by pore-rhombs or pectini-
rhombs. There is therefore justification for the old divisions
DIPLOPORITA and RHOMBIFERA, as orders, in a restricted sense.

The Diploporita (p. 70) show a gradually increasing regularity
of structure in the food-grooves, and in their relations to the
theca, leading almost imperceptibly to the Blastoidea. So much
is this the case that it seems well to separate from the Cystidea
certain forms in which " the radial polymeric symmetry " is " in
complete correlation with the radial symmetry of the ambulacra "
(see definition, p. 39), and to refer them to the Blastoids as an
order Protoblastoidea (p. 79). The only alternative is to make
the Blastoids an order of the Cystidea.

In many of the Rhombifera (p. 52) a peculiar modification of
the food-grooves takes place, in that they are continued over the
theca, not directly on the thecal plates themselves, but by a
proliferation of plates from the mouth region. The grooves thus
formed have been termed " recumbent arms " or " pseudambulacra,"
and are fringed with brachioles. This type of ambulacral structure
was independently developed in this order more than once ; but it
is most common in the group of genera characterised by pectini-
rhombs and by peritamerism in the theca (family Glyptocystidae,
p. 58). A group with pore-rhombs highly developed inside the
theca, and with hexamerous symmetry, is distinguished as the
family Caryocrinidae (p. 65). In it the food-grooves tend to be
enclosed by thecal plates (" hypothecal ").

The orders already mentioned do not include all genera that
come under the terms of our definition of Cystidea. From early
forms of Rhombiferi, or perhaps even directly from Amphoridea,
there arose a small group in which neither diplopores nor pore-
rhombs were developed, at all events to the same extent, but the
number of thecal plates was greatly lessened and exothecal
brachioles were developed. The best known of these is Cryptocrinus
(p. 69). One might adopt APORITA (sens, sir.) as an ordinal name.

ORDER 1. Amphoridea, Haeckel (1896, pars).

Primitive Cystidea in which radial symmetry has affected
neither food-grooves, nor thecal plates, nor, probably, nerves,
ambulacral vessels, nor gonads.

Haeckel included under this name rather more forms than are
here referred to it, and separated them from the Cystidea as a

44

THE CYSTIDEA

primitive class of Echinoderma, comparable to the Pentactula
stage passed through in the development of all their descendants.
The more characteristic and undoubted Amphoridea, however,
represent only a stage in the development of the Echinoderm type
and not a divergence ; they are too intimately connected with
more specialised Cystidea to warrant separation as a class. It
should also be remembered that, though such a stage as this
probably was actually passed through, still forms are liable to be
referred here, owing to our ignorance of their true structure.

FAMILY!. ARISTOCYSTIDAE. Amphoridea without extension of food-
groovee, epithecally, cndothecally, or on exothecal skeletal processes.
Theca composed of numerous plates without regular arrangement or
specialised structure. No stem. Genera — Aristocystis, Barrande (1887),
Ordovician, Bohemia (Fig. II.), is in many respects the simplest Echino-
derm known. The ovoid
theca is composed of 150-
200 plates, of no definite
shape or arrangement, but
with a tendency in the nar-
rower, aboral half of the
theca to form transverse rows
of elongate hexagons. The
animal usually fixed itself to
some solid body by a portion
of the theca at or near the
lower pole (B). At the upper
pole is the mouth (0), a
wide slit in the transversal
plane, with slightly raised

Aristocystis bohemicm. 1, side view; 2, oral view, edges' About a third °f the

both i nat. size ; 3, basp, showing impression of Gastro- way down the theca is the

pod shell, x 3. The lettering is explained in the adjoin- . / A \ o o

ing text. (All adapted from Barrande.) round anal Opening (As), 6-8

mm. in diameter, closed by

six or seven triangular plates, meeting in the centre, and known as
"the valvular pyramid." Between mouth and anus, and usually a
little to the left, are two smaller openings — a transverse slit (M)
close to the mouth ; and a round pore (G) close to the anus. Of these M
is regarded by P. H. Carpenter (1891) as the hydropore, and G as the
gonopore, a view accepted by Haeckel (1896) and adopted here. There is
no trace of calcified arms or brachioles, whether jointed or solid, nor even
of epithecal or hypothecal extensions from the hydrocoel ring or from the
mouth. The hydroplwres palm&s described by Barrande, and supposed by
Neumayr (1889) and P. H. Carpenter (1891) to be subtegminal ambulacra,
are really epithecal food-grooves, and have not tteen proved to belong to
this genus. The plates of the theca are thick, especially at its lower end ;
they are said by Barrande to be composed of three layers (Fig. III.) :
(e) outer, thin, smooth, and solid ; (ra) middle, thick, pierced by irregular
canals, more or less at right angles to the outer surface ; (K) inner, thin,

B— J

FJO. 11.

THE CYSTIDEA

45

passing up into the suture lines and into the ends of the canals, smooth
on its inner surface. Doubt has been cast on the existence of these three
layers, but examination of well-preserved specimens suggests the follow-
ing interpretation : — (m) is homologous with the " mesostereom " of all
Echinoderms, its plates were deposited in a stroma of connective tissue,
and presented not only largv^ meshes, but continuous larger canals for
the passage of strands of stroma, and perhaps of lacunar blood-vessels :
the stroma was thus continuous throughout each plate, and strands often
passed over the outer surface, uniting the larger strands and sometimes
producing the grooves bent in horse-shoe, as at 2 in Fig. I. ; (e) is not

Section of test in Aristocystis, in the region
of u suture (s). Shows epistereom (e), meso-
Htereom (m), and hypostereom (h). m is pierced
by haplopores which are .joined above at 2, 2.
(Original diagram, magnified about 10 diani.)

st .-.:

Fio. IV.

Calix Sedgtricki, from a recon-
struction by Rouault ; about
nat. size.

A transverse section of a plate of Calix Murchisoni,
the stroma -strands replaced by infilling matrix, the
stereom dissolved away, x 0 diain. (After Rouault.)

a truly calcined epistereom, but probably represents a hard epidermis ;
there is little doubt that some structure did actually cover the outer ends
of the canals ; (h) represents the inner layer of the integument, which
towards the margins of each plate was often differentiated into elastic or
muscular strands, uniting adjacent plates and giving flexibility to the
theca ; if calcined, this layer would be homologous with the "hypo-
stereom" of many Crinoids, but, as in the case of (e), there is only
preserved to us the space which it occupied, filled with subsequent
infiltration or with iron oxide precipitated during the decomposition of
the organic matter. The primitive features of Aristocystis are then :
indefinite shape of theca, indefinite arrangement of plates, undifferentiated

46

THE CYSTIDEA

structure of stereom, absence of stem and of definite base of attachment,
absence of arms, absence of ambulacra, a single and independent gonopore
as in Holothurians. Ccdix, Rouault (1851-78 ; syn. Craterina, Barr. pars,
1887), Ordovician, France and Bohemia (Fig. IV.). The theca is a bowl
or vase of very thick plates, with an oral covering of very thin ones
rarely preserved. There is usually a marked hollow at the lower end,

FICJ. VI.

1, Cnlix bohe mica, from tlie side ; towards the base the pores are covered by a smooth epi-
dermis. 2, C'. excavata, the tegmen. Botli nat. size. (After Barrande.)

but 6'. Sedgwicki, Renault's type- species, has a short, stem-like promi-
nence (St). The canals in the plates tend to definiteness of arrangement,
especially at the lower end, where they seem to radiate from the hollow.
The connection of the canals at their outer ends, to form pairs, is often
marked (Fig. V.). All these characters are exaggerations of some already
noted in Aristocystis. The tenuity of the upper surface, and its conse-
quent disappearance in most specimens, have
permitted the recognition of only one aperture,
which is pentagonal, and probably represents the
anal pyramid (Fig. VI. 2). Specimens with hydro-
phores palmtes have been referred to this genus,
but belong to (p. 73) Diploporita. Pilocystis,
Lapillocystis, and Acantlwcystis, Cambrian, and
BaculocystiSj Ordovician, Bohemia, all described
by Barrande, are probably referable to this order
if not to this family. Lodamlla, Kayser (1885),
Lowest Devonian, Germany, though called a
sponge, is very like Calix. Deutocystis, Barr.
(1887), Ordovician, Bohemia (Fig. VII.), is dis-
tinguished from Aristocystis by the greater irregu-
larity in size of the thecal plates (comparable to
the arrangement of plates in the carapace of some

^_ ™™, restored extinct edentates, Glyptodontidae), and by the
on the evidence of Barrande, absence of an independent gonopore, this having
Fig.1!!."' 8' Lettering "8 in fused either with the anus or with the hydropore.
There are signs of fixation by the aboral end of
the theca (#), where the plates are larger and tend to lie in rows. The

Fio. VII.

THE CYSTIDEA 47

mouth (0) is surrounded by five large plates, forming a slight projection,
somewhat eccentric. The anal pyramid (As) has five plates. The hydro-
pore (M), a little to the right of the line joining mouth and anus, was
covered by a small pyramid of three plates, the impressions of which
on the internal cast (the only part preserved) have been regarded as three
openings. The canals in the thecal plates are more numerous near their
margins. Certain species, in which the plates seem more rounded, not
closely apposed, and perhaps without canals, in which the anal pyramid
had four plates, and in which the hydropore was not tripartite (i.e. had no
valvular plates), have been separated by Haeckel (1 896) as a genus, Amphora-
cystis. Piracy stis, Barrande (1887), Ordovician,
Bohemia (Fig. VI II.), had a pear-shaped theca trun-
cate below for fixation (St\ but still without true
stem. The anal pyramid had six plates (As) ; other-
wise it was much like Deutocystis. The regularity
of the adoral plates in these two genera suggests
that they may eventually prove to be early forms of
Diploporita or Rhombifera.

The Lower Niagara rocks (Silurian) of Indiana
and neighbouring states have yielded numerous forms
resembling Aristocystis in external appearance and
structure of theca, but with an ambulacral system fit —
apparently presenting three grades of organisation. FIG. vin.

They have all been described under the generic Pirocystis pirum, re-
name Holocystites or Holocystis, a name previously ev^dlnce^of^Barrande]
given to a coral, and therefore bound to yield to the g1.- 2J- Lettering as in
alternative Megacystis, Hall (1864-65). Some of the
so-called species described by Hall and S. A. Miller seem to agree with
Aristocystis in the entire absence of arms and food-grooves, in the similar
position and structure of mouth ("ambulacral orifice," S. A. M.), and anus
("mouth," S. A. M.), while a hydropore ("anus," S. A. M.) is often ob-
servable, and occasionally a fourth opening (? gonopore) ; the positions
of the two latter are at varying distances between mouth and anus.
Miller has described other species with similar structure, but with four
or five of the plates surrounding the mouth raised into .elliptical facets,
apparently for the support of spines like those of Placocystis (p. 51);
no groove connects these facets with the mouth, although in some species
the mouth assumes a tetragonal or pentagonal outline, with angles
directed towards these facets. The third and higher stage of organisa-
tion, possibly developed from this one, is seen in Holocystites gyrinus,
Miller and Gurley (1894), and must be referred to the Sphaeronidae
(seep. 72).

FAMILY 2. DENDROCYSTIDAE. Amphoridea with a single oral skeletal
process, theca composed of numerous irregular plates, extending below
gradually into a stem. The single genus, Dendrocystis, Barrande (1887),
Ordovician, Bohemia and Russia (Fig. IX.), has a theca in shape and
intimate structure not far removed from that of Aristocystidae ; of equal
thinness all over, its plates irregularly polygonal, and their strands of
rnesostroma not so well-defined. The following differences are of great

48

THE CYSTIDEA

importance : — The theca suddenly thins below to about one-third its
width, forming a tubular extension (&£), the walls of which contain
numerous small plates, which gradually become
larger and more definite in arrangement, and
merge into a tube with a narrow lumen en-
closed by comparatively large solid plates. Com-
parison with more highly developed genera
shows that this extension is a stem (columiia),
and its presence indicates a more fixed habit
than was, perhaps, assumed by Aristocystidae.
Further influence of fixation is clear in the de-
velopment at the opposite pole of the theca
of a movable, jointed tube (Br\ composed of
four or five rows of small plates, wider than
high, and often alternating ; this tube tapers to
a rounded end, in which no opening is per-
ceptible ; neither are there openings between
its plates ; the plates may, however, have
opened along one of the vertical lines, thus
converting the tube into a groove, exactly as
figured in Fig. 5 of Barrande's Plate XXVI. (see
Fig. IX.). This organ was regarded by Barrande
and Trautschold as a tubus ventralis for genital
(not faecal) products; by Neumayr (1889) as
an arm, with a double row of ambulacral pores ;
by Haeckel (1896) as an oral proboscis, or
possibly the stem. It is here regarded as an
extension from the mouth, bearing a ciliated
food-groove that could be closed by plates, and
perhaps also an extension from the water-ring.
Other thecal openings are doubtful ; an anal
pyramid may have existed in the lower third
of the theca (As), but Barrande's figures and
descriptions are inconsistent ; hydropore and
gonopore quite unknown. Folds or ridges
radiate from the centre to the edges of each
thecal plate ; besides strengthening the plates,
these folds, like similar ones in later stalked
forms, may indicate the concentration of a
nervous layer in the integument into definite

DendrocystiK Sedgivicki. re- , / • i \ ..• .1 Ai_ i

stored on the evidence of tracts (axial nerves) putting the stem, thecal

Barrande. Jir the arm-like an- plates, and plates of food - grooves into C011-
pendage, copied from Barrande. *

nection. Cujara, Barrande (1887), resembles

the stem of Dendrocystis, and suggests its occurrence in the Cambrian.
Syringocrinus paradoxus, E. Billings (1859), is the same thing from Quebec.
FAMILY 3. EOCYSTIDAE. Established to include certain obscure forms
from the Lower and Middle Cambrian of Great Britain and North America.
Eocystis, Billings (1868), and Protocystis, Hicks (1872, see also Salter. 1873),
have never been properly described or figured ; but since they cannot

FIG. IX.

THE CYSTIDEA

49

well be distinguished from Eocystites (?) longidactylus, Walcott (1886), that
species must be taken as the example of the family (Fig. X.). Thecal
plates numerous, irregular, " varying in form, size, and surface characters
on the same body." The two important points are : the varying develop-
ment of radiating stereom-folds on some of the plates ; the presence around
the mouth of not less than ten biserial brachioles, with long covering-
plates (" short pinnulae," Walcott). This type, therefore, is intermediate
between Amphoridea and Rhombifera, and its occurrence at so low a
horizon is fortunate for the phylogenist. This family Eocystidae in no
way corresponds to Haeckel's Eocystida, which, like his Eocystis> is a
purely imaginary creation of no systematic validity.

FAMILY 4. ANOMALOCYSTIDAE. Theca compressed in the plane of the

FIG. X.

Eocystis longulactitlus. 1, portion of test much enlarged, showing variation in size, outline,
and markings of plates, some of which have apparent pores at their edges ; 2, upper part of
a specimen without test, showing portions of brachioles and impressions of covering-plates.
With kind permission of Dr. C. D. Walcott.

thecal apertures, one side tending to be convex, the. other concave.
Plates of the two sides enclosed by a common frame of marginals. Plates
of concave side tend to be fewer and more regular than those of convex,
but never achieve bilateral symmetry as do the latter. Tapering stem of
polymeric columnals at one end of theca ; at the opposite end are the
apertures, with function still uncertain. In some genera, spines ("arms"
of most writers) are known, one at each upper angle of the theca. Orna-
ment of granules, which on the theca tend to run in transverse, wavy,
sub-parallel lines, simulating the scale-markings of some Crustacea. No
pores. J. Walther (1886) and Haeckel (1 896) have considered the bilateral
symmetry primitive, and homologous with that of the Dipleurula ;
but M. Neumayr (1889) maintained that the symmetry of the two was
different. The evidence suggests that the evolution was towards greater

THE CYSTIDEA

FIG. XI.
TrocJutcysti: boliemicus.

bilateral Asymmetry, and therefore started from the usual sack-like form.
Genera— Trochocystis, Barrande (1859-87 ; syn. Trigonocystis, Haeckel),
Cambrian, Bohemia, France, Spain (Fig. XI.), is
the most primitive ; its theca is bounded by
twelve stout maiginals (wra), conspicuous on one
side mere than on the other, and forming a
circular, elliptical, or subtriangular frame ; the
enclosed space on either side is occupied by a
mosaic of 80-160 somatic plates, hexagonal
except where truncated by the marginals. At
the oral end of the frame three openings, nearly
in the broad median plane, pass from the thecal
cavity through or between the marginals to the
exterior ; the middle opening (? hydropore and
gonopore) is widest, and is protected by a
hood-like projecting plate (M), (? madreporite) ;
of the other two openings, one (0) in the broad
plane is the wider (? mouth), the other (As)
slightly out of that plane is smaller (? anus) ;
they appear to be connected by a canal (? for
reception of gut) running round the thecal
^ cavity on the inside of the marginals. At the

8toredvonvithe evidence of Bar- aboral end of the frame the marginals pass
theca on'fhrrighfof^e^gure into a snort» tapering stem (Sf) of subtri-
is removed showing the interior, angular section, composed of rows (3 or 5 ?)
The arrows show the supposed 7° ,. . . , -,

direction of the gut. The letters of alternating ossicles, its lumen commum-

are^ explained in the adjoining cating with the thecal cavity. Trochocystis

may be regarded as a differentiation from

such a form as Aristocystis by lateral compression, so that its broad
median plane is morphologically the
sagittal, and the flat sicLs are the
primitive right and left, the pro-
jecting plate M being on the left side.
Mitrocystis, Barrande (1887), Cam-
brian and Ordovician (Fig. XII.),
has twelve marginals (mm), but on
the (left ?) side that corresponds with
the convex side of later forms, the
junction of the stem with the somatic
plates lies between two of them ;
while the median adoral plate of
this side (M) is vertically grooved on
the interior, but exteriorly resembles
the somatic plates, which on this
side, though larger and fewer (50-60) Fl°- XIL

than in Trochocysti,, still form a gld^^ "^ J- SFJEfftuSSK

mosaic of hexagons; tWO, adjoining of upper part of the latter, showing the folded

the stem, are larger than the others. plate"' (After Barrando->

On the other side, which corresponds with the concave side of later forms,

THE CYST1DEA

the marginals have extended far over the area formerly occupied by somatic
plates, and these latter number from three to six, one of which is much the
largest and of irregular shape. Openings other than the median not distinct.
Stem of about four alternating rows of plates, often provided with thorn-like
processes, and each overlapping its distal neighbour ; total length about
equal to that of the theca, the proximal third with a wide lumen. No
other appendages observed. Atelecystis (A. Huxleyi], Billings (1858), is
imperfectly known ; Anomtdocystis cornutus, Hall (1859), may be con-
generic, as usually supposed ; but A. disparilis, Hall, probably belongs to
Placocystis (vide iufrci). All the species described by Barrande (1887) as
Anomalocijstis are doubtful. The specimens described by Meek (1873)
and Wetherby (1879) are separable generically under the latter's name,
Enoplnum ; the species are E. lulanoides, Meek sp., and E. Crustacea,
Haeckel sp.1 In all these one traces the gradual diminution in number of
plates, especially of somatic plates, and the evolution of the granular
ornament into wavy ridges. Some of these also show traces of adoral
spines. Belemnocystis is placed in this family by Miller and Gurley (1894),
probably with justice, though its exact affinities are obscure. Platycystis,
S. A. Miller (1889), is based on a worn Anomalocystid of indeterminable
affinities. Placocystis, de Koninck (1869), from the English Wenlock beds
of Silurian age (Fig. XIII.), is the most specialised form of this family. It
was redescribed by H. Woodward
(1880); since then fresh know-
ledge has been gained. On either
side of the concave face (Fig.
XIII. 1) are three marginals
(mm) which pass over on to the
convex side ; at the columnal end
are two marginals, at the oral
end are three, and none of these
five continue on to the convex
side, although corresponding
plates occur there. The median
adoral marginal of the convex
side (Fig. XIII. 2) is the plate
M ; its free edge is occasionally
denticulate (cf. ridgings in Mitro-
tystis). The somatic plates of .

the concave side are one large (^reconstructed from various specimens in British
/•o^fral orirl rmn email of ita Museum ; 2, from Brit. Mas., E7545, enlarged ; 3,

central, and one small at its from the type-specimen, Brit. Mas., KH6&)
left upper corner : those of the

convex side are eleven, viz. two ad-columnal, as in Mitrocystis, supporting
one median which does not touch the column as it does in Enoploura ;
a transverse row of five, the median of which is small and often quite
surrounded by its two neighbours (it is not an anal structure, as supposed) ;
a row of three adjoining the adoral marginals. Stem much as in Mitro-
cystis. The three marginals that meet at each adoral angle of the theca
(Fig. XIII. 3) form an ?~ticular surface (Br) for the support of a spine

Fio. XIII.
Placocystis Forbesianus. 1, concave or supposed

1 This name must he restricted to Wetherby's Fig. 1, d, e, /, and 1g.

52 THE CYSTIDEA

(Br) which may attain 2/3 the length of the theca, without signs of a
joint. The spine is subcircular in section, and has no groove and no
accessory plates ; none the less it may have served as an arm, i.e. as the
bearer of a tentaculate extension of the water-system, and of a ciliated
path to the mouth. The mouth, anus, and hydropore were probably
situated in the integument uniting the two sides of the thecal opening
that seems to stretch between the spines. Haeckel (1896) supposes that
the Apiis-form of the Anomalocystidae was correlated with locomotion,
and that the stem was a locomotor organ — a suggestion by no means far-
fetched ; but his statement that the anus was ad-columnal must have
been based on Pleurocystis (p. 64), which, though belonging to a different
order, simulates Placocystis in many features.

ORDER 2. Rhombifera, Zittel (1879, emend.)

Cystidea in which radial symmetry affects the food-grooves,
and, in the more advanced families, the thecal plates ; probably also
the nerves and ambulacral vessels, but not the gonads. The food-
grooves are exothecal, i.e. are stretched out from the theca on jointed
skeletal processes (brachioles). These either are close to the mouth
or are removed from it upon a series of ambulacral or subambulacral
plates not derived immediately from thecal plates, or are separated
from the oral centre by hypothecal passages passing beneath teg-
minal plates. The stereom and stroma become arranged in folds and
strands at right angles to the sutures of the thecal plates ; in higher
forms the stereom-folds are in part specialised as pectini-rhombs.

The chief reason for the establishment of this order is the
recognition of a distinct line of development in the skeletal
structures bearing food-grooves. That this represents a true phylo-
genetic series is confirmed by the structure of the test, although
there may be some indefiniteness in this respect shown by the
earlier genera. The difficulty of classifying forms at the parting of
the ways is not one to be lessened by advance of knowledge. We
note also the gradual decrease in number of thecal plates, their
increasing subjection to a radial symmetry, and greater develop-
ment of a stem. Hence, as is to be expected in a natural classifica-
tion, there is a far greater difference apparent between extremes
in the same series, say, the Callocystinae and the Echinosphaeridae,
than between the latter and the Diploporite family Sphaeronidae,
which constitute initial forms of different series. The radial
symmetry or actinism of the order is trimerous, pentamerous, or
hexamerous, but may undergo secondary modification through
atrophy of a ray (e.g. Comarocystis, Lepadocrinus).

FAMILY 1. ECHINOSP^AERIDAE. Rhombifera in which the thecal plates
are numerous and indefinitely arranged. So-called pore-rhombs are de-
veloped, but no pectini-rhombs. Brachioles confined to neighbourhood
of mouth, unbranched. Stem, when present, not composed of a single series

THE CYSTIDEA

53

of columnals. This family includes some of the Cystidea earliest
studied. EchinospJiaera and Sphaeronis (p. 71) are found in the rocks
around the Baltic as round balls filled with radiating crystals of calcite.
Hence they were known to the older Scandinavian naturalists as "crystal
apples," arid were, as such, included by Linnaeus in his Mineral Kingdom,
under the name Aetites. T,heir animal nature was first demonstrated
by the youthful Gyllenhal (1772) in an admirable paper. He further
recognised, not merely their echinoderm affinities, but also that essential
difference between the tests of the two forms which was emphasised by
Joh. Miiller eighty-two years later, served as the basis of Zittel's classifica-
tion, and is still regarded as an ordinal character. So little were these
forms understood that Kb'nig (1825) placed Echinosphaera aurantium near
the Ascidian, Boltenia, under the name Leucopliihalmus Strangwaysi ; while
in 1845 M'Coy compared a Sphaeronis to the Ascidian, Chelyosoma. These

Br'

Echinosphaera aurantium, after Vol-
borth. The lettering is explained in the
adjoining text. Nat. size.

FIG. XV.

Peristomial areas of Echinosphaera
aurantium, showing variation in origin of
brachioles (Br). (After Volborth.) En-
larged.

comparisons, though based on superficial similarity, with bearing on the
supposed relationship between Echinoderma and Enteropneusta, have again
been brought forward by Haeckel (1896). Genera — Echinosphaera, Wahlen-
berg(1818 ; synn. Orystallocystis, Citrocystis, Trinemacystis, Haeckel), Ordo-
vician, Europe, type Echinus aurantium, Gyll. (Fig. XIV.). The smooth,
spheroidal theca is composed of some hundreds of irregular plates, mostly
hexagonal. At the aboral pole the plates, arranged in one or two fairly
regular circlets, form a slight projection (Sf), by which probably the theca
was attached, but there was no definite stem. At the oral pole is another
projection (0), very variable in size and shape ; this, as shown by Volborth
(1846), supports arms. From the figures published by Volborth (1846),
Miiller (1854), Quenstedt (1876), Angelin (1878), and Haeckel (1896), it
appears that the plates forming the oral projection, as well as the arms them-
selves, vary in number and position (Fig. XV.). The primitive number of
arms appears to be three, one anterior, i.e. opposite the anus, and two
lateral. The two lateral may fork, thus producing five branches in all

54

THE CYSTIDEA

(cf. p. 11). In either of these cases the anterior arm may diminish in size
and finally disappear, leaving either two lateral, or two antero-lateral + two
postero-lateral = four branches in all. The length of the arms and their
subsequent branching, if any, are unknown. The portions preserved are
formed of series of brachials, bearing on their adoral (upper) surfaces a
groove leading to the central mouth, and roofed by small covering-plates.
The anus (As), with its pyramid of four to ten plates, lies 1/3 or 1/4 of the
way down the theca. Between anus and mouth, to the right, is the hydro-
pore (M). The plates of the theca were united by strands of mesostroma as
in the next genus. Arachnocystis, Neumayr (1889), of Ordovician age,
has for type the Echinospliaerites infaustus of Barrande (1887). The
pear-shaped theca is composed of 200-800 plates of irregular shape and
arrangement, mostly small, but with a few larger plates interspersed. At
the narrower end of the theca is a stem, about 40 mm. long, composed
of five (?) alternating rows of hexagonal plates. At the opposite pole lies
the mouth, on a slight elevation of larger, irregular plates (not five orals
as sometimes stated). From these are given off three arms, composed of
two alternating rows of plates (biserial), and with a ventral groove roofed
by small covering-plates ; they may reach 100 mm. in length. About a
third of the way down the theca is the large anal opening, closed by a
pyramid of five plates. The hydropore has not been observed. The
structure of the thecal plates is clearly shown in the Bohemian fossils ;
between the thin non-porous epistereom and hypostereom lies the ineso-
stereom, penetrated by canals left by the strands of mesostroma that ran
at right angles across the sutures and united the plates. A trace of the
original path of these strands, as seen in Aristocystis, remains in the form

of canals passing down to the hypostereom,
one at the ad-central end of each transverse
canal, and one on either side the suture.
The centre of each plate is solid, and often
raised in an umbo ; by ideal lines drawn
from the umbo to the angles of each plate,
the canals are grouped in triangles, and
the adjacent triangles of two neighbouring
plates form a " pore-rhomb." Palaeocystis,
Billings (1858), Ordovician, Canada, has
thecal plates of similar structure. Orocystis,
Barrande (1887), Ordovician, Bohemia
(Fig. XVI.), has an oviform theca, with a
small hexagonal stem (St) of unknown
length, and near the other pole three
eccentric openings : 0 generally, and
probably with right, regarded as the
restored on mouth ; Ait the anus ; M the hydropore.
Th* thecal plates are marked with strong
axial folds, parallel with which are smaller
ridges, all at right angles to the sutures. The folds are probably the
superficial indications of axial nerves, and it is noteworthy that strongly
marked folds radiate from the six angles of the stem, and that six

" As

FIG. xvi.

Orocyatis

THE CYST1DEA

55

similar folds lead to the oral aperture (possibly conveying nerves to
six brachioles). The minor ridges are the superficial indications of strands
of mesostroma, uniting plate to plate across the sutures, and emerging on
the inner surface in rows of apparent pores. Heliocrinus, Eichw. (1840 ;
syn. Heliocystis, Haeck.), Ordovician, Europe, has for type Echinosphaerites
balticus, and therefore includes numerous species of similar structure,
usually referred to Caryocystis. It differs from Echinosphaera in the more
pronouncedly pentagonal, though minute stem, and in the greater orna-
mentation of the cup by axial folds and ridges. Stichocystis, Jaekel (1899),
has apparent pores in the ridges. Caryocystis was . founded by von
Buch (1844 and 1845) to receive Echinosphaera granatum, Wahlenberg
(Spliaeronites testudinarius, Hisinger, non Auctt.\ and another species which

von Buch and all subsequent authors have
incorrectly supposed to be S. testudinarius,
Hisinger. Eichwald (1859), aware that C.
granatum belonged to his own Heliocrinus, justly
took as the type the second species mentioned by
von Buch, adding to it C. pumila, an Echino-
encrinid. The type of the genus is therefore
the species universally and erroneously known as
C. testudinarius, which name yields to C. An-
yelini, Haeckel (Fig. XVII.). Amorphocystit,
Jaekel in Koken (1896), is a simple synonym.
Caryocystis differs from Heliocrinus in the elon-
gation of the oral and aboral poles, and the
elongation of the mouth in the sagittal plane.
At each end of the mouth-slit, Angelin's figure
(1878; our Fig. XVII.) shows two facets for
brachioles (Br) ; it also seems to show two
openings (hydropore, M, and gonopore .?, G)
between mouth and anus.

FAMILY 2. COMAROCYSTIDAE. Rhombifera in
which thecal plates are numerous and inde-
finitely arranged. Radial structure of stereom
strongly marked, but no definite pore-rhombs or
pectinirhombs. Food-grooves on free exothecal
brachioliferous processes. Columnals in a single
series. Genera — Comarocystis, Billings (1854),
Ordovician, Canada. Theca ovate, may be over
7 cm. high, composed of about 150 mostly
hexagonal plates, with strongly marked radial
striation of the stereom, especially towards the

-Br'

— M

margins, which are raised above the umbones dinarius, Auctt.). (After An-

p ii -I.. -»r .1 TJ. '±1 ~ 8e'in-)

r..st

Fio. XVII.

Caryocystis Angelini (C. testu-
s, Au

of the plates. Mouth -slit transverse, with a
pair of uniserial brachioliferous arms at either end. Theca flattened
in mouth-plane. Anus below arms on right side. Hydropore above level
of anus, near the posterior margin of the mouth. Stem longer than
theca ; columnals low, circular, with moderately wide lumen. Achradocystis,
Volborth (1870), Ordovician, Russia, appears to have an anal pyramid of

THE CYSTIDEA

seven plates, and a very highly developed stem. Thecal plates have
strong radiating ridges, marginal concentric ridges, and suture margins
toothed on the inside in correspondence with the ridges. Since this struc-
ture is not unlike that in Conutrocystis, the genus is provisionally placed hero.
FAMILY 3. MACROCYSTELLIDAE. Rliombifera in which the theca
consists of three or four circlets of plates, subjected to a more or less
regular pentamerism. The stereom is strongly radiately ridged or folded,
but no so-called "pores" or pectinirhombs are developed. Brachioles
are borne by the upper circlet of plates, and within these there may have
been tegminal plates over the mouth. There is perhaps no intimate
connection between the two genera referred to this family. But they
are both Cambrian, and show an early development of that tendency to
reduce the number of plates, which eventually evolved the Glypto-
^ystidae from a different branch of the Rhombiferi. The Macrocystellidai'
were probably derived from Eocystidae without passing through an
Echinosphaerid stage. Genera — Macrocystella, Calla-
way (1877), Upper Cambrian, Shropshire. Theca
seen from the side (Fig. XVIII. 1) shows four circlets
of plates, apparently five in each circlet. Those of
the aboral circlet are low and pentagonal ; those of
the second and third circlets hexagonal and rela-
2 tively large ; those of the fourth circlet about half
the size of those in the third, sub-pentagonal, and
each bears a brachiole. These almost immediately
bifurcate, making ten branches in all, about as long
as the theca is high, and apparently biserial; covering-
plates are distinct. There were probably tegminal
plates above the origins of the brachioles. Thecal
plates strongly marked with radiating folds, which
divide the surface into triangles ; between them are
smaller folds. ' No fine rhomb structure is seen.
Anus unknown. Stem rapidly tapering, about half
as long again as total length of crown ; proximal
columnals low and imbricating, with very wide
lumen ; distal columnals long and narrow. Mimo-
cystisj Barr. (1887), Ordovician, Bohemia, does not, so
far as can be gathered from the published description,
differ from Macrocystella in any essential. Lichenoides,
Barr. (1846, 1887; Pompeckj, 1896; syn. Licheno-

AlacrocysteUa Marine. , • TT ii\/-< i_ • T>V • JT>

i, from ' side (recon- cystis, Haeckel), Cambrian, Bohemia and Bavaria.

Mu8CtE7523frand E7?24)' ^hew composed of rounded plates of very different

x s;' 2-4, portion 'of a sizes, but semi-regular in arrangement. At the base

s™e,hdoreai, anc^Veitnii are nve to twelve minute plates, indicating absence of

surfaces ; 5, single plate stem and probably of fixation, at all events in adult.

of a large specimen (Brit. AU *,v • • i A. f^ • 11

Mus., E7.V2S), nnt. size. Above those is a circlet of five irregularly pentagonal
large plates. Resting on, and to a certain extent
alternating with these, are six or seven plates of similar size. A circlet
of smaller plates, alternating with the last mentioned, forms the summit,
and bears about eight biserial unbranched brachioles. No anus observed.

THE CYSTWEA

All plates except the minute ones at the base were united by strong
stroma-strands across their sutures, and these form pore-rhombs.

FAMILY 4. TIARACRINIDAE. Rhombifera in which the plates forming
the sides of the theca are arranged in not more than two circlets, and the
plates of each circlet are transversely united by strongly marked pore-
rhombs. The genera, though clearly separated as offshoots from the
Rhombifera, do not form a very coherent group, and need careful description
by a well-informed worker. Tiuracriiiux, Schultze (1867, Syn. titaurosoma,
Barrande, 1887), Devonian, Eifel, and Bohemia. Theca cup-shaped or a
truncate spheroid, the sides composed of four large, interradially situated
plates, one of which is (? only in some cases) horizontally bisected. These
plates are united by strongly marked stereom- folds, raised above the
surface, and forming demi-rhombs. The composition of the oral surface is
unknown, but there seems to have been a central mouth, with food-grooves
radiating from it towards the margin, where brachioles probably arose.
An anus seems to have pierced the margin at the summit of one of the
triangular side-plates. Stem- facet four-sided, with angles radial ; lumen
small, lihombifera, Barrande (1867 and 1887), Ordovician, Bohemia.
Theca elongate, triangular in section ; appears composed of two circlets — a
lower, of three plates united by strong stereom-folds, visible exteriorly only
as terminal pores outlining " pore-rhombs " ; an upper, of six (?) plates, of
which three pairs are united by pore-rhombs, similar to those of the
lower circlet, and vertically above, not alternating with them. Oral
region unknown. Aboral region passes gradually, by smaller plates, into
a cylindrical stem. The structure of the pore-rhombs and the trimerous
symmetry suggest comparison with Caryocrinidae. Rhomb ifeiu mira,
Barr., is usually considered to be a
Stephanocrinus. Aethocystis, S. A. Miller
(1892), Silurian, Indiana, may be placed
here provisionally.

FAMILY 5. MALOCYSTIDAE. Rhombi-
fera in which thecal plates are numerous
and indefinitely arranged. Radial folds
of stereom often pronounced, but minor
rhomb-like striae not clearly seen. Food-
grooves on exothecal processes passing
over the theca and bearing brachioles.
Columnals (when known) in a single
series. This Ordovician family shows
the independent evolution of a structure
common in a later family, Glyptocystidae,
viz. the extension from the mouth over

the theca of series of alternating plates, Amygdalocystis florealis. 1, from side;

supporting a single series of brachioles. ^^SSSS^SSfS^
The alternating series is not so com- ^^^^^£^ff^
plicated as in Glyptocystidae, and the belonging to Dr. G. J. Hinde.
food-groove passes, not on the top of it,

but along its sides. The main grooves appear in all cases to be reduced
to two ; but these may branch and wind round the theca as in the later

58 THE CYSTIDEA

form, Sphaerocystis (p. 63), or may remain simple, and stretch in the
transversal plane. Genera — Malocystis, Billings (1858). Chazy Lime-
stone, Canada. Theca globular. In the type-species the grooves branch.
Amygdalocystis, Billings (1854), Trenton Limestone, Canada. Theca
(Fig. XIX.) flattened in plane of food-grooves, and elongate. Grooves
never branched.

FAMILY 6. GLYPTOCYSTIDAE. Rhombifera with stem, theca, and
brachioles. The theca composed of five circlets of alternating plates,
typically five in each circlet. In first (aboral) circlet, right posterior
(r. post) plate is always fused with right antero-lateral (r. ant.) plate.
Anus, with valvular pyramid, between second and third circlets, in right
posterior interradius (r. post IR). Hydropore in adoral circlet, always
opposite unpaired arm-groove, and thus defining posterior interradius
(post. IR). The trans-sutural foldings of the stereom (pore-rhombs) are
restricted in distribution but exaggerated in structure (" pectinirhombs ").

Fio. XX.

explained
Glyptocystidae. The plates numbered as in Fig. XX

Actual distribution of pore-rhombs explained by supposed course of gut in primitive

One of them invariably unites the left posterior plate of the first aboral
row with the left anterior plate of the second row. Mouth central ; from
it over the theca radiate food-grooves, primitively 5, by reduction 4 or 2 ;
these are bordered by the plates of the adoral circlet, or by plates
derived from their proliferation ; and these side-plates bear facets for
brachioles of biserial structure. This family is perhaps descended from
primitive Echinosphaeridae, in which a natural tendency to the develop-
ment of five food-grooves (one ant, single ; two lateral, paired) was
accompanied by decrease in number, and increase in size and thickness,
of the thecal plates, together with their arrangement in five alternating
circlets of five (as in Mimocystis). The diminution of the thecal cavity,
we may suppose, pressed the coil of the gut against the body wall ;
thus the respiratory function was hindered in the pore-rhombs along this
tract, so that they disappeared, while it was thrown more on the remain-
ing pore-rhombs, which became highly developed (Fig. XX.). The process
continued till only three pair of intensely folded areas were left, one at
the base on the side opposite the anus, the others above the anus to

THE CYSTIDEA

59

the right and left of it ; from their likeness to combs, these areas are
called pectinated rhombs, more shortly " pectinirhombs."

We may reconstruct a form like that analysed in Fig. XXL as a type
(probably ancestral) from which every known genus of this family may
easily be derived. In this archetype two plates (r. post, and r. ant.)
of the aboral circlet are fused together, so that the total number is
twenty-four ; numbers are attached to these in the diagram, after a plan
originated by Forbes. The following statements may safely be made
concerning this archetype, those applying equally to the whole family
being italicised : — 3 is the double plate ; a pectinirhomb joins 1 <£• 5,
while other pectinirhombs were probably distributed as in Fig. XX. ; the
anus is in right posterior IB, between plates 7 and 8, and below 13 ; it is
protected by a valvular pyramid, surrounded by a riug of smaller plates ;
20-24 lie between food -grooves, and are therefore interradial ; the

Fio. XXI.

Arrangement of plates in the supposed archetype of the Glyptocystidae. The attached
numbers are those given by Forbes. The letters at the top, read from left to right, denote :
left anterior interradius & radius ; left posterior interradius & radius ; posterior interradius ;
right posterior interradius & radius ; right anterior interradius & radius ; anterior radius. Plate
23 be;irs the hydropore and gonopore ; the notches in plates 15 to 19 represent the primitive
position of five food-grooves ; between 7, 8, and 18 lies the anus. Pectinirhombs not shown.

hydropore, denning the posterior interradius, is in 23, and above it is
a semi-lunar pore of unknown function (genital or excretory) ; opposite
these is the unbranched anterior food-groove, passing to plate 15 ; the
branches of the right groove pass to 18 and 19, those of the left groove
to 16 and 17 ; at the end of each of the five main grooves is a facet,
bearing a brachiole composed of two alternating series of ossicles ("biserial");
the food-grooves of both brachioles and tegmen are protected by small covering-
plates ; the theca is borne on a stem, ivhich at its proximal end has low
columnals with a wide lumen, and which tapers distally. Modifications of
this type take place in the following directions : — Enlargement of the
ring of small plates around the anus, into a region of scaly, flexible
integument (Cheirocrinus, Olyptocystis, Pleurocystis) ; the accentuation of
the relations of the five main grooves to plates 15-19, coupled with
the sinking of those plates between those of the third series (Cheirocrinus,

6o

THE CYSTJDEA

Cystoblastus) ; the extension of the subvective system over the thecal
plates, by the proliferation from plates 20-24 of alternating series
of plates, in which every other plate bears a brachiole (Schizocysti*,
Glyptocystis, Lepadocystis, Callocystis, Sphaerocystis, Lepadocrinus, Pseudo-
crinus) ; the atrophy of the anterior groove (partially in Glyptocystis, more
so in Prunocystis and Schizocystis, wholly in Lepadocrinus, Sphaerocystis,
and Strobilocystis) ; atrophy of two side grooves in addition (Schizocystis,
Pseudocrinus, Pleurocystis) ; restriction of pectinirhombs to sutures
between 1 & 5, 14 & 15, and 12 & 18 (Prunocystis, Schizocystis, Pseudo-
crinus, Lepadocrinus, Callocystis). None of these characters can be taken
as a basis of classification ; each group so formed would include genera
very diverse in other respects ; doubtless the same structures have in
many cases been independently attained. Happily certain relationships
seern clear, and round them the genera may be gathered into sub-families.
SUB- FAMILY 1, ECHINOENCRININAE, passes from the simple form

FIG. XXII.

Echinoencriniis Sencken-
bergi, after Jaekel.

Fio. XXIII.

Analysis of Echinocncrinus, original, after
specimen E12C5 in Brit. Mus.

Fin. XXIV.

Prunocystis Fletcheri (fronr.
Brit. Mus., 40207). Sh,
pectinirhorab. x 4 cliam.

Echinoencrinus, in the direction of extension of two of the lateral grooves
over the theca, through Prunocystis to Schizocystis ; pectinirhombs 1 & 5 and
14 & 15 always, 12 & 18 frequently present, but no others, except 1 & 6
in Echinoencrinus. Genera — Echinoencrinus, H. v. Meyer (1826;synn.
Gonocrinus, Eichwald, 1840, and Sycocystis, von Buch, 1845 ; see also
Volborth, 1842), Ordovician, Russia, differs from the imagined archetype
in restriction of pectinirhombs to 1 & 5, 1 & 6, and 14 & 15, and
in apparent bisection of plate 23 (Fig. XXIIL). The main grooves
may support five brachioles, or only two, or may branch yet more ; in
any case the facets are always close around the mouth. Anal region often
prominent (Erinocystis, Jaekel, 1899). The plates usually have strong
radiating folds, often crossed by finer concentric ridges (Fig. XXII.).
Here Jaekel (1899) places his Glaphyrocystis and Scoliocystis, Ordovician,
Russia. Prunocystis, Forbes (1848), Silurian, England, includes P.
Fletcheri and Echinoencrinus baccatiw, Forbes (Fig. XXIV.). Theca
" shaped like the fruit of a dog-rose." Adoral row of plates increased
in number. Pectinirhombs on 1 & 5, 14 & 15, and 12 & 18 only.

THE CYSTIDE/i

61

Brachioles clustered round mouth, probably five, slender, and comparatively
long. Schizocystis, Jaekel (1895), Silurian, England. Type, Echinoencrinus
armatus, Forbes (Figs. XXV.-XXVIL). Arrangement of plates and rhombs
as in Prunocystis ; rhomb 12 & 18 sometimes absent. Ant., r. ant., and
1. post, food-grooves almost or entirely atrophied ; 1. ant. groove extends
nearly half-way down the theca on to plate 1 1 , is floored with alternating

Rh

Br'

& XXVI.

chizocystis armatus (from
. Mus., E7589 & E7590).
. XXV. from above;
/I. from the a banal «
. As, anus ; 7Jr', facets °l
brachioles ; fp, flooring
es of subvective groove ;
ig. XXVI. the impressions
hese are seen in the distal
part of the groove ; Rh, the
pectinirhomb of plates 14 &
15 ; 67, part of stem.

CD

Fio. XXVII.

XXVI.

Analysis of Hchizocystis. Plates 11 & 13
are notched by the food-grooves. Plate 23 is
double, and bears hydropore and gonopore.

plates, bears about five brachioles ; a similar extension of r. post,
groove to plate 13 is checked by anus, and bears only three brachioles.
SUB-FAMILY 2, CALLOCYSTINAE, starts from a form in which all five sub-
vective grooves, with the usual flooring plates and brachioles, pass over
the theca, as in Lepadocystis ; and diverges (a) by suppression of grooves,
without branching, into Lepadocrinus and Pseudocrinus ; (b) by branching
of grooves, with subsequent suppression, into Callocystis, Sphaerocystis, and
Strobilocystis ; pectinirhombs 1 & 5, 14 & 15, 12 & 18 probably always
present. Genera — Lepa-
docystis, P. H. Carpenter
(1891 ; syn. Meekocystis,
Jaekel, 1899), Ordovician,
Indiana (Fig. XXVIII.).
Theca ovoid ; plates around
anus slightly altered from
the archetypal position.
Pectinirhombs on 1 & 5,
12 & 18, 14 & 15, and
10 & 15. Main grooves
five, stretching a short dis-
tance over theca, one to
each plate of fourth circlet.
Lepadocrinus, Mather
(1843?; Conrad, 1840; Hall, 1859: synn. Apiocystis, Forbes, 1848;
Staurocystis, Haeckel, 1897 ; ? Hallicystis, Jaekel,

CD

FIG. XXVIII.

Analysis of Lejiadocystis Moorei, based on Meek's figures
(1871 & 1873). Plates 15 to 19 notched by food-grooves ;
hydropore on plate 23.

1899; includes

62

THE CYSTIDEA

Pseudocrinus quadrifasciatus and P. oblongus), Silurian, North America,
England, Scandinavia (Fig. XXX.). Plates much as in Lepadocystis.
Pectinirhombs onl&5, 12&18, 14&15 only. Main grooves four,
the anterior being aborted, extend over theca to a degree that differs
in different species, and is always less in younger stages. Pseudocrinus,

Pio. XXIX.

Analysis of Pseudocrinus, based on specimens in
the British Museum, especially 40193.

FIG. XXX.

Restored Lepadocrinus quadrifasdatm. The arm-
lets of the outer rows are erect ; those of the middle
row depressed. Near the top of the left-hand quarter
is the anus ; near the top of the right-hand quarter
is a pectinirhon.b. By permission of the Keeper of
the Geological Dept. , British Museum.

Pearce (1842, redescr. Forbes, 1848), Silurian, England (Fig. XXIX.).
Resembles a Lepadocrinus in which all main grooves are aborted except
apparently r. post, and 1. ant. The grooves are the same as those of
Sihizocystis, but their relation to the thecal plates and anus is as in
Lepadocrinus. The theca is compressed so as to have two almost flat

oval sides, fringed by the
brachioles of these grooves ;
one side contains the anus
and pectinirhomb 14 & 15,
the other contains pectini-
rhombs 1 & 5, and 12 & 18.
Callocystis, Hall (1852, syn.
Anthocystis, Haeckel), Silurian,
North America (Fig. XXXI.).
Plates 10, 12, and 14 are
sunk down between 6 and 7,
8 an-1 9 respectively, so as to
rest on 2 and 3 ; thus there
appear to be eight plates in
the second circlet. Con trari wise, by the vertical elongation of 7 and 8,
13 is raised between 18 and 19. The remaining plate of the third

<D

FIG. XXXI.
Analysis of Callocystis Jewetti based on Hall's ligures.

THE CYSTJDEA

circlet is reduced in size and sometimes quite atrophied. Pectinirhombs
on 1 & 5, 12 & 18, 14 & 15, as usual. Main grooves five, pass nearly to
base of theca ; one or more of the grooves may bifurcate once towards
the distal end. Sphaerocystis, Hall (1859), Silurian, Maryland. Theca
spheroidal. Main grooves four, each with three to six branches. Plates
undetermined ; pectinirhombs, anus, and hydropore situated as in Lepado-
crinus. Strobilocystis, White (1876), Devonian, Iowa, resembles Sphaero-
cystis, but the branches have become "small secondary arm -grooves
extending obliquely downward from each side of the principal grooves."
SUB-FAMILY 3. GLYPTOCYSTINAE. Besides the negative characters of
irregularity in the shape of the plates and in the distribution of pectini-
rhombs, the only feature common to the included genera is the evidence
they offer of descent from a simple form like CJieirocrinus. The number
of pectinirhombs connects Cheirocrimis with Cystoblastus and Glyptocystis ;
the enlargement of the anal area connects it with Glyptocystis and. Pleurocystis;
the relation of the third circlet of plates to the fourth is another link with
Cystoblastus. Genera — Cheirocrinus, Eichwald (1856 and 1859 ; see also

<£b

FIG. XXXII.
Analysis of Cheirocrinus penniger, modified from Fr. Schmidt.

Schmidt, 1874, under Glyptocystis), Ordovician, Russia, and North America.
The chief departures from the archetypal arrangement of plates are
correlated with an increase in size of the anal area, which is covered with
numerous small plates, and surrounded by plates 7, 8, 12, 13, and 14.
This in C. penniger, the type-species, induces the vertical fission of 18
(or intercalation of 18a), and sinking of 12 on to 2 ; the pushing of 15
to the right over 10, and sinking of 10 between 9 and 5 on to 4 ; the
pushing of 17 to the left over 11; the consequent sinking of 16 between
10 and 11 on to 5. In G. Volborthi the plates of the third row are not
sunk, but raised between those of the fourth row ; this is a change in
the direction of Cystoblastus. The distribution of the numerous pectini-
rhombs varies with the species, and, to some extent, with the individual ;
those on 1 & 5, 1 & 6, with the demi-rhombs 1 & 4, 1 & 2, are constant.
The arrangement in one specimen of G. penniger is shown in Fig. XXXII.
Reversions *o the older and simpler type of rhomb structure occa-
sionally occur (e.g. 13 & 18a in the figure). The five main grooves
pass between plates 20-24 to plates 15, 16, 17, 18a, and 19 ; they are
rather wide and fringed with short brachioles. The threefold division of

64

THE CYSTIDEA

plates 20-24 in C. penniger is the first stage in the proliferation of those
plates to floor the food-grooves ; more advanced stages occur in other
species, thus leading to Glypton/sti^ Homocystis, Barrande (1877), Ordo-
vician and Silurian, Bohemia, resembles Cheirocrinus in the shape of the
theca, the number of pectinirhombs, and the position of the anus. The
material does not permit more precise comparison. Glyptocystis, Billings
(1854 and 1858), Ordovician, Canada. Theca ovoid (Fig. XXXIIL). Anal

area less large than in Cheiro-
crinus, enclosed by 8, 13, and 14.
Plate 16 sunk on to 5, as in C.
penniger; thus 11 is pushed to
the right, and both 11 and 12
are reduced in size. Pectini-
rhombs 1 & 5, 1 & 6, 10 & 14,
10 & 15, 13 & 17, 15 & 16, 16
& 17, 17 & 18, 18 & 19; demi-
rhomb 10 & 16 ; imperfect
rhombs, 7 & 8, 11&12, 15 &
19. Main grooves five, passing
from over the theca, by way of plates
15-19 ; all nearly reach the base
except anterior groove, which is checked by pectinirhomb 10 & 15.
The resemblance of the grooves to those of Callocystinae suggests that
that sub-family was derived from Glyptocystis itself; but this is nega-
tived by the different modification of the thecal plates. Pleurocydis,
Billings (1854 and 1858), Ordovician, N. America and Britain (Fig.
XXXIV.). Anal area so large as to occupy almost all one side of the theca
with its numerous small plates, the anus being at lower right-hand
corner of the area, which is bordered by plates 3, 2, 7, 12, 13, 14, 8.
Correlated with this is a flattening of the theca in 1. post, and r.
ant. plane, the atrophy of all main grooves except the two in that
plane, great diminution in size of plates 15-19, as well as 13. Pectini-
rhombs 1 & 5, 11 & 12, 10 & 14; the two latter are analogous
to, not homogenetic with, pectinirhombs 12 & 18, 14 & 15, in
Callocystinae. Each main groove ends in a single long and sturdy
brachiole, of the usual biserial structure, with stout covering -plates.
The stem is longer than usual in the family. The curious homo-
plastic resemblance to the Anomalocystidae (p. 52) led Haeckel to
place Pleurocystis in that family, though it differs in every essential
structure. Cystoblastus, Volborth (1867 and 1870), Ordovician, Russia
(Fig. XXXV.). Imagine a Cheirocrinus in which plates 10, 11, 12,
and 14 are still further pushed up between those of the fourth circlet
than they are in C. Volborthi, in which plate 13 has entirely dis-
appeared so as to compensate for the asymmetry induced by the
anus ; in which the arm-grooves, of the usual structure, are stretched well
into plates 15-19 and limited to those plates, as in C. sculptus: then you
will have such a form as Cystoblastus. It need only be added that the
anus is surrounded by plates 8, 14, and 19 ; and that pectinirhombs
1 & 5, and 1 & 6 remain as before, while demi-rhoiubs unite the plates

THE CYSTIDEA

which now form the third circlet, viz. in order, 14, 15, 10, 16, 11, 17,
12, 18, 19. The superficial resemblance of Cystoblastus to certain
Blastoids has led most
writers to imagine a true
relationship. This involves
the entire disappearance
of plates 5-9 ; the homolo-
gising of the plates here
called 10, 11, 12, and 14,
as well as the absent 13,
with plates 20-24, and the
consequent disappearance
of plates 10-14 also ; the
violent supposition that
the horizontal transverse
or tangential folds of the
demi-rhombs in Cystoblastus
originated the radial or
vertical folds of the hydro-
spires in Codaster; as well
as such minor points as
the shifting of both anus
and hydropore, and the
fusion of two pair of basals,
neither of them the same
pair as composes plate 3 of
Cystoblastus.

The structures of the
subvective grooves in the
more highly specialised
genera of this family have
often been spoken of as
"recumbent arms." They
differ, however, from the
arms of Crinoidea in origin
as well as recumbency. From
the subvective structures of
Glyptosphaeridae and Pro-
toblastoidea, they differ
in that the brachioliferous

FIG. XXXIV.

Pleurocystis filitex-
tns. 1, analysis ; 2,
view from antanal
side, restored from
original observations ;
3, view from anal side,
after Jaekel.

plates are not thecal plates
or even intercalated be-
tween such plates, but lie
outside them and often
transgress their sutures.

FAMILY 7. CARYOCRIN- 2

JDAE. Rhombifera in which

the theca is composed primitively of four circlets of plates, comparable to
the infrabasal (IB), basal (B), radial (R), and deltoid (A) circlets of a crinoid

5

66

THE CYSTIDEA

with dicyclic base (p. 99), and dominated by trimerous symmetry. IBB 4,
the two on either side the anal inter-radius apparently being produced by
fusion of two original pairs. BB 6 (ten in Heterocystis). Alternating with
BB are 6 RR, between which, on the anterior side of the cup, two or three
interradials (iR) may be developed. (In Heterocystis the iRR also alter-
nate with the BB.) On the interior of all these plates the stereom is
thrown into strong folds, forming bundles of laminae at right angles to
the sutures ; on the exterior the ends of the folds are marked by pores,
each surrounded as a rule by a raised rim, and sometimes broken up into
two or more smaller pores. Since the laminae correspond in position and

Fio. XXXV.

Cyitoblastus Leuchtenbergi, after Volborth. 1, oral surface ; 2, posterior ; 8, aboral surface ;
4, analysis ; s.p, plates flooring arm-grooves ; 0, mouth ; As, anus, the position of which is
indicated by * in 1 ; M, hydropore between 18 & 19 ; St', attachment of stem.

essential structure to ordinary pore-rhombs, the pores necessarily run in
lines from the umbones to the angles of each plate (Fig. XXXVI. 3, 4) ; these
structures may have helped in respiration ; but are in no sense homo-
logous with Blastoid hydrospires. The three primitive food-grooves (ant.,
r. and 1.) are of equal size and bifurcate in like manner. The proximal
portion tends to be hypothecal ; then a short portion lies on the theca
like the " recumbent arms " of the Glyptocystidae ; the distal portion is
freely erothecal, biserial, and brachioliferous. Stem well developed,
circular in section. Genera — Hemicosmites, von Buch (1840, redescribed
1845 ; see also Joh. Miiller, 1854 ; syn. Hexalacystis, Haeckel), Lower to
Upper Ordovician, Russia, and perhaps Silurian, N. America. Com-
paring the dorsal cup (Fig. XXXVI. 3) with a crinoid cup, we may

THE CYSTIDEA 67

imagine the anterior radius to be doubled, concurrently with the
bifurcation of the anterior arm, and may therefore speak of the left and
right anterior RR and IBB and the anterior B, while the other plates
retain their usual names. The three iRR rest on the truncated upper
margins of the anterior and right and left antero-lateral BB, and are half
the width of the RR, but quite as high. The anus, with valvular
pyramid, is left of the posterior interradius, being in fact below L post.
R, and between 1. post, and post. BB. The tegrnen is solidly covered
with nine plates alternating with the nine plates of the radial circlet
(Fig. XXXVI. 1). These plates have no pores like those of the dorsal
cup (but the posterior one is rugose at its adoral end, and appears to have
been pierced by a hydropore). From the central mouth three food-
grooves run over these plates towards the anterior and the right and left
antero-lateral RR. The mouth and grooves are protected by relatively
large irregular covering-plates. At the end of each groove is an oval
area over which passed the base of a biserial arm, which became free almost
immediately. Caryocrinus, Say (1825, see von Buch, 1845, and Hall, 1852 ;
synn. Stribalocystis, S. A. Miller; Enneacystis, Haeckel), Upper Ordovician,
Scandinavia, and Silurian, N. America. Dorsal cup (Fig. XXXVI. 4) differs
from that of Hemicosmites only in absence of anterior iR, and of anus, the
latter having moved up to the tegmen. The covering-plates of the food-
grooves have become larger and incorporated in the tegmen (Fig. XXXVI. 2),
so that the grooves are subtegminal (cf. Crinoidea Camerata) ; but
end in facets on the margins of the RR and iRR. Since the grooves
branch while beneath the tegmen, these facets are more than three, though
still distributed into a left, right, and anterior group. In the growth of
the individual, Hall has noticed successive stages with from three to
fourteen facets ; six and nine are fairly common, but for adults thirteen
is the usual number, the thirteenth being added on the right of the anus.
Hall describes the arms as composed proximally of "semicircular, and
scarcely interlocking " ossicles ; distally the brachials alternate so that
the arm is biserial. Each brachial bears a grooved pinnule, perhaps also
biserial. In Corylocrinus and Juglandocrinus (both von Koenen, 1886),
Upper Ordovician, South France, the composition of the dorsal cup (Fig.
XXXVI. 6) is the same as in Caryocrinus. The tegmen of Corylocrinus
is composed of four plates, of which the right, left, and anterior bear arm
facets, apparently as in Hemicosmites ; the posterior tegminal plate is very
thick and porous (a madreporite). In Juglandocrinus (Fig. XXXVI. 5), as
in a young Caryocrinus, the food-grooves are subtegminal, and come to the
surface on the upper margins of three large plates, which are right, left,
and anterior in position, and alternate with 'three smaller plates. These
six plates exactly correspond to the tegmiual plates of Hemicosmites; a
central plate over the mouth, and three plates covering the food-grooves,
represent the covering-plates of that genus. This interpretation of von
Koenen's obscure genera was partly suggested by P. H. Carpenter (1891).
An anus was doubtless present, though not observed in the imperfect
specimens. Heterocystis, Hall (1852), Silurian, New York (Fig. XXXVI. 7),
appears to have been derived from Hemicosmites by the vertical bisection of
4 BB (viz. r.post., r. ant.-lat, ant., 1. ant.-lat.), thus producing 10 BB ;

Caryocrinidae. 1, teginen
of Jfemicosmites, x $ ; the
plates covering the mouth
are removed. (From Brit.
MUH. E7591.) 2, tegmen of
Caryocrinns, x2. (Modified
from P. H. Carpenter.) 3,
analysis of Uemicosmites.
(Original.) 4, analysis of Ca-
ryocrintis (based on various
figures). 5, tegmen, and (>,
analysis, of Juglnndocrinus
(based on von Koenen's
ligurns). 7, analysis of He-
tcrocystis (based on Hall's
figures).

The three primitive rays
are distinguished as ant., r.,
and I. The four circlets are
marked Hi, 1), R, & T (teg-
minals). Among the teg-
ininals, that marked M is
connected with the hydro-
pore. Plates intercalated
in R circlet are marked iR ;
Dr', facets for brachioles.

iB

a.i»t

THE CYSTIDEA 69

the corresponding broadening of 1. ant. B ; the sinking of the 6 RR,
and 3 iRR, so as to alternate with the 10 BB, except in posterior
IR, which is entirely occupied by anal plates ; the repeated bifurcation
of the arms, as in Caryocrinus, but probably to a greater extent, accom-
panied by an increase in number of tegminal and accessory plates, the
arrangement of which is unknown. Stereom-folds are visible. Thecal
plates nodose (cf. Hemicosmites), hence the name of H. armatus given to
the only specimen known.

Throughout this family there is no strict correlation between the arms
and the cup-plates. There is in both structures a dominance of the
number three or six, it is true ; but each arm has not its own radial plate
supporting it, as in Crinoidea. Indeed, an arm may be borne by a plate
that, on all other grounds, would be considered as interradial. Moreover,
the trend of evolution in the family is parallel to the probable evolution
of early Camerata, rather than towards that or any other Crinoid type.

ORDER 3. Aporita, Zittel (1879, restr.)

Cystidea in which pentamerous symmetry affects the food-
grooves and thecal plates, probably also the nerves and ambulacra!
vessels, but not the gonads. The food-grooves are exothecal and
circumoral. The stereom and stroma show no trace of folds,
rhombs, diplopores, or anything other than the finely porous
structure characteristic of all Echinoderm stereom.

One may regard this order as a backwater in the stream of pro-
gress, derived perhaps from the Khombiferi, but leading nowhere
in particular, and only retained because the forms referred to it
cannot be placed elsewhere. The arrangement of the thecal plates,
and a homoplastic resemblance to Hypocrinus (p. 178), have sug-
gested to some a connection with the Crinoidea.

There is only one family, the CRYPTOCRINIDAE, and in it the thecal
plates are arranged in four circlets. Cryptocrinus, von Buch (1840 and 1845),
Ordovician, Russia (Fig. XXXVII.). Theca small, irregularly spheroidal,
composed of four circlets of plates. Aboral Hrclet of three unequal plates,
produced by fusion of an original five, the ifused plate being in right
anterior interradius. Above these is a circlet of five rather large hexagonal
plates, following on and alternating with which are five smaller sub-
pentagonal plates. These surround an irregular pentagon in which are
the minute plates of the fourth circlet, and other small tegminal plates.
Five main food-grooves lead from the mouth to facets borne by these
adoral plates. The free brachioles rising from the facets must have been
slender. The anus, with valvular pyramid, lies between two plates of
the third circlet, either supported on a plate of the second circlet, or
separated therefrom by a small supplementary plate. The hydropore
appears to have been in the adoral plate opposite the anterior food-groove,
and left of the anus, which therefore occupies much the same position as
in Glyptocystidae. Another pore, perhaps excretory, lay in the adoral

70

THE CYSTIDEA

plate on the left, that is, opposite the anus. There was a slender stem
of circular columnals. Lysocystis, S. A. Miller (1889, proposed for Echino-
cystis, Hall non Wyville Thomson; syn. Scolocystis, Gregory, 1897),
Silurian, N. America. Theca sub-pentagonal, composed of four circlets of
plates. Aboral circlet of four (f) small plates, followed by two circlets of
five plates each, regularly alternating, and an adoral circlet, number un-
known. Apparently three free brachioles, possibly becoming five by the
usual bifurcation. Anus at adjacent upper angles of two plates of third
circlet. In the only known species the plates of second and third circlets
were strongly nodose. Stem unknown.

-Br'

FIG. XXXVII.

Cryptocrinus. 1, from oral surface ; 2, enlarged view of oral region, the tegminal plates
removed; 3, abpral view; 4, side view. (All diagrammatised from Jaekel.) As, anus; fir',
facets for brachioles ; M, hydropore ; p, another pore ; Stf , facet for stem.

ORDER 4. Diploporita, Zittel (1879, emend.)

Cystidea in which radial symmetry affects the food-grooves,
and by degrees the thecal plates connected therewith, but not the
interradial thecal plates ; probably also the nerves and ambulacral
vessels, but not the gonads. The food-grooves are epithecal, i.e.
are extended over the thecal plates themselves without intermedi-
ate flooring ; they are also prolonged on to exothecal brachioles,
which line the epithecal grooves. The stereom of the thecal plates
may be thrown into folds, but the mesostroma does not so much
tend to lie in strands traversing the sutures, nor are pectinirhombs
or " pore-rhombs " developed ; diplopores are always present in
the mesostereom, but often restricted to definite tracts or plates,
especially in higher forms.

While some descendants of the Aristocystidae were seeking in
vain to perpetuate their race by assuming the flattened carapace
of Anomalocystidae, and while others, becoming more reconciled
to a sedentary life, were stretching out from their mouths longer
and longer arms into the food-bearing sea, raising themselves too
on loftier columns, there were yet others that hit upon another
way of meeting the needs of a fixed existence. The chief need
was to expose food-collecting surface in greater amount and over a

THE CYSTIDEA 71

wider area. The way found was to stretch ciliated grooves from
the mouth over the surface of the theca, and then to raise them on
armlets or brachiola placed at fit intervals. At the same time,
the hydrocoel, it may be inferred from its constant connection
with such grooves, sent out branches corresponding with the food-
grooves ; from these branches were given off the podia, serving
for both respiration and the prehension of food. This wandering
of the brachiola away from the mouth marks then the develop-
ment of two structures previously unknown among Echinoderms :
(1) canals radiating from the hydrocoel along the theca ; (2) an
intrabrachial, circumoral, or ventral region of the theca, such as
in Crinoids is called the tegmen. These two structures, in one
form or another, characterise all Echinoderma other than Amphor-
idea and a few Rhombiferi. But it seems that they were inde-
pendently developed along many lines. What in the Crinoidea
is a mere "tegmen" or pot-lid, comes in Glyptosphaera, in the
Edrioasteroidea, and in the Echinoidea, to form the greater part of
the test ; and with it the perradial ambulacral vessels extend ;
whereas the theca of Aristocystidae, which in Crinoidea becomes the
specialised and important calyx or dorsal cup, is in the other orders
more and more reduced until in some cases no more of it can be re-
cognised than the plates of the anal pyramid or their homologues.
Another structure characterising this order, viz. diplopores (p. 41),
may possibly have assisted respiration by bringing lacunar blood-
vessels into closer contact with the sea- water ; this may have been
connected with a less development of podia. The functions of
diplopores have not as yet been satisfactorily explained by reference
to recent Echinoderms.

FAMILY 1. SPHAERONIDAE. Primitive Diploporita, in which the food-
grooves do not extend from the mouth beyond the adoral circlet of plates.
Diplopores diffuse. The included genera show the early stages of tegminal
and ambulacral development ; in none of them does more than a single
cycle of plates intervene between the oral pole and the bases of the
brachiola. The five plates forming the cycle are interradially placed,
being separated by the grooves proceeding from the mouth to the brachioles.
The direction of these grooves is primitively the same as that of the food-
grooves in Echinospliaera, viz. one anterior, opposed to the anus, two
lateral, each of which soon branches, thus making five grooves in all,
with bilateral symmetry. Thus the shape and position of the five inter-
vening plates are those characteristic of true orals (see p. 1 24). Genera —
Sphaeronis, Hisinger (1828 and 1837) ; the name was proposed to replace
Echinosphaera of Wahlenberg (1818) for no assigned reason, but was
restricted to forms agreeing with Echinus pomum, Gyllenhal, by Joh.
Miiller (1854). The species referred to this genus by Angelin (1878) are
all Ordovician, and agree in the following characters (Fig. XXXVIII.) : —
A spheroid or ovoid theca, sessile on a broad base, composed of irregu-
lar plates, the mesostereom of which is pierced by regularly formed

72

THE CYSTIDEA

A6

Fio. XXXVIII.

Sphaeronis globulus, after Angelin. 1, from side, nat.
size ; 2, tegmen, enlarged.

diplopores ; the anus close to the peristome, and with a valvular pyramid ;

small hydropore, perhaps combined with gonopore, between mouth and

anus, to the left ; five orals
separated by food - grooves
with primitive bilateral
arrangement. The various
species may be arranged
in groups, according to
the number of times the
grooves branch, and the
number of brachiola given
off from them. Haeckel
(1896) has sought to separ-
ate as genera (Pomonites,
Pomocystis, Pomosphaera}
those with one, two, three,

and four brachiola to each ray ; but until Angelin's notoriously

inaccurate figures shall have been corrected by observation instead

of by hypothesis, these names can rest on no sure ground. Moreover,

Love"n's figure of the type-
species, S. pomum, repro-
duced in our Fig. XXXIX.,

shows that the number of

branches visible may be

two, three, or four in a

single individual. Eu-

cystis, Angelin (1878), Or-

dovician, Sweden (Fig.

XL.), sends its grooves

farther down the theca

than Sphaeronis, over one

or two circlets of thecal

plates. From the distal

end of each ray a brachiole

was given off, while others

of uncertain number and

position arose along the

side of the grooves. Prob-
ably some of the forms

described by S. A. Miller Flo Xxxix.

as Holocystites should be Adoral region of ' Sphaeronis pomum (from Loven, "Om

placed here (e.g. TrematO- Leskia mirabilis," Ofv. Vct.-Akad. Forhandl. 1867, p. 434).

riistijt TflpVp^ nltlinnrrh °» orals covering mouth ; vg, section of food-groove running

Cl/StW, Jaefcei;, altnougn from mouj.h ^ B/, brachiole-facets, some of which are pierced

their orals are not known, by an axial canal ; As, plates closing over anus ; G, pro-

IT • TV/ril s n i minence with two pores, which Loven considered gono-

Jtl. gynnus, Miller & hurley pores; AT, ridge which Loven thought might indicate a

(1894), presents a Stage in '"^reporite (or 0 may represent combined gonopore and
hydropore) ; diplopores surround the whole area.

the development of food-
grooves that in some respects is more advanced, although the peristomial
plates have no regular arrangement (Fig. XLIL). Allocystis, Miller (1 889),

BT

THE CYSTIDEA

73

Silurian, Indiana, may go here. Proteocystis, Barrande (1887), Lower
Devonian, Bohemia (Fig. XLL), differs from Sphaeronis mainly in the ir-
regular branching of the food-grooves, which stretch farther over the theca,
though whether the tegmen ever contained more than one cycle of plates
cannot be determined from the published figures. Apparently the hydro-
pore formed a slit between the small gonopore and the mouth ; and the
base, broader than in Spliaeronis, is said to have been prolonged into a stem.

Flo. XL.

Eucystis raripitnctata, oral surface!
after Angelin, enlarged.

FIG. XLL

Protencyst is Jlami, oral sur-
face, slightly restored from
JJarrande, enlarged.

Fio. XI. 1 1.

Mouth and food-
grooves of " Holn-

after Miller' i& GUI -
ley.

Its geological age forbids us to regard Proteocystis as a link bc-tween Sphae-
ronitidae and Glyptosphaeridae, but it certainly has points of likeness to
the latter family. Carpocystis, Oehlert (1887), Lower Devonian, W. France,
is a simple spheroid with large stem-attachment. Palmacystis, Haeckel
(1896), Archegocystis, & Codiacystis, Jaekel (1899), are forms with the
epithecal branched food-grooVes, described by Barrande as hydroplwrcs
palmees of Piracy stis, Craterina, & Aristocystis.

FAMILY 2. GLYPTOSPHAERIDAE. Diploporita in which the food-grooves
extend over the theca well beyond the adoral circlet, and irregularly
transgress the sutures between the
thecal plates. Diplopores diffuse.
These represent a further advance
on the type of structure originated
in Sphaeronitidae. Genera —
Glyptospkaera, J. Miiller (1854),
Orclovician, Baltic countries (Fig.
XLIIL), has for type the species
first figured by the Duke of
Leuchtenberg (1843), and dis-
tinguished by Volborth (1846) as
Sphaeronites Leuchtenbergi. The
spherical theca, reaching a diameter
of 7 cm., and larger than any other
cystid, is composed of irregularly
arranged polygonal plates, bearing
diplopores. The mouth is covered
by five orals (0) with characteristic bilateral symmetry, and from between
them the anterior unpaired, and lateral paired, grooves radiate about

0 -.„

G -

FIG. XLIIL

Glyptosphaera Leuchtenbergi, after Volborth, nat.
size. For lettering, see adjoining text.

74

THE CYSTIDEA

,_ ,

half-way over the theca, crossing the sutures of the plates, and giving off
short branches on either side, or sometimes on one side only, at irregular
intervals. At the ends of the branches are facets (Br) for the support
of brachioles. The grooves are shallow and have minute covering-plates.
The anus (As), of which the valvular pyramid is rarely preserved, lies about
a third of the way down the theca ; between it and the mouth are two
openings, a little to the left, viz. a small round gonopore (G) and a madreporite
(M). The latter, the representative of the hydropore, is always close to
the first brachiole-facet of the left posterior groove, at the junction of three
plates ; it consists of folds (slits ?) running at right angles to the sutures,
and is bounded by a slight ridge forming a triangle or trapezoid. According
to Volborth, there was at the aboral pole a stem, —^ to £ width of theca,
with a wide lumen, and low columnals with five longitudinal sutures, and
with encrusting root-expansion. Eichwald, however, could not find more
than a short conical extension of the theca, and this agrees better with
the appearances of specimens sent by Volborth to the British Museum.
Funyocystis, Barrande (1887), Ordovician, Bohemia (Fig. XLIV.), differs

from Glyptosphaera in the broad base,
hollowed for attachment to some marine
object, and forming an angle with the long
axis of the theca ; the paucity of diplo-
pores ; and the regular alternation of the
groove branches, making each groove a zig-
zag. The test itself is not preserved ; the
internal casts show no sign of the external
grooves and brachioles ; but they show
between anus and mouth a curved eleva-
tion (madreporite t).

FAMILY 3. PROTOCRINIDAE. Diploporita
in which the food-grooves extend over the
theca almost to the aboral pole, and are
regularly bordered by alternating thecal
plates ("adambulacrals"), on which are
the brachiole-facets. Diplopores diffuse or
confined to adambulacrals, from which they
are never absent. These represent a further

advance in precisely the same direction as previous families. Genera —
Protocrimis, Eichwald (1840), Ordovician, Russia (Fig. XLV.), was well
described by Volborth (1846). Theca spheroidal or ovoid, attached by
a stem in the young, but free in old age and losing all traces of attach-
ment (cf. Lichenoides). Thecal plates larger, stouter, and more swollen
than in Glyptosphaera ; all bear diplopores, which may become somewhat
at right angles to main food-groove, on the adambulacrals. The main
grooves are rather straighter than in Glyptosphaera, lying regularly between
large alternating thecal plates (adambulacrals), each of which bears a
brachiole, except one or two of the proximal ones on the side towards the
direction of the clock-hands. Hydropore minute, above anus. Proteroblastiis,
Jaekel (1895; syn. Dactylocystis, 1899), Ordovician, Russia (Fig. XLVI.).
Theca ovoid, sometimes prolonged gradually into a stem (cf. Dendrocystis).

Barrande's figures ; enlarged.

THE CYSTIDEA

75

Thecal plates clearly differentiated into : (a) smooth, irregular, and
depressed interambulacrals ; (b) transversely elongate adambulacrals.
Diplopores at right angles to main food -groove, and confined to inner

Fio. XLV.

Protocrinus oviformis, after Volborth. 1, oral surface, showing food-grooves partly covered
by ambnlacrals, x jj ; 2, aboral surface of young individual, showing stein -attachment; 3,
aboral surface of old individual, without stem.

portions of adambulacrals. Each adambulacral bears a brachiole-facet ;
there are about thirty-six in each ray.

FAMILY 4. MESOCYSTIDAE. Diploporita in which the food- grooves
extend over the theca almost to the aboral pole, and are regularly
bordered by alternating brachioliferous adambulacrals, raised above and
outside the adjacent interambulacrals. Diplopores confined to1 inter-
ambulacrals. Five interradial deltoids (A) sur-
round the peristome. In this family we reach
the final stage of the Diploporita, although a
branch parallel with the Mesocystidae passes on
in the direction of the Eublastoidea, beyond the
boundary of the Cystidea. Genera — Mesocystis,
Bather (Jan. 1898, =Mesites, Hoffmann, 1866;
Nikitin, 1877 ; Agelacrinus, Schmidt, 1874),
Ordovician, Esthonia. Theca (Fig. XLVII. 1)
simulates that of a regular echinoid, or still more,
Edrioaster, since the mouth is on the upper
surface ; from it narrow food-grooves, protected
by covering -plates, pass straight down to the
margin of the flattened aboral surface (Fig.
XLVII. 2). Adambulacrals raised above the (Amb) bordered by adambu-
general surface of the theca, by the pushing in JS^er Sef"™18
under them of the adjacent interambulacrals ;

thus they outwardly resemble the sub-ambulacrals or side -plates of
Callocystinae (Fig. XLVII. 3). The interambulacrals do not, however,
absolutely meet underneath the adambulacrals, but leave an irregular

Fio. XLVI.

THE CYSTIDEA

canal (which may have contained a nerve of the aboral nerve -system).
Interambulacrals numerous and irregular, all pierced by diplopores.
Mouth lies in a depression (cf. Edrioaster) ; surrounded by five interradial,
slightly forked A, apparently continuous with the adambulacrals, and
homologous with the five plates similarly situated in Sphaeronis. Posterior

As

idAmb

Br

iAmb

Fio. XLVII.

Mesocystis Pusirtfskii. 1, general form, restored after Hoffmann and Nikitin. 2, oral sur-
face, after Hoffmann, enlarged ; the deltoids are not clearly shown. 3, structure of the adoral
end of a food-groove, modified from Jaekel ; the covering-plates removed from the lower right-
hand part ; much enlarged. 4, transverse section of a food -groove, after Jaekel. As, anus ;
adAmli, adambulacrals or side-plates (s.p) ; cp, covering-plates or ambulacrals ; Br, dotted
outlines of brachioles, borne by Br", brachiole-facet ; /</, food-groove ; iAmb, interambulacrals,
pierced by diplopores (p) ; M, supposed hydropores, probably only due to a boring parasite ;
0, mouth.

A pierced by hydropore (and ? gonopore). Anal pyramid in upper part of
posterior interambulacrum. Aboral surface of theca composed of numerous
small plates, but the structure of its central region is still unknown.

FAMILY ft. GOMPHOCYSTIDAE. Diploporita in which extension of food-
collecting surface is provided by the curving of the five main grooves
around the theca and not by their prolongation on to brachioles.

THE CYSTJDEA

77

Fio. XLVIII.

Gomphocystis tenax, from above
showing course of food-grooves,
and from side with outlines of
plates shown only in upper part.
After Hall, x J.

Gomphocystis, Hall (1864), Silurian, N. America
XLVIII.). Theca flattened above, greatly
elongate below, composed of many irregular
plates, pierced by diplopores. From a central
mouth five food -grooves radiate over the
theca, curving sinistrally around the upper
part, and occasionally descending a short dis-
tance on the stem-like base. Covering-plates
often strongly developed, and grooves lowered
beneath thecal surface. Jaekel (1895) states
that small side -grooves, but no brachioles,
occur in a Gotland species. Anus close to
mouth, in an interradius. Attachment ap-
pears to have been by the base, as in Aristo-
cystis. The curving of the food-grooves and the
asserted absence of brachioles cause Gompho-
cystis to resemble many Edrioasteroidea. But
the structure of the grooves seems to be that
which obtains in Diploporita, while the
presence of marked diplopores confirms the
reference to that order. In any case the family is out of the main line
of evolution.

APPENDIX TO CYSTIDEA.

The following names have been supposed to refer to Cystids : —
Ascocystis, Barr. , probably a Camerate Crinoid.
Balanocystis, Barr., indeterminable.
Cainarocrinus, Hall (syn. Lobolithus, Barr.), root of a Crinoid (Scyphocrinus,

apud Jaekel).

Cardiocystis, Barr., indeterminable.
Crinocystis, Hall, probably a Camerate Crinoid.
Cydocrinus, Eichwald (Pasccolus, Billings), now regarded as calcareous Algae ;

at any rate not Echinoderms.
Cystidea, a name used by Barrande for any indeterminable fragment, and not

intended as a zoological genus.
Dictyocrinus, Conrad, is a Receptaculite.
Hyponome, Loveii, the ejected viscera and disc of an Antedon.
Hypocrinus, Beyrich, an Inadunatc Crinoid (see p. 178).
Lichenocrinus, Hall, the root of a Pelmatozoan (see p. 133).
Mespilocystis, Barr., probably Stephanocrinus (see p. 96).
Neocystis, Barr., probably the root of a Pelmatozoan.
Porocrinus, an Inadimate Crinoid (see p. 172).
Rhombifera mira, Barr., is a Stephanocrinus (see p. 96).

A fairly complete Bibliography of the Cystidea was given in—
Barrande, «/., 1887. Systeme Silurien du centre de la Boheme, lre Partie :
Recherches Paleontologiques. Vol. VII. Classe des Echinodermes. Ordre
des Cystidees, 4to, xvii. and 232 pp., xxxix. pis. Prague.
The literature is so interwoven with that of Crinoidea and Blastoidea, that
other references are reserved for the list at the end of Pelmatozoa (p. 211).

CHAPTER X.

THE BLASTOIDEA.1

CLASS II. BLASTOIDEA, SAY (1825, sensu extenso).

GRADE A. Protoblastoidea.
„ B. Eublastoidea.

PELMATOZOA in which five (by atrophy four) epithecal ciliated
grooves, lying on a lancet-shaped plate (? always), radiate from a
central peristome between five interradial deltoid plates (A), and
are edged by alternating side-plates bearing brachioles, to which side-
branches pass from the grooves. Grooves and peristome protected
by small plates, which can open over the grooves. The generative
organs and coelom probably did not send extensions along the rays
into the brachioles ; but apparently nerves from the aboral centre,
after passing through the thecal plates, met in a circumoral ring,
from which branches passed into the plate under the main food-
grooves, and thence supplied the brachioles. The thecal plates,
however irregular in some species, always show defined basals (B)
and a distinct plate (" radial," E,) at the end of each ambulacrum ;
they are in all cases so far affected by pentamerous symmetry
that their sutures never cross the ambulacra.2

The more primitive of these forms can hardly be distinguished
from their immediate ancestors among the Cystids, such as Pro-
teroblastus and Mesocystis, except by the more developed basals and
radials; and it is this greater intimacy of correlation between
ambulacral and thecal structures that necessitates their removal
from the class Cystidea as here defined. Those general relations
of the ambulacra to the theca, shared by Blastoidea with Diplo-
porita, serve to distinguish them from the Callocystinae, with
which some of the genera have been allied by naturalists. To
these characters may be added the presence of diplopores, which
are still to be found in the most primitive genus. From the

1 By F. A. Bather, M.A.

2 The term "ambulacrum" has been loosely used in the Blastoidea for the
thecal elements connected with the food-groove. " Pseudambulacrum " is more
correct and more cumbrous. The relations of the true ambulacral system are doubtful.

THE BLASTOIDEA

79

Edrioasteroidea they are separated by the presence of brachioles and
absence of ambulacral pores. Their line of evolution, though in
some respects parallel to those of the Callocystinae and Edrioas-
teroidea, was independently derived through the Diploporita from
the primitive Amphoridea.

Within the class itself may be traced the increase of penta-
merism, combined with a lessening in number of the thecal plates
until a very definite arrangement is reached. At the same time,
there is a concentration of semi-porous structures into the inter-

Fio. I

Asterollastus. 1, oral surface of A. stellatus ; not quite correct, as anal area is not shown.
2, side view of A. Volborlhi. Both figures adapted from Schmidt, and x 2 diain. ami,, so-
called ambulacrum ; br, brachioles, supported on W, brachiole-facets ; f.g, food-groove ; pp,
pore-phite, at adoral end of which is deltoid ; R, radial, at end of ambulacrum ; tit, stem.

ambulacral region, and the evolution therefrom of elaborate
respiratory organs (hydrospires) ; this is probably connected with
the fact that all external connection with the water-ring (by means
of a madreporite or hydropore) seems to have disappeared in the
more specialised forms, and that, if the negative results of research
may be trusted, there were no extensions of the water-vascular
system into the brachioles.

A somewhat arbitrary line may be drawn below forms that
have acquired the normal definite number of plates and the
hydrospire- folds hanging far into the thecal cavity ; and while
such forms constitute a grade Eublastoidea or Blastoidea sensu
stricto, those below the line are Protoblastoidea.

GRADE A. Protoblastoidea, Bather (1899).

Blastoidea without interambulacral groups of hydrospire-folds
hanging into the thecal cavity.

8o THE BLASTOIDEA

FAMILY 1. ASTEROBLASTIDAE. Protoblastoidea with an indefinite
number of thecal plates, bearing diplopores, and with a pore-plate adjacent
to, but distinct from, the deltoid in each interanibulacrum. Genus —
Asteroblastus, Eichwald (1862), emend. Schmidt (1874), Ordovician, Russia
(Fig. I.). Theca pentagonal, on a relatively small round stem ; its hemi-
spherical dorsal portion is composed of 4 BB and 5 RR, between which are
from twenty-five (A. Volborthi) to fifty (A stellatus, type-sp.) polygonal
plates; all plates have radiating ribs ("axial folds"), between which are
diplopores. Round the mouth are 5 A, between which food-grooves
pass to the ambulacra. Each ambulacrum consists of two rows of side-
plates (adambulacrals), set alternately on either side of the food-groove,
branches from which run to right and left between adjacent side-plates.
Any underlying plate there may be is entirely covered by the side-plates.
At the ends of the grooves are sockets, by which uniserial brachioles were
attached .to the side-plates. Grooves down the brachioles joined the food-
grooves of the ray, and all probably were covered with small alternating
plates (ambulacrals). Adjoining each A and
separating the adjacent ambulacra, is a large
diplopore- bearing plate ; in A. tuberculatus
this is much like other plates of the theca
(on which ground Haeckel has separated the
species as Asterocystis, Fig. II.) ; but in A.
stellatus and A. Volborthi it is enlarged, has
a pronounced median crest, and is almost
separated by the ambulacra from the ordinary
thecal plates. [The anus appears to have

Portion of the oral surface of been at the distal end of this " pore plate "
:T£^eS^s±un1: in one °f ^ interambulacra, which is wider

ing the peristome ; s.p, side-plates than the Others.] FAMILY 2. BLASTOIDO-

CEINIDAE. Protoblastoidea with a definite
number of thecal plates, without diplopores,
and without distinct pore- plate. Genus — Blastoidocrinus, E. Billings
(1859), Canadian, and Schmidt (1874), Russian, Ordovician, is only
known from imperfect specimens, so that only the following details can
be given (Fig. III.) : — The small round stem is inserted into a consider
ably invaginated base, as is the case in several crinoids. BB, at least 3.
RR, 5, not notched for reception of ambulacra, but with truncate upper
margin ; they support 5 large interradially situate A. Each A consists of
two parts : an orad, subtriangular piece, like that called deltoid in Astero-
blastus, and a lower, triangular piece corresponding to the pore-plate in
Asteroblastus ; the connection between the two parts is still slight ; indeed,
in Canada the orad portions are often fused with one another into a pen-
tagonal frame. The lower portions show a marked striation, due to stroma
strands, at right angles to the radio-deltoid suture, and on the sutural
edge the stereom is thrown into folds (incipient " hydrospires ") which,
however, have not been observed to pass into the RR. The food-grooves
pass out between the A ; the lancet-plat* supporting them is covered by
alternating side-plates, which receive branches from the main groove, and
are provided with a clear brachiole-facet ; the remains of brachioles are

THE BLASTOIDEA

81

preserved, as in Asteroblastus and many Eublastoids. These known facts
afford no character, other than the less development of the hydrospires,
by which Blastoidocrinus should be separated from the Eublastoidea, so

br

sp

br-

Fio. ill.

Blastoidocrinus carchariaeilens ; restored on the evidence of E. Billings. 1, oral surface ; 3
-ays show the side-plates (sp) with their facets for brachioles (br1) ; 1 ray shows the brachioles
(br) attached ; the 5th ray shows lancet-plate (L) exposed by removal of side-plates. 2, from
the side ; the upward extension of the stem (St) and the invagiuated basals (B) are indicated by
dotted lines, h, incipient hydrospires.

long as that order includes such a form as Cadaster. But till the structure
of the base and the position of the anus are known, the genus may be
kept with its rather more primitive allies, Asteroblastus and Asterocystis.

GRADE B. Eublastoidea, Bather (1899).
( = BLASTOIDEA, A-uctt.).

Blastoidea in which the thecal plates have assumed a definite
number and position in three circlets, as follows : — 3 BB, 2 large
(formed by fusion of two pairs of the primitive 5 BB) and 1 small,
in r. ant. IR ; 5 RR, often fork-shaped, forming a closed circlet ;
5 A, interradial in position, supported on the shoulders or the pro-
cesses of the RR, and often surrounding the peristome with their
oral ends. The stereom of the RR and A on either side of the
ambulacra is thrown into folds running across the radio-deltoid
suture ; these folds hang down into the thecal cavity, forming the
hydrospires.

Beginning in the Silurian with Cadaster and Troostocrinus, we
may trace the gradual modification of a simple type,, and the
evolution of the numerous complicated structures characteristic of
so specialised a form as Pentremites. Starting afresh with Elae-
acrinus we shall study the elaboration of the structures that
characterise Orbitremites ( = Granatocrinus), which represents the
acme of the Eublastoidea in Britain. Thus the morphology and
the classification will be unfolded along with the phylogeny.

6

82

THE BLASTOIDEA

Codaster, M'Coy (1849), Silurian to Carboniferous, Britain and N.
America, is the least specialised of all Eublastoids (Fig. V.). The 3 BB
form a conical cup, on which follows the more cylindrical circlet of 5
RR, the processes of which bend in
above almost at right angles, to form
part of the truncated summit or oral
surface. Following on the upper mar-
gins of the radial processes come the 5 A,
which surround the pentagonal mouth-
opening (Fig. V. 1). These A probably
represent the combined A and pore-
plates of Asteroblastus ; like those pore-
plates they have a pronounced median
crest (" oral ridge," Etheridge & Car- Radiala i -m m?/m
penter). Between adjacent A, and \lf_mr

passing down into the sinus of each Basals
R, is a long plate, known from its
shape as the " lancet-plate " (Fig. V. 2) ;
the edges of this partly overlap the R
and A ; its upper surface bears a groove
which passes between the adjacent A
to the mouth. It is pierced by a cen-
tral canal, comparable to the central

hollow seen in Mesocystis, and probably

filled by a nerve from the aboral

system (Fig. V. 6). Along the sides

of the lancet-plate lie small "side-
plates," not, however, in single series,

but in zigzag, so that each pair forms

a rhombohedron bisected by the suture

between them (Fig. VI.). On these

sutures are brachiole - facets ; and

branches from the ventral groove pass

alternately to the side-plates, and on

to the brachioles. The whole groove

was covered by small movable plates,

the impressions of which are seen along

its sides. The peristome was similarly

plated over. The apposed edges of

the radial processes and the A are

thrown into a set of strongly-marked

folds at right angles to the radio-
deltoid sutures (Fig. V. 6). These

folds in the stereom may be an ex-
aggeration of the axial folds SO con-

flnirmmq in Asternhln<itiix • tliA istprpmr. mis> from Kinderliook beds of Iowa. By
n ASteroblastllS , the Stereom permission of the Keeper of the Geological
forms a thin -folded wall, the ends Department of the British Museum.

of the folds dipping far down into

the thecal cavity. One infers from similar foldings in other animals

Brachioles.

Theca.

-Stem.

Root.

Fio. IV.

Reconstruction of a typical Eublastoid,

THE BLASTOIDEA

that the object was an increase of surface for respiratory purposes, the
outer oxygenated sea-water passing down into the folds and the inner
coelomic fluid passing up into the alternate folds. Hence to such struc-
tures E. Billings gave the name "hydrospires." They are similar, in
essential structure and in position across the suture lines, to the pectini-
rhombs of Callocystinae ; and similarly are a development of the normal

FIG. V.

Cadaster trilobatus. 1, oral surface of young individual of var. acutus, with 3 hydrospires
each side of a food-groove and traces of them in anal interradius (Brit. Mus. E8020). 2, a theca
ground down from the oral surface, thus bringing out the sutures (Brit. Mus. E8023). 3,
analysis of the main thecal elements. 4, theca seen from the left posterior radius. 5, the
same theca from below, with posterior interradius uppermost. (Both from Brit. Mus. E8013).
x$. 6, slightly restored section across part of a radius, amb, so-called ambulacrum or
pseudambulacrum, with food-groove ; As, aperture for anus ; B, basal ; br, brachiole ; c.p,
covering-plates ; h, hydrospire-folds ; L, lancet-plate ; 0, mouth-aperture ; R, radial ; s.p, side-
plate ; A, deltoid.

structure of the test. The marked pentamerous symmetry of the thecal
plates (with the apparent exception of BB) and of the hydrospires is
disturbed only by the anus, which makes an opening between the pos-
terior A and the adjacent radial processes. From this IE. hydrospire-folds
are said to be absent (but one or two may be seen in some specimens,
Fig. V. 2). The anal opening was closed by small plates. The essential
structures of Cadaster are all to be found in Protoblastoidea ; the absence
of interambulacrals, in Blastoidocrinus ; the ambulacral structures, in

THE BLASTO1DEA

Asteroblastus and Blastoidocrinus ; the position of the anus, in Asteroblastus;
the hydrospires, though fur less developed, in Blastoidocrinus. So much
is this the case that Cadaster has been referred to the Cystidca by several
writers of eminence. Further arguments for such action are found in the
fact that Cadaster has no " spiracles " at the proximal ends of the ambu-
lacra, and no hydrospire-pores along their sides,
structures which are often considered character-
istic of the Eublastoidea. The gradual evolution
of these structures may, however, be traced
within the order.

Phaenoschisma, Etheridge & Carpenter (1882,
-86), Devonian and Carboniferous, Europe and N.
America, differs from Codaster mainly in the fact
that the hydrospire-folds become more concen-
trated, and pushed in under, or overgrown by,
the side-plates and part of the lancet-plate, so
that as a rule only their ends are visible (Fig.
VII. 2) ; they are also well developed in the
anal 1R. The side-plates may lie at the sides
of the lancet-plate, as in Codaster, or they may
lie on it ; in cases where the proximal side-plates
project far over the hydrospires, they may
entirely roof in the depression in which the
hydrospires lie, only leaving a small communi-
cation with the exterior on either side the
proximal end of the ambulacrum (Fig. VII. 6).
The openings thus formed are the rudiments
of " spiracles." The side-plates that in Codaster
formed the outer halves of the rhombohedral
pairs are here diminished in size and pushed out-
wards (Fig. VII. 1). They are now distinguished

as " outer side-plates " (cf. similar structures in the recumbent arms of
Callocystinae). Cryptoschisma, Etheridge & Carpenter (1886), Lower
Devonian, Spain, differs from Phaenoschisma and Codaster in little but the
greater breadth of the ambulacra, which entirely conceal the eight hydro-
spire-slits on either side. Sometimes the ambulacra do not reach to the
tops of the deltoid crests (or oral ridges), and thus there is a spiracle on
either side of each crest, or ten in all ; but sometimes they reach right up
the oral ridges, so that the spiracles of adjacent ambulacra become con-
fluent, and there are five in all (Fig. VII. 8). This is expressed by saying
spiracles " double " or " single " respectively. Orophocrinus, von Seebach
(1864 ; synn. Dimorphicrinus, d'Orb. ; Codonites, Meek & Worth), Carboni-
ferous, Britain, Belgium, and N. America (Fig. IV.). The hydrospire-slits
are more concentrated at the bottom of the depression for the ambulacra
(" radial sinus ") than in previous genera ; but since the ambulacra are
narrower, some of the slits may be exposed, and the spiracles are merely
clefts, often extending all along the sides of the ambulacra (Fig. VII. 7).
The concentration of the hydrospire-slits causes the inner walls of the
two nearest the median line of the ambulacrum to meet along that line,

FIG. VI.

Food -groove and associated
structures of Codaster triloba-
tus, greatly enlarged, o.s.p,
outer side-plate; s.p, side-
plate ; br, brachiole - facet,
from which passes a small
groove to/.<7, the main food-
groove, along which are im-
pressions of covering-plates.
These structures are borne
by L, lancet-plate, in which
is n.c, a possible nerve-canal.
Beneath this emerges h.p,
the plate forming the side
of h, the first hydrospi re-fold.
(Based on Brit. Mus. ES027.)

THE BLASTOIDEA

and so to form a new structure known as the "under lancet-plate"
(" sub-lancet "). In this genus the oral surface is raised into more or less
of a dome, so that the A are visible in side view. Post. A encloses the anus.
Pentremitidea, d'Orbigny (1849, emend. Eth. & Carp., 1882,-86), Dev-
onian, Spain, the Eife], and N. America. In the form of the calyx, the
complexity of the hydrospires, and other points there is great variability.
P. Paillettei, de Vern., however, type of the genus, is not far removed from
Phaenoschisma and Cryptoschisma. Eight hydrospire-folds lie on each side
of an ambulacrum, entirely covered by the broad lancet-plate. The
side-plates are on top of the lancet-plate, so that none of it is visible.
The outer side-plates are wedged in between the side-plates and do not

Fro. VII.

1, Phac.noschisma Verneuili, food -groove seen from above. lp, lancet-plate ; sp, side-plates,
between which are the small outer side-plates. 2, section across part of a radius of the same ;
the hydrospires are here in the deltoid, and only a process of the radial is seen (R.pr). 3, part
of an ambulacrum of Pentremites, seen from above, x 6 diam. 4, left side of the same further
enlarged, showing some of the covering-plates (diagrammatised from Steinmann). 5, section
across a radius of Pentremites, x 6 diam. C-8 are diagrams showing the evolution of spiracles :
6, in Phaenoschtima ; ends of hydrospire-folds still visible, only an incipient- spiracle. 7, in
Orophocrinns ; a long spiracle-slit is enclosed between side-plates and deltoid ridge. 8, in Crypto-
schisma; on the left the spiracles are distinctly separated by the deltoid ridge ; on the right
the side-platos of adjacent ambulacra meet and the deltoids sink, so that the spiracles appear
single, br, brachiole ; br', articular facet for same ; c.p, covering-plates ; A, deltoid crest ;
f.g, food-groove ; L, lancet-plate ; 0, peristome ; o.s.p, outer side-plate ; p, pores ; R, radial ;
s, spiracle ; s.l, sub-lancet ; s.p, side-plate.

project beyond their edges. A covered by radial processes, so that only
the crests are exposed. The spiracles are clefts on either side the
ambulacra, between the deltoid crests and the proximal side-plates. The
anus pierces post. A, which therefore has no crest, so that the adjacent
spiracles are confluent with the anus ; the opening so formed is called " the
anal spiracle." In some species (esp. P. angulata) the side-plates do not
cover the lancet-plate any more than they do in some species of Phaeno-
schisma. Within the limits of Pentremitidea (as defined by Etheridge &
Carpenter) two changes of importance take place. The hydrospires no
longer remain extended, with each fold opening into the space below the

86 THE BLASTOIDEA

overspreading ambulacrum, but the admedian folds become gradually
lowered into the thecal cavity and open into a common canal (e.g. P.
clavata) ; thus each system of folds forms a " pendent" hydrospire-sac.
The second change is the slight notching or bevelling of the ends of the
side-plates and the projection of the outer side-plates so as to touch the
wall of the radial sinus ; thus is formed along each side of the ambulacrum
a series of openings by which water is admitted to the hydrospires, which
it bathes and then passes out again by the spiracles (e.g. P. lusitanica) ;
these openings are called " pores," but are not comparable to the haplo-
pores or diplopores of the Cystidea, to the water-pores of Crinoidea, or to
the ambulacral pores of Echinoidea. Pentremites, Say (1820, originally
Pentremite), Carboniferous, N. America (Fig. XV. 1), appears to be a
descendant of the American species of Pcntremitidea. The chief difference
is that the side-plates do not cover the lancet-plate, but rest against its
edge ; thus the ambulacrum becomes much broader and assumes a petaloid
shape (Fig. VII. 3, 4). A sub-lancet is developed, as in Orophocrinus ;
hydrospire-folds 3 to 9 according to the species, freely pendent (Fig. VII. 5),
as in some Pentremitidea ; but in some species the floor of the radial sinus
meets below the hydrospires at the distal end of the ambulacrum. A
canal runs between the hydrospires and the side-plates, emerging through
spiracles which may be single or double ; there is an anal spiracle, as in
Pentremitidea. Though limited in geographical and geological distribution,
Pentremites has more species and individuals than has any other Blastoid
genus; for this reason, and because it was the first of its class to be
introduced to science, it has usually been treated as the type of the Blas-
toidea. Zoologically considered, however, it merely represents the acme
of one particular line which thereafter died out.

We return, therefore, to the Silurian, and take up the beginning of
another line of development, apparently connected in origin with that
which has just been traced and not very divergent therefrom.

Troostocrinus, Shumard (1866, emend. Eth. & Carp., 1886), Silurian,
N. America, might be expected from its geological position to be a primi-
tive form ; and that it is such is shown by the structure of the ambulacra
(Fig. VIII.). The chief differentiation from the Codaster type lies in the
elevation of the radial processes ; the restriction of all A, except post. A,
to a very small truncate summit, so that the hydrospire-folds are almost
entirely formed out of the radial stereorn ; the narrowing of the radial
sinus, so that (as in some species of Phaenoschisma) the side-plates are
pushed up on to the top of the lancet-plate, while the hydrospire-folds
are pushed beneath it The outer side-plates are small, subtriangular,
squeezed out to the edge of the side-plates, with which they alternate ;
they touch the wall of the radial sinus so that "pores" are formed
between them. The hydrospire-folds are midway between the Codasterid
type and the pendent type ; their admedian walls, which support the
lancet-plate, are thickened, and tend to form " hydrospire plates," covering
the immediately adjacent folds. The canal that runs along above the
hydrospires and below the lancet and outer side -plates comes to the
surface at the oral end through a spiracle bounded by the A and lancet-
plate, and since the deltoid crests are slight, the spiracles are almost

THE BLASTOIDEA

single. The anus opens through, the posterior spiracle, i.e. through the
oral end of post. A ; the position is less primitive than in Cadaster, but the
greater size of post. A seems primitive. Metablastus, Eth. & Carp. (1886),
is represented by doubtful species in the Silurian of N. America, less
doubtful from Devonian of Europe, and undoubted from Lower Carboni-
ferous of N. America. It closely resembles Troostocrinus in form ; but
the A are all equal, small, and confined to the summit, while the two

Fio. VIII.

Troostocrinus Reinwardti. 1, section across a radius, much enlarged. 2, upper part of
theca, posterior view. 3, oral surface of theca. Lettering as before. As+sp, anus and con-
fluent spiracles. (2 and 3 slightly altered from Etheridge & Carpenter.)

posterior spiracles are distinct from the anus. In structure the ambulacra
scarcely differ from those of Troostocrinus; but the lancet-plate of
M. lineatus is said to have three longitudinal canals •.*, the meaning
of which is not obvious (cf. Schizoblastus). The base is elongate, and
triangular in section, with flattened sides. Tricoelocrinus, Meek & Worthen
(1868), Carboniferous, N. America, and (?) Queensland (Fig. XY. 2), owes
its name to three excavations along the interbasal sutures, corresponding to
the three flattened sides of Metablastus. A feature of more importance is
the enclosure of the distal portion of the hydrospires within the thickness
of the radial plate ; this is probably due to the upward growth of the
floor of the sinus, as has taken place, independently and to a less extent,
in Pentremites. Eleutherocrinus, Shumard & Yandell (1856), Devonian,
N. America, is probably a descendant of Troostocrinus. It has, however,
lost its stem and adopted some mode of life that has affected its symmetry.
The place of the stem is occupied by the small r. ant. B. The two other
BB are pushed a little to the side and produced up the theca to meet the
broadened and shortened 1. post. R. The remaining RR, with their ambu-
lacra, are much like those of Troostocrinus and Metablastus; this 1. post.
R, however, bears a short ambulacrum with only seven side -plates on
either side, but these greatly widened transversely to the median groove, and
curved distalwards (Fig. IX.). There are small A, and (apud Shumard)
spiracles at their oral ends on either side of each normal ambulacrum.
The anus is stated by Wachsmuth (in Whiteaves, 1889) to lie on the
right upper margin of the abnormal R, and therefore at the aboral end
of the A. Sections across the ambulacra (Eth. & Carp., 1886) show a
lancet-plate with large longitudinal canal, and seven hydrospire- folds,
arranged as in Troostocrinus, on either side of each ambulacrum in its

88

THE BLASTOIDEA

upper part ; at the aboral end of the ambulacrum the floor of the radial

sinus comes below the lancet-plate (cf. Tricoelocrinus).

Nucleocrimts, Conrad (1842,
synn. Elaeacrinus, Roemer, 1851;
and Olivanitet, Troost, MS.,
1849), Devonian, North America,
differs much from the genera
hitherto described (Fig. X.). Al-
though the theca is lofty (often
olive - shaped) the BB and RR
reach a very short way up it ;
the BB and lower part of RR
are pressed inwards, and the
radial sinus receives only the
distal extremity of the ambula-
crum. The greater part of each
ambulacrum lies between certain
int?rradial plates. The posterior
interradius contains three such
plates — a median plate, at the
oral end of which lies the anus,
and a plate on either side. These
lateral pieces meet at their oral

Eleutlwrocrinus Cassedayi ; oral view x 0 diam.
The abnormal, 1. post, radius turned to the
observer; r. post, ray has covering-plates (c.p)
which in life would have covered over the oral
centre; in r. ant. ray the c.p are removed
and one sees the whole of the side-plates (s.p) ;
these again are removed from ant. ray, exposing

the lancet-plate (/,) ; removal of 7, in 1. ant. ray
shows the hydrospires (A.s). (Based on Shumard
& Yandell, and Whiteaves.)

ends, enclosing the anus, and
stretch down between the central
piece and the ambulacra. The

central interradial is smooth, with a median vertical ridge ; but the lateral
pieces are transversely grooved, each groove corresponding to a pore

Fio. X.

Nucleocrinus VtrneMill. 1, from anterior radius, x J. 2, oral view, show-ins large cover-
ing-plates over the peristome (from Brit. Mus. 120), x |. 3, from posterior interradius, x j}.
At, anus ; cp, covering-plates, which were continuous over the food-groove (/</) ; IK, iiiterradiii)
pieces ; R, radials ; gp, spiracles ; A, deltoids, the limbs of which flank the ambulacra.

in the ambulacrum. As regards the structure of the other inter-
ambulacra there is a difference of opinion. Each contains a lanceo-

THE BLASTOIDEA

late central area, with a median ridge, separated from the ambulacra by
transversely grooved areas meeting adorally. Some specimens show
appearances of sutures between these areas, especially at the aboral end.
Hence Lyon (1857), from the study of many hundred specimens, and
Billings (1870) concluded that these interambulacra also contained three
plates, as a rule fused together. Koemer and Etheridge & Carpenter,
however, believed that there was only one plate, and that the markings
were merely ornament. The am-
bulacra are narrow ; their struc- u.r...^&

ture is shown in Figs. X. 2 and ^=JL>™W

XL They dip down to the mouth

underneath a roof of strong plates.

The hydrospires are pendent, their

folds reduced to two ; they emerge

through large spiracles, separated

by the interambulacral plates.

As for the homologies of the

thecal plates, those who believe

that there is only one in each

interambulacrum regard that one

as the A, which in the posterior

interradius is split in two by an Flo XI

intercalated anal plate. This gection of ambulacrum O'f yllfleocrinV8 Ver_

makes Nudeocrinus a highly neuili, x 10 diam. b.r, brachiole; /,, lancet-
SDCciiilisGcl foi'iii into wliicli is k ' *i *• i * * i * *"~i *^^ »

suddenly introduced an element iuterainbulacwb.
found in no other Eublastoid. If,

however, we accept the view that each interambulacrum has essentially
the same composition, viz. three plates, we are able to institute compari-
sons with Protoblastoidea. These suggest that the true homologues of the
A in Nudeocrinus are the proximal portions only of the plates called deltoid
by Etheridge & Carpenter. This latter explanation of the structure of
Nudeocrinus permits us to regard it as primitive in all except tne hydro-
spires, and consists better with its geological age. SdiizoUastus, Eth. & Carp.
(1882-86), Carboniferous, Britain and N. America, i^ay be described
as a Nudeocrinus in which there is only one plate in each of the inter-
ambulacra ; this therefore may be called a deltoid, but may well represent
the three plates some suppose to exist in Nudeocrinus, since it preserves
their peculiar sculpturing. In some species it is of much less relative
size. The hydrospires, as in Nudeocrinus, are of simple structure, with
one to four folds. In two American species the posterior spiracles
are separate from the anus, as in Nudeocrinus; in others they are con-
fluent. The plates roofing the mouth, though not quite so stout as in
Nudeocrinus, are usually well preserved. Cryptoblastus, Eth. & Carp.
(1886), Carboniferous, N. America, is like a Schizoblastus with email
A. The hydrospires differ from those of Nudeocrinus and Sckizoblastut
only in the development of hydrospire- plates (cf. Troostocrinus) which
extend right up the sides of the lancet- plate, separating it from the
folds and from the walls of the radial sinus. But where the A are

THE DLASTOIDEA

Pio. XII.

Section of ambulacrum of Orbitremites

reached, the lancet-plates abut directly on them, without leaving any
pores ; the hydrospire-canc^s pass up beneath the lancet-plates, and open
through spiracles on either side of each A. Orbitremites, T. &. T. Austin

(1842, generally known as Granato-
crinus, Hall, 1862), Carboniferous,
England, N. America, and (?) Queens-
land (Fig. XV. 3, 4). The general
shape and relations of the plates are
as in Schizoblastus and Cryptoblastus.
Hydrospires the same as in Crypto-
blastus (Fig. XII). The hydrospire-
canals here are surrounded by the
A, and pass through them to five
single spiracles (the posterior confluent

Norwood}" xiQ~diaiinr"hp ,"hydrospIre-piateT with the anus) at the apices of the

A, which sometimes project as short

tubes. The plates roofing the mouth are minute and usually irregular.
Heteroblastus, Eth. & Carp. (1886), Carboniferous, England, and (?) N.
America, differs from Orbitremites in that the hydrospire -canals pass
beneath the A, and then curve outwards on either side of each A. The
adoral end of the A is produced upwards as a short stout process. Meso-
blastm, Eth. & Carp. (1886), Carboniferous, England, Belgium, (?) N.
America and Queensland, differs from Orbitremites in that the hydrospire-
canals are continued upwards over the flattened lateral portions of the
A, and open at the sides of the A crest. . In some species the crest is
almost absent, and the spiracles therefore almost single. Acentrotremites,
Eth. & Carp. (1883-86), Carboniferous, Somerset, is only known from
one specimen, but is a far more distinct genus than those just described
(Fig. XIII.). The general
form is that of Orbitremites.
The hydrospires do not pass
up into the A, but the canal
opens immediately on the
suture between A and R ;
there are thus ten spiracles,
as in Schizoblastus Sayi. The
anus pierces post. A, close to
its adoral end. The number
of hydrospire -folds is un-
known ; the inner walls form
a structure meeting in the

median line, and thus like a sub-lancet ; but also passing up the sides of
the lancet-plate, and thus like a hydrospire-plate. The lancet-plate is
small and perforate ; it is covered by small side-plates and large outer
side-plates.

Pentephyllum, Haughton (1859), Carboniferous, Limerick, is based
on a single internal cast. It is said to be unstalked, and this con-
dition, if obtaining, is correlated with the asymmetry shown in the
slight shortening of one ambulacrum and the slight curve of the

Fio. XIII.

Transverse section of ambulacrum of Actntrotremita
cllipticus.

THE BLASTOIDEA 91

adjacent ambulacra towards the opposite side. The linear ambulacra, the

large A, and the strictly pentagonal base remove it from the other

irregular Blastoids, and suggest affinities with the group of genera last

discussed. »Zygocrinus, Bronn (1848, syn. Astrocrinus, T. & T. Austin,

1843, non Conrad, 1840, nee Asterocrinus, Miinster), Carboniferous, Britain

(Fig. XIV.). This highly asymmetrical form

has been elaborately described by K Etheridge,

fil. (1876), and by Etheridge & Carpenter

(1886), but its affinities remain uncertain.

Its resemblance to Eleutherocrinus lies only in

secondary characters induced by a sessile life.

The theca is depressed, sternless, and produced

into four lobes ; on the shortest of these lies

an ambulacrum modified much as in Eleuthero-

crinus, and towards it the two larger BB

stretch up as in that genus. The four normal

ambulacra lie in the depressions between the

lobes. Three A are large and stretch out over

the lobes; two are small and flank the Etife^T

abnormal ambulacrum. Anus and spiracles £iai?- Oral view» modified

, ~ , , . „,, .. , food-groove (Arab) towards the

unknown. One hydrospire-told lies on each observer ; o, peristome ; AS ?,

side of the radial sinus, and is enclosed by it fette^alfulua?6™118' °ther
at the distal end (cf. Pentremites, Tricoelocrinus).

The surface is covered with strong tubercles which bear minute spines
(cf. Hystricrinus = Arthracantha, p. 158).

Classification. — We have now reviewed every known genus of
Eublastoidea, in an order approximating, as near as our very imperfect
knowledge of many types will allow, to that of their race-history. This
seems to show three main branches : one leading from Cadaster, through
Phaenoschisma, to Pentremitidea and Pentremites, with offshoots Crypto-
schisma and Orophocrinus ; the second from Troostocrinus, through Meta-
blastus, to Tricoelocrinus, with the offshoot Eleutherocrinus ; the third from
Nucleocrinus, through Schizoblastus, to Cryptoblastus, Orbitremites, Meso-
blastus, and Heteroblastus, with an offshoot Acentrotremites, and probably
Pentephyllum. Zygocrinus also is perhaps connected with this third line
of descent.

The classification of Etheridge & Carpenter does not agree very
well with the phylogeny here outlined. The erection of an "Order
Irregulares" is no more likely to be correct for Blastoidea than for
Crinoidea. With the exception of the Codasteridae, their families of the
" Regulares " are based almost entirely on the relations of the hydrospire
canals to the deltoids, relations which may vary considerably even in an
individual, while they take no account of important differences in the
relations of the hydrospires to the ambulacra. Moreover, in the con-
struction of family names, these authors have contravened the laws of
nomenclature.

The following classification attempts to overcome the above objections
while making as little change as possible : —

Series A. CODONOBLASTIDA. FAMILY 1. CODASTERIDAE. Hydrospire-

THE BLASTOIDEA

folds distinctly portions of the thecal plates, coming to the surface of the
radial sinus. No distinct hydrospire-canal or pores ; spiracles developed
imperfectly or not at all. Genera — Cadaster, PJuienoschisma, Cryptoschisma,
Orophocrinus. This family coincides with that of Etheridge & Carpenter.
FAMILY 2. PINTRKMITIDAE. Hydrospire-folds, usually numerous, concen-
trated at the lowest part of the radial sinus, and partly or wholly pendent.
Hydrospire-canal opens through spiracles bounded distally by side-plates.
Base convex. Ambulacra rather broad. Genera — Pentremitidea, Pentre-
mites. This equals the Pentremitidae of Etheridge & Carpenter, minus
Mesoblastus.

FIG. XV.

Theeas of typical Blastoids. 1, a Codonoblasticl — Pentremites robustus (from Brit. Mus. ES145).
2, a Troostoblastid — Tricoelocrinus woodmani (from Brit. Mus. E8171). 3 and 4, a Granato-
blastid— Orbitremites orbicularis (from the type-specimens, Brit. Mus. E8135). 3, from the side.
4, from below, with posterior interradius uppermost. All figures nat. size.

Series B. TROOSTOBLASTIDA. FAMILY 1. TROOSTOCRINJDAE. Elongate
forms with linear ambulacra descending sharply outwards from the much
restricted peristome. Hydrospire-folds only slightly concentrated, but
communicate with exterior through pores, and through spiracles bounded
by A and lancet-plates. Genera — Troostocrinus, Metablastus, Tricoelocrinus
This equals the Troostoblastidae of Etheridge & Carpenter. FAMILY 2.
ELEUTHEROCRINIDAE. Elongate, stemless, asymmetrical, with four narrow
ambulacra, accompanied by unconcentrated hydrospires. Fifth ambula-
crum shortened and widened. A minute. Genus — Eknthcrocrinus.

Series C. GRANATOBLASTIDA. FAMILY 1. NUCLEOCRINIDAE. Interam-
bulacra show traces of a primitive triad of plates. Ambulacra linear, and
stretching far down the theca, which is ovoid. Hydrospire-folds few and
pendent. Spiracles double. Mouth roofed by large plates firmly united
into a tegmen. Genera — Nuclcocrinus, Schizoblattus, This family equals
Etheridge & Carpenter's Nucleoblastidae, minus Cryptoblastus and Acentro-
tremitcs. FAMILY 2. OUBITRKMITIDAE. Theca globular with concave or
flattened base. Ambulacra linear, stretching down to concavity of base.
Hydrospire-folds few and pendent ; a hydrospi re-plate always present
(unknown in Hetcroblastus). Hydrospire-folds rarely penetrate A, but long
canals pass onward, through, beside, or under them, except in Acentro-
tremites. Genera — Orbit remitcs, Cryptoblastus, Heteroblastus, Mesoblastnx,
Acentrotremites. This corresponds to Etheridgc & Carpenter's Gfanato-
blastidae, plus Cryptoblastus, Mesoblastua, and Acentrotrcmites. FAMILY 3.

THE BLASTOIDEA 93

PENTEPHYLLIDAE. Theca subpentagonal, stemless ; RR asymmetrical.
Ambulacra linear, stretching down to base. One shorter than the rest.
Genus — Pentephyllum. FAMILY 4. ZYGOCRINIDAE. Theca depressed, stem-
less, asymmetrical, quadrilobate. Four ambulacra between the lobes,
accompanied by a single hydrospire on either side. Fifth ambulacrum
shortened and widened. A large. Genus — Zyyocrinus.

A Bibliography of the Eublastoidea was given by R. Etheridge, jun.,
and P. H. Carpenter, " Catalogue of the Blastoidea in the . . . British
Museum," London, 1886. A complete index of names with references
to literature is furnished by F. A. Bather, " The Genera and Species of
Blastoidea, with a List of the Specimens in the British Museum," London,
1899. For other references see Nos. 16, 17, 25, 27, 30, 34, 38, 45, 50,
00, 64, 69, 71, 73, 75, 76, 77, 78, 82, 84, 85, 94, 95, 96, in the list at
the end of Pelmatozoa (p. 211).

CHAPTER XL

THE CRINOIDEA.1

CLASS III. CRINOIDEA, MILLER (1821)

( = CRINOIDEA BRACHIATA, Audi, veterum ; EUCRINOIDEA, Zittel,

1879).

SUB-CLASS 1. MONOCYCLICA.

Order 1. Inadunata.
„ 2. Adunata.
„ 3. Camerata.
Sub-Order 1. Melocrinoidea.

„ 2. Batocrinoidea.

„ 3. Actinocrinoidea.

SUB-CLASS 2. DICYCLICA.

Order 1. Inadunata.

Sub-Order 1. Cyathocrinoidea.
„ 2. Dendrocrinoidea.
Order 2. Flexibilia.

Grade 1. Impinnata.

„ 2. Pinnata.
„ 3. Camerata.

PELMATOZOA in which epithecal extensions of the food-grooves,
ambulacrals, superficial oral nervous system, blood-vascular and
water-vascular systems, coelom, and genital system are continued
exothecally upon jointed outgrowths of the abactinal thecal plates
(brachia), carrying with them extensions of the abactinal nerve-
system. The number of these processes is primitively and norm-
ally five, but may become less by atrophy. The brachia rise from
a corresponding number of thecal plates, " radials (RR)." Below
these is always a circlet, or traces of a circlet, of plates alternating
with the radials, i.e. interradial, and called " basals (BB)." Through

1 By F. A. Bather, M.A.

THE CRINOIDEA

95

all modifications, which are numerous and vastly divergent, these
elements persist. A circlet of radially situate infrabasals may
also be present. Below basals or infrabasals there follows a stem,
which, however, may be atrophied or totally lost.

Although many Rhombifera simulate Crinoidea in the penta-
merism of the theca or the possession of exothecal extensions of
the food-grooves, yet in none are those extensions supported by
plates that are clearly outgrowths from the abactinal system of
thecal plates ; in none is there the intimate correlation between
brachia and radialia that obtains in the Crinoidea. This class
therefore cannot be derived from the Rhombifera, as many
structures might otherwise lead us to suppose ; the presence of
brachia also forms a clear distinction between it and Diploporita
and Blastoidea. A further difficulty in tracing the origin of the
Crinoidea is furnished by the occurrence of perfectly developed

DO

FIG. I.

Analysis of the cup and brachial elements of Hybocystis problematic™. The outlines of the
food -grooves (vg) are dotted.

species in the Lower Cambrian, while representatives of all crinoid
orders are plentiful in Ordovician rocks. Further research, how-
ever, may throw back the origins of other Echinoderm classes ; in
any case, negative evidence when Cambrian rocks are concerned
counts for little.

Certain features in some of the older Crinoidea seem to throw
light on their ancestry. Such are the presence of hydrospires,
comparable to those of Cadaster, in Carabocrinus (p. 172) and
Hybocrinus (p. 145); the presence in these and other genera of
well-developed deltoids (A), over the edges of which pass the ambu-
lacra, while the posterior A frequently shows signs of a hydropore
(Fig. XXXVI.) ; the absence of a brachium from certain radials in
Baerocrinus (Fig. LVII. 4) ; the greater development of three radials
in many Inadunata Monocyclica (see p. 144). We are thus led to
a form not unlike that which isy actually presented by Hybocystis,
Wetherby (1880), from the Ordovician of Kentucky (Fig. L).
This has 5 large subequal basals, 5 radials, and 5 deltoids. The

96 THE CRINOIDEA

anus lies between the posterior A and the radial circlet, being
separated from the latter by a special anal plate (a:). The right
posterior radial is transversely bisected ; its upper smaller portion
(Rs) being pushed a little to the right by x. The striking
peculiarity of this form is the continuation of the food -grooves
over the thecal plates, as in Diploporita and Blastoidea. In the
right and left antero-lateral rays these pass over the edges of the
deltoids, over the radials, on to the underlying basals. In the
anterior and the right and left posterior rays there are two ossicles,
each as high as wide, supported on the summits of the radials ;
the grooves pass between the deltoids, over these ossicles, down on
to the outer surfaces of the radials. These ossicles form exothecal,
jointed outgrowths of the abactinal thecal plates ; a deep notch on
their inner surfaces, leading into the cup by a hole between the
deltoids, suggests that they bore, besides the ambulacral structures,
also extensions of the abactinal nerve-system. Therefore, although
incipient, they constitute true brachia, such as are found in no
Echinoderma except Crinoidea, and they show us the probable way
in which brachia originated. Hydrospires have not been described;
but, considering their occurrence in the closely allied Hybocrinus
(Fig. XXXVI. ), they are likely to be found along the radio-deltoid
sutures, as in Codaster. Brachioles fringing the grooves do not
seem to have been present, nor has a lancet-plate been observed.
These facts, as well as the five basals, prove that Hybocystis is not
an offshoot from Eublastoidea, as indeed its geological age forbids ;
but it may well be derived from early forms of Protoblastoidea. If
Hybocystis be admitted as actually ancestral, then the development
of brachia in only three rays sheds light on corresponding irregu-
larities of development in many simple and ancient crinoids, connect-
ing them in this respect with primitive Cystidea (see pp. 11, 53).
Another form suggestive of the connection of the Crinoidea with
the Blastoidea is Stephanocrinus, Conrad (1842), Silurian of N.
America and England. C. F. Roemer (1851), Joh. Miiller (1853),
and Pictet (1857), regarded it as a cystid ; Etheridge & Carpenter
(1883) and S. A. Miller as a blastoid ; Dujardin & Hupe" (1862)
and Hall (1851) as a crinoid ; Zittel (1879) as doubtfully a
blastoid. Wachsmuth £ Springer (1886) proved the presence of
brachia, which make it unquestionably a crinoid, but said, "It
agrees by its oral and anal pyramid with certain forms of the
Cystids, while in its general habitus and in the position of the
ambulacra it agrees with the Blastoids." Stephanocrinus (Fig. II.)
has 3BB, 5RK, and 5A, arranged as in Eublastoidea, especially
Codasteridae. The radial processes are often prolonged into spear-
like spines (S), one in each interradius. Each ambulacral groove lies
in a deep sinus between the deltoids and radial processes, and it is
continued on to an arm, which rises from a single brachial at the end

THE CRINOIDEA 97

of the sinus, and immediately bifurcates, the groove forking with it.
The edges of the deltoids meet beneath the groove, but a space for the
mouth (" peristome ") is left in the middle. This space, as well as
the whole groove, is covered by ambulacrals ; these often fuse into
a single plate on either side the groove, where it passes over the oral
surface of the theca, and form five plates, likewise often fused, over
the peristome (P). The anus is between posterior A and the adjacent
radial processes, and is closed 'by a valve of four to six small plates.
Certain pits in a similar position in other interradii possibly are
atrophied hydrospires (h). The ornament of the cup-plates is strongly
reminiscent of that in Eublastoidea ; but there are clearly marked
axial folds passing up from the basals to the arm-facets, perhaps due
to the greater development of the abactinal nerve- system in the
brachiate form. There were neither brachioles nor a lancet-plate.
Stephanocrinus undoubtedly belongs to the simplest and most
primitive group of the Crinoidea, and it is hard to believe that its

FIG. II.

Stephanocrimis angulatus. 1,
from anterior radius, x 2 diam. ;
2, from oral surface, x 4 diam.
(from Brit. Mus. E6715). As, anus,
covered by plates ; AJ; axillare,
from which are supposed to spring
two arm-rami ; Br1, position from
which these spring, between 6',
spines formed by radial processes,
broken off in 2, and showing (/i)
supposed atrophied hydrospires ;
P, large covering - plates over the
peristome, which is surrounded by
the five orals or deltoids (0).

remarkable resemblances to Eublastoidea are merely homoplastic,
especially since the position of the small basal is not one which
usually occurs in other Crinoidea that have fused basals.

However the crinoid or brachiate stage in the history of the
Pelmatozoa may have been reached, it will be useful to recapitu-
late here the common pelmatozoic characters as well as those
distinctive of the Crinoidea, as manifested in a normal crinoid of
simple structure. The specialisation of those characters will be
shown historically in the systematic part; but since many
structures have been produced or modified in the same way more
than once, a general account of the processes may be given here.
We can speak more decidedly on questions of development and
internal anatomy in this class, since the differences between extinct
and recent genera are not such as to hinder interpretation.

A normal Crinoid was thus described in 1821 by J. S. Miller,
the founder of the class : " An animal with a round, oval, or
angular column, composed of numerous articulating joints,
supporting at its summit a series of plates or joints forming a

7

98

THE CRINOIDEA

cup-like body containing the viscera, from whose upper rim
proceed five articulated arms, dividing into tentaculated fingers,

Anal liJ}e

i

A simple form of Crinoid : Botryo-
crinus decadactylus of the Wenlock
Limestone, seen from posterior inter-
radius.

more or less numerous, surrounding the aperture of the mouth,
situated in the centre of a plated integument, which extends over
the abdominal cavity, and is capable of being contracted into a
conic or proboscal shape. Some species of these animals ascer-

THE CRINOIDEA 99

tained to be permanently attached to extraneous bodies, whilst
others appear to have been capable of locomotion." So little is
amiss with this description, that we need do no more than trans-
late it into modern terminology, as follows : —

A normal Crinoid (Fig. III.) consists of a " crown " (corona)
attached by its dorsal (i.e. aboral) extremity to a " stem " (columna),
which is fixed to the sea-floor or to some solid body by a "root "
(radix). The crown consists of a theca (or calyx, in the sense
of Wachsmuth & Springer) containing the viscera, and of 5
" arms " (brachia), which may be more or less branched. That
part of the theca below the origins of the free arms is called the
" dorsal cup " (or shortly " cup ") ; that part above the origins of
the free arms, i.e. the oral surface, is called the tegmen (sometimes
" disc," sometimes "vault," between which a distinction erroneously
used to be imagined). The skeletal and many of the other
systems have a radiate arrangement, of which 5 is the dominant

FIG. IV.
Imaginary analyses of the structure of the dorsal cup in two simple types of Crinoid.

number. Thus the whole animal can be divided into 5 cor-
responding and almost symmetrical sections, " pentameres," by
5 imaginary "perradial planes," starting from the vertical
dorso-ventral axis and passing through the origins of the arms.
The skeletal elements are either perradial or interradial in
position.

The Dorsal Cup in its simplest form is composed of 2 or 3
circlets of 5 plates, those in one circlet alternating with the 5 in
the adjacent circlet (Fig. IV.). Of these the most important are those
that support the brachia, and to them the term radialia (RR) is
restricted. The interradial plates below these are called basalia
(BB), since in many crinoids they form the base and rest on the
stem. In some crinoids a circlet of perradial infrabasalia (IBB)
occurs beneath the BB (which latter are then called pardbasalia
by some writers). The former type of base is called " monocyclic ";
the latter " dicyclic."

The Tegmen in its simplest form is likewise composed of 5

100

THF CR1KOIDEA

plates, deltoidea (A), here regarded as synonymous with oralia (0),
alternating with the RR (Fig. V. 1). But there are nearly always
also present "ambulacralia" (Amb or c.p), covering the grooves
that lead between or over the apposed edges of the A to the

,P

As

Three stages in the evolution of the Tegmen. 1, orals only: 2. orals anil ambulucruls ;
8, orals, ambulacrals, and enlarged peristomial ambulacrals.

brachia (Fig. V. 2). The mouth is either beneath the A or in a
space between them ; in the latter case it is covered by ambulacrals,
often 5 in number and interradial in position, and taken to be
orals by some writers (P in Fig. V. 3). The posterior A in many
primitive forms seems to have been pierced by a
hydropore, the walls of which may be folded so as
to form a madreporite (cf. Stelleroidea and Echi-
noidea). The anus (As) lies between post. A and
the adjacent RR, and is closed by a valvular
pyramid, often surrounded by, or raised on, small
plates.

The Brachia in their simplest form consist of a
series of ossicles called brachialia (Br), which con-
tinue straight up from the radials (Fig. VI.). The
surface of the radial to which the proximal brachial
is attached is called the "radial- or arm-facet."
Each brachial is rounded on the outer or dorsal sur-
face, and grooved on the inner or ventral surface.
FIG. vi. The ventral or brachial groove contains the follow-

branched'arm that *n& s0^ Parts> ta^en in order from ventral to
of iiybocrinus. ^A1, dorsal (Fig. VII.). On the surface, the food-groove
(f.g), lined with ciliated epithelium (et), which directs
a stream of water towards the mouth ; an epi-
thelial nervous band (nl) stretching from the
superficial or oral nerve-system; a blood-vessel (b), "radial
pseudhaemal canal"; a water - vessel (w), which gives off tubes
(p) to a series of podia or " tentacles " (t) that fringe the
food -groove and subserve sensation and respiration; two "sub-
tentacular canals " (s.f.c), extensions from the body-cavity ; a canal

covU-
°r an'"

THE CR1NOIDEA

101

containing the genital cord or rachis (g.c) • and a " coeliac canal "
(c.c). On either side of the water-vessel, beneath the tentacles, is a
senso-motor nerve (/i2), giving off branches to the muscles of the

Fio. VIT.
Diagram of solid section of a Crinoid arm. For e:

wtion of letters see adjoining text.

water-canals and to sensory papillae (s) on the tentacles. Below
all these, in a special groove on the very floor of the main brachial
groove, lies another nerve, the " axial cord " (a.c), proceeding from
the aboral or dorsal motor nerve- system ; the groove in which this
lies is often (as in Fig. VII.) separ-
ated from the brachial groove
during individual development
by an outgrowth of stereom, and
is then known as the " dorsal " or
" axial canal " (Fig. VIII.). The
axial cord sends off branches (w3)
to all the muscles of the arms,
and to supposed sensory endings in the ectoderm (ed), and is con-
nected with the subtentacular nerves. All these soft structures
in the ventral groove are protected by covering-plates (c.p\ also

Stapes in the separation of an axial canal,
exemplified by brachials of Gissocrinun rjnnio-
dactyl us. 1 is youngest ; 4, oldest. X <• diam.

102

THE CRINOIDEA

called " ambulacrals " (Amb), which can open or close as occasion
demands (Fig. IX.).

We may now trace the various
extensions of the Body Systems into
the Thecal Cavity.

The food -grooves and associated
structures, except the axial cord, pass
over the tegmen to the mouth, into
which the food -grooves drive their
streams of water. The mouth leads
into a gut, which makes a dextral coil
down to the bottom of the cup, and
Pio IX then rises along the side of the cup to

Ambulacrals0' 1, ventral view of the anUS '> this system, then, is DOt

two brachiais of Giaomnvs sqnami- affected by radiate symmetry (see

fer, with c.p closed above and re- T,. _7TT nx * \

moved below, x 8 diain. 2, side view r Ig. V 11. p. y ).

The epithelial nerves on the floor
of the food -grooves also pass to the
are'seen'"notchevs'for isaccuiiVisf),' mouth, where they join an epithelial

ridge encircling the mouth ; from

this "oral ring," nerves pass to the walls of the gut. The
paired subtentacular nerves run down to a subepithelial,
" circumoesophageal nerve-ring," below the oral nerve-ring. From
this ring proceeds, in each interradius, a pair of nerves which
innervate the tegmen and the mesenteries of the body-cavity. This
nerve-system is connected with the aboral nerve-system in a manner
explained below.

The radial pseudhaemal canals join a " pseudhaemal ring "
round the oesophagus beneath the oral nerve-ring ; these structures
are hard to distinguish, and even in other classes, where they are
better developed, their origin is not yet clear. There is, however,
surrounding the oesophagus a " lacunar plexus " belonging to what
is generally called the blood-vascular system. The circumoeso-
phageal ring is connected with two vascular trunks leading from
the plexus that surrounds the intestine and that absorbs nutrient
substances therefrom ; these substances appear to be worked up
into corpuscles by a "spongy organ" in the oesophageal ring. The
ring is also connected with a plexus that passes down the vertical
axis of the theca, through the coil of the gut, to the base ; this
surrounds the " axial organ " (vide infra).

The water- vessels (perradial ambulacral canals) meet in a
circumoesophageal water -vascular ring (hydrodrcus) ; these struc-
tures have longitudinal muscle-bands, as well as muscle-fibres
traversing the lumen ; no ampullae or valves are differentiated,
as they are in forms where this system has a locomotor function.
In so simple a crinoid as is here in question, there is good reason to

THE CRINOIDEA 103

believe that the water-ring opened into the body-cavity by a single
ciliated canal in the posterior interradius ; and that the body-cavity
communicated with the exterior by a single hydropore in post. A,
sometimes merged in the anal opening (as probably in Blastoidea).
This system was not as yet completely affected by radiate sym-
metry ; but in some forms it became so by the development of a
similar canal with corresponding hydro-
pore in each of the other interradii (Fig.
X., compare Figs. XXXIV., XLVL, and
CXIV.).

The two subtentacular canals of each
arm enter a division of the coelom that
passes down the vertical axis through
the coil of the gut, and is known as the Fl0t X-

"axial sinus." The dorsal coeliac canal
passes into a division of the coelom that

SUrrOUnds both axial sinUS and ffUt, and connective tissue fibres ; g, wall of

11 j ,-, ,, ••,.,• i •, » gut; n, circumoesophageal nerve ;

IS Called the pen - intestinal Cavity. oe, oesophagus; p, pore; r.c, ring

The remainder of the coelom, surround- ^.su, stone canal ; r, stereom
ing the latter, is called the " subtegu-

mentary cavity." All these divisions of the body -cavity are
lined by endothelium, and are separated from each other, as
well as penetrated, by connective tissue, in which spicules are
often richly developed. From the peri -intestinal cavity, at its
aboral end, there are in this way cut off five chambers, which
surround the axial sinus, and are themselves covered on all sides
by epithelium, containing ganglion -cells and nerve-fibres ; the
whole structure is called " the chambered organ " (see Fig. XX.
p. 24).

The genital rachis of each arm is connected with a complex of
twisted, fine canals, called the "axial organ" (see p. 23). This
passes down the axial sinus, widening in the middle of its course,
and then narrowing to a thin strand as it passes between the
five chambers just mentioned.

The axial nerve-cord of the arm does not, as all the organs yet
dealt with, pass to the oral centre, but enters the vtheca over
the radial. If there is a separate axial canal, it may be continued
through the radial facet into the substance of the thecal plates.
The cords ultimately pass into the epithelial covering of the
chambered organ, but their passage is not a direct one (Figs. XI.
and XII. ). Each cord is really a double structure, connected
at intervals by chiasmas, and so soon as it enters the radial it
divides into two branches, one of which proceeds to the basal on
the right, the other to that on the left. In addition the branches
are connected with each other and with those of the other radii by
a series of commissures that form rings all round the cup. One

104

THE CRINOIDEA

such ring is at the level of the radials. If the crinoid have a
monocyclic base, the cords that pass to the basals join one another
in a ring immediately surrounding the chambered organ, the

Fie. XI.

Course of axiul nerve-cords in Iscx-rinns. Diagrammatised from sections figured by P. H.
Carpenter. It, basal ; 7t', radial ; oar, axial organ ; ch, live chambers of chambered organ ; «1,
nerve-cord from 7? to 1> ; «2, cord passing down II / w3, cord from 7> to radially placed lobes of
chambered organ.

lobes of which in this case correspond with the basals, i.e. are
interradial. If the base be dicyclic, the ring forms a commissure
at the level of the centres of the basals ; and from these points

IB--

St

Fm. XII.

Course of axial nerve-cords in Dicyclic (I)), Pseudomonocyclic (!'), and Monocyclic (M)
Crinoids. c.o, lobes of chambered organ, the connecting nervous sheath omitted for greater
clearness ; r.r, ring commissure in radials ; other letters as usual.

the cords again fork towards the adjacent irifrabasals, where
they join in another ring round the chambered organ, the lobes
of which in this case correspond with the infrabasals, i.e. are

THE CRINOIDEA

105

radial. Nerves from these cords are given off to the stroma of
the cup-plates. This system, as experimentally proved, chiefly
by \V. B. Carpenter (1876 and 1884), in opposition to the
scientific opinion of his time, is a senso- motor nerve -system,
governed from the nervous capsule of the chambered organ.
By means of the commissures the motion of all muscles is cor-
related. Primitively the cords lie on the inner surface of the
cup; they then become bordered by ridges of stereom, and
finally enclosed within the cup-walls. Branches from these nerves
unite with the interradial nerves that proceed from the circum-
oesophageal nerve-ring.

To turn to the Stem. We have already traced its probable
origin as an evagination of the many-plated theca of Amphoridea,
and the gradual introduction of order into the irregular plates (p. 4 8).
In the pentamerous Crinoidea, these naturally became subjected to
pentamerism ; and evidence of many of the older crinoids shows that
the plates were at first hexagonal and arranged in alternating circlets

FIG. XIII.

Evolution of Pentamerism in the
Stein. IJointsurface of a columnal com-
posed of five sections, which alternate
with the angles of the stem-lumen ; 2,
portion of a stem composed of hexag-
onal alternating plates, which in 3 be-
come arranged more definitely in hori-
zontal rows ; 4, continuance of the
process results in columnals of five pen-
tameres. The figures are of Botryo-
crinus stems (after Bather).

of 5, just as plates of the theca (Fig. XIII.). The next stage was
that in which the plates no longer alternated, but were arranged in
horizontal rows divided by five longitudinal sutures. Finally, the
pentameres of each row became fused to form a " columnal," still
pierced by a wide lumen. This regularity was perhaps connected
with the extension into the lumen of a vessel from each of the five
lobes of the chambered organ, with its nerve-sheath (axial cord) ;
the five cords surrounded a prolongation of the axial organ. In a
monocyclic crinoid the axial cords would be interradial, as are the
lobes of the chambered organ, while the pentameres would alternate
with the basals and be radial. In a dicyclic crinoid the cords would
be radial, the pentameres interradial. The exterior angles of the
stem usually correspond with the pentameres, but not always.
The cirri, or side-arms of the stem, correspond, for reasons that
will appear presently, with the axial cords. The lumen of the
stem is often split up into grooves by ingrowths of stereom ; and
these grooves primarily contain the axial cords, and may even form
closed canals containing the cords, but this correspondence is not
inevitable. The so-called "law of Wachsmuth & Springer,"

io6

THE CRINOIDEA

summarised in the annexed diagram (Fig. XIV.) and table, is in itself
empirical, applicable only to pentagonal stems or lumens, and even
then liable to exceptions (marked * in table) ; but by attending

DICYCLIC

MONOCYCLIC

Fio. XIV.

Comparison of the Dicyclic and Monocyclic base. B, basal ; Br, brachials marking the five
rays ; ci, cirri, only three out of the tive are shown ; co, pentameres of column ; IB, infrabasal ;
n, nerves going to cirri from extensions of capsule ; R, radial ; s, sutures between penta-
meres of stem.

(as is here done) chiefly to the relations of the axial cords, we shall
have a surer guide for discrimination between monocyclic and
dicyclic crinoids in the many doubtful cases that occur.

DICYCLIC.

MONOCYCLIC.

BB [lobes of capsule in Monocyclica] .

Interradial

Interradial

IBB [lobes of capsule in Dicyclica] .

Radial

Pentameres of stem (p) , «. ;

Interradial

Radial

* Outer angles of stem .

Interradial

Radial

Vertical sutures of stem (s) .

Radial

Interradial

* Sides of stem .

Radial

Interradial

* Angles of lumen of stem

Radial

Interradial

Cirri, when present (c)

Radial

Interradial

[Axial cords] .

Radial

Interradial

The primitive crinoid is attached by the distal end of its stem ;
and it is supposed by many, from the evidence of the embryo
Antedon (Fig. XV.), that there is developed at that point a special
fixing plate, to which they apply the term " dorso-central," which
must not be confused with * 'centre- dorsal " (see especially the
writings of P. H. Carpenter). Palaeontology does not lead us to
regard such a structure as primitive, or to ascribe to it any morpho-
logical importance. As a rule, skeletal growth takes place at the

THE CRINOIDEA

107

distal end of the stem after two main plans: (1) Deposition of

solid, unjointed stereom, around the distal columnals, forming an

encrusting plate or mass (Fig. CXIII. 2) ; this

occurs on rocky bottoms. (2) Outgrowth of jointed

branches from the plated end, forming " radical

cirri," often with traces of polymeres like those of

the primitive stem, often very long and branching

again, and always with a lumen which contains

an extension of the axial cord (Fig. CXVI.) ; this

is adapted to a muddy bottom. The radical cirri

arise from the vertical suture-lines of the stem, by

the intercalation and outgrowth of small plates,

and the extrusion of the axial cord (Fig. XVI. 3). fifa, after wyviiie-

T ,v r i • , ,, . . j n Thomson, x 20 diam.

In the course ot race-history the cirri gradually ap- Coi, fascicuiar stereom

FIG. XV.

Distal end of stem
of larval Antedon bi-

pear higher and higher up the stem (Fig. XVI. 1),

and at the same time become shorter, more mobile,

and eventually arranged in whorls (Fig. XVI. 5). At these levels

the axial cords of the stem swell out, forming a repetition of

the chambered organ (Fig. XVI. 4).

The columnals are rarely all of the same height ; certain con-
spicuously larger ones, including those that bear cirri, are termed

Fir,. XVI.

Evolution of cirri. 1, part of stem of .1 Silurian crinoid with large, branching cirri (Brit.
Mus. E1354). 2, section across stem of a Carboniferous crinoid, showing branch from axial
canal to cirrus (Brit. Mus. E6708). 3, root of Jlarycrinus, with cirri originating between
pentameres (modified from Wachsmuth & Springer). 4, section across stem of /MCfiHtt*
WyvUle-ThmnMni at, lovel of cirrus-whorl, the central portions disproportionately enlarged
for greater clearness. 5, part of stem of Isocrinus decants, with cirri in whorls of live.

"nodals"; those between them " internodals " (Fig. XVII. 1).
Nodals are the first columnals to be formed during growth ;
internodals are subsequently intercalated between them, and
again fresh internodals between the first formed internodals, and
so on. Fresh nodals are developed at the proximal end of the
stem, so that in that region are more nodals, while distally are
more internodals (Fig. XVII. 2 and 3). In one type nodals are
introduced immediately beneath the base of the cup, so that the

io8

THE CRINOIDEA

proximal columnal is always the youngest. In another type the
proximal columnal is one of the first formed, and remains attached
to the cup, new nodals being introduced below it (Fig. XVII.
4-7). This proximal columnal is called article basal by P. de
Loriol, and " centro- dorsal" by others, erroneously. It is here
called the proximale.

The Connection between the Elements of the Crinoid Skeleton
is primarily by means of the fibrils of the stroma in which they
are deposited. This condition persists in the " primitive suture,"
and from it development proceeds in the direction of either
greater rigidity or greater flexibility. Towards rigidity we have :
(1) "Close suture," in which the fibres are short, and their ends
surrounded by denser layers of stereom on the apposed surfaces of

4

Fi<;. XVII.

The Relations of the Stem. 1, portion of stem of Gastrocriwts jwtulus (modified from Jaekel),
the latest formed columnals are numbered 1, the oldest 6 ; ,r> and G bear cirri. 2, proximal, and
^distal, regions of stem of Isocrinns decorus (of. 5 in Fig. CI.), /)., nodals (modified from P. H.

Springer)

5, proximale of Apii*-rinn* dcynns, Defr., showing
0, cup and part of stem of A. elegans, showing

Carpenter). 4, proximal region of stem of OnifchHcrinus (after Wachsmnth & Springer), show-

,le

proximale and other enlarged columnals (based on Brit. .Mns. Er,70i» and E6710)." 7, portion of

ing infrabasals (IJi) fused to proximale (}').
depressions for Bl> (Brit. Mus. E0711).

cnp and stem of Millcricrintis polyihtctitlim fittodiflad from 1'. de Loriol), showing minute
infrabasals attached to proximale, also new columnals forming (1).

the ossicles, which are thus closely and immovably fitted together,
though separable by alkalies. (2) " Syzygy," a special case of close
suture between brachials or columnals (Fig. XVIII. 2, 3, and 4),
in which the upper ossicle, " epizygal," bears a pinnule or cirri,
as the case may be, and the lower one, " hypozygal," bears none.
(3) " Anchylosis " or fusion, when two ossicles are immovably
cemented by an unbroken deposit of stereom, which, however,
is less solid than that of the plates themselves. Towards flexibility
we have: (1) That form of "loose suture" in which the stroma-
fibrils lie at right angles to the suture, and the stereom is thrown
into corresponding folds (cf. pore-rhombs of cystids), or that form
in which there is a slight facet, either smooth or striated (Fig.
CXI. 3), or interlocking crenulations (Fig. XVIII. 1). (2) " Imper-
forate articulation," in which there is a slight facet, or a toothed

THE CRINOIDEA

109

articular surface (Fig. XVIII. 5) ; the fibrils are at first developed
into elastic ligaments, and later into true muscles. (3) "Perforate
articulation," in which there is a highly developed facet, with fulcral
ridge, ligamentar depressions, and muscles innervated from an axial
cord which perforates the ridge (Fig. XVIII. 6).

In such a simple crinoid as that under discussion the cup-plates
would be united by close suture ; the tegminals, probably by
primitive suture, or loose suture of the rhomb type ; the brachials,
by imperf orate articulation, perforate being a later development ;
the columnalsj by loose suture of the striate type.

The Skeletal Elements of a crinoid may be thus classified :
" Primary Elements," the first to be developed in both ontogeny
and phylogeny, divided into — " abactinal," developed on the
right or aboral coelom, and directly innervated from the chambered

Fiu. XVlll.

Forms of Joint. 1, sutural mar-
gin of cup -plate of Marsupitex
(original). 2, brachials from dorsal
side, x 10. 3, ditto from ventral
side, x Y. 4. distal face of HBr^,
a hypozygal, x *£. ('2, 3, and 4 are
of Antedon Infida, after W. B. Car-
penter.) 5, articular facet of radial
of Pisocrinus ollula, x 3 (after
Bather). C, articular facet of radial
of Jiatkycrinus aldrichianus, x S
(after P. H. Carpenter), ac, axial
canal for nerve ; dl, dorsal liga-
ment ; /, fulcral ridge ; il, inter-
articular ligament ; mf, muscle
fossa ; n. notcli for axial nerve ;
p, pinnule; p', pinnule-facet; ,<••,

organ, viz. columnals, cirrals, IBB, BB, ER, Br, and pinnulars
(vide infra) ; " actinal," developed on the left or oral coelom, and
connected with the various oral ring-systems, viz. A (orals) and
ambulacrals (Chapter VIII., Fig. X.). "Secondary or Supple-
mentary Elements," which may be intercalated between the primary
pieces ; these have not yet been discussed ; they include " inter-
brachials" (iBr), " interambulacrals " (iAmb), " interaxillaries "
(iAx), some " anals," and a few others of no special importance.

The terms " proximal " and " distal " are reckoned from the
plane separating stem from crown, so that the infrabasals and top
columnal are the proximal elements of crown and stem respec-
tively. The actinal elements, however, start from the oral centre
as proximal point.

For orientation the Crinoid is placed in its natural position,
mouth upwards, and is viewed from the anal side. The anal
interradius is then posterior ; the radius opposite it is anterior ;

no

THE CRINOIDEA

right and left correspond with the right and left of the observer.
To preserve this orientation when the crown is viewed from above,
the anal side must be nearest the observer (downwards in a
figure) ; when viewed from below, the anal side must be away
from the observer (upwards in a figure). Such is the rule followed
in the drawings illustrating this book, while in the various analyses
the anterior radius is always placed on the right of the figure.
Various modes of designating the radii have been attempted. To
extend to the Crinoidea Loven's Echinoid numeration, is to postu-
late an homology that is far from proven. The annexed table
compares with other systems the symbols here used : —

Orientation as above described.

Symbols here
used.

|!

|_H Loven for Echin-
oidea, Jaekel for
I""1 Crinoidea.

Ijjl

Anterior Radius

ant. R.

A

ant. 1

Right Anterior Interradius
Right Antero-lateral Radius .
Right Postero-lateral Interradius

r. ant. IR.
r. ant. R.
r. post. IB.

A-B
B

B-C

I

I

Is

3
IV.

4

1. ant.
1. ant.-lat. J £
1. post.-Iat.

Right Posterior Radius

r. post. R.

C d

V.

1. post. 1 m

Posterior Interradius ... .

post. IR.

C-D

5

a
post.

Left Posterior Radius

1. post. R.

D 1

I.

r. post. ) *

Left Postero-lateral Interradius

1. post. IR.

D-E

£

1

r. post.-lat.

Left Antero-lateral Radius
Left Anterior Interradius .

1. ant. R.
1. ant. IR.

E

E-A J

3

II.

2

r. ant.-lat. ) |

r!

r. ant. J £

We have now run through the chief characters of a normal
crinoid of simple structure. Few are so simple as this, but
various modifications have occurred in the history of the class,
some perhaps only once, others at different geological periods in
races of diverse origin. Some of these, especially when of im-
portance for classification, must now be discussed.

Some crinoids have a Dicyclic, others a Monocyclic Base (p.
99). The value of this in classification is disputed. Among
various early genera, placed by Wachsmuth & Springer in a
single family (Keteocrinidae), some are with, some without infra-
basals, having the angles of the stem-lumen respectively radial
and interradial. But the yet simpler genera, from which presum-

THE CRINOIDEA in

ably these genera descended, probably differed in the same way ;
and resemblances, undoubtedly of secondary nature, should not
lead us to place together forms of diverse origin. The distinction to
be drawn between monocyclic and dicyclic genera is more obvious
in the simpler crinoids ; but here too there are parallel stages
passed through — the monocyclic locrinus and Heterocrinus (p. 145)
correspond with the dicyclic Merocrinus and Ottawacrinus (p. 178)
respectively. Since the presence or absence of infrabasals is cor-
related with the radial or interradial position of the lobes of the
chambered organ, the derivation of one type from the other
involves more change than the mere atrophy or appearance of
certain plates. Hence monocyclic and dicyclic genera should not
be placed in the same line of descent, unless this change can be
proved : there is no reason why they should not have been inde-
pendently evolved. The origin of Dicyclica from Monocyclica
is, in fact, opposed by the available evidence ; but Monocyclica
may conceivably have been derived from Dicyclica in one of two
ways, outlined in the next two paragraphs.

There are, especially among the later crinoids, several genera
known as " pseudomonocyclic," because though infrabasals are in-
visible or absent, at all events in the adult, the evidence of the
axial cords (e.g. in Bhizocrinus) of palaeontology (e.g. Apiocrinus
and Pentacrinus), or of embryology (e.g. Antedon\ demonstrates the
existence of infrabasals sither in the young or in near ancestors.
Discoveries of this nature have strengthened Wachsmuth &
Springer's law by affording a rational explanation of apparent
exceptions. But suppose secondary growth of stereom to occur
in a pseudomonocyclic genus, converting the angles of the stem
from interradial to radial, and the angles of the stem-lumen from
radial to interradial. Then, if recent examples were known, the
law, as emended above (p. 106), might be applied successfully;
but it would not tell the truth if only fossils were available, and
the crinoid would pose as monocyclic. Such changes are actually
observed in the growth of Antedon, while in Isocrinus, which other
facts prove pseudomonocyclic (Fig. XL), the angles of the stem-
lumen in the proximal region are interradial, as if the genus were
truly monocyclic, though the downward prolongations of the
chambered organ are radial. In Glyptocrinus Fornshelli the angles
of both stem and axial canal are radial; since there are actually.
no infrabasals, we may suppose secondary ingrowth of stereom,
and this is confirmed by S. A. Miller's description of the columnals
(1874).

The changes just described leave the essential distinction
between monocyclic and dicyclic genera untouched (as shown in
Fig. XII.); and among the earliest crinoids there is little evidence
of pseudomonocyclic forms. There is, however, a possibility that

THE CRINO1DEA

the change from Dicyclic to Monocyclic may have taken place,
not by compression and atrophy, but by torsion and fusion. Many
monocyclic genera of Ordovician and Silurian age have some
radials (usually r. post., r. ant., and 1. ant. RR) transversely bisected ;
the upper part is called "super-radial" (Rs) ; the lower part,
" inferradial " (Ei), (see Fig. LVIIL). Now, in some dicyclic
genera (e.g. Ottawacrinus, Fig. XCVL), perhaps in consequence
of the introduction of fresh plates in the anal interradius, the
radials are shifted to right and left so as to lie almost vertically
above the basals. The suggestion then is that the inferradials and
basals of Monocyclica represent basals and infrabasals respectively
of Dicyclica. If then the Rs and R?' fuse, a truly monocyclic type
is produced with one circlet of BB and one of RR. One obvious
objection to this theory is the presence in many Dicyclica of a
plate (the radianal, RA), which is now generally regarded as a
slightly modified inferradial (r. post. Re), (Fig. XXVI.). Other
objections to this and similar views, based by P. H. Carpenter
(1878) on alleged homologies with the apical system of Echinoidea,
have, it is true, been somewhat discredited by modern embryological
and palaeontological research. Nevertheless, for the present the
gulf between Monocyclica vera and Dicyclica is unbridged, and
must be recognised in classification.

The dorsal cup of a simple crinoid consists of two or three
circlets, but there is often a tendency for the proximal brachials to
be so joined to the radials and to one another as to form part of
the cup (Fig. XIX.). There is, however, a supposed
morphological distinction between these "fixed
brachials " (Br) and the radials : the latter are de-
veloped in the Antedon larva as expanded sieve-
like films ; but all brachials begin as " imperfect
rings, which soon become filled up with lengthening
fasciculated tissue" (P. H. Carpenter, 1884; see Fig.
XIX.). In a form where many brachials enter the
dorsal cup, it is convenient to have a common tern
FIO. xix. f°r the primitive elements of the cup (IBB, BB,
R, radial, crib- RR) : some call them the " apical system," postu-
rifonn stereom; lating a homology with the plates so called in

IBr, pnmibrach, .» . .111 ,- • i T j-

fasciculate ste- Echinoidea ; the old term patina is shorter and safer.
3™m&i/Ma, x "75. To understand the extension of the cup beyond
by w °i flcare the Patina> ifc is necessary first to study the Arms
penter!) Or Bracliia (Fig. XX.). These are rarely single.

The first step in advance is a bifurcation, con-
stantly repeated in a regular manner (regular dichotomy or
isotomy). Modifications of this occur through the suppression
of a bifurcation at definite points (irregular dichotomy or hetero-
tomy). In each half of the arm, the first branch on the right,

THE CRINOIDEA

then the first on the left, and so on, may be smaller than the other
branch. Thus there arises a main trunk giving off smaller branches

FIG. XX.

Specialisation of arm-branching. 1, a non-pinnulate, regularly dichotomous arm (isotomous) ;
2, a less regular dichotomous arm, a type common in Cyathocrinoidea ; 3, 4, two stages in
the evolution of unilateral heterotomy ; 5, 6, 7, three stages in the evolution of bilateral
heterotomy, culminating in pinnulation. For other types, see Fig. CX.

right and left alternately. These smaller branches may themselves
undergo a similar process, and so form armlets (ramuli) borne by
the main arm -branch
(minus). When the
ramules cease to branch
themselves, and are reg-
ularly placed on alter-
nate sides of successive
brachials of the main
branch, they are called
piunulae, and the arm is
" pinnulate." This pro-
cess of evolution has
been phylogenetically
traced in Botryocrinus
(Fig. XXL), while the
primitively dichotomous
origin of the pinnules
may be seen in the
developing Antedon.
The pinnule, as P. H.
Carpenter said, is an
arm in miniature ; it

rliflfovo in nrktViinrr Vmt bear ramuli, of which 'all except the proximal one are un-

QinerS in notnmg \ >Ut branched) and are almost regularly disposed. In B. pin-

Position from the Small nulatus the ramuli have become regular unbranched pin-

' . . , nules. (After Bather, 1891.)

end-branches of a simple

dichotomous arm ; but, in a pinnulate arm, it differs from
the ramus by the restriction to it of the fertile portions

8

FIQ. XXI.

Evolution of pinnules in Botryocrinus. 1, B. ramosus,
a species in which each arm has two rami bearing branched
ramuli ; 2, B. decadactylus, a species in which the two rami

ii4 THE CRINOIDEA

of the genital rachis. The pinnule differs in origin, and prob-
ably in structure and function, from the brachiole, which is an
independent exothecal process (see p. 41). Whether such struc-
tures as brachioles occur in Crinoidea is disputed. Wachsmuth &
Springer (1897) appear to regard pinnules as independent develop-
ments, rejecting the above theory of their origin. Jaekel (1894)
accepts the theory for most Inadunata and Neozoic crinoids, and
speaks of the organs as ramuli, restricting the term " pinnulae "
to the similar organs of the Palaeozoic Camerata (Cladocrinoidea,
Jaekel) and to the cystid brachioles, which he regards as homo-
logous. His view is admissible, but lacks proof.

In simple dichotomous arms, each brachial that supports two
branches does so by two upper sloping sides, or shoulders, of
equal size, each notched by the ventral groove, and pierced by the
axial canal, which branches with the arm. Such a brachial is
called axillare (ax), and all that part of the arm borne by any
single axillary is a " dichotom." The branches of the axial cord
are united at their bifurcation by a criss-cross of nerve-strands
(chiasma) serving to correlate their activities. As the size of one
half of a dichotom is reduced, the supporting shoulder of the
axillare is narrowed. Continuance of the process tends to bring
the wider shoulder more parallel to the under joint-surface of the
axillary. Thus a pinnulate arm of primitive structure consists
of a series of axillaries in which the alternate right and left
shoulders are wide and almost parallel, while the others are
greatly reduced and bear pinnules.

A pinnulate arm may consist of two rami, or each ramus may
bifurcate just as in a simple arm, though never to the same
extent. The axillaries on which the rami fork remain unmodified,
and with equal shoulders. It is convenient to distinguish these
as " main-axils " (Ax).

Owing to the great variation in the branching of the brachia,
it is extraordinarily difficult to devise a consistent terminology,
or to denote any particular ossicle in a concise and intelligible
manner (for fuller discussion, see Wachsmuth & Springer, 1897;
and Bather, Ann. Mag. Nat. Hist. Jan. 1892; and Geol. Mag.
July 1898). In a non-pinnulate dichotomous arm all brachials up
to and including the first axillary may be styled primilwachialia or
"primibrachs" (IBr), the axillary being distinguished as "primaxil "
(lax); the following Br in each branch are " secundibrachs " (IIBr),
with a " secundaxil " (Ilax) ; then succeed " tertibrachs " (IIIBr),
" quartibrachs " (IVBr), and so on. In a pinnulate dichotomous
arm the IBr do not as a rule bear pinnules, and are therefore
homologous with the IBr of a simple arm ; but of the next series
only the proximal brachial of each ramus is strictly homologous
with the IIBr of a simple arm, the pinnule borne by it, together

THE CRINOIDEA

with the next Br, representing the IIIBr of a simple arm. To use
the same terms is especially perplexing in intermediate forms ;
but solutions of the difficulty, though proposed, have not gained
general approval.

For descriptive purposes, a dichotomous arm is viewed from
the dorsal, i.e. aboral surface, and " right " and " left " equal right
and left of the observer. The mediad rami are called interior ;
those to the sides, exterior. A particular brachial in any series
is denoted by placing a small Arabic numeral after the symbol —
IIBr4, IVBr0, IIIBr5. The number of Br in a series may be
expressed either by stating it with a large Arabic numeral, e.g.
IIIBr, 7, and IIBr,10; or by giving the number of the axillare,
thus IIIaxT, and IIax10. The ossicles of the distal rami which do
not branch again are called " finials " (F).

The ambulacrals (Amb) l necessarily branch with the brachials,
and the several series or orders may be designated as lAmb,
IIAmb, etc. Their simplest form is that of a line of small
plates on either side the groove, capable of being raised or
depressed ; and when closed, meeting in the median line by a
zigzag suture due to their alternating arrangement. They vary
in size, both absolutely, and relatively to the brachials. Each
ambulacral may be divided by one or more tranverse sutures,
parallel to the long axis of the arm ; this produces the appearance
of short pinnules, for which these structures have been mistaken
by more than one author. The transverse sutures may come, to
lie at an angle, and the portions to alternate with one another.
Thus arose the side-plates or adambulacrals, which are a persistent
feature in many of the later crinoids (cf. Fig. IX. 2). There
may also be developed minute but distinct ossicles beneath the
outer covering-plates and alternating with them. The complicated
structure thus developed in Cyathocrinus and Gissocrinus has been
exquisitely worked out and illustrated by G-. Liljevall (Bather,
1893, pis. vii.-ix.).

Brachia, rami, ramuli, and pinnulae, in which the ossicles lie in
a single row, with more or less parallel joints, are termed " uni-
serial." Simple arms in Crinoidea are always uniserial. Pinnu-
late arms undergo a modification. Since in such arms the joints
slope alternately to right and left (p. 114), the brachials tend to
assume a wedge -shape; in process of growth of either the in-
dividual or the race, a complete wedge-shape is assumed, So that
the joint-lines between the ossicles form a " zigzag." Lastly, the
brachials come to lie in two alternating rows, in which case the arm
is termed "biserial." This development doubles the number of
pinnules in a given length of arm, and thus aids the collection

1 The terras "ambulacral" and " adambulacral" must not be held to imply any
homology with elements thus named in Stelleroidea and Echinoidea.

THE CRINOIDEA

Fir;. XXII.

Evolution from uniserial, through zigzag,
to biserial brachials.

of food. The axial cords and ventral groove at first swing from
side to side ; but this would be almost impossible in biserial arms,
so here a common straight ventral groove is formed, and the axial
cord lies at the bottom of it. The change from uniserial to

biserial, just as the evolution
of pinnules, begins in both
ontogeny and phylogeny at
the growing tip of the arm,
and proceeds gradually proxi-
mal wards (Fig. XXIL).

Still further develop-
ment occurs in the Camer-
ata. The adjacent right and
left ossicles of a biserial arm
may fuse, so as to form a compound brachial, and this necessarily
bears two pinnules, one on either side. Further than this, it appears
as though two or more compound brachials could fuse, and so form
a triply compound ossicle bearing two or three pinnules on either
side. At the same time, the pinnulars themselves may come to lie
in zigzag or biserial fashion, in the same way as do the ossicles of
so many cystid brachioles. It is this structure that influenced
Jaekel in his distinction between "pinnulae" and "ramuli" (supra) ;
but the facts are explicable as the final stage in a regular
evolution (Fig. LXXIX.).

Fusion of brachials either laterally, or in vertical series, or both,
may occur in any crinoid race in which it proves advantageous.
In some Gissocrini the IIBr, and possibly IIIBr, of each arm were
laterally united by suture ; in Orotalocrinus (Fig. XCII.) all brachials
of an arm are suturally united by projections at the distal margin
of each brachial ; in Petalocrinus (Fig. XCI.) all brachials of an arm
except IBr are fused into a single petaloid plate. Compare also
Melocrinus (Fig. LXXIV.) and Eucladocrinus (Fig. LXXI. 4).

Brachials primitively, and pinnulars nearly always, are
united by loose suture (compare Fig. XVIII.). The next stage is
imperforate articulation. In the final stage, perforate articulation
(Fig. XXIII. 1), there is a well-marked transverse fulcral ridge,
pierced by the axial canal ; the ventral groove comes nearly up to
the ridge at this point. On each side of the ventral groove, and
often separated by a slight vertical, i.e. dorso-ventral, ridge run-
ning down to the axial canal, are two depressions, fossae ; the ventral
pair lodges muscle fibres ("muscular fossae"); and the doisal
pair, interarticular ligament (" ligamentar fossae "). Dorsad of
the fulcrum is a deep "dorsal fossa," lodging elastic ligament.
This type may be modified by the disappearance of the ventral
muscles, the increase of the interarticular ligaments and their
fossae, and of the vertical ridge separating them, which now

THE CRINOIDEA 117

passes dorsad of the axial canal, and the concentration of the
dorsal ligament in a pit at the end of the vertical ridge (Fig.
XXIII. 2). This is called " trifascial articulation." Further
increase of the ligamentar fossae and of the vertical ridge,
with the disappearance of the dorsal fossa, produces the
" bifascial articulation," adapted only for lateral movement (Fig.
XXIII. 3). These forms of articulation may be bilaterally
symmetrical, but in pinnuliferous brachials the fulcral ridge is
skewed, so that on the distal joint-surface the end of the ridge
towards the pinnule is moved dorsalwards.

The syzygy (J. Miiller, 1841; P. H. Carpenter, 1884;
Bather, 1896) is an immovable sutural union between two brachials
of a pinnulate arm, accompanied with loss of the pinnule on the

Fio. XXIII.

Ann-joints. 1, brachial of Tsom'nMs nstfria, after Job. Miiller. 2, distal facn of IBr^ of
Jlfttkyi'/'iniis Aldrifhutnus (cf. (i in Fig. XVIII.), x 8 diam. 3, the same in Isocrinus Blakei, x 3J
(litiin. 4, syzygy of Jlhizncrinus liawsoni — a, epizygal from its under surface ; b, hypozygal
from its upper surface, X 7£ diam. 5, syzygy of Isocrinus Blakei — a, upper surface of hypozygal ;
/', epizygal ; and c, hypozygal in their relative positions, seen from side, x 3£ diam. (2-5 are
after P. H. Carpenter.)

w, axial canal ; ill, dorsal ligament fossa ; il, interarticular ligament fossa ; mf, musclft
fossa ; 21', facet for pinnule ; -eg, ventral groove.

hypozygal (compare Fig. XVIII.). Immobility may be effected
in various ways. The apposed faces may be smooth (some
Pentacrinids), striated (Uinfacrinus, most Antedonidae), or dotted
(some Adinoindrae) ; in Jlhizo&'inits a peg projects from the dorsal
region of the epizygal into a pit in the hypozygal (Fig. XXIII. 4),
and in some Pentacrinids a dorso-ventral ridge on the epizygal
fits into the hypozygal (Fig. XXIII. 5). The former type of
syzygy facilitates fracture along the suture, and is specially
developed in locomotive forms liable to entangle their arms.
The latter type appears different in origin and function.

We now return to the extension of the Dorsal Cup. This may
be effected, as in Ichthyocrinus (Fig. CVIII.), by the direct lateral
union of the proximal brachials. At the same time, the proximal
ambulacrals enter the tegmen, so that the thecal cavity stretches out
further between the actinal and abactinal elements. In many living
crinoids the proximal brachials are united by a flexible integument

THE CRINOIDEA

containing minute supplementary plates (Isocrinus, Calamocrinus) ;
thus the thecal cavity is enclosed by secondary as well as primary
elements. Similar plated membrane may occur between the IIBr

ilHBr

IHBr

Fia. XXIV.

Calamocrimis diometleae. 1, posterior view of cup, showing attachment of anal tube to
pinnules and proximal brachials, x f ; 2, radial and proximal brachials seen from inside of
cup, showing attachment of interbrachials (much enlarged) ; 3, a similar portion seen from the
side, showing the interbrachials (unshaded) attached to the brachials and pinnulars (shaded).
Enlarged. (All after Al. Agassiz.) As, anus ; Ji, basal ; IBr, primibrachs ; Uir, interbnichiuls ;•
•jin, pinnulars.

or IIIBr of a single arm. The small plates may increase in size and
firmly bind together the arms and rami (Fig. XXIVa) ; those
between brachia are "interbrachials" (iBr); between IIBr, "inter-

secundibrachs " (illBr), and so on.
Similarly there are interambula-
crals of various orders ; the ilAmb,
merging at the sides of the theca
into the iBr ; the illAmb separated
by the thecal cavity from illBr ;
the illlAmb, and so on. The
interbrachials sometimes, though
rarely, descend between the
radials. Not merely brachials,
but also pinnulars may be incor-
porated in the cup, and between
pinnules are developed " interpin-
nulars" (e.g. Uintacrinns,'Fig.ClII.)')
it is often hard to distinguish fixed
(From Brit. pinnu]ars fr0m supplementary
plates. Except in primitive forms
(e.g. Reteocrinidae), the interbrachials have a similar arrangement
in each interradius of an individual. Indeed, the arrangement often
serves as a means of distinguishing species and genera. This

illBr

FIG. XXIVo.
Sagenocriniis expansus, to show incorpora-

lettering see adjoining text.
Mus. specimen 57147.) x <j.

THE CR IN OWE A 119

regularity is, however, often modified in the posterior interradius,
which is widened by the insertion of "anal plates," so called
because they afford room for the anus, and are continuous with
the plates supporting the anal tube when that organ is present.

The Anal Plates of the Camerata appear as a median line
splitting the posterior interbrachials, and forming as it were a
sixth ray to the cup. They are rarely developed in forms in
which the anus is central or comparatively small; a slight
enlargement of the posterior calycal elements then sufficed. They
are, therefore, regarded as supplementary pieces developed as
occasion arose in the position
where they are found. The anal
tube is an outgrowth of the pos-
terior interambulacrum, and is, in YnLngtyjLjU. jRr
Reteocrinidae, Glyptocrinus, and sim-
ilar forms, supported by a dorsal
line of ridged plates continuous
with the anals (Fig. XXV.). The
ridge is connected with the ridges
that unite the posterior basal to FIO. xxv.

the right and left posterior radials, ciyptocrinu* deca<iactnins. i, ivom pos-
and this indicates that an axial
cord passed up it to govern the
motions of the tube. In later
Camerata, where the interradii and tegmen are less flexible, this
differentiation disappears.

The anals of the Inadunata and Flexibilia (Fig. XXVI.) have
been much discussed (see summary in Bather, 1890 and 1899 ; and
Wachsmuth <fe Springer, 1897). At least one of them, the
radianal (RA), is admittedly a primary, abactinal, radial element,
being in fact the modified lower half of the right posterior radial
(r. post. Hi). By the introduction of other plates, and notably
one special anal (x) into the posterior IR, the r. post. Rs is pushed
to the right, so that RA comes into contact with y, and helps in
the widening of the area and the support of the anal tube. The
theory, originated by Wachsmuth and Springer, that the tube is a
modified arm, has since been rejected by them. The less extreme
view that the dorsal median line of ossicles supporting the tube
represents the proximal left ramus of the right ^sterior arm, up
which the interambulacral peristome around the aiius gradually
stretched, has the following facts among others in its favour : —
The tube is admittedly in close connection with the right posterior
ray (Fig. XXVII.) ; it is up this side that the rectum passes ; in
locrinus, Merocriiuts, a-nd Castocriiius, and, to a less extent, in
Heterocrinus, Edenocrinus, and Ohiocriiius, the proximal plate of
this median row rests on r. post. R ; in locrinus the articulation

120

THE CRINOIDEA

between r. post. 11 and this proximal plate differs from that between
r. post. R & IBrj only in size (Fig. XXVIII.) ; in locrinus the
ventral groove of the median series coalesces in r. post. R with the
ventral groove of right posterior arm ; this, and other evidence
from Heterocrinidae, shows that the median anal series was in-
nervated from the axial cord of r. post. R. The only argument

4. Scopliiocriwts elcyans. 5. Cyfithocriniwlongiinanu*, 6. Pnnsomnns.

anal area.

FIG. XXVI.

Diagrams of the anal urea in various Inadunate Crinoids. (From Bather, after Waclisniiith
& Springer, Hall, and Angeliu.)

against the view, is the improbability of a change of function in
the ramus ; still the view is not proven.

A distinct question is whether the anal x, which frequently
occurs in the posterior interradius of the cup, is a secondary
element suddenly introduced, as are the anals of Camerata, or
whether it is the proximal median plate of the tube (as in locrinus,
Heterocrinidae, etc.), that has gradually sunk down into the cup.
Wachsmuth & Springer hold the former view, believing that
x is homologous with the strictly interradial anal of Camerata,

THE CRINOIDEA

121

and that the proximal median plate of locrinus and Heterocrinidae
(which they call t) is represented in genera with a special anal x
by a tube-plate (?•/) on the left shoulder of r. post. R (Fig. XXVI.
2-6) ; they assert that " in the earlier and simpler forms, the tube
consists of only five [vertical] series, one to each interradius, that
of the anal side resting upon t. Later on, as the tube grew larger,
a new row of plates was introduced with plate x supporting it.
When there are three series [at the posterior side], as in Dendro-
critius, the third generally rests upon one side of the left posterior
radial. The arrangement of the plates within the rows is so
regular, that if a sinking of the plate t had taken place, it would
certainly be indicated by some disturbance among the lower plates
in the tube." In this sentence the proximal median plate t (our x)

Anal area of Ilctcro-
rrinus isodactylus,
.showing close connec-
tion of anal (x) with
right posterior super-
radial (r.p.li). En-
larged. (From Bather.)

FIG. XXVIII.

locrinus, showing connection of anals .and brachials. 1,
part of the r. posterior ray seen from inside the cup ; "2,
upper articular surface of Hi ; the groove on the left goes to
x, that OTI the right to brachials ; 3, left upper articular
surface of Its, supporting x; 4, right upper articular sur-
face of Jin, supporting IKr\. Enlarged. (From Bather.)

of Ileterocrmus is identified with rt of Dendrocrinus ; but no proof
has yet been given that the added series may not be those starting
from rt and It, rather than from x and It. On the other hand, the
view that the series x is homologous with the series tf, is supported
by the general size and appearance of the two, and by the inferred
relations of the axial cords. And the homology of x with t is
supported by the facts that the position of t with reference to r.
post. Rs, does vary from a higher to a lower level in early genera,
while the position of x to the adjacent radials likewise varies.
In late Carboniferous genera of Dicyclic Inadunata, x certainly
appears to pass up out of the cup (Fig. XXIX.), and this inter-
pretation is confirmed by the migration of the anal in the develop-
ment of Aniedon, which anal is universally homologised with x
(Fig. XXX.) ; but if a plate can pass up, it can also pass down, as

122

THE CRINOIDEA

is further proved by the phylogeny of the Calceocrinidae. The
fact that x is wholly or partly in the cup, and t partly or wholly
outside, does not make them different morphological elements ; for
there is now admitted to be no difference between interambulacrals
and interbrachials, or between fixed and free brachials. Con-
sequently in this work the symbol x will always be applied to the

Fio. XXIX.

Upward iwssage of anal (x) in Ulowi niis. 1, anal area of " Ulocritnts" niairi; 2, posterior
view of U. Bnttxi; 3 and 4, U. Kansasensis, the cup from posterior and from above, x $•
(From Bather, after Miller & Gurley.)

proximal plate of the median line of the anal tube, whatever its
position.

Modification of the cup is not confined to the fixed brachials
and interbrachials, but also affects the patina. We have discussed
the disappearance of IBB. We have also to note a tendency to
fusion in the plates of the proximal circlet, whether IBB or BB,
and their change of shape due to the introduction of anals into the

3 4

Fio. XXX.

(After W. B. Carpenter and M. Sure.) Originating
. -"—,--- „ " > two small

fragments remain at the base of the anal tube (t). The dotted
gut. These figures also show change in shape of RR, and atrophy of orals (0).

Migration of the anal in Antedon.

between RR, the plate x gradually moves upwards^eventually atrophying till only two small

ed lines in 1 show the course of the

patina. The first stage is the fusion of one pair, producing 1
large and 3 small plates (Fig. XXXI. 2). This is almost entirely
restricted to monocyclic genera, where the plates that fuse are the
right and left anterior basals. Next comes the fusion of two pair,
producing 1 small and 2 large plates (Fig. XXXI. 3). This occurs
in both Mono- and Di-cyclica. In the former the small plate is
the left anterior basal, or rarely left posterior basal ; whereas
in Eublastoidea it is the right anterior basal (Fig. XXXI. 4).
In Dicyclica three infrabasals have been observed only among

THE CRINOIDEA

123

Inadunata and Flexibilia ; in the former group the small plate
is often, but not always, the anterior infrabasal (Fig. XXXI. 7) ;
in the latter it is (apnd W. and Sp.) always the right posterior
infrabasal (Fig. XXXI. 8). A bipartite base is formed only in a
few Monocyclica ; the two plates lie on the right and left sides of
the cup (Fig. XXXI. 9). Finally, all plates of the proximal circlet
may fuse into a solid ring, both in Mono- and Di-cyclica. The
infrabasals may fuse with the proximal columnal in Flexibilia,
thus forming a pseudomonocyclic type. The basals may be over-
grown by, and incorporated with, the radials, as in Eufjcniacrinus.
The symmetry of the base is modified by the presence of anals.
An anal resting on the basal circlet causes one of the basals to
double in width, so that the base becomes hexagonal instead of
pentagonal. Thus the quadripartite base comes to consist of a

FIG. XXXI.

Bases and their modifications. 1-0
and 9, inonocyclic ; 7 and 8, dicyclic ;
1-4, pentagonal, unaffected by anal ;
5, 6, 9, hexagonal, affected by anal.
In all the anal side is uppermost, and
the plates are numbered on Jaekel's
plan (see table, p. 110) ; the imagin-
ary additional piece is inarked + .
1, 5 BB ; 2,4 Un ; 3, 3 JiJf, Crinoid
type ; 4, 3 JiJS, Blastoid type ; 5,
4 J:B ; 6, 3 Bit ; 7, 3 IttB, as usual in
Dicyclica inadunata; 8, 3 I Jill, as
usual in Flexibilia imjrinnata ; 9,
2 JiJi. (Adapted from Wachsmuth &
Springer.)

posterior and anterior large plate, and two small lateral plates (Fig.
XXXI. 5). These tend to approximate in size. In Xenocrinus
(Fig. LXXVIIL), interbrachials as well as anals come down between
the radials, so that the basals are nearly equal in size, but irregular
in shape, and make the base decagonal. Removal of anals and
interbrachials from the radial circlet leaves a pentagonal quadri-
partite base, such as is found in Melocrinidae (p. 161). An anal
resting on a tripartite base is accompanied by increased width in
the small left anterior basal (Fig. XXXI. 6). But in the bipartite
base the small basal fuses with the combined posterior and left
posterior basals, while the combined right-hand basals increase in
width (Fig. XXXI. 9). In most Dicyclica the infrabasals do not
assume a hexagonal outline ; for the anals do not occur in the
basal circlet, but x truncates the upper surface of the posterior
basal. Exceptions are Sagenocrinus, Cardbocrinus (Fig. LXXXIV.),
Strophocrinus, and Thenarocrinus (Fig. XCVL).

124

THE CRINOIDEA

The enlargement of anal structures was not the only factor in
the modification of the typical pentamerism. Allusion need not
be made to the (apparently sudden) dropping of a radius to
form Tetracrinus (p. 153), or the duplication of the radii to form
Promachocrinus (p. 195), and similar cases. Nor need more be
said as to the enlargement of certain radii (e.g. Pigocrinws), the
bisection of others (e.g. Heterocrinus), and so forth, since in these
cases the outwardly symmetrical appearance of the cup usually
remained unaltered. But certain factors, probably of physical
environment, such as currents and direction of food-supply, or
possibly connected with locomotion, have at different times pro-
duced similar results in different families. A bending over of the
cup, accompanied by diminution of certain radials, was common
in Eugeniacrinidae, as well shown by Jaekel (1891). In the
remarkable Calceocrinidae the crown was bent towards the right
posterior interradius, and far-reaching changes brought about in
both cup and arms (p. 148). Even in an unattached species,
apparently of Agassizocrinus, similar growth of one side took place

at the expense of the other. These
cases are comparable to the irregu-
lar Eublastoids.

Concomitant with modifications
in the dorsal cup were modifica-
tions in the-^ Tegmen. Just as
brachials entered the cup, so their
covering-plates (Amb) entered the
tegmen, prolonging the food-
grooves over its surface. And cor-
responding to interbrachials in the
cup, there arose interambulacrals
(iAmb) in the tegmen (Fig.
XXXII.).

Other changes that took place
are difficult to describe without

Amb

JAmb

FIG. XXXII.

Tegmen of Marsipwritms nidiattt.*, show.

Wachsmuth & springer, isi»7.)

of the Plates ^vering the
mouth. In Antedoii five inter-
radial plates (0) are developed

before the radials and at the same time as the basals, upon which
they rest (Fig. XXXIII. 1). Between these two circlets appear
the radials, upon the shoulders of which the five adoral interradials
then rest (Fig. XXXIII. 2), forming a pyramid closed over the
oral centre, but soon opening at the apex to expose the entrance
to the mouth (tentacular vestibule). The posterior of these plates
surrounds the hydropore. At a more advanced stage they become
separated from the radials by ambulacrals and interambulacrals

THE CRINOIDEA

125

(Fig. XXXIII. 5), and finally, in most species of Antedon are
resorbed (cf. Fig. XXX. 4). These plates are called oralia (0).
Their prominence in early stages shows them to be primary
elements of the theca, probably well developed in the adult of

FIG. XXXIII.

Development of orals in Antedon. 1 (after Bury), 77? still visible, no RR yet formed. 2
(after Allman), 0 closed ; between them and BB are the developing RR. 3 (after Allman), 0 open,
exposing oral tentacles ; no arms yet exist. 4 (after W. B. Carpenter), 0 now separated from
BB by RR, which support arms. 5 (after W. B. Carpenter), arms cut off above first brachial so
as to show O, which now surround the mouth ; the shaded portion represents integument, in
which ambulacrals and interambulacrals are developed.

Note also gradual decrease in size of BB. A further stage in oral history is seen in Fig.
XXX. 4, which is from another species.

primitive forms. Five triangular plates that cover the mouth in
the recent Holopus, Hyocrinus, Rhizocrinus, and Thaumatocrinus,
are by all writers homologised with orals (Fig. XXXIV.). In
all Antedonidae, Bathycrinus, and Calamocrinus, they are almost

Amb

iAmb,

Fio. XXXIV.

Teginen of Holopus. The arms
are removed from the side nearest
the observer, showing the articu-
lar surfaces of R, the radials,
which are fused together. Amb,
ambulacrals, which pass down the
brachials to the tegmen and a little
way up between the orals ; iAmb,
interambulacrals, partly separat-
ing brachials from orals ; 0, five
perforate orals; Rp, processes of
fused RR. (After P. H. Carpenter,
1888.) x |.

Fia. XXXV.

Tegmen of Haplo-
crinv.t mespiliformis.
Br, first brachial; 0,
oral ; p, pore in post.
O ; R, radial. (After
Wachsmuth & Sprin-
ger, 1888.) X 6.

FIG. XXXVI.

Tegmen of Hybocriiius coni-
cus. Small irregular ambula-
crals overlie the apposed edges
of 0, the orals, which are
shaded. Post. 0 has a hydro-
pore, and is therefore a madre-
porite, M. As, anus, lies be-
tween this and x, the anal
plate of the cup. R, radials,
with arm -facets shaded, and
nerve channel black. (Based
partly on MS. drawings by W.
R. Billings. Natural size.)

or entirely resorbed in the adult. The ambulacra pass to the
mouth between or below the orals.

In Hybocrinus, Haplpcrinus, Carabocrinus, and other primitive
genera are five interradial plates precisely resembling the orals of
the larval Antedon in shape and position (Figs. XXXV., XXXVI.).

126 THE CRINOIDEA

The generally accepted view that these are orals is confirmed by
the frequent presence of a pore or pores in the posterior one, as in
larval Antedon and adult Hyocrinus, and by the situation of the
anus between this plate and the adjacent RR ; in Haplocrinus the
pore and the anus appear to be combined near the oral end of this
plate. These plates have in preceding pages been spoken of as del-
toids (A). They meet close around the mouth like the A of many
Blastoids, and in Hybocrinus and Carabocrinus they show traces of
hydrospires of Cadaster type. On the other hand, they are
homologised, and justly so, with five similar plates that occur in
CoccocrinuSy S'/mbatkocrinns, Pisocrinus, Allagecrinus, Myrtillocrinus,

some Platycrini, and the specimen of
Taxocrinus intermedius (Fig. XXXVII.)
described by Wachsmuth & Springer
(Nov. 1888). In most of these
genera the orals (or A) cover the
mouth, and the food -grooves pass in
under them; but in Taxocrinus the
mouth is open, and the grooves with
ambulacrals pass between the orals ;
while in Hybocrinus and Carabocrinus

FIG. xxxvn. the apposed orals, showing that the

Tegmcn of Taxocrinus intermulin*. grOOVCS Were actually above those
A, anal ridge ; 7Jr, edges of brachials ; °
iAmb, UIAmb, HIIAmb, interaiiibiiln- plates.

r'S™ SsT^i^'di^e^on; From this primitive Palaeozoic type,
the peristome. (Alter Wachsmuth & three lines of evolution start: (1)

bpnnger, 1686.) ALT J.T. j r A.

Ambulacra pass over the edges of the

orals, while ambulacrals and sometimes interambulacrals gradually
cover the orals, which seem thus to sink below the surface and to
diminish in size : the posterior oral, however, usually remains large
and is pierced by hydropores, while the increased size of the anal tube
pushes it more towards the oral centre (e.g. Euspirocrinus, Cyatho-
cnnus, Cupressacrinus, Figs. XXXVIII., XXXIX.). (2) Ambulacra
and ambulacrals pass between the orals, leaving an open mouth,
while the orals gradually atrophy (e.g. Taxocrinusy Fig. XXXVIL,
probably other Palaeozoic Flexibilia, and certainly many of their
Neozoic descendants). (3) Ambulacra pass beneath the orals,
and gradually also beneath other tegminal plates, which are
developed pan passu with the incorporation of brachials in the
cup, and which thus separate the orals from the periphery
of the tegmen (e.g. Adunata and Camerata, Fig. XL. ; cf. Caryo-
crinidae, p. 66).

At the same time, in types (1) and (3) a modification of the
ambulacrals takes place. The proximal ambulacrals covering the
mouth in (1) become large, and assume a pentagonal arrangement

THE CRINOIDEA

127

simulating that of primitive orals (e.g. Gissocrinus, Fig. XLL, Crotalo-
crinus, Fig. XCII. ; cf . Eublastoidea). Other ambulacrals, especially

FIG. XXXVIII.

Tegmen ofEuspirocrinvs spiralis, showing four cordi-
form deltoids or orals, and a mudreporito, with ambu-
lacrals overlying their apposed edges, lu the pos-
terior interradius is the base of the broken oft* anal
tube. (From Bather, 1893.) x 3.

Fio. XXXIX.

Teamen of Citathocrinus. 1, C.
plaints with ambulacrals and inter-
ambulacrals removed, exposing
app'-r.-j-J edges of orals (A) and
madieporite (M). 2, C. mammillae-is
with ambulacrals (rj>) and interam-
bulacrals (in) almost entirely cover-
ing orals (A) and peristome. (From
Bather, 1892.) x 2.

axillary Amb, increase in size, and form prominent bosses on the
tegmen, called "radial dome-plates" (Fig. XLIL); the effect of
this is enhanced by the sinking of the other ambulacrals (e.g.

Fio. XL.

Cylwocriniis nodosus, to exemplify the simplest
type of Camerate tegmen. 1, from 1. post, radius ;
2, from above. As, anus ; !•', basals ; br, opening
for food -groove and underlying canals; ib, large
interbrachials (suborals, Jaekel). Other letters as
usual. (Ai; ,;.,r Joh. Miiller, 1855.) x 2.

XLL

Tegmen of Gissocriniis arthriticus,
showing ambulacrals (Amb) passing
down arms and over apposed orals
(0),and becoming enlarged over peri-
stome. Post. O remains as a folded
madreporite; t, anal tube. Other
letters as usual. (From Brit. Mus.
46457.) X 3.

Actinocrinidae). These two facts suggest that the proximal dome-
plates of Camerata, regarded as orals by Wachsmuth & Springer,
and so quoted under head (3), may after all be ipr-^nsd ambulacrals.

128

THE CRINOIDEA

FJG. XLII.

Tegmen of
nodomui, sliowing
dome - plates, d.
th &

radial

The most conflicting views have been held from time to time

by the same and by different writers as to the homologies of these

plates. That here put forward agrees in the main with Neumayr's

(1889), but is based on facts not accessible to him. Wachsmuth
& Springer (1897) deny the homology of the
deltoids in Eublastoidea, Hybocrinus, and Cya-
thocrinidae, with the orals in Haplocrinus and
Antedon ; the plates here regarded as enlarged
ambulacrals (e.g. in Eublastoidea, Cyatho-
crinidae, Fig. XLIIL, Crotalocrinus) are taken
by them to be orals, and they imagine that
they undergo resorption, fission, and other
changes, stating that they are relatively larger
in young specimens. As to the origin and
(After homologies of the large interradial plates in
Spnnger, inac]unata (here called A or 0), those authors

are undecided.
The gradual sinking of the ambulacra and their covering-plates

below not only the orals but other tegminal plates, has given rise

in the typical Camerata to structures so differentiated that they

were long misunderstood, and their chief elucidator, Wachsmuth,

believed in 1877 that the tegmen of Palaeozoic crinoids was "a

solid vault or dome," which could not " in the remotest degree be

homologised with the soft

peristome of " recent crinoids.

" It forms," he said, " a part

of the abactinal system " ; "a

continuation of the radial and

interradial series of the dorsal

side, and serves merely as a

covering and protection for

the organs underneath."

From this it was generally

inferred that an originally

flexible tegmen (" disc " it was

called, as in recent crinoids)

had been overgrown by "a

free arch which braces the

entire oral side of the body

without the aid of oral plates" (W. & Sp. 1881). The

Fia. XLII I.

Tegmen of Cyathocrinus ramosns. The large
tegminal plates are not homologous with the del-
toids, but the squarish central one may be the
niadreporite. (From Bather, 1893.) x 3.

disc
and

remained as an "inner test," in which were ambulacra
possibly orals. Because of this structure, supposed to obtain to a
greater or less extent in all Palaeozoic crinoids, but not in their
successors, the Crinoidea were divided into Palaeocrinoidea and
Stomatocrinoidea, the latter term being altered by P. H. Carpenter
to Neocrinoidea.

THE CRINOIDEA 129

As fresh facts kept coming to light there was a good deal of
shifting of ground and mutual criticism on the part of Wachsmuth
& Springer and Carpenter. The supposed difference, and con-
sequently the classification, were rejected by Neumayr (1889) and
Bather (1889-90). Independently and synchronously Wachsmuth
& Springer (1889) concluded, chiefly on the evidence of Taxocrinus
intermedius (p. 126), that " in some Palaeozoic crinoids-the mouth is
exposed, and there is no vault aside of the orals " ; also that " all
attempts to subdivide the Crinoidea by separating the Palaeozoic
from the Mesozoic and later forms as natural divisions will have
to be abandoned." But it was not till 1891 that they published
their recantation of the view that " the Camerata had a vault and
a subtegminal disk."

The explanation of the Camerate tegmen given by Wachsmuth
& Springer in 1891 was readily accepted and now prevails. It
may be condensed as follows : — The plates of the tegmen were at
first small and yielding, as in the Ichthyocrinidae and in most
recent crinoids ; in this state when the arms are open the ventral
surface is depressed, when they are closed it bulges upwards. To
afford better protection to the viscera the tegminal plates became
more solid ; the tegmen being thus less flexible was fixed perforce
in its protruded state. The covering-plates of the ambulacra had
perhaps been closed from the begin-
ning, but as, through the upswelling of
the tegmen, the grooves were now more
exposed, further protection was needed.
Consequently they were lowered be-
neath the surface and, starting from
the solid orals, interambulacral plates
closed in over them. Certain of the
covering- plates, however, especially,
it would appear, the axillary pieces,
which perhaps could not so easily be
covered by other plates, became much
stouter, and were still exposed on the
surface as solid radial dome-plates. In FlG- XLIV-

any form highly developed along these
lines, e.g. Cactocrinus (Fig. XLIV.) the

food-grooves, Water-Vessels, and blood- one side of tegmen broken away,

i • i , v ,1 4.1.-. showing how the food-canals pass from

Vessels are SUnk right beneath the the arm - openings (Brr) under the

farrmon anrl ova onnlncArl in a tnViP tegmen to the upper end of the con-

tegmen, and are enclosed m a tuc , voiuted organ (after F. B. Meek, ms).
consisting of alternating ambulacrals x §. 2, one of the food-canals, from

., Till i • j i above, further magnified (after Meek).

above and adambulacrals or side-plates 3, the convoluted organ, from below.
below. The interambulacral plates of

the tegmen send curious extensions into the interior of the calyx, and
these extensions, spreading out, form what used to be regarded as

130 THE CRINOIDEA

a disc. We may, with Wachsmuth & Springer, regard the exten-
sions as caused by the perforation of the plates for water-canals ;
or we may regard them as simple processes for the purpose of
adding strength, without forgoing lightness, by a system of girders.
The supposition just quoted, as to the existence in Camerata
of a complicated water -vascular system, is supported by the

connection of the internal passages
with small pores near the arm-bases
(Fig. XLV.). Such have been ob-
served by Wachsmuth & Springer
in Actinocrinidae, Batocrinidae,
Rhodocrinidae, and Melocrinidae ;
l 2 they are placed in the cup-wall at

*"10' XLV the level of the tegmen, between the

Pores in Camerata. 1, DafotorflMU t ^-t - • i i •

Lyoni, an interrudins, showing siit-iike armsand their rami, and their canals

are separated from the subtegminal
arm-grooves by a thin partition.

(from Brit. Mns. specimen 755!»2). x t). In JlatOCnniLS, StrotOCnilUS, StegaUO-
.ecal cavity. ^^ Eudadocrinu^ and others in

which the arms branch off alternately, there is a pore to each ramus
that springs directly from the dorsal cup. Dolatocrinus may have
four to six in each interradius, and two to four between each IIBr
series. Other genera have only ten pores. In Gilbertsocrinus these
are at the end of long tubular extensions of the iriterradial areas
(Fig. CXXVIL). The facts are so plain, that the introduction of
water into the thecal cavity for aeration of the viscera seems prob-
able ; but the connection of these passages with the hydrocoel or
with branches thereof is a different question. The pores may pos-
sibly have replaced the hydropore or the madreporite of certain
Inadunata. In many recent crinoids pores pierce not only the 0.
but the iAmb, often in great numbers, being least numerous in
the posterior IR. Antedon bifida (Fig. XLVI.) is said to have 1500.
They may also occur on the edge of the theca between the arms.
In Actinometra they are chiefly developed near the ventral grooves,
and even on the pinnules. These pores communicate with the
coelom or its extensions (Fig. XLVIL), and so indirectly with the
water-ring. Where there are few pores (e.g. Rhizocrinus, Fig. X.),
a process (stone-canal) stretches out towards each from the ring ;
but, when numerous, there is no correspondence between stone-
canals and pores.

The statement has repeatedly been made (by Trautschold,
Lov^n, Wachsmuth & Springer) that pores occur on the suture-
lines between the plates composing the anal tube of many
Inadunata. In the cases to which the last-named authors now
restrict the statement, the tube-plates have strong axial folds, being
no doubt connected along these by thicker ligament, innervated from

THE CR1NOIDEA

the axial cords (Fig. XL VIII.). The depressions between these folds
are often deep, and it is in them that the pores are said to lie. It
is supposed that such genera have no madreporite, and that the
pores aerated the rectum or a blind extension thereof, for the

w-p

Flo. XLVI.

Ventral surface of Aiitedon bijulu
diagrammatised from various author!
ties, x 4 diam. As, anus ; 0, mouth
p, pinnules ; s, saeculi ; w.p, water
pores.

Vertical section through tegmen and under-
lying structures of A ntedon bijidu. Hhows four
pores, with ciliated funnels ((/), piercing the
integument (teg), and communicating by a
water-vessel (ice) with lacunar vessels (/) in the
connective tissue of the mesentery (m). The
funnels are cut through in different directions.
/, fibrous layer of integument ; c, coelomic
cavities ; j/, gut- wall, snowing outer fibrous
layer, epithelium, and inner cuticle. (After
Vogt & Jung, 188(3.) Greatly magnified.

anus often opens low down on the anterior face of the tube. The
statement has been definitely disproved for many forms hitherto
said to have such pores. But Wachsmuth & Springer (1897,
pi. vii., figs. 26, 5, 6, 9) support it by figures which, if correct and

FIG. XLVIII.

Structure of the anal tube in an Inadunate Crinoid, Mustigocriniis loreus. 1, plates in
normal position, from left edge of distal third of tube ; /, transverse folds connecting the main
axial ridges, r. 2, plates from the proximal third, disturbed and exposing the articular facets
(art) of the axial ridge. (After Bather, 1892.) x 10 diam.

correctly interpreted, prove it for some species up to the hilt —
and much further. For they show pores not only on the sutures,
but penetrating the plates ; not only in the interaxial depressions,
but on the axial folds ; not only in the tube, but in the dorsal
cup.

The last organ of which the modifications need be considered

I32

THE CRINOIDEA

is the Stem. The simplest form of columnal (after the fusion of
the pentameres) is circular, but with a tendency to pentagonal
outline. The joint-surface is radiately striated. The lumen may
be large and circular (Fig. XLIX. 1), or small, and circular or
five-rayed (Fig. XLIX. 2 and 3). The assumption of a pentagonal
outline is often accompanied by a restriction of the striation to

the margin and the concentration
of the longitudinal ligament-fibres
in five bands (Figs. XLIX. 4 and
CXI. 7-10). The joint-surfaces
may become elliptical, with a
fulcral ridge in the long diameter
and ligamentar fossae on either
side ; in this case the long diameter
of one end is set at an angle to
that of the other end, and the stem
thus gets a corkscrew twist (e.g.
PlatycrinuSy Fig. XLIX. 5), and so
can bend in any direction. In a
Russian Carboniferous Platycrinoid
the columnals are square in section,
and the ridges form diagonals at
right angles to one another. In
Bourgueticrinidae, Bathycrinus, and
Rhizocrinus this form of joint is
strongly marked, and the columnals

T»* are usua"y long and dice-box

MUS. 50978, x2-5; 5, riatyerinm, Brit, shaped (Fig. XLIX. 7): the length

Mus. 400; 6, PlatycH-Mts, showing twist v . e ,

of stem, Brit. MUS. 75!>o<5, x f ; 7, Rhizo- appears in some cases to be pro-

crinvs, from distal region of stem, x 8; ^,,,,^,1 KTT fnairm rvf twn nt n Q\r-7vo-v

8, Antolo* sam, larva, articular surface of dUCCd by lUSlOn OI tWO at a S} zygy.

one of the columnals seen from the side in The Stem of *he larval Antedon

9 (after M. Sars). Much enlarged. /T_. ArT Txr ... . . . , ,

(Fig. XLIX. 8, 9) has ossicles of the

Bourgueticrinus type and is very flexible. In two genera of
distinct origin — Herpetocrinus allied to Heterocrinus, and Campto-
crinus allied to Dichocrinus — the stem is rolled up round the crown
as shown in Fig. LIX. ; the cirri are, over the greater part of the
stem, confined to two rows along the sides of the ossicles and
directed towards the axis of the coil. In Herpetocrinus the ossicles
become hollowed towards the inside of the coil, and there is a
fulcral ridge parallel to this side ; strong (muscular ?) ligament?
were developed towards the outer margin. The stem could
uncoil and the crown be projected. The structure of the
columnals in Camptocrinus has not been described, but is said to
be similar. Wachsmuth & Springer say that such stems are
also found among allies of Poteriocrinus.

The axial canal, which in recent crinoids serves to transmit tne

Fio. XLIX.

THE CRINOIDEA

'33

vascular and nervous prolongations of the chambered organ and
axial organ to columnals and cirrals, may in some earlier forms
have served other purposes. The lumen is sometimes extra-
ordinarily wide (40 mm. in the root of Barycrinus, Fig. XVI. 3).
Pores sometimes appear to exist between or through the colum-
nals (e.g. Barycrinus, Crotalocrinus, Fig. L., Traumatocrinus). The
distal ends of the cirri sometimes appear to have been open, so
that the large axial canal communicated with the sea -water
(e.g. Barycrinus, Eucalyptocrinus, Cystocrinus, Fig. L. 2). The flat
under surface of encrusting roots is often ridged, as though
grooves put the axial canal in connection with the exterior (e.g.
Lichenocrinus, in which the upper surface is formed of poly-
gonal plates, supported beneath by numerous radiating lamellae).
These bases of attachment are, say Miller & Gurley, " as full of

FIG. L.

The development of "pores" from
cirri. 1, Crotulocrinus, portion of root,
with branching cirri below, and attach-
ments of cirri in upper part. These
latter show the axial canal that passes
from the main axial canal of the stem,
through the thickness of the colnmnals,
to each cirrus, and continues to the end
of the cirrus. Nat. size. 2, Cystncrinus
tennesseensis, part of stem, showing
stumpy aborted cirri, wit'.i axial canals
opening at their ends. Nat. size. 3,
Crotalocrinns, part of stem, showing
crenulate sutures between columnals,
and on the columnals the atrophied
attachments of cirri ;• compare with the
extreme upper part of tig. 1. x 5 diam.
4, Crotalncrimis, part of stem, showing
total disappearance of cirrus-attachment,
and only the axial canals remaining as
"pores" piercing the columnals. x 5
diam. (From Bather, 1898.)

pores as sponges." On the theory here adopted as to the origin
of the stem, a greater exten. n of the viscera into it in early
forms is probable ; the chambered organ itself may have been
placed some way down it (compare evolution of siphuncle from
visceral cone in Cephalopoda). Assuming that the soft structures
contained in the stem-lumen needed aeration, Wachsmuth &
Springer have supposed that streams of sea - water entered
by these pores. This suggestion seems no more happy than
Miller & Gurley's idea that "the mucous or fluid substance,
that contained the material for the base, passed through the
columnar canal into the pores of the base and was deposited in
a softer state than it afterward assumed.'' We may, however,
suppose that these passages served the double purpose of trans-
mitting nutrient fluid to the mesoderm cells depositing the outer
layers of stereom as the stem and root grew wider by concentric
accretion, and of aerating the same fluid by bringing it near the
oxygenated sea-water.

-C*

134

THE CRINOIDEA

In some genera, and especially, as Jaekel has suggested, in
those exposed to rough water or currents, the stem shortened con-
siderably while its attachment was preserved (e.g. Eugeniacrinidae,
Fig.CXX., Cupressocrinus). Cotylederma, Eudesicrinus, and Cyathidium
are fossil genera, Holopus (Fig. CXXL), a recent genus, in all which
the stem is reduced to a mere mass of stereom cementing the cup
to some solid object.

Although the Crinoidea are the Echinoderms in which the
Pelmatozoan habit has had most effect on the anatomy, yet they
present a constant tendency to relinquish the attached mode of
life and to lose that typical organ, the stem. So early as the
Ordovician, stems are found that during the life of the animal
were separated from the root, and became attached to other
objects, either by the remaining cirri, or
by winding around them. Often the
distal end of the stem formed a coiled
support like a serpent's tail (e.g. Acantho-
crinus rex, Jaekel, 1895). In some Silu-
rian genera (e.g. Calceocrinus, Mastigo-
crinus) stems have been described that
were rounded off at the distal end during
life. In Herpetocrinus the stem was rarely
if ever attached by anything except its
cirri ; while in the species described by
Hall as Brachiocrinus nodosarius, the stem
ends distally in a bulb. In the Devonian
Myrtillocrinus the stem ended in a four-
fluked grapnel ( = Ancyrocrinus, Fig. LI.).
Similar detachment took place in many
Carboniferous and Mesozoic crinoids ; the
recent Isocrinus ( = Pentacrinus) is known
to change its place, probably by swimming
with its arms, and the lower surface of
the distal columnal is "smoothed and
rounded" (Wyville Thomson, 1873).

Addiction to this habit led to the
gradual shortening of the stem ; in
stages have been described by P. H.
Carpenter (1882), from a stem of seventy columnals over 50
mm. long, down to a single ossicle, the proximale (Fig. LIL).
Continuance of this process led to the evolution, along many
different lines, of crinoids that are generally described as un-
stalked, and for which older writers were wont to erect an order,
Astylida. These fall into three groups : First, those in which a
portion of the stem remains, becoming compressed and fused, with or
without the infrabasals, into a cirrus-bearing compound ossicle, to

Fio. LI.

Grapnel of Myrtillocrinia (after
Hall).

Millericrinus Pratti all

THE CRINOIDEA

135

which the term " centro-dorsal " was originally applied, and to which
it must be restricted (e.g. Antedon, Eudiocrinus, Thaumatocrinus ; see
Figs. CXVII.-CXIX.). These forms anchor themselves by their
cirri, and though capable of crawling, climbing, and swimming, do
not often exercise their faculty of locomotion. Secondly, the group
in which either a portion of remaining stem, or the lower part of
the cup (i.e. BB or IBB), becomes solidified, usually by additional
deposition of stereom, into a knob, which, one may suppose, serves
as ballast or as a sea-anchor ; such forms are Agassizocrinus (p. 181),
Edriocrmns (Fig. CXIL), and Millericrinus Pratti (Fig. LIL). Both
of these groups have a small calycal cavity with thick walls, and
there can be little doubt but that all are attached by a stem in the
earlier stages of ontogeny. The third group, comprising Marsupites
(Fig. CIV.), Saccocoma (Fig. LXVIIL), and Uintacrinus (Fig. CIIL),

Fia. LIL

Stages in the loss of the stem by
Millericrinus Pratti. 1, cnp with stem 01
seventy coluinnals ( x 3). 2, distal end of a
stem, with apparent root (natural size).
3, cup with fairly long stem, with inter-
calated new coluninals (natural size). 4,
c.up with stem of twenty coluinnals (x :]).
5, cup with stem of live coluinnals ( x 2).
<i, lower part of crown, with stem reduced
to a pentagonal plate (c), with slight
trace of atrophied next columnal (x i).
7, base, closed below by a single plate
(c), with no trace of lumen or of other
coluinnals. This plate is the proximate
(p of 3 and 4), but is covered by second-
ary stereom (x •_>). (All after P. H. Car-
penter.)

has no trace of a stem or of any anchoring structure, but is in all
respects adapted for free locomotion ; the calycal cavity is large in
proportion to the thickness of the arms, and is enclosed by thin
flexible walls. Of^ these three genera, Saccocoma is the most special-
ised, and is supposed by Jaekel (1893) to have been pelagic, living
in swarms. Uintacrinus, with its extraordinarily long and movable
arms, may also have been pelagic. The genera of this third group,
although of origin as diverse as those in the other groups, resemble
one another in the presence of a central, pentagonal, apical plate.
This in Saccocoma may be the fused basals ; in Uintacrinus and Marsu-
pites it represents neither basals nor infrabasals, but may be the
proximale, or the supposed distal columnar plate ("dorso-central"),
or a new supplementary plate. It is safest to call it centrale.

Another curious modification, perhaps connected with a free-
floating existence, was presented by the root of Scyphocrinu*.
This swelled out into a hollow, chambered, balloon-like body,
referred by Barrande to an independent class of Echinoderms under

136 THE CRINOIDEA

the name Lobolithus, and described by Hall as a float, which he
called Camarocrinus.

As regards the internal organs of the crinoid not much can be
said. The most remarkable modifications are those affecting the
Gut. In most recent crinoids this makes a simple dextral coil
around the thecal cavity, from central mouth to eccentric anus.
The mouth may be slightly shifted anteriorly by increase in size
of the anus, or by the anal tube coming to occupy the centre of
the tegmen, as in Batocrinus, or even to pass beyond it towards
the anterior margin, as in Siphonocrinus (p. 199). But the mouth
remains in the axis of the coil, and such forms are called " endo-
cyclic." In Adinometra (p. 196) the gut winds in the same way,
but instead of issuing immediately the first coil is completed,
it continues to coil, not however around the axis of the
mouth but around the axis of the anus. The mouth, with its
annular accompaniments, therefore lies between the outer coil and
the next one, and not in the axis of the coil ; such a form is called
"exocyclic." This type of coiling does not correspond to the
two coils of the echinoid gut, since those are formed by a loop
returning on itself, in the way that any tube or cord fixed at the
extremities is necessarily lengthened. The coil of the gut in
Adinometra is therefore doubly peculiar. Yet in the number of
its coils it finds a parallel among the Camerata. In many of these
(e.g. Teleiocrinus, Cadocrinus, Batocrinus, Strotocrinus, Macrocrinus,
Eutrochocrinus, Habrocrinus, and Dimerocrinus) the gut seems to
have been supported by a loose, spicular calcification of the
connective tissue around the axial sinus, forming a " convoluted
organ " not unlike the shell of Bulla (Fig. XLIV. 3). Probably the
oesophagus passed down the hollow axis, then the gut coiled
dextrally in a widening spire, and the rectum passed up outside,
often along a thickened rim. The number of coils was at least three
in a Batocrinus figured by Wachsmuth and Springer (1897, pi. v.
fig. 6). It is remarkable that two of their figures (ib. figs. 5
and 7), if correctly described, show a sinistral coil. There is no
reason to suppose that the coil of the gut was ever other than
dextral in any class or order of Echinoderma, though Jaekel
(1897) has made an unconvincing attempt to prove that it was
sinistral in Camerata, Cystidea, and Blastoidea.

In Bathycrinus, Ehizocrinus, and the larval Anicdon, the mid-gut,
at the bottom of the thecal cavity, is widened into a stomach.
In Bathycrinus and Rhizocrinus are also interradial diverticula from
the outer side of the coil, supported by processes from the brachialg
(Fig. LIIL). Such diverticula were present in the Silurian Habro-
crinus, if the evidence of the convoluted organ and of Angelin
(1878, pi. xxvi. f. 12) can be relied on. In Pentacrinus and
Antedonidae, and to a less extent in Bathycrinus, the gut-wall on

THE CRINOIDEA

137

the inner side of the coil is thrown into folds or villi (Fig. LIV.).
In Adinometra, however, with its long coil, such plication is slight.
Sacculi (Fig. LV.) are structures confined to this class. They
are globular sacs surrounded by mesoderm, but lying close beneath
the epithelium, usually of the external surface. The lower wall
of each sacculus is clothed with rather large nucleated cells,
apparently derived from mesoderm, and from these grow up
processes filled with refringent granules of albuminoid substance.
Each process elongates and becomes attached to the upper (i.e.
outer) wall of the sacculus by a filament. The granular portion
of the cell may then separate from the nucleated base, and finally
may burst, setting free the granules. These granules, colourless
in life, are stained on death by the yellow pigment of the peri-
some, and show strong affinity for most staining reagents. The

Fio. LIII.

Section across the tliecal
cavity of IMhycrinus Al-
drichianus, at the level of
second primibrachs, show-
ing interradial processes of
stomach (*t), the rectum (r),
and the axial organ (cut).
(Diagrammatised from P.
H. Carpenter.) x 7 diam.

Fio. LIV.

Longitudinal section of
wall of anal tube of Antcdou
bifida, showing villi (v) of
rectal wall and their gland
cells (g). In the connective
tissue layer are plates of
stereoi n (si), and outside is
epithelium (e). (Diagram-
matised from Hamann.)
Greatly magnified.

Section of a nearly ripe sac-
culus of Antedon bifiiJu. int,
superficial layer of integument;
/, Hbrillar attachments of con-
tained cells to integument ; citf,
cutis surrounding sacculus, and
containing nuclei (w.v) ; g, granu-
lar masses of the contained cells ;
the large nuclei of these cells are
seen below in the lining epi-
thelium (cp) of the sacculus.
(After Cuenot.) x 200 diam.

sacculi occur chiefly at the edges of the food-grooves ; and the
side-plates, when highly developed, are notched for their reception
(as seen in Fig. IX. 2). They have also been observed in the
walls of the gut, in the mesenteries, and above the chambered
organ. They are developed so soon as the larva begins to feed.
Sacculi have been regarded as calcigenous glands (Wyville Thomson),
mucous glands (Bury), excretory organs (Ludwig), symbiotic algae,
" Zooxanthellae " (Vogt & Jung), and accumulations of reserve
material (Walther, and especially Cuenot, 1891). Since the com-
prehensive account by the last-named, little has been written on
the subject, and his view has found general favour. Sacculi occur
in Antedon, Promachocrinus, Eudiocrinus, Atelecrinus, Bhizocrinus,
Bathycrinus, and Pentacrinus ; they are certainly absent from Actino-
metra, and probably from all other recent genera. This fact renders
their occurrence of taxonomic value.

138 THE CRINOIDEA

THE CLASSIFICATION OF THE CRINOIDEA.

When the modifications above described have been grasped, when it
is remembered that these are only the more usual among the changes
that take place, and that there are others even more remarkable, and
when it is learned that most of these may affect members of any group
at any period, then it will be understood that the decipherment of the
few and fragmentary leaves of crinoid history that have been preserved
to us has been a long and difficult task, full of vain attempt and
rejected theory, and that classification after classification has been raised
but to fall ; and still the leading writers cannot agree, even provisionally.

An admirable account of the literature on this class, from Agricola,
in 1558, to C. F. Koemer, in 1853, was given by de Koninck and Lc
Hon (1854), and supplemented by W. B. Carpenter (1866). The
nomenclature of genera and species dates, of course, from 1758, the year
of publication of the tenth edition of Linnaeus's Systema Naturae ; but
neither Linnaeus nor his immediate successors were more happy in their
dealings with this then little known group than more ancient authors.
It was J. S. Miller of Dantzig and Bristol, who, in 1821, laid the
foundation for a scientific knowledge and classification of the CRINOIDEA,
as he was the first to name them. Accounts of the subsequent growth of
knowledge and theory are so accessible in Zittel (1879, 1899), P. H.
Carpenter (1884), and Wachsmuth and Springer (1897), that only the-
main stages need recalling, and that briefly.

Miller's CRINOIDEA excludes unstalked genera ; the others known to
him are divided into : — ARTICULATA : ossicula forming the cup articula-
ting, Apiocrinus, Pentacrimis, Encrinus. SEMI- ARTICULATA : plates of cup
articulating imperfectly, Poteriocrinus. INARTICULATA : plates adhering by
sutures, lined by muscular integument, Cyathocrinus, Adinocrinus, Rhodo-
crinus, Platycrinus. COADUNATA : proximal ossicles of cup anchylosed to
proximal columnal, Eugeniacrinus. The principles of classification here
adopted profoundly influenced subsequent attempts, while the genera
form the types of modern families.

Joh. Miiller (1843) meant by "Crinoidea" all Pelmatozoa, dis
tinguishing the crinoids proper as Crinoidea brachiata. Among these
he retained Miller's ARTICULATA, adding to it the Antedonidae, and
stating that the rays developed from the base of the cup, and merged
into the free arms ; that the proximal plates of the rays were laterally
united by an integument continuous with that of the ventral surface ;
that the radial and first primibrach, the primaxil and first secundi-
brach, were joined by muscles, but the first and second primibrachs by
muscles or syzygy ; and that food - grooves, mouth, and anus were
visible on the tegmen. The peculiar genus Saccocoma, unknown
to Miller, was made the type of the COSTATA : stemless, without centro-
dorsal, and with processes from the brachials (pinnulae oppositae, Miiller).
Haplocrinus mespiliformis was separated under the head TESTACEA : cup
and tegmen forming a firm, connected test, with five ambulacra running
up to the mouth. Holopiis was regarded as an entirely independent and
peculiar division of the Crinoidea (sensu lato), on account of its sessile cup

THE CRINOIDEA 139

and the supposed absence of an anus ; but no group name was proposed.
All other crinoids then knewn were placed in the TESSELLATA, which
included the unstalked Marsupites ; their cup was said to be composed
entirely of plates, to which names were given (BB, RR, Ax, etc.), and
their tegmen was said to be solid, with only one opening, and with no
food-grooves. This classification, while retaining as a guide the mode of
union of the plates, took also into consideration the structure of the tegmen.

Zittel (1879) divided the Crinoidea brachiata, EUCRIXOIDEA, as he
called them, into three sub-orders, on the basis of Miiller's classification,
but merging Holopus and the Testacea in the Articulata. TESSELATA :
cup-plates thin, immovably united by simple suture ; tegmen solidly
plated ; mouth subtegminal ; Marsupites] Uintacrinus, and all Palaeozoic
crinoids. ARTICULATA : cup-plates usually very thick, united by articu-
lating or plane sutures ; tegmen integumentary, rarely plated, with open
food-grooves and central mouth ; all Neozoic forms, except those here
mentioned under other sub-orders. COSTATA : Saccocoma (see under
Miiller). By this time several families had been founded, notably by
C. F. Roemer (1855) and Angelin (1878). These were added to by
Zittel, and those of the Tesselata arranged in groups, chiefly according
to the construction of the tegmen. Most of the families were well
founded, but the characters of the Tesselata and Articulata, though
applicable to a few genera, were not really capable of extension to all the
forms that had become known since the time of Miiller.

Wachsmuth:s establishment of the PALAEOCRINOIDEA in 1877, "to
include those forms in which the disc is roofed by a second integument,
which he supposed to exist in all Palaeozoic crinoids " (W. & Sp.),
has already been noticed. The order covered nearly the same ground
as the Tessellata, and opposed to it was the order Stomatocrinoidea
or NEOCRINOIDEA, corresponding roughly to the Articulata. Carpenter
& Etheridge (1881) accepted the division, but laid more stress on the
asymmetry of the posterior interradius in Palaeocrinoidea, and therefore
suggested IRREGULARIA and REGDLARIA.

Wachsmuth & Springer (1885) divided their Palaeocrinoidea into
three sub-orders, originally suggested by Wachsmuth's study of the tegmen.
CAMERATA : tegmen rigid, formed of rather large thick plates ; brachia
in part rigidly incorporated in cup by means of interbrachials ; Reteo-
crinidae, Rhodocrinidae, Thysanocrinidae, Glyptasteridae, Melocrinidae,
Actinocrinidae, Platycrinidae, Hexacrinidae, Eucalyptocrinidae, Barrandeo-
crinidae, and Acrocrinidae. ARTICULATA : tegrnen flexible, formed of
minute plates ; brachia may or may not be partly incorporated in cup,
but not rigidly ; Ichthyocrinidae, Crotalocrinidae.1 INADUNATA : brachia
not incorporated in cup. These were divided into LARVIFORMIA :
tegmen of few plates; "disc" covered by "vault"; Haplocrinidae.
Cupressocrinidae, Gasterocomidae,2 Stephanocrinidae ; and FISTULATA :
"disc "not entirely covered by "vault," but passes out as a porous
" ventral sac " ; Hybocrinidae, Heterocrinidae, Anomalocrinidae, Belemno-
crinidae, Cyathocrinidae, Calceocrinidae, Catillocrinidae, Poteriocrinidae,
Encrinidae, Astylocrinidae.

1 Removed to Camerata in 1888. 2 Removed to Fistulata in 1890.

140 THE CRINOIDEA

It must always have been obvious that the Neocrinoidea were a
polyphyletic group derived from the Palaeocrinoidea ; the difficulty has
been to trace the relationship. This task, however, was forced upon us
when those orders were finally rejected. Since that rejection did not
carry with it the overthrow of Wachsmuth and Springer's sub-orders, the
practical result was the raising of them to the rank of orders, in which
Neozoic crinoids had to be appropriately placed.

P. H. Carpenter (1889) referred all Neozoic crinoids to the Articu-
lata (W. and Sp.) as a sub-order, PINNATA, with pinnules ; while the
Ichthyocrinidae constituted a sub-order, IMPINNATA, without pinnules.
This had the advantage of making Articulata, W. & Sp., very nearly
the same as Articulata, Miiller. But there is reason to believe that
the Pentacrinidae are descended directly from the dicyclic Inadunata ;
and since Pentacrinus (i.e. Isocrinus) was Miiller's type, one can hardly
escape confusion in using the term Articulata for a group that
excludes the Pentacrinidae. FLEXIBILIA, Zittel (1895), is superior
and prior to Wachsmuth & Springer's proposed substitute ARTICULOSA
(1897), which, moreover, was used in a different sense by Jaekel (1894).

One reason for the above reference of the Pentacrinidae is the
discovery by Wachsmuth & Springer (1897) that in the Flexibilia the
top columnal is not the latest formed, but a persistent proximale, usually
fused with the infrabasals ; whereas in Pentacrinidae, as in Camerata and
Inadunata, the top columnal is merely the latest formed, and continually
moves from its proximal position as new columnals develop. This, along
with the other characters, seems to confirm the independent nature of the
order Flexibilia ; at the same time, the resemblance of early genera
to contemporaneous Inadunata is so striking, that one must suppose the
Flexibilia Irnpinnata to be derived from non-pinnulate dicyclic Inadunata.
Then the want of links between Impinnata and Pinnata suggests that
the process may have been repeated, and that Pinnata were derived
from Triassic pinnulate Inadunata (Fistulata, W. & Sp.). Whether the
mode of stem-growth indicates affinity or parallel modification is a point
that demands investigation.

The Larviformia of Wachsmuth & Springer are generally accepted as
the most primitive Criiioidea, and as representing the ancestral type of
all the orders. On this ground the group might be retained, not
as a sub-order of Inadunata, but as an Urgruppe, or an independent
order (Zittel, 1895). But their geological age forbids us to regard the
known genera and species as themselves ancestral to far older forms. If
then they be placed with other Inadunata, the question arises whether
the distinction of the dicyclic and rnonocyclic base is not more funda-
mental than the varying development of the tegmen. A complex tegmen
is a development from a simple one, but between dicyclic and monocyclic
the barrier runs back to archaean ignorance (Bather, 1893).

The Camerata, like the Flexibilia, form an order fairly well defined
on a morphological basis. But here too there are grounds for suggesting
that the modifications may have occurred more than once. The evolution
of pinnules, of biserial arms, of fixed brachials, interbrachials, and the
like, even of a solid tegmen, may all be traced among Inadunata (e.g.

THE CRINOIDEA 141

Jjotryocriniis, Encrinus, Uintacrinus, Crotalocri?ius, Cyathocrinus). The
Reteocrinidae (W. & Sp.) present us with the Articulate or Flexible stage
of the Camerata, and include both Dicyclica and Monocyclica. It is
reasonable to suppose that the monocyclic Melocrinidae, Calyptocrinidae,
Batocrinidae, and Actinocrinidae were derived from monocyclic ancestors,
whether Reteocrinidae or others, and that the dicyclic Thysanocrinidae
and Rhodocrinidae were derived from dicyclic Reteocrinidae and similar
forms. On the other hand, the Platycrinidae, and their descendants,
Hexucrinidae and Acrocrinidae, are closer to the Inadunata, and sprang
probably from monocyclic genera, independently and at a later period.
The Crotalocrinidae, which Wachsnmth and Springer now place in the
Camerata, are at any rate of totally different origin from all other Camerata,
and are intimately allied to Cyathocrinus and Gissocrinus among the
dicyclic Inadunata.

Previous authors have attempted to make their classification serve as
a key to structure rather than as an epitome of descent. And the above
considerations suggest that Wachsmuth & Springer's system, though far
the best from an anatomical standpoint, is not the phylogenetic classifica-
tion sought by the modern biologist.

A bold attempt at a phylogenetic classification is shortly to be brought
out by Jaekel. From his preliminary notice (1894) and subsequent
writings it appears that he separates the Camerata (W. & Sp.), under the
name CLADOIDEA, from the rest, to which he restricts the name CRINOIDEA.
The Crinoidea have radials, five arms with brachials, and ramuli (Jaekel,
supra), and anal plates connected with right posterior radial ; their origin
is from such a simple type as has been described above. The Cladoidea
originated independently from a many-plated cystid, in which numerous
arms produced as many rows of supporting plates (costalia) in the cup, not
homologous with radials and brachials ; the free costalia bear pinnulae
(Jaekel, supra). The resemblances between Cladoidea and Crinoidea are
admittedly great, but all to be explained comfortably by homoplasy and con-
vergence. The Crinoidea (Jaekel) are divided into : — FISTULATA : the root-
group, with primitive, simple calyx ; including the generally accepted genera,
also Porocrinus, Crotalocrinidae, and Marsupites. LARVATA : essentially
the same as Larviformia (W. & Sp.), but supposed to be derived from
Fistulata, perhaps through Heterocrinidae. COSTATA : " arms give off
undivided, alternating side-branches, serving in part for reception of
gonads ; cup usually thin-walled and spacious, composed of a circlet of
large RR and a tripartite or fused base. No anal plates or tube ; tegmen
simple, of five 0, to which suboralia l may be added " ; Hybocrinidae, Hapa-
locrinidae, Plicatocrinidae, Hyocrinidae, Saccocomidae, Rhizocrinidae (?).
ARTICULOSA : the old Articulata, W. & Sp. non Miiller ; Lecanocrinidae,
Ichthyocrinidae, Taxocrinidae. ARTICULATA, Miiller non W. & Sp. :
derived from Fistulata not from Articulosa ; Encrinidae, Pentacrinidae,
Apiocrinidae, Eugeniacrinidae, Antedonidae. JaekePs proposals are here
alluded to in order that the student may understand the terms used in
his forthcoming finely illustrated work. They are valuable as suggesting
research for a large amount of confirmatory evidence.

1 "Die Oralia mit den Radialien verbindenden Plattchen."

142

THE CRINO1DEA

The Classification here adopted keeps accepted terms so far as
possible, but the distribution of the groups is very different from that
hitherto accepted. Dividing all Crinoidea into MONOCYCLICA and
DICYCLICA, we trace in each order a gradual and to some extent parallel
modification, here and there diverging in somewhat similar directions.
Thus the simplest forms in each order are INADUNATA, with free distinct
arms, and pass from a Larviform stage, with simple tegmen, to a Fistulate
stage, with more complex anal tube and tegmen. At an early period
(? Cambrian) in the history of the Monocyclica, the Camerate modification,
viz. rigid incorporation of brachials in cup and anibulacrals in tegmen,
'fleeted a few forms, and thus arose MONOCYCLICA CAMERATA. At a later
p *od (Silurian) was a repetition of this modification, but one affecting the

MONOCYCLICA. DICYCLICA

INADUNATA. INADUNATA.

k

Cambrian. (LARVIFORMIA).

Ordovician.

Silurian.

Devonian.

Carboniferous.

Permian.

*^*+ CAMERATA. ^^ A

ADUNATA. B\ FLEXIBILIA.

A A I IMPINNATA

\ M (DISTINCT A)

(*?* •

** PINNATA.

(ARTICULATA).

Fio. LVI. •

ipposed Relations of the Orders and •

Sub-Orders of Crinoidea. ^L

CreUceoufc Fio. LVI.

Tertiary Supposed Relations of the Orders and

Sub-Orders of Crinoidea.
Recent

cup to a far less extent, and resulting chiefly in a solid tegmen and biserial
arms ; thus arose the MONOCYCLICA ADUNATA (or Platycrinoidea), which
even Wachsmuth and Springer find a difficulty in placing with the
Camerata. These two highly specialised branches died out before the
close of the Palaeozoic epoch, the Adunata outliving the Carnerata ; but
the simpler Inadunate forms continued, and reached a high degree of
specialisation in their Jurassic descendants, to which the living Hyocrinus
is closely related. The Dicyclica Inadunata similarly gave off the
DICYCLICA CAMERATA, which persisted only a little less long than their
monocyclic convergents. The dicyclic Crotalocrinidae of the Silurian
are curiously parallel to the Monocyclica Adunata, but it is not worth
while to separate them from the typical Inadunata. About the same time
arose among the Dicyclic Inadunata the modification that resulted in
the FLEXIBILIA, with brachials loosely incorporated in dorsal cup. The

THE CR1NO1DEA

143

Dicyclic Inadunata caiue to their acme in Carboniferous times, and only
those forms persisted to Neozoic and Recent periods which assumed an
Articulate modification, viz. a loose lateral union of proximal brachials,
as seen in Pentacrinidae, which are convergents of the Neozoic Flexibilia.
The latter sub -order was represented during Palaeozoic times by the
Ichthyocrinoidea (Impinnata) ; between them and the Neozoic Apio-
crinidae, Bourgueticrinidae, etc. (Pinnata), the links are missing, but may
yet be found among Permian and Triassic crinoids (cf. p. 140). At any
rate, the Neozoic Flexibilia, when they assumed the free-swimming habit,
took a new lease of life and have their acme in our own day.

The systematic account of the Crinoidea may be prefaced .by a table
of the terms and symbols employed in technical description, as explained
for the most part in the preceding pages.

PLATES OF PATINA ..... Seepage 112

Infrabasal . . .IB . . ,,99

Basal . . . B , . . „ 99

Superbasal . . SB ., ,, 159

Radial. . . . R

Inferradial Ri

Superradial . . R« . •.

Radianal . . RA .

Pararadial . PR

BRACHIALS . . free, Br ; fixed, Br .

Primibrachs . . ,, IBr ,. IBr .

First primibrach . IBr,

Second primibrach . IBr2 i ;

Primaxil . ?. ,,.• . I Ax

Secundibrachs . . „ IIBr ,, IIBr .

Secundaxil . . IIAx

Tertibrachs . . . IIIBr

Quartibrachs . . IVBr

and so on to the

Finials . F

AMBULACRALS . . . Amb or c. p.

Primambulacs . . I Amb

Secundambulacs . . IIAmb .

and so on.
INTERRADIAL AND SUPPLEMENTARY

ELEMENTS

Ai^ interradius . . IR

Interbrachials . . . jBr

Intersecundibrachs . illBr

ana so on.
Special anal plate (or proximal

median plate of anal tube) . x
Proximal tube -plate on right . rt
Corresponding plate on left . It
Deltoids . . .A

here regarded as homologous with
Orals O

Interambulacrals . . iAmb

Interprimambulacs . ilAmb

Intersecundambulacs . illAmb .

The various radii and interradii are denoted as explained in the table
on p. 110.

112

112

119

150

100, 112

114

115

115

114

114

114

114

114

115

100, 115
115
115

109

20

118

118

119
121
121
100

124
118
118
118

144 THE CRINOIDEA

We now proceed to review the sub-classes, orders, and families
of Crinoidea.

SUB-GLASS 1. MONOCYCLICA, Bather (1899).

Crinoidea in which the base consists of BB only, the aboral prolonga-
tions of the chambered organ being interradial ; new columnals are
introduced at the extreme proximal end of the stem.

ORDER l. Monocyclica Inadunata

( = INADUNATA, W. & Sp. pars, emend.)

Monocyclica in which the dorsal cup is confined to the patina and
occasional intercalated anals ; such Amb or iAmb as enter the tegmen
remain supra-tegminal and not rigidly united.

Thirty -one genera are here referred to this order. Of these, 20
diverge from the normal symmetry to a greater extent than by the intro-
duction of anals, viz. 10 through the horizontal bisection of certain RR,
other than r. post. R, usually r. and 1. ant. RR, while the remaining RR
often increase in width ; 8 through such increase in width of certain RR
(usually 1. post. R and ant. R), often accompanied by variation in the
number of arms directly springing from RR ; 1 by disappearance of a
radial (an occurrence also found in some of the other genera), and apparent
increase in number of arms springing from RR (as in some other genera).
Only 2 of the genera have regularly pinnulate arms, and in 1 5 the arms
are unbranched. In 14, 0 are not separated from one another or from
RR ; 1 1 at least have no anal tube, while in 6 of those the anus if it
existed must have pierced post. 0. In the rest the anal tube was always
supported by a well-defined median ridge of plates, simulating Br. In
Dicyclica Inadunata such characters as these are confined to very few
forms, and those the oldest. Therefore, as well as on anatomical grounds,
we infer that the prevalence of these characters denotes a primitive order.

There is considerable parallelism of development between early In-
adunate forms of Dicyclica and Monocyclica ; nevertheless, most of the
Monocyclic genera fall into clearly marked groups, with which no Dicyclica
can be confused. The Hybocrinidae and Stephanocrinus are here taken
first, because of their simple structure and resemblance to Eublastoidea.
At a very early period the Heterocrinid type must have arisen, through
horizontal bisection of RR (antea, p. 112) and branching of arms. locrinus
and Anomalocrinus may be early offshoots from that line of descent ;
Herpetocrinus and the Calceocrinidae are other remarkably specialised off-
shoots, the latter surviving to Carboniferous times. Starting afresh from
the Hybocrinidae, the exaggeration of size in certain RR became more
pronounced in the Pisocrinidae, which, by addition of supplementary
radials and arms, lead up through Calycanthocrinus and Mycocrinus to
Catillocrinus, a type presenting a strange convergence to Halysiocrinus.
Zophocrinus, Allagecrinus, and certainly Haplocrinus appear to be aberrant
from this line of descent. The more regular Symbathocrinidae, with
their simple arms, represent either the direct descenuants of the Hybo-
crinidae or a return from the Pisocrinidae to a more nor"mi structure.

THE CRINOIDEA 145

The Carboniferous Belemnocrinus simulates many Dicyclica in the com-
plexity of its anal tube, but in arm-structure may be regarded as inter-
mediate between the Heterocrinidae and certain Neozoic Monocyclica.
Of these latter, the first to appear are the Plicatocrinidae, and from them
spring the Jurassic Saccocoma and the Recent Hyocrinus.

FAMILY 1. HYBOCRINIDAE. Monocyclica Inadunata, in which RR differ
but slightly in size and shape, except r. post. R, which supports on its
left shoulder the proximal plate (x) of the rudimentary anal tube, when
present, while its right shoulder is separated by a slanting suture from
the lower part of the plate and bears an arm. BB, 5. Arms, 3-5,
unbranched, uniserial, nou-pinnulate, less wide than RR. O, so far as
known, large, with traces of hydrospires, not separated from RR by
supplementary plates ; post. O cut into by anus and pierced by hydropore.
Amb rest on adjacent edges of O. Genera — Hybocystis, Wetherby (1880,
see p. 95, Fig. I.) ; Hybocrimis, Billings (185 7, -59 ; syn. Indianocrinus,
Mill. & Gurl.), Ordovician and Silurian, N. America (Figs. LVII. 3, VI.
and XXXVI.) ; Hoplocrinus, Grewingk (1867), Ordovician, Russia, includes
Baerocrinus (Fig. LVII. 1, 2, 4). For chief literature see W. & Sp. (91).

Fio. LVII.

Hybocrinidae. 1, Hoplocrinus dipentas (type) ; 2, Hoplocrinus dipentas (var.) ; 3, Hjlocrinut
tuinidus; 4, " Baerocrinus " Ungerni.

FAMILY 2. STEPHANOCRINIDAE. Monocyclica Inadunata, with 5 equal RR,
each bearing an arm in a deep radial sinus. BB, 3, the unfused B being
usually r. ant. as in Blastoidea. Arms branched. 0 large, with traces of
hydrospires (?) at their junction with RR ; anus at junction of post. 0 with
RR, with minute valvular covering. Amb, often fused, rest on adjacent
edges of 0, and 5 large ones, often fused, cover the peristome. Genus —
Stephanocrinus, Conrad (1842, see p. 96, Fig. II.), Silurian, N. America
and Britain, and as " Rhombifera mira" in Bohemia. The arms are
said to branch, but no two observers agree as to the mode of branching,
and in fact it seems to differ in the various ambulacra of the few
specimens that preserve the arms (R. P. Whitfield in litt.).

FAMILY 3. HETEROCRINIDAE. Monocyclica Inadunata, with x usually
resting on left shoulder of r. post. Rs, and partly on right shoulder of
1. post. R, but sometimes sunk between the two, and always supporting a
long anal tube. BB, 5. Arms, 5, with proximal Br usually full width of
RR ; isotomous or heterotomous, or bifurcate with armlets ; non-pinnu-
late. 5 O, without perisomic plates, have been observed in Heterocrinus
heterodactylus juvenis, but the relations of Amb to tegmen are unknown.
Genera— locrinus, Hall (1866), Ordovician, N. America (Figs. XXVI. 1 ;
XXVIII. ; and LVIII. 1) ; primitive in its isotomous arms, and RR similar
in shape and undivided,, except r. post. R, from which an upper portion,
in shape like an axillare, is separated by a horizontal suture, and bears an
Arm on its r. shoulder, and on its 1. shoulder a line of ossicles supporting

10

146

THE CRINOIDEA

an anal tube. Therefore for this line of descent, such a position of the
median dorsal rib of the anal tube is considered primitive (among Dicyclica
a similar stage is presented by Merocrinus). Anal tube somewhat com-
plicated. Stem markedly pentagonal, with 5 interradial sutures. Hetero-
crinus, Hall (1843, em. S. A. Miller, 1889, syn. Stenocrinus, W. & Sp.) ;
Ectenocrinus, Miller (1889); and Ohiocrinus, W. & Sp. (1886), are best
known from Ordovician, N. America, but certainly had representatives
in Europe, where also they were preceded in Cambrian seas by " Dendro-
crinus cambriensis" Hicks (1873), which seems allied to locrinus. They
agree in the transverse bisection of r. and 1. ant. RR, in addition to
r. post. R (exceptionally 1. ant. R is simple, and ant. R may be compound in
its stead) ; in the partial entry of the proximal plate of the anal tube into
the cup, sinde it now rests partly on 1. post. R, though more intimately
connected with r. post. R* (Fig. XXVII.) ; in the departure from isotomy
in the direction of two rami with armlets. In Heterocrinus, Ohiocrinus,

P\

FIG. LVIII.

Heterocrinidae. 1, locrinits; 2, Heterocrinus bellevillensis ; 3, Ectcnocrinus; 4, Anor>ialo-
crinus.

and locrinus the stem is pentagonal and quinquepartite ; in Ectenocrinus
almost circular and tripartite. The anal tube is straight and narrow
in Heterocriniis and Ectenocrinus, spirally coiled in Ohiocrinus (cf.
Botryocrinus, Streptocrinus). Ectenocrinus has almost reached a pinnulate
stage of arm -branching, since the Br form syzygial pairs. Anomalo-
cri?ius, Meek & Worthen (1885 ; syn. Ataxocrinus, Lyon), Ordovician, N.
America, differs from the last three genera : (1) in greater irregularity of
RR, 1. post. R being larger than the others and often bisected vertically
(1. ant. R or ant. R, as well as r. post. R, are horizontally bisected as usual) ; (2)
in its peculiar arm-branching (Fig. LVIII. 4). Herpetocrinus, Salter (1873 ;
synn. Ophiocrinus, Charlesworth ; Myelodactylus, Ang. and '[?] Hall ; [?}
Brachiocrinus, Hall. Fully described, Bather, 1893), Silurian, Europe,
and probably N. America. Crown bent, in the transversal plane, back
on the stem, which is then coiled around it in the opposite direction
(Fig. LIX.). The coil could be tight, or could be uncoiled and the crown
exposed ; the movement was effected by an elastic ligament towards the
inner margin of the columnals, counteracted by strong muscles towards
the outer margin, the fulcrum being a transverse ridge. Columnals show

THE CRINOIDEA

147

traces of original pentarnerisra, but over the greater part of the stem are
crescentic in section, the concavity being on the inner margin. Cirri,
borne on the horns of the crescent, vary in their arrangement in different
species. No root in adult. One ray missing ; other RR all compound,
except 1. post. R sometimes ; x lower than in other Heterocrinidae, partly
rests on r. post. Rt (the radianal, RA). The tube outwardly resembles a
series of Br and covering -plates. Arms slightly heterotomous. The

Fio. LIX.

Herpetocrinus Flctcheri, in its natural coiled position, the cirri which covered the crown
having been removed. C, cirri ; S, longitudinal suture of stem ; t, anal tube. (After Bather,
1893.) x ?.

remarkable resemblance of the coiled cirriferous stem to a pinnulate,
canaliculate arm has misled most writers ; for the crown is rarely visible
(vide -p. 134). FAMILY 4. CALCEOCRINIDAE. Monocyclica Inadunata, with
the essential characters of typical Heterocrinidae, the 1. ant., r. ant., and
r. post. RR being compound, and the arms branching primitively on the
plan of Heterocrinus heterodactylus ; but with the following modifications
induced by the bending down of the crown : — the r. post. IR lies alongside
the stem ; the 1. ant. R lies away from the stem ; the plane thus marked
is one of a gradually increasing bilateral symmetry ; r. post, arm always

148

THE CRINOIDEA

absent, its place being occupied by anal tube ; the tube encroaches on
r. ant. arm, so that this too disappears ; r. post. Rs and r. ant. Rs fuse
to form a T-shaped piece supporting anal tube ; the T-piece atrophies,
and the tube then rests on the corresponding iuferradials ; the simple RR
(1. post, and ant.) increase in size, forming the sides of the cup and bend-
ing round on the adcolumnal side where the anal tube is, as well as on
the acolumnal side, where they eventually meet between the two halves
of 1. ant R ; the arm borne by 1. ant. R may fork once, but usually

l.a. Br

JAx

a. Br

VAx

FIG. LX.

Calceocrinus tucanus, from the anterior side.
7?, the base, hinged to left anterior radial,
which is out of sight, and flexibly jointed to
r.a.Ri, right anterior inferradial ; the super-
radials of this and of the corresponding radial
on the other side are hidden by the series of
main-axils, lAx to VAx, which support the
branches of the two large side -arms; the
visible side-arm, n.Br, is the anterior, and
springs from a.R, anterior radial ; its diminished
branches are seen at y and S, its enlarged
branches at F ; the single arm along the upper
side is the left anterior (l.a.Br). (After Bather,
1893.) Natural size.

remains simple, owing to the extraordinary development of the arms
borne by the two large RR on either side of it ; the adcolumnal ramus
of each of those arms is reduced to a series of 3-8 axillaries (main-axilsj
which lie side by side, curving round towards the anal tube ; each main-
axil gives off a branch which itself bears armlets, which in turn may
assume the regular nature of pinnules, and are rarely visible on the
exterior of the folded arms ; the acolumnal ramus dwindles in size and
becomes hidden by the other branches ; r. post. B atrophies, 1. post, and
1. ant. BB fuse, and the three plates thus left form a triangular base,
which is regularly hinged to the 1. ant. Rt, so that the crown could move
up and down in the sagittal plane of its bilateral symmetry ; on the other
hand, the columnals have each a slight curvature in this plane, and this

Fio. LXI.

Diagrams illustrating the structure of the posterior area in 1, Castocrinus ; 2, Euchirocrinus ;
8, Calceocrinus; 4, Halysiocrinus.

kept the stem itself rigid. The various stages in the evolution are
marked by 4 genera (Fig. LXI.) : Castocrinus, Ringueb. (1889), Ordovician,
N. America; Euchirocrinus, Meek & Worth. (1873 ; synn. Cheirocrinus,
Hall non Eichw. ; Cremacrinus, Ulr. ; Proclivocrinus, Ringueb.), Ordovician
and Silurian, N. America; Calceocrinus, Hall (1852, em. Ringueb., 1889 ;
synn. Cheirocrinus, Salter, nom. nud. ; Pendulocrinus, Austin, MS. ; 1 Delta-
crinus, Ulr.), Silurian and Devonian, N America and Europe ; Halysio-
crinus, Ulrich (1886, em. Bather, 1893), Carboniferous, N. America. For
history and morphology see Bather (1893).

THE CRINOIDEA

149

FAMILY 5. PISOCRINIDAE. Monocyclica Inadunata with 1. post. R
and ant. R much larger than the other RR ; r. post. R is the only
compound R transversely bisected, and its lower half separates r. post. Rs,
and r. ant R from BB, and forms with the two large RR the greater part
of the dorsal cup. Arms unbranched and non-pinnulate. An anal tube

1. Pisocrinus.

i-L^ri
Sj^/ 1 post

'/••'* \ /*»** VRA";: r\*\ :

2. Triacrinus.

}. Calycantho-
crinus.

4. Mycocrinus.

5. Catillocrinus.

FIG. LX1I.

Pisocrinidae and Catillocrinidae. PJZ, pararadials ; t, anal tube plate = x; R', radianal
=.RA ; other letters as usual. In Mycocrinus the radianal and lower parts of RR are hidden
and therefore represented by dotted lines.

usually present, resting on post. RR or on their processes. Genera — Pisct-
crinus, de Koninck (1858), Silurian, N.-W. Europe and N. America (Fig.
LXII. 1 ; see also XVIII. 5), and Triacrinus, Miinster (1839, syn. Tricho-
crinus, J. Miillpr, 1856), Deitonian, Germany (Fig. LXII. 2), differ in that
the former has 5 BB, the latter 3. But incipient fusion of BB is seen in
P. ollulat and 5 BB occur in some Triacrinus. The anal tube closely

ISO THE CRINOIDEA

resembles an arm, and its presence was 'first notified in 1893. The relations
of the cup-plates, till then misunderstood, were thus shown to be essentially
the same as in the majority of Monocyclica, while the origin of Calycantho-
crinus, Mycocrinus, and Catillocrinus became clear. This was confirmed by
Jaekel (1895). Pisocrinus was shown by Wachsmuth to have 5 O sur-
rounding a peristomial space ; a groove passed along each interoral suture,
and probably conveyed the food-grooves to the central opening. The
rectum passed into the anal tube between post. 0 and adjacent RR. An
anal tube is said by Jaekel to have been absent from the thinly plated
species of Triacrinus in Devonian slates. Calycanthocrinus, Follmann
(1887), Lower Devonian, Germany, shows a remarkable modification,
in the introduction of small arm -bearing plates ("pararadials" PR),
in the positions shown in Fig. LXII. 3. This may be compared with
the vertical bisection of a radial in Anomalocrinus or the addition
of 5 "interradial radials" in Promachocrinus ; but it is the first stage
of a process continued in Catillocrinidae. It is hard to see how this
process could have been inaugurated except as a discontinuous meristic
variation (cf. Bateson, Materials for tlie Study of Variation, chap. xvii.
1894). FAMILY 6. CATILLOCRINIDAE. Monocyclica Inadunata, in which
1. post. R and ant R are much larger than the other RR, and bear PR,
usually fused to them ; no RA visible. Arms unbranched and non-
pinnulate. Anal tube resembling arms, but stouter, rests on left process
of r. post R. Genera — Mycocrinus, L. Schultze (1866, W. & Sp., 1886 ;
Jaekel, 1895), Middle Devonian, Germany (Fig. LXII. 4), differs from Caly-
canthocrinus in the suppression of RA and the PR borne by it ; the still
greater size of 1. post. R and ant. R, which now bear 6-7 arms apiece, i.e.
15-17 arms to the whole crown. The 3 BB become fused, while the
10-12 PR are usually fused with the large RR on which they rest. The
increase in size of the large RR takes place chiefly in the upper part ; all
the RR rest on the basal circlet, which forms a knob sharply separated
from them. Catillocrinus, Shuniard ex Troost (1860, syn. Nematocrinus,
Meek & Worth en ; see W. & Sp., 1886), Lower Carboniferous, N.
America (Fig. LXII. 5). The cup has here resumed a shallow basin-
shape, and differs from Mycocrinus in the complete fusion of BB and the
still greater lateral extension of the 1. post R and ant. R in their upper
regions. Those large RR now bear 15-31 arms apiece, there being
usually more on ant. R than on 1. post R (cf. Calycanthocrinus). The PR
are absolutely fused with RR, or were never developed at all. The basal
circlet projects upward on 1. ant. side, but is almost hidden by the stem on
the other side. The arrangement and shapes of the plates curiously
resemble those which obtain in Halysiocrinus, while the arms may be com-
pared to the branches that spring from the main-axils in that genus. A
similar result has been attained by two lines of development, which, in
their initial stages, were as different from one another and from the
normal type as could well be. The relationship of these forms is a
problem that would repay yet deeper study. FAMILY 7. ZOPHOCRINIDAB.
Monocyclica Inadunata, with 3 BB, 2 equal and 1 larger ; 4 RR, 1 rather
larger than the rest, and probably equals r. and 1. post fused ; 5 O form
solid teginen, post 0 being the largest, and r. and 1. ant O the smallest

THE CRINOIDEA

and, what is rather unusual, not meeting post. O in centre ; anus unknown ;
5 groups of short arms lie where the interoral sutures meet the RR,
which is not in the middle of any RR except ant. R (see Fig. LXIII. 2) ;
each arm-group appears to consist normally of 3 elements, 2 inner and
1 outer, all springing equally from the calyx ; stem with small axial canal.
Genus — Zophocrinus, S. A. Miller (1891), Silurian, Indiana (Fig. LXIII.).
Theca pear-shaped ; Miller says there are 5 arm -plates in each group, and

Fio. LXIII.

Zophocrinus Iwwardi, from a specimen in the collection of Hon. F. Springer. 1, from
posterior, x f ; 2, oral surface, x ^' ; 3, dissection of plates.

this may be so in some specimens ; these arm-plates may be compared
with the arms of Catillocrinidae, but whether they bore further Br is
uncertain. The tetramerism of this genus affects the cup only, and was pro-
duced by fusion, not by atrophy as was probably the case in Herpetocrinus.
FAMILY 8. HAPLOCRINIDAE. Monocyclica Inadunata, with 5 BB, 5
RR, of which 1. ant., r. post., and r. ant. are compound ; tegmen composed
solely of 5 0, one being pierced by a pore (? anus + hydropore) ; 5 arms,
unbranched and non-pinnulate, the food-grooves supposed to be subteg-
minal. Genus — Haplocrinus^ Steininger (1837 ; Aplocrinus, d'Orb. ; see
W. & Sp., 1886), Devonian, Europe and N. America (Figs. XXXV.
and LXIV.) ; resembles the typi-
cal Heterocrinidae in its cup,
the Pisocrinidae in its arms,
and Allayecrinus in its tegmen.
The pore in post. O was dis-
cussed on p. 126. FAMILY 9.
ALLAGECRINIDAE. Monocyclica
Inadunata, with 5 BB, 5 RR all
simple, but of unequal and vari-
able size, the larger ones often
bearing 2 arms, while some arms

may occasionally be absent or diminished in size ; tegmen composed in
the young of 5 0, one pierced by a pore (? anus + hydropore) ; 5 arms un-
branched and non-pinnulate, the food -grooves supposed to be subtegminal.
Genus — Allagecrinus, Etheridge & Carpenter (1881, restricted by W. & Sp.,
1886), Lower Carboniferous, N. America, Upper Carboniferous, Scotland,
resembles Haplocrinus in its tegmen, while the duplication of arms is

RR

Fio. LXIV.
Dissection of cup of Haplocrinus. R' = radianal.

152 THE CRINO1DEA

reminiscent of Calycanthocrinus ; here, however, the RR are axillary,
and PR not developed. Rowley (1895) has described calyces less than
•5 mm. high. FAMILY 10. SYMBATHOCRINIDAE. Monocyclica Inadunata,
with 5 simple RR, bearing 5 unbranched, non-pinnulate arms, which rest
as a rule on broad articular facets projecting adorally in "muscle -plates,"
corresponding to the muscle-fossae. Genera — Phimocrinus (Low. and Mid.
Devonian, Europe), Stylocrinus and Stortingocrinus (Mid. Devonian, Europe),
Symbathocrinvj, including Lageniocrinus, which is probably its young (Upper
Devonian and Lower Carboniferous, Europe and N. America). Phimocrinus,
Schultze (1866), is primitive in having 5 BB, while all the rest have 3 ;
the oldest species, P. Jouberti, Oehlert, shows clear traces of horizontal
suture in r. post., r. ant, and 1. ant. RR, the usual compound RR of typical
Heterocrinidae ; the proximal plate of the anal tube is placed as in those
forms. The latter feature is also found in Stortingocrinus, Schultze (1866),
which is notable for having the small unfused B in 1. post. IR, and the
angles of the columnar canal radial in position, though the genus is Mono-
cyclic. Stylocrinus, Sandberger (1850), has no facet on RR for support
of an anal tube, nor is the usual passage for the rectum seen in the cup-

RR

Fm. LXV.

Dissection of cup of Xi/mbathocrinuf. The projections on the llll are muscle-plates.

wall ; therefore one cannot orient the small B ; the anus probably pierced
post. O. Symbathocrinus, Phillips (1836), has small B in 1. ant. IR, as in
Platycrinidae ; the anal tube rests on the shoulder of r. post. R ; 5 O,
all small, post. O being the largest and separated from the muscle-plates
of the adjacent RR by the passage of the rectum (Fig. LXV.). The family
is not clearly denned, and its Middle Devonian members present remarkable,
though superficial, resemblances to the contemporary Cupressocrinidae.

FAMILY 11. BELEMNOCRINIDAE. Monocyclica Inadunata, with 5
BB, 5 simple RR, each supporting by a slightly excavate facet an arm
which forks on IBr4 or 5 ; each ramus, with numerous syzygies in the
proximal portion, bearing ramuli on alternate sides of most epizygals
and of all ordinary distal brachials ; single narrow anal in radial circlet,
resting on post. B, and supporting a large anal tube composed of alter-
nating hexagonal plates, folded on their lateral margins. Genus — Belem-
nocrinus, White (1862), Lower Carboniferous, N. America. Dorsal cup
elongate, its plates solid, and thecal cavity a narrow canal below and
shallow excavation above, thus resembling that of many Neozoic crinoids.
B. typus has no cirri on the stem, which is circular in section ; B. florifer
has a stellate stem of true Monocyclic type, with 3 or 4 cirri to each node.

THE C KINO IDE A

153

Missouricrinus, S. A. Miller (1891), Burlington group, if truly Mono-
cyclic, must be placed near here.

FAMILY 12. PLICATOCRINIDAE. Monocyclica Inadunata, with BB fused
into a knob-like support ; 4 or 6 (exceptionally 3, 5, 7, or 8) simple RR,
enclosing wide thecal cavity, each supporting an arm, which forks on
IBi-j ; IIBr wedge-shaped, united by perforate articulation, not, so far as
known, by syzygy, regularly pinnulate ; pinnulars tend to fuse. Tegmen
unknown. Columnals cylindrical, with radiately striated joint-surface.
For full account, and for the evidence of the axial canals passing from RR
into the fused basal circlet, so proving that it is no columnal, see Jaekel
(1893). Genera — Plicatocrinus, Miinster (1839), Upper Jura, Germany
(Fig. LXVL), thin RR enclose a wide and deep thecal cavity, facets from
^ to | width of R, crescentic, without
large muscle -plates ; proximal pinnules
composed of 3 pinnulars, which in suc-
ceeding pinnules are fused to a solid piece.
Tetracrinus, Miinster (1839), Upper Jura,
France and Germany, has thick RR with
strong muscle-fossae. The occasional oc-
currence of 3, 5, 7, or 8 rays shows that
the normal 4 and 6 arose from the more
usual 5 as sports (i.e. discontinuous meris-
tic variations). FAMILY 13. HYOCRINIDAE.
Monocyclica Inadunata, with 3 thin BB,
the smaller one in 1. post. IR ; 5 RR, thin,
broad, and spade-shaped, with a slight axial
fold running straight up the middle from
the subjacent B and ending in a narrow
facet ; 5 arms, bearing unbranched alter-
nating ramuli, which may be modified in
their proximal portions for reception of
ripe gonads; Br long, cylindrical, with iss2, x j|.)
deep, narrow ventral groove ; below the

first ramule, which is on the right of each arm, are 6 Br, with the
joint between each pair a syzygy ; between successive ramuli are 3
Br, united by syzygies ; proximal ramuli the longest, and the follow-
ing ones proportionately shorter, so that they all terminate on the
same level as the arm ends ; 5 triangular 0, separated from RR by
iAmb and Amb, cover the wide mouth and circumoral tentacles ; each 0
as a rule pierced by a water-pore, and post. O by 2 ; water-pores also
pierce several iAmb, but not in post. IR ; anal tube reduced to a small
cone towards the left of the post, interambulacrum ; columnals cylindrical,
slightly higher than wide, united by discs of ligament- fibres, joint-
surfaces hollowed and plain or indistinctly striate, with stellate axial
canal ; no cirri ; root unknown. The sole living representative of the
order is the unique species Hi/ocrinus bethellianus, W. -Thorn son (1876,
Fig. LXVIL), dredged by H.M.S. Challenger 30 miles W. of Crozet Is.,
in the Southern Ocean, while colunmals are said to have been found
in Mid-Atlantic just N. of the Equator ; see P. H. Carpenter (1884).

FIG. LXVI.
Plicatocrinus Fraasi, from aborai

154

THE CRINOIDEA

Evidence for its truly Monocyclic nature is wanting; but in the
absence of proof to the contrary, it must be
left in this, apparently natural, position. The
stem is not unlike that of some Plicatocrinidae ;
the branching and syzygies of the unforked
arms remind one of the structure in the
arm-rami of Belemnocrinus and Saccocoma ; the
cup is not unlike that of Plicatocrinus Fraasi.
The only specimen at all complete is a male,
with testes probably mature and swelling out
in the proximal portions of the proximal
ramuli ; in these regions each ramular sup-
ports 2 or 3 square side-plates on either side,
and these support the covering-plates. The
orals could probably open to expose the
funnel-shaped gullet, which leads into a
narrow gut with single dextral coil ; glandular
ridges line the gullet and first part of the gut.
The intra-thecal connective tissue contains no
spicules. FAMILY 14. SACCOCOMIDAE. Mono-
cyclica Inadunata, in which 5 RR and a
minute centrale enclose a large spheroidal
thecal cavity ; each R has a prominent median
ridge ending in a narrow facet, which sup-
ports a thin arm, forking on IBr2 ; beginning
at about IIBrJ5, each ramus gives off from
every 3rd Br unbranched alternating ramuli
arranged as in Hyocrinus ; the rami and
ramuli are usually found rolled up in their
distal portions ; Br cylindrical, elongate ;
lax and the more proximal IIBr may bear
lateral, paired, wing-like expansions, which
in the more distal Br and the ramulars are
always represented by delicate trellised pro-
cesses, with thicker upper and under mar-
gins, which it is conjectured supported a
continuous membrane ; no stem ; all skeletal
elements very thin and coarsely reticulate.
Genus — Saccocoma, L. 'Agassiz (1834; syn.
Euryale, Konig), Solenhofen Lithographic
Fio. LXVII. Stone, Upper Jura (Fig. LXVIII.). Jaekel,
Hyocrinus bethel- who has admirably elucidated the structure
Chaito'cr (Narra" and amnities of tJlis wonderfully specialised
tive, xf.) crinoid (1893), considers that .the arms were

swimming-organs, the food-groove, ambulacral,
and genital systems being atrophied at their
distal ends, and that the animal was pelagic,

floating in enormous swarms in the peaceful lagoons of Eichstadt and

Solenhofen.

THE CRINOIDEA

155

Fio. LXVIII.

Saccocomn. 1, S. teneUa, from aboral surface (after Jaekel, x {). 2, S. pectin/ata, from
aboral surface, to show coiling of arm-branches (x £). 3,'S. pectinata, cup and proximal
brachials from side (after Zittel, x |).

ORDER 2. Adunata (Bather, 1899).

Monocyclica with dorsal cup primitively confined to the patina and
an occasional single anal ; tegmen solid ; portions of the proximal Br
and their Amb tend to be rigidly incorporated in the theca. Arms fork
once to thrice, and bear pinnules on each or on every other Br. BB fused
to 3, 2, or 1. (Eucladocrinus and Acrocrinidae offer peculiar exceptions
to this diagnosis.)

The genera of this order have been referred, in whole or part, to the
Camerata or their equivalents by most recent writers. But most of the
Silurian genera here included are admitted as close allies of Monocyclica
Inadunata. If so, then they were derived from Inadunata ages after the
typical Camerata had appeared. Moreover, though the Devonian and
Carboniferous descendants of these Silurian genera became modified after
the fashion of the Camerata, there is scarcely one in which the modification
is so great as in the simplest Camerate. The order is divisible into two
groups — A. with RR in contact all round the cup, and base consequently
pentagonal in outline ; B. with RR separated by an anal plate in post.
IR, and base consequently hexagonal in outline. The relations of
these 'groups are not clear. B. may be derived from A. by sudden inter-
calation of an anal, or A. and B. may have descended independently from
Inadunata ; the absence of B. from Silurian rocks renders the former
hypothesis more probable. There is no difficulty in imagining the descent
of early genera of -4., such as Cordylocrinus and Goccocrinus, from Hybo-
crinidae in which the cup had become as symmetrical as in, say, Stephano-
crinus while the arms had passed through regular dichotomy to a stage
with ramuli, as in JEctenocrinus. The actual intermediate form is not yet
known, but that it will be found among Ordovician crinoids is a
legitimate inference.

THE CRINOIDEA

GROUP A. FAMILY 1. PLATYCRINIDAE. No anal. BB 3 (= 2 fused
pairs and 1 unfused, this being usually the left, sometimes the right,
antero-lateral) forming a pentagon. Cordylocrinus, Hapalocrinus, Cocco-
crinust and Cylicocrinus, which appear in the Silurian and continue to
Devonian, are more primitive than the rest. Marsipocrinus, though also
Silurian, is considerably more advanced, and may be regarded as a
first attempt at the typical Platycrinus structure ; it was apparently an
unsuccessful attempt, since it left no descendants, although Wachsmuth
& Springer and Jaekel seem to regard Platycrinus as such. Platycrinus
came in the Carboniferous, with its offshoot Eucladocrinus, and, unless
we are to imagine a reversal of the general trend of evolution, must be
derived from a simpler form than Marsipocrinus. These facts are best
presented by instituting 3 sub-families. SUB-FAMILY 1. COCCOCRININAE.
Platycrinidae with IBr 2 (3 in Hapalocrinus Victoriae) ; IIBr more than
2 ; few Amb and iAmb in tegmen ; (?) anal tube rarely present ; stem cir-
cular in section ; lumen small and round. Genera — Coccocrinus, J. M tiller
(1855, syn. Amblacrinus, d'Orbigny pars, 1849), Silurian of America,
Devonian of Europe ; O large, symmetrical, almost covering the tegminal
ambulacra ; iAmb 1 or 3 ; anus between O and iAmb ; arms apparently
delicate, fork once, distal portions unknown. Cylicocrinus, J. Miiller
(1855, as Culicocrinus; Jaekel, 1895), Devonian, Germany (Fig. XL.),
differs from' Coccocrinus chiefly in having heavy biserial arms. Hapalo-
crinus, Jaekel (1895, em. Bather, 1897 ; includes Agriocrinus, Thallocrinus,
and Clematocrinus, Jaekel), Silurian of England, Australia, and (?) N.
America, Devonian of Germany (Fig. LXIX.) ; 0 small ; iAmb more than

1 ; Amb visible between 0 and iAmb ;
arms fork once, sometimes twice, varying
in this respect in the same species, or even
individual ; IIBr (and IIlBr when pre-
sent) uniserial, or slightly in zigzag, bear-
ing alternately disposed pinnules, either
on each or on every other Br ; cirri at
the root, and often on nodals. Cordylo-
crinus, Angelin (1878 ; W. & Sp., 1897 ;
Bather, 1897), Silurian of Gotland and
England, Lower Devonian of N. America ;
differs from Hapalocrinus in having com-
pound IIBr, each of which bears a pin-
nule on each side. SUB - FAMILY 2.
MARSIPOCKININAE. Platycrinidae with
one IBr ; IIBr, if not finials, one or two ;
many Amb and iAmb in tegmen ; no
anal tube; stem circular in section; lumen
large and quinquelobate. Genus — Mar-

Hapalocrinus retiarius, from Brit. . e . £ .. ^ ,, ...

Mus. E5615. c, cirri; iBr, interbra- sipocrinus, Bather (1889, nom. nov. pro

Marsupiocrinus, Phillips, 1839, non de
Blainville, 1830 ; syn. Cypellocrinus vel
Cupellaecrinus, Shumard, 1866, ex Troost MS., non Steininger), Silurian,
N.-W. Europe and N. America ; arrangement of cup-plates shown in Fig.

THE CRINOIDEA

157

LXX., and of tegmen in Fig. XXXII. ; 0 rather small, pushed anteriorly,
and asymmetrical; tegminal lAmb and IIAmb distinctly visible ; arms fork
once or twice, stout, biserial, with large pinnules ; no cirri on column. The
large size of the cup is accompanied by development of stroma-strands across
the sutures. The large tegmen permitted a Gastropod (P7 'yceras) to attach
itself above the anus and live on the excrement ; this is seen in many of
the larger Adunata and Camerata and a few other crinoids (Keyes, 1888).
SUB-FAMILY 3. PLATYCRININAE. Platycrinidae with IBr 1 (rarely 2), IIBr 2
(rarely 3) ; tegminal Amb and iAmb usually more than in Coccocrininae,

FIG. LXX.
Marsipocrinus, from Brit. Mus. E6519. pn, pinnules ; other letters as usual. Nat. size.

fewer than in Marsipocrininae ; anal tube often present ; section of stem
circular near the cup, elliptic or rhomboid below, with fulcral ridge following
long diameter ; lumen small and round. Genera — Platycrinus, Miller
(1821, W. & Sp., 1897; synn. Centrocrinus and Pleurocrinus, Austin;
Edwardsocrinus, d'Orb.), Devonian (1 species), Carboniferous, Europe and
N. America (Fig. LXXI.) ; 1-3 iAmb always rest on the adjacent shoulders
of the RR in each IR, and consolidate a varying number of Br and Amb
with the calyx ; arms fork once to thrice (exceptionally more), the dichotomy
often irregular, producing 6 rami to the arm ; columnals have a slight
skew, so that the fulcral ridge of the proximal surface lies at an angle to
that of the distal surface ; thus the flattened stem, which otherwise could

I58

THE CRINOIDEA

move only in one plane, has a spiral twist that enables it to bend in any
direction (Fig. XLIX. 5, 6). Eucladocrinus, Meek (1871, W. & Sp.,
1897), Carboniferous, N. America ; appears late in the history of Platy-
crinus, from which it was evolved polygenetically by modification of the
arms ; two main rami to each arm (sometimes only one ramus) composed
of biserial Br, with large almost rigid Amb, form tubular extensions of the
thecal cavity, and give off on alternate sides short, biserial ramuli, which
in turn bear pinnules (Fig. LXXI. 4). This sketch of the evolution of the
Platycrinidae is confirmed by the ontogeny of Platycrinus. In young
stages (Fig. LXXI. 2) the basal cup is relatively shallower ; RR less high ;
columnals circular in section ; Br uniserial, later on zigzag, and longer ;
pinnules relatively stouter and wider apart ; 0 relatively larger, and

Fio. LXXI.

Platycrininae. 1, aboral view of cup and proximal brachials of Platycrinus subspinosus (after
Wachsinuth & Springer). 2, crown of young P. Huntsvillae, from Brit. Mus. E6778. x I. 8,
calyx of P. eminulus, anterior view (after W. & Sp.). 4, arms of Eucladocrinus millebrachiatus,
oral surface showing covering-plates (c.p) and aboral surface (after W. & Sp.). Br', articular facet
for arms ; Rm, ramus ; r, ramule ; other letters as usual. All (exc. 2) two-thirds natural size.

occupying greater part of teginen. From this obviously Coccocrinine stage,
the change to the mature Platycrinus has been observed in many species.
GROUP B. FAMILY 2. HEXACRINIDAE. Cup formed of 1, 2, or 3 BB,
forming a hexagon ; 5 RR ; and 1 anal in line with RR. The family
is closely related to the Platycrinidae, but differs in the hexamerous
symmetry ; also, as a consequence of this, in the presence of an interbasal
suture in post. IR, whereas in Platycrinidae the nearest to that position
is in r. post, radius. Genera — Hexacrinus, Austin (1843, Schultze, 1867,
W. & Sp., 1897), Devonian, Europe and N. America (Fig. LXXII.) ;
BB 3 ; IBr, 2 united by syzygy in America, only 1 in Europe ; arms with
2 rami, bearing ramuli on one or both sides at intervals ; Br uniserial,
and all except axillaries bear pinnules ; tegmen as in Platycrinus ; stem
circular in section, with small round lumen. Arthracantha, Williams
(1883, syn. Hystricrinus, Hiude, 1885), has theca'of same structure, but
armed with movable spines borne on tubercles ; arms dichotomous, biserial ;

THE CRINOIDEA

159

stem circular in section. Dichocrinus, Miinster (1838, W. & Sp., 1897,
syn. Cotyledonocrinus, Casseday & Lyon, 1860), Carboniferous, Europe and
N. America ; BB 2 ; IBr 2, united by syzygy ; IIB^ ^ 2 and, when arms
fork again, IIIBrj and 2 are united by syzygy ; arms thin, uniserial or
biserial, occasionally pendent ; stem circular in section. Camptocrinus,
W. & Sp. (1897), Carboniferous, America, differs from Dichocrinus

Fia. LXXII
Hexacrinus. 1, analysis of cup ; 2, //. pateraeformis, cup from below. (After L. Schultze.) x L

only in the structure of the stem, which in its crescentic section and 2
series of cirri resembles that of Herpetocrinus (p. 147). Talarocrinus, W.
& Sp. (1881-97), Carboniferous, N. America ; differs from its ancestor
Dichocrinus in its more massive plates and in having but one IBr to an
arm, and that small ; anal resembles ant. R in shape and size ; arms
branch twice, biserial, free from IIBr inclusive. Pterotocrinus, Lyon &
Casseday (1859, W. & Sp., 1897 ; syn. Asterocrinus, Lyon non Miinster),
Carboniferous, N. America ; a remarkable modification of Talarocrinus,
with Br up to IIlBr incorporated in cup ; large
wing-like processes spread out from tegmen, and
probably represent the hypertrophied axillary
lAmb. With this exaggerated type, the nearest
approach to the true Camerata, the Hexacrinidae
become extinct. FAMILY 3. ACROCRINIDAE. Cup
formed of 2 BB, forming a hexagon ; 5 RR ; 1
anal in line with RR ; and a large belt of acces-
sory plates between BB and RR. Acrocrinus,
Yandell (1855, W. & Sp., 1897), later Carboni-
ferous, N. America (Fig. LXXIII.) ; derived from
Dichocrinus, which it otherwise resembles, by the
gradual intercalation of 6-20 circlets of supple-
mentary plates, " superbasals " (SB) ; the SB im-
mediately above BB are always the latest formed
and the smallest ; the SB supporting the anal and
ant. R are in single series, the rest alternate.
In A. amphora the arms were recumbent on the cup and apparently
immovable. This remarkable family was the last to appear, and sur-
vived all other Adunata and all Camerata.

ORDER 3. Monocyclica Camerata

( = CAMERATA, W. & Sp. pars).

Monocyclica in which IBr, two in each ray(exc. Stereocrinus, Hadrocrinus^
Alloprosallocrinus) and often succeeding orders of Br, are incorporated by

Fir.. LXXlIf.

Aerocrinns amphora. (Re-
constructed from Wachs-
niuth & Springer.) X J.

i6o THE CRINOIDEA

iBr in the dorsal cup (becoming " fixed brachials," 13r), while the corre-
sponding Amb are either incorporated in, or pressed below, the tegmen
by iAmb ; all thecal plates united by suture, somewhat loose in the
earliest forms, but speedily becoming close, and producing a rigid theca ;
mouth and tegmiual food-grooves closed ; arms pinnulate.

The families may be grouped somewhat after the plan of Wachsmuth
and Springer (1897), thus : —

A. No anal plate in radial circlet of patina. BB usually form a

pentagon. Melocrinoideu.

B. An anal plate between RR in post. IR. BB form a hexagon.

1. Proximal anal heptagonal, succeeded by one or more in the

same vertical series, between the ordinary iBr. Batocrin-
oidea.

2. Proximal anal hexagonal, succeeded by no vertical series, but

by 2 iBr. Actinocrinoidea.

Whether Group B. was derived from some genus in Group A. is un-
certain ; representatives of both groups are found in the Ordovician. It
is more probable that j£>,2 was derived from #, 1, by way of the Periecho-
crinidae. Among Melocrinoideu the Patelliocrinidae are scarcely removed
from the Inadunata, and some of their genera might almost be placed
with the early Adunata, with which they were contemporary. Although
their occurrence in the Ordovician is doubtful, yet the existence of the
family points to the path along which the Carnerata ascended. The
Glyptocrinidae and Melocrinidae, which differ in little but number of
BB, and both have anals in the dorsal cup, may have been derived from
such a form as Stelidiocrinus, which also has anals. With less doubt we
infer that the Patelliocrinidae without anals gave rise to the similarly
constituted Clonocrinidae, from which sprang Eucalyptocrinidae, and
probably also Dolatocrinidae. Among Batocrinoidea the simplest and
one of the oldest genera is Tanaocrinus, with 5 BB, and not far from
it come the earlier Xenocrinidae, with 4 BB, which perhaps led on to
Periechocrinidae, and so to Actinocrinoidea. The important ancestral
family, however, is the Silurian Carpocrinidae, in which Acacocrinus is
nearest to the Inadunate type. From them arose Barrandeocrinus with
its recumbent arms, then the Coeliocrinidae without anal tube or respira-
tory pores, and, later on, the Batocrinidae possessed of both those structures.
No members of this order survived the Lower Carboniferous, but during
their history they developed some of the most numerous in individuals
and species among crinoid genera, and in Barrandeocrinus, Eucalyptocrinus,
Agaricocrinus, and Strotocrinus, some of the most remarkable of all
Echinoderma.

SUB-ORDER 1. Melocrinoidea. Monocyclica Camerata with RR in
contact all round ; IBrj usually quadrangular.

FAMILY 1. GLYPTOCRINIDAE. Melocrinoidea with 5 BB ; in each half-
ray 2-8 HBr, sometimes IIlBr ; free arms rarely branch beyond IIIBr,
and may be uniserial, zigzag, or biserial ; iBr numerous but definite ;
illBr numerous, less definite ; a ridge of anals in post. IR ; tegmen
of numerous small plates ; stem round, with pentagonal lumen. All
Ordovician of N. America, European representatives doubtful. Genera —

THE CRINOIDEA

161

Glyptocrinus, Hall (1847 ; W. & Sp., 1897 ; synn. Canistrocrinus, W. &
Sp. ; Pycnocrinus, S. A. Miller), has small BB and arms in zigzag (Fig. XXV.).
Schizocrinus, Hall (1847 ; W. & Sp., 1881 ; (?) syn. Scyphocrinus, Hall
non Zenker), doubtful, but close to Glyptocrinus, arms uniserial. Periglypto-
crinus, W. & Sp. (1897), has large BB and biserial arms. FAMILY 2. MELO-
CRINIDAE. Melocrinoidea with 4 BB ; in each half-ray 2-5 IlBr ; these
support 2 or 4 main rami giving off pinnules or pinnulate ramuli ; iBr,
illBr, and post. IR as in Glyptocrinidae ; tegmen of numerous, small, and
irregular plates ; stem round. Genera — Scyphocrinus, Zenker (1833 ;
non Hall, nee Pictet), Silurian, Bohemia and (?) N. America. Cup very
large, including many of the proximal ramuli, which enter the iBr
and illBr areas ; subsequent rami are free and divergent ; rami and
ramuli uniserial ; root a large hollow spheroid strengthened by internal
septa, regarded as a float ( = Camarocrmus) by Hall, as a cystid ( = Lobolithus)
by Barrande. Mariacrinus, Hall (1859, -estr. W. & Sp., 1881-97 ; syn.
Zenkericrinus, Waag. & Jahn), Silurian, Europe, N. America. Free arms
composed of wedge-shaped IIIBr, divergent, may or may not bear a
few ramuli on their amedian sides (Fig. LXXIV. 1). Melocrinus,
Goldfuss (1826; W. & Sp., 1897;
synn. Astrocrinus, Conrad ; Turbino-
crinus, Troost ; Castanocrinus, and
Cytocrinus, C. F. Roem. ; Clonocrinus,
Oehlert), Devonian, Europe and N.
America ; the 2 main rami of each arm
are laterally fused into one trunk with
single large ventral groove ; this bears
paired biserial pinnulate ramuli (Fig.
LXXIV. 2) ; thus the genus is related
to Mariacrinus as Eucladocrinus (p. 158)
to Platycrinus, and as Steganocrinus (p,
170) to Adinocrinus. Ctenocrinus, Bronn
(1840, em. Follmann, 1887), Lower
Devonian, W. Europe, 'is distinguished
by Jaekel (1895), but merged with
Melocrinus by W. & Sp. (1897) ; the
ossicles of the rami are compound, and
each may bear 2 pinnules.

FAMILY 3. PATELLIOCRINIDAE. Melo- ger). Rm, ramus ; r, ramule ; p, "pin-
. T. . nule. Supplementary plates are shaded.

crmoidea with but few Br incorpor-
ated in cup ; BB usually 3, unequal, may be the original 5, or may
fuse to 1 ; both IBr resemble free brachials ; in each half-ray 1, or
generally 2, IIBr, merging into free arms, which may be uniserial, zig-
zag, or biserial ; iBr 2-6, a single one rests on RR ; stem small and
round. This early Palaeozoic family contains genera with and without
additional plates in post. IR ; but all are simple forms, scarcely more
removed from the Inadunate type than are the Silurian Adunata. They
may be intermediate between Inadunata and some more advanced
Camerata, e.g. Clonocrinidae ; but it is not probable that they represent
the ancestors of Glyptocrinidae or Melocrinidae. Genera — Stelidicorinus,

ii

FIG. LXXIV.

Rays of Melocrinidae. 1, Mariacrinus,
from Brit. Mus. 57475. 2, Melocrinus
nobilissimus (after Wachsiuuth & Sprin-

162

THE CRINOIDEA

Angelin (1878; [?] syn. Harmocrinus, Ang.), Ordovician, N. America,
and Silurian, Gotland (Fig. LXXV. 1,2); 5 BB, uniserial arms, anals in

B...

iBr

FIG. LXXV.

Patelliocrinidae. 1 and 2, Stelidiocriniis capitulum. 1, calyx from posterior ; 2, oral surface ;
3, PateUiocrinus pachydactylus. (All after Angelin, about nat size.)

post. IR, large orals. Macrostylocrinus, Hall (1852, W. & Sp., 1897), Silu-
rian, N. America ; 3 BB, biserial arms, anals in post. IR, small plates in

tegmen. Allocrinus, W. & Sp. (1889-
97), Silurian, N. America; 3 BB,
uniserial arms, no anals in post. IR.
Patelliocrinus, Angelin (1878), Silurian,
Gotland (Fig. LXXV. 3) ; 3 BB, arms
zigzag or biserial, no anals. Briaro-
crinus, Angelin (1878), Silurian, Got-
land ; 3 BB, arms secondarily uni-
serial, i.e. I IBr are compound plates,
anals uncertain. Centriocrinus, Bather
(1899, nom. mut. pro " Centrocrinus,"
W. & Sp., 1881, non Austin, 1843,
nee Worthen, 1890); BB fused, arms
unknown, no anals in post. IR.
FAMILY 4. CLONOCRINIDAE. Melo-
crinoidea with 4 BB ; in each half-
ray 1-2* IIBr, with varying number
of II IBr and even occasionally IVBr ;
free arms biserial, sometimes forking
as far as VIIBr ; iBr few and
definite ; illBr few, not always
present ; occasional illlBr ; no anals
in post. IR ; tegmen unknown ; stem
round or sub -pentagonal. Genera —
ClonocrinuSy Quenstedt (1876, non
Fio LXXVI. Oehlert, 1879 ; syn. Corymbocrinus,

Oonocrin«s poiydactyius.' i, from side Ang.), Silurian, England, Gotland, and
(from Brit. Mus. 40257) ; 2, vertical section N. America ; base concave ; free arms
(from Brit. Mus. B6489). Lettering as . TTT_, A TTTTT. -^ .

usual. Nat. size. isotomous from IIIBr to VIIBr ; iBr ma

single vertical row, the two proximal

large and definite. This genus leads towards Eucalyptocrinidae. Poly-
peltes, Angelin (1878), Silurian, Gotland, seems to have pinnules incorporated

THE CRINOIDEA

163

in the cup. It leads on to TryUiocrinus, Geinitz (1867; syn. Spyridio-
crinus, Oehlert, 1891), Lower Devonian, Germany and France, which
also has a concave base. Technocrinus, Hall (1859, W. & Sp., 1897),
Devonian, Md., U.S.A., base convex, arms not branching beyond IIIBr,
no illBr ; the cup plates have axial folds as in Dolatocrinidae, which

FIG. LXXVII.

Eucalyptocrinidae. 1, Callicrinus murchisonianus from the side, arms removed except on
the right ; 2, plates from distal end of anal tube ; 3, the same plates from C. costatus ; 4, similar
plates described by Hall as Cryptodiscus ; 5, calyx of Callicrinus costatus; 1, 2, 3, k denote
successive circlets of the anal tube ; i, ii, iii, denote the areas of origin of interbrachial
processes. Other letters as usual. 1, 2, 3, 5, after Angelin ; 4, after Hall.

family may be thus connected with Corymbocrinidae. FAMILY 5.
EDCALYPTOCRINIDAE. Melocrinoidea with usually concave patina of 4 BB
and 5 RR ; in each half-ray are 2 IIBr, supporting HIB^ (and in
Eucalyptocrinu* IIIBr0), followed by IIIBr, proximally uniserial, distally
biserial ; iBr 3, 2 resting on 1 ; illBr 1 ; tegmen elevated in a central
anal tube, and composed of 4 circlets of large plates, variously shaped

1 64 THE CRINOIDEA

and bearing processes. When stripped of arms and processes, the theca
resembles a sherry-decanter with a kick at the bottom. At the junction
of cup and tegmen are the 10 pairs of ambulacral openings for food-grooves
and associated vessels, which pass beneath the tegmen to a central
mouth suspended beneath the neck of the decanter. From this a gut
winds dextrally down, around, then up, and out at the mouth of the
decanter. The processes borne by the tegmen are vertical partitions
rising from the fixed iBr and illBr to varying heights, while smaller
partitions are between the IIIBr series. Genera — Callicrinus, d'Orb.
(1849), Silurian, Europe and N. America (Fig. LXXVIL, see Angelin,
1878, W. & Sp., 1897). The cup is built on almost the same plan as
that of Clonocrinus, the difference lying in the vertical fission of the 2nd
iBr. The partitions do not rise very high between the arm-branches ;
on the other hand, large spinous processes are frequently given off from
plates of both cup and tegmen, while the upper circlet of the anal
tube is often extended in 4 quadrant-shaped horizontal extensions
(Cryptodiscus, Hall'; see Weller, 1898). Eucalyptocrinus, Goldfuss (1826 ;
syn. Hypanthocrinus, Phillips), Silurian of Europe, N. America, and
(?) Victoria, Devonian of Eifel (see Angelin, 1878, and W. & Sp., 1897) ;
differs from Callicrinus in the great development of the vertical partitions,
which form compartments in which the arms rest right up to their tips,
so that the closed crown is almost egg-shaped, especially as it has not the
obtrusive ornament of Callicrinus. Angelin has figured laminae of
stereom within the theca ; these served to support the subtegminal
food-grooves and mouth, and are pierced by the gut ; they are not
hydrospires. This family was richer in species than any other of the
Silurian ; while the main structure was fixed, the ornament varied
greatly. FAMILY 6. DOLATOCRINIDAE. Melocrinoidea with large base,
flattened or concave, 3 BB, nearly always fused ; in each half-ray, 1-4
IIBr, and sometimes IIIBr and IVBr ; free arms biserial, and may fork ;
'Br,3 or more ranges ; illBr usually present, merging with illAnib ; an
Additional iBr exceptionally present in post. IR ; tegmen, when known,
solid, with large plates ; stem round and large. All Devonian of N.
America. Genera — Dolatocrinus, Lyon (1857 ; W. & Sp., 1897 ; syn.
Cacabocrinus, Hall), has 2 IBr, a stout almost central anal tube, respiratory
(?) slits in interbrachial areas at junction of cup with tegmen (Fig. XLV. 1),
and plates usually with axial folds. Stereocrinus, Barris (1878, W. &
Sp., 1897), has 1 IBr, and BB unfused. Hadrocrinus, Lyon (1869 ;
W. & Sp., 1897 ; syn. [?] Coronocrinus, Hall), has 1 IBr and large iBr
variable in number and arrangement ; very large, and imperfectly known.
The single primibrachs of Stereocrinus and Hadrocrinus probably represent
the two IBr of Dolatocrinus fused (cf. Alloprosallocrinus}.

SUB-ORDER 2. Batocrinoidea. RR in lateral contact except in post. IR.
Proximal anal heptagonal. IB^ quadrangular, except in Periechocrinidae.

FAMILY 1. TANAOCRINIDAE. Batocrinoidea with 6 BB ; the proximal
IIBr fixed, but outwardly resemble free brachials ; arms fork still further,
brachials wedge-shaped ; iBr and illBr numerous, irregular, occupying
depressed areas connected with tegmen, which was probably flexible and
composed of small plates ; in post. IR is a vertical ridge of anals ;

THE CRINOIDEA

165

stem relatively large and -sub-pentagonal. Genus — Tanaocrinus, W. &
Sp. (1897), Ordovician, Ohio (Fig. LXXVIII. 1, 2). In structure and in
time this genus is well fitted to be taken as an ancestor of the Carpo-
crinidae. FAMILY 2. XENOCRINIDAE. Batocrinoidea with 4 BB ; each
half-ray contains IIBr, and sometimes IIIBr ; free arms in zigzag or
biserial. iBr and illBr, also illlBr when present, numerous. Post. IR
wider than the others, divided by a longitudinal row of ridged anals.
Tegmen of minute irregular plates. Stem quadrangular to circular in
section, with pentagonal lumen. Genera — Xenocrinm, S. A. Miller

FIG. LXXVIII.

Tanaocrinidae and Xenocrinidae. 1, Tanocriniis typm, posterior view, x 2 diam. 2, anterior
view, x 2 diam. 3, Xenocrinus, analysis of cup. 4, Compsocrinus harrisi, cup from posterior,
diagrammatic, x 2 diam. (All adapted from W. & Sp.)

(1881 ; W. & Sp., 1897 ; Fig. LXXVIII. 3) ; iBr sink between RR, so
as almost to rest on BB ; IIBr are finials. Compsocrinus, S. A. Miller
(1883 ; W. & Sp., 1897 ; Fig. LXXVIII. 4), has RR in contact except
on anal side ; IIIBr sometimes fixed ; iBr stouter than in Xenocrinus.
Both Ordovician, Ohio. Abacocrinus, Angelin (1878 ; syn. ? Carolicrinus,
Waag. & Jahn), Silurian, Gotland and (?) Bohemia, is more highly developed.
Between it and Compsocrinus we must imagine a form in which the free
Br became biserial, while the free rami forked several times. In Abaco-
crinus the proximal biserial brachials (IIIBr) with their pinnules are in-
corporated in the dorsal cup ; the stem has changed from sub-quadrangular

1 66

THE CRINOIDEA

to circular, but the columnals still alternate in size. From the imagined
intermediate form (not from Abacocrinus itself } Periechocrinus may have been
derived by fusion of 2 BB. FAMILY 3. CARPOCRINIDAE. Batocrinoidea
with 3 BB (? fused in Macarocrinus) ; RR rather large ; each half-ray
contains 1-3 IIBr usually passing into the free arms, which are usually
2, occasionally 3, to each ray ; iBr 2-5, in contact with iAmb ; illBr
may be present in limited and definite number ; tegmen of numerous
small plates, with a few larger ones ; in post. IR a vertical row of anals ;
stem large and round, usually with small pentagonal lumen. Genera —
Acacocrinus, W. & Sp. (1897), Silurian, Indiana, has 2 arms to each
ray ; brachials wedge-shaped, each bearing one pinnule. Desmidocrinus,
Angelin (1878), Silurian, Europe ; in each ray one of the rami forks
again almost immediately. Macarocrimis, Jaekel (1895), Lower Devonian,

Fio. LXXIX.

Anns of Carpocrinidae. 1,
Macarocrinus Springeri (after
Jaekol). 2, Carpoci-imis sim-
pler- (based on Brit. Mus. 5536
& 57360).

FIG. LXXX.

Barrci ndcocr inns sccptrnm. 1, reconstruction
of the crown, with three of the rami removed
(after G. Liljevall in W. & Sp.). Nat. size. 2,
diagram of the calyx. Lettering as usual.

Eifel, differs in greater length of Br, in fusion of BB, and presence
of one IIBr instead of two (Fig. LXXIX. 1). Carpocrinus, J. Miiller
(1841 ; synn. Phoenicocrinus, Austin ; Abracrinus, d'Orb. ; Hak'ocrinus,
Pionocriniu, and [?] Leptocrinus, Angelin, 1878. See W. & Sp., 1881),
Silurian, Europe ; derived from Desmidocrinus by fusion of free Br to
form uniserial ossicles each supporting 2 or more pinnules, and by atrophy
of the unpaired ramus, which resembles a large proximal pinnule (Fig.
LXXIX. 2). FAMILY 4. BARRANDEOCRINIDAE. Batocrinoidea with
3 BB ; RR irregularly shaped ; each half-ray containing 1 IIBr, which
supports a free arm, biserial, and recumbent over dorsal cup ; iBr, 3, in
contact with iAmb ; tegmen solid with large sub-spinous O, alternating
with 5 radial dome-plates ; stem large and round. . Genus — Barrandeo-
crinus, Angelin (1878 ; syn. [?] Cylicocrinus, S. A. Miller non Miiller ; see
W. & Sp., 1897), Silurian, Gotland and (?) N. America (Fig. LXXX.).

THE CRINOIDEA 167

The chief feature is the fixed recumbency of the arms, as in some Hexa-
crinidae and Acrocrinus ; but differing from those forms in that the
pinnules were very close-set, and folded in two rolls over the ventral
groove when closed. Except for this and the consequently more solid
tegmen, the genus is not far removed from the Carpocrinidae. Ant R is
hexagonal ; r. and 1. ant. RR heptagonal ; r. and 1. post. RR pentagonal.

FAMILY 5. COELOCRINIDAE. Batocrinoidea with 3 BB ; in each half-
ray, 1 or 2 IIBr ; free arms separated by 3 or more iBr in contact with
iAmb, 2-4 biserial rami to each ray ; a row of anals in post. IR, sup-
porting no tube ; but anus opening marginally from a slight prominence ;
tegmen solid, with large O, esp. post. 0 ; no respiratory pores known.
Genera — Coelocrinus, Meek & Worth. (1865-66 ; synn. Sphaerocrinus, M.
& W. non Roem. ; Aorocrinus, W. & Sp., 1897), Devonian and Carboni-
ferous, N. America, and (?) Devonian, Europe (Fig. LXXXL). 2 or 4
rami to each ray, each with independent opening into the theca ; cup extend-
ing below arm-bases. Dorycrinus, C. F. Roemer (1854, W. & Sp., 1897),
Carboniferous, N. America ; derived from Coelocrinus with convex base ;
rami paired, 2 to a single opening ; radial dome-plates of tegmen bear
large spines. Agaricocrinus, Hall ex Troost (1858 ; W. & Sp., 1897),
Carboniferous, N. America ; derived from Coelocrinus with concave base,
that feature being greatly exaggerated so that,
the cup does not project below the arm-bases ;
rami 2-4 in each ray, each with independent
opening. FAMILY 6. BATOCRINIDAE. Bato-
crinoidea with 3 BB ; each half-ray contains
1-5 IIBr, also IIIBr, 1-5, always in adanal
series, and sometimes in the rest, but rarely in
anterior series ; free arms with 1-4 biserial
rami in each ray ; iBr, 1-15, usually uncon-
nected with iAmb, except in post. IR, and
sometimes separated from them by the brachials

there also; plates in posterior IR 2-19, per- codocrin i^nat^s, from
haps more • anal tube long, usually stout, and Brit. Mns. E1773. x $.
central ; tegmen solid, with O and radial dome-
plates usually prominent, post. O always pronounced ; 10-20 respiratory
pores characteristic (Fig. XLV. 2), but doubtful in Dizygocrinus and very
doubtful in Hyperocrinus ; stem round, usually stout, with small pent-
agonal lumen. The family is confined to the Lower Carboniferous
rocks of central N. America, where it flourished exuberantly along
with the parallel family Coelocrinidae. Batocrinus came first (Kinder-
hook) and went last (St. Louis) ; the rest are confined to Burlington
and Keokuk Groups. Genera — Batocrinus, Casseday (1854 ; em. W.
& Sp., 1897), has 1-3 iBr separated from iAmb by IIIBr; 18-26
short arm-rami set regularly round periphery (Fig. LXXXI I.). JEretmo-
crinus, Lyon & Casseday (1859 ; W. & Sp., 1897), like Batocrinus,
but has 12-26 long paddle-shaped arm-rami, and eccentric anal tube.
AHoprosallocrinw, Ly. & Cass. (1860, W. & Sp., 1897), constructed like
Batocrinus, but converges in outward shape towards Agaricocrinus.
Eutrochocrinus, W. £ Sp. (1897), 18-40 short rami, set regularly around

1 68 THE CR1NOIDEA

periphery like spokes of a wheel ; iBr may or may not join iAmb ;
illBr often present Dizygocrinus, W. & Sp. (1897), like Eutrocho-
crinug, but with rounded calyx, more variable in composition, anal tube
and stem rather slender. Hyperocrinus, Meek & Worthen (1865, as
Uperocrinus; syn. Lobocrinus, W. & Sp., 1897); 18-22 free rami, arranged
in arm-groups, separated by iBr which join iAmb ; illBr sometimes
present ; lofty tegmen ; respiratory pores unknown, flfacrocrinus, W.
& Sp. (1897) ; 12-16 rami, in groups, but not separated by iBr
except in post. IR ; anal tube tapering ; stem slender ; tegmen coni-
cal or hemispherical ; respiratory pores, 10. The study of the relations
between these genera is a fertile and unappropriated field.

FAMILY 7. PERIECHOCRINIDAE. Batocrmoidea with 3 BB ; IBi^
hexagonal; each half-ray contains 1-5 IIBr, and sometimes 2-6 IIIBr ;
free arm -rami biserial, usually branching ; iBr numerous and merging

FlO. LXXX1I.

Batocrinus. 1, B. ii'nsidtictylus, calyx from the side (after Casseday, nat. size). 2, cup
seen from aboral side, from Brit. Mus. B5055. Supplementary plates shaded, x J.

into iAmb ; post. IR wide, with plates in successive rows of 1,3, 4-6, etc. ;
tegmen of numerous small plates in which O, Amb, and radial dome-plates
are sometimes to be distinguished, especially post. O ; anus without tube,
from sub-central to marginal ; stem large, round, with wide lumen, round
or 5-lobed. Members of this family, esp. Gennaeocrinm, are liable to be
confounded with Actinocrinidae, but differ in the presence of 3 plates
(not 2) in the second row of post. IR, in which respect they resemble
Xenocrinidae, Carpocrinidae, Coelocrinidae, and Batocrinidae. Genera —
Periechocrinus, Austin (1843 ; synn. Geocrinus, d'Orbigny ; Saccocrinus,
Hall ex Troost ; Pyxidocrinus, J. Muller, pars ; [?] Trochocrinus, Portlock
and [?] Pander ; [?] Pradocrinus, Verneuil, 1850 ; see W. & Sp., 1897),
Silurian to Carboniferous, Europe, N. America, Australia ; elongate cup
of thin long plates, usually with axial folds, and depressed theca in which
0 are not distinguishable ; arms fork and are less advanced than in
Abacocrinus (p. 165). Beyrichocrinus, Waag. & Jahn (1899), Silurian,
Bohemia, little known. Megistocrinus, Owen & Shumard (1852 ; W. &
Sp., 1897), Devonian and Carboniferous, N. America, (?) Carboniferous,

THE CRINOIDEA 169

England ; globose cup of heavy short plates, with tegmen from flat to
conical ; 0, radial dome-plates, and Amb usually distinguishable (Fig.
XLII.) ; free rami grouped in pairs, have covering-plates and side-plates.
Gennaeocrinus, W. & Sp. (1881-97), Devonian, N. America ; low cup of
thin plates with axial folds ; theca rather depressed and lobed in arm
regions ; O small but visible, as also are Amb ; rami, 8 to a ray, branch
from alternate sides of arms ; in form like the Actinocrinid Physetocrinus.

SUB-ORDER 3. Actinocrinoidea. RR in lateral contact except in
post. IR ; proximal anal hexagonal ; IBrj usually hexagonal ; BB 3
equal, forming a hexagon.

FAMILY 1. ACTINOCRINIDAE. Cup conical or bowl-shaped, with orna-
ment of axial folds ; only 1 IIBr, which is axillary ; free arm-rami branch
on alternate sides of the half-rays, starting either from lax or I lax ; all
free portions of arms are biserial ; proximal pinnulars bear hooks ; iBr
numerous, primitively merge into iAmb, but become separated therefrom

Fio. LXXX111.

Actinocrinus. 1, analysis of part of cup, showing post, and r. post, interradii. 2, cup from
below ; the supplementary plates shaded.

by development and lateral contact of fixed brachials ; post. IR wide,
with proximal anal followed by 2 plates in second row ; tegmen
solid ; O not prominent, but usually distinguishable ; iAmb may cover
the tegminal Amb ; anus either on a tube or piercing tegmen, either
central or eccentric ; stem round, with small 5 -lobed lumen. The
family is confined to the Lower Carboniferous, and flourished chiefly
in N. America. Actinocrinns, with its descendant Steyanocrinus, and
Cactocrinus, with its descendant Teleiocrinus, form a group characterised by
a long anal tube. In Physetocrinus and its descendant Strotocrinus the anus
pierces the tegmen directly. For complete account of the family and
revision of genera see W. & Sp., 1897. Genera — Actinocrinus, Miller (1821;
synn. Amphora, Cumb. ; Blairocrinus, S. A. Miller; Fig. LXXXIIL) ; cup-
plates with axial folds and parallel ridges ; theca lobed at arm-regions, the
numerous iBr being depressed, while the higher orders of Br with their
Amb form 5 broad rays, sometimes including some iBr ; these rays fork and
give off free rami to alternate sides every second or third plate, first on outer
side from Ilax, then on inner side from 1 1 lax, and so on ; the ramimay branch
again ; pinnules long, the proximal pinnulars armed with a small hook pro-
jecting from the middle ; tegmen solid, rising centrally into a stout tube with

i;o THE CRINOIDEA

anus at distal end ; 0 small and eccentric ; Amb visible in tegmen, either
as two rows of large covering-plates which pass out from between 0 and
follow the branching of the food-grooves, or as large single plates (lAmb
and IIAmb), succeeded by small covering-plates, which pass on to the free
arms. Steganocrinus, Meek & Worthen (1866), differs from Actinocrinus
in the extension of each ray into 1 or 2 rigid tubular rami, from
which biserial ramuli are given off alternately on opposite sides, either
from every ossicle or from every other ; covering-plates consist of side-
pieces and spinous Amb. Cactocrinus (W. & Sp., 1897) differs from
Actinocrinus in having the arm-rami given off in a continuous row
around the theca, the brachials meeting laterally, so that iBr are not in
contact with iAmb (cf. Eutrochocrinus) ; bifurcation takes place on each
successive plate, i.e. all Br, except the finials, are axillary ; all pinnulars,
except the extreme distal ones, bear hooks which imbricate over the adjacent
pinnules (cf. structure of a feather) ; covering-plates consist of side-pieces
and Amb ; Br and tegminal plates spinous (cf. Fig. XLIV.). Teleiocrinus, W.
& Sp. (1881 ; syn. Calatliocrinus, Hall, pars, non v. Meyer) ; a modified
Cactocrinus in which the arms have become so numerous and crowded
that they are pushed outward, while their bases have become united
and extended as a broad rim at the top of the theca ; between the Amb
and iAmb forming the roof of this rim, and the Br forming its floor, are
developed processes of stereom serving as girders ; the ambulacra with
their ambulacrals are mostly depressed below the tegmen and covered by
iAmb ; a respiratory pore is at the side of each ambulacral opening into
the theca. Physetocrinus, Meek & Worthen (1869), differs from Actino-
crinus in the absence of an anal tube, and in bifurcation of arms on each
successive Br, as in Cactocrinus. Strotocrinus, Meek & Worthen (1866),
often confounded with Teleiocrinus, towards which it is convergent, bears
to Physetocrinus precisely the same relation as Teleiocrinus bears to Cacto-
crinus. Sampsonocrinus, Mill. & Gurl. (1895), Carboniferous, Missouri,
has r. and L post. iBr truncating BB, and only 1 IBr in r. and 1. ant. radii ;
the unique specimen is best regarded as an abnormal Actinocrinus. Com-
pare Phillipsocrinus, M'Coy (1844 or 1862), Carboniferous, Ireland, in
which also some iBr truncate BB ; but since BB are here 4, this may
be a more direct descendant of Xenocrinidae, perhaps by way of the
Silurian Laubeocrinus, Waagen & Jahn (1899), which seems a link
between early Batocrinoidea and Actinocrinoidea. FAMILY 2. AMPHORA-
CRINIDAE. Actinocrinoidea with cup depressed and tegmen much elevated,
accompanied by downward projection of proximal regions of arm-rays ;
cup-plates with granule -vermicular ornament; iBr few; in other
respects like Actinocrinidae. Genus — Amphoracrinus, Austin (1848),
Carboniferous, Europe and N. America ; arms free from lax or IIBrj ;
anal tube short and eccentric. Two American species occasionally have
3 plates (instead of 2) succeeding the proximal anal, but the middle of
these is small and wedge-shaped, barely touching the anal ; this may be
a plate of the ensuing row pressed down, or it may represent the middle
plate in Batocrinoidea.

THE CRINOIDEA 171

SUB-CLASS 2. DICYCLICA, Bather (1899).

Crinoidea in which the base consists of BB and IBB, the latter being
liable to atrophy or fusion with the proximale, but the aboral prolonga-
tions of the chambered organ are always radial ; new columnals may or
may not be introduced at the proximal end of the stem.

ORDER l. Dicyclica Inadunata
( = INADUNATA, W. & Sp. pars, emend.}.

Dicyclica in which the dorsal cup primitively is confined to the patina
and occasional intercalated anals, and no other plates ever occur between
RR (Grade : Distincta) ; Br may be incorporated in the cup, with or
without iBr, but never rigidly, and their corresponding Amb remain
supra-tegminal (Grade : Articulata) ; new columnals are introduced at
the extreme proximal end of the stem.

This order, so far as its Palaeozoic genera are concerned, corresponds
roughly to the Inadunata Fistulata of Wachsinuth & Springer, and
entirely to the Inadunata Dicyclica of Bather with an error or two
corrected ; but it includes also some of Miiller's Articulata and some of
Wachsmuth & Springer's Larviformia. The latter authors have them-
selves proved the connection of the Encrinidae and Pentacrinidae with
their Fistulata. The distribution of the 70 genera into families would
present no great difficulty, were a purely morphological classification our
aim. One might use, as has been done, such characters as the presence
or absence of pinnules, of an anal tube, of a radianal, of articulation
between cup-plates, or of simply bifurcate as contrasted with dichotomous
arms. But there is every gradation in the development of these characters,
pinnulate forms being derived from non-pinnulate, the radianal gradually
disappearing, articulation of plates developed as need arose, and so on.
Hence the great division into Cyathocrinidae and Poteriocrinidae (W. &
Sp., 1886 ; Zittel, 1895) cannot meet the needs of the phylogenist. An
attempt to sketch the actual race-history (Bather, 1890) resulted in the
recognition of a distinction between Dendrocrinus and its allies, with their
broad radial facets and thin tegmen on the one hand, and Cyatlwcrinus
and its allies, with narrower facets and more solid tegmen 6n the other,
while the pinnulate forms were all derived from the Dendrocrinidae.
This distinction, subsequently strengthened (Bather, 1893), has been
made much of by Jaekel (1895), who divides all his Fistulata into
Cyathocrinacea, Dendrocrinacea, and Poteriocrinacea, the last group being
derived from the Dendrocrinacea, and giving rise to the Articulata
(Jaekel). It is therefore but a slight step to establish two sub-orders,
Dendrocrinoidea and Cyathocrinoidea. Of these the latter were the first
to be specialised and the first to disappear. The Dendrocrinoidea moved
more slowly and went further — even to our own day — undergoing modi-
fication in the development of the anal tube, in the pinnulation of the
arms, and in the relation of arms to cup. Moreover, from them branched
off the order Flexibilia, probably on more than one occasion.

172

THE CRINOIDEA

FIG. LXXXIV.
Analysis of cup of Carabocrinus.

SUB-ORDER 1. Cyathocrinoidea. Dicyclica Inadunata, with a fairly
stout tegmen, in which 5 orals (A, sub-ambulacrals, interradials, con-
solidating apparatus, of authors) are usually conspicuous, helping to
stiffen the tegmen, supporting the ambulacra on their adjacent edges, and
enclosing but not covering the peristome ; post. O frequently a madre-
porite ; radial facet usually narrow ; arms distinct from dorsal cup, un-
branched or dichotomous ; none attain the pinnulate stage, but the
presence of pinnules would not in itself remove a genus from the sub- order.
FAMILY 1. CARABOCRINIDAE. Cyathocrinoidea with one or more
large anals in line with RR ; RA supporting these, and resting on a
supplemental plate intercalated between post. B and r. post B and rest-
ing on IBB. Anus, surrounded by a few small plates, pierces tegmen
between x and post O. Strong stereom -folds pass across the radio-
oral sutures. Post O pierced by hydro-
pore (a madreporite). Arms stout and short.
Genera — Carabocrinus, Billings (1856,-59) ;
Strophocriniis, Sardeson (1899), both Ordo-
vician, N. America (Fig. LXXXIV.). The
radio -oral folds are probably vestigial
hydrospires (cf. Hybocrinus). The inter-
radial tegminal plates are admittedly homologous with orals, and ambu-
lacrals rest on their apposed edges (as in Fig. V. 2 ; see also p. 126).
FAMILY 2. PALAEOCRINIDAE. Cyathocrinoidea with anal x in line with
RR ; RA smaller and rhomboidal, abutting on x and not separating
r. post Rs from r. post B ; anal tube slightly developed ; 5 0 surround a
pentagonal peristome ; post 0 a madreporite ; arms narrow, rising from
well-defined facet, axial canal not separate from ventral groove, but passes
into thecal cavity through a large opening between R and adjacent
O. Genera — Palaeocrinus, Billings (1859), Ordovician, Canada (Fig.
LXXXV). ; usually regarded as a synonym of Dendrocrinus, to which it
is closely allied ; but it differs in shape of RA, the defined radial facet,
the anal tube composed
of but 4 or 5 vertical
rows of plates, and above
all, the solid orals sup-
porting ambulacrals, —
differentiae which place
it in this sub -order.
Stem shows 5 radial
sutures ; cup - plates
usually folded ; arms
isotomous to IVBr.
Porocrinus, Billings (1856-59 ; see J. Grant, 1881, and Beyrich),
Ordovician, Canada and Russia (Fig. LXXXVL); arms unbranched.
Deep folds lie at the angles of all thecal plates, directed towards the
angle, and not passing at right angles across the middle of the sutures,
thus differing from hydrospires of Eublastoidea and from pectinirhombs ;
Beyrich imagined them to be separated by suture from the rest of the
plate, and Grant described a membrane [? a film of epistereom] covering

IB

Fio. LXXXV.

Analysis of cup of Palaeocrinus, showing the axial
folds of the plates. R' radianaT.

THE CRINOIDEA

173

Or

them. Bactrocrinus, Schnur in Steininger (1849), Devonian, Eifel ; usu-
ally made a synonym of Homocrinus, but separated by Zittel (1879), Bather
(1893), and Jaekel (1895) ; differs from Palaeocrinus only in the occa-
sionally wider radial facet and rather more de-
veloped anal tube, in which points it approaches
Homocrinus. FAMILY 3. EUSPIROCRINIDAE.
Cyathocrinoidea with anal x hexagonal or
heptagonal, resting on post. B, but rising
above level of RR ; with RA pentagonal,
resting on post, and r. post. BB, supporting
x on one side, r. post. Rs on the other, and
a plate of the anal tube (r£) sunk into the
cup between them ; anus at end of a massive
anal tube ; post. O a madreporite ; arms
dichotomous, axial canal not separate from ven-
tral groove. Genera — Euspirocrinus, Angelin
(1878; see Bather, 1893), Ordovician of
Canada, Silurian of Gotland (Figs. LXXXVII.
and XXXVIII.). The usual text-book figure
of E. spiralis is reversed. Closterocrinus, Hall
(1852), and Ampheristocrinus, Hall (1882),
both Silurian, N. America ; 3 IBB ; imper-
fectly known (Fig. LXXXVIIL). The anal
area of the family resembles that of the
advanced Dendrocrinoidea. FAMILY 4. SPHAE-
ROCRINIDAE. Cyathocrinoidea, with 3 anal

plates as in Empirocrinus but differing in reconstructioil) seen from right
that RA is comparatively large, x not rising posterior radius, the arm of
above RR, rt small and not, or hardly at all, £" tTSi SS2£
rising above RR ; post. 0 a madreporite ; (Based on specimens belonging
arms isotomous ; axial canal separate in Br j£ai£r- 2?opUcai^ection across
and R. Genera— Spliaerocrinus, C. F. Roemer a. brachiai (Br), showing reia-

' tions of covering - plates (c.p).

(1851; for history see Bather, 1892; for x 4 diam. other lettering as
structure, Jaekel, 1895), Devonian, Germany usuaL
and England ; the anus pierces the tegmen directly through a ring
of small plates. Parisocrinus, W. & Sp. (1879), Devonian of Ger-
many, Carboniferous of England and N. America (Fig. XXVI. 6), has a
well-developed anal tube of hexagonal plates, which are folded at the edges.

Sf-

Fio. LXXXV1.
Porocrinus Smithi. 1, partial

Fio. LXXXVII. FIG. LXXXVIIL

Analysis of cup of Euspirocrinus sidrcdis. Analysis of cup of Amplieristocrintis.

FAMILY 5. CYATHOCRINIDAE. Cyathocrinoidea, with no anal except xy
which is in line with RR, and usually supports a large tube ; post. O a
madreporite ; arms isotomous, axial canal separate or not. Genera —
Cyathocrinus, Miller (1821 ; see Bather, 1892-93, for full revision and

174

THE CRINOIDEA

details), Silurian to Carboniferous, world-wide. This, being the best
known genus of the sub-order, demands closer description. The dorsal
cup (analysed in Fig. LXXXIX. 5) consists of: 5 equal IBB ; 5 large
BB, all hexagonal except post B, which is truncate above ; 5 shield-
shaped RE, each with a facet usually \ width of plate, the articular
surface being either smooth and imperforate, or having a slight transverse
ridge pierced by the axial canal (Fig. LXXXIX. 2); a square anal x, in
line with RR and resting on truncate post. B. Tegmen consists of 5 O
resting on the incurved shoulders of RR, and surrounding a pentagonal

O O

Fi<:. LXXXIX.

Cyathocrinw. I, C. muUibnuhiatvt, seen from posterior. (Brit. HIM. E5402.) 2, radial
of C. visbycensit, showing articular facet, x 2. 3, joint-surface of columnal of C. acinotubut,
x 2. 4, joint-stirface of a brachial of same, with covering-plates in position, x 9. 5, analysis
of the cap. a.c, axial canal ; /, fulcral ridge ; t, intercalated plate ; j, joint-surface ; t.c* and
I c* the twp halves of a covering-plate of the left side ; nn, nodals ; r.c1, proximal half of a
covering-plate of the right side ; t, anal tube ; v. g, ventral groove. (2-5 after Bather.)

peristome (Fig. XXXIX. 1); post. O being large, conspicuous, and pierced
by numerous water -pores, the other 0 being smaller and often almost
entirely hidden, partly by ambulacrals resting on the apposed edge of all
O, partly by small interambulacrals (Fig. XXXIX. 2). The proximal Amb
meet over the peristome and often grow to a large size, sometimes fusing
and simulating orals (for which elements they are taken by W. & Sp.,
see Fig. XLIII.). The anal tube consists of more or less hexagonal plates,
arranged in fairly regular longitudinal rows ; it varies greatly in width,
length, and width of lumen ; the anus is at its distal end ; the plates may
be slightly folded, but are not transversely elongate, nor are there pores,

THE CRINO1DEA

175

FIG. XC.
Analysis of cup of Giuocrinu*.

or even the appearances of pores, between them (Fig. XXVI. 5). Arms
dichotomise 5-7 times, and in each series there are more brachials in the
admedian branch of the dichotom (Fig. LXXXIX. 1). The covering-plates
are well developed ; in their simplest form they are conical, in both out-
line and longitudinal section, regularly alternating, and each reaching
about !• across the ventral surface ; each covering-plate may, however, be
transversely divided, and the parts may come to be arranged in a manner
too complicated for description here (Fig. LXXXIX. 4). Stem round,
with lumen usually 5-lobed ; stem and lumen vary in width ; columnals
low, usually alternating in thickness and height, the smaller ones being
those last formed; joint-surface radiately striate (Fig. LXXXIX. 3); no
longitudinal sutures and no cirri. Gissocrinus, Angelin (1878 ; em. Bather,
1893), Silurian, Europe and possibly
America (Fig. XC. ; see also Figs. VIII., IX.,
and XLL); connected with Palcwocrinus ;
one or two pairs of IBB usually fuse ; cup-
plates have clear axial ridges ; distal mar-
gins of Br usually project ; anal tube com-
pressed antero-posteriorly, its plates trans-
versely elongate and folded. Arachnocrinus, Meek & Worthen (1866),
Silurian to Devonian, America and Europe ; small cup and heavy arms,
which, together with anal tube, spread out horizontally from the cup.
Lecythocrimis, Miiller (1858, em. Zittel, 1879 = Taxocrinus briareus,
Schultze, 1866), Devonian, Eifel ; stem subquadrangular, with one large
central and 4 smaller peripheral canals (cf. Cupressocrinidae). FAMILY 6.
PETALOCRINIDAE. Cyathocrinoidea without x in dorsal cup, and with

Fio. XCI.

Petalocrinus. 1, partial recon-
struction of P. mirabilis, with one
arm removed to expose radial
facet, and other arms devoid of
covering-plates; the root is im-
aginary, x 2 diain. 2, section
across four grooves of an ami-fan,
showing traces of the original
sutures (s) between them, covering-
plates (c.p) closed and open, also
stages in the separation of the
axial canal (o.c) from the ventral
groove (v.g), compare Fig. VIII.,
x 5 cliam. 3, articular facet of
arm -fan of P. visbycensis ,• ro,
muscle - fossa ; I, dorsal ligament-
fossa; r, fulcral ridge, x |. 4,
dorsal view of cup and proximal
regions of amis of P. mirabilis ;
Br, arm-fan ; St, proximal colum-
nal ; other letters as usual, x 3
diain. (3 and 4 are after Bather.)

arms fused into solid arm-fans. Genus — Petalocrinus, Weller (1896),
Silurian, Gotland and N. America, appears to have been derived from
Arachnocrinus by lateral fusion of the rami of each arm to form a blade

1 76

THE CR1NOIDEA

or fan articulated to the R by a single IBr (Fig. XCI., see Bather, 1898).
FAMILY 7. CROTALOCRINIDAK. Cyathocrinoidea with cup as in Gyatho-
crinus ; anal tube when present constructed like that of Gissocrinus ;
tegmen almost entirely composed of Amb, some of which are much
modified ; the orals seem to have been covered by these and to have
atrophied, except post. O, which remains as a conspicuous plate, ap-
parently madreporic (Fig. XCII. 3) ; the entry of Amb into the tegmen is
connected with the shortening up of proximal portions of arms, so that
IBr, IIBr, IIIBr, and sometimes IVBr, all partially rest on R, and are
firmly united by suture with it (Fig. XCII. 1). Arms repeatedly iso-
tomous ; axial canal distinct, except sometimes in extreme distal region.

IR

VBr,

ac.—

Fin. XCII.

Crotaloctinidae. 1 and 2, Enattocriniis scriptus (after Wachsm. & Spr.). 1, posterior view
of cup and arm-bases. 2, enlarged view of more distal portions of anus ; a from side, b from
back, showing cornice-like projections. 3, tegmen of Crotalocriniis pulclier, from a specimen at
Stockholm. The arm-branches are united by the lateral processes of the brachials, the spaces
between being represented in solid black ; the interradii (IR), along which adjacent arms unite,
appear as five depressed lanceolate areas, in the posterior of which lies the short anal tube (A*) ;
the axial canals (a.c) are separate from the ventral grooves (v.g)', the latter are protected by
covering-plates (c.?<), which become larger towards the centre, and four proximal ones (P) meet
around the madreporite (M), x 2 diam.

Stem large, round, with wide lumen (Fig. L. 1, 3, 4). All Silurian.
Genera — Enallocrinus, d'Orbigny (1850), Gotland and England ; arm-
branches distinct, often with a pronounced cornice at distal margins of Br
(Fig. XCII. 2), in this as in other respects closely resembling Gissocrinus.
Crotalocrinus, Austin (1842 ; syn. Anihocrinus, Muller, 1853), Gotland,
England, and N. America ; arm-branches united by lateral processes from
each Br, so as to form a flexible network, which may be continuous all
round the crown, or divided into 5 broad arm-fans. The family is referred
to the Camerata by Wachsmuth and Springer (188 8); but the resemblance
to Marsipocrinus is homoplastic, and the connection with Cyathocrinidae
scarcely admits of question. The Crotalocrinidae might be called the
Adunata of the Dicyclica, just as Platycrinus and its allies are of the
Monocyclica. FAMILY 8. CODIACRINIDAE. Cyathocrinoidea with no anal

THE CRINOIDEA

177

'O

FIG. XCIII.

plates in dorsal cup ; with diehotomous arms relatively slightly developed.
Genera — Codiacrinus, Schultze (1867 ; Follmann, 1887), Devonian, Ger-
many (Fig. XCIIL). Lecythiocrinus, White,
(1880), Coal Measures of Illinois and Kansas.
Both genera little known ; Codiacrinus is
compared by Schultze with Myrtillocrinus,
by W. & Sp. with Achradocrinus. FAMILY 9.
CUPKESSOCRINIDAE. Cyathocrinoidea with
no anal plates in dorsal cup ; anus piercing
tegmen ; arms unbranched (or [?] forking few
times) ; stem square in section, with an axial and 4 peripheral canals (cf.
Lecythocrinus, p. 175). Genera — Cupressocrinus, Goldfuss (1826 ; synn.
Halocrinus & Cypellocrinus of Steininger ; see W. & Sp., 1886, and
Neumayr, 1889), Devonian, Germany and England (Fig. XCIV.). A
massive form with basin-shaped cup ; IBB fused (by some held to be a
proximale, Fig. XCIV. 3) ; a stout arm, composed of a few large Br in
single series, rests on a facet the full width of each R ; large ambulacrals
cover the arm-grooves and are taken for pinnules by Zittel ; the solidity of
the close-fitting arms renders a plated tegmen unnecessary, but the 5 orals

Analysis of cup of Codiacrinits
Schultzei. * shows position of anus.

FIG. XCIV.

Cupressocrinus (after L.
Schultze). All two -thirds nat.
size. 1, C. inflatus, complete
crown. 2, C. abbrcviatus, ventral
surface of calyx, slightly modi-
fied ; shows " consolidating ap-
paratus" of live plates (A), here
regarded as orals, between which
are the passages (v.g) for organs
of the ventral groove other than
the food-groove ; a.c, axial canal
of radial ; As, passage for rectum.
3, infrabasal circlet of same, from
below, showing sutural surfaces
(s) for basals, axial canal, and
peripheral canals (p.c).

characteristic of Cyathocrinoidea are retained as a frame (" consolidating
apparatus" of authors) around the ventral surface of the cup, post. 0
being larger than the others and pierced for the rectum (Fig. XCIV. 2).
Myrtillocrinus, Sandberger (1855, syn. Ancyrocrinus, Hall, see p. 134,
Fig. LI.), Devonian, Germany, and N. America ; may have closer affinities
with Gasterocomidae. FAMILY 10. GASTEROCOMIDAE. Cyathocrinoidea with
anus in side of cup, at a level varying with the genus, either above or
below a;, which is always within limits of cup. Arms apparently small,
borne on a narrow horseshoe-shaped facet, with distinct axial canal. O
for most part covered by Amb, post. 0 a large madreporite. IBB small,
often fused into 3 plates or 1. Stem usually of Cupressocrinid type.
Genera — Gasterocoma, Goldfuss (1839, W. & Sp., 1886 ; Jaekel, 1895 ;
synn. Epactocrinus and Ceramocrinus, J. Miiller, 1855), Devonian, Ger-
many ; anus below xt which is in line with RR ; IBB fused into one.
Nanocrinus, J. Miiller (1856, Schultze, 1867), Devonian, Eifel ; like

12

THE CRINOIDEA

Fio. XCV.

Analysis of cup of Achradocrinus
veutrosus.

Gasterocoma, but ant. R small, without facet ; r. ant. R with 2 facets. Scolio-
crinus, Jaekel (1895), Devonian, Eifel ; still more bilateral, in that ant.

and 1. ant. RR are larger than the rest, and
alone bear arms ; anus below r. post. R,
and x between post. B and r. post. B.
Achradocrinus, Schultze (1867), Devonian,
Eifel (Fig. XCV.) ; x below anus, and
resting on post B ; IBB 5 ; stem round,
with single canal. Hypocrinus, Beyrich
(1864), Carboniferous, Timor ; differs from
Achradocrinus in having IBB fused to 3 ; referred by most authors to
Cystidea Aporita.

SUB-ORDER 2. Dendrocrinoidea. Dicyclica Inadunata with a thin
flexible tegmen, or with the ventral surface almost entirely occupied by
a large anal tube or ventral sac (the latter name being needed if the
extension contained more than the mere rectum) ; orals inconspicuous or
entirely atrophied in the adult ; no madreporite ; radial facet often wide,
so that the distinctness of arms from dorsal cup is not maintained ; arms
dichotomous, the dichotomy often irregular, leading up to a pinnulate
stage.

Whereas the genera of the Cyathocrinoidea all have the arms either
quite distinct from each other above the level of the patina, or at least
not united by supplementary plates, the Dendrocrinoidea gradually attain
a stage of development in which the arms are thus partially united.
Below this stage we may draw a somewhat arbitrary line, separating the
former as a grade, Distincta, from the latter grade — Articulata. This
line happens to correspond with the break between Palaeozoic and
Mesozoic time. We deal first with the Dendrocrinoidea Distincta.
FAMILY 1. DENDROCRINIDAE. Dendrocrinoidea with regularly dich-
otomous, non-pinnulate arms, with anal x, and large RA in its

Honoerinns. Fio. XCVI. Thenarocrinu*.

Dendrocrinidae. Analyses of cups.

primitive position as inferradial (Fig. XCVI.); stem quinquepartite.
Genera— Merocrinus, Walcott (1883), Ordovician, N. America and England,
resembles locrinus (p. 145) in all but the presence of IBB ; anal tube sup-
ported by the left shoulder of r. post Rs. Ottawacrinus, W. R. Billing*

THE CR IN OWE A 179

(1887), Ordovician, Canada; a; rests on post. B, RA immediately above
r. post. B ; r. post. Rs is above general level of RR and may be !Brr These
two genera suggest that RA of Dicyclica may not be strictly homologous
with RA of Monocyclica. Dendrocrinus, Hall (1852), Ordovician and
Silurian, N. America ; large anal sac with folded plates making wide anal
area in cup (see also Fig. XXVI. 2, 3). Homocrinus, Hall (1852. em.
Bather, 1893), Silurian and Devonian, N. America and Europe ; RA
rhomboid and smaller. Thenarocrinus, Bather (1890), Silurian, England ;
large sac, anal area widened by RA sinking between BB. FAMILY 2.
BOTRYOCRINIDAE. Dendrocrinoidea with arms bifurcating in two main
rami with armlets or pinnules ; RA usually small and quadrangular
or absent. This family connects Dendrocrinidae with Decadocrinidae,
and it is difficult to diagnose it so as to include all of the closely
related forms. Genera — Botryocrinus, Angelin (1878, em. Bather, 1891 ;
syn. Sicyocrinus, Ang.), Silurian and Devonian, Gotland, England, Canada;
small RA, large anal sac, often coiled ; arms range from irregularly
dichotomous, through ramuliferous, to pinnulate (Fig. XCVIL, see also
Figs. III., XIII., and XXL). Gothocrinus, Bather (1893), Silurian, Gotland,
has a cup like Dendrocrinus with ramuliferous arms. Mastigocrinus,
Bather (1892), Silurian, England, has very long, finely dichotomous arms,

,RA

FIG. XCVII. FIG. XCVIII.

Analysis of cup in Botryocrinus. Analysis of cup in Atdcstoerimis.

no RA (for anal sac, see Fig. XLVIII.). Gastrocrinus and Rhadinocrinus,
Jaekel (1895), Devonian, Germany, are, respectively, like a Botryocrinus
with cirri in whorls of 5 (Fig. XVII. 1), and a Gothocrinus with minute
ramules. Cosmocrinus, Jaekel (1898), Devonian, Germany and N. America,
has broad cup of Botryocrinus type, arms pinnulate, and with rarnuli on the
inner side of the two main rami. Vasocrinus, Lyon (1857, em. W. &
Sp., 1879), Devonian to Carboniferous, N. America, scarcely differs from
Botryocrinus. Barycrinus, Meek & Worthen (1868, em. W. & Sp., 1879),
Carboniferous, N. America and England, has heavier cup-plates and large
stem-lumen. Goniocrinus, Miller & Gurley (1890), Waverly Group,
Iowa, has no RA, and has cirri like Gastrocrinus. Atelestocrinus, W. & Sp.
(1886), Carboniferous, N. America, is distinguished by absence of an arm
from ant. R; large RA supporting rt (Fig. XCVIII.) ; arms ramuliferous.
Streptocrinus, W. & Sp. (1886 ; redescr. Bather, 1893 ; syn. Ophiocrinus,
Ang. non Salter), has branched arms, coiling inwards, with pinnule-like
processes arising irregularly from sides of Br ; 110 RA, anal sac coiled.
FAMILY 3. LOPHOCRINIDAE. Dendrocrinoidea with only 1 ramus to each
arm, with ramuli springing from the alternate sides of every second Br ;
with no RA, but x supporting on its shoulders 2 plates of a delicate
anal sac. Genus — Lophocrinus, H. v. Meyer (1858 ; redescr. Jaekel,
1895 ; syn. ? Carduocrinus, v. Koenen), Upper Carboniferous, Germany.

i8o THE CRINOIDEA

A similar evolution of arm-structure to that seen in Botryocrinidae
produced a long series of genera with pinnulate arms, for the most part
clearly forking into two rami, and rarely branching more than once
again, but in some genera branching more often. All at first had an
anal area more complicated than that of Dendrocrinus, in that a. fresh
plate (rt) was inserted between x and r. post. R so as to rest on RA,
while a corresponding plate (It) appeared on the left of x. FAMILY 4.
SCAPHIOCRINIDAE ( = Poteriocrinidae, Anctt., greatly restricted). Dendro-
crinoidea with dichotomous, usually much branched, pinnulate, stout arms,
with facet occupying nearly full width of R ; with x, RA, rt, and It, in
anal area of cup, supporting a large plicated sac (Fig. XCIX.). Carbon-
iferous of N. America and Europe, a few Devonian. Genera — Scaphio-
crinus, Hall (1858, em. W. & Sp., 1886 ; synn. Hydriocrinus, Trautschold ;
Abrotocrinus, Mill. & Gurl.), and Poteriocrinus, Miller (1821, em. W. & Sp.,
1881), differ in little but the greater definiteness and less width of the
facet in the latter; each has a long anal sac (Fig. XXVI. 4). Woodocrinus, de
Koninck (1854 ; synn. Philocrinus, de Kon. ; Pachylocrinus, W. & Sp.), has
shorter cup, arms, and sac. In Anlocrinus, W. & Sp. (1897), the sac forks.
Zeacrinns, Hall (1858, em. W. & Sp., 1886), has a short, stout sac, around

Poteriocriints. FIG. XCIX. Zeacrinus.

.ScaphiocrinMuc. Analyses of cups.

which the wide arms fit closely. Coeliocrinus, White (1863), and Hydreiono-
crinus, de Kon. (1858), differ from Zeacrinus in having the sac respectively
balloon-shaped and mushroom-shaped. Bursacrinus, Meek & Worthen
(1861 ; syn. Synyphocrinus, Trautschold), has arms like Zeacrinus, but no
anal except x in the cup ; it is in the latter respect the morphological
equivalent of Graphiocrinus (infra). FAMILY 5. SCYTALECRINIDAE. Den-
drocrinoidea with forked, pinnulate, slender arms ; anal structures as in
Scapliiocrinus. Genera — Scytalecrinus, W. & Sp. (1879, syn. Dactylocrinus,
Sladen non Quenst.), and Decadocritms, W. & Sp. (1879), both Carboniferous,
differ chiefly in shape of cup, conical in the former, saucer-shaped in the
latter, which thus leads on to : FAMILY 6. GRAPHIOCRINIDAE. Dendrocrin-
oidea with forked pinnulate arms and saucer-shaped cup, concave at base,
and containing x, but no RA (Fig. C.). In many points resemble the earlier
Encrinidae, but have not such thick cup-plates or large muscle-fossae.
Middle and Upper Carboniferous. Genera — Graphiocrinus, de Koninck
(1853), Aesiocrinus, Miller & Gurley (1890), Delocrinus, Miller & Gurley
(1890 ; syn. Ceriocrinus, White non Desor). In this and the succeeding
families the biserial arrangement of Br first assumes prominence ; it had
already appeared as an occasional gerontic character at the distal end of
the rami, out now is found in all except their most proximal portions,
accompanied by shortening of the arm. FAMILY 7. CROMYOGRINIDAE.
Dendrocrinoidea with simple or bifurcating, stout, pinnulate, usually

THE CRINOIDEA

181

biserial amis ; with cup- bowl- or saucer-shaped, composed of stout plates,
RR having muscle-fossae gradually more pronounced ; with x always, and
RA usually, present in cup, RA often supporting rt (Fig. C.). There is
much confusion in the nomenclature of this and allied families, and the
following names may not be Used in what will ultimately prove the correct
sense. Genera — Cromyocrinus, Trautschold (1867), and Eupachy crimes,
Meek & Worthen (1855, em. W. & Sp., 1886), Middle and Upper Car-
boniferous, Europe and N. America, are closely allied ; x, RA, and rt in
anal area. Ayassizocrinus, Shumard ex Troost MS. (1853; syn. Astylocrinus,
C. F. Roemer), Kaskaskia group, N. America, is a Cromyocrinus that loses
its stem in adult life, while IBB fuse to a solid mass (cf. Edriocrinue,
p. 191). Tribmchiocrinus, M'Coy (1847, redescr. R. Etheridge, fil., 1892 ;
syn. Pentadia, Dana, pars), Per mo-Carboniferous, Australia, has 1. and r.

Krisocriiins. FIG. C. Tribmchiocrinus.

Analyses of cups of Graphiocrinidae, Cromyocrinidae, and Eucrinidae.

pairs of IBB fused ; single arms borne by ant., r. post., and 1. post.
RR ; whether the other RR bore arms is a moot point. Phialocrinus,
Trautschold (1879, em. R. Etheridge, fil., 1892, non Eichwald ; syn.
Pentadia, Dana, pars), Permo- Carboniferous, Australia, Russia, India ;
differs from Acsiocrinus and Ceriocrinus in little but greater thickness of
cup-plates, especially RR, and has, as they, only x in cup. Ulocrinus,
Miller & Gurley (1890), Upper Carboniferous, N. America, has large
RA, but very small x, which rises partly above level of RR (Fig. XXIX.).
From these genera we pass almost insensibly to : FAMILY 8. ENCRINIDAE.
Dendrocrinoidea with forked, pinnulate, biserial arms, saucer-shaped cup,
with stout plates and well-developed muscle-fossae ; with no anals in cup,
and with sac diminished or absent (Fig. C.). Genera — Steminatocrinus,
Trautschold (1867), Upper Carboniferous Limestone, Russia, has IBB fused
into one. Erisocrinus, Meek and Worthen (1865), Lower to Upper Car-
boniferous, N. America, has 5 small IBB covered by the stem, and x rests
on upper surface of adjacent r. and 1. post. RR. Encrinus, C. F. Schulze

1 82 THE CRINOIDEA

(1760, synn. CMocrinus and [?] Calathocrinus, v. Meyer, Flabellocrinus,
Klipstein, [?] Cassianocrinus, Laube, [?] Traumatocrimis, Wohrmann, which
= Porocrinus, Dittmar non Bill.), Trias, Europe, has 5 minute IBB, a
lofty plated tegmen, small tube, but no distinct anal plate.

Another line of evolution, probably continuing that of the Graphio-
crinidae, introduces the Articulate Grade of Dendrocrinoidea, of which
the most important family and the first to appear is the FAMILY 9.
PENTACRINIDAE. Dendrocrinoidea with pinnulate, uniserial arms, forking
once or dichotomising many times, either regularly, or irregularly so that
the minor branches become pinnulate ramuli ; with a small, usually
depressed patina, in which IBB are often minute or atrophied in adult
(pseudomonocyclic, or, as Bigot has recently expressed it, "cryptodi-
cyclic ") ; but the flexible tegmen extends some way up the arms, so
that the proximal IBr, and sometimes some IIBr, are incorporated loosely
in the dorsal cup ; a slight anal tube or cone, but no distinct anal plates
in either cup or tegmen ; stem pentagonal or sub-pentagonal, usually
with cirri in whorls of 5. Genera — Dadocrinus, v. Meyer (1847 ; see v.
Koenen, 1887-95), Trias, Middle Europe, has a round or sub-pentagonal
stem without cirri, pinnulate arms forking once, 2 IBr, which are united
by several small iBr. Holocrinus, W. & Sp. (1886, em. Jaekel, 1893),
Trias, Germany, has whorls of cirri, slender arms forking once, 3-4 IBr,
not united by iBr, but tegmen stretches up to about IBr, cup high and
constricted above. The family characters become more definite in later
forms, which may be associated as a SUB-FAMILY — PENTACRININAE ; IBB
always minute or atrophied ; stem bears cirri in whorls attached to the
epizygal of a syzygial pair ; its internodes transversed by 5 ligament-
bundles, which are interradially disposed and give rise to a more or less
petaloid figure on the joint-faces \ root-attachment may exist in young,
but is relinquished in adult (Fig. CI.). Genera — Pentacrinus, Blumenbach
(1804 ; syiin. Polycenis, Fischer, 1811, pars; Extracriniis, Austin, 1847 ;
[?] Chladocrinus, L. Agassiz; see especially Quenstedt, 1875), Lias and Jura,
Europe. Petaloid sectors of stem linear with delicately crenulate edges ;
cirri elliptical or compressed in section, in close-set whorls ; IBB present in
adult ; RR prolonged downwards over proximal columnals, the prolonga-
tions being jointed ; each arm has at least 4 rami with large pinnulate
ramules on one side only. Isocrinus, v. Meyer (1837 ; synn. Isis, Lin-
naeus, pars ; Cainocrinus, Forbes ; Picteticrinus, de Loriol ; Cenocnnus and
Neocrinus, Wy v. Thomson ; and Pentacrinus, sensu P. H. Carpenter J),
Trias to Recent, now chiefly Caribbean and Pacific. Sectors of stem dis-
tinctly petaloid, with coarsely crenulate edges ; cirri circular in section,
the whorls further apart (Figs. XVI. 4, 5 ; XVII. 2, 3) ; IBB are visible
in various Jurassic species, but become obsolete in later times (cf. Fig. XI.) ;
BB may form a complete circlet (on which feature Cainocrinus was based),
or may be minute and separated by RR ; IBr 2, non-pinnulate ; arms
isotomous or nearly so ; sacculi occur sparingly. See Guettard (1761),
J. Miiller (1843), P. H. Carpenter (1884), and Fig. XXIII. 3, 5, on p. 117.
BalanocriniLs, Agass. in Desor (1845 ; em. de I^or. 1879), Trias to Eocene,

1 Carpenter chose to ignore all writers on Crinoidea before J. S. Miller. For
the history of these names, see Natural Science, April 1898.

THE CRINOIDEA

183

has columnals of circular or basaltifonn section, with crenellations round the
edge only, not along the sides of the sectors. Austinocrinus, de Loriol (1889),
Cretaceous ; columnals have a joint-surface like that of Isocrinus, but with
finer striae radiating from the petals to the circular periphery. Metacrinus,
P. H. Carp. (1884), W. Pacific, differs from Isocrinus in having 5-8 IBr,
of which IBr2 and 3 always, and IBr5 an(i 6 usually, are united by syzygy,
while each, except IBrx and the hypozygals, bears a pinnule ; BB form a
complete circlet. The members of this sub-family live in colonies, but
can move about and anchor by the cirri at the distal end of the stem.

10

FIG. CI.

Pentacrininae. 1, 1'entncrinus fossil is, portion of stem, patina, and portion of arm, showing
rami, ramuli, and pinnules (p). 2, the same ; portion of a cirrus and articular facet of a cirral.
3, the corresponding parts of Isocrinus asteria. 4, Mctacrinus Moseleyi, cup and proximal por-
tion of an arm. 5, ImrfMMJKiuitdKt. cup seen from below, with portion of stem, bearing cirri,
still attached to it, and with proximal brachials. 6, a radius of the same, showing isotomy of arm.
7, Isocrinus amblyscalaris, joint-surface of an internodal columnal. 8, the same of Jtalanocrinus
mtbteres. 9, the same of Pentacrinus fossiii*. 10, the same of Isocrinus asteri.a. (From Bather,
after P. H. Carpenter, de Loriol, von Meyer, and original.) 2, 3, 7, 8, 0, and 10 are slightly
enlarged.

The stem is least specialised in Balanocrinus, most in Pentacrinus, in
which it attained a length of 18 feet (Quenstedt thought 70). FAMILY
10. UINTACRINIDAE. Dendrocrinoidea in which the arms fork once on
IBr.,, are long and pinnulate, with numerous syzygies, and are incor-
porated in their proximal regions, together with proximal pinnules, in
the dorsal cup, by means of iBr, illBr, and interpinnulars ; there is
no stem, but a centrale. Genus — Uintacrinus, Grinnell (1876), Upper
Cretaceous, N. America, Germany, and England (Fig. GUI.), has a rela-
tively large flexible theca and long arms. It was free-swimming and
possibly pelagic. IBB usually obsolete. For detailed account, see Bather
(1896). FAMILY 11. MARSUPITIDAE. Dendrocrinoidea (?) with pinnulate

1 84

THE CRINOIDEA

Fu;. CI1.

7socritt?«s nsteria. (From A.
Agassiz, after P. H. Cn rpeiiter.)
Rather less than half nat. sixo.

THE CRINOIDEA

185

arms, short in proportion to cup, borne on a sharply denned radial facet,
and forking on IBr2 (further branching unknown) ; they are loosely

Fio. CHI.

Uintacrimis stx-ialis, Upper Cretaceous of America, f; nat. size. 1, from the side. 2, from
below, c, centrale; Jl, basal s ; 11, radials; 77*/-j, first primibrach ; A.r, primaxil ; 1, 2, 3, etc.,
seciuulibrachs, bearing p, pinnules, some of which are included in the walls of the cup, viz.
f.p. ; .s, sy/ygy. The intercalated plates, which bind these elements together, are shaded.

united by iBr, but do not merge in the dorsal cup. Cup large, com-
posed of 5 RR, 5 BB, 5 IBB, and a large centrale, with no trace of a
stem. Genus — Marsupites, J. S. Miller ex Mantell MS. (1821 ; synn.

sc-

--- c

Fir.. CIV.

MurtiipitKs testudinarins. 1, the cup from the side, showing the character of the ornament ;
om Brit. Mus. specimen, E201S), x jj. 2, the radials and proximal region of the arms (from
it. Mus. 5482), nat. sixe. C, centrale; /, fulcral ridge of radial facet; p, pinnules; s,
r.ygies. Other letters as usual.

Sitidaria, Cumberland ; Marsupiocrinus, de Blainville), Upper Cretaceous,
England and Germany (Fig. CIV. ; see also Fig. XVII. 4). The relations
of the genus are not yet clear. FAMILY 12. BATHYCRINIDAE. Dendro-

1 86

THE CRINOIDEA

crinoidea (?) in which the arms fork once on IBr0 ; IBi-j and 2 and all
IIBr, except IlBr3, 6, and 9, are united in pairs by'trifascial articulation
(apud Carpenter, see Fig. XXIII. 2), which may become syzygial (apud
Danielssen) ; only the distal brachial of each pair bears a pinnule, and
there are no pinnules on the first few pairs ; the arms are loosely incor-
porated in the cup to half-way up IIBr:J ; lax has large muscle-fossae on
strong wing-like processes (cf. Fig. XVIII. 6). BB fused in adult to a
single discoidal ossicle ; RR also become closely united ; IBB obsolete.

Interambulacral areas of tegmen contain
scattered small plates, and sometimes each
has a large plate, which may be an oral.
Columnals dicebox- shaped and twisted,
with bifascial articular surfaces ; each
said by Danielssen to be formed by
fusion of two columnals ; those in the
younger, proximal region are thin and
discoidal. The root branches. Genus —
Bathyci-inus, Wyv. Thomson (1872; see
P. H. Carpenter, 1884, and Danielssen,
1892; syn. Ilycrinus, Danielssen & Koren,
1877), North Atlantic and Southern
Ocean, at 750-1500 fathoms (Fig. CV.).
Carpenter places this in the Bourgueti-
crinidae on the grounds of its resem-
blance to Rhizocrinus, while admitting

FIG. cv.

Iktthycrinus. 1, crown of B. Aldri-
chiunus, nat. size (after P. H. Car-
penter). 2, 3, and 4, B. Carpenteri
(after Danielssen). 2, new radials and
anus on an old base and stem, x 4
diani, 3 and 4, stein fragments, nat.
size, pn, pinnules ; rt, root • cirri ;
.S'*], proximal region of stem; 67.,,
median ditto ; 8(3, distal ditto.

FJO. CVI.

Transverse section of the dot-Hal nervous
system in Bathycrinus Carpenttri, diagram-
matised from Danielssen. IR, interradial cords
which pass up between radial* ; JR, radial cords,
connected by re, the ring-commissure.

that "the differences between the two genera are much greater than
their resemblances." If the absence of a proximale have the value
claimed for it by Wachsmuth & Springer, the two genera must go into
different orders. The interradial axial nerve-cords correlated with the BB
fork first within the sutures between RR, and, in Carpenter's opinion, the
basi-radial strands take the place of a ring-commissure ; but Danielssen
describes a ring-commissure (see Fig. CVI.). There 'are 3 water-pores
in each IR. The crown separates very easily from the BB and stem,
and may be replaced by a fresh crown (Fig. CV. 2). Sacculi occur. (See
also Fig. LIII.)

THE CRINOIDEA 187

ORDER 2. Flexibilia, Zittel
( = ARTICULATA, W. & Sp. non Miiller).

Dicyclica in which proximal Br are incorporated in the dorsal cup,
either by their own sides, or by iBr, or by a finely plated skin, but
never rigidly ; plates may occur between RR. Tegmen flexible, with
distinct Amb and numerous small iAmb ; mouth and food-grooves remain
supra-tegminal and open. The top columnal is a persistent proximale,
often fusing with IBB, which are frequently atrophied in the adult.
Arms non-pinnulate (Grade Impinnata), or pinmilate (Grade Pinnata),
but always uniserial.

As in the case of the Distincta and Articulata among the Dicyclica
Inadunata, so the line between the grades Pinnata and Impinnata corre-
sponds roughly with that between Palaeozoic and Mesozoic time. But
in the present order the grades are more self-contained and the gap
between them greater. In fact, we are by no means certain that they
are rightly described as mere grades ; in other words, that the Pinnata
are the lineal descendants of Impinnata. The two divisions may have
arisen from Inadunata independently, springing from pinnulate and non-
pinnulate forms respectively.

GRADE 1. Impinnata.

Flexibilia, in which all plates of the crown are united by loose
suture or muscular articulation. IBB 3, the primitive r. post, remaining
as the small unfused IB. Br usually united by waving sutures, the lower
edge of each frequently with a projection that fits into a depression on
the plate below, and often becomes a separate patelloid plate. Arms
isotomous, or rami may bear ramules on one or both sides, but no
pinnules. Ventral groove wide and shallow ; axial canal separated from
it in proximal region. 5 O, between which food -grooves pass to the
mouth. Stem round ; proximal columnals very short, and usually wider
than the others.

Many of the earlier genera can be distinguished from Dicyclica
Inadunata only by the greater thickness and more elaborate sutural
union of their plates, and the greater width and less length of the arms.
It is the combination of massiveness with flexibility that characterises
the Grade. There is never an elaborate anal sac. Within the Grade
Impinnata can be traced the evolution of heterotomous arm-branching of
two types, also an increase in number of iBr. The genera seem to
merge into one another, and are as yet too ill-defined to be grouped into
families on a sure genetic basis. The following arrangement represents
similarities of structure rather than lines of descent.

FAMILY 1. ICHTHYOCRINIDAE. Impinnata with no iBr, with isotomous
arms, closely abutting by their sides. Genera — Pycnosaccus, Angelin (1878;
syn. Oncocrinus, Bather), Silurian, Gotland, England, and N. America, has a
cup like that of Barycrinus, with x and RA, and with strong axial folds
(Fig. CVIL); arms, though abutting above, are distinct below and do

1 88

THE CRINOIDEA

FIG. CVII.

Analysis of cup in P.i/cnosoeciis.
(After Bather.)

not interlock ; columnals solid, moniliform, alternating in thickness.

Lecanocrinus, Hall (1852), Silurian and Devonian, N. America and Europe,

differs only in greater smoothness of
cup and approximation of arms,
which may even interlock by the
alternating edges of the Br. Cyrtido-
crinus, Angelin (1878), only known
from cup, which appears to be like
that of Lecanocrinut, but is said to
have 4 BB. Clidochirus, Ang. (1878),

Silurian, Gotland ; arms abut, but do not interlock ; no RA, but 3 anals

in vertical series rest on truncate post. B. Mespilocrinus, de Koninck

(1853), Lower Carboniferous, Belgium, England, and N. America, differs

from last-mentioned in having each arm curved

over to the right, so that they all fold with a

sinistral twist (as seen from above) ; IBrx wedge-
shaped, broader on left. Nipterocrinus, Wachs-

muth in Meek & Worthen (1868), Carboniferous,

N. America, should perhaps come here. Ichthyo-

crinus, Conrad (1842), Silurian, N. America and

Europe, has no anals ; arms abut all round and

interlock by edges of Br (Fig. CVIIL). FAMILY 2.

GAZACKINIDAE. Impinnata (?) with a single large

iBr in each interradius, the posterior (anal) resting

on truncate edge of large post. B ; arms isotomous ;

tegmen of 5 O of Cyathocrinoid type, surrounding

a peristorne which is covered by fused proximal

Amb, and supporting Amb along their edges ;

food -grooves fork on the tegmen (Fig. CIX.).

Genus — Gazacrinus, S. A. Miller (1892 ; syn.

FIG. CVIII.

Iclitliyocrinus pirlformis,

Idiocrinus. W. & Sp.), Silurian, N. America, is slightly restored from Brit.

_, ., ~ . . , Mus. 40214. (Nat. size.)

referred to the Camerate Family Dimerocrmulae

by W. & Sp. (1897). Its dorsal cup does not differ essentially from that of

Anisocnnus (infra), but the number of IBB is uncertain. The tegmen,

HBrJ

Gazacrinus. (Diagrammatised from Wachsmuth & Springer), x 2 diam. 1, G. inornatns,
tegmen with almost all ambulacrals removed. 2, G. ventricosus, dorsal view. 3, same, side
view, with proximal ambulacrals (P) in position. As, passage for rectum through posterior
oral ; v.g, ventral grooves passing between edges of orals. Other letters as usual.

scarcely of either Flexible or Cainerate type, suggests recent descent from
'nadunata. FAMILY 3. TAXOCRINIDAE. Impinnata with iBr, which usually

THE CRINOIDEA

189

are few and have the proximal larger than the rest ; with isotomous
arms which may abut but do not interlock ; anals form a well-defined
vertical series. Genera — Gnorimocrinus, W. & Sp. (1879, em. Bather,
1899), Silurian, Gotland ; arms distinct, with 0-4 small iBr, and
sometimes a few illBr ; post. B reaches up to top of patina, between
it and r. post. R is a RA, supporting the greater part of cc, which bears
a vertical series of 1 or 2 rows. Anisocrinus Angelin (1878), Silurian,
Gotland, has abutting but not interlocking arms, a very large proximal
iBr, with small triangular piece above ; x differs from iBr only in
resting on truncate edge of large post. B ; no RA. Taxocrinus, Phillips
in Morris (1843; synn. Isocrinus, Phill. non v. Meyer; Cladocriniis,
Austin non Agass. ; Euryalecrinus, Aust. ; Forbesiocrinus, de Kon. non W.
& Sp.), Silurian to Carboniferous, Europe and N. America, has arms
more distinct, with few or no iBr, and occasional small illBr and illlBr ;
anals form a vertical series resting on truncated post. B, and seem to

Fia. CX.

Diagrams of the arm-branching in Taxocrinidae and Dactylocrinidae. 1 — 2 — 3

1, Taxocrinus tuberculutus ; 2, Litkocrinus ; 3, Ccdpiocrinus ; 4, Dactylo- /\

criniis; 5, Synerwrinus ; 6, Onychocrinus exsculptus. These do not form / 4

a continuous evolutionary series, but their relationship may be indicated / |

thus— 6' 5

I

have supported a small free tube (Figs. CX. 1, and XXXVII.). Homalo-
crinus, Ang. (1878), Silurian, Gotland, has very small BB. abutting armsr
large proximal iBr, followed by 1 or 2 in vertical series, occasionally 2
illBr, anals as iBr but resting on post. B ; except for the isotomy of its
arms, this genus closely resembles Ccdpiocrinus. FAMILY 4. DACTYLO-
CRINIDAE. Impinnata with iBr either few or numerous ; with hetero-
tomous arms, the rami bearing ramules ; with anals in vertical series.
Genera — Calpiocrinus, Ang. (1878), Silurian, Gotland and England, has
minute often obsolete IBB, but fairly large BB ; iBr few and variable,
illBr occasionally present ; anals 3-5, x resting on the small post. B ; the
IIBr rami bear unbranched ramules on their inner sides, the proximal
ramule much larger than the rest (Fig. CX. 3). Lithocrinus, W. & Sp.
(1879, emend. Bather, 1899 ; syn. Forbcsiocrinus, Ang. non. de Kon.),
Silurian, Gotland, differs from Calpiocrinus in larger size of BB, greater
number of iBr, branching of ramules ; the latter characters make the
arms less apposed to one another (Fig. CX. 2). Dactylocrinus, Quenstedt

190

THE CRINO1DEA

(1876, based on Dimerocrinus oligoptilus, Pacht ; probably includes Aristo-
crinus or Callawaycrinus, Rowley, 1895), Silurian (?) and Devonian,
N.-W. Europe and N. America ; iBr few, proximal large ; x rests on large
truncate post. B, a little to the right, and supports numerous smaller
plates in somewhat irregular vertical series ; the IIIBr rami bear ramules
on the sides towards the middle of the dichotom, the proximal raruule

branches again (Fig. CX. 4).
Synerocrinus, Jaekel (1897),
Carboniferous, Europe, has arms
like Dadylocrinus, except that no
ramules branch ; it also differs in
having 3-8 iBr, perhaps more, with
occasional illBr and illlBr ; y rests
on post. B (Figs. CX. 5, and CXI.).
This genus probably includes the
Belgian species erroneously referred
by de Koninck to Taxocrinus nobilis
when erecting Forbesiocrinus. Euryo-
crinus, Phillips (1836), Carboni-
ferous, England, is probably a close
ally, but its anus are not well
known. Onychocrinus, Lyon &
Casseday (1859), Carboniferous, N.
America and Ireland, has a large
proximal iBr followed by small
plates, often numerous, merging
with the tegmen and ventral cover-
ing of the arms, but leaving the
arms more free than usual in the
sub-order ; I IBr rami bear branch-
ing ramules, either along each side
or in clusters at end, but the
heterotomy is always bilateral, not
MUS. unilateral as in all the genera just
mentioned — a fact suggesting that
Onychocrinus should form a distinct

basal is .just visible on the left; dl, fossa for VfTT

dorsal ligament ; mf, muscle-fossa ; pt' , surface (see also Fig. XVII. 4).

FIG. CXI.

Synerocrinus incurvus, from Brit.
E6707. 1, from the right anterior side, x J.
2, joint-surface of a brachial, x 3 diam. 3, a
radial, showing articular surface for ttrst primi-

brach, and on either side the surfaces of loose r .. , , . ..

suture with interbrachials, x i!. B, basais, family or, at least, Bub-family (Fig.
which pass umler the cup, the larger posterior CX g) . analg 3.5 resting on post g

FAMILY 5.

SAGENOCRINIDAE. Impinnata with
over 20 iBr, 6 or more illBr, and

variable number of illlBr; arms isotomous or almost so, to VIBr
or beyond ; anals not a distinct series, but represented by greater width
or number of iBr in post. IR. Genera — Sagenocrinus, Austin (1843),
Silurian, Europe and N. America, has RA sunk between post, and
r. post BB, so that proximal iBr is supported between post. B and RA
(Fig. XXIVa). In " Forbesiocrinus Agassizi" Carboniferous, N. America,
which may be placed in this family, post. B supports 2 iBr, and there
are considerably more iBr in this IR Patelloid plates are richly

THE CRI NO IDE A 191

developed, but are absent from Sagenocrinus. Otherwise the two genera
agree closely.

The following genera are placed provisionally in the Impinnata :

Edriocrinus, Hall (1859), Devonian, N. America (Fig. CXII.), when young
is attached by BB, but is
free - floating in adult ; BB
become fused into a bowl-
shaped mass, supporting 5
RR and x ; arms broad, with
low Br, isotomous. Cleio-
crinus, E. Billings (1856 ; see
W. & Sp., 1886), Ordovician,

Canada ; IBB and BB hidden V Flo CXII

by stem ; RR small and sepa- I M Edriocritlus pirif(mnis, nat.

rated by a large pentagonal size. (After Hall.) i, from

. . ,. , . , \---.Mi side> showing concavity for

interradial ; arms isotomous to V- AY attachment at end of fused

about VIIBr, and all appear V^-B basais (B). 2, the cup from

. above, showing anal (x) and

to interlock and to be joined facets of radiais.

by close suture ; post. IR sup-
ports a vertical series of anals, which reach the full length of the arms.
RhopalocrinuS) W. & Sp. (1879), is based on " Taxocrinus gracilis"
Schultze, Devonian, Eifel ; it perhaps belongs to Dicyclica Inadunata.

GRADE 2. Pinnata.

Flexibilia with BB and RR united by close suture, RR and proximal
Br by muscular articulation or syzygy ; pseudomonocyclic ; arms pinnu-
late and either simple or isotomous ; axial canal separate from ventral
groove throughout ; lax is generally IBr2, rarely IB^ ; 5 0 present in
early stages and sometimes in adult, but usually atrophy ; anals do not
form part of the dorsal cup in the adult. Stem round, pentagonal, or
elliptical in section, proximal columnals often forming a widened cone.

This group is confined to Mesozoic and later times, and there is no
evidence that it is descended from the Palaeozoic Impinnata ; it may be
an offshoot from Triassic Dicyclica Inadunata, from which it is dis-
tinguished by the mode of growth of the proximal columnals.

FAMILY 1. APIOCRINIDAE. Pinnata in which the patina consists of 5
BB and 5 RR ; arms incorporated to a very variable extent in the cup,
and iBr may be present ; columnals round or pentagonal in section, their
joint-surfaces marked with radiating striae, and sometimes tubercles in the
middle ; no cirri ; root, when present, encrusting. (For fossil forms, see
de Loriol, 1883.) Genera — Millericrinus, d'Orb. (1840 ; synn. Ceriocrinus
and Pomatocrinus, Desor ex Konig), Trias (?) to Lower Cretaceous, Europe ;
IBrx united to R by muscular articulation, and to lax by close suture ; arms
isotomous, free from IBr, or proximal Br united by tegmen, or a few small
iBr developed ; except for the proximale, the upper columnals are rarely
widened (Fig. XVII. 7). In some species (Fig. LII.) the crown breaks off
from the root, the stem is gradually resorbed, and a free-floating stage attained.
Apiocrinus, Miller (1821), Jurassic, Europe ; IBrx united to R probably
by ligament, not by muscular articulation, and to lax by incomplete

192

THE CRINO1DEA

syzygy ; IIBrj united to lax by articulation ; arms isotomous, incorporated
in cup at least up to IIBr., ; iBr few and irregular ; the upper columnals
widen gradually, and, with the proximale, form a cone passing into the
cup (Fig. XVII. 5, 6). Guettardicrinus, d'Orbigny (1840), Upper Jurassic,
differs from Apiocrinus only in the union of IIBr1 to lax by close suture,
the incorporation of a greater number of IIBr in the cup, and the presence

2 1 3

Fio. CXIII.

Calamocrinus Diomedcif. 1, the crown and proximal portion of the stem from the right
posterior intcrradius, x jj. 2, the root, nat. size. 3, the interior of the basal circlet, from
above, showing the anchylosed sutures (S) and the axial cords (cur) radiating from the central
chambered organ, x f. (After Agassiz.)

of more iBr ; it is the acme of this line of development. Acrochordocrinus,
Trautschold (1859; synn. Cyclocrinus, d'Orb. non Eichw. ; Mespilocrinus,
Quenst. non de Kon.), Jurassic and Lower Cretaceous; columnals only.
Calamocrinus, A. Agassiz (1890,-92), 392-782 fathoms, Galapagos Is. and B.
of Panama (Fig. CXIII.). Patina distinct, owing to restriction of facet to
$ width of R ; BB tend to be fused ; RR laterally united by ligament ;
r. and 1. post. RR slightly longer than the others. IBfj united to R by

THE CRINOIDEA

193

muscular articulation, and to IBr2 by incomplete syzygy. Arms hetero-

tomous ; each gives off 5 unbranching rami nearly as stout as the main

stem, 2 to right and 3

to left, or vice versa; the

first main -axil is the

10th or llth brachial,

succeeding branches are

at much greater inter-

vals ; pinnules occur

below lAx, on IBr4> 6 g

or IBr4> 6> 7 9 those

being as a rule epizy-

gals. There are dis-

tinct Amb and adAmb,

both in arms and teg-

men. The proximal

region of the arms is

fixed by the tegmen,

so that the grooves of

the proximal pinnules

rise from the tegminal

food - groove (Fig.

XXIV.). Small iBr

, -i -11 c -,->-,-> legmen or uaiamocrinus Dlomedac, witli the arms and pin-

rest on snouiaers OI KK nuies cut off down to the level at which they become fixed in

'

/Br_.

P2

P3

FIG. CXIV.
Tegmeu of Ccdamocrinus Diomedat, with the arms and

ambulacrals or covering-plates'; ^4s, anus

• A i j • nx, axial canal of arm ; cc, coeliac canal of arm ; fy, food-groove

lorate lAmb, and these of anterior arm where it joins the tegminal food-groove ; IBr,

and merge into imper-
"

primibrach ; iAmb, iuterambulacrals, both perforate and im-
perforate ; pi, first or proximal pinnule ; p2, second ; p3, third
iAmb : 5 imperforate. pinnule ; stc, subtentacular canal of anterior arm ; w.p, water-

plates at the interraclial pore-- <After Asassiz>- x

angles of the peristorne are taken by Agassiz for 0 (Fig. CXIV.).
Stem long, smooth ; root encrusting (Fig. CXIII. 2). FAMILY 2.
BOURGUETICRINIDAE. Pinnata in which the patina consists of 5 BB

a.c

FIG. CXV.

Bourgueticrinns. 1, B. aeqiuilis (from Brit. Mus. E6705). 2, ventral view of same (from
E6706). 3, joint-surface of proximale (P), by which it articulates with adjoining columnal (C).
4, vertical median section of B. ellipticus (after d'Orbigny). All x 4 diam.

13

194

THE CRINOIDEA

forming a closed ring and 5 RR
with high muscle-plates ; arms in-
corporated to a very slight extent
in the cup ; and then only loosely
without development of iBr ; all
columnals, or all except those in
the proximal region, have ellipti-
cal joint-surfaces with a grooved
and toothed fulcral ridge in the
long diameter and ligament-fossae
on either side of it ; each colum-
nal is twisted so that the ridge at one end
lies at an angle to that at the other end ; cirri
may be present at the root end or in the
middle of the stem. Sacculi occur in recent
forms. Genera — Bouryueticrinus, d'Orbigny
(1840), Cretaceous, Europe and Alabama (Fig.
CXV.) ; sides of cup vertical or sloping in-
wards above ; BB about half height of RR ;
radial facet small, horizontal, with small
dorsal ligament - fossa ; IBr^ and 2 laterally
connected with adjoining rays, arms un-
known ; proximale enlarged, circular ; colum-
nals of proximal region circular, widening
upwards ; those of middle and distal regions
twisted, elliptical ; fulcral ridge continuous
around axial canal, ligament-fossa broad and
shallow ; cirri rare except at root. Meso-
crinus, P. H. Carpenter (1881), Cretaceous,
Sweden, Germany ; sides of cup slope out-
wards ; BB short ; radial facet large, slopes
upwards to centre, with larger ligament-fossa ;
arms unknown; proximale small and circular;
proximal columnals circular, narrowing up-
wards ; the rest as in Boitrgucticrinus, but liga-
ment-fossa deeper ; cirri sometimes numerous.
Rhizocrinus, M. Sars (1864,-68 ; synn. Cono-
cnntw, d'Orb. ; Democrinw, Perr.), Eocene
to Recent ; Atlantic, 73 to (?) 1900 fathoms :
sides of cup slope slightly outwards ; BB
high, may be 7 times height of RR, often
fused ; radial facet slopes upwards to centre,
with small ligament - fossa ; IBr^ ^d 2 free
laterally, arms not forking; proximale thin
and discoidal ; proximal columnals discoidal,
those of middle region slender, fulcral ridge
broken at axial canal, around which the
ligament-fossa is concentrated (Fig. XLIX.
7) ; no cirri except at root (Fig. CXVI. ; see

Rliizocrinvs lofoten-
sis, North Sea, about
twice nat. size. (From
Narrative of the Crui»e
of Il.Af.S. "Cltallen-

THE CRINOIDEA 195

also X., XXIII. 4). FAMILY 3. ANTEDONIDAE. Pinnata in which the
patina consists of 5 small BB, not forming a closed circlet, and RE with
large high muscle-plates and facet approaching horizontal ; cavity enclosed
by RR is minute ; mouth endocyclic ; proximal Br loosely incorporated in
cup ; columnals as in Bourgueticrinidae, but usually lost in the adult, ex-
cept the proximale and adjoining columnals which fuse with one another
and with IBB to form a single ossicle, the " centrodorsal," which bears
cirri ; root when present, encrusting. 0 and x present in brephic stage,
but not in adult. Sacculi almost always present. (See structural details of
Antedon, Figs. IX., XV., XVIII. 2, 3, 4, XIX., XXX., XXXIIL, XLVL,
XLVIL, XLIX. 8, 9, LIV, LV.) Genera— Thiolliericrimis, Etallon (1859),
Jurassic and Cretaceous of Switzerland, France, and Portugal (Fig. CX VI I.) ;
scarcely differs from Mesocrinus, except in the reduction of BB and
presence of a cirriferous centrodorsal at the top of the fairly stout stem ;
our knowledge of this most important form is due to de LorioL The
remaining Anted onidae retain the portion of stem below the centrodorsal
only in the brephic stage, while their BB further diminish during geo-
logical periods, their adcentral portions fusing into a small 10-rayed plate,

FIG. CXVII.

Thiolliericrinus. 1, T.fletuosns,
cup seen from below, no basals
visible (from Brit. Mus. 4922-2r<).
2, T. Ilibeiroi, from the side,
showing basals and facets for
cirri (reconstructed from de
Loriol's figures). CD, centro-
dorsal, still bearing facet (Sf)
for attachment to stem, x 2
diam.

the " rosette," which lies above the chambered organ, and in some species
of Antedon is all that remains of BB. There is also traceable in the arms
a gradual attenuation and, in many cases, increase of forking, with a
partial or entire loss of calcified covering-plates. Antedon, de Fremin-
ville (1811 ; synn. Alecto, Leach ; Comatula, Lamarck, pars ; Hibernula,
Fleming ; Phytocrinus, de Blainville ; Solanocrinus, Goldfuss ; Hyponome,
Loven ; Geocoma, Fraas non d'Orb. ; et alia), Lias to Recent, almost all seas,
littoral to 2900 fathoms. Arms fork once or more ; Amb usually present,
especially on pinnules. The genus is divisible into 9 groups, differing in
arm-structure and distribution. Eudiocrinus, P. H. Carp. (1882 ; syn.
Ophiocrinus, Semper non Salter), Neocomian (?) to Recent, Pacific and B. of
Biscay, 50 to 900 fathoms ; differs from Antedon only in non-forking of
arms. Promachocrinus, P. H. Carp. (1879), Pacific and South Sea, 70 to
1800 fathoms; 10 RR, probably a persistent meristic variation from
more than one species of Antedon. FAMILY 4. ATELECRINIDAE. Pin-
nata with patina of 5 BB forming closed circlet and no rosette, 5 RR
with high muscle-plates; arms fork once, IIBr long, with no pinnules
on first 8 or 16 ; no stem, but acorn -shaped centrodorsal, with cirri
alternating in 5 vertical double rows ; sacculi present. Genus —
Atelecrinus, P. H. Carp. (1881), Cretaceous (?) to Recent, tropical Atlantic

196

THE CRINOIDEA

and Pacific (Fig. CXVIIL). FAMILY 5. ACTINOMETRIDAE. Pinnata with
dorsal cup and centrodorsal on the Antedonid plan, but differing in the
following points: — Mouth exocyclic and gut much coiled, with con-
sequent larger cavity between RR, small muscle-plates, and facet approach-
ing the vertical ; further asymmetry shown in unequal and variable
tegminal food -grooves, and in occasional ungrooved structure of some
posterior rami, which may be shorter than the rest and without podia ;
no sacculi ; no calcified Amb ; proximal pinnules have a terminal serrated
margin — " comb " ; centrodorsal discoid, with cirri few, almost limited
to its margin, and sometimes atrophied. Genus — Actinometra, Miiller
(1841, em. P. H. Carp., 1887 ; synn. [?] Comaster, L. Agassiz ; Comatula,

Fio. CXVIIL

Atelecriiius bidanoides, with two
cirri partly preserved and arms
imperfect. (From A. Agassiz, after
P. H. Carpenter.) x 2 diam.

FIG. CXIX

Thaumatocrinus rcnomltis, from the anal
side, eta, anal appendage ; u/t, interamlmla-
crals ; at, anal tube ; b, basal ; b%, second
brachial ; cd, centrodorsal ; t, interradial ; r,
radial. (After P. H. Carpenter.) x ¥•

purs; Plumoyenia, Love"n), Lower Jurassic to Recent, almost all seas,
littoral to 800 fathoms. Divisible into 8 groups, differing in arm-
structure and distribution. FAMILY 6. THAUMATOCRINIDAE. Pinnata
with cup of 5 BB forming a closed ring, and 5 RR separated by 5 inter-
radials resting on BB ; Br not incorporated in cup ; arms do not fork ;
5 0, separated from cup-plates by relatively large iAmb ; no Amb ; no
eacculi ; anal tube in post. IR ; post, interradial followed by 5 plates
in vertical series forming a free appendage (Fig. CXIX.). Genus —
Thaumatocrinus, P. H. Carp. (1883), a unique individual, probably young,
South Sea, 1800 fathoms. Differs greatly from all other Pinnata; the
structure of the cup is as in Xenocrinut (p. 165), the anal appendage is
paralleled in some Taxocrini (cf. Fig. XXXVIL).

THE CRINOIDEA

197

FAMILY 7. EDGENIACRINIDAE. Pinnata with patina of 5 ER only,
BB having been overgrown by RR and absorbed by a continuance of such
a process as produced the rosette of Antedon. RR united by close suture,
often fused. IBr2 axillary, united to IBrx by syzygy or fusion. Rami
10, robust, incurving. Stem short; colunmals cylindrical, high, with
joint-surface granulate, or marginally striate ; no cirri ; root encrusting,
lobed. All European (see de Loriol,
1883, and Jaekel, 1891). Genera —
Engeniacrinus, J. S. Miller (1821 ; syn.
Symphytocrinus, Konig ; Caryophyllites,
Auctt. pre-Linn.), Bathonian to Lower
Cretaceous (Fig. CXX.). Patina cylin-
drical or clove - shaped, with shallow
depression for viscera ; radial facets
separated by processes ; lax rising above
origin of IIBr into a 3 -sided spine,
which perhaps helped to support the
tegmen ; IIBr small. Torynocrinus,
Seeley (1866 ; synn. Cyrtocrinus, Jaek. ;
? Hemicrinus, d'Orb.), Upper Jurassic to
Lower Cretaceous ; patina and proximale
fused, the ventral surface bent to one side
and bearing stout arms. Gammarocrinus,
Quenst. (1858 ; syn. Sclerocrinus, Jaek.),
Upper Jurassic ; patina massive, concave
below. .Gymnocrinus, de Lor. (1879, em.
Jaek., 1891); Upper Jurassic; lax re-
markably developed. PhyHocrimis, d'Orb.
(1849), Bajocian to Neocomian ; RR have
small facets and long spines. Tormocrinus,
Jaek. (1891), Eocene, and Trigonocrinus,
Bather (1889), Oxfordian, have very
small radial facets, rounded spines, and
a deep tubular cup-cavity ; the latter
differs in the loss and atrophy of certain
rays. Dolichocrinus, de Loriol (1891 ;
syn. Tetanocrinus, Jaek.), Upper Jurassic ;
RR form a tube 10. mm. long, 2-25 mm.
wide, with interradial re-entrant angles
at its base; the radial facets are of
Bourgueticrine or Antedonid type, and if
BB were present, as de Loriol supposes, the genus must be removed from
this family. FAMILY 8. HOLOPODIDAE. Pinnata with patina of 5 RR,
usually fused, and enclosing a relatively wide cavity. BB presumably as in
Eugeniacrinidae. 0 large, surrounded by a few iAmb. lax fused to IBij
in adult, and supports stout, incurved, unbranching rami ; no stem ;
attachment by base of patina. No sacculi. The arm-structure led Jaekel
(1891) to combine these forms with the Eugeniacrinidae. Genera —
Holopus, d'Orb. (1837), Tertiary of Italy, Recent, Caribbean Sea, shallow

FIG. CXX.

Eugeniacrimts caryophyllattts, partial
reconstruction (x 2 diam.), and in-
terior view of a primaxil (x 3 diam.),
on which are seen the articular i'acets
(IIBr') for the secundibrachs ; the
latter are only partially known. Other
letters as usual. (Based on the obser-
vations of Jaekel.)

1 98 THE CRINOIDEA

water (Figs. CXXI. and XXXIV.). Cyathidium, Steenstrup (1846 ; syn.

Micropocrimis, Michelin), Uppermost Cretaceous to Miocene, Denmark

and Italy. FAMILY 9. EUDESICRINIDAE.
Pinnata (?) with patina of 5 RR, enclosing
funnel-shaped cavity open below, resting
on a solid mass which may represent
fused BB or a proximale. lax articulated
to IBi-j, and supporting 2 stout rami
which abut on adjoining rami. Axial
cords lie close to inner walls of RR.
Genera — Eudesicrinus, de Loriol (1882) ;
and Cotylederma, Quenst. (1852 ; syn.
Cotylecrinus, E. Deslongch.), both Lias.
Jaekel would place these genera near
the Plicatocrinidae ; they are usually
referred to Eugeniacrinidae or Holo-
podidae.

FIG. cxxi. ORDER 3. Dicyclica Camerata

Adult Holopns Rang!, from anterior. / PAHTTPPATA W & ^r» rtnr«\

(From A. Agassiz, after P. H. Carpenter.) I " ^MERATA, W . & bp. pal l>

Enlarged by one-twelfth. . ..««,.* ,

Dicyclica in which all IBr and usually

IIBr are incorporated in the dorsal cup by iBr, at first loosely, but
afterwards by close suture. IBB always the primitive 5. A plate
always between r. and 1. post. RR, resting on post. B, and followed by
others leading up to the anus. Mouth and ambulacra subtegminal. Arms
pinnulate.

FAMILY 1. RETEOCRINIDAE. Dicyclica Camerata with RR and Br sepa-
rated by supplementary plates irregular in size, shape, and arrangement, and
forming depressed interradial areas, the posterior of which is divided by a
single yertical series of prominent plates leading from post.B to the eccentric
anus. Genus — Reteocrinus, Billings (1859 ; see W. & Sp., 1897), Ordo-
vician, N. America (Fig._CXXII.) ; 2-3 IBr ; about 6 IIBr, of which the
pinnules, borne from IIBr2 onwards, are also fixed in the interradial areas ;
the pinnule borne by IIBr2 is in some species represented by a ramus,
partly fixed ; the rami may branch again after becoming free, and are
uniserial or slightly in zigzag ; tegmen a low dome of minute irregular
plates; stem pentagonal. FAMILY 2. DIMEROCRINIDAE. Dicyclica Camerata
with RR in contact except at posterior side, with 2 IBr and a varying
number of IIBr, separated by a large proximal iBr, which rests on
shoulders of RR and supports 2 plates, usually followed by smaller
ones merging into iAmb ; post. IR wider, and its proximal plate (anal)
supports 3 plates followed by others, and leading up to anus, which
has no tube ; illBr usually present. Genera — Ptychocrinus, W. & Sp.
(1885, em. 1897), Ordovician, N. America ; arms fork twice, are slender and
uniserial as in Reteocrinus. Orthocrinus, Jaekel (1895), Devonian, Germany ;
arms fork once, are free from lax, stout and uniserial. Dimerocrinus,
Phillips (1836 ; synn. Glyptaster and Thysanocrinus, Hall em. W. & Sp.,

THE CRINOIDEA

199

1897 ; Eucrinus, Ang.), Silurian, Europe and N. America (Fig. CXXIIL) ;
arms fork once or twice, are stout, biserial, and directed upwards. Cypho-
crinus, S. A. Miller (1892 ; syn. Hyptiocrinus, W. & Sp.), Silurian, Indiana ;

IB

FIG. CXX1I.

Reteocrinvs Onealli, anterior view.
p, pinnules ; j3, fixed pinnules ;
other letters as usual. (After
Wachsmuth & Springer.) x §.

nn

Fio. CXXIII.

DimerocHnvs ilecculfcti/lus, from
Brit. Mus. EOT07 ; seen from pos-
terior interred ius. nn, nodals of
stem, x 2 diain.

As

arms fork at least once, are stout, biserial, and pendent, thus exposing
the tegmen which is spinous. FAMILY 3. LAMPTEROCRINIDAE. Dicyclica
Camerata with a dorsal cup in general structure like that of Dimero-
crinidae, but with asymmetry introduced by the
development of an anal tube and consequent
bulging of IR, and shifting of mouth anteriorly.
All from Silurian, N. America. Genera —
Lampterocrinus, C. F. Roemer (1860, W. & Sp.,
1897) ; IBB large, anchylosed ; anal tube cen-
tral ; arms supposed to be 5 rami bearing alter-
nate ramules, but are not known beyond IIIBrj
(Fig. CXXIV.). Siphmiocrimts, S. A. Miller
(1888, em. W. & Sp., 1897) ; IBB small ; rectum
forms an asymmetric protuberance below, then
curves subtegminally either to a central anal
tube, or right across to an anterior opening at
or even beneath the arm-bases.

FAMILY 4. RHODOCRINIDAE. Dicyclica Cam-
erata with RR separated by a single distinct
plate in each IR, followed by well-defined iBr
regularly arranged (some individuals of Lyrio- c. F. Roemer.) x j.
crinus have RR not quite separated, and

some species of Diabolocrinus have not the single distinct interradial) ;
the anal area is not always distinct, and but rarely has a vertical series of
plates. IBr, 2, in all except the rather doubtful Anthemocrinus. Arms free

Fro. CXXIV.

Isampterocrinus ten nesseensis,

200

THE CRINOIDEA

above IIBr (except in Thylacocrinus) ; illBr may or may not be present.

The plates of the tegmen are small and usually irregular. The family

begins in the Ordovician, probably as an independent development from

Reteocrinidae or similar forms, and runs parallel to the Dimerocrinidae

through the Silurian, but, unlike them, persists to Carboniferous times.

There is seen in it an increase in definiteness of iBr, and the origin of

biserial arms, which are usually isotomous but exceptionally bear ramules.

Genera — Rhaphanocrinus, W. & Sp. (1885 ; syn. Coelocrinus, Salter),

Ordovician, N. America, Gt. Britain, has uniserial arms and numerous iBr

and illBr. Archaeocrimis, W. & Sp.

(1881), Ordovician, N. America, has

Br in zigzag, numerous iBr more

regular in arrangement, with no verti-

cal series in anal IR (Fig. CXXV.).

Diabolocrinus, W. & Sp. (1897), Ordo-

vician, N. America, has the larger

and more regular iBr surrounded by

supplementary plates ; the anus is

at the end of a strong, subcentral tube ;

the arms bear ramules. Lyriocrinus,

Hall (1852 ; syn. Marsupiocrinus, Hall

non de Blainv. nee Phill.), Silurian, N.

America and England, has but 2 rami

IllBr

Fio. CXXVI.

Thylucocrimis yannioti (from Brit.
Mus. E6642). Seen from posterior inter-
radius. Owing to slight crushing all
live iufrabasals (IB) are seen, and the
absence of a stein-facet may be noted.
Supplementary plates are shaded. The
section of an arm with pinnules (pn) is
after Oehlert. x i

IllBr

Kio. CXXV.

Archacocrinus desiderntuf, from the left
posterior radius. (Diagrammatised from
Wachsmuth & Springer.) x A.

to each ray, biserial; iBr, 1 + 2 + 1, not always quite separating RR ;
anal area usually of similar structure. Anthemocrinus, W. & Sp. (1881).
Silurian, Gotland, has only one IBr, and the first large iBr is followed by
only 1 or 2 small ones ; the biserial arms fork 2 or 3 times. Diameno-
crinus, Oehlert (1891, em. Jaekel, 1895), Devonian, France and Germany,
has repeatedly isotomous arms, with Br in zigzag ; 6-8 IIBr, and a long
narrow series of iBr. Thylacocrinus, Oehlert (1878), Devonian, France
and New York (Fig. CXXVL), has arms fixed up to IIIBr and sometimes
IVBr, after which follow long unbranched biserial rami ; supplementary
plates occur between all fixed brachial series ; stem minute or (?) absent.
Lahiiseniocrinus, Tschernyschew (1892,-93), Lower Devonian, Ural,
appears allied to the preceding, but base and free orachials are unknown.

THE CRINOIDEA

201

Rhipidocrinus, Zittel (1879, ex Beyrich, MS.), Devonian, Germany, has
stout uniserial rami giving off biserial ramuli on alternate sides. Acantho-
crinus, C. F. Roemer (1850, em. Jaekel, 1895), Devonian, Germany, does
not differ greatly from the next genus. Rhodocrinus, J. S. Miller (1821,
restricted W. & Sp., 1881), Carboniferous and (?) Devonian, Europe and

As

FIG. CXXVll.

Gilbertsocrinus. 1, G. lyonanus (=G. dispansus, W. & Sp.), ventral view, showing the
tegminal appendages and the proximal portions of the arms emerging from beneath them.
2, G. tuberculosus, the appendages mostly broken off, but more of the arms preserved. 3, G. cal-
caratus, showing how the appendages of the European forms are double where they issue from
the teginen. 4, G. typus, patina from below ; ap, tegminal appendages ; op', their points of
origin, showing central canal ; Ax, primaxil ; Br, pinnulate arms, which issue from the dorsal
cup at Br'. Other letters as usual. (All after Wachsmnth & Springer.) x 3.

N. America ; has a vertical series of anals and isotomous arms, all the
free parts of which or only the finials are biserial. Condylocrinus,
Eichwald (1859, em. Tschernyschew, 1893), Lower Devonian, Ural,
differs, if at all, in iBr series, which runs — 1, 1, 2, 3. Ophiocrinus,
Salter (1856 ; non Charlesworth, nee Semper, nee Angelin), Devonian,
S. Africa, has numerous iBr and several illBr ; IIIBr free and in zigzag.
Gilbertsocrinus, Phillips (1836; synn. Ollacrinus, Cumberland MS.;
Goniasteroidocrinus, Ly. & Cass. ; Trematocrinus, Hall), Devonian and
Carboniferous, Europe and N. America (Fig. CXXVll.), distinguished by
small pendent arms, and large, branching, horizontal, hollow extensions
of the interambulacral areas of the tegmen, of uncertain function.

The following incorrect or indeterminable Names may be added to those
already quoted as synonyms or incertac sedis : —

Adelocrinus, Phillips (1841), Devonian, Britain, is probably syn. of either

Hexacrinus or Arthracantha (see Whidborne, 1899).
Allionia, Michelotti (1861), syn. of Antedon.
Aporocrinus, Austin (1842), is an Agelacrinid, fide label on A. gyratus in

Bristol Museum.

Aspidocrinus, Hall (1859), Silurian, New York ; encrusting root of a crinoid.
Asteria, Auctt. vett., a Pentacrinoid columnal ( = Pentacrinos, Agricola).
Asterias, Linn, pars (1758) ; syn. of Antedon and Actinometra.
Asteriatites, Schloth. (1813) ; A. pennatus and A. rosaceus, syn. of Antedon.
Asterocrinus, Miinster (1838), "aus dem Orthoceratiten-kalk bei Elbersreutb,"

if a Crinoid at all, is an Inadunate.

202 THE CRINOIDEA

Astrios, Troost, nom. nud. (1850) ; not known.
Astrocoma, de Blainville (1830) ; syn. of Antedon and Actinomelra.
Astrocrinus, Cumberland (1829), fide. Waclismuth & Springer, not known.
Astropodia, Llhuyd et Auctt. pre-Linn., for Crinoid arms; but Ure (1797)

applied it also to various cup-plates, and Defrance (1819) to Apiocrinus.
Atocrinus, M'Coy (1844 or 1862), Carboniferous, Ireland ; referred to Platy-

crinus by de Koninck & C. F. Roemer, to Cyathocrinus by Waclismuth

& Springer. The former probably right.

Ealanocrinus, Troost, nom. nud. (1850) ; syn. of Lampterocrinus.
Bohemicocrinus, Waagen & Jahn (1899), Silurian, Bohemia ; cup alone known,

probably Carpocriniis or Desmidocrinus ; RR shaped much as in Barrandeo-

crinus.
Caleidocrinus, Waagen & Jahn (1899), Ordovician, Bohemia ; like Taxocrimts,

but apparently no anal between RK, and no anal ridge.
Calocrinus, Steininger (1849), Devonian, Eifel. Dicyclic Inadunate ; IBB fused ;

no anal ; perhaps a Cuprcssocrinus.
Campanulitcs, G. Troost, nom. nud. (1850).
Ccntrocrinus, Worthen (1890), non Austin & Meek & Worthen, nee W. & Sp. ;

?syn. of Gazacrinus.

Codonocrinus,Troost, nom. nud. (1850) ; syn. of Ptcrotocrinus, apwdShumard(1866).
Coinatulina, d'Orbigny (1852), Oxfordian, C. costata = Solanocrinust Goldf., which

is syn. of Antedon.

Comaturclla, Miinster (1839) ; syn. of Antedon.

Conocrimis, Troost, nom. nud. (1850) ; syn. of Alloprosallocrinus, apud. W. & Sp.
Cophinus, Murchison, ex Konig MS. (1839), Silurian, England. Impressions of

stems.

Crumenaecrinus, Troost, nom. nud. (1850).
Cupulocrinus, d'Orb. (1850), based on ticypkocrinus Jieterocostalis, Hall (1847),

which is referred by Waclismuth & Springer to Flexibilia Impinnata.
Cypressocrinus, a variant and possibly preferable spelling of Cuprcssocrinus.
Cystocrinus, C. F. Roemer (1860), Silurian, Tennessee. Stem only known, see

Fig. L. 2.
Daemonocrinus, Troost, nom. nud. (1850) ; syn. of Pterotocrinus, apud Shumard

(1866).

Decacnemos, Bronn ex Linck (1837) ; syn. of Antedon.
Decadactylocrinus, D. D. Owen, nom. nud. (1843) ; syn. of Hcterocrinus, apud

Shumard (1866).

Decameros, d'Orbigny ex Linck (1850) ; syn. of Antedon.
Ditnorphicrinus, d'Orbigny (1849), based on Platycrinus pentangularis, Miller ;

syn. of Oi'ophocrinus, p. 84.
Doliolocrinus, Troost MS., fide Hall (1858) ; D. ovalis is syn. of Dichocrinus

simplex, Shumard.

Donacicrinus, Troost, nom. nud. (1850).
Echinocriniis, L. Agassiz, 1841, and T. & T. Austin (1842) ; not a crinoid, but an

echinoid = Archaeocidaris, M'Coy, 1844.
Emperocrinus, Miller & Gurley (1895), Silurian, Indiana ; probably syn. of

Anthemocrinus ; but has 3 IBB, and was placed by its authors with

doubt in Taxocrinidae.
Encrinos, Agricola (1558), a stem-fragment composed of Pcntacrini or Astcriac

(q.v. supra),
s, Agricola (1558), a stem-fragment composed of Trochitae (q.v. infra).

THE CRINO1DEA 203

Ganymeda, J. E. Gray (1834) ; centrodorsal of Antedon, apud L. Agassiz (1841).

Gaurocrinus, S. A. Miller (1883), Ordovician, N. America ; referred by Wachs-
muth & Springer part to lleteocrinus, part to Ptychocrinus.

Glenotremites, Goldfuss (1832), centrodorsal of Antedon.

Gnathocrinus, T. & T. Austin, nom. nud. (1842). The genotype G. fusiformis
is proved by Austin's MS. drawings to be syn. of Millericrinus Pratti.

Goldfussia, Norman (1891, non Castelnau, 1843), proposed for Comatula vel Corn-
aster multiradiata, Goldf. w<wLinn., nee Lam., a quite unrecognisable form.

Grammocrinus, Eichwald (1859), Ordovician ; colmnnals incertae sedis.

Halophenix "of our British Museum," fide Cumberland (1826); syn. of
Pentacrinus.

Helmintholiihus entrochus, Linn. (1768) ; stems of various Palaeozoic Crinoids.

Helmintholithus portentosus, Linn. (1768); syn. of Pentacrinus.

Hertha, v. Hagenow (1840) ; centrodorsal of Antedon.

Icosidactylocrinus, D. D. Owen, nom. nud. (1843) ; referred to Glyptocriniu by
Shumard (1866).

Kallispongia, Wright (1877) ; larva of Antedon.

Koninckocrinus, Seeley, nom. nud. (1864) ; syn. of Torynocrinus and Acrochordo-
crinus.

Afedusacrinus, T. & T. Austin, nom. nud., fide "Wachsmuth & Springer (1881).

Microcrinus, Emmons (1858), Eocene, N. Carolina ; probably centrodorsal of
Atelecrinus.

Mitrocrinus, Miller & Gurley (1894), Ordovician, Tennessee ; based on a six-
rayed individual, probably abnormal, and a Periechocrinid or Carpocrinid.

Pachyantedon, Jaekel (1891), Upper Cretaceous, N. Germany. Based on impres-
sion of stout arms, and a few cirri ; Br and cirrals in zigzag.

Pachycrinus, Eichwald (1840), Carboniferous, Russia ; columnals only, in part
a syn. of Platycrinus (see p. 132).

Pachyocrimis, E. Billings (1859), Ordovician, Canada ; founded on a single base,
with no diagnostic features.

Pentagonites, Rafmesque (1819), based on pentagonal coluinnals, probably of a
Heterocrinid.

Perischodomus magnus, Tornquist (1894) ; syn. of Adelocrinus hystrix, Phill.

Petinocrinus, Hall (1859), Geol. Iowa, vol. i. Suppl. p. 49. Not seen.

Phialocrinus, ' Eichwald (1860), Ordovician, Russia ; encrusting roots with ridged
under surface (see p. 133).

PlatyspJiaerites, Trenkner (1868), doubtfully crinoidal.

Proteuryale, J. Miiller ex C. F. Roemer MS. (1855) ; syn. of Cylicocrinus con-
fluentinus, apud Jaekel (1895).

Pterocoma, L. Agassiz (1835) ; syn. of Antedon.

Ehodocalix, Trenkner (1868), doubtfully Crinoidal.

Shumardocrinus, Miller & Gurley (1895), based on imperfect specimens of
Steganocrinus concinnns.

Solacrinus, L. Agassiz (1835), for Solanocrinus, Goldf. (1832), which is syn. of
Antedon.

Sphenocrinus, Eichwald (1856), quinquepartite columnals.

Sycocrinus, T. & T. Austin (1843) ; the authors' MS. drawings suggest that
S. clausus — Lageniocrinus, S. Jacksoni=Cryptocrinus, and S. anapepta-
tnenos — Hypocrinus.

Syringocrinus, E. Billings (1859), Ordovician, Canada ; stem of Dendrocystis.

Tetracrinus, T. & T. Austin, nom. nud. (1842, non Miinster). Not known.

204 THE CRINOIDEA

Tetramerocrinus, T. & T. Austin (1843) ;. shown by labels and MS. to be syn.

of Melocrinus or Mariacrinw.
Thalamocrinus, Miller & Gurley (1895), Silurian, Tennessee. Cup alone

known ; like Bactrocrinus or Homocrinus, without RA.
Trianisites, Rafinesque, fide L. Agassiz (1841) ; not known.
TriplaricriniM, Goldfuss, MS. label for Hexacrinus pateraeformis,fide L, Schultze

(1867).

Trochita, Auctt. vett., a round columnal with radiating striae.
VT,etavicrinus, Waagen & Jahn (1899), Silurian, Bohemia ; if monocyclic, must

be a Batocrinoid ; and if each Br bears more than one pinnule, is probably

a Carpocrinid.
Xenocrinus, Jahn (1892) non Miller; altered to Zenkericrinus, Waagen &

Jahn (1899), Silurian, Bohemia ; based on an imperfect cup, which does

not differ from Mariacrimis (p. 161).

A few of the Terms used by some other writers, and not previously alluded
to, may be correlated with those of the present work : —

Subradialia, de Koninck and Americans = BB of Dicyclica.

Subradiale, Jaekel = RA [and presumably all Rt] in Monocyclica Inadunata.

Subannlc, Jaekel = RA in Dicyclica Inadunata.

Azygos plates, Billings et alii = anals in general, and RA in particular.

Costalm = IBB in Marsupitcs (J. S. Miller) ; IBB & BB in Dendrocrinus (Hall) ;

BB in clearly dicyclic Inadunata and Flexibilia, and in Dimerocrinus

(J. S. Miller, Hall) ; BB of all Crinoidea (Loven) ; Ri in Heterocrinus

(Hall) ; RR in Scyphocrinns, Lyriocrimis, Macrostylocrinus (Hall) ; BR &

IBri in monocyclic or j>seudomonocyclic Camerata and Articulate (J. S.

Miller, Hall) ; RR & IBr in Cladocrinoidea only (Jaekel) ; IBr in all

Crinoidea (Bather in earliest papers ; Wachsmuth & Springer, 1897) ;

iBrt in Eucalyptocrinus (Hall).
Scapula, J. S. Miller, Hall ; Radiale articulare, L. Schultze, Zittel = the

proximal plate in a ray tha.t has an articular facet for the arms, therefore

may be R, IBrlf IBr2, or lAx.
Radialia, Miiller, Zittel ; Primary Radials, Wachsmuth & Springer (1879-81)

= all plates in a ray up to first forking, i.e. R & IBr up to lAx.
Brachialia of 1st, 2nd, 3rd, etc. order, Miiller, Zittel, Wachsmuth & Springer

l^e fore 1890 ; articles brachiaux, de Koninck = free Br only.
Brachialia, Dibrachialia, Tribrachialia, Jaekel = IBr, IIBr, IIIB-
Brachialia, P. H. Carpenter, before 1890 = Finials.
Radialia distichalia or Distichalia, Miiller, Zittel, Waagen & Jahii ; Secondary

Radials, Wachsmuth .& Springer (1879-81) ; Dicostalia, Jaekel = IIBr.
Distichalia, P. H. Carpenter (1879), Wachsmuth & Springer (1897) = IIBr.,

fixed or free.
Palmaria, Huxley (1877), P. H.'Carpenter (1879), Wachsmuth & Springer (1897)

Postpalmaria of 1st, 2nd, 3rd, etc. order, Carpenter, Wachsmuth & Springer
(1897) = IVBr, VBr, VIBr, etc.

The corresponding terms for supplementary plates — Intercostal, Inter scapular,
Interradial, * Inter axillary, Interdistichal, Intcrpalmar, and Interbrachial — have
also been used in diverse senses.

For Literature of Crinoidea see p. 211.

CHAPTER XII.

THE EDRIOASTEROIDEA.1

CLASS IV. EDRIOASTEROIDEA, E. BILLINGS (1854,-58;
HUXLEY, 1877; and BATHER, 1899)

( = THYROLDA, Chapman, 1860; AGELACRINOLDEA, S. A. Miller,
1877-83 ; Worthen, 1883; CYSTASTEROIDEA, Steinmann, 1888;
F. Bernard, 1893; THECOIDEA, Jaekel, 1895).

Not divided into Orders.

PELMATOZOA in which the theca is composed of an indefinite
number of irregular plates, some of which are variously differen-
tiated in different genera ; with no subvective skeletal appendages,
but with central mouth, from which there radiate through the
theca five unbranched ambulacra, composed of a double series of
alternating plates (covering-plates), sometimes supported by an
outer series of larger alternating plates (side -plates or flooring-
plates). Pores between (not through) the ambulacral elements,
or between them and the thecal plates, permitted the passage of
extensions from the perradial water-vessels. Anus in posterior
interradius, on oral surface, closed by valvular pyramid. Hydro-
pore (usually, if not always, present) between mouth and anus.

Reducing the characters of the Edrioasteroidea to their simplest
expression, one may imagine a schematic type of the following
nature : — A flexible theca of sack form, composed of numerous
irregular, polygonal plates deposited in the integument, probably
with the stroma- strands between them still plainly visible (cf.
Stromatocystis) ; it would have a mouth in the centre of the upper
surface, and would be attached by some indefinite portion of the
lower surface ; the anus, with its valvular covering, would lie on
the upper surface, and there would probably be a hydropore
between it and the mouth. So far this type would present primi-
tive characters like those of the earlier Amphoridea, from which

1 By F. A. Bather, M.A.

206

THE EDR10ASTEROIDEA

one may suppose it to have been derived. But the structure and
relations of the ambulacra, even in their least specialised form, at
once remove the type from primitive simplicity, and place it on a
road different from that traversed by other Pelmatozoa. The evi-
dence suggests the existence of a circumoesophageal water-ring, with
five perradial canals, and their associated nerves and blood-vessels,
passing between or below the thecal plates, and underlying a
ciliated food-groove, which was covered by an alternating series
of movable plates (covering-plates = ambulacrals of Crinoidea, but
probably not those of Echinoidea and Asteroidea). Pores between
the plates lining or flooring the groove (adambulacrals of Pelma-
tozoa, but perhaps = superambulacrals of Asteroidea) permitted
the passage either of podia or ampullae. In Cystidea, Blastoidea,
and primitive Crinoidea, on the other hand, there was a free exit
for the ambulacral organs only through the peristome ; in fact,
Blastoidea and Cystidea present no evidence that they possessed
perradial water-vessels and podia at all.

AS 4

Fio. II.

Cyntttster granulatus, from pos-
terior, showing oral surface in per-
Fici. I. spective. The two left- hand rays

Stnmatocystii pentangular is, oral surface. retain the covering-plates, which are

As, anus; 0, peristomial plates ; c.p, covering- °8t from the others Lettering as

plates ; s.p, side-plates ; ia, interambulacrals. '» Fl8- \ (Reconstructed from Hall's

(Reconstructed from Pompeckj's figures.) tigures.) x 3 diam.

Slightly enlarged.

The primitive sack form did not long persist, but the follow-
ing characters were, as a rule, impressed upon it : a sessile habit,
the consequent assumption of a circular, flattened form, the differ-
entiation of the upper and under surfaces, the development of
marginals or concentric frame-plates, and the tendency to increase
the food-gathering surface by spiral coiling of the ambulacra in
either sinistral or dextral direction. According to the varied extent
of these several modifications, the Edrioasteroidea are divisible into
3 families — Agelacrinidae, Cyathocystidae, Edrioasteridae. But
to these must be added a fourth, Steganoblastidae, in which the
development of a short stem was correlated with greater concentra-
tion and regularity of the thecal elements.

THE EDRIOASTEROIDEA

207

FAMILY 1. AGELACRINIDAE. Edrioasteroidea with a theca composed
mostly of thin plates, flexible, attached temporarily or permanently by
the greater part of the aboral surface ; with ambulacra confined to the oral
surface. Genera — Stromatocystis, Pompeckj (1896), Cambrian, Bohemia,
seems to have had a somewhat flexible theca of non-imbricate plates
(Fig. I.). Rays straight and extending to the margin, imposing on the
theca a subpentagonal outline ; composed of stout and long alternating
side-plates, along the outer margins of which are pores [for podia 1]
while along their inner margins is a groove, protected by minute
covering-plates. Four large and many small plates [modified side- and
covering-plates] surround and cover the mouth. Interambulacrals hex-
agonal, united by stroma- strands. Under side of theca composed of
irregular plates, without evidence of stroma-strands, and larger towards
the middle. The animal was probably sessile on its under surface, but
perhaps not fixed permanently. Cystaster, Hall (1872, Thecocystis, Jak.),
Ordovician, Ohio, is also primitive (Fig. II.). Theca sac-like, composed of

Fi<;. III.

Aijelacrlnus hamiltonensis, oral surface,
wwft, marginals, consisting of the large
internal (i) and the small external (c) ; M,
supposed madreporite. Other letters as
before. (After Hall.) x 2.

Fio. IV.

Lcpidodiscus cincinnaten.-'is, oral surface.
Lettering as before. (After Hall.) x \ .

minute plates, not always attached (?) ; rays straight, composed of alternat-
ing covering -plates supported on side -plates ; no hydropore observed.
Hemicystis, Hall (1852), Ordovician and Silurian, N. America and
Bohemia, shows an advance on Cystaster in the imbrication of the
thecal plates, and their differentiation into larger interambulacrals,
and a zone of smaller marginals. This leads on to Agelacrinus
and its allies Streptaster and Lepidodiscus, all which are characterised
by the curvature of the rays, sinistrally, dextrally, or both, by the
elaboration of the marginal zone, and by their flattened sessile habit,
being usually attached to brachiopod shells. The type -species of
Agelacrinus is the Devonian A. hamiltonensis, Vanuxem (1842), (Fig.
III.), in which the anterior and two left-hand rays curve sinistrally,
the two others dextrally ; the interambulacrals are large, non-imbricate,
and radiately ridged ; there is a border of large plates, with an outer

208 THE EDRIOASTEROIDEA

border of small plates. Lepidodiscus, Meek & Worthen (1868), Ordovician
to Carboniferous, has imbricating interambulacrals and a border of small
imbricating plates ; plates forming the floor of the ambulacra, and a pen-
tagonal internal frame around the mouth, have been described by Miller &
Faber (1892), but more details are needed; Hall figures right posterior
ray as dextral, and the rest as sinistral (Fig. IV.) ; other arrangements
may obtain, but the rays adjacent to the anus always curve towards it.
The Devonian Haplocystis, C. F. Roemer (1855), is known from an internal
cast, apparently proving that the ambulacra were floored by a single
row of plates [? fused adambulacrals], between the edges of which were
pores [for passage of podia]. The Ordovician Streptaster, Hall (1872),
has all its rays sinistral. Discocystis, Gregory (1896), based on Echino-
discus optatus, Worthen & Miller, is doubtful.

FAMILY 2. CYATHOCYSTIDAE. Edrioasteroidea with the theca composed
on the oral surface of five deltoids surrounded by marginals, but below
of a fused solid mass of stereom, with irregular longitudinal sutures ; per-
manently attached by the aboral surface as by an encrusting root ;
ambulacra confined to oral surface. Genus — Cyathocystis, Schmidt (1880),
Ordovician, Esthonia (Fig. V.), is not far removed from Stromatocystis and
Cystasterj but the trend of its evolution is quite away from that of either
Edrioaster or Ayelacrinus. Upper ambulacral surface of theca bordered
by a pentagonal frame of 40 marginals ; rays straight, with a single
series of covering-plates ; five larger plates cover the mouth. The
minute plates that formed the theca of Cystaster are here fused into solid
masses ; thus there arise between the ambulacra 5 large A, and below the
marginals a massive cup, fixed to some foreign body by its base, and
occasionally marked by obscure longitudinal sutures, irregular in number
and position, but never more than 5. Anus, with pyramid of 5 plates,
lies between a A and the adjacent marginals ; no hydropore observed.

c.p.

Cyathncystis Plaittinae. 1, two
individuals growing on a pebble of
rolled coral, and seen from the side.
(After Schmidt.) Nat. size. 2,
oral surfacp. As, anus ; A, inter-
ambulacrals fused into deltoids ;
in-ill, marginals ; c.p,covering-plates,
partly removed from the posterior
rays exposing fp, which are not a
distinct series of flooring -plates,
as represented, but ledges project-
ing from the under side of the
deltoids. (Diagrammatised from
Schmidt.) x 3 diam.

FAMILY 3. EDRIOASTERIDAE. Edrioasteroidea with flexible theca com-
posed of thin plates ; attached, if at all, by a small central portion of the
excavate aboral surface ; ambulacra pass on to aboral surface. Genera —
Aesiocystis, Miller & Gurley (1894), Ordovician, Kentucky, has a sub-
pentagonal theca, with height two-thirds the width ; rays wide and straight,
with large covering-plates ; interambulacrals non-imbricate. Podial pores,
madreporite, and abactinal surface unknown ; but the genus appears to

THE EDRIOASTEROIDEA

209

connect this family with the more primitive forms of Agelacrinidae.
Edrioaster, Billings (1858), Ordovician, Canada (Fig. VI.), also includes
" Agelacrinite* Buchianus" Forbes (1848), from Wales. Theca about half
as high as wide, depressed slightly at oral pole. Kays curved, all or
some sinistral or dextral, passing on to under surface of theca ; arnbu-
lacral groove floored by alternating plates [? adambulacrals], between which
were pores [? for podia] ; with covering -plates, i.e. ambulacrals, over
the groove. Interambulacrals non- imbricate. Under- surface of theca
excavate, its central region composed of a flexible membrane set with
minute imbricating plates, and in a frame of about 11 large plates.
Whether the animal was attached by the central part of this membrane
is doubtful ; immediately round the centre this is evaginated in five lobes,
apparently caused by the pressure of some internal organs [?gonads],

amb — -

amb

ad

FIG. VI.

Edrioaster Bigsbyi. 1, oral surface, with covering-plates (amb) on the anterior and left
anterior grooves, but removed from the others, which show only the side- or flooring-plates
(ad), between which are pores (p). The greater part of the subpentagonal peristome (ps) is
roofed by enlarged covering-plates, ia, interambulacrals, one of which is a madreporite (Af),
2, section across the same specimen through the right anterior radius and left posterior
interradius. The covering-plates are removed except just over the peristome, and in the am-
bulacrum seen in section on the left. /, frame of stouter plates ; m, membrane with imbricating
plates, thrown into five lobes (I). 3, section across an ambulacrum, with plates in situ
covering ventral groove (w/). Dotted surfaces are the natural edges of the plates, ruled surfaces
are cut through the plates. (All slightly diagrammatised from a specimen belonging to the
Canadian Geological Survey.) 1 and 2 are nat. size.

which must have acquired pentamerous symmetry. Dinocystis, Bather
(1898), Uppermost Devonian, Belgium, has a slighter frame on the aboral
surface, and the surrounding region composed of a thin flexible integument
containing narrow imbricating ossicles ; otherwise like Edrioaster.

FAMILY 4. STEGANOBLASTIDAE. Edrioasteroidea, with a rigid theca
composed of plates relatively larger and thicker than in other families of
this class ; these include elements comparable to the RR and BB of
Blastoidea ; BB attached to a stem, probably short ; ambulacra descend
into the radials. Genus — Steyanoblastus, Whiteaves (1897, originally
described as Astrocystites, name preoccupied), Ordovician, Canada (I .g.
VII.). The remarkable resemblance to Asteroblastusy insisted on by its
founder, suggested the reference of Steyanoblastus to the Protoblastoidea
(Bather, 1899); but tlie ambulacra are now known to have essentially
the same structure as in Edrioaster, while the absence of brachioles may
be maintained with confidence. Theca piriform, its plates strongly marked

14

210

THE EDRIOASTEROIDEA

with axial folds, and consisting of : BB (5 ?, sutures not clear) ; RR, 5,
alternating with BB, and receiving the distal ends of the ambulacra ;
interambulacrals, one large one and an uncertain number of smaller ones,
in each interradius ; 5 slightly-pitted plates of spear-head shape, stretching
up between the ambulacra to the oral pole and simulating 0 of Crinoidea
or A of Blastoidea, but perhaps being only proximal covering-plates.
The anus pierces one interambulacrum, and slightly disturbs the
pentamerous symmetry of the theca. From the mouth 5 ambulacra
stretch about half-way down the theca ; the adambulacrals (side- or
flooring-plates) appear almost anchylosed, but the pores between them
are very clear, and one can trace the original median line of suture ;
the ambulacrals or covering-plates were stout, at least in the proximal

6mb

amb

— St

Steganoblastus ottau-aensis, slightly restored from the type-specimens, and x 3 diani. 1,
oral surface ; 2, from 1. post, radius, udamb, adambulacrals or side-plates ; amb, ambulacrals
or covering-plates, mostly removed ; As, anus surrounded by small plates ; li, basal ; 1R, largo
median interradial ; 0, 5 orals or proximal ambulacrals; p, pores between side -plates; 2!,
radial ; St, fragment of stem.

regions, where they seem to have combined with the spear-head plates to
form a solid roof over mouth and food-grooves. Stem small, round, with
lumen less than half the diameter. Fifty years ago Steyanoblastus would
have been described as a generalised or synthetic type, with Cystid,
Blastoid, Crinoid, and Asteroid affinities ; it is simpler to regard it
as a specialised Edrioasteroid, in which features common in stalked
genera of other classes have been evolved independently under similar
conditions of existence.

CyclocystoideSy Billings and Salter (1858), Ordovician of N. America
and Britain (Fig. VIII.), probably belongs to this class, though not to any
of the recognised families. It is hardly well enough known to make
the type of an order as yet A ring of stout ossicles, more regular than
that seen in Agelacrinus and Edrioaster, forms a frame between which are

THE EDR1OASTEROIDEA

211

stretched two thinly plated membranes, rarely preserved. The dorsal

membrane contains irregular non-imbricating plates (Miller & Faber,

1892). The plates of the ventral membrane (apud J. Hall, 1866) were

delicately reticulate, and arranged in

numerous rays passing from a central

pyramid of minute plates [mouth ?].

Eccentrically, between two rays, is an

oval opening [anus ?]. Outside the

frame is a border of smaller plates

as in Agelacrinus. The animal was

not permanently sessile [but could

probably fix itself like a limpet].

As-

Cijclocystoides Sulti-ri. ,4s, supposed
region of anus ; mm, frame of larger mar-
ginals. The plated aspect of the flooring,
is based on
ise the figure
pied

The classificatory position here
assigned to the Edrioasteroidea
is not that usually accepted by
zoologists, although many have FUJ. vm.

given them equal, nearly equal,
or greater classificatory value
under various names (see heading, «» seen through the openi

,,,, Miller & Faber; othei-wb

p. 205). Whatever may eventu- copied from Hail, x >.=.
ally prove to be the value of the

characters insisted on in the present work or in others, even if
there be traced a closer connection with Diploporita than is as yet
apparent, the Edrioasteroidea can never be thought a less distinct
or less homogeneous group than, say, the Blastoidea. The zoologi-
cal importance of the Edrioasteroidea is another reason for raising
them to this position. Many zoologists, e.g. Forbes, Billings, Neu-
mayr, Steinmann, have regarded them as connecting either the
Echinoidea or the Asteroidea with the Cystidea. On the hypothesis,
ever becoming more probable, that all eleutherozoic Echinoderms
are descended from some pelmatozoic ancestor or ancestors, then
the Edrioasteroidea are alone among Pelmatozoa in presenting a
type of ambulacrum from which the holothurian, stellerid, and
echinoid types may readily be derived.

LITERATURE OF PELMATOZOA.

1. Agassiz, A. 1892. (Calamocrinus Diomcdac . . . with notes on the apical

system and the homologies of Echinoderms.) Mem. Mus. Comp. Zool.
Harvard, vol. xvii., No. 2, 95 pp., xxxii. pis.

2. Angelin, N. P. 1878. Iconographia Crinoideorum in Stratis Sueciae

Siluricis Fossilium, 4to, iv. and 62 pp. , xxix. pis. Holmiae.

3. Austin, T. and T. 1843. (Descriptions of several new genera and species of

urinoidea.) Ann. Mag. Nat. Hist. (1) vol. xi. pp. 195-207. (Palaeo-
zoic Crinoids and Blastoids.)

4. 1843-1849. Monograph on recent and fossil Crinoidea, 4to, 128 pp.,

xvi. pis., London and Bristol.

212 THE EDRIOASTEROIDEA

5. Barrande, J., continued by Waagen, W. andJahn, J. J. 1887, -99. Systeme

Silurien . . . de la Boheme. Recherches Pale"ont., vol. vii. Echinodennes.
I. Cystide'es, xvii. + 233 pp., xxxix. pis. n. Crinoides, vi. + 216 pp., xl.
pis. Prague.

6. Bather, F. A. 1890-1892. (British Fossil Crinoids, i.-viii.) Ann. Mag.

Nat. Hist. (6), vol. v. pp. 306-334, pi. xiv., 373-388, pi. xv., 485, 486 ;
vol. vi. pp. 222-235, pi. x. ; vol. vii. pp. 35-40, pi. i., 389-413, pi: xii. ;
vol. ix. pp. 189-236, pis. xi.-xiii. (Inadunata, chiefly Silurian.)

7. 1893. (The Crinoidea of Gotland. Tart I. The Crinoidea Inadu-
nata.) K. Svenska Vet.-Akad. Handl. vol. xxv. No. 2, 200 pp., x. pis.
(All Silurian ; with Bibliography.)

8. 1896. (Uintacrinus : a morphological study.) Proc. Zool. Soc.

London, for 1895, pp. 974-1004, pis. liv.-lvi.

9. 1897. (Hapalocrinus Vidoriae, n.s., Silurian, Melbourne, and its

relation to the Platycrinidae.) Geol. Mag., n.s., dec. iv. vol. iv. pp. 337-
345, pi. xv.

10. 1898-99. (Wachsmuth and Springer's Monograph on Crinoids.) Geol.

Mag., n.s., dec. iv. vols. v., vi. ; also separately published, London. (The
separate issue contains a Bibliography of Wachsmuth.)

11. 1898-99. (Studies in Edrioasteroidea. ) Geol. Mag., n.s., dec. iv.

vols. v. and vi. In progress.

12. 1899. (A phylogenetic classification of the Pelmatozoa.) Rep. Brit.

Assoc. for 1898, pp. 916-923.

13. Beyrich, H. E. 1858. (Ueber die Crinoiden des Muschelkalks.) Abh. Akad.

Wiss. Berlin, Phys. Klasse, 1857, pp. 1-49, pis. i.-ii.

14. 1871. (Ueber die Basis der Crinoidea Brachiata.) Monatsber. Akad.

Wiss. Berlin, pp. 33-55. Translation in Ann. Mag. Nat. Hist. (4), vol. xi.
pp. 393-411.

15. Billings, E. 1858. (On the Cystideae of the Lower Silurian rocks of

Canada.) Canadian Organic Remains, dec. iii. pp. 9-74, pis. i.-vii.,
Montreal.

16. 1859. (On the Crinoideae of the Lower Silurian rocks of Canada.)

Op. ciL, dec. iv. pp. v., vi., and 7-66, pis. i.-x.

17. 1869-70. (Notes on the structure" of the Crinoidea, Cystidea, and

Blastoidea.) Amer. Journ. Sci. (2), vol. xlviii. pp. 69-83 ; xlix. pp. 51-
58 ; and 1. pp. 225-240, and 436. Reprint in Ann. Mag. Nat. Hist. (4),
vol. v. pp. 251-266, 409-416, and vii. 142-158 ; and in Palaeozoic Fossils,
vol. ii. pp. 90-128, Montreal.

18. Billings, W. R. 1881-87. Various papers in Trans. Ottawa Field-Nat. Club,

vols. i. and ii. ; and Ottawa Naturalist, vol. i. (Cystids and Crinoids.)

19. Buck, C. L. von. 1845. (Ueber Cystideen eingeleitet durch die Entwicke-

luug der Eigenthiimlichkeiten von Caryocrinus ornatus, Say.) Abhandl.
Akad. Wiss. Berlin, Phys. KL, 1844, pp. 89-116, pis. i. and ii. Trans-
lation in Quart. Journ. Geol. Soc. London, vol. ii., part ii., pp. 20-42,
pis. iii., iv.

20. Carpenter, P. H. 1884, -88. (Report on the Crinoidea, etc.) Challenger

Report, Zoology, vol. xi., No. 26, and vol. xxvi., No. 60. (With Biblio-
graphy of Recent Crinoids. The first volume contains much on Crinoids
generally. )

21. 1891. (On certain points in the Morphology of the Cystidea.)

Journ. Linn. Soc. London (Zool.), vol. xxiv. pp. 1-52, pi. i.

THE EDRIOASTEROIDEA 213

22. Carpenter, P. H. 1891. Geol. Mag., n.s., dec. iii., vol. viii. pp. 573-575.

(Obituary notice, with supplementary Bibliography. )

23. Carpenter, W. B. 1866-76. (Structure, physiology, and development of

Antedon rosaceus.) Phil. Trans. R. Soc. 1866, pp. 671-756, pis. xxxi.-
xliii. ; and Proc. R. Soc. vol. xxiv. pp. 211-231, pis. viii., ix. (Part 1
contains a valuable Historical Summary.)

24. 1884. (On the nervous system of the Crinoidea. ) Proc. R. Soc., No.

232, pp. 67-76.

25. Cumberland, G. 1826. Reliquiae Conservatae, etc., 4to, x. and 45 pp., xiii.

pis. , Bristol. ( Blastoids and Crinoids. )

26. Danielssen, D. C. 1892. (Crinoida) Norske Nordhavs-Exped. Zoologi, Part

xxi., 28 pp., v. pis., 1 map, Christiania. (Bathycrinus.)

27. Dujardin, F. and Hupe, H. 1862. Hietoire Naturelle des Zoophytes Echino-

dermes, etc., 8vo, 627 pp., x. pis., in Suites a Buffon, Paris. (A con-
venient summary to date.)

28. Eiclnvald, E. v. 1859. Lethaea Rossica . . . Ancienne Periode, Livr. v.

(Palaeozoic Crinoids and Cystids.)

29. Ethcridge, E., fil. 1892. (Monograph of the Carboniferous and Permo-

Carboniferous Invertebrata of New South Wales, II. Echinodermata, etc.)
Mem. Geol. Surv. N.S.W., Pal., No. 5. (Crinoids.)

30. 1892. Chapters xxii., xxxii., xxxiv., and xxxix. in Jack and Etheridge,

"The Geology and Palaeontology of Queensland and New Guinea," 8vo,
Brisbane. (Crinoids, Blastoids, and Echinoids.)

31. and Carpenter, P. H. 1886. Catalogue of the Blastoidea in the . . .

British Museum, 4to, xvi. +322 pp., xx. pis., London.

32. Follmann, 0. 1887. (Unterdevonische Crinoiden.) Verhandl. naturhist.

Ver. preuss. Rheinl. Jahrg. xliv. pp. 113-138, pis. ii., iii.

33. Forbes, E. 1848. (On the Cystideae of the Silurian rocks of the British

Islands.) Mem. Geol. Surv. Great Brit. vol. ii., part 2, pp. 483-538, pis.
xi.-xxiii.

34. (,'oldfuss, G. A. 1826-1833. Petrefacta Germaniae, etc., Erster Theil, fol.

x. and 252 pp. frontisp. and Ixxii. pis., of which xl.-xliii. are double.
(Div. II., Radiariorum Reliqq. pp. 115-214, pis. xxxix. -Ixiv.)

35. Gijlloih-ahl, J. A. 1772. (Beskrifning pa de sa kallade Crystall-aplen ock

kalk-bollar, sasom petreficerade Djur af Echini genus, eller dess nar-
inaste slagtingar. ) K. Svenska Vetenskaps-Acad. Handl. vol. xxxiii.
pp. 239-261, pis. viii. and ix. (Sphaeronis and Echinosphaera.)

36. Haeckd, E. 1896. (Die Amphorideen und Cystoideen, etc.) Festschr. fiir

Gegenbaur, No. 1, 179 pp., v. pis. (Bibliography.)

37. Hall, J. 1847, 1852, 1859. Palaeontology of New York, vols. i., ii., iii.

in Natural History of New York, 4to, Albany. (Palaeozoic Crinoids and
Cystids.)

38. 1848 onwards. (Annual Reports of the State Geologist.) Regents'

Reports on State Cab. Nat. Hist. N.Y., 8vo, Albany. (Palaeozoic
Pelmatozoa of all classes. )

39. 1858. Report on the Geological Survey of the State of Iowa, vol. i.,

Part ii., Palaeontology. (Palaeozoic Crinoids, Cystids, and an Edrio-
asteroid.)

40. Hinde, G. J. 1885. (New species of Crinoids with articulating spines.)

Ann. Mag. Nat. Hist. (5), vol. xv. pp. 157-173, pi. vi. (Hystricrinus^
Arthracantha.}

214 THE EDR1OASTEROIDEA

41. Hisinger, W. 1837, 1840, 1841. Lethaea suecica, 124 pp., pis. A, B, C,

i.-xxxvi. Supplementum secundum, 11 pp., pis. xxxviii. - xxxix.
Supplement! secundi continuatio, 6 pp., pis. xl.-xlii., 4to. Holmiae.

42. JaekelyO. 1891. (Ueber Holopocriniden [ = Eugeniacrinidae and Holopodi-

dae.]) Zeitschr. deutsch. geol. Ges., vol. xliii. pp. 557-671, pis. xxxiv.-xliii.

43. 1893. (Ueber Plicatocriniden, Hyocrinus, und Saccocoma.) Zeitschr.

deutsch. geol. Ges., vol. xliv. pp. 619-696, pis. xxv.-xxx.

44. 1894. (Entwurf einer Morphogenie und Phylogenie der Crinoiden.)

Sitzber. Ges. naturf. Freunde Berlin, Jahrg. 1894, pp. 101-121.

45. 1895. (Beitriige zur Kenntniss der palaozoischen Crinoiden Deutsch-

lands.) Palaeont. Abhandl. Jena, neue Folge, vol. iii., 116 pp., x. pis.

46. 1895. (Organisation der Cystoideen.) Verhdl. deutsch. Zool. Ges.,

1895, pp. 109-121. Prelim, to Stammesgeschichte der Pelmatozoen, I.
1899.

47. Koenen, A. v. 1886. (Neue Cystideen aus den Caradoc-schichten der

Gegend von Montpellicr.) Neues Jahrb. f. Min., 1886, II. pp. 246-254,
pis. viii., ix.

48. 1887. (Beitrag zur Kenntniss der Crinoiden des Musckelkalks.)

Abhandl. K. Ges. Wiss. Gottingen, vol. xxxiv., Phys. Kl. pp. 1-44, pi. i.

49. 1895. (Ueber die Entwickelung von Dadocrinus yracilis und Holocrinus

Wagiwit etc.) Nachr. K. Ges. Wiss. Gottingen, Math. -Phys. KL, 1895,
pp. 283-293.

50. Koniiick, L. G. de, ai\d Le Hon, H. 1854. (Recherches sur les Criuoides

du terrain Carbonifere de la Belgique.) Mem. Acad. Roy. Belgiijue, vol.
xxvii., Mem. 3, 215 pp., vii. pis. (Ciinoids and Blastoids. With very
full Bibliography.)

51. Loriol (afterwards Loriol le Fort], P. dc. 1877,-79. (Monographie des

Crinoides Fossiles de la Suisse.) Mem. Soc. Pal. Suisse, vols. iv.-vi.

52. 1882-89. Paleontologie Fraucaisc, Ire serie . . . Invert6brds.

Terrain Jurassique. Tome XL, Ire partie, Crinoides, 8vo, Paris.

53. Liidwig, H. 1877. ( Beitriige zur Anatomic der Crinoideen.) Zeitschr. wiss.

Zool. vol. xxviii. pp. 255-353, pis. xii.-xix.

54. 1877. (Zur Anatomie des Rhizocrinits lofotciisis.) Op. cit. vol.

xxix. pp. 47-76, pis. v., vi.

55. Marshall, A. M. 1884. (On the nervous system of Antcdon rosaceus.)

Quart. Journ. Micr. Sci., n.s., vol. xxiv. pp. 507-548, pi. xxxv.

56. Meek, F. B. and Worthen, A. H. 1869. (Notes on some points in the

structure and habits of the Palaeozoic Crinoidea. ) Proc. Acad. Nat. Sci.
Philadelphia, for 1868, pp. 323-334 ; reprinted in Amer. Journ. Sci. (2),
vol. xlviii. pp. 23-40, and in Canad. Naturalist, n.s., vol. iv. pp. 434-452.
56a. See Worthen, A. H., and others.

57. Meyer, H. v. 1826. (Beschreibung des Echino-Encrinites Senkenbergii,

einer neu enddeckten Versteinerung. ) Kastner's Archiv f. d. gesammte
Naturlehre, vol. vii. pp. 185-192, pi. ii. (Nachtrag), op. cit. vol. viii. pp.
232-237.

58. 1837. (Isocrinus and Chelocrinus.) Mus. Senkenberg. vol. ii. pp.

251-263, pi. xvi.

59. Miller, J. S. 1821. A Natural History of the Crinoidea, etc., 4to, viii.

+ 150 pp., 1. pis., Bristol.

60. Miller, S. A. 1889,-92,-97. North American Geology and Palaeontology,

etc., with two Appendices, 8vo, 793 pp., Cincinnati. (Index to first

THE EDRIOASTEROIDEA 215

descriptions of all N. American fossil Echinoderms ; including references
to Miller's numerous papers in Journ. Cincinnati Soc. Nat. Hist., and in
the publications of the Illinois, Indiana, and Missouri Geol. Surveys.)

61. Mutter, Joh. 1843. (Ueber den Ban des Pentacrinus caput Medusae.}

Abhandl. Akad. Wiss. Berlin, Phys. Kl. vol. for 1841, pp. 177-248, vi. pis.

62. 1857. (Ueber neue Echinodermen des Eifeler Kalkes.) Abhandl.

Akad. Wiss. Berlin, Phys. Kl. vol. for 1856, pp. 243-268, pis. i.-iv.
(Crinoids. )

63. Murchison, R. I. 1839. The Silurian System, etc., 4to, xxxii. and 768 pp.

London. (Crinoidea, by J. Phillips.)

64. Ncumayr, M. 1889. Die Stamme des Thierreiches, vol. i. (all published),

8vo, vi. + 603 pp., Vienna. (Phylogeny of all Echinoderms.)

65. Oehlert, D. P. 1879. (Deux nuiiveaux genres de Crinoides du terrain

de'vonien de la Mayenne.) Bull. Soc. geol. France (3), vol. vii. pp. 6-10,

pis. i., ii. (Thylacocrinus.)
66. 1882. (Crinoides nouveaux du Devonien, etc.) Ser. cit. vol. x. pp.

352-363, pis. viii., ix.
67. 1891. (Spyridiocrinus), ser. cit. vol. xix. pp. 220-227, pis. vii., viii.

68. 1892. (Deux Crinoides nouveaux du Devonien), torn. cit. pp. 834-853,

pi. xviii. (Diamenocrinus.)

69. 1896. (Fossiles deVoniens de Sta. Lucia), ser. cit. vol. xxiv. pp. 814-

875, pis. xxvi.-xxviii. (Crinoids and Blastoids.)

70. Orbigny, A. D. de. 1840, -58. Histoire naturelle . . . des Crinoides, etc.,

4to, 98 pp., xviii. pis., Paris. (Apiocrinidae, Bourgueticrinidae.)

71. 1849-52. Cours elementaire de Paleontologie, etc., 2 vols. and Atlas.

Prodrome de Paleontologie ... des animaux mollusques et rayonnes.
3 vols., Paris.

72. Perrier, J. 0. E, 1886, -90. (Memoire sur 1'organisation et le developpe-

ment de la Comatule de la Mediterranee. ) Nouv. Arch. Mus. Hist. Nat.
(2), vol. ix. pp. 53-348, x. pis., and (3), vol. i. pp. 169-286, vol. ii. pp.
1-86, pis. i., ii.

73. Pictct, F. J. 1857. Traite de Paleontologie, 2e edit. vol. iv. Oinoides

(Blastoides et Cystidees), pp. 278-345, Paris.

74. Pompcckj, J. F. 1896. (Die Fauna des Cambrium von TejTovic und Skrej

in Bbhmen.) Jahrb. geol. Reichsanst. Wien, vol. xlv. pp. 495-614, pis.
xiii. -xvii. (Cystidea and Edrioasteroidea..)

75. Qucnstedt, F. A. 1874-76. Die Asteriden und Encriniden nebst Cysti-und

Blastoideen, Petrefactenkunde Deutschlands, 1 Abth. vol. iv., 8vo, viii.

and 742 pp., Atlas pis. xc.-cxiv., Leipzig.
76. 1885. Handbuch dcr Petrefactenkunde, Dritte Auflage, 4 Abth. pp.

801-988, pis. Ixiii.-lxxx., Tiibingen.
77. Roemcr, 0. F. 1851. (Monographic der . . . Blastoideen, etc.) Arch.

Naturges. vol. xvii. (1), pp. 323-352, pis. iv.-viii.
78. 1855. (Crinoidea, etc.) Lethaea Geognostica, 3rd ed., vol. ii. pp.

210-280, Stuttgart. (All Palaeozoic Pelmatozoa.)

79. Rouault, M. 1883. (Euvres posthumes. Amorphozoaires siluriens, publiees

par Lebesconte, 4to, 73 pp., xxii. pis., Rennes. ( Amphoridea. )

80. Salter, J. W. , and Billings, E. 1858. (On Cydocystoides, etc.) Canad.org.

Remains, dec. iii., pp. 86-90, pi. x. bis.

81. Sars, M. 1868. Memoires pour servir a la connaissance des Crinoides

vivants, 4to, 65 pp., vi. pis., Christiania. (Rhizocrinus, Antedon.}

216 THE EDRIOASTEROIDEA

82. Schmidt, F. B. von. 1874. (Miscellanea Silnrica, II. Ueber einige neue

und wenig bekannte Baltisch-Silurische Petrefacten. ) Mem. Acad. Imp.
Sci. St-Pe"tersbourg (7), vol. xxi. pp. 1-48, pis. i.-iv.

83. 1880. (Ueber Cyathocystis Plautinae, etc. ) Verhandl. Mineral. Ges.

St. Petersburg (2), vol. xv. pp. 1-7.

84. Schultze, L. 1867. (Monographic der Echinodermen des Eifler Kalkes.)

Denkschr. Akad. Wiss. Wien, Math.-nat. Kl. vol. xxvi. pp. 113-230,
pis. i.-xiii.

85. Shumard, B. F. 1866. (A Catalogue of the Palaeozoic Fossils of N.

America.) Trans. Acad. Sci. St. Louis, vol. ii. pp. 334-407. (Contains
also "A Chronological List of Works which contain Descriptions or Notices
of N. American Palaeozoic Echinodermata " from 1811 to 1865.)

86. Trautschold, H. 1879. (Die Kalkbriiche von Mjatschkowa . . . Schluss.)

Nouv. Mem. Soc. Nat. Moscou, vol. xiv. pp. 101-180, pis. xii.-xviii.
(Carboniferous Crinoidea.)

87. Polborth, A. v. 1842. (Ueber der Echinoencrinen. ) Bull. Scient. Acad

Imp. Sci. St-Pe'tersbourg, vol. x. pp. 293-303, pis. i. and ii.

88. 1844. (Ueber die Arnie der bisher zu dem armlosen Crinoiden

gezahlten Echino-Encrinen.) Bull. Classe Phys.-Math. Acad. Imp. Sci.
St-Petersbourg, vol. iii. No. 6, columns 91-96, with 1 plate.

89. 1846. (Ueber die russischen Sphaeroniten, eingeleitet durch einige

Betrachtungen ueber die Arme der Cystideen.) Verhandl. Mineral. Ges.
St. Petersburg, 1845-46, pp. 161-198, pis. ix., x.

90. 1870. (Ueber Achradocystites und Cystoblastns . . . eingeleitet durch

kritischo Betrachtungen ueber die Organe des Cystideen.) Mem. Acad.
Sci. St-Petersbourg (7), vol. xvi., No. 2, 15 pp.

91. Wachsmuth, C., and Springer, F. 1879-1886. (Revision of the Palaeo-

crinoidea.) Proc. Acad. Nat. Sci. Philadelphia, for 1879, pp. 226-378,
pis. xv.-xvii. ; for 1881, pp. 177-411, pis. xvii.-xix. ; for 1885, pp. 225-364,
pis. iv.-ix. ; for 1886, pp. 64-226. (Index to all genera and species of
Palaeozoic Crinoids, with references to literature.)

92. 1897. (North American Crinoidea Camcrata.) Mem. Mus. Harvard,

vol. xx. and xxi. (Contains also chapters on morphology and classifica-
tion of Crinoids in general.) For other writings see under 10.

93. Woodward, H. 1880. (Notes on the Anomalocystidae, etc.) Geol. Mag.,

n.s., dec. ii., vol. vii. pp. 193-201, pi. vi.

94. Warlhcn, A. H., with Meek, F. B. ; Miller, S. A.; Wachsmuth, C. ;

Barns, W. H. ; Springer, F., and others. 1866-90. Palaeontology of
Illinois, in Geol. Survey Illinois, vols. ii., iii., v., vi., vii., and viii.
(Palaeozoic Echinoderma, especially Pelmatozoa, with references to the
numerous preliminary papers in Proc. Acad. Nat. Sci. Philadelphia.
Vol. viii. contains complete Index and a Bibliography of A. H. Worthen,
both by J. Lindahl )

95. ZiUel, K. A. v. 1879. Handbuch der Palaeontologie, I. Band. Palaeo-

zoologie, I. Band, 1 Abth., pp. 308-560, Miinchen. (All fossil Echino-
derma.)

96. 1895. Grundzuge der Palaeontologie (Palaeozoologie), 8vo, viii.

and 971 pp., Miinchen. First portion translated as "Text-Book of
Palaeontology," by C. Jl. Eastman, 1896, with revision for Pelmatozoa by
C. IVachsmuth and others, 8vo, New York. Reissue, 1900, London.

See also Nos. 5, 21, and 36 in Literature of Echinoderma generally (p. 35).

CHAPTER XIII.

ELEUTHEROZOA THE HOLOTHURIOIDEA.

GRADE B. ELEUTHEROZOA.

CLASS I. HOLOTHURIOIDEA.
„ II. STELLEROIDEA.
„ III. ECHINOIDEA.

ECHINODERMA in which the theca, which may be but slightly or
not at all calcified, is not attached by any portion of its surface,
but is usually placed with the oral surface downwards or in the
direction of forward locomotion. Food is not conveyed by a
subvective system of ciliated grooves, but is taken in directly by
the mouth. The anus when present is typically aboral and
approaches the mouth only in a few specialised forms. The aboral
nervous system, if indeed it be present at all, is very slightly
developed. The circumoesophageal water-ring may lose its connec-
tion with the exterior medium ; the podia (absent only in some
exceptional forms) may be locomotor, respiratory, or sensory in
function, but usually are locomotor tube-feet.

As explained on p. 33, the genetic affinity between the three
classes now to be dealt with is not so obvious as that between the
classes of Pelmatozoa. Some writers have taken the Holothurioidea
to be the most primitive class of Echinoderms, or at least to be
widely separated from the Stelleroidea and Echinoidea. Without
denying a large measure of truth to such statements, it is here
maintained that all these three classes bear the impress of a Pelma-
tozoic ancestry. And, though they arose from the Pelmatozoan
stem, probably at different periods, and possibly from different
branches thereof, yet the trend of the evolution of each was in the
same direction — a direction opposed to that of the Pelmatozoa.

At any rate the plain facts set down in the above definition
are conveniently connoted and emphasised by the adoption of the
term Eleutherozoa, whatever be the precise systematic value attached
to it. In discussing the Eleutherozoan classes, the order followed
is from the more primitive to the more specialised.

218 THE HOLOTHURIOIDEA

CLASS I. HOLOTHURIOIDEA, C. T. v. SIEBOLD (1848) 1

( = FISTULIDES, Lamarck, 1801 ; SCYTODERMATA, Burmeister, 1837;
ASCIDIASTELLA, T. & T. Austin, 1840 ; SCYTACTINOTA, Bronn,
1860).

OKDER 1. Actinopoda.
„ 2. Paractinopoda.

Eleutherozoic Echinoderms normally elongate along the oro-anal
axis, which axis and the dorsal hydropore lie in the sagittal plane
of a secondary bilateral symmetry. The calcareous skeleton, which
may be entirely absent, is usually in the form of minute spicules.
sometimes of small irregular plates with no trace of a calycinal or
apical system ; to these is added a ring of pieces radiately arranged
round the oesophagus. Ambulacral appendages take the form of :
(1) circumoral tentacles, (2) sucking-feet, (3) papillae; of these (1)
alone is always present. The gonads are not radiately disposed.

The Holothurians have long been known to man. Many of
the common v/rms are large and conspicuous animals, which are
frequently caught in the dredge or thrown up on our shores. It
is generally supposed that these are the marine animals to which
Aristotle gave the name of oXoOovpiov, from which their present
scientific name has been derived. Pliny, mentioning a species of
Citcumaria, calls it Cucumis marinus, or sea-cucumber, a name often
applied to Holothurians at the present day.

Belon, in 1553, first described a Holothurian, recognising some of
its affinities to the starfish and sea-urchins ; and Rondelet, two years
later, gave some figures of two species. Bohadsch, Pallas, Fabricius,
Cuvier, and others increased our systematic and anatomical know-
ledge of this group during the later part of the eighteenth century.

In 1816 Tiedemann published an excellent account of the
anatomy of Holothuria tubulosa, C4mel. (12), and since then the
study of the structure, development, and classification of the class
has been greatly extended by a large number of naturalists, amongst
whom one may mention A. Baur (1), Selenka (7), Semper (10), Joh.
Miiller, Metschnikoff, and Kowalevsky. Within recent years the
finer anatomy and histology of Holothurians have been studied by
O. Hamann (3), Herouard (4), Cue"not, and many others, and their
embryology by Selenka (7), Semon (9), and Bury (2). Our
knowledge of that most interesting group, the Elpidiidae, dates
almost entirely from the publication by Th4el of the Report on
the collections made during the Challenger Expedition (11). The
family Pelagothuriidae has only lately been described by H.
Ludwig (6) An admirable and comprehensive treatise on the

1 By E. S. Goodrich, M. A.

THE HOLOTHURIOIDEA

219

FIG. I.

l.—Holothuria forskali, D. Ch., opened along the right side of the median "dorsal" line;
the blood-vascular system is not represented ; 2.— Left side view ; and 3.— Anterior view.

4. — Calcareous ring of H. forskali; the anterior ends are turned inwards.

5.— Diagram of the water-vascular system of H. forskali, illustrating the Actinopod plan.

6. — Larva of a Holothurian. (After Bal four.) 7. — Auricularia larva ; and 8. — Barrel-
shaped larva of Synapta digitata. (Both after Semon ; of. Chapter VIII., Figs. III. & XVII.)

9. — Portion of the intestine with the accompanying blood-vessels and rete mirabile of
Holothuria tubidosa. (After Tiedemann.)

10.— Left side view of Ankyroderma affine, Dan. and Kor. ; and

11.— Spicules of the same. (After Danielssen and Koren.)

n, anus ; an, anchor ; a.v, anti-mesenterial blood-vessel ; br, brim ; c.b, ciliated band ; c.c,
circular water- vascular canal ; cl, rectal "cloaca " ; c.o, Cuvierian organs ; c.v, collateral blood-
vessel ; d.mes, dorsal mesentery ; g, gonad ; i, intestine ; i.p, interradial piece ; l.d.r, left dorsal
radius ; l.v.r, left ventral radius ; 'l.r.t, left respiratory tree ; ro, mouth ; md, madreporite ; m.r,
median ventral radius ; m.v, mesenterial blood-vessel ; ot, otocyst ; p, podium ; p.c, podial
canal ; pp, papilla ; p.v, Polian vesicle ; r.cc, radial canal ; rd, spatulaterod ; r.d.r, right dorsal
radius; re, rete mirabile ; r.p, radial piece; r.r.t, right respiratory tree; r.v.r, right ventral
radius; s.c, stone-canal; sp, spicules ; t, tentacle ; t.a, tentacle ampulla ; t.c, tentacle canal ;
x, oesophagus.

220 THE HOLOTHURIOIDEA

structure, development, and taxonomy of the Holothurioidea has
recently been written by the same author (5), to whose works the
present account is greatly indebted.

As a typical example of the Holothurioidea the common Cotton
spinner, Holothuria forskali, Delle Chiaje (H. nigra, auctt.), may be
taken. It is about 20 cm. long, almost cylindrical in shape (Fig. I. 2),
pale yellowish -brown beneath, and black above. The "ventral"
or lower surface is covered with closely set retractile podia, by
means of which the animal creeps along (2, p) ; the lateral and
" dorsal " surfaces are covered with small and large conical papillae
(2, pp). The five "radii" running longitudinally from mouth
to anus are scarcely distinguishable on the outer surface; there
are two dorsal radii (bivium), and one median ventral, and two
lateral ventral radii (trivium). Near the anterior end the body-
wall is produced into an irregular brim (2, br) ; beyond this the
animal terminates in a smooth pale-coloured area, in the centre of
which lies the mouth, somewhat ventral in position (2 and 3, m).
Surrounding the mouth is a set of twenty semi-retractile tentacles
(2 and 3, /). When in a state of contraction the brim closes
over the retracted tentacles and mouth. The anus is situated at
the posterior pole of the animal.

The body -wall of Holothuria is thick, tough, and leathery.
Externally a thin transparent cuticle covers the epidermis, which is
not ciliated, and is composed of a layer of columnar cells, with
scattered gland cells and sensory cells. The thick underlying cutis
is formed of connective tissue cells and numerous fibres lying in a
homogeneous ground substance. "Wandering cells are found in the
cutis, while the greater part of the pigment is in its outermost layer.
Below the cutis is the layer of circular muscles (Fig. II. 5, c.m) l
interrupted at the radii, except immediately round the anus, where
it forms a sphincter muscle. A paired band of longitudinal muscles
runs along each radius from the posterior end of the animal to
the anterior (Fig. II. 1 & 5, Lin) ; here it is attached to the radial
plate of the calcareous ring to be described below. Internally the
body-wall is lined with ciliated coelomic epithelium.

The skeleton of Holothuria forskali consists chiefly of calcareous
spicules, knotted or branching rods, and perforated discs (Fig. II.
6, A, B\ secreted by connective tissue cells and forming a thin
layer in the outer region of the cutis, especially on the papillae and
ambulacral appendages (podia and tentacles). Spicules may also be
found in the connective tissue throughout the body, and they are
very numerous in the wall of the stone -canal and madreporite. A
large perforated plate is situated at the extremity of each podium.
A ring of ten calcareous pieces surrounds the oesophagus; the five
radial pieces are notched and larger than the interradial (Fig. I. 4).
1 This figure refers to the allied species Holothuria tubulosa, Gmeliu.

THE HOLOTHURIOIDEA 221

The nervous system consists of a " superficial " ring surrounding the
mouth and giving off five radial nerves which run backwards to the
posterior end. The tentacles and viscera are supplied by nerves
from the ring ; the podia by branches from the radial nerves. The
ring and the radial nerves are sunk below the surface, and lie on
the inner side of the cutis. An epineural canal lies outside the
radial nerves. Following each radial nerve on its inner surface is
a small " deep " nerve, from which are supplied the muscles of the
body-wall (Fig. II. 4).

The water -vascular system consists of a circular canal sur-
rounding the oesophagus behind the calcareous ring, and giving
off five radial canals (Fig. II. 5, c.c and r.c). Each radial canal
passes forward between the calcareous ring and oesophagus, and
then outwards, through the notch in the radial piece, to the body-
wall ; it then runs backwards below the radial nerve. The radial
canals send off a branch on either side to supply a pair of tentacles,
each tentacle being provided with a long ampulla freely projecting
into the body - cavity (Fig. I. 5, t.a., and Fig. II. 5). A Polian
vesicle is attached to the circular canal in the left ventral in-
terradius. Branches are given off from the radial canals in the
body - wall to supply the podia and papillae ; small ampullae are
here present. A twisted stone - canal lying in the median dorsal
line leads from the circular canal to a madreporite, pierced with
many small apertures, and lying in the body-cavity at the anterior
edge of the dorsal mesentery (Fig. I. 1 and 5, s.c and ind).

The alimentary canal is looped (Fig. I. 1, i, and Fig. II. 1),
coiled in the direction of the watch-hand as viewed from the anterior
end, and supported almost throughout by a mesentery attached to the
body-wall. The mouth leads into a wide oesophagus, which narrows
on passing backwards out of the calcareous ring. A scarcely dis-
tinguishable stomach succeeds the oesophagus and passes into the
long intestine, which finally ends in a rectal " cloaca " opening at
the anus. The oesophagus, stomach, and part of the intestine
form the first region of the alimentary canal, running backwards
and supported by the median dorsal mesentery. Near the hinder
end of the body the intestine and mesentery cross over from the
dorsal interradius to the left lateral dorsal interradius, up which
they run. Near the anterior end this second section of the
intestine, with its mesentery, crosses to the right ventral inter-
radius, down which the third and last portion of the intestine runs
straight to the anus. Round the oesophagus, bounded externally
by the calcareous ring and the radial canals, a portion of the coelom
is incompletely shut off from the general body -cavity (Fig. II.
5, pr) ; connective tissue strands run across it from one wall
to the other. The enlarged rectum, or cloaca, is fastened on all
sides to the borlv-wall by muscular strands (Fig. I. 1, d).

222

THE HOLOTHURIOIDEA

Opening into the anterior and dorsal region of the cloaca by a
common duct are the respiratory trees, two large tubes which give

j^^rii: ^

Fia. II.

1.— Transverse section of one of the Holothurinae. (Modified from Liuhvig and Lang.)

2. — Transverse section of the body-wall of Holothuria impaticns, Forskal.

8.— (.4) "table " and (h) " biscuit " spicnle of the same.

4. — Transverse section of the radius of an Actinopod Holothurian. (After Lang.)

5.— Left side view of the anterior region of Holothuria tubitlosa, Gmel., from which the
loft side of the body-wall has been removed.

0.— (A) perforated disc, and (B) branching rod spicule of Ilolothitriu forskdli.

a, ampulla ; a-.v, anti-mesenterial blood-vessel ; b, " biscuit " spicule ; /*.r, blood-vessel ; b.n',
body-wall ; c, coelom ; car, calcareous ring ; c.c, circular canal ; c.ep, coelomic epithelium ; c.i»,
circular muscles ; r.p, podial canal ; ct, cutis ; d.mes, dorsal mesentery ; d.n, deep nerve ; tp,
epidermis ; ep.c, epineural canal ; g, gonad ; g.d, genital duct ; g.p, genital pore ; i, intestine ;
l.d.r, lea dorsal radius ; l.m, longitudinal muscles ; l.mes, left mesenterial ; l.r.t, left respira-
tory tree ; l.v.r, left ventral radius ; wrf, madreporite ; m.r, median ventral radius ; m.r, mesen-
terial blood-vessel ; p, podium ; p.c, pseudhaemal canal ; p.n, podial nerve ; pr, peripharyngo.il
coelom ; p.v, polian vesicle ; r.c, radial canal ; r.rf.r, right dorsal radius ; r.mes, right
fold ;
(.a,

fold ; r.r.t, right respiratory tree ; r.v.r, right ventral radius ; s.n, superficial nerve; t, tentacle ;
(.a, tentacle ampulla; ta, "table" spicule in 2.

off numerous side - branches ending blindly in tufts of slender
twigs (Fig. I. 1, l.r.t and r.r.t). Coming off from the base of the

THE HOLOTHURIOIDEA 223

left respiratory tree are a large number of long slender pro-
cesses, the Cuvierian organs (Fig. I. 1, c.o). These remarkable
organs, formed of spirally coiled fibrous tissue covered with a
layer of modified coelomic epithelium which secretes a sticky
substance, are shot out of the anus through the wall of the cloaca
by the living animal when irritated. The organs swiftly elongate,
forming adhesive white threads, to which this Holothurian owes its
name of Cotton-spinner.

The blood-vascular system consists of a circular vessel round
the oesophagus, giving off' five radial vessels which run between
the water-vascular canal and the pseudhaemal canal, covering the
radial nerve internally. The anterior region and genital organs
are supplied from the circular vessel. The alimentary canal is
provided with two longitudinal trunks coming from the circular
vessel — one, the "dorsal" or mesenterial vessel, runs along the
region where the mesentery is attached to the intestine ; the other,
the " ventral " or antimesenterial, runs along the alimentary canal
on the opposite side (Fig. II. 1 and 5, m.v and a.v). A cross
vessel passes from the mesenterial vessel on the first region of
the intestine to the antimesenterial vessel on the second region.
In connection with the mesenterial vessel (especially along the
second region of the intestine) is an extensive and rich plexus of
blood-vessels, the rete mirabile, overlying the left respiratory tree
(Fig. I. 9). Blood lacunae extend in the walls of the alimentary
canal, respiratory trees, gonads, and other parts.

A closed pseudhaemal system extends over the inner surface
of the nerve ring and radial nerves.

The body -cavity forms a continuous coelomic space lined
throughout by epithelium, which is generally ciliated.

The sexes are separate in this species. The genital organs consist
of a bunch of tubes, with free blind ends lying in the coelom, and
uniting at their base to open into a duct running forwards in the
dorsal mesentery (Fig. II. 1 and 5, g and g.d). The genital duct
opens by a median dorsal pore behind the tentacles (5, g.-p).

Variation in the Holothurioidea. — In outward form the
more specialised and highly modified genera depart very widely
from the elongated and somewhat pentagonal shape which appears
to be the more primitive. In organisation, however, this group is
fairly constant, and clearly denned from the other classes of Echino-
derma. The epidermis is not ciliated in the adult. The body-wall
is remarkable for its thickness and leathery consistency, the cal-
careous skeleton being rarely in the form of plates or scales, and
more often as minute spicules of various shapes (there is no apical
system of plates). When these spicules are of different kinds they
generally form distinct layers in the cutis — as, for instance, in
Holothuria impatiens, where the " tables " lie on the surface, whilst

224 THE HOLOTHURIOIDEA

the "biscuit" spicules are disposed in a thick inner layer
(Fig. II. 2 and 3, ta, b). The spicules are deposited by cells in
the cutis, and develop first as small rods which become branched
at each end ; by the increase of the number of branches and their
repeated fusion arise the innumerable varieties of spicules found in
different species of Holothurians. These calcareous bodies are of
great value to the systematist in classifying the smaller groups,
such as genera and species.1 Although their general characteristics
are fairly similar within the several families, the different shapes
of spicules are not sufficiently constant to be used as diagnostic
characters of such large divisions.

Very characteristic of the Holothurians is the calcareous ring
formed of radial and interradial pieces surrounding the oesophagus.
Occupying the same position as " Aristotle's lantern " in the Echi-
noids, it may possibly be homologous with that organ.

A well-developed muscular system is present in the body-wall,
whereby the animal can move and alter its shape. The longitudinal
muscles, generally paired radial bands, often form special retractors
for the withdrawal of the anterior region of the body. In the
Synaptidae alone the circular muscles are not interrupted at the radii.

Although the nervous and water-vascular systems are distinctly
built according to the pentagonal radiate Echinoderm type, yet the
latter system is generally modified in relation to the very frequent
differentiation of a dorsal surface occupying the three upper interradii,
and of a flattened ventral surface or creeping sole occupying the
two lower interradii. The ambulacral appendages, the podia, may
become modified from typical sucking-feet into pointed papillae.
Either the tube-feet or the papillae may extend over the entire inter-
radial space. On the other hand, both tube-feet and papillae may be
absent (Molpadiidae, Pelagothuriidae, and Synaptidae), and even the
radial canals may disappear in the adult (Synaptidae). The oral
tentacles, so characteristic of the Holothuiians, are no doubt modified
ambulacral appendages homologous with the podia. They vary
greatly in number and shape, and are of great taxonomic value,
being almost invariably more or less peltate in the Holothuriidae and
Elpidiidae, arborescent in the Cucumariidae, and digitate or pennate
in the Molpadiidae and Synaptidae.

The Polian vesicle is usually single, and situated in the left
ventral interradius ; but there are sometimes more than one. The
stone-canal, generally single, lies in the median dorsal mesentery.
Although in some of the Holothuriidae, Elpidiidae, and Pelago-
thuriidae, and in the larvae of other forms, the stone- canal retains
its opening to the exterior in the median dorsal line, in most Holo-
thurians this condition is modified, in that the primitive madreporite

1 The structure of the spicules is liable to alter during the lifetime of the
individual (Herouard, Mitsukuri, Ostergren).

THE HOLOTHURIOIDEA 225

disappears, the connection with the body-wall is lost, and the canal
opens by a new madreporite into the body-cavity. In some species
numerous accessory stone-canals and madreporites may be developed
(Fig. II. 5, md).

The disposition of the organs in the radius, as seen in transverse
section, is distinctive (Fig. II. 4). The superficial radial nerve is
separated from the epidermis by the thick cutis and a space
(epineural canal). The deep radial nerve is separated from the
ambulacral radial canal by a pseudhaemal canal and the radial
blood-vessel. Over all lies the longitudinal muscle internally. This
arrangement resembles most that found in the Echinoidea.

The alimentary canal and the mesentery which supports it have
the dextral coil characteristic of Echinoderma, as described above
for Holotlmria (p. 221).

The respiratory trees, organs quite peculiar to this class, are by
no means of universal occurrence, being absent in the Elpidiidae,
Pelagothuriidae, and Synaptidae. It is interesting to note, however,
that in certain of the Elpidiidae the rectum is provided with a
caecum, which may represent a vestige or a rudiment of the respira-
tory trees (Fig. III. 10, coe).

The Cuvierian organs (p. 223) appear to be modified branches
of the respiratory trees.

The Holothurioidea are distinguished from the remainder of the
Echinoderma by the structure of the genital organs. These always
consist of a single, or of a right and left, tuft of tubules leading into
a common duct, which runs in the dorsal mesentery and opens to
the exterior in the median dorsal line near the anterior extremity
of the body. There is no axial organ or sinus, and no trace of
radial structure in connection with the gonads (see p. 24).

As a rule, the genital products are shed in the sea, where fertilisa-
tion and development take place. Rarely, as in Chiridota rotifera
(Pourt.) and Pht/llophorus urna (Grube), the young develop in the
body-cavity of the parent. Brood chambers are formed in Psolus
ephippifer (W. Thomson) and some species of Cucumaria.

A total and almost equal segmentation of the fertilised egg leads
to the formation of a ciliated blastula and gastrula, from which is
developed the characteristic Auricularia larva (Fig. I. 6, 7). The
free-swimming Auricularia has an alimentary canal provided with a
mouth and anus, and the cilia are restricted to a single large
circumoral band and a small adoral band within the mouth. The
archenteron now gives rise to the hydro-enterocoel, opening by the
primary stone- canal at a pore a little to the left of the median
dorsal line. Later the circumoral ciliated band becomes greatly
folded, and then converted into the circular ciliated rings of the
barrel-shaped larva or pupal stage (Fig. I. 8). The mouth shifts
to the anterior pole, round which are developed the tentacles as the

15

226 THE HOLOTHURIOIDEA

first appendages of the water-vascular system. The radial canals
and podia (when present) are now formed, and the young Holo-
thurian assumes the adult form. Rarely the Auricularia stage is
omitted, the ciliated gastrula developing more or less directly into
the barrel-shaped larva.

Very little is known of the extinct Holothurians. Some spicules
have been found, many of which belong to the Synaptidae, in
deposits ranging from the Carboniferous to Tertiary strata.

With advancing knowledge the Classification of the Holothurians
has undergone many changes since it was first attempted at the
beginning of this century. It now appears to be firmly established
on deep-seated structural characters. In 1815 Oken divided the
few species then known according to the shape of the body, whilst
soon after Lamarck made use of the tentacles as a taxonomic
character, a system afterwards perfected by Grube, who founded
the family Aspidochirotae for forms with peltate tentacles (Holo-
thuriidae), and Dendrochirotae for forms with arborescent tentacles
(Cucumariidae). Cuvier and others followed Okcn, taking into
account the occurrence and distribution of the podia ; and Brandt
divided the Holothuria into Pedatae with podia, and Apodes
without podia. Selenka, in 1867, following Jaeger, formed the two
orders Pneumophora and Apneumona, the first for the modern
Holothuriidae, Cucumariidae, and Molpadiidae, provided with re-
spiratory trees or " lungs," the second for the Synaptidae without
" lungs." The"el adopted the orders Apoda and Pedata, adding the
order Elasipoda for the newly discovered Elpidiidae. Ludwig has
clearty shown that the classifications founded on the mere presence
or absence of podia or of respiratory trees are artificial; first,
separating off the Synaptidae, which differ in important respects
from all the other families, as the Paractinopoda, he divides the
remainder of the Holothurians, the Actinopoda, into five families,
as shown in the following table : —

ORDER 1. Actinopoda. Radial canals sup-
plying tentacles and podia.

A. With respiratory trees.

r \ w-fi, A- /FAM. 1. HOLOTHURIIDAE.

(a) With podia. |FAM 4 CUCUMARIIDAE.

(6) Without podia FAM. 5. MOLPADIIDAE.

B. Without respiratory trees,

(a) With podia. FAM. 2. ELPIDIIDAE.

(6) Without podia. FAM. 3. PELAGOTHURIIDAE.

ORDER 2. Paractinopoda. Neither radial
canals nor podia. Tentacles
supplied from circular canal.

FAM. SYNAPTIDAE.

THE HOLOTHURIOIDEA 227

ORDER l. Actinopoda, Ludwig.

The five radial canals of the water-vascular system, springing from
the circular canal, supply the tentacles and podia (Fig. I. 5).

FAMILY 1. HOLOTHURIIDAE. Body more or less flattened ventrally
to form a creeping sole. Mouth generally somewhat ventral in position.
Spicules chiefly in the form of tables, rods, and perforated plates. Cal-
careous ring of five radial and five interradial pieces. Ambulacral
appendages present in the shape of tube-feet, papillae, and peltate tentacles
(generally twenty). Kespiratory trees well developed. SUB -FAMILY 1.
SYNALLACTINAE. No tentacular ampullae. Stone-canal single, and joined
to the body-wall. No rete mirabile. Genera — A. With genital tubes in
a right and left bundle — Pseudostichopus, Theel ; Paelogatides, Theel ;
Meseres, Ludwig; Synallactes, Ludwig; Bathyplotes, Ostergren. B.
Genital tubes in a tuft on the left side only — Mesothuria, Ludwig.
SUB-FAMILY 2. HOLOTHURIINAE. Well-developed tentacular ampullae.
Stone - canals often numerous, and with complex madreporites not con-
nected with the body-wall (exc. Holothuria lactea). The left respiratory
tree is enveloped in a vascular network, the rete mirabile. Cuvier's
organs often present. Genera — A. With genital tubes in a tuft on the
left side only — Muclleria, Jager, mostly tropical ; Labidorlema?, Selenka,
tropical ; Holothuria, Linn., to which genus belong the British Cotton-
spinner described above (p. 220), and H. edulis, one of the kinds of edible
Trepang. B. With genital tubes in a right and left tuft — Stichopus,
Brandt.

The Holothuriidae are characterised not only by the possession of well-
developed respiratory trees, and of from fifteen to thirty peltate tentacles
(formed of a thick stem branching at its extremity), but also by the
absence of retractor muscles, and of auditory vesicles. The tube-feet and
papillae may be in double radial rows, as in Labidodemas, in several rows,
as in Stichopus, or scattered, as in Holothuria. The calcareous ring is
generally composed of short, compact pieces ; it appears to be absent in
Paelopatides and Synallactes. The anus in Pseudostichopus is situated in
a deep vertical furrow, whilst in Muelleria it is surrounded by five
pointed, calcareous plates. The longitudinal muscles are in paired bands,
except in Pseudostichopus.

The Holothuriinae are a well-differentiated group distinguished by
the possession of large tentacular ampullae projecting freely into the
body-cavity (Fig. II. 5, t.a), internal madreporites (5, md), and a vascular
plexus surrounding the left respiratory tree. The Synallactinae, on the
other hand, are in many respects intermediate between the former
sub -family and the Elpidiidae. The absence of tentacular ampullae,
and of the rete mirabile, and the presence of a single stone-canal con-
nected with the body- wall and probably retaining its primitive opening
to the exterior, are all characters uniting the Synallactinae to the
Elpidiidae, with which they have indeed been classified (Ostergren).

FAMILY 2. ELPIDIIDAE. Body generally flattened ventrally. Mouth
more or less ventral. Tentacles, ten to twenty, more or less peltate. Ventral

228

THE HOLOTHURIOIDEA

tube-feet and dorsal papillae present (exc. Capheira). Often auditory
vesicles on the radial nerves. Stone-canal single ; madreporite frequently
opening to the exterior. Respiratory trees absent (or rudimentary).

Fio. III.

1.— Left side view of Deima validum, Theel. (After Theel.)

2.—Ilyodaemon maculatus, Th. (After Theel.)

3. — (A) C-shaped spicule of Scotoplanes albuia, Th. ; (B) wheel of Pannychia moscleyi, Th.;
(C) four-armed spicule of Elpidia rirjida, Th. (After Theel.)

4.—Peniagone wyvillii, Th. (After Theel.)

5.—Psychropotes longicanda, Th. (After Theel.)

6.— Ventral view of ElpUlta glacial is, Th. (After Thlel.)

7 and 8.— Side view and oral view of Pelagothuria natatrix, Ludw. (Modified from Lndwig.)

9. — Section through the madreporite and genital papilla of Ilyodaemon maculatus, Th.
(After Theel.)

10. — Portion of the intestine and rectal "cloaca," with its caecum, of Benthodytes
sanguinolenta, Th. (After Ludwig.)

a, anus ; br, brim ; b.w, body-wall ; cl, rectal " cloaca" ; coe, caecum ; d, disc ; d.a, dorsal
appendage ; g.d, genital duct ; g.h, genital aperture ; g.p, genital papilla ; i, intestine ; m,
mouth ; op, opening of pore canal ; p, tube-foot ; p.c, pore canal ; pp, papilla ; pr, anterior
procejss ; r, ray of the disc ; s, creeping sole ; s.c, stone canal ; t, tentacle.

Calcareous ring of five or ten pieces. Spicules various, frequently in the
form of wheels and four-rayed spicules. SUB-FAMILY 1. PSYCHROPOTINAE.
Body surrounded by a brim. Mouth quite ventral Dorsal papillae
usually small Tube-feet small, as a rule on the three ventral radii.
Calcareous ring incomplete, the interradial pieces being represented by

THE HOLOTHUR10WEA 229

small spicules. Genital tubes in a right and left tuft. Genera — Psychro-
trephes, Theel ; Euphronides, Theel ; Psychropotes, Tlieel, Fig. III. 5 ;
Benthodytes, Tlieel. SUB-FAMILY 2. DEIMATINAE. Moutli sub - ventral.
Papillae large and numerous. Tube-feet large, generally only on the lateral
ventral radii. Calcareous ring of five radial and five interradial pieces.
Genital tubes in a right and left tuft. Genera — A. With a row of large
papillae above the lateral ventral podia. Median ventral podia rudimentary
or absent Twenty tentacles — Deima, Theel (Fig. III. 1) ; Oneirophanta,
Theel ; Orphnurgus, Theel ; Scotodeima, Ludwig. B. Without a distinct
row of lateral ventral papillae, fifteen to twenty tentacles. Wheel
spicules — Laetmogone, Tlieel ; Ilyodaemon, Theel (Fig. III. 2 ; Pan-
nychia, Theel ; Capheira, Ludwig. SUB -FAMILY 3. ELPIDIINAE. Mouth
generally sub-ventral. Papillae usually few and large. Podia only on
the lateral ventral radii. Calcareous ring of five radial pieces. Genital
tubes in one or two tufts. Genera — A. With ten tentacles — Parelpidia,
Theel ; Elpidia, Theel (Fig. III. 6) ; Scotoplanes, Theel ; Kolga, Dan.
and Koren ; Irpa, Dan. and Koren ; Peniayone, Theel (Fig. III. 4) ;
Scotoanassa, Theel. B. With more than ten tentacles — Achlyonice,
Theel ; Enypniastes, Theel.

The Elpidiidae are a deep-sea group of wonderfully diverse outward
form. The body is generally flat or even concave ventrally ; it is often
produced along its lateral edge into a brim, which may be posterior (as in
Scotoplanes), anterior (as in Elpidia purpurea, Theel), both anterior and
posterior (as in Scotoanassa), or all round (as in the Psychropotinae).
The mouth may be quite ventral, and some way behind the anterior
edge of the body (Fig. III. 5). The anus is terminal, dorsal, or ventral.
Tube-feet occur on the ventral radii only, and are often remarkable for
their large size and paired arrangement (Fig. III. 6), resembling the feet
of a segmented animal. The dorsal papillae also may be distinctly
paired and very large (Fig. III. 1). Peculiar posterior dorsal appendages,
sometimes of huge size (Fig. III. 5), are developed in Euphronides and
Psychropotes. Somewhat similar anterior appendages occur in Peniagone
(Fig. III. 4). These outgrowths of the body-wall are often supplied with
right and left canals from the radial, water -vascular system. The most
characteristic spicules are C -shaped, four -armed wheels (Fig. III. 3),
and perforated plates. The calcareous ring is generally but slightly
developed ; in the Elpidiinae the five radial pieces only are represented
as single-branched spicules. Auditory vesicles or otocysts are sometimes
situated in the Elpidiinae along the circular and radial nerves, but chiefly
along the lateral ventral radii. Of great interest is the relation of the
stone-canal to the body-wall in this family. The primitive condition in
which it opens by a single pore in front of the genital aperture is found
in Elpidia (some species), Kolya, and other genera. In some forms, e.g.
Laetmoyone and Ilyodaemon (Fig. III. 9), it opens by a number of pores to
the exterior. The external opening has been lost in Irpa, Benthodytes,
and others, the canal opening to the coelom, but being still connected
with the body-wall. There are no retractor muscles, and no Cuvierian
organs. Although the Elpidiidae differ from the preceding family by the
generally simple, unpaired structure of the longitudinal muscles (p. 224),

230 THE HOLOTHURIOIDEA

and the absenco of well-developed respiratory trees (pp. 222, 225), yet
they are undoubtedly closely related to them, and more especially to
the Synallactinae,

FAMILY 3. PELAGOTHURIIDAE. Body cylindrical, and produced at the
base of the crown of tentacles into an umbrella-like disc, drawn out
into long slender rays. Mouth and anus terminal. No podia ; branches
of the tentacle canals extending into the rays of the disc. No retractors
and no respiratory trees ; a single stone-canal opening to the exterior.
Right and left tufts of genital tubes. No calcareous skeleton. Genus —
Pelagothuria, Ludwig (Fig. III. 7 and 8), represented only by a remarkable
free-swimming Holothurian, Pelagothuria natulrix, recently discovered in
the Pacific. The thirteen to sixteen tentacles are forked and beset with
papillae, and the disc is produced into a corresponding number of rays,
each containing a branch of the tentacular canal. These rays may repre-
sent the modified tentacular ampullae of the Holothuriidae ; or the disc
may be derived from the anterior brim, which occurs in the two preceding
families. The five radial vessels are normally developed, in spite of the
absence of podia. The longiuidinal muscles are in simple bands. Neither
calcareous ring nor calcareous spicules are developed.

This highly modified form undoubtedly belongs to the Actinopoda,
since there are radial canals, and the circular muscles are interrupted
at the radii. The absence of respiratory trees, and of free tentacular
ampullae, and the simple longitudinal muscles remove it from the Mol-
padidae ; whilst the first of these characters, combined with the absence
of podia and retractors, separate it from the Cucumariidae. With the
Elpidiidae, on the other hand, it has many characters in common, such
as the single stone-canal opening to the exterior, the simple longitudinal
muscles, the absence of respiratory trees, and the reduction of free ten-
tacular ampullae and the calcareous ring. Pelagothuria, therefore, is
probably a free-swimming form derived from an Elpidiid ancestor.

FAMILY 4. CUCUMARIIDAE. Podia generally tube-feet only and no
papillae. Mouth and anus terminal or dorsal. Tentacles eight to thirty,
branched. Tentacle ampullae rudimentary or absent. Madreporite in-
ternal Calcareous ring of five radial and five interradial pieces. Re-
tractor muscles. Longitudinal muscles generally simple. Respiratory
trees well developed. Cuvierian organs rare. Right and left tufts of
genital tubes. Spicules chiefly rods and knobbed, perforated plates.
Genera — A. With distinct creeping sole. Ventral podia restricted to
radii — Colochirus, Troschel ; Psolidium, Ludwig; Theelia, Ludwig
(Fig. IV. 10); Psolus, Oken (Fig. IV. 8). B. Without distinct creeping
sole — Thyone, Oken (Fig. IV. 1 1) ; Orcula, Troschel ; Phyllophorus, Grube ;
all with scattered podia (generally). With podia more or less restricted
to the radii ; Cucumaria, Blainville (Fig. IV. 1, 2, 3, 4, and 5 ; Pseudo-
cucumis, Ludwig ; A ctinocucumis, Ludwig; Echinocucumis, Sars; Sph^.ero-
thuria, Ludwig (Fig. IV. 9). C. Flask-shaped, with mouth and anus close
together — Rhopalodina, Gray (Fig. IV. 7).

The Cucumariidae are distinguished by the possession of delicate re-
tractile arborescent tentacles (Fig. IV. 3 and 4). Frequently the two
median ventral tentacles are smaller than the others (Fig. IV 3), as in species

THE HOLOTHURIOIDEA

231

KfTL

FlO. IV.

1. — Cucumarla pentactes (Linn.), opened along the right of the median dorsal line ; the
right respiratory tree has been cut short. 2.— Spicule ; 3.— Oral view ; and 4.— Left side view
of the same.

5.— Inner view of a retracted C. pcntactes, from which the dorsal region of the body-wall
has been removed, and the viscera extracted, to show the action of the retractor muscles.

0. — Portion of the calcareous ring of Oraila tenera, Liidw. (After Ludwig.)

7. — Diagrammatic drawing of Rhopalodina lageniformis, Gray. (After Lang.)

8.—" Dorsal " view of Psolas antarctinis (Philippi). (After Theel.)

9.— Left side view of Sphaerothuria Mtentaculata, Ludw. (After Ludwig.)

10.— Theelia ambulatrix, Bell.

11.— Thyone fiisus (O. F. Mitller).

a, anus ; o. v, anti-mesenterial blood-vessel ; b.w, body- wall ; ca.r, calcareous ring ; c.c, circular
water-vascular canal ; cl, cloaca ; g, gonad ; g.d, genital duct ; g.p, genital pore ; t, intestine ;
i.p, interradial piece ; l.d.r, left dorsal radius ; l.r.t, left respiratory tree ; l.v.r, left ventral
radius ; m, mouth ; m.r, median ventral radius ; m.v, mesenterial blood-vessel ; p, podium ;
p.v, polian vesicle ; r.c, radial canal ; r.d.r, right dorsal radius ; r.m retractor muscle ; r.p,
radial piece ; r.r.t, right respiratory tree ; s. creeping sole ; s.c, stone canal ; t, tentacle.

232 THE HOLOTHURIOIDEA

of Cucumaria and Psolus ; in other cases several tentacles may be small,
and these may even form an inner ring surrounded by an outer ring
of large tentacles, as in Phyllophorus and Pseudocucumis. The stone-
canal may be single or multiple. Five powerful retractor muscles, reach-
ing from the body -wall to the radial pieces of the calcareous ring
(Fig. IV. 1 and 5, r.m), serve to invaginate the anterior region of the body
so as to withdraw the mouth and tentacles out of sight (Fig. IV. 5). The
body-wall in the anterior region is thinner and often devoid of podia
(Fig. IV. 1 1). Every stage is exhibited between the arrangement of the
podia in a double row along the radii, as in Cucumaria pentactcs
(Fig. IV. 4), and their distribution over the interradial areas, as in Thyone
fusus (Fig. IV. 11). In Theelia they are nearly, and in Psolus entirely,
reduced on the "dorsal" surface. This is correlated with the develop-
ment of a very distinct creeping sole, to which the podia are restricted
(Fig. IV. 10 and 8). In these genera the calcareous deposits in the dorsal
region form large plates or scales ; five triangular so-called " oral " plates
close over the introverted anterior region, and small plates may surround
the anus. In Colochirus and Actinocucumis somewhat similar valves are
developed in front. Although the spicules may be in the form of large
perforated plates, sometimes produced into spines, as in Sphaerothuria
bitentaculata, Ludwig (Fig. IV. 9), they are more usually rods or knobbed
buttons and plates (Fig. IV. 2).

The calcareous ring in the Cucumariidae is very well developed, the
pieces being large, and frequently made up of several plates fitting
together. The radial segments, to which the retractors are attached, are
generally produced backwards into two long processes (Fig. IV. 6). A
more or less pronounced bilateral symmetry is often brought about by
the unequal development or fusion of the pieces. Cuvierian organs
have been described in Cucumaria frondosa and C. nigricans. The genital
tufts are paired, and load into a duct which usually opens to the exterior
on a papilla within the circlet of tentacles (Fig. IV. 1 and 3, g.p).

In Sphaerothuria (Fig. IV. 9) the dorsal surface is reduced, the
mouth and anus being approximated and the ventral radii much curved ;
this process is carried to an extreme in the extraordinary genus Rho-
palodina (Fig. IV. 7), where the body has assumed a flask-shape, the
mouth and anus are close together at the small end, while the radii are
bent round. This appearance misled the early observers, who described
Rhopalodina as having ten radii.

Whilst the Cucumariidae resemble the Holothuriidae in the possession
of well-developed respiratory trees, they differ markedly from all the
preceding families in the shape of the tentacles and the presence of
retractor muscles.

FAMILY 5. MOLPADIIDAE. Neither tube-feet nor papillae. The
posterior region generally tapering. Mouth and anus terminal. Generally
fifteen simple or digitate tentacles. Tentacle ampullae well developed.
Calcareous ring of five radial and five interradial pieces. Single stone-
canal with an internal madreporite. Longitudinal muscle bands more
or less paired. Respiratory trees present; Cuvierian organs rare.
Genital tubes in right and left tufta Genera — A. With well-developed

THE HOLOTHURIOIDEA

233

retractor muscles ; tentacles digitate — Molpadia, Cuvier. B. Retractors

absent or rudimentary — Eupyrgus, Liitken ; Haplodactyla, Grube ;

Caudina, Stimpson ; Trochostoma, Dan. and Kor. ; Ankyroderma, Dan. and
Kor. (Fig. I. 10).

vim-

1.— Left side view of Synapta digitata, Mont.

2.— The same opened up to the right of the median " dorsal " line.

3. — Diagram of the water-vascular system of S. digitata illustrating the Paractinopod plan.

4.— Ciliated urns; and

5. — Transverse section through the radius and otocysts of S. digitata. (After Cuenot.)

(>. — Anchor and anchor-plate of .S'. digitata.

7. — Wheel of Myriotrochus Rinkii, Steenstrup. (After Danielssen and Koren.)

<r, anus ; on, anchor ; an.p, anchor-plate ; b.?r, body-wall ; c.c, circular canal ; c.f, coelomic
epithelium ; c.m, circular muscles ; c.u, ciliated urn ; d.mcs, dorsal mesentery ; e.s, epineural
sinus ; g.p, genital pore ; i, intestine ; l.g, left gonad ; m, mouth ; mes, mesentery ; n, nerve ;
o.c, otocyst ; oe, oesophagus ; o.l, otolith ; ps.c, pseudhaemal canal ; r, rectum ; r.g, right gonad ;
r.mcs, right mesenterial fold ; r.n, radial nerve ; s.c, stone-canal ; st, stomach ; t, tentacle ; t.c,
tentacle canal ; v.l.m, median ventral longitudinal muscle.

The Molpadiidae are rounded forms, with generally a tapering, tail-
like, posterior end. The tentacles are simple processes, or of lobed or
digitate shape. Their ampullae are large and free, except in Eupyrgus (T).
There are no podia, unless they be represented by the anal papillae. As

234 THE HOLOTHURIOIDEA

in the preceding family, the calcareous ring is well developed, often
bilaterally symmetrical, and with radial pieces strongly forked behind.
In Trochostoma and Ankyroderma the stone-canal is still connected with
the body -wall Cuvierian organs have been described in Molpadia
chilensis, J. MiilL The calcareous spicules are very similar to those of
the Cucumariidae, except in Ankyroderma, which has peculiar anchor-like
spicules (Fig. I. 11), somewhat resembling those found in Synapta
(Fig. V. 6). Trochostoma has red calcareous deposits.

The presence of retractor muscles, the structure of the calcareous
ring, and the general anatomy of the Molpadiidae indicate their close
relationship to the Cucumariidae.

ORDER 2. Paractinopoda, Ludwig.

The five radial canals have disappeared in the adult. There are no
tube-feet or papillae, and the tentacles are supplied directly from the
circular canal (Fig. V. 3).

FAMILY, SYNAPTIDAE. Body cylindrical and elongated. Mouth and
anus terminal. Tentacles pennate or digitate. Tentacle ampullae small
or rudimentary. Calcareous ving of five radial, and frequently more
than five interradial, pi^es. One or more internal madreporites.
Auditory vesicles on the radial nerves. Circular muscles uninterrupted
at the radii. Longitudinal muscles as a rule impaired. Neither respira-
tory trees nor Cuvierian organs. Ciliated funnels on the coelomic epi-
thelium. Genital tubes in right and left tufts, often hermaphrodite.
Calcareous spicules as wheels, anchors, etc. Genera — A. Spicules various ;
no wheels in the adult. Synapta, Eschscholtz (Fig. V. 1) ; Anapta,
Semper. B. With wheels — Chiridota, Eschscholtz ; Trochodota, Ludwig ;
Trochoderma, Theel ; Myriotrochus, Steenstrup ; Acanthotrochus, Danielssen
& Koren.

The tentacles of the Synaptidae vary in number from ten to twenty-
five. When they exceed ten in number there is usually a corresponding
increase in the number of the interradial pieces of the calcareous ring.
The radial pieces are often pierced anteriorly. There are some very
characteristic spicules, such as the many-spoked "wheels" (Fig. V. 7)
of some genera (and in the larva of Synapta), and the anchors and
anchor-plates in Synapta (Fig. V. 6, an, an.p). The perforated anchor-
plates lie in the cutis parallel with the surface, whilst the "shank"
of the anchors rest against them, the "arms" projecting towards the
surface. They aid locomotion (Ostergren, 1897). A pair of auditory
vesicles, or otocysts, has been found at the base of the five radial
nerves in Synapta and other forms (V. o.c). Sense organs, some pig-
mented and perhaps representing eyes, occur in some species on the
tentacles. The radial water-vascular canals, which are absent in the
adult, are temporarily developed in the larva (Fig. I. 8, r.c.c). The
tentacular canals, coining from the circular canal, may branch and
supply several tentacles. The layer of circular muscles is not inter-
rupted at the radii, as in the Actinopoda. Retractor muscles are de-

THE HOLOTHURIOIDEA 235

veloped in some species of Synapta and Chiridota. Very remarkable are
the funnels or ciliated urns (Fig. V. 4, c.u) situated on the mesentery,
and sometimes also on the inner surface of the body-wall. These cup-
shaped organs, the function of which is obscure, are attached by stalks to
the epithelium, and may be joined together into large bunches (Chiridota).
The genera Synapta, Anapta, Chiridota, and Trochodota are herma-
phrodite. The genital pore is situated behind the tentacles (Fig. V. 2).

The absence of radial water- vascular canals, and of interruptions in the
circular musculature, as well as the presence of ciliated urns, distinguish
the Synaptidae from all the preceding familiea The first of these
characters has no doubt been secondarily acquired, since the canals are
present in the larva; correlated with this reduction is the method of
progression by means of the tentacles and of contractions of the body-
wall, accompanied by the entire disappearance of podia. On the other
hand, the shape of the tentacles, the spicules, the presence of retractors,
indicate a distant relationship to the Molpadiidae.

Phylogeny of the Holothurioidea. — It has been shown above
that the Actinopod Holothurians fall into two groups. In the first,
containing the Holothuriidae, Elpidiidae, and Pelagothuriidae, the
tentacles are more or less peltate ; the calcareous ring is radially
symmetrical and of simple structure, it may be reduced and even
absent; the stone-canal often retains its primitive opening to the
exterior ; the genital tubes are sometimes restricted to the left side ;
there are never retractor muscles. In the second group, containing
the Cucumariidae and Molpadiidae, the tentacles are simple or
branched, never peltate ; the calcareous ring is much developed,
with posterior prolongations, and often bilaterally symmetrical ; the
stone-canal always opens internally ; there are always right and left
tufts of genital tubes ; retractors are generally developed. The
two groups, then, probably represent two diverging stems, arising
from a common ancestor possessed Of respiratory trees. The Syn-
aptidae would appear to have come off far down the Cucumarian
stem, perhaps in common with the Molpadiidae. These relation-
ships may be expressed in the following diagram : —

Pelagothuriidae.

Elpidiidae.

olothuriidae.

Holothurioidean.

ancestor. . _

.Cucumariidae.

Molpadiidae.
Synaptidae.

236 THE HOLOTHUR1OIDEA

Little is known concerning the origin of the class. It may,
however, safely be conjectured that the Holothurians are derived
from an Echinoderm ancestor with five typically developed radii,
along which ran branches of the nervous and water-vascular systems,
the latter provided with podia. The outward bilateral symmetry
of the Holothurians seems to have been secondarily acquired in
connection with their free-moving mode of life.1

LIST OF WOIIKS.

1. Baur, A. 1864. (Beitrage zur Naturgeschichte der Synapta digitata.}

Nova Acta Acad. Leop. -Carol, vol. xxxi.

2. Bury, If. 1889. (Studies in the Embryology of Echinoderms.) Quart.

Journ. Micr. Sci., n.s., vol. xxix. pp. 409-449, pis. xxxvii.-xxxix. 1895.
(The Metamorphosis of Echinoderms.) Op. cit. vol. xxxviii. pp. 45-135,
pis. iii.-ix.

3. Hamann, 0. 1883. (Beitrage zur Histologie der Echinodermen. I. Die

Holotlmrien (Pedata), u.s.w.) Zeit. f. Wiss. Zool. vol. xxxix. pp.
145-190, pis. x.-xii. (II. Das Nervensystem der Pedaten Holotlmrien.
Die Cuvierschen Organe. Nervensystem und Sinnesorgane der Ape-
daten), torn. cit. pp. 309-333, pis. xx.-xxii.

4. Htrouard, E. 1889. (Recherches snr les Holothuries des cotes de France.)

Arch. Zool. Exper. (2), vol. vii. pp. 535-704, pis. xxv.-xxxii.

5. Ludwig, H. 1889-92. (Die Seewalzen.) Bronn's Klassen und Ordmmgen

des Thierreichs, Echinodermen, I. Klasse. [Gives a full Bibliography.]

6. 1894. (Reports on an exploration ... by the . . . Albatross . . .

XII. The Holothurioidea. ) Mem. Mus. Harvard, vol. xvii. No. 3.

7. Sclenka, E. 1867. (Beitrage zur Anatomic und Systematik der Holo-

tlmrien.) Zeit. f. Wiss. Zool. vol. xvii. pp. 291-374, 4 pis.

8. 1876. (Zur Entwickelung der Holotlmrien.) Ibid. vol. xxvii. pp. 155-

178, pis. ix.-xiii.

9. Scmon9R. 1888. (Die Entwickelung der Synapta digitata.) Jena. Zeitschr.

vol. xxii. pp. 175-309, pis. vi.-xii.

10. Semper, C. 1868. Reisen in Archipel der Philippines II. Theil. Vol. i.

Holothurien, Leipzig.

11. Theel, Hj. 1882, 1886. (Report on the Holothurioidea.) Challenger

Reports, Zoology, vol. iv. part xiii., and vol. xiv. part xxxix.

12. Ticdcmann, F. 1816. Anatomie der Rb'hren-Holothurie, u.s.w. Fol.

Landslmt.

See also Nos. 2, 3, 19, 25a, and 28, in Literature of Echinoderma generally,
Chapter VIII. p. 35.

1 See Chapter VIII. It should not be forgotten that the plane of symmetry of
the adult does not correspond exactly \vith that of the larva, and that the ali-
mentary canal and the mesentery are invariably twisted spirally from left to right.

CHAPTER XIV.

THE STELLEROIDEA.1

CLASS II. STELLEROIDEA.

SUB-CLASS 1. ASTEROIDEA.

Order 1. Phanerozonia.
„ 2. Cryptozonia.

SUB-CLASS 2. OPHIUROIDEA.

Order 1. Lysophiurae.

„ 2. Streptophiurae.

„ 3. Cladophiurae.

„ 4. Zygophiurae.

THE class of the Stelleroidea includes the starfish, brittle stars,
sand stars, basket-fish, and branching stars, all of which are
characterised by a depressed stellate body composed of a central
disc, whence radiates a number of rays or arms. They have
radiately arranged genital organs (i.e. are actinogonidial) ; they are
not attached by the aboral surface, nor is the oral surface furnished
with ciliated food-grooves (i.e. are eleutherozoic) ; and they usually
have the podia limited to the lower half of the body (i.e. are
lysactinic), instead of having them continued upward to the apical
plates, as in typical sea-urchins (which are desmactinic). The
radial water-vascular vessels lie along the under sides of the arms,
and are exterior to the ambulacral ossicles. The aperture of
the water-vascular system or madreporite is not connected with
the apical plates, as it is in all the Echinoids except Echinocystis.

This list of characters is quite sufficient to mark off' the
Stelleroids from all other Echinoderms. The class is divided into
two groups — the Asteroidea and Ophiuroidea — each of which is
usually ranked as a distinct class, but no definite line of separa-
tion can be drawn between them. The two characters on which
reliance is generally placed are the presence of an ambulacral

1 By J. W. Gregory, D.Sc. MS. closed at end of 1896.

238 THE STELLEROIDEA

groove in the Asteroids and its absence in Ophiuroids ; and the
restriction of the digestive and generative organs to the disc, and
consequent sharp distinction between body and arms, in the latter.
The first character is unreliable, as in the living Ophioteresis there
are no ventral plates, and a shallow ambulacral furrow is accord-
ingly present. One order of fossil Ophiuroids — the Lysophiurae
— has the same feature more strongly marked. The differentia-
tion of the body into disc and arms happens in most Ophiuroids,
but also in some Asteroids, e.g. Freyella. The restriction of the
digestive organs to the disc appears to offer a more reliable
character ; but in Astrophiura the arms are sharply marked off
from the disc and contain no digestive caeca, while the ambulacral
ossicles are asteroid. Similarly the digestive and genital systems
of the Asteroids, Colpaster, and some species of Freijella must be
limited either to the well-marked disc, or at most to the bases of
the arms ; while the arm structure is practically identical with
that of some Palaeozoic Ophiuroids. It must be remembered,
moreover, that the digestive sac of Ophiuroids is marked by a
series of radial bulgings, which may be homologous with the
radial caeca of Asteroids. The position of the madreporite is relied
on by Perrier, but it will not serve ; in the Asteroids, Asterina,
and Palasteriscus it is ventral, as in most Ophiuroids.

Not only is there no character which serves to separate the
Ophiuroids and Asteroids, but the whole structure of the body is
on the same plan in both groups. It consists in both of a central
disc and a series of (usually five) radial rays. The skeleton in
each ray consists essentially of two series of plates — the ambu-
lacral and adambulacral. The former lie internal to the radial
water- vascular vessel, and the furrow which this occupies is later-
ally protected by the adambulacral plates. Additional elements
may occur, but are not found in all members of either division.
The mouth armature consists of a ring round the mouth, formed
by the union of one or more pairs of ambulacral and adambulacral
plates for each arm, and bearing spines modified to serve as teeth.
The body is protected by accessory plates or granules in the integu-
ment ; these plates may be protected by spines and pedicellariae.

The alimentary system consists of a central digestive sac,
opening by a mouth at the centre of the ventral surface ; the size
of the digestive sac is increased by radial bulgings, of which there
are as many pairs as the Stelleroid has arms ; these bulgings may be
short and limited to the disc or base of the arms, or extend up
the arms. There may or may not be an anus.

The water -vascular system consists of a ring round the
oesophagus ; a radial vessel runs up each arm from the ring, which
also bears a series of Polian vesicles, or sac-like diverticula. The
radial vessels give off a pair of branches in each segment ; each

THE STELLEROIDEA 239

branch ends in a podion, which may be pointed, or may end in a
sucker. Connected with each podion there may be a globular
ampulla (absent from some Asteroids, e.g. Brisinga and from all
Ophiuroids). The madreporite in either group is dorsal, marginal,
or ventral. The nervous system comprises a circumoesophageal
ring and a radial branch along each arm or ray.

The reproductive organs consist of strands connected with the
axial coelomic system ; there is a central ring whence a pair of
strands pass to each ray. A number of small gonads occur on
each strand ; the gonads of each strand may be grouped into one
bundle, with a common aperture at the margin of the disc (as in the
Asteroid Asterias and the Ophiuroid Ophiothrix) ; or they may occur
only at the bases of the arms, as in Freyella ; or they may occur as
a series of distinct gonads with separate apertures, as in Brisinga and
Ophiactis. In most Ophiuroids the gonads discharge into a bursa.

As both Ophiuroids and Asteroids are therefore constructed
upon the same fundamental plan, as they contain the same varia-
tions from the typical arrangement, and as there is not a single
constant difference between them, it seems indispensable that they
should be united into one class, the STELLEROIDP;A, which may be
diagnosed as follows : —

Eleutherozoic, actinogonidial, and lysactinic Echinoderma in
which ambulacral plates lie internal to the radial ambulacral vessels.
The madreporite is not connected with an apical system of plates.
The body is more or less depressed, and is markedly stellate.

In spite of the fact that the separate treatment of the Asteroids
and Ophiuroids has led to many unfortunate errors, and still
hampers the classification of the group, it seems advisable here
to consider the sub-classes separately.

SUB-CLASS 1. ASTEROIDEA.

The sub-class Asteroidea includes the Echinoderms known as
Starfish. The animals consist of a central body marked on the
ventral side by a series of radial furrows which are usually continued
outward along prolongations of the body known as arms. They
live on the sea-floor and creep about by means of suckers or podia in
the radial furrows, the side containing which is always placed down-
wards. The general aspect of starfish is therefore very different
from that of any of the previously described groups of Echinoderms.
They are closely related to the Ophiuroids, and no very satisfactory
line of demarcation can be drawn between the two sub-classes ; but
the Asteroids usually have diverticula from the alimentary canal
extending along the arms, which pass gradually into the disc.

Before 1841 the Asteroids and Ophiuroids were always considered
members of one group. Thus J. H. Linck (23), who in 1733 began

240

THE STELLEROIDEA

the systematic study of the Stelleroids, included members of both
sub-classes in his group "Stella." He separated the Asteroids as the
Stellae fissae, which he divided according to the number of arms or
rays into such divisions as Trisactis, Tetractis, Hexadis, etc. Linnaeus
in 1766 included both sub-classes as well as some unstalked
Crinoids in his genus Asterias, grouping all the starfish together as
"Stellatae." Lamarck in 1816 (21), de Blainville (1830 and
1834), Nardo and Brandt (1834), and L. Agassiz (1835) proposed
various divisions of Asterias, which these authors recognised to be
a family or order. In 1840 Gray's Synopsis, and Miiller and
Troschel's essay, Ueber die Gattunyen der Asteiien (35), first prepared
the way for a scientific classification. In 1842 the latter authors'
System der Asteriden (36) laid the foundation for all later work.

After that date additions to our knowledge of the recent Asteroids
have been made by many authors, especially A. Agassiz, Barrett,
Bell, Danielssen, Fewkes, Forbes, Gray, Jullien, Koren, de Loriol
le Fort, Liitken, Marenzeller, Martens, Moebius, Perrier, Philippi,
Sars, Sladen, E. A. Smith, Stimpson, Studer, Thompson, Verrill,
Viguier. The fossil forms have been described by E. Billings, Eck,
Forbes, E. Fraas, Goldfuss, Heller, Hall, de Loriol le Fort, Miiller,
C. F. Roemer, Salter, Simonowitsch, Stiirtz, Wright, and others.

The study of the anatomy of Asteroids received its first stimulus
from the researches of Joh. Miiller (34). The skeleton has been
described by Meckel (1828), Duvernoy (1848), Gaudry (13), and
especially Viguier (52). Perrier has devoted most attention to the
pedicellariae (1875 and 1884). The study of the visceral anatomy
was begun by Tiedemann, Delle Chiaje, and Meckel ; and of the

nervous system by Krohn.
The present position of the
subject is due mainly to
Ludwig (25), Cuenot (8, 9),
and Hamann (17).

The embryology has
been worked out by many
authors, the study of the
early stages being unusually
easy ; among contributions
to this branch of the subject
are those of Joh. Miiller
(1848 - 55), A. Agassiz
(1877), Ludwig (1882),
MacBride(1893 and 1894),
(32, 33), and Bury (1889
and 1895).

The principal classifications after those of Gray, and of Miiller
and Troschel, are those of Viguier (1878), (52), based on the oral

Fio.

Actinal surface of Asterias ri<l>ens with a podion (o)
enlarged.

THE STELLEROIDEA

241

FIG. II.

Asterias ruliens, part of abac-
tinul skeleton.

skeleton, of Perrier (1884 and 1894), (38), on the pedicellariae,
and of Sladen (1889), (48).

Structure of a Typical Asteroid. — The common English star-
fish (Asterias rubens, L.) is a convenient type of the Asteroidea. It
has a flattened body composed of a central disc from which radiate
five arms (Fig. I.). The upper or abactinal surface of the Body is
covered by an integument or perisoma (Fig. II.), composed of a
network of calcareous rods, the meshes between which are closed
by tough membrane. The anus opens almost
in the centre of the abactinal surface.
Between the anus and one of the angles
between the rays occurs the madreporite, a
thick grooved plate, perforated by pores
leading to the water-vascular system.

The ventral surface of each Arm is tra-
versed by a broad groove ; since the grooves
radiate from the mouth to the ends of the
arms, the ventral surface is known as the
actinal surface. The mouth is at the centre
of this side of the body, and is surrounded by spines (the " mouth
papillae "). The grooves are occupied by four rows of suckers or
podia, and therefore correspond to the ambulacral areas of the
Echinoid. On either side of the grooves are three rows of spines.
Dissection is necessary for the recognition of any further points
in the structure of the Asterias. By the removal of some of the
podia, the ambulacral grooves may be seen to lie outside a series
of pairs of narrow plates — the ambulacral ossicles (Fig. III.). The

two series of ossicles meet
in the middle line ; laterally
they abut against a row
of adambulacnil ossicles, be-
yond which are further rows
of interambulacral and mar-
ginal ossicles, all of which
are comparatively small. The
ossicles are protected by spines and pedicellariae similar to, but
simpler than, those of Echinoidea.

The Oral Skeleton (or actinostomial ring) consists of a solid
calcareous ring around the mouth. It is composed of thirty plates
in a quinqueradiate starfish, there being always six times as many
plates as there are rays. Each segment of the oral skeleton con-
sists of two pairs of ambulacral, and of one pair of adambulacral
ossicles. In Asterias the ambulacral plates are more prominent
than the adambulacrals, and project into the oral cavity. The
mouth armature is therefore on the ambulacral type (Viguier,
52).

16

FIG. III.

Asterias rulens, ambulacral and adainbulacral plates,
a, apertures for podia.

242

THE STELLEROIDEA

The Alimentary System consists of a mouth at the centre of
the actinal surface of the starfish. The oesophagus is very short

Fio. IV.

Asterias nibens. I, ray from which the skin of the abnormal surface has been removed
and the outgrowths (c) displaced, showing the ambulacra! ossicles (<u>); the suckers (x),
g, gonads in ray II ; a, anus; s;>, stomach, the folded arrangement of the walls of which are
shown by removal of part of the upper wall at the base of ray V. From F. J. Bell, Catalogue
of British Echinoderms in the Brit. Mvs. (Nut. Hist.).

THE STELLEROIDEA 243

and leads to a large stomach, which occupies most of the disc.
From the stomach five branches pass off, one to each ray. Each
branch divides into two caeca, which lie one on either side of the
ray. From the stomach a short rectum leads upward to the anus,
which opens on the abactinal surface at a little distance from the
centre. Two small longitudinally folded diverticula from the
rectum occur below the anus. These rectal caeca occupy a
similar position to the " respiratory trees " of Holothurians, with
which they may be homologous.

The Water- Vascular System consists of a circular vessel round
the oesophagus (the circumoesophageal canal or water- vascular ring),
from which, in a five-rayed starfish, there are eleven offsets. The
most important are the five radial canals, one of which passes
along each ray, just external to the ambulacral ossicles. From
these radial canals branches are given off on either side; each
branch ends in a tubular podion, which consists of an internal
reservoir or ampulla situated above the ambulacral ossicles, and of
an external tube or sucker. Valves occur on the transverse branches,
and prevent water, expelled from the ampulla, returning into
the radial vessel ; they thus direct it into the sucker. The next
set of offsets from the circumoesophageal canal are five sac-like
" Polian vesicles," one in each interradius ; they act as reservoirs
for the water-vascular system. The last (eleventh) vessel on the
circumoesophageal canal is the "stone-canal," which runs from
the base of one of the Polian vesicles to the upper surface of the
starfish. It expands above, and its upper end is attached to the
madreporite, through the pores in which water enters the water-
vascular system.

The circumoesophageal canal also supports nine tufts of
tubules known as Tiedemann's bodies ; there is a pair of tufts
in each interradius, one on either side of the base of the " Polian
vesicle," but in the interradius containing the stone-canal there is
only a single Tiedemann's body.

The presence of a blood-vascular system in Asteroids is not yet
determined, the organs described as such belonging to the Pseud-
haemal System (cf. pp. 22, 26). The main organ in this system
is the "axial sinus," which is a rather thick vertical tube sur-
rounding the stone-canal. It communicates below with a ring (the
circumoesophageal pseudhaemal ring), which surrounds the mouth
and gives off five radial branches, which pass one along the upper
side of each ray. At its upper end the axial sinus communicates
with the genital ring ; attached to this ring are five pairs of short
processes, while an additional pair passes beside a prolongation of
the axial sinus leading to the madreporite. Some of the pores of
the madreporite open to the axial sinus, and there is no known
direct communication between the latter and the stone-canal.

244

THE STELLEROIDEA

FIG. V.
Brisinga coronate. Abactinal surface.

The Nervous System consists of three disconnected sets of
nerves. The most important set is composed of a circumoral

THE STELLEROIDEA

245

ring, from which a branch runs up each ray between the rows of
tube-feet, and external to the pseudhaemal radial vessel. Each
branch gives off nerves to the tube-feet and ectoderm of its ray.
The second set of nerves consists of bands lying internal to the
radial branches of the first set ; they lie along the perihaemal canal,
which surrounds the pseudhaemal vessel ; the branches from these
nerves supply the muscles of the ambulacral ossicles. The third
group of nerves is abactinal in position ; there is a ring round
the anus, giving off a branch along the upper side of each ray, and
innervating the muscles of the body wall.

The Genital Organs of Asterius consist of ten branched glands,

a pair being situated on the dorsal side of each ray close to the
disc. Each gland has a separate aperture on the abactinal surface.

The only special respiratory organs are a series of long, tubular
prolongations of the body - cavity, known as papulae, dermal
branchiae, lymph gills, or branchial vesicles. In Asterias they
occur on all sides of the rays and disc.

The most interesting sensory organ of Asterias is a small
eye-spot on the terminal podion of each ray (Fig. XXV. on
p. 30). The animal has an olfactory sense, for it will follow
food, even after the eye-spots have been destroyed ; the situation
of this sense is diffuse, for any part of a starfish arm that can
move independently will follow food (Romanes) ; but Prouho
limits the sense of smell to a few podia near the end of the ray,
which he calls " palps."

THE STELLERO1DEA

The Variations in Structure from the typical genus are less
remarkable among the Asteroids than among the Echinoids and
Ophiuroids.

In shape the extremes are genera such as Brisinga (Fig. V.) or
Freyella, and Culcita (Fig. VI.). In the first two the arms are
numerous and slender, and sharply marked oft' from the disc. In
the last the body is bun-shaped, and the ambulacra extend for a
short distance over the abactinal surface.

The most conspicuous variation from Asterias. in skeletal
structure is due to the presence of a series of thick plates round
the margin both of arms and disc. These marginal plates are
best developed in the order Phanerozonia. There may be two
series, one above the other, and known respectively as supra-
marginals and infra -marginals (Fig. VII. d and c) ; in some
genera intermarginal plates occur between these two series.

The spaces (actinal interbrachial
areas) between the marginal
plates and the ambulacra] fur-
rows may be occupied by a
series of accessory plates, form-
ing a pavement like mosaic.

The accessory plates on the
abactinal side may be large or
small, equal or unequal. In
Fio vn some species the central plate

Segment of an,, of Astropcctc* irre<,nl«ri«. a, ls 1™ZG> and SOmG °f the re'

supra-ainbulacral plates ; b, ambulacra! plat«s ; inaming plates are arranged in

c. and (J, inferior nnd superior lateral plates; r. • i i ., mi i , f

dorsal plates with paxillae. Circles TOUlld it. The plates of

three of these circles have been

regarded as the homologues of the calycinal plates of Crinoids, and
are accordingly sometimes named the radials, basals, and infra-
basals (but see p. 14). In Cnemidiaster wywllei the central, and
the so-called radials and basals are present ; in Zoroaster fulgens
there are infra-basals as well. The order of development of these
plates is as follows : — First five plates appear round the centre
of the abactinal surface ; these plates move outwards to the arm-
tips, and form the terminals ; a second ring appears, the plates of
which are the " basals " ; the central plate next develops in the
centre of this ring ; the radial circle follows, and after that come
the "infra-basals." Between the plates of the last set and the
central some genera have additional plates, which cannot be
homologised with any in Crinoids.

The Oral Skeleton consists of a ring of plates round the mouth.
The ring is composed of as many segments as the starfish has rays,
and each segment is interradial, and forms an " oral angle " ; it
consists of two sets of plates, usually three in number.

THE STELLEROIDEA 247

The oral skeleton is described as either ambulacral or adam-
bulacral, according as the ambulacral or adambulacral plates
are the more prominent. Viguier, in his important memoir on
the skeleton of the Asteroids, pointed out the existence of these
two types, and based his classification upon this character. As we
have seen, in the genus Asterias the oral skeleton consists of a
solid ring, in which the ambulacral plates form five prominences,
while the adambulacral plates are small, and occur in the angles
between them. Such a mouth-structure is described by Viguier as
"ambulacral." In Pentaceros (Oreaster), on the other hand, the
ambulacral plates are inconspicuous, and the adambulacral plates
project into the buccal cavity and form the jaws. This type is
Viguier's "adambulacral mouth." In both cases above the adam-
bulacral plates is a basal, interbrachial, or oral plate which is
called by Viguier the odontophore, although it does not bear the
teeth ; its value is of secondary importance.

The Pedicellariae of Asteroids are of four main types. The
simplest form consists of a row of pairs of small, sessile, opposajjle
spines; these are the "pseudo-pedicellariae." The members .o'f
the second set are " sessile." The next advance is the develop-
ment of a stalk ; of these pedunculate pedicellariae there are
two varieties: (1) the " forficiform," in which the two hooks are
attached to the nearest end of the basal plate nearest to them ;
(2) the " forcipiform," in which the two hooks cross one another
and are attached to the end of the basal plate furthest from them.
Perrier has used the pedicellariae as the basis of his classification
of the Asteroids, on the ground that they are the degenerate repre-
sentatives of organs once more important. Other authors, however,
regard them as modified and elaborated spines, in which case they
are of little taxonomic value.

Another type of spines occurs as part of the structures known
as " paxillae." Each paxilla consists of a thick plate supporting a
number of short, calcareous pillars, the summit of each of which
is covered by a group of small spines. In some Phanerozonia, such
as Astropecten, the paxillae occupy almost the whole abactinal
surface of the Asteroid (Fig. VII.).

The Water- Vascular System is on the same plan as in Asterias,
but there are the following modifications : — In most starfish there
is a pair of Polian vesicles in each interradius ; they rise from the
circumoesophageal vessel beside the Tiedemann's body. The
radial branches and its podia are simplified by the absence of
ampullae, as in Brisinga, or by the podia ending in points instead
of suckers, as in the Astropectinidae. Suckers are improvised in
such pointed podia by a contraction of the walls below the end.

The number of madreporites is very variable among the
Asteroids ; in most of those with many rays there are several

248 THE STELLEROIDEA

madreporites, but in the Solasteridae there is but one ; on the
other hand, many five-rayed starfish, such as Asterias capensis, have
more than one. The stone-canal is repeated, especially in forms
which reproduce asexually.

There is no anus in the Astropectinidae, the members of which
are more primitive than Asterias in many respects. The radial caeca
of the stomach remain constant, but the rectal caeca vary in number
and arrangement ; they are increased to five in many genera, and
in Cnhita each of the five caeca branches into two.

The "papulae" or branchial vesicles, which, in Asterias, rise
from all parts of the external surface, are limited among Phaner-
ozonia to the abactinal surface, and to the area enclosed by the
supra-marginal plates.

The Genital Organs are, as a rule, less concentrated than in
Asterias. The glands are repeated along the arms ; in the simplest
cases each series discharges its products by a single sieve-plate.
The extreme case is in Brisinga, where there are a series of separate
glands along the arm, one pair to each segment, and each gland
discharges by a separate aperture.

In Asterina gtttlosa the genital orifice is on the ventral or
actinal surface, as in Ophiuroids.

The development of the Asteroids is generally indirect, the larva
passing through a metamorphosis. The typical form of larva is
the Bipinnurin, which passes through an Auricularia stage and thus
resembles the larvae of the Holothurians rather than of the
Echinoidea or Ophiuroidea (see p. 5). In some cases the Bipinnaria
develops into a Brachiolaria by the division of the frontal process
of the larva into three branches. In some genera, such as Asterina
(the development of which has been studied with especial care),
the Bipinnaria stage is never developed, although the larvae are
free-swimming and undergo metamorphosis. In other cases, e.g.
in Blakiaster and Pteraster, the development is direct; in the
former case, the ova develop in " arcade-like spaces " between the
paxillae of the abactinal surface ; in the latter there is a large
marsupium formed by the presence of a supra-dorsal membrane
rising above the abactinal surface.

Asexual reproduction is not uncommon ; it results either from
a division of the body, approximately into halves, or by regrowth
of the disc from an arm that has been thrown off. The latter
method occurs especially in Linrkia, and with the first appearance
of the disc the starfish is said to assume the comet form.

Proceeding to the Systematic Description of the orders and families,
we may sum up the foregoing characters in the following Diagnosis of
the Sub-Class.1 The Asteroidea are elcutherozoic, actinogonidial, and

1 Emended from Bell (4), p. 19 ; the terms are explained, anted, p. 237.

THE STELLEROIDEA 249

lysartinie Kchinodi'rms in which there is an ambulacral groove. The
arms gene rally pass gradually into the disc, but in some eases are
sharply marked oil' from it. The digestive system generally has un anus,
and shares in the stellate disposition of the body. Pentameric repetition
is more often exceeded in this than in any other class, and asexual repro-
duction is not uncommon. Respiration diffuse. The madreporic aperture
is generally ahactinal.

This diagnosis at once sharply separates the Asteroids from all
Echinoderms except Ophiuroids, between which, as we have seen (p. 238),
it is not possible at present to draw any precise line of separation. The
Asteroid eo, however, always have an open actinal groove, whereas this
is exceptional among the Ophiuroidea ; the arms usually pass gradually
into the disc, and generally contain throughout prolongations of the genital
and alimentary systems.

OitDKit 1. Phanerozonia, Sladen.

Asteroidea with large marginal plates, and with the dermal branchiae
or lymph gills limited to the abactinal surface.

This order includes a group of starfish which began in the Cambrian
age and has lived on till the present time. Its members can be readily
distinguished by the large si/e of the marginal plates. The limitation of
the dermal branchiae to the abactinal surface is a more primitive con-
dition than that met with in the Cryptoxonia. Embryological evidence,
and the greater importance of the order in the Palaeozoic and Mesoxoic
eras also suggest that this is the simpler of the two orders of star-
fish.

The Palaeozoic genera appear to be normal members of this order,
and some of them may be included in existing families. They are, how-
ever, often all grouped together as an order, the " Pahvasteroidea," and
separated from all the later, or " Euasteroidea." The character on which
this separation is based is the alternation of the ambulacral ossicles in the
former, whereas they are said to be always opposite in post- Palaeozoic
Asteroids. This character is of great importance among Ophiuroids, for
when the ossicles are alternately arranged, they cannot be united into verte-
bral ossicles. But when the ossicles are narrow, thin plates, closely packed
into two series, one on either side of a ray, and when the separate ossicles
meet those of the other side only by their narrow ends, then alternation
is very likely to arise from growth pressure. In fact, one part of an
arm may have the ambulacral ossicles alternate, while in another part
they may be opposite. The character, moreover, is one on which little
reliance can be placed when applied to fossils, for a slight movement is
sufficient to alter the relative positions of the two series. It is difficult
to explain the relative positions of the ambulacral ossicles in different
arms of the same Asteroid, except on the assumption that the two series
moved past one another during the lateral bending of the arm. Alterna-
tion of the ossicles was probably an original character ; but as the arms
became flexible with the reduction of the external skeleton, and as the

250 THE STELLEROIDEA

ambulacral ossicles became narrower, greater freedom was gained by the
opposition of the two plates of a pair. The artificial nature of the
divisions based on this character is shown by Stiirtz's action in dividing
several of his families into halves and placing members of the same
family in two different orders. Thus he founded a family Palajechin-
asteridae for the genera Eckinasterella and Palasteriscus ; but he included
the latter among the division with alternate ambulacral ossicles, and the
former in the division in which these ossicles are opposite. As the
" Palocasteroidea " include representatives of both Phanerozonate and
Cryptozonate Asteroids, and of several families of each, it is necessary to
dismember such an artificial group.

FAMILY 1. PALJEASTERIDAE. Phanerozonia with most or all of the
ambulacral ossicles alternate ; the madreporite is dorsal. Oral armament
adambulacral. Abactinal skeleton tessellate. Rays long, disc small.
This family includes a series of Asteroids occurring in the Lower
Palaeozoic. In most of them the ambulacral ossicles are alternate,
but in some cases these plates are opposite, either for a part, or for the
whole length of the arm. Hence this character does not seem to be of
the value assigned to it. The marginal ossicles are always conspicuous,
and, as far as is known, the madreporite is fairly small and dorsal in
position. The oral skeleton consists of a ring in which the adambulacral
plates are most conspicuous. There are two sub-families. SUB-FAMILY 1.
PAL^ASTERINAE, in which the ambulacral ossicles are alternate. Genera
— Palwaster, Hall; Argaster, Hall; (?) Monaster, Eth. jnr. ; and (?)
Pctraster, Billings pars. The genera are all Palaeozoic, ranging from the
Cambrian to the Devonian. Among existing families the nearest ally is
the Archasteridae. SUB- FAMILY 2. XEN^STERINAE, including those
with inooi, of the ambulacral ossicles in opposite pairs. Genera — Xcnaster,
Simonowitsch ; Tetraster, Eth. jnr. & Nich.

FAMILY 2. PALJEASTERINIDAE. Phanerozonia with alternate ambu-
lacral ossicles and small marginal plates. The oral armature is adam-
bulacral. The madreporite is abactinal. The rays are short and are
separated by large interradial areas. Genera — Palaasterina, M'Coy ;
tichoenaster, Meek & Worthen.

FAMILY 3. ASPIDOSOMATIDAE. Phanerozonia with alternate ambula-
cral ossicles ; large marginal ossicles and large interradial areas ; rays
massive, petaloid, sub-petaloid, or tapering. On the abactinal surface
there are large depressed areas between the marginal ossicles and the
outermost of the longitudinal series of large plates which run along the
arms. Genera — Aspidosoma, Goldf. (the type-species is A. Arnoldi, but
as this is imperfectly known, A. pdaloides, Simonowitsch, may be
accepted provisionally) ; Palaeonedria, Stiirtz ; Palaeostella, Stiirtz ; Trich-
asteropsi^ Eck ; Archanterias, Mull., may belong here, but the genus is
insufficiently known.

FAMILY 4. TAENIASTERIDAE. Phanerozonia with alternate ambulacral
ossicles. There are neither disc nor interbrachial areas. The adambu-
lacral plates are large and act as marginal plates. The axes of the
marginal plates are parallel and the rays petaloid (as in Stenaster), or the
axes of the marginal plates are convergent ; these plates bear spines on

THE STELLEROIDEA 251

their free end?, and the rays taper gradually (Taeniaster). Genera —
Tacniastcr, Billings ; Stenaster, Billings, pars (S. salteri, but not S. pul-
chellus) ; both Ordovician of Canada ; Satteraster and Urasterella, M'Coy,
Silurian, England ; Protasteracanthion, Stiirtz, Devonian, Germany.

FAMILY 5. ARCHASTERIDAE. Phanerozonia with opposite ambulacral
plates. There is an anus, but no super-ambulacral plates. Pedicellariae
are generally present. The abactinal plates are spiniform or paxilliform.
The adainbulacral plates are large. This family includes a large number
of Neozoic starfish, ranging from the Lower Oolites to the present day.
They have frequently been included with the Astropectinidae, from
which they differ by the presence of an anus, by the large size of the
adainbulacral ossicles, and by the absence of supra-ambulacral plates.
There are four sub-families, including sixteen genera. SUB-FAMILY 1.
PARACHASTERINAE, comprising those in which the branchial vesicles or
papulae are limited to an area at the base of the rays, and in which
the actinal interradial plates are absent or are very few in number.
Genera — Cheiraster, Studer ; Pararchaster, Sladen ; Pectinaster, Perrier ;
and Pontaster, Sladen. SUB -FAMILY 2. PLUTONASTERINAE, including
those with papulae scattered over the whole abactinal surface. There
are numerous actinal interradial plates. Genera — Crenaster, Per. (non
Ag.) ; Dytaster, Slad. ; Goniopecten, Per. ; Lonchotaster, Slad. ; Persephon-
aster, Mason & Alcock ; Plutonaster, Slad.; Tethyaster, Slad. SUB-
FAMILY 3. PSEUDARCHASTERINAE, including those with a definite median
line of plates along the rays, and with the abactinal plates arranged
in series parallel to the central line. There are no pedicellariae.
Genera — Pseudarchaster, Slad. ; Aphroditaster, Slad. SUB-FAMILY 4. ARCH-
ASTERINAE, including those in which there is a definite median line of
abactinal plates, and the remainder are arranged in oblique rows.
Pedicellariae present. Genera — Acanthar "chaster, Verrill ; Archaster, Verr. ;
Isaster, Verr. The following genera are included by Sladen in this
family, but not divided among the sub -families : — Benthopeden, Verr. ;
Blakiaster, Per. ; Luidiaster, Stud.

FAMILY 6. PORCELLANASTERIDAE. Phanero/.onia with opposite ambu-
lacral plates ; with thin lamelliform marginal plates, traversed by cribri-
form organs. There are two sub-families. SUB- FAMILY 1. PORCELL-
ANASTERINAE, in which the cribriform organs are highly developed and
limited to a few plates, and there are no fimbriated channels in the
actinal interradial areas. Genera — Caulaster, Per. ; Porcellanaster, Wyv.
Thomson (Fig. VII I.); Styracaster, Hyplialaster, Thoracastcr, and Pseud-
aster, Sladen. SUB-FAMILY 2. CTENODISCINAE, in which the cribriform
organs are simple and occur on the margins of each pair of marginal
plates ; continuations from the cribriform organs run through the actinal
interradial areas as " fimbriated channels." Genus — Ctenodiscus, Miiller &
Troschel. The most interesting features of this family are the develop-
ment of the cribriform organs, densely packed groups of small spinelets or
lamellae on some or all of the marginal plates. Their function is uncer-
tain, but, according to Sladen, " it is not improbable they act as percolators."
In some species of Astropecten the marginal ossicles are bordered by
fringes of small spines, which A. Agassiz has compared to the fascicles of

252 THE STELLEROIDEA

Spatangoids (see p. 319). In Ctetiodiscus the grooves between the marginal
ossicles are bordered by bands of lamellae forming the simplest type of
cribriform organs. In the rest of the Porcellanasteridae these organs are
confined to a few special plates. In some species of this family the
anus opens on the summit of a small tube. It rises from the abactinal
surface of the starfish, and has been accordingly compared to the stem
of Crinoids, an homology which is quite inadmissible.

FAMILY 7. ASTROPECTIXIDAE. Phanerozonia with opposite ambula-
cral plates and paxilliform abactinal plates. Super-ambulacral plates are
present, and the adambulacral plates are compressed. There is no anus,
and pedicellariae are rarely present. The compressed adambulacral plates
and the presence of a series of super-ami mlacral plates, which occur inside

Fie;. VIII.
Porcdlanastcr cacrulcus. Abactinal surface showing cribriform organs and anal tube.

the arms above the ambulacrals, are the two most striking features of
the Astropectinidae. The position of the super-ambulacral plates is
shown in Fig. VII. There are two sub -families and nine genera.
SUB-FAMILY 1. ASTROPECTININAE, including those members of the family
in which the adambulacral plates touch the infero- marginal plates
along the ray. The marginal and adambulacral plates do not correspond
in length or number. Genera — Astropertea, Schul/e ; Bathybiaster,
Danielssen and Koren ; Craspidaster, Slad. ; Dipsacaster, Alcock ; Blakiastcr,
Per. (syn. Leptoptychaster, Smith), (Fig. IX.) ; Moriatter, Phoraster, and
Psilaster, Sladen. SUB-FAMILY 2. LUIDIINAE. Astropectinidae with a
row of small plates separating the adambulacral and infero- marginal
plates. Genera — Astrdhi, Per. ; Luidia, Forbes ; Platasterias, Gray.

FAMILY 8. PENTAGONASTKRIDAE. Phanero/onia with opposite ambu-
Incral plates, large marginal plates, and tessc-llate abactinal skeleton.

THE STELLEROIDEA 253

The arms are short and the shape of the starfish is generally penta-
gonal. SUB -FAMILY 1. PENTAGON ASTERINAE. Pentagonasteridae with
rounded, polygonal, or paxilliform abactinal plates. Genera — Anthenoides,
Per. ; Asterodon, Per. ; Astrogonium, Miiller & Troschel ; Calliastcr, Gray ;
Calliderma, Gray ; Chitonaster, Slad. ; Comptonia, Gray ; Dorigona, Gray
(Nymphaster, Slad.) ; Goniodon, Per. ; Hoplaster, Per. ; Iconaster, Slad. ;
Leptaster, Lor. ; Mediaster, Stimps. ; Metopaster, Slad. ; Mitelephaster, Alcock ;
Mitraster, Slad. ; Nedria, Gray ; Odontaster, Verr. (Gnathaster, Slad.) ;
Paragonaster, Slad. ; Pentagonaster, Schulze ; Phaneraster, Per. ; Pycnaster,
Slad. ; Eosaster, Per. ; Stephanaster, Ayres. SUB-FAMILY 2. GONIODISCINAE.
Pentagonasteridae with flat, stellate, abactinal plates. Genera — Goniodiscus,

FIG. IX.
Abactinal view of Hippasterias phrygiana.

Miill. & Trosch. ; Leptogoncister, Slad. (included by Perrier in Archasteridae) ;
Stellaster, Gray ; Ogmaster, v. Martens. SUB-FAMILY 3. MIMASTERINAE.
Pentagonasteridae with small, stellate, paxilliferous abactinal plates.
The plates of the actinal intermediate areas imbricate over one another.
Genus — Mimaster, Slad. Hoplaster, Per., is included by Sladen in the
Pentagonasteridae, but Perrier places it with Goniodon, Gnathaster, and
Asterodon, in a special sub-family of Archasteridae.

FAMILY 9. ANTHENEIDAE, Per. Phanerozonia with opposite ambu-
lacral ossicles and massive marginal plates. Abactinal skeleton stellato-
reticulate ; the actinal intermediate plates bear large valvate pedicellariae.
Genera — Anthenea, Gray ; Goniaster, Ag. (em. Per.) ; Hippasterias, Gray
(Fig. IX.).

FAMILY 10. PENTACEROTIDAE. Phanerozonia with opposite ambu-

254 THE STELLEROIDEA

lacral ossicles and irregular marginal plates ; the upper series are often
covered. The abactinal skeleton is reticulate, and the plates bear large
tubercles. There are no valvate peclicellariae on the actinal interradial
areas. Genera — Amphiaster, Verr. ; Asterodiscus, Gray ; Choriaster, Liitken ;
Culcita, Ag. (Fig. VI.) ; Nidorellia, Gray ; Paulia, Gray ; Pentaceropsis,
Slad. ; Pentaceros, Schulxe (Oreastw, Miill. & Tr.) ; Sphaeraster, Quenst. ;
Sphaerites, Quenst.

FAMILY 11. GYMNASTERIIDAE. Phanerozonia with opposite ambu-
lacral ossicles and unequally developed marginal plates. Abactinal
skeleton tessellate, but its plates are irregular and only partially in
contact. The actinal interradial areas contain large plates. The whole
test covered with membrane. Genera — Asteropsis, Miill. & Tr. ; Derm-
asterias, Per. ; Gymnasteria, Gray ; Lasiaster, Slad. ; Maryinaster, Per. ;
Porania, Gray ; Poraniomorpha, Dan. & Kor. ; Itheyaster, Slad. ; Tylaster,
Dan. & Kor.

FAMILY 12. ASTERINIDAE. Phanerozonia with opposite ambulacral
ossicles, and with small, inconspicuous marginal plates, the axes of which
are convergent. Intermediate plates imbricate ; those on the abac-
tinal side lamelliform. Pedicellaria absent. SUB-FAMILY 1. GANERIINAE,
with large marginal plates. Genera — Cycethra, Bell; Ganeria, Gray;
Lebrunaster, Per. ; Radiaster, Per. SUB-FAMILY 2. ASTERINIDAE, in which
the marginal plates do not exceed the remaining plates in size. Dermal
branchiae arise from any part of the abactinal surface. Genera — Asterina,
Nardo, which has often been described as a typical Asteroid, and its embry-
ology carefully studied; Disasterina, Per.; Nepanthia, Gray ; Patiria, Gray.
SUB-FAMILY 3. PALMIPEDINAE, with dermal branchiae confined to the
radial regions. Genera — Palmipes, Ag. ; Stegmaster, Slad. Tremaster,
Verr., is also assigned to this family.

ORDER 2. Cryptozonia, Sloden.

Asteroidea with the marginal plates small and inconspicuous or
absent ; when present the upper and lower rows do not touch. Dermal
branchiae not limited to the abactinal surface. Ambulacral plates are
generally crowded and narrow. Either the ambulacral or adambulacral
plates are the more prominent in the oral skeleton.

This order is characterised by three main characters : ( 1 ) The
insignificance of the marginal plates, whence the name Cryptozonia ;
(2) the occurrence of papulae or dermal branchiae beyond the abactinal
surface, whence the order is described as " adetopneusic " ; and (3) the
crowding and narrowness of the ambulacral plates, whence the order is
said to be " leptostroterate." The last character is not developed in some
Palaeozoic genera that appear to belong to this order.

FAMILY 1. PALAEOCOMIDAE. Cryptozonia with alternate ambulacral
ossicles and numerous long free actino-lateral spines. There is a web
supported by a reticular calcareous skeleton. The adambulacral plates
are large, and the spines are borne on the plates adjoining them.
Qenus — Palaeocoma, Salter. Bdellacoma, Salter, is placed by its author
as a sub-genus of Palaeocoma ; its affinities are doubtful, but are certainly
not with this family.

THE STELLEROIDEA 255

FAMILY 2. LEPIDASTERIDAE. Cryptozonia with alternately arranged
ambulacral ossicles. Disc large ; rays comparatively short, thick, and
blunt. No lateral spines. The abactinal skeleton consists of closely set,
irregular, granular plates. Genus — Lepidaster, Forbes (Fig. X.).

FAMILY 3. TROPIDASTERIDAE. Cryptozonia with ambulacral ossicles
opposite or sub-alternate. Rays short, broad, arid Hat. Ambulacral
ossicles narrow ; adambulacral ossicles broad and thin.
Abactinal surface granular. Genera — Tropidaster,
Forbes ; (?) Compsaster, Worthen & Miller. The name
of the type genus was suggested from the apparent
occurrence of a keel along the abactinal side of the
rays. This, however, is only due to the ambulacral
plates being exposed by the loss of part of the
granular abactinal integument. Compsaster has a
narrow ambulacral groove, and is bordered by
imbricating actinal plates ; its general characters Wenlock Limestone,
resemble Tropidaster, but it is insufficiently known for its affinities to
be definitely settled.

FAMILY 4. LINCKIIDAE. Cryptozonia with opposite ambulacral
ossicles, comparatively well-developed marginal plates. Disc small, with
long cylindrical rays. Abactinal skeleton tessellate with granular integu-
mentary deposits. SUB-FAMILY 1. ROEMERASTERIXAE, with spines on
the marginal plates and disconnected granular plates arranged in longi-
tudinal series on the abactinal surface and sides of the arms. Genus —
Roemeraster, Stiirtz. SUB-FAMILY 2. LINCKIINAE, with abactinal plates
devoid of internal supplementary plates. Abactinal and marginal plates
granulose and not bearing spines. Genera — Ferdina, Gray ; Fromia, Gray ;
Leiaster, Peters ; Linckia, Gray; Narcissia, Gray; Nardoa, Gray; Ophidi-
aster, Ag.; Pharia, Gray; Phataria, Gray; (?) Arthraster, Forbes. SUB-
FAMILY 3. CHAETASTERINAE, with internal supplementary plates and
paxilliform tabulae. Genus — Chaetaster, Mull. & Tr. In this genus
there are remarkable groups of spines borne on the ends of disc-like
pillars, rising from the external plates ; these are known as paxilliform
tabulae. SUB-FAMILY 4. METRODIRINAE, with the marginal and abactinal
plates covered by membrane. There are neither internal supplementary
plates nor paxilliform tabulae. Genus — Metrodira, Gray.

FAMILY 5. STICHASTERIDAE. Cryptozonia with opposite ambulacral
ossicles and contingent marginal plates. Disc small ; rays long and
cylindrical. The plates of the abactinal surface are large, closely packed,
and regularly arranged. SUB-FAMILY 1. STICHASTERINAE. The adambu-
lacral plates are equal, have a simple armature, and have no ridges.
Genera — Calycaster, Per. ; Coelasterias, Verr. ; Neomorphaster, Slad. ; Stick-
aster, Mull. & Tr. ; Tarsaster, Slad. ; Tonia, Gray. SUB-FAMILY 2.
ZOROASTERINAE. The adambulacral plates are unequal, and their armature
is complex ; alternate plates have ridges. Genera — Cnemidiaster, Slad. ;
Mammaster, Per. ; Pholidaster, Slad. ; Prognaster, Per. ; Zoroaster, Wyv.
Thorns.

FAMILY 6. SOLASTERIDAE. Cryptozonia with opposite ambulacral
ossicles, and a reticulate, abaCtinal skeleton bearing paxilliform groups of

256 THE STELLEROIDEA

spines. Marginal plates obscure. No pedicellariae. Genera — Crossaster,
Mull. & Tr. ; Ctenaster, Per.; Lophaster, Verr.; Rhipidaster, Slad.; Solaster,
Forbes (Fig. XL).

FAMILY 7. KORETHRASTERIDAE, Dan. & Kor. Characters — " Crypto-
zonia allied to the Asterinidae, but distinguished by the complete absence
of interbrachial spaces, and by the possession of a continuous, calcareous
plating, and the formation of the paxillae" (Bell, 4, p. 23). Genera —
Korethraster, Wyv. Thorns. ; Peribolaster, Slad. ; Remaster, Per.

FAMILY 8. PTERASTERIDAE. " Cryptozonia in which the dorsal ossicles
carry a spine crowned by long diverging spines which support a more
or less well-developed membrane ; this forms a marsupial recess for the

Fin. XI.
Abactinal surface of Solaster pappoaus.

young. No actinal intermediate plates, interbrachial septa, or pedicel-
lariae" (Bell, 4, p. 23). In this family the most remarkable feature is the
development of a large marsupium in which the young are reared. This
is formed by a large veil above the abactinal surface, from which it is
raised by numerous long paxillae. The dorsal membrane is perforated by
many small pores (" spiracula," Sars), and has a large central opening, the
"oscular orifice." In some genera, e.g. Pttraster, there are also openings to
the actinal surface, known as " segmental apertures." There are often
long spine?, attached to the rays close by the adambulacral plates ; these
are known as the actino-lateral spines, and they are either enclosed in the
membrane of the actinal surface or in a marginal web. SUB-FAMILY 1.
CHEIROPTASTERINAE, with alternate ambulacral ossicles and short actino-
lateral spines. (The presence of the marsupium is not certain.) Form

THE STELLEROIDEA 257

pentagonal. Genera — Cheiroptaster, Stiirtz ; Loriolaster, Stiirtz ; (?) Rho-
palocoma, Salter. SUB-FAMILY 2. PTERASTERINAE, with a well-developed
supra-dorsal membrane, opposite ambulacral plates, and actino-lateral
spines. Form pentagonal. Genera — Benthaster, Slad. ; Caly piaster, Slad. ;
Cryptaster, Per. ; Flexaster, Per. ; Hymenaster, Wyv. Thorns. (Fig. XII.) ;
Marsipaster, Slad. ; Myxaster, Per. ; Stemster, Mull. & Tr. ; Retaster, Per.
SUB-FAMILY 3. PYTHONASTERINAE, with stellate form, opposite ambulacral
ossicles, and rudimentary marsupium formed by five triangular fan-like
valves ; no segmental apertures or actino-lateral spines. Genus — Mayr-
aster, Per. ; I'ythonaster, Slad.

FAMILY 9. PALASTERISCIDAE. Cryptozonia with the ambulacral
ossicles alternate for at least part of the arms. The madreporite is

FIG. XII.

Hymenaster jvlhtddvs (after Wyv. Thomson).

large and ventral in position. Actino-lateral spines are present. The
dorsal integument is granular. The form is stellate with small inter-
brachial areas. Genera — Palasteriscus, Stiirtz ; Echinasterella, Stiirtz.
This family is remarkable for the abnormal position of the madreporite,
which, unlike that of recent Asteroids, is ventral in position. This char-
acter is possibly due to the development of a granular integument over
the whole of the abactinal surface. Large spine-like paxillae occur, and
it is quite possible that the granular integument was the roof of a large
marsupium. There can be no question as to the position of the madre-
porite, for the actinal and abactinal surfaces of the same specimen are
shown in examples in the British Museum.

FAMILY 10. ECHINASTERIDAE. Cryptozonia with a reticulate abac-
tinal skeleton of small imbricating plates. Ambulacral ossicles opposite ;
the pores biserial ; oral armament adambulacral. Interbrachial septa

17

258 7V7.fi: STELLEROIDEA

single if present. SUB-FAMILY 1. ACANTHABTERINAE, with large disc and
numerous rays; numerous madreporites. Genus — Acanthaster, Qerv.
SUB-FAMILY 2. MITHRODIINAE, with a small disc, usually five rays. One
madreporite ; no interbrachial septa. Armed with large scaly spines.
Genus — Mithrodia, Gray. SUB-FAMILY 3. ECHINASTERINAE, with a small
disc and five or six rays. Spines small and simple. No pedicellariae.
Genera — Dictyaster, Mason & Alcock ; Echinaster, Mull. & Tr. ; Henricia,
Gray (syn. Cribrella, Ag.) ; Perknaster, Slad. ; Plectaster, Slad. SUB-FAMILY 4.
VALVASTERINAE, with the marginal plates bearing large valvate pedi-
cellariae. The disc is moderate in size, and there are five rays. Genus —
Valvaster, Per.

FAMILY 11. HELIASTERIDAE. Cryptozonia with opposite ambulacral
ossicles and double interbrachial septa. The disc is large and bears
very numerous short rays. SUB-FAMILY 1. HELIANTHASTERINAE. The
abactinal skeleton is granular in the disc and bases of the arms ; at the
arm tips it is tessellate, but the plates not in contact. Arms separate
at their bases, so that the two sets of plates which form the interbrachial
septa are separated by parts of the disc. Infero-marginal plates occur
round the disc. Madreporite marginal. Genus — Helianthaster, F. Roem.
This includes two Devonian species, of which only one is adequately
known, and is of somewhat uncertain affinities. It was originally
regarded as an Asteroid, but Stiirtz, in his latest description of its
anatomy, referred it to the Euryalidae. Stiirtz's specimens, now in the
British Museum, show nothing to separate them from the Asteroidea, of
which however they are very abnormal representatives. The ambulacral
plates are thin and L-shaped, but not crowded as in Heliaster, while the
abactinal skeleton is different. The species appears, however, to be a
primitive form of the Heliasteridae, in which the arms do not occupy the
whole margin of the disc, and are separated from one another through-
out; the ambulacral plates are not crowded, so that the podia are
biserial. SUB-FAMILY 2. HELIASTERINAE, with reticulate, abactinal
skeleton and arms in contact with one another at their bases, there being
no interbrachial spaces. The ambulacral ossicles are crowded and the
pores quadriserial. Genus — Heliaster, Gray.

FAMILY 12. PEDICELLASTERIDAE. Cryptozonia witn opposite ambu-
lacral ossicles, a small disc and narrow sub -cylindrical rays. Podia
biserial. The abactinal skeleton consists of narrow plates forming a
quadrangular network. Genera — Coronaster, Per. ; Gastraster, Per. ;
Lytaster, Per. ; Pedicellaster, Sars.

FAMILY 13. ASTERIIDAE. Cryptozonia with opposite ambulacral
ossicles. Podia quadriserial. Abactinal skeleton reticular and composed
of small unequal plates. Genera (including sub-genera of Asterias) —
Anasteriai, Per. ; Asterias, Linn. ; Calvasterias, Per. ; Coscinasterias, Verr. ;
Cosmasterias, Slad. ; Diplasterias, Per. ; Hydrasterias, Slad. ; Leptasterias,
Verr. ; Podasterias, Per. ; Polyasterias, Per. ; Pycnopodia, Stimps. ; Scler-
asterias, Per. ; Smilastei-ias, Slad. ; Sporasterias, Per. ; Uniophora, Gray.
The anatomy of Asterias is described on pp. 241-245.

FAMILY 14. BRISINGIDAE. Cryptozonia with opposite ambulacral
ossicles ; marginal plates absent or rudimentary. Rays numerous and

THE STELLEROIDEA 259

sharply marked off from the disc. Aboctinal skeleton absent, or present
only on the ovarial regions at the bases of the arms. No intermediate
actinal plates or interbrachial septa. Genera — Brisinga, Asbjornsen; (?)
Brisingaster, Lor. ; Colpaster, Slad. ; Freyella, Slad. ; (?) Gymnobrisinga, Stud.;
(?) Hymenodiscus, Per. ; Labidiaster, Lutken ; Medusaster, Stiirtz ; Odinia,
Per. This remarkable family was originally founded for the genus Brisinga,
which has many primitive characters. The arms are small and sharply
marked off from the disc. There are no ampullae connected to the podia ;
the generative organs consist of a series of small isolated glands along the
arms. The genus was accordingly at first regarded as very primitive in
structure and affording in some ways a link with the Ophiuroids. Later
authors, however, such as Ludwig and Sladen, entirely repel this view
and regard the Brisingidae as allied to the Asteriidae, and extremely
specialised rather than primitive. Sladen concludes, " In my opinion
the Brisingidae are true cryptozonate Asterids, very nearly related to the
Asteriidae, Pedicellasteridae, Heliasteridae, and Echinasteridae, and prob-
ably derived from a common ancestor, the divergence of form and the
peculiarities of structure now exhibited by Brisinga being the result of
modification produced by the extreme isolation and the exigencies of
the abyssal depths in which the family has existed" (48, p. 593). But
in Colpaster and Freyella the genital glands are limited to swellings
at the base of the arms ; and although the arms are sharply marked
off from the disc, at least six-sevenths of the arm has no extensions of
either alimentary canal or generative organs. The arm-ossicles of Freyella
tuberculata are identical in character with those of Ophiurina, and they
differ from Ophiogeron only by the absence of adambulacral ossicles.
Sladen gives no diagnosis of the class Asteroidea, so that it is not
quite clear on what characters he would base the separation of Asteroids
and Ophiuroids. But no known diagnosis of the Asteroidea would
include Colpaster and Freyella, and exclude forms universally admitted to
be Ophiuroids. If the more primitive types of Brisingidae are a recent
degenerate offshoot from the Asteriidae or some allied family, then a type
of structure, practically indistinguishable from that of the Ophiuroidea,
has been twice independently evolved. It seems therefore that Perrier is
probably correct when he regards the Brisingidae as the most primitive
instead of as the most special of living Asteroids.

The following fossil Asteroids are not sufficiently known for determina-
tion : —

Calliaster, Trautschold ; Coelaster, Sandb. non Agass. ; Cribrellites, Tate ;
Cupulaster, Fritsch ; Plumaster, Wright ; TricJwtaster, Wright.

SUB-CLASS 2. OPHIUROIDEA.

The Ophiuroidea form a sub-class of the Stelleroidea, including
the sand-stars, brittle stars, and branching stars. The typical
members of the class differ from the typical Asteroids by having
the arms sharply marked off from the disc as appendages, and by
not having a groove along the ventral side of the arms. These

26o THE STELLEROIDEA

differences are so important, anatomically, that the Ophiuroidea
and Asteroidea are often regarded as distinct classes. Owing to
the general external resemblance between the two groups, the first
naturalists who described them made no attempt to separate them.
Thus Linck, who in 1733 (23) figured several Ophiuroids,
included most of them, along with some Asteroids, in his genus
Stella. The common British species, Ophiura ciliaris, Linn, sp., he
named Stella lacertosa, and Ophioderma longicaudu he described as
Stella lumbricalis longicauda. He, however, recognised that his group
" Stella " must be broken up into several divisions, for he separ-
ated Ophiothrix frayilis, 0. F. Miiller sp., under the name of Ilosuht,
and the branching forms under the name of Astrophyton. Linnaeus,
on the other hand, did not grant the Ophiuroids even generic
distinction, and included them all in the genus Asterias. It
was not until 1816 that Lamarck (21) definitely founded the
genera Ophiura and Euryale. De Blainville in 1834, L. Agassiz in
1836, Dujardin in 1840, and Muller and Troschel in 1844, added
greatly to the systematic knowledge of the Ophiuroids, which they
retained in the order Stellerida. The first proposal to separate
the group as one of the primary divisions of the Echinoderms was
made by Forbes in 1840 (11); he founded the order " Spinigrada "
for the Ophiuroids, while the Asteroids he named the " Cirrigrada."
Gray, in 1840 and 1848, kept the two groups as separate orders,
but united them in one class, Hypostoma. Since this date the
view that the Ophiuroids are a distinct class has been widely
adopted.

The sub-class includes over one hundred genera, most of which
are recent. The earliest fossil forms occur in the Ordovician, and
representatives are known from all later systems.

The recent species have been described by many authors,
especially by Bell, Diiben, Forbes, Grieg, Grube, Heller, de Loriol
le Fort, Ljungman, Liitken, Lyman, Marenzeller, v. Martens,
Mortensen, J. Muller, W. Peters, G. 0. Sars, Sladen, Stimpson,
Studer, and Verrill. Lyman's report on the Cliallenger Collection
is a complete synopsis up to 1882 (31).

The fossil Ophiuroids are rare, and, as a rule, badly preserved.
The literature consists mainly of the description of isolated
specimens, as by E. Billings, Bohm, Forbes, S. A. Miller, Pohlig,
C. F. Roemer, H. Woodward, Worthen, and Wright. Stiirtz and
Roemer have described the remarkable fauna from Bundenbach in
Germany, and Salter that from the English Silurian. References
to the American Palaeozoic species are given in Miller's North
American Geology and Paleontology.

The first classifications of the Ophiuroids were artificial ; the
basis of a natural arrangement was laid by Ljungman in 1867
(24). Previous to that date the group had been divided into

THE STELLEROIDEA

261

two families — the Ophiuridae, including the forms with simple
arms, and the Astrophytidae, the members of which have branched
arms. Raised to sub-orders and divided into families, these two
divisions have long survived. In 1892 Bell (3) proposed a classi-
fication into three orders, to which a fourth has been added to
include some fossil forms (Gregory, 15).

The first important contributions to the anatomy of the
Ophiuroids were an account of the structure of the Euryaleae by
L. Agassiz, and of the skeleton of various genera by Miiller
( 1 854). Further researches have been carried out by Cue'not (1888)
and Hamann (1889). The two authors, however, to whom we are
most indebted are Lyman, who, in a long series of papers, has
described the skeletal structures, and Ludwig, whose masterly

FIG. XIII.

Ophinpholis aculeata, abactiual surface.

memoirs gave the first detailed account of the visceral anatomy.
Genera of exceptional interest have been described by Bell,
Ludwig, Simroth, and Sladen, to whom we owe memoirs
respectively, on Ophioteresis, Trichaster, Ophiadis, and Astrophiura.
The development was first studied by Miiller and Krohn (1851);
Ludwig in several memoirs, and Apostolides (1882), Russo (1891),
Cue'not (1892), and Bury, have studied additional forms with modern
methods; while MacBride, in 1896, has shaken, if not destroyed,
faith in the theory of the homology of the "calycinal" plates
(p. 14) as taught by Lov4n, Carpenter, and Sladen.

The body of an Ophiuroid, like that of an Asteroid, consists of
a central disc, from which radiate several (generally five) arms (Fig.
XIII.). The disc, however, is not formed, as it is in Asteroids, by

262 THE STELLEROIDEA

the union of the bases of the arms, but is sharply marked off
from them, and they are attached to it as appendages ; there
is not, moreover, in the Ophiuroids the ventral groove of the star-
fish. These characters, however, are not absolutely to be relied
on ; thus in some species of Astroschema there is no sharp separa-
tion between the arms and the disc ; while in Ophioteresis (Fig.
XIV.) the radial ambulacral vessels and nerve-trunks lie in shallow
grooves on the ventral surface of the arms.

An idea of the Structure of a typical Ophiuroid may probably
be best obtained by the careful examination of a representative
species, for which purpose the commonest English brittle star
(Ophiura ciliaris, Linn, sp.) is a convenient type.

This Ophiuroid consists of a round, flat, scale-covered disc, from
which radiate five long, tapering arms. The Arms are composed
of a series of jointed segments, each containing six plates. Two
of these are fused together into a single " vertebral ossicle," and

Firs. XIV.

Ophioteresis (after Boll). Aboral surface of an nnu ossic e ; rf, the double dorsal arm-plates ;
o, articular cavities ; I, lateral arm-plates.

a series of these forms the axis of the arm. The remaining four
plates form an external tube round the vertebral ossicle. One
pair occurs at the sides, and is known as the lateral arm-plates or
shields. Another of the four plates lies above the vertebral
ossicle, and is accordingly known as the dorsal arm -plate (or
dorsal shield) ; the fourth lies on the lower surface of the arm,
and is accordingly known as the ventral arm -plate (or ventral
shield). Each lateral arm-plate bears seven short spines.

The plates forming the central chain of the arm are known
as " vertebral ossicles," because, in typical Ophiuroids such as
Ophiura ciliaris, they articulate by a series of knobs and sockets
like the bones of a vertebral column. In a typical vertebral
ossicle the two articular surfaces are very different ; in the proximal
or adoral surface (that nearer to the disc) the most conspicuous
features are the prominent central umbo (Fig. XV. u) and two broad
" lower muscle fields " (Fig. XV. l.m) at the two lower angles.
Above the umbo there is a narrow " upper canal furrow " (u.f), while
a corresponding " lower canal furrow " (/./) occurs between the two

THE STELLEROIDEA

263

muscle fields. Above the lower furrow there is a depression to
receive a prominence on the distal face of the adjoining ossicle, arid
on each side of the depression there is a small knob, which similarly
fits into depressions on the adjacent ossicle. On the distal or
aboral surface there is a deep " umbonal socket " (w.s), and also
the prominence and two hollows already referred to; the two

it.

1. 2.

FIG. XV.

Vertebral ossicle of Ophiura cilittris (after Miiller). 1, aboral surface; 2, adoral surface,
w, umbo; l.m, lower muscle field; u.f, upi>er, and /./, lower canal furrows; u.s, umbonal
socket ; c, canal for the podion (shown by removal of part of muscle Held on right side).

lower angles are occupied by broad expanded surfaces, under
which pass the tubes of the podia (the surface on one side in figure
is broken away to show the canal for the podion, c).

Passing from the arms to the Disc, the skeleton is seen to
consist of two sets of plates, one belonging to the external integu-
ment, and the other to the oral system.

The Oral Skeleton is complex. It may be most readily con-
ceived as resulting from the fusion of the elements of two segments
in each of the arms. The entire oral
skeleton surrounds the mouth, and con-
sists of as many segments as there are
arms. Each segment is roughly triangu-
lar in shape, the apex pointing inwards,
and being separated by the deep " buccal
fissures." The principal element in each
segment is a pair of " syngnaths " (Fig.
XVL), also known as oral angle-plates, FIO. xvi.

iaws, mouth-plates, scutella or alia, etc. ; syngnatiw of opu

7T*. . , , V , ' (after Miiller). j, jaw ; m.f, mouth

the name Syngnath IS Suggested for them, frame ; n.g, groove for circum-

aa thmr nrmeiat f\f twn narfc o-Pnprallv oesophageal nerve ring; p.d, pore

as tney COnSlSt OI tWO parts generally and depression for oral tentacle.

completely fused together. The larger

piece of the syngnath is the mouth frame (m.f), which unites with

the corresponding plate of the next segment across the buccal

264 THE STELLER01DEA

fissure. The smaller piece is the jaw, which unites with the jaw
of the next arm to form the angle of the oral segment. Each jaw
has a depression, in which rest the oral tube-feet, and is notched by
a groove for the circumoral nerve ring.

The union of two adjoining jaws is strengthened by a small
plate at the apex, known as the " jaw plate " or torus angularis.
This plate supports the teeth, of which, in Ophiura ciliaris, there are
five in each segment. A series of similar processes occur along
the side of the jaw, projecting into the buccal fissure ; these are
the oral papillae (mouth papillae or buccal papillae), of which, in
Ophiura ciliaris, there are ten or more in each series.

The angle between the two jaws of one segment is occupied by
a shield-shaped plate, known as the " buccal shield " (oral plate,
mouth-shield, or scutum buccale). These plates are interradial in
position ; the smaller plates corresponding to them, but radial in
position, are the first ventral arm-plates. Between the mouth-
frames and the buccal shields are five pairs of long, bar-shaped
" peristomial plates," which cross the interradial spaces from arm
to arm ; they cannot be seen from without, as they are covered
by the shields, and by two plates beside the latter, known as the
lateral buccal shields (or scAita adoralia).

The oral skeleton of the Ophiuroid is therefore very different
from that of the Asteroid. But both are formed by the modification
of similar parts, viz. the ambulacral and adambulacral ossicles of
the arm segments. In the Ophiuroid the supposed homologies of
the principal parts are as follows : —

Vertebral ( = ambulacra 1 Lateral arm-plates

ossicles). ( = adambulacral ossicles).

1st arm segment peristomial plates jaws

2nd „ mouth-frames lateral buccal shields.

The remaining elements in the skeleton of Ophiura belong to
the exoskeleton. On the upper or aboral side of the disc there are
five pairs of large and somewhat pyriform plates known as " radial
shields " (Fig. XXIII. r.s). Between them are a few smaller plates,
and the rest of the disc of 0. ciliaris is covered with small,
irregular, imbricating scales.

On the lower surface of the disc there are two pairs of long,
thin plates beside the bases of the arms. A narrow cleft, the
" bursal slit " (known also as the bursal aperture, genital slit,
genital cleft, and ovarian aperture), separates these plates. This
slit leads into the bursal cavity, into which the generative products
are discharged from the gonads. Of the two bar -like plates
bounding the bursal slit, the larger is the "genital plate," and the
smaller is the " genital scale."

The Alimentary Canal of the Ophiuroids is simpler than in

THE STELLEROIDEA

265

either Asteroids or Echinoids. The mouth is situated in the centre
of the lower surface of the disc, and from it the buccal fissures
radiate. The mouth opens to a short oesophagus, above which
is a large digestive sac, occupying nearly the whole cavity of the
disc. Neither anus nor diverticula along the arms are present.

The Water-vascular System consists essentially of a circum-
oesophageal ring which gives rise to five radial vessels bearing tube-
feet, and five vesicles, one of which communicates with the exterior.
In Ophiura ciliaris the circumoesophageal ring lies over the syn-
gnaths, through which pass branches (Fig. XVII. o.v) leading to

o.t. L

J;P-

r. 11.

Fio. XVII.

Diagram of circumoral region of Ophiura ciliaris (after Miiller). c.ce.n, circumoesophageal
nerve ring; c.ce.v, circumoesophageal water- vascular ring; e.n, nerve branch to integu-
ment; i.v.m, intervertebral muscles; i.v.n, intervertebral nerve; j, jaw ; j.p, jaw plate;
o.n, oral nerve; o.p, oral papillae; o.t, oral tentacle; o.v, branch of water vessel to oral
tentacle; p.n, podia! nerve; p.v, Polian vesicle; r.n, radial nerve; t, teeth; t.g, tentacular
groove; t.n, tentacular nerve ; r.o, vertebral ossicles.

the mouth-tentacles (o.t) ; radial vessels run along the under sides
of the arms. In each of four of the interradial (or interbrachial)
spaces there is a "polian vesicle" (p.v); in the fifth interradius there
is a short expanded stone-canal (Fig. XVIII. s.c) which opens to the
exterior by a single madreporite on a buccal shield. The radial
vessels lie in the lower canal furrows of the vertebral ossicles (Fig.
XV. If) ; branches from the radial vessel pass through the vertebral
ossicles; they emerge at the lower angle of the ossicle, and the
podia pass to the exterior through a space between the shields.
There are no ampullae, but the flow of water is regulated by

266

THE STELLEROIDEA

o.b.

FIG. XVIII.

valves. There are no suckers, so the podia are useless in loco-
motion.

The Nervous System of Ophiura dliaris consists of a circum-
oesophageal ring, from which radiate five radial nerve trunks (Fig.

XVII. c.ce.n and r.n) ; it also gives
oft' small branches to the oral ten-
tacles, teeth, and oral muscles. The
radial nerve trunks consist of two
separate nerve bands, one above the
other, both lying in the " epineural
canal " (Fig. XXI. e.r.n and i.r.n).
The two nerve bands are not con-
nected, though situated very close
to one another. Both radial nerve
trunks thicken into ganglion -like
swellings, of which there is one in
each arm segment. From the ganglia
of the upper or internal nerve band,
branches are given off to the muscles
between the vertebral ossicles.
From each ganglion of the lower
water-vascular ring of Ophiura dliaris. or external nerve trunk, two pairs

<>.b, branch to oral tentacles ; c.o.v, circuni- f • «• •

oesophagcal vessel; m, inadreporite ; p. r, Of IiervCS are given OH ; One pair

JlfSone'cSf. r'*' "*M "**" v**™1 ' supplies the tentacles, and the other

the integumont and spines.

The last element in the Ophiuroid nerve system consists of a
"genital nerve ring," which lies along the "aboral circular sinus."

Respiration in the Ophiuroids is effected only by the walls of
the bursae, and by the podia.

The Reproductive Organs consist of a series of small pear-
shaped gonads (Fig. XIX. y) which do not open directly to the
exterior, but into the bursae, where, in some cases, the development
takes place. In Ophiura there are more than forty gonads to each
bursa.

The Coelomic Sinus. There appears to be no true blood-
vascular system in the Ophiuroids ; the vessels which have been
described as such are connected with the axial sinus, and are there-
fore coelomic spaces and not vessels (MacBride, 32, and Russo).
The axial sinus (Fig. XX. x.s) is the most important ; it is the space
through which runs the stone-canal, and it contains a gland in con-
tact with the stone-canal, which is known as the " ovoid gland " or
"axial organ " (z.o), (see p. 23). From the upper end of the axial
canal runs the "aboral circular sinus." The course of this sinus
is very sinuous : above the arms it is aboral in position, but
beside the arms it bends downwards until, in the interradii, it
is quite ventral in position. The main sinus passes on the inter-

THE STELLEROIDEA

267

radial side of the bursae, and sends off a branch along the radial
side of each bursa. The parts of the sinus beside the bursae bear
the gonads.

Fio. XIX.
Gonads (</) and bursae (6) of Ophioglypha albida. (After Ludwig.)

Arm Structure. As we have seen, there are no prolonga-
tions of the digestive or reproductive organs in the arms,
while branches of the nervous and water -vascular systems run
along the ventral side between , s

the vertebral ossicles and the — !__

ventral arm-shields. It follows
therefore that the vertebral
ossicles are directly supra-
ambulacral, and are homologous
with the ambulacral ossicles,
and not with the supra-ambu-
lacral ossicles of such Asteroids
as Astropecten. A transverse
section across an Ophiuroid arm

is Shown in Fig. XXL (cf. p. 1 5). Diagrammatic section through oral region of

Ophiuroid (after MacBride). amp, ampulla ; 1>,
^. , . bursa ; i.m, interradial muscle of the disc ; 0.s,

From SUch a type as Uplliura genital sinus ; g.r, genital rhachis ; m, mouth ;

•7 • • • , i ., . i n , m.p, madreporic pore ; n.r, circumoesophageal

CUtam, Wltll Its SmOOtn, nat, nerve ring . pmtt peristomal sinus ; st.c, stone

circular disc, its irregular, scaly ^•g^^fS^1^^'^
plates, its comparatively short,

straight, unbranched, and regularly tapering arms, there are many
striking deviations, both in aspect and structure.

The Exoskeleton may be either more or less complete than in

268

THE STELLEROWEA

Ophiura. The most important addition to the plates found in that
genus is formed by the plates of the so-called "apical system."
In some species, such as Ophiomusium validum, there is a central plate

br.c.

-pa.

FIG. XXI.

Diagrammatic section through arm of Ophiuroid. ftc.c, dorsal branch of brachial coelomic
Kystem ; d.s, dorsal arm-plate; e.r.n, external radial nerve; i.r.n, internal radial nerve; La,
lateral arm-plate ; pd, portion ; pd.n, podial nerve ; r.e.c, radial epinenral canal, and r.p.c, radial
pseudhaemal canal, together forming ventral branch of the brachial coelomic system ; r.v, radiul
vessel ; v.o, vertebral ossicle ; v.s, ventral arm-plate.

surrounded by three circlets, each of five plates ; these were once
regarded as homologous with plates in Asteroids and Oinoids, and
therefore named the infra- basals, basals, and radials. In some
species, such as Ophiomitra exigua, the two circlets of radially situated

plates are absent, and only the
central and basal plates are
present. In Ophiocrene and
Ophiura convexa, Lym. sp., there
are only the central plate and
the radials; in Ophiura minuta,
Lym. sp., there are the central,
radial, and basal plates. The
last possible combination occurs
in a species of Pectinura (Fig.
XXII. ), in which the two circles
of plates round the central have
been described as the radials
and infra-basals. In some other
forms, such as Ophioceramis ob-
stricta, Lym. (Fig. XXIII.), there
is a remarkable uniformity in the
plates on the dorsal aspect, but
these are not arranged on the
calicular plan. In some other cases, such as Ophiura inornata, Lam.
sp. (Fig. XXIV.), there are two " infra-basal " plates side by side

Fio. XXII.
Dorsal plates of Pectinura, sp. (After Bell.)

THE STELLERO1DEA

269

below each radial. The genus in which these supposed calycinal
plates most resemble a Crinoid cup is the genus Ophiopyrgus

Via. XX lit.

Abactinal plates of Ophioccrmnis olmtrii-ta, Lyin. (after Lymau). c, central plate ; r.it, radial
shields ; il.a.p and l.a.p, dorsal and lateral nnu-|>lntes.

(Fig. XXV.), in which the dorsal side is raised into a conical
projection.

The irregularity of these " calycinal plates " both in develop-

FI.I. XXIV.
Abactinal plates of disc of Ojthiwa inornata (Lain.), c, central, r.n, radial shields.

ment and arrangement discredits their supposed homology with
the three circlets of plates in the calyx of a Crinoid. It is only
natural that among the many variations in the grouping of the

270

THE STELLEROIDEA

Fio. XXV.

Ophinpyrgus, seen from the side (after Lyinan).
c, central, b, basal, r, radial plates.

dorsal plates of Ophiuroids, that one or more pentamerous rings
should be more conspicuous than the rest.

In the previous cases the dorsal plates of the disc are more pro-
minent than in Ophiura. In some other Ophiuroids, however, these

plates are less important.
Thus the radial shields may
be buried beneath the in-
tegument, or may be absent
altogether, as in Neoplax ;
the remaining plates may
also be reduced to scales
or small granules embedded
in a soft integument, as in
the Gorgonocephalidae.

Variations in the Arm
structure offer important
indications as to the affinity
between Ophiuroids and As-
teroids. The differences be-
tween such types as Ophiura
and Asterias are very great ;
but many intermediate
stages occur between these
extremes ; and in the case of some extinct genera it is doubtful
to which sub-class they belong.

The vertebral ossicles of some existing deep-sea Ophiuroids, such
as Ophiogeron (Fig. XXVI.) or Ophiohelus (Fig. XXVII.), consist of
two separate bars similar to the ambulacral ossicles of Asteroids. The
same arrangement holds in some fossil species, such as Lapworthura
Miltoni. A further approximation to the arm struc-
ture of Asteroids occurs in some genera from which
the ventral arm-plates are absent. The radial water-
vascular vessels then lie along open grooves, as in the
Asteroids. This condition is found among recent
Ophiuroids in the genus Ophioteresis, and among fossil
members of the class in the extinct Lysophiurae.

In such cases it is generally easy to recognise that
the lateral arm-plates are homologous with the adam-
bulacral plates of Asteroids, a point also well shown by
the living Ophioteresis (Fig. XIV.)

The lateral arm-plates, though often smaller than
the ventral or dorsal arm-plates, are morphologically
the most important. They are always present, except when the
arm is covered by a soft or granular integument, as in the
Cladophiurae and some genera of Streptophiurae.

The dorsal arm-plates vary more than either of the other sets.

l.a.p

Fio. XXVI.

piate;

pate; i.a.p, iat-
^"1 "p

THE STELLEROIDEA

271

They are double in Ophioteresis and Ophiomyxa pentagona, are split up
into several pieces in Astroschema (Fig. XXVIII.), and into a mosaic
of small plates in Hemieuryale ; or they are absent altogether from
many genera, such as Neoplax, Ophiobyrsa, Protaster, Lapworthura.

d.a.p.

La.p

Fio. XXVII.

Anu-plates of Ophiohelns umbrella, l.a.p, lateral anu-plates ; d.a.p, dorsal arm-plates ;
a, ambulacral plate ; s, spines.

Another character by which the recent Asteroids and Ophiuroids
differ is the alternation of the ambulacral ossicles in the former,
whereas in the latter the pairs are opposite. The Palaeozoic
genera, however, also bridge this gap, as
in the order Lysophiurae the ambulacral
ossicles are alternate as in the Asteroids.
This is shown in the arm structure of
Palaeophiura.

While Protaster and its allies on the
one hand approximate to the Asteroids,
another group of Ophiuroids has arms which
appear more like those of Crinoids. In
Ophiura the arms are fairly straight and
only capable of a limited movement in a
horizontal plane, for the vertebral ossicles
are " zygospondyline," i.e. are attached to
one another by a series of knobs, closely
fitting into sockets (e.g. Fig. XV.) In the group of the Strepto-
phiurae much more play of the arms is possible, as the ossicles are
" streptospondyline," i.e. articulate by simple ball-and-socket joints,

FIG. XXVIII.

The external arm-plates of a
segment of Astroschema (after
Lyman). v.o, ventral arm-
plates; v.a.p, lateral arm-plates;
the remaining six plates re-
present the dorsal arm-plates.

272

THE STELLEROIDEA

without lateral pits and processes (Fig. XIV.) ; while in the Clado-
phiurae the ossicles articulate by hour-glass shaped surfaces (Fig.
XXIX.), and the four external shields are replaced by a soft,
granular integument ; in this group the arms are therefore capable
of movement in any direction, the arms frequently being able to
coil round any support, as in Astrosckema. A further change is

FMJ. XXIX.
Articular surfaces of vertebral ossicles of A*lm*i-liriint.

introduced by the branching of the arms ; this occurs in two
groups of the Cladophiurae, viz. in the Trichasteridae, in which
the arms branch a few times at their free ends, and in the Gorgono-
cephalidae, in which the arms branch repeatedly (Fig. XXXII.).

The Oral Skeleton of most Ophiuroids is on the same plan as
that of Ophiura; the details however differ considerably. The
nature of the parts is shown very clearly in some Palaeozoic
genera, such as Stnrtzura. The apparatus here consists of a
syngnath, apparently composed of the plates of only a single arm
segment ; one of the plates of the central
vertebral pair is elongated to a bar, and forms
the mouth-frame, which is therefore clearly
ambulacral. The jaw plate attached to this
mouth-frame is the adambulacral plate of the
same segment, while the teeth upon this are
homologous with spines.

In many recent Ophiuroids there is an
element in the armature in addition to those
of Ophiura ; this is the set of " dental papillae "
which are situated at the oral ends of the jaws
and project toward the mouth above the
teeth.

I" Ophiura and its allies there are no
Pedicellariae, but a very primitive type of
., them occurs in tne Cladophiurae, as in Tri-

and retractor 'muscles, chaster (Fig. XXX.).

Turning to the internal anatomy we find the

most important changes in connection with the Water -vascular
System. In many genera of the Cladophiurae there is more than one
madeprorite, of which there may be one in each interradius. In

FIO xxx

Pediceiiaria of 'in-

THE STELLEROIDEA 273

Trichaster the stone-canals are thus repeated, but there is only one
madreporite. A repetition of the stone -canals also occurs in
Ophiactis, in which the value of this character is apparent, for the
animal reproduces by fission ; in the same genus there are also
several Polian vesicles in each interbrachial space. In development
the water-pore of recent Ophiuroids originally opens on the dorsal
surface (Bury, 6, pp. 422, 423, pi. xxxvii. f. 2), and then works round
to the ventral side, where it becomes attached to an oral plate.

The Alimentary System remains very simple in all the Ophiuroids,
consisting simply of a large chamber, divided into a series of short,
blunt, sac-like protuberances by radial constrictions of the walls.

Respiration in the Ophiuroids is generally effected by the
genital bursae and the podia ; but when the bursae are absent,
their place may be taken by an extra series of Polian vesicles, as
in Ophiactis ; or the general body-cavity may be used both for the
protection of the ova and for respiration, as in Gorgonocephalus.

Reproduction. — The genital bursae in some Ophiuroids also act
as brood chambers ; the eggs pass through all stages of development
in them, and such Ophiuroids are therefore viviparous ; Amphiura
squamata and Ophiomyxa vivipara are examples of this condition.
Asexual reproduction occasionally occurs in Ophiuroids either
normally by fission, as in Ophiactis, or abnormally by regeneration
of lost parts when the disc of an Ophiuroid is cut into halves.
Broken arms are readily replaced, but a broken arm cannot
reproduce a complete animal as can be done in the Asteroids
(Semon and Ludwig).

The Development of the Ophiuroids agrees in the early stages
with that of the Asteroids, but the larval form is a Pluteus (p. 6)
and not a Bipinnaria (p. 5). It, therefore, in this stage offers a
greater resemblance to the larvae of the Echinoids than of the
Asteroids. There is no doubt, however, that this larval form is
a secondary development and does not represent any stage in
phylogeny of the group ; it therefore does not indicate any
affinity with the Echinoids.

Distribution. — The Ophiuroids range from shallow and estuarine
waters to abyssal depths. Their distribution in space is wide,
species such as Ophiocten sericeum occurring at 80° N. latitude ; but
the largest forms are tropical. The order was first represented by
species of Protaster from Ordovician rocks. Representatives of the
Lysophiurae and Streptophiurae occur in the Silurian, Devonian,
and Carboniferous strata ; Zygophiurae begin in the Trias ; and
Cladophiurae in the Jurassic.

On the characters discussed in the preceding pages may be based the
following Diagnosis of the Sub-Glass : J — The Ophiuroidea are " eleuther-
ozoic," " actinogonidial," and " lysactinic," Echinoderms which usually
1 Emended from Bell (4), p. 215. The terms are explained antea, p. 287.

18

274 THE STELLEROIDEA

have no ambulacral groove. The arms are generally sharply marked off
from the disc, are generally five in number, and are sometimes elaborately
branched. The digestive system, which is aproctous, and the generative
system are both confined to the disc ; so also is the special respiratory
apparatus which takes the form of deep clefts.

This diagnosis at once indicates that the Ophiuroids are more nearly
allied to the Asteroids than to other Echinoderms, for both classes are
actinogonidial, eleutherozoic, and lysactinic. Moreover, neither of the
main characters which separate the two classes hold in all cases ; for in
Ophioteresis and Protaster there is a ventral groove, and in some species of
Astroschema the arms pass gradually into the disc. Similarly the
Asteroids of the family Astropectinidae have no anus ; and the Ophiuroids,
Gorgonocephalus and Ophiactis, have no genital bursae.

ORDER 1. Lysophiurae, Gregory (1897).

Ophiuroidea in which the ambulacral ossicles are alternate, and are
not united into vertebral ossicles, but those of each segment are separate.
There are no ventral arm -plates, and the ventral side of the arm is
occupied by an ambulacral furrow.

This order includes a group of fossil Ophiuroids, in which the arm
structure is asteroid in plan, for there are no ventral arm-plates; there is
an ambulacral groove ; and the ambulacral plates are in alternate pairs.
The members of the order differ from the Asteroids by having the arms
sharply marked off from the disc, and the alimentary canal was doubtless
confined entirely to the disc.

FAMILY 1. PROTASTERIDAE. Lysophiurae with boot-shaped ambulacral
ossicles, each composed of a " body " in the median line of the arm, and a
lateral "wing" at right angles to it. Genera — Protaster, Forbes, and
Bundenbachia, Stlirtz. There is a well -marked scale -covered disc and
five flexible arms. The axial portion or " body " of each ambulacral
ossicle is marked off into two parts by a depression which probably served
for the attachment of powerful ventral muscles. Stout adambulacral
plates occur in all the members of the family, and support lateral spines.

FAMILY 2. PALAEOPHIURIDAE. Lysophiurae in which the ambulacral
ossicles consist of a bar -shaped or subquadrate "body" without wings.
Genera — Sturtzura, Greg. ; Eugaster, Hall ; Ptilonaster, Hall ; Taeniura,
Greg. ; all Silurian ; and the Devonian Palaeophiura, Stiirtz. This family
is characterised by its alternate rod-shaped ambulacral ossicles, shown in
Palaeophiura lymani, Stiirtz.

ORDER 2. Streptophiurae, Bell (1892).

Ophiuroidea in which the ambulacral ossicles are opposite, and generally
united to form vertebral ossicles. In such cases the vertebral ossicles
articulate with one another by means of a more or less simple ball-and-
socket joint. The covering plates are more or less regularly developed as
superior, inferior, and two lateral, the last of which bear spines.

THE STELLEROIDEA

275

The diagnosis of this order is necessarily vague, as it includes
a series of simple forms, among which there are great divergences in
structure. One living genus included in this order has no ventral
arm-plates, but a small ambulacral furrow, and thus agrees with the
Palaeozoic genera, for which Stiirtz in 1885 proposed the family Ophio-
encrinasteriae. The main character of the order is the simple articulation
of the arm ossicles. The ambulacral plates are always opposite instead of
being alternate, as in the Lysophiurae. In some of the most primitive
genera the two ambulacral ossicles in an arm segment are separate, as in
Ophiurina, or incompletely joined, as in Ophiohelus. In the Lapworth-
uridae they are firmly united, and in the recent genera Ophiomyces and
its allies they are specialised into definite ambulacral ossicles, with a
simple ball-and-socket articulation.

FAMILY 1. OPHIURINIDAE. Streptophiurae without ventral arm-plates
and with separate ambulacral ossicles. Genera — Ophiurina, Stiirtz,
Devonian. Tremataster, Worthen & Miller.

FAMILY 2. LAPWORTHURIDAE. Strepto-
phiurae without ventral arm-plates or buccal
shields. Ambulacral ossicles fused, but their
articulating surfaces are plain. Genera — Lap-
worthura, Greg., Silurian, W. of England ; Fur-
caster, Stiirtz; Palastropecten, Stiirtz. The oral
armament is typically Ophiuroid. The struc-
ture in the genus Furcaster is shown in Fig.
XXXI. ; mouth-frames, small jaws, jaw plate,
and teeth are all present.

FAMILY 3. EOLUIDIDAE. Streptophiurae
with ambulacral ossi<ties united to form verte-
bral ossicles. Ventral arm -plates present,
but there are neither buccal shields nor
dorsal arm -plates. Genera — Eoluidia, Stiirtz ;
Eospondylus, Greg. ; Miospondylus, Greg. ;
Aganaster, Miller & Gurley (syn. Ophiopege,
Bohm) ; Cholaster, Worthen & Miller.

FAMILY 4. ONYCHASTERIDAE. Strepto-
phiurae with well-developed vertebral ossicles,

Fia. XXXI.

Furcaster palaeozoicus, Stiirtz,
showing skeletal elements of the
arms and oral armament, ad,
adambulacral ossicles bearing
spines ; am, ambulacral ossicles ;
j, jaw ; j.p, jaw plate ; m.f, mouth-
frame ; t, teeth. (After Stiirtz.)

and very flexible, unbranched arms. There
are no external arm -plates, the integument only containing granules.
Genus — Onychaster, Carboniferous, Illinois, has previously been included
among the Euryalid group, but its ainbulacral ossicles (as shown by Meek
and Worthen's later figures, Geol. Surv. Illinois, vol. v., 1873, pL xvi.
fig. 3d) are Streptospondyline.

FAMILY 5. EDCLADIIDAE. Streptophiurae with contorted branching
arms. Five pairs of large plates round the centre of the side exposed in
the fossil have been regarded either as jaws or as radial shields. Madre-
porite on same side as the plates. Arms have no external arm-plates but
a granular integument. The type-species, Eucladia Johnsoni, was care-
fully figured and described by H. Woodward in 1869. Owing to its
flexible, branched arms, and soft integument, it has generally been

276 THE STELLERO1DEA

regarded as an ally of Euryale, but some authors have urged its removal
to the Asteroidea, ascribing a dorsal position to the madreporite. It is
no doubt in external appearance more like the Cladophiurae than the
Streptophiurae, but the external resemblance to Euryale is probably due
to homoplastic modifications.

FAMILY 6. THE LIVING STREPTOPHIDRAE. No attempt has yet been
made to arrange the living Streptophiurae into families, and this cannot be
done with advantage without further knowledge about several points. Some
of the most remarkable genera are known only by single specimens,
which are very small and probably immature. Ophiohelus and Ophiotholia,
present a certain resemblance to the Lysophiurae in the structure of the
ambulacral ossicles. Ophiogeron, with its long, rod-shaped, ambulacral
ossicles lying in opposite pairs, is much like Ophiurina, but the evidence
available at present is insufficient to justify its inclusion in the Ophiuri-
nidae. Hence it is advisable at present to leave the living Streptophiurae
arranged according to Bell's scheme (3). When the classification is
attempted, probably Ophioteresis will form one family, and Ophiosciasma
another ; Hemieuryale, Sigsbeia, Ophwchvndrus, and Ophiomyces may con-
stitute a third.

A. No under arm-plates — Ophiotcresis, Bell.

B. Under arm-plates imperfect — Ophiosciasma, Lym.

C. Under arm-plates moderate or well developed.

a. No upper arm-plates — I. No radial shield — Neoplax ; Ophiohelus,
Lym.; Ophiotliolia, Lym. (?). II. Radial shields present —
Ophioscolex, Miill. & Trosch. ; Ophiambix, Lym. ; Ophiogeron,
Lym..; Ophiobyrsa, Lym. ; Ophiomyxa (pars), Miill. & Trosch.

/?. Upper arm-plates minute or formed of scattered plates — Ophio-
myxa (pars), Miill. & Trosch.; Ophiomyces, Lym. ; Ophiochondrus,
Lym. ; Hemieuryale, Martens ; Sigsbeia, Lym.

ORDER 3. Cladophiurae, Bell (1892).

Ophiuroidea in which the vertebral ossicles articulate with one another
by means of hour-glass-shaped surfaces (Fig. XXXIIL), and are covered
by granular deposits in the thick integument. The arms may be simple
or branched repeatedly.

FAMILY 1. ASTRONYCIDAE. Cladophiurae with simple unhranched
arms. GROUP 1. With large disc — Astrotoma, Lym.; Astromyx, Mull.
& Trosch. ; Astrochele, Verr. GROUP 2. Disc of medium size — Astrogomphus,
Lym. ; Astroporpa, Oersted & Liitken. GROUP 3. Disc small — Ophiocreas,
Lym. ; Astroschema, Oerst. & Liitk. ; Astroceras, Lym. FAMILY 2.
TRICHASTERIDAE. Cladophiurae with arms branching a few times near
their free ends. Genera, recent — Trichaster, L. Ag. ; Astroclon, Lym. ;
Astrocnida, Lym. Fossil — Astrocnida, Lym. FAMILY 3. GORGONOCEPHA-
LIDAE. Cladophiurae with arms branching much and from their base.
Euryale, Lam. ; Gorgonocephalus, Lym. ; Astrophyton, Ag. (Fig. XXXII.)
Ophiocrene, Bell.

THE STELLEROIDEA

277

ORDER 4. Zygophiurae, Bell (1892).

Ophiuroidea in which the movement of the ossicles on one another is
limited by the development of lateral processes and pits. Superior,

Fio. XXXII.

Astrophyton Llncki. Abactinal surface. (From Wyv. Thomson.)

The

inferior, and lateral spine-bearing arm-plates are always present
arms are simple and cannot coil round straight rods.

FAMILY 1. OPHIODERMATIDAE. Zygophiurae with oral papillae
numerous, and none infra-dental. Arm incisures on the disc. Dental
papillae absent. Genera— Ophioderma, Mull. & Trosch. ; Ophiopeza, Peters ;
Ophiarachna, Mull. & Trosch.; Ophiocoeta, Liitk.; Ophioconis, Liitk.; Ophio-
plax, Lym. ; Ophivyona, Stud. ; (?) Ophiopyrgus, Lym. ; Ophiopyren, Lym.

278 THE STELLEROIDEA

FAMILY 2. OPHIOLEPIDIDAE. Zygophiurae with oral papillae from
three to six, of which the last may be infra-dental. Arm incisures on the
disc. Dental papillae absent. Genera — Ophiolepis, Miill. & Trosch. ;
Ophioden, Liitk. ; Ophiura, L. Ag. ; Ophioglypha, Lym. ; Ophioceramis,
Lym. ; (?) Ophiochiton, Lym. ; Ophiopaepale, Lym. ; Ophiozona, Lym. ;
Ophioplinthus, Dan. ; Ophiernus, Lym. ; Amphiglypha, Pohl. ; Geocoma,
d'Orb.

FAMILY 3. AMPHIURIDAE. Zygophiurae with oral papillae from one
to five, of which the last is generally infra-dental. Arms inserted on
ventral side of disc. Dental papillae absent. Genera — Amphiura, Forbes ;
Ophiocnida, Lym. ; Ophiomusium, Lym. ; Ophiopeltis, Daniels. & Kor. ;
Ophiocentrus, Ljung. ; Amphilejns, Ljung. ; Ophiolepis, Lym. ; Ophiomartus,

FiO. XXXI II.
Abactinal view of Hemipholis cordifera, Lym. c, central plate ; r, plates of radial circlet.

Lym. ; Ophiophyllum, Lym. ; Ophiotrochus, Lym. ; Hemipholis, Lym. (Fig.
XXXIII.); Ophiactis, Liitk.; Ophiopus, Ljung.; Ophiopholis, Liitk. (Fig.
XIII.); Ophiacantha, Miill. & Trosch. ; Ophiotrema, Koehl.; Pcctinura, Heller
(non Forbes) ; Ophioplocus, Lym. ; Ophionereis, Liitk. ; Amphipholis, Ljung. ;
Ophiophragoniw, Lym. ; Ophiostigma, Liitk. ; Ophioblenna, Liitk. ; Ophio-
cymbium, Lym. ; Ophiochytra, Lym. ; Ophiolebes, Lym. ; Ophiomitra, Lym. ;
Uphiocamax, Lym. ; Ophiothamnus, Lym. ; Acronra, Ag. ; Aspidura, Ag. ;
Hemiylypha, Pohl. ; Polypholis, Dune.

FAMILY 4. OPHIOCOMIDAE. Zygophiurae with both oral and dental
papillae. Genera — Ophiocoma, L. Ag. ; Ophiomastix, Miill. & Trosch. ;
Ophiarthrum, Pet. ; Ophiopsila, Forbes ; Ophiopteris, E. A. Smith.

FAMILY 5. OPHIOTHRICIDAE. Zygophiurae with dental papillae, but
no oral papillae. Genera — Ophiothrix, Miill. & Trosch. ; Ophiocnemis, Miill.
& Trosch. ; Ophiogymna, Ljung. ; Ophionema, Liitk. ; Ophionephthys, Liitk. ;
Ophiomaza, Lym.; Ophiothela, Verr. ; Ophiopsammium, Lym. ; Ophiopteron,
Ludw. ; (?) Ophiurella, Ag.

THE STELLEROIDEA 279

THE LITERATURE OF STELLEROIDEA.

1. Agassiz, A. 1877. (North American Starfishes. ) Mem. Mus. Com p. Zool.-

Harvard, vol. v. No. 1, v. and 136 pp., xx. pis.

2. Apostolidts, N. C. 1882. (Anatomic et developpement des Ophiures.)

Arch. Zool. Exper., ser. 1, torn. x. pp. 121-224, pis. vii.-xii.

3. Bell, F. J. 1892. (A Contribution to the Classification of Ophiuroids, &c.)

Proc. Zool. Soc. for 1892, pp. 175-183, pis. xi., xii.

4. 1892. Catalogue of the British Echinoderms in the British Museum

(Natural History), 8vo, xvii. and 202 pp., xvi. pis. London.

5. Billings, K 1858. (On the Asteriadae of the Lower Silurian Rocks of

Canada.) Canad. Org. Rem., dec. iii. pp. 75-85, pis. viii.-x.

6. Bury, H. 1889. (Studies in the Embryology of the Echinoderms.) Quart.

Journ. Micro. Sci., N.S. vol. xxix. pp. 409-449, pis. xxxvii.-xxxix.

7. Carpenter, P. H. 1882. (Notes on Echinoderm Morphology. No. V. On

the Homologies of the Apical System, with some Remarks upon the
Blood- Vessels.) Quart. Journ. Micr. Sci., N.S. vol. xxii. pp. 371-386.

8. Ciienot, L. 1887. (Contribution a 1'etude anatomique des Asterides.) Arch.

Zool. Exper., ser. 2, vol. v. bis. suppl., 144 pp., ix. pis.

9. 1888. (Etudes anatomiques et morphologiques sur es Ophiures.)

Arch. Zool. Exper., ser. 2, vol. vi. pp. 33-82, pis. iii.-v.

10. Danielssen, D. C., and Korcn, J. 1884. The Norwegian North Atlantic

Expedition, 1876-78. Zoology, Asteroidea, 119 pp., xv. pis.

11. Forbes, E. 1840-41. A History of British Starfishes, and Other Animals

of the Class Echinodermata, 8vo, xx. and 270 pp. London.

12. 1849-50-56. Figures and Descriptions of British Organic Remains.

Decades i., iii., v.

13. Gaudry, A. 1852. (Memoire sur les pieces solides des Stellerides.) Ann.

Sci. Nat. Zool., se>. 3, vol. xvi. pp. 339-379.

14. Gray, J. E. 1866. Synopsis of the Species of Starfish in the British

Museum, 8vo. London.

15. Gregory, J. W. (The Classification of the Palaeozoic Echinoderms of the

Group Ophiuroidea.) Proc. Zool. Soc. for 1896, pp. 1028-1044.

16. Hall, James. 1867. (Note upon the Genus Palaeaster and Other Fossil

Starfishes.) Reg. Rep. State Cab. Nat. Hist. N.Y., No. xx. pp. 282-303,
pi. ix.

17. Hamann, 0. 1885-89. Beitrage zur Histologie der Echinodermen. Heft

2. Die Asteriden, &c., 8vo, Jena. Heft 4. (Anat. und Histol. der
Ophiuren und Crinoiden), Jena. Zeitschr., vol. xxiii. pp. 233-388, pis.
xii.-xxiii.

18. Heller, Camill. 1858. (UeberneuefossileStelleriden.) Sitz.-ber. math.-nat.

Cl. Akad. Wiss. Wien, vol. xxviii. pp. 155-170, pis. i.-v.

19. Hoffmann, C. K. 1874. (Zur Anatomie der Asteriden.) Niederl. Arch.

Zool., vol. ii. pp. 1-32, pi. i.-ii.

20. Koehler, R. 1887. (Recherches sur 1'appareil circulatoire des Ophiures.)

Ann. Sci. Nat. Zool., ser. 7, vol. ii. pp. 101-158, pis. vii.-ix.

21. Lamarck, J. B. P. A. de M. 1816. Histoire naturelle des Animaux sans

Vertebres, vol. ii., 8vo. Paris.

22. Lange, W. 1876. (Beitrage zur Anatomie und Histiologie der Asterien

und Ophiuren.) Morph. Jahrb., vol. ii. pp. 241-286, pis. xv.-xvii.

28o THE STELLEROIDEA

23. Linck, J. H. 1733. De Stellis, Marinis etc. Fol., xxiv. and 107 pp.,

xlii. pis. Lipsiae.

24. Ljungman, A. 1866. (Ophiuroidea viventia hue usque cognita.) Ofvers.

Svensk. Vet. Akad. Fbrhandl., 1866, No. 9, pp. 303-336.

25. Ludwig, H. 1877-82. Morphologische Studien an Echinodermen, 8vo.

Leipzig. Reprinted from Zeitschr. Wiss. Zool. vol. xxx. pp. 99-162,
pis. v.-viii. ; vol. xxxi. pp. 59-67, pi. v. ; pp. 216-234, pi. xv. ; pp. 346-
400, pis. xxiv.-xxviii. ; vol. xxxii. pp. 672-688 ; vol. xxxiv. pp. 333-366
pis. xiv.-xvi. ; vol. xxxvi. pp. 181-200, pis. x., xi. ; vol. xxxvii. pp. 1-98
pis. i.-viii.

26. 1888. (Ophiopteron elegans, eine neue, wahrsclieinlich schwimmeude

Ophiuridenform. ) Zeit. Wiss. Zool. vol. xlvii. pp. 459-464, pi. xxxv.

27. Liitken, C. F. 1858-59. Additamenta ad historian* Ophiuridarum,

Videnskab. Selsk. Skrifter (5), Naturvid. og Math. Afd. vol. v. pp. 1-74,
pis. i., ii. ; pp. 177-272, pis. i.-v.

28. Lyman, Th. 1865. (Ophiuridae and Astrophytidae. ) 111. Cat. Mus.

Comp. Zool. Harvard, No. 1, viii. 200 pp., ii. pis.

29. 1871. (Supplement to the Ophiuridae and Astrophytidae.) 111. Cat.

Mus. Comp. Zool. Harvard, No. 6, 18 pp., ii. pis.

30. 1874. (Ophiuridae and Astrophytae, new and old.) Bull. Mus. Comp.

Zool. Harvard, vol. iii. No. 10, pp. 221-272, vii. pis.

31. (Report on the Ophiuroidea dredged by H.M.S. Cliallcnger during

the years 1873-76.) Rep. Chall. Zool. vol. v. 1882, 4°, pp. 387, 48 pis.

32. MacBridc, E. W. 1892. (The development of the genital organs, ovoid

gland, axial and aboral sinuses in Amphiura sqiiamata, together with
some remarks on Lud wig's haemal system in this Ophiurid.) Quart.
Journ. Micr. Sci., N.S., vol. xxxiv. pp. 129-153, pis. xvi.-xviii.

33. 1896. (The Development of Asterina gibbosa.) Quart. Journ. Micr.

Sci., N.S., vol. xxxviii. pp. 339-411, pis. xviii.-xxix.

34. Miillcr, JoJi. 1854. (Uber den Ban der Echinodermen.) Phys. Abh.

Akad. Wiss. Berlin, Jahrg. 1853, pp. 123-219, pis. i.-ix.

35. <fc Troschel, F. H. 1840. (tlber die Gattungen der Asterien, und

der Ophiuren.) Arch. f. Naturg., Jahrg. 6, Bd. i. pp. 138-328,
367-368.

36. 1842. System der Asteriden. 4to, pp. xx. and 135, xii. pis. Braun-
schweig.

37. Perrier, J. 0. E. 1875-76. (Revision de la collection de Stellerides.)

Arch. Zool. Exper., vols. iv. and v.

38. 1894. Expeditions Scientifiques du Travailleur et du Talisman

pendant les annees 1880, 1881, 1882, 1883. Echinodermes, pt. 1. 4°, pp.
iv. and 431, xxvi. pis.

39. 1891. Mission Scientifique du Cap Horn, 1882-83, torn. vi. Zoologie

Echinodermes, I., Stellerides. 4to, Paris, 198 pp., xiii. pis.

40. 1891. (Stellerides nouveaux provenant des Campagnes du yacht

I'Hirvndelk.) Me"m. Soc. Zool. France, vol. iv. pp. 258-271.

41. ProuJio, H. 1890. (Du sens de 1'odorat chez les Etoiles de Mer.) Compi.

Rend. Ac»d. Paris, vol. ex. pp. 1343-46.

42. Salter, J. W. 1857. (On some new Palaeozoic Starfishes.) Ann. Mag. Nat.

Hist. ser. 2, vol. xx. pp. 321-334, pi. ix.

43. Sars, G. 0. 1875. Researches on the Structure and Affinities of the Genus

JBrisinga, etc. 4to, 111 pp., vii. pis. Christiania.

THE STELLEROIDEA 281

44. Schulze, C. F. 1760. Betrachtung der versteinerten Seesterne und ihrer

Theile. 4to, vi. and 58 pp., ii. pis. Warschau und Dresden.

45. SimonowUschy Spiridon. 1871. (Ueber einige Asteroiden der rheinischen

Grauwacke.) Sitzber. math.-phys. Cl. Akad. Wiss. Wien, vol. Ixiv.,
Abt. 1, pp. 77-122, iv. pis.

46. Simroth, H. 1876-77. (Anatomic und Schizogonie der Ophiactis virens.)

Zeitschr. Wiss. Zool., vol. xxvii. pp. 417-485, pis. xxxi.-xxxv. ; vol. xxviii.
pp. 419-526, pis. xxii.-xxv.

47. Sladen,W.P. 1879. (On the Structure of Astrophiura, a new and Aberrant

Genus of Echinoderms. ) Ann. Mag. Nat. Hist., ser. 5, vol. iv. pp. 401-
415, pi. xx.

48. 1889. (Report on the Asteroidea collected by H.M.S. Challenger

during the years 1873-76.) Rep. Chall. Zool., vol. xxx. 893 pp., cxvii. pis.

49. Stiirtz, B. 1886. (Beitrag zur Kenntniss palaeozoischer Seesterne.)

Palaeontographica, vol. xxxii. pp. 75-98, pis. viii.-xiv.

50. 1890. (Neuer Beitrag zur Kenntniss palaeozoischer Seesterne.)

Palaeontographica, vol. xxxvi. pp. 203-247, pis. xxvi.-xxxi.

51. 1893. (Ueber versteinerte und lebende Seesterne.) Verh. Naturli.

Ver. Preuss. Rheinl., vol. 1. pp. 1-92, pi. i.

52. Viguier, C. 1879. (Anatomic comparee du Squelette des Stellerides. )

Arch. Zool. Exper., vol. vii. 1878, pp. 33-250, pis. v.-xvi.

53. Woodward, H. 1869. (On Eucladia, a new genus of Ophiuridae.) Geol.

Mag., N.S., vol. vi. pp. 241-245, pi. viii.

54. Wright, Th. British Fossil Echinodermata of the Oolitic Formations.

Vol. ii. The Asteroidea and Ophiuroidea. Palaeontogr. Soc. 1863-80.

CHAPTER XV,

THE ECHINOIDEA.1

CLASS III. ECHINOIDEA.
SUB-CLASS 1. REGULARIA ENDOBRANCHIATA.
Order 1. Bothriocidaroida.
„ 2. Cystocidaroida.
„ 3. Cidaroida.
„ 4. Melonitoida.
„ 5. Plesiocidaroida.

SUB-CLASS 2. REGULARIA ECTOBRANCHIATA.
Order 1. Diademoida.
Sub-Order 1. Calycina.
„ 2. Arbacina.
,, 3. Diademina.
„ 4. Echinina.

SUB-CLASS 3. IRREGULARIA.

Order 1. Gnathostomata.

Sub-Order 1. Holectypina.
„ 2. Clypeastrina.
Order 2. Atelostomata,
Sub-Order 1. Asternata.
„ 2. Sternata.

KNOWLEDGE of the Echinoidea or Sea Urchins is more complete
than that of any other of the classes of Echinoderma. The
class is widely distributed at the present day, members of it
living in seas of all parts of the world and at all depths. It is of
great antiquity, and fossil Echinoids occur abundantly in rocks of
all periods since the Devonian, while a few are known from that
system and from the Silurian. Moreover, as the classification of
the Echinoids depends either upon the skeleton, or upon structures,
the characters of which can be inferred from those of the skeleton,
the systematic position of a fossil can generally be determined with
as much accuracy as that of a recent specimen.

1 By J. W. Gregory, D.Sc. MS. closed at end of 1896.

THE ECHINOIDEA 283

The very varied habits of the Echinoids further increase the
value of this class. Some sea-urchins burrow into sand, and others
bore into rocks. Some seek the shelter of rock-pools or the quiet
of a muddy sea-floor below the limit of tidal action. Others
cling to rocks between tide-lines, choosing the positions that are
most exposed to the buffeting of a tropical surf ; others again
crawl over, or lie half -buried in, the ooze of abyssal oceanic
depths. Some feed on algae ; others swallow mud and live
on the organic matter it contains. Of some the young develop
directly, of others indirectly, the latter undergoing a metamor-
phosis during development. The modifications in structure by
which Echinoids are able to adapt themselves to these different
habits are so well marked, that the conditions under which fossil
sea-urchins lived can generally be determined. Hence the group is
of great value to the geologist. Such rich series of Echinoid faunas
are known, that the life-history of the class can be written with
greater completeness than that of any other group of Echinoderms,
and as completely as that of any class in the animal kingdom.

Little, however, is certainly known as to the relations of the
Echinoidea to other Echinoderms. The group is so compact and
well marked, that the distinction between it and the other classes
was known to Aristotle, and has never been in doubt. Moreover,
the class has been as well defined as it now is since Silurian times.
The recognition of the sea-urchins as a distinct group of Echino-
derms is, therefore, as old as zoology. A sketch of the history of
work upon the Echinoids need only consider the determination of
the main points in their anatomy, life-history, and classification.

The earliest account of the natural history of the sea-urchins
is in Aristotle's History of Animals, wherein the common edible
species of the Mediterranean (Echinus esculentus) is described
with considerable accuracy. Aristotle called this animal Echinus
(ExtVos), the Greek for hedgehog, a term subsequently given to the
best known genus, and used as the root of the name for the class.
In the same book three other types of sea-urchins are mentioned,
viz. Brissus, Spatangns, and Echinometra ; and these names are
still used in Echinoid nomenclature. Aristotle's account shows
that he had studied both the habits and anatomy of the animals ;
thus he knew that Echinus could walk upon the tip of its spines,
had five teeth, five unpaired ovaries, a pharynx, and stomach. In
mediaeval times Echinoids were described by Rondelet (1554) and
Aldrovandus (1606 and 1648); but it was not till the beginning
of the eighteenth century that any observations of scientific value
were published. The first important post-classical contributions
to the subject were the works of Breynius (1732) and Klein
(1734). The former in his De Echinis et Echinitis, and the
latter in his Naturalis Dispositio Echinodermatum, figured and

284 THE ECHINOIDEA

described many genera and species of Echinoids; but, as their
names were not binominal, they are not accepted. Gualtieri in
1742, and Seba in 1758, used the names and methods of their
predecessors, and described additional species. In the latter year
Linnaeus adopted the binominal system of nomenclature in the
tenth edition of his Systema Naturae, but in other respects his
treatment of the Echinoids was retrograde. He accepted sixteen
species ; but although among these there were such different types
as Echinus esculentus, Cidaris, Echinometra, Colobocentrotus, Arachnoides,
Clypeaster, etc., he included them all within a single genus. He
thus threw back the study of Echinoids for twenty years. It was
not till 1778 that Leske (48) reintroduced the sound system of
work adopted by Breynius and Klein, which had been discarded by
Linnaeus. Leske's edition of Klein, with his own " Additamenta,"
therefore forms the real starting-point of systematic Echinology.
No important advance from this was made until the publication
of Lamarck's Histoire Naturelle des Animaux sans Ferttbres in 1816.

From this date progress was rapid ; J the principal contribu-
tions being made by Gray (1825), de Blainville (1830), Desmoulins
(1837), L. Agassiz (1836, 1840, 1841, 1842), and Desor (1842).
In 1846 and 1847 the last two authors published a complete
synopsis of knowledge up to that date in their Catalogue Raisonnt
des Echinides. Since then the literature of the Echinoidea has
been voluminous. The existing species have been described by
Liitken, Diiben and Koren, Lov£n, Leuckart, Peters, Grube,
Doderlein, Thomson, Bell, and especially by A. Agassiz. Our
knowledge of Palaeozoic Echinoids is mainly due to M'Coy, C. F.
Romer, J. Miiller, Desor, Meek, Worthen, Hall, Neumayr, Duncan,
Keeping, and Jackson ; of the Mesozoic faunas to d'Orbigny,
Cotteau, Wright, de Loriol le Fort, Clark, Schultze, Lambert,
Gras, Forbes, S. P. Woodward, etc. ; of the Cainozoic to Dames,
Laube, Cotteau, de Loriol, Forbes, Bittner, Gauthier, Peron,
Pomel, Duncan, Sladen, Hutton, Sismonda, Michelin, Grateloup.

Dujardin and Hupe" (21) in 1862 attempted a synopsis of
the whole of the Echinoidea. Desor's Synopsis des Echinides
Fossiles (1854-58) is the last reliable summary of the fossil species,
as is Agassiz's Revision of the Echini of the recent. Revisions of the
genera were arranged by Pomel in 1883 and Duncan in 1890 (24).

The morphology of the Echinoidea was first seriously studied by
L. Agassiz and Valentin, whose account in 1842 (5) of the skeleton
and visceral anatomy was based mainly on Stronyylocentrotus lividus,
Lam. sp. The circulation had been previously described by Tiede-
mann (1815) and Delle Chiaje (1825), and the nerves by Krohn

1 Reference to part of the literature is given 011 pp. 328-332 ; a bibliography up to
1872 is given by A. Agassiz (1), pp. 1-9. Some of the general works are included in
the lists for Stelleroidea (p. 279) and for Ecliinodenua generally (p. 35).

THE ECH1NOIDEA 285

(1841). The next important advance was made by the memoirs of
Joh. Miiller (1850-57) who gave a full account of Cidaris. Among
later general contributions to anatomy may be mentioned those of
Teuscher (1876), Koehler (1883), Prouho (1887), Fredericq (1876),
Perrier (1869 and 1875), Hamann (1889), and the last part
of A. Agassiz's Revision. Special types have been described in
separate memoirs, such as Cidaris and its internal branchiae, by
Stewart (1880); Spatangus, by Hoffmann (1871); Asthenosoma,
by P. and F. Sarasin (1888); Phormosoma, by Bell (1889);
Dorocidaris, by Prouho (1888); Pourtalesia, by Loven (1883).

Knowledge of the anatomy of the skeleton is due to many
students, but especially to the masterly series of memoirs by
Lov6n. The structure of the spines has been described and their
taxonomic value shown by Mackintosh (63) and A. Agassiz (1, 2) ;
the fascioles were first used in classification by Liitken. The pedi-
cellariae were first described by O. F. Miiller, who regarded them
as parasites, a view disproved by Delle Chiaje ; L. Agassiz con-
sidered them to be young Echini, which was shown to be erroneous
by Valentin's detailed figures. An attempt to use pedicellariae as
a basis for classification was made by Perrier (1869-70); their
function has been the subject of a long controversy.

The study of the embryology of Echinoidea was begun by Derbes
in 1847, and by Miiller in a series of memoirs (1848-55); it was
continued by Krohn (1849-57), A. Agassiz (1864), Metschnikoff
(1868-69), Bury (1889 and 1895), Theel (1892), MacBride
(1896), and others.

The term Echinoidea, now used as the name of the class, was
applied by Aristotle to animals that resembled Echinus, which has
always been regarded as the most representative genus. Though
in some ways rather complex, it may be conveniently studied as
i Typical Sea-Urchin, since specimens can be easily obtained.

The Skeleton. If we examine a specimen of the common
British species Echinus esculentus, we first notice that it is
covered all over by short, bluntly pointed spines, which are
coloured violet at the tips. If we pull off the longest spines,
we find between them a number of smaller " secondary " spines
(compare Fig. XXXIIL). After removal of all the spines, the
general character of the main shell can be seen. It is com-
posed of closely fitting plates, which together form the rigid
" test " or shell. The shape is rounded in transverse section ;
the surface on which the mouth opens is flattened, while the
upper half of the test may be either well rounded or sub-conical.

On the centre of the lower flattened surface there is a large
flexible membranous area covered with loose plates which bear
short spines. This is the " peristome " ; the mouth opens in the
centre of this space ; its exact position can be determined by

286

THE ECHINOIDEA

the presence of five sharp teeth, which, if not extruded so as to be
seen, can at once be felt. At the other end of the axis of the

body there is a smaller membranous
area, the " periproct," in the centre
of which is the anus. The periproct
is surrounded by a circle of five plates
(Fig. I,), one of which is larger than
the rest ; this large plate bears a
small perforated prominence, called
the " madreporiform tubercle " or
" madreporite " (m). Through the
no. i. pores in this plate water passes to

csvdentus Linn apical the series of canals known as the
e^^ water-vascular system. Each of the

tai pore ; o, ocular plates ; an, anal other four plates of this apical circle

plates covering the periproct, in the . r . j , <• ^r

centre of which opens the anus, is perforated by one of the pores

through which the genital products

escape to the exterior. These five plates are therefore known as
the " genitals " (g). Alternating with them are five smaller plates
(o), each perforated by a pore, through which passes a process
ending in a sensory eye-spot. These plates are therefore called
the " oculars." The whole ten plates form a group known as the
" apical system." Between the apical system and the peristome is
the main test. Ten lines of suckers may be seen in a fresh
specimen, radiating from the five ocular plates ; two lines start
from each ocular, and pass in a straight series round the test to
the edge of the peristome. As the two series of suckers are
arranged like the rows of trees in an avenue, the area bounded by

Fio. II.

Ambulacral (1) and interam-
bulacral (2) plates of Echinus
uculentus. fl, fl are the primary
and secondary tubercles ; g,
miliary granules ; p, primary
ambulacral plate ; d, epipodium
with a pore pair; b, boss; m,
mamelon.

them has been called an ambulacrum.1 As there is one ambula-
crum to each ocular, there are five in the complete test, separated
by five broader interambulacra.

Each ambulacrum and interambulacrum consists of a double
row of plates in vertical series running from the apical system to
the peristome. Each interambulacral plate (Fig. II. 2) is irregu-
larly pentagonal in form ; the angles are sharp and regular, but
the plate is elongated in a horizontal direction. Each plate bears
a number of " tubercles," of which there are three sizes — primary

1 From ambulacrum, an avenue or a walk between trees.

THE ECHINOIDEA

287

(tl), secondary (t2), and miliary (g). The primary tubercles are
the largest, and bear the primary spines ; each tubercle consists
of a rounded base or "boss" (6), on the centre of which is a small
pimple or " mamelon " (m). Around the base of each primary
tubercle is a smooth, level surface called the " scrobicule," which is
generally surrounded by a circle of granules called the " scro-
bicular circle." Scattered irregularly over the plates are the
smaller secondary tubercles which bear the secondary spines, and
between these are large numbers of smaller elevations known as
" miliary granules " (g).

The function of these tubercles and granules is the support of

g

Diagram of gemmiform pedicellaria of
Echinus acutus (after Hamann). a.m, ad-
ductor muscles ; e.m, extensor muscles ;
/.TO, flexor muscles ; gl, gland ; gl.c,
glandular epithelium ; s.o, stem ossicle.

FIG. III.

Appendages of Echinus. Transverse
section of spine (magnified).

the appendages known as "spines"
and "pedicellariae." The Spines,
like the tubercles, are of three
sizes — primary, secondary, and
tertiary — the structure of eaqh of
which is fundamentally the same.
Each consists of a long shaft,
marked by longitudinal flutings ; the base^ of the spine is hollowed
into a cup or condyle, which fits over the mamelon of the tubercle.
Around the condyle is the collar of the spine which projects above
the level of the shaft, and serves for the attachment of the muscles
which fix it to the test. A transverse section of the spine (Fig.
III.) shows that it is made up of a number of solid wedges, radiating
from a central axis, and separated by bands of porous tissue.1 The
smallest spines consist only of tiny needles of porous tissue, and
agree in structure with the spines of the second type of appendage.
The pedicellariae (Fig. IV.) consist of two or three beak-like

1 The spine is therefore, according to the terminology of Mackintosh (63), poly-
cylic and acanthosphenote.

288 THE ECHINO1DEA

valves attached to the end of a flexible stem. The valves open
and shut like a bird's beak, or like the avicularia of Bryozoa.
They have been seen to seize particles of the excreta of the urchin
and pass them on from one pedicellaria to another, until they fall
over the margin of the test. Their main function, however, appears
to be defensive. When a starfish attacks a sea-urchin, the latter
bends down its spines and thus exposes its pedicellariae; these
seize hold of the tube-feet of the starfish, which their bites appear
to hurt. The pedicellariae, however, are always torn away, as they
cannot relax their hold ; and thus, if an urchin is attacked by
a series of starfish, it is in time rendered defenceless (Prouho, 70).
The sphaeridia are also modified spines ; they are globular in
form, and lie in pits around the peristome (Fig. V.).

The ambulacral plates (Fig. II. 1 ) are ornamented by tubercles

and granules like those of the interambulacral plates, though

_^ smaller; but the plates differ

in structure. The interam-
bulacrals are solid, whereas
the ambulacral plates are
pierced by small holes,
through which pass the
suckers or podia of the am-

Tmnsverse section through peristonual plates of bulaCra« Thf P0™ OCCUr
Clypeaster showing a globular sphaeridium in its pit. Ill pairs, and each pair IS

surrounded by a raised

ring forming an epipodium. There is one epipodium to each
primary ambulacral plate ; but, except at the summit of the am-
bulacrum, the primary plates are united into compound plates,
each of which has as many epipodia as there are primary plates
in it. In one of the plates in the middle of the ambulacral series
there are three pairs of pores, which occur in a curved series or
arc. This arrangement of the pairs is due to the crowding of the
plates during growth ; owing to the same cause, the elementary
ambulacral plates no longer always extend right across the half
of the ambulacrum to which they belong, but are cut off from
the median suture by tlie union of adjoining plates behind them.
Plates thus cut off from the central suture line are known as
" demi-plates." In an ordinary Echinus esculentus each compound
ambulacral plate (Fig. XII. 4) consists of one central demi-plate
(d) between two primary plates (p).

The principal remaining elements in the skeleton are those of
the five jaws, and of the internal processes, to which the muscles
that work the jaws are attached. The Dental Apparatus (Fig. VI.)
is a conical structure which is placed apex downwards over
the mouth. The axis is hollow and contains the pharynx or
commencement of the oesophagus. The dental apparatus consists

THE ECHINOIDEA

289

of twenty pieces. The largest of these are the five pyramids (Fig.
VI. p) ; they are shaped like the sectors of a cone, being pointed
at one end, having two flat sides, and with a curved outer margin.
The flat sides are marked by transverse ridges, which serve for the
attachment of the muscles that bind the pyramids to one another.
The pyramids are hollow, and in each of them lies a long, curved,
keeled tooth, the hard point of which projects through a hole
at the pointed lower
end of the pyramid.
Above the suture be-
tween a pair of pyra-
mids rests a short, thick
bar known as the brace ;
above the brace (b) is
a curved bifid process
or " compass " (cp).

Round the inside of
the peristome is a hard
raised rim (Fig. VII.
p.r) which rises into an
arch over each of the
ambulacra. This is the
" perignathic girdle,"
and to it are attached
the muscles which work
the masticatory appara-

tUSj One Set Of muscles

ni-ta^ViPrl tn thp Tnrramirl
atta( ie Pyrami I

pulls the jaws apart ;

other sets, attached to the braces and compasses, pull them down-

ward and drive the teeth together.

The remaining skeletal structures are small and unimportant,
and consist only of scattered calcareous plates and spicules which
are distributed through the tissues. The principal of these are
the " rosettes " or " pellions " (rings of plates which support the
suckers), and spicules in the stems of the podia.

The Internal anatomy of the Echinus may be most conveniently
studied by the removal of the upper half of the test. The five
ovaries will then be seen lying in the upper parts of the inter-
radia, and the great coiled intestine occupying most of the
interior. The main features in the internal anatomy are shown
in Fig. VII.

The Alimentary Canal begins with a mouth (mt) situated at
the centre of the peristomial membrane (pst) ; from the mouth the
muscular pharynx (ph) passes upwards through the central tube
of the masticatory apparatus. The oesophagus (oe) runs horizon-

'9

Fl(J> VJ<

A pyramid

rom ell GO > P> Pyr
aml e> its epjphysis ; t, tooth ; b, brace ; cp, compass.

Dental apparatus of Echinus csculentus, L.
seen from tl)e side W' and from behilld GO > P> Pyramid;

290

THE ECHIN01DEA

tally from the upper end of the pharynx to the lobed stomach (s),
which is twisted in a spiral twice round the body cavity ; it con-
tracts above to the rectum (r), which opens to the exterior by an
anus (a), situated in the centre of the apical system.

The Water -vascular System. Above the top of the dental
apparatus a vessel may be seen running round the oesophagus.
This is the circumoesophageal ring of the water-vascular system
(c.ce.v). From it five radial vessels pass downwards outside the
masticatory apparatus to the peristomial membrane; there they
bend upwards, pass beneath the arches of the perignathic girdle

sp.

tb.

pdc.

FIG. VII.

Diagrammatic vertical section through Echinus, a, anus; amp, ambulacral plate; ap,
ampulla ; fc, branchiae ; c.oc.v, circumoesophageal vessel of water-vascular system ; d.o, dorsal
organ ; g, gonad ; m, madreporite ; int, mouth ; n.r, nerve ring ; ce, oesophagus ; p.a, arch of
perignathic girdle ; pd, podia ; ph, pharynx ; pdc, pedicellaria ; pp, periproct ; p.r, ridge of
perignathic girdle ; pst, peristome; py, pyramids of masticatory apparatus ; r, rectum; r.n.c,
radial nerve cord; s, stomach; s.e, stone canal; si>/i, sphaeridia ; sp, si>. e; st, Stewart's
organ ; tb, tubercle ; r, Polian vesicle on circumoesophageal vessel.

(p.a), and one runs up the inside of each ambulacrum. Branches
from the radial vessels are given off alternately to right and left ;
one of these branches passes to each pore-pair, below which it opens
to a pair of pocket-shaped vesicles or ampullae (ap). From each
ampulla a small tube passes through both pores ; the two tubes
unite on the exterior to form the shaft of the tube-foot or
podion (pd).

The water-vascular system round the peristome of an irregular
Echinoid is shown in Fig. XLIV. The water-vascular ring (w.v.r)
lies above the peristomial membrane, and just above the circumoeso-

THE ECHINOIDEA 291

phageal nerve ring, which can be seen in the figure, connecting the
five ambulacra! nerve cords (am.n).

The walls of the podia are strengthened by calcareous spicules,
and expand at the end into a sucker. The function of the tube-
feet is to help in locomotion. The sucker is pressed against a
smooth surface ; water from the reservoir or ampulla is driven
into the podion, and the tube-foot is thus rendered tense and rigid.
The rosette of plates in the sucker is pulled backward, when a
vacuum is left between the sucker and the surface against which it
is pressed. Firm attachment is thus secured, and the animal can
drag itself along (for action, see Fig. XLIL). So powerful are
these suckers that, by their means, the Echinoids of the genus Colobo-
centrotus (Fig. XXXIV.) can cling to exposed rock surfaces, fully
exposed to the surf of a coral reef.

The supply of water to the water- vascular system is introduced
by a membranous vessel — the stone-canal (s.e) — which rises from
the circumoesophageal ring, and is attached to the plate of the
apical system which bears the madreporite (m). As we have seen,
this plate is perforated by many small pores, through which water
can pass into the stone -canal. Owing to the small size of the
pores, the water is filtered as it enters. In Echinus esculentus the
stone-canal is membranous, and the name therefore appears inap-
propriate. The name was first applied to this canal in the genus
Cidaris, in which it is hard and calcareous. The flow of water in
the interior is regulated by five "Polian vesicles," situated on
the circumoesophageal ring, and acting as reservoirs. The latest
account of the function of these vessels is given by Uexkiill (82).

The main points in the distribution of the water- vascular vessel
can be. easily verified ; but the Blood-vascular or haemal System
is more obscure. The most important structure is an ovoid body
(Fig. VII. d.o), situated beside the stone-canal ; it is known as
the " dorsal or axial organ," and by other names (see Chapter VIII.
pp. 23, 25). Its canal joins above with the stone-canal; it is said
to open to the exterior through the madreporite, but this is denied
by some authorities (as Hamann). Round the upper end of the
" canal of the dorsal organ " is a circular canal known as the
"genital ring," which appears to be connected with a series of
haemal vessels or lacunae which surround the dorsal organ. In
this case it must be also connected with a ring round the oesophagus,
from which five branches pass downward beside the pharynx, and
then run up along the test below the ambulacra. The "circum-
oesophageal haemal vessel " is connected with a haemal vessel which
runs along the inner side of the stomach.

A third group of canals or vessels consists of a circular sinus
round the oesophagus, from which five branches run up the ambu-
lacra between the radial water- vascular vessels and the radial nerve

292 THE ECHINOIDEA

cords. This group of sinuses is known as the "pseudhaemal
system."

The function of the haemal ard pseudhaemal systems has been
much debated, and the relations of their various members are still
uncertain. The dorsal organ is sometimes said to be a kidney,
as by Hamann (38) and the Sarasins (72). Hartog (39) has
supported this by claiming that the circulation is outward
through the madreporite; but Cuenot and Ludwig (62) main-
tain that the current is inhalent. Leipoldt and Prouho (70)
point to the absence of any glandular epithelium and of any
connection between the cavity of the organ and the general body-
cavity. They therefore deny that the organ is a gland, and regard
its function as the making of amoeboid cells for the perivisceral
fluid. It therefore seems most probable that the haemal system
distributes nutrient material through the body, both in solution
and by corpuscles (see also Durham, 11 of previous list, p. 36).

A third circumoesophageal ring is that of the Nervous System
(Fig. VII. n.r). This is placed below the water- vascular ring. From
it five radial nerve cords (r.n.c), the ambulacral nerves, pass up the
inside of the test, between it and the ambulacral water-vascular
vessel. Branches from the ambulacral nerve cord pass right and left
to the ampullae, and give off smaller branches which pass through

the pores to the suckers
(Fig. VIII.). The branches
fork, one half running up
n.j>d. the podion and the other

expanding over the surface
of the test as a plexus,
which controls the move-
ments of the spines and
the pedicellariae. A small
nerve ring (n.sp) surrounds
Fl°- viii. the base of each spine.

Diagram of innervation of spines in Echinus. plt TViA ftpnArutivP Orwana

ambulacral plate ; ep, epidermis ; m, muscles of Dative Urgans

spines ; n, branch of nerve passing out through are large and simple They
pore; n.pd, branch of nerve to tube foot; n.x, branch . ' . £ , . J

of nerve running across test to n.sp. the spinal COttSlSt Ol nve branching

nerve ; pd, podion ; sp, base of spine ; pi, ambulacral ^lor^e, /!?;„ 17TT /,\ ,,,^;,a,

plate with <^boss of tubercle and (m)mamelon. glands (* Ig. Vll. g), Which

lie attached to the inter-

ambulacral plates in the upper part of the body-cavity. Each of
the five organs opens to the exterior by a single tube which passes
through the pore in a genital plate. The young of Echinus are
free-swimming plutei, and undergo a metamorphosis during the
development and resorption of the pluteal skeleton and its append-
ages (cf . development of Echinocyamus, Chapter VIII. pp. 15-1 7).

Respiration is largely effected by the aeration of water in the
podia ; but in addition to this there is a series of five pairs of small

THE ECHINOIDEA

293

folded branchiae or gills (Fig. VII. b) lying on the margins of the
peristomial membrane. These are diverticula from the general
body-cavity, and pass out by the ten notches in the peristomial
margin of the test. There are, moreover, five large internal
vesicles rising from the upper edge of the masticatory apparatus ;
these are known as "Stewart's organs" (Fig. VII. st), and may
act as internal gills.

The ordinary Echinus, then, has the following characters : — It
consists of a skeleton, which is mostly external, and is com-
posed of numerous closely fitting polygonal plates, bearing spines.
Within, it has a simple, coiled, alimentary canal, with mouth and
anus at the opposite poles ; it has five generative glands ; an
elaborate series of water-vascular vessels, provided with podia,
ending in suckers. This water-vascular system, the blood-vascular
system, and the nervous system each consist essentially of a
ring round the mouth, from which five branches pass outward,
one up the inside of each ambulacrum.

Among the Echinoids the Variations in Structure from this
simple type are very diverse. Thus the form, instead of being
globular, may be depressed into a thin, flat sheet, in which the
wide, low roof has to be supported by pillars, as in Scutella (Fig.
XXXV. 4). In some of these thin forms the posterior margin of
the test is lobed and digitate, as in
Rotula (Fig. IX.) ; in some cases the
ends of the processes unite, leaving
perforations or " lunules." In such
depressed forms, owing to the sharp
division of the test into upper and
lower halves, the central podia of
the former are useless for purposes
of locomotion, and are specialised
to serve as branchiae ; thus the am-
bulacra become modified into petaloid
and extra - petaloid portions. The
ejection of the excreta through an
anus situated in the middle of these
branchiae would be disadvantageous,
owing to the consequent pollution of
the water. Hence, in such forms, the anus has passed backwards,
and opens in the hinder part of the test. This backward move-
ment of the anus is usually balanced by the forward movement
of the mouth, and thus the Echinoid loses its quinqueradiate
symmetry and becomes bilaterally symmetrical.

This change affects not only the position of the external
apertures, but the development of the internal organs. Owing to
the invasion of the posterior interradius by the anus, the generative

Fio. IX.

Rotiila Avgusti, with posterior "digita-
ions " and a pair of anterior " lunules."

294 THE ECHINOIDEA

gonad there aborts ; the gonads thus occur as two lateral pairs, and
increase the bilateral symmetry of the Echinoid. This is further
strengthened by changes in the structure of the apical system
of plates. This system, as in Cidaris or Echinus (Fig. I.), consists
of a double circle of plates. The inner circle is formed by the
five genital plates, which are often called the basals. As the
latter name has been given to them on the ground that they are
the homologues of the basal plates in Crinoids, the older name of
genitals is here retained. The outer circle of plates occur in the
angles between the genitals, and as they bear the " oculi," they
are called the oculars ; they have in turn been regarded as homo-
logous with the radial plates of the Crinoid cup and called "radials,"
while Cu6not, on a different theory of homology, calls them the
" terminals." As the double circle of plates surrounds the anus,
this form of apical system is known as " endocyclic." From its
typical arrangement, as seen in Cidaris, variation takes place in
two directions. The extreme of one line is seen in Tiarechinus (Fig.
XVIII.) and Lysechinus, in which the apical system of plates forms
either half or nearly the whole of the test. In the other direction the
plates become less important ; in Aspidodiadema (Fig. XXII.) they
form a single ring of ten plates ; and in Asthenosoma (Fig. X.) they
_. are either reduced to ten rudimentary plates of no

^©^H? functional importance, or are altogether absent.
^o *^?$ The membrane which lies between the anus

a—t^—O ff and the genital plates is generally covered by num-
^^*S^ erous small plates, known as the "anal plates."
FIG. x. I*1 some genera, such as Acrosalenia, some

Apical system of of the anal plates are large and thick, and are
a^anai'^opS: firmly attached to the genitals. One of these
ashn P* and F' ***' Plates may increase at the expense of the rest,
until, as in the genus Salenia, there is one large
suranal plate attached to the genitals (Fig. XIX.). This plate
has been regarded as the homologue of the imaginary dorso-
central plate of the Crinoids. This plate pushes the anus back-
ward from its originally central position.

A tendency towards the retrogression of the anus is shown
in all groups of Echinoids. One of its effects is the pulling
out of the posterior plates of the apical system, and the consoli-
dation and increase of those in front. Thus in Zeuglopleurus the
anterior genitals meet along the middle line, while the posterior
genitals become narrow and are completely separated from one
another by the oculars. In Pygaster (Fig. XI. 1 ) the retrogression
has become so marked that the anus lies just outside the apical
system, which is therefore " exocyclic." In Clypeus (Fig. XL 2, 3)
the anus becomes completely detached from the apical system,
which is no longer a ring of plates, but a compact group.

THE ECHINO1DEA

295

In Echinus the madreporite (Fig. I. w), or the opening of the
water- vascular system, is on the right anterior genital plate. After
the anus has receded from the apical system, the madreporite
begins to follow it. In a simple compact apical system the pores
of the water-vascular aperture occur only in the right anterior
corner of the system. Such a system is said to be " ethmophract,"
as in Discoidea or Micraster (Fig. XL 4). In more advanced forms
the pores and the plate on which they open extend backward until

4 Fia. XL 5

Apical systems of Echinoidea. 1, Pygastcr umbrella ; 2, Clypeus sinuatus ; 3, Clypeus Hugi ;
4, Discoidea conica,- 5, Spatangus purpureus. In 1 and 2 the anna is still in contact with the
apical system ; in 3 it is free ; 4 is ethmophract, and 5 ethmolysian.

they separate the two postero-lateral genitals, as in Spatangus
purpureus (Fig. XL 5). Such an apical system is said to be
" ethmolysian."

Another change in the apical system is wrought by the elonga-
tion of the test in the antero- posterior direction, whereby the
apical system becomes elongated. The plates of the two pairs of
genitals become adjacent, and completely separate the anterior
and the two pairs of oculars from each other ; the three anterior
ambulacra usually meet close together, and are separated from the
two posterior ambulacra by a wide space. The three anterior

296

THE ECHINOIDEA

ambulacra then form the " trivium," and the two posterior form
the "bivium." This may be clearly seen in the common Chalk
Echinoid, Echinocorys scutatus, Leske (syn. Ananchytes ovatus, Lam.),
(Fig. XXXV. 2). In some cases the separation of the trivium and
bivium becomes greater; it is finally completed in the Jurassic
genus Collyrites, and the living genus Pourtalesia, in which the
apical system is broken up into two parts, separated by a zone of
ordinary interambulacral plates.

Important changes also take place in the plates of the test,
affecting both their structure and arrangement.

The Interambulacral Plates are biserial in the great majority
of Echinoidea, but they may be irftiserial, triserial, or multiserial.
Each plate may bear one tubercle or more than one ; to increase
the strength of the muscular attachment of the spine, the mamelon

4 Pio. XII. 5

Types of ambulacral plates. 1, cidaroid ; 2, diademoid ; 3, arbacioid ; 4, cchinoid ;
5, cyphosomoid. rf, demiplates ; p, primary plates.

is perforated by a small hollow, and the boss becomes irregular,
owing to a series of crenulations.

Thus among other variations of interambulacral plates are the
following: — Those which are unituberculate with the tubercles
either plain or crenulate, perforate or imperforate; those which
are bi- or multi - tuberculate, and have the tubercles either
perforate or imperforate ; those which are granulate, as in
Palceechinus ; those which have their edges bevelled, as in Echino-
thuria.

The principal variation in the Ambulacral Plates is in the
number of elementary plates (shown by the number of pore-pairs)
in a compound plate. There are five main types : —

1. The cidaroidt when all the plates are low, simple primaries,
as in the Cidaridae, Orthopsinae (Fig. XII. 1).

2. The diademoid, when all the plates are primaries, but they
are united in sets of threes into compound plates (Fig. XII. 2).

THE ECHINOIDEA

297

3. The arbacioid, when the compound plates are formed of
three simple plates, the middle one being a large primary, while
the other two are small demi-plates (Fig. XII. 3).

4. The echinoid, when the compound plates are formed of three
simple plates, but the middle plate is a small demi-plate, and the
two others are primaries (Fig. XII. 4).

5. The cyphosomoid type, when the compound plates are formed
of many simple plates arranged in arcs, in which the middle com-
ponents are demi-plates (Fig. XII. 5).

In most of the compound ambulacral plates, one or more of the
constituents become " demi-plates " by losing their contact with
the vertical suture on either side of the series. In some Echinoids
some of the plates are further reduced by growth-pressure, so that
they lie along the horizontal sutures between the primaries, as in
the Echinothuridae, or form broad areas of numerous small plates,
as in the Melonitidae. For these plates the name of " klasma-
plates " has been suggested.

One important variation affects both the ambulacral and inter-
ambulacral plates. In some forms, such as Asthenosoma, the plates
are thin and attached to powerful lateral muscles (Fig. XXX.), by
which the test can be contracted and
expanded. In such Echinoids the plates
are not closely fitted like stones in a
mosaic, but the edges are bevelled, so
that the plates overlap like slates on a
roof.

The Mouth Armature also under-
goes great changes, which may be best
seen by the nature of the perignathic
girdle, of which there are five main
types. In Cidaris (Fig. XIII. 1) it is
" disconnected," consisting only of an
erect "ridge" situated on the inter-
radial plates around the peristome.
In Salenia (Fig. XIII. 2), in addition to
the ridge, there are small " processes "
on either side of the ridge — the pro-
cesses arise from the ambulacral plates.
In the Diadematidae and Echinidae
(Fig. XIII. 3) the ridge becomes insig-
nificant and the processes important ;
they lengthen and form an arch across
the ambulacra. The perignathic girdle
is then said to be " continuous." The extreme form is met with
in such genera as Echinometra (Fig. XIII. 4), where the arch is
strengthened by a strong cap. In those Echinoids in which the

FIG. XIII.

Types of perignathic girdles. 1,
disconnected type of Cidaris / 2, dis-
connected type of Salenia ; 3, simple
arch of Diadema; 4, capped arch of
Echinmnetra. a, ambulacral plates ;
our, auricle ; c, cap ; i, interainbula-
cral plates ; i.a.p, interambulaeral
processes.

298 THE ECHINOIDEA

apical system is exocyclic, the jaws become less important. In
the Clypeasters the jaws are massive, but their power of move-
ment slight, as they are poised on small vertical processes which
fit into a socket on the pyramids.

In the Holectypina we see a gradual reduction of the peri-
gnathic girdle, the processes becoming low and the ridge important
(as in Discoidea), until in Galerites the ambulacral processes are
absent, and there are only five low interradial thickenings which
act as ridges. Finally, from the Atelostomata the perignathic
girdle and jaws are completely absent.

The Generative Glands are fairly constant in character, but their
number varies, one or more being lost in many Irregular Echinoids.
The young are generally free-swimming plutei, but some species
are viviparous, the young being nursed in marsupial depressions
(Fig. XLIIL, see Wyv. Thomson, 81).

In spite of these great variations in structure there are several
characters common to all Echinoids. By the selection of those
structures which are found in all the Echinoids (except when they
have been lost by obviously secondary modification) it is possible to
conceive a schematic Echinoid. This is useful, as it helps us to an
idea of the primitive ancestor of the class, and as it brings into pro-
minence the features which separate the Echinoidea from the other
Echinoderma.

The Primitive Echinoid — for which various names have been
suggested — probably had a globular muscular body, covered by an
irregular series of polygonal plates. It must have had a simple
alimentary canal rising from a mouth situated at the centre of the
lower surface ; at first, possibly, it may not have had an anus,
which, when it came, opened on the upper surface. Three rings
surrounded the oesophagus, and from each ring five branches passed
up the test to the aboral pole. These three rings with their
branches formed the water-vascular, blood-vascular, and nervous
systems. Branches from the radial vessels of the water-vascular
system passed between the plates of the skeleton to the exterior
and acted as suckers. The suckers, by absorption of parts of
the plates, .at length passed out through pores, instead of through
the sutures. The perforated plates were therefore marked off
from the others and formed the five ambulacra, while the imper-
forate plates between constituted the interambulacra. A tube
connected the water-vascular ring with the exterior, and allowed
the entrance of the necessary water by a single pore. Five un-
paired gonads occupied the interambulacral areas. The primitive
Echinoid did not have either a stalk, apical system of plates,
masticatory apparatus, or perignathic girdle.

Such an animal would have been regarded as an Echinoid, as it
was not attached by the aboral surface, but on the contrary had

THE ECHINOIDEA 299

the mouth downwards ; as its gonads were quinqueradiate ; and as
its ambulacra extended from the peristome almost to the aboral
pole. The union of these three characters separates the Echinoidea
from the rest of the Echinoderma.

Proceeding to discuss the Sub-Classes, Orders, and Families of the
Sea-urchins, we may sum up their common characters in the following : —

Diagnosis of the Class.1 — The Echinoidea are eleutherozoic Echino-
derma which are actinogonidial (i.e. having the gonads quinqueradiate)
and zygopodous (the podia extending from the peristome to near the
aboral pole). The body is covered by numerous series of plates, usually
polygonal and in vertical series. The apical system may be absent,
rudimentary, well developed, or very extensive. The gonads are un-
paired and interradial. The body is spherical, or flat, or bilaterally
symmetrical, and is covered by spines which may be long, stout, and
strong, or present every stage of reduction to such as are fine and
silky. An anus is always present, but its position is variable ; but
it is either at the aboral pole or in the posterior interradius. Respiration
is partly by gills and partly by podia. Development is either direct or
indirect.

The usual primary subdivision of the Echinoidea is into two sub-
classes— PAL^EECHINOIDEA and EUECHINOIDEA ; the former including ap-
proximately all the Palaeozoic, and the latter all the Neozoic Echinoidea.
The last formal attempt to define the two groups was that of Duncan
(24, p. 4), which has been accepted by Jackson (41).

Excluding from Duncan's diagnoses characters common to the two
sub-classes, we find that the only distinction between them is that the
Palseechinoidea have either one or more than two vertical rows of plates
in each interambulacrum, and two or more vertical rows of plates in each
ambulacrum ; while the Euechinoidea have two vertical rows of plates in
each interambulacrum and in each ambulacrum.

This classification is open to two fatal objections. The rule is not
absolute. Thus in the Cretaceous genus Tetracidaris there are four rows
of plates in each interambulacrum ; and in such genera as the Euechinoid
Tripneustes, it is no more correct to say that there are only two rows of
vertical plates in the ambulacra, than it would be to say so of the Palae-
echinid Palceechinus which is described as having more than two vertical
series. In the second place, the classification separates fairly close allies,
and brings together extremely divergent forms. Thus, such a species as
the Liassic Euechinoid Gidaris edwardsi is far more closely allied to such
a Palseechinoid as Archaeocidaris than the latter is allied to Tiarechinus.

The separation of these two sub-classes was originally based on several
very definite characters, such as the imbrication of the plates, the flexi-
bility of the test, and the number of pores in the genital plates. One by
one these characters have been shown to be valueless for the purpose for
which they were used, but the classification based on them has been re-
tained. It is preferable to return to the division of the Echinoidea into
— REGULARIA and IRREGDLARIA.

1 Emended from Bell (10), p. 214.

300 THE ECHINOIDEA

SOB-CLASS 1. REQULARIA ENDOBRANCHIATA.

Mouth and anus at opposite poles. Anus surrounded by the apical
system of plates, when they are present. No external gills.

ORDER 1. Bothriocidaroida, Schmidt.

Loven has shown that in many young Echinoids the interambulacrum
begins and ends with a single plate. In the oldest known Echinoids the
whole interambulacrum consists of a single vertical series. The Echinoids
in question are two species from the Ordovician rocks of Esthonia. They
belong to the genus Bothriocidaris, Schmidt.

Bothriocidaris has a small test, on the top of which is an apical
system (Fig. XIV. 2), composed of a ring of five large ocular plates, in
the angles between which are five small imperforate genital plates.
Each ocular plate has two pores. The anus is in the centre of the
apical system, and the periproct is covered by six or eight anal plates.
The test is mainly formed by the ambulacra. Each ambulacrum consists

. Bothriocidaris Pahleni, Schmidt ; Ordovician, Russia. 1, from the side ; 2, apical system ;
3, peristoiniiil plates. The interambulacral plates are shaded.

of two vertical series of hexagonal plates, each perforated by one or two
pore-pairs. The interambulacra are narrower than the ambulacra, and
consist of a single series of plates which do not extend to the peristome,
from which they are cut off by the expansion of the peristomial ambulacral
plates (Fig. XIV. 3). The most recent description of the genus is by
Jaekel (42).

ORDER 2. Cystocidaroida, Zittel.

Echinoidea Regularia Endobranchiata with test ovoid, flexible. No
apical system of plates. Madreporite and anus (when present) open in-
dependently in the posterior interambulacrum. Mouth central. Ambu-
lacra of low, closely packed plates. Interambulacra broad, of numerous,
thin, and irregularly arranged plates, bearing short spines. A masticatory
apparatus present.

FAMILY 1. PALAEODISCIDAE. Cystocidaroida with depressed, discoid
body. The ambulacral plates are biserial, crowded, and narrow ; on the
oral surface they are not perforated by pores, but the podia pass out

THE ECHINOIDEA 301

through the sutures between the plates. Near the aboral pole the
ambulacra are narrower and pores occur in the plates. Genus —
Palaeodiscus, Salter, from the Silurian of Ludlow. It is the most
primitive of known Echinoids and has been frequently assigned to the
Stelleroida. The main radial water-vascular vessels appear, however,
to have passed along the inside of the test instead of below or outside
the ambulacral plates, ae in Stelleroids (Salter, 42 on p. 280, ante; Neu-
mayr, 64 ; Gregory, 36).

FAMILY 2. ECHINQCYSTIDAE. Cystocidaroida, in which the ambulacra
consist of narrow plates, each perforated by a pore-pair. The pore-pairs
are biserial ; most of the plates are low primaries, but demi-plates also
occur. Genus — Echinocystis, Wyv. Thorns, (non Hall), Silurian; one of
the most remarkable of known Echinoids. It has no apical system of
plates, and the anus and madreporite both open independently in the
posterior interradius. The genus is therefore sometimes described as
exocyclic, but it is really acyclic (Gregory, 36).

ORDER 3. Cidaroida.

Echinoidea Regularia Endobranchiata, in which the peristome is
central ; the periproct is central on the aboral surface of the body, and is
surrounded by the apical system of plates. The ambulacra each consist
of two vertical series of simple narrow plates, some of which may be
demi-plates. The interambulacral plates are unituberculate, bearing large
spines. There is a dental apparatus.

In the Devonian system the Echinoids are scarce, but their characters in-
dicate a marked advance upon the Silurian species in the strength of the tests,
owing to the greater thickness
and regularity of the plates.
Two main lines of differen-
tiation are apparent. In the
first the increase takes place
in the interambulacral plates,
in the second the ambulacral
plates become more import-
ant. The former is the order
Cidaroida, the latter is the
order Melonitoida.

There are four families
of Cidaroida, of which three
are extinct. The most typical
genus is Cidaris ; the earliest D of ^^ of plates near peri8tome

and most primitive IS Lepido- in Lepidocentrus. Devonian, Germany.

centrus.

FAMILY 1. LEPIDOCENTRIDAE. Cidaroida with ambulacral pore-pairs
in a single series. Interambulacral plates in more than two vertical rows.
Test flexible, owing to imbrication of the plates. No interambulacral
plates pass on to the peristomial membrane. This family is represented
by four Palaeozoic genera — Lepidocentrus, Miiller, Devonian ; Lepidechinus,

302

THE ECH1NOIDEA

Hall, Devonian and Lower Carboniferous ; Koninckocidaris, Dollo &
Buisseret, and Perischodomus, M'Coy, both Carboniferous. The main
character of the family is that none of the interambulacral plates occur
detached from the test on the peristomial membrane (Fig. XV.).1

FAMILY 2. ARCHAEOCIDARIDAE. Cidaroida with ambulacral pore-
pairs in a single series. Interambulacral plates in more than two vertical
rows. Test slightly flexible, owing to slight imbrication of plates.
Peristome large, several rows of the interambulacral plates as well as of
the ambulacral passing on to the peristomial membrane. The main char-
acter of this family is that, while the interambulacral plates remain in
more than two series and somewhat imbricated, in both of which
features it agrees with- the Lepidocentridae, it has acquired the peristomial
characters of the true Cidaridae. Genera — Archaeocidaris, M'Coy (Fig.

Fio. XVI.

1, Palceechiniis sphaeriwts, M'Coy; Carboniferous. 2, A plate and spine of Archaeocidaris
iirii, Fleni. ; Carboniferous. 3, Cidaris ylandifera, Golclf. ; Jurassic. 4, Hemicidaris intermedia,
Fleiu. ; Mid. Jurassic. 5, Salenia ]>etalifera, Uesni. ; Cretaceous. 6, Dysaster ringens, Ag. ;
Jurassic. 7, Enallaster Greenovi, Forbes ; Cretaceous. 8, Catopyffiis columbarius, Lam. ;
Cretaceous.

XVI. 2), and Lepidocidaris, Meek & Worthen, both Carboniferous. In
the latter some of the ambulacral plates are demi-plates. Xenocidaris,
known from spines only, may also belong here.

FAMILY 3. CIDARIDAE. Cidaroida with ambulacral pore-pairs uni-
serial and plates all primaries. Interambulacral plates in two vertical
series in each area. Test rigid, as the plates do not imbricate. Several
rows of interambulacral and ambulacral plates pass on to the peristomial
membrane. The family includes the living genus Cidaris, Leske, with
its numerous subdivisions — Rhabdoridaris, Chondrocidaris, Stereocidaris,
Discoddaris, Tylocidaris, Typocidaris, Dorocidaris, etc. Goniocidaris, Des.;
Orthocidaris, Cott ; Temnocidaris, Cott ; Polycidaris, Quenst, are also
genera of this family. Cidaris is one of the most primitive of recent
Echinoids, and therefore one of the most instructive.

(41).

The systematic value of this character is shown in Jackson's interesting paper

THE ECHINOIDEA

3<>3

Fig. XVII. gives the aboral surface of Cidaris (Stereocidaris) subvesi-
culosa, from the Chalk, showing its primitive dicyclic apical system ; its
massive interambulacral plates separated by very narrow ambulacra, com-
posed only of low, simple primaries (Fig. XII. 1). The arrangement of
the peristomial plates in this genus is very important ; the peristomial
membrane is covered by loose plates which include representatives of
both the ambulacral and interambulacral series. There are no arched
processes over the ambulacra, the perignathic girdle consisting only of
interradial ridges (Fig. XIII. 1). The internal gills,1 or Stewart's organs,
are well developed.

FAMILY 4. DIPLOCIDARIDAE. Cidaroida with ambulacral pore-pairs
biserial. Interambulacral plates in two or more vertical series in each

Fiu. XVII.
Cidaris (Stertotidaris) suhvesiculvsar d'Orbigny ; from the Chalk.

area. Peristomial plates as in Cidaridae. This family includes the
interesting genus Tetracidaris, Cotteau, which has four rows of plates in
each interarnbulacrum. This Paloeechinoid character is associated with
a type of ambulacrum which, for the Cidaridae, is remarkably specialised.
A second genus is Diplocidaris, Desor.

ORDER 4. Melonitoida.

Echinoidea Regularia Endobranchiata, in which the peristome is cen-
tral on the lower surface, and the periproct central on the upper surface,
surrounded by the apical system of plates. The ambulacra each consist of
two or more rows of simple plates, of which some or all may be demi-
plates, or klasma-plates. The interambulacral plates are covered by
granules bearing short, eimple spines ; but occasional tubercles may occur.

1 It should be remembered that the respiratory function of these organs is still
hypothetical.

304 THE ECHINOIDEA

There are more than two vertical series in each interambulacrum. There
is a masticatory apparatus, but no external gills. The order is therefore
endocyclic, gnathostomate, anectobranchiate, with simple ambulacral
plates, and having granulate interambulacrals.

This order represents an offshoot from the main Cidarid stem. It
began in the Silurian, attained its maximum in the Carboniferous, and
became extinct in the Permian. There are three families in the order,
and these form an evolutionary series. All differ from the Cidaroida,
by having granular instead of unituberculate plates, which, by itself,
however, is not a character of ordinal importance. The main feature
is the great increase in the importance of the ambulacral areas, reminding
us of Bothriocidaris. This character is well developed in the two more
specialised families, but in the Palaeechinidae it is only just appearing.
Thus the family named is closely allied to the members of the Cidaroida,
and is separated from that order only as it marks the beginning of a
very remarkable type of Echinoid structure.

FAMILY 1. PAL/EECHINIDAE. Melonitoida, in which the ambulacral
plates are essentially biserial (or in one case triserial). Most of the
plates are primaries, and the remainder long, narrow, demi-plates. The
plates of the test are rigidly attached. One row of interambulacral
plates passes on to the peristomial membrane. Genera — Palceechinus,
M'Coy (pars), and Rhoechinus, Keeping; and perhaps also Perischocidaris,
Neumayr (syn. Homotoeclius, Sollas). The family is separated from the
Melonitidae, owing to the great difference in the characters of the
ambulacra; but it is regarded as the ancestral group from which that
family was derived. RJwechinus is the simplest genus, and includes those
with the pore-pairs in a single series. Palceechinus, which ranges from
the Silurian to the Carboniferous, includes those in which the pore-pairs
are biserial, and demi-plates occur (Fig. XVI. 1).

FAMILY 2. MELONITIDAE. Melonitoida, in which the ambulacral
plates are all small, simple klasma-plates, which are multiserial in arrange-
ment. These form broad areas. The test is rigid. One row of inter-
ambulacral plates passes on to the pBiistomial membrane. This family
represents a marked advance on the previous one. The tendency
towards the crowding of the ambulacral plates and the reduction of many
of them into klasma-plates has made great progress. Genera — Oligoporus,
Meek & Worthen, Carboniferous ; the plates agree in general character
with those of Palceechinus, but the ambulacral plates are quadriserial
instead of biserial. Melonites, Norwood & Owen, Carboniferous; the
process has gone further, and each ambulacrum consists of from six
to sixteen vertical series.

FAMILY 3. LEPIDESTHIDAE. Melonitoida, in which the ambulacral
plates are small klasma-plates, multiserial in arrangement. The plates
of the test imbricate. None of the interambulacral plates pass on to the
peristomial membrane. This family is the extreme type of the Meloni-
toida, and represents a condition in which the plate arrangement becomes
most irregular. It includes the species with the greatest number of plates
in the ambulacra. The plates being thin and small, the test is necessarily
fragile, a danger to the animal obviated by the imbrication of the plates.

THE ECHINOIDEA 305

As is the rule among all thin-plated, flexible Echinoids, there is a marked
tendency to irregularity, especially shown in Pholidocidaris. Genera —
Pholidoddaris and Lepidcsthes, both of Meek & Worthen, from the Lower
Carboniferous.

ORDER 5. Plesiocidaroida, Duncan.

Echinoidea Regularia Endobranchiata with a small rigid test ,
peristome and periproct central and opposite. Periproct in the centre
of an apical system of large plates, which forms half of the whole
test. The ambulacral areas are short and biserial. Their plates are
all simple primaries. The interambulacra have each a single peristomial
plate.

FAMILY 1. TIARECHINIDAE. Plesiocidaroida with desmactinic ambu-
lacra (£«. ambulacra continuous from peristome to apical system). Each
interambulacrum consists of four plates, viz. a single peristomial plate
and three tall, vertical plates in a horizontal row. Genus — Tiarechinus,

Tiarechinus princeps, Neuuiayr ; Trias, Tyrol. 1, from the side ; 2, from below ;
3, the apical system (magnified).

Neumayr ; Trias, Tyrol. The figures (Fig. XVIII.) show its enormous
apical disc, small ambulacra, and vertical interambulacral plates.

FAMILY 2. LYSECHINIDAE. Plesiocidaroida with ambulacra limited to
grooves on lower surface of the test Each interambulacrum begins with
a single peristomial plate succeeded by a row of two plates, and this by
one or more containing three plates. Genus — Lysechinus, Greg. (34) ;
Trias, Tyrol.

This small order includes the two most aberrant of all known
Echinoids. In Tiarechinus there are three vertical plates in each inter-
ambulacrum, while the calyx is much larger and more crinoidal in
aspect than in any other Echinoid. It has hence been regarded as an
argument in favour of the origin of the Echinoidea from an ancestor in
which the apical system was of great importance. Both known genera of
Plesiocidaroida are small forms, and they appear to have lived under
unfavourable conditions in Triassic lagoons, for the Echinoids with which
they are associated are all dwarfed. Hence it seems more natural to
dismiss Tiarechinus and Lysechinus as two aberrant genera, in which the
test was strengthened by the development of the apical plates. Thus they
have no bearing on the character of the ancestral Echinoid.

20

306 THE ECHINOIDEA

SUB-CLASS 2. REGULARIA ECTOBRANCHIATA.

Ecliinoidea with mouth and anus at opposite poles. Endocyclic, with
external gills.

ORDER 1. Diademoida, Duncan.

Echinoidea Regularia Ectobranchiata in which the mouth and anus
are both central and opposite. The anus opens in the centre of the apical
system (which may however be rudimentary). The external branchiae
pass out through the buccal clefts. A dental apparatus is present.
There are no interambulacral plates on the peristomial membrane. The
ambulacral plates are generally compound.

This order marks a great advance on any of those previously
denned. The ambulacral plates in some forms remain as simple
primaries, but in the majority they unite into compound plates, different
from anything met with iiA the preceding groups. At the same time
external gills appear, with or without internal gills, and none of the
interambulacral plates occur on the peristomial membrane. The order is
accepted practically as denned by Duncan,1 but his method of division is
not followed.

SUB-ORDKR 1. CALYCINA.

Diademoida in which the apical system is very large and includes
one or more supplementary suranal plates.

As we have seen la the description of Tiarcchinus and Lyscchinus,
these small forms gained strength by the development of a series of
large apical plates. In one group of the Diademoida the same result
is obtained by the incorporation of one or more "suranal plates"
in the apical system Like Tiarcchinus,
the Echinoids in which this feature first
appeared are very small. The character has
persisted from the Trias to the present time,
but the Echinoidc in which it occurs are never
large.

FAMILY 1. SALENIIDAE. Calycina in
which th: ambulacral plates are simple
primaries, and there is one large suranal
plate in the apical system. Genera — tialenia,
Gray ; Heterosaletiia, Cott ; Peltastes, L. Ag.
no. xix. (Fig> XVI. 5 ; Fig. XIX.) ; Goniopharvs, L. Ag. ;

Salenia ***)££ Abactinal Baucria, Ebert.

FAMILY 2. ACROSALENIIDAE. Calycina in

which there are one or more suranal plates in the apical system; the
ambulacral plates are simple primaries near the apical system, but com-
pound, with demi-plates near the peristome. Genus — Acrosalenia, L. Ag.

1 The periguathic girdle is not always continuous as stated in his diagnosis. —
Duncan (24), p. 24.

THE ECHINO1DEA 307

SUB-ORDER 2. ARBACINA.

Diademoida in which the ambulacral plates are simple primaries
near the apical area ; at the ambitus they are compound. Some or
all of the compound plates consist of a large central primary, on either
side of which is a small demi-plate. (These plates are on the " arbacioid "
type of Duncan.)

The Echinoids of this sub-order contain forms characterised by remark-
able simplicity of structure. The interambulacral plates are large and
generally of the Cidarid type. The peristome is large. The ambulacra
ure narrow except on the ambitus and near the peristome, where they
often expand somewhat suddenly. The apical system is large and simple.
There are two families, one of which is typically Jurassic, and the
other typically Cainozoic. This difference in age has probably delayed
the recognition of the resemblances between the two families. There
are, however, several Cretaceous genera which link the Jurassic and the
Cainozoic forms, and thus support the idea that the Arbaciidae are the
descendants of the Hemicidaridae.

FAMILY 1. HEMICIDARIDAE. Arbacina in which the ambulacral
plates are narrow, and consist of simple primary plates for some
distance from the apex. The compound plates are few in number,
and irregular in arrangement ; the arbacioid
type of plate is not extensively developed,
many of the compound plates being dia-
demoid, though with the sutures approach-
ing the arbacioid character (Fig. XX.). This
family is not well marked off from the
Arbaciidae. It represents the characters of
that family imperfectly developed. Genera
—Hemicidaris, L. Ag. (Fig. XVI. 4 ; Fig.
XX.), is the most important ; Acrocidaris,
L. Ag.; Gonio,,,jylu,, L. Ag. j flfaa^^aSV.SSEBSfc

Polliel ; GidarOpVW. Cott. ; GlypttCUS. L. Ag. ; Klein. ; .Inrassic. Showing the
T A -j • t\ t. ij. i i. arbacioid structure of the plates.

LeptocidariSy Quenstedt, and several sub- (After Duncan.)
genera.

FAMILY 2. ARBACIIDAE. Arbacina with small, generally sub-conical
tests. They are ornamented by numerous granules ; a bare space occurs
in the middle of the upper part of each interambulacrum. The am-
bulacral plates are mainly on the arbacioid type, but there are some
primaries near the apical system, and a few diademoid plates between
the primaries and the arbacioid plates. Ocular pores double. This
family includes four primitive genera ; of these two are only known
living, one occurs in the North American Cainozoic, and the fourth
ranges from the Eocene to the present day. The main distinctions
from the Hemicidaridae are that in the ambulacra there are fewer
primaries and more compound plates, and that the union of the inter-
ambulacral plates is strengthened by the development of a series of
knobs and sockets. These occur on the facets of the plates, the knobs
of one plate fitting into the sockets of the next (Fig. XXI. 2).

3o8

THE ECHINOIDEA

Genera — Arbacia, Gray ; Echinocidaris, Dune, non Desmoulins ; Coelo-
pleurus, Ag. ; Podocidaris, A. Ag.

SUB-ORDER 3. DIADEMINA.

Diademoida including a series of Echinids in which the compound
ambulacral plates gradually increase in complexity. In the simplest
forms all the plates are simple primaries ; in others, some of the plates

Ooo
o o o

O O c

Fio. XXI.

Arbacia nigra. 1, ambulacra! plate ; 2, interambulaoral plate showing articular
pfta 00 and knobs (*•).

lire compound, each being formed of three primaries (the diademoid
type) ; in others, again, some of the plates consist of three or more prim-
aries and one or more demi-plates, which occur between the aboral and
middle primaries.

This is the largest of the sub -orders of regular Echinoidea and
includes important families. It represents an evolutionary series from
the primitive Eodiadema to the complex Cyphosomatidae or the
abnormal Echinothuridae. The simplest members of the sub-order can-
not be distinguished from the Saleniidae by ambulacral structure alone,

but they are clearly separated by
the absence in this group of any
suranal plate.

FAMILY 1. ORTHOPSIDAE.
Diademina with the ambulacral
plates all simple primaries, and
the pore-pairs in a simple series.
This interesting family includes
1 FIO. xxn. 2 a series of simple Echinoids,

AspidodiailematonsHm (&tter A. Agassiz). 1, from which form the beginning of the
the side; 2, from above showing the apical system 0,,i -.j--. r\-nJ « T4-

composed of a single ring of ten plates! sub - order Diademina. It in-

cludes seven genera ranging from

the Middle Lias to the present day, viz. Eodiadema, Dune. ; Archaeo-
diadema, Greg. ; Orthopsis, Cott. ; Gymnodiadema, Lor. ; Peronia, Dune. ;
EchinopsiSy L. Ag. ; Aspidodiadema, A. Ag. The only living genus is
Aspidodiadema, a deep-sea form dredged by the Challenger. It lias
been made the type of a special family by Duncan, but it seems to

THE ECHINOIDEA

309

be a survival of the Orthopsidae. The apical plates form a single
circle, within which are five large anal plates around the anus (Fig.
XXII.).

FAMILY 2. DIADEMATIDAE. Diademina in which the ambulacral
plates at the ambitus are compound, and consist of three (Fig. XXIII.)
or more (Fig. XXIV.) primaries fused together with an occasional demi-

FK;. XXI1J.

Compound ambula-
i-ral plates of Psewlo-
<liadenui hemispJuterica ;
the simple diademoicl
type of three primaries.

FK;. XXIV.

Compound ambulacral
plate of Pseudodiadema
hemisphacrica, containing a
demi-plate (rf).

<*3» >^*x_,_ ___^ ^

(Isrr0^p$

FIG. XXV.

Compound ambulacral plates
of Plncodiademn Mlchelini (after
Duncan) ; one formed of four ami
tlic other of live components.

plate (d). This family includes fourteen genera ranging from the Lias
to recent times, viz. Diadema, Gray ; Pseudodiadema, Desor (Figs. XXIII.,
XXIV.) ; Microdiadema, Cott. ; Diademopsis, Desor ; Heniipcdina, Wright ;
Kchinodiadema, Cott. ; Pleurodiadema, Lor. ; Placodiadema, Dune. (Fig.
XXV.) ; Heterodiadema, Cott. ; Codiopsis, L. Ag. ; Magnosia, Mich. ;
Cottaldia, Des. ; Centrostephanus, Peters ; Helikodiadema, Gregory.

FAMILY 3. DIPLOPODIIDAE. Diademina in which the ambulacral

d

FIG. XXVI.

Ambulacra! plates of Diplopodia
/•fi-sipora (after Uuncan), showing the
iHserial arrangement of the pore-pairs.

Fia. XXVII.

Compound ambulacral plates
of Peuina SmitJii ; one plate
consists of threo primaries,
and one of a central primary
and two demi-platos ((/) ; the
pore -pairs occur in oblique
series, with three pairs in each.

plates are compound ; the pore-pairs are biserial either throughout the
area, or at least near the peristome (Fig. XXVI.). Genera — Diplopodia,
M'Coy ; Phymechinus, Des. ; Asteropsis, Cott. ; Diplotayma, Schliiter ;
Plistophyma, Peron & Gauthier ; (?) Acanthechinus, Duncan & Sladen ;
Micropyga, A. Ag.

FAMILY 4. PEDINIDAE. Diademina in which the ambulacral plates
are compound and the pore-pairs are triserial (Fig. XXVII.). Genera —
Pedina, Ag. ; Pscndopedina, Cott. ; Heterocidaris, Cott. ; Stomechinus, Des. :

310 THE RCHINOIDEA

Polycyphn*, Ag. ; Pedinothnria, Greg. ; Micropcdina-, Cott. ; Codechinus,
Des. ; Echinopediiia, Cott. ; Echinoihrix, Pet. ; Astropyga, Gray.

FAMILY 5. CYPHOSOMATIDAE. Diademina in which the ambulacral
plates are compound ; they are high with from three to seven pore-pairs
in an arc ; the ndoral and supra-adoral, and sometimes also the aboral
plates are primaries. The remaining constituents
are demi-plates (Fig. XXVIIL). Genera — Actino-
phyma, Cott. & Gauth. ; Cyphosonw, Ag. ; Leiosoma,
Cott. & Triger ; Gauthieria, Lamb. ; Th.ylechinust
Pomel ; Coptosoma, Des. ; Micropsis, Cott. ; Orthe-
Fio xxvin chinus, Gauthier (syn. Gagaria, Dune.) ; Triplacidia,

Ambulacral plate of Bittner.
Cypkosoma, composed of In the normal compound ambulacral plates of

three primaries and two ,v TV i .L-J i i i • L a A.I

demi-plates. the Diadematidae each plate consists of three prim-

aries ; but, as in the case of Fig. XXIV., an extra

demi-plate sometimes appears below the uppermost primary. This is the
link between the typical Diadematidae and Cyphosomatidae. Demi -plates
appear in Diadematids in the Middle Oolites ; the Cyphosomatidae begin
in the Upper Oolites and attain their maximum in the Cretaceous. The
last members of the family lived in the Eocene.

FAMILY 6. ECHINOTHURIDAK. Diademina in which the test is more
or less flexible ; the plates are thin and usually separated by membrane.
Apical system rudimentary (Fig. X.) ; ambulacral plates triserial, arranged
typically in triplets of a central primary between two klasma-plates. In
»>ne genus three triplets unite together to form a single plate. This
interesting family was founded by Wyville Thomson to include the
Chalk fossil Echiiiothiiria, S. P. Woodw., and some living Echinoids
dredged by the Porcupine Expedition. As the tests are flexible and
the plates overlap, the family was at first compared with the Palaeozoic
Echinoids. P. and F. Sarasin argued from the rudimentary nature of the
apical disc, and from the great size of the " Stewart's organs " and the
presence of powerful radial muscles (the two latteA characters being very
noticeable in a new species of Asthenosoma described by those authors) that
the Echinothuridae were a primitive family of Echinoids, and helped to
establish the origin of this class from a Holothuroid ancestor. Neviani
accepted this conclusion. But, as has recently been shown (Gregory,
35), tue family is an offshoot from the Pedinidae ; the genus Pcdiiw-
thuria helps to bridge the gap between Pedina and the oldest known
Echinothurid — Pelaneckinus.

SUB-FAMILY 1. PELANECHININAE. Echinothurids of which the am-
bulacral plates are compound ; those near the apex consist of two demi-
plates and a large middle primary. Those at the ambitus consist of
three sets of three plates united into a single polyporous plate ; each
triplet of this compound plate consists of a primary between two demi-
plates or klasma- plates. Genus — Pelanechinus, Keeping; Corallian,
Wiltshire. An admirable account of the genus has been given by Groom
(37). SUB-FAMILY 2. ECHINOTHURINAE. Echinothuridae in which the
ambulacral plates are simple and free ; they consist of triplets, each com-
posed of a large central primary, and with an isolated klasma-plate above

THE ECHINOIDEA

and below it. Genera — Echinothuria, S. P. Woodw. ; from the Chalk ;
Asthenosoma, Grube (Fig. XXIX.), a living genus with large " Stewart's
organs " and powerful radial muscles ; and Phormosoma, Wy v. Thomson,
with rudimentary Stewart's organs and without powerful radial muscles!
In the Echinothurinae the reduction in the calcification of the test|
which had begun in Pelancchinus, has been carried so far, that all the
ambulacral plates are disunited, but are held together by strong muscular
lining. In Asthenosoma there is, in addition, a series of powerful radial
muscles (Fig. XXX.), which give a panting motion to the test. The

FIG. XXIX.

Asthenosoma hystrix. Oral surface ; the tips of the jaws are seen protruding through
the peristoniiul membrane.

spines are covered by epithelium, and when handled can inflict a sharp
sting.

SUB-ORDER 4. ECHININA.

Diademoida in which the ambulacral plates typically consist of an
aboral and adoral primary, between which are one or more demi-plates.
The sub-order includes a series of Echinoids, in the simplest of which the
compound ambulacral plates consist of three primaries, and are separated
from one another by free primaries.

In the sub-order Arbacina the plates of the test are often fixed
together by sockets and knobs (Fig. XXI. 2), while in some genera, such
as Glypticus, there is a great development of the subsidiary ornament. In

312

THE ECHINOIDEA

Fio. XXX.

Radial muscle of Agthenosoma.
(After P. and F. Sarasin.)

the simplest forms of the Echinina both characters are further developed. In
the Chalk genera, Glyphocyphus and Zeuglopleurus (Fig. XXXL), the sutures

are excavated by deep hollows, while the plates
are thickened by granules and ridges. From
such forms as these there is a gradual transi-
tion to others with deep pits, which dowel
into the plates, as in Temnopleurus (Fig.
XXXII.). The transition from the diademine
to echinine type of ambulacral plates is
shown by Zeuylopleurus and Ortholophus. In
the former the plates consist of three fused
primaries, separated by free primaries. The
middle primary is often very small, and in
Ortholophus is often reduced to a demi-plate.
The plates then have the arrangement typical
of Echinus. From this oligoporous plate
the polyporous strongylocentrotoid type is
produced by the development of one or
more demi-plates between the aboral and
adoral primaries.
This sub-order began in the Cretaceous.

FAMILY 1. TEMNOPLEURIDAE. Echinina in which the compound
ambulacral plates are formed of three constituents. In the oldest and
most primitive forms the three plates are all primaries ; in the later and
more specialised types the middle plate is crowded into a demi-plate
(i.e. the plates are on the Echinoid type). There is a great development
of superficial ornamentation, and the plates are
hollowed or undermined by depressions or pits.
SUB -FAMILY 1. GLYPHOCYPHINAE. Temnopleuridae
in which the compound ambulacral plates are com-
posed of three primaries ; the plates are united by
do welling, but there are no sutural pits. Genera —
Glyphocyphus, Haime (syn. Rhabdopleurus, Cott.) ;
Zeuglopleurus, Greg. ; Echinoci/phus, Cott. ; Paradox-
echinus, Laube ; Leiocyphus, Cott. SUB- FAMILY 2.
ORTHOLOPHINAE. Temnopleuridae in which the com-
pound ambulacral plates are composed of two primaries
and an intermediate demi-plate. The plates are united
by dowelling, but there are no sutural pits. Genera
— Ortholophus, Dune. ; Coptophyma, Peron & Gauth. ;
Lepidopleurus, Dune. & Slad. ; (?) Trigonocidaris, A.
Ag. ; Dictyopleurus, Dune. & Slad. ; Arachnioplev.rus,
Dune. & Slad. (Radiocyphus, Cott). SUB-FAMILY 3.
TKMNOPLEURINAE. Temnopleuridae in which the
compound ambulacral plates are composed of two lacrum" showing simple
primaries and an intermediate demi-plate. True JJJj
sutural pits occur, and these often undermine the
plates (Fig. XXXII.). Genera — Temnopleuru*, Ag. ; Temnechinus,
Forbes ; Opechvnui, Desor ; Pleurechinus, Ag. ; Salmacis, Ag. ; Salma-

Fio. XXXI.

Zeui
tus,
upper part of an ambu-

Zeuglopleurus costula-
s, Greg. ; plates of

THE ECHINOIDEA 313

copsis, Doderlein ; Mespilia, Desor ; Microcyphus, Ag. ; Amblypncustes,
Ag. ; Goniopneustes, Dune. ; (?) Holopneustes, Ag. ; (?) Grammechinus, Dune.
& Slad.

FAMILY 2. TRIPLECHINIDAE. Echinina in which the arabulacral plates
consist of two primaries and an intermediate demi-plate. The three pairs
of pores are arranged in arcs of triplets ; the sutural faces of all plates
are smooth ; and there are no pits or grooves in their substance ; so that
in these three respects the Triplechinidae differ from the Temnopleuridae,
though an approach to this family is shown by Grammechinus. Genera —
Echinus, Linn. ; Psammechinus, Ag. ; Micropsina, Cott. ; Leiopedina, Cott. ;
Tripneustes, Ag. ( = Hipponoe) ; StirechinitA, Desor ; Glyptechinns, Lor. ;
Hybechinus, Desor ; Toxopneustes, A. Ag. ; Boletia, Desor ; Evechinus,
Verrill ; Pedinopsis, Cott. .

Fio. XXXII.
Temnopleurus tweumaticus (after Agassiz), showing the pitted test.

FAMILY 3. STRONGYLOCENTROTIDAE. Echinina with more than three
constituents in each ambulacral plate, the pores being in high curved
arcs. Genera — Strongylocentrotus, Brandt ; Sphaerechinus, Desor ; Echino-
strephuSj A. Ag. ; Eurypneustes, Duncan & Sladen ; Pseudobolctia, Trosch. ;
Aeolopneustes, Dune. & Slad.

FAMILY 4. ECHINOMETRIDAE. Echinina with three or more con-
stituents in each ambulacral plate. The test is elongate, and the long
axis does not coincide with the antero-posterior axis. Genera — Echino-
metra. Gray; Stomopneustes, Ag. ; Heterocentrotus and Colobocentrotiis,
Brandt ; Parasalenia, A. Ag. The elliptical shape of the test is the most
remarkable character in this family. The perignathic girdle is very
powerful, each arch being surmounted by a cap (Fig. XIII. 4). The
spines are large and very varied in form. In Heterocentrotus the
secondary spines form a fur below the primary spines (Fig. XXXIII.) ;
in Colobocentrotus the spines are stout and end in flat surfaces ; they
are so crowded together as to form a natural armour-plate (Fig.

THE ECHINOWEA

Fio. XXXIII.
Hcterocentrotitf mammillatus (after Agassiz), showing relation of primary and secondary spines.

Fio. XXXIV.
Colobocentrotui atratug, showing tlie tessellate arrangement of the spines.

XXXIV.)- The spines of the Triassic genus, Anaulocidaris, Zitt., had
a similar arrangement.

THE ECHINOIDEA

315

SUB-CLASS 3. IRREGULAUIA, Desor.

Echinoidea in whicli the anus lies outside the apical system of plates
in the posterior interradius.

ORDER 1. Gnathostomata, Zittel.

Echinoidea Irregularia with a central peristome surrounded by a
perignathic girdle : jaws present, but sometimes rudimentary. Ambu-
lacra all similar.

This is the first of the two orders of Irregular Echinoidea, and
differs from the other order — the Spaiangoida — by the presence of a
perignathic girdle and jaws. As in the Regular Echinoids, there is a
marked tendency for the anus to pass backward out of the apical
plates. In Pygaster the peristome is much like that of Stomechinus,
and the ambulacral plates are sometimes compound ; the jaws are fragile,
but otherwise normal. The only
character that excludes the genus
from the Diademoida is that
the anus opens outside the apical
plates (Fig. XXXV. 1). Pygaster is
thus the nearest form \ve know
to the ancestor of all the Clype-
astroid and Spatangoid sea-urchins.

The order Gnathostomata in-
cludes three main series. The first
was typically Mesozoic, and was
characterised by the reduction in
the functional importance of the
jaws, and the formation of the
perignathic girdle into a high tubu-
lar peristome. From this series

hrnnrhpq divorced in nrmnqitp Irregular Echinoidea. 1, Pygaster semisul-

rittnx, Phil. ; Jurassic. «, Echinaeorys scvtotvs,
directions. In one the jaws dis- Leske; Chalk. 3, Galeritts allioffalerus, Leske ;

appeared and the perignathic girdle : *' '*I"J'H°

became rudimentary ; while the

ambulacral plates remained as in Pyyaster. This branch culminated in
the aberrant group of the Galeritidae. Later on a second branch was
given off ; in this the jaws became of greater power ; the ambulacra
became complex, parts of them expanding into petals, the podia of which
net as branchiae. The Gnathostomata may accordingly be divided into
two sub-orders.

SUB-ORDER 1. HOLECTYPINA.

Gnathostomata in which the jaws are reduced in size and strength.
In the most primitive members the jaws are arranged as in the
Diademoida ; but in later forms they are inside a tubular perignathic
girdle. The jaws do not worK on sockets.

316 THE ECH1NOIDEA

This sub-order is difficult to characterise, as it includes the primitive
Irregular Echinoids, as well as one series of these forms which has
continued to the present day.

FAMILY 1. PYGASTERIDAE. Holectypina in which the peristome is
large, and the perignathic girdle consists of disconnected processes.
The ambulacra are simple and apetaloid. Genera — Pygaster, Ag.
(Fig. XXXV. 1) ; Pileus, Desor ; Pygastrides, Love'n ; Holectypus, Desor ;
(?) Pachyclypeus, Desor; Galeropygus, Cott. FAMILY 2. DISCOIDIIDAE.
Holectypina in which the peristome is small and the perignatliic girdle
tubular. Jaws unknown. Ambulacra apetaloid. Genera — Discoidea,
Gray ; Protocyamus,1 noni. nov. FAMILY 3. GALKRITIDAE. Holectypina
in which the perignathic girdle is rudimentary, jaws are absent, and
their place taken by ten buccal plates. Genera — Galerites, Lamk.
(Fig. XXXV. 3) ; Lanieria, Dune. ; Adelopneustes, Gauth. ; (?) Copto-
discu*, Cott. & Gauth. FAMILY 4. CONOCLYPEIDAE. Holectypina in
which the peristome is .small ; the perignathic girdle tubular and high,
surrounded by five bourrelets. Genera — Conoclypeus, Ag. ; Oviclypeiis,
Dames.

SUB-ORDER 2. CLYPEASTRINA.

Gnathostomafci in which the jaws are powerful. The teeth are
placed in pyramids, which articulate by a socket fitting on to vertical
processes ; the jaws only move horizontally, and have neither braces nor
compasses (p. 289). The ambulacra are petaloid.

This sub -order includes a series of striking variations from the
ordinary Echinoid type. Echinocyamus is the most primitive form, and
appears to have developed from an ancestor closely allied to Protocyamus.
The great advance in Echinocyamus is the expansion of parts of the
ambulacra into rudimentary petals (Fig. XXXVI.) ; in the upper part
of the ambulacra the outer pores of the pore-pairs have increased to small
slits, and occur along curved lines, enclosing somewhat leaf-shaped areas.
Beyond these only a single pore occurs in each ambulacra! plate. The
perignathic girdle of Echinocyamus consists of five vertical pegs, rising
from the interradial peristomial plates ; this reminds us of Galerites, in
which the perignathic girdle is reduced to five interradial thickenings.
The structural differences between Protocyamus and Echinocyamus are
small, and their importance is exaggerated by the different shape of the tests.
But Echinocyamus was succeeded by a very divergent series. Most of the
members of the group are long, broad, and low ; some are thin and flat.
In these cases the upper surface regains the support it loses owing to
departure from the dome-shaped form, by the development of pillars which
pass from floor to roof. The ambulacra in the typical forms are petaloid,
and the podia in these areas expand to act as branchiae (Fig. XXX VI.). In
some cases pores only occur in the petals ; in others they are scattered over
the test, occurring on both radial and interradial plates. In some genera,
such as Laganum, though the petaloid portions of the ambulacra are broad,

1 A name suggested in lieu of Echinites proposed by Duncan, but preoccupied l>y
Leske for Echinoids, and by Mullerand Troschel for Asteroidea ; the name is selected
to indicate the affinity of this cchinoid with the K'hinttcyamus series.

THE ECHINOIDEA

317

they are narrower than the ambulacral plates in the extra-petaloid portions
which expand laterally, and are much broader than the interambulacra.
In the simplest members of this group the interambulacra are, however,
" continuous " from apical system to peristome (as in Laganum) ; but in
the more advanced, such as Rotula, the interambulacra are " discontinuous,"
the ambulacra meeting one another and cutting off the interambulacral
peristomial plate from its connection with the rest of the interambulacrum.
Another feature peculiar to this sub-order is the presence of a series of
furrows on the lower surface of the test ; these are known as the actinal
furrows, and they are either straight, as in the Clypeastridae, or bifurcating,
as in the Scutellidae.

FAMILY 1. FIBULAUIIDAE. Clypeastrina with ambulacra in rudi-
mentary open petals. The interambulacral plates are continuous. The
pillars are slightly developed. The perignathic girdle consists of five

FIG. XXXVI.

Branchial podia from peta-
loid portion of ambulacrum of
Clypeaster.

1 2

FIG. XXXVII.

Eclt i nofyninii* pnsillnft. 1 , upper
surface showing simple petals and
four genital pores ; 2, the lower
surface showing the anus inter-
mediate between the central
mouth and the posterior border.

single interradial processes. The peristomial interradial plate ia large.
Genera — Echinocyamus, Phelsum (Fig. XXXVII.) ; Scutellinat Ag. ; Sis-
mondia, Desor; Fibularia, Lam.; Tkegaster, Pome).

FAMILY 2. LAGANIDAE. Clypeastrina with ambulacra petaloid ;
numerous pores for prehensile podia occur in addition to the large pores
for the respiratory podia. The interambulacral plates are " continuous " ;
the peristomial plate is medium in size and bears a single perignathic
process. The actinal furrows are simple and straight. Genus — Laganum,
Blainv.

FAMILY 3. SCUTELLIDAE. Flat Clypeastrina with closed petaloid
ambulacra. The interambulacral plates are " discontinuous " in some or
all of the areas ; the peristomial plate is large and bears a single peri-
gnathic process. The actinal furrows are bifurcating. Genera — Scutella,
Lam. (Fig. XXXV. 4) ; Echinarachnius, Leske ; Echinodiscus, Ag. ; Encope,
Ag.; Monophora, Ag. ; Mellita, Ag. ; Melitella, Dune. ; Astriclypeus, Verrill ;
Lenita, Desor ; Mortonia, Desor ; Rotula, Ag. (Fig. IX.) ; Rotuloidea,
Etheridge ; Moulinsia, Ag. ; (?) Runa, Ag. The most striking feature in
this family is the extreme thinness and flatness of the teats. In some
species, such as Scutella striatula, the test may be 100 mm. in diameter,
and only 10 mm. in height. The upper surface accordingly needs greater

318 THE ECHINOIDEA

support than it could obtain from the margin of the test ; this is given by
numerous pillars which connect the upper and lower walls. The external
margin is often notched, as in Scutella ; the notches may deepen into
" slits " separated by finger-shaped processes, as in Uotuki (Fig. IX.). In
some genera two adjacent processes unite at their free ends, and a hole
is left through the test ; such holes are known as " lunules," and occur in
Mellita, Monophom, etc. All the interambulacral areas are discontinuous
in some genera, e.g. Encope, but in Rotula and Mellita one or two of the
areas may be continuous from peristome to apex.

FAMILY 4. CLYPEASTKIDAE. Clypeastrina with closed petaloid
ambulacra. The interambulacral plates are discontinuous ; the peri-
stomial plate is small. There are two perignathic processes in each
area, and they are ambulacra! in position. The actinal furrows arc
straight. SCO-FAMILY 1. CLYPEASTUINAE. Massive Clypeastridae with
closed petals; usually high. Genera — Ctypea*Urt Lam.; Echinanthus,
Leske (= Diplotht'canthn,*, Dune.); 1'lesiantku*, Dune.; Anonutlanthus,
Bell; Monottyfhiit, Laube. SUB-FAMILY 2. AHACHXOIDINAE. Flat, low
GMypeastridae with open petals. Genera — Arachnoides, Ag. ; Alexandria^
Pfetfer. These genera are usually included as a sub- family of
Scutellidae, which they resemble in external form. Their structure,
however, allies them more nearly with the Clypeasters, with which they
agree in all fundamental characters. They differ from the Scutellidae
by having (1) a very small pcristomial interambulacral plate, which in
some species may be absent in several areas ; (2) straight, simple, actinal
furrows ; (3) five pairs of ambulacral perignathic processes.

ORDER 2. Atelostomata, Zittel.

Echinoidea Irregularia, in which there are no jaws, teeth, perignathic
girdle, or external branchiae.

The Atelostomata introduce three additional structures, upon which
the classification within the order depends. These are the sternum,
tioscelle, and fascicle. In Echinoids previously considered the mouth is
central or sub-central, and the five areas around it are of equal import-
ance; but as the mouth becomes eccentric in position, one interradius
necessarily becomes longer than the rest. The anus is situated in this
interradius, which requires some modification of the plates for the sake of
increased strength. In the simplest of the Atelostomata the plates of the
posterior interradius are but slightly different from those of the other areas
but the plates are larger, and dovetail more deeply into one another.
In Collyrites there is a slight advance on this plan, and in genera such as
Echinocorys and Holaster the plates dovetail so deeply as to form a strong
sternum along the under side of the test This type is known as the
" meridosternous " (Fig. XXXVIII. 1). In the next stage the first pair of
plates in the interambulacrum increase in length, and both are in contact
with the peristomial plate of the same area, as in Toxaster (Fig. XXXVIII. 2).
This is the "amphisternous" type, the extreme form of which we see in
Spatanyns purpureus (Fig. XXXVIII. 4).

THE ECHINOIDEA

319

We have seen that the upper parts of the ambulacra of Glype<ister, etc.,
are modified into petals. In one section of the Atelostomata there is an
analogous expansion of the ambulacra round the peristome into " fioscelles."

m
m

FIG. XXXVJU.

The types of Sputunyoid Sterna. 1, meiklosternons ; 3, 4, arnphistermms ; -2, intermediate.
1, Eckinocorys scutatus ; -2, ToMtxter ricordm tius ; 3, Caxsiilnlns pudfieux .• 4, Sputc.iHjua pur^niretty.
ni, month ; o, anus.

The peristomial interambulacral plates are raised into projecting ridges
known us " bourielets " ; while the ambulacra are expanded into leaf-shaped
areas known as " phyllodes." The pore-pairs of the phyllodes are much
larger than those of the rest of the ambulacrum. The five bourrelets
and five 'phyllodes together form the lloscelle, which is typical of the
Cassidulidae.

Among the Atelostomatu large spines like those of Cidaris are never
found. The spines are very
numerous and generally small,
forming a fur over the test.
In some cases specially modi-
fied spines occur crowded
together along bands (Figs.
XXXIX.-XLL), forming "fas-
cicles." There are five dif-
ferent varieties : (1) The
" peripetalous," which encloses
the petaloid portions of the
ambulacra; (2) the "subanal"
(«/in Fig. XXXVIII. 4), which
encloses a space, or "plastron," below the anus; (3) the "marginal," along
the border of the test ; (4) the " internal," which crosses the petaloid
portions of the ambulacra ; and (5) the two " lateral," which run from
the peripetalous to below the anus. All five kinds of fascicles never
occur together in the same Echinoid.

FIG. XXXIX.

Simple fascicle on Agassizia.

320

THE ECHINOIDEA

The Atelostomata include two main divisions, which develop along
somewhat parallel lines.

Fin. XL.
Compound fasciole on part of test of MftcroptUKtte*. (After A^assiz.)

SUB-ORDER 1. ASTEHXATA.

Atelostomata with the peristorne central and never bilabiate ; the
ambulacra simple, sub-petaloid, or petaloid, and generally all five are
similar. Floscelle generally present. No sternum.

FAMILY 1. ECHINONBIDAE. Asternata with narrow, apetaloid, similar
ambulacra, and without floscelle. Genera — Echinoneus, Van Phelsum ;
Galeroclypeus, Cott. ; Galeropygus, Cott. ; Hyboclypeus, Ag. ; Infraclypeus,

an

s.a.

Fio. XLI.

o, pen

Diagram of a Spatangoid Kchinoid, showing arrangement of the fascioles. i, internal fas
)eripetalous fasciole ; lt lateral fasciole ; w, marginal fasciole ; s.a, subanal fasciole ; a.n,

internal fasciole ;
anua.

Qauth. ; Nucleopygus, Ag. ; Pileus, Desor ; Pachyclypeus, Desor ; Pyrina,
Desmoulins.

FAMILY 2. NUCLEOLITIDAE. Asternata with sub-petaloid ambulacra
and no floscelle. Genera — Amblypygus, Ag. ; Anochanus, Grube ; Anor-
thopygus, Cott. ; Botriopygus, d'Orb. ; Caratomus, Ag. ; (1) Desordla, Cott. ;
Haimea, Michelin ; Ilariona, Dames ; Nucleolites, Lam. (syn. EchinobrUsiui) ;
Oligopodia, Dune. ; Oligopygw, Lor. ; Pygaulus, Ag. ; Trematopyguay d'Orb.

THE ECHIN01DEA 321

FAMILY 3. CASSIDDLIDAE. Asternata with closed, petaloid ambulacra
a floscelle is present. SUB-FAMILY 1. CLYPEIN A. Genera — Clypeus, Leske ;
Clypeopygus and Faujasia, d'Orb. ; Pseudodesorella, Etallon ; Pygurostoma,
Cott. & Gauth. ; Pygurus, Ag. SUB-FAMILY 2. CASSIDULINAE. Genera —
Australanthus, Bittner ; Breynella, Greg. ; Cassidulus, Lam. ; Eurhodia,
Archiac & Haime ; Paralampas, Dune. & Slad. ; Pygorhynchns, Ag. ;
Ehynchopygus, d'Orb. ; Stigmatopygus, d'Orb. S^J- FAMILY 3. CATO-
PYGINAE. Genera — Catopygus, Ag. (Fig. XVI. 8) ; Neocatopygus, Dune.
& Slad. ; Phyllobrissus, Cott. ; Pseudocatopygus, Cott. & Gauth. ; Stu-
derui, Dune. SUB -FAMILY 4. ECHINOLAMPINAE. Genera — Conolampas,
A. Ag. ; Echinolampas, Gray; Heteroclypeus, Cott.; Microlampas, Cott;
Craterolampas, Cott. ; Milletia, Dune. ; Neolarnpas, A. Ag. ; Oriolampas,
Munier-Chalmas ; Palaeolampas, Bell ; Phylloclypeus, Lor. ; Plesiolampas,
Dune. & Slad. ; Vologesia, Cott. & Gauth. SUB- FAMILY 5. EOLAM-
PINAE. Genera — Archiacia, Ag. ; Asterostoma, Ag. ; Claviaster, d'Orb. ;
Eolampas, Dune. & Slad.

SUB-ORDER 2. STERNATA.

Atelostomata with the peristome eccentric anteriorly (usually bilabiate).
No floscelle ; anterior ambulacrum different from the rest. A sternum is
present. Fascicles sometimes present.

FAMILY 1. COLLYRITIDAE. Sternata without floscelle. There is a
rudimentary meridosternum. The anterior ambulacrum is narrower
than the others. Apical system disjunct ; the three anterior ambulacra
grouped together as the "trivium," and the two postero - lateral am-
bulacra as the "bivium." There are no fascicles. Genera — Collyrites,
Desmoulins ; Dysaster, Ag. (Fig. XVI. 6) ; Grasia and (?) Metaporhimis,
Michelin ; Pygorhytis, Pomel. Owing to the disjunct apical system, this
family has completely lost the radial symmetry, and presents some
remarkable resemblances to the Pourtalesiidae. It appears, however,
probable that while the Collyritidae have descended from some primitive
asternate form allied to Hyboclypus, the Pourtalesiidae are degenerate
forms of Prymnodesmian Sternata.

FAMILY 2. ECHINOCORYTHIDAE. Meridosternous, labiate Sternata, with
an elongate apical system, and the ambulacra separated into a bivium
and trivium. Fascicles present in some genera. Genera — Calymne,
Wyv. Thorns. ; Cardiaster, Forbes; Coraster, Cott. ; Cystechinus, A. Ag. ;
Echinocorys, Leske (syn. Ananchytes, Lam., Fig. XXXV. 2) ; Enallo-
pneustes, Pomel ; Enichaster, Lor. ; Entomaster, Gauth. ; Galeaster, Seunes ;
Guettaria, Gauth. ; Hagenovia, Dune. ; Hemipneustes, Ag. ; Holaster, Ag.
(sub -gen. Sternotaxisj Lamb.); Infulaster, Hagenow ; Jeronia, Seunes;
Lampadaster, Cott. ; Offaster, Desor ; Oolaster, Laube ; Ovulaster, Cott. ;
Stegaster, Pomel ; Stenonia, Desor ; Tholaster, Seunes ; Urechinus, A. Ag.

FAMILY 3. SPATANGIDAE. Sternata with anterior ambulacrum re-
duced ; apical system compact ; sternum either amphisternous or merido-
sternous. SECTION 1. ADETINAE. Spatangidae without fascicles. Genera —
Archaeopneustes, Greg. ; Clypeanthus, Cott. ; Echinocrepis, A. Ag. ; Enallaster,
d'Orb. (Fig. XVI. 7) ; Epiaster, d'Orb. ; Genicopatagus, A. Ag. ; Hemipatctgus,
Desor; Heterolampas, Cott.; Isaster, Desor ; Macraster, Roem. ; (?) Megalaster,

322

THE ECHINOIDEA

Dune. ; Palaeobrissus, A. Ag. ; Palaeopneustes, A. Ag. ; Platybrissus, Grube ;
Spatagocystis, A. Ag. ; Toxaster, Ag. SECTION 2. PRYMNADETINAE. Spa-

Fia. XIJI.

Brissopsis lyrifera, Forbes, sp., showing podia in action, and peripetalous fascicle.
(After Loven).

Fio. XLIII.
Hemiaster Philippii, with young echinoids in the marsupia. (After Wyville Thomson.)

tangidae with fascicles, but no subanal fascicle. Genera — Abatus, Desor ;
Aceste and Aerope, Wyv. Thorns. ; Agassizia, Valentin ; Brissopsis, Ag.
(Fig. XLII.); Coraster, Cott. ; Dipneustes, Cott. ; Faorina, Gray; Hemi-

THE ECHINOIDEA

323

aster, Desor (Fig. XLIII.) ; Homoeaster and Hypsospatagus, Pomel ;
Iraniaster, Cott. & Gauth. ; Lambertiaster, Gauth. ; Linthia, Merian ;
Moira, A. Ag. ; Moiropsis, A. Ag. ; Ornithast&r, Cott. ; Pericosmus,
Ag. ; Prenaster, Desor ; Schizaster, Ag. SECTION 3. PRYMNODESMINAE.
Spatangidae with subanal fascicle. Genera — Argopatagus, A. Ag. ;
Brissopatagus, Cott. ; Brissus, Leske ; Breynia, Desor ; Cionobrissus, A. Ag. ;
Cleistechinus, Lor. ; Cydaster, Cott. ; Echinocardium, Gray ; Eupatagus, Ag. ;
Gaultieria, Desor ; Gibbaster, Gauth. ; Homolampas, A. Ag. ; Isopneustes,
Pomel ; Linopneustes, A. Ag. ; Loveuia, Ag. & Desor ; Linchophorus,
Dames ; Macropneustes, Ag. ; Maretia, Meoma, and Metalia, Gray ; Jfi-
craster, Ag. ; Nacospatangus, A. Ag. ; Neopneustes, Dune. ; Palaeotropus,

UI

FIG. XLIV.

Peristomial region of S^mtangus jmrpnrcu* setMi from inside the test (after Lovon). I-V, the
tive ambulacra in which the vertical series are each marked a or &.

Lov^n ; Pygospatangus, Cott. ; lihinoibrissus, A. Ag. ; Sarsella, Pomel ;
Spatangomorpha, Bohm ; Spatangus, Leske (Figs. XLIV., XLV.) ; Stomo-
porus, Cott. ; Tuberaster, Peron & Gauthier.

FAMILY 4. PALAEOSTOMIDAE. Sternata with a pentagonal, alabiate
peristome, provided with five buccal plates ; a peripetalous fascicle.
Genus — Palaeostoma, Lov^n (Leskia, Gray).

FAMILY 5. POURTALESIIDAE. Sternata with apetalous, flush ambulacra.
Peristome in a deep anterior recess. Form elongate ; flat or oral surface.
The ambulacral plates are uniporous (Fig. XLVI.). This family includes
perhaps the three most perplexing of recent Echinoids ; but owing to the
extreme fragility of the tests their study is difficult, and owing to their
great variability the classification is at present unsatisfactory. The typical
genus Pourtalesia is the subject of an elaborate memoir by Lov6n (58). It

324

THE ECHINOIDEA

has a disconnected apical system, the postero-lateral interambulacry meet-
ing across it. In this and some other respects it resembles Collyrites ; but
it has a sub-anal fascicle, and is probably to be regarded as a degenerate
Spatangid rather than a direct descendant of the Meso/oic Collyritidae.

J'xu XLV.
The Heart Urehiu (Spata/

Abactinal side.

The second genus Spatagocystis agrees with Pourtalesia in the disruption
of the apical system, but it has no fasciole. The third genus Echinocrcpi*
also has no fasciole, but the apical system is compact.

FIG. XLV I.
Pourtalesia Jefreysi (Wyv. Thomson).

We are now in a position to discuss briefly, first, the relations
of the Echinoidea to the other classes of Echinoderma, and
secondly, the lines of evolution within the class itself. (Compare
Chapter VIII. pp. 17, 33.)

THE ECHINOIDEA 325

Each of the main groups of Echinoderma has been at one time
regarded as the ancestor or nearest ally of the Echinoids, and the
question is still highly conjectural. Embryology gives very little
assistance. Study of the development of a young Echinothurid,
Echinocyamus, or Hemiaster (Fig. XL VII.) teaches important lessons
as to the affinities between those forms and other Echinoids. It
shows that the young Echinothurid resembles the Diademoida,
and that the young Hemiaster is endocyclic. But the earlier larval
stages have been so affected
by secondary variations
that they give no satis-
factory information as to
whether the Echinoids are
nearest to the Cystids,
Crinoids, Holothurians, or
Stelleroids.

The Crinoids are so
unlike the Echinoids in ap-
pearance and structure,
that we know of no form Flc- X

that ar>npnr<3 tn linV flip Larval form of Hcmiaste.r camrnosa, in the endocyclic

mat appears to 8tage- (Aftcr Agassiz.)

two classes. ' Neverthe-
less, the Echinoidea have been regarded as descended from a
Crinoid-like ancestor. The acquirement of a radial symmetry
was unquestionably the most important event in the develop-
ment of the ancestral Echinoderm ; it is easiest to explain this
as the result of fixation, and therefore the fixed, stalked forms
have been claimed as the ancestors of the free forms. It is further
argued that this conclusion is supported by the occurrence on the
abactinal side of some Echinoids and Stelleroids of a series of plates
known as the apical system. This system includes a central plate
surrounded by two circles of plates. The theory has been urged,
especially by the late P. H. Carpenter and by W. P. Sladen, that
the plates of this apical system are homologous with those of the
calyx of the Pelmatozoa, and are to be regarded as relics from a
period when these plates were of great functional importance.
Unfortunately for this view, however, the calycinal or apical plates
are either absent or unimportant in the oldest Echinoids and
Asteroids ; and it is in later groups, such as the Saleniidae and
Cidaridae, that the plates are developed on the supposed ancestral
plan. Moreover, instead of Tiarechinus — in which the apical plates
are most important — being ancestral, it is almost certainly an
aberrant, and somewhat degenerate offshoot.

The last blow to the idea of the apical plates of Echinoids
being homologous with the calyx plates of Crinoids, has been
given by MacBride, who, on embryological grounds, urges that

326 THE ECHINOIDEA

the abactinal poles of Asteroids and Crinoids, on which these plates
develop, have nothing to do with one another. MacBride recognises
the influence of fixation on the anatomy of Starfish and Echinoids,
but maintains that such fixation was by the actinal surface in
these two classes, whereas in Crinoids it was by the abactinal
surface. (See, however, Chapter VIII. p. 14.)

The question as to which of the groups of organs first acquired
the radial character is of great importance in connection with the
origin of the Echinoderma. The Sarasins, who made a detailed
study of the Echinothuridae, were much impressed with the
importance of the radial muscles, and suggested that it was the
muscles that first became pentamerous. There are many striking
points of resemblance between such a form as Asthenosoma and the
Holothurians. The Sarasins therefore ridiculed the supposed Crinoid
ancestor as a " Crinoid phantom," and derived the Echinoidea from
the Holothurians. This argument is based on the idea that the
primitive characters of the Echinothuridae are due to inheritance
from the ancestral Echinoid. But it appears most probable that
the Echinothuridae arose from the Diademoid Pedina, or from
some close ally of that genus. The primitive characters of the
Echinothuridae are therefore secondarily acquired, and are not
original. The immediate ancestor of the family lived in the
Jurassic, and not in the Palaeozoic seas. To accept this con-
clusion means to abandon the derivation of the Echinoids from the
Holothurians.

Leuckart, in 1848, separated the Echinoderma into the three
groups of the Pelmatozoa, Echinozoa, and Scytodermata, and this
classification is still generally used in practice. In the two latest
arrangements of the Echinoderma, those of Bell and Haeckel, the
Echinoids are still grouped with the Stelleroids. They undoubtedly
agree in several important characters, the members of both classes
having the gonads pentamerous, the oral surface kept downward,
and power of locomotion.

The class Stelleroidea is older than that of the Echinoidea, but
we know of no member of the former that can be regarded as the
ancestor of the latter.

The evidence in favour of the origin of Echinoids from Cystids
or allied forms is more weighty. Neumayr advocated this view
(64, 65), and it has recently received fresh support from Haeckel
(36 on p. 213). Neumayr included Echinocystis in the Cystidea.
The genus is, however, here included among the Echinoids. The
uncertainty as to its position shows that there is an approximation
between the two classes. We are, therefore, forced to the position
that one group l of primitive Pelmatozoa diverged from the mam
stem and approximated to the Echinoids ; and that it was succeeded

1 Separated in Chapter XII. as a Class— Edrioasteroidea.

THE ECHINOIDEA 327

by Echinoids so similar in structure that it is hard to draw a satis-
factory line of separation.

Although the ancestral Echinoid is still unknown, the main
lines of evolution in the class are clearly recognisable. The
Echinoids began with forms having small, sac-like bodies, and a
mouth and anus at opposite poles ; the body was muscular,
supported by a series of angular plates, of which five pairs were
perforated by pores. At the summit of the" test occurred the
apertures of the alimentary, generative, and water - vascular
systems ; and the apertures of these systems were supported and
held in place by a series of special plates.

At first the palaeontological record is incomplete, the plates of
the test being thin, fragile, and loosely fitted together. Hence
there is a gap between the Echinoid just described and its next
known successors, in which the interradial plates are irregular,
and the apical system of plates is absent. But as the skeleton
thickens, fossils become more abundant and better preserved.
We can see the increase in the number of interambulacral
and of ambulacral plates up to forms such as Melonites. Then,
as the plates became stouter, the flexibility of the test was
lost ; thus the advantage of having small, numerous plates was
lost. Hence the Echinoids with more than twenty rows of plates
disappeared, and were succeeded by a group, the main feature of
which was the consolidation of the test. About the same time
there appeared an offshoot from the main stem, in which the test
was strengthened by a great development of the apical system ;
this arrangement reached its highest development in two aberrant
genera (Tiarechinus and Lysechinus) which lived in the Triassic
coral lagoons of the Tyrol. The Melonitoida and Plesiocidaroida
apparently left no issue, and all existing and post-Triassic Echinoids
appear to have descended from the primitive genera of Cidaroida.

From Cidaris, with its ambulacra of simple primary plates, the
more complex types were developed by the crowding of the pore-
pairs, and the decrease in size and increase in number of the
spines ; hence the ambulacral plates become compound, and the
interambulacral plates bore numerous tubercles and granules, and
thus gave rise to the various groups of Regular Echinoidea. In
some deep-sea forms the calcification of the external skeleton is
imperfect ; the plates are thin and the muscles strong ; by the
imbrication and isolation of the plates there is a return to some of
the features of the flexible Palaeozoic Echinoids.

The main departure from the type of regular Echinoids is
due to the backward movement of the anus interfering with
the originally quinqueradiate arrangement of the organs (Fig.
XL VI.). The mouth passes forward, the jaws disappear, the
ambulacral podia become specialised for respiration as well as

328 THE ECHINOIDEA

locomotion, the apical system of plates becomes compact, elongate,
disjunct, or rudimentary, and a bilateral symmetry replaces the
primitive, pentameral symmetry. At first the jaws are retained ;
but as the body becomes bilateral, the mouth is constricted, and
room for play of the jaws is lessened. No doubt all Echinoids get
a proportion of nourishment from the mud and sand which they
swallow ; but as their jaws become smaller, and they can browse
less effectively, the* importance of this food-supply becomes more
important. The development of a projecting under lip below the
mouth was an advantage to the Echinoid, by enabling it to swallow
more food. Hence the Irregular Echinoids began with teeth and a
central mouth — a type first met with in the lower Jurassic ; later
on, in the Jurassic, came the second type, in which there are no
jaws and the mouth is eccentric ; the former is the order Gnatho-
stomata, and the latter the order Atelostomata. The Gnathostomata
began in the Jurassic with the genus Pygaster, which differs from
the regular ectobranchiate Echinoids only by the anus opening
behind the apical system ; the Pygastridae were succeeded in the
Cretaceous by the Discoidiidae, from one genus of which, Proto-
cyamus, there is an easy passage to the Fibulariidae, and thus to
the sub-order Clypeastrina.

The order Atelostomata has apparently also been derived from
a genus allied to Pygaster. The jaws are lost, and the apical
system either remains compact or becomes elongate : the former
series possibly began with Galeropygus, whence the rest of the
asternate forms were derived. The series with elongate apical
systems began with some such genus as Hyboclypeus, which led the
way to the Collyritidae and Echinocorythidae, whence the higher
Spatangid Echinoids descended.

LITERATURE OF ECHINOIDEA.

1. Agauiz, A. 1872-74. (Revision of the Echini.) 111. Cat. Mus. Comp. Zool.-

Harvard, No. vii. 2 vols. 762 pp. 94 pis.

2. - 1881. (Report on the Echinoidea dredged by H.M.S. Challenger

during the years 1873-76.) Rep. Chall. Exped. Zool., vol. iii. 321 pp.
45 pis.

3. 1883. (Report on the Echini. Results of dredging by the Blake,

No. xxiv. pt. 1.) Mem. Mus. Comp. Zool. Harvard, vol. x. No. i. 94
pp. 32 pis.

4. Agassiz, L. 1836. (Prodrome d'une Monographic des Radiaires ou Echino-

dermes.) Mem. Soc. Sci. Nat. Neuchatel, vol. i. pp. 168-199.

5. Agassiz, L.t Dcsor, £., and Valentin, G. 1838-42. Monographies d'Echino-

< ir nut's vivans et fossiles, Nos. i.-iv. ; and Anatomie, No. i. 4to, Neuchatel.

6. Agassiz, L.y and Dcsor, E. 1846-48. (Catalogue raisonne des families, des

genres et des espcces de la classe des E*chinodermes.) Ann. Sci. Nat.
Zool., ser. 3, vols. vi., vii., and viii. 167 pp. 2 pis.

THE ECHINOIDEA 329

7. Bell, F. J. 1881. (The Echinometridae.) Proc. Zool. Soc. for 1881, pp.

410-433.

8. 1889. (Report of a Deep-sea Trawling Cruise off the S.W. Coast of

Ireland. Echinodermata. ) Ann. Mag. Nat. Hist., ser. 6, vol. iv. pp.
432-445, pis. xviii., xix.

9. 1891. (On the Arrangement and Inter-relations of the Classes of the

Echinodermata.) Ann. Mag. Nat. Hist., ser. 6, vol. viii. pp. 206-215.

1 0. 1892. Catalogue of die British Echinoderms in the British Museum,

xvii. and 202 pp. 16 pis.

11. Biitschli, 0. 1892. (Versuch des Ableitung des Echinoderms aus einer

bilateralen Urform.) Zeitschr. Wiss. Zool., vol. liii. Suppl. pp. 136-160,
pi. ix.

12. Clark, W. B. 1893. (The Mesozoic Echinodermata of the United States.)

Bull. U.S. Geol. Surv. No. 97, 207 pp. 50 pis.

13. Cotteau, G. H. 1862-67. (Echinides reguliers.) Paleontologie fran$aise,

Terrains cretacees, vol. vii. 894 pp. 198 pis.

14. 1867-85. (Echinides irreguliers et reguliers.) Paleont. fransaise,

Terrains juras.siques, vol. ix. 1867-74, 552 pp. 142 pis. ; vol. x. pt. i.
1875-80, 468 pp. 120 pis. ; vol. x. pt. ii. 1880-85, 960 pp. 258 pis.

15. 1885-94. (Echinides.) Paleont. fran^aise, Terrains tcrtiaires

(Eocene), vol. i. 1885-89, 672 pp. 200 pis. ; vol. ii. 1889-94, 788 pp. 176 pis.

16. Cotteau, G. H., Peron, P. A., and Gauthier, V. 1876-91. Echinides fos-

siles de 1'Algerie, 3 vols. 8vo, Paris.

17. Cotteau, Q. H., and Triger. 1857. Echinides du departement de la

Sarthe. 8vo, Paris, 458 pp. 75 pis.

18. Dcsmoulins, C. 1835-37. Memoires stir les Echinides. 8vo, Bordeaux,

520 pp. 2 pis.

19. Desor, K 1855, 1857-58. Synopsis des Echinides fossiles. 4to, Geneve,

Paris, and Wiesbaden, Ixiv. and 490 pp. 43 pis.

20. Dodcrlein, L. 1887. Die Japanesischen Seeigel, Theil 1. 4to, Stuttgart,

59 pp. 11 pis.

21 . Dujardin F. , and Hupe, H. 1862. Histoire naturelle des Zoophytes £ehino-

dermes. 8vo, Paris, 628 pp. 10 pis.

22. Duncan, P. M. 1885. (Structure of the Ambulacra of some Fossil Genera

and Species of Regular Echinoidea.) Quart. Journ. Geol. Soc., vol. xli.
1885, pp. 419-453.

23. 1888. (On some Points in the Anatomy of the Temnopleuridae. )

Ann. Mag. Nat. Hist., ser. 6, vol. i. pp. 109-131, pi. xi.

24. 1890. (A Revision of the Genera and Great Groups of the Echinoidea. )

Journ. Linn. Soc. Zool., vol. xxiii. pp. 1-311.

25. Duncan, P. J/., and Sladen, W. P. 1882-86. (The Fossil Echinoidea of

Sind.) Pal. Ind., ser. xiv. vol. i. pt. 3, 382 pp. 58 pis.

26. - 1883. (The Fossil Echinoidea of Kachh and Kattywar.) Pal.

Ind., ser. xiv. vol. i. pt. 4, 104 pp. 13 pis.

27. Gauthier, V. 1889. (Description des Echinides fossiles recueillis . . . dans

la region sud des hauts-plateaux de la Tunisie, par M. P. Thomas.)
Exploration Scient. Tunisie. Paris, 116 pp. in 8vo, 6 pis. in 4to.

28. 1892. Notes sur les Echinides cretaces recueillis en Tunisie, par

M. Aubert. 8vo, Paris, annexe de la carte geol. de la Tunisie, 52 pp. 4 pis.

29. Gray, J. E. 1825. (An Attempt to Divide the Echinida or Sea-Eggs into

Natural Families.) Ann. Phil., vol. x. pp. 423-431.

330 THE ECHINOIDEA

30. Gray, J. E. 1855. Catalogue of the Recent Echinida or Sea-Eggs in the

Collection of the British Museum. 8vo, London, 69 pp. 6 pis.

31. Gregory, J. W. 1891. (The Maltese Fossil Echinoidea.) Trans. Roy. Soc.

Edinb., vol. xxxvi. pt. 3, pp. 585-639, 2 pis.

32. 1891. (The Relations of the American and European Echiuoid

Faunas.) Bull. Geol. Soc. Amer., vol. iii. pp. 101-108.

33. 1891. (A Revision of the British Fossil Cainozoic Echinoidea.)

Proc. Geol. Assoc., vol. xii. pp. 16-60, pis. L, ii.

34. 1897. (On Lysechinus, a new Genus of Fossil Echinoderms from the

Tyrolese Trias.) Proc. Zool. Soc. for 1896, pp. 1000-1005, pi. li.

35. 1897. (On the Affinities of the Echinothuridae.) Quart. Journ.

Geol. Soc., vol. liii. pp. 112-122, pi. vii.

36. 1897. (On Echinocystis and Palaeodiscus — two Silurian genera of

Echinoidea.) Tom. cit. pp. 123-136, pi. vii.

37. Groom, T. T. 1887. (On some new Features in Pelanechiims corallinits.)

Quart. Journ. Geol. Soc., vol. xliii. pp. 703-714, pi. xxviii.

38. ffamann, 0. 1887. (Beitrage zur Histologie der Echinodermen, Heft 3.

Anatomie und Histologie der Echiniden und Spatangiden. ) Jena.
Zeitschr., vol. xxi. pp. 87-266, pis. vi.-xviii.

39. Hartog, M. M. 1887. (The True Nature of the " Madreporic System " of

Echinodermata.) Ann. Mag. Nat. Hist. (5), vol. x. pp. 321-326.

40. Hoffmann, C. K. 1871. (Zur Anatomie der Echinen und Spatangen.)

Niederl. Arch. Zool., vol. i. pp. 11-112, pis. iii.-x.

41. Jackson, 11. T. 1896. (Studies of Palseechinoidea.) Bull. Geol. Soc. Amer.,

vol. vii. pp. 171-254, pis. ii.-ix.

42. Jacket, 0. 1894. (Ueber Bothriocidaris. ) Sitz. Ges. naturf. Berlin fur

1894, pp. 243-255.

43. Kohler, R. 1883. (Recherche* sur les Echinides des cotes de Provence.)

Ann. Mus. Hist. Nat. Marseilles, vol. i. 167 pp. 6 pis.

44. Lambert, J. 1889. (Note sur le developpement de F Echinospatangus

ncocomicnsis.) Bull. Soc. Sci. Yonne, vol. xliii. 19 pp.

45. 1892. (Recherches sur les Echinides de 1'Aptien de Grandpre.)

Bull. Soc. Geol. France, ser. 3, vol. xx. pp. 38-100, pis. ii.-iv.

46. 1893. (Etudes morphologiques sur le Plastron des Spatangides.)

Bull. Soc. Sci. Yonne, vol. xlvi. section ii. pp. 55-98 and folding
table.

47. Lany, A. 1894. Lehrbuch der vergleichenden Anatomie, Vierte Abthei-

lung. 8vo, Jena. Translated by H. M. and M. Bernard, as Text-book
of Comparative Anatomy, pt. ii. 1896, London.

48. Lcskc, N. G. 1778. Additamenta ad Jacobi Theodori Kleinii Naturalem

Dispositionem Echinodermatum. 4to, 286 pp. 54 pis. Lipsiae.

49. Loriol (afterwards Loriol-Lcfort), P. de. 1880. (Monographie des Echinides

contenus dans les couches nummulitiques de 1'Egypte.) Mem. Soc. Phys.

Geneve, vol. xxvii. pp. 59-148, 11 pis.
00. 1883. (Eocaene Echinoideen aus Aegypten und der libyschen Wiiste.)

Palaeontographica, vol. xxx.
61. 1884-97. (Notes pour servir ti 1'etude des Echinodermes, Nos. i.-vi.)

Recueil Zool. Suisse, Mem. Soc. Phys. Geneve, and Revue Suisse de

Zoologie.
52. 1887-88. (Faune Cretacique du Portugal. Vol. ii. Echinodermes),

123 pp. 22 pis. Commiss. trav. geol. Portugal.

THE ECHINOIDEA 331

53. Loriol-Lcfort, P. dc. 1890-91. (Faune jurassique du Portugal, Em-

branchement des Echinodermes), 179 pp. 29 pis. Commiss. trav. ge"ol.
Portugal.

54. 1896. (Description des Echinodermes tertiaires du Portugal), 50 pp.

13 pis. Mem. trav. geol. Portugal.

55. Loriol, P. de, and Desor, E. 1868-85. Echinologie Helve"tique, vol. i.

Wiesbaden, 1868-72, 441 pp. 61 pis. ; vol. ii. 1873, in Pictet, Materiaux
Paleont. Suisse, ser. 6, No. 2, 398 pp. 33 pis. ; vol. iii. 1875, in Me"m. Soc.
pal. Suisse, vol. ii. 142 pp. 23 pis. ; lre Suppl. 1885, 24 pp. 3 pis. op. cit.
vol. xii.

56. Lovtn, S. 1867. (Om LesMa mirabilis, Gray.) Ofv. Svensk. Vet.-Akad.

Forhdl., vol. xxiv. pp. 431-440.

57. 1874. (Etudes sur les Echinoidees.) Svensk. Vet.-Akad. Handl.,

vol. xi. No. 7, 91 pp. 53 pis.

58. 1883. (On Pourtalesia, a genus of Echinoidea.) Op. cit. vol. xix.

No. 7, 95 pp., 21 pis.
59. 1887. (On the Species of Echinoidea described by Linnaeus.)

Bihang Svensk. Vet.-Akad. Handl., vol. xiii. Afd. iv. No. 5, 185 pp. 9 pis.

60. - 1892. (Echinologica.) Op. cit. vol. xviii. Afd. iv. No. 1, 74 pp.

12 pis.

61. Ludwig, H. 1877. (Ueber bewegliche Schalenplatten bei Echinoideen. )

Zeitschr. Wiss. Zool., vol. xxix. pp. 77-86, pi. vii.

62. 1890. (Ueber die Function der Madreporenplatte und des Stein-
canals der Echinodermen.) Zool. Anz., vol. xiii. pp. 377-379.

63. Mackintosh, H. W. 1879-83. (The Acanthology of the Desmosticha. )

Pt. I. Trans. R. Irish Acad., vol. xxvi. pp. 475-490, pis. ix.-xi. Pt. II.
op. cit. vol. xxviii. pp. 241-258, pis. v.-viii. Pt. III. op. cit. vol. xxviii.
pp. 259-266, pis. ix.-x.

64. Neumayr, M. 1881. (Morphologische Studien iiber fossile Echinodermen.)

Sitzber. Akad. Wiss. Wien, vol. Ixxxiv. pt. i. pp. 143-176, pis. i., ii.

65. 1889. Die Stamme des Thierreiches, vol. i. 8vo, vi. and 603 pp.

Wien.

66. Orbigny, A. D. a". 1853-55. (Echinodermes.) Paleont. fra^aise, Ter-

rains Cretacees, vol. vi. 597 pp. 206 pis.

67. Perrier, J. 0. E. 1875. (Recherches sur 1'appareil circulatoire des Our-

sins.) Arch. Zool. Exper., vol. iv. pp. 605-643, pis. xxiii., xxiv.

68. 1893. Traite de Zoologie, fasc. ii. (Echinoderms on pp. 781-864.)

8vo, Paris.

69. Pomel, A. 1883. Classification methodique et genera des Echinides

vivants et fossiles, 132 pp. 1 pi. These doctorale, Paris. Also in
Paleont. de 1'Algerie, Zoophytes, fasc. 2.

70. Prouho, H. 1888. (Recherches sur le Dorocidaris papillata. ) Arch. Zool.

Exper., ser. 2, vol. v. pp. 213-380, pis. xiv.-xxvi.

71. 1890. (Du role des Pedicellaires Gemmiformes chez les Oursins.)

Compt. Rend. Acad. Sci. Paris, vol. cxi. p. 62.

72. Sarasin, P. B. and C. F. 1887. (Die Augen und das Integument der

Diadematiden. ) Ergeb. Naturw. Forsch. Ceylon, vol. i. Heft 1, pp.
1-18, pis, i.-iii.

73. 1888. (Ueber die Anatomie der Echinothuriden und die Phylogenie

der Echinodermen.) Tom. cit. Heft 3, pp. 83-154, pis. x.-xvii.

74. Schliiter, 01. 1883-92. (Die regularen Echiniden der norddeutschen

332 THE ECHINOIDEA

Kreide.) Abh. preuss. geol. Landesanst., vol. iv. Heft 1 ; and new ser.
vol. v. ; in all 315 pp. 21 pis.

75. SchliUer, Cl. 1897. (Ueber einige von Goldfuss beschriebene Spatangideu.)

Zeitschr. deutsch. geol. Ges., vol. xlviii. pp. 963-975.

76. 1897. (Ueber einige exocyclische Echiniden der baltischen Kreide

und deren Bett.) Zeitschr. deutsch. geol. Ges., vol. xlix. pp. 18-50,
pis. i., ii.

77. Stewart, C. (On Certain Organs of the Cidaridae.) Trans. Linn. Soc.

Lond., ser. 2, Zool., vol. i. 1879, pp. 569-572.

78. Thccl, H. 1892. (On the Development of Echinocyamus pusillus. ) Nova

Acta R. Soc. Sci. Upsala, ser. 3, vol. xv. fasc. 1, No. 6.

79. Thomson, C. IVyville. 1861. (On a New Palaeozoic Group of Echino-

dermata.) Edin. New Phil. Journ., n.s., vol. xiii. pp. 106-117, pis.
iii., iv.

80. 1874. (On the Echinoidea of the Porcupine Deep -Sea Dredging

Expeditions.) Phil. Trans., vol. clxiv. pp. 719-756.

81. — — 1876. (Notice of some Peculiarities in the Mode of Propagation of

certain Echinoderms of the Southern Sea.) Journ. Linn. Soc. Zool., vol.
xiii. pp. 55-79.

82. Ucxkiill, J. von. 1896. (Ueber die Function der Polischen Blasen am

Kauapparat der reguliiren Seeigel.) Mitth. Zool. Stat. Naples, vol. xii.
pp. 463-476, pi. xxi.

83. Wright, T. 1857-78. Monograph of the British Fossil Echinodermata of

the Oolitic Formations, vol. i. Echinoidea, xiv. and 481 pp. 43 pis.
London : Palaeontogr. Soc.

84. — — 1864-82. Monograph of the British Fossil Echinodermata of the

Cretaceous Formations, vol. i. Echinoidea, xxviii. and 371 pp. 80 pis.
London : Palaeontogr. Soc.

85. Zitttl, K. A. Handbuch der Palaontologie, vol. i. pt. iii. 1879.

INDEX

To names of Classes, Orders, Sub-Orders, and Genera ; to technical terms ; and to
names of Authors discussed in the text.

Abacocrinus, 165
abactinal skeletal elements,

109

Abutus, 322
aboral circular sinus (Oph.)*

266

Abracrinitu, 166
AbrotocrittHfi, 180
Acacocrinus, 166
Acanthar chaster, 251
Acanthaster, 258
Acanthechinas, 309
Acanthocrinus, 201
. I canthocystis, 4 6
aeanthosphenote, 287
Acanthotrochus, 234
accessory intestine, 27

— plates ( Ast. ), 246
Acentrotremites, 90, 92
Acute, 322
Achlyouice, 229
Ackradocrimis, 178
Achradocystis, 55

A crochordocrinits, 192

Acrocularis, 307

^4 crocrinus, 159

Acrosalenia, 306

.4c«wra, 278

actinal furrows (Ech.), 317

— skeletal elements, 109
Actinocrinus, 169
Actinocucumis, 230
actinogonidial, 237
actino-h'teral spines, 256
Actinomctra, 196
Actinophyma, 310
Actinopoda, 227
actinostomial ring (Ast.),

241

Actinozoa (Leuckart), 33
adambulacral, 20

— mouth, 247
adambulacrals (Crin.), 115 ;

(Ast.), 241 ; (Oph.), 270
Adelocrinus, 201

j AdelopneuxteS) 316
! Adethme, 321
adetopueustic, 254
adradii, 20
Aduuata, 155
Aeolopaeiistes, 313
Aerope, 322
Aesiucrinns, 180
Aes-iocystis, 208
Aethocystiti, 57
Aetites, 53
Aganaster, 275
Agaricocritius, 167
Agassiz, A., 240, 285
Agassiz, L., 240, 260, 261,

284

Agassizia, 322
Agassizocrinus, 181
Agelacrinoidea, 205
Agelacrinus, 207
Agelacystis, 207
Agelcutiscus, 208
Agriocrinus, 156
Aldrovandus, U., 283
^1/ecto, 195
Alexandria, 318
Allagecrimis, 151
Allionia, 201
AUocrinus, 162
Allocystis, 72
Alloprosallocrimis, 167
Amblacrimts, 156
Amblypneustes, 313
Ainblypygus, 320
ambulacra, 2, 15 ; (Blast.),

78 ; (Ech.), 286
ambulacralia, ambulacrals

(Crin.), 100,115; (Ast.),

241; (Ech.), 288, 296
ambulacral mouth, 247
amnion (Ech.), 15
amoebocytes, 22
Amorphocystis, 55
Amplieristocrinus, 173
Amphiaster, 254

yl mphiglypha, 278
Amphilepis, 278
J mphipholis, 27 8
amphisternous, 318
Amphiura, 278
Amphora, 169
.4 wphoracrinns, 170
AMjrfioracystis, 47
Amphoridea, 43
anjpulla, 290
Awyydalocystiti, 58
anal plates (Crin.), 119;
(Ech.), 294

— spiracle, 85

— tube, 131

— *•, 119
Atianchytes, 321
Atutpta, 234
Anasterias, 258
Anauloddaris, 314
anchylosis, 108

^1 ncyrocrintis, 177
anectobrauchiate, 304
Anisocrinns, 189
Ankyrodervia, 233
Anoc/utnus, 320
J nomalanthits, 318
Anomalocrinus, 146
AnovialocystiSi 51
Anorthopygus, 320
Antedon, 195
Anthenea, 253
Anthenoides, 253
Antliocystis, 62
Anthemocrinus, 200
AnthocrinuSi 176
antimeres, 19
vlorocrmw^, 167
Aphroditaster, 251
apical nervous system, 30

— system (Crin.), 112 ;
(Oph.), 268 ; (Ech.), 286,
294. See also calycinal

Apiocrinits, 191
Apiocystis, 61

334

INDEX

Aplocrimts, 151

Asterodon, 253

Bathybiaster, 252

Apneuinona, 226

Asteroidea, 239 ; diag-

Bathycrinus, 186

Apodes, 226

nosed, 248

Bathyplotes, 227

Aporita, 69

Asteropsis (Cott.), 309

Batocrinus, 167

Aporocrinus, 201

— (M. & T.), 254

Baiieria, 306

Apostolides, N. C., 261

Asterostoina, 321

Baur, A., 218

Arachniopleunis, 312

Ast/ienosoma, 311

Baur's vesicles, 32

Arachnocrimis, 175

Astrella, 252

Bdellacoma, 254

Arachnocystis, 54

Astridypeus, 317

Belemnocrinus, 152

Arachnoids, 318

^rios, 202

Belemnocystis, 51

Arbacia, 308

Astroceras, 276

Bell, F. J., 261, 285, 326

Arbacina, 307

Astrochele, 276

Belon, P., 218

arcade -like spaces (Ast.),

Astrocniila, 276

Bent/taster, 257

248

Astrocoma, 202

Benthodytes, 229

Arcliaeocidaris, 302

Astrncrinns (Austin), 91

Benthopecten, 251

Arc/taeocriniis, 200

— (Conr.), 161

Beyrichocrinns, 168

ArcJuieodiadeitia, 308

— (Cumb.), 202

bifascial articulation, 117

Archaeopnetistcs, 321

Astrocystites, 209

Bipinnaria, 5, 248

Archnstcr, 251

Astrodon, 276

biscuit spicules (Hoi.), 224

Arc/ui#te,-u(s. 250

Astrofjomphus, 276

biserial arms, 115

Archegwyiitui, 73

Antnyoninm, 253

bivium, 11 ; (Hoi.), 220 ;

Archiacia, 321

A sir onyx, 276

(Ech.), 296

Argaxtei; 250

Astropectcn, 252

Blainville, H. D. de, 240,

Argopatagus, 323

Astrophyton, 276

260

Aristocrinus, 190

Astropodia, 202

Blairocrinus, 169

Aristoci/stt'.i, 44

Astroporpa, 27 Q

Blakiaster, 251

Aristotle, 218, 283

Astropyga, 310

Blastoidea diagnosed, 78

arm (Criu.), 112; (Ast.),

Astroschema, 276

Blastoidocrinus, 80

241 ; (Oph.), 262, 267,

Astrotonta, 276

Bohadsch, J. B., 218

270. See also brachia

Astylocrinus, 181

Bohemicocrinus, 201

— facet, 100

Ataxocrinus, 146

JBo^/rt, 313

— pores, 130

Atdecrinus, 195

boss (Ech.), 287

Arthracantha, 158

Atelecystis, 51

BothriocidariSi 300

Arthraster, 255

Atelestocrinns, 179

Bothriocidaroida, 300

article basal, 108

Atelostomata, 318

Botriopygus, 320

articles brachiaux, 204

Atwrinus, 202

Bntryocrin MS, 179

Articulata (Bather), 178 ;

Aulocrinits, 180

7Jo */ r^r?f ei icrinus, 194

(Jaekel), HI ; (Miller),

Anricularia, 5, 225

bourrelet, 319

138 ; (Miiller), 138 ; (W.

Anstniocritius, 183

brace, 289

& Sp.), 139

Anstndanthus, 321

brachia, 3, 100, 112. See

articulation (Grin.), 108

axial canal, 101

also arm

Articulosa (Jaekel), 141 ;

— cord, 12, 103

brachialia, 100, 204

(W. & Sp.) 140

— organ, 23 ; (Ech.), 291

Brack iocrin us, 1 4 6

Ascidiastella, 218

— sinus, 22 ; (Grin.), 103 ;

brachiola, brachioles, 3- 41

Ascocystis, 77

(Ast.) 243 ; (Oph.), 266

Brachiolaria, 6, 248

Aspidocrinus, 201

axillare, axillary, 114

branchiae (Ech.), 293

Aspulodiadcnia, 308

Ayers, H., 32

branchial vesicles, 245

Aspidosoma, 250

azygos plates, 204

Brandt, J. F., 226, 240

Aspidura, 278

Breynia, 323

Astoria, 201

Jiactrocrinus, 173

Breynius, J. P., 283

Asterias, 201, 258

Baculocystis, 46

Briarocrinus, 162

— rubetis, 241

Baerocrinus, 145

Brisinga, 259

Asteriatites, 201

BaUtnocrinus (Ag.), 182

Br is ing aster, 259

Asterid plane, 21

— (Troost), 202

Brissopatagus, 323

Asterina, 254

nalanocystis, 77

Brissopsis, 322

Asternata, 320

Barrande, J., 44, 48

Brissiis, 323

Asteroblastus, 80

Barrandeocrinuj, 166

Bronn, H. G., 40

Asterocrinut (Lyon), 159

Barycrinus, 179

buccal fissures, 263

— (Munst.), 201

basalia, basals (Criu.), 99,

— papillae, 264

Asterocystis, 80

122 ; (Ast.), 246 ;

— shield, 264

AslerodiscMS, 254

(Ech.), 294 ; (Oph.), 268

Buch, L. v., 40

INDEX

335

Bundeiibachia, 274

Catillocrinus, 150

Cionocrinus (Quenst.), 162

Bursacrinus, 180

Gatopygus, 321

close suture, 108

bursae, 26

Caudina, 233

Closterocrinus, 173

.bursal slit, 264

Caidaster, 251

Clypeanthus, 321

Bury, H., 218, 240, 261,

Ge)wcrinus, 182

Clypeaster, 318

285

central blood-plexus, 23

Clypeastrina, 316

central plate (Oph.), 268

Glypeopygus, 321

Cacabocrinus, 164

centrale, 135

Clypeits, 321

Cactocrinus, 170

Gentriocrinus, 162

Cnemidiaster, 255

caeca (Ast.), 243

Gentrocrinns (Austin), 157

Coadunata, 138

Cainocrinus, 182

— (W. & Sp.), 162

(Joccocri mis, 156

Gcdamocrinus, 192

-- (Worthen), 202

Cadaster, 82, 92

Calathocrinus (Hall), 170

centro-dorsal, 135

Codechiwtu, 310

— (Meyer), 182

Ccntrosteplianus, 309

Codidcrinns, 177

Galceocrinus, 148

Ceratiwcrinns, 177

Godiacystis, 73

Caleidocrinus, 201

CeriocriRus (Desor), 191

Godiopsis, 309

Calix, 46

— (White), 180

Codonites, 84

Callawayerinus, 190

Cliaetaster, 255

Codonoblastida, 91

Calliastcr (Gray), 253

Cheiraster, 251

C<xloiiocritius, 202

— (Trautsuh), 259

Clieii'ocritius (Eiclnv. ), 63

Coelastcr (Saiulb. ), 259

Catticrinut, 164

- (Hall), 148

Coelaster-iati, 255

Galliderma, 253

— (Salter), 148

coeliac canal, 101

Callocystis, 62

Cheiroptastcr, 257

CocliocriHu-s, 180

Calocriiius, 202

Chelocriww, 182

Guelocrinus (M. & W.),

Calpiocrimis, 189

Chiaje, S. Delle, 240, 284,

167

Calvasterias, 258

285

- (Salter), 200

Calycanthocriniis, 150

chiasma (Crin.), 114

coeloni, 7

Galycaster, 255

Chiridota, 234

Goelopleurua, 308

Calycina, 306

ChUonastei-, 253

collateral organ, 23

calycinal system, 14 ;

Chladocriuus, 182

Golly rites, 321

(Oph.), 268. See also

Cholaster, 275

Coloboceiitrotux, 313

apical

dwndrocidaris, 302

Golochirus, 230

Calymne, 321

Clwriaster, 254

Golpatiter, 259

Calyptaster, 257

chromatogen organ, 23

coluinua, 48, 99. See also

calyx, 99

Cidaris, 302

stem

Camarocrinus, 77, 161

Cidaroida, 301

columnals, 105, 132

Camerata (W. & Sp.), 139

Cidaropsis, 307

Comarucytitis, 55

— Dicyclica, 198

Ciffara, 48

Conwxtei; 196

— Monocyclica, 159

ciliated urns, 235

CoiHMtula, 195, 196

Campanulites, 202

Cionobrissus, 323

Gomatulina, 202

Camptocrinus, 159

Uircopeltis, 307

Comaturella, 202

canal furrow (Oph.), 262

circuinoesophageal canal

compass (Ech.), 289

Ganistrocrinus, 161

(Ast.), 243; (Oph.),

Gnwpucuiter, 255

Caphcira, 229

265. See also Hydro-

Computer in us, 1 65

(Jarabocrinus, 172

circus

Co)ii2)to)iia, 253

Garatomus, 320

— nerve-ring (Crin.), 102 ;

Coiidylocriiius, 201

Card-faster, 321

(Ast), 245

connective tissue, 28

Cardiocystis, 77

cirri, 107, 133

Conodypeiis, 316

GardiMcrinus , 179

Cirrigrada, 260

Gonocrinus (d'Orb), 194

Carolicrinus, 165

Citrocystis, 53

— (Troost), 202

Carpenter, P. H., 14, 44,

Cladocrinus (Aust.), 189

Conolampas, 321

91, 139, 140, 182, 325

Cladoidea (Jaekel), 141

convoluted organ, 136

Carpocrinus, 166

Cladophiurae, 276

Cophinus, 202

Carpocystis, 73

Claviaster, 321

Coptodiscus, 316

Garyocrinus, 67

clavulae, 31

Goptophyma, 312

Caryocysiis, 55

Oleiocrinus, 191

Goptosoma, 310

Caryophyllites, 197

Gleistechinus, 323

Coraster, 322

Cassianocrinus, 182

Clematocrimts, 156

Cordylocrinus, 156

Cassidulus, 321

Clidochirus, 188

corona (Crin.), 99

Castanocrinus, 161

cloaca (Hoi.), 221

Coronaster, 258

Castocrinus, 148

Glonocrinus (Oehl.), 161

Coronocrimts, 164

336

INDEX

Corylocrinus, 67

Cyclaster, 323

Diademoida, 306

Corymbocrinus, 162

Oydocrinus (d'Orb.), 192

Uiademopsis, 309

Coscinasterias, 258

— (Eichw.). 77

Diamenocrinus, 200

Cosmasterias, 258

Cydocystoides, 210

dibrachialia, 204

Cosmocrinus, 179

Cylicocrinus (S. A. Mill.),

Dichocrinus, 159

costalia, 141, 204

166

dichotomy of arms, 112

Costata (Miiller), 138 ;

Oylicocrinus (Miiller), 156

dicostalia, 204

(Jaekel), 141

Cypellocrinns (Shum.),

Dictyaster, 258

Cottaldia, 309

156

Dictyocrinus, 77

Cotylecrinus, 198

— (Stein.), 177

Itictyoplenrus, 312

Cotyledenna, 198

Cypliocrinus, 199

dicyclic, 99, 110

Cotyledonocrinus, 159

Cyphosoma, 310

Dicyclica, diagnosed, 171

covering- plates, 101. See

Cypressocrinus, 202

— Camerata, 198

also ambulacrals

Cyrtidocrinus, 188

— Inadunata, 171

Craspidaster, 252

Cyrtocrinus, 197

Dimerocrinus (Pacht), 190

Craterina (Barr.), 46

Cystaster, 207

— (Phill.), 198

Craterolampas, 321

Cystasteroidea, 205

JJimorphicrinus, 84, 202

Cremacrinus, 148

Cystechinus, 321

Ditwcyttis, 209

Crenaster (Perr.), 251

Cystidea, diagnosed, 39

Diplasterias, 258

Cribrella, 258

Cystidea (Barr. ), 77

IHpleunda, 4

Cribrellites (Tate), 259

Cystoblastus, 64

Diplocidaris, 303

cribriform organs, 251

Cystocidaroida, 300

Diplopodia, 309

Orinocy stigt 77

Cystocrinus, 202

diplopores, 41

Crinoidea, diagnosed, 94 ;

Cytocrinus, 161

Diploporita, 70

(Jaekel), 141 ; (Muller),

Diplotagnna, 309

138 ; (Roemer), 33

Dactylocrinus (Quenst.),

Diplotfacanthns, 318

— braclriata, 94, 138

189

Dipneustea, 322

Cromyocrinus, 181

— (Slad.), 180

Dipsacoster, 252

Crossaster, 256

Dactylocystis, 74

Disasterina, 254

Crotcdocriuus, 176

Dadocrinus, 182

disc. (Ast.), 2 ; (Crin.),

crown, 99

Daemonocrinus, 202

99 ; (Oph.), 262

Crumenaecrinus, 202

Decacnemos, 202

IHscocidaris, 302

Cryptaster, 257

Decadactylocrinus, 202

Discocystis, 208

Cryptoblastus, 89, 92

Decadocrinus, 180

Discoidea, 316

Cryptocritius, 69

Decameros, 202

distal (Grin.), 109

Cryptodis&is, 164

^ewAia, 229

distichalia, 204

Cryptoschisma, 84, 92

deeper oral nervous system,

Distincta, 178

Cryptozonia, 254

30

Dizygocrinus, 168

Crystallocystis, 53

Delocrinus, 180

Dolatocrinus, 164

Ctenaster, 256

Deliacrinus, 148

Dolichocrinus, 197

Ctenocrinus, 161

deltoidea, 100. See also

Doliolocrinus, 202

Ctenodiscus, 251

or ali a

Donacicrimis, 202

Cuciimaria, 230

demi-plates, 288

Dorigona, 253

Cuenot, L., 33, 218, 240,

Democrinus, 194

D&rocidaris, 302

261, 294

Dendrocrinacea, 171

dorsal canal, 101

Culcita, 254

Dendrocrin&idea, 178

— cup, 99

Culicocrinus, 156

Dendrocrinus, 179

— fossa, 116

Cupellaecrinus, 156

— cambriensis, 146

— organ, 23

Cuprcssocrinus, 177

Dendrocystis, 47

— shield, 262

Cupulaster, 259

dental papillae (Oph.), 272

dorso-central, 14, 106

Cupulocrinus, 202

Derbos, A., 285

Dorycrinus, 167

Cuvier, G., 1, 218, 226

dermal branchiae, 245

Dujardin, F., 40, 260, 284

Cuvierian organs, 223

Dennasterias, 254

Duncan, P. M., 284, 299,

Cyathidium, 198

desmactinic, 237

306

Cyathocrinacea, 171

Dtsmidocrinus, 166

Duvernoy, G. L., 240

Cyathocrinidae (W. & Sp.),

Desor, K, 284

Dysaster, 321

171

Desordla, 320

Dytaster, 251

Cyathocrinoidea, 172

Deutocystis, 46

Cyathocrinus, 173

Diabolocrinus, 200

EchinanthTis, 318

Cyathocysti*, 208

Diadeina, 309

Echinarachnius, 317

Cycethra, 254

Diademina, 308

Echinatter, 258

INDEX

337

Echinasterella, 257

Enichaster, 321

Faujasia, 321

Echinid plane, 21

Enneacystis, 67

Ferdina, 255

Echinina, 311

Enoploura, 51

Fibularia, 317

Echinites (Duncan), 316

JEntomaster, 321

fimbriated channels (Oph.),

Echinobrissus, 320

Entrochus, 202

251

Echinocardium, 323

Enypniastes, 229

finials, 115

Echinochrome, 23

Eocystis, 48

Fistulata (Jaekel), 141 ;

Echinocidaris (Dune.), 308

Eodiadenia, 308

(W. & Sp.) 139

Echinocorys, 321

Eolampas, 321

Fistulides, 218

Echinocrepis, 324

Eoluidia, 275

fixed brachials, 112

Echinocrinus (L. Ag.), 202

Eospondylus, 275

Flabellocrinus, 182

Echinocucumis, 230

Epactocrimis, 177

Flexaster, 257

Echinocyamus, 317

Epiaster, 321

Flexibilia, 140, 187

Echinocyphus, 312

epineural canal, 15, 30 ;

floscelle, 319

Echinocystis (Hall), 70

(Oph.), 266

Forbes, E., 260

— (W. Thorns.), 301

epipodium, 288

Forbesiocrinus (Ang.), 189

Echiuoderma, diagnosed, 34

epithecal food-grooves, 41

— (deKon.), 189

Echinodiadema, 309

epizygal, 108, 117

— ^4^«5sm, 190

Echinodiscus (Ag.), 317

Eretmocrinus, 167

forcipifonn pedicellariae,

— (Worth. & Mill.), 208

Erinocystis, 60

247

Echinoencrinus, 60

Erisocrinus, 181

forficiform pedicellariae,

Echinoidea, 282

Etheridge, R. til., 91, 139

247

— diagnosed, 299

ethmolysian, 295

food -groove, 100

Echinolampas, 321

ethmophract, 295

Freyella, 259

Echinometra, 313

Euasteroidea, 249

Fromia, 255

Echinoneus, 320

Eublastoidea, 81

Fungocystis, 74

Echinopedina, 310

Eucalyptocrinus, 164

Furcaster, 275

Echinopsis, 308

Euchirorrinus, 148

Echinosphaera, 53

Eucladia, 275

Gagaria, 310

Echinostrephus, 313

Eucladocrimis, 158

Oaleaster, 321

Eckinothrix, 310

Eucrinoidea, 139

Galerites, 316

Echinothuria, 311

Eucrinits, 199

Galerodypeus, 320

Echinozoa, 33

Eucystis, 72

Galeropygus, 316

Echinus, 313

Eudesicrinus, 198

Gammarocrinus, 197

— esculentus, 285

Eudiocrinus, 195

Ganeria, 254

Ectenocrinus, 146

Euechiuoidea, 299

Ganymeda, 203

Ectobranchiata, 306

Engaster, 274

Gasterocoma, 177

ectoderm, 30

Eugeniacrinus, 197

Gastraster, 258

Edrioaster, 209

Eupachycrinus, 181

Gastrocrinus, 179

Edrioasteroidea, diagnosed,

Eupatagus, 323

Gaudry, A., 240

205

Euphronides, 227

Gaurocrinus, 203

Edriocrinus, 191

Eupyrgus, 233

Gauthieria, 310

Edriocystis, 209

Eurhodia, 321

Gazacrinus, 188

Edwardsocrinus, 157

Euryale (Konig.), 154

Genicopatagits, 321

Elaeacrinus, 88

— (Lam.), 276

genital nerve ring, 266

Elasipoda, 226

Euryalecrinus, 189

— organs, 24

Eleutherocrinus, 87, 92

Euryocrinus, 190

— plate, 264

Eleutherozoa, 33

Eurypneustes, 313

— rachis, 101

— diagnosed, 217

Euspirocrinus, 173

— ring (Ech.), 291

Elpidia, 229

Eiitrochocrinus, 167

— scale, 264

Emperocrinus, 202

Evechinus, 313

— stolon, 23

Enallaster, 321

exocyclic (Ech.), 294

genitals (Ech.), 286, 294

Enallocrinus, 176

— gut (Grin.), 136

Gennaeocrinus, 169

Enallopneustes, 321

exothecal processes, 41

Geocoma (Fraas), 195

Encope, 317

external gills, 26

— (d'Orb.), 278

Encrinos, 202

Extracrinus, 182

Geocrinus, 168

Encrinus, 181

eye, 32, 245

Gibbaster, 323

Endobranchiata, 300

Gilbertsocrinus, 201

endocyclic (Ech.), 294

Fabricius, 0., 218

Gissocrinus, 175

— gut (Grin.), 136

Faorina, 322

glandular organ, 23

endothecal food-grooves, 41 ' fasciole, 319

— pedicellariae, 31

22

338

INDEX

(flaphyrocystis, 60

Ifalysiocrinus, 148

Qlcnotremites, 203

Hamaim, 0., 218, 240, 261

GlypJwcyphits, 312

Hapalocrinus, 156

Uly plaster, 198

Ihiplocrinus, 151

(ilyptechinus, 313

Jfaplocystis, 208

<;l>/pticus, 307

JIaplodactyla, 233

Ulyptocrinus, 161

haplopores, 41

Glyptocystidae, 58

Hamwcrinus, 162

Glyptocystis, 64

heart, 23

(Jlyptospliaera, 73

Hdianthaster, 258

Giiathaster, 253

Ildiaster, 258

Giiathocrinus, 203

Helikodiadema, 309

Gnathostomata, 315

Ifeliocrinus, 55

guathostoniate, 304

Hdiocystis, 55

Gnorimocrinus, 189

Helmintholithus, 203

Goldfussia (Norman), 203

Hcmiaster, 322

Gomphocystis, 77

ffemicularis, 307

gonads, 25

Hemicosmites, 66

Goniaster, 253

Hemicrinus, 197

Goniasteroidocrinus, 201

Hemicystis, 207

Goniocidaris, 302

Hemieuryale, 276

Goniocrinus, 179

Hemiglypka, 278

Goniodiscus, 253

Hemipatagtts, 321

Goniodon, 253

Hemipedina, 309

Goniopecten, 251

Hemipholis, 278

Goniophorua, 306

Hemipneustes, 321

Goniopneustes, 313

Henricia, 258

Goniopygus, 307

Herouard, E., 218

Gonocrinus, 60

Herpetocrinus, 146

Gorgonocephalus, 276

tfertAa, 203

Gothocrinus, 179

Heteroblastus, 90, 92

Grammechinus, 313

Heteiocentrotus, 313

Grammocrinus, 203

ffeterocidaris, 309

Granatoblastida, 92

Heteroclypeus, 321

Granatocrinus, 90

Heterocrinus, 146

Graphiocrinus, 180

Heterocystis, 67

Grasia, 321

Heterodiadema, 309

Gray, J. E., 240, 260

Heterolampas, 321

Gregory, J. W., 261, 310

Heterosalenia, 306

Grube, A. E., 226

heterotomy, 112

Gualtieri, N., 284

Hexacrimis, 158

(fualtieria, 323

I/exact is, 240

Guettardicrimis, 192

Hexalacystis, 66

Guettaria, 321

Uibernula, 195

gut, 136

Hippasterias, 253

Gyllenhal, J. A., 53

Hipponoe, 313

Gymnasteria, 254

Hoffmann, C. K., 285

Gyinnobrisinga, 259

Holaster, 321

Gymnocrinus, 197

Holectypiua, 315

Gyinnodiadema, 308

Holectypus, 316

Holocrinus, 182

Habrocrinus, 166

Jfolocystis, 47

fladrocrinus, 164

"Jfolocystites," 72

Haeckel, K, 40, 43, 48,

Molopneustes, 313

49, 52, 326

Holopus, 197

haemal strands, 23

Ifolothuria, 227

ffagenovia, 321

—f&rskali, 220

Jfaimea, 320

Holothurian plane, 20

Hallicystis, 61

Holothurioidea, diagnosed,

Jfalocrimis, 177

218

Jfalophentc, 203

Homalocrinus, 189

ffomoerinva, 179

HonuKystis, 64
Homoeaster, 323
Horn olam pos, 323
Hoplaster, 253
/loplocrinus, 145
Hupe, H., 40, 284
Huxley, T. H., 33
Jlybechinus, 313
Hyboclypeus, 320
Hybocrinus, 145
Hybocystis, 95, 145
Hydrasterias, 258
Hydreionocrinusi 180
Jfydriocrinus, 180
hydrocircus, 12, 102
hydrocoel, 7
hydrophores palmees, 44,

46, 73

hydropore, 8, 103
hydrospire plates, 86
hydrospires, 83
Hymenaster, 257
Hymenodiscus, 259
Hyocrinus, 153
Hypanthocrinus, 164
ffyperocrinus, 168
Hyphalaster, 251
Hypocrinus, 178
Hyponome, 77, 195
hypostereom, 45
Hypostoma, 260
hypothecal, 2
hypozygal, 108, 117
Ilypsospatagus, 323
Jfyptiocrin us, 199
Hystricrinus, 158

Ichthyocrinus, 188
Iconaster, 253
Icosidactylocrinus, 203
Idiocrinus, 188
llariona, 320
Ifycriniis, 186
llyodaemon, 229
imperforate articulation,

108
Impinnata, 187 ; (of P. H.

C.), 140
Inaduuata dieyr.lica, 171

— monocyclica, 144

— (W. & Sp.), 139
Inarticulata, 138
Indianocrinus, 145
iuferradial, 112
infrabasaliu, iiifrabasals

(Grin.), 99, 122 ; (Ast.),

246 ; (Oph.), 268
Infradypeus, 320
infra-marginals, 246
Infvlaster, 321

INDEX

339

interambulacra (Ech.), 286

lancet-plate, 82

Lor iol aster, 257

interanibulacrals (Crin. ),

Lanieria, 316

Loven, S., 14, 32, 285

109, 124 ; (Ast.), 241 ;

Lankester, E. Ray, 33

Loven's plane, 21

(Ech.), 296

Lapillocystis, 46

Lovenia, 323

interarticular substance, 28

Lapworthura, 275

Ludwig, H., 218, 226, 240,

interaxillaries, 109, 204

Larvata, 141

261

iuterbrachial areas, 246

Larviformia, 139

Luidia, 252

iuierbrachials, 109, 118, 204

Lasiaster, 254

Luidiaster, 251

intercellular substance, 28

lateral arm-plates, 270

lunules, 293

intermarginals, 246

Laubeocrinus, 170

Lyman, T., 261

interned als, 107

Lebrunaster, 254

lymph gills, 245

interpinnulars, 118

Lecanocrinus, 188

— gland, 23, 27

interradii, 20

Lecythiocrinus, 177

Lyriocrinus, 200

intersecundibrachs, 118

Lecythocrinus, 175

lysactinic, 237

focrinus, 145

Leiaster, 255

Lysechinus, 305

Iraniaster, 323

Leiocyphw, 312

I.ysocystis, 70

Irpa, 229

Leiopedina, 313

Lysophiurae, 274

Irregulares (Blastoidea), 91

Leiosoma, 310

Ly taster, 258

Irregularia (Crin.), 139 ;

Lenifa, 317

(Ech.), 315

Le doer in us, 61

M plane, 20

Isastcr (Verr. ), 251

Lepadocystis, 61

Macnrocnnus, 166

— (Desor.), 321

Lepidaster, 255

MacBride, E. W., 13, 14,

Ms, 182

Lepidesthes, 305

240, 261, 285, 325

Isocrinns (Meyer), 18-

Lepidocentrus, 301

Mackintosh, H. W., 285

— (Phill.), 189

Lepidocidaris, 302

Macraster, 321

Isopneustes, 323

Lepiiiodiscus, 208

Macrocrinus, 168

isotomy, 112

Lepidopleurus, 312

Macrocy stella, 56

Lepocrinites, 61

Macropneustes, 323

Jackson, R. T., 299, 302

Leptaster, 253

Macrostylocrinus, 162

Jaeger, G. F., 226

Leptasterias, 258

inadreporic gland, 23

Jaekel, 0., 114, 116, 141,

Leptocidaris, 307

madreporite, 11 ; (Ast.),

154, 171, 300

Leptocrinus, 166

241 ; (Ech.), 286. See

jaw (Oph.), 263

Leptocystis, 204

also hydropore

— plate, 264

Leptogonctster, 253

Maynosia, 309

Jeronia, 321

Leptopty chaster, 252

main-axil, 114

Juglandocrinus, 67

leptostroterate, 254

Malocystis, 58

Leske, N. G., 284

mamelon, 287

Kattispvngia, 203

Leskia, 323

Mammaster, 255

kidney, 23

Leuckart, R., 2, 33, 326

Maretia, 323

klasma-plates, 297

faucophtlialmus, 53

marginals, 241, 246

Klein, J. T., 283

Lichenocrinus, 133

Marginaster, 254

Kohja, 229

Lichenocystis, 56

Mariacrinus, 161

Koninckocidaris, 302

Lichenoides, 56

Marsipaster, 257

Koninckocrinus, 203

ligament tissue, 28

Marsipocrinus, 156

Korethraster, 256

ligamentar fossae, 116

Marsupiocrinus (Blainv.),

Kowalevsky, A., 218

Linchoplwrus, 323

185

Krohn, A., 240, 261, 284

Linck, J. H., 239, 260

— (Hall), 200

Linckia, 255

— (Phill.), 156

Labidiaster, 259

Linnaeus, C., 53, 240, 260,

Marsupites, 185

Labidndemas, 227

284

marsupium, 256

lacunar plexus, 102

Linopneustes, 323

Mastigocrinws, 179

— system, 26

Linthia, 323

Mayraster, 257

Laetmogone, 229

Lithocrinus, 189

Meckel, J. F., 240

Laganum, 317

Ljungrnan, A., 260

Mediaster, 253

Lageniocrinus, 152

Lobocrinus, 168

Medusacrinus, 203

Lahuseniocrinus, 200

Lobolithus, 77, 136, 161

Medusaster, 259

Lamarck, J. P. B., 226,

Lodanella, 46

Meekocystis, 61

240, 260, 284

Lonchotaster, 251

Megacystis, 47

Lambertiaster, 323

loose suture, 108

Megalaster, 321

Lampadaster, 321

LopJiaster, 256

Megistocrinus, 168

Lampterocrinus, 199

Lophocrinus, 179

Melitella, 317

340

INDEX

Mdlita, 317

Moriaster, 252

Oolaster, 321

Melocrinoidea, 160

Mortonia, 317

Opechinns, 312

Mdocrimts, 161

Moulinsia, 317

Ophiacantlia, 278

Melonites, 304

mouth -frame, 263

Ophiactis, 278

Melonitoida, 303

mouth papillae, 241

Ophiarachna, 277

Meoma, 323

Mndleria, 227

Ophiarthrum, 278

meridosternous, 318

Miiller, Joh., 40, 138, 218,

Ophidiaster, 255

Merocrinus, 178

240, 260, 261, 285

Ophiemus, 278

Meseres, 227

muscle fibres, 27

Ophioblenna, 278

MesoUastus, 90, 92

muscle fields, 262

Ophiobyrsa, 276

Mesocrinus, 194

— fossae, 116

Ophiocamax, 278

Mesocystis, 75

Mycocrinus, 150

Ophiocentrus, 278

mesostereom, 45

Myelodactylus, 146

Ophioceramis, 278

Mesothnria, 227

Alyriotrochus, 234

Ophiochiton, 278

Mespilia, 313

AfyrtillocrinuSi 177

Ophiochondrus, 276

Afespilocrinus (de Kon.),

Myxaster, 257

Ophiochytra, 278

188

Ophiocnemis, 278

— (Quenst.), 192

Nacospatangiis, 323

Ophiocnida, 278

Mespttocystis, 77

Arano&°inust 177

Ophiocoeta, 277

Metablastus, 87, 92

Narcissia, 255

Ophiocvma, 278

Metacrinus, 183

Nardo, J. D., 240

Ophioconis, 277

Metalia, 323

Nardoa, 255

Ophiocreas, 276

Metapvrhinus, 321

A7ectria, 253

Ophiocrene, 276

Metopaster, 253

Nematocrinus, 150

Ophiocrinus (Ang.), 179

Metrodira, 255

Neocatopygus, 321

— (Charlesw.), 146

Metschnikoff, E., 218, 285

Neocrinoidea, 139

— (Salter), 201

Micraster, 323

NeocriiiHS, 182

— (Semper), 195

Microcrinus, 203

Xeocystis, 77

Ophiocten, 278

Microcyphus, 313

Neolampas, 321

Ophiocymbium, 278

Microdiadema, 309

Neomorphaster, 255

Ophiodenna, 277

Microlampas, 321

Neoplax, 276

Ophiogerwi, 276

Micropedina, 310

Neopneustes, 323

Ophioglypha, 278

Micropocrinns, 198

Nepanthia, 254

Ophiogona, 277

Micropsina, 313

Neumayr, M., 44, 48, 49,

Ophiogymna, 278

Micropsis, 310

326

(fphiohdus, 276

Micropyga, 309

Neviani, A., 310

Ophiolebes, 278

miliary granules, 287

Nidorellia, 254

Ophiolepis (Lvman), 278

Miller,- J. S., 97, 138

Nipterocrinus, 188

— (M. & T.), 278

Miller, S. A., 47, 260

nodals, 107

Ophiomartvs, 278

Mttlericrinus, 191

Nucleocrinus, 88, 92

Ophiomastix, 278

MiUetvi, 321

Xudeolites, 320

Ophiomaza, 278

Mimaster, 253

Nucleopygus, 320

Ophiamitra, 278

Mimocystis, 56

Xyinphaster, 253

Ophiomusium, 278

Miospondylus, 275

Ophiomyces, 276

Missouricrinus, 153

oculars, 286, 294

Ophiomyxa, 276

MiteU])haster, 253

Ocftnia, 259

Ophionema, 278

Mithrodia, 258

Odontaster, 253

Ophi&nephthys, 278

Mitraster, 253

odontophore (Ast.), 247

Ophionereit, 278

Hfitrocrinus, 203

0/aster, 321

Ophiopaepale, 278

Mitrocystis, 50

Ogmaster, 253

Ophiopege, 275

3/otVo, 323

Ohiocrinus, 146

Ophiopeltis, 278

Afoiropsis, 323

Oken, L., 226

Ophiopeza, 277

Afolpadia, 233

Oligopodia, 320

Ophiophdi*, 278

Jfonaster, 250

Oligoporus, 304

Ophwphragonus, 278

monocyclic, 99, 110

Oligopygus, 320 Ophiophyllum, 278

Monocyclica, diagnosed.

Olivanites, 88

Ophioplax, 277

144

QUaerinua, 201

Ophioplinthus, 278

— Camerata, 159

Oncocrinus, 187

Ophioplocus, 278

— Inadunata, 144

Oneirophanta, 229

Ophiopsainninrrij 278

Monophora, 317

Onychaster, 275

Ophiopsila, 278

Monostychia, 318

Onychocrinus, 190

Ophiopteris, 278

INDEX

34i

Ophwpteron, 278

Palaeechinasteridae, 250

Pentaceropsis, 254

Ophiopu*, 278

Palaeechinoidea, 299

Pentaceros, 254

OphiopyrcA, 277

PaUutchinv*, 304

Pentacrinug, 182

Ophiopyrg**, 277

Po^o«*riMu«, 322

Pentactaea, 12

Ophiosciasma, 276

PI^MOOWMI, 254

Pentad ula, 18

Ophiostigma, 278

Palaeocrinoidea, 139

Pentadia, 181

fr*ttmu,276

Palacocrinus, 172

Pentagonatter, 253

OohioUiamJuu. 278

Palaeocvstis. 54

Pejitagonitzt, 203

Ophiothda, 278

.* WMnAsy a<rl«, »/•«

Palatodiscu*, 301

pentameres, 99

OokMtholia. 276

Palaeolampat 321

Pentephyllum, 90, 93

Sfjf* f . . . . . . -t-. —4 J

Ophiothrix, 278

Palaeonectria, 250

Pentrcautc*, 86, 92

Ophwtrema, 278

PaUuophiura, 274

Pentremitideo, 85, 92

OpUotrocbu, 278

Palacopneustes, 322

perforate articulation, 109,

Qpfcuwmo, 278

PalaeogteUa, 250

116

Gfr**«cni,278

Palaeostvnia, 323

PeriboUuter, 256

— d/tari*, 262

Palaeotropus, 323

Pericosmus, 323

Opkiurella, 27S

Pcdasterisats, 257

Periechocrinus, 168

Ophiurina, 275

Palastropccten, 275

PeriglyptocriniiA, 161

Ophiuroidea, 259

Pallas, P. S^ 218

perignathic girdle, 289,

— diagnosed, 273

Palmacystis, 73

297

oral angle (AsU 246

palmaria, 204

peri- intestinal cavity, 103

plates (Oph.), 263

Palmipe*, 254

perioesophageal sinus, 22

— nerve-ring, 102

palps (AsL), 245

periproct, 286

— nervous system, 30

PannycJiia, 229

Pcrischodomus, 302

- papillae, 264

papulae, 26, 245, 248

— mognus, 203

- ridge (Blast), 82

parabasalia, 99

perisoma, 241

— skeleton (Ast), 241

Paractinopoda, 234

peristome (Crin.), 97 ;

orals, 124. See also del-

ParadaaxehinHt, 312

(Ech.), 285

toids

Paragonastcr, 253

peristomial plates (Oph.).

OrbUnmtia, 90, 92

Paralampas, 321

264

Orttete, 230

pararadials, 150

Perknoster, 258

Greater, 254 Panwrfcuter, 251

Peronio, 308

orientation of crinoid, 110 Parasalenia, 313

perradii, 19

Oriolamptu, 321 Pardpidia, 229

Perrier, E., 241, 259, 285

Onuthafter, 323 parietal canal, 9

Perttphonoster, 251

oro-centnl, 14 ParisocrinHs, 173

Pctolvcriniu, 175

Orocyrfw, 54 Pcuceolus. 77

Petinocrintu, 203

OropAocriiMW, 84, 92

PatdliocriKu*, 162

Pctnuter, 250

Ortfodfcuuu, 310

patina, 112

PA^fuMcAtnna, 84, 92

Ortkoddari^ 302

Patina, 254

Phanenuter, 253

OrtAoeriaiu, 198

Pov/uz,254

Phanerozonia, 249

Ortkolopk**, 312

parillae, 247

Phanogenia, 196

OrthopsU, 308

PcctinouUr, 251

PAorui,255

OMrite cr:-;e. 25o

peetinirhombs 43, 58

pharyngeal vesicles, 27

otocy«ts,32

Pcctinura, 278

Phialocrinus (Eichw.), 203

OtfteMurtiuu, 178

Pedatae,226

— (Traatsch.), 181

outer side-plates (Blast), 84

pedkellariae, 31; (Ast),

PhiUipsocrinus, 170

Ondfepou, 316

247 ; (Ech.), 287

Philocriniu, 180

OToid gland, 23. See also

Pedicdlastcr, 258

Phivwerinv*, 152

axial organ

/'w'i'-.i KM

Phoeniooerinut, 166

<MMbr.Sn

Pedinopsi*, 313

PholuUuter, 255

P«iM0a«ria, 310

Pholidoddarix, 305

Packyan^d^ 203

Pelagvthuria, 230

Pkormoama, 311

PmAgdtr** 316

Pdancckinu,, 310

Phoxatter, 252

ftadfcripiui 203

pellions, 289

PhyUobrusut, 321

Packyteerinui, 180

PelmatoBom, 33

PhyUodypeuf, 321

PoeAfocriitvc, 203

- diagnosed, 38

PhyOocrimu*, 197

Padanahde* 227

T v .--v^,-;^ o-> Q

^^••^M^HU«UCa9 £>£•%

PmlmeaUr, 250

MfasiBt, 3: 6

pnjuoue, oi»
PkyOoptona, 230

P.:,".7^/.7or->-in.:-'. '2T 0

Pcvkdocri***, 148

PAyswcUnu, 309

PllMMlMidM, M

Pentagon*, 229 ! Phytctocrinut, 170

342

INDEX

Phytocrinus, 195

pores, ambulacral (Ech.),

i'terocoma, 203

Pictet, F. J., 40

287

Pterotocrinus, 159

Picteticrinus, 1 82

— of Blastoids, 86

Ptilonaster, 274

J'ileus, 316, 320

Porocrinus (Bill.), 172

Ptychocrinus, 198

Pilocystis, 46

— (Dittm.), 182

Pupa, 5

Pinnata (P. H. Carp.), 140 ;

postpalmaria, 204

Pycnaster, 253

(Bather), 191

Poteriocrinacea, 171

Pycnocrinus, 161

pinnulae, 113

Poteriocrinidae (W. & Sp.),

Pycnopodia, 258

Pionocrinns, 166

171

Pycnosaccus, 187

Piracy stis, 47

Poteriocrinus, 180

Pygaster, 316

Pisocrinus, 149

Pourtalesia, 323

Pygastrides, 316

Placocystis, 51

Pradocrimis, 168

Pygaulus, 320

Placodiadema, 309

Prenaster, 323

Pygorhynchus, 321

plastidogeu organ, 23

pre-oral lobe, 4, 8

Pygorhytis, 321

Platasterias, 252

primary skeletal elements,

Pygospatangus, 323

Platybrissus, 322

109

Pygurostoma, 321

Platyceras, 157

primaxil, 114

Pygurus, 321

Platycriniis, 157

primibrachs, 114

pyramids (Ech.), 289

Platycystis, 51

primitive suture,, 108

Pyrina, 320

Platysphaerites, 203

Proclivocrinus, 148

Pyrocystis. See Piracy si >'s

Plectaster, 258

Prognaster, 255

Pythonaster, 257

Plesianthus, 318

Promachocrintts, 195

Pyxidocrinns, 168

Plesiocidaroida, 305

Protaster, 274

Plesiolampas, 321

Protasteracanthioii, 251

radial canals, 243. See

Pleurechinus, 312

Proteocysiis, 73

also perradial

Pleurocrinus, 157

ProteroUastus, 74

— dome-plates, 127

Pleurocystis, 64

Proteuryale, 203

— facet, 100

Pleurodiadema, 309

Protoblastoidea, 79

— pseudhaemal canal, 100

plexiform gland, 23

Protocrinus, 74

— shields, 264

Plicatocrinus, 153

Protocyamus, 316

— sinus, 84

Pliny, 218

Protocystis, 48

— vesicles, 27

Plistophyma, 309

Prouho, H., 245, 285, 288

radialia, radiivls (Crin.), 99,

Plumaster, 259

proximal (Crin.), 109

204; (Ast.), 240 ; (Ech.),

PJ^ews, 6, 16 ; (Oph.), 2/3

proximale, 108

294; (Oph.), 268

Plutonaster, 251

Prunocystis, 60

radianal, 112, 119

Pneumophora, 226

Prymnadetiuae, 322

Radiaster, 254

Podasterias, 258

Prymnodesminae, 323

Radiata, 2

podia, 1, 2; (Ast.), 243;

Psammechinus, 313

radical cirri, 1C7

(Ech.), 291

pseudambulacra (Cyst. ), 43 ;

Radiocyphus, 312

Podocidaris, 308

(Blast.), 78

radix, 99

Polian vesicles, 27 ; (Ast),

PseudarcJiaster, 251

rami, 113

243 ; (Ech.), 291

Pseudaster, 251

ramuli, 113

Polyasterias, 258

pseudhaemal ring, 102

rectal caeca, 243

Polycerus, 182

— system, 25 ; (Ast.), 243 ;

recumbent arms, 43, 65

Polycidaris, 302

(Ech.), 292

Regulares (Blast.), 91

polycyclic spine, 287

Pseudoboletia, 313

Regularia (Crin.), 139

Polycyphus, 310

Pseudocatopygus, 321

— Ectobranchiata, 306

Polypdtes, 162

Pseudocrinns, 62

— Endobranchiata, 300

Polypholis, 278

Pseudocucnmis, 230

Remaster, 256

Pomatocrinus, 191

Pseudodesorella, 321

respiratory trees, 27, 222

Pomel, A., 284

Pseudodiadema, 309

Retaster, 257

Pomocystis, 72

pseudo-heart, 23

rete mirabile, 223

Ponwnites, 72

pseudomonocyclic, 111

Reteocrinus, 198

Pomosphaera, 72

pseudo-pedicellariae, 247

ItJiabdocidaris, 302

Pontaster, 251

Pseudopedina, 309

Rhabdopleurus, 312

Porania, 254

Pseudostichopus, 227

Rliadinocrinus, 179

Poraniomorpha, 254

Psilaster, 252

Rhaphanocrimts, 200

Porcellanaster, 251

Psolidium, 230

Rfogaster, 254

pore canal, 8

Pso^M*, 230

Rhinobrisms, 323

— plate, 80

Psychropotes, 229

Rhipidaster, 256

— rhombs, 42 i Psychrotrcphes, 229

Rhipidocriws, 201

INDEX

343

Rhiwcrinus, 194

Scyphocrinus (Zenker), 161

Steiidiocrinus, 161

Jihodocalix, 203

Scytactinota, 218

^e//a, 260

Jihodocrinus, 201

Scytalecrinus, 180

Xtellae Jissae, 240

Ithoechinus, 304

Scytodermata, 33, 218

Stellaster, 253

Rhombifera, 52

Seba, A., 284

Stelleroidea, 237

Rhombifera (Barr.), 57

secondary skeletal elements,

— diagnosed, 239

— mira, 77, 145

109, 204

stem, 48, 99, 105, 132.

Rhopalocoma, 257

— radials, 204

See also columna

Rhopalocrinus, 191

secundibrachs, 114

Stemmatocrinus, 181

Rhopalodina, 230

segmental apertures, 256

Stenaster, 251

Rhynchopygus, 321

Selenka, E., 218, 226

Stenocrinus, 146

Roemeraster, 255

Semi-Articulata, 138

Stenonia, 321

llomanes, G. J., 245

Semon, R., 12, 18, 218

Stephanaster, 253

Rondelet, W., 218, 283

Semper, K., 218

Stephanocrinus, 96, 145

root, 99

sensory buds, 31

Steraster, 257

Rosaster, 253

sessile pedicellariae, 247

Stereocidaris, 302

rosette (Grin.), 195 ; (Ech.),

Shumardocrimis, 203

Stereocrinus, 164

289

side-plates, 82

stereom, 28

.Rotate, 317

Sigsbeia, 276

Sternata, 321

Rotuloidea, 317

Simroth, H., 261

Sternotaxis, 321

Rouault, M., 45

Siphonocrinus, 199

sternum (Ech.), 318

.ftwn«, 317

Sismondia, 317

Stewart, C., 285

Russo, A., 261

Sitularia, 185

Stewart's organs, 303

Sladen, W. P., 14, 240,

Xtichastcr, 255

Saccocoma, 154

251, 259, 261, 325

Stichocystis, 55

Saccocrimts, 168

Smilasterias, 258

Michopus, 227

sacculi, 137

Solacrimis, 203

Stigmatopygns, 321

Sagenocrinus, 190

Solanocrinus, 195

Stirechinus, 313

sagittal plane, 20

Solaster, 256

stomach (Ast. ), 243

Salmacis, 312

Spatagocystis, 324

Stomatocrinoidea, 139

Salmacopsis, 312

Spatangomorpha, 323

Stomechinus, 309

Salenia, 306

Spatangus, 323

Stomopneustes, 313

Salter, J., 260

Sphaeraster, 254

Stomoporus, 323

Salteraster, 251

Sphaerechinus, 313

stone -canal, 8; (Ast),

Sampsonocrimis, 170

sphaeridia, 32

243 ; (Ech.), 291

Sarasin, P. & F., 32, 285,

Sphaerites (Quenst.), 19,

Stortingocrinus, 1 52

310, 326

254

Streptaster, 208

SarseUa, 323

Sphaerocrinus (M. & W.),

Streptocrinus, 179

Scaphiocrinus, 180

167

Streptophiurae, 274

scapula, 204

— (Roem.), 173

streptospondyline, 271

Schizaster, 323

Sphaerocystis, 63

Stribalocystis, 67

Schizoblastus, 89, 92

Sphaeronis, 71

Strobilocystis, 63

schizocoelic system, 26

Sphaerothuria, 230

Stromatocystis, 207

Schizocrinus, 161

Sphenocrinus, 203

Strongylocentrotus, 313

Schizocystis, 61

spines (Ech.), 285, 287

Strophocrinus, 172

Scfioenaster, 250

Spinigrada, 260

Strotocrinus, 170

Sclerasterias, 258

spiracula, spiracles (Blast.),

Studeria, 321

Sclerocriniis, 197

84 ; (Ast.), 256

Stiirtz, B., 250, 260

Scoliocrinus, 178

spongy organ, 27, 102

Sturtzura, 274

Scoliocystis, 60

Sporasterias, 258

Stylocrinus, 152

Scolocystis, 70

Springer, F. See Wachs-

Styracaster, 251

Scotoanassa, 229

muth

subanale, 204

Scotoplanes, 229

Spyridiocrinus, 163

subepithelial nervous sys

scrobicular circle, 287

Statozoa, 33

tern, 30

scrobicule, 287

Staurocystis, 61

sub-lancet, 85

scuta adoralia, 264

Staurosoma, 57

suboralia, 141

Seutella, 317

Steganobtastus, 209

subradiale, 204

scutella oralia, 263

Steganocrimis, 170

subradialia, 204

ScuteUina, 317

Stegaster, 321

subtegumentary cavity, 103

scutum buccale, 264

Stegmaster, 254

sub tentacular canals, 100

Scyphocrinus (Hall), 161

Steinmann, G., 40

superbasals, 159

344

INDEX

super-radial, 112

Thiolliericrinus, 195

Trybliocrinus, 163

supplementary plates, 109,

Tholaster, 321

Tuberaster, 323

204

Thoracaster, 251

tubercles (Ech.), 286

supra-raarginals, 246

Thylacocrinus, 200

Turbinocrinus, 161

suraual, 294

Thylechinus, 310

Tylaster, 254 •

sutures, 108

Thyone, 230

Tylocidaris, 302

Sycocrinus, 203

Thyroida, 205

Typocidaris, 302

Sycocystis, 60

Thysanocrinus, 198

Syinbathocrinus, 152

Tiaracrinus, 57

Uintacrinus, 183

Symphytocrimts, 197

Tiarechiniis, 305

Ulocrinus, 181

Synallactes, 227

Tiedemann, F., 218, 240,

umbo (Ech.), 262

Synapta, 234

284

umbonal socket, 263

Synerocrimis, 190

Tiedemann's bodies, 27,

under lancet-plate, 85

syngnaths, 263

243

Uniophora, 258

Synyphocrinus, 180

Tcmta, 255

uniserial arms, 115

Syringocrinus, 48, 203

Tormocrinus, 197

Uperocrinus, 168

syzygy. ios, 117

Tornaria, 5

Urasterella, 251

torus angularis, 264

Urechinus, 321

tables (Hoi.), 223

Torynocrinus, 197

Taeniatter, 251

Toxaster, 322

Valentin, G., 284

Taeniura, 274

Toxopneustes, 313

Valvaster, 258

Talarocrinus, 159

Traumatocrinus, 182

Trosocrt'?m5, 179

Tanaocrinus, 165

Trautschold, H., 130

vault, 128

Tarsaster, 255

Tremaster, 254

ventral sac, 27

Taxocrinus, 189

Tremataster, 275

— shield, 262

Technocrinus, 1 63

Trematocrinus, 201

vertebral ossicle (Oph.), 262

tegmen, 99, 129

Trematocystis, 72

vestibule, 11

Teleiocrinus, 170

Trematopygus, 320

Viguier, C., 240, 247

Temnechinus, 312

Triacrinus, 149

Vletavicrinus, 204

Teinnocidaris, 302

Trianisites, 204

Volborth, A., 53

Temnopleurus, 312

tribrachialia, 204

Vologesia, 321

tentacles (Hoi.), 224

Tribrachiocrinus, 181

terminal tentacle, 31

Trichaster, 276

Wachsmuth, C., 139

terminals (Ech.), 294

Trichasteropsis, 250

Wachsmuth & Springer,

tertibrachs, 114

Trichocrinus, 149

114, 139, 171

Tesselata (Zittel), 139

Trichotaster, 259

Wachsmuth & Springer's

Tessellata (Miillar), 139

Tricoelocrinus, 87, 92

Law, 106

test (Ech.), 285

trifascial articulation, 117

Walther, J., 49

Testacea, 138

Trigonoddaris, 312

water - vascular system

Tetanocrinus, 197

Trigonocrinus, 197

(AsU, 243 ; (Oph.), 272 ;

Tethyaster, 251

Trigonocystis, 50

(Ech.), 290

Tetracidaris, 303

Trinemacystis, 53

Woodocrinus, 180

Tetracrinus (Austin), 203

Triplacidia, 310

— (Miinst.), 153

Triplaricrinus, 204

Xenaster, 250

Tetractis, 240

Tripneustes, 313

Xenocrinus (Jahn), 204

Tetramerocrinus (Austin),

Trisactis, 240

— (S. A. Miller), 165

204

trivium, 11 ; (Hoi.), 220 ;

Tetraster, 250

(Ech.), 296

Zeacrinus, 180

TkalamocrinuS) 204

Trochita, 204

Zenkericrinus, 161, 204

Thallocriniis, 156

Trochocrimis, 168

Zeuglopleurus, 3J2

Thaumatocrinus, 196

Trochocystis, 50

zigzag arms, 115

theca, 99

Trochoderma, 234

Zittel, K. A. v., 40, 139,171

Thecocystis, 207

Trochodota, 234

Zophocrinus, 151

Thecoidea, 205

Trochostoma, 233

Zoroaster, 255

Theel, H., 15, 218, 226

Troostoblastida, 92

Zygocrinus, 91, 93

Thedia, 230

Troostocrinus, 86, 92

Zygophiurae, 277

Thegastfr, 317

Tropidaster, 255

Zygospondyline, 271

Thenarocrinus, 179

Troschel, F. H., 240, 260

THE UNIVERSITY LIBRARY

UNIVERSITY OF CALIFORNIA, SANTA CRUZ

SCIENCE LIBRARY

This book is due on the last DATE stamped below.
To renew by phone, call 429-2050

MAR 1 4 1973
REC'O MAR 1 6

MAY 1 6 1973

I » L,v U

MAR 2 8 1980
REC'D MAR 1 8 1980

30m-9,'72(Q458588)— 3A-1

TORED AT NRI P