to ee res 2, La —
+ Veet Ores err eres ~e Cee mee ws
Se aiid * en

eer cower
Frecce seeccewewe
Goceeten
os - oe
soe veer evvew ewes
reowevureces 10ers perverse Vere

eee ewe owe eenwet wvewee
ee ew ewww o9.

oo were
yewveesweces

ove wer rewrre re
- —~~
—

<o- eevee wee reese wee
“~~ worry

iveew eee we

cS ~

- *

Were sere rrs

baat AL AM vewvyys { sega MNES wees cattle gad te eT Mc cy: watt

a A hoe AAA Jv" MAS, ‘y A "tel oUF -
| f i? eww « : a & oy wy ‘\ C% ef 42. “hed Be iad weve Yew ives
"t he = iinet peal YY ics, eet thy ye MEM
wT ye ew ‘. ty

} . IUCN eeu TEI. ‘ vey! Uwe ye ; Wve yyn'
RS aaah ad TCOS — Y) WW yyudeee wy Ai | AL) LM wr ate ln “l~~ Wyyeve

wv Ah ew we > a VaJd¥ id ys vv"
eee? He eee We ve. aan one iy yvererilinne Cade 0S
Mite eC dv ~*~ ye! ve Pd v's “=~ AEM AT A i
Need si PRM Ansty thle PA MY IM etd wl ideg vit
aN a AAJ re USUI “ | \ ‘av Aydd @ da

yuh ww ~ e v | 4 hae Ze
vole Hevea th se Wihhadehngchaet Me siege 4 Liv ;
Wy ae vd Weevese very fis sagad vc ae
Winey diasaliet se HH re eeu" Yemen Yee we ve
ae aga ee ONTO AY : nd gti aye wer ieaeT oe eee ng

1 v
yee ’ vive huey ewe. hh hd
wide WV yeti Sey JJ pebAR | P Sao wey awivyuy ste wry TYYIT
ve Leecnutuvery LEes tes ~ a ~d vue Bae, id Wy 1) ey ‘vw a SUsy Get Sen gree ee wv"
ve J MAAF AG LAAAN PAS ¥) Gores She we So ely p> Y
‘ dese wel : hE RA aaa | Re ete real i eer see Ue wee Een Aes,

ei | Se dNmceelmig 10g cet ae ceed
yee Sa hh ie OF awe pet yuyttayvevwelgyvye TONNES

¢

were wee

a vv_¥ ste? Bae
ESE PELE) Le Matt dc saa Th swe tegetyuadaaedevedyar
fen ~\. : z)
pepe Patebst I VA. eye yse Ww jw" ; YIMANA,

we?

Peaster ee a metal Ween

~~~ Wauewee

prs

Paha Wedgie ew ety a ye ett ed ees veuuve
SI Mggagrow Madani Se oe weuees us vw ee, Wu Ln vyeusvdtieg” tebie

sity Ags ie Foray jod SS iene id yeild se 9 ' <p wa AS wry

oS os Cig ae ea na sees tadie ds rtrhitn oa eer Swi! = tl Je ieee Cetera

So WOT IN

’
~

ove

yy -~ AAS Ls 2°
Mregysree eth weve Se ay

vw a

ne bad
we wv

’

Wey vere, ww

Nw eeetey ee, +4

#iyert ;
v dvyr" i] wor verve A Ae . dv
wd i¥w ww" We eeceee’ vie NY
: tit yovdtw ete OS Ww wet 7 ve hate. oP} veel wd wet vvewe regou were tered TT
Ud . a “wr w’s vw 3 A Nth ef ae Sige ese

bids Sane weeny eo, see. I A Ell
ws posvlgpeyd A OPTED ec ange pvr vete ede
yet vi

- ~~. w
SN) nd ed wt’ JIVWW yw"

ae z -
~
hy ! WAV Ay Uveew .
j= ’ Je oH Uv ~

a et
pattie gitatddddd? ddédouyee

Lu VWI. eer ale ics ret
aH pegerertcesnceWPowwins ciety aug : acl i
FOO to Es eae Hie casey Sy f rtpet
Sept PE oe ai pr intaene its
peredetteeeeeleuttneyy © ec reinnNeNy. create anaceetinentett
. fa “vice stwes ded Cady,

WV o, Vcclliinelias: oes EY IS WVvy ad Vdvi¥e ~~~"

PALAONTOGRAPHICAL SOCIETY.

INSTITUTED MDCCCXLVII.

LONDON

MONOGRAPH OF THE BRITISH FOSSIL ECHINODERMATA FROM
THE CRETACEOUS FORMATIONS.

Vou. [].—Astreroipea AND OPHIUROIDEA.

Mr. Sladen is the author of pages 1—66, Plates I—XVI, while Mr. Spencer
is the author of the remainder of the Volume. Mr. Spencer desires to express
his indebtedness to Dr. F. A. Bather for much help and advice in his share of
the work.

ORDER OF BINDING AND DATES OF. PUBLICATION.

ISSUED IN VOL.

PAGES | PLATES ener S PUBLISHED
|
Title-page | — | 1908 ' December, 1908
1—28 | I—VIII i; 1890 | April, 1891
29—66 IX—XVI | 1893 December, 1893
67—90 XVII—XXVI | 1905 | November, 1905
91—132 | XXVII—XXIx | 1907 December, 1907
138—138 (including Index) o | 1908 December, 1908

The Plates are intended to be collected and bound at the end of the Volume.

a MON OG RA PH

ON THE

bet itsk Fossit

KCHINODERMATA

FROM

THE CRETACEOUS FORMATIONS.

VOLUME SECOND.
THE ASTEHROIDEA AND OPHIUROIDEA.

BY

We ekneyY SLADEN, F.L.S., F.G.S.,

AND

Weoeheee NCE, B.A, F.G.S,

RALZEG
od ae ees
Letion
LONDON:
PRINTED FOR THE PALMHONTOGRAPHICAL SOCIETY.

1891—1908.

PRINTED BY ADLARD AND SON, LONDON AND DORKING.

PALAONTOGRAPHICAL SOCIETY.

INSTITUTED MDCCCXLVIL.

VOLUME FOR 1890.

BON DO Ns

A MONOGRAPH

ON THE

Bria FOSS i

ECHINODERMATA

FROM

THE CRETACEOUS FORMATIONS.

VOLUME SECOND.

THE ASTEHROIDEA.

BY

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

SECRETARY OF THE LINNEAN SOCIETY.

PART FIRST.

Paces 1—28; Puatres I—VIII.

LONDON:
PRINTED FOR THE PALZONTOGRAPHICAL SOCIETY.
1891.

PRINTED BY
ADLARD AND SON, BARTHOLOMEW CLOSE.

A MONOGRAPH

ON THE

FOSSIL ECHINODERMATA

OF THE

CRETACEOUS FORMATIONS.

THE ASTEROIDEA.

INTRODUCTORY REMARKS.

Ir was the intention of the late Dr. Thomas Wright to have continued his
magnificent series of Monographs on the British Fossil Echinodermata of the
Oolitic and Cretaceous Formations, which have already appeared in the volumes of
the Palzontographical Society, by the publication of a Monograph on the Creta-
ceous Asteroidea. With this object in view anumber of plates had been prepared
under Dr. Wright’s directions, and some preparatory notes for the letterpress had
been written, when the work was cut short by the lamented death of the author.
Subsequently the Council of the Palaontographical Society did me the honour of
inviting me to undertake the memoir. The plates and notes above mentioned
were placed at my disposal, but the latter proved to be for the main part merely
summaries or transcripts of descriptions already published, and were unfortunately
unsuitable to form part of the letterpress. For the whole of the latter I am

l

2 FOSSIL ASTEROIDEA.

therefore responsible. The plates which were drawn on stone have all been
utilised, although the specimens illustrated were not in every case those which I
should have selected, nor the order in which the figures are associated on some of
the plates that which I should have followed. This, however, is a comparatively
small matter, and the remark is not intended in any way as disparaging the
excellence of the illustrations. Indeed, I would here bear unqualified testimony
to the careful and accurate way in which the fossils have been delineated by
Mr. A. H. Searle. His plates are monuments of patient study of morphological
detail, and of exquisite technical execution as examples of lithographic drawing.

In his Monograph on the Oolitic Asteroidea, Dr. Wright gave as an intro-
duction a general account of the structure of the main divisions of the Asteroidea
then known, recent as well as fossil, with special reference to the calcareous
framework ; and he also gave a summary of the different systems of classification
which had been formulated by previous writers on the subject. It would there-
fore, in my opinion, be out of place, and in a certain measure superfluous, to preface
the present memoir with a similar introduction; but, as the knowledge of recent
Starfishes has been considerably extended since the date of Dr. Wright’s contribu-
tion, I propose to give in an appendix to this monograph my views on the classi-
fication of the Asteroidea, with special reference to the fossil forms.

At the commencement of his splendid Monograph on the Cretaceous
Echinoidea—to which the present memoir is a sequel—Dr. Wright gave a valuable
stratigraphical summary of the Cretaceous Formations in Britain. It is conse-
quently altogether needless to burden the pages of the Society’s publications with
a repetition of these details. I shall, however, if necessary on the completion of
my work, give a synopsis of the distribution in time of the various species dealt
with, together with such remarks on their occurrence and associations as occasion
may require.

FOSSIL ASTEROIDEA.

DESCRIPTION OF THE CRETACEOUS SPECIES.

Sub-class—EUASTEROIDEA, Sladen, 1886.
Order—PHANEROZONTA, Sladen, 1886.
Family—PENTAGONASTERIDA, Perrier, 1884.

Phanerozonate Asterids, with thick and massive marginal plates, which may
be either naked, or bear granules or spiniform papillae. Disk largely developed.
Apical plates often increscent. Abactinal surface tessellate, with rounded, poly-
gonal or stellate plates, which may be tabulate or paxilliform. Actinal inter-
radial areas largely developed, covered with pavement-like plates, which may be
naked or covered with membrane, or may bear granules or spinelets.

The family Pentagonasteridz, as defined by Prof. Edmond Perrier’ in 1884,
was separated from a larger and more comprehensive group of genera which had
been previously recognised by him’ as constituting the family Goniasteridew. The
name Goniasteridz was not retained for any of the groups or families into which
that incongruous series of genera was divided. Previous to 1875, even the generic
name of Goniaster had been very loosely and incorrectly applied. The vaguest
notions as to the limits or characters of the genus seem to have been held. The
mere form of the body, and the applicability of the significant name, irrespective
of structural details, appear to have alone determined the reference of a large
number of the species which have at different times borne the generic name of
Goniaster.

1 «Nouv. Archives Mus. Hist. Nat.,’ 2e série, 1884, t. vi, p. 165.
2 « Révis. Stell. Mus.,’ p. 25 (‘ Archives de Zool. expér.,’ 1875, t. iv, p. 289).

4 FOSSIL ASTEROIDEA.

M. Perrier showed that none of the recent species ranked as Goniaster pre-
vious to 1875 had any right to be so called. He consequently employed the name
in a new and restricted sense, taking the Asterias obtusangula of Lamarck as the
type of the genus. No other species is at present known which can be regarded
as congeneric with that form.

A large number of fossil Starfishes have been named as species of Goniaster,
but none of them present characters which justify their reference to that genus
in its new sense, and none of them invalidate the course taken by Prof. Perrier.
It will therefore be unnecessary in the following pages to discuss in each case
separately the reasons for removing the large number of species which have
from time to time been ranked under the name of Goniaster.

Subfamily—PENTAGONASTERINE, Sladen.

PrenTAGONASTERIN®, Sladen. Zool. Chall. Exped., part li, Report on the Aste-
roidea, 1889, pp. xxxi, 262.

Pentagonasteride with the abactinal area paved with rounded, polygonal, or
paxilliform plates. Granules or spinelets when present co-ordinated.

Genus—CALLIDERMA, Gray, 1847.

CaLLIpERMA, Gray. Proc. Zool. Soc. Lond., part xv, 1847, p. 76; Ann. and Mag.
Nat. Hist., 1847, vol. xx, p. 198; Synop. Spec. Starf.
Brit. Mus., 1866, p. 7.

Marginal contour stellato-pentagonal. General form depressed. Disk large
and flat. Rays moderately elongated and tapering. Marginal plates forming a
broad border to the disk, and may be united along the median abactinal line of
the ray throughout [or, in some fossil species, may be separated by one or more
series of medio-radial plates, at least at the base of the ray]. The marginal
plates of both series are granulated. [In recent species the supero-marginal
plates bear some small papilliform spinelets on the margin where the abactinal
and lateral surfaces of the plate unite; and the infero-marginal plates have a
number of similar, but larger and more fully developed, spinelets irregularly dis-
tributed amongst the granulation of the actinal surface.] Abactinal area of the
disk covered with small and regularly arranged plates, hexagonal in the radial
areas, bearing co-ordinated granules, and some with a larger, globular, central,

CALLIDERMA. 5

tubercle-like granule. Actinal interradial areas large, confined to the disk.
Actinal intermediate plates large, covered with granules [and in the recent species
bearing one or occasionally two compressed acute papilliform spinelets]. Arma-
ture of the adambulacral plates arranged in longitudinal series.

This genus was established by Dr. J. E. Gray for the reception of a recent
Starfish, the type of which is preserved in the British Museum. It was described
under the name of Calliderma Hmma. In his remarks which follow the diagnosis,
Dr. Gray observes' that “there is a fossil species, very like the one here
described, found in the chalk, and figured in Mr. Dixon’s work on the fossils of
Worthing, which I propose to call Calliderma Dixonii.” I have not been able to
trace which of the fossil species is here referred to, but that is a circumstance of
no great importance, as the forms figured in Mr. Dixon’s work on ‘ The Geology
of Sussex’ were described and named by the late Prof. Edward Forbes. It is inter-
esting, however, to note that the resemblance of some of the Cretaceous forms to
the genus Calliderma had actually been observed by the author of that genus.

Thanks to the careful study and critical insight of Mr. J. Walter Gregory, of
the British Museum, a number of the examples which now form part of the
National Collection have been correctly, as I think, referred to the genus Calli-
derma, and already bear that name upon the manuscript labels attached by him to
the specimens.

There are, however, some differences between the fossil forms and the recent
type. The most notable perhaps being the character presented by the spinulation
of the marginal, the abactinal, and the actinal intermediate plates in the recent
species, as compared with the fossil examples, whose state of preservation does not
permit of our positively asserting whether the same character was present in their
case or not. I am inclined to think that this uncertainty does not necessarily
invalidate the reference of the fossil forms to the genus, and I consider it highly
probable that species might exist which did not bear incipient spinelets on the
plates in which they are found in the solitary existing species with which we are
acquainted. The peculiar pits found upon the plates in some of the fossil examples
may indicate the former presence of these spinelets, although, for my own part, I
am more disposed to believe that in the majority of cases the depressions in
question are structures associated with a pedicellarian apparatus. (See, for ex-
ample, Pl. I, figs. la, 1b, 1c, 1d; Pl. IU, fig. 3a; Pl. V, figs. 2a, 26, 2 d.)
In other cases it is certain that little spinelets did exist, as in the tip of the ray
shown in PI. VIII, fig. 2 a; also, but perhaps more doubtfully, in Pl. VII, figs.
4a, 4c.

1 «Proc. Zool. Soc. Lond.,’ part xv, 1847, p. 76; ‘Synop. Spec. Starf. Brit. Mus.,’ London,
1866, p. 7.

6 FOSSIL ASTEROIDEA.

Another point of difference is to be found in some of the fossil forms which are
referred in the following pages to the genus Calliderma ; and this consists in the
separation of the supero-marginal plates, at least at the base of the ray, by one or
more series of medio-radial plates. It is a character whose importance is not to be
under-estimated, but too little as yet is known of the morphological plasticity of
the genus to justify in my opinion the separation of the forms on this ground
alone. I prefer, therefore, to regard this extension of the abactinal plating as a
transitional character, and I believe that this opinion is warranted by the range of
plasticity observed in other genera of recent Asteroidea.

1. CatiiperMa Smitat1#, Forbes, sp. Pl. I, figs. 1 a—1f; Pl. VIII, figs, 2 a—2c.

GonrastEr (AsrRoconium) Smiruit, Forbes, 1848. Memoirs of the Geological
Survey of Great Britain, vol. i,
p- 474.

— _ _ — 1850. In Dixon’s Geology and
Fossils of the Tertiary and Cre-
taceous Formations of Sussex,
London, 4to., p. 334, pl. xxii,
figs. 1 and 2.

— — Smitutaz, Morris, 1854. Catalogue of British Fossils,
2nd ed., p. 80.

ASTROGONIUM SMITHII, Dujardin and Hupé, 1862. Hist. Nat. Zooph.
Echin. (Suites 4 Buffon), p. 399.

GoNIASTER SMITHI, Quenstedt, 1876. Petrefactenkunde Deutsch-
lands, I. Abthl., Bd. iv, p. 64.

_ (AsrroGgoniuM) SmirHta, Forbes, 1878. In Dixon’s Geology of Sussex
(new edition, Jones), p. 367,
pl. xxv, figs. 1, 2, 2a.

Body of large size. General form depressed. Abactinal area probably capable
of slight inflation, and more or less flexible: a slight carination being present in
the radial abactinal regions. Actinal surface flat. Marginal contour stellato-
pentagonal, the major radius measuring rather more than twice the minor radius.
Rays broad at the base and tapering gradually to the extremity. Interbrachial
arcs well rounded and forming a regular curve. Margin thick, with a well-defined
channel traversing the line of junction of the supero-marginal and infero-marginal
series of plates, formed by the tumid character of the marginal surface of both
series of plates.

The infero-marginal plates are about twenty or twenty-one in number, counting

CALLIDERMA SMITHLA. ‘j

from the median interradial line to the extremity. ‘They form a broad conspicuous
border to the actinal area, the breadth of which diminishes gradually from the
median interradial line to the extremity. The largest infero-marginal plates near
the median interradial line measure 9 mm. in breadth, and 4 mm. in length; the
length increases a little between this point and the base of the ray, where it is
again4 mm. ‘The breadth decreases step by step from the median interradial line,
and at the base of the ray is less than 4 mm., and further out the breadth of the
plates is less than the length. The height of the infero-marginal plates as seen in
the margin is greater than the length of the plate, the proportions near the
median interradial line being as 3: 2 approximately. The proportion of the
height decreases at the extremity of the ray. The infero-marginal plates are
slightly convex on their actinal surface and distinctly tumid on their marginal
surface. The whole superficies is covered with small, hexagonal, closely-placed
punctations, upon which granules were previously borne, probably uniform in size
and compactly placed. On a number of the plates are one or more subcircular or
irregular shallow concavities, quite irregular in size, position, and occurrence,
which I believe to have been caused by the presence of a pedicellarian apparatus,
perhaps the cavities of ordinary foraminate pedicellariz enlarged by weathering.
These are seen in PI. I, figs. 1 a, 1b, 1¢,1d. I scarcely think that they are the
marks left by tubercles or enlarged granules. In the example, however, figured
on Pl. VIII, fig. 2 a, small spinelets were undoubtedly present.

The adambulacral plates are broader than long, their dimensions at a short
distance from the mouth being 3°25 mm. broad and 1°75 mm. long. They bear
upon their surface four or five ridges, parallel or sub-parallel to the ambu-
lacral furrow, each with five or six articulatory elevations upon which spines had
previously been borne. A number of these spines are still preserved, irregularly
strewed over the surface of the plates. They are short, tolerably robust, slightly
flattened, slightly tapering and abruptly rounded at the tip. The longest
measures about 1°5 mm. in length, or a little more; their surface is finely striate,
in fact so fine that the character is perhaps mainly due to the effect of weathering
upon the structural texture of the spine.

The actinal interradial areas are large and are covered with a great number of
small, regular, quadrangular or rhomboid intermediate plates, which are arranged
in series parallel to the ambulacral furrow, and form a compact tessellated pave-
ment. The average size of the plates is about 2 mm. in diameter, but the plates
of the series adjacent to the adambulacral plates are somewhat broader, and the
plates near to the infero-marginal plates become smaller and irregular. The plates
extend at the base of the ray to about the eighth infero-marginal plate, counting
from the median interradial line. The surface of the plates is covered with large,
rather widely spaced, hexagonal punctations—the marking left by the granules

8 FOSSIL ASTEROIDEA.

previously borne upon the plates, which appear to have been rather large and
uniform.

Very few, if any, pedicallariz: appear to have been borne on the actinal inter-
mediate plates.

The character of the mouth-plates and their armature cannot be made out in
any of the examples I have examined.

The supero-marginal plates are only exposed in the marginal view of the
example under description. Their height is seen to be less than that of the
infero-marginal plates, and they are also rather smaller both in length and breadth ;
about twenty-two appear to be present between the median interradial line and the
extremity of the ray. In their general character and ornamentation they resemble
closely those of the infero-marginal series.

From another example, also contained in the British-Museum Collection, and
bearing the registration mark “ EH. 2037,” in which the abactinal surface is shown,
the following details are supplemented.

The abactinal area is covered with small, regular, hexagonal plates or paxillar
tabula, which are slightly rounded superficially, and a little bevelled on the
margin of the tabulum. The surface of the tabulum is covered with punctations
or marks left by the granules originally borne on the plate, and here and there
small pedicellarian foramina may be seen, usually near the margin of the plate.
The plates (or paxille) of the median radial series are broader than any of the
others, the largest measuring about 2 mm. in breadth, and a little more than 1
mm. in length. The other tabula are true hexagons, measuring about 1 mm. in
diameter or a trifle more, and they are arranged in longitudinal series parallel to
the median radial series. Hight or nine series are present on each side of the
median radial series on the disk. Opposite the eighth supero-marginal plate,
counting from the median interradial line, only the median radial series and one
lateral series of tabula on each side of it are present. The median radial series
then extends alone and is present at the fourteenth plate, where the ray is broken
in the example under description—and looks like continuing further—pro-
bably reaching to the extremity of the ray.

Dimensions.—In the type specimen (figured on Pl. I) the major radius is
98 mm., the minor radius 48 mm., the thickness of the margin from 9 to 10 mm.
The breadth of a ray between the fifth and sixth infero-marginal plates measures
about 28 mm.

Locality and Stratigraphical Position—The specimen figured on PI. I is from
the Lower or Grey Chalk at Burham, in Kent. The species has also been ob-
tained from the Upper Chalk at Brighton (Coll. Brit. Mus.) ; and it is sometimes

CALLIDERMA MOSAICUM. 9

found in ferric sulphide at Amberley Pit, Sussex. The fragment figured on
Pl. VIII, figs. 2 a, 2b, 2c, is from the Lower Chalk of Dover.

History.—The type specimen, from which this species was originally described
by the late Edward Forbes in 1848, formed part of the collection of Mrs. Smith, of
Tunbridge Wells. It is now preserved in the British Museum. It was first
figured in Dixon’s ‘Geology and Fossils of the Tertiary and Cretaceous Forma-
tions of Sussex,’ London, 1850. Pl. I of the present memoir is a faithful draw-
ing of the same beautiful specimen.

2. CattierMa Mosaicum, Forbes, sp. Pl. V, figs. 2a—2e; Pl. VI, figs. 1 and 2
a, b,c; Pl. VII, figs. 4 a, d, c.

Gontaster (AsTRogonIuM) Mosarcus, Forbes, 1848. Memoirs of the Geological
Survey of Great Britain, vol. ii,

p- 475.
= _ mosaicus, Forbes, 1850. In Dixon’s Geology and
Fossils of the Tertiary and Cre-
taceous Formations of Sussex,
London, 4to., p, 334, pl. xxiv,

fig. 26.
— _ mosaicus, Morris, 1854. Catalogue of British Fossils,
2nd ed., p. 80.
ASTROGONIUM MOSAICUM, Dujardin and Hupé, 1862. Hist. Nat.
Zooph. Echin. (Suites 4 Buffon),
p. 399.

Gontaster (AsTRoGoNIUM) Mosaicus, Forbes, 1878. In Dixon’s Geology of Sussex,
(new edition, Jones), p. 367,
pl. xxvii, fig. 26.

GONIASTER MOSAICUS, Etheridge, 1885. In Phillips’s Manual of
Geology (new edition), part ii,
by R. Etheridge, p. 560.

Body of large size. Disk large. Rays narrow at the base and well produced,
General form depressed and thin. Abactinal area probably capable of slight
inflation, and more or less flexible; some carination present in the radial abactinal
regions. Actinal surface flat. Marginal contour stellato-pentagonal, the major
radius measuring more than twice and a half the minor radius. Rays narrow,
the supero-marginal plates being united in the median radial line. Interbrachial
arcs wide and with their curvature more or less flattened, which gives a distinctly
pentagonal character to the disk. Margin rather thin, and with the lateral wall
perpendicular.

9

~

10 FOSSIL ASTEROIDEA.

The supero-marginal plates are about twenty-eight in number, counting from
the median’ interradial line to the extremity. (This number is taken from the
fragment figured on Pl. VII, fig. 4 a; in the larger example drawn on PI. V,
fig. 2 a, twenty-two may be counted up to the place where the ray is broken.)
They form a well-defined, conspicuous border, but the breadth of this is distinctly
less in proportion to the size of the disk when compared with the breadth of the
marginal plates in Calliderma Smithiez. The largest supero-marginal plates in the
specimen figured on PI. V, fig. 2 a, near the median interradial line, measure 5°25 mm.
in breadth and 3°25 mm. in length. The breadth diminishes very slightly as
the plates approach the base of the ray, but from that part outward the length of
the plates becomes much reduced—the breadth remaining the greater dimension
throughout the ray.

The supero-marginal plates are comparatively flat on the abactinal surface and
only slightly depressed along their margins of juncture. The general surface of
the whole series has the character of sloping at a small angle to the margin of the
disk, to which it gives a slightly bevelled appearance. The marginal surface of
the plate is almost vertical, the junction of the abactinal and marginal surfaces is
well rounded but not tumid, and there is very slight, if any, convexity on the
marginal surface, at least along the disk. The height of the plates as seen in the
margin is only a little greater than the length, and the diminution in height is
only very trifling as the plates proceed along the ray. The whole superficies of
the plates is covered with small hexagonal punctations upon which granules were
previously borne. Small foraminate pedicellariz are occasionally present here
and there upon the plates; the foramen is small and oval, and is surrounded by a
definite margin or lip. Sometimes more than one are present on one plate.
The example figured on PI. VII, fig. 4 a, is remarkable for the presence of the
prominent teat-like eminences, in the centre of which the pedicellarian foramen is
situated. These eminences at first sight look like tubercles for the articulation of
spines (see Pl. VII, figs. 4a, 4c). A similar character is also seen in the example
drawn on PI. V, fig. 2 a, but is less strongly marked (see fig. 2 d).

The abactinal area of the disk is covered with small, regular, hexagonal and
tetragonal plates or paxillar tabula; those in the radial areas being regularly
hexagonal and larger than those in the intermediate regions, which are rhomboid,
and all diminish in size as they approach the margin. The abactinal plates or
paxille do not appear to extend beyond the twelfth supero-marginal plate, counting
from the median interradial line ; the supero-marginal plates of the two sides of
the ray meeting in the median radial line beyond this point. The plates or paxille
of the median radial series are larger and broader than any of the others; they
are succeeded on each side by five or six longitudinal series of hexagonal plates,
those of the second or third series from the median series measuring about 1°5 mm.

CALLIDERMA MOSAICUM. ll

in diameter. The remaining plates which occupy the intermediate areas are
tetragonal or rhomboid. All the plates have their surface marked with rather
widely-spaced punctations—the impressions of the granules previously present.
Small foraminate pedicellarie are also frequently present here and there, usually
near the margin of the plate.

The madreporiform body is flat, distinct, and polygonal in outline; it is situated
near the centre of the disk. Its surface is marked by fine straight stria, which
radiate regularly centrifugally from the centre to the margin (see PI. V, fig. 2 e).

Other specimens show that the infero-marginal plates in this species are more
nearly subequal to the supero-marginal series than in Calliderma Smithiz, that the
actinal intermediate plates are relatively larger than in that species and a good
deal larger than the abactinal paxillar plates or tabula. The actinal intermediate
plates originally bore granules only, judging from the character of the punctations
with which their surface is ornamented. A fragment belonging to the British
Museum Collection (which bears the register number ‘‘ K 373’), in which the spines
that formed the armature of the adambulacral plates are preserved, indicates that
these spines are smaller, shorter, and perhaps more numerous than in Calliderma
Smithiz.

In the example drawn on Pl. VI, fig. 2 a, the supero-marginal plates are pre-
served, but the whole of the abactinal plating has been removed, leaving exposed
the inner surface of some of the actinal intermediate plates and the adambulacral
plates. Magnified details of these plates are given, and they represent the cha-
racters of the structures preserved better than any verbal description.

Dimensions —The large example figured on Pl. V, fig. 2 a, has the following
measurements :—Major radius 82 + mm. (all the rays are broken and imperfect,
and the full dimensions cannot therefore be given); minor radius 36 mm.; thick-
ness of the margin about 8mm. Breadth of a ray between the eighth and ninth
supero-marginal plates about 15 mm., or a trifle more.

Locality and Stratigraphical Position.—The example figured on Pl. V, fig. 2 a,
is labelled from the Lower Chalk, but the locality is not recorded. It formed
part of one of the old collections preserved in the British Museum. Other
examples in the British Museum are from the Grey Chalk or Chalk Marl of Dover,
from the Lower Chalk of Glynde in Sussex, and from the Lower Chalk of
Amberley Pit, Arundel. There is also a magnificent specimen in the Museum of
Practical Geology, Jermyn Street, from the Lower Chalk of Dover.

12 FOSSIL ASTEROIDEA.

3. CALLIpERMA LatuM, Forbes, sp. Pl. II, figs. la—le, 2a—2d; Pl. III,
figs. la—le, 2a, 2b, 3a, 30.

GonrasTER (AsTROGONIUM) LATUS, Forbes, 1848. Memoirs of the Geological
Survey of Great Britain, vol. ii,
p- 474.

— — — — 1850. in Dixon’s Geology and
Fossils of the Tertiary and Creta-
ceous Formations of Snssex,
London, 4to., p. 333, pl. xxiii,
figs. 4, 5.

— — — Morris, 1854, Catalogue of British Fossils,
2nd ed., p. 80.

ASTROGONIUM LATUM, Dujardin and Hupé,1862. Hist. Nat. Zooph.
Echin. (Suites 4 Buffon), p. 399.

GontastER (AsTROGONIUM) LATUS, Forbes, 1878. In Dixon’s Geology of Sussex
(new edition, Jones), p. 367,
pl. xxvi, figs. 4, 5.

Body of large or moderate size. General form depressed. Abactinal surface
probably capable of some degree of inflation. Actinal surface flat. Marginal
contour stellato-pentagonal, the major radius probably not exceeding the minor
radius by more than one half. Rays narrow at the base, short, not greatly pro-
duced, and probably tapered to a pointed extremity. Interbrachial arcs very
wide and flattened, which gives a strongly marked pentagonal outline to the disk.
Margin of uniform thickness.

The infero-marginal plates are more than sixteen in number, counting from the
median interradial line to the extremity (the tip of the ray being broken in all the
specimens examined). They forma remarkably broad margin to the actinal area of
the disk, which diminishes rather rapidly in width at the base of the rays, and then
slightly to the extremity. The largest infero-marginal plates near the median
interradial line measure about 13 mm. in breadth and 4°5 mm. in length. The
length is nearly uniform throughout, or at any rate till well out on the free part
of the ray; but the breadth diminishes until the plates at the base of the ray are
6°5 mm., and the diminution proceeds to a certain extent along the ray. The
infero-marginal plates are slightly convex along their line of breadth, by which
means the separate plates are distinctly marked. They are well rounded at the
junction of the actinal and lateral surfaces, and are slightly tumid in the margin.
The outline of their inner or adcentral edge is also rounded. The height of the

CALLIDERMA LATUM. 13

infero-marginal plates as seen in the margin is a little greater than their length.
The height of the supero-marginal plates is, however, somewhat greater.

The whole superficies of the plates is covered with circular punctations of
irregular size rather than hexagons, as in the other forms, and the irregularity
caused by the presence of larger punctations here and there is remarkable. This
character seems to indicate the former presence of an irregular-sized granulation.

The supero-marginal plates are similar in character to the infero-marginal
series, but the large irregular punctations are larger and more numerous.

The adambulacral plates are broader than long, and they bear upon their
surface five or six ridges parallel or subparallel to the ambulacral furrow, each with
prominent well-defined granulations or elevations, all uniform and closely placed,
upon which the adambulacral armature of spines was previously borne (see Pl. I,
fig. 1d; Pl. III, fig. 2d).

The actinal interradial areas are large, and are covered with comparatively
large polygonal and rhomboid intermediate plates, which are arranged in series
parallel to the ambulacral furrow, and originally formed a compact tessellated
pavement. Ina number of the fossils of this species, however, these plates are
often separated and displaced, which leads to the inference that in life the plates
were not so intimately connected as in other species, and that membrane or con-
nective tissue was more largely developed. The one or more series of plates
adjacent to the adambulacral plates are much larger than the others, and none of
the intermediate plates extend beyond the base of the ray. The surface of the
plates is covered with large, irregular, and rather deeply sunken pits, the
character of which leads to the inference that the granulation originally present
was also somewhat irregular in size and coarse in character (see PI. II, fig. le;
Pl. TH, fig. 12).

In some examples (notably in that figured on PI. ITI, fig. 2 @) small oval fora-
minate pedicellariz, distinctly lipped at the margin of the foramen, are present on
the actinal intermediate plates.

The mouth-plates are elongate, about three times as long as broad, triangular
in shape, with the two outer sides subequal. Their surface is covered with large,
coarse, irregular, tuberculose elevations (see Pl. I, fig. 1 e), which suggest the
inference that the armature of the mouth-plates consisted of large, irregular,
papilliform granules.

In some examples a portion of the actinal floor has been removed, exposing the
inner surface of the abactinal floor. In these cases the stellate base of the abac-
tinal plates or paxillz are seen (see PI. I, figs. 2a, 2d; Pl. III, figs. 34,36). It
will be noticed that there is a difference in the form of the stellate bases in these
examples, which may indicate a specific or varietal difference, but I do not feel
justified from this character alone in recognising either of these fragments as the

14 FOSSIL ASTEROIDEA.

type of a distinct species. More material is needed before such a step would be

warranted.

Dimensions. —The large example figured on PI. II, fig. 1 a, has a major radial
measurement of from 80 to 95 mm. or more, with a minor radius of about 52 mm.
The breadth of the ray between the sixth and seventh infero-marginal plates,
counting from the median interradial line, is about 17 or 18 mm.

Locality and Stratigraphical Position.—This species appears to be confined to
the Lower Chalk. Examples have been collected from Washington, Amberley,
Southerham, and Glynde, in Sussex. Also from the Lower Chalk of Folkestone,
and the Chalk Marl of Dover.

History.—Two examples of this species were first figured by Forbes in Dixon’s
‘Geology and Fossils of the Tertiary and Cretaceous Formations of Sussex,’
London, 1850, pl. xxiii, figs. 4 and 5. Both these specimens are now preserved
in the British Museum. One example, which is from Amberley, is drawn on
Pl. II, fig. 1a. The other, which is from Washington, is accurately represented
by fig. 3 a of the same plate.

Variations.—In addition to the difference noted above in the form of the
stellate bases of the abactinal plates or paxille, other minor differences may be
observed. In some examples the breadth of the border formed by the marginal
plates on the disk area is not relatively so great as in other examples, and the
proportions of length to breadth, as well as the amount of tumidity of the
component plates, are subject to variation. In some examples, again, the irregu-
larity in the granulation of the marginal plates, arising from the former presence
of coarser granules interspersed amongst the average granulation, is more marked
than in others. These differences will be more readily noticed by turning to the
figures given on PI. II and Pl. III than by a lengthy verbal description. Some of
the examples come from different beds and different localities—circumstances
which I consider to be sufficient to account for the variation.

Genus—NYMPHASTER, Sladen, 1885.

Nympnasrer, Sladen. In Narr. Chall. Exp., 1885, vol. i, p. 612; Zool. Chall.
Exped., part li, Report on the Asteroidea, 1889, p. 294.

Disk large and flat. Rays elongate, slender, tapering, and almost square in
section. Marginal plates forming a broad border to the disk, and either united

NYMPHASTER COOMBII. 15

along the median abactinal line of the ray throughout, or separated only by a single
series of medio-radial plates. The marginal plates of both series are granulated,
and bear no spines (normally, but occasional incipient spinelets may be present).
Abactinal area of the disk covered with large and regularly arranged plates, those
in the radial areas well separated, usually hexagonal, more or less tabulate and
paxilliform, and frequently bearing an excavate or entrenched pedicellaria.
Actinal interradial areas large, confined to the disk. Actinal intermediate plates
well defined, covered with uniform granules, and occasionally bearing pedicellariz.
Armature of the adambulacral plates arranged in longitudinal series. Madre-
poriform body exposed and situated within one third of the distance from the
centre to the margin. Large entrenched pedicellariz are frequently present on
the marginal plates in some species.

This genus includes a number of recent species brought to light by the deep-
sea explorations of late years. It has been found in the Atlantic, the Pacific, and
the Eastern Archipelago. The Atlantic species pass into the abyssal zone, but
those inhabiting the Pacific and Eastern Archipelago do not, so far as at present
known, extend beyond the continental zone, or in other words they live in depths
of less than 500 fathoms.

The structure and character of the Cretaceous species described in the
following pages, so far as they can be made out from the fragmentary condition
of the fossils, appear to me to warrant their inclusion in the genus Nymphaster.

1. Nympnaster Coomsn, Forbes, sp. Pl. VII, figs. 1—3; Pl. VIII, figs. la, 1.

Gontaster (AsrrogonruM) CoomBi, Forbes, 1848. Memoirs of the Geological

Survey of Great Britain, vol. 1,
p. 474.

= = —_ — 1850. In Dixon’s Geology and
Fossils of the Tertiary and Cre-
taceous Formations of Sussex,
London, 4to., p. 334, pl. xxiii,
fig. 6.

— — — Morris, 1854. Catalogue of British Fos-
sils, 2nd ed., p. 80.

Asrroconrum Comsn, Dujardin and Hupé, 1862. Hist. Nat.
‘ Zooph. Echin. (Suites 4 Buffon),
p. 399,

Gon1asteR (AsTROGONIUM) CoomBil, Forbes, 1878. In Dixon’s Geology of
Sussex (new edition, Jones),
p- 367, pl. xxvi, fig. 6.

16 FOSSIL ASTEROIDEA.

Body of medium size. Disk moderately large. Rays well produced, rather
broad at the base and tapering to the extremity. General form depressed and
thin. Marginal contour stellato-pentagonal, the major radius measuring more
than twice and a half the minor radius. Marginal plates broad, the supero-
marginal series of the two sides of the ray meeting in the median radial line.
Interbrachial arcs deeply indented and well rounded. Margin rather thin.

The infero-marginal plates are more than fifteen in number, counting from the
interradial line to the extremity. They form a broad conspicuous border to the
actinal area, which is relatively broad in proportion to the size of the disk. The
largest infero-marginal plates near the median interradial line measure about
5°5 mm. in breadth, and about 2°5 to 2°75 mm. in length. The breadth decreases
slightly from this point as the plates approach the base of the ray, and then
much more rapidly, the plates on the outer part of the ray having the length
considerably in excess of the breadth. The plates are tumid and roundly
bevelled at the lateral edges, but are flatly rounded at the margin of the disk, and
without tumidity there. The whole superficies of the plate is covered with large,
rather deeply depressed, hexagonal punctations, closely placed, which give
somewhat of a honeycomb appearance to the plate (see Pl. VIII, fig. 16). These
are the marks left by the granules previously borne upon the plate. Upona
number of the plates in the example figured on Pl. VIII, fig. 1 a, the granules are
still preserved im situ. They are large and closely placed. The punctations,
and consequently the granules, in this species are coarser than in any of the other
Cretaceous forms known to me. I have not been able to assure myself of the
presence in this example of any pedicellariz on the infero-marginal plates.

The adambulacral plates are broader than long, except on the outer part of the
ray, and their armature appears to have consisted of five or six regular series of
spinelets. This is indicated by the presence upon the surface of the plate of that
number of ridges, running parallel or subparallel to the ambulacral furrow, each
having four or five articulatory elevations and intervening pits upon which
spinelets had previously been borne. The spinelets were probably short, and
similar to those described in Calliderma Smithixw and Calliderma mosaicum, but I
have not found any preserved in specimens which I consider to be undoubted
examples of Nyinphaster Coombii.

Dimensions.—In the type specimen, figured on PI. VIII, fig. 1 a, the major
radius is more than 56 mm., and the minor radius23mm. The breadth of the ray
between the fourth and fifth infero-marginal plates measures about 15 mm.

Locality and Stratigraphical Position.—The specimen figured is from the Lower
Chalk of Balcombe Pit, Amberley. The species has also been obtained from the

NYMPHASTER COOMBII. 17

Lower Chalk of Glynde, Sussex; as well as from the Lower Chalk of Dover and
the Isle of Wight. Other specimens of Nymphaster, as to the reference of which
to N. Coombii I am more or less doubtful, which show certain differences in
structural details, are from the Grey Chalk of Folkestone and Dover, and from
the Lower Chalk of Betchworth. Several examples in the Museum of Practical
Geology, Jermyn Street, are labelled from the “ Upper Chalk,” but I am inclined
to think that their reference to that horizon is more or less doubtful.

History.—The type of this species was found by Mr. G. Coombe at Balcombe
Pit, Amberley, and formed part of Mr. Dixon’s collection. It is now preserved in
the British Museum. It was first figured by Edward Forbes in Dixon’s ‘ Geology
and Fossils of the Tertiary and Cretaceous Formations of Sussex,’ London, 1850,
pl. xxiii, fig. 6. The same specimen is carefully represented on PI. VIII, figs. 1 a,
1 } of this memoir.

Doubtful Examples of this Species—Three specimens are figured on Pl. VII,
which I only place provisionally and with very great doubt under this species. 1
do not, therefore, at present propose to describe them in detail, or to definitely
assign the characters they present as supplementary to those already given as
belonging to Nymphaster Coombii.

1. An example from the Lower Chalk of Betchworth, in which a portion of the
actinal surface is preserved (Pl. VII, figs. 1 a—1e). This specimen shows large
infero-marginal plates somewhat longer in proportion to their breadth than in the
type specimen, and their surface is covered with an extremely fine uniform puncta-
tion. The latter character is altogether unlike that of examples which I consider
to be true forms of Nymphaster Coombii. But from this character alone, which
recent forms show to be one subject to considerable variation, I shrink from
taking any more definite step, at least until further material is available for study.
This example has some of the adambulacral plates and actinal intermediate plates
well preserved. ‘The adambulacral plates (see Pl. VII, fig. 1c) conform to the
description given above. The actinal intermediate plates are rhomboid in form,
and their surface is covered with deep, large, well-spaced pits, which indicate the
former presence of a coarse uniform granulation. These plates are shown on
Pl. VII, fig. le. The margin of this example is quite characteristic of Nym-
phaster Coombii. The infero-marginal plates are seen to be low and more or less
bevelled or sloping towards the margin; whilst the supero-marginal plates are
relatively rather higher and more abruptly bent at the junction of the actinal and
lateral surfaces (see Pl. VII, fig. 1 b).

2. This is a badly preserved specimen from the Grey Chalk of Folkestone, in
which nothing but the supero-marginal plates and the general outline are available

3

18 FOSSIL ASTEROIDEA.

for determination (see Pl. VII, figs. 2 a, 2). The marginal plates resemble in
character those of the specimen just mentioned, and they are like them covered
with a very fine punctation, unlike that of the typical Nymphaster Coombii. There
are also fewer plates in that portion of the interbrachial are which may be said to
belong to the disk than in Nymphaster Coombii, but as the example is smaller, this
may probably be only a question of age ; or it may, like the punctation of this and
the preceding example, be attributed to variation, which I am disposed to consider
a not improbable reason for the differences, when regard is had to the horizon
from which the fossils were obtained, and consequently the changed conditions of
existence in which those Asterids probably lived.

3. This specimen (figured on Pl. VII, figs. 3 a, 3) is from the Lower Chalk
of Glynde, Sussex, and I consider that its reference to Nymphaster Coombii is less
doubtful than that of either of the two preceding examples. The fragment repre-
sents a portion of the abactinal surface. The supero-marginal plates are large,
and are covered with the characteristic coarse punctation of Nymphaster Coombii
(see Pl. VII, fig. 34). The supero-marginal plates of the two sides of a ray
meet in the median radial line from the very base of the ray, distinctly charac-
teristic of the genus Nymphaster. Comparing this example with the typical form
of the species, there appear to be a much smaller number of supero-marginal
plates in the interbrachial arc belonging to the true disk, and on these grounds I
hesitate from accepting it positively as an undoubted example of this species until
further material is forthcoming to throw light upon the amount of plasticity
which may be accredited to this species.

2. NympHaster marcinatus, Sladen. Pl. VIII, figs. 4a, 4 b.

Body of medium size. General form depressed. Marginal contour stellato-
pentagonal. Rays well produced, rather broad at the base, and tapering gradu-
ally to a pointed extremity. Interbrachial arcs deep and rounded, the sweep of
the curve from the tip of one ray to the tip of the neighbouring ray being of a
paraboloid character. Margin rather thin.

The supero-marginal plates form a broad and massive border to the abactinal
area of the disk. There are six plates on each side of the disk counting from the
base of one ray to the base of the neighbouring ray. All the succeeding plates
along the ray meet the corresponding plate of the opposite side of the ray in the
median radial line. The abactinal surface of the ray is thus occupied entirely by
the supero-marginal plates throughout its length.

NYMPHASTER OLIGOPLAX. 19

All the supero-marginal plates are of uniform height, excepting the normal
diminution towards the extremity of the ray ; and all have the breadth greater
than the length. The plates on each side of the median interradial line measure
about 4 mm. in breadth and about 2 mm. in length; and this length is maintained
with very slight diminution until about midway between the base and the extre-
mity of the ray, where the length is 1°75 mm., and the breadth is between 2°75
and 3mm. Sixteen supero-marginal plates are preserved in the longest ray of
the specimen under description, counting from the median interradial line to the
broken extremity. A few plates are apparently missing. Measured in the
margin, the height of the plates is about 2°5 mm.

All the supero-marginal plates are distinctly convex in the direction of their
median line of breadth, by which each plate is very clearly marked out, and a
highly ornate character is given to the species. The plates are also tumid and
well rounded on their marginal surface. The whole surface of the plates is
covered with rather large, widely spaced punctations or pits, which have a pecu-
liarly isolated appearance, unlike that of any other species (see Pl. VIII, fig. 4 ).
I have not detected the presence of any pedicellariz upon this example.

The remains of a few isolated plates are preserved on the abactinal area of the
fossil figured. They are all small and out of position, and are not available for
description.

Dimensions.—The specimen figured on Pl. VIII, fig. 4 a, has a minor radius of
about 12 mm.; and the longest fragment of a ray preserved measures about
35mm. The extremity is wanting. The thickness of the marginis 45mm. The
breadth of the ray at the base between the third and fourth supero-marginal
plates counting from the median interradial line is from 8 to 8°5 mm.

Locality and Stratigraphical Position—The example described is from the
Upper Chalk near Bromley. It is preserved in the British Museum, and bears the
registration number 35,484.

3. Nympnasrer oxicortax, Sladen. PI. VIII, figs. 3a, 3b.
Body of medium size. General form depressed and thin. Marginal contour

stellato-pentagonal. Rays narrow at the base and produced. Interbrachial ares
wide and rounded. Margin thin.

The supero-marginal plates form a broad border to the abactinal area of the

20 FOSSIL ASTEROIDEA.

disk. There are only three (or possibly four) supero-marginal plates between the
median interradial line and the base of the ray—that is to say, about six plates on
each side of the disk. The fourth (or perhaps the fifth) plate counting from the
median interradial line, and all the succeeding plates along the ray, appear
normally to meet the corresponding plate of the opposite side of the ray in the
median radial line. The abactinal surface of the ray is thus occupied entirely by
the supero-marginal plates throughout its length. In one of the rays preserved
there appear, however, to be traces of a few abactinal plates which interfere with
the union of the supero-marginal plates in the median radial line near the base of
the ray. As to how far this is normal I am unable to say.

All the supero-marginal plates are comparatively low and flat. The plates on
each side of the median interradial line are 3°5 mm. in breadth, and from 3 to
3°5 mm. in length, and are thus practically square. Their abactinal surface is
slightly convex; and their height as seen in the margin is less than the length,
and the abactinal surface is gradually bevelled to the inferior margin which abuts
upon the infero-marginal plates. The other plates which form the border of the
disk-area are of the same size and character as those adjacent to the median inter-
radial line. ‘The supero-marginal plates along the ray have the breadth greater
than the length, the fifth plate counting from the median interradial line
measuring about 3°75 mm. in breadth and 2°5 mm. in length. Their character is
similar to that of the plates above described. The surface of the plates is covered
with small well-spaced punctations, and there is a distinct smooth border on the
inner and two lateral margins of each plate on which no punctations or pits are
present.

Large trench-like pedicellariz, which are nearly the length of the plate, are
present in this species; they occur more frequently on the infero-marginal plates
than on those of the superior series ; in fact, only one or two are present on the
latter series of plates in the example under description.

No other portions of this fragment are available for description.

Dimensions.—The fragment figured in PI. VIII, fig. 3 a, has a minor radius of
about 15 mm. The longest portion of the major radius preserved is about
33 mm., and the ray is broken off abruptly. The thickness of the margin is
between 4and 5mm. The breadth of the ray at the base is about 8 mm.

Locality and Stratigraphical Position.—The fragment described, which is,
unfortunately, all that I have seen, is from the Upper Chalk of Bromley. It is

preserved in the British Museum, and bears the registration number 40,178.

Remarks.—The character of the marginal plates, as regards both their form

PYCNASTER ANGUSTATUS. 21

and their ornamentation, as well as the presence of the peculiar pedicellarie,
and indeed the whole facies of this fossil, lend strong support to the presumption
that this species may ultimately need to be placed in a distinct genus, but I do
not feel warranted in taking that step on the basis of such scanty material.

Genus-—PYCNASTER, Sladen.

Disk relatively small and pentagonal. Abactinal surface more or less convex,
and was probably somewhat inflated during life. Margin thick, and highest
in the region of the disk. Rays elongate, narrow, and robust. Marginal plates
forming a broad border to the disk, and united along the median abactinal line of
the ray throughout. The marginal plates are high and very robust, those of the
superior series being prominently convex abactinally in the median line of breadth
and height, which imparts a well-rounded character to the ray. ‘The marginal
plates of both series are finely granulated, and probably bore no spines. Actinal
intermediate plates large, covered with uniform granules. Armature of the
adambulacral plates arranged in longitudinal series. Foraminate pedicellariz
with radiating channels may be present on the marginal plates.

The fragmentary state of the fossils which I have referred to this type
unfortunately does not permit of a complete diagnosis of the genus being drawn
up. The characters above given appear, however, to me to be sufficient to
warrant the recognition of the possessors of them as the representatives of a
distinct genus. The small high disk, the massive convex marginal plates,
and the large actinal intermediate plates, together with the form of the rays,
produce a facies alone sufficient to stamp its individuality, irrespective of other
details of structure.

1. Pyonaster ancustatus, Forbes, sp. Pl. IX, figs. 1 a, 1b.

Gonzaster (AsTROGONIUM) ANGuUSTATUS, Forbes, 1848. Memoirs of the Geolo-
gical Survey of Great Britain,
vol. il, p. 474.
_ — = _ 1850. In Dixon’s Geology and
Fossils of the Tertiary and
Cretaceous Formations of
Sussex, London, 4to., p. 335,
pl. xxiii, fig. 10.

22 FOSSIL ASTEROIDEA.

GontasteR (AsTROGONIUM) ANGusTATUS, Morris, 1854 Catalogue of British
Fossils, 2nd ed., p. 80.

AsTROGONIUM ANGUSTATUM, Dujardin and Hupé, 1862. Hist. Nat.
Zooph.Kchin.(Suitesa Buffon),
p- 399.

GontasTER (AsTROGONIUM) aNneusTaTUS, Forbes, 1878. In Dixon’s Geology of
Sussex (new edition, Jones),
p- 368, pl. xxvi, fig. 10.

Disk of medium size or relatively small and pentagonal. Rays elongate,
narrow, robust, and, though tapering, nearly uniform in breadth throughout.
Marginal contour stellate. Interbrachial ares more or less flattened, which
emphasises the pentagonal outline of the disk. Margin thick and robust, much
highest in the region of the disk. Abactinal surface more or less convex, and was
probably somewhat inflated during life. Actinal intermediate plates very large,
covered with uniform granules.

The supero-marginal plates are thick and massive, and they form a high and
broad border to the disk. There are only three supero-marginal plates between
the median interradial line and the base of the ray—that is to say, six plates on
each side of the disk. The fourth plate counting from the median interradial
line, and all the succeeding plates along the ray, meet the corresponding plate of
the opposite side of the ray in the median radial line. The abactinal surface
of the ray is thus occupied entirely by the supero-marginal plates throughout
its length.

The supero-marginal plates which form the border of the disk are much larger
in the direction of height than any of the others. The plates on each side of the
median interradial line are about 4°75 mm. in breadth as seen on the abactinal sur-
face, and about the same measurement in length. They are convex abactinally,
and well rounded at the junction of the abactinal and lateral surfaces. Measured
in the margin their height is 8 mm., and their lateral surface (which forms the
vertical wall of the margin) is distinctly convex or pulvinate, but to a less degree
than their abactinal surface.

The supero-marginal plates of the ray are not so high as those of the disk,
although their height is greater than their length. The height of the sixth plate
from the median interradial line is about 5mm. Their abactinal and lateral sur-
faces form together a true segment of a circle, and this imparts a well-rounded
character to the ray. The plates are deeply bevelled at their junction with the
adjacent plates, and consequently distinctly pulvinate in the median line of breadth
and height. The surface of the plates is covered with minute punctations, but

PYCNASTER ANGUSTATUS. 23

these are so extremely faint that they are seen with difficulty. They are probably
weather-worn in the example under notice.

The infero-marginal plates, as seen in the direct lateral view of the margin, are
much smaller in height than the supero-marginal series in the type specimen.
The plates which form the margin of the disk are higher than long, the height
being about 5 mm. and the length about 3°25 mm. in those adjacent to the
median interradial line; the succeeding plates on the margin of the disk are each
less in height than the preceding plate, the third or fourth plate, counting from
the median interradial line, having the height and length about equal. The infero-
marginal plates along the ray have the length greater than the height. The
surface of the infero-marginal plates resembles that of the superior series in the
character of its punctation.

Traces of small excavate pedicellariz are present on occasional plates, but
these appear to have been very few in the example under description.

On the abactinal surface of the disk a few isolated and displaced plates are
present. Some of these seem rather thick and tuberculous in character, but the
state of the preservation of this part of the fossil is unfortunately quite unfitted
for description.

There is a fine fragment of this species preserved in the Museum of Practical
Geology, Jermyn Street, from the Upper Chalk of Bromley, which shows part of
the actinal surface. The infero-marginal plates in the disk are very high in this
example, and five of them in an interbrachial are bear a small foraminate pedi-
cellaria. This is situated near the upper margin of the plate, about equidistant
from that margin and the lateral margins of the plate, and consists of a small
round foramen situated in the middle of a very shallow concavity, and with five
or six faint channels radiating from the foramen to the margin of the concavity,
gradually thinning and dying out there. The channels radiate lke the spokes of
a wheel, or a five-rayed star, and produce a facies unlike that of any other pedi-
cellarian apparatus with which I am acquainted. The actinal intermediate plates
are very large, and not more than three series are present. The plates of the
series next to the adambulacral plates are much larger than the others, and are
broader than long, The adambulacral plates are broader than long, and their
surface is marked with three or four ridges parallel to the furrow, upon which
spinelets were previously borne. The furrow series consists of about five
spinelets. A few of these spinelets are preserved, and they are rather short,
cylindrical, and slightly tapering. The mouth-plates are very small and narrow.

Dimensions.—The specimen figured on PI. IX, fig. 1 a, has a minor radius of
about 23 mm. The longest portion of a major radius preserved is 53 mm. ; the
ray is broken abruptly, and there is very slight diminution in the breadth at the

24. FOSSIL ASTEROIDEA.

broken extremity as compared with the breadth at the base ; there would appear to
be every indication that only a small part of the ray is preserved.

The thickness of the margin at the median interradial line is 13 mm., and
at the base of the ray 85 mm. The breadth of the ray at the base is about
12 mm.

Locality and Stratigraphical Position—The example above described, which has
been drawn on PI. IX, fig. 1a, was obtained from the Upper Chalk in Kent, but
unfortunately the exact locality is unknown. It is preserved in the British
Museum. A fine fragment preserved in the Museum of Practical Geology,
Jermyn Street, was obtained from the Upper Chalk of Bromley. The species has
also been found in the Upper Chalk of Sussex.

History.—The type of this species was first described by Forbes under the
name of Goniaster (Astrogoniwm) angustatus, and was afterwards figured by him in
Dixon’s ‘Geology and Fossils of the Tertiary and Cretaceous Formations of
Sussex,’ London, 1850, pl. xxiii, fig. 10. That illustration does not, however,
give a good idea of the facies of the species.

Genus—PENTAGONASTER, Linck, 1733.

PENTAGONASTER, Linck. De Stellis marinis, 1733, p. 20.
— Schiilze. Betrachtung der versteinerten Seesterne und ihrer

Theile, Warschau u. Dresden, 1760, p. 50.

GontastTer (pars), ZL. Agassiz. Prod. Mon. Radiaires, Mém. Soc. Sci. Nat.
Neuchatel, 1835, t. i, p. 191.

Asrrogontum (pars), Miiller and Troschel. System der Asteriden, 1842, p. 52.

Gontoprscus (pars), Miiller and Troschel. Ibid., 1842, p. 57.

Hosta (pars), Gray. Ann. and Mag. Nat. Hist., 1840, vol. vi, p. 279.

Tosta, Gray. Ibid., 1840, vol. vi, p. 281.

Body depressed and pentagonal in contour, or with the rays slightly produced.
Marginal plates smooth or granular, ordinarily few in number. Supero-marginal
plates form a broad border to the disk, and, when the ray is produced, are sepa-
rated throughout by abactinal plates. Abactinal area covered with rounded or
polygonal plates, which may either be smooth or bear co-ordinated granules.
Actinal intermediate plates and infero-marginal plates smooth or granulose, devoid
of prominent spinelets.

Much diversity of opinion has existed, unnecessarily it seems to me, as to the

PENTAGONASTER LUNATUS. 25

character and limits of this genus. Two species were originally referred to
Pentagonaster by its founder. The type of one of these is now lost, and its identi-
fication rests only on surmise. The second species, however, Pentagonaster semi-
lunatus, is a well-known and widely distributed recent form, about which there is
no doubt. I therefore consider that this form has every claim to be regarded as
the type of the genus. The existing species of Pentagonaster are found in the
Atlantic, the Pacific, the Indian and the Southern Oceans, and in the Hastern
Archipelago; and the bathymetrical range of the genus extends from 20 to 1500
fathoms or more.

1. Pentaconaster Lunatus, Woodward, sp. Pl. IV, figs. 1 a—1 c.

AsTeRIAs Lunatus, Woodward, 1833. An Outline of the Geology of Norfolk,
p- 52, pl. v, fig. 1.
Tosa LUNATA, Morris, 1843. Catalogue of British Fossils, p. 60.
_ _ Bronn, 1848. Index Palwontologicus, Nomenclator, p. 1274.

Body of medium size. General form depressed. Abactinal and actinal areas
flat. Marginal contour stellato-pentagonal, the major radius measuring nearly
twice the minor radius. Rays short and moderately produced, rather narrow at
the base and tapering to the extremity. Interbrachial ares deeply indented and
well rounded.

The infero-marginal plates are twelve (or more) in number, counting from the
median interradial line to the extremity. They form a broad border to the actinal
area of the disk, the breadth of which diminishes rather rapidly plate by plate as
they recede from the median interradial line. The largest infero-marginal plates
adjacent to the median interradial line measure about 5°25 mm. in breadth and about
3 mm. or a little more in length. The length and breadth decrease as each plate
proceeds outward until about midway on the ray, where these dimensions are
subequal. On the outer part of the ray the length is greater than the breadth.
The infero-marginal plates are distinctly convex on their actinal surface in the
direction of the median line of breadth, and are slightly tumid at the margin.
Their surface is covered with small, uniform, closely placed, and deeply sunken
moniliform punctations, upon which small granules were previously borne, probably
uniform in size and closely placed (see Pl. IV, fig. 1c). I am not aware that
traces of any pedicellariz have been detected on these plates.

The adambulacral plates are small and oblong, and bear on their surface ridges

4

26 FOSSIL ASTEROIDEA.

of alternating granuliform eminences and depressions, upon which the spinelets
constituting the armature of the adambulacral plates were originally borue.

The actinal interradial areas are small, and are covered with regular pentagonal
or rhomboid intermediate plates, which are arranged in series parallel to the
ambulacral furrow, and form a compact, mosaic-like pavement. The actinal inter-
mediate plates are moderately large in relation to the size of the disk. The plates
of the series adjacent to the adambulacral plates are sensibly larger than any of the
others, and the plates of the next series are also larger than those which form the
rest of the pavement. Near the infero-marginal plates the actinal intermediate
plates become small and more or less irregular. The intermediate plates extend
at the base of the ray to about the fifth infero-marginal plate, counting from the
median interradial line. The surface of the plates is covered with rather large,
widely spaced, and deeply sunken punctations, upon which granules were pre-
viously borne, and these would appear to have been comparatively large in size and
uniform (see Pl. IV, fig. 1 5).

From what is visible of the margin.of this example it is seen that the supero-
marginal series of plates are nearly of the same height as the infero-marginal
series, and are similar in structure.

Unfortunately no other portions of this fragment are available for description.

Locality and Stratigraphical Position—The specimen upon which this species
was founded was collected by Mr. Samuel Woodward, from the Upper White
Chalk, near Norwich.

Dimensions.—In the type specimen (figured on Pl. IV, fig. 1 a) the major
radius is about 35 mm., and the minor radius about 18 mm. Breadth of the ray
between the fifth and sixth infero-marginal plates about 10 mm.

History.—The type specimen was figured by Woodward in his ‘ Outline of the
Geology of Norfolk,’ pl. v, fig. 1, and is now preserved in the collection of the
Norfolk and Norwich Museum. It was kindly lent by the committee of that
institution to Dr. Wright for the purpose of this monograph. It has been
carefully drawn on Pl. IV, figs. 1 a—1 ¢. An admirably executed cast of this
specimen is in the British Museum. Iam not at present aware of the existence
of any other examples of this rare form.

Remarks.—The example referred by Forbes to this species, and figured by
him in Dixon’s ‘Geology and Fossils of the Tertiary and Cretaceous Formations
of Sussex,’ London, 1850, pl. xxiii, fig. 9, belongs to a distinct species, which I
have named Pentagonaster megaloplav. A number of other specimens in other

PENTAGONASTER MEGALOPLAX. 27

collections have, following Forbes, been erroneously referred to Pentagonaster
lunatus, which are in reality examples of Pentagonaster megaloplaw. This is
unfortunate, for the latter form has thus become comparatively well known under
the name of Pentagonaster lunatus, a name which they must now cease to bear, as
the real Pentagonaster luwnatus is quite a different form, and there is no doubt
whatever either as to the type (which is preserved in Norwich) or the priority.
The differences between the two species will be further noticed under the descrip-
tion of Pentagonaster megaloplaz.

2. PENTAGONASTER MEGALOPLAX, Sladen. PI. IV, figs. 2—4.

GontAsteR (AsTROGONIUM) LUNATUS, Forbes, 1850. In Dixon’s Geology and
Fossils of the Tertiary and Cre-
taceous Formations of Sussex,
London, 4to., p. 358, pl. xxiii,
fig. 9 (non Asterias lunatus,
Woodward, 1833).

Body of medium size. General form depressed. Abactinal and actinal areas
flat. Marginal contour stellato-pentagonal, the major radius measuring a little
more than once and a half the minor radius. Rays short and not greatly pro-
duced, tapering gradually to the extremity. Interbrachial ares regularly rounded,
curving gradually from the tip of one ray to that of the adjacent ray, which gives
a distinctly lunate character to the disk. Margin of uniform thickness.

The infero-marginal plates are only five or rarely six in number, counting from
the median interradial line to the extremity. They form a very broad border to
the actinal area of the disk in relation to its size, and the breadth is maintained
until near the extremity. The largest infero-marginal plates adjacent to the median
interradial line measure 7 mm. in breadth, and about 6°5 mm. or nearly 7 mm. in
length ; they are consequently almost square. The proportion of breadth dimin-
ishes in the succeeding plates as they recede from the median interradial line.
The infero-marginal plates have a more or less pulvinate appearance actinally,
consequent on being rounded or bevelled at the edges; and they are slightly
tumid in the margin. Their whole actinal surface is covered with large, weli-
spaced, deeply sunken pits, in the centre of which is a slight eminence—a struc-
ture which has almost the character of a granule surrounded by a scrobicule (see
Pl. IV, fig. 2 b). On the surface which stands in the margin the punctations
are fewer and more widely spaced on the upper half of the surface—that is to say,

28 FOSSIL ASTEROIDEA.

the half adjacent to the supero-marginal series (see Pl. IV, fig. 4c). I have not
found any pedicellariz on these plates.

The adambulacral plates are broader than long, except on the outer part of
the ray, and bear on the surface four or five ridges with granuliform eminences,
upon which the spinelets constituting the adambulacral armature were originally
borne. In one well-preserved specimen these small articulatory tubercles are seen
to have each a small microscopic central puncture (see Pl. IV, fig. 3c), but I am
not certain whether this is always present.

The actinal interradial areas are small, and are covered with a comparatively
small number of large pentagonal or tetragonal intermediate plates, which are
arranged in series parallel to the ambulacral furrow, and form a compact
tessellated pavement. The actinal intermediate plates are larger in relation to the
size of the disk than in the species above described. The plates of the series
adjacent to the adambulacral plates, and a few of the plates of the succeeding
series within the angle towards the mouth, are larger than the others. The
intermediate plates do not extend beyond the second, or at most a short distance
along the margin of the third infero-marginal plate, counting from the median
interradial line. The surface of the intermediate plates, excepting a border
round the margin of the plate, is covered with large punctations, which are nearly
confluent, and in some cases almost give the appearance of a coarse reticulate
superficial ornamentation; the border round the margin of the plate above
mentioned is marked with a concentric crenulation (see Pl. IV, fig. 3). Within
the pits are more or less definite elevations. In other examples the reticulate
character is less marked, and the margin of the pit is then prominently lipped,
and the marginal crenulation is not so strongly shown (see Pl. IV, fig. 4 e).

In the marginal view of the type specimen the supero-marginal plates are seen
to be higher than the infero-marginal plates, and that their height is greater than
their length, whereas in the infero-marginal series of plates the height is less than
the length (see Pl. IV, fig. 2c).

In other examples the mouth-plates are preserved. These are rather small,
triangular, and covered with rather large, irregular, tuberculose eminences for
the attachment of the mouth-plate armature.

Variations.—Three examples of this species are figured on Plate IV. ‘These
present a number of minor differences, which will be readily noticed on referring
to the figures.

The example which is shown in fig. 3a has the marginal border of the infero-
marginal plates rather less broad than in the type form, and it is especially
remarkable for the peculiar retiform and crenulated ornamentation of the actinal
intermediate plates already noticed. The disposition of the armature of the

PALAONTOGRAPHICAL SOCIETY.

[INSTITUTED MDCCCXLVILI.

VOLUME FOR 1893. oe AEE PR

LONDON:

MbDCCCXCIIL

A MONOGRAPH

ON THE

Bat iisH LOSS) I

ECHINODERMATA

FROM

THE CRETACEOUS FORMATIONS.

VOLUME SECOND.

Paw ASTHROLDE A.

BY

W. PERCY SLADEN,. F.LS.,. F.GS..&e:,

SECRETARY OF THE LINNEAN SOCIETY.

PART SECOND.

Pages 29—66 ;

Prates IX—XVI.

LONDON:
PRINTED FOR THE PALZHZONTOGRAPHICAL SOCIETY.
1893.

PRINTED BY ADLARD AND SON,
BARTHOLOMEW CLOSE, E.C., AND 20, HANOVER SQUARE, W.

PENTAGONASTER MEGALOPLAX. 29

adambulacral plates is also noteworthy. This seems to indicate the former
presence of a distinct furrow series of spinelets or granules much smaller than
usual, followed by granules or spinelets borne on the outer part of the plate,
more irregularly placed than in the other forms described, and articulated on
punctured eminences.

The example which is represented in fig. 4a also has a narrower marginal
border of infero-marginal plates than the type. The punctation of the infero-
marginal plates is smaller than in the type, and does not present the striking
scrobiculate character noticed in that example. The markings are rather to
be described as lipped pits, and some granules are still in situ. The actinal
intermediate plates do not have the retiform and crenulate ornamentation shown
in the plates belonging to the specimen figured in 3a, but the margins of the
punctations are strongly lipped. The supero-marginal plates are less regular
and much less high than in the type specimen, but they are not perfectly
preserved.

Dimensions.—In the type specimen (figured on Pl. IV, fig. 2a) the major
radius is about 41 mm., and the minor radius 26 mm. Breadth of a ray between
the third and fourth infero-marginal plates, counting from the median interradial
line, about 12 mm. or rather more. Thickness of the margin about 85 mm.

The specimen given in fig. 3a has a major radius of about 39 mm. and a
minor radius of 24 mm.

The specimen given in fig. 4a has a major radius of about 41 mm. and a
minor radius of 25°5 mm. Breadth of the ray between the third and fourth
infero-marginal plates about 11 to 12 mm., or rather more.

Locality and Stratigraphical Position.—The type specimen, which is now pre-
served in the British Museum, is labelled from the ‘‘ Lower Chalk” of ‘ Sussex,”
but is stated by Forbes to have been obtained from the Upper Chalk. Other
examples of the species have been collected from the Upper Chalk of Bromley,
Sittingbourne, Purfleet, Gravesend, Sussex, and Wiltshire. Fine series are pre-

served in the British Museum and in the Museum of Practical Geology, Jermyn
Street.

History.—The specimen which I have taken as the type of this species was
originally referred by Forbes to the Asterias lwnatus of Woodward, and was
figured by him as that species in Dixon’s ‘Geology and Fossils of the Tertiary
and Cretaceous Formations of Sussex’ (pl. xxiii, fig. 9). The same example is
carefully represented on PI. IV, fig. 2 a, of this memoir.

5

30 FOSSIL ASTEROIDEA.

Remarks.—It will be at once seen on referring to the figures that Woodward’s
Pentagonaster lunatus, which is drawn on PI. IV, fig. 1 a, of this Monograph, is a
distinct species. The rays are more produced, and are narrower at the base. The
infero-marginal plates are twice as numerous, the marginal border is less broad,
and the plates are much shorter in proportion to their breadth. Their punctation
is also different. The actinal intermediate plates are smaller in relation to the size
of the actinal interradial areas, and their punctation is different from that which
characterises Pentagonaster megaloplax. The armature of the adambulacral plates
also appears to be more regular in its arrangement.

Under these circumstances I have no hesitation in considering the form under
description a distinct species. I much regret having to impose a new name, as
this form has for a long time been known under the specific name of lunatus ;
the course, however, seems unavoidable, as the actual type of the real Pentago-
naster lunatus described by Woodward is in existence, and there can, in my
opinion, be no question as to its being a different species.

Since the preceding sheet was printed off I have found several specimens in
the British Museum which show the abactinal aspect of the disk. I have little
hesitation in referring these examples to Pentagonaster megaloplax, and a drawing
of one of them is given on PI. XIII, fig. 1a. The infero-marginal plates all show
more or less distinctly the characteristic ‘‘ scrobiculate ’’ or areolated pits already
described. A similar ornamentation also extends upon the supero-marginal
plates, but is confined to the lateral wall which falls in the margin of the disk.
The curvature which unites the abactinal and lateral areas of the plate is more or
less abrupt, and the lateral wall of the disk is consequently vertical and not
rounded, as arule. The abactinal area of the supero-marginal plates is covered
with small, uniform, granular eminences (see Pl. XIII, fig. 1%). Two or three
supero-marginal plates at the extremity of the ray meet the corresponding plates
of the opposite side of the ray in the median radial line, and a rapid diminution
in breadth occurs as they approach the extremity.

Genus—METOPASTER, Sladen.

[Mérwrov = a cheek-piece. |

Body depressed and pentagonal or stellato-pentagonal in contour, the rays
being produced to a very slight degree. Marginal plates covered with well-
spaced uniform punctations, upon which granules were originally borne, and
surrounded by a narrow depressed border with very minute and crowded puncta-

METOPASTER PARKINSONI. 31

tions for the articulation of setw. Supero-marginal plates ordinarily few in
number, and form a broad border to the disk. Ultimate paired supero-marginal
plates the largest of the series. Abactinal area covered with polygonal, and
usually hexagonal, plates (some of which may have stellate bases), and upon the
tabulz are borne small, more or less co-ordinated granules. Infero-marginal
plates more numerous than the supero-marginal series, and decreasing in size as
they approach the extremity of the ray. Their ornamentation similar to that of
the superior series. Actinal intermediate plates small, polygonal, covered with
uniform granules. Armature of the adambulacral plates arranged in longitudinal
lines. Small entrenched pedicellarix may be present occasionally on the plates.

Metopaster differs from Pentagonaster by the large ultimate paired supero-
marginal plates, by the comparatively small number of the supero-marginal plates,
which are also fewer in number than the infero-marginal series, and by the
character of the ornamentation of the marginal plates of both series.

The forms ranked under this genus were all classed by the late Professor
Edward Forbes under Goniodiscus, which he considered to be a sub-genus of
Goniaster. There is, however, no justification whatever in my opinion for regard-
ing any of the Cretaceous starfishes hitherto described as belonging to either the
genus (foniaster or Goniodiscus. The species which may be considered as the types
of each of these genera are existing forms, and no Cretaceous forms agreeing in
structural detail have, so far as I am aware, been discovered. Itis also erroneous
to rank Goniodiscus as a sub-genus of (roniaster. The two genera belong to
different families; and I am in perfect accord with Professor Edmond Perrier as
to the limitation of the two genera. His view appears to me to be perfectly
logical, and to be the result of careful and impartial judgment. I also consider
that the fossil forms under consideration are quite distinct from the recent genus
Astrogoniwm, as limited by me elsewhere.’

1. Meropasrer Parkinson, Forbes, sp. Pl. IX, figs. 2a—2c; Pl. X, figs. 1 a—
5c; Pl. XI, figs. 1a—2ec; Pl. XII,
figs. la—ld; Pl. XVI, figs. 2a, 2b.

PENTAGONASTER REGULARIS, Parkinson, 1811. Organic Remains, vol. iii, p. 3, pl.i,
fig. 3 (non Linck).
Tosi REGULARIS, Morris, 1843. Catalogue of British Fossils, p. 60.

1 Zool. Chall. Exped.,’ part li, “ Report on the Asteroidea,” 1889, p. 285.

32 FOSSIL ASTEROIDEA.

GontastER (Gonropiscus) ParKinsont, Forbes, 1848. Memoirs of the Geological

Survey of Great Britain, vol. ii,
p- 472.

_ — RECTILINEUS, McCoy, 1848. Ann. and Mag. Nat.
Hist., ser. 2, vol. ii, p. 408.

oa — PaRKINSoNI, Forbes, 1850. In Dixon’s Geology and
Fossils of the Tertiary and Cre-
taceous Formations of Sussex,
London, 4to, p. 332, pl. xxi,
figs. 10, 11; pl. xxii, figs. 5—7.

= — — Morris, 1854. Catalogue of British
Fossils, 2nd ed., p. 81.

- — RECTILINEUS, McCoy, 1854. Contrib. Brit. Pal., p. 55.

—_— — — Morris, 1854. Catalogue of British
Fossils, 2nd ed., p. 81.

ASTROGONIUM PaRKINSONI, Dujardin and Hupé, 1862. Hist. Nat.
Zooph. chin. (Suites 4 Buffon),

p- 399.
— RECTILINEUM, Dujardin and Hupé, 1862. Ibid., p. 400.

GontastER (Gonropiscus) ParKrnsont, Forbes, 1878. In Dixon’s Geology of
Sussex (new edition, Jones),
p- 365, pl. xxiv, figs. 10, 11;
pl. xxv, figs. 5—7.

PENTAGONASTER RECTILINEUS, Woods, 1891. Catalogue of the Type
Fossils in the Woodwardian
Museum, Cambridge, p. 36.

Body of medium size. General form depressed. Abactinal surface flat, with
a tendency, however, for the extremity of the rays to be slightly upturned; as
found in the fossil state the area occupied by the abactinal plates is usually at a
lower level than the marginal plates, which leads to the assumption that the
abactinal floor had collapsed or fallen to a certain extent on the death of the
animal. Actinal surface slightly convex. Marginal contour pentagonal, with
slightly lunate sides, the curvature being often flattened at right angles to the
median interradial line. The major radius measures about one-third more than
the minor radius, and frequently less than one-third; the rays are consequently
very feebly produced. Margin thick and well rounded.

The supero-marginal plates are four in number, counting from the median
interradial line to the extremity, or eight from the tip of one ray to the tip of the
adjacent ray, exclusive of the odd terminal or “ ocular’ plate in each case.
They form a broad border to the abactinal area of the disk of uniform breadth
throughout, which measures about 9 mm. at the median interradial line in an
example whose minor radius measures 30 mm. Excepting the ultimate paired
plates all the supero-marginal plates are of equal size, the breadth being about

METOPASTER PARKINSONI. 33

twice the length, the actual measurements in the specimen under notice being,
length 4°75 mm. and breadth 9°5 mm. respectively, i.e. as 1:2. The abactinal
surface of these plates is distinctly convex, with a slight depression along their
margins of juncture, formed by a well-defined bevel along the sides and adcentral
end. The general surface of the whole series is well rounded, the curvature
being regular and uninterrupted between the adcentral margin of the plate and
the margin in the lateral wall adjacent to the infero-marginal plates. The height
of the plates as seen in the margin is a little greater than their length, and there
is no diminution in height as the plates approach the extremity of the ray—in fact,
the ultimate paired plate is not unfrequently higher than the other plates in
consequence of a tendency to become gibbous on its abactinal surface. The
whole superficies of the plates is covered with small, widely spaced, equidistant,
uniform punctations, and there is a depressed border along the margin of the
plate, varying slightly in breadth in different examples, covered with much
smaller and closely crowded punctations, upon which much smaller granules than
those upon the median area of the plate were originally borne. Traces of these
granules may occasionally be found i situ.

The ultimate paired plate is larger than any of the other supero-marginal
plates, and is of a different shape. It is subtriangular in form as seen from
above, and one margin touches the corresponding plate of the adjacent side of the
disk throughout, the junction coinciding with the median radial line of the disk.
The length of this margin of the plate is subequal to or only slightly greater than
the breadth of the preceding marginal plates. In small specimens the subequal
measures are the rule, whilst in larger examples the plate becomes more elongate
and produced in the direction of the prolongation of the ray. When viewed in
the margin of the test the form of the ultimate plate strikingly resembles that of
the carapace of some Coleoptera. The length of the plate in this aspect, measured
from its outer extremity to the margin adjacent to the penultimate plate, is in
small and medium sized specimens about once and a half the length of the other
marginal plates, but in large examples it may be as much as, or even exceed,
twice their length. The surface of the ultimate plate bears a similar ornamenta-
tion to that on the other supero-marginal plates.

The odd terminal plate is very small, appearing externally when denuded of
granules like a truncate cylinder, having a fanciful resemblance to a cannon
projecting from a porthole. This plate seems to be very rarely preserved in situ
in the fossil state. In a remarkably good specimen belonging to the British
Museum Collection (marked ‘“ E 2034’’) (see Pl. XVI, figs. 2a, 26) each of the
terminal plates preserved bears at its outer truncate extremity a single horizon-
tally placed entrenched pedicellaria. Whether this regularly placed pedicellaria is
always present on the odd terminal plate in this species I am unable to say.

34 FOSSIL ASTEROIDEA.

The abactinal area of the disk within the boundary of the marginal plates is
covered with small sub-regular plates or paxillar tabulz, an hexagonal form pre-
dominating especially in the radial regions ; and a small but distinct diminution in
size takes place as the plates approach the margin of the disk. All the plates
have their surface marked with minute, shallow, and closely placed punctations—
the impressions of the attachment of the granules previously present. The
primary basal plates are larger than any of the other abactinal plates, and they
are well shown in several of the drawings illustrative of this species (see Pl. X,
figs. 1, 2a). Occasional plates bear small entrenched pedicellariew, the normal
form consisting of a small, central, hpped foramen with a lateral trench on each
side. It frequently happens, however, that the organ exhibits a more complex
development, and assumes a stellate form in consequence of the presence of addi-
tional trenches—five or six being not an unusual number—which radiate from the
central foramen ; the whole being placed on a small hemispherical elevation, and
producing an appearance shown on Pl. X, fig. 2 d.

‘he madreporiform body is large and subtriangular in outline; and its surface
is sculptured by very fine striations which radiate from the centre to the margin,
with more or less wavy lines here and there (see Pl. X, fig. 2c). The margin
of the plate is surrounded by three large plates, one on the adcentral side of
the madreporite towards which it presents a straight suture; the other two
plates are on the remaining sides of the triangular body, and they have a
concave curve directed towards the madreporite to correspond with the
convexity of its sides. The position of the madreporite is nearer the centre of
the disk than the margin.

The infero-marginal plates are seven in number, counting from the median
interradial line to the extremity—that is to say, there are fourteen for the whole
side of the disk, as against eight in the supero-marginal series. The length of
the three innermost plates on each side of the median interradial line is equal to
that of the superior series, but there are four infero-marginal plates corresponding
to the large ultimate supero-marginal. As seen in the lateral wall of the disk the
height of the infero-marginal plates is less than that of the supero-marginal series.
The breadth of these plates adjacent to the median interradial line on the actinal
surface is 7°5 mm. in an example whose major radius measures 36°5 mm. and the
minor radius 27°5 mm. The breadth of the marginal border rapidly diminishes
towards the extremity of the ray. The surface of the plates is ornamented in a
precisely similar manner to that of the supero-marginal plates. A narrow border
of smaller granulation is also present round the whole margin of the plate,
similar in all respects to that already described in the case of the superior
series of plates.

The adambulacral plates are small, about or nearly twice as broad as long, and

METOPASTER PARKINSONI. 35

their surface is traversed by four or five ridges running parallel to the ambulacral
furrow, with punctures upon which the spinelets composing the armature were
articulated. There were four or five spinelets in each lineal series. ‘The spinelets
are short, their length being about equal to the length of the plate, stumpy, com-
pressed, slightly tapering and rounded at the extremity; and all appear to have
been uniform.

The actinal intermediate plates are rather large for the genus ; those adjacent to
the adambulacral plates are pentagonal in form, but elsewhere they are subhexagonal,
or perhaps more correctly polygonal. ‘The plates are very large on the inner
portion of the area, but diminish greatly in size at the outer margin of the disk
adjacent to the marginal plates. The surface of the intermediate plates is entirely
covered with small, equidistant punctations, upon which a uniform close granulation
was previously attached. Remains of this granulation are still occasionally to be
seen in situ on plates here and there in the example under notice. Hntrenched
pedicellariz similar to those above described on the plates of the abactinal surface
occur on a number of the plates in the series adjacent to the adambulacral plates,
but the organ does not appear to diverge, or only very rarely, from the normal
form of a central foramen and two lateral trenches.

Dimensions.—In the specimen figured on Pl. X, fig. 2a, the major radius is_,
38 mm., and the minor radius 30 mm. Other examples have the following
approximate measurements: R=35 mm., r=27 mm.; R=43 mm., r=34 mm.
The diameter of the disk (R-+ 7) in well-grown tests ranges, therefore, from
60mm. to 80mm. The thickness of the margin is about 11°75 mm.

Locality and Stratigraphical Position—All the examples figured in this
Monograph were obtained from the Upper Chalk, near Bromley. The species is
a characteristic Upper Chalk fossil in the south of England, and has been found
in beds of that age at Brighton, Charlton, Gravesend, Kent, and other localities.
It is stated by Forbes to occur in the Lower Chalk of Sussex, but I have not seen
any examples from that horizon.

History.—A fossil starfish which has been generally considered to be this
species was figured by Parkinson in his ‘ Organic Remains of a Former World,’
vol. iii, pl. i, fig. 3, but it was referred by that author to the Pentagonaster
regularis of Linck. The last named has, however, been supposed to be a recent
species, but the type has unfortunately been lost, and the form has not subse-
quently been recognised definitely. Apart from this the fossil starfishes now
under notice are certainly distinct from the form indicated by Linck’s figure, and
this view was taken by Forbes, who named the species after Parkinson in his
memoir ‘On the Asteriade found fossil in British Strata,’ and figures of the

36 FOSSIL ASTEROIDEA.

fossils so named by him were given in Dixon’s ‘Geology and Fossils of the
Tertiary and Cretaceous Formations of Sussex,’ London, 1850. One, if not
more, of the specimens delineated in that work is now preserved in the British
Museum.

Variations.—In some examples the breadth of the border formed by the
supero-marginal plates on the abactinal surface is greater in relation to the disk-
area than in others, and this variety was noted by Forbes (‘Mem. Geol. Surv.,’
vol. ii, p. 472). I have not been able to establish the relation of this modification
with any other permanent morphological character, nor to associate it with any
special locality or stratum, and I am therefore led to consider, for the present at
least, that the variation in question is one affecting individual examples of the
species independently of other structural modifications which would warrant
recognition by name.

Two other variations are to be noted in this species which are superficially
much more striking, and either of them would, if only isolated examples were
known, lend a strong temptation to the separation of their possessor from the
normal form of the species. One of the variations in question affects the large
ultimate paired supero-marginal plates. On comparing the examples drawn on
Pl. X, fig. 1 and fig. 2a, with fig. 1a, Pl. XII, and fig. la, Pl. XI, it will be seen
that the ultimate plates are relatively much larger than the adjacent supero-
marginal plates and are more produced at the extremity; whilst in the specimen
delineated in fig. 2a, Pl. XI, this modification is carried to such an extent that
at first sight it would appear scarcely possible to believe that this fossil belongs
to the same species as, for example, fig. 2a, Pl. X. I have, however, been unable
to find any other constant variation from what has been considered the typical
form of Metopaster Parkinsoni associated with this modification in the size and
shape of the ultimate plates ; and as the most complete gradation between the two
extremes may be traced in the splendid series of specimens now preserved in the
British Museum, all obtained from the same locality and the same horizon, no
reasonable doubt can be entertained that the variation in question is of a
comparatively trivial character, affecting the individual independently, and that it
is not stamped by correlation with other structural modification with sufficient
importance to justify the forms being separated from the species, or even a name
being given to the variety. All the examples referred to in the foregoing remarks
and figured in the plates accompanying this Memoir are from the Upper Chalk,
and were obtained from the same locality near Bromley.

The fine specimen with large and greatly produced ultimate plates drawn on
Pl. XI, fig. 2a, is also an example of the second variation in the structure of this
species, to which I have alluded. This manifests itself in the presence of an

METOPASTER PARKINSONI. 37

additional supero-marginal plate. It will be seen that there are nine supero-marginal
plates, exclusive of the terminals (or so-called oculars) on each side of the penta-
gonal disk. At first sight the presence of this odd intermediate marginal plate in
association with the strikingly modified ultimate plates found in this example
would appear sufficient to indicate a well-marked variety, if not actually a distinct
species. A careful examination of the collection in the British Museum shows,
however, that this assumption is untenable, for in another example (which bears
the Museum number “ 46,765 ’’), which also possesses an odd intermediate marginal
plate on two of its sides, the ultimate plates are in no way specially abnormal in
their form or size. The presence of the additional supero-marginal plate would
in this case, therefore, appear to be only an occasional variation, and, so far as |
am able to detect, occurring independently of other structural modifications. I
have observed no differences either in the proportions or the ornamentation of the
intermediate marginal plates worthy of remark. This view is strengthened, if not
absolutely confirmed, by the presence in another specimen (British Museum number
*° 46,796’), drawn on Pl. XII, fig. 1 a, of nine supero-marginal plates on two sides
only of the disk, the remaining three sides having the normal number of eight
marginal plates. In this example the ultimate plates are distinctly larger and more
produced than in the truly normal forms of the species, but it will be noticed that
their development is unequal, as is also the case in No. ‘* 46,765; ” in other words,
in three of the rays one of the ultimate plates is smaller than the corresponding
ultimate plate to which it is adjacent. The two sides which have an additional
marginal plate are also the sides which have one of the smaller ultimate plates,
and the inference naturally follows that the additional plate is to balance or com-
pensate for the smaller size of the ultimate plate ; and I am inclined to think that
the additional plate is not in this case a true odd interradial marginal plate at all,
such as occurs in Gnathaster, Sladen, but that it is only a supplementary plate
formed because its primitive or embryonic rudiment has not been included in the
series merged together during development to form the structurally compound
ultimate plate. That these ultimate plates are compound, or formed by the union
of several embryonic plates, there is in my opinion little doubt when regard is had
to the embryonic history of the ultimate plate and the associated infero-marginal
plates.

38 FOSSIL ASTEROIDEA.

2. MeropasteR Manteuui, Forbes, sp. Pl. XIII, figs. 2a—4b.

PENTAGONASTER SEMILUNATUS, Parkinson, 1811. Organie Remains of a
Former World, vol. i, p. 3,
pl. i, fig. 1 (non Linck).

? GONIASTER SEMILUNATA, ; Mantell, 1844. The Medals of Creation,
London, vol. i, p. 388, lign. 75
(non Linck, ? nee Parkinson).

GontastER (Gonropiscus) Manrenu1, Forbes, 1848. Memoirs of the Geological
Survey of Great Britain,
vol. ii, p. 472.

— _ — Forbes, 1850. In Dixon’s Geology and
Fossils of the Tertiary and
Cretaceous Formations of
Sussex, London, 4to, p. 332,
pl. xxiii, figs. 11, 12.

— _— = Morris, 1854, Catalogue of British Fos-
sils, 2nd ed., p. 81.

? — MANrTrELLI, Mantell, 1854. The Medals of Creation,
London, 2nd ed., vol. i, p. 307,

lign. 98.
Asrroconium Manvexui, Dujardin and Hupé, 1862. Hist. Nat.

Zooph. ebin. (Suites a
Buffon), p. 399.

GontastER (Gontopiscus) Manreit, Forbes, 1878. In Dixon’s Geology of Sussex
(new edition, Jones), p. 366,
pl. xxvi, figs. 11, 12.

Body of medium or rather small size. General form depressed. Abactinal
surface flat, with the area circumscribed by the supero-marginal plates plain and
rather sunken below the level of the median convexity of the border formed by
these plates. Actinal surface plain, or may be a little convex in consequence of a
tendency to a slight upturning of the rays in some examples. Marginal contour
pentagonal, with sides faintly lunate, or in small examples they may be almost
straight. The major radius measures scarcely one-third more than the minor
radius. Margin rather thin, but well rounded.

The supero-marginal plates are four in number, counting from the median
interradial line to the extremity, or eight from the tip of one ray to the tip of
the adjacent ray, exclusive of the odd terminal or ‘ ocular” plate in each case.
In one example, which is figured by Forbes in Dixcn’s ‘Geology of Sussex,’

METOPASTER MANTELLI. 39

there would appear to be only seven plates on the entire side, but only one
radial angle is preserved intact, and much displacement of plates has occurred.
The circumstance is in any case of comparatively trifling importance. The
supero-marginal plates form a conspicuous and moderately broad border to the
abactinal area of the disk, of uniform breadth throughout, and measuring about
6 mm. at the median interradial line in an example whose minor radius measures
about 25 mm.

Excepting the ultimate paired plate all the supero-marginal plates are nearly
subequal in size ; the plates, however, adjacent to the median interradial line are
slightly longer in relation to their breadth than those adjacent to the ultimate
paired plate. The plates adjacent to the median interradial line have the appear-
ance of being nearly square in outline as seen from above, the actual dimensions
in the specimen under notice, whose side measures 38 mm., being length 5 mm.,
and breadth 6 mm. Ina smaller example, with a side measurement of 29°5 mm.,
the corresponding plate is 3°5 mm. long and 45 mm. broad. In Forbes’s type
the measurements are, length 3°75 mm., and breadth 5 mm.

The abactinal surface of the supero-marginal plates is slightly tumid, and the
general surface of the whole series forms a well-rounded regular curve from the
adcentral margin to the margin in the lateral wall adjacent to the infero-marginal
plates. The height of the plates as seen in the margin is less than their length,
the actual measurement being about 3 mm. Their abactinal contour is distinctly
convex, but not gibbous. The whole superficies of the plates is covered with
small, widely spaced, equidistant, uniform punctations. In some examples which
have been subjected to much weathering the punctations are almost obliterated,
as in the case of the fine specimen shown on Pl. XIII, fig. 2a. A narrow
depressed border surrounds the margin of the plate, bearing very small, closely
crowded punctations, those adjacent to the main or median area being in serial
arrangement. Occasionally a small entrenched pedicellaria may be detected on
the median area of the plate (see Pl. XIII, fig. 2d).

The ultimate paired plate is small and triangular in outline as seen from above,
and one margin touches the corresponding plate of the adjacent side of the disk
throughout, the junction coinciding with the median radial line of the disk. The
length of the plate—that is to say, of the side of the plate which falls in the margin
of the disk—is a trifle greater than the length of the largest supero-marginal
plate, measuring in the example under notice nearly 5°5 mm., whereas the
breadth of the plate, or measurement of the side adjacent to the penultimate
supero-marginal plate, is not more than 5mm. Near the outer or apical extremity
of this plate when seen from above there is frequently a more or less strongly
developed tendency to gibbosity present.

The abactinal area of the disk within the boundary of the marginal plates is

40 FOSSIL ASTEROIDEA.

covered with small subregular hexagonal plates or paxillar tabule, which are,
however, comparatively large for the size of the species. The primary apical
plates are large and conspicuous, and all the plates in the central area of the disk
and in the interradial areas are considerably larger than elsewhere ; all diminish
in size as they approach the margin of the disk. All the plates have their surface
covered with a small, uniform, closely placed granulation, in which an indefinite
subcircular arrangement in relation to the centre of the plate is discernible (see
Pl. XIII, fig. 2 d).

The madreporiform body is small and subtriangular in outline ; its surface is
sculptured by fine striations which radiate from the centre to the margin. The
madreporite is usually enclosed by three plates, but four may be present in conse-
quence of the division or retarded development of one of them, as is the case in
the example figured on Pl. XIII, fig. 2c. The position of the madreporite is
rather nearer the centre of the disk than the margin.

The infero-marginal plates are at least six in number, counting from the
median interradial line to the extremity,—that is to say, there are twelve (or
perhaps more) for the whole side of the disk, as against eight in the supero-
marginal series. The length of the three innermost plates on each side of the
median interradial line is equal to that of the superior series, but there are three
infero-marginal plates corresponding to the ultimate paired supero-marginal. As
seen in the lateral wall of the disk, the height of the infero-marginal plates is greater
than that of the supero-marginal series. The breadth of these plates, adjacent to
the median interradial line on the actinal surface, is 5 mm. in an example whose
minor radius measures about 15 mm. ‘The breadth of the marginal border appears
to be well maintained till near the extremity. The surface of the plates is
ornamented in a similar manner to that of the supero-marginal plates, excepting
that the punctations on the main area are rather more numerous, and that the
finely punctate depressed border round the margin of the plate is much broader
than in the plates of the superior series; the border is broader on the adcentral
margin of the plate than elsewhere. (Compare figs. 3b and 46 on Pl. XIII.)

The adambulacral plates appear to be comparatively small, but their preserva-
tion in the examples examined is not sufficiently good to permit of description.

The actinal intermediate plates, which are small and hexagonal, are covered
with small, closely crowded, uniform granules. All are much displaced in the
specimens under notice.

Dimensions.—In the example figured on Pl. XIII, fig. 2a, the major radius is
about 32 mm., and the minor radius about 25 mm. ‘The length of the side is 38
mm. The thickness of the margin is about 8mm. In Forbes’s type (Pl. XIII,
fig. 3 a) the major radius is about 20 mm., and the minor radius about 15°5 mm. ;

METOPASTER MANTELLI. 41

the length of the side about 25°5 mm. In a very finely preserved cast from the
Upper Cretaceous beds of Haldon the length of the side is 29°5 mm.

Locality and Stratigraphical Position All the examples of this form with
which I am acquainted are from the Upper Chalk. One of Forbes’s type is from
Gravesend, but the locality of the other is not recorded. The large example
figured on Pl. XIII, fig. 2 a, is from the Upper Chalk near Bromley.

An extremely well-preserved cast in flint from the Upper Cretaceous beds
of Haldon (Devonshire) is in the collection of the Albert Museum, Exeter; and a
cast of this example may be seen in the Museum of the Geological Survey, Jermyn
Street. /

History.—This form appears to have been first recognised by Parkinson, who
erroneously referred it to the Pentagonaster semilunatus of Linck. The latter is a
well-known recent species, and quite distinct from the fossil under consideration.
Mantell, following Parkinson’s determination, referred to the form under the
name of Goniaster semilunata. Forbes was the first to indicate that these views
of his predecessors were incorrect, and diagnosed the species in his memoir ‘ On
the Asteriade found fossil in British Strata’ under the name of Goniaster (Gonio-
discus) Mantelli ; and figures of two examples were subsequently given in Dixon’s
‘Geology and Fossils of the Tertiary and Cretaceous Formations of Sussex,’
London, 1850. Both these specimens are now preserved in the British Museum.
Careful drawings of each fossil are given on Pl. XIII, figs. 3a and 4a.

Remarks.—lIt is not without hesitation that I maintain this species of Forbes’s
as independent from Metopaster Parkinsoni. For the present, however, I
consider it to be distinguished by the smaller size, the comparative squareness
of the supero-marginal plates, the small size of the ultimate paired plates, as
well as by the character of the ornamentation of the supero-marginal plates and
of the abactinal plates. Whether a more extensive series of specimens will break
down or uphold these distinctions I do not feel prepared to say. It is undoubted
that the two forms are very nearly allied.

I feel considerable doubt as to whether one of Forbes’s types—that shown in
Pl. XIII, fig. 4a—really belongs to the same species as the examples illustrated in
figs. 2a@ and 3 a on the same plate.

42 FOSSIL ASTEROIDEA.

3. Meroraster BowErBankil, Forbes, sp. Pl. XV, figs. 2a—2d; Pl. XVI, figs.
la—1 d.

GontastEr (Gonropiscus) BowErBankul, Forbes, 1848. Memoirs of the Geolo-
gical Survey of Great Britain,
vol. ii, p. 473.
— — — Forbes, 1850. In Dixon’s Geology and
Fossils of the Tertiary and
Cretaceous Formations of
Sussex, London, 4:to, p. 333,
pl. xxii, fig. 4.
— — — Morris, 1854. Catalogue of British
Fossils, 2nd ed., p. 81.

AsTROGONIUM BowERBANKII, Dujardin and Hupé, 1862. Hist. Nat.
Zooph. Hehin. (Suites 4 Buffon),
p. 399.

GontastEerR (Gontopiscus) BowerBANnkKil, Forbes, 1878. In Dixon’s Geology of
Sussex (new edition, Jones),
p- 366, pl. xxv, fig. 4.

Body of medium or rather large size. General form depressed. Abactinal
surface flat ; actinal surface also flat, or with a slight tendency to become convex.
Marginal contour pentagonal, with a very small amount of lunation in the sides.
Margin thick and well rounded.

The supero-marginal plates are five in number counting from the median
interradial line to the extremity, or ten from the tip of one ray to the tip of the
adjacent ray, exclusive of the odd terminal or ‘‘ ocular ” plate in each case. They
form a broad border to the abactinal area of the disk of uniform breadth through-
out, which measures about 8°5 mm. at the median interradial line in the type-
specimen described by Forbes from the collection of the late Dr. Bowerbank.
Excepting the ultimate paired plate all the supero-marginal plates are of equal
size. They are short and broad, the breadth being nearly twice and a half the
length, the actual measurements in the type being length 3°5 mm. or a trifle more,
and breadth 8°5 mm., in which case the dimensions are in the proportion of 7:17.
The abactinal surface of the plates is slightly convex along the direction of the
breadth, sufficient to define each plate distinctly. The general surface of the
whole series is gently arched towards the margin adjacent to the infero-marginal
plates. The height of the plates as seen in the margin is about 6 mm., and there
is scarcely any or only the slightest diminution in height as the plates approach

METOPASTER BOWERBANKII. 43

the extremity of the ray, and the ultimate paired plate is not prominent or gibbous
abactinally. The whole superficies of the supero-marginal plates is covered with
small, widely spaced, equidistant, uniform punctations, and along the entire
margin of the plate is a very narrow and deeply depressed border of fairly
uniform breadth, covered with much smaller and closely crowded punctations,
upon which much smaller granules than those which occupied the central area of
the plate were originally borne. One or two or even three small entrenched
pedicellarize may be present on the central area of a plate, irregularly
disposed.

The ultimate paired plate is triangular in form as seen from above. The
margin or side which represents its length and coincides with the margin of the
disk measures a little more than once and a half or nearly twice the length of the
adjacent supero-marginal plates; and the margin representing the breadth of the
plate which abuts on the penultimate supero-marginal plate is shorter than the
margin of the latter plate. The remaining side of the plate which touches the
corresponding ultimate plate of the adjacent side of the disk, and falls in the
median radial line of the disk, is subequal to or even slightly shorter than the
breadth of the preceding marginal plates. The ultimate plate in this species is
not elongated, and no prolongation beyond the normal pentagonal contour of the
disk occurs in the extension of the median radial line. The surface of the ultimate
plate is covered with punctations and margined with a finer and closely crowded
series precisely similar to those on the other supero-marginal plates.

The abactinal area of the disk within the boundary of the marginal plates is
covered with small, subregular, hexagonal plates or paxillar tabula, which have
their surface marked with minute, low, subhemispherical, and closely placed
miliary granulations, which do not, however, extend quite to the margins of the
plates. A number of the plates bear in the centre a rather large entrenched
pedicellaria, consisting of a central foramen and normally two lateral fosse, and
there is usually a circular series of coalesced granules in which the foss are
included, which imparts a very characteristic appearance to the organ in this
species (see Pl. XVI, fig. 1 7).

The following description of the characters of the actinal area of the disk of
this species is taken from an example preserved in the Museum of Practical
Geology, Jermyn Street, and figures of which are given on Pl. XV, figs. 2a—2d.

The infero-marginal plates are eight in number counting from the median
interradial line to the extremity,—that is to say, there are sixteen for the whole
side of the disk, as against ten in the supero-marginal series. The length of the
four innermost plates on each side of the median interradial line is slightly greater
than that of the corresponding plates of the supero-marginal series; and there
are four infero-marginal plates much smaller than those preceding, corresponding

44 FOSSIL ASTEROIDEA.

to the ultimate supero-marginal plate. As seen in the lateral wall of the disk the
height of the infero-marginal plates is slightly greater than that of the supero-
marginal series. The breadth of these plates adjacent to the median interradial
line on the actinal surface is 6 mm., or even a little more in an example whose
major radius measures 37 mm. and the minor radius 27 mm. The breadth of the
marginal border rapidly diminishes towards the extremity of the ray. The
surface of the infero-marginal plates is ornamented with extremely small and
closely crowded punctations, upon which traces of a minute, closely crowded, and
uniform granulation are preserved here and there.

The adambulacral plates are small, about twice as broad as long, and their
surface is traversed by about three ridges, bearing punctures, running parallel to
the ambulacral furrow, upon which the spinelets composing the adambulacral
armature were articulated. There were about five or six spinelets in each lineal
series.

The mouth-plates are regularly triangular, about twice and a half as long as
broad, and the two adjacent plates which constitute a pair form together a regular
rhomboid or lozenge-shaped figure. Their surface is covered with small, crowded,
rather coarse, irregularly disposed tubercles or granules (see Pl. XV, fig. 2 d).

The actinal intermediate plates are fairly large, and there is a distinct diminu-
tion in size towards the outer margin of the disk adjacent to the marginal plates.
The plates adjacent to the adambulacral plates are pentagonal, but a subhexagonal
or polygonal form elsewhere is the rule, with comparatively little irregularity.
The surface of the actinal intermediate plates is entirely covered with very small
equidistant punctations, upon which a uniform close granulation was previously
borne. Occasional small excavate pedicellarize are present here and there, the
lateral fossee being slightly curved.

Dimensions.—The type specimen is unfortunately fragmentary, and the radial
dimensions cannot be given. The length of one side of the disk, measured from
the tip of one ray to the tip of the adjacent ray, is about 41 mm. ; the breadth of
the supero-marginal plates adjacent to the median interradial line is 9 mm. ; and
the thickness of the margin is about 12 mm.

In the fine example preserved in the Museum of Practical Geology, Jermyn
Street, figured on Pl. XV, fig. 2a, the major radius measures 37 mm., and the
minor radius 27 mm.; the length of one side of the disk measured about
42 mm., or probably rather more when complete; the breadth of the infero-
marginal plates adjacent to the median interradial line is 6 mm.; and the
thickness of the margin is 8°56 mm.

Locality and Stratigraphical Position.—The type specimen is stated by Forbes

METOPASTER ZONATUS. 45

to have been obtained from the Upper Chalk of Kent, but no indication of locality
is now preserved on the label. The example belonging to the Museum of
Practical Geology is from the Upper Chalk of Gravesend.

History—The type of this species, which was described by Forbes in his
memoir ‘On the Asteriade found fossil in British Strata’ (‘ Mem. Geol. Surv.,’
vol. ii, p. 473, 1848), originally formed part of Dr. Bowerbank’s Collection, and
was first figured by Forbes in Dixon’s ‘Geology and Fossils of the Tertiary and
Cretaceous Formations of Sussex,’ London, 1850, pl. xxii, fig. 4. That specimen
is now preserved in the British Museum, where it bears the register number
** 2578.” An accurate drawing of the type is given on Pl. XVI, fig. 1 a, of the
present work.

4. Meropaster zonatus, Sladen. Pl. XII, figs. 2 a—2 ce.

Body of small or medium size. General form depressed. Abactinal surface in
the fossil condition, as at present known, essentially concave in consequence of
the conspicuously upturned extremities of the rays. Actinal surface conformably
convex. Marginal contour pentagonal, with the radial angles slightly produced
and obtusely rounded, and the sides distinctly lunate. The major radius measures
nearly one-half more than the minor radius, or in the proportion of 3:2 approxi-
mately ; the actual dimensions in the example under notice being R = 27 mm.,
r = 19 mm. approximately. Margin very thick in relation to the size of the disk,
and regularly rounded.

The supero-marginal plates are four in number, counting from the median
interradial line to the extremity, or eight from the tip of one ray to the tip of the
adjacent ray, exclusive of the odd terminal or “ ocular” plate ineach case. They
form a broad border to the abactinal surface of the disk, of uniform breadth
throughout, which measures about 8 mm. at the median interradial line, in an
example whose minor radius measures 19 mm. Excepting the ultimate paired
plate all the supero-marginal plates are subequal, the breadth being nearly three
times the length, the actual measurements in the example under notice being
length 2°75—3 mm.,and breadthabout 8 mm. The abactinal surface of the plates
is distinctly convex along the median line of breadth, by which means each plate
is conspicuously defined. The general surface of the whole supero-marginal
series between the two ultimate plates is regularly rounded, and forms an uninter-

7

46 FOSSIL ASTEROIDEA.

rupted regular curve between the adcentral margin of the plate and the margin
in the lateral wall adjacent to the infero-marginal plates. The height of the
plates as seen in the lateral view of the disk is greater than their length, and
there is an apparent increase in height as the plates approach the extremity of
the ray, the ultimate plate being still higher and distinctly tumid (see Pl. XII,
fig. 2b). The whole superficies of the plates is covered with small, widely spaced,
equidistant, uniform punctations; and there is a narrow depressed border sur-
rounding the margin of the plate with much smaller and closely crowded puncta-
tions, upon which minute miliary granules were previously borne.

The ultimate paired plate is much larger than any of the other supero-marginal
plates, its length as measured on the outer margin being more than twice the
length of the other supero-marginal plates. Its breadth is about equal to that of
the adjacent supero-marginal plate. It is subtriangular in form as seen from
above, and the line of junction with the corresponding companion plate of the
adjacent side of the disk is complete throughout, and coincides with the median
radial line. The convexity of the plate falls in a line parallel and adjacent to this
margin of the plate, and its height is greatest there. From this convexity the
surface slopes gradually and regularly in conformity to the curve of the superficies
of the other supero-marginal plates. The actual dimensions of the ultimate plate
in the example under notice are, length 7 mm., breadth 7 mm.; greatest height as
seen inthemarginalview,about 8°5 mm. The surface of the ultimate plate is marked
with a precisely similar ornamentation to that on the other supero-marginal plates.

The odd terminal plate is very small, and though only traces are present in
the type it is well preserved in other examples. It is prominent, cylindrical, and
abruptly truncate, resembling in all respects the form described in Metopaster
Parkinsoni.

The madreporiform body, which is only partially exposed in the example under
notice, is apparently subtriangular in outline, and is marked with very fine centri-
fugally radiating striations (see Pl. XII, fig. 2 c).

The infero-marginal plates are seven in number, counting from the median
interradial line to the extremity,—that is to say, there are fourteen for the whole
side of the disk, as against eight in the supero-marginal series. The length of the
three innermost plates on each side of the median interradial line is subequal to
that of the superior series, but their height as seen in the marginal view of the
disk is much greater, the thickness of the whole margin, 1. e. both series of plates
together, being 10 mm. in an example whose minor radius is19mm. Four plates
underlie the superior ultimate plate, the last two being very small and triangular
inform. All the four are adjacent to the adambulacral series of plates. The
ornamentation or surface-marking of the infero-marginal plates is precisely
similar to that of the supero-marginal series.

METOPASTER UNCATUS. 47

Dimensions.—In the example figured on Pl. XII, fig. 2a, the major radius is
27 mm. and the minor radius about 19 mm., the thickness of the margin 10 mm.

Locality and Stratigraphical Position.—The figured example is from the Upper
Chalk near Bromley. A number of other examples of this species are in the
collection of the British Museum, all from the Upper Chalk, and most of them
from the same locality, the locality of the remainder being not recorded.

Remark:s.—I was at first somewhat in doubt as to whether to rank this form
as a variety of Metopaster Parkinsoni, or as a distinct species. I have taken
the latter course. Although there are many points of resemblance in general
structure, as well as in various details when considered independently, the facies
of this form is so distinct from that of Metopaster Parkinsoni, and is so readily
recognisable, that there seems to me full justification for considering Metopaster
zonatus specifically distinct. The great breadth of the supero-marginal plates as
compared with their length, the form and character of the ultimate plates, the
great thickness of the margin, and the relative proportions of the marginal plates,
as well as the general habit of the abactinal surface, apart either from the
individual characteristics of the plates independently or from other special details,
are amply sufficient to distinguish the form from its nearest ally, Metopaster
Parkinson.

5. Meropaster unoatus, Forbes, sp. Pl. XI, figs. 3a, 3b; Pl. XIV, figs. 1 a—3;
Pl. XV, figs. 1a, 1b.

GontastER (Gontopiscus) uncatus, Forbes, 1848. Memoirs of the Geological
Survey of Great Britain, vol. ii,
p- 472.

a _ — Forbes, 1850. In Dixon’s Geology and
Fossils of the Tertiary and Creta-
ceous Formations of Sussex, London,
4to, p. 331, pl. xxi, figs. 4, 5, 8.

— _ _ Morris, 1854. Catalogue of British Fossils,
2nd ed., p. 81.

ASTROGONIUM UNCATUM, Dujardin and Hupé, 1862, Hist. Nat. Zooph.
Behin. (Suites 4 Buffon), p. 399.

GontastEer (Gonropiscus) uncatus, Forbes, 1878. In Dixon's Geology of Sussex
(new edition, Jones), p. 365, pl. xxiv,
figs. 4, 5, 8.

48 FOSSIL ASTEROIDEA.

Body rather small, or at most only of medium size. General form depressed.
Abactinal surface flat, or with a tendency for the rays to be slightly directed
upward at the extremities. Actinal surface slightly convex. Marginal contour
pentagonal, with the sides slightly lunate, though the curvature is often more or
less flattened; the extremity of the rays is only slightly produced. The major
radius measures about one-third more than the minor radius, the major dimension
being proportionately rather greater in large tests than in small ones. The
margin is thick, and the lateral wall has more of a precipitous than a rounded
character, although the infero-marginal plates are well rounded on the actinal
surface.

The supero-marginal plates are three in number, counting from the median
interradial line to the extremity, or six from the tip of one ray to the tip of the
adjacent ray, exclusive of the odd terminal or “ ocular”’ plate in each case. They
form a broad border to the abactinal area of the disk, of uniform breadth through-
out, which measures about 7 mm. at the median interradial line in an example
whose major radius measures 36°5 mm. and minor radius 27 mm. (Pl. XIV, fig. 2 a).
Excepting the ultimate paired plates, the four intermediate supero-marginal plates on
each side of the disk are of equal size, the breadth being about once and a half the
length, the actual measurements in the specimen under notice being length 5 mm.,
and breadth 7mm. The abactinal surface of these plates is tumidly convex, while
the lateral wall is plane and vertical, and the abactinal tumidity commences
abruptly at a little distance from the adcentral margin of the plate, which leaves a
small level area at the rounded end of the plate abutting on the abactinal plates or
paxillar tabulz. On the surface of this level band near the adcentral margin, and
forming a more or less definite series running parallel to it, are three or four
irregular tubercular eminences or granules, but very indistinct and more or less
weatherworn (see Pl. XIV, fig. 2c). The entire margin of the plate is surrounded
by a very narrow depressed border, with very fine, closely crowded, uniserially
disposed punctations, upon which a small miliary granulation was previously borne.
The whole general superficies of the plate is smooth and weatherworn in every
example I have seen. The height of the plates as seen in the margin is as great
as or even slightly greater than their length, and the prominently tumid character
of the plates abactinally causes them to appear in the lateral view somewhat like
truncate cones abruptly rounded (see Pl. XIV, fig. 20).

The ultimate paired plate is larger and longer than any of the other plates,
and is of a different and very peculiar shape. It is subtriangular in form as seen
from above, produced and pointed at the extremity, and to a certain extent recalls
the form of a ploughshare or coulter in consequence of a peculiar nipped-in
appearance caused by the extension of a tumid region which runs parallel to the

METOPASTER UNCATUS. 49

outer margin and rises abruptly from a level area which occupies the inner half of
the plate along the margin touching the corresponding plate of the adjacent ray,
—the line of junction of the two plates coinciding with the median radial line.
The length of the ultimate plate is nearly twice that of the other marginal plates,
measuring 9°25 mm. in an example whose major radius is 36°5 mm. and minor
radius 27 mm. As seen in the marginal view of the test, the ultimate is not
higher or more tumid than the other marginal plates (see Pl. XIV, fig. 2). The
outer margin of the ultimate plate has a slight concave curvature, and the inner
margin adjacent to the corresponding plate is curved convexly towards the
proximal end of the plate. In consequence of this rounding the two ultimate
plates in a pair do not unite throughout their entire length, but are separated by
a small notch at the end of the suture adjacent to the abactinal paxillar area of
the disk. On the small level area of the ultimate plate are a number of small
irregular tubercular eminences ; four or five larger than the others form a sort of
series parallel to the rounded margin, and a longitudinal series of eight or nine
much smaller miliary granules run along the flank of the longitudinal tumidity of
the plate; and several additional granules of intermediate size may be present in
the space between the two series just described. Excepting these granules, the
sutface of the ultimate plate is smooth like that of the other supero-marginal plates.

The abactinal area of the disk within the boundary of the marginal plates is
covered with small subregular plates or paxillar tabulz, an hexagonal form predomi-
nating. ‘I'he plates in the median interradial areas are much larger than the other
plates on the disk, and a marked diminution in size in all the plates takes place as
they approach the margin. All the plates have their surface marked with a very
fine granulation. Small entrenched pedicellariz are occasionally present, but
there appear to have been very few.

The primary basal plates are larger than any of the other abactinal plates.
They are well seen in an example from the Upper Chalk of Kent, in which the
inner side of the abactinal wall is exposed by the removal of the actinal floor and
ambulacral plates. This specimen, which is preserved in the British Museum, and
bears the registration number “35,496,” is drawn on Pl. XI, fig. 3a. The
example in question is further interesting in showing that the plates of the radial
regions have stellate bases, whereas the larger plates of the interradial regions are
sharply hexagonal, and fit closely to their adjacent plates (see fig. 3d).

The madreporiform body is very small, and is subsagittiform or irregularly
lozenge-shaped in outline; in the example under notice it is embedded, all except
two straight sides, in one large adcentrally placed basal plate (see Pl. XIV,
fig. 2d); the two straight sides are bounded each by one large plate. The
surface of the madreporite is sculptured by very fine striations, which though
more or less wavy are directed subparallel to the adcentral sides of the body.

50 FOSSIL ASTEROIDEA.

The position of the madreporite is nearer the centre of the disk than the
margin.

The infero-marginal plates are six in number, counting from the median inter-
radial line to the extremity,—that is to say, there are twelve for the whole side of
the disk, as against six in the supero-marginal series. The length of the two
innermost plates on each side of the median interradial line is subequal to that of
the superior series, but the four succeeding infero-marginal plates are much shorter,
and together correspond to the large ultimate supero-marginal plate. As seen in the
lateral wall of the disk the height of the infero-marginal plates is very little more
than one-half that of the superior series, the actual proportion in the specimen
under notice being less than three-fifths. The breadth of the plates adjacent to
the median interradial line on the actinal surface is 7°5 mm. in an example whose
major radius measures 25 mm. and the minor radius 20 mm. The breadth of the
marginal border diminishes so rapidly on each side of the median interradial line
towards the extremities of the rays that the inner margin of the series of infero-
marginal plates of each side of the disk is practically a sector of a circle. The
surface of the plates is marked with small widely spaced punctations, and there is
a narrow depressed border round the entire margin of the plate which is very finely
punctate; the border is broader at the adcentral end of the plate and the adjacent
corners than elsewhere.

The adambulacral plates are small, broader than long, and their surface is
marked with about three oblique series, and an outer irregular group of fine
punctations upon which the spinelets composing the armature of the plates were
articulated. There are about six punctations in the innermost ridge or series
adjacent to the ambulacral furrow. A few spinelets are still preserved in an
example from the Upper Chalk near Bromley, from which this description is taken,
which is in the British Museum collection, and bears the registration number,
«* #2613.” The spinelets are cylindrical, truncated, and do not appear to taper ;
and their length is about equal to the length of the plates.

The actinal intermediate plates are rhomboid and hexagonal or polygonal in
form, and their surface is entirely covered with small equidistant punctations,
upon which small uniform granules were previously attached. Remains of this
granulation are occasionally to be found preserved in situ, and may be seen in the
example referred to in the preceding paragraph. Small entrenched pedicellarie
are occasionally present on the actinal intermediate plates.

Dimensions.—The example figured on Pl. XIV, fig. 2a, has a major radius of
36°5 mm. and a minor radius of 27 mm. The length of the side is about 40 mm.
Another specimen, shown in fig. 1 a on the same plate, is rather larger, the length
of the side being about 45 mm. Much displacement of the plates has occurred in

METOPASTER SUBLUNATUS. 51

this fossil, and the radial dimensions can only be calculated approximately. The
specimen figured by Forbes, which I have unfortunately not been able to trace, is
smaller than either of the above. The drawing represents a test with the major
radius measuring about 25 mm., the minor radius 20 mm., and the length of the
side about 28 mm. The figure of the actinal aspect of this example is reproduced
on Pl. XIV, fig. 3, of this memoir.

Locality and Stratigraphical Position —This is a well-known Upper-Chalk form.
It is recorded by Forbes from Kent, Sussex, and Wiltshire. Specimens in the
British Museum bear the localities of Charlton, Gravesend, Bromley, and
*« Kent.”

History.—This species, primarily described by Forbes in his memoir ‘ On the
Asteriade found fossil in British Strata,’ was first figured in Dixon’s ‘ Geology
and Fossils of the Tertiary and Cretaceous Formations of Sussex,’ London, 1850.
The example which I consider must undoubtedly have been the type (Dixon,
pl. xxi, figs. 4 and 5) I have hitherto unfortunately not been able to find. It is
stated by Forbes to have formed part of the late Mr. Pearce’s collection, and to
have been obtained from the Wiltshire Chalk. A badly drawn fragment (Dixon,
pl. xxi, fig. 8) is preserved in the British Museum, but I do not consider it to be
correctly referred to this species. In my opinion it belongs to a distinct species,
Metopaster cingulatus, of which a description is given at p. 53.

The examples of this species which have been drawn on Pl. XIV are preserved
in the British Museum.

6. Meropaster susLtunatus, Porbes, sp.

Gontaster (GoNTODISCUS) SUBLUNATUS, Forbes, 1848. Memoirs of the Geological
Survey of Great Britain, vol. ii,
p. 472.

— _ _ Forbes, 1850. In Dixon’s Geology and
Fossils of the Tertiary and Cretace-
ous Formations of Sussex, London,
4to, p. 331.

_ — — Morris, 1854. Catalogue of British Fos-
sils, 2nd ed., p. 81.

AsTROGONIUM SUBLUNATUM, Dujardin and Hupé, 1862. Hist. Nat.
Zooph. Echin. (Suites 4 Buffon),
p. 399.

GontastER (Gontopiscus) suBLUNATUS, Forbes, 1878. In Dixon’s Geology of
Sussex (new edition, Jones),
p- 365.

52 FOSSIL ASTEROIDEA.

This species was described by the late Professor Edward Forbes in Dixon’s
‘Geology of Sussex’ in 1850 in the following words :

“ Body pentagonal, with gently lunated sides. Superior intermediate marginal
plates four, nearly equal, plain, smooth, or minutely punctate. Inferiors similar.
Superior oculars mitrate, large, triangular, acuminated. Ossicula of dise punc-
tate.

“Easily distinguished from the last species [wncatus] by its flattened mar-
ginals and from the next [ Hunteri] by its lunated sides.

**Mus. Bowerbank, from the white chalk; also in the collection of the
Geological Survey ”’ (op. cit., p. 331).

No figure of the species has ever been published, and no record exists as to
the specimen or specimens seen or used by Forbes as type. I have been unable to
find any example from the Bowerbank Collection to which this name has been or can
be applied ; and the only examples which I have seen referred to this species at all
are four fragmentary specimens now preserved in the Museum of the Geological
Survey in Jermyn Street, and not more than two of these could have been in that
collection in 1850.

After a careful study of these specimens I am bound to confess that 1 find no
character by which they can be separated from Metopaster wncatus; and if the
diagnosis is correct I am led to consider that I have certainly not seen Forbes’s
type. The swpero-marginal plates in the specimens in question are not flattened,
and cannot be said to differ in character from those of Metopaster wncatus.

A possible explanation suggests itself in the supposition that Forbes inad-
vertently mistook the actinal for the abactinal surface of the disk, a mistake which
might easily be made by a less careful observer than the author of this species
when dealing with a badly preserved fragment. If, however, what is really the
actinal surface has been described as the abactinal surface the difficulty is prac-
tically solved, for the infero-marginal plates in the fragments under notice are
“plain, smooth, or minutely punctate.’’ That this is not an improbable explanation
I would submit the following facts :—(1) that in the original diagnosis of 18481
Forbes states that the infero-marginal plates are unknown; (2) that in the
diagnosis of 1850, given above, the infero-marginal plates are stated to be
“similar” (to the supero-marginals) ; and (3) that notwithstanding these state-
ments all the examples in the Geological Survey Collection are essentially actinal
presentments of the disk, and therefore the infero-marginal plates are the plates

1 The following is the diagnosis in full :—‘‘G. corpore pentagonali, lateribus lunatis ; ossiculis
lateralibus superioribus 4, subequalibus, planis, minutissime punctatis; inferioribus? Ossiculis
ocularibus superioribus magnis, triangularibus, mitratis, tumidis, acuminatis.’ (‘ Mem. Geological

Survey of Great Britain,’ vol. ii, p. 472.)

METOPASTER CINGULATUS. 53

conspicuously available for description. To my mind it follows with little doubt
either that Forbes has described the infero-marginal plates as supero-marginals,
or else that I have not seen his type specimens.

With my reverence for all that Forbes has written I prefer to leave the species
as described by him, together with the record of these remarks, rather than strike
a ruthless pen through any species created by so careful and accurate a worker.
Time will pronounce the verdict.

I consider it quite unnecessary to figure any of the fragments, here referred to, as
being in the Jermyn Street Collection. The differences they present as compared
with a series of Metopaster wncatus do not in my opinion amount to even varietal rank,
and are confined to the slightly less developed tumidity of the abactinal surface of
the supero-marginal plates, and to the external contour of the mitrate ultimate paired
supero-marginal plates, which is slightly convex marginally rather than incurved
to produce the characteristically claw-shaped form of Metopaster uncatus. These
are, in my Opinion, merely individual differences.

7. Meropaster cineutatus, Sladen. Pl. XIV, figs. 4 a—4d.

Body of small size. General form depressed. Abactinal surface slightly
concave, actinal surface flat. Marginal contour pentagonal, with the sides slightly
lunate. The rays are not produced beyond the contour of a true pentagon, and
the radial angles are not rounded. The major radius is proportional to the minor
radius as 100: 77:5, the actual dimensions in the example described being, major
radius 20 mm., minor radius 15°5 mm, approximately. The margin is thick, and
though well rounded has more or less of a precipitous character.

The supero-marginal plates are three in number, counting from the median
interradial line to the extremity, or six from the tip of one ray to the tip of the
adjacent ray, exclusive of the odd terminal or “ ocular”’ plate in each case. They
form a broad border to the abactinal area of the disk of uniform breadth through-
out, which measures about 6°5 mm. at the median interradial line in an example
whose diameter (R +r) measures from about 36 mm. to 37 mm.

Excepting the ultimate paired plates, the four intermediate supero-marginal
plates on each side of the disk are approximately equal in size, the breadth being
more than twice and a half the length, the actual dimensions in the example under
notice being length 2°25 mm., and breadth 6°5 mm., in the plate adjacent to the
median interradial line. The abactinal surface of these plates is distinctly
tumid, a subtubercular eminence rising in the median third of the abactinal area
of the plate. ‘The slope of the tumidity descends gradually on the outer side, and

8

54 FOSSIL ASTEROIDEA.

merges in the rounding of the high lateral wall of the plate. On the level area
of the plate, which is consequently the inner or adcentral part of the surface, are
several low tubercular eminences of irregular shape and disposition, but which
appear to assume a more or less distinct biserial arrangement at right angles to
the adcentral margin of the plate. They appear to be enlarged irregular granules,
and in all the examples I have examined they have become to a certain extent
ill-defined, owing either to growth or to weather-wearing. The general character
of the ornamentation is shown in Pl. XIV, fig. 4c. The entire margin of the
plate is surrounded by a very narrow depressed border, which is very minutely
punctate, probably only in a single lineal series. The general superficies of the
plate beyond the ornamentation mentioned is smooth, as if weatherworn, in all
the examples I have seen, but in some specimens there appear to be traces of a
more or less granulous character, and in some instances suggest the impression
that an ornamentation similar to that noticed in Mitraster rugatus was probably
present on at least a part of the surface of the plate. The height of the supero-
marginal plates as seen in the margin is greater than their length, and their
prominent abactinal tumidity has a distinctly conical character from this point
of view.

The ultimate paired plate is fully twice as long as the other supero-marginal
plates measured on the outer margin, and its breadth is equal to that of the
adjacent supero-marginal. It is triangular in form, and the line of junction with
the companion ultimate plate of the adjacent side coincides with the median
radial line. The actual dimensions in the example under description are, length
6°2 mm., breadth about 6 mm. As seen in the lateral view of the disk the
ultimate plates are distinctly tumid abactinally (see Pl. XIV, fig. 40). The
abactinal surface of the plate is ornamented by a number of miliary tubercles or
granules, more or less serially disposed parallel to the margin adjacent to the
companion plate, and more numerous at the adcentral end of that margin.
Beyond this the surface of the ultimate plate is smooth, like that of the other
supero-marginal plates.

The abactinal area of the disk within the boundary of the marginal plates is
covered with hexagonal or polygonal plates or paxillar tabulz, which are small in
size generally, excepting the primary apical plates, which are comparatively very
large. All the plates have their surface covered with a fine, uniform granulation.
The primary apical plates have a small central area of low elevation, not higher
than if a number of granules had become coalesced—the structure being sugges-
tive of a tubercle in process of disappearance,—in other words, the scar left by a
tubercle which had existed in an earlier stage of growth (see Pl. XIV, fig. 4d).

Dimensions.—The example figured on Pl. XIV, fig. 4a, has the following

METOPASTER CORNUTUS. 55

measurements :—major radius, 20 mm.; minor radius, 15°5 mm; length of the
side, about 22 mm. The thickness of the margin is 7 mm. in a specimen whose
side is 23°5 mm.

Locality and Stratigraphical Position—The specimen described above and
figured on Pl. XIV, fig. 4a, is from the Upper Chalk, near Bromley. It is
preserved in the British Museum, and bears the register-number “ 46,776.’ Other
examples which I refer to the same species are also in the National Collection,
and one, if not more, was obtained from the same locality.

Remarks.—A fragmentary example which was figured in Dixon’s ‘ Geology and
Fossils of the Tertiary and Cretaceous Formations of Sussex,’ pl. xxi, fig. 8, and
was referred by Forbes to Goniaster (Goniodiscus) wncatus, now forms part of the
collection in the British Museum, and bears the registered number “ E 2577.”
This specimen appears to me to belong to the present species, and not to the true
Metopaster wncatus. I can scarcely think that Forbes would have intentionally
ranked the two forms as belonging to one species, and I have not thought it
needful on the sole grounds of this fragment having been figured to regard
Metopaster cingulatus as a dismemberment of Forbes’ Metopaster wncatus.

Metopaster cingulatus is readily distinguished from Metopaster wncatus by its
very short broad supero-marginal plates, by their more limited and more conical
tumidity, by their greater height as seen in the margin, as well as by their
different ornamentation. The ultimate plates are triangular, and do not present
the peculiar form characteristic of Metopaster wneatus. In many respects
Metopaster cingulatus appears to hold an intermediate position between Mitraster
rugatus and Metopaster uncatus.

8. Meropaster cornutus, Sladen. Pl. XIV, figs. 5a—5 d.

Body of small size. General form depressed. Abactinal surface slightly
concave, the extremity of the rays being directed slightly upward. Actinal surface
slightly convex. Marginal contour pentagonal, with the sides slightly lunate
and the extremity of the rays slightly produced. The major radius is proportional
to the minor radius as 100 : 74, the actual dimensions in the example described
being, major radius about 19 mm., minor radius about 14 mm. approximately.
The length of the side is 225 mm. The margin is moderately thick and
apparently well rounded.

56 FOSSIL ASTEROIDEA.

The supero-marginal plates are only two in number, counting from the median
interradial line to the extremity, or four from the tip of one ray to the tip of the
adjacent ray, exclusive of the odd terminal or “ ocular” plate in each case. They
form a broad border to the abactinal area of the disk. The supero-marginal plate
adjacent to the median interradial line is broader than long, the dimensions in
the example described being length about 3°25 mm., and breadth about 5°75 mm.
The abactinal surface of this plate is slightly convex in the median line of
breadth. The entire margin of the plate is surrounded by a very narrow
depressed border with very minute punctations, closely crowded and uniserially
disposed near the inner edge of the border, upon which a small miliary granulation
was previously borne. The general superficies of the plate is covered with minute
punctations irregularly disposed. The height of the two intermediate supero-
marginal plates as seen in the lateral view of the margin (Pl. XIV, fig. 56) is
scarcely half their length, and is less than half the height of the underlying infero-
marginal plates.

The ultimate paired plate is very much larger than the small intermediate
plate just described, being nearly three times as long, the actual measurement
being about 9°25 mm. in the example under notice. It is of a peculiar form, being
shaped somewhat like an irregularly formed elytron of an insect as seen from
above, the proximal margin being truncate where it joins the companion inter-
mediate plate, and the distal extremity obtusely rounded (Pl. XIV, fig. 5c). At
a point situated three-fourths of the distance from the proximal to the distal
extremity rises an abrupt truncate teat-like eminence. The peculiar form of the
plate as seen in the lateral view of the margin and the character of the eminence
just described will be better appreciated by reference to Pl. XIV, fig. 5d, than by
verbal description. The superficies of the ultimate plate is studded with rather
widely spaced punctations disposed in groups, which do not extend over the whole
of the area which falls in the lateral wall of the disk. As seen in the marginal
view of the test the ultimate is very much higher than the intermediate marginal
plates (see Pl. XIV, fig. 5d).

The odd terminal plate is unknown to me, no trace being preserved in the
example under notice.

In like manner the whole of the abactinal plating within the boundary of the
supero-marginal plates has been removed, excepting only a few isolated fragments
of plates out of position.

The infero-marginal plates are large and high. Four are preserved between
the median interradial line and the extremity, but the series is incomplete,
probably one or more being wanting at the distal end of the series. This would
increase the number eight now preserved on the whole side to ten or perhaps
twelve. The two infero-marginal plates on each side of the median interradial

MITRASTER. 57

line are much larger and higher than the other plates. The plate adjacent to the
median interradial line measures about 4 mm. in length and 4 mm. in height;
the next plate 5 mm. in length, and from 3°5 mm. to 3°75 mm. in height. The
surface of the plates is marked with small widely spaced punctations, and there is
a narrow depressed border round the entire margin of the plate, which is very
finely punctate.

The actinal area is unknown to me.

History.—The fossil delineated on Pl. XIV, fig. 5a, was drawn by Mr. A. H.
Searle under Dr. Wright’s instructions, but I regret that I have not been able to
find any trace of the specimen. I am therefore led to believe that the type
belonged to Dr. Wright’s private collection, which has been distributed since his
lamented death. Knowing by experience the extreme care and fidelity which
characterise all Mr. Searle’s work, I have ventured to describe the species from
his drawings alone, for it seemed undesirable to leave such an interesting form
without notice; and I am hopeful that the publication of the figure and the
description of its characteristic features will lead to the detection of the type. I
am unable to give any information as to the locality or stratigraphical position
from which the fossil was obtained.

Remarks.—The rather small size of this example and the small number of
supero-marginal plates—only four for the whole side of the disk—would not
unnaturally suggest at first sight that this was possibly animmature form. After
careful study, however, I do not consider such to be the case, or at any rate I am
unable to regard the fossil under notice as the young of any of the species with
which I am acquainted. The large and characteristically developed ultimate
supero-marginal plates in conjunction with the presence of a normal number of
infero-marginal plates, together with the fact that the size of the test is not less
than that of another perfectly characterised species, lead me to rank this as a
distinct species with little hesitation. The general proportions as well as the
character of the different plates, and the facies of the form as a whole, appear to
me to fully warrant this view.

Genus—MITRASTER, Sladen.

(Mirpa = a broad belt, or girdle.)

Body depressed and cycloid, or cyclo-pentagonoid in contour. Marginal plates
with co-ordinated granulose elevations and punctations, and a surrounding narrow

58 FOSSIL ASTEROIDEA.

depressed border with very minute punctations for the articulation of sete.
Supero-marginal plates few in number and all subequal in size, forming a broad
uniform border to the disk. Abactinal area covered with polygonal plates, some
of which may have stellate or substellate bases, and upon the tabulz are borne
small, more or less co-ordinated granules. Infero-marginal plates more numerous
than the supero-marginal series, and decreasing in size as they approach the
extremity; the surface marked with punctations, which may be co-ordinated in a
similar manner to that of those on the supero-marginal series, and may be associated
with granulose elevations. Actinal intermediate plates small, polygonal, covered
with uniform, crowded, shallow punctations, upon which granules were originally
borne. Armature of the adambulacral plates arranged in longitudinal lines, which
may be slightly oblique. Small entrenched pedicellarizs may be present occasion-
ally on the actinal intermediate plates.

Mitraster is characterised by its cycloid contour, by the equality in size of the
supero-marginal plates, which do not diminish towards the extremity, and by the
character of the ornamentation of the marginal plates, especially of the superior
series.

The main characters which distinguish this and the two preceding genera are
distinctly relative, and may be here conveniently compared. In Pentagonaster the
rays are more or less produced, the supero-marginal plates are more or less
numerous, and decrease in size as they approach the extremity of the ray,
and are devoid of a marginal border of sete. In Metopaster the rays are very
slightly produced, the contour being pentagonal and only slightly extended.
The supero-marginal plates are few in number, and do not decrease in size as they
approach the extremity, the ultimate paired plate being larger than the others.
All are furnished with a marginal border of setew. In Mitraster the contour
is cycloid almost to the obliteration of the pentagonal form. The supero-marginal
plates are few in number, but neither decrease nor diminish in size, being subequal
throughout ; and they are furnished with a marginal border of sete.

I consider that these differences indicate structural characters of sufficient
morphological significance to render the forms presenting them worthy of recog-
nition as distinct genera.

MITRASTER HUNTERI. 59

1. Mirraster Hounvert, Forbes, sp. PI. IX, figs. 3a—3e; Pl. XII, figs. 3a—se;
Pl. XV, figs. 3a—d b.

GoNIASTER REGULARIS, Mantell, 1844. Medals of Creation, vol. i,
p- 335, lign. 73 (non Linck).
GonrasTER (Gonropiscus) Hunrert, Forbes, 1848. Memoirs of the Geological
Survey of Great Britain, vol. ii,
p. 471.
= _ _ Forbes, 1850. In Dixon’s Geology and
Fossils of the Tertiary and Creta-
ceous Formations of Sussex,
London, 4to, p. 331, pl. xxi, fig. 1.
— Morris, 1854. Catalogue of British Fossils,

2nd ed., p. 81.

Astroconium Hunter, Dujardin and Hupé, 1862. Hist. Nat.
Zooph. Echin. (Suites 4 Buffon),
p- 399.

Gonyzaster (Gontopiscus) Hunrert, Forbes, 1878. In Dixon’s Geology of Sussex
(new edition, Jones), p. 365,
pl. xxiv, fig. 1.

Body of small size. General form depressed. Abactinal surface flat and
apparently depressed within the boundary of the supero-marginal plates, the
gibbosity of the latter and their centrally sloping surface giving a concave character
to the area generally in fossil examples, which was probably always more or less
present. Actinal surface flat or with a tendency to be slightly convex. Marginal
contour subcircularly pentagonoid, the sides being slightly convex or bulging out-
ward. The major radius is proportional to the minor radius as 100 : 90°47 approxi-
mately ; and the rays are not produced, the angles of the pentagon being obtusely
rounded. Margin thick and well rounded, the slope being more gradual on the
actinal surface.

The supero-marginal plates are three in number counting from the median
interradial line to the extremity, or six from the tip of one ray to the tip of the
adjacent ray, exclusive of the odd terminal or “‘ ocular” plate in each case. They
form a broad border to the abactinal area of the disk, of uniform breadth through-
out, which measures about 6°3 mm. at the median interradial line in an example
whose diameter (R + 7) measures about 40 mm. Excepting the ultimate plates
all the supero-marginal plates are of equal size and uniform, the breadth being
a little more than once and a half the length, the actual measurements in the

60 FOSSIL ASTEROIDEA.

specimen under notice being length 4 mm. and breadth 63 mm., in the plate
adjacent to the median interradial line. The length of the plate at the end which
falls in the margin of the disk is a shade greater than the adcentral or inner end,
and the plates are consequently faintly wedge-shaped, but so slightly that the
character is scarcely noticed at first sight. The ultimate plates, however, are
distinctly wedge-shaped, the length at the outer margin being a little greater
than that of the other marginal plates, while the length of the inner end is rather
less—often not more than one-half the length of the same end in the other plates.
The breadth of the ultimate plates is the same as that of all the supero-marginal
plates ; and the corresponding plates of the two adjacent sides touch one another
throughout, the line of junction falling in the median line of the ray. The abactinal
surface of the plates is distinctly convex, and the character is more conspicuously
emphasised by the plate becoming rapidly gibbous on the outer half, the outer
side of the eminence forming the rapid bend to the lateral wall of the plate.
The height of the plates as seen in the margin is usually equal to, or even
rather greater than their length, but may occasionally be less. There is no
diminution in height as the plates approach the extremity of the ray, and the
ultimate paired plate has a tendency to appear even a trifle higher and more
gibbous than the others, but the character is derived probably more from the
position in which the plate sometimes is than from an actual increase in size or
gibbosity. The abactinal surface of the plates is covered with coarse tuberculiform
mammillations which gradually die out before reaching the apex of the gibbosity.
In the interspaces between the eminences are small, more or less widely spaced
punctations, and these extend over the whole surface of the plate, and are
consequently present on the outer portions as well as on the lateral wall. There
is a narrow depressed border round the entire margin of the plate, which is very
minutely punctate (see Pl. XII, fig. 3c). In smaller examples there often appears
to be only one or two rows of punctations. The ornamentation of the ultimate
paired plates is precisely similar to that on the other supero-marginal plates.

The odd terminal or “ocular” plate is very small, and, so far as I can make
out, resembles superficially a short truncated cylinder which protrudes somewhat
cannon-like from a small triangular space left by the ultimate paired plates similar
to what I have already described in Metopaster.

The abactinal area of the disk within the boundary of the marginal plates is
covered with comparatively large polygonal plates, with closely crowded, rather
coarse, uniform granulations, upon which miliary granules or spinelets were
previously borne. The primary apical plates are remarkably large, and the plates
in the interradial areas are larger than the plates in the radial areas. Of the
plates in the radial areas at least the median series and two series on each side
have bases of a six-rayed, substellate form. These are admirably seen in an

MITRASTER HUNTERI. 61

example preserved in the British Museum bearing the register-number “ 46,772,”
in which the inner surface of the abactinal floor is exposed.

The madreporiform body, in the examples in which I have detected its
presence, appears to be small and subtriangular, and is marked with very coarse
striations.

The infero-marginal plates are five in number, counting from the median inter-
radial line to the extremity,—that is to say, there are ten for the whole side of the
disk as against six in the supero-marginal series. ‘The length of the two innermost
plates on each side of the median interradial line is a little greater than that of
the corresponding superior plates; the breadth of the plate adjacent to the
median interradial line is 7 mm. in the specimen bearing the British Museum
register-number “46,766,” and is a little greater than the breadth of the corre-
sponding superior plate as seen in the example bearing the British Museum register
number ‘ E 2583.”

The second plate, counting from the median interradial line, is a little less
broad, and the third is slightly more diminished in breadth, and its adcentral
margin merges with a sweeping curve into the lateral distal margin, which gives
the plate a more or less cuneiform shape. A large portion of the distal lateral
margin of this plate abuts on the adambulacral plates. The fourth infero-marginal
plate is very small and triangular in form, with the apex directed adcentrally,
and with one side abutting entirely on the adambulacral plates. ‘The fifth plate is
smaller still. In consequence of the triangular shape of the fourth and fifth
plates the greatest length of the third plate is opposite the apex of the fourth
plate, and the length of the third plate gradually diminishes up to the outer
margin. ‘The breadth of the third plate is 5°6 mm., and that of the fourth plate
is only 3°2 mm.

The surface of the infero-marginal plates, which is more or less plain, is orna-
mented with small, widely spaced,and more or less equidistant punctations, and
there is no trace of the tubercular mammillation present on the surface of the
supero-marginal plates. On the inner third of the plate the punctations are more
closely placed, and they have the appearance of falling into a more or less
distinct reticulated arrangement (see Pl. XII, fig. 3d). A narrow depressed
border surrounds the entire margin of each plate, which is very minutely
punctate.

The adambulacral plates, which are small, are broader than long, and have
their surface traversed by several ridges placed slightly obliquely, but I am unable
to define the armature.

The actinal intermediate plates, which are somewhat large in relation to the
size of the disk, are rhomboid or polygonal in outline. Their surface is covered
with small, closely crowded, uniform pits, upon which miliary granules or spinelets

9

62 FOSSIL ASTEROIDEA,

were previously borne (see Pl. XII, fig. 3¢). A small entrenched pedicellaria may
be present on an occasional plate here and there.

Dimensions.—The example figured on Pl. XII, fig. 3a, has a major radius of
about 30 mm., and a minor radius of about 28mm. The length of the side is about
36mm. The thickness of the margin is about 10 mm. In the specimen given on
Pl. XV, fig. 4a, the major radius measures 21 mm., and the minor radius 19 mm.
approximately. The length of the side is about 25 mm.

Locality and Stratigraphical Position.—This species is a characteristic Upper
Chalk form. The majority of examples are from Kent and Sussex. A fine series
from Bromley is preserved in the British Museum.

History.—This species was referred to by Dr. Mantell under the name of
Goniaster regularis. The specific name was, however, already preoccupied—at
least in literature—for an existing starfish; and although the latter is not now
recognisable, no doubt can possibly exist that Mantell’s fossil species is certainly
a different thing from the starfish to which Linck gave the name of Pentagonaster
regularis.

Forbes was the first to describe the species under the name of Goniaster
(Goniodiscus) Hunteri, and his type-specimen, which is figured in Dixon’s ‘ Geology
and Fossils of the Tertiary and Cretaceous Formations of Sussex,’ pl. xxi, fig. 1,
formed part of John Hunter’s Collection, now preserved in the Museum of the
Royal College of Surgeons. The examples illustrated in the present work are all
preserved in the British Museum.

Variation.—There are at least two well-preserved examples in the collection of
the British Museum which I consider to be varieties of this species. They are
characterised by the presence of only four supero-marginal plates on each side of
the disk, exclusive of the odd terminal plates, as against six plates in typical
examples. Beyond this difference in number and the relatively greater length of
the supero-marginal plates in proportion to their breadth, I can indicate no
character worthy of being noted which would distinguish the examples in question
from the typical form of Mitraster Hunteri. The specimens under notice measure
35 mm, and 30 mm. in diameter (R-+7) respectively. Both are from the Upper
Chalk, one from near Bromley, the other being only labelled ‘‘ Kent.” Figures
of the first-mentioned are given on PI. IX, figs. 3a—3e.

MITRASTER RUGATUS. 63

2. Mrrraster rucatus, Forbes, sp. Pl. XVI, figs. 3a—é d.

GontasteR (Gonropiscus) ruGaTUS, Forbes, 1848. Memoirs of the Geological
Survey of Great Britain, vol. ii,
p- 471.

— _ _ Forbes, 1850. In Dixon’s Geology and
Fossils of the Tertiary and Creta-
ceous Formations of Sussex, London,
4to, p. 380, pl. xxi, figs. 2, 2*;
pl. xxiii, fig. 15.

— _ — Morris, 1854. Catalogue of British Fossils,

2nd ed., p. 81. ‘

ASTROGONIUM RUGATUM, Dujardin and Hupé, 1862. Hist. Nat.
Zooph. Echin. (Suites 4 Buffon),
p. 399.

Gontaster (Gontopiscus) ruGaAtuS, Forbes, 1878. In Dixon’s Geology of Sussex
(new edition, Jones), p. 364, pl. xxiv,
figs. 2, 2*; pl. xxvi, fig. 15.

Body of small size. General formdepressed. Abactinal surface flat. Actinal
surface also flat. Marginal contour pentagonal with almost straight sides. The
major radius is proportional to the minor radius as 100: 75. The actual
measurements in an example from the Upper Chalk of Gravesend, preserved in
the British Museum, being, major radius, 17°5 mm.; minor radius, 13°75 mm.
The rays are not produced beyond the contour of a true pentagon, and the radial
angles are not rounded. Margin moderately thick, the lateral wall being almost
vertical in consequence of the rapid bend from the abactinal and actinal surfaces ;
the bend to the actinal surface is, however, rather more gradual.

The supero-marginal plates are three in number, counting from the median
interradial line to the extremity, or six from the tip of one ray to the tip of the
adjacent ray, exclusive of the odd terminal or ‘‘ ocular” plate in each case. They
form a broad border to the abactinal area of the disk, of uniform breadth through-
out, which measures 6 mm. at the median interradial line in an example whose
diameter (R+ 7) measures about 31 mm. Excepting the ultimate plates, the
supero-marginal plates are approximately equal in size, the breadth being nearly
twice the length, the actual measurements in the above-mentioned specimen of
31 mm. in diameter being, length 3°5 to 3°6 mm., and breadth 6 mm. in the plate
adjacent to the median interradial line. The corresponding plate in another
example has a breadth of 5 mm. and length 2°75 mm., and in a third, breadth
5°5 mm. and length 2°75 to 2°8 mm. In the plate adjacent to the median inter-

64 FOSSIL ASTEROIDEA.

radial line the length at the outer and at the adcentral margin is equal, but in the
second plate the length at the adcentral margin is greater than at the outer
margin in consequence of a slight extension of the plate to make up for the
diminution in size of the ultimate plate, the distal lateral margin of the second
plate appearing to be slightly hollowed out for the reception of the ultimate plate,
and the corresponding second plates of two adjacent rays consequently touch at
their adcentral ends. The ultimate plate is triangular or wedge-shaped in form,
and its breadth is less than that of the other supero-marginal plates, and the
corresponding plates of the two adjacent sides touch one another thoughout, the
line of junction falling in the median line of the ray. ‘The ultimate plate is not
unfrequently a little longer at the outer margin than the other supero-marginal
plates.

The abactinal surface of the plates is regularly convex along the line of
breadth, almost resembling the segment of a cylinder, and no special gibbosity is
developed. The abactinal area and the lateral area of the plate form a right angle,
and the uniting curve is short and abrupt. The height of the supero-marginal
plates, as seen in the margin, is less than their length, the height being 2°25 mm.
where the length is 2°75 mm. There is no diminution in height as the plates
approach the extremity of the ray, and no special prominence is noticeable in the
ultimate plate. The abactinal surface of the plates is ornamented with irregular
and frequently elongate eminences or ridges, which are usually disposed trans-
versely or in the direction of the length of the plate. Two or three series may
be present, as in Pl. XVI, fig. 4, or the lear ridges may be bent at a right
angle, or otherwise curved—a character which is probably the result of several
eminences being merged together, by which means a peculiar hieroglyphic-like
marking is produced, which fancifully resembles Chinese writing to a certain
extent (see Pl. XVI, fig. 5c). This character is emphasised in weathered
specimens. The ornamentation covers the whole abactinal area of the plate, but
does not extend upon the lateral wall. Sometimes the form of the eminences is
more rounded and tuberculiform, as in Mitraster Hunteri; this is distinctly the
case in one example, and minute punctations are present in the channels which
intervene, but it is to be noted that the tubercles and pits cover the whole
abactinal area of the plate, and that there is no gibbosity, with its outer flank
devoid of tubercles, as in Mitraster Hunteri. There is a very narrow depressed
border round the entire margin of the plate, from which the regular convexity of
the plate rapidly rises. The border is very minutely punctate, and there appears
to be usually only asingle regular lineal series of punctations. The ornamentation
of the ultimate paired plates is similar to that on the other supero-marginal
plates, but is more rounded and tubercle-like even in an example in which the
broad lineal ornamentation occurs on the other plates.

MITRASTER RUGATUS. 65

The odd terminal or so-called “ ocular” plate must be very small in order to
fit the space left by two adjacent ultimate supero-marginal plates; but I am
unable to describe the plate, as I have only found a trace of it in one instance, and
have not seen a perfect one preserved in any of the examples of the species I have
examined.

The abactinal area of the disk within the boundary of the marginal plates is
paved with large polygonal plates, whose surface is marked with a very minute,
closely crowded uniform granulation, upon which small miliary granules were
previously borne. Traces of these granules are still occasionally present in some
examples. The abactinal plates in this species are all large in relation to the size
of the disk, and the primary apical plates are much larger than any of the others.
The primary apical plates and several of the other plates have a central, small,
low, irregular eminence, such as might be formed by the merging together and
partial grinding down of a number of granules. I am unable to explain this
structure, or to consider it as associated with a pedicellarian apparatus, of which
I see no trace. It may possibly mark the remains of a tubercle which had
existed in an early stage of the animal’s life, but had disappeared and been
outgrown at a later stage.

The madreporiform body, which is very small, is circularly subtriangular in
shape, and is marked with fine regularly radiating striations. In the example
under notice two large plates surround two-thirds of the circumference of the
madreporite. In this specimen the madreporiform body measures 1°5 mm. in
diameter, and the largest primary plate is 3°2 mm. in diameter.

The infero-marginal plates are five in number, counting from the median
interradial line to the extremity,—that is to say, there are ten for the whole side of
the disk as against six in the supero-marginal series. The length of the two
innermost plates on each side of the median interradial line is greater than that of
the corresponding supero-marginal plates, consequently the second infero-marginal
plate extends a little way beneath the ultimate supero-marginal plate, and the re-
maining three infero-marginal plates are all under the ultimate supero-marginal
plate. In some examples I am inclined to think that probably only four infero-
marginal plates were present, counting from the median interradial line to the
extremity, in which case the edge of the second and the remaining two plates were
under the ultimate supero-marginal plate.

I have unfortunately not seen any example of this species in which the actinal
surface of the disk is exposed, I am therefore unable to give the dimensions of the
infero-marginal plates on the actinal surface, or to describe their ornamentation.
For the same reason I am prevented from offering any remarks on the adam-
bulacral and actinal intermediate plates.

10

66 FOSSIL ASTEROIDEA.

Dimensions.—The example figured on Pl. XVI, fig. 3 a, has a major radius of
about 17°5 mm., and the minor radius is about 13°75 mm. The length of the side
is about 20 mm. The fragmentary type-specimen figured by Forbes, which is
drawn on Pl. XVI, fig. 5a, was probably about the same size, judged by compu-
tation of the half-side.

Locality and Stratigraphical Position.—One of the type specimens figured by
Forbes is stated to have been obtained from the Upper Chalk of Wiltshire, but
the second specimen, which is now preserved in the British Museum, bears no
record of any locality. Forbes also records the species from Kent and Sussex.
Authentic examples from the Upper Chalk from Gravesend and “ Kent” are
preserved in the British Museum.

History.—This species was described by Forbes in his memoir “On the
Asteriade found fossil in British Strata”? (‘Mem. Geol. Surv.,’ vol. ii, p. 471,
1848), and figures of two examples were given in Dixon’s ‘ Geology and Fossils of
the Tertiary and Cretaceous Formations of Sussex,’ London, 1850, pl. xxi,
figs. 2, 2*; pl. xxii, fig. 15. The latter specimen is now preserved in the
British Museum (register-number ‘‘ HE 2585”’), and is illustrated on Pl. XVI,
fig. 5a, of the present work. I have not been able to find any trace of the other
example figured by Forbes, which originally formed part of the collection belonging
to the late Mr. Channing Pearce. Forbes states that it was found in Wiltshire.

Remarks.—Although at first sight the differences between Mitraster Hunteri
and Mitraster rugatus appear well marked, I am not perfectly satisfied as to the
species being altogether independent. When the types alone are examined there
appears to be no need for any doubt upon this question. But examples occur
which are exceedingly difficult to determine on account of presenting features
which seem to break down some character which has been regarded as diagnostic
of the other species. In illustration of this difficulty I have drawn on Pl. XVI,
fig. 3a, an example which I have ranked under Mitraster rugatus, but which
presents considerable superficial resemblance to Mitraster Hunteri in the character
of the ornamentation of the supero-marginal plates. I believe, however, that the
proportions of the plates, the absence of any abactinal gibbosity, and the exten-
sion of the tuberculation over the whole abactinal area constitute, inter alia, a
justification for regarding the example as Mitraster rugatus.

Turning, on the other hand, to a series of Mitraster Hunteri, considerable
variation is to be noted in the relative length and breadth of the supero-marginal
plates, as well as in the amount of gibbosity developed on the abactinal area of
the plate. In such an example as that figured on Pl. XV, fig. 3 a, the proportions

Paleontographical Society, 1905.

A MONOGRAPH

ON THE

BRELTISH. POSS i

ECHINODERMATA

FROM

THE CRETACEOUS FORMATIONS.

VOLUME SECOND.
THE ASTHROIDEA.

BY

W. K. SPENCER, B.A., F.G.S.

PART THIRD.

Paces 67—90; Prares XVII—XXVI.

LONDON:
PRINTED FOR THE PALHONTOGRAPHICAL SOCIETY.

L905.

MITRASTER COMPACTUS. 67

of length to breadth are in the case of the interradial supero-marginal plates as
56 mm. is to 3°7 mm. Further, the tuberculation in this specimen may or may
not extend over the whole of the abactinal area, the variations being on adjacent
plates, and the abactinal gibbosity is not strongly developed. In the example
figured on PI. IX, fig. 3, the proportions of length to breadth in the case of the
interradial supero-marginal plates are as 4°5 mm. is to 3°6 mm.; the granulated
areas more generally stop short of the distal edge of the supero-marginal plate and
the abactinal gibbosity is well pronounced. In view of these considerable variations
it is difficult to refer these forms to more than one species.

3. Mirraster compactus, Forbes, sp. Pl. XVII, fig. 2; Pl. XXVI, figs. 3, 3a, 3b, 3c.

GontasTeR compacts, Forbes, 1848. Mem. Geol. Surv. Gt. Brit., vol. ii, p. 468.
_- Forbes, 1850. In Dixon's Geology and Fossils of the Ter-
tiary and Cretaceous Formations of Sussex, London,
p- 333, pl. xxii, fig. 3.
Astrocontum compactum, Dujardin et Hupé, 1862. Hist. Nat. Zooph. Echin.
(Suites a Buffon), p. 399.
GontasrER compactvs, Forbes, 1878. In Dixon’s Geology of Sussex (new edition,
Jones), p. 366, pl. xxii, fig. 3.

Specific Characters—Outline pentagonal, slightly cycloidal. Eight supero-
marginal plates on each side of the pentagon. Supero-marginal plates form a broad
margin, and the breadth of each is about four times its length. Base of ultimate
paired supero-marginal plate twice as long as the other, more proximal, supero-
marginal plates. Ten corresponding infero-marginalia.

Matevial—Only one specimen of this species is known. This formed a portion
of Mr. Willett’s collection and is now preserved in the Brighton Museum, It
apparently escaped the observation of the late Dr. Wright, for the figure on Plate
XVII is copied from that in Dixon’s ‘ Geology of Sussex.’ As this figure is slightly
inaccurate I have had it redrawn and further details added on Plate XXVI.

Description.—The dorsal surface of the dise is covered with a number of small,
subequal, closely-fitting plates. It is considerably sunk in the specimen known.

The supero-marginalia bounding the dise form a uniform margin 5°15 mm.
broad. They are eight in number along each side, exclusive of the odd terminal
or ocular plates. The six middle plates are about 1°2 mm. long. Their breadth
is rather more than four times their length, a feature which distinguishes them

ll

68 FOSSIL ASTEROIDEA.

from Mitraster Hunteri and M. rugatus. A further distinctive feature is the size
of the distal paired plate. This plate is triangular. The base of the triangle
measures 2‘7 mm., giving the plate twice the length of the more proximal plates.
The plate is gibbous at its outer extremity as in M. Hunteri. All the plates are
ornamented with a single or double marginal row of small spinelets. The ocular
is a small conical plate barely visible in abactinal view. It fits imto notches on
the lower surface of the distal paired plates, and is, as usual, notched on its
inferior surface for the purpose of protecting the unpaired terminal tube foot.

The median infero-marginal plates are rather longer than the corresponding
members of the superior series. The first two, reckoning from the median inter-
radial line, are 1°85 mm. long, and 8°5 mm. broad. The third is only 1°8 mm. long
and not quite as broad. The fourth has approximately the same length but is
subtriangular in form. The fifth is a small triangular plate. Two infero-marginals
and a portion of a third are situated underneath the distal paired supero-marginal
plate.

The ventro-lateralia visible are small hexagonal plates covered with a fine uniform
granulation. The adambulacralia are small oblong plates. The margin of the dise
is very abrupt, but the transition from infero-marginalia to the actinal surface is
more gradual than that of the supero-marginalia to the upper surface. A number
of small granules are irregularly distributed between the plates.

Remarks.—Unfortunately, the specimen is slightly distorted, so that the pro-
nouncedly cycloidal appearance in the figure is partially due to the unnatural
position of the marginal plates, which has brought the inferior series into the dorsal
view. The supero-marginal plates appear to have been straight and the inferior
series but slightly cycloidal. This, together with the large comparative size of
the ultimate paired plate, would bring the species very near to the genus Meto-
paster. Forbes remarked upon the fact that it appeared to be intermediate between
Goniaster (Metopaster, Sladen) wncatus and Goniaster (Mitraster, Sladen) rugatus. I
have therefore considerable doubt as to the validity of the separation of these two

genera.

Locality and Stratigraphical Position —Upper Chalk of Haughton, Sussex.

COMPTONIA COMPTONT. 69

Genus—COMPTONTA, Gray.

Gray, 1840. Annals and Magazine of Natural History, vol. vi, p. 278.
— 1866. Synopsis of Starfishes in the British Museum.

Body depressed, with produced tapering rays. Disc covered abactinally and
actinally with numerous polygonal plates which possess a uniform granulation.
Marginal plates numerous. Supero-marginal plates equal in number to the infero-
marginals, and forming a moderately broad border to the disc. Infero-marginal
plates (as well as all other plates) devoid of spines. Radialia present throughout
length of ray.

This genus apparently differs from Sfellaster only in the absence of spines on
the infero-marginalia. It is thus similar to, as well as prior to Ogmaster (von
Martens, 1865) and Dorigona (Gray, 1866).

1. Compronta Comptont, Forbes, sp. Pl. XVII, figs. 3, 3a, and 3b; Pl. XVIII,

figs. 2, 2a, 2b, 2c, 2d.

STELLASTER CoMPTONI, Forbes, 1848. Mem. Geol. Surv. Gt. Brit., vol. ii, p. 476.

— — Forbes, 1850. In Dixon’s Geology and Fossils of the
Tertiary and Cretaceous Formations of Sussex,
pl. xxu, fig. 8, p. 335.

— — Dujardin et Hupé, 1862. Hist. Nat. Zooph. Echin. (Suites
i Buffon), p. 408.

-- _ Forbes, 1878. In Dixon’s Geology of Sussex (new edition,
Jones), pl. xxii, fig. 8, p. 368, 370.

Specific Characters.—Disc large and interbrachial arcs wide, giving the dise a
distinctly pentagonal appearance. Major radius rather more than twice the length
of the minor radius. Arms elliptical in cross section. Large valvate pedicellariz
present.

Material.—Two specimens (the two cotypes) of this species are in existence.
One (formerly in the Bowerbank Collection) displays the actinal aspect (Pl. XVII,
fig. 3), and is preserved in the British Museum of Natural History (34311).
The other (Pl. XVIII, fig. 2), which shows the dorsal aspect, is preserved in the
Northampton Museum. This is the specimen figured in Dixon.

Description —The large pentagonal disc is covered dorsally with numerous

FOSSIL ASTEROIDEA.

NTI

small closely-fittimg plates. In the radial areas these plates are polygonal and are
about 1‘8 mm. in diameter. In the interradial areas the plates measure only 1 mm.
in diameter and are rhomboidal. All the plates are covered with a fine uniform
granulation (Pl. XVIII, fig. 2a). Upon very many of the plates are valvate
pedicellariz. Post-mortem changes have produced a sinking in of the plates over
the interradial areas. Depressions, doubtless due to similar causes, appear in
recent forms when dried, as also in C. elegans. I have been unable to distinguish
either the madreporite or the anus.

The arms are not so much produced as in C. elegans.
R:r:: 62mm. : 29mm. in the specimen at Northampton.
R:r:: 556mm, : 258mm. in the British Museum (Natural History)
specimen.
The width of the arms at the sixth supero-marginal (reckoning from the inter-
radius) is 11°5 mm.

The supero-marginalia are oblong in shape. In the interradial areas they are of
fairly constant size, measuring 5°2 mm. in breadth and 2mm. in length. They
diminish in size distalwards. They are eighteen in number, and often bear one
or more valvate pedicellariz. The margin is rounded and is about 8 mm. high.

The infero-marginalia are equal in number and similar in appearance to the
superior series.

The actinal interradial areas are large and filled proximally with a number of
small rhomboidal plates about 1:2 mm. in average breadth. The more distal
plates are crowded, smaller, and polygonal in appearance. ‘Traces of a fine
granulation are visibie.

The adambulacrals are a series of small oblong prominent plates. The largest
are about 1°6 mm. in length and 1:2mm. in breadth. Remains of their armature
are still present. The mouth-angle plates are small and but slightly prominent.
They also bear traces of armature. Valvate pedicellariz are scattered apparently
irregularly over all these various plates.

Locality and Stratigraphical Position—Upper Greensand of Blackdown.

Remarks.—Forbes considered this species was equivalent to Asterias Schultzit,
Roemer.’ In this latter species, however, the superomarginalia meet across the
dorsal surface of the ray, which would disprove Forbes’ statement.

! Roemer, ‘ Versteinerungen des Norddeutschen Kreidegebirges,’ pl. vi, fig. 21.

COMPTONIA ELEGANS. 71

2. CoMPTONIA ELEGANS, Gray. Pl. XVII, figs 4 and 4a.

CoMPTONIA ELEGANS, Gray, 1840. Ann, & Mag. Nat. Hist., vol. vi,
p- 278.

STELLASTER ELEGANS, Forbes, 1848. Mem. Geol. Surv. Gt. Brit., vol. ii,
p. 476.

a _- Forbes, 1850. In Dixon’s Geology and Fossils of
the Tertiary and Cretaceous Formations
of Sussex, London, p. 336, pl. xxii, fig. 9.
CoMPTONIA ELEGANS, Morris, 1854. Catalogue of British Fossils, 2nd ed.,
p- 50.
— — Dujardin et Hupé, 1862. Hist. Nat. Zooph.
Echin. (Suites 4 Buffon), p. 408.
STELLASTER ELEGANS, Forbes, 1878. In Dixon’s Geology of Sussex
(new edition, Jones), pp. 369, 370,
pl. xxu, fig, 9.

Specific Characters.—Dise strongly convex, covered with small polygonal
plates. Actinal interradial areas large. Arms well produced, the major radius
being at least three times as long as the minor radius. Interbrachial ares
paraboloid.

Material—The specimen figured by Dixon, at that time in the Bowerbank
Collection, is now preserved in the British Museum of Natural History (E. 2567).
Both dorsal and ventral aspects are exposed. Another specimen showing an
impression of the ventral surface exists in the Oxford University Museum.

Dixon’s specimen, however, can hardly be the type, since Gray (1840) stated that
the specimens described by him were in the British Museum or in the collection
of the Zoological Society. Forbes (1848) refers only to specimens in the British
Museum and the collection of the Marquess of Northampton. No part of the
Bowerbank Collection is known to have come to the British Museum before 1865.
The type specimen therefore must be either lost or still unrecognised in the
national collection. Since it was never figured it could never be identified with
certainty. It is therefore advisable to take the specimen KH. 2567 as type.

Description.—The dise is high in the central and radial regions. In the inter-
radial areas, however, post-mortem changes have caused a collapse of the test
and the consequent production of deep triangular depressions. The plates
covering the disc are minute, polygonal, and closely fitting. The centrale is the

72 FOSSIL ASTEROIDEA.

only plate of the dorsal surface which is larger or more conspicuous than the
remainder; all are covered with a minute uniform granulation of a quite charac-
teristic appearance. The anus is almost central in position. It is surrounded
by a circlet of plates, amongst which is the centrale. The madreporite is, as
usual, situated in the next (clockwise) interradius, almost halfway between the
centrale and the margin. It is a triangular plate, the apex of the triangle being
a markedly acute angle.

The arms are well produced. R:r::30mm.+:9 mm. Their breadth at the
base is 9mm. Radialia, adradialia, and dorso-lateralia extend into the base of the
arms. The dorso-lateralia soon disappear, but the adradialia persist as far as the
seventh or eighth supero-marginal plate. When the adradials disappear the radialia
become larger. They are at this point 1 mm. broad and 1:1 mm. long and
therefore appear almost square.

The supero-marginalia are oblong plates of curiously uniform size in the portions
of the specimen preserved. They are 1°6 mm. long and from 1:2mm. to 1:3 mm.
broad. The infero-marginalia are of the same length and are opposite to the
supero-marginalia. In lateral view the supero-marginalia appear higher than the
inferior series. Both series are ornamented with a number of small, fine granules
which are uniformly distributed over their surfaces.

The ventral surface is concave. The ventro-lateral plates are rhomboidal in
the region of the mouth. .They become polygonal and crowded as they approach
the margin. Some of these plates extend into the base of the arms. Around the
edges of the plates spinelets are visible. The spines of the adambulacral plates are
still present. Unfortunately, it is not possible to make out their exact distribution.
The mouth-angle plates are not prominent.

There is no trace in this species of such valvate pedicellarize as characterise
C. Comptoni.

Remarks.—Gray compared this species with Celuster, Agassiz.! The rather
vague diagnosis of Calaster given by Agassiz renders exact identification impossible.

Locality and Stratigraphical Position—Upper Greensand of Blackdown. Also
in the Upper Greensand at Folkestone (observed by Forbes).

‘ Agassiz, ‘Annales des Sciences Naturelles,’ 1837. Translated in ‘Annals and Magazine of
Natural History,’ vol. 1, 1838.

PENTAGONASTER ROBUSTUS. 73

Genus—NYMPHASTER, Sladen, 1885. (See p. 14.)
4, NYMPHASTER RADIATUS, n. sp. Pl. XXV, figs. 1, la, 1.

Specific Characters—Arms very much produced. R:r:: 150 mm. : 10 mm,
Supero-marginalia in contact almost the whole length of arm.

Material.—The only specimen of this species, formerly in the collection of Mr.
J. Starkie Gardner, is preserved in the British Museum of Natural History (KE. 375).
The plates have, unfortunately, disappeared from the dise. Practically all that
remains is the greater portion of one arm.

Description.—At the base of the arm the supero-marginalia are oblong. Each
measures 2°8 mm. in breadth, 2 mm. in length, and 3 mm. in height. Distally
these plates become almost square. They are ornamented by small granules
which tend to run together transversally to the length of the arm (Pl. XXV,
fic. 1), and articulations for spines and deep depressions for pedicellariz are also
present. They are about twenty-five in number. The infero-marginalia equal
in number and size and oppose the supero-marginalia. Further they are about
the same height in marginal view.

The breadth of the arm at the fourth supero-marginal is 6 mm.

Stratigraphical Position—Lower Chalk. Locality uncertain.

Genus—PENTAGONASTER, Linck. (See p. 24.)
3. PENTAGONASTER RoBUSTUS, n. sp. Pl. XXI, figs. 2, 2a.

Specific Characters.—Dise covered with small rounded plates. Margin high.
Rays short, high and robust. R:r::21°5mm.:9°9 mm. The supero-marginal
plates meet along the median line throughout almost the whole length of the arm,
and form a broad border to the dise. Interbrachial ares paraboloid.

Material—The only specimen of this species is the one here described, formerly
in the Mantell collection and now preserved in the British Museum of Natural
History (48085). The locality from which it was derived is stated rather

74 FOSSIL ASTEROIDEA.

vaguely as Upper Chalk, Sussex. The specimen is somewhat imperfect, consisting
only of the dorsal view of three arms and a portion of the disc.

Description.—The disc appears to have been covered on its dorsal surface by a
large number of small, rounded, closely-fitting plates. Mostly they are subequal
in size and have an approximate diameter of 2 mm. An uncertain number of
even smaller granule-like plates exist scattered between these.

Both radialia and adradialia extend into the base of the arms, but only as far
as the third supero-marginal plate, counting from the median interradial line.
The arms themselves are short and high. The height of the specimen in the
interradius is 9°2mm. From this pomt the height gradually diminishes to the
extremity of the ray, where it is 5 mm.

The supero-marginalia are about twelve in number. They form a_ broad
margin to the disc and rays. Hach supero-marginal plate is high, and is very
convex dorsally. Hence every plate is very distinct. The six proximal supero-
marginals diminish only slightly in size distalward along the ray. The next six,
however, diminish much more rapidly. The supero-marginal nearest the inter-
radius has the followme measurements: height, 5°75 mm.; breadth, 4°5 mm. ;
length, 3 mm.

The ocular plate has broken away, and there is no trace of a madreporite.

The infero-marginalia alternate with the supero-marginal series. They are
not so high and much squarer in appearance. They decrease in size much more
rapidly than the upper series (see Pl. XXI, fig. 2). The infero-marginal plates,
nearest the interradius, measure 4°5 mm. high and 3:2 mm. im length. Any orna-
ment that may have existed has disappeared from all parts.

Locality and Stratigraphical Position —Upper Chalk, Sussex.

4, PENTAGONASTER oBTUSUS, Forbes, sp. Pl. XXHL, figs. 1, 1 a, 1b, 2, 2a, 3a, 36,

o¢, 38a, 3 e, 37, 3g.

OREASTER OBTUSUS, Forbes, 1848. Mem. Geol. Surv. Gt. Brit., vol. ii, p. 468.

= — Forbes, 1850. In Dixon’s Geology and Fossils of the Tertiary
and Cretaceous Formations of Sussex, London, p. 330,
pl. xxi, fig. 12.

= — Dujardin et Hupé, 1862. Hist. Nat. Zooph. Echin. (Suites a
Buffon), p. 389.

= — Forbes, 1878. In Dixon’s Geology of Sussex (new edition,
Jones), pl. xxi, fig. 12, pp. 364, 370.

PENTAGONASTER OBTUSUS.

JI
or

Specific Characters.—Dise slightly convex. Majority of the plates covering
the disc of subequal size and closely set. R:r::25 mm.:12mm. Extremities
of arms obtuse. Interbrachial ares only slightly rounded, giving the dise a
distinctly pentagonal appearance.

Material—The two extremities of the arms from which Forbes originally
described the species when in the Dixon collection, are now preserved in the
British Museum (Natural History}. They are not on the same slab of chalk as
represented by Forbes, but are and probably always have been two independent
specimens, E. 5038 (Pl. XXII, figs. 3b, c,d), and E. 5039 (fig. 3a). More complete
specimens have since been added to the collection (40400, Pl. XXII, fig. 1 ex
J. Simmons’ Coll., and 35481, Pl. XXII, fig. 2, ee H. W. Taylor’s Coll.). Two

extremities of arms are also preserved in the Brighton Museum.

Description.—The disc is high and distinctly pentagonal. At the edge of the
disc the dorsal covering plates are crowded and polygonal in appearance. Towards
the centre they become slightly smaller and rounded. The average diameter of
these plates is 1°7 mm.

The madreporite is subcentral in position. It is about the same size as the
other plates of the dise and is pentagonal in shape (Pl. XXII, fig. 1a).

The arms are stoutly built. A triple row of polygonal plates enters their
bases. The adradial series soon disappears, leaving the single radial series, which
appears to persist until it reaches that part of the ray which is obtuse. From this
point the supero-marginal plates may or may not be adjunct up to the end of the
ray. Considerable variation appears to exist as to this point in the single
specimen examined. The arms are distinctly obtuse in their distal half. This has
given the species its name.

The supero-marginalia form a rather broad border to the disc. There
are nine supero-marginals from the median interradius to the extremity of the
ray. Towards the end of the ray the plates of this series become narrower, more
oblong in appearance, and distinctly convex.

The infero-marginalia are equal in number and situated generally alternating
with the supero-marginal series. They are distinctly square in shape, especially
at the obtuse extremities of the ray. Their ornament is in some specimens not
so coarse as that of the supero-marginals.

The actinal interradial areas are very distinct and are occupied by four series
of ventro-lateral plates. As usual, the actinal plates near the mouth are larger and
more rhomboidal than the distal plates. Ventro-lateral plates only extend

throughout about a quarter of the length of the arm.
12

76 FOSSIL ASTEROIDEA.

The mouth-angle plates are not prominent. The adambulacral plates possess
a triple row of spines.

Dimensions.—Specimens 35,481 and 40,400.—The greatest width of the ray
varies from 8°3 to 6°5 mm., and the least width from 7-2 to 5°7 mm. The supero-
marginal plates are 3°2 mm. broad near the interradii.

Specimen E. 5038.—Greatest width of ray 10°5 mm.
Specimen in Brighton Museum.—Greatest width of ray 9°2 mm.

Locality and Stratigraphical Position.—Upper Chalk, Lancing, Sussex, and
also from the Upper Chalk of Kent.

Familyj—PENTACEROTID (Gray) emend. Perrier, 1884.

Phanerozonate Asteroids with unequally developed marginal plates, the superior
series being frequently masked or hidden in membrane. Abactinal skeleton
reticulate. Plates with large isolated tubercles, or spinelets, or granulose, or
covered with membrane. Actinal interradial areas with large pavement-like
plates which bear unequal-sized granules.

Genus—PENTACEROS, Schulze, 1760.

PENTACEROS, Schulze, 1760. Betrachtung der versteinerten Seesterne und ihrer
Theile, Warschau u. Dresden, p. 50.

GonTAsTER (pars), Agassiz, 1835. Mem. Soe. Sci. Nat. Neuchatel, t. i, p. 191.

PENTACEROS, Gray, 1840. Ann. and Mag. Nat. Hist., vol. vi, p. 276.

OREASTER, Miiller and Troschel, 1842. System der Asteriden, p. 44.

Generic Characters.—Form stellate, marginal plates conspicuous, defining the
ambitus. Abactinal plates regular, with more or less definite intermediate papular
areas. Prominent localised mammillated tubercles or spines present.

All the fossil species of this genus possess intermarginalia, but do not other-
wise approach Sladens’ genus Pentaceropsis which possesses this character. In
view of the fact that intermarginalia may occur as a variation in unmistakable
recent species of Pentaceros this character cannot invalidate the admission to the
present genus of the species about to be described.

PENTACEROS BULBIFERUS. 77

1, Pentaceros BuLbirervs, Forbes, sp. Pl. XX, figs. 1,1a, 1b, and 1 ¢; figs. 2, 2a,
and 2b; Pl. XXI, figs. 1, la, 1b, 3, 34,
4,4a; Pl. XXIII, figs. 2, 2a

OREASTER BULBIFERUS, Forbes, 1848. Mem. Geol. Surv. Gt. Brit., vol. ii, p. 468.

_- — Forbes, 1850. In Dixon’s Geology and Fossils of the
Tertiary and Cretaceous Formations of Sussex,
pp. 328, 329, pl. xxiv, fig. 7.

— = Dujardin et Hupé, 1862. Hist. Nat. Zooph. Echin. (Suites
i Buffon), p. 389.

= -— Forbes, 1878. In Dixon’s Geology of Sussex (new
edition, Jones), pp. 363, 370, pl. xxiv, fig. 7.

-- — P. H. Carpenter, 1882. Geol. Mag., p. 12.

Specific Characters.—Dise and arms very convex. The centrale and primary
interradialia large and tuberculiform. The major radius approximately twice the
minor radius. Radialia of the arm conspicuous. Extremities of the arms swollen.

Material_—The specimens figured and described are all preserved in the
British Museum (Natural History). E. 5040 (Pl. XXI, fig. 1), 40175 (Pl. XX,
fig. 1), 48748 (Pl. XX, fig. 2), and E. 5041 (Pl. XXI, fig. 3), which were bought
from J. Simmons, and 40399 (Pl. XXI, fig. 4), from the collection of E. Charlesworth,
are all labelled as coming from the Upper Chalk of Bromley, Kent, which,
however, seems to be an inexact dealer’s locality, probably intentionally mis-
leading. E. 5042 (Pl. XXIII, fig. 2), also bought of J. Simmons, is labelled
* Upper Chalk, Charlton, Kent.”

Other specimens are known in the Sedgwick Museum, Cambridge, Northamp-
ton Museum, and Brighton Museum. Specimens have also been described by
Valette from the South of France,

This seems to be much the commonest as well as the most graceful of the
Chalk Pentacerotide.

Description —The general aspect of the plates of the disc gives this species a
very characteristic appearance, for the five primary interradialia and the centrale
are very prominent. They have a lobed widely-spreading base, and are swollen
on the upper surface into an almost spherical form. Their weathered surfaces
are pitted in a very regular manner, the pits indicating the former presence of

78 FOSSIL ASTEROIDEA.

granules. Sometimes the granules are still present in situ. The pits are
separated in the example figured Pl. XXI, fig. 1, on an average rather less
than their own diameter apart. There may or may not be a slight margin to the
plate. The centrale in a specimen R:r:: 40: 20 measures 8°5 mm. in diameter.
The primary interradialia are rather smaller, being 6°7 mm.in diameter. Radially
the most conspicuous plates of the disc are the proximal radialia. They have a
very characteristic appearance, their general shape reminding one of a breastplate.
The remainder of the plates of the disc are of very various sizes and distributed
in a fairly regular manner. The general arrangement of these plates is given in
the general account at the conclusion of these volumes.

The madreporite is a conspicuous plate lying at the distal end of a primary
interradial. The two neighbouring adradialia are notched for its reception.

The arms are moderately produced, the major radius being about twice the
minor radius. Measurements of five specimens give the following :

R r
40 mim. i 20 mm.
i é 35 mm. i. ip ie
50 mm. ae 95 oo <
ca 50 mm. 25 mm.
i 50 mm. 20 mm.

At the base of each arm there are five series of plates visible on the dorsal
surface—the radialia, adradialia, and supramarginalia. All the plates at the base
of the arm overlap. They are of a type which may be derived from the breast-
plate shape mentioned above. They gradually become narrowed in length and
increased in breadth until they are shaped somewhat like an inverted T (Pl. XX,
fic. 2)). The granulation is generally confined to the central region of each plate.

The arm about halfway along its length becomes swollen and the plates no
longer overlap but are contiguous. They lose their L-shaped form, become
almost oblong, and at the same time rather tumid. This is especially noticeable
in the case of the radialia. The form of the plates is, however, rarely absolutely
regular, but one which is generally derivable from the breastplate shape.

If we examine a cross-section of the arm, we see that the base of the plates
of the dorsal intermediate series is prolonged inwards (and ventralward), so that
a single isolated plate appears club-shaped.

All the plates are pitted for granules except at the extreme margin.

The supero-marginal plates are from twelve to thirteen in number, the infero-
marginals from thirteen to fourteen in number. The arm is very high and both

PENTACEROS BULBIFERUS. 79

infero- and supero-marginal plates appear in dorsal view. In fact, the infero-
marginal plates do not take any part in the formation of the actinal surface. This
is parallelled in modern species of Pentaceros, e.g. P. clavatus. In marginal view
the supero- and infero-marginals at the extremity of the ray very distinctly
alternate. This alternation persists at the base of the arms, but here it is not
always so obvious.

The supero-marginals are much higher than the infero-marginals, and also more
oblong in shape. Both supero- and infero-marginal plates are regularly but
coarsely pitted for granules.

In specimen figured on Pl. XX, fig. 1, we obtain the following measurements :

Breadth of fifth infero-marginal from the extremity of ray. 5°8 mm.

Length Me a He i = . 4 -
Breadth ,, | supero-marginal __,, x. . SOAs
Length = x Ss A 5 ye oe
Breadth ,, _ radialia 5 es =) EO ny
Length Fe Xe * #5 - . aA,
Width of ambulacral groove’. : : SS ae

The ocular is visible in this specimen. It is about 1:6 mm. in length and
breadth. The extremity is slightly pointed, and its ventral surface is hollowed out.

A ventral view is figured on Pl. XXI, fig. 1. Ventro-lateral plates extend
almost to the extremity of the arms. These are, as usual, rather greater in breadth
than in length. The adambulacral plates appear to be about half the length of
the bordering actinal plates. Their armature consists of several rows of spinelets
arranged in pairs.

A few intermarginalia are present in the interradii. They, as usual, press the
supero- and infero-marginalia on to the abactinal and actinal surfaces of the disc
respectively.

Locality and Stratigraphical Position—Upper Chalk, Bromley, Kent; according
to Dr. Rowe, probably from the Chislehurst caves near that locality.

Variations.—Variations occur amongst al] the specimens, especially with
regard to the ornamentation of the plates and the madreporite. The British
Museum specimens, 48748, which occur together in a slab, are especially note-
worthy, inasmuch as the lowest situated individual possesses on the disc no plate,
which is bulbiform or raised conspicuously above the remainder.

80 FOSSIL ASTEROIDEA.

2, Penraceros Borst, Forbes, sp. Pl. XXII, figs. 4, 4a, 4b, 4c; Pl. XXIII,
figs. 1, 1a, 16; Pl. XXVI, figs. 2, 2a, 25.

OreastER Boysu, Forbes, 1848. Mem. Geol. Surv. Gt. Brit., vol. ii, p. 468.

= — Forbes, 1850. In Dixon’s Geology and Fossils of the Tertiary
and Cretaceous Formations of Sussex, p. 328, pl. xxi,
fig. 6.

- — Dujardin et Hupé, 1862. Hist. Nat. Zooph. Echin. (Suites A
Buffon), p. 389.

— — Forbes, 1878. In Dixon’s Geology of Sussex (new edition,
Jones), pp. 362, 370, pl. xxi, fig. 6.

Specific Characters —The primary radialia and interradialia are large hemi-
spheroid punctate tubercles. R:r::80mm.:18 mm. Rays well produced, steep-
sided, almost square in section, and tapering gradually to the extremity. Only a
few of the plates of the disc enter the base of the arm. Supero- and infero-
marginal plates adjunct, the intermarginalia being represented only by a few
scattered granules.

Matevial.—The type specimen was said by Forbes (1848) to be in the collection
of the Marquess of Northampton. ‘The specimen figured and described in Dixon’s
‘Geology of Sussex’ (see reference) was said by Forbes to have been “ discovered
by Major Boys and formed part of his interesting collection.” This statement
does not preclude the hypothesis that the specimen figured was also the type
specimen. Neither specimen (if there were two) can now be traced. The following
description is based chiefly on a specimen in the Sedgwick Museum, Cambridge
(Pl. XXVI, fig. 2), which shows the actinal surface of the arms and a portion
of the disc. It is supplemented by reference to a less nearly perfect specimen
preserved in the British Museum of Natural History (J. Simmons’ Coll., 46600),
which presents views of isolated rays (Pl. XXII, fig. 4), and an isolated ray seen
from the dorsal surface (Pl. XXIII, fig. 1) in the same museum (Dixon Coll., 48083).

Description—The dise is covered with a number of rounded or irregularly-
shaped plates. A circlet of large tubercles is very distinct and characteristic of the
species. These tubercles are hemispherical and not so swollen as those of
P. bulbiferus. They are smooth, and possess a fine distinct ornament, thus
distinguishing them from the circlet of P. coronatus. Their diameter is about
85 mm., and they seem to be arranged radially and interradially, making a total
of ten. The madreporite was figured by Forbes. It is roughly triangular in shape.

PENTACEROS BOYSII. 81

The arms are well produced. R=80 mm. and r==18 mm., the major radius
being thus about four and a half times the minor radius. They taper gradually
to the extremity. The breadth of the ray about the fourth supra-marginal plate
is 68mm. The height of the ray at the same spot is almost exactly the same.
The rays are steep-sided, and consequently appear almost square in cross section.

The supero-marginalia are adjunct throughout almost the whole length of the
ray, for only one or two single radialia enter the base of the ray. At the base of
the ray they are flat and slightly rhomboidal. They possess an anterior indenta-
tion on their inner surface and are about 3°5 mm. in breadth. They gradually
diminish in size distally and at the same time become distinctly swollen. They
number about twenty-eight.

The infero-marginal plates are approximately of the same size and number as
the supero-marginals. Both series imbricate shghtly. The ornament of these
plates consists of a number of fine granules in the centre, while there is a distinct
margin without granulations.

Between the supero- and infero-marginal plates a few scattered granules repre-
sent a slight development of the intermarginalia.

The adambulacrals are a series of small oblong plates. They border the infero-
marginals from about the eleventh supra-marginal onward. ‘They are much worn,
and but slight traces of their armature remain. About five adambulacrals occupy
the same length as two infero-marginal plates. Proximally there is a single row of
small plates which separate the two series.

Only a few scattered ossicles of the actinal surface of the dise remain.

Locality and Stratigraphical Position —Upper Chalk, Kent.

Remarks.—Valette (‘ Bull. Soc. Yonne,’ 1902) has described a number of
species of starfishes from the Senonian of the South of France. The remains are
found as scattered ossicles. Some of these are grouped by Valette as a new
species which he calls P. senonensis. They are noticed by the author to resemble
P. Boysii except that they are smooth and therefore do not have the ornament
possessed by P. Boysii. Valette regards this absence of ornament as rendering
them specifically distinct from P. Boysii, as other ossicles found in close
proximity still possess the ornament. In view of the vagaries of the way in
which solution may occur, I cannot admit this contention and consider that it
is much more probable that the ossicles at one time possessed ornament and
were identical with P. Boysii. All the other ossicles except those of the so-called
Arthraster senonensis (vide infra, p. 92) were identified with English Cretaceous
genera, which would support this contention.

82 FOSSIL ASTEROIDEA.

3. PENTACEROS coronatus, Forbes, sp. Pl. XIX, figs. 1, la; Pl. XXIV, figs. 2,
24,25, 2¢5 Pl. XXV 5 hese,

OREASTER CORONATUS, Forbes, 1848. Mem. Geol. Surv. Gt. Brit., vol. 11, p. 467.

-- a Forbes, 1850. In Dixon’s Geology and Fossils of the Ter-
tiary and Cretaceous Formations of Sussex, pp. 327,
328, pl. xxi, fig. 7 a—d.

— — Dujardin et Hupé, 1862. Hist. Nat. Zooph. Echin. (Suites
a Buffon), p. 389.

— — Forbes, 1878. In Dixon’s Geology of Sussex (new edition,
Jones), pp. 862, 370, pl. xxi, figs. 7, 7 a—d.

Specific Characters.—Disc large, with conspicuous nodular primary radialia and
interradialia. The major radius is about five times the length of the minor radius.
Sides of arms very steep, so that the arm appears to be square in cross section.
A triple row of intermarginalia present in the interbrachial areas.

Material_—The type specimen of this species is preserved in the British Museum
of Natural History (Dixon’s Coll., 35480). Unfortunately, only one arm and a
portion of the dise are preserved. A further specimen, registered EH. 2562, from the
cabinet of Mrs. Smith, of Tunbridge Wells, is preserved in the same museum, and
another example is to be seen in the Museum of Practical Geology, Jermyn Street.

Description.—The most conspicuous feature of the dise is the circlet of ten
“large, more or less polygonal nodose pyramidal tubercles.”' These are the
primary radialia and interradialia. The interradial tubercles are rather larger
than the radial tubercles, the former measuring 9°2 mm., the latter 7°7 mm. in
diameter. The remainder of the disc is covered by irregularly shaped plates.

The madreporite has been broken away from the disc of the specimen no. 35480.
It is figured Plate XXV, fig. 9.

R:r::58 mm.:19-+ mm. in the type specimen where the single arm is
broken short. In specimen no. E. 2562 R:r:: 100 mm.:20 mm. The arms
are 30 mm. broad at the base. Their surface is flat, and the sides slope away
at right angles, so that a cross section of the arm is square.

Both radial and adradial plates are present in the base of the ray. The
adradials are irregular in shape and soon disappear. The radials are roughly
oblong in appearance, and exist throughout that portion of the arm preserved.
They diminish in size, however, distally.

The supero-marginal plates are indented on their anterior median surface.

1 Forbes, in Dixon’s ‘ Geology of Sussex,’ p. 327.

PENTACEROS SQUAMATUS. 83

They appear to imbricate slightly at their margins. The breadth of the fourth
supero-marginal is 7 mm., the length 4 mm., and the height 3°5 mm. The height
of the ray at this point is 12°2 mm.

The infero-marginal plates are opposite to the supero-marginals. They are
approximately about the same size and number. Between the supero- and infero-
marginal series a triple series of intermarginalia occurs in the interradial areas.
The inner and larger intermarginals persist throughout the greater part of
the length of the arm. It is this intercalated series which gives to the arm its
great proportionate depth. The outer and smaller series disappear at about the
seventh and ninth infero-marginal plates.

The ornamentation of the plates appears to have been worn away, although
upon many of the plates a distinct marginal area may be seen.

Upon most of the plates there occur small entrenched pedicellaria which are
very characteristic of this species of Pentaceros. They consist of a small pit from
which radiate two fine entrenchments (see Pl. XXIV, fig. 2 @).

One of the rows of specimen no. E. 2562 is distorted so as to bring the ventral
surface into view. This shows that the ventro-lateral plates extend well towards,
and perhaps all the way to, the extremities of the arm.

Locality and Stratigraphical Position.—The locality of the type specimen is
given as Lower Chalk, Washington, Sussex. The specimen registered E. 2562 is
from the Lower Chalk, Burham, Kent, and the specimen in the Museum of
Practical Geology is from the Lower Chalk, Dover.

Remarks.—The specimen registered E. 2562 presents only one or two pedi-
cellariz, which are so characteristic and numerous on the other two specimens.

4. Penvaceros squamatus, Forbes, sp. Pl. XXV, figs. 3, 3a, 3b, 3.

OREASTER SQUAMATUS, Forbes, 1848. Mem. Geol. Surv. Gt. Brit., vol. ii, p. 468.

— — Forbes, 1850. In Dixon’s Geology and Fossils of the
Tertiary and Cretaceous Formations of Sussex,
p- 328, pl. xxiii, fig. 7.

— -- Dujardin et Hupé, 1862. Hist. Nat. Zooph. Echin. (Suites
i Buffon), p. 389.

_ _ Forbes, 1878. In Dixon’s Geology of Sussex (new edition,
Jones), pp. 363, 370, pl. xxiii, fig. 7.

Specific Characters. — Dise high, with conspicuous primary radialia, inter-

13

84 FOSSIL ASTEROIDEA.

radialia and centrale. Major radius about four times the length of the minor
radius. Only radialia enter the base of the arm. Dorsal surfaces of arms flat,
sides slope away at an obtuse angle from this. Ossicles distinctly imbricating.
A few intermarginalia present.

Material_—The only specimen of this species is preserved in the Brighton
Museum. ‘The specimen consists of the disc and a portion of three arms. On the
whole little displacement of the ossicles has taken place.

Description—The dise is strongly convex, and is covered with the circlet of
primary radiaha and interradialia which are disposed around the centrale. All
these ossicles appear shaped like a breast-plate. The centrale has a diameter
of 42 mm. The primary interradialia are larger, possessing a diameter of
5°3 mm., whilst the primary radialia are the smallest of the series, measuring only
3°7 mm. across. Between the centrale and the primary interradialia a number
of irregularly distributed plates appear. In the next right-hand interradius
to the madreporite a number of these appear to have surrounded an anal opening.
The primary interradialia almost touch one another, and the radialia consequently
rest on the bases of pairs of ossicles. A few adradialia are present, but they are
confined to the disc. A pair of them help to enclose the madreporite, which
is a polygonal plate 9 mm. in greatest diameter. The ornamentation of the
ossicles is rather coarse when present, but usually it is very much worn away.

The arms are well produced. R:r::30-+--mm.:7:'8mm. They are 1:3 mm.
in breadth at the base. After the fourth or fifth radiale the remainder become
minute but persist throughout the length of the arm preserved.

The supero-marginalia are finger-shaped; they, as also the infero-marginalia,
distinctly imbricate. The dimensions of the third supero-marginal, reckoning
from the median interradial line, are as follows: length 2°3 mm., breadth 3°1 mm.
The long axes of the supero-marginal plates slope away distally, thus causing
pairs of plates to assume the shape of arms of a V. They are at least thirteen
in number.

The infero-marginal plates are similar in size and number to the supero-
marginal series. In the interradii a few intermarginalia are present. These force
the supero-marginal series to the surface of the disc.

Nothing is known of the ventral surface.

Locality and Stratigraphical Position.—Upper Chalk, Woolwich.

PENTACEROS OCELLATUS. 85

5. PENTACEROS OCELLATUS, Forbes, sp. Pl. XXV, figs. 4, 4a.

OREASTER OCELLATUS, Forbes, 1848. Mem. Geol. Surv. Great Brit., vol. ii,
p. 468.
— — Forbes, 1850. In Dixon’s Geology and Fossils of the
Tertiary and Cretaceous Formations of Sussex,
p- 329, pl. xxi, fig. 13.
_- -- Dujardin et Hupé, 1862. Hist. Nat. Zooph. Echin.
(Suites 4 Buffon), p. 389.
— —- Forbes, 1878. In Dixon’s Geology of Sussex (new edition,
Jones), pp. 364, 370, pl. xxi, fig. 3.
PENTACEROS — McPherson, W., 1902. Rep. Brighton Nat. Hist. Soe.

Specifie Chavacters.—Ventro-lateral plates (as probably also the dorsal plates)
depressed and finely striated on their truncated surface so as to simulate the
surface of a madreporite, with sides rugged and ocellato-punctate. Between these
plates smaller ossicles of a similar character are interspersed.

Matevial—But one specimen of this species was known to Forbes. This is
preserved in the British Museum of Natural History (Dixon Coll., E. 2571). It is
a mass of ossicles which look as if they were derived from the dorsal surface of the
disc. They are more spheroidal and somewhat larger than the ossicles of the
ventral surface of the more nearly perfect example discovered by Mr. William
McPherson in the Senonian Marsupites band at Brighton. This he presented
to the British Museum (Natural History) in 1901 (KE. 5012).

Description—The dise and arms are unknown. The specimen no. E. 5012 shows
a well-preserved portion of the ventral surface. The mouth-angles were occupied
by single initial rhomboidal ossicles. To these succeed the ventro-lateral ossicles
which border the ambulacral groove. ‘These are pentagonal ossicles of very
uniform size. The length of the exposed sides of the ossicles bordering the
groove is 4°4mm. and the greatest breadth of an ossicle 42mm. The remaining
ventro-lateral plates are hexagonal, but of almost the same dimensions, although
the plates appear to become a little larger distally.

The plates overlap one another considerably, rendering precise measurement
difficult. Between the larger plates are interspersed large numbers of smaller
and more irregular ossicles which fill up the angles between their sides. The
whole test would be thus very strongly built.

Both larger and smaller plates are curiously similar in appearance. The

86 FOSSIL ASTEROIDEA.

madreporiform striations on the truncated summits and the ocellato-punctate sides
give a most characteristic appearance and render the species unmistakably distinct
from all known species of Pentaceros.

The ambulacral groove is 3°5mm. wide. The adambulacrals are difficult of
recognition and have probably for the most part been lost, but a large number of
the hour-glass shaped ambulacrals may be seen.

Locality and Stratigraphical Horizonn—Upper Chalk, Kent; Upper Senonian,
Brighton.

6. PENTACEROS ABBREVIATUS, n. sp. Pl. XXIV, figs. 1, 1a, 1b, le.

Specific Characters.——Body of medium size. Arms moderately produced, but
their breadth making them appear stumpy, rounded at the extremities, and hemi-
spherical in cross section. Five series of dorsal ossicles enter their base. Of these
the radialia and adradialia persist throughout the length of the arm. A few small
intermarginaha are present.

Material—There is only one specimen known of this species, and of this
practically all that remains are two arms. It is preserved in the British Museum
of Natural History (J. Tennant’s Coll., 57538).

Description.—These arms are characteristically wide, the width of the arm at
the base being 31mm. They narrow very gradually towards the extremity.
Throughout the ray the ossicles, except for the differences noted below, are very
similar in appearance. Atthe base of the ray, where dorso-lateralia also enter into
the composition of the dorsal skeleton, they are rounded and possess interspaces
of considerable extent. These interspaces are often filled by smaller granules
arranged irregularly. At times, however, between two radialia or adradialia one
of the smaller ossicles is arranged in a very regular and alternating manner. Both
large and small ossicles are finely granulated, and the large ossicles alone are
perforated for pedicellariz. The average size of the larger ossicles at the base of
the arms is about 6 mm. Towards the extremity of the ray the radialia, adradialia,
and marginalia become hexagonal, and fit very closely so as to make a compact
skeleton. The terminal ocular plate is hexagonal and conspicuous. It has a
flattened articulation which undoubtedly was originally occupied by a spine.
Several of the other dorsal plates in the distal portion of the ray also possess
similar articulation.

PENTACEROS BISPINOSUS. 87

The supero-marginalia and infero-marginalia are equal in number. There
were probably thirteen of each in the space between an interradius and an
extremity of an arm.

In the interbrachial are there is a series of minute granular intermarginalia.

The traces of the dise which are present suggest that the ossicles of this region
were oval in shape and minute in size. I exposed a portion of the ventral surface
of the arm, but, unfortunately, little trace of structure was shown. The ventro-
lateralia extended to the extremity of the ray. The ridges of the adambulacral
armature are lost.

Locality and Stratigraphical Position.—Upper Chalk, Charlton, Kent.

7. PENTACEROS BISPINOSUS, n.sp. Pl. XXITI, figs. 3, 3.4, 3, 3c.

Specific Characters—Dise large. Arms moderately produced. Single isolated
y p g

marginal ossicles vertebra-shaped with biconcave extremities. Ventro-lateral
plates with strongly marked sockets for two or more spines.

Material.—The only specimen of this species is that preserved in the British
Museum of Natural History (H. W. Taylor’s Coll., 35482). Only the ventral
surface is exposed, and this is very much distorted.

Description —The disc appears to have been large. Its actinal surface is
covered with a number of sub-equal oblong or polygonal plates, which possessed
sockets in which fitted spines (Pl. XXIII, fig. 3c). These plates are 48 mm.
long, and 3:1 mm. wide.

The arms are moderately broad, and at least four series of ventro-lateral plates
enter at the base. R:r:: 60 mm.: 20 mm. (approximately), the major radius
therefore measuring about three times the minor radius. ‘The marginal ossicles
are shaped very much like the centrum of a vertebra, and are biconcave. They
possess a distinct granulation in their central region, which is surrounded by ¢
wide margin. The infero-marginals at the base of the ray are about 3°2 mm. wide
and 21 mm. long. There were probably sixteen of them from the interradius to

the extremity of the ray.
The specimen is otherwise so distorted that little can be made of its structure.

Locality and Stratigraphical Position —Upper Chalk, Sittingbourne, Kent.
U] gray PI g

88 FOSSIL ASTEROIDEA.

The following are placed provisionally in the genus Pentaceros :

8. PENTACEROS PUNCTATUS, n.sp. Pl. XXVI, figs. 1, la, 1D.

Specific Characters.—Body of large size. Marginal series of plates possessing
well-developed foraminate pedicellarize. Intermarginalia present.

Material—The only example of this species is a fragmentary portion of an
arm preserved in the British Museum (Natural History) and bearing the
registered number H. 2561.

Description.—The body of the starfish must have been of large size. The
supero-marginals are in contact in the extremity of the arm, although interspersed
granular plates occur. Proximally, at least, radial plates were present. The
largest supero-marginal present is 11:2 mm. high and 6:5 mm. long in its widest
point. It is of rather irregular shape and possesses two foraminate pedicellariz.
The infero-marginal plates alternate with the supero-marginals. They are of the
same height as the supero-marginals but only 5 mm. broad. They are oblong in
shape; the two interior corners of the oblong, however, are cut away, making the
ossicles six-sided. The foramina once occupied by the pedicellariz are deep and
often situated in a depression. From the foramen itself ridges may run out,
which probably served for the attachment of muscles.

The infero-marginal series border only the side of the arm and take but little
part in the formation of the ventral surface.

An intermarginal series of rounded granular plates occurs.

Locality and Stratigraphical Position—Upper Chalk.

9, PENTACEROS PISTILLIFERUS, Forbes sp. Pl. XXV, fig. 5.

OREASTER PISTILLIFERUS, Forbes, 1848. Mem. Geol. Surv. Great Brit., vol. ii, p. 467.
— — Forbes, 1850. In Dixon’s Geology and Fossils of the
Tertiary and Cretaceous Formations of Sussex,
p. 329, pl. xxi, fig. 15.
— PISTILLIFORMIS, Dujardin et Hupé, 1862. Hist. Nat. Zooph. Echin. (Suites
i Buffon), p. 389.
-- PISTILLIFERUS, Forbes, 1878. In Dixon’s Geology of Sussex (new edition,
Jones), pp. 363, 370, pl. xxi, fig. 15.

PENTACEROS. 89

Specific Characters.—Primary radialia (or interradialia) large with a dilated
summit which possesses no ornament and is excavated into pits. ‘ Ossicles of the
arm are narrow, shuttle-shaped, tumid in the centre and slightly impressed
towards each extremity” (Forbes).

Material.—Several fragmentary remains of this species are known. The most
nearly perfect remains are those in the Museum of Practical Geology, Jermyn
Street. Other specimens are in the British Museum (Natural History), registered
E. 5037, 57624, E. 2564 (all Pl. XXV, fig. 5), 7600, E. 25637, E. 2565.

Description.—Nothing is known further than the description given in the
diagnosis. Forbes’ description reads as if he had described the species from the
specimen in the Museum of Practical Geology. This originally was in the collec-
tion of the Marquis of Northampton. Forbes seems to have described the large
ossicles upside down. He also says they were in a circlet of five, which is not
apparent in any specimen known. The roughened and pitted surface recalls in
some respects the primary radialia and interradialia of P. coronatus.

Locality and Stratigraphical Position Upper Chalk, Kent and Sussex.
q gray PI

10. Penraceros, sp. Pl. XXV, fig. 7.

This specimen is preserved in the British Museum of Natural History (no. 5514).

It consists of five marginal plates which are 12°3 mm. high, and have an
average length of 55 mm. The plates are rugged in appearance and the orna-
ment is worn away. A cirral of a crinoid (probably Bourguetiocrinus) has become
fixed between two of these plates.

11. Penraceros, sp. Pl. XXV, fig. 8.

The only specimen is preserved in the Brighton Museum. It consists of a few
marginal plates. The supero-marginals are rather irregular in shape, some being
almost wedge-shaped. On an average they are 4 mm. high and 3-2 mm. long.
The infero-marginals are opposite and equal in length to the supero-marginals.

They are only 2°9 mm. high. The plates possess a distinct margin, but the

90 FOSSIL ASTEROIDEA.

ornament otherwise is worn away. There are a few small granular inter-
marginalia.

With these plates is associated a large plate which appears to be a worn radial
or interradial of P. Boysii.

Family—ASTROPECTINID (Gray, 1840), emend. Sladen, 1886.

Phanerozonate Asteroids with large marginal plates bearing spines or spiniform
papille. Abactinal skeleton with true columnar papillz. Actinal interradial areas
small, interradial plates when present spinose. Ambulacral plates short and more
or less compressed. Superambulacral plates present. Aproctuchous. Pedicellariz
rarely present.

Genus—ASTROPECTEN, C. F. Schulze, 1760.

Adambulacral plates touching the infero-marginal plates along the ray.
Marginal and adambulacral plate not correspondent in length and number, Supero-
marginal plates more or less well developed. Marginal plates long and more or
less quadrate. Superior and inferior series subequal.

ASTROPECTEN, sp. Pl. XXV, figs. 2, 2 a.

Material.—There is one specimen in the Sedgwick Museum at Cambridge, which
looks like an Astropecten. It is figured on Pl. XXV, figs. 2 and 2a. Practically
only the marginal plates are preserved.

Description —R :r::45 mm.:15 mm. The interbrachial arcs are well rounded.
he supero-marginalia are remarkably uniform in size throughout the greater
portion of the ray. Their breadth is 4 mm, and length 1:7 mm. About thirty of
these are present from the interradius to the extremity. At the apex of the ray
these plates are adjunct. The upper surface of each plate is rounded.

The infero-marginalia are equal in size, opposite to, and, as far as one can
judge, similar in appearance to, the superior series. There is a distinct groove
between the two series.

Locality and Stratigraphical Position—Upper Greensand, Blackdown (?).

Palzxontograpbical Society, 1907.

A MONOGRAPH

ON THE

match FOSSIL

HCHINODERMATA

FROM

THE CRETACEOUS FORMATIONS.

VOLUME SECOND.
THE ASTEROIDEA AND OPHIUROIDEA.

BY

Make oem NCHR, B.A,’ F:.G.S8.

PART FOURTH.
Paces 91—132; Prares XXVII—XXIX.

LONDON:
PRINTED FOR THE PALHZONTOGRAPHICAL SOCIETY.
1907.

. aon ,
Fs \ ft %
Fin | 4 .
press ach ? S
PRINTED BY ADLARD AND SON, LONDON AND DORKING.
'
‘=

ARTHRASTER DIXONI. 91

FAMILY UNCERTAIN.

Genus—ARTHRASTER, Forbes, 1848.

Arms stout and long. Radialia, marginalia, and ventro-lateralia form an alter-
nating series of seven very completely articulating similar ossicles, which fit so
closely as to leave no conspicuous interstices. Each ossicle consists of an oblong
and flattened base with a surmounting ridge. Ventro-lateral plates on actinal
surface of dise small and mammiform. Ossicles on abactinal surface of disc
hemispheroid with a crenulated edge. All the ossicles possess, as ornament,
hemispherical granular prominences.

1. Arruraster Drxont, Forbes. Pl. XVIII, figs. 1 and la; Pl. XXIX, figs. 11

and lla.

Arruraster Drxont, Forbes, 1848. Mem. Geol. Surv. Gt. Brit., vol. ii, p. 467.
— — Forbes, 1850. In Dixon’s Geology of Sussex, p. 336,
pl. xxiii, fig. 1.
— — Dujardin et Hupé, 1862. Hist. Nat. Zooph. Echin. (Suites a
Buffon), p. 437.
_— — Forbes, 1878. In Dixon’s Geology of Sussex (new ed., Jones),
pp. 369 and 370, pl. xxiii, fig. 1.

Specific Characters—Dorsal ridge of all of the arm ossicles well rounded.
No spines present except on the adambulacral plates.

Material—The best example of this very peculiar starfish is preserved in the
British Museum of Natural History, Dixon Coll., 47000. The specimen consists
of the remains of four arms, only one of which is at all well preserved.
It is the type described by Forbes, and is figured in this Monograph on
Pl. XVIII. A well-preserved fragment of an arm is also in the possession of
Dr. Rowe, of Margate. I have referred two fragmentary specimens presented to
the British Museum (E 5023 and E 5024) by Mr. W. McPherson, F.G.S., to this
species (vide infra).

Description —A section of the arm is similar at all points. It shows seven
ossicles, namely, a radial, the pairs of supero- and infero-marginalia, and a pair of
ventro-lateralia. These ossicles all alternate in series, and they closely fit the
corresponding neighbouring plates in their respective series. The edges of the
plates possess articulations which assist in forming this close union. All the plates

14

92 FOSSIL ASTEROIDEA.

are generally similar in appearance. They differ, however, in measurements as
detailed below.

Breadth of radialia .. , : . 82 mm.
Length ,,_ ,, : ; 5 a ROD eiss
Breadth of supero-marginalia—. : o. Opens
Length ,, * ; ; ho cpa
Breadth of infero-marginaha : F DOE
Length ,, - ; te ORD ep
Breadth of ventro-lateralia d é » Uaee ay
» 3 J =
Length ,, “p : : UOtOE as

It will thus be seen that the breadth of the plates diminishes as we proceed
ventralwards. The ridges, also, on the plates, become more rounded in the same
direction.

The ornament of the plates consists of hemispherical granular prominences
of moderate size. They appear to have been especially prominent at the base
of the ridge. No spine-pits are present.

The height of the ray is 16°5 mm., and the breadth is about the same. Post-
mortem contraction has brought the ventro-lateralia of opposite sides into close
approximation, in some cases totally obliterating the ambulacral groove. Along
one or two of the arms some of the adambulacrals are still visible. They are
1:8 mm. broad and 1:2 mm. long. The portion of the plate nearest the ambulacral
groove is depressed, giving the plate a two-storied appearance.

A few robust, rhomboidal, smallish ventro-lateralia are present at the base of
the arm. Their breadth is 1:8 mm. They are mammiform. <A few similar
plates also enter the base of the arm.

The two collections of isolated ossicles presented by Mr. McPherson referred
to above are very interesting. Each specimen consists of a single ossicle
simulating one of the abactinal bulbiform ossicles of Pentaceros, but possessing
the distinct Arthraster ornament, associated with plates which exactly match the
ventro-lateralia of A. Divoni and other plates which resemble the arm plates
of this species except that the surmounting ridge is not so high. There is no
doubt that the plates are those of a species of Arthraster. I have little hesitation,
in spite of their occurrence in the Upper Chalk, in referring the ossicles to A.
Divoni, especially as it is a matter of common experience that species of Chalk
starfish have a wide stratigraphical range.

Locality and Stratigraphical Position—Forbes’ type is from the Lower Chalk,
Baleombe, Sussex. The specimen in the possession of Dr. Rowe was collected in
the zone of Terebratulina gracilis i Devon. The specimens presented to the
British Museum by Mr. McPherson are from the Marsupites zone, Brighton.

ARTHRASTER CRISTATUS. 93

Remarks.—Forbes compared the genus Arthraster with the modern genus
Ophidiaster. The larger amount of material known since that time does not allow
us to recognise such affinity.

Valette (see above, p. 81) has described certain isolated plates, which are
similar in form and size, as belonging to the genus Arthraster, and has called
the species A. senonensis. These plates are smooth and show no trace of the
surmounting longitudinal ridge or ornament which is so characteristic of Arthraster.
Dom Aurélien Valette kindly lent these plates to Dr. Bather in order that I might
examine them. I am therefore enabled to state that the plates are those of
Pycinaster angustatus.

2. ARTHRASTER CRISTATUS, n. sp. Pl. XXIX, figs. 10, 10 a, 10d.

Specific Characters.—Ridges of the radialia and supero-marginalia cristate.
Upper surface of the ridge of all arm-ossicles possessing lipped pits formerly
occupied by small spines.

Material—The specimen figured on Pl. XXIX was restored by Dr. Blackmore,
of Salisbury, from a number of isolated ossicles in his collection which were found
ina single mass of chalk. These ossicles are the only material known of the species.

Description.—The dimensions of the ossicles are as follow:

Breadth of radialia . : : 5'4 mm.
Length ,, 5 : : ‘ Se oe
Breadth of supero-marginalia . : 48,
Length ,, - ; , 52 ss
Breadth of infero-marginalia . ‘ Aco ves
Length ,, = E ‘ mae as
Breadth of ventro-lateralia : : Shep mon
Length ,, i : big ane

Just as in A. Divoni the breadth of the plates diminishes ventralwards, and the
ridges on the plates become more rounded in the same direction. The cristate
ridges of the more dorsal plates are, however, very characteristic of the species, as
are also the lipped pits on the summit of the ridges. The pits were formerly
occupied by small spines. The base of the ridges of the plates possesses the
granular hemispherical prominences, such as are also met with in A. Divoni and
characterise the genus.

Locality and Stratigraphical Position—Micheldever, Hants. Zone of Micraster

cor-anguinum.

94. FOSSIL ASTEROIDEA.

ADDENDUM (to Phanerozonate Asteroids).

Further investigation, as a result of the privilege of investigating the fine
collection of Chalk Asteroids in the possession of Dr. Blackmore, of Salisbury, has
enabled me to describe several new species belonging to genera which have been
dealt with in previous pages. Some of these species had been recognised but not
described by Dr. Blackinore, to whom I am indebted for very many valuable
suggestions.

Family—PENTAGONASTERIDA, Perrier, 1884. (See p. 3.)
Genus—N YMPHASTER, Sladen, 1885. (See p. 14.)

5. NYMPHASTER RUGOSUS, n. sp. Pl. XXIX, figs. 7, 7 a.

Specific Charactervs—All marginalia covered with granular prominences, which
are closely crowded, and in no case arranged in a linear series. No spine-pits on
the marginalia. Margin of disc lunate.

Material—Two specimens are known of this species. They are preserved in
the British Museum (Nat. Hist.), and bear the registered numbers 57516 (pur-
chased of W. Griffiths) and 76002 (Capron Coll.). Both specimens are imperfect.
The first-named specimen is figured Pl. XXIX, fig. 7, and is taken as the type.

Description.—The two specimens show the disc to have been small. The minor
radius in the specimen 57516 measures about 11 mm. The arms are broken off
short in both specimens, and therefore it is not possible to give the major radius.

There are about eight infero-marginalia in each interbrachial arc. These are
all approximately equal in size, being 2°7 mm. long and 1°8 mm. broad. In shape
they are oblong.

The margin of the disc is lunate. It is this character and the character of the
granular prominences which distinguish the species from N. radiatus.

Locality and Stratigraphical Position—Lower Chalk, Dover and Folkestone.

PYCINASTER SENONENSIS. 95

Genus—PYCINASTER,' nom. nov.

Pyenaster, Sladen, 1891 (see p. 21), non Pomel, 1883. Classif. méthod. Echin., p. 42.

1. Pycrvasrer ancustatus, Sladen sp. (see p. 21). Pl. IX, figs. la, 1b; Pl. XXI,
fies. 2, 203 Fl. ARV, fig. 7;
Pl. XXVI, figs. 4, 4a, 4b.

This species appears to be quite common in the Upper Chalk. An exceedingly
well-preserved specimen is in the collection of Dr. Blackmore, of Salisbury.

The following specimens, which belong to the genus Pycinaster, and probably to
this species, have been erroneously ascribed by me to other genera and species in
Part III of this volume (pp. 67-90).

The single specimen described on p. 73 as a new species, Pentagonaster robustus,
is probably an immature form of this species. A collection of five ossicles described
(p. 89) as Pentaceros, sp., and the specimen figured on Pl. XXVI as Calliderma
mosaicum, also belong to the species. The latter specimen should be described as
from the Upper Chalk.

Dom Aurélien Valette has courteously enabled me to examine the syn-types of
his Arthraster senonensis (¢ Bull. Soc. Sci. Yonne,’ 1902, p. 25). They prove to be
marginals of Pycinaster angustatus. Four ossicles referred by him to his Pentaceros
senonensis (vide infra) also belong to the present species.

9. PYCINASTER SENONENSIS, Valette, sp. Pl. XXVI, figs. 1, la, 1b; Pl. XXIX,
figs. 6, 6a.
PENTACEROS SENONENSIS, Valette, 1902. Bull. Soc. Sci. Yonne, vol. lvi, pp. 17, 18,

figs. 1, 2 (non 3—7).
_ punctatus, Spencer, 1905. Antfea, p. 88.

Dr. Blackmore’s material enables me definitely to ascribe this species to
Pycinaster, and to add the following new diagnosis and details :

Specific Characters.—Body of large size. Breadth of marginalia more than
twice their thickness. All marginalia smooth or with very shallow hexagonal
spine-pits.

Description —The marginalia may be as much as 20 mm. high. They appear
to be distinguished from the marginalia of P. angustatus, not only by their

1 Tuxevds, compact, Homeric form of ruxvds. Dom Aurélien Valette kindly pointed out the
prior use of Pycnaster to Dr. F. A. Bather, who suggests the above emendation.

96 FOSSIL ASTEROIDEA.

magnitude, but also by the manner in which the upper surface is turned over so as
to make the ossicle f-shaped (compare Text-fig. 23, p. 119).

Associated with these marginalia are found rounded, smooth ossicles, which are
correspondingly large, being as much as 12 mm. in diameter, and which are
undoubtedly ossicles from the abactinal surface of the disc. ‘Their size and form
(see Pl. XXIX, fig. 6) render them liable to be mistaken for ossicles of Stawrander-
aster (see p. 125). They do not, however, possess spine-pits, and species of Stanr-
anderaster which do not possess spine-pits are very distinct, having nodular abactinal
ossicles of a very characteristic appearance (compare S. coronatus, Pl. XXIV, fig. 2).
There appears therefore to be no doubt that these ossicles should be referred to
the genus Pycinaster, and probably to P. punctatus. The base of the abactinal
ossicles of P. angustatus is produced, as also in P. crassus (Pl. XXIX, fig. 4a), and
quite different from the flattened base of these ossicles.

Remarks.—It might be urged that the differences which separate these ossicles
from those ascribed to P. aingustatus, are not sufficient warrant for a new species.
I regard the differences, however, given above as important, and though several
well-preserved specimens of P. angustatus are known, none approaches the large
size which P. punctatus must have attained. On p. 81, Pentaceros senonensis,
Valette, was regarded as probably identical with P. Boysii. Hxamination of the
original specimens, which I owe to the courtesy of Dom Aurélien, shows that they
belong to four different species: Stawranderaster coronalus, S. argus, Pycinaster
angustatus, and my “ Pentaceros punctatus.” The last species is represented by
two dorsal ossicles from Les Clérimois (figs. 1 and 2). With the concurrence of
Dom Aurélien, I therefore take the original of his fig. 1 as type.

Locality and Stratigraphical Position—The specimens in the possession of
Dr. Blackmore are from East Harnham, Wilts., zone of Actinocamax quadratus.

3. PYCINASTER CRASSUS, n. sp. Pl. XXIX, figs. 1, 2, 2a, 3, 3a, 4, 4a, 9.

Specific Characters.—Body of large size. Height of marginalia not twice their
thickness. Median marginalia smooth. More distal marginalia with prominent
mammilations.

Material—There are about eight fragmentary specimens of this species in the
British Museum (Nat. Hist.). The specimen registered E. 2576 (Mantell Coll.)
shows considerable portions of the actinal surface, and that registered 35498
(Taylor Coll.) the dorsal view of a well-preserved portion of one arm. Both these
are figured on Pl. XXIX. Another specimen, registered E. 2628 (Mantell Coll.),
shows a portion of the abactinal surface of the disc. ‘The other specimens are

METOPASTER QUADRATUS. 97

mostly collections of isolated plates. The specimen registered 35498 is taken as
the type.

Description.—The abactinal surface of the disc appears to have been covered
with a number of plates of generally uniform size, with an average diameter of
about 3°38 mm. A few plates exceed this size, but in no case are they as large as
the corresponding plates in P. punctatus.

No specimen is sufficiently well preserved to give the proportionate lengths of
the major and minor radii, but there is no doubt that the arms were considerably
produced. <A row of hexagonal tabulate radialia are present throughout the greater
portion of the arm. The breadth of the arm at the base in the specimen registered
E. 2576 is about 22 mm. This specimen, however, judging by the dimensions of
the marginal plates, does not by any means appear to have attained the usual size
of the species. The length of its minor radius is 18°5 mm.

The median supero-marginalia are quite smooth and are distinguished from
those of all other species of the genus by their thickness (Text-fig. 24«).
In full-grown specimens they appear to be 18 mm. in breadth, 6 mm. in length,
and 10 mm. in thickness. More distally the supero-marginalia acquire large
mammilate rugosities.

The infero-marginalia are similar in character to the superior series.

Two rows of ventro-lateralia enter the base of the arm. Most of the ventro-
lateralia appear to have been rhomboidal in shape and of uniform size. They
possess very shallow hexagonal fittings indicating the former possession of
granules.

Locality and Stratigraphical Position —Upper Chalk, Kent.

Genus—METOPASTER, Sladen. (See p. 30.)
9. Meropaster QuADRATUS, n. sp. Text-figs. 1, 2, p. 98.

Specific Characters—Marginal plates in interbrachial areas almost square.
Raised area on marginal plates without spine-pits. Supero-marginal plates rugose
on interior surface. Ultimate supero-marginal plates may or may not be the
largest of the series, variation in this respect being especially marked. Abactinal
plates of dise with distinct stellate marking.

Material.—There are three fairly perfect and four fragmentary specimens of
this species in the collection of Dr. Blackmore, of Salisbury. Two of these are
figured in Text-figs. 1,2. The species was discovered by Dr. Blackmore, who
suggested the specific name “ quadratus” on account of the characteristic shape

98 FOSSIL ASTEROIDEA.

of the majority of the marginal plates. The type is the specimen figured in Text-
fig. 2.

Description.—The abactinal area of the disc is covered with hexagonal plates,
which have a distinctly stellate appearance on their upper surface.

The arms are distinctly produced. In the specimen figured in Text-fig. 1,
R:r::41 mm.:28 mm. In the specimen figured in Text-fig. 2, R:r:: 41 mm.:

Text-Fia. 1. Text-Fia. 2.

Trext-ric. 1.—a, Abactinal view of a specimen of M. quadratus, nat. size; b, actinal view of the
same specimen; c, view of infero-marginal plate magnified two diameters; d, view of two
supero-marginal plates magnified two diameters.

Trex-Fia. 2.—a, Abactinal view of another specimen of M. quadratus, nat. size; b, actinal view of
the same specimen; c, view of infero-marginal plate magnified two diameters ; d, view of two
supero-marginal plates magnified two diameters.

30 mm. The length of the side in the first-named specimen is 51 mm., in the
second-named specimen 56 mm.

The supero-marginalia are either five or six in number, counting from the
median inter-radial line to the extremity of the arm. In the inter-brachial area
they are distinctly quadrate in character, and are from 7 to 8 mm. in width and
from 6 to 7 mm. long.

The terminal paired supero-marginalia present very curious features. In
the specimen figured in Text-fig. 1 some of these plates are large and _ tri-

STAURANDERASTER ARGUS. 99

angular, just as in a typical Metopaster. Other terminal plates are, however,
small and approximate to those characteristic generally of Asteroidea. The
specimen figured in Text-fig. 2 presents no terminal plate which has a resemblance
to those typical of Metopaster. In all other respects the specimens are almost
exactly similar to one another.

The infero-marginalia are smooth and slightly concave in the centre. They
are eight in number.

The actinal area of the dise is covered with sub-equal plates, which are four-
sided in the inter-radial regions and tend to become hexagonal radially.

Locality and Stratigraphical Position—Zone of Actinocamax quadratus, East
Harnham, Salisbury.

Remarks.—The ornament of this species is identical with that of I. wncatus.
The important differences between the species lie in the shape of the marginal
plates, the character of the terminal supero-marginalia which show their especial
peculiarities in all the specimens, and the ornament of the abactinal plates of
the disc.

Family—PENTACEROTIDA (Gray), emend. Perrier, 1884. (See p. 76.)
Genus—STAURANDERASTER, novum. (See p. 125.)

12. S. arcus,n. sp. Pl. XXV, figs. 6,6a; Pl. XXIX, figs. 8, 8a, 9, 9a.

Specific Characters.—Ossicles ocellato-punctate. Surface of ossicles very
rarely truncate. If truncate, the flattened surface is not striated so as to simulate
the asteroid madreporite.

Material_—Only very fragmentary specimens of this species are known. The
best preserved specimen is in the collection of Dr. Blackmore, of Salisbury, and is
figured on P]. XXIX. Dr. Blackmore also possesses other specimens belonging
to this species. Two specimens of the species are also preserved in the British
Museum (Nat. Hist.), and bear the registered numbers KE. 5019 and E. 2566
respectively. The first-named specimen was presented by Mr. W. McPherson.
The second specimen was originally figured by Forbes in Dixon’s ‘ Geology of
Sussex,’ pl. xxi, fig. 16, as a “fragment of an Oreaster.” In Part III of this
Monograph it was figured under my direction (Pl. XXV, figs. 6, 6a) as Genus (?),
Sp. (?). I have now isolated two or three ossicles from the specimen, and they
are figured on Pl. XXIX, fig. 9. They show that the ossicles as originally figured
merely present their interior aspect. The specimen KH. 5019 is taken as the type.

15

100 FOSSIL ASTEROIDEHA.

Description.—The state of preservation of the fragmentary specimens of this
species only allows adequate description of the ossicles of the disc. These are
very uniform in character, and only differ from those of Stawranderaster ocellatus
in the absence of the truncated summit with madreporiform striations. The
largest ossicle measures about 6 mm. in diameter.

The isolated ossicles figured on Pl. XXIX show the characteristic shape of
marginal ossicles belonging to the genus Stawranderaster (see p. 120), and assist
us in ascribing not only this species, but also S. ocellatus, to which the species is
closely allied, to that genus.

Locality aud Stratigraphical Position—The specimen presented to the British
Museum (Nat. Hist.) by Mr. W. McPherson, is from the Marsupites zone at
Brighton. The specimens in the collection of Dr. Blackmore are from
Micheldever, Hants (zone of Micraster cor-anguinum).

Order—CRYPTOZONITA, Sladen, 1886.

Family—LINCKIIDAS, Perrier, 1875.

Cryptozonate Asteroidea, with comparatively well-developed marginal plates,
always contingent. Disc small, rays long and cylindrical. Abactimal skeleton
tessellate. Tegumentary developments granulate, superambulacral plates usually
present. Pedicellariz (rarely present) excavate or foraminate.

Genus—LINCKIA, Nardo, 1834.

Lincx1, Nardo, 1834. De Asteriis, Oken’s Isis, p. 717.

OPHIDIASTER (pars), Miiller and Troschel, 1890, Monatsber k. preuss. Akad. Wiss.
Berlin, p. 103.

Lincx1a, Gray, 1841. Ann. Mag. Nat. Hist., vol. vi, p. 284.

Acauta (sub. gen.), Gray, 1841. Tom. cit., p. 285.

Arms more or less cylindrical. Dorsal plates small, not arranged regularly in
longitudinal series. Two or three rows of granules on the adambulacral plates.
Superambulacral plates present. Papular areas distributed irregularly between
the dorsal plates.

1. Linckia, Psp. Pl. XXVII, figs. 1, 1a.

Material—aA distorted specimen, which very probably belongs to the genus
Linckia, is preserved in the British Museum (Nat. Hist.) (E. 5055, Capron Coll.).

OPHIURA. 10]

Description—The dise is small and very much distorted. The ventral
aspect of one arm is the only portion of the star-fish which affords much
opportunity for description. The arm is about 18 mm. long and 4mm. broad,
and possesses the cylindrical ¢haracteristic shape of the genus. It is composed
of a large number of square ossicles, which are superposed in the cross-section of
the arm, so far as it can be seen. They possess no spines, but a regular granula-
tion appears to run lineally across their breadth.

Remarks.—This specimen is not sufficiently well preserved to ascribe it to a
definite species of the genus.

Stratigraphical Position —Lower Chalk.

CRETACEOUS OPHIUROIDEA.
Order—ZYGOPHIURA, Bell (1892).

Ophiuroidea, in which the movement of the ossicles on one another is limited
by the development of lateral processes and pits. Superior, inferior, and lateral
spine-bearing arm-plates are always present. ‘The arms are simple and cannot
coil round straight rods.

Family—OPHIOLEPIDID As.

Zygophiure with oral papillae from three to six, of which the last may be
infradental. Arm incisures on the disc. Dental papille absent.

Genus—OPHIURA, Lamarck, 1801.

OpuHIuRA, Lamarck. Histoire Naturelle des Animaux sans Vertebres, vol. ii.
Ornioueris, Miller and Troschel, 1842. System der Asteriden (Braunschweig).
OruiociyPHa, Lyman, 1865. (Ophiuride and Astrophytide) Il. Cat. Mus. Comp.
Zool. Harvard, Nos. i, viii; 1882, Rep. Challenger Zool.,
vol. v.
— Verrill, 1899. Report on the Ophiuroidea collected by the Bahama
Expedition in 1893; Iowa City, Bull. Labor. Nat. Hist., p. 4.

Dise covered with plates or scales which are for the most part swollen.

Radial shields naked and swollen. Teeth. The inner mouth-papille long but
16

102 FOSSIL ASTEROIDEA.

becoming smaller and shorter towards the distal oral region, where they are
almost hidden by the scales of the mouth-tentacles. Arm-spines smooth and
short, seldom longer than an arm-segment. ‘Tentacle-scales numerous. The
innermost pair of tentacle-pores narrow, surrounded by numerous tentacle-scales,
and opening obliquely into the oral slits. In the back of the disc, where the arm
joins it, a notch usually edged with papillae. Two genital slits arise from the
sides of the mouth-shields.

The following species are placed provisionally in this genus, to which the
known characters would approximate them. The evidence, however, in every
case 1s incomplete.

1. Oruiura sERRATA, Roemer. Pl. XXVIL, figs. 3, 3.4, 3b, 3c, 3d, 3e.

OpHrurA sERRATA, Roemer, F. A., 1841. Die Versteinerungen norddeutsch.

Kreidegeb., p. 28, pl. vi, fig. 23.

— — Forbes, 1843. Proc. Geol. Soc., vol. iv, p. 234, fig. 2.

— — Reuss, 1845-6. Die Verstein. béhm. Kreide-form., vol. ii, p.
58, pl. xx, fig. 26.

— — Forbes, 1850. In Dixon’s Geology and Fossils of the Tertiary
and Cretaceous Formations of Sussex, p. 337, pl. xxiii,
figs. 2, 3, 3a, 3.

== — Morris, 1854. Cat. Brit. Foss., ed. 2, p. 84.

-- — Forbes, 1878. In Dixon’s Geology of Sussex (new edition,
Jones), p. 369, pl. xxii, figs. 2,3, 3a, 3b.

Specific Characters.—Conspicuous pear-shaped radials. Remainder of dorsal
surface of disc covered with scales. Upper arm-plates occupying great propor-
tionate width of arm. Six, occasionally seven ?, spines on each side of arm-segment.

Material—Only two fragments of arms were originally available for descrip-
tion by Roemer. The specimens described by Forbes were more nearly complete,
one specimen showing a considerable portion of the disc and the proximal portion
of four arms (figured on Pl. XXVII of this Monograph), now in the British
Museum (Nat. Hist.), and bearing the registered number E. 5043 (Dixon Coll.).
There are fragments of arms in several collections which can apparently be

ascribed to this species.

Description.—The disc is 15 mm. in diameter. Almost the whole of the dorsal
covering has disappeared in the specimen figured, thus exposing the inner surface
of the mouth-plates. The jaws (oral angle plates) are clearly seen. ‘lhey are
long and slender and do not meet inter-radially. The grooves for the water

OPHIURA PARVISENTUM. 103

vascular canal and for the nerve-ring; and the depressions for the first mouth-
tentacle and the entrance to the branch of the water vascular system are clearly
shown. Fragments of the scaly covering of the disc are seen scattered over the
dise. The peristomial plates are not recognisable, as is also the case in modern
species of Ophiura. The arms are 3 mm. wide at the base. The upper arm-plates
near the disc are broad. Six (or seven?) spines, which are in length about a
third of the length of the arm-segment, are present.

The vertebral ossicles are figured on Pl]. XXVII, figs. 3¢, 3d, 3e. They show
the typical Ophiurid structure as displayed by modern species of Zygophiurids.

Locality and Stratigraphical Position—Upper Chalk, Bromley, Kent.

2. Opntura Fitcant, n. sp., ee Forbes, MS. Pl. XXVII, figs. 2, 2a, 2b.

Specific Characters.—Body large and stoutly built. Dise covered with large
swollen plates. Radials contiguous, large, kidney-shaped. Upper and lower arm-
plates small.

Material—An external cast in flint, the sole remains of this species, is
preserved in the Norwich Museum (No. 2294).

Deseription.—Very little can be made out concerning the structure of this
ophiurid. The disc appears to have been surrounded with a circlet of large
radials. It has a diameter of about 16 mm.

A plasticine mould of the lower portion of the cast is figured on Pl. XXVII,
fig. 2a. The jaws (oral angle plates) are distinctly seen. They were long and
slender, and similar in form to those of O. serrata. Similarly a peristomial plate is
not visible. The inter-radial rounded buccal shield is clearly seen. ‘The arm is
45 mm. broad at the base. The impression of the cast of the arm appears to
indicate that both upper and lower arm-plates were small, as the side arm-plates
appear to meet in the dorsal and ventral median lines.

Locality and Stratigraphical Position —From flint gravel, Mousehold, Norwich.

3. OPHIURA PARVISENTUM, n. sp. Pl. XXVII, figs. 4, 4c.
’ P g

Specific Characters.—Dise covered with small plates, radials inconspicuous.
Proximal upper arm-plates only occupying about one third of dorsal surface of
arm. Five spines on each side of arm-segment.

104 FOSSIL ASTEROIDEA.

Material.—There is only one specimen of this species. This is preserved in
the British Museum (Nat. Hist.), and bears the registered number E. 5052
(purchased of Simmons).

Description.—The disc is about 15 mm. in diameter. It is covered with a large
number of small plates, which are rather scattered. The arms are 3 mm. broad
at the base. The upper arm-plate is much narrower than in O. serrata. There
are five spines, about three quarters the length of an arm-segment, on each side

of the arm.

Locality and Stratigraphical Position—Upper Chalk, Bromley, Kent.

Genus—OPHIOTITANOS,: novum.

Dise covered with plates which are small and sub-equal. Radial shields small,
triangular, naked, scarcely swollen. Arm-spines small. Mouth-shields large.
Side mouth-shields small, widely separated.

1. Opuio1rranos TENUIS, n. sp. Pl. XXVIII, figs. 1, 1a, 2, 2a.

Proximal upper arm-plates longer than broad. Spines very short, five in

number. Disc, with the exception of the radial plates, covered with an extensive
granulation.

Material.—The material for the description of this species consists of several
specimens in the British Museum (Nat. Hist.), The specimen E. 5056, which is
the type, is figured on P]. XXVIII, fig. 1, and the specimen E. 5057 on Pl. XXVIII,
fic. 2. There are also specimens registered E. 5058, H. 5059, 57512. All are ina
fair state of preservation. The first four specimens are from the Capron Coll.,
and the latter specimen was purchased from W. Griffiths. There are further
examples of the species in the Sedgwick Museum, Cambridge.

Description—The dise is flat on the dorsal surface, and its diameter in the
largest specimen is 4°7 mm. Lach pair of radial shields is separated by three
ornamented plates.

The mouth-shields are almost oval in shape. ‘The side mouth-shields are
small, and lie widely separated on the outer edges of the mouth-shields. There

1 Tiravoc= chalk.

OPHIOTITANOS LAVIS. 105

is a large number of granules in the oral region, but the tips of the jaws (oral
angle plates) can be seen just above the mouth-shields. There appear to be five
(or six) squarish mouth-papille.

The first under arm-plate is small and with rounded edges. Distally the
under arm-plates are at first almost square, then roughly pentagonal, and finally
triangular, with the apex pointing towards the disc. The side arm-plates meet
below at about the seventeenth arm-plate. The upper arm-plates are at. first
hexagonal, but rapidly become roughly triangular. They are tumid in appearance
and have a rounded base. They rapidly become smaller, and allow the arm-plates
to meet dorsally.

There are on each arm-segment five small smooth spines considerably shorter
than the length of a segment.

There appear to be two tentacle-scales, but the exact number is rather
difficult to determine.

Locality and Stratigraphical Position-—Lower Chalk, Folkestone and Dover.

2. OPHIOTITANOS L&VIS, n. sp. Pl. XXVIII, figs. 3, 3a, 4, 4a.

Specifie Characters.—Spines longer than arm-segments. Upper arm-plates
broader than long. Plates of dorsal surface of dise not hidden by granules.

Material—Only one specimen is known. This is preserved in the British
Museum (Nat. Hist.), no. E. 5053 (purchased of Mr. Griffiths), and shows the
dorsal aspect.

Description—The diameter of the dise is 48 mm., being thus about the same
size as Oplhiotitanos tenuis. The arm is, however, not so broad at the base,
measuring here only 1°5 mm. across. The specific characters given above
separate it sharply from this last-named species. I have not been able to
determine the number of spines on each arm-segment.

Locality and Stratigraphical Position —Lower Chalk, Dover.

Remarks.—A small specimen upon the slab, no. E. 5058 (Pl. XXVIII, figs. 4,
4), may be a young member of this species, as it possesses a strong general
resemblance to the above. It is peculiar in having only four arms, probably an
abnormality. As explained on p. 111, Ophioglypha bridgerensis, Meek, is very
similar in appearance to this species.

106 FOSSIL ASTEROIDEA.

3. OPHIOTITANOS MAGNUS, n. sp. Pl. XXVIII, figs. 5,5 a; Pl. XXIX, fig. 13.

Specific Characters—Proximal upper arm-plates broader than long. Spimes
very short, seven spines on each arm-segment. Disc, with the exception of the
radial plates, covered with an extensive granulation.

Material.—There are several specimens belonging to this species in the British
Museum (Nat. Hist.) and in the Sedgwick Museum at Cambridge. The specimens
in the British Museum bear the registered numbers E. 5060 (Capron Coll.), E. 5050,
K. 370, and E. 371 (all from J. Starkie Gardner Coll.). The first-named speci-
men, which shows the ventral aspect, is figured on Pl. XXVIII as the type. A
specimen from the Sedgwick Museum showing the dorsal aspect is figured on PI.
XXIX.

Description.—This species is the largest of all the known Chalk Ophiuroids, the
diameter of the disc being about 37 mm., and the breadth of the arm at the base
5 mm. There appear to have been two tentacle-scales.

Remarks.—Portions of the arm of this species are very similar in form to those
of Ophiura serrata. Unless the dise is present it is difficult to separate this
species from that form.

Locality and Stratigraphical Position —Lower Chalk.

NECTOPHIURA.

Sub-order.

Spines situated at an angle to the arm.

Family—AMPHIURIDA, Ljungman, 1867.

Zygophiure with oral papillae from one to five, of which the last is generally
infradental. Arms inserted on ventral side of disc. Dental papillae absent.

Genus—AMPHIURA, Forbes, 1842.
Disc small, delicate, covered with naked overlapping scales, and furnished

with uncovered radial shields. Teeth. Mouth-angles small and narrow. Arms
long, slender, even, and more or less flattened. Arm-spines short and regular.

EXTRA-BRITISH ASTEROIDEA. 107

1, AMPHIURA CRETACEA, n. sp. Pl. XXVIII, figs. 6, 6a.

o

Specific Characters.—Five (or six) mouth-papille. Two tentacle-scales. Five
arm-spines. Mouth-shields triangular in shape with a curved convex base, a
fair proportion of the. jaws (oral angle plates) showing on the ventral surface.

Material.—The one specimen known of this species is preserved in the British
Museum (Nat. Hist.), no. HE. 5059 (Capron Coll.).

Description.—The dise is 3 mm. in diameter.

There are five (or six) moderately stout blunt mouth-papille on either side of
the mouth-angle, and a triangular papilla situated infradentally. The side mouth-
shields are long and narrow, with proximal and distal sides parallel. They meet
in the middle. The extremities of the jaws are very obvious above these.

The arms are long and slender, and are 2 mm. broad at base. The width of
arm close to disc is 16 mm. The first under arm-plate is small, with slight
re-entering sides and a prominent median groove. The remaining under arm-
plates have a pointed proximal apex, and their sides are re-enteringly curved.
The distal side is convex. The side arm-plates meet proximally on the under
surface in the middle line.

The five stout tapering spines are rather longer than a joint of the arm, and
are situated on a proximal ridge.

Locality and Stratigraphical Position.—Lower Chalk, Folkestone.

CRETACEOUS ASTEROIDEA AND OPHIUROIDEA FROM
EXTRA-BRITISH LOCALITIES.

The following are the principal species from extra-British localities which
have not been mentioned in the previous portions of the Monograph. It will be
seen that where accurate information is available, these species are almost entirely
identical with British species.

ASTERIAS QUINQUELOBA, Goldfuss, 1826. ‘ Petrefacta Germanix,’ p. 209, pl. Lxiii,
figs. 5 a—u.

The illustrations given of this species are very beautiful and accurate, and
enable one to identify the ossicles ascribed to it as a mixture of ossicles of

108 FOSSIL ASTEROIDEA.

Metopaster Parkinsoni (a—p), Stawranderaster ocellatus (q—r), and Pentagonaster
megaloplax (s—u).

Various German writers have utilised the description of Goldfuss and this
specific name for the identification of isolated ossicles—e. gy. in Roemer, 1841,
‘Die Verstein. norddeutsch. Kreidegeb.,’ pl. vi, fig. 20, the ossicles ascribed to
A, quinqueloba are really ossicles of P. megaloplax; while in Reuss, 1845—6,
‘Verstein. béhm. Kreideform.,’ p. 58, pl. xlii, figs. 15—20, and in Geinitz,
1872—5, ‘ Paleontographica,’ vol. xx, pt. 2, pl. vi, fig. 7, the ossicles ascribed to
A, quinqueloba are really ossicles of M. Parkinsont.

Since the name quingueloba is prior to all the other names mentioned above, it
must be used instead of one of them. The simplest course appears to be to limit
it in the sense of Roemer, by fixing on the specimen represented in Goldfuss’s
plate Ixiii, fig. 54, wu, as type of Asterias quinqueloba. The result of this is to
replace the name Pentagonaster megaloplax, Sladen (antea p. 27, Pl. IV, figs. 2—4,
Pl. XIU, figs.la, 1), by Pentagonaster quinqueloba (Goldfuss).

AsveriAs JuRENSIS, Minster, 1826. In Goldfuss, ‘ Petrefacta Germanie,’ p. 210,
pl. Ixiu, figs. 6 a—h.

Figs. 6a—e represent a fragment and isolated ossicles which closely resemble
Calliderma Smithix, and figs. 6 f—h represent isolated ossicles bearing an equally
strong resemblance to Stauranderaster Boysti. The species, however, is said to be
“e calcareo jurassi Wurthembergia et Baruthino,’ whereas C. Smithiz and S.
Boysti are typical Cretaceous species, and have not in any other work been
described from Jurassic rocks. Without seeing the original specimens no one
should assert that Miimster was so far mistaken as to the horizon and locality of
the fossils before him. We can only suspend judgment.

ASTERIAS TABULATA, Goldfuss, 1826. ‘ Petrefacta Germanie,’ p. 210, pl. Ixiii.
figs. 7 a—g.

Figs. 7 a—b are illustrations of isolated ossicles of Stawranderaster argus
which are found in the Upper Chalk (zone of Micraster. cor-anguinum). I am
unable to recognise the illustrations of the remaining ossicles as appertaining to
any English Cretaceous species. The plates are said to be “e stratis argillaceis
superioribus calearei jurassi Baruthini.” - It is just possible that the locality and
stratigraphical horizon are wrongly given in the case of a and b, and the name
Asterias tabulata should be restricted to figs. c—g, one of those specimens being
taken as type.

EXTRA-BRITISH ASTEROIDEA. 109

Asrentas Scnvunzi, Cotta.

This is described and figured in Roemer, 1841, ‘ Die Verstein. Norddeutsch.
Kreidegeb.,’ p. 28, pl. vi, fig. 21, as follows: “ Fiinfeckig mit fiinf kurzen Strahlen,
unten vertieft und in der Mitte mit fiinf Erhabenheiten; der vorstehende Rand
gewolbt und zwischen je zwei Strahlenspitzen mit etwa 45 schmalen Tiifelchen
besetzt.””

The description reads as if the specimen belonged to the genus Sfauranderaster,
but neither this nor the figure given is much aid in the exact identification of the
species.

A cast of a fossil Asteroid is ascribed to the same species under the name of
Stellaster Schulzii, by Geinitz, ‘ Palzeontographica,’ vol. xx, pt. 2, pl. v, figs. 3, 4
This cast, however, looks like a cast of a species of either Calliderma or
Nymphaster.

Asrertas? Dounxert, Roemer, 1841. ‘Die Verstein. norddeutsch. Kreidegeb.,’
De Ads

This species is described as follows: ‘ Die Flichentiifelchen sind linger als
breit, 4—6 eckig, schriigrandig fein gek6rnt und nahe am obern Rande durchbohrt.”
Only isolated plates were known. Iam unable to identify the species with other
known Asteroids. The illustration given is not very clear.

According to Roemer, these isolated plates were described by Dunker and
Koch, 1837, ‘ Norddeutsch. Oolithgeb.,’ as plates of Cidaris variabilis.

Canaster Covunont, Agassiz, 1836. ‘Mém. Soc. Sci. Nat. Neuchatel,’ vol. i, p. 191.

No figure was given of this species by Agassiz. A fossil Asteroid figured by
Fritel, ‘ Le Naturaliste,’ vol. xvi, 1902, p. 79, as this species, appears to be some-
what like Nymphaster marginatus.

CupuLaster pauper, Frié, 1893. ‘Arch. Landesdf. Béhmen,’ ix, no. 1, p. 112.

This species is named from a specimen 3mm. in diameter, which so obviously
presents the large terminal plates which are common to all very young forms
of starfish, that it is useless to speculate as to its identity.

Locality and Stratigraphical Position. — Cretaceous (Priesener Schichten),
Waldek, near Bensen, Bohemia.

17

110 FOSSIL ASTEROIDEA.

GonIASTER MARGINATUS, Feuss, 1845—6. ‘ Verstemm. béhm. Kreideform.,’ p. 58,
pl. 43, figs. 21—32.

The illustrations of the isolated ossicles described as this species bear strong
resemblance to those of M. Parkinsont.

GONIAS'ER MAMMILLATA, Gabb, described by Clark, 1892, in ‘ Bull. U.S. Geological
Survey,’ no. 97, p. 32.

“ Determinative Characters——Body pentagonal, provided with a dorsal and a
ventral row of marginal plates that are narrower than igh, and distinctly tumid
on their outer surface. Only detached marginal plates preserved.”

Remarks.—The isolated plates undoubtedly belong to a species of Pycinaster
and bear a strong resemblance to those of Pycinaster angustatus.

* Locality and Geological Position.—Yellow Limestone of the middle marl bed
of the Cretaceous from Vincentown, New Jersey.”

Pentaceros pinaratus, S. Meunier, 1906. ‘Le Naturaliste,’ (2), vol. xx, p. 117.

The specimen described under this name is an external cast in flint, of which
a plaster cast has been presented to the British Museum (Nat. Hist.), E. 13075.
Owing to the courtesy of Professor Stanislas Meunier I have been able to examine
the original specimen. It shows a well-preserved impression of the abactinal
surface. The ossicles of the disc are rhomboidal or hexagonal, contiguous, of
almost uniform size, about 2 mm. in diameter. The specimen, therefore, cannot
be placed in the genus Pentaceros, and its appearance, measurements, and type
of ornament enable me to ascribe it to Pentagonaster obtusus.

OPHIOGLYPHA BRIDGERENSIS, Meek, described and figured by Clark, 1892. ‘ Bull.
U.S. Geological Survey,’ no. 97, p. 29.

‘ Determinative Characters—Dise composed of numerous small imbricating
plates. Upper arm-plates wider than long, the outer angles sharp and expanding

between the side arm-plates, which are slightly smaller. Under arm-plates long
and nearly rectangular in shape.

EXTRA-BRITISH ASTEROIDEA. 111

* Dimensions.—Diameter of dise 6 mm. Length of arm 20 mm. Width of
arm near dise 1:25 mm.”

Remarks.—The figures given and dimensions of this species appear to me to
indicate that it closely resembles Ophiotitanos levis. The upper arm-plates are
described as hexagonal, and the more proximal upper arm-plates of the British
species present this appearance.

Locality and Stratigraphical Position —Cretaceous, Fort Ellis, Montana.

OpHioGLyPHA TEXANA, Clark, 1892. ‘Bull. U.S. Geol. Survey,’ no. 97, p. 50.

“ Determinative Characters.—Dise round; composition indistinct. Arms long,
with wedge-shaped under arm-plates about as wide as long; upper arm-plates
about twice as wide as long.

“ Dimensions.—Diameter of disc 15 mm. Length of arm 50 mm. Width of
arm at disc 2 mm.”

Remarks.—It is difficult to identify this with a British species, but the illus-
trations appear to indicate that it is somewhat similar to Anphiwra cretacea.

Locality and Stratigraphical Position —Cretaceous, six miles north of Fort
Worth.

Opuiwura GRanutosa, Ioemer, 1841. ‘Die Verstein. norddeutsch. Kreidegeb.,’ p.
28, pl. vi, fig. 22.
This species is described as follows : ;
“ Die Arme sind walsenférmig und bestehen aus gewélbten, seitlich durch ein
Furche getrennten, deutlich gekérnten Seitenschildchen; wo sich deren vier
beriihren liegt ein kleines, dreieckiges Schildchen dazwischen.”

Remarl:s—The small fragment of the arm figured shows it to be a portion
of the distal extremity—quite possibly the distal extremity of some species of
Ophiotitanos.

Locality and Stratigraphical Position—Lower Chalk, Lindener Berg, near
Hanover.

112 FOSSIL ASTEROIDEA.

>

OpHiurA ? PULCHERRIMA, Fric, 1893. ‘Arch. Landesdf. Béhmen,’ vol. ix, no. 1,

p- 113.

Remarks.—No description is given. The figure does not show any resemblance
of this species to other known forms, but the specimen was obviously imperfect.
The upper arm-plates appear to be V-shaped.

Locality and Stratigraphical Position—Cretaceous (Priesener Schichten),
Waldek, near Bensen, Bohemia.

STELLASTER ALBENSIS, Greinitz, 1872—5. ‘ Paleeontographica,’ vol. xx, pt. 2, p. 16,

pl. vi, fig. 3.
Remarks.—This species is only known from a cast from the Quadersandstein.
In the absence of determinative characters it is impossible to say whether it is
identical with or differs from more fully described species.

-

Srettaster Coombu, Forbes, sp., m Geinitz, 187
pt. 2, p. 17, pl. vi, figs. 4—6.

2—5, * Paleontographica,’ vol. xx,

The specimens illustrated here as 8. Coombii certainly do not belong to Forbes’
species of that name. They appear to be ossicles of various species, but I am
unable to identify them from the figures given.

SPECIFIC AND GENERIC CHARACTERS IN CHALK ASTEROIDEA.

When I commenced this account of Cretaceous Asteroids I endeavoured, so
far as possible, to follow the generic classification of previous authors and
especially to preserve the continuity of Mr. Sladen’s work. More recent work,
however, has led me to believe that the shape of the marginal plates, together
with their ornament, affords the best determinative generic and specific characters,
and further enables us to identify almost all Cretaceous starfishes from single
isolated plates.

Some necessary revision as to nomenclature in both genera and species is given
below, together with an illustrated key-table, which it is hoped will enable zonal
collectors to identify the isolated asteroid plates which are commonly met with in

ASTEROIDEA. 113

almost all exposures. Up to the present our knowledge of the zonal occurrence
of these forms has been limited, as complete specimens of starfishes are exceed-
ingly rare. There appears to be no reason now, however, why our knowledge
of the zonal distribution of these forms should uot become as nearly complete as
it is, for example, in the case of Kchinoids.

I must thank Dr. Blackmore, of Salisbury, for his invaluable suggestions to me
concerning this means of identification.

Ornament.—The ornament of starfishes consists of calcareous pieces, which
may be spinous in form, or scaly, or granular. These may occur :

(1) Embedded in the living tissues outside the general body-plates, but not in
contact with the plates themselves. On the death of the animal they become
dispersed on the disintegration of the living tissues, and such ornament is therefore
rarely visible in fossil specimens.

(2) Articulated to the plates. In this case they are situated :

(a) Hither in a depression of the plate ;

(b) Or in a depression upon a raised eminence
of the plate. Occasionally in this latter case
the depression may be excavate in the centre
in order to allow a strong muscular attach-
ment. In this case the eminence may

2 ee re simulate the perforate tubercle of an Echi-
Nymphaster Coombii, showing spines noid such as Cidavis.
— In almost all cases in Cretaceous Asteroids the
ornament is of the type 2 a.

Generally the movable articulated pieces have
disappeared, but in such cases the depression on
the plate which they formerly occupied is readily
visible (compare Text-fig. 3).

_I purpose to call all such movable articulated
pieces, whether they are spinous or granular in
character, “spines,” and, at the suggestion of

-. Bather, the depressions on the plate “ spine-pits.”
Texr-ric. 4.—Isolated marginal of Dr. Bather, t 2 I ; : P I i

Riawrandevaster, bulbiferus, showing The character of the spine-pits appears to be

the “ pustulate” type of ornament. ; . a :
constant in character in each individual species.
Thus, e.g. in N. Coombii (Text-fig. 3) they show a coarse honeycomb structure,
uniform in character over the whole of the plate. In Stauwranderaster bulbiferus
(Text-fig. 4) the spine-pits are circular and widely spaced. This latter type is
interesting, as it apparently occurs only in the genera Metopaster and Sfauran-
deraster. The spine is very small, and barely projects over the edge of the deep

114 FOSSIL ASTEROIDEA.

spine-pit, thus giving the plates an embossed appearance. This type of ornament
I call the “ pustulate”’ type.

In previous portions of this Monograph both Mr. Sladen and myself have

assumed that if no spine-pits are present on a plate they have been weathered

away. It now appears that the absence

of spine-pits is such a constant character

in certain species that this supposition

. can no longer be held, and the absence of

spine-pits indicates an original absence

Text-ric. 5.—Pedicellaria of Nymphaster of spines, or, at any rate, spies articu-

nigoniae lated to the plates. In support of such a
conclusion it can be urged that, generally
e
'] é ®) « speaking, Chalk fossils are but little
i) : :
weathered, and that there is evidence
Text-ric. 6.—Three pedicellaria from Pentago- derived from a study of recent forms
nasler quinqueloba on the right. A pedicellaria ;
from Hadranderaster abbreviatus on the left. Pedicellariz.—As can readily be under-

stood, only pedicellarize which are articu-
———<™—— lated in depressions of the plate are pre-
served in Cretaceous Asteroids. Purse-
Text-ric. 7.—A pedicellaria from Metopaster like, valvate pedicellariz (Text-fig. 5)

Purkinsoni in the centre, on the left a pedi-

cellaria of Pycinaster senonensis, on the right a of a generalised type are common to
pedicellaria of Stawranderaster coronatus. :

many genera. More specialised pedicel-
larize, however, peculiar to the genera Metopaster and Pycinaster are also met
with (Text-fig. 7).

KEY-TABLE FOR THE IDENTIFICATION OF CRETACEOUS
ASTEROIDS.

The following key-table is based on the shape of the marginal plates and the
character of the spine-pits on them. Generally speaking both superior and
inferior series are similar in these respects, but when otherwise a note is made in
the table.

A short description is also given of various plates which cannot be adequately
treated in the table.

All the plates figured in the table are magnified 4 diameters.

It is convenient to consider the Chalk (Cenomanian-Senonian) species
separately from the Upper Greensand forms. No Cretaceous Asteroidea have
been described from below this horizon.

—

KEY-TABLE. 115

CuHatk (CENOMANIAN-SENONIAN) SPECIES.
I.—Marginals four-sided, with sides rectilinear or almost rectilinear; broad.

a. Without a rabbet-edge. Calliderma, Nymphaster, Pentagonaster.
s. With a rabbet-edge. Metopaster, Mitraster.
c. With a distinct ridge. Arthraster.

Il.—Marginals either hexagonal or rounded; very thick, Hud randeraster.
Il].—Marginals wedge-shaped, high, spine-pits very shallow or absent. Pycinaster.
IV.—Marginals breast-plate-shaped. Stawranderaster.

V.—Miscellaneous plates.

I.—Marginals four-sided, with sides rectilinear or almost rectilinear. Pedi-
cellarie when present of a simple bi-valvate character.

a. Without a rabbet-edge. Calliderma, Nymphaster, Pentagonaster.

1. Spine-pits shallow, hexagonal, giving a honeycomb appearance.

a. Honeycomb medium or fine. C. Smithise (see p. 123).
Text-fig. 8 a.—Variety with medium-sized spine-pits
(see p. 123).
Text-fig. 8 B.—Variety with fine spine-pits (see
p. 1238).

TExtT-Fia. 8,

b. Honeycomb coarse. N. Coombii. Text-fig. 3, p. 113.

i =
(= )

Be 5% c. Honeycomb confined to a portion of the plate (or
absent). P. obtusus. Text-fig. 9.

Text-ric. 9.

116 FOSSIL ASTEROIDEA.

2. Spine-pits shallow, circular, circles variable in size, adjoming.

One species, C. latum. Text-fig. 10.

Text-Fia. 10.

3. Spine-pits deep, circular, not adjoining.

a. Spine-pits coarse. N. marginatus. Text-fig. 11.
b. Spine-pits fine, not on margin. WN. oligoplax. Text-fig. 12.

Text-Fia.11. | Terxt-ria. 12.

c. Spine-pits fine, uniformly over the whole of plate. P. lunatus.
Text-fig. 13.

Trext-Fic. 13.

4, Spine-pits on outer edge of plate with raised margins.

One species, P. quinqueloba (see p. 108). Text-fig. 14.

Trxt-Fia. 14,

KEY-TABLE. 117
| 5. No spine-pits present.
(4) (8)
a> a. Proximal marginalia almost smooth ; distal
Ue with granular rugosities arranged in a linear
series. N.radiatus. Text-fig. 15, (a) pro-

ximal marginal, (B) distal marginal.

Text-ria. 15.

b. Marginalia with rugosities not arranged in
linear series. N. rugosus. Text-fig. 16.

Trext-ria. 16,

B. Marginals with a rabbet-edge. Rabbet-edge covered with small spine-
pits. Pedicellariz when present “ winged” (see Text-fig. 7).

1. Spine-pits on central raised area.

a. Central raised area smooth on both superior
and inferior marginalia, Spine-pits on
infero-marginalia uniformly situated. M.
Parkinsoni. Text-fig. 17, (a) outer view
of marginal, (8) side view of marginal.

Text-Fic. 17.

b. Central raised area of supero-marginalia
rugose; that of infero-marginalia smooth,
with spine-pits in form of network. M.
Hunteri. Text-fig. 18, (4) supero-marginal,
(8) infero-marginal.

ce. Central raised area of supero-marginalia
smooth in young, rugose in mature indi-
viduals. Lower Chalk form. M. cornutus
(see p. 124).

Text-rio. 18.

18

118 FOSSIL ASTEROIDEA.

2. No spine-pits on central raised area.

a. Rugosities present on greater portion of surface
of supero-marginalia. M. rugatus. Text-fig.
19, (A) supero-marginal, (B) infero-marginal.

b. Rugosities confined to inner edge of supero-
marginalia or absent; outer portion of plate
tumid. a. Plates oblong. M. wncatus. Text-
fig. 20, (4) supero-marginal, (B) infero-mar-
ginal. £8. Platessquare. M. quadratus.

c. No rugosities present on the supero-marginalia
which are not tumid on their outer portion.
M. compactus. See Pl. XXVI, fig. 35.

Text-Fig. 20.
N.B.—The infero-marginalia of the above species are difficult to distinguish except by
their dimensions. The reader is advised to refer to the detailed description for these.

c. Marginalia with a distinct ridge which has granular elevations along
its base. Arthraster.

(a) (8)
Brn ; >> 1. Upper surface of ridge smooth. A. Dizoni.
/ { Text-fig. 21, (a) outer view of marginal, (B)
Ea, | side view of marginal.
| ? 2. Upper surface of ridge with spine-pits. A. cris-
ed tatus. See Pl. XXIX, fig. 10.

Text-Fic. 21.

II.—Plates either hexagonal or rounded ; very thick. Spine-pits form a distinct,
fine honeycomb marking. Hadranderaster (see p. 125).

There is only one species. H. abbreviatus. Text-
fig. 22, (a) outer view of marginal, (B) side
view of marginal.

Text-FiG. 22.

KEY-TABLE. 119

Ill.—Marginals generally wedge-shaped, high, spine-pits very shallow or absent.
Pedicellarie when present with five valves round a deep central
depression. § Pycinaster.

1. Plates of maximum height, 10
mm. P. angustatus. See Pl.
IX, fig. 1 a.

2. Plates of maximum height, 20
mm., f-shaped in profile. P.
senonensis. Text-fig. 23, (a)
outer view of marginal, (B)
interior view of marginal, (c)
side view.

Text-ric. 23.

3. Plates very thick, often oblong,
not wedge-shaped. P. crassus.
Pl. XXIX, fig. 3, exterior view
of marginal. Text-fig. 24, side
view of marginal.

Text-Fia. 24,

IV.—Marginal plates breast-plate-shaped. Stawranderaster (see p. 125).

A. With spine-pits.

1. Spine-pits fairly coarse. S. bul-
biferus (commonly occurring).
Text-fig. 25.

2. Spine-pits medium. S. bipune-
tatus (only one specimen
known). ‘Text-fig. 26.

Text-Fic. 25. Text-Fic. 26.

120 FOSSIL ASTEROIDEA.

3. Spine-pits fine. S. Boysi. Text-fig.
27, (A) outer view of marginal, (B)
interior view of marginal.

Trext-Fic, 27.

b. Without spine-pits.

1. Maximum size about 7 mm. _ S. coro-
natus. Text-fig. 28.
2. Maximum size about 2°55 mm. _ S.

a —_ squamatus.
w 3. Maximum size about 2°5 mm. S.
= “4 pistilliferus. Text-fig. 29.

N.B—It is impossible to distinguish
isolated marginalia of these two species.
Compare V, 3 (p. 121).

TextT-Fia. 29.

TrExt-FIG. 28.

V.—The collector may also come into possession of the following plates :

1. Large triangular plates having the characteristic ornament of Metopaster.

These are the ultimate supero-marginalia which characterise the genus. See e. g.
Pl. XVI, fig. 2 a.

2. Large hemispherical plates with a flattened base. These are met with in
the following species:
A. With spine-pits.

Stawranderaster bulbiferus, S. Boysi,
and ? S. bipunctatus. These are
distinguished from one another
by their spine-pits, which are of the
same character as those met with
in the marginalia. Text-fig. 30=
a primary inter-radial of S. bulbi-
ferus.

B. Without spine-pits.
Pycinaster senonensis, distinguished by
the smooth surface, Pl. XXIX,
fig. 6.
Arthraster Dizoni, distinguished by the
rugose surface, Pl. XXIX, fig. 11.

Trxt-Fia. 80.

KEY-TABLE. 121

3. Nodular plates with an excavate margin met with in the following species :
A. With flattened base, from 7—9 mm. in diameter, S. coronatus, Pl. XXIV,
fig. 2.
3—4 mm. Ps S. squamatus, Pl. XXV,
fig. 3.
p. With produced base, N. pistilliferus, Pl. XXV, fig. 5.

399 22 99

4. Oblong or almost oblong plates having the characteristic ornament of NS.
bulbiferus. These are the more distal marginals, or in some cases the ventro-
lateralia of this species. Generally the marginals are indented at the corners, and
their shape can be decided from the characteristic breast-plate form.

Text-fig. 4, p. 113.—Outer view of plate,
Text-fig. 31 —Side view of plate.

Text-ria. 31.
5. Irregularly rounded or polygonal plates with ocellate depressions having a

raised ridge.

A, Without madreporiform markings on summit. S.argus. Pl. XXIX,
fig. 8a.

s. With madreporiform markings on summit. S. ocellatus. Text-
fig. 32.

Text-Fria. 32.

6. One specimen of Astropecten? sp. and Linckia sp.? respectively have also
been described, see pp. 90 and 100.

Upper GRrEENSAND Forms.

The genus Comptonia appears to be characteristic of this horizon. The plates
are very similar in shape to those of Calliderma, except that they are more
rounded in profile. Unfortunately, no specimen at present known shows the
ornament of the marginalia.

122 FOSSIL ASTEROIDEA.

I have also met with C. Smithiw and P. punctatus from this horizon. They
are readily distinguished by the characters which have already been given in
the key-table.

NOTES ON THE KEY-TABLE.

Genera—CatLipermMaA, NyMPHASTER, PENTAGONASTER.

The following are the chief distinctive characters which separate these three
genera in recent forms.

CaLLIDERMA. PENTAGONASTER. NYMPHASTER.
1. Arms well produced. Arms slightly produced. As in Calliderma.
2. Abactinal area covered with closely As in Calliderma. Abactinal area covered with
fitting plates. paxille, which are not
closely fitting in the radial
areas.

3. Ventro-lateral and infero-marginal Ventro-lateral and infero- As in Pentagonaster.
plates with prominent spines. marginal plates without
prominent spines.

4. Armature of the adambulacral Armature of the adambu- As in Pentagonaster.

plates consisting of 14-16 small lacral plates arranged in
spines arranged uniserially, longitudinal series. Series
with three or four rows of on the whole uniform in
larger spines arranged rather character.

irregularly.

The fossil species of Calliderma possess the characters numbered 1 and 2, but
differ to a greater or less extent in characters 3 and 4, in which they resemble
Pentagonaster. The genus Calliderma was, however, founded by Gray on one
species—O. emma. No other recent species has been assigned to the genus. It is
difficult, therefore, to say how far the distinctive characters of the Cretaceous
genera should have generic value. The question is debated by Mr. Sladen on p. 5
of this Monograph, and the very striking general resemblance of the fossil species
to the recent C. emma influenced him in his decision to group them under this
genus. There appears no great reason to dispute this assignment, but I am in
more doubt as to the systematic position of the species which have been placed in
the genera Pentagonaster and Nymphaster. It will be seen that as the fossil
species of Calliderma resemble Pentagonaster in characters 3 and 4, the only
distinctive character which remains between the two genera is the length of the

—

NOTES ON GENERA. 123

arm. Two of the fossil species of Pentagonaster, namely P. lunatus and obtusus,
have all arms which are well produced (the arms in the specimen of P. lunatus
figured on Pl. IV, fig. 1, are broken off short). The third and remaining species,
P. quinqueloba, is usually much more pentagonal in shape, although a specimen in
the possession of Dr. Blackmore has a major radius at least twice the magnitude
of that of the minor radius.

The species assigned to the genus Nymphaster by Sladen were so assigned
because their structure and character, so far as they could be made out from the
fragmentary condition of the fossils, appeared to warrant their inclusion in the
genus Nymphaster (see p. 15).

It appears to me that these species have the same generalised characters as
those assigned to the genera Calliderma and Pentagonaster. The distinctive
character of the genus Nymphaster is the possession of paxille on the abactinal
plates. No fossil species is sufficiently well preserved to show whether these were
absent or present, and it is impossible therefore to confirm or deny Sladen’s
suggestion.

It will be seen from the above that there is no certain evidence which entitles
us to distribute the Cretaceous species amongst the three genera, and it may be
the task of a future observer to place them in one new genus. I have, however,
in order to secure uniformity, utilised all these generic names even for the
description of new species. The following species also appear to require revision.

Calliderma Smithix, C. mosaicum.

After examination of the fairly numerous specimens of the fossils assigned to
these species in the British Museum (Nat. Hist.) I cannot confirm the specific
distinctions made by Sladen on pp. 10 and 11 of this Monograph. All the
characters mentioned vary greatly in individual specimens. ‘The ornament,
however, is common in character to both species, and I should prefer to unite them
in one species, namely, U. Smithix, as this has prior place in the original account
given by Forbes.

The specimens figured on Pl. VII, figs. 1 and 2, and stated by Sladen to be in
his opinion doubtful examples of N. Coombii (p. 17), should in my opinion be
assigned to CO. Smithix, as should also the specimen figured on Pl. XIX, fig. 3.
These examples possess a finer type of honeycomb structures on their marginalia
than is usually met with in C. Smithix, and they may, therefore, be a distinct
variety of this species (see 'Text-fig. 8).

A specimen preserved in the British Museum (Nat. Hist.) E. 5063, was figured
by Sladen on PI. V, fig. 1 a, of this Monograph as V'omidaster sulcatus. Apparently
it was the intention of Sladen to make a new genus and species for the reception

124 FOSSIL ASTEROIDEA.

of this fossil. The dimensions of the marginalia, compared with the minor radius
and the ornament, are precisely the same as in C. Smithix, from which the
specimen only differs in the possession of numerous valvate pedicellariz. In view
of the somewhat freakish way in which pedicellariz occur in starfishes, it does not
appear to me that this character alone entitles us to make a specific or generic
distinction on behalf of this specimen.

There is another specimen also in the collection of the British Museum (E. 1116)
which has similarly numerous valvate pedicellariz, but which shows the abactinal
aspect. In all other respects the specimen cannot be distinguished from a
specimen of C. Sinithiz.

Genus—MEVOPASTER.

In the key-table I have only distinguished four species of Metopaster, namely,
M. Parkinson’, M. uncatus, M. quadratus, and M. cornutus, the latter being a
doubtful species. If one examines collections of Cretaceous Asteroids, one finds
that practically all the specimens have been rightly assigned to these species.
Specimens which could be assigned to the species M. Mantelli, M. Bowerbankt,
M. zonatus, M. sublunatus, M. cingulatus (see pp. 38-55), are very rarely met with.

The very considerable variation which occurs in undoubted specimens of
M. Parkinsont m the number of the supero-mar-

i caer, . ginalia, their form, amount of ornament, and the
‘ 7,” shape of the ultimate plates of this series, makes
., specific characters founded solely upon these charac-
| tae ters of doubtful validity, particularly as such varia-

tions occur even in an individual specimen, and it
Homage Meat ean pga te is upon a rather extreme variation of these charac-
EE cag pucckcnncd aid ters occurring in very few specimens that this large
that usually met with. number of species have been described. On the
other hand, the presence or absence of spine-pits on the raised central area of the
plate is a constant character in species of Metopaster. Two specimens figured in
the Monograph appear to belie this statement. The specimen figured on Pl. X,
fig. 4a, shows no spine-pits on its infero-marginalia, but is figured as M. Parl:insoni.
I have isolated a dorsal ossicle, which shows the specimen undoubtedly to belong
to M. wicatus. The specimen figured on Pl. XI, fig. 3 a, as M. wneatus, shows
spine-pits on its supero-marginalia in one inter-radius only. After very careful
examination of this specimen, I have come to the conclusion that this inter-radius
—the right-hand upper inter-radius of the figure

has been added by a dealer
from a collection of ossicles of M. Parkinsoni to an imperfect specimen of M. wncatus.

NOTES ON GENERA. 1

bo
cr

(Jenus—STAURANDERASTER,’ NOVUM.

Prnvaceros (pars). Pp. 76-89 of this Monograph.

The species bulbiferus, Boysii, coronatus, ocellatus, bispinosus, pistilliferus, and
squamatus (pp. 76—89 of this Monograph), which have been formerly placed in the
genus Pentaceros, together with the new species argus (p. 99), should, I think,
now be ascribed to a new genus. The plates of these species are breast-plate-
shaped, at times almost cross-shaped, and bear a characteristic type of ornament
(see p. 113). In both these respects and in the absence of papular areas the
species differ widely from species of recent Pentaceros, with which the only feature
they have in common is the circlet of raised plates on the abactinal surface of the
dise.

The type species of the new genus is Stawranderaster Boysii, and its diagnostic
characters are :

Arms high, well produced, marginalia breast-plate-shaped, generally with a
rabbet-edge free from ornament, and a raised interior area, either smooth or with
ornament of an embossed type. A circlet of swollen plates present on the
abactinal surface of the disc.

This genus, with the following genus, may be placed provisionally in the family
Pentacerotidee.

Genus—HADRANDERASTER,” NOVuM.

Penvaceros (pars), p. 86 of this Monograph.

The species described as Pentaceros abbreviatus on p. 86 of this Monograph
differs so considerably in the shape of its marginalia and their ornament from the
species of Stawranderaster and recent species of Pentaceros that 1 have placed it in
a new genus.

The type species is Hadranderaster abbreviatus, and the diagnostic characters
of the new genus are:

Arms high, well produced, marginalia either hexagonal or rounded, very thick,
ornament spread uniformly over the surface of the plate. Pedicellaria bi-valvate.

1 Gravpds = a cross, dvdnpoy = a raised garden border.
* dépdés = stout, avdqpor = a raised garden border.

19

126 FOSSIL ASTEROIDEA.

GENERAL CHARACTERISTICS OF CRETACEOUS ASTEROIDEA AND
OPHIUROIDEA.

The majority of Cretaceous starfishes belong to the Phanerozonate forms
included in the families Pentagonasteridze and Pentacerotidz. Modern forms of
the genera of these families are widely distributed geographically, but, generally
speaking, they are characteristic of warmer waters than those of the English
Channel of to-day.

The Chalk starfishes are specialised types which, although approximating to,
are not identical with, modern genera. The differences, at any rate in some cases,
appear to be distinctly physiologically advantageous.

Metopaster and Mitraster, the most abundant of all Chalk starfishes, possess
not only a specialised type of ornament but also characteristic massive plates.
The arms tend to become shortened and the disc correspondingly enlarged.

The Chalk species of the Pentacerotide differ from the modern forms,
inasmuch as they are more strongly built; the abactinal areas are not reticulate,
and all species possess intermarginalia which cause the characteristically deep
body of these forms. ;

The Chalk is a deposit formed in seas which were sufficiently distant from land
to avoid any great admixture of clay or sand. Globigerinw and other forms of
pelagic Foraminifera floated in abundance on the surface of the sea, which, because
of its temperature, must have been exceedingly favourable to prolific forms.
In the circumstances there must have been an abundance of food for starfishes,
and we find, therefore, that the long-armed, comparatively active Astropectinide,
which were so characteristic of the Jurassic shallow water deposits, are displaced
by more sedentary forms which tend to specialise, so as to obtain, by the enlarge-
ment of the disc or development of intermarginalia, the largest possible space for
their digestive organs.

The irregular Echinoids which are so characteristic of the Cretaceous seas are
similarly sedentary forms.

The fossil Ophiuroidea also closely resemble modern forms. The isolated
vertebral ossicles of O. serrata, figured Pl. XXVII, figs. 3 ¢, 3 d, 3 e, cannot be
distinguished from the ossicles of recent Ophiuroids. Complete specimens of
Ophiuroidea and Asteroidea are rare, but isolated plates are very numerous in
the Upper Chalk. They are, on the contrary, rare in the Lower Chalk,
according to experienced collectors, as, for example, Mr. Dibley.

The following starfishes are found in the zones indicated. The list is compiled

PHYLOGENY. 127

from the papers of Dr. Rowe (‘ Proc. Geol. Assoc.,’ vols. xvi, xvii), and also
from notes furnished by Dr. H. P. Blackmore and Mr. T. H. Withers.

| = 7 =
sre is |.3] a et
Sees es) e.g |s |SS) 33 Seles
S | Sa) 52] S| se] _se/8 [&8| 2 [Re] se|
Bg /SFs Sse SS SES /SES |S e/Sslsse ssl ss
NSEINMEINGTSINE (NAS [NSE IN SINGS INGO IN EINE |
| = =e SS SS le
| Nymphaster Coombei, Forbes the Ir x
— oligoplax, Sladen | x
Pentagonaster quinqueloba,
Goldfuss | ... | x m x x ee See et)
= obtusus, Forbes - ea x x te -
Metopaster Parkinsoni, Forbes | x x x ee | ees:
— uncatus, Forbes . ~ x x x
— quadratus, 1. sp. x —
— cornutus, Sladen oe x
Mitraster Hunteri, Forbes . x Dae Be
— rugatus, Forbes. |) cee x x So) Fone
Pycinaster angustatus, Forbes as me x x x 3 |
— senonensis, Valette .| ... ve x eal
Stauranderaster bulbiferus, Forbes |...) xX x xX a
— Boysi, Forbes Peal cee x ,
— ocellatus, Forbes .| ... | X > x x Sas
-- pistilliferus, Forbes)... | X x Soa Il oe x
— argus, Q. Sp. . NY ee era bene ve Ss x aes ees ae
Arthraster Dizoni, Forbes . a lGoes te se x brs re el Wate lhe 4
= cristatus, 0. sp. . Billesse ae rs Bi x

THE PHYLOGENY OF THE CRETACEOUS ASTEROIDEA.

If we examine the various species of a genus or group of related genera of
Cretaceous Asteroids, we find that there is a similar transition from smooth to
spinous forms through an intermediate form, to that which has been observed in
Ammonites and Brachiopods.

Both in Ammonites and Brachiopods single specimens show the whole life-
history of the individual, for the shell of the animal is not materially altered in
character after it has once been formed. It is therefore possible to show, e. q.,
that the character of the ornament of the shell of an Ammonite was smooth
in its infancy, costate in its adolescence, spinous in the adult, and it has also been
shown that this life-history depicted by the individual is an epitome of the
phylogenetic history of the species (Buckman, ‘Mon. Ammon. Inf. Oolite,’ Pal.
Soc., 1905, p. cc). Similar observations have been made with regard to Brachio-
pods (Buckman, ‘ Quart. Journ. Geol. Soc.,’ vol. lxiii, 1907, p. 338) ; primitively the
Brachiopod shell is smooth externally, more advanced forms are progressively
costate and then spinous. Occasionally species may regress towards a primitive

128 FOSSIL ASTEROIDEA.

plain form through a costate phase. The progression or elaboration is known as
“ anagenetic ’ development, and the retrogression as “ catagenetic.”

It is regrettable that our present state of knowledge of Chalk Asteroids does
not allow us to recognise such definite phylogenetic series as those obtainable in
Brachiopods or Ammonites. The plates of an Asteroid are constantly being eaten
away and replaced by new calcareous matter, so that the adult plate may differ
considerably in character from its young phase. An opportunity for study,
however, is afforded by the fact that all the plates are not formed at once. The
more distal plates are younger than the proximal plates, and therefore resemble
more closely those of the young form. The resemblance is not, however,
quite exact, as they are formed later in the life of the individual, and may show
consequently characters which have appeared later in the history of the species.
Doubtless, if it were not for the paucity of the well-preserved specimens of
Cretaceous Asteroids much might be still made out by a comparative study along
these lines.

The following paragraphs are only suggestions made in the hope that more
material may come to light at a future date. The great majority of starfishes are
and have been spmous forms, and I propose to assume that the origmal ancestor
in each group was spinous.

Genera—METOPASTER AND MITRASTER.

METOPASTERS. MirRastTeErs.
M. quadratus M. uncatus M. Parkinsoni M. Hunteri M. rugatus M. compactus
(rugose ) (rugose) (spinous) (spinous, with (rugose) (smooth)
rugosities )

M. cornutus
(spinous, with rugosities)

Original spinous form.

It is convenient in this group to consider the ornament on the raised central
area of the marginalia. It will be seen from the above diagram that the spinous

PHYLOGENY. 129

form, which was the ancestor of these two groups, early gave rise to two offshoots,
one of which includes the Metopaster species, the other the Mitraster species.
Both genera retained evidence of common ancestry by the possession of similar
ornament of the specialised “ pustulate” type, and by the similar appearance of
their ultimate supero-marginalia, which, instead of being smaller than the rest of
the superior marginal series, are as large or larger than these. MM. cornutus, which
is the only species found in the lower zones of the chalk (Turonian), shows that
early specialisation set in. This form is spinous without rugosities when young
(see p. 55, Pl. XIV, fig. 5), but older specimens (see Pl. XXIX, fig. 12) acquire
rugosities.

The species of Metopaster, M. quadratus and M. wneatus, have lost all spines
from the raised central area of their supero-marginalia, and, instead, possess
rugosities. The raised central areas of the infero-marginalia possess neither
spines nor rugosities, but are quite smooth. M. quadratus has also acquired, as
a frequent variation, a primitive type of ultimate supero-marginalia (see p. 98).
The type of ornament shows the species to be highly specialised, and this fact,
together with its occurrence in the higher zones of the Chalk, affords us an
explanation of the remarkable peculiarity of the terminal supero-marginalia on
the supposition that it is a catagenetic tendency.

The species of Mitraster show even more decided evidence of the three phases
—spinous, rugose, smooth. These alterations only occur on the raised central area
of the supero-marginalia. The infero-marginalia appear to pass directly from the
spinous to the smooth stage without the intervention of a rugose stage.

Genera—CatLiperMA, NYMPHASTER, AND PENTAGONASTER.

This group tends to become smooth both in the Senonian and in the Turonian-
Cenomanian.

Turonian-Cenomanian forms include C. Smithiv, C. latum, N. Coombii, N.
oligoplaa, and N. marginatus, which are spinous; N. rugosus, which is rugose; NV.
radiatus, which is smooth on the older proximal plates, but rugose on the younger
distal plates.

Senonian forms include P. quinqueloba and P. lunatus, which are spinous; P.
obtusus, which very often possesses marginalia which have lost the majority of the
spines and are almost smooth.

130 FOSSIL ASTEROIDEA.

Genus—STAURANDERASTER.

S.argus 8S. ocellatus
(spinous) (spinous)

S. squamatus
(smooth) <2

S. pistilliferus | S. Boysii S. bipunctatus S. bulbiferus
(smooth) | (spinous) (spinous) (spinous)

S. coronatus
(smooth)

Original spinous form.

The left-hand stem and branches of the diagram above are occupied by the
generalised species which have long tapering arms. They show the transition
from spinous to smooth forms. The spinous form of these species is S. Boysit.
In the lower zones of the Chalk one species (8. coronatus) appears. This has
neither spines nor rugosities on the majority of its marginalia, although a few
distal marginalia are rugose. S. squamatus, which is almost identical in character
with S. coronatus, except that it is of smaller size, and NS. pistilliferus (arms not
known) are the smooth forms which characterise the upper or middle zones of
the Chalk.

The right-hand stem and branches are occupied by various specialised forms
from the upper and middle zones of the Chalk. WS. bulbiferus shows a specialisa-
tion in the bulbiform character of the extremity of its arms, and S. bipunetatus in
the character of the spines on the ventro-lateralia. S. ocellatus and S. argus are
specialised in the peculiar nature of their armature. All these forms are spinous,
their specialisation lying in other directions.

Genus—HADRANDERASTER.

The majority of the plates on this form are spinous, although a few distal
plates are smooth, probably indicating a catagenetic tendency in this direction.

GLOSSARY. 131

Genus—PYCINASTER.

Spines are very feebly developed in this genus. The spine-pits when present
are very shallow, and often they are absent altogether. Spine-pits are often
visible on the actinal plates after they have disappeared from the abactinal series.

P. crassus possesses rugosities on the distal marginalia.

Genus—ARTHRASTER.

A, cristatus possesses both spines and rugosities; A. Divont is rugose without

spines.

Speculation as to phylogeny in these latter three genera, in view of the state
of our knowledge, would be valueless.

GLOSSARY.

The following glossary and diagram (Text-fig. 34) is added to aid the geologist
who has but little acquaintance with modern zoological terms.

Abactinal.—Applied to the surface which is uppermost when the starfish walks
on its tube feet; the term “ dorsal” is used by some authors in the same sense.

Actinal——Applied to the surface which is undermost when the starfish walks.
On this surface are situated the mouth and the ambulacral grooves. The term is
used synonymously with “ ventral”? by some authors.

Adambulacralia or Adambulacral Plates.—The ossicles which are adjacent to the
ambulacral ossicles. In the order “ Phanerozonia,”’ to which the great majority of
Chalk Asteroids belong, these ossicles are visible on the actinal surface, bordering
the ambulacral groove and hiding the ambulacral ossicles. Adambulacralia may
be recognised by their prominent armature of spines.

Adradialia.—Ossicles situated on either side of the radialia (7. v.).

Ambulacral.—The ambulacral groove is the groove stretching from the mouth
to the extremities of the arm. It is formed by the ambulacral ossicles, which meet
in the middle so as to form an arch. The tube feet project through the arch and
into the groove.

132 FOSSIL ASTEROIDEA.

Centrale-——The most central ossicle on the abactinal surface of the disc.
This ossicle, together with five ossicles situated inter-radially and called the
“Primary Inter-radialia,” are especially prominent in the young form, in which
they often occupy almost the whole of the abactinal surface. Generally speaking

fe the five a
7 primary interradialia

(Se
pe

\
\ Radialia

Gieea ialia
~--S uperomarginalta

Text-ria. 34.—Abactinal view of a specimen of Stawranderaster bulbiferus, natural size, slightly restored from
the specimen from Charlton, Kent, registered E. 4344 in the British Museum (Nat. Hist.).

they can be distinguished in the adult form by their larger size, and occasionally
they are especially prominent, as in species of Stawranderaster (Text-fig. 34) and

of the recent genus Pentaceros.

Marginalia—In adult forms of the order ‘ Phanerozonia,” which includes

Palxontographical Society, 1908.

A MONOGRAPH

ON THE

BAT PSH rae si L

HCHINODERMATA

FROM

THE CRETACEOUS FORMATIONS.

VOLUME SECOND.
THE ASTEROIDEA AND OPHIUROIDEA.

BY

Wek, see NCL, B.A, F.GS:

PART FIFTH.

Pages 133—138; TirLe-pace anv INDEX.

PRINTED FOR THE PALHONTOGRAPHICAL SOCIETY.

1908.

S 4
to
. =
;
PRINTED BY ADLARD AND SON, LONDON AND DORKING.
7
A :
4
7
i
7
= .
. a

ADDENDA ET CORRIGENDA. 133

almost all Cretaceous Asteroids, and in quite young forms of the order * Crypto-
zonia,” the margin of the dise and arms is bordered by specially prominent plates
by

—the “ Marginalia.” The abactinal series are called “ Supero-marginalia,” and the

actinal series the “ Infero-marginalia.”
Primary Inter-radialia—See Centrale.

Radialia.—The abactinal series of plates along a major radius are called the
radialia.

Radius.—A line drawn from the central point of the dise to an extremity of
the arm is called the “ Major radius,” R.A line drawn from the central point of
the dise to a point half-way between two radii is called the “ Minor radius,” vr.
This is sometimes called an “ Inter-radius.”

Spine-pits.—Depressions in a plate for the articulation of spines (see p. 118).

Ventro-lateralia.—The plates on the actinal surface of the Asteroid excluding
the infero-marginalia and the adambulacralia. In the inter-radial regions these
plates are often rhomboidal. A typical view of an isolated plate of this descrip-

tion is given (Pl. XXIX, fig. 4).

ADDENDA ET CORRIGENDA.

Page 24, line 19, for Scliilze read C. I’. Schulze.

Page 26, Locality, ete., for Upper White Chalk near Norwich read Hard Chalk,
West Norfolk, the precise locality wikiown.

Page 67, line 11, for Goniaster compactus read Goniaster (Goniodiscus) compactus,
and omit all reference to Forbes, 1848.

Page 69, line 15, for Stellaster comptoni read Coniaster (Stellaster) Comptont.

Page 71, line 4, for Stellaster elegans read Goniaster (Stellaster) elegans.

Page 89, line 8 from end, for Bourguetiocrinus read Bourqueticrinus.

Page 90, last line, for Upper Greensand read Lower Chalk.

Page 95, line 3, for Sladen read Forbes.

Page 95, line 13, for (p. 89) read (p. 89, Pl. XXV, fig. 7).

Page 101, line 9 from end, the first reference should read Orntura, Lamarck,
1801. Systeme des Animaux sans Vertebrés, p. 350. The date of the reference
given is 1816,

Page 102, line 12, for 1841 read 1840.

20

134 FOSSIL ASTEROIDEA.
Page 103, line 10, add cor-anguinum zone, Northfleet, Kent, and Blandford,
Dorset.

Page 103, line 11, add in synonym: Oputura serrata? Forbes, 1843. Proce.
Geol. Soc., vol. iv, p. 234.

Page 1035, line 4 from end, for parvisentum read parvisentis.

Page 106, line 7, after 15060 add and 5061; under Locality insert Polkestone.

Page 117, line 12 from end, for M. read Metopaster.

Page 117, line 7 from end, for M. read Mitraster.

Page 117, line 2 from end, for M. read Metopaster.

Page 118, line 3 from top, for M. read Mitraster.

Page 118, line 7 from top, for M. read Metopaster.

Page 118, line 9 from top, for M. read Metopaster.

Page 118, line 12 from top, for M. read Mitraster.

Page 119, line 3 from end, for bipunctatus read bispinosus.

Page 120, line 13 from end, for bipunctatus read bispinosus.

Page 130, table and line 7 from end, for bipunctatus read bispinosus.

Page 121, lines 2, 4, 6, for S. read Stawranderaster.

Pl. IV, figs. 2—4, for Lower read Upper.

PL. V, fig. 1, for Tomidaster sulcatus read Calliderma Smithix (see p. 123).

Pl. VU, figs. la and 2a, for ? Nymphaster Coombii read Calliderma Smithix
(see p. 122).

Pl. X, fig. 4, for Metopaster Parkinsoni read Metopaster wncatus (see p. 124).

Pl. XIX, fig. 3, for Nymphaster Coombii read Calliderma Simithix (see p. 122).

Pl. XXT, fig. 2, for Pentagonaster robustus read ? a young form of Pycinaster
angustatus (see p. 95).

Pl. XXIV, fig. 1, for Pentaceros abbreviatus read Hadranderaster abbreviatus
(see p. 125).

Pl. XXV, fig. 2, for Upper Greensand read Lower Chalk!

Pl. XXV, fig. 6, for Genus ? sp. ? (p. 93) read ? Stauranderaster arqus (p. 99).

Pl. XXV, fig. 7, for Pentaceros ? vn. sp. (p. 89) read Pycinaster angustatus
(pp. 89, 95).

Pl. XXV, fig. 8, for marginal read internal.

Pl. XXVI, fig. 1, for Pentaceros punctatus read Pycinaster senonensis (see p. 95).

Pl. XXVI, fig. 4, for Calliderma mosaicum read Pycinaster angustatus (see
p. 95).

Pl. XXVI, fig. 4, for Mom the Lower Chalk read From the Upper Chalk.

Pl. XXVII, fig. 3, for abactinal read actinal or adoral.

Pl. XXVIT, fig. 3.¢, for side read right side.

' Mr. H. Woods informs me that recently he has been able to match the matrix in which this

fossil is embedded.

se

ADDENDA ET CORRIGENDA. 135

Pl. XXVIL for Ophiura parvisentum read Ophiura parvisentis.

Pl. XXVII, fig. 4, before natural size insert slightly less than.

Pl. XXLX, fig. 12, for Sladen sp. read Sladen.

On all Plates (except XXVI) for Caulliderma mosaicum read C. Smithix (see
p- 122).

On all Plates for Metopaster Bowerbankii, M. Mantelli, M. zonatus, read M.
Parkinsoni (see p. 124).

On all Plates for Metopaster cingulatus read M. uncatus (see p. 124).

On all Plates for Pentaceros bulbiferus, I’. Boysti, P. coronatus, P. bipunctatus,
P. squamatus, P. pistilliferus, P. ocellatus, P. argqus, read corresponding species of
Stauranderaster (see p. 125).

On all Plates for Pentagonaster megaloplax, read P. quinqueloba (see p. 108).

LN DEX.

PAGE PAGE
Abactinal ey _.. 181 | Calliderma ... ne 235 ... A, 122, 129
Acalia,...*° : “aa sae ON) — latam? "=. of 12,116; 11, m
Actinal ra yas an isi - mosaicum on ON 123 eysevaseval
Adambulacral plates fe: i a BHI — Smithiz ... ae 6. e238 ear van
Adradialia ... Ff ee _ ... 131 | Centrale ae = a. A 3 vn Be
Ambulacral ... ibe aa had ... 181 | Ceelaster Couloni ... a! ahs may HDS,
Amphiura... oer Aor he ... 106 | Comptonia ... see aa ae i 169
= cretacea ... aa sen WOhApesrerraubt — comptoni wee 69 ; xvii, xvill
Amphiuride ... Re apr re: pa) O06 — elegans ... oe se 71; xvii
Arthraster... : : : 91,131 | Cryptozonia ... is Sh ae sn) LOD
— cristatus ... we 93,118; xxix | Cupulaster pauper... He a ... 109
— Dixoni... ... 91,118; xvili, xxix
senonensis ss 2h rie 98 | Dorigona Rte “ie wea Lae ee (GE
Asterias Dunkeri ... eo as LOO
— jurensis... DE oe soo 1Mhs) :
— lunatus ee Bar ae oe 2D Huasteroidea.... 3
— quinqueloba a so a OF
— Schulz ... cn: me ... 109 | General characteristics 5 Boe oe lgo
— tabulata ... us is ... 108 | Generic and specific characters... ee elle
Astrogonium ... ie ie .. 24 | Glossary i Be Sis es: celal
— angustatum SA _. 22 | Goniaster ae Lae ase a 24, 76
Bowerbankii ... aes a AD -- angustatus a he ee neal
= Coombii EAE he oe pele) — Bowerbankii ae Nf sie a
_- compact —... a Se OY, — compactus a ae OG
Hunteri a at oa) 59) -- Coombii ... ah on a malo
latum ... 12 — Hunteri... — Me pt)
lunatum ar ones — latus aS ae Fic oe aes
Mantelli = ea Co aitaies — lunatus... $k a Pe ek
—— mosaicum oh i : 9 — mammillata ee rt: a peg
Parkinsoni ... : oe | — Mantelli ... as re ». 938
—- rectilineun —... ai eet ea — marginatus 44 mi fie (0)
— rugatum = ae os — mosaicus ... ae of - 9
Smithii a Rac fa 6 — Parkinsoni fie Ae me Oe
sublunatum 51 — rectilineus ne au eon:
uncatum ihe a rere FEL, -- regularis ... sa: ae Sin ROS
Astropecten ... : mE San -- rugatus ... a aah oS
Astropectinidie * ae ae e 90 a semilunata xe es ns 38

PAGE
Goniaster Smithii 6
— sublunatus 51
_ uneatus 47
Goniodiscus ee 24
-—— Bowerbankii 42
-- Hunteri 59
-- Mantelli 38
— Parkinsoni 32
-- rectilineus 32
-- rugatus ...
= sublunatus 51
— uneatus... 47
Hadranderaster oa a 125, 13
abbreviatus... 114, 118, 125
Hosia ... 24
Key-table 114
Linckia 100
Linckiide 100
Marginalia i ee ERR)
Metopaster ; 30, 97, 124, 128
— Bowerbankii ... 423 XV, Xvi
— cingulatus 53; xiv
— cornutus ... TAO Ais AV |
— Mantel, ... ... 88; xiii |
—- Parkinsoni 31, 114,117, 124; ix—xii,
Xvi
— quadratus 97, 118
= sublunatus vs ad ; 51
— uneatus ... ... 47,118; xi, xiv, xv
— zonatus 45 ; xii
Mitraster as ee : 57, 128
— compactus ... ... 67, 118; xvii, xxvi
— Hunteri SOO es Ik, KU) Xy.|
-— rugatus 63, 118; xvi
Nectophiure: ... at : 106 |
Nymphaster ... ... 14, 73, 94, 122, 129
— Coombii 15, 113 ; vii, viii
— marginatus ... 18,116; viii
—_ oligoplax 19, 114, 116 ; viii |
= radiatus 73, 117; xxv
— rugosus 94, 117; xxix

INDEX.

Ogmaster
Ophidiaster
Ophioglypha ...
a bridgerensis
texana ...
Ophiolepididee
Oplhiolepis
Ophiotitanos ...
laevis
-- magnus
— tenuis ...

Ophiura

— Fitch
eranulosa
parvisentum
— pulcherrima
— serrata
Ophiuroidea ...
Oreaster

— Boysii

— bulbiferus
— coronatus

- ocellatus

— obtusus :
— pistilliformis
— pistilliferus ...

Ornament

Pedicellarix

Pentaceros ser
— abbreviatus
— bispinosus

— Boysii ... 80

bulbiferus
— coronatus
— dilatatus
— ocellatus ...
— pistilliferus
punctatus
— senonensis
-—- squamatus
Pentacerotide
Pentagonaster
— lunatus
megaloplax
obtusus

quinqueloba ...

PAGE

69

100

101

110

11]

101

101

.. 104

105; xxviii

106 ; xxviii, xxix
104; xxviii

101
XXVil
111
XXVil
112
102: xxvii
101

76

80

108 ~

103:

114
Fe, 76
86, 125; xxiv

87, 125; xxiii

5; XNil, XXill, XXvVi

110

85, 125; xxv
88, 125; xxv
... 88,95; xxvi
81, 95

83, 125; xxv
te 76, 99
24, 73, 122, 129
25, 116; iv

. 27,108; iv
74, 115; xxii
108, 114, 116

9]
-

>
2
v

138
PAGE
Pentagonaster rectilineus... e fon
— regularis —... ne we) xo
~- robustus 733 XXxi
— semilunatus ... .: we le
Pentagonasteride ... 1B sf Se titce ta
Pentagonasterme ... : 5 ee 4,
Phanerozonia ... oe c ec 3
Phylogeny 127
Primary inter-radialia 133

95, 131

Pycinaster ok :
= angustatus 93, 95, 119; ix, xxi, xxv,
XXxvl
— senonensis 95, 114, 119; xxvi, xxix
Pyenaster : nen 21, 95
— angustatus 21, 95; ix, xxi, Xxv, xxvi
= crassus 96, 119; xxix
Radialia 133
Radius 133 |
Specific and generic characters 112
Spine-pits

. 99, 125, 130
99; xxv, Xxix

Stauranderaster

= argus

133 |

INDEX.

PAGE
Stauranderaster bipunctatus (= bispinosus) 119
ae es
— Boysii 120, 125
= bulbiferus 113,119, 120, 121, 125

-— bispinosus ...

— coronatus 114, 120, 125

— ocellatus 121, 125

— pistilliferus 120, 125

~- squamatus ... 120, 125

| Stellaster sh. a cud == 969
— albensis ... ee en oer le

— Comptoni ... we or eGo)

—- Coombii 112

— elegans... a3 = ae

-— Schulzii 109
Tomidaster sulcatus ... ae one noe eS
Tosia <.. sat ihe ae a ee ee
— lunata ... ase oa) an see. eo
— regularis oe a3 ee ae oll
Ventro-lateralia 133
Zygophiure ... ee a Se an 20

ADLARD AND SON, IMPR., LONDON AND DORKING,

PLATE I.
CaLLIpeRMA Smiraia#, Forbes, sp. (P. 6.)

From the Lower Chalk.
Fra.
la. Actinal aspect; natural size. (Coll. Brit. Mus.)
b. Lateral view of the margin ; natural size.
c. Infero-marginal plates ; magnified 2 diameters.
d. Lateral view of the marginal plates ; magnified.
e. Adambulacral plates; magnified.
f. Actinal intermediate plates; magnified 3 diameters.

delet hth

PLATE II.
CaLuipEeRMA LATUM, Forbes, sp. (P. 12.)

From the Lower Chalk.

. Actinal aspect; natural size. (Coll. Brit. Mus.)

. Infero-marginal plates; magnified 2 diameters.

. Actinal intermediate plates; magnified 3 diameters.

. Adambulacral plates ; magnified.

. A mouth-plate; magnified.

. Actinal aspect of another example, with a portion of the actinal floor removed,

showing the stellate bases of the abactinal plates or paxillz ; natural size.
(Coll. Brit. Mus.)

. Actinal surface of part of the ray; magnified 15 diameters.
. Infero-marginal plates; magnified 3 diameters.
. Stellate bases of the abactinal plates or paxille; magnified 4 diameters.

sen LE >

yi
7

Fie.

PLATE ITI.
CaLLIDERMA LATUM, Forées, sp. (P. 12.)

From the Lower Chalk.

. Actinal aspect of an example from Amberley ; natural size. (Coll. Brit. Mus.)
. Lateral view of the margin ; natural size.

. Lateral surfaces of the marginal plates ; magnified.

. An infero-marginal plate ; magnified.

. An actinal intermediate plate ; magnified.

. Actinal aspect of an example from the Chalk Marl of Dover ; natural size.

(Coll. Brit. Mus.)

b. An adambulacral plate; magnified.

. Actinal aspect of an example from Washington, with a portion of the

actinal floor removed, showing the stellate bases of the abactinal plates or
paxille ; natural size. (Coll. Brit. Mus.)

. Stellate bases of the abactinal plates or paxille ; magnified.

ill

Pl

ab

AH Searle delet lith

a. r
9 ae a a

Mh ie

b.

. An infero-marginal plate ; magnified.

xo Qa OS

PLATE: TV.
PENTAGONASTER LUNATUS, Woodward, sp. (P. 25.)
From the Upper Chalk.
Actinal aspect of the type specimen; natural size. (Coll. Norfolk and

Norwich Mus.)
An actinal intermediate plate ; magnified.

PENTAGONASTER MEGALOPLAX, Sladen. (P. 27.)

From the Lower Chalk.

. Actinal aspect of the example figured by Forbes, under the name of (roniaster

(Astrogonium) lunatus ; natural size. (Coll. Brit. Mus.)

. An infero-marginal plate ; magnified 3 diameters.

. Lateral view of the margin; natural size.

. Actinal aspect of another example ; natural size. (Coll. Brit. Mus.)

. An actinal intermediate plate ; magnified 6 diameters.

. Adambulacral plates ; magnified 6 diameters.

. Actinal aspect of an example from the Upper Chalk of Bromley; natural

size. (Coll. Brit. Mus.)

. Lateral view of the margin; natural size.

. Lateral surface of an infero-marginal plate ; magnified.
. An infero-marginal plate ; magnified.

. An actinal intermediate plate ; magnified 6 diameters.

IV

AH Searle delet hth

so Qa om 8

PLATE V.
TomipasTer suLcatus, Sladen.

From the Grey Chalk.

. Actinal aspect ; natural size. (Coll. Brit. Mus.)
. Actinal intermediate plates; magnified.
. Adambulacral plates ; magnified 3 diameters.

CaLLIDERMA Mosatoum, Forbes, sp. (P. 9.)

From the Lower Chalk.

. Abactinal aspect ; natural size. (Coll. Brit. Mus.)

. Lateral view of the margin; natural size.

. Abactinal plates; magnified.

. Supero-marginal plates ; magnified.

. Madreporiform body and surrounding plates; magnified.

PI

dearle delet hth

AH.

Spy LW ae

e-Gs Vivt
ie Ai,

PLATE VI.
CALLIDERMA mosalcuM, Forbes, sp. (P. 9.)

From the Grey Chalk.
Fie.
1. Abactinal aspect of an example from which the abactinal plates have been
removed; natural size. (Coll. Brit. Mus.)
2a. Abactinal aspect of another example from which the abactinal plates have
been removed, showing the inner surface of some of the actinal intermediate
plates and adambulacral plates ; natural size. (Coll. Brit. Mus.)
b. Adambulacral plates, seen from within ; magnified.
c. Actinal intermediate plates, inner surface ; magnified.

PI

VI

Fia.

. Actinal aspect; natural size. (Coll. Brit. Mus.)
. Lateral view of the margin; natural size.

. Adambulacral plates; magnified.

. An infero-marginal plate ; magnified.

PLATE VII.
(?) Nympaasrer Coomsu, Forbes, sp. (P. 15.)

From the Lower Chalk.

. Actinal intermediate plates ; magnified.
. Abactinal aspect of an example from the Grey Chalk at Folkestone; natural

size. (Coll. Brit. Mus.)

. A supero-marginal plate; magnified.
. Abactinal aspect of an example from the Lower Chalk of Glynde ; natural

size. (Coll. Brit. Mus.)

. A supero-marginal plate ; magnified.

CaLLIDERMA Mosarcum, Forbes, sp. (P. 9.)

From the Lower Chalk.

. Abactinal aspect; natural size. (Coll. Brit. Mus.)
. Lateral view of the margin; natural size.
. A supero-marginal plate; magnified.

P) Vol

earie Geil. et utr

>>

a A

: ot ’ y, LD
; ioat 1 PYGs hy

4 a ant y

PLATE VIII.
Nympuaster Coomsit, Forbes, sp. (P. 15.)
From the Lower Chalk.
Fia.

la. Actinal aspect of the type specimen; natural size. (Coll. Brit. Mus.)
b. An infero-marginal plate; magnified 3 diameters.

CaLtiperMa Smitui#, Forbes, sp. (P. 6.)
From the Lower Chalk.

2a. Profile view of a fragment of a ray; natural size. (Coll. Brit. Mus.)
b. Actinal view of the same; natural size.
c. An adambulacral plate; magnified 6 diameters.

NyYMPHASTER oLIGoPLAX, Sladen. (P. 19.)
From the Upper Chalk.

3a. Abactinal aspect ; natural size. (Coll. Brit. Mus.)
b. A supero-marginal plate; magnified.

NYMPHASTER MARGINATUS, Sladen. (P. 18.)
From the Upper Chalk.

4a. Abactinal aspect; natural size. (Coll. Brit. Mus.)
b. A supero-marginal plate ; magnified 4 diameters.

PI. Vill

AH Searle delet, lith

PLATE IX.
PYONASTER aNGUSTATUS, Forbes, sp. (Page 21.)

From the Upper Chalk.

. Abactinal aspect; natural size. (Coll. Brit. Mus.)
. Lateral view of the margin ; natural size.

Meropaster Parxinsoni, Forbes, sp. (Page 31.)

From the Upper Chatk.

. Actinal aspect; natural size. (Coll. Brit. Mus.)
. Lateral view of the margin ; natural size.

An infero-marginal plate ; magnified.

Mirraster Hunter, Forbes, sp. (Page 59.)

From the Upper Chalk.

. Abactinal aspect of a small example with four supero-marginal plates ;

natural size. (Coll. Brit. Mus.)

. Actinal aspect of the same ; natural size.

Lateral view of the margin; natural size.

. A supero-marginal plate; magnified.

An infero-marginal plate ; magnified.

PIX

PLATE X.
Mevoraster Parxginsoni, Morbes, sp. (Page 31.)
From the Upper Chalk.

Abactinal aspect ; natural size. (Coll. Brit. Mus.)

. Abactinal aspect of another example; natural size. (Coll. Brit. Mus.)
. A supero-marginal plate; magnified 3 diameters.

The madreporiform body; magnified.

. An abactinal intermediate plate, with pedicellarian apparatus; magnified.
. Actinal aspect of another example; natural size. (Coll. Brit. Mus.)
. Lateral view of the margin; natural size.

Adambulacral plates ; magnified.

. Actinal aspect of another example; natural size. (Coll. Brit. Mus.)
. An infero-marginal plate ; magnified.

c. An actinal intermediate plate ; magnified.

. Abactinal aspect of a young (?) example; natural size.

. A supero-marginal plate of the same ; magnified 4 diameters.

. Abactinal intermediate plates ; magnified.

3a.

PLATE XI:
Meropaster Parxrnsoni, Forbes, sp. (Page 31.)

From the Upper Chalk.

. Abactinal aspect; natural size. (Coll. Brit. Mus.)

. A supero-marginal plate ; magnified 3 diameters.

>, The madreporiform body ; magnified 6 diameters.

. Abactinal aspect of another example; natural size. (Coll. Brit. Mus.)
. A supero-marginal plate; maguified.

An abactinal intermediate plate ; magnified.

MeropastEr uncatus, Forbes, sp. (Page 47.)
From the Upper Chalk.

Actinal aspect of a specimen from which the whole actinal floor has been
removed, showing the inner surface of the abactinal floor; natural size.
(Coll. Brit. Mus.)

. Abactinal intermediate plates, inner surface ; magnified.

delet: hth

CaP ie

AH

PLATE XII.
MeropastEeR Parxinsoni, Forbes, sp. (Page 31.)

From the Upper Chalk.

. Abactinal aspect; natural size. (Coll. Brit. Mus.)
. Lateral view of the margin; natural size.

. A supero-marginal plate ; magnified.

. The madreporiform body ; magnified.

. Abactinal intermediate plates ; magnified.

Meropaster zonatus, Sladen. (Page 45.)

From the Upper Chalk.

. Abactinal aspect; natural size. (Coll. Brit. Mus.)
. Lateral view of the margin; natural size.

The madreporiform body ; magnified.

Mitraster Huntert, Forbes, sp. (Page 59.)

From the Upper Chalk.

. Actinal aspect ; natural size. (Coll. Brit. Mus.)
. Lateral view of the margin; natural size.

. A supero-marginal plate ; magnified.

. An infero-marginal plate ; magnified.

. Actinal intermediate plates ; magnified.

CaLLIDERMA MosatcuM, Forbes, sp. (Page 9.)

From the Lower Chalk.

. Actinal aspect ; natural size. (Coll. Brit. Mus.)
. Lateral view of the margin ; natural size.

Mouth-plates ; magnified.

1. An infero-marginal plate ; magnified.

HSearle del et hth

Hanhart imp.

ath

Fia.
IL a.
b.

b.
4a.

PLATE XIII.
PENTAGONASTER MEGALOPLAX, Sladen. (Page 27.)
From the Upper Chalk.

Abactinal aspect; natural size. (Coll. Brit. Mus.)
A supero-marginal plate; magnified 3 diameters.

Meropaster Mantenti, Forbes, sp. (Page 38.)

From the Upper Chalk.

. Abactinal aspect ; natural size. (Coll. Brit. Mus.)

A supero-marginal plate (much weathered) ; magnified 3 diameters.
The madreporiform body and surrounding plates ; magnified.

. Abactinal intermediate plates ; magnified 6 diameters.
. Abactinal aspect of the example figured by Forbes; natural size.

Brit. Mus.)
A supero-marginal plate ; magnified 3 diameters.

Actinal aspect of another example figured by Forbes; natural size.

Brit. Mus.)

. An infero-marginal plate; magnified 3 diameters.

(Coll.

(Coll.

Fa,

Hanhart imp

earle del et lith

Ss
te)

AH

Aa.

d.

PLATE XIV.
Meropaster uncatus, Forbes, sp. (Page 47.)

From the Upper Chalk.

. Abactinal aspect ; natural size. (Coll. Brit. Mus.).
. An ultimate paired supero-marginal plate ; magnified.
. Abactinal aspect of another example; natural size. (Coll. Brit. Mus.)

Lateral view of the margin; natural size.
A supero-marginal plate ; maguified.
The madreporiform body ; magnified 4 diameters.

Actinal aspect ; copied from the figure given by Forbes in Dixon’s ‘ Geology

of Sussex’ (pl. xxi, fig. 5).

Mevoraster cincunatus, Sladen. (Page 53.)
From the Upper Chalk.
Abactinal aspect; natural size. (Coll. Brit. Mus.)
Lateral view of the margin ; natural size.

A supero-marginal plate; magnified.
An abactinal intermediate plate ; magnified.

Mertopaster cornutus, Sladen. (Page 55.)

. Abactinal aspect; natural size.

Lateral view of the margin; natural size.
An ultimate paired supero-marginal plate seen from above; magnified.
The same seen in profile (lateral view); magnified.

AH Searle. del.ebJith Hanhart. imr

: vores
ak os

aoe
J

PLATE XV.

MErtopastER uNcaATUS, Forbes, sp. (Page 47.)
Fie.
la. Actinal aspect; natural size. (Coll. Wright.)
b. Lateral view of the margin; natural size.

Meropaster BoweErBaNKil, Forbes, sp. (Page 42.)
From the Upper Chalk.

2a. Actinal aspect; natural size. (Coll. Geol. Survey.)
b. Lateral view of the margin; natural size.
c. An infero-marginal plate (weathered) ; magnified.
d. Mouth-plates ; magnified.

Mirraster Huntert, Forbes, sp. (Page 59.)
From the Upper Chalk.

3a. Abactinal aspect; natural size. (Coll. Brit. Mus.)

b. A supero-marginal plate; magnified 3 diameters.
4a. Actinal aspect of another example; natural size. (Coll. Brit. Mus.)

b. An infero-marginal plate ; magnified 3 diameters.

c. Extremity of the radial region; magnified 3 diameters.

d. Abactinal intermediate plates, seen from within ; magnified 6 diameters.
5a, Actinal aspect of another example; natural size. (Coll. Brit. Mus.)

b. An infero-marginal plate; magnified 3 diameters.

AHSearle dalet lith

Hankart imp

| )

o hag? : Pb,
ta

= a At Ovi Siameat :

3 4
ifs

2, inks inte ge eh ah thebg nenela fees, a ot ;
at rit 5%; a

mn : ce) Sua er ae ne % |
a 7 7 See cian at ? “We cv) es fae r rs Pa tal gata A ie
. ; mith SSF) 4 a7 v a haley mn bale vg owrcuate Hise n
; ! i 7 7 ‘ :
: 4 7 ’

; _
a Yen ya) eee baci, see
'

a

oY bewiegey A lint wall © acy» chi eRe ees Rta’ tau i te “8

fa Met J) trea es vildryeee tft ty
PERT L Seats intl iz Wee ee ae hy ee ey ee
7 as ow a 7 ’
mn > Code ee ey” .
7 7 . - ; ‘ 5 - a 7 a
rd ug CAS BME? igh dee Col Pee a :
a i eS 7 — hele
n : oe - age
7 7 ' 44 o
a i.) eA ed slap
~ A a 4 *

" “9 be na BS ne sbi did h Saris i a il a ve ae ts nT 1 he y

| : _ sei DY it ees % pel a) yeaar -_ ot
Bice tanets &: a signe ar Bb leety bi, honlst a
: a : if uu vin -

Ji die

) “e4eh hee
‘Bai . jaar
74) ie ss :
iat oP RING seats 7 Ly 2
ee) a i : Speed (| LC. wns ek aa rt .
aa zh ‘ il ' ia ie ‘i yk) ay ‘e. Py)
Ress udopiyeeriertiatl

Bayi 4 ree

PLATE XVI.
Meropaster Bowerpanktl, Forbes, sp. (Page 42.)

From the Upper (?) Chalk.
Fia.
la. Abactinal aspect of the example figured by Forbes, natural size. (Coll.
Brit. Mus.)
b. Lateral view of the margin ; natural size.
c. The last three supero-marginal plates ; magnified 2 diameters.
d. Abactinal intermediate plates ; magnified 6 diameters.

Meropaster Parxrnsont, Forbes, sp. (Page 31.)
From the Upper Chalk.

2a. A pair of ultimate supero-marginal plates and the odd terminal or “ ocular”
plate ; magnified 2 diameters. (Coll. Brit. Mus.)
b. Lateral or marginal view of the same ; magnified 2 diameters.

Mirraster rveatus, Forbes, sp. (Page 63.)
From the Upper Chalk.

3a. Abactinal aspect; natural size. (Coll. Brit. Mus.)
b. A supero-marginal plate ; magnified 3 diameters.
4, Supero-marginal plates of another example ; magnified 2 diameters. (Coll.
Brit. Mus.)
5a. Abactinal aspect of the example figured by Forbes; natural size. (Coll.
Brit. Mus.)
b. Lateral view of the margin; natural size.
c. A supero-marginal plate; magnified 3 diameters.
d. Profile or sectional view of the margin, showing a supero-marginal and an
infero-marginal plate; magnified 3 diameters.

AH Searle delet lith Hanhart imp

PLATE XVII.
Metopaster Manrext, Forbes, sp. (Page 38.)
From the Upper Chalk.
Fic.

1. Actinal aspect; natural size. (Coll. Brit. Mus., 40402.)
a. Infero-marginal plate; magnified 4 diameters.

Mirraster compacts, Forbes, sp. (Page 67.)
From the Upper Chalk.
2. Actinal aspect; copied from Forbes in Dixon’s ‘Geology of Sussex,’ pl. xx,
fig. 3.
Compronia Compton, Forbes, sp. (Page 69.)
From the Upper Greensand.
3. Actinal aspect; natural size. (Coll. Brit. Mus., 34511.)

a. Infero-marginal plate; magnified 5 diameters.
b. Lateral view of interbrachial arc; natural size.

CoMPTONIA ELEGANS, Gray. (Page 71.)
From the Upper Greensand.

4. Abactinal view ; natural size. (Coll. Brit. Mus., E. 2567.)
a. Actinal view of same specimen; natural size.

AH Searle del. et. ith Pitcher L‘* imp

GRETACEOUS ASTEROIDEA .

PLATE XVIII.

ARTHRASTER Drxont, Morbes. (Page 91.)

From the Lower Chalk.

Fic.

il (Coll. Brit. Mus., 47000.)

Actinal aspect ; natural size.
a. Supero-marginal plate ; magnified 3 diameters.

Compronia Compront, Morbes, sp. (Page 69.)

From the Upper Greensand.

2. Abactinal view; natural size. (Coll. Northampton Mus.)

a. Isolated plate of interradial portion of disc; magnified 6 diameters
b. Isolated radial; magnified 6 diameters.

c. Lateral view of interbrachial are ; natural size.

d. Supero-marginal plate; magnified 5 diameters.

AH Searle del et.hth Pitcher Lt imp

YYTA VT T
CRETACEOU

a oa _s A - /. we _ eee a ee Al nr oe SS

PLATE XIX.
Penraceros coronatus, Forbes, sp. (Page 82).

From the Lower Chalk.

Fie.
1. Abactinal aspect; natural size. (Coll. Brit. Mus., EH. 2562.)

a. Lateral view of arm; natural size.

Meropaster Parxinsont, Forbes, sp. (Page 31.)
From the Upper Chalk.

a

2. Actinal aspect; natural size. (Coll. Brit. Mus., E. 5027.)

a. Lateral view ; natural size.
b. Ventro-lateral plate ; magnified 5 diameters, showing entrenched pedicellaria.

c. Supero-marginal plate; magnified 3 diameters.

Nympuaster Coomepi, Morbes, sp. (Page 15.)
From the Upper Greensand,

3. Actinal aspect; natural size. (Coll. Brit. Mus., 48620.)

Plxe

a

ponte Fa -

We

'’

pmanope acer oes
oe |

_ we c 4

mp

ere t
Pitcher L*4 im,

earle del et lth

CRETACEOUS ASTEROIDEA .

aot ee —— = — J 2 ee ee a > Se

PLATE XX,
PENTACEROS BULBIFERUS, Forbes, sp. (Page 77.)

From the Upper Chatk.
Fig.
1. Abactinal view; natural size. (Coll. Brit. Mus., 40175.
a. Plate from dorsal part of disc ; magnified 5 diameters.
b. Supero-marginal plate; magnified 5 diameters.
c. Plate from dorsal part of disc ; magnified 2 diameters.
2. <Abactinal view of two specimens; natural size. (Coll. Brit. Mus., 48748.)
a. Madreporite; magnified 6 diameters.
}. Proximal supero-marginal plate ; magnified 4 diameters.

A.H Searle del.et lith .

Pitcher L*4 imp

CRETACEOUS ASTEROIDEA

PLATE XXII.
PENTACEROS BULBIFERUS, Forbes, sp. (Page 77.)

From the Upper Chalk.
Fic.
1. View of the extremities of three arms and portion of disc; natural size.
(Coll. Brit. Mus., E. 5040.)
a. Ventro-lateral plate; magnified 6 diameters.
b. Actinal view of the extremities of two of the above arms; natural size.

PENTAGONASTER ROBUSTUS, n. sp. (Page 73.)
From the Upper Chalk.

2. Abactinal aspect; natural size. (Coll. Brit. Mus., 48085.)

a. Lateral view of an arm; natural size.

PENTACEROS BULBIFERUS, Forbes, sp. (Page 77.)
From the Upper Chalk.

3. Abactinal aspect; natural size. (Coll. Brit. Mus., E. 5041.)
a. Madreporite ; magnified 6 diameters.

4. Abactinal aspect; natural size. (Coll. Brit. Mus., 40399.)
a. Madreporite; magnified 6 diameters.

AH. Searle del et lith Pi
CRETACEOUS ASTEROIDEA .

Leher Loy imp

Fie.

4,

a.
b.
C.

PLATE XXII.
PENTAGONASTER oOBTUSUS, Forbes, sp. (Page 74.)
From the Upper Chalk.

Abactinal aspect ; natural size. (Coll. Brit. Mus., 40400.)

. Madreporite; magnified 10 diameters.

Lateral view of margin, abactinal side upwards ; natural size.

Actinal aspect of two specimens; natural size. (Coll. Brit. Mus., 35481.)
Lateral view of extremity of arm of the underlying specimen; natural size.
Dorsal view of extremity of arm; natural size. (Coll. Brit. Mus., E. 5039.)
End view of extremity of arm; natural size. (Coll. Brit. Mus., E. 5038.)
Side view of extremity of same arm; natural size.

. Ventral end of extremity of same arm ; natural size.
. Supero-marginal plate; magnified 3 diameters.

Infero-marginal plate; magnified 3 diameters.
Adambulacral plate ; magnified 5 diameters.

Pentacrros Boystt, Forbes, sp. (Page 80.)
From the Upper Chalk.

Abactinal view; natural size. (Coll. Brit. Mus., 46600.)
Isolated ossicle of disc; magnified 4 diameters.

Isolated ossicle of disc ; magnified 4 diameters.

Lateral view of arm; natural size.

!4 Searle delet lth. Pitcher Ltd imp

CRETACEOUS ASTEROIDEA

PLATE XXIII.

Penraceros Boysu, Forbes, sp. (Page 80.)

From the Upper Chalk.

Fie.
1. Supero-marginal plate ; magnified 4 diameters. (Coll. Brit. Mus., 48083.)
a. Lateral view of extremity of arm from which the above supero-marginal plate

was drawn; natural size.
b. Dorsal view of extremity of same arm; natural size.

PENTACEROS BULBIFERUS, Forbes, sp. (Page 77.)

From the Upper Chalk.

Abactinal aspect; natural size. (Coll. Brit. Mus., EH. 5042.)

9,
a. Madreporite; magnified 5 diameters.

=.

PENTACEROS BISPINOSUS, n. sp. (Page 87.)

From the Upper Chalk.

3. Actinal aspect; natural size. (Coll. Brit. Mus., 35482.)
Actinal view of extremity of arm ; natural size.

Infero-marginal plate; magnified 3 diameters.
Ventro-lateral plate; magnified 4 diameters.

Pe OGM.

AH Searle del.et Irth Pitcher L*¢ imp

CRETACEOUS ASTEROIDEA

PLATE XXIV.

PENTACEROS ABBREVIATUS, n. Sp. (Page 86.)

From the Upper Chalk.

Fia.
1. Abactinal aspect ; natural size. (Coll. Brit. Mus., 57538.)

a. Lateral view of arm; natural size.

b. Radial plate; magnified 5 diameters.
c. Two succeeding supero-marginal plates; magnified 5 chameters.

PENTACEROS CoRONATUS, Forbes, sp. (Page 82.)

From the Lower Chalk.

2. <Abactinal aspect; natural size. (Coll. Brit. Mus., 35480.)

a. Plate of disc; magnified 3 diameters.
b. Supero-marginal plate ; magnified 3 diameters.
c. Lateral view of arm; natural size.

[See also Pl. XXV, fig. 9.]

PL

¢

NI

AH.Searle delet hth

PLATE XXV.
NyMPHASTER RADIATUS, n. sp. (Page 73.)
From the Lower Chalk.

1. Abactinal view ; natural size. (Coll. Brit. Mus., E. 375.)
a. Lateral view of margin; natural size.
b. Supero-marginal ossicle; magnified 4 diameters.
AsTROPECTEN ? n.sp. (Page 90.)
From the Upper Greensand.
9. Abactinal view; natural size. (Coll. Sedgwick Museum, Cambridge.)
a. Isolated marginal ossicle ; magnified 4 diameters.
Penraceros squamatus, Forbes, sp. (Page 83.)
From the Upper Chalk.
3, Abactinal view; natural size. (Willett Coll., Brighton Mus.)
a. Madreporite; magnified 6 diameters.
b. Marginal view of arm ; natural size.
c. Supero-marginal ossicle; magnified 6 diameters.
Pentaceros ocELLarus, Morbes, sp. (Page 85.)
From the Upper Chalk.

4. Actinal view; natural size. (Coll. Brit. Mus., EH. 5012).
a. Ventro-lateral ossicle; magnified 4 diameters.

PrNTACEROS PISTILLIFERUS, Morbes, sp. (Page 88.)
From the Upper Chalk.
5. Ossicles of disc; natural size. (Coll. Brit. Mus.; from left to right the
register numbers are H. 5037, 57634, H. 2564.)
Genus? sp.? (Page 93.)
From the Chalk.

6. Ossicles; natural size. (Coll. Brit. Mus.)
a. Isolated ossicle ; magnified 4 diameters.

Penraceros P n. sp. (Page 89.)
From the Chalk.

7. Marginal view of ossicles of arm; natural size. (Coll. Brit. Mus., 5514.)

PENTACEROS? n.sp. (Page 89.)
From the Chalk.

8. Marginal view of ossicles; natural size. (Willett Coll., Brighton Mus.)

Prenraceros coronatus, Morbes sp. (Page 82.)
9, Madreporite; magnified 4 diameters. (Coll. Brit. Mus., 35480.)

KXV.

PL

~

A.H. Searle del.et lith

A

4

ROIDE

EK

;OUS AST

TACE

1

CRE

: | a pee eee

PLATE XXVI.
PrENTACEROS PUNCTATUS, n. sp. (Page 88.)
From the Upper Chalk.

1. Abactinal view of remains of arms; natural size, (Coll. Brit. Mus., E. 2561.)
a. Lateral view of margin; natural size.
). Enlarged view of single ossicle ; magnified 2 diameters.
Pentaceros Boystt, Forbes, sp. (Page 80.)

From the Upper Chalk.

2, Actinal view; natural size. (Coll. Sedgwick Mus., Cambridge.)
a. Lateral view of margin; natural size.
b. View of isolated ossicle ; magnified 4 diameters.

Mirraster compactus, Forbes, sp. (Page 67.)

From the Upper Chalk.

Oo

Abactinal view ; natural size. (Willett Coll., Brighton Mus.)
a. View of end of arm; magnified + diameters.
b. Lateral view of supero-marginal ossicles ; magnified 4 diameters.
. Lateral view of infero-marginal ossicles; magnified 4 diameters.

S

Cattierma mosatcum, Forbes, sp. (Page 9.)
From the Lower Chalk.

4. Actinal view of ambulacral groove; magnified 4 diameters. (Coll. Sedgwick
Museum, Cambridge.)
a. Actinal view; natural size.
b. Infero-marginal ossicle ; magnified 4 diameters.

AEE) OOYIL.

A.H.Searle delet lith . Pitcher Ltd imp
CRETACEOUS ASTEROIDEA

ijn

i
a
(
’
Pi
i
=

PLATE XXVII.
LINCKIA sp., n. sp. (Page 100.)
From the Lower Chalk.

Actinal aspect; natural size. (Coll. Brit. Mus., E. 5055.)

. Actinal aspect of portion of arm; magnified 4 diameters.

Orniuka Frrcuu, un. sp., ev Forbes, MS. (Page 103.)
From the Flint Gravel.

Actinal aspect, natural size. (Coll. Norwich Mus.)

a. Cast of actinal aspect in region of mouth; magnified 4 diameters.

. Abactinal aspect; natural size.

OpuIuRA seRRATA, Roemer. (Page 102.)
From the Upper Chalk.

Abactinal aspect; natural size. (Coll. Brit. Mus., E. 5043.)

. Abactinal aspect of segment of disc and two arms; magnified 4 diameters.

Abactinal view of two isolated vertebral ossicles of another specimen; mag-
nified 9 diameters. (Coll. Brit. Mus., HE. 5046.)

. Side view of the same ossicles ; magnified 9 diameters.
. Anterior view of the same ossicles ; magnified 9 diameters.
». Posterior view of the same ossicles ; magnified 9 diameters.

OPHIURA PARVISENTUM, n. sp. (Page 103.)
From the Upper Chalk.

Abactinal aspect of type specimen; natural size. (Coll. Brit. Mus., E. 5052.)

. Abactinal aspect of portion of one arm; magnified 6 diameters.

AH. Searle.deletlith

PLATE XXVIII.
OPHIOTITANOS TENUIS, 2. sp. (Page 104.)

From the Lower Chalk.
Fic.
1. Abactinal aspect of type specimen; natural size. (Coll. Brit. Mus., EH. 5056.)
a. Abactinal aspect of a segment of disc and two arms ; magnified 8 diameters
(shghtly restored).
2. Actinal aspect of another example; natural size. (Coll. Brit. Mus., E. 5057.)
a. Actinal aspect of a segment of dise and two arms; magnified 8 diameters
(shghtly restored).

OpHIOTITANOS LEVIS, n. sp. (Page 105.)
From the Lower Chatk.

3. Abactinal aspect; natural size. (Coll. Brit. Mus., H. 5053.
a. Abactinal aspect of a segment of dise and two arms; magnified 8 diameters
(slightly restored).
4. Abactinal aspect of another small example. (Coll. Brit. Mus., E. 5058.)
a. Abactinal aspect of a segment of disc and one arm; magnified 10 diameters
(shehtly restored).

OPHIOTITANOS MAGNUS, n. sp. (Page 106.)
From the Lower Chalk.

5. Actinal aspect of type specimen ; natural size. (Coll. Brit. Mus., E. 5060.)
v. Actinal aspect of portion of arm; magnified 4 diameters (slightly restored).

AMPHIURA CRETACHA, n. sp. (Page 107.)
From the Lower Chalk.
6. Actinal aspect of type specimen; natural size. (Coll. Brit. Mus., E. 5059.)

a. Actinal aspect of a segment of disc and two arms; magnified 10 diameters
(slightly restored).

AH Searle.del.etlith Pitcher Lt¢ imp.
CRETACEOUS ASTEROIDBEA.

“J

8),

a.

10.

11.

12.

13.

PLATE XXIX.
PYCINASTER CRASSUS, n. Sp. (Page 96.)
From the Upper Chalk.

Actinal aspect of a small specimen; natural size. (Coll. Brit. Mus., E. 2576.)
Abactinal aspect of surface of arm of type specimen; natural size. (Coll. Brit. Mus., 35498.)

. Supero-marginal plate; magnified 3 diameters.

Supero-marginal plate ; natural size. (Coll. Brit. Mus., EH. 2631.)

. Side view of same ossicle ; natural size.

Isolated ventro-lateral plate; magnified 2 diameters. (Coll. Brit. Mus., E. 2632.)

. Isolated plate from abactinal surface of disc; magnified 2 diameters.

Madreporite ; natural size. (Coll. Brit. Mus., E. 2628.)

PYCINASTER SENONENSIS, Valette, sp. (Page 95.)
From the Upper Chalk.

Outer view of abactinal ossicle of disc; natural size. (Dr. Blackmore’s Collection.)

. Side view of same ossicle; natural size.

NYMPHASTER RUGOSUS, n. sp. (Page 94.)
From the Lower Chalk.
Actinal aspect; natural size. (Coll. Brit. Mus., 57516.)

. Infero-marginal plate; magnified 6 diameters.

STAURANDERASTER ARGUS, n. sp. (Page 99.)
From the Upper Chalk.

Actinal aspect; natural size. (Dr. Blackmore’s Collection.)

. One of the ventro-lateral plates; magnified 2 diameters.

Inner view of three ossicles isolated froma specimen in the British Museum ; figured by Forbes ;
magnified 2 diameters. EH. 2566. [See also Plate XXV, fig. 6.]
Outer view of same ossicles; magnified 2 diameters.
ARTHRASTER CRISTATUS, n. sp. (Page 93.)
From the Upper Chalk.

Side view of segment of arm, restored; natural size. (Dr. Blackmore’s Collection.)

1. Outer view of a radiale; magnified 4 diameters.
. Side view of same ossicle; magnified 4 diameters.

Arturaster Dixoni, Forbes, sp. (Page 91.)
From the Upper Chalk.
Ossicle from abactinal surface of disc; natural size. (Coll. Brit. Mus., E. 5024.)

. The same ossicle; magnified 3 diameters.

Muropaster cornutus, Sladen, sp. (Page 117.)
From the Upper Chalk.

Ultimate supero-marginal plate of adult specimen; magnified 2 diameters. (Dr. Rowe's
Collection.)

OPHIOTITANOS MAGNUS, n. sp. (Page 106.)
Fyrom the Lower Chalk.
Abactinal aspect; magnified 2 diameters. (Coll. Sedgwick Mus., Cambridge.)

del et lth

Searl

c
x

Md ee

ot

PAA AAR, pPRAARP EAS. Th TW Pe Ja p ie Ps 7 =_ a oe a. = -~ Wee Ah \ A. MAAR my Aa VY WY Oey — ~

“ ALORS Aen ngnt 2 ARARR: RARipprere AAAAA a ag
2 paper” haphAas mr 2 apar lA, at - OF AAR & = sage Pa “ae Apa? ap- y;
ppppaPrAnah yA AAR 2% Aa “aah PAanaal nw? naar aN TTY ARN Gi A ARAB ABAhe
A NARA ra NA * APA paraln> an Aa wy Ar A a! Aap, . > N KARO ST a parcnh’ sAaaanaee AA ‘aan
ap, aPrpanhnrhet MARY Ey key ite ord nro mR ah sana’ panne At asananen nA aaan ciitvaasstlapaieee-
a - mee’. Pin Png Naar ep a ; - 4 ee
nem OP letial t JAMEL +" aA A APA Rap Pa. yyy acl
] p paal a r Pa An in a mole ar, - Aa a A
PAneh Aa, prPAphh MMR Banat eet pant A A Tate Ama. gaan tl* “nha
VAN. MAA) & we Z i pnt at ANOVA Rare Nan aran tn”

a , nAne lal i PAIN EN Bm ie
oeFGetemered £ tate! ARAN WAVY TL rahe an, ag tag hlt® AAA ARR? VA Onn Vanna
f ar pp pry thr ran sant aan Ana ~~, aPinaaariricy” nar Ae pr:

PPAR APP rRRRA ALT Arte. nr, RY alan pee Bien iksnaas nanan

NARA A» sPAR sar “f AA ry!
. piper . al alala ee : UT AANARMARAAAA AAA

Sabri ale AAA akan nn cA ARAM

aA el leieiey ¥ arn A pe Ana MAPA lala RAANA aamAra ah
ee ee An AAs on APA APARy RAM an AAnABarA
DPA PRPIPAPP Ara apaaann” Si noANn pane Roe ' > a “AAA” " ASP a AAAs aA A mannan AAA RAR aa AAR:
24 Ae haak Rak AR AA Ama paar nA ‘TT: Abtia : ate Naagar MAA
VA Wit haaaanatiere de n Ae ace Ws NAN AP ARARR SER A aanr a“ he Lge
\ =) pipe Bete irk ble es Radetagote, Eben 5 Aj. BOIL WON nt baAReeaae
AT Y aap Ae ae att A AR \n a 2 8 pa
fats: SLL Val wet Wola inlet LL ee paamaneacrancle A
aanpAsAtaaar’ 0? AAA» F a oh any “PP Perak AAAA i aaP
anna, Annnct A mraaphet enna? Ap Pama), caer aatattaat maar naradal gaannnstbAphAl't
aa” A

nak
ybagnnncanneaet nae oz apnmnnltAaqhanAAAanagah®
Aan wae pelt yt at lal Pare aR , arAr ApaPA
AOAARAPPAPRORARRER neti HULU ui rye Hit tae
AANA Una ; PARIIAAA AA, , Cnn iaalhe AprAll paper mance WAN AA AA Ahaa

ane ar

AA ALAAMIAAARAY ANA SAREES os pannnltanan

fl ee ala al ; 3 ' i e \ ~ “a A
Aa) pr’ rt pan. ‘ 4 paarr of cae
Laan R ocdat caller yi _pAahaaan ¥ PLis arnar ;
anaAnaat mPAAAnass prey WAT aaa Ease a rita
=e hy Aaa. ant ai 49 NV 5 ks f

HSetA SE ing “peaae | nec talre tanh ar hae, anathens AMAAAAAa

Ger Rated EL EHEETET Th % ‘ ‘ tT lay ppelio pea eg geet ;
ee. } 1% s aay AA AM Asaaa, 4 MAA, olf

rato. A) nan Wakil and Rae peter e A 4 7

Paap ; RPh 2 ey.
‘ £ “i p - ABAAA Pay oe Pa
1 anes A ~ act + f Ann, al -
pee i ge A \ is r PAL ia Ppt alatste Rr arpa” % Af": AAR «

A pelrl p Aneel eye VaqQAs
Vener pn tb Wa Cte et ae cy ROD s 28f: ph aaahy As PrnPAAa 445
bn 4? Ai AmAN TASS s@asPamere > P rene a err 4 Med .* yaar? > meee ey
Ae pre Of Vda : PrAaaiat renga A ane of a *t amen ae r an® TSR Valli eels
PACS latte pathnnpiPnn aufais nary Pane. or aagG men hash Pe a Pr caaantel f
a Siena ee # lanl rT ma
etna naan? et ARMIN ALPACAAAA: nnanh’ Pan pA 3 A peasceensnepent) : sera
ha, AAPAAC 43 i- ay arerl tha pan’ Aen nnnehs Dann AY 3 Aap -- a aa an Apr A 3 Presenter
< ae eae AAT 8 AAA Aye) tt anne :
- r ~ 22-0 oa | ‘ an A7Aa at an |
iP PA ARAAAAAAD OVAMAGCR?? idlin aaa givangan nati
‘ + A ‘ ae
Cw al tall \— lal -\ omer alalal
to) SRE CE WANN nent ne, TA Ac caaae ag
: y ; BR Be 5 Serta ArT. ies BY aA rq

JARAn-.-

f
2
>.
>»
>2
>
>
>
>>
<
—
>
P
»
>»
>
»
>»
>)
>
>»

An TAn Aa Manan’
~A ek) Aa TNA ar a A ay
ole, Aieoer eee Selena an

2 EDA an ROSSA WAN A ERASE Ray,

pe mp
AR am gp Aal f

AP AAARS Ana. | Lal

La)

Aan!

Amn Aa A Cora

, o | rn

slalel CT

nee Sih nani none ghananenc :
me ad . entisaaicagi pee wAAAaa aA. aunts

nanang ef Oe a Aaenar!
Mea ale PARR RR NSAOP ATOR gaa
SRR Miaaeat thy atte ete oA tA aah Cet eree aman anrana BAe enanans om
SITY PA, Aza AMAi: Bah aT FH aank
OOO Am A R&S a7 ,- sg tees
teranathenstenbnana

MAAS aaete MUAAIARA,
~Aaapar
GGrginee:* Bae Ty, al PMANA AAA, -AaAnarans.
r MAN senna, pore” aa 7 pr
PAAR AAA, .ARAAAP EA.

EN IY er
1 pepe COTCELER CTY HTT
Mp ann nn op - Ae POA nn Ahh APAP a ALtATf AA apr
Ammen ETT RT aaa ttt AA, gents cntrann™ Oe eae aC

Ana gnpnnnnn nnn nna Aa aanP’Aansa man’ * a7 8 Aas anna’ AnaannnnnnQAtns PARARE?

ANON OY a a he ana’ ONE Ae Ae aPPah anman pan RAS nr’
raawa Mannan nnnAnt naar’ Rann pra Rananr™
aAAAaa’ cna eagha hc sannee

ad
AR ARE: AA pene =
‘As AAMAMAAGA LipReee crate neg barra roe aaa "ath
Propel ~AAAre raf . AA AMA yn aanrev ny Ann “pena
A AaAaa a; AA, ar ARs nA, nnn
ita oft tag ttt'™ ay bP aphaapnnrenone ie aan
By AR rye yy ted
f , AA eAnteestecectens
aadielal eee Ty!

a ee
ARAAAARArpRRAAAA AAR as na

WE sas aamansnnbeeh Mn
opp pAAAAAY

fa inf
AR MAA AE FABLE
A eats
saiae AAADAAR PAM AMARA Ann ‘
yi be aL laaieb! V4 oonnnttte NARA AAA Age a i :
Annn 4 ann bad Pa a Sar NEAR AR AR? .49 nase
Mana habits ashes phe ka FP atten pe SITLL) notes na
Bt ~\ SMAARAR AR, ap , op % RAR A AnAA Aap
He EERE : A\EEDR> RARARB Aah an~ae Pla als AUVs) | Aaa agrraas!
one RE searnten aon Pant a TAA ANA a YY Talalala ae pannel PAagnne)
PAA aR mo ror Ons abA 4
AeA at An haa nannnthet ah aAAaa a AAAAAA Bases

~ 8 Anal A an"paar?
Seat ag
aaa aMea sans, crane NAAN WAM Ano aapnent nnn nae am
AA A par anon” AiO at TAY An rere Tae t RasParaii in f
MA ng mn Mean ree ae parternspnre renee
Naa WT mh Riel neat eatin asa anita paar. aapecet

“anne me vn Peppprarl hh Raper’
lala weer A NTE Mee noactacepartec RCRD
PRA acne. 4 tenet oaannt ROME "a my ea 1 Vavaaa

anal PP a pRanPA hans 3 3000) annallinan a oo. ite al “

pet ston Kaaba ream “annem nna en MAAR A
ent aaa r “ BATRA AAO aoa nn Peps
pnAne* ORR PRP aga

Any AIMAAPARa nana it»

An A

Meaty Per Ey Lye As

Ao a espa se
Bad i
_ pl AAAana,n, -~f Fr
a |
Anse v,
> ARR ERIAR ee ae)
‘Pan, _ Ap _

BRA!
|

ty
11

81

ra papers es ui
MOANA
3 9088 00579