Sas
G ar at
Pe ‘ bi
ie MB. Th MEMOIRS: | : .
ae BRITISH MARINE PLANTS: & ANIMALS ; \

Cate,

‘EDITED ay W. Aw ~HERDMAN, D.S¢.s ERS. ‘

‘ * .

%; Oe " ate aM #e % P
Be > eed s Pages

~ m . my ae

x j ‘ all
~ . X, aa cf
F . arf *
ay « ice e.
i {

¢ Sor ‘ e AWRY e aan, a:
ee BY a a “i a, +5 ; aN & 7
; y R. AINSWORTH. D AVIS, MA.
Pins Zoology in the esa Cal @ ye Woes, bien
et +: and) Nagy ee. ae
H. 4. FLEURB, B§ ee gee
Fellow of the eee of Vi sae) ar ‘
i a wa as Kis " 43,
, : by ‘ om Hid
. ns | (With 4 Plates).
2 a

Price Two SHILLINGS AND SIXPENCE

_e LONDON eae i
ph’ ie WititraMs & NOoRGATE ANE a eRe NE 5
=~ ies o- Ria

May, 1903.3 pees ae
id “S ee fan

> ie ei

Ea

4

in

/

IF

Fie

4

EX LIBRIS
Wilham Healey Dall

rE

4

hele

Division of Mollusks
Sectional Library

Sectional Library

eves C. wi ie MORK S.

—_—

X.

Toa iI A:

NOTICE.

Tur Committee desire to intimate that no copies of these

Memoirs will be presented or exchanged, as the prices

have been fixed on such a scale that most of the copies will

have to be sold to meet the cost of production.

The Memoirs may be obtained, post free at the nett

prices stated, from Messrs. Williams and Norgate, 14,
Henrietta Street, Covent Garden, London.

Memoir I.

93

99

EM:

iBUE

Ascidia—published in October, 1899, 60 pp.
and five plates, price 2s.
Cardium—published in December, 1899,
92 pp., six plates and a map, price 2s. 6d.
Echinus—published in” February, 1900,
36 pp. and five plates, price 2s.
Codium—published in April, 1900, 26 pp.
and three plates, price Ls. 6d.
Aleyonium—published in January, 1901,
30 pp. and three plates, price 1s. 6d.
Lepeophtheirus and Lernea

published in
March, 1901, 62 pp. and five plates,
price 2s.

Lineus—published in April, 1901, 40 pp. and
four plates, price 2s.

Pleuronectes—published in December, 1901,
260 pp. and eleven plates, price 7s.

Chondrus—-published in July, 1902, 50. pp.
and seven plates, price 2s. 6d.

Patella—now ready (May, 1903), 84 pp.
and four plates, price 2s. 6d.

Liverpool Marine Biology Committee.

1s... IE VOLES

ON TYPICAL BRITISH MARINE PLANTS & ANIMALS

EDITED: BY W. A. HeERDIMAN, D:Sc., FURS:

X.

BATELEA ,

(THE CoMMoN LIMPET)

bi BY
eB yor
J’ R AINSWORTH DAVIS, M.A.,

Professor of Zoology in the University College of Wales, Aberystwyth ;
AND
Wo *"PLEURE,, BiSes
Fellow of the University of Wales.

(With 4 Plates)

Price Two SHILLINGS AND SIXPENCE

LONDON
WILLIAMS & NORGATE

MAY, 1903

EDITOR’S PREFACE.
‘ne Liverpool Marine Biology Committee was constituted
in 1885, with the object of investigating the Fauna and
Flora of the Irish Sea.

The dredging, trawling, and other collecting expeditions
organised by the Committee have been carried on inter-
mittently since that time, and a considerable amount
of material, both published and unpublished, has been
accumulated. Sixteen Annual Reports of the Committee
and five volumes dealing with the ‘ Fauna and Flora”
have been issued. At an early stage of the investigations
it became evident that a Biological Station or Laboratory
on the sea-shore nearer the usual collecting grounds than
Liverpool would be a material assistance in the work.
Consequently the Committee, in 1887, established the
Puftin Island Biological Station on the North Coast ot
Anglesey, and later on, in 1892, moved to the more
commodious and accessible Station at Port Krin in the
centre of the rich collecting grounds of the south end of
the Isle of Man. A new and larger Biological Station and
Fish Hatchery, on a more convenient site, has now been
erected, and was opened for work last summer, in July,
1902.

In these sixteen years’ experience of a Biological Station
(five years at Puffin Island and eleven at Port Erin),
where College students and young amateurs form a large
proportion of the workers, the want has been frequently
felt of a series of detailed descriptions of the structure
of certain common typical animals and plants, chosen
as representatives of their groups, and dealt with by
specialists. The same want has probably been felt in other
similar institutions and in many College laboratories.

V1.

The objects of the Committee and of the workers at the
Biological Station have hitherto been chiefly faunistic and
speciographic. ‘lhe work must necessarily be so at first
when opening up a new district. Some of the workers
have published papers on morphological points, or on
embryology and observations on life-histories and habits ;
but the majority of the papers in the volumes on the
“Fauna and Flora of Liverpool Bay” have been, as was
intended from the first, occupied with the names and
characteristics and distribution of the many different kinds
of marine plants and animals in our district. And this
faunistic work will still go on. It is far from finished,
and the Committee hope in the future to add greatly to
the records of the Fauna and Flora. But the papers in
the present series are quite distinct from these previous
publications in name, in treatment, and in purpose. ‘They
are called the ‘“L.M.B.C. Memoirs,” each treats of
one type, and they are issued separately as they are
ready, and will be obtainable Memoir by Memoir as they
appear, or later bound up in convenient volumes. It is
hoped that such a series of special studies, written by
those who are thoroughly familiar with the forms of which
they treat, will be found of value by students of Biology
in laboratories and in Marine Stations, and will be
welcomed by many others working privately at Marine
Natural History.

The forms selected are, as far as possible, common
L.M.B.C. (Irish Sea) animals and plants of which no
adequate account already exists in the text-books.
Probably most of the specialists who have taken part in
the L.M.B.C. work in the past will prepare accounts of one
or more representatives of their groups. The following
have performed ov promised their services, and in many

cases the Memoir is either issued or far advanced. The

vii.

first Memoir appeared in October and the second in
December, 1899, the third in February, and the fourth in
April, 1900, the fifth in January, the sixth in March, the
seventh in April, and the eighth in December, 1901, the
ninth in July, 1902, while this tenth one will be ready in
May, 1903, and an eleventh later in the summer;
others will follow, it is hoped, in rapid suecession.
Probably Arenicola, Myxine, Gammarus, and the Oyster
will be ready next.

Memoir I. Ascrp1a, W. A. Herdman, 60 pp., 5 Pls., 2s.
» II. Carpivm, J. Johnstone, 92 pp., 7 Pls., 2s. 6d.
» III. Ecuinus, H. C. Chadwick, 36 pp., 5 Pls., 2s.
» LV. Coprum, R. J. H. Gibson and Helen Auld,
Zowpp.aze Lls.,<liss 6d.
a VY. Axcyontum, 8. J. Hickson, 30 pp., 3 Pls., 1s. 6d.
» VI. Lerrorpntuerrrus anp Lernzza, Andrew Scott,
62: pp., o' Pls:,..28:
» VII. Lingus, R. C. Punnett, 40 pp., 4 Pls., 2s.
» VIII. Pratcr, F. J. Cole and J. Johnstone, 260 pp.,
LE Bis, Ts
» LX, Cuonprus, O. V. Darbishire, 50 pp., 7 Pls.,
2s. 6d.
e A Parncna, J. RR. A. Davis and H: J. Fleure,
84 pp., 4 Pls., 2s. 6d.
ARENIcorA, J. H. Ashworth.
Myxinz, F. J. Cole.
Buccinum, W. B. Randles.
Bucuna, Laura R. Thornely.
Oyster, W. A. Herdman and J. T. Jenkins.
Osrracop (CyTHERE), Andrew Scott.
Portrunuvs, J. Pearson and W. Tattersall.
Cycrtopvorus, F. F. Laidlaw.
Drnpronotvus, J. A. Clubb.

Vill.

Prriprnians, G. Murray and F. G. Whitting.
ZosteRA, R. J. Harvey Gibson.
Himantuatta, C. E. Jones.

Dratoms, F. KE. Weiss.

Fucus, J. B. Farmer.

Antepon, H. C. Chadwick.

Gammarus, M. Cussans.

Borrytioipes, W. A. Herdman.
Currie-Fisu (ELEponE), W. E. Hoyle.
Caxanus, I. C. Thompson.

Actinia, J. A. Clubb.

Hyprorp, K. T. Browne.

CALCAREOUS SPONGE, R. Hanitsch.

In addition to these, other Memoirs will be arranged
for, on suitable types, such as Sagitta (by Mr. Cole), a
Cestode (by Mr. Shipley), an Tsopod, and a Pycnogonid.

As announced in the preface to Ascrpta, a donation
from Mr. F. H. Gossage of Woolton met the expense
of preparing the plates in illustration of the first few
Memoirs, and so enabled the Committee to commence
the publication of the series sooner than would otherwise
have been possible. Other donations received since from
Mr. Gossage, from the Publications Committee of the
Victoria University, and from Mrs. Holt, are regarded by the
Committee as a welcome encouragement, and have been a
ereat help in carrying on the work.

W. A. HerpMAN.

University College, Liverpool,

April, 1903.

|Lete Jkt ie onl

INGOs ox

MEMOIRS.

PATELLA.

BY

Professor J. R. Atnswortu Davis,

AND

H. J. Frevre, B.Se.

CONTENTS.

PAGE PAGE
PREFACE ..... neadqaubops annsatasand 1 Circulatory Organs and Coelom 46
General Description ............ 3 Respiratory Organs............+6 53
External Characters ............ 6 Excretory Organs ........esceeee 56
Body Wall, Muscular System, Reproductive Organs ............ 58
INIEWOUWNy, GUO Goocoodponcsnoddea Development cecwncosceeececceee se 60
Digestive Organs .............e000 20 @onlelusionisterecesseeceece steerer 63
Neryous System, Sense Organs 36 Explanation of the Plates...... 69
PREFACE.

Tue chief objects of this Memoir are : —

(1) To provide a reliable account of the anatomy of

Patella vulgata.

(2) To treat the subject matter so as to show the place
among Gastropods which this type occupies.

The work has been carried out in the Zoological
Laboratory of the University College of Wales; and no

facts are included that we have been unable to personally

B

2

verify, except those of the short résumé we have given
from Patten with regard to the development.

It is unnecessary to refer to the many authors whose
works we have had to consult, but we wish to acknowledge
our special indebtedness to the researches and theories
of Ray Lankester, J. W. Spengel, Pelseneer, Rémy
Perrier, Bela Haller, Lacaze Duthiers, Boutan, Grobben,
and Patten.

The matter which we believe to be new includes the
following chief points :-

(1) A lateral glandular streak has been found along
each side of the foot of young specimens, resembling that
found in Nacella and its allies.

(2) A muscular zone (which we name the internal
pallial muscle) has been found extending in the mantle
between the tips of the shell muscle.

(3) The structure of the Crop, and inferences drawn
therefrom as to special torsion of the viscera of Docoglossa
during consolidation of the visceral hump.

(4) The respiratory function of the nuchal cavity as
regards damp air.

(5) Discussion of the evolution of the present topo-
graphical relations of rectum, kidneys, pericardium, and
heart.

(6) Details of mantle innervation and pallial tentacles.

Our best thanks are due to Professor W. A. Herdman,
D.Se., F.R.S., the Editor of this series, for his general
advice and special suggestions, and to Prof. J. Travis
Jenkins, D.Se., Ph.D., late of Aberystwyth, for working
out several intricate details of mantle innervation as well
as other points. The greater number of the drawings are
original, but a few are based on those given in Lankester’s
article ‘* Mollusca,” in the Kneyclopedia Britannica (Ith
edition), and those of the development are after Patten.

THE COMMON LIMPET (Patella vulgata).
GENERAL DeEscrIPTION.

The Limpet belongs to the asymmetrical class of
Mollusea (Gastropoda), and, though it has lost both
etenidia, is referred to the order Prosobranchia, the
members of which have the ctenidium (or ctenidia)
anterior to the heart. Among the Prosobranchia its place
is in the group of Docoglossa, while the absence of etenidia
and presence of a compensating circlet of pallial gills show
that it belongs to the sub-group Cyclobranchiata.

Limpet shells are so simple and their few characters
are so subject to variation, that it is not easy to classify
the British forms into varieties and species. Of these,
however, two are generally recognised : —

P. vulgata has its shell substance greyish or yellowish,
but never white. The margin is fairly simple, and, though
the radiating ribs vary a good deal, they are not usually
very strongly developed. Aged shells have their vertices
more central than those of young ones.

P. athletica has its shell substance white, and possesses
numerous elevated ribs with regular series of projecting
scales, the interstices being stained with brown. The
shell is usually depressed. Forbes and Hanley state that
Clark found the pallial tentacles shorter and thicker and
the pallial gills longer than in P. vulgata. ‘This species
is fairly abundant but is locally distributed. It is usually
found far down the intertidal zone, and its flesh is said to
be tougher than P. vulgata. Jeffreys considers P.
athletica to be merely a variety, and he enumerates three
other varieties: —P. e/evata—small, round, high; P. picta
——small, thin, alternate ribs reddish and dark blue; P.
intermedia—smaller, flatter, and oval, finer ribs, orange

4

crown, inside of shell golden yellow or flesh-tint, fine ribs
towards edges.

Hasrratr.— Patella inhabits the intertidal zone, affixing
itself to rocks with such force that it is dislodged only
with some difficulty. The method of this adhesion is not
known with absolute certainty. Limpets will hold on
very tenaciously to a surface smaller than the foot, so it
seems improbable that the latter acts like a sucker as has
been suggested. Another view is that the animal is
fixed to the rock by means of a glutinous substance
secreted by its foot, but, from examination of specimens
allowed to fix themselves to plate glass, it seems that this
idea is also ill-founded. The most plausible explanation,
and the one supported by our evidence, is that it is a
case of adhesion of two very closely apposed surfaces, the
foot being, so to speak, rolled out on the rock.

When the young lhmpet finds a suitable spot, it makes a
home of it, returning thither even after wandering some
distance in search of food. This “ homing” faculty, noted
by observers since the days of Aristotle, is one of the most
remarkable endowments of the limpet, and it will be
referred to further on, when the sense organs are described.
One of us has studied this matter by observing animals,
specially marked with enamel paint, which were moved
and watched at suitable intervals.

As a result of continued residence on one spot an oval
depression-is formed in the rock-surface, at any rate on
the softer and smoother rocks. This is called the limpet’s
scar. It has been suggested that the scar is made by the
chemical action of an acid secretion of the foot, but there
is no proof that any such secretion is produced, and well-
marked scars are formed upon siliceous rocks. The
distinctness of the scar appears to vary inyersely as the

hardness of the rocks, and its peripheral portion is

5

deepest. Probably the excavation results from the
mechanical action of the foot surface and of the bevelled
edge of the shell. During life the animal constantly
leaves and returns to its home, and even the comparatively
soft foot must, therefore, bring about an appreciable
amount of wear, especially as fixation is usually preceded
by a certain amount of shuffling or twisting round on the
sear. The deeper margin of the scar clearly indicates
shell action.

An allied form, /Zelcion pellucidum, found at the top of
the Laminarian zone, lives, when adult, in a sheltered
“home” excavated in the bulb or, more rarely, in the
frond of Laminaria, and possesses the same faculty of
returning thereto after excursions. The depressions in
the Laminaria surface are deeply eaten out, and have
tracks leading from them. ‘The shell is thinner than
that of Patella.

Foop anbd ENnemies.—The chief food of limpets consists
of the minute Algve covering the rocks, but they also eat
the larger forms Corallina, Melobesia, Fucus, and Lamin-
aria. With this food they must also take in a number of
minute animals. The details of the feeding process are
discussed later, in the description of the alimentary canal.
They feed, at least partly, while the tide is low, returning
to their scars with its advance to avoid danger from the
tidal wash, but it seems likely that they also feed a good
deal when covered by water; some, indeed, have actually
been watched doing so, and, moreover, the finest limpets
are often found low down on the shore where their
uncovered period is necessarily very limited. Aquarium
specimens also can be kept submerged for weeks,
apparently without ill effect. On the other hand, tiny lim-
pets and feeble sears occur in abundance high up the shore.

The rock barnacles (Balanus) compete with the limpets for

6

space, and in this, perhaps, les the need of the develop-
ment of the homing faculty, since a smooth surface is
required for fixation. Limpets cannot, moreover, success-
fully invade an area already occupied by these
competitors. Fixation is a protective measure against
the wash of the tide and the attacks of enemies, and the
normal adhesion can be largely increased when the shell
is touched.

The limpet, with these habits, does not seem to have
much to fear from active enemies when once the manifold
dangers of youth are past; but it is to some extent eaten
by seabirds, by the common starfish, and even by rats,

while man is, in some districts, a serious enemy.

XTERNAL CHARACTERS.

I. Sueti.—The shell is conical and its base has an
elliptical outline, the anteroposterior diameter being
greater than the transverse. The apex of the cone is in
the sagittal plane and nearer the front end; its exact form
varies greatly, being sometimes, especially in young
limpets, bent over and pointing forwards. The external
markings of the shell vary remarkably within the limits
of the species, but there are always some lines radiating
from the apex and other lines concentric about that
pot. The former are raised ribs running to the shell
margin, but they vary considerably in number and in
prominence; the latter are lines of growth. The edge of
the shell is notched proportionately to the prominence of
the radiating ribs, and its rim is roughly chisel-shaped,
with the bevelled edge inside, and forming an efficient
instrument for enlarging the scar and deepening its
margin. There is considerable variation in the general
shape of limpet shells, those of animals living far up the
shore and on exposed flat surfaces being typically low and

7

broad, while those of animals living nedrer the low tide
limits and in sheltered positions are often high and
narrow, but these rules are not without exceptions.
Plate L., fig. 1, shows extreme types of shell.

The larval shell shows the beginning of a spiral. This
shell grows chiefly by additions to its posterior border
and its mouth is thus lengthened and broadened until it
has a cap-shape with the apex (the remains of the spiral)
far forwards. This apex then breaks off and the hole is
closed by secretion of nacreous material on the inside.
In further growth the anterior border seems to have a
larger share and older shells thus usually have the apex
further back than young ones. This, however, depends
iargely on the precise form of the shell, the broad low
cones above mentioned having the apex further back than
the tall narrow ones. The transition from spiral to
conical shell, with the correlated consolidation of the
visceral hump and the broadening of the foot imto an
oval adhesive sole, are all ‘adaptations to this animal’s
special mode of life. Limpets are peculiarly exposed to
the wash of the tide and of storms, and it is an advantage
to them to be able to adhere firmly to the rock and
present as small an area as possible to the waves. Hence
the thick, plate-like foot and the preparation of a
smooth surface to which it can effectively cling. The
sie of the shell offers a minimum amount of resistance
to waves and tide; at the same time the shell is an
effective protection against enemies, for it can be pulled
down so as to completely cover the animal, while the deep
margin of the scar increases the difficulty of detachment.
The peripheral position of the shell-musele, and its
symmetrical disposition are adaptations to this protective
arrangement.

The spiral- shelled Gastropods protect themselves by

8

retreating far into the shell, the part of the animal
nearest the opening being further protected in many cases
by an operculum. When thus retracted, the animal can
be knocked about without incurring serious risks. This
form of protection is suited to an actively creeping
animal. In a cap-shelled form the power of retraction is
much less and the operculum has gone, so the animal
draws down its cap and remains tightly affixed to the
spot until the danger has passed.

Internally the shell shows markings, of which the most
conspicuous is the horse-shoe shaped impression, with
expanded front ends, of the shell muscle. This impres-
sion is divided into partially distinct areas corresponding
to the component bundles or fasciculi making up this
muscle. On the outer side of the horse-shoe muscle we
see the pallial impression which is continuous across the
front of the shell.

Sections of the shell show its structure to be as follows
(Plate I., fig. 2):—

A.—A thin irregular external layer of brownish colour.

B.—A middle layer made up of very compact crystalline
material the external part of which (a) is penetrated in
every direction by branching minute canals while its
inner part (b) is comparatively clear.

C.—An internal layer made up of numerous smaller
layers extending for varying distances. Each such layer
is, in its turn, made up of parallel but obliquely arranged
lamelle. This portion is thickest under the dome. It
is the nacreous or pearly layer.

Accounts of the shell structure have been given by
Gibson and by Boutan. Gibson noted the minute
branching canals in the external part of the shell and
ventured to suggest that they were due to boring Alge,
which our observations make us think highly probable.

9

Boutan speaks of an intermediate layer between the
middle and internal, but, unless this corresponds with
the inner and clearer portion (B, b) spoken of above, we
have not found it.

II. Sorr Parts or Exrerror._Before removal of the
shell the prominent ventral foot, ill-defined head, and
continuous mantle-skirt can be made out in ventral
(fig. 3) and side view of the body.

(a) The oval muscular Foor has a well-defined wavy
edge which separates a smooth ventral surface from the
smooth sides. Laterally and posteriorly the latter rise
almost vertically to the insertion of the mantle, and
anteriorly become continuous with the head.

(b) The Heap, as seen from the side, is a prominent
muscular projection overhanging the front end of the
foot (fig. 5). In shape it may be compared to a truncated
demi-cone with the cut end turned forwards, rounded off,
and bent somewhat downwards to form a snout, the
circular muscle in the wall of which is very well
developed, and is practically a sphincter. The flat side of
the demi-cone faces downwards, and is marked off by a
distinct transverse groove from the snout. On either side
of the groove is a well-marked retractor muscle band con-
necting the head with the side of the foot. The upwardly
facing curved surface of the demi-cone is continued back-
wards till it passes into and becomes the floor of a cavity
of which the wide opening can be seen above the head.
This is the Nuchal cavity.

The transversely oval mouth situated at the front end
of the head is bounded by a continuous frilled lip, having
a reniform outline resulting from an indentation of its
ventral half. The lip is clearly divided into a long dorso-
lateral portion and a short sharply ineurved ventral

portion. There is a depression at the junction of these two

10

parts on either side. The mouth area faces downwards,
and is placed on the end of the freely-projecting truncated

‘

end of the demi-cone, to which the term “‘ snout” has been
applied. The form of the mouth and lps is evidently
adapted to the process of scraping forwards by which the
animal feeds, and there is a marked contrast with the
vertically elongated seizing mouth of a Haliotis ov a
Pleurotomaria. The snout is devoid of any power of intro-
version, and has no claim to be styled a “ proboscis.”

An elongated cephalic tentacle, swollen at its base, 1s
attached on either side just behind the snout im line with
the front end of the retractor muscle band, which runs
back from the head to the foot. The eye is visible as a
small black pigment spot on the posterior side of the
thickened base of the tentacle.

(c) The Manvve is a continuous flap or skirt, running
completely round the body and ning the marginal part
of the shell. The palhal impression is due to muscle
fibres which take origin not far from the mantle edge and |
are inserted into the shell over a continuous band-lke
area a little above the margin. The mantle is of greatest
extent in front, since here it roofs the nuchal cavity, the
opening of which has already been seen. The edge of
the mantle is pigmented and slightly thickened, and it
bears numerous pallial tentacles each arising from the
bottom of a pit into which it can be retracted. In a
living animal, observed when covered with sea water,
these organs can usually be seen projecting as a delicate
fringe beyond the edge of the shell. The pallial tentacles
are of two kinds, large and small, arranged in a regular
manner, the latter being by far the more numerous.
There may be over three hundred tentacles in a large
specimen, and the number appears to increase with age.

The mantle bounds externally a deep groove, the pallial

11

cavity, the inner side of which, except in front, is formed
by the side of the foot. Above the head a large transverse
opening leads into the nuchal cavity already mentioned.
This may be regarded as a local deepening of the pallial
vavity, originally formed as a “branchial cavity” to
shelter Ctenidia.

Projecting from the inner surface of the mantle are a
large number of plate-like pallial gills, which are flattened
in a direction perpendicular to the curve of the mantle
outline. They form a continuous series right round the
body, except at the left end of the nuchal opening (fig. 5) ;
at this point the pallial vein, which completely encircles
the gills, runs in to enter the heart. These pallial gills
are secondary structures, which have replaced the primary
gills, or ctenidia, just mentioned.

After removal of the shell the following points can be
made out in dorsal view (fig. 4) :—

(a) The shell muscle shaped lke a horse-shoe with
forwardly directed thickened ends. It is divided into a
variable number (12-17) of fasciculi, separated by
transverse fibrous septa.

(>) The mantle skirt running right round the body
external to the shell muscle. In front it extends back-
wards for a short distance between the ends of this
muscle to constitute the roof of the nuchal cavity. Just
external to the shell muscle is seen the band of attach-
ment of the pallial muscle to the shell. Just internal to
the anterior part of this attachment band there is a
very well marked second line of attachment running
between the tips of the shell muscle and indicating the front
boundary of the nuchal cavity. This is the internal pallial —
muscle. The mantle is pigmented dorsally near its edge, and
so also is its band of muscular attachment, but the attach-
ment zone of the internal pallial muscle is not pigmented.

12

(c) The low visceral hump occupies nearly the whole
space within the shell muscle, and is therefore oval in
outline, but the front end is truncated where it adjoins
the back of the nuchal cavity. The hump is covered with
a very thin layer of black pigmented epithehum which
can be brushed off without difficulty; limpets are largely
exposed to the baking action of the sun and this pigment
may have been developed as a _ protective measure.
After its removal the following organs are seen by
transparency (fig. 4) :—

(7) The dorsal surface of the digestive gland occupying
the central region.

(c) A superficial loop of intestine dividing off an
obliquely oval patch in the centre. The two limbs of this
loop run out to the edge of the mass near its right anterior
corner.

(f) More or less of the coiled stomach, external to the-
left Limb of the loop (e) above and concentric with it.

(g) The rectum running transversely a little way in
front of the stomach, and bending forward on the right
to reach the nuchal cavity.

(h) The somewhat triangular pericardium occupying
the left front of the visceral mass. One side of it abuts
against the left end of the shell muscle, a second side runs
obliquely inwards from the shell muscle on the left to
the nuchal cavity on the right, while the third side abuts
upon the nuchal cavity. The second side is approximately
parallel to the rectum which is a little way from it. The
left front corner of the triangle is continued along the
rounded left end of the shell muscle as far as the
attachment of the internal pallial muscle.

(¢) The kidneys, two in number, (1) a small left kidney
between the rectum and the right side of the pericardium ;
(2) a very large right kidney, of which two diverging lobes

13

almost completely encircle the visceral hump except at
the extreme left front. The visible parts of the kidneys
lie very superficially, closely bound to the visceral
integument, and their boundaries can, as a rule, only be
clearly seen in injected specimens, since, in uninjected
ones, the digestive gland often shows through by trans-
parency. As the renal waste is dark coloured, the kidney
is fairly often naturally tinted so as to make its outline clear.

TIT. Nucwat Caviry anp RetateD APERTURES.—The
continuous mantle cavity which lodges the pallial gills
deepens in front to form a fair sized nuchal cavity, so-
called because it hes above what may be called the neck
region of the animal. It opens by a wide slit between the
anterior ends of the shell muscle. Its roof is formed by
a backward extension of the mantle, and its floor by the
muscular body wall of the neck region. With the
exception of the mouth, all the apertures of the body
open into it, ze. the anus, the left renal opening, and
the right renal opening (figs. 4 and 5). When the cavity
is opened by careful removal of its roof the following
structures can be made out :—

(a) The anal papilla opening into the back of the
cavity upon the animal’s right shoulder.

(b) On either side of this a renal papilla, the right
being’ rather more conspicuous than the left.

(c) The pericardium, which bulges into the cavity along
a line from the left tip of the shell muscle nearly to the
left renal papilla.

(d) The pallial vein (also visible from outside before
opening the nuchal cavity) at the left side of the cavity,
running into the pericardium.

(e) The small “ pulmonary veins ’”’ running in from the
roof of the nuchal cavity (visible before opening) and
piercing the pericardial wall to the right of d.

14

(f) The pair of sensory patches or osphradia, which
probably cover the vestiges of the ancestral pair of
ctenidia. Hach structure is a minute orange or reddish
coloured elevation on the floor of the nuchal cavity close
against the shell muscle on either side. It is nearer the
back of the nuchal cavity than the tip of the shell muscle.
These structures prove that the nuchal cavity is really a

branchial cavity of which the etenidia have degenerated.

Bopy Watt, Muscunar System, ann MANTLE.

The body of Patella is covered by a layer of columnar
epithelium which in most parts possesses a well-developed
cuticle. That covering the visceral hump is mostly black-
pigmented as already mentioned; that covering the regions
of muscular attachment is, as might be expected, much
flattened. A great deal of the epithelium is ciliated in
young animals. Unicellular glands are not common in
the epidermis, but a few occur in the foot and mantle;
they are of the usual goblet type. ‘The covering tissues
will be discussed in greater detail in connection with the
various organs.

Mvscunar Sysrem.—Though muscle fibres are found in
almost every part of the animal, there are special aggre-
gations of them which need separate mention. ‘These are
the body-wall muscles of the head and neck, the foot, the
shell muscle, the pallial muscles, and the muscles of the
odontophore. The last of these will be described in
treating of the odontophore.

The muscles of the dorsal body-wall in the head region
comprise the following layers :—

(a) An outer transverse layer, () a layer of longitudinal
and obliquely arranged fibres which are head retractors.
There are also some fibres running transversely internal
to (b).

fal
Ot

The ventral body-wall in the head region is thicker,
but possesses almost solely transverse and oblique fibres.
The muscle of this region is continuous with that of the
foot and shell muscle.

The general appearance of the foot has already been
described. It is essentially a thickening of the ventral
body wall made up, for the most part, of bundles of
muscle fibres bound together by connective tissue. In
the ventral part the fibres run in all directions, while in
the dorsal part they are mostly horizontal and oblique.
The fibres in the outer portion of the foot are steeply
oblique, and these, together with some of the dorsal
fibres, are continuous with the shell muscle. The
absence of a horizontal circular muscle, and the many
differences between the disposition of fibres here and in -
the suckers of various animals, tell against the hypothesis
of a sucker-like action of the foot, which, as was said
above, is also not supported by observation. ‘The spaces
between the bundles of muscle fibres contain blood and,
when full, make the foot swell far beyond its usual size.
Two of the blood spaces above mentioned are of special
importance, they run along in the vicinity of the pedal
nerve-cords, and it is from them that blood spreads to
the remainder of the organ (fig. 87).

In very young specimens (about } in. long) there is a
distinct glandular outgrowth from either side of the foot,
which disappears more or less completely when the animal
gets older. The outgrowth is, from its position and
histological characters, the homologue of the “ lateral
streak” in Nacella, etc. There is also a rudimentary

flap covering the glandular tissue in the anterior
region, where also the whole structure is best developed.
The persistence of this glandular tissue is possibly con-

nected with the protection of the young animal against

16

desiccation, to which it is specially liable in its habitat
far up the tidal zone, and because of the greater ratio of
surface to volume in smaller than in larger forms.

The foot (fig. 8a) is covered externally by a layer of
columnar epithelium, which possesses more or less of a
cuticle, with goblet cells here and there. The cells on the
side of the foot are small, but inerease in height towards
the ventral surface, where they are fairly long and regular,
with nuclei usually near the base. The ventral covering,
as might be expected from the amount of wear to which
it is subjected, seems to be shed frequently. The
epithelium of the sides of the foot is thrown into a
number of small ridges, which, in very young specimens,
are found nearly up to the mantle attachment, but in
older animals are mainly restricted to the part near the
ventral edge. Internal to the epithelium of the sides of
the foot is a deeply staining layer with nuclei, which is
apparently a sub-epithelial nerve plexus, and from it
there proceed outwards narrow deeply staining cells
which are probably sensory. Beneath this covering tissue
is a dermis, an irregular network of connective tissue
with blood spaces which are more numerous internally.
The dermis is thin, at the sides of the foot. Along the
anterior border of the foot there occur several small glands
opening near the edge of this organ.

Internal to the dermis are found the muscle bundles
which make up the main mass of the foot. Most of these
bundles are continuous with those of the shell muscle,
and, in fact, the horse-shoe impression might be spoken
of as the area of attachment of the muscle fibres of the
foot. Of these bundles the outer ones proceed downwards
in a steeply oblique direction, while the inner ones
curve in and become the slightly oblique bundles which
form the dorsal part of the foot. Between these bundles

17

are numerous blood spaces, the most important pair of
which have already been mentioned. It is the quantity
of blood in these spaces that governs the turgidity of the
foot and therefore its functional activity. Besides the
above muscle fibres there are others which run horizontally
and obliquely especially in the ventral part. The muscle
bundles are wrapped in connective tissue.

The lateral glandular streak is a much larger projection
than the puckerings of the side surface of the foot and it
occurs amongst these just below the ventral limit of the
mantle cavity, extending backwards a variable distance
on each side. It consists mainly of tubular gland tissue,
the upper surface of which is covered by small columnar
epithelium as also is the under surface except where the
glands open. There is very little connective tissue
between the glandular tubes. The epidermal and sub-
epidermal tissue just ventral to the projection is to some
extent vacuolated. A variable number of muscle fibres
go out to the projection, and dorsal to it, in the anterior
region, is found a vestige of a flap formed by muscle fibres
and connective tissue covered by epithelium.

The muscles are composed of smooth fibres. These may
be scattered, when they are seen to be each an elongated
cell with lateral nucleus, or aggregated into bundles, as
in the shell muscle. None of the fibres of this muscular
tissue show striation.

The Suett Muscre.—The bundles of oblique fibres
forming the dorsal part of the foot are continued around
its dorsal rim into the shell muscle, so called because its
fibres are attached to the shell (by means of a compressed
epidermis). This muscle encircles the visceral hump
except at the anterior end where it is wanting across the
neck region, it is therefore of a horse-shoe shape and the
anteriorly placed ends of the horse-shoe are enlarged,

C

18

The fibres of this muscle are closely packed and agere-
gated into 12-17 bundles, each surrounded*by connective
tissue. Through this connective tissue run the afferent
blood spaces of the mantle, gills and mantle. skirt.
Iimbryological evidence, derived from Acmea, shows us
that this type of muscle has arisen by enlargement back-
wards of a pair of lateral muscles. Such a paired condition
exists only in /aliotis and Scissure/la among recent Gastro-
pods, and in the former the left member of the pair is
almost vestigial. The columellar muscle of the typical
Prosobranch is the modified right member of the ancestral
pur. The outer more steeply oblique fibres of the shell-
muscle in Patella pull down the edge of the shell by their
contraction, while the medianly running inner ones exert
downward and inward traction which must greatly
strengthen resistance to lateral blows. The latter fibres
are more numerous in Patella than in more primitive
Docoglossa (e.g., Acmea virginea).

The Patirat Muscres attach the mantle skirt to the
shell all round and will be described in the account of the
mantle which follows. Their insertion into the shell has
already been mentioned.

The MantLe Sxrrt-is covered dorsally by a layer of
columnar epithelium which passes gradually into the flat
pigmented epithelium covering the visceral hump. Peri-
pherally it is thrown into a large number of folds (fig. 8)
parallel to the free edge, the cells on the crests of these
folds being very much elongated. Near the extreme edge
of the mantle on the dorsal side the epithelium is not
folded in this way. There is a broad pigmented band
outside and concentric with the shell muscle and a
narrower and less continuous band of the same character
a few cells from the free edge of the mantle, the mantle

edge itself is also pigmented (fig. 8a). The epithehum

19

covering the attachment of the pallial muscle is com-
pressed and pigmented, and that covering the internal
pallial muscle between the tips of the shell muscle is
compressed but not pigmented. The epithelium clothing
the under side of the mantle skirt is also columnar, folded
to a less extent, and not pigmented. The ventral
epithelium of the nuchal roof is shortly columnar. Many
of the epithelial cells of both the dorsal and the ventral
epithelium, are clear and glandular, and near the edge of
the mantle skirt a great deal of sub-epithelial glandular
tissue is present, one set of glands opening at the extreme
edge of the mantle and the others further ventrally.
These glands secrete the material for the outer shell
layers. Many sense cells are present in the epithelium
of the mantle and of its tentacles, which are best described
ina separate paragraph. These are particularly abundant
in the epithelium of the dorsal side just near the free edge
which, as has been stated, is not folded.

The substance of the mantle, including the roof of the
nuchal cavity, is made up of a basis of connective tissue
containing numerous blood spaces and traversed by
muscles and nerves The blood spaces in the roof of the
nuchal cavity are specially large and are traversed by
distinct trabecule of connective tissue. This fact, com-
bined with the comparative thinness of the epithelium,
is in itself adequate evidence of the respiratory function
of the nuchal roof. The other more regular blood spaces
are best considered along with the remainder of the
circulatory system and with the pallial gills.

The chief aggregates of muscle fibres are those forming
the pallial muscle and the transverse band which we have
‘alled the internal pallial muscle.

The fibres of the pallial muscle take origin in the
pallial impression of the shell and then travel outwards,

20

radiating into the mantle edge which they serve to retract
and into the sub-epithelial tissue of which they are
inserted. They also furnish the retractor muscle slips
which run into the pallial tentacles. The fibres of the
internal pallial muscle take origin in the corresponding
transverse impression, and form a definite layer of
radiating fibres, which are also inserted into the mantle
skirt nearer the edge than the level of origin of the other
pallial muscles. A continuous ring-nerve runs round in
the tissues of the mantle skirt. It is formed in front by
the anastomosis of the anterior branches of the palhal
nerves, and behind by the fusion of the posterior branches
of the same nerves (fig. 24). From the outer side of this
ring-nerve numerous twigs are given off, which break up
into a network from which branches pass into the pallial
tentacles.

DiGEstTIvE OrGans (figs. 4, 6, 10, 11-22).

These consist of a very long and complex Gut and
some important Glands which open into it. The mouth
opens into a buccal mass (containing the Odontophore),
which is succeeded by gullet, crop, stomach and long
intestine, the last part of which is the rectum. There
are buccal (‘salivary’) glands opening into the bucval
mass and a large digestive gland or hepato-pancreas
(“liver”) opening into the stomach.

I. Tur Gur.—This complicated tube is formed of many
coils arranged in a characteristic manner. We now
proceed to the detailed description of its various parts.

The Buccal Mass (fig. 6)—This is best exposed by
lateral incisions of the anterior part of the body wall.
Connect these by a transverse incision near the tentacle
bases and carefully peel off the dorsal body wall from
before backwards, noting that the gullet is very liable to

21

be damaged in the process on account of its intimate
connection, by means of fibres, with the body wall.

The buccal mass, as seen in dorsal view (fig. 12), is
roughly rectangular in outline with the broad sides
lateral. The anterior quarter of the rectangle contains
the buccal cavity and the remainder constitutes the
broadened anterior end of the radular sac, the rest of
which will be seen later. The gullet is a thin-walled
tube which opens at the junction of these two regions,
and then runs backwards covering and concealing from
view the radular sac. The four buccal ducts (fig. 16) are
seen running forwards with a somewhat wavy course, the
two outer lie freely at the sides of the gullet, the inner
run above the dorsal wall of the gullet to which they are
closely attached. All four ducts open at about the level
of the beginning of the gullet by piercing two thickened
dorso-lateral areas, which will be discussed later on in con-
nection with the glands. The gut cavity is wider at this
level than further back, and these thickenings have been
homologised with the dorgo-lateral buccal pouches of
Haliotis.

The Buccal Cavity is best studied by comparing the
results of median dorsal incision, lateral incision,
longitudinal median section, and transverse section im a
few specimens (figs. 6 and 12). The following points are
to be noted :—

(a) The situation of the cavity.—As the mouth 1s
ventral, the morphologically front and back ends of the
cavity are topographically ventral and dorsal, while the
morphological roof and floor are respectively anterior
and posterior. In describing the cavity we shall refer to
the parts in their morphological relations. The main or
palatal buccal cavity has a posterior extension, which we
shall call the post-palatal buccal cavity, situated on the

29

dorsal side of the odontophore under the commencement
of the gullet (fig. 12). This cavity narrows behind into
the radular sac.

(6) Within the dorsal and ventral outer lps an
inner lip is present on either side as an upwardly growing
fold (fig. 6). Boutan refers to this pair of lips as “ une
levre en forme de fer a@ cheval.” These two lips can be
approximated so as to close the mouth almost entirely.
A median papilla, or “licker,’ belonging to the
odontophore, projects between the lips so as to com-
pletely fill the chink between them when the mouth
is closed.

(c) The dorsal palatal plate is a chitinous sub-epithelial
thickening forming an arch over the region of the inner
lips. It imparts firmness to the lp and also serves to
htt the tissues connected with it out of reach of the
rasping radula. It is prolonged into two pairs of
expansions, one pair dorso-lateral and one pair ventro-
lateral (fig. 11). This plate is most probably a pair of
jaws united by a median dorsal piece.

(d) The front end of the odontophore or rasping organ
projects into the palatal section of the buccal cavity from
its postero-ventral wall, and the front end of the radula,
a horny tooth-studded ribbon, occupies a narrow median
strip on this projection (fig. 12), extending backwards on
the posterior part of the odontophore cushion, which is
placed beneath the floor of the post-palatal section of the
buceal cavity (a@ above) underlying the gullet. The
epithelium covering the tip of the odontophore, anterior
to the radula, has grown out into a transversely ribbed
projection, which is the papilla, or “ licker.”

(c) The sub-lingual pouch is a recess below the front
end of the odontophore (fig. 6). It is found in Chiton
and many other forms. In Patella it is lined by

23

yellowish epithelium and contains mucus-secreting cells
in its roof.

(f) The floor of the buccal cavity is muscular with a
slight sub-epithelial median development of chitin.

(g) The odontophore and its muscles are described later on.

The Gullet (figs. 6, 10, 15, 16, 22) runs back as a thin-
walled tube from its point of junction with the buccal
mass, overlying the central part of the posterior two-
thirds of this. Immediately behind the buccal mass it
shifts slightly to the left of the median line and merges
into the thick-walled crop, dilating into two ventro-
lateral cesophageal pouches just before the point of
junction (fig. 10). The esophageal pouches are in part
spare tissue for the purpose of avoiding strain when the
buceal mass is extruded for feeding. They are probably
homologous with the lateral cesophageal pouches of
Pleurotomaria and /Talvotis, though their functions cannot
be quite the same, as in these two forms they are lined by
ridged glandular epithelium, while in Patella they are
comparatively smooth, and not markedly glandular. The
gullet has various folds (figs. 15 and 22), as follows :—

(a) A transverse section taken near the posterior end
of the buccal mass, shows the gullet as a dorso-ventrally
flattened structure, the central portion of which is marked
off from the sides by two long folds projecting downwards
from the dorsal wall.

(6) There are a pair of longitudinai ventral folds
starting slightly behind the junction of the gullet and
the post-palatal section of the buccal cavity. These folds
converge as they run back, and the triangular space
between their front ends is marked by shght transverse
ribs. It seems to correspond with the much _ better
developed ‘‘ ventral valve” of this region in most primi-
tive Prosobranchs.

24

(c) A number of small oblique folds lie external to each
ventral fold.

The great width of a posterior section through the
gullet is due to its ventro-lateral extension into the
cesophageal pouches. These occupy part of the space at
the back of the buccal mass and are not sharply marked
oft, being, as it were, mere bulgings of the gullet. The
lumen of each pouch is continued forwards into that part
of the gullet cavity lateral to the dorsal fold of its side.
The gullet is, to some extent, asymmetrical on account
of the shift to the left where it merges into the crop.

The Crop (figs. 10 and 18)—On passing into the
visceral mass the food tube shifts further to the left of
the median line and its walls become very much
thickened. This thick-walled region is called the crop.
It is imbedded in the visceral mass and care is needed to
expose without damaging it. The buccal and digestive
glands adhere to its walls, and various gut coils also hide
it from view. In this imbedded position it runs back
some distance, becomes much narrower, and then passes
to the right and runs forward,

The lining of the crop projects internally to form the
following folds :—

(a) The continuations of the longitudinal dorsal folds
of the gullet. These folds pass down the left side of the
gut-wall and become mid-ventral, thus indicating that
the crop has been subjected to a torsion of 180° relative
to the head. Further back, these folds run up the right
side showing an additional torsion of 90° or more
(fig. 22).

(b) The continuations of the ventral longitudinal folds
of the gullet. These have fused basally, so we find a
single outgrowth branching distally into two. This keeps
at a uniform space of 180° from the dorsal folds and so

25

is twisted round through the same angular distance
(fig. 22).

(c) Obliquely transverse folds running across the wall
space from the single divided fold (4) almost to the pair
of folds (a) (fig. 18).

Where the crop narrows down posteriorly the transverse
folds of its walls cease, but the longitudinal ones are
continued until the gut begins to run forward on the
right side. At a certain level all folding ceases, and
this point should probably be taken as the termination of
the crop.

The Stomach (figs. 4, 6, 10, 19, 20)-—Beyond the crop,
the now thin-walled gut runs forward for a_ short
distance and then bends over on itself and becomes a
large coil which encircles the dorsal surface of the
digestive gland. The whole of this region may be ralled
stomach. At the point where the stomach bends over on
itself the duct of the digestive gland opens into it on the
left (fig. 19). This opening is fairly conspicuous, and
the wall of the duct has a groove with thickened pro-
jecting sides (figs. 19 and 20). This groove Is continued
along the inner wall of the stomach throughout the whole
of the superficial coil, and becomes indistinct where the
stomach narrows down and passes into the intestine at
the right side of the visceral mass. At the point of
bending, near the opening of the duct, there is an inward
flap-like projection of the posterior wall (fig. 19).

Intestine—Numerous coils of intestine (figs. 4, 6, 10,
21) form the remainder of the gut.

(a) A coil following the stomach, running back along
the right side, then round just ventral to the crop,
returning finally to the right anterior corner of the
visceral mass. This coil has a relatively wide cavity (fig.

10, Int. 1).

26

(6) A coil on the dorsal surface of the visceral mass
within the stomach coil (fig. 10, Znf. 2). The anterior
limb of this coil communicates with the end of a, and the
posterior limb with the posterior limb of ¢.

(c) A small coil ventral to (a) (fig. 10, Znét. 3). Its
anterior limb passes, at the right anterior corner of the
visceral mass, into

(d) a long coil (fig. 10, Zné. 4) which first passes, over
the right anterior corner of the visceral mass and then
proceeds beneath the pericardium and around the
external border of the stomach to about the middle of
the right side, when it turns back and forms

(e) the last coil (fig. 10, Znt. 5), which is in close
contact with (7) until it reaches the pericardium, when it
proceeds along the posterior edge of this to the animal’s
vight shoulder and opens on the anal papilla. The last
part of this coil (e) is the rectum. Its ventro-lateral wall
has two inwardly projecting ridges with a gutter between
them. The anus opens at the tip of the anal papilla
which projects into the nuchal cavity on the right
shoulder (fig. 4). The walls of this projection are much
thickened, and its lining is papillated. The ventro-
lateral ridges of the rectal wall become more strongly
marked as the walls thicken towards the anus (fig. 21).
Judging from the condition in other Gastropods it would
seem probable that the rectum once ran forward in or
near the median plane traversing the pericardium. The
shifting of the anus to the right side, and of the peri-
cardium to the left, has made the rectum lie across the
body, and has pulled out the whole of the last coil of the
gut so that it now lies on the extreme outside of the
hump around the left side of the latter.

Il. THe Larcr Guanps or tue Gur.—The Buccal
Glands (figs. 6, 15, 16) are four compound orange-

bo

i

coloured tubular glands occupying the front of the
visceral hump below the pericardium and rectum. ‘They
are usually united practically into one mass, but
occasional specimens show, anteriorly, more or less
division into four (fig. 16). The two inner ducts run in
the grooves formed externally by those inward projections
into the gullet cavity, called above the dorsal longitudinal
folds. . Free at first, they become, towards the front end,
involved in the gullet wall (fig. 15), but they run along
in it and only open at about the same level as the outer
pair. The two outer ducts run freely at the sides of the
gullet, and open into slight lateral pouches in the buccal
cavity behind the palate; they often have ampulle placed
at irregular intervals along them (fig. 16). Cuvier noticed
small patches of yellowish tissue near the openings of the

‘

bueeal ducts, and named them “ salivary glands.” It is
interesting to note that similar yellowish glandular masses
occur in the pharyngeal wall of /vssurella near the open-
ing of its (one pair of) buccal ducts; these Boutan has
unhesitatingly called traces of another pair of buccal
glands.

The Digestive gland, or Hepato-pancreas (‘‘ liver ”’),
is a large racemose gland, occupying the centre of
the visceral mass (figs. 4 and 6). The bilobed condition,
which is primitive for Mollusca, may have disappeared
long before the differentiation of the Docoglossa, or may
be obscured by the consolidation and compression which
the viscera have undergone in this group. The ducts of
this gland converge into two main ducts, which unite
just before they open into the stomach, as described
above. The ridge-bounded groove on the internal wall
of the stomach is continued on the floor of the main duct
and even of its first branches.

Eves tology ot sthhe (Gat) anda G lamas:

The digestive tube is lined throughout by columnar
epithelium, many of the cells being ciliated, others
glandular, and a cuticle being developed in some parts.
Beneath the epithelium the wall consists of connective
tissue and elongated muscle fibres. The size of the lumen
of the gut varies considerably in different parts, as also
does the thickness of the walls. The cavity of the gullet |
is extensive (fig. 15), while its walls are very delicate
and closely bound dorsally, by strands of connective
tissue, to the outer body wall covering the head. The
crop has thick walls and a correspondingly reduced
cavity (fig. 22), while the stomach has a very wide cavity
with rather thin walls; the succeeding coil of the gut is
narrower and thin-walled, and the remaining coils are
still narrower; the rectum is thicker walled (fig. 21).

The sub-lngual pouch, like the rest of the gut, is lined
by columnar epithelium, which here has a_ distinct
yellow colour, is largely glandular, and probably mucus-
secreting. The region of the buccal ‘avity, into which
the buccal glands open, has its walls yellow tinged and
somewhat thickened, the thickening being due to the

presence of the small tubular glands previously

mentioned. Their cells resemble those of the buccal
glands. The crop has a greater development of sub-

epithelial tissue than is possessed by other parts of the
gut, but the major part of the thickening is due to the
epithelial folds projecting into-its cavity.

The longitudinal folds are covered by high columnar
epithelium, which includes a large number of glandular
cells. Many of the cells contain rounded granular
highly refractive bodies. Under the epithelium of each
fold is a distinct, mainly fibrous, band which can be

29

traced outwards through the thickness of the crop wall.
The transverse folds are infoldings of the side walls
between the longitudinal folds just mentioned. They
connect with the single divided longitudinal fold, but
not with the pair of longitudinal folds. Each of these
folds has secondary foldings on itself, and these run
approximately along lines radiating from the centre of
the crop. Near the inner and free edge ot the fold are
smaller papilla-like outgrowths (fig. 18).

The stomach is lined by columnar epithelium, with
subjacent connective tissue which is not very abundant.
The epithelium is fairly even all round except along a
line on the internal wall where we find a groove formed
by the upgrowth of two ridges. The height of these
ridges is due mainly to the unusual height of the
epithelium along them, while the cells lining the groove
are very short (fig. 20). This description is correct
for a young lmpet, but as the animal grows older the
groove deepens.

Much of the intestine has ridged walls. The rectum
shows two ventro-lateral folds, which are covered by
fairly high columnar epithelium, growing into its cavity
and forming a gutter between them. These folds become
much more distinct on the anal papilla (fig. 21), and here
they branch into secondary foldings. Feces seem to be
extruded only from the upper section of the rectum,
dorsal to the “ gutter” and folds.

The buceal glands are made up of much branched tubes
which are imbedded in connective tissue, and the glands,
for this reason, have the appearance of one large mass.
The cells lining the gland tubules are large and contain
numerous granules, they are fairly equal in size and have
large and distinct basal nuclei. The cells near the apices

of the tubules appear not to be ciliated, but those further

BO

towards the ducts possess cilia and have smaller
nuclei.

The digestive gland is similarly a great mass of
branching tubules imbedded in connective tissue. The
cells lining these tubules vary a good deal in size and
shape but are very distinctly glandular, and usually so
gorged with droplets of a yellowish-green secretion and
fine granules that they burst while being prepared for
microscopic examination (fig. 17). The ducts of the
digestive gland are lined by ciliated columnar epithelium
resembling that of the general gut lining. Miss
Newbigin finds that, in sections of specimens hardened
in formalin, the epithelium lining the intestine has a
band of brownish green pigment near the inner margin
of the cells. When examined under a higher power the
pigment is seen to occur in minute closely packed
granules, brownish green in mass, green when viewed
singly. The cells of the digestive gland vary in size,
the large cells near their inner surface contain several
of the characteristically molluscan pigmented vesicles,
usually of a brownish yellow colour, while scattered
through the protoplasm occur numerous oil drops. The
presence of varying amounts of a pigment, called
enterochlorophyll, causes great variations in the colour
of the digestive gland. The same pigment also occurs in
the feces.

Ill. Tur OponrorHore (figs. 6, 12, 13, 14)—The
Odontophore, or rasping organ, characteristic of all
cephalous Molluses, arises in development as a ventral
pouch of the fore gut, the ectoderm of which, together
with the underlying mesoderm, ultimately gives rise
to :—

(a) A projection on the floor of the buccal cavity—the
Cushion (Mesodermic).

dl

(b) A pouch—the Radular Sac, formed of ectodermic
epithelium, with a little subjacent connective tissue.

(c) A horny tooth-studded ribbon, the Radula, of which
the part in use rests on the surface of the cushion, while
the parts in reserve and in course of formation are lodged
in the radular sae. :

The Cushion is the projection on which the radula rests,
and whose movements enable that organ to perform its
scraping function. It consists essentially of muscles and
four pairs of cartilages, and is covered above by the fore
eut epithelium and that of its diverticulum, the radular
sac. The radula is placed in a median groove on the
dorsal surface between the paired cartilages (fig. 12).

The cartilages are best named Anterior, Antero-lateral,
Ventro-lateral and Posterior. Their shapes and relations
are most suitably explained by diagrams (fig. 15). Hach
anterior cartilage has a ridge towards its front end, acting
as a lateral pad for the radula. The cartilage is of the
type called spongy, as it consists of fairly large, clear,
nucleated cells, separated by a comparatively small amount
of structureless intercellular substance. ‘The antero-
lateral are placed at the sides of the anterior cartilages
near their front ends, and give the cushion its charac-
teristic broad front. The ventro-lateral cartilages, first
noticed by Amaudrut, are very small, and on the ventral
surface of the cushion. The posterior cartilages are well
developed, and not so ventral as in some Rhipidoglossa.

The muscles of the Odontophore include (1) Extrinsic
Muscles, connecting the odontophore with the walls of the
buccal chamber, (2) Intrinsic Muscles, connecting the
cartilages with one another and with the sub-radular
membrane.

The chief Extrinsic Muscles are :—

(a) The large ventral protractor muscles attached

32

posteriorly to the ventral part of the posterior cartilages
and spreading out towards their anterior origin behind
and at the base of the inner lips.

(b) Dorso-lateral protractor muscles attached at the
front end around the expansions of the dorsal palatal
plate, and, behind, to the sides of the posterior cartilage.

(¢) Antero-lateral retractor muscles going from the
sides of the anterior cartilages to the floor of the neck
cavity. These may be divaricators of the cartilages.

(d) Numerous fibres, more or less irregularly arranged,
and including some posterior ventral retractors going
from the odontophore to the floor of the neck cavity.
The protractor muscles seem to be by far the most
important, and probably, therefore, the act of retraction
is mainly a natural relapse to normal conditions from a
state of strain.

The chief Intrinsic Muscles (fig. 156) are : —

(a) Two transverse layer-like bands connecting the
ventral sides of the anterior and lateral cartilages together.
These are separated by (6).

(b) A pair of muscle bands going forwards from the
posterior cartilages to the epithelium beneath the front
end of the radula.

(c) Muscle bands connecting the anterior and antero-
lateral cartilages dorsally, and

(d) Muscle bands connecting these cartilages ventrally.

Besides the above muscles, various bands of connective
tissue bind the antero-lateral and the posterior cartilages
to the anterior, and some further minute details are shown
in the figures (figs. la, 4, ©).

The Radular Sac (figs. 6, 12) is a long eylindroidal
diverticulum of the post-palatal section of the buccal
cavity. Arising at the upper limit of the back of the

cushion, it first runs downwards close to this, and then,

Do

after a swerve to the left, bends back along about two-
thirds of the under surface of the visceral mass in the
median plane. It then curves well round to the right
side, taking a forward course to about the middle of the
visceral mass, after which it bends sharply upon itself
and retraverses the greater part of its former course.
This second part is closely applied to the right side of
the first, and its caecal tip extends forwards almost as
far as the back of the cushion. The anterior part, at
least, of the radular sac and the odontophore, are sur-
rounded by a large blood space, whence blood travels
both to the visceral hump and to the foot.

The Radula is altogether about twice the length of the
animal, and is a narrow belt-like structure with rows
of teeth, having at its anterior end a flat plate-like ex-
pansion on either side. This plate is bent over the front
end of the cushion where the covering epithelium has
secreted it. Behind this the radula sinks into a groove along
the middle of the odontophore and runs back into its sheath.

The teeth of the radula are arranged in convexly curved
transverse rows, the convexity being forwards on the dorsal
surface of the cushion. There are twelve members to
each row (fig. 14). Of these twelve the four central ones
are similar, although the middle two are slightly smaller
than the others. Hach consists of a yellow stalk and a
black-tipped brown “ claw,” the claw having its concavity
directed backwards. Next to these four and behind their
level is another tooth of the same kind but larger and
with three claws. Lateral to these again we have three
pairs of teeth without claws, these have their ends curved
slightly backwards. They are at the same level as the
four central teeth.

The absence of one definitely differentiated median
tooth characterises all the Docoglossa except the Lepetide,

D

Bet

and other special features are the fewness of teeth in each
row, and the high degree of specialisation of the
individual tooth. These characters sharply differentiate
the Docoglossa from other Gastropods, for the specialisa-
tion of their radula is certainly along a line diverging
absolutely from that adopted by the Tzenioglossa, where
the fewness of teeth per row is due to several having
fused to form compound ones.

To this account of the alimentary canal we may add
a short description of the relations of the various parts
of the fore-gut to the feeding process. The generalised
diagram of a median longitudinal section of Patella, elven
in fig. 6, is intended to illustrate these relations.

By contraction of its strong ventral protractor muscles
the tip of the odontophore is extruded from between the
lips, and the radula is rubbed along the rock surface from
behind torwards to scrape off minute Algw. The outer
lips, aided by the “ licker,” seize and hold any fragments
torn from the rock, the dorsal palatal plate greatly
strengthening the dorsal outer lip for this purpose, besides
lifting the roof of the buecal cavity out of reach of the
rasping radula. The food, consisting of small Algee and
tiny organisms of various kinds, with an admixture of
rock substance, is then passed into the buccal cavity from
which its exit is barred by the closure of the inner lips.
The mouth parts are almost always examined when both
they and the head are in a retracted condition, and it
may be that the cesophageal pouches are, in part, spare
folds of tissue allowing the protrusion of the odontophore
without breakage of any of the gut lining. — Specimens
paralysed by a dilute solution of chloral hydrate im sea
water often die with the head partly expanded, and they
certainly seem to show less folding in the region of the

cesophageal pouches.

35

Patella also feeds in another way by gripping a piece
of seaweed with its outer lips (aided by the palate) and
then scraping off fragments from it by the rasping action
of the radula. The mechanism for retracting the
odontophore is not as manifest as that for its extrusion,
and this process is probably in part a return to normal
conditions from a strained state of expansion. The
slightly developed retractor fibres, the cessation of con-
traction of the protractors, and the antero-lateral muscles
probably all help, and assistance must also be given by the
contraction of the muscular snout.

While the tip of the radula is scraping food from the
rock, another part of it is, by the same motion, working
against the jaw, or palatal plate, and thus helping to grind
the food already obtained. This working of the radula
against the jaws continues after actual feeding has ceased,
and the cesophageal pouches seem to retain the food frag-
ments temporarily till this function has been completed,
and till they can be passed on into the crop. This passing
on is necessarily slow on account of the many folds project-
ing into the cavity of the crop. Meanwhile, also, the food
becomes mixed with the secretion of the buccal glands.

It is noteworthy that the esophageal pouches of Patella
do not show the development of folds and papille which
characterises presumably homologous structures in the
primitive Rhipidoglossa, and another feature is that the
dorsal and ventral valves, which in the latter, and even in
Acmea (dorsal valve), prevent passage of the food from
the pouches back into the buccal cavity, are very much
reduced or absent. The pouches are therefore probably,
in part, spare tissue and, in part, temporary stores, and
not highly specialised secretory regions, as in Halzotis, and
other Rhipidoglossa. The corresponding secretory func-
tion is performed by the crop, which is extremely

.

36

specialised among the Cyclobranch Docoglossa, and thereby
gives evidence of the difficulty of digesting the tough
vegetable food; the high development of folds must also
enable this region of the gut to act as a strainer, preventing
the further transport of large fragments into the thin
walled regions which follow. The great length and size
of the stomach, which is the region where the secretion of
the great digestive gland becomes mixed with the food
is a further testimony to the slowness and ditticulty of the
digestive process.

NERVOUS SYSTEM AND SENSE ORGANS.

The nervous system of Patella vulgata (fig. 25) may be
considered under three headings.

(a) The Cireumeesophageal ring with its cerebral,
pleural and pedal ganglia, and the principal nerves thence
given off except—

(b) The Bueeal and Labial nervous systems, connected
to the ring at the cerebral ganglia.

(c) The Visceral Loop and nerves thence given off,
connected to the ring at the pleural ganglia.

The Circumesophagea! ting.—-This 1s
seen, after exposure of the dorsal surface of the gullet,
on pressing apart the gut tissues and body wall in the
region of the head.and neck. When completely exposed
it is observed to be roughly four-sided, the plane of the
quadrilateral sloping downwards and backwards, while
its upper and lower sides are curved outwards. The
quadrangular form of the ring is due to its having accom-
modated itself to the outlines of the buccal mass. Though
certain swellings on the ring are referred to as ganglia,
it must not be supposed that nerve cells are not.found in
other parts, the concentration of these being by no means

complete. The two upper corners of the quadrilateral

a7

ave set near the tentacle bases and are swollen to form
the well-marked cerebral ganglia. These are connected
by the cerebral commissure (forming the upper side of
the quadrilateral) which runs across very far forward
just beneath the dorsal outer lip.

The right and left sides of the quadrilateral (continuous
above with the respective cerebral ganglia) have become
double by separation of the connectives going to the
pleural from those going to the pedal ganglia.

The lower side of the quadrilateral is a very short pedal
commissure thickening at either end into the commence-
ment of the great pedal nerve cord, which is ganghonic
for a considerable portion of its length. The outer side of
this anterior end of the pedal nerve cord is continuous with
a short stout pleuro-pedal connective, and this latter
thickens at the side and a little to the front into a pleural
ganglhon. From the ganglionic centres a munber of
nerves are given off, as follows (see fig. 25):

(a) From each cerebral ganghon:

(1) The cerebral commissure, the cerebro-labial con-
nective, and the cerebro-pedal and cerebro-pleural
connectives.

(2) The branching tentacular nerve,

(3) The fine optic nerve supplying the eye which is
placed on the posterior side of the base of the tentacle.

(4) Fine nerves (5 in number) going to the snout.

(5) A nerve which, for some distance, runs along with
the cerebro-pedal connective, but which diverges from this
at about the level of the pleural ganglia, going thence to
the dorsal body wall.

(6) A nerve which is at first fused with the cerebro-
pleural connective but soon becomes distinct from it and
supplies the otocyst which lies at the base of the pleural

ganglion against the pleuro-pedal connective.

38

(6) From each pedal cord, the anterior end of which
may be called an ill-defined pedal ganglion :—

(1) From the anterior end, a nerve going forwards to
the foot and ventral body wall.

(2) Nerves to the dorsal layers of the foot and to the
shell muscle.

(5) Stout nerves to the ventral part (sole) of the foot.

(4) Fine nerves, those from one cord going towards
those from the other. These nerves from the two cords
unite in two cases forming anastomoses, of which one
is close behind the pedal commissure, and the other
quite posterior. It is probable that these nerves and
anastomoses are vestiges of an ancestral ladder-like
condition of the pedal cords and their interconnections.

(c) From each pleural ganglion :—

(1) Two small nerves (one more strictly from the
cerebro-pleural connective and one from the ganglon
itself) arising close together and supplying the dorsal
body wall.

(2) A distinct nerve which seems to supply the anterior
end of the shell muscle.

(3) A small nerve arising near the lateral extremity of
the ganglion close to the origin of (4).

(4) The great pallial nerve, which soon divides into
two: —(a) The anterior branch—this again divides usually
into two; and (b) The stouter posterior branch—this
runs outwards with the anterior branch and then turns
backwards. At this point it may be seen just beneath
the tissue covering the foot when the visceral hump 1s
lifted off. This branch gives off, from its outer side,
several nerves which run outwards to the mantle. Dr.
J. Travis Jenkins has shown by minute dissection that an
anastomosis does exist between the anterior and the
posterior branches of the pallial nerve (fig. 24). It 1s,

39

however, hardly as direct as in the more primitive
Docoglossa (e.g., Acmea). From the pleural centres are
also given off the nerve bands which together form the
visceral loop.

A first examination seemed to show a slender nerve
connecting the two pleural centres together, and running
just in front of the pedal commissure. — Microscopic
investigation, however, shows that this slender cord is a
fibrous band connecting the two otocysts and extending
beyond them to end in the dorsal surface of the foot.

The Labial and Buceal Nervous
System.—A fairly thick connective arises from the
inner border of each cerebral ganglion, and goes along
the floor of the snout, very soon swelling into what may be
called a labial ganglion. The commissure between the
labial gangha is thin and markedly curved, with the
coneayity forwards, the curve being underneath the
sub-lingual pouch. The labial ganglia give rise to nerves
going forwards to the lps, and, at the back, to a few
supplying the snout muscles, &c. (see fig. 25). From each
of these ganglia also arises a fine sinuous labio-buccal con-
nective, which enters the tissue of the odontophoral
cushion. When it reaches the dorsal surface of this organ
it swells into an elongated buccal ganglion. The two buccal
ganglia converge to some extent posteriorly, and are
connected with one another by a well-marked commissure
entering their posterior extremities. The buccal nervous
system is easily seen on removal of the gullet and the
radular sac, but the connection with the labial centres
is difficult to trace. The labio-buccal connective gives
off a nerve to the anterior parts of the odontophore. Hach
buccal ganglion gives rise, at its anterior end, to a nerve
supplying the lateral muscles of the odontophore, and
several small nerves arise from its inner border. <A well-

40

marked nerve arises from the posterior border of the
ganglion and supplies the posterior muscles of the
odontophore; it gives off a branch to the radular sae.

The Visceral Loop.—rTracing this loop from
the right pleural ganglion, which is easily exposed, we
find that it takes an upward course somewhat towards the
left, and soon enters the visceral region, wherein it travels
backwards and towards the left, through the salivary
glands for a short distance till it reaches the supra-
intestinal ganglion (so-called for phylogenetic reasons,
because it is on the connective which, in pre-Docoglossan
forms, ran across above the gut). This ganglion
varies very much in size and distinctness, but from it is
always given off a long and slender connective which runs
across to the left side above the crop and terminates in
a ganghon just beneath the left osphradium. This
osphradial ganglion supplies the sense organs near it, and
Bouvier has also found a slender nerve arising from it
which loses itself near the anterior wall of the peri-
cardium. ‘This, he says, is the vestigial ctenidial nerve,
and it has no connection with the osphradium. From
this, and from histological considerations, he seems
disposed to argue that the osphradium, so-called, is
entirely such, and not, as is commonly believed, both
osphradium and ctenidial vestige. Resuming our con-
sideration of the visceral loop we find that, from the
supra-@sophageal ganglion, it goes backwards and
towards the right for a short distance and then runs into
the visceral ganglion. This visceral ganglion gives rise
to the following chief nerves :

(a) The great visceral nerve which is richiy branched
and supplies the heart, left kidney, rectum and various
viscera,

(b) A nerve to the right kidney.

41

(c) Smaller nerves with doubtful distribution.

From the visceral ganglion the loop goes on towards
the animal’s right side, attaining its extreme position in
this direction near where it emerges in front from the
visceral mass, thence it goes across beneath the other
part of the loop and above the pedal cords to the’ left
pleural ganglion. At the extreme right point a few
ganglion cells can be found, and from this point arises a
thin and short nerve going to the right osphradiun,
beneath which there is a small osphradial ganglion. ‘This
extreme point would be the proper place for the usual
sub-intestinal ganglon, which, therefore, may be
represented by the few ganglion cells found there. It is
possible that the vanished ganglion is now included in
the left pleural ganglion, which is much elongated and
partly divided into two. The ‘“sub-intestinal ” part
of the visceral loop, that is the part next the left pleural
ganglion, gives off a couple of slender nerves, one near
the pleural ganglion and one further along (fig. 25).

It is characteristic of the Docoglossa that the visceral
loop lies entirely to the right of the fore gut, instead of
over it and near the right side, as in many other Gastro-
pods. This is in relation to the extra torsion of the fore
gut which has taken place in the group.

SENSE OreAns (figs. 25, 26, 27, 28) —A sub-epidermal
plexus of primitive type, with connected sensory cells,
still remains in some parts of the body, notably in the
sides of the foot of the young animal, but we also have
concentrations to form specialised sense organs. These
are the cephalic tentacles, the eyes, the otocysts, the
osphradia, and the pallial tentacles.

The Cephalic Tentacles ave a pair of specialised sensory
out-growths, situated at the sides of the snout (figs. 4 and 5),
They are extremely extensible, and their surface appears

42

corrugated; the apical part is pigmented. At the base of
the outer side is a small pit (the eye) lined by deeply
pigmented epithelium; this is discussed below.

The tentacle (fig. 25) is covered by a layer of columnar
epithelium, the cells of which are long and narrow, with
elongated nuclei, and they do not appear to be closely
packed when the tentacle is moderately extended. They
have a cuticle, which stains yellow with picric acid. There
are ordinary epithelial cells, sensory cells and goblet cells.
The sub-epithelial layer is a very compact felt-work of
fibres, many of which are undoubtedly nervous. Beneath
this is a mass of muscle fibres of the usual type, arranged
in bundles surrounded by connective tissue. Most of
them go from base to tip, and so are longitudinal tentacle-
retractors; there are also a few oblique fibres, but no
circular muscle occurs. The tentacle nerve goes down
the centre, receiving its fibres from the sub-epithelial
region, and finally entering the cerebral ganglion,
which is at the base of the tentacle. ‘There is a
good deal of loose tissue in the tentacle, as might
be expected in an organ with such a high degree of
contractility.

The animal waves its tentacles as it moves along, the
lateral surface near the tip just barely grazing the rock
surface over which it is creeping. This lateral surface
near the tip is the region of maximum sensitiveness. The
tentacles are undoubtedly tactile, and Professor Lloyd
Morgan considers them as the organs of the well-known
* homing” sense, but, though they may assist in that
function, his conclusion seems more than is warranted by
evidence. Limpets with the tentacles cut short have
‘homed ” successfully in several cases, and two animals
were observed at Granton doing the same, though the

entire tentacles had been removed. The pigment on the

43

tentacles is probably of use as a protection, lke that on
the visceral hump.

The Myes are a pair of organs situated on the outer
sides of the swollen bases of the tentacles (figs. + and 25),
but their shape and position does not seem to be affected
by the expansion, contraction and motion of these organs.

The eye is really a simple pit, lined by a continuation
of the general surface epithelium, part of which has
become modified in connection with the sense of light-
perception. This modified epithelium is found over a
shell-like area, which approaches nearer to the opening
of the pit on the lower and outer than on the tentacular
and central side (fig. 25). In this modified epithelium
we find two kinds of cells (fig. 26):——(1) Elongated sensory
cells, with swollen bases, and long clear processes directed
towards the surface; and (2) long Pigment cells sur-
rounding and filling up the interspaces between the
sensory cells. Hach cell consists of (a) an internal
tapering region, fairly clear, and containing a nucleus
in its outer part; (6) a broader middle region crowded
with minute pigment-granules, and (¢) a clear outer
region. The Pigment is black and resembles that of
the mautle.

Nerve fibres, going to the optic nerve, are found
beneath this sensory epithelium. The cuticle is well
developed over the epithelium of the sensory region. ‘The
eye of the Monobranch Docoglossa is more highly
developed, its cavity is filled with a jelly-like substance,
and the opening is a narrow slit. It is, therefore, almost
certain that the eye of Patella is degenerate, and this is
what might be expected from the conditions of life, since
the head remains under the shade of the conical shell.

The Otocyst, like the eye, first appears as a depression
in the epidermis at the side of the head. With further

tt

development it comes to lie far in, and its connection
with the exterior is lost. It 1s found finally just posterior
to the pleuro-pedal connective on either side (fig. 25).

The otocysts of the two sides are bound to one another
by a fibrous band passing ventral to the pleuro-pedal
connectives, and just in front of the pedal ganglia.
This band is continued a short distance beyond the
otocysts and terminates in the dorsal surface of the foot. Itis
probably related to the equilibrating function of the otocysts.

The otocyst nerve passes ventral to the pleuro-pedal
connective and goes forward between the cerebro-pedal
and cerebro-pleural connectives, fusing with the latter
not far from the middle of its length. Its fibres enter
the cerebral ganglion. The otocyst itself is a cavity lined
by ciliated epithelium, the cells of which ave in intimate
communication with the underlying nerve fibres. The
nucleus of these cells is more voluminous than in epithelial
cells, and in both Patella and Haliotis the cells are smaller
and longer than in many other forms. The otoliths are
small, usually rounded and numerous; the otocyst nerve
is hollow, and the otoliths may be found in its cavity some
distance away from the main cavity of the otocyst (fig. 27).

The Osphradia are patches of brown-pigmented
epithelium, situated at the sides of the posterior part of
the nuchal cavity, as already stated.

The component cells (fig. 28) are elongated and ciliated,
and, beneath them, we find a group of multipolar
ganglion cells (osphradial ganglion). In the immediate
neighbourhood of the osphradia, and covered by a con-
tinuation of this epithelium, are projections, supposed by
some to be the last vestiges of the ctenidia of the limpet’s
ancestors. Their lacune are encumbered with corpuscles,
and Boutan thinks they have some special glandular
function connected with the blood.

45

The Pallial Tentacles round the mantle edge have
already been mentioned in general terms, and it has been
said that they are of two kinds, large and small, of
similar structure. A large tentacle can extend as much
as } of an inch beyond the mantle edge. The surface of
a tentacle presents a number of fine encircling ridges,
and, by the action of muscle strands, it can be retracted
into a pit, when its surface is thrown into prominent
circular folds. The pits for the large tentacles are
situated further ventral than those for the small ones
(figs. 8 and 24). The tentacle (fig. 8) is covered by
columnar epitheium made up of two kinds of cells
resting on a basement membrane :—(a) Relatively broad
goblet cells with a basal nucleus; and (b) Sensory cells
which are slender, and narrowest in the middle where the
nucleus is situated. The outer end bears a number of
stiff sensory processes, so that the cell is of the brush type
(prnselzelle) described by Flemming.

Underneath the basement membrane scattered ganglion
cells and nerve fibres can be made out; the latter converge
to the axis of the tentacle, where they form a nerve
which is connected with the peripheral part of the nervous
network of the mantle. At the points of fusion of the
tentacle nerves with this network a few ganglion cells are
found. Where a tentacle is retracted its nerve is thrown
into a coil.

Beneath the epithelium of the tentacle is a faily
continuous sheath of longitudinal muscle fibres, the
majority of which are inserted into the tip, though some
end at the basement membrane. Most of them collect at
the base into two bands which run into the mantle in
different directions, those of a laterally placed tentacle
being relatively more or less anterior and_ posterior,

though not necessarily in the same horizontal plane. The

46

core of the tentacle is a loose network of connective tissue,
muscle fibre, and nerve, and contains large blood spaces.

The above described arrangement of the muscle fibres
of the palhal tentacles accords well with our supposition
as to their nature, for, better than any other arrangement,
it allows of their sweeping over the rock surface, thus
enabling the animal to recognise it in some way, and so
subserve the “homing” faculty. When the animal
“ shuffles” round on the sear on returning from an excur-

sion they are in active use.

CIRCULATORY ORGANS AND Ca@Lom.

Broop Sysrrmu.—<As in Molluses generally, this is to ¢
large extent lacunar, and is greatly developed at the
expense of the eewlom, which is reduced to small dimen-
sions. The so-called body-cavity is a hemoceele,
consisting of blood spaces, and the ceelom is reduced
to the pericardial cavity, which is, therefore, not a blood
space. The blood is a colourless fluid in which float
amoeboid corpuscles. The parts concerned in blood
circulation are the Heart, the Arteries, and the irregular
Blood Spaces.

The Heart may be described as a specialised portion of
the heemocele, which has projected itself into the eelomic
space called the pericardium.

In the primitive Chiton, therefore, the feebly difteren-
tiated heart is surrounded by ecelomic epithelium, which
goes up on either side to the dorsal wall of the pericardium,
7.e., the heart is, as it were, suspended from the dorsal wall
of the ceelom in an infolding of the lining of that cavity.
In Patella the heart (ventricle) is also connected with the
pericardial roof, but the connection is not so regular and
complete as in Chiton. In most Gastropods, this connec-
tion has disappeared, and it is even possible that it is a

47

secondary consequence of the special reduction of the
pericardium in Patella.

The Heart.—TYo expose the heart of Patella, carefully
make a transverse incision in the front part of the roof
of the pericardium. Then remoye this roof, first cutting
the fibres connecting it with the heart. The heart, thus
exposed, will be seen to consist of a thin-walled auricle
g. 29):

The auricle receives blood from the pallial gills and

in front, and a thicker walled ventricle behind (fi

related mantle skirt, from the roof of the nuchal cavity,
and, perhaps, is directly connected with the reduced left
kidney. Blood from the pallial gills and mantle skirt
is returned by the great pallial vein into the left front
of the auricle, while that from the nuchal cavity is con-
veyed by a number of small channels opening into the
right front part of the auricle. Perhaps among these
small channels, which can be very distinctly seen in a
fresh uninjured specimen, some bring blood from the
left kidney; at any rate, when the auricle is filled by
coloured injection the colour is communicated to this
kidney.

On opening the auricle by a transverse incision, the
large orifice of the pallial vein can be noted, and, to the
right of this, a linear series of small apertures from the
httle. channels just mentioned. At the back is the
auriculo-ventricular septum, pierced by a transversely
oval opening, the edge of which is thickened. The
junction of auricle and ventricle is shown by a well-
marked constriction, which corresponds to the thickened
margin of the auriculo-ventricular septum.

The ventricle stretches right across the pericardium,
and its antero-dorsal wall is thicker than the postero-
ventral. As has already been said, its dorsal wall is
connected by fibres with the roof of the pericardium,

48

along a line going obliquely from right to left (i.e., the
long axis of the ventricle), and representing what was
the main or antero-posterior axis of the ventricle in
ancestral Molluses.

On opening this chamber of the heart, the interwoven
muscle fibres, which form much of its wall, are seen
radiating from the thickened margin of the auriculo-
ventricular aperture, which is situated near the middle
of its antero-dorsal side. The aperture is guarded by
two valvular flaps which project into the ventricle cavity.

The main aorta and its posterior branch, the visceral
artery, diverge in opposite directions right from the
origin, and as they run parallel to, and in close connection
with, the ventricle wall, we get an appearance as of a
third chamber of the heart. This appearance is
emphasised by the fact that the beginnings of the aorta
and visceral artery are swollen, and that this pseudo-
chamber resembles the ventricle in general characters.
The heart of Patella was described as three-chambered
by Wegmann, who gave the name ‘ Intermediate
Chamber” to that which is here called “ Ventricle,” and

the name “ Ventricle”’ to the pseudo-chamber or aortic
bulb formed by the swollen arterial bases. | Wegmann’s
determination of parts would make the lmpet’s heart
practically suz generis, and would remove the genug very
far from the Rhipidoglossa, and even from its nearer allies.
The present account, on the other hand, shows the essential
similarity between Patella and the other Docoglossa in
this respect, and agrees with the conclusions of Haller and
of Boutan. The aperture between the ventricle and the
aortic bulb is guarded by a valvular flap which projects
into the cavity of the latter.

Arteries—Vhe Aorta, of which the origin has just

heen described, runs to the right end of the pericardium,

19

and then goes on in front of the rectum, soon curving
downwards and forwards until it becomes continuous
with an anterior sinus, surrounding at least the front
part of the radular sac, and enlarging forwards so as to
enclose the cushion of the odontophore.

The movements of the odontophore must affect this
sinus, especially as it connects with the pedal sinuses
and other blood channels. It is in this way that these
latter sinuses probably got blood pumped into them.
This opinion is greatly strengthened by the observation
of the head of living specimens, in which the odontophore
can sometimes be seen moving forwards and backwards
almost rhythmically. It is interesting to note that a
similar and similarly connected sinus exists in Chiton,
and the pumping work thus done outside the heart
may be correlated with the feeble muscularity of the
ventricle in so many of the lower Molluscan forms. The
existence of arterial branches of the aorta, other than
the posterior or visceral artery, is doubtful, though
possibly one goes into the salivary glands.

The Visceral Artery seems to branch almost imme-
diately after leaving the left posterior corner of the
pericardium, one branch certainly goes to the gonad
region and the other also seems to enter the visceral
hum.

The course of the arteries cannot be traced far from the
pericardium, the greater part of the blood system being
lucunar, 7.e., consists of spaces which have not a definite
epithelial lining.

Lacune and Sinuses (Blood Spaces ).—Blood from the
Anterior Aorta goes into the great anterior sinus sur-
rounding the radular sac and odontophore cushion.
Thence it proceeds into

(a) The Pedal Sinuses (fig. 8a).—These are fairly

E

50

distinct spaces running back through the foot on either
side just internal to the nerve cord. The anterior sinus
opens into these near the pedal gangha.

(b) Spaces in the visceral hump, from that part of the
sinus which surrounds the radula.

The foot is, therefore, supphed from the anterior aorta
vd the anterior sinus, while the visceral hump is supplied
partly in the same way and partly by the posterior artery.
The shell muscle seems to be furnished with blood from
the pedal sinuses. The impure blood from the foot and
that from the visceral hump goes to the perivisceral
sinus which, as its name implies, surrounds the viscera.
From this sinus some of the blood goes to the blood
spaces in the trabecule, etc., of the large right kidney,
and, more doubtfully, to those of the left kidney. For
this reason the perivisceral sinus and large kidney are
about co-extensive. Practically the whole of the blood
from the perivisceral sinus, including that which has
traversed the kidney trabecule, etc., ultimately finds its
way out to the mantle. Some of it goes to the nuchal
mantle, while the remainder reaches the mantle skirt
by way of channels running between the fasciculi of the
shell muscle.

Before describing the oxygenation of the blood, it 1s
convenient to describe the circulation in the head, in
the nuchal mantle, and in the left kidney. The main
channels in the head region are the parts and ramifications
of the anterior sinus, ?.e., spaces below the lning epithe-
lium of the gut, and particularly of the odontophore.
From this space, blood seems to go all over the head
around the body wall. The inner lps, which are
puckerings of this gut epithelium and subjacent. tissue,
contain each a blood space so that their opening and

closing may be brought about in part by changes in blood

51

pressure. It is impossible to say whether the circulation
in the head is a loop system, or whether it is merely an
enlargement of the great blood space around the
odontophore—probably both interpretations are to some
extent correct, and, if so, the foot is partly supplied with
blood that has already been around the head. Possibly,
however, the blood in the head becomes re-purified where
it runs near the surface in places covered by delicate
skin (e.g., tentacles, etc.).

From the anterior end of the perivisceral sinus some
blood goes into a network of spaces in the nuchal mantle,
and these lead off ultimately into the auricle through
the linear series of small apertures seen on opening the
latter. This blood must be partly oxygenated. Blood
channels from the mantle, in more primitive forms,
probably opened, as they do in the Rhipidoglossa, into the
efferent ctenidial veins, but as these latter disappeared
they have become directly connected with the auricle, and
have undergone compensating development.

The blood reaching the left kidney, in a more primitive
form, would have gone out to the corresponding ctenidium
for oxygenation, returning thence by the left branchial
vein to the left auricle. A change has necessarily had
to follow the disappearance of the left ctenidium, and
now in Patella there is some doubt us to the circulatory
arrangements of the left kidney. Possibly it receives
blood from the perivisceral sinus and passes it on almost
direct to the auricle. If this is not the case, this organ
must be in direct dependence on the auricle for its blood

supply—a condition which its homologue in /Zalzot’s seems
to have attained. The circulatory arrangements of the

left kidney, even of the primitive Patella, show how far this
organ has departed from the ordinary condition of a kidney
among the lower Gastropods.

D2,

The blood which does not reach the heart through
the minute channels of the nuchal mantle, or vid the
left kidney, goes out to the mantle skirt and pallial
gills through channels running between the fasciculi
of the shell muscle, and, afterwards, in the substance of
the mantle skirt. After oxygenation in the pallial gills
and mantle edge, blood is returned to the large palhal
vein by small veinlets projecting on the ventral surface
of the mantle skirt (fig. 7). Those from the edge (V1. fig.
7) unite with those from the gills (G.V. fig. 7), and the
channels thus formed turn outwards to open into the
large pallial vein. The veinlets from the mantle edge
pan, therefore, be seen at intervals crossing over the
ventral surface of this large pallial vei, which runs
completely round the mantle skirt just external to the
attachment of the pallial gills (fig. 5). The completeness
of the circle is broken at one point directly anterior to
the left front end of the shell muscle; here the two sides
of the vein bend inwards and fuse into one trunk which
runs along the inner side of the shell muscle to the left
front corner of the pericardium. :

‘

It must be remembered that, though the name ‘ vein ”’
is apphed to them, none of the blood channels in the mantle
are true vessels, even the great pallial vein being of the
nature of a lacuna.

Catomic System.—The extensive development of the
blood system has entailed a corresponding reduction of
the celom, of which there remains practically nothing
except the pericardium. The form and position of the
pericardium have already been described, and it only
remains to say now that the pericardial gland seems to
be absent. The pericardium communicates with the
large right kidney, and, as we can positively state,
with the small left one as well, but this matter will

58

be discussed later in dealing with the excretory
organs.

Haller described a pair of celomic cavities between the
visceral hump and the foot, but Pelseneer does not con-
firm this, and our results tend to the conclusion that the
only epithelium-lined cavity in this region is that of the
large kidney.

The cavity of the gonad, visible only in a very young
specimen in which the sex products have not yet been
much differentiated, is necessarily a remnant of the
celom. It becomes practically obliterated at a later stage.

RESPIRATORY ORGANS.

With the. specialisation of the right side of the
branchial chamber as an excurrent channel for waste
products, the right ctenidium, we may suppose, dis-
appeared at an early stage in the descent of the
Docoglossa. In the less modified members of this group,
Acmea, etec., the work of respiration is, therefore, per-
formed mainly by the surviving left gill, but in part also
by the mantle skirt, which has increased in importance as
the shell became more cup-like, and its projecting edge
spread farther out. The mantle skirt in these Mono-
branch forms already shows a tendency to the formation
of a series of transverse ridges, constituting incipient
secondary or mantle gills (Lotta and Scurria). In Patella
both primary gills are reduced, being represented only
by vestiges (fig. 4). The nuchal chamber, in which these
vestiges are contained, is equivalent to the branchial
chamber of a Pleurotomaria, Iissurella, or Acma@a in a
reduced condition. This chamber, however, still plays
a subordinate part in respiration, although that function
is mainly effected by the circlet of pallial gills, which
have now attained a high degree of development.

54

It will be convenient to give details under three
headings:—Nuchal Roof, Vestigial Ctenidia, Pallial
Gills,

The Roof of the Nuchal Chamber has already been
mentioned, and it has been stated to be permeated by a
network of blood channels; its histological structure has
also been discussed. The abundance of blood channels,
the general structure, and the fact that blood goes direct
from it to the auricle makes it probable that this tissue
acts as a respiratory organ. Probably the movements
of the head, nuchal floor, etc., enable the cavity to
function as an imperfect sort of lung when the animal
is left uncovered by the tide. Individuals lying far up
the shore are uncovered for the greater part of the time,
and such a specialisation would undoubtedly be
advantageous.

The supposed Ctentdial Vestiges are probably entirely
functionless, but there is a large osphradium, with under-
lying osphradial ganglion, in connection with each. It
is generally thought that the function of an osphradium
is the qualitative testing of the respiratory medium, and
the retention of these organs in Patella is an argument in
favour of the respiratory activity of the nuchal chamber.
The osphradium has been described in the account of the
sense organs, where it is also stated that Bouvier does not
accept as such what are here described as ctenidial
vestiges. The vestige is a mass of connective tissue con-
taining blood spaces, and situated near the osphradium ;
these blood spaces contain numerous corpuscles, and
Boutan remarks that the mass resembles a lymph gland.

The Secondary or Pallial Gils have already been
mentioned. ach gill is triangular in form, with the
base attached to the mantle. The inner side is curved and
runs from the mantle to the outwardly projecting apex,

55

and, from this apex, the third side runs straight up to
the mantle (fig. Sa). A pallial gill is merely a down-
growth from the mantle skirt, and both, therefore, have
an essentially similar structure. It is covered by
epithelium, with many cells ciliated, some glandular, and
some sensory (fig. 9b); beneath this is a nerve plexus,
with a few multipolar ganglion cells (fig. 96). Within
this we find a comparatively small amount of muscle
fibre and other tissue enclosing a large blood space sub-
divided by fibrous trabecule (fig. 9a). Blood enters the
inner border of the gill and leaves from its outer border,
and in these two positions we, therefore, find rather large
and distinct blood spaces.

We may now shortly summarise what has been said
concerning respiration and the circulation of the blood
in the mantle.

Blood comes in from lacunze running between the
fasciculi of the shell muscle; it is then distri-
buted both to the pallial gills (through a blood
channel running ventrally and reaching the inner
border of the gill) and to the mantle proper (through
lacune running further dorsally). In the latter set of
lacunz, the blood seems mostly to reach the manile edge
whence it returns through channels which project like
veinlets on the ventral surface of the mantle. In the
former set of lacunz it is distributed over the gills and
is then collected into ventrally placed efferent lacunie
(fig. 7), which leave the outer border of the gill
and join the veinlets from the mantle edge. The united
sinus opens into the inner side of the great pallial vei
(fig. 7). Though the pallial gills and the nuchal roof are
the only distinctly respiratory organs, it must not be
supposed that they are the only places in which blood is
oxygenated. The skin in several parts seems to be

56

sufficiently delicate to allow the necessary diffusion to
proceed; such places, for example, are the surface of the
tentacles, of the inner lips, ete.

EXCRETORY ORGANS.

Two kidneys are present, of which the right is much
the larger.

The much-reduced Left Kidney (see figs. 4, 5, 8, 31)
is a small compact sac with only a limited excretory activity.
The wall of this kidney is thin where it abuts on the rectum.
but is much thickened where it adjoins the pericardium. It
possesses renal epithelium like that of the right kidney, and
the amount of excretory matter visible is fairly large. The
concensus of opinion is in favour of its possessing a reno-
pericardial pore, a conclusion which our preparations
confirm. ‘The thickened wall between the left kidney and
the pericardium is a heterogeneous mass made up of a
few muscle fibres and connective tissue, with occasional
granular cells of lymphatic function. In this mass are
numerous cavities which are blood lacune.

The Right Kidney, which seems to be the main
excretory organ throughout the Gastropod series, has, in
Patella, been flattened out and spread around the visceral
hump, in correlation and connection with the great
perivisceral blood sinus. It is a large structure made
up of several lobes (figs. 4, 5, 8, 31) :—

(a) An anterior dorsal lobe, extending superficially over
the front of the visceral hump behind the pericardium
and rectum.

(b) A posterior or perivisceral lobe, extending around
the visceral hump backwards along the right side, and
then a fair distance forwards along the left. In very old
limpets (a) and (6) are united distally.

(c) A ventral lobe, extending on the ventral surface

i

Or

of about the right half of the visceral hump and connected
laterally with (0).

(d) A sub-rectal lobe stretching beneath the rectum
and the back part of the left kidney, and with its apical
portion contiguous to the pericardium.

When coloured by excretory matter or by injection,
the outlines of the dorsal parts of the kidne$ become
visible, making it appear to be a much branched
arborescent gland. On opening, this impression is cor-
rected, for we find the structure characteristic of the
excretory organs of Mollusca. It is essentially a sac
lined by renal epithelium, but the space is much
obstructed by the growth across it of subjacent tissue,
forming pillars over which the renal epithelium is, of
course, continued. These pillars or trabecule contain
extensions of the part of the perivisceral sinus in contact
with the kidney (see fig. 32a). They increase the
excretory surface, and in this way add to the efficiency
of the organ. It was long supposed that the blood
channels actually opened into the kidney, and that the
blood, in this way, received water; ‘but this idea is
now quite discredited, and we know that the renal
epithelium forms a boundary everywhere between the
blood sinus and the kidney cavity. ‘This renal epithelium
consists of a single layer of cells of variable size. Many
of them are ciliated, especially when young. As they
grow older, these cells amass concretions, some developing
a number of small ones, other a few larger ones (fig. 520).
These concretions contain, mainly, nitrogenous waste,
and are dark brownish green in colour. The relation of
the kidney to the perivisceral blood sinus is that typical
for Molluscan kidneys generally, the renal epithelium
forming a much complicated partition between a blood
space on the one hand, and, on the other, a cavity

58

opening to the exterior. The excretory papille of the
kidneys have a small central canal lined by ciliated
epithelium, outside which is a fairly strong ring of
circular muscle. The blood supply of the kidneys has
already been spoken of in the account of the circulation,
to which reference should be made.

The pericardial communication of this right kidney
has often been discussed, the difficulty of observation
having led to the enunciation of conflicting opinions by
different workers. It is generally admitted that the
reno-pericardial canal opens into the sub-rectal lobe of
the kidney, but the exact position of this opening, and
the length of the canal, are subjects of dispute.
Cunningham makes the canal open into the dorsal
surtace of the sub-rectal lobe some distance to the left
of the rectum, while Goodrich and Pelseneer find the
opening on the ventral surface of that lobe practically
ventral to the rectum. Our results (fig. 31) support the
latter view, and we find that the long canal opens into
the extreme right end of ‘the pericardium. Most of our
sections differ somewhat from those of Pelseneer and
Goodrich as regards the space relations of kidneys, peri-
eardium and nuchal cavity, but we think these relations
vary a good deal in different specimens, and also change
somewhat with the age of the individual.

REPRODUCTIVE ORGANS,

The gonad occupies the ventral face of the left side of
the visceral mass (figs. 6, 8) in both sexes. It varies very
much in size at*different seasons. In a very young form
it is practically a pouch, the cavity of which is celomice.
This cavity is hned by germinal epithelium, which, with
its underlying tissue, grows in as folds, some of which

unite into trabecule (fig. 35a), thus converting the pouch

59

into a mass of sex cells covering the connective basis. As
the gland grows in each season towards maturity it pushes
forwards, sometimes as far as the level of the csophageal
pouches, and often extends across the median line towards
the right side (fig. 8). At the time of complete maturity
the ova are surrounded by a tough coat, which possesses
a micropylar opening. The ovum contains numerous
yolk spherules (fig. 33>). The sperms (figs. 34+) are very
minute and consist of head and tail as usual.

The gonad expels its products by rupture into the
cavity of the right kidney, and they thus make their way
to the exterior. The gland seems peculiarly liable to
overgrowth, and, among specimens collected in autumn
on the Welsh coast, various ruptures can frequently be
noticed, sometimes between the shell muscle and the
foot, sometimes above the shell muscle, though this latter
does not seem to be of any advantage for the expulsion
of sex products. The season of sexual maturity is the
autumn; Boutan finds it to be about September at Roscoft ;
at Aberystwyth we think it is somewhat later. <A few
limpets have been found by Gemmill with male and
female regions in the gonad, and he also notes that the
percentage of the two sexes do not depend on tidal level.

Though the gonad is situated on the animal’s left side
it must not, therefore, be supposed to be the (post-
torsional) left member of an ancestral pair, for there have
been such changes in connection with the consolidation
of the hump that the present position cannot be taken as
a guide. The evidence rather points to the view that we
have here the (post-torsional) right member of the
ancestral pair, for :—

(1) In all other classes of Gastropods the (post-torsional)
right gonad, and, in Cephalopods, the one corresponding
to this (ae., the left), is the one which survives.

60

(2) The blood supply of the gonad (from the posterior
artery) is identical with that of the gonad of other
primitive Gastropods, in which that organ distinetly
belongs to the post-torsional right side. Possibly, how-
ever, the other member of the ancestral pre-torsional pair
of gonads has fused with it.

(3) The sex products are expelled into the right kidney,
and, perhaps, the extension of this organ on the ventral
surface of the hump is, in part, an attempt to preserve
its ancient connection with the gonad, an attempt which,
as we have seen, 1s partially unsuccessful.

It is, however, not impossible that the apparently single
gonad has arisen from the fusion of the primitively
distinct pair.

DEVELOPMENT.

See figs. 35 to 42)
co)

The ovum has already been described as a small
globular body covered by a tough coat, which is incom-
plete at one spot—the micropyle. The sex products are
extruded into the surrounding sea water and fertilization
occurs in a haphazard fashion, the sperms entering vd
the micropyle. Sperms seem to make their way
occasionally into the females, as ciliated embryos are
occasionally found which have not yet been extruded. It
is not impossible that hermaphroditism, to a very small
extent, may be rather more common than is supposed,
and this kind of variation would seem to be advantageous
to the animal, though self-fertilization has its draw-
backs. Fertilization may be obtained artificially by
injecting the sex products of one kind into the nuchal
cavity of a mature member of the other sex, or by
mixing both kinds of sex products in a small volume of
sea water. By observation of artificially fertilized ova,

61

Patten has been able to follow the early stages of
development.

The fertilized ovum segments into two, and then into
four fairly equal cells. Subsequent divisions are
markedly unequal, giving rise to numerous small cells
(future ectoderm), and, less rapidly, to a few larger cells
(future endoderm and mesoderm). The “ Gastrula ”
stage is initiated by growth of the smaller (ectoderm) cells
over the large cells, 2.e., by epiboly. The blastopore is
the spot over which the small cells do not spread. By
differential growth a blastoccele is now formed, into which
the inner ends of the large cells grow. These cells bud
off smaller ones towards the blastoeccele, and from one of
these are developed the two primary mesoderm cells.

The ectoderm is mostly ciliated at first, but two adjacent
transverse rows of cells soon develop larger and more
conspicuous cilia, thus forming two (pre-blastoporal)
ciliated rings around the embryo. These together form
the ‘‘ Prototroch,” the first rudiment of the “ Velum.”
The apical cells of the embryo also develop very long
cilia, and those around them shorter but still conspicuous
ones. These ciliated cells lengthen and sink, forming a
fairly flat apical plate. At the opposite pole of the
embryo two ectoderm cells increase in size and also bear
cilia. These are called the anal ciliated cells. Another
group of ectoderm cells, posterior to the Prototroch and
on the dorsal side, become depressed and lengthened.
These form the shell gland which increases in extent
especially at the back, making this part of the surface
of the embryo convex, and shifting the ventrally-placed
blastopore relatively forwards. Meanwhile the blastopore
changes its relative position, becoming U-shaped and then
slit-like. In the position of its most anterior portion,
there occurs an insinking of ectoderm, which pushes this

62

part of the blastopore inwards, and its ultimate destiny
is to form the opening from the (ectodermic) stomodceum
into the (endodermic) mesenteron. The remainder of the
blastopore has closed. On either side of the blastopore
in its slit-like stage we find one of the mesoderm cells,
which divides and gives rise to a pur of rounded
elevations. ‘These two elevations are the rudiments of
the foot.

At this stage the larva is practically a trochosphere.
Its preoral region is large and possesses a well-developed
prototroch and an apical plate. Near the apical plate
there appear two small ciliated elevations each consisting
of a single cell with cilia. The shell gland secretes a
shell which becomes more and more convex outwards,
attaiming a rounded cup-shape. The originally solid
mass of endoderm becomes hollowed out and arranges
itself as the lining of the midgut. It gives off a diver-
ticulum postero-ventrally, which ultimately opens to the
exterior forming the rectum. The fore-gut gives off a
diverticulum ventrally, which is the rudiment of the
radular sac.

The ectoderm thickens and projects along a band just
external to the shell margin, and this is the first trace
of the mantle skirt. The anal diverticulum reaches the
ectoderm ventral to this, and between anus and mouth
we now find the well-marked pair of foot rudiments.
A pair of depressions occur at the sides of the mouth
and grow inwards along the sides of the foot, becoming
closed sacs and forming the otocysts.

As the shell and foot continue to grow, the anal diver-
ticulum and the rudiment of the branchial cavity become
confined between them, and we then observe their
migration from the postero-ventral position, along the
animal’s right side, ventral to the mantle skirt, up to an

63

antero-dorsal position behind the prototroch, which has
now grown out and become the “ Velum,” the differ-
entiated locomotor organ of the larva. The larva is
bound to its cap-shaped shell at first by a strand of tissue
inserted into the apex of the cap. Later on, Boutan has
observed in Acmea the development of a pair of antero-
lateral muscular strands inserted near the shell margin
and functionally replacing the apical strand. This pair
of muscles spreads backwards, and ultimately form the
horse-shoe shell muscle of the adult.

The shell now grows chiefly by additions to its posterior
border so that its mouth is much widened and the apex
comes to lie relatively far forwards. The apical portion,
which may show the beginnings of development of a
symmetrical spire, breaks off and the hole is closed by a
secretion of nacreous material. The next stage known
has the foot broadened into the adhesive sole of the adult,
the shell muscle horse-shoe shaped, and the shell conical.
This stage possesses a glandular streak on each side of the
foot, homologous with that found in the adult of Naeella
and its allies; it disappears later.

In further growth of the shell, the anterior margin
seems to share more largely, so the apex subsequently
undergoes relative motion towards the back. This is
more marked in limpets which live in positions much
exposed to. the action of the waves, such limpet-shells
being lower cones than those in which the growth is more
equal,

CONCLUSIONS.

Of the features which have now been described in Patella,
the following characterise the Cyclobranchs, the specialised
group of the Docoglossa to which our type belongs :—

(1) The highly differentiated crop placed so as to form a

64

curve with concavity towards the right anterior corner of
the visceral mass.

(2) The intestinal coil (Int. 2) placed on the dorsal
surface of the visceral mass. Its position varies in different
forms, being often much further to the left than it is in the
lim pet.

(3) The presence of three uncini and nearly always, at
any rate, three lateral teeth on each side of each row of teeth
in the radula.

(4) The strong odontophore with its squarish front, its
highly differentiated transverse muscles, and its numerous
cartilages (more than the usual two pairs).

(5) The large development of buccal glandular tissue.

(6) The pseudo-chamber (or aortic bulb) formed by the
varying amount of swelling of the basis of the aorta and
posterior artery.

(7) The practically transverse posterior boundary of the
triangular pericardium.

(8) The absence of a ctenidium, the smallness of the
branchial (nuchal) cavity, and the occurrence of a circle of
pallial gills. This circle is incomplete above the head in
some Cyclobranchs (/7elcion), and in others the gills in that
position are small (Nacella).

(9) The presence of a lateral or epipodial streak. It has
not been found in Helcion and its allies, and it disappears
in the adult Patella, but persists throughout life in Nacella,
It is not known in any of the more primitive Docoglossa
(Monobranchs and Lepetide).

(10) The specially degenerate eye.

(11) The numerous differentiated large and small pallial
tentacles.

The following features characterise the whole group of
the Docoglossa :—

(12) The apex of the simple conical shell is typically bent

65

over so that it points forwards and is usually placed in front
of the middle of the shell. In adult Cyelobranchs growth
of the front edge of the shell often makes these features less
distinctly marked.

(13) The long raking radula with a limited number of
specialised clawed teeth. The median tooth is not specially
differentiated (except in some Lepetidz), and is typically
reduced and often absent.

(14) The disposition of the parts of the gut shown in
Fig. 10a, except as regards details mentioned in (1) above.
This is not quite constant throughout the group, but is
found in several types and the disposition of the stomach-
gull 1s quite characteristic.

(15) The position of the visceral loop of the nervous
system well to the right of the median line.

(16) The practically triangular pericardium completely
filled by the heart and situated at the left anterior corner of
the dorsal surface of the visceral hump.

(17) The superficial extension of the right kidney (almost
characteristic) and the retention of a certain amount
functional excretory tissue by the tiny left kidney.

The following features show that Patella and the Doco-
elossa rank among the lower Gastropods :—

(19) The presence of a horse-shoe muscle derived, as the
development of Acmea shows, from the backward extension
of a pair of laterally placed muscles.

(20) The retention of external symmetry throughout the
development, as far as is known.

(21) The feeble concentration of ganglia in the nervous
system and the anastomosing commissures between the
pedal cords.

(22) The presence of a complete labial commissure
and the form both of this and of the buccal nervous
systeni.

F

66

(23) The possession of a pair of osphradia placed right
and left in the nuchal cavity.

(24) The possession of two kidneys right and left, the
left still possessing some excretory tissue.

(25) The extrusion of the sexual products through the
right kidney and the absence of all accessory sexual organs.

Of those characters which are peculiar to the Docoglossa
amongst Gastropods—

No. (12) is an adaptation to the adhesive habit, while
(14), (15), and (17) are consequences of the consolidation of
the visceral hump involved in the development of (12).

No. (16) is due to the disappearance of the right etenidium
and the subsequent shifting of the heart to the left side so
that it might he behind the remaining etenidium (which has
also disappeared in Patella).

No. (18) is a specialisation enabling the animal to gather
its food by scraping the rock or other surface over which
it creeps.

Among the Cyclobranchs:—

Nos. (1) and (5) are specialisations to overcome the
difficulty of digestion of the tough food.

Nos. (2) and (7) are special consequences of the further
compression of the parts of the visceral hump among these
forms.

No. (3) is an undoubted characteristic, but is not easy to
understand,

No. (4) is an adaptation to the habit of extruding the
tip of the odontophore for raking purposes and to the
consequent need of a flat dorsal surface, a broad front. and
facilities for adjustment.

The reduction of the pericardium has led to the develop-
mont of No. (6), this arrangement promoting the regular
ud unimpeded circulation of the blood.

No. (8) is the feature which gives the group of ‘“ Cyelo-

67

branchiata” its name. The etenidium in a fairly deep
branchial cavity could not be kept well rinsed among
sluggish forms often left uncovered by the tide. The
ctenidium, therefore, became inefficient and has disappeared,
the pallial gills being a new and compensating development.

No. (9) is an interesting feature not easy to account for.
Perhaps the glandular secretion improves the animal's
power of holding on and helps it to withstand desiccation
in the exposed spots which these forms typically inhabit.

No. (10) is a consequence of having the head always
under the shadow of the shell.

No. (11) is a specialisation which helps the animal to
obtain a topographical acquaintance with its immediate
neighbourhood.

It will thus be seen that the Docoglossa, though un-
doubtedly correctly included among the lower Gastropods,
are yet specialised on lines of their own in connection with
their adoption of the habit of adhering to exposed surfaces
and making limited excursions for the purpose of raking
up food.

They cannot be said to be directly and closely related to any
of the other primitive Prosobranchs, the connection in each
case being due to descent from a not very remote common
ancestor. Plewrotomaria and the Trochidee have specialised
on quite other lines as regards the shell, foot, and shell
muscle, and this is true also of Haliotis, which has adapted
itself to creeping about in chinks and confined spaces. The
great contrasts in the respiratory systems and in the gereral
disposition of organs show the distance that separates these
forms phylogenetically, The Fissurellide have also evolved
on quite other lines though they have a shell muscle and
visceral hump closely resembling that of the Docoglossa
externally. The most striking contrasts are the presence of
a pair of ctenidia in the Fissurellide and the shortening of

68

the path of the excurrent stream of the branchial cavity in
these forms by the deepening of the slit or its conversion
into a hole at or near the apex of the conical shell.

Scissurella is certainly one of the most primitive Proso-
branchs known and may be more nearly related to the
Docoglossa than are the other groups named, but even in
this form the left kidney is specialised in the same way as
in Pleurotomaria, ete.

The fossil Bellerophontide were very primitive Gastropods,
though we cannot know enough about them to say much of
their relationships. It is, however, possible to conceive of
the evolution of the Docoglossa from forms something like
them, the symmetrical spire undergoing relative reduction as

¢
C

the posterior margin of the peristome grew out.

EXPLANATION OF THE PLATES.

Reference Letters used in the Plates.

Note.—The letters d—n in Figs. 13b and 13c refer to the minute

muscles of the odontophore, and are not included here.

A, = Thin brown external layer
of shell.
A.C. = Anterior

odontophore.
A.-D. Lobe = Antero-dorsal lobe
of right kidney.

cartilage of

A.-L.C. = Antero-lateral car-
tilage of odontophore.

An. C. = Anal cells.

Ant. Ao. = Anterior aorta.

Ap. D. = Apical disc.

B. = Thick middle layer of shell.

(a) Outer part traversed by
branching canals.

(b) Clear inner part.
B.C. = Buccal cavity.
Bp. = Blastopore,
Buc. = Buccal commissure.
JE}QVOn ID) =
Bue-G. =
Buce.Gl. = Buccal glands,

Duct of buceal gland.
Buccal ganglion,

C. = Internal laminated layer of
shell,

Cer. = Cerebral ganglion.

Cer.C.

Cil.R.
ring which becomes proto-
troch,

= Cerebral commissure.
= Pre-blastoporal ciliated

C.-Ped. = Cerebro-pedal con-
nective.

C.-Pl. = Cerebro-pleural con-
nective.

D.E.G. = Dorsal fold of gullet
wall.

D.H.P. = Duct of hepato-
pancreas.

D.P, = Dorsal palatal plate.

D.Pal.Gl. = Dorsal pallial gland-
tissue.

ict. = Ketoderm.

H.-M. = Endo-mesoderm.

Hp. fF. = Epithelial fold.

Fl. = Flap of lateral (‘‘ epipo-
dial ’’) streak.

E.R, = Foot rudiments (meso-

blastic).
.-l, = Gland tubules of lateral

[ont
a2

(‘‘ epipodial ’’) streak,
G.M.T. = Greater pallial ten-
tacle.
G.M.V. = Great pallial vein.
Go. = Gonad.
G.V. = Veinlet
gill.
H, = Head.
H.-P, = Hepato-pancreas.

from pallial

Int. = Intestine (the coils are
numbered in Figs. 10a and
b).

T.L. = Inner lip (only a line is
shewn as neither lip comes
in median section).

K.C. = Kidney cavity.

Lab. = Labial commissure.

Lat.St. = Lateral (‘epipodial’’)
streak.

L. Au. = (Morphologically left)
auricle.

L.K. = Left kidney.

L.K.Ap. = External aperture of
left kidney.

L.M.T. = Lesser mantle ten-
tacle.

L.Osph.G. = Left osphradial
ganglion.

M. = Mantle or pallium.

Mcp. = Micropyle.

Mesen = Mesenteron (endoder-
mic).

Meso. = Foot rudiments (meso-
dermic).

M.Gang.C.=Multipolar ganglion
cell.

M.T. = Pallial tentacle.

Mth. = Mouth.

M.T.N. = Nerve of pallial ten-
tacle.

N.C. = Nuchal cavity.

N.R. = Nuchal roof.

N.R.V. = Veinlets from nuchal
roof.

Odont.C. = Odontophore cushion,

Oes.P. = Oesophageal pouch.

O.L. = Outer lip.

Op.N. = Optic nerve.

Os. = Osphradium.

Os.E. = Osphradial epithelium,

Osph. G. = Osphradial ganglion.
Of. = Otocyst.
Ot.N. = Nerve of otocyst.

P. = Pericardium.
Pal.Gl. = Pallial gland-tissue.
Pal.M. = Pallial muscle.

Pal.N. = Nerve in mantle tissue.

Pap. = Ridged papella or
“lieker.”

Pap.Cr.F. = Papille of internal
edge of transverse crop
folds.

P.C. = Posterior cartilage of
odontophore. =~

P.Cell. = Pigment cell of eye.

Ped. = Pedal nerve cord.

Ped. Anas. = Pedal anastomosis.

Ped.S. = Pedal sinus.

Pig. Bd, = Pigment band.

Pl. = Pleural ganglion.

P1.Ped.= Pleuro-pedal connective.

Post.A. = Posterior artery.

P.Pal.N.= Posterior pallial nerve.

P.p.b.c. = Post palatal buccal
cavity.

Py.Tr. = Prototroch.

P.-V.Lobe. = Perivisceral lobe of
right kidney.

P.Z. = Pigment zone of pigment
cell of eye.

R. = Rectum.

Rad, = Radula.

Rect. = Rectal evagination of
mid-gut.

R.G. = Rudimentary gill (cten-
idium).

RK. = Right kidney.

R.K.Ap. = External aperture;of
right kidney. a

R.Osph.G. = Right osphradial
ganglion.

R.R.P.= Renal aperture of reno-

fond

1

I’, Visceral loop.

pericardial canal of right Vac.H. = Vacuolated epithelium
kidney. beneath lateral streak.
R.S. = Radular sac. V.F.G. = Ventral fold of gullet.
S.C. = Sense cell of eye. V.G. = Visceral ganglion.
Sec.F.Cr. = Radial secondary Vii. = Visceral hump.

folds on the transverse crop
folds.

Sec.G. = Secondary or pallial gills.

S.L.P. = Sub-lingual pouch.

I’.£.C,.= Ventro-lateral cartilage
of the odontophore.
V.-l.F. = Ventro-lateral folds of

wall of rectum.

S.M. = Shell muscle. Vl. = Veinlet from mantle edge.
S.-R. Lobe. = Sub-rectal lobe of V.-Lobe. = Ventral lobé of right
right kidney. kidney.
S.R.M.= Sub-radular membrane. Vn. = Ventricle.
V.St. = Valve hindering flow of

St. = Stomach.

Si.Gr.= Groove in stomach-wall.

Sub-Hp.N.L. =
nervous layer.

Swb-Oes.C. =
connective of visceral loop.

Sup.-Int. =
ganglion of visceral loop.

Sub-epithelial
Sub-oesophageal

Supra - intestinal

secretion of hepato-pancreas ™-
into anterior part of gut.

X is the point of connection
between Int. 1 (Fig. 10a) and
Int. 2 (Fig. 100).

Y. is the point at which the

folds itself

stomach on

S.Z. = Sensory zone of mantle (Fig. 10a).
epithelium. Z. is the anastomosis between

Tf. = Tentacle.
T.N. = Tentacular nerve.

the posterior pallial nerves

of the two sides.

The terms right and left are always used to signify post-torsional

relations unless otherwise specified.

Extreme forms of shell of Patella vulgata. [rom

specimens collected in the Gouliot Caves, Sark.

Thick middle layer:

C Internal laminated layer.

A Thin brown external

(“)

(b) clear inner

outer part

<< illo:

Puate I.
Bie
Fig. 2. Section through the shell.
layer. B
traversed by branching canals ;
part.
Hig, 3.
Hig. 4.

- The animal in its shell—Ventral surface. H. = Head.

The
pigment layer of the mantle has been rubbed off

Dorsal view to show positions of organs.

Fig.

Fig.

Fig.

Fig.

72

and the nuchal roof removed. MW.7. = Pallial
tentacle. Os. = Osphradium. x 11.

The animal seen from the left anterior point of view

(diagrammatic and based upon published figures).
The nuchal roof has been removed. x 24.

Jiagrammatic sag sectlon—most o 1e blooc
Diagrammatic sagittal section—most of the blood

sinuses, also the kidneys and nerves are omitted
for the sake of clearness. 1.C. = Buccal cavity.
D.P. = Dorsal palatal plate. J.0. = Inner lip
(not in median section). O.L. = Outer lip.
N.R. = Nuchal roof. Oes.P, = Tissue of oeso-
phageal pouch. Pap. = Ridged papilla. S.L.P.
= Sublingual pouch. x 26.

A small portion of the ventral surface of the mantle

showing the circulatory arrangements. V/. =
Veinlet from mantle edge. G.V. = Veinlet from
pallial gill, x 10.

Puate Il.

8a. Diagrammatie transverse section of a very young

specimen to show the lateral streak, the arrange-
ment of fibres in the foot, the relations of the
kidney cavity, the structure of the mantle skirt,
x 85. D.Pal.Gl. = Dorsal pallial gland-tissue.
Hip.F’. = Epithelial fold. Fl. = Flap of Lateral
(** Epipodial’’) streak. Gl. = Gland tubules of
Lateral (‘‘ Epipodial ”’) streak. Lat.St. = Lateral
(** Epipodial”’) streak. Pal.Gl. = Pallial gland-
tissue. Pa!.M. = Pallial muscle. Pal.N. = Nerve
in mantle tissue. Ped.S. = Pedal sinus. Swb-
Ep.N.L. = Sub-epithelial nervous plexus. S.Z7.=
Sensory zone of mantle epithelium. Vac.H. =
Vacuolated epithelium beneath the lateral streak.

8). Epithelium of one of the folds of a pallial tentacle in

cross-section (after Haller). Highly magnified.

73

9a. Horizontal section of a pallial gill, x 30.

9b. Part of the sub-epithelial nerve plexus of a pallial

gill (after Haller), showing connection with sense
cell and multipolar ganglion cell (MW. Gang. C.).
Highly magnified.

Fig. 10. The alimentary canal of Patella. See description,

ar
ig. 13

#18;

fee

age)

b.

ig. 13e.

14
ig. 1d.

= Li:

SAT:

G

page 20. For. the sake of clearness this is
represented in two parts, the first three coils are
shown in Fig. 10@ and the third (Int. 1) connects
at x with the fourth (Int. 2) in Fig. 10b. Yin
Fig. 10a is the point at which the stomach folds
on itself. The coils of intestine are labelled
nib simte 2 eter Se Ik

Dorsal palatal plate. x 10.

Diagram showing the radular sac, ete. Rad. =
Radula. x 3.
Dorsal view of the odontophoral cartilages. x 3.

Ventral view of odontophoral cartilages and
muscles. See p. 31 for description. Some
muscles are supposed to have been removed.
Somewhat diagrammatic. x 3.

The inturned (median) surface of left half of odonto-
phore cushion. Some muscles have been removed.
x oe

One row of teeth from the radula. x 25.

Diagrammatic transverse section of the gullet to
show the positions, ete., of the ducts of the
buccal glands. x 15.

Diagram of the gullet. and buccal glands in a
specimen which had the glands unusually well
marked off from one another. x 3.

Section of a lobule of the hepato-pancreas (after
Haller). Highly magnified.

Surface of one of the transverse folds of the crop.
Pap.Cr.F.= Papille of internal edge of transverse

Fig.

Fig,

a
Fig.

Fig.

(2)

Fig. 23.

74

crop folds. Sec.F'.Cr. = Radial secondary folds
on the transverse crop folds. x 16.

19. The bend of the stomach (Y in fig. 10a) showing the

opening of the duet of the hepato-pancreas
(D.H.P.). St.Gr. = Groove in stomach wall.
V.St. = Valve hindering back flow of the secretion
of the hepato-pancreas. x 3.

20. Transverse section of ridges and groove in stomach

wall of a young specimen (after Haller). Highly
magnified.

Prate III.

Transverse section across the rectal papilla. x 20.
22a—-f. A series of transverse sections of gullet and crop

(the posterior faces of the sections are drawn)
showing the migration of the points of attach-
ment of the longitudinal folds (D.F.G. and
V.F.G.) through an angular distance of over 270°.
D.F.G. = Dorsal fold of gullet. V..G. = Ventral
fold of gullet. x 10.

Nervous system of Patella vulgata ;—Buc. = Buccal

commissure. Buc.G. = Buccal ganglion. Cer.=
Cerebral ganglion. Cer.C. = Cerebral commis-
sure. C.-Ped. = Cerebro-pedal connective. C.-Pl.
= Cerebro-pleural connective. Lab. = Labial
commissure. LL. Osph. G. = Left osphradial
ganglion. Op.N. = Optic nerve. Ot.N. = Otocyst
nerve. Ot. = Otocyst. Ped. = Pedal nerve-cord.
Pl = Pleural ganglion. Pl.- Ped. = Pleuro-pedal
connective. Ped.Anas. = Pedal anastomoses.
There are two such anastomoses. f.Osph.G@. =
Right osphradial ganglion. Suwbes.C. = Sub-
cesophageal connective of the visceral loop.
Supint. — Supra-intestinal ganglion of the
visceral loop. J.N. = Tentacular nerve. V. =
Visceral loop. V.G. = Visceral ganglion. x 2.

Fig.

24,

ig. 29.

530,

ie. ole

75

Dissection of mantle nerves in posterior region to
show the anastomosis (7) between the posterior
pallial nerves (P.Pal.N.) of the two sides. The
dissection is supposed to be made from the
ventral surface, and a small portion of the skin is
left to show the pallial tentacles (see also Fig. 8).
x 8. (After a drawing by J. T. Jenkins).

Longitudinal section of the cephalic tentacle. 7..V. =
Tentacular nerve. x 16.

Section of eye epithelium. x 800.

Pigment cell of eye epithelium. x 1,500.

. Sense cell of do. S.C.=Sense cell. P.C. = Pig-

ment cel!. P.Z.= Pigment zone. x 1,500.
Section of otocyst and nerve (after de Lacaze-
Duthiers). x 16.

. Epithelium of otocyst cavity.

Section of osphradium, ete. Osph.G. = Osphradial
ganglion. Os... = Osphradialepitheium. L.G.
= Rudimentary gill (ctenidium). x 180.

Pruate LY.
Heart of Patella. N.R.V. = Veins from the nuchal
root. Ant.Ado. = Anterior aorta. L.Au. =
Auricle (morphologically left). © Post. A. =

Posterior artery. Vn.= Ventricle. x 3.

Right kidney (after Lankester), d.-D.Lobe = Antero-
dorsal lobe. P.-V.Lobe = Peri-visceral lobe.
S.-R. Lobe = Sub-rectal lobe. V.Lobe = Ventral
lobe. x 14.

Diagrammatic transverse section showing the position
of the renal aperture (f£.£.P.) of the renoperi-
cardial canal of the right kidney. The anterior
face of the section is drawn. x 80,

. Diagram illustrating the structure of the kidney.

K.C. = Kidney cavity. x 300.

76

Fig. 32b. Epithelium of kidney. x 1,500.
Fig. 33a. A fragment of the ovary in section. x 300.

Figs. 35—49.

g. 33b. Section of young ovum. x 600.
ig. 34a, b, ¢.

Diagrams showing the development of sperms
in the testis (after Gibson). x ca. 350.

Figures of the development of Patella (after
Patten). Enlarged to various degrees. Fig, 36
is a ventral view. Figs. 37—40 are sections.
Figs. 41 and 42 are views of the oral (ventral)
side of the embryo. An.C. = Anal cells. <Ap.D.
= Apicaldise. Bp. = Blastopore. Cil.R. = Pre-
blastoporal ring becoming the prototroch. Het.=
Ketodetm. H.-M. = Endo-mesoderm. Mcp. =
Micropyle. Meso, or F'.-R. = Foot rudiments
(mesoblastic). Mesen. = Mesenteron (endoder-
mic). Pr.-Tr. = Prototroch. Hect. = Rectal
evagination of midgut. Rad. = Invagination to
form the radular sae.

C. TInnInG AND‘ Co., Printers, 58, Victoria STREET, LIvERPooL,

L.M.B.C. Menor X. Prats I.

SCY

HSF del - Fig.6 SB ith

PATELLA.

L.M.B.C. Menmorr X. Prats II.

2 Pal. Gl.

Pig Ba
Pali. Pal Gl}

Pig BY.
Pig. Bd 7 of as

HSF del S5 sith

PATELLA.

L.M.B.C. Memor X. Pruate IIT.

=
dd

PATELLA. Cm

L.M.B.C. Menor X. Pruate LV.

‘th ‘iN

KM cl

/

'
Meso.(F-R) @)

Ap. 0.

ich

Nh

PATELLA.

SMITHSONIAN INSTIT INSTITUTION

iii

= oe 8095

5