HARVARD UNIVERSITY

LIBRARY

OF THE

Museum of Comparative Zoology

L.M.B.C. MEMOIRS.

XVI.

CANCER.

NOTICE.

The 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 at the nett prices stated,
from Messrs. Williams and Norgate, 14, Henrietta Street,
Covent (iarden, London.

Memoir 1. Ascidia — published in October, 1899, 00 pp.

and live plates, price 2s.
,, II. Cardium — published in December, 1899,

92 pp., six plates and a map, price 2s. b'd.
,, III. Echinus — published in February, 1900,

30 pp. and five plates, price 2s.
,, IV. Codium — published in April, l'JUU, 20 pp.

and three plates, price Is. Ud.
'" Y. Alcyonium — published in January, 1901,

30 pp. and three plates, price Is. 6d.
„ VI. Lepeophtheirus and Lernsea — published in

March, 1901, 62 pp. and five plates,

price 2s.
,, \ll. Lineus — published in April, 1901, 4U pp. and

four plates, price 2s.
,, VIII. Pleuronectes — published in December, 1901,

26U pp. and eleven plates, price 7s.
,, IX. Chondrus — published in July, 19U2, 50 pp.

and seven plates, price 2s. 6d.
X. Patella— published in May, 1903, 84 pp. and

four plates, price 2s. 6d.
,, XI. Arenicola — published in March, 1904, 126

pp. and eight plates, price 4s. 6d.
,, XII. Gammarus — published in July, 1904. 55 pp.

and four plates, price 2s.
,, XIII. Anurida— published in October, 1906, 105 pp.

and seven plates, price 4s.
,, XIV. Ligia — published in January, 1907, 45 pp.

and four plates, price 2s.
,, XV. Antedon — published in June, 1907, 55 pp.

and seven plates, price 2s. 6d.
„ XVI. Cancer— published in June, 1908, 218 pp.

and thirteen plates, price 6s. 6d.

Xtverpool flDanne UBiologv Committee.
L.M.B.C. MEMOIRS

on Typical British Marine Plants &, animals

EDITED BY W. A. HERDMAN, D.Sc, F.R.S.

xvi.

CANCER

JOSEPH PEARSON, M.Sc,

Demonstrator in Zoology, University of Liverpool.

(With 13 Plates)
Price Six Shillings and Sixpence

LONDON

Williams & Norgate

June, iqoS

EDITOR'S PREFACE.

The Liverpool Marine Biology Committee was constituted
in 188-3, 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material, both published and unpublished, has been accu-
mulated. Twenty-one 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
Puffin Island Biological Station on the North Coast of
Anglesey, and later on, in 1892, moved to the more
commodious and accessible Station at Port Erin 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 since been
erected, and was opened for work in July, 1902.

In these twenty-one years' experience of a Biological
Station (five years at Puffin Island and sixteen at Port
Erin), where College students and young amateurs form a
large proportion of the workers, the want has been fre-
quently 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.

VI.

The objects of the Committee and of the workers at the
Biological Station were at first chiefly faunistic and
speciographic. The work must necessarily he so when
opening up a new district. Some of the workers have
published papers on morphological points, or on embry-
ology 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 still further to the
records of the Fauna and Flora. But the papers in the
present series, started in 1899, are quite distinct from these
previous publications in name, in treatment, and in pur-
pose. They are called " L.M.B.C. Memoirs/1 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
list shows those who have either performed or promised.

Memoirs from I. to XVI. have now been published.

Vll.

Pecten, by Mr. W. -I. Dakin; Eledone, by Dr. W. E.

Movie: and Doris, by Sir ('. Eliot, arc now far advanced
and ought to lie out during' 1908. It is hoped that
Cucuniaria, Puecinum, and the Oyster will follow soon.

Memoir I. Ascidia, W. A. Herdman, 60 pp., 5 Pis., 2s.
„ II. Oarditjm, J. Johnstone, 92 pp., 7 Pis., 2s. Gd.
,, III. Echinus, H. C. Chadwick, 36 pp., 5 Pis., 2s.
,, IV. Codium, P. -J. II. Gibson and Helen Auld,

26 pp., 3 Pis., Is. 6d.
,, V. Alcyoxium, S. J. Hickson, 30 pp., 3 Pis., ls.6d,
,, VI. Lepeophtheirus am) Lerx.ea, Andrew Scott,

62 pp., 5 Pis., 2s.
„ VII. Linetjs, R. C. Punnett, 40 pp., 4 Pis., 2s.
„ VIII. Plaice, F. J. Cole and J. Johnstone, 260 pp.,

11 Pis., Ts.
,, IX. Chondrtjs, (). V. Darbishire, -30 pp., 7 Pis.,
2s. 6d.
X. Patella, J. P. A. Davis and II. -I. Fleure,
84 pp., 4 Pis., 2s. 6d.
„ XL Arenicola, J. II. Ashworth, 126 pp., 8 Pis.,

4s. (id.
,, XII. Cjammarus, M. Cussans, 55 pp., 4 Pis., 2s.
,, XIII. Anurlda, A. D. Imms, 107 pp., 8 Pis., 4s.
„ XIV. Ligia, C, G. Hewitt, 43 pp., 4 Pis., 2s.
,, XV. Axtedox, II. C. Chadwick.
,, XVI. Cancer, J. Pearson.

Doris, Sir Charles Eliot.

Pecten, W. J. Dakin.

Eledone, W. E. Hoyle.

Cucumaria, E. II indie.

Oyster, W. A. Herdman and J. T. Jenkins.

( )stracod (Cytheru), Andrew Scott.

Buccinum, W. 13. Handles.

Vlll.

Bugtjla, Laura R. Thornely.
Zostera, II. J. Harvey Gibson.
Himanthalia, F. J. Lewis.
Diatoms, F. E. Weiss.
Fucus, J. B. Farmer.
Botrylloides, W. A. Herdman.
Actinia, J. A. Clubb.
IIydroid, E. T. Browne.
Haliciiondria and Sycon, A. Bendy .

In addition to these, other Memoirs will be arranged
for, on suitable types, such as Pagurus, Sagibta, Pontob-
della, a Cestode and a Pycnogonid.

As announced in the preface to Ascidia, a donation
from the late Mr. F. 11. Crossage, 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
Mrs. Holt, Sir John Brunner, and others, are regarded by
the Committee as a welcome encouragement, and have
been a great help in carrying on the work.

W. A. Herdman.

University of Liverpool,

Jane, 1908.

L.M.B.C. MEMOIRS

No. XVI. CANCER.

JOSEPH PEARSON, M.Sc,

Demonstrator in Zoology, University of Liverpool.

CONTENTS.

PAGE

Introduction 2

External Characters ... 6

Appendages 22

Endophragmal System . . 31

Structure of Integument . 45

Ecdysis 53

Autotomy 56

Muscular System ....

Coelom and Body Cavity

Alimentary Canal : —

General Description and
Histology .

Digestive Gland .

Ossicles of Fore-gut

Muscles of Pore-out

65

84

85

92

97

103

PAGE

Blood Vascular System . . \ 10
Respiratory System . . . 126
Excretory System .... 136
Nervous System .... 149

Sense Organs 156

Reproductive System . . .163

Development 169

Economics and Bionomics : —

General Habits .... 173
' The Crab Fisheries . . .177

Size of Crabs at Maturity 177

Distribution and

Migration 179

Bionomics of Ecdysis . . 183

Description of Plates . . . 195

Introduction.

Cancee is a genus which has a world-wide distribution.
Only one species, however, is found in Europe, viz., Cancer
pagurus, the subject of the present memoir.* This species
is found on almost every part of the coasts of Europe from
Norway to Greece, and it is particularly abundant on the
shores of North- West Europe (France, Germany and the
British Isles).

Cancer pagurus, the edible crab, has been chosen as
the subject of the present memoir partly on account of its
economic importance, and also because, as a type for
dissection, it is easily procurable and is of a convenient
size. The account given below, however, may be applied
with very few alterations to any of the brachyurous
Decapod Crustaceans, such as the common shore crab
(Carcinus) or the swimming crab (Portimus).

The edible crab is found in great abundance on the
coasts of the British Isles, especially on those parts which
are rocky, and gives rise to an important fishing industry.
The large crabs live in fairly deep water, but the young-
representatives of this species may be readily obtained
between tide-marks. Cancer is mainly carnivorous in its
habits and feeding. It is particularly fond of dead fish,
and it probably also feeds on other Crustaceans in a small
degree. There is, however, no evidence to show that it
has cannibalistic instincts. (For further particulars with
regard to habits, distribution, crab fishery, &c, see section
on Economics.)

Cancel' pagurus was first named by Lumens, who
established both the genus and the species. In his

* The investigation has been assisted by a grant of £25 from the
Board of Agriculture and Fisheries, and the expense of producing the
lithographed plates has; been met in part by a grant of £30 from the
" Treasury Grant for Research " of the University of Liverpool.

" Histoire naturelle des Crustaces," Milne-Edwards
named it Platycarcinus pagurus. This latter name appears
to have been retained in many continental works up to
quite recent years.

I have followed the classification of Borradaile,* and
T give below a table compiled from the results of his work.

Crustacea Decapoda

N;itanti;i Reptantia (sub-orders)

Palinura Astacura Anomura Brachyura (tribes)

r

iii.

Oxystomata Drorniacea Brachygnatha (sub-tribes)

Oxyrhyncha Brachyrhynca (super-families)

Corystidae Portunidae, etc. Cancridae (families)

Pirimelinae Cancrinae (sub-families)

r

Cancel* (genus)

The main characters of the various divisions of the
Decapod Crustacea are given below. t

Natantia.- — Rostrum well developed and compressed.
Body compressed. First abdominal somite equal
to rest. Stylocerite present. Second antenna!
scale large. In the legs basis and ischium
never fused, and one fixed point in the carpo-
propodal articulation. Male genital opening
arthrodial. Abdominal limbs 1-5 well developed
and used for swimming.

* Borradaile, L. A. " On the Classification of the Decapod
Crustaceans." Ann. and Man. Nat. Hist. (7), Vol. XIX, June, 1907.

f These characters are abstracted from Borradaile's paper.

Reptantia. — Rostrum reduced or absent, depressed if
present. Body depressed. First abdominal
somite smaller than rest. Stylocerite absent.
Second antennal scale reduced or absent. In the
legs generally a basi-ischium, and two fixed
points in the carpo-propodal articulation. Male
genital opening coxal and sternal. Abdominal
limbs 1-5 reduced or absent, and not used for
swimming.
The four tribes belonging to the Reptantia are
divided into two groups.

I. Third legs like the first. Abdomen macrurous.
Gnathobases of second maxillae narrow. Exopodites of
maxillipedes with lash directed forward. Grills numerous.

(1) Palintjra. — Carapace fused to epistoma. Rostrum

small or absent. Inner lobes of second
maxillae and first maxillipedes reduced. Body
depressed.

(2) Astacura. — Carapace free from the epistoma.

Rostrum large. Inner lobes of second maxillae
and first maxillipedes not reduced. Body sub-
cylindrical.

II. Third legs unlike the first, never chelate. Abdo-
men rarely macrurous. Gnathobases of second maxillae
broad. Exopodites of maxillipedes with lash directed
inward. Gills few.

(3) Anomura. — Carapace not fused with epistoma.

Last thoracic sternum free, its legs differing from
the others. Abdomen anomurous. Movable
antennal scale. Third maxillipedes narrow.

(4) Brachyura. — Carapace fused with epistoma. Last

thoracic sternum fused with rest, its legs like the
others. Abdomen brachyurous. No movable
antennal scale. Third maxillipedes broad.

The following are the sub-tribes of the Brachyura :
Ox vstom ata.— Mouth-Held prolonged forward as a

gutter. No female first abdominal limbs. Gills

few. Female openings sternal.
Dromiacea.— Mouth-field square. First abdominal

limbs present in female. Gills many. Female

openings coxal.
Brachygnatiia. — Mouth-field square. Female openings

sternal. No first abdominal limbs in female.

Gills few.

The Brachygnatha are divided into two super-
families.

Oxyrhyncha. — Front part of body narrow. Distinct
rostrum. Body more or less triangular. Orbits
incomplete.
Brachyrhyncha.— Front part of body broad. Rostrum
reduced or wanting. Body oval. Orbits com-
plete.
The Brachyrhyncha are sub-divided into fourteen
families. I give here the chief characters of the one
family — the Cancridae.

Cancridae. — Marine crabs with the branchial region
not greatly swollen. Carapace broadly oval or
hexagonal. Rostrum often wanting. Orbits com-
plete. Male openings coxal. Second antennal
flagella short. First antennae folded length-
wise. Inner lobe on the endopodite in the first
maxillipedes wanting. Legs generally nut
adapted for swimming.
The two sub-families of the Cancridae are as
follows : —

Pirimelinae.— Carapace hexagonal. Epistoma sunken.
Cancrinae. — Carapace broadly oval. Epistoma not
sunken.

EXTERNAL CI EARACTERS

(PL I, %s. 1, 2, 3).

The whole of the exterior of the animal is covered
by a thick continuous chitinous exoskeleton or shell,
which is highly calcified except between the movable
somites of the abdomen and between the movable podo-
nieres in the appendages.

The body may be conveniently divided into an
anterior region — the Cephalon, a middle region — the
Thorax, and a posterior region — the Abdomen. As in
all the Decapoda the Cephalon and Thorax are fused to
form the Cephalothorax.

The Cephalothorax is by far the largest portion of
the body, and is the only part seen from the dorsal
surface. The Abdomen is much reduced and is a flap-
like structure closely applied to the ventral region of
the cephalothorax between the bases of the walking legs.

There is every reason to believe that the crabs and
their relatives have arisen from primitive Crustaceans
having a body divided up into a number of movable
segments or somites. Extreme specialisation has taken
place, especially in the cephalothoracic region, and it is
in the Abdomen that one sees the nearest approach to
this primitive external segmentation. A careful exami-
nation, however, reveals the fact that there are five
somites in the cephalic region, eight in the thorax and
six in the abdomen — nineteen in all.

Before entering on a description of the more complex
parts it will be useful to examine the structure of a
typical abdominal somite.

The third abdominal somite of the female may be
taken as a type (see text fig. 1).

This somite is flattened dorso-ventrallv. On the

dorsal side there is thick and highly calcified exoskeleton,
but the ventral wall is membranous.

The dorsal wall consists of a median portion — the
tergum (text fig. 1, t.) — which is continued into two
broad lateral portions — the pleura. The median ventral
wall is known as the sternum, and from each of the outer
portions of the sternum arises an appendage. Between
the point of attachment of each appendage and the
pleuron the ventral wall is known as the epimeron.

■t.

Fig. 1. — Diagrammatic section through female abdomen.

t.= tergum. m.a. =muscles of abdominal

p. = pleuron. appendage.

s. = sternum. h.g. =hind gut.

e. = epimeron. pr. =protopodite.

e.m. = extensor muscles of abdomen, ex. = exopodite.
/. m. = flexor muscles of abdomen. en. = endopodite.

The segment is connected with the two neighbouring
segments by a thin uncalcified part of the exoskeleton —
the arthrodial membrane. This allows of free movement
between the segments. Each segment articulates with
the one in front by means of a pair of hinges placed at
the outer and anterior part of the pleuron at each side.

The appendages will be described in detail later.

Cephalothorax.

1. Carapace.

The terga and pleura of the cephalothorax are fused
to form the large Carapace. This is a broad shield the
width of which is about 1^ times as great as the length.

8

Instead of taking an even sweep downwards the carapace
passes outwards almost horizontally and then suddenly
bends inwards and passes down to the bases of the walking
legs. An examination of a rough section of the animal
will show that at the base of the legs the carapace turns
suddenly upwards and is continuous with the membranous
wall of the spacious Branchial Chamber (PI. IV, fig. 56,
br. ch.). The space between the ventral part of the
carapace and the base of the legs is so very small, and
moreover is so well guarded by long setae that no water
can enter the branchial chamber at this border, as is the
case in the Macrura. There are, however, two openings
into the branchial chamber — the small posterior inhalent
branchial aperture, above the coxopodite of the last
pereiopod, and the larger anterior inhalent branchial
aperture, situated immediately in front of the coxopodite
of the chela. The ventral part of the carapace turns
forward in front of the latter opening, and passing around
the mouth it fuses with the pre-oral cephalic sterna. The
portion of the carapace which passes around the mouth is
turned inwards at each side to form a chamber which lies
immediately in front of the Branchial Chamber. This
may be called the Pre branchial Chamber. Its roof is
membranous and is fused on its inner side with the
epistoma and with the endopleurites of the two post-oral
cephalic somites, and probably represents the epimera of
the third, fourth and fifth cephalic somites. The Pre-
branchial Chamber will be described in detail in the
seel ion on Respiration.

Both the dorsal and the inflected portions of the
lateral region of the carapace were designated the
" branchiostegite " by Milne-Edwards because they
enclose the branchial cavity.

Anteriorly the dorsal surface of the carapace is

9

bounded by a median portion between the orbits and two
lateral portions. Similarly the posterior boundary
consists of a median portion and two lateral portions. So
that we may speak of these borders as the anterior, antero-
lateral, posterior and postero-lateral respectively.

The Anterior Border is situated between the orbits.
The rostrum, which occupies the median portion of this
region, consists of a median and two lateral lobes. It is
continued ventrally as a median plate which separates the
two cavities in which are lodged the eye peduncles. Each
of the lateral lobes of the rostrum passes downwards as
the supraciliary lobe, which fuses with the anterior and
inner region of the second antenna (PI. Ill, fig. 20, S.I.).
Passing outwards from the rostrum the anterior border of
the carapace divides at each side into the supra-orbital
and infra-orbital portions which form the boundary of the
orbit. On its inner side the supra-orbital edge has the
prominent supra-orbital lobe which is close to the lateral
lobe of the rostrum. The inner boundary of the orbit is
fused with the outer portion of the second antenna.

The Antero-lateral Borders form an arc of a circle the
centre of which is at the junction of the two outer grooves
bounding the epibranchial region of the carapace (see
below). Each antero-lateral border is divided up into
nine lobes by well-defined ridges. There is no definite
distinction between the antero-lateral border and the
postero-lateral border, but the latter may be said to
commence at the posterior end of the ninth lobe. There
is also a feebly marked lobe on the outer portion of the
postero-lateral ridge.

The Postero-lateral Border passes backwards and
inwards. This border is well rounded and not so. clearly
defined as the anterior and antero-lateral borders.

Immediately in front of this border there is the

10

postero-lateral viclge, which is continuous mi its outer side
with the antero-lateral border. At its outer edge it is
coincident with the postero-lateral border, but as it passes
inwards it becomes quite distinct from the latter and
dies away near the median line in front of the posterior
border.

The Posterior Border of the carapace is horizontal,
and is continuous behind with the tergum of the first
abdominal segment.

Areas of the Carapace. (Text fig. 2.)

The dorsal surface of the carapace is divided up by
means of small depressions into areas.

The Cervical groove (C. gr.) separates the cephalic
region of the carapace from the thoracic region. This
groove is seen as a transverse median depression a little
more than half way down the carapace. The width of this
median groove is almost equal to the distance between the
two supra-orbital lobes.

At each of its outer edges the median groove is
continuous with a well-marked depression which
commences at the posterior end of the fifth lobe of the
antero-lateral border. This depression is curved, the
convexity being in front. The median groove and its
two lateral extensions together form the cervical groove.

The Cephalic portion of the carapace is divided into
the Facial and Gastric regions.

The Facial region is separated from the rest of the
cephalon by a faint transverse depression near the front
of the carapace. The outer ends of the depression bend
forward and terminate on the second lobe of the antero-
lateral border. This region is divided into a median
Frontal region (Fr.) and two lateral Orbital regions (Orb.).

The Gastric region is bounded behind by the

11

Cervical groove, and is composed of a median triangular
portion having the apex pointing backwards and two
lateral portions which end at the antero-lateral border.
The median portion is divided into two anterior Proto-
gastric regions (Pg-), a median anterior Mesogastric
region (Mg.), a pair of posterior Metagastric regions
(Mfg.), and a median posterior Urogastric region

Fig. 2. — Areas of the Carapace.

Fr. = Frontal region.
Orb. = Orbital
Hep. = Hepatic
Pg. = Protogastric
Mg. = Mesogastric
Mtg.= Metagastric
Ug. = Urogastric

Mb. =Mesobranchial region.
Mtb. = Metabranchial , ,
Eb. =Epibranchial ,,

Card. = Gardiac ,,

C.gr. = Cervical groove.
B.gr. — Branchio cardiac groove.

(Ug.). The lateral portions are known as the Hepatic
regions (Hep.). Each extends outwards to the antero-
lateral border and is bounded behind by the outer part
of the cervical groove.

The Thoracic portion comprises that part of the
carapace which lies behind the cervical groove. It is

12

divided by two longitudinal grooves the Branchio-
cardiac grooves {li.tjr.) — into a median Cardiac region
[Card.) and two lateral branchial regions. Each branchial
region is made up of an anterior Mesobranchial region
[Mb.) and a posterior Metabranchial region (Mtb.) and a
small inner Epibranchial region (Eb.).

The ventral inflected portion of the carapace is
divided into two parts by a well denned groove, which
may be termed the pleural groove, as it probably marks
the separation between the cephalo-thoracic terga and
pleura. It is along this groove that the carapace splits
during ecdysis (see section on Ecdysis). The pleural
groove commences at the epistoma and passes outwards
and slightly backwards until it almost reaches the
posterior end of the seventh lobe of the antero-lateral
border. Here it turns backward and runs parallel to the
postero -lateral ridge, finally reaching the posterior border
with which it becomes continuous. Thus the pleural
groove divides the inflected portion of the carapace into
an outer, or Sub-hepatic region, and an inner, or Sub-
branchial region. The sub-hepatic region may be
considered as an inflected portion of the tergum, and the
sub-branchial as belonging to the pleural region. Milne-
Edwards regarded the sub-branchial region as part of the
cephalo-thoracic epimera, but the inner walls of the
branchial chambers undoubtedly represent the epimera.

2. Pre-oral Cephalo-thoracic Sterna.
(PI. Ill, figs. 19, 20.)

-Yentrally the median lobe of the rostrum passes
backwards as a triangular plate, the apex of which points
posteriorly. This plate, which is separated at its posterior
end from the first sternum [antennulary sternum] (S1)

13

by a well-defined suture, forms a septum* between the
articular cavities of the two optic peduncles.

From the dorsal side of the sternal region the septum
he! ween the above-mentioned articular cavities is short
and broad. On a level with the posterior end of these
cavities there is a well-marked suture separating the
septum from the first sternum.

Immediately in front of the dorsal side of the
ophthalmic articular cavities (o.m.c.) are two short
calcareous plates near the median line, which stretch
across to the roof of the carapace. These are the
Procephalic -processes (p.c.p.) to which are attached the
anterior gastric muscles.

The First Sternum (Sl) lies in the segment of the first
antennae (cmtIc.) and separates the articular cavities of
these appendages. Owing to the depth of the sternum in
this region its relationship to the articular cavities is best
seen from the dorsal side of the sternum.

It consists of a median piece lying between the
articular cavities of the antennules, and of two lateral
expansions which form the posterior boundaries of the
articular cavities. Viewed laterally the sternum is seen
to have a comparatively great depth. About half way
down the anterior face of this sternum is a concavity into
which fits a process from the septum between the
ophthalmic articular cavities.

From the ventral side the first sternum bounds the
posterior and inner sides of the sockets of the antennules,
and the lateral prolongations extend as far as the bases of
the second antennae.

* In the present Memoir the optic peduncles are not regarded
as modified appendages, and I shall not regard the region of the body
from which the eyes arise as the first segment, nor shall I speak of
the septum between the articular cavities of the optic peduncles as
the first sternum.

14

Huxley* in his short account of the crab states that
the ophthalmic and antennulary sterna are fused, and
that the suture is between the fused sterna and the
rostrum. I am of the opinion that the whole of the
sternum behind the suture belongs to the antennulary
somite alone, and that the septum separating- the
articular cavities of the optic peduncles is a posterior
prolongation of the rostrum, as described above.
An examination of the sternum from the dorsal side
(PI. Ill, fig. 19) shows that the suture is on a level with
the posterior boundary of the ophthalmic articular cavity;
so that, if Huxley's view be accepted, we have the
ophthalmic sternum entirely behind the articular cavity
of its own somite ! It is much more reasonable to
conclude that the suture separates' the ophthalmic septum
from the antennulary sternum.

The Epistoma is a broad plate in front of the mouth
and immediately behind the first sternum. It represents
the united sterna of the second (antennary) (S'2) and third
(mandibular) ($3) somites. Its anterior border is concave
in front. The two lateral borders gradually slope inwards
towards their posterior ends. The posterior border is
deeply concave behind and bounds the front edge of the
mouth.

The middle part of the anterior border touches the
posterior edge of the first sternum and the two outer
portions bound the posterior edge of the second antenna.
The lateral borders are in contact with the membranous
roof of the pre-branchial chamber.

From the middle of the anterior border of the ventral
side of the epistoma a median groove passes backwards
but does not extend as far as the posterior border. From

* T. H. Huxley, Manual of the Anatomy of Invertebrated Animals,
1877, pp. 340-345. '

15

the posterior edge of this groove a slight depression passes
outwards at each side parallel to the anterior border.
Tli is depression probably marks the boundary between the
antenna ry and mandibular sterna. This groove is better
defined on the dorsal side of the epistoma.

The Labrum (PL III, fig. 20, tab.) is a soft fleshy
lobe attached to the middle region of the posterior border
of the epistoma. It is surrounded near the middle by a
calcareous ring which gives off a median posterior
prolongation. At each side of this median plate is a
soft fold.

•r>. Post-oral C e p h a 1 o - t h o r a c i c Sterna
(PL I, figs. 2, 3, Text fig. 3).

These are all fused together as a single oval-shaped
plate situated between the bases of the paired post-oral
cephalothoracic appendages. Transverse grooves are
present which mark the division of this region into
segments or somites, and which mark the places at which
the sterna grow inwards to form the endosternites of the
endophragmal system.

The surface of the fused sterna is concave laterally
in order to accommodate the abdomen, which is always in
a flexed condition. This concavity is especially well
marked in the males. The surface of the sterna is, how-
ever, convex antero-posteriorly.

On the sternum of fifth thoracic somite are two small
tubercles (P.) which fit into two concavities on the
abdomen and thus form an effective locking apparatus
which keeps the abdomen in position. These are
especially large in the males.

The sternum of the sixth thoracic somite of the
female bears a pair of large openings which are the
external genital openings.

16

The sterna of the last four thoracic somites are
characterised by a median groove which marks the place
at which the " Median plate " of the endophragmal system
has grown inwards. The anterior end of this median
groove is marked by a very deep depression which is
situated at the posterior end of the fourth thoracic
sternum.

At the outer and posterior corners of each
sternum are backwardly directed areas — the " episterna *'

Fig. 3. — Ventral view of post-oral
cephalothoracic sterna (male).

F.S. = Fused sterna of the two
post-oral cephalic and the first
three thoracic somites. 4.ts. — 8.ts.
= Sterna of the 4th to 8th thoracic
somites. Ep.s. = Episternum.
P. = Sternal papilla of the
abdominal locking apparatus.

(Ej)s.) which run for a short distance alongside the
following sternum, from which they are separated by
distinct sutures. Between each episternum and the
corresponding sternum is a slight groove which is not very
distinct in the edible crab. In Portunus and other crabs,
however, this groove is much more distinct so as to
suggest a complete separation between the sternum and
episternum. This probably explains why Brooks* states

Brooks, Handbook of Invertebrate Zoology,

17

that the episternum is (interior to the outer end of its own
sternum. He has evidently mistaken the groove mentioned
above for a true suture, and has therefore concluded that
the episternum belongs to the following sternum. The
last thoracic sternum has no episterna.

At the anterior end of each episternum is a small
concavity into which fits the ventral hinge of the
coxopodite of the appendage of that somite.

The sterna of the last two cephalic and the first four
thoracic somites (F.S.) are fused together, without any
sign of separation into distinct segments as in the posterior
region of the thorax. t There is, however, a slight evidence
of a division in front of the fourth sternum of the thorax.

That portion of the thoracic sterna which is covered
by the abdomen is characterised by the absence of setae.
There are long setae along the outer edges of the
episterna, and also on that portion of the fused sterna
belonging to the two last cephalic segments and the first
four thoracic segments.

In front of the fourth thoracic sternum the outer
edges of the sterna are turned up vertically.

The first post-oral cephalic sternum has two lateral
processes which project forwards and give support to the
Metastoma. The metastoma is a fleshy lobe forming the
posterior lip of the mouth.

4. Cephalic epimera.

In the first two cephalic somites it is difficult to
identify the epimera, but the latter are probably repre-
sented by the region between the outer portion of the
articular cavities of these somites and the carapace.

f The two post-oral cephalic sterna, which are represented by two
narrow bars at the extreme anterior end of the fused post-oral sterna,
are separated from each other, however, by transverse sutures.

18

The epimera of the third (mandibular) and the two
last cephalic somites (maxillary) are probably represented
by the membranous roof of the pre-branchial chamber at
each side (PL III, fig. 18, r.br.). This is continuous
behind with the thoracic epimera.

5. Thoracic epimera
(Plate III, tig. 18, epm. 6-12).

The thoracic epimera are represented by a continuous
plate at each side forming the inner wall of the branchial
chamber. This is the " flanc " of Milne-Edwards. The
lower border of the epimera commences immediately
above the base of the thoracic appendages. They pass
upwards and inwards and are continuous above with the
membranous roof of th^ branchial chamber. At the
posterior end they extend upwards almost to the carapace,
from which they are only separated by short muscles which
pass from the summit of the epimera to the carapace. At
the anterior end the epimera are much shallower and
become continuous with the roof of the branchial chamber
some distance below the carapace.

The fused thoracic epimera form an extremely
convex wall Avhich is divided up into segments by vertical
sutures, which correspond to the lines of separation
between the various somites of the thorax. In this way
the epimera are divided up into seven portions. The
epimera of the first and second thoracic somites are
completely fused, and there is no groove separating them,
but apart from this there is one segment of the fused
epimera for each of the remaining thoracic somites. The
epimeron of the fourth somite is particularly broad. That
of the last thoracic somite is not bounded posteriorly by

1!)

a groove, but is continuous with the sternum of the same
segment.

In their natural position the gills lie upon the
thoracic epimera.

The Abdomen.

The abdomen is continuous with the posterior part
of the cephalothorax. The connection is effected by means
of an arthrodial membrane, which allows of considerable
movement between the two regions. The abdomen is
small and in its natural position is closely applied to the
sternal region of the thorax. This region differs in the
two sexes, being much broader in the female than in the
male. This character provides a useful and ready method
of distinguishing between the two sexes. There are other
differences which require a more detailed examination.

Female.
(PI. I, fig 2, PI. IV, fig. 32, PL V, fig. 34.^

This consists of six somites and the telson, all of
which are freely movable. When lying in position it
extends as far forward as the posterior end of the sternum
of the third thoracic somite. The locking arrangement
for keeping the abdomen closely applied to the thoracic
sternum is not so well developed as in the male. It
consists of two extremely small tubercles on the fifth
thoracic sternum which fit into two slight depressions at
the postero-lateral corners of the ventral side of the sixth
abdominal somite. The total length of the abdomen is
2\ times as much as its greatest width.

There are four pairs of appendages, one pair being
borne on the second and on each of the three following
somites, respectively.

20

On the dorsal side of the abdomen the terga are
separated from the pleura by two longitudinal grooves.
Ventrally the median portion is covered by a thin
uncalcified cuticle and the outline of the hind-gut is
clearly seen.

The hind-gut opens on the ventral side of the telson
at the anus.

Only the first somite requires any special comment.
The upper side of this somite is prolonged forward as a
thin triangular flap, the apex of which points anteriorly.
This triangular portion is covered by the carapace, only a
narrow region at the posterior end of the somite being
exposed. All around the anterior edge of this somite is a
thin membrane which is continuous with the posterior
region of the cephalothorax.

The first two somites are hollowed out laterally to
provide free movement for the last pair of thoracic legs.

Male (PL I, fig. 3).

When lying in position the abdomen extends slightly
in front of the middle of the fourth thoracic sternum.
It is slightly shorter than the female abdomen and much
narrower. As in the latter, the sides of the first two
somites (and also part of the third) are hollowed out and
pass round the inner sides of the last pair of thoracic
appendages.

The first somite has the same arrangement as in the
female. The anterior part of the dorsal side is triangular
and is covered by the carapace.

The third, fourth and fifth somites are fused together
so that there is absolutely no movement between them.
The sutures marking their separation still persist.

There are only two pairs of appendages which are
present on the first and second somites respectively.

21

These appendages are peculiarly modified to act as
eopulatory organs (see sections on Appendages and Repro-
ductive System).

The male abdomen is much more closely applied to
the thorax than is the case in the female. This is partly
due to the small number of appendages and also to the
very effective locking apparatus. The latter is similar
to that described in the female and the position of the
parts is the same, but the tubercles on the thoracic
sternum are much larger, as is also the case with the
concavities on the sixth abdominal somite.

Below are e-iven the measurements of the abdomen
of a male and female, both having a carapace breadth of
23-5 cm.

Female.

Male.

No. of

Somite.

Greatest

Greatest

Greatest

( rreatest

length.

width.

length.

width.

mm.

mm.

mm .

mm .

1

17 25

17

22

2

8 22

8

17

3

7 30

7

23

4

8 32

8

20

5

10 35

9

17

(')

20 35

13

16

Telson

18 21

13

13

Total length, 88 mm. '

Total length, 75 mm.

External apertures.

The external openings are as follows: —
The Mouth — a median aperture on the ventral side
of the cephalic region between the mandibles.

22

The Anus — a median aperture on the ventral side of
the telson.

The Excretory Openings one pair. These are
situated at the base of the second antennae on the ventral
side. Each is covered by an operculum.

The Female Reproductive Openings — one pair. These
are two large apertures on the sternum of the sixth
thoracic somite.

The Male Reproductive Openings — one pair. These
are situated on the ventral side of the coxopodites of the
last pair of thoracic appendages.

Appendages (Plate II).
There are five pairs of appendages on the head and
eight pairs on the thorax. There are four pairs of
abdominal appendages in the female and only two pairs
in the male. The appendages are as follows:

Somite

C epha Ion.

T hor a x.

I. — 1st Antennae ( Antennules) .

II. — 2nd Antennae.
III.— Mandibles.
IV.— 1st Maxillae.
V.— "2nd Maxillae.

Somite VI. — 1st Maxillipedes.
„ VII. — 2nd Maxillipedes.

VIII. — 3rd Maxillipedes.
IX. — 1st Pereiopods.
„ X. — 2nd Pereiopods.

„ XL — 3rd Pereiopods.

„ XII. — 4th Pereiopods.

„ XIII. — 5th Pereiopods.

A b d o men.

Female.

Semite XIV. — Absent.

XV. — 1st Pleopods.
XVI. — 2nd Pleopods.
XVII.— 3rd Pleopods.
XVIII.— 1th Pleopods.
XIX.— Absent.

Male.

1st Pleopods.

2nd Pleopods.

Absent.

Absent.

Absent.

Absent.

•2:5

The First Antenna or Antennule (PI. II, fig. 4, PI.
Ill, fig. 20) is situated in a deep depression on the ventral
side of the cephalic sternum (s. a.1). This depression or
socket is bounded in front by the rostrum (rost.), and
behind by the lateral expansion of the first sternum ($] .
The outer boundary is formed by the inner edge of the
second antenna (ant.), and the inner boundary by the
median portion of the first sternum. The appendage
consists of a broad basal joint, from which is given oft on
its inner side a two-jointed portion. These three pieces
together form the protopodite (prot.). From the end of
the distal segment of the protopodite arise two many-
jointed flagella — an inner endopudite [end.) and an outer
exopodite (ex.). The exopodite is the larger of the two,
and bears on its inner side a tuft of long setae. The
" olfactory " setae are small setae on the ventral side of
the exopodite (see section on Sense Organs).

( )n the dorsal side of the basal segment of the
protopodite is a longitudinal groove covered with long
setae. This groove marks the place where the auditory
sac opens to the exterior in the young animal. In the
adult crab this groove is completely closed, although it
remains open a short time after eedysis.

In their natural position the three parts of thr
protopodite are folded on one another. The second
segment is closely applied to the inner side of the basal
segment. The third segment is bent back along the
dorsal side of the second, and its distal end lies in an
excavation made for its reception in the dorsal wall of the
basal segment.

Second Antenna (PI. II, fig. 3, PI. Ill, fig. 20). This
consists of a large basal portion (prot.) which is fused to
the carapace, and a distal flagellum, which consists of
two long basal segments and a number of short rings,

24

arising from the anterior and inner region of the basal
portion. At the posterior and outer corner of the large
basal segment is the operculum (op.), which covers the
external excretory opening. This operculum probably
represents the coxopodite and the larger basal portion the
basipodite, the two together forming the protopodite. The
flagellum probably represents the endopodite.

The outer edge of the basipodite is fused to the sub-
hepatic region of the carapace. Backward processes from
the supraciliary (S.l.) and supra-orbital lobes fuse with
the anterior end of the basipodite. The inner and
posterior corner of this segment is in contact with the
lateral portion of the first sternum, and the posterior
border of the same segment is in contact with the
epistoma.

The Mandible (PI. II, fig. 6) lies at the side of the
mouth. The main portion is an elongated strongly
calcified structure which is divided into two parts — an
inner part, which projects over the ventral region of the
mouth, and acts as the " jaw," and an outer part, the
apophysis (apoph.), to which are attached the tendons of
the mandibular muscles. At the outer extremity is the
tendon of the external adductor (t.ex.ad.). Behind this,
attached to a small projection, is the tendon of the
external abductor (t.ex.ab.). To the posterior and inner
side of the apophysis is attached the tendon of the internal
adductor (t.int.ad.). The internal abductor arises from
the apophysis on the inner side of the base of the tendon
of the external adductor. There is no tendon for the
internal abductor. Anteriorly the mandibular palp (md.
palp.) arises from the inner side of the apophysis. The
mandible is hinged to the epistoma by means of a small
projection below the palp. There is no definite hinge
posteriorly, but the posterior border of the inner region

25

of the apophysis is attached to the metastoma by means
of a somewhat flexible membrane.

The First Maxilla (PL II, tig-. 7, PL IV, fig. 26), which
arises immediately behind the mandible, is small and is
made up of a protopodite and endopodite. The exopodite
is absent. The protopodite is on the inner side and is
composed of two distinct pieces — a narrow proximal
coxopodite (C.) and a larger basipodite (B.) which is
external to the coxopodite. The endopodite (end.) arises
from the onter side of the basipodite, and consists of a
broad proximal leaf-like region and a narrower distal
region. From the distal extremity of both parts of the
protopodite arise fairly strong setae.

The Second Maxilla (PL II, tig. 8, PL IV, tig. 27)
consists of an inner protopodite, a median endopodite
(end.) and an outer e.vopodite (Scaph.). The protopodite
is composed of a coxopodite (C.) and a basipodite (B.),
each of which is bilobed. The two lobes of the coxopodite
are long and slender, and are clearly separated from one
another. Those of the basipodite are broader, and the
separation between the two lobes is only partial. On the
outer side of the basipodite is the small endopodite, which
ends in a long narrow process. On the outer side of the
endopodite and arising from the basipodite, is the large
modified exopodite which is known as the scaphognaihite
(Scaph.). This is a broad plate of irregular shape which
lies in the pre-branchial chamber. By means of its rapid
and complicated movement it bales the water out of the
branchial chamber.

In the First Maxillipede (PL II, fig. 9) the protopodite
is on the inner side. The coxopodite (C.) is small and
richly clothed with setae, and the basipodite (B.) is a long
lamella having two rows ot setae on its outer edge. The
endopodite (end.) is between the exopodite and the proto-

26

podite. It is membranous, the proximal half being
flattened laterally and the distal half dorso-ventrally.
The exopodite (ex.) is long and slender and consists of a
long proximal segment, which is as long as the endo-
podite, and a distal many-jointed fiagellum (flag-)
which in its natural position projects inwards at right
angles to the proximal segment. During life this
Hagellum is exceedingly active. From the outer side of
the protopodite arises the long flabellum (flab.) (or
epipodite). This is a long narrow membranous plate which
passes back into the branchial chamber above the gills.
The proximal portion of the flabellum is broad and leaf-
like.

The Second Maxillipede (PL II, fig. 10) has the
exopodite and flabellum in the same position as in the
previous appendage. The flabellum (flab.), however, is
much shorter than that of the first maxillipede, and lies
along the upper portion of the thoracic epimera and below
the gills. The protopodite is much reduced, but the
proximal coxopodite* (C.) and the distal basipodite (B.)
can still be made out. The endopodite is comparatively
larger than the same part in the first maxillipede. It
arises from the basipodite and is divided into five
movable segments. The first or proximal segment— the
ischinjtodite (I.) is small. The second segment or
meropodite (M.) is the longest, and equal in length to the
other four segments. The three distal segments are
small, and between the second and third segment the
endopodite turns inwards, the distal segments being at
right angles to the meropodite. The names of the third,
fourth and fifth segments are carpopodite (C.1), pvopo-

* The following abbreviations are sometimes used : — coxa =
coxopodite; basis = basipodite; ischium = ischiopodite ; meros =
meropodite; carpos = carpopodite; propos = propodite ; dactylos =
dactylopodite.

27

dite (P.) and dactylopodite (D.) respectively. Arising
from the appendage immediately in front of the base of
the nabellum is a podobranch (pod. br.) (see section on
Respiratory Organs).

The Third Maxillipede (PI. II, fig. 11, PL IV, fig. 30)
is built on a similar plan to the previous appendage. The
basis and the ischium are fused together to form the
basi-ischium (B.-I.). The podobrancli (pod. br.) is very
small and arises from the coxopodite. The nabellum
(flab.) lies on the lower part of the thoracic epimera below
the gills. The endopodite and exopodite (ex.) are closely
applied together and are much flattened so as to form
with the same appendage of the other side an effective
operculum closing over the remaining mouth parts, and
preventing the exhalent current of water from the
branchial chamber from passing out except in front of the
scaphognathite.

The mandibles, maxillae and maxillipedes all lie
around the month in the large depression between the
anterior parts of the sub-branchial regions of the
carapace. The ventral side of the third maxillipedes is
on a level with the sub-branchial region. As the
nabellum of this appendage passes back into the branchial
cavity, it passes along the front of the anterior inhalent
branchial aperture, and reduces the size of the aperture
considerably. At this point the nabellum* is also richly
clothed with strong setae, which probably act as a
"strainer" in conjunction with the setae present on the
front part of the coxa of the chela (see section on
Respiratory Organs).

The First Pereiopod (or chela) (PL II, fig. 12, PL III,

* The coxopodite of the third maxillipede is prolonged outwards,
and bounds the inner part of the inhalent aperture. The nabellum
bounds the outer part. Both are richly clothed with setae on their
posterior faces.

28

fig. 21) is the largest appendage in the body. It consists
of seven segments (or podomeres). A comparison with
the third maxillipede indicates that the two proximal
segments belong to the protopodite, and the remaining
five to the endopodite. There is no exopodite present.
The seven segments have the same names as the similar
parts in the third maxillipede. With the exception of
the second and third segments, which are fused together
to form the basi-isckium (B-l.), all the parts are freely
movable. The basi-ischium lias a thin groove running
around it, which marks the separation of this fused
portion into its two constituent parts. This groove is
known as the fracture plane because it is at this point that
the animal fractures the limb during the process of self-
amputation (see section on Autotomy). The two distal
segments of the limb are slightly modified to form the
pincer which constitutes an effective prehensile organ.

Each of the movable segments swings in a different
plane, so that the combined movement of the whole
appendage is a very complete one. The coxopodite (C)
articulates with the body by means of two hinges, one
being dorsal (PI. III., fig. 21, d.) and the other ventral
(v.). The dorsal hinge is attached to the antero-ventral
corner of the epimeron of the fourth thoracic somite, and
the ventral hinge articulates at the postero-lateral corner
of the sternum of the same segment. Thus the motion of
the coxopodite is in a horizontal plane, moving backward
and forward. The fused basi-ischium (B.-I.) articulates
with the coxa by an antero-dorsal (d1) and a postero-
ventral hinge (v1), and the movement is upwards and
downwards in a plane making an angle of about 45° with
the vertical. The meros (-1/.) has very little movement.
Its two hinges are antero-dorsal (d2) and postero-ventral
respectively (r2), and the small degree of movement of

29

which this segment is capable is almost in a vertical
plane. The two hinges of the carpos (C.1) are situated
dorsal ly (d3) and vent rally (r!), and the segment moves
forward and downward. The propodite (P.) has two
hinges dorsal (<IX) and ventral (vi). The former is
external to the latter, and the segment moves forward
and slightly upward. In the dactylos (/>.) the hinges are
horizontal and the segment swings in a vertical plane.

The dimensions of the various segments of the chela
in a female crab (carapace breadth 23-5 cm.) are as
follows : —

Anterior length.

Posterior length.

Coxopodite . . .

7 mm.

11 mm.

Basi-ischiopodite

... 17 „

6 „

Meropodite ...

... 19 „

... 32 „

Carpopodite . . .

... io „

... 30 „

Propodite

... 50 „

... 20 „

Dactylopodite

'20 „

... 30 „

The dorsal sides of the basi-ischium and of the meros
are flattened so that they can be closely applied to the
anterior portion of the sub-branchial and sub- hepatic
regions of the carapace, and in these places setae are
absent from the carapace.

Between the meros and the earpos the limb is capable
of bending on itself, so that the anterior borders of the
propodite and the carpos become closely applied to the
anterior borders of the basi-ischium and the meros.

On the dorsal side of the basi-ischium and meros
there are irregular grooves. These are the lines of
absorption (PI. II, fig. 12, abs.) (see section on Ecdysis).

Pereiopods 2-5 (PI. II, fig. 13). These are known as
the " walking legs." Their essential structure is the
same as that of the chela. The one obvious difference is
that in all the walking legs the propodite has not an

30

outgrowth which, in conjunction with the dactylos, forms
a pincer. In other words, the walking legs terminate in
a single claw, and are not chelate.

The three terminal segments are capable of being
flexed upon the proximal segments. This flexion is in a
vertical plane.

Pleopods (Female) (PI. II, fig. IT). There are four
pairs of appendages on the female abdomen, one pair
being situated on each of the second, third, fourth and
fifth somites respectively. They are all similar in
structure. Each pleopod is attached to the abdomen by
a basal piece — the pvotopodvte (jjrot.). From this arise
two long pieces — an outer exopodite (ex.) and an inner
endopodite (end.). The exopodite is almost cylindrical
in section, and about half as long as the abdomen.

From the outer and inner edges of the exopodite
rows of setae arise. Each seta has short fine branches
given off from each side of the central stem. The endo-
podite is about as long as the exopodite. About one-third
of its length from the base is a well-defined transverse
groove. The setae are arranged, as in the exopodite,
along the outer and inner edges, but they arise in small
bundles. The setae are very long and do not bear off-
shoots except near the tip, where there are a few very
fine short branches. The eggs are attached to the
endopoditic setae.

Pleopods (Male) (PL II, figs. 14, 15, 16). There are
two pairs of abdominal appendages in the male, which are
situated on the first two somites. Both pairs are greatly
modified and act as copulatory organs (see section on
lieproductive Organs).

First pair (PI. II, fig. 14). Each consists of two
parts — a broad basal portion, probably the protopodite
(prot.), and an elongated distal portion, which is rolled

31

on itself longitudinally to form a tube. This distal
portion probably represents tlie endopodite (end.). The
two basal portions fuse in the middle line, thus forming
a tunnel-like structure extending backward below the
second somite. Below the fused basal portions of the first
pleopods arise the second pair of appendages (PI. II,
fig. 15). Each consists of two parts — a horizontal rod
(prot.) projecting posteriorly, and a vertical rod (end.)
attached to the posterior end of the first portion. The
vertical rod is divided into two parts by a transverse
groove. The horizontal rod probably represents the
protopodite, and the vertical portion is the endopodite.
There is no trace of an exopodite on any of the male
pleopods.

The vertical rod-like portion of the second pleopod fits
into the tube of the first pleopod.

ENDOPHRAGMAL SKELETON
(PI. Ill, fig. IS, Text figs. 4, 5, 0, 7).

The post-oral region of the cephalothorax has an
extremely complex system of internal plates, known
as the endophragmal skeleton. Essentially
tli is system may be said to consist of a number of
inwardly-projecting plates arranged transversely so as to
divide up the interior of the cephalothorax into a
series of irregular compartments. Each partition, or
a r t h r o p h r a g m, arises at the junction of two somites,
and is formed by an infolding of the sternal and epimeral
exoskeleton between these somites. Thus, each plate of
the endophragmal skeleton is double, and is composed of
two flattened plates of exoskeleton which are closely
applied together.

The primary function of the endophragmal system is

32

to afford attachment for the muscles of the proximal
region of the appendages in this region of the body. It is
also useful in supporting and protecting certain portions
of the viscera.

Although at first sight the arthrophragms of the first
five post-oral cephalothoracic somites differ in a marked
degree from those of the posterior thoracic region, it will
be shown that all are built on the same plan.

Description of a Typical Arthrophragm
(Text fig. 4).

The fourth thoracic arthrophragm (between the
fourth and fifth thoracic somites) niay be taken as a type.
It is a vertical partition extending inwards at each side
from the line of junction of the fourth and fifth thoracic
epimera. The portion of the partition in contact with the
thoracic sternum arises between the fourth and fifth
thoracic sterna. Thus we may distinguish between two
kinds of plates, viz., those growing inwards from the
epimera — the endopleurites (PL III, fig. 18, ep., also
Text figs. 4, 5, 7), and those arising from the inner side
of the sternum the endosternites. Each arthrophragm
consists, therefore, of an outer endopleurite and an inner
endosternite at each side of the middle line. The two
endosternites in the arthrophragm under discussion are
separated from each other in the middle line by the
median plate (fig. 18, med. />.), which is an ingrowth from
the median suture present on the last four thoracic sterna.
The plates of which the arthrophragm is composed are
sometimes known as the " apodemata."

The Endosternite is irregular in shape and has
five principal borders.

The median border is vertical, and is the part of the
endosternite in contact with the median plate.

33

The sternal border (Text fig-. 4, Sb. s.) is in contact
with the sternum, and forms the ventral boundary of the
endosternite.

The articular harder {Ah. s.) passes upwards and out-
wards, and is equal in length to the sternal border. It is
connected with the arthrodial membrane in contact
with the coxopodites of the fourth and fifth thoracic
appendages.

-Pl

Endoplmritt, £2 EndosUrniU,

Fig. 4. — Diagram of a typical arthrophragm.

Db.p. = dorsal border of the endopleurite.

Db.s. = dorsal border of the endosternite.

Eb.p. = epimeral border of the endopleurite.

Ab.p. = articular border of the endopleurite

Ab.s, = articular border of the endosternite.

Sb.s. = sternal border of the endosternite.

A.F . = apodemal foramen.

The outer border passes imvards and upwards, and at
its upper and lower ends fuses with the endopleurite.
This fusion is interrupted in the middle region of the
border by the large apodemal foramen (A.F.) which lies
between the endosternite and endopleurite.

The inner part of the dorsal border (L)b. s.) passes

34

upwards and outwards from the median line, describing
almost a semi-circde. The upper edge of the semi-
circle almost reaches the median line. Thus the inner
portion of the dorsal border of each side surrounds an
almost closed cavity, which corresponds to the " sternal
canal " of the Macrura.

The E ndopleurite is rectangular in shape, and
its length is about twice as great as its width. Four
borders ma}* be distinguished.

The inner border is in contact with the outer border
of the endosternite at its upper and lower ends. In the
intermediate region it forms the outer boundary of the
apodemal foramen.

The articular border (Ab. j>.) is in contact with the
upper part of the arthrodial membrane connecting the
coxopodites of the fourth and fifth thoracic appendages.

The epimeral border [Kb. p.) is in contact with the
fourth and fifth thoracic epimera.

The dorsal border (Db. p.) bounds the dorsal free end
of the endopleurite

Above the apodemal foramen the endopleurite
becomes fused on its posterior face with the following
arthrophragm, and the anterior face of the arthrophragin
under discussion becomes fused with the preceding
endopleurite.

All the arthrophragms of the post-oral cephalo-
thoracic region are built on the above plan, that is to
say, each somite has one endosternite and one endopleurite
at each side. But in some cases the homology is very
much disguised.

The last five thoracic arthrophragms are very similar
to the one described, but the anterior arthrophragms are
extremely reduced. It is, therefore, advisable to describe
the endophragmal skeleton in two parts.

85

(1) The last five thoracic arthrophragms, beginning
in front and working backward (posterior thoracic).

(2) The two post-oral cephalic arthrophragms and
the first three thoracic arthrophragms, beginning behind
and working forward (anterior post-oral).

(1) Posterior Thoracic Endopiiragmal System
(PL III, fig. 18, and text, fig. 5).

This consists of the arthrophragms of the last five
thoracic somites

The median, 'plate commences at the posterior end of
the fourth thoracic sternum. At first it is extremely
shallow, but as it proceeds posteriorly it increases in
height. It is present in the last four thoracic somites.

As in other parts of the endopiiragmal system, the
median plate is composed of two closely applied portions
of the exoskeleton. In the fifth thoracic somite these two
parts remain separate, and the cavity between them opens
to the exterior at the posterior end of the fourth thoracic
sternum.

Each endosternite is at right angles to that part of
the sternum from which it arises, and similarly each
endopleurite arises at right angles to the epimeron. If
the sternum were horizontal throughout its entire length,
and also if the epimeral wall at each side were vertical,
the endopiiragmal system would be represented by a series
of vertical partitions arranged one behind the other. This
is the case in the Macrura. In the Brachyura, however,
neither the sterna nor epimera follow this arrangement.
The thoracic sternum is extremely convex antero-
posterior^ and has an extreme upward tilt at its posterior
end. The epimeral wall, instead of having a flat surface,
is extremely convex on its outer face. The shape of the
sternum and of the epimeral wall gives rise to much

86

complexity in the endophragmal system of the last five
thoracic somites. The endosternite and the endopleurite
of the same arthrophragm instead of heing in the same
plane, as in the Mariana, may be situated at a consider-
able angle to each other, so that it is difficult to believe
that they belong to the same segment. The fifth endos-
ternite is almost vertical, but the succeeding endosternites
incline more and more forward until the last arthro-
phragm is practically horizontal.

The endopleurites of each arthrophragm become fused
with the anterior face of the following arthrophragm, and
thus we have each somite divided into four chambers.
There is an outer chamber at each side lying between two
consecutive endopleurites, and bounded on the outer side
by the epimeron and on the inner side by the backward
growth of the endopleurite. These chambers may be
called the Pleural muscle chambers (Text fig. (i. I'.).

There is also an pinner chamber at each side, lying
between two consecutive endosternites, and separated from
one another by the median plate. We designate these the
Sternal muscle chambers (Text fig. 6, S.).

These pleural and sternal muscle chambers contain
the muscles which work the two basal segments of the
appendages in this region.

The muscle chamber of the last walking leg is not
divided into parts owing to the absence of a separate
endopleurite in this somite. Therefore this last muscle
chamber may be known as the Pleuro-Sternal muscle
chamber (Text fig 0, PS.). Each of these chambers has
an antero-lateral prolongation, which extends forward as
far as the posterior face of the fourth thoracic arthro-
phragm.

The fourth thoracic arthrophragm (Text fig. 5, A.)
arises between the fourth and fifth thoracic somites. This
arthrophragm has already been described.

37

The fifth thoracic arthrophragm (Text fig-. 5, B.)
arises between the fifth and sixth thoracic somites. All
the parts are very similar to those described in the fourth
arthrophragm. The sternal border is slightly more

c.

Fie. 5. — Anterior view of the left side of thoracic arthrophragms.
A. = 4th thoracic arthrophragm. B. =5th thoracic arthrophragm.

C. = 6th thoracic arthrophragm. D. =7th thoracic arthrophragm.

(The parts are shaded as in Fig. 4).
g. =line of fusion with the following arthrophragm.
//.=line of fusion with the 3rd thoracic endopleurite.
k. =line of fusion with the 4th thoracic endopleurite.
I. = line of fusion with posterior face of the preceding thoracic endosternite.
m. =antero-lateral extension of pleuro -sternal muscle chamber.
H.=line of fusion with the 5th thoracic endopleurite.
o. = line of fusion with the (5th thoracic endopleurite.
p. =line of fusion with the 6th thoracic endosternite

arched. The apodemal foramen is not quite so large.
There is an additional cavity left in the dorsal side of the
endosternite at each side (m.) This is formed by the

38

anterior prolongation of the last pleuro-sternal mnscle
chamber.

The endosternite is almost vertical, but there is a
slight forward tilt. The endopleurite is slightly concave
on its anterior face, and its plane is slightly posterior to
thai of the endosternite. The border of the endosternite
surrounding the sternal canal has a slight notch. Around
the foramen of the pleuro-sternal muscle chamber the
endosternite of this arthrophragm fuses in front with the
fourth endosternite (I.) and posteriorly with the sixth
endosternite. Similarly along the outer border of the
endosternite, above the apodemal foramen, this arthro-
phragm is fused in front with the fourth endopleurite (k)
and behind with the sixth endopleurite.

The sixth thoracic arthrophragm (Text fig. 5, C.)
arises between the sixth and seventh thoracic somites.

The endosternite is convex on its anterior face, and
its plane is inclined considerably forward. Its median
border is longer than in the previous arthrophragms. The
notch in the sternal canal is much more pronounced than
in the fifth arthrophragm. The upper border of the
sternal canal is bent posteriorly, and its inner tip becomes
fused with the upper edge of the median plate in the last
somite. The dorsal border is fused with the anterior (or
dorsal) edge of the last arthrophragm. In this way the
seventh endosternite is completely roofed over and cannot
be seen clearly until the sixth endosternite is removed.
As in the previous arthrophragm, there is a large foramen
in the dorsal region of the endosternite through which
the pleuro-sternal muscle chamber passes (m.).

The endopleurite is very similar to that of the fourth
arthrophragm. At the junction of the endosternite and
endopleurite this arthrophragm fuses in front with the
fifth endopleurite (».), and behind with the seventh endo-

39

pleurite. Also at the edge oi the foramen bounding the

pleuro-sternal muscle chamber this endosternite is
fused in front to the fifth and behind to the seventh
endosternite (/.).

The seventh thoracic arthrophragm (Text tig. 5, D.)
lies between the seventh and eighth thoracic somites.

The endosternite is inclined at an angle of 50° to the
vertical, the upper border being anterior. It is almost
completely covered by the overhanging sixth endosternite.
The median border is very deep and the sternal canal is
very small. The dorsal border is almost level, and partly
bounds the ventral side of the pleuro-sternal muscle
chamber. The endosternite does not completely surround
this chamber as in the two previous arthrophragms. The
sternal border is inclined at a considerable angle to the
horizontal. The apodemal foramen is small.

The plane of the endopleurite is almost at right
angles to tuat of the endosternite. At the junction of the
endosternite and endopleurite this arthrophragm fuses in
front with the sixth endopleurite (o.), and where the
endosternite borders the pleuro-sternal muscle chamber
there is a fusion with the sixth endosternite (p.).

The eighth thoracic arthrophragm (PL III, fig. 18,
e. st. 13) lies at the posterior end of the last thoracic
somite. In this somite there is no separate epimeron.
It is probably fused with the sternum. The arthro-
phragm, therefore, shows no division into endosternite
and endopleurite. It may be accepted, however, that this
arthrophragm represents the fused endosternite and endo-
pleurite. It consists of two halves, which are separated
in the median line by the posterior end of the median
plate. This arthrophragm is practically horizontal, and
was designated the " sella turcica " by Milne-Edwards.
As already stated, the last arthrophragm fuses in front

40

with the dorsal border of the sixth endosternite. Between
this arthrophragm and the seventh endoplenrite there is
a large foramen.

Pig. 6 — -Diagrammatic plan
of the endophragmal
system to show the
muscle chambers. (The
corresponding parts
are shaded similarly
throughout).

1 — 7 = thoracic arthro-
phragms 1 to 7.

LP. - VII. P. = pleural
muscle chambers of
thoracic somites 1 — 7.

I.S. — VI 1. 8. = sternal
muscle chambers of
thoracic somites 1 — 7.

VIII. P. S. = pleuro-sternal
muscle chambers of last
thoracic somite.

Ep. = epimeron.

E.s. = endosternite.

PI. — endoplenrite.

St. = sternum.

m. = median plate.

(2) Anterior Post-oral Endophragmal System

(Text fig. 7).

This consists of the last three thoracic and the two
post-oral cephalic arthrophragms.

The nature of the epimera and sterna in this region
naturally decides the form and extent of the corresponding
arthrophragms.

The apparently great differences between the endo-
phragmal system of this region and that of the posterior
thoracic somites is due to three main causes.

41

(1) There is no median plate. This only begins at
the level of the fourth thoracic arthrophragm.

(2) The post-oral sterna anterior to the fourth
thoracic arthrophragm are all fused together. Con-
sequently there can be no broad plate-like endosternites
formed as in growths between the somites. The small
endosternite present in each somite of this region is rod-
like, and represents merely the articular border of the
typical endosternite. (The third thoracic endosternite
has the form of a fairly broad and deep plate, and is
therefore an exception to this rule.)

(3) The epimera in front of the second thoracic
arthrophragm are fused together, so that the endopleurite
in these somites are extremely reduced and represent only
the articular border of the typical endopleurite.

Here, as in the posterior thoracic region, each
endopleurite gives off a posterior out-growth, which fuses
with the following arthrophragm. So that pleural
muscle chambers and sternal muscle chambers may be
made out, but owing to the rod-like nature of their con-
stituent parts, they have a very different appearance from
the muscle chambers of the posterior somites of the
thorax (see Text tig. 6, also PI. Ill, tig. 18).

The third thoracic arthrophragm (Text tig. 7, E.)
arises between the third and fourth thoracic somites.

Each endosternite (PI. Ill, fig. 18, e.rt.S) differs from
that of the typical arthrophragm described above. It
arises merely from the outer edge of the sternum, so that
the two endosternites are separated from each other by
the entire width of the sternum in this region. The
endosternite is a broad and deep plate facing downwards
and backwards. The dorsal and inner corner is prolonged
inwards and backwards and almost meets the similar part
from the other side.

42

Fig. 7.

A. = anterior view of 1st post-oral cephalic arthrophragm.

B. — E. = anterior views of the left side of the following arthrophragm.

B. = 2nd post-oral cephalic arthrophragm.
G. = 1st thoracic arthrophragm.

D. = 2nd thoracic arthrophragm. E. = 3rd thoracic arthrophragm.
(The parts are shaded as in Fig. 4).

a. = junction of 2nd cephalic endopleurite with 1st thoracic endosternite.

b. = point of fusion between 2nd thoracic endosternite and 2nd ceph a lie

endopleurite.

c. = line of fusion between 2nd thoracic endopleurite and 3rd thoracic

endosternite.

d. = point of fusion with the 1st thoracic endopleurite.

e. = line of fusion between 3rd thoracic endopleurite and the 4th thoracic

endosternite.
/. = line of fusion between 3rd thoracic endosternite and the 2nd

thoracic endopleurite.
x. = point of fusion between the 2nd cephalic endopleurite and the

2nd thoracic endosternite.
y. = point of fusion with 1st cephalic endopleurite.

(The dotted line in Fig. D represents the 3rd thoracic endosternite.)

43

On its anterior face the endosternite is connected
with the narrow plate-like second thoracic endoplenrite
(/.). K"ear the point of junction of these two plates the
dorsal and articular borders are prolonged backwards as
rod-shaped pieces, each of which comes into contact with
anterior rod-like outgrowths from the corresponding
borders of the third thoracic endopleurite.

The endopleurite arises betwen the third and fourth
epimera. It gives off two short anterior rod-like
prolongations from the dorsal and articular borders which
fuse with the rod-like extensions of the endosternite
mentioned above. The main part of the endopleurite,
however, consists of a broad plate, which passes backwards
and fuses with the fourth thoracic arthrophragm (e.).

Second thoracic arthrophragm (Text fig. 7, D.).
The endosternite is much more reduced than that of the
third thoracic arthrophragm. It arises from the
upturned edge of the sternum in this somite, and has a
very irregular shape. Its inner portion passes upwards,
and fuses with a narrow membranous process projecting
downwards from the last cephalic endopleurite (&.).

The articular border is prolonged outwards as a rod-
like process, which fuses with the extremely small
articular border of the endopleurite of the same arthro-
phragm. About half wTay along the articular border the
endosternite fuses with a posterior rod-like extension of
the first thoracic endopleurite (//.).

The endopleurite of this arthrophragm is a deep
narrow plate arising at the junction of the second and
third thoracic epimera. From its lower end it sends
forward a short process which fuses with the outer part of
the endosternite. The main part of the endopleurite
passes backwards and becomes fused with the third
thoracic endosternite (('.), as described above.

44

First thoracic arthrophragm (Text fig. 7, C). The
endosteruite arises from the upturned edge of the
sternum. It consists of a simple rod which passes back-
wards, upwards and outwards parallel to the articular
border of the second thoracic endosternite. It fuses with
the endopleurite of the same arthrophragm, but
immediately before doing so it comes into contact, on its
anterior side, with a posterior prolongation from the last
cephalic endopleurite (a.).

The epimera of the first and second thoracic somites
are fused together, and the endopleurite of this arthro-
phragm arises from the ventral edge of the fused epimera
immediately in front of the origin of the second thoracic
endopleurite. It is rod-like, and passes forwards and
inwards in precisely the same plane as the first thoracic
endosternite, with which it fuses. Near its fusion with
the latter, the endopleurite gives rise to a posterior
process which fuses with the second thoracic endosternite.

Last cephalic arthrophragm (Text fig. 7, B.). The
endosternites of the two post-oral cephalic arthrophragms
are fused together, but there is a distinct longitudinal
suture present, which assists in the identification of the
two parts.* The fused endosternites pass outwards and
backwards parallel to the first thoracic endosternite.
After a short distance the last cephalic endosternite
becomes distinct from the anterior endosternite, and at
the point of separation a prolongation from the first
cephalic endopleurite fuses with the endosternites (y. ).
From this point the posterior endosternite passes outwards
and fuses with the lower border of the last cephalic
endopleurite.

The last cephalic endopleurite is an irregular

:;: There is also a well-marked groove separating the sternum of
these two cephalic somites.

45

membranous plate divided into a dorsal and a ventral
portion. The dorsal portion has on its inner side a down-
wardly projecting' process which fuses with the upper
part of the second thoracic endosternite (./.) as described
above. The ventral portion of the endopleurite has an
upper crescent-shaped region and a lower part which
fuses with the endosternite.

From the posterior side of the lower portion of the
endopleurite is given off a rod-like process which fuses
with the first thoracic endosternite.

First post-oral cephalic arthrophragm (Text fig.
7,^4.). In addition to the portion fused with the last
cephalic endosternite, the endosternite of the above
arthrophragm lias an anterior process at each side which
form the skeleton of the metastoma (PI. Ill, fig. IN, met.)
or posterior lip of the mouth.

The endopleurite arises from the soft membranous
epimeron immediately behind the insertion of the external
abductor muscle of the mandible. It passes backwards
and gives rise to a small upwardly directed process, and
afterwards becomes joined to the fused eirdosternites.

Integument (Text fig. 8).

The crab is covered by a continuous chitinous

exoskeleton, which serves partly as a protective covering
and also as a means of attachment for the muscles. The
main portion of this exoskeleton is strongly calcified.
Between the movable somites of the abdomen, however,
and also between the articulating segments of the
appendages, the exoskeleton remains nncalcified in order
to allow of free movement, and has the appearance of a
thin chitinous membrane, known as the " arthrodial
membrane."

The exoskeleton of the ventral reg-ion of the abdomen

46

is but feebly calcined. The cuter walls, the floor and
roof of the branchial chamber and the roof of the prc-
branchial chamber are also extremely thin and
membranous. The chitinous linings of the fore-gut and
hind-gut, which are continuous with the exoskeleton, are
uncalcified, except in those regions of the fore-gut where
the ossicles are present. The gills, also, are covered by
an extremely fine chitinous layer.

The integument of the crab consists of an epidermis,
below which lies the dermis. On the outer side of the
epidermis is a chitinous layer, the thickness of which
differs considerably in various parts of the body. This
outer chitinous layer is a product of the epidermis, and
constitutes the exoskeleton already referred to. The
chitinous layer may be impregnated with calcareous
salts.

The epidermis [Chitogenous epithelium, Vitzou*]
(Text fig. 8, e.) consists of a single row of columnar cells
resting upon a basement membrane (/.). These cells differ
in their appearance in various parts of the body, and also
show marked changes during the interval between one act
of ecdysis and the next. In the dorsal integument of the
hard crab, for example, the cells of the epidermis are
only moderately columnar, but in the oesophagus of the
same animal the cells are extremely elongated. In the
soft crabs the cells are of much greater length compara-
tively than in the hard crabs.

In some regions where we have two parts of the

integument coming close together, such as at the edge of

the carapace and also the gill lamellae, the cells of the

epidermis sometimes become extremely elongated and

pass across the dermis to fuse with similar cells from the

* Vitzou, A. N. "Eecherchos sur la structure ct la formation des
teguments chez les Crustaces Pecapodes." Arch, de Zooloqie exp&r. et
g&n., T. X. [1882], p. 451.

47

epidermis of the opposite side. Vitzou lias termed such
cells " colonnades de soutien.''

The cells of the epidermis are also elongated at the
point where the muscle fibres arise.

k:^

Fig. S. — Diagrammatic section through the integument of a hard crab.

a. = cuticle.

b, = pigmented layer,
r. = calcified layer.
d = non-calcified layer.
e. — epidermis.

I.

f. = basement membrane.

g. = dermis.

h. = duct of the cutaneous gland.
k. =' muscle attached to the base-
ment membrane.

cutaneous gland

The dermis [g.) lies below the basement membrane,
and varies in thickness. It consists mainly of a network
of connective tissue fibres and scattered cells. There is
a laver oi pigment cells close to the basement membrane.

48

As pointed out by Cuenot,* there are two kinds of cells
present in the connective tissue which contain reserve
material. These are the " cellules de Leydig " and the
" cellules proteiques." The former contain glycogen, and
are present in great numbers, especially when the period
of ecdysis approaches. The latter are also present in
great abundance, and contain proteid material.

The cutaneous glands are also embedded in the
dermis. The muscle fibres stretch across the dermis, and
are attached to the inner side of the basement membrane.
The muscles when dissected appear to be attached to the
exoskeleton, but an examination of sections reveals the
fact that they do not extend farther than the basement
membrane.

The chitinous layer of the integument is situated
on the outer side of the epidermis, and consists of several
layers. Commencing from the outside, these are as
follows : —

(1) The cuticle (Text fig. 8, a.) is an extremely thin
structureless layer covering the whole of the chitinous
exoskeleton. Its continuity is interrupted at intervals
where the ducts of the cutaneous glands open to the
exterior. From the cuticle there arise numerous small
papillae, which are only seen when examined under the
microscope. These must not be confused with the setae,
which are visible to the naked eye and which have an
entirely different structure (see below).

(2) The pigmented layer (b.) is a moderately-thick
layer containing pigment. In the hard parts of the
exoskeleton this layer is calcified. It has a laminated
structure, and the numerous layers of which this portion
of the exoskeleton is composed are parallel to the surface.

(3) The calcified layer (c.) is the broadest layer of all

* Cuenot, L. " Etudes phvsiologiques sur les Crustaces Deca-
podes." Archives de Biologie, T. XIII, p. 245.

49

in the fully-formed exoskeleton. It is colourless and
richly impregnated with calcareous salts. Like the
previous layer, it exhibits striations parallel to the
surface, but the laminae are generally broader than in the
pigmented layer. It is to this layer that the great thick-
ness and hardness of the shell in a hard crab are due, as
new laminae are constantly being added to this region
until the exoskeleton attains its maximum thickness.

(4) The non-calcified layer (d.) is a very thin layer
composed of delicate laminae parallel to the surface. This
layer remains in a very soft condition, and is not formed
until the calcified layer has attained its maximum width.

Vertical sections through the integument reveal the
fact that there are striations in the chitin at right angles
to the surface, as well as the horizontal lamellae already
referred to. Also, as Vitzou has pointed out, in horizontal
sections the chitinous integument is divided up into small
hexagonal areas, and in each of these areas small pores are
present. Vitzou determined that these areas were of the
same size and shape as the horizontal sections through
the cells of the epidermis. He concluded, therefore,
that the exoskeleton is composed of innumerable hexagonal
prisms packed side by side, having their long axes at
right angles to the surface of the body. Furthermore,
each of these chitinous prisms is in contact with the
outer end of an epidermal cell. So that for every cell of
the epidermis there is a corresponding prism forming a
unit of the chitinous exoskeleton. Such an explanation
accounts for the presence of the vertical striations in
vertical sections, and for the polygonal areas in the
horizontal sections. The small pores in the middle of
these areas are due to the presence of numerous fine
canals traversing each prism from the epidermis to the
exterior.

50

H o d e of Formation of the Exoskeleton.

Originally the exoskeleton was believed to be pro-
duced by a secretion from the cells of the epidermis.
Yitzou, however, claimed that the process is effected in a
different manner. According to him the new shell is
produced in the following way. The contents of each
epidermal cell becomes modified at the outer margin.
This outer part becomes cut off from the rest of the cell.
Thus at this stage the epidermis is covered by a thin layer,
which, however is not one homogeneous whole, but is
divided up into numerous polygonal areas, each area
corresponding in shape and position to an epidermal cell.
The process is repeated ; the outer part of each cell is
again cut off, and at this stage we have a two-layered
polygonal cylinder above each cell. This process is
repeated until we have built up over each cell a multi-
layered cylinder. Since the cells of the epidermis lie
close together, the chitinous cylinders are also tightly
packed and form what appears to be a continuous
exoskeleton. The striations parallel to the surface of the
integument represent the successive lines of growth.
Thus, according to Yitzou, the process of formation of the
chitinous integument consists in the successive thickening
of the outer walls of the epidermal cells.

Setae.

These are long hair-like processes which project from
the exoskeleton in various regions of the bodv. In
sections each seta is seen to arise from the region of the
epidermis as a narrow tube enclosing a cavity. This tube
passes through the chitinous layers and projects from the
exterior as a long narrow process. Its walls are cuticular
and are continuous with the thin structureless cuticle

51

covering the chitinous integument. So that wherever a
seta arises the continuity of the thick exoskeleton is
broken in order to allow this tube-like prolongation to
iea eh the exterior. The contents of the setae are proto-
plasmic and are connected with the epidermis. In some
regions of the body the setae have nerve fibres passing to
their interior. These are the sensory setae, of which there
are several kinds (see section on Sense Organs). The setae
may be simple prolongations, or they may consist of a
central axis, from which arise off-shoots. In the latter
case the cavity of the central axis is not continued into
the lateral out-growths. In addition to the setae
described above, there are small papillae on the surface
of the shell, which are merely thickenings of the cuticle
and do not contain a cavity. Vitzou states that in
Portunus these cuticular processes are comparatively
long. In Cancer, however, they are extremely small, and
can only be detected under the microscope. Yitzou
affirms that the long " setae," present in the walls of the
fore-gut, have no central cavity, and are probably merely
extremely large cuticular prolongations and not true setae.
These " setae " in the fore-gut act as strainers. Where
the sub-branchial region of the carapace is closely applied
to the bases of the thoracic legs there is a rich growth of
setae. These probably assist in preventing the water
from entering the branchial chamber at the base of the
thoracic legs.

The inhalent branchial opening is also well guarded
by long setae, both on the flabellum of the third maxilli-
pede and on the anterior border of the coxopodite of the
chela. The setae on the endopodites of the pleopods in
the female are used for the attachment of the eggs.

52

Cutaneous (or tegumentary) glands.

Scattered throughout the connective tissue, near the
basement membrane, are globular masses of cells, each
cellular clump being connected with the exterior by a fine
duct. These are the cutaneous glands. Each cell of the
globular mass is in contact with a small cavity on its
inner side. This central cavity of the glandular mass
receives the secretion from the various cells. The cavity is
connected with the duct, and thus the glandular secretion
is enabled to pass to the exterior. The duct is lined by
a tine protoplasmic wall. The wall of the duct probably
represents a single cell, in which case the cavity of the
duct is intracellular. The gland cells and the duct cell
are all modified epidermal cells.

The cutaneous glands are scattered throughout the
integument, and in some regions are extraordinarily
abundant. The glands present in the mandibles and in
the walls of the oesophagus, and also those in the hind-
gut, have a similar structure to the ordinary glands on the
surface of the body. They are, in fact, modified cutaneous
tegumentary glands.

Immediately in front of the mouth there is a compact
mass of cutaneous glands at each side, which open on the
surface of the epistoma. These glands have the structure
of the typical cutaneous glands, but are extremely large.
They are about four times as large as those present in the
walls of the oesophagus (see section on Alimentary
Canal). Similar glands are found also in the metastoma,
packed very closely together. Herrick* has also observed
them in the same regions in the lobster. (See fig. 60.)

In the floor of the branchial chamber there is a well-
defined transverse ridge lying in front of the inhalent

* Herrick. "The American Lobster." Bull. U.S. Fish Com.,
Vol. XV., 1895.

53

branchial aperture. The epidermis in this region
presents a very interesting condition, and there appear to
be numerous modified cutaneous glands.

There are also great numbers of cutaneous glands
present on the endopodites of all the maxillipedes.

On the endopodites of the pleopods of the female there
are closely-packed tegumentary glands. According to
Herrick, these secrete the cement which attaches the eggs
to the endopodites of the abdominal appendages.

The function of the various tegumentary glands in
various parts of the body is not clearly known. Lang*
states that some have an excretory function. There is
little doubt that the functions of these glands differ in
various regions of the body. Those, for example, on the
pleopods are extremely specialised. It is not incon-
ceivable that the glands in the integument of the
epistoma and metastoma may produce a secretion which
is poured on the food as it enters the mouth.

The glands in the walls of the oesophagus are
probably salivary glands. Herrick thinks that this
explanation of their function no longer holds good, since
glands of similar structure have been found in the walls
of the hind-gut. This argument, however, does not carry
much weight, if we recognise that all the tegumentary
glands (both on the surface and in the walls of the
alimentary canal) have the same essential structure, and
yet are capable of performing different functions in
various regions of the body.

ECDYSIS.

The epidermis of all Arthropods is covered by a
continuous layer of chitinous integument, which may
become calcified in certain regions. This outer integu-

* Lang. Text-book of Comparative Anatomy, Part I.

54

ment is continuous with the cliitinous lining of the fore-
gut and hind-gut. The body, therefore, may he said to
be enclosed in an inflexible coat, which prevents the
tissues from expanding. The growth of the animal cannot
be gradual, but can only take place when the animal
breaks through the stiff outer covering. Immediately
after exuviation, the animal, which is then only covered
by an extremely thin flexible membrane, will increase in
size. This process of casting, or ecdysis, is characteristic
of all Arthropods. Ecdysis takes place periodically, and
growth can only take place while the animal is in a
" soft " condition.

In Cancer, when ecdysis is about to take place, the
carapace opens along the pleural groove at each side.
These two longitudinal splits become connected posteriorly
with a transverse opening, which makes its appearance
between the posterior border of the carapace and the
tergum of the first abdominal somite. Thus the tergal
region of the carapace is free from the remainder of the
exoskeleton, except along a line marking the posterior
boundary of the first cephalic sternum. The carapace,
therefore, acts like a lid of a box, and is hinged anteriorly.
The first part of the body to be withdrawn from the old
shell is the abdomen, which is followed by the various
legs. When all the parts are completely free the crab
emerges from beneath the hinged carapace.

On the dorsal sides of the basi-ischium and meros of
the chela there are faint grooves (PL II, fig. 12, abs.).
These are the " lines of absorption," and at the time of
ecdysis the exoskeleton of the chela loses its calcification
at these points. In this way the withdrawal of the large
claw is effected, as it would be extremely difficult for the
chela to be withdrawn if the integument at the base of the
limb remained hard.

55

As pointed out by Vitzou,* the method of ecdysis in
the Macrura differs from that found in the Brachyura,
because in the latter the abdomen is withdrawn first. In
the Macrura the thorax is first withdrawn, and the
abdomen leaves the old shell last.

The tissues of the animal become greatly changed
immediately before ecdysis. The blood increases
enormously in volume, and Wittent suggested that the
increase is due to the absorption of water by means of the
digestive gland. He presumed that this excess of blood
plasma produced the internal pressure necessary for
ecdysis and growth. The muscles become very soft and
semi-fluid, and the fibres lose their well-defined outlines
and cross-striations.

The digestive gland probably increases in size during
ecdysis. The fat cells are stocked with glycogen, the
ferment cells are much bigger, and the colour of the
ferment vesicle is of a deep brown colour, thus giving
the digestive gland a deeper colour at this period.
The reproductive organs are generally in an immature
condition at the time of ecdysis.

Immediately before and after ecdysis the crabs are
unfit for food. They are " watery " and have a bitter
taste. Keference is made in the Economic section to the
" Granny " crabs, which are considered by the fishermen
to be diseased crabs. I have reason to believe that they
are merely crabs preparing for ecdysis.

One of the most interesting changes which accom-
pany ecdysis is probably the formation of the new integu-
ment, as a result of the extreme activity of the
epidermal cells. This new exoskeleton is already formed
when the hard shell is discarded.

* Vitzou. Arch. zool. exp. et gen., T. X, 1882.
t Report on the Scientific Investigations, Northumberland Sea-
Fisheries Committee, 1903.

56

Before ecdysis the cells of the epidermis become
greatly elongated, and in the underlying dermis the cells
of Ley dig, which are rich in glycogen, become extremely
numerous. The supply of reserve food material in these
cells is evidently of the utmost importance at a time when
growth and regeneration of the tissues is taking place.
At the time when the crab is preparing to cast, a new
chitinous layer is formed by the epidermis. This new
layer is separated from the old shell by a gelatinous
fluid. The chitinous layer, which is the first appearance
of the new exoskeleton, consists of two parts — an outer
structureless cuticular layer, and an inner chitinous layer
containing pigment. This inner layer represents the
pigmented layer. The calcified and non-calcified layers
are not produced until after ecdysis. The calcified layer
grows throughout the greater part of the period until the
next ecdysis, and it is to this layer that the hardness and
increasing thickness of the shell is due.

Vitzou's theory explaining the method of formation
of the exoskeleton has been described above (see section
on Integument).

The frequency of casting and other problems
connected with ecdysis are discussed below in the section
on Economics.

AUTOTOMY AND REGENERATION OF LlMBS.

One of the most interesting and characteristic
features in the natural history of the crab is the power
the animal possesses of throwing off injured limbs
(autotomy) and of forming new limbs to replace the old
(regeneration).

The processes associated with these phenomena may
be briefly stated as follows : —

If the distal portion of one of the pereiopods be

57

seriously injured, the crab immediately throws off part of
the limb. The whole limb is not sacrificed. The self-
amputation always takes place along the thin groove
present on the basi-ischium representing the line of
separation between the basipodite and ischiopodite. This
groove, therefore, may be said to surround the fracture
plane (PI. II, fig. 12, f.p.).

When autotomy has been effected, the fracture plane
is seen to be covered by a thin membrane, or diaphragm,
which is perforated, slightly below the centre, by a small
foramen. The blood flows out through this small opening,
but soon coagulates, forming a clot over the mouth of the
foramen and also on the outer surface of the diaphragm.

The diaphragm with its outer coating of coagulated
blood assumes a dark brown colour in a few clays, and
ultimately becomes quite black. This black coating is
worn away in course of time, and reveals a thin membrane
extending across the stump.

Beneath the membrane a small papilla makes
its appearance, and marks the commencement of the
regeneration of the limb.*

Conditions necessary for Autotomy.
The successful performance of self-amputation in
Cancer depends upon several conditions, of which the most
important are discussed below.

1. The crab must be healthy.

This is a most important factor. Animals which are
in a diseased or weak condition, or which have been kept
out of water for a considerable time, and in which, as a

* According to Williamson, the regeneration only takes place
when the crah is preparing for ecdysis. The limb does not attain
its full size at the first moult after regeneration. Two or three
moulting processes must take place before the limb attains its normal
size.

58

consequence, the nerve responses are feeble, do not
perform autotoniy very readily.

2. The nerve of the limb must be sufficiently stimulated.
This appears to be a self-evident proposition. What-
ever may be the cause of autotoniy, and whatever may be
the reason of this complex phenomenon, it is without
doubt the result of nervous stimulation. But the question
as to wluit is a " sufficient " stimulation cannot be disposed
of so easily (see below under the general discussion on
Autotoniy).

3. The thoracic nerve mass must remain intact.

We are indebted to Fredericq* for his investigations
on the physiological processes involved in autotoniy. He
has proved that the latter is the result of a reflex, and that
the thoracic ganglion belonging to the appendage is the
centre of this reflex. The brain is the seat of voluntary
and co-ordinated movement in Cancer, and if the brain be
removed autotoniy will still take place. If, on the other
hand, the thoracic nerve mass be removed or destroyed,
self-amputation cannot proceed. The afferent nerve
fibres which are stimulated as the result of injury to the
limb are connected with the ganglion cells of the thoracic
nerve mass, and from these the efferent fibres pass to the
extensor muscle of the basi-ischium. This muscle is the
one -concerned in the autotoniy, and thus we are provided
with a fourth condition.

4. The integrity of the extensor muscle of tlie basi-ischium

must be maintained.

The first movement after the limb lias been injured
is the extension of the basi-ischium; i.e., it moves in a
dorsal direction. This movement continues until the

* Fredericq, L. " Nouvelles recherches sur l'autotomie chez le
crabe." Archives de Biologie, T. XII, 1892.

59

distal portion of the limb comes into contact with the
carapace or with some other fixed object, when the limb
bieaks at the fracture plane. That this upward move-
ment, or extension, of the basi-ischinm is necessary for
autotomy may be proved by cutting the extensor muscle
(or muscles), and then injuring' the limb. No self-
amputation will then take place. If the flexor muscle
be cut, and the extensor remain uninjured, autotomy will
proceed.

5. The distal portion of the limb must come into contact

with some point of resistance.

This condition has been emphasised above. As soon
as that part of the limb on the distal side of the fracture
plane comes into contact with some point of resistance
(e.g., the carapace) the upward movement of this portion
is stopped. The proximal portion of the basi-ischium,
however, still continues to move upwards under the
influence of the extensor muscle. Thus there are two
forces acting on the fused basi-ischium — a force at the
proximal end tending to move the segment upward, and
a force at the distal extremity preventing this upward
movement. A great strain is produced on the basi-
ischium and it snaps at its weakest point, which is the
fracture plane.

6. The stimulation to produce autotomy must be applied

between the fracture plane and the distal end of the
propodite.

The nerve does not pass into the dactylopodite, so
that if the latter segment be wounded the nerve will not
be stimulated. It is equally futile if the limb be
damaged on the proximal side of the fracture plane.

60

Amongst the Bracliyura two kinds of autotomy have
been recognised.

(1) If the crab is captured by means of one of its
limbs it will throw off the limb in order to escape from
its enemy (" evasive antotoniy ").

It is evident that all Decapods do not act similarly
nnder such conditions. It does not appear to be the case
in Cancer or Carcinns. Fredericq's researches led him
to believe that crabs did not throw off legs in order to
escape from enemies, but his experiments were confined
to a few species. Taking all the evidence available, it
would appear that autotomy does take place under the
above conditions in some crabs, such as the Maiidae and
the Grapsidae.

(2) If one of the legs of a crab be severely wounded,
the limb will be thrown off. This probably occurs without
exception in the Bracliyura.

It is well to remember that in both cases we are
probably dealing with essentially similar physiological
conditions. In both cases the autotomy is produced as
the result of the stimulation of the nerve of the leg, and
the difference appears rather to be one of degree than of
kind. In both the above cases the autotomy is produced
as the result of a reflex, and the seat of this reflex is in the
ganglion of the somite to which the autotomised leg
belongs.

Quite recently, Pieron* has concluded that there is
still another kind of autotomy which is purely voluntary,
and will not take place after the commissures connecting
the cerebral ganglia with the thoracic mass have been cut.
One of his experiments with Grapsus was as follows : — A
leg of the crab was tied to a stake within view of a

* Pieron, H. C.R. Soc. Biol, 11th May, 1907. Ibid., T. LXIII
(1907), Nos. 33 and 34.

61

sheltered crevice in the rocks. The crab discarded the
imprisoned limb as a conscious effort in order to reach the
inviting shelter. This did not happen if the brain were
destroyed or if the commissures were cut (" psychic
autotomy ").

It is difficult at the present juncture to accept
Pieron's explanation, as Mile. Drzewinat has also per-
formed similar experiments with Grapsus, and has
obtained entirely different results. But so far as Cancer
is concerned, the " psychic autotomy " does not appear to
be present.

It is not possible in the present state of our
knowledge to arrive at a definite conclusion with regard
to the full significance of the processes involved in
autotomy. But whatever may have been the lines along
which autotomy has been evolved, there is no doubt that
one of its most important objects is the prevention of
bleeding. If the arthrodial membrane between an
appendage and the body be cut, the crab will probably
bleed to death, and this appears to be one of the greatest
dangers with which the animal has to contend. The
limbs, on account of their position and size, are
continually in danger of being torn or crushed. If the
limb were seriously injured, and autotomy did not take
place, the crab would bleed to death, because the wounded
surface would probably be too large to allow coagulation
to take place. This difficulty is surmounted by the limb
being thrown off at the fracture plane, across which, as
we have already seen, a membrane is stretched. This
membrane is perforated by a small foramen through
which pass the nerve and blood streams connecting the
proximal and distal parts of the appendage. Over this
foramen a clot may readily be formed, and thus tic
excessive bleeding may be prevented.

f Drzewina, A. C.R. Soc. Biol., T. LXIII (1907), Xos. 33 and 34.

62

Histology. (Text fig. 9).

Before autotomy. A longitudinal section through
the basi-ischium of a pereiopod in which autotomy has
not been effected displays the following structure (9J..).

In the region of the fracture plane the exoskeleton
is discontinuous, the plane of the discontinuity being at
right angles to the longitudinal axis of the basi-ischium.
The break is not always easily detected, as the two parts
fit very closely together.

On the inner side of the exoskeleton is the normal
layer of epidermis (cp.). At the plane of breaking the
epidermis turns inward both at the distal extremity of the
basipoditic region and also at the proximal end of the
ischiopodite. These ingrowths extend as far as the
central nerve and blood vessels, where the epidermal
ingrowth of the basipodite (i.) becomes continuous with
that of the ischiopodite (o.). In other words, across the
plane of fracture the epidermis underlying the exo-
skeleton is not directly continuous, but becomes turned
inward as far as the central nerve of the leg.

Thus there is a double diaphragm stretching across
the leg in the fracture plane, and near the centre of this
double membrane there is a small opening which permits
of the passage of the nerve (».) and blood vessels from
one side to the other. The walls of this narrow opening
are composed of a cellular membrane, which connects
the proximal and the distal diaphragms.

After autotomy. The ischial portion of the exo-
skeleton is broken away at the fracture plane, and the
underlying structures belonging to the ischium have also
been torn away. These include the epidermis of the
ischium and also the distal portion of the diaphragm.
Stretching across the broken end of the stump (Text

63

fig. 9, /?.) is a membrane representing the proximal
portion of the double diaphragm (i.). Near the centre
of this is a small foramen. In sections taken immediately
after antotomy there is a layer of coagulated blood (b.)
on the outer side of the diaphragm.

The torn edge of the diaphragm in contact with the
foramen appears to grow over the latter. Thus, shortly
after the autotomy has been effected, there is a continuous
membrane or diaphragm covering the broken stump
(9 C). This membrane is composed of a single layer of
epidermal cells, which is continuous with the epidermis
underlying the exoskeleton of the basipodite. On the
outer side of the membrane is a layer of coagulated blood.
On the inner side of the ectoderm of this membrane, and
lying close to it, there appears to be a continuous layer of
connective tissue fibres. Miss Reed* describes also the
presence of a dense mass of blood cells immediately
beneath the membrane.

Regenerative process. Shortly after autotomy has
taken place the cells of the diaphragm begin to degenerate
(Text fig. 9, D.). Ultimately there is on the outside of the
stump a layer of dead tissue, formed of an outer layer of
coagulated blood, beneath which is the layer of degenerate
epidermal cells. According to Miss Reed, there is also
an inner layer of degenerate blood cells. The dead
epidermal cells of the diaphragm become disconnected
from the epidermis underlying the exoskeleton of the
limb, and this epidermis grows inward beneath the dead
outer layer. This takes place from all sides, and the
in-growing cells meet in the centre and form a single

* Unfortunately I did not have access to Miss Reed's paper on
the histological processes in connection with autotomy until after rny
own observations had been made. My results, in the main, however,
bear out the conclusions arrived at in her paper (Bryn Mawr College
Monographs, Reprint Series, Vol. V, 1905).

A.

64

B.

w—ep.

C. D. E.

Fig. 9. — Diagrams to illustrate the histology of the structures in the
fracture plane before and after autotomy. (In all the diagrams
the proximal end of the limb is to the left).

A. = longitudinal section through basi-ischium before autotomy,

showing the double nature of the diaphragm.

B. = longitudinal section through the basipodite immediately after

autotomy. Showing the single diaphragm and the foramen.

C. = longitudinal section through the basipodite shortly after

autotomy. The epidermis of the diaphragm has grown over the
foramen.

D. = the degeneration of the diaphragm and the formation of a new

layer of epidermis beneath.

E. — Formation of a thin cuticle (which is continuous with the
exoskeleton) by the new epidermis.

ep.

ep. =

epidermis.

exoskeleton.

proximal part of diaphragm.

distal part of diaphragm.

new epidermis.

nerve of the appendage.
b. = coagulated blood.
d. = degenerated diaphragm
and blood tissue.
cut. = new cuticle formed by the
new epidermis.

65

layer of cells beneath the outer dead layer. Eventually
a thin layer of ehitin is secreted on the outer side of these
cells (Text fig. 9, E.), and this layer of ehitin is continuous
with the exoskeleton. The old membrane, which is now
almost black, becomes worn off, and this new chitinous
membrane is exposed.

The cells in the new layer of epidermis become
extremely active, and increase in number internally. At
first an undifferentiated mass of cells is formed beneath
the membrane, but gradually differentiation takes place
and the new parts of the limb are laid down in miniature.
As they increase in size they grow on t ward, and form a
small papilla on the stump.

MUSCULAR SYSTEM.
(Pis. Ill, IV).

Muscles of the Cephalothorax.

I. Eye. The ocular peduncle consists of two parts
— an inner rod-like portion extending inwards as far as the
middle line, and an outer swollen j)ortion at the free end
of which is the visual organ. The outer portion articulates
with the inner, and is connected with the latter by means
of a flexible membrane. The movement of the outer
portion is effected by two small muscles — a ventral flexor
and a dorsal extensor.

II. First antenna. The muscles are extremely
small. The basal segment of the protopodite has a dorsal
extensor and a ventral flexor. In their natural condition
the second and third segments are flexed. In both cases
the extensor is on the inner side and the flexor muscle is
on the outer side.

66

III. Second antenna. The basal region of
this appendage is fused with the carapace, and the muscles
have degenerated. The operculum, which probably
represents the coxopodite, is still freely movable, but its
extensor and flexor muscles have now another function in
connection with the raising and closing of the operculum.
The whole question of the homology of the opercular
muscles has been fully discussed by Marchal.* The
flagellum has not much movement, and its muscles are
very small.

IV. Ma n d i b 1 e (fig. 31). There are two sets of
muscles — the adductors for closing the mandibles and the
abductors for opening the mandibles.

External adductor muscle (e.a. md.). Arises as a broad
band from the anterior and outer portion of the' sub-
hepatic region of the carapace. It passes inwards and
upwards, and is inserted on a long tendon attached to the
outer part of the mandibular apophysis.

Internal adductor muscle {i.a. md.). Arises from the
urogastric region of the carapace. It passes downwards
and forwards as a short broad muscle, and is inserted on
an extremely long narrow tendon attached to the posterior
margin of the mandible.

External abductor muscle (e.b. md.). Arises from the
posterior and inner corner of the hepatic region of the
carapace. It passes directly downwards, and is inserted on
a narrow tendon attached to the posterior side of the
apophysis. This muscle is comparatively small.

Internal abductor muscle [i.b. md.). Arises from the
top of the vertical rod-like portion of the first post-oral
endopleurite. It passes outwards and forwards, and is
attached to the outer part of the apophysis.

* Marchal. " Appareil excreteur des Crustaces Decapodes."
Archives Zool. exp. et gen. (Ser. 2), T. X, 1892.

67

V. First maxilla (fig-. 26). There are two
extensors and two flexors.

Flexors. One outer (o.e.ra.) and one inner (i.e.m.)
muscle, which run together and arise from the outer
portion of the protogastric region of the carapace. They
pass directly downwards together, and when near the
maxilla the two separate and are inserted on the outer and
inner parts of the coxopodite respectively.

Extensors. One outer (o.f.m.) and one inner (i.f.m.)
The tops of the two pillar-like portions of the first post-
oral endopleurites are joined by a strand of tissue.
Beneath the arch thus formed the two muscles arise near
the middle line. They pass downwards and slightly
inwards, diverging somewhat as they approach the
appendage. They are inserted on the coxopodite at the
same point as the corresponding flexor muscle.

VI. Second maxilla. There are two extensors
and two flexors.

Extensors. The inner extensor arises from the
posterior face of the first post-oral endopleurite. It s a
short muscle which passes downwards and slightly
outwards, and is inserted on the outer side of the coxo-
podite.

The outer extensor is a long narrow muscle. It
arises from the epimeron of this somite just in front of the
last cephalic endopleurite. It passes inwards and down-
wards across the anterior face of the flexors of the
scaphognathite, and is inserted close to the small inner
extensor.

The two flexors are small, and arise close together
near to the origin of the outer extensor. They pass
directly downwards, and are inserted near together on the
inner side of the coxopodite.

68

The Scaphognathite (figs. 27, 28) has a complex
movement, and the plane of motion is roughly at right
angles to its long axis. There are two sets of muscles —
extensors which pull the organ downwards, and flexors
which draw it up again to its natural position. In the
upward movement the scaphognathite does not remain flat,
as when in a position of rest, but it becomes curved so that
the upper side is concave. This is effected by a set of
accessory muscles. The latter extend into the leaf-like
portion of the scaphognathite, and do not stop at the edge
of the organ, as do the other muscles.

All the flexors arise from the anterior face of the
last cephalic endopleurite. Their names have been given
according to the position of insertion on the scapho-
gnathite. The flexors are inserted on the anterior wall
of the base of the scaphognathite.

Inner flexor (i. e. s.) A long narrow muscle
arising from the upper part of the endopleurite. It
passes down the latter and, turning slightly inwards, it is
inserted on the innermost part of the base of the scapho-
gnathite. It has a small branch which arises from the
side of the epimeron.

Older flexor (p. e. s.) An extremely broad muscle,
which arises immediately beneath the origin of the
previous muscle and also on its inner side. It passes
down the endopleurite parallel to the epimeron, and is
inserted on the extreme outer edge of the base of the
scaphognathite.

Outer median flexor (o. m. e.) A long and fairly
broad muscle, arising from the extreme inner border of
the endopleurite above the large foramen of the latter. It
passes downwards and outwards across the front of the
foramen, and is inserted on the base of the scaphognathite
on the inner side of the previous muscle.

69

Inner median flexor (i.m.e.) A very short muscle
arising from the endopleurite near the middle of the base
of the foramen. It passes downwards below the other
flexors and is inserted at the base of the scaphognataite
on the outer side of the inner flexor.

The extensor muscles are situated beneath the flexors,
so that it is necessary to cut away the latter in order to
expose the extensors. There is one extensor corresponding
to each flexor, and the insertion of each extensor is near to
that of the corresponding flexor All the extensors arise
at the base of the anterior face of the last cephalic
endopleurite. They are all short muscles, and are
inserted on the posterior wall of the base of the
scaphognathite.

Inner extensor (i.f.) Arises at the base of the endo-
pleurite. It passes inwards and is inserted close to the
inner flexor.

Inner median extensor [I. m.f.) It lies immediately
below the inner median flexor, and above the three other
extensors. It arises about the middle of the base of the
endopleurite and passes downwards and forwards. Its
insertion is close to that of the inner median flexor.

Outer median extensor (o. m.f.) Its origin is on the
inner side of the previous muscle. It passes obliquely
outwards and forward beneath the previous muscle, and
is inserted close to the insertion of the outer median
flexor.

Outer extensor (<>./.) A very short and broad muscle
arising on the outer side of the origin of the inner
median extensor. It passes outwards and downwards and
is attached to the scaphognathite close to the insertion
of the outer flexor.

The last two muscles have their insertions situated
in the thickened bulb-like portion of the scaphognathite.

70

The accessory muscles are situated entirely within
the scaphognathite itself. Their function is to bend the
scaphognathite during the process of lifting up or extend-
ing the latter.

There are two muscles which arise close together on
the inner side of the insertion of the inner flexor of
the scaphognathite. They extend outwards into the
middle and inner portions of the scaphognathite.

The anterior accessory (a. ace.) divides into two parts,
each of which is attached to the anterior wall of the
scaphognathite.

The posterior accessory (p. ace.) also divides into two
parts. One division appears to be inserted on the
posterior wall and the other on the anterior wall.

VII. First maxillipede (fig. 29). Only the
extensor and flexor of the Coxopodite need be noted here.

The extensor (e. C.) arises on the epimeron of this
somite. It passes downwards and inwards between the
two flexors of the flabellum and behind the extensor of
the latter (see below). It is inserted near the outer and
posterior margin of the coxopodite.

The flexor (/. C.) arises at the upper side of the
posterior face of the last cephalic endopleurite on the
inner side of the point where the latter fuses with the
second thoracic endosternite. It passes directly down-
wards as a narrow muscle, and is inserted on the
anterior margin of the coxopodite.

The muscles of the exupudite have the same arrange-
ment as the similar parts in the third maxillipede.

The muscles of the flabellum are large and powerful.
The fiabellurn lies on the dorsal side of the gills, and by
repeated rhythmical movements keeps the surface of the
gills free from sand and mud.

71

The extensor [ex. ft.) is a short broad muscle
arising from the upturned edge of the sternum in this
somite. It passes outwards across the mouth of the cavity
of the coxopodite, and is inserted on the inner side of the
base of the fiabellum.

The anterior flexor (a.f.fi.) is a fairly broad muscle
arising from the posterior face of the last cephalic endo-
pleurite, immediately above the foramen of the latter. It
passes downwards, and is inserted on the anterior edge of
the base of the fiabellum.

The posterior flexor [p-f-fl-) is an extremely broad
muscle arising from the last cephalic endopleurite, above
the origin of the previous muscle. It runs downwards
behind the extensor of the coxopodite, and is inserted on
the posterior edge of the base of the fiabellum.

VIII. Second maxillipede. This appendage
is similar to the third maxillipede in structure, and its
muscles have the same arrangement (see below).

The extensor muscle of the coxopodite arises from the
inner side of the second thoracic epimeron, and the flexor
arises from the upper end of the anterior face of the
second thoracic endosternite. All the muscles of the
basi-ischium are attached to the lower end of the anterior
face of the second thoracic endosternite. The muscles of
the fiabellum are quite small, but have the same parts as
described in the first maxillipede.

IX. Third maxillipede (fig. 30).
Coxopodite. There is a small extensor and a larger

flexor.

The extensor muscle arises on the inner side of the
third thoracic epimeron, and is inserted on the outer side
of the coxopodite by means of a narrow tendon (ex. C).

The flexor muscle is attached to the anterior wall of

72

the third endosternite. It passes outwards, and is inserted
on a broad tendon (/. C.) at the inner side of the coxo-
podite.

Basi-ischium. There are two chief muscles.

The extensor is a small muscle arising from the
anterior wall of the third thoracic endosternite. It is
inserted on the ventral side of the basi-ischium by means
of a small tendon (ex. Ji.).

The flexor is a larger muscle arising near the origin
of the extensor. It is inserted on a long tendon (/. B.)
on the dorsal side of the basi-ischium. There is also a
small accessory flexor inserted on the outer side of the
larger flexor.

There is one flexor and one extensor for each of the
remaining segments of the endopodite. The muscles of
the meropodite and carpopodite are fairly large. Those
of the propodite and the dactylopodite are small.

The exopodite has two small muscles— a dorsal
extensor and a ventral flexor.

The flagellum of the exopodite is flexed in its
natural position. There is a large extensor muscle (ex. fl.)
running the whole length of the exopodite, which is
attached to the outer edge of the flagellum and by its
action raises the latter. I have not been able to make
out a flexor muscle. Probably the flae'ellum falls back
into its natural flexed condition by means of the elasticity
of the arthrodial membrane.

The flagellum in each of the maxillipedes is very
active, and is constantly moving with great rapidity.

X. Chela .* (fig. 21, Text fig. 10).

Coxopodite. There are two muscles — a posterior

* In all the pereiopods it is probable that those muscles which are
situated in the dorsal region of the pleural muscle chambers arise, not
only from the walls of the latter, but also from the carapace in this
region (PI. IX, fig. 5&,fl.m.)

73

extensor pulling the coxopodite backward and an anterior
flexor pulling it forward.

The extensor is situated in the outer and posterior
region of the fourth thoracic pleural muscle chamber. It
arises from the anterior and posterior walls of the latter
and passes forward and downward. Its insertion is on a
long- narrow tendon arising from the posterior side of the
coxopodite (fig. 21, ex. ('., Text fig. 10, d.).

The flexor is a much larger muscle than the extensor,
and lies in the fourth thoracic sternal muscle chamber.
It arises from three parts of the endophragmal system —
(1) from the posterior face of the third thoracic
endosternite ; (2) from the inner side of the third endo-
pleurite ; and (•'!) from the anterior face of the fourth
thoracic endosternite. It passes downward and forward,
and is inserted on an extremely broad tendon on the
anterior side of the coxopodite (fig. 21, /. C, Text fig.
10, c).

Basi-ischiopodite. There are two extensors and three
flexors inserted on the proximal region of this segment.

The anterior extensor is situated in the anterior and
ventral portion of the fourth thoracic pleural muscle
chamber. It arises from the ventral part of the anterior
and inner walls of the chamber. It runs outward, and is
inserted on a long and narrow tendon situated immediately
above the anterior hinge (fig. 21, a. ex. B, Text fig. 10, /.).

The 'posterior extensor is a small muscle situated
almost entirely in the base of the coxopodite. It is
attached to the ventral part of the wall of the fourth
pleural muscle chamber. It passes outward, and is
inserted on a small tendon which is immediately above
that of the anterior extensor (fig. 21, p. ex. li., Text fig.
10, e.)

The anterior flexor lies in the ventral region of the

74

fourth sternal muscle chamber, and arises from the fourth
thoracic sternum. Its course is outward, upward and
backward, and at its outer extremity it is inserted on a
broad tendon lying on the ventral side of the basi-ischium,
mid-way between the two hinges (fig. 21, a. f. B., Text
fig. 10, *.).

Fig. 10. — The proximal end of the chela, after all the soft tissues have
been removed. The tendons of the muscles moving the coxa and
basi-ischium are shown. (The upper side of the figure is dorsal,
and the right side is the anterior end.)

Cozopodite. — a. = dorsal hinge ; b. = ventral hinge ; c. = tendon of
flexor muscle ; d. = tendon of extensor muscle.

Basi-ischium. — e. = posterior extensor muscle; /. = anterior extensor
muscle ; g. = lesser posterior flexor muscle ; h. = greater posterior
flexor muscle ; k. = anterior flexor muscle; I. = posterior hinge ;
m. = anterior hinge.

The greater posterior flexor lies in the outer and
anterior portion of the fourth pleural muscle chamber,

75

and its origin is on the anterior wall of the latter. It
passes downward behind the anterior extensor, and is
inserted on an extremely long tendon which arises from
the joint immediately behind the tendon of the anterior
flexor (fig. 21, p. f. B., Text fig. 10, h.).

The lesser posterior flexor is situated on the inner and
anterior region of the fourth pleural chamber. It passes
outward, and is inserted on a small tendon in front of the
posterior hinge (Text fig. 10, g.).

The Meropodite has very little movement, and the
muscles are extremely small.

The extensor is a small muscle arising from the
posterior wall of the ischium. It passes downward and
backward, and is inserted on a small tendon on the ventral
side of the meros (fig. 21, t. ex. M.).

There does not appear to be a definite flexor muscle,
but when the extensor muscle relaxes, uie weight of the
distal portion of the limb is sufficient to produce the small
amount of flexion necessary.

Carpopodite. There is an anterior flexor and a
posterior extensor muscle.

The extensor arises from the posterior walls of the
meros throughout its entire length. The insertion of
the muscle is on a long tendon situated near the dorsal
hinge of the carpos. This tendon lies in the dorsal part
of the meros and extends almost to the proximal end of
the latter (ex. C.1).

The flexor has its origin on the anterior walls of the
meros. Its tendon is similar in size to that of the
extensor. It lies in the ventral region of the meros, and
arises from the antero-ventral border of the carpos (/. C.1).

Propodite. There is a posterior extensor and an
anterior flexor muscle.

The extensor arises from the posterior walls of the

76

earpos, and is inserted on a broad tendon at the posterior
side of the propodite (eat. P.).

The flexor has its origin on the anterior wall of the
earpos, and is inserted on a large tendon at the anterior
side of the propodite (J. P.).

Dactylopodite. There is a dorsal extensor and a
ventral flexor.

The extensor is comparatively small, and arises from
the dorsal or posterior walls of the propodite. It is
inserted on to a narrow tendon which is situated
immediately above and between the two hinges of the
dactylos (ext. D.).

The flexor is an extremely large muscle which
occupies the greater portion of the propodite. It arises
from the ventral and anterior walls of the latter, and is
inserted on a very broad tendon which is attached to the
ventral side of the dactylopodite (f. D.).

XL First walking leg (PI. Ill, fig. 22).

Coxopodite. There is a posterior extensor and an
anterior flexor.

The extensor is situated in the fifth pleural muscle
chamber. It passes downward, and is inserted on a long
narrow tendon immediately behind the dorsal hinge
{ex. C).

The flexor is a large muscle situated in the anterior
and upper portions of the fifth sternal muscle chamber.
Its origin is partly on the anterior wall of the chamber
and partly on the median plate. It passes outward and
downward, and is inserted on a broad tendon on the
anterior portion of the coxopodite (/. ('.).

Basi-ischiopodite. There is one dorsal extensor and
one ventral flexor muscle.

The extensor occupies the dorsal and posterior portion

77

of the fifth sternal muscle chamber. It arises from the
anterior face of the fifth thoracic endosternite, and passes
outward and downward, and is inserted on a long tendon
on the dorsal side of the basi-ischium (ex. B.).

The flexor lies in the ventral and posterior part of the
fifth sternal muscle chamber. It arises from the median
plate, and passes downward and outward. Its insertion
is on a fairly broad tendon on the ventral side of the joint
(/. B.).

In addition to the above flexor there is a small
accessory flexor muscle on each side of the former. They
are probably comparable to the anterior flexor and lesser
posterior flexor of the corresponding segment of the chela.

The muscles of the other segments of this limb are
very similar to those described above in the chela. The
two muscles of the dactylopodite, however, are extremely
small, and almost equal in size.

XII. Second walking leg.

The muscles here are similar to those of the first
walking leg.

Coxopodite. The extensor arises from the sixth
pleural muscle chamber. The flexor arises from the sixth
sternal muscle chamber.

In the Basi-ischiopodite both muscles arise from the
sixth sternal muscle chamber.

XIII. Third walking leg.

The muscles are similar to those of the two previous
appendages.

Coxopodite. The extensor arises from the seventh
pleural muscle chamber. The flexor arises from the
seventh sternal muscle chamber.

In the Basi-ischiopodite both muscles arise from the
seventh sternal muscle chamber.

78

XIV. Fourth walking leg (PL III, fig. 23).

In the coxopodite and basi-ischium there are the
same muscles as in the other walking legs. There is,
however, an additional extensor of the coxopodite. This
ventral extensor is inserted on a small narrow tendon
[v. ex. C.) immediately below the insertion of the dorsal
extensor.

The muscle chamber in this somite is not divided into
pleural and sternal regions. Hence it may be designated
the pleuro-sternal muscle chamber (see section on
Endophragmal System). All the muscles of the coxa and
the basi-ischium arise from this chamber.

The dorsal extensor of the coxopodite arises from the
anterior wall of the chamber ; the ventral extensor from
the median plate ; the flexor of the coxopodite from the
anterior end of this muscle chamber. The extensor of the
basi-ischium arises from the anterior and inner corner of
the chamber, and the flexor has its origin on the median
plate.

The muscles of the remaining parts of this appendage
are similar to those of the other walking legs.

In the chela the coxa swings horizontally. In the
first walking leg the coxa is slightly tilted, so that it
swings forward and slightly downward. In each of the
succeeding walking legs the corresponding part is more
tilted, and in the last walking leg the ventral hinge of the
coxa is more posterior and the dorsal hinge anterior. So
that, instead of swinging horizontally as in the chela, the
coxa swings in a plane inclined at a considerable angle to
the vertical, and during the movement of extension the
limb is capable of being turned almost on to the dorsal
side of the carapace. The presence of the additional
extensor muscle undoubtedly aids such a movement.

This freedom of movement is probably not of much

79

value in Cancer. In the swimming crabs, however, where
the last thoracic appendages are flattened and oar-like
and are utilised as an effective rowing organ, such an
arrangement is of no mean importance.

Muscles of the fore-gut. These are
described in the section on the Alimentary Canal.

The Dor so-ventral muscles are described
in the section on Respiration.

It is of interest to note in passing that in the case of
those muscles arising from the carapace, there are definite
marks on the outside of the shell corresponding in shape
and size to the areas of the origin of these muscles. It
has been stated above (section on Integument) that the
muscle is not directly attached to the chitinous exo-
skeleton, but arises from the basement membrane under-
lying the epidermis. How, then, can the marks of the
muscle attachments be duplicated on the outer side of the
exoskeleton ?

The most probable explanation is that in those
regions where the muscles are attached to the basement
membrane the cells of the epidermis are in some way
affected by the underlying muscles. In this manner the
rate of secretion of the integument may have been slightly
reduced in these localised areas, thus producing the marks
on the outer side of the carapace.

Muscles of the Abdomen.

On account of the third, fourth and fifth somites
being fused together, the abdominal muscles of the
male abdomen differ from those of the female.

Female abdomen (PL IV, figs. 32, 33).

There are two sets of muscles working each somite,

80

a pair of dorsal extensors, and a pair of ventral flexor
muscles.

Somite I. Each extensor {ex. 1) arises as a broad band
of muscle from the top of the epimeral region of the last
two thoracic somites. It passes inward and backward,
and is inserted on the side of the anterior triangular
portion of the tergum of the first somite.

Each flexor (/. 1) muscle arises from the "sella
turcica " in the thorax, and passes backward near the
median line. It is inserted on a small ingrowth of the
sternum immediately in front of the arthrodial membrane
separating the first sternum from the second.

Somite II. Each extensor [ex. 2) is inserted on a
small concavity in the tergal region of the first somite.
It passes backward, and is inserted near the middle line
on a tendon which is attached to the anterior extremity of
the second somite.

Each flexor (/. 2) arises from the posterior face of the
ingrowth, or tendon, at the posterior end of the first
sternum. It passes backward near the median line, and
is inserted on the anterior face of a similar tendon at the
posterior extremity of the second sternum.

Somite III.-YI. and Telson. The muscles in the
succeeding abdominal somites have the same arrange-
ment as those of the second somite. The flexor of the last
abdominal somite is not inserted on a well-marked tendon.
The same applies to both the origin and insertion of the
flexor muscle of the telson.

Uropod of female (PL IV, fig. 25).

In their natural position the uropods are extended
and lie almost horizontal, with their distal extremities
pointing toward the posterior end of the abdomen.

The Protopodite is capable of two kinds of move-
ment. First there is a movement in an antero-posterior

81

plane. The flexor muscle, which has very little power,
moves the protopodite forward, and the extensor acts in
the opposite direction.

The extensor (ex. prot.) arises from the posterior
region of the tergum of the somite to which the uropod
belongs. It passes forward and downward, and is inserted
on a small tendon on the anterior wall of the protopodite,
some distance below the proximal border of the latter.

The flexor (/. prot.) arises from the anterior region of
the tergum. It passes downward and inward, and its
insertion is on a tendon arising from the outer border of
the protopodite.

The protopodite also has a slight movement from side
to side. The flexor muscle draws the appendage toward
the middle line and the extensor pulls it outward.

The lateral extensor (I. ex. prot.) arises from the outer
region of the tergum, and passes downward and inward.
Its insertion is on a tendon arising from the outer border
of the protopodite.

The lateral flexor (I. f. prot.) arises from the inner
part of the tergum. It passes downward and outward, and
is inserted on a tendon arising from the inner border of
the protopodite.

Exopodite. The movement of the exopodite is lateral.
There is one extensor and two flexors.

The extensor (ext. ex.) arises from the outer edge of the
protopodite. It passes downward, and is inserted on the
inner wall of the exopodite some distance below the
arthrodial membrane.

The flexors (d. f. ex., v. f. ex.) arise from the inner wall
of the protopodite. They pass outward, and converge to
a single insertion on the inner edge of the exopodite.

The endopodite is fused to the protopodite, and has no
muscles.

82

M a I 6 a 1» d 0 in <' n •

There are no extensors between the third and fourth
somites, and also between the fourth and fifth somites.

According to Williamson, there are only two Long
flexors a1 each side arising from the thorax. One is
inserted on the sternum of the united third, fourth and
fifth somites, and the other is inserted on the telson.

Uropods of Male. These muscles have been described
by Williamson.* In the ftrsl appendage llie endopodite
has n strong flexor muscle. The extensor is extremely
small, and probably the flexion is effected by the elasticity
of the arthrodial membrane. The protopodite lias two

small muscles, one of which Ilexes and the other rotates

the Limb. In the second appendage there is also ;i strong
flexor in the endopodite. I he protopodite has a system of
small muscles which rotate, extend and Ilex the appendage.

Eistology of Muscle.

The in use I es of Cancer ;i re coin posed of striated fibres.

Each fibre is an elongated multi-nuclear cell which
reveals, in Longitudinal sections ;ind in stained prepara-
tions, two kinds of striations Longitudinal and transverse.
As a rule, the cross striations are the more obvious, and

produce the "striped" appearance so characteristic of
A rl hropod m usele fibres.

Each fibre is composed of numerous Longitudinal
fibrils, which give rise to the Longitudinal striations. In n

transverse section across a fibre it is seen thai the filu'ils

have an unequal distribution, and arc usually grouped

together into polyhedral areas (Cohnheim's areas). The
bundle of fibrils constituting a single area is known as a
muscle column. The various muscle columns are separated

Williamson, II. ('. Twenty-second Annual Report of the
Fishery Board for Scotland, p. lot.

88

from one another by the sarcoplasm (protoplasm), which
varies in quantity in different kinds of fibres.

In stained preparations the muscle fibre reveals
alternate light and dark cross-striations. At its centre,
each light band is interrupted by a transverse line
(Krause's membrane). There is also a transverse line
stretching across the middle of the dark band (Hensen's
line). The latter is only seen with difficulty. Each
portion of a fibril between two adjacent Krause's
membranes is known as a " sarcomere."'

Haycraft's* experiments led him to believe that the
cross striations are due to regularly-occurring varicosites,
and some of the preparations made in the course of the
present work appear to show this. It is doubtful whether
this structure (even if admitted) is alone sufficient to
account for the fact that the cross-striations seen in fresh
tissue are accentuated under the action of various
staining reagents. It is highly probable, as suggested by
Schafer, that the cross striatum of the fibrils is due to the
heterogeneous nature of the latter.

An examination of the fresh muscles of a crab reveals
the interesting fact that — as in the vertebrates -some of
the muscles are of a darker colour than others. Sections
across the muscle fibres show that, generally speaking, the
" dark " muscles have much more sarcoplasm than the
" light " ones. Hence Knoll distinguished between
plasmic (dark) and aplasmic (light) fibres. Biedermannf
has shown that there is a definite relationship between the
amount of sarcoplasm present in a fibre and the nature of
the work performed by the fibre. He has, furthermore,
stated that " the elements of those muscles which serve

* Haycraft, J. B. "Cause of Striation of Voluntary Muscular
Tissue," Q.J. M.S., Vol. XXI (1881).

t Biedermann, W. Electro-physiology , Vol. I,

84

the most persistent or most strenuous action are richest in
sarcoplasm."

In Cancer the muscles present all grades of colour,
from an opaque yellowish brown (muscles of scapho-
gnathite) to a transparent white .(muscles of appendages).
Undoubtedly, the most active muscles of the body are
those of the seaphognathite, and probably the flexors and
extensors of the abdomen are the most sluggish (in the
Macrura the abdominal muscles are very strenuous). I
append a list of muscles, commencing with the most
strenuous and darkest in colour and finishing with the
least active. In all cases the position of a muscle on the
list for colour agrees with its position regarding its
activity.

1. Muscles of the scaphognathite.

2. Muscles of the heart.

3. Mandibular muscles.

4. Anterior cardiac muscles.

5. Extensor muscles of flagella of maxillipedes.

6. Gastric muscles (other than the anterior cardiacs).

7. Muscles of the appendages.

8. Extensor and flexor muscles of the abdomen.

COELOM AND BODY CAVITY.

Arthropods in general are characterised by the
presence of a greatly reduced coelom in the adult. This
reduction of the coelom is the result of the increase in the
blood-holding spaces or sinuses. This system of swollen
sinuses, which contain the venous blood, has produced a
series of cavities lying between the various organs of the
body, and has been termed by Lankester a haemocoel. The
theory of Phleboedesis formulated by Lankester to account
for the development of the haemocoel is as follows : —

85

" The system of blood-containing spaces pervading the
body in Mollusca and Arthropoda is not, as sometimes
supposed, equivalent to the coelom or perivisceral space of
such animals as the Chaetopoda and the Yertebrata, but
is in reality a distended and irregularly swollen vascular
system — the equivalent of the blood- vascular system of
Chaetopoda and Yertebrata." ,T

In Cancer the only remnants of the true coelom are
the gonadial sacs and the end sacs of the antennarv
glands. The labyrinth and bladder of the excretory
system are lined by cells derived from epiblast.

ALIMENTARY CAXAL

(Pis. V, YI, VII).

The alimentary canal extends from the mouth, which
is situated on the ventral side of the cephalic region
between the mandibles, to the anus on the ventral side
of the telson. The nature of the development of the
alimentary canal suggests a natural division into three
parts: — (1) The fore-gut, which is the embryonic
stomodaeum, (2) the mid-gut, the archenteron of the
embryo, and (3) the hind-gut, which is the embryonic
proctodaeum.

Fore -.gut.

The fore-gut commences at the mouth and is formed
of the embryonic epiblast. It is lined throughout by a
cuticle which is continuous with the exoskeleton around
the mouth. The mouth leads into a short oesophagus
(PI. YI, fig. 40, oe.) which opens into the so-called
" stomach," which is continuous behind with the mid-gut.

The mouth is situated on the ventral surface of the

* Lankester, E. Ray. " The Enterocoela and the Coelomocoela,"
A Treatise on Zoology, Part II.

86

cephalic region, and is bounded in front by a fleshy lobe — ■
the labrum (PI. Ill, figs. 18, 20, lab.) or anterior lip —
and behind by the metastoma {met.) or posterior lip. On
each side of the mouth are the mandibles.

Both the labrum and metastoma have closely packed
glands which have the appearance and structure of the
" salivary glands " found in the walls of the oesophagus.
It is not inconceivable that they have the same function
as the oesophageal glands. The mandibles also have at
their base a mass of glands which are continuous with
those in the ventral portion of the oesophageal walls.

In the oesophagus (PL V, fig. 35) the epidermal cells
[eh. ep.) are of great length. In a soft crab with a
carapace 25 mm. in width these cells are 90 /* in length
and only 3 n wide. On the outer side of the epidermis
is a thin chitmous layer about 8 /* in width. This consists
of two layers — a thin outer structureless layer, the cuticle,
and a broader inner layer showing evidences of longi-
tudinal striations. On the inner side of the epidermis is
a well-marked basement membrane. Below the basement
membrane is a layer of connective tissue (der.) about 370 n
in width. This is composed of a dense reticulate mass
formed of intercrossing connective tissue fibres. There
are also small connective tissue cells scattered about.
Embedded in the connective tissue are numerous glands
which may conceivably be salivary glands, but which I
designate the oesophageal glands (sal. g.).

On the outer side of the connective tissue of the
oesophagus is a layer of circular muscles — the constrictors
of the oesophagus (c. oe.) — and passing through the
connective tissue and attached to the basement membrane
are numerous muscle bundles — the dilators of the
oesophagus (oe. I.).

Each oesophageal gland is globular and consists of

87

numerous large conical cells, the apex of each cell
pointing to the centre of the mass. Each cell has a well-
defined nucleus near its outer side. In the centre of each
gland mass is a small cavity into which the secretion from
the individual cells is poured. This small central cavity
is connected with the lumen of the oesophagus by means
of a long narrow duct which passes between the cells of
the epidermis. The duct and its walls is probably formed
of a single cell, in which case the gland duct is intra-
cellular. These glands are scattered through the
connective tissue of the oesophageal wall. They take the
stain distinctly, and have a diameter of 25 n to 35 // in
the small crab mentioned above. From the above
description it would appear that the oesophageal glands
are merely modified cutaneous glands.

At the extreme ventral end of the oesophageal wall
at each side there is an additional mass of such glands
which are very closely packed together (v.oe.g.). As
mentioned above, these glands are continuous with those
of the mandible.

The large and spacious region of the fore-gut which
follows the oesophagus is generally termed the
" stomach,"* and is divided into a large anterior portion
— the cardiac chamber (card.) and a smaller posterior
portion — the 'pyloric chamber (pyl.).

The Cardiac fore-gut (PI. V, fig. 34, PI. VI, fig. 40,

curd.) is a large simple sac roughly spherical in shape.

The cuticle lining this part of the alimentary canal

* The term "stomach" is an unsatisfactory one, as this part of
the fore-gut is neither embryologically nor physiologically what is
generally recognised as a stomach. Also the terms "cardiac" and
" pylroic " have no meaning when applied to the Malacostraca, seeing
that the cardiac region — as pointed out by Huxley — is the farthest
from the heart. It would be inconvenient, however, to reject the
terms " cardiac " and " pyloric," as such a change would also involve
an alteration in the names of numerous ossicles and muscles connected
with the fore-gut. The terms " cardiac fore-gut " and " pyloric fore-
gut " will be used in the present Memoir.

88

presents numerous thickening's to form an elaborate
system of plates and teeth known as the " Grastric Mill."
Tli is will be described below.

The posterior wall of the cardiac fore-gut is
invaginated on its ventral surface to form the cardio-
pyloric valve separating the cardiac region from the
pyloric" region.

The Pyloric fore-gut (PI. Y, figs. 34, 36) presents a
very complicated arrangement.

In the posterior two-thirds of the pyloric region the
cliitin of the ventral wall is thickened at each side to
form the pyloric ampullae (amp.). These are clearly
seen from the outside as swellings on the floor of the
pyloric region. The ampullae have their chitinous lining
thrown into well-defined longitudinal parallel ridges.
From the summits of the ridges there are numerous fine
setae projecting into the cavity of the pyloric chamber.
Each seta has numerous small hook-like branches. The
two ampullae meet in the mid-ventral line in a well-
defined ridge — the inter-ampullary fold (PL Y, fig. 36,
i.a.f.).

The ventro-lateral walls immediately above the
ampullae have the chitin enormously thickened at each
side to form cushion-like pads projecting into the cavity
of the pyloric chamber immediately above the ampullae.
These are the supra-ampullary ridges [ " voiite arnpul-
laire," Mocquard] (amp. c). Each cushion has a convex
surface which faces inwards and downwards, and the
upper parts almost meet in the middle line. The presence
of the supra-ampullary ridges and the inter-ampullary
fold causes the cavity and the ventral part of the pyloric
chamber to be reduced to a narrow two-rayed fissure. The
supra-ampullarjr ridges are covered with numerous fine
setae, which stretch across the narrow lumen of this

89

portion of the pyloric chamber, so as to form — as Huxley
suggested — a very effective filtering apparatus.

The dorsal part of the pyloric chamber has a
comparatively large cavity. In transverse sections
through the pyloric region the lateral walls of the dorsal
portion are roughly at right angles to one another.

The supra-arnpullary wall (fig. 36, s. amp.) is
immediately above the supra-ampullary ridges, and is
almost horizontal, thus forming the floor of the upper
region.

The pleuro-pyloric wall (pp.) is on the outer side of
the supra-ampullary wall and turns upwards almost at
right angles to the latter. This portion of the wall may
be complicated by the presence of folds (up. /.).

The dorsal wall is simple in structure.

Thus in the posterior two-thirds of the pyloric region
the lumen is divided into a wide dorsal portion and a
narrow ventral portion, the two parts being capable of
complete separation by the concrescence of the inner
portions of the supra-ampullary ridges.

The anterior third of the pyloric region is compara-
tively simple, and shows no such division into dorsal and
ventral portions.

It is probable that in the anterior part of the pyloric
region the contents undergo a certain amount of separa-
tion. For instance, any hard shell-like structures
belonging to the creatures taken in as food will be
separated from the soft and nutritious parts. The hard
parts pass backwards along the wide dorsal chamber, and
by means of an elaborate system of valves they are carried
directly into the hind-gut without coming into contact
with the unprotected walls of the mid-gut (fig. 40, vol.).
The valves are flap-like structures projecting backwards
from the upper side of the posterior end of the fore-gut.

90

As suggested by Huxley and Mocquard the function of
these valves may be partly to prevent the waste matter
from passing back into the fore-gut, but Cuenot has
claimed that the valves are also used, as described above,
for carrying the hard waste pieces directly into the hind-
gut. The mid-gut is not lined with chitin, and con-
sequently the sharp pieces present amongst the food in
the alimentary canal would be liable to tear the walls of
the mid-gut.

The soft parts of the aliments are passed through the
narrow ventral portion of the pyloric region, where they
are sieved by the setae stretching across the lumen, and
near the posterior region of this region the food comes
into contact with the secretion from the digestive glands.
The latter open into the ventro-lateral wall of the
mid-gut immediately behind the ampullae, and the
digestive fluid flows forward and mixes with the food in
the ventral part of the pyloric region.

In the cardiac and pyloric regions Ave have essentially
the same histological arrangements as in the oesophagus.
The epidermis consists of columnar epithelium of much
less length than the cells of the oesophageal epidermis.
The chitin ous layer is very thin except in the regions of
the ossicles of the gastric mill, which are merely thick
portions of the chitin which have become strongly
calcified. The basement membrane is well marked and
the connective tissue is a very thin layer. Embedded hi
this layer are very thin bauds of circular and longitudinal
muscles.

M i d - g u t for Mesenteron) (PL Y, fig. 34, m. g.).

The mid-gut is an extremely short portion of the
alimentary canal, being only about 10 mm. long in a
full-grown crab. This is the only part of the alimentary

n

canal which is derived from the archenteron and is lined
by cells formed from the hypoblast.

From the mesenteron arise a pair of caeca — -the
so-called " pyloric caeca." This is an unfortunate
designation, as they do not arise from the pyloric region
of the fore-gut. I therefore propose to substitute the
name of Mid-gut caeca (fig. 34, caec). Each caecum
arises from the side of the anterior part of the mid-gut.
It passes forward as a narrow tube alongside the pyloric
chamber, and is closely applied to the postero-lateral
region of the cardiac chamber of the fore-gut. On a level
with the widest part of the latter the caecum terminates
in a much convoluted portion.

The digestive glands (fig. 3-1, di.gl.) arise at each
side from the ventro-lateral region of the mid-gut,
immediately behind the origin of the mid-gut caeca.
These will be described more fully below.

The epithelium lining the mid-gut (PI. X, fig. (il)
consists of columnar cells having a length of 55 /j- in an
adult crab. There is no cuticular lining to the epithelium
of this region, but each cell has an outer striated border
from 1 fj. to 2 n in thickness. This is similar to the
" Harchensaum " present in the mid-gut of Anurida* and
in the duodenum of many vertebrates. It is very probable
that this striated border is characteristic of the mid-gut
epithelium of arthropods in general, and probably the
thin cuticle lining the mid-gut of Ligia described by
Hewittf is a similar structure.

In many of the epithelial cells are refractive bodies,
probably the fat globules mentioned by Cuenot. Beneath
the epithelium is a thick basement membrane. In the
comparatively broad layer of connective tissue beneath
the basement membrane are thin layers of circular and
longitudinal muscles.

* Imms. L.M.B.C. Memoir, " Anurida."

t Hewitt. L.M.B.C. Memoir, " Ligia.'"

92

Hind -gut (PI. Y, fig. 34, h.g.)

The hind-gut or intestine is a long narrow tube
extending from the posterior end of the mid-gut to the
anus which opens on the ventral surface of the telson.
Near to the mid-gut it passes below the median bridge-
like portion of the reproductive organs : passing further
back it runs beneath the pericardium, and a short distance
behind the latter it enters the abdomen, along which it
pursues a straight course. Just before entering the
abdomen the hind-gut gives off from its right side a long
coiled tube — the hind-gut caecum (fig. -34, i. caec).

The caecum lies above the hind-gut, and the coils,
which are packed very closely together, extend into the
first segment of the abdomen.

The hind-gut has very pronounced columnar
epithelium (PI. X, fig. 62). This is lined by a thin
chitinous layer, consisting — as in the fore-gut — of an
outer cuticle, and a layer longitudinally striated which
appears to be continuous at the anus with the pigment
layer of the exoskeleton. The epithelium rests upon a
basement membrane outside of which are thin layers of
circular and longitudinal muscles.

In the walls of the hind-gut immediately behind the
mid-gut there are closely packed glands very similar in
structure to those present in the walls of the oesophagus.

There are also glands, having a similar structure to
the above, present in the walls of the hind-gut in the
abdominal region. They are not closely packed (fig. 62).

Digestive Gland.

The. digestive gland ("liver," " hepato-pancreas ")
(PI. Y, figs. 34, 37, 38, 39) is a large yellowish-brown*

* This colour is due to lipochrome. (Miss Newbigin, Journ.
Physiol., Vol. XXI., p. 237, 1897.)

93

organ occupying nearly the whole of the ventral side of
the anterior region of the cephalothorax. It is a lolralated
structure composed of a large number of digitate tubular
outgrowths, and arises from the mid-gut at each side.
The front edge of the gland sweeps backward close to the
antero-lateral border of the carapace and resembles the
latter in having a notched edge. The posterior border of
the gland is generally on a level with the anterior region
of the branchial chamber : in other words, the branchial
chamber is only covered b}' the digestive gland at its
anterior end. Posteriorly the gland occupies the ventral
part of the region between the muscles of the thoracic
walking legs, and below the pericardium and hind-gut.
Throughout the digestive gland is covered by the gonads
(PI. VIII, fig. 51). The gland does not extend into the
abdomen.

Arising from the mid-gut at each side there are three
main ducts which communicate with smaller ducts.
These branch repeatedly, and ultimately end in the
cavities of the tubules of which the main part of the
gland is composed. Thus the tubules have a cavity which
is continuous with that of the mid-gut, and both the
ducts and the tubules are lined by cells derived from the
embryonic hypoblast. Each of the main ducts mentioned
above receives the digestive ferments from one of the
three main lobes into which each half of the gland is
divided. These lobes are as follows (fig. 34) : — (1) An
antero-lateral lobe having its outer border marked by
notches which correspond to the markings of the antero-
lateral border of the carapace; (2) a postero-lateral
lobe lying above the anterior part of the branchial
chamber ; and (3) a posterior lobe which lies between the
muscles of the thoracic walking legs.

94

Histology of the Digestive Gland
(PI. V, figs. 37, 38, 39).

In sections the tubules are seen to be closely packed
together, generally being separated only by a very thin
layer of connective tissue (fig. 37, c. t.) or a small blood
sinus. Sometimes, however, the walls of the tubules are
not separated from each other by any tissue. The lumen
of the tubules has four angles in transverse section, and
the cells at the angles are much shorter than the others.

The ducts are lined by a single layer of large
columnar and non-glandular cells. In the sections
stained with methyl-blue eosin these cells take the stain
more readily than the cells of the tubules.

The tubules (fig. 37) have three kinds of cells.

(1) Fat cells (fig 37, /.«,, fig. 38). These are
columnar cells from 70 f* to 120 /x in length and
15 /u wide. The contents of the cells are vacuolated, due
to the presence of fat globules (fig. 38, g. /".). The border
of each cell in contact with the lumen of the tubule is
striated (sb.). The nucleus (/?.) is generally situated in
the inner portion of the cell.

(2) Ferment cells (fig. 37, fm. c, fig. 39). These are
not quite as long as the fat cells, but they are about four
times as broad. Each cell contains a large globular mass
(tig. 39, /. v.) which nearly fills the whole cell. These
masses are yellowish-brown and are responsible for the
characteristic colour of the digestive gland. According
to Frenzel each mass is enclosed in a bladder, and the
vesicles are more abundant during feeding time than
during the fasting periods. On the side of the cell in
contact with the lumen of the tubule there is a small
amount of vacuolated protoplasm which exhibits striation.
The border of the cell in contact with the lumen was

95

described by MacMunn as being- ciliated. This is highly
improbable, and it is more likely that we have in both
the ferment cells and the fat cells a striated hem
(Harcheusauni) (sh.) similar to that already described in
the mid-gut epithelium. The nucleus is situated in that
portion of the cell farthest removed from the lumen.

(3) Young cells (fig. 37, y. c). These are small cells
found between the larger cells near the periphery of the
tubules. These young cells stain deeply and will
eventuallv give rise to the fat cells or ferment cells.

Physiology of the Digestive Gland.

As pointed out by Cuenot, the digestive gland has
many functions, which may be summarised as Digestion,
Absorption, Excretion, Elimination and Regulation.

Digestive function. According to MacMunn,
Frenzel and others the gland is a pancreas, and the
ferments produced (proteolytic and amylolytic) are poured
into the ducts of the gland and thence into the mid-gut.
The ferments are produced entirely in the ferment cells.

The fat cells have the power of forming and storing
fat.

Roaf* found that the action of the extract of the
digestive gland was as follows: —

It does not digest coagulated white of egg. It
digests fibrin most actively in alkaline solution, but not
actively in acid solution. It converts starch into sugar,
and inverts cane sugar. It does not hydrolise olive oil,
but it hydrolises methyl acetate.

Function of absorption. According to Cuenot the

digestive gland is of great importance as an accessory

organ for absorbing the products of digestion. The

-;-Roaf, H. E. "A Contribution to the Study of the Digestive
Gland in Mollusca and Decapod Crustacea." Bio-Chemical Journal,
Vol. I, Nos. 8 and 9.

96

inid-gut is the only portion of the alimentary canal not
lined with chitin, and therefore the absorption of the
soluble products of digestion can only take place in this
region. It is inconceivable that the short mid-gut, even
with the mid-gut caeca, can be the only region where the
process of absorption is carried on. The digestive gland,
which is merely an outgrowth from the mid-gut, is richly
supplied with blood, and it is an easy matter for the fluids
to pass from the mid-gut into the tubules and through
the cells into the blood stream. Thus the digestive gland
becomes an accessory absorptive organ of no mean
importance.

Excretory function. It was observed by Cuenot and
MacMunn that when a Crustacean was injected with
certain colouring matters, the latter were discovered in
the ferment cells of the digestive gland as well as in the
cells of the recognised excretory organ. Cuenot is of the
opinion that the pigment contained in the excretory cells
is of an excretory nature, and that when the contents of
these cells ultimately find their wray into the alimentary
canal, the excretory pigment becomes separated from the
ferments and passes down the hind-gut to the exterior.

Function of elimination. During the process of
absorption, Cuenot states that the cells of the digestive
gland keep back many useless products which are after-
wards carried to the exterior together with the excretory
products. This is quite distinct from the excretory
function.

Function of regulation. In addition to the other
functions it is probable that the digestive gland is capable
of regulating the composition of the blood, especially with
regard to the quantity of water contained in the blood.

Summary. As the food enters in at the mouth
it will come into contact with the secretion from the

97

oesophageal glands. In the cardiac region of the fore-gut
the food is broken up in a very effective manner. Passing
back into the pyloric chamber, the food encounters the
cardio-pyloric valve. Here the large pieces are prevented
from passing into the pyloric chamber. The food which
passes into the latter chamber probably undergoes a
further process of sifting, the useless material passing
along the dorsal portion of the pyloric chamber and the
food being passed along the ventral portion. In this
ventral region the food first comes into contact with the
digestive ferments. Both are well mixed by the action
of the muscles of the p}Toric chamber.

As already stated, the probable regions of absorption
are the mid-gut, mid-gut caeca and the tubules of the
digestive gland. The waste products pass down the long
hind-gut to the exterior.

Ossicles of the Fore-Out.
(PI. VI, figs. 40, 41, 43, 44.)

In certain regions of the fore-gut the chitinous lining
is thickened and strongly calcified to form ossicles. These
ossicles give attachment to muscles. One set of ossicles
in the dorsal and lateral walls of the cardiac region are
connected with three tooth-bearing ossicles. This system
of plates which is worked by the anterior and posterior
gastric muscles (see section on Muscles of the Fore-gut)
forms a very effective apparatus for breaking up the food
which has passed into the cardiac fore-gut. Hence the
name gastric mill.

In addition to the ossicles of the gastric mill there
are " supporting ossicles "' in both the cardiac and pyloric
regions. To these supporting ossicles are attached the
various muscles of the fore-gut.

98

Ossicles of the Gastric Mill.

The Mesocardiac Ossicle (r/i.c.) is a small median
ossicle in the dorsal wall of the cardiac region. It is
triangular in shape, with the apex pointing forwards. It
is not clearly separated from the urocardiac ossicle which
passes posteriorly, and it is only partially separated from
the pterocardiac ossicle which extends laterally. The
ossicle is thicker dorso-ventrally at its posterior end, and
to the thickened posterior edge the anterior ends of the
cardio-pyloric muscles are attached. In the Macrura and
the Anomura the mesocardiac ossicle is much larger than
in the Brachyura, and the pterocardiac pieces are much
smaller.

One pair of Pterocardiac Ossicles (pt.c). They are
situated to the right and left of the mesocardiac piece and
in contact with it. The posterior border is almost
straight, and the anterior border is curved. Each ossicle
is broadest on its inner side, and tapers towards its outer
extremity. Near the inner border of each ossicle is a
smooth area where the anterior gastric muscle is inserted.
Each ossicle extends outwards, and its outer extremity
articulates with the zygocardiac ossicle by means of the
antero -lateral ligament (lig.).

One pair of Zygocardiac Ossicles (z.c.) lying in the
supero-lateral wall of the cardiac region of the
fore-gut. Each passes backwards and inwards, and
comes into contact at its posterior end with the exopyloric
ossicle, thus forming a connecting link between the
ossicle of the cardiac and the pyloric regions. The
zygocardiac ossicle is irregular in shape. The anterior
part is rod-like, but the ossicle becomes gradually broader
as it passes backwards, and the posterior portion is a broad
rectangular plate which bears the lateral tooth. One side

99

of the ossicle points inwards, and the other faces
outwards. The inner face is concave and the outer face is
convex. The inner edge of the ossicle folds outwards so
as to produce a deep groove on the outer side below the
convexity. The ossicle has four borders. The anterior
border is concave and terminates at its posterior extremity
in the large anterior tooth. The dorsal border, which can
be seen through the dorsal wall of the stomach, is also
concave. It passes backwards and inwards and ends at
the posterior border. The posterior border has a large
indentation into which the anterior border of the
exopyloric ossicle fits. The inner border lies obliquely,
being nearer the middle line at its anterior end. The
ossicle appears to be much thicker at its inner border than
in any other region. This thickness is not real, but is
merely due to the ossicle folding outwards at its inner
border. This inner border bears the de?iticles. Anteriorly
there is a large single denticle, which is followed by about
seven smaller denticles, which point inwards and decrease
in size from before backwards. The folded edge of the
inner border is crossed by about twenty-four transverse
ridges. This system of denticles and ridges on the
zygocardiac ossicle is known as the lateral tooth (lat. t.).

The Exopyloric Ossicles (ex. py.) are a pair of small
triangular plates, each of which lies between the posterior
border of the zygocardiac ossicle and the pyloric ossicle.
The superior border gives support to the posterior end of
the outer part of the cardio-pyloric muscle, and on its
external face it provides insertion for the external part
of the posterior gastric muscle.

The Urocardiac Ossicle (u.c.) is a median plate more
or less fused with the mesocardiac ossicle in front. It
passes backwards and downwards as a broad, thin
rectangular plate. At its posterior end, which articulates

100

with the propyloric ossicle, it bears the large blunt
median tooth (med. t.) on its ventral surface.

The Propyloric Ossicle (pr.p.) is a small median plate
situated almost vertically. When the gastric mill is at
rest the lower end of this ossicle is considerably behind
its upper end. Its upper end articulates with the front
portion of the pyloric ossicle, and its lower end is in
contact with the posterior end of the urocardiac ossicle.
The plate is roughly triangular in shape and its apex,
which points downwards, is bifurcated. The base of the
triangle is dorsal and is extremely concave. The ossicle
is highly calcified around the edges, but in the centre it
is almost membranous.

The Pyloric Ossicle (o.py.) is a median ossicle lying
between the two exopyloric ossicles which articulate with
it at each side. It covers the anterior part of the pyloric
region of the stomach. Its central portion is membranous,
but laterally it is slightly calcified. These lateral
calcifications indicate that the pyloric ossicle is really a
paired structure. In the Macrura all signs of the double
origin disappear.

Cardiac ''Supporting Ossicles.''
The Pectineal Ossicles (pec.) are a pair of irregular
hammer-shaped ossicles, each lying in the lateral wall of
the fore-gut beneath the posterior portion of the
zygocardiac ossicle. The curved " handle " of the hammer
points anteriorly. On the inner side of the " head " of the
hammer are three claw-iike teeth. These are the lateral
accessory teeth (a.t.l.). | "" Infero-lateral cardiac teeth,"

Huxley.]

The Prepectineal Ossicles (p>.ptc.) are a pair of long
narrow rod-like ossicles, each being concave on its inner
border and extending upwards from the pectineal ossicle

101

to the outer edge of the zygocardiac ossicle with which it
articulates by means of a ligament.

One pair of Post-pectineal Ossicles (pt. pec.) Each is
a narrow rod-like ossicle which passes backwards from the
pectineal ossicle to the posterior wall of the stomach. It
then suddenly turns downwards and runs down the
posterior wall of the cardiac fore-gut as a straight rod.
At its lower end the ossicle turns forwards for a short
distance. On the internal border of the ossicle there is a
row of setae projecting into the stomach.

The Infero- lateral Cardiac Ossicles (?'./.) are a pair of
long rod-like ossicles, each of which lies immediately
behind and parallel to the rod-like portion of the post-
pectineal ossicle. Dorsally the ossicle is in contact with
the sub-dentary ossicle, and ventrally it terminates on a
level with the lower end of the post-pectineal ossicle.
The ossicle is broader at its upper end and tapers
gradually towards its lower extremity.

There is one pair of Sub-dentary Ossicles (s.dt.) At
its anterior end each ossicle is in contact with the inner
border of the zygocardiac ossicle. The ossicle passes
downwards and backwards as a somewhat curved rod, and
its posterior end touches the upper end of the infero-
lateral cardiac ossicle.

The Lateral Cardio-pyloric Ossicles are a pair of small
ossicles articulating with the posterior and upper end of
the infero-lateral cardiac ossicles.

Postero-lateral Cardiac Plates (cd. pi.). These are a
pair of broad plates roughly quadrangular in shape, each
lying in front of the post-pectineal ossicle. It is a
membranous area having no decided calcification, but
being distinctly thicker than the ordinary wall of the
stomach. There are two rows of long setae arranged along
the posterior edge of each plate and projecting into the
cavity of the stomach.

10"2

The Antero-lateral Cardiac P\a,tes (cd.al.) are a pair of
thickened areas in the side walls of the stomach in front
of the posterolateral cardiac plates, but they are not so
well denned as the latter.

The Cardio -pyloric valve (c.p.v.) is the thickened
median portion of the posterior wall of the cardiac
fore-gut. Its upper end is invaginated into the floor
of the fore-gut so as to form an incomplete partition
between the cavities of the cardiac and pyloric regions of
the fore-gut. The top of the cardio-pyloric valve is richly
clothed with setae.

Pyloric "Supporting Ossicles."

In the dorsal wall of the pyloric fore-gut there are
three pairs of ossicles.

The Anterior Mesopyloric Ossicles (a.mes) are a pair
of small ossicles lying immediately behind the pyloric
ossicle near the median dorsal line.

The Posterior Mesopyloric Ossicles (p.mes.) are a pair
of small ossicles lying behind the anterior pair.

The Uropyloric Ossicles {u.py.) are a pair of small
ossicles lying in the roof of the posterior part of the
pyloric region and immediately behind the posterior
mesopyloric ossicles.

In the ventral wall of the pyloric fore-gut the main
supporting ossicles are as follows : —

The Antero-inferior Pyloric Ossicle (a.i.p.) is a
median plate shaped somewhat like a truncated triangle.
The base of the triangle points forwards and conies into
contact with the cardio-pyloric valve. This ossicle lies in
front of the inter-ampullary groove.

The Pre-ampullary Ossicles are a pair of small plates.
Each lies at the side of the antero-inferior pyloric ossicle
and immediately in front of the pyloric ampulla.

L03

The Postero-inferior Pyloric Ossicle is a median
curved rod-like ossicle. It is concave anteriorly, and is
situated behind the pyloric ampullae.

In the lateral walls of the pyloric fore-gut there are
the following principal ossicles : —

On the supra-ampullary walls there are three pairs of
ossicles, viz., the Anterior (a.s.a.), Middle (m.s.a), and
Posterior [p.s.a.) Supra-ampullary Ossicles.

There are also three pairs of ossicles in the pleuro-
pyloric walls, viz., the Anterior, Middle and Posterior
Pleuropyloric Ossicles.

The positions of these six pairs of ossicles are
indicated by the names.

Muscles of the Fore-gut. (PI. VII.)

Mocquard* has divided the muscles of the fore-gut into
two kinds. The extrinsic muscles are those muscles which
have points of origin on some part of the skeletal system
outside the fore-gut, and which are inserted on to ossicles
lying in the walls of the fore-gut. The intrinsic muscles
are attached at both ends to ossicles lying in the walls of
the fore-gut.

E xtrinsic Muscles.

Anterior Gastric Muscles (g.a.) — one pair. Each
muscle has its origin on the procephalic process. Both
pass directly backwards near the median line, being only
slightly separated from one another, and are inserted on
the front of the pterocardiac ossicles near the middle line.

Inner Posterior Gastric Muscles (g.p.i.) — -one pair.
They arise from two small calcareous projections, almost
median in position, situated on the under side of the
mesogastric region "of the carapace. Each muscle passes

* Mocquard, Annates Sciences Naturelles, 6 ser., t. 16, 1883,
p. 238.

104

downwards and forwards, and is inserted on the front part
of the pyloric ossicle.

Outer Posterior Gastric Muscles {cj.p.e.) — two pairs.
The two muscles at each side iiin together so that they
may he mistaken for a single muscle. They arise from
the under side of the nasogastric region of the carapace,
some distance in front of the origin of the dorsal
pyloric dilator muscles, but not so near the middle line.
They pass downwards, forwards and inwards, and are
inserted on the external face of the exopyloric ossicle.

The above three sets of muscles are concerned in the
working of the gastric mill. In addition to these the
intrinsic muscles — the cardio-pyloric muscles — to be
described later, are also used in connection with the
gastric mill.

The following muscles serve to dilate the fore-gut : —

Upper Anterior Dilator Muscles (a.s.) — one pair. Each
arises from the inner side of the cephalic sternum
immediately behind the orbit. The muscle is not a
compact one, but passes backwards, upwards and inwards
as a series of muscular strands which gradually diverge.
They are inserted on the anterior and outer corner of the
fore-gut.

Lower Anterior Dilator Muscles (d.ai.) — one pair.
These are a smaller pair of muscles than the preceding.
They are very close to the middle line so as to appear
almost as a single median muscle. Each arises on the
upper side of the epistoma near the middle line and passes
backwards and slightly upwards, being inserted on the
lower part of the front wall of the fore-gut near the
median line. As in the preceding case, the muscle is
composed of several separate strands which diverge as
they approach the point of insertion.

Antero-lateral Dilator Muscles (d.la.) — one pair.

105

These are narrow muscle bands each arising about half
way along the outer edge of the roof of the pre-branchial
chamber. Each passes inwards and downwards parallel
to the front edge of the carapace and is inserted on the
lateral wall of the cardiac region of the fore-gut, above
the oesophagus.

Postero- lateral Dilator Muscles (d.lp.) — one pair.
These are broad muscles arising near the point of origin
of the preceding muscles. Each passes directly inwards
and slightly backwards and downwards. The muscle
broadens considerably as it approaches the fore-gut. Its
insertion is on the anterior edge of the postero-lateral
cardiac plate.

Dorsal Pyloric Dilator Muscles (d.sup.) — two pairs —
anterior and posterior The two muscles at each side
run close together so that it is difficult to distinguish the
separate muscles. They arise close together from the
under side of the carapace just behind the origin of the
outer posterior gastric muscles. They pass downwards
and slightly forwards and are inserted on the ossicles of
the dorsal wall of the pyloric region of the fore-gut.
The anterior muscles are inserted on the posterior nieso-
pyloric ossicles and the posterior muscles on the uropyloric
ossicles.

Ventral Pyloric Dilator Muscles. Two pairs — outer and
inner. Each of the inner pair (i.py.i.) is a long narrow
muscle arising near the base of the mandibular apophysis.
It passes upwards and slightly backwards on the inner
side of the posterior oesophageal dilator muscle and runs
very close to the posterior wall of the cardiac region. It
is inserted on the anteroinferior pyloric ossicle in the
ventral wall of the pyloric region. Each of the outer
pair (i.py.e.) is much shorter than the inner pair. Its
origin is on the endopleurite of the first maxillary

106

segment. From this the muscle passes upwards and is
inserted on the ventral pyloric wall on the outside of the
insertion of the inner pair.

The folloAving muscles dilate the oesophagus : —

Upper Anterior Oesophageal Dilator Muscles (oe.as.)
— one pair. Each of these muscles arises from the
epistoma close to the origin of the upper anterior dilator
muscle. Passing backwards and slightly upwards below
the latter muscle, it is inserted on the anterior wall of the
oesophagus. The muscle is not compact, but is made up
of separate strands which diverge as they approach their
insertion.

Lower Anterior Oesophageal Dilator Muscles (oe.ai.).
One pair of very small muscles. Each arises from a small
eminence on the posterior part of the epistoma near the
middle line. These eminences are behind the origin of
the previous muscle. The muscle passes backwards below
the previous muscle, and its insertion on the anterior wall
of the oesophagus is immediately below that of the
previous muscle.

Lateral Oesophageal Dilator Muscles (oe.l.) — one pair.
Each of these muscles is made up of three distinct bands
of muscle fibres. Near its origin the muscle is compact,
but the fibres diverge as they approach the oesophagus.
Each muscle arises near the extreme posterior angle of
the epistoma and passes inwards below the upper muscle.
Its insertion is on the lateral wall of the oesophagus.

Posterior Oesophageal Dilator Muscles (oe.p.) — one
pair. Each muscle arises from the top of the pillar-like
portion of the endopleurite of the first maxillary
segment and passes inwards and downwards. It runs
external to the inner ventral pyloric dilator, and crossing
over that muscle it is inserted on the posterior wall of the
oesophagus.

107

Intrinsic muscles.

Cardio-pyloric Muscles (c.py.) These consist of one
median and two lateral muscles. The median muscle
extends from the thickened posterior border of the meso-
cardiac ossicle to the upper edge of the propyloric ossicle.
The lateral muscles extend from the mesocardiac ossicle
to the exopyloric ossicle. These muscles are used in
connection with the gastric mill and are concerned in
bringing' the ossicles of the mill back to their original
position after each series of complicated movements
effected by means of the gastric muscles.

Lateral Cardiac Muscles (c.lat.) — three pairs. The
three muscles at each side may be distinguished as the
upper, middle and lower muscles respectively. The
upper muscle arises from the upper edge of the infero-
lateral cardiac ossicle and passes upwards and forwards
as a broad sheet of muscle to the dorsal border of the
z}Tgocardiac ossicle. The middle muscle also arises from
the upper edge of the infero-lateral cardiac ossicle below
the origin of the upper muscle and passes upwards and
forward parallel to this muscle. It is inserted on the
prepectineal ossicle, and also on the anterior part of the
dorsal border of the zygocardiac ossicle. This muscle is
much narrower than the previous one. Both sheets of
muscle are broader at their insertion than at their origin.
The lower muscle is a short broad sheet arising from the
side of the infero-lateral cardiac ossicle, and passing
across the upper portion of the postero-lateral cardiac
plate. Its insertion is on the antero-superior border of
this plate. According to Mocquard, these muscles raise
the cardio-pyloric valve.

The Postero-inferior Cardiac Muscle (Fig. 48, c.i.).
This is a median broad sheet of muscle covering the

108

posterior wall of the cardiac region of the fore-gut. It
is attached at each side to the posterior border of the
infero-lateral cardiac ossicle.

Anterior Cardiac Muscle (cant.). This is a median
muscle extending as a broad and thin sheet down the
front wall of the fore-gut. It arises in the median line
on the front of the mesocardiac ossicle and passes
forwards. As it passes downwards along the front wall
of the fore-gut it divides into two main branches, which
are attached separately to the front wall of the fore-gut.

The above muscle must not be confused with the
muscle of the same name described by Mocquard. The
latter muscle is on the antero-lateral wall. I therefore
designate it the Antero-lateral Cardiac Muscle (c.al.).
There is one pair of these muscles, each being situated on
the antero-lateral wall of the fore-gut immediately above
the oesophagus. It is attached to the anterior border of
the membranous antero-lateral cardiac plate, and passes
upwards almost to the median line.

The above two sets of muscles act as constrictors of
the cardiac portion of the fore-gut.

Circular Oesophageal Muscles (c.oe.). These are
present as a broad band running around the oesophagus
and acting as constrictors of the oesophagus.

Lateral Pyloric Muscles (py.lat). There are several
pairs of muscles — some broad and others very small —
arising at each side from the upper part of the post-
pectineal ossicle and the infero-lateral cardiac ossicle.
They pass upwards and are inserted on the various
ossicles of the lateral and dorsal walls of the pyloric
region of the fore-gut. These muscles serve as constrictors
of this region of the fore-gut.

109

The Mechanism of the Gastric Mill.

According to Huxley* the movement of the gastric
mill is effected by means of both the anterior and posterior
gastric muscles. By the contraction of these muscles the
urocardiac tooth is thrown forward, and simultaneously
the zygocardiac teeth are rotated inwards and the three
teeth meet in the middle line.

Mocquard has been fortunate enough to observe the
movements in a living Stenorhyncus having a remarkably
transparent carapace. Tie states that the active move-
ment is brought about almost solely by means of the
anterior gastric muscles. If the posterior muscles act at
all, it is only very feebly and spasmodically. When the
anterior gastric muscles contract, the urocardiac ossicle
and the median tooth are thrown forward. The movement
is slightly complicated because of the connection between
the posterior part of the urocardiac ossicle and the lower
part of the propyloric ossicle. When in a state of rest
the lower part of the latter ossicle lies considerably behind
its upper border. As a result of the contraction of the
anterior gastric muscle the lower part is drawn forward so
that the ossicle takes up a vertical position. The median
urocardiac tooth, if not in contact with the propyloric
ossicle, would have a simple backward and forward
movement. But the connection between the two ossicles
causes the median tooth to move in an arc the convexity
of which points downwards.

Since the anterior gastric muscles are inserted on the
inner ends of the pterocardiac ossicles, the latter are also
drawn forward when the muscles contract. This move-
ment causes the outer ends of the ossicles to turn down-
wards and inwards. Because of the connection between

* Huxley, T. H. The Crayfish. [International Science Series-]

110

the pterocardiac and the zygocardiac ossicles the anterior
handle-like portion of the latter are also drawn down-
wards and inwards. Posteriorly the zygocardiac ossicles
are in contact with the exopyloric ossicles, which in their
turn articulate with the pyloric ossicle. Therefore if we
consider the zygocardiac and the exopyloric ossicles as a
single rod, we have a lever of the second order, the
fulcrum being at the anterior end and the weight in the
region of the zygocardiac tooth. Thus, the application
of the force at the anterior end rotates the tooth down-
wards and inwards, and the three sets of teeth meet in the
middle line. When the muscles relax the ossicles spring
back into their original position, partly because of the
elasticity of their joints, but mainly by means of the
action of the cardio-pyloric muscles.

THE BLOOD VASCULAR SYSTEM

(PI. VII, figs. 49, 50; Pis. VIII, IX).

Briefly stated the scheme of circulation is as follows.
The pure blood returning from the gills passes into the
Pericardium by means of the Branchio-cardiac veins.
From the Pericardium the blood enters the heart through
the ostia. From the anterior end of the heart there
arise five arteries carrying the blood to the gonads, diges-
tive glands, fore-gut, and the front part of the body.
From the posterior region of the heart two median
arteries arise which supply the abdomen and the appen-
dages. The impure blood returning from the system does
not pass to the gills along definite vessels, but flows
through irregular spaces or sinuses between the various
organs. The blood frsm the sinuses eventually reaches
the gills and passes along the Afferent Branchial Sinuses
on the outside of the grills. The blood is distributed to

Ill

the various gill lamellae where it is oxygenated. The
pure blood leaves the gills by the Efferent Branchial
Veins, running along the inside of the gills and which
pass into the Branchio-cardiac veins.

The Pericardii! m (PI. IX, figs. 54, 56, Per.) is
a closed cavity surrounding the heart and having thin
transparent walls. It is situated immediately beneath
the cardiac region of the carapace, and between the
" flancs." It lies above the hind-gut and covers the
posterior portions of the digestive gland and gonads.
When viewed from above the shape of the pericardium
is roughly pentagonal (fig. 54). The base of the pentagon
is anterior and the apex is posterior.

The Branchio-cardiac veins (PI. IX, fig. 54, he 1-5)
enter each side of the pericardium by means of three wide
openings which have no valves. The first opening is
situated at the anterior corner of the pericardium and
receives the first and second branchio-cardiac veins. The
second opening is situated a little behind the first and
receives the third branchio-cardiac vein. The third
opening is situated at the postero-lateral corner, and at
this point the fourth and fifth branchio-cardiac veins
enter the pericardium.

The Heart (PI. VII, figs. 49, 50) is a white semi-
transparent body, pentagonal in shape when seen dorsally
and having a rectangular shape when viewed from the
side. Two angles of the pentagon are anterior and the
other three are posterior. The heart is suspended in the
pericardium by means of the alae cordis (fig. 49, 50,
cd.l-G), which are bands of fibrous connective tissue
stretching across from the angles of the heart to the wall
of the pericardium. Each ala cordis appears to have a
short band of muscle fibres attached to its outer
extremity.

112

The alae cordis are eleven in number (fig. 49, 50).

Dorsal Antero-lateral (cd.l)— one pair. Stretching
from the dorsal side of each of the antero-lateral corners
of the heart to the corresponding corner of the
pericardium.

Ventral Antero-lateral (cd.2) — one pair. Immediately
beneath the dorsal antero-lateral band. Extending from
the ventral side of each of the antero-lateral corners of the
heart to the corresponding corner of the pericardium.

Dorsal Poster o-lateraJ (cd.3) — one pair. Extending
from the dorsal side of each of the postero-lateral angles
of the heart to the corresponding angle of the pericardium.

Ventral Postero-lateral (cdA) — one pair. Having a
similar position to the dorsal postero-lateral band, but
lying immediately beneath it.

Median Posterior (cd.b) — A single band arising from
the dorsal side of the posterior angle of the heart and
stretching across to the posterior angle of the pericardium.

Posterior (cd.ti) — one pair. Arising ventrally from
the postero-lateral side of the heart and crossing to the
postero-lateral side of the pericardium.

The walls of the heart are also very muscular, and
the cavity of the heart is crossed by numerous strands of
muscle.

The blood enters the heart from the pericardium by
means of the Ostia. There are three pairs of ostia — one
pair at the anterior end of the dorsal wall of the heart
(Figs. 49, 50. a.ost.), one pair at the posterior end of the
dorsal wall (p.ost.), and the third pair are found in the
lateral walls of the heart — one ostium at each side (Lost.).
Each ostium is valved so as to prevent the blood from
returning to the pericardium.

113

The Arteries. (Pis. VIII, IX.)

The following arteries are given off from the heart :
At the anterior end (1) the median Cephalic artery, and on
each side of this is (2) a Lateral artery, and (3) a Hepatic
artery, all passing forwards. At the posterior end there
are two median arteries arising about the same point, (4)
the Descending artery passing downwards, and (5) the
Posterior Aorta passing backwards above the intestine.

At its origin from the heart each artery is valved, so
as to prevent the blood from returning.

Cephalic artery [Ophthalmic artery] (PI. VIII, fig.
51. o.art.). This is a median artery arising from the
anterior end of the heart. From its origin it passes
directly forward above that portion of the gonads situated
between the internal adductor muscles of the mandibles.
It pursues a straight course over the pjdoric and cardiac
regions of the stomach and between the anterior gastric
muscles. So far the course of the artery has been entirely
superficial, but near the anterior end of the cardiac
fore-gut it dips downward and divides into two
branches immediately above the brain. Each branch
passes outward and supplies the eyes and also the various
parts of the front region of the head.

Lateral artery [Antennary artery] (PL VIII, Fig.
51. a.art.) There is one pair of lateral arteries, each of
which arises from the anterior end of the heart a little
outside of the origin of the cephalic artery. On leaving
the heart the lateral artery passes outward, making an
angle of about 40° with the cephalic artery. Almost
immediately it passes through the outer portion of the
internal adductor muscle of the mandible (i. a. md.). It
then curves outward, sweeping around the stomach until
it reaches the external abductor muscle of the mandible
{e. b. md.). Here it divides into two parts — (1) an outer

i

114

portion, the Ovarian (ov. a.) [or Spermatic] artery, which
follows the course of the gonads, and (2) an anterior
portion, the Antennary artery (a. art.), which passes
around the front of the fore-gut and supplies the organs
in the region of the head.

The main portion of the lateral artery, after leaving
the heart, dips down gradually until it reaches the
external abductor muscle of the mandible (e. a. md.). The
antennary artery still continues to pursue a deeper course,
but the ovarian [or spermatic] artery becomes more super-
ficial. When the gonads are well developed the main
artery and the gonadial branch are partly embedded in
the substance of the gonad, but when the reproductive
organs are small these parts of the artery are quite
superficial.

Branches of Main Lateral artery —

Branch to the Hind-gut. This arises on the inner
side of the artery just behind the internal adductor
muscle of the mandible. It passes downward and inward
to that portion of the hind-gut beneath the front part of
the heart.

Branch to the Digestive Gland. This is a large branch
arising immediately in front of the branch to the hind-
gut. It passes outward and gives off numerous branches
to the digestive gland and also to the hypodermis in this
region.

Branch to the Cardiac Fore-gut. In front of the
internal adductor muscle of the mandible a large branch
is given off on the inner side. It passes through the
substance of the gonad and breaks up into a complicated
network on the lateral and dorsal walls of the cardiac
fore-gut'. Tin's branch supplies the muscles of the
fore-gut.

Branches to the Hypodermis. Throughout the whole

115

course of the lateral artery and its branches small arteries
are given off which supply the hypodermis.

The Ovarian [or Spermatic] Branch (ov. a.). This
follows the course of the gonads, and sweeps round near
to the outer edge of the carapace. Numerous branches
are given off to the gonads and also to the hypodermis.

The Antennary Branch (a. art.). This passes
anteriorly and dips downward as it sweeps around the
fore-gut. It passes over the paragastric lobe of the
bladder and divides into an outer and an inner portion.
The outer branch dips downward and outward and gives
branches to the external adductor muscle of the mandible.
It then passes outward to the hypodermis and supplies
also the hepatic lobe of the bladder. The inner branch
passes inward and supplies the antennae and the front
part of the head. It also sends branches to the anterior
gastric muscles and to the main vesicle and the para-
gastric and oesophageal lobes of the bladder.

Hepatic artery (PL VIII, fig. 51, h. art.). Owing to
(he fact that this artery dips down immediately on leaving
the heart, and becomes deeply embedded in the digestive
gland, it is rather difficult to locate. Hence several workers
at the Brachyura have neither figured nor described this
artery, and some have described other arteries as the
hepatic artery. Milne-Edwards,* in his description of
Maia, has designated as the hepatic artery those branches
of the sternal artery which supply the digestive gland,
lirookst has called the lateral (ophthalmic) artery by the
name of hepatic artery.

The hepatic artery of each side arises from thf
ventral side of the anterior region of the heart, its origin
being beneath and slightly external to that of the lateral

* Milne-Edwards. Hist. Nat. des Crustaces.
t Brooks. Handbook of Invertebrate Zoology.

116

artery. Immediately on leaving the heart it dips down-
ward and makes an outward sweep in the deeper parts of
the digestive gland. Near its origin on its inner side it
gives oft' a branch which goes to the hind-gut. There are
also other small branches which supply various parts of
the gland. The main artery, however, divides into two
branches, the posterior of which sweeps outward em-
bedded in the posterior part of the digestive gland. The
anterior branch passes beneath the gonad and the
external adductor muscle of the mandible and supplies
the anterior portion of the gland.

Posterior aorta (Superior abdominal artery) (PI.
VIII, fig. 51, PL IX, tig. 53, sa. art.). This arises as a
median vessel from the posterior end of the heart. Just
after leaving the heart it gives off at each side a small
vessel. This passes forward and downward beneath the
pericardium, and supplies those parts of the reproductive
organs lying beneath the pericardium, and also some of
the muscles of the Basi -Ischium of the Chela.

Immediately above the hind-gut caecum a second
pair of branches arises. These are the Postero-lateral
arteries (pi. a rt.), and have several complicated branches.
The main branch passes outwards above the coils of the
caecum and divides into two vessels. The. anterior vessel
passes forward and gives off branches to the extensor
muscles of the coxopodites of the four walking legs. The
posterior branch supplies the muscles extending from the
" flancs " to the carapace, and also gives a rich blood
supply to the coils of the hind-gut caecum.

Behind the origin of the postero-lateral arteries the
posterior aorta enters the abdomen. As the arrangement
in the two sexes is somewhat different, these will be
described separately.

Female. The aorta passes down the abdomen

117

above the hind-gut, but not in the median line. As it
passes backward it gradually crosses over to the right
side. In each of the second, third, fourth and fifth
abdominal segments a pair of arteries is given off to the
appendages. At the posterior end of the fifth segment
the aorta divides into two branches. The right branch
follows the course of the aorta. The left branch crosses
over the hind-gut and is continued down the left side of
the hind-gut parallel to the right branch. Both
branches pass into the telson, where they break up into
fine branches. Throughout the abdomen small branches
are given off from the posterior aorta to the hind-gut, to
the abdominal muscles and the muscles of the abdominal
appendages.

Male. There are only two pairs of arteries given
off from the posterior aorta. These are in the first and
second abdominal segments and supply the two pairs of
appendages. As the aorta passes backward it crosses over
to the left side, and in the fifth segment it bifurcates, the
right branch crossing over the hind-gut. As in the
female, there are also innumerable small arteries given off
to the hind-gut and to the abdominal muscles.

Descending artery (PI. IX, fig. 56, d.art.). This
leaves the heart at the posterior end close to the
origin of the posterior aorta. It is an extremely
wide vessel which passes almost directly downward
on the right side of the hind-gut until it nearly
reaches the anterior part of the " sella turcica," where it
turns suddenly forward. It continues to pass downward
and forward, and between the muscles of the fourth and
fifth thoracic appendages it passes through the foramen

* The term sternal artery is generally given to this artery, as well
as to its continuation along the ventral side of the thorax. I have,
however, thought it more fitting to apply the term " sternal artery "
only to the ventral portion.

118

of the coalesced thoracic ganglia. Immediately beneath
the nerve chain the artery divides into a short and broad
posterior branch and a much longer and narrower anterior
branch. Both these branches are continuous and
horizontal, and are known together as the sternal artery.

Sternal artery (PI. IX, fig. 52, S.art.). This is a
large and well-defined median artery lying in the thorax
below the nerve cord and between the muscles of the
thoracic appendages. It is broadest at its posterior
portion just behind its connection with the descending
artery (j.d.). In front of this connection it is continued
forward as a much narrower vessel, from wh'ich are given
off arteries to each of the post-oral cephalo-thoracic
appendages, and also to the digestive gland. The arteries
supplying these appendages arise separately from the
sternal artery, except in the case of the last two pairs of
thoracic appendages. The arteries supplying these
appendages arise as a single vessel at each side of the
posterior part of the sternal artery, each of which divides
into two branches, each branch going to an appendage.

The blood supply of the last five pairs of thoracic
appendages is very similar, and the artery supplying the
chela may be taken as typical of all. This artery (art.ti.)
arises singly from the sternal artery and passes outwards.
A short distance from its origin it gives off a large ventral
branch which supplies the muscles of the coxopodite and
basi-ischium. The main artery passes to the extreme end
of the appendage, giving off various small branches to the
various muscles of the limb.

Each of the arteries to the maxillipedes (art. 5), after
giving off its ventral branch, enters the appendage and
bifurcates, one branch going to the endopodite and the
other to the exopodite.

The arteries supplying the first and second maxilli-

119

pedes (art. 3, art. 4) are not symmetrical. The origin of
the artery going" to the first maxillipede of the right side
is posterior to that of the left. In the case of the second
maxillipede the origin of the artery on the right side is
anterior to that of the artery on the left.

• Just in front of the branches to the first maxillipede
the sternal artery bifurcates, each part passing forward
and uniting again behind the mouth, thus forming a ring.
From this ring arteries are given off to the mandibles and
the first and second maxillae (art. 1, art. 2).

Inferior Abdominal artery (PL IX, fig. 52, ia.art.).
The posterior part of the broad sternal artery is continued
as a narrow vessel which runs backwards over the " sella
turcica " and down the abdomen beneath the nerve cord
and the hind-gut. It gives off a few small branches to the
hind-gut and to the flexor muscles of the abdomen.

Blood Sinuses and Veins. (PL IX, figs. 54, 55, 56.)

The blood returning from the various parts of the
body to the gills is not enclosed in definite vessels, but
flows through irregular spaces known as sinuses. Generally
speaking, all the main organs of the body, such as the
alimentary canal, digestive gland, reproductive organs,
muscles, &c, have blood sinuses in close contact with
them.

Above the cardiac fore-gut there is a large sinus —
the Dorsal sinus — which is situated above the epigastric
lobe of the bladder. There is also a smaller sinus
between this lobe of the bladder and the fore-gut. These
two sinuses are connected in front.

The dorsal sinus passes down the front of the fore-gut
and is connected ventrally with a sinus which passes
backward beneath the stomach. At the level of the
oesophagus this sinus divides into a right and left portion.

120

These Sternal sinuses pass backward at each side of the
ventral part of the thorax between the leg muscles and
beneath the pericardium. The two sternal sinuses are
not definitely separated from one another, but are
connected here and there by irregular sinuses. On the
outer side each sternal sinus sends offshoots down the
pleural muscle chambers to the base of the gills (fig. 56,
br. S. 4). These branches or branchial sinuses again unite
into a long sinus which runs along the base of the gills.
This is the infra-branchial sinus (fig. 55, i. s.). The blood
sinuses from each of the thoracic legs also pass into the
infra-branchial sinus.

From the infra-branchial sinus the blood passes along
the afferent branchial sinuses on the outside of each gill.

At the posterior end of the thorax the sternal sinuses
are connected with a small abdominal blood sinus.

The sinuses in connection with the digestive gland —
the Hepatic sinus and the reproductive organs — the
Ovarian [or Spermatic] sinus — open into the sternal sinus
at each side. There is a very large sinus below the
posterior part of the digestive gland.

The Branchial sinuses (br. s. 1-5) connect the
sternal sinus with the infra-branchial sinus at each side.
They are five in number at each side. The first branchial
sinus (br. s. 1) commences below the anterior end of the
pericardium. It passes down the pleural muscle
chamber of the second thoracic segment and opens
into the infra-branchial sinus near its anterior end.
Similarly the second branchial sinus (br. s. 2) passes
down the third thoracic pleural muscle chamber, the
third sinus (br. s. 'A) is in the fourth pleural chamber, the
fourth sintis (br. s. 4) is in the fifth pleural chamber, and
the last branchial sinus (br. s. 5), which commences below
the posterior end of the pericardium, passes down the

121

pleural muscle chamber of the sixth thoracic segment.
All the pleural sinuses are below the hranchio-cardiac
veins, which also run down the pleural muscle chambers.

The Infra-branchial sinus (fig. 55, i. s.) is a long
sinus which runs along each side of the thorax at the
base of the thoracic legs and the gills. Posteriorly it
extends as far as the last walking leg, and anteriorly
almost as far as the metastoma. Into it flow the five
branchial sinuses from above, and from its outer and ventral
side there enters a narrow sinus from each of the thoracic
appendages. From its outer and dorsal side there is
given off an afferent branchial sinus to each of the gills.

Afferent Branchial sinuses (fig. ob). These sinuses
run along the outer side of each gill.

First (iff event branchial sinus (af. 1) goes to the podo-
branch of the second thoracic somite.

Second afferent branchial sinus (af. 2) goes to the
arthrobranch of the second thoracic somite.

Third afferent branchial sinus (af. 3) goes to the small
podobranch of the third thoracic somite.

Fourth afferent branchial sinus (af. 4) goes to the
anterior arthrobranch of the third thoracic somite.

Fifth afferent branchial sinus (af. 5) goes to the
posterior arthrobranch of the third thoracic somite.

Sixth afferent branchial sinus (af. 0) goes to the
anterior arthrobranch of the fourth thoracic somite.

Seventh afferent branchial sinus (af. 7) goes to the
posterior arthrobranch of the fourth thoracic somite.

Eighth afferent branchial sinus (af. 8) goes to the
pleurobranch of the fifth thoracic somite.

Ninth afferent branchial sinus (af. 9) goes to the
pleurobranch of the sixth thoracic somite.

The Efferent Branchial veins (fig. 54) receive the
pure blood from the branchial lamellae. They pass down

122

the inside of each gill. Each efferent branchial vein is a
blood vessel having a definite wall.

There are nine efferent branchial veins (ef. 1-9)
corresponding to the nine afferent branchial sinuses — one
for each gill.

The Branchio-cardiac Yeins (fig. 54) convey the pure
blood from the efferent branchial veins to the peri-
cardium. There are five branchio-cardiac veins at each
side. They have definite walls.

First Branchio-cardiac vein (be. 1) receives the first
and second efferent branchial veins. It passes up the
outer side of the pleural muscle chamber of the second
thoracic segment, above the first pleural sinus.

Second Branchio-cardiac vein (be. 2) receives the third,
fourth and fifth efferent, branchial veins. It passes up the
outer side of the third thoracic pleural muscle chamber.
The first and second branchio-cardiac veins enter the
pericardium together through the first opening (see section
on Pericardium).

Third Branchio-cardiac vein (be. 3) receives the sixth
and seventh efferent branchial veins. It passes up the
fourth thoracic pleural muscle chamber. It enters the
pericardium by means of the second pericardial opening.

Fourth Branchio-cardiac vein (be. 4) receives the
eighth efferent branchial vein, and passes up the fifth
pleural muscle chamber.

Fifth Branchio-cardiac vein (be. 5) receives the ninth
efferent branchial vein and passes up the sixth pleural
muscle chamber. The fourth and fifth branchio-cardiac
veins enter the pericardium through the third pericardial
opening.

There are no valves between the branchio-cardiac
veins and the pericardium

123

p

P

•rH

f

hj o Q q

J Ph Ph Ph

HH OQ S ^H

h i—i cq co

P O O O OQ

% pq pq pq pqpq
c

fa

o

w
W
o

vK AAAAAAil

N

O a a1 a- a* e5 a- a- a- a1

s n n ° ^2

3^^ ojS

m.Sl'eSPQ

-*-> ^ r-P rP • Cf\ • O *

H ce P s-<+^^ £ ^ i^d ~ „u

^ i—l <N CO "1< »0 ^Q^tt'c^^-

P

m £R ^P «G ^P <P ;:P «P iP ::P pq

g H(McOTfU5!Ot>00O5 ^

o

to "a -s ^ ,~

„^

_^

pSj

^

.2 £

"-" "i -

-*~

i-H GS} CO ^JH »0

to

^

JO

c^

2 : : o

o

„ „0 cS

i

15 -^ 2

CO

"M " CO rg

O O

CQ

° 3 ° cs

s S s 2

cS o d :

u o ^ 2

.2 r3 ,2 ®

•-d 13 t3 -g

o o

a p

^ , »- 2 S
.2 o.2^

1j .« m o o

(T> J-< Q) S-l *-l

29 "S 25 S S

CQ OQ CO CQ CQ

(§ 0$ (^ (^ (^

*» 'S Is -< -«

i-~( O* Co >Ji *o

OUOCO-,-;-,;--,

II II II II II II II II II

i- « W «* W tO h- 00 0)

124

The Blood.

The blood is an almost transparent fluid having a
slight pinkish-blue tint (due to the presence of haemo-
cyanin). The colour deepens on exposure to air. The
blood consists of an almost colourless lymph in which are
found numerous small cells or amoebocytes. There are two
principal kinds of amoebocytes: — (1) Semi-transparent
cells, which are amoeboid and have finely granular proto-
plasm. There is a well-defined nucleus. (2) Globular
cells containing refringent granules. As pointed out by
Cuenot,* these granules are similar to the eosinophilous
granules recognised by Ehrlich in the leucocytes of
various vertebrates. Hence Cuenot designates the second
kind the Eosinophilous amoebocytes.^ They are composed
of an albuminous material.

In the neighbourhood of the cephalic artery, above
the fore-gut, is a cellular mass which, according to
Cuenot, is a lymphatic gland in which the amoebocytes
are formed.

Cuenot has recognised five kinds of amoebocytes in
the blood, which are all stages in the transformation of
the clear amoebocytes mentioned above. The eosino-
philous amoebocytes are also formed from the clear
amoebocytes, and mark a stage in the degeneration of the
cell. The eosinophilous granules present in the amoebo-
cyte are small and few in number at first. They become
comparatively large in size and very numerous until the
entire cell is filled with a solid mass. The cell then
degenerates rapidly and finally disappears. The granules

* Cuenot, L. " Etudes physiologiques sur les Crustaces Deca-
pod es." Archives de Biologie, T. XIII, 1895, p. 245.

t The granules readily take the following stains :— Picric acid,
eosin, indigo-carmine, fuchsine acid and " orange G." They remain
absolutely colourless under the following stains : — Methyl green,
dahlia, crystal violet, methylene blue and safranin. (Cuenot).

125

are probably dissolved in the lymph. The cells them-
selves are eaten by the young clear amoebocytes, which
thus act as phagocytes. The phagocytic function is
limited to the young cells only. After the appearance
of the eosinophilous granules they cease to act as
phagocytes.

Cuenot* claims also to have discovered a phagocytic
gland which is quite distinct from the lymphatic gland.
It is a swollen mass of cells situated on the terminal
branches of the hepatic artery. The cells resemble the
free amoebocytes and probably act as phagocytes.

When the blood ceases to flow, coagulation takes
place. This is effected by the clear amoebocytes. These
cells become changed in their appearance, and they send
out numerous fine pseudopodia which unite with those of
the neighbouring cells to form a network, in which all
the cells are united together. Thus a clot is formed.

The "Pericardial Pouch."

At each of the postero-lateral corners of the peri-
cardium there is a structure which Cuenotf designated
the " poche pericardiale." In Cancer each pouch lies on
the upper part of the posterior thoracic epimera and pro-
jects into the branchial chamber. Externally, each pouch
is covered with a cuticle which is continuous with the
chitinous wall of the branchial chamber. The cavity of
the pouch, which is continuous with the pericardial sinus,
is to some extent broken up by connective tissue cells
and by muscle hbres. The function of these pouches is
unknown.

* Cuenot, L. Comptes Rendus, 1903 (No. 137), p. 619.
f Cuenot, L. ''Etudes physiologiques sur les Crustaces
Decapodes." Archives de Biologic, t. XIII (1895).

126

INSPIRATORY SYSTEM.

(PL X, figs. 63, 64; PL XI, figs. 65, 66, 67).

Respiratory Mechanism.

Respiration is effected by means of gills or branchiae,
which are outgrowths from the Avails of the thorax. The
gills do not project directly on to the exterior, but are
situated in the branchial chambers at each side of the
cephalothorax.

The branchial chamber s — one pair. The
cavity of each chamber is morphologically a part of the
exterior, and its walls are formed by the downgrowth of
the carapace at each side. The sub-branchial region of
the carapace is closely applied to the coxopodites of the
pereiopods, and here turns inward to form the wall of the
branchial chamber. In transverse section the chamber
has a triangular shape (PL IX, fig. 56), and its walls may
be spoken of as ventral, dorso-lateral and inner. The
two former are membranous and are continuations of the
inturned edge of the sub-branchial region.

The postero-lateral portion of the digestive gland
rests upon the roof of the branchial chamber. Between
the floor of the chamber and the sub-branchial region of
the carapace there is a mass of connective tissue. In the
anterior region of the chamber the floor is raised into a
well defined transverse ridge. In a full-sized crab this
groove is about ^ inch in front of the anterior inhalent
aperture. In sections through this ridge the epidermis is
greatly elongated and has a glandular structure. There
are also tegumentary glands below the epidermis. The
podobranch of the second thoracic somite is closely
applied to the posterior side of this ridge. For the sake
of convenience I designate the latter the branchial ridge.
Its probable function will be discussed below.

127

The inner wall is well calcified, and is formed by the
thoracic epimera (" flancs "). The gills rest on the inner
wall.

The development of the branchial chambers would
lend to produce a stagnant layer of water around the
respiratory organs. But the nature of their function
requires that the gills should be in contact with water
containing a normal amount of dissolved oxygen. So
that, correlated with the formation of the branchial
chamber, an arrangement has been effected for producing
a constant stream of water over the gills. This necessi-
tates two things (1) inhalent and exhalent openings in
connection with each branchial chamber, and (2) some
mechanism for producing the current of water through
the chambers.

In the Macrura the inhalent opening is situated
between the inner edge of the branchiostegite and the
base of the thorax. In Cancer, however, this opening is
considerably smaller. Between the chela and the last
pereiopod the sub-branchial region is closely applied to
the base of the thorax, and the line of separation between
the two is guarded by a thick growth of long setae, so
that it is highly improbable that any water can gain
entrance to the branchial chamber in this region. Above
the last walking leg, however, there is a small slit opening
into the posterior region of the branchial cavity. This
is the posterior inhalent aperture. In front of the
coxopodite of the chela there is a well-defined transverse
opening leading into the anterior part of the branchial
cavity. This is the anterior inhalent aperture. The
latter is guarded in front by the coxa and flabellum of the
third maxillipede, which, on their posterior borders, are
clothed with long setae. There are also numerous setae
on the anterior face of the coxopodite of the chela. These

128

two sets of setae probably strain the water as it passes
through to the branchial chamber.

At its anterior end the branchial chamber is
extremely shallow on account of the roof sloping down at
a considerable angle. Above the branchial ridge, already
referred to, there is an extremely narrow cavity between
the top of the ridge and the roof of the chamber. This
cavity is continued forward and inward into the pre-
branchial chamber.

P r e - b r a n c h i a 1 chambers. One pair.
They are situated at the side of the mouth, each being in
front of and connected with the branchial chamber of the
same side. Each chamber is produced by the ingrowth of
the inner edge of the anterior part of the sub-branchial
region of the carapace. Its walls, therefore, are con-
tinuous with those of the branchial chamber. The pre-
branchial chamber is much smaller and shallower than the
branchial chamber. On its anterior and inner side it is
connected with the exterior by means of a wide opening
the exhalent aperture.

The current through the branchial and pre-branchial
chambers is caused by the vigorous action of the seapho-
gnathite. The latter lies in the pre-branchial chamber,
and when at rest the anterior surface faces upward.
Normally the scaphognathite displays the following
movements: — The action of the extensor muscles tends to
pull the ventral surface backward. This is followed by a
sharp forward blow of the outer lobe of the scapho-
gnathite, caused by the action of the outer flexors. This
is immediately followed by an undulating movement of
the inner lobe, caused by the accessory muscles and inner
flexors. In this way the water is baled out of the exhalent
aperture. This current from behind forward is probably
assisted by the energetic action of the exopoditic flagella

129

of the maxillipedes. The extremely active motion of
these flagella is quite obvious, and they probably
form an accessory current-producing organ of no mean
importance.

The normal current, as we have seen, flows from
behind forwards, entering at the inhalent apertures and
leaving by means of the exhalent aperture. In Corystes,
Atelecyclas and Portumnus, Garstang* observed that the
branchial current was sometimes reversed. Bohnt has
extended these observations, and finds this phenomenon is
of universal occurrence throughout the Brachyura. In
Cancer the habit of reversing the branchial current does
not appear to be very strongly developed. Bohn suggests
that the reversal takes place in order to rest the fatigued
muscles of the scaphognathite, as the energetic action is
performed by different muscles in the two cases.

The flabella (epipodites) of the maxillipedes pass
backward into the branchial chamber. That of the first
maxillipede (f.m.1) is by far the largest, and extends
backward throughout the whole length of the branchial
chamber lying upon the gills. The fiabellum of the
second maxillipede (f.m.2) lies below the gills towards the
dorsal side of the epimera. It only extends as far forward
as the middle of the epimeron of the fourth thoracic
somite. The fiabellum of the third maxillipede (f.m.3)
also lies below the gills ana on the ventral and outer side
of the second fiabellum. Its proximal portion forms part
of the anterior boundary of the anterior inhalent aperture.
It extends backward to the posterior end of the epimeron

* Garstang. " The Habits and Respiratory Mechanism of Corystes
cassivelaunus ." — Journal Marine Biological Association, Vol." IV
fN.S.), p. 223.

"The Respiratory Phenomena of Portumnus nasutus." Journal
Marine Biological Association, Vol. IV (N.S.), p. 402.

t Bohn, G. " Sur la Respiration des Decapodes." Bull. Sci.
France et Belg. T. XXXVI (Ser. 6), 1902, p. 178.

130

of the fifth thoracic somite. All three flabella are richly
clothed with long setae.

In the living animal the flabella have a slow motion
over the surface of the gills. Their main function is,
undoubtedly, to keep the surface of the gills free from
sand and mud which may be suspended in the water
carried into the branchial chamber. In a crab from Port
Erin the flabellum of the first maxillipede of the right
side had been destroyed. In consequence of this the
outer surface of the gills of the right side was covered
with a layer of fine mud, which must have rendered the
outer portions of the gills inoperative. It is doubtful
whether the flabella have any function with regard to the
formation or regulation of the current of water over the
gills. At any rate, this function, if present, has not the
importance ascribed to it by Claus*

The description of the respiratory mechanism of the
Brachyura given by Milne-Edwards has become almost
classic, and has been accepted by most workers on the
subject. According to this explanation, the water enters
the branchial chamber at one place, viz., in front of the
coxopodite of the chela. On entering the branchial
chamber the current passes backward below the gills and
then forward above the gills and out to the exterior
through the pre-branchial passage.

This explanation has been disputed by Bohn,f who
states that the water enters the branchial chamber
throughout the entire length of the inner edge of the
sub-branchial region, the entrance being especially
marked at the anterior and posterior inhalent apertures
(" l'orifice inspirateur anterieur et posterieur," Bohn).
According to Bohn, the water entering by the anterior

* Claus. Arbeit. Zool. Tnstit. Wim, Bd. VI., Hft. 1.
t Bohn. Op. cit.

13]

inhalent aperture does not pass backward but passes
directly forward to the pre-branchial chamber, and only
bathes the anterior part of the sixth gill and all the gills
in front of this. The posterior gills are supplied by water
entering the posterior inlialent aperture. He denies that
there is a backward current caused by the flabella.

My own observations on these points are as follows :
— There are two inhalent apertures — the anterior and
posterior. Between these two apertures the inner border
of the sub-branchial region is closely applied to the side
of the thorax, and there appears to be absolutely no inflow
of water along this border. Of the two inhalent
apertures, the anterior is decidedly the most important.
The current of water flowring in through the posterior
aperture is very small. I think it extremely probable
that some of the water drawn in at the anterior inhalent
aperture passes backward, but I do not accept the
explanation of Claus — that the backward current is caused
by the flabella of the maxillipedes. The presence of the
well-defined branchial ridge (see above) in the anterior
part of the branchial cavity has suggested another
explanation. The ridge is situated on the floor of the
branchial chamber immediately in front of the anterior
inhalent aperture. It arises near the inner side of the
chamber, and passes in front of the aperture as a
transverse wall. ( hi the outer side of the aperture it turns
backward and outward, and after extending half way
down the branchial chamber, it gradually dies away.
At its anterior end the branchial chamber is exceedingly
shallow, so that the ridge almost extends to the roof of
the chamber, leaving only a narrow slit which communi-
cates with the pre-branchial cavity. As the water flows
in through the anterior inhalent aperture, it will be drawn
forward by the vigorous action of the scaphognathite.

132

The branchial ridge, however, will act as a formidable
barrier, and although some of the water will undoubtedly
pass directly over into the pre-branchial chamber, it is
reasonable to suppose that some of it will have its course
changed by the branchial ridge and will pass backward,
following the direction of the latter.

The Gills (Figs. 63, 64).

The gills arise from each side of the thorax and lie
upon the inner wall of each branchial chamber, i.e., on
the thoracic epimera. According to the terminology
introduced by Huxley, the gills may be placed in three
categories — the podob ranch arising from an appendage,
the arihrobranch arising between an appendage and the
epimeron, and the pleurobranch arising from the epimeron.

In Cancer there are only nine gills at each side. The
following is the branchial formula.

Thoracic somites 1

2

3

4

5

6

7

8

Total.

1
1

(1)

1

1
1

(1)

1
1

1

1

—

—

2

Anterior arthrobranck .
Posterior arthi obranch .

3

2
2

(1)

(3)

Total

(1)

2 + (l)

3 + (l)

2

1

1

— i —

9 + (3)

First gill. Podobranch of the second thoracic somite
(tigs. 63, 64, g. 1). It arises from the coxopodite of the
second maxillipede between the exopodite and the flabellum.
It lies with its outer face in contact with the posterior
side of the branchial groove, and its inner face is closely
applied to the basal portions of the gills 2 to 6. Its apex
points backward and outward. Length 22 mm.*

Second gill. Arthrobranch of the second thoracic

somite (g. 2). Arises from the arthrodial membrane of

* The measurements of the gills are taken from a crab having a
carapace breadth of 12 cm.

133

the second maxillipede. It passes directly backward and
lies upon the fused epimera of the first and second
thoracic somites. Its long axis is at right angles to that
of the first gill. Length 25 rani.

Third gill. Podobranch of the third thoracic somite
\(j. •'!). Arises from the elongated coxopodite of the third
maxillipede. It is extremely short and is wedged in
between the first and fifth gills, at the base of the latter.
Length 7 mm.

Fourth gill. Anterior arthrobranch of the third
thoracic somite (g. 4). Arises, together with the fifth
gill, from the arthrodial membrane of this somite. It
lies on the thoracic epimera, immediately behind the
second gill. Length 26 mm.

Fifth gill. Posterior arthrobranch of the third
thoracic somite (g. 5). Arises from the same place as the
fourth gill, and lies immediately behind the latter. Its
base is notched in order to receive the third gill. Length
25 mm.

Sixth gill. Anterior arthrobranch of the fourth
thoracic somite (g. G). Arises together with the seventh
gill from the arthrodial membrane between the chela and
the fourth thoracic somite. It lies behind the fifth gill.
Total length 37 mm.

Seventh gill. Posterior arthrobranch of the fourth
thoracic somite (g. 7). Arises from the same place as the
sixth gill and lies immediately behind it. Total length
37 mm.

Eighth gill. Pleurobranch of the fifth thoracic
somite (g. 8). Arises from the epimeron of this somite.
Total length 28 mm.

Ninth gill. Pleurobranch of the sixth thoracic
somite (g. 9). Arises from the epimeron of this somite,
upon which it lies, immediately behind the eighth gill.

J 34

Structure of a gill (tigs. 05, GO, 67).

Each gill is of the phyllo-branchiate type. With the
exception of the third gill, they are all pyramidal in
shape, their apices pointing upward (with the exception of
the first gill, in which the apex points backward). Along
the outer side of each gill runs the afferent branchial
vessel, and the efferent vessel is situated on the inner side.
The gill is composed of numerous lamellae, which have
the appearance of the leaves of a book. Each lamella is
covered with a thin layer of chitin. This layer is also
continued on the outside of the afferent and efferent
vessels. The gills, therefore, are covered by part of the
general chitinous exoskeleton, and at ecdysis this outer
chitinous layer is cast with the remainder of the
exoskeleton. In transverse section each gill is triangular
(iig. (J5). The efferent vessel is situated at the apex of
the triangle, and the afferent vessel lies in the middle of
the base of the triangle. Stretching across from the
afferent to the efferent vessels is the branchial septum
(i.b.s.), which separates the anterior from the posterior
lamellae.

In the branchial septum between the afferent and

efferent vessels transverse sections reveal the presence of

scattered cells, generally having brown contents (fig. 07,

br. c). These are excretory cells, and together constitute

the branchial excretory organ (see section on Excretory

System). (Atenot* found that when a crab is injected

with ammonium carminate or methylene green these

substances are taken up by the excretory cells of the

branchial septum. These cells, therefore, have the same

reaction as the end-sac epithelium.

I have not been able to find any trace of the branchial

glands discovered by Allent in Palaemonetes. Cuenot,

* Cuenot. Arch, de Biol, T. XIII, p. 245.
t Allen. Q.J.M.S., Vol. XXX IV, p. 75.

135

however, has found them close to the efferent vessels in
several of the Bracyhura. They do not appear to be
present in Cancer.

In longitudinal sections through the gill (fig. 66)
the lamellae are seen to be lined by epidermal cells
(eh. cp.). There is a narrow cavity containing blood
separating the upper and lower layer of cells. This
cavity, or lamellar sinus (l.s.) is bridged over in certain
parts by the junction of the two layers of epidermal cells.
At the free edge of each lamella the lamellar sinus is
continuous with the larger outer lamellar sinus (p. l.s.).
This runs around the edge of the lamella, and the
epidermal cells in this region are extremely flattened.
Each lamellar sinus is in contact with the afferent
branchial vessel on the outer side, and the efferent
branchial vessel on the inner side. It is in the lamellae
that the aeration of the blood is effected.

Dorso- ventral muscles (PI. VIII, fig. 51,
PL IX, fig. 56, d. v. m.).
Extending upward from the membranous roof of the
branchial chamber to the carapace, is a series of muscles
which may be termed the dor so-ventral muscles. The
arrangement of these muscles will not be described in
detail. There are three sets of muscles at each side (see
fig. 51). The outer and middle series are arranged in two
parallel lines running antero-posteriorly. The inner set
is small, and is situated above the inner region of the
branchial chamber. The roof of the chamber is consider-
ably lower at the anterior end than in the posterior region,
so that the anterior muscles are consequently longer than
the posterior muscles.

Since the roof of the branchial chamber is soft and
membranous, it is capable of considerable movement. The

136

contraction of the dorso-ventral muscles will effect the
raising- of the branchial roof, and thus produce a corres-
ponding increase in the capacity of the branchial
chamber. When the muscles relax, the weight of the
superimposed digestive gland and gonad will be sufficient
to depress the roof and decrease the volume of the
branchial chamber.

Although it is difficult to understand the precise
function of these muscles, it must be conceded that their
action may be of supreme importance in connection with
the branchial chamber, either as a current regulator or as
an accessory current-producing organ. It will not be
surprising if additional investigations on this point throw-
new light on some of the problems discussed above.

EXCRETORY SYSTEM

(PI. X, figs. 57, 58, 59, Text figs. 11 and 12).

Excretion is performed in three different parts of the
body.

(i) by the Antennary glands and their connections.
(ii) by the Ferment cells of the digestive gland.
(in) by the Branchial excretory organ.

(ij The Antennary Glands and Connections.

These form a complicated system of organs at each
side of the body. The right and left sides, which are
similar to one another are absolutely separate, although
in certain places the two parts are in very close contact.
This excretory organ is a coelomoduct*, and may be
divided into three portions on each side. The first part,
or the antennary gland (" green gland," " rein anten-
narie "), is situated in the cephalic region immediately
behind the eye socket. It is a small spongy mass of a

* Goodrich. Various papers in Q.J.M.S., Vols. XXXVII-XLV.

137

light green colour, having a triangular shape when
viewed from above. At its posterior and inner corner it
is connected with the second portion — the bladder
(" vessie," Marchal ; " nephro-peritoneal sac," Weldon).
This is an extensive thin-walled sac having several large
branches. It is easily made out because of the dark
brown colour of its walls. Immediately in front of the
antennary gland the main portion of the bladder is
connected ventrally with the third part — the ureter.
This is a spacious tube leading downwards and opening
to the exterior beneath the operculum, which is situated
on the ventral side of the basal portion of the second
antenna.

(1) The Ante n nary G 1 a n d (PI. X, tigs. 58, 59,
Text fig. 11) is made up of two portions. On the dorsal
side is a small vesicle — the end sac [" saccule," Marchal]
(fig. 58, end. s., Text fig. 11, e.s.). From the floor of the
end sac are given off numerous blind prolongations, which
may either be simple or branched. The epithelium lining
the end sac (e.es.) is composed of flattened irregularly-
shaped cells, some of which project more than others into
the cavity of the end sac. Many of these cells contain
small yellow oil globules. Marchal speaks of the
epithelium of the end sac in Maia as being columnar, but
in Cancer it has a decidedly squamous appearance.
Marchal also states that the walls of the end sac are more
than one cell thick in places. This does not appear to be
the case in Cancer. The cells of the end-sac epithelium
do not stain so deeply as the epithelial cells of the lower
part of the antennary gland.

The ventral portion of the antennary gland is much
larger than the dorsal end sac. In sections this lower
portion is seen to have a very complicated structure, and
is, therefore, known as the Labyrinth (fig. 58, Text fig. 11

188

Lab.). The essential part of the labyrinth is the Renal
tube (tu.), the cavity of which is connected in front with
the end sac and behind with the bladder. The roof of the
renal tube is in close contact with the floor of the end sac.
It may be conceived that the lumen of the renal tube was
primitively quite simple, so that in such a condition of
things the ventral part of the antennary gland would
show none of the complicated structure which we
designate the labyrinth. The complexity has been
produced in two ways. As already mentioned, the floor of
the end sac sends downwards numerous branched tubes,
the cavity of each tube being connected with that of the
end sac, and its walls being lined by the squamous
epithelium typical of the end sac. The floor of the end
sac is closely applied to the roof of the renal tube, so that
these prolongations push before them the epithelium of
the renal tube, at the same time breaking up its lumen.
Invaginations also appear in the ventral and lateral walls
of the renal tube, giving rise to partitions across the
lumen of the tube known as trabeculae. These ventral and
lateral ingrowths are not caused by the extension of the
end sac. In sections the dorsal ingrowths of the end sac can
always be distinguished from the ventral ingrowths by the
fact that the former appear to be lined by two rows of
epithelial cells — the squamous epithelium of the end
sac, carrying before it the epithelium of the renal tube,
the two only being separated by a narrow blood sinus.
The ventral ingrowths are only lined by the epithelium of
the renal tube, and enclose portions of connective tissue
which have been drawn in from tissue surrounding the
gland. The epithelium of the renal tube (tig. 59, e. tu.) is
distinctly columnar, and the protoplasm has a finely
striated appearance. The cells are lined by a thin border,
which is generally described as a cuticle.

139

In sections through the antennary gland the
following structures may, therefore, be made out:

Dorsal, (a) The main portion of the end sac. This
is a simple cavity lined by squamous epithelium, the cells
of which do not take the stain well.

Ventral. The labyrinth made up of the following
parts : —

(b) Numerous small spaces with an inner lining of
squamous epithelium (ventral prolongation of the
end sac), and an outer lining of deeply stained
columnar epithelium.
(<•) Irregular spaces surrounded by columnar epithelium
which is deeply stained. On the side facing the
lumen the cells are lined by a fine border. These
are portions of the renal tube.

(d) Small spaces lined by the renal tube epithelium.
In the spaces are connective tissue cells and blood.
These are the trabecular ingrowths. In some parts
the epithelium of the trabeculae fuses with the
epithelium of the roof of the renal tube, so that
the blood sinus passes right through the labyrinth.

(e) Between the epithelium of (/>) and (c) may be made

out small blood sinuses.
(2) The bladder (Fig. 57, Text figs. 11, 12) is
extensive and complicated. It is a thin-walled sac
readily made out on account of its deep brown colour.

The Main Vesicle is situated above the antennary
gland, and its cavity is continuous with that of the renal
tube at the inner and posterior end of the antennary
gland In front of the gland it is connected with the
ureter. From the main vesicle are given off the follow-
ing lobes : —

(a) at the anterior end — Epigastric lobe, Progastric
lobe, Antero-lateral lobe, Cerebral lobe.

140

(/3) at the posterior end — Hepatic lobe, Supra-
hepatic lobe, Paragastric lobe, Oesophageal
lobe.
The Main Vesicle ["sac vesical," Marchal] (31. V.)
is situated below the anterior and outer corner of the
stomach and above the antennarv gland, with which it is
connected at the posterior of the latter. On its ventral
side the main vesicle is connected with the ureter
immediately in front of the antennary gland.

Fig. 11. — Diagrammatic longitudinal section of the excretory system.

Ep. = Epigastric lobe.

Pro. = Progastric ,,

Par. = Paragastric „
S.H. = Supra-hepatic .,
Hep. = Hepatic
a.l.i. = Inner part of antero-
lateral lobe.

M.V. = Main vesicle.
Mus. = External adductor
muscle of mandible.
e.s. = End sac.
Lab. = Labyrinth.
Ur. = Ureter.
U. ap. = External aperture of
ureter.

The Epigastric lobe (Fig. 57, e. lb., Text fig. 11,
Ep*) is the most obvious part of the bladder when dis-
secting the crab from above. It arises from the anterior
end of the Main Vesicle. It passes forwards and upwards
along the front wall of the fore-gut and turns back
along the dorsal wall, extending as far back as the

141

pyloric region. It is clearly separated in the middle line
from the corresponding' lobe of the other side. It is broad
anteriorly, and becomes narrower and more irregular in
shape towards the posterior end. The outer part of the
broad anterior portion is slightly reflected down the side
wall of the fore-gait and comes into contact with the
paragastric lobe.

The Progastric lobe (Fig. 57, p. lb., Text fig. 11,
Pro.), is, strictly speaking, part of the Epigastric lobe.
It is situated close to the connection between the Main
Vesicle and the Epigastric lobe. It is on the inner side
of the latter, and is closely applied to the front wall of
the stomach. The lobes of each side come into very
close contact in the median line, and it is only by very
careful dissection that they are seen to be separate.

The Antero-lateral lobe ["lobe du muscle adducteur,"
Marchal] (Fig. 57, a. lb., Text fig. 11, a. !. i.) arises from
the anterior and outer corner of the Main Vesicle. It
consists of two parts: —

(a) An inner lobe (Text fig. 12, 13, a. I. i.), which
passes inwards beneath the antennary gland.

(b) An outer lobe (Text fig. 12, li, a.l.o.), which
passes outwards beneath the digestive gland as far as the
origin of the outer adductor muscle of the mandible.

The Cerebral lobe (fig. 57, c. lob.) arises from the
anterior and inner corner of the Main Vesicle. It passes
inwards above the cerebral ganglia, and almost meets the
corresponding lobe of the other side.

The Hepatic lobe [" arriere vessie " (part), Marchal]
(fig. 57, h. U>., Text fig. 11, Hep.) arises from the posterior
and outer corner of the Main Vesicle by a very narrow
portion, which passes beneath the outer adductor muscle
of the mandible close to its insertion on the mandibular
apophysis. The main part of the lobe passes outwards

14-2

beneath the digestive gland and follows the course of the
gland to its extreme outer and posterior corner. Along
the outer edge of the digestive gland the lobe turns
upwards and covers the outer part of the dorsal portion
of the gland. Owing to its position beneath the digestive
gland the hepatic lobe is not readily seen, in spite of its
large size. Near the origin of this lobe from the Main
Vesicle a small inner lobe is given off, which ends blindly
near the posterior oesophageal lobe.

The Supra-hepatic lobe (tig. 57, s.Ib., Text fig. 11.
S. H.) is not well developed in Cancer. It arises from
the Main Vesicle on the inner side of the origin of the
hepatic lobe, and passes above the digestive gland on
each side of the fore-gut.

The Paragastric lobe (fig. 57, g. lb., Text fig. 11,
Par.) arises from the posterior end of the Main Vesicle
near to the origin of the supra-hepatic lobe. Its outer
side is applied to the mandibular apophysis, and on its
inner side it comes into contact with the side wall of the
fore-gut. It passes up the side of the latter and
touches the epigastric lobe.

The Oesophageal lobe arises from the inner and
posterior corner of the Main Vesicle. It passes inwards
and divides into anterior (as. lb.) and posterior (po. lb.)
portions which wrap around the oesophagus, touching
the corresponding lobe of the other side in the middle
line.

In sections through a young crab (width of carapace
15 mm.) the epithelium of the bladder (PI. XII, fig. 82)
consists of columnar cells 20 fi long and 15 /x wide. The
protoplasm is denser near the outer portion of each cell,
and the inner portion of the protoplasm is greatly vacuo-
lated. In the outer region of the cell the protoplasm is
arranged in longitudinal strands, which gives rise to the

143

at. 'U.np.

Fn;. 12.— Diagrammatic transverse sections through the left half of
the body to show the excretory system.

A. = at the level of the external excretory aperture.

B. = at the level of the " green gland." (The gland is shown as a

dotted line.)

C. = at the level of the oesophagus. (For S.H. read Hep.)
Card. = cardiac region of the fore-gut. o.l. = oesophageal lobe.

o. = oesophagus. nid. = mandible.

a.l.o. = outer portion of anterodateral lobe.

(Other references as in Fig. 11).

144

striated appearance noted by Weldon and Allen. In some
cases these striations are carried through to the inner
region of the cell, but generally the protoplasm is divided
into a deeply-stained outer portion, which exhibits the
striations mentioned above, and a more lightly stained
inner portion in which the strands of protoplasm are few
in number, thus causing the vacuolated appearance. The
nucleus of the bladder epithelial cells has a diameter of
6ju, and is situated near the centre of the cells.

(3) The Ureter (Text tig. 11, 12, A, Ur.) is a
spacious sac situated on the inner side of the antennary
gland, partly in front of and partly below the gland. The
connection with the main vesicle is a narrow opening
immediately in front of the anterior and inner corner of
the gland. The ureter passes downwards, and opens to
the exterior on the ventral surface of the proximal portion
of the second antenna. It is lined with epithelial cells,
which are distinctly larger than those of the bladder and
take the stain more distinctly.

The relation of the parts in the neighbourhood of the

external orifice is of interest. The proximal region of the

ventral side of the second antenna is occupied by a small,

irregularly-shaped plate — the operculum (PL II, fig. 5,

op.), which is freely movable. The movement is possible

because of its connection, by means of a flexible

membrane, with the surrounding hard parts. The

membrane forms a pocket-like invagination around the

operculum, and is deepest on the inner and posterior side,

so that the movement is greatest on this side when the

operculum is raised. When the operculum is elevated it

is seen that the membrane is perforated by a distinct

orifice on the inner and posterior side. This is the

excretory orifice. Hence Marchal* termed the membrane

::' Marchal. " Appareil excreteur des Crustaces Decapodes."
Arch. Zool. exp. et gen., T. X (Ser. 2), 1892.

145

surrounding the orifice the excretory membrane. He states
that the excretory fluid ,-s not ejected as the result of
muscular contraction around the walls of the sac-like
ureter. There are no muscles in the walls of the ureter.
The outflow is of a passive character, and takes place
whenever the excretory orifice is opened. Although it is
probable that the operculum does not fit tightly enough
to prevent the outflow of the excretory fluid, yet it is when
the operculum is closed that the fluid ceases to escape.
"When the operculum is drawn down closely into its
socket, the lips of the excretory orifice are pressed together
because of the contraction of the excretory membrane.
On the contrary, when the operculum is raised the
membrane is extended and the orifice opens. As shown
by Marchal, the movements of the operculum are under
the control of two muscles — an elevator muscle for
extending the operculum membrane, and n depressor
muscle for drawing the operculum down into its socket.

Mode of Excretion..

According to Marchal, the excretory fluid is not
produced by diffusion across the epithelial cells of the
antennary gland and bladder, but is the result of an
actual breaking away of a part of each of the epithelial
cells. An examination of the excretory fluid will show
that there are numerous bladder-like vesicles and also
cells floating in the fluid. In sections of the bladder,
and also in teased preparations, one can see, as stated
above, that the protoplasm is highly vacuolated on the
inner side of each epithelial cell. Moreover, in some
cells there is one large vacuole, or vesicle, projecting
into the lumen of the bladder and surrounded by an
extremely thin layer of protoplasm. In the interior of
the vesicle is a fluid sometimes containing refringent

L

146

granules. Each vesicle which contains excretory fluid
and excretory granules ultimately breaks free from the
cell and floats away in the excretory fluid.

In addition to this method, it is highly probable that
some of the excretory fluid passes through the epithelial
cells into the cavity of the bladder by the ordinary process
of diffusion.

The above mode of excretion is not only performed
by the cells of the bladder, but also, according to
Marchal, by the cells of the labyrinth, in a slightly
modified manner.

As stated above, the cells of the renal tube in the
labyrinth appear to be lined by a thin cuticle. Marchal
states that it is not a cuticle, but that the appearance is
due to a row of very small vacuoles lining the inner side
of each epithelial cell. These vacuoles gradually increase
in size, fuse together, and the single large vesicle formed
is liberated into the lumen of the renal tube, in a very
similar manner to the method described above in the case
of the bladder.

In the cavity of the end sac the fluid contains small
vesicles containing yellow oil globules. These have been
excreted from the epithelial cells of the end sac, which,
as mentioned above, often contain yellow oil globules.

In addition to these, Marchal states that in Maia
entire epithelial cells break free from the walls of the end
sac.

At first Marchal believed that in Maia the wall of the
end sac was composed of a single layer of cells, and he
expressed surprise that it was possible for cells to break
away bodily from such a layer without breaking the
continuity of the walls of the end sac. Finally he decided
that the wall was several cells thick in certain places, and
that it was from these places that the cells found in the

147

lumen of the end sac. had broken free. Whatever may be
the condition of things in Maia, there is no doubt that in
Cancer the walls of the end sac are uniformly only of one
cell in thickness. As in Maia, there are cells in the
cavity of the end sac, and I agree with Marchal that these
are epithelial cells of the end sac which have broken away.
There appears to be nothing surprising that certain cells
of this single layer of epithelial cells should be gradually
nipped oft by the activity of the surrounding cells and
thus shed into the lumen of the end sac. An examination
of serial sections through the end sac reveals cells in
every stage of this process of shedding. This explains
why certain cells of the end sac epithelium project farther
into the lumen than others, as described above.

Cuenot investigated the excretorv organs of Cms-
tacea by injecting various colouring matters into the body
of the living animal. lie has placed these colouring
matters into three groups, according to where they were
eliminated.

(1) Fuchsin acid, Bismarck brown, safranin, indigo-

carmine, etc.

(2) Methylene green, ammonium carminate, etc.
(-3) Methylene blue.

After such injections it wTas found that the cells of
the renal tube of the labyrinth and the cells of the bladder
have a decidedly alkaline reaction and excrete the
substances of the first category. The cells of the end sac,
on the other hand, have an acid reaction and eliminate
. the substances of the second category.

By the above method of injection Cuenot discovered
that in addition to the antennary gland and its connec-
tions there are two other kinds of excretory cells, viz. : —
The ferment cells of the digestive gland, and the cells
in the branchial septum.

148

Development of the Antennary Gland.
According to Waite,* who studied the development of
the gland in Homarus, the end sac alone arises from the
mesoderm when the embryo is only five days old, and its
lumen is for a long time completely enclosed by meso-
dermal cells. An ectodermal ingrowth (which ultimately
becomes the renal tube of the labyrinth) occurs at about
the twenty-eighth day. The lumen of the end sac does
not become continuous with that of the labyrinth until the
embryo is nearly 300 days old. The complications in the
walls of the renal tube to form the labyrinth do not occur
until the third larval stage. The bladder is formed by a
dorsal outgrowth of the ectodermal tube. Therefore, only
the cavity of the end sac represents part of the original
coelom. The renal tube of the labyrinth, the bladder and
the ureter are all ectodermal in origin. Although the
development of the green gland in the Brachyura has
not been investigated, there is no reason to believe that
it presents any striking differences from that of the
Ma crura.

(ii) Ferment Cells of the Digestive Gland.

Part of the contents of the ferment cells are of an
excretory nature, and when the contents of these cells
pass down the digestive tubules into the mid-gut the
excretory products are separated away and are carried
to the exterior along with the faeces. The ferment cells
of the digestive gland lake up methylene blue when this
colouring matter is injected into the body of the living
animal. The large coloured mass inside each ferment
cell is coloured light blue. Inside this mass there are
small bodies, which take a dark blue stain. According
to Cuenot, in five days after the experiment the blue

* Waite. Bull. Mm. Harv., Vol. XXXV, No. 7 (1899).

14!)

stain will pass out into the lumen of the tubule and thence
to the alimentary canal, where it is got rid of along with
the excrement.

(iii) Branchial Excretory Organ.

This is found in the gills in that portion of the tissue
situated between the afferent and efferent branchial veins.
According to Cuenot, the excretory cells are continued
along the sides of the branchio-cardiac veins When the
crab is subject to Cuenot's system of injection, it is found
that the cells of the branchial excretory organ act like
the cells of the end sac They have an acid reaction and
eliminate substances of the second category.

NERVOUS SYSTEM.

The nervous system of the Brachyura (Carcinus
maenas) has been investigated in detail by Bethe.* It is
proposed to give here only a short account of the nervous
system of Cancer.

The nervous system may be described briefly as
consisting of two main nerve masses the fused pre-oral
cerebral ganglia and the fused ganglia of the post-oral
region. The two are connected by a pair of commissures
which pass round the oesophagus. With the cerebral
ganglia are connected the nerves supplying the eye,
antennules and antennae. All the post-oral appendages
and somites are innervated from the posterior nerve mass
which lies in the thorax. At each side of the oesophagus
there is a ganglion on the commissure, from which arises
the stornatogastric nerves supplying the fore-gut.

The brain is the centre of co-ordinated movement,

and each ganglion of the ventral nerve mass is the reflex

centre for the appendage which it supplies.

* Bethe. Arch. f. Mikr. Anal., Bd. XLTV (1895), pp. 579-622;
Bel. L (1897), pp. 460-516, 589-639 ; Bd. LI (1898), pp. 382-452.

150

The Brain.

The Brain (cerebral ganglia, e.g.) is situated above
the anterior end oi the epistoma. It is roughly
rectangular when viewed from above, and is formed of a
eomplex mass of nerve cells and fibres. The anterior half
of the brain is connected with the nerve fibres supplying
the eye and optic peduncle. From the posterior half
arise the nerves innervating the anteimules, antennae, the
sense organs of the anteimules, and also the nerves
supplying the integument. The nerve fibres from the
different parts become aggregated in the brain in definite
masses or neuropiles (" Punktsubstanz," Leydig).
Between the neuropiles there are numerous nerve cells
more or less closely packed. The main neuropiles are as
follows : —

Connected with the Optic fibres. The antero-superior,
median and posterior optic neuropiles, all situated in the
anterior half of the brain. Also the inferior optic
neuropile extending below the others.

Connected with the oculomotor fibres: The paired
lateral oculomotor neuropiles situated on the outer side of
the optic neuropiles, and also a median oculomotor
neuropile lying in the middle line behind the optic
neuropiles.

Connected with the first antenna (antennule). The
median neuropile of the first antenna situated in the middle
line below the median oculomotor neuropile. The lateral
neuropiles of the first antenna situated ventrally at each
side, behind the inferior optic neuropiles.

Connected with the second antenna. There are three
neuropiles situated at each side of the posterior region of
the brain, viz., the median, posterior and lateral neuro-
piles of the second antenna.

Connected with the tegumental'}- nerves. The

* 151

superior and inferior leyumentary neuropiles situated in
the posterior region of the brain, slightly anterior to the
neuropiles of the second antenna.

At each side of the brain there is a globular mass of
radiating fibres lying between the oculomotor and the
tegumentary nerves. This is the globulus. There is a
tract of nerve fibres extending1 from the globulus to the
median optic neuropile. It is probable that some of the
fibres of the otocyst nerve arise from the globulus.

The chief groups of nerve cells in the brain are as
follows : —

The supero-median cells are on the inner side of the
antero-dorsal region of the brain.

The infero-median cells are situated on the ventral
side of the posterior part of the brain near the median
line.

The supero-lateral cells are situated dorsally on the
inner and anterior side of the globulus.

The infero-lateral cells extend along the inner side of
the globulus on the ventral side of the brain.

The anterior globular cells are situated ventrally on
the inner and anterior side of the globulus.

The posterior globular cells lie above and behind the
globulus at its outer and posterior side.

The following nerves arc connected with the brain
(fig. 68):-

Optic nerve [p. n.). One pair. Each arises from the
dorsal side of the brain at its anterior corner. The fibres
are continuous with those of the various optic neuropiles.
It runs outward and forward at an angle of 45° with the
longitudinal axis, and passes into the distal part of the
optic peduncle, where the latter articulates with the
proximal part of the peduncle. In the swollen part of
the peduncle the nerve becomes enlarged to form the optic

152 *

ganglion (fig. G9, o.g.). From this ganglion nerve fibres
are given off, which pass through the basement membrane
and innervate the various ommatidia of the eye (see figs.
69, TO).

Oculomotor nerve [qm. n.). One pair. These arise
from the brain immediately behind the optic nerve. Each
passes outward behind the optic nerve, and supplies the
optic peduncle and the muscles connected with it.

Antennulary nerve [<i.ln.). One pair. At each side
this nerve appears to be single, but it is composed of two
different kinds of fibres, having different functions and
arising from different centres in the brain. For this
reason the two nerves should be described separately, as
in the case of the optic and oculomotor nerves. The main
branch innervates the antennule, and is connected in
the brain with the fibrils of the median neuropiles. The
fibres innervating the olfactory and tactile setae are
probably connected with the lateral neuropiles. The
otocyst branch is connected with the lateral neuropiles,
and also with the globulus. It passes out from the brain
together with the fibres of the main branch, but the two
kinds soon separate and those innervating the otocyst pass
outwards, and the fibres are connected with ganglion cells
at the base of the otocyst setae. The antennulary nerve
arises from the middle of the ventral surface of the brain,
but in Cancer it apparently leaves the brain near the
anterior end.

Tegumentary nerve (/. n.). One pair of broad nerves,
which leave the brain near the postero-lateral corner.
Each passes outwards almost at right angles to the longi-
tudinal axis, and divides into two main branches. The
anterior branch supplies the integument in region of the
rostrum. The outward branch sweeps backwards and out-
wards, giving off small branches to the integument.

153

Antennary nerve (a.2 n.). One pair of small nerves,
which arise from the ventral side of the brain immediately
behind the tegumentary nerves. They are connected
with the tibres of the neuropiles of the second antenna.
Immediately on leaving the brain each passes beneath the
tesrumentary nerve and enters the muscle chamber of the
second antenna.

The commissures (Com.). One pair of rather large
nerves, which arise from the posterior side of the brain.
They pass posteriorly on each side of the oesophagus, and
are connected with the ventral nerve mass in the thorax.
At each side of the oesophagus there is a paraesophageal
ganglion (p. </.), and behind the oesophagus the two
commissures communicate by means of a transverse post-
oesophageal connective (n. po.). From the paroesophageal
ganglia arise the stomatogastrie nerves, and immediately
behind the ganglion there is a small nerve (n. m.)
supplying the muscles of the mandibles.

A' e n t r a 1 Nerve Mass.

The ventral nerve mass (t.g.) is situated above the
sternal artery in the third and fourth thoracic somites.
It is partly supported by the median plate of the endo-
phragmal system. It contains nerve elements repre-
senting the post-oral cephalothoracic ganglia and the
abdominal ganglia. It is not possible to make out all
these ganglia from dissections, but they may be
distinguished in stained preparations and in sections.
Slightly behind the centre of this fused ganglionic mass
is a foramen (n. f.), through which passes the descending
artery.

At its anterior end, where it is connected with the
commissure, the ventral nerve mass is narrower than in
its posterior portion. In this anterior portion it is

154

possible to make out, in suitable preparations, evidenees
of six pairs of ganglia belonging to the last three cephalic
and the first three thoracic somites. Behind this region
five pairs of larger ganglia may be distinguished more
readily. From these arise the nerves supplying the live
pairs of pereiopods and the last five thoracic somites.
Between the last two ganglia is a median swelling, which
is composed of the six ganglia belonging to the abdominal
somites.

The following nerves arise from the ventral nerve
mass : —

First nerve (//. 1). One pair. Each arises close to
the connections between the commissure and the ventral
nerve mass. It contains fibres from the hist three ganglia
of the nerve mass. It passes forwards parallel to the
commissure and divides into three branches, passing
to the mandible, first maxilla and second maxilla
respectively.

Second nerve (n. 2). One pair. Each arises behind
the first nerve, and its fibres are connected with the cells
of the fourth ganglion of the nerve mass. It passes
forward and slightly outward, and supplies the various
parts of the first maxillipede.

Nerves 3-9 (n. •'!-/;. 9). These are all paired, and are
connected with the fifth and following; ganglia of the
ventral mass. Each nerve supplies the appendage of its
own somite, and its course is very similar to that of the
second nerve. Nerves 5-9, however, are larger than the
others, and the posterior nerves pass outwards and back-
wards.

Abdominal nerve (n. ah.). This is a median nerve,
which passes backward in the median line. It arises from
the fused abdominal ganglia at the posterior end of the
ventral nerve mass, and contains nerve fibres from all the

155

abdominal ganglia. It passes backward along the median
line of the "sella turcica" and along the ventral region
of the abdomen. Small nerves are given off to each
somite of the abdomen.

According to Jolyet and Yiallanes,* the centres of the
moderator and accelerator nerves of the heart are in the
anterior part of the ventral nerve mass. The cardiac
nerve of the Macrura does not appear to be present in the
crabs.

The 8 t o m a t o gastric System.

This consists mainly of nerves arising from the
paroesophageal ganglia. These nerves fuse to form the
stomatogastric nerve, which supplies the anterior and
dorsal regions of the fore-gut. There is also the postero-
lateral nerve arising behind each paroesophageal ganglion.

There are two main nerves arising from each
paroesophageal ganglion. These pass below the com-
missure, and, passing forward, they fuse with each other
and with the similar nerves of the other side to form the
main stomatogastric nerve. The ventral nerve (st. i.)
arises from the outer side of the ganglion, and
immediately passes below the lateral dilator muscle of the
oesophagus, to which it gives small branches. It passes
below the commissure and innervates the anterior dilator
muscles of the oesophagus. It then turns upward and
fuses with the similar nerve of the other side and with the
dorsal nerve. The latter nerve (st. s.) arises from the
anterior and outer side of the paroesophageal ganglion.
It passes immediately below the commissure, and gives off
a small nerve to the anterior Avails of the oesophagus. In
the median line it fuses with the corresponding nerve of
the opposite side, and also with the ventral nerve.

The stomatogastric nerve (si. n.) is formed by the

* Jolyet ct Yiallanes. Comptes Bendus, CXIV (1892), p. 189.

156

fusion of the two ventral and dorsal roots, as described
above. It passes up the anterior wall of the cardiac fore-
gut as a median nerve. About half way along this wall
the nerve enlarges to form the stomato gastric ganglion
(st. g.). From this point the nerve proceeds backward
along the dorsal side of the fore-gut, and almost
immediately gives off two large branches — the lateral
gastric nerves (7. g. n.). Each lateral gastric nerve passes
over the anterior cardiac muscle, which it innervates, and
gives rise to a nerve plexus in the dorsal walls of the
cardiac fore-gut. Both nerves also give rise to an internal
branch, which passes inwards and joins again with the
stomatogastric to form a large ganglion. Behind this
ganglion the stomatogastric nerve bifurcates. Each
branch, which is known as the posterior gastric nerve
(p. </./!.), passes backward to the pyloric fore-gut. Here
the two nerves join again, thus forming a ring. From the
posterior end of this ring three nerves are given off — one
to the hind-gut (n. /.), one to the digestive gland (n. I.) and
one to the integument (n. t.). The muscles in this region
of the fore-gut are also innervated from the posterior
gastric nerves.

The postero-lateral nerve (p. n.) of each side arises
from the inner side of the commissure, immediately
behind the paroesophageal ganglion. It passes backward,
and innervates the posterior dilator muscle of the
oesophagus. It then passes upward along the posterior
wall of the fore-gut, and supplies the muscles in this

region.

Sense Oegans.

The Eye.

Each eye is situated at the distal extremity of the
long optic peduncle. The peduncle is clearly divided into
two portions: — (1) A long narrow proximal part, which

157

meets the peduncle of the other side in the median line
above the first cephalic sternum. This proximal portion
is not seen from the exterior. (2) A shorter and broader
distal portion, which is the part seen from the exterior.
The two parts are separated by a region of soft chitin,
which allows of considerable movement between the two
parts. At the distal extremity of the peduncle is a small
black area — the cornea — which defines the external
surface of the eye. The cornea would be almost circular
were it not for the presence of two small tubercles which
invade the dorsal border of the black area, and thus give
it a somewhat irregular shape. The cornea is composed
of a thin layer of chitin, which is continuous with
the thick, strongly calcified integument covering the
remainder of the distal part of the peduncle.

The thick calcified integument has the usual layer of
epidermis beneath, resting upon a basement membrane.
In the region of the eye, the cellular layer on the inner
side of the cornea is extremely thick, and is known as the
ommateum. On the inner side of the ommate.um is a well-
defined basement membrane, which is continuous with
that of the ordinary epidermis. The ommateum,
therefore, may be regarded as a differentiation of the
epidermis, with which it is continuous.

An examination of the surface of the cornea reveals
the presence of numerous small hexagonal facets. Vertical
sections through the eye show that this sub-division of the
cornea is not merely superficial, but is continued through
to the inner side of the chitinous layer. The cornea may,
therefore, be said to be composed of numerous hexagonal
prisms, which are packed closely together. On the inner
side of the cornea the ommateum is divided into numerous
elements, which correspond in number and position to the
corneal facets. Each element of the ommatenrn is known

158

as an ommatidium and fits below a corneal facet, and
extends from the cornea to the basement membrane. All
the ommatidia have the same essential structure.

Structure of <tn ommatidium.

Each ommatidium may be conveniently divided into
proximal and distal regions. The outer parts of botli
these regions are defined by the presence of pigment
(fig. 69, pg. L, pg. o.).

Underlying the cornea is the flattened corneagen.
Below this are the vitrellar cells, the distal region of
which enclose the crystalline cone. The vitrellae are
surrounded by pigment cells, which are densest on a level
with the proximal region of the crystalline cone. At the
proximal end of the vitrellae are the pigmented retinulae
which surround the rhabdome. The retinulae are in
contact with the nerve fibres from the optic ganglion, and
the proximal end of the rhabdome is in contact with the
basement membrane.

The corneagen [" corneal hypodermis," Parker]
(fig. TO, corn.) lies immediately below the cornea, which
is a product of the corneagen cells. It consits of two
flattened tile-like cells.

The Yitrellae [" cone cells," Parker] (vit.) lie
immediately below the corneagen. Transverse sections
show that there are four vitrellae in each ommatidium.
Their distal extremities are applied to the base of the
corneagen. Passing inward the cells become narrower.
Distally they enclose the crystalline cone, which is
secreted by these cells. Surrounding the vitrellae are
two pigment cells (pg. c), which are continued inward as
fine processes and eventually come into contact with the
retinulae. (Fig. 72 is a section through the vitrellae of a
single ommatidium, and shows that there are four vitrellar

159

cells surrounded by two pigment cells.) Between the
vitrellae and the pigment cells there are intercellular
spaces, which are larger at the proximal end of the
vitrellae.

The retinulae (ret.) are seven pigment cells which
surround the rhabdome, and extend from the proximal
portions of the vitrellae to the basement membrane.
Distally each retinula ends in a rounded knob, which
contains a nucleus (fig. 71 shows the disposition of the
retinulae around the rhabdome). At their distal ends
the retinulae are extremely large, and surround the
rhabdome in this region so as to completely hide it. The
concentration of the pigment in this region gives rise to a
well-marked pigment band (fig. 69, pg. i.). Proximally
the retinulae are not so large, and the rhabdome can be
seen quite distinctly between them. The rhabdome has a
peculiarly striated appearance, which is caused by the
arrangement of the pigment granules of the retinulae.
The optic nerve fibres pass into the retinular cells, so that
it is this part of the eye which is sensitive to light.

The rhabdome (rhab.) is a rod-like structure in the
centre of the retinulae. According to Watase,* the
rhabdome is a chitinous structure produced as a secretion
from the retinulae. Parkert states that transverse
sections show that the rhabdome is composed of four
parallel rods, and he further affirms that these four rods
are the inner extremities of the four vitrellae (cone cells).

The optic nerve fibres perforate the basement
membrane and end in the retinulae.

Each ommatidium, therefore, may be said to consist

* Watase, S. "On the Morphology of the Compound Eves of
Arthropods," Q. J. M. S., Vol. XXXI, p. 143.

t Parker, G. H. " The Histology and Development of the Eye in
the Lobster," Bull. Mus. Comp. Zool. Harv., Vol. XX, No. 1.

"The Compound Eyes in Crustaceans," loc. cit., Vol. XXI,

No. 2.

160

of a series of cells which on their outer side receive and
concentrate the rays of light. These are transmitted to
the retinulae by means of the transparent vitrellae. The
retinulae are connected with the nerve fibres, and are,
therefore, the important part of the eye. Each
ommatidium is surrounded by a layer of pigment, so that
the sensitive retinulae are situated at the bottom of a
tube, which is completely separated from the tubes of the
other ommatidia because of the presence of the pigment.
Therefore, the retinulae of an ommatidium can only
receive lie-lit through the small corneal facet at the distal
extremity of the particular ommatidium.

The cornea (en.) is composed of three layers. On the
outside is a thin structureless layer — the cuticle (cut.).
Beneath this is an outer pigmented layer (pig. I.), to which
the colour of the cornea is due. Below the pigment layer
is the deeper layer. Both inner layers exhibit longi-
tudinal striations.

According to Watase, the corneagen, vitrellae and
retinulae are all modified epidermal cells, which, in the
case of the vitrellae and retinulae, have grown inward.
The cornea is a chitinous secretion of the corneagen. The
crystalline cone is primarily a chitinous secretion of the
vitrellae, and the rhabdome is a chitinous rod secreted by
the cells of the retinulae.

The otocyst.

The otocyst is found in the basal segment of the
antennule. It is a sac lined with a layer of chitin, which
is continuous with the outer chitinous integument. In
the megalopa stage and in the young crabs the otocyst is
open to the exterior, and in these stages otoliths are
present in the sac. In large crabs, however, there are no
otoliths, and the sac is completely closed except

161

immediately after ecdysis, when the otoeyst is connected
with the exterior by means of a transverse slit on the
dorsal side of the basal segment of the antennnle. This
slit is quite obvious in the hard crab. The walls of the
sac project into the lumen at three places— (1) in the
postero-lateral wall (2) in the posterior wall, and (3) the
anterior part of the floor.

There are three kinds of setae situated on the walls of
the otoeyst and projecting' into the lumen.

The hooked setae [" Hakenkaare," Hensen] (PI. XI,
fig. 75) are found on the convex portion of the posterior
wall of the otoeyst. These setae are characterised by
having the distal extremity inclined at a considerable
angle to the proximal portion, sometimes as much as 90°.
The distal portion of the shaft has fine barbs arising from
it. The base of the shaft is sunk into a socket-like
depression. The total length of the hook hairs is abont
55//.

The thread setae [" Fadenhaare," Hensen] are
found on the anterior part of the floor. They are the
largest setae present in the otoeyst, and are about six
times as long as the hook hairs. There are well-defined
barbs arising from the shaft. The base of each seta lies
in a cup-shaped depression of the chitinous wall.

The grouped setae [' Gruppenhaare," Hensen]
(fig. 76) are situated on the lateral walls of the otoeyst,
below the closed entrance of the sac. They are about
140/* in length, and have blunt shafts. There are no
barbs present.

The three kinds of setae are innervated from the

otoeyst nerve, and according to Prentiss* there is one

nerve element for each seta. The same author states that

* Prentiss, C. W. "The Otoeyst of Decapod Crustacea: its
structure, development, and functions," Bull. Mtis. Comp. Zool. Harv.,
Vol. XXXVI, No. 7.

M

162

the otoeyst does not become functional until the megalopa
stage.

Hensen and the early observers believed that the
otoeyst was an auditory organ. Hensen found that the
" auditory " setae were individually sensitive to sound
vibrations of different frequency, and concluded that in
the lobster the auditory organ had a range of three
octaves.

Kreidl was the first to deny that the otocysts
possessed an auditory function. He substituted iron
filings for the otoliths, and the experiments led
him to believe that this organ served the function of
equilibration.

Bethe, while accepting Kreidl's results, did not reject
the idea that the otoeyst was also an auditory organ.

The researches of Prentiss led him to believe that the
otoeyst was a static organ solely. It is probable that in
the Brachyura the hooked setae and grouped setae have
lost most of their functional activity owing to the absence
of otoliths. The thread setae are, undoubtedly, the most
important sensory organs of the otoeyst. (For a further
discussion of this subject the reader is referred to the
paper by Prentiss.)

The sensory setae of Cancer are of two kinds.

The tactile setae may be present in various parts of
the body. They are found on the antennules and
antennae, and on most of the appendages. The setae of
the otocysts are probably modified tactile setae. They
are characterised by having a long tapering shaft which
bears barbs, and each seta is innervated by a single nerve
fibre and a single nerve cell.

The olfactory setae (fig. 74) are short and blunt, and
are much more firmly attached to the integument than
the tactile setae. Each seta is divided into proximal and

163

distal portions by a transverse suture. At the distal end
the seta is either perforated or the membranous covering
is extremely thin. The olfactory setae differ from the
tactile setae in having numerous nerve elements for each
seta. They are present on the exopodite of the antennule
on the side opposite to the long setae. They are extremely
small, and there are only one or two on each ring- of the
exopodite. On the endopodite of the antennule there are
also a few small setae on the dorsal side of each ring,
which have the appearance of olfactory setae. The
gustatory setae present in the region of the mouth are
modified olfactory setae.

REPRODUCTIVE ORGANS

(Plate XII).

1. Male (fig. 78).

The abdomen is much narrower than in the female,
and the third, fourth and fifth abdominal somites are
fused together. Abdominal appendages are present only
on the first and second abdominal somites, and these are
modified to form copulatory organs. The external genital
apertures are paired, and each is situated at the tip of a
membranous papilla on the ventral side of the coxopodite
of the last walking leg. The thoracic sterna are deeply
concave, in order to receive the closely-applied abdomen.
The locking apparatus of the male abdomen is much
better developed than in the female. The dorsal side of
the carapace is much flatter in the male than in the female.
As Williamson has pointed out, the lobed antero-lateral
border of the carapace is slightly turned up in mature
males. The chela of the mature male are larger than
those of mature females of the same size.

The two abdominal appendages of each side form a

164

single copulatory organ (see fig. 16). The anterior
appendage (p. 1) is tubular, and into the tube the
posterior rod-like appendage (/>. 2) is inserted during
copulation. The copulatory organ of each side is
introduced into one of the vulvae of the females during
fertilisation. The genital papillae of the male are too
short to reach the vulvae, and the abdominal appendages
have become modified to form sexual organs in conse-
quence of this. During copulation each genital papilla
tits into the base of the tubular appendage and the
spermatophores are poured into the tube. The rod-like
second appendage is constantly working up and down the
tube, and thus forces the male sexual products into the
spermatheca of the female.

The testes (test.) are paired and symmetrical, and the
two halves are connected immediately behind the fore-
gut. Each testis is a compact lobulated organ situated
in the antero-lateral region of the cephalothorax. It is
superficial in position, and lies immediately below the
dermis and above the digestive gland. The size and
shape of the testis varies considerably according to the
condition of the animal. In immature crabs it may be
extremely small, but in the mature specimens the testis
is massive and lobulated, and may cover almost the whole
of the digestive gland. Its blood supply is obtained from
the large spermatic branch of the lateral artery. At its
inner extremity, near the cardiac fore-gut, each testis is
connected with the vas deferens (v. d.). The main part
of the testis gives off a posterior branch which passes
backward beneath the vas deferens alongside the mid-gut
caecum. Above the mid-gut this prolongation turns
inward and joins with the similar portion from the other
side to form a bridge behind the pyloric fore-gut.

The vasa deferent in are a pair of long convoluted

165

tubes passing- backward from the testes to the posterior
region of (lie thorax, where each opens to the exterior on
(he coxopodite of the last walking leg.

In a mature crab the course of each of the vasa
deferentia is as follows :— Where it arises from the inner
portion of the main lobe of the testis the vas deferens is
an extremely convoluted and narrow tube. As it sweeps
round the outer side of the cardiac stomach the tube
grows broader and the convolutions become less compli-
cated, so that in this region the course of the vas deferens
may be traced without much difficulty. At the level of
the pyloric fore-gut the duct turns suddenly inwards and
covers the lateral walls of this region of the alimentary
canal. From this point the vas deferens passes backward
as a white convoluted tube above the hind-gut and below
the pericardium. The maximum width is attained below
the anterior region of the heart. Behind this point the
duct gradually becomes narrower, and is then known as
the ejaculatory duct [e. <!.). At the posterior end of the
pericardium the duct dips downward through the foramen
on the outer side of the " sella turcica." It then passes
behind the dorsal extensor muscle of the coxa of the last
walking leg and in front of the extensor muscle of the
basi-ischium of the same appendage. Below the latter
muscle it passes beneath the flexor of the basi-ischium,
and opens to the exterior at the end of a soft papilla
situated on the ventral surface of the coxa of the last
walking leg.

In sections through the testis the gland is seen to be
composed of numerous follicles which are closely packed
together. Each follicle is lined by epithelial cells which
surround a central cavity, and this cavity is continuous
with that of the vas deferens. In an early stage the cells
of the follicles are not differentiated and are quite small.

166

Eventually some of these become spermatoblasts and
increase greatly in size. Each spermatoblast gives rise
to a large number of spermatozoa. In sections through
a mature testis it is difficult to distinguish the follicles,
as the cavities of the latter become almost obliterated by
the growth of the spermatoblasts. When the spermatozoa
are ripe they break free and are carried down into the
vas deferens. Here they become collected together in
small groups, and each group becomes surrounded by a
capsule to form a spevmatophore.

The spermatozoa (fig. 81) are non-motile. They
consist of a central dark portion containing a nucleus
and an outer clear margin. In side view the clear margin
is seen to be much thinner than the central part of the
cell, and has the appearance shown in fig. 81. The
diameter of the cell is about b>.

The presence of the spermatophores gives rise to the
white appearance so characteristic of the vasa deferentia
of mature male crabs. In small immature crabs the vas
deferens is narrow and almost transparent, due to the
absence of spermatophores. As pointed out by
Williamson, " the condition of maturity in the male is
the presence of ripe male elements in the vas deferens."
It is probable that most male crabs above 4^ inches in
width are mature, although Williamson has given
instances of male crabs attaining maturity at a much
smaller size.

2. Female (fig. 77).
The abdomen is broad, and all the somites are freely
movable. There is one pair of appendages on each of the
second, third, fourth and fifth somites. After the ova are
spawned they are attached to the endopodite setae of the
abdominal appendages. The external genital apertures

167

are a pair of Large opening's situated on the1 sternum of the

sixth thoracic somite. The thoracic sterna are not so
concave as those of the male, and the abdomen is not so
closely applied to the thorax. The abdominal locking
apparatus is poorly developed. The dorsal side of the
carapace is well arched in the mature females, and the
antero-lateral border of the carapace is not upturned as in
the males.

The ovaries (ov.) are paired, and lie in a similar
position to the male reproductive organs. There is, how-
ever, a considerable posterior prolongation of each ovary.
As in the male, the two antero-lateral portions are
connected behind the pyloric fore-gut by a strand of
gonadial tissue, which forms a bridge over the mid-gut.
Behind this transverse connection each ovary is prolonged
backward as a narrow strip, which extends to the extreme
posterior end of the thorax. At the posterior extremity
the two prolongations fuse together in mature specimens.
These backward extensions of the ovaries occupy a
similar position to the vasa deferentia of the male; that
is to say, they lie above the hind-gut and below the
pericardium. Beneath the anterior end of the latter each
posterior branch is connected on its outer side with a
large sac, the spermatheca* (spt.). Each spermatheca is
continued into a short ocidwt (ovd.), which opens to the
exterior by means of the vulva on the sternum of the
sixth thoracic somite.

The condition of the ova in the ovary naturally
depends upon the degree of maturity attained by the
gonads. The immature gonads are small and pale, and
no evidence of the presence of eggs can be detected by the
naked eye. The mature gonads, however, fill almost the

* In young crabs the spermatheca is extremely small, and can
only be made out with difficulty.

168

whole of the dorsal side of the cephalothorax. They are
of an orange-red colour, and the separate eggs can be
readily distinguished. The red colour is due to the
presence of the food-yolk. The yolk granules form the
main part of the mature ovum (fig. 79). With regard to
the condition of the ovaries between two processes of
ecdysis, the reader is referred to the section on Bionomics.

Copulation takes place immediately after the female
has cast, and while it is still in a soft condition.
Apparently the spermatozoa burst free from the spermato-
phores as soon as they leave the vas deferens. In the
spermatheca only free spermatozoa are found. After
copulation the cells lining the spermatheca secrete a fluid
which fills the cavity of the oviduct. This secretion
hardens upon contact with the sea water, and thus the
oviducts become effectively plugged, and the contents of
the spermatheca cannot escape. It is a remarkable fact
that the spermatozoa remain inside the spermatheca for
many months before they fertilise the ova.

The eggs are spawned in the winter. Upon reaching
the exterior, each egg is probably surrounded by two
membranes — an inner vitelline membrane and an outer
chorion. Between the two is a perivitelline space, which,
according to Williamson, contains a fluid possessing
adhesive properties. The eggs become attached to the
endopodite setae of the abdominal appendages. The
interesting question regarding the mode of attachment of
the eggs to the endopodite setae is not yet conclusively
settled. Some of the early observers believed that a
sticky substance was secreted around the eggs as they
were being shed. This, however, would not explain why
the eggs become attached only to the endopodite setae.
ITerriek's* explanation is that the tegumentary glands of
* Herrick. " The American Lobster," Dull. U.S. Fish Com., 1895.

169

the endopodites of the pleopods secrete an adhesive fluid.
Williamson* gives a detailed explanation of the method
of attachment. He suggests that the endopoditic seta
penetrates the chorion of the egg in two places, and thus
the egg becomes skewered on the seta. The piercing of
the chorion liberates the adhesive perivitelline fluid,
which assists in making the attachment more permanent.
The chorion eventually becomes drawn out at the point of
attachment, and the egg appears to be attached to the seta
by a stalk (see hg. 80).

According to AVilliamson, the number of eggs
attached to the abdomen may vary from half a million in
a small mature female to three millions in a large crab.

Development.

It is a surprising fact that the development of
Cancer 'pagurus has never been satisfactorily investigated.
The internal changes do not appear to have been followed
in any Brachyurous embryos, and although the general
characters of the larval developments in the Brachyura
are well known, our knowledge of these stages in Cancer
pagurus is extremely scanty.

The development may be divided into three stages —
embryonic, larval and post-larval.

The embryonic development takes place while the
embryo is attached to the pleopods of the female, and,
therefore, extends over a period of about seven months.
The internal development during this period has not been
investigated, but it is very probable that the Brachyura
do not differ from the Macrura in this respect. The early
development has been thoroughly investigated in the
Macrura.t

* Williamson. 23rd Report, Scotch Fishery Board.

f See Herrick.F.H. " The Development of the American Lobster,"
Johns Hopkins T~iiiv. Circ, Vol. IX, 1890, No. 80.

Reichenbach, H. " Studien zur Entwicklungsgi schichte des
Flusskrebses," Ab. Senkenberg. Nat. Ges. Frankfurt, Bd. XIV, 1886.

170

While working at the Biological Station, Heligoland,
I was able to examine the early larval stages of Cancer.
[ was fortunate enough to obtain berried crabs on which
the embryos were ready for hatching. The process of
hatching lasted several hours, and during this time the crab
assisted by moving its abdomen backwards and forwards.
The last walking legs were also used for the purpose
of detaching the larvae from the pleopods. Unfortunately,
I was not successful in keeping the larvae alive longer
than the first zoea stage.

I give below a summary of the characters of the
larval stages of the Brachyura in general. I have
utilised the results of other investigators,* and have also
added my own observations. The larval stages may be
divided into Protozoea, Zoea (four kinds) and Megalopa.

Protozoea (PI. XIII, figs. 83, 84). Hatching takes
place at this stage, which is of very short duration.
There are no frontal or dorsal spines present. The lateral
spines (sp. I .) are present one on each side of the cephalo-
thoracic shield. The large paired eyes are present in the
head. The abdomen is well defined, but only five somites
and the telson can be distinguished. The sixth somite
is at this stage fused with the telson. The telson is
forked, and each branch bears strong spines (fig. 84).
All the cephalic and the first two pairs of thoracic limbs
are present. The antennule is blunt and consists of two
segments, of which the distal is the larger. The antenna
consists of a broad basal joint, from which is given off a
short pointed process. The mandible is a small rounded
outgrowth. The first and second pairs of maxillae are

* Williamson, H. C. " On the Larval and Early Stages and Rate
of Growth of the Shore Crah (Carcinus mamas)," Tiventy-first Annual
Report, Fishery Board for Scotland, p. 136.

Korschelt and Heider. Text Book of Embryoloqy (Invertebrates,
Part II). '

171

similar, and are beginning to show evidences of a
biramose structure. The first and second pairs of niaxilli-
pedes are large and biramose. The protopodite is large.
The endopodite and exopodite have few setae. Towards
the end of the protozoea stage the cuticular covering
becomes very loose, and beneath can be seen the
developing organs of the first zoea stage. Chromato-
phores are present at the sides of the body.

First zoea (PI. XIII, figs. 85, 80, 87). This stage
is generally seen about three or four hours after hatching,
and probably lasts for at least ten days. The important
difference between this and the previous stage is the
presence of the large frontal and dorsal spines. The
latter is about half as long as the body, and the frontal
spine is about two-thirds the length of the body. Both
are tipped with a red pigment. The lateral spines are well
developed. The branched chromatophores are well
developed. The number of appendages appears to be the
same as in the previous stage, but they are more highly
developed. Each antennule bears a group of setae
at its tip. The antennae and both pairs of
maxillae are biramose. The exopodites of the maxilli-
pedes each bear four long setae. the third maxillipedes
and the pereiopods and the associated gills are present as
extremely small buds, which are hidden beneath the
cephalothoracic shield. The pleopods may be seen for the
first time as extremely small tubercles. In some examples
of both the protozoea and the first zoea there was a pair of
tubercles present on the second abdominal somite. The
telson differs from that of the previous stage in having
two extremely long posterior branches.

Zoea stages II, III and IY. I have not been able to
identify these stages in Cancer pagurus, but in Carcinus
maenas they have the following essential characters. They

172

are very similar to the Hist zoea in appearance. Between
the two branches of the antenna an outgrowth takes place
in the second stages, and ultimately developes into the
long flagellum. The maxillae are not very different from
those of the first stage. The setae on tlie exopodites of the
first and second maxillipedes increase in number at every
stage. The last six pairs of thoracic appendages and
their gills gradually increase in size, but never become
functional during the zoea stages. The pleopods
gradually develop until at the fourth stage there are five
pairs present. There is not a pair present on the sixth
abdominal somite in Carcinus. At the third stage the
sixth abdominal somite becomes separate from the telson.
The rostral and dorsal spines gradually become shorter.

In his account of the larval stages of Cancer irroratus,
Smith describes the following characters in the last zoea
stage: — Rostral and dorsal spines short. The abdominal
legs are seen as stumpy outgrowths. The third niaxilli-
pede is well developed, but the other posterior thoracic
appendages do not project below the edge of the cephalo-
thoracic shield. The flagellum of the antenna is present.

Megalopa stage. The main points of difference
between the zoea and megalopa are as follows: — The
frontal and lateral spines disappear. According to
Smith, the dorsal spine of Cancer irroratus persists as a
small backwar dry-projecting process. The carapace is
broader. The pereiopods are well developed, and the
gills are probably functional. The pereiopods are never
biramose as in the Macrura. The abdomen is macrurous,
and the pleopods are used for swimming. The telson is
much shorter, and loses its spines. The megalopa is still
a pelagic stage. Its pereiopods, however, may be used for

' Smith, S. " The Invertebrate Fauna of Vineyard Sound," U. S.
Fiali Commission Report, 1871-72 (published 1873).

173

walking on the bottom. The last pereiopod in Carcinus
maenas and Cancer irroratus has a tuft of setae on the
dactylos. According to Smith, the megalopa stage is
very short, and at the first moult it changes into a young
ad nit.

Post-laryal stages. The young adult differs from the
megalopa in having the abdomen tucked beneath the
thorax. It is no longer a pelagic animal, but lives on the
bottom, and uses the last four pairs of pereiopods for the
purposes of locomotion. As pointed out by Smith* and
Cunningham, t the early post-larval stages differ con-
siderably from the larger specimens. The carapace is
elongated and the rostral region is well developed. The
lobes of the antero-lateral border are sharp. Cunningham
pointed out that it is difficult to distinguish the early
stages of Cancer from those of Atelecyclus heterodon. At
each succeeding moult the transverse axis of the carapace
increases more rapidly than the longitudinal axis.

ECONOMICS AND BIONOMICS.!

The main features in the life-history of the edible
crab may be briefly summarised as follows: —

Cancer pagurus is found all round the coasts of the
British Isles, being especially abundant on the rocky
coasts. The size at which maturity is attained is variable,
but most crabs above five inches in breadth may be said
to be mature. There appear to be no records of large

* Smith. Op. cit.

t Cunningham, J. T. "On the Early Post-larval Stages of the
Common Crab (Cancer pagurus), and on the Affinity of that species
with. Atelecyclus heterodon," Proc. Zool. Soc, 1898, Part IT, p. 204.

I For further information on these subjects see —

Cunningham. Cornwall Count]! Council : Report of the Executive
Committee for Fisheries, 1897-8. Penzance, 1898.

Williamson. Fishery Board for Scotland : 18th, 22nd, and 23rd
Annual Reports.

Wilson. Northumberland Sea Fisheries Committee: Reports on

174

edible crabs, but I have seen several specimens in shops
having a carapace breadth of about twelve inches. A crab
measuring nine inches would be considered a large one.
The crabs are captured by means of crab-pots (" creels,"
Williamson; " creaves," Wilson), which are baited with
fish. In Port Erin the chief fishing season is from March
to September, but the crab fishery is continued through-
out the winter. Generally speaking, the chief fishing
season in the British Isles lasts from the early spring to
the autumn. In some districts, such as the North-East
of England, there is a close season.

Fishermen are not allowed to sell crabs below four
and a half inches, berried crabs or soft crabs, but these
restrictions do not hold good concerning crabs used for
bait. Since, in some parts of the country, crabs are used
for bait to a large extent, these laws for the protection of
the crab fishery to some extent fail in their purpose.
The size limit may be increased at the discretion of the
local committees. In the Lancashire District the
minimum size is five inches.

In the summer the mature crabs frequent the inshore
waters, and in the winter they occur in the deeper off-shore
waters. The immature crabs do not take part in this
annual migratory cycle. The mature crabs cast in the
autumn, and the females are fertilised when " soft."
Spawning takes place in the deeper water in winter. The
larvae are hatched in the following summer in the inshore
waters. Wilson is of the opinion that the " berried "

the Crab Fishery, 1893 and 1895. Also Proe. P.. Soe. Edin., Vol. XX,
1894, p. 309.

Meek. Northumberland Sea Fislierics Committee, 1897-190G.

Buekiand, Walpole and Young. Reports on the Crab and Lobster
Fisheries of England and Wales, of Scotland, and of Ireland
[C. 1695], 1877, p. 56.

Statistics regarding the crab fishery may be obtained from the
various Annual Reports of Inspectors, Sea Fisheries (England and
Wales).

175

crabs feed very little, and lie records instances of such
crabs being' covered with sand. Williamson's observa-
tions appear to support this statement.

Baudouin* gives an interesting account of how Ihe
phenomenon of autotomy is utilised by the fishermen of
Southern Spain. The common edible crab of that region
is Gelasimus tangeri. When the crabs are captured the
large claws are removed up to the fracture plane, and th<>
crab is put back in the water. Only the claws are sent to
market, the crab being1 returned to the sea to grow new
ones.

The statistics published annually by the Fishery
Inspectors for England and Wales are very scanty, ami
appear to have but little value in the elucidation of the
numerous problems connected with the natural history
of the edible crab.

I give below a summary of the figures published in
the Annual Reports of the Fishery Inspectors for England
and Wales since 1887, merely giving the total number of
crabs caught in each year and their approximate value.

Year.

Total Number.

Approximate Value.

1887-1897 (average)
1898

4,669,861
5,628,114
4,918,184
5,177,350
5,325,974

4,923,536
4,580,318
5,106,345

£55,082
£67,895

1899

£62,494

1900

£56,822

1901

£58,743

1902

1903

£54,327

1904

£52,556

1905

£59,479

1906

* Baudouin, M. " Utilisation de l'autotomie chez un Crabe.
Revne scientifique (Ser. V), T. VI, No. 10.

176

Through the kindness of Dr. Jenkins, Superinten-
dent of the Lancashire and Western Sea Fisheries
District, I am able to give the following statistics dealing
with crabs landed in that Sea Fisheries District during
the years 1900 to 1906, inclusive:-

New Quay.

Aberdovey.

Pwllheli.

Holyhead.

Liverpool.

Year

No.

£'s.

No.

£'s.

No.

£'s.

No.

£'s.

No.

£'s.

1900

6,000

180

600

6

1901

—

—

—

—

2,000

60

800

8

—

—

1902

—

—

—

—

10,204

112

1,100

11

■ — | —

1903

—

—

—

—

11,143

135

1.000

11

600

5

1904

50

1

—

—

10,710

125

615

6

900

8

1905

40

2

—

—

8,851

109

760

8

466

6

1906

509

11

435

7

9.347

113

—

—

800

8

The crabs returned as having been landed at
Liverpool were certainly not caught in that
District.

The Harbour Master of Port Erin has kindly
provided me with the following figures, which give the
number of crabs captured by the Port Erin fishermen
during the years 1904-5-6: —

1904.

1905.

1906.

January

?

?
7,000
7,500
7,500
6,100
7,000
6,000
3,000
1,600
500
1,200

800
3,400
6,500
8,000
9,400
5,000
6,500
2.000
2.000
1,500
400
800

1,200

February

2,200

March

April

Mav

4,000
6,500
7,000

June

July

4,000
7,000

August

5,500

September

October

November

December

2,500

2,500

300

300

Total

47,400*

46,300

43,000

* The figures for January and February are not given.

177

Fishery regulations.*

In section 8 of the Fisheries (Oysters, Crabs and
Lobsters) Act (40 & 41 Vict. ch. 42), the following restric-
tions are imposed: —

A person shall not take or sell : —

(1) Any edible crab which measures less than four
inches and a quarter across the broadest part of the back.

(2) Any edible crab carrying spawn.

(3) Any edible crab which has recently cast.
Such crabs may, however, be used for bait.

In the Lancashire and Western Sea Fisheries
District the minimum legal size has been raised to five
inches. (Bye-law 2bb : "No person shall remove from a
fishery any edible crab measuring less than five inches
across the broadest part of the back.")

Size of crabs at maturity.

Female. — There appears to be some difference of
opinion with regard to the size at which the female
becomes mature. Wilson t had reason to believe that on
the Northumberland Coast the size of maturity is about
six inches. Williamson's! investigations, on the other
hand, show that the crabs of the east coast of Scotland
become mature when about four and a half inches in
width. I have examined crabs from Port Erin which
had been fertilised when about this size. It is probable
that a crab is mature when it has attained a size of
four and a half inches, but in many cases fertilisation may
not be effected until after the next casting. There is no

* For a discussion on this subject read Williamson, 18tJi Annual
Report, p. 134. Other literature on this point is given by him on
p. 78 of the same report.

f Wilson. Northumberland and Sea-Fisheries Committee, 1893,
1895 ; Proceedings Royal Society, Edinburgh, 1892-3, p. 309.

t Williamson. Eighteenth Report, p. 77.

178

doubt that many crabs do not bear their first batch of eggs
until attaining a size of six inches.

Male. — The male crabs evidently attain maturity at
a smaller size than is the case in the females. Probably
all males above four and a half inches are mature, and
Williamson has found mature males below this size.

Fertilisation takes place in the inshore waters during
the late summer and autumn, and is effected immediately
after the female has cast. The one supply of spermatozoa
is probably sufficient for two successive batches of eggs,
and even three in the older crabs. The spermatozoa
remain in the spermathecae, and the entrances to the
latter are closed by plugs, which are probably formed by
a secretion from the walls of the spermathecae which
hardens in contact with water.

The spawning of the eggs is effected in the deeper off-
shore waters during the winter. The eggs are attached
to the endopoditic setae of the pleopods, and remain there
until the following summer, when they are hatched in the
inshore waters.

The crab probably does not cast after the larvae are
hatched, but a second batch of eggs are spawned in the
following winter in the offshore waters. As with the
first batch, the developing embryos will be retained on the
abdominal appendages until the following summer, when
the hatching process will again take place in the inshore
waters. After the second hatching the female probably
casts, and is fertilised.

The developing embryos probably remain attached to
the pleopods for about seven months. The various zoea
stages and the megalopa stage may extend over a period of
two months, but our knowledge of the larval stages of
Cancer is remarkably scanty. It is probable that the
larvae hatched at the end of June will be in the first
adult stage about the end of August.

179

Distribution and Migration.

According to Williamson,* the crabs after the larval
stages may be placed in four different groups according
to their distribution.

Group I includes the young stages up to ^-inch in
breadth. These are probably restricted to the shallow
shore waters.

Group 11 includes the crabs found on the beach
between tide-marks. From |~inch to 2\ inches in breadth.

Group III includes the crabs living in the littoral
waters beyond low-water mark. From 2\ to 4 inches.

Group IV includes all the crabs above 4 inches in
breadth. These crabs are mostly mature, and migrate
from the inshore waters in the summer to the deeper
offshore waters in the winter.

With regard to Group I, my own observations
confirm those of Williamson. In the spring and summer,
when small specimens of Cancer must be very abundant,
they are very rarely found between tide-marks. On the
other hand, they are frequently taken in the dredge close
to the shore.

The migration of the mature crabs has been long
known to fishermen, and our knowledge with regard to
this subject is now fairly complete, thanks to the work of
Williamson, Meek and others. Only the crabs of
Group IV are concerned in the migration, which may be
divided into an offshore migration in the autumn and an
inshore migration in the spring. Both hard and soft
crabs begin to move outward into the deeper water in
September (see Text fig. 13). The extreme depth to
which they travel must necessarily vary with the locality,
but it is generally between twenty and thirty fathoms.
* Williamson. Eighteenth Annual Report,

180

The crabs probably stay in the offshore waters from
December to February, and it is here that the females
spawn. The inshore migration begins in February, and
in May the bulk of the crabs are probably back in the
inshore waters again. The hatching of the larvae takes
place in the warm inshore waters, and casting process is
performed, and, in the case of the females, fertilisation is
effected immediately after ecdysis. In the autumn the
offshore migration again commences, and the cycle is
repeated.

Inshore migration

hpnl March '

Pig. 13. — Diagram illustrating trie annual migratory cycle of mature
crabs. The three processes — casting, spawning and hatching —
probably do not take place in one cycle.

It must be remembered, however, that in the mature
female crab casting and fertilisation in the autumn is not
necessarily followed by spawning in the same winter.
Furthermore, casting will only take place once in two or
three years. But when spawning does take place it is in
the offshore waters, and the hatching is always in the
inshore waters.

As suggested by Williamson, the main reason for

181

this regular migration is the influence of temperature.
In the winter the deeper layers of the offshore waters are
warmer than the inshore waters, and hence the former
are most suitable for the spawning time. In the summer
the shallower inshore waters become much warmer, and
here the young larvae are hatched. It is also reasonable to
suppose that the warm water of the inshore regions is
specially favourable for the somewhat critical period of
casting. The necessity for food may also have an
important bearing on the yearly migration.

I give below a summary of the life-history of an
adult female crab between two successive processes of
ecdysis. The condition of the gonads at the various stages
is discussed. In the example given, it is assumed that
the crab did not spawn in the winter following fertilisa-
tion. The period elapsing between fertilisation and the
first spawning appears to depend altogether on the
condition of the gonads. As a rule, the ovary is very
small and pale when ecdysis takes place, and in such a
circumstance the eggs cannot possibly be ready for
spawning in the course of three or four months.
Consequently the eggs are not shed until the second
winter, i.e. fourteen months after fertilisation. In some
crabs the ovary is fairly ripe when ecdysis takes
place, and in such cases the eggs will probably be
extruded in a few months.

First Year. September — The crab casts and is fertilised.

It then migrates to offshore waters for the winter.
It returns to inshore waters in the spring of the second
year. In the following autumn offshore migration again
takes place.

Condition of ovary. At first it is poorly
developed and pale in colour, and the eggs are without
food-yolk. The ovary gradually ripens, first becoming

182

pink in colour and finally a bright orange. This change

in colour is due to the development of food-yolk. The

eggs first become yolked in the spring of the second

year.

Second Year. December— 7v///s extruded (first spawning).

When the spawning takes place the crabs are in the
deeper offshore waters. In the spring of the third year
the inshore migration commences.

Condition of ovary. Immediately before
spawning the eggs are large and of a bright orange
colour, due to the food-yolk. After the extrusion of the
eggs the ovary is shrunken and of a pale colour. The
majority of the eggs are small and without food-yolk.
There may be, however, a few ripe eggs present which
have failed to escape to the exterior. As the spring
advances the eggs gradually become more mature, and
present a pinkish colour.
Third Year. July — First hatching takes place.

In the autumn of the third year the offshore
migration again takes place.

Condition of ovary. At the time of the
first hatching the ovary is generally of a red colour, and
the eggs are about half ripe.

Third Year. December —Eggs extruded (second spawning) .*
During spawning time the crab is in the offshore

waters. In the spring of the fourth year the inshore

migration commences.

Condition of ovary. Very similar to the

appearance at the time of the first extrusion.

Fourth Year. July — Second hutching takes place.

The crab is now in the inshore waters.

* Williamson gives instances in which the second spawning did
not take place for about fifteen months after the first hatching. It is
difficult to sav whether this is of regular occurrence.

183

Condition of ovary. During the spring
and summer the ovary remains unusually small, and has
a pale colour. The eggs are mostly small and without
yolk.

Fourth Year. September — The crab casts and is fertilised.

During these processes the crab is in the inshore
waters. Immediately after fertilisation the offshore
migration will commence, and the whole of the above
history will be repeated.

Condition of ovary. Similar to that at the
first casting.

Bionomics of Ecdysis.

As already seen, the growth of the crab can only take
place by the exuviation of hard shell or exoskeleton.
Immediately after ecdysis has taken place the body — now
covered by a soft flexible membrane — increases consider-
ably in size. In the mature female fertilisation is also
effected while the crab is soft. The soft crabs are not fit
for food, and it is illegal to expose them for sale when in
this condition.

Season of casting.

The young crabs cast at various times of the year, but
after the third year ecdysis generally takes place in the
autumn. According to Williamson, the main casting
period on the East Coast of Scotland is from July to
September, but it may be extended to December. From
August to Xovember appears to be the chief casting time
in the Isle of Man. The casting is effected in the warmer
inshore waters. The hardening process in the mature
crab lasts from three to five months, according to
Williamson.

184

Rate of Gr r o w t h .

From the evidence obtained by Williamson and
Waddington, and also from observations taken in the
course of the present work, it would appear that the rate
of growth varies considerably. For example, in a series
of seventeen crabs measured in Heligoland I found that
the fraction of increase varied from ^ to -^ (see table
on p. 382). In Waddington's series published by
Williamson* the fraction varied from \ to £j, and in the
table published by Williamsont the ratio of increase
varies from ^ to ~. But, speaking generally, it may be
said that this ratio is between ^ and \.

It is not surprising that this rate of growth should
be a variable one. The main factors to be considered are
probably the general health of the animal (i.e. the
condition of the tissues) the amount of food, the purity of
the water and the temperature of the water, and many
other causes. Any one of these factors would alone be
capable of affecting very considerably the rate of growth.

I give a table (p. 382) showing the rate of increase in
seventeen different crabs. These measurements were
taken from specimens in the Xordsee Museum, Heligo-
land, by kind permission of the Director, Professor
Heincke.

Frequency of casting.

Speaking generally, it may be said that in the
younger stages the moultings are frequent, but that as
the crab grows older the period between each process of
ecdysis and the next becomes longer.

In the earliest stages the frequency of casting differs in

* Williamson. Twenty-second Annual Report, Fishery Board for
Scotland, p. 135.

t Williamson. Eighteenth Annual Report, Fishery Board for
Scotland, p. 110 (see also Twenty-second Report, p. 122).

185

a marked degree in individuals of the same approximate
age. As stated above, the hatching period lasts from

Carapace breadth

before easting.

Cm.

Carapace breadth

after casting.

Cm.

Ratio of
Increase.

1

L-55

2-05

l

81

2

2-15

2-80

l

3'3

3

2-25

2-95

t

3" 2

4

3-10

4-12

1
30

5

3-12

3-82

1

4'1

6

3-40

4-58

1
2'8

7

4-00

5-25

1
32

8

425

5-50

1
3'4

9

445

5-55

1
40

10

4-5G

5-70

1
4'0

11

5-20

6-60

1
37

12

555

6-50

1
., s

13

5-55

7-20

1
33

14

5-70

7-20

1
38

15

6-50

9-10

1
25

16

7-80

9-75

1
40

17

12-80

15-50

1
4-7

June to August. The " early " larvae, therefore, will
have a considerable advantage over the " late " larvae.

186

Consequently, the early crabs will probably cast eight or
nine times before the following summer, while the late
crabs may have cast only five times. There will, also, be
a corresponding difference in size.

It will be realised, therefore, that it is quite
impossible to state the age of a crab with any degree of
certainty. The size of the crab not only depends upon
the frequency of casting, but also upon the ratio of
increase at each act of ecdysis. As we have already seen,
both these factors are subject to a considerable amount of
variation. The most valuable information on this point
is to be obtained from continued observations of crabs
kept in an aquarium. This has been done to a certain
extent by Mr. Waddington, of Bournemouth. But, even
in such cases, we are not justified in establishing any
broad principles on the results obtained. In the first
place, there is no doubt that captivity affects the frequency
of casting. There is also the same difficulty that obtains
amongst crabs living amongst natural conditions, viz.,
that ratio and frequency of casting vary in individual
crabs. This is demonstrated quite clearly by an examina-
tion of AVaddington's :" three series, in which both factors
vary considerably.

But in spite of the impossibility of forming any
definite laws with regard to the frequency of casting, a
careful examination of all the available facts and figures
enables one to give a general outline of the life-history of
Cancer with regard to this particular point.

I propose, therefore, to give such an outline, but, in
doing so, I must emphasise what I have already said —
that we are dealing with factors which are by no means
constant. .

I intend taking a purely hypothetical case, utilising,

* Williamson. Twenty -second Annual Report, p. 135.

187

however, the stages of Waddington's Series A for the
first two years.

First year. The larva was probably hatched in
June. During the first year the crab cast eight times,
and at the end of the first year (June to June) it was
o0'75 mm. broad.

Second Year. The crab cast twice (September and
March), and was 45*75 mm. broad at the end of its second
year. (In the first two years the ratio of increase varied
considerably. It will also continue to vary throughout
life, but for practical purposes I intend taking the ratio
of increase to be uniformly \, which is an average ratio.)

Third year. There will probably be two castings, so
that at the end of the year the crab will be 70"G mm.
broad.

Fourth year. There will be only one ecdysis. At
the end of the year the crab will be 88 mm. broad.

Fifth year. Only one casting. Size at the end of
the year, 110 mm. The crab will now be mature, and if
a female will probably be fertilised while in the " soft "
condition.

Sixth year. The crab will not cast.*

Seventh year. The crab will cast once, and will be
137"5 mm. in width after ecdysis. Fertilisation will
again be effected when the crab is " soft."

Eighth year. The crab will not cast.

Ninth year. The crab will cast, and after ecdysis
will be 17T8 mm. broad. Fertilisation will take place
immediately after ecdysis.

I believe it to be highly probable that after the
female crab attains a size of six inches (150 mm.) ecdysis
will only take place once in three years (see below under

* There is every reason to belie\e that after attaining maturity
the crabs only cast once in two years.

188

" Granny " Crabs). If a male, it will probably cast once
in two years.

First Yeah.

Number of Casting-
Width in mm

.. 1
4-75

2
5-75

3

8-5

(i)

4
10-75

(!)

5
14-5
(1)

6
19-5

(*)

7
24-5

(1)

8
30-75

Width in inches

(1«

Second Yeah.

Third Year.

Number of Casting

Width in mm

9
36-5
(li)

10

45-75

(1!)

11

56-5
(2i)

12

70-6

Width in inches

(2|)

Fourth Year. Fifth Year. Sixth Year.

Number of Casting
Width in mm.
Width in inches ...

14
110-0

(H)

Did not cast

Seventh Year. Eighth Year. Ninth Year-

Number of Casting-
Width in mm

Width in inches

15

137-5
(H)

Did not cast

16

171-8

(«*)

Tenth Year. Eleventh Year. Twelfth Year.

Number of Casting Did not cast
Width in mm.
Width in inches ...

Did not cast

17

214-7

(81)

The approximate calculations given above do not
lend support to Williamson's statement that " a crab of
four and a quarter inches across would be not less than
three years, nor probably more than four years old."
Even if we take a growth ratio of one-third, which is
rather high, we find that the crab would not reach the

189

size of four and a quarter inches until the fourth year at
the earliest.

There is abundant evidence to show that the adult
male and female crabs do not cast every year, but
probably only once in two years. There appears to be
little doubt, moreover, that in the older crabs ecdysis may
take place less frequently than once in two years, and
Williamson has given several instances of crabs in which
the shell is undoubtedly more than two years old. In the
Nordsee Museum, Heligoland, there is a female Cancer,
with a carapace breadth of IT cm., having attached to the
shell, an Anomia 5"8 cm. in width.

Although there appears to be a stage in the- life-
history of Cancer after which ecdysis is only triennial, no
attempt has been made to determine at which period the
change from a biennial to a triennial growth takes
place.

Such an investigation presents innumerable
difficulties, and the only way in which the problem can be
satisfactorily solved is either by having large crabs under
observation in captivity for many years, or by the careful
analysis of a great number of exact measurements made
for the purpose. Neither of these methods has been
followed, and although Williamson has been able to
gather a mass of extremely useful information, his
statistics do not appear to be of much value in the
elucidation of this particular problem.

Our present knowledge with regard to the frequency
of casting may be briefly surnniarised as follows. In the
early stages the young crab casts frequently. As it grows
older the periods between successive castings become
longer, and it is probable that after reaching maturity
both males and females cast only once in two years.
There is scattered evidence to show that many large crabs

190

cast less frequently than once in two years, and a stage is
ultimately reached when the crab ceases to cast.

It is my belief that after attaining a size of about six
inches carapace breadth the females, as a general rule,
only cast once in three years, but that the males continue
to cast biennially for a considerably longer period. It
must be admitted that the figures on which this
suggestion is based are not sufficiently large to warrant
any conclusive statements on the subject. So that,
although I feel that my explanation is a reasonable one,
and is, furthermore, supported by many facts, I bring it
forward merely as a provisional hypothesis. I further
recognise that, just as in the case of the younger stages,
there can be no rigid law regarding the frequency of
casting. It is quite possible that many males over six
inches cast less frequently than once in two years, and
that many females above this size may cast biennially.*

" Granny " Crabs.

The assumption that female crabs above six inches
cast only once in three years was first suggested to me
when examining into the nature of " granny " crabs.
This name is given by the fishermen of Port Erin to
crabs occurring in the late summer and early autumn
which have dirty and discoloured shells and broken claws.
These crabs, if eaten, have a bitter taste and a powerful
purgative effect. The fishermen believe they are diseased
crabs, and always kill such when captured. There
appears to be no adequate reason for believing that these
crabs are diseased. The fact that they occur in consider-
able numbers every year during the late summer and
early autumn is sufficient to render this suggestion
doubtful. The general condition of these crabs leads me

* Williamson has pointed out that if the soft mature female fails
to be fertilised, it will probably cast again in the following year.

191

to believe that they are merely crabs ready for casting.*
Mr. T. N. Cregeen, of the Biological Station, Port
Erin, kindly examined and measured for me a number of
crabs during the summer of 1907. The table inserted
below gives a summary of the results obtained.

One thousand and ninety-four crabs were examined,
and of this number 318 were males and the remainder
females. Of the 77(i females, 112 were "grannies." It
will be observed that there are no male " grannies " in the
list, 'and that practically the whole of the female
" grannies " are six inches or more in width. General
information from Port Erin bears this out. Male
'' grannies " and small female " grannies " are almost
unknown.

Crabs obtained between the Calf of Man and Bradda
Head, Port Erin, during the summer (-July to September)
of 1907.

Greatest breadth

of Carapace

Males.

Fern

ales.
" Granny."

Total.

(in inches).

Normal.

3i-4

—

2

2

4-41

27

16 —

43

4J-5

63

31 —

94

5 -H

67

90

157

5|— 6

54

153 3

210

6 -6£

38

164 18

220

61-7

23

100 36

159

7 -7i

16

58 31

105

7£-8

15

30- 19

64

8-81

13

14 4

31

8|-9

1

6 1

8

9 —91

1

—

—

1

Totals

318

664

112

1,094

-,< in " granny " crabs that I have examined there has been a well-
defined cuticle beneath the hard exoskeleton. This condition is
found in crabs preparing for ecdysis.

192

As already stated, mature crabs generally cast from
August to November, and immediately before and after
ecdysis their flesh is " watery " and has a bitter taste, and
is, therefore, unfit for food. The above symptoms are
also characteristic of the " granny " crab.

" Grannies " are found only amongst females above
six inches in breadth. The explanation of this is that
those females below this size dc not retain their shells
long enough for the latter to acquire the discoloured and
broken appearance. An interval of two years between
successive castings is evidently insufficient to produce this
effect.

This diminution in the frequency of casting in the
females above six inches may be due to the general rule
that as the crab grows older the period between successive
castings tends to become longer until a time is reached
when ecdysis ceases altogether. But this explanation
does not appear to be sufficient to account for the supposed
difference between males and females in this respect. An
important factor in determining the time of casting is the
condition of the reproductive organs. Ecdysis will not
take place in the female so long as there is a supply of
spermatozoa in the spermathecae. The supply of
spermatozoa received by the soft female is generally
sufficient to fertilise at least two batches of eggs in
successive winters, and it is highly probable that the
older crabs will spawn three times between each moult.
Williamson has emphasised this point, and has also stated
that the soft female which, for any reason, does not
become fertilised, will cast in the following summer.
The frequency of ecdysis, then, in mature females is
influenced by the condition of the spermathecae, and it is
probable that on this account the older females will cast
less frequently than the younger ones.

193

These conditions do not affect the male. In the
female the casting- time is of double importance, as it is
at that period that fertilisation is effected. In the male,
however, ecdysis would appear to be important only as a
period of growth. It appears to me, therefore, from an
examination of all the evidence obtainable, that male
crabs above six inches in breadth continue to cast once in
two years, and it is probable that this is the case until
ecdysis stops. This rule cannot be an invariable one, as
there are a few records of male crabs which have not
undergone the casting process for at least three years.

My suggestion that the males above six inches cast
more frequently than the females of a similar size, and,
therefore, do not become " grannies," is borne out, not
only by an examination of the statistics given above, but
also by some figures which I quote from Williamson.*
These figures give a comparison of hard and soft mature
crabs taken off Dunbar. I have only included those
figures dealing with the casting period.

Date

Hard crabs above 4Jins.

Soft crabs above 4] ins.

Males.

Females.

Males. Females.

Aug. 4, 1899

26

76

83

269

117

196

60

77
104
160
206

60
200
110

1

235
140
64
66
56
22

8

Sept. 23, 1893

144

Oct. 27, 1899

87

Nov. 5, 1898

32

Nov. 17, 1899

48

Dec. 13, 1899

42

Dec. 20, 1897

4

Totals

827

917

584 365

Total number of males 1,411

females 1,282

In other words, for every 100 males (hard and soft)
there are !)1 females (hard and soft). For every 100 soft

* Williamson. Eighteenth Report, page 102, Table V.
o

194

males there are only 62 soft females. This points to the
fact that the mature males oast more often than the
females. It is probable that both males and females of
between four and a quarter inches and six inches cast once
in two years, so that it is in the older crabs that the males
cast more frequently than the females. I do not attach
too much importance to the above figures, as the numbers
are too small to justify their use in the question of the
frequency of casting. So far as they go, however, they
confirm my statement that the female crabs over six
inches do not cast so often as the males of a similar size.
It is, therefore, probable that the " granny " crabs are not
diseased, but are merely females ready for casting. They
are only found amongst females of over six inches in
breadth. The reason why smaller females and males of
all sizes do not become " grannies " is because they cast
at least once in two years.

I have been unable to find any reference to
" granny " crabs in the literature* of the subject, but
Williamson says " as a rule the shell of the old female
crab is much more dirty than that of the male."t He
attributes this dirtiness to the fact that the female when
carrying eggs lies half buried in the mud. This explana-
tion may have some truth in it, but it does not solve the
problem as to why only female crabs above six inches in
breadth become discoloured.

It is evident that the various processes which are

characteristic of the life-history of the edible crab are

subject to considerable variation, and it is necessary that

further investigations should be made before the numerous

problems can be regarded as being solved.

* With the exception of a preliminary notice by Professor
Herdman in Twenty-first Annual Report of the L.M. B.C. (Port Erin
Marine Biological Station), p. 25.

f Williamson. Eighteenth Report, p. 110.

195

EXPLANATION OF PLATES.
Reference Letters.

a. ace. — Anterior accessory muscle of the scaphognathite.

a. art. — Antennary artery.

cd). 1 — 6. — Abdominal somites 1 to 6.

abd. — Abdomen.

abs. — Lines of absorption.

a. ex. B. — Tendon of the anterior extensor of basi-ischium of chela.

af. — Afferent branchial sinus.

af. 1. — Do. of the podobranch of the second thoracic somite.

af. 2. — Do. of the arthrobraneh of the second thoracic somite.

af. 3. — Do. of the podobranch of the third thoracic somite.

af. 4. — Do. of the anterior arthrobraneh of the third thoracic somite.

af. 5. — Do. of the posterior arthrobraneh of the third thoracic somite.

af. 6. — Do. of the anterior arthrobraneh of the fourth thoi'acic somite.

af. 7. — Do. of the posterior arthrobraneh of the fourth thoracic somite.

af. 8. — Do. of the pleurobranch of the fifth thoi'acic. somite.

af. 9. — Do. of the sixth thoracic somite.

a. f. ft. — Anterior flexor of the flabellum of the first maxillipede.

a. gl. — Antennary gland.

a. i. p. — Antero-inferior pyloric ossicle.

a. lb. — Antero-lateral lobe of the bladder.

a. mes. — Anterior mesopyloric ossicle.

amp. — Pyloric ampulla.

amp. C. — Supra-ampullary ridge.

a1, n. — Nerve of the first antenna : a', n. — Nerve of the second antenna.

ant. — Second antenna : ant1'' — First antenna (antennule).

a. oe. lb. — Anterior oesophageal lobe of the bladder.

ao. lb. — Outer portion of the antero-lateral lobe of the bladder.

a. ost. — Anterior ostia of the heart.

a. pi. — Anterior pleuropyloric ossicle.

apoph. — Apophysis of the mandible.

art. 1 — etc. — Branches of the sternal artery supplying the first and

following post-oral cephalothoracic appendages.
a. s. — Antero- superior dilator muscles of the fore-gut.

a. s. a. — Anterior supra-ampullary ossicle.

as. lb. — Anterior oesophageal lobe of the bladder.
'/. /. /. — Lateral accessory tooth.

B. — Basipodite.

be. 1 — 5. — Branchio-cardiac veins 1 to ;">.

B-I. — Basi-ischiopodite.

bl.— Bladder.

b. m. — Basement membrane.
br. — Gills.

br. ch. — Branchial chamber.

br. e. — Branchial excretory organ.

br. s. 1 — 5. — Branchial sinuses 1 to 5.

C. — Coxopodite.
C1. — Carpopodite.
race. — Mid-gut caecum.

c. al. — Antero-lateral cardiac muscles.
( . ant. — Anterior cardiac muscles.
card. — Cardiac portion of the fore-gut.

cd. 1. — Dorsal antero-lateral muscles of the heart.
cd. 2. — Ventral antero-lateral muscles of the heart.

196

cd. 3. — Dorsal postero-lateral muscles of the heart.

cd. 4. — Ventral postero-lateral muscles of the heart.

cd. 5. — Posterior muscle of the heart.

cd. (i. — Lateral posterior muscle of the heart.

cd. cd. — Antero-lateral cardiac plate.

cd. pi. — Postero-lateral cardiac plate.

c. g. — Cerebral ganglia.

ch.— Chela.

ch. ep. — Epidermis (chitogenous epithelium).

c. i. — Postero-inferior cardiac muscle.

c. lat. — Lateral cardiac muscles.

c. m. — Circular muscles.

c. lob. — Cerebral lobe of the bladder.

en. — Cornea.

c. oe. — Constrictor muscles of the oesophagus.

Com. — Commissure.

corn. — Corneagen (cells of ommatidium).

c. p. v. — Cardio-pyloric valve.

c. py. — Cardio-pyloric muscles,
r. /. — Connective tissue.

ct. gl. — Cutaneous gland.
cut. — Cuticle.

D. — Dactylopodite.

d. ai. — Antero-inferior dilator muscle of the cardiac fore-gut.
d. art. — Descending artery.

Der. — Dermis.

d. ex. G. — Tendon of the dorsal extensor muscle of the coxopodite.

d. f. ex. — Dorsal flexor muscle of the exopodite.

d. g. — Duct of the gland.

di. gl. — Digestive gland.

d. I. — Deeper layer of the cornea.

d. la. — Antero-lateral dilator muscle of the cardiac fore-gut.

d. Ip. — Postero-lateral dilator muscle of the cardiac fore-gut.

d. sup. — Dorsal pyloric muscle.

d. v. m. — Dorso-veutral muscles.

E. — Epistoma.

e. a. md. — External adductor muscle of the mandible.
e. b. md. — External abductor muscle of the mandible,
e. bl. — Epithelium of the bladder.

e. d. — Ejaculatory duct.

e. es. — Epithelium of the end-sac.

ef. — Efferent vessel.

ef. 1 — 9. — Efferent vessels of gills 1 to 9.

e. lb. — Epigastric lobe of the bladder.

end. — Endopodite.

end. s. — End sac.

ep. 4 — 12. — Endopleurites of somites 4 to 12.

epm. 1 — 19. — Epimera of somites 1 — 19.

e. st. 4 — 12. — Endosternites of somites 4 to 12.

e. st. 13. — Last thoracic arthrophragm (" sella turcica ? ).

e. fu. — Epithelium of the renal tubule of the labyrinth.

ex. — Exopodite.

ex. 1 — 6. — Extensor muscles of the abdominal somites 1 to 6.

ex. B. — Tendon of the extensor muscle of the basi-ischiun.

ex. C. — Tendon of the extensor muscle of the coxopodite.

ex. C1. — Tendon of the extensor muscle of the carpopodite.

ex. /I. (Pig. 29). — Extensor muscle of the flabellum.

197

ex. fl. (Fig. 30). — Extensor muscle of the flagellum.

ex. prot. — Extensor muscle of the protopodite.

ex. py. — Exopyloric ossicle.

ext. D. — Tendon of the extensor muscle of the dactylopodite.

ext. ex. — Extensor muscle of the exopodite.

ext. P. — Tendon of the extensor muscle of the propodite.

ex. tel. — Extensor muscle of the telson.

/. 1 — 6. — Flexor muscles of the abdominal somites 1 to 6.

/. B. — Tendon of the flexor muscle of the basi-ischium.

/. c. — Fat cell of the digestive gland.

/. C. — Tendon of the flexor muscle of the coxopodite.

/. 6". — Tendon of the flexor muscle of the carpopodite.

/. D. — Tendon of the flexor muscle of the dactylopodite.

flab. — Flabellum (epipodite).

flag. — Flagellum.

fl. m. — Muscles from the top of the " gland " to the carapace.

/. m. 1. — Flabellum of the first maxillipede.

/. in. 2. — Flabellum of the second maxillipede.

/. m. 3. — Flabellum of the third maxillipede.

/. m. c. — Ferment cell of the digestive gland.

/. o. — External female genital opening.

/. p. — Fracture plane.

/. P. — Tendon of the flexor muscle of the propodite.

/. prot. — Flexor muscle of the protopodite.

/. tel. — Flexor muscle of the telson.

/. v. — Ferment vesicle.

g. 1. — First gill. Podobranch of the second thoracic somite.

g. 2. — Second gill. Arthrobranch of the second thoracic somite.

g. 3. — Third gill. Podobranch of the third thoracic somite.

g. 4. — Fourth gill. Anterior arthrobranch of the third thoracic somite.

g. 5. — Fifth gill. Posterior arthrobranch of the third thoracic somite.

g. 6. — Sixth gill. Anterior arthrobranch of the fourth thoracic somite.

g. 7. — Seventh gill. Posterior arthrobranch of fourth thoracic somite.

g. 8. — Eighth gill. Plcurobraneh of the lifth thoracic somite.

g. 9. — Ninth gill. Pleurobranch of the sixth thoracic somite.

g. a. — Anterior gastric muscle.

g. /. — Fat globules.

g. lb. — Paragastric lobe of the bladder.

g. p. c. — External posterior gastric muscle.

g. p. i. — Internal posterior gastric muscle.

h. — Seta of the endopodite of the pleopod of the female.

h. art. — Hepatic artery.

h. g — Hind gut.

h. I. — Hepatic lobe of the bladder.

/. — Isehiopodite.

ia. art. — Inferior abdominal artery.

i. a. /. — Inter ampullary fold.

i. a. md. — Internal adductor muscle of the mandible.

i. b. md. — Internal abductor muscle of the mandible.

i. b. s. — Inter-branchial septum.

i. caec. — Hind-gut caecum.

i. c. m. — Inner flexor muscle of the first maxilla.

i. ex. — Inner flexor muscle of the scaphognathitc (t. ex. s., Figs. 27, 31).

i. f. — Inner extensor muscle of scaphognathite.

i. j. m. — Inner extensor muscle of first maxilla.

198

i. gl. — Glands of the hind-gut.

i. I. — Infero-lateral cardiac tooth.

/'. m. ex. — Inner median flexor muscle of the scaphognathite.

i. m. /. — Inner median extensor muscle of the scaphognathite.

i. pi/, e. — External inferior dilator muscle of the pyloric fore-gut.

i. pi/, i. — Internal inferior dilator muscle of tire pyloric fore-gut.

i. s. — Infra-branchial sinus.

j. d. — Junction between the descending artery and the sternal artery.

/.—Gill lamella.

lab. — Labrum.

lot. t. — Lateral teeth.

I. ex. prot. — Lateral extensoi muscle of the protopodite.

I. f. prot. — Lateral flexor muscle of the protopodite.

/. g. n. — Lateral gastric nerve.

lig. — Ligament.

I. m. — Longitudinal muscle.

I. pi/. — Lateral pyloric ossicle.

I. s. — Lamella blood sinus.

.1/. — Meropodite.

m.a1. — Muscle chamber of first antenna.

m.a2. — Muscle chamber of second antenna.

in (uid. — Mandible.

in. c. — Mesocardiac ossicle.

md. palp. — Mandibular palp.

med. p. — Median plate of endophragmal system.

met. — Me tastoma .

m. ex. C". — Extensor muscle of the carpopodite.]

m. ex. M. — Extensor muscle of the meropodite.

m. ext. D. — Extensor muscle of the dactylopodite.

m. ext. P. — Extensor muscle of the propodite.

to. /. C". — Flexor muscle of the carpopodite.

in. f. D. — Flexor muscle of the dactylopodite.

in. f. M. — Flexor muscle of the meropodite.

//;. /. P. — Flexor muscle of the propodite.

to. g. — Mid-gut.

m. o. — Male genital opening.

to. py. — Middle pyloric ossicle.

m. s. a. — Middle supra-ampullary ossicle.

M. V. — Main vesicle.

mx1. — First maxillipede.

mx2. — Second maxillipede.

mx'-K — Third maxillipede.

n. — Nucleus.

«. 1. — n. !>. — Nerves arising from the ventral nerve mass of the thorax

v. ah. — Abdominal nerve.

n. c. — Nerve commissure.

n. f. — Foramen of ventral nerve mass for descending artery

n. i. — Nerve to hind-gut.

n. I. — Nerve to digestive gland.

n. in. — Nerve to the mandibular muscles.

n. po. — Transverse post-oesophageal connective.

n. t. — Nerve to the integument.

o. — Eye.

o. art. — Ophthalmic artery.

oe. — Oesophagus.

199

oe. at. — An tero -inferior dilator muscle of the oesophagus.
oe. as. — AnteroT-superior dilator muscle of the oesophagus.

oe. I. — Lateral dilator muscle of the oesophagus.

o. c. in. — Outer flexor muscle of the first maxilla.

or. p. — Posterior dilator muscle of the oesophagus.

o. er. — Outer flexor muscle of the scaphog lathite (o.ex.s., Fig<. 27, 31).

o. f. — Outer extensor muscle of the scaphognathite.

o. /. m. — Outer extensor muscle of the first maxilla.

o. g. — Optic ganglion.

o. I. s. — Outer lamellar sinus.

o. m. c. — Ophthalmic muscle chamber.

o. in. ex. — Outer median flexor muscle of the scaphog -athite.

o. in. /. — Outer median extensor muscle of the scaphognathite.

oin. ii, — Oculo-motor nerve.

o. ii. — Optic nerve.

o. /*. fib. — Optic nerve tibres.

op. — Operculum of excretory opening.

op\ — Inner side of operculum.

up. Ii. ■<■■. — Ophthalmic blood sinus.

o. ped. — Optic peduncle.

o. py. — Pyloric ossicle.

orb. — Orbit.

ov. — Ovary.

"/•. art. — < Ovarian artery.

ovd. — Oviduct.

P. — Propodite.

P. 1—4.— Walking legs 1 to 4.

p. 1 — 2. — 1st and 2nd abdominal appendages of male

p. ace — Posterior accessory muscle of the scaphognathite.

p. c. p. — Pro-cephalic processes.

pec. — Pectineal ossirlc.

ped. — Eye peduncle.

Per. — Pericardium.

per. gl. — " Pericardial pouch."

p. ex. B. — Tendon of the posterior extensor muscle of the basi-ischium.

p. j. B. — Tendon of the posterior flexor muscle of the basi-ischium.

p. (j. — Paro; sophageal ganglion.

pg. c, — Pigment cell.

pg. i. — Inner pigmented layer of the eye.

p. g. n. — Posterior gastric nerve.

pg. o. — Outer pigmented layer of the eye.

pig. I. — Pigment layer.

pi. art. — Postero-lateral artery.

p. lb. — Progastric lobe of the bladder.

p. mes. — Posterior mesopyloric ossicle.

p. n. — Postero-lateral nerve.

pod. br. — Podobranch.

p. o. lb. — Post-oesophageal lobe of the bladder.

p. ost, — Posterior ostia of the heart.

pp. — Pleuropyloric wall.

p. pec. — Prepectineal ossicle.

prot. — Protopodite.

pr. p. — Pro pyloric ossicle.

p. s. a, — Posterior supra-ampullary ossicle.

pt. c, — Pterocardiac ossicle.

pt. pec. — Post pectineal ossicle.

pyl. — Pyloric region of the fore-gut.

py. lat. — Lateral pyloric muscles.

200

r. br. — Roof of the pro-branchial chamber.
ret. — Cells of the retinula.
rhab. — Rhabdome.
rost. — Rostrum.

S1. —S19. —Sterna of somites 1 to 19.

s. a ' . — Socket of first antenna.

s. a-. — Socket of second antenna.

sa. art. — Superior abdominal artery.

s. amp. — Supra-ampullary wall of the pyloric region of the fore-gut.

sal. g. — Salivary (oesophageal) glands.

s. art. — Sternal artery.

sb. — Striated border of the cell.

sea ph. — Scaphognathite.

s. ch. — Blood sinus from the chela.

s. dt. — Subdentary ossicle.

s. /(., s. hr. — Sub -hepatic region of carapace.

s. I. — Supraciliary lobe.

s. lb. — Supra-hepatic lobe of the bladder.

s. mx. 2. — Blood sinus from the second maxillipede.

s. mx. 3. — Blood sinus from the third maxillipede.

s. Pi. — s. P. 4. — Blood sinus from the walking legs 1 to 4.

sp. a. — Frontal spine.

sp. d. — Dorsal spine.

sp. I. — Lateral spine.

apt. — Spcrmatheca.

st. — Stalk of the egg attached to the seta of the endopodite.

st. g. — Stomatogastric ganglion.

st. i. — Inferior root of the stomatogastric nerve.

st. n. — Stomatogastric nerve.

st. s. — Superior root of the stomatogastric nerve.

t. c. ex. — Tendon of the extensor muscle of the exopodite.

tel — Telson.

test. — Testis.

/. ex. db. — Tendon of the external abductor muscle of the mandible.

/. ex. ad. — Tendon of the external adductor muscle of the mandible.

/. ear. 21. — Tendon of the extensor muscle of the mcropodite.

t. f. ex. — Tendon of the flexor muscle of the exopodite.

t. g.. t. gm. — Ventral thoracic nerve mas-;.

f. int. ad. — Tendon of the internal adductor muscle of the mandible.

t. n. — Tegumcntary nerve.

tit. — Renal tubule of the labyrinth.

u. c. — Urocardiac ossicle.
up. f. — Uropyloric fold.
it. py. — Uropyloric ossicle.

v. — Globules in the epithelial cells of the mid-gut.

vol. — Pyloric valves.

v. d. — Vas deferens.

v. ex. C. — Tendon of the ventral extensor muscle of the coxopodite.

/'. /. ex. — Ventral flexor muscle of the exopodite.

rit. — Vitrella (cells of the ommatidium).

v. m. — Egg membranes.

v. oe. g. — Ventral oesophageal cutaneous glands.

//. r. — Young cells of the tubules of the digestive gland.
y. gr. — Yolk granules.

z. c. — Zygocardiac ossicle.

201

Plate I.

Fig. 1. Cancer paguvus, from above. Small specimen.

Fig. 2. ('. pagurus, female, from below. Only the
slumps of the pereiopods are shown. This
figure shows the shape of the abdomen, and
also the " pleural groove." x i.

Fig. 3. C. paguvus, male, from below. Only the
stumps of the pereiopods are shown. x i.

Plate II.

Pig. 4. First antenna (antennule) of right side, seen
from below. x 2.

Fig. >5. Right second antenna, from below. x 2.

Fig. (J. Eight mandible, from below. The apophysis
is also shown with the tendons of the external
adductor, the internal adductor and the
external abductor muscles. x 1.

Fig-. 7. height hist maxilla, from below. x 1.

Fig. 8. Right second maxilla and seaphognathite, from
below. x 1.

Fig. 9. Right first maxillipede, from below. x 1.

Fig. 10. Right second maxillipede, with podobranch,
from below. x 1.

Fig. 11. Right third maxillipede, with podobranch.
from belowT. x 1.

Fig. 12. Anterior view of the right chela (first pereio-
pod). x 1.

Fig*. 13. Anterior view of the rig-lit third walking1 leg1
(fourth pereiopod). x 1.

Fig. 14. First abdominal appendage of male of right
side. x 1.

Fig. 15. Second abdominal appendage of male of right
side. x 1.

Fig. 16. First and second abdominal appendage of male,
viewed from left side. x 1.

Fig. 17. Anterior view of a right abdominal appendage
of female. x 1.

202

Plate 111.

Fig. 18. The endophragmal system viewed from above.
On the right side the thoracic epimera have
been removed in order to display the endo-
pleurites. The following parts have also been
removed:— The carapace, with the exception
of the anterior portion of the sub-hepatic
region ; the membranous roof of the branchial
chamber: the abdomen; the cephalothoracic
appendages (with the exception of the
mandibles) ; the gills ; and all the soft parts of
the body. x 1.

Fig. 19. The sternum of the pre-oral cephalic region,
viewed from above. The dorsal part of the cara-
pace has been removed, as well as the pre-oral
cephalic appendages and the soft parts. x 2.

Fig. 20. The sternum of the pre-oral cephalic region,
from below. The first and second antennae of
the left side have been removed to display their
sockets and muscle chambers. The long
peduncle of the left eye is shown, and also the
labrum. x 2.

Fig. 21. Anterior and ventral view of the right chela.
The anterior wall of each segment has been
cut out, and the soft parts removed in order to
display the tendons. x |.

Fig. 22. Posterior view of the basal portion of the second
walking leg (right side). The posterior wall
of the coxopodite has been removed in order to
display the tendons of that segment and also
of the basi-ischium. x 1.

Fig. 2->. Posterior view of the base of the last walking
leg of the right side. The posterior wall of the
coxopodite has been removed in order to
display the tendons of the coxopodite and of
the basi-ischium. x 1.

Plate IV.

Fig. 24. Vertical section through the integument of a
soft crab, x 550.

203

Fig. 25. Right abdominal appendage of female, viewed
from behind. To show the muscles of the
protopodite aud exopodite, the posterior walls
of the protopodite, exopodite and endopodite
have been removed. x 3.

Fig. 26. Anterior view of the right first maxilla, in order
to show the muscles. x 2.

Fig. 27. Anterior view of the right second maxilla, to
show the extensor muscles of the seapho-
gnathite. x 1-A-.

Fig. 28. Anterior view of the right second maxilla, in
order to show the muscles of the scapho-
gnathite. The extensor muscles have been
cut short, and the basal and inner portions of
the scaphognathite have been opened in order
to show the flexors and accessory muscles,
x 1^. (To see this appendage in its natural
position, the hgure must be rotated to the left
through an angle of 45°.)

Fig. "rl(J. Posterior view of part of the first maxillipede
of the right side, in order to show the muscles
of the maxillipede and flabellum. The posterior
flexor and the extensor muscles of the
flabellum are cut through. x 2.

Fig. 30. Posterior view of the third maxillipede of the
left side. The ventral walls of the various
parts have been removed in order to display
the muscles. x 1^. (The base of the
appendage is to the left of the figure.)

Fig. 31. A dissection of the anterior part of the cephalo-
thorax to display the muscles of that region.
The dorsal portion of the carapace has been
removed, and also the soft parts, with the
exception of the muscles. x 2.

Fig. 32. Dissection of the female abdomen from the
ventral side to display the extensor muscles.
The ventral wall and appendages have been
removed, as well as all the soft parts. x i.

Fig. 33. Semi - diagrammatic longitudinal section
through the abdomen of the female, to show
the extensor and flexor muscles. x i.

204

Plate V.

Fig. 34. A dissection from above to display the various
parts of the alimentary canal. The dorsal
part of the body has been removed, as well as
the gonads. The gills are not shown. The
digestive gland is only shown on the left side.

Fig. 35. Vertical transverse section through the
oesophagus, to show especially the oesophageal
glands. x 35.

Fig. 36. Transverse section through the pyloric region
of the fore-gut. x 30.

Fig. 37. Transverse section through a tubule of the
digestive gland. x 120.

Fig. 38. A "fat cell" of the digestive gland. x 430.

Fig. 39. A " ferment cell " of digestive gland, x 430.

Plate VI.

Fig. 40. The fore-gut from the left side, showing the
ossicles. x ]A.

Fig. 41. The pyloric region of fore-gut, dorsal view, x 3.

Fig. 4L/. Anterior view of the pro-pyloric ossicle, x 4.

Fig. 42. Transverse section through the region of the
green gland, to show the parts of the bladder.
Only one side shown. x 8.

Fig. 43. The ossicles of gastric mill, from above, x 2.

Fig. 44. The ossicles of the gastric mill, from below.
The fore-gut has been opened ventrally. The
left zygocardiac ossicle has been rotated in
order to show the lateral teeth. x 3.

Plate VII.

Fig. 45. The fore-gut from the left side, showing the
intrinsic and extrinsic muscles. x 3.

Fig. 40. Anterior view of the fore-gut, to show the
muscles. x 11.

Fig. 47. The fore-gut from above, showing the intrinsic
and extrinsic muscles. x li.

205

Fig. 48. The fore-gut from behind, showing the intrinsic
and extrinsic muscles. The postero-lateral
dilator has been cut near its insertion at the
right side in order to show the antero-lateral
dilator muscle. The right posterior dilator of
the oesophagus has also been cut in order to
expose the lateral dilators of oesophagus, x.'i.

Fig. 49. Dorsal view of the heart. x 2.

Fig. 50. View of the heart from the left side. x 2.

Plate YIII.

Fig. 51. General view of the blood system from above.
The dorsal region of the carapace has been
removed. On the right side the organs remain
intact, but on the left side the ovary and
digestive gland have been removed. One gill
on the left side has also been turned outward
in order to show the afferent branchial vein
and the branchio-cardiac vein. In this region,
also, the flabella of the second and third
maxillipedes are seen lying beneath the gills.
The course of the bladder is shown on the left
side. x 1.

Plate IX.

Fig. 52. View of ventral region of post-oral cephalo-
thorax to show the sternal artery and its
branches. All the muscles have been removed.
On the right side the arteries going to the legs
are cut short. The inferior abdominal artery
is also cut short. x i.

Fig. 5-3. Dissection of abdomen from the dorsal side to
show the superior abdominal artery. The
tergal region of the abdomen has been
removed. x 1.

Fig. 54. Dissection of the posterior region of the thorax
to show the pericardium, heart, branchio-
cardiac veins and efferent branchial veins.
Only seven of the gills are shown on the left
side, and on the right side only the roots of the

206

nine gills are shown. On the right side the
epimeral wall has been removed in order to
show the course of the branchio-cardiac veins,
and also to show the connection between the
latter and the pericardium. x 1.

Fig. 55. A dissection of the epimeral region of the
thorax from the right side, in order to show
the positions of the branchial sinnses, the
infra-branchial sinus, the afferent branchial
sinuses, and also the sinuses coming from the
thoracic legs. x 1.

Fig. 56. Diagrammatic section through the thorax in
the region of the heart, to show the blood
system and the general arrangement of the
organs. The tendons of the first walking legs
are shown. x 1.

Plate X.

Fig. 57. Dissection of the left side to show the exten-
sions of the bladder. The dorsal region of the
carapace has been removed. The fore-gut has
been cut through at the oesophagus and
removed. The gonads and the digestive gland
have also been taken away. The antennary
gland is represented by a dotted circle. x 1.

Fig. 58: Semi-diagrammatic sagittal (vertical longi-
tudinal) section of the antennary gland. The
anterior part of the gland is to the right,
x 60.

Fig. 59. Detailed drawing of part of previous figure,
showing the epithelial cells of the bladder, end
sac and the renal tubule, x 90.

Fig. TiO. Section through the mass of cutaneous glands
opening on to the epistoma. x 210.

Fig. 61. Section showing the epithelial cells of the mid-
gut, with characteristic striated border, x 550.

Fig. 62. Longitudinal section through part of the wall
of the hind-gut, about the middle of the
abdomen, showing cutaneous glands, x 165.

207

Fig. 60. Dorsal view of the gills of the left side in their
natural position, lying upon the thoracic
epimera. The roofs of the branchial and pre-
branchial chambers have been removed. The
scaphognathite has been turned over to the
inner side, in order to show the maxillipedes.
The flabellum of the first maxillipede is shown
lying upon the gills (the longitudinal axis of
the body is at an angle of GO0 with the long
axis of this plate). x 1.

Fig. 64. View of the branchial chamber of the left side.
Each gill has been removed at its base, so that
only the points of attachment of the gills are
shown. The scaphognathite and first maxilli-
pede (with flabellum) have been removed. By
removing the gills the flabella of the second
and third maxillipedes are exposed. x 1.

Plate XL

Fig. 65. Diagrammatic transverse section across a gill,
showing the branchial septum, the afferent and
efferent vessels and the lamellae. x 3.

Fig. 06. Longitudinal section through a gill in the
region of the afferent vessel. x 100.

Fig. 07. Transveise section through a grill, to show
especially the branchial excretory cells in the
septum, x 24.

Fig. (iS. A dissection of the nervous system from above.
The alimentary canal has been cut through the
region of the oesophagus and removed, but the
stomatogastric system is shown. The nerves
arising from the left side of the cerebral
ganglia are alone shown. Those of the right
side have been cut, only the nerves arising
from the right side of the thoracic ventral mass
are shown. The abdominal nerve is only
shown as far as the anterior region of the
abdomen. x 1.

208

Fig. 09. Longitudinal section through the eye. The
optic ganglion is also shown. x 12.

Fig. 70. Enlarged drawing of two of the ommatidia
from the previous figure. x 450.

Fig. 71. Transverse section across one of the ommatidia
in the previous figure, through the rhabdomo
and retinulae. x 1000.

Fig. 12. Transverse section across one of the ommatidia
in the previous figure, through the vitrella and
the pigment cells. x 1000.

Fig. 73. Surface view of the cornea, showing one of the
corneal facets. x 750.

Fig. 74. Olfactory seta taken from the exopodite of the
first antenna. x 1100.

Fig. 75. Auditory seta (" hooked seta ") from the audi-
tory sac of the first antenna, x 400.

Fig. 70. Auditory seta (" group seta ") from the auditory
sac of the first antenna. x 300.

Plate XII.

Fig. 77. reproductive system of a fairly mature female,

showing the double ovary, the spermathecae

and the oviducts. x 1.
Fig. 78. Reproductive system of an immature male,

showing the paired testes, the vas deferens and

the ejaculatory duct. The fore-gut is also

shown in position. x 2.
Fig. 79. Section through an almost ripe ovary, showing

the eggs filled with yolk granules. The

details are only inserted in one egg. x 190.
Fig. 80. A very earl}* stage in the development of the

embryo, attached to a seta of an endopodite

of a female pleopod. This drawing shows the

method of attachment. x 50.
Fig. 81. Ripe spermatozoon, showing the two processes.

Taken from spermatheca of female, x 1400.
Fig. 82. Some of the epithelial cells of the bladder of the

excretory system. The striated nature of the

protoplasm is shown. x 050.

209

Plate XIII.

Fig. 83. Protozoea of Cancer pagurus, two hours after
hatching. View from the left side. x 120.

Fig. 84. Telsou of the same stage, from above. x 175.

Fig. 85. Zoea of Cancer pagurus, thirty hours after
hatching, showing the maximum development
of the frontal and dorsal spines. x 90.

Fig. 86. Dorsal view of the cephalothorax of the same
stage. The dorsal and frontal spines are cut
off short, x 110.

Fig. 87. Telson of the same stage, from above. x 160.

C. Tinling & Co., Ltd., Printers, 53, Victoria Street, Liverpool.

L. M. B. C. Memoir XVI.

Plate I.

Fig.

Fig
2.

Fig
3.

CANCER

L.M. B. C. MemoirXVI.

:k J I

L.M. B. C. Memoir XVI. Plate III

rost o.m.o . m.<z. 2

op.' ?:' ?-<*■ /.c.

meat t ~ ■* acne -

f '■**■* H- : ?°St- -s.a'.

r^s^5^? — ,6ei/.

■m-.a. '
.4,V ^ 4-«.-m.«*

of.

Fig. 19. M- ,^ I-',, 20

■J rp extP

1 C ' ' •*

*-**■■£• I \ d2 ^■-■'~^^i^r-^—-~-~^—™T^ ■

p.f.B-. V, -. \\ /: ■■- "' ■"-■•. ^ '*»i^te=^._

,--^%. ^' — ^^ "

f.C. -\-- \\|| I

• : v: p

M fC- Fig. 21.

C. a?. f.p\ v.' 'v2-~,

■' 0-X t.ex.?!

CANCER.

L.M. B. C. Memoir XVI.

Plate IV

%:$.

?v

t—

" "

lr

■

■ T jf f

i ' i i f i ft titti *

er

ii

Fi£28

p.QLCC

CANCER.

\ J Fig. 29. ''ful-

"'..,.. . '.:■ ikin< i-.th Sin

L.M B C. Memoir XV]

I ' 1 . \TE V.

Kiu 34.

ca.rd.

Fig. 39. Fjg.38.lv

ynand. ^

CANCER.

M'Farlioc 1 Erslme lilh Ed'.n*

L.M.B. C. Memoir XVI.

Plate VI

Fig. 43

p.pec.

*m.fi.

cd. del.

Fijg. 40

cd.pl.

Dorsal ■

Fig. 42

CANCER.

0.^. TL.C.

- LitTi Edii

B. C. Memoir XVI.

Plate VII.

I. Fig. 50.

CcL.'4- d.'ctrt. '-cd.6

CANCER

L.M. B C. Memoir XVI.

Plate VIII

IfKarlane lErjlme L.tdtdm'

CANCER

L M B. C. Memoir W

ep-m,. 9 e.t>. 9

Plate K

\ ,' .' J. MI J \

oif.6 ■ OL.f.4 '. of. 2

if 8 i / ay? 7 ay?.* a/: J

t->J- HTirUtie i Erstuit . Lltk Eta!

CANCER

L.M. B. C. MemoirXVI.

Plat i- ; X

Fig. 63.

epn%IZ-

J . P. del .

CANCER.

1

M'FarlajieiErsku.:

L.M. B. C.

Plate XI.

N'Farlanc lErsHine.UUi. EdmT

CANCER,

L.M. B. C. Memoir XVI.

Plate XII.

Fid. 77.

Spt.

card.

CANCER.

L MB. C. Memoir K.VJ

Platj

M'Fa-lacf L ErsVm. Litl ESoT

CANCER.

Date Due