a

fa
= Sq)

ss

My-§

HARVARD UNIVERSITY.

LIBRARY

OF THE

MUSEUM OF COMPARATIVE ZOOLOGY.
M1524
GIFT OF i

ALEX. AGASSIZ.

aes au 396 Noanehe tt Isq¢

QUARTERLY JOURNAL

OF

MICROSCOPICAL SCIENCE.

EDITED BY

E. RAY LANKESTER, M.A., LL.D., F.R.S.,

DIRECTOR OF THE NATURAL HISTORY DEPARTMENTS OF THE BRITISH MUSEUM; HONORARY FELLOW OF EXETER COLLEGE
OXFORD ; CORRESPONDING MEMBER OF THE IMPERIAL ACADEMY OF SCIENCES OF ST. PETERSBURG, AND OF THE
ACADEMY OF SCIENCES OF PHILADELPHIA; FOREIGN MEMBER OF THE ROYAL BOHEMIAN SOCIFRTY OF
SCIENCES, AND UF THE ACADEMY OF THE LINCEI OF ROME; ASSOCIATE OF THE ROYAL ACADEMY
Or BELGIUM; HONORARY MEMBER OF THE NEW YORK ACADEMY OF SCIENCES, AND OF
THE CAMBRIDGE PHILOSOPHICAL SOCIETY AND OF THE ROYAL PHYSICAL SOCIETY
OF EDINBURGH} ASSOCIATE MEMBER OF THE BIOLOGICAL SOCIETY
OF PARIS} FULLERIAN PROFESSOR OF PHYSIOLOGY IN THE
ROYAL INSTITUTION OF LONDON.

WITH THE CO-OPERATION OF

ADAM SEDGWICK, M.A., F.R.S.,

FELLOW AND TUTOR OF TRINITY COLLEGE, CAMBRIDGE 3

AND

Wer Re WEEDON, MA, PRS;

JODRELL PROFESSOR OF ZOOLOGY AND COMPARATIVE ANATOMY IN UNIVERSITY COLLEGE, LONDON; LATE
’
FELLOW OF ST. JOHN S COLLEGE, CAMBRIDGE.

VOLUME 41.—New Sertes.
With Aithographic Plates and Engrabings on Wood.

LONDON:
J. & A. CHURCHILL, 7, GREAT MARLBOROUGH STREET.
1899,

CONTENTS.

CONTENTS OF No. 161, N.S., MARCH, 1898.
MEMOIRS :

The Habits and Structure of Arenicola marina. By F. W.
GamBLE, M.Sc., and J. H. Asuwortu, B.Sc., Demonstrators and
Assistant Lecturers in Zoology, Owens College, Manchester.
(With Plates 1—5)

The Aseptic Cultivation of Mycetozoa. By Casper O. Mituzr,
M.D. (With Plates 6 and 7)

On the Development of Tubulipora, and on some British and
Northern Species of thisGenus. By Stpnsny F. Harmer, Se.D.,
Fellow of King’s College, Cambridge ; Superintendent of the
University Museum of Zoology. (With Plates 8—10) .

The Molluses of the Great African Lakes.—I. Distribution. By
J. E.8. Moore . : ;

The Molluses of the Great African Lakes.—II. The Anatomy of the
Typhobias, with a description of the New Genus Bathanalia. By
J. E.S. Moorz. (With Plates 11—14)

CONTENTS OF No. 162, N.S., JUNE, 1898.
MEMOIRS:

The Segmentation of the Ovum of the Sheep, with Observations on
the Hypothesis of a Hypoblastie Origin for the Trophoblast. By
Ricuarp AssHEeTon, M.A. (With Plates 15—18)

On the Heart-body and Ceelomic Fluid of Certain Polychaeta. By
Lioness James Picton, B.A. (With Plates 19—22)

PAGE

43

73

159

181

205

263

iv CONTEN''S.

On the Hypothesis that, Lake Tanganyika represents an Old Jurassic
Sea. By J. HE. S. Moors. (With Plate 23)

On the Reno-pericardial Canals in Patella. By Epwin 8. Goop-
ricH, B.A., Aldrichian Demonstrator of Comparative Anatomy,
Oxford. (With Plate 24)

CONTENTS OF No. 163, N.S., NOVEMBER, 1898.

MEMOIRS:

The Development of the Pig during the First Ten Days. By
Ricuarp AssHeton, M.A. (With Plates 25—28) ‘

The Structure of the Mammalian Gastric Glands. By R. R.
Buns ey, B.A., M.B., Assistant Demonstrator in ee Uni-
versity of Mornin, (With Plate 29) ‘

On Certain Green (Chlorophylloid) Pigments in Invertebrates.
By Marion I. Newsiein, D.Se., Lecturer on Zoology in the
Kdinburgh College of Medicine for Women. (Krom the Laboratory
of the Royal College of ee meu (With Plates
30 and 31) x ; :

Note on a (? Stomatopod) ee Larva. _ J. J. Lister,
M.A., Demonstrator of Comparative Anatomy in the University
of Cambridge

On the Nephridia of the Polychata.—Part II. Glycera and Goniada.
By Epwiy 8. Goopricu, B.A., Aldrichian Demonstrator of
Comparative Anatomy, Oxford. (With Plates 32—35) .

CONTENTS OF No. 164, N.S., JANUARY, 1899.
MEMOIRS :

On Differences in the Histological Structure of Teeth occurring
within a Single Family—the Gadide. By Cuartes 8. Tomes,
M.A., F.R.S. (With Plate 36) .

A Description of ''wo New Species of Spongilla from Lake
Tanganyika. By Ricuarp Evans, B.A. (With Plates 37
and 388) . 3 ; i . :

PAGE

303

323

329

361

391

433

439

459

471

CONTENTS.

On Tetracotyle petromyzontis, a Parasite of the Brain of
Ammoccetes. By Atzert W. Browy, B.A., F.L.S., formerly
Exhibitioner of Christ Church, Oxford. (With Plate 39)

Studies on the Structure and Formation of the Caleareous Skeleton
of the Anthozoa. By Gitzert C. Bourne, M.A., F.L.S.,
Fellow and Tutor of New College, Oxford; University Lecturer
in Comparative Anatomy. (With Plates 40—43)

The Structure and Development of the Hairs of Monotremes and
Marsupials.—Part I. Monotremes. By Batpwin Spencer,
M.A., Professor of Biology in the University of Melbourne, and
Grorcina Sweet, M.Sc., University of Melbourne. (With
Plates 44—46)

Trophoblast and Serosa; a contribution to the Morphology of
the Embryonic Membranes of Insects. By ArtHur WILLEY,
D.Sc.Lond., Hon. M.A. Cantab., Balfour Student of the Uni-
versity of Cambridge

TirLe, INDEX, AND ConvTENTS.

Vv

PAGE

489

499

549

589

APR 27 1898

The Habits and Structure of Arenicola
marina.

By

F. W. Gamble, M.Sc.,
and

J. H. Ashworth, B.Sc.,

Demonstrators and Assistant Lecturers in Zoology, Owens College,
Manchester.

With Plates 1—5.

ConTENTS. -
PAGE PAGE
1. Distribution: Varieties : 7. Gills : ; 3 wee
Habits . : : : 1 | 8. Nervous System and Sense-
2. External Features: Seg- | organs . : : x Ok
mentation: Skin: Sete. 5 9. Nephridia MAD ae ek:
3. General Anatomy of the | 10. Celom. : : 2. 26
Internal Organs. ; 9 | 11. Reproductive Organs -; =O
4. Musculature . : . 10 12. General Summary . . 34
5. Alimentary Canal . . 12 = 18. Literature pee 27
6. Vascular System. a a le |

1. Distribution: Varieties: Habits.

Tue common lugworm and its coiled castings of sand are
familiar objects on almost all the sandy and muddy shores of
Western Europe, but the exact geographical range of the
species is doubtful. It has been recorded from the shores of
North Siberia, Spitzbergen, Iceland, and Greenland (Wirén,
1883; Levinsen, 1883). On the north-east coast of America
it has been found from the Bay of Fundy to Long Island

vou. 41, PART 1.—NEW SER. A

2 F. W. GAMBLE AND J. H. ASHWORTH.

(Verrill, 1881). On both sides of the Atlantic, latitude 40° N.
marks approximately the southern limit of Arenicola
marina. South of this it is replaced in the Mediterranean
by A. Claparédii, Lev., and by A. cristata, Stimps., the
latter also ranging on the west side of the Atlantic from Cape
May (N. J.) to the Caribbean Sea. Its reputed occurrence on
the north coast of Alaska (Murdoch!), at Vancouver Island
(Marenzeller, 1887), Coquimbo, and South Africa requires
confirmation.

An abundant, widely ranging, and undoubtedly old form
such as Arenicola, might be expected to vary considerably
in its habits and structure, though it has not hitherto been
ascertained how far this is the case. Having paid special
attention to this point, we have found that there are (at least
on the Lancashire coast) two varieties of A. marina, differ-
ing in habits, structure, and times of maturity, and that there
is, in addition, considerable individual variability.

(1) From high-water mark down to the beginning of the
Laminarian zone, the common shore lugworms (or “ lugs,” as
fishermen call them, in contradistinction to the second variety,
or “‘ worms”) sink their U-shaped burrows to a depth of from
one to two feet below the surface. One end of the burrow is
marked by a casting, the other by a “countersunk”’ hole,
through which the head of the lugworm is protruded when
the tide comes in. The size and colour of the animal vary
with the amount of muddy organic matter in the sand. Where
there is comparatively little mud, the Arenicola average
about seven inches in length and are somewhat transparent,
so that the superficial blood-vessels can be clearly seen through
the thin body-wall. The gills, which are not very strongly
developed, are composed of nine to eleven branches, each pro-
vided with three to five pairs of short lateral twigs (Pl. 1,
fig. 3). The proboscis and prostomium are only slightly
pigmented, and being very vascular, appear red in colour.

Where, however, the amount of organic matter is consider-
able, the worms are usually about ten inches long, and their

1 * Proc. U. S. Nat. Museum,’ Washington, vol. vii, 1884, p. 522.

HABITS AND STRUCTURE OF ARENICOLA MARINA. 3

prostomium, proboscis, gills, and epidermis are black. The
gills are better developed than those of worms living in purer
sand. These differences are probably due to more abundant
nutrition. The time of maturity of both these forms of the
littoral variety on the Lancashire coast is the summer, while
at St. Andrews they are found mature from February to
September.

(2) The second variety occurs on the Lancashire coast
at the upper part of the Laminarian zone. Almost all the
Arenicola from this zone (which can accordingly be obtained
only at low spring-tides) are of this kind, which when fully
mature, as it is from February to May, is probably one of the
largest Polychets of our shores, measuring as much as fifteen
inches in length and three in girth. It is almost black, the
prostomium, proboscis, and the base of the gills being markedly
so. The tail is shorter in proportion to the length of the
body than in the littoral variety. The burrows are of con-
siderable length, three feet or more, and are not U-shaped,
but simply vertical. Like those of the littoral variety, they
are lined by a greenish coating of mucus. The dark ‘‘ worms”
appear to keep nearer the surface of the sand in cold weather
than in summer,—at least, during the winter of 1893-4 large
numbers were thrown up on the beach at Blackpool.

The most distinctive character, however, of this ‘‘ Lami-
narian” variety is the gill (Pl. 1, fig. 2), which presents a
structure hitherto only known in Arenicola cristata,
Stimps. Instead of the somewhat simple gill seen in the shore
lugworms, there is in the ‘‘ Laminarian” variety a highly
developed pinnate structure, consisting of about twelve
branches united by a connecting membrane at their bases, and
bearing ten or more pinnules on each side of the main axis.
Such a gill is undoubtedly a much more efficient respiratory
organ than the gill of a shore lugworm, though it does not
appear to possess the same power of contractility as the latter,
and hence probably does not contribute so much to the move-
ment of the blood. In some old specimens the gills lose many
of their finer branches, perhaps owing to friction or to the

A F. W. GAMBLE AND J. H. ASHWORTH.

attacks of enemies,! and in such cases there is an approxima-
tion to the type of gill seen in the littoral variety, though a
certain amount of difference is always observable.

Thus there appear to be two varieties of the common
lugworm on the Lancashire coast, distinguished by their
habits, external features, and periods of maturity, but there
are no important structural points of difference.

The habits of Arenicola marina at the breeding season
are still to a large extent unknown, and developing eggs have
not hitherto been obtained. It has been stated that, when
mature, the animal is in the habit of swimming freely (Ehlers,
1892, a), but we are unable to confirm this. The post-larval
stage, however, appears to be, for a short time, pelagic (Benham,
1893).

The curved burrow of the shore lugworm is formed by the
combined action of the proboscis, the swollen anterior region
of the body, and the waves of muscular contraction which pass
along the body from behind forwards. When the proboscis is |
everted and pressed into the sand, the prostomium is slightly
retracted into the body. The proboscis is withdrawn full of
sand, again everted, and the body is thrust forward, partly by
contraction of the longitudinal muscles, partly by a peristaltic
wave produced by the circular ones. The anterior end is in
this way rendered swollen and tense, and is able to enlarge the
burrow, and thus a passage is gradually eaten through the sand,
smoothened by contact with the skin, and lined by the mucous
secretion of the epidermis. The gill region being narrower
than that which precedes it, is thus, to a certain extent, pro-
tected from friction, while, as if to ensure this, the notopodial
pencils of bristles are directed so as to protect the gills. After
burrowing vertically downwards for a depth of from one to
two feet, the worm forms a horizontal or oblique gallery, and
then a second vertical one which ends at the ‘‘ countersunk ”’
hole, through which the anterior part of the worm may pro-
trude, and so bathe the gills in fresh sea water.

* See the curious account of the ravages of Corophium longicorne,
by d’Orbigny, ‘Journal de Physique,’ 1821.

HABITS AND STRUCTURE OF ARENICOLA MARINA. D

The amount and value of the work done by lugworms has
been estimated on the shore of Holy Island by Mr. Davison
(1891), and has also been adverted to by Mr. Hornell under the
name of “ cleansing of the littoral.” Mr. Davison finds that
the castings are larger and more numerous above than below
half-tide; and as the result of several estimates and measure-
ments he calculates that on the Holy Island Sands, the entire
layer of sand, to a depth of two feet, passes through the bodies
of the lugworms which live in it, once in twenty-two months,
and that in a year the average volume of sand per acre, which
is brought to the surface in the form of castings, is 1911 tons,
representing, when spread out, a layer of thirteen inches in
thickness over the surface of the sands.

2. External Features.

Segmentation.—The body is divided into an anterior
chetigerous portion, a middle branchial one, and a posterior
caudal region or tail. The first region begins with the pros-
tomium, and is followed by a short achetous portion (fig. 1,
MET), which in many specimens appears to be composed of
four annuli, divided, however, by secondary circular markings.
The first chetigerous annulus is produced into a strongly
marked ridge, just behind which the notopodial sete (Chn.')
are inserted, the corresponding neuropodia (Nm.') being very
short and containing only a few sete. The intervals between
the chetigerous annuli are subdivided into rings, of which
there are, in the ‘ Laminarian” variety,22444..., and
in the littoral variety 23444... respectively.

The cheetigerous annuli do not mark the true somites into
which the body is divided. From a consideration of the
internal anatomy (see p. 10) we have reasons for believing
that, in the middle region of the body, the second groove
behind each chetigerous annulus marks the boundary between
the somites. A somite is, therefore, composed of a chetigerous
annulus together with three annuli in front of, and one behind,
it. The parapodia are not situated at the beginning, but
slightly behind, the middle of the somites to which they

6 F. W. GAMBLE AND J. H. ASHWORTH.

belong, thus confirming Benham’s observations on the post-
larval stage (1893).

The anterior region of the body is thus composed of the
prostomium, six chetigerous somites, and a region between
these, made up probably of two somites, but the exact number
is somewhat doubtful. (See Plate 1, fig. 1, and explanation,
p- 39.)

The second or branchial region of the body is composed of
thirteen somites, and is distinguished by the presence of gills,
a pair of which are attached to a slight fold of the skin just
behind the notopodia. ‘The first gill is variable, usually fairly
well developed, but always smaller than the rest and sometimes
absent. The gills about the middle of the branchial region
are frequently, but not always, the largest. Both the gills
and notopodia are very sensitive, and are retracted from time
to time on the application of stimuli, such as a strong light.
This contraction of the gills proceeds sometimes as a wave
down the body, and as Milne Edwards (1888) pointed out in
his classical paper, considerably assists the circulation of the
blood. The neuropodia in the branchial region extend towards
the mid-ventral line, so as almost to meet, and are only
separated by a groove which marks the line of the nerve-cord.
This groove is continued on to the prostomium by a pair of
diverging arms (‘‘ Metastomial grooves”) underlying the cir-
cum-cesophageal nerve connectives (Pl. 4, fig. 19, C. MZ.).

The tail, which is devoid of setz and gills, is marked by a
large number of secondary annuli, crowded together at first,
but arranged in distinct somites of about five each, towards the
hinder end. The caudal region varies much in length; some
specimens have about thirty somites, but the number is not
constant, possibly owing to the tendency of the worm to throw
off the last few segments when irritated.

There is no change in the internal organs to mark the
somite which bears the first gill, but the transition from the
branchial to the caudal region is accompanied by the loss of
parapodia, oblique muscles, and branchial vessels.

External apertures,—The mouth (Pl. 4, fig. 19, C, 70.),

HABITS AND STRUCTURE OF ARENICOLA MARINA. fi

when the proboscis is withdrawn, is a slightly crescentic trans-
verse slit, bordered by papille and somewhat overhung by an
upper lip. The anus, which is terminal, is often protruded,
and the thin vascular swollen lips of the aperture project
behind the last caudal segment.

The opening of the ‘ nuchal organ” is a fairly wide slit on
the upper and hinder border of the prostomium (PI. 4, fig. 19,
aands, NV.). Through this aperture, sea water (or a mixture
of sea water and the secretion of the surrounding glandular
cells) is probably introduced.

The openings of the otocysts are difficult to see. They lie
behind the prostomium on each side of the anterior end in the
position marked OT’. (Pl. 4, fig. 19, a and B). Each is placed
at the point of intersection of the first transverse groove
following the prostomium, with the oblique ‘‘ metastomial ”
groove which marks the position of the nerve commis-
sure.

The nephridial openings (fig. 1, NO), six in number on
each side, though not so distinct as in some species (e.g. A.
Claparédii), are not difficult to find. The first is placed
behind and at the upper edge of the fourth neuropodium, and
the other five in corresponding positions on the succeeding
somites. They are minute slightly oblique slits, sometimes
exhibiting tumid lips.

Skin.—The skin is subdivided into raised polygonal areas
separated by corresponding shallow grooves, and is noteworthy
in being devoid of special glands. Wirén (1887) has shown
that the grooves are composed of columnar cells containing
pigment granules, the raised areas being made up partly of
larger cells containing still greater quantities of pigment
granules and partly of clavate mucus-forming cells, which
produce the slimy covering of the animal with which the
burrow is lined.

The 5 per cent. formalin solution of the epidermal pigment
is fluorescent, but does not yield any absorption bands, merely
cutting off the rays at the blue end of the spectrum. In suc-
cessively thicker layers of this solution, first the violet, then

8 F. W. GAMBLE AND J. H. ASHWORTH.

the blue, and lastly the green portions of the spectrum were
cut off.

MacMunn (1889), however, has shown that the alcoholic
extract of the integumental pigment shows a band in the blue
and green (A 503—468) ; that the residue of this solution if
dissolved in ether or chloroform yields two bands, X 503—474,
and A 465—446; and that the residue of this solution again
being dissolved in nitric acid gives two bands, A 500—468,
and A 472—443, so that a chlorophan-lke lipochrome is
present. It is probable that the pigment (melanin) of the
skin is derived from the lipochrome of the yellow “glandular”
tissue of the stomach, since the alcoholic extract of the latter
yields a similar absorption spectrum.

Further investigation will be required to show in what way
the transference of the pigment from the yellow peritoneal
cells to the epidermis is brought about, and whether the dark-
coloured, hairy-looking investment of the ventral vessel and its
branches (Pl. 2, fig. 5) contributes to the melanin of the
skin. In this connection the intermuscular extension of the
ccelom, bringing it almost into contact with the epidermis at
certain points, must be borne in mind (see p. 29).

Setz.—The notopodial sete are long capillary structures
averaging 6 mm. in length, and bearing several rows of minute
free and pointed hair-like processes (Pl. 3, fig. 10). The
neuropodia in the anterior somites, which at first contain few
sete, gradually extend by addition of new ones at their ventral
edge, so as to almost reach the mid-ventral line (Pl. 1, fig. 1).
By isolating the entire band of the sete the different stages
in their development may be seen. The youngest sete are
always at the lower end of the series ; the point of each seta
is formed first, then the toothed ridge, and lastly the shaft.
The fully-developed ventral seta is frequently almost smooth,
owing to the wearing down of the teeth behind the apex.
The middle of the shaft is straight, the inner end bent ven-
trally, and the outer end bent slightly dorsally, ending with a
finger-shaped process bordered on the convex side by a toothed
ridge, while on the concave side it is slightly produced at one

HABITS AND STRUCTURE OF ARENICOLA MARINA. 9

point into a minute process (Pl. 3, fig. 12, proc.). This
process is more constant in the Laminarian than in the littoral
variety. It appears to correspond, in position, to the eharac-
teristic tuft of hairs on the ventral setz of the Maldanide.

According to the age of the specimen the ventral setz
differ in shape, and in the development of the toothed ridge.
In sete from a small specimen (17 mm. long) the apex was
bent more sharply on the shaft than in old examples, and the
teeth were very prominent (Pl. 3, fig. 9). Apparently the
production of fresh ventral sete goes on slowly throughout
life, and the form which they assume before being cast out of
the body, varies at different ages. Their size of course varies
with the age of the worm to which they belong (see Pl. 3),
but in a worm of average size their length is about ‘5 to
‘8 mm.

3. General Anatomy of the Internal Organs (Pl. 2).

In opening the body-cavity by a dorsal incision, the middle
part of the alimentary canal is usually forced out through the
cut by the pressure of the somewhat viscous ccelomic fluid.
Normally this portion of the canal, being longer than the
section of the celom in which it les, is swung to and fro by
the movements of the body. This freedom of motion is ensured
by the absence of mesenteries, by the absence of any vessels
running from the body-wall into the dorsal vessel, and by
the length and flexibility of the branchial and nephridial
vessels, which are the only connection between the stomach
and the body-wall.

The ccelom is exceedingly spacious, and continuous from one
end of the body to the other. In front it is divided trans-
versely by the origins of the buccal retractors (B. Sh.), which
form a sheath round the proboscis, and by three septa or
diaphragms (Pls. 2 and 3, figs. 5 and 6). The first of these
septa (Dphm.) is placed obliquely, arising below behind the
level of the first neuropodium, and being inserted dorsally in
front of the first notopodial sacs. The result of this arrange-
ment is that between the first and second diaphragms two pairs

10 F. W. GAMBLE AND J. H. ASHWORTH.

of setal sacs occur, caused by the forward shifting of the upper
edge of the first diaphragm (fig. 5). The second and third are
inserted both above and below, opposite the second groove
behind the second and third chetigerous annuli. Between the
first and second diaphragms, dorsal and ventral mesenteries
occur, supporting the corresponding vessels; and it will be
noticed that the dorsal mesentery ends in front, exactly where
the first diaphragm would be inserted if it corresponded with
the other two. The third diaphragm is perforated by the
funnels of the first nephridia. There are, then, three diaphragms
and not, as so often stated, four, and, while affording valuable
evidence of the extent of the first and second chetigerous
somites, they do not help in determining the number of seg-
ments which compose the achetous portion following the
prostomium.

Behind the last diaphragm the body-cavity is unsegmented
up to the base of the tail. The segmental arrangement of the
organs, however, can be recognised by taking the funnels of
the nephridia as marking the anterior ends of the somites.
The slight amount of connective tissue supporting the long
afferent and efferent vessels (segmental vessels) (Pl. 2, fig. 5)
of the nephridia and gills, may be regarded as the remains of
the septa. Allied species of Arenicola fully confirm this
view.

At the level of the thirteenth pair of notopodial sacs, the
segmental afferent and efferent blood-vessels, which have
hitherto run nearly parallel across the coelom, diverge. At
the base of the tail, the connective tissue between them in-
creases slightly in amount, septa forming which are continued
down to the end of the body (fig. 5, C. Sp.).

4, Musculature.

The muscles of the body-wall are arranged in (1) an outer
circular sheath, subdivided in the anterior and middle regions
of the body into hoops, which cause the annulation of the
skin; and (2) an inner longitudinal sheath of considerable
strength and thickness divided by the nerve-cord and lines of

HABITS AND STRUOTURE OF ARENIOCOLA MARINA. it

insertion of the notopodial sacs into three parts, two ventro-
lateral and one dorsal (Pl. 4, fig. 23). The intermuscular
spaces are filled by ccelomic fluid, and are probably lined by a
delicate peritoneum.

In the anterior region of the body there are a few circular
muscle-bands which are stronger and more obvious than the
rest (fig. 5, Df. Cire.).

The oblique muscles, which divide the ccelom longitudinally
into three compartments, commence behind the third dia-
phragm, and disappear at the base of the tail. These muscles
are arranged in thin broad bands, arising at the sides of the
nerve-cord, and are inserted right and left into the body-wall
at the level of the notopodial sacs. They partly cover the
nephridia, and in some specimens a muscle-band is attached
to each nephrostome.

The musculature of the buccal mass consists of a strong
sheath of fibres derived from the longitudinal layer just behind
the first diaphragm. This sheath, which is loosely attached to
the proboscis by slips which run through the cceelomic space
between the two structures (PI. 3, fig. 6, B. Sh.), is inserted into
the anterior part of the proboscis. Pressure of the ccelomic
fluid at this point causes eversion of the buccal mass, which
is withdrawn by the contraction of its muscular sheath.

The prostomium is retracted by a small sheet of muscle
which arises partly from the longitudinal layer dorsally, and
partly from the muscular covering of the circumcsophageal
connectives ventrally, and it is inserted into the ventral surface
of the brain, and the ventral and hinder edge of the nuchal
organ (Pl. 3, fig. 6, Nu. Tr.).

The parapodial muscles are modifications of the longitudinal
layer. One, the retractor of each notopodium, is remarkably
long, reaching to the side of the nerve-cord (Pl. 3, fig. 18,
Rn.). The protractors (Pn.) of the notopodia are six to eight
in number, three to four being placed in front of, and three
to four behind, the setigerous sac. They arise from the body-
wall just below the dorsal longitudinal vessel, and are inserted
into the base of each sac,

12 F. W. GAMBLE AND J. H. ASHWORTH.

The position and relations of the three anterior septa or dia-
phragms, of the dorsal and ventral mesenteries between the
first two of these, and the presence of regularly arranged septa
in the tail region, have already been noted. It may be added
that a pair of outgrowths from the first diaphragm lie under
the cesophagus, opening anteriorly into the coelomic space in
front of the first septum. They are very vascular, and con-
tract rhythmically every three or four seconds during life, and
are doubtless of use in everting the proboscis (Pl. 2 and 8, figs.
5 and 6, Dph. Ph.).

In the caudal region the intestine is attached both above
and below to the body-wall by mesenteries, in which the dorsal
and ventral vessels lie.

5. Alimentary Canal (PI. 2).

This consists (1) of an eversible buccal mass (Bucc. M.),
of a pinkish or greenish-brown colour, which lies in front of
the first septum; (2) of an esophagns, of a light brown colour,
provided with a pair of glandular pouches behind the third
diaphragm ; (3) of a gastric region, with yellow glandular walls,
extending from the level of the heart to about that of the
twelfth or thirteenth notopodium ; and (4) of an intestine, of a
dark brown or almost black colour, folded in a concertina-like
manner by the caudal septa, and opening at the terminal anus.

During life the buccal mass (or “ proboscis”) is constantly
being everted and withdrawn, carrying sand into the cesopha-
gus. During eversion several rows of curved, pointed, vascular
papille (B. Pap.) are first extruded. Thése papille (Pl. 3,
fig. 7) in old specimens are tipped with chitin, and recall the
armature of the proboscis in certain Sipunculids (e.g. Phas-
colion collare!). ‘Then the more globular portion of the
buccal mass, covered with minute rounded processes, is pro-
truded. Finally, when fully everted, the buccal aperture is
surrounded by a few pointed pigmented papillz, which are
continuous with the lining of the first part of the cesophagus.

1 Selenka, ‘ Die Sipunculiden,’ 1883, pl. vi, fig. 74.

HABITS AND STRUCTURE OF ARENICOLA MARINA. 13

The csophagus! itself is slightly looped behind the second
diaphragm. It is a thin-walled distensible tube, the first part
of which is lined by non-ciliated mucus-forming cells. The
middle portion is lined by a cuticle, and the posterior part by
cells resembling those of the stomach in bearing cilia. The
cesophageal pouches (Oe. Gl.) are somewhat flask-shaped, and
open into the cavity of the csophagus by a short tubular
stalk. They are usually greenish in colour, but have a slight
reddish tinge on account of their very large blood-supply.
Their blood-vessels are connected with the lateral cesophageal
and dorsal vessels. The cavity of the pouch is subdivided by
twenty-five to thirty incomplete partitions, produced by in-
folding of the wall of the pouch, and therefore covered on each
side by the epithelial lining of the pouch (Pl. 4, fig. 22).
Between the epithelial lamelle is a blood-sinus, which is
slightly enlarged at the inner end and slightly thickened at the
edge of each partition. The csophageal pouches are lined by
ciliated epithelium, covered with a fairly stout cuticle, and
contain glandular cells. The walls of the cesophagus are
marked by longitudinal and circular muscular impressions.

The stomach, marked out by the patches of yellow tissue
on its walls, extends from the level of the heart to about the
twelfth notopodial sete. As we have already stated (p. 9),
the stomach is bent upon itself and loosely attached to the
body-wall. The patches of “chlorogogenous” tissue are at
first arranged in symmetrical oval areas right and left of the
dorsal blood-vessel, while more ventrally they are placed in
two or three less regular series, and are separated from one
another by a network of blood-vessels.2 About the level of
the tenth setz these yellow areas all become subequal and
arranged in a spiral manner, ending at the level of the four-
teenth setz.

Stomach and Intestine.—The muscular wall of the

* The histology of the alimentary canal has been carefully investigated by
Wirén (1887, p. 31). Our results agree very closely with his.

? This network is considered by Wirén and others to be parts of a con-
tinuous sinus. We are not convinced, however, that this is really the case,
and our reasons will be found on p. 17 infra.

14 F. W. GAMBLE AND J. H. ASHWORTH.

gastric region is exceedingly thin, and composed purely of
circular fibres, which appear to confer very slight powers of
peristalsis upon the stomach.

The mucous lining is strongly folded, and is composed of
several kinds of cells. Some of the cells in all parts of the
stomach are ciliated, others are apparently digestive, and a
large number appear to secrete a mucus similar to that of
the cesophagus, the cells themselves being discharged into the
mucus which they help to form.

Commencing about the middle of the stomach (that is
between the ninth and tenth segments) is a ventral groove
formed by a couple of folds of its inner and lower surface.
This groove! (Pl. 4, fig. 238, Gv.) is provided with specially
long cilia, which produce a current of mucus from before back-
wards. There are other smaller grooves on the side walls of
the stomach and the anterior part of the intestine, whose
general direction is downwards and backwards, and which
open into the median ventral groove. ‘The direction of the
current in all these is from before backwards. The ventral
groove is continued back to the anus. ‘The intestine is dark
brown or nearly black in colour externally. Its mucous lining
is somewhat similar to that of the stomach, but is covered by
a thin cuticle, and is not ciliated.

The process of digestion in the lugworm has not been at all
fully investigated, but the series of events appear to be some-
what as follows. The sand or mud is mixed with the mucous
secretion of the esophagus, and is slowly carried backwards by
peristaltic contraction. At the junction of the stomach and
cesophagus the secretion of the wsophageal pouches is poured
upon the sand. Wirén regards the contents of these pouches
as acid and digestive. In several cases we have found the fluid
neutral. In the stomach several changes occur. ‘The secre-
tion of the gastric cells proper is probably digestive, and this,
together with a further amount of mucus, is mixed with the
sand, and shaken together by the swing of the loose gastric
loop. In this way the food, which apparently consists of the

1 This groove has only hitherto been noticed by Wirén (1887).

HABITS AND STRUCTURE OF ARENICOLA MARINA. 15

organic substances! in the sand, is brought into contact with
the digestive secretion. The ciliary action of the lateral and
ventral grooves probably separates the digested substances
from the sand and carries them slowly downwards and _ back-
wards. ‘The lining of the stomach is very thin, and the lateral
and ventral grooves are in specially close contact with the
blood-plexus, in which the flow is, probably, slowly forwards,
more rapidly in the sub-intestinal vessels. It seems probable,
therefore, that the blood in the visceral plexus conveys the
nutritive material to the hearts, which pump it along the
ventral vessel to the various parts of the body.

The action of the chlorogogenous tissue round the stomach,
and particularly of that in the neighbourhood of the ventral
vessel and its branches, is uncertain.

6. Vascular System (Pl. 2, fig. 5).

The blood-vascular system of Arenicola attains a high
degree of perfection. The large size of the chief vessels,
the great development of the capillary system (especially
on the walls of the alimentary canal), and the mechanism
for promoting the flow of the blood, are features that dis-
tinguish it.

There are two chief vessels running, one above, and the
other below, the alimentary tract from end to end,—the dorsal
vessel, which contracts fairly rhythmically from behind for-
wards; and the ventral vessel, which is feebly, if at all, con-
tractile. The walls of the gastric and intestinal portions of
the gut areenclosed in a blood-plexus, and the cesophageal region
is supplied by lateral vessels. The gastric vessels are connected
with the ventral vessel by a pair of “‘ hearts ”’ placed a short dis-
tance behind the esophageal pouches (fig.5, V.). These hearts
drive the blood from the gastric vessels into the ventral vessei.

The dorsal vessel (D V) arises near the anus, and as it runs
along the intestine gives off in each somite a pair of branches

1 Saint Joseph found in an Arenicola a whole Nereis almost digested.
‘ Ann. Sci. Nat.,’ series vii, t. xvil, 1894, p. 127.

16 F, W. GAMBLE AND J. H. ASHWORTH.

which are attached to the anterior face of the caudal septa, and
which run downwards and forwards to open into the ventral
vessel (Pl. 2, fig. 5). Of these there may be twenty-seven to
thirty pairs. In front of the caudal region each of the last
seven pairs of gills returns an efferent branch to the dorsal
vessel, and between these there are three or two pairs of
smaller branches which run round the alimentary canal from
the ventral vessel to open into the dorsal one. From the level
of the twelfth sete to the csophageal pouches the dorsal
vessel does not receive any segmental vessels from the gills or
nephridia, nor does it open directly into the heart (fig. 5). It
merely receives numerous branches from the gastric plexus.
In front of the heart it receives on each side a branch from the
third nephridium and the fifth setigerous sac; a branch from
the cesophageal pouches ; and one from the second nephridium
and fourth setigerous sac. It then runs on and, piercing the
third diaphragm, receives a branch running on the anterior
face of the diaphragm from the first nephridium and third
setigerous sac. On reaching the second diaphragm it receives
a branch from the second setigerous sac, and after piercing the
first diaphragm receives a branch from the muscles forming
the buccal sheath. Thence the dorsal vessel breaks up into
capillaries around the buccal musculature, prostomium, and
otocysts. From these capillaries the ventral vessel takes its
origin. It gives off a small unpaired branch running in the
first diaphragm and to its pouches; a paired branch arising
about midway between the first and second diaphragms to the
neural vessels and second setigerous sac; a single small vessel
supplying the second diaphragm and the neural vessels; an
unpaired vessel to the third diaphragm, to the neural vessels
in that region, and to the first nephridia; a pair of branches
to the neural vessels and second nephridia; and lastly, a pair
to the neural vessels and third nephridia. From this point
ouwards the ventral vessel supplies the setigerous sacs, body-
wall, nephridia (if present), and gills by large segmental vessels.
The ventral vessel is very large and turgid in the gastric region,
and is surrounded by tufts of dark brown chlorogogenous tissue,

HABITS AND STRUCTURE OF ARENICOLA MARINA. 7

which are also found in older specimens on the vessels running
to the body-wall. This chlorogogenous tissue is first seen 01
the ventral vessel about the level of the eighth pair of sete.
In the tail the ventral vessel ends in the obliquely p'aced
intestinal vessels which encircle the intestine, and which form,
along with the capillaries from its median terminal portion,
the commencement of the dorsal vessel.

Visceral Plexus.—Wirén (1887) maintains that the in-
testine and stomach are enclosed in a blood-sinus, thickened
along certain lines which have been called the dorsal, gastric,
and subintestinal ‘vessels.’ We are, however, of the opinion
that the so-called sinus is a close plexus of vessels, some of
which appear to have a distinct cellular lining. The dorsal
vessel is, at any rate, a perfectly distinct structure with
proper walls.

The subintestinal vessels (fig. 5, S. V.), which commence
just behind the heart and run backwards, are moderately large
up to the level of the thirteenth sete, but then taper rapidly
and gradually disappear. They each receive seven segmental
vessels. The first of these comes from the fourth nephridium,
the second from the fifth nephridium and the first gill, the
third from the sixth nephridium and second gill, and the other
four from the third, fourth, fifth, and sixth gills. The sub-
intestinal vessels open.through the plexus into the lateral
gastric ones, and so into the heart. The flow in these vessels
is probably slowly forwards.

The gastric vessels give off from the “auricle,” into which
they expand, a lateral cesophageal vessel (Oe. Lat.), which, after
giving off a stout branch to the cesophageal pouches, runs
forwards to the buccal mass, supplying the wall of the ceso-
phagus, as it does so, with numerous small branches.

Neural Vessels.—These are a pair of small vessels lying
one on each side of the ventral nerve-cord, and accompanying
it from one end of the body to the other. They arise round the
nerve-connectives from the brain from capillaries of the dorsal
vessel, and receive several branches from the ventral vessel (1)
midway between the first and second diaphragms, (2) from

vot. 41, part 1.—NEW SER. B

18 F. W. GAMBLE AND J. H. ASHWORTH.

the vessel running in the second diaphragm, (3) from a vessel
just behind the third diaphragm, (4 and 5) from the vessels to
the second and third nephridia. Near the middle of each
somite the two neural vessels are united by cross connections,
which also supply the nerve-cord (Pls. 2, 3, fig. 18, N. V.,
IN: CV);

Behind the third diaphragm the neural vessels supply the
oblique muscles by branches which run the whole length of
the bands, and are connected with the outer longitudinal
parietal vessel (fig. 13).

Vessels of the Body-wall.—This parietal system of true
vessels is highly developed in Arenicola marina. It con-
sists of two longitudinal vessels, (1) the nephridial longitudinal
vessel (fig. 22, N. LZ. V.) running just below the level of the
nephridiopores, and (2) the more important dorsal longitudinal
vessel (fig. 13, D. L. V.), which runs just above the level of
the insertion of the notopodial setal sacs. Both arise just
behind the first sete, and increase in size as they pass back-
wards. The former receive vessels from the nephridia, just
behind which they taper and disappear. The latter, which
may be traced to the anus, and are largest in the branchial
region, receive branches in each somite: (1) from the segmental
vessels ; (2) from its fellow of the opposite side. The body-wall
in the dorsal and lateral regions derives its blood-supply from
the nephridial and dorsal longitudinal vessels, and in the
ventral region from the neural vessels. These parietal vessels
(Par. V.) run just within the layer of circular muscles in
almost every groove between adjacent longitudinal muscle-
bands of the body-wall, are chiefly longitudinal in direction,
but at frequent intervals there are cross connections. Branches
from these vessels ramify between the bases of the epidermal
cells, and are accompanied by extensions of the coelom.

Hearts.—The hearts are a pair of muscular bulbous swell-
ings connecting the visceral plexus with the ventral vessel on
each side. Each commences with the thin-walled expansion
of the gastric vessel (“auricle,” fig. 5, d.v.) which, after
giving off the lateral cesophageal branch, opens into the ventricle

HABITS AND STRUCTURE OF ARENICOLA MARINA. 19

(V.). The cavity of the ventricle is small and broken up by
a spongy mass of cells. The ventricular walls are muscular,
and contract from above downwards, forcing the blood into
the ventral vessel. (We have sometimes seen an apparent
reversal of the heart’s action.) The spongy cardiac body
arises by ingrowths from the wall of the ventricle, chiefly in
the middle and ventral regions. It gradually encroaches on
the blood space, so as to reduce it considerably (PI. 5, fig. 36,
Card. B.) in an old specimen. The cardiac body in a young
specimen (fig. 38) is much smaller, and extends obliquely across
the heart, its general direction being downwards and back-
wards. The cells of the cardiac body in an old specimen
which we have examined are loosely arranged, so as to cause
the formation of a large number of intercellular spaces, some
of which are of considerable size, and which are in life filled
with blood (Figs. 36—38, B.S.). Between the cells there are
numerous fibres, which are probably muscular. The cells are
apparently of two kinds, which, however, merge into each
other: (1) cells whose protoplasm has a very vacuolated appear-
ance, and which contain few or no granules ( Vac. C.); (2) cells
which contain a large number of yellowish granules in the
protoplasm (G.C.). These latter cells are possibly glandular,
and correspond to those found in the cardiac body of other
Polychets. The function of the cardiac body may be, as
Schaeppi (1894) suggests, to prevent regurgitation of the blood
from the ventral vessel into the heart when the diastole com-
mences. The “cardiac body” of Polychets, as hitherto
described, is an unpaired structure lying in the dorsal vessel.
That of Arenicola, however, is paired and in no way con-
nected with the dorsal vessel. Hence a strict homology is
scarcely probable.

Blood.—As Professor Lankester was the first to point out,
the blood of Arenicola is strongly impregnated with hzmo-
globin, but there has been no thorough investigation of the
constituents of the plasma. Krukenberg (1882), it is true, made
some experiments which led him to believe that there were no
coagulable albumens in the blood of his specimens ; but as they

20 F. WwW. GAMBLE AND J. H. ASHWORTH.

were in a starving condition, a fresh examination is very
desirable. A large quantity of albumen is certainly present,
which when the specimens are fixed becomes very hard and
brittle.

We have seen small cells (4 4 in diameter) in the blood-
vessels of the nephridia, but it is doubtful if these are the
blood-corpuscles, which we have not been able to demonstrate.!

General Remarks on the Circulatory System.—No
other system of organs shows the true segmentation of the
body of Arenicola so well as this. The lines of demarca-
tion between the somites from one end of the body to the
other are marked by the segmental vessels passing from the
ventral to the dorsal vessel and breaking up on their way in
the body-wall, nephridia, or gills. Throughout the gastric
region, however, this arrangement is somewhat disguised, owing
to the loss of the connection with the dorsal vessel, an altera-
tion caused probably by the necessity for leaving this part of
the alimentary canal freely moveable.

Wirén evidently believes that there is no capillary system
except in the gills and the alimentary canal. He suggests
that the assimilation of food and oxygen by the tissues is
effected chiefly through the mediation of the ceelom, which he
points out is parcelled off in the intermuscular spaces, by
a channelling out of the subepidermic connective tissue, into
* perihzemal canals.” Though this suggestion isa valuable and
correct one, we have found a very perfect system of capillaries
in the skin in all parts of the body, and in the nephridia and
septa the same is the case. The extension of the ccelom into
the intermuscular and subdermal spaces has, however, all the
appearance of acting as the equivalent of lymph-spaces of
higher forms. The transformation of the constituents of the
blood into coelomic fluid takes place in all probability with
especial rapidity in the neighbourhood of the dark chlorogo-
genous processes of the ventral vessel (cf. Cuénot, 1891).

1 Since writing this we have discovered that these small cells are the
blood-corpuscles.

HABITS AND STRUCTURE OF ARENICOLA MARINA. 21

7. The Gills (Pl. 1, figs. 2—4).

The general characters of these organs have been mentioned
in the introductory part of this paper, and little remains to be
added.

There are thirteen pairs of gills from the seventh to the
nineteenth cheetigerous somites inclusive. The shape varies
from the short dendritic type of the littoral form to the delicate,
richly- branched gill of the Laminarian variety. The gills are
hollow, being outgrowths of the body-wall enclosing an exten-
sion of the clom, and what little evidence we have of their
development (see Benham, 1893) points to their being inde-
pendent structures, and not modified dorsal cirri.

The walls of the gills, though thin, are muscular, and there
are also muscular bands stretching across the cavity of the gill
(fig. 23); and Milne Edwards has pointed out that the contrac-
tion of the gills, which often proceeds like a wave from before
backwards down the sides of the body, must exert a powerful
influence in propelling the blood partly into the efferent vessels,
and partly to the parietal capillaries.

The ventral vessel supplies all the gills with their afferent
branches. The first seven pairs return the blood to the sub-
intestinal vessels, and so to the heart; while the efferent
branches of the remainder open into the dorsal vessel.

8. Nervous System and Sense-organs.

This system is composed of the brain, the cesophageal con-
nectives, the ventral nerve-cord, and the nerves arising from
these. We have not been able to demonstrate a visceral
nervous system.

The brain (Pl. 5, figs. 25, 26) is placed in the prostomium, of
which it forms the chief part, being only separated from the
epidermis by blood-vessels lying in extensions of the ccelom.
It is a small elongated structure, measuring ‘75 mm. in length
in ordinary shore lugs, and 1 mm. in the large ‘‘ Laminarian”’
variety. At its anterior end the brain is divided into two stout
cornua (A. Cr.), separated by a cleft containing blood-vessels.
About the middle of the brain the cornua unite, but only for a

22 F. W. GAMBLE AND J. H. ASHWORTH.

very short distance, a second connective-tissue partition divid-
ing the smaller posterior cornua (P. Cr.), which gradually taper
off and end at the hinder edge of the nuchal organ (Pl. 5,
fig. 25).

Sections of the prostomium of the littoral variety of
Arenicola (immature specimens, 4” long) exhibit a thick
covering of ganglion- and glia-cells, forming the dorsal surface
of the brain (fig. 24) ; a central fibrous portion; and a strong
ventral membrane, into which the greater part of the pro-
stomial muscles are inserted, though a few fibres are attached
in front of and between the anterior cornua (PI. 5, fig. 25).
In older specimens, and particularly in mature examples of the
“ Laminarian” variety, the ganglion-cells are more scattered,
and in other ways the brain shows greater differentiation. The
anterior cornua, for example, are not only deep and thick, but
give off from their dorsal surface short stout branches, along
which the ganglion-cells are scattered, and which supply the
prostomium. The central fibrous part of the brain also grows
out ventrally in these large examples, separating the hitherto
compact layer of cells and carrying them outwards or leaving
them in clumps, and not evenly arranged as in young Areni-
cola.

From the anterior cornua a large nerve arises on each side,
in front of the origin of the esophageal connectives. It passes
out to the under surface of the epidermis, and supplies the
papillz on the upper surface and the sides of the mouth. The
epidermis of the prostomium itself is in close contact through-
out its whole length with the ganglionic covering of the pro-
cesses arising from the dorsal and lateral surfaces of the brain.
The posterior cornua seem to be specially connected with the
nuchal organ, against which they lie and terminate (Pl. 4,
fig. 21).

The most remarkable histological feature of the brain is the
close contact between the large ganglion-cells of its upper
surface and the sensory epithelium of the prostomium (figs. 20
and 24). Racovitza (1896) has figured (PI. 5, figs. 48 and 49)
a similar condition in Clymene. It is only at this point

HABITS AND STRUCTURE OF ARENICOLA MARINA. 23

that the nervous system of the adult Arenicola marina can
be said to have an epidermal position. Elsewhere it is separated
from the epidermis by the circular musculature.

The circum-cesophageal nerve-connectives arise from the large
anterior cornua in the form of two thick cords, covered on their
outer surfaces by ganglion-cells (figs. 20, 21, 25, Oe. Comm.).
From them a pair of short nerves (fig. 26, OT. N.) arise supply-
ing the otocysts, and several longer ones are distributed to the
oral papille of the ventral region of the mouth. The line of
the connectives is marked externally by the ‘ metastomial
groove” (Pl. 4, fig. 19, C.), and the commencement of the ven-
tral cord by the junction of these grooves, which occurs on the
ventral surface just in front of the first chetigerous annulus.
The nerve-cord is protected by a delicate connective-tissue
sheath, a thin sheath of circular muscle, and a thin layer of
epidermis. Though nearly circular in section it is somewhat
flattened from above downwards, but exhibits scarcely a trace
of segmentation externally or internally. The ganglion-cells
are arranged in two veutral groups, while the fibrous portion
of the cord is dorsal. In the tail the ganglionic masses in-
crease in size, and are separated from the skin by a thicker
layer of circular muscle-fibres. Two “giant-fibres” are
present in the branchial region, a single one only in the
anterior and tail region.

From the cord a paired series of nerves is given off with
great regularity, one opposite each groove separating the annuli
of the somites, so that there are five nerves on each side of the
body in each somite. These lie in the body-wall just beneath
the circular layer of muscle, and, in some places where this layer
becomes obsolete, they lie just under the epidermis. Dorsally
these nerves thin out and become very difficult to trace.

Sense-organs.—There is no doubt that the prostomial
lobes, the nuchal organ, and the otocysts are sense-organs ; but
there are, in addition, certain other structures, such as the
sete! of the notopodia and some of the buccal papille, which,

1 Retzius has described free nerve-endings on these sete. ‘ Biologiska
Foreningens Forhandlingar,’ Bd. ii, Hefte 4—6, 1891, p. 85.

24 F. W. GAMBLE AND J. H. ASHWORTH.

on account of their position, movements, and the nerves ending
in them, may be considered as probably belonging to this
category.

Professor Ehlers’ (1892) account of the nuchal organ and
otocysts is an almost exhaustive description of these organs in
Arenicola. We have worked over the whole subject again,
however, and are able to add a few points to this important
paper.

The nuchal organ belongs to the prostomium, whereas the
otocysts belong to the metastomium. The prostomium and the
nuchal organ are found, in varying degrees of complexity, in
nearly all Polychets; the otocysts, however, occur in few and
widely separated families.

The general appearance of the prostomial lobes and the
opening of the nuchal organs have already been described.
Seen from the dorsal surface the former consists of a small
median papilla and two larger lateral prominences (Pl. 4,
fig. 19), which together correspond with the single prostomial
papilla of allied forms (cf. Racovitza’s figure of Leiocephalus,
1896, pl. v, fig. 5). In young Arenicola these lobes are
transparent, and therefore red from the underlying blood-
vessels. In old specimens they become dark-coloured and
opaque from the deposition of pigment in them. In no species
of Arenicola have eyes been discovered, although they are
known to occur on these lobes in many related genera.

The prostomial epithelium is a complex of several distinct
kinds of cells,—unaltered columnar elements, fusiform sense-
cells, each ending in a conical prominence, glandular cells, and
apparently also ‘ wandering cells” from the body-cavity.
Underneath the epithelium is a connective tissue continuous
with the supporting tissue, the neuroglia of the brain, which
binds together the large ganglion-cells of the cornua of the
brain. The prostomial sensory structure thus formed is very
sensitive to light, but what function it subserves has not been
determined with accuracy.

Nuchal Organ.—To the outer side of the lateral prosto-
mial lobes is a depression guarded externally by a fold (just

HABITS AND STRUCTURE OF ARENICOLA MARINA. 25

above Nu., Pl. 4, fig. 19, B.). These two pits form the beginning
of the nuchal organ and indicate its paired origin. Further
back they unite to form a transverse groove (bordered by the
hinder edge of the prostomium), which is continued inwards as
a deep pit to the hinder margin of the brain (PI. 5, fig. 25).
From the posterior cornua of the latter the nuchal organ is
innervated.

In its paired form and under the names ‘“‘ Wimperorgane,”
“ Wimpergriibschen,” the nuchal organ is well known in
almost all families of Polychets, and a similarly placed organ
is found in Sipunculids,! not to mention other more distantly
related groups. It is always associated with the posterior lobe
of the brain, and arises as a pair of pits from the surface of
the prostomium. Of its development in Arenicola, however,
we have no evidence, but the two depressions in front of the
main part of the organ, together with the paired nerve-supply,
point to its double nature.

The epithelium of this deep, pigmented pit (PI. 4, fig. 21, Nw.)
is composed of long columnar ciliated cells, glandular cells
which secrete the mucus in which the cilia work, and slender
sense-cells. It seems probable that the whole organ is olfac-
tory in function.

Otocysts.—The otocysts of Arenicola marina are a pair
of flask-shaped structures projecting into the body-cavity close
to the outer edge of the csophageal nerve-commissures.
They open externally by a couple of apertures (Pl. 4, figs. 19,4
and B, OT.), at that point on the “ metastomial groove ” where
the latter is crossed by the first groove of the body following
the prostomium. The body of the flask is placed at an angle
with the ‘‘ neck,” and contains the otoliths. It is lined by
non-ciliated columnar sense-cells and supporting cells, which
are surrounded by the nerve-fibres and connective-tissue fibrils,
figured by Ehlers (1892, pl. xii). The neck of the otocyst is
made up of a columnar epithelium covered with a thick cuticle,
which gradually merges into the epidermis of the external
surface, and ciliated cells only occur in its lower portion. A

1 Ward, ‘ Bull. Mus. Harvard,’ vol. xxi, 1891, p. 148.

26 F. W. GAMBLE AND J. H. ASHWORTH.

short nerve from the csophageal commissure supplies the
otocyst.

If the otocyst of a fresh shore lugworm be rapidly dissected
out under sea water and mounted, the sand-grains will be seen
to execute a most extraordinary movement. Each one is
rotating slowly and jostling its fellows, so that the whole
contents of the flask are in a state of commotion. ‘The fluid
in which the otoliths move is slightly viscous, and is a secretion
of the walls of the otocyst, mixed with a little sea water. The
sand-grains are covered with a distinct layer of some chitinoid
substance soluble in boiling potash. Acids have no appreciable
effect upon these grains, and under the polariscope they react
as quartz does. Hence it seems clear that the otoliths of
Arenicola marina (the other species of the genus differ
most remarkably in this respect, as well as amongst them-
selves) are quartz grains covered by an organic film, and sur-
rounded by a fluid which is not merely sea water.

Large specimens of the “‘ Laminarian”’ variety were examined
without being opened under sea water, and the otocysts were
mounted by us in celomic fluid. No movement of the oto-
liths was observed even in specimens which were perfectly
healthy in all respects. The otoliths sometimes filled the
_ expanded part of the organ, and it is possible that they had no
room to turn round. But it appears to us more likely that if
we assume the cause of the rotation to be the diffusion caused
by liquids so different as sea water, in which the preparation
was first mounted, and the somewhat viscous, perhaps albu-
minous fluid inside the otocyst; then if we mount the otocysts
in the same kind of fluid which they contain, no movement
should occur; and the experiment showed that in these cases
no movement did occur. The whole matter is one of very great
interest, especially in view of the probable functions of such an
organ as the otocyst. Ehlers has suggested that the movement
is due to the cilia at the bottom of the neck of the otocyst ;
but the same extraordinary movements are seen in the otocyst
of A. Grubii, which is closed and has no cilia. We quite
agree with Ehlers that there are no cilia in the expanded part

HABITS AND STRUCTURE OF ARENICOLA MARINA. 27

of the otocyst where the movement has been noticed, but we
are of the opinion that the quivering motion of the otoliths is
not a normal phenomenon, but is due to diffusion currents.

9. Nephridia.

There are six pairs of nephridia, belonging to somites 4 to 9.
Of these the first pair seems to be unrepresented in any other
species of Arenicola, and its variation in A. marina points
clearly to a gradual degeneration which it appears to be under-
going at the present time. It is not only the smallest of the
series, but is sometimes represented merely by a funuel or by
the secretory and terminal portions. Very rarely both the first
nephridia are mere funnels, and again one may be fully
developed and the other rudimentary, but they are never abso-
lutely wanting. Their small funnels, which are of a bright
pink colour, are placed on the anterior face of the third
diaphragm with the long axes vertical (Pl. 2, figs. 13 and 14).
One lip (the outer) is produced into processes corresponding to
the dorsal lip of the other nephridia. The secretory portion is
elongated, narrow, and usually brownish in colour, and the
terminal portion opens just above the fourth neuropodium
(Pl. 1, fig. 1) at a decidedly lower level than is the case in the
succeeding nephridiopores.

The remaining five pairs are always in adults fully de-
veloped. They are attached to the body-wall partly by connec-
tive tissue, partly by the broad bands of oblique muscle
which obscure them at first sight (Pl. 2, fig. 5). The nephro-
stomes are very long, and bent upon the rest of the organ.
The narrow slit-like aperture has a dorsal vascular lip bearing
finger-shaped or spatulate ciliated processes, and an entire
ventral one. The cilia just within the mouth of the funnel
are exceedingly long, and produce a current tending to carry
celomic fluid and corpuscles into the cavity of the organ.
The middle or secreting portion is brownish (in old worms
almost black), owing to the excretory granules which are
formed in its cells. ‘The terminal rosette-shaped bladder, which
is slightly lighter in colour, opens by a minute slit-like aper-

28 F. W. GAMBLE AND J. H. ASHWORTH.

ture through the body-wall, which thins out at this point
(Pls. 1 and 4, figs. 1 and 22, NO.).

The blood-supply to the nephridia (Pl. 4, fig. 18) is derived
from the ventral segmented vessels, which divide, one branch
going to the funnel of the nephridium and the other to the
body-wall. The former traverses the funnel, sending a vessel
into each of the ciliated processes, and giving off numerous
small branches to the lips of the funnel. After traversing the
funnel the vessel runs over the secreting portions of the nephri-
dium, supplying the genital strand in its course, and finally
ramifies on the terminal portion. The blood is collected again
into small vessels, which open into the dorsal longitudinal or
nephridial longitudinal vessels of the body-wall, from which it
is returned largely to the dorsal or subintestinal vessels, but
in part passes into the parietal vessels.

In young specimens the funnels are naturally simpler, but
have similar positions and relations, as may be seen in figs. 16
—18, which show nephridia from worms 29°5 and 44 mm.
long, in which the processes on the dorsal lip are being formed.
In the post-larval stage (Benham, 1893) the nephridia have no
funnels, the development of which has still to be investigated.

10. Celom.

The colom of Arenicola is well developed, and continuous
in all its parts. Not only does it form the space between the
alimentary tract and body-wall from one end of the body to
the other, but it is carried along with the blood-vessels into
the intermuscular spaces. Thus the blood-vessels of the pro-
stomium, of the buccal sheath, and of the body-wall generally,
are accompanied by ccelomic canals which very probably serve
as lymphatic spaces from which nutritive matters can be
absorbed by the surrounding tissue, and into which waste
nitrogenous substances may be excreted.

The segmentation of the body-cavity is very faintly marked.
Anteriorly three diaphragms, perforated just above the nerve-
cord, are present, whose position and relations are indicated

HABITS AND STRUCTURE OF ARENICOLA MARINA. 29

on fig. 5, and Pl. 3, fig.6. The whole middle region of the
body is devoid of septa, which, however, reappear on the last
two somites of the branchial region, and are present through-
out the tail in a complete form, though they are perforated to
allow of the more thorough circulation of the celomie fluid.

Arenicola fresh from the sand exhibits a series of peri-
staltic waves of the body-wall from behind forwards, which
can be easily seen if the gonads are sufficiently developed to
cause slight swellings, which each wave carries forwards.
These waves of fluid are probably of considerable physio-
logical value. They assist the circulation of the fluid, the
celomic cells, and the developing reproductive cells. They
inflate the anterior digging part of the worm, and thus assist
in burrowing. By their action the contents of the gut will
tend to travel slowly backwards, the weak visceral muscu-
lature being probably insufficient by itself to cause the
requisite amount of movement of the sticky sand: while in
defeecation the main agent is doubtless the pressure of the
ccelomic fluid on the intestine, brought about by violent con-
tractions of the body-wall.

The ccelom is lined by a very thin layer of flattened cells,
which undergo remarkable changes in certain parts of the
body, resulting in the formation of (1) chlorogogenous tissue,
(2) ova or spermatozoa, (3) coelomic corpuscles.

The ccelomic fluid is a mixture of sea-water and globulins,
among which only paraglobulin has hitherto been detected
(Krukenberg, 1882, p. 87). We find that the specific gravity
of the fresh fluid (including corpuscles) varies slightly, but is
on the average 1°0288.1

On exposure to air this fluid coagulates, and a delicate
fibrous network is formed, binding the corpuscles together.
If carmine is injected into the celom, it is removed by the
coelomic corpuscles, by the cells lining the celom and by the

' It was found to be least (1:0270) in specimens which had been kept for
some time in sea water, and greatest (1°0311) in those which had been kept

for thirty-six hours in moist seaweed only. The specific gravity of the sea
water used was 1°0264,

30 F. W. GAMBLE AND J. H. ASHWORTH.

nephridia, and there is no trace of carmine in the ccelom after
forty-eight hours.!

Celomic Corpuscles.—These abundant cells occur in
two chief forms, which probably pass into one another. The
first varies from 8 to 20 w in length, is ameeboid, and usually
contains yellow or brown granules of a very highly refractive
character. The pseudopodia are often grouped at the two
ends of the cell (PI. 5, fig. 24). The longer forms of this kind
of corpuscle pass into the second or spindle-shaped cells of the
ccelom, which measure as much as 50 wu in length, and contain
no coloured granules. These fusiform elements are most abun-
dant, and constitute the most characteristic features of the
coelomic contents.

The chlorogogenous tissue of the ventral vessel and its
branches in the body-wall consist of groups of cells about 20 u
in length, full of large slightly yellow or deep brown granules,
which are not highly refractive. The tissue in old black
worms is immensely developed, so as to completely cover the
vessel by the masses of hair-like threads, each thread consist-
ing of a small blind diverticulum of the vessel surrounded by
the chlorogogenous cells.

11. Reproductive Organs.

Thanks to the researches of Cosmovici (1880), Cunningham
(1887), Kyle (1896), and others, the true ovaries and testes of
Arenicola marina are now known to arise by proliferation
of the peritoneal covering of an extension of the blood-vessel
supplying the funnels of the nephridia. It is not certain that
there is a corresponding gonad on the first pair of nephridia,
but on each of the following five pairs the gonads are present
during the breeding season. In both sexes the organ is a mass
of cells, from which the ova or spermatoblasts break away at
a very early stage, to ripen in the celom. The rachis is con-
tinuous with the posterior angle of the nephrostome, and is
developed around a backwardly projecting process of the

1 Schneider, ‘Arbeit. Naturf. Gesellschaft,’ St. Petersburg, Bd. xxvii,
Heft 1, 1890.

HABITS AND STRUCTURE OF ARENICOLA MARINA, 31

nephridial vessel which comes off segmentally from the ventral
vessel (Pl. 4, fig. 18, G. V.).

In large Arenicola, at certain seasons, the vascular process
has no gonad, and it is possible, as Cuénot (1891) suggests,
that a formation of the ameeboid corpuscles of the ceelom takes
place at this point when the animal is not breeding.

After passing through the earliest stages of their develop-
ment in the genital rachis, the young reproductive cells may
be found at the breeding season in all stages of development
in thecelom. The ova do not exhibit any considerable changes
except in size in attaining maturity. They are nourished
either directly from the celomic fluid, or possibly (Cuénot,
1891) by the ameeboid cells acting as follicle-cells, though
we have seen nothing to support this view. Extrusion of a
polar cell (?) has been observed by us in an ovum only about
half the definitive size (Pl. 5, fig.35, aands). In thespherical
ripe ova (which measure ‘16 mm. in diameter) a distinct but
very thin vitelline membrane is present, and a small quantity
of food-yolk in the form of very small granules in the proto-
plasm. The production of ova by the fertile vascular pro-
cesses of the nephrostomes must be extraordinarily great,
since the spacious body-cavity of a large worm is eventually
filled to bursting with them by the end of February.

We have not followed the development of the spermatozoa
in great detail. The youngest stage which we have found in
the coelom contained eight spermatoblasts arranged round a
vesicular-looking blastophore (PI. 5, fig.30). Further division
and elongation of the outer ends of the cells to form the tails
of the spermatozoa produces the stages seen in figs. 31 to 34.
The masses of spermatids are not spherical, but disc-shaped,
their thickness being only about one quarter of their long
diameter. They contain a cavity, the remains of the blasto-
phore, together with a small quantity of a slightly fibrous
coagulum in the centre of the cavity. Curiously enough,
perfectly ripe males were comparatively rare in March and May
of this year, when mature females were abundant. In most
cases the body-cavity was full of spermatids in great bundles,

32 F. W. GAMBLE AND J. H. ASHWORTH.

as in fig. 34, The ripe spermatozoa closely resemble those
of A. Grubii, which have been accurately figured by Claparéde.t
They measure ‘058 mm. in length, and possess a curiously
shaped head,‘004 mm. in length, and an extremely long slender
tail (054 mm. long). The head (figs. 28 and 29) is divisible
into three regions,—a rounded disc-like cap (S.) at the anterior
end, which is partially divided by a median groove ; the nucleus
(N.), which is large and oval in shape; and the “ middle piece ”
(M.), which bears posteriorly a depression into which the tail
is inserted. This depression is formed only at the time when
the spermatozoa are fully ripe. The tail (7.) in the specimens
which we have been able to obtain appeared to be a somewhat
stiff filament, which could only be bent to a comparatively
small extent.

The breeding season of the “‘ Laminarian” variety of Are-
nicola marina lasts from February to May on the Lancashire
coast. The large black ‘‘ worms” which may be dug out
during the great spring tides of these months are then dis-
tended with ova or spermatozoa. Males and females are not
distinguished by external characters, but owing to the slight
discharge of gonads from the nephridiopores consequent on
the tense condition of the body, it is often possible to distin-
guish the sex of an example without dissection. It is at
present impossible to state how long these Arenicola live
and how many times they breed. :

The ordinary littoral lugworms of the Lancashire coast and
of the Isle of Man are not mature in the spring, and contain
at most a few very small eggs. In the summer (August) of
1896 we found mature specimens, and we believe that this
variety breeds through the summer, commencing at about the
time when the deeper water form has ceased.

Relation of the Nephridia to the Reproductive
System.—As is well known, the ova and spermatozoa escape
by the nephridiopores, but it does not seem to have been noticed
before, that in both males and females the bladders of the last
five pairs of nephridia are specially enlarged (Pl. 3, fig. 15, B/.),

1 « Annélides de Naples,’ 1868, pl. xix, fig. 2, C.

HABITS AND STRUCTURE OF ARENICOLA MARINA. 33

and contain mature ova or spermatozoa, so that upon irritation
a simultaneous discharge through all these apertures may
occur. In one worm only eight inches in length the bladder
of the nephridium was swollen with ova so as to measure
14 mm. in length and 6 mm. in width. During the discharge
of ova from the female the eggs are caught by the slimy mucus
covering of the body, and, owing to the movements of the
animal, collect in strings round the body. We have not
observed the formation of gelatinous capsules in which the
eggs may be laid, since we have not worked at the oviposition
of this species, about which nothing is at present known. At
certain times of the year, chiefly in the spring, the nets used
by shrimpers on the sandy coast near Lytham are almost
choked by the balls of eggs, each moored by two “ cables ” to
the sand. Whether these eggs belong to Arenicola remains
to be seen, but their form differs from that of Phyllodoce
found so commonly in early spring.

It has generally been assumed that the number of nephridia
and gonads occurring in Arenicola marina is typical or
fairly typical of the genus, and it is usually stated that the
number of both these organs is a small one (five or six),
An investigation of several other species of Arenicola, the
results of which we hope shortly to publish in full, have
shown that A. Grubii and A. Claparédii have five pairs of
nephridia, and apparently the same number of gonads, whereas
A. ecaudata has no less than thirteen pairs of nephridia,
twelve of which bear large and complicated gonads of a size
and complexity which is scarcely equalled by any other Poly-
chet. What relations exist between A. marina and the other
species of the genus cannot be discussed here, but it may be
stated generally that the genus exhibits greater variety in the
development of several systems of organs than has been hitherto
suspected, and that it is no longer possible to exemplify the
characters of Arenicola as a genus by using their particular
grade of development in A. marina as a type.

vol. 41, part 1.—NmW SER. Cc

34 F. W. GAMBLE AND J. H. ASHWORTH.

12. General Summary.

The following is a recapitulation of the new points which we
have found in Arenicola marina.

1. On the Lancashire coast, and probably elsewhere, two
well-marked varieties of Arenicola marina occur, differing,
as the following table shows, in general appearance, in their
habits, in the structure of their gills, and periods of maturity.

Colour.

Name. Habitat. ——— Gills. ran
Adult. | Young. ;

“Shore lugs,””/The sandy and muddy/Greenish | Semi- | Moderately} July,
or littoral | shores of bays, estu-/brown or) transpa- developed. | August.
variety, 6—8") aries, and harbours,| reddish |rent, yel- Branches
long, excep- | extending from high| black jlowish or with 3—5
tionally 10” | water mark to and brown | pairs of
sometimes beyond _gill-plumes
low tide level
Burrow U-shaped

Worms,” or/The sandy shore ex- Black orDark red, Very well | January
Laminarian | posedat extreme low very dark opaque developed. | to May.

variety, | spring tides, occa-| brown | Branches
8—15'in | sionally extending with usually
length above this. | about 12
‘Burrow a_ vertical, pairs of di-
shaft | | chotomous-
ly arranged

| plumes

2. The cilia lining the central or gastric region of the
alimentary canal are specially arranged (1) on the sides of a
ventral groove which is continued to the anus, and (2) on
curved shallow grooves running downwards and backwards
into the former. The current caused by the action of these
cilia carries a stream of mucus and of digested food slowly
backwards and away from contact with the mass of sand in
the gut. As these grooves are in close connection with parts
of the visceral plexus, absorption may take place from them.

While the ventral groove is morphologically equivalent to the
similar structure of Oligognathus (described by Spengel?),

1“ QOligonathus Bonellie,” ‘ Mitt. Zool. Stat. Neapel,’ iii, 1882.

HABITS AND STRUOTURE OF ARENICOLA MARINA. 35

and probably to the ‘‘ siphon ” of Capitellids, we have seen no
reason for regarding it or any other part of the alimentary
canal as “‘ respiratory ”’ in function.

3. In the circulatory system the two hearts each contain a
cardiac body. This structure is composed of masses of granular
and vacuolated cells, projecting into the cavity of each ventricle.
Functionally they may be regarded as glandular valves pre-
venting the reflux of blood into the gastric sinuses. While
previously unknown in Arenicola, the ‘cardiac body” has
been long known in allied genera (Ophelia, Tyrophonia,
Chlorhzema), but as an unpaired structure in the dorsal vessel
(Schaeppi, 1894). Hence, though histologically similar, it is
very doubtful whether the paired structure of Arenicola,
which has no connection with the dorsal vessel directly, is
homologous with the unpaired organ of other Polychets.

Contrary to Wirén (1896), we regard the dorsal vessel as a
distinct structure, the gastric blood-system as a plexus, and
we find that the nephridia and body-wall, as well as the gills,
are well supplied with capillaries.

4. Both the large pinnately-branching, and the smaller
dendritic, types of gill occurin A. marina. The usual state-
ment that the latter type of gill characterises this species, and
that the former type is characteristic of A. cristata, must
therefore be modified.

5. The brain is divided by a narrow cleft throughout the
greater part of its length. The anterior cornua supply the
prostomium, the buccal papille, and give off the cesophageal
nerve-connectives. ‘The middle region of the brain supplies
the upper part of the prostomium, and the posterior cornua
innervate the nuchal organ.

_ In young specimens the almost uniform covering of ganglion-

cells of the brain is in close contact with the peculiar and
complex sensory epithelium of the prostomium, but in old
specimens of the “ Laminarian” variety fibrous outgrowths
from the dorsal and lateral surfaces of the brain scatter this
ganglionated covering.

6. The nuchal organ, though apparently single, shows traces

36 F. W. GAMBLE AND J. H. ASHWORTH.

of a double origin. It is probably an olfactory organ, and is
developed from the posterior region of the prostomium.

7. The otoliths consist of quartz grains surrounded by a deli-
cate chitinoid film, as Ehlers stated. The peculiar commotion
observed in otocysts mounted in sea water was not noticed in
others examined in celomic fluid. Hence the motion is
probably a result of diffusion currents.

8. The first pair of nephridia are in process of reduction.
In the others the form of the funnel at an early stage is
described and figured. In adult examples the terminal
portions of the nephridia act as receptacles for the ripe ova
or spermatozoa.

9. The specific gravity of the coelomic fluid varies slightly,
but is on the average (including the corpuscles) 1:0288, thus
being only very slightly denser than sea water (1°0264).

10. The general analogies of Arenicola with certain other
limnivorous Cheetopods are very striking. With the Sipunculids
the Arenicolidz agree in the chitinous spines tipping the pro-
boscis papille, the buccal papillz, the strong retractors of the
“ proboscis,” the capacious and largely unsegmented ceelom, the
general character of the musculature, the thin-walled looped
alimentary canal with its ciliated ventral groove, the action of
the body-wall in producing waves of coelomic fluid auxiliary to
the process of burrowing and defzecation, and lastly, the pig-
meuted nuchal organ. If we acknowledge the many points
of agreement, which have for the most part arisen indepen-
dently, between these two distantly related families under
similar conditions of life, the true relationship between
Arenicola and other genera of Polychets can only be ascer-
tained by exercising the greatest caution in not confusing
convergent adaptational characters with true genetic resem-
blances.

1838.

1880.
1881.
1882.

1883.

1883.

1887.
1887.

1887.
1888.

1889.
1891.
1891.
1892.

HABITS AND STRUCTURE OF ARENICOLA MARINA. 37

13. LITERATURE.
Mitne-Epwarps, A.—“ Recherches sur le circulation des Annélides,”’
* Ann. Sci. Nat.,’ sér. 2, t. x, 1838.
Cosmovici.— Arch. Zool. Expt.,’ vol. viii, 1879-80.
VeRRILL, A. H.—‘ Trans. Connecticut Academy,’ vol. iv, pt. 2, 1881.

KRuKENBERG, C. F. W.—‘ Vergleich. Anat. Studien,’ Zweite Reihe,
Zweite Abtheil., Heidelberg, 1882, p. 87.

Wrireiw.— Chetopoder friin Sibiriska,” ‘ Vega-expeditionens Vetensk
Takttag,’ Bd. ii, 1883.

Levinsen.—“ Systematisk-geograph. Oversigt over de Nordiska Annu-
lata,” ‘ Vidensk. Meddels. Kjobenhavn,’ 1882-3.

CunnINGHAM.—‘ Quart. Journ. Micros. Sci.,’ xxviii, 1887-8, p. 239.

MaRENZELLER.—‘ Zoologische Jahrbiicher,’ Abth. f. Systematik, Bd. iii,
1887, p. 12.

Winty.—‘ Kong]. Vetenskamps-Akad. Handlinger,’ 1886-7.

CunNINGHAM AND Ramacr.—‘ Trans. Roy. Soc. Edinburgh,’ vol.
xxxill, 1888.

MacMuny.—‘ Quart. Journ. Micros. Sci.,’ xxx, 1889-90, p. 74.
Cutnwot.—‘ Arch. Zool. Expt.,’ sér. 2, vol. ix, 1891.
Davison.—‘ Geol. Magazine,’ vol. viii, 1891, p. 489.
Euters.—‘ Zeit. f. wiss. Zool.,’ vol. liti, Suppl., 1892.

18924, Huters.—‘ Gottingen Nachrichten,’ No. 12, July 27th, 1892.

1893.
1894.
1896.
1896.

Brenuam.—‘ Journ. Marine Biol. Assoc.,’ N.S., vol. iii, p. 48.
ScHaEPri.— Jenaische Zeitschr.,’ Bd. xxviii, 1894, p. 247.
Racovitza.—‘ Arch. Zool. Expt.,’ sér. 3, vol. iv, 1896.
Kytz.—‘ Annals and Mag. Nat. Hist.,’ ser. 6, vol. xviii, 1896.

38 F. W. GAMBLE AND J. H. ASHWORTH.

EXPLANATION OF PLATES 1—5,

Illustrating Mr. F. W. Gamble’s and Mr. J. H. Ashworth’s
paper on “The Habits and Structure of Arenicola
marina.”

List or REFERENCE LETTERS.

A. Cr, Anterior cornua of the brain. dz. Anus. Az, “ Auricle’”’ of the
heart. B/7. Bladder or terminal part of the nephridia. Blph. Blastophore of
spermatoblast. 8B. Pap. Papille of the buccal mass. Br. Gills. Br. Af.
Branchial afferent vessels. Br. Hf. Branchial efferent vessels. BR. Brain.
B.S. Blood spaces in the heart. 2B. Sf. Sheath of retractor muscle en-
closing the buccal mass. Bucc. M. Buccal mass. Card. B. Cardiac body.
Chi. Tiss. Chlorogogenous tissue on the stomach and ventral vessel. Chu.
Notopodial chete. C.F. Cardiac fibres. C.Sp. Caudal septa. D. L. V.
Dorsal longitudinal vessel. D.Nph. Dorsal lip of the nephrostome. Dp. Ph.
Diaphragmatic pouch. Dphm. }~* Diaphragms or anterior septa. Zp. Kpi-
dermis. G. Refringent granules in ccelomic cells. Ga. Ganglion-cells of
the brain. Gast. Lat. Lateral gastric vessel. Gast. V. Gastric vessels.
G. F. “Giant fibres.” Gl. Op. Opening of the esophageal glands into the
cesophagus. G.S. Granular cells of the heart. Gz. Ventral groove of ali-
mentary canal. G.V. Gonidial vessel. Int. V. Intestinal vessels. MV.
“Middle piece” of spermatozoon. JM. Circ. Circular muscles. Mes. D. and
Mes. V. Mesenteries supporting the dorsal and ventral vessels between
the first and second diaphragms. M27. “ Metastomium,” or achetous
portion of the body immediately following the prostomium. J. Long. Longi-
tudinal muscles. I/O. Mouth. J. Od. Oblique muscles. MM. Pav. Parapodial
muscles. Jf, Pr. Retractors of the prostomium. WV. Nucleus. WV. 4f-
Afferent vessel to the nephridia. 4. C. Ventral nerve-cord. V. Cap. Ne-
phridial capillaries. WV. Hf Efferent vessel from the nephridia. NWZV. Ne-
phridial longitudinal vessel. Nm.1!~!° Neuropodia. No. !-* Nephridiopores.
NPH.** Nephridia. MPHM. Nephrostomes. WS. Nervous elements and
connective tissue round otocyst. Nw. Nuchal organ. Ww. Tr. Retractor
muscle of nuchal organ. JV. V. Neural vessels. Oe. @isophagus. Oc. Comm.
Circumcesophageal nerve-connectives. Oe. G/, Hsophageal glands. Oe. Gi. V.
Vessel of cesophageal glands. Oe. Lat. Lateral cesophageal vessel. O. Of.
External opening of otocyst. O7' Otocysts. OZ", Neck of otocyst. Odh.
Otolith. OZ. N. Nerve to otocyst. Par. V. Parietal vessels. P. Cr. Posterior
cornua of brain. Px. Protractor of notopodium. Pr. Prostomial lobes.
Rn. Retractor of notopodium. §. Cap of spermatozoon, 8S. VY. Sub-
intestinal vessels, 7, Tail of spermatozoon, V. Ventricle of the heart,

HABITS AND STRUCTURE OF ARENICOLA MARINA. 39

Vac. Vacuole. Vac. C. Vacuolated cells of the heart. V. Nph. Ventral lip

of nephrostome. V.V. Ventral vessel. J. 7. ITI. IV. &c. Somites beginning
with the first chetigerous.

PLATE 1.

Fic. 1.—The anterior end of a large specimen of the “ Laminarian ”
variety seen from the left side, to show the external features, the seg-
mentation of the body-wall in relation to the internal metamerism, the
nephridial apertures, and the commencement of the branchial region. The
acheetous region following the fully everted buccal mass (Buce. MW.) extends
forwards as far as the groove indicating the insertion of the first diaphragm
dorsally (Dphm.!). We have considered the first chetigerous annulus and the
annulus behind this, as composing the first chetigerous somite (JZ), although
we are fully aware that, owing to the obliquity of the first diaphragm, and the
absence of landmarks in the achetous region in front of this septum, it is
somewhat hazardous to delimit this first chetigerous somite. x 4%.

Fig. 2.—View from the right side of two somites from the anterior part of
the branchial region of a specimen of the ‘‘ Laminarian ”’ variety 7 inches long.
The fourth gill is shown in detail, while the third and fifth are cut down to
the base of the main branches. The large size of the spreading branches
and the somewhat pinnate arrangement of the lateral twigs distinguish the
gill of this variety of A. marina from that of the ordinary shore lugworm
seen in figs. 8 and 4. The webbing at the bases of the branches is generally
much more marked in old black examples than in immature dark red
specimens such as the present. x 14.

Fic. 3.—Fifth gill of the right side of a shore lugworm 8 inches long, to
show the features characteristic ef the littoral variety of Arenicola
marina. The branches are united by extensive connecting membranes,
between which the blood-vessels of the gill are faintly visible. x 14.

Fic. 4.—The first gill of the right side from the same specimen as Fig. 38.
The ventral branches are apparently the last to develop, and are only just
budding off the secondary leaflets. x 14.

PLATE 2.

Fic. 5.—Dissection of a large “ Laminarian” variety, to show the general
characters of the internal anatomy (conf. pp. 9 to 10). The body-wall has
been cut along the mid-dorsal line, the flaps pinned back, and the alimentary
canal turned over to the left side. The special features shown are the
vascular system, the nephridia, the septa, and muscles. xX 2.

40 F. W. GAMBLE AND J. H. ASHWORTH.

PLATE 3.

Fic. 6.—View of a vertical longitudinal section of Arenicola marina
taken somewhat to the left of the middle line. The thickness of the body-
wall is exaggerated. The stomach has been cut away behind the heart,
to show the oblique muscles and the second nephridium. The main blood-
vessels only are indicated, the object of the figure being to show the exact
position of the three diaphragms (Dphm. 1%), of the buccal or proboscidal
sheath (B.Sh.), and the relations of these to the external segmentation. x 3.

Fie. 7.—Chitinoid spines covering the buccal papille of that part of the
proboscis which is first protruded during eversion. ‘They may be compared
with the figures of ‘ hooks” from the proboscis of Sipunculids (e.g. Phasco-
lion) shown in Selenka, ‘Die Sipunculiden.’ Caustic potash preparation.
x 50.

Fie, 8.—Papille in situ on the base of the proboscis of young worm.
x 6.

Fie. 9.—A gronp of neuropodial sete from a very young Arenicola
marina 16mm. long. ‘The shape and strongly-toothed ridge distinguish
these sete from those of the adult (figs. 1l and 12). ‘The youngest sete are
on the left side of the figure. x 300.

Fie. 10.—Notopodial seta 6 mm. long. x 16.

Fic. 10a.—The tip magnified. x 50.

Fre. 108.—The toothing on the notopodial seta highly magnified. x 450.

Fic. 11.—Neuropodial seta (x 20), and enlarged (x 120).

Fic. 12.—A group of developing neuropodial sete in situ in the neuro-
podium (Vm.) of a “ Laminarian” specimen. x 70. Proc. is referred to on
De ws

Fic. 13.—The fourth and fifth chetigerous segments of the left side of
a large mature “‘ Laminarian ” specimen. The first two nephridia are shown.
The figure is a study of the blood-vessels of the nerve-cord, of the oblique
muscles, and of the connection between the nephrostomial and the dorsal
longitudinal vessels (D. Z.V.). xX 3%.

Fie. 14.—The first nephridium from the specimen shown in fig. 13,
seen from the dorsal surface, to show the gonidial vessel (G. V'.) bearing
blind, vascular processes. The gonidial vessel on this nephridium is sterile.
x 4.

Fic. 15.—Fifth right nephridium of an adult male, to show the bladder
distended with spermatozoa. The nephrostome is widely open. Seen on
February 24th, 1897. x 4.

Fic, 16.—The second nephridium of the right side of a specimen 29°5 mm.

HABITS AND STRUCTURE OF ARENICOLA MARINA. 41

long, seen from the dorsal surface, to show the gonidial vessel (G. V.), the
commencing processes of the dorsal lip, and the position of the external
opening (Wo.2) with regard to the neuropodium (Vm*.). The ventral lip
of the nephrostome (V. Vph?.) is seen through the dorsal one. x 60.

Fre. 17.—Funnel of the first left nephridium from the same specimen as
fig. 16, seen from the right side, to show the vertical position of the nephro-
stome and the commencing processes on the anterior lip. x 90.

PLATE 4.

Fic. 18.—The second right nephridium from a specimen 44 mm. long.
Dorsal view, to show the remarkably complete capillary circulation and the
extension of the vessel of the dorsal lip to form the gonidial vessel (@. V.).
The ventral lip (V. WpA?.) is seen by transparency. X 65.

Fic. 19.—Three views of the anterior end of a specimen 8 inches long
(littoral variety), to show the prostomium, nuchal organ, openings of the
otocysts, and the secondary annulation of the skin. a. From the left side.
B. From above. c. From below. x 12.

Fre. 20.—Transverse section across the middle of the prostomium to show
the brain, the three prostomial lobes, and the rich blood-supply of this region.
The brain lies in the central prostomial lobe, and its covering of ganglion-
cells is closely applied to the overlying sensory epithelium. The section is
cut across in the region of the posterior cerebral cornua (P. Cr.). X 65.

Fic. 21.—Transverse section of the same series as Fig. 20, across the
nuchal organ, Vw., the hinder cornua of the brain, and one otocyst, with its
contained otoliths. x 65.

Fic. 22.—Transverse section of the body a short distance behind the third
diaphragm at the level of the openings of the esophageal pouches (GZ. Op.).
The external aperture of the second nephridium is shown on the right side.
The subdivision of the body-cavity into three longitudinal portions, and the
structure of the cesophageal pouches, are well seen. X 38.

Fic. 23.—Transverse section of the body in the branchial region at the
level of a parapodium. ‘lhe neuropodium is cut through its entire length on
the left side. On one side of the nerve-cord a retractor muscle from the
notopodium arises, on the other an oblique muscle. The vascular supply of
the body-wall, sete, and gills is well seen. x 38.

PLATE 5.

Fic. 24.—Ameeboid and spindle-shaped cells of the celom. x 1000.

Fig. 25.—Sagittal section of the brain slightly to the left of the middle
line, from a young littoral form about 3 inches long. The mass of ganglion-

42 F. W. GAMBLE AND J. H. ASHWORTH.

and glia-cells underlying the epithelium of the prostomium is distinct ; some
of the cells of the latter are shown bearing sensory processes. The nuchal
organ, Vw., is cut at its full depth. x 85.

Fre. 26.—View of a dissection of the brain, cesophageal connectives, oto-
cysts, and the buccal sheath. The commencement of the neural vessels from
capillaries of the organs just mentioned, is shown. Seen from the dorsal
surface. The buccal mass, cut transversely, lies in the centre of the figure.
x 6.

Fic. 27.—An otocyst with the otoliths composed of quartz grains. The
sensory epithelium and the surrounding nervous and supporting cells are seen.
x 160.

Fig. 28.—Otoliths to show the chitinoid covering of the quartz grains.
x 500.

Fre. 29.—Ripe spermatozoon seen on March 10th, 1897. Length of head
4p, length of tail 54 4. x 3000.

Fic. 294,—Head and portion of tail of an immature spermatozoon seen
on February 22nd, 1897. x. 38000.

Fries. 30—34.—Stages in the development of the spermatozoa.

Fig. 30.—The 8-celled stage, in which the spermatoblasts leave the
tesiis.. =< 500.

Figs. 31 and 32. Later stages. x 500.

Fig. 33.—Two cells from a stage much later than the preceding, showing
the commencement of the tail. x 2000.

Fig. 34.—Discoidal mass of almost ripe spermatozoa. xX 500.

Fic. 35.—Developing ova. (a and 4) show a polar body (?) x 250.
(c) is a ripe ovum enlarged 125 times.

Fic. 36.—Longitudinal section of the heart of an Arenicola ou mm,
in length, to show the cardiac body. x 82.

Fre. 37.—Histology of a portion of the cardiac body of fig. 36. x 500.

Fic. 38.—Longitudinal section of the heart of a young Arenicola 65 mm.
in length, to show the cardiac body at an early stage of development. x 50.

THE ASEPTIC CULTIVATION OF MYCKETOZOA. 43

The Aseptic Cultivation of Mycetozoa.
By

Casper O. Miller, M.D.

With Plates 6 and 7.

OBSERVATIONS ON THE CULTIVATION OF MyYCETOZOA.

Untit the work of de Bary nothing was known about the
development of Mycetozoa further than that they appeared as
a slimy mass from which the sporangia were formed. He made
a short report (1) on the development of the zoospores from
the spores at the “ Naturforscherversammlung,” in Gottingen,
in 1854, which was followed by his other publications (2, 3).
He speaks (8) of keeping portions of plasmodia in glass dishes
containing water, or on slides, but they died in a few days
without forming sporangia. Sporesof Aithalium septicum,
planted on moistened tan on the 2nd of May, showed at the
beginning of July colourless plasmodia, which continued
through July without further development. Another culture
of spores of the same plasmodium, planted the 13th of August,
developed many zoospores, and on the 8th of October plas-
modia were seen. Spores of Lycogala, planted in a dish
containing water and decaying pine-wood, developed zoospores
within twenty-four hours ; about the fourteenth day there were
plasmodia present, which at the end of a fortnight had died
without forming sporangia. He also planted spores of Stemo-
nitis obtusata on decaying pine-wood, and found plasmodia
on the fourteenth day, but they did not develop further. De
Bary was unable to determine whether the plasmodia de-

44, CASPER O. MILLER.

veloped from a single zoospore or by the fusion of a number
of zoospores.

Cienkowski (6 and 7) planted spores of Licea pannorum,
Wallr., on decomposing carrots, and obtained plasmodia.
He also planted the spores in water placed on slides, and saw
the zoospores fuse to form plasmodia. Spores of Physarum
album=Chondrioderma difforme, planted on microscopic-
ally small portions of vegetable fibre, developed plasmodia on
the fourth day, and twenty-four hours later they fructified,
so that under good conditions they completed their cycle of
development in five days.

Lieberktihn (9) described a plasmodium which he found in
the bottom of a glass vessel in which spongillia were being
cultivated,

Cienkowski (16) cultivated Didymium libertianum in
water. In one to two weeks plasmodia appeared in the water
or creeping on the wall of the vessel.

He also found a plasmodium in fresh water containing alge.
He studied it in hanging drop-cultures and on the slide. He
thought it probably was the same species which Lieberkiihn
had studied. Sporangia did not form in any of his cultures.

Stahl (22) cultivated Aithalium septicum on moist tan,
and saw a species of Physarum form small-stalked sporangia
on a filter-paper culture. He did not use any aseptic pre-
cautions, and does not state how long it took the sporangia to
form after planting the spores.

Ward (25) found a plasmodium which formed sporangia on
the roots of hyacinths which he was cultivating in water con-
taining a small percentage of salts of lime, magnesia, potash,
and soda. He then made a decoction of hyacinth roots, which
he boiled and used to make drop-cultures. By planting the
spores he succeeded in getting the zoospores and plasmodia in
drop-cultures and on slides without other forms than bacteria.
The cover-glasses were heated, and the cardboard used in
making the moist chambers was boiled.

Strasburger (26) obtained Chondrioderma diff. by
placing macerated stalks of Vicia faba on moistened filter-

THE ASEPTIC CULTIVATION OF MYCETOZOA, A5

paper under a bell-jar ; the sporangia developed after a few days.
He also made drop-cultures of the spores of Chond. diff.
in a decoction of cabbage-leaves or bean-stalks, leaving frag-
ments of vegetable fibre in the fluid. He heated the cover-
glasses and needle used in making the inoculations, but added
alge and bacteria to the cultures. In many of the cultures
the development did not go further than the formation of
microcysts, but in more favorable cultures plasmodia de-
veloped which fused with each other, and on the fourth or
fifth day they crawled out from beneath the cover-glass and
formed sporangia.

Wingate (82), in describing Enteridium rozeanum,
says that Roze (12) cultivated plasmodia in earthenware dishes
filled with sphagnum and water, into which he thrust dead
branches of trees, pieces of decayed stumps, &c., which were
taken from the neighbourhood of Paris to America. He ob-
tained various plasmodia, and studied them until they formed
sporangia. I have unfortunately not been able to procure the
original work by Roze.

Lister (84) cultivated Chond. diff., and obtained the
sporangia in from ten to fourteen days after planting the
spores. ‘The writer has only seen a short report of the paper
in Just’s ‘ Jahresbericht,? so that he does not know what
methods were employed.

Celakovski (88) used the method which Pfeffer (35) found
useful for obtaining plasmodia. He placed dried stalks of
Vicia faba, or the leaves and stalks of other plants, par-
ticularly of Typha latiflora, in broad crystallising dishes,
poured enough water in the dishes to cover the greater portion
of the nutrient material, covering the dishes with suitable lids,
and sterilised them at a boiling temperature. He then planted
spores of Chondrioderma diff. and Didymium macro-
carpon. In from six to fourteen days plasmodia of the former
were found in the cultures. He frequently obtained the two
plasmodia together by simply moistening the stalks of Vicia
faba, and placing them in a covered dish. By repeatedly
transplanting he obtained the Didymium alone, without the

46 CASPER O. MILLER.

Chond. diff. He fails to mention how long the interval was
between the planting of the spores and the formation of the
sporangia. He also planted the spores of Arcyria punicea,
Pers., Trichia nutans, Libert, and Stemonitis dictyo-
spora, Rostaf., on sterilised decayed beech-wood in flat
crystallising dishes, containing water to the depth of ‘5 cm.
He did not see the plasmodia of the first two in the water;
they developed in the interior of the wood, and only appeared
on the surface when the sporangia were formed. The Ste-
monitis developed plasmodia in the water, and fourteen days
after their first appearance the sporangia were formed. He
fails to state how long it took for the plasmodia to develop
after the spores were planted.

Although Celakovski sterilised his nutrient media and the
vessels, no observations were made as to the presence or
absence of contamination. It is very difficult to prevent the
contamination of cultures in a large flat dish, when the lid is
removed or lifted for the purpose of examining the culture.

My cultures were first made as controls for another series
of experiments, but the results seem of sufficient interest to
publish as a separate paper.

Ture MertTHopS EMPLOYED.

In the summer of 1890, while making some experiments at
the pathological laboratory of the Johns Hopkins Hospital, to
determine what Protozoa one finds in the air, a number of
flasks, containing sterilised water with 2 per cent. of milk
added, were left uncorked for a number of days. Of the
flasks one showed zoospores of Mycetozoa, which were trans-
planted a number of times. The zoospores and plasmodia
developed, but no sporangia appeared.

The first systematic attempt of the writer to cultivate Myce-
tozoa was made at the Zoological laboratory at Heidelberg
in 1893.

A culture was prepared in the laboratory for the study of
Infusoria, by simply placing unsterilised hay in a glass jar with

THE ASEPTIC CULTIVATION OF MYCETOZOA. 4.7

unsterilised hydrant water, the jar being covered by a glass
plate. On examining the culture ten or twelve days later
zoospores of plasmodia were found in the water, and sporangia
of plasmodia developed on the hay a few days later.

A number of similar cultures without aseptic precaution
were then made of hay gotten from different sources, and they
all showed the presence of plasmodia.

Next a series of cultures were made in tall narrow beakers,
they being first closed with a large plug of cotton and sterilised
in a hot-air steriliser. The beakers were then filled about
half full with unsterilised hay. Care was taken to first wash
the hands and sterilise the scissors, so as to be moderately
certain that no spores of plasmodia were introduced from the
hands or instruments. Water which had been sterilised in
flasks was then poured into the beakers, until most of the hay
was submerged, care being taken not to cover it completely.

In a few days the hay projecting from the surface of the
water was covered with mould fungi. A pair of sterilised
forceps was then used to remove the stalks of hay covered by
the fungi, care being used to loosen up the hay so as to have
some of it projecting above the water. If the hay is entirely
submerged plasmodia may not develop, but when prepared as
above, all of the cultures prepared with hay, whether gotten in
Heidelberg or Baltimore, developed plasmodia. It would
appear that plasmodia are constantly present on hay in one
form or another.

Cultures prepared in the same way with the stalks of wild
carrot picked out from the hay did not develop plasmodia.

A series of cultures were made by putting dried chestnut
and oak leaves in sterilised Erlenmeyer flasks with sterilised
hydrant water. In a number of these cultures plasmodia de-
veloped. —

Elsewhere (45) the writer has described the aseptic methods
employed in the cultivation of Protozoa, but for Mycetozoa
some modifications are necessary. They will grow in sterilised
dilute hay infusion, or 2 per cent. of milk in water, but for the
formation of sporangia it is in general advantageous, and for

48 CASPER 0. MILLER.

some forms essential, to furnish them a mechanical support
as a means of getting out of the water.

The medium which has proven the most generally useful is
prepared as follows. A handful of hay is placed in a jar and
washed repeatedly until the water remains colourless. It is
then covered with fresh water and allowed to soak overnight.
The following day the water is poured off, filtered, diluted with
fresh water until it is of a white-wine colour, and 2 per cent.
of milk is added to the infusion. It is then filtered, put into
a flask, and sterilised for future use. The macerated hay is cut
and placed in Erlenmeyer flasks ; the first portion is cut short
enough so as to form a tolerably compact layer in the bottom
of the flask to the depth of 1 cm.; the rest is cut sufficiently
long to form a very loose layer reaching about two thirds the
way up the sides of the flask, care being taken not to allow any
of the stems to reach the cotton. Sufficient water is placed in
the flasks to cover the hay, and they are sterilised for fifteen
minutes. On the following day fresh water is substituted, and
they are again sterilised. The water is once more poured off,
and enough of the hay infusion and milk previously prepared
is added until it is about 1 cm. deep. ‘The flasks are then
sterilised in a steam steriliser for ten minutes on three suc-
cessive days. They are then ready for use.

After soaking the hay for twenty-four hours in water, and
boiling it several times in fresh water, about all of the soluble
substance has been extracted, and the diluted hay infusion
with 2 per cent. of milk is added; we thus have a medium of
tolerably uniform composition.

Of the cultures gotten from the air several contained mould
fungi, which were eliminated by putting the cultures in the
oven at a temperature of 37° C.

One culture contained chroococci, and these were eliminated
by keeping a series of cultures in a dark closet. It is not
possible in every case to eliminate other protozoic forms that
may be present, but one may at times succeed by taking ad-
vantage of the fact that the encysted forms withstand drying.
In this way one may sometimes succeed in separating Myce-

THE ASEPTIC CULTIVATION OF MYCETOZOA. 49

tozoa from the Infusoria, Amcebe, and other protozoic forms
found in hay infusions.

The cultures are usually transplanted by means of a sterilised
pipette.

Bacteria are found in all the cultures, and studies have been
made with the view of finding out what effect bacteria have on
the growth of Mycetozoa, and what bacteria, if any, are more
favorable to their growth.

It is not the writer’s purpose to discuss the influence of the
bacteria in this connection, but he will leave it for a future
communication.

THe Mycrtozoa cULTIVATED.
Physarum cinereum.

This was the first plasmodium from the air which was culti-
vated. It will grow and form plasmodia in water with 2 per
cent. milk or in dilute hay infusion. The best cultures are
obtained when the hay also is present as described above.

In all the cultures where sporangia are formed, the plas-
modia grew in the fluid and crawled on the side of the flask
above the fluid preparatory to the formation of the sporangia.
Although the largest plasmodia form in cultures containing
hay, yet the sporangia only form on the glass.

The plasmodia spread out on the glass in the form of a
yellowish-white network, consisting of primary trunks from
which run branches anastomosing with each other, the net-
work becoming finer as the periphery is approached. At the
periphery there is a more or less flattened perforated proto-
plasmic plate with a scalloped border. In the cultures not
containing hay the principal trunks extend to the water; in
cultures containing hay the plasmodia spread out from stems
of hay leaning against the side of the flask (fig. 2), and it
cannot be determined whether branches extend to the water.

In the more vigorous cultures the plasmodia are large
enough to cover the whole inner surface of the flask above the
water, but do not pass to the cotton plug.

voL, 41, part 1.—NEW SERIES. D

50 CASPER O. MILLER.

After remaining on the glass above the water for from two
to twelve days, the protoplasm collects at one or a number of
points at the periphery of the network, and forms sporangia,
leaving behind a so-called hypothallus, retaining the shape and
outlines of the original network, but much paler in appear-
ance. The sporangia vary in number according to the size
and vigour of the plasmodia. In one culture there were only
two sporangia; in other cultures the sporangia form groups,
the larger of which may contain from seventy to eighty spo-
rangia. In the first stage of the formation of the sporangia
the protoplasm is of a more yellow colour than that of the net-
work. As the sporangia assume their completed shape the
colour becomes a brownish red, which changes to a greyish
white when the development is completed.

The sporangia are sessile, resting on a broad base. When
isolated they are round, oval, or kidney-shaped. At times they
are united, forming a long drawn-out sporangium with con-
strictions at irregular intervals. The small oval or round
sporangia may measure as little as 0°5 mm. in diameter, the
long drawn-out ones may measure as much as 7 mm. On
examination with the low power by reflected light the surface
shows irregularly shaped small white elevations, between which
are darker areas. Under the high power these white areas
are seen to consist of aggregations of coarse granules, which
dissolve on the addition of hydrochloric acid with the forma-
tion of gas bubbles. The sporangia have no columella, and
the sporangium wall is colourless. The capillitium is made up
of a network of thin, colourless fibres attached to the wall of
the sporangium. At the point of communication of the fibres
there is a more or less flattened triangular or polygonal thick-
ening, containing granules of lime. The spores are smooth
and of a brownish-violet colour, measuring 8°5—13°5 w in
diameter. The majority of the spores are spherical, but occa-
sionally there are oval or irregular forms. From a study of
the structure and the arrangement of the sporangia of this
plasmodium, it would appear that it is identical with Phy-
sarum cinereum, Pers.

THE ASEPTIC CULTIVATION OF MYCETOZOA. 51

Stemonitis.

In July, 1892, another series of flasks containing sterilised
milk, 2 per cent.in water, was exposed for a month to the air;
they were then closed with cotton and examined. In three of
the flasks were flagellate bodies which the writer thought
corresponded to the description given by Bitschli (29) of
Mastigameba. From these flasks cultures were made, and in
one of the transplantations plasmodia developed. At that
time the writer had not studied plasmodia sufficiently to
recognise the relationship between the zoospores and the
plasmodia, and inasmuch as there were similar flagellates in
all three flasks, he concluded that the plasmodia and the
flagellates were independent forms. Nothing further was
done with the cultures until July, 1893, when they were again
transplanted, and plasmodia developed in all three cultures.
At that time they were cultivated with 2 per cent. of milk in
water, and in hay infusion. The zoospores and the plasmodia
grew, but there was no formation of sporangia. The cultures
were then made in flasks containing hay, with the idea that
the plasmodia would be enabled to get out of the water to
form the sporangia. Upon placing hay in the flasks a number
of cultures formed sporangia.

All three plasmodia belong to the genus Stemonitis.

From a study of the sporangia there is no difficulty in
deciding that two of them are distinct, while it is not so evi-
dent that the third one differs from one of the others, but the
writer is inclined to the opinion that it is also a distinct
species. The writer is not familiar with the American My-
cetozoa, and has not been able to get all of the literature on
the subject ; it is possible that they agree with species already
named. It will be necessary to refer to each of these cultures,
and it will be more convenient simply to designate them as
Stemonitis A, B, and C.

At a variable period after the inoculation of the cultures
there appears, rather suddenly, a large yellowish-white plas-
modium lying on the hay at the surface of the water, which

52 OASPER O. MILLER.

may cover an area measuring about 1 by 2 cm. They are
composed of a network of short, thick, anastomosing branches,
from the periphery of which extend branching, sausage-,
horn-, or club-shaped prolongations. There is not much
change in the appearance of the plasmodium for forty-eight
hours ; during this time it may change its location on the hay,
but the motion is a slow one. At the end of this time the
motion becomes more rapid. The plasmodium moves some
distance from the surface of the water, and settles upon the hay
or on the glass. In one of the cultures the whole plasmodium
had moved 6 cm.in four hours. When it has found a suitable
place the peripheral prolongations are drawn in, there is no
longer any evidence of the presence of a network; it then
appears as an oval or rounded, conical or flat, yellowish-white
mass, the surface of which is covered by a number of closely
crowded small hemispherical prominences. From each of
these prominences is formed a cylindrical sporangium. Soon
after the sporangia assume their permanent form, the yellow-
ish-white colour begins at the base to change to a reddish
colour, which gradually ascends to the apex, and finally
becomes a reddish or dark brown colour.

It takes from twelve to eighteen hours from the time the
plasmodium leaves the water until the sporangia are fully
developed.

The well-developed sporangia are cylindrical, closely
crowded, and placed more or less perpendicular to the mem-
branous hypothallus, from which extends a branch going to the
surface of the water, indicating the route which the plasmo-
dium took. When the sporangia are formed on the glass the
plasmodia take an oblique course up the side of the glass.

There are slight differences in the appearance of the plas-
modia of the three cultures, but the difference is not alone
sufficient to enable one to say that they are distinct species.
The peripheral prolongations of Stemonitis B are usually
longer and thicker than Stemonitis A. The network of
Stemonitis C is a more open one than that of the other
two.

THE ASEPTIC CULTIVATION OF MYOETOZOA. 53

In some cultures the sporangia are imperfectly developed.

A typical sporangium of Stemonitis A measures about
3 mm. in height, including the stalk, which is 0:167mm. The
diameter of the sporangium is about 0°3 mm., and is usually
uniform throughout. The sporangium may be thicker toward
the apex or base, The apex is usually rounded, but at times
is more acute ; the base may or may not be symmetrical. The
measurement of the stalk given above is about the average,
and applies to fig. 9. In a few instances the capillitium
extends to the hypothallus; in other instances the stalk may
be 0°5 mm. long. The columella tapers gradually from the
base to near the apex, where it divides into several branches,
becoming continuous with the capillitium. Occasionally one
finds a spindle-shaped thickening of the columella. The
primary branches of the capillitium usually come off at an
acute angle from the columella, forming one series of anasto-
moses, and then divide into smailer branches, which go
obliquely to the surface network. The surface network usually
extends over the entire sporangium. ‘The meshes of the net-
work average from 8 to 33. On the surface network are
distributed small wart-like thickenings. The colour of the
capillitium is a brownish violet. The spores measure 7—18 1,
and are of a violet-brown colour; the membrane is finely warted.

The sporangia of Stemonitis B (fig. 10) measures 3°5—
3°83 mm. in height, not including the stalk, which is about
1°37 mm. long. They are tolerably uniform in thickness,
measuring about 0°27 mm. in diameter. The columella
tapers gradually from the base to near the apex, dividing into
branches which become continuous with the capillitium. The
capillitium fibres come off at right angles to the columella,
forming one series of anastomosing arches from which pass
out secondary fibres placed perpendicular to the surface ; they
break up into branches which become continuous with the
surface network. The capillitium is of a dark violet-brown
colour. At the point where the primary fibres anastomose
one frequently finds membranous expansions which are more
marked than in the sporangia of Stemonitis A, but these

54 CASPER O. MILLER.

membranous expansions vary a good deal in sporangia from
the same culture. The spores are of a light violet colour ; they
have a smooth membrane, and are tolerably uniform in diameter,
measuring 7°9—8°5 mw.

The sporangia of Stemonitis C resemble those of Stemo-
nitis B. The sporangia of Stemonitis C measure 3°3—
3'7 mm. in length, and 0°37 mm. in thickness. The columella
is not infrequently bent on itself at about the upper four fifths.
The secondary fibres of the capillitium are longer than in the
sporangia of Stem. B. The stalk measures 0°68—1:16 mm.
in length. The spores are smooth, of a brownish-violet colour,
measuring 7'4—1] w in diameter.

It would therefore appear that the differences between the
sporangia of Stemonitis B and C are not less than those
which separate some of the forms which are described in
works on the subject under different names. It is possible
that further cultivation may show that they are the same.

Hay anp Lear Cu.ururges.

In cultures made with unsterilised hay in jars without
aseptic precautions, or in flasks with aseptic precautions,
one finds bacteria, fungi, monadina, infusoria, and plasmodia
developing with uniform regularity. Chondrioderma dif-
forme and some species of Didymium, usually micro-
car pon, appear together orsingly, the Chondrioderma being
most frequently present. As has been stated before, some plas-
modium appears in every culture made with unsterilised hay.

By the drying method the Chondrioderma diff. and
Didymium microcarpon have been separated and culti-
vated aseptically in flasks. They both form sporangia on the
hay, and on the glass above the hay.

In a culture of Chond. diff. made in dilute hay infusion
with 2 per cent. milk added, which had been kept in the dark
for several weeks and then placed in the light, sporangia
formed under the surface of the water. The sporangia were
small, round, or pear-shaped, and did not show the presence of

THE ASEPTIC CULTIVATION OF MYCETOZOA. 55

any granules of lime in the sporangium wall. In all the other
cultures observed the sporangia formed on the hay or on the
sides of the flask above the level of the fluid.

In speaking of the classification of his plasmodium, Ward
(25) says, ‘It is, indeed, not improbable that we have here
an aquatic form of Didymium difforme, one of the com-
monest of our Myxomycetes; and if so, we have another proof
of the all but uselessness of attempting to classify the lower
organisms until we know more of their habits under vary-
ing conditions.” From the writer’s experience, he questions
whether Ward did not really have the ordinary form of Didy-
mium diff.= Chondrioderma diff. in his cultures, and
whether the character of the fluid in which they grew, and
the other conditions surrounding them, did not cause the
sporangia to form only on the roots under the water or on the
moist roots above the water.

Didymium farinaceum was obtained from a culture made
with unsterilised leaves taken from the forest.

In one flask containing leaves, and in two containing pine-
needles, plasmodia developed and formed sclerotia above the
water on the side of the flask, but no sporangia appeared, so
it was not possible to determine what species they were.

Spores of Aithalium septicum obtained from a tan-pile
were planted in flasks, and yellowish plasmodia developed,
but no sporangia formed. Spores from several varieties of
Stemonitis collected at Heidelberg were planted in flasks.
The zoospores and plasmodia developed, but only one of them
formed sporangia.

Spores of Ceratium porioides, gotten from a pine stump,
dried and planted aseptically, developed zoospores which have
been cultivated for about four years, and as yet the writer has
failed to find any plasmodia or sporangia. So far as I have
been able to discover, no one has succeeded in cultivating
. plasmodia of any of the Ceratiomyxa.

56 CASPER O. MILLER.

The Time of the Appearance of the Large Plasmodia,
and of the Formation of the Sporangia.

Plasmodia, as we usually find them in nature, appear rather
suddenly on decaying wood, tan, or leaves, and within a short
time they form sporangia. We know little about their pre-
vious growth.

Some Mycetozoa may form sporangia during any of the
warm mouths, while, according to de Bary (21), others are
characterised by forming sporangia only during a short time
in the year. As has already been mentioned, Cienkowski
and Strasburger obtained sporangia on the fifth day after
planting the spores of Chond. diff., and Lister obtained
them in from ten to fourteen days. Celakovski (88) men-
tions that the sporangia of Stemonitis dictyospora,
Rostaf., developed fourteen days after the appearance of the
plasmodia, but does not state how long it took the plasmodia
to develop.

Rex (36) mentions having seen Stemonitis Bauerlinii
form sporangia op a decayed log in the autumn, and the next
summer the same species formed sporangia three times on
the same log at intervals of a month. One cannot say that
the spores fell back on the log, developed zoospores, and from
these new plasmodia grew and formed sporangia.

In the cultures made with unsterilised hay in water the
conditions are practically the same. The forms of the Myce-
tozoa, whether microcyst, sclerotia, encysted zoospores, or
spores, have been dried for months. The hay is placed in the
water and kept at the room temperature. The sporangia of
Chond. diff. appeared on the hay from the twenty-fourth to
the twenty-ninth day. Crops of sporangia continue to be
formed on the hay every few days for from two to four weeks.

Didymium microcarpon first show sporangia on the hay
from the twenty-first to the twenty-fourth day, and continue to
form sporangia for several weeks.

When there were only a few stalks of hay projecting above
the surface of the water the sporangia appeared, but were less

THE ASEPTIC CULTIVATION OF MYCETOZOA. 57

numerous than when a good many of the stalks projected.
The time from the planting of the cultures until the sporangia
form varies considerably.

Cultures of Stemonitis A, B, and C formed sporangia as
early as the thirtieth, and as late as the seventy-sixth day.
Two cultures made from the same parent culture in the same
media developed sporangia on the thirty-second and seventy-
sixth day respectively.

As a rule but one set of sporangia developed in the same
culture. Sporangia do not develop in all the cultures; at
times large plasmodia form on the hay and degenerate without
forming sporangia.

Physarum cinereum formed sporangia from the twenty-
second to the sixty-fourth day.

Didymium farinaceum formed sporangia on the dried
leaves on the fifty-seventh day.

AEthalium septicum formed large plasmodia about the
fifty-fifth day, remained on the side of the flask for about ten
days, and then degenerated without forming sporangia.

Plasmodia under natural conditions leave their moist or wet
habitat, crawl to the surface when it is dry, aud they are ex-
posed to the light. In some of my cultures the formation of
the sporangia seems to have been delayed by keeping the
culture in the dark; some of the cultures were kept in the
dark six weeks, and after being in the light for several weeks
formed sporangia.

The zoospores develop readily in the oven at 37° C., but no
sporangia formed in any of the cultures. The absence of light
may have had something to do with the result.

Time of the Day at which the Sporangia develop.

De Bary (8) studied the formation of the sporangia of
Physarum sulphureum, Didymium serpula, Mthalium
septicum, and Stemonitis ferruginea, and found that
usually the sporangia began to form in the afternoon or late
evening, and the development was completed in some cases by

58 CASPER O. MILLER.

the next morning ; in others not until near the middle of the
day.

In one observation by the writer, the plasmodium lying on
the hay at the surface of the water began about noon to crawl
up the side of the flask. By 6 p.m. the plasmodium had
collected at the point where the sporangia formed; by 7 p.m.
the branches were drawn in, and the surface was covered by a
number of hemispherical projections; and by 6 a.m. the
following day the sporangia were fully formed. In other
cultures observed the plasmodia were resting at the surface of
the water at 6 p.m.; by 9 o’clock the next morning they were
out of the water, and the sporangia had begun to assume a
cylindrical shape. By 11 a.m. the shape of the sporangia
was fully developed; the colour appeared first in the base
of the columella, gradually going to the apex. By 2 p.m.
the sporangia were of a brownish-red colour except at the
apex, which was yet a yellowish-white on the surface. By
5 p.m. the colour was fully developed and the sporangia were
completed.

The sporangia of Phys. cinereum, so far as observed,
began to be developed at 3—6 p.m., and were completed by
the next morning. The sporangia of Chond. diff., Didym.
microcarpon, and Didym. farinaceum also developed for
the most part at night.

Observations and Speculations concerning the
Formation and Growth of the Plasmodia.

In his first studies De Bary failed to show how the plasmodia
develop, whether by growth from a single zoospore or by the
fusion of a number of zoospores.

Cienkowski (6, 7) described and pictured the fusion of
the zoospores to form small plasmodia, and he saw plasmodia
which had later taken in foreign particles, spores, and micro-
cysts.

De Bary (8, 21) accepted Cienkowski’s results, although
he never saw the zoospores fuse.

Ward (25), in speaking of the fusion of the zoospores to

THE ASEPTIC CULTIVATION OF MYCETOZOA. 59

form plasmodia, says, ‘‘ The inference becomes almost a cer-
tainty after watching the specimens under cultivation ;” but
he did not actually see them fuse.

Strasburger (26) also describes the fusion of the zoospores
to form Myxameeba.

The writer has not been fortunate enough to observe the
fusion of the zoospores, but the accuracy of the observations
of such competent observers as Cienkowski, Strasburger,
Lister, and others can hardly be doubted. In the cultures,
as the writer has studied them, however, he does question
whether the fusion of the zoospores is the chief mode by which
the plasmodia grow.

If a few drops of a culture containing microcysts of Stemo-
nitis, with suitable bacteria, be inoculated in a flask con-
taining sterilised water, with milk 2 per cent., the bacteria
multiply at the expense of the milk. Within two or three
days the fluid loses the slight opalescent appearance which it
had, and on microscopic examination there are no longer milk
globules present. I think, from our knowledge of bacteria,
we can conclude that at least a portion of the milk has been
consumed by them. During this time the zoospores have
nultiplied by division; they feed on the bacteria, and possibly
some elements of the milk which the bacteria may not have
appropriated. In a few days the zoospores begin to encyst,
and by the end of the second week the majority of the zoospores
are encysted, while a smaller number remain active. If control
cultures are made from the flask, it will be found that there are
not near so many bacteria present as there would be in a flask
containing a similar medium inoculated with the bacteria
alone which grow with the zoospores. In from ten to fourteen
days small plasmodia may appear; they increase in numbers
and in size, and later large plasmodia are present.

In cultures made in flasks containing hay, with milk 2 per
cent.in hay infusion, essentially the same changes take place, but
the hay interferes somewhat with the examination. If examined
about the end of the second week one finds bacteria, encysted
zoospores, active zoospores, and a few small plasmodia. The

60 CASPER O. MILLER.

plasmodia increase’ in number and in size, but they are not seen
macroscopically, If the culture be one which forms sporangia
on the thirtieth day, and it is examined about the twenty-sixth
day, one finds more small plasmodia and a smaller number of
microcysts present in the fluid than at the previous examina-
tious. <A large plasmodium appears, rather suddenly, on the
twenty-eighth day, lying on the hay at the surface of the fluid.
It does not increase noticeably in size for two days, and then
passes up the side of the flask to form sporangia. If the fluid
is examined the small plasmodia have disappeared for the most
part from the fluid.

One may have exainined the culture the previous day without
having macroscopically observed the presence of a plasmodium.
It must have originated by the fusion of a number of small
plasmodia, or have grown as a large plasmodium in the interior
of the stalks of hay. One stalk of hay is not large enough to
accommodate the plasmodium, and no branches of plasmodia
are seen connecting the various stalks. The writer is of
the opinion that it originated by the fusion of a number
of small plasmodia. Plasmodia large enough to be seen macro-
scopically have been observed by the writer to fuse on the
slide.

What takes place in the culture seems to be as follows :—
the bacteria multiply at the expense of a portion of the nutrient
material: the zoospores multiply at the expense of the bacteria,
and possibly some nutrient material which was not consumed
by the bacteria; the majority of the zoospores encyst; small
plasmodia develop from a single zoospore or by the fusion of
several zoopores ; the plasmodia take in and digest active
and excysted zoospores and bacteria; finally, the small plas-
modia fuse to form the large plasmodium.

Celakovski (38) studied the action of the plasmodia Chond.
diff., Didymium microcarpon, and Aithalium septicum
on various substances placed in the fluid with them. He saw
them take in microcysts which, after ingestion, were not found
in vacuoles, but were simply surrounded by protoplasm. After
two days the microcysts were expelled unchanged ; if dried

THE ASEPTIC CULTIVATION OF MYCETOZOA. 61

and again moistened they gave origin to active zoospores. He
thus reached the conclusion that the plasmodia did not digest
the microcysts.

It is well to consider the condition under which he placed
the plasmodia. He removed them from the fluid to which they
were accustomed, washed them in fresh water, and placed the
microcysts, spores, &c., on or near the plasmodia. In some
instances he washed them several times. To the writer this
seems harsh treatment. It cannot be wondered at that they
were not in a condition to digest foreign substances, ana that
under normal conditions he got peculiar results.

The writer has observed living plasmodia which had taken
in microcysts and rounded off zoospores which had not yet
formed a cyst wall. In these instances the zoospores were
lying in vacuoles. Plasmodia placed on slides under a
cover-glass (with active and encysted zoospores in the same fluid
in which they grew) and allowed to spread out were killed with
picric and acetic acid, and stained with picro-carmine. One
finds in such specimens microcysts and rounded-off zoospores
lying in vacuoles. They are in various stages of degenera-
tion, and stained with varying degrees of intensity. From a
study of the specimens the writer does not see how one can
reach any other conclusion but that the microcysts and
zoospores are digested. If one examines a culture after
having developed sporangia, there are a smaller number of
microcysts present than there was some time previously.

If one places a few drops of a culture containing zoospores
of Stemonitis on a slide under a cover-glass, placing it in a
moist chamber for somé hours, on examination he will find
many of the zoospores creeping around on the slide, feeding
on the bacteria. If, now, a point is examined at one side of
the cover-glass, and a drop of sterilised water be added to the
culture at the opposite side of the cover-glass, it will be
observed that the zoospores instantly draw themselves
together, many of the vacuoles will disappear, and the bacteria
or undigested granules which were in the vacuoles will appear
as granular particles enclosed in the protoplasm. It will be

62 CASPER O. MILLER.

some minutes before the vacuoles reappear and the zoospores
begin to feed again.

It is not an infrequent experience that some protozoic forms
are killed by simply placing them in fresh water. The plas-
modia may not be as sensitive as the zoospores. They may
have sufficient vitality not to be killed by the treatment to
which Celakovski subjected them, but the writer questions
whether the results obtained under the conditions in his
experiment can be used as a basis for conclusions as to what
plasmodia do in the fluids in which they thrive. His studies
also showed that the plasmodia did not digest the encysted or
active Colpoda which they had ingested. By the study of
plasmodia taken from hay cultures and placed on a slide with
a few drops of fluid containing Colpoda, the writer obtained
results which showed that they do digest Colpoda.

The observations of Lister (44) also show that plasmodia do
take in microcysts, enclose them in vacuoles, and digest them.

Observations on the Ingestion of Foreign Substances
and the Multiplication of the Zoospores.

Lister (80) described the taking up of bacteria by the
zoospores of Stemonitis fusca and Trichia fallax. The
bacteria were drawn in by pseudopodial prolongations, which
always came off from the posterior extremity, and were
carried to vacuoles near the nucleus, where they remained
until digested. He also gave them particles of carmine, which
were ingested but not digested.

The zoospores of Stemonitis A, B, or C have not been
observed to put out pseudopodial prolongations and draw in food
particles, but the writer has seen the zoospores of Stemonitis
A, B, C, and Ceratium porioides take up bacteria and par-
ticles of carmine by means of a kind of vacuole which forms at
the surface of the body (fig. 12). It is situated most frequently
in the anterior half, but may form at any portion of the surface.
The formation of these vacuoles can best be studied in a culture
which has been on a slide for some hours, If a drop is taken

THE ASEPTIC CULTIVATION OF MYCETOZOA. 63

from a flask culture, placed on the slide, and immediately
examined, the vacuoles are not always found. In favorable
cultures the zoospores are found creeping on the slide and
changing their shape; the flagellum is in active motion. At
the posterior portion of the body a pseudopodial prolongation
is put out, by means of which they adhere to the slide. Not
far from the attachment of the flagellum there arises a pro-
jection, which is placed more or less perpendicular to the surface
of the body. At first sight it appears to be simply a conical
or papillary projection from the surface layer of protoplasm.
On closer examination a thin fold of protoplasm is seen to
arise from the whole length of the projection, and extends
forward toward the flagellum, where a second similar projec-
tion arises. At this stage the two projections, with the thin
fold-like connections, form a funnel-shaped depression. ‘The
apices of the two projections approach each other and fuse,
converting the funnel-shaped depression into a closed vacuole,
which passes backward, and is lost among the vacuoles near
the centre of the body. The flagellum is in active motion,
and throws the bacteria or particles of carmine into the funnel-
shaped depression, which then closes and forms the vacuole.
The writer has seen bacteria and particles of carmine taken
up in this way. When these vacuoles are located in the
posterior region of the body, the flagellum has not been
observed to throw the bacteria into them. Fig. 12, a—j, shows
the different stages in the formation of these vacuoles. The
zoospores of Phys. cinereum have not been observed to form
these vacuoles.

The ameeboid stage of the zoospores of Phys. cinereum is
much more pronounced than that of Stemonitis A, B, C, and
Cerat. porioides. A large part of the active existence of the
zoospores of Phys. cinereum is passed without the presence
of flagella. They put out pseudopodia which are more or Jess
angular, and their change of shape resembles more that of an
amoeba. They may ingest their food after the manner of ameebe.

The zoospores of Stemonitis A, B, C, and of Ceratium
porioides are characterised by rarely being without a fla-

64 CASPER O. MILLER.

gellum, and when they do change their shape the pseudopodia
are more rounded.

The zoospores of plasmodia usually have a single flagellum,
but they may have two or four. The writer has not seen
zoospores of the Endosporia with more than one nucleus,
whereas large zoospores of Cerat. porioides have occa-
sionally been found which have two distinct nuclei (fig. 12, /.).

In all species of plasmodia which the writer has examined,
one can distinguish two forms of microcysts in the cultures.
In the simple form the cyst wall is made up of a single homo-
geneous membrane, closely applied to the protoplasm, as indi-
cated in fig. 12, m. In the second form the cyst wall is made
up of an outer thick membrane, irregular or scalloped in
outline, and an inner, thinner membrane, which is closely
applied to the protoplasm (see fig. 12, ”., 0.). One finds cysts
intermediate between these two varieties. The simple micro-
cysts of Stemonitis A measure 5—7 uw in diameter, the thick-
walled cysts measure 10—14 pm in diameter.

In cultures of Stemonitis A, B, and C, and Phys. cin-
ereum made in hay infusion, the thin-walled microcysts
remain unstained, whereas the membrane of the thick-walled
cysts may be stained a brownish colour.

In old cultures made in hay infusion at times one sees
microcysts with dark brownish, almost black pigmented
granules in their interior.

Lister (44) speaks of the spores of Ceratiomyxa as having
four “ nucleus-like ” bodies, and pictures them indistinctly in
his plates. The writer’s observations show, from staining with
picro-carmine, that these bodies are true nuclei.

Famintzin and Woronin (10) showed that after the proto-
plasm escaped from the spores of Ceratium it remains at one
place for some time, undergoing ameeboid changes. It then
divides into four round segments, each of which divides and
gives origin to two zoospores, Lister (44) describes the
naked spore dividing into eight spherical segments which re-
main attached to each other. These develop flagella, and then
separate,

THE ASEPTIC CULTIVATION OF MYOETOZOA. 65

The writer has seen them divide after the manner described
by Famintzin and Woronin (see fig. 13).

Microscopical Appearance of the Plasmodia and
the Structure of the Nuclei.

Cultures made in fluids without the presence of hay offer
the best facilities for studying the plasmodia. The smaller
plasmodia are usually found lying in or upon a clump of
microcysts and bacteria. The larger plasmodia can be seen
spread out on the bottom or sides of the flask. The closeness
of the network, the size of the branches, and the peripheral
arrangement of the network can be studied.

For microscopical study, the plasmodia, with a few drops of
the fluid in which they grow, are placed on a slide by means
of a pipette. A cover-glass is carefully laid on, and is supported
by small bits of wax at each corner to prevent injuring the
plasmodia by pressure. The specimen can be immediately
examined, and then placed in a moist chamber for twelve to
twenty-four hours, after which it may again be examined.

If it is desired to preserve the specimen, the plasmodium
can be fixed and hardened on the slide and stained by any of
the usual methods. Hardening in picric and acetic acids, and
then staining in picro-carmine, give good results.

Fig. 6 represents a segment of a plasmodium taken from a
culture of Stemonitis A. Some of the clumps containing
microcysts, bacteria, and plasmodia were placed on a slide. The
specimen was examined at the expiration of an hour, and a
number of rather long, finger-lhke, blunt, unbranched proto-
plasmic processes were seen radiating from the periphery of
the clumps. At the expiration of twenty-four hours the
clumps were surrounded by a network of protoplasmic branches.
The primary trunks are large, and extend from the clump
toward the periphery of the network. They anastomose with
each other, and are also connected by means of a finer network
of secondary fibres. At the periphery of the network are
irregular, angular, flattened, protoplasmic expansions, which
at times unite and form irregular plates, as shown in the

vot. 41, part 1.—NEW SERIES. E

66 CASPER O. MILLER.

figure. Occasionally one gets specimens where the fusion of
these expansions is more extensive.

When the plasmodia spread out as in the figure, one may
see microcysts which have been taken up from the clump by
the plasmodium and carried along the branches toward the
periphery.

One frequently finds small plasmodia composed of only a
few branches which do not form as close a network as in the
figure, and the ends of the branches are not so angular.

The writer has not, as yet, sufficiently studied plasmodia from
Stemonitis A, B, and C to be able to point out any con-
stantly marked characteristics which distinguish them. They
are all more transparent and not so granular as the other
plasmodia which have been studied.

Fig. 7 represents a portion of the periphery of a small plas-
modium of Phys. cinereum taken from a culture and allowed
to spread out in the same way. It has more the appearance of
a spread-out, perforated, protoplasmic plate than a network
of branches, and one cannot distinguish between primary and
secondary branches.

Fig. 8 represents the peripheral expansion of a plas-
modium taken from a hay culture in which were Chon. diff.
and Didym. microcarpon. Here there are several large
trunks going to a broad peripheral expansion, less perforated
than Phys. cin.

The nuclei of the plasmodia of Stem. A are distributed
irregularly along the branches of the network and in the
peripheral expansions. In some of the larger branches they
may be collected in groups, where, as in other branches, they
are situated at irregular intervals. In the larger branches
the most of the nuclei lie in the peripheral layer of proto-
plasma, often immediately under the surface. As a rule one
does not see the nuclei in the living plasmodia, but occasion-
ally, if the nucleus lies at the side of the branch, a nucleus
can be seen as a spindle-shaped body just beneath the surface,
and may cause a bulging at that point.

The shape of the nuclei is usually that of a flattened oval

THE ASEPTIC CULTIVATION OF MYCETOZOA. 67

disc. When seen on the edge they appear spindle-shaped.
They contain from one to six or seven small bodies, which
may provisionally be called nucleoli. One finds nuclei con-
taining two nucleoli. These nuclei are at times constricted,
and suggest stages of division.

Strasburger (23) studied the division of the nuclei of
Trichia fallax during the formation of the sporangia. He
succeeded in finding stages which showed karyokinesis.
Rosen (89) studied the division of the nuclei in the forming
ethalia of Aithalium septicum, but found the division of
the nuclei simpler than that described by Strasburger.

Lister (44) describes the division of the nuclei of zoospores
and of the nuclei of the plasmodia by karyokinesis, but also
concludes that they divide by direct division. The writer has
not been so fortunate as to find nuclei dividing by karyokinesis,
although frequent search was made for them.

The zoospores can frequently be seen to divide and form
two zoospores, and to the writer there seems to be evidence
that at times the protoplasm of the microcysts breaks up into
a number of small segments, and these segments enlarge and
develop zoospores.

The nuclei of Phys. cinereum are smaller than those of
Stemonitis A. They are spherical, containing one or more
nucleoli, and are distributed in every part of the protoplasm,
but are more abundant in some portions than in others.

Before concluding the writer wishes to acknowledge the
invaluable assistance which he received from Professor Biitschli
while pursuing these studies in Heidelberg, also to Professor
Wladimir Schewiakoff, then assistant at the laboratory of
Heidelberg, for his assistance in the preparation of the
plates.

68

fo)

10.

11.

12.

13.

14.

15.

16.

if
18.

CASPER O. MILLER.

BIBLIOGRAPHY.

. DE Bary, L.—‘ Flora,’ 1854, p. 648.
. DE Bary, L.—“‘ Ueber die Myxomyceten,” ‘ Botanische Zeitung,’

1858, p. 357.

. DE Bary, L.—“ Die Mycetozoen,” ‘ Zeitschrift fiir wiss. Zoolog.,’ 1860,

p. 88.

. Crenkowsk1, L.— Ueber parasitische Schlauche auf Crustaceen und

einigen Insectenlarven,” ‘ Botanische Zeitung,’ No. 25, 1861, p. 170.

. Wicanp, A.—“ Zur Morphologie und Systematik der Gattungen Trichia

und Arcyria Pringsheim,” ‘ Jahrbiicher fr. wiss. Botanik,’ Bd. iii,
1863,

. Crenxowsk1, L.—“ Zur Entwickelungsgeschichte der Myxomyceten”

(same Journal), p. 325.

. Crenkowsk1, L.—‘‘ Das Plasmodium ” (same Journal), p. 325.
. DE Bary.—‘ Die Mycetozoen’ (Schleimpilze), 1864.
. Lirperkiinn.—‘ Ueber Bewegenserscheinung der Zellen,’ p. 376, Tafel iv,

fig.38; ‘Schriften d. Gesellsch. ziir Bef. der ges. Naturw. zu Marburg,’
November, 1873.

Famintzin, A., und Woronin, M.—‘ Ceratium hydnoides und C.

porioides als zwei neue Formen von Schleimpilze,’ Botanische Zeitung,
No. 34, 1872.

Famintzin, A.—‘‘ Ueber zwei neue Formen von Schleimpilze,” ‘ Mém.
de l’Acad. Imp. des Se. de St. Pétersb.,’ ser. 7, tome xx, 1873.

Rozz, E.*—“Des Myxomycétes et de leurs place dans la systéme,”
‘ Bulletin de la Soc. Bot. de France,’ tome xx, 1873.

RostaFinskI, J. F.—‘ Versuch eines Systems der Mycetozoen’ (Inau-
gural Dissertation), Strasburg, 1873.

RostarFinskI, J. F.*—‘ Sluzowce (Mycetozoa) Monografia,’ Paris, 1875;
‘ Dodatek I. do Monografii Sluzowcow,’ 1876.

Scuuuzg, F. E.—< Rhizopodenstudien V.,” ‘ Arch. fiir mikrosk. Anatom.,’
Bd. xi, 1875.

Cienkowsk1, L.—‘ Ueber einige Rhizopoden und verwandte Organis-
men,” ‘ Arch. f. mikr. Anatomie,’ Bd. xii, 1876.
Cooks, M. C.—‘ Myxomycetes of Great Britain,’ London, 1877.

Bitscu11, O.—“ Beitrige zur Kenntniss der Flagellaten und verwandter
Organismen,”’ ‘ Zeitschr. f. wiss. Zoologie,’ Bd. xx, 1878.

References marked * have not been read.

19.

20.
21.

22.

23.

24.

25.

26.

27.
28.

29.

30.

31.

32.

33.

34,

35.

36.

37.

38.

THE ASEPTIC CULTIVATION OF MYCETOZOA. 69

Gruser, A.—“Dimorpha mutans,” ‘Zeitschr. f. wiss. Zoologie,’
Bd. xxxvi, 1881.

Kent, 8.—‘ A Manual of Infusoria,’ London, 1880-82.

DE Bary.—‘ Vergleichende Morphologie und Biologie der Pilze, Myce-
tozoen, und Bacterien,’ Leipzig, 1884.

Staut, E.—‘‘ Zur Biologie der Myxomyceten,’’ ‘ Botanische Zeitung,’
1884.

SrrasBuRGER, H.—‘‘ Zur Entwicklungsgeschichte der Sporangien von
Trichia fallax,” ‘ Botanische Zeitung, 1884.

Zorr, W.—“ Die Pilzthiere oder Schleimpilze,” ‘ Handbuch der Botanik,’
Schenk., Hefte 15 und 16, 1884.

Warp, H. M.—‘* The Morphology and Physiology of an Aquatic Myxo-
mycete,” ‘Studies from the Biological Laboratories of the Owens
College,’ vol. i, 1886.

STRASBURGER, E.—‘ Das Botanische Practicum,’ Jene, 1887.

Ravyxizr, C.*—“ Myxomycetes Danie,” ‘ Bot. Tidsskr.,’ 1888-89.

Lister, A.*—‘‘ Notes on the Plasmodia of Badhamia utricularis and
Brefeldia maxima,” ‘ Annals of Botany,’ vol. ii, 1888 and 1889.

Birscuui, O.—‘‘ Protozoa,” in Bronn’s ‘Klassen und Ordnung des
Thierreichs,’ 1889.

Lister, A.— Notes on the Ingestion of Food-materials by the Swarm-
cells of Mycetozoa,” ‘Journ. Linn. Soc. London (Botany),’ vol. xxv
1889, p. 435.

Scuroter, J.—‘ Myxogasteres (eigentliche Myxomyceten) in die natiir-
lichen Pflanzenfamilien von Engler und Prantl,’ 36 Lieferung, 1889.
Wivneate, H.—‘‘ Notes on Enteridium Rozeanum,” ‘Proceedings

of the Acad. of Nat. Sci. of Philadelphia,’ 1889.

Zorr, W.—‘ Vorkommen von Fettfarbstoffen bei Pilzthieren (Myceto-
zoen),” ‘Flora,’ 1889, p. 3538.

Lister, A.—‘“ Notes on Chondrioderma difforme and other Myce-
tozoa,” ‘Annals of Botany,’ vol. iv, 1890, pp. 281—298.

Prerrer, W.*—‘ Ueber Aufnahme und Ausgabe ungeléster Korper,’
1890, p. 154.

Rex, G. A.—‘‘ A Remarkable Variation of Stemonitis Bauerlinii,
Mass.,” ‘ Proce. Acad. Nat. Sci. Philadelphia,’ 1890, p. 36.

Rex, G. A.—‘‘ New American Myxomycetes,” ‘Proc. Acad. Nat. Sci.
Philadelphia,’ 1891, pp. 389—398.

CxrLakovskI, L., jun.—‘‘ Ueber die Aufnahme lebender und todter verdau-

licher K6rper in die Plasmodien der Myxomyceten,” ‘ Flora,’ Bd. lxxvi
(Erg. Bd.), 1892.

70 CASPER O. MILLER.

39. von Rosen, F.—-‘‘ Studien tiber die Kerne und die Membranbildung bei
Myxomyceten und Pilzen,”’ ‘ Beitrage zur Biologie der Pflanzen,’
Bd. vi, Heft 2, 1892.

40. Hertwie, R.—‘ Lehrbuch der Zoologie,’ Jene, 1892.

41. Mass, G.—‘ A Monograph of the Myxogastres,’ London, 1892.

42. McBrips, T. H.—“ The Myxomycetes of Eastern Iowa,” ‘ Bulletin
from the Laboratory of Nat. Hist. State Univ. of Iowa,’ vol. ii, 1892.

43. Ceaxovskl, L., jun.—‘ Die Myxomyceten Bohmen’s,’ Prag, 1893.

44. Lister, AA—‘ A Monograph of the Mycetozoa,’ London, 1894.

45. Mitirr, C. O.—< Ueber aseptische Protozoenkulturen und die dazu
verwendeten Methoden,” ‘Centralb, f. Bak. u. Parasitenk.,’ Bd. xvi,
1894, No. 7.

EXPLANATION OF PLATES 6 and 7,

Illustrating Mr. Casper O. Miller’s paper on “ The Aseptic
Cultivation of Mycetozoa.”

Fic. 1.—Sporangia of Chondrioderma diff. on the glass above the
fluid, from a culture made with unsterilised hay.

Fic. 2.— Physarum cinereum, showing sporangia at the periphery of
the plasmodial network which radiated from the end of a stalk of hay leaning
against the glass.

Hic. 3.—Stemonitis A, with the sporangia in process of formation on
the side of the flask at a point where a stalk of hay touched the glass. The
colour had not fully developed. The course which the plasmodium took is
shown by the thread which descends obliquely to the water.

Fic. 4.—Sporangia of Stemonitis B, fully developed on stalks of hay.

Fie. 5.—A plasmodium of Stemonitis lying on the hay at the surface
of the water before it ascends to form sporangia.

Fic. 6.—A portion of plasmodium of Stemonitis A., which had spread
out under a cover-glass. The trunks radiate from a clump of encysted zoo-
spores, some of which were enclosed in the protoplasm. Drawn with a
camera lucida. xX 180.

Fic. 7.—A portion of a plasmodium of Physarum cinereum spread out
under a cover-glass. x 1380.

Fig. 8.—A portion of a plasmodium gotten from hay. x 130.

THE ASEPTIC CULTIVATION OF MYCETOZOA. fa

Fic. 9.—A sporangium of Stemonitis A, drawn with a camera lucida
x 66.

Fic. 10.—A sporangium of Stemonitis C. x 66.

Fic. 11.—A sporangium gotten on a stump at Heidelberg. It has not been
identified with any of the species described in works on Mycetozoa.

Fic. 12.—Zoospores of Stemonitis. 12 a—A, A zoospore in the act of
forming a nutrient vacuole, and taking in a granule of carmine. 12 %, J,
showing the location of the nutrient vacuoli posteriorly. 12 4. A zoospore
with four flagella. 127. A zoospore of Ceratium porioides, with two
flagella and two nuclei. 12m. An encysted zoospore, with a thin wall.
12 x, o. Encysted zoospores, with thick walls.

Fic. 13 shows different forms of spores of Ceratium porioides, with
the four nuclei and four zoospores developing from one spore.

cf - - ‘ y c a 2

ON THE DEVELOPMENT OF TUBULIPORA. 73

On the Development of Tubulipora, and on some
British and Northern Species of this Genus.

By

Sidney F. Harmer, Sc.D.,

Fellow of King’s College, Cambridge ; Superintendent of the University
Museum of Zoology.

With Plates 8—10.

Introduction.

THE principal object of the observations described in the
present paper was to test the conclusion, at which I formerly
arrived, that a process of embryonic fission is of normal occur-
rence in Cyclostomatous Polyzoa. The process had already
been demonstrated in Crisia (15) and in Lichenopora (16),
and I am now able to show that the development of Tubuli-
pora! takes place on essentially the same lines. In the course
of these observations it became apparent that the discrimina-
tion of the British species of Tubulipora had not been
characterised with sufficient precision.

The satisfactory determination of the species of Cyclo-
stomatous Polyzoa is not an easy task. There are perhaps
few cases, even in this group, in which the synonymy is in a
more involved state than in the genus Tubulipora. Many
of the specific diagnoses which have hitherto been given are
based on immature specimens, or are for other reasons insuf-
ficient for purposes of identification, and the compilation of
unduly long lists of synonyms is for this reason undesirable.

1 A preliminary note (17) on this subject was published, in which the
genus appears as Idmonea.

74, SIDNEY F. HARMER.

The question is further complicated by difficulties in deciding
which names have the greatest claim to be assigned to parti-
cular species.

Before proceeding to the accounts of the species the scope
of the paper may be stated. I have given diagnoses of two
Northern species, one of which I believe to be new, and of all
the forms ordinarily recognised as British species of Tubuli-
pora, with the exception of T. lobulata, Hassall.

I have not had a sufficient supply of this supposed species to
enable me to come to any clear conclusion about it. Many of
the colonies which I have found on shells in the deeper Ply-
mouth dredgings (20—30 fathoms) agree well enough with
Hincks’s description of this form. I have also seen similar
specimens from the Liverpool district, kindly lent to me by
Professor Herdman; but I have not been able to examine un-
injured mature colonies. In the absence of perfect ovicells,
it is possible that some of the specimens supposed to be T.
lobulata may be worn examples of other species of Tubuli-
pora. I prefer, therefore, to express no opinion with regard
to this form.

The paper is divided in the following way :

I. Structure of the colony and of the ovicell. (The terms
“ ogeciostome” and “ oceciopore”’ are here proposed.)

II. History of the species and genus.

III. Synonymy, diagnoses, and accounts of the species.
(New species, T. aperta.)

IV. The nature of certain vesicles found in the tentacles and
other parts, with a few statements relating to the budding and
the structure of the adult zocecium.

V. Description of the development. (The terms “ axial
lobe” and “ lateral lobes”’ are here proposed.)

VI. The morphology of the internal parts of the ovicell.

I. Structure of the Colony and of the Ovicell.

A colony of Tubulipora, as of other Cyclostomes, takes
its origin in the well-known circular primitive disc (Pl. 8,

ON THE DEVELOPMENT OF TUBULIPORA. 75

fig. 2) formed by the calcification of the body-wall of the
metamorphosed larva. The early zoarial development is well
described by Barrois (1, p. 70, pl. iv, fig.), whose account
perhaps refers partly to T. phalangea and partly to T.
plumosa; and it results in all species in the formation of a
colony which is at first pyriform or fan-shaped. The terms
“proximal” and “ distal’? will be used with relation to the
primitive disc. The distal curved margin of the colony is
bounded by the ‘terminal membrane” (16, p. 93), which
consists partly of an uncalcified cuticle (ectocyst) and partly of
underlying protoplasmic structures. The terminal membrane
does not remain as a continuous sheet, since portions of it are
continually cut off by the upgrowth of the calcareous septa
which form the lateral walls of the zocecia or ovicells. Each
of these structures is thus closed by a derivative of the original
terminal membrane, and this name may consequeutly be ap-
plied also to the uncalcified distal wall of each unit of the
colony (fig. 24). Both the zocecia and the ovicells are added
to, after their first formation at the growing edge, only by the
prolongation of parts of the calcareous tubes which have
already been developed, though this is not altogether true of
the Lichenoporide, in which the character of the colony is
altered by the subsequent development of “cancelli.’ It thus
follows that a young Tubulipora colony is identical, so far
as its calcareous structures are concerned, with the proximal
part of the same colony at a later stage of its existence; and
that in order to understand what a particular colony or ovicell
was like at an early period of its development, it is only neces-
sary to imagine the distal parts of the colony suppressed.

The form of a particular colony is due to the behaviour of
its terminal membrane, which is identical in its extent with
the growing margin. Should this remain undivided, and con-
tinue to grow symmetrically, a pyriform or flabelliform colony
will result, according to the relative activity of growth in the
longitudinal and transverse directions. Under certain circum-
stances, probably due to unfavourable conditions, the colony
becomes mature by developing an ovicell without losing its

76 SIDNEY F. HARMER.

pyriform shape. I have found mature colonies of this type
commonly in at least three of the species described in the
present paper, namely in T. phalangea, T. flabellaris
(fig. 4), and T. aperta (fig. 2).

In other cases the transverse growth of the terminal mem-
brane is more active, and the flabelliform character becomes
more marked (figs. 1 and 5), its lateral edges commonly
growing proximally so as more or less to encircle the primi-
tive disc. This flabelliform shape typically occurs in well-
grown colonies of T. flabellaris and T. aperta. In other
species the terminal membrane commonly divides at an early
stage, so that there is a cessation of growth between the two
parts of the divided membrane. The colony then grows into
two lobes, which usually diverge from one another. This
form of growth is well indicated by Pallas (83, p. 248) in his
original account of T. liliacea (= T. serpens, auctt.).
Colonies of this species, as well as those of T. phalangea
and T. plumosa, often become mature in this condition ; but
in all three species further divisions of the terminal mem-
brane! usually take place, the colony thus becoming variously
lobed. In T. liliacea alone of the forms described in this
paper, there is in many cases a marked tendency for the lobes
to become free and erect; and an Idmonea-like form thus
results.

The common basal part of a colony consists of a mass of
pyramidal tubes with pointed proximal ends, the arrangement
being comparable with that of a honeycomb, all the “ cells” of
which look in one direction, represented in Tubulipora by
the growing margin of the lobe. The base of the colony, or
basal lamina, is the sum of the lower walls of the more
proximal parts of the zocecia. It is completely adherent to
the substratum in many species of Tubulipora, but it may
grow out freely from it, and the colony thus becomes erect.
Each zocecium is developed at the growing edge by the forma-
tion of a new radial septum in connection with the basal

1 A commencing division of the terminal membrane is indicated on’ the
right side of fig. lL.

ON THE DEVELOPMENT OF TYBULIPORA. 77

lamina, and grows at first in a more or less horizontal, centri-
fugal direction. As its length is added to at the growing
edge, its upper wall gradually rises above the general level of
the colony, and some of the zoccia thus form projecting ridges
(fig. 1). As the zocecium rises up in this way, its basal wall
splits off from the basal lamina, between which and itself a
cavity thus originates. This cavity is the beginning of a new
zocecium, which continues to grow in the same manner.

It follows from this description that each zocecium proxi-
mally reaches the basal lamina, as in Lichenopora (16, wood-
cut, p. 84); a younger zoccium always originating from the
basal side of an older one, and making its appearance distally
to it after a certain time.

As the zoccium continues to grow longer, its distal part be-
comes free from the common basal mass of the colony. It may
become completely free on all sides, and it then, by the activity
of its own terminal membrane, develops into a curved tube,
which is generally cylindrical, though in some cases with an
oval transverse section and orifice. Very commonly in Tu-
bulipora the zoccium does not become free on its basal
side, where it remains connected with one or more of its
younger neighbours. In this way it forms the beginning of a
row, usually uniserial, of connate zoecia, as show in figs. 1, 5,
and 9. The adjacent zocecia of a series are separated by a
flat septum, corresponding with the intersecting plane of two
cylinders. Any one of the middle zoccia of a simple series
will thus have four flat walls (fig. 1).

The shape of the orifice corresponds with that of the
zocecium, being at first angular in the case of serial zoccia.
If, however, growth of these goes on actively, the series may
be resolved above into its constituent units, each of which then
becomes cylindrical with a round orifice.

The ovicell of Tubulipora is an enlarged zocecium. This
is shown by several facts, the most striking of which is that
when young it has a polypide, and then differs in no way from
an ordinary zoecium. The proximal end of the ovicell thus
takes part in the formation of the basal lamina. On looking

78 SIDNEY F. HARMER.

down into the mouth of the open calcareous funnel (fig. 4.7),
which is the condition of the young ovicell at the time when
it commences to expand distally, a tubular cavity can be seen,
which passes down to the basal lamina and represents the ovi-
cell in its zocecium stage.

The upper wall of the expanded distal part of the ovicell is
thickly perforated by pores (fig. 1). These pass through the
calcareous layer; they are wider internally and constricted ex-
ternally, where they are closed by a cuticle, the pore being
entirely filled with a few cells. The pores are always much
more numerous in the upper wall of the ovicell than in the
walls of the zoccia, as Waters has pointed out (45, p. 277,
and in other papers). It is perhaps probable that the function
of these pores is mainly respiratory, to provide for the gas-
exchange which must be necessary for the development of the
great mass of larve which are formed inside the ovicell. The
definitive assumption of the ovicell-character by a zocecium is
thus marked by a great increase of the porosity of its upper
wall,

In some species, and notably in T. flabellaris (fig. 4,
ovicell 2) and T. aperta, the more porous part can in some
cases be seen to be separated from the less porous part by a
sharp line. ‘The proximal part has the appearance of a
zocecium ; and we thus have an external indication of the time
at which the fertile zocecium became definitely an ovicell.

An ordinary zocecium was seen to become, sooner or later,
free from the common basal part of the colony, unless it
remained connected with it by forming one of a connate series
or of a fasciculus of zocecia. The corresponding process in the
ovicell is the formation of its tubular portion ending in the
orifice. This part, in all respects comparable with the up-
standing free part of an ordinary zocecium, is, however,
developed late, and can only be seen in ovicells which are
nearly mature. The ovicell, in fact, remains a part of the
common basal mass of the colony for a long period, during
which its distal end is expanding and increasing the space
available for the growth of the embryos. Before the distal

ON THE DEVELOPMENT OF TUBULIPORA. 79

expansion begins the ovicell has become separated by younger
zocecia from the basal lamina, and its floor is thus formed by
their proximal parts. The upper part or roof of the ovicell
spreads out horizontally as the conspicuous, porous, calcareous
film usually described as the ovicell. If no fresh zocecia were
formed at the growing edge, the roof of the ovicell wouid be a
simple fan-shaped film. But new zoccia continue to grow in
the same way as if no ovicell were present (Pl. 10, fig. 32).
The floor of the ovicell, formed by the upper walls of the
younger zoecia, or really by the septa dividing its own body-
cavity from that of these zocecia, thus rises into radial ridges
which encroach on its cavity. As these become more vertical
they meet the growing horizontal roof of the ovicell, and with
further prolongation stand up from it at a right or obtuse
angle.

In most of the species investigated the zocecia which are
younger than the ovicell are arranged in connate series. But
whereas the proximal parts of the fertile lobes have their
zocecia in obliquely transverse, alternate rows (fig. 5), these
series usually become radial in the region of the ovicell. In
other words, a young zoccium which has reached the roof of
the ovicell in the manner just indicated does not become frée,
but remains connected with the expanding growing edge by
a still younger zoecium. A radial series of connate zoccia
thus results, and the roof of the ovicell is divided into two
lobes, one on each side of the zocecial series. After a time the
series may become completely free from the growing edge, and
the two adjacent lobes of the roof of the ovicell may then unite
distally to the series.

In my preliminary note (17, p. 212) I have stated that the
zocecia may thus be left as columns passing freely through the
cavity of the ovicell. I believe that this statement is not
correct for Tubulipora. The cavity of the ovicell may indeed
surround the zocecium, or series of zocecia, on the proximal,
lateral, and to some extent on the distal sides; but I have
found no evidence that two lobes which have become con-
tiguous on the distal side of a zocecium ever really unite, as

80 SIDNEY F. HARMER. :

they certainly doin Lichenopora. In transparent prepara-
tions it can be seen that the cavities of the two lobes remain
separated by a vertical radial septum, the upper edge of which
is generally indicated as a line on the roof of the ovicell even
in dried colonies (fig. 5.s.).

As the radial series of zocecia are formed all round the peri-
phery, the ovicell itself consists of a cavity which branches
dichotomously at the proximal end of each series. In large,
actively growing colonies (fig. 5) the ovicell usually branches
in a palmate manner, and the primary lobes may undergo
further dichotomy once or more. The extent to which the
ovicell branches is not a specific character, but is probably
dependent partly on nutrition and partly on temperature.
Figs. 2 and 4 represent mature fertile colonies of T. aperta
and T. flabellaris respectively, and fig. 1 a fertile lobe of T.
plumosa. ‘The ovicell shown in the last figure is much
more complicated than that shown in the former two; but it
must be distinctly understood that the ovicell of T. flabel-
laris and T. aperta may be large and much branched in
vigorous colonies.

After the ovicell has reached a certain size it develops its
tubular portion, ending in the orifice through which the larve
make their escape. Here, however, there is some modification
in the course of the development as compared with an ordinary
zocecium. While the orifice of the latter is merely the part
which, for the time being, forms the distal end of the zocecium,
the lobes of the ovicell undergo a considerable amount of
growth, in most cases, after the orifice is fully formed. Sooner
or later, however, the peripheral ends of the lobes, closed
merely by a soft terminal membrane during growth, become
completely calcified, except in the abnormal specimens of T.
aperta (fig. 2), described under the heading of that species,
in which accessory openings may be formed by the outgrowth
of the lobes into tubular passages which more or less resemble
the real orifice of the ovicell. The complete calcification of
the peripheral ends! the lobes of the ovicell takes place, as
in other Cyclostomes, by the encroachment of the calcareous

ON THE DEVELOPMENT OF TUBULIPORA. 81

roof on the terminal membrane, which is, so to speak, gra-
dually constricted until it completely disappears.

My study of Tubulipora has fully confirmed the conclusion
stated in my paper on Crisia (18, p. 128), that the form of the
orifice of the ovicell is of great importance for the discrimination
of the species. As I am convinced that this structure will become
increasingly important in the systematic study of the Cyclo-
stomata, I venture to think that a special terminology may be
convenient for descriptive purposes; and I therefore propose
to term the passage by which the larve escape from the
ovicell the “oceciostome,”’ and its actual external orifice the
“oeciopore.” The oceciostome is usually a tubular or funnel-
shaped structure, as in the species of Tubulipora here de-
scribed, and in many other Cyclostomes. The part connecting
the cavity of the ovicell with the oceciopore may be termed
the ‘‘ tube” of the oceciostome, whatever its shape. The tube
is not a necessary part of the occiostome, since it is absent in
Crisia aculeata, Hassall, in which the oeciopore represents
the entire oceciostome.

The structure of the wall of the tube is commonly different
at its two ends. The proximal portion is often pierced by
pores identical with those in the roof of the ovicell, of which
it is a direct continuation (fig. 1), while the distal portion
is imperforate. Valuable specific characters are afforded by
the shape, size, and relations of the oceciostome, and by the
size and position of its tube and oeciopore.

The constancy of the characters of the oceciostome has been
verified by the examination of numerous specimens of several
species from various localities. While the shape and size of
the entire colony, and therefore of the ovicells, has been shown
to be highly variable, the oceciostome retains its character,
whatever the condition of the colony. I do not, of course,
mean to assert that there is no variation in the oeciostome.
Variation does occur, sometimes within fairly wide limits; but
it is usually possible to decide the species at once by an
inspection of its occiostome. In order to do this it may be
necessary to have a fully developed oceciostome, since the

vou. 41, parr 1.—NEW SER. Fr

82 SIDNEY F. HARMER.

specific characters are not completely shown in the immature
condition of this structure. There are occasional cases in
which the occiostome of a given colony may appear inter-
mediate between two species. In most of these cases com-
parison with other specimens will, however, usually leave no
doubt of the existence of a definite character for the oceciostome
of each species, in spite of the occasional occurrence of difficult
cases.

The value of the characters of the oceciostome in diagnosing
a species is strikingly illustrated by the specimens of T.
flabellaris, Fabr., which have come into my hands. Fig. 4
is an Arctic specimen of this species, dredged by Colonel H. W.
Feilden on July Ist, 1895, in Barents Sea, and presented by
him to the Cambridge Museum of Zoology. Although the
colony is very much smaller than the form of this species
figured by Smitt (40, pl. ix, fig. 6),1t possesses no less than six
complete ovicells, each with a fully formed oceciostome (1—6),
five of which are visible in the position in which the colony
lies, besides one immature ovicell (7). The ovicells are
crowded and are very small, exhibiting hardly any of the
lobing which may occur in this species as in others. The
principal characteristic of the species, the flattened shape of
the oceciostome, is nevertheless as well marked in each ovicell
as it is in the lobed ovicells of a larger colony, sent to me by
Dr. Nordgaard from Hammerfest.

The above case can hardly be regarded as a mere alee
abnormality. I have examined four specimens (including the
above) from Colonel Feilden’s collection, and all of them
showed the same peculiarities. ‘Two other colonies (about the
same size as the above) had each four small, contiguous ovi-
cells with fully formed oceciostomes, and an immature fifth
ovicell. The last consisted (unlike the others) of two diverg-
ing lobes, which may, however, have been formed from two
larve. One lobe had two ovicells with occiostomes, and at
least one immature ovicell, and the other lobe had one ovicell
with a fully developed oceciostome. In every ovicell the long
flattened oceciostome, which I regard as the distinguishing

ON THE DEVELOPMENT OF TUBULIPORA. 83

feature of this species, was developed on precisely the same
type.

A case of the abnormal development of four contiguous ovi-
cells has been described by me (18, p. 166, pl. xii, fig. 13) in
Crisia ramosa. These cases are interesting as showing the
essential similarity of the ovicell and the zocecium. The pro-
duction of an ovicell is probably induced by the development
of an embryo from one of the eggs which so commonly occur
in young zocecia, most of which do not ordinarily become
fertile.

The simple ovicells of the specimens of T. flabellaris
from Barents Sea are similar to those which were figured by
Savigny (87, pl. vi, figs. 4.2 and 5.2), and described by Audouin
under the names of Proboscina Boryi, and P. Lamou-
rouxii. Ovicells of a similar type are known in Diasto-
pora suborbicularis, Hincks, and in many fossil forms.
They are figured, for example, by Gregory (12, pl. 1, fig. 6;
pl. iii, fig. 8), and this type of ovicell is given by Walford (41,
p. 78) as a characteristic of his genus Pergensia, although I
cannot agree with him in thinking that the character shows
any approach to the Cheilostome genera Lekythopora and
Pecilopora. It appears to me probable that the stunted
specimens of T. flabellaris above described have reverted to
a more primitive condition (a conclusion borne out by the
evidence of fossils) in developing a less complicated form of
ovicell than is usually produced in other recent species, and
even in T. flabellaris itself. This reversion may be re-
garded as due to the small size of the colony, in conjunction
with the successful development of primary embryos in several
contiguous zoccia. It is probable that this points to a time
when the ovicells were ordinary zocecia, and that the restric-
tion in the number of fertile zocecia now so characteristic of
Cyclostomes is not a primitive feature of the group.

References to the oceciostome in any form hitherto described
as Tubulipora or Idmonea are curiously rare; and most
authors do not seem to have considered the possibility of the
occurrence of this essential part of the ovicell. Smutt (40)

84 SIDNEY F. HARMER.

has described it in Idmonea atlantica (p. 443), in I. ser-
pens [= T. liliacea of the present paper] (p. 446), in T.
fimbria [= 'T. aperta] (p. 455), in T. flabellaris (pp:
456, 457), and in T. lobulata (p. 457). The descriptions
there given are not, however, sufficiently precise to be used in
discriminating the species, nor are they illustrated by figures
which really show their character. Waters (44, pp. 256—259,
and 48, p. 339) has described the occiostome in one or two
species of Tubulipora or Idmonea; and Kirkpatrick (22, p.
22, pl. iv, figs. 6 a, 6 6) has described and figured the ocecio-
stome of I. pulcherrima. These are almost the only refer-
ences I have been able to find to the occiostome of recent
species of Tubulipora. It appears to me that a Tubuli-
pora colony reaches its fully adult condition when the ovicell,
with its oceciostome, is fully formed. If this view be correct,
the importance of the oceciostome as a specific character is
intelligible. Colonies in which this structure is not yet deve-
loped are in a state of immaturity. It is perhaps as reason-
able to characterise a species of Tubulipora without taking
account of its oceciostome as it would be to describe a new
species of Cervus without describing the form of the antlers.
It is unfortunately impossible to describe the oceciostome of a
Cyclostome in the majority of cases, and particularly in fossil
specimens; but I think the description of this structure should
be an essential part of a diagnosis wherever it can be given.

The absence of the oceciostome in a given case may depend
on the season of the year at which the specimen was obtained.
It is, for instance, somewhat difficult to find the oceciostomes
of the British species in colonies collected in the spring, in
which the formation of the ovicell is beginning; but it is easy
to find them in uninjured healthy colonies collected in the
summer.

The following table illustrates the use which can be made
of the oceciostome in distinguishing the species treated of in
this paper. The “lower” surface of the colony is that by
which it is fixed, so that “ upwards” means away from the
basal surface :

ON THE DEVELOPMENT OF TUBULIPORA. 85

at least in well-developed colonies 5 ne

[te more or less obviously arranged in connate series,
Zocecia not in connate series, even in the fertile lobes.

: Oceciostome with a well-developed tube, usually more

{

| or less free, the oceciopore being larger than an orifice

and looking upwards T. aperta (Pl. 8, fig. 2)

Oceciopore much larger than an orifice, opening upwards
or obliquely horizontally

T. plumosa, W. Thomps. (fig. 1)

Oceciostome or occiopore not larger or only slightly

larger than an orifice ; : tno

Tube of the occiostome recumbent on a zoccium or

series of zowcia, the oceciopore opening horizontally

or downwards : . 4

Tube of the occiostome free, eaecile ciranieeneeal the
oceciopore being slit-like

T. flabellaris, Fabr. (fig. 4)

Oceciopore larger than an orifice, opening horizontally

T. liliacea, Pall. (=T. serpens, auctt.) (figs. 7, 8)

Oceciopore concealed, smaller than an orifice, looking

downwards T. phalangea, Couch (figs. 5, 6)

The average size (in thousandths of a millimetre = ph) of
the oceciopores and of the orifices is stated in the following
table, the greater diameter of the oceciostome being added
in T, phalangea, the only species in which the occiopore is
narrower than the occiostome:

Oceciopores. Orifices.
T.phalangea. - 120 Shits aie . 160
T.flabellaris . . 165 2 75
T. liliacea . ‘ . 260 ; : : G5
Taperta, <9 . 280 : : 5 . 170
T, plumosa. : . 93855 185

It should be expressly noted that the ree of the oldest
zocecia of a colony may be much smaller than that of the
younger zocecia. The above averages are taken from zocecia
which have reached their full size.

86 SIDNEY F. HARMER.

II. History of the Species and Genus.

The tenth edition of the ‘Systema Nature’ gives the
diagnosis of certain species referred to the genus Tubulipora,
of which the first or type species is T. musica, the organ-
pipe coral. Amongst these (p. 790) occurs T. serpens.

The twelfth edition contains (p. 1271) a diagnosis of T.
serpens, as “T. tubulis cylindricis erectis brevissimis dis-
tautibus axillaribus, basi repente dichotoma divaricata.” This
is practically identical with that given in the tenth edition, to
which a reference is given, with the further reference ‘ Ameen.
Acad.,’ i, p. 105, t. 4, f. 26.

The species is thrown up on the shores of the Baltic, while
a similar but smaller form occurs in the Mediterranean, the
only locality mentioned in the former edition.

A reference to the first figure which is quoted by Linneus,
namely, to’ that contained in his ‘ Amecenitates Academic’
(vol. i, 1749), leads to the conclusion that the original descrip-
tion did not refer to the species which is now usually known
as Idmonea serpens. The figure is on a plate headed
“», 812,” and the description is on p. 209 [not 105]. The
figure, with which the description agrees, represents a stone
bearing a closely adherent species, consisting of an open
network of tubes with single pores at considerable intervals,
usually at the angles of the meshes. It is hardly possible to
recognise any similarity to any species of Tubulipora or
Idmonea; but, on the contrary, the figure is strikingly sug-
gestive of the Alcyonarian Sarcodictyon catenatum,
Forbes, and closely resembles the figure of that species given
by Herdman in the ‘ Proc. Liverpool Biol. Soc.,’ vol. ix, 1895,
pl. viii, fig. 2.

Whether the description of Linneus referred to Sarco-
dictyon or to an Alecto-like form must be left an open
question,! but I think that it can have had no connection with
any species of Tubulipora or Idmonea.

In 1755 (9, p. 74) Ellis described, under the name of the

1 Milne Edwards (29, p. 331) believed that it referred to an “ Aulopore,”

ON THE DEVELOPMENT OF TUBULIPORA. 87

‘small purple Eschara,” the form which is now com-
monly known as Idmonea serpens. The purple colour and
the parallel arrangement of the zoccia are expressly men-
tioned. The figures e and £ on pl. xxvii represent the species
as growing on a substance which is doubtless the stem of the
“sickle coralline”? (Hydrallmania falcata), as appears
from the description, on p. 75, of the Cellepora shown on
the same stem. The occurrence on this species of Hydroid is
eminently characteristic of the ‘small purple Eschara.”

In 1766 (33, pp. 248, 249) Pallas described the same species
under the name of Millepora liliacea, referring to Ellis’s
description and figures; and it is given as Millepora tubu-
losa in the well-known work of Ellis and Solander (10, p. 136).

In the enlarged thirteenth edition of the ‘ Systema Nature ’
(1788) Gmelin complicates the question by describing the
species under no less than three different names.! The first
of these (tom. 1, part 6, p. 3754) is Tubulipora serpens.
Linneus’s diagnosis is repeated, but a reference to former
editions of the ‘Systema’ is omitted, while the small purple
Eschara of Ellis, and Millepora liliacea of Pallas are
given as synonyms. The second (p. 3790) is Millepora
tubulosa, Ell. and Sol., the small purple Eschara of Ellis
appearing a second time as a synonym. The third (p. 3790)
is Millepora liliacea, Pall., and Tubulipora serpens,
Linn., is given as a synonym.

It appears to me that the Linnzan name must be rejected, and,
following the ordinary laws of priority, that the choice must lie
between Tubulipora liliacea, Pall. (1766), and T. tubulosa,
Ell. and Sol. (1788). Ifthe tenth edition of the ‘Systema Na-
ture’ is adopted as the commencement of the binomial system,
Pallas’s name has the right to be accepted ; while the adoption
of the twelfth edition as the starting-point would necessitate
the employment of T. tubulosa, Ell. and Sol. I shall follow
the example of Mr. Hincks® in regarding as valid Pallas’s

1 As has already been pointed out by Lamouroux (25), p. 66.
2 See his remarks on Flustra securifrons, Pall., on p. 122 of the
‘ British Marine Polyzoa’ (1880),

88 SIDNEY F. HARMER.

names, published in the year before the part of the twelfth edi-
tion of the ‘ Systema Nature’ which referred to Zoophytes.

I have come to the conclusion, from my study of the de-
velopment, that it is not possible to separate ‘‘Idmonea”
liliacea generically from the British forms recognised as
“Tubulipora,” although I claim no novelty in that con-
clusion. It thus becomes necessary to consider whether the
genus which includes the two species should be called Idmo-
nea or Tubulipora.

The genus Tubulipora was founded by Lamarck (23) in
1816, while Idmonea is due to Lamouroux (25), and dates
from 1821. Lamarck’s type-species is T. transversa, said
to be found on Fucus in the Mediterranean. Thesmall purple
Eschara of Ellis, and Millepora tubulosa of Ellis and
Solander, are given as synonyms, and from this, with the
diagnosis, it might be concluded that T. liliacea, Pall., is the
type-species of the genus Tubulipora. H. Milne Edwards
has, however, figured (80, pl. ix, figs. 3 and 3a) a specimen
from the Paris Museum with the statement (p. 218, note)
that it is the one from which Lamarck’s description was
taken. He regarded the species as an Idmonea (29, p. 382),
a course which is hardly justifiable considering that it was the
type-species of the earlier genus Tubulipora. If, then, we
are to accept Milne Edwards’ figures as a correct representa-
tion of Lamarck’s species, [dmonea becomes, on his showing,
a synonym of Tubulipora.

If a generic distinction between Tubulipora and Id-
monea, in the ordinarily understood sense, can really be
maintained, this is a regrettable conclusion, since it results in
the substitution of Tubulipora for Idmonea, and would
necessitate the use of some other generic name for the species
usually understood to belong to Tubulipora. If Lamarck’s
type-specimen is still in existence, and the evidence that it is
the type-specimen is satisfactory, I suppose there is no option
but to regard the synonyms which he himself gave for T.
transversa as erroneous. But as the evidence is perhaps
not quite certain, and as, moreover, it is not clear that any

ON THE DEVELOPMENT OF TUBULIPORA. 89

generic difference between Tubulipora and Idmonea can
be maintained, I shall regard the species described in this
paper as members of the genus Tubulipora.

The type-species of Lamouroux’ genus is Idmonea tri-
quetra, a fossil form from the “terrain a polypiers” (Ba-
thonian) of Caen. The description might lead to the inference
that the species is erect, but Gregory (12, p. 134) states that it
is always an encrusting form, and has justly remarked that it
is therefore impossible to define Idmonea as consisting only
of erect species. If this is so, it becomes very difficult to draw
any line between 'Tubulipora and Idmonea. Dr. Gregory’s
catalogue includes no species of Tubulipora, but he dis-
tinguishes (p. 134) a family Idmoniidz from the Tubuliporide
mainly by the existence of regular transverse rows of zowcia
in the former. I do not think that this distinction can be
maintained, either as the character of a family or even of a
genus. Lamarck’s type-species of Tubulipora was defined as
having its zoccia in transverse series, and this feature is
strongly marked in other recent species which are ordinarily
included in that genus. The only character in Dr. Gregory’s
diagnosis of _Idmonea (p. 184) which is not applicable to
many species of Tubulipora is the ridged or triangular cross-
section of the branches, and it is very doubtful if this is really
a valid generic difference.

From a superficial examination of the ovicells of Idmonea
atlantica, and from a consideration of Smitt’s description
and figures (40, p. 443, pl. iv, figs. 5, 7), it appears to me that
this form at least is closely allied to T. liliacea.

III. Synonymy, Diagnoses, and Accounts of the Species.

The material on which the following account was based was
collected mainly in the Salcombe Estuary, in South Devon, in
March and April. I have to thank Dr. A. M. Norman for
having recommended me to choose that place as a base of
operations. Other specimens were collected by myself while
working at the Plymouth Laboratory and on other parts of

90 SIDNEY F. HARMER.

the English coast, and in Norway. I have to express my
great indebtedness to my friends who have kindly given or
lent me specimens from other localities ; and particularly to
Professor Herdman and Miss Thornely for specimens from the
Liverpool district; to Professor M‘Intosh for material from
Scotland; to Dr.O. Nordgaard for specimens from Norway; and
to Dr. F. M. Turner for material from Guernsey. I must also
express my obligation to Mr. A. H. Church for determining
one or two seaweeds on which my specimens were found, and
for some observations on the amount of annual growth of the
colony; and to Mr. S. D. Scott for some observations on the
excretory vesicles.

Tubulipora, Lamarck.

Zoarium with a distinct basal lamina, adnate or erect, beginning as a pyri-
form or flabelliform colony, which may become lobed by the division of the
terminal membrane. Lobes short and adherent, or longer and dichotomously
divided once or more often, sometimes becoming erect. Zocecia with a free,
cylindrical, terminal portion ; or connate in obliquely transverse series, in
which they are separated by flat septa corresponding with the intersection of
two cylindrical zocecia. The series are arranged alternately on opposite sides
of the axial line of the lobe, but the transverse arrangement usually becomes
radial in the distal part of the fertile lobes. Ovicell an enlarged zocecium,
which extends into the intervals between the parallel or radial series.

The number of the tentacles is usually eleven or twelve in
the three species I have studied by means of sections. Of
these, T. phalangea and T. plumosa seem to have eleven
tentacles in most cases, Milne Edwards (80, p. 195, note),
giving the number as twelve for the former. In T. liliacea
I have counted twelve tentacles in most cases, a number agree-
ing with Dalyell’s statement (6, p. 86); but one polypide had
thirteen, and several had eleven.

T. liliacea, Pallas (figs. 7—9).

Tubulipora and Idmonea serpens, auctt. (not Tubipora serpens,
Linn. [27, p. 1271], nor Fabr. [11, p. 428).

Small purple Eschara, Ellis (9, p. 74, pl. xxvii, figs. e, ).

ON THE DEVELOPMENT OF TUBULIPORA. 91

Millepora liliacea, Pallas (33, p. 248).

Millepora tubulosa, Ell. and Sol. (10, p. 136).

Millepora tubulosa and M. liliacea, Linn. Gmel. (28, p. 3790).
Tubipora serpens, Dalyell (6, p. 85, pl. xviii, figs. 11—15).

Tubulipora serpens, Johnst. (19, pl. xxxi, figs. 4—6; and 20, pl. xlvii,
figs. 4—6). Couch (5, pp. 105 [part], 106). Smitt (40, pp. 399, 444,
[part], pl. iii, figs. 4a—4e, 5a, 54; pl. ix, figs. 1, 2a, 26). Busk (2,
pp. 25, 26 [part], pl. xxii, figs. 1—3).

Tdmonea serpens, Hincks (18, p. 453, pl. Ixi, figs. 2, 3). Levinsen (26,
p. 76, pl. vii, figs. 6—10).

Zoarium adnate or erect, its form being greatly influenced by the sub-
stance on which it is growing; commonly dividing several times dichoto-
mously. Zocecia curved for the most part in one plane, with the serially
connate, alternate arrangement strongly marked, though sometimes obscured
in small or irregular colonies. In well-branched colonies the inner zoccia are
much longer than the outer ones, so that the height of the transverse series
diminishes greatly in passing from the inner to the outer side. Ectocyst
usually vitreous and hyaline. Oceciostome about 260 » in diameter, slightly
larger than the orifice of a zocecium, opening horizontally.

Common on Hydroids (especially Hydrallmania falcata), from 20 to 40
fathoms ; but also found on Cellaria from the same depth, and on red sea-
weeds from shallower water.

This is the “small purple Eschara” of Ellis, and the
Tubulipora or Idmonea serpens of most writers. I have
explained on p. 86 the reasons for rejecting the familiar
specific name.

The distinctive feature of this species is the form of the
oceciostome (figs. 7 and 8). In size it is intermediate between
the corresponding structure of T. plumosa and that of T.
phalangea, and is somewhat larger than the orifice of an
ordinary zoccium. The difference between the oceciostome of
T. liliacea and that of T. phalangea (figs. 5 and 6) is not
always apparent at first sight, the oceciopore being often con-
cealed in both cases. But whereas in the latter species it is
seldom possible to see the oceciopore in any position in which
the uninjured colony may be placed, it is nearly always

92 SIDNEY F. HARMER.

possible in T. liliacea to see it by inclining the colony in a
suitable direction. The oceciopore typically opens hori-
zoutally ; or, in other words, its plane is vertical to the upper
or exposed surface of the ovicell. The edge of the oceciopore
may be slightly everted, so as to form a narrow brim, or there
may be no eversion; and the upper lip may be horizontal and
more prolonged than the lower lip.

Tn all cases the occiopore is relatively large, varying from
230 w to 270 pu, that of an ordinary zowcium being 145 p to
190 u. When the structure is placed in a suitable position it
is possible to see some way down the tube of the oceciostome
(fig. 7). This cannot be done in T. phalangea.

The tube is moderately long, and is recumbent on a zocecium
or series of zocecia. It may occur on the proximal side of the
series, and look towards the oldest part of the colony, or it
may be placed on the distal side and look towards the growing
edge.

T. liliacea is very common on certain Hydroids, parti-
cularly on Hydrallmania falcata, a fact familiar to many
of the older naturalists, and on Sertularella. It is easily
obtained from the masses of Hydroids brought up by trawlers
in water of twenty fathoms or more. The form of the entire
colony varies a good deal. It may remain closely attached to
the narrow stem of the Hydroid, its basal lamina curving
round the stem and thus giving rise to very irregular colonies ;
or it may remain attached merely by a small central area, and
grow out into free, erect branches, as in the var. radiata of
Hincks. In this condition it assumes a typical _ Idmonea-
form, having a very strongly marked alternate arrangement of
connate plates of zocecia.

In a particularly fine specimen of this form, from the
Trondhjem Fjord, which I owe to the kindness of Dr. Nord-
gaard, a single series consists in some cases of as many as
eight zoccia, the inner ones being very much taller than the
outer ones. The tip of the longest branch is about 11 mm.
from the centre of the colony. The occiostome belonging to
this branch is on the distal side of the seventh series (of one

ON THE DEVELOPMENT OF TUBULIPORA. 93

side of the branch) from the bifurcation preceding the ovicell ;
whereas in another branch the oeciostome is on the proximal
side of the eighth series of one side, the ovicell itself beginning
immediately after the fifth series.

The ovicells of this colony extend through a region of four
or five transverse series of zocecia on each side of the fertile
lobe. The shape of the ovicell is of course affected by the
strongly marked alternate arrangement of the series, and its
roof is thus a comparatively narrow, curvedly zigzag band,
running along the middle of the lobe, and giving off an inter-
serial lobe on the convex side of each bend. The ovicell may
thus be described as consisting of a regularly undulating axis,
with an alternately pinnate arrangement of simple lobes ex-
tending between the series of zoccia. The ends of the
branches are bifurcated, but the same ovicell extends into both
halves of the fork by division of its main axis. The same
arrangement is figured by Smitt (40, pl. iv, figs. 5 and 7) in
Idmonea atlantica.

A great contrast to this Idmonea-like colony was afforded
by a fine specimen from the Liverpool district kindly lent to
me by Professor Herdman. A narrow branch suddenly ex-
panded into a nearly semicircular fertile lobe, 6 mm. in trans-
verse diameter. The zoccia in this lobe had a Tubulipora-
like arrangement, consisting of radial series, which showed a
distinct tendency to become frayed out into separate zocecia at
their upper borders. The occiostome had the typical form,
and other colonies from the same locality were in no way dif-
ferent from the more ordinary type of T. liliacea.

T think there can be no doubt that the form of oceciostome
which I have described is quite characteristic of this species.
Although I first noticed it in a number of specimens from the
Plymouth district, it is not a local peculiarity, since I have
found precisely the same form in the specimens which have
just been alluded to from Trondhjem and Liverpool, as well
as in a series of colonies from Hydroids dredged in St.
Andrews Bay (20 to 27 fathoms), kindly given to me by Pro-
fessor M‘Intosh. The variations in the size of the ocwciopore

94 SIDNEY F. HARMER.

are indicated by the following list of measurements of the
transverse diameter :

Plymouth (on red seaweed) : : . 220 p.
St. Andrews (on Hydroid) ‘ ‘ . 230 p.
Plymouth (on Hydroid) ' ; . 260 pz.
Trondhjem . : ; - 265 p.
Plymouth (on Hiydroid) . . 280 p
Liverpool (probably on Hydroidy : - 300 x.

The average of this series of measurements is 260 yu.

T. phalangea, Couch (figs. 5, 6).

Tubulipora phalangea, Couch (5, p. 106, pl. xix, fig. 7 [figure bad]).
Johnston (20, p. 273 [part], pl. xlvi, figs. 1—4). Busk (2, p. 25,
pl. xxiii, fig. 2).

Tubulipora flabellaris, Hincks (18, p. 446, pl. Ixiv, figs. 1—3).

Tubulipora verrucaria and T. verrucosa, Milne Edwards (29, pp.
337, 328, 323, pl. xii, fig. 1).

Zoarium entirely adnate, variously lobed, sometimes consisting of a series
of divaricated lobes, sometimes almost circular in outline, and then reaching a
maximum diameter of at least 15 mm. In stunted specimens the terminal
membrane may not divide, but gives rise to a single small fertile lobe, the

whole colony being pear-shaped. Zocecia serially connate, the series alternate
near the base of elongated branches, but becoming radial in fertile lobes.
The series are commonly resolved above into their component elements, the
zocecia having a longer or shorter free cylindrical portion; but they may
remain entirely connate to their ends. Zocecia narrow and long compared
with those of most other species. Oceciostome (fig. 5) about as large, at its
widest point, as an orifice, averaging 150 p in diameter, the tube bent com-
pletely round, so that the oceciopore (fig. 6), which averages only 120 y in its
longer diameter, looks down on to the roof of the ovicell, and can rarely be
seen without dissection of the colony. The tube of the oceciostome is adnate
to a series of zocecia, and its upper exposed surface is convex. The primary
zoccium diverges from the plane of attachment to a greater extent than in
most species, and the proximal part of the colony is usually rather deep and
narrow.

Common (in Devonshire) on red seaweeds, shells, and stones from about
three fathoms to moderately deep water. I have seen one specimen from the
Outer Hebrides.

The reasons for maintaining that this species is distinct

ON THE DEVELOPMENT OF TUBULIPORA. 95

from T. flabellaris, Fabr., are given below, under the ac-
count of that species.

The occiostome of T. phalangea is shown in figs. 5, 6.
The convex upper surface is really the outer part of the wall
of the tube, which has the form of a NM, one limb of which
rises vertically from the roof of the ovicell, the other being
shortened and opening downwards. It results from this
arrangement that it is quite impossible to see the adult ocecio-
pore in the great majority of cases without breaking off the
series of zocecia which bears the tube, and turning it over until
the occiopore becomes visible. It is then seen to have the
form shown in fig. 6, being conspicuously smaller than that of
T. liliacea. The diameter of the occiopore varied from 110
to 135 yw in five specimens measured (average 118 yw), the
widest part of the entire tube varying from 110 to 180 pu
(average 139 mw) in the same specimens. The occiostome may
be quite symmetrical, or it may be distorted so that the occio-
pore looks obliquely downwards. When the occiostome is
typically developed (as in the great majority of cases), it differs
to a striking extent from that of any other species described in
this paper.

In its typically developed form this species is distinguish-
able by its very long and slender zoecia, the ends of which
are more commonly dissociated from their neighbours (and
are therefore completely cylindrical) than in T. plumosa,
with which it commonly occurs. The character of the oldest
zocecia is of some value in distinguishing the species. While
T. plumosa is usually depressed in the oldest part of the
colony, this species has rather the opposite tendency, and the
primary zocecium usually grows upwards at a considerable
angle from the plane of support, the interval being occupied
by the proximal ends of the next zoecia.

The general characters are well described by Couch (5, pp.
106, 107), though his fig. 7, pl. xix (wrongly given as fig. 8
in the text), is excessively bad. H. Milne Edwards (29, pl. xii,
figs. 1, 16, &c.) gives excellent figures of this species, accom-
panied by some anatomical details (pp. 323, &c.), under the

96 SIDNEY F. HARMER.

name of T. verrucaria, Fabr., accidentally given as T. ver-
rucosa in one place on p. 328. The name employed by Milne
Edwards cannot be retained, since the Madrepora verru-
caria of Fabricius is a Lichenopora. From an inspection
of the original description and figures I can see no sufficient
reason for believing that the fossil Diastopora plumula,
Reuss, is identical with the present species or with T.
flabellaris, Fabr., although Pergens (34, p. 9) considers the
specific name given by Reuss to be the correct name of
one of the forms to which the name T. flabellaris has
been ascribed.

T. phalangea is common in the Salcombe Estuary, at a
depth of 3 to 5 fathoms, on red seaweeds, where it occurs in
company with T. plumosa, and on dead shells. It is equally
common at Plymouth from 3 to 15 fathoms; and I believe"
that the greater number of specimens of Tubulipora found
on shells in the shallower parts of the Plymouth district
belong to this species. In the deeper water (20 to 30 fathoms)
a considerable proportion of the specimens may belong to the
form identified by Mr. Hincks as T. lobulata, Hassall; but
I am at present unable to express any positive opinion with
regard to Hassall’s species.

I have seen a typical specimen of T. phalangea, kindly
lent to me by Professor M‘Intosh, from the Outer Hebrides ;
but the rest of my material has been obtained from Devonshire.
T. phalangea is very variable in the form assumed by the
colony. It may consist merely of a single small, fertile
lobe, the whole colony being then pear-shaped, and closely
resembling the Obelia tubulifera of Lamouroux (26, p. 81,
pl. lxxx, figs. 7,8), a Mediterranean form with which it may be
identical. It may consist of a small number of well-separated
lobes, or it may have an almost completely circular outline.
Colonies of the last type may reach a diameter of nearly an
inch. The complexity of the ovicell varies greatly with the
size of the colony, the large colonies having very complicated
ovicells with numerous palmately arranged lobes extending
between the radial series of zocecia; fig. 5 being by no means

ON THE DEVELOPMENT OF TUBULIPORA. 97

an extreme case of this kind. The smaller colonies have
simpler ovicells, but this is also the case in other species.

My Salcombe specimens were dredged in March and April ;
and an examination of this material gives some hints with
regard to the meaning of the differences in size. A very iarge
number of colonies found on shells were small and pyriform,
although in many cases possessing a mature ovicell; while
other small colonies consisted of only two or three lobes.
These colonies were nearly all brown, and more or less en-
crusted with foreign matter. Here and there an old lobe had
recommenced to grow, and had given rise to a fresh and clean
lobe, whose brilliant white colour (in spirit specimens) forms
the most striking contrast with the older lobes. The colonies
of Tubulipora have in fact the power, which is probably
common to all Cyclostomes, of regenerating new zowcia from
various parts of the old colony (cf. 18, p. 141). In the species
now under consideration a small part of the edge of the old
colony here and there becomes active, so that a series of fresh
lobes, with narrow bases, may be seen growing out from
various parts. These lobes have in many cases acquired a
considerable size by the beginning of April, and have developed
mature ovicells. A few quite young, healthy colonies were
found amongst these specimens. It is probable that these
last were colonies which had recently commenced their ex-
istence, that the brown specimens belonged to a previous year,
and that the fresh lobes proceeding from them were entirely
recent growths. If this is a correct inference, it may be
suggested that the small brown colonies were produced late
in the year when the temperature was becoming low, so that
although they became mature so far as the external characters
of their ovicells were concerned, they were unable to grow large.
There was of course no doubt that the finely developed speci-
mens were actively growing when they were dredged.

I think it probable, therefore, that the difference in the
development of the entire colony may, in some cases at least,
be of a seasonal nature.

I may here refer to some interesting remarks which have

vou. 41, part 1.—NEW SER. G

98 SIDNEY F. HARMER.

been made to me by Mr. A. H. Church, whom I had consulted
on the growth of Tubulipora. Mr. Church informs me that
Rhodymenia ciliata, a red seaweed on which I obtained
nearly all the material (T. phalangea and T. plumosa)
which I have used for sections, is an annual. It follows,
therefore, that even the largest colonies (some 12—15 mm.)
found on this seaweed must represent the growth of one year.
Mr. Church informs me that R. ciliata usually dies in the
winter, the middle of which period may be regarded (for Algz)
as February, but that the specimens I dredged (in a sheltered
estuary) at the beginning of April must have grown in the
preceding year. I have not noticed processes of regeneration
in colonies growing on seaweeds, the evidence from which is
not entirely concordant with that from shells. The specimens
growing on Rhodymenia (and the same is true of specimens
of T. plumosa collected at the same season on Saccorhiza
bulbosa) show no apparent discontinuity of growth, nor were
any stunted mature colonies observed in this situation, The
colonies were probably far too large to have grown in the year
in which they were collected; and although the growth may
have been less active or dormant during the winter, there was
no interruption sufficient to give rise to a marked discon-
tinuity, as in the specimens growing on shells. It may, how-
ever, be noticed that there is no reason for assuming that
these latter were the growth of the year immediately preced-
ing. They may be evidence of unfavorable conditions more
than one year before, in which case the absence of similar
colonies on the Rhodymenia growing in the same locality
would be due to the fact that this plant is an annual. I
think it follows from the observed facts that growth may
start at any time when the species is breeding, and that a
colony which begins existence in the summer continues
to live through the winter, and produces ovicells in the
spring.

Mr. Church informs me that in some cases of regeneration
which he has observed (on a glass bottle) the central part of
the colony had decayed, leaving the new growths with a vacant

ON THE DEVELOPMENT OF TUBULIPORA. 99

space in the centre. I have had no opportunity of examining
cases of this kind.

Hincks (18, p. 447) has referred to a curious lobed Tubuli-
pora, about an inch in diameter, which he has met with in
Salcombe Bay. I have obtained some specimens which appear
to correspond with Mr. Hincks’s description. One or two of
these colonies were from the Salcombe Estuary, and had the
oceciostomes of T.,phalangea. Two others were sent to me
from Plymouth by Mr. Church, and had the occiostomes of
T. plumosa. The variety is a very curious one, and is
characterised by its nearly circular outline and by the great
crowding of the zoccia, the series being placed very close
together, and the lobes of the well-branched ovicells being
correspondingly narrow. On closer examination a consider-
able difference (no doubt specific) between the Plymouth and
the Salcombe specimens becomes apparent. A perfect colony
of the former measures about 15 mm. in diameter ; it is com-
posed of twelve well-marked lobes, and closely resembles the
form of T. plumosa which is found on Saccorhiza. These
lobes can be readily made out without any magnification,
whereas the Salcombe specimens do not appear obviously lobed
when examined with the naked eye.

I regard this variety as due to an excessive growth of the
edge of the colony, resulting, by the mutual pressure of the
lobes, in a crowding of the zocecia, and in the acquirement of
a circular form. This appears to me to be a further instance
of the tendency of different species of Tubulipora to assume
the same general form as the result of some unknown factors
in the environment.

T. flabellaris, Fabricius (fig. 4, described on p. 82).

Tubipora flabellaris, Fabr. (11, p. 430).
Tubulipora flabellaris, Smitt (40, p. 401, pl. ix, figs. 6—8).
? T. flabellaris, Levinsen (26, p. 76, pl. vii, figs. 1—38).
Zoarium entirely adnate, more or less fan-shaped in form in well-developed

specimens. Stunted colonies may occur, as in the preceding species. Some
of the zocecia are free, others are in connate series, which are more or less

100 SIDNEY F. HARMER.

developed, and become radial in fertile lobes. Ovceciostome consisting of a
greatly compressed tube, whose oceciopore is slit-like. The longer diameter
of the slit averages about 165 » (130—180 pz), being equal to or somewhat
less than the diameter of an orifice. The tube is not recumbent on a series
of zocecia, but stands up freely from the roof of the ovicell, its two narrow
edges being placed in a radius of the colony.

This seems to be an essentially Northern species, and I have no evidence of
its occurrence in British waters. I have examined specimens from Greenland
(the locality from which the type-specimens came), Barents Sea (50 fathoms,
growing on Cellularia peachii), and Hammerfest ; and I have also obtained
what I believe to be a young form of this species from Godésund, Bjorne Fjord,
Norway. The specimen figured (fig. 4) is a stunted, somewhat abnormal form
of this species from Barents Sea (see the description on p. 82), but it well
illustrates its characteristics by showing five perfectly typical oceciostomes.

Professor Smitt (40, pp. 400—402, 454, 458), who is fol-
lowed in this respect by Mr. Hincks, regards T. phalangea of
Johnston as identical with T. flabellaris, although he believes
the original T. phalangea of Couch to be identical with T.
lobulata, Hassall. It appears to me that the characters
of the oceciostome are amply sufficient to separate T. pha-
langea from T. flabellaris. I first became acquainted with
the oceciostome of the latter in two colonies from Hammer-
fest, kindly sent to me by Dr. Nordgaard. The collection of
Polyzoa given to the University Museum of Zoology by Miss
E. C. Jelly contained a moderate number of specimens on sea-
weed from Greenland. The form and size of the oceciostome
were identical in these and in the Hammerfest specimens.
The colonies from Greenland, although very small and
stunted (one of them, with only twenty-seven zocecia, possess-
ing a complete ovicell), did not resemble the Barents Sea
specimens from deeper water in possessing a multiplicity of
ovicells.

It can hardly be doubted that the Greenland specimens, sent
by Miss Jelly, belong to Fabricius’ species. The flabelliform
shape, the connate series of zocecia, the occurrence on sea-
weed, and the small size (given by Fabricius as 13 lin. in trans-
verse diameter), all agree closely with the original description.
The species has been more fully described by Smitt, who gives
excellent figures. In two of these (40, pl. ix, figs. 6, 8), re-

ON THE DEVELOPMENT OF TUBULIPORA. 101

presenting specimens from Spitzbergen, the peculiar form of
the flattened oceciostomes, with their radially arranged flat
sides, is indicated, while the flattened form is expressly men-
tioned on p. 457. I have not seen specimens so finely de-
veloped as that shown by Smitt in his fig. 6, in which the
radial serial arrangement of the zocecia is strongly marked (as
many as twenty being stated to occur in one series). The
difference between Smitt’s specimens and those examined by
me may, however, be seasonal, as suggested under the last
species.

It may be remarked that Smitt’s conclusion that T. pha-
langea is asynonym, partly of T. lobulata, Hass., and partly
of T. flabellaris, Fabr., does not appear to have been based
on an examination of actual specimens of the first-named
species.

Tubulipora aperta, n.sp. (figs. 2, 3).

Tubulipora fimbria, Smitt (40, pp. 401, 452, pl. ix, fig. 5).
? Tubulipora fimbria, Levinsen (26, p. 75, pl. vi, figs. 45—50).

Zoarium entirely adnate, pyriform, flabelliform, or lobed. Zocecia not
serially connate, or only exceptionally united in very short series. Ectocyst
with few pores. Oceciostome about 280 » in diameter, larger than an orifice,
more or less funnel-shaped ; the oceciopore opens upwards, and is circular or
oval. Tube of the oceciostome usually more or less free, and diverging from
the zocecium on which its base is recumbent, the edge of the oceciopore often
resting on the wall of another zoccium. Accessory openings sometimes
present at the ends of the lobes of the ovicell.

Common on the fronds of Laminaria saccharina in Norway. My
largest colony is 5°25 mm. in transverse diameter.

This species, which I believe to have hitherto received no
distinctive specific name, has been described and figured by
Smitt under the name of T. fimbria. This name, applied by
Hincks to T. plumosa, Thomps., was given by Lamarck to
an immature specimen of Tubulipora, of which the locality
is not recorded, and I give my reasons for not accepting it
on page 107. The name aperta is suggested in reference to
the wide occiopore, which is usuall clearly visible from above,

102 SIDNEY F. HARMER.

and is not concealed either by the zocecia or by other parts of
the oceciostome. The specimens on which my account of this
species is based were found principally at Godésund, a small
island off the north of Tysnas6, at the entrance to the Bjorne
Fjord in Norway. They were not uncommon on the fronds of
Laminaria saccharina, where they occurred in company
with Lichenopora verrucaria, Fabr. Most of the speci-
mens collected at the end of June had fully developed ovicells.
Specimens collected at Lervik, in the Hardanger Fjord, at the
same period in a previous year were, however, not provided
with ovicells. Smitt describes the species as occurring on
Laminarie and other Alge from the Gullmar Fjord (in the
south of Sweden) to Spitzbergen, so that the species may fairly
be regarded as a Northern one.

I think there can be no doubt of the specific distinctness of
this very beautiful form, which is more easily recognised, at all
stages of its growth, than are most of the European species of
Tubulipora. It differs strikingly from the other species here
described in having its zocecia isolated, or only associated two
or three together. The connate arrangement found in other
forms is usually completely absent; and even in the fertile
lobes the zocecia stand up for the most part singly from the
roof of the ovicell, the lobes of which may thus unite distally
to the zoccium. Even in these cases the suture or septum
between the two ovicell-lobes is easily seen near the growing
edge, andI have no reason to suppose that contiguous lobes
really fuse at any time.

The zocecia are more clearly marked off from one another in
the basal part of the colony than is the case in most species,
the proximal part of a zocecium being very convex as far as the
suture where it joins another zoccium. In other species
parts of the zocecia may form more or less extensive flat surfaces,
as is well seen in the connate series or in the basal adherent
parts of the colony. The ectocyst of T. aperta has a delicate
appearance, and there are noticeably fewer pores on the zocecia
than in most other species. Numerous concentric lines of
growth, passing transversely across the zoccia, are clearly

ON THE DEVELOPMENT OF TUBULIPORA. 103

marked, as Smitt points out. The zoccia are relatively large,
averaging about 175 mw in diameter.

The oceciostome more nearly resembles that of T. phumosa
than that of the other species here described, but it differs
from it in the fact that the basal region of the tube, which is
perforated by pores, is commonly partially free (fig. 3), whereas
in T. plumosa, as in T. phalangea and T. liliacea, the
porous part is wholly adnate to a zocecium, and only the non-
porous termination of the tube is free. The occiopore typically
opens upwards, or away from the basal surface of the colony.
The proximal part of the tube is attached to a zocecium, and is
followed by a considerable free length of tube (part of which is
porous), which dilates upwards, so as to be more or less funnel-
shaped. This funnel-shaped portion commonly grows away
from the first zocecium and towards the free part of a second
zocecium, on which its edge rests. In T. plumosa the whole
of the tube is recumbent on the same zocecium or zocecia. In
some specimens of T. aperta, and especially when the tube is
wedged in between two zocecia which are near one another,
the arrangement above described is not so obvious. The owcio-
pore is circular or oval, and is of considerable size, its longer
diameter usually varying from 200 p» to 370 pw, and averaging
about 280 ». One abnormally small oceciopore was only 135 p,
and one unusually large one was 435 mu, but the usual range is
given by the figures first stated. The edge of the occiopore
may be slightly everted, or evenly circular or oval. In some
cases it is somewhat indented on one or both sides, and may
then closely resemble that of T. plumosa.

In one or two colonies the proximal part of the exposed
portion of the ovicell had the characters of an ordinary zoe-
cium, the pores being very few and widely scattered. At the
level where the ovicell commences to enlarge, in these cases,
the pores become suddenly numerous, and the zocecium takes
on the character of an ovicell. The line where the pores
become more numerous probably corresponds with a time at
which the embryonic structures reached a certain stage of
development. The same peculiarity in the ovicell was also

104 SIDNEY F. HARMER.

noticed in some of the proximal ovicells of the specimens of T.
flabellaris from Barents Sea described on p. 82.

I have observed one or two interesting variations in this
species. In two or three colonies, zoccia of double the
normal width occur, with an orifice measuring as much as
380 in its major axis, but having the normal diameter in its
minor axis. In one of these cases a groove down the large
zocecium indicated that its size was due to the absence of the
septum which should normally have divided it into two zocecia.
A similar large zocecium was found in a colony of T. flabel-
laris from Greenland. I have not seen more than one of
these giant-zoecia in a single colony, and they do not occur at
all in most cases. It is not impossible that they may be
zocecia which made an abortive attempt to develop into ovi-
cells.

In one instance an oceciostome was recumbent on a giant
zocecium (cf. the remarks given in Waters’ paper, 45, p. 277),
but there was no similar relation in the other cases. Walford
(42, p. 80) has characterised his genus Cisternifera (said to
be Cheilostomatous) by the occurrence of giant-zocecia in the
colony, and has in the same place (p. 79, pl. v, figs. 14, 15)
described similar giant-zoccia or “ cistern-cells”’ in “so-called
Diastopore” from the Great Oolite. These latter, at any
rate, appear to me to correspond with the large zoccia of T.
aperta and T. flabellaris.

A more interesting variation was noticed in two colonies in
which an ovicell had developed accessory occiostomes (fig. 2).
Since the ovicell of Tubulipora is an enlarged zocecium, its
oceciopore is homologous with a normal orifice, and it follows
that an ovicell should have only one occiostome. ‘This is
actually the case in most species, and in most colonies which I
have seen of T. aperta. When the growth of the ovicell is
completed, each of the lobes other than that ending in the
oceciostome normally becomes closed by a porous calcareous
film. In the abnormal colonies of T. aperta, however, one or
more of these lobes has not closed completely, but has grown
at its distal end into a tube, which may completely simulate

ON THE DEVELOPMENT OF TUBULIPORA. 105

the normal occiostome, which can be seen in the more proxi-
mal part of the ovicell.

One of these cases is a small pyriform colony (fig. 2), be-
ginning with the usual primitive disc, which measures 185
in diameter. One ovicell is present, with a normal ocio-
stome, its tube (fig. 3) being long and to a large extent free,
and the oceciopore measuring 265 uw. To the apparent right
of the oceciostome one of the lobes of the ovicell ends in an
accessory oceciostome (fig. 2.1), 105 w in diameter; while on
the other side three lobes have accessory oceciostomes. Of
these, No. 2 measures 165 uw; No. 3 is not yet fully formed ;
and No. 5 is 90 win diameter. A smaller tube (4), at the end
of a very small ovicell-lobe, had closed except for a minute
terminal pore. ;

A second case is that of a portion of a larger colony. The
fragment is rather more than 2 mm. in transverse diameter,
and has five ovicell-lobes, each of which has an accessory ocecio-
stome, a normal one being present more proximally. One
of these closely resembles the normal occiostome, while
another has about the same diameter at its base, but then
becomes much constricted, finally opening by an orifice much
smaller than a normal orifice.

I have not had time to examine sections of this species, but
it appears to me highly probable that the accessory occio-
stomes are functional in providing a means of escape for the
larve which find their way into the lobes to which they
respectively belong.

Tubulipora plumosa,' W. Thompson (fig. 1).

Tubulipora plumosa, W. Thompson, in Johnston (20, p. 274 [immature]).

Tubulipora phalangea (part), Johnston (20, p. 273, especially the state-
ments given on the authority of Mr. Peach).

1 Thompson’s name first appeared in 1847. In the same year Reuss
(“‘ Foss. Polyparien d. Wiener Tertiarbeckens,” in Haidinger’s ‘ Naturwiss.
Abhandl.,’ Band ii, Abth. 1, 1848, p. 39) described Defrancia pluma and
(p. 51) Diastopora plumula. The former, at any rate, is doubtless a
Tubulipora, and is given as a member of this genus by Manzoni (‘ Denkschr.

106 SIDNEY F. HARMER.

Tubulipora flabellaris, Johnston (20, p. 274, pl. xlvi, figs. 5, 6 [imma-
ture]). Busk (2, p. 25 [part], pl. xxiv, figs. 1—3; pl. xxv, fig. 2).
? Tubulipora flabellaris, Busk (8, p. 28, pl. v, figs. 1, la—le).

The locality given in the text is Station 315 [Falkland Islands], and
the specimens figured presumably came from that locality. On this
account I regard the determination of the species as uncertain, in the
absence of any description of the oceciostome.

Tubulipora fimbria, Hincks (18, p. 448, pl. Ix, figs. 3, 3a [immature]).

Zoarium completely adnate, variously lobed, reaching a diameter of an inch
in the finest specimens. The serial, connate arrangement of the zocecia is
strougly marked, the series becoming radial in fertile lobes. Zocecia large.
Oceciopore much larger than an ordinary orifice, averaging about 355 p in its
greatest diameter. The tube is funnel-shaped, and the oeciopore opens
directly or obliquely upwards, one of its edges being usually somewhat
indented. Tentacles containing (homogeneous) excretory vesicles (see p. 113).

Common in the cavities and on the outside of the remarkable nodular
rooting bulbs of Laminaria (Saccorhiza) bulbosa, where it reaches its
greatest size, and on red seaweeds from shallow water.

I have examined large numbers of specimens of this species from Salcombe,
Devonshire, and from Plymouth. I have also obtained typical specimens
from Guernsey and from Roscoff.

This well-marked species appears frequently in the literature
of the subject in its immature form, in which condition it has
been supposed to be adult. Its real adult condition has been
presumably mistaken for T. liliacea (T. serpens, auctt.)
by most writers; but it is clearly alluded to by Johnston (20,
p- 273), who quotes Mr. Peach as the authority for stating
that it luxuriates in the bulb of Laminaria bulbosa, and
that it reaches the diameter of nearly an inch. It is there
given as a form of T. phalangea, while on the next page
Johnston describes the species in its young state as T. fla-
bellaris.

These specimens were sent to Johnston as T. plumosa by
K. Acad. Wien,’ xxxviii, Abth. 2, 1878, p. 20, sep. copy) ; it has, moreover, a
considerable resemblance to Thompson’s species. The specific names pluma,
plumula, and plumosa are similar in meaning, but they are sufliciently
distinct in sound. I think there is no practical inconvenience in reviving
Thompson’s name, and in a group where the synonymy is so involved it is

advisable to retain any name that can possibly be used when it is moderately
certain what was originally meant by it,

ON THE DEVELOPMENT OF TUBULIPORA. 107

Mr. W. Thompson, whose name I have adopted. The de-
scription and figures given by Johnston leave no doubt in my
mind that Thompson’s T. plumosa is the form of which I
have given a diagnosis above. The flat central region of the
colony, which alone is developed in Johnston’s figs. 5 and 6
(pl. xlvi), is eminently characteristic of the present species.
This form is indeed not invariably assumed by it, but is very
distinctive when it does occur. In this condition the zoccia
are often strongly ridged transversely, in the manner de-
scribed in Johnston’s work, and the serial arrangement of the
zoccia is hardly apparent unless the colony is looked at from
its distal edge. The shortness of the tubes alluded to by
Johnston is of course due to the fact that they are im-
mature. °

Specimens with this depressed central region are usually
easy to distinguish from T. phalangea, with which the species
is associated on red seaweeds. The radial ridges formed by
the more projecting zocecia are the commencements of the
series of zoccia. As the colony increases in diameter, the
freshly added parts of these projecting zocecia project more
and more, younger zoccia are developed connately with them
on their lower side, and the serial arrangement becomes as
strongly marked as in any other species of Tubulipora.

T. plumosa is described by Hincks (18, p. 448, pl. Ix,
figs. 3 and 3a) from immature specimens as T. fimbria, Lamk.
Fig. 3 of this author shows the beginning of an ovicell on
the right side, although the septa between the ovicell and
the contiguous zoccia are not all indicated. Mr. Hincks
expresses a doubt as to the identity of this species with
T. fimbria, Lamk., but accepts the name with some hesi-
tation on Smitt’s authority. Smitt’s T. fimbria is, however,
in my opinion, a different species, and it is here described as
T. aperta. The figure of T. ‘“fimbria,” given by Milne
Edwards (29, pl. xiv, fig. 2) from a specimen labelled by
Lamarck, does not specially resemble T. plumosa,

The distinctive feature of T. plumosa is the owciostome
(fig. 1), which is a wide, funnel-shaped structure, looking up-

108 SIDNEY F. HARMER.

wards or obliquely upwards, the oceciopore being conspicuously
larger than an orifice. Its tube is recumbent on a series of
zocecia, and does not often become completely free, as is nor-
mally the case in T. aperta. The shape of the occiopore is
usually complicated by a slight fold or indentation of one side,
as shown in fig. 1, or of both sides. The size of the oceciopore
varies from 270 p to 430 pw, the average being 355 wu.

T. plumosa reaches its largest size in the cavities of the
bulb-like bases of Laminaria (Saccorhiza) bulbosa. One
of the finest colonies I have seen was approximately circular,
with a diameter of 21 mm., and its peripheral parts were
formed of sixteen separate lobes, some of them bifurcated.
These colonies are attached to the seaweed by remarkable
ridges on their under side,! similar to those which have been
figured by Waters (44, pl. vii, figs. 2,3) in an Australian form,
termed by him T. fimbria, forma pulchra, MacG., and by
Busk (8, pl. v, fig. 1.0) in T. “flabellaris.” The ridges do
not project greatly, and their general arrangement is radially
dichotomous. As the colony increases in diameter, the ridges
increase in length, often dividing or undergoing a cessation of
growth. When the ridge is re-formed it commonly grows in
such a way as to prolong the direction of its more proximal
portion. In one case a ridge was observed measuring 1°36 mm.
without interruption, but the ridges are usually much shorter
than this. If the ridge is narrow its axis may be occupied by
a single row of pores, which correspond with the ordinary
zocecial pores, but are considerably larger. If the ridge is
broader the number of rows of pores is correspondingly in-
creased, and large areas of the lower side are sometimes in
the form of broad, irregular, porous areas, which radiate out
into narrower ridges towards the distal part of the colony.
In some cases the ridges are broken up into short pieces, only
large enough to contain a few pores, sometimes only a single
pore. If the base of the colony is, by reason of an irregularity
in the surface of the seaweed or for any other cause, separated

1 The colonies can be conveniently removed from the seaweed by prolonged
boiling with caustic potash.

ON THE DEVELOPMENT OF TUBULIPORA. 109

from its base of support, these short ridges may grow out into
short foot-like columns of support.

The lobes of this form of T. plumosa are often narrow,
with nearly parallel radial sides. In well-grown colonies the
parallel sides of adjacent lobes are commonly in close proximity,
and they may then unite by means of these attaching pro-
cesses, which grow towards one another and fuse. This may
occur distally, but not proximally, so that narrow fenestre
are left between the united lobes, and these are most easily
seen from the back of the colony. A colony whose lobes have
thus united may have an almost completely circular edge, and
may at first sight appear to have an undivided terminal mem-
brane. ‘That this is not the case can usually be seen by careful
examination, and the original lobes are indicated not only by
continuing to develop their zoccia in alternate, connate series,
but also by a slight undulation of the growing edge, due to the
fact that the distal border of each lobe has a convexity which
has a shorter radius than that of the entire colony. Somewhat
similar colonies of this species growing on shells have been
described above (p. 99).

I have not found these attaching ridges or processes on
specimens of this species growing on other seaweeds, nor have
I found them in any other species. It appears, therefore,
possible that the ridges are, from some cause, a special adapta-
tion connected with the occurrence of T. plumosa on this
particular seaweed.

T. plumosa appears to be a shallow water form. Its
fondness for Saccorhiza bulbosa has been already referred
to. It is dredged at Plymouth in great numbers on Cysto-
seira granulata from a few fathoms depth. On this sea-
weed the colonies are small and irregular in their growth,
owing to the small diameter of the stems round which they
grow. I obtained numerous specimens, which were much
more convenient for examination, growing with T. phalangea
on Rhodymenia ciliata from the Salcombe Estuary. The
flat surfaces of this seaweed induce a regular growth in the
Tubulipora, which are thus well adapted for the preparation

110 SIDNEY F. HARMER.

of sections which it is desired to obtain in given planes. My
study of the development of this form has been entirely based
on material obtained from this source.

I have found T. plumosa on the empty carapace of a large
crab (Cancer pagurus), but I have not obtained many spe-
cimens which I could certainly refer to this species on shells
or stones. Its typical habitat may be considered to be the
surface of seaweeds from shallow water, but some of the
specimens of Tubulipora from shells in deeper water may
also belong to the same species. Should it, in fact, prove
identical with T. lobulata, Hass., this name would have
priority over T. plumosa.

IV. The Nature of certain Vesicles found in the Tentacles and
other Parts, with Remarks on the Structure of the Adult
Zocecium and on the Budding.

The account of the occurrence of these vesicles must be pre-
ceded by some statements with regard to the development of
the polypide-buds and the structures connected with the orifice
of the adult zocecium.

a. Budding and Structure of Orifice.

There are no really detailed accounts of the budding of
Cyclostomes, though some information is given by Ostroumoff
(32).

I have not satisfied myself with regard to the manner in
which the first rudiment of the bud is derived from the ter-
minal membrane. The young bud (fig. 23) is bounded exter-
nally by a more or less solid mass of cells of some thickness,
which may be regarded as the protoplasmic part of its terminal
membrane, and is probably in the main ectodermic. Within
this, two or three excretory vesicles appear at a very early
stage in all the three species investigated. More internally is
a deeply staining two-layered mass of cells, at first forming a
hollowed plate, concave externally, but soon taking the form
of the well-known vesicular polypide-bud of Ectoprocta. In
correlation with the fact that the proximal ends of the zoecia

ON THE DEVELOPMENT OF TUBULIPORA. 111

are narrow and pointed, the younger stages of the bud are
longer than is usually the case in the Gymnolemata; and a
longitudinal section of the entire mass of protoplasmic struc-
tures appears as a sharply acute-angled triangle, filling up the
narrow pointed tube which is at present the only representative
of the future zocecium.

It may be assumed, on the analogy of other cases,! that the
inner layer of the vesicular bud is ectodermic, and the outer
layer mesodermic. The distal part of the bud gives rise to the
tentacle-sheath, into which the tentacles project; while the
proximal part develops into the remainder of the polypide, in
much the same manner as in other Ectoprocta.

Immediately on the distal side of the two-layered polypide-
bud, which can be distinguished by the great readiness with
which it takes up colouring matters, there appears at an early
stage a cavity, lined by a thin layer of cells, which ultimately
gives rise to that part of the introvert which les between the
calcareous “orifice” of the zoecium and the ‘ diaphragm”
which forms the distal end of the tentacle-sheath. This cavity
is figured and described by many writers on the Ectoprocta.
It is shown by Nitsche (81, pl. xxxv, fig. 2) and by Prouho
(36, pl. xxiii, figs. 1 and 3), and it appears to be of general
occurrence throughout the Ectoprocta. Davenport (7, and
elsewhere) having termed the cavity of the tentacle-sheath the
“atrium,” this space, with its wall, will be alluded to in
future as the ‘vestibule,’ without thereby implying any
exact homology with the similarly named part of Polyzoon
larvee. I have previously figured the vestibule of a young
Cyclostome-bud (15, pl. xxii, fig. 1), although the space in
question was then erroneously described as the tentacle-sheath.

In some cases the young vestibule appears as an invagina-
tion of the ectoderm of the terminal membrane, but it is often
difficult to obtain direct evidence on this point.

The vestibule of the adult zoccium is usually a space of
considerable length, the diaphragm being a constriction be-
tween it and the tentacle-sheath, as originally described by

' Cf. especially Seeliger (38).

112 SIDNEY F. HARMER.

Nitsche in Membranipora membranacea. The vestibule
is lined with a flexible cuticle, which can be made out in
longitudinal sections through an adult orifice as a collapsed
tube continuous externally with the chitinous part of the ter-
minal membrane of the zocecium, and internally with the dia-
phragm, through which the tentacles are protruded.

In a living, perfectly healthy colony, each zoccium is seen
to be closed by a delicate terminal membrane, perforated by a
minute hole at its centre, the membrane and its perforation
being stretched across the extreme end of the calcareous
zocecium. A similar membrane closes the young ovicells, and
appears to be stretched continuously over the edges of the
calcareous septa which are forming the walls of new zoccia at
the growing margin of the colony. The size of the perforation
in the membrane of the adult zocecium can be altered in a way
which suggests the alterations in the diameter of the pupil of
the eye, though I have not succeeded in demonstrating the
mechanism of the movements. Before protrusion of the ten-
tacles the pupil, or in reality the opening of the introvert, is
widely dilated, and the tentacles are pushed through it.

The perforated terminal membrane appears to be similar to
what Jullien (21, p. 38) has described as the ‘“‘irisoidea”’ in
Cheilostomes (Microporella malusii), although Pergens
(35, p.509) has adversely criticised Jullien’s results on this point.

In colonies which are still alive, but less healthy, the ter-
minal membrane may no longer appear flush with the surface
of the zocecia, but may be sunk down to some little distance
from the orifice (fig. 29), and be deeply concave distally or
upwards. There is little doubt that this is the result of an
unhealthy condition; but it is important to notice that the
position of the terminal membrane can vary in the tube be-
cause in sections and in colonies which have been mounted
whole the distal ends of the zocecia are often completely empty
of cellular structures for a distance equal to once or twice the
diameter of the zoccium. The same phenomenon is noticed
in the ovicells, in which the growing edge of the living body-
wall is usually at some distance from the edge of the cal-

ON THE DEVELOPMENT OF TUBULIPORA. 1138

careous part. I regard these as post-mortem changes due
to the action of reagents, and I think it probable that in the
healthy zocecium or ovicell the distal end of the body-wall is
in the immediate neighbourhood of the edge of the calcareous
wall which it secretes.

The pupil-like opening seen in the terminal membrane of
living colonies is doubtless the external opening of the vesti-
bule. When the membrane is in its proper place it is probable
that the body-wall of this region is of no great thickness ; but
in the more usual retracted condition, which is probably arti-
ficial, the irisoid (adopting Jullien’s name), or in other words
the terminal membrane of the zoccium, generally appears as a
thick mass of nucleated protoplasm.

The young zoccium with its polypide-bud has been seen to
consist of the following parts (cf. 15, pl. xxii, fig. 1) :—an ex-
ternal terminal membrane (irisoid), a vestibule, and the two-
layered vesicular bud. Further examination has shown, how-
ever, that the outer layer of the bud is reflected at its distal
end into a thin membrane (fig. 23, s.m.), which encloses the
whole bud, a cavity occurring between it and the bud. In
older polypides the alimentary canal, the retractor muscles,
and the reproductive organ, whether ovary or testis, lie in this
space, which is the body-cavity. The reflected layer seen in the
young bud may probably be regarded as the somatic mesoderm.
This perhaps throws some light on the morphology of the
cavity lined by the inner layer of cells of the secondary embryo
(cf. 15, p. 228, pl. xxiv, fig. 23). The inner layer applies itself
closely to the dorsal ectoderm in the region which will give
rise to the primary polypide bud. If that bud were developed
by an invagination of the entire dorsal wall, and the shape of
the cavity of the inner layer were simplified by the eversion of
the sucker, the cavity in question would closely resemble what
has just been described as the body-cavity of the bud.

6b. Excretory Vesicles.
The most casual inspection of sections of the three species
of Tubulipora which I have specially studied reveals the
vou. 41, part 1.—NEW SER. H

114 SIDNEY F. HARMER.

presence of remarkable brown vesicles in various parts of
the polypides and ovicells (see figs. 20, 22), provided that the
material was prepared with certain reagents. As I have,
moreover, found these vesicles also in decalcified preparations
of T. aperta, it is highly probable that they are of normal
occurrence in a portion of the genus at least, and perhaps
throughout the whole genus, and that they play some important
part in the physiology of the colonies. What that part may
be is somewhat doubtful. I bave hesitated whether to consider
them as reserve-stores of nutritive material or as excretory
bodies; but the balance of evidence appears to me to incline
towards the latter view.

I first found the vesicles in sections of material which had
been prepared with corrosive sublimate. The vesicles are
with this preparation excessively resistent, and therefore retain
their form and their yellowish-brown colour in sections of
colonies which have been decalcified with nitric acid, embedded
in paraffin, and stained with hematoxylin or borax carmine.
They are found principally in one of three places: (a) beneath
or in the terminal membranes of the zocecia and buds; (0)
beneath or in the terminal membrane of the ovicell; (c) in the
tentacles. They occur in the first two situations in T. plumosa,
T. liliacea, and T. phalangea; and in the first, at least, in
T. aperta, of which I have not examined sections. So far as
I know at present the occurrence of the vesicles in the tentacles
is almost restricted to 'T. plumosa; and so constant have I
found this character (during the spring months) that I have
used it as a means of distinguishing sections of this species
from those of T. phalangea. In one case, however, excretory
vesicles occurred in the tentacles of a specimen of T. pha-
langea, determined as such by the characters of its oceciostome.
The colony contained an old ovicell (early stage G), but most
if not all of its polypides were degenerating, and the whole
colony was unusually heavily charged with excretory vesicles.
It thus appears that the vesicles may occur in the tentacles of
T. phalangea under certain circumstances; but I think it is
none the less true that in the early spring, on the coast of

ON THE DEVELOPMENT OF TUBULIPORA. 1S

Devonshire, the tentacles of T. plumosa normally contain
these vesicles, and those of T. liliacea and T. phalangea
do not normaliy contain them.

I spent some time at the Plymouth Laboratory during the
spring, for the purpose of endeavouring to decide the nature of
these vesicles. Material was unfortunately not plentiful, but
some results were obtained. The most convenient way of
studying the vesicles was to dissect out the fresh polypides of
T. plumosa, and after separating the tentacles from one
another to examine them in sea water. Although there is
some difficulty in dissecting out the polypides, decalcification
cannot well be employed, because the vesicles are at first
readily altered by reageuts.

The vesicles of T. plumosa do not occur in the lumen of
the tentacles, as might at first be supposed; but they are
situated in the external epithelium on the abaxial side of the
lumen (fig. 26). In specimens obtained in the early spring
no trace is seen, in fresh material, of the typical vesicles found
in sections, but the epithelium contains a row of greenish
refractive vesicles, which are commonest near the tips of the
tentacles, but may also occur more proximally. These vesicles
are not bounded by any distinct membraue, but appear rather
as drops of a fluid substance contained in the epithelium.
Other smaller granules of yellowish pigment (p.) occur here
and there in the epithelial cells, but these have a more solid
appearance, and are not affected by reagents in the way
characteristic of the larger vesicles.

In sections, on the contrary, the vesicles appear to be
bounded by a sharply defined membrane ; and it was therefore
necessary to consider whether they might be symbiotic Alge.
With the view of testing for starch and cellulose, I added a
solution of iodine in potassium iodide to the fresh tentacles,
with somewhat surprising results. The vesicles, which were
at first more or less irregular in outline, immediately rounded
themselves off and became darker in colour, an active Brownian
movement at once becoming apparent in their interior (show-
ing that they are really fluid). This lasted for a brief period,

116 SIDNEY F. HARMER.

at the end of which (fig. 27) each of the homogeneous greenish
vesicles took on the appearance of the sharply contoured
spherical bodies seen in sections, a certain number of brownish
granules being precipitated in their interior, and usually coming
to rest on the inside of the wall of the vesicle, the fluid con-
tents of which are now colourless.! The subsequent addition
of strong sulphuric acid gave no blue colour, and did not
greatly alter the vesicles. These changes took place in the
same way in vesicles which had been freed from the polypides
during the dissection.

The greenish vesicles were found in the fresh material in
the tentacles (in T. plumosa alone) and beneath the terminal
membranes, i.e. in precisely the same places as those in which
they had been previously noticed in sections.

The above-described reaction was due to the iodine and not
to the potassic iodide. This was shown by the fact that a
solution of potassic iodide (a) in distilled water, (6) in sea
water, was added to other specimens without producing any
effect. On adding a fragment of iodine to either solution the
characteristic precipitation at once took place. The same
result was produced by adding sea water in which a fragment
of iodine had been rubbed up.

It soon became apparent, however, that iodine was not alone
in precipitating the contents of the homogeneous vesicles.
The same result was produced by the following reagents :—
distilled water, solution of corrosive sublimate (in fresh water
or sea water), strong ammonia, osmic acid.

The darker (browner) colour assumed by the vesicles in the
first experiment is not merely the effect of staining by the
iodine, since it occurs with other reagents, even with distilled
water. In all cases the colour of the homogeneous vesicle at
once becomes darker, and an appreciable interval occurs
during which nothing else can be made out. The Brownian
movement then commences, the granules being at first invisible
individually. The vesicle soon contains numerous minute

1 The vesicles are very similar to those which have been figured by Durham
(Quart. Journ. Mie. Sci.,’ xxxiii, pl. 1, fig. 3) in Spatangids.

ON THE DEVELOPMENT OF TUBULIPORA. 117

granules in active movement. These become larger, and on
coming to rest form the granules seen even in sections in the
interior of the vesicles. One fresh vesicle always gives rise to
one precipitated vesicle.

The amount of the precipitate produced by various reagents
is not identical. Ammonia gives rise to a specially copious
precipitation, the vesicles now looking very dark, and some-
times dissolving after a prolonged action of the ammonia.
The addition of strong potash now broke up the precipitated
granules, with some indication of solution, during which a
purplish colour appeared. Osmic acid similarly gives rise to
a very dark-coloured precipitate ; but when added to vesicles
which have recently been precipitated by corrosive sublimate,
it merely slightly darkens their granules.

In most of the above-described experiments no trace was
seen in the fresh material of precipitated vesicles, but these
were observed in one or two specimens. It is, I think,
probable that precipitation of the granules takes place during
life in old vesicles ; but it seems clear that the young stages of
these bodies consist merely of a homogeneous fluid.

While the above reagents produced precipitation of the
vesicles, others dissolved them completely. 90 per cent.
alcohol at once dissolves them without precipitation, and the
same effect is produced by 30 per cent. alcohol, which does
not affect the yellow granules above described in other parts of
the tentacles. The vesicle is indicated as an empty space in
the tentacle after the addition of alcohol. Nitric acid dis-
solves the fresh vesicles, giving rise to a red-brown mass.
Dilute acetic acid has no effect.

When the vesicles have been precipitated, their behaviour
towards reagents is entirely different from that of fresh
vesicles. Alcohol, even weak alcohol, readily dissolves the
latter, but it has no effect on vesicles which have been pre-
cipitated with corrosive sublimate, as is obvious enough from
the fact that they are seen in sections.

The vesicles of some sublimate-material which had been
kept in 70 per cent. spirit for seven or eight months were ex-

118 SIDNEY F. HARMER.

traordinarily resistent. After decalcification of the colonies
with dilute nitric acid the vesicles were practically unaffected
by the following reagents :—sulphuric acid (dilute and pure),
nitric acid (even when the strong acid is heated), strong
ammonia, 4 per cent. potash, ether, chloroform, benzole.
Gmelin’s test for bile-pigments was tried without any results,
and the murexide test for uric acid was also negative.

The freshly precipitated vesicles, on the contrary, were less
resistent. Thus, after precipitation with distilled water, 90
per cent. alcohol had some effect on them, while a 4 per cent.
solution of potash at once dissolved them. After the action
of corrosive sublimate (15 minutes), followed by distilled water
(25 minutes), even strong potash had practically no effect.

In some cases, though not in all, ammonia added to the
vesicles, freshly precipitated by distilled water, first turned them
purplish, and then dissolved them. Certain pigmented granules
or cells observed in the terminal membranes of the zocecia of
T. plumosa also turned purplish on adding distilled water.

T. phalangea and T. liliacea.—Although the tentacles of
these species do not usually contain homogeneous vesicles,
they possess in their places structures which consist of a large
number of minute vesicles (fig. 28). Each compound vesicle
probably corresponds with a single cell, and its reactions are
different from those of the homogeneous vesicles, which occur
in the terminal membranes of the same species, as in T. plu-
mosa. The homogeneous vesicles are precipitated by various
reagents, and then take on the form seen in sections, whether
they occur in the tentacles (T. plumosa) or in the terminal
membranes. The compound vesicles of T. phalangea and T.
liliacea under no circumstances give rise to what I may call
the “ formed ”’ excretory vesicles.

The tentacles of T. phalangea, as of other Cyclostomes I
have examined, contain pigmented granules in various parts of
their external epithelium. ‘lwo of these (p.) are seen in the
distal part of the tentacle shown in fig. 28. They do not give
the reactions characteristic of the other structures.

The fresh compound vesicles of the tentacles of T, pha-

ON THE DEVELOPMENT OF TUBULIPORA. 119

langea are dissolved by strong ammonia, by 4 per cent.
potash, by 90 per cent. alcohol, and by nitric acid. Osmic acid
also destroys them.

They are not affected by corrosive sublimate, nor are they
broken up by acetic acid. After being fixed with corrosive
sublimate, they are no longer affected by osmic acid, nor by
distilled water.

On adding distilled water the fresh compound vesicles lost
the boundaries of their constituent vesicles, and turned a
diffuse red-purple, that colour becoming soon restricted to
numerous granules, which disappeared on adding ammonia.
The homogeneous vesicles of the terminal membrane were
precipitated by distilled water; some of these (probably the
more delicate ones) were dissolved by ammonia, while others
remained and turned purple inside.

Several facts in the above experiments suggest that there is
some connection between the well-known purple colour of
certain species of Tubulipora and the homogeneous or com-
pound vesicles. The change from a greenish to a purplish
colour was specially marked on adding distilled water to the
compound vesicles, and it also occurred in the homogeneous
vesicles, under certain circumstances, as the result of treat-
ment with ammonia or potash. The purple colour is, how-
ever, not entirely due to the action of reagents. The branches
of T. liliacea, for instance, are commonly purple during life,
the colour occurring as a purple pigment in the cells of the
terminal membrane, as correctly stated by Smitt (39, p. 22),
and in other parts. This purple pigment was not affected by
the successive action of distilled water, ammonia, and potash,
nor by acetic acid.

It seems to me probable, however, that the purple colour
often seen in dry preparations of species of Tubulipora may
in many cases be due to post-mortem changes of the excre-
tory vesicles. This view was confirmed by an observation on
a colony of T. liliacea which, when alive, had no trace of
purple, but assumed that colour after being washed with fresh
water and dried. The colour is not seen in colonies preserved

120 SIDNEY F. HARMER.

in spirit (which dissolves the vesicles), nor in those treated
with corrosive sublimate (which precipitates the homogeneous
vesicles) ; but colonies which have been allowed to dry, after
washing with fresh water, are commonly of a characteristic
purple colour. This colour is usually not noticeable during
life in T. plumosa, healthy colonies of which are yellowish ;
but it is conspicuous in specimens which have been dried
without being put in spirit.

I have found very pale greenish, homogeneous vesicles in
Diastopora patina in the brown bodies, and apparently in
the body-cavities. These were precipitated by sublimate,
ammonia, or iodine in potassic iodide; but the details of the
reactions were not altogether similar to those in Tubulipora.
I have also seen structures somewhat resembling the com-
pound vesicles of T. phalangea in the tentacles of a Stoma-
topora, perhaps 8. major.

Experiments made with indigo-carmine, carminate of am-
monia, and Bismarck-brown on living specimens of Tubuli-
pora gave only negative results. The vesicles did not appear
to take up any of these pigments, although control-experi-
ments, made with species of Bugula, gave results similar to
those I had previously arrived at (14). The cells which take up
indigo-carmine in B. flabellata are the same as the “ gelbe
Tropfen ” described by Claparéde (4, pl. viii, fig. 1s, ¢.). Since
these are normally of a yellow or yellowish-green colour, they
are not unlike the homogeneous vesicles of Tubulipora.
They are not precipitated, however, by distilled water, iodine,
sublimate, or ammonia; and taking this fact in conjunction
with the difference in their behaviour to indigo-carmine, it
may be concluded that they do not closely resemble the vesicles
of Tubulipora.

The result of the previous reactions seems to be that there
is some substance in solution in the homogeneous vesicles
which is precipitated by any reagent (not being a solvent of
that substance) which alters the density or constitution of the
solvent. The action of distilled water may be due to osmosis
from the vesicle, while: it-is: possible that substances like cor-

ON THE DEVELOPMENT OF TUBULIPORA. 121

rosive sublimate have some specific action on the substance
itself. The precipitate, once formed, behaves to reagents
quite differently from its antecedent state in solution in the
vesicle. Alcohol, for instance, dissolves the contents of the
fresh vesicles without precipitation; but the precipitated con-
tents become entirely resistent to alcohol, except immediately
after their formation by distilled water.

Different colonies vary a good deal in the extent to which
the vesicles are developed ; but those obtained in the summer
appear to have a much greater development of the vesicles than
those of the same species found in the spring. I have exa-
mined summer specimens in preserved material only ; but in
many of them the excretory vesicles are very much more
conspicuous than in the spring colonies. ‘Their walls become
very dark, and may be so thickened as to greatly diminish the
lumen, while the contents may appear as a single solid con-
cretion. The number of the vesicles may often be very greatly
increased in the summer. This was observed especially in
T. plumosa, in which summer specimens had almost the
entire length of the tentacle occupied by a single row of
closely packed vesicles.

These facts seem to indicate that the vesicles are not nor-
mally discharged to the exterior, although it would not appear
difficult for their fluid contents to escape through the epi-
dermis of the tentacles. In many of my preparations ex-
cretory vesicles appear to have been forced to the outside of
the epithelia which normally cover them. I am inclined to
regard this as an artefact, the fluid vesicles being squeezed out
as the result of the contraction induced at death by the action
of reagents, and being precipitated outside the body by those
reagents. This is indicated by the fact that I have never seen
them on the morphologically outer side of the ectocyst of the
vestibules of the zocecia or of the terminal membrane of the
ovicell.

The increase of the number of the vesicles in the later part
of the year is a reason for regarding them as excretory, and
this view is confirmed by their dark colour and by the con-

125 SIDNEY F. HARMER.

cretion-like contents which may appear within them. Their
occurrence in the terminal membranes of buds (even of young
ones) as well as of the polypides and ovicells might suggest
the view that they were nutritive.

Davenport (7, pp. 34—, pl. vi, figs. 54, 56, 57, 59) has
described certain mesoderm cells (which have also been ob-
served by Braem) in the budding regions of Paludicella.
These contain refractive bodies which are not altogether unlike
the vesicles which I have described above, and they are re-
garded by both Braem and Davenport as nutritive. It appears
to me that these bodies are not comparable with the vesicles
of Tubulipora; but there is enough appearance of similarity
to make it worth while to call attention to the resemblance.

I can hardly doubt, however, that the Tubulipora vesicles
have a close similarity to the “ Exkretblaschen” which have
been described by Eisig (8, pp. 725—) in Capitellide. The
reactions of these bodies, and particularly the resistent way
in which some of them behaved to mineral acids and potash,
led Eisig to the belief that they contained chitin as one of
their constituents, and that this substance was to be regarded
as one of the normal nitrogenous excreta. The reactions
which I have described agree so well with Hisig’s results that
the view that the vesicles are of an excretory nature appears
to me to be greatly strengthened thereby. If the insoluble
substance which is precipitated by various reagents is really
chitin, or some similar body, the normal homogeneous vesicles
may perhaps be considered to contain a substance which could
be called chitinogen, which, though itself soluble, readily
passes into an insoluble state. The vesicles might conceivably
be employed in giving rise to the chitinous parts of the ter-
minal membranes or vestibules of the zocecia, although their
number would appear disproportionately large on this hypo-
thesis. It is perhaps more likely that their occurrence
in the tentacles and terminal membranes is due to a ten-
dency to deposit the excretory substances in the peripheral
parts of the colony, and to the advantage of removing waste
products of metabolism from the polypide-buds and embryo-

ON THE DEVELOPMENT OF TUBULIPORA. 123

phores, where specially important processes of growth are
taking place.

It would be interesting to know the fate of the vesicles when
a brown body is formed. I have not as yet obtained any
satisfactory evidence on this point, the colonies found in the
spring having been almost completely devoid of brown bodies.
It is not improbable that the vesicles found in the tentacles, at
least, are in some way got rid of when a brown body is formed.

V. Description of the Development.

The ovicell of Tubulipora is in its development, in certain
respects, intermediate between that of Crisia and that of
Lichenopora, a conclusion which might be expected from a
consideration of the adult characters of the three genera.
The account given in my preliminary note (17) of “ _Idmonea
serpens” really refers mainly to what is here termed Tubu-
lipora plumosa, though I now find that the sections on
which that account was based were partly prepared from T.
phalangea.

The following account refers indifferently to T. plumosa
and T. phalangea if no species is mentioned. Where state-
ments are made with regard to T. phalangea, it must be
understood that the evidence for the specific determination
was, unless the contrary is stated, the absence of excretory
vesicles in the tentacles. This negative evidence is not so
satisfactory as the positive evidence, afforded by their presence,
that a given specimen belongs to T. plumosa; since it is
possible to overlook in the sections vesicles which are not very
numerous. It may be noted that the material used for the
study of the development of these two species was derived
almost entirely from a single haul of the dredge which brought
up Rhodymenia ciliata, on which T. plumosa and T.
phalangea alone were discovered.

The specimens of T. liliacea were from Hydroids in deeper
water at Plymouth, and the species was determined before
they were decalcified,

124 SIDNEY F. HARMER.

The account of the development will, as far as possible, be
described as a series of stages corresponding with those which
I have formerly described in Lichenopora (16, p. 99).
Those stages were characterised partly by the degree of
development of the embryo and embryophore,! and partly by
the condition of the fertile polypide. This latter degenerates
at an earlier stage in Tubulipora than in Lichenopora,
and the correspondence between the developments is therefore
not exact.

The stages selected for descriptive purposes are thus as
follows:

Stage A. Formation of the definitive egg (fig. 10).

Stage B. Division of the egg, and degeneration of the fertile
polypide (figs. 11—18).

Stage C. [Not represented in Tubulipora.]

Stage D. Formation of the embryophore. The fertile
zocecium is still cylindrical, or slightly expanded distally, and
its brown body becomes closely surrounded by a distinct cel-
lular investment (figs. 14—18, 21).

Stage E. The investment of the brown body becomes vacuo-
lated, so as to give rise to a cavity. The ovicell expands dis-
tally (figs. 19, 20,:22, 25, 32).

Stage F. Commencement of embryonic fission.

Stage G. Fully formed ovicell (figs. 1—8, 38).

Stage A.—Formation of the Definitive Egg.

The eggs are developed at a very early stage by the polypide-
buds, as in Lichenopora and Crisia. This precocious
occurrence of the eggs appears to be in some way correlated
with a colonial habit, and cases of this kind are familiar to
every student of colonial Ascidians and of Hydroids. The

1 The term ‘‘embryophore” will be used below to denote the structures, in
relation with the primary embryo and its derivative secondary embryos, which
are on the proximal side of the vestibule. The principal contents of the
ovicell are thus the embryophore, containing the embryo and the fertile brown
body, the vestibule, and the terminal membrane.

ON THE DEVELOPMENT OF TUBULIPORA. 125

phenomenon has been described in Cheilostomes (Bugula) by
Claparéde (4, p. 166).

The eggs appear as part of the outer or mesodermic layer of
the polypide-bud. In some colonies they are very numerous.
Thus in one case (T. plumosa) a young bilobed colony con-
tained twenty-three individuals in which an egg or eggs had
become unmistakable. Some of these were fully formed
polypides, others very young buds, the youngest belonging to
a stage when the bud consisted of a two-layered cell-plate
which had not yet become vesicular. All stages intermediate
between these extremes were represented.

In most of the zocecia there is only a single egg ; in six cases
there are two, and in one case three eggs; the position being
at or near the proximal end of the polypide-bud or the cecum
of the polypide, as the case may be. Each egg (whatever the
number in a zocecium) is surrounded by a very distinct follicle
(fig. 10) of cells, in which a follicle-cavity is usually visible.

There can be no doubt of the normal occurrence of eggs in
the manner above described, since they are repeatedly noticed
in the sections. I think, moreover, that there is clear evidence
that the eggs occur especially in young lobes. When a lobe
possesses an ovicell belonging to one of the later stages,
although eggs may not be altogether wanting, they cease to
be the conspicuous feature that they form in young lobes
during the breeding period. It is thus probable that the suc-
cessful development of an egg, with the resulting modification
of the fertile zocecium into an ovicell, diverts the energies of
the lobe from the production of fresh eggs to the nutrition of
the embryos, although new lobes may be formed from the
same colony and give rise to new ovicells.

The mature egg reaches a diameter of 24 uw or even 28 uv
(T. plumosa), and this measurement is larger than those pre-
viously recorded for Crisia (17°6 u) and Lichenopora
(16 »). It possesses a distinct germinal vesicle and germinal
spot, and it commonly has a paranuclear body lying in its
protoplasm. This body might possibly be a male pronucleus,
against which is to be set the fact that the nucleus of the egg

126 SIDNEY F. HARMER.

shows no sign of alteration; or it might be a centrosome, a
view which hardly appears likely, from the fact that it stains
readily with hematoxylin. I have not discovered the nature
of this body.

In stage B (fig. 13) I have seen what I believe to be sperma-
tozoa in the follicle cavity; but I have not observed the ferti-
lisation of the egg.

I have never found a testis in an ovigerous zocecium, contrary
to what may happen in Lichenopora. The testes appear in
the majority of zoccia, and usually in all those which produce
no eggs, in the same position as the ovaries of the female
polypides; the colony being thus monecious. The young
testis is distinguishable from the young ovary owing to the
fact that it consists of a small group of nuclei at the proximal
apex of the cecum of the polypide. The testes grow con-
currently with the polypide to which they belong, and they
occur at all stages of the development of the ovicell. Owing
to the great size they reach, they form a very conspicuous
feature of sections, and can readily be seen in colonies mounted
whole. They fill up the proximal end of the zoccia with a
great mass of developing or mature spermatozoa, and they may
reach the length of 670 u (T. plumosa). When the testis is
mature, the ripe spermatozoa may be seen extending up the
side of the polypide, and may even be massed between its
tentacle-sheath and body-wall. I have not ascertained the
mode of escape of the spermatozoa.

The testes are thus produced in the great majority of the
zocecia, while ovaries are developed in but few, and an embryo
in an extremely restricted number. This seems to make it
probable that cross-fertilisation takes place, the great number
of spermatozoa that are produced being probably in some way
discharged into the surrounding water before fertilisation is
effected.

The occurrence of large testes had no relation to the age of
the corresponding ovicell; whilst in Lichenopora (16, p.
127) there was evidence of the disappearance of the testes
with increasing age of the ovicell. The apparent difference

ON THE DEVELOPMENT OF TUBULIPORA. 127

between the two genera may be due to the fact that in
Lichenopora the ovicell dominates the entire colony, whereas
in Tubulipora new fertile lobes can be developed in colonies
which possess an old ovicell (fig. 1). It must further be noted
that the specimens of Tubulipora were collected early in the
year, and therefore early in the breeding season.

Stage B.—Division of the Egg, and Degeneration of the Fertile
Polypide.

The number of polypides which become actually fertile is a
strictly limited one, as in Lichenopora. The bilobed con-
dition so often characteristic of the young colony is usually
correlated with the development of two ovicells. These are at
first ordinary zoccia, as in Lichenopora; and in T. flabel-
laris and T. aperta the proximal end of the ovicell not un-
commonly has the external characters of a zocecium, the
increased number of pores which denote the ovicell beginning
in some cases suddenly when the zocecium has reached a
certain length. But in all species it is easy to see, by looking
down into the young ovicell at a time when it is commencing
to expand, that the dilated part of the ovicell is continuous
with a prismatic or pyramidal cavity which runs proximally
into the general series of zocecia. A convenient way to demon-
strate the fact that the proximal end of the ovicell is morpho-
logically a zocecium is to stain a colony which possesses a
young ovicell without decalcifying, and to embed it in paraffin.
By scraping away the lower or basal wall of the colony, and
then dissolving out the paraffin and mounting whole, a view of
the lower surface can be obtained without having the details
obscured by the relatively thick, calcareous, basal lamina.
In preparations of ovicells of a suitable age made in this way
(fig. 80) the young embryophore may be demonstrated in a
part of the ovicell whose floor is the basal lamina, and in fact
ina cavity which in no way differs from an ordinary zoccium.

The fertile zocecia of Lichenopora are usually differentiated
at an extremely early stage in the life of the colony. In

128 SIDNEY F. HARMER.

Tubulipora the entire colony cannot be looked upon as an
individual of the third order to the same extent as can that of
Lichenopora; but a certain amount of individuality may be
recognised in each of the ovicell-bearing lobes. The fertile
zocecium is, in fact, differentiated at a very early stage in the
development of a lobe. This may be understood by reference
to fig. 1, in which a young ovicell is beginning to develop at
the right of the figure. The development of the first ovicells
of the colony begins in T. plumosa in the immature stage
which was considered by Thompson and Johnston as the adult
condition of this species.

The phenomena of the selection of the fertile zocecia are
probably more primitive in Tubulipora than in Licheno-
pora, in which the differentiation of the ovicell may have
been thrown back to an early stage in the development of the
colony, in correlation with the high degree of individuality
which is possessed by the entire colony in that genus.

Internal evidence that the ovicell is at first a zocecium is
afforded by the invariable presence in it of a brown body,
indicating the previous existence of a polypide. This may be
called the “ fertile brown body,” as in the case of Licheno-
pora.

In the latter the fertile polypide is not the first inhabitant of
its zocecium, as is indicated by the simultaneous presence in it
of a brown body, a functional polypide, and an embryophore.
In Tubulipora, on the contrary, the first polypide of the
fertile zocecium becomes fertile, and no new polypide is deve-
loped in the normal ovicell after the first brown body is formed.
In cases of abnormal development, however, the entire embryo-
phore may degenerate, and a new polypide-bud may make its
appearance. In a single case only I have found a brown body
(in addition to the normal brown body) on the proximal side
of an embryophore in stage E. This may be interpreted as
evidence of the degeneration of a polypide before the formation
of the definitive fertile polypide.

It is not easy to obtain evidence showing the exact stage
at which degeneration of the fertile polypide takes place in

ON THE DEVELOPMENT OF TUBULIPORA. 129

Tubulipora, and this is probably because the stage is of
short duration. Two possibilities have to be considered,— (a)
that degeneration takes place while the polypide is still a
bud ; (4) that it occurs after the polypide has become functional.
The latter view is probably the correct one, and if this is so,
Tubulipora occupies an intermediate position in this respect
between Crisia, in which the polypide degenerates while it is
still a two-layered bud, and Lichenopora, in which two
functional polypides successively occupy the young ovicell.

The question clearly turns on the observation of the stage
at which the egg begins to develop. No trace of development
is found in the eggs of polypide-buds, and on the assumption
that the brown body is formed by the degeneration of a bud,
it would be necessary to assume that the large eggs which are
so commonly noticed in the ovaries of polypides have missed
their chance of developing, and would later have degenerated.
Positive evidence that the fertile brown body is formed from a
polypide, and not from a mere bud, is afforded by the fact that
the youngest embryophores found with a brown body and
partially developed embryo occur in fully formed, long zocecia,
and not merely in immature zocecia still in process of develop-
ment at the growing edge.

Even more direct evidence is, however, afforded by the stage
shown in fig. 11 (T. plumosa), which represents what is
certainly a case of the division of the egg. The darkly stained
bodies marked f probably represent degenerating eggs or
their follicles. This would indicate that in the event of more
than one egg occurring in a single ovary, only one of the eggs
actually develops. The figure shows the cecum of the fertile
polypide, which from its size, and from the avidity with which
its tissues have taken up hematoxylin, was clearly only just
mature. Its rectum contains Diatoms, a fact which proves
that the polypide had commenced to feed. A second, pre-
cisely similar polypide, containing Diatoms, and provided with
an egg in the same condition as that of the first specimen, also
occurred in the same colony. Most of the buds and young
polypides in this colony, if not all of them, possessed either

vou. 41, part 1.—NEW SHRIES, I

130 SIDNEY F. HARMER.

one or two eggs, and one unmistakable ovicell was also
present ; no other development had taken place.

This is the only colony in which I have obtained certain
evidence of the period at which the egg begins to develop ; but
taken in connection with the next described series of cases,
the conclusion may be drawn that the degeneration of the
fertile polypide normally takes place shortly after the polypide
has become functional. This is in accordance with the fact
that ovaries or eggs are not found in the older polypides of a
colony, indicating that they are in some way absorbed or
degenerate if development does not begin in them shortly
after their parent polypide has become mature.

The occurrence of more than one polypide in the fertile
zoecium of Lichenopora is associated with the fact that
each of the ordinary zocecia of the young colonies also pos-
sesses a brown body and a polypide. This is not the case in
Tubulipora, in which there are normally no brown bodies
(in the species investigated) in the young colonies. It is not
impossible that this may have something to do with the season
at which my material was collected. My specimens were
obtained during March and April, a time when growth is
taking place with great energy, probably after a period of
winter rest. I have found brown bodies commonly in colonies
of Tubulipora obtained in Devonshire during the summer
months. It is not impossible that the excretory vesicles
which have been described above provide a means of eliminat-
ing excretory matters which would otherwise accumulate in
the stomachs, and induce the formation of brown bodies.
Whether this be the case or not, the entire absence of brown
bodies in a large proportion of the specimens of Tubulipora,
even in quite large colonies, is a fact in striking contrast with
the normal occurrence of brown bodies in the young zoccia of
Lichenopora.

Evidence of the degeneration of a fertile polypide was
obtained in ten cases, four of which appear to belong to
T. plumosa and six to T. phalangea. The results of the
examination of these cases were quite concordant, all of them

ON THE DEVELOPMENT OF TUBULIPORA. kl

pointing to the degeneration of a polypide at a stage when the
embryo consists of about 2—4 blastomeres, at first enclosed
in the original follicle of the egg, the follicle ceasing to be
distinct in the later stages. In two of these cases additional
evidence that the degeneration was a normal stage in the
development of the ovicell was afforded by the fact that it was
in the position relatively to an obvious ovicell which it might
have been expected to occupy, from the fact that two ovicells
are so commonly found in young colonies. Fig. 12 shows the
beginning of the degeneration of the fertile polypide. The
section cuts the edge of the mass of tentacles, which in the
next few sections are obscurely defined, showing that degenera-
tion is taking place. Parts of the alimentary canal are visible
in the section figured. The nuclei of the follicle are not so
regularly arranged as in the case of unfertilised eggs. The
embryo has two cells, which lie in a distinct follicle-cavity.

Fig. 13 (T. phalangea) shows a later stage, drawn with a
higher power. The brown body, which is only indicated in the
figure, still shows some traces of the parts of the alimentary
canal of the fertile polypide, whereas a fully formed fertile brown
body shows no such traces, and consists of a granular pigmented
mass, containing a few nuclei. ‘The embryo certainly possesses
three blastomeres, while the elongated nuclei of two of the cells
of the egg-follicle are clearly seen. The follicle contains a
minute deeply stained body, which is almost certainly the
remains of a spermatozoon.

The terminal membrane of the fertile zocecium plays an
important part in the development of the future ovicell. The
evidence afforded by this and the next stage appears to be
that the orifice of the vestibule closes at the degeneration of
the polypide, and that the terminal membrane retires some
little distance within the calcareous orifice. It is, however,
difficult to be certain how far this regression is normal, and
how far it is an artefact.

[Stage C (functional polypide+embryophore) of Licheno-
pora is not represented in the development of Tubulipora.]

132 SIDNEY F. HARMER.

Stage D.—Definitive Formation of the Embryophore.

The appearance of the ovicell in this stage is very charac-
teristic, and from the frequency with which it occurs in my
sections I conclude that the stage is of relatively long dura-
tion. I have examined about seventeen ovicells in this stage
of T. plumosa, twenty-seven of T. phalangea, and one of
T. liliacea. It appears to me that the embryophore of the
first species is normally distinctly larger than that of the
second, while that of the third species differs from the other
two in its great length.

Fig. 15 shows an embryophore of T. plumosa at the be-
ginning of this stage. The brown body is fully formed, but
appears young, and it is not yet surrounded by any very
definite investment of cells.) The embryo has two obvious
blastomeres, and the follicle-cavity in which they lie is in the
immediate neighbourhood of the brown body. The cord of
cells running proximally from the follicle is the shrunken
remains of the somatic mesoderm of the fertile polypide. The
terminal membrane is thickened, and its staining properties
indicate that it is in a state of active growth. It is continuous
internally with a mass of cells which extend from it to the
brown body. This mass contains four excretory vesicles,
though a larger number were visible in some of the other sec-
tions of the same ovicell.

The length of the embryophore, from the proximal end of
the follicle to the distal tip of the terminal membrane, is
200 pu.

Fig. 14 (T. phalangea) illustrates the condition in which
the terminal membrane is usually found in this stage. The
distal end of the embryophore projects into the cavity of the
young ovicell as a knob, which appears to be quite free from
the wall of the ovicell. The brown body is fully formed, and
has a more compact appearance than that of the former speci-
men.

Fig. 18 is a slightly older embryophore of T. plumosa,
from a colony in which excretory vesicles were specially

ON THE DEVELOPMENT OF TUBULIPORA. 1338

numerous. In the section drawn they form a large axial
mass, lying in the solid tissue on the distal side of the brown
body. The details of this tissue are not shown, but it does
not differ in any essential respect from the similar region of
fig. 16.

Two embryonic cells are seen, the two large nuclei shown
to the right of the follicle-cavity probably belonging to the
original egg-follicle. While in the younger ovicells shown in
figs. 14 and 15 the follicle is in the immediate neighbourhood
of the brown body, it is here separated from it by some inter-
vening tissue, the existence of which, and the occurrence of a
definite cellular investment to the brown body, are indications
that the embryophore is more advanced in its development
than was that of the former figures.

The investment of the brown body has in fact become a
perfectly definite structure, sharply marked off from the sur-
rounding tissues. The part of it which lies distally to the
brown body has become thickened, and the nuclei are here
arranged in several layers, while they occur in a single layer
at the sides of the brown body. Distally to the thickened
part is a slit-like cavity, which will be termed the vestibule.
The morphology of these parts will be considered in the final
part of the paper.

Fig. 16 is a still later stage (T. phalangea), as is shown
by the more advanced development of the embryophore and of
the embryo. The latter now consists of a considerable number
of cells, still lying in a follicle-cavity. The embryophore is
rather longer than before, and the investment of the brown
body is now thickened laterally and proximally, as well as
distally. A special growth of the investing tissue is taking
place proximally to form the “nutritive tissue,” destined
hereafter to form the reticulum in which the secondary
embryos lie. As it appears probable that this reticulum is
directly or indirectly concerned in the nutrition of the rapidly
growing mass of embryos, the term here suggested may con-
veniently be used for descriptive purposes. The investment
of the brown body in fig. 16 shows some signs distally of

134 SIDNEY F. HARMER.

being continuous with the vestibule. The terminal membrane
is deeply invaginated medianly ; and this is in fact a general
feature of ovicells in this stage. In sections which are not
median the invagination may not be seen; and in some cases,
as in fig. 18, it is obscured by the great development of the
excretory vesicles. The existence of the invagination may,
however, be looked on as the general rule; and it is probable
that the vestibule is really continuous with, and has been
developed by invagination from, the terminal invagination,
though it is not easy to demonstrate this continuity. The
wall of the vestibule, during the beginning of stage D, is
closely surrounded by the other tissues of the solid distal half
of the embryophore, so that the whole of its limits cannot, in
most cases, be made out with certainty. Excretory vesicles
were but slightly developed in this ovicell (fig. 16).

Fig. 17 is a somewhat oblique longitudinal section of the
distal half of an ovicell in about the same stage as fig. 16.
Part of the median invagination of the terminal membrane is
seen as a slit-like cavity. The section shows the investment of
the brown body, and the way in which it is connected with
the vestibule, the junction being constricted like the neck of
a flask. This is a normal arrangement, as is also (in T.
phalangea at least) the oblique position of the neck of the
flask-like connection.

Fig. 24 (Pl. 10) represents an ovicell of T. plumosa in
an unusual condition. While the embryo and the proximal
half of the embryophore are in stage D, the remainder of the
ovicell has the form characteristic of stage E.

It can hardly be doubted, from the sections of stage D, that
the ovicell at this time does not differ, or hardly differs,
externally from a zoecium. The young ovicell les completely
in series with the ordinary zocecia, and there is no indication
of a distal dilatation of its cavity.

In Fig. 24, however, the distal end of the ovicell is clearly
dilated, and this expansion has been accompanied by the
vacuolation of the previously solid or semi-solid distal half of
the ovicell, which now contains a spacious cavity, with excre-

ON THE DEVELOPMENT OF TUBULIPORA. 135

tory vesicles and a few cellular contents. The vestibule has
well-defined walls lying in the body-cavity of the ovicell. The
terminal membrane has become distinct; it closes the body-
cavity at its distal end, and is somewhat thickened, except at
the middle.

Fig. 21 represents an ovicell of T. liliacea in stage D.
Although I have only one series of sections of this species in
the stage in question, its characters correspond so closely with
those of the succeeding stage (fig. 20) that I regard it as a
normal ovicell. If this be admitted, it follows that there is a
marked difference between T. liliacea and the other two
species (T. plumosa and T. phalangea), its embryophore
being much longer than anything which occurs in them, and
being in fact as well developed as that of Lichenopora.
Fig. 21 is drawn to the same scale as figs. 16 and 18, so
that comparison is easy.

The other features of the section resemble those found in
the other species. The distal part of the ovicell is still solid,
and contains but few excretory vesicles (not seen in this
section) ; the vestibule can be made out, and the brown body
has a thick investment continuous proximally with the embryo-
phore, and having distally a spht leading to the vestibule.
The existence of the nutritive tissue shows that the ovicell is
at the end of stage D, a conclusion which is also indicated by
the length of the embryophore.

The general features of stage D are thus as follows :—The
egg-follicle becomes replaced by what may be termed the
embryonic follicle, or simply the follicle. This becomes sepa-
rated from the brown body by intervening cells, which are
specially well developed in T. liliacea. The brown body, at
first without any distinct cellular investment, becomes sur-
rounded by a mass of cells which mark it out sharply from
the other parts of the ovicell ; and the proximal part of this
investment gives rise to a special “ nutritive tissue.” The
vestibule makes its appearance, at first surrounded by a nearly
solid mass of cells containing excretory vesicles, into which a
median terminal invagination of the ectoderm projects. The

136 SIDNEY F. HARMER.

ovicell is still in the main cylindrical (or really pyramidal),
and does not yet differ materially from an ordinary zocecium.

Stage E.—Development of the Cavity of the Embryophore and
Enlargement of the Ovicell.

The general external features of the ovicell in this stage
are seen from figs. 31 and 82, both taken from one colony.
The embryophore of fig. 81 is shown in back view in fig. 30,
and is a rather late representative of stage D, as can be seen
by comparison with fig. 16, the nutritive tissue being clearly
indicated on the proximal side of the brown body. The
combination of an embryophore in stage D with an ovicell in
stage E has already been seen in fig. 24. The embryophore
of fig. 32 is in stage E.

The cavity of the ovicell is beginning to expand distally in
figs. 81 and 32, in preparation for the great increase in the
size of the embryophore which is to take place during the
succeeding stage. The distal end of the embryophore in both
ovicells lies at the level marked Emd., and it is thus interest-
ing to note that the embryophore is still in the part of the
ovicell which represents its zocecium-stage.

The growing margin of the ovicell has already acquired
the characters of stage E, consisting of a deeply staining mass
of tissue which extends round the growing edge of the ovicell.

Fig. 32 shows that the growing edge is uniformly curved in
the upper wall of the ovicell, while below it has an undulating
course, due to the fact that the floor of the ovicell is ridged
by the upgrowth of the young zoccia in the way that has
already been described (p. 79).

I believe that the position of the growing edge shown in
figs. 31 and 382 is not completely normal, but that the action
of reagents has caused it to shrink away from the edge of the
calcareous parts. This seems to be indicated by fig. 31, in
which the protoplasmic growing edge practically coincides with
the distal margin of the roof of the ovicell on the left of the
figure, though it has shrunk away from it to the right.
Sections show that the stained line seen in the figures is

ON THE DEVELOPMENT OF TUBULIPORA. 137

really the thickened edge of the terminal membrane, which is
deeply invaginated in the middle, as was the case in the
earlier stage. A longitudinal section of the ovicell thus has
the form seen in fig. 22, in which the middle of the
terminal membrane is greatly depressed, so as to be widely
removed from the chitinous part of the membrane (ectocyst),
which is tightly stretched across the open part of the calcare-
ous funnel. I am doubtful how far this is normal. Some
amount of retraction of the living tissues from the orifices of
the zocecia almost certainly takes place when the animals are
killed, and it is possible that the protoplasmic terminal
membrane is normally in close contact with its chitinous
ectocyst. It is not easy otherwise to see how the ectocyst is
formed, unless the terminal membrane is capable of attaching
itself from time to time to the ectocyst, which, it must be
remembered, increases in extent so long as the ovicell con-
tinues to expand.

Decalcification is a further source of alteration. Bubbles of
gas accumulate during this process in various parts, and are
responsible for a good deal of distortion of the protoplasmic
structures. The bubbles often find much difficulty in making
their way through or past the ectocyst, and an accumulation
of gas between the latter and the living part of the terminal
membrane may be responsible for a considerable amount of
depression of the former. As, however, the tissues were well
hardened before being decalcified, I think it probable that the
gas would rather tear the tissues than alter the entire position
of an epithelium which had lost its flexibility as the result of
long immersion in spirit.

I have examined sections of nearly seventy ovicells in this
stage. About half that number belonged to T. plumosa,
eight to T. liliacea, and the remainder to T. phalangea.

Fig. 19 is a longitudinal section of an ovicell of T. plumosa
at the beginning of stage EH. The terminal membrane is
deeply invaginated in a neighbouring section. The distal part
of the ovicell is still nearly solid, and contains numerous
excretory vesicles. The vestibule is somewhat more distinct

138 SIDNEY F. HARMER.

than before; the brown body has diminished in size, but the
embryo is not greatly changed. The most noticeable difference
between this and the earlier stage is in the embryophore, the
proximal part of which is much more developed than before.
The nutritive tissue is, in fact, now much increased in amount,
and the embryo is thereby further removed from the brown
body.

The nutritive tissue is becoming vacuolated in fig. 19, and
further spaces are originating between it and the outermost
wall of the embryophore. This leads to the condition of
fig. 22, a considerably older stage (of the same species) drawn
to the same scale.

Fig. 22 is from a distinctly bilobed colony, each lobe of
which contains an ovicell in stage E. The two ovicells con-
verge proximally, as can be seen from the sections, which are
parallel to the basal lamina of the colony. Their proximal
ends extend far down into the part of the colony which is
common to the two lobes, so that it is clear that the colony
was in an unlobed state when the development of the ovicells
commenced.

The most important difference between fig. 22 and fig. 19 is
that the part of the embryophore containing the nutritive
tissue consists in the former of a long, cylindrical, thin-walled
portion, containing a loose inner mass of cells which do not
nearly fill its cavity. There is still no great increase in the
development of the embryo.

Fig. 20 is a section of an ovicell of T. liliacea in the same
stage, and it confirms the conclusion drawn from fig. 21 that
T. liliacea is characterised by the great length of its embryo-
phore. The ovicell here figured is about as much developed as
fig. 19 (T. plumosa). The distance of the brown body from
the embryo is very different in the two cases.

The condition of the ovicell towards the end of this stage is
illustrated by fig. 25, representing an ovicell, probably of T.
phalangea, cut horizontally, as is shown by its considerable
breadth. The embryo has now enlarged to a marked extent,
and measures 80 p in its greatest length, which is about three

ON THE DEVELOPMENT OF TUBULIPORA. 139

times the length of the embryo in fig. 19. Several giant-cells
are seen in the immediate neighbourhood of the embryo, as in
Crisia and Lichenopora. The function of these is not
certain, but there is no evidence that they take any direct part
in the future development.

The cavity of the embryophore has greatly enlarged, and
now fills up most of the ovicell. Parts of the nutritive tissue
are seen in its proximal region, and a few scattered cells
belonging to this tissue occur in the middle of the cavity.
The brown body is at the distal end of the cavity ; and this is
its usual, though not its invariable position. In some cases,
during this or earlier stages, it may lie partly in the cavity of
the embryophore and partly in the vestibule, demonstrating
the existence of a communication between these two cavities.
The vestibule is much shorter than before, so that the distal
end of the cavity of the embryophore is now very near the
terminal membrane. The point where the vestibule joins the
terminal membrane will become the future occiopore, the
morphological ‘‘ orifice ”’ of the ovicell.

The terminal membrane is still a good deal thickened and
folded at its edge, particularly on the left side of the section,
and it contains the usual excretory vesicles.

The general features of stage E are thus as follows :—The
fertile zocecium becomes definitively an ovicell, and becomes
obvious externally by the dilatation of its distal end. The
part of the embryophore immediately distal to the follicle be-
comes more or less hollowed out, so that a passage is prepared
by which the embryo can pass into the nutritive tissue, which
is developed from the proximal part of the investment of the
brown body. The part of the embryophore containing the
nutritive tissue becomes vacuolated, and finally forms a wide
space.

The brown body is generally left at the distal end of this
space, surrounded by a mass of cells in close connection with
the vestibule; but as by the vacuolation of the surrounding
tissue it becomes free from other tissues, its position towards
the end of this stage is variable. It may pass into the proximal

140 SIDNEY F. HARMER.

part of the cavity of the embryophore, and in one or two cases
it was found in the vestibule.

The series of developmental stages grouped under the head-
ing of stage E is really a very long one; and there are great
differences between ovicells at the two ends of the series.
This difference is marked in the measurements of all the parts
of the ovicell.

In the early part of stage E the embryo remains for some
time in the position in which it was found in the preceding
stage, and its size does not at first materially increase. It is
now more or less spherical, and has a diameter of about 20—
25 mw. Later in stage E the follicle becomes continuous with
the cavity of the embryophore. The embryo now begins to
elongate in the direction of the main axis of the ovicell, and
soon reaches a length of 40—50 yn, its transverse diameter
being at first small. With the increase of this diameter it
becomes ovoid, and then rapidly increases in size. Measure-
ments of the embryo fairly late in stage E amounted to 80 u
for the major axis, and 50 w for the minor axis of the same
embryo. The oldest ovicell which I have found in this stage
had a pear-shaped embryo 135 m long, with its narrow end
situated proximally. The ovicell had reached a considerable
size, its greatest transverse diameter, as measured in a series
of horizontal sections, being 1:25 mm. (= 1250 nu). The
cavity of the embryophore was spacious, and was cylindrical
for about its proximal half, the distal end dilating into the
form of a trefoil consisting of three lobes, the middle lobe
being connected with the vestibule.

In most of the later ovicells studied in this stage the cavity
of the embryophore was either cylindrical (as in fig. 22), or was
somewhat dilated distally without being distinctly lobed. Be-
ginning this stage with a length? of about 150 u, the embryo-
phore may reach a length of 500 u by the end of this stage ;
and after the establishment of its cavity, its transverse diameter
may become as much as 450 yu at its distal end.

1 This measurement is taken between the points C and B in the figures on
Plate 9.

ON THE DEVELOPMENT OF TUBULIPORA. 141

In a case of rapid and continuous growth like that of the
ovicell of Tubulipora there is of course no special signifi-
cance in the measurements quoted, but they will give some
idea of the dimensions of the parts during the stage in ques-
tion. The measurements here given refer to T. plumosa.

Throughout stage E the embryo is found at the proximal
end of the embryophore. At first in its follicle, the embryo
begins to extend into the adjacent part of the embryophore as
it begins to lengthen. Later in the stage it lies somewhat
more distally in the embryophore, its long axis coinciding
with that of the ovicell. The distinctness of the follicle be- *
comes lost towards the end of this stage, and the fertile brown
body diminishes in size.

Stage F.—Commencement of Embryonic Fission.

I have not given any figure of this stage, which does not
differ materially from that of Lichenopora. The primary
embryo increases largely in size, and divides in much the same
way as in that genus (see 16, pl. x, figs. 832—35).

The details of the lobing of the ovicell are not always the
same, as is shown by the varying position of the oceciostome
in the adult ovicell ; but in several cases which I have observed
in stage F the distal end of the embryophore was merely a
later stage of the oldest embryophore, with a trefoil-like
lumen, described in stage HE. This trifid division of the
embryophore is certainly a common arrangement, the median
lobe being connected with the vestibule, and therefore later
with the occiostome ; while the other lobes form the more
lateral parts of the ovicell. The lobe connected with the
vestibule may be termed the “ axial lobe,’ because, although
not necessarily in the axis of symmetry of the ovicell, it is
part of its morphological long axis. Other lobes will be
termed “ lateral lobes.”

Ovicells in stage F may exhibit no subdivision of the primary
lobes, but in other cases the end of each of the lateral lobes
becomes bilobed. This subdivision of the embryophore is in

142 SIDNEY F. HARMER.

all cases due to the upgrowth of a zoccium which interrupts
the growth of the ovicell.

The ovicell contains (as in stage E) two principal cavities,—
firstly, the cavity of the embryophore; and secondly, the
cavity in which the embryophore lies. The latter may be
regarded as the body-cavity of the original fertile zocecium,
and it is constantly extending itself between the series of
zocecia by the growth of the edge of the ovicell. The growth
of this cavity is more energetic during this stage than that of
the embryophore ; so that each lobe of the latter merely enters
the base of the corresponding lobe of the entire ovicell.

In one ovicell in stage F where this arrangement was
obvious, the axial lobe of the entire ovicell had elongated
greatly, while the corresponding lobe of the embryophore only
just entered its base. The vestibule had accordingly been
greatly elongated (to 400 4), so as to retain its connection
both with the terminal membrane and with the embryophore.
The vestibule is lined with a chitinous ectocyst, and opens to
the exterior by joining the terminal membrane. The fertile
brown body lies freely in the middle of the large cavity of the
distal end of the embryophore. The embryo, about 225 u
long, lies in the proximal cylindrical part of the embryophore,
and is clearly dividing into a number of secondary embryos.
Giant-cells, similar to those of Crisia and of Lichenopora,
occur in the position of the original embryonic follicle.

The total length of a trifid embryophore during this stage
was 800 py, and the distance from tip to tip of its lateral lobes
was 720 uw. The brown body remains distinguishable during
this stage, usually lying freely in the cavity of the embryo-
phore. The partially divided primary embryo in one case
formed a large, more or less spherical mass, with a diameter
of 305 p.

Stage G.—Fully formed Ovicell.

This stage, which corresponds with the figures given of the
mature colonies of the several species, is illustrated by fig. 33,
from a decalcified preparation of T. plumosa. The embryo-

ON THE DEVELOPMENT OF 'TUBULIPORA. 143

phore is seen to be considerably lobed in a more or less palmate
way. The proximal part is cylindrical, and occupies the cavity
which represents the fertile zocecium before it became an ovi-
cell. This part, together with the greater part of the digitate
lobes, is practically solid; and consists, as in Crisia and
Lichenopora, of a reticulum of nutritive tissue, containing a
very large number of secondary embryos in all stages of devel-
opment. The distal ends of some of the lobes can be seen to
be hollow, although the thickness of the colony makes it im-
possible to make out all the details. The oceciostome is at the
end of the lobe marked o., and this is hence the axial lobe.
The greatest length of the solid part of the embryophore, from
the proximal end to the distal tip of the most projecting lobe,
is 2°5 mm.

Sections through ovicells in this stage are readily intel-
ligible by comparison with stage F. A comparatively young
ovicell, which had a transverse diameter of about 1°5 mm.
(measured in the sections), had a five-lobed embryophore.
The middle lobe was in connection with the vestibule, and the
two lobes of each side had resulted from the bifurcation of the
primary lateral lobes.- The distal end of each of the five lobes
of the embryophore was hollow, while the proximal part was
filled up with a mass of young secondary embryos. Excretory
vesicles were numerous in the growing edge.

The fertile brown body can still be discovered, lying freely
in the cavity of the embryophore, in some of the earlier ovicells
belonging to this stage.

The mass of embryonic tissue increases very greatly, and in
the younger ovicells it is easy to see that embryonic fission is
or has recently been taking place. In the most satisfactory
series I possess illustrating this point, the ovicell contained
(in addition to numerous young secondary embryos) an embry-
onic mass, elongated in the direction of the axis of the ovicell,
and measuring about 160 yw in length. This mass was very
similar to the corresponding structure which I have described
at the beginning of embryonic fission in Crisia (15, pl. xxiii,
fig. 11). MKaryokinetic figures were clearly seen in a large

144 SIDNEY F. HARMER.

number of the nuclei, and the mass was constricting off secon-
dary embryos on all sides, and was further surrounded by a
good many young secondary embryos of about the same size as
those which were still in connection with itself.

Demonstrative evidence of the occurrence of embryonic
fission during the early part of this stage was obtained in both
T. plumosa and in T. phalangea. In the case of the best
series of sections obtained of the latter, the species was not
merely inferred from the absence of excretory vesicles in the
tentacles, but had been determined, before decalcification, by
the characters of the owciostome. It appears to me that the
nutritive tissue is much more abundant during this stage in
T. phalangea than in T. plumosa.

The younger ovicells in this stage contain a large number of
young secondary embryos, all in about the same stage as those
which are being constricted off from the larger embryonic mass
already described. Comparing the entire embryophore to a
hand, the secondary embryos at first occupy the part which
corresponds to the palm, where they form a dense mass, com-
posed mainly of secondary embryos, which are separated from
one another by a certain amount of nutritive tissue. Measure-
ments made of this mass, in sections cut in a suitable plane,
gave 250—560 m as its greatest transverse diameter in par-
ticular cases.

As in Lichenopora, I am not able to say how far a secon-
dary embryo, once formed from a larger embryonic mass, is
capable of further fission. In some cases this process seemed
to be clearly indicated. But even in the oldest ovicells, con-
taining mature larvee, ready to escape, and even actually in the
tube of the oceciostome, I have in one or two cases found larger
masses of embryonic tissue clearly giving rise to more than
two or three secondary embryos. One of these masses noticed
in an old ovicell was 90 « long, which is about the same as
the average length of a larva ready to escape from the ovicell.
The dividing embryonic masses are similar to those of Crisia,
consisting of a more or less clearly defined outer layer of cells,
surrounding a central solid mass containing specially large

ON THE DEVELOPMENT OF TUBULIPORA. 145

nuclei (8 4). These nuclei appear to be characteristic of the
growing tissue of the primary embryo, since they occur at a
part where no separation of secondary embryos is taking place.
When traced to a part where a secondary embryo is being
constricted off, these large nuclei are seen to become smaller by
division, and to form the smaller nuclei (3°5 u) of the inner
layer of the secondary embryos. Evidence of fission was not
obtained in all the old ovicells.

The young secondary embryos are always embedded in
nutritive tissue, whatever the age of the ovicell. In many
cases the occurrence of a group of young secondary embryos
in close proximity to one another probably implies recent
embryonic fission. As development continues, the secondary
embryos become ciliated externally, and may then become
quite free in the cavity of the embryophore, the wall of which
may be reduced to a thin nucleated film of protoplasm in old
ovicells.

The cavity of the ovicell is ultimately almost completely
filled by the embryophore, the lobes of which may be several
times divided. Even the axial lobe may give off secondary
lobes (fig. 33), and it is clear that this must take place in
many cases from the external conformation of the ovicell. The
growing edge of the ovicell is usually conspicuous at the end
of any lobe which is still incomplete, but it disappears with
the complete closure of the lobe by a calcareous wall. Some
of the older ovicells suggest as a possibility that the number of
excretory vesicles may be reduced in some way during stage
G, but others were provided with numerous vesicles.

VI. The Morphology of the Internal Parts of the Ovicell.

There can be little doubt that the embryophore of Tubu-
lipora corresponds in general with that of Crisia and
Lichenopora; but it is not easy to decide how far the homo-
logy is an exact one.

In Crisia (15) the ovicell is at no period of its life an
ordinary zocecium. ‘This would almost follow from the fact
that its proximal end has not the character of a simple, cylin-

vou. 41, part 1,.—NEW SERIES. K

146 SIDNEY F, HARMER.

drical zocecium, but begins to widen from its commencement
(cf. 18, pl. xii, fig. 6). The ovicell develops a polypide-bud,
which does not become a polypide, but gives rise to the tissues
immediately surrounding the primary embryo. There is thus
no fertile brown body. In Tubulipora the ovicell is at first
a zocecium with a functional polypide. Degeneration of the
latter takes place at the beginning of the development of the
embryo, and at a time when the polypide is still moderately
young; and a brown body results. In Lichenopora (16)
also the young ovicell is a zocecium, but it has two successive
polypides, the second of which is present until the primary
embryo has developed for some time, and the embryophore has
been formed.

The investigation of Crisia appeared to show that the
rudimentary polypide-bud of the ovicell gave rise to a tentacle-
sheath, which communicated with the exterior by means of a
vestibule (described as “ aperture” in 15). The cavity of the
embryophore in Tubulipora is so similar to the “ tentacle-
sheath ” of the ovicell of Crisia that at first I had no doubt
that the two spaces were homologous. The cavity in Tubu-
lipora is, however, a completely new formation, formed after
degeneration of the fertile polypide. If then it is a tentacle-
sheath at all, it must be that of a newly formed bud.

In my paper on Lichenopora (16, p. 114) I have alluded
to this possibility for that genus. The evidence afforded by
Tubulipora seems to be in favour of the hypothesis that the
parts of the embryophore correspond with parts of a polypide-
bud. ‘This view is supported, for instance, by the great
resemblance of the vestibule of the ovicell to that of the
zocecia, in its structure and in its relation to the terminal
membrane. The polypide-bud would be represented by the
cellular investment which appears round the brown body.
This mode of origin is probably similar to those cases which
have been recorded by Smitt and Hincks (18, p. lvii) among
Cheilostomes, in which the young polypide-bud appears to
grow out of the brown body.

The relations of the cavity of the embryophore to the vesti-

ON THE DEVELOPMENT OF TUBULIPORA. 147

bule at first sight seem to indicate that the walls of the space
represent the tentacle-sheath. There is, however, one fact
which suggests a different explanation. The distal end of the
wall of the embryophore in stages D and E is invariably
reflected inwards, as shown in figs. 22 and 25, the reflected
part usually terminating in cells which are loosely connected
and lie in the cavity of the embryophore. This relation is
very similar to that of what I have called above the somatic
mesoderm of the young polypide-buds (fig. 23, s. m.). If the
wall of the embryophore-cavity can really be compared with
this layer, the general history of the ovicell would appear to
be somewhat as follows. The ovicell is at first an ordinary
zocecium, whose first polypide develops a functional alimentary
canal and anovary. An egg begins to develop, probably while
the polypide is still functional, but still comparatively young.
The polypide then degenerates and forms a brown body.
Certain cells arrange themselves as a definite investment round
the brown body; and these cells, whose origin is obscure,
probably represent a polypide-bud. A vestibule makes its
appearance distally to the brown body. The body-cavity of
the ovicell-bud appears as a space just inside an outer epithe-
lial layer, which represents its somatic mesoderm. This layer
lies freely in the old body-cavity, and its outer surface may be
more or less covered with cells belonging to that cavity. The
greater part of the rudimentary polypide-bud of the ovicell
becomes a mass of nutritive tissue, in the meshes of which the
secondary embryos are afterwards contained. The brown body
comes to lie freely in the new body-cavity by the breaking up
of the nutritive tissue. By the enlargement of the body-cavity,
its somatic mesoderm is brought close to the outer calcareous
wall of the ovicell, and the new body-cavity thus replaces the
old one. The cavity becomes lobed in correspondence with
the lobing of the entire ovicell. The lumen is visible distally,
even in advanced stages, while it becomes filled up proximally
by the great development of the nutritive tissue and of the
secondary embryos. One of the lobes of the ovicell, the
“axial lobe,” is connected with the exterior by means of the

148 SIDNEY F. HARMER.

oceciostome, where the vestibule opens to the exterior ; and the
secondary embryos ultimately escape by this passage.

The junction of the embryophore with the vestibule is some-
what complicated, and from stage D onwards there is a
cellular plug between the brown body and the vestibule. This
is shown in fig. 20 (y.), where it is still solid. In a later stage
(fig. 22, y.) it becomes hollow, its cavity sometimes appearing
completely closed, sometimes opening towards the brown body,
and sometimes communicating with the vestibule. These
different conditions appear to be found in ovicells of one and
the same stage, and the first of them is shown in fig. 25. This
figure illustrates the way in which the wall of the embryo-
phore is reflected inwards, and the similarity with a polypide-
bud (fig. 23) is marked.

It is possible that this plug of cells is the morphological
representative of a tentacle-sheath. On this view, the brown
body of fig. 25 does not lie freely in the tentacle-sheath (which
would be a somewhat anomalous position for it to occupy), but
in the body-cavity.

Sooner or later the vestibule becomes continuous with the
*‘ tentacle-sheath ” (y.), and this with the body-cavity; the
communication between the two latter being uninterrupted
after the migration of the brown body to some other part of
the body-cavity. A communication between the body-cavity
and the exterior is not a very unusual occurrence in the
Polyzoa, since the reproductive bodies escape more or less
directly from the former to the outside in several cases. This
is shown, in Phylactolemata, by the escape of the statoblasts
after the decay of the polypide has left an open passage to the
exterior, and in certain Gymnolemata by the occurrence of a
special passage, the intertentacular organ.

The cavity of the embryophore in Lichenopora is probably
comparable with the cavityin Tubulipora. The stage shown
in pl. ix, fig. 27, of my former paper (16) is very similar to a
Tubulipora at the end of stage D. The vestibule is repre-
sented by the invagination there marked “ orifice ; ” the brown
body with its investment requires no special explanation, while

ON THE DEVELOPMENT OF TUBULIPORA. 149

the ‘‘ suspensor,” constituted by the inner cells between the
brown body and the embryo, is probably represented in Tubu-
lipora by the nutritive tissue.

The comparison with Crisia is less easy, but the main
difference—the absence in the ovicell of that genus of a
functional polypide degenerating to a brown body—has been
already commented on.

Fig. 15 of pl. xxiv (15) shows a young ovicell with a bud
consisting of vestibule (distally), “ tentacle-sheath,” and thick
proximal portion, corresponding with the alimentary canal and
tentacles of an ordinary polypide. Fig. 1 of the same paper is
probably younger, the part there marked “ tentacle-sheath ”
being more probably the vestibule, which is developed in
Cyclostomes before the tentacle-sheath. Fig. 2 is considerably
later, and its ‘‘ tentacle-sheath ” corresponds with the similarly
marked space in the later stages; this is quite evident from
the fact that a vestibule like that of fig. 3 is present in the
ovicell which fig. 2 represents. Although fig. 2 is a good deal
later than fig. 1, it will be noticed that its egg is in much the
same state. The later ovicells formed a perfectly uninterrupted
series. It thus appears to me to be sufficiently established
that the vesicular bud of the ovicell of Crisia (15, fig. 1) gives
rise to the “ follicle ” (figs. 3, 5, 6) of later stages. If this is
the case, the cavity into which the follicle projects may be
really the tentacle-sheath, which would, on this hypothesis, be
much more developed than in Tubulipora. It seems to me
more probable, however, that the cavity of the embryophore
of Tubulipora is identical with the cavity of the “ tentacle-
sheath ” of Crisia, as is indicated by a comparison of fig. 25
of the present paper with fig. 8 of my former paper (15). The
apparent difference in the early stages is probably due to the
fact that the development of the so-called “ tentacle-sheath ”
of Crisia was not observed.

The junction of the vestibule and embryophore in the ovicell
of Crisia (15, fig. 8) is marked by a thickening, which is by
no means unlike the junction between the vestibule and the
embryophore in Tubulipora. I can hardly doubt, therefore,

150 SIDNEY F. HARMER.

that the part which was described as “ tentacle-sheath ” in the
ovicell of Crisia corresponds with the thin-walled embryo-
phore of Stage E of Tubulipora (fig. 25), whatever may be
the morphological character of this space.

The separation of the genera of Cyclostomes is notoriously
difficult. In Crisia, Tubulipora, and Lichenopora we
have three extreme cases, which there can be no difficulty in
recognising as distinct. The fact that their ovicells belong to
three entirely different types renders the ultimate definition of
the genera of recent Cyclostomes a more hopeful task than it is
sometimes supposed to be.t Crisia may be characterised as a
genus in which the ovicells are modified zocecia dilated into a
pear-like form, the region of the oceciostome, as in other
Cyclostomes, not sharing in this dilatation. The ovicell is
from the first an ovicell; and although its morphology is
indicated by the appearance of a polypide-bud, the bud never
becomes a functional polypide.

In Tubulipora the dilatation of the ovicell is usually
much more marked, and the ovicell is commonly lobed. Its
lobes are developed owing to the formation of zocecia distally
to the ovicell, and the latter is accordingly obliged to divide
into two portions which grow round the sides of the zocecium
or series of zocecia. The degree of lobing corresponds with the
extent to which young zoccia are developed distally to the
ovicell. Two lobes may become contiguous on the distal side
of a zoecium, but probably do not fuse ; and the embryophore
also is composed of a series of divaricated lobes which do not
unite (in the species examined). The young ovicell differs
from that of Crisia in beginning life as an ordinary zocecium ;
and the duration of this period is accurately indicated by the
extent of the proximal cylindrical (or pyramidal) portion of
the ovicell.

In Lichenopora, as in Tubulipora, the ovicell is at first
a zocecium, but has more than one functional polypide (in L.
verrucaria). The principal characteristic of the ovicell is,
however, the mode of growth by the addition of peripheral

1 Cf. Gregory, No. 12, p. 21.

ON THE DEVELOPMENT OF TUBULIPORA. 151

“alveoli,” which are at first distinct from the cavity of the
ovicell. The lobes of the ovicell, and even of the embryophore,
unite with one another on the distal side of the zocecia, which
thus pass through the cavity of the ovicell as completely free
columns.

Further investigations will be necessary in order to ascertain

how far the distinction of genera and species of Cyclostomes
can be based on the characters of the ovicells.

REFERENCES.

. Barrots, J.—‘* Recherches sur l|’Embryologie des Bryozoaires,” Lille
? 5 . ’ >

1877.

. Busk, G.—“Cat. of the Cyclostomatous Polyzoa in the British

Museum,” 1875.

. Busx, G.—“* Challenger Reports,” part 1 (‘ Polyzoa,” II), 1886.

. CLaPaREDE, H.—“ Beitrage zur Anat. u. Entwicklungsgeschichte d.

Seebryozoen,” ‘ Zeits. f. wiss. Zool.,’ xxi, 1870, p. 187.

. Coucn, R. Q.—“ A Cornish Fauna,” part iii, Truro, 1844.
. DALYELL, Sir J.—‘‘ Rare and Remarkable Animals of Scotland,” ii, 1848.

. Davenport, C. B.— Observations on Budding in Paludicella and some

other Bryozoa,” ‘ Bull. Mus. Comp. Zool.,’ xxii, 1891-2, p. 1.

. Eistc, H.—“ Monogr. d. Capitelliden,” ‘Fauna u. Flora G. v. Neapel,’

xvi, 1887.

. Exuis, J.—‘‘ An Essay towards a Natural History of the Corallines,”

1755.

. Extis, J., and Sonanper, D.—‘The Natural History of Zoophytes,”

1786.

. Fasricius, O.—‘* Fauna Greenlandica,” 1780.
. Grecory, J. W.— Cat. of the Fossil Bryozoa in the British Museum,”

1896.

. Harmer, 8. F.—“On the British Species of Crisia,’ ‘ Quart. Journ.

Mier. Sci.,’ xxxii, 1891, p. 127.

. Harmer, 8. F.—‘ On the Nature of the Excretory Processes in Marine

Polyzoa,” ‘Quart. Journ. Micr. Sci.,’ xxxiii, 1892, p. 123.

. Harmer, 8. F.—* On the Occurrence of Embryonic Fission in Cyclosto-

matous Polyzoa,” ‘Quart. Journ. Micr. Sci.,’ xxxiv, 1898, p. 199.

152

16.

Ve

18.
19.
20.
2l.

22.

23.

24.

25.

26.

27.
28.
29.

30.

31.

32.

33.
34.

35.

36.

37.

SIDNEY F. HARMER.

Harmer, 8S. F.—‘‘On the Development of Lichenopora verrucaria,
Fabr.,” ‘ Quart. Journ. Mier. Sei,’ xxxix, 1897, p. 71.

Harmer, 8. F.—“ Notes on Cyclostomatous Polyzoa,” ‘ Proc. Cambridge
Phil. Soc.,’ ix, part 4, 1896, p. 208.

Hincxs, T.—‘ A History of the British Marine Polyzoa,”’ 1880.

Jounston, G.—‘‘ A History of the British Zoophytes,” 1838.

Jounston, G.—‘ A History of the British Zoophytes,” 2nd ed., 1847.

JutiiEeN, J — Mission Sci. du Cap Horn” (Zool.), VI. ‘‘ Bryozoaires,”
1888.

Kirkpatrick, R.—‘ Report upon the Hydrozoa and Polyzoa... .
China Sea,” ‘ Ann. Mag. Nat. Hist.’ (6), v, 1890, 11.

Lamarck, J. B. P. A. de.—** Histoire Naturelle des Animaux sans
Vertébres,” ii, 1816.

Lamarck, J. B. P. A. de.—‘ Histoire Naturelle des Animaux sans
Vertébres,” 2nd éd., ii, 1836.

Lamovuroux, J.—< Expos. méthodique des Genres . . . . des Polypiers,”
1821.

Levinsen, G. M. R.—“‘ Zoologia Danica,” Bd. 4, Afd. 1, ‘ Mosdyr,’
Copenhagen, 1894.

Linyzus, C.— Systema Nature,” ed. 12, T. i, Pars ii, 1767.

Linnavs, C.—‘ Systema Nature,” ed. Gmelin, T. i, Pars vi.

Mitne Epwarps, H.—“ Mém. sur les Polypes du Genre des Tubuli-
pores,” ‘Ann. Sci. Nat.’ (2), ‘ Zool.,’ vol. vill, 1837, p. 321.

Mitne Epwarps, H.—‘‘ Mém. sur les Crisies....,” ‘Ann. Nat.
Sci.’ (2), ‘Zool.,’ vol. ix, 1838, p. 198.

Nirscuz, H.—“ Beitrage zur Kenntniss der Bryozoen,” ‘Zeits. f. wiss.
Zool.,’ xxi, 1870, p. 416.

Ostroumorr, O.—< Zur Entwickelungsgeschichte d. Cyclostomen See-
bryozoen,” ‘ Mitt. Zool. Stat. Neapel,’ vii, 1887, p. 177.

Patuas, P. S.—‘ Elenchus Zoophytorum,” 1766.

Percens, E.—“ Pliocine Bryozoén von Rhodos,” ‘ Ann. Hofmus. Wien,’
i, USS7ipae.

Pereens, E.—*‘ Untersuchungen an Seebryozoen,” ‘ Zool. Anzeiger,’
xii, 1889, p. 504.

Provuo, H.— Contribution a l Histoire des Bryozoaires,” ‘Arch. Zool.
Exp.’ (2), x, 1892, p. 557.

Savieny, J. C.—‘‘ Description de Egypt,” ‘ Polypes” [plates only],
“Explication” by V. Audouin, 1826 [ef. C. Davies Sherborn, ‘ Proc.
Zool. Soc.,’ 1897, p. 287].

ON THE DEVELOPMENT OF TUBULIPORA. 153

838. SEELIGER, O.—‘‘ Bemerk. zur Knospenentwicklung der Bryozoen,”
‘Zeits. f. wiss. Zool.,’ 1, 1890, p. 560.

39. Smrrr, F. A—“ Om Hafs-Bryozoernas utveckl. och fettkroppar,” ‘ Ofv.
Vet.-Akad. Forh.,’ 1865, No. 1.

40. Smirt, F. A.—“ Kritisk forteckning 6fver Skandinaviens Hafs-Bryozoer,”
II, ‘ Ofv. Vet.-Akad. Forh.,’ 1866, p. 395.

41. Watrorp, EH. A.—‘On some Bryozoa from the Inferior Oolite of
Shipton Gorge,” LI, ‘ Quart. Journ. Geol. Soc.,’ 1, 1894, p. 72.

42. Watrorp, E. A.—‘‘On Cheilostomatous Bryozoa from the Middle Lias,”’
‘Quart. Journ. Geol. Soc.,’ |, 1894, p. 79.

43. Waters, A. W.—‘‘On Tertiary Cyclostomatous Bryozoa from New
Zealand,” ‘Quart. Journ. Geol. Soc.,’ xiii, 1887, p. 337.

44, Waters, A. W.—“ Bryozoa from New South Wales,” III, ‘ Ann. Mag.
Nat. Hist.’ (5), xx, 1887, p. 253.

45, Waters, A. W.—‘On some Ovicells of Cyclostomatous Bryozoa,”
‘Journ. Linn. Soce.,’ ‘ Zool.,’ xx, 1890, p. 275.

EXPLANATION OF PLATES 8—10,

Illustrating Mr. Sidney F. Harmer’s paper “ On the Develop-
ment of Tubulipora, and on some British and Northern
Species of this Genus.”

PLATE 8.

The figures on this plate were all drawn to the same scale (camera lucida,
Zeiss A obj., with front lens removed; afterwards x 3). The pores of the
zocecia are in most cases not indicated.

Fig. 1.—Tubulipora plumosa, W. Thomps. (p. 105). Fertile lobe,
with one ovicell and the beginning of a second ovicell; from a bilobed colony
(proximal ends of oldest zocecia, at the bottom of the figure, obscured by
foreign substances). Salcombe estuary, 4—5 fathoms, March—April, on
Rhodymenia ciliata.

Fic. 2.—T. aperta, n. sp. (p. 101). A small colony with a single ovicell,
which is abnormal in possessing four accessory oceciostomes, numbered 1, 2,
3, 5. 4 is a similar structure, but its terminal membrane is completely

154 SIDNEY F. HARMER.

calcified with the exception of a minute central pore. The colony is seen to
originate in the “primitive disc,” the calcified body-wall of the larva.
Godésund, Bjorne Fjord, Norway, June, on Laminaria saccharina.

Fic. 3.—T. aperta. Distal view of the same colony, to show the tube of
the oceciostome. ‘T'wo of the accessory oceciostomes (numbered as in fig. 2)
are also seen.

Fic. 4.—T. flabellaris, Fabr. (p. 99). A stunted colony, in which the
proximal ends of the oldest zocecia were obscured by foreign substances. The
lateral parts of the colony are bent backwards round the Cellularia peachii
on which the specimen is growing. 1—6, occiostomes of the six mature
ovicells, oceciostome 4 being concealed from view by the zoccia. The ovicells
to which the occiostomes 5 and 6 belong are seen not to have completed
their growth. 7 is a young ovicell. Barents Sea, 50 fathoms, July 1;
dredged by Colonel H. W. Feilden. Greatest diameter of colony, 2°36 mm.

Fic. 5.—T. phalangea, Couch (p. 94). Fertile lobe with a single ovi-
cell. s. septum between two contiguous lobes of the ovicell. Salcombe
estuary, 4—5 fathoms, March—April, on Rhodymenia ciliata.

Fic. 6.—T. phalangea. Zoccium, with the oceciostome, broken off from
a colony, and placed in such a position as to show the oceciopore. Salcombe
estuary, 4—5 fathoms, March—April, on shell.

Fie. 7.—T. liliacea, Pall. (p. 90). Part of ovicell, with two series of
zocecia and an occiostome. Plymouth, 30—40 fathoms, March—April, on a
Hydroid.

Fie. 8.—T. liliacea. Fertile lobe seen from its distal end, to show the
oceciostome, the upper lip of which projects so as to conceal the oceciopore,
which opens horizontally. The proximal end of the lobe is the lower side of
the figure. Plymouth, obtained with the last specimen on a Hydroid.

Fie. 9.—T. liliacea. Lobe in which no ovicell is yet apparent, to show
the biserial, Idmonea-like arrangement of the zocecia. Plymouth, obtained
with the last specimen on a Hydroid.

PLATE 9.

The sections figured are all longitudinal. Fig. 10 was drawn with Zeiss F,
figs. 11 and 13 with >, oil-immersion, figs. 12—22 with DD. All the
figures were afterwards reduced 3.

(The microscopical sections which are referred to as T. plumosa
are those in which excretory vesicles were discovered in the
tentacles; those referred to as T. phalangea are the specimens in
which no excretory vesicles were seen in that position. Unless
otherwise stated, the discrimination of these two species in
sections depended entirely on this character.]

ON THE DEVELOPMENT OF TUBULIPORA. 155

Fie. 10.—T. plumosa. Ovarian egg, in follicle (stage A).

Fie. 11—T. plumosa. Stage B, the embryo consisting of two blasto-
meres, The cecum of the fertile polypide (which has not yet degenerated) is
seen ; f, and the corresponding structure to the left are probably the degene-
rating follicles of eggs which are not developing.

Fic. 12.—T. phalangea. Degeneration of the fertile polypide (stage B).
The tentacles, which are better seen in neighbouring sections, have lost their
distinct outlines, and are obviously degenerating. AB=300 pz.

Fic. 13.—T. phalangea, stage B. Three blastomeres are seen in the
follicle-cavity, which also contains a spermatozoon. The brown body still
shows traces of the alimentary canal of the fertile polypide.

Fie. 14.—T. phalangea. Early stage D. AB=165 pp; CB=75 yp.

Fie. 15.—T. plumosa. Early stage D. A B= 200 p.

Fie. 16.—T. phalangea. Late stage D, as shown by the considerable
development of the cellular investment of the brown body, and particularly by
the developmeut of the nutritive tissue on the proximal side. The terminal
membrane is deeply invaginated, and the vestibule is visible. A B= 250 p,
CE 110" u.

Fie. 17.—T. phalangea. Late stage D; obliquely longitudinal section
of the distal end of the ovicell, to show the oblique, constricted junction of the
vestibule with the embryophore. ‘The invagination of the terminal membrane
is not cut so as to show its opening to the exterior.

Fie. 18.—T. plumosa. Rather early stage D, with very numerous excre-
tory vesicles. The vestibule is small. The cellular investment of the brown
body is thickened distally (X). There is at present no nutritive tissue.
AO B= 225 p, C B= 8b pe.

Fie. 19.—T. plumosa. Larly stage E. The nutritive tissue is largely
developed, and the embryophore is becoming vacuolated. The embryo has
increased in size. The section passes on one side of the median invagination
of the terminal membrane. A B = 460 », C B = 165 up.

Fig. 20.—T. liliacea. arly stage E. The embryophore is longer than
in the other species. y. Plug of cells between vestibule and embryophore
(see p. 148). A B= 630 p, C B= 255 yp.

Fie. 21.—T. liliacea. Younger ovicell (late stage D). The embryophore
is much elongated, as in Fig. 20.

Fic. 22.—T. plumosa. Ovicell at middle of stage EH. y, corresponding
with the similarly lettered part in figs. 20 and 25, has now acquired a lumen,
A B= 960 p, C B= 355 pw.

156 SIDNEY F. HARMER.

PLATE 10.

Fig. 23 was drawn with Zeiss DD, and was not afterwards reduced. The
other figures were reduced two thirds after being drawn. Figs. 24 and 25
were drawn with DD; Figs. 26—28 with F; Figs. 30—32 with A; Fig. 33
with A, the front lens of which was removed. Fig. 29 was not drawn with a
camera lucida.

Fic. 23.—T. plumosa. Polypide-bud, partly diagrammatic ; for compari-
son with the ovicells shown in Figs. 22, 25, &c. s. m. Somatic mesoderm,
reflected on to the very thin tentacle-sheath, which becomes continuous with
the tentacles near their proximal end. The epithelium of the stomach (below
the tentacles) is derived from the inner layer of the bud, and is covered by
the outer layer. The distinction between the two layers is not easily made
out in the mass marked ‘‘ tentacle.”

Fie. 24.—T. plumosa. The embryophore is in stage D, and the distal
part of the ovicell is in stage EK. Numerous excretory vesicles occur distally.
A B= 460, CB = 110) D Ri 275i:

Fic, 25.—T. phalangea. Advanced stage H. The growing edge is much
lobed and thickened on the left side of the figure. The vestibule is much
shortened; it is separated by the space y (cf. Figs. 20 and 22, and p. 148)
from the brown body, which is now almost free in the cavity of the embryo-
phore. The outer wall of the embryophore is reflected over the wall of the
space y (cf. the polypide-bud, Fig. 23). The embryo has greatly enlarged,
and three giant-cells are shown. C B = 360 yn, F G= 385 yp; greatest
length of embryo = 80 p.

Fic. 26.—T. plumosa. Tentacle, fresh, with homogeneous excretory
vesicles and pigment granules (p.).

Fic. 27.—T. plumosa. Another tentacle belonging to the same polypide,
after the addition of iodine in potassic iodide. The contents of the homo-
geneous vesicles have been precipitated.

Fic. 28.—T. phalangea. ‘Tentacle, fresh, with compound vesicles and
pigment granules (p.).

Fie. 29.—T. plumosa. Orifices of living zoccia. The terminal mem-
brane has been somewhat retracted.

Fie. 30.—T. plumosa. Not decalcified (see p. 136). View of a colony
from below, after the basal lamina has been scraped away. X and Y are zocecia,
which are similarly marked in Fig. 31. G is the thickened edge of the
growing margin of the ovicell (in stage E), and corresponds with G in Fig. 31.
The embryophore (in stage D) lies in the proximal undilated part of the
ovicell. The fertile brown body, the embryo, the beginning of the nutritive
tissue, and the distal thickening of the investment of the brown body, can be
made out. A B= 900 p.

ON THE DEVELOPMENT OF TUBULIPORA. 157

Fie. 31—T. plumosa. Upper view of the same lobe. mb. Leve of
the distal end of theembryophore. X, Y, and G are the parts similarly marked
in Fig. 30.

Fie. 32.—T. plumosa. A similar preparation of an older ovicell in stage
E. The zocecia D project upwards into the floor of the ovicell, and the
thickened edge of the terminal membrane is looped over these projecting
parts. Hmb. Level of the distal end of the embryophore.

Fie. 33.—T. plumosa. Decalcified preparation of an old ovicell (stage G),
with nearly solid embryophore containing numerous secondary embryos. The
axial lobe of the embryophore ends in the occiostome at o., and gives off
another lobe to the right. The main lateral lobe of the right side is only
obscurely bifurcated ; that of the left side is divided into five lobes. 2. Proxi-
mal part of ovicell, corresponding with a zoecium. Greatest length of solid
part of embryophore to tip of most projecting lobe, 2°5 mm.

Postscript.—Braem’s interesting work, ‘Die geschlechtliche Entwickelung
von Plumatella fungosa” (‘ Zoologica,’ Heft 23, 1897), did not appear
until after my MS. was in the hands of the printers; and I am unable
to refer further to his observations and conclusions on the present
occasion.

me
a a)

a is
<b
Sa
a
5 - = |
- i] Li) P ; si/ 4 ~ orl ifs Ava avg ee
_ eee? 9's) k 12) bi a rms > wig 4 rh ns :
vx a PR Shas }
le ie insite cain wt ¢ ; tar wlce a fi

“8h os 0) ). UTR Pee Beale Mil hie

t
at ey i ‘ae Wi L = ay K CP ort, ae
oss se
- . <> Bee FP Bios Lien eal wie alr
herty ivy p ui Gai 74s
7 ee ow §
iy 1h} = } er Rie / ely cell
iD Ho Peat yi hel aie are
Q . i ‘, . > wa
7) P ge ‘ 4 4 yar iyi atig ipa) iyche We ae a
: — : ‘ie
ie ; ' weil ‘ 4
ae)
: ‘ Ps ania iy 2
sf = 7 —_ .-
a pe ay te
~~ " ‘t r I.
= _s e f :
UL pone 4 Pal
au wo
jai?
|
A -
= ry a
S, = fe

-
=
>
*
}

$e

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 159

The Molluscs of the Great African Lakes.
I. Distribution.

By

J. E. S. Moore.

THE present paper forms the first instalment of the
zoological report of an expedition which, through the
generous support of the Royal Society and the British
Association, I was able to make to Lake Tanganyika during
1895 and 1896.

The primary objects of this expedition were—

1. To study the unique fauna of Lake Tanganyika on the
spot ;

2. To make what observations were possible in the Nyassa
region while I was en route; and—

3. To bring back properly preserved material for the com-
plete morphological investigation of the more remarkable
lake organisms after I returned.

Before proceeding to the purely zoological matters with
which I propose to deal, it is appropriate that I should here
express my sincere thanks to Professor Ray Lankester, to
whom I have been indebted for the primary suggestion of the
whole inquiry, and for much kindly help since my return. I
have also to thank Professor G. B. Howes for the use of the
Huxley Laboratory and invaluable advice, without both of
which I should never have been able to get the subject
through, while I have been very materially indebted to Sir
John Kirk, who procured for the expedition the necessary
introductions to the administrative gentlemen through whose
districts it had to pass. And last, but not least, I have to

160 J. BE. S. MOORE.

thank Sir Harry Johnston for the very effective support he
lent the expedition, and without which it would have been
impossible for me to attain the objects which I had in view.

Excluding the polar regions proper, there exists in the
fresh waters of the different continents a type of fauna which
in the character of its constituents is essentially the same.
Certain forms are added and others are omitted as we pass
from the more temperate to the equatorial zones, but beneath
these changes there exists a substantial similarity, so easily
recognisable and so marked that geologists have not hesitated
to distinguish between fresh-water and marine fossiliferous
deposits wherever they may be found. On the other hand,
that there is a hard and fast demarcation between fresh-
water and marine faunas is not true, for there are many
instances of animals—for example, of prawns and crabs, which
in this country are purely oceanic—having made their way up
the rivers into inland and elevated fresh-water lakes. Further,
there are a number of animals that belong neither to salt nor
fresh water, but are inhabitants of the brackish regions which
lie between inland fresh waters and the sea. That the well-
established and more permanent fresh-water organisms of
the present day are descended from older phyla that were
once marine, is an accepted truth. This view is necessitated
by the theory of common descent, and it is supported, as in
the case of the ganoid fishes, by the similarity of numerous
living fresh-water organisms to older oceanic types. It is sig-
nificant, however, that with few or no exceptions, all the well-
established fresh-water organisms of to-day are not directly
referable to the earliest oceanic forms, but rather to those
which in their temporal distribution stand intermediate be-
tween then and now. It seems that the fresh-water molluscs
of the present day first make their definite appearance in
Tertiary times, for much doubt has recently been raised as to
the genuineness of the so-called carboniferous Unio, Ticho-
gonia, aud Planorbis; these forms being now regarded
as more nearly related to Anthracosia, Avicula, and

THE MOLLUSCS OF THE GREAT AFRICAN LAKHS. 161

Serpula respectively. The family Limneide, which is
now so universal in its distribution, does not certainly extend
further back than the Jurassic period. The same is true
of the fresh-water Melaniide and of the Paludinid, but
I need scarcely point out that it is necessary to use the
greatest caution in drawing any inferences respecting the
date of origin of the true fresh-water forms from these apparent
facts. It may, however, be taken as approximating to the
truth tosay that although the typical and universal fresh-water
molluscs of the present do not appear upon the stage of life as
such before the Jurassic period, they almost certainly origi-
nated from a series of marine types which had become com-
pletely differentiated from their oceanic associates long before
this time. Some of these antecedent organisms are probably
represented in the paleontological record by those extinct
genera with which the earliest known modern fresh-water
types are usually associated.!

The facts of morphology are themselves in harmony with
such a view, for in their anatomy the living fresh-water
molluscs do not approximate to any of the more modern
marine genera; they hark back to those more permanent
marine types which were in existence long before Jurassic
times. They certainly bear no resemblance to the generalised
conceptions or archetypes of the more modern marine genera
which appeared during Tertiary and post-Tertiary times, such
as Strombus, Pteroceros, Rostellaria, Conus, Mitra,
Chenopus, and the like. It is this fact that is of first
importance to us here; for if it should be found that in
some district at the present time there exists a fresh-water
fauna which departs from the normal and universal type in
the possession of genera which approximate to those that are
undoubtedly modern and marine, we shall have very strong
prima facie evidence for regarding these organisms as recent

* It is quite possible that many of the old so-called fresh-water deposits are
in reality marine, since the forms which became exclusively fresh water as
time went on probably made their appearance in the sea first, as so many of
the more recently derived fresh-water types have done,—prawns, for example.

VoL. 41, part 1,—NEW SERIES, L

162 J. E. S. MOORE.

importations from the sea. Now such a fauna is presented to us
in that of Tanganyika at the present day, for in this lake there
have been known to exist ever since 1859 what appear to be
the shells of some six genera of Gastropods, which are entirely
unlike any known fresh-water forms, while their shells at
the same time simulate several modern oceanic types. The
interest in these strange molluscs, which have been known
hitherto only by their conchological characters, was greatly
augmented when in 1883 Boehm found jelly-fish in the lake;
and during my recent expedition I have been able to add
deep-water crabs, prawns, sponges,' and Protozoa to
this anomalous list of organisms, all of which appear to
possess the same marine affinities.

It is my object in the present paper, to ascertain to
what conclusions as to the nature and origin of this anoma-
lous series the facts of distribution lead ; while in those which
follow I shall deal with the morphological affinities of the
hitherto unknown individual forms, and thus determine whether
the conclusions to which these facts of distribution seem to point
are really sound. Almost no definite observations have been
hitherto available for the study of this subject, and consequently
the material contained in this and the following papers will be
mostly new. I would, therefore, invite particular attention to
the positive character of the evidence which I shall bring
forward in support of the recent marine origin of a number of
the animals contained in Tanganyika, as compared with the
wholly negative character of that upon which the geological
speculations of Murchison respecting the ‘‘ permanence of
terrestrial conditions ” in the African interior at present rest.

To believe that the marine animals of Tanganyika are among
the few remaining indications of a sea that once extended to
the very heart of the African continent is to come into the
most uncompromising conflict with the theory put forward by

1 The affinities of the deep-water sponge which I obtained in Tanganyika
have not yet been determined, but its striking external character and its
remarkable deep-water habitat have inclined me to regard it as a member of
the anomalous section of the fauna which the lake presents.

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 1638

Sir Roderick Murchison in 1852. Yet such a view is now
supported by the strongest kind of zoological evidence it is
possible to get.

In order to arrive at a sufficiently complete comprehension
of the general type of the molluscan fauna which characterises
the individual lakes, it is quite unnecessary and even prejudicial
to discuss the question of the occurrence or non-occurrence of
many of the so-called specific forms, since a large number of
these are merely geographical varieties, while others have
been based on minute and often purely fanciful conchological
distinctions. The four species of Hylacantha, described
by Bourguignat in 1890, for example, are certainly nothing
more than the rather remarkable polymorphs of the original
Typhobia Horei described by Smith in 1881. But what is
true of the polymorphic Typhobia is equally true of the
polymorphic Paramelania and Neothauma. I shall there-
fore consider the distribution of the genera alone, or the main
issue will become lost in the pursuit of really non-existent types.

In Nyassa, which was the first great lake I reached,
and which has hitherto been better known than all the rest,
there have been recorded some sixteen genera of molluscs,
namely, Limnza, Isodora, Physa, Physopsis, Planorbis,
Ancylus, Ampullaria, Lanistes, Vivipara, Cleopatra,
Bythinia, Melania, Spatha, Iridina, Corbicula, and
Unio.

In the smaller lakes occurring in this district, such as
Shirwa, there have been found a fewer number of the sarne
genera. In Kela, which is within twenty miles of the south
end of Lake Tanganyika, I found Planorbis and Limnea.
In Mwero there have been recorded Unio, Ampullaria,
Lanistes, Vivipara, Cleopatra, Bythinia, and Melania;
while in Bangweolo, according to Lieutenant Weatherly, there
are no shelled molluscs, but it is hardly credible that their
absence in this lake will be maintained. It is thus certain
that the generic forms occurring in Nyassa completely
cover the mollusca in a very large number of African

1 Ann, des Sci. Nat.,’ septiéme série, ix, x, 126, 1890.

164 J. E. Ss. MOORE.

lakes indeed. In the Victoria Nyanza all the Nyassan
genera have been recorded, and more or fewer of the con-
stituents of the same Nyassa list constitute the faunas of the
remaining members of this more northern group of lakes,—as,
for example, the Albert Nyanza, the Albert Edward, Beringo,
and the like.

To facilitate comparison I have arranged the names of all
the lakes about which anything definite is known in the tabular
form given on p. 166. On the left-hand side will be found a
list of all the genera hitherto known to be contained in each.
From this table it will be seen that more or fewer members of
the Nyassa list of genera are contained in every lake, but that
there is a curious reduction of the number of the genera as we
pass from the greater lakes to the less. This is probably due
to the impermanence of the conditions in the smaller lakes,
for we find in Shirwa, which is salt, and in Kela, which has
periodically dried up, only those forms which can stand a wide
amount of change. If we pass momentarily, however, from
the study of the genera among the lakes to that of the specific
forms, it will be found that there is a certain amount of
variation in the specific representation in the genera contained ;
and that when the lakes are widely separated—as, for example,
Nyassa and Lake Mwero—such specific variations are often
strongly marked. Judged, however, by the genera alone, it
will be seen that there is a remarkable uniformity in the
character of the African fresh-water molluscs over an immense
area of ground. To this rule of uniformity in type which
characterises the molluscan fauna of all the lakes about which
anything is known, Tanganyika seems to form a solitary and
striking exception. But the differences which this lake pre-
sents are in one sense more illusory than real, for on inspec-
tion of the table it will be seen that Tanganyika does contain,
and fully represented, the great lake list of molluscs found in
the Nyassa to the south, and the Victoria Nyanza to the
north. It differs from the other lakes in there being here
added to the otherwise universal list a number of entirely new
forms. The genera which compose this superadded series

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 165

comprise among others the six genera of Gasteropods which
have been known hitherto only by their empty shells, namely,
Ty phobia, Paramelania, the so-called Lythoglyphus of
Tanganyika, Syrnolopsis, Nasopsis, and Limnotrochus.
I was, during my recent expedition, enabled to add to this
isolated series at least two entirely generic forms, for which
I have proposed the names Bathanalia and Bythoceras.
We have, therefore, now in Tanganyika some eight genera of
Gasteropods which are not found in any of the other lakes,
and to this isolated list of molluses there should prohably be
further added among the Lamellibranchiata the so-called
Unio Burtoni, and one of the Tanganyika Spathas. Con-
sequently there are now known to exist in Tanganyika ten
genera of molluscs, which appear to be restricted to the lake
in which they were originally found.

Although I am concerned here primarily with the dis-
tribution of the molluscs in these lakes, it must be clearly
understood that the marine organisms, such as jelly-fish, crabs,
prawns, sponges, and Protozoa, with which the above molluscs
are associated, share equally the same geographical limitation.
The ten quasi-marine molluscan genera being, in fact, only one
section of a complete fauna, containing widely separated types,
which in Tanganyika exists along with the normal fresh-water
stock the lake contains. The fauna of Tanganyika is thus a
double series, and to distinguish its apparently marine con-
stituents from the more normal lake animals I shall speak of
them in future as the Halolimnic group.

Now it can be confidently affirmed that there are no Halo-
limnic animals in Nyassa, Shirwa, or Kela, all of which I
visited and dredged; and they are certainly not present in
Bangweolo or Mwero. Yet these organisms are so con-
spicuous and common, when they do occur, that they would
certainly have been recorded from the Victoria Nyanza, the
Albert Edward, and the Albert, if any Halolimnie animals
had existed in these more northern lakes. So far as is at
present known, then, the Halolimnic fauna is entirely re-
stricted to the confines of Tanganyika, in which lake it was

166 J. E. S. MOORE.

originally found. The remarkable isolation and independ-
ence of the Halolimnic fauna which these facts disclose is

Taste I.—Showing the Molluscan Genera which have
hitherto been recorded in the Principal African
Lakes.

3

~
‘ =

: 3
aon Te 5 .
. i=} o oOo « Aas Pe ae

ee Sei ees eee Pp i Rp eS Gee
ES rey Sires) ese Sa ee Fo ie Cee ee ccve OmemnD
Aye) Sul ola oa ees Wee eres
a—| G3) — wt |
Ss aoa oO Mo me msl nm de 40 4 ee

Wor oe. ahaa ael ee Saleen li arscral le eusctl ese sel rete leh lhe eal re
Satin < s. mcilbenellgwarcaleateal ronsulll sine Mages ted Paracas ellie ae ba.) —|—}|—
(OTA ANIC, ese wallace lhopadltaaod eae: locdalltsee esol | spe lteea kcal |p ef)
Gero bie e heume ae wel oou.| ater! sees Manat lose tooellpanel deol aval eos dl aod Ih)
Ibitenimesel oe 60 ollaoo'|) oo.|c00 bom {ode Il ood |I="Ihoga Il soo || co0-|lbod | == | = | —
| E=Ye¥s (0) ccs mee PRU eameres Vieterea! [eA licen aearil hoi. Gla biesael Rotel ocala itsoaalleaaed ie | =I —
PHYSOPSIB: ts 52 als oAoeaallereet| cancel taeier|| Gectall gest beet lostae leery omelet Uae ara
PAG oreoyeie, oe ollocodt op |enedlcod "|| sec || font oe |=——"|| evo ll == = || |
PCy LOC haa eter cht ese WiScib acl fcr [ocieeel kcioal aoe paeonl coal Sane esas Ifiode, || =
Anmipallllarian peas Fai|ises'l feel ate arse emer il ieteeellh ee eal mele” ee a
1 DENTE pence ean | (ere Peres optic hence ainae [eaoPell terete ===! lasetel lege letacia |=) ——|h
Vivipara 2. fee fess fee foe | eee [eee we f—l|—}—}— |} lo
Cleo paticatge.. 7 allitcra|(Prwe th ice cage l|etewsy hier panty cee dl Retell et td ell
Bithyuia . oil astar|Naioe il BAY vel | retell 435

Melania . . «|eec|cce ese} cee | vee | oe | 2 P—|—} — | = | = | =
INIGLOV TENT TIE Gere” eal cope Re ESonll Panel lane. Wiseee lean isda. Agee eeoul snail imoaoel| atone
GMa eo) eG Weare ges |leeiceal Perera cul/Aar cP cscen | acre Peaor eeanl astm Lanoe[ectss rcs’ ||xeoo ul)
INAS OWSISE cose © ce: lkeettetl| econ seven ltescesy ll eens |llecteteall ceo cereal iszeloa | ceeteal lero || eect ceca
ATUINITLO CROC MUS ee ead een t eecseal tester ee alle kad lesees Hl lreetesy| Reese | etcrcell peta Fees ele
DVGMOLO SIS a5 ei fowl nee latwartl soa eoeret leery Aleve Ili. geregll eran leet roi stead Ui
ThE Oey US sees heal ccice| aera | Gea ae oat fret | con Rev Pastel beers lie tee te
MGA INV VAL ates enc ee hail Pet” [Pataten Hine stl ciel sted || penta ocitl a eed eenal ee | cee Pea
ararmel na elle seccillp ace Mdecre alt ctete | aceseell sescec ell cs cll bce ty eu eee era li eee ee
IS VPDOCERAS © yotA rails. coer lsat ates | gael lcesill er llama I teats Neste eect cee eal

Lake Dumi contains Ampullaria and Mutela. Lake Elmentella, Limnea
and Physa. I have not thought it worth while including these small lakes in
the above list.

of the first importance when we attempt to ascertain from
what source they may have sprung ; it is extremely important,
therefore, that the conclusions which result from the study of
the geographical distribution of these forms can be corrobo-
rated from another point of view.

Normal
Lacustrine
Series.

Halolimnic
Group.

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 167

When comparatively examined, the observations which I
was enabled to make respecting the bathymetric distribution
of the molluscs in the lakes will be seen to show quite as
clearly as the facts of their geographical distribution that the
Halolimnic series is something entirely distinct from the
normal fresh-water population of the lake. Through the kind-
ness of Sir Harry Johnston I was enabled to attach myself to
one of the Nyassa gunboats, and thus to become acquainted
with the facts of molluscan distribution in Nyassa before I
reached Tanganyika on my way north; and the observations
made on this voyage have been of the utmost value as material
for comparison with what I subsequently saw.

Nyassa is a relatively narrow lake of great and unknown
depth ; soundings of 300 fathoms, no bottom, having been
obtained throughout a great proportion of its area. In total
length it covers some 340 miles, and it varies from 20 to 40
miles across. The shores are of the most varied description,
steep and precipitous in some places, in others bounded by
extensive flats. The lake has a free outlet down the Shire
River and the Murchison cataracts, the water being conse-
quently clean and fresh. Owing to the lake’s great extent,
the shores are often swept by a heavy surf, and the fairly
strong currents which are observed are probably produced by
the trade winds, for they seem generally on the surface to set
from south to north. The fauna of such a lake is exposed to
the same conditions as those to which it wouid be on an open
oceanic coast. The enormous depth of the lake in many places
rendered it impossible to dredge, and whether with an efficient
deep-water apparatus anything further could be obtained from
the abysmal recesses it is impossible to say. But it was very
soon apparent that the molluscan portion of the population
rapidly thinned out with increasing depth and distance from
the coast, and that beyond 100 feet one could often dredge
for miles over rocks and sand and mud without securing a
single shell.

In many places the lake was floored with compact drifted
masses of shells and shell-fragments, consisting chiefly of the

168 J. E. S. MOORE.

Nyassa Viviparas. No sponges grew upon these shells, and
beyond an occasional Melania such stretches of the lake were
without life of any sort or kind. The curious diminution of
the molluscs of Nyassa beyond the immediate coast-line is a
striking feature of the lake throughout, and a tolerably correct
idea of the bathymetric distribution of the individual genera
will be gathered from the accompanying table, which contains
an epitome of the observations made.

TaspLe I1.—Bathymetric Distribution of Molluscs

Limnea
Tsodora.
Physopsis .
Planorbis .
Ampullaria

Lanistes
Vivipara
Bithynia
Melania .
Unio.
Spatha .
Utela

| Ibexecbuag, ~

100 200 3800 400 500 600 700 800 900 1000
The numbers at the foot indicate the approximate limits of depth in feet at
which the molluscan genera are found, and the thickness of the lines
shows in what depth of water they most abundantly occur.

The lines representing the bathymetric extent of each genus
are thickened so as to show where it is found in the greatest
abundance, and approximately at what depth it ceases to exist.

From this table it will be seen that the purely fresh-water
molluscan population of Nyassa is more or less completely re-
stricted to a littoral band along the shores, and that the great
majority of the genera do not extend in depth beyond 200
feet.

Thus Lake Nyassa, so far as molluses go, is thinly peopled,
and great extents of its shore and deep bottom are altogether
uninhabited. The above facts of distribution, indeed, give the

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 169

impression that the molluscan population of Nyassa retains its
character of an importation from the ponds and streams which
in the vast lake is, as it were, completely out of place.

Let us now turn to Lake Tanganyika, and compare the dis-
tribution of the molluscs in this lake with the observations I
have just described in relation to Nyassa. But let me first
point out that the physiographical features of Lake Tanganyika
are slightly different from those obtaining in the great Nyassan
valleys. Tanganyika is 2700 feet instead of 1500 feet above
the level of the sea; but notwithstanding this greater eleva-
tion, the climate of Tanganyika is appreciably hotter than that
of Nyassa.

The shores of Tanganyika are, perhaps, on the whole more
precipitous, and are certainly less extensively fringed with the
broad coast belts of modern alluvium so characteristic of the
margin of Nyassa. The southern half of Lake Tanganyika is
not nearly so deep as Nyassa, the water being generally little
more than from 900 to 1200 feet in depth; and it is not so
pure, being always impregnated with an appreciable taint of
several salts.

These slight physiographical differences which exist between
Nyassa and Tanganyika cannot, however, be considered as
having any potency to modify the individual fauna which the
lakes contain. For there is a far greater difference between
the physiographical features of Nyassa and the Victoria
Nyanza than between Nyassa and Tanganyika. Yet the
molluscan population of both the Victoria Nyanza and Nyassa
are essentially the same. We cannot, therefore, regard the less
amount of difference which is perceptible between Nyassa and
Tanganyika as in any way responsible for the wide faunistic
differences which exist between these Jakes. This view is also
in accordance with the very important fact that all the genera
found in Nyassa exist in Tanganyika also. There is no more
difference between the normal, non-Halolimnic molluscan popu-
lation of Tanganyika and Nyassa than there is between that of
Nyassa and Victoria Nyanza—a state of things which seems
to indicate clearly that the difference in physiographical

170 J. EH. S. MOORE.

features which exists between these lakes are incapable of
greatly modifying the forms they contain. Therefore whatever
wide difference in type and mode of distribution is apparent
between the faunas of Nyassa and Tanganyika must be the
expression of the difference of the animals themselves, and not
of the slight differences of condition under which they live.

In confirmation of this view, we find that all the Nyassan
molluscan genera are distributed both in Tanganyika and
Nyassa in a similar way, i.e. they are more or less restricted
to the coast-line and the shallow, sheltered places, such as
creeks and bays. For comparison my observations on the bathy-
metric distribution of the molluscs in Tanganyika have been

Tasie II].—Bathymetric Distribution of Molluscs
in Tanganyika.

Unio
Mutela
Tridina
Limnza
Tsodora
Physopsis
Planorbis .

Ampullaria .
Lanistes .
Cleopatra

Melania
Neothauma .
Typhobia .
Bathyanalia .
Paramelania .
Nasopsis .
Limnotrochus
Syenolopsis .
Lithoglyphus :
Unio Burtoni? .

Halolimnie

Group.

100 200 300 400 500 600 700 800 900 1000

The numbers at the foot indicate the approximate limits of depth in feet to
which the molluscan genera known to occur in Tanganyika extend, and
the thickness of the individual lines shows at what depth each genus
most abundantly occurs,

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. Let

epitomised in the same tabular form I used while speaking
of Nyassa. From this table it will be clearly seen that if we
exclude the Halolimnic fauna altogether, the bathymetric dis-
tribution of the molluscs both in Nyassa and Tanganyika is
approximately the same. In both cases the fresh-water
molluses are restricted to the sheltered, shallow portions of
the lakes. But directly we pass from the consideration of
the normal fauna to that of the Halolimnic forms the most
striking changes are at once observed. Instead of the Halo-
limnic molluscs being restricted to the shallow creeks and bays
about the coast, they swarm on the rough surf-swept rocks and
on the open beach. And what is more remarkable than this,
they extend in great profusion to the deepest portions of the
lake. Thus, dredging in water which varied in depth from
800 to 1200 feet, I always obtained plenty of Typhobia,
Paramelania, Bathynalia, and Bythoceras among the
Gastropods, as well as the so-called Unio Burtoni among
the Lamellibranchiata; and how far these genera extended be-
yond these depths I cannot say, but they showed no signs
of dying out, but rather the reverse. On the lake floors which
were not so deep as this, from 200 to 300 feet below the sur-
face, but which were yet deep enough to have yielded nothing
by dredging in Nyassa, there was an abundance of Limno-
trochus, Syrnolopsis and Neothauma, together with
those varieties of Melania which inhabit Tanganyika. It is
thus rendered apparent by these observations that the Halo-
limnic molluses are all either surf-swept rock dwellers, or
entirely deep-water forms. Unfortunately we are as yet
entirely ignorant of the distribution of the molluscs in any of
the great lakes besides the two which I have named. But as
the normal fresh-water fauna of Nyassa and Tanganyika have
the same bathymetric distribution, it is probable that these
same genera inhabiting the remaining lakes which have not
yet been investigated will be found to be similarly disposed.
It is thus apparent that the Halolimnic molluscs are com-
pletely dissociated from the normal fresh-water forms, along
with which they exist in Tanganyika, not only by their singular

172 J. E. S: MOORE.

geographical isolation, but by their bathymetric distribution
also; the conclusions to which the facts of their geographical
distribution seem to point being thus completely substantiated
from another point of view.

There are, however, yet other ways in which the fact that
the Halolimnic fauna is entirely distinct from, and unconnected
with the more normal series becomes clear. For in many
branches of biological inquiry we are often rightly guided
by impressions which, like the types of human physiognomy,
are real enough, but quite incapable of definite expres-
sion. Impressions of this character are at once produced on
reaching Tanganyika, as I did, after studying the fauna of
several neighbouring lakes. For there is a singular and oceanic
profusion of life in Tanganyika, which is quite peculiar, and it
quickly becomes evident that this numerical increase in the
aquatic population does not affect the normal fresh-water stock,
it is solely produced by the astonishing abundance of the
members of the Halolimnic group. In contrast with the shallows
of Nyassa, the creeks and bays of Tanganyika swarm with crabs
and prawns, and the open sandy beaches are strewn with
empty Halolimnic shells ; dead detached fragments of the deep-
water sponges are tossed up by hundreds on the shore. And
on the extensive rocky coasts the barely submerged stones are
covered with the so-called Lithoglyphus and Nasopsis, just
as the half-tide rocks swarm with Natica and Litorina on
an English beach. Further, on putting out into the lake
itself, the deep open water is filled and discoloured with clouds
of pelagic Protozoa (chiefly Peridinia and Condylostoma);
and during the dry season swarms of the lake jelly-fish are
seen pulsating at all depths.

Recapitulating, it may be said, then, that the facts of
the geographical and bathymetric distribution of the great
lake molluscs lead to the following results :—That among all
the fresh-water lakes of the African continent which have
hitherto been explored there exists a type of fauna which is
curiously similar throughout. It differs only in the specific
representation of the same genera which these lakes contain.

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 1738

This generalised African lake fauna contains only those families
and genera of molluscs which would be regarded as typically
fresh-water, lake, river, and pond dwellers, in whatever con-
tinent the fresh water might occur. In one African lake,
however, but in one lake only, there have been found to exist,
superadded to this normal lacustrine stock, a number of
Gastropods which do not closely resemble any other forms
either living or extinct; these molluscs are also completely
dissociated from the remaining normal series of the lake in which
they occur by their modes of life. Together these molluscs
constitute the molluscan section of a whole faunistic series,
which in Tanganyika is added to the normal fresh-water
stock the lake contains. This fauna forms what I have called
the Halolimnic group, and the tout ensemble of all the
Halolimnic genera is marine.

To account for the presence of the Halolimnic organisms
in Tanganyika, only three hypotheses which are even tempo-
rarily tenable can be found. It may be supposed—

1. That they have arisen as modifications of the ordinary
fresh-water fauua through prolonged isolation in the lake ;

2. That they are the surviving representatives of an extinct
fresh-water stock ; or—

3. That they are comparatively recent importations from
the sea.

Let us examine each of these three possible explanations
in the light of the new facts of distribution which have just
been detailed. Unless the conditions affecting the fauna of
Tanganyika have been permanent for a greater period of
time than has been the case with any of the other lakes, they
could not have produced the Halolimnic fauna which this lake
now presents. Unless we make the further suppositions (1) that
the conditions in Tanganyika have been permanent, while those
affecting the fauna in all the other lakes have changed so much
as to kill off the Halolimnic forms they once possessed ; or
(2) that all the other lakes are much younger than Tanganyika,
and that therefore the Halolimnic fauna has not had time to
develop in them yet. There is no evidence for either of these

174 J. EY Ss. MOORE:

views, and there is direct evidence to show that Nyassa has been
a fresh-water lake longer than Tanganyika. On the shores of
Nyassa there are old raised beaches, forming white limestone
cliffs which contain the fossilised remains of the shells now
living in the lake. But in these old lake beds there are no
traces of any Halolimnic forms, and this is all the more con-
clusive as the shells of the Halolimnic molluscs are much more
solid and durable than those of the fossilised fresh-water
forms.

The second hypothesis, that which suggests that the Halo-
limnic fauna may be the surviving representative of an ancient
fresh-water stock which has become extinct, has great attrac-
tions, as it conforms to a famous geological speculation,
and has at first sight the appearance of a certain modicum
of positive support. For the shells of the Paramelanias of
Tanganyika have been independently supposed by White and
by Tausch to be identical with the extinct estuarine or
brackish Pyrguliferas of cretaceous Europe, America, or
Africa.!

The type of shells possessed by these forms has been,
however, repeated so often by so many widely separated
molluscan types, such as in the Melanias, Litorinas, Pur-

1 On further examination it appears—(1) That the genus Paramelania
of Tanganyika is similar to the eretaceous Pyrgulifera; (2) but that the
genus Pyrgulifera, so far as some of its representatives go, is concho-
logically indistinguishable from the old marine Jurassic genus Purpurina,
and that the Nanopsis of Tanganyika corresponds to one section of this
genus, the Paramelania to the other. (See Hudleston’s figs., Plates i
and ii, and text p. 85—95, ‘ Jurassic Gasteropoda,’ Paleeontographical Society,
vol. xli, 1887.) It would thus appear that the marine genus Purpurina
became a fresh-water form, as so often happens in Cretaceous times. We
find, however, that other Halolimnic Gasteropods, Bathanalia, the so-
called Lithoglyphus, and Limnotrochus, are also indistinguishable
from marine Jurassic forms, which are not found in any Cretaceous forma-
tion, fresh-water or marine. Consequently the geological evidence on this
matter distinctly favours the old marine origin of the Halolimnic fauna; but
it places their original marine existence much further back than I had even
dared to suggest. I shall discuss this most interesting line of investigation
fully in a special memoir.

THE MOLLUSCS OF THE GREAT AFRIOAN LAKES. 175

purinas, and the like, that it is pardonable if zoologists re-
quire something more than merely conchological characters to
establish an identity among these forms.'' But this supposed
homology between the shells of a living and extinct species of
Gasteropod (about the anatomy of neither of which up to the
present anything whatever has been known) is the one
fragment of positive evidence which can be produced in favour
of the relation of the Halolimnic fauna to an extinct fresh-water
stock.

The hypothesis, moreover, is combated by the same objec-
tions emanating from the facts of distribution of the Halolimnic
animals that were fatal to the first hypothesis, and they have
here equal force. If the Halolimnic fauna of Tanganyika is
the remnant of an old African fresh-water stock, it must have
been present at one time in all the lakes which are as old as
Tanganyika; but we have seen that with respect to Lake
Nyassa this does not appear to have been the case. It is very
improbable that many of the remaining so-called rift-valley
lakes are not as old as Tanganyika, yet we have seen that they
do not contain the Halolimnic forms. Therefore, in order to
support this second hypothesis, we shall be obliged to have
recourse to hypothetical catastrophes which must be supposed
to have destroyed the Halolimnic fauna in every lake but one.
Hypotheses of this sort spring, however, from the carcass of a
theory only after it is dead, and our second hypothesis is there-
fore opposed to the facts of distribution as they at present stand;
its acceptance would, moreover, be revolutionary to many
zoological conceptions of the present time. It would neces-
sarily lead us to believe that deep-water crabs may be indige-
nous fresh-water forms; that deep-water Gastropods and
sponges were common in Cretaceous times ; that jelly-fish
were once fresh-water organisms, and so on through a number
of consequences, which, when the nature of the evidence sup-
porting the original hypothesis is weighed, must seem little
better than grotesque. On the other hand, all the facts of

1 Tam quite aware that this statement cuts at the roots of many geological
determinations ; but I am prepared to maintain that the criticism is sound.

176 J. E. S. MOORE.

distribution and the like, as well as the superficial character
of the Halolimnic animals themselves, are absolutely in accord
with the third hypothesis, i.e. that the Halolimnic fauna is a
relatively recent importation from the sea.

But before accepting this conclusion, as the natural teaching
of the facts and observations which we have been discussing,
it is absolutely necessary to be quite sure that in the nature of
the country itself—that is, in the past geological history of
Africa—there is nothing which renders impossible the realisa-
tion of sucha theory in fact. Now, on turning to the geological
aspect of the questions we have just discussed, it is apparent
that there is an accumulation of negative evidence drawn from
what is now known to geologists of the nature of the African
interior, which, although it does not specifically favour the
view of the ancient fresh-water origin of the Halolimnic forms
certainly renders evident a gap in the confirmation of the
theory of their marine origin.

In 1852 Sir Roderick Murchison! advanced the hypothesis
that Africa, south of the Sahara, was a continent of great
antiquity aud simplicity, the greater part of which has never
been changed or covered by the sea, at any rate since the age
of the formation of the new red sandstone. This theory has
appeared to be supported by the discoveries of Livingstone,
Burton and Speke, and Speke and Grant, and it was finally
re-advanced and summarised by Murchison in 1864,? when
he described this part of Africa as geographically unique “ in
the long conservation of ancient terrestrial conditions.” But
he immediately fell into the now exploded error of assuming
that ‘this impression is further supported by the con-
comitant absence throughout all the larger portion of this
vast area, i.e. south of the equator, of any of those volcanic
rocks which are so often associated with oscillations of the
terra firma.” This latter speculation is now shown to be in
no sense true, for there is abundant evidence of volcanic action
and of voleanic materials all the way from Kilimanjaro and

1 Murchison, President’s Address, ‘ Journ. Royal Geog. Soc.,’ vol. xxii, 1852.
2 «Journ. Royal Geog. Soc.,’ vol. xxxix, 1864, pl. xxxvii, pp. 201—205.

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. Yat

Ruwanzori in the north, to the little group of volcanic cones
near the coast of Lake Nyassa. “The first part of Murchison’s
theory, however, which affirms that Central Africa has never
been below the level of the sea, is still in harmony with the
known geological facts, for no deposits of a certainly marine
origin have as yet been discovered in the interior.” The sedi-
mentary rocks described by Burton and Speke to the west of
the Victoria Nyanza have yielded no fossils to indicate the
conditions under which they were formed. The triassic
Ganoids and Gastropods unearthed by Drummoud! at the
north end of Lake Nyassa have been generally regarded as
fresh-water forms.2 The great red sandstones and shales
which stretch from the north of Nyassa far up the coasts of
Tanganyika, which were examined by Joseph Thomson, and
more recently by myself, have not yet been found to contain
any animal forms; the only indication which might lead to a
belief that fossiliferous rocks occur in these regions being the
fact that the natives of the west coast of Tanganyika are said
to wear necklaces of beads which they dig out of limestone
rocks, and which, if this statement is true, are probably the
disarticulated segments of crinoid stems.

Marine, Jurassic, triassic, and probably carboniferous
deposits have been found along the coast at many points from
Mombasa to the Cape, but these have never been shown to
extend any distance inland, and they seem to have no connec-
tion with the great sedimentary deposits of the interior, such
as those north of Lake Nyassa, which underlie Drummond’s
fresh-water triassic (?) beds. There is thus at present no
geological evidence of the sea, or of even an arm of the sea,
ever having been in the region of Tanganyika within reason-
able geological times.

There has, however, been steadily accumulating a mass of
observations relating to the formation of the so-called rift
valleys, the general tenor of which has been to reveal a

1 Drummond, ‘ Tropical Africa.’
2 It appears, however, that these fossils have been by no means satisfac-
torily described.

vou. 41, part 1.—NEW SERIES. M

178 J. E. 8. MOORE.

great instability of the regions in which Tanganyika lies; an
instability which has been quite sufficient to coitveivably
account for any amount of upheavals and depressions which
may have been requisite for the marine contamination of that
lake. The valleys in which the north of Nyassa, Lakes Tan-
ganyika, Albert Edward, Albert, Baringo, Rudolph, and about
twenty-seven minor lakes lie, are really part of a connected
series of depressions formed by faults which run approximately
north and south through an immense distance, and can be
traced as far as Berbera on the Red Sea, thence north along
the Red Sea shore itself, the coasts of which are to a great
extent of similar formation, and they terminate finally in or
about the Dead Sea, and the valleys of the tributaries of the
Jordan.t All the country traversed by this immense series of
faults from the north of Nyassa in the south, to the ancient
sites of Sodom and Gomorrah in the north, is filled with
native traditions of catastrophes, of floods, of earthquakes, of
volcanic outbursts, and the like ; and the geological investiga-
tious of Gregory 2 and others have shown that much of the
above faulting and volcanic activity must have occurred,
geologically speaking, in quite recent times. The existence
of these singular rift-valley faults has divided the African
lakes into two distinct series ; one series of lakes being always,
like Nyassa, Tanganyika, and Rudolph, long, narrow, and deep ;
the other, like the Victoria Nyanza, Bangweolo, and Shirwa,
broad, shallow, and round.

We have, therefore, evidence of great geological instability
in the very regions in which the Halolimnic animals now live ;
but, like the paleontological record, it affords no insight as to
how or when, if Tanganyika ever was connected with the sea,
this connection could really have been made; but it is a sin-
gular fact that the one lake in which the Halolimnic animals
now live is that which les at the bottom of the biggest and
most conspicuous inland rift.3

1 See Suess, ‘Die Briicke des Oust Afrika.’
+ The Great Rift Valley.

* The southern two thirds of Nyassa is not in a rift; and in contrast to

THE MOLLUSOS OF THE GREAT AFRICAN LAKES. 179

From all this it will be seen that, unless we are to assume
that the Halolimnic group came into Tanganyika from the sea in
very ancient times indeed, and that they are far older than their
characters in any way appear to warrant;! we are without any
direct evidence from geology that the sea, or even an arm of
the sea, has ever been in the Tanganyika region of the interior.
So far as positive evidence goes, geology is absolutely silent
upon this subject, it offers no evidence of any sort; and the
theory of marine contamination—if it occurred, let us say,
during a later period than Jurassic times—is thus diametrically
opposed to a geological theory of the nature of the African
interior which is at present accepted by many competent
authorities.

The only way in which the nature and origin of the Halolim-
nic group can be really satisfactorily determined is, therefore,
through a minute knowledge of the morphology of the in-
dividual members of the group themselves, and the best types
belonging to the Halolimnic group for this kind of work are
the Gastropods, because these organisms, unlike the lake
Medusze, can be more or less directly compared with all sorts
of analogous organisms, ancient and modern, fresh-water and
salt. Ifit can be shown from the study of their morphology
that the Halolimnic Gastropods in Tanganyika are really mor-
phologically most closely related to the fresh-water Gastropods
at present known, then the theory of the ancient fresh-water
origin of the Halolimnic group is probably true. If, on the
other hand, it turns out that the Halolimnic Gastropods are
really most closely related to typicaliy marine genera, then
there will be little doubt that the Halolimnic group originated
in the lake through marine contamination, and geological
conceptions will have to make room for the fact of the interior
of Africa having been connected with the sea as best they can.

In arriving at the conclusions contained in the preceding
Gregory I do not believe that the north of Nyassa lies in the main eastern
rift, but in one which through Lake Rukwa is continuous with the western
Tanganyika series. See Gregory’s ‘ Rift Valley,’ p. 7; also my paper in the
‘ Journal of the Royal Geographical Society,’ September, 1897,

See foot-note on page 161.

180 J; E. S) MOORE:

paragraphs I have virtually fulfilled the object which I had
before me in collecting and examining the facts concerning
the distribution of the African lake faunas, before entering
upon any detailed examination of the evidence which can be
gathered from the study of the morphology of the Halolimnic
animals themselves. We have seen that the collateral evidence
afforded by the facts of distribution and the like at once clear
away the likelihood of the Halolimnic group having originated
at any time or in any manner, de novo, in Tanganyika, and
that there is finally brought on a more or less direct issue
between the supposition of an ancient marine contamina-
tion of Lake Tanganyika and the ancient fresh-water
origin of the Halolimnic group. All the facts of dis-
tribution which we have examined appear to me to strongly
favour the former of these hypotheses; and although we are
at present ignorant of the precise manner in which the marine
contamination of Lake Tanganyika may have been effected,
there is no positive geological objection to the view that it has
occurred, while there is the certainty of a sufficiently great
geological instability throughout the very districts in which
Tanganyika lies to have easily accounted for it.

THE MOLLUSCS OF THER GREAT AFRICAN LAKES. 181

The Molluscs of the Great African Lakes. —
II. The Anatomy of the Typhobias, with a
Description of the New Genus (Batanalia).

By

J. E. S. Moore.

With Plates 11—14.

No entire specimen of Typhobia has hitherto been described,
and we have consequently remained entirely in the dark as to
the real morphological character of what is probably the most
remarkable fresh-water Gastropod at present known.

Presumably from the characters of its empty shell this genus
has been classed by the conchologists! with the Melanias, as
a new sub-section of that group.? But into what serious error
determinations of this sort may lead, when based on concho-
logical evidence alone, the present paper, which contains the
first anatomical description of the mollusc, will suffice to show.

It will be seen that the structural features of the Typho-
bias, so far from establishing the above conchological antici-
pations, in every way confirm the conclusions at which I
arrived respecting the marine origin of these molluscs from a
study of the distribution of the African lake fauna in general.’
Hence the actual facts of anatomy are, as I anticipated from

1 Smith, ‘ Proc. Zool. Soc.,’ 1881, p. 276.
® Fischer, ‘ Manuel de Conchyliologie,’ p. 705. These determinations have
been particularly unfortunate, as they have masked the marine, and conse-
quently intensely interesting character of the molluscs of the lake.
“The Molluscs of the Great African Lakes.” I. “ Distribution,”
‘Quart, Journ. Mier, Sci.,’ present number, p. 159,

182 J. E. 8S. MOORE.

the facts of distribution that they would be, directly in conflict
with all those geological speculations respecting the interior of
Africa that have been hitherto more or less generally held.
In the paper to which I have referred! it was seen that the
Typhobias belong to, and are, one of the most remarkable
constituents of the gwasi-marine or Halolimnic section of
the Tanganyika fauna. They consequently share, along with
the other members of this group, the strange geographical
isolation which is its distinctive mark.

Like nearly all the Halolimnic animals, Typhobia is found
pretty abundantly in Tanganyika, occurring in some places in
the most astonishing profusion, but, so far as it is at present
known, the mollusc is found living nowhere else in the world.
I first obtained the empty shells of T. Horei on the long sandy
beaches near the south-west corner of the lake, and subsequently
on the southern shore of the deep Kituta Bay. They were
readily recognised by the head men of the villages, who told
me they had never seen the Gastropod alive, but only the shells
when washed up empty along the beach. From this statement
of the natives, and from the spinous character of the shells,
I thought it probable that they would be found living on mud,
but I was unable to find them in the muddy reaches among
the Kinyamkolo Islands,in depths of fifty to one hundred feet,
nor indeed in any portions of the lake that were of similar
depth. It was not until I had extemporised a primitive deep-
water dredging apparatus that I obtained the Gastropod alive.”

In June, 1896, we were on the west coast of Tanganyika
and on the southern shores of Cameron Bay, and here I was
able to obtain the strong bark rope used by the Wafipa fisher-
men for their nets. As these nets are hauled by rows of men
on the ropes at either end, the ropes themselves are strong
enough to drag a heavy net with all its weights and stretchers
over several hundred yards of ground; and with them I was

Loc. cit
2 My ordinary dredges were smashed almost at once by the sharp rock-
ridges which protrude through the muddy floor of the lake. For this deep
water I used a native basket, weighted down with stones.

THE MOLLUSOS OF THE GREAT AFRICAN LAKES. 183

consequently enabled to dredge in water that varied from
500 to 850 feet in depth. Eventually in this manner, during
the months of June and July, about a hundred Typhobias were
obtained alive. Of these some were examined on the spot, some
preserved in various ways, stored in spirit, and eventually brought
back. The living Typhobias were associated with another
deep-water Gastropod, also alive, a brief description of which
will be found at the end of the descriptive part of this paper.

Except in the characters of the shell, this new genus 1s
almost identical anatomically with T. Horei, consequently it
is unnecessary that I should do more than point out in what
it differs from the form already known.

External Characters.—The general appearance of a
living Typhobia is seen in fig. 1. They are always very active
when brought up from the deep water they inhabit, probably
being uncomfortable through the decreasing pressure. The
tentacles are very long and slender, and the eyes completely at
their base; the snout is wrinkled and very much pigmented
on the upper surface, and it is so long and slender as to suggest
the ordinary protrusible snout or introvert of Prosobranchs ;
on dissection, however, it is seen to be simply elongated ex-
ternally, very retractile, but, like that of Pterocera, net intro-
vertible in any sense. The foot is very broad, and of the same
pale semi-transparent yellow as that of Anodonta. The mantle
is prolonged into the well-marked anterior and posterior
siphons (fig. 2).

As is the case with many other fresh-water Gastropods, the
shells of Typhobia Horei vary to a remarkable degree ;
indeed, the extreme forms when isolated, and not linked to-
gether by the innumerable intermediate forms which actually
occur in Tanganyika, differ so widely that the French concho-
logist | Bourguignat regarded these differences as sufficient to
split the genus up into four species, under the new name
of Hylacantha ; his four so-called species being respectively
H. Horei, H. Bourguignati, H. longirostris, and H.
Jubertii. Had this author, however, been able to obtain

1 «Ann. des Sci. Nat.,’ septiéme série, ix, x, 1890, p. 125.

184. J. E. S. MOORE.

large numbers of these shells on the spot, and to have made
collections of the extreme and intermediate varieties, it is hardly
conceivable that he would have ever regarded their variations
as specifically distinct. Be thisas it may, however, it is quite
evident, when large numbers of these shells are studied, that
their varieties cannot be regarded as specific forms. The shells
of Typhobia Horei are, however, undoubtedly polymorphic,
for there are about four well-marked varieties, into one or
another of which the great majority of the specimens I collected
tend to fall (figs. 18, 16, 23, 25). When the shells are very
young, about the time of birth (figs. 27, 28), they are desti-
tute of all but the merest trace of spines as well as of the
pronounced-rostral beak, which is a marked structural charac-
teristic of the majority, but by no means of all the older
shells. Now the fact that in some adult shells the rostral
beak is entirely wanting (figs. 16 and 20) shows that, in
this respect, such shells have not deviated from their embryo-
logical character, All the four extreme types of variation
graduate off into forms which, in the more or less complete
absence of a rostral beak, approximate to the shells repre-
sented in figs. 16 and 25. Such shells, therefore, may be said
to represent the extreme of least specialisation. The remaining
extreme polymorphs (figs. 18, 23) can be all traced through
successive stages from the types represented in figs. 12, 17,
18, and 21. Thus between the extremes, figs. 13 and 16,
there are intermediate forms, such as figs. 12 and 17; the
extreme, fig. 23, has intermediates, such as figs. 18 and 21,
while the extreme, fig. 25, is connected up by forms similar
to that represented in fig. 20 and perhaps 21. Any number
of intermediate stages could have been represented for each
series ; but for obvious reasons those only have been selected
which seemed to best express the transition from any one type
to the next. It would thus appear that Bourguignat’s four
species were formed in the absence of an adequate quantity ot
material to work upon, and this conclusion is finally clinched
by the fact that the anatomical characters of the soft parts of
the extreme variations are indistinguishable from one another.

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 185

We may, therefore, conclude that the name Ty phobia Horei,
as given by Smith,! stands rightly for but one species after all.
The nervous system.—The nervous system of Ty phobia,
both in its general relationships and in the details of its different
constituent parts, is almost entirely unique. Viewed from above,
it is at once obvious that there is a great condensation and
fusing together of the chief ganglionic masses, no commissure
being visible externally between the cerebral ganglia (fig. 5, 5).
Each cerebral ganglion gives off anteriorly two sets of nerves,
one obliquely above the other (fig. 34, 7,3); the upper and
external arises from a prolongation of the ganglion comparable
to the “ saillie labiale.” These nerves are distributed to the
tentacles and the eyes. From each cerebral ganglion below
the “ saillie” there arises another set of buccal nerves, the two
innermost of which pass forward, enlarge into the buccal
ganglia (fig. 34, 78, and fig. 5, Z), and unite again below the
mouth. The remaining members of this set of nerves are dis-
tributed to the buccal mass, and to the parietes of the anterior
portion of the head and snout. Ganglionic cells extend along
the buccal nerve trunks as far as the buccal ganglia. Late-
rally each cerebral ganglion gives off a number of small nerves
distributed to the head (figs. 84, 19). Towards the posterior
upper surface of the cerebral ganglion the paired otocyst
nerves arise, and pass obliquely backward over the pleuro-
pedal commissures to the enormous otocysts (fig. 35, 6).
Below, the cerebral ganglia are connected with the pedal
ganglia by the rather loug cerebro-pedal commissures (figs. 6,
and 35, 20). Immediately behind these there is found on each
side a second commissure, which at first sight appears to pass
from the cerebral ganglion also (fig. 835, 77). In reality this
is the pleuro-pedal commissure, the pleural ganglion being dis-
placed forward, so as to lie closely applied to and immediately
beneath the cerebral ganglia. In order to understand the rela-
tions of these ganglia it is necessary to examine several sections,
as they cannot be seen by the ordinary methods of dissection.
Fig. 9 represents a section taken through a point marked X in

1 € Proc. Zool. Soc.,’ loc. cit.

186 J. E. S. MOORE.

the general figure of the nervous system given in fig. 35. It
shows the cerebral ganglion (a), separated from the pleuro-
pedal commissure (0), while th fore-part of the pleural ganglion
is seen as a continuation of th.s commissure (c). Fig. 10 isa
little further back, and shows the pleural ganglion (a) and the
posterior portion of the pleuro-pedal counective (6) But the
section is still in front of the connection between the pleural
and cerebral ganglia, both ganglia appearing separate. Fig. 11
is slightly further back again, and shows the pleuro-cerebral
connective (a). There is visible also the posterior portion of
the cerebral ganglion immediately before it passes downwards
and is merged in the pleural ganglion itself. The pleural ganglia
are thus seen to be displaced, and their real position is indicated
by shading in the general arrangement of the nervous system
(figs. 834 and 85). The pleuro-pedal connective is consequently
shorter than it would be if the pleural ganglia were in their
normal positions, while the pleuro-cerebral connective, as
such, may be said to be almost entirely wanting.

On the left, the pleural ganglion is continued below and
very slightly across the subcesophageal space (figs. 5, 6, 6, 3,
34, 16), as an enormous ganglionic trunk, in the course of which
the subintestinal ganglion is superficially quite indistinguish-
able, but the locus of this ganglion is marked by the great
pallial nerve (fig. 5, 7). The ganglionic character of this
left cord continues, as is shown by the presence of ganglionic
cells, for a long distance, the appearance it presents at the
point marked x’ (fig. 35) being represented in section at fig.
37, 1. The right pleural ganglion gives off a relatively small
nerve, which, after passing ecbliquely over the cesophagus,
carries the supra-intestinal ganglion, from which nerves
branch to the gill, osphradial ganglion, and to the left pallial
anastomosis (figs. 34, 85). This anastomosis is formed in the
usual way by a rather large nerve passing out from the left
pleural ganglion, and meeting the branch from the supra-
intestinal ganglion near the angle of the gill (fig. 35, 76).
The nervous system is, therefore, dialyneurous on the left, and
the relations of the nerves here give no indication of the

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 187

extraordinary state of asymmetry which is encountered in the
same region on the right. On this side the pleural ganglion
gives off a nerve (fig. 34, 8) which appears at first sight as if
it would form the right pallial anastomosis, either in the
region of the subintestinal ganglion or along the course of the
-right pallial nerve. This nerve, however, after passing directly
-outwards for some distance bends sharply forwards, branches
once, and each of the twigs diminishes rapidly, and dies out in
the parietes of the mantle and the body-wall. The great
mantle nerve which is given off from the subintestinal gan-
glionic trunk (fig. 34, 20) passes outwards, and is also distri-
buted, without forming any connection with the _ pleural
branch, to the mantle of the right side. Neither is there any
connection between the pleural and subintestinal ganglion
beneath the cesophagus. We have thus on the right side a
condition of things which is almost unique among all the
Streptoneurous Prosobranchiata which have hitherto been
investigated. It is neither zygoneurous nor yet dialyneurous,
and the condition of these parts finds its only analogy in the
rather unsatisfactory descriptions given by Bouvier! of the
nervous system of Solarium and the Scalarids. In general ar-
rangement, and apart from the above singular feature, the
character and arrangement of the cerebral, pleural, and intes-
tinal ganglia with their nerves and connections show a marked
and indisputable similarity to those of the corresponding parts
in such forms as Strombus, Pterocera, Cancellaria,
Voluta, and their associates. The wide distribution of the
ganglionic cells in the nerve-cords of Typhobia is a remark-
able and undoubtedly primitive feature ; while the fact that
the nervous system of Typhobia foreshadows and is similar
to those of several rather widely separated modern marine
genera is direct and incontrovertible evidence, so far as it goes,
that these molluscs are old and modified marine forms.

The pedal ganglia project forwards and are curiously ex-
tended in front by two colossal nerves (fig. 85, 13), which at
their points of origin possess ladder-like connections one with

' © Ann. des Sci. Nat.,’ ili, iv, 1887, pp. 156—167.

188 J. E. S. MOORE.

another (fig. 35, 74), and thus approximate to the primitive
type of pedal nerves possessed by the Helicinide. The pedal
ganglia themselves are united together by a great transverse
commissure (fig. 835) which contains ganglionic cells ; and on
their postero-lateral surfaces they give off four or five large
nerves which pass into the foot. The pedal ganglia are con-
nected with the cerebral, and the pleural ganglia by the cerebro-
pedal and pleuro-pedal connectives already described.

The otocysts of the Typhobias are relatively immense, and
each is innervated by two fine nerves springing from the upper
portion of the cerebral ganglia (fig. 35, 6). The position
occupied by the otocysts is very anomalous, being completely
above and separated from the pedal ganglia, close to the cerebro-
pleural ganglionic mass (fig. 35, 0). The otoliths are numerous
and small (fig. 36), and the otocyst is lined by a well-marked
sensory epithelium, from the free internal surfaces of the cells
composing which there are given off fine ‘‘ sensory processes ”
projecting into the cavity of the sac (fig. 36, 7). The posi-
tion and character of the otocysts, and the prolonged pedal
ganglia, are features in keeping with the generally primitive
characters which the other parts of the nervous system seem to
possess.

The digestive system presents points which, like those
appertaining to the nerves, are at once interesting and new.
The buccal mass is exceedingly small, and the radular sac is
very short. There are no horny jaws, and the radular dentition
is unique and peculiar in the extreme. A single transverse
row of teeth is represented in fig. 43; also in the upper figure
on page 189. The salivary glands are long and branched, the
cesophagus being long, slender, and longitudinally folded. In-
ternally it is lined by ciliated and glandular cells. Posteriorly
the cesoph agus opens into the right side of the stomach, which
is divided into an anterior and posterior chamber, the cesophagus
opening intothe latter. This posterior chamber of the stomach
is traversed by several marked folds, the most conspicuous of
which extends longitudinally (fig. 44, 7). On the floor of the
stomach, to the right of this fold, are found the openings of the

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 189

“ bile-ducts ” (fig. 44, 2,3). The anterior chamber is separated
from the posterior chamber of the stomach by a constricted annu-
lus, and the anterior chamber encloses and is almost filled by a

Central and three lateral teeth of the radula of Typhobia (upper figure) and of
Bathanalia (lower figure). Three lateral teeth on each side and a central
tooth constitute the unit, which is repeated row after row along the radula.

large crystalline style, represented in fig. 44,5). The whole
arrangement of the stomach, the position of the folds and aper-
tures, the separation into a posterior and an anterior chamber
or cecum, and the presence in the latter of a crystalline style
are all similar to the condition of things obtaining in Pterocera.
The style and its sac are undoubtedly homologous with the

190 J, EVES MOORE:

structures described by Collier, Huxley,” Haller,? and others,
and the morphological conclusions which can be drawn from
the nature of the stomach are in harmony with those which I
have pointed out in reference to this Gastropod’s ganglia and
nerves (see p. 187).

On leaving the stomach the intestine bends twice in the
manner represented in fig. 42, and towards its recta] extremity
it is considerably enlarged (fig. 49, Z). This enlargement
contains the curious glandular fold represented in fig. 46, 1.
The anus is carried on a slight projection of the rectum from
the mantle wall, and during life is slightly in advance of the
margin of the mantle (fig. 2, 7).

The “Liver” occupies the lower portion of the upper
whorls of the shell (fig. 2, 6), and has the usual characters of
a digestive gland. ‘The “ bile-ducts” open by two orifices in
the fioor of the stomach, behind the pyloric aperture.

The Kidney occupies the region behind and to the left of
the heart (fig. 3, 6), and opens by a single minute pore, quite
at the posterior extremity of the mantle cavity.

The Heart and Gills.—The heart is simple, lying rather
obliquely at the end of the mantle cavity. There is a large
pericardial chamber (fig. 3, 5). The ventricle tapers from
before backwards, and is surmounted by a large, rather thin-
walled auricle, which in turn receives the pulmonary vein.
There are well-formed valves between the auricle and the
ventricle (fig. 48, 7), and between the ventricle and the aortic
trunk (fig. 48, 4). From the aortic trunk the anterior and
posterior aorte diverge in the usual way (fig. 48, 7, 8).

The gill in Typhobia (fig. 8,7) is very long, extending
from the base to the margin of the mantle cavity. It is com-
posed of simple broad-based triangular leaves, the apices of
which are elongated. The osphradium lies at the base of a
groove; it is long and simple, not fimbriated or gill-like, in
fact a mere ridge (fig. 8, 8). This ridge is innervated by a

1 Collier, ‘ Edin. New Phil. Journ.,’ vol. vii, 1829, pp. 230, 231.

2 Huxley, ‘Phil: Trans; 1853, ps 10:
3 Haller, ‘ Morph. Jahr.,’? Bd. xix, 1893, pp. 582—584.

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 191

nerve which springs from the small osphradial ganglion.
Externally the ridge is covered with ciliated and glandular
cells, the relations of which and the characters of the osphradial
nerve are shown in section in fig.8. There is nothing peculiar
about the gills except their great length.

The Reproductive Apparatus.—In Typhobia the sexes
are distinct and the female viviparous, the whole reproductive
apparatus being simple but somewhat peculiar. The ovaries
and testes occupy the upper surface of the last two whorls of
the spire, and in the female the eggs, with their bright green
yolk, pass directly into the simple oviduct (fig. 47, 7). From
this they reach the lower expansion of the oviduct (fig. 47, 2) ;
and in this sac, which functions as a brood chamber or uterus,
they go through the greater part of their development. The
walls of the sac are very thin, and while the animal is alive the
bright green yolk of the eggs is distinctly seen through the
delicate semi-transparent shell, so that the sexes can be distin-
guished at a glance. The sac opens near the rectum, at the
junction of the mantle and the body wall (fig. 3, 3). The
mollusc breeds during the months of June and July.

The testis (fig. 54, 6) opens by several small collecting
channels into the simple vas deferens (fig. 54, 4). This tube
becomes somewhat but not much convoluted on its way, and
ultimately expands into the curious enlargement represented
in fig. 54, 2. On opening this it was seen in every case to bear
about six singular parallel folds (fig. 49, 6). Beyond this ex-
pansion there is a curious finger-like outgrowth, extending from
the duct into the mantle wall; this process contains a muscular
mass, which has all the appearance of, and probably is, an
introvertible penis (figs. 45 and 46, 4). The lower extremity
of the male genital duct opens by an elongate slit (fig. 49, 5).
The possession of a penis in the mantle wall is a most curious
fact, which would seem to indicate that that organ is analogous
to, though very likely not homologous with, the penis of the
Ampullariz and other pulmonate Prosobranchs. In Typhobia
the male genital gland is extremely interesting from a cyto-
logical point of view, as this genus is one of those Prosobranchs

192 J. E. S. MOORE.

which, like Murox and Paludina, possess two forms of
spermatozoa.

The small normal variety appear to arise one division
after the heterotype which terminates the synaptic phase,
and the cells out of which these normal spermatozoa are
directly formed are, as in most cases, extremely small when the
actual characters of the spermatozoa are taken on. On the
other hand, the cells which directly metamorphose into the
large spermatozoa, or megasperms, are very large, being
similar in size and character to the synaptic (growing-cells)
themselves. As in a former paper! I have advanced the view
that after the synapsis any cellular generations which may
exist are to be considered as potentially ova or spermatozoa,
as the case may be, this fact that during the course of the
spermatogenesis in Typhobia the small spermatozoa appear to
arise two divisions after the formation of the synapsis, while
the megasperms appear to be produced directly from the
synaptic cells themselves, is extremely interesting.

Bathanalia.

The new generic type among the Gastropoda for which I
propose the name of Bathanalia is at present represented by
one specific form, Bathanalia Howesi, Pl. 12, figs. 29, 30,
31, 83. As the name implies, this species is found in associa-
tion with T. Horei in the deep water of Lake Tanganyika,
while in its anatomical features it is so similar to Typhobia
that no special anatomical description is required.

There is only one point in Bathanalia that needs mention.
I found after great difficulty, and by the help of sections, that
in this genus there is a very slight but quite distinct pallial
anastomosis on the right side.

It is entirely on account of the remarkable characters of the
shell that I have thought it necessary to separate Bathanalia
as a distinct genus from Typhobia.

The shell, as will be seen from fig. 29, is conical, composed of
eight angular whorls, which from apex to base carry numerous

1 Quart. Journ. Micr. Sci.,’ vol. xxxviii, p. 292.

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 193

short spines. The whorls are strongly sculptured, and the
columella is open (fig. 33).

The mouth is ovoid, somewhat angular, and the mantle
during life is prolonged into the last spine, forming a kind of
false siphon (fig. 30, 7). The radula is shown in the process
block, p. 189, and also in Pl. 12, fig. 32.

Comparative.

In establishing by comparison the true affinities of the
Typhobias, I have purposely left undeveloped all those ques-
tions respecting the validity of the current systems of classifi-
cation which such a comparison will inevitably raise. This
course has been taken because I am now confident that the
Halolimnic molluscs are among the few remaining indications
of an ancient sea that once extended to or near the Tanganyika
region of the present day. The anatomical features of the
Halolimnic molluscs when studied together should therefore
throw a most important light on the inter-relationships of
those more modern marine genera and families which it has
hitherto been so hard to solve. It would be premature and
most unsatisfactory to view these questions from the anatomy
of a single Halolimnic type; for this reason I have reserved a
discussion of these wider matters until I have had time to publish
anatomical descriptions of the remaining Halolimnic forms.

The Typhobias have been classed by the conchologists
among the Melanias, being regarded by Fischer! as a section
of this group equivalent to Faunus or Melanopsis. Judged
by their conchological characters alone it is by no means easy
at first sight to understand why such a classification was ever
made, as the Typhobia shell is almost as unlike any known
Melania as that of a Pteroceras ora Cone. In assigning
a systematic position to any animal concerning which the
morphological study is incomplete, investigators are, however,
always influenced, and often rightly influenced, by whatever
collateral evidence respecting the habitat or modes of occur-

1 Fischer and Smith, loc. cit.
vot. 4], PART 1,—NEW SERIES, N

194 J. E. 8. MOORE.

rence of such a form may be to hand. It was thus well
known when the strange Typhobia shells were first described,
that they came from a great equatorial fresh-water lake, and it
appears to me that the early investigators only did the best
they could with the purely conchological material they had
before them, in concluding that although the shell of Ty-
phobia had few characters in common with those of the
Melaniide, they probably belonged to this group all the same.
The Typhobias, however, happen to be one of those rare
organisms in dealing with which, unless there is ample mor-
phological material to draw upon, common sense anticipations
such as the above are almost certain to be wrong. ‘There
was no reason when the Typhobias were originally described
to suppose that Tanganyika, the great fresh-water lake in the
centre of the African continent, had ever been connected with
the sea. It was not known then that jelly-fish inhabited the
lake, or that the Typhobias were only one in a long series of
Gastropods which are not known to be living anywhere else
in the world. Progressive zoological exploration has com-
pletely changed our views. The study of the distribution of
the molluscs in the great African lakes points strongly, as I
have shown, in the direction of the marine origin of the Halo-
limnic group of animals. We might, therefore, now with
reason tend to be prejudiced in the opposite direction, i.e. in
favour of the marine affinities of all the Halolimnic forms.
Such a conception will, however, as I pointed out in my paper
on the distribution of these forms,! require the very strongest
morphological support, since it comes into the most uncom-
promising conflict with all those geological speculations re-
specting the character of the interior of Africa which were
started by Murchison, and which affirm that the African in-
terior has never been beneath the sea, at least since the period
of the New Red Sandstone.” It is necessary, therefore, to use
the greatest caution in determining what the affinities of the

1 This Journal, p. 159.
2 See also Gregory’s re-statement of this view contained on p. 214 of his

work ‘The Great Rift Valley,’ published in 1896.

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 195

Typhobias and the other members of the Halolimnic group
may really be.

It will have been seen from the foregoing anatomical descrip-
tion that unless the family of the Melaniidz! is to be considered
as an utterly heterogeneous group, the Typhobias are structurally
near, if they are not at, the opposite end of the whole Tenio-
glossate series. The Melanias, as they stand at present, are
certainly by no means homogeneous, and as Bouvier very
justly remarks, ‘ La famille est une des plus mal etablies dans
tout le groupe des Prosobranches,”’ but they do contain a sub-
stratum, possibly a majority of naturally associated forms,
and although it will be most important, when dealing with
other Halolimnic molluscs, to set limits to this group, the
question of its heterogeneity does not obtrude upon the present
discussion, the Typhobias being sufficiently distinct to be at
once dissociated from all those Melanian forms which have up
to the present time been anatomically examined. Whether
they may have relations among those numerous so-called
Melanias, the anatomy of which is utterly unknown, need not
be discussed.

The unique and characteristic nervous system of Typhobia
at once dissociates this form from all the ordinary fresh-water
types. The great subintestinal ganglionic cord presents no
analogy even to the zygoneurous types of nervous system, such
as those of Potamides, Cerithidea obtusa, and Pyrgus
sulcatus, which have been rightly regarded by Bouvier and
others as representing the transitional links between the Palu-
dina, Bythinia (?), and true Melanian types of nervous system
on the one hand, and Haller’s generally marine “ longicommis-
surate””? families on the other. In Typhobia Horei the
pleuro-subintestinal cord is in a most extraordinary condition,
at ouce primitive, specialised, and unique. It is specialised in
having lost the left pallial anastomosis, being thus neither
zygoneurous nor dyaloneurous on the left side, a condition of
things which finds its only parallel in the rather doubtful

‘ See Bouvier’s description of nerves of Melania, ‘Ann. des Sci. Nat.,’
ser. 7, 1887, pp. 125—181.

196 J. BH. S.- MOORE:

descriptions given by Bouvier! of the nervous systems of the
Scalarids and Solarium. It is unique in the enormous deve-
lopment of the pleuro-subintestinal cord, the whole of this
side of the nervous system being so disproportionate to the
other as to distinctly foreshadow the secondarily acquired
orthoneury of the Helicinide. The almost complete fusion of
the cerebral ganglia in Typhobia, and the reduced and
shortened-up pleuro-cerebral connectives, are conditions un-
doubtedly analogous to those obtaining in the Strombi, the
Pteroceras, the Cancellarida, and other forms. The displace-
ment of the pleural ganglia and the almost complete disappear-
ance of the cerebro-pleural commissure on both sides, appear
to be characters peculiar to Ty phobia alone, while the position
of the otocysts in the head, and not in the foot, is most primi-
tive, but may have been accentuated by the forward displace-
ment of the pleural ganglia, and the consequent necessity for
the otocyst nerves to pass over these ganglia before they reach
the otocysts. On the other hand, the otocyst nerves are very
short, and even if the pleural ganglia were in their normal
position, the otocysts would still be very high up in the head.
The complete fusion between the pedal ganglia, and the pre-
sence of ganglionic cells in what remains of the pedal commis-
sure, lead to the same inferences as do the characters of the
cerebral ganglia. The great forward prolongation of the pedal
ganglia, and the ladder-like connections between the proximate
portions of the great anterior pedal nerves, are far more primi-
tive characters. These do, in fact, suggest that the approxima-
tion in the posterior portion of the nervous system to that
condition, of secondarily acquired orthoneury witnessed in the
Helicinide, may not be altogether illusory after all. The
characters of the nervous system of Typhobia show thus in
a manner which does not appear to be capable of serious dis-
putation, that this Gastropod has no relation to, nor indeed any
but the most remote phylogenetic connection with, the hitherto
recognised fresh-water forms. Nor has the nervous system any
of those characters which could be regarded as possibly pos-

1 Loe. cit.

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 197

sessed by the forerunners of the Meianias, the Paludinas, the
Bythinias, or indeed any of the recognised fresh-water types.
Therefore, so far as the nerves go, the anatomy of the Typho-
bias gives a flat contradiction to the view that these Gastropods
may be the survivors of any extinct fresh-water stock.! The
nervous system of Typhobia exhibits, on the other hand, the
characters of some ancient but more especially of several
modern marine genera. Therefore the evidence which can be
gathered from the anatomy of the nerves is exactly in accord
with the deductions which were drawn from the study of the
distribution of these Gastropods, the Typhobias appearing to
be among the survivors of some old, but not geologically
ancient, marine types.?, What is true of the nervous system is,
however, true of the remainder of the soft parts. Beginning
with the digestive organs, it will be seen on reference to
fig. 43 that the Typhobia radula, although very singular and
self-contained, is still comparable to that of several marine
Teenioglossa. Thus in the massive characters of the admedian
teeth, the long slender character of the laterals, as well as the
form of the median tooth, this radula approaches to those
of Chenopus, Zenophora, Trochiformis, Pteroceras,
Strombus, and Pustularia, while it has many characters in
common with the radule of Crepidula, Trochita, Hyponix,
Turritella, and Cassis, and it resembles in a less degree
those of Vermetus, Triton, Ranella, and Natica.

The characters of the salivary glands, the relation of the
stomach to the cesophagus, of the intestine to the stomach, the
position of the apertures in the stomach as well as the cha-
racter of the pronounced median fold in the posterior stomachic
chamber, are all characters which are strictly analogous to

those obtaining in the Strombi and Pteroceras. The crystal-

1 The view that some of the Halolimnic forms are the remains of an old
fresh-water stock, as advocated by White, Tausch, and others, will be found
discussed in my paper on distribution, loc. cit.

2 I beg that I may not be misunderstood in this: it is one thing to say that
the Typhobias are old, since the lake in which they now live must have been
cut off from the sea for a great many years; it is quite another thing to say
that the Typhobias were contemporary with geologically ancient forms.

198 J. E. 8S. MOORE.

line style which I found in Typhobia and in certain other
Tanganyika Gastropods, and more especially the pyloric
cecum, in which this style is contained, requires more atten-
tion than it has hitherto received. The existence of these
structures in connection with the alimentary canal has long
been known in the Lamellibranchiata, and it was formerly
supposed to be confined to them. It is further well known
that in the Lamellibranchiata the style has by no means the
same relations in them all. In Anodonta and Mautela it is
free in the intestine and not contained in a pyloric cecum as
in Lutraria and many other forms, nor is the cecum present
in the former types. When, however, the style and the caecum
are both present, the latter structure apparently has invariably
the relations represented in fig. 53, the cecum being a long
stomachic appendiculum. This cecum does not necessarily
contain a style, and thus of the Lamellibranchiata it may be
said that in some the style has indefinite relations, and there
is no pyloric cecum present; while in others the cecum is
present, but does not necessarily contain a style. It was long
ago pointed out—first, I believe, by Collier! in 1829—that in
several Gastropods, the Strombide, also in species of Trochus
and of Murex “ there is an organ, the crystalline styletto,
confined erroneously by a celebrated naturalist (Cuvier) to
the bivalves. It is enclosed in a sheath that passes parallel
to and by the side of the csophagus to the stomach, into
which the styletto enters, leaving its coverings.”

This interesting observation was subsequently confirmed
and extended by Huxley? to Pterocera, and the organ was
again more completely described by M. F. Woodward, the
relations of the style and cecum to the stomach as this author
describes them being shown in fig.52. Haller? has confirmed
Collier’s observation especting the existence of this structure
in Strombus, and has extended them to Rostellaria,
but he did not recognise the significance of the structure
in relation to that of the Lamellibranchs; nor does he
appear to have been aware of Collier’s and Huxley’s obser-
1 Loc. cit. See p. 190. ? Loc. cit. Seep. 190. * Loc. cit. See p. 190.

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 199

vations on this point. Lastly, I have found the style and its
sac to be present in Typhobia, where it has exactly the same
relations to a stomachic cecum as in Strombus, Pterocera,
or in those Lamellibranchiata in which this structure is
present. I may also remark that the crystalline style and its
czcum are present in the so-called Lithoglyphus of Tan-
ganyika, the affinities of which Gastropod have been entirely
misinterpreted.

From the complete similarity of the style, and more especi-
ally of the stomachic ceecum, in those Lamellibranchiata which
possess it, and in those Gastropods where it is also present,
there can be little doubt, as Collier, Huxley, and Woodward
have already clearly seen, that the structures are in reality
strictly homologous throughout. But the great importance
and suggestive character of this conclusion has been much
obscured by Fischer! and others who confuse the true crystalline
style in its sac with the doubtful structure known as the
*« Fléche tricuspide.” There is little doubt that the ‘“ Fléche”
has in the majority, if not in all cases been merely the cuti-
cular lining of the stomach which has become detached, as it
most readily does. With the appearance thus produced are to
be classed the bodies described by Fischer in the stomachs of
Cyclostoma and Paludina. Young? also describes in
Helix pomatia a cuticular lining to the intestine, which he
erroneously compares with the crystalline style of the Lamelli-
branchs. There appears to be no similarity between the caecum
described by Cuvier and Keferstein® in Buecinum and that
in Strombus and the Typhobias. Further investigation
is, however, undoubtedly required.*

1 § Manuel d. Conchyliologie,’ p. 41.

* «Mém. Cour. Acad. Belg.,’ 4to, t. xlix, No. 1, p. 34.

3 Bronn’s ‘ Klassen u. Ordnung. d. Thier-Reichs,’ Bd. iii, Abth. 2, Mala-
cozoa, 1862-66.

4 Apart from their bearing on the affinities of the Typhobias, the above
observations show that the generally taught hypothesis which originated with
Meckel and Garner, and depicts the cecum in the Lamellibranchiata as

homologous with the radular sac of the Gastropods, and the style of the former
with some part of the ddontophore of the latter, must be utterly unsound.

200 J. E. S. MOORE.

It will be seen from all this that the Typhobias and other
Tanganyika Gastropods possess crystalline styles and ceca
which have identically the same relations, and are structures
which are undoubtedly homologous with the similar forma-
tions present in numerous Lamellibranchiata, and in a few
other Gastropods as well. Now the practical importance of
these facts to the present inquiry is this, that the particular
Gastropods in which as yet the cecum has been indubitably
recorded, are Strombus, Pterocera, Rostellaria, Murex
vertagus, Trochus turritus, the two species of Typhobia
at present known, and the so-called Tanganyika Litho-
glyphus. It may also possibly be present in Bythinia,
Once more, then, the Typhobias in the characters of their
stomachs and their related ceca are structurally near to those
marine families with which by the character of their nerves
and radule they were seen to be akin. In the possession of
a style and its sac they further exhibit anatomical features
possessed by the Lamellibranchs on the one hand, and by the
connecting link between the Lamellibranchs and the Proso-
branchs, the diotocardiate Trochi, on the other.

The gills in Typhobia are very similar to those in
Strombus and Pterocera, and the osphradium resembles
completely the same structure in all those Strombi which I
have examined. The heart, as will be seen from reference to
page 190, possesses the characters which are exhibited by
nearly all the Tzenioglossa.

The siphon possessed by the Typhobias is a structure of
doubtful value from a classificatory point of view, and even in
its narrower application it is by no means to be trusted, as both
Bouvier and Haller have already shown. An interesting
example of the impossibility of separating the holostomous
from the siphonostomous Prosobranchs has come _ before
me during the present investigation, for while examining
one of the Melanias which the authorities of the British
Museum generously placed at my disposal for compa-
rison, I found in one, the exact species of which was doubtful,
and which had been collected by Mr. Cuming in the Philippine

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 201

Islands, the small but quite apparent siphonal extension of
the mantle represented in fig. 4. This Melania had in every
other respect the true characters of the group, but from the
existence of the siphon it would, according to the old arrange-
ment, have to be removed from the Melaniidz and associated
with those families of the Tzenioglossa to which it most cer-
tainly does not belong. ‘The distinct but small anterior pro-
longation of the mantle in Typhobia (fig. 8, Z), does not
therefore appear to be of primary morphological importance,
but its existence is undoubtedly another indication of the
general similarity of the Typhobias to the forms which I have
named.

The reproductive apparatus in the Typhobias has been pro-
bably much modified through changed conditions, and the
peculiar position of the penis is possibly more the result of
extreme specialisation than the retention of any primitive con-
dition.

From all this it will be seen that the Typhobias can hardly
be said to be archaic forms, but they do, as in the character of
the nerves and the otocysts, possess some undoubtedly archaic
characters. They are far less specialised in the characters of
the foot and mantle than Strombus and Pterocera, to
which in other respects they appear to be clusely allied.
They certainly possess none of those characters which would
suggest that they can by any possibility be regarded as the
persistent representatives of an old fresh-water stock. They
do, however, simulate and retain the characters of the nerves
of the Solarium and the Scalarids, and they probably indi-
cate the road by which the more modern marine genera of the
Strombide and their associates have been evolved. But to my
mind the most remarkable features which they present are
those which I have pointed out as indicating an approxima-
tion to several forms which have been generally regarded as
recent productions; that is, they distinctly bridge the gap be-
tween several twigs which are well up in the phylogenetic
tree.

Lastly it will have been seen that in many ways the

vot. 41, parr 1.—NEW SERIES. 0)

202 J. E. S. MOORE.

Typhobias are self-contained, and have undoubtedly under-
gone individual specialisation of their own. It will there-
fore be most expedient, most natural, and most expressive of
the actual anatomical facts, to separate these two genera of
Typhobias as a family by themselves, the members of which
have affinities with, and stand in the relation of fore-
runners of, those more modern forms which group them-
selves about the Strombidze. They have been seen also to
exhibit more or fewer of the characters of a wider range of
forms, more especially of the Aporrhaide, Xenophoride,
Cypreide, and that ill-defined group the Ptenoglossa.

For this family I propose the name Typhobiide; Typhobia
Horei and Bathanalia Howesi represent the two generic
forms at present known.

The Typhobias are intensely interesting forms; their
affinities show that they have without doubt been cut off
from an exclusively marine stock at what is, geologically
speaking, no very remote period of time.

DESCRIPTION OF PLATES 11—14,

Illustrating Mr. J. E. S. Moore’s paper on ‘“ The Molluses
of the Great African Lakes.”

PLATE 11.

Fie. 1.—Living Typhobia. 2. Tentacles. 2. Eyes. 38. Operculum.

Fie. 2.—Animal removed from shell. 7. Anterior, 2. Posterior siphon.
3. Embryos seen through the thin wall of the ovisac. 4. Stomach. 45. Ovary.
6. Liver. 7. Anus. 8. Gills.

Fic. 3.—Interior of the mantle cavity. 72. Siphon. 2. Anus. 3. Genital
aperture. 4. Ovisac. 5. Heart. 6. Kidney. 7. Gills. 8. Osphradium.
9. Muscles of mantle wall.

Fic. 4.—Mantle cavity of Melania, species? from Philippine Islands.
1. Siphon. 2. Gills. 8. Osphradium.

Fic. 5.—Nervous system of Typhobia Horei dissected from above.
1. Buccal ganglion, 2. Buccal mass. 3. Tentacular nerve. 4, Pedal

THE MOLLUSCS OF THE GREAT AFRICAN LAKES. 2038

ganglion. 5. Cerebral ganglion. 6. Left pleuro-subintestinal ganglionic
trunk. 7. Pallial nerve. 98. @sophagus. 9. Superintestinal ganglion.
10. Osphradial nerve. 1. Siphon.

Fic. 6.—Nervous system dissected from the right side. Z. Buccal mass,
2. Buccal ganglion, 3. Cerebral ganglion. 4. Pedal ganglion. 5. Pleuro-
subintestinal trunk. 6. Otocyst. 7

Fic. 7.—Section through cerebral ganglion, showing—Z. Cerebral ganglion.
2. Pleuro-pedal connective. 3. Gsophagus. 4. Anterior otocyst nerve.
5. Calcareous bodies in connective tissue. 6. Otocyst with otoliths.

Fie. 8.—Section through osphradium. J, Osphradial nerve. 2. Osphra-
dial epithelium. 3. Osphradial ganglion.

Figs. 9—11.—Sections showing relation of the cerebral and pleural ganglia
(see text).

PLATE 12.

Fies. 12—26.—Variations and polymorphs of shell of Typhobia Horei
(see text).

Fies. 27, 28.—Typhobia shells at time of birth.

Figs. 29, 30.—Shells of Bathanalia Howesi.

Fic. 31.—Animal of Bathanalia removed from shell. 1. Operculum.
Fie. 32.—A single row of teeth from the radula of Bathanalia.

Fie. 33.—Base of shell of Bathanalia, showing the open columella.

PLATE 13.

Fic. 34.—Nervous system of Typhobia Horei, viewed from above.
1. Buccal nerves. 2. Tentacular nerves, 3. Opticnerves. 4,5, 6. Cerebral
ganglion. 7. Otocyst. 98. Nerve from pleural ganglion, which does not form
aright pallial anastomosis. 9. Pleural ganglion. 10. Right pallial nerve.
11. Visceral nerve. 12. Superintestinal ganglion. 13. Branch of visceral
nerve. 14, Left visceral nerve. 15. Superintestinal commissure. 16. Co-
lumella nerve. 17. Left pleural nerve going to form left pallial anastomosis,

Fig. 35.—Nervous system of Typhobia Horei, viewed from the side.
1. Buccal nerve. 2. Cerebral ganglion. 3. Pleural ganglion. 4. Super-
intestinal commissure. 5. Right pallial nerve. 6. Otocyst nerves. 7. Co-
lumella nerve. 8. Pleural nerve, going to form pallial anastomosis. 9. Left
pleural ganglion. 10. Otocyst. 11. Pleuro-pedal connective. 12. Lateral
pedal nerves. 13. Great anterior pedal nerves. 14. Ladder-like connections
between the bases of the great anterior pedal nerves. 15. Superintestinal
ganglion. 17. Osphradial nerve. 18. Right visceral nerve. 19. Left visceral
nerve. 20. Cerebro-pedal connective.

204 J. E. 8S. MOORE.

Fic, 36.—Otocyst in section, showing sensory epithelium and otoliths and
sensory processes, 1.

Fie, 37.—Section through, showing ganglionic character of the right
visceral cord at the point marked x’ in Fig. 35.

Fie. 38.—Section through cerebral ganglion in the region of the otocyst
nerves.

Fic. 39.—Sensory epithelium of the otocyst in surface view.

Fic, 40.—Section through anterior pedal nerves and ganglion, showing the
ladder-like connection between the roots of the anterior pedal nerves.

Fie. 41.—Section through snout, showing the buccal ganglia.

PLATE 14.

Fic. 42.—Semi-diagrammatic representation of the alimentary canal of
Typhobia Horei. 72. Gsophagus. 2. Opening of the oviduct. 38. Ova
in ovisac. 4. Rectum. 5. Stomach. 6. Opening of the oesophagus into the
stomach. 7. Pyloric aperture. 8. Crystalline style.

Fie. 43.—A single row of teeth from the radula of Typhobia Horei.

Fie. 44.—Dissection of the stomach. 1. Bristle passed through the
opening of the esophagus into the stomach. 2 and 3. Ditto, passed through
the opening of the bile-ducts into the stomach. 4. Bristle passed through
the pyloric aperture into the stomach. 5. Crystalline style. 6. Intestine.
7. Median foldin stomach. 8. Smaller fold. 9. Constricted annulus dividing
the stomach proper from the cecum containing the crystalline style.

Fic, 45.—Rectum and genital aperture in the male. 7. Buccal mass.
2, Anus. 8. Genital aperture. 4. Penis. 5. Glandular folds in the cavity
of the rectum.

Fie. 46.—Same. 1. Penis opened to show the muscular core.

Fie, 47.—Illustrating the course of the oviduct in a female.

Fie, 48.—Heart dissected. 1. Cavity of ventricle. 2. Auricle. 3. Auri-
cular ventricular valve. 4. Valve between the ventricle and the aortic trunk.
5 and 6. Openings into the anterior and posterior aorta. 7 and 8. Anterior
and posterior aortee.

Fic. 49.—Dissection of male, showing :—Z. Enlargement of rectum.
2. Gills. 38. Penis. 4. Anus. 4. Genital aperture. 6. Enlargement of
lower extremity of vas deferens, with parallel folds. 7. Upper portion of the
vas deferens.

Fies. 50—53.—Semi-diagrammatic representation of the stomachs and
erystalline styles in Lithoglyphus, Typhobia, Pterocera, and Lutraria.

I'te, 54.—Male genital apparatus dissected out. 1. Aperture. 2. Anus.
3. Penis. 4. Vas deferens. 5, Collecting tubes. 6. Testes.

SEP 23 Se

SEGMENTATION OF THE OVUM OF THE SHEEP. 205

The Segmentation of the Ovum of the Sheep,
with Observations on the Hypothesis of a
Hypoblastic Origin for the Trophoblast.

By

Richard Assheton, M.A.
With Plates 15—18.

INTRODUCTION.

A searcH for the earliest stages of the development of the
sheep has been attempted several times, but, so far as I know,
those efforts have not been rewarded with much success.

For instance, von Baer writes of some previous attempts :
‘Haller verband sich zu diesem Zwecke mit seinem Schiler
Kuhlemann, und beide untersuchten Schaafe sehr haufig und
von Tag zu Tage mehrere, fanden aber zu ihrer und der anato-
mischen Welt Verwunderung vor dem 12ten Tage gar nichts,
dann etwas Schleim, der sich mehrte, am 17ten die ersten
Spuren des Eies, und am 19ten schon ein sehr grosses Ei mit
dem Embryo.” But von Baer himself succeeded in finding
an ovum in the oviduct towards the end of the first day. Of
this specimen he merely says, “ Die Keimschicht war sehr
aufgelockert und verringert.”

The following description I believe also applies to the sheep,
although it is rather uncertain to what animal he is referring,
whether to the sheep alone or to Mammalia in general,
(vol. ii, p. 183) :

“ Schon im Eileiter wird, wahrend das Ei durchgent, etwas
mehr Feuchtigkeit ergossen, als gewohnlich. So kommt es in

vou. 41, parT 2.—NEW SER. P

206 RICHARD ASSHETON.

den Fruchthalter. Es ist noch immer eine blosse Dotterkugel,
doch scheint der Dotter schon etwas Feuchtigkeit aufgesogen
zu haben, da er weniger gefarbt ist. Die Haut, die den
Dotter umgiebt, ist zwar ziemlich dick, doch, wie der Erfolg
lehrt, nur Oberhaut zu nennen. Es liegt wenigstens bei
Hunden und Schaafen noch ein ganz unregelmassiger Stoff
darauf, den man fiir einen Rest der Keimschicht ansehen muss.
Unter der Dotterhaut is wahrscheinlich ein Keim, denn die
Dottersubstanz klebt nicht an der Oberhaut an, und das
Mikroscop lasst auch erkennen, dass an der Oberflache die
Dotterkornchen continuirlich zusammenhangen. Das ist der
Charachter eines Keimes.”

I do not think that there is any other description of these
early stages than this of von Baer.

Bonnet made an attempt in 1884 to secure the stages of seg-
mentation, but failed to find anything before the twelfth day.

He writes, “Ich habe neun Schafe und viel Zeit verge-
blich geopfort und wiinsche eventuellen Nachuntersuchern
meine Geduld und—mehr Gliick.”

“Mehr Gliick” I may at any rate claim to have had.
There are, of course, some gaps in my material, but I have
succeeded in obtaining a very fairly perfect series of specimens
from the time of fertilisation up to shortly before the stage
with which Bonnet’s account begins.

The ova pass very rapidly down the oviduct and enter the
uterus at an early stage of segmentation without having acquired
any mucous or albuminous coat.

While in the Fallopian tube they may be obtained very
easily by scraping off the whole of the mucous membrane
with a scalpel, and spreading it out on a glass slide, where the
ova can be found under the microscope.

It is as well to search the Fallopian tube during the first
four days, but it is not likely that any ova will be found
after the first three days. If the actual point of rupture of
the Graafian follicle is still present in the corpus luteum as a
minute bright red spot, the ovum will be in the Fallopian
tube.

SEGMENTATION OF THE OVUM OF THE SHEEP. 207

When the ova have reached the uterus, which is usually early
on the third day, the surest way of finding them is the adop-
tion of the method I made use of in the acquisition of the
early embryos of the pig. A ligature was placed round the
lower ends of the Fallopian tubes, a cannula was inserted and
tightly tied into the mouth of the uterus, both horns of which
were then slowly filled with -25 per cent. or ‘5 per cent.
chromic acid.

The uterus, when distended to its uttermost and its lower
end ligatured, was left in chromic acid of the same strength
for one to three days. | . .

The contents were then let out and the uterus thoroughly
washed, and the contents and washings were searched through
under the microscope. ;

This is an excessively tedious process, because during the
fourth to seventh days the uterus contains more or less of a
milky secretion, which renders the contents turbid, and greatly
increases the difficulty of finding so small an object as the
mammalian ovum.

In this way about three out of five may be obtained.

I received the uteri about three quarters of an hour to one
and a half hours after the death of the sheep, and filled them
at once with chromic acid.

The specimens thus obtained were stained in carmalum,
hemalum, Kleinenberg’s hematoxylin, or borax carmine, and
passed through the usual grades of alcohol, and through cedar
oil into paraffin and cut into series of sections, which varied
from ‘005 mm. to ‘01 mm. in thickness.

The dates which I give are only approximate. I am not
sure that there is so much variation in the time that may
elapse between the moment of sexual union and fertilisation as
in the case of the pig. It is, however, very difficult to obtain
any one given stage. Circumstances did not permit of a con-
tinuous watch being kept upon the flock of sheep. The dates
which I giveare not from my own observations, but are derived
from the information given me by the shepherd.

For the purpose of this paper I killed forty-one sheep, and

208 RICHARD ASSHETON.

obtained from them forty specimens. Of these thirty-one
were perfect embryos, and nine were apparently unfertilised
ova.

From fourteen of these sheep I obtained two specimens in each
case, from twelve only one apiece. In the remaining fourteen
I was unable to find any embryo, although I think the
majority probably were pregnant. The approximate ages of
these were as follows:

Embryos. Unfertilised ova.

2 days 3 0
22 » 3 0
Ss ep 6 1
4 POD 2 0
aes y) 4
be os 2 0
6 ” 5 2
7 ne 3 0
St 2 0
OnE tet. 1 0
10>; 1 0
Ts 1 2

Total . 31 9

SEGMENTATION OF THE Ovum.

In one respect I have been very unlucky. I tried with six
sheep to obtain the earliest stages of segmentation. I
have, however, not succeeded in getting any stage between the
fertilised but unsegmented ovum, and a stage with six segments.
The earliest specimens I have are two taken from the oviducts
of sheep killed during the second day. Each ovary showed a
small bright red corpus luteum. In each side an ovum was
found about halfway down its oviduct. In each case the ovum
was slightly retracted from the zona radiata, leaving a space at
one side. The ovum of the sheep is large, measuring as much
as ‘15 mm. in diameter, or including the zona radiata ‘18 mm.
while fresh.

These specimens were preserved in ‘5 per cent. chromic

SEGMENTATION OF THE OVUM OF THE SHEEP. 209

acid stained with carmalum, cut and mounted in series of
sections. Each specimen proved to be on the point of com-
pletion of the process of fertilisation.

Pl. 15, fig. 2, is a section of one of these two. The section
passes through both male and female pronucleus.

I have not been able to identify a polar body. The contents
of the ovum are for the most part of a very uniform texture.
At three points, however, the granulations are coarser and
more concentrated and are more deeply stained. The section
passes through two of these. The third, which is the least
conspicuous, is at the periphery opposite, and the next section
passes through it.

The accumulation of deeply staining particles envelops the
inner face of one pronucleus only. The opposing face of the
other pronucleus seems to be surrounded by a less concen-
trated area. Close to this latter nucleus is a small, round,
deeply staining body, which does not occur in the section from
which fig. 1 has been drawn. The other specimen mentioned
above is in exactly the same condition, and presents the same
features, with the exception that the small deeply staining
body is not present. This specimen is cut in a plane at right
angles to the previously described section.

In this specimen and in others there are certain spherical
bodies near to the periphery of the cell, which stain rather
deeply and homogeneously. They appear to be of the nature
of yolk granules, and are shown in my drawing of the next
specimen which I shall describe.

This one (fig. 3) was presumably about two days old, and
was found halfway down the oviduct. It was cut into sections,
and was found to be in the act of dividing. The plane of the
section is, I think, at right angles to the direction of the
spindle. The chromatin threads are arranged near the centre
of the ovum. The spherical bodies mentioned above, two of
which can be seen at Y in my figure, are not scattered so evenly
over the surface of the ovum as in the former stage. They are
all close to the surface, andare aggregated round the two poles.
There are rather more in one hemisphere than in the other—

210 RICHARD ASSHETON.

sixteen to twelve. A curved band of them connects the two
polar groups.

The six-segment specimen which I possess was found in the
upper part of the Fallopian tube of a sheep killed upon the
third day. It was examined in the liquid from an ovarian
follicle and drawn while fresh (fig. 25). It was composed of
two large segments and four small segments.

I could not make out that there was any difference in the
nature of the segments, either while fresh or after treatment
with Flemming’s fluid and carmalum. The four small segments
were nearly of equal size. One of the large segments was
appreciably larger than the other (v. fig. 25).

The relative sizes of the six segments may be gathered from
the following figures, the result of measurement while it was
still fresh in the follicle liquid :

15 x 16

15 x 16 1 21 x 21
Four small segments { 4, ) 15 Two large segments Be oa:

16 x 16

Fig. 4 is a section passing through two of the small and the
two large segments. The two latter appear to be upon the
point of division.

I cannot clearly recognise any yolk-like spheres such as
those described in fig. 8. Unfortunately the specimen is more
deeply stained than the other. In the two large segments
there are some larger looking bodies of more irregular shape,
which may be something of a similar nature.

From the same oviduct I obtained another embryo, which
was in eight segments. This was placed at once into chromic
acid, where it remained for four hours. It was then stained
with carmalum, cut, and mounted.

As regards size the segments do not differ greatly ; there is,
however, one segment which differs most markedly in texture
and in colour from all the remaining seven. The whole
appearance is of very much lighter colour. The minute
structure is finer, and the protoplasm has a tendency to shrink
from its walls, which in other segments it has not.

SEGMENTATION OF THE OVUM OF THE SHEEP, 211

Fig. 7 represents a section passing through the centre of
this segment. This separation of one segment differing from
the other seven is very remarkable, because there is no trace
of such a difference in the former specimen (fig. 4).

Unfortunately I had preserved the six-segment specimen in
Flemming’s solution, and so a comparison between the two
is not so perfect as might be.

I have another specimen in the eight-segment stage taken
from another sheep. It was found in the upper end of one
horn of the uterus of a sheep killed on the fourth day. It was
in chromic acid for forty-eight hours.

This specimen was stained with borax carmine and mounted in
two groups of four cells in Canada balsam, as shown in fig. 24.

The variation in size of the several segments may be recog-
nised in the drawing. I could see no difference at all as
regards colour. In one segment, which is slightly larger than
any of the others, there is a group of spherical bodies which
are stained a little more deeply than the protoplasm in which
they are embedded.

It is a noteworthy fact that in two specimens of the eight-
celled stage of the embryo one of the segments differs from all
the others. On the other hand, I have two specimens in the
eight-cell stage taken from the same horn of a uterus, neither
of which shows any sign of difference between the several seg-
ments. Both, however, were stained differently from the before-
mentioned specimens.

Fig. 5 represents the middle section of a series cut through
one of them. The cells are arranged so as to leave a distinct
cavity, which may be called a segmentation cavity. This cavity,
which would seem to persist for some time, always contains
spherical and otherwise shaped masses of a substance in every
way similar to the cytoplasm of the segments.

I have specimens of 15, 15, 16, and 17 segments respec-
tively, all of which show features similar to the eight-celled
specimen shown in fig. 5. They may be described as “ blas-
tule,” with the segmentation cavity filled with fragments of
cells.

B19 RICHARD ASSHETON.

In three of the specimens there may be seen a deeply
stained spot, resembling the chromatin of the nuclei, either in
one of the spherical masses or amidst the general detritus.

Fig. 6 represents the above condition in its most obvious
phase. This is a section of a specimen with sixteen segments.
There was nothing in its outward appearance to suggest that
it was in any way abnormal. The segments are approxi-
mately of the same size and of similar colour. The chromatin-
like granules in the interior resembled a nucleus more closely
than in the other specimens.

In the other three specimens the condition of the zona led
me to think that the preserving fluid had not perhaps reached
them, or had not acted efficiently for some other reason,
although the general form of the embryo itself seen as a whole
presented no feature of abnormality. Fig. 23 shows the
appearance of one of those which were made up of fifteen
segments.

I cannot give any explanation of the fragments of cells seen
within the segmentation cavity, nor of the origin of the nucleus-
like body. Nor can I offer any suggestion for the absence of
the difference in colour between the several segments which
is so marked in fig. 7, and again in a later stage containing
thirty segments (fig. 8), to which I shall refer presently.

I had one other specimen between the specimens just de-
scribed and this one which had about twenty-five segments.
In this there was no central cavity; but I do not think that
there was any marked difference either in the size or the
colour of the segments. It was unfortunately very feebly
stained, and in trying to remedy the defect I lost it.

Instead of the uncertainty which surrounds the interpreta-
tion of the last four specimens, namely, those between the
eight-segment and twenty-five-segment stages, I have now to
deal with a short period in which the embryos are quite dia-
grammatic in their clearness.

In a thirty-cell specimen obtained from the uterus of a sheep
of four days there can be no question as regards the differ-
ence between certain segments. Six are larger than the rest,

SEGMENTATION OF THE OVUM OF THE SHEEP. 213

and stain more lightly, and occupy a more central position.
The darker smaller cells, which partially surround the inner
core of lighter cells, number twenty-four. The segments are
not now so spherical as they were; they do not leave spaces
between each other, but are indeed pressed closely up against
each other (v. fig. 8). The difference in texture is exactly the
same as it was in the case of the eight-segment stage described
above (fig. 7). ‘he light inner cells are left exposed over
about one fifth of the surface.

The next stage is, I believe, represented by an embryo
obtained from the uterus of a sheep of six days. A median
section is represented in fig. 9. Here, again, there can be no
doubt about the different nature of certain of the segments.
A layer of darkly stained cells, which are on the whole
rather smaller, completely surrounds a group of lightly stained
cells. The light cells no longer appear on the surface at any
point. The small dark-staining cells number thirty-three,
and the inner light-coloured cells are five in number. Two
of these, however, are so nearly divided that one may say they
number seven. The outer layer of dark cells is decidedly
thicker at one point than elsewhere. In the preceding figure
(fig. 8) the middle cells of the outer layer are just a little
larger than those at the edge. If these two thickened parts
correspond, it may be concluded that the part of the inner
core which remained longest uncovered is that part imme-
diately opposite the thickened part of the outer layer.

My fig. 10 represents the middle section of a series through
a specimen taken from the same sheep as that represented by
fig. 9. It was taken from the other horn of the uterus. This
specimen is made up of forty-four small dark cells and uine or
ten rather larger inner light cells. The outer dark layer of
cells has become doubled at one part, which I think we may
presume to be the part which had previously shown a thicken-
ing (fig. 9). It was stained with hemalum, which for the
purpose of differentiating between the two kinds of cells is less
favorable than some other stains. It would seem to be in about
the same stage as the preceding specimen (fig. 9), or perhaps

214. RICHARD ASSHETON.

it shows a slight advance. The light-coloured cells are smaller
and rather more numerous.

In my next embryo, which is from a sheep of six days, the
inner mass of light-coloured cells is quite unmistakable. Its cells
number six. Of the dark outer layer there are now forty-six.

As to the last three specimens, there can be no difficulty in
determining to which group any cell belongs, but from this
time forwards the work of tracing the lightly coloured cells is
far more intricate and the result less certain. I have tried to
represent in my figures as accurately as I can typical sections
of these stages.

The all-important question to decide is, whereabouts in a
specimen such as those represented in figs. 10 and 11 does
the cavity of the blastodermic vesicle arise? I have two
specimens in which this cavity is just beginning to be formed.
One was from a sheep of about six days; of the age of the
other I have no record. My figs. 12 and 13 are both drawn
from sections of the former embryo.

There is without doubt a marked difference in colour
between the several segments; but there is no longer a sharply
defined boundary between the dark cells and the light cells, as
there is in figs. 8,9,and 10. And yet, on the whole, the light-
coloured segments are aggregated and pressed up against one
part of the wall of the vesicle (#)—a thin part of the wall—
much as they are in fig. 11, at the spot marked 2 Amongst
the light cells are undoubtedly scattered darker ones. As an
explanation it may be said that, where the knife happened to
pass parallel to the surface of a cell boundary which becomes
included in the section resulting therefrom, darker areas will
appear among the lighter.

In this specimen many strands may be seen passing between
the separating cells, which suggests that the cavity was only
just beginning to be formed at the moment of preservation.
It indicates also that there is protoplasmic connection between
at any rate certain of the segments at this time.

Another specimen of unknown age, but seemingly a little in
advance of the one just described, shows similar features.

SEGMENTATION OF THE OVUM OF THE SHEEP. 215

In this specimen the strands of protoplasm seen in the former
are less numerous. The cavity is rather larger (fig. 14). The
light cells are not so numerous nor so well differentiated as in
the former case; they form an irregular mass, partially sur-
rounded by the darker cells. I cannot be sure that there are
more than eight or nine of the lighter cells. The dark cells
number about seventy-five. It is, however, very difficult to
count them.

Fig. 15 is a drawing from a section of a rather more ad-
vanced specimen. In this the cavity of blastocyst is well
developed, and the connecting strands of protoplasm are only
found in the narrow space round the embryonic pole.

There is now very little difference in colour between the
several segments. I think my figure represents very fairly
accurately such difference as there is. This is a very slightly
lighter colour of the segments lying between the outer wall
and the innermost layer of the inner mass.

The embryo now fills the whole of the space inside the zona
radiata, the thickness of which has not materially altered since
it left the ovary. There has been no additional investment.

From this moment the internal pressure increases rapidly,
and the vesicle expands, and causes the zona radiata to become
very much attenuated. Such is the condition of my next
embryo, which is shown in fig. 16. This embryo was probably
eight days old; the age was not noted. The embryo formed a
spherical, transparent, perfectly typical mammalian _blasto-
dermic vesicle. It differs from the foregoing specimen in the
great attenuation of the zona radiata and outer layer of cells,
and in a general reduction in the size of the inner mass.

It is now impossible to distinguish any lighter coloured
area extending over more than a very limited space. There
are no visible boundaries to the cells. I am not sure that the
outer layer is distinct from the inner mass.

Fig. 17 has been drawn from a section of a specimen taken
from the same uterus, but not the same side as the preceding
specimen. It is undoubtedly slightly more advanced. The
zona radiata had ruptured or had been absorbed, and the

216 RICHARD ASSHETON.

embryo was lying naked and rather crumpled in the cavity of
the uterus. In the embryonic mass an oval body is sharply
marked out, which probably gives rise to the epiblast of the
later stages. I have, unfortunately, not been able! to obtain
the stages between my fig. 21 and Bonnet’s earliest stage, in
which there was no trace of the trophoblast layer over the
embryonic epiblast. I cannot see that there can be any reason-
able doubt as to the fate of the mass E. The more loosely
arranged masses lying at the sides and lining the inner surface
of the epiblastic knob can be recognised as the hypoblast.

Fig. 17 probably represents a stage only very little older
than the former embryo (fig. 16), for both are from the same
sheep ; yet the change in appearance of the inner mass is very
marked. One cannot help thinking that the very decided
change is due in some way to the loss of the zona radiata,
aud that in all probability some such delimitation as seen in
fig. 17 really already exists in fig. 16, but that, owing to the
greater tension under which the whole structure must be, the
delimitation is masked for the time being.

In my account of the development of the pig? I have laid some
stress upon the sudden splitting up of the embryonic epiblast
aud hypoblast into distinct layers, and I noted the coincidence
of this phenomenon with the loss of the zona radiata. In this
respect the sheep and pig correspond.

If it is possible, as I believe it to be, to identify the oval
knob of fig. 17 with the lighter cells of fig. 15, and again these
with the extremely plainly defined lighter cells of fig. 11, we
have here a most striking developmental history. The hypoblast

1 Since writing the above I have obtained a twin specimen of the sheep,
i.e. a blastocyst with two embryonic masses, each apparently normal, with
the exception of being only half the usual size. In one of these the tropho-
blast has disappeared over the centre of the epiblast, which at this spot shows
indications of a pitting in which recalls the condition of other mammals
(Tupaia, Sus) accompanying the rupture of the trophoblast over the epiblastic
knob. A description of this specimen is given in the ‘Journal of Anatomy
and Physiology,’ vol. xxxii, April, 1898.

2 This account is already in print, and will be published in the next number
of this Journal.

SEGMENTATION OF THE OVUM OF THE SHEEP. AWS

and trophoblast are shown to be derived from the same group
of segments, which are at an early period differentiated from
those which give rise to the epiblast. The trophoblast may,
therefore, be said to be hypoblastic, and the hypoblast may be
said to completely surround the epiblast at one period of
development.

Fig. 18 is a drawing of an embryo found in the uterus'of a
sheep of nine days. This specimen is very much larger than
any hitherto described. The vesicle was only slightly crumpled.
It was oval in shape, and the embryonic mass was plainly
visible on the longer face.

Fig. 19 is a drawing of a section taken along the line a—d.
The trophoblast is quite plainly distinct from the embryonic
epiblast, which has grown very considerably, and forms a solid
lenticular mass. The hypoblast is stretched tightly against
the inner surface of the epiblast, and even extends some short
distance beyond the limits of the epiblast. It must be remem-
bered that fig. 19 is taken rather obliquely, and so gives the
impression of a further extension of hypoblast than really
exists. The hypoblast is of a remarkably close texture, which
is very unusual at such an early stage.

The condition of the hypoblast as seen in fig. 19 may very
readily be derived from that seen in fig. 17 by the expansion
of the walls of the blastocyst round the embryonic pole. At
ten days there is no great change. I have only one specimen
of this age. It was considerably crumpled (fig. 20). The
embryonal area has a longer and shorter axis.

Fig. 21 is a section of the embryonal area. The tropho-
blast (T) is very thin, and so much stretched that not more
than six nuclei overlie the epiblast (E).

The only change to be noted in the epiblast is a greater
activity apparent, and the lower cells seem to show a tendency
to become arranged in a rather columnar or stellate fashion.
There is a layer of smaller cells on the inner surface of the
epiblast, and one of very much larger cells near the outer
surface, which may be compared to the state of things in the
corresponding stages of the pig or tupaia. The hypoblast is a

218 RICHARD ASSHETON.

continuous sheet extending nearly halfway round the inner
surface of the vesicle (v. fig. 22).

This is the oldest stage which I shall describe. My next
specimens are twelve and thirteen days old, and already much
elongated.

The specimen, fig. 20, is younger than the youngest de-
scribed by Bonnet. I regret very much that I have nothing
between this and Bonnet’s, for in Bonnet’s specimen the
Rauber layer had gone.

That the layer H is the same as Bonnet’s layer Er (fig. 3)
there can be no reasonable doubt. Nor can there be much
doubt that the mass E in my fig. 16 gives rise to most if not
all of the ‘‘ ectoblast ’ E in Bonnet’s fig. 3. But of the fate
of the Rauber cells overlying this in my fig. 16 I have no
record (vide note on p. 216).

In the complete absence of zona radiata I cannot think that
their fate can be very different from what it is in the pig or
tupaia, with whose development the sheep has many points in
common. The separation between the trophoblast cells and
epiblast is quite complete in my specimens (figs. 19, 21).

OBSERVATIONS ON THE MAMMALIAN BLAstTocyst.

The interpretation which I believe may be placed upon
the specimens described above and in my paper on the pig
may be briefly stated. At an early stage, in the sheep per-
haps as early as the eight-segment stage, the future epiblast
and hypoblast are differentiated.

The hypoblast surrounds the epiblast, so that at the morula
stage the embryo consists of a few epiblast cells surrounded
by hypoblast cells, which at one pole form a thicker investment
than elsewhere. In the middle of this thickened mass of
hypoblast the cavity of the blastodermic vesicle arises.

Subsequently, by the rupture of the hypoblast over the
epiblast, the latter comes to the surface.

Therefore the greater part of the wall of the blastodermic
vesicle is hypoblast. The hypoblast is, owing to mechanical

SEGMENTATION OF THE OVUM OF THE SHERP. 219

and physical causes, double over part of the wall in the majo-
rity of forms.

The cavity of the blastocyst is not the segmentation cavity,
but the archenteron. It is bounded on all sides by hypoblast,
though at the part where it stretches across the epiblastic
disc it is perhaps a network of cells rather than a continuous
sheet at the time when the blastocyst cavity is very rapidly
expanding, as during the stages of figs. 12—16.

There would seem to be nothing in the pig to cause a
differentiation in staining, but in the early stages there is a
distinct suggestion of a growth of smaller cells round a group
of larger ones.

The bursting through of the epiblast and rejection of the
pieces of trophoblast torn apart in the process are particularly
clear in the pig. I regret very much that I have not got the
stages which may include the corresponding process in the
sheep (vide foot-note on p. 216).

Concerning the homologies of the several parts of the
segmented ovum and the blastodermic vesicle of mammals
very diverse views have been held.

It is only because the facts which I have discovered in the
early development of the sheep, which are corroborated by
events in the development of the pig, seem to give promise of
an interpretation in some ways more satisfactory than any
hitherto advanced, that I venture to enter upon a short discus-
sion of the subject.

I am perfectly well aware that there is much to be said in
favour of the views expressed by embryologists far more com-
petent to deal with the subject than am I. As recently as
during the last two years important papers dealing with the
question of the segmentation of the mammalian ovum have
appeared from the pens of such noted embryologists as Pro-
fessors Hubrecht and Duval. Although the respective views of
these two authors were quite different, yet in each case most
plausible explanations were offered for many of the problems
of mammalian embryology.

But to my mind neither of their views can be maintained

220 RICHARD ASSHETON.

for the sheep; and apart from this, there seems to me to be
such a strong prima facie case for the hypoblastic blasto-
dermic vesicle theory that I believe it to be deserving of more
attention than it has received. Hitherto no mammal with a
typical blastodermic vesicle stage has been examined which in
any way hinted at the explanation which is suggested by the
study of the segmentation of the sheep’s ovum.

While my hypothesis is not entirely in accord with that of
any former author, it resembles those of Minot (48) and
Robinson (46), and indeed owes much to the work of the latter
upon the rat and mouse, and to the theoretical conclusions of
his suggestive essay in vol. xxxili of the ‘Quarterly Journal
of Microscopical Science.’

The chief feature of my hypothesis is that (like Minot and
Robinson) I regard the main wall of the blastodermic vesicle
as entodermic. I differ from them, however, in that I regard
as entodermic a greater portion of it than they do.

I differ from Minot in considering the inner layer of the
inner mass as entoderm as well as the whole of the subzonal
membrane (v. Minot’s ‘Human Embryology,’ pages 107 and
108). Hence the cavity is not, as he regards it, the segmenta-
tion cavity, but the archenteron, which arises as a split within
the entoderm as shown in my diagrams 2 and 3, and is not com-
parable to the segmentation cavity of amphibians.

Haddon’s (27) theory, although called by Minot a similar
explanation to his, seems, however, to me to be very different.
Haddon regards the subzonal epithelium, including “ Decken-
zellen,”’ as the equivalent of the extra-embryonic epiblast of
the bird’s egg; so that the cavity of the blastodermic vesicle,
which he compares with the yolk-sac, is supposed to be
bounded on all sides except the embryonic pole by “ extra-
embryonic” epiblast, and not by hypoblast. Keibel’s (88)
suggestion is practically the same as that of Haddon. With
Minot I should agree in so far as supposing that, to quote his
words, ‘‘ there is, then, a complete inversion of the germinal
layers in all placental mammals.” This I imagine to occur in
most forms at a very early stage (pig, sheep, rabbit, mole,

SEGMENTATION OF THE OVUM OF THE SHEEP. 221

&c.), but rather later in some (hedgehog, bats). According to
my hypothesis the epiblast never grows round the blastocyst
in Monodelphic Mammals. From Robinson I should differ in
supposing that the whole of the subzonal epithelium is hypo-
blastic instead of only the “ greater part” of it, and further in
denying that the epiblast ever extends round a_ previously
existing hypoblastic vesicle. Both layers of the extra-
embryonic part of the wall of the didermic vesicle are on
this hypothesis hypoblastic.

The accompanying diagram (p. 222) will render clear what I
believe to be a perfectly legitimate interpretation of the facts
derived from the study of the segmentation of the ovum and
formation of the embryonal area in the sheep and pig.

Such a conception of the early embryo of the mammal is very
closely in accord with the meroblastic egg of the Sauropsida.

If we regard the cavity of the blastocyst as the segmentation
cavity, it is then impossible to compare the blastocyst with the
egg of the Sauropsida. But if the cavity is regarded as the
equivalent of the subgerminal cavity the comparison is com-
plete. The origin and the fate of the subgerminal cavity of
the bird and reptile are exactly the same as the origin and fate
of the cavity of the blastocyst of mammals.

In each case it begins as a split amongst hypoblast cells,
and becomes enormously enlarged and distended by fluid, and
provides ample space for the development of the embryo free
from pressure. <A portion of it in each case becomes the gut
cavity of theembryo. The diagrams A, B, C, on Pl. 18 repre-
sent the bird’s egg during the first few hours of development
according to Duval’s description.

Fig. A represents an early stage of segmentation. The
segmentation cavity is seen as a narrow slit between the epi-
blast, which is coloured pink, and the hypoblast, which is dis-
tinguished by a green tint. The segmentation cavity does not
occur in the placental mammals, except perhaps in the case of
the rat (Robinson) and sheep; otherwise figs. A and C are
exactly comparable with the first and last of the diagrams upon
the following page.

voL. 41, pART 2.—NEW SERIES, Q

222 RICHARD ASSHETON.

5

The light cells are epiblast (g), the darker cells are hypoblast (#).

SEGMENTATION OF THE OVUM OF THE SHEEP. 223

In the bird’s egg at the time of laying, and for the first few
hours of incubation, there is said to be a discontinuity between
the hypoblast and the “yolk” (B). The yolk at this time is
nevertheless laden with nuclei (Duval, 23). After the first few
hours of incubation the yolk again becomes continuous with the
hypoblast, as in diagram C.

The passage which is caused by this temporary loss of con-
tinuity, and which extends round the blastoderm, has been re-
garded as blastopore, and has been said to become converted
into the primitive groove. Some experiments made with a
view of testing this point by myself (4) proved the assumption
that it becomes converted into the primitive groove to be un-
tenable.

It is, I think, reasonable to suppose that this loss of con-
nection between the two portions of hypoblast is in some way
related to the act of laying of the egg, and to the cessation of
development attendant on it.

Fig. B, Pl. 18, can be compared with the second diagram
(p. 222) as regards the doubling of the hypoblast cells, amongst
which, in the next stage, the great cavities—the subgerminal
cavity, and the cavity of the blastodermic vesicle—subse-
quently appear. In most mammals, however, the hypoblast
temporarily extends over the epiblast, so that the latter is com-
pletely withdrawn from the surface during a certain period of
development.

The exact method by means of which this overgrowth in the
mammalian ovum is effected, and the method by which it is
subsequently rectified, so as to bring the epiblast again into
its proper position, seem to vary with almost every species.
The series of small diagrams on Pl. 18 illustrate some of these
diverse methods, and their relations to each other.

Fig. D, Pl. 18, represents the segmenting ovum of possibly
all placental mammals.

The pink represents epiblast, the green hypoblast. The
black ring is the zona radiata, and other investing coats which
may be present.

Immediately below this is the next stage (X) of a certain

224. RICHARD ASSHETON.

type of development in which a central core of epiblast is sur-
rounded by a single layer of hypoblast cells. This group con-
tains Ovis, Sus, Tupaia, Talpa, Sorex, Lepus, and probably Cavia.
These all have a strong zona radiata.

Fig. Y shows the next stage of all these forms, in which the
hypoblast has become doubled over the ab-embryonic pole. At
this point one of the number, Cavia, loses its zona radiata, and
branches off on a line of development of its own (C,_,.)

Continuing down the page is a figure (Z) which may be
supposed to be a stage common still to the remaining members
of this group, which, however, after this point take three rather
distinct courses. On the right side of the plate are shown
Lepus and Sorex, L,_, and §,_».

These two are characterised by the looser arrangement of
the epiblast, which condition results in a gradual expansion
and formation of a flat epiblastic disc. Both forms retain the
zona radiata uutil after the growth of the epiblastic disc has
caused the rupture of the overlying hypoblast cells (Rauber
layer). This leads to the inclusion (either permanent or
temporary) of these fragments among the epiblastic disc
cells.

On the left of the page, diverging also from Z, is another
group containing Sus and Tupaia (Su,;_; Tu,_,), and probably
Ovis. These are characterised by the formation of a more
solid epiblastic knob, which does not become drawn out, but
undergoes a process of doubling up previous to assuming its
final dise-like form. The zona radiata is thrown off, so that
when the rupture of the overlying hypoblast cells occurs there
is nothing to prevent the fragments from being lost. Accord-
ingly none become incorporated in the epiblastic disc.

Intermediate between these is placed Talpa (T,_,), which is
like Tupaia and Sus as regards its epiblastic characteristics,
but like Lepus and Sorex retains the zona radiata, with the
result that the fragments of overlying hypoblast become incor-
porated with the epiblastic disc (Heape, 30).

On turning again to the first figure, D, there will be found
upon the right of the plate a series of five figures, M,_;, which

SEGMENTATION OF THE OVUM OF THE SHEEP. 225

I have taken with some modification from Robinson’s paper on
the rat.

I take the trophoblast to be hypoblastic instead of epiblastic.
The true epiblast is cut out from the surrounding trophoblast
at an early stage (v. Robinson’s [46] fig. 8). In the majority
of his figures he certainly shows the trophoblast cells to be
darker than the hypoblast and more like the true epiblast. But
this is not so in all. It is not so in fig. 6. Ido not see why
the thin side walls of the “‘ segmentation cavity ” of this figure
should not be derived from the hypoblastic mass, in which
case, since they subsequently give rise to the trophoblast
(figs. 8, 9), the trophoblast may be said to be hypoblastic.
Selenka (49) and Duval (25) have given very different accounts
of the rat’s development. That there is much to be said on
both sides is evident from Weysse’s (56) remark that, from his
own work upon the subject, he is “ still undecided.”

Sobotta (54) describes and figures a separation of the
segments into groups of small and large segments, but he does
not trace their fate.

In this case we see that there is no zona radiata. There is
a segmentation cavity, which, however, quickly becomes oblite-
rated. The very early loss of zona may account for there being
no growth round by the “ hypoblast’’ during the early stages.
If Selenka’s or Duval’s description is adopted, it would be open
to a similar interpretation to that suggested in diagram C,_, for
the guinea-pig. The same may be said for Mus sylvaticus
(Selenka, 50). In this the epiblast is seen to be cut out very
early as a definite knob (fig. 36).

On the left of the plate, also diverging from D, are shown
two developmental histories, which seem to form a rather
special group. Inthese the envelopment of the epiblast by the
“hypoblast”’” is supposed to take place rather less quickly.
Figs. V to V, represent what seems to be a possible interpre-
tation of the ontogeny of the bat. In this case the surrounding
does not occur until after the appearance of the cavity of the
blastocyst.

Fig. V, is taken from van Beneden and Julin (18), the

226 RICHARD ASSHETON.

remainder from Duval’s (26) account of the bat. As, however,
Duval’s explanation is quite different from the above, I have
given below at some length my reasons for dissenting from
his views.

Erinaceus on this hypothesis would be in some respects
similar.

Unfortunately the earliest stages of the hedgehog, with
the exception of a two-segment stage described by Keibel (87),
are not known. Fig. E is therefore quite hypothetical for
the hedgehog, and should probably be drawn without a zona,
and possibly at this stage the hypoblast should be shown
already extended over the epiblast.

If the increase of the hypoblast cells is very much greater
proportionally to the epiblast than it is in the sheep, it is quite
possible that when the cavity of the blastocyst arises its walls
may be more than one layer thick, some of the inner cells
giving rise to the inner sac described by Hubrecht (88).

In Hubrecht’s earliest figures there is certainly no distinc-
tion between the trophoblast and epiblast, but in the slightly
older ones which are to be compared with my diagram E the
distinction is clear, as a glance at Hubrecht’s figs. 39, 8, 8a
will show. Figure E may be compared with fig. 15 or 16.

I have drawn Figure E, according to Hubrecht’s figs. 20 6
and 51, although I am not clear whether he describes the forma-
tion of the true amnion as the advance of a free edge, as would
appear from fig. 51, or whether, as the wording of the text
seems to indicate, (see pp. 289—296, and again 374, where
he says, “ We have seen that after the completion of the
germinal area its epiblast remains attached to the trophoblast
[from which it has split off] along a circular line of insertion.
At the time of the formation of the amnion this line of attach-
ment travels upwards, the circle becoming smaller and smaller,
until at the completion of the amnion no connection between
embryonic epiblast and trophoblast any longer exists,” it is
really a process of splitting off from the overlying trophoblast).

If, therefore, it is really a case of splitting, the amnion cavity
shown in Figures EH, and E, should have been drawn within

SEGMENTATION OF THE OVUM OF THE SHEEP. 227

the pink mass instead of between the pink and green. In this
case the amnion formation of Erinaceus would resemble more
closely that of Pteropus or Cavia ; whereas, as I have interpreted
it, it is more like Vespertilio.

Such is briefly the hypothesis to which I venture to draw
the attention of those who are interested in the problems of
mammalian embryology. I cannot deny that there are diffi-
culties, but these do not seem to me to be formidable.

Let it be noted that we have now to face the fact, based
upon actual sections, that there is in certain mammals a clear
separation of segments at an early stage into two groups, one
of which eventually completely surrounds the other. Until
the last few years this has not been so. It is true that van
Beneden in 1880 described this process in the rabbit. His
interpretation, however, led to many difficulties, which were
only partially removed by his subsequent modification of it.
I showed in 1894 that van Beneden’s description, derived from
surface views and optical sections only, was not supported by
my actual sections, and urged that the phenomena of the
development of the rabbit’s ovum were open to a more strictly
epigenetic explanation. Since then Duval’s work on the bat
and Hubrecht’s on Tupaia, and my own observations on the
sheep, have convinced me that such a process does occur in
certain mammals.

What does this phenomenon mean? It must surely have some
most profound significance. Van Beneden and Duval maintain
that it is the division into epiblast and hypoblast, the former
being the external layer. Hubrecht considers it is a division
into epiblastic trophoblast and embryo. Sobotta (54) connected
the lighter appearance of certain cells in the early stages of
the mouse’s development with the state of division of the cells.
It is, I think, quite impossible to explain the differences in my
specimens in this manner. The present hypothesis supposes
that in cases where this differentiation does clearly occur it is
a division into epiblast and hypoblast, the latter being the
external layer; and that if this is or at some time has been
the mode of development in all placental mammals, it is then

228 RICHARD ASSHETON.

suggested that the phenomenon is capable of interpretation in
some such way as that indicated by my Plate 18.

The obvious difficulties which arise from this last hypothesis
may be stated thus:

(i) The alternative explanations given of individual de-
developmental histories by other authors.

(ii) The double-layered condition of the hypoblast.

(iii) (@) The inference that the chorion of placental mammals
is formed by the fusion of allantois and extra-embryonic hypo-
blast (trophoblast), whereas in birds and reptiles the corre-
sponding membrane is formed by the fusion of the allantois
and extra-embryonic epiblast, with, of course, in each case the
intervening mesoblast. (4) Although the true amnion may in
every case be either wholly or partly formed of true epiblast,
in all cases the false amnion would appear to be formed of
hypoblast.

These difficulties are discussed below. The chief features of
my hypothesis which seem to me to be most likely to commend
it to the attention of embryologists may be stated thus:

1. It is an explanation of the early differentiation of the
segments into two groups of cells, which avoids the difficulty
which attends that of van Beneden.

2. It is an explanation of the van Beneden blastopore.

3. It suggests a reason for the rejection of the Rauber layer
celis by the embryonic epiblast, which is so evident a pheno-
menon in some cases.

4. It demands no change in function of the two primary
layers during the substitution of the yolk granules by maternal
fluids. The hypoblast cells maintain their function as a nutri-
tive organ, entering into direct connection with the maternal
tissues, in nearly all animals, in the various forms of trager,
ectoplacenta, and villi, and thereby supporting the embryo
until the placenta is established.

5. The morphological relation of the epiblast and hypoblast
and the cavity from which the future gut is formed agree on
this explanation absolutely with the conditions of the Saurop-
sidan and Monotreme egg.

SEGMENTATION OF THE OVUM OF THE SHEEP. 229

It suggests an explanation of the formation and origin of
the Eutherian amnion, which leads to the same conclusion
as Hubrecht’s view, namely, that the more primitive form of
amnion is probably that which is found in mammals like
Erinaceus, Cavia, or, as I think more truly, in Vespertilio,
and is of a different origin to the amnion of the Sauropsida
and the Monotremata ; but it does not, on that account, demand
the removal of the placental mammals any further from those
groups than any other known anatomical differences do.

CoNSIDERATION OF THE OBJECTIONS WHICH MAY BE URGED TO
THis View oF A Hyposuastic ORIGIN FOR THE TROPHOBLAST.

Tupaia.

Hubrecht (38) in Tupaia javanica recognised an early
arrangement of some cells round others. He finds no differ-
ences during the first two generations, but when there are appa-
rently about a dozen cells or more he finds a differentiation into
an inner core of one lightly staining cell, which is completely
surrounded by an outer layer of darker cells. “ Auf dieses
vierzellige Stadium folgt sehr rasch die Gruppirung der
weiteren Furchungsderivate, nicht um eine centrale Hohlung,
sondern um eine centrale Zelle. Diese centrale Zelle fallt in
mehreren Schnitten durch ihre hellere Farbe auf, gleichsam
alsob ihr Protoplasma in etwas anderer Weise auf das inc-
tionsmittel reagirt habe, wie dasjenige der peripheren Zellen.”’
This rather later differentiation into two kinds of cells is very
like what I have described for the sheep, in which case I gave
some evidence to show that the differentiation occurred at the
third generation.

According to Hubrecht, this differentiation marks embryo
from ‘‘ trophoblast.” The central cell, which soon multiplies,
gives rise to the whole of the epiblast and hypoblast, while the
outer layer is only ‘‘ trophoblast,” and is destined to form no
part of the embryo proper. As appears in the latter part of

230 RICHARD ASSHETON.

the paper, Hubrecht considers the trophoblast to be of
epiblastic origin, and he compares it to the outer layer of
epiblast of certain amphibians. He derives the mammalian
egg from a small egg like that of amphibians, with compara-
tively little yolk.

Apart from all question of homology and mammalian de-
scent the term trophoblast seems to me most useful in
describing the events of mammalian development, and as such
1 have adopted it in my descriptions of the sheep and pig; but
I do not adopt his hypothesis, because—

(1) It offers no explanation of the mode of origin of the
trophoblast. On Hubrecht’s hypothesis we should have ex-
pected a process of delamination, whereas in the sheep and
bat, where the two groups of cells are the most clearly defined,
it is an essentially horizontal division, as in the meroblastic
eggs of birds and reptiles.

(2) It does not account for the throwing off of the tropho-
blast cells over the embryonic area, which is so evident in
certain forms; which is the more surprising since in the Anura
(to the epidermic layer of epiblast of which these cells are
supposed to owe their origin) there seems to be permanent
fusion between the epidermic and nervous layers, as described
by myself (2).

(3) It is not applicable to my specimens of the sheep.

In order to make my observations accord with Hubrecht’s
hypothesis it is necessary to suppose that the inner layer of
dark cells in my fig. 11 has been derived from the light-
coloured cells. The difference is so great that I think it is
almost impossible to ascribe such an origin to them. Or else
we must suppose that these cells subsequently pass into the
outer layer again after the stages of figs. 12 and 15. On the
other hand, it seems easier to make the description and figures
of Tupaia agree with my description of the sheep.

Hubrecht’s figures in his pl. 1 show only a small differ-
euce in colour between the inner cell and the outer wall. For
instance, in the figs. 19—25 there is very little evidence to
prove that none of the inner cells were derived from the outer

SEGMENTATION OF THE OVUM OF THE SHEEP. 231

wall. As far as it is possible to judge from these figures there
is nothing to absolutely condemn such an interpretation.

The Bat.

An attempt to reinstate van Beneden’s (8, 9) metagastrula
theory for the development of the mammalian ovum has quite
recently been made by Professor Matthias Duval in his work
upon bats in the ‘ Journal de l’Anatomie et de la Physiologie’
for the years 1895, 1896.

Duval found in the early stage of the development of Ves-
pertilio murinus a stage exactly comparable to that which
van Beneden described for the rabbit.

Quite in accordance with van Beneden’s earliest account of
the segmentation of the ovum in the rabbit, Duval believes
that in the bat the whole of the inner cell mass becomes hypo-
blast, and the outer layer alone becomes epiblast. Like van
Beneden he traces the origin of the distinction between the
hypoblast and epiblast to the first division in the process of
segmentation of the ovum.

From the first two segments a group of larger and more
slowly dividing cells arises, and forms eventually the whole of
the inner mass, which is surrounded by the smaller and more
rapidly dividing descendants of the other primary segment.

The outer layer is from the moment of its first formation a
single cell in thickness, and it remains so until a cavity
appears between the inner, mass of cells and the outer layer,
and converts the embryo into a typical blastodermic vesicle.
The vesicle expands, and the inner mass is almost entirely
flattened out, so that at one time a condition is attained when
the vesicle at its upper pole consists of an outer layer of a
single cell in thickness—the epiblast, and an inner layer also
of a single layer in thickness—the hypoblast.

The outer layer now begins to thicken, and forms the em-
bryonic knob. In this knob an irregular cavity subsequently
appears, portions of the roof of which are thrown off and
absorbed, and the floor remains as the permanent epiblast.

This description as regards the segmentation of the ovum.

232 RICHARD ASSHETON.

and origin of the epiblast is exactly comparable to that of van
Beneden for the rabbit, which, however, has been abandoned
by van Beneden himself. Duval believes that it is neverthe-
less true for the rabbit. It follows in that case that the per-
manent epiblast in that animal has a different origin from that
which at present it is held to have.

Duval suggests that its true origin has been missed by all
the workers on the rabbit hitherto, and that the whole of the
inner mass becomes flattened out as hypoblast (v. Duval [26],
fig. 28, p. 163).

The rabbit has formed an object of research for so many
persons that such an important stage is very unlikely to have
been missed. Duval suggests that it is a stage which occupies
so short a time that no one has chanced to catch it. But
think what the process must involve! Cells which are ex-
tremely attenuated and under considerable tension must sud-
denly grow so rapidly as to bud off large cells in a direction
contrary to that in which they have hitherto given rise to cells,
and then again resume their attenuated form ; and while this
is going on, the large rounded cells previously existing between
the outer layer and the hypoblast have to pass into the hypo-
blast and become suddenly stretched and flattened, like those
already there. These would surely be changes more profound
than anything that occurs during the first seven days of deve-
lopment! It is difficult to believe that they could have
escaped notice if they had existed. If there is a true “ meta-
gastrula” in the rabbit, and if the outer layer of this meta-
gastrula gives rise to the epiblast in a way similar to Duval’s
description of the process in bats, the thickening of the outer
layer to form the permanent epiblast must occur at a much
earlier stage, and possibly before the formation of the cavity
of the blastodermic vesicle,—as, for instance, the thickening
of the outer layer does in the similar stage in the sheep
(fig. 9).

Here is, at first sight, another possible interpretation of my
sheep specimens, if the differences in colour indicated in my
drawings of the later specimens (figs. 12—15) are held to

SEGMENTATION OF THE OVUM OF THE SHEEP. 230

be of insufficient evidence to prove the descent of the light
and dark cells of the later stages from those similarly coloured
in the earlier figs. 8,9, 11. This would be in accordance with
Duval’s view.

But if I attempt to take this interpretation, and say that
the dark cells of figs. 8, 9, 11, are all epiblast and the light
cells hypoblast, I find that, as may be seenin the figs. 12—
15, there is no tendency whatever for the dark cells to
become grouped together into a solid mass, such as the em-
bryonic epiblast is in the later stages (figs. 17, 19). On the con-
trary, the darker cells of the inner mass in every section show
a looser and a more peripheral arrangement to the inner mass.
I find it is not possible to trace either the dark outer layer
cells of fig. 11 into a compact inner mass, or to trace the
lighter inner mass cells of fig. 11 into a loose investing
sheath as at fig. 17. Whereas for the reverse course there
is, as I have shown in the earlier pages of this paper, not
a little evidence.

Is it possible to put an interpretation upon the ontogeny of
the bat similar to that which I believe should be placed on the
ontogeny of the sheep, such as suggested in my diagrams
(V—V, on Pl. 18) ?

I believe that a closer examination of Duval’s and van
Beneden and Julin’s works indicates that such an interpre-
tation is possible, as I will now attempt to show.

The account given by Duval is by far the most complete
description of the development of any bat hitherto published.
In some most important features it differs from that given by
van Beneden and Julin (18).

The authors are, however, agreed that there is from the
first division of the ovum a distinct difference, by which the
segment which will ultimately give rise to the epiblast cells
may be recognised from that which will give rise to the hypo-
blast.

They agree that the four-segment stage consists of two
larger spheres and two smaller spheres.

Van Beneden and Julin say that two are smaller, darker,

234 RICHARD ASSHETON.

and more granular and duller than the other two. This
description refers to two specimens of Vespertilio murinus
examined in the fresh state. The same holds good for speci-
mens of Rhinolophus ferrum equinum.

On the other hand, Duval, who describes two specimens of
the same species Vespertilio murinus and speaks from pre-
served and stained material studied in section, asserts that the
two larger spheres are the darker ones, ‘“‘ deux plus petits et
plus clairs ; deux plus gros et plus foncés.”

According to Duval the two small clear spheres give rise to
the ectoderm, Van Beneden and Julin had unfortunately no
stages between the four-segment stage and the completed
blastodermic vesicle. It would seem, however, that they con-
sidered the large clear cells to be ectodermal, and the smaller
darker cells hypoblast.

Again, there is a disagreement as to the position of the
inner mass on the wall of the blastodermic vesicle. Duval
concludes that the inner mass is attached to the base of the
cup formed by the enveloping outer layer, whereas van Beneden
and Julin find it at the mouth of the cup, which two accounts
are quite irreconcilable. In Duval’s case it is Vespertilio
murinus which is being described, and in van Beneden and
Julin’s it is Rhinolophus ferrum equinum. But the
difference is so great that I can hardly think it should be
accepted as a generic variation. It is impossible to derive
van Beneden and Julin’s figs. 5 and 6 from Duval’s fig. 24, or
vice versa.

Van Beneden and Julin’s account is derived from two living
specimens, and Duval’s from one specimen which had been
subjected to the process of preservation and staining, and had
been made into sections.

Duval unfortunately never shows the presence of zona
radiata in any figure, and I do not think that he ever makes
any mention of it. It is difficult to imagine such an embryo
existing in a uterus unless there is a zona present. Van
Beneden and Julin, however, show the zona in all their figures.

One could have felt more confident that this specimen was

SEGMENTATION OF THE OVUM OF THE SHEEP. 235

quite normal and uninjured if Duval had mentioned whether
the zona radiata was present and quite sound.

Of course Duval notices this all-important difference, but
cannot explain it. He says (p. 155), ‘‘ Quant au blastopore
primitif (ombilic ombilical) du Murin, nous devons aussi
remarquer que nos préparations nous le montrent avec des
rapports différents de ceux observés par van Beneden. Cet
auteur l’a decrit et figuré comme placé au centre de la région
ou la masse endodermique adhére a la face interne de la sphére
ectodermique. Nous |’avons vu, ou, pour mieux dire, nous avons
vu un orifice, précisément dans la région opposée, au centre de
V’autre hémisphére. Nous ne saurions pour le moment ex-
pliquer cette contradiction. Nous sommes bien convaincu de
la valeur et de l’exactitude de la fig. 24 (pl. i), Poeuf que l’a
donnée, étant conservé en coupes, qui ont pu étre étudiées a
diverses reprises. Mais nos observations ne sont pas assez
nombreuses, car, entre le stade F et la stade H nous n’avons
qu’une observation.”

It seems to me that there is a greater weight of evidence in
favour of van Beneden’s view.

Taking the combined results of Duval’s and van Beneden’s
papers, the interpretation offered in my diagram, V—V,,
Pl. 18, seems to be by no means an impossible one. W is
Duval’s fig. 15, 17, 18, or 20, reversed, E V is derived from
fig. 21, and the description of the same on p. 140. Duval
concludes that his fig. 21, H., is a section taken through such a
specimen as his fig. 20, at right angles to the direction of the
section fig. 20, close to one end, and that it cuts the ectodermal
cells only. Now according to Duval this ectodermal layer is
only one cell thick. Is the section Duval has drawn (fig.
21, H.) taken as close to the surface as he supposes? Is it not
possible that the ectoderm is really two layers thick (or perhaps
more) at this end of the embryo like my diagram EV? Fig.
21, H., undoubtedly shows an inner core of cells, and when we
consider how convex a surface it is which is supposed to be
cut, it is remarkable that in no less than fifteen out of seventeen
cells through which the section passes the nuclei should have

236 RICHARD ASSHETON.

been cut, if it is only one layer in thickness. The diagram
V, is derived from fig. 6 of van Beneden and Julin’s work, or
from fig. 32 of Duval’s work.

If there is such a distinct difference in colour between the
outer layer and inner layer of the inner mass, as shown in
Duval’s figures, my interpretation can hardly hold good.
Duval, however, explains that his figures are to a certain
extent diagrammatic. He writes on p. 184, ‘ Sur nos figures,
nous avons exagéré et schématisé ces différences d’aspect entre
les segments petits et clairs (cellules ectodermiques) et les
segments plus foncés, plus granuleux (cellules endodermiques),
et a cet effet nous avons, d’une maniére conventionelle, donné
aux cellules ectodermique un corps clair, avec un noyau foncé
et aux cellules endodermiques un corps foncé avec noyau clair,
de fagon a permettre de saisir au premier coup d’cil la maniére
dont se comportent ces cellules les unes vis-a-vis des autres
dans les stades successifs que nous figurons.”

Another extremely important stage from Duval’s point of
view is that of his fig. 36.

_ Here the outer layer consists ‘“ généralement d’une seule

couche de cellules cubiques.”? At the embryonic pole, however,
these cells tend to become arranged in two layers. It is,
however, a very slight thickening. The inner mass is very
much flattened, and extends halfway round the walls of the
vesicles.

At the embryonic pole the inner mass is, although distinct
from the outer wall, two layers thick, which thickness, however,
extends over a short area only.

Duval assumes that the whole of this inner mass is hypo-
blast, and that it eventually becomes completely flattened ont
into a single layer, as he shows in his diagram B on p. 106
(vol. 1896). But he apparently has not found this stage—it
is not shown in any of his figures.

May we not, on the other hand, suppose that in the double-
layered condition of the inner mass of this stage the inner
layer is the true hypoblast, and the cells lying between this
and the outer layer are the true epiblast cells? May not

SEGMENTATION OF THE OVUM OF THE SHEEP. 237

fig. V;, Pl. 18 be as likely an interpretation of his fig. 36 as
that given by his diagram B, p. 106?

This stage is at once succeeded by one in which a rapid
increase in the mass occurs at this point, which Duval con-
siders to be entirely due to the activity of the outer layer.
So great is the increase that it is impossible to distinguish
between it and the hypoblast ; there is “ une soudure des deux
feuillets avec engrénement des cellules de l’un dans celles de
Vautre.” Is it decisively proved that the intermediate cells
take no part in this increase? I hardly think it is, and suggest
that his fig. 41 is equally open to the interpretation given by
my fig. V;.

The subsequent history tends to support my interpretation.
In speaking of the increase in bulk of the general embryonic
mass Duval says, p. 430, “La cavité amniotique du Murin
prend naissance d’une maniere singuliére. I] se forme d’abord
un épaissisement massif de l’ectoderme, et ce processus rapelle
la masse amniotique pleine du cochon d’Inde; mais au lieu
que cette masse se creuse, comme chez ce rongeur, d’une
cavité centrale close de tous cdtés, elle se disloque irréguliére-
ment, chez le Murin, et s’ouvre a la surface, de l’oeuf, figurant
une bourse largement etalée, dont les bords se relévent alors
selon le type classique de replis amniotique et produisent
Yocclusion de ’amnios par leur rapprochment et soudure.”’

He then explains how after the roof of the amniotic cavity
has been separated off in small fragments from the solid floor
(figs. V;, Vg, Pl. 18) the true amnion is eventually formed by the
advance and fusion of the free edges of the true epiblastic mass
as shown in diagram V,,.

On my interpretation, the rapid growth of cells which
produces the ‘‘ amniotic mass” is due chiefly to the inner mass
cells, which cause the rupture of the overlying trophoblast
cells which now grow rapidly for a time, and form irregular
pieces which ultimately are thrown off and lost, leaving the
epiblast exposed on the surface. It has no connection at all
with the amnion formation, which is brought about subse-
quently by folds of the true epiblast, and not by trophoblast.

VoL, 41, PART 2,—NEW SERIES. R

238 RICHARD ASSHETON,

Tue Raspit, THE Moe, THE SHREW.

The growth of an outer layer of cells round an inner core
of cells of rather different texture was in the first instance
described by van Beneden, 1875, which process he regarded as
a kind of gastrulation, and to the form produced thereby he
gave the name of metagastrula. The outer layer he called
epiblast, the inner core hypoblast. Van Beneden’s description
is very complete, and was at first accepted, and his figures
passed into all text-books.

From the researches of Rauber and Kolliker, however, it
was seen that this interpretation could not be maintained,
because, as they showed, and as was confirmed by van Beneden
also in a later paper (van Beneden and Julin, 14), the
greater part of the inner core is transformed into the epiblast
of the actual embryo, while only a small portion becomes
hypoblast. But although the interpretation of the meta-
gastrula as given by van Beneden was thus shown to be in-
correct, does it necessitate the total abandonment of the
description given by van Beneden? Does a “ metagastrula”
stage not exist at all? Duval, on finding in his bats an
identical process to that described by van Beneden for the
rabbit, re-adopts the metagastrula for both rabbit and bat.

He writes (p. 153) : “ Cette gastrulation est celle, sauf une
différence a préciser plus loin, que van Beneden a décrite
chez le lapin sous le nom de Métagastrula, et qu’il a ensuite
admise pour les Chéiroptéres, d’aprés l’étude d’une vésicule
blastodermique du Rhinolophe, vésicule formée d’une sphére
externe d’ectoderme et d’une mass interne d’endoderme. Or,
depuis ses deux mémoires de 1880, van Beneden a abandonné
sa conception de la métagastrula des mammiféres. Nous devons
déclarer purement et simplement que nous reprenons cette
conception, et que nous nous préparons a la défendre.”

The evidence in favour of the occurrence of the metagastrula
in the rabbit may be briefly tabulated.

(1) Foremost is van Beneden’s (8, 9) original account. The

SEGMENTATION OF THE OVUM OF THE SHEEP. 239

difference between outer and inner cells dates from the first
division of the ovum.

(2) Keibel (38) states that he has been able to observe in a
number of cases in the rabbit the blastopore of van Beneden.

(3) Heape (29) has seen a similar appearance in the seg-
mented ovum of the mole.

_ (4) Duval has described the formation of a metagastrula as
a result of observations by sections in the case of the bat’s egg.

(5) In the sheep, also, I have found a stage exactly com-
parable to the metagastrula, though I place a different inter-
pretation upon it.

On the other hand, the value of van Beneden’s and Heape’s
observations must be to a certain degree discounted, because
they were based upon optical sections only.

Keibel does not state whether his observations are from real
or optical sections.

Then, again, although Heape finds a metagastrula stage at
the end of segmentation, he does not confirm van Beneden’s
account of its formation. On the contrary, he considers that
it is formed by the migration of yolk granules from the outer-
most segments to the innermost at the close of segmentation.

He writes (p. 168): “I have myself examined segmenting
ova of the rabbit, and have isolated the segments one from the
other, in order the more clearly to compare them, and in no
case have I been able to distinguish the slightest difference in
the density or constitution of these segments. If my observa-
tions are correct, then the differentiation of the segmentation
spheres into two layers in the fully segmented ovum is not
a primary differentiation, such as Beneden discerns, but a
secondary differentiation, due to the peculiar circumstances of
nutrition and development attending the formation of the
Mammalian embryo.”

According to Heape, it is only at the close of segmentation
that the segments become divided into two kinds. ‘“ A single
layer of cubical hyaline segments completely surrounds, except
at one point, an inner mass of rounded or polygonal densely
granular segments.”’ Thus Heape will not admit that there is

240 RICHARD ASSHETON.

any such process as a growth of one group of cells round
another group in the rabbit.

In 1894 a renewed study of this interesting stage was under-
taken by myself (1). I examined a very large number of
segmenting ova, both fresh, and by actual section. I was able
to confirm Heape’s account of the irregularity of the segmenta-
tion process, and quite failed to find any decisive evidence of
the metagastrula stage at all. In fact, the balance of evidence
was all against it, and I came to the conclusion that van
Beneden’s description was not supported by a study of the egg
by means of sections.

When, therefore, I read Hubrecht’s paper on Tupaia and
Duval’s on bats, and made my own observations upon the sheep,
I was not a little puzzled to know what to think of the rabbit.

To accept van Beneden’s account, as Duval proposes to do,
leads us again into all the old difficulties regarding the fate of
the inner mass, and, as explained above, it is, I think, impossible
to admit Duval’s method of escape from them.

Let it be supposed, on the other hand, that we admit van
Beneden’s view as far as the formation of the metagastrula
is concerned: is it not possible that this stage is succeeded
by a stage in which the outer layer becomes doubled, as in
my fig. 11 of the sheep, and that the cavity of the blastodermic
vesicle appears really between the cells of this doubled outer
layer? For instance, van Beneden’s (9) pl. iv, figs. 4, II, is very
much like my figure of the sheep (fig. 9). The outer layer is
distinctly thicker at one side than the other. May not this be
the commencement of a reduplication of the outer layer?

In the absence of any difference in colour or other recog-
nisable characteristics of the two groups of cells it is not
possible to affirm that this is the course of events, but it is
equally impossible to deny it.

I have again gone over the same ground in the rabbit with
new material, and have treated it as far as possible in the
same way that I treated the sheep embryos.

On examination of these embryos I am forced to come to
the same conclusion to which I came three years ago. In the

SEGMENTATION OF THE OVUM OF THE SHEEP. 241

majority of specimens there is no trace of a metagastrula stage.
In two specimens I find that some cells are distinctly darker than
others ; these cells are mostly internally placed, but they come
to the surface at more than one place and at nearly opposite
poles of the embryo. I have drawn similar sections in my
former paper (1, figs. 18, 19).

So, although I would not now assert that a metagastrula
stage does not exist in the rabbit, I must reiterate that I have
been unsuccessful in finding satisfactory evidence of it. But if
it does exist, I believe it to be open to the same interpretation
as that which I have placed upon the similar condition in the
sheep.

In the same way it is not difficult to explain the develop-
ment of the mole as described by Heape. In the absence of
any very marked difference between the characters of the two
groups of cells it is not easy to deny that any of the cells of
the inner mass have been derived from the outer wall of cells.
In the sheep the characters are markedly different. The
doubling of the outer layer at one point is clear.

In the mole (as in the pig) there is no such marked differ-
ence. Hence it is not possible to deny or to affirm such a
course of events.

Heape gives the subsequent separation of the hypoblast in
these words :— Certain of the cells bordering the blasto-
dermic cavity become separated off from the main portion of
the inner mass, and form a single layer of cells bordering the
mass on the inner side. This layer is the hypoblast. The
hypoblast is therefore derived from cells which result from the
multiplication of the inner cell mass present in the fully
segmented ovum.”

But is it not possible that these cells may have originally
been derived from the outer layer? Unfortunately Heape had
no real sections of any stage younger than his fig. 17, in
which the cavity of the blastodermic vesicle had attained
a great size.

The difference which Heape detected between the “ single
layer of cubical hyaline segments” surrounding “an inner

242 RICHARD ASSHETON.

mass of densely granular segments” (29, p. 166) had, accord-
ing to his description, a strictly physiological significance only,
and was due, he suggested, to the transmission of yolk con-
tained in the outer segments to the inner segments, this
transmission being performed in order that the changes about
to “take place in the constitution of the ovum may more
readily be performed” (p. 167). And the inner mass as it
appears in his figure of this stage (fig. 1) gives rise to definite
epiblast, definite hypoblast, and a part of the walls of the
blastodermic vesicle (vide 30, p. 418).

It is, however, possible that a stage similar to my figs. 8, 9,
11, of the sheep may have been missed, and that the metagas-
trula stage of the mole may be open to the same interpretation
as that which I offer for the sheep.

In Sorex the earliest specimen described by Hubrecht (34)
is one in which the cavity of the blastocyst is well established.
The segmentation stages are unfortunately not known. ;

It is, I think, by no means difficult to interpret Hubrecht’s
youngest stage (PI. 36, fig. 5) in the way that is indicated in
diagram Z (Pl. 18). There is, moreover, a distinct difference
in size between the cells comprising the embryonic knob.
Two are very much larger, and two more are larger still than
the others. Are these the epiblast, and all the rest hypoblast ?
(vide Hubrecht’s remarks upon it, pp. 506, 507).

Tue Rat, tHE Movse.

Robinson’s hypothesis, to which I have already alluded, was
based chiefly upon his interpretation of rat and mouse em-
bryos. I have suggested only a slight modification of his views
in connection with these animals. His idea with regard to
other animals was that the epiblast subsequently grew round
a pre-existing hypoblastic vesicle. It was in connection with
this inference that he met with so little sympathy.

In a former paper on the rabbit (1) I wrote, “ There is no
evidence in support of Robinson’s speculations concerning the
existence of a hypoblastic wall to the blastocyst surrounded
subsequently by the epiblast.”

SEGMENTATION OF THE OVUM OF THE SHEEP. 243

In making that statement I was confident that Robinson
was not right in the manner of application of his suggestions
to the rabbit. His remarks on the blastocyst of the rabbit
will be found on pp. 420—422 (46), part of which I must
quote here. ‘It is stated that the flat cells on the outer
surface of the inner mass either disappear entirely, or they
fuse with the cells immediately beneath them to form the
epiblast of the germ; and in either case, after the flat cells
over the outer surface of the inner mass have disappeared, it is
just as possible that the remainder of the vesicle wall hangs
in continuity with the peripheral flattened cells derived from
the inner layer of the inner mass—that is, with the hypoblast,
as with those which constitute the epiblast; but in the case of
the rabbit there is no evidence which will completely substan-
tiate a statement that either the one or the other of those
possibilities occurs. It is certain that the portion of the
vesicle wall, which is at first formed by a single layer of
flattened cells eventually becomes didermic, and that the
change from the single to the double layered condition com-
mences in the vicinity of the inner mass, whence it gradually
extends to the opposite pole of the ovum. It is stated that
the didermic condition is produced by the extension of the
hypoblast round the inner surface of the primitive wall, but
no satisfactory proof has been brought forward in support of
this statement. The cells of the extending layer are from
the first flattened, like those over which they are extending,
and nuclear division has not been clearly demonstrated either
in the inner or the outer layer.”

*“It appears to me, therefore, that the evidence which has
been obtained from the study of the rabbit’s ovum does not
conclusively substantiate the statement that the outer wall of
the primitive blastocyst is epiblastic in nature, and that the
hypoblast extends round its inner surface.”

I paid much attention to this, and I can have no doubt that the
rabbit’s blastocyst is not capable of receiving the interpretation
that Robinson suggests. The cells, which apparently ‘‘ extend
round ” the blastocyst, are beyond all doubt on the inside.

244 RICHARD ASSHETON.

Hubrecht (385) says in reference to Tupaia javanica, in
speaking of Robinson’s “ ingenious speculations,” ‘ According
to these views, the outer layer of the monodermic blastocyst
is in reality a hypoblastic layer. The didermic phase essen-
tially originates out of this by a gradual spreading of epiblast
cells outside the more primitive hypoblastic wall. However
ingenious these speculations may be, the author holds them to
be erroneous.”’ He then proceeds to give an account of the
development of Tupaia javanica “as an example of a
mammal, the early development of which furnishes us with
decisive evidence in this respect.”

So, although Robinson’s conclusions cannot be maintained
for the rabbit when applied in the way he did, there seems to
me to be no great objection to placing an interpretation upon
a much earlier stage, in accordance with my discoveries in the
sheep, as explained above.

I would, therefore, suggest a modification of Minot’s and
Robinson’s theories of a hypoblastic blastocyst as follows :

1. The whole of the subzonal epithelium is entodermic.

2. The central portion of the inner mass of the mammalian
blastocyst is ectodermic, but the surrounding layer is ento-
dermic.

3. “The cavity of the mammalian blastocyst does not
correspond with the segmentation cavity of the lower Verte-
brates, but with the archenteron”’ (Robinson).

4. The archenteron is at first a closed vesicle; there is no
blastopore.

In criticising Robinson’s suggestions with regard to the
rabbit, it must be remembered that he has described a stage in
the development of the ferret (47), in which the blastodermic
vesicle of the eleventh day is said to be a thin-walled vesicle of
hypoblast, bearing at one pole and on the outside, i. e. between
the main wall of the vesicle and the zona radiata, a small disc
of cells—the epiblast.

Thus Robinson finds in the ferret what he expected to be
the case in the rabbit.

If the blastocyst of the ferret becomes, as we may suppose,

SEGMENTATION OF THE OVUM OF THE SHEEP. 245

converted into a didermic vesicle, like those of other Carnivora,
we must conclude that it is brought about by the growth
round of the epiblastic disc cells. The process in the ferret
is, then, diametrically opposed to that by which an extremely
similar condition is brought about in the rabbit, Tupaia,
Talpa, &e.

It is, of course, just possible that Robinson may have been
deceived by the appearances of the sections he examined.
The stage which he had is an extremely difficult one, and in
this case was quite isolated; for the author had neither older
nor younger material. He had as many as seven specimens of
this one stage.

If we are to suppose that Robinson was quite right, we must
conclude that in the ferret the course of development seems
to be very different from that of any other mammal. But a
more extended research in the ferret should be made before
this conclusion is adopted.

Tur Two-LAYERED CONDITION OF THE HypoBLast.

With reference to the difficulty of the double-layered con-
dition of the hypoblast, I would point to my figs. 15 and 17
on Pl. 16, and ask what else but a two-layered condition could
result upon the occurrence of the expansion of the blastocyst
without a corresponding extension of the epiblastic knob? (v.
figs. 17 and 19).

The simple expansion of the blastocyst would be sufficient
to set up the two-layered condition by drawing out the angles
as shown diagrammatically in the figure on p. 222. The two-
layered condition once set up, the more complete separation of
fig. 19 might arise owing to difference in tension and so forth,
under which the layers might reasonably be supposed to exist.
In some cases where the expansion of the blastocyst does
not occur in this way there is no double layer of hypoblast,
e.g. Cavia, Mus (Robinson).

If, as suggested in the sequel, it is thought to be better to
compare the trophoblast more strictly with the yolk-bearing

246 RICHARD ASSHETON.

hypoblast of the meroblastic egg, then a distinction might
perhaps be drawn between true hypoblast and trophoblast, and
compared with the true hypoblast and yolk nuclei of the bird’s
egg, by whose union in the bird’s egg the bourellet endo-
dermique is formed (Duval).

Tuer RELATION OF THE ALLANTOIS TO THE AMNION.

In a former paper (1) on the early stages of the develop-
ment of the rabbit I ascribed the apparent extension of the
hypoblast round the inner surface of the blastocyst to the
presence of a region of special activity in the extra-embryonic
epiblast, or, as I would now rather call it, the trophoblast,
which begins to be manifest as early as the fifth day, and cul-
minates at last ‘in the production of the ectoplacental area,
or part of it” (p. 146). When discussing in a subsequent
paper (8) the mode of extension of the primitive streak meso-
blast, | was not able to understand how the extension of that
layer in the form of a wide-meshed reticulum, such as it is in
its peripheral parts, could be produced except by its being
dragged away from the primitive streak by the expanding
walls of the blastocyst in a way analogous to that assigned for
the spreading of the hypoblast.

Accordingly I thought that the overlying layer must also
have come from the primitive streak, and wrote, ““ We may
take the outline of the primitive streak mesoblast as indicat-
ing also the outline of the epiblast derived from the primitive
streak, at any rate for the stages up to about the one drawn
for fig. 6. At this time cells are separated from the hypo-
blast, both at points where there is already existing primitive
streak mesoblast, and at points where, up to now, there has
been no mesoblast.”” So far, I think, I was right; but in
drawing the diagrams on Pl. 22, and in writing the descrip-
tion of them, I seem to have neglected to consider the meso-
blast of hypoblastic origin in the region round the primitive
streak (originally described by Hensen), and so fell into the
error of taking the extreme outline of the mesoblast as an
indication of the limit of cells of primitive streak origin at this

SEGMENTATION OF THE OVUM OF THE SHEEP, 247

more advanced stage. If this were so, the greater part of the
ectoplacenta would be of primitive streak origin, and not tro-
phoblastic, which would be adverse to the hypothesis I have
been tempted to put forward as the result of my observations
on the segmentation of the sheep’s ovum.

From a renewed examination I am quite convinced that the
outskirts of the mesoblast surrounding the hinder end of the
embryonal area of the rabbit about the time the primitive
streak attains its maximum elongation, are in part made up of
cells of hypoblastic origin, though whether to such a large
extent as Bonnet describes for the sheep, and Hubrecht for
Sorex, I cannot say.

In the sheep Bonnet says the first formed mesoblast cells
are those of hypoblastic origin, and in this case it is clear that
the primitive streak mesoblast does not extend far beyond the
embryonic area; whereas in the rabbit the first formed cells
are undoubtedly those of epiblastic origin, and if these are
very quickly removed from the immediate region of the primi-
tive streak, as seems probable, it is impossible at present to
determine the actual boundary between the two sets of cells.

If the greater part of mesoblast which underlies the ecto-
placenta is of hypoblastic origin, and if the true primitive
streak mesoblast does not really extend beyond the inner
limits of the ectoplacenta—i.e. not further than the circle 4
or 3 in fig. 44 of the paper referred to above (8),—there would
then be no reason for supposing the ectoplacenta to be any-
thing but trophoblast, and the rabbit would agree with other
mammals in this respect.

Dr. Robinson tells me that in the ferret the ectoplacenta is
formed, and attachment to the uterus thereby effected, before
the mesoblast has appeared beneath the primitive streak. In
this case there is, therefore, still less reason for doubting a
trophoblastic origin for the ectoplacenta.

In my diagrams on Pl. 18, | have shown how the true
amnion is in most cases formed entirely of true epiblast; that
is to say, it can be shown to have been derived from the
embryonal area, and not the trophoblast.

248 RICHARD ASSHETON.

This is pretty obvious in cases like those of the guinea-pig
and Pteropus, in which the amnion is formed by the hollowing
out of a portion of the embryonic epiblast, and not by folding.

Duval’s account of the development of the amnion in the
bat almost as clearly proves it to be of true epiblastic origin.

In the rat also it is clearly the embryonic epiblast which
forms the amnion.

In all these cases the subzonal membrane which takes part
in the formation of the chorion is trophoblastic, and the true
amnion alone is epiblastic. But how is it in such cases as
those of the rabbit, pig, and mole? Is the amnion in these
cases also true epiblast ?

As regards the rabbit, I have nothing further to say, beyond
what I have written above. The amniotic folds arise between the
hinder end of the embryo and the ectoplacental region. Where
exactly the junction between epiblast and trophoblast lies cannot
be determined. In my diagram (L,) I have left it uncertain.

In the pig, however, I believe I have been able to trace the
junction. In an embryo thirteen days old, which exhibits the
earliest sign of amnion formation, the boundary between the
embryonal area and the trophoblast is just recognisable.

The embryonal area cells are still slightly hghter in colour,
the nuclei are on an average larger, and at many places the
remains of the membrana hypoblastica limitans are still
visible (figs. 26, 27).

From this specimen it appears that the amniotic folds arise
at about the junction of the embryonal area and the tropho-
blast. It gives one the impression that when completed the
true amnion will be formed of embryonal area epiblast and the
false amnion of trophoblast.

But, although it is exceedingly difficult to follow, I do not
think that this is exactly what happens. I fancy that in a
specimen in which the amnion, with the exception of the
* amnionnabelstrang,” is completed I can trace the junction
between epiblast and trophoblast within the limits of the
true amnion, as shown in my diagram Su,, and figs. 28,

PARMA ADE

SEGMENTATION OF THB OVUM OF THE SHEEP. 249

I have no doubt at all that the true epiblast takes some
part in the formation of the true amnion, but I am not at ail
sure that it forms the whole; in fact, I think it does not.

Since I hold that the trophoblast is probably of hypoblastic
origin in all these forms, it follows that the true chorion in
mammals is formed by the fusion of the allantois with hypo-
blast (i.e. trophoblast), whereas the analogous membrane in
birds is formed by the fusion of allantois with epiblast. To
some this may appear to be a fatal objection, as it implies
that the false amnion of a bird is not strictly homologous
with the false amnion of the pig.

To this I would reply, firstly, that the more primitive form of
amnion for mammals may be that which we find in the guinea-
pig or hedgehog, in which the parts which serve as false amnion
are developed quite apart from the production of the true
amnion. From these we get through stages such as are found
successively in the bat, pig, and rabbit, to a method of pro-
duction very similar to that of the bird, as the condition of
development became more and more alike.

Secondly, the essential feature of the hypothesis is that the
epiblast never grows round the ovum in placental mammals,
and with the exception of the true amnion it takes no part at
all in the formation of any part of the foetal membranes. In
birds the epiblast grows round the ovum, ultimately forming
a firm wall to a sac which enables the all-important organ,
the allantois, to become rapidly and effectively developed and
extended.

In mammals the sac is sufficiently well produced at an
earlier stage, and the conditions necessary for the growth of
the allantois are provided without need of the extension of the
epiblast.

The epiblast cells which grow round the yolk in the bird do
not take any part in the formation of the actual embryo; their
function ceases at the time of hatching, and they are thrown
off or absorbed.

The progress of the epiblast over the yolk in the bird is
described as being due to the multiplication of the epiblast

250 RICHARD ASSHETON.

cells themselves. It is an actual sliding of the free edge of
the epiblast over the yolk, and is therefore quite different from
the extension of the epiblast over the amphibian egg, where it
is not a sliding of one layer over the yolk, but a conversion in
situ of nuclei from the yolk into epiblast nuclei.

Let us, on the supposition that the mammalian ancestor had
a large yolked egg like that of a reptile or bird with vitelline
membrane, consider the probable steps which converted such a
condition to the one suggested in my paper.

Firstly, it must be remembered that the epiblast is cut off
from the rest of the ovum at a very early period—in the bird
about the time of laying. From this moment it never receives
any nuclei or cells from the yolk, but spreads by its own
interstitial growth. It does not of itself form a closed vesicle
until quite late in embryonic life.

On the other hand, the lower layer cells, after their separa-
tion from the epiblast, become continuous with yolk along the
germinal wall (bourrelet entodermo-vitellin—Duval),
and form at once a closed vesicle, which now rapidly fills with
fluid and expands.

This closed vesicle, which is formed by true hypoblast and
yolk-containing hypoblast, has two functions in addition to
that of supplying actual cell material,—namely, of providing
nourishment, and of supplying a roomy cavity for the reception
of the embryo while forming.

If the same condition appertained in the large-yolked egg of
the mammalian ancestor, is it not more likely that the same
group of cells should have continued during the substitution
of maternal fluids for yolk the offices of supplying nutriment
as before, and of the maintenance of a large cavity for the
reception of the embryo, than that the other group of cells
should acquire them ?

Duval (24) has described a thickening, and a series of villous
processes on the edges of the extending epiblast of the bird’s
egg, which according to him form with the allantois a “pla-
centa,’”’ by means of which nourishment is obtained from the
albumen towards the end of incubation. Nevertheless by far

SEGMENTATION OF THE OVUM OF THE SHEEP. 251

the greater portion of nourishment is obtained from the yolk
contained in the hypoblast cells.

The difference in size between the fertilised ovum of a reptile
or bird, and of a mammal is very great; but the difference in
size between the embryo of, say, a bird, with one pair of meso-
blastic somites, and of a mammal of the same age, is compara-
tively small. This means that nearly the same space is
required for the production of the mammalian embryo as of
the Sauropsidan, and has to be provided.

As the egg diminished in size, owing to the loss of yolk, a
corresponding increase in activity of the cells of the wall of the
vesicle must have taken place to keep up the size of the vesicle.
The smaller the egg became, the longer would be the time
occupied in the attainment by the vesicle of the requisite size.

Now, as the epiblast plays the more prominent part in the
formation of the bulk of the embryo during the earliest stages,
it clearly would be useless for the embryonic part to exhibit
much energy of growth until the old conditions were to a
certain extent regained; hence the lethargy exhibited by the
embryonic epiblast in mammals during the first week of
development. No feature of the early stages of the mammalian
embryo is more striking than this inertness of the embryonic
epiblast—or, as I should now prefer to call it, simply epiblast
—during the first few days.

A need for the extension of the edges of the epiblast as a
protective layer, asin the Sauropsida, has been obviated by the
development among mammals of the strong zona radiata; and
any tendency which existed for it to spread may very easily be
conceived to have been prevented by the increased activity of
the hypoblast cells, now no longer laden with yolk. This in
many cases resulted in a temporary overflow—as it were—
round the epiblast, aided perhaps originally, if not actually in
present embryos, by the restricted confines of the zona radiata.
Some such intermediate stage is shown in fig. D on Pl. 18.

It is probable that for the more primitive history of the
relation of trophoblast to the uterus and placenta we ought
to turn to the smallest mammals; and it is there that we see

252 RICHARD ASSHETON.

the closest connection between the walls of the blastocyst and
the mother before the formation of the placenta.

In large animals, such as the sheep and pig, it is doubtful
whether there is any connection quite comparable to the
ectoplacenta or trager eating its way into the maternal
tissue. The energy is present, but shows itself in an enormous
growth of the vesicle, which no doubt answers the purpose of
acquiring nutriment, but in a rather different way.

In many of the smallest mammals we find also great
differences in the mode of formation of the amnion. For
instance, the amnion of Erinaceus, Mus, Cavia, Vespertilio,
differ much from each other and from the Sauropsidan type.
In the larger animals, Sus, Ovis, and in those smaller ones
which retain the zona radiata for some considerable time,
Lepus, Sorex, Talpa, the amnion formation is apparently more
like that of the bird.

Do the smaller mammals really show us the more primitive
type of amnion development for the mammalia ?

One is tempted to ask, did a continuance of the retardation
of the epiblast and acceleration of the hypoblastic mass with
its tendency to surround the former, lead to a curving upwards
of the edges of the epiblast, which, by means of their subsequent
expansion, gave rise to an amnion which is essentially different
from the amnion of the Sauropsida,—an amnion dependent
in no way on the force of gravity, and less on the action
of the internal fluid than in the Sauropsida, and therefore less
likely to be so much affected by the frequent alteration of
position which must occur in the mammalian uterus? Such
an amnion formation is actually found in the case of Vesper-
tilio (Duval).

On this view the free edges of the advancing amnion fold,
as described by Duval, would be both morphologically and
physiologically equivalent to the advancing free edges of the
epiblast of the blastoderm of the Sauropsidan egg.

A hastening of this process might lead to the condition,in
Pteropus (52) or Cavia in which the amnion is from the first
appearance perfect.

SEGMENTATION OF THE OVUM OF THE SHEEP. 253

A delay in the accomplishment of the formation of the
amnion until after the complete attachment of the edges of the
true epiblast to the trophoblast subsequent to its rupture would
cause a condition as in the pig or sheep, when the first starting
of the amnion seems to be due to the bending up of the edges
of the embryonic area, but the full completion of the amnion
to be due to other causes, such as the pressure of the enclosed
fluids. Thus a state analogous to the Sauropsida is attained.
In this case (pig), as I have shown in my figs. Su,, Su;, Pl. 18,
the true amnion is only partly formed by the true epiblast.
If this is a correct view, it is interesting to note that there is
not the same tendency for the edges of the amnion to fuse in this
case, pig (also sheep), as where the free edges of the epiblast
coalesce.

It seems not improbable that in the rabbit this latter process
has gone even further. It differs much in its formation from
that of the pig. It arises much later, and is practically formed
entirely from the hinder fold.

It is much to be regretted that we have not yet a detailed
account of the development of the monotreme blastoderm.
Semon’s account of the segmentation of the ovum does not
extend as far as the separation of the true hypoblast as a
permanent layer. There is no trace of an archenteron in his
oldest stage on pl. ix of his work.

I shall not attempt to reconcile this hypothesis with the
description of the opossum given by Selenka. The develop-
ment of this animal is very unlike that of any placental
mammal. Although it has what at first sight seems to be a
typical blastodermic vesicle (v. Selenka [51], pl. xix, figs. 1, 5),
this, nevertheless, differs from all forms hitherto observed by
the absence of a Rauber layer over the formative epiblast. It
is clear from this fact, and from the presence of a large “ seg-
mentation cavity ” which never disappears, the evident ‘ blasto-
pore,’ and the peculiar origin of the entoderm, that the
Marsupials in their development differ more widely from the
placental mammals than the members of the latter group differ
among themselves. Selenka’s account is by no means com-

vo. 41, PART 2.—NEW SERIES. s

DBA RICHARD ASSHETON.

plete, and in the absence of a more extended research into
the process of segmentation of the ovum of the Marsupialia
and Monotremata it is useless to discuss at any length the
present hypothesis with regard to those groups of mammals.
As far as we can gather from the little which Caldwell
and Semon have published concerning the segmentation of
the Monotreme egg, we must expect that the eggs of these
mammals resemble those of the Sauropsida, and that in them
the epiblast does grow round the yolk, producing a double-
walled sac consisting of an outer layer of epiblast, and an
inner layer of large yolk-laden hypoblast, as described by
Hill and Martin (31) in a specimen obtained from the uterus
of an ornithorhynchus, in which seventeen mesoblastic somites
were present, How this sac is formed we do not know.

Whether the Marsupial development resembles that of the
Monotreme or that of the Placentalia more closely cannot at
present be decided.

Whichever view is taken of Selenka’s description of the
opossum, many obvious difficulties remain, for the solution of
which no satisfactory suggestion can as yet be offered. If, as
I believe, the development of the placental mammals shows us,
the protomammalian had a large-yolked egg, the Monotreme’s
development would be of the greatest possible interest.
Hubrecht has, in the very interesting paper to which I have
frequently referred, most ably advanced the opposite view
that the protomammalian egg was a small egg of the amphi-
bian type, and that the Monotreme is an aberrant offshoot. If
I prefer to support the other view I am only adhering
to the opinion expressed by Balfour in his great work on
‘Comparative Embryology,’ 2nd ed., p. 189 :—‘* The features
of the development of the placental Mammalia receive their
most satisfactory explanation on the hypothesis that their
ancestors were provided with a large-yolked ovum like that of
the Sauropsida. . . . The embryonic evidence of the common
origin of Mammalia and Sauropsida, both as concerns the
formation of the layers and of the embryonic membranes, is as
clear as it can be, The only difficulty about the early

SEGMENTATION OF THE OVUM OF THE SHEEP. 200

development of Mammalia is presented by the epibolic gas-
trula and the formation of the blastodermic vesicle. . . . No
satisfactory phylogenetic explanation of the mammalian
gastrula has, in my opinion, as yet been offered.”

Although my inferences as to the homologies of the amniotic
folds cannot be said to have been contemplated by the writer of
these words, I hope it may be thought that my work upon the
segmentation of the sheep might have removed from his mind
the difficulty of the supposed epibolic gastrula of the mammal
by showing that such a phenomenon does not occur, and that the
appearance which has given rise to the misconception is capable
of quite a different interpretation in complete accord with his
view that the development of the mammalian ovum bears
traces of a descent from an ancestor with a large-yolked egg.

REFERENCES.

1. AssHEToN, R.—“ A Re-investigation into the Early Stages of the Deve-
lopment of the Rabbit,” ‘ Quart. Journ. Micr. Sci.,’ vol. xxxvii, 1894.

2. AssHEToN, R.—On the Phenomenon of the Fusion of the Epiblastic
Layers in the Rabbit and in the Frog,’ ‘Quart. Journ. Mier. Sci.,’
vol. xxxvil, 1894.

8. AssHETON, R.—“ The Primitive Streak of the Rabbit ; the Causes which
may determine its Shape, and the Part of the Embryo formed by its
Activity,” ‘Quart. Journ. Micr. Sci.,’ vol. xxxvii, 1894.

4, AssHeTon, R.—“ An Experimental Examination into the Growth of the
Blastoderm of the Chick,” ‘ Proc. Roy. Soc.,’ vol. lx, 1896.

5. Assueton, R.—“ The First Ten Days of the Development of the Pig,”
(in the press) ‘Quart. Journ. Micr. Sci.,’ the present volume, 1898.

6. Batrour, F. M.—‘ Comparative Embryology,’ 2nd edit., London, 1885.

7. von Barr, K. H.—‘ Ueber Entwickelungsgeschichte der Thiere,’? Kén-
igsberg, 1828.

8. van Benepen, E.—“ De la maturation de l’ceuf, de la fécondation et des
premiers phénomenes embryonnaires chez les Mammiféres d’aprés les
observations faites chez le lapin,” ‘ Bulletins de l’Académie Royale
des Sciences de Belgique,’ 1875.

9. vaN BrenepEN, E.—“ Recherches sur l’embryologie des Mammifeéres, la
formation des feuillets chez le lapin,” ‘Archives de Biologie,’ tome i
1880.

256 RICHARD ASSHETON.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24,

25.

vaAN Brengepren, E.—“ De la formation et de la constitution du placenta
chez le Murin,’ ‘ Bulletins de Académie Royale des Sciences de
Belgique,’ 1888.

vAN BENEDEN, E.—“ De la fixation du blastocyste a la muqueuse utérins
chez le Murin,” ‘ Bulletins de Académie Royale des Sciences de
Belgique,’ 1888.

vaN Brnepen, E.—‘ Untersuchungen iiber die Blatterbildung den
Chordakanal, und die Gastrulation bei den Saugethiere, Kaninchen,
und Vespertilio Murinus betreffend,” ‘ Anatomische Anzeiger,’
1889.

vAN BENEDEN, E., et Jutin, C.—‘‘ Observations sur la Maturation, la
Fécondation, et la Segmentation de l’ceuf chez les Cheiroptéres,”
‘Archives de Biologie,’ tome i, 1880.

vAN BENEDEN, E., et Juin, C.—‘‘ Recherches sur la formation des
annexes foetales chez les Mammifeéres,” ‘ Archives de Biologie,’ t. v.

Biscnorr, T. L. W.—‘ Entwickelungsgeschichte des Kaninchen-Hies,’
Braunschweig, 1842.

Biscuorr, T. L. W.—‘ Entwickelungsgeschichte des Hundes-Hies,’
Braunschweig, 1845.

Biscuorr, T. L. W.—‘ Entwickelungsgeschichte des Meerschweinchens,
Giessen, 1852.

Biscuorr, T. L. W.—‘ Entwickelungsgeschichte des Rehes,’ Giessen,
1854.

Bonnet, R.—‘ Beitrage zur Embryologie der Wiederkaiier Gewonen
am Schafel,” ‘Archiv fiir Anatomie und Physiologie,’ Anatomische
Abtheilung, 1884.

Bonnet, R.—“ Beitrage zur Embryologie der Wiederkaiier,” ‘ Archiv
fir Anatomie und Physiologie,’ Anatomische Abtheilung, 1889.

Bonnet, R.—‘Grundriss der Entwicklungsgeschichte der Haussauge-
thiere,’ Berlin, 1891.

Catpwe1t1L, W. H.—* The Embryology of Monotremata and Marsupialia,”
pt. 1, ‘ Philosophical Transactions of the Royal Society,’ London,
vol. elxxvili, 1887.

Duvat, M.—‘‘ De la formation du Blastoderme dans |’ceuf d’Oiseau,”
‘Annales des Sciences Naturelles, Zoologie,’ vol. xviii, 1884.

Duvar, M.—* Etudes Histologiques et Morphologiques sur les annexes
des Embryons d’Oiseau,” ‘Journal de l’Anatomie et de la Physiologie,’
vol. xx, 1884.

DouvaL, M.—“ Le Placenta des Rongeurs,” ‘Journal de l’Anatomie et
de la Physiologie,’ 1889—1892.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

SEGMENTATION OF THE OVUM OF THE SHEBP. 257

Duvat, M.—“ Etudes sur ’Embryologie des Chéiroptéres,” ‘Journal de
Anatomie et de la Physiologie,’ vols. xxxi, xxxii, 1895, 1896.

Happon, A. C.—‘‘ Note on the Blastodermic Vesicle of Mammals,”
‘The Scientific Proceedings of the Royal Dublin Society,’ 1885.

HensEn, V.— Beobachtungen iiber die Befruchtung und Entwickelung
des Kaninchens und Meerschweinchens,” ‘ Zeitschrift fiir Anatomie
und Entwickelungsgeschichte,’ 1875, 1876.

Heart, Watter.—‘ The Development of the Mole (Talpa europea);
the Ovarian Ovum, and Segmentation of the Ovum,” ‘ Quart. Journ.
Micr. Sci.,’ vol. xxvi.

Hearse, Watter.— The Development of the Mole (Talpa europea) ;
the Formation of the Germinal Layers, and Harly Development of the
Medullary Groove and Notochord,”’ ‘Quart. Journ. Micr. Sci.,’
vol. xxiii.

Hint, J. P., and Martin, C. J.—“ On a Platypus Embryo from the
Intra-uterine Egg,” ‘ Proceedings of the Linnean Society of New
South Wales,’ vol. x, 1895.

Husrecut, A. A. W.—“ Die Erste Aunlage des Hypoblastes bei den
Saugethieren,” ‘ Anatomischer Anzeiger,’ vol. iii, 1888,

Husrecut, A. A. W.—<Studies in Mammalian Embryology. I. The
Placentation of Hrinaceus europeus, with Remarks on the
Physiology of the Placenta,” ‘Quart. Journ. Micr. Sci.,’ vol. xxx,
1889.

Husrecut, A. A. W.—“ Studies in Mammalian Embryology. II. The
Development of the Germinal Layers of Sorex vulgaris,” ‘ Quart.
Journ. Micr. Sci.,’ vol. xxxi, 1890.

Husrecut, A. A. W.—“‘ On the Didermic Blastocyst of the Mammalia,”
‘Report of British Association,’ 1894.

Husrecut, A. A. W.— Die Phylogenese des Amnions und die Bedeut-
ung des Trophoblastes,” ‘ Verhandelingen der Koninklijke Akademie
van Weterschappen te Amsterdam,’ 1895.

KerseL, F.—“ Zur Entwickelungsgeschichte des Igels (Wrinacens
europeus),” ‘Anatomischer Anzeiger,’ vol. ili, 1888.

KxiseL, F.—“ Van Beneden’s Blastoprus und die Rauber’sche Deck-
schicht,” ‘ Anatomischer Anzeiger,’ ii, 1887.

Keiser, F.—“ Zur Entwickelungsgeschichte der Chorda bei Saugern
(Meerschweinchen und Kaninchen),” ‘Archiv fiir Anatomie und
Physiologie,’ Anatomische Abtheilung, 1889.

KeriBeL, F.—‘ Studien zur Entwickelungsgeschichte des Schweines,’’ I,
‘ Morphologische Arbeiten,’ Bd. iu.

258

4l.

42.

43.

44.

45.

46.

47.

48.

49.

50.

61.
52.

53.

54.

55.

56.

RICHARD ASSHETON.

Keiser, F.—“ Studien zur Entwickelungsgeschichte des Schweines,”’ II
‘Morphologische Arbeiten,’ Bd. v.

Kouirker, A.—“ Die Entwickelung der Keimblatter des Kaninchens,”
‘Zeitschrift zur Feier des zoojahrigen Bestehens der Julius-Maximilians-
Universitat zu Wurzburg,’ 1882.

Minot, C. 8.—‘ Human Embryology,’ New York, 1892.

RavuBeR.— Die Entwickelung des Kaninchens,” ‘ Sitzungsberichte der
Naturforschenden Gesellschaft,’ 1875.

Roginson, A.—“ On the Nutritive Importance of the Yolk-sac,” ‘ Journal
of Anatomy and Physiology,’ 1892.

Roginson, A.—‘ Observations upon the Development of the Segmenta-
tion Cavity, the Archenteron, the Germinal Layers, aud the Amnion in
Mammals,” ‘ Quart. Journ. Mier. Sci.,’ vol. xxxiii, 1891, 1892.

Roxzinson, A.—‘‘ Observations upon the Development of the Common
Ferret, Mustela ferox,” ‘Anatomischer Anzeiger,’ vol. vill, 1893.

Scuirer, EH. A.—‘ Description of a Mammalian Ovum in an Harly Con-
dition of Development,” ‘Proceedings of the Royal Society of
London,’ 1876.

Srtenka, H.—‘ Keimblatter und Primitivorgane der Maus,’ Wiesbaden,
1883.

SELENKA, H.—“ Die Blatterumkehrung im Hi der Nagethiere,’ Wies-
baden, 1884.

SELENKA, li.—‘ Das Opossum,’ Wiesbaden, 1886, 1887.

SELENKA, H.—‘ Studien tiber Eutwickelungsgeschichte der Tiere,’ Wies-
baden, 1891.

Semon, R.—‘ Zoologische Forschungsreisen in Australien,’ vol. ii,
“* Monotremen u. Marsupialien,’ Jena, 1894.

Sozorra, J.—‘‘ Die Befruchtung und Furchung des Hies der Maus,”
‘ Archiv fiir mikroskopische Anatomie,’ 1895.

Tarani, A.— La Fécondation et la Segmentation étudiées dans les ceufs
des Rats,” ‘ Archives Ital. de Biologie,’ 1889.

Weyssz, A. W.—‘ On the Blastodermic Vesicle of Sus scrofa-domes-
ticus,” ‘American Academy of Arts and Sciences,’ vol. xxx, 1894.

SEGMENTATION OF THE OVUM OF THE SHEEP. 259

DESCRIPTION OF PLATES 15—18,

Illustrating Mr. Richard Assheton’s paper “ On the Segmen-
tation of the Ovum of the Sheep, with Observations on the
Hypothesis of a Hypoblastic Origin for the Trophoblast.”

List or REFERENCE LETTERS.

K. Epiblast. H. Hypoblast. M. Mesoblast. M.H.L. Membrana hypo-
blastica limitans. T. Trophoblast. Y. Spherical bodies found near the
surface of the unsegmented egg. Z. Zona radiata.

PLATE 15.

Fic. 1 (Ovis 3, No. 1).—Unfertilised ovum from the uterus.

Fie. 2 (Ovis 36, No. 1).—Age 14 days. Section of the fertilised ovum of
a sheep, showing male and female pronucleus. Found in the Fallopian tube.
Preserved in ‘5 per cent. chromic acid. Stain carmalum. Thickness of
section 0075 mm. x 420.

Fie, 3 (Ovis 35, No. 2).—Age 3 days. Section of the fertilised ovum of a
sheep. Fertilisation completed, and the division of the ovum is taking place.
Preserved in 1 per cent. chromic acid. Stain, Klein, hematox. Thickness
of section 0075 mm. x 3880.

Fie. 4 (Ovis 33, No. 1).—Age 24 days. Section of a segmented ovum of
a sheep with six segments. Preserved weak Flemming. Stain carmalum.,
Thickness of section (0075 mm. x 880,

Fig. 5 (Ovis 49, No. 1).—Age between 4 and 6 days. Section of a
segmented ovum of a sheep with eight segments. Stain carmalum and pieric
acid. Thickness of section 006 mm. x 380.

Fic. 6 (Ovis 63, No. 1).—Age 4 days. Section of a segmented ovum of a
sheep with sixteen segments. Stain carmalum. Thickness of section
005 mm. x 380.

Fie. 7 (Ovis 33, No. 2).—Age 23 days. Section of an ovum of the sheep
with eight segments. Preserved in *5 per cent. chromic acid. Stain car-
malum. Thickness of section ‘0075 mm. x 420.

Fic. 8 (Ovis 18, No. 1).—Age 4 days. Section of the ovum of a sheep
with thirty segments, six of which are large and light in colour. Preserved
in 5 per cent. chromicacid. Staincarmalum. Thickness of section ‘005 mm.
x 420.

Fic. 9 (Ovis 14, No. 1).—Age 6 days. Section of the ovum of a sheep in
which the large light-coloured cells have been completely surrounded by the

260 RICHARD ASSHBE'ION.

small darker segments. Preserved in chromic acid. Stain borax carmine.
Thickness of section ‘(005 mm. x 420.

Fie. 10 (Ovis 14, No. 2).—Age 6 days. Section through an embryo in
which the dark outer layer has become doubled at one pole. Preserved in
chomic acid. Stain hemalum. Thickness of section‘01 mm. x 380.

Fie. 11 (Ovis 6, No. 1).—Age 7 days. Section of the segmenting ovum
of the sheep in which the doubling of the outer layer of dark cells is very
distinct, and the same layer has become thinner at the opposite pole. Pre-
served in chromic acid. Stain carmalum. Thickness of section ‘0075 mm.
xX 420.

Fig. 12 (Ovis 4, No. 1).—Age 53 days. Section through an embryo of
the sheep in which the cavity of the blastocyst is beginning to be formed.
It apparently appears amongst the dark cells. Preserved in chromic acid.
Stain carmalum. Thickness of section ‘0075 mm. x 380.

PLATE 16.

Fie. 13 (Ovis 4, No. 1).—Another section through the same embryo as
fig. 12. x 380.

Fie. 14 (Ovis 48, No. 1).—Age uncertain. Section through the embryo
of a sheep with cavity of the blastocyst well established. Preserved in
chromic acid. Stain carmalum. Thickness of section ‘0075 mm. x 380.

Fig. 15 (Ovis 4, No. 2).—Age 53 days. Section through the embryo of a
sheep slightly more advanced than the above. Preserved in chromic acid.
Stain carmalum. x 880.

Fie. 16 (Ovis 34, No. 1).—Age unknown. A section through the embryo
of a sheep just before the rupture of the zona radiata. Preserved in chromic
acid. Stain carmalum. Thickness of section‘0075 mm. x 380.

Fic. 17 (Ovis 34, No. 2).—Age unknown. A section through the embryo
of a sheep immediately after the rupture of the zona radiata. Thickness of
section ‘0075 mm. x 380.

Fie. 18 (Ovis 42).—Age 9 days.

Fie. 19.—Section through above along the line a—d.

Fic. 20 (Ovis 8).—Age 10 days.

Fre. 21.—Section through above specimen (Fig. 20) along the line v—d.
Preserved in chromic acid. Stain carmalum. Thickness of section ‘0075 mm.
x 380.

Fic. 22.—Camera outline drawing of above section to show the extension
of the inner hypoblast layer.

SEGMENTATION OF THE OVUM OF THE SHEEP. 26]

PLATE 17.

Fic. 23 (Ovis 40, No. 1).—Age 6 days (?). Segmenting ovum with fifteen
segments, which surrounded a central cavity comparable to that shown in
fig. 6 on Plate 15.

Fic. 24 (Ovis 40, No. 2).—Age 6 days (?). An eight-segment embryo
separated into two groups of four cells. One segment contained certain
round bodies which are not found in any of the others. Stain borax carmine.

Fie. 25 (Ovis 33, No. 1).—Age 23 days. Six-segment stage of sheep.
A section of this is shown in fig. 4, Plate 15.

Fie. 26.—Section of the side of an embryonal area of the pig of thirteen
days; to show the difference in the character of the epiblast and trophoblast,
at the place where the first sign of the amnion fold can be detected.

Fic. 27.—Another section of a similar embryo of the pig.

Fic. 28.—The whole section, from a part of which Fig. 29 was drawn. Pig
of about 17 days. The amnion and false amnion folds are seen.

Fic. 29.—A section of a portion of the amnion fold of a pig of about 17
days. A very sudden alteration in the character of the epiblast of the amnion
between the letters T and E is supposed to indicate a transition from tropho-
blast to true epiblast of Lhe embryonal area.

PLATE 18.

In the diagrams on this plate, pink indicates epiblast, green hypoblast. The
black ring is the zona radiata.

Fie. A represents the ovum of a bird while in the oviduct, showing the
segmentation cavity. Compare Duval (28), fig. 29.

Fie. B.—The ovum of a bird about the time of laying. Compare Duval’s
‘Atlas,’ fig. 33.

Fic. C.—The ovum of a bird after aefew hours of incubation. The sub-
germinal cavity is now well established. Compare Duval’s ‘Atlas,’ figs.
49—55, and Duval (28), 50—53.

Fic. D.—A diagram to illustrate a supposed intermediate condition of a
protoentherian ovum in which a tendency to grow over the epiblast is
exhibited by cell masses of hypoblastic origin.

Fig. W.—The segmenting ovum of a sheep and bat, and by hypothesis of
all placental mammals.

Fic. X.—Another later stage of a segmenting ovum of Ovis, Tupaia, and

by hypothesis of some other mammals. The hypoblast has completely
surrounded the epiblast.

262 RICHARD ASSHETON.

Fic. Y.—A stage found in Ovis, and by hypothesis common to many other
mammals. The hypoblast is doubled over one pole.

Fie. Z.—A typical mammalian blastodermic vesicle interpreted according
to the specimens figured on Plate 2 of this paper, supposed to be common to
many mammals besides the sheep.

Fies. T,—T,.—Four diagrams illustrating the interpretation according to
the hypothesis of some of the succeeding stages of the development of the
mole (Talpa europea). Compare Heape (80), figs. 18, 23, 26, 27.

Fics. Tu—Tv,.—Three diagrams illnstrating the stages subsequent to Z of
Tupaia javanica. Compare Hubrecht (86), figs. 16, 23, 33, 41, 62, 68,
with diagrams X to Tv,.

Kies. Su—Su,.—Six diagrams illustrating the stages subsequent to Z of
Sus scrofa. Compare Assheton (5), figs. 19, 31, 39, and Keibel (40),
and figs. on Plate 17 of this paper.

Fics. L—L,.—Five diagrams illustrating the stages of the rabbit (Lepus
cunicula) subsequent to Z.

Fries. E,—E,.—Four diagrams illustrating the stages subsequent to W of
Erinaceus interpreted according to hypothesis. Compare with Hubrecht (83),
figs. 25, 39, 15, 20 4, 51, 49, &e.

Figs. V—V;.—Hight diagrams illustrating the development of Vespertilio.
Compare figs. W, V, with Duval (26), figs. 15, 21, fig. V, with Van Beneden
and Julin (18), fig. 6, and V,—V, with Duval (26), figs. 33, 36, 41, 50, 76, 102.

Fics. M,—M,.—Five diagrams of Mus. Compare with Robinson (46),
figs. 4, 6, 8, 10, 13.

Fies. C,—C,.—Four diagrams of Cavia. Compare with Selenka (49),
pls. xi, xii, figs. 1, 3, 9, 13.

HEART-BODY, ETC., OF CERTAIN POLYCHATA. 263

On the Heart-body and Celomic Fluid of
certain Polycheta.

By

Lionel James Picton, B.A.

With Plates 19—22.

[This work was done at Naples during my tenure of the Oxford Scholar-
ship there in 1896-7.

Besides the subjects mentioned in the title, there is a description of a new
bow-shaped corpuscle in Notomastus profundus on p. 268, foot-
note; and an appendix at the end of the paper concerning intra-vitam

staining of nuclei of the gut-wail with carmine in sea-water.—
sds)

In several groups of Polychzta there is found in the dorsal
vessel or “ branchial heart” a rod-like structure usually
attached anteriorly and posteriorly, and sometimes at other
points, to the vascular walls ; but otherwise lying in the lumen
of the tube freely. It is in many instances of a deep brown
colour, due to the presence of numerous pigment granules
deposited in it. The cells in which these are contained are
small and thick-walled. They have little protoplasm, but
well-marked nuclei. This organ, under the various names
“corps cardiaque,’ ‘‘ Herz-Korper,” “second cesophagus,”
“cecum gastro-cesophagien,” and others, has been discussed
by several authors.

264 LIONEL JAMES PICTON.

Claparéde (9, A, p. 399)? mentions in the dorsal vessel of
Terebella multisetosa “ une substance d’un noir profond,
distribuée en cordons irréguliers;” and describes a similar
structure also in the Cirratulide. He further notes (9, B)
that chloragogen, the so-called “ liver” tissue of Annelids,
which usually clothes the gut and the exterior of the chief
blood-vessels, is entirely absent in those forms in which “les
masses intravasculaires ” are known to exist.

Salensky (28) describes its state in the larva of Terebella
Meckelii. Though not observed at the earliest period, at a
certain stage an opening in the wall of the vessel was noted,
which led into the cardiac body. Salensky does not state the
exact position of this orifice, but says, “Il s’ouvre d’abord
dans un petit tube bien étroit et se continue ensuite dans le
corps cardiaque, qui a ce stade du développement représente
un organe de forme cylindrique.” For the rest, the state of
the heart-body is described as being much the same as in
the adult.

In the Chlorhemide many naturalists have observed the
peculiar dorsal vessel which appears to spring from the wall
of the gut, and is of a very deep colour on account of the
heart-body contained in it. Horst (16) describes the mis-
taken views of Otto, who thought it was a second cesophagus ;
of Dujardin, who regarded it as a mere blood-vessel; of
Claparéde, who thought it was a gland opening into the
pharynx, and of others; and himself considers that it is not
merely the heart, nor a gland only, but that it is a gland
contained in the heart. He thus combines the views of
Dujardin and Claparéde. He further holds that the heart-
body is derived from a pouch of the gut comparable with that
of Enchytreeide.

Kennel (20) considers a structure in the dorsal vessel of _
Ctenodrilus as the homologue of the heart-body of Terebella.

Steen (82, p. 42), in the case of Terebellides Stroemii,
thinks the heart-body has a valvular function, preventing a

1 The numbers refer to the Bibliography at the end of the paper.

HEART-BODY, ETC., OF CERTAIN POLYCHATA, 265

back-flow caused by the contracting gills. He describes it as
enclosed in a fine membrane.

Kisig (12, p. 691) mentions the statements of the earlier
writers on the heart-body. First, ordinary chloragogen is
absent in the forms which have a heart-body, whilst the
heart-body is very similar to chloragogen tissue. He there-
fore proposes to call it “ intra-vascular chloragogen.”” Secondly,
he considers that Salensky’s statements make it very probable
that it is derived from the wall of the dorsal vessel and from
peritoneal tissue, from which the ccelomic corpuscles! are also
produced. This is equivalent to saying that chloragogen,
which is a modification of peritoneal tissue, instead of remain-
ing on the outside of the dorsal vessel, becomes folded so as
to occupy its interior. Both to celomic corpuscles and to
chloragogen excretory functions have been attributed, and this
view regards them as having a morphological connection.

Cunningham (10) describes the heart-body in Cirratulide,
Chlorhemidez, and Terebellide, and mentions its occurrence
in Amparetide and Amphictenidz. He is inclined to regard
it as a ductless gland of mesoblastic origin, and he suggests
its possible homology with the “ notochord ” of Balanoglossus.

E. Meyer (24, B) describes the heart-body of Chetozone
szetosa, and expresses the opinion that its function consists
in the preparation of the pigment which is dissolved in the
red blood-fluid.

Jourdan (17),in his work on Siphonostoma diplochetos,
gives an account of the “czcum gastro-cesophagien,” or heart-
body. He speaks of it as ‘‘communiquant avec le tube
digestif,” which is denied by Bles (5), who supports the view
that the heart-body is of mesoblastic origin, remarking that
chloragogen is peritoneal, and that the connections of the

1 « Haemolymph ” is the word used, but it evidently refers to the coelomic
fluid. Would it not be better to restrict the term “hemolymph” to such
fluids as represent the combined blood and lymph, such as that of the Capitel-
lide, to which Hisig previously applied the word? It was first used by
Professor Ray Lankester for a fluid consisting, like Vertebrate blood, of both
“hema” and “lymph,” when discussing the presence of corpuscles in what
was then called the “‘ pseud-hemal ” system of Cheetopods (this Journal, 1878).

266 LIONEL JAMES PICTON.

heart-body with the vascular walls suggest a similar source
of origin for it. He regards the organ as a blood-pigment
gland.

Miss Buchanan (6,A) found a heart-body in Hekatero-
branchus Shrubsolii; and also (7, B) a structure, which is
wanting in the adult, in the dorsal vessel of the developing
Magelona. She regards it as a larval heart-body, and supposes
that on its dissolution it forms the blood-corpuscles.

Cuenot (9”), who examined chiefly Nicolea venustala,
describes cells in the heart-body as enclosing a number of
refringent green granules. ‘‘J’ai vu,” he says, “ plusieurs
fois les cellules vertes de la glande émettre de courts pseudo-
podes et faire saillie au-dessus de leurs voisines pour étre, sans
aucun doute, entrainées plus tard par les courants sanguins; le
corps cardiaque est donc une glande lymphatique parfaitement
caracterisée.” Further, he agrees with Meyer in regarding it
as also effecting a transformation of colourless albuminoids
into hemoglobin.

Schaeppi (29), in his interesting paper on the chloragogen
of Ophelia, though his chief purpose is a study of the re-
markable chitinous rods which occupy certain of the coelomic
fluid corpuscles of that worm, yet gives in addition a careful
description of the heart-body, and also of certain peculiar
intra-vascular connective tissue which occupies the perivisceral
sinus posterior to that organ. This latter tissue begins where
the heart-body ends; and, though the author shows that it
contains different substances from those in the heart-body, the
description suggests very strongly that it is an accessory part
of the heart-body prolonged backwards from the short anterior
dorsal vessel or heart into the sinus which surrounds the gut.
The heart-body itself is a flattened band, of a bluish-white
colour, occupying the cavity and attached to the walls of the _
heart. It consists of a homogeneous groundwork strewn with
connective-tissue cells. Besides these there also occur cells
occupied with peculiar green pigment granules, similar to
others found in the thoracic sinus, but only rarely in the rest
of the blood system. There are green granules also in some

HEART-BODY, ETC., OF CERTAIN POLYCHATA. 267

of the ccelomic fluid ceils; and Schaeppi found that in both
cases they resist the action of alkalies and cold acids. He there-
fore regarded them as consisting of chitinous chloragogen.
They gave no uric acid reaction. The cells of the intra-sinus
connective tissue, however, he found to contain guanin granules,
which, he suggests, are passed through the sinus wall into the
peritoneum, the cells of which he describes as falling off in
clusters from time to time. This process would give an
opportunity for the guanin granules to be carried into the
nephridia and excreted. With regard to the function of the
heart-body itself, on the other hand, he remarks that the
central position of the granules which occur in it, and the
absence of a secretory epithelium, are objections to any theory
of glandular action. He considers that the organ, by swelling
up at systole, owing to the pressure of the blood into its
meshes, acts as a valve.!

' Some such mechanical function may very well belong to the organ, though
it seems unlikely that the blood should enter its meshes and swell it up;
but still the presence of the chloragogen granules remains to be explained.
In the first place their material is evidently drawn from the blood; but
though Schaeppi admits its excretory nature, yet he objects on several grounds
to the inference that it is taking part in a blood-cleansing process. He
leaves the question in this dilemma, that the granules are neither accumulated
in the organ, for in old individuals it is often comparatively free from them ;
nor found in its periphery, as if in the course of being removed into the
body-cavity by leucocytes. Nevertheless the bulk of the evidence, and a
comparison with other animals, seem to point to the removal of the granules,
not necessarily intact, into the body-cavity, by a process similar to that which
Schaeppi describes for the guanin chloragogen of the intra-sinus connective
tissue. That there, in the ccelomic fluid, they or their débris should be
absorbed by the lymph-cells, the green granules of which are shown to have
the same properties, appears a natural theory.

Schaeppi gives an account of the way the chloragogen granules of the small
lymph-cells fall into an alignment, fuse, and form the well-known rods which
characterise the Jarge lymph-cells. On account of their proximity to the
nucleus, however, he considers that the granules are originally derived only
from katabolic products of the nuclein; which is equivalent to saying that
the admittedly excretory function of the rods, which it must be remembered
reach such a size as to be visible to the naked eye, is limited to dealing with
the nuclear refuse of the single cells which contain them. Comparing the

268 LIONEL JAMES PICTON.

Beddard (4) supports Horst’s view that the organ is hypo-
blastic, and suggests comparisons of it with the thyroid and
spleen of Mammalia.

39

parallel case of the “ciliated pots ” of Sipunculus, which collect excretory
matter from all sides, this view is surely untenable. The granules must have
some other origin; and their source from chloragogen, such as that of the
heart-body and thoracic sinus, is at any rate not disproved.

Concerning the fate of the rods, the only evidence he adduces is that a
collection of them in the body is nowhere found, though clear signs of
degeneration are noticeable (p. 298). But since Ed. Meyer has shown, in
Terebellids, that lymph-corpuscles degenerate and are found in the lumen of
the nephridium [24, B, p. 648], and as such bodies as the rods could not be
carried through the nephridial tubes, to maintain their excretory function it
is necessary to suppose that they degenerate. Schaeppi mentions cell-heaps full
of granules, and three times the size of the rod-cells, floating in the celomic
fluid, and occurring also in the cavity of the nephridium itself. He regards
them as portions of the chloragogen-bearing peritoneum from around the
intestinal sinus, which have fallen off into the body-cavity. Some of them
probably bear this explanation ; but in cases, which he overlooks, when they
contain rods identical with those of the lymph-corpuscles, it is evident that
they are not derived from the peritoneum. In such instances they may con-
sist of a plasmodium of rod-cells, and the granules may be formed by
degeneration of the rods. Further—and this can be seen in single corpuscles—
the rods appear to undergo a peculiar method of degeneration. Often in
larger and older corpuscles the ends of the rods are notched. This notching
is such a frequent occurrence that it would appear to have some special
significance (fig. 61). Occasionally the part beyond the notch has fallen off,
and may be seen lying in the cell-body (fig. 62). This is perhaps a regular
process, and the rod having reached full size continues to grow at the ends
into projections which from time to time are detached. The projections are
often of a paler colour than the rest of the rod, which may be taken as an
indication of recent growth. Degeneration in plasmodia may be the final
stage of the rod-cells, when their life-history is complete.

Corpuscles containing Chitinous Bow-shaped Rods in Noto-
mastus profundus (fig. 60).

Hitherto the chitinous rods of Ophelia have been unique, no structures
resembling them having been found elsewhere among the Annelida. In
the hemolymph of Notomastus profundus, however, I found a cell
containing a refringent bow-shaped rod, which had a strong general resem-
blance to the rod-cells of Ophelia. It is not mentioned in the excellent
account of the hemolymph in Hisig’s monograph on the Capitellide; but
the description there deals only with the hemolymph in its fresh state,

HEART-BODY, ETC., OF CERTAIN POLYCHATA. 269

Guido Schneider (31,B) shows that in the heart-body of
Pectinaria hyperborea iron is present, and increases in
quantity after injection of “ Ferrum oxidatum saccharatum ”
into the body-cavity ; but only the preliminary notice of his
work is published, and his interesting researches may be ex-
pected to throw much light on the subject.

Monticelli (25) gives a very minute description of the heart
and heart-body of Polyophthalmus, correcting the earlier de-
scription by E. Meyer (23, A, p. 815).

Finally, in a recent paper on the Ampharetide (13), Fauvel
deals with the heart-body of that group. He describes it as a
solid organ bifurcated at the base, where it is inserted into the
dorsal face of the stomach. He regards it as fulfilling several
functions. It supports the heart in the first place, especially
when the pharynx is everted ; and it also acts asa valve. It

whilst I first noticed the bow-shaped corpuscle in a fixed and stained film.
In such preparations they are much easier to find than in fresh hemolymph,
though I subsequently found them in the fresh state. I may say that I found
at least one bow-shaped corpuscle in every specimen of hemolymph of each
individual of Notomastus profundus which I examined.

The bow is of a pale yellow colour, and highly refringent. Dr. Paul Mayer
kindly examined it with polarised light, and pointed out that it was always
of a different colour from the rest of the field. Its shape is that of a strung bow,
the tips being slightly recurved. Like the large brown granules which
occupy the hemoglobinous corpuscles, it is insoluble in cold caustic potash,
whether in 25 per cent. solution or weaker; nor was it soluble when heated
in potash solution up to 120°C. It is therefore probably of a chitinous
nature, if not actually chitin. The protoplasm completely encloses the rod,
forming, however, a very thin layer over the ends and the convex side. It is
colourless, often slightly vacuolated, and frequently exhibits long filamentous
projections, which may be seen in the fresh state making slow movements.
It is noteworthy that these projections extend from the protoplasm on the
convex side of the rod, as well as from the body of the cell, which, of course,
lies on the concave side. The nucleus lies on the concave side, and
slightly lateral to the rod. It is stained, though usually not strongly, with
hematoxylin. The rest of the protoplasm is almost unstained, except at the
ends of the rod, the cap which covers the latter being often intensely coloured.

Small fusiform corpuscles containing a clear straight rod occur also in the
hemolymph, and may be the young stages of the rod-cells. Dr. Hisig, how-
ever, was inclined to think that the rod-containing corpuscles are parasitic.

vou. 41, PART 2,—NEW SERIES, T

270 LIONEL JAMES PICTON.

removes, he thinks, waste products from the blood, and stores
them up as pigment. The observation from which he draws
this conclusion is that pigment is more abundant in the organ
in older than in younger individuals. Another function which
he suggests it may perform is the secretion of chlorocruorin,
the colouring matter of the blood. The embryology of the
group is at present unknown; but the connection which he
notes of the heart-body with the wall of the gut, leads Fauvel
to conclude that Horst’s view of the hypoblastic origin of the
organ holds good for the Ampharetide. He considers, how-
ever, that the heart-body may have a different origin in
different groups; that is to say that the heart-body is not
homologous throughout the families of Polycheta in which it
occurs.

In the Ampharetide, then, Fauvel considers the heart-body
as a gland annexed to the digestive tube, and performing a
function more or less hepatic. Contrary to Cuénot, he thinks
it has no connection with amcebocytes.

It will be evident from this recapitulation of the history of
the subject, that both on the morphology and functions of this
curious organ widely divergent views are held.

In the present paper a description of its anatomy and histo-
logy in some of the chief groups in which it occurs will be
given, with some observations on the chemical nature of the
granules, and on their ultimate fate in connection with the
coelomic corpuscles. Matters about which a good account is
already existing will be treated briefly. A description of the
organ in the Cirratulide, where it reaches its maximum deve-
lopment, will afford the best opportunities for pointing out sig-
nificant features which reappear in the different forms. Finally,
an account of its principal characters, and of those of the
coelomic fluid, in the Chlorhemidez, Terebellide, and Amphic-
tenide, and an instance of its development, will be given.

Besides in the groups just mentioned, the heart-body occurs
also in the Ampharetide, Spionide, Magelonidz, and Hermel-
lide. Its degree of development is very unequal, in Magelo-
nide the organ being merely larval and transitory, whilst in

HEART-BODY, ETC., OF CERTAIN POLYCHATA. 271

the Cirratulide the dorsal vessel, which often reaches a dia-
meter as great as that of the gut, is almost blocked by it.

The Heart-body of the Cirratulide.

The blood-system of a Cirratulid has been well figured
by Ed. Meyer (24, B),and his observations are confirmed by an
examination of young living individuals, and specimens dis-
sected and afterwards fixed and cleared. In Audouinia
filigera, as in all the group, the dorsal vessel is not confined
to the anterior end of the body, but runs back into the
abdominal region. At about the fifth segment it gives off a
pair of lateral recurrent vessels, which supply branchial vessels
in each gill-bearing segment (fig. 1). In an adult stained
specimen several vessels are also seen going off directly to the
gills, whilst the dorsal vessel itself continues anteriorly as a
much-diminished trunk. The heart-body extends forwards to
the point of origin of the lateral vessels. There it usually
ends in irregular projections which appear on the point of
becoming detached ; but it may continue as a broken cord ex-
tending for the space of a segment into the fine anterior
dorsal vessel. The three brown cords, of which it is typically
composed, are constricted at the intersegmental septa, and
branched, united, and folded irregularly (fig. 2). Sections
reveal the fact that they are connected with the heart-wall at
many points by means of fine processes. The greatest develop-
ment of the heart-body is at about one third of its length
from its anterior end. In the small transparent Cirratulus
chrysoderma it may be seen with the low power that at this
point, on systole, it almost entirely blocks the lumen of the
vessel, the action of which as a blood-propelling organ must be
greatly modified. The suggestion by Schaeppi in the case of
Ophelia, and by Steen in that of Terebellides Stroemii,
that it has a valvular function has been mentioned above.
Beddard, in his monograph on Oligocheta, refers to Michaelsen
also as saying that it ‘serves to ease the contractions of the
dorsal vessel.” Michaelsen’s view seems rightly applicable to

272 LIONEL JAMES PICTON.

the heart-body generally. The mechanical principle here in-
volved is the same as in the case of the right ventricle of
the mammalian heart, which contracts upon the convex
solid wall of the left ventricle. In the same way the dorsal
vessel of the worms in question contracts on systole, so as
practically to obliterate the lumen between its wall and the
solid heart-body, so that the whole of the blood which the
vessel contained at diastole is expelled. In the gill-vessels of
Sternaspis cellular rods are found which confer on them a
similar mechanical advantage.

Between the cords of a piece of a fresh heart-body, flattened
under a cover-slip, the minute, fusiform, non-nucleated blood-
corpuscles are often observed with a high power. Claparéde
states that they are of the same colour as the plasma in which
they float, that is, yellow. Lankester (18) says that they are
colourless. They appear to mea bright pink, unlike hemoglobin.

The histological structure of the heart-body shows points of
great interest, but has been for the most part neglected in the
literature of the subject. Three layers may be distinctly re-
cognised (fig. 5),—an endothelium clothing each cord, a cortex,
and a central or medullary tissue. The outermost layer, or
endothelium, is directly bathed by the blood. It is composed
of a single layer of cells, the nuclei of which are seen at
intervals. The cortex is composed of cells, elongated in
shape, the walls of which are very well marked (figs. 2 and 5).
The nuclei frequently have a deeply-stained nucleolus; but,
except for some refringent yellow granules, there are few cell-
contents. In the periphery of the organ the cells are closely
set together, with their long axes at right angles to its surface,
and the nuclei are towards the outer side. Beneath the peri-
pheral layer the cells are irregular and nuclei are rarer.

The development of the medullavaries in different individuals;
whilst in the same individual it is reduced in amount, or absent,
at the intersegmental constrictions. It consists of a highly
granulated tissue. In this, although at some points nuclei can be
recognised with a comparatively low power (D, Zeiss, figs. 5, 8),
yet the numbers of granules which occupy the structure render

HEART-BODY, ETC., OF CERTAIN POLYCHETA. 278

it very difficult to distinguish the cell walls. On examination
with an oil immersion, however, a framework of cells is seen to
support the granules (fig. 11), some of which are enclosed in
these medullary cells, whilst others appear to be intercellular.
The granules are greenish yellow in fresh tissue; in sections they
remain unstained with hematein, and are of a bright yellow
colour. They very frequeutly occur in heaps, constituting
larger granules (4). Occasionally a granule may be noticed
set in a vacuole (fig. 5, a). Besides these there are large
homogeneous granules, often kidney-shaped, which stain in-
tensely with Ehrlich’s hematein.}

There still remain to be mentioned the most remarkable
structures in the heart-body of this worm, and which, on
first looking at a section of the organ through a microscope, at
once arrest the attention by their striking appearance.

The medullary tissue contains numerous spherical cavities
(figs. 2 and 5), the majority of which are occupied by from one
to eight round or oat-shaped bodies. In those which are round
a “nucleus” is frequently well marked; and were it not for
their large size, which gives them a resemblance to ova, and
the fact that the “ nucleus ” is the only spot stainable with most
dyes, there would be little need for hesitation in pronouncing
them to be single cells. Those which are oval in shape are
much creased and folded; they stain irregularly with Ehrlich’s
hzematein or with fuchsin, some folds colouring intensely, others
hardly at all. Picric acid, used after hematein, stains the
blood, and also stains parts of these bodies ; but eosin, which
likewise stains the blood, does not affect them.

Some of the spherical spaces are empty, whilst others again
contain a colourless refringent, unstainable mass, dotted with
numerous dark points. The mass does not quite fill the cavity,
but the remaining space is partially occupied by a fine crumpled
membrane which appears to ensheathe the colourless bodies.
Smaller masses of the same unstainable material are seen in
some sections embedded in the groundwork of the organ.

1 T.e. Ehrlich’s hematoxylin, with the substitution of hematein for
hematoxylin, as Mayer recommends for all cases.

274 LIONEL JAMES PICTON.

The line of demarcation between the central granular or
medullary part of the organ and the external or cortical
tissue is fairly well defined, and at some points a split between
the two occurs. This is usually crossed by branching and
highly granular trabecule (fig. 5). Sometimes it is con-
siderably widened (fig. 2, y.), though in no case is there any
sign of blood within it. On the other hand, the heart-body is
frequently so folded as, in section, to show spaces surrounded
by a proper superficial cortical tissue, which contains blood
(fig. 2, #.); nevertheless it may be said that the blood never
penetrates the organ,! but only flows between its folds and
strands.

Between all the granules occurring in the heart-body a
series of steps can be traced, so that it would appear probable
that they represent different stages in some process which it
must be the chief function of the heart-body to carry on. An
examination of the shapes and properties of these granules
will help to throw some light on the nature of that process.
The methods used in this study, besides that of sections, were
teasing fresh tissue and also tissue that had been macerated in
Béla-Haller’s fluid, tests of solubility in different reagents,
and various micro-chemical tests, such as those for iron, fat,
glycogen, chitin, and uric acid.

In the following description the granules are treated in the
order in which they appear to pass step by step into one
another. But whether the history indicated be in fact correct
must be left an open question. Certain unstainable refringent
bodies have been mentioned above as occurring in the medulla
of the heart-body, embedded in the groundwork (fig. 5, e) ;
they are not conspicuous, and are apt to be overlooked, but as
they increase in size a space appears around them, which
renders them very noticeable (f). This is the origin of the
spherical spaces so characteristic of the medulla, Under an
oil-immersion objective these bodies are seen to consist of a

1 Except in a few doubtful cases, where I have noticed in sections a small
patch of what appears to be coagulated blood in a cavity in the medullary
tissue,

HEART-BODY, ETC., OF CERTAIN POLYCHATA. 275

stained spot, sometimes surrounded by a ring (fig. 8), which
may represent the nucleus, embedded in a spherical granular
mass composed of globules of a colourless material, in the
centre of each of which is a minute refringent granule, un-
stained, but naturally of a yellow or pink colour (fig. 10). In
teased preparations it is well seen that each spherical space
containing such a mass is enclosed in a definite membrane
(figs. 6 and 7). Frequently from two to eight masses occupy
a single space, in which case they are smaller than single
masses, by the division of which they are probably formed
(fig. 9). The “nuclei” are frequently well marked at this
stage (fig. 11). About this stage also the bodies begin to show
a slight general stainability with Ehrlich’s hematein. The
limits of each mass, as well as of the spaces, are defined by a
membrane, which is often developed unequally in several
masses in the same space (fig. 10); the appearance of these
membranes is very striking,—indeed, they are most conspicuous
objects in a section of the heart-body (fig. 2). They are oval
in shape, and much creased and folded along the long axis ;
the surface is often scattered with granules (fig. 12), and some
of the folds or contents of the folds stain very intensely.
Meanwhile the granular mass itself round which the membrane
was formed has disappeared. Minute holes in the membrane,
revealed, as will be seen later, by the action of caustic potash,
may be the way of exit for this matter (fig. 13) ; at any rate, it
appears to be extruded into the surrounding space (fig. 14),
and eventually gets into the groundwork, where it undergoes a
transformation which gives rise to the rich granulation that
characterises the heart-body. ‘The refringent globules, colour-
less or with a slight tinge of green, form a morula-like mass
(figs. 15 and 16), in which some of them are turning brown.
Eventually the whole mass becomes a deep greenish-brown
colour, and forms a large composite granule (fig. 17). ‘The
minute yellow points, which were mentioned above as occurring
in the centre of each globule, probably survive in the form of
a dark spot, to be seen in fresh teased tissue in the centre of
some of the larger granules (figs. 18 and 19). Besides brown,

276 LIONEL JAMES PICTON.

minute pink and green granules are found, and also large
bright red masses, that probably have some office in connection
with the blood-pigment, possibly that of reserve. One more
structure remains to be spoken of—the large, deeply stained,
homogeneous granules mentioned above (fig. 5,c). The fact that
they most frequently occur in richly granulated places suggests
that they are other products of the same activity that gave
rise to the granules. And indeed, although their mode of
origin is obscure, some cases are found in which they appear
to be in a state of formation from the deeply staining portions
of several of the crumpled, ovoid bodies. This is at any rate
possible, seeing that the chief portions of the latter require to
be accounted for, since spherical spaces are often found empty,
except for a scrap of degenerating membrane.! The homo-
geneous bodies occur in considerable numbers, and occasionally
one is found in the cortex of the organ.

The reactions of the brown granules serve to differentiate
them from the sorts of chloragogen described by Schaeppi.
In the following tests for solubility the reagents were applied
directly, and not drawn under a cover-slip. Absolute alcohol
had but little effect on the granules, except to slightly contract
them. After the addition of a saturated solution of oxalic
acid to fresh teased tissue, in three or four minutes those
granules which naturally contain a central dark spot sur-
rounded by a lighter portion had become homogeneous and
somewhat reduced in size. There was no further change.
When ether is poured over a hand-section of material pre-
served in alchol, as evaporation goes on some large yellow
drops appear, a pigmented fat having been dissolved out of the
tissue. Strong acetic acid, whether cold or boiling, has no
effect. ‘The same is true in the case of hydrochloric acid
diluted to 50 per cent. Strong hydrochloric acid, on the other
hand, dissolves the brown granules in the cold when allowed to
stand for some time; boiled on a slide or in a test-tube the
granules are at once dissolved by it. Sulphuric acid gives a

1 On the other hand, it is possible that the empty spaces have lost their
contents during manipulation.

HEART-BODY, ETC., OF CERTAIN POLYCHHTA. 277

brown solution on boiling. Nitric acid completely dissolves the
granules after standing in the cold, and at once on boiling ;
the solution on evaporation leaves a yellow residue, which
gives the xantho-protein reaction, turning yellow on addition of
ammonium hydrate; but in many trials I never saw the purple
coloration of the murexide reaction.

On standing in the cold in caustic soda the brown pigment
granules are dissolved, giving a fluid of a brownish-yellow
colour. On boiling a piece of fresh heart-body with caustic
potash, the pigment having been dissolved, some masses of
insoluble material remain floating about in the liquid. Con-
tinued boiling has no further effect on them ; and on removing
them to a slide and examining them microscopically they are
shown to be composed of some irregular and granular débris,
but chiefly of clusters of bodies of an irregularly oval shape,
considerably creased and folded, translucent, in places colour-
less, but at some points brown, especially where they are
creased (fig. 13). Sometimes brown dots or circular markings
are scattered over their surface, whilst occasionally there is a
clear appearance of small tubular perforations of the walls
resembling the transverse striations of some egg membranes.
These bodies are evidently the crumpled oval envelopes so
conspicuous in the spherical spaces of the heart-body ; but the
appearance of perforations is a new feature in them, revealed
by the action of the boiling alkali. It has already been sug-
gested that the openings thus shown may be the ways by which
the granular contents of the bodies are extruded. The re-
markable resistance of these bodies towards boiling alkalies
at once refers their material to a group of organic substances
of a chitinous nature.

I asked Prof. Paul Mayer whether any further tests could
be applied to solve the nature of this substance, which he was
so good as to examine, and he pointed out that its effect on
polarised light corresponds with that of chitin. He referred
me to Ambronn’s paper on the cellulose reaction of chitin (2),
and was so good as to perform the test there recommended
himself. I collected the insoluble remnants of half a dozen

278 LIONEL JAMES PICTON.

heart-bodies after boiling in caustic alkali solution, and by his
direction bleached a portion by his chlorate of potash and
hydrochloric acid method. The acid was introduced by a
pipette at the bottom of a tube of alcohol, so that the proble-
matical bodies, which were suspended in the tube in a piece of
silk gauze, could be but little affected by it. Dr. Mayer
tested both bleached and unbleached material, the unpig-
mented portions of the latter being quite sufficient to indicate
any colour reaction. The substance was first washed in dis-
tilled water, and then a drop of the double salt, chloride and
iodide of zinc, was put upon it. Some of the bodies turned
deep indigo-blue, others took a bluish tinge, whilst others
again did not show the least stain. In a cluster of three, two
might be markedly coloured, and the third not at all. With
the double salt of calcium, in the case of unbleached material,
some of the bodies turned light violet, others yellow. These
reactions show that the nature of the substance is closely
similar to that of chitin, but in some of the bodies it is in
a different state from that in others, since the stain does not
occur uniformly. This fact helps to confirm the view that
these structures, which may now be referred to as “ chitinous
bodies,” are in process of formation in the heart-body. The
occurrence of the chitin-cellulose reaction in Annelida is in-
teresting, since Ambronn, who used Spirographis, did not find
it in this group.

Osmic acid darkens many of the granules, but in a pig-
mented structure such as the heart-body it shows nothing
conclusively. Ranvier gives as a reaction of fat a blue colora-
tion with quinoline blue, and its preservation in glycerine
[26], and Rosa confirms his statement [27]. When a piece of
a heart-body is teased, treated on the slide with alcohol, and
then with an alcoholic solution of quinoline blue diluted with
water as far as is possible without precipitation of the colouring
matter, the granules are stained greenish blue, and also a
number of globules are rendered intensely blue. After standing
in glycerine under a cover-slip for three weeks the blue of the
globules has not faded, but the colour of the granules has

HEART-BODY, ETC., OF CERTAIN POLYCHMTA. 279

become green. It appears, then, that a number of fat globules
exist side by side with the granules—a state of things which
accords with the effect of ether, mentioned above, of dissolving
out fat from the heart-body. This view is clearly confirmed
by the use of Soudan III. The reaction of this aniline colour
on fat was recently discovered by Daddi [11]. Professor Mayer
informed me that he had examined it, and found it appeared
to be a reliable stain for all fats, rendering them bright
orange. In a hand-section of material fixed in alcohol, and
stained on the slide with a solution of Soudan in 70 per cent.
alcohol, and then teased in glycerine into fragments, numbers
of fat droplets stained a bright orange colour are seen distri-
buted in the cells. They are especially abundant in the outer
layers. In the heart-bodies of Audouinia that had been about
a month in the aquarium tank, fat had entirely disappeared.

The finding by Guido Schneider of iron naturally occurring
in the heart-body of Pectinaria has been mentioned above.
I found that.in the case of Audouinia, hand-sections of
material fixed in alcohol, when laid in ferrocyanide of
potash solution, acidulated with hydrochloric acid, showed a
blue coloration of the gut wall at once ; but in the heart-body
when teased only a few insignificant fragments were bright
blue. MacAllum’s ammonium-hydrogen sulphide method !
for micro-chemical detection of iron [21], although it of course
shows inorganic and albuminate compounds of iron, is espe-
cially designed to reveal iron in higher organic compounds,
such as nucleins, which exist, as well as in nuclei, also in
secreting cells, yolk of egg, heematoblasts, and other such bodies.
On jaying a hand-section of the heart-body in a drop of 50 per
cent. glycerine thoroughly mixed with two drops of ammonium-
hydrogen sulphide, there was no immediate reaction except a
slight general greenish tinge. The tissue was then teased by
means of drawn-out glass rods with broken points, instead of
needles, and, covered with a cover-slip, was left in a water-bath

1 Since this work was done MacAllum has published a new and easier
method for the micro-chemical detection of iron, in the ‘Journal of Physio-
logy,’ September, 1897.

280 LIONEL JAMES PICTON.

at 60° C. forten hours. It was then found to be coloured green,
and numerous black dots were scattered about it.! On exa-
mination with a ;4;-inch objective these dots were seen to be
intensely stained granules, a crescent-shaped portion round
the edge of each of which was of a deeper colour than the rest.
Besides these, numbers of clearly defined and minute granules
were stained, especially in the outer cells of a strand of the
tissue. After 3 per cent. nitric-acid-alcohol acting for some
hours at 35° C. the results were the same, except that the
crescentic marking of the larger granules was not shown. In
tissue subjected to the action of Bunge’s fluid for thirty hours
no trace of the reaction takes place. On the criteria laid
down by MacAllum, the facts that iron is not, or only slightly,
revealed by the ferrocyanide reaction, but yet that it is shown
by the ammonium-sulphide method, whilst it is easily removed
by Bunge’s fluid, go to indicate that it is in organic (*‘ mas-
kirt ”), but not in very elaborate combination. The presence
of iron-containing bodies and also of fat globules in especial
numbers in the peripheral cells is noticeable. It has been
mentioned that the fat dissolved from the heart-body by ether
was pigmented, and it may be here suggested that possibly this
pigment contains the iron. It is likely that the iron-holding
bodies are connected with the formation of the blood-pigment.

Barfuth’s method of demonstrating glycogen microchemi-
cally (3) was applied to the heart-body, his mixture of glycerine
and iodine dissolved in potassium iodide solution being used.
Material fresh from the sea was chosen, and also worms placed
in an aérated mixture of starch and sea water. The worms
took up the starch, but neither in this nor in any case was a
clear red-brown colour obtained with the reagent, sufficient to
prove the presence of glycogen.

The indications of all these reactions point to the following
conclusions as to the nature of the contents of the heart-body :
that the brown granules, being soluble in caustic alkali solu-

1 The nuclei did not markedly show the iron reaction. No doubt iron is

present in them, but for its demonstration, according to MacAllum’s recipe,
the cells would have to be completely isolated from one another.

HEART-BODY, ETC., OF CERTAIN POLYCHATA. 281

tion, are not chitin; that they are not guanin, since they give
no murexide reaction (in these points they differ from both
kinds of chloragogen described by Schaeppi in Ophelia); that
the ovoid crumpled structures are chitinous bodies ; that fat
is distributed in globules in the heart-body, especially in its
periphery; that iron is distributed in larger and smaller
granules, also especially in the periphery; and finally, that
glycogen is probably absent.

Before leaving the Cirratulide the relation of the heart-
body to its surroundings suggests several matters for con-
sideration. On passing out of the dorsal vessel the great bulk
of the blood must first go to the gills; so that if the heart-
body in any way modify the constitution of the blood, that
change has probably some relation with the respiratory organs.
In connection with this point it is interesting to note the
pigment lymph-glands described by Ed. Meyer (24, B, p. 701)
ou the blood-vessels, particularly the arteries, of the gills of
Cirratulide. They consist, he says, of a single glandular cell-
layer belonging to the peritoneal wall of the respiratory
vessels, and contain brown pigment.! He also mentions them
in Terebellide (p. 645) and elsewhere. His view with regard
to them is that they take some fluid ingredient out of the
blood, and cast it into the ccelomic fluid. This theory,
however, does not account for the situation of the organs in
the gills. Is it to be inferred that the excretory matter is cast
to the exterior through the gill-wall?? At any rate, gill

1 Tron-holding patches, scattered at some points in the epidermis of the
gills, are demonstrated both by the Prussian blue and the ammonium sulphide
reactions; the iron is removed by Bunge’s fluid at 60°C. in an hour. No
iron reaction was noticed in the pigment lymph-glands.

* Gill glands for excretory purposes have been demonstrated by Allen (1)
in Crustacea; whilst amongst Annelida yellow concretions regarded as cer-
tainly excretory have been found by Miss Buchanan (8, C) in the gills of a
branchiate Polynoid. She finds a similar phenomenon in several other worms.
In sections of the posterior gills of Sternaspis I find numbers of small
refringent bodies crowding the epidermal cells. It may be suggested that
the important functions of the dorsal pores of earthworms, of removing waste
products, may be represented in branchiate worms by an excretory activity of
the gills.

282 LIONEL JAMES PICTON,

excretion is a possible mode of removal for waste products
thrown into the blood by the heart-body, chloragogen, or other
organs connected with the vascular system.

A remark of Schaeppi’s has already been mentioned, that
the intra-sinus guanin chloragogen of Ophelia is excreted by
being carried through the sinus wall into the peritoneal tissue,
clusters of the cells of which eventually become detached and
fall into the ccelom, and are removed by the nephridia. Quite
independently of this, anyone glancing at sections of Audou-
inia filigera would suspect some connection between the
heart-body and peritoneum. The latter is remarkably thick-
ened on the mesenteries which cross the coelom from the gut
and dorsal vessel to the body-wall (fig. 2); its cells are well
developed ; their protoplasm is concentrated in their central
part, and crowded with yellow or green granulations, and,
what is here to the purpose, this tissue is often found lying on
the wall of the heart, the inner side of which in many places
is connected with the heart-body (fig. 20). Thus it is often
found that the peritoneum and heart-body are only separated
by the thin muscular wall of the heart ; and that a transfer of
granules takes place from one tissue to the other is not im-
probable. The granules of the peritoneum, it is true, are not
so large as the larger in the heart-body ; but it is reasonable to
suppose that the small and disintegrated granules would be
the most suitable for excretion. Granular heaps are found
floating in the ccelomic fluid, and I have especially noticed
them in the anterior part of the animal,—that is to say, in the
neighbourhood of the great anterior pair of nephridia. The
coelomic corpuscles also, which are large and ameeboid, fre-
quently contain brown granules, though many are occupied by
globules of a reddish-brown pigment! (figs. 21 and 22). These
different places where brown granules are found*—the heart-

1 Tn the season when I examined the ccelomic fluid of Audouinia it was so
filled with genital products that the ccelomic corpuscles were present in
relatively very small numbers.

2 Numerous small, homogeneous, spherical, dull green granules are found
also in the gut wall. But they are always on the inner side of the nuclei of

HEART-BODY, ETC., OF CERTAIN POLYCHATA. 283

body, the peritoneum, the detached cell-heaps, the ameeboid
corpuscles, and finally the great anterior nephridia—trace out a
path by which they may be removed from the blood to the
body-cavity, and thence to the exterior.

Characteristics of the Heart-body in other Poly-
cheta.—The organ in the Chlorhemide, Terebellide, and
Amphictenidz has much in common with that in the Cirra-
tulidz, but each group shows peculiarities. Of the two other
groups which are included in Dr. Benham’s classification,
under the head of Terebelliformia, Sternaspis has no heart-
body proper, and the Ampharetide are rare at Naples.

Chlorhemide.—The dorsal vessel is, unlike that of the
Cirratulide, confined to the anterior end of the worm. It
arises at the opposite side of the digestive tube to the point of
attachment of the stomach, and the blood from the sinus
which surrounds the digestive organs flows into it. It is a
broad trunk, lying dorsal to the cesophagus, and after its point
of origin quite separate from the alimentary canal. It narrows
towards its anterior end, and eventually divides into two
branchial vessels. In Siphonostoma, Cunningham (10) aptly
describes the heart-body which occupies this vessel as a folded
band, which in section branches dorsally ; he also states that it
contains no lumen, though the cells of the opposite sides are
in a definite line of contact. Jourdan (17), writing the same
year, 1887, stated, on the contrary, that there is a lumen, and
gave a figure showing it.

Now, on examining a series of transverse sections, it is
clearly seen that the “ band” is formed of a cylindrical tube,
the walls of which have been compressed together, and that
though the single cell-layers which form the walls! are in

the cells that contain them, that is the side towards the gut, and are probably
of a different nature from that of those discussed.

1 If an endothelium covering the heart-body exist, it must be extremely
attenuated. One or two nuclei on the surface of the organ seem to suggest
its presence. Bles states that the gut wall is directly bathed in blood, but
there, too, it is not impossible that an extremely attenuated membrane is
interposed. All the structures of Siphonostoma are so thin, the solid jelly

284 LIONEL JAMES PICTON.

places in contact, yet a considerable space is generally left
between them (fig. 23). This lumen appears so constantly
that I think it is natural, and not due to fixation and handling,
nor has it the appearance of an artefact. Moreover it is pos-
sible in a fresh specimen to snip the heart longitudinally with
a pair of fine scissors, so as partially to slit open the heart-
body, so that on microscopic examination a single thickness of
the wall is viewed. If this be really the case, as I suppose,
then the lumen is probably greater in life than it appears in
sections. There is another even more important discrepancy
between the descriptions of different naturalists. Whilst
Jourdan speaks of the lumen as “ communiquant avec le tube
digestif,” Bles (5), on the contrary, states that it has no con-
nection, not even contact, with the gut. As a matter of fact,
a series of sections clearly shows that in its posterior part it is
attached to the gut, and that this connection extends for a con-
siderable distance; on the other hand, no open duct or pas-
sage from the one into the other is seen to pierce the gut-wall.
The attachments to the heart-wall are mentioned by Bles, who
considers that they point to the peritoneal origin of the organ.
These occur irregularly, and consist of one or two finely drawn-
out cells, much resembling the muscular cells of the heart-
wall, to which they are fastened on the one side, whilst on the
other they are inserted into projections of the heart-body,
which they have pulled out (Fig. 34).

The cells, which are uniform throughout the organ, are
columnar, though when found isolated in teased preparations
of fresh tissue they have assumed a spherical shape. It may
be inferred from this that their connection and packing to-
gether not only induces, but maintains their cubical form.
With regard to their granulations, Bles remarks that in
hardened specimens they are crowded with green granules,
which are also found in clotted blood. The natural appearance
of the granules, however, is very striking. There are three
kinds in each cell (fig. 25). First there is an irregular homo-

in which the animal is enwrapped rendering strength unnecessary, that the
usual morphological layers are very difficult to recognise.

HFART-BODY, ETC., OF CERTAIN POLYCHATA. 285

geneous granule of a clear red colour, about a quarter of the
diameter of the cell; then there are usually two emerald-green
globules, each of which is somewhat less than half the size of
the red granule; and finally, there are numbers of still smaller
granulations, deep brown in colour, which are usually con-
centrated at the poles of the cell, as seen in isolated cells of
macerated and teased tissue (fig. 26). The nuclei are large,
and are placed towards the inside or lumen of the organ,
whilst the large granulations are towards the outside (fig. 28).

The green globules look like oil droplets, but, like the red
and brown granules, are insoluble in ether. They blacken
somewhat with osmic acid ; and with acetic both they and the
red granules lose their colour. With cold 50 per cent. hydro-
chloric acid the red pigment is dissolved and stains the cells,
and the brown granules at a certain focus look purple. The
latter are not dissolved even on boiling with the dilute nor on
standing in the strong acid in the cold, though they disappear
on heating. The residue of a solution in nitric acid fails to
give the murexide reaction. Cold caustic soda solution re-
moves the red and green bodies, but only causes the brown to
swell and turn a paler shade. On boiling, however, they too
dissolve, but a small residue of insoluble bodies remains, some-
what resembling the chitinous bodies of Audouinia.! In
some of these, which had been boiled on the slide with alco-
holic potash, the double salt of chloride and iodide of lime
produced a yellow, and in some cases a violet tinge (Dr. Mayer).

Fat was not found in the heart-body.

Tron is shown by the Prussian blue reaction in some of the
granules. The ammonium sulphide test occurs only on
heating, but after six hours at a temperature of 60° C. tissue
treated with this reagent is dotted over with clearly defined
greenish-black patches, which I believe to represent the red
granulations. After the action of Bunge’s fluid for an hour at
60° C., followed by nitric-alcohol, ammonium sulphide fails to
produce any coloration, so that the iron must be in such

1 These are not noticeable in sections, but I have found pale-staining bodies
in the lumen which may represent them.

voL. 41, PART 2.—NEW SERIES. U

286 LIONEL JAMES PICTON.

combination as to be easily removable by hydrochloric acid.
In tissue treated for ten hours with nitric alcohol the granu-
lations are very markedly stained by either method.

In Stylaroides hirsutus the heart-body folds have
become rounded or flattened strings, which may run for con-
siderable distances independently of one another. The cells in
sections of 4 are clearly defined, and usually occupy the
whole thickness of a strand, though occasionally a lumen is
left. Except for brown granules, some of which are large
and contain a dark central spot, there are few cell-contents.

Cunningham describes the cords in Trophonia plumosa
as hollow, and formed of three layers of cells, the most
internal of which is composed of spherical elements which
project irregularly into the lumen as in a section of a nephri-
dium, which the whole structure much resembles.

Horst studied the heart-body in Brada (16). He is struck
with the lack of demarcation between the cells: “ Bei einem
jungen Examplare von Brada villosa war an der Peripherie
der Strange die Zellgrenze ziemlich deutlich, der centrale Theil
aber wurde gebildet von einer mit braunen Kérnchen gefiillten
Grundsubstanz, worin keine deutlichen Zellen nachzuweisen
waren. Bei den erwachsenen Individuen zeigen die Strange
auf dem Querschnitt nur ein unregelmassiges Netz von
Fasern, in dessen Knotenpuncten deutliche Kerne liegen,
wahrend in der durchsichtigen Grundsubstanz der Maschen
die braunen Kornchen zerstreut sind.” If “ Fasern” means
cellular fibres, such an arrangement would be similar to that
above described in the medulla of Audouinia ; but, on the other
hand, if, as is more probable, the so-called fibres are only cell-
walls seen in section, the arrangement agrees with that seen
in Stylaroides, which is characteristic of the Chlorhzemide.
Haswell, however, who studied Australian Chlorhzemide (15),

1 In fresh heart-bodies a considerable amount of débris and a number of
crystals are noticeable. The forms of the crystals were not characteristic of
any special substance. Certain of them, hexagonal in shape, stained indigo
blue with iodine dissolved in potassium iodide. Dr. Mayer pointed out to
me that they have no effect on polarised light.

HEART-BODY, ETC., OF CERTAIN POLYCHATA. 287

states that in Coppingeria the cellular structure is only repre-
sented by nuclei.

In Siphonostoma, Stylaroides, and Trophonia there is no
such development of the coelomic epithelium as that in Cir-
ratulide. If excretory products be passed through the
heart-wall, as suggested in the case of that group, they would
be received directly by the coelomic fluid or its cellular ele-
ments. The latter consist in Siphonostoma of colourless cells,
containing large nuclei, in some cases with very little sur-
rounding protoplasm, in others, however, with a considerable
cell body of a pointed oval shape (fig. 37), staining strongly with
eosin, usually somewhat vacuolated, and containing one or two
small dark spherules. In Trophonia, besides small vacuolated
fusiform cells (fig. 40), which are often drawn out to a fine
point at one end, as in the corpuscles described by Good-
rich in Enchytreus hortensis (14, p. 56), where the
processes in an early stage of the corpuscles attached them to
the walls of the celom ; there are also somewhat larger cells
(fig. 41) with nuclei of the same size, but with their protoplasm,
in specimens rapidly fixed, entirely drawn out into broad
flattened projections which radiate from the nucleus like the
petals of a flower. In both kinds of corpuscles, when exa-
mined fresh, vacuoles and yellow granules are observed, the
latter being sometimes as large as the chitinous granules in
the ordinary corpuscles of Capitellide, and with a similar con-
centric marking.!

1 PROBLEMATICAL BoDIES CONSTANTLY PRESENT IN THE C@LOoMIC Fiuip

or SIPHONOSTOMA AND TROPHONIA.

Jourdan (17, p. 33) describes in Siphonostoma crystalline calculi con-
tained in the mesenteries, especially that of the stomach. I find similar
structures, with their cellular envelope around them, constantly present,
floating freely in the ccelom (figs. 38 and 39). The calculi are usually spherical,
and tend to split into segments radially. ‘They are pale green in colour.

In Trophonia bodies occur in the ccelomic fluid which are possibly of a
similar nature. They are colourless rods, usually shaped somewhat in the
form of a figure of 8, and are enclosed in an envelope formed of a few cells,

in the substance of which are some globules of fat (fig. 43). In the centre of
the rod is an elongated dark streak. The whole structure has a striking

288 LIONEL JAMES PICTON.

A very striking fact has been noticed by Bles in Siphono-
stoma, which, so far as I have seen, holds good in other
Chlorhemide: the great single anterior pair of nephridia is
totally deprived of blood-vessels. Their near proximity to the
heart-body, the occurrence of the similar granules in both, and
in the ceelomic corpuscles, coupled with this singular fact, are
very strong arguments in favour of the way of excretion of
waste blood-substances proposed. On the other hand, pigment
lymph-glands similar to those mentioned by Ed. Meyer in Capi-
tellidae and Terebellide are described by Haswell in the gills
of Stylaroides cinctus, accompanying the branchial vessels,
and containing a reddish-brown pigment; and these offer an
alternative possibility as a way of excretion from the blood.

Terebellide.

The arrangement of the blood-vessel is similar to that in the
Chlorhemide, and the branchial heart, formed by the union
above the gut of a pair of peri-intestinal sinuses, runs forward
freely suspended in the body-cavity by a dorsal mesentery, and
ends, after piercing the dissepiment, by dividing into two
branches which carry blood to the three pairs of gills (fig. 3).
In some species a small median vessel proceeds forwards from
the point of division ; but in Polymnia this is represented by a
median vessel which springs from the base of the heart-body,
and runs forward between that and the gullet.1 Thus in
Polymnia all the blood which flows over the heart-body goes to
the gills. The heart-body itself is in the form of a cylindrical
rod, extending from the posterior end of the heart to a point a
short distance beyond the dissepiment. It is attached to the
heart wall by fine processes, both at its extremities and some-
times at other points. Amongst the many Polymnia which I
examined, a single specimen showed a peculiar abnormality of
the organ. Instead of ending at the posterior extremity of the
resemblance to bodies described by Metchnikoff in the spleen of the Jerboa
(22, pl. iii). In that case the figure-of-eight shaped bodies represent reactions
against bacteria.

1 Milne-Edwards’ otherwise excellent description of the blood-system is
wrong on this point (Ann. Sci, Nat.,’ 2e sér., t. x, p. 199).

HEART-BODY, ETC., OF CERTAIN POLYCHATA. 289

heart, the heart-body extended in two thick cords into the
peri-intestinal sinuses. In Terebellides Stroemii the
organ is hollow; but in most Terebellide it is solid. In
Lanice conchilega it is in the form of a number of cords
similar to those of Stylaroides.

The colour of the heart-body in the Terebellide is deep
brown, and varies in intensity with the general depth of pigmen-
tation in the rest of the animal. Oval masses of coarser pig-
ment are set in it here and there transversely to its length.
The organ is enclosed in a fine endothelium, as in the case of
Audouinia. Cortex and medulla are not distinguishable; but
the elongated cells of which the structure is composed form a
sort of network, radiating out from the central axis. There is
no great lumen; but numerous intercellular spaces appear. The
pigment granules are of a greenish-yellow colour, and each is
usually surrounded by a vacuole (figs. 27 and 32). Each of the
oval patches of granules mentioned above is contained in an en-
velope, and perhaps represents a much-distended cell (fig. 28).

In sections of Polymnia nebulosa numerous blood-cor-
puscles are to be seen in the coagulated blood in the vessels.
They appear to be non-nucleated ; but each contains a minute
granule or crystal.

The ccelomic corpuscles of Terebellide may be classified into
three kinds,—ameeboid, fusiform, and eleocytes. To what extent
these are distinct in origin I do not know ; but, both in young
and mature specimens, they may be clearly distinguished from
one another. All three forms may contain brown pigment
granules. In Polymnia the fusiform corpuscles! average about
25 w in length, whilst the amceboid are of somewhat smaller
diameter (figs. 57 and 54). One end of the cell-body of the
former is sometimes expanded so as to contain a circular body,
deeply staining with eosin, in which is placed a granule
staining blue with methyl green (fig. 55).

1 In Amphitrite the fusiform corpuscles are sometimes drawn out to a
considerable length (fig. 44), recalling the statement of Lim Boon Keng (19)
that the spindle-shaped cells of Lumbricus may project their ends for great
distances.

290 LIONEL JAMES PICTON.

The third kind of corpuscle appears in relatively small
numbers in Polymnia, but is very abundant in Amphitrite
(A. rubra and A. variabilis). On making a small incision
in the body-wall of a fresh specimen of this worm, a thick
pink liquid comes out. A drop of this examined under the
microscope is seen to contain numbers of corpuscles of a
striking size and appearance. About 35 m in their greatest
diameter, they are of a flattened oval form (fig. 50), and are
filled with pale yellowish spheres which resemble oil droplets,
and evidently are the cause of the pink colour of the coelomic
fluid in bulk. These droplets are undoubtedly fat. In a
cover-slip preparation of coelomic fluid, fixed in absolute
alcohol, and stained with Soudan III dissolved in 70 per cent.
alcohol, washed with dilute alcohol, and mounted in glycerine,
the globules are stained a bright orange-red (fig. 52). If,
instead of washing with alcohol, the preparation be washed
with ether, the stained fat is removed from the corpuscles,
leaving the spherical cavities which contained it empty.
There is a large nucleus, somewhat eccentrically placed, and
visible only on staining,! whilst the groundwork of the cell is
occupied in the fresh condition with minute greenish granules
which do not stain with Soudan III, and may be compared with
the discrete granules described by Lim Boon Keng in Lum-
bricus. Strongly defined granules of yellow pigment also occur,
especially in smaller eleocytes, with little fat (fig. 51); and,
finally, the cell is enclosed in a distinct cell membrane.
These large fat-containing corpuscles may be compared with
the eleocytes of Oligochzeta, in which Rosa (27) states, on the
evidence of Ranvier’s fat test, that the granules consist of fat.*

1 In preparations treated with Ehrlich’s hematein, stained curved lines were
often seen radiating irregularly from the central point in the eleocytes of
Amphitrite. I supposed at first that at this point there was a centrosphere,
the whole arrangement being similar to that described by Rosa (27) in the
eleocytes of Allobophora. In preparations treated with iron hematoxylin,
however, no centrosphere is brought out, so that the appearance remains
unexplained.

2 T have confirmed this observation by the use of Soudan III, on the eleo-
cytes of an Allobophora, a common worm in the Villa Reale, Naples.

HEART-BODY, ETO., OF CERTAIN POLYCHATA. 291

With regard to the relation of the heart-body to the coelomic
fluid, the position of the dissepiment sacs must be noted.
These are diverticula of the dissepiment, or great anterior
septum, which extends transversely across the body-cavity
behind the nephrostomes of the most posterior pair of the
great anterior nephridia. The heart-body runs in the heart
to a point a little beyond where the latter pierces the dissepi-
ment. The sacs consequently occupy a position on each side
of the heart—a fact which is especially noteworthy in consider-
ing the connection of the heart-body with leucocytes, since the
dissepiment sacs are generally crowded with corpuscles.’

Some of the species of Polycirrus are remarkable among the
Terebellide in lacking blood-vessels, and consequently also
gills and heart-body. It is, therefore, not without interest to
observe an accumulation of coelomic corpuscles upon a mem-
brane supporting the gut in the anterior dorsal region. An
area for the origin of the coloured corpuscles may survive at
this point.

These coloured corpuscles, as is well known, occur mingled
with the ordinary coelomic corpuscles, the resulting coelomic
fluid being termed hemolymph (Kisig and Meyer). Meyer
states (24, B. p. 645, note i), “ Abgesehen von den eigentlichen
Lymphkorperchen, noch andere auch freie Zellen in grosser
Menge in der Leibesfliissigkeit umherschwimmen, welche
durch und durch von einer Art Blutpigment dirchtraukt
sind und somit die Rolle von gefarbten Blutkérperchen zu
iibernehmen scheinen.”” The ordinary corpuscles of which he
speaks are ameeboid, and contain considerable masses of amber-
coloured pigment insoluble in ether (fig. 29). The ‘ blood-
corpuscles” are elongated and oval, or sometimes very small
and round, and are pale yellow in colour, whilst in vacuoles
placed, in the case of the elongated form, at the extremities of
the cell, there are often small droplets of a madder-pink colour
(figs. 80, 381, 57, 58). Cuénot {9”, p. 414) speaks of these

! It is remarkable in young Polymnia that whilst numbers of eleocytes occur
behind the dissepiment, they are almost entirely absent from the part of the
ccelom anterior to it.

292 LIONEL JAMES PICTON.

droplets as colourless. They seem to me to bear the same
colour relation to the yellow colour of the cell-body as the
blood-corpuscles of Audouinia bear to the colouring matter
dissolved in the plasma of the latter worm.

Development of the Heart-body in Polymnia nebu-
losa.—The work of Salensky, who described the primitive
cavity in the heart-body of Terebella Meckelii, and the
communication between this and the ccelom ; and also Hisig’s
conjecture based on that observation, that the organ is due to
an infolding of the heart wall, have already been mentioned.

Although much discussion has taken place on the question
as to whether the heart-body be a hypoblastic or a mesoblastic
organ, no further embryological work has been published on
the point.

In Polymnia nebulosa! the matter can be settled by an
examination of transverse and longitudinal sections of the
heart-body shortly after its first appearance in the larva.

As in Terebella Meckelii, the blood-system is earliest
represented by a peri-intestinal blood-sinus; but whereas in
that species the dorsal and ventral vessels made their
appearance only in a larva of twenty segments, in Polymnia,
at a stage when nine pairs of chet were counted, a ventral
vessel was clearly defined ; whilst in a larva of thirteen pairs
of cheeta a dorsal vessel also had appeared in the anterior part
of the body. But as yet there was no heart-body. There were,
however, cells in the vessel walls, which were swollen, and con-
tained collections of greenish-yellow pigment (fig. 33). The
blood-pigment which gradually makes its appearance is pro-
bably derived from this.

The exact moment at which the heart-body forms is difficult
to determine. In a larva about 15 cm. in length it has

1 The eggs of Polymnia are obtainable in great numbers at Naples, and
develop well in captivity. Also large numbers of the larve are found amongst
the alge from certain beds. The tube is very early secreted, and is removed
with great difficulty. Cav. Dr. Lo Bianco, however, told me that, on
allowing them to stand for a few hours in stagnant sea water, the larvee came

out of their own accord. By this means they may be obtained living and
without their tube, with great ease.

HBART-BODY, HTC., OF CERTAIN POLYCHETA. 298

appeared as a cluster of cells with large nuclei in the dorsal
vessel. From the first it shows signs of pigmentation. Even in
the living state a cavity can be recognised in it, whilst sections
show that part at least of this cavity opens directly into the
celom on the ventral side of the heart just anterior to its origin.
In other words, the heart-body is an in-pushing of the heart-
wall (fig. 36). It shows no connection whatsoever with the
hypoblast. Later (fig. 35) the open connection with the ceelom
appears to be narrowed, and finally obliterated.

Sections through the organ at the stage when this is going
on are difficult to interpret except in the light of the earlier
stage. Nevertheless the heart wall may be traced on the
ventral side to a point where it is invaginated, and becomes
continuous with the wall of the heart-body (w.). It is not clear
whether the lumen which primitively exists at the posterior
end of the heart-body, and which at first is open to—in fact, is
a part of—the ceelom, is continuous with that in the anterior
part of the organ. It is not unlikely that, an invagination
of the heart-wall having taken place, proliferation of cells
from the anterior end of the invagination gives rise to the
anterior part of the organ, and that in this a secondary cavity
is formed.

However that may be, the main fact is evident that the
heart-body in Polymnia is a purely mesoblastic structure.

The Heart-body in the Amphictenide.

In Pectinaria, Claparéde (9, A, p. 879) states, “Le gros
vaisseau intestinal est accompagné dans toute sa longeur par un
cordon jaune, irrégulier, adhérant a sa paroi, et formé de cel-
lules larges de 11 micr., et pourvues d’un nucléus clair. Cet
organe est sans doute comparable aux boyaux celluleux qui
adhérent au vaisseau dorsal chez tant d’Annélides. J’insiste
sur ce fait, parceque ce vaisseau dorsal de l’intestin joue chez les
Pectinaires le role d’un vaisseau dorsal véritable.”

Wirén (83), in his figure of the blood-system of Pectinaria,
shows that this great intestinal vessel gives off the gill vessels.

294. LIONEL JAMES PICTON.

Thus in position in the circulation, as well as in appearance,
this irregular cord (fig. 4) corresponds to a heart-body. It
contains numerous yellowish pigment granules.

Summary and Conclusions.

It is undesirable to attempt as yet to form a theory of the
action of the heart-body accounting for the details of its
structure. The facts relate themselves to one another, and
suggest resemblances with other animals; but the part the
organ plays in the economy cannot be defined: still there are
several indications of the broad outlines of its nature.

First of all, with regard to its homologies, its mesoblastic
origin, shown at least in one instance, Polymnia, must be
considered. The organ is not homologous with the diver-
ticulum of the gut which projects into the dorsal vessel of
Buchholzia and other Oligocheta, and the theories of Horst
and Beddard to this effect fall to the ground. Nevertheless,
considering the fact that in those forms the diverticulum of
the gut, as it pushes into the heart, must carry the heart wall
with its coelomic epithelial covering in front of it, it is evident
that the latter must form a part of the so-called “ heart-body.”
It is, perhaps, this mesoblastic part from which the anterior
part of the heart-body of Buchholzia is divided, and were this
the case there would be a partial homology of the organ with
that of Polycheta. Further, the heart-body described by
Nusbaum and Rakowski in Fridrica (25’) appears to have no
connection whatsoever with the gut, and accordingly re-
sembles that of Polychta in being entirely mesoblastic.

The mesoblastic origin being established, the organ may be
regarded, as Hisig suggests, as of the nature of intra-vascular
chloragogen,-—that is, as modified peritoneal tissue, primitively
clothing the outside of the dorsal vessel, but becoming folded
so as to lie withinit. The word “ chloragogen,”’ however, must
be understood only as a descriptive term, Schaeppi having
shown that the granules occurring in the tissues included
under that designation are of different chemical natures,

HEART-BODY, ETC., OF CERTAIN POLYCHATA. 295

The function of chloragogen is a much-debated question.
Certain text-books speak of the tissue frankly as “ liver cells.”
Schneider sums up the facts which point to such a conclusion :
** Den Leberzellen ahneln die Chloragogenzellen in folgenden
Punkten. Sie nehmen Pigmente (Kikenthal) und albuminoide
Substanzen (Cuénot) wahrscheinlich aus dem Blut auf und
absorbiren Indigkarmin und Eisen aus injicirten Lésungen ”
(30, A, p. 386). The analogy with the liver of Vertebrates
is emphasised when the chloragogen is in the form of a
heart-body ; it is then situated in the stream of blood from
the alimentary to the respiratory organs—in the portal blood,
in fact; and it is reasonable to conclude that its functions
play a similar part in the economy of the worm to that under-
taken by the liver in as far as the latter may be regarded as a
ductless gland.

The analogy cannot be followed into details. Glycogen,
though occurring in Sipunculus (Jourdan), and described even
in the chloragogen of Oligocheta (Cuénot, 1897), appears to
be absent in the heart-body. Other substances, however, may
exist in it, representing elaborated and stored products. Fat
and iron have been shown to be present, and the latter, it can
hardly be doubted, is associated with the hematopoietic func-
tion. The functions of the pigment and the chitinous bodies
are obscure, and if they be excretory their mode of removal is
difficult to understand. It has been suggested above that
waste products may be carried through the heart wall, and
thence to the nephridia by leucocytes, or emptied into the
blood and removed by gill excretion; the latter alternative
seems only applicable to fluid substances.

With regard to the mechanical functions of the heart-body
we are on surer ground. ‘The vessel wall contracting upon it
obliterates the heart cavity at systole, the whole of the blood
passing on to the gills.

In conclusion I wish to offer my sincere thanks for the
assistance I have received from the Staff of the Zoological
Statiou at Naples. Professor Paul Mayer gave me invaluable

296 LIONEL JAMES PICTON.

help, and I am also much indebted to Professor Hisig. I have
to thank Professor Lankester and Mr. E. 8S. Goodrich of
Oxford for suggestions and criticism.

APPENDIX.

Note on intra-vitam Staining of Nuclei with
Carmine in Sea-Water.

The question whether nuclei can be stained whilst alive is
still undecided. Galeotti (‘Z. f. wiss. Mikr.,’ xi, p. 172) says
that the processes of coloration in the normal condition ‘mai
si mostrano nel nucleo.’’? Pfeffer (‘ Untersuch. Bot. Inst.
Tibingen,’ vol. ii, p. 825) says that nuclei will only stain with
methylene blue after the cells have been damaged. Bethe
(‘ Biol. Central.,’ xv, 1895, p. 141) states that the epithelial
nuclei of a living piece of a Ctenophore may be stained with
methylene blue. Lauterborn (‘ Untersuchungen tber Dia-
tomen,’ Leipzig, i, 1896, p. 36) speaks of seeing moribund
diatoms, the nuclei of which were stained, move along in the
field of the microscope. Finally, Przesmycki (‘ Biol. Central.,’
xvii, 1897, p. 321), writes of “ intra-vitale Farbung des Kerns
und des Protoplasmas,” with ‘neutral red’? and methylene
blue by means of a special technique, to be explained in a
later paper.

During my work on the heart-body I had occasion to cut
sections of Siphonostoma which had been for a week in a jar
of aérated sea-water in which some powdered carmine had been
stirred up, aud was surprised to find that in sections mounted
without further stain the nuclei and especially the nucleoli of the
gut-cells were coloured red. The fixative used was corrosive
sublimate and acetic acid, and to the action of this on the car-
mine in the gut of the animal, Dr. Meyer attributed the stain.
I found, however, that the stain still occurred after fixation in
90 per cent. alcohol; and even in living tissue’ some nuclei
were found stained pink. The majority of the gut cells are so
loaded with granules in the fresh state that it is impossible to

1 Ciliary movement was going on in the preparation.

HEART-BODY, ETC., OF CERTAIN POLYCHATA. 297

make out the nucleus in them; but in some pale yellow cells,
occurring just posterior to the stomach, it can be seen. That
it was the nucleus that was stained in these cells was shown by
the action of acetic acid, which revealed no other nuclei. Dr.
Mayer, to whom I showed a preparation, described the appear-
ance as a faint but distinct nuclear stain, though he was
inclined to regard the carmine as in the form of granules in
the nuclei.

It should be stated that carmine partially dissolves in sea-
water, and this solution will stain fixed tissues; and also that
the contents of the gut of Siphonostoma are faintly alkaline.

BIBLIOGRAPHY.

1. ALLEN.—‘ Quart. Journ. Micr. Sci.,’ vol. xxxiv. (Gill excretion in Crus-
tacea. )
2. Ampronn.—‘ Mitt. Zool. Stat. Neap.,’ vol. ix, 1890, p. 475. (Cellulose
reaction in chitin.)
3. BarrutH.— Arch. fiir mik. Anat.,’ Bd. xxv, 1885, p. 269.
4, Bepparp.—‘ Monograph on Oligocheta’ (Clarendon Press), p. 77.
(“‘ Heart-body ” in Oligocheta.)
5. Bues.—‘ Report of Brit. Assoc.,’ 1891, p. 873. (Siphonostoma.)
6. Bucuanan (Miss).—(A) ‘ Quart. Journ. Micr. Sci., xxxi. (Hekatero-
branchus.)
7. BucHanan (Miss).—(B) ‘ Report of Brit. Assoc.,’ 1895, p. 469. (De-
velopment of Magelona.)
8. Bucuanan (Miss).—(C) ‘Quart. Journ. Mier. Sci.,’ xxxv. (Polynoid
with branchiz. )
9. CLAPAREDE.—(A) ‘ Annélides Chétopodes du Golfe de Naples,’ Geneva,
1868.
9’. CLAPAREDE.—(B) ‘ Rech. Annélides Séd.,’ p. 94.
9”. Cutnot.—‘ Arch. de Zool. Expt.,’ 2¢ sér., tom. ix, p. 438.
10. CunnincHaM.—‘ Quart. Journ. Micr. Sci.,’ xxviii, 1888, p. 259. (Heart-
body.)
11. Dapp1.—‘ Arch. Ital. Biol.,’ xxvi, 1896, p. 143. (Fat test.)

298 LIONEL JAMES PICTON,

12.

13.

14.

15.

16.

Ve

18.

19.

20.

21.

22.
23.

24.

~

24’.

25.

~

25’.

26.
27.

28.
29.

30.

Ers1¢.—“ Die Capitelliden des Golfes von Neapel,”’ ‘ Fauna und Flora des
Golfes von Neap.,’ xvi, p. 690.

Fauvet.—‘ Bulletin scientifique de la France et de la Belgique,’ tom.
xxx, 2° partie, p. 387, 1897. (Ampharetide.)

Goopricu.—‘ Quart. Journ. Micr. Sci.,’ xxxix. (Cclomic corpuscles of
Oligocheeta.)

HasweE.u.—‘ Proc. Linnean Soc. New 8. Wales’ (2), vi, 1892, p. 347.
(Chlorhemide.)

Horst.—‘ Z. Anzeiger,’ 8 Jahrg., 1885, pp. 12—15. (Hin rathselhaftes
Organ bei den Chloremiden.)

JourpAN.—‘ Annal. Musée d’Hist. Nat. de Marseille, Zoo.,’ vol. iii,
Mém. 2, 1887. (Siphonostoma.)

LANKESTER.— Quart. Journ. Micr. Sci.,’ 1878, vol. xviii. (Red
vascular fluid of the earthworm a corpusculated fluid.)

Lim Boon Kene.—‘ Trans. Roy. Soc.,’ exxxvi B., 1895, p. 383. (Ceelomic
fluid of Lumbricus.)

Kennet.—‘ Arb. Z. Inst. Wirzburg,’ Bd. v, 1882, pp. 386—388.
(Ueber Ctenodrilus.)

MacAtium.—‘ Quart. Journ. Mier. Sci.,’ xxxvili, 1895, p. 175. (Micro-
chemical detection of iron.)

Merrtcunikorr.— Legons sur |’ Inflammation.’

Meyer (Ep.).—(A) ‘Arch. fiir mik. Anat.,’? Bd. xxi, p. 815. (Poly-
ophthalmus.)

Meyer (Ep.).—(B) ‘ Mitt. Zool. Stat. Neapel,’ vol. vii, 1887. (Studien
iiber den Korperbau der Anneliden.)

Miine-Epwarps.—‘ Ann. Sci. Nat. Zoo.,’ 2° sér., tom. x, p. 199.
(Circulation of Polymnia.)

MonticEtui.— Boll. della Soc. di Nat. in Napoli,’ vol. x, 1897, p. 47.
(“ Polyophthalmus.”’)

NusBpaum und Raxowsk1.—‘ Biol. Centralblatt,’ Bd. xvii, 1897, p. 260.
(Enchytreide.)

RanviER.— Traité technique d’Histologie,’ p. 102. (Fat test.)

Rosa.—‘ Mem. Reale Acad. Sci. Torino,’ xlvi, 1896, p. 149. (Calomic
fluid of Oligocheta. Fat test.)

SaLtensky.— Archiv. de Biol.,’ iv, p. 252. (Terebella Meckelii.)

Suarpe1.—‘ Jen. Zeitschr. fiir Nat. Wiss.,’ 1894, pp. 247—293.
(“Ophelia radiata.”)

ScHNEIDER (GuIDo).—(A) ‘ Zeitschrift fiir wiss. Zool.,’ Bd, lxi, Heft 3.
(‘‘ Chloragogenzellen.’’)

HEART-BODY, ETC., OF CERTAIN POLYCHMTA. 299

31. ScuneIpER (Gurpo).—(B.) ‘Arb. Nat. Geo. Petersburg,’ Bd. xxvii,
Heft 1. (On the segmental organs and heart-body of Polycheta, with
a German translation.)

32. StrEN.—Jena, 1883, ‘Terebellides Stroemii’ (Inaugural Disser-
tation).

33. Wirtn.—‘ Kongl. Svenska Vetenskaps-Akad. Handlingar,’ Stockholm,
1885, 21, No. 7, pl. vi.

EXPLANATION OF PLATES 19—22,

Illustrating Mr. Lionel James Picton’s paper ‘ On the Heart-
body and Ceelomic Fluid of certain Polycheta.”

PLATE 19.

Fic. 1.—Audouinia filigera. Diagram of a dissection from the dorsal
surface. The head and posterior half of the worm are not represented.
The heart is represented as cut open, revealing the heart-body within it.
The arrows indicate the direction of the circulation.

Fie. 2.—Audouinia filigera. Transverse section, x 115, through
widest part of heart-body, The cut strands of the heart-body are represented
lying in the heart, bathed by the blood (red). «. Blood between folds of the
heart-body. y. Lumen in heart-body.

Fig. 3.—Anterior portion of young Polymnia nebulosa. Dorsal view ;
drawn whilst alive; under low power. The heart-body is a dark rod inside
the dorsal vessel (heart).

Fic. 4.—Pectinaria belgica. Transverse section of heart-body.
D objective (Zeiss).

PLATE 20.
All the figures in this plate refer to Audouinia filigera,

Fic. 5.—Transverse section of a strand of the heart-body, showing endo-
thelium, cortex, and medulla. The last is occupied by yellow granules and
spherical spaces, which contain chitinous bodies. a. Granule in vacuole.
b. Large yellow granule. c. Granule deeply stained with hematein. d.
Nuclei of medullary cells (the outlines of these cells are not evident). e.
Granular mass.

300 LIONEL JAMES PICTON.

Fic. 6.—Granular mass and its envelope. (Béla-Haller’s fluid.) 33,’ ob-
jective.
Fic. 7.—Ditto.

Fre. 8.—Granular mass and its envelope, from a section stained with
Ehrlich’s hematein, showing “nucleus.” 7’.

Fic. 9.—A mass divided into two. (Béla-Haller’s fluid.)

Fic. 10.—A mass divided into three, from a section. Parts are deeply
stained with hematein.

Fie. 11.—A portion of the medulla. 4’. From a section.

Fic. 12.—Four chitinous bodies in a common envelope. (Béla-Haller’s
fluid.) 4’.

Fic. 13 —Chitinous bodies after being boiled in caustic potash solution.
D (Zeiss).

Fic. 14.—Chitinous body, surrounded by granules. ;y’.

Fies. 15—19.—Formation of brown granules. ;,’ objective.

Fic. 20.—Heart-body, heart-wall, and adjacent peritoneum. Transverse
section. 5}.

Fics. 21 and 22.—Leucocytes with pigment, fresh. 2’.

PLATE 21.

Fic. 23.—Siphonostoma diplochetos. ‘Transverse section through
heart-body at the origin of the heart. x 85.

Fic. 24.—S. diplochetos. Transverse section of heart-body. jy’. The
nuclei stained with hematein (the dark granules, red in the fresh tissue, were
stained deep greenish blue with hematein; these are shown in the figure in

black).

Fie. 25.—An isolated cell of the heart-body of 8. diplochetos, in its
natural colours. jy’.

Fic. 26.—Similar cells, after maceration and teasing in Béla-Haller’s
fluid. 3.

Fie. 27.—Isolated cell of heart-body of Lanice conchilega. jy’. Béla-
Haller’s fluid.

Fic. 28.—Lanice conchilega. Composite granule. ;;’.

Fic. 29.—Polycirrus, amoeboid corpuscle, with amber-coloured granules ;
fresh. jh’.

Fies. 30 and 31.—Polycirrus, pigmented corpuscles; fresh. 3’.

Fic. 32.—lIsolated cell of heart-body. Terebellides Stroemii. Béla-
Haller’s fluid. Stained with methyl green.

HEART-BODY, ETC., OF CERTAIN POLYCHmTA. 3801

Fic. 88,—Polymnia nebulosa. Larva (ten pairs of chete). Pigmented
patch on wall of ventral vessel.

Fic. 34.—Siphonostoma diplochetos. Sketch showing an attach-
ment of the heart-body to the heart-wall; fresh. D (Zeiss).

Fic. 35.—Polymnia nebulosa. Sagittal section of posterior end of
heart-body in the larva, showing communication of its cavity with the celom
in course of obliteration. ,’. Blood red, nuclei blue. zx. Point where the
heart-wall is continuous with the wall of the heart-body.

Fic. 36.—Polymnia nebulosa. Sagittal section of posterior end of

heart-body at an earlier stage than in Fig. 35. 4’. Blood yellow, nuclei
blue.

PLATE 22.
Catomic FLvip.
Siphonostoma diplochetos.

Fie. 37.—Oval corpuscle (fixed in alcohol ; stained heemalum and eosin). ;,’.

Fie, 38.—Problematical bodies (hamalum, &c.). D (Zeiss).

Fic. 39.—Problematical body ; fresh. ,’.

Trophonia.,

Fic. 40.—Fusiform cclomic corpuscle, with stalk. (Methyl green and
eosin.) y’.

Fic. 41.—Stellate corpuscle. (Formol, alcohol, and acetic mixture, &.) j5'.

Fic. 42.—Ameeboid corpuscle; fresh. 4.

Fie. 43.—Problematical body ; fresh. D (Zeiss), In the cellular envelope
fat droplets are seen.

Fics. 44—46.—Amphitrite variabilis. Fusiform corpuscles. 1’.
Fig. 46 shows a granule stained with hemalum.

Fic. 47.—Young Polymnia nebulosa. Celomiccorpuscle. 5.
Amphitrite (variabilis and rubra).
Fic. 48.—Granular corpuscle. 4’.
Fic, 49.—Granular ameeboid corpuscle; fresh. ,'.
Fig. 50.—Eleocyte ; fresh, showing fat and pigment spherules, and granu-
lar protoplasm. The cell-envelope is distinct.

Fic. 51.—Smaller eleocytes, with little fat and well-marked pigment granules.
Osmic, bichromate of potash, hematein. ;,’.

Fic, 52.—Eleocyte. Fixed in absolute alcohol, stained with hemalum ;
fat stained with Soudan iii; granulated cell-protoplasm unstained.

voL. 41, PART 2.—NEW SERIES. x

3802 LIONEL JAMES PICTON.

Polymnia nebulosa.

Fic. 54.—Fusiform corpuscle. Alcohol, hemalum. 4’.

Fic. 55.—Fusiform corpuscle. Stained methyl green and eosin. j,/. The
contained circular body is stained deeply with eosin, its ‘‘nucleus” blue
with methyl green.

Fie. 56.—Vacuolated fusiform corpuscle. y’.
Fie. 57.—Ameeboid corpuscle ; fresh. 4’.

>

Fies. 58 and 59.—* Blood-corpuscles”” of Polycirrus. Hemalum. ,!

mae

Fie. 60.—New bow-shaped corpuscles in Notomastus profundus.

Absolute alcohol, hemalum, eosin. 7’.

Ophelia radiata.
Fic. 61.—Rod-cell, showing notches in rod. A (Zeiss).
Fic. 62.—Rod-cell, showing deciduous process. A (Zeiss).

LAKE TANGANYIKA——AN OLD JURASSIC SEA, 303

On the Hypothesis that Lake Tanganyika
represents an Old Jurassic Sea.

By

J. E. S&S. Moore.

With Plate 23.

“For anything that geology or paleontology can show to
the contrary, a Devonian fauna and flora in the British Islands
may have been contemporaneous with the Silurian life in
North America, and with a Carboniferous fauna in Africa.
Geological provinces and zones may have been as clearly
marked in the Paleozoic epoch as at present, and those seem-
ingly sudden appearances of new genera and species which we
ascribe to new creation may be simply due to migration.”

If the statements contained in this remarkable passage
express the truth—and no one acquainted with the forcible
arguments which Huxley brought forward! in their support
will doubt that such is actually the case—it follows as a sort of
natural corollary that the existence of our modern fauna and
flora may not be incompatible with the co-existence in certain
places of extremely ancient types. In several former papers I ?
have laid especial emphasis upon the very singular fact that
the fauna of Lake Tanganyika is a double series, that it is in
reality composed of two entirely dissimilar faunas which co-
exist in the great lake side by side.

1 Anniversary address to the Geological Society, 1862.

2 «Nature,’ July, 1897, p. 198; ‘Science Progress,’ October, 1897; “ The
Molluscs of the Great African Lakes :—Distribution.” ‘Quart. Journ. Mier.
Sci.,’ vol. 41, pp. 159—180.

304 J. E. S. MOORE.

I! have shown that one of these two faunas consists of the
normal and ubiquitous fresh-water stock, which is distributed
throughout the whole African continent, and indeed through-
out the world. The second fauna is altogether different from
this, and in the appearance of its widely divergent constituents
is utterly unlike any modification of the normal fresh-water
fauna that is known. It has long ago been recognised that
the superficial facies of the molluscan shells belonging to this
series are those of a marine rather than a fresh-water stock,
and in recognition of the more complete marine affinities
which a closer scrutiny of the internal anatomy of these
animals has revealed, I? have here, as elsewhere, spoken of the
whole series of forms in Tanganyika which exhibit these quasi-
marine characters as members of the halolimnic group.®

1 «On the Zoological Evidence for the Connection of Lake Tanganyika with
the Sea,” ‘ Proc. Roy. Soce.,’ vol. Ixii, 1898, pp. 452—458.

2 «The Molluses of the Great African Lakes.—II. The Anatomy of the
Typhobias, &c.,” ‘Quart. Journ. Micr. Sci.,’ vol. 41, 1898, pp. 181—202.

3 Tf the practical distinction between fresh-water and marine faunas in general
were not a well-established and accepted fact, it would have been impossible
for geologists to separate, as they have done, fresh-water from marine deposits
by the characters of the animals they contain. It is generally assumed that
the modern fresh-water fauna has gradually originated far back in time by
organisms having one by one acquired characters which have enabled them to
successfully colonise fresh water in connection with the sea; but the actual
phylogenetic descent of most of the true fresh-water organisms, except in a very
broad sense, is lost in antiquity and hopelessly obscure. In some Crustacea,
in the Ganoids, and some other fishes we have enough paleontological
evidence to demonstrate their actual migration from the sea, and such evidence
forms part of the ground whence it is argued from analogy that all fresh-water
organisms have originated in a similar way. Further evidence of this kind is
afforded by those cases, at once remarkable and few, where animals that are
eenerally marine exhibit a wonderful capacity to migrate inland, there being
every reason to believe that such organisms constitute the ‘modern instances ”
of the origin of new fresh-water types. The true fresh water fauna of any period
is thus a heterogeneous assemblage of organisms, all of which have, so to speak,
voluntarily acquired the habit of living in fresh-water, and, excepting in this
peculiarity, they have no necessary relation with each other. The constant
facies which the fresh-water fauna presents all over the world are due primarily
to the universal distribution of its heterogeneous constituents, and secondarily
to the direct similar effect produced on organisms by a fresh-water life.

LAKE TANGANYIKA—AN OLD JURASSIC SEA. 305

It is perhaps needless for me here to reiterate the great im-
portance of arriving at a final decision as to the real nature of
the halolimnie forms, for it will be obvious that if they have
nothing to do with the normal fresh-water series, and are to
be regarded as the remnant of an ancient sea, our views
respecting the past history of the African interior must be
greatly changed.

Having obtained the animals, it appeared, therefore, in the
first place to be incumbent on me to ascertain, by a careful
study of their anatomy, whether this superficially marine
appearance was real, and indicative of their common origin
from the sea, or whether it was merely, so to speak, skin deep,
and to be regarded as wholly the result of modification, or to
the persistence of characters which belonged to some old fresh-
water stock. So far as zoologists are concerned, the evidence
which I have now accumulated on this point will be found to
be conclusive ; but since, with the exception of my paper on
the Typhobias (loc. cit.), the detailed accounts of the anatomy
of the Halolimnic Gasteropods have not yet been published, I
will briefly recapitulate the facts. It has been found‘—

1 Tn order that the significance of the new classificatory outline, given in the
text immediately below, may be fully appreciated, it should be clearly under-
stood that before my return from Tanganyika no account of the anatomy of any
of the halolimnic molluses was in existence,—indeed, so far as I can ascertain,
with the exception of the brief description by Smith of a few badly preserved
Nasopses brought home by Captain Hore, the animals contained in any of
these shells had never been seen before. Consequently their conchological
classification was, as, indeed, Smith frankly implied in 1881, entirely pro-
visional.

All these Halolimnic Gasteropods appear to be rigidly restricted to the
confines of Lake Tanganyika, and the only molluses from this lake, halolimnic
or otherwise, of which we have had any anatomical description, is the common
Reiodon, the morphological characters of which were described by Professor
Pelseneer from specimens which were brought back by the officials of the
Congo Free State *(‘ Bull. de Musée Royale d’Hist. Nat. de Belg,.’ Brussels,
1886, iv, p. 108).

The marine appearance of the genus Nassopsis was pointed out by
S. P. Woodward in 1857, but with curious inconsistency he regarded this
form as a Melania belonging to the sub-genus Melanella. Itis to Smith that
we owe the first definite assertion of the possibility that these Gasteropods

306 J. E. S. MOORE.

1. That in the genera Bathanalia and Typhobia we
have a type of Gasteropods which stands very much in the
same relation to the modern Strombide that the early Equide
do to the modern horse.

2. That in the so-called Spekia zonatus we have a form
which even in its most minute anatomical details, as well as
in its shell structure, is an unquestionable Naticoid of the

Lamellarian type.

3. That the so-called Tan ganyicia rufofilosa is closely
related to the oceanic Planaxids, and that it is antecedent to a
certain section of the heterogeneous Melanoid group, much in
the same way that Littorina is antecedent to another.

4. That the genus Limnotrochus is really compounded

might, when their anatomy became known, turn out to be marine derivatives.
Smith was unfortunate, however, in his forecast of the affinities of Typhobia
as a Melania, since it is obvious, from the character of the radula of this
molluse alone, that it has no affinity with that group (see my figs., ‘ Quart.
Journ. Mier. Sci.,’ vol. 41, pt. 1, p. 189). Great credit is, however, due to
Smith for his shrewd guess at the marine nature of the halolimnic shells
with which he was then acquainted, and more especially so because he was not,
as it were, frightened out of his better judgment as a naturalist by the existing
geological preconceptions respecting the past history of the African interior.
The later classification of the Tanganyika shells given by Bourguignat (‘ Ann.
des Sci. Nat.,’ t. x, pp. 1—267) is quite unintelligible either as to the means
by which his endless species are distinguished from each other, or as to their
affiliation in his so-called natural groups. Indeed, as an example of the utter
confusion and obscuration of the facts which may be produced by the un-
restrained application of the conchological method of determining molluscan
affinities when the animals contained in shells are quite unknown, this work is
perhaps unrivalled. In M. Fischer’s excellent conchological treatise, on the
other hand, there will be found a careful estimation in each case of the probable
affinities of those Tanganyika shells which were known. But each of these is,
of necessity, simply drawn from conchological data, and the caution with which
the author proceeds in the absence of all morphological information is
most marked. In order that the reader may obtain a clear conception
of the points in which what may be called the newer classification given in
the text of this paper differs from and extends that which could be arrived
at by the study of the empty shells, I give here in parallel columns for com-
parison a list of the families and genera with which the Halolimnic Gas-
teropods are incorporated in M. Fischer’s work, and those to which I should
myself refer them after a study of the morphology of each. In this list I

LAKE TANGANYIKA—AN OLD JURASSIC SBA.

307

of two distinct types, one of which, represented by L.
Thompsoni, is closely similar to Bathanalia; while the other,
represented by the unique L. Kirkii, is the only fresh-water
Xenophora (Onustus) at present known.

5. That in the Paramelanian group, composed of the genera
Paramelania, Nassopsis, and Bythoceras, we have forms

have used the family name Purpurinide to include the genera Parame-
lania, Nasopsis, and Bythoceras, and the new generic name Chytra for
the old generic name of Limnotrochus, in the case of L. Kirkii.

Conchological Classification
according to M. Fischer.

Fam.—Melaniide.
Genus.—Typhobia (Smith).
T. Horei, Smith.
Genus.—Paramelania (Smith).
P. Damoni, Smith.
M. nassa, 8. P. Woodw.
Fam.—Hydrobiide.
Genus.—? ? Syrnolopsis (Smith).
S. lacustris, Smith.
Genus.—Spekia (Bourguignat).
S. zonata, 8. P. Woodw.
Genus.—Tanganyicia (Crosse).

T. rufofilosa, S. P. Woodw.

Genus.—Limnotrochus (Smith).
L. Thomsoni, Smith.
L. Kirkii, Smith.

Redetermination from the
Characters of the Animals
themselves.

Fam.—Typhobiide (Moore).
Genus.—Typhobia.
T. Horei, Smith.
Genus.—Bathanalia (Moore).
B. Howesii, Moore.
Genus.—Limnotrochus (Smith).
L. Thomsoni, Smith.
Fam.—Planaxide.
Genus.—Tanganyicia (Crosse).
T. rufofilosa, 8S. P, Woodw.
F'am.—Xenophoride.
Genus.—Chytra (Moore).
C. Kirkii, Smith.
Fam.—Purpurinide.
Genus.—Paramelania (Smith)
= Purpurina (Hudl.).
P. Damoni, Smith.
P. crassigranulata, Smith.
Genus.—Nassopsis (Smith).
N. nassa, 8. P. Woodw.
Genus.—Bythoceras (Moore).
B. iridescens, Moore.
Fam.—Naticide.
Genus.—Spekia (Bourguignat).
S. zonata.

In the above list I have placed the genus Chytra among the Xenopho-
ride on account of its conchological similarity to numerous fossils which are

referred to this group.

The notes of interrogation in the older list are those of M. Fischer.

308 J. KE. S. MOORE,

which, judged by their anatomical (as well as by their con-
chological) features, do not appear to be living elsewhere now,
but their shells approximate in a most remarkable degree to
those of the extinct marine Jurassic genus Purpurina, whilst
at the same time they possess the nervous system of a Cyclo-
phorus. They thus appear not only to come of a marine
stock, but also to indicate the hitherto unknown road by which
the Cyclophoran nervous system has been evolved. These living
Teenioglossa stand in much the same relation to their extinct
marine ancestors as the living Cyclostoma has been shown
(by the beautiful investigations of Lacaze-Duthiers and Bouvier)
to stand in relation to the common periwinkle of our shores.

No Stromboid, Naticoid, or Xenophoran molluscs have
been found hitherto in any fresh water that is known; and
when we remember that these truly marine Gasteropods are
associated in Tanganyika with other and widely different
marine forms, such as sponges, meduse, crabs, and prawns, it
is impossible to avoid the conclusion that these animals can be
anything but the dwarfed and stunted remnant of a fauna that
the sea has left behind.

That the halolimnic animals still living in Tanganyika are the
remains of an extensive sea fauna that once existed there, is thus
the plain and unequivocal testimony afforded by the morphology
of the widely different types of which this fauna is composed.

This being so, in the present paper I shall attempt to show,
from a variety of considerations relating to the similarity
between the halolimnic shells and certain fossils, and upon
more general grounds, to what old sea fauna the halolimnic
series once belonged. It will probably have been seen that
the above conclusion by no means exhausts the information
which can be gathered from the joint study of the distribution
and the morphology of the halolimnic group. We need only
refer to what is now generally known respecting the gross
physical features of the African continent, and especially of
the regions about the great lakes, to see that it is impossible
for most if not all the halolimnic forms either to have made
their way up to, or to have been left in, Tanganyika in recent

LAKE TANGANYIKA—AN OLD JURASSIC SBA. 309

times. The lake is now 2700 feet above the level of the sea,
and is more than 700 miles from any coast; there is but one
effluent, and the course of this river is beset with rapids and
with falls even long before it reaches the lower channels of the
Congo on its way towards the sea; and finally, there are no
true representatives of the halolimnic fauna, except, perhaps,
the universal Xenophoride, in that part of the Atlantic into
which the Congo flows.

The physiographical features of the continent point directly, ~
therefore, to the conclusion that the halolimnic fauna must be
very old. It must have been left in the great valley of
Tanganyika long before that part of the continent had at-
tained its present altitude, and when the surface of the water
was approximately at the level of the sea. In exact conformity
with this indication, it will have been seen that the halolimnic
animals as they now exist, although closely allied to different
marine genera, are not exactly similar to any oceanic species
that we know ; and finally, it has been shown that the halolimnic
Gasteropods, at any rate, stand in the relationship of immediate
ancestors to several of our well-known oceanic forms. There
thus exists evidence which appears to be practically conclusive
that the halolimnic animals retain the facies of a sea fauna
that has elsewhere disappeared, and consequently, unless they
have become modified out of all semblance to their original
marine progenitors, it is only natural to expect that on some
marine fossiliferous horizon we shall again encounter in a
fossilised condition similar molluscan shells. The hope that we
may in this way be able actually to “locate ” the halolimnic
fauna of Lake Tanganyika with that particular marine stock
from which it sprang is all the greater, on account of the very
striking facies which the shells of the molluscs belonging to it
invariably present. But in actually searching among marine
deposits for the particular sea fauna to which the halolimnic
animals may correspond, it is essential that we bear in mind
the caution that single comparisons are likely to be of little
service as affording any indication that two such faunas are
the same. There must be in the old stock, to which we are

310 J. E. 8. MOORE.

going to compare the halolimnic fauna, at least a sufficient
number of types which are similar to individual halolimnie
forms to. correspond with a majority of the forms the halo-
limnic fauna now contains. I have emphasised this point
because certain comparisons have already been instituted
between the shells of the Paramelanias of Tanganyika and
forms occurring in the fresh-water cretaceous beds.

In 1883, White,! in an extremely short paragraph, pointed out
that, speaking conchologically, there is not much to distinguish
the shell of the genus Paramelania (Smith) from that be-
longing to the extinct fresh-water Pyrgulifera, which he
obtained from the Green River deposits of the United States.
So far as the outward forms of these shells go, there are slight
differences as to sculpture, and so forth (compare Pl. 23, figs.
1 and 7). But I do not know that such dissimilarities as
these would justify even a conchologist in regarding the genera
as distinct, and that this comparison of a single halolimnic
and cretaceous shell is, in the absence of any possibility of
information respecting the nature of the contained animals or
their associates, ‘so far, so good,” seems to be the total net
result of the further observations made upon the subject by
Tausch? and Oppenheim,® except that these latter authors
appear to have had at their command more extensive and
better preserved material than that which White examined.
Speaking conchologically, then, there is one type in the creta-
ceous fresh-water deposits and one in the African halolimnic
fauna which are similar in form. But even in the case of the
single correspondence which presents itself Tausch’s work
appears to have rendered it extremely doubtful whether the
two forms can be still considered as even conchologically the
same. He showed, after examining hundreds of Pyrguliferas
from the upper cretaceous beds from Ajka in Hungary, that

1 «Proc. U.S.A. Nat. Mus.,’ 8. 98, Washington, 1882, p. 98 (published in
1883).

«Sitz. Ber. d. k. Acad. Math. Wien,’ 1885, Bd. xe, p. 57.

3 Zeitschrift. der Deutsch. Geol. Gesell.,’? 1892, Bd. xliv, p. 697; for

diagnosis of Pyrgulifera see ‘U.S. Geol. Surv.,’ 40 parallel, vol. iv, p. 146,
pl. 7, fig. 19.

LAKE TANGANYIKA—AN OLD JURASSIC SBA. 311

their shells could be sorted out into several groups which in
their extreme forms were quite distinct, but which were really
indissolubly connected together by innumerable transitional
types. Thus one type of Pyrgulifera agrees with Palu-
domus Pichleri (Hoern.) from the ‘‘ Gosauformation,”
another with P. armatus (Math.) from the French chalk, a
third with P. lyra (Math.) from the same, a fourth with
Pyrgulifera humerosa (Meek) from the Laramie of North
America; while toa fifth and sixth Ajka variety there seem to be
no known corresponding forms. Since all these types are stated
by Tausch to run completely into one another, they can but be
regarded as connected polymorphs of one and the same generic
type, whatever the actual organisation of this genus may have
been. Tausch further points out that in Paludomus
Pichleri there are certain characters at the base of the
mouth which have led to this shell being described both as a
Paludomus and a Melanopsis. This melanopsid “ mouth ”
is not found, according to Tausch, in P. Stephanus (Bens),
but it is present in P. humerosa and in the Paramelania
Damoni of Tanganyika. ‘Tausch therefore argues that the
Paramelania Damoni, Pyrgulifera humerosa, and those
forms of Paludomus which possess this peculiarity of
“mouth ” are, together with certain forms of Melanopsis,
merely varieties of a single polymorphic type. This type
embraces also in its other modifications forms approximating
to Melania amarula, Lamarck’s type of the genus Melania.
I can fully confirm the observation of the remarkable similarity
of some of the Pyrguliferas collected by Dr. Oppenheim,
and now in the British Museum, to M. amarula; in fact,
some of these forms approximate far more closely to the
living M.amarula of Madagascar than they do to the Para-
melania of Tanganyika. Thus, whatever the dead Pyrguli-
fera may have been, its shells in their different modifications
agree with a great number of living types, and if it be really
legitimate to draw any conclusion from this complexity of
corresponding forms, it can only be said that Tausch’s work
has shown that there appears to have existed in the fresh water

o12 J. E. 8. MOORE.

of the upper cretaceous series a form which united by insensible
gradations the conchological characters of Melanopsis,
Paludomus, Pyrgulifera, Melania (amarula) and Para-
melania. But if on this ground it should be maintained that
the living representatives of these different groups have any
immediate phylogenetic relationship with each other, all that
can be said by anyone acquainted with the morphology of such
of them as now exist must be that although a deduction of
this kind from the characters of living and extinct shells may
be conchologically correct, it is also at the same time morpho-
logically nonsense ; there is no sort of morphological similarity
between Melanopsis and a Melania amarula. These
forms, as the investigations of Bouvier have shown, should by
right be placed in different families. Paludomus differs
from them both, while the Paramelania of Tanganyika is
altogether unlike any of the three. Thus if the genus
Pyrgulifera corresponds to any of these types which now
exist, it differs from all the rest which I have named. If
Pyrgulifera humerosa was morphologically similar to
Melanopsis, it was not a Paramelania. If, on the other
hand, it was a Paramelania, it was neither a Melampus
paludomus nor a Melania proper. There is thus really no
direct reason why the Pyrgulifera of the chalk should not
have beena Paramelania; but since the genus Pyrgulifera
has been shown by Tausch to correspond equally to three
widely distinct living types, it is clearly more than three to
one that such was not the case.

As to the question of the identity of the entire fresh-water
fauna with which the Pyrguliferas are connected in the upper
chalk, and that consisting of the halolimnic group in Tan-
ganiyka, whether we regard the Paramelanias and Pyrguliferas
as similar or not, it will be obvious that as there are no other
forms in these faunas bearing the slightest resemblance to one
another, the question of their general identity is ipso facto
out of court. Not only do the halolimnic animals differ from
those of the fresh-water fauna individually, but the whole
halolimnic fauna differs entirely from the cretaceous or any

LAKE TANGANYIKA—AN OLD JURASSIC SBA, 313

other fresh-water stock in the general facies it presents. These
old eretaceous beds present the facies of a true fresh-water
fauna, otherwise they could not be identified as such; they
contain no crabs or prawns, there are no impressions of jelly-
fish in the soft grey mud of which they are generally composed ;
they contain no shore sponges, Lamellariidz, no Xenophoride
or other marine Gasteropods, all of which are still living in
the slightly brackish water of Tanganyika at the present time.
In fact, the halolimnic fauna differs from that occurring in
these fresh-water cretaceous beds just in those features which
distinguish fresh-water from marine stocks in general, and there
is not the slightest doubt that had the halolimnic fauna occurred
fossilised it would have been regarded as unquestionably marine.

The halolimnic fauna of Tanganyika, then, is not the remnant
of a cretaceous fresh-water stock, neither is it likeany cretaceous
marine fauna which we know, nor is it represented in any of
the upper Mesozoic beds. Itis only when we compare the shells
of the halolimnic molluscs with those in several of the lowest
secondary formations that any substantial similarity appears.

In fig. 1 a, are represented two remarkably fine examples
of the marine Jurassic genus Purpurina; the figure is copied
from a specimen P. bellona courteously placed at my disposal
for this purpose by Mr. Hudleston from his magnificent collec-
tion of Jurassic fossils. The genus has a somewhat curious
history in literature, which will be found fully dealt with in
Mr. Hudleston’s! monograph, ‘ The Jurassic Gasteropoda.’ As
amended in this work for P. elaborata, the diagnosis of the
genus runs as follows:

* Shell ovate conoidal, apex acute, whorls about five or six,
posterior area tabulate, sides moderately tumid. The orna-
ments consist of about eighteen longitudinal costz, which are
feebly developed on the tabular area, rise up into spinous nodes
on the keel, and are strong and regular on the flanks of the
whorls. The coste have a tendency to die out anteriorly on
the body-whorl ; the costz decussate with regular and closely

1’ «A Monograph of the British Jurassic Gasteropoda.’ Palzeont. Soc., 1887,
Part 1, No. 2, p. 86.

314 J. E. S. MOORE.

set spirals, which extend down to the base of the shell. No
spirals are seen on the flat area. Aperture oval to subquad-
rate, columella moderately reflexed, so as to produce anteriorly
a wide and shallow groove towards the point. Umbilical slits
scarcely indicated.”

Side by side with these old Jurassic shells I have had drawn
two corresponding views of Smith’s Paramelania Damoni
from Tanganyika (fig. 1), the generic diagnosis of which runs
as follows :

‘Shell solid, ovate, conical, imperforate, longitudinally
ribbed, transversely lyrate, covered with a thin epidermis.
Aperture ovate, entire, indistinctly effuse at the base, last
whorl sometimes slightly prolonged inferiorly. Peristome
thick, margins joined by a callosity, operculum like that of
Ty phobia.””!

The striking similarity of the two shells from these descrip-
tions will be at once apparent; in fact, as Mr. Hudleston
remarked while we were examining the recent shells and fossil
side by side, “they are not only generically the same, but
specifically identical.”’

The shells of the genus Paramelania were, however,
shown by the German authors I have quoted to be similar to
the Cretaceous genus Pyrgulifera, and, as objects which are
like the same thing are necessarily like each other, it becomes
a question for the systematists and the conchologists whether
or not the genus Pyrgulifera and Paramelania should be
quashed, and both replaced by the older genus Purpurina.
There are slight differences between the shells of the genera
Pyrgulifera, Purpurina, and Paramelania when they
are carefully examined side by side; but these are not at all
sufficient to separate the specimens from one another as
specifically distinct, and, as Dr. Woodward pointed out to me,
those of the genus Paramelania approximate more closely to
the shells of the Jurassic genus Purpurina than they do to
the more recent Pyrgulifera type.

In Hudleston’s monograph there are represented two rather

1 “Proc, Zool. Soc.,’ 1881, p, 559.

LAKE TANGANYIKA—AN OLD JURASSIC SEA. 315

distinct types of Purpurina shell, one characterised by the
P. bellona (fig. 14), the other by the P. inflata given
in fig. 24. Hudleston did not separate these forms as generi-
cally distinet, but figured the types of which they are charac-
teristic on separate plates. How closely similar this inflata
type of Purpurina is to the living Nassopsis of Tanganyika
will at once be apparent from figs. 2 and 2a. The genus
Nassopsis was separated by Smith! from Paramelania on
account of the difference in the operculum, but it is doubtful
if this distinction can be maintained from their anatomy ; in-
deed, I should be inclined to place Paramelania, Bytho-
ceras, and Nassopsis as species of one new family, the
Paramelanide. The fact that there is more constant distinction
between the Tanganyika Paramelania and Nassopsis now
than that which used to exist between the bellona and
inflata types of Purpurina is just what we might expect,
since it is probable that these two forms would become less
transmutable as time went on.”

The Tanganyika Paramelania and Nassopsis are thus
identical with two forms occurring in the old Jurassic beds, and
the Paramelania corresponds more closely to the Bellona type
of Purpurina than it does to the Pyrgulifera of the chalk.

In the same Jurassic series there is another characteristic
genus, Amberlya, which is specifically very variable in size,
sculpture, and in the character of its spines. Two forms are
represented in fig. 3a, the upper one from the collection in
the British Museum, the lower from Mr. Hudleston’s collec-
tion. The history of this genus Amberlya is peculiar and
instructive, and will be found fully set forth in Hudleston’s
monograph.* The genus was originally founded by Morris
and Lycett, but was subsequently modified by Hudleston, and

1 © Proc. Zool. Soc.,’ 1881, p. 559.

2 There is a peculiarity in the base of the columella of some of the
Nassopsis shells which is not represented in those of the genus Purpurina,
but which is a permanent feature of the Jurassic Monodonta. So far as

Nassopsis goes this is an unimportant feature, since it is not constant in
the genus.

* Loe. cit., part 1, No. 6, pp. 274—279.

316 J. E. S. MOORE.

as amended by him the diagnosis runs—“ Shell turbinate,
more rarely trochoid, rather thin, imperforate or nearly so;
subelongate, frequently turreted ; sutural space wide; orna-
mented with spiral bands, usually spinulous or nodular, some
of which are prominent. The interspaces are finely striated,
the striz being slightly oblique to the axis; sometimes these
fine lines are strong enough to represent fine axial ribs. Base
rounded, spirally ribbed, and marked by fine radial striz ; aper-
ture suboval, but varying according to age, in the adult more
or less rounded, so as to become suboval or subcircular; there
is usually a considerable deposit of callus; outer lip thin, often
crenulate.”

This description would certainly answer for that of one of
the new types which I found in Tanganyika, and for which I
have proposed the generic name Bathanalia (fig. 3), for
although the Jurassic genus Amberlya shows a considerable
range of specific variation, all its species have essentially the
same characteristics as the two represented in fig. 3a, upper
and lower. The thin shell, the absence of all trace of epi-
dermis, and the character of the whorls, as well as the sculpture
and the character of the mouth, are all essentially the same in
Bathanalia as they are inAmberlya; the only point in
which they differ is in the columella, that of Bathanalia
being generally open, while that of Amberlya is always closed.
I have, however, consulted Mr. Edgar Smith and others about
this, and he assures me that such differences cannot be upheld
as generically distinctive, more especially as the amount of
umbilical opening in Bathanalia varies a good deal in extent
from shell to shell. We may, therefore, conclude that con-
chologically Bathanalia and Amberlya are the same.

The next example of the close similarity existing between
the living shells in Tanganyika and the marine Jurassic types
is that afforded by the Limnotrochus Thompsoni of the
one and certain so-called Littorinas of the other. In fig. 5
are represented two views of L. Thompsoni, while in fig. 5a
are given similar views of the Jurassic species Littorina
sulcata.

LAKE TANGANYIKA—AN OLD JURASSIC SEA. 317

Smith’s generic diagnosis of Limnotrochus runs thus :—
“Shell trochoid, umbilicated, without an epidermis, spirally
ribbed ; body-whorl keeled round the middle; aperture non-
lyrate within; with the outer lip oblique, the basal margin
broadly sinuated, and the columella edge somewhat reflexed.
Operculum horny, paucispiral, litterinoid.”! This description
would not do for the living Littorinas of our shores, but it
covers the two forms, one from Tanganyika and the other
from the marine Jurassic beds, just described.

I would next direct attention to the very obvious con-
chological similarity between the so-called Limnotrochus
Kirkii (fig. 6) and the marine genus Xenophora (Onustus),
a form which has extended in the ocean from the Devonian to
the present time. This genus is not, therefore, typical of the
Jurassic period, specially those which I have already described,
but it forms one more remarkable example of the marine
character of the halolimnic forms. I have represented side by
side the L. Kirkii (fig. 6) and an example of Onustus
(fig. 6 a), a typical Jurassic form.!

I have already stated that the so-called Lithoglyphus
(Spekia) zonatus of Tanganyika (fig. 4) is unquestionably,
from the characters of its anatomy, a Naticoid ; and in the infe-
rior Oolite there are forms which it would be quite legitimate to
regard as coming near this genus. ‘To illustrate this fact I
have figured the so-called Neridomus (fig. 4) of the inferior
Oolite, the affinities of which are doubtful in a high degree.

Lastly, in Tanganyika there exists a remarkable longitudi-
nally sulecated shell known as Melania admirabilis (Smith) ;
how closely this form corresponds to those remarkable Oolite
shells known as Cerithium subscalariforme will be seen
on comparing their respective shells. I did not find this species
myself in Tanganyika, and as the animal it contains is not

1 * Proc. Zool. Soc.,’ 1881, p. 285.

2 It is needless for me to point out that the two forms of Xenophora here
figured from Tanganyika and from the Inferior Oolite are not specifically the
same. The so-called Limnotrochus Kirkii of Tanganyika being much

more like several modern examples of the genus Onustus (Xenophora).
The figures only illustrate the general similarity of such shells.

voL. 41, PART 2.—NEW SERIES. i

818 J. BE. S. MOORE.

known I have not thought it necessary that I should give
figures of them here.

Besides the above marine types the halolimnic fauna
contains two forms, Syrnolopsis and Turbonella terebri-
formis, which, although they do not resemble any known
Jurassic shells, nevertheless exemplify in a remarkable
manner the marine affinities of the halolimnic mollusca as a
whole ; for the shell of the first of these species is practically
undistinguishable from that of the genus Syrnola, a form
found in the tropical seas, the second from that of the genus
Terebra.

It is thus apparent that with the exception of Ty phobia,!
and possibly of Bythoceras, all the halolimnic genera now
living in Lake Tanganyika are generically identical with
Jurassic forms, while two of these, Paramelania and Nas-
sopsis, contain forms which are specifically indistinguishable
from their corresponding Jurassic types.

Curious and startling as the foregoing comparison undoubt-
edly appears, I might still have had some hesitation in
bringing it forward as evidence of the origin of the halo-
limnic fauna, had not the three authors, White, Tausch, and
Oppenheim, practically foreed my hand by attempting the
comparison of which I have spoken between the living halo-
limnic and the old cretaceous fresh-water stocks. Whatever
may be the real value of evidence which is based upon shell
structure alone (and this certainly becomes more and more
questionable as time goes on), it will have been rendered
clear that the amount of this kind of evidence favouring the
similarity of the halolimnic and old cretaceous fresh-water
stocks is utterly insignificant beside that which can be pro-
duced in favour of the similarity of the halolimnic and old
marine Jurassic forms.

So far as I am concerned, therefore, this paper will have

1 The genus Typhobia, as I have shown, is, however, closely related to
Bathanalia, and there is very little doubt that it simply represents a modifi-
cation of the former form. It may be that the genus Typhobia in reality
represents the Jurassic form Purpuroidea.

LAKE TANGANYIKA—AN OLD JURASSIC SHA. 319

fulfilled its purpose if it acts merely as a sort of counterpoise
to the altogether disproportionate importance which has been
attached to the apparent similarity between the Paramelania
of Lake Tanganyika and the Pyrgulifera of the upper chalk.

Whatever opinion those competent to judge may form of the
comparisons which I have just instituted between the marine
Jurassic and the halolimnic faunas, it is obvious that these com-
parisons are nothing like so rash an undertaking as that
attempted by the three authors I have named. The view that
the Tanganyika fauna corresponds to a fresh-water cretaceous
stock rests on nothing but the similarity of a single type of
shell common to the halolimnic and cretaceous fresh-water
series ; and, as we have seen, the possibility of even this single
point of similarity being due to anything more than mere con-
vergence of external form has been rendered so extremely
doubtful by the more extended observations which one of
the authors named has already made, that any attempts
to pursue the question further would be simply waste of
time. Even if the far more weighty evidence for the cor-
respondence of the halolimnic fauna with that of the Jurassic
seas rested solely on the similarity of their respective shells,
although such evidence would be as good as that forthcoming
for many sweeping geological deductions, I should, for my part,
be highly sceptical that it afforded any trustworthy indication
that the hypothesis is true. When, however, we view the
supplementary facts of this comparison, when we regard it in
the light of what I have ascertained respecting the distribution,
and especially the comparative morphology of the halolimnic
forms, it is very clearly apparent that the theory of their
similarity is not without much collateral support.

We know now that the morphological characters of the
halolimnic fauna are those of an early oceanic stock, that they
do not stand midway between the living fresh-water faunas and
their marine beginnings, for they do not foreshadow any known
fresh-water types; on the other hand, we have seen that they
do very distinctly foreshadow many living oceanic types, each
individually uniting the characters of several modern oceanic

320 J. BE. S. MOORE.

species. We are sure, therefore, that the halolimnic group
represents an old sea stock that became detached from the
general oceanic fauna of which it was a part, far back in time.
Like the Oolite molluscs, those of this halolimnic fauna have
a striking type of shell, and when, after reviewing the facies
presented by the marine fauna of the successive geological
periods, we find such types represented abundantly nowhere
except in the Jurassic seas, and that these seas present forms
corresponding to them all, the comparison appears to be some-
thing more than a mere coincidence. It rather appears as the
fulfilment of an expectation raised simultaneously by the three
chief lines of search relating to their distribution, morphology,
and affinity with existing types.

I offer this comparison, therefore, as the probable explana-
tion of the singularly interesting problem presented by the
mixed fauna which Tanganyika now contains, and I have all
the more confidence in so doing since much study of the
question, in the light of every suggestion which I could either
invent or borrow, has convinced me that no other even mo-
meutarily tenable explanation is likely to be found.

DESCRIPTION OF PLATE 23,

Illustrating Mr. J. E. 8. Moore’s paper ‘On the Hypothesis
that Lake Tanganyika represents an Old Jurassic Sea.”

Fie, 1.—Front and back view of the shell of Paramelania Damoni
(Smith), Tanganyika.

Fig. 14.—Front and back view of the shell of Purpurina bellona (Hudl.),
the corresponding Jurassic form.

Fic, 2.—Front and back view of the shell of Nassopsis nassa (Smith),
Tanganyika.

Fie. 24.—Front and back view of the shell of Purpurina inflata (Hudl.),
the corresponding Jurassic form.

LAKE TANGANYIKA—AN OLD JURASSIC SHA. B yal |

Fic. 3.—Front and back view of Bathanalia Howsei (Moore), Tan-
ganyika.

Fie. 8a.—Upper figure, back view of Amberlya, sp.? from the Inferior
Oolite, British Museum ; lower, front view of Amberlya, sp. ? from the Lias,
the corresponding Jurassic forms.

Fig. 4.—Front and back view of Spekia zonatus (Woodward), Tanganyika.

Fig. 44.—Front and back view of Niridomus minutus, var. tumidulus
(Phill.), the corresponding Jurassic form.

Fic. 5.—Front and back_view of Limnotrochus Thompsoni (Smith),
Tanganyika.

Fic. 54.—Front and back view of Littorina sulcata, the corresponding
Jurassic form.

Fie. 6.—Back view and base of Limnotrochus Kirkii (Smith), Tan-
ganyika.

Fic. 64.—Back view and base of Xenophora (Onustus), from the Inferior
Oolite.

Fic. 7.—Back view of the shell of Pyrgulifera-humerosa (Meek), from
the fresh-water deposits of the upper chalk.

a cate
i és

Tid Ree Tere |

>

ON THE RENO-PERICARDIAL CANALS IN PATELLA. 320

On the Reno-pericardial Canals in Patella,

By

.

Edwin 8. Goodrich, B.A.,
Aldrichian Demonstrator of Comparative Anatomy, Oxford.

With Plate 24.

Srranee indeed, and happily unique in the annals of
comparative anatomy, has been the history of our knowledge
of the reno-pericardial canals of Patella. Although discovered
more than thirty years ago, and investigated by many
observers since, not only is their structure insufficiently known,
but their very existence has been called in question, and even
positively denied !

Wishing to find out definitely whether these ducts really
existed or not, I undertook this work, which was carried out
in Oxford, on material obtained from Plymouth and Naples.
In this short paper I hope to establish clearly, and beyond the
possibility of doubt, the fact that there are reno-pericardial
canals leading from the pericardium to the right kidney and
to the left kidney in Patella.

A communication between the pericardial and the renal ce-
lom of Patella vulgata was originally described by Professor
Lankester in 1867. ‘“‘ By most careful dissection,” he tells us,
“Dr. Rolleston and myself detected what appears to be a
minute opening from the pericardium into the supra-anal
reticulated sac lying in the curve of the rectum [left kidney].
The orifice I found first by opening the pericardium, when it
was seen between the bifurcation of the auricle at the right
side of the cavity, and was then traced from both the peri-
cardium and supraanal sac in other specimens.” Dr. von

324 EDWIN S. GOODRICH.

Jhering ten years later, in an important paper on the kidneys
of molluscs, states that he was unable to find a reno-pericardial
communication : “ Die Pericardialoffnung sah ich nicht” (1877,
Jhering). In 1881 Lankester and Bourne together reinves-
tigated the question, and described the canal thus :—“ Its
presence can be demonstrated both by injections which pass
from the pericardium sometimes into the right, sometimes
into the left renal sac, and by dissection. The orifice leads
directly into a narrow subanal tract of the further or
right renal sac, and not directly into the left or small renal
sac”’ (1881, Lankester). It will be seen that, curiously enough,
although the presence of the canal previously described as
leading into the left kidney is not actually denied, yet the author
seems not to be convinced of its existence. Shortly after, Mr.
J. T. Cunningham undertook the study of these canals by
means of series of sections (1883, Cunningham). In this paper,
to the details of which we shall refer later on, two pericardial
canals are described leading into the small left and large right
kidneys respectively in Patella ceerulea. The fact that only
a diagram of the canals is given, and that Cunningham made
use chiefly of injected material, perhaps contributed to weaken
the evidence brought forward. The main facts were, however,
confirmed by Mr. Harvey Gibson in his studies on the anatomy
of Patella vulgata (1887, Gibson). Animportant memoir on
the kidneys of the Prosobranchs was brought out by M. Rémy
Perrier in 1889; in this work the author states that although
he made use of sections, and found the right reno-pericardial
canal, yet he was unable to find a canal opening into the left
kidney: “ Je n’ai pu retrouver la communication du péricarde
avec le rein gauche.” Perrier concludes that the right kidney
alone communicates with the pericardium (1889, Perrier).

We now come to the most sensational chapter in our story.
In 1892 Dr. R. von Erlanger published an elaborate work on
the ‘Paired Nephridia of Prosobranchs’ (1892, Erlanger), in
which he positively denied the existence of any reno-pericardial
duct in Patella. The author, criticising the injection method,
maintained that previous observers had been misled by the

ON THE RENO-PERICARDIAL CANALS IN PATELLA. 325

injection being forced through weak spots where the kidneys
reached the wall of the pericardium. In conclusion, von
Erlanger stated that in Patella there is “no reno-pericardiac
duct whatever.”

The absence of the communicating canals seemed now to be
as firmly established as their presence had appeared to be but a
short time before. The matter was not long allowed to rest in
this condition. Hardly had naturalists become reconciled to
v. Erlanger’s view, when Dr. Béla Haller published some elabo-
rate studies on Prosobranchs (1894, Haller), in which he
describes in considerable detail a right reno-pericardial canal in
Patella magellanica. A dissection is figured showing the
apertures of this canal. As for the left canal, Haller denies its
existence: ‘“‘ Wie wir wissen, hat Cunningham auch fiir die linke
Niere eine pericardiale Mindung behauptet, darum war ich,
obgleich dieses mir nach dem Verhalten bei der Monobranchen
hdchst unwahrsheinlich vorkam, doch bemiiht dieses unbe-
fungen zu verfolgen. Auf Totalpraparaten war dieses in Folge
der subtilen Verhaltnisse nicht recht moglich und darum
beniitzte ich hiefiir meine Querschnittserien, doch konnte ich
bei keiner der untersuchten Formen eine Miindung der linken
Niere in das Pericard auffinden. Eine solche fehlt ganz ent-
schieden.”

Having thus briefly reviewed the history of the subject, I
must now give a short account of my own observations, which
are founded on the examination of complete series of transverse
sections. The structure and relations of the small left and
large right kidney are now so well known that they need not
again be mentioned. I shall, therefore, merely describe the
selected sections figured on Pl. 24.

In fig. 1 is represented a section through the two kidneys,
rectum, and pericardium, some little way behind the posterior
limit of the mantle chamber. It will be seen that from the
right ventral corner of the pericardium proceeds a diverticulum,
which, in fact, is the beginning of the right reno-pericardial
canal. A section taken farther forward (fig. 2), so as just to
cut through the hinder region of the mantle cavity, shows the

voL, 41, PART 2.—NEW SERIES. Z

326 EDWIN S. GOODRICH.

right reno-pericardial canal separated off from the pericardium,
and lying close to the wall of the subrectal portion of the
right kidney. From the pericardium a second diverticulum is
seen coming off at a higher level than the first,—it is the begin-
ning of the left reno-pericardial canal. If we compare this
figure (fig. 2) with fig. 25 6, pl. xxxvil of Erlanger’s memoir,
we can hardly doubt but that he actually figured the origin of
the left canal without understanding its true nature ; for no-
where else does the right wall of the pericardium become
folded or pushed out at this level as v. Erlanger represented.
Neither of the canals opens straight into the renal celom,—on
the contrary, they bend forwards and extend along the walls
of the kidneys for some considerable distance. The right
reno-pericardial canal is especially long. If we follow the left
canal to about one third of the way between its place of origin
and the external aperture of the left kidney, we find that it
opens into this kidney (fig. 3) through its left wall. Tracing
out the right or lower canal farther forwards, we find it opening
into the subrectal region of the right kidney, about two thirds
of the way from its origin to the right renal pore (fig. 4). In
both cases the reno-pericardial opening is situated at the end
of a papilla projecting into the renal ccelom, and forming a
ciliated funnel-like spout.

The excretory epithelium of the kidneys (fig. 9) is formed of
a layer of very tall cells, the free ends of which are much
swollen, and often broken off. They sometimes bear cilia. A
round nucleus is situated towards the base, and outside it are
numerous excretory granules; the swollen distal end appears
almost empty—an effect due, no doubt, to the reagents. At the
rim of the funnel (figs. 6 and 7, and the reconstruction in
fig. 5) this epithelium changes suddenly into ordinary high
columnar epithelium, provided with numerous long and powerful
cilia directed towards the renal cavity. Near the base of the
funnel the ciliated epithelium passes into the flat coelomic
epithelium lining the canal (fig. 8). The pericardial epi-
thelium itself is identical and perfectly continuous with that
of the canal.

ON THE RENO-PERICARDIAL CANALS IN PATELLA. 327

Comparing this description with that given by previous
observers, it may be remarked that in the main my results
agree with and confirm those of Cunningham; yet neither he
nor Gibson appears to have seen the well-marked funnels.
On the other hand, like Gibson, I am unable to find the
‘triangular piece of tissue” described by Cunningham as
forming “a sort of valve’’ over the opening. It is difficult,
indeed, to see how such a flap would act in connection with the
papilla. Haller describes a ciliated funnel at the right reno-
pericardial aperture without figuring a section through it ; but
he further states that the canal itself is lined with high columnar
cells. This is certainly not the case in the species I have inves-
tigated, and I cannot help thinking that he may have mistaken
in this instance a branch of the ramifying kidney for the reno-
pericardial canal, Both the kidneys give off numerous branches
lined with epithelium similar to that of the main’ renal
chambers.

In the four series of sections of Patella vulgata examined
I have always found the two reno-pericardial canals present,
and well developed. In Patella coerulea I have observed
two canals of essentially similar structure,—in fact, the
description given above applies equally well to either species.

Summary.

In the foregoing pages it has been shown that in Patella
vulgata and cerulea there are two reno-pericardial canals,
opening by means of projecting ciliated funnels! from the
pericardium into the right and left kidneys respectively.

List oF REFERENCES.

1883. Cunnincnam, J. T.—‘ The Renal Organs of Patella,” ‘ Quart. Journ.
Micros, Sci.,’ vol. xxiii, 1883.

1892. Ertancer, R. von.—‘‘On the Paired Nephridia of Prosobranchs,”
‘Quart. Journ. Micros. Sci.,’ vol. xxxiii, 1892.

Similar but better developed structures are found at the mouth of the
reno-pericardial canals of Chiton and other molluscs,

328 EDWIN S. GOODRICH.

1887. Gipson, R. J. Harvey.—‘‘ Anatomy and Physiology of Patella
vulgata,” ‘Trans. Roy. Soc. Edinburgh,’ ser. 3, vol. xxxii, 1887.

1894, Hatter, Bira.—‘ Studien tiber Docoglosse und Rhipidoglosse Proso-
branchier,’ Leipzig, 1894.

1877. Juerine, H. von.—‘‘ Zur Morphologie der Niere der sog. Mollusken,’
‘Zeit. wiss. Zool.,’ vol. xxix, 1877.

1867, LanxesteR, HE. Ray.—‘‘ On some Undescribed Points in the Anatomy
of the Limpet,” ‘Annals and Mag. Nat. Hist.,’ ser. 3, vol. xx, 1867.

1881. Lanxester, EH. Ray.— On the originally Bilateral Character of the
Renal Organ of the Prosobranchia,’’ ‘ Annals and Mag. Nat. Hist.,’
ser. 5, vol. vii, 1881.

1889. Perrier, Remy.—“ Recherches sur l’Anatomie et |’ Histologie du Rein
des Gastropodes Prosobranches,” ‘Ann. Se. Nat. Zool.,’ vol. viiis
1889.

EXPLANATION OF PLATE 24,

Illustrating Mr, Edwin S. Goodrich’s paper on “ The Reno-
pericardial Canals in Patella.”

All the figures refer to Patella vulgata.

Fies. 1, 2, 3, and 4.—Four transverse sections through the region of the
rectum and pericardium, showing the course of the two reno-pericardial canals.
Drawn with the camera. X 9.

Fie. 5.—Reconstruction of the opening of the right canal into the right
kidney.

Fie. 6.—Section through the opening of the right canal into the right
kidney. Cam. xX 180.

Fie. 7.—Section through the opening of the left canal into the left kidney.
Cam. X 180.

Fic. 8.—Section through the wall of a reno-pericardial canal, showing the
flat coelomic epithelium on the inside and blood space outside.

Fic. 9.—Section through the wall of a kidney, showing the large excretory
cells.

JAN 9 1899.

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 9329

The Development of the Pig during the First
Ten Days.

By

Richard Assheton, i.A.

With Plates 25—28.

INTRODUCTION.

Tue history of the development of the roe deer, published
by Bischoff in 1854, contains almost the only account there is
of the early stages of the development of any Ungulate. The
description which that distinguished embryologist was able to
give was, however, somewhat meagre, and does not offer much
assistance towards the solution of problems which occupy the
attention of embryologists of the present day.

He was able to describe the external features of some of the
stages of segmentation, and made the very interesting dis-
covery, that in the deer the embryo, upon reaching the fully
segmented or morula stage, enters upon a period of quiescence,
and remains for some weeks unaltered.

Von Baer (2), who made most careful observations upon the
later stages of development of both the sheep and pig, found
the egg of a sheep in the oviduct as early as the end of the
first day. His descriptions, however, of the earliest stages of
these animals, although historically interesting, are of little
use for our present purpose.

Between 1884 and 1889 Bonnet (10, 11) published his work
upon the sheep, but in this he dealt with nothing earlier than
the twelfth day.

vo. 41, PARY 3,—NEW SERIES, AA

330 RICHARD ASSHETON.

Keibel’s (18, 19) work, which treats of the pig (Sus scrofa
domesticus), contains no reference to earlier embryos than
those of fourteen days after fertilisation. At fourteen days
the embryonal area is well defined and oval; the primitive
streak is distinct, and the whole blastodermic vesicle has
grown to an extraordinary extent, varying from half to a
whole metre in length.

A younger stage of the pig’s development has been described
by Weysse (25), who obtained some thirty specimens of about
the ninth to eleventh days, which he states present “ pheno-
mena unusual in the ontogeny of the Mammalia.”

My specimens, which I propose to describe at the present
time, are of stages prior to any described by the above authors,
and date from the second day after fertilisation to the tenth.

They include the earliest phases of segmentation, the forma-
tion of the blastodermic vesicle, and the definite establishment
of the epiblast and hypoblast layers of the future embryo.

Method of treating the Specimens.

The great difficulty of finding the embryos has made it
almost impossible to examine many in the fresh state. The
size of the cavity of the uterus of a pig is so great, and the
foldings of the mucous membrane are so complicated, that
a search to find them in situ proved most unprofitable.
Accordingly, with very few exceptions, the whole of the
material upon which I have worked was preserved in the way
I am about to describe before any examination of it was made.

The ova, which, while in the Fallopian tube, can of course
be very easily found, unfortunately in the pig travel very
rapidly down the Fallopian tube, and pass into the uterus
during about the third day, before segmentation has proceeded
further than the four-segment stage.

The ova are extremely small, and receive no additional
investment from the walls of the oviduct, and are quite
invisible as they lie among the folds of the uterus wall. By
scraping the mucous membrane of the uterus a specimen may
occasionally be procured, but by far the most satisfactory

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 391]
method of procedure is the slow injection of the cavity of the
upper part of the uterus with some preserving fluid, and a
subsequent search through the contents under the microscope.

Within a space of time after the death of the sow, varying
from three quarters of an hour to three hours, I very slowly
injected ‘25 or ‘5 per cent. chromic acid into the upper half or
third of the uterus, having previously tied up the upper end
of the piece of uterus. I allowed as much fluid to flow in
as was possible without running the risk of bursting the
uterus, in order that the folds of the mucous membrane should
be reduced toa minimum. In this way the ova were caused
to float freely in the fluid.

The lower end, through which the injection had been made,
was then ligatured, and the whole was placed in ‘5 per cent.
chromic acid for two or more days. The contents of the
uterus were then let out into glass vessels, and the specimens
were easily found under the microscope.

Chromic acid used in this way was found to be the most
satisfactory, on account of its clearness and the absence of
any precipitate.

In a few cases I used Flemming’s solution with equal success ;
but other fluids, such as Perenyi’s, or solution of nitrate of
silver, were unsatisfactory, owing to the great precipitation
which occurs rendering the search for the embryos too
laborious.

All my specimens were stained before they were cut into
sections. The stains used were hemalum, carmalum, hema-
toxylin, and borax carmine. They were embedded in paraffin
by means of cedar oil.

Material.

The following account is based upon the examination of
about 100 embryos obtained from sixteen sows. In about ten
other sows I failed to find any specimens. The majority of
those I have no doubt were really pregnant. My specimens
may be tabulated thus:

332 RICHARD ASSHETON.

List of embryos obtained arranged according to age :

Under 4.days . j . 2 | From7to 8days . 6: Adib
From 4to5days . . 28 aye, ee BOLD ase ‘ «i val
”? 5 ” 6 22 O 2 9 oe) 9 ” 10 2 7 . 10
ss Os has : ~ a4 roe UU airs a Rs a hg

List of embryos arranged according to stage of develop-
ment:

A. Prior to the 5-segment stage . é » 18
B. The 5-segment stage to morula, with suiare Ar ae
mately equal ; : 12
C. Commencement of pletedecaia aide to lalaest intye
zona . . : . 33
D. From rupture of zona ate Papbatte of Ranber uae : o20

E. Formation of embryonal area to first sign of primitive streak 22

Embryos of Stages A and B.

The dates which are given must be regarded as approximate
only.

In the pig it would seem that fertilisation may occur about
the end of the first day, or may be postponed two or three
days after union.

I have made no observations upon the process of fertilisa-
tion, and merely draw the above conclusions from finding
specimens in identical conditions in pigs killed upon the
fourth, fifth, or sixth day. It is accordingly very difficult to
obtain any given stage.

Embryos frequently differ very greatly in the same uterus.
For instance, in one case I had ova in two segments, nine
segments, and completed morule from the same horn of the
uterus.

The youngest stage that I have is a two-segment stage
(Pl. 25, figs. 1 and 2). This was taken from the uterus of a
pig on the sixth day. It was preserved in Flemming’s solution
(weak formula). The embryo is protected by a thin zona

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 833

radiata, which is not thickened by the addition of any mucous
or albuminous layer.

The slight and very thin irregular external coat to the zona
radiata is seen upon higher magnification to be caused by
numerous spermatozoa, which seem to have adhered to the
zona radiata,

The embryo lies freely within the zona radiata. Both
segments appear plainly divisible into two parts, an outer
clearer layer which extends round the greater part of an inner
much darker part, within which lies the nucleus.

The outer layer, which shows a very fine reticulation or
granulation, is only very slightly darkened by the fixing solu-
tion. On the other hand, the inner portion is very much
blackened, owing to the action of the osmic acid upon the
numerous oil globules whose presence characterises this part
of the segment.

In a fresh specimen these globules are rather dark and of an
olive-green colour. Ordinary stains have no effect upon them.
Osmic acid is the only reagent that I have found to colour
them. In this respect, among others, they differ very much
from the yolk granules of reptilian, avian, and elasmobranch
eggs.

The nucleus, like the outer layer, has none of these oil
globules, and so the centre of each segment appears more
transparent. The nucleus of each of the segments of this
specimen is distinctly visible, and is spherical.

Two polar bodies are present.

Fig. 2, which is a side view, shows very distinctly the rela-
tive extent and position of the outer clearer zone and the
inner oil-bearing mass.

Fig. 3 was drawn from a fresh specimen taken from the
upper end of the uterus within 10 mm. of the opening of the
Fallopian tube. It was found with three other specimens, two
of which were in the four-segment stage, the other in five
segments.

The nuclei could not be distinctly seen, but their presence was
indicated by the slightly greater transparency of the central

304 RICHARD ASSHETON.

parts of each of the three segments. The same characters of
clear and darker portions are recognisable as in the two-
segment stage.

Fig. 4 is a drawing from a fresh specimen taken from the
Fallopian tube of another sow. This and one other, which
was also in the four-segment stage, are the only two embryos
I have ever obtained from the Fallopian tube of the pig.
The same features were present as regards inner and outer
portions of each segment.

The four segments were not of equal magnitude.

The following measurements were made while the specimen
was quite fresh :

Diameter of the whole specimen from outer

edges of zona radiata : : . ‘164 mm.
Thickness of zona radiata . s : + (OLGex,
Diameters of the largest sphere : . 064 4, x *066 mm.
Diameters of the smallest sphere. - (060) 35) >< s0O0ke:

These two segments formed one pair in contact with each
other. At right angles to this pair was the pair formed by the
other two segments, which were of equal size and did not
quite touch each other.

Diameters of each . . : : - ‘060 mm. X ‘064 mm.
The diameters of one polar body were - “OlLSS 4s, Sor OlOme
The second polar body was too indistinct to be measured.

I was not able to detect any difference in the colour or in
the composition of the several segments, either while fresh or
after the application of osmic acid and stains. There would
seem to be distinct difference in size.

The following measurements from another specimen made
after fixation give similar results.

Pair in contact :

Diameters of the larger segment : . ‘056mm. x °058 mm.
Diameters of the smaller segment. = 006) 5, De. oa,

Pair widely separated and at right angles to the preceding :

Diameters of larger segment . : . °050 mm. X :061 mm.
Diameters of smaller segment . : DA 50 xX S0D8 ae

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 335

The polar bodies of this specimen were especially large.
They were spherical, and measured ‘025 mm. and ‘014 mm.
respectively.

In the five-segment stage (fig. 5) there is one segment dis-
tinctly larger and one distinctly smaller than all the others.
The remaining three are approximately equal. There is no
means of determining to which generation the segments re-
spectively belong. It is, however, interesting and important
to note that at this early stage there is a very great difference
in size between the several segments.

The next stage I have is one of six days three hours.
I had several in this stage, all of which had been treated
with Flemming’s fluid, and were therefore considerably
blackened, owing to the action of the osmic acid on the
oil vacuoles. This makes it difficult to determine the
number of segments in surface view. The chief feature of
all these specimens was easily discernible. This was the
presence of one segment very much larger than any of the
others.

One specimen was stained in carmalum and cut into sections,
a drawing of one of which is given as fig. 14, Pl. 26. In this
specimen there are altogether nine segments. One of these is
far larger than any of the others. The nucleus of this seg-
ment is in an early stage of mitosis. The section I have
drawn, however, does not pass through it. The remaining
eight segments are of very different sizes. The nucleus in
each is in its resting condition.

The specimen of which fig. 12 represents a section was, when
seen in surface view, apparently in a similar stage to that
represented by fig. 138. It seems, at any rate, possible
that it was really in the two-segment stage, and that one of
the segments had by some means burst.

The section which I have drawn shows one very large and
complete segment, and partially surrounding this there is a
mass which contains not a trace of nucleus nor any arrange-
ment of the oil globules comparable to those of the large intact
segment or to those of fig. 13. If this is really the case,

336 RICHARD ASSHETON.

the large segment will represent very well the more minute
structure of the embryo when in the two-segment stage, as in
figs. 1 and 2.

Towards the periphery the protoplasm is very finely reticular,
whilst more centrally the meshes are much larger. It would
seem really to be vesicular, the vesicles being filled with oily
“yolk,” as mentioned above.

The less vacuolated cortical layer makes inroads at two
points into the central portion. The nucleus is spherical and
sharply defined by a very deeply staining membrane. A plate
of very fine chromatin granules passes through its centre.
There are no oily vacuoles in the nucleus.

In fig. 6 I give a drawing of a rather more advanced embryo.
In this there were about twelve to fourteen segments. There
was a very great disparity in the size of the several segments.
Such a stage may be easily derived from the foregoing condi-
tion of fig. 13. The large segment has divided, as well as
several of the smaller ones. This specimen, together with
the next one which I shall describe, and those of which figs.
1 and 2, 12 and 13 are drawings, were all obtained from the
same horn of a uterus. Their age was six days. The same
uterus contained embryos in which the cavity of the blasto-
dermic vesicle was established.

Fig. 8 represents a specimen which presented a unique
stage. Segmentation had proceeded very much further than in
the preceding case. At one point a single large segment
protruded beyond the other. Over the rest of the embryo
the outlines of the segments formed tolerably regular figures,
showing that the segments on the surface were approxi-
mately uniform in size and somewhat compressed by their
surroundings.

This is the only specimen I have found in this condition.
It was unfortunately lost.

This stage is succeeded by one which may be described as
a typical mammalian morula (fig. 7). In surface view the
segments present no great disparity of size. The embryo as a
whole is spherical. A series of sections shows that there is a

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 387

very considerable difference in the minute structure of the
segments as compared with an earlier stage.

Although the oily ‘‘ yolk ” vesicles are still present, but in
diminished numbers, there is certainly a tendency to the
formation of other spaces or vacuoles. I cannot say whether
these spaces are the commencement of the cavity of the blasto-
dermic vesicle or not.

Fig. 14 is a median section through a specimen such as
that represented by fig. 7. It was taken from the uterus at
six days.

It is very difficult to detect in sections any boundaries
between the cells. It would not, however, be right to say
that the embryo is a vacuolated protoplasmic mass containing
nuclei, for surface-view specimens treated with silver nitrate
show the lines of division between the cells very clearly. Here
and there, also, there is in some of the sections a trace of a
cell boundary, more especially where it would mark off a peri-
pheral from an inner mass cell.

There is also a distinctly radiate arrangement of the proto-
plasm round certain centres, by which the sections (v. fig. 14)
are marked out into cellular areas. The arrangement of the
nuclei round the periphery should be noted.

This concludes all the evidence I have to offer with regard
to the process of segmentation of the ovum in the pig.

The most noteworthy feature is the very great dissimilarity
of size of the segments after the four-segment stage. I un-
fortunately have only one certain specimen in the two-segment
stage. In this I can see no difference in size or nature
between the two.

In the four-celled stage there is always a slight difference
in size, but at the eight- to twelve-segment stage the difference
is most marked, and recalls the great difference which we find
in many molluscan ova. In no case, however, have I found
any difference perceptible except in size. The smaller seg-
ments form a cap upon the larger.

This great inequality is more marked than in any other
mammalian segmentation. Bischoff shows nothing like it in

338 RICHARD ASSHETON.

his figures of the segmenting ova of the dog, or guinea-pig, or
deer, and only in a less degree in the rabbit. Van Beneden
(4) for the rabbit described a cap of smaller segments, which
ultimately surround a group of larger cells. I was ina former
paper unable to offer evidence in support of this. Heape (15)
found nothing of the kind in the mole.

When, however, we take into consideration the recent
account given by Duval (13) for the bat, and Hubrecht (16)
for Tupaia, of a growth of slightly smaller cells round a core of
slightly larger cells, which is made much more obvious by
reason of a difference in the affinity to stains of the two groups
of segments, and an account of a similar process in the sheep,
which I hope to give in an accompanying paper, together with
the original description of the rabbit as given by Van Beneden
(although subsequent observers have been unable to confirm
his statements), it is clear that there is much evidence in
favour of this original discovery of Van Beneden ; and in the
present case I have to consider whether the specimens I have
described above of the segmentation in the pig may indicate
an identical wrapping round one group of segments by another
or not. It seems extremely probable that there is such a pro-
cess. Its significance, however, I have discussed in my paper
on the Development of the Sheep (1a), in the last number
of this Journal, whose development I believe shows that it is
really the hypoblast which grows round the epiblast, and not
vice versa, as Van Beneden supposed.

Formation of Cavity of Blastodermic Vesicle to
Rupture of Zona radiata—Stage C.

The surrounding membrane consists still of the zona
radiata, which is becoming slightly stretched (fig. 15). In this
figure the vacuolation seems to be more pronounced at one
part than elsewhere.

I am inclined to think that this more intense vacuolation
eventually leads to the origin of the cavity of the blastodermic

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 339

vesicle, which soon after becomes established. If this is so, the
embryo in this particular behaves like that of the rat (Robinson
[23], Duval [12], or Tupaia javanica, Hubrecht [17]),
whereas in forms like the rabbit and mole the blastodermic
vesicle cavity arises as a clean cleft between the several seg-
ments, which are always more sharply separated than in the
case of the former animals. In some instances I have seen
strands of protoplasm connecting the inner mass with the outer
wall of the ab-embryonic pole after the cavity has attained as
great a size as in fig. 16.

Small spherical globules can often be found within the
cavity which have the appearance of being oil drops liberated
during the process of vacuolation that has resulted in the
formation of the cavity of the blastodermic vesicle.

In fig. 15 the cavity is seen distinctly in two places. In
fig. 16 the cavity is large and eccentrically placed. The
embryo may be described as a spherical hollow ball, the wall
of which is much thickened at one spot. There is as yet no
distinct inner mass clearly divided from the outer wall.

As the embryo grows and the blastodermic vesicle cavity
enlarges, the innermost of the nuclei of the thickened portion
of the wall become, with their surrounding protoplasm, more
and more free and more rounded; while the outer layer, which
is more directly connected with the remainder of the wall, is
found to be clearly separated off from the former, and it
becomes alone directly continuous with the outer wall of the
vesicle (compare figs. 17 and 18). These changes occur
very rapidly, and without much increase in the number of
cells.

The oily yolk globules are still present in little groups in
the meshes of protoplasm, chiefly round the nuclei. The
whole protoplasm is very much vacuolated, and retains this
condition until after the rupture of the zona radiata.

The inner mass consists now of very few cells. In one
specimen, in size between those shown in figs. 18 and 19, the
inner mass was completely flattened up against the outer wall,
so that the embryonic pole of the blastocyst was not more

340 RICHARD ASSHETON.

than twice as thick as the ab-embryonic pole, instead of being
six or seven times as thick as it is in fig. 17.

Pl. 25, fig. 9,is an outline drawing of the embryonic pole of
this specimen. The nuclei of the outer layer are shaded with
lines, the nuclei of the inner mass with dots. The solitary.
nucleus of the outer layer which overlies the inner mass is
smaller than the others. The inner mass is there seen to con-
sist of six cells only.

Boundaries of cells cannot be distinguished at this stage,
except in silver nitrate specimens.

I have obtained a large number of specimens between the
stages represented by my figs. 16 and 20. In these I find a
great inconstancy in the relative size of the inner mass to the
outer layer. Although the inner mass may consist of as few
as six cells, it is more usually of greater size than this. I
cannot say whether there is a reduction in the number of
cells of the inner mass after it has become separated off
from the outer layer.

Heape (14) described an occurrence of this nature in the
mole, and assumed that cells passed from the inner mass into
the outer layer.

Just before the destruction of the zona radiata comes about
the inner mass shows a marked increase in activity, and
increases in size and becomes more compact (figs. 19 and 20). -

I have not seen in any specimen anything to indicate, before
the rupture of the zona radiata, which cells will become the
definite epiblast and which the hypoblast. The diameter of
the whole blastodermic vesicle just before the rupture is
about ‘15 mm. Until the rupture the blastocyst is spherical.
I cannot say how the rupture is effected. The zona radiata
becomes exceedingly attenuated, and I am inclined to think
that it becomes torn and broken in many places, and is not
thrown off in large pieces.

Unless I have missed the intermediate stages, the dissolu-
tion of the zona radiata is accompanied by very marked
changes in the embryo. It loses its spherical form, and is
no longer tensely distended. The nature of its cells is

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 341

changed. Previously they contained many vacuoles, but now
they are without vacuoles, and far more homogeneous in
their minute structure. There are no longer any thinly
drawn-out portions such as can be seen in fig. 19.

The inner mass has undergone a complete change. Instead
of being a compact lenticular mass, it is now divided sharply
into (a) a small group of cells, very often lying quite separated
from the outer layer of cells, (6) a loose layer or network of
cells lying apparently but very slightly attached to the inner
surface of the group (a), and extending beyond its limits on
to the “ outer layer” (figs. 21 and 24). -So slight is the con-
nection between (6) and (a), that very often the two appear to
be entirely separated (v. figs. 22 and 23, 25 and 26). It seems
probable that the conditions of figs. 21 and 24 are normal,
and that the separation of the two layers seen in figs. 22, 23,
and 25, 26, may be due to the disturbances during preserva-
tion. It may therefore be accidental, but the specimens which
show the separation were to all appearances quite perfect when
they were examined after having been stained and cleared.
In any case it seems to point to a very slight connection
between those two layers. This is curious because in the
immediately preceding stages (figs. 19, 20) there is hardly
anything which indicates it. Fig. 19 shows a whiter core, Z.,
which may perhaps suggest such a separation.

I shall throughout this paper talk of these three separate
cell areas as trophoblast, epiblast, and hypoblast, but in
adopting Hubrecht’s terminology—which, as he remarks, is
extremely convenient—I do not adopt the conclusions to
which he comes regarding the homology of the trophoblast as
enunciated in his work, ‘ Die Phylogenese des Amnions und
die Bedeutung des Trophoblastes.? In the dissertation fol-
lowing my description of the sheep’s development (la) I give
reasons for regarding the trophoblast as really part of the
hypoblast. So we may describe the embryo, shortly after
the loss of the zona radiata, as a closed vesicle of very varied
shape, whose wall is formed by a single layer of cells which are
nearly cubical in shape. This layer is the trophoblast. At one

342 RICHARD ASSHETON.

point on the inner surface a small group of cells is loosely
attached to it, which is the epiblast, and lying along the inner
surface of the epiblast is a loose network of irregular-shaped
cells, the hypoblast.

In figs. 21 and 24 the hypoblast (H.) forms a continuous
layer beneath the epiblast (#.). I do not find that the embry-
onic hypoblast in the pig has a later origin from the inner
mass than that which the extra-embryonic hypoblast has in
the way described by Hubrecht (17) for Tupaia.

The smallest specimen which I have, from which the zona
radiata has gone, is one (figs. 22, 23) in size intermediate
between fig. 19 and fig. 21. Its anatomy is similar to the
latter specimen.

The increase in size between the stages of fig. 19 and fig.
22 or 23 is by no means great, which leads me to suppose
that the latter specimens are only slightly advanced on the
former.

The epiblast cells in the specimen represented by figs. 22,
23, number twelve or fourteen, and the hypoblast cells seven.

The actual ‘“‘ ages” of the specimens, figs. 18—24, are as
follows :

Fig. 18 was. ‘ : ‘ 2 5 days.
99 1) 39 5 99
2” 20 ” ° D9
Site 4 eras : 4 ; : adr

Figs. 22 and 23 were. ; ‘ {he aaes

Fig. 24 was ; (eer

We may say that, as a rule, the zona radiata is lost during
the seventh day, and on the same day the separation of the
inner mass into epiblast and hypoblast occurs. Even if this
separation is not so sudden as I have supposed, there can be
no doubt that the two inner cell layers are derived from the
inner mass of figs. 19 and 20.

From this time—that is to say, during the eighth and until
the closing hours of the ninth day—the shape of the blasto-
dermic vesicle becomes extraordinarily irregular.

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 3483

The degree to which the embryos become crumpled is rather
interesting. No doubt the rupture of the zona radiata very
greatly affects the osmotic properties of the walls as a whole.
Up to this time it would seem that the accumulation of the
fluid within the cavity of the blastodermic vesicle was in excess
of the growth of the cell walls, and so caused a great tension
to be placed upon them. Now it seems that the accumulation
is unable to keep pace with the growth of the walls, and con-
sequently the crumpling which they undergo is quite extra-
ordinary. It reaches its maximum about the middle of the
ninth day, after which the fluid gains upon the excess of wall
growth, and so the vesicle becomes again distended, and by
the eleventh day the creases have to a large extent become
smoothed out.

During the eighth and ninth days a fluid is poured out into
the cavity of the uterus from the cells lining the cavity of the
uterus and the glands, which contains many corpuscles, which
may often be found adhering to the walls of the vesicles,
These corpuscles are clear and colourless, and do not readily
take stains, The amount of this fluid is, however, slight in
comparison to that poured out into the cavity of the sheep’s
uterus.

Eighth and Ninth Days—Stage D.

The changes that occur during the eighth and ninth days
are not very remarkable. They are chiefly visible in the large
increase in size of the whole blastodermic vesicle, in the
spreading of the hypoblast layer, and consolidation of the
epiblast.

The hypoblast at the moment of its first separation from
the inner mass forms a kind of network, which lines at first
the embryonic pole only of the blastocyst (fig.21, H.). During
the eighth and ninth days it extends its borders, and before
the end of the eleventh day it practically lines the entire
blastocyst. At this time the part lining the lower pole of the
blastocyst is still a network, while at the embryonic pole it is

344, RICHARD ASSHETON.

a continuous membrane (H, figs. 86—42). On the earlier
parts of the eighth and ninth days it is, however, a network
over all parts, and is still absent from the lower pole. Whether
this extension is effected by an active advance of its edge, or is
only an apparent advance due to the more active growth of the
circum-embryonic part of the trophoblast, as I believe to be the
case in the rabbit, I cannot say. During these two days the
blastocyst increases in size from about ‘2 mm. to about 15 mm.
The chief interest undoubtedly lies in the fate of the epi-
blastic portion of the inner mass and the part of the tropho-
blast which at first overlies it and is quite distinct from it.

Fig. 27 may be taken to represent quite typically the
condition of things at the end of the eighth day. The tropho-
blast cells are cubical with clearly marked boundaries, both in
section and in surface view. There is no change observable for
some days in the nature of these cells beyond the limits of
the embryonic epiblast.

In that part of the trophoblast which covers the epiblastic
portion of the inner mass the following changes occur con-
currently with certain changes in the epiblast. I will describe
the two together.

The diameter of the epiblast—that is to say, of the embryonal
area—as seen in surface view increases from about ‘04 mm. to
°25 mm. As it grows it becomes more firmly pressed against
the trophoblast, which at this spot becomes distinctly thinner
and more compressed, and its nuclei become much flattened
(fig. 28).

The epiblast begins to show signs of more active growth,
and the cells which at first were loosely arranged and equal in
size, and showed no sign of differentiation, may now, in some
cases, be seen to be arranged in an inner layer of small nuclei
lying next to the cavity of the blastodermic vesicle, and a layer
of very much larger nuclei between this and the trophoblast
layer (fig. 29).

This inner Jayer of smaller nuclei, which also stain rather
more deeply, is so very distinct in one of my specimens which
I obtained from a sow killed at eight days, that I thought it

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 345

possible that this internal layer might split off and give rise
to the embryonic hypoblast in the way Hubrecht describes for
Tupaia.

As somewhat supporting this view I may draw attention to
the fact that the hypoblast is extremely attenuated, and very
reticular in the region of the embryonic area at this time. I
cannot positively assert that the hypoblast in many of my
specimens can for certain be seen lying against the region in
question.

Also in the same sow there were two other specimens in
which the epiblast seemed to be reduced to little-more than a
single layer of large cells (or rather nuclei). The small nuclei
were, however, nowhere to be seen, and the hypoblast was not
in any degree more evident.

As militating against this view is the fact that in such speci-
mens as that drawn for fig. 28, the whole mass (Z.) here is un-
doubtedly epiblast. This is proved by the presence of the thin
membrane (JZ.H.L.), as will appear in the description of the
older stages. In this case there is also a tendency to the for-
mation of an inner layer of smaller nuclei. This is probably
an older stage, its age being nine and a half days, but the
hypoblast here is by no means a continuous layer, such as it
ison the next day. A similar layer of small nuclei is also
plainly visible long after the hypoblast has become a well-
developed layer on the twelfth and thirteenth days. So,
although there are certain features which suggest a further
splitting off of the embryonic hypoblast comparable to what
Hubrecht has described for Tupaia, yet, from a consideration
of the evidence on both sides of the question, I am of the
opinion that this phenomenon does not occur in the pig, but
that the whole hypoblast is split off at the same time immedi-
ately after the rupture of the zona radiata ; and that the con-
dition shown in my figure is comparable to the condition shown
in Hubrecht’s (17) figures 57 to 59, or my figures of the sheep,
and have nothing to do with the hypoblast formation.

vo. 41, PART 3,—NEW SERIES. BB

346 RICHARD ASSHETON.

The Rupture of the Rauber Layer. The Tenth and
Eleventh Days.

I have now reached the stage of which we have recently had
a description from Weysse (24). In this paper Weysse
described a very remarkable overgrowth, which, starting from
the edges which he assumed to be the posterior and lateral
margin, grew forwards over the embryonal area.

According to his description both the embryonic epiblast
and the trophoblast cells were concerned in this process.

It must be remembered, however, that Weysse did not
regard this as the real trophoblast, but as the definite extra-
embryonal area epiblast. He thought that he had evidence of
a third layer, the true Rauber layer, outside this again. On
this I shall have some more remarks to make.

The overgrowth, according to him, formed a complete cover-
ing, or, as he termed it, a bridge, over the hinder part of the
embryonic epiblast. Ultimately the bridge was said to fuse
with the underlying embryonic epiblast, and to give rise to the
embryonic area.

Weysse compares this overgrowth to the apparent growth
of the epiblast over the neural plate from the hinder part of
the embryo of Amphioxus. ‘To this conclusion he seems to
have been brought chiefly by finding in certain sections a very
narrow canal leading from the exterior under his bridge into
the space between the epiblast and the hypoblast, or rather into
the space between the epiblast and a non-cellular membrane,
which is apparently the homologue of that which Schafer (24)
found in the cat, and called the membrana limitans hypo-
blastica. This canal Weysse considered to be equivalent to
the neurenteric canal. The important difference that the sup-
posed canal leads into the space noticed above, and not into
the archenteron, does not seem to have influenced his opinion.

Duval (18) and Hubrecht (17) both comment on Weysse’s
discovery, and each interprets the facts differently.

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 347

Weysse will probably stand alone in instituting a com-
parison with Amphioxus, but as regards the actual facts 1 am
able to support him to a certain extent.

It is true that what I have found is in detail extremely
unlike the majority of Weysse’s specimeus. Yet my own
specimens differ so greatly one from another that it is probable
that almost any form may be assumed by the cell masses in
question.

The course of events would seem to be as follows. During
the ninth day the epiblastic knob is a compact mass of cells,
nearly as thick as itis broad. The inner surface is characterised
by the presence of many small nuclei, while towards the surface
the nuclei are larger (fig. 29). (Compare Hubrecht [17], figs.
57, 59.) These features probably indicate a more rapid
growth of the inner parts of the epiblastic knob. The result
is that the knob acquires a convex surface on its inner face
(figs. 32, 33). An exaggeration of this process might no
doubt lead to a doubling up as in Tupaia or Talpa. Some-
thing of the kind does occur in the pig, but I do not
think it is so evident a phenomenon as it is in those animals.
Figs. 830—33 are drawings of. four sections through one
specimen. There are thirteen sections in all; fig. 30 is the
third. In this there is a distinct doubling up, and the
trophoblast is broken. In fig. 31, which is the fifth section,
the epiblastic knob is curved and bent away from the over-
lying trophoblast, which, however, is not broken. Figs. 32
and 33 are the ninth and tenth sections, and are therefore
near the other end of the series. These show that this part
of the epiblastic knob is not doubled up, but forms a solid
plate, which is not actually fused with the trophoblast above
it. The trophoblast overlying the epiblast is broken in
several places.

Fig. 35, which represents a section of another specimen
from the same uterus, shows a very slight curvature of the
epiblastic plate. The overlying trophoblast is broken, and
pieces of it (7.R.) are to be seen lying upon the epiblast. In
another specimen (fig. 34) there is no curvature at all. The

348 RICHARD ASSHETON,

trophoblast is lying over the surface of the epiblast, and is
broken. The age of these three specimens was nine days
eighteen hours.

I have killed several animals in attempting to acquire a
more perfect series of this stage, but without success. There
is no possibility of obtaining with certainty embryos of a
given age.

If we take the specimens of figs. 30—33 as representing the
ordinary course of development, we see there is evidence of a
doubling up of, at any rate, a certain part of the epiblastic
plate ; but I do not think that it ever amounts to such a com-
plete folding as it does in Tupaia or Talpa. In any case it
is clear that considerable pieces of the ruptured trophoblast
may be left upon and more or less attached to the epiblastic
knob (v. figs. 81—35), which give rise to the conditions
described below, and which, I think, account for the appear-
ances described by Weysse.

There is no regularity in the shapes of the pieces thus sepa-
rated, some more and some less completely, from each other,
and from the edge of the trophoblast adjoining the embryonic
area.

In a certain specimen (fig. 10), the age of which was ten
days two hours, the trophoblast at one spot is seen in a section
to be overlying the embryonic epiblast (fig. 37, 7.R.). The tro-
phoblast stains more darkly than the embryonic epiblast. On
following this overlapping edge through the series of sections,
it is found to run towards the centre of the embryonal area as
a narrow strip quite free from and rising away from it as it
advances towards the centre. At the centre it expands into a
fan-shaped plate, from the floor of which a pillar runs to the
embryonal area below, and is clearly fused with it at this spot
(fig36, 07.)

It is evident that here is a structure even more deserving of
the name of “ bridge ” than the lateral expansions of Weysse,
which may be likened rather to balconies.

Over another part of the same embryonal area are a few
cells which lie more loosely upon the epiblast.

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 349

Another specimen (PI. 25, fig. 12, A) shows a large isolated
fragment not actually fused nor continuous with the edge of
the trophoblast. Fig. 89 represents a section taken through
this specimen. At no point is there any complete fusion
between the fragment and the underlying epiblast, as in the
_ last case (fig. 36).

Another specimen has a long fragment, clearly a continua-
tion of the trophoblast, running across the surface of the
embryonal area (as in the case of figs. 33, 37), but in this
instance it ends freely, and at no point is it attached to the
embryonal area below. In the same section (fig. 39) a single
cell (7.R.) may be seen, which from its appearance, colour,
position, and prominence above the others seems undoubtedly
to owe its origin to the trophoblast layer. Such cells are not
rare but by no means numerous. The number of them com-
pared with that of the cells of undoubted inner mass origin is,
however, quite insignificant.

Now it seems to me that such structures as the above-
mentioned irregular masses, and the appearances described by
Weysse, are to be explained in this way.

After the trophoblast has ruptured during the tenth day, in
the accomplishment of which process I assume that the local
tension produced by the growth and lateral expansion of the
epiblastic mass plays an important part, the ragged edges of
the ruptured trophoblast and the isolated cells of the same
layer are by no means forced to die or to degenerate imme-
diately. On the contrary, they grow and multiply, and grow
into almost any shape, and form balconies, bridges, &c. After
some hours they become, for the most part, rubbed off or
swept away by the action of the uterine ciliated cells, and by
contact with the walls of the uterus. Some few cells which
have become more closely attached, hke 7.R. in fig. 39, or
those at the point of fusion (7.R.) in fig. 36, remain fora longer
time as part of the embryonal area. Whether, however, any
of them become permanently part of it I am quite unable to
say. I cannot trace any further than the condition of the
single cell (7.R’.) in fig. 38 or 39,

350 RICHARD ASSHETON,

Some time ago (1) I supported Balfour’s (3) and Heape’s
(14) views upon the fate of Rauber’s cells in the rabbit.

It must be remembered that the conditions in the pig are
very different from those of the rabbit. In the pig the
epiblastic disc is a compact mass of small area, and has no
investing membrane. In the rabbit the epiblastic disc is a
loose layer of cells, stretching over a large area, and outside
the trophoblast layer is a firm investing membrane, the
albumen layer, against which all the cellular layers are pressed
by the force of internal fluid. So whereas in the pig there
is nothing to prevent the broken fragments from being brushed
away or otherwise lost, except such original attachment as
they may have to the underlying epiblast, in the rabbit these
fragments are kept tightly pressed down to the surface of the
epiblast, and may even be forced into the interstices between
the loosely arranged cells as shown in my figures of the rabbit
(1) (Pl. 16, figs. 30, 32, 33).

Although there is very good evidence (Balfour [8], Heape
[14], Lieberkiihn [21], &c.) that in certain mammals a fusion
occurs, it is equally clear that in other cases (Verspertilio
murinus, Duval [13], Tupaia, Hubrecht [17], and in the
pig) most if not all of these cells are lost. It is only in those
cases in which there is, at the time in question, a well-developed
albuminous or other investment that we find evidence of a
fusion between these layers.

So perhaps we may say that there is no inherent tendency
for the “ Rauber” cells to fuse with the formative epiblastic
disc (except, of course, round its margin), but that in certain
cases (e. g. the rabbit, mole, and perhaps Sorex [ Hubrecht, 16,
17]) in which extraneous conditions in the form of zona
radiata and albumen layer occur—and in the case of the
rabbit, the existence of interstices between the epiblast cells—
many of the cells are included with and may form part of the
permanent epiblast.

It may be asked, Is the rejection of the cells in the pig,
Tupaia, and bat, due to the loss of continuity caused by the
physical and mechanical conditions of the earlier development,

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 351

or is there really a more profound difference in the nature of
the cells which prevents trophoblast and epiblast cells from
mingling? If (as I have argued in a discussion on the
development of the sheep) the trophoblast is really to be
regarded as hypoblast, and if there is in the pig so real a
difference between the two layers, and dating from so early a
period as appears to be the case in the sheep, one may well
consider that the trophoblast cells overlying the epiblast are
incapable of permanent incorporation as epiblast cells; and
that fusions which are brought about by the presence of
investing layers as described in the cases of the rabbit, mole,
and Sorex are secondary phenomena, or may even be only
apparent, and are not permanent fusious.

Weysse took a quite different view of the homology of the
layer he termed extra-germinal outer layer. The innermost
layer of cells (H., fig. 36, &c.) he calls entoderm. The outer
layer (7. R.) he did not regard as the original outer wall of the
blastodermic vesicle, but cailed it the extra-germinal ectoderm,
and considered that it was continuous at all times with the
ectoderm of the germiual area (£.).

Outside this he believed that a third layer, the true Rauber
layer, had existed during an earlier stage, and he describes the
remains of it as appearing at intervals over the surface of the
“ extra-germinal ectoderm.”

It is clear that this was a rather rash supposition, as
the youngest embryo which he possessed was ten days old,
—that is to say, the same age as those of which my figs. 10,
36—89, are drawings. That certain cells existed outside
the ‘‘ extra-germinal ectoderm ”’ he was quite convinced, and
describes and gives drawings of them. Hubrecht (17) con-
sidered that these cells must be cells derived from the uterine
walls.

Weysse is perfectly right in saying that at the time of the
process of “ bridge” formation on the embryonic area, there are
at certain places cells outside the outer wall of the vesicle.
For instance, in the specimen (fig. 10) there are no less than
forty-three perfectly distinct cells with nuclei, scattered at

352 RICHARD ASSHETON.

intervals, some singly, others in groups of two, three, or, in
one case, as many as seven cells, upon the lower or ab-
embryonic pole.

Figs. 40 and 41 are drawings of some of these seen in
section. The trophoblast layer is thick (7. #.), the hypoblast
layer (H.) is very uncertain, and is probably a uetwork, and
on the outer surface of the trophoblast a few isolated cells
(7. R.”) form conspicuous objects. These cells are in every
way exactly comparable in their colour, texture, and size of
nucleus with the cells of the trophoblast layer. They are
never present before this time ; they disappear very quickly
afterwards.

Iam quite sure they are not derived from the walls of the
uterus. They are not leucocytes, nor are they corpuscles of
the secretion of the uterine glands. Such are, indeed, present
at this and an earlier period, but cannot be mistaken for
embryonic cells. The epithelium of the uterus is still per-
fectly sound, and shows no sign of degeneration until after
the fourteenth day. It becomes detached about the seventeenth
day.

I have very little doubt of the origin of these scattered cells.
I think they are the detached fragments of the trophoblast
from the embryonal area.

The uterus at this time contains ciliated cells, whose action,
no doubt, helps the passage of the blastodermic vesicles down
the uterus. It seems very likely that the broken fragments
are swept along by the action of the cilia, and adhere for some
time to the walls of the blastodermic vesicle until ultimately
swept or rubbed off altogether.

Fig. 42 represents a section taken through a larger piece,
which was only a short distance from the embryonal area
of another specimen. The clear line which invariably
marks them off from the walls of the blastocyst makes it
unreasonable to suppose that they have been budded off
in situ.

I think there can be no doubt that the cells which I find and
have just described are the same as those described by Weysse,

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 358

At the same time his description can hardly be said to agree
with mine. He speaks of the cells being indistinct and
flattened, the cytoplasm “scarcely stainable” at all. In my
case they are mostly sharp and rounded, and they stain with
carmalum and hematoxylin as deeply as any others in the
embryo. A difference in methods of preservation and staining
may account for this.

However this may be, there can be no question of the pre-
sence of a third complete layer outside the layer which he calls
extra-germinal ectoderm, and I call trophoblast ; such a layer
never exists.

Before the completion of the eleventh day all trace of the
loose fragments of the torn edge of the outer layer has dis-
appeared from the embryonal area and elsewhere.

The formative epiblast is now continuous all round its
margin with the trophoblast, and forms a disc-like plate,
slightly thicker near its centre than at its edge.

The hypoblast immediately beneath the epiblast is thicker,
and contains more nuclei than elsewhere. Its characters are
shown in figs. 36—42.

There is one structure which is very puzzling. Weysse has
noticed this. This is the membrane shown in figs. 30—33 and
36—39, M. H.L. Ihave not been able to determine its origin.
Weysse (25) gives a correct description of it on pages 297 and
298. It certainly has in my sections the appearance of being
part of the inner margin of the cells of the trophoblast and
the epiblast. It is structureless, but whether secreted by
the epiblast or trophoblast and the hypoblast I do not know.
Schifer (24) considers it of hypoblastic origin in the cat.
Its probable function would seem to be in some way con-
nected with the rupture of the outer layer, and the subsequent
fusion of its edges with those of the epiblastic disc.

A time may very well occur when there would be weakness
in the general wall of the blastocyst during this process. The
membrana hypoblastica limitans seems to serve to hold the
disc in its place while this fusion is effected. The membrane
is tightly fixed to the inner surface of the epiblastic disc, but

354 RICHARD ASSHETON,

more loosely to that of the trophoblast. The fluid inside the
blastocyst would force the dise up into its place.

The changes of the blastodermic vesicle and its great in-
crease in length, which occur during the twelve to fourteen
days, have been related by Von Bacr, Coste, and Bischoff, and
more recently, together with a detailed account of the changes
in the embryonal area, by Keibel. A brief account from me
will suffice in confirmation of their discoveries.

From the moment of the rupture of the zona pellucida upon
the sixth day of development, there has been a constant ten-
dency for the vesicle to become elongated. By the twelfth
day it is 10 or 12 mm. long and only 38 mm. broad. It now
grows out with exceedingly great rapidity, and by the thir-
teenth or fourteenth day each embryo may measure as much
as a foot in length (30 cm.).

The hypoblastic vesicle extends to nearly the same length
as the outer wall. Hence the whole blastocyst can be said to
be didermic.

By the seventeenth and eighteenth days the vesicles have
attained to about their greatest length, and completely fill the
cavity of the two horns of the uterus, The length of each

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 355

vesicle depends, it would seem, directly upon the length of the
uterus, number of embryos, and the position of the embryos in
the uterus.

The walls of the uterus are very soft, and the foldings of its
mucous membrane very complicated, and obliterate the lumen
of the tube almost completely.

The several blastodermic vesicles do not overlap each other.
The highest up on each side is always longer than the others
in the same horn of the uterus. The longer ones are thinner
than the shorter ones. The vesicle is not folded back upon
itself, but crumpled like the sides of wind bellows. In one
sow killed upon the eighteenth day, the one which occupied
the highest position of one side, measured when removed
and unfolded forty-two inches. The length of the piece of
uterus which contained it was twelve inches. The two at
the lower ends of the horns of the uterus extend into the
vagina, and in one case were fused with each other. I
have never found any fusion between any of the vesicles
within the uterus, although their ends were in contact.

REFERENCES.

1. AssHETon, R.—‘‘ A Re-investigation into the Early Stages of the
Development of the Rabbit,” ‘Quart. Journ. Micr. Sci.,’ vol. xxxvii,
1894.

la. AssHeTon, R.—‘‘ The Segmentation of the Ovum of the Sheep, with
Observations on the Hypothesis of a Hypoblastic Origin for the
Trophoblast,” ‘Quart. Journ. Mier. Sci.,’ vol. xli, 1898.

2. von Baer, K. E.—‘ Ueber Entwickelungsgeschichte der Thiere,’
Konigsberg, 1828.

3. Batrour, F. M.—‘ A Treatise on Comparative Embryology,’ 2nd edit.,
London, 1885.

4. Van Benepen, E.—‘ De la maturation d’ceuf, de la fécondation et des
premiers phénoménes embryonaires chez. les Mammiféres d’apreés les
observations faites chez le lapin,” ‘ Bulletin de l’Académie Royale des
Sciences de Belgique,’ 1875.

5. Van Bernepen, E.—‘ Recherches sur |’embryologie des Mammiferes.
La formation des feuillets chez le lapin,” ‘Archives de Biologie,’
tome i, 1880.

356 RICHARD ASSHETON.

6.

10.

11.

12.

13.

14.

15.

16.

Wil.

18.

19.

20.

21.

Biscuorr, T. L. W.— Entwickelungsgeschichte des Kaninchen-Hies,’
Braunschweig, 1842.

. Biscuorr, T. L. W.—‘ Entwickelungsgeschichte des Hundes-Kies,’

Braunschweig, 1845.

. Biscnorr, T. L. W.—‘ Entwickelungsgeschichte des Meerschweinchens,’

Giessen, 1852.

. Biscnorr, T. L. W.—‘ Entwickelungsgeschichte des Rehes,’ Giessen,

1854.
Bonnet, R.—“ Beitrage zur Embryologie der Wiederkauer gewonnen

am Schafei,” ‘Archiv fiir Anatomie und Physiologie,’ Anatomische
Abtheiiung, 1884.

Bonnet, R.—* Beitrige zur Embryologie der Wiederkauer, gewonnen
am Schafei,” ‘Archiv fiir Anatomie und Physiologie,’ Anatomische
Abtheilung, 1889.

Duvat, M.— Le Placenta des Rongeurs,” ‘Journal d’Anatomie et de
la Physiologie,’ 1889-92.

Duvat, M.—“ Etudes sur l’Embryologie des Chéiropterés,”’ ‘Journal de
l’Anatomie et de la Physiologie,’ vols. xxxi, xxxil, 1895-6.

Heare, W.—* The Development of the Mole (Talpa europea), the
Formation of the Germinal Layers, and Early Development of the
Medullary Groove and Notochord,’”’ ‘ Quart. Journ. Mie. Sci.,’ vol.
xxill, 1883.

Heare, W.—‘‘ The Development of the Mole (Talpa europea), the
Ovarian Ovum, and Segmentation of the Ovum,” ‘Quart. Journ. Mic.
Sci.,’ vol. xxvi, 1886.

Husrecut, A. A. W.— Studies in Mammalian Embryology. II. The
Development of the Germinal Layers of Sorex vulgaris,” ‘ Quart.
Journ. Mie. Sci.,’ vol. xxxi, 1890.

Husrecut, A. A. W.—‘ Die Phylogenese des Amnions und die Bedeu-
tung des Trophoblastes,” ‘ Verhandelingen der Koninklijke Akademie
van Wetenschappen te Amsterdam,’ 1895.

Kerpex, F.— Studien zur Entwickelungsgeschichte des Schweines,” I,
‘Morphologische Arbeiten,’ Bd. iit.

Keiset, '.—“ Studien zur Entwickelungsgeschichte des Schweines,” II,
‘ Morphologische Arbeiten,’ Bd. v.

Koutixer, A.—‘ Die Entwickelung der Keimblatten des Kaninchens,”
‘Zeitschrift zur Feier des Zoojahrigen bestehens des Julius-Maxi-
milians- Universitat zu Wirzburg,’ 1882.

Linzerkiiun, N.—‘ Ueber die Keimblatter der Saugethiere zur der
Funfzigjahrigen doctor-juvelfeier des Herr Hermann Nasse,’ Marburg,
1879.

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 357

22. Rauper.—“ Die erste Entwickelung des Kaninchens,” ‘ Sitzungsberichte
der Naturforschenden Gesellschaft zu Leipzig,’ 1875.

23. Ropinson, A.—“ Observations upon the Development of the Segmenta-
tion Cavity, the Archenteron, the Germinal Layers, and the Amnion
in Mammals,’ ‘ Quart. Journ. Mic. Sci.,’ vol. xxxiii, 1891-2.

24. Scuirer, EH, A.—“ Description of a Mammalian Ovum in an Early
Condition of Development,” ‘Proceedings of the Royal Society of
London,’ 1876.

25. Wrysszt, A. W.—“‘On the Blastodermic Vesicle of Sus scrofa

domesticus,” ‘American Academy of Arts and Sciences,’ vol. xxx,
1894.

DESCRIPTION OF PLATES 25—28,

Illustrating Mr. Assheton’s paper on “The Development of
the Pig during the First Ten Days.”

Compete List oF REFERENCE LETTERS.

C. Bl. Cavity of the blastocyst. #. Epiblast. HZ. Hypoblast. 7. /. Inner
mass. J/. H.Z. Membrana hypoblastica limitans. SP. Spermatozoa. 7.
Trophoblast. 7... Trophoblast cells which overlie the embryonic knob.
Z. Zona radiata.

PLATE 25.

Fig. 1 (Sus 25, No. 1).—Embryo of the pig in the two-segment stage.
Found in the uterus. Killed with Flemming’s solution. The oily globules
scattered throughout the inner part of each segment have been blackened by
the osmic acid. Age 6days. x 380.

Fic. 2.—Same embryo as the former, but seen edgewise. x 380.

Fig. 3 (Sus 12, No. 1).—Kmbryo of pig in the three-segment stage.
Found in the upper part of the uterus. Examined and drawn while fresh.
Age 4 days. xX 380.

Fig. 4 (Sus 1, No. 1).—Embryo of pig in the four-segment stage. Found
in the Fallopian tube. Hxamined and drawn while fresh. Age three days.
x 380.

Fic. 5 (Sus 12, No. 4).—Embryo of pig in the five-segment stage. Found
in the uppermost part of the uterus. KHxamined and drawn while fresh. Age
4 days. x 380.

Fic. 6.—Embryo of pig. Found in the uterus. Drawn after the specimen
had been treated with chromic acid. Age 6 days. x 380.

358 RICHARD ASSHETON.

Fie. 7 (Sus 25, No. 6).—Embryo of pig in morula stage. Found in uterus.
Drawn after the specimen had been treated with chromic acid. Age 6 days.
x 380.

Fie. 8 (Sus 25, No. 7).—Embryo of pig. Morula stage probably not quite
attained. Age 6 days. x 380.

Fie. 9 (Sus 26, No. 13).—Hmbryo of pig. Advanced blastodermic vesicle
stage. Age 5 days. x 380.

Fie. 10 (Sus 28, No. 2).—Embryo of pig. The blastodermic vesicle is now
a large thin-walled sac of irregular shape. This is double-layered; and the em-
bryonal area is a circular or sometimes slightly oval patch upon one side of it.
Age 10 days 23 hours. x 20.

Fic. 11 (Sus 33, No. 1)—Embryo of pig. The blastodermic vesicle is
thrown into many folds. Age 8 days. x 20.

Fie. 12 (Sus 28, Nos. 5, 3).—The embryonal areas of two specimens
showing the pieces of trophoblast, 7’. 2., lying upon the dise of true epiblast.
The specimen from which the smaller embryonal area was taken measured
about 3 mm. in length. The embryonal area itself measured 215 x ‘190mm.
The specimen from which the larger figure (B) was taken measured 3°5 mm.
by 2°94 mm., and the embryonal area was about *22 mm. in diameter. The
two figures were drawn with the camera lucida, but under different magnifying
power.

PLATE 26.

All the figures on this plate are camera drawings of sections of embryos of the
pig. ‘The magnification in every case is 380 times.

Fie. 13 (Sus 23, No. 3).—Age 6 days 3 hours.
Fte. 14 (Sus 23, No. 1).—Age 6 days 3 hours.
Fie. 15 (Sus 25, No. 10).—Age 6 days.

Fie. 16 (Sus 29, No. 1).—Age 5 days.

Fie. 17 (Sus 19, No. 2).-—Age 5 days 3 hours.
Fie. 18 (Sus 26, No. 2).—Age 5 days.

Fic. 19 (Sus 51, No. 8).—Age 43 days.

Fic. 20 (Sus 27, No. 1).—Age 5 days.

In all the foregoing the zona radiata is present, but after this point it has
ruptured and disappeared.

Fie, 21 (Sus 32, No. 4).—Age 7 days.
Fie. 22 (Sus 14, No. 6).—Age 7} days.
Fic. 23 (Sus 14, No. 6).—Age 73 days.
Fic, 24 (Sus 33, No. 6).—Age 7 days.

DEVELOPMENT OF THE PIG DURING FIRST TEN DAYS. 3099

PLATE 27.

All the figures on this plate are camera drawings of sections of embryos,
or of the embryonal area of embryos of the pig. The magnification in
each case is 380 times.

Fig. 25 (Sus 32, No. 1).—Age 7 days.

Fic. 26 (Sus 32, No. 1).—Age 7 days. The hypoblast in this specimen,
as in several others, is removed from the epiblastic knob.

Fic. 27 (Sus 14, No. 2).—Age 7% days.

Fic. 28 (Sus 14, No. 1).—Age 73 days.

Fie. 29 (Sus 33, No. 3).—Age 8 days.

Fies. 30, 31, 32, 33 (Sus 18, No. 8).—Age 9% days. These are four
sections through one specimen, namely, the 3rd, 5th, 9th, and 10th of a
series of thirteen sections.

Fic. 34 (Sus 18, No. 10).—Age 93 days.

Fie. 35 (Sus 18, No. 9).—Age 93 days. Diameter of embryonal area
1 mm.

PLATE 28.

All the figures on this plate are camera drawings of sections through
portions of the walls of the blastocyst of the pig. All are magnified 380
times.

Fie. 36 (Sus 28, No. 2).—The same embryo as Fig. 10. Age 10 days
2% hours. Section through embryonal area. Whole embryo measured 3°57
xX 2°31 mm.

Fie. 37 (Sus 28, No. 2).—Another section through the same embryonal
area,

Fie. 38 (Sus 28, No. 4).—Age 10 days 23 hours. Section through the
embryonal area. Blastocyst measured 4°06 x 2°60 mm.

Fig. 39 (Sus 28, No. 5)—Age 10 days 23 hours. Section through em-
bryonal area.

Fig. 40 (Sus 28, No. 2).—Age 10 days 23 hours. Section through the ab-
embryonic pole.

Fig. 41 (Sus 28, No. 2).—Age 10 days 23 hours. Another section simi-
larly located to above (Fig. 40).

Fig. 42 (Sus 28, No. 7).—Age 10 days 23 hours. Section through the
wall of the blastocyst a short distance from the embryonal area.

aa

bg ; rd
on3 epiit

=~}
‘ wy

313 Se al *

ips \ 118
sal 1eGhevs sal a ray

t
*
2S S-
a oa
Tr i
is (ade
irs a ;
: o
zy
= fy a
i
\ ;
‘
7 .
; r

nd at ae. - Oe
;
=h
if - i
2g, 1S
chgty * Gk
‘
;

j

a |
- ‘

"
i
,
,

F

4 a ;

;
-
,
‘

=

aaNet

STRUCTURE OF THE MAMMALIAN GASTRIC GLANDS. 361

The Structure of the Mammalian Gastric
Glands.

By

R. R. Bensléy,’ B.A., M.B.,
Assistant Demonstrator in Biology, University of Toronto.

With Plate 29.

.

In a preliminary notice, published in the ‘ Proceedings of
the Canadian Institute,’ vol. i, Part I, I gave a brief account
of some new points in the structure of the gastric glands of
mammals, which appear to afford a solution of the question of
the morphological significance of the pyloric glands.

The view of Heidenhain,! Ebstein,? and Griitzner,® that the
pyloric glands are simply peptic glands without border cells,
and that the pyloric gland cells are identical with the chief or
central cells of the fundus glands, is no longer tenable.

Heidenhain* himself noted that in the fresh condition the
pyloric gland-cells are finely granular, whilst the chief cells of
the fundus glands are coarsely granular.

Langley and Sewall® observed the same feature, but con-
sidered that the undoubted presence in the pyloric glands of
pepsin in small amount was sufficient evidence of their pepsi-
nogenic character. They concluded, therefore, not that the
pyloric gland-cells are different from the chief cells of the

1 ¢ Arch. f. mik. Anat.,’ Bd. vi.

* Tbid.

3 *Pfliiger’s Archiv,’ Bde. vi and viii.

4 Herrmann’s ‘ Handbuch d. Phys.,’ Bd, v.
5 ¢ Journal of Physiol.,’ vol. il.

vou. 41, PART 3.—NEW SERIES. rons!

362 R. R. BENSLEY.

fundus glands, but that “pepsin formation is not necessarily
connected with the formation of coarse granules ;” and further,
that “the chief cells of the fundus are a highly differentiated
form of the pyloric gland-cells.”’

The introduction of new methods has made us acquainted
with new points of difference between the two kinds of cells.

Schiefferdecker! found that the pyloric gland cells of the
pig and man stained intensely in dahlia, a property which is
not shared by the chief cells of the fundus glands, or by the
glands of the cesophagus and mouth. He found, moreover,
that the pyloric glands of the cat and dog did not stain in
dahlia, a fact which is of importance as indicating a difference
between the pyloric glands of different mammals, and further,
that these glands differed anatomically from those of the pig
and man.

Bonnet? also has studied, by means of aniline dyes, the
staining reactions of the various cells of the stomach. He
finds that the pyloric gland cells stain differently from the
surface epithelium and from the chief cells in methyl violet,
Congo red, and acid fuchsin.

R. Krause ® has observed that the cells of the pyloric glands,
in common with many mucous cells, stain metachromatically
in thionin.

On the other hand, the results of research by purely physio-
logical methods seem to point to a functional relationship
between the pyloric gland cells and the chief cells. Of the
host of observers who have examined the pyloric mucous mem-
brane of the dog for pepsin, few have failed to find it, and its
presence in the secretion of the pylorus has been shown by
Klemensiewicz* and Heidenhain,> who established pyloric
fistule, and found abundant evidence of the presence of a

1 ‘Nachrichten d. Gottingen Gesellsch.,’ 1884, p. 303.

2 «Berichte d. Oberhessisch. Gesellsch.,’ xxix, 1893.

3 R. Krause, “Zur Histologie der Speicheldrusen,” ‘ Arch. f. mik, Anat.,’
Bad. xlv.

4 « Sitzungsber. d. k. Akad. d. Wissenseh.,’ Bd. lxxi.

5 « Pfliger’s Archiv,’ Bd. xviil.

STRUCTURE OF THE MAMMALIAN GASTRIC-GLANDS. 9863

proteolytic ferment, in Heidenhain’s case even after the lapse
of five months.

In his recent exhaustive compilation of the literature of
this subject, Oppel! attempts to reconcile the conflicting re-
sults of these two lines of research. He concludes that the
pyloric gland cells are cells sui generis, differing both from
the surface epithelium and from the chief cells, and engaged in
the secretion of pepsin holding gastric juice.

The present state of our knowledge does not permit of any
comparison between the pyloric glands and the other glands
of the stomach, nor is it possible to compare them with any of
the gastric glands of lower Vertebrates.

Further, the researches of Edelmann” have shown that there
exists in the cardiac region of the stomach of many mammals
a peculiar kind of gland, called by him the cardiac gland,
differing both from the fundus glands and the pyloric glands,
and concerning which we are even more in the dark.

The application of new methods to the study of the gastric
glands has convinced me that the pyloric and cardiac glands
of various animals are closely allied to one another, and that
the various kinds of cells one meets are but the results of
differentiation along divergent lines from a single primitive
type. The pyloric gland cells, furthermore, are in most
mammals closely allied to, and in the cat, dog, and rabbit
identical with, certain cells in the neck of the fundus gland,
which, up to the present, have been regarded as ordinary chief
cells.

A convenient starting-point for the descriptions which
follow is afforded by the gastric glands of the frog.

A fundus gland of this animal may be divided into three
portions ; the duct or stomach pit, lined by mucus-secreting
cylindrical cells similar to those of the surface ; the neck,
occupied by very large vesicular-looking cells, which, although
different from the surface cells, are also regarded as mucous
cells, and the body of the gland, occupied by granular proto-

1 «Lehrbuch der vergleichenden mik. Anat.,’ 1896.
2 ‘Deutsch. Zeitschr. f. Thiermedizin,’ Bd, xv.

364 R. R. BENSLEY.

plasmic cells, which secrete both acid and pepsin. In sections
stained in the muchematein solution of Mayer, the muci-
genous border of the cylindrical cells of the surface and the
whole of the large vesicular neck cells stain intensely, indi-
cating beyond doubt that the latter are mucin-secreting cells.
Langley! has shown that during digestion these cells exhibit
the usual secretion changes.

The pyloric glands of the frog are made up of only two
kinds of cells, those of the body and those of the duct. The
former bear so strong a resemblance to the large mucus-
secreting neck cells of the fundus glands, that one cannot
avoid the conclusion, with Partsch, that they are of the same
nature. They, too, stain intensely in muchematein.

It is generally admitted that the two main kinds of cell of
the mammalian fundus gland are the result of the differentia-
tion of the one kind found in the body of the gland of lower
Vertebrates. The view advanced by Oppel,’ that the mucous
neck cells of batrachian and reptilian glands correspond to the
chief cells of the mammalian gland has been, however, the
only attempt to find in the glands of mammals a morphological
equivalent for these peculiar cells.

My studies have enabled me to establish what has hitherto
been unsuspected, namely, that there exists in the fundus
glands of many mammals cells which are morphologically and
physiologically equivalent to the mucous neck cells of the
batrachian gland, and that the same relationship exists between
these and the pyloric gland-cells as obtains in the Anura.

These cells in the neck of the gland have received little
attention from histologists owing to their small size, and to
their being overshadowed by the large and numerous border
cells of this region of the gland; they are generally regarded
as small pepsin-secreting chief cells.

The fact that they are different from the cells lower down
in the gland has not, however, entirely escaped notice.
Bizzozero® noticed in the dog that the chief cells of the neck

1 «Phil. Trans. Roy. Soc.,’ vol. clxxii.

2 * Anat. Anzeig.,’ Bd. xi.
3 Arch, f. mik. Anat.,’ Bd. xlii.

STRUCTURE OF THE MAMMALIAN GASTRIO GLANDS. 365

of the gland have a more transparent protoplasm than those of
the body of the gland, and a nucleus compressed against the
base of the cell. Similar features have also been noted for the
glands of the badger and hedgehog by Oppel,! who calls
attention to the fact that in the hedgehog the neck cells
contain less protoplasm than the chief cells of the bottom of
the gland, and that this stains less readily with haematoxylin.
Bizzozero suggests that these cells may be a transitional type
between the cells of the gland duct and the fully developed
chief cells of the deeper portions.

My attention was first attracted to these cells in the glands
of the greater curvature of the rabbit, in sections of which,
stained in hematoxylin, the chief cells appear as comparatively
large cubical cells with deeply staining protoplasm, whilst the
neck cells are small pyramidal structures which stain but
feebly. The question naturally arose whether this difference
was due to a different functional condition of the cells, or to
the cells being essentially different, and I turned for a solution
of the question to a study of the distribution of zymogen
granules in the gland.

The stomach of the rabbit did not lend itself very readily to
this investigation on account of the comparatively short neck
that the glands of this animal possess, and because I was not
then able to fix the granules in any but the lowest portions of
the glands. I therefore resorted to a study of the glands of
the cat and dog, in which the ueck region is relatively long.
I subsequently discovered that it was possible to fix perfectly
the granules in all parts of the glands of many mammals
by means of a modification of Foa’s blood-fixing fluid, prepared
by mixing equal parts of a saturated solution of mercuric
chloride in 95 per cent. spirit, and a two to four per cent.
aqueous solution of potassium bichromate. I was also fortunate
enough to discover a means of staining in a distinctive fashion
with indulin these peculiar neck cells, and the use of these
methods has enabled me to extend the facts discovered in the
cat and dog to the rabbit and other mammals.

1 “Lehrbuch d. vergleich. mik. Anat.,’ 1896.

366 RB. B. BENSLEY.

My methods are briefly as follows :—Small pieces of the
gastric mucosa are snipped off with scissors, and dropped
into the sublimate bichromate mixture, where they remain
from one half to two hours, according to their thickness.
They are then transferred to 70 per cent. alcohol, in which
they remain twenty-four hours, or until all the free bichromate
is extracted, then to 95 per cent. alcohol. Sections of 3—5
micra are cut after embedding in paraffin by the oil of ber-
gamot method, fastened to the slide, and stained. The results
obtained by this method of fixation were controlled by the
study of pieces fixed in alcohol, in aqueous bichloride solutions,
and in the osmic acid fixing fluids of Hermann and yom Rath.
The staining methods employed will be indicated in connec-
tion with the special descriptions.

I have chosen for special description in the present memoir
the gastric glands of the cat and dog, because these present
the most highly differentiated form of the gastric gland, and
because the relationship obtaining between the pyloric and
fundus glands corresponds so closely to that found in the
highly specialised Anura.

A. The Gastric Glands of the Cat.

The fundus glands are elongated tubular structures, opening
into shallow depressions of the surface lined by mucus-secreting
cylindrical cells, and called the stomach pits or gland ducts.
The glands consist of two kinds of epithelial cells, the central
or chief cells and the parietal or border cells, and are divisible
into two portions, a narrower superficial part called the gland
neck, in which the border cells are in excess, and a deeper,
wider portion called the body of the gland, in which the chief
cells predominate.

That the difference between the body and neck of the gland
is of a more profound nature than a mere difference in relative
size, or in the relative numbers of the constituent cells, may
be readily determined by the study of the fresh mucous
membrane in an indifferent fluid. If, in a freshly killed cat,

STRUCTURE OF THE MAMMALIAN GASTRIC GLANDS. 867

a piece of mucous membrane be snipped off, and as thin a
section as possible prepared with a razor moistened with
aqueous humour, and mounted under a cover in a drop of the
same fluid, it may be observed, under a low power and small
diaphragm, that the mucous membrane is divided into a
superficial transparent zone and a deeper more opaque zone
(fig. 1).

Under a high power (Zeiss, apo. 2 mm. and 8 oc.) the
opacity of the deeper zone is seen to be due to the presence
of numerous large, coarse granules of zymogen. These
granules are entirely absent from the superficial zone, although
many minute fat globules may be seen in both kinds of cells.

The superficial granule-free zone includes not only the pits,
but a large portion of the glands themselves; and it may be
inferred that the chief cells of the neck of the gland do not
contain zymogen in the form of granules.

This peculiarity of the distribution of zymogen granules in
the mucosa did not escape the notice of Langley and Sewall,!
as their figure (14) of the neck of the gland of the cat clearly
indicates. They, however, attributed the absence of granules
from the neck of the gland to the comparatively infrequent
occurrence of chief cells here.

It will be seen from what follows that the absence of the
granules from the gland neck is rather to be ascribed to the
fact that the chief cells are different from those of the body
of the gland, and are engaged in the secretion of a quite
different product. In order that the differences between the
two kinds of chief cells may be clearly defined, it will be
necessary to describe accurately the ordinary chief cell of the
body of the gland.

In sections from the greater curvature of the stomach of a
eat that has fasted one to three days, the chief cells of the
body of the gland present the appearance indicated in fig. 2.
They are pyramidal or wedge-shaped, and so appear cubical or
triangular, according to the direction in which they are cut.
The contents of the cell exhibit an exceedingly regular network

«Journal of Physiology,’ vol. ii.

368 R. R. BENSLEY.

of large meshes, and in thick sections present a vacuolated
appearance. The protoplasmic strands which compose this
network are very coarse, and stain readily in hematoxylin, a
feature which is particularly noticeable at the thickened nodal
points.

In sections stained in gentian violet or safranin the ap-
pearance depends on the degree of success attained in fixing
the zymogen granules. If they are well preserved they stain
intensely in these dyes, and the cell is then seen to be filled
with large deeply stained granules, between which may be
seen running the trabeculz of the protoplasmic framework.
If the fixation is less successful the granules are found to
have swollen up, so that the whole cell stains diffusely—afford-
ing, however, unmistakable evidence of the presence of
zymogen.

The relation between the granules and the protoplasm may
be clearly seen in sections stained in the Biondi three-colour
mixture, in which the granules stain a pale blue and the proto-
plasm red. In sections thus stained each granule is found to
correspond to a mesh of the protoplasmic network. This is,
then, not a true network, but simply the optical expression of
the fact that the zymogen granules occupy small cavities in
the cell, which are separated from one another by thin films of
the protoplasm of the cell. In hardened cells there is usually
a clear space surrounding each zymogen granule, but it is to
be inferred that in the living resting cell the granule com-
pletely fills the cavity in the protoplasm which it occupies.

In the base of the cell, even after a prolonged fast, there
may usually be seen a small quantity of protoplasm which, on
account of the peculiar properties it presents, seems to merit
a more extended description than is usually accorded it.

Langley ! observed that the protoplasmic zone of the active
cell contains a substance which stains more readily with osmic
acid than ordinary protoplasm, and which he inferred to be
one of the earlier steps or mesostates in the formation of the

1 «Phil. Trans. Roy. Soc.,’ vol. elxxii,

STRUCTURE OF THE MAMMALIAN GASTRIC GLANDS. 369

zymogen. Griitzner! also noted this peculiar staining with
osmic acid.

Some information as to the nature of the substance in
question is afforded by the researches of Macallum,? who
describes the difference in staining properties exhibited by the
resting and exhausting pancreatic cell, and explains the dif-
ference as follows :—‘‘ The chromatin of the nucleus gives rise
to a substance which we may call prozymogen, sometimes
dissolved in the nuclear substance, sometimes collected in
masses (plasmosomata), and finally diffused into the cell
protoplasm, uniting with a constituent of the latter as
zymogen.” Ina subsequent research into the distribution of
assimilated iron compounds in animal and vegetable cells,? he
found, in the outer protoplasmic zone of the pancreatic and
many other gland cells, a firm organic compound of iron,
which he regards as the prozymogen of his earlier investiga-
tion. A similar view is taken by Mouret of the nature of the
fibrillar chromophilous element in the outer zone of the pan-
creatic cell, and the term “ prézymogen” is applied by this
observer to the substance in question.

I have made a series of experimental studies of the gastric
and many other glands, with a view of determining the relation
of this substance to the formation of zymogen granules, and
also its source in the cell. The results of these studies will be
contributed in a separate paper, and I will content myself at
present with a recital of the facts that are of importance from
the stand-point of determining the morphological relationships
of the cells.

The prozymogen is co-extensive with the protoplasm of the
cell, and, even in cells which possess only a small outer zone,
usually presents quite definite staining and structural charac-
ters, which enable one to decide with ease as to its presence.
The most favorable material for studying its characters is offered
* Pfliiger’s Archiv,’ Bd. xx.

‘Trans. Canadian Institute,’ vol. i, part ii, 1891.

‘Quart. Journ. Micr. Science,’ vol. xxxvili, part ii, new series.

1
z
3
4 ¢ Journal de |’Anat. et de la Physiol.,’ année xxxi, 1895.

370 . R. R. BENSLEY.

by the glands of animals that have been in active digestion for
ten to twelve hours, and therefore exhibit a well-marked outer
protoplasmic zone (fig. 3). In sections from such glands,
stained in freshly prepared Mayer’s hzmalum, a pure nuclear
stain is obtained in all the cells, with the exception of the
chief cells of the body of the gland, the outer protoplasmic
zone of which also stains blue. A more vigorous stain of this
portion of the cell may be obtained by the use of Ehrlich’s
acid hematoxylin, diluted for use with a 5 per cent. solution
of ammonia alum in water. The sections, after staining in
this fluid, are washed in tap water, then dehydrated and mounted
by the usual methods. Staining in very dilute solutions of
methylene blue, gentian violet, or safranin, followed by
rapid dehydration in absolute alcohol, and clearing in benzole,
also gives a very serviceable stain of the outer zone of the
cell. In sections so stained the outer zone of the cell
exhibits an obscurely fibrillated structure, which reminds one
strongly at first of the striated epithelial cells in the intra-
lobular ducts of the salivary glands (fig. 3). On closer exami-
nation it may be seen that the fibrillation in the outer zone of
the chief cell is not so regular, nor are the fibrille so distinct
from one another as in the salivary ducts.

The strong affinity for nuclear stains exhibited by the outer
protoplasmic zone of the chief cell is due to the presence in it
of a chromatin or firm organic compound of iron, the prozy-
mogen of Macallum, as may be shown by the reactions for the
presence of iron. If a section of a piece of mucous membrane
that has been hardened in absolute alcohol be treated with
ammonium hydrosulphide, or an acid solution of potassium
ferrocyanide, no reaction occurs, indicating that no inorganic
iron is present in the cell. If, however, the sections be first
treated with a solution of pure sulphuric acid in alcohol, con-
taining four volumes per cent. of the former, for a period of
three to six hours at a temperature of 37° C., and then, after
thorough washing in fresh alcohol, transferred to ammonium
hydrosulphide or acid ferrocyanide solution, a strong reaction
for iron is obtained, not only in the chromatin of the nucleus,

STRUCTURE OF THE MAMMALIAN GASTRIC GLANDS. 371

but also in the outer protoplasmic zone of the chief cells of the
body of the gland. The reaction in the protoplasm is about
equal in intensity to that obtained in the oxyphile nucleolus.
The best method of demonstrating the presence of prozymogen
is by means of the hematoxylin iron reaction, recently an-
nounced by Macallum.! In this method, after unmasking the
iron by means of sulphuric acid alcohol, the sections are carefully
rinsed in alcohol to remove all the free acid, then transferred to
0°5 per cent. solution of pure hematoxylin in water, which turns
every portion of the section containing iron a peculiar slate-blue
colour similar to that obtained in staining by the iron-alum
hematoxylin method of M. Heidenhain. This is a much more
sensitive test than the ammonium hydrosulphide or ferro-
cyanide reactions, and serves extremely well to exhibit the iron
when present in only minute quantities in the protoplasm. In
sections so treated the outer protoplasmic zone of the chief
cell shows a distinct blue colour, indicating that it contains
a considerable amount of unmasked iron. A reaction is also
obtained in the protoplasm of the portion of the cell occupied
by the granules of zymogen, although not equal in intensity to
that in the outer clear zone.

The presence of a large quantity of masked iron in the cell
protoplasm is a feature which serves to distinguish the chief
cells of the body of the fundus gland from all other glandular
cells in the stomach. The protoplasm of the border cells
shows no reaction whatever when treated in the manner indi-
cated above; and that of the pyloric gland cells, of the cylin-
drical surface cells, and of the chief cells of the neck of the
gland gives only a faint reaction for iron.

The prozymogen or cytoplasmic chromatin differs from the
nuclear chromatin in some respects, as is shown by its relation
to stains. In sections stained in gentian violet the prozy-
mogen takes a reddish metachromatic stain, which contrasts
very well with the colder blue of the nucleus and zymogen
granules (fig. 4). In Biondi solutions, which give a good basic
nuclear stain, the prozymogen stains reddish—not, however, a

1 «Journal of Physiology,’ vol. xxii, 1897.

O72 R. R. BENSLEY.

purerubin stain. Furthermore, washing in alcohol after stain-
ing in safranin will extract the safranin from the prozymogen,
and, as Macallum!? pointed out, from the oxyphile nucleolus,
long before it is extracted from the basophile chromatin of the
nucleus. I have made use of this property in connection with
the hematoxylin iron reaction to determine the distribution of
prozymogen in the glandular cells of the stomach. In working
with the hematoxylin method alone one is often in doubt
whether an apparent reaction is a real one, or simply the
result of a nucleus lying in an upper or lower plane of the
section, and out of focus, acting as a light filter. If, however,
the section after treatment with the hematoxylin be well
washed and transferred to a dilute solution of safranin in 30
per cent. alcohol, then extracted in alcohol until the safranin
is nearly all removed, cleaned in benzole, and examined, it is
found that whilst the nuclear chromatin has taken on a red-
dish-blue tinge from the safranin, the oxyphile nucleolus and
the prozymogen have retained a slaty-blue colour, which it
is impossible to mistake even in thick sections (fig. 5).

The fibrillated appearance presented by the outer clear zone
of the chief cell is of adventitious origin, and not in itself of
importance. A study of the mode of growth of the zone
shows that the first indication of an increase of protoplasm
is a thickening of the trabecule separating the granule-
containing spaces in the outer ends of the cells. Then,
as the granules disappear from the outer ends of the cells,
this thickening becomes more apparent, and affects more par-
ticularly those trabecule which are arranged in a direction
parallel to the long axis of the cell, so that these give in
optical section the impression of longitudinal bars or fibrillze.

The fine fibrillation which Eberth and Miiller,? Mouret,®
and others figure in the pancreatic cell may be seen in the
gastric chief cell, only in the small amount of unused proto-
plasm which is usually seen in the resting cell. This not

’ «Quart. Journ. Mic. Science,’ vol. xxxviil, part il, new series.

2 « Zeitschr. f. wissensch. Zool.,’ Bd. hii, supplement.
> Op. cit.

STRUCTURE OF THE MAMMALIAN GASTRIC GLANDS. 373

rarely presents peculiar sheaf-like and concentric forms
which are not unlike the figures published by Macallum,'
Mouret,” and others, of the nebenkerne in the pancreatic cells
of Batrachia.

A coarse fibrillation, similar to that exhibited by the outer
protoplasmic zone of the chief cell, I have also observed in
the cells of the cesophageal glands of the frog, and in the
serous glands of the gustatory region of the tongue of the
rabbit and dog, the fibrille being strongly chromophile and
iron-holding in each case. Solger® has described in the serous
cells of the human submaxillary gland, and Erik Miiller* in
the submaxillary glands of the guinea-pig, rod-shaped elements,
placed vertically in the bases of the cells, which stain intensely
in hematoxylin. It appears probable that these also are small
masses of protoplasm which owe their affinity for hematoxylin
to the fact that they are strongly impregnated with prozymogen.

Between the coarse fibrils in the base of the cell may be
seen small vacuoles containing fluid, and in the pepsin-secret-
ing cells of the stomachs of Batrachia the outer zone appears
rather vacuolated than regularly fibrillated. In osmic acid
specimens minute fat droplets may usually be seen in the outer
ends of the cells, the border granules of Langley.

The nucleus of the chief cell of the body of the fundus gland
is placed near the base of the cell in the resting condition.
It is spherical or slightly oval in shape, frequently exhibiting
slight irregularities of contour. <A well-defined chromatin
network and one or two large oxyphile nucleoli may be made
out in all stages of secretion. The latter are always invested
by a thin layer of basic chromatin, which frequently collects
in small masses at certain points on the periphery of the
nucleolus, as has occasionally been observed in the nuclei of
nerve-cells.

It will be seen from the foregoing that there are two salient

1 «Trans. Canadian Institute,’ vol. i, part 1, 1891.
2 Op. cit.

3 * Anat. Anzeiger,’ Bd. xi.

4 *Arch. f. mik, Anat.,’ Bd. xlv.

374 R. R. BENSLEY.

features in the structure of the chief cells of the body of the
gland, namely, the presence of granules of zymogen in a por-
tion of the cell of varying extent next the lumen, and the
presence in large amount in the protoplasm of the outer end
of the cell, and in that between the granules, of a kind of
chromatin called prozymogen, which stains strongly with
hematoxylin and other nuclear stains, and gives, after treat-
ment with sulphuric acid alcohol, a strong reaction for iron.
Further, the outer protoplasmic prozymogen-impregnated zone
exhibits a coarse fibrillar structure, which is a quite frequent
morphological feature of the ferment-secreting cell.

I have not been able to make out any primary structure in
the protoplasm, that is any differentiation into a firmer frame-
work and hyaline interstitial substance, although such may
exist and be masked in the ferment-secreting cell by the large
amount of deeply staining prozymogen present.

In the neck of the gland the chief cells are of smaller size
and more pyramidal in shape, and are present to the number
of one to four between each pair of border cells.

It has already been indicated in the description of the fresh
gland that the neck of the gland is devoid of zymogen granules.
This may be readily verified by the examination of sections
prepared after fixation in the alcohol bichromate sublimate
mixture, and stained in gentian violet. Such preparations
exhibit exactly the same division of the mucosa into two zones
as has been observed in the fresh material, due in this case,
however, to the fact that the zymogen granules with which
the chief cells of the body of the glands are filled stain
intensely in the dye, and thus give a deep stain to the body of
the gland, whilst the cells of the neck of the gland which con-
tain no granules have only their nuclei stained. Under the
high power not a single granule may be found in the chief
cells of the neck of the gland.

This absence of granules is not due to imperfect fixation, as
might be inferred, for one may see at the junction of the neck
and body-cells of both kinds, side by side, some containing
granules of zymogen, perfectly preserved and staining readily,

STRUCTURE OF THE MAMMALIAN GASTRIC GLANDS. 375

and others entirely devoid of these. Moreover one can
readily distinguish the two kinds of cells from one another,
even when, as occasionally happens, the fixation of the granules
has been imperfect. In such cases the zymogen-holding cell
still stains strongly, although diffusely, in gentian violet, whilst
in the neck cell only the nucleus stains.

In sections from the stomach of an animal that has been
digesting for several hours and stained in hematoxylin, the
division into two zones may again be observed. In this case
the extensive prozymogen-holding outer zone of the chief cells
of the body of the gland stains strongly, showing that the
distribution of the prozymogen corresponds to that of the
zymogen. The protoplasm of the chief cells of the neck of
the gland exhibits but little affinity for hematoxylin, and it is
only by means of the hematoxylin-iron reaction that it is
possible to demonstrate the presence of any cytoplasmic chro-
matin at all in these cells. In sections of alcohol-hardened
material, treated with sulphuric acid alcohol for three to six
hours in the warm oven, ammonium hydrosulphide or acid
ferrocyanide produce a scarcely recognisable reaction in the
protoplasmic portions of the neck cells. If, however, the
sections be treated with aqueous hematoxylin, according to
the method already outlined, a slight blue colour is obtained,
which, however, is not to be compared in intensity with the
strong reaction observed in the chief cells at the bottom of
the gland. The protoplasm of the cells of the surface, and of
the duct of the gland, gives a similar faint reaction for iron.

The chief cells of the neck of the gland, therefore, lack the
two most important features of the chief cells of the body of
the gland; they contain no granular zymogen, and they contain
prozymogen in such small amount that they are rather to be
compared in this respect with the mucus-secreting cells of the
surface and gland duct. It might be urged that these are
young or imperfectly differentiated zymogenic cells, which as
yet secrete their ferment in such small amount that its ante-
cedents do not appear in the form of granules in the cell. It
is not at present possible to determine whether or not these

376 R. RB. BENSLEY.

cells do secrete small quantities of pepsin, but a study of the
positive characters of the cell reveals that they possess different
staining characters from the chief cells lower down, and are
engaged in a different kind of secretion.

I shall describe in the first place the cells from the upper
portion of the gland neck (fig. 6, a). These are conical or
pyramidal in shape, and wedged in between the large ovoidal
border cells of this portion of the gland in such a way that in
vertical sections their broad end is usually directed towards
the lumen. In sections stained in hematoxylin and eosin the
cell is divided into two zones, an outer protoplasmic zone
staining readily in eosin, and an inner zone engaged in secre-
tion which stains with difficulty. The outer zone consists of
two elements, fine fibrille which join one another to form a
network, and a-hyaline substance filling up the interstices of
this network. The inner zone exhibits a structure similar to
that of the chief cells of the body of the gland, and for similar
reasons the accumulation of droplets of secretion forces the
protoplasm to simulate the appearance of a reticulum, although
the bars which compose it are not nearly so thick nor so
regularly arranged as in the chief cells lower down.

The secretion in the cells of the gland neck stains intensely
in Bordeaux R and indulin, and these dyes, particularly the
latter, have rendered me considerable service in determining
the distribution of this kind of secretion in the stomach, and
also in studying the secretive processes in the cells containing
it. Ihave found the most convenient method of applying this
dye to be in the form of Huber’s blood-staining fluid, con-
sisting of two grammes each of aurantia, eosin, and indulin,
rubbed up in a mortar with thirty grammes of pure glycerine.
This fluid is diluted with from 100 to 400 times its volume of
distilled water before use, and allowed to act on the sections
transferred to it from water for five to thirty minutes. The
sections are then washed in water, dehydrated, cleared in
benzole, and mounted. On examination it is found that the
red blood-corpuscles are stained yellow, the nuclei of all cells
a faint hematoxylin tint, the border cells and chief cells of

STRUCTURE OF THE MAMMALIAN GASTRIC GLANDS. 377

the body of the gland red, and the white fibres of connective
tissue blue. The protoplasm of the chief cells of the neck of
the gland also stains red, but the portion of these cells con-
taining secretion stains intensely blue (fig. 6). Staining in
indulin solutions alone does not give successful results, as the
stain in this form is also taken up diffusely by the other cells ;
but mixtures containing only eosin and indulin, or orange G
and indulin, give fairly good results. This colour reaction is
very little affected by the mode of fixation, as it may be
obtained in exactly the same features as indicated in fig. 6, in
preparations hardened in absolute or dilute alcohol, or aqueous
corrosive sublimate.

In sections stained in the indulin mixture the inner zone of
the cell appears vacuolated, exhibiting in optical section a
network of thick bars separating spaces, in which lies blue-
stained secretion. The network also stains, as a rule, much
more intensely than the secretion, indicating that it too is
impregnated with the indulinophilous substance. Not rarely
one may notice irregular clumps or flakes of deeply stained
substance lying in the spaces of the cell network, giving the
cell a coarsely granular appearance. This is never observed in
cells fixed in aqueous sublimate, and is probably due to the
rapid extraction of water by the alcoholic fixative, and a
consequent precipitation in this form of the solids of the
secretion.

It will be noted that the inner zone of these cells, when
stained in the indulin mixture, exhibits a coarse network, while
in hematoxylin eosin sections the network is finer than in the
chief cells of the body of the gland. The reason appears to
be that the solids of the secretion are precipitated along the
bars of the protoplasmic network, and when intensely stained
give these a triple thickness. In sections staimed in concen-
trated solutions of the indulin mixture, one sometimes finds
the true protoplasmic portion of the network stained red, and
it then presents the same characters as in the hematoxylin
eosin sections.

As one follows the neck of the gland downwards, it is seen

vou. 41, pART 3.—NEW SERIES, DD

378 R. R. BENSLEY.

that (fig. 6, 6) the portion of the cell engaged in secretion
becomes relatively greater, and at the lowest portion of the
neck of the resting gland two zones are no longer to be
recognised in the cell, the whole of which is concerned in the
formation of the indulinophilous secretion. These cells of the
lower portion of the neck of the gland are in structure exactly
like the cells of the mucous salivary glands in the secretion-
filled phase.

Following the gland in the direction of the free surface, it
is found that the cells of the uppermost portion of the neck,
and of a varying portion of the duct or stomach pit, are also
engaged in the formation of the indulinophilous secretion, the
portion of the cell thus engaged becoming relatively less, and
the protoplasmic portion greater as one approaches the surface.
The cells also, as the border cells become fewer, assume the
cylindrical shape and become longer. In these cells (fig. 6, e)
the secretion is found, for the most part, as a spherical mass in
the middle of the protoplasm of the cell near the nucleus.
There is always, however, a small amount of indulinophilous
substance diffused through the protoplasm intervening between
this and the free surface of the cell, and along the free surface.
Passing up the duct of the gland this mass gradually approaches
the free border of the cell again, and becomes less stainable
in indulin, thus passing by a gradual transition into the muci-
genous border of the surface cylindrical cells, which in the cat
stains but faintly in the indulin mixture.

A further difference between the cells of the gland neck
and those of the free surface is that the protoplasm in the
latter cells is of a denser character than that in the cells of
the gland neck,—that is, it contains a larger proportion of
the fibrillar element, and a smaller amount of interfibrillar
substance.

Thus, although the cells of the gland neck differ from those
of the free surface, both in general appearance and in staining
properties, it is impossible to discover at any point interven-
ing between the lowest part of the gland neck and the free
surface au abrupt change in the character of the cells. This

STRUCTURE OF THE MAMMALIAN GASTRIC GLANDS. 379

in itself would suggest a similarity in the nature of the
secretion products of the cells, as well as a similarity in their
mode of origin. There are, however, additional reasons for
regarding the secretion of the neck cells as of a mucous
nature. Staining in thionin gives a faint metachromatic red
stain to the secretion inside the cell, and in Mayer’s muche-
matein the secretion of these cells stains even more intensely
than that of the surface epithelium. I may add that in the
latter fluid the chief cells of the body of the gland stain not
at all.

The nuclei of these cells vary in shape with the amount of
secretion present. In those cells which have a well-defined
outer protoplasmic zone it is round or oval, and of regular
contour. In the cells that are filled with secretion the nucleus
presents the irregular, compressed, sometimes crescentic out-
line usually observed in mucus-secreting cells.

Mitotic divisions are, as Bizzozero! pointed out, most abun-
dant in the cells of the bottom of the duct and of the upper
portion of the gland neck, but they are by no means infre-
quent in the chief cells of the gland neck themselves, and
I have frequently observed cells completely filled up with
secretion with their nuclei in the various phases of indirect
division.

The nature of the difference between the mucus secreted by
the cells of the gland neck and of the surface epithelium is a
point of some interest, but one which it is very difficult to
determine. The fact that the cells in which mitoses are most
frequent contain secretion which stains with indulin, and that
the cells of the gland neck containing a similar secretion also
divide frequently, while the cells of the surface rarely undergo
division, would suggest that the difference is partly of the
nature which Bizzozero? has found to exist between the
secretions of the voung and old mucus-secreting cells of the
Lieberkuhnian glands of the intestine. This is not, however,
a sufficient explanation, as similar differences in staining re-

1 © Arch. f. mik. Anat.,’ Bd. xlii.
2 Tbid., Bd. xl.

380 R. R. BENSLEY.

actions are found in cells which are strictly comparable mor-
phologically. For example, the mucous border of the surface
epithelium of the stomach of the rabbit stains in indulin almost
as readily as the neck cells of the gland of the cat, and in the
intermediary zone of the rabbit’s stomach there are peculiar
transparent glands, the cells of which have all the features of
a mucous cell, but which stain differently from the cells of
the neck of the fundus gland, from the pyloric gland cells, and
from the surface epithelium. The only fluid which stains the
secretion of these cells is the muchematein solution of Mayer,
in which it becomes intensely blue.

The indulinophilous mucus-secreting cells do not cease
abruptly at the lower end of the gland neck, but a few may
be found among the ferment-secreting cells of the upper por-
tion of the body of the gland, as indicated in fig. 6, d, and
rarely one finds them even in the lowest parts of the gland.
These cells in haematoxylin and eosin stained sections stain
red, contrasting strongly with the more blue stained ferment-
forming cells. It seems to me very probable that these are
the cells observed by Pilliet,! Trinkler,? aud others, and re-
garded as stages in the transformation of chief cells into
border cells, or vice versa.

The cells of the neck of the gland exhibit the usual secre-
tion changes. In the first hours of digestion very little change
is to be noticed; but after twelve hours of secretion, and more
particularly when the stomach is mechanically stimulated by
sponge feeding, the secretion is seen to have been largely
passed out into the lumen of the gland, where it stains readily
with indulin and muchematein, and exhibits the strimgy or
spongy texture usually presented by mucus outside the cell.
At the same time the cell becomes reduced in size, and contains
relatively more protoplasm. Secretion changes may also be
observed in the cells of the bottom of the gland duct, the
rounded mass of mucus moving forward to the free surface
of the cell, where it is partly discharged.

1 © Journal de |’Anat.,’ &., 1887.
2 * Arch. f. mik. Anat.,’ Bd. xxiv.

STRUCTURE OF THE MAMMALIAN GASTRIC GLANDS. 381

It has seemed to me that there is an increase of indulino-
philous cells in the body of the gland during the first hours of
digestion, but I have not yet given this matter enough atten-
tion to enable me to speak with certainty concerning it.

The discovery of the different nature of the cells of the
neck of the gland affords a cytological basis for the division of
the gland into two regions, a neck and body.

The neck of the gland, as determined by the distribution of
the indulinophilous cells, varies in length in the different por-
tions of the stomach. It is shortest in the glands of the
greater curvature, where it forms about one third of the gland,
and longest in the lesser curvature, where it may comprise as
much as four-fifths of the entire gland.

B. The Pyloric Glands of the Cat.

These glands are, as Toldt! pointed out, branched tubular
glands, consisting of a deep pit or duct lined by a continua-
tion of the surface epithelium, and, opening into this, branched
tubules of various lengths exhibiting ‘a tortuous course in the
deeper layers of the mucosa.

Disregarding the anatomical divisions of the gland, one may
divide it, on a basis of the nature of the cells, into three
regions: a funnel-shaped duct, lined by a continuation of the
surface epithelium (fig. 7, a); the tubular body of the gland,
lined by the true pyloric gland cells (fig. 7, ¢) ; and a short por-
tion connecting these, and lined by cells of an intermediate
type (fig. 7, 5).

The true pyloric gland cells are, except as regards shape,
identical with the chief cells of the neck of the fundus gland.
They present all the characteristic staining properties of the
latter, and, if one compares similar secretion phases, are of
the same structure. They contain a secretion which stains in-
tensely in Bordeaux R, in indulin and Mayer’s muchematein,
and, as R. Krause has already pointed out, gives a meta-
chromatic red stain in thionin. ‘This secretion I regard, for
similar reasons, as of a mucous nature.

* «Sitzungsber, d, k, Akad. d, Wissensch., Wien,’ 1881.

382 R. RB. BENSLEY.

The cells of the pyloric glands contain neither in the fresh
condition nor in the hardened and stained gland granules
of zymogen. Furthermore, the protoplasm of these cells is
not chromophilous to nuclear dyes, and examination for masked
iron by the methods already indicated reveals but a trace of
prozymogen. The cells of the transitional portion of the
gland resemble most closely those of the surface, but contain
a relatively larger amount of protoplasm. By means of Mayer’s
muchematein it may be shown that these cells always contain
a small mass of mucus at their free ends, but this is so small
that, when using less decisive staining methods, it is readily
overlooked, and the cell regarded as entirely protoplasmic. In
these cells mitoses are fairly numerous, although they also
occur in all parts of the gland proper.

These pyloric glands present exactly the same features as
the neck and duct of the fundus gland,—that is, although
the true pyloric gland cells differ both in general appear-
ance and in structure from the cells of the surface, in
passing from one to the other along the gland at no point
can one discover an abrupt change from one type of cell to
the other.

The secretion phases in the pyloric glands of the cat are
peculiar, and are represented in figs. 8—10. Fig. 8 shows
the condition of the gland after a fast of twenty-four hours’
duration. The cells are comparatively long, with a spherical or
oval nucleus placed near the base of the cell. The stored-up
secretion is found, for the most part, in the form of a narrow
zone along the lumen of the gland, although it is not uncommon
to find a second mass of secretion in the deeper portion of the
cell near the nucleus (fig. 8,a@). It is also quite common to
find this second mass of secretion more stainable in indulin
than that along the free border, the cell then presenting an
appearance closely resembling that already described for the
cells of the lowest portions of the duct of the fundus glands.
Sometimes, as in the neck of the fundus glands, the mucin is
precipitated in the form of irregular flakes or granules. A pro-
longation of the fasting period even to four days does not

STRUCTURE OF THE MAMMALIAN GASTRIC GLANDS. 383

result in an increase of the amount of secretion stored up in
the cell.

Fig. 9 shows a similar gland from the pylorus of an animal
killed six hours after a copiots meal of meat. The cells are
increased in length, and the nuclei exhibit a tendency to become
irregular in outline and flattened in a direction at right angles
to the long axis of the cell. The portion of the cell between
the nucleus and the lumen is now entirely filled up with a
coarse meshwork, which stains strongly in indulin, and contains
a similarly staining secretion, although in many cells a division
of this into two masses, partly separated from one another by
a band of protoplasm, is still obvious.

Fig. 10 is from an animal killed twelve hours after a meal
consisting of several pieces of sponge soaked in beef juice. The
lumen of the gland is now increased in size, and the cells much
shorter than in the preceding phase. The nuclei have regained
their spherical shape, and are situated nearer the middle of the
cell. The secretion has been nearly all cast out of the cell, so
that only an extremely small amount on the free border is now
to be recognised.

The pyloric glands of the cat illustrate in a very striking
manner the fact that in gland cells the period of greatest
loading does not always coincide with the end of a normal
period of rest. Here it appears that the growth that takes
place in the cell during the rest period following the comple-
tion of digestion results merely in an increase of the proto=
plasm of the cell, and that during the first hours of digestion a
large portion of this is rapidly transformed into mucigen, so
that the real period of greatest loading is reached some hours
after digestion begins.

Further evidence in support of the view that the indulino-
philous cells of the neck of the fundus glands and the pyloric
gland cells of the cat are identical is afforded by an examina-
tion of the short intermediary zone. The first change that
one notices in passing from the greater curvature in the
direction of the pylorus is an increase of the indulinophilous
cells in the body of the gland. At the same time the neck of

384 R. R. BENSLEY.

the gland increases in length. As the pylorus is approached
these cells gradually replace entirely the zymogen-forming
chief cell, and the proximal portion of the pyloric region is
occupied by short glands, consisting of indulinophilous cells
and a few border cells. Ultimately the latter also disappear,
and we have the true pyloric gland.

C. The Gastric Glands of the Dog.

A section of the fresh mucosa of the greater curvature of
the stomach of the dog, examined in aqueous humour, reveals
much the same features as a similar preparation of the cat’s
stomach. The superficial portion of the mucosa, including
not only the pits but a large portion of the glands themselves,
is entirely free from the zymogen granules that crowd the
lower portion of the gland. The granules are smaller than in
the cat, and much more difficult to preserve, so that I have
been compelled to rely on the examination of the fresh glands
for control of my results.

In sections hardened in corrosive sublimate, or in the
bichromate sublimate mixture, the chief cells exhibit the same
structural features as those of the cat, namely, a regular mesh-
work separating spaces, which one may infer to have been
occupied in the fresh cell by the zymogen granules (fig. 11),
and in glands from a digesting stomach (fig. 12) of an outer
protoplasmic zone, which stains intensely and readily in
hematoxylin, exhibits a coarse fibrillar structure, and gives,
after treatment with sulphuric acid alcohol for three or four
hours at a temperature of 37° C., a well-marked reaction for
iron. In short, the cell contains both zymogen and prozy-
mogen in large amount.

The chief cells of the neck of the gland, on the other hand,
contain no granular zymogen, and only a trace of prozymogen.
These cells are of the same nature as in the cat, and present
an inner secretion zone of variable extent, which stains intensely
in indulin and muchzmatein. As in the cat, also, a gradual
change in the character of the cells is seen as the surface of
the mucosa is approached.

STRUCTURE OF THE MAMMALIAN GASTRIC GLANDS. 385

It is an interesting fact that among the other cells of the
neck of the fundus gland of the dog, cells may be found which
are in every respect similar to the cells of Stohr of the pyloric
glands, as described by Hamburger.! These are characterised
by their fusiform shape and their faintly staining protoplasm,
in which wavy fibrille may be seen. The pointed end of the
cell which reaches the lumen stains intensely red both in the
Biondi mixture and in eosin.

A point of some physiological interest is the varying length
of the gland neck in the stomachs of different dogs. It will be
seen from the following measurements that the differences in
the thickness of the mucous membrane are mainly due to
variations in the length of the body of the gland:

Thickness Length Length Percentage of

of of gland of body whole occupied

mucosa. neck. of gland. in zymogenesis.
No. 1 OG) Mims.) “26. Ini. =, 2 4. mnt 58°8
ee. TCU Pe 276 e, ee ear) ee rn bAcG
Pie haeviy 2652 cite coke Gal PENWR EMT pOGum: ) Kul SBbaR
eA eet G hs oy 0h 2868s epee a tu hs foes OTD,
tee oS fame Be) Me errrea were ad Uae : 73'3

The above measurements are taken from vertical sections
from the middle of the greater curvature, stained in the
indulin mixture, so that the junction of the body and neck of
the gland was sharply indicated by the cessation of the
indulinophilous cells. It is obvious that such differences in
the percentage of the whole gland engaged in ferment secretion
must be taken into consideration in estimating the relative
amounts of pepsin in the mucous membrane in different
periods of digestion.

The pyloric glands of the dog ‘are in all essential features
similar to those of the cat, the main difference being that the
ducts are much longer and rather cylindrical than funnel-
shaped, and the gland cells are much longer. The pyloric gland
cells contain a secretion which stains intensely in Bordeaux
R, in indulin, and in muchematein, and are in every respect
similar to the chief cells of the neck of the fundus gland.

1 * Arch, f, mik. Anat.,’ Bd, xxxiy.

386 R. R. BENSLEY.

The secretion phases are represented in figs. 138 and 14.
The resting cell (fig. 13) contains already a great deal of
reserve secretion, the nucleus being crescentic and compressed
against the base of the cell. In the exhausted phase (fig. 14)
the secretion is for the most part confined to the free border
of the cell, although many of the cells exhibit spherical masses
in the neighbourhood of the now spherical nucleus. The
protoplasm is also very much increased in amount.

All observers are agreed that the fresh pyloric glands of the
dog do not contain coarse granules. The only granules that
may be observed are small globules of fat, which are
particularly numerous in the bases of the cells. The mucous
secretion of the cell, too, occasionally, as in the cat, precipitates
in a granular or flaky form, although I have never observed
this in secretions hardened in aqueous sublimate solutions.
The protoplasm of the cell contains only a trace of prozymogen.

Conclusions.

In the cat and dog the fundus glands contain two kinds of
chief cells, those of the body and those of the neck of the
gland. ‘The former are engaged in the secretion of ferment,
and are characterised by the possession of a large number of
zymogen granules which occupy a portion of the cell of varying
extent near the lumen, and a protoplasmic outer zone of
various size which stains intensely in nuclear dyes such as
hematoxylin, and presents a coarse fibrillar structure. The
staining properties of this outer protoplasmic zone are due to
the presence in it of a kind of chromatin, which may stand in
a genetic relation to zymogen, and which has been named pro-
zymogen.

I regard the cells of the neck of the gland, and the pyloric
gland cells in the cat and dog, to be of the same nature, for
the following reasons :

The chief cells of the neck of the fundus gland and the pyloric
gland cells do not contain at any period of digestion zymogen
in the form of granules, and prozymogen is present only in
traces.

STRUCTURE OF THE MAMMALIAN GASTRIC GLANDS. 387

Both groups of cells are engaged in forming and secreting a
substance which stains intensely in Bordeaux R, in indulin, and
in Mayer’s specific mucin stain muchzematein, and which gives
the usual metachromatic mucin stain with thionin.

Both may be traced through a gradual transition to the
cylindrical cells of the surface.

The assertion that the secretion of these cells is mucin rests
as yet on the evidence afforded by the structure of the cell,
and its staining in thionin and Mayer’s muchematein. No
method has suggested itself to me for securing the secretion of
these cells unmixed with that of the surface epithelium, and
testing it for mucin. It is, however, on evidence of the same
character that the large transparent neck cells of the batra-
chian gland have been accepted as mucus-secreting cells.

The chief cells of the neck of the fundus glands, and the
pyloric gland cells of the cat and dog, are the physiological and
morphological homologues of the mucous neck cells and pyloric
gland cells of the frog, with which they correspond in situation,
structure, functional changes, and staining. It is interesting
to note that these cells in the frog and other Batrachia stain
intensely in indulin when treated with mixtures of eosin,
indulin, and aurantia.

The cells of the lowest portion of the gland duct or pit are
of a type intermediate between the surface epithelium and the
mucin-secreting neck cells or pyloric gland cells, and form
the fundamental type to which the origin of both must be
traced. Moreover the researches of Bizzozero! have made it
extremely probable that these, although themselves secreting
cells, are constantly dividing, and so form the young elements
from which both the cells of the surface and the neck cells, or
the pyloric gland cells, are recruited, although it must be
admitted that the two latter are partially replaced by division
among themselves.

Mucin-secreting neck cells are present in the fundus glands
of the mink, rabbit, mouse, rat, squirrel, ground hog (Arctomys),
chipmonk (Tamias striata), pig, and sheep. The relation

1 «Arch. f, mik. Anat.,’ Bd. xlii,

388 R. R. BENSLEY.

existing between these and the pyloric gland cells on the one
hand, and the cardiac glands on the other, will be discussed in
a second paper now in course of preparation.

EXPLANATION OF PLATE 29,

Illustrating Mr. R. R. Bensley’s paper on “ The Structure of
the Mammalian Gastric Glands.”

Nore.—All figures are drawn with the aid of the camera lucida, and, with
the exception of figs. 1 and 7, as seen under the Zeiss apochromatic objective
(2 mm. apert.) and compensation ocular 8. In fig. 7 the same objective and
compensation ocular 4 were employed. Figs. 2, 3, 11, and 12 are reproduced
from grey pencil drawings, and do not indicate the colours of the original
preparations.

Fic. 1.—A photomicrograph of a fresh unstained section of the mucosa of
the greater curvature of the stomach of the cat.

Fie. 2.—Transverse sections of the Jower ends of three fundus glands of
the cat after a fast of twenty-four hours’ duration ; hematoxylin and eosin.

Kia. 8.—Transverse sections of the lower ends of three fundus glands of
a digesting cat, stained in hematoxylin and eosin. a. Indulinophilous mucin-
secreting cells. All other chief cells show a deeply staining outer zone with
coarse fibrillar structure.

Fie. 4.—Fundus gland from a digesting cat, stained in gentian violet. The
outer prozymogen-holding zone stains metachromatically.

Fie. 5.—Fundus gland of cat fixed in alcohol. Section treated with sul-
phurie acid alcohol, then with aqueous hematoxylin, and finally stained faintly
in safranin. The prozymogen-holding outer zone gives a strong reaction for
iron, which has been unmasked by the action of the acid alcohol.

Fic. 6.—Vertical section of the duct, neck, and upper portion of the body
of a fundus gland ofa fasting cat ; indulin, eosin, and aurantia. a, d. Mucin-
secreting chief cells of the neck. d. Ferment-secreting chief cells of the
body of the gland. f£ Mucin-secreting cell of the body of the gland.

Fic. 7.—Pylorie gland of cat aftera fast of twenty-four hours’ duration ;
indulin and eosin. a. Duct. 4. Intermediate portion. ¢. Body of gland.

Exhibits the gradual transition in passing from the body of the gland to the
duct,

STRUOTURE OF THE MAMMALIAN GASTRIC GLANDS. 389

Fies. 8, 9, and 10 are sections from the lowest portion of the pyloric gland
of the cat in different secretion phases, and stained in indulin and eosin.

Fig. 8.—From an animal after twenty-four hours’ fast.

Fic. 9.—From an animal killed six hours after feeding.

Fic. 10.—From an animal killed twelve hours after sponge-feeding.

Wie. 11.—Body of fundus gland of dog after a fast of twenty-seven hours ;
hematoxylin and eosin.
' Fie. 12.—The same from a digesting dog; the cells here contain an un-
usually large amount of prozymogen-holding protoplasm.

Fig. 13.—Cells from the pyloric gland of the dog after twenty-seven hours’
fast ; indulin and eosin.

Fig. 14.—Cells from the pyloric gland of the dog twelve hours after sponge-
feeding; indulin and eosin.

a a die ray Tiir.: mM * gstadaggane doth Py

HES. noe 7 a

a

on)
‘ag Nena EA ae peas PY ay
vA oie ie ee + Me haa oe

; ‘pm eats Wier Vn ret 15h a ia I=, aie ie

ere rw Pa aot oe bE Nien ee nohie Aa

atet amie a fhe eA Sea rt oF Gl tee & &f-

, 7% ieee mii i ‘ist = de ie » pte. ih a
ee ad a er eee vats We yh
ond wear ok By ans oars alt

>

paaliy saath ty Sacig'| ipineiie Sie ell

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATER. 391

On Certain Green (Chlorophylloid) Pigments
in Invertebrates.
By

Marion I. Newbigin, D.Sc.,
Lecturer on Zoology in the Edinburgh College of Medicine for Women.
(From the Laboratory of the Royal College of Physicians, Edinburgh.)

With Plates 30 and 31.

ConTENTs.
PRELIMINARY.

A. Chetopterin.
(1) General characters.
(2) Action of acids.
Comparison with bonellin.
General characters of acidified solutions.
(az) 'The ether solution.
(4) The blue acid solution.
(c) The green acid solution.
General conclusions as to action of acids.
Characters of the green acid derivative.
(3) Action of alkalies.
(a) Ammonia.
(4) Caustic soda and potash.
(4) Action of salts.
Summary.
B. “ Enterochlorophyll.”
(1) Previous investigations.
(2) Mode of occurrence.
(3) Characters of solutions.
Relation to chlorophyll.
Characters of the associated lipochrome.
(4) Action of acids.
(5) Other reactions.
(6) Relation to chetopterin.
(7) Distribution.
c. Bonellin.
(1) Comparison with cheetopterin.
(2) Relation to thalassemin.

(3) Other green pigments in Invertebrates.
ConcLusion.

392 MARION I. NEWBIGIN.

PRELIMINARY.

THE existence in various Invertebrates of pigments present-
ing a remarkable resemblance to chlorophyll has long been
known. Of these pigments the most familiar are the green
pigments of the worms Bonellia and Chetopterus, and
the pigment described by Dr. MacMunn as enterochlorophyll.
In each case an identity with chlorophyll has been asserted by
different authors. The green pigment of Bonellia has been
described as chlorophyll by Schmarda, Schenk, and others,
although the researches of Sorby, Geddes, and Krukenberg
long ago demonstrated the erroneous nature of the description.
Similarly, Professor E. Ray Lankester, who discovered the
peculiar pigment of Chetopterus, called it a “chloro-
phylloid” substance, and at one time placed Chetopterus
in the list of chlorophyll-containing animals, although he no
jonger maintains this position. In the case of “ entero-
chlorophyll,” however, Dr. MacMunn’s application of the term
chlorophyll has not been seriously challenged.

During some work on the pigments of the Crustacea I
came across a statement by Dr. MacMunn that enterochloro-
phyll occurs at least in some cases in the “liver” of these
forms. Not being able to find it there readily, I resolved to
study its properties in the organs in which it was first de-
scribed, namely, the digestive glands of the Mollusca. Soon
after I had begun this work Professor Lankester sent me
some solutions of the pigments chzetopterin and bonellin, and
asked me to make a chemical examination of them. I found
so much resemblance between the cheetopterin solutions and
solutions of enterochlorophyll as to make it desirable to study
the two pigments side by side. In the interim Professor
Lankester’s own paper on chetopterin appeared, with spectro-
scopic observations by Drs. Benham and Engelmann.

Iam much indebted to Professor Lankester for the solutions
of cheetopterin and bonellin which he sent, to Professor W. A.

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 395
Herdman for solutions of his new pigment, thalassemin, and
to Professor W. M‘Intosh for various green Invertebrates.

A. CHAETOPTERIN.
(1) General Characters.

As cheetopterin occurs both in larger amount and in a purer
state than enterochlorophyll, it is convenient to begin with
some account of it.

It is not necessary to say much with regard to its general
characters and mode of occurrence, for these points are fully
discussed in Professor Lankester’s paper (5). As shown
there, the pigment occurs in the form of numerous minute
granules in the cells of the mid-region of the gut. In addition
to these small green granules, there occur in some of the gut
cells much larger round vesicles of a more brownish tint.
These bodies seem to me also to contain the pigment. The
pigment further occurs in oily drops mingled with the contents
of the gut.

An interesting microchemical reaction which indicates the
presence of chetopterin, or the related pigments, is afforded
by the use of strong acid, preferably hydrochloric. Ifa drop
of this acid be placed on the gut wall of Chetopterus when
spread out on a slide, the cells, previously greenish brown,
turn a vivid green, or more rarely blue. ‘The test is not
absolutely diagnostic, inasmuch as certain of the lipochromes
give a dirty greenish tint on the addition of strong hydro-
chloric acid, but the lipochromes in animal tissues are usually
uniformly diffused and not granular, and they only give the
reaction in the dry condition, the addition of water or alcohol
destroying the colour. Chetopterm granules do not lose
their colour on the addition of water or alcohol, except in so
far as a green pigment dissolves out. Further, in the case
of chzetopterin the green colour can be removed by the addi-
tion of a little alkali, and the process repeated indefinitely,
which is again impossible in the case of the fleeting lipochrome
colour. In the general case also the lipochrome would be dis-

voL, 41, PART 3.—NEW SER, EE

394 MARION I. NEWBIGIN.

tinguished by its yellow or orange tint. The reaction, which
does not seem to have been previously described, is of some
importance, because it affords a readily available method of
recognising the pigment in cases where the amount may be
too small to make it easy to demonstrate its nature in any
other way.

As shown by Professor Lankester, the pigment readily dis-
solves in cold methylated spirit to form a solution varying in
colour according to its strength. Very strong solutions are a
deep brownish-yellow colour with greenish lights; more dilute
ones are grey-green, while those which are very weak may be
almost pure green. All the solutions show a strong blood-red
fluorescence. The want of definiteness in the colour is very
characteristic, and is no doubt due to the complex spectrum,—
that is, to the differential absorption. The bands are not all
of equal intensity, and as those at the violet end disappear
when the solution is diluted before those at the left end of the
spectrum, the effect of diluting is naturally to increase the
amount of green in the tint.

As shown by Professor Lankester, the spectrum of the
freshly extracted solution shows four bands, a very strong one
over the line C, and three others lying respectively to the left
of the lines D, E, and F. In strong solutions there is also a
shading at the right of the D line. For the details of the
spectrum reference should be made to Professor Lankester’s
paper; it is figured in Plate 30, fig. 1, for convenience of
comparison.

On the addition of a considerable amount of acid to the
solution the colour changes to a dusky blue without loss of
the blood-red fluorescence. The spectrum shows the original
four bands, of which, however, the first two at least have
shifted slightly to the right, and an additional band, which
lies to the right of the D line, in the position of the shading
already noticed in the normal solution. Besides the slight
alteration in position, the original bands in the green and
violet have greatly diminished in intensity. ‘This is probably
due to the dilution and the slight turbidity of the solution,

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 395

It occurs to such an extent as often to almost obliterate the
F band; thus this band is figured as absent in Dr. Benham’s
maps, though Dr. Engelmann’s chart indicates its presence.

By neutralisation with alkali the original spectrum and
colour may be restored—a marked difference from the pheno-
mena exhibited by solutions of chlorophyll.

(2) Action of Acids on Chetopterin.

The changes which solutions of cheetopterin undergo on the
addition of acids are of so striking a nature that they form a
natural starting-point for any investigation.

The acid usually employed was hydrochloric acid in con-
ceutrated solution. The employment of gaseous hydrochloric
acid produced the same effect as the solution; it did not
produce a precipitate, as it does in the case of solutions of
chlorophyll. It was found, further, that, as is the case with
bonellin according to Krukenberg, when other acids are
employed the amount necessary to produce the blue colour
depends upon the strength of the acid, and not upon its
chemical nature. Thus a much larger amount of acetic acid
is necessary than of hydrochloric, while excess of nitric acid
is apt to carry the reaction beyond the blue stage, and pro-
duce a brown solution from which the original pigment cannot
be recovered. This is a product of decomposition.

Comparison with Bonellin.—In its relation to acids
cheetopterin shows a marked analogy to bonellin. The changes
which bonellin undergoes have been studied by Krukenberg
(3), and his results may be briefly detailed.

Krukenberg found that the addition of a considerable
amount of strong acid to a solution of bonellin turned his
bright green solution violet without destroying the fluorescence,
while a large excess turned it pure blue without any trace of
fluorescence. He figures three sets of spectra corresponding
to three stages in the action of the acid. ‘The first of these
he describes as indicating the existence in the solution of a
mixture of bonellin and an acid derivative; the other two as
representing two distinct acid derivatives, which he calls

396 MARION I. NEWBIGIN.

respectively bonellidin and acidobonellin. Now Professor
Lankester has shown that the solution which Krukenberg
regarded as that of neutral bonellin was in reality alkaline
bonellin, the neutral solution being not green but dusky grey.
We are thus entitled to omit Krukenberg’s first stage from
our comparison with chzetopterin, and consider only his
bonellidin and acidobonellin solutions. We have already
described the colours of these solutions, and as the pigments
were not isolated there remains only the spectroscopic
characters. It is not necessary at present to discuss these in
detail, for an account of the alteration in position undergone
by the dominant band in the red is sufficient for our purpose.
Bonellin, like cheetopterin, exhibits in neutral solution a very
strong band in the red. On the addition of acid this band,
without marked diminution in intensity, shifts its position
towards the right. This is an old observation. Krukenberg
found, however, that a further addition of acid in very large
excess caused the band to move back until it almost occupied
its original position. It was this double movement, com-
bined with the colour changes in his solutions, which induced
him to reject Sorby’s suggestion that the acid has merely a
physical effect, and to put forward the theory of the existence
of two acid compounds. The subject has not apparently been
again studied.

Passing from solutions of bonellin to those of chztopterin,
we find that here again acid has a duplex effect. While a
small amount of acid produces the blue colour already described
by Professor Lankester, I find that a very large excess turns
the blue solution a pure clear green, with diminished fluores-
cence. Further, an examination of the spectra shows that, as
in bonellin, the band in the red shifts first to the right, then
with excess of acid back to its original position. The band of
chetopterin does not occupy exactly the same position as that
of bonellin, nor is the movement so extensive, but there is at
least an analogy.

Characters of Acidified Solutions.—The question
whether the changes which occur on the successive addition

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 397

of acid do or do not indicate the existence of acid derivatives
may seem at first sight one which can be readily determined.
When, however, it is recollected that we are dealing not with
a well-defined chemical substance capable of being invariably
recognised by definite reactions, but with an unstable, unknown
substance, which, apart from the presence of impurities, may
be a mechanical mixture of several pigments; when, further,
it is found that virtually the one available test is that of the
spectrum, whose validity as a test is the point to be proved, it
is then possible to obtain some notion of the difficulties. It is
curious to note that in point of fact, in spite of the frequent
occurrence of pigments whose spectra change on the addition
of acids, Schunck’s beautiful work on chlorophyll (10) seems
to be the only case where the reasons for the changes have
been fully investigated. The difficulties mentioned above
perhaps afford a ready explanation of the blanks in our know-
ledge of such pigments.

The most obvious characteristics of the acidified solutions
are, of course, their spectra. The spectrum of the blue acid
solution has been completely mapped by Professor Engelmann,
whose method shows its peculiar characteristics ina very striking
manner. My ownobservations were made both with a Sorby’s
microspectroscope and the large double prism spectroscope of
the Cambridge Instrument Company. The results are recorded
here (see fig. 2) only because they are essential to the course
of the argument; they agree very closely with those of
Engelmann, but on account of the indefiniteness of the margins
of the bands are less accurate,—that is, the point given as the
centre of the band does not always coincide with the point of
maximum absorption.

The general characters of the spectrum of the blue acid
solution have been already described. In regard to detail, the
most important point is the character of the dominant band in
the red. In freshly extracted solutions of chzetopterin this
band, as measured by a table spectroscope, has the following
position :

\ 679 — A 643, centre = A 661 (see the first band of fig. 1).

398 MARION I. NEWBIGIN.

On the addition of a considerable amount of hydrochloric acid
its limits are as follows :

669 — X 637, c. = X 653 (see the first band of fig. 2).

It will thus be seen that in spite of the fact that the solution has
been very considerably diluted, the actual extent of the band
has not been greatly diminished. Further, while the left-hand
side of the band remains fairly sharp and well defined, the
right-hand side is very indefinite, shading gradually off, so that
exact measurement is virtually impossible. This peculiar
appearance has been described by Dr. MacMunn (6) in the
case of solutions of enterochlorophyll as “a band super-
imposed upon a shading.” Now if excess of acid be added to
this solution until it turns green, this shading at the right
hand disappears, and the band recovers approximately its
original position. Thus it may stand as follows:

A 677 — Xd 647, c. = X 662.

As to the other bands of this green acid solution, they may be
present as in the blue acid one, but in the general case they are
exceedingly faint, and present merely as traces. Their changes,
if they do change, can be followed with much less certainty
than those which are undergone by the very distinct band in
the red. These changes are not sudden, but take place very
gradually, and can be watched step by step when acid is added
to a solution suspended in front of the slit of the spectroscope.
As the acid is added the band moves to the right until the
movement reaches its maximum, and then on further addition
of acid it moves back to its original position. Further, if
alkali be added to a very strongly acid solution the band shows
the same change of position as when little acid is added to a
normal solution, and excess of alkali restores it to its original
position, just as does excess of acid in the other case. In other
words, dilute acid produces a change in the position of the band,
which is reversed by strong acid ; and it is unimportant whether
the dilute acid is directly added to the solution, or produced by
removing some of the acid from a highly acidified solution. In
the case of bonellin, Krukenberg states that the strongly

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 399

acidified solution is without fluorescence. In the case of
chetopterin the fluorescence ceases to be marked when the
solution is very strongly acid, but when a ray of bright light
is thrown upon the vessel containing the solution placed against
adark background, it rarely fails to show a trace of the charac-
teristic blood-red colour. The more detailed characters of the
blue and green solutions it may be well to study separately.

The Blue Acid Solution.—When acid is added to an
alcoholic solution of chtopterin the solution becomes more
or less turbid in appearance. If a considerable amount of acid
be added, and then water, and the whole shaken in a separation
funnel, the superficial layer of ether becomes pale green, the
lower layer a pure clear blue with marked fluorescence.

(2) The ether gives more or less distinctly the original
four bands of chetopterin without the band at the right of D.
When weak the band in the red has its centre about A 661—
the position of that of the original solution ; but if the ether
contains much pigment the centre of the band tends to shift to
the right, and to approach more closely that of the band of the
acid solution. In other words, the spectrum may be that of
fig. 1, but there is a tendency for the first band of it to be
replaced by the first band of fig. 2, or by a band intermediate
between the two. When the ether is evaporated and the pig-
ment dissolved in methylated spirit a green solution is formed,
with little fluorescence. On adding acid the solution does not
turn blue; it shows at first little alteration, and later becomes
brownish. The pigment is thus altered,—is not identical
with neutral chetopterin. At the same time it is to be noticed
that the amount of alteration varies greatly. In some cases
the ether seems to give the full four-banded spectrum of fig. 1
without alteration; in other cases it may give only a band in
the red, corresponding to the first of fig. 2. The green tint
and the absence of the power of giving a blue colour with acid
are the most constant characters.

(6) The blue acid solution left after shaking with ether
gives the same spectrum as it did before the process. When
strong it is a beautiful pure blue colour, but in some cases

400 MARION I. NEWBIGIN.

nearly all the colour can be removed by successive shaking
with ether.

If ammonia be added to this blue solution until it remains
only slightly acid, the blue colour greatly diminishes in in-
tensity, and on shaking with ether the ether becomes brownish
green in colour, and gives beautifully the original four-banded
spectrum (fig. 1). On evaporation the ether leaves a dull
green pigment, which dissolves in methylated spirit to form a
brownish-green strongly fluorescent solution, which turns blue
with acid, and shows all the characters of the original solution.
The whole of the pigment cannot, however, be extracted from
the acid solution in this way, for when ammonia is added in
slight excess ether does not extract any pigment, chztopterin
being readily soluble in ammonia to form a solution in which
the band in the red only is distinct. Instead of attempting to
neutralise the acid solution, an easy method of precipitating its
pigment is to add pieces of marble to it. Violent effervescence
occurs, the blue colour is completely lost, and a dull green
precipitate falls, leaving the solution colourless. The pre-
cipitate after washing readily dissolves in methylated spirit,
and yields a solution of blue-green colour, more or less
fluorescent, and giving a four-banded spectrum. The band in
the red is distinct and broad. In strong solution it has the fol-
lowing position :

r 669 — A639, c. = A 654;

that is, it has almost the same position as that of the band in
the red in the blue acid solution. The other three bands have
the same position as in neutral chetopterin solutions; the
band to the right of Dis absent. The spectrum is thus that
of fig. 1, but with its first band replaced by that of fig. 2. On
standing with acid this solution turns brown; the position of the
band in the red does not alter, although the band at the right
of D may appear. ‘The pigment thus appears to be identical
with that obtained by shaking the acid solution with ether.

In attempting to explain these reactions, the first point is
to consider whether they support the view that in the case of

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 401

chetopterin there is an acid derivative corresponding to
Krukenberg’s bonellidin. As already seen, bonellidin is the
pigment supposed to be present in a solution of bonellin turned
violet by acid; it is characterised by its colour and its spec-
trum. The spectrum is distinguished from that of neutral
bonellin by the different position of especially the band in the
red, the appearance of a new band anaiogous to that at the right
of D in the case of chzetopterin, as well as by the loss of the band
in the violet. Blue acid solutions of chztopterin are character-
ised by the (slight) change of position of the band in the red, the
apparent loss of the band in the violet, and a slight change in
the position of the other bands, as well as by a colour-change
as contrasted with neutral solutions. But the band to the
right of D is present as a shadow in neutral solutions, and is
only marked when the solution examined actually contains
acid. Thus, if a solution of the pigment in anhydrous ether
be vigorously shaken with acid, this band, previously a mere
shadow, becomes suddenly distinct. If the ether be then
carefully washed with distilled water to remove any trace of
acid, the band disappears. This seems to me to prove that the
appearance of this band is not due to the formation of a com-
pound, for it is difficult to believe that a true compound could
be so extraordinarily unstable. Krukenberg noticed a similar
fact, that an evaporation of solutions of bonellin containing a
volatile acid caused the disappearance of the corresponding
band; he calls this the regeneration of bonellin from bonellidin,
but I cannot agree to this conclusion.

Again, the loss of the band in the violet is a point of no
importance, for it may be quite easily explained as the result
of the slight turbidity of the solution, and the change in posi-
tion of the other bands is too slight to be of any moment. I
thus dissent from the view that the spectrum and the colour
of the blue acid solution are together diagnostic of the ex-
istence of a new pigment defined by these characters, and
believe that the point which requires explanation is that the
action of dilute acid is to produce a permanent alteration in
the position of the band in the red solutions in cheetopterin, and

402 MARION I. NEWBIGIN.

a permanent loss of the power of giving a blue colour with
acid. To these points we shall return after considering the
characters of the green acid solution.

(c) The Green Acid Solution.—The deep green solution
produced by adding excess of acid yields no pigment to ether.
On adding ammonia, and again shaking with ether, unaltered
cheetopterin of brownish-green tint is extracted by the ether.
On the addition of marble to the acid solution a precipitate
falls of dull green colour. This dissolves in methylated spirit
to form a green solution with a mere trace of fluorescence.
The spectrum shows four bands: the band in the red is exceed-
ingly strong; the other three have the same position as in un-
altered cheetopterin, and vary greatly in intensity. The position
of the band in the red is as follows:

\ 666 — 4634, c. = 1650.

The band has thus much the same position as the first band in
fig. 2 would have if the shadow to the right were to become
distinct. The addition of acid to this solution does not change
the position of the band in the red, but it ultimately turns the
solution a brownish colour.

When the dull green precipitate obtained by the addition of
marble is treated with methylated spirit, there remains un-
dissolved a brownish residue which is insoluble in ether or
methylated spirit, and which is unaffected by acid or alkali.
A trace of similar residue remains when the precipitate from
the blue acid solution is treated with methylated spirit or
ether.

General Conclusions as to the Action of Acids.

If we survey generally the action of acid on solutions of
cheetopterin, we see that the pigments precipitated respectively
from the blue and green acid solutions by the action of marble,
agree with one another, and differ from the original cheetopterin
in forming in methylated spirit solutions which are pure green
or bluish green in colour, very slightly fluorescent, and which
turn brown and not blue or green on the addition of acid.
They differ from chetopterin and from each other in the

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 403

position of the band in the red, about \ 653 in the one case
and 1.650 in the other, while they may or may not exhibit
distinctly the other bands of the original chzetopterin solution.
In this last respect there is much variation, but it is unneces-
sary here to repeat the numerous spectroscopic observations
which were made in order to find, if possible, a valid explana-
tion of the variation. There is also variation in tint in both
solutions. When the chetopterin bands are distinct the tint
of the solution exhibits an approximation towards the indefinite
brown-green colour of the original chetopterin ; when less dis-
tinct the solution isa bright blue-green; when they are barely
visible the solution is a pure dull green. These observations
ultimately convinced me that both solutions contain a mixture
of pigments, the two components varying in amount. After
many trials I succeeded in proving this as follows.

(1) The precipitate from the blue acid solution obtained by
adding marble was dissolved in ether, the ether placed in a
separation funnel, and concentrated hydrochloric acid cautiously
poured in. The acid became indigo-blue at the line of junc-
tion, deep green further down, and ultimately green through-
out; the ether remained pure pale green. When examined
with a spectroscope the ether showed two bands in the red
(see fig. 3), one with the centre at (661—the band of the
original chetopterin ; and another with the centre at about
\ 641, an entirely new band. If the ether be examined in very
thick layer the two bands approach so near as to be distin-
guished with difficulty, and then appear like one broad band
with a centre about 4 653 (cf. figs. 2 and 3). In addition to
these two bands the ether in thick layer shows a trace of a
band at about A 601, which corresponds to the second band of
the original chetopterin solution. The acid gives a spectrum
showing the band in the red characteristic of chetopterin, and
traces of the other bands,—that is in essence the spectrum of
fig. 1.

(2) The precipitate obtained by the addition of marble to
the green acid solution if dissolved in ether and treated with
concentrated acid gives the same results. Again, the ether

404 MARION I. NEWBIGIN.

shows a double band in the red, the one to the right being
more distinct. The right-hand band is entirely new, the
other is one of the bands of chzetopterin. By successive
treatment with acid it is possible to remove the left-hand
band from the ether almost entirely, and leave only the new
band at 1 641,—that is the second band of fig. 3; the solution
is then pure green without fluorescence.

The changes which chzetopterin undergoes on the addition
of acid I therefore explain as follows:

When a small amount of acid is added to a neutral solution,
a small amount of a new pigment of a green colour is formed,
which shows one band with its centre at (641. This band,
however, is so near the broad band of chetopterin that the
two overlap, and in strong solution the apparent result is to
shift the red band to the left, as well as to produce that inde-
finiteness of margin of which we have already spoken (see
fig. 2). The other slight peculiarities of the spectrum of
acidified solutions I believe to be analogous to the changes
seen in the position of the bands of bonellin when the solvent
is varied, and not to indicate chemical change.

When excess of acid is added to the solution the green
derivative is not destroyed, but, owing apparently to its imper-
fect solubility in strong acid, its band disappears, leaving
only the original chetopterin band. This is, as I believe, the
explanation of that shifting of bands of which we have already
spoken (p. 896). The disappearance of the band of the deriva-
tive in strong acid solution is paralleled by that of the other
chetopterin bands, which become exceedingly indistinct when
much acid is present in the solution, although by removal of
the acid the original pigment can be recovered.

The changes which occur in the spectrum of chzetopterin
solutions when acidified I thus believe to be due to the blend-
ing of the spectrum of unaltered chzetopterin and that of an
acid derivative. The colour-change is, I believe, similarly due
to the mixture of pigments present.

Characters of the Green Acid Derivative.—The
green derivative formed by the addition of acid to a solution

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 405

of cheetopterin is not very easy to obtain unmixed with cheto-
pterin. If the pigment precipitated by marble from an acidi-
fied solution be treated with ether, the ether becomes green as
already stated, leaving behind a brownish insoluble residue.
If the ether be repeatedly treated with concentrated acid, the
acid removes at least the greater part of the chetopterin,
leaving the ether pure green; the reaction depends on the
fact that chetopterin is more soluble in concentrated acid
than in ether ; while the derivative is, on the other hand, solu-
ble in ether, but not in strong acid. The spectroscopic exami-
nation of the ether usually, however, shows a trace of a band
at A 661,—that is the strong band of cheetopterin, in addition to
the band at (641 characteristic of the derivative (see fig. 3).
Another method is to treat the precipitate from the acid solu-
tion with methylated spirit, and add lead acetate to the solu-
tion so obtained. The chetopterin is precipitated, leaving the
derivative in the solution.

The green derivative obtained in either of these ways forms
in ether or alcohol a pure green solution without fluorescence,
and giving a one-banded spectrum. On evaporation it is
obtained in the form of green oily drops, which are readily
soluble in dilute ammonia. It is not soluble in pure water.
It was not obtained in large amount, and is apparently formed
by the acid splitting chetopterin into a brown insoluble sub-
stance and the derivative: the derivative cannot be recon-
verted into chetopterin. In the absence of detailed chemical
investigation this pigment will be simply called the acid
derivative.

(3) Action of Alkalies.

(a) Ammonia.—The action of ammonia upon solutions of
chetopterin is somewhat peculiar. Ifa few drops of ammonia
be added to the alcoholic solution, there is, as noticed by
Professor Lankester in the case of caustic soda, a colour-change
to yellow-green without distinct change in the spectrum. If
the alkaline solution be allowed to stand for some time, how-
ever, the colour becomes a deep pure green with strong fluores-

4.06 MARION I. NEWBIGIN.

cence, and the spectrum shows, in addition to the four bands of
chetopterin, a new band at A 625 (see fig. 4); the first band
also tends to shift slightly to the right. There is at the same
time a slight precipitate of a green colour. On prolonged
standing with ammonia the solution loses its fluorescence, and
the bands, with the exception of two in the red, tend to
disappear (see fig. 5). The addition of acid in excess to this
solution does not produce a blue colour. A more speedy and
certain method of attaining these results, however, is to boil
solutions of chetopterin with ammonia for some time on the
water-bath, then evaporate to dryness, and extract the residue
with methylated spirit.

Only a part of the residue is soluble, a portion remaining
undissolved, and being of a green colour.. This is identical
with the precipitate already described as being formed when
ammonia is added to alcoholic solutions, and is apparently an
ammonia compound; it is insoluble in alcohol and ether.
After treatment with dilute acid it yields to methylated spirit
a solution containing a mixture of chetopterin and its acid
derivative. The reason for this will be considered under the
action of salts on chetopterin; it is sufficient for the present
purpose to note that chetopterin, like certain of the lipo-
chromes, is at least in part precipitated by ammonia, the
compound being insoluble in alcohol or ether.

While heating with ammonia converts in this way a portion
of the chetopterin into an ammonia compound, another portion
is converted into a derivative. This is present in the alcoholic
extract of the residue after evaporation of the alkaline solution.
There is usually present in addition some unaltered cheto-
pterin, which can be precipitated by addition of lead acetate,
when the derivative remains in solution. The derivative is
characterised by its green colour, its peculiar spectrum con-
sisting of two bands in the red, one about A 650 and the other
at \ 625 (see fig. 5), its want of fluorescence, and finally its
solubility in water as well as in dilute alkali. It may be
formed, as already indicated, by allowing alkaline solutions of
chetopterin to stand for some time in the cold, but usually

ON OERTAIN GREEN PIGMENTS IN INVERTEBRATES. 407

the presence of the derivative in such solutions is only
indicated by the appearance of a new band at A 625, and
sufficient unaltered cheetopterin remains for the solution to
retain its fluorescence and its original four bands in addition
to the new one. The point is emphasised because it seems to
have some bearing upon the characters of bonellin. When
the ammonia derivative is formed by heating chetopterin with
ammonia, the band at A 625 is often difficult to demonstrate.

(6) Caustic Soda and Potash.—Very dilute solutions of
these when added to the alcoholic solution produce the same
effect as ammonia,—that is, there is a slight green insoluble pre-
cipitate, and the solution remains pure green, showing two
bands in the red, one at A 661 and one at A 625, with more or
less distinct traces of other bands, and yields a pigment which
is readily soluble in water or dilute alkali. Ifa considerable
amount of caustic soda or potash be added, however, the
solution turns dull yellowish brown, and when shaken with
ether, the ether removes a yellow pigment which gives a faint
band at \ 661, the band of chetopterin. This pigment seems
to me to be a product of decomposition, showing that
chetopterin is destroyed by strong alkali; a mere trace of
unaltered chetopterin would be sufficient to yield the band.

It thus appears that dilute alkalies havea double effect upon
chetopterin. In the first place, the pigment forms with
alkalies a compound of green colour which is insoluble in
water, alcohol, and ether, and from which original chzetopterin
can be recovered by the action of acid. In the second place,
alkalies give rise to an alkaline derivative of green colour,
which is soluble in water as well as in dilute alkali and alcohol.
Solutions of this in alcohol are not fluorescent, and show two
bands, one at A 650 and the other at A 625.

(4) Action of Salts.

The addition of salts like lead acetate or copper acetate or
sulphate to alcoholic solutions of cheetopterin causes a precipi-
tation of pigment. In the case of lead acetate the precipi-
tated pigment is bright yellow-green in colour, and is insoluble

408 MARION I. NEWBIGIN.

in ether or alcohol. If the precipitate is treated with a little
dilute acid the colour changes, and, after washing with water,
the pigment can be dissolved in ether or alcohol. The alco-
holic solution is blue-green in colour, but the blue is not
intensified by acid, and spectroscopic examination shows that
the solution consists of a mixture of unaltered chetopterin
and the green acid derivative already described. The colour
and fluorescence vary according to the amount of the deriva-
tive present, and this depends upon the amount of acid
employed to decompose the compound. If a considerable
amount of the derivative is present the fluorescence becomes
indistinct, and an appreciable amount of a brown insoluble
residue remains behind when the acidified precipitate is treated
with methylated spirit. These results show that precipitation
with lead acetate cannot be readily employed as a means of
purifying chetopterin, for when the insoluble lead compound
is treated with acid a portion of cheetopterin as it is set free is
converted into the green acid derivative (cf. the effect of acid
on the ammonia compound as noted above, p. 406) ; other salts
give similar results.

The solution obtained in methylated spirit from the acidi-
fied precipitate after lead acetate is of a singularly pure and
beautiful blue-green colour, and rarely gives more than one
definite band, though the band in the yellow may be repre-
sented by a shading. The band at C has its apparent centre
about A 654 (ef. the first band of fig. 2), but it is of course
really the result of the apposition of the band of chetopterin
and that of the green derivative. Before the relation of che-
topterin to acids was fully understood this solution was
thought to contain a single pigment, and a considerable quan-
tity of the dried pigment was tested for nitrogen by igniting
with metallic sodium. The test showed the presence of
nitrogen, but, as the pigment was a mixture of the acid deri-
vative and original chetopterin, the result is only to show that
chetopterin itself contains nitrogen. In this respect it resem-
bles bonellin. It would be of interest to know whether, when
cheetopterin splits into the green acid derivative and the brown

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 409

insoluble residue, the nitrogen does or does not enter into the
composition of the green pigment, or whether it is all con-
tained in the brown substance.

Summary.

The pigment chetopterin dissolves readily in cold methy-
lated spirit or in ether to form solutions, which are indefinite
in colour, strongly fluorescent, and which give a spectrum
consisting of four distinct bands and a shading.

When acid is added to the solution in methylated spirit, it
becomes first blue and then green.

The blue solution is fluorescent, and distinguished by certain
peculiarities of the spectrum, but the fact that it can be made
to yield only original chetopterin and a single-banded green
derivative seems to prove that its peculiarities are not in
themselves diagnostic of the existence of an acid derivative
comparable to Krukenberg’s bonellidin, but are due to the
combination of the two pigments present.

The green acid solution shows little fluorescence, and yields
similarly a mixture of chetopterin and the one-banded acid
derivative.

This acid derivative is characterised by its single faint band,
its green colour, and the absence of fluorescence. It is appa-
rently formed by the splitting of chztopterin into an insoluble
brown residue and this derivative.

Dilute alkalies have a twofold action upon chetopterin.
They in part precipitate it as a compound insoluble in alcohol,
ether, and water, and in part convert it into a green derivative.
This derivative is characterised by its colour, its spectrum
consisting of two bands in the red, and its solubility in water.

Salts of the metals, such as lead acetate, precipitate cheeto-
pterin from its solutions, forming compounds which are in-
soluble in water, alcohol, or ether. By the action of dilute
acid on these compounds chetopterin can be regenerated, but
it is liable to be intermixed with the acid derivative.

Chetopterin, like bonellin, contains nitrogen.

The points which seem to me of special importance are that

vol. 41, PART 3,—NEW SER. FF

410 MARION I. NEWBIGIN.

while chzetopterin itself is indefinite in colour and strongly
fluorescent, and exhibits a complex spectrum, the action of
reagents is to tend to produce pigments of bright definite tint
and simple spectrum, which may be soluble in water and are
without fluorescence. Certain points of resemblance to
bonellin are also of much interest.

B. ENTEROCHLOROPHYLL.
(1) Previous Investigations.

The name enterochlorophyll was given by Dr. MacMunn (6)
to a pigment, found in the digestive glands of Mollusca and
other Invertebrates, which turns green on the addition of
acid, and then gives a spectrum resembling that of acid chloro-
phyll. The same pigment is apparently denoted by Kruken-
berg’s term hepatochrome, but Krukenberg did not isolate the
pigment or define it clearly.

Dr. MacMunn’s observations may be briefly summarised as
follows :—He found that the epithelium lining the “liver ”
tubules in Mollusca contains pigmented oil drops and granules
which dissolve in alcohol to form a greenish-yellow solution
with strong red fluorescence. The solution gives a spectrum
with three bands, one in the red, one to the left of D, and one
to the left of E, and also strong absorption of the violet end.
In some cases in dilute solution there may be one or two very
faint bands in the violet, but these are always ill-defined as
compared with the dominant three. The addition of strong
acid turns the solution grass-green, shifts the bands slightly to
the right, adds an additional band at the right of the D line,
and diminishes the absorption at the violet end, so that one
clear band to the left of F is now visible there. It is this five-
banded spectrum which Dr. MacMunn compares to the spectrum
of chlorophyll. In his first paper (1883) he compares it to the
spectrum given by an acidified solution of chlorophyll,—that is,
to the spectrum given by a mixture of chlorophyll and phyllo-
cyanin; but in the second communication (7) he dwells upon
its resemblance to the spectrum of pure chlorophyll. There is

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. All

thus a slight ambiguity in the use of the term enterochlorophyll,
for it is not quite apparent whether it is to be used to designate
the pigment contained in the yellowish extract of molluscan
liver, which gives a three-banded spectrum, or to that in the
green solution produced by the addition of acid to this solution,
which has a five-banded spectrum.

In his second paper Dr. MacMunn applied the method of
saponification to the pigment, and showed that solutions of his
enterochlorophyll, like solutions of plant chlorophyll, contain a
yellow lipochrome pigment in addition to a greenish constituent.
In the case of plant chlorophyll it is now known that the associa-
tion of “chlorophyll green” and the lipochrome, xanthophyle,
is merely incidental, and the term chlorophyll is restricted to
the former. It is not quite clear whether Dr. MacMunn regards
“enterochlorophyll” as a combination of a lipochrome and a
green constituent, or whether he regards them as associated
pigments ; but he was not able to obtain complete separation,
and believes that “in enterochlorophyll there is probably a
more intimate union between the constituents than in plant
chlorophyll.”

Krukenberg’s (2) observations are much less detailed. He
found in the “ bile” of Mollusca what he regards as evidences
of three pigments. One of these, he says, is a lipochrome, and
gives two bands in the violet ; another gives a strong band in the
red and one in the green, and is a “ hepatochrome:” a band at
the beginning of the green appeared in some of his solutions
and puzzled him greatly, but he believed that it belonged to a
third unknown pigment. He evidently worked with very small
amounts of pigment, and did not go beyond this point. In-
complete as the observations are, however, Krukenberg was
right in every one of his inferences. The bands in the red and
the green are two of the bands of “ enterochlorophyll ;” the
occasional band in the beginning of the green is that band of
enterochlorophyll which is only distinct in acidified solutions ;
the violet bands are due to the presence of an additional yellow
pigment.

412 MARION I, NEWBIGIN.

(2) Mode of Occurrence.

Dr. MacMunn describes enterochlorophyll in various Kehi-
noderms as well as in the Mollusca; my own observations
were made entirely on the Mollusca. As the pigment occurs
in relatively small amount, it is necessary to choose a form
which can be obtained in large quantity. The garden snail
(Helix aspersa) and the common limpet (Patella vulgata)
both fulfil this condition, but I soon found that the snail
contains a very much smaller amount of pigment than the
limpet, and in consequence, in spite of the greater difficulty
in dissection, the latter was employed in all the later observa-
tions.

In the limpet the pigment has been described by Dr. Mac-
Munn in the “liver” and in its secretion; I find it also in the
cells of the gut, in the contents of the gut, and in very pure
condition in the feces. In all these situations it can be re-
cognised by the microchemical reaction already described for
cheetopterin (p. 393)—the vivid green colour with hydrochloric
acid.

In order to study the distribution of the pigment in the
“liver” and intestine, portions of the visceral hump were
hardened in formalin, and sections cut through both the
digestive gland and the coils of the intestine. The sections of
the gut show with low power a band of brownish-green pig-
ment placed in the epithelial cells near their inner margin
(see fig. 9). When examined under a higher power (fig. 10)
the pigment is seen to occur in minute closely packed granules,
brownish green in mass, green when viewed singly. They in
all respects resemble the granules in the cells of the gut in
Chetopterus, but the amount of pigment is much smaller.
The sections of the tubules of the digestive gland (fig. 11) do
not show cells having this peculiar granular appearance. The
large cells near their inner surface contain several of the
characteristic molluscan pigmented vesicles, usually of a
brownish-yellow colour, while scattered through the proto-

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 413

plasm, as described by Dr. MacMunn, coloured oil-drops
occur. Sections of the digestive gland are by no means easy
to cut when, as in the present instance, the ordinary harden-
ing agents are inadmissible. Many of the cells are ruptured
in the process, and mingled with the débris there occur green
oil-drops like those found mixed with the contents of the gut
in Chetopterus. I am of opinion that these contain the
pigment enterochlorophyll, while the brownish vesicles pro-
bably contain it intermixed with yellow pigment, which may
also occur diffused through the protoplasm. The colour of the
liver varies greatly in different specimens, the differences being
apparently due to variations in the amount of enterochloro-
phyll present. The presence of the pigment in the cells of the
gut and in the feeces seems to me of prime importance from a
comparative point of view.

(83) Characters of Solutions.

In the earlier experiments great care was taken to diminish
the risk of contamination of the solutions by plant chlorophyll
from the gut. In the limpet portions of the liver were care-
fully dissected away from the coils of the gut and dropped into
methylated spirit. As it takes some thirty or forty limpets to
yield a very moderate amount of solution, the process is a
somewhat tedious one. It was found later that the risk of an
intermixture with plant chlorophyll is in reality very small if
care be taken to remove the “manyplies” and its contents,
for the digestive juices seem to very speedily destroy the
chlorophyll, and it rarely occurs in the small intestine.
After the discovery of the presence of the pigment in the
feeces was made these were employed as sources of the pig-
ment. The absence of lipochrome pigment and of fat in
solutions obtained from them greatly simplifies further opera-
tions.

To obtain the pigment the parts employed, whether liver or
feeces, should be dried and powdered, and the powder extracted
with cold methylated spirit, in which the pigment is exceed-

414, MARION I. NEWBIGIN.

ingly soluble. Solutions obtained from the digestive gland
are yellowish in colour, from the feces greenish brown ; both
have strong red fluorescence. The spectroscopic characters of
the former solution have been already described by Dr. Mac-
Munn; the latter differs chiefly in showing less absorption of
the violet end, and a more or less distinct band in the neigh-
bourhood of the F line: the differences are associated with the
absence or diminished amount of lipochrome pigment in the
solution from the feces. The “liver” extract, as shown by Dr.
MacMunn, turns green on the addition of acid; the extract of
the feces, on the other hand, turns first bluish, and then green
on further addition of acid.

Relation to Chlorophyll.—The resemblances to cheeto-
pterin which have been incidentally noted in the above de-
scription tend to disprove the identity of ‘ enterochlorophyll,”
and plant chlorophyll, but it may be well before proceeding
further to state more clearly the difficulties which the sugges-
tion has to encounter. ‘ Enterochlorophyll,” in the first place,
differs from true chlorophyll in giving the peculiar green re-
action already described. Further, the solutions are much
more stable than those of chlorophyll. As is well known, the
decomposition of a solution of chlorophyll in bright light is a
matter of minutes, while even in obscurity the fading is rapid.
Detailed observations on the effect of sunlight on entero-
chlorophyll were not made, but solutions did not show marked
change when left to stand in the diffused light of the laboratory,
and remained unaltered during months of standing in a cup-
board, while an extract of green leaves standing in the same
cupboard lost its green colour entirely. More striking is the fact
that, although the colour and spectrum of “ enterochlorophyll ”
solutions change on the addition of acid, the original spectrum
can be restored by alkali, and the process repeated any number
of times; this is impossible in the case of chlorophyll. The
addition of aqueous or gaseous hydrochloric acid does not
produce a precipitate, or only to a very slight extent, and if
precipitated the pigment shows the character of the original
enterochlorophyll. Gaseous hydrochloric acid when introduced

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 415

into solutions of chlorophyll produces a precipitate of phyllo-
eyanin which differs markedly from chlorophyll, and cannot be
reconverted into it. It is probably unnecessary to pursue these
contrasts further; a comparison of the properties of “ entero-
chlorophyll” as detailed in the present paper with Schunck’s
results (10) in the case of chlorophyll and phyllocyanin will
show that enterochlorophyll is a much less complex pigment.
The apparent points of resemblance are the fluorescence, the
association with a yellow lipochrome, and the spectrum. Of
the fluorescence, a not uncommon character amongst certain
classes of pigments, it is not necessary to say anything. The
analogy of cheetopterin, and the conditions seen in the feeces of
Patella, show that the association of the lipochrome is an
accidental character of no significance. With regard to the
third point, the spectrum, there is more difficulty. In his first
paper Dr. MacMunn compared the spectra of acidified solutions
of chlorophyll and acidified alcoholic extracts of the “ liver ” of
Ostrea, and found them almost identical. This is at first
sight a very striking result, but the reflection that the first
solution contained a mixture of chlorophyll and phyllocyanin,
and the second a mixture of “ enterochlorophyll”’ and an acid
derivative, somewhat diminishes the force of the comparison. In
his second paper Dr. MacMunn compares the spectra of un-
altered chlorophyll and “enterochlorophyll” directly. The most
striking difference is, then, the absence in the latter of a band
to the right of D. Such a band appears in enterochlorophyll
solutions on the addition of acid, which Dr. MacMunn regards
as evidence of the existence of “ enterochlorophyll” in the
‘reduced condition, or in the form of a chromogen.’ He
found this band in the normal extract in some cases, cf. the
observations of Krukenberg as quoted above (p. 411). The
occasional appearance of this band in both cases I am inclined
to regard as due to the presence in the solution of traces of
acid, probably derived from the gut or its contents, or to the
solution containing an unusually large amount of pigment
(cf. cheetopterin, where the band is present in very strong solu-
tions, p. 394). The resemblance between the spectrum of

416 MARION It NEWBIGIN.

“ enterochlorophyll ” when this band is present, and the spec-
trum of true chlorophyll, I believe to be merely a striking co-
incidence, emphasising the danger of relying upon spectroscopic
observations unsupported by chemical investigation, rather
than indicating true affinity. In view of these facts Prof.
Lankester suggests the emendation “‘ enterochlor,” or “ ente-
roverdin,” in place of the term ‘ enterochlorophyll,” but it is
to be noticed that neither in the natural condition nor in
solutions is the greenness of the pigment at all well marked,
except after the addition of acid.

Characters of the Associated Lipochrome.—Al-
though the yellow pigment found in company with entero-
chlorophyll was not subjected to a detailed examination, it may
be useful to point out in what respects its presence modifies the
reactions of enterochlorophyll. In the first place it modifies
the spectroscopic characters, in that it produces marked
absorption of the violet end, and so blurs the fourth band of
enterochlorophyll, that is the band in the neighbourhood of
the F line. Like other yellow lipochrome pigments, it is de-
colourised by the addition of hydrochloric or other acids to
the alcoholic solution. The result is that the addition of acid
makes the fourth band of enterochlorophyll distinct by dimin-
ishing the absorption of the violet. The yellow pigment also
exercises a marked and often very puzzling effect on the colour
of the solutions. When it is virtually absent solutions of
enterochlorophyll, like those of chzetopterin, turn blue on the
addition of acid, though the tendency for the blue to pass into
green is much stronger in the former than in the latter. When
the yellow pigment is present acid turns the solution green
without trace of blue. The reason for this is that though the
yellow pigment is destroyed by acid, yet the amount of acid
necessary to completely remove it is also sufficient to turn the
enterochlorophyll solution green instead of blue (cf. cheto-
pterin, p. 396). It is exceedingly difficult to remove the yellow
pigment from solutions of enterochlorophyll, for although
acid decolourises it, the colour returns on the addition of
alkali, so that the pigment is liable to appear at very unexpected

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 417

points. Dr. MacMunn has shown that saponification with caustic
soda or potash is ineffective, precipitation with acetate of lead is
better, for the yellow pigment remains in the solution. ‘Traces
of it are, however, always liable to be carried down with the
enterochlorophyll, and the difficulty of re-obtaining unaltered
enterochlorophyll from the precipitate greatly diminishes the
utility of the method. The most useful method is, perhaps,
to add excess of acid, and shake with ether several times, then,
rejecting the ether, add alkali to the acid solution in too small
amount to neutralise the solution, and shake again with ether.
The ether takes up almost pure enterochlorophyll, which after
careful washing may be used for further experiments. The
method is, of course, wasteful, and not entirely effective, and
it is better, when possible, to obtain a solution from the feeces
where the yellow pigment is virtually absent. The purification
of enterochlorophyll from fats, and the other impurities with
which it is associated, is a matter of great difficulty.

(4) Action of Acids.

Enterochlorophyll resemble chetopterin so closely that it is
not necessary to do more than note its characters, referring to
the description of chzetopterin for details.

The bands shown by a solution of enterochlorophyl! differ
slightly from those of chetopterin, but the difference is not
marked. A neutral solution shows the following bands:

I A 667, IT A 604, IIIT A 539, IV A 508 (fig. 6).

If acid is added to a solution containing little lipochrome a
bluish colour develops, and the solution becomes five-banded
with the bands as follows:

ITA 657, ILA 599, IIT X 567, 1V A 534, VA 500.

This is the “enterochlorophyll” spectrum so frequently
figured by Dr. MacMunn. The fifth band, as in the case of
cheetopterin, tends to be indistinct. On further addition of
acid the solution turns green, the right-hand bands tend to
disappear, and the band in the red shifts back to its original
position at about A 667, just as occurs in the case of cheto-

418 MARION I. NEWBIGIN.

pterin. The association between colour and spectrum is not,
however, so close as in the case of cheetopterin, for the solution
even when only slightly acidified shows a strong tendency to
become green.

When the acidified solution is diluted with water and shaken
with ether, the ether extracts a considerable amount of
pigment, more than is the case with chztopterin, entero-
chlorophyll being apparently somewhat less soluble in dilute
acid than is chetopterin.

The ether is pale green in colour; it usually displays four
bands, but the first band tends to be about A 657 instead of
\ 667, as in the original solution. That is, the spectrum is
that of fig. 6, but the first band tends to be replaced by the
first band of fig. 7. When the ether is evaporated and the
residue dissolved in methylated spirit, a green solution is
formed, which turns brown and not green with acid. If
the ether before evaporation be placed in a separation funnel
and concentrated hydrochloric acid poured in, the acid becomes
bright deep green, and the ether remains yellowish green.
When examined with the spectroscope it then shows two bands
in the red, one at \ 667, and one at about A 650 (fig. 8), which
in a very thick layer overlap and produce the appearance of a
broad band at X 657. j

It is thus obvious that acid has the same effect on solutions
of enterochlorophyll as it has on those of chetopterin. That
is, it produces small amounts of a one-banded acid derivative
which is not fluorescent, and is readily soluble in ether and
alcohol.

It may perhaps be well to notice, in regard to the action of
acid on these pigments, that the statement that they differ
from chlorophyll in that they can be converted into the acid
form, and then reconverted to the normal by means of alkali any
number of times, is not strictly speaking accurate, for acid acts
on the pigment very slowly, and the “‘ reconversion ” is in the
general case merely due to the removal of acid from the
solution; in the case of the Patella pigment, indeed, the
normal pigment often precipitates from a dilute acid solution

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 419

on standing, which can hardly be described as a “‘ reconversion.”
When the pigment is actually converted into the one-banded
acid derivative it cannot be reconverted into the normal
pigment. The difference between chlorophyll and the pig-
ments enterochlorophyll and chetopterin as to the action of
acids, is more accurately expressed by saying that the latter
are relatively insensitive to the action of acids, while the
former is exceedingly sensitive.

Enterochlorophyll can be precipitated from acid solution by
the addition of marble, just as cheetopterin can. Solutions of
the former, however, when obtained from the liver and
intestine always contain fat and other impurities, which may
more or less disguise the reactions.

(5) Other Reactions,

In its other reactions also “ enterochlorophyll” shows a
remarkable resemblance to chetopterin. The action of
ammonia is to turn the pigment green, although the colour
may be concealed by the presence of the yellow pigment; to
alter its spectrum; and to render it soluble in water. As to
the spectrum of this ammonia derivative, I have only been
able to find one band at about 2 655, instead of the two of the
corresponding chetopterin derivative ; but as the second band
is often difficult to demonstrate in the case of chetopterin, I
am not disposed to lay much stress upon this fact. In regard
to it and to some other trifling differences from chetopterin,
it seems only necessary to point out that not only does
enterochlorophyll occur in smaller amount than chetopterin,
but also the solutions contain a large admixture of foreign
substances, which are of course increased in relative amount
with every increase in the strength of the solution. These
intermixed substances greatly increase the difficulty of making
observations, especially in regard to solubilities and so forth.

Enterochlorophyll is to some extent precipitated by ammonia
just as cheetopterin is. The addition of acetate of copper or
lead to a solution of enterochlorophyll causes a precipitation

420 MARION I. NEWBIGIN.

as in the case of chetopterin, the precipitate being a bright
green colour. The precipitate is insoluble in alcohol, and
after treatment with dilute acid dissolves in alcohol to form a
solution which contains a mixture of enterochlorophyll and
its one-banded acid derivative. Lead acetate affords a useful
test for the presence of enterochlorophyll in a solution which
contains so much lipochrome as to disguise the ordinary reac-
tions. Such a solution may be pure yellow, but with lead
acetate a green precipitate at once forms, the colour being in
striking contrast to that of the solution.

(6) Relation to Chetopterin.

Enterochlorophyll is so closely related to chetopterin that
the question at once arises whether or not it is identical with
that pigment. While leaving that question open, I may point
out one or two differences between the two.

First, as to the spectrum, the bands in enterochlorophyll are
slightly nearer the red end than those of chetopterin, and the
difference appears also in the derivatives. According to Dr.
MacMunn, however, there is some variation in the spectrum
of the pigment in different animals.

Second, as to colour, solutions of enterochlorophyll even
when apparently free from lipochrome pigment, do not give so
marked a blue on the addition of a little acid as do those of
cheetopterin, the colour always inclining towards green. Solu-
tions of enterochlorophyll in concentrated acid are of a bright
green colour, but the green inclines towards yellow, while
that of chzetopterin solutions inclines towards blue. The deri-
vatives show analogous differences.

Finally, as to the solubility, enterochlorophyll seems to be
distinctly less soluble in dilute acid than chetopterin. I am
inclined to suspect, however, that this is in part due to the
large amount of fat in most solutions of enterochlorophyll.
The differences are thus not very well marked. In relation to
acids, and in giving rise to a one-banded acid derivative ; in
the action of alkalies, and the production of a soluble alkaline
derivative ; in the action of salts; and in the general spectro-

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 421

scopic properties, the pigments show much resemblance to one
another.

(7) Distribution of Enterochlorophyll.

I have not yet made observations on this subject, but it
seems desirable to briefly consider the literature. Dr. MacMunn
describes the pigment in the Mollusca and Echinoderma as
well as in some other cases. In the two groups mentioned it
seems exceedingly common, and usually present in considerable
amount. Its presence in the latter is especially interesting
because of the existence in the group of Moseley’s (9) penta-
crinin and antedonin. I am strongly of opinion that these
two pigments are in some way related to enterochlorophyll,
but Iam not able to explain why they should present such
striking differences in spectra and so forth. In Echinoderms
“enterochlorophyll” is said to occur chiefly in the so-called
digestive gland, but in sea-urchins it is said also to be found in
the perivisceral fluid, although I have not succeeded in obtaining
the full spectrum there.

As to the presence of enterochlorophyll in the great group
of Arthropods there is much less certainty. Dr. MacMunn
speaks of it as existing occasionally in the digestive gland in
Crustacea, but he does not seem to have obtained the full
spectrum, or to have isolated the pigment. I have examined
the pigment of the digestive glands of a few Crustacea, and
have not succeeded in obtaining the reactions of enterochloro-
phyll. If present it can only be in small amount, and the
most abundant pigment of the “liver” is certainly not
*©enterochlorophyll”’ in the sense in which that name has
been used here. The question whether the pigment does exist
in Arthropods in considerable amount is one worthy of further
investigation.

In the Celenterata there seems little reason to doubt that a
pigment related to enterochlorophyll is widely distributed.
Such a pigment probably exists, for example, in Anthea
cereus, though it appears difficult to isolate, and presents
certain peculiarities (see Krukenberg [4] and MacMunn [8)).

422 MARION I. NEWBIGIN.

Moseley’s polyperythrin (9), which is widely spread in certain
corals and sea-anemones, is perhaps an allied pigment.

There is thus much reason to believe that pigments related
to cheetopterin and “ enterochlorophyll” are widely spread in
Invertebrates.

c. BoNnELLIN.
(1) Comparison with Chetopterin.

The amount of bonellin at my disposal was so limited that I
have not been able to make a full investigation of it. There
are, however, certain points which seem to suggest a close
relation to chetopterin, and are worthy of note. First, as to
the spectra, I quote from Engelmann the following points of
maximum absorption in neutral solutions of bonellin and
chetopterin. It will be noted that, for the reasons already
stated, the points in the case of chetopterin do not exactly
correspond to the apparent centres of the bands as determined
by an ordinary spectroscope.

Bonellin . 12685, II 1585, III 1520, IV A490.
Chetopterin I 4655, II 4600, III 1535, IV A500.

When compared together the two sets show some curious
analogies. Thus both pigments have four bands placed in
similar parts of the spectrum, and, curiously enough, the dis-
tances between the bands are almost identical in the two cases,
as a little calculation will show.

In addition to its four bands, chetopterin in strong solution
shows also a shading, not yet described for neutral bonellin. -

Then, as to the spectra of the acid solutions, I quote again
from Engelmann :

Bonellin .. 1 X613, IL \ 570, 11) 145. 1V) bis.
Chetopterin I ’ 650, II \597, III 1560, IV 1533, V A500?

On comparing these spectra, no such relation as that noted
for the neutral solutions is observed, but there are several
points which require notice. In the first place, it is obvious
that the first band of acidified bonellin does not correspond to
the apparent first band of acidified chetopterin. We have

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 425

already seen that the apparent single band of the latter is in
reality formed by the apposition of two bands, of which the
second has its centre about 1641. It is, as I think, this band
which corresponds to the first band of acid bonellin. In other
respects there is, as already seen, considerable analogy between
the two sets of spectra. Thus the band at 545 in acidified
bonellin obviously corresponds to the band at 560 in acidified
chetopterin ; the disappearance of the violet band of neutral
bonellin is paralleled by the dimness of the corresponding band
in the case of chetopterin. On the fact that the yellow and
green bands of chetopterin change little on the addition of
acid, while those of bonellin show considerable movement, I
am not inclined to lay any stress. My own observations
showed considerable discrepancy from those of Engelmann on
this subject, and I am strongly of opinion that the movement
varies with the amount of acid in the solution.

The most apparent difference from chetopterin which
bonellin shows is in the characters of its alkaline solution.
Professor Lankester has made the exceedingly interesting dis-
covery that normal bonellin is alkaline,—that is, that the
pigment apparently occurs in the animal in the alkaline
condition. This alkaline solution gives no less than six bands,
of which four are those of neutral bonellin, while the other
two have their maximum points of absorption at A614 and
A551 respectively. The solution is bright pure green, with
strong fluorescence. Now we have already seen that when
cheetopterin stands for some time with alkali it becomes pure
rich green with undiminished fluorescence, and then shows a
five-banded spectrum (fig. 4) with two bands in the red, the
original one and a new one at A625; at the same time the
_ bands in the yellow and green show a marked diminution.
When one passes from the study of normal cheetopterin to
that of normal (alkaline) bonellin, the most striking differences
are, in the latter case, the definiteness of the colour and the
indistinctness of the bands, except that in the red and to a
less degree that in the violet. In working at the action of
alkalies on chetopterin these same characteristics, definiteness

A424, MARION I. NEWBIGIN.

of colour and indefiniteness of bands, reappear; when it is
known that bonellin normally occurs in the alkaline state it is
difficult not to regard this fact as an additional proof of affinity.
As to the additional band at 1550 in bonellin, it is to be re-
collected that it is very faint, so faint that it was missed by
Sorby (11) entirely; I believe that it is the same band as that
described at A 545 in acidified solutions. It will be recollected
that in solutions of cheetopterin a similar band occurs, which
is only distinct in acidified solutions, but is represented in
normal ones by a faint shadow.

It may be well to summarise briefly these facts.

1. Neutral solutions of chetopterin and of bonellin re-
semble one another in their indefiniteness of tint, in their
strong fluorescence, and in possessing a spectrum of four bands
occupying similar but not identical positions. Chetopterin
solutions show, in addition, a faint shadow in the green, not
yet described in neutral bonellin solutions. A similar band,
however, occurs in normal (alkaline) solutions of bonellin.

2. The addition of a little acid turns chetopterin solu-
tions blue, bonellin solutions violet, without diminution of the
fluorescence. The spectrum is considerably altered, but the
alteration in the two cases is similar in so far as in each a new
band appears or becomes distinct, and the fourth band of the
original spectrum tends to grow faint or disappear. The most
striking difference is seen in the fact that in chetopterin the
position of the red band only alters slightly, in bonellin it
alters much; but the difficulty is diminished by the fact that
the acid cheetopterin solution can be proved to possess a band
analogous to that of acid bonellin, which is concealed by the
presence of the original band in the red. Further addition
of acid in both cases produces a further change of colour, and
the reappearance of the original band in the red. |

8. Normal (alkaline) bonellin and chetopterin which has
stood with alkali resemble one another in their deep green
colour and fluorescence, and in showing the four-banded
spectrum of the neutral form, plus an additional faint band in
the red. They differ from one another in that bonellin shows

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 425

in addition a faint band in the green, probably corresponding
to the band in this position in acidified solutions. But a
similar band is represented in the normal solution of cheto-
pterin, and apparently is absent in the alkaline solution only
because this is moredilute. The balance of evidence, therefore,
seems to me to be in favour of an affinity between the two
pigments.

In their distribution, however, the two pigments seem to
differ markedly. Chzetopterin occurs in the Annelid Cheto-
pterus. ‘ Enterochlorophyll,” which is undoubtedly at least
nearly related, occurs in Mollusca and Echinoderma—not to
mention other doubtful cases. In these three sets of animals,
certainly not nearly related, the pigment when carefully
studied has been found to occur in endodermal tissues, in
connection with the alimentary tract or its outgrowths.
There is, I believe, much evidence that bonellin is related to
cheetopterin and enterochlorophyll, and yet we find it described
as occurring in the epidermis and in “ subepidermic cells
apparently belonging to the connective tissue,’—that is, in
ectodermal and mesodermal tissues. The latter position is
that also described for Professor Herdman’s (1) thalassemin—
a pigment differing markedly from bonellin. In view of these
facts the possibility suggests itself that in Bonellia a pigment
similar to cheetopterin occurs in the cells lining the gut; that,
instead of being eliminated intact with the feces, as is entero-
chlorophyll in Patella, the pigment undergoes modification,
forming a green derivative which is deposited in granules in
the epidermis and underlying tissues. When Bonellia is
placed in alcohol both the original pigment and its derivative
may dissolve out, resulting in the formation of the solution
called alkaline bonellin. The suggestion seems to me to be
supported by a comparison with Professor Herdman’s thalas-
semin.

(2) Relation to Thalassemin.
The fact that although a green colour is common in the
Echiuride, the peculiar pigment bonellin has only been
voL. 41, PARY 3.—NEW SERIES. GG

426 MARION I. NEWBIGIN.

described in Boiellia, makes the characters of the green pig-
ment of Thalassema a question of some interest. I do not
propose to repeat here my notes on the subject (see Professor
Herdman’s paper), but wish merely to point to the special
differences from bonellin. The pigment is greenish blue in
tint, the blue being accentuated by acid—a point I did not at
first notice, but which is of some interest. It is readily
soluble in formalin or in water, and the residue after the eva-
poration of the formalin is soluble in alcohol, although Pro-
fessor Herdman did not find that the worm itself yielded a
coloured solution when placed in alcohol ; this may, however,
have been due to the small amount of pigment present. None
of the solutions are fluorescent, and the spectrum shows a
single band at about A 617 (cf. the band at A 614 in alkaline
bonellin). If we compare these characters with those given
previously, as tending to characterise the derivatives of cheto-
pterin, we find that the solubility in water, the loss of fluor-
escence, the definiteness of colour, the simple spectrum, all
reappear. It seems to me not improbable that thalassemin,
which is apparently common among the Echiuride, is a deri-
vative of a pigment allied to chetopterin, and that it quite
possibly occurs in Bonellia itself, in addition to a cheto-
pterin-like pigment.!

(3) Other Green Pigments.

In view of the simultaneous occurrence in the Echiuride of
an apparently complex pigment like bonellin and a simple one
like thalassemin, it is interesting to inquire whether similar
simple pigments do not occur in association with enterochlo-
rophyll and cheetopterin. There is one marked case of this
kind which does not appear to have been yet described.

The tortoiseshell limpet, Acmza testudinalis, so common

1 In connection with thalassemin it may be well to mention that I find that
the green reaction with nitric acid described in my previous notes is an error.
The reaction is the result of the action of impure nitric acid on alcohol, and
has no connection with the pigment. (See Dastre et Floresco, ‘C. R. Soc. d.
Biol.,’ x, 1898.)

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 427

on Scottish coasts, is remarkable in having the epithelium
which covers the visceral hump coloured a bright vivid green,
which passes into brown at the margin of the mantle skirt.
The epithelial layer which is turned in to cover the dorsal
surface of foot, is also coloured by the same pigment. The
pigment occurs as minute granules in the epithelial cells, and
in life varies from blue to green in colour. Sections of the
mantle skirt show that it is the superficial cells only which are
pigmented. At its margin the green pigment passes gradually
into a brown one of similar distribution. The pigment of the
shell, it will be remembered, is deep brown. The green pig-
ment readily dissolves in sea water, in distilled water, or in
formalin. It is also soluble in alcohol, but is then apt to be
mingled with enterochlorophyll derived from the liver and gut.
When a number of limpets are preserved in formalin, the
formalin becomes clear greenish blue. The solution re-
sembles that of thalassemin in its colour, in turning distinctly
blue on the addition of acid, in being, at least in part, de-
colourised by alkalies, which tend to turn the pigment yellow.
The pigment also, of course, resembles thalassemin in its so-
lubility in water; it differs from it in that I have not succeeded
in obtaining any band in its spectrum, and in that it is very un-
stable. In view of the effects of reagents on enterochlorophyll,
it does not seem to me improbable that this pigment is a
derivative of enterochlorophyll. There can be little doubt
that the brown pigment of the margin of the mantle is derived
from the green, and that it is identical with the pigment of the
shell; this suggests the possibility that the enterochlorophyll
of the gut and liver, instead of being entirely eliminated with
the feeces, may give rise in the Mollusca to the bright pigments
of shell and mantle. There seems little doubt that soluble
green pigmeuts, like those of Acmza and Thalassema, are
widely distributed in Invertebrates; whether they usually
originate in the way suggested here must, of course, remain at
present undetermined. Such a green pigment occurs, as I
believe, in many Annelids, notably in Eulalia viridis,
especially in the eggs, but is there not easy to extract.

428 MARION I. NEWBIGIN.

Whether the suggestion as to the origin here made is correct
or not, it is at least interesting to note that in many groups of
Invertebrates, either in the same animal or in related forms,
there may occur two different sets of green pigments, distinctly
marked off from one another, but connected by the derivatives
of the more complex series. Such are bonellin and thalassemin
in the Echiuride, chetopterin and the pigment of Eulalia
in the Chetopoda, enterochlorophyll and the pigment of
Acméea in the Mollusca; it is probable that there are many
other similar cases.

CONCLUSION.

If the observation and deductions set forth in this paper are
correct they go to prove that there exists in Invertebrates a
widely spread group of pigments occurring primarily in con-
nection with the alimentary tract or its outgrowths, and
characterised by forming in alcohol fluorescent solutions
of indefinite colour which exhibit a complex spectrum, con-
sisting when fully developed of five bands. In the fluores-
cence and in the complex spectrum these pigments resemble
chlorophyll, but the other characters, and especially the
relation to acids and alkalies, show that this resemblance
is entirely superficial. Of such pigments, chetopterin, and
the pigment or pigments described as “ enterochlorophyll,” are
typical examples. Whatever the primary function of the pig-
ments—and of this I have nothing to say—they at least so far
resemble the bile pigments of Vertebrates that they occur
mingled with the contents of the gut, and at least in some
cases are eliminated with the feces. It seems desirable to
have a general name to designate these pigments ; and in view
of our present ignorance of function, the position in which
they occur, and the fact that they give rise to highly coloured
derivatives, seem to be the only available characters from which
a name can be based. Krukenberg’s term hepatochrome is in
many respects very suitable, but it has the great disadvantage
that he himself used it for “liver” pigments giving banded
spectra, without, so far as I can find, defining it clearly.

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 429

Perhaps the term “enterochrome” might be found more
suitable. I should suggest it as a general term for a series of
pigments occurring in connection with the alimentary tract
and its outgrowths in Invertebrates, and characterised by
solubility in cold alcohol to form strongly fluorescent solutions,
usually of greenish tint, which give in neutral solution a four-
banded spectrum with a trace of a fifth band which is
rendered more distinct by the addition of acid. Acid also
changes the colour of the solution, and produces other changes
in the position of the bands. As further characters are to be
noted the power of yielding derivatives which may be soluble
in water, are of bright colour, exhibit simple spectra consisting
of one or two bands, and are without fluorescence in alcoholic
solution. It is probable that it will be found that such deriva-
tives are numerous.

Further, while the enterochromes may be eliminated intact
from the body, as in Patella, the fact that forms containing
them not infrequently also exhibit bright green pigments,
soluble in alcohol without fluorescence, and often soluble in
water in addition, suggests the possibility that the entero-
chromes may in some cases, instead of being eliminated, give
rise to brightly coloured derivatives, capable of being employed
in surface coloration.

The peculiar pigment bonellin resembles the enterochromes
in most of its characters, but is entirely absent from the ali-
mentary canal of the form in which it occurs. The same is
true of the pigments Pentacrinin and Antedonin, and many
others. Hence perhaps any reference to the “enteron” in
the group-name of these pigments is misleading, and such a
term as “ polychromes ” would be preferable.

REFERENCES.

1. HerpMan, W. A.—‘‘Note on a New British Echiuroid Gephyrean,
with Remarks on the Genera Thalassema and Hamingia,” ‘ Quart.
Journ. Micr. Sci.,’ vol. xl (1897), pp. 367—3884, 2 pls.

2. Kruxenperc, C. Fr. W.—‘‘ Ueber das Helicorubin und die Leberpig-

mente von Helix pomatia,” ‘ Vergleich. Physiol. Studien,’ 2te
Reihe, 2te Abt. (1882), pp. 63—69, 1 fig.

430 MARION I. NEWBIGIN.

3. Krukenpere, C. Fr. W.—“ Ueber das Bonellein und seine Derivate,”
‘Vergleich. Physiol. Studien,’ pp. 70—80, 1 fig. (Gives earlier
references.)

4. Krukensere, C. Fr. W.—“ Beitrage zur Kenntniss der Actinien-
farbstoffe,” ‘ Vergleich. Physiol. Studien,’ 3te Abt. (1882), pp. 72—87,
pk

5. Lankester, HE. Ray.—“ On the Green Pigment of the Intestinal Wall of
the Annelid Chetopterus,” ‘Quart. Journ. Micr. Sci.,’ vol. xl
(1897), pp. 447—468, 4 pls.

6. MacMunn, C. A.— Observations on the Colouring Matters of the So-
called Bile of Invertebrates, on those of the Bile of Vertebrates, and
on some Unusual Urine Pigments, &c.,’’ ‘Proc. Roy. Soe. London,’
xxxv (1883), pp. 370—403, 1 chart.

7. MacMunn, C. A.—“ Further Observations on Enterochlorophyll and
Allied Pigments,” ‘Trans. Roy. Soc. London’ (1886), elxxvii, pp.
235—266, 2 pls.

8. MacMuny, C. A.—“ Observations on the Chromatology of Actiniz,”
‘Trans. Roy. Soc. London’ (1885), clxxvi, pp. 641—668, 2 pls.

9. Mosr.ey, H. N.—“ On the Colouring Matters of Various Animals, and
especially of Deep-sea Forms, dredged by H.M.S. ‘Challenger,’ ”
‘Quart. Journ. Micr. Sci.’ (1877), xvii, pp. 1—28, 2 pls.

10. Scuunck, Epwarp.—“ Contributions to the Chemistry of Chlorophyll,”
‘Proce. Roy. Soc. London,’ xxxviii (1885), pp. 386—340.

11. Sorsy, H. C.—‘On the Colouring Matter of Bonellia viridis,”
‘Quart. Journ. Mier. Sci.,’ xv (1875), pp. 166—172, 1 fig.

EXPLANATION OF PLATES 30 & 81,

Illustrating Marion I, Newbigin’s paper “On Certain Green
(Chlorophylloid) Pigments in Invertebrates.”

Fic. 1.— Absorption spectrum of normai chetopterin for comparison with its
derivatives, showing four distinct bands and a trace of the band to the right
of D.

Vie. 2.—Spectrum of a blue acid solution of chetopterin, showing the
result of an intermixture of the green acid derivative and acidified Cheto-
pterin.

ON CERTAIN GREEN PIGMENTS IN INVERTEBRATES. 431

Fig. 3.—Spectrum of the acid derivative obtained by dissolving the pre-
cipitate from an acid solution in ether, and treating the ether with concentrated
acid. The band to the right of the C line is the band of the derivative, the
other two are due to the presence of traces of normal cheetopterin.

Fig. 4.—Spectrum of a solution of chetopterin which has stood for some
time with ammonia; the band between those near C and D indicates the
presence of the ammonia derivative, the other four are the bands of
cheetopterin.

Fic. 5.—Spectrum of the alkaline derivative obtained by allowing cheto-
pterin to stand for a prolonged period with dilute alkali. The solution no
longer contains unaltered chetopterin.

Fic. 6.—Spectrum of a solution of “enterochlorophyll” obtained from the
feeces of Patella; the solution contains little or no lipochrome.

Fic. 7.—Spectrum of the same solution after the addition of hydrochloric
acid ; note especially the position of the band in the red.

Fic. 8.—Spectrum of the acid derivative of enterochlorophyll intermixed
with a trace of the original pigment. The spectrum corresponds to that
figured in 3 for cheetopterin, and the solution was obtained in a similar way.
The second band is that of the derivative. The solution was too dilute to
show the band corresponding to the third band of Fig. 3.

Fig. 9.—Section of visceral hump of Patella, showing the epithelial cells
of the intestine with their pigment granules, and sections of the liver tubules.
t. Intestine. 7. ¢. Liver tubules.

Fie. 10.—A few cells from the intestine, more highly magnified to show the
pigment granules. xz. Nucleus. p. Pigment granules.

Fic. 11.—Section of liver tubule more highly magnified, showing the pig-
mented vesicles, v., and the numerous oil-drops scattered through the proto-
plasm. 2. Nucleus.

Ly os ao Pg glia GR Tie kg ca Tae liek

=-%" aa ;
: ; ie 2 oe
> oe Ay vite te%—l AA nae ae

= = ’ 7 7 » ?

7 . a ay : . a ‘a yr Te he = Ry »
BU) uae "2% Nepales aplasia ati? i

het i Sw abel p> ag AND ed a eet AO! ae
a ere assis) 20d Pe. oh a ¥. er lithe 1 >
a o 4 & -

ag} _ oe 4 ge ri a wD gid . a | Ve tap) day > ant
4 ’ Ge _ i rr : eis a. ie, as a iP i

| 2a te De Ge) cos! iggy, Agana ie a

_ =v iP
i f

2 : y " cae

: ry ; ids Piha. 48,
: os Lye 4 '
ds sh} 2 Zh, “ona

Sia v4 ‘ inn se

a |

NOTE ON A (P STOMATOPOD) METANAUPLIUS LARVA. 433

Note on a (? Stomatopod) Metanauplius
Larva.

By

J. J. Lister, M.A.,
Demonstrator of Comparative Anatomy in the University of Cambridge.

Tue larva to which I wish here to invite attention was
caught in a tow-net out at sea, off the south coast of Tasmania,

A. The larva here described. (The right anterior antenna was omitted in
the original drawing, and in order that the figure may be as far as possible a
reproduction of this it is omitted here.)

B. The Erichthoidina stage of a Stomatopod. (From H. J. Hansen,
Isopoden, Cumaceen, und Stomatapoden der Plankton Expedition, ‘ Ergebnisse
der Plankton Expedition,’ Bd. ii, G.c., pl. viii, fig. 14. a’, a’. First and second
antenne. abd’. First abdominal appendages. ¢.f. Caudal fork. p./.s. Postero-
lateral, and p.m.s., postero-median spines (the latter is rather too large in A).
r. Rostrum.

during daylight on December 25th, 1886. I had at that time
very small acquaintance with larval forms of Crustacea, and

434 Jo J. LISTER.

did not recognise the features of particular interest which
this larva presented. Having made the drawing which is here
reproduced (A) and a few notes, I paid no further attention
toit. To my regret, I cannot now find the specimen. Although
the evidence in my possession is thus very imperfect, I have,
after some hesitation, decided to publish it, because it appears
to throw some light, though far from a bright one, on an
obscure corner of crustacean larval history.

As may be seen from the figure, the body is enclosed in a
large transparent shield, produced anteriorly into a strong
rostral spine, and posteriorly into a smaller median spine. On
either side the lateral parts of the carapace fold round the
body of the larva, and where the ventral and posterior borders
meet, a small, backward pointing, postero-lateral spine is
situated. A median eye is present beneath (in a ventral view)
a low eminence, and two large globular compound eyes project
on either side of the base of the rostral spine. Of the two pairs
of antennee, the first seems to have been simple, and the second
is biramous, possessing a short endopodite, and a well-developed
swimming exopodite, jointed and beset with long sete. The
body appears to have been unsegmented, and the posterior part
is small, free from the dorsal shield, and, in the position drawn,
strongly flexed ventrally. It terminates in a caudal fork, the
divisions of which are articulated and setose. The dorsal
region of the posterior part of the body was tinged with red.
A note attached to the drawing calls attention to “ rudimentary
appendages”’ behind the second antennz, and states that a
heart was to be seen (in the position indicated) under the hinder
part of the carapace. There is some indication in the drawing
of an upper lip, between and a little behind the second antenne,
and the “‘ rudimentary appendages” are shown to the number
of perhaps three, between this and the flexed posterior part of
the body.

It will, I think, be admitted that the larva is in the meta-
nauplius stage. It seems improbable that the mandibles were
really rudimentary, but the mandibular palp was at any rate
inconspicuous, and two, perhaps, of the succeeding pairs of

NOTH ON A (P STOMATOPOD) MBTANAUPLIUS LARVA. 435

limbs had already made their appearance. As far as the
development of the limbs is concerned, the stage appears to
correspond with that of Euphausia pellucida, represented
by Metschnikoff in the ‘ Zeitschrift fiir Wiss. Zoologie,’ Bd.
xxi, plate 34, fig. 6. Further, it is clear from the character of
the eyes that we have to do with a Thoracostracan form. Itis,
then, a Thoracostracan larva at about the metanauplius stage.

In the Cumacea the young leave the brood-pouch nearly in
the form of the adult. The young of the Thysanopoda,
among the Schizopods (as shown by Metschnikoff), as well,
probably, as those of the Decapod Peneus (Fritz Muller 2) are
hatched as nauplii, while those of Lucifer (Brooks 8) appear
as metanaupli. But in all these the paired eyes are absent in
the metanauplius stage, and are not fully developed until a long
and fully segmented abdomen has been formed. ‘The carapace
is without spines in the metanauplius stage, and though in
Kuphausia, and also in Lucifer, spines make their appear-
ance in later stages, corresponding in position with those above
described, their shape in Kuphausia and the shape of the
shield in both genera are markedly different.

The remaining Thoracostracan group is the Stomatopoda.
In Squilla (Paul Mayer 4) and Gonodactylus (Brooks and
Herrick °) the eggs have been seen to hatch as Alima larvee
in a stage which has been compared with the Zocea stage of
Decapods. But, as was shown by Claus,® larve of Stomato-
pods also occur in another form, the Erichthus, of which
stages are known prior to that at which the Alima larva is

1: Loc. cit.

? “Die Verwandlung der Garneelen,” Erster Beitrag, ‘Arch. fiir Natur-
geschichte,’ Jahrg. 29, 1863.

° “ueifer: a Study in Morphology,” ‘ Phil. Trans.,’ vol. elxxiii, 1882,

. 57.
: 4 Carcinologische Mittheilungen, IX,” ‘Mittheilungen aus dem Zool.
Stat. zu Neapel,’ vol. ii, p. 219.

* “Embryology and Metamorphosis of the Macroura,” ‘ Memoirs of the
Nat. Acad. of Sciences,’ vol. v, p. 353.

6 «Die Metamorphose der Squilliden,” ‘ Konig. Gesell, d. Wissenschaften
zu Gottingen,’ Bd, xvi, 1871,

436 J. d. LISTER:

hatched. The youngest known member of this series of forms,
the Erichthoidina, is represented in fig. B.

The carapace is furnished with spines resembling those of
the larva shown in fig. A, and well-developed compound eyes
are present. The first antenne are obscurely biramous, while
the second are uniramous. The thorax is distinctly seg-
mented, and the five anterior segments bear biramous swim-
ming feet, while the three following segments are without
appendages. The first pair of abdominal feet are present, but
behind the segment bearing them, the abdomen is unseg-
mented, and ends in a large truncated telson.

In succeeding stages the Erichthoidina larva changes into a
Zocea-like form comparable with the Alima larva of Squilla
and Gonodactylus, but the stages which precede the Erich-
thoidina are unknown.

Now the development of the Erichthus larva differs from
that of the Schizopod and Decapod larve with which we are
acquainted, in one respect, among others, namely, that in a
stage, the Hrichthoidina, antecedent by many moults to the
Zocea stage, the paired eyes are already well developed. In this
respect, as well as in the shape and large size of the carapace,
the disposition and direction of its spines, and in the fact that
it is a reduplication exclusively of the cephalic region, the
larva under consideration resembles the Erichthoidina.

I would submit then, that it is rather probable that this larva
is a Stomatopod larva at a stage prior to the Erichthoidina
stage ; that it is, in fact, a Stomatopod metanauplius.

The condition of the antenne, the first apparently simple,
and the posterior biramous, differs from that found in the
Erichthoidina stage, but the difference is precisely that which
from analogy with other nauplii we should expect to find in
the metanauplius form. The first antenna of the stage figured
by Hansen appears to be just acquiring its biramous character.
The larva appears to be unique among those hitherto de-
scribed in possessing well-developed compound eyes in the
metanauplius stage. The articulated condition of the divi-
sions of the caudal fork is also, so far as I know, peculiar

NOTE ON A (P STOMATOPOD) METANAUPLIUS LARVA. 487

among the Malacostraca; and is interesting in view of the
great prominence of these processes in that isolated and pri-
mitive Malacostracan form Nebalia, and in the Phyllopods
A pus (in which again they are articulated) and Branchipus,
with which Nebalia forms a connecting link.

ON THE NEPHRIDIA OF THE POLYCHATA. 439

On the Nephridia of the Polycheta.
Part II.—Glycera and Goniada.'

By

Edwin 8. Goodrich, B.A.,
Aldrichian Demonstrator of Comparative Anatomy, Oxford.

With Plates 32—35.

In the following pages are described the nephridia and
“ciliated organs” of the two closely allied genera, Glycera
aud Goniada. The observations on the first genus were begun
on fresh and preserved material obtained at Naples in 1896
and 1898, where I examined the three common species,
Glycera unicornis, Sav., Gl. siphonostoma, D. Ch., and
Gl. convolutus, Keferst, continued on the coast of Nor-
mandy, where Gl. convolutus is fairly abundant, and com-
pleted here in Oxford. Unfortunately I have not been able to
obtain living specimens of the genus Goniada, and the work on
this form has been therefore entirely carried out on preserved
specimens of the species G. emerita, Aud. and Edw., and G.
maculata, Oerst.?

To Dr. Lo Bianco I am indebted for a fine specimen of the
rare Goniada emerita, to Prof. Benham for a ripe specimen
of G. maculata, and to Dr. Levinsen, of Copenhagen, Dr.

1 Part I is published in vol. 40 of this Journal (1897), and contains an
account of the nephridia of Nephthys.

2 T am glad to have the opportunity of thanking Professor Dohrn for
generously placiug a table at my disposal last winter in the Stazione

Zoologica, and to Professor Perrier for the kind way in which he received
me in his laboratory near St. Vaast-la-Hougue last summer.

440 EDWIN S. GOODRICH.

Appelof, of Bergen, and Dr. Thiel, of Stockholm, for other
specimens of the same species.

GLYCERA.

In this genus the “ciliated organ” is so closely connected

with the true nephridium, although it does not actually open
into its lumen, that the two structures have been confounded
by Ehlers (2), the only observer who gives an account of the
excretory organs of Glycera.

The nephridium, the ciliated organ, and a peculiar organ
which I shall call the nephridial sac, are all united into a
single structure which may for convenience be called the
nephridial complex. Ehlers, in his well-known work on the
Polychzeta, described the position of this nephridial complex
on the anterior surface of the septum accurately enough ; but
he failed to make out its real structure, partly no doubt owing
to the fact that he studied it in spirit specimens only. What
he describes as the duct of the segmental organ would appear
to be, judging from his figure, the extended outer lip of the
‘ciliated organ ;” and what he took for an internal opening
may perhaps be another portion of the same. The nephridium
itself escaped his observation ; and, indeed, it is very difficult
to make out in preserved specimens.

Before describing these organs in detail it is best to give a
general account of their distribution, shape, and mutual re-
lations.

On examining a fresh specimen of Glycera siphonostoma
which has been opened up dorsally, and from which the gut
has been removed, small round bodies can be seen on either
side near the base of each parapodium, attached to the front
face of the septum, and lying below and partially hidden by the
bundle of chet with its muscles (fig. 1). One pair of these
bodies occurs in every segment excepting the first few and the
last one or two; in a full-grown specimen they appeared to be
absent in the first twenty segments. ach body constitutes
what I have called the nephridial complex, consisting of a
hollow sac, the “nephridial sac,” forming a flattened disc-like

ON THE NEPHRIDIA OF THE POLYCHATA. 441

organ, into which opens the “ ciliated organ.” The outer lip of
the latter stretches out to the body-wall. The true nephridium is
spread over the surface of the nephridial sac, and has no internal
opening (ef. figs. 1, 3, and 30). The description applies to
both sexes.

The Nephridium.—The structure of the nephridia in the
three species of Glycera I have studied differs only in detail.
The nephridium opens to the exterior ventrally by a minute
pore, situated just outside the limit of the large bundles of
ventral longitudinal muscles. The nephridiopore leads into a
very slender canal, difficult to follow in sections, which passes
through the body-wall into the septum above; then running
inwards and piercing the septum, the canal reaches its anterior
face, where it soon joins the nephridial sac. On arriving here
the canal divides repeatedly, giving off branches which spread
over the outer surface of the sac.

In Glycera convolutus, where the nephridium is small
and the sac scarcely developed (figs. 13 and 21), the body of
the nephridium forms a somewhat flattened pear-shaped mass
applied to the ciliated organ. A fresh nephridium dissected
out and examined in sea water appears to consist of a proto-
plasmic mass in which the lumen of the canal branches, form-
ing a sort of sponge-work. Along the course of the canals
are rounded diverticula or chambers, projecting towards the
free ceelomic surface of the organ (figs. 13 and 12).

The wall of the lumen of the canal leading to the exterior,
and of its main branches, is provided with long cilia (fig. 12) ;
the smaller branches leading from the chambers are also
ciliated, but to a less extent. By their action the cilia tend to
drive a current from the chambers towards the external pore.

The protoplasmic walls of the nephridium are very granular,
being more or less loaded with excretory matter in the form of
granules or droplets, small and large, which often give the
organ a yellowish colour. There are nuclei here and there,
but I have seen no distinct cell outlines.

The outer wall or roof of the chambers projecting towards
the ccelom is very thin, and arising from near the centre of

vot, 41, PART 3,—NEW SERIES. HH

44.2 EDWIN 8S. GOODRICH.

each is a “ tube-bearing ” flagellated cell, somewhat similar in
structure to those I have described in Nephthys (4). Occa-
sionally two or even three tubes may open into the same
chamber (figs. 12, 14, and 21).

Each of these peculiar cells consists of a little rounded mass
of finely granular colourless protoplasm, in which is placed the
round nucleus, supported at the free end of a long conical
tube. As in the case of Nephthys, so in Glycera, the nuclei
of the tube-bearing cells have the property of staining very
deeply and rapidly. The tube itself is formed of a thin layer
of cuticular substance; it is flattened from side to side,! in-
serted by its narrow end into its cell, and by its broad end into
the roof of the nephridial chamber. Thin longitudinal lines
give it the appearance of being delicately fluted. A long
flagellum attached to the cell at the apex of the tube works
rapidly within the latter, reaching into the underlying nephri-
dial cavity.

The tube-bearing cells rarely, if ever, stand alone. They are
not ranged in rows as in Nephthys, but are dispersed over
the whole surface of the nephridium in pairs, or in groups of
three, four, or even five cells, resting against each other, no
doubt for mutual support. In this position the cells form
roundish masses without actually fusing, and project into the
coelomic fluid, standing on their tubes as on stilts. The tube
is so delicate, and the entire apparatus so slender, that the
mere action of the flagellum inside often makes the whole cell
waggle backwards and forwards. In a teased preparation of a
nephridium the tube-bearing cells may break away from the
chambers, and move about in the fluid actuated by the long
projecting flagellum: in this condition their resemblance to
the collar-cells of sponges is most striking (fig. 5). For these
peculiar nephridial cells bearing a tube and a flagellum, which
I have hitherto designated by the cumbrous descriptive term
“‘ tube-bearing cell,” both in the Nephthyide and in the Gly-
ceridee, I now propose the more convenient term solenocyte
cwAiv, a pipe). As far as I have been able to make out

1 The tube is shown in profile in fig. 14, at the left-hand lower corner,

ON THE NEPHRIDIA OF THE POLYCHATA. 443

there is no communication of the lumen of the ne-
phridium with the celom, either directly or indi-
rectly, through the nephridial sac, in this or any other species
of Glycera.

Over the chamber-bearing surface of the actual nephridium
there appears to be no regular layer of ccelomic epithelium.
An occasional nucleus here and there may indicate its remains,
but the nephridium seems to have made its way through the
epithelium, as in the case of the nephrostome of an ordinary
earthworm.

In the much larger species, Gl. siphonostoma, the nephri-
dial complex is a structure of considerable size (figs. 8 and
30), visible even to the naked eye. Figs. 13 and 8 represent
these organs in G1. convolutus and siphonostoma respec-
tively, drawn to the same scale. Here the nephridial sac is
very much more developed, and the nephridium itself extends
almost all over its surface, forming a sort of outer layer or
shell (fig. 30). The system of branching canals is extremely
complicated, and the number of chambers immense. Their
structure is best studied in the very similarly developed species,
Gl. unicornis.

The solenocytes (tube-bearing cells) in Gl. siphonostoma,
though so much more numerous, closely resemble those just
described in Gl. convolutus. Occasionally I have noticed
little ameeboid processes with thickened ends radiating from
the cells (fig. 4); they remind one of the pointed proto-
plasmic processes originating from the tube-bearing cells of
Nephthys (fig. 4), but appear to arise rather from the bases of
the cells where these are applied to each other.

The third and last form which I shall describe, Gl. uni-
cornis, is intermediate in size between the first two species.
Here the sac is large, but the nephridium does not, as a rule,
cover over its whole surface, being specially developed along
the rim of the disc (figs. 2, 11, 16, and 26). There seems,
however, to be considerable variation in the extent to which
the nephridium spreads over the sac; in some specimens it
forms a layer covering its whole surface, as in Gl. siphono-

444, EDWIN S. GOODRICH.

stoma, whilst in others this is not the case. These different
conditions may be dependent upon the state of expansion of
the thin-walled underlying nephridial sac.

The structure of the nephridium can be very well studied in
this species. The nephridial canal is divided into several
branches which spread over the surface of the organ; the
branches all converge towards and finally open into the canal
leading to the exterior, and their lumen is provided with
powerful long cilia (fig. 24).

Coming off from the inner surface of these branches are
numerous secondary canals, which branch repeatedly, and
form a network throughout the substance of the nephridium
leading from one chamber to another (figs. 24, 29, 32, and fig.
15 of Gl. siphonostoma). The secondary canals appear not
only to branch, but to anastomose ; the canals coming from one
main branch, however, do not seem to open into those coming
from another. The system, except for the anastomosis, may
perhaps be compared to a river, tributaries of which are
separated by watersheds. Here and there the secondary
canals lead up to the very numerous chambers into which
open the tube-bearing cells.

In Gl. unicornis the solenocytes are generally distributed
in pairs, never in groups, and are intermediate in structure
between those of Nephthys and those of the species of
Glycera described above. Instead of being entirely sup-
ported by the tubes, as in the latter, the cells are attached
at their base to the wall of the nephridium (figs. 9 and 32) by
a short stalk. The cell is more elongated, and a neck of
considerable length bends round from the body of the cell to
the top of the tube. Although in some specimens this neck is
quite long, yet it is always much shorter than in Nephthys
(fig. 10). ‘The nucleus is large and oval.

Such is the structure of a typical tube-bearing cell in G1.
unicornis ; but in many cases these cells seem to resemble
those of Gl. siphonostoma (fig. 29). These apparent
exceptions may possibly be due to the worms having been
wrongly identified; I am inclined to believe, however, that

ON THE NEPHRIDIA OF THE POLYCHATA. 445

there is considerable variation in the structure of the nephridia,
perhaps owing to the specimens being of different ages or in
different stages of maturity.!

The Ciliated Organ.—The ciliated organ in G1. unicornis
and siphonostoma forms a considerable part of the nephridial
complex. It is of essentially the same structure in both these
species, and can be seen in a dissected specimen as a thick
band running from the nephridial sac, in front of the septum,
to the body-wall near the base of the parapodium (figs. 1
and 2).

This band is hollowed out on its upper and anterior surface
by a groove which runs longitudinally along it, becoming
deeper at its inner end, where it reaches the sac and passes
into its interior (figs. 16 and 80).

Having entered the mouth of the sac, which aperture it
entirely surrounds, the ciliated organ is bent back on itself
so as to extend into the *‘ cecum,” a region of the nephridial
sac which will be described farther on.

Outside the sac the edges of the grooved ciliated organ are
drawn out into two pointed flaps guarding the opening (fig. 8,
ji.). This structure forms a sort of one-sided funnel, and from
the free edge inwards extend slight ridges (fig. 8, 7.), more
conspicuous in living specimens, which probably represent in
a very rudimentary condition the high ridges characteristic of
the ciliated organs in so many forms, such as Nereis and
Hesione (3 and 4).

The whole of the grooved surface of the organ is provided
with a dense covering of cilia, by the action of which floating
bodies are driven into the sac. The ciliated organ is formed
on its inner surface of ordinary ciliated columnar epithelium,
consisting of narrow cells with oval nuclei, which is continuous
at the edges with the flat coelomic epithelium, by means of
which it is attached to the septum, muscles, and body-wall
along its course (figs. 16, 17, 26, 27, and 28).

It should be noticed that the band-like grooved outer lip of

1 I have found the typical fixed solenocytes in quite small and apparently
young specimens of GI, unicornis.

44.6 EDWIN S. GOODRICH.

the ciliated organ varies considerably in development in
different individuals. Not always does it reach the body-wall,
and I am inclined to think, after examining a very large
number of worms, that it is more developed in those animals
which approach sexual maturity.

In the small species, Gl. convolutus, the ciliated organ is
relatively little developed, without the pointed flaps on either
side of the entrance into the nephridial sac, and never reaching
the body-wall so far as I have been able to make out (figs. 13
and 21).

The Nephridial Sac.—As already mentioned, this in-
teresting organ consists of a round flattened sac, over which
the nephridium lies. It must be clearly understood that
structurally, and no doubt also in its origin, it is quite
distinct from the nephridium, and is only called
‘‘ nephridial ” because of its close connection with that organ.

It is, in fact, a hollow sac formed by a layer of flattened
epithelium continuous with the epithelium of the ciliated
organ, and communicating only with the celom by means of a
single opening (figs. 2, 3, 16, and 30).

Just as the ciliated organ is to be considered as formed of
differentiated coelomic epithelium, so no doubt the sac is
essentially a mere pouching of this same epithelium covering
the nephridium at that region towards which the ciliary action
of the ciliated organ converges—or, in other words, at the point
where this organ accumulates the cells floating in the cceelom.

In Gl. convolutus the nephridial sac is quite a small
chamber (figs. 13 and 21), into which the lip of the ciliated
organ is produced, forming its wall on one side.

In connection with the quite anterior nephridia of Gl.
unicornis and Gl. siphonostoma the sac is not much more
developed than in the first species (fig. 11); but farther back
it gets larger and larger, until it becomes a relatively huge and
almost spherical structure (fig. 3).

Moreover these large nephridial sacs become differentiated
into two regions—a main rounded sac which opens to the
ceelom, and a diverticulum from the main sac, to the blind end

ON THE NEPHRIDIA OF THE POLYCHA'A. 44:7

of which extends the lip of the ciliated organ (figs. 16, 17, and
30).

The latter division, which I shall call the cecum, extends
between the nephridial duct and the main limb of the ciliated
organ. It is best developed in G1. unicornis.

On examining this cecum more closely its cavity is found
to be subdivided by means of thin walls, formed apparently by
folds of the epithelium, projecting inwards from the side
opposite to that on which the ciliated organ is situated (figs. 19,
20, and 380). The chambers thus formed, resembling somewhat
the cells of a honeycomb, are partially, but not entirely, cut off
from the main cavity by roofing extensions of the walls (fig.
19);

At the blind extremity of the cecum the chambers become
small, and disappear near the ciliated epithelium. At the
opposite open end they extend into the main cavity of the sac,
becoming more and more shallow, and soon dying out (fig. 30).

The contents of the sac will be described below.

On the Functions of the Ciliated Organs, Nephridial
Sac,and Nephridium, and ontheCelomic Fluid of the
Glyceridw.—The functions of the organs forming the nephri-
dial complex are no doubt connected with excretion. When
the worm attains sexual maturity the ciliated organ probably
acts as a genital duct; of this, however, I have no direct proof,
and the discussion of the matter must be reserved for a future
paper.

The nephridium itself is of course a kidney excreting waste
matter. In freshly killed specimens the walls of this organ
are found, as already mentioned, to be full of dark granules
and paler droplets of varying size, which sometimes flow
together, forming quite large drops embedded in the proto-
plasm. As might be inferred from the fact that the nephridium
does not communicate with the ceelom, it seems to be entirely
concerned in the excretion not of solid particles, but of substances
dissolved in the coelomic fluid. These appear to be stored up
as granules and droplets, which are subsequently discharged
into the lumen of the nephridial duct.

44.8 EDWIN 8. GOODRICH.

This view is supported by the following experiment :—If a
living worm be injected with a mixture of Indian ink and
carmine in sea water, and opened a few hours after, it will be
found that of the mixture which has entered the celom the
solid particles have been ingested by the ameeboid cells, whilst
the small quantity of carmine which was dissolved has been
taken up by the nephridium. In such a specimen the nephridia
are tinged a delicate pink colour, which can be distinctly seen
with a lens.

The solenocytes do not appear to be in any way affected by
the injection, and the pink colour is entirely due to the carmine
having been deposited in globules occupying the same position
in the nephridial cells as the yellow excretory matter in an
uninjected specimen. Never have I found solid particles of
carmine or Indian ink in these cells.

The nephridial sac, on the other hand, seems to be concerned
with the elimination of solid waste products. Before discussing
this question, however, I wish to make a digression on the
subject of the coelomic fluid of Glycera.

It is well known that in this genus there is no separate
canalicular blood system, and that the coelomic fluid contains
numerous hematocytes, or round flattened nucleated cells,
stained red with hemoglobin.

In the celomic fluid of Gl. convolutus are found a large
number of these round hematocytes deeply stained with
hemoglobin, a relatively small but yet considerable number of
white amoeboid cells, leucocytes, and a number of rather larger
oval and flattened cells containing minute colourless granules
(these cells are quite similar to those found in Gl. unicornis,
shown in fig. 6).

On examining the ccelomic fluid of specimens which have
been injected with carmine and Indian ink, it is found that
the foreign granules have been taken up rapidly by the
leucocytes, which soon become filled with them. No particles
occur in the red or in the oval cells; these would appear,
then, to be neither ameeboid nor phagocytal.

In Gl. siphonostoma the majority of celomic cells are

ON THE NEPHRIDIA OF THE POLYCHATA. 449

faintly tinged with hemoglobin and of very irregular shape,
being generally covered with amceboid processes. Normal
rounded hzematocytes are occasionally present, especially in
young specimens. The numerous processes on the hemoglo-
binous cells give them a spiny appearance ; but, as a matter of
fact, the pseudopodia are not merely spine-like in shape, being
really thickenings in thin expansions of protoplasm. When
these cells are watched under the microscope the pseudopodia
can be seen to begin as little rounded knobs, which gradually
expand, spreading out in thin sheets, supported here and there
by ribs or thickenings. It is these which, on a casual glance,
have the appearance of freely outstanding processes (fig. 22).
The granular oval cells occur also in small numbers, but the
ordinary white amceboid corpuscles appear to be very rare or
entirely absent in this species. On injecting a Glycera
siphonostoma with carmine or Indian ink, we find that the
granules are taken up in quantities by the hemoglobinous cells,
Just as in the previous species they are absorbed by the leuco-
cytes. The hematocytes are therefore both ameboid
and phagocytal. This, so far as I am aware, is the first
instance in which a free cell containing hemoglobin has been
shown to ingest foreign particles.

Gl. unicornis is, in respect to its celomic cells, interme-
diate between the two species described above. As a rule, in
Gl. unicornis normal round hematocytes occur in the celom,
together with a number of amoeboid leucocytes and the usual
granular oval cells (fig. 6). Such specimens, when injected,
show that the particles of carmine or Indian ink are taken up
exclusively by the leucocytes.

On the other hand, it may be frequently observed that the
leucocytes are very rare, and that the hematocytes have a
more or less pronounced tendency to produce pseudopodia. In
accordance with this it is often found in injected worms that the
particles have been ingested by the red hemoglobinous cells.

I have not met with any evidence distinctly supporting the
view that the hematocytes are modified leucocytes, or vice
versa; yet there seems to be no doubt that the functions of

4.50 EDWIN 8S. GOODRICH.

ordinary leucocytes are assumed by the hematocytes in Gl.
siphonostoma, and to a lesser extent in Gl. unicornis.

We can now return to the study of the function of the
nephridial sac.

In the large species Gl. unicornis the nephridial sac is
always more or less filled with a mass of cells (figs. 3, 8, 16, and
26), consisting of some hematocytes and a large number of
amceboid cells. In Gl. siphonostoma the cells in the nephri-
dial sac are almost all amceboid hematocytes. These cells are
no doubt brought in by the action of the ciliated organ from
the celom. They penetrate also into the cecum, but in fewer
numbers. In the sac are generally found masses of waste matter
in the form of concretions, brown or yellow granules, and irre-
gular aggregations, together with bits of chetz and any other
particles which may occur in the ceelom (figs. 7 and 25).

Similar but smaller aggregations of waste material may be
found floating in the celom, especially towards the posterior
end of the body ; they are either actually within or surrounded
by a number of ameeboid cells,! which evidently move about as
scavengers, and eventually find their way either singly or in
masses into the nephridial sac.

The accumulation of such waste materials in the nephridial
sacs gives these organs a brown or black colour in the poste-
rior region of the body. The waste materials do not as a rule
penetrate into the cecum.”

Experimental proof of the account given above is afforded on
examining specimens which have been injected with powdered

1 T can confirm Cuénot’s observation that in such cases the leucocytes
secrete a chitinous substance round the foreign body, apparently as a protec-
tive measure. For instance, a broken piece of cheta will be found with the
jagged ends covered over with concentric layers of secreted substance.

2 The ciliated organ has already been described as producing a current
from without inwards; occasionally, however, the action of the cilia appears
to be reversed, leading from the cecum outwards. The whole mass inside
the sac is sometimes seen to rotate, and it is possible that this is the normal
action of the cilia in the living worm. When the nephridia are dissected out
and placed on a slide under a cover-slip they are naturally subject to pressure,
which would be sufficient to impede the rotation,

ON THE NEPHRIDIA OF THE POLYCHATA. 451

carmine or Indian ink. In such worms the nephridial sacs are
found to be crammed with cells loaded with particles of these
substances. I have noticed that many of the loaded ameeboid
cells make their way into the cecum.

There can be no doubt, then, that solid waste matters are
accumulated in the nephridial sac. We may now ask what
becomes of them when they have reached this cul-de-sac.

In connection with this question I may now describe
another peculiar variety of cell which occurs in the nephridial
sac and its cecum. These cells are large, and generally of an
irregularly oval flattened shape; they are distinguished by the
possession of an immense number of minute colourless
granules, giving the whole cell a characteristic greyish appear-
ance (fig. 23). They are seen in numbers creeping over the
inner surface of the nephridial sac (figs. 3, 18, and 19, gr. ¢.),
and also in the secondary chambers of the sac and cecum. In
the latter position, indeed, they are always found, sometimes
being flattened against the walls, and at other times piled up
in rows (figs. 19 and 20, gr. c.). The nucleus of these finely
granular cells is often of very remarkable appearance, being
most irregular in form, like a hollow sphere, or very frequently
bent round so as to form a horse-shoe shape (figs. 18 and 20).

Appearances have led me to believe that these cells originate
in the cecum, but I have never been able to find convincing
proof of this supposition. They bear a great resemblance to
the granular oval cells already described as occurring in the
ccelom, and it is quite possible that they are really derived from
this source.

Never are foreign particles ingested by them, and they do
not appear to be at all of a phagocytal nature, although occa-
sionally amoeboid. The fine granules in them are evidently an
endoplastic product, and I should suggest that the cells secrete
a ferment which helps to dissolve the waste material in the
sac. The matter in solution would then be carried through the
wall of the sac to the nephridium to be excreted. However
this may be, it is certain that the cells aggregated in the sac die
and undergo degeneration ; this can be clearly seen in sec-

452 EDWIN S. GOODRICH.

tions through the organ, where cells and nuclei can be found in
every stage of dissolution (figs. 26 and 25).

It would appear, then, that the ciliated organ and nephridial
sac are concerned in the gathering up, through the agency of
phagocytes, of the solid waste products found in the celom,
whilst the function of the nephridium is to eliminate the
soluble excretory material derived from the ccelomic fluid, and
also perhaps from the sac. The function of the solenocytes
or tube-bearing cells themselves is possibly analogous to that
of the Malpighian capsules in the Vertebrate, namely, to
excrete liquid, which presumably can pass by osmosis through
the thin wall of the tube.

GONIADA.

The Nephridium.—As in Nephthys and Glycera, so also
in Goniada, the nephridium is an organ without opening into
the celom, and with a branching termination provided with
tube-bearing flagellated cells. The organ can best be studied
in the large species G. emerita. It consists of a ciliated
canal leading to the exterior, which, except near the internal
end, is much wider than in Glycera. This canal passes
through the septum and emerges on the anterior face, where
it branches to form a lobed terminal organ almost exactly
intermediate in structure between that of Nephthys and that
of Glycera. For whereas in the latter this region spreads out
to form a plate, the surface of which is evenly covered with
nephridial chambers and scattered solenocytes, and in the
former it divides into long branches with a regular row of
solenocytes on either side facing each other, in Goniada
emerita the terminal organ consists of massive short branches
or lobes, on which the solenocytes are set in somewhat regular
rows, generally, but perhaps not invariably, facing each other
(figs. 833 and 35). These cells are also themselves of inter-
mediate character : as in Nephthys, they are fixed at the base,
have oval nuclei, long necks, and long tubes; but the tubes
and necks are relatively short, resembling those described
above in Glycera unicornis,

ON THE NEPHRIDIA OF THE POLYCHATA. 453

In G. maculata! the terminal organ of the nephridium is
much smaller, and appears in section as little more than a
bunch of solenocytes (fig. 39).

The Ciliated Organ.—It has been my good fortune to
obtain a perfectly ripe male specimen of Goniada maculata,
which proves, I think, beyond the possibility of doubt the fact
that the ciliated organ acts as a genital duct or
funnel,—in this species, at all events. This is all the more
gratifying to me, since I suggested that this is its function
when I first described this organ in the Lycoridea five years
ago (3).

It will be remembered that Goniada is divided into two
dissimilar regions, somewhat as the heteronereid phase of
Nereis. Sections show that in the first few segments both
nephridia and segmental organs are absent; that in the
anterior region generally the nephridium is present, but the
ciliated organ not at all, or scarcely, developed. It is only in
the posterior region that this organ is fully formed in the
adult. In these segments it takes the shape of a wide-
mouthed funnel, a trumpet-shaped structure, the lips of
which spread and gradually thin out on the septum. Here
the thickened ciliated epithelium, of which the wall of the
funnel is formed, flattens out and passes into the ordinary
celomic epithelium lining the body-cavity (figs. 36, 39, and
31). Bunches of cilia may be seen on this epithelium some
way up the septum (fig. 36).

The ciliated cells of the funnel are remarkably striated, and
the cilia numerous and very powerful (fig. 40).

The funnel leads into a wide tube, which passes backwards
to join the nephridial duct behind the septum (fig. 36).

The ciliated organ opens here into the lumen of
the nephridial canal, the junction or grafting of the two
organs being marked by quite a sudden change in the character
of the tissues. Behind this point the nephridial duct widens
out, and affords an easy outlet to the spermatozoa, which in

1 The description applies to both sexes.

454 EDWIN S. GOODRICH.

this specimen can be actually seen to pass out by its means
(fig. 38).

In front of the communication between the two organs the
nephridial canal passes forwards to the lip of the genital funnel,
round which it runs inwards to the terminal nephridial organ
beset with solenocytes (figs. 37 and 81). Nephridium and
genital funnel are, therefore, really quite independent, except
at the point of junction behind.

In other specimens, unripe males and females, the ciliated
organ is much less developed, being represented by a scarcely
opened funnel, the epithelium of which is composed of rela-
tively small closely packed cells, often not yet provided with
cilia. Posteriorly this rudimentary organ abuts on, but does
not actually open into, the nephridial duct.

In the specimen of G. emerita, which shows no signs of
maturity, the funnel is even less developed, being completely
closed and entirely without cilia (figs. 33 and 34).

Owing to the unexpected discovery of nephridia with “ tube-
bearing cells” or solenocytes in the Phyllodocide, I have been
obliged to postpone the general summary and theoretical
conclusions of these researches to a third part, which I hope
to publish in the near future. I may so far anticipate these
conclusions as to point out that it has been shown in the fore-
going pages that the “ciliated organ” is the morphological
representative of the peritoneal or genital funnel of other
Annelids, and probably of the Ccelomata in general.

List oF REFERENCES.

1. Cuénor, L.—* Etudes physiologiques sur les Oligochétes,” ‘ Arch. Biol.,’
vol. xv, 1897.

2. Enters, h.—‘ Die Borstenwirmer,’ Leipzig, 1864-8.

8. GoopricH, HE. §.—“ On a New Organ in the Lycoridea,” ‘ Quart. Journ.
Micr. Sci.,’ vol. xxxiv, 1893.

4. Goopricn, EK. §.—‘* On the Nephridia of the Polycheta,”’ ‘ Quart. Journ.
Mier. Sci.,’ vol. xl, 1897.

ON THE NEPHRIDIA OF THE POLYCHATA. 455

EXPLANATION OF PLATES 32—35,

Illustrating Mr. Edwin 8. Goodrich’s paper ‘“‘On the
Nephridia of the Polychzta.”’

List of Reference Letters.

ap.t. Aperture for the tube of the solenocyte. 4. White amcboid cor-
puscle. ¢. White granular corpuscle. ch. Fragment of cheta. cil. org.
Ciliated organ. deg. nuclei. Degenerating nuclei. exer. gr. Excretory
granules. fl. Pointed flap of the ciliated organ. g7.c. Granular cell. hem.
Hemoglobinous celomic cell. izé. Intestine. x. Nucleus. xeph. Ne-
phridium. 7. Ridge. ¢.b.c. Tube-bearing cell or solenocyte.

PLATE 322.

Fic. 1.—Inner view of a portion of Glycera siphonostoma opened
along the mid-dorsal line and stretched out. The gut, part of the longitudinal
dorsal muscles on one side, and one bundle of chete, have been removed to
expose the nephridia.

Fic. 2.—View of the “nephridial complex” in two segments of a similarly
dissected Gl. unicornis.

Fic. 3.—Nephridial complex of GI]. siphonostoma dissected out. The
outer lip of the ciliated organ is torn off short. The contents of the ne-
phridial sac are seen by transparency. Fresh. Cam. x 100.

Fic. 4.—Edge of the nephridium of a Gl. siphonostoma, showing the
tube-bearing cells, or solenocytes, with small processes. [resh.

Fic. 5.—Three solenocytes broken off free, from a teased nephridium of Gl.
convolutus. Fresh. Cam.]. Oil im., oc. 3.

Fic. 6.—Corpuscles in the ccelomic fluid of Gl. unicornis. Fresh. Cam.
x 400.

Fic. 7.—Optical section of the edge of the nephridial sac, and overlying
nephridium. The “excretory mass” is composed of amoeboid cells, degene-
rating coelomic cells, and refuse; from a posterior nephridium of Gl. uni-
cornis. Fresh. Cam. x 400.

Fic. 8.—Transverse section of G]. unicornis. Cam. Z. aa, oc. 3.

Fic. 9.—A pair of solenocytes of Gl. unicornis. The flagella are seen by
transparency in the tube and underlying chamber. Fresh.

Fic. 10.—Another variety of the same cells, Fresh. Cam.1. Oil im., oe. 3.

456 EDWIN 8. GOODRICH.

Fie. 11.—Nephridial complex of an anterior segment of Gl. unicornis.
The nephridial chambers and canals are partially shown by transparency.
Fresh. Cam. |. 4, oc. 3.

Fie. 12.—Portion of the nephridium of Gl. convolutus near the origin of
the duct. Fresh. Cam. x 400.

Fic. 13.—Nephridial complex of G]. convolutus. Fresh. Cam. x 100.

PLATE 33.

Fie. 14.—Surface of the nephridium of Gl. convolutus. Fresh. L. Oil
im., oc. 3.

Fic. 15.—Diagrammatic reconstruction of a portion of the nephridium of
Gl. siphonostoma, cut so as to show the course of the canals, &c. The
inner surface would be applied to the wall of the nephridial sac.

Fies. 16 and 17.—Two sections through the same nephridial complex of
Gl. unicornis (longitudinal through the animal). Cam. x 140.

Fie. 18.—Optical section of the edge of the nephridial sac (near the cecum)
of Gl. unicornis. Perenyi, paracarmine. Cam.1. Oil im., oe. 3.

Fic. 19.—Section taken almost transversely near the extremity of the
cecum of the nephridial sac of Gl. unicornis, showing the secondary
chambers separated by thin partitions. Cam.]. Oil im., oc. 3.

Fic. 20.—Section along the cecum of a similar specimen. Cam. x 400.

Fic. 21.—Section across the nephridial complex of Gl. convolutus.
Cam.1. Oil im., oe. 3.

Fie. 22.—Two ameeboid hemoglobinous ceelomic cells of Gl. siphono-
stoma, fresh from a specimen eighteen hours after injection with Indian ink
and carmine. Cam. xX 500.

Fig. 23.—Three of the ‘granular cells” from the nephridial sac of Gl.
siphonostoma. Fresh. Cam. x 400.

PLATE 34.

Fic. 24.—Portion of a preserved nephridium of G]. unicornis seen from
the outer surface. The tube-bearing cells are not drawn ; the canals, &c., are
seen by transparency. Cam.1. Oil im., oc, Z. 4 ¢.

Fic. 25.—Portion of a section through the nephridial sac of G]. unicornis.
Cam. Z. D, oc. 3.

Fies. 26, 27, and 28.—Three longitudinal sections of Gl. unicornis taken
from within outwards, showing the nephridium, the nephridial sac, and the
ciliated organ, extending on to the body-wall. Cam. x 200.

Fic. 29.—Portion of the nephridium from a section similar to that drawn

ON THE NEPHRIDIA OF THE POLYCHATA. 457

in Fig. 26, showing the canals, chambers, and solenocytes. An irregular
cavity is left between the nephridium and the wall of the nephridial sac, cav.
Cam.

Fic. 30.—Diagrammatic reconstruction of the nephridial complex of Gl.
siphonostoma, showing the nephridial sac (represented as empty), the
ciliated organ, and the nephridium (yellow).

Fie, 31.—Diagrammatic reconstruction of the nephridium (yellow) and
genital funnel (ciliated organ) of a ripe Goniada maculata.

PLATE 35.

Fig. 32.—Portion of a section through the wall of the nephridial sac and
overlying nephridium of Gl. unicornis. Cam. x 1600.

Fic. 33.—Longitudinal section through Goniada emerita (median
region), showing the septa, nephridia, and undeveloped or rudimentary genital
funnels. Cam. x 100,

Fie. 34.—More enlarged view of a portion of a similar section. Cam.
x 400.

Fic. 35.—Section through the lobed inner extremity of the nephridium of
G. emerita, showing the solenocytes on the surface. Cam. x 400.

Fie. 36.—Longitudinal section (lower half) of Goniada maculata,
showing the genital funnels opening behind into the nephridial ducts. Cam.
x 100.

Fries. 37 and 38.—Sections from the same series taken farther outwards,
showing the course of the nephridium carrying the spermatozoa to the ex-
terior. Cam. x 100.

Fic. 39.—Section from the same series taken farther inwards than Fig. 36,
and showing the lip of the genital funnel and the inner extremity of the
nephridium. Cam. x 100.

Fic. 40.—Enlarged view of the lip of the genital funnel and the underlying
nephridial canal. Cam. x 400.

voL. 41, pant 3.—NeEW SERIES. Il

. ; = ‘i ; a *
7a. ray) ets

ge oP kee or ieee en Ausra

ave ve
a= P I at

ice . 7 ¥ - ; a
; ip ol Si hee eee ae ae eal Ba ®
- ae ie (og eguean Tee egal andra VS alt aA PS iy
ral : . ‘ < ne
el : 4 Ue tiper pa Os ge ahah een or
. " ‘ “ a eA) eS Liat ae iis wena “aw dD S
need 1. ee ( ey Mi toh ihe ' \ fis
be | T3\ % ! ‘ 1<tays) ee Se |
* re
a Gi weep 4 * faa nh aes .
a ’
Ps f
‘
j

MAR 11 1899

DIFFERENCES IN HISTOLOGICAL STRUCTURE OF TEETH, 459

On Differences in the Histological Structure of
Teeth occurring within a Single Family—the
Gadide.

By

Charles 8S. Tomes, M.A., F.R.S.

With Plate 36.

Some years ago, whilst engaged upon an investigation which
had a different object, I was struck by the fact that the teeth
of various members of the Gadide presented marked differences
in histological structure; but, partly from want of material
and partly from other causes, I have only lately examined a
sufficient number of the genera which are grouped together
in this family to enable me to draw any conclusions about
them.

It was, of course, only to be expected that fishes differing
so much in respect of food, habits, size, and external form as
the various Gadidz should have dentitions which in naked-eye
characters differ much from one another, and, as a matter of
fact, the family shows examples of adaptive modifications of
an interesting and very complete kind. And it is compara-
tively easy to see how these adaptive characters have arisen
by gradual slight modifications of a type which runs through
them all.

But it is not so easy to understand how the group of
influences known as natural selection should have operated in
the direction of producing differences of minute structure, for
it would seem as though it mattered little what the histological
structure of a tooth might be, as far as the exercise of its

vou. 41, PART 4.—NEW SER. K K

460 CHARLES §. TOMES.

functions goes, so long as it is sufficiently strong, sufficiently
sharp, and of an appropriate shape. Hence it appeared to bea
matter of interest to ascertain what the extent of this variation
in structure is, and to see how far these differences are found
to coincide with the lines of classification which have been
adopted on general grounds.

As a preliminary to this inquiry I sought to ascertain how
far the Gadide are to be regarded as a good natural family ; and
being perfectly unable to myself form any opinion upon this
point, I consulted Dr. Giinther and Mr. Boulenger, who will
be acknowledged to be in a position to speak with the highest
authority.

They tell me that the Gadide are a well-marked and natural
group, and therefore well adapted for the purpose of such an in-
vestigation ; to their kindness I am indebted for the opportunity
of examining several genera not otherwise accessible to me.

The teeth of the Gadide all consist of that modification of
dentine which is known as vasodentine proper, and, as there is
still a little confusion in the application of this term by various
authors, it seems desirable to very briefly define what the
meaning of the term really is; this I did some years ago at
greater length (1).

Retzius (3) was the first to describe the tissue accurately,
but he did not give to it a distinctive name. Owen (2), fol-
lowing in his footsteps, gave it the name of vasodentine, but
at the same time included certain other forms of dentine which
hardly come under the same category.

In 1878 (1) I endeavoured to carry further the existing
descriptions, and to point out that the term vasodentine should
be limited to those dentines in which, in the words of Retzius,
the larger canals “ formed, with others of tlie contiguous tubes,
large loop-shaped anastomoses, and the outer -extremities
entered also into closed anastomoses, almost like the more
minute blood-vessels in the villi of the abdominal canal.”

I also pointed out that in the fresh state these canals
contained capillary blood-vessels through which red blood
circulated, so that a typical vasodentine tooth in the living

DIFFERENCES IN HISTOLOGICAL STRUCTURE OF TEETH. 461

fish is quite red, and that, the canals being of just the
size of the blood-vessels, they contained practically nothing
else.

A typical vasodentine, therefore, consists of a matrix per-
meated by a rich plexus of blood-carrying channels, and contains
no dentinal tubes whatever of the ordinary kind; the matrix is
laminated, but is quite solid (4).

It would, however, be out of place to recapitulate here all the
details which are, for the most part, to be found elsewhere, and
it will suffice to describe those points only which are relevant
to the title of the paper.

The teeth of the ling (Molva) may be taken as typical of
that group in which the vascular system attains to its fullest
development ; they are long conical teeth, slightly curved and
very sharp, and have large axial pulp chambers which are
also elongated cones, and extend through the greater part of
the length of the teeth.

From these pulp chambers the vascular canals run outwards,
almost at right angles to the surface in the lower part of the
teeth, whilst near to their apices they run obliquely upwards;
they are disposed with the utmost regularity, and anastomose
with one another to some extent.

They do not reach to the surface of the dentine, but all
terminate in loops at exactly the same distance from the
surface, the loops in which they terminate being flattened, so
that the terminal canals lie parallel with the surface, and
there is an appearance of a bounding channel parallel to the
exterior of the tooth (figs. 1—3).

This disposition of the vascular canals is equally charac-
teristic in the hake (Merlucius) ; in both fish the outer solid
layer of the dentine shows a faint striation parallel with its
surface, which appears to be due to slight lamination.

The apex of the tooth is kept sharp bya little spear-point of
enamel, neatly fitted on to the dentine in such wise that it
does not much increase its size, although itself of material
thickness. This spear-point of enamel is possessed by all the
family, and does not show much distinct structure, though in

462 CHARLES 8S. TOMES.

favourable specimens fine lines may be seen to run through it
more or less at right angles to its surface.

The ling (Molva) and hake (Merlucius) are the two genera
in which the teeth are largest and most conspicuous ; they are
also the genera in which the development of the vascular net-
work is most complete. In the genus Gadus the arrangement
of the vascular canals is somewhat different, the loops are more
rounded, and the peripheral loops less flattened, so that there
is no appearance of a circumferential vessel such as was seen
in the ling and hake, and consequently the outer non-vascular
layer of the dentine is less sharply marked off (figs. 5 and 6).

In the common cod (G. morrhua) the teeth are fairly large
and are not very firmly fixed, having a certain degree of motion,
though nothing approaching to the very definite hinge (4,
p- 224) of Merlucius is to be found. The vascular network is
rich, but it takes the form of isolated loops to a great extent
(fig. 5); the lamination already spoken of exists, but is
perhaps rather less marked than in the ling or in the hake, and
there is the usual enamel tip to the teeth. The whiting pout
(Gadus luscus) has (fig. 6) a similar structure in its teeth, as
has also the poor cod (Gadus minutus), although in this latter
the canals are less abundant (fig. 7).

But in the remaining species of the genus Gadus the reduc-
tion goes much further; thus the pollack (Gadus pollachinus)
has dentine in which the vascular loops remain much more
distinct from one another, and there are larger intervals of
laminated matrix without any (fig. 8). In the haddock
(Gadus eglefinus) the whole upper part of the tooth is
often composed of finely laminated dentine without any vas-
cular canals, while in the lower portion they do occur as sparse
and isolated loops (fig. 9).

In the whiting (Gadus merlangus) the loops are present
in about the same proportion as in the pollack, while in the
coal-fish (Gadus virens) they are still more scanty, being
confined to a few loops here and there (fig. 10).

Thus of the whole genus Gadus it may be said that in none
do the vascular channels present the beautiful and complete

DIFFERENCES IN HISTOLOGICAL STRUCTURE OF TEETH. 463

network characteristic of the ling and hake, and that within
the limits of this single genus many stages in the reduction of
the vascularity are exemplified, although in none has it wholly
disappeared.

In the burbot (Lota vulgaris), the only fresh-water
representative of the family, there is an interesting arrange-
ment. This fish is generally regarded as closely allied to
the ling, of which it is sometimes spoken of as the fresh-
water representative. In it the vascular channels are few, and
the upper part of the tooth is generally free from them, but
where they do exist they present the flattened exteriors of the
loops. They thus recall the circumferential canals of the ling,
and differ from those teeth found in the genus Gadus, in which
the canals are reduced to a similar extent. So far, then,
as dentine structure goes, the relationship of Lota to the
ling appears to be somewhat confirmed (fig. 11). A similar
but richer disposition of vascular canals occurs in Brosmius
(fig. 3).

In Lotella, usually regarded as allied to the hakes, the vas-
cular canals have utterly disappeared, and the dentine presents
no structure save a well-marked lamination (fig. 12).

To sum up the differences which have been so far described,
they are of two kinds: the one the diminution in abundance
of the vascular canals; the other a slight difference in their
arrangement, consisting in the absence of the distinctly
bounding canal which is formed in the cases of the hake and
ling by great flattening of the exterior loops and their anasto-
moses with their neighbours. In each case, therefore, it is a
difference of degree rather than of kind; nevertheless no one
could possibly mistake the teeth of the hake or of the ling for
those of the genus Gadus ; and, as has been already pointed out,
the difference in the form of the loops is still perceptible,
even when they have become greatly reduced in number, as in
the burbot (Lota).

The teeth which I have examined are those of the following
genera :—Gadus, Merlucius, Molva, Lota, Lotella, Uraleptus,
Phycis, Motella, Raniceps, and Brosmius ; whilst of the genus

A464 CHARLES Ss. TOMES.

Gadus, the species wglefinus, luscus, minutus, tomcodus, mer-
langus, virens, and pollachius have been investigated.

These genera are usually arranged, with but little deviation
on part of the various writers, in the following order (5) :

1. Gadus.

2. Merlucius.
3. Lotella.
4. Uraleptus.
5. Phycis.
6. Lota.
7. Molva.
8. Motella.
9. Raniceps.
10. Brosmius.

Were we to attempt to classify them by their tooth structure
alone we should arrive at something like the following arrange-
ment :

2. Merlucius Highly developed systems of vascular canals, the outer

7. AON non-vascular portion being sharply marked off by the

10. Brosmius A : sibs ; :

6 Lot bounding canal: ” in Brosmius the canal system is
. Lota .

reduced somewhat, and in Lota it is disappearing.
? 4. Uraleptus i Bi eesrer

The vascular canals, abundant in G. morrhua, becoming

De HeHIPER Ss less frequent in other species, but always retaining
de Ee sae rounded external loops; in some species of Gadus
8. Motella they are sparse, and in Motella there are hardly any.

5. Phycis . . ( The vascular loops are so reduced that little can be said
? 4. Uraleptus as to their arrangement. In Lotella they are quite
3. Lotella absent.

It will be noticed that the teeth of Brosmius (fig. 3, the
torsk) partake pretty closely of the structural characters of
those of the ling and the hake, whereas in the accepted classi-
fication it does not come near to them ; and this is perhaps the
most conspicuous instance of incompatibility between the tooth
structure and the usually accepted affinities of the creature met
with in the family.

DIFFERENCES IN HISTOLOGICAL STRUCTURE OF TEETH. 460

For the dentines of Phycis and Motella (the rockling), figs.
15 and 16, which fish do not stand far apart from the ling
and hake, have the vascularity so far reduced that its pattern can
hardly be positively deciphered, though, so far as any evidence
does exist, it appears to resemble that of the genus Gadus more
closely than that of the ling.

It is, of course, fully recognised that no linear arrangement
of genera and species can ever truly represent relationships,
but still the classification by tooth structure departs somewhat
more widely from that adopted upon general grounds for the
deviation to be thus accounted for; whilst if we were to take
into account simply the quantity of vascular canals present we
should have still more discrepant results, for it would then
come about that some species of the genus Gadus would be
interspersed amongst other widely different genera, as is here
shown.

2. Merlucius )_.
7. Molva f (rich system of canals).
1. Gadus morrhua.
9. Raniceps.
. Brosmius.
. Gadus luscus.

|
=)

Gadus minutus,

. Gadus pollachius.

. Gadus eglefinus.

. Lota.

Gadus virens.

. Uraleptus.

. Phycis.

. Motella.

. Lotella (no vascular canals at all).

It is true that the genus Gadus has been sometimes divided
into several genera, but even in that case they are very closely
allied genera, and cannot possibly be regarded as interspersed
amongst other genera with any propriety, so that the conclusion
is irresistible that these differences of tooth structure only to a
limited extent follow the lines of the general affinities of the

OO RH Oe eee

(Ss)

466 CHARLES S. TOMES.

animals, and the question next presents itself, what limes do
they follow ?

The first thing that would suggest itself is that as the denti-
tions become less pronounced, and the individual teeth smaller,
so does their intimate structure become simplified.

Some facts may be adduced in support of this view; thus
the most elaborate tooth structure is found in the two genera
which have the largest teeth, namely, the hake and the ling
(figs. 1 and 2); while in the Lota (fig. 11), which is regarded as
allied to the ling, the teeth are both much smaller and simplified
in structure. Similarly, Gadus minutus, which has teeth
large for the size of the fish, has a richer vascular network than
many of the genus possess. But if this be a truth, it is only
a partial truth, for, at all events in the same individual, the
size of the tooth does not determine its structure.

For the hake is peculiar in having comparatively long and
moveable gill-rakers upon its first gill arch, and these gill
rakers are beset with tiny teeth; nevertheless these minute
teeth are exactly, so far as space will allow, of the pattern of
the large teeth which are twenty times their size, so that here
at least reduction in size has not been accompanied by any
simplification of structure (fig. 2*). Again Uraleptus (fig. 13)
has teeth of quite considerable size, but they are almost devoid
of vascularity, while teeth from its branchial arches exactly
agree in structure with its large teeth (fig. 14).

A similar inference may be drawn from the teeth of the
Cheetodonts (4, p. 249) in which the teeth, although attenu-
ated to an extreme degree, yet retain a full supply of
vascular canals whenever there is room for them. Hence
there seems to be some degree of fixity in the vascular pattern
of dentine, and it does not seem to be quite lightly abandoned,
as might at first be inferred from the fact that within the
single genus Gadus so many grades of vascularity are met with.
It seems possible that the fact that a large tooth like that of
Uraleptus should present that absence of vascularity which
pertains to the reduced dentitions, might be accounted for by
the supposition that the ancestors of this genus had suffered

DIFFERENCES IN HISTOLOGICAL STRUCTURE OF TEETH. 467

great reduction in their teeth, and that a more powerful den-
tition had again been evolved, without any recovery of the
minute structure which belongs to the larger teeth of those
which have never lost strong dentitions. The teeth round the
margins of the mouth in Uraleptus are villiform, with large ones
interspersed. However, these and similar speculations may be
more safely indulged in by those who have a more intimate
acquaintance with these fishes than I can pretend to, and it
will suffice for me to merely point out the facts.

It may, perhaps, seem that in this communication a very
small point has been laboured, but it appears to me that such
facts as these are worthy of being recorded, for, so far as I
know, very little has been done in the way of investigating the
existence or non-existence of structural change in tissues under
the influence of evolution. And the existence of these struc-
tural differences seems to me to have some bearings of a wide
nature.

The teeth of the Gadide appear to furnish an argument
against the adequacy of the purely mechanical theory of the
evolution of tooth forms, so warmly advocated by Cope under the
name of kinetogenesis, and adopted in its entirety by a large
number of the American school of naturalists. For if the form
of a tooth is the direct consequence of the direction of stimu-
lation that it has received by use in successive generations,
then a tooth which is subject to the very minimum of use, such
as that of the gill-rakers of the hake, ought not to be so exact
a copy of the teeth round the margins of the mouth. And if
very considerable use be essential to the maintenance of elabo-
rate structure, then we might expect, on the one hand, that
the teeth in the gill-rakers of the hake should be of very
simple structure, which is not the case, and, on the other hand,
that the large teeth of Uraleptus, which must be held to be
important in function, and so to receive the stimulus of use,
should not have lost the structure typical for the family whilst
retaining the size, and more, indeed, than the average size.
Hence mechanical theories do not suffice to account for
the structural degradation. For such it must be termed

468 CHARLES 8S. TOMES.

when we get a dentine which neither presents a vascular canal
system nor ordinary dentinal tubes, structures which a wider
investigation and the observation of cases in which both co-exist,
as in the teeth of Pleuronectide, seem to stand in a comple-
mentary relation to one another.

To sum up the results of these observations, true dentinal
tubes are not met with in the Gadide. The spear-point enamel
tip exists in all of the family, even upon the smallest teeth.
Those members of the family which have the largest teeth,
either fixed to the jaws by anchylosis (ling, some of the teeth
of the hake) or by a highly elaborated hinge (the hinge-teeth
of the hake), have the vascular canal system in its highest deve-
lopment. Those which have the teeth small in relation to the
size of the animal, as happens ina large number of Gadide, and
those in which the teeth are not very firmly attached, show also
a simplification of the minute structure of the teeth. But mere
smallness of size does not necessarily involve simplification of
structure, as is exemplified by the tiny teeth of the gill-rakers
of the hake. Nor does large size necessarily involve elabora-
tion of structure, as is instanced by the large teeth of Uraleptus.
On the other hand, large and small teeth in the same creature
present identical structure ; and whilst the differences in tooth
structure in some instances follow the lines of the accepted
classification of the genera, in others they do not.

BiIBLioGRAPHY.
1. Tomrs, C. S—*On the Structure and Development of Vascular Den-
tine,” § Phil. Trans.,’ 1878.
2. Owen, R.—‘ Odontography,’ 1840—1845.
3. Rerzius, as quoted by ‘ Nasmyth on the Teeth,’ 1835.
4, Tomzs, C. S.—‘ Dental Anatomy,’ fifth edition, 1898.
5. Ginter, A.—‘ The Study of Fishes,’ 1880.

DIFFERENCES IN HISTOLOGICAL STRUCTURE OF TERTH. 469

DESCRIPTION OF PLATE 36,

Illustrating Mr, Charles S. Tomes’ paper “On Differences in
the Histological Structure of Teeth occurring within a
Single Family—the Gadide.”

Fic. 1.—Apex of tooth of a ling (Molva) with its enamel cap. xX 40.
Fie. 2.—Apex of tooth of a hake (Merlucius). x 40.
(Figs. 1 and 2 comprise about one fourth of the entire length of the teeth.)

Fie. 2*,—Portion of a moveable gill-raker from the first branchial arch of
ahake. x 100.

Fic. 3.—Hntire tooth of Brosmius. x 30.

Fic. 4.—Nearly entire tooth of Raniceps. x 50.

Fie. 5.—Portion of dentine of cod (G. morrhua). x 90.
Fic. 6.—Tooth of Gadus luscus. x 50.

Fic. 7.—Tooth of Gadus minutus. x 50.

Fie. 8.—Tooth of pollack (G. pollachius). x 50.

Fie. 9.—Tooth of haddock (G. eglefinus). x 60.

Fie. 10.—Tooth of coal-fish (G. virens). x 60.

Fig. 11.—Tooth of burbot (Lota vulgaris). x 80.
Fic. 12.—Tooth of Lotella. x 45.

Fie. 13.—Portion of tooth of Uraleptus. x 50.

Fig. 14.—Teeth from branchial arch of Uraleptus. x 80.
Fic. 15.—Tooth of Phycis. x 80.

Fie. 16.—Tooth of Motella. x 50,

: lll Do Soe, Pe eee vee
7 : . ris ee e @ a ~ a -
av b - « = =a > 7

Lise pc ne ee . oe on Re
am tse Lee
Y i A ie! ee Sa at aed : _
_ ac a =». a 7 ra “7 -_
Cte bale Oy ter ae ee

7 oe x - ” : i sd J -
= oe rh é "5 a a
9 le 4 ¥e 7 AR - o.
i eee ohio Cae ees

- t j -
* Baie ~~ :
oe x Ae % ibe” wi ee * te

So (Conia AeA wen tpa eat

i. ° LP RA ;
aa : ft Te eet MT. a ed tee hae aly ‘
id ¢ j ettun sth tee 4 Janae i a ae {capone a mae |

. va Baek ofl Ge vile ee aA

i
L { Li bo tet
i (
¢ ’
{ 1 ‘ LiTAa
a
\ ‘ >
» ; for |
ac
: MIA 4 fer ill
, in
i
‘y
>
s ‘
id ”
'
Z -
i
4
Py i:
i { ;
.
‘
|
- 1
ra
.
’
4
.
‘
‘ « ‘ ;
1
¢

TWO NEW SPONGILLZ FROM LAKE TANGANYIKA.

471

A Description of Two New Species of Spongilla

from Lake Tanganyika.
By

Richard Evans, B.A.

With Plates 37 and 38.

ContTENTS.

I. Introduction
II. Description piaoneallas moorei ;
(1) Habits of growth and external form.
(2) Skeleton 9 5
(a) Spicules

(z) Arrangement of Rpienles to fait fires &e.

(c) Spongin
(3) Canal system
(4) Gemmules
Ill. Affinities of Spongilla moorei :
IV. Description of Spongilla Sas ges :
(1) Skeleton :
(A) Spicules ;
(s) Arrangement of apieules Se...
(c) Spongin
(2) Gemmules
V. Affinities of Spongilla fan pangikee

I. INTRODUCTION.

PAGE
471
472
472
473
473
475
476
477
478
478
481
481
481
481
482
482
482

Tur two species of Spongilla described in this paper were
collected by Mr. J. E. 8S. Moore, of the Royal College of
Science, during his visit to Lake Tanganyika, in the summer of

1896, and were taken at a depth of 350 fathoms.

Mr. Moore entrusted the material in question to Mr. E. A.

4.72 RIGHARD EVANS.

Minchin, Fellow of Merton College, Oxford, who in turn
handed it over to me that I might study it. There were,
besides four specimens preserved in alcohol and kept in
separate hottles, a number of fragments which had been most
carefully preserved, some in corrosive sublimate, and the rest
in Flemming’s fluid, and which were kept apart in separate
tubes.

On examination all the material, with the exception of one
small piece of that which had been preserved in corrosive sub-
limate, proved to belong to the same species. The small piece
mentioned above must have been cut from a sponge which in
its external appearance was almost exactly similar to the one
to which the bulk of the material belonged, however different
it may be in its skeletal characters, or else Mr. Moore would
have noticed the difference. However great the external
similarity, a single glance at a section suffices to distinguish
them as belonging to two entirely different species.

I have given these species the names Spongilla moorei and
Spongilla tanganyike respectively. The former species
is represented by the bulk of the material, and is named in
honour of its discoverer.

Of the latter I had but a small fragment, and have chosen
its designation from its locality.

II. Description or SPONGILLA MOOREL.

(1) Habits of Growth and External Form.—Spon-
gilla moorei grows on shells of various mollusca, and par-
tially covers them as a crust. The upper surface is raised into
lobes or mound-like elevations, which in no case are more than
half an inch above the general surface, and which are usually
no more than an eighth of an inch above the shallow depres-
sions which separate them. The surface texture of the pre-
served sponge is somewhat woolly in appearance, though this
is probably the result of the broken condition of the dermal
membrane, for it has been observed that some of the fragments
preserved in Flemming’s fluid are smooth, and the spicules of

TWO NEW SPONGILLA FROM LAKE TANGANYIKA. 473

the skeleton, though supporting the dermal membrane, do not
in the natural condition penetrate it.

An osculum is situated at the tip of each of the lobes or
mound-like elevations of the surface of the sponge. This
opening measures about an eighth of an inch in diameter, and
underlying it there is a fairly large gastral cavity. The
dermal pores are small, as usual, and are situated on the flanks
of the lobes as well as in the intermediate depressions.

(2) The Skeleton.—In treating of the skeleton or the
supporting part of the sponge, first, the spicules will be de-
scribed; secondly, the arrangement of the spicules to form
fibres, and of the fibres at large to form the skeleton; and
thirdly, the spongin which binds the fibres together.

(a) The Spicules.—In order to facilitate description the
spicules will be divided into three classes, the ordinary division
into ‘“‘ megascleres ” and “ microscleres”’ being intentionally
avoided, because it is—to say the least—doubtful whether the
small smooth spicules are microscleres or young megascleres.

The three classes of spicules are—

(a) Diactinal monaxons which taper to a sharp point either
gradually (amphioxea), or more rapidly (amphitornota), and
are without swellings on their shaft. The former are always
straight, the latter curved (Pl. 37, fig. 3).

(3) Similar straight amphioxea or curved amphitornota with
distinct swellings on the shaft (Pl. 37, fig. 4).

(y) Irregular systems formed by the fusion of spicules
belonging to class a (Pl. 37, fig. 2).

(a) The straight amphioxea taper gradually into a sharp-
pointed end (Pl. 37, fig. 3, a and 6), while the curved am-
phitornota, which are far more numerous, taper much more
abruptly into a similar point (Pl. 37, fig. 8, c—e). Both the
straight amphioxea and the curved amphitornota are highly
variable in thickness, and exhibit all stages of development.
The axial thread is of even thickness throughout its whole
length in all these spicules.

(8) In addition to being slightly more slender than the

Cf. Schulze and Lendenfeld’s nomenclature (10).

474 RICHARD EVANS.

spicules already described, the main feature of these spicules
is the presence of a number of swellings, which varies from
one to five. As a rule they are situated symmetrically with
regard to the middle point of the spicule; that is, if there is
only one swelling, it is situated at that point, but if there are
two they are placed one on each side of that point, and at equal
distance from it ; and similarly the symmetry is retained when
there are three, four, or five swellings. The absence of the
symmetrical arrangement as seen in fig. 4, d, is very exceptional.
The axial thread, in contrast to spicules of class a, present a
dilatation corresponding to each swelling on the spicule.

(vy) The spicules in this class are of variable and irregular
form, since the individual amphioxea or amphitornota which
form them may fuse at any point, and at any angle (Pl. 37,
fig. 2, a—/, especially fig. 2,d). As arule, these compound
systems are formed from spicules of class a, though occa-
sionally a spicule of class (3 is found to take part in their
formation.

With regard to their origin, two suppositions are possible ;
first, that they are the result of irregular growth, and branching
of a single spicule derived entirely from a single scleroblast ;
secondly, that they arise by fusion of spicules primitively
distinct, and formed each by its own scleroblast. Fig. 2, e—y,
might be taken as evidence for the former view, but such
forms as that represented in fig. 2, d, render such a supposition
highly improbable, to say the least. The view that these
spicular systems are of compound origin receives strong
support from the way in which their axial threads cross one
another instead of branching. If these irregularities arose as
outgrowths from one spicule formed in one mother cell, it
might well be expected that their axial threads should be also
formed as outgrowths from that of the main spicule, and would
therefore be continuous throughout the system, but this is
certainly not the case in many spicules of our Spongilla, as can
be seen from the figures. In another sponge, which is probably
a monaxonid of the family Axinellide, viz. Tricentrium
muricatum (Pallas, 1756), Ehlers, 1870 (=Plectronella

TWO NEW SPONGILLZ FROM LAKE TANGANYIKA. A475

papillosa, Sollas, 1879) [11], there are branched spicules in
which the axial threads are continuous throughout, a fact which
may indicate that the spicules themselves owe their form to
branching. It seems clear, therefore, that the irregular
spicules of Spongilla moorei have in many cases been pro-
duced by fusion. Judgment must be suspended for the
present with regard to those systems in which no discontinuity
can be detected in the axial threads of the component spicule
rays; such spicules may be simply branched. The question
cannot be decided until the actual origin of the spicules has
been studied, and the same may be said for Tricentrium.
Since now it has been shown that the triradiates and quad-
radiates of the Ascons are formed by fusion, there is no
inherent improbability in a similar process occurring in other
cases [8].

Spicules of a similar character to the compound systems here
described have been figured by many authors in various Spon-
gillide (Spongilla aspinosa, Potts [9]; Lubomirskia
intermedia, Dybowski [4]). All these authors regard them as
abnormalities, but in moorei they are so frequent that they
must be considered as a normal feature of the species. It is
possible that in other Spongillide these systems have not
received the attention they deserve.

In addition to the spicules described above, there are small
masses of silica in Spongilla moorei, comparable with those
found in Spongilla aspinosa (Pl. 37, fig. 5).

(B) The Arrangement of the Spicules to form
Fibres, &c.—The spicules which form the _ polyspiculous
fibres belong mainly to the first and third classes above
described. Spicules of the first class form the greater part of
the fibres, while others lie about in the sponge tissue,
presenting for the most part an irregular method of arrange-
ment, though many such spicules are placed so as to bridge
over the spaces between the fibres in a perfectly definite way.
Spicules of the second class, which are far less numerous than
those of the first, seldom participate in the formation of the
fibres, but, as a rule, lie scattered irregularly between the

vou. 41, pARY 4,—NEW SURIES. LL

4:76 RICHARD EVANS.

fibres. The spicular systems of the third class are seldom
found in any other position than in the fibres.

As a rule, the spicules are arranged in the fibres with their
axes parallel to one another, and in the deeper parts of the
sponge the connecting spicules are rather numerous, and
more strongly developed than in the more superficial parts.
The connecting spicules are usually the most strongly developed
spicules in the whole sponge as regards size, differing, how-
ever, only in thickness from the smooth curved amphitornota
which constitute the fibres (Pl. 37, fig. 3, d—g). Speaking
generally, the largest spicules of the first class, together with a
few of the second and all the third, form the fibres and the
connecting links between them; while the smaller spicules
belonging to the first, and nearly all those belonging to the
second class, are scattered about irregularly in the meshes
between the fibres. The smallest spicules of all appear to be
absolutely independent of the skeletal meshwork, and this is the
strongest argument that can be adduced in favour of the view
that they are microscleres, and not young megascleres.

The arrangement of the skeleton at large is reticulate. The
most prominent feature of the general conformation of the
fibres is the way they pass from the surface of fixation of the
sponge to the dermal membrane which they support. Along
their course from one surface to the other they present a
wavy appearance, often dividing and again reuniting, ap-
proaching the dermal membrane nearly always at right angles,
and in many cases expanding into a kind of brush-like struc-
ture which supports that membrane (Pl. 38, fig. 6). In some
of the largest lobes of the sponge the fibres nearest the centre
pursue a straight course, while those furthest from that
position curve outwards, so as to form supports to the dermal
membrane which covers the flanks of these mound-like
elevations. Owing to this arrangement a longitudinal radial
section of one of these lobes presents an almost fan-like
appearance as regards the skeletal fibres.

(c) The Spongin.—All the skeletal fibres of this sponge
are enclosed in a distinct sheath of spongin, which is greatly

TWO NEW SPONGILLZ FROM LAKE TANGANYIKA. 477

thickened at the points where the connecting spicules occur,
these being either partially or completely surrounded by it
(Pl. 38, figs. 6 and 7). Not only are the fibres and the con-
necting spicules enclosed in a sheath of spongin, but the
surface of the sponge is covered by a thin layer or cuticle of
the same substance, which dips down between the cells of the
dermal membrane, and communicates with that which envelops
the fibres (Pl. 38, fig. 6).

(8) The Canal System.—Owing to the fact that the
material which had been preserved for histological study of the
sponge had been shaken considerably in moving from place to
place, a great number of cells had apparently become loose,
and were found lying in the spaces of the canal system, In
consequence it was impossible to make a complete and thorough
study of that system, though individual cells were in many
places nicely preserved ; nor is Spongilla, for other reasons, a
favourable object for the study of the canals in the Monaxonida.

The canal system in Spongilla moorei belongs to the
type usually described as the third. The dermal pores, which
are situated on the flanks of the mound-like elevations of the
surface and in the intermediate depressions, are small, and open
into the subdermal cavity, which is lined by flattened epithelium,
and considerably reduced by the passing through of the skeletal
fibres, which are enclosed in a sheath of spongin, which is
covered by cells of the epithelial layer.

The inhalant canals which pass from the subdermal cavity
into the chambers are narrow and difficult to make out. In
some cases these canals are short, owing to the flagellated
chambers being situated close to the floor of the subdermal
cavity. Those canals which pass into the chambers which are
situated more deeply in the sponge are long and narrow, follow-
ing a winding course, and keeping nearly always between the
chambers and the fibres of the skeleton. On their way down
into the deeper parts of the sponge they give off branches which
open into the chambers by way of prosopyles, which are so small
that it is almost impossible to make them out. The apopyle
was easily distinguishable as a wide opening, communicating

478 RICHARD EVANS.

directly with the wide exhalant canals, and occupying nearly
a fourth of the surface of the otherwise almost spherical
flagellated chamber, which is lined by collar-cells with nuclei
situated at their bases. The canals of the exhalant system
are much wider than those of the inhalant, and, as a rule, occupy
a central position between the fibres. As they pass down into
the deeper parts of the sponge they converge and unite together,
forming wider canals, which are few in number, and which
open into the somewhat spacious gastral cavity, which commu-
nicates with the exterior by way of an osculum situated at the
summit of each of the mound-like elevations of the surface.
(4) The Gemmule.—The gemmules, which are few in
number and scattered about singly, are spherical in shape and
small in size, measuring only *35 mm. in diameter. They
possess a thin coat, which is not surrounded by spicules
specially characteristic of the gemmule, but by the ordinary
skeleton spicules. Their cellular contents present the same
characters as do those of the common species of Spongilla, and
each individual cell is full of the two kinds of granules which
are quite characteristic of the cells of Spongillid gemmules. It
is just possible that had the material been preserved later in
the year the gemmules would have been more numerous, though
there would appear to be no absolute necessity for the produc-
tion of gemmules, since the sponge lives at a depth of 300
fathoms, and cannot possibly be either dried or frozen.

Ill. Tur AFFINITIES OF SPONGILLA MOOREI.

The presence of the gemmule is the most important character
tending to fix the position of Spongilla moorei among the
Spongillidee. Gemmules have been described in marine sponges,
and this fact diminishes the importance of the existence of gem-
mules in a newly discovered sponge as a character supposed to be
distinctive of the Spongillidz (Topsent [12]). It appears that
there is no special feature in the structure of the skeleton of
Spongilla moorei that would cause it to be separated from
the Chalinidz had it been a marine sponge. It most decidedly

TWO NEW SPONGILLZ FROM LAKE TANGANYIKA. 479

possesses more spongin than the Spongillide are usually sup-
posed to have. As a matter of fact, it is difficult to make out
what structural reasons there are for retaining the family
Spongillide. It is not at all improbable that when they are
more carefully studied they will be distributed among the
several genera of the Homorrhaphide. But as our knowledge
has not yet attained a stage which will enable us to do this, it
is deemed advisable for the present to place this new species
among the Spongillidz, and to retain that assemblage of sponges
as a family, however artificial it may be.

The characters of the gemmule of Spongilla moorei place
this species among the sub-family Spongillinz, and not among
the Meyinine or the Lubomirskinz. They lack the amphidises
which surround the gemmule of the Meyeninz, while, on the
other hand, the Lubomirskine is a sub-family which has been
created for the purpose of including certain fresh-water sponges
in which, up to the present, the gemmule has not been discovered.

Spongilla moorei appears to be more closely related to
Spongilla aspinosa (Potts) than to any other species of
the Spongilline. Both species agree in possessing spicules
which are smooth, straight or curved, and for the most part
rather abruptly pointed. Malformed spicules, as they are
described by Potts [9], are found in both, but they appear to be
more numerous and more complicated in Spongilla moorei
than in Spongilla aspinosa. Further, both species pro-
duce gemmules which are small in size, spherical in shape, and
supplied with a thin crust which is not protected by spicules
characteristic of the gemmule, but by the ordinary skeleton
spicules. Though the gemmules are few in Spongilla aspi-
nosa, they are more numerous than in Spongilla moorei,
a feature which may be explained either by the lesser import-
ance and consequent scarcity of the gemmule in the latter
species, or simply by the season at which the material was
collected.

Spongilla aspinosa differs from Spongilla moorei in
that it possesses small flesh spicules, which lie on the dermal
membrane and among the smooth, slender skeleton spicules.

480 RICHARD EVANS.

These small spicules are not found in Spongilla moorei,
unless they are represented by those drawn in PI. 37, fig. 3,
j—I, and fig. 4, j—n, which is probably the case. However,
it must be admitted, as has been done by Potts, that in both
cases these small spicules may be young megascleres, and not
microscleres. The only distinction obtaining between mega-
scleres and microscleres, viz. that the former are bound up
in the general skeleton of the sponge while the latter lie
scattered about freely, is a functional rather than a morpho-
logical character, and seems to break down in the Spongillide,
whose Homorrhaphid ancestors were probably without micro-
scleres. The consequence of this is the impossibility of
deciding definitely the true character of certain spicules. It
seems, however, a safe conclusion that these small spicules
are the same in Spongilla moorei as in Spongilla aspi-
nosa, though in the former they are not found in the dermal
membrane, their place being taken by the cuticular layer of
spongin which covers the surface.

The form of growth of these two species appears to differ.
Spongilla aspinosa is provided with long, slender, cylin-
drical branches which occasionally subdivide. These branches
grow from a thick basal membrane. Spongilla aspinosa,
however, at times forms merely a sheet which envelops the
support on which it grows, while Spongilla moorei in all
the specimens examined presented this appearance.

The spongin has not been described in Spongilla aspi-
nosa, and therefore neither comparison nor contrast is
possible.

The colour of Spongilla aspinosa is said to be green,
a fact which is the result of the position in which it grows,
for Spongilla lacustris and Ephydatia milleri and
fluviatilis may be either green or pale, according as they
grow in direct sunlight or in the shade. Owing to the depth
at which Spongilla moorei lives, the green colour of Spon -
gilla aspinosa would appear to be wanting.

TWO NEW SPONGILLE FROM LAKE TANGANYIKA. 48]

IV. Description oF SPONGILLA TANGANYIKZ.

Owing to the fact that there was but a small piece of this
sponge among the material presented to me for investigation,

is impossible to make any statement with regard to its
external form and habits of growth. However, it may be
conjectured that in both respects it must have resembled
Spongilla moorei, since Mr. Moore failed to detect it as
a distinct form. Though the two species are probably similar
to one another in their habits of growth and external characters,
they are strikingly dissimilar in the characters of their indivi-
dual spicules, though the general arrangement of the spicules
in the fibres and of the fibres at large is strikingly alike.

The description of this species must, of necessity, be brief.
The same plan will be followed as far as possible as in the case
of the description given of Spongilla moorei.

(1) The skeleton will be described under the following
heading :

(a) Spicules.
(s) Arrangement of spicules, Xc.
(c) Spongin.

(a) Spicules.—It may be safely stated that there are
megascleres and microscleres in this sponge. The megascleres
consist of amphistrongyla and amphitornota, which are for
the most part thickly covered with small spines. In addition
to these there are a few smooth or sparsely spined amphioxea
(Pl. 38, fig. 10, a, f, g, AR—Il, and o—gq). A few irregularly
shaped spicules, which appear to be the result of fusion,
are present (Pl. 38, fig. 10, m). The microscleres are much
slenderer than the megascleres, though they almost equal
them in length. They are always smooth and slightly curved
(Pl. 38, fig. 10, n).

(8) The General Arrangement.—The arrangement of the
spicules does not differ materially from that already described
in Spongilla moorei. The spiny amphistrongyla and
amphitornota, together with a few smooth or sparsely spined
amphioxea, are arranged with their axes parallel to one another

482 RICHARD EVANS.

to form the skeletal fibres. These divide and again reunite,
producing an arrangement which is usually described as being
reticulate. The fibres are connected together in many places
by spicules which bridge over the intermediate spaces. These
spicules are the largest in the whole sponge, as a rule, as was
found to be the case in Spongilla moorei. In addition
to these there are many spicules, both spiny and smooth, which
appear to lie about more or less freely in the tissues. The
slender microscleres are nowhere connected with the fibres, but
lie absolutely free in the tissues.

(c) The Spongin.—The spongin is not so highly deve-
loped in Spongilla tanganyike as in Spongilla moorei.
The former, therefore, in this respect resembles more closely
the ordinary species of the Spongillide than the latter appears
to do. The spongin does not appear to extend to the surface,
and the layer which covers the fibres is correspondingly thin.
The greater development of spongin occurs at the points where
the fibres branch or reunite, and at the places where the con-
necting spicules penetrate the fibres.

2. The Gem mule.—Though there was but a small piece of
this sponge it happened to contain several gemmules. These
are devoid of special spicules, but are surrounded by the
ordinary skeletal spicules and the microscleres. They possess
a thin coat as in Spongilla moorei, and are spherical and of
small size. As regards their cellular contents they present
the ordinary characters of the Spongillid gemmule.

V. Tue AFFINITIES OF SPONGILLA TANGANYIKA.

This subject must be considered from two aspects. In the
first place, the characters of the gemmule must be taken into
consideration, since the grouping of the Spongillide into the
three sub-families, Spongilline, Meyenine, and Lubomirskine,
and the division of the sub-families into genera, usually adopted,
depends on these characters. In the second place, the spicules
are of great importance as presenting a close resemblance to the
spicules of Lubomirskia intermedia var. a(Dybowski, ef. pl.
iv, fig. 3, 6, [4]), which belongs to the sub-family Lubomirskine.

TWO NEW SPONGILLZ FROM LAKE TANGANYIKA. 483

(4) The Gemmule.—The gemmule of Spongilla tangan-
yik lacks the amphidiscs which surround the gemmule of the
Meyenine. It therefore appears that this species cannot belong
to that sub-family. But it equally lacks the small spicules
which are usually found in close relation with the gemmule of
the Spongilline. Potts, however, places Spongilla aspinosa
among the Spongillinz, in spite of the fact that its gemmules
lack characteristic spicules. If this arrangement be followed,
the absence of such spicules from Spongilla moorei and
Spongilla tanganyike should not be considered as a barrier
against including these species among the Spongilline. But the
inclusion does away with the importance of the presence of
special gemmule spicules as a sub-family character.

The thin coat of the gemmule resembles that found in
Spongilla moorei, Spongilla aspinosa, and others of the
Spongilline, and has no similarity to the thick coat of the
gemmule of the Meyenine. The characters of the gemmule,
therefore, as far as they go, point to this new African species
discovered by Mr. Moore being one of the Spongillinz.

(3) The Spicules.—lIt is generally stated that the skeletal
spicules of the several species of the Spongillidz have no charac-
ters of higher than specific value. It is difficult to make out from
the literature of the family how far such a statement is justified.
However, the spicules of Spongilla tanganyikz possess
such characters that it is almost impossible to believe that they
have not a wider application. This sponge, considered from the
point of view of the skeleton, seems to present a certain amount
of affinity with a few species of the Spongilline on the one
hand, and of the Lubomirskinz on the other.

The megascleres of the greater number of species arrayed
under the sub-family Spongilline are sharp-pointed,—that is,
they are either amphioxea or amphitornota. There are, how-
ever, a few species which possess spicules with rounded ends,
that is, amphistrongyla. The species in question are Spon-
gilla nitens (Carter [8]), Spongilla béhmii (Hilgendorf
[5], and Spongilla loricata (Weltner [13]), to which may
be added Spongilla tanganyike, now described for the first

484, RICHARD EVANS.

time. Spongilla tanganyike, therefore, seems to be more
closely related to these species, so far as the characters of the
skeleton are concerned, than to any other species of the
Spongilline. Of the three species named above, it appears to
present closer affinity with Spongilla béhmii than with
either of the other two, for in Spongilla nitens and in
Spongilla loricata the amphistrongyla are smooth, while
in both Spongilla béhmii and Spongilla tanganyike
they are spiny. In the former the spines are more thickly
set at the end, which is a special feature of the megascleres of
some species of the Lubomirskine, and which may point to a
certain amount of affinity in that direction, while in the latter
they are evenly distributed over the whole spicule. In Spon-
gilla béhmii the megascleres are curved as in Spongilla
nitens, Spongilla loricata, and most of the Lubomirskine,
while in Spongilla tanganyike they are straight. How-
ever, there is among the Lubomirskine a variety of Lubo-
mirskia intermedia, described by Dybowski as var. a, in
which the spicules are spiny and almost straight. The spines
are evenly distributed, and the ends of the spicules in many
cases present the amphistrongylote character. Another feature
of Lubomirskia intermedia agreeing with Spongilla
tanganyike is that the microscleres are smooth, and almost
equal the megascleres in length. In Lubomirskia baci-
lifera and Lubomirskia papyracea the spicules are Amphi-
strongylote, though in the former the spines are arranged at
the ends of the spicules, in contrast with those of Spongilla
tanganyike, but to a certain extent agreeing with those of
Spongilla bohmii, while in the latter the spines are evenly
distributed over the shaft of the spicule, in contrast with those
of Spongilla béhmii, but similar to those of Spongilla
tanganyike.

From these points of comparison it seems that Spongilla
tanganyike as well as Spongilla béhmii must be closely
related to the Lubomirskinz. Had it not been for the presence
of the gemmule in the small piece of Spongilla tanganyike
at my disposal, I would certainly have placed it among the

TWO NEW SPONGILL® FROM LAKE TANGANYIKA. 485

Lubomirskinez. On the other hand, were gemmules to be
found in any species of the Lubomirskine, it would have to be
removed from that sub-family as at present defined. Conse-
quently I venture to suggest that the sub-family Lubomirskinz
should be abolished, and the species contained in it placed under
the Spongillinz, which then could be rearranged into a number
of genera according to the characters of the megascleres.

APPENDIX ON SOME SPONGE SPICULES FOUND IN THE
Mup or Lake TANGANYIKA.

Along with the sponge material which Mr. Moore entrusted
to Mr. Minchin, there was a microscopical slide with some of
the mud of Lake Tanganyika mounted on it. There were on
the slide, among other things, some sponge spicules which in
shape resemble those of the genera Uruguaya (Carter [3]) and
Potamolepis (Marshall [7]).. They vary from *14 to’31 mm. in
length, and from ‘013 to ‘05 in breadth. A great number of
intermediate stages between the two extremes are present.
Some of the spicules seem to be “‘ micropunctate.”” They are
nearly always curved, though the amount of curvature varies
considerably. The smallest spicules are of even thickness
throughout, being amphistrongylote. The spicules of inter-
mediate size in some cases present the same form as the small
ones, but differ in other cases in that they are slightly swollen
at the ends. The largest spicules are in all cases club-shaped.
In passing from the smallest spicules to the largest there seems
to be a gradual change from amphistrongyla to amphitylota
(Pl. 38, fig. 11).

It is difficult even to suggest what these spicules are. From
their characters they might well be the spicules of a species
belonging to the genus Uruguaya. But as the species of this
genus belong to the New World, and those of Potamolepis to
the Old, their locality seems to favour the view that they are
the spicules of some Potamolepid sponge. The variation in
size increases our difficulty, for it is impossible to decide
whether the smallest forms are young spicules, or a different
class of spicules, belonging either to the same or to an entirely

486 RICHARD EVANS.

different species. If, however, the largest spicules are looked
upon as fully developed forms, and the smallest ones as imma-
ture forms, the size of the normal spicule, that is the fully
developed form, agrees with those of Potamolepis leubnit-
zize rather than with the spicules of any other Potamolepid
sponge. Still there is a real difference in that these spicules
are distinctly swollen at the ends, while those of Potamo-
lepis leubnitziz are not. It seems, though closely resem-
bling the spicules of the species mentioned above, that the
spicules now discussed may belong to another species. The
matter, however, must be left open for the present, as the
above feature, when considered alone, is not a sufficient reason
for the formation of a new species.

BIBLIOGRAPHY.

e

BoweErpank, J. §8.—‘*A Monograph of the Spongillide,” ‘ Proc. Zool.
Soc. London,’ 1863, p. 440.

2. Carter, H. J.—‘ Contributions to our Knowledge of Spongida,” ‘ Ann.
Mag. Nat. Hist.” (5), im, p. 284, 1879.

3. Carter, H. J.—“‘ History and Classification of known Species of Spon-
gilla,” ‘Ann, Mag. Nat. Hist.’ (5), vii, p. 77, 1881.

DyzBowskI, W.—“ Studien iiber die Spongien des russischen Reiches, &c.,”
‘ Mém. d. |. Acad. Imp. d. Sci. d. St. Pétersbourg ’ (vii £), xxvii, No. 6,
1880.

5. Hiteenporr, M.—‘ On Two Fresh-water Sponges (Spongilla nitens,
Carter, and Spongilla Bohmii, Hilgéndorf),” ‘Ann. Mag. Nat.
Hist.’ (5), xii, p. 120, 1883.

. Hrypz, G. J.—‘* On some New Species of Uruguaya, &c.,” ‘ Ann. Mag.
Nat. Hist.” (6), ii, p. 1, 1888.

7. Marsnatt, W.—“‘On some New Siliceous Sponges, &c.,”” ‘ Ann. Mag.
Nat. Hist.’ (5), xii, p. 391, 1883.

. Mincuty, E. A.—“ Materials for a Monograph of the Ascons,” part i,
‘Quart. Journ. Micros. Sci.,’ n.s., xl, p. 469, 1898.

9. Ports, E.—‘ Contributions towards a Synopsis of the American Forms
of Fresh-water Sponges, &.,” ‘Proc. Acad. Nat. Sci. Philadelphia,’
p. 158, 1887.

ib

fo)

oO

TWO NEW SPONGILLE FROM LAKE TANGANYIKA. 487

10. Scuuuzs, F. E., und LenpEenretp, R. v.— Ueber die Bezeichnung der
Spongiennodeln,” ‘Abhandl. d. k. preuss. Akad. d. wiss. Berlin,’
Abh. ii, S. 1, pp. 1—35, 1889.

11. Soutas, W. J.—“Plectronella papillosa,” ‘Ann. Mag. Nat. Hist.’
(5), ii, p. 17, 1879.

12. Torsent, E.— Contribution a V’étude des Clionedes,” ‘Arch. Zool.
Expér. et Gen.’ (2), v bis., mem. iv, 1890.

13. Wetrner, W. S.—*Spongilliden Studien,” iii, ‘Arch. f. Naturgesch.,’
vol. Ixi, 1, p. 114, 1895.

EXPLANATION OF PLATES 37 and 38,

Illustrating Mr. Richard Evans’ paper “On Two New
Species of Spongilla from Lake Tanganyika.”

The figs. 1—8 on Plates 37 and 38 refer to Spongilla moorei, figs. 9 and
10 on Plate 38 to Spongilla tanganyika, and fig. 11 to Potamolepis from
the mud of Lake Tanganyika.

PLATE 37.
Fic. 1.—(x 1.) Spongilla moorei growing on a molluscan shell.

Fic. 2.—(a—e and e—m x 300; a’, U', c’, and d x 750.) A number of
irregularly shaped spicules showing both the variety and the irregularity of
form which they present. The spicules a’, J’, c’ represent a, 4, and ¢ on a
larger scale of magnification, and together with the spicule d show the relation
which exists between the axial threads in these compound spicules.

Fie. 3. —( x 300.) Amphioxea_ and amphitornota without swelling.
a, b, and J. Straight amphioxea. c—h. Curved amphitornota. 7, 4, and /
are possibly microscleres, or they may be young megascleres.

Fic. 4.—(x 300.) Amphioxea and amphitornota with swellings. a.
Straight amphioxea with four swellings. 4—/. Curved amphitornota with a
variable number of swellings from one to five. @ shows the swellings asym-
metrically arranged. 4—z. Hither microscleres or young megascleres with
one swelling.

Fie. 5.—(x 750.) Masses of silica found in Spongilla moorei,
varying both in size and shape.

488 RICHARD EVANS.

PLATE 38.

Fic. 6.—(x 200.) The skeleton near the surface as seen in section.
The spongin is shown forming a thin layer at the surface and covering the
fibres in the interior.

Fie. 7.—(x 800.) <A portion of two fibres with a large connecting
spicule and a much finer one with aswelling. The spongin is greatly thickened
at the points where the connecting spicules enter the main fibres.

Fic. 8.—( x 45.) The skeleton as seen in a section passing from the
base to the upper surface. A portion of the shell on which the sponge was
growing is seen at the bottom of the figure.

Fic. 9.—(x 200). A portion of the skeleton of Spongilla tan-
ganyike seen in section near the surface, showing the main fibres formed
almost entirely of spiny spicules, while both spiny and smooth ones are found
scattered about irregularly.

Fie. 10.—(x 400.) Spicules of Spongilla tanganyike. a—/
Straight spiny amphistrongyla. gaudj. Straight, spiny amphitornota. 4,
k, and/. Straight, sparsely spined amphioxea. 0, p, and g. Straight, smooth
amphioxea. m. Irregularly shaped, spiny spicule. 7. Microsclere.

Fie. 11.—Potamolepis spicules from the mud of Lake Tanganyika. a and
6. Two large spicules with a curved shaft and swollen ends, i.e. curved
amphitylota. ¢, d,ande. Three intermediate spicules with curved shaft and
rounded ends, but not swollen, i.e. curved amphistrongyla. / and g. ‘l'wo
small spicules similar to ¢, d, and e, i.e. amphistrongyla.

ON TETRACOTYLE PETROMYZONTIS. 489

On Tetracotyle petromyzontis, a Parasite of
the Brain of Ammocectes.

By

Albert W. Brown, B.A., F.L.S.,
Formerly Exhibitioner of Christ Church, Oxford.

Sd

With Plate 39.

Wuusr I was working as a student in the Department of
Comparative Anatomy at Oxford, in the Histology Class of
Mr. G. C. Bourne, my attention was drawn to a very extra-
ordinary appearance in some sections of the brain of Ammo-
coctes. Examination of sections of several different individuals
revealed a constancy in the appearance at first sight suggestive
of some highly complex abnormality, but a more thorough
examination of series of sections resulted in the discovery that
I had to deal with a parasitic form, which eventually proved
to be a Holostomid trematode.

Now the occurrence of a Trematode in the brain cavities of
a vertebrate is a unique phenomenon, occurring, so far as I can
discover, in no other case, and it therefore seems desirable to
publish some sketches and a description of the structure of
the parasite.

At the outset I wish to tender my thanks to Mr. G. C.
Bourne for his help during the early stages of the work. My
best thanks are also due, and cordially given, to Professor
Ray Lankester, in the stimulating atmosphere of whose
laboratory the work was done. Professor W. B. Benham,
recently lost to Oxford, gave me much assistance throughout
my work, and I am especially grateful to him for drawing my

490 ALBERT W. BROWN.

attention to some obscure references of great importance
shortly after I had begun to work on this form.

In O. F. Miller’s ‘ Vergleich. Anat. der Myxinoiden’ a
statement occurs to the effect that the fourth ventricle of the
brain contains many Diplosomids, and this is the earliest
statement yet found. My discovery therefore is not new.

George Gulliver, in papers dealing with the blood-corpuscles
of the lamprey published nearly thirty years ago, noticed
these forms, and stated their occurrence. He promised a
further description, but, so far as I can discover, did nothing
more than give them the rather fancy name of Neuronaia
lampetre. The genus Neuronaia (or Neuronaina!) was
founded by Goodsir for a parasitic form from the nerves of a
shark which he called Neuronaia munroi, and Gulliver’s
association of his species with this appears to have no justifica-
tion beyond a distant similarity of habitat.

The name Neuronaia is in every way an undesirable one.
The form here described is immature, and has therefore no
right to be considered even a new species till its adult form
has been determined. This I have, unfortunately, not been
able to do, but it probably belongs to an already named
Holostomid genus. In the second place, it is now the custom
to call such immature forms by the generic name of Tetra-
cotyle, and I propose therefore, in order to bring this form
into harmony with present-day nomenclature, to call it Tetra-
cotyle petromyzontis. The most confirmed devotee of
the laws of priority can scarcely object to this, seeing that
Gulliver gave the form a name without attempting to deter-
mine its real position. If every fancy name is to be retained,
all hope of reducing our zoological nomenclature to a simple
uniform system must be abandoned.

I may here point out the remarkable similarity in appear-
ance between Tetracotyle petromyzontis and the form
known as Diplostomum volvens.’ Indeed, at first I was

1 There appears to be some irregularity in the spelling of this word.

2 Nordmann’s figure of Diplostomum volvens is reproduced in the
‘Cambridge Natural History,’ vol. ii, p. 64, fig. 31.

ON TETRACOTYLE PETROMYZONTIS. 491

almost inclined to regard it as the same. Seeing, however,
that species of these Trematodes are determined a great deal
by the structure and arrangement of the generative organs, it
is, of course, possible that two forms might bear a close
resemblance to one another in their early stages, and yet
belong to different species. On the whole, a separate name
seems the most desirable. A list and descriptions of these
immature forms is very much needed as a supplement to
G. Brandes’ admirable work on the Holostomee.

Occurrence.—Tetracotyle petromyzontis is found in
great numbers in the brain cavities of Ammoceetes. I have
examined many specimens from the river near Oxford, and in
only one case have found it absent. Last April I found a
number of young lampreys in a stream in Sussex, but not a
single parasite was present in them. Gulliver states that
they are never absent from the brain either in the young or
adult conditions, but he does not say whence he obtained his
lampreys. Kupfer, Shipley, Gaskell, and others make no
mention of any parasites in the brain, and it seems to me
probable that Gaskell would have seen them if present. I
believe a good many of his lampreys came from Surrey.
For myself I have never found Tetracotyle petromyzontis
in the adult lamprey nor in any lamprey of greater length
than six to seven inches. Its distribution therefore does not
seem to be as universal as Gulliver thought, and I can hardly
believe that many competent observers would have over-
looked it if present in the lampreys examined by them.

In many Ammoceetes the brain is bulged out completely by
these tiny animals, and I have sometimes seen them squirted
eighteen inches into the air when the roof of the brain is
divided suddenly with a sharp knife. They are found mostly
in the region of the fourth ventricle, and especially under
the choroid plexuses between the foldings of the roof of the
brain, whence they apparently draw their nourishment.

In spite of all this, the Ammoceetes appear to live on quite
comfortably and to suffer no great inconvenience, although
the choroid plexuses of the brain often have an inflamed

voL. 41, parry 4.—NEW SERIES. MM

492 ALBERT W. BROWN.

appearance when the parasite is present in large numbers.
I have kept several Ammoceetes alive for nearly three months,
during which time they have lived and grown quite normally.
On killing and opening, the brain was found packed full of
parasites.

Tetracotyle petromyzontis will live for some time
outside the brain; Gulliver says for several days, but mine
have never lasted more than three hours. Whilst alive they
fix themselves by the ventral sucker, and, contracting rhyth-
mically, undergo rapid changes of shape. Outline sketches to
illustrate these changes are seen ou Pl. 39, fig. 2.

ExrernaL Cuaracters.— Tetracotyle petromyzontis
has an average length of about 42 mm. It possesses a typical
fluke-like form. At the posterior end of the body, which is
slightly broader than the anterior, the rudiments of the tail,
in which the adult generative organs are developed, are visible.
The mouth is at the anterior end, and is surrounded by the
oral sucker. On each side of it are two retractile ear-shaped
projections with which numerous gland-cells are connected.
About midway on the ventral surface is situated the ventral
sucker, and just posterior to this the large glandular adhesive
organ. At the posterior end of the body is the excretory
pore.

The intestine, which is easily seen under a low power of the
microscope, is Trematode in character. Some little distance
behind the mouth it expands into a powerful pharynx which
seems to perform a kind of pumping action during life.
Behind the pharynx the intestine forks, each fork ending
blindly just posterior to the glandular adhesive organ.

The body appears filled with a number of bright globular
bodies. These are contained in the terminal vesicles of the
excretory system.

Careful focussing shows the integument dotted all over the
surface with minute black dots arranged in regular rows.
Sections prove these to be minute canals pee the cuticular
layer (Pl. 39, fig. 3).

InterNaAL AnNatomy.—Below the ectoderm come the usual

ON TETRACOTYLE PETROMYZONTIS. 493

layers of muscles. The longitudinal bands of muscle are
quite easily visible in the animal without the aid of sections.
They extend almost the entire length of the body, except in
the region of mouth and lateral adhesive organs, where the
musculature has a circular arrangement. The general mass
of the body is filled with a highly vacuolated parenchymatous
tissue, the ultimate structure of which is very difficult to make
out. In this he the main organs of the body.

The intestine does not call for special mention. The fork
of the intestine is occupied by a mass of gland-cells which
extend on each side, forming a zone across the animal. These
cells stain very readily, and show up in strong contrast with
the general parenchyma of the body. They are filled with
material of a slightly more yellow tinge. The space between
the intestinal fork and the ventral sucker is entirely filled
with them.

These gland-cells communicate with the retractile ear-
shaped structures on the sides of the oral region, and open
there to the exterior by fine canals. The “ ears”? themselves
are retracted by a series of longitudinal muscles shown in
fig. 6, which appear to be specially differentiated longitudinal
muscles.

I was able to trace out the fine tubes leading from the
gland-cells in several sections. Their connection with the
cells in the intestinal fork is not so easily seen, but I am
convinced that they do communicate with part of them at
least, both groups of cells being similar in _ histological
characters. Both group of cells, then, are connected either
wholly or in part with the antero-lateral retractile pro-
cesses.

Excretory System.—This system of organs commands the
most interest. Fraipont, in ‘ Archives de Biologie, I,’ has
given an account of the excretory system of Diplostomum
volvens which agrees pretty closely with this. It consists of
two parts—a coarser, and a finer and more dorsal portion.

The coarser part leads from the excretory pore in the form
of two main trunks, whose terminal portions are dilated to

4.94, ALBERT W. BROWN.

form the excretory bladder. These large trunks run the
whole length of the animal. About halfway each gives off a
branch vessel, which runs near the lateral surface of the body
backward nearly to the excretory pore.

The two main branches run on to the anterior end of the
animal, when they become transverse, and, meeting in the mid-
line, form a vessel running backwards as far as the ventral
sucker.

All these vessels give off at intervals, and finally break up
into smaller and smaller branches, each of which ultimately
ends in a dilated vesicle containing a rounded body called
by Fraipont ‘calcareous body.” The calcareous bodies
appear to be in a semi-fluid condition, and contain a good
deal of calcium carbonate. On treatment with acid they
dissolve and disappear, and at the same time give off
gas. They are non-crystalline, and occupy nearly the
entire terminal vesicle. In the centre of these calcareous
bodies may be seen one or two bright dots looking like
granules. In some cases I have found the terminal vesicles
quite empty.

In Diplostomum volvens Fraipont described a more
dorsal network composed of very fine tubules ending in flame-
cells. This finer network joined the main trunk of the
coarser at about one third of the entire length from the
anterior end of the animal. Now I have seen flame-cells
throughout the entire body of Tetracotyle petromyzontis
in the dorsal region, but I am quite unable to assert the
existence of a definite network joining the coarser part at a
given point. My endeavours to see such a network have
ended fruitlessly, although I have made many attempts, for
the tubes are excessively minute and seem capable of occlu-
sion at times. Consequently my observations only enable me
to assert the existence of flame-cells forming scattered portions
of network, but whether these communicate with the rest of
the system, as in Diplostomum volvens, or at many points
near the terminal vesicles—as I have imagined them to do in
several cases—I am unable to say definitely.

ON TETRACOTYLE PETROMYZONTIS. 4.95

The flame-cells are best seen by staining with very weak
methylene blue.

I may mention that terminal vesicles of the excretory
system appeared surrounded in parts by subsidiary vesicles.
As mentioned above, the addition of acid to the animal causes
the calcareous bodies to disappear gradually with evolution of
gas which comes off at the excretory pore. The solution of
these bodies is far too gradual to allow of them being con-
sidered entirely calcareous. Probably they contain a certain
amount of organic matter as well. As to their formation I
have nothing to say.

The physiology of this excretory system seems to me well
worth further study, in view of the existence of calcareous
bodies in the parenchyma of other Platyhelminthes. Some
light might be thrown on the origin of the latter, and possibly
even the older view that they were derived from the excretory
system be re-affirmed.

Tail Region.—Reference was made earlier in this paper to
the tail region of the body which grows out of the posterior
region, and in adult forms contains the reproductive organs.
In the younger stage the excretory bladders seem to occupy
a great deal of it.

The tail is not of the same length in all individuals examined
by me. In many it is hardly perceptible, whilst in others it
may be nearly one third the length of the animal. A study of
sections of the posterior end of the animal makes manifest the
great number of nuclei aggregated there. They are scattered
about pretty uniformly in individuals with an inconspicuous
tail, but in those with longer tails they are already becoming
gathered into groups—some of them very defined—and these
are probably rudiments of the future generative organs.

Part, at any rate, of the development of Tetracotyle
petromyzontis takes place in the brain of Ammoceetes, but
does not proceed very far.

I have been quite unable to trace any connection of Tetra-
cotyle petromyzontis with an adult form. I do not quite
know how to trace it till more knowledge of the animals

4.96 ALBERT W. BROWN.

preying on Ammoccetes is obtainable. Most Holostomids
have birds as final hosts, but in the places from which my
Ammocetes were brought there are not many birds to prey
on them.

Yarrel, in his ‘British Fishes,’ states that eels prey on
Ammoceetes. I obtained some large eels, starved them for a
time, and then tried to feed them with Ammoceetes, but
neither alive nor dead would they touch them.

The most puzzling question of all to decide is how the
parasites get into the brain. Possibly they reach it by means
of the blood-vessels, but this seems rather improbable in
spite of the apparent ease with which Ammoceetes can stand
injury to the brain. It seems more probable that they get in
before communication between the brain cavities and the
external world is interrupted. There seems, however, no
possibility of them getting out again without killing the
animal, and their ultimate fate is at present a mystery. Do
they kill off the forms they infest after a certain time? To
determine this it would be necessary to keep a large stock of
Ammoceetes and examine carefully all who died off.

It is almost incredible that any vertebrate could live on,
apparently without discomfort, whilst its brain is packed with
hundreds of flukes. It is true they do not appear to damage
the brain substance, but they must consume a great deal of
nourishment intended for the brain itself, and their excretory
products must surely interfere in some way with the proper
brain functions.

I have thought it well to publish these brief notes on this
interesting form, although my attempts to solve its life-history
have thus far proved ineffectual. The existence of so highly
organised a form as a Trematode in the brain cavity of a
vertebrate is a unique phenomenon in many ways, and this
must commend it to the interest of all naturalists. For one
thing it shows that even the brain is not exempt from the
attacks of Trematodes.

Gulliver’s statement of the occurrence of this form did not
attract the attention it deserved, and he only gave an outline

ON TETRACOTYLE PETROMYZON'IS. 497

sketch of it. I hope, therefore, that the accompanying figures
and this description may help to gain for Tetracotyle
petromyzontis the interest it merits.

REFERENCES.

Branpzs, G.— Die Familie der Holostomiden,” ‘Zool. Jahrbicher,’ Ab. fiir
Systematik, v, pp. 549—604, pl. xxxix—xli.

Braun, Max.—“ Zur Entwickelungsgeschichte der Holostomiden,” ‘ Zool.
Anzeiger,’ xvii, pp. 165—167.

Goopstr, H. S.—‘ Transactions Roy. Soc. Edinburgh,’ xv, p. 561, 1844, and
‘Proceedings Roy. Soc. Edinburgh,’ i, p. 466.

Guiiiver, G.—‘ Quart. Journ. Mier. Sci.” Nov., 1872.

Miter, J.—‘ Vergleichende Anatomie der Myxinoiden,’ Berlin, 1835.

Norpmann, A.—‘ Mikrographische Beitrage zur Naturgeschichte der Wirbel-

losen Thiere,’ Berlin, 1852.

EXPLANATION OF PLATE 39,

Illustrating Mr. A. W. Brown’s paper, “On Tetracotyle
petromyzontis.”

Kry To THE LETTERING.

at. Atrium of the glandular adhesive body. ev. 4, Excretory bladder.
ex. p. Excretory pore. ex. v. Main vessel of the excretory system. g/. d.
Glandular adhesive body. g/.c. Gland-cells. dé. Intestine. /. s. Lateral
adhesive organ. m. Mouth. muse. Retractor muscles of the lateral adhesive
organ. 2. ¢. Nerve-cord. «@s. Hsophagus. 07. s. Oral sucker. p. v. Poste-
rior lateral branch of main excretory vessel. ¢. Tail region of body. tub.
Tubules leading from gland-cells to lateral adhesive organs. 7. ». Terminal
vesicle of the excretory system. v. s. Ventral sucker.

Fic. 1.—View from the ventral surface of the entire animal. Leitz, ce. 1,
obj. 5.

Fic. 2.—Outline sketches, showing changes of shape at four consecutive
short intervals.

498 ALBERT W. BROWN.

Fic. 3.—(a) Surface view of integument, showing dots caused by ends of
canals.
(6) Section of cuticle, showing minute canals.

Fie. 4.—Camera drawing of a horizontal section, to show tubules connecting
gland-cells with the lateral adhesive organs.

Fie. 5.—Diagram to show arrangement of the gland-cells of the lateral
adhesive organs.

Fie. 6.—Horizontal section, showing retractor muscles of the lateral
adhesive organs.

Fic. 7.—Semi-diagrammatic view of the coarser ventral portion of the
excretory system. ‘Terminal vesicles and finer tubules shown only on the
right side of the figure.

Fic. 8.—More highly magnified part of the excretory system, to show the
terminal vesicles and calcareous bodies contained in them.

Fic. 9.—Four calcareous bodies.

Fic. 10.—Longitudinal vertical section, to show the greater number of
nuclei in the posterior region of the body. Leitz, oc. 1, obj. 7.

Fie. 11.—Schematic plan—combined from several horizontal sections—of
the chief organs of the bodies.

CALCAREOUS SKELETON OF THE ANTHOZOA. 499

Studies on the Structure and Formation of the
Calcareous Skeleton of the Anthozoa.

By

Gilbert C. Bourne, M.A., F.L.S.,

Fellow and Tutor of New College, Oxford; University Lecturer in Compara-
tive Anatomy.

With Plates 40—43.

Mitne Epwarps and Haime, and the older authors on coral
structure, regarded the calcareous skeleton of the Madreporaria
as a calcified mesoderm. In 1881 Dr. A. von Heider, of
Graz (11), described a layer of rounded cells underlying the
corallum of Cladocora, and named them ecalicoblasts, since he
judged from their position that they must be the coral-forming
elements. In the following year G, von Koch (17), in a
memoir on the development of Astroides calycularis, con-
firmed the observation of von Heider, and further demonstrated
the fact that the calicoblast layer is a product of the basal
ectoderm.

Von Koch described the first elements of the corallum as
making their appearance between the basal ectoderm and the
surface to which the larva attaches itself, and he described the
first formed calcareous particles as ‘‘ crystalline spheroids,”
which increase in number and eventually fuse together. “ Bei
Astroides bildet sich aus dem Ektoderm der Fussscheibe zuerst
eine diinne Kalkplatte, welche sich irgend einer Unterlage
anlegt, und die Fussplatte des spatern Polypars darstellt. Sie
ensteht aus krystallinischen spheroiden Ko6rperchen, welche

500 GILBERT C. BOURNE.

wie sich aus ihrer Lage ergibt, aus Ausscheidungen der Ekto-
dermzellen hervorgehen.”

. After the discovery made almost simultaneously by von
Heider and von Koch of the calicoblast layer, these cells were
recognised by several other authors working on the same
subject. In 1886 Fowler (7) described and gave somewhat
diagrammatic figures of the calicoblasts of Flabellum pata-
gonichum, and in the same year von Heider (12) published a
careful memoir on the anatomy and histology of Astroides
calycularis and Dendrophylliaramea. The calicoblasts
of the former species were not described, but special attention
was paid to them in Dendrophyllia. Von Heider describes the
calicoblasts in this species as being in many places incon-
spicuous, in some quite invisible, but easily discoverable in
others. Where they occur they are of two kinds: the most
numerous are polygonal or fusiform cells, one or two layers deep,
with a well-defined nucleus and granular contents. In other
places, and especially where the mesoderm of the body-wall
turns inwards to form the mesenteries, von Heider found cells of
more or less wedge shape, their pointed ends turned towards the
mesogloea. He says that usually these cells have no nucleus,
and that their inner ends are filled with exceedingly fine stria-
tions. Some cells adjoining these had less conspicuous stria-
tions, more granular contents, and a well-defined nucleus, and
these he regarded as being transitional between the first and
second kind of calicoblast. Finally, he concluded that the
striz were spicules of calcium carbonate formed within the cell
much asare the spicules of Alcyonaria. Von Heider does not
say whether he examined his hypothetical spicules with a
polariscope.

In the same year W. L. Sclater (26), in a paper on Stepha-
notrochus Mosleyanus, described a tissue or series of cells
as occurring everywhere between the corallum and the meso-
derm. These, he says, are very different from the calicoblasts
figured by von Koch, being of irregular shape, separated from
one another by intervals so as to seldom form a definite layer,
and striated in an extraordinary way. Sclater’s figures leave

CALCAREOUS SKELETON OF THE ANTHOZOA. 501

no doubt that the structures in question are identical with von
Heider’s second kind of calicoblasts.

In 1888, in a paper on the anatomy of Mussa and Ku-
phyllia (2), I described the presence of elongate columnar
calicoblasts on the upper and peripheral parts of the septa of
Mussa distans, and at the same time I criticised von
Heider’s interpretation of the striated wedge-shaped bodies as
calicoblasts, pointing out that I had found precisely similar
structures in Mussa, but never in the region of most active
coral growth. On the contrary, they were always associated
with the mesogloea of the mesenteries ; and, moreover, the strize
could not be due to the presence of spicules of carbonate of
lime, for they were unaltered after prolonged decalcification.
At the same time Fowler (8) referred to the existence of
similar wedge-shaped structures in Pocillopora brevi-
cornis, and gave his opinion that they were rather connected
with the attachment of the mesentery to the corallum, than
with the secretion of carbonate of lime. In the same paper
Fowler described and gave a diagrammatic figure of elongated
columnar calicoblasts in Lophohelia prolifera, Ina sub-
sequent paper (9) Fowler described and gave careful drawings
of the striated structures in Stephanophyllia formo-
sissima and Flabellum alabastrum, and pronounced them
to be nothing more than offshoots of the homogeneous meso-
gloa of the mesentery, possessing neither nucleus nor cell-
wall. He held that their occurrence only in the neighbour-
hood of the lines of attachment of the meseuteries, their
position, shape, and striation, indicated that their function was
to provide an increased surface for fixation of the mesentery,
and a firmer fulcrum for the action of the powerful retractor
muscles. No further observations of importance were made on
the subject till 1896, when Mrs. Gordon (formerly Miss Maria
M. Ogilvie) published a long and minute account of the micro-
scopic structure of various types of Madreporaria (25), and
suggested an entirely new classification of the group founded
upon the microscopic characters. Mrs. Gordon rejects
Fowler’s and my accounts of the striated structures, and says

502 GILBERT C. BOURNE.

(loc. cit., p. 102), “I find, however, that Koch, Fowler, and
Bourne are wrong in their conception of the calicoblast. My
investigation of the superficial layers of the skeleton have
given the fullest confirmation of Heider’s idea, that the calico-
blasts were actually converted into calcareous groups of fibres.
I hope to show, in the course of this paper, that the fibre-
containing calicoblasts which lie next to the ske-
leton are shed off, so to speak, from the polyp, new cells
taking their place in the ectoderm by cell division. The
shed calicoblasts build up successive layers of calcified cells,
which hang together at first by their cell-walls, and ultimately,
as crystalline changes continue, form the individual lamine of
the skeletal structures. The whole Madreporarian skeleton is
formed of such laminz ; any apparent variation in microscopic
structure is accounted for by some difference in the shape and
position, locally, of the ectoderm.’’ The quotation shows that
Mrs. Gordon attached great importance to the conception of
the formation of the corallum from a number of calcified cells,
and she made it the basis of a number of interesting specula-
tions concerning the origin of the different types of coral
structure which she described.

On p. 115 of the same work Mrs. Gordon gives an account
of the isolated skeletal elements of the Madreporarian
corallum. These she discovered to be scale-like structures
entirely filled with a bush of minute fibres ; others had granular
contents, or partly fibrous, partly granular. The resemblance
of these scales to the striated structures described as calico-
blasts by von Heider led Mrs. Gordon to seek for proof of the
identity of the two. I give Mrs. Gordon’s proof in her own
words. ‘It has been frequently mentioned by writers on
corals that organic remnants after removal of the polyp may
be found on the skeleton. Wherever on the skeleton I found
such remnants they consisted of calicoblasts, which showed in
shape, size, and contents the varieties already drawn by von
Heider. ‘he cells were round or obovate. The contents
varied from yellowish organic-cell material to the inorganic
fibrous condition. Comparison of my own observations on

CALCAREOUS SKELETON OF THE ANTHOZOA. 503

several corals with the figures given by von Heider left no
doubt that the isolated skeletal element was a cal-
cified calicoblast cell. . . . We may look upon the super-
ficial layers of the skeletal elements, and of incompletely
calcified calicoblasts, as the outer layers of a many-layered
ectoderm. The ectoderm of the Madreporarian polyp” (I
presume Mrs. Gordon here means the calicoblast layer) “is
figured by Koch, Fowler, and Bourne as a simple layer of cells.
I have observed, on the contrary, that a section through soft
and hard parts shows an ectoderm, sometimes composed of a
simple cell layer, sometimes several cells deep. Heider’s
figures indicate a similar observation. The calicoblasts re-
main adherent to one another in dense groups, or may be
uniformly distributed. And in this manner they are gradu-
ally left behind on the skeleton and completely calcify, while
active cell division develops constantly new ectodermal cells.
The calicoblasts adherent to the skeleton represent such as
were already in the course of losing living continuity with the
polyp, at the time when the polyp was removed from the
skeleton. The above observations made on the skeletal sur-
faces will be seen to carry out fully the opinion of Heider,
that the skeletal deposit forms, in the first place, within the
calicoblast. I pointed out in the introduction that Koch’s
view of the extra-cellular deposition of the skeleton had been
accepted by Fowler and Bourne. The figures given by the
two last-named authorities are confined entirely to prepara-
tions made from completely decalcified specimens. My obser-
vations have been made partly on dry skeletons, partly on
skeletons from which the soft parts have been freshly removed,
organic tissues remaining here and there adherent, and partly
on preparations showing the body-wall in position.”
Accompanying this statement are figures of individual scales
taken from the dissepiments of Galaxea, which Mrs. Gordon
claims to represent calicoblasts in various stages of calcifica-
tion. ‘The scales as figured show a distinct striation, due to
their being composed of a tuft of elongate radiating calcareous
crystals, and several of them exhibit some dark blotches which

504: GILBERT C. BOURNE.

are doubtless traces of organic matter. The figures certainly
do not show to demonstration that the scales are what Mrs.
Gordon claims them to be, calcified cells. And though she
identifies the scales with the striated structures called calico-
blasts by von Heider, she does not bring forward any new
evidence, either in drawings or in the text, to show that the
striated structure of the scales of Galaxea, and of von Heider’s
calicoblasts, is due to one and the same cause, viz. the presence
of spicules of carbonate of lime.

It is easy to make a preparation from the vesicular exotheca
of Galaxea, showing the scale-like calcareous structures figured
by Mrs. Gordon (loc. cit., figs. 84, 8B), and it may be conceded
that their resemblance to the striated structures identified by
von Heider as calicoblasts is sufficiently striking. But we
have no evidence that the striated structures are found in this
position in the fresh polyp; we have no evidence that the
striations of von Heider’s calicoblasts are due to the formation
of spicules within them ; but, as I shall show directly, we have
evidence to the contrary, and from the analogy of other corals
there is every reason to believe that the striated structures are
not present in the soft tissues covering the exotheca.

Nor can I entirely agree with Mrs. Gordon’s description of
the superficial emergences of the corallum in Galaxea and
other corals as “ scales.”

It will appear in this paper that the surfaces of certain corals
present an appearance like that described for Galaxea, but
that the apparent scales are really due to the emergence of
bundles of parallel crystalline fibres at the surface of the
corallum. But before going further into this question I will
give the results of my observations on the structure and
formation of the spicules of Aleyonaria. Since we know that
these spicules are entoplastic products, it seemed to me that a
knowledge of the minute structure of the spicule might assist
in the interpretation of the minute structure of the Madre-
porarian corallum. Throughout this paper the word “ spicule ”
denotes an entoplastic product of a single cell or of a coeno-
cyte.

CALCAREOUS SKELETON OF THE ANTHOZOA. 505

Kolliker, in his ‘Tcones Histologice,’ p. 119, has given the
results of a thorough investigation of the spicules of Alcyo-
narians. He described them as consisting of an organic and
an inorganic part. The organic part consists, according to
him, of a thin cuticular membrane, but he denied the existence
of any central organic basis. ‘‘ Wenn ich iibrigens vorhin
bemerkte dass das Innere der Kalkkérper durch Sauren sich
auflose, so muss ich bemerken, dass es sehr schwer ist, in
dieser Beziehung volkommen ins Reine zu kommen. Nach
Anwendung verdiinnter Saiuren bleibt in der Regel noch ein
kérnig streifige, oft wie aus Faserchen zusammengesetze Masse
im Inneren, und brachte mich dies anfianglich zu der Meinung,
dass auch Inneren ein organischer Riickstand bleibe. Setze
ich dann aber eine concentrirte Sdure in geniigender Menge
zu, So loste sich auch dieser Rtickstand, und habe ich in eine
Reihe von sorgfaltig untersuchte Fallen mich davon tiberzengt,
dass schliesslich nichts als die Cuticula sich erhalt, in welcher
Beziehung ich iibrigens noch bemerken will dass Salzsiure
viel entschiedener wirkt als Essigsiiure. Aus diese Thatsache
folgt iibrigens nicht dass im inneren der Kalkkérper gar keine
organische Substanz enthalten ist, und zeigen schon die
mannichfachen Farbstoffe dieser Kérper, dass sie nicht einzig
und allein aus Erdsalzen bestehen. Auch der Zahmschmelz
enthalt etwas organische Materie, obschon er in Siuren keinen
nennenswerthen Rickstand Jasst.” Though this statement is
cautious enough, he is more definite further on, where he says
(p. 121), “ Ein centraler organischer Faden, wie ihn viele
Kieselnadeln von Spongien besitzen, mangelt den Kalkkérpern
der Polypen ganz und gar.”

The intimate structure of the spicule is described by Kélliker
as crystalline. The most conspicuous feature in the spicules
is a fine concentric striation parallel] to the surface, due to the
apposition of concentric lamellz. There is also a fine striation
and punctation in the longitudinal axis, from which one may
conclude that there is a special structure, as also by the fact
that after prolonged action of weak acids the inner part of the
spicule breaks up into small crystalloid needles. Kolliker also

506 GILBERT C. BOURNE.

noticed and figured (loc. cit., pl. xvii, fig. 7) that in the
spicules of. certain Gorgonids the superficial warts and pro-
jections were continued through the substance of the spicule
nearly to its centre, as it were by roots, and that these internal
prolongations were often forked or branched.

K6lliker entered at great length into the forms of Alcyo-
narian spicules, of which he distinguished two main classes,
the smooth and the warty spicules. The smooth spicules he
divided into spheroids, spindles, lenticular spicules, and
cylinders ; the warty spicules into (a) simple spicules, which
include spindles, clubs, double clubs, and double spheroids,
double stars, double wheels, double spindles and scales, with
several subdivisions of the first four kinds, and into (6) com-
posite spicules, subdivided into doublets, triplets, quadruplets,
&e. Kolliker also denied the intra-cellular development of
Alcyonarian spicules, a mistake which was subsequently
corrected by Kowalevsky and Marion and von Koch.

The spicules of recent Alcyonaria, with the single exception
of Heliopora, are, as is well known from the work of Kow-
alevsky and Marion (28) and von Koch (16), true spicules.
They are formed in certain scleroblastic cells which originate
in the ectoderm, and either remain there (some species of
Clavularia, Xenia) or migrate into the mesoglea. I have
been at some pains to verify previous statements on the subject,
and to discover, if possible, the earliest stages of spicular for-
mation, and to trace them upwards.

Whilst working at Roscoff, by the kind permission of M. H.
de Lacaze-Duthiers, | was able to make very satisfactory
preparations of Alcyonium digitatum and Gorgonia
cavolinii illustrating this point. The specimens were killed
in an expanded condition by rapid immersion in a ‘5 per cent.
solution of osmic acid in sea water, were thoroughly washed with
distilled water, and stained for twenty minutes with Ranvier’s
picro-carmine. The expanded polyps were cut off close to
their bases, placed in dilute glycerine, and laid open. The
ectoderm and endoderm having been removed with a camel’s-
hair brush, very satisfactory flat preparations were obtained

CALCAREOUS SKELETON OF ''HE ANTHOZOA. 507

illustrating the formation of the spicules in the lower moieties
of the exserted portions of the polyps.

The scleroblasts of Aleyonium digitatum (Pl. 40, fig. 1)
have the form of irregularly polygonal, ovate, or amcebiform
cells, varying very much in size and shape, but resembling
one another in the coarsely reticulated structure of the proto-
plasm, and in their dark granular appearance. Vacuoles of
small size (not the ultimate vacuoles of Biitschli) are visible
in the protoplasm, and not infrequently a vacuole contains a
single microsome. The scleroblasts run in strands and
patches through the mesogloea at the bases of the expanded
polyps, and may be found, though they are not easily studied,
in the thickened mesogloea of the coenenchyme. They are
always accompanied by two other kinds of cells whose function
is obscure. The one kind, marked gr. in fig. 1, are rather
smaller than, but of similar shape to, the scleroblasts, and are
filled with minute highly refringent granules; their nucleus
is rarely to be seen, being hidden by the granules. My studies
on other Alcyonaria lead me to believe that their function is
to secrete the gelatinoid substance of the mesoglea. The
second kind of bodies I call provisionally the ovoid bodies.
Each of these is about ‘0075 mm. in length, is surrounded by
a protoplasmic sheath, and has a relatively large nucleus on one
side of it (fig. 1, ov.). These were noticed but not well figured
by Hickson (18), and similar structures were observed by von
Heider in Cladocora. Hickson suggests that they may
possibly be parasitic sporozoa, but my observations do not lend
any support to his suggestion. Their contents are clear and
highly refringent, but are not calcareous. They are not
affected by prolonged treatment with dilute acids; they are
unaffected by treatment with acetate of potash (used with the
view of precipitating lime if any was held in solution) ; they
do not light up when viewed through the polariscope with
crossed Nicols. They stain deeply with hematoxylin, but are
unaffected by other dyes. I have met with similar ovoid
bodies among the scleroblastic cells of all the Anthozoa which
I have examined carefully, and I am unable to guess their

vou, 41, pARY 4,—NEW SERIES. NN

508 GILBERT C. BOURNE.

function. I have, however, observed in some Madreporaria
traces of a coiled filament within the ovoid body, which leads
me to the conclusion that they are degenerate nematocysts. I
have not been able to discover any traces of a coiled filament
in Aleyonium, Gorgonia, or Heliopora, but the ovoid bodies
have in other respects the characters of developing nematocysts,
aud it is not unreasonable to conclude that the scleroblasts in
the course of their migration from the ectoderm have carried
with them some of the most characteristic constituents of that
layer in the shape of nematocysts, which, being functionless,
have undergone degeneration. .

One of the smallest sclerites which I was able to discover is
shown in fig. 1, sel. It consists of a single minute crystal,
‘01 mm. long by ‘0025 mm. broad, lying in a distinct vacuole
in the centre of the scleroblast. Adjoining it is another
scleroblast, in which there is a minute nodule, some ‘005 mm.
in its greatest diameter, in addition to a relatively large and
already irregular spicule.

Figs. 2 and 38 represent larger sclerites enclosed in the cells
which have given rise to them. Fig. 4 shows what are ap-
parently two cells containing sclerites fusing together, but it
is probable that this is the result of division of the nucleus of
the scleroblast without corresponding division of the cytoplasm,
and the two sclerites which are contained in the binuclear cell,
after reaching a certain size and form, are becoming joined
together. Fig. 5 shows the normal condition of a young spicule
of Aleyonium ; the sclerite has the characteristic shape, some-
what resembling a caudal vertebra, and is enclosed in a proto-
plasmic sheath containing two nuclei. In older and more
complicated spicules I have counted three or four nuclei, and
where two are present they are generally at the ends of the
spicule, and not at the sides as in fig. 5. Though the exist-
encé of two, three, or four nuclei suggests the presence of as
many cells, I do not think that more than one cell, i.e. more
than one continuous body of cytoplasm, is concerned in the
formation of a single spicule in Aleyonium digitatum.
The scleroblasts, as is seen in fig. 1, are often ccenocytes con-

CALCAREOUS SKELETON OF THE ANTHOZOA. 509

taining two, three, or more nuclei; where two sclerites are
formed in a single scleroblast two nuclei are usually present,
and if only one sclerite is formed the nucleus apparently
divides when the sclerite has attained a certain size, and the
division is repeated as growth continues, without any corre-
sponding division of the cytoplasm.

So far as the formation of the spicules is concerned, all that
is true of Aleyonium is also true of Gorgonia Cavolinii,
and these are the only two Alcyonarians which I have been
able to obtain in a living condition, and to preserve’by appro-
priate methods.

The spicules which I have examined belong to the smooth
lenticular, the spindle-shaped and scale-like warty kinds de-
scribed by Kolliker.

The lenticular spicules were obtained from Clavularia
coerulea, Ehrh., and from Xenia and Heteroxenia. In
Clavularia cerulea these spicules are formed within ecto-
dermic cells, which do not migrate into the mesoglea. Each
spicule is of flat ovate shape, measuring some 0:02 mm. in its
longer and 0:01 mm. in its shorter diameter. These lenticular
simple spicules are remarkable for the large amount of organic
matter which they contain, an amount so great that, after
slow decalcification, the organic basis retains the shape and
size of the spicule, and can readily be stained by diffuse stains
such as hematoxylin or acid fuchsin. The undecalcified
spicule, when examined by the polariscope with crossed Nicols,
grows fainter in four positions at 90° to one another, but is
never wholly extinguished. This suggests that it is made up
of a number of crystals, the majority of which are parallel to
one another. In some few of the spicules I could detect signs
of a dark cross when the prisms are crossed, but this is only
an occasional phenomenon. In some specimens which had
been treated for a short time with very dilute acetic acid, I
could detect minute crystalline fibres which seemed to diverge
fan-wise from the centre towards the two extremities of the
oval spicule. But the structure is so minute that one cannot
see anything distinctly, even with an immersion lens. I can

510 GILBERT C. BOURNE.

only say with certainty that the spicule is composed of a
number of very minute fibro-crystals. The organic matrix
left after decalcification consists of a close feltwork of ex-
ceedingly fine fibres, the whole enclosed in a distinct and
apparently structureless spicule sheath. The similar spicules
of Xenia and Heteroxenia, which I have described in
another place (4), behave under polarised light in exactly the
same way as those of Clavularia cerulea, and their organic
basis is also similar.

Secondly, there are the flat scale-lke spicules which occur
in Primnoa, Plumarella, and in many other Alcyonarians.
Examination by polarised light shows that the scale-like
spicule of Primnoa and Plumarella is made up of a
number of crystals of calcium carbonate lying in nearly the
same plane, and radiating from a common centre. Fig. 6
is a drawing of a single spicule of Plumarella delicatis-
sima, Wright and Studer, and fig. 7 is a drawing on a
smaller scale of the same spicule when viewed through the
polariscope with the prisms uncrossed. The characteristic black
cross leaves no doubt as to the crystalline arrangement of
the spicule, and exactly the same effect is obtained with many
inorganic crystalline deposits, e.g. with platino-cyanide of
magnesium. So far as I have been able to determine, these
scale-like spicules in Primnoa and Plumarella are formed
by several cells, or at least by a comparatively large coeno-
cytial investment containing many nuclei. The material at
my disposal is not sufficiently well preserved to allow me
to speak with certainty on this point, but sections made
through the zooids of both genera show that the scale-like
spicules, which form opercula corresponding in number to the
tentacles, and situated at the bases of these organs, are each
enclosed in a cavity in the mesogloea, which cavity is lined by
a protoplasmic layer in which many nuclei are embedded. ‘The
specimens were too much macerated by the prolonged action of
spirit to enable me to say definitely that no cell outlines are
present in the protoplasmic lining, but at all events I was un-
able to discover any in sections stained by various methods and

CALCAREOUS SKELETON OF THE ANTHOZOA. 5

examined with the highest powers. The distinction between
the ccenocytial layer and the cell layer is, perhaps, of no great
importance, but it is of more importance to note that the
arrangement of the crystals composing the spicule bears little
relation to the structure and arrangement of the cells from
which the spicule is formed.

The actual crystalline structure is but slightly indicated in
the fresh spicule, but it becomes evident after treatment with
dilute acids. A spicule which has thus been partially decalci-
fied or ‘‘ etched ” is seen to be composed of a number of minute
calcareous fibres which radiate from the centre towards the
periphery. Their arrangement, however, is not perfectly radial.
They cross and overlap one another to some extent, and many
of the fibres crop out on the surface of the sclerite, and do not
reach its edge. Hence there is some amount of interference,
and the dark cross shown in fig. 7 is not perfectly symmetrical,
nor are all parts of its limbs equally opaque. Some of the
crystalline fibres may be isolated at the edges of the etched
sclerite. Each is a minute acicular crystal about 0°03 mm.
long, and behaves under polarised light as a true uniaxial
crystal.

I did not succeed in isolating the organic constituents of the
spicules of Primnoa or Plumarella.

Thirdly, there is the spindle-shaped spicule in its many
varieties, usually covered with spines, warts, and projections of
different kinds. My observations on this form of spicule were
almost entirely confined to Spongodes and Siphongorgia
mirabilis, whose spicules are of comparatively simple form,
and large enough to admit of my making transverse and longi-
tudinal sections.

The spicules of Spongodes (sp. incert.) are shown in fig. 9.
They are long fusiform structures, covered with warty pro-
jections and of very variable length. Some of the larger
spicules, which serve as supports for the individual zooid heads,
are as much as 3°5 mm. long; the smaller, which lie in the
walls of the zooids, are about 0°2 mm. long. They are
variously coloured, crimson, orange, or pale yellow, according

bl? GILBERT C. BOURNE.

to the species and the place from which the spicule is taken in
each species. The colour is deepest along an axial line, and is
considerably lighter or altogether absent in the more peri-
pheral parts of the spicule. Most of the spicules show traces
of a dark core running along their axes.

The structure of such a spicule may be studied by transverse
and longitudinal sections, by crushed and ‘ etched” prepara-
tions. <A transverse section is shown in fig. 9. Centrally
there is alittle cluster of dark pores, which comparison with a
longitudinal section shows to be the expression of longitudinal
canals running along the axis of the spicule. Around this are
concentric striations, which become darker and more evident
as they approach the periphery. In some of these minute
spaces may be detected. The most characteristic feature in
the section is the presence of a number of radial cords, some
of which are simple, others forked. They originate in the
very centre, starting apparently from the central axial canals,
aud they end peripherally in the warty projections at the
surface of the spicule. On examining the section by polarised
light one finds, when the Nicols are crossed, that the whole
section is bright, but that the radial cords stand out brighter
than the remainder when in certain positions. Fixing one’s
attention on a single cord and the tissue immediately con-
tiguous to it, one finds that as the object is rotated the cord
becomes alternately brighter and darker without ever being
wholly obscured, whilst there is no great change in the sur-
rounding substance, though ill-defined dark shadows sweep
across it. In certain positions the radial cords, besides being
bright, show brilliant iridescent colours, due to interference.
I gather from this that the whole structure, cords and ground
substance alike, is crystalline, and that the axes of the
crystals composing the cords are situated in a different plane
from those composing the ground substance. Examination of
a longitudinal section (fig. 10) shows the structure of ground
substance and cords even better than a transverse section.
The axis of the spicule is occupied by a number of very dark
lines diverging at very acute angles from the centre. The

CALCAREOUS SKELETON OF THE ANTHOZOA. 5io

rest of the spicule exhibits longitudinal striations, more con-
spicuous and closer packed together in some places than others,
so that the whole section has a banded appearance. In the
section figured one may distinguish (1) a medullary portion
surrounding the axial diverging lines, which is very finely
striated and of a deep pink colour. Outside this is (2) an
intermediate portion of lighter substance, coloured a faint
pink, and showing few but distinct longitudinal striations. It
should be observed that the intermediate portion ends on the
left-hand side of the figure in a blunt point, and that the
dark axial lines diverge in a brush-like manner as they ap-
proach the point. Both the medullary part, and the inter-
mediate part forming the core of the spicule, have the shape
and characters of spicules enclosed within spicules, and are
without doubt the expression of former stages of growth.
(3) The cortical portion, as is shown by its regular longitudinal
bands, has clearly been added in successive layers to the inter-
mediate portion. It consists of alternately darker and lighter
bands, the darker bands being due to the presence of more
numerous dark longitudinal striz. The complete correspond-
ence between the longitudinal and transverse sections is
evident, the concentric lamine of the latter being represented
by the longitudinal bands of the former.

The radiating cords are well exposed in the longitudinal
section. It should be noticed that none of them are forked;
their bifurcation occurs only in the transverse plane, and
cannot be seen in longitudinal section. Each cord starts from
the dark axial lines, diverging but slightly from the longitu-
dinal axis at first, but soon turning sharply outwards to run
at right angles to the surface, on which it emerges as a warty
prominence. A spicule which has been partly ground down
and afterwards crushed is seen to be composed of a number of
longitudinal fibrous strands, interrupted here and there by the
radial cords which traverse them at right angles. Each strand
is composed of numerous fibres, which run nearly parallel to
one another, and parallel with the long axis of the spicule.
Some of these are isolated in a crushed specimen, and each when

514 GILBERT C. BOURNE.

isolated behaves as a single crystal under the polariscope.
The fibrous structure, however, is better shown in etched
specimens.

A spicule of Spongodes, which has been carefully treated
with very dilute acid, breaks up into bundles of exceedingly
fine acicular crystals, arranged with their long axes nearly
parallel to the long axis of the spicule; they are not quite
parallel, but are fitted together and interwoven so that any
one bundle of fibres gives brilliant interference colours when
viewed through the polariscope. The radial cords are not
easily studied in etched specimens, but it can be seen that the
longitudinal feltwork of crystalline fibres is interwoven with
the radial cords much as a fabric would be which had very few
and coarse strands in the warp, and very numerous and fine
strands in the woof.

When a longitudinal section of a spicule is examined by
polarised light with the Nicols crossed, the radial cords are
seen, in certain positions, to stand out brightly; whilst the
ground substance is nearly dark, though lit up here and there
by iridescent tints. As one rotates the section the radial
cords become dark, whilst the ground substance, still retaining
its iridescence, becomes lighter, and these alternations take
place regularly for every 90° through which the section is
rotated. It has been shown that the ground substance is
composed of nearly parallel longitudinal crystalline fibres, and
it may be inferred that the radial cords are composed of similar
crystalline fibres, whose axes are at right angles to the fibres
of the ground substance.

A single crystalline fibre measures some 0:03 mm. in length,
and has the optical characters of a single crystal, being extin-
guished in four positions at right angles to one another when
rotated between the crossed Nicols.

As none of my sections were thin enough to obtain a section
of a radial cord with two plane surfaces, and as the cords are
neither straight nor of uniform diameter (see figs. 9 and 10),
but swollen in places so as to be almost moniliform, they ex-
hibited interference phenomena under polarised light in every

CALCAREOUS SKELETON OF THE ANTHOZOA. 515

position of the Nicols ; these of course were due to the irregu-
larity of the crystalline fibresof which the cords are composed. I
have dwelt at some length on the structure of the inorganic part
of the spicule of Spongodes, because I am able to show
that all its details are moulded, as it were, upon an organic
matrix.

Kolliker, as has been stated above, denied the existence of
axial fibres in the spicules of Aleyonaria. Some fortunate
preparations of the partition walls of the stem canals in
Ammothea arborea and Siphonogorgia mirabilis re-
vealed the presence of an organic filament in the spicules of
the partition walls of these species. Following up this obser-
vation I was able to discover similar axial and also radial fibres
in the spicules of Spongodes, but not without considerable
trouble. Kolliker remarked that hydrochloric acid gave much
more uncertain effects than acetic. I found that treatment with
even very dilute solutions of hydrochloric or nitric acid removed
all traces of the axial organic basis. Decalcification with picric
or very weak picro-nitric acid had the same effect, but very
careful decalcification with a ‘5 per cent. solution of acetic acid
and subsequent staining with aqueous safranin and _picro-
nigrosin brought out the structure shown in fig. 12. The
spicules, which were removed with needles from the colony,
always had a certain amount of mesogloea adherent to them.
This was stained blue by the picro-nigrosin, and thus differen-
tiated from the spicular sheath, which was stained crimson by
the safranin. The sheath retained more or less faithfully the
shape of the spicule, and the warts or projections due to the
emergence of the radial cords on the surface were clearly
seen. I could detect no trace of fibrillar or other structure
in the uninjured sheath, though it broke up readily into
longitudinal fibrille when torn or otherwise injured. Within
the spicule sheath is an arrangement of organic fibrillz,
represented as faithfully as possible in fig. 12. These
fibrille are stained deep orange by safranin, and stand out
conspicuously from the spicule sheath. The axis of the
spicule is occupied by a dense feltwork of longitudinal fibrillz,

516 GILBERT C. BOURNE.

whose general direction is parallel to the long axis of the
spicule. There can be no doubt that these fibrille occupy the
spaces represented by the axial dark lines or dots in longi-
tudinal and transverse sections. From the central bundle
fibrille diverge outwards, and after making several connections
with other fibres, whose direction is approximately longitudinal,
they run straight towards the surface of the spicule, meeting it
at right angles and ending in the warty projections on the
surface of the sheath. The whole structure will be best under-
stood from an examination of fig. 12. It will be seen that what
I have described as the medullary portion of the spicule is
occupied by a strand of closely apposed fibres forming a net-
work, whose meshes are greatly extended longitudinally ; that
what I have described as the intermediate portion is occupied
by a much more open network of fibrillz, and that the cortical
portion is traversed by the radial fibrille, connected only here
and there by alongitudinal fibril. The correspondence between
the organic and inorganic elements 1s obvious, and it is clear that
the crystalline structure of the whole spicule is dominated by
the organic matrix of fibrille. The needles of calcium car-
bonate are arranged parallel to the organic fibrils, and this
circumstance explains why the long axes of the crystalline
fibres composing the radial cords are set at right angles to those
composing the ground substance. Professor Sollas kindly
determined the specific gravity of the Spongodes spicules for
me, and found it to le between 2°63 and 2-7. They appear,
therefore, to consist of calcite (sp. gr. 2°7), and the lightness
of some of the spicules was probably due to their still retaining
traces of organic matter in spite of prolonged treatment with
Kau de Javelle.

In the case of the large spicules of Spongodes, Ammothea,
and Siphonogorgia I could find no trace of a cell or cells en-
closing the spicule. The last named are deeply embedded in the
mesogloea, and no trace of cellular structure could be discovered
in connection with them. There can, however, be no doubt
that they were originally developed, like the spicules of
Alcyonium, Gorgonia, and Clavularia, within cells, and

CALCAREOUS SKELETON OF THE ANTHOZOA, 517

indeed I have observed the small spicules from the body-walls
of the polyps of Spongodes to be enclosed in cells. All that
has been stated of the spicules of Spongodes is true, mutatis
mutandis, for the spicules of Siphonogorgia, and Kdlliker’s
figures of the spicules of Eunicea leave no doubt that the same
is true of this genus also.

It is of importance to my present argument that the spicules
of the Alecyonaria show a definite and complex crystalline
structure, the details of which are, indeed, moulded upon and
dominated by an equally complex organic matrix, but are not
the expression of any particular arrangement of the cells from
which the spicules are formed. It cannot fora moment be sug-
gested that the radial cords and their warty emergences, or the
longitudinal bundles of crystalline fibres, are formations due to a
particular arrangement of calcified cells. The spiculoblast forms
an organic matrix, and secretes carbonate of lime. The latter
seems to be crystallised out, pari passu with the growth of
the former, and the crystalline growth conforms to the organic
growth, but that is all.

I have elsewhere (4) given some details of the structure of
the skeleton of Heliopora cerulea. As this Aleyonarian
differs from all its allies in having not a spicular but a so-called
lamellar skeleton, it seemed to me that a renewed investiga-
tion of its structure might throw some light on the questions
raised by Mrs. Gordon’s paper. Nor have I been disappointed.
The skeleton of Heliopora proves on closer examination to
be remarkably instructive, and to throw considerable light on
vexed questions regarding the Madreporarian skeleton. In
examining Heliopora I had recourse to flat and macerated
preparations of the growing tips of the corallum. These were
always stained with picro-carmine. I also made sections of
the growing tips after very careful decalcification with 1 per
cent. acetic acid, staining my preparations with picro-carmine
and picro-nigrosin or with Heidenhain’s iron hematoxylin
followed by acid fuchsin. I cut some sections of hard and
soft parts together, the fragile and delicate character of the
corallum at the growing tips enabling me to do this without

518 GILBERT ©. BOURNE.

entire destruction of the soft tissues, and for the study of the
corallum itself I made use of sections, minute transparent
fragments, crushed and “ etched ” preparations.

Most careful examination with high powers (Zeiss, ;4, oil
immersion, with comp. oculars 4 and 8) convinced me that
there is no question of any “ spicular ” structure in Helio-
pora. The corallum is, as I have before stated, a secretion
product of a definite layer of cells derived from the ectoderm,
which I have called, like those of Madre poraria, calicoblasts.

So far as the derivation and general arrangement of the
calicoblasts is concerned I have nothing to add to what is
contained in my previous paper, but improved histological
methods and the use of higher powers of the microscope
enable me to add a good deal in the way of detail. The cali-
coblasts form a somewhat thin covering over the coenenchymal
tubes, and the proximal moieties of the zooids, and are every-
where present between the living tissues and the corallum.
But they seldom are arranged in a single layer. In some
places, it is true, only a single row of cubical or fusiform cells
of coarsely vacuolated structure with numerous granules can
be distinguished. In such cases the calicoblasts always appear
to have an external limiting membrane, turned towards the
corallum, and this membrane, like the mesogloea, stains bright
blue with picro-nigrosin. But in all places where the tissues
are well preserved, and show no indications of having been dis-
placed by the action of bubbles formed during decalcification,
a layer of flattened cells, fusiform in section, can be distin-
guished outside of the external limiting membrane, and there-
fore abutting on the corallum. These are shown in fig. 13,
ca. e. They are found in this particular condition only in
those places where coral secretion is least active. At the
bottom of the superficial canals, where active centrifugal
erowth of the corallum is taking place, and at the lower blind
ends of the coenenchymal tubes and zooids (the regions where
the tabule are formed), the calicoblasts form a layer several
cells deep, as I have shown in my previous paper (loc. cit., pl.
10, fig. 4). In these regions the calicoblasts are grouped

CALCAREOUS SKELETON OF THE ANTHOZOA. 519

together in a manner which will be best understood by refer-
ence to fig. 20. The cells are very irregular in shape, and are
separated from one another by considerable spaces. Some of
the cells nearest to the edges of a group appear to be under-
going disintegration. In these groups of calicoblasts two
other kinds of elements may be distinguished. The first kind
comprises the “ ovoid bodies,” which I have already described
in Alcyonium digitatum. These bodies, which form a
constant element among the calicoblasts of Heliopora, do
not differ in any essential particulars from the similar bodies
in Aleyonium, except that, as is shown in fig. 14, some of
them show a distinct vacuolated structure. Though I have
devoted much time to them, I have been unable to throw any
light on their history or function; they may frequently be
observed in close association with the mesogloeal processes
which I am about to describe (vide fig. 18).

Amongst the calicoblastic groups structures such as are
represented in fig. 15 constantly occur. They consist of a
much vacuolated cell with a nucleus and central contents,
stained blue with picro-nigrosin, and exhibiting a concentric
striation which may be due to the presence of a feltwork of
fibrille. These are the earliest stages of structures which are
clearly homologous with the striated calicoblasts of von Heider.
They are shown in a somewhat later stage of development in
fiz. 16, and in later stages in figs. 17, 18, and 19. In fig. 16
two such structures are shown: one is so placed that its
nucleus is clearly visible; there is no longer any trace of the
surrounding layer of vacuolated protoplasm, and it appears as
an ovoid body with contents feebly stained by picro-nigrosin
and traces of striation. The other has no nucleus—probably
it was not included in the section,—but it has a clearly
defined striation along its lower margin, the stria in this case
being stained by the picro-carmine. It is noticeable that both
the structures lie outside of the layer of calicoblasts. Fig. 17,
which is an optical section drawn from a flat preparation,
shows a further stage of development of one of these structures,
which I shall henceforth call desmocytes. The previously

520 GILBERT C. BOURNE.

ovoid body has become somewhat flattened, and a process of
mesoglea, apparently formed by the cells contiguous to it,
stretches downward towards the mesogloeal lamina, to which,
however, it is not yet united. A somewhat later stage is shown
in fig. 18,in which there are several such mesoglceal processes,
apparently in course of formation from the cells overlying
them, and the final stage is shown in fig. 19. In the last-
named figure a shallow cup with irregular margins and with
striz deeply stained with picro-carmine is seen to be attached
to the mesogleal lamina by a band-like process of the latter.
To the edges of the cup is attached an external cuticular
membrane, stained blue by picro-nigrosin. The desmocyte
shows different staining reactions according as the section has
been stained for a short or long time in picro-nigrosin. It
always stains in picro-carmine: if it is left for a short time in
picro-nigrosin, the carmine is not displaced ; if for a long time
the picro-nigrosin displaces the picro-carmine and the desmocyte
and its striations are stained blue. The same is the case with
the similar structures in Madreporaria.

I was long inclined to attribute to these structures the pre-
dominant share in the formation of the corallum, the more so
because their earlier ovoid shape and their final cup shape sug-
gested that they were similar to goblet-cells, the cup shape
being the expression of a goblet whose contents had been voided.
But further study convinced me that this was not the case.
The desmocytes are most abundant and best developed in places
where coral growth is least active ; only their earlier stages are
to be observed among the groups of calicoblasts in regions of
active coral growth. ‘They are scarce relatively to the granular
calicoblasts, and in my flat preparations of hard and soft parts
at the growing points of the corallum I could not detect any
of them amongst the numerous calicoblasts clothing the ob-
viously newly formed points of coral. Moreover I am of the
opinion that the corallum is formed as a result of the disinte-
gration of calicoblasts. This is suggested by fig. 20, and fig, 21
is a drawing made from a partially decalcified preparation
stained with eosin and methylene blue. The calicoblasts are

CALCAREOUS SKELETON OF THE ANTHOZOA. 521

seen to be full of eosinophilous granules lying in small vacuoles.
At the bottom of the drawing is a sort of loose feltwork of un-
stained tissue, which is probably the organic matrix of the
corallum. Minute crystals are adherent to this feltwork, and
in places it is covered with minute black granules, which are
invariably present on similar organic remnants in partially de-
calcified specimens. ‘To the left are some bundles of acicular
crystals of carbonate of lime, broken up by the action of the
acid. It appears as if some of the calicobiasts in the figure were
in the course of disintegration, and I have observed this break-
ing up of the cells not only in many other instances in Helio-
pora, but also in many different species of Madreporaria.
In the latter case the disintegration is not a result of macera-
tion, for it is conspicuous in preparations of Caryophyllia,
freshly killed by various methods, in which the remaining
tissues were excellently preserved. Though I have hunted
through scores of preparations with the polariscope, I have
never been able to find in Heliopora any trace of the formation
of spicules or of crystals within cells, such as may be readily
discovered in Alcyonium or Gorgonia. I have specially
examined the desmocytes, for I was for a long time inclined to
the opinion that von Heider and Mrs. Gordon must be right,
and that I should find evidences of spicule formation within
these structures. But my results were always negative. The
calcareous skeleton of Heliopora is not formed from spicules
developed within cells, but is a crystalline structure formed by
crystallisation of carbonate of lime, probably in the form of
aragonite, in an organic matrix produced by the disintegra-
tion of cells which I have described as calicoblasts.

This view is confirmed by a minute study of the corallum
itself. In carefully made maceration preparations of hard and
soft parts the most delicate growing tips of the corallum are
preserved uninjured, and in close contiguity to the soft tissues
from which they originated. Figs. 22 and 238 are careful
drawings made with the camera lucida of the spinous projec-
tions at the tips of a growing frond of Heliopora. They are
drawn under different magnifications; fig. 22 shows more

522 GILBERT C. BOURNE.

clearly how the component crystals are arranged along the edge
of such a spine; fig. 23, which is a surface view, shows the
arrangement of the crystalline components on a larger scale.
Both drawings might be taken for representations of an
ordinary inorganic crystalline deposit. In fig. 22 the crystal-
line elements are seen to diverge, more or less regularly, from
a centre of growth, and this is the usual condition; but in
fig. 23 the components are arranged pell-mell, the long axes of
the crystals crossing one another at every angle: only at the
right-hand edge is there an indication of a more SEL! diver-
gent growth from a centre.

Although the structure of the corallum of Heliopora is
much coarser than that of any madrepore which I have
examined, it is difficult to isolate the crystals, so as to
identify the crystal system to which they belong. On crush-
ing, the corallum breaks up into fragments, some of the
most regular of which are represented in fig. 24, a, 6, and
c. The smallest fragments, such as 6 and c, behave under
polarised light like single crystals, being regularly extin-
guished in four positions 90° apart when rotated between
the crossed Nicols. Treatment with dilute acids breaks
up the crystals into needle-like fibres resembling those of
Spongodes, and every such fibre behaves as a single crystal
under the polariscope.

There is no evidence whatever that the crystalline elements
are formed within cells. In my previous paper I gave a
drawing of a section through the corallum at the tip of a
growing frond of Heliopora. The arrangement of the crystal-
line fibres was only indicated by faint diverging lines. By
preparing thinner sections, and examining them with higher
powers and polarised light, I was able to make out the dis-
position of the crystalline elements sufficiently clearly. The
coarser structure of Heliopora makes it a good object for
study. Fig. 25 is a representation of a portion of a transverse
section magnified 420 times. The first thing to be noticed is
the absence of dark lines or “ centres of calcification,” such as
occur in Madreporaria, and have been so well figured and

CALCAREOUS SKELETON OF THE ANTHOZOA. 523

described by Mrs. Gordon. But centres of calcification are
clearly present, though not represented by dark spots. The
section clearly shows radiating systems of crystals diverging
from centres a, a, the latter appearing white and translucent
when viewed by transmitted light. From these centres crys-
talline fibres diverge upwards and outwards, spreading out to
greater distances from the centre in three positions, which are
approximately at 120° to one another. These extensions meet
and dovetail with similar extensions from adjoining centres,
and thus produce the characteristic skeleton of the ccenen-
chymal tubes. The section figured is not from the extreme
growing point, but rather below it. At the extreme growing
point it would be seen that each centre is really trifid, pro-
ducing the arrangement which I have described before, and
which I need not further particularise. A little below the
extreme growing edge the separate centres unite in the angle
formed between three adjacent coenenchymal tubes, and there
form a more or less Y-shaped centre of calcification, from
which the crystalline fibres diverge in three directions, meeting
in sutural junctions with similar diverging systems with the
diverging fibres from adjacent systems. The whole structure
is displayed in the annexed diagram. From this it is seen

Y JI;

UMMM MMM

on
x)
qt

MUI

Diagram illustrating the arrangement of the crystalline fibres forming the
corallum in Heliopora cewrulea. ct. Coenenchymal tubes. ¢r. Tra-
becule. s. Sutures.

von. 41, pART 4,—NEW SERIES. 00

524 GILBERT C. BOURNE.

that at some little distance below the surface the coenenchymal
tubes are bounded by trabecule, which run at 90° to the
surface of the colony. Each trabecula is triradiate in trans-
verse section, and is seen in longitudinal section through its
centre to be composed of crystalline fibres diverging upwards
and outwards from a centre of calcification. In any given
transverse section the fibres emerge at the surface of the
section, and thus a series of concentric lines is formed around
each centre of calcification, which are exactly similar to the
“* growth-lines ” described by Mrs. Gordon in various Madre-
porarian sections. It is quite clear, from what has gone
before, that these concentric striations in Heliopora are
not the expression of layers of calcified cells which have
been added to the thickness of the corailum. I shall hope to
show that the same is the case in the Madreporaria.

Owing to the comparative coarseness of its component crys-
tals, there is very little doubt as to the cause of the concentric
markings in the Helioporan skeleton, and for the same reason
the markings themselves are not so delicate, and betray their
origin more clearly than do those of Madreporaria.

Professor Sollas kindly determined the specific gravity of
the skeleton of Heliopora for me, and found it to be 2°82,
calcite being 2:7 and aragonite 29. The skeleton of Heliopora
therefore appears to be composed of aragonite, and not of
calcite, as are the spicules of other Alcyonaria. The difference
between the specific gravities of the Helioporan skeleton and
aragonite is to be accounted for by the presence of a consider-
able amount of organic matter in the former. This organic
matter, as | showed in my previous paper (4), is associated
with the blue colouring matter characteristic of the species.
The coralla of Euphyllia and Madrepora are, by Professor
Sollas’s determination of specific gravities, of 2°77 and 2°78
respectively, thus standing nearly halfway between calcite and
aragonite. In this case, too, the difference between the coral
fragments and aragonite is probably due to the fragments being
full of organic matter, particularly of the filaments of a parasitic
boring fungus, and a boring sponge of the genus Clione.

CALCAREOUS SKELETON OF THE ANTHOZOA. 5)

i)
or

To sum up the characters of Heliopora :

1. The corallum is composed, like that of Madreporaria,
of vertical or nearly vertical trabecule; but instead of
the trabecule being in apposition by their lateral surfaces,
as in Madreporaria, they are separated by ccenenchymal
tubes. Each trabecula gives out arms at angles of 120°,
and the arms unite suturally with similar arms from adjacent
trabecule.

2. Each trabecula consists of crystalline fibres diverging
outwards and upwards from a centre of calcification situated in
the centre of the trabecula.

3. In transverse section the emergent crystalline fibres give
rise to the appearance of striz concentric with the centres of
calcification. Such striz are most conspicuous in the secondary
deposits in the interiors of the calyces.

4, The separate crystals or bundles of crystalline fibres are
not formed within cells, but are formed by crystallisation in con-
nection with an organic basis produced by the disintegration of
calicoblast cells.

5. Special structures, here called “‘ desmocytes,” are formed
amongst the calicoblasts. They take their origin from certain
cells in the calicoblast layer, and become secondarily attached
to the mesogloea by processes. Their function is to bind the
soft tissues to the corallum.

MADREPORARIA.

As regards the shape and position of the calicoblasts and of
the desmocytes there is no great amount of variety among the
Madreporaria, though it will appear that my observations con-
firm previous statements by Fowler and myself as to the
presence of unusually elongated calicoblasts in Mussa and
Lophohelia. I have examined a considerable number of
species, including Caryophyllia Smithii, Mussa distans,
Lophohelia prolifera, EKuphyllia (sp. incert.), Porites
bulbosus, Madrepora subulata, rosacea, and hyacin-

526 GILBERT C. BOURNE.

thus. The results obtained were so nearly the same in all that
I will not give details of all the species examined ; to do so
would involve a wearisome repetition.

In a previous paper (2) I described the calicoblasts of Mussa
distans as being, in the regions of most active coral growth,
long, narrow, and columnar, but elsewhere rounded or poly-
gonal, with nuclei which stain faintly in borax carmine. At the
same time I described structures which I identified with the
striated calicoblasts of von Heider, but now with the desmo-
cytes of Heliopora.

As it was very difficult to get good preparations of calicoblasts
from decalcified specimens of Mussa, and as they could scarcely
be studied in sections of hard and soft parts together, I had
recourse to the following method :—Thick sections of hard
and soft parts together were prepared in the usual manner by
grinding. After examination of these sections by low powers
suitable parts were cut out and the balsam was dissolved out by
xylol. They were then placed in a very dilute solution of
acetic acid, and the soft tissues were carefully removed as soon
as decalcification had proceeded far enough to loosen their hold
on the corallum. ‘The tissues thus separated were either
arranged as flat preparations, or were again embedded in paraffin
and cut into very thin sections. For staining I used picro-car-
mine and nigrosin, or iron hematoxylin and acid fuchsin.

Fig. 27 is a drawing of part of a transverse section made as
above. Sep. and Dissep. mark the positions occupied by a
septum and a dissepiment respectively. Mes. are mesenteries.
The calicoblasts (ca.) are seen to be tall vacuolated columnar
cells, with distinct nuclei. At the places where the mesen-
teries join on to the tissues investing the corallum are seen
groups of structures marked dc. They are the striated calico-
blasts of von Heider and Mrs. Gordon. AsI have no doubt
of their homology with similar structures in Heliopora, I shall
call them desmocytes. This section is characteristic in so far
as the position of the desmocytes is concerned. They are
always grouped in the greatest numbers about the insertions of
the mesenteries, a fact which has already been recorded by von

CALCAREOUS SKELETON OF THE ANTHOZOA. 527

Heider, Fowler, and myself. The fact is shown even better in
a flat preparation of the edge-zone of Caryophyllia. The
desmocytes appear as a number of thick bands, corresponding
to the extra-thecal continuations of the mesenteries; the
spaces between are occupied by the irregular vacuolated calico-
blasts which will be described further on. I may here insist
on the arguments which were put forward with great clearness
by Fowler (9). The lines along which the desmocytes occur
correspond neither to the septa nor to the cost, but in corals
in which there is a “ pseudotheca’’ they correspond to the lines
along which adjacent septa have met and fused. These are not
lines of active growth, but contrariwise The regions of active
growth, the septa and costz, are not covered by desmocytes, but
by a quite different kind of cells, whose characters vary accord-
ing as the position is or is not one of active coral growth. The
position of the desmocytes affords strong presumptive evidence
against their being the active agents in coral secretion. Were
this the case the corallum would grow most rapidly not along
the edges of the septa or costz, but along the lines between the
septa and cost where the mesenteries are attached to it. But
these are exactly the places where its growth is least notice-
able. A few desmocytes, it is true, may be observed abutting
on the lateral surfaces of the septa. Such a one is shown in
fig. 27, on the left-hand side of the drawing; but in such
places the desmocytes never form a continuous layer, but are
placed at intervals far apart from one another.

A surface view of a group of desmocytes is given in fig. 35,
and a section made at right angles to their surfaces is repre-
sented in fig. 28. In the latter figure the striz are seen to be
dark branching fibres, which run with a wavy course from the
base to the surface of the desmocyte. These strize stain deeply
with iron hematoxylin, eosin, Bismarck brown, picro-carmine,
or picro-nigrosin. In fact, almost any diffuse stain will bring
them out. Neither their shape nor their staining properties
suggest their being spicular, as von Heider suggested, nor does
a surface view lend any further support to the idea. Viewed
from above, the desmocytes have a curious “ machine turned ”

528 GILBERT C. BOURNE.

appearance, due to the arrangement of the fibres, which will be
best understood by reference to the figure. Both the surface
view and section show that there is a small remnant of proto-
plasm with a nucleus in connection with each desmocyte. Asa
rule the desmocyte is cup-shaped, just as in Heliopora; and in
such a case its centre appears dark when viewed from above.
I have examined many hundreds of these desmocytes, freshly
removed from the coral without decalcification, to see whether
I could find any trace of crystalline structures in them with
the aid of the polariscope. I have never found any
either in Mussa, Euphyllia, Lophohelia, Madrepora,
Astrea, Fungia (all of which were spirit specimens), or in
the freshly removed tissues of a newly killed Caryophyllia.
The striations, whatever they are, are not due to crystals of
carbonate of lime.

It would be tedious if I were to give details of structure and
position of these desmocytes in all the forms which I have exa-
mined. Fig. 27 shows their relation to the corallum in Mussa;
fig. 83 shows their relation to the peripheral parts of
the mesenteries in the edge-zone of Kuphyllia; fig. 42 shows
one of them in a radial connecting canal of Madrepora.
Figs. 86 to 39 show various phases of their development in
Caryophyllia Smithii. In the last-named species the
calicoblasts, except at the extreme growing edges of the
corallum, are small, highly vacuolated, and without definite
cell outlines. At a spot where a desmocyte is about to be
formed, one, two, or three nuclei become surrounded with a
mass of darker, finely granular protoplasm. The next phase
is the appearance of a band-shaped or ovoid body in the centre
of the granular protoplasm, which already shows faint signs of
striation, and stains readily with picro-nigrosin or acid fuchsin,
the striz being usually coloured by picro-carmine if this has
been used. Usually one nucleus remains in close association
with this body ; the others (if more than one combine to form
the granular protoplasmic mass) appear to be concerned in the
formation of the mesoglceal process which will join the desmo-
cyte to the mesogleal lamina. The striations next become

CALCAREOUS SKELETON OF THE ANTHOZOA. 529

more defined, and the desmocyte, which was at first separate
from the mesogloea, becomes attached to it by a process deve-
loped, as it seems, at the expense of neighbouring cells. In its
final condition the desmocyte may be of various shapes—
twisted as in fig. 39, or of a well-marked goblet shape. Ihave
no doubt that Fowler (9) was right when he ascribed to the
desmocytes the function of fastening the soft tissues to the
corallum, though they do not appear to be mere processes of
the mesoglea, as he described, but structures sui generis,
developed at the expense of cells otherwise indistinguishable
from calicoblasts.

I claim, therefore, that not only their position, but also their
microscopical structure, their development, and their behaviour
under the polariscope, show that the desmocytes are not coral-
secreting cells.

But if they are not, some further information is required as
to how the corallum is formed, and as to what are the histolo-
gical characters of the calicoblasts which form it. And the
existence of the scale-like calcareous structures described and
figured by Mrs. Gordon has to be accounted for, since it is evident
from what has preceded that these cannot be calcified desmocytes.

In order to determine these questions | made a number of ob-
servations on CaryophylliaSmithii, this being the only coral
which I could obtain living. If a Caryophyllia be killed in
a five per cent. solution of osmic acid, stained for twenty
minutes in picro-carmine, and then placed in glycerine, the soft
tissues may, with the exercise of a little care, be readily re-
moved from the exsert portions of the septa; but it is best to
cut out the septa carefully with a stout pair of scissors, and to
remove the tissues from the fragments of the septa with needles
and acamel’s-hair brush. The coralium is found to be every-
where covered by an exceedingly fine membrane, consisting on
the one side of very flat endoderm cells, forming a sort of pave-
ment epithelium. I noticed in these cells a peculiar structure,
which is shown in fig. 26, and it may be described in passing.
Each endoderm cell bears a rather long flagellum, and the
flagellum may be traced to a minute prominence in the neigh-

5380 GILBERT ©. BOURNE.

bourhood of the nucleus. On focussing carefully the promi-
nence exhibits radial striz, as shown in the figure, surrounding
a central clear space, and the flagellum seems to end in a
minute granule in the midst of the clear space. The cytoplasm
of the cells is much vacuolated, but usually there is a mass of
denser protoplasm round the nucleus in which the star-shaped
structure is situated. The invariable proximity of the star-
shaped body to the nucleus suggests that the flagellum is con-
nected with the centrosome. If this is so it has the same
relations as the flagella of the spermatids of Elasmobranchs
described by J. E. 8S. Moore. I was unable to find any example
of these cells in course of division,

These endoderm cells are placed upon an exceedingly delicate
lamina of mesoglea, so delicate that it is easily overlooked,
but it may be made evident by picro-nigrosin. On the other
side of the mesoglea—that is to say, next to the corallum—are
a number of nuclei, round and clear in appearance, staining
with great difficulty, and showing no chromatin network, but a
single nucleolus.

Surrounding each nucleus is a thin layer of finely granular
protoplasm, connected by fine processes with adjoining cells of
the same character. This is all that represents the calicoblast
layer over the greater part of the corallum, and so thin and
delicate is it that it is hardly to be distinguished in transverse
sections. But at the edge of a septum, where the sheet of
tissue is folded over it, the calicoblast layer in the angle of the
fold assumes more definite characters. It is still exceedingly
thin, but the cytoplasm surrounding the nuclei increases in
bulk, and exhibits a distinct vesicular structure in addition to
the coarser vacuoles which are abundantly developed in it.
The nuclei also show chromatin networks. Sections made
through a Caryophyllia which has been killed in a mixture of
8 per cent. formol in sea water, to which is added about 3 per
cent. of saturated solution of corrosive sublimate, and after-
wards carefully decalcified in weak acetic acid, show the same
results, and also that the calicoblast laver is more definite in
the immediate neighbourhood of the mesenteries, both in the

CALCAREOUS SKELETON OF THE ANTHGZOA. 531

edge zone and in the intra-calicular part of the polyp. (Not
all specimens of Caryophyllia have aa edge-zone, but it is well
developed in some.)

In the latter situation the calicoblasts seem to serve chiefly
for the formation of desmocytes, for they thin out to an almost
unrecognisable layer in the areas between the mesenteries.
But in the angles formed by the folds of tissue over the edges
of the septa the larger granular cells may be distinguished just
as in the macerated preparations. The remarkable thing about
these calicoblasts is that they do not form a single or a regular
layer of cells. There are, in fact, no separate cells, but there is
a layer of very much vacuolated protoplasm containing nuclei—
some I have observed undergoing division,—and the protoplasm
seems to be “concentrated,” i.e. to be more finely vacuolar
and granular in certain places. In many places, as seen in
surface view, there are hollows in the protoplasmic mass sur-
rounded by ridges of the more vacuolar protoplasm, and the
hollows are lined with finely granular cells, if one may apply
the name to such a tissue as is shown in fig. 40.

Two other structural elements could be distinguished. The
one consisted of the ovoid bodies with which I was familiar
in Aleyonarians. In Caryophyllia these generally showed
traces of a filament coiled up spirally within, and hence my
conjecture that they are degraded nematocysts. The other
structure is shown in fig. 30. It consists of a little eminence
in the vacuolar protoplasm forming a ring or half-hoop, with
a concavity in the centre, and the walls surrounding the
concavity are striated as shown in the figure. These only
occur at the sides of, and not actually on the edge of a septum,
and they are probably rudimentary desmocytes, serving the
same purpose of attaching the soft tissues to the corallum.

The soft tissues covering the pali in Caryophyllia corre-
spond exactly with those covering the septa. Among the corals
which formed a part of the collection of the late Mr. George
Brook were some very well-preserved specimens of a species of
Euphyllia. As I only have specimens embedded in paraffin
or in balsam I have been unable to determine the species ; a

582 GILBERT 0. BOURNE.

portion of a section of this coral is represented in fig. 83. The
calicoblasts have the same characters as in Caryophyllia, and
where they form a deeper layer the same incoherent vacuolar
structure is observable. Iig. 34 represents part of the tissues
clothing the upper and inner edge of a septum in this species.
The endoderm cells are well preserved, but the central mass of
calicoblasts look as if they bad undergone prolonged maceration.
I do not, however, believe that this is the case, but that the
loose vacuolar appearance of the protoplasm—one can hardly
speak of cells, for no cell outlines are visible—really represents
the structure of the calicoblast layer. lor the same characters
are present wherever the calicoblast layer attains any thickness,
and the only places where it does attain any thickness are at
the edges of the septa, and deep down in the angles between
the septa where the edge-zone overlaps the edges of the theca.
In all cases the surrounding tissues were very well preserved,
aud the state of their preservation may be judged from fig, 33.

In Mr. Brook’s collection were also many sections of
Madrepora. I am unable to determine the species for the
same reasons as stated for Kuphyllia. Iowler has spoken (7)
of a “distinct ” layer of calicoblasts in Madrepora Dur-
villei. The layer is distinct enough in my specimens so far
as the presence of nuclei and surrounding cytoplasm goes, but
there are no cell outlines, and the cytoplasm is vacuolated and
irregular as shown in fig. 42.

I have noticed the same phenomena in several other corals,
viz. extreme attenuation of the calicoblast layer in all but
certain well-defined’ places, and in those a mass of vacuolar
protoplasm of irregular outline with no cell boundaries. ‘These
characters, therefore, appear to be normal for a considerable
number of corals.

On the other hand, there are the elongate cells of Mussa, and
those described by Fowler (9) in Lophohelia prolifera.
‘To satisfy myself as to the latter species I have made a series
of preparations, but the tissues were too much injured by pro-
longed action of spirit to allow me to speak with certainty
about details. It is sufficient to say that 1 found elongate

JALGARKOUS SKELETON OF THE ANTHOZOA. 533

“ cells ” in the positions described by Fowler ; and, as far as I
could determine, the “cells” in question were not neat
columnar cells with definite cell boundaries, as figured im
Fowler’s diagram, but resembled those of Mussa (vide fig. 29),
—being, in fact, long amoeboid, vacuolated, and partially fused
masses of protoplasm containing nuclei. The cellular cha-
‘acter of the calicoblasts is, however, more evident im Lopho:
helia and in Mussa than in any other forms which I have
examined.

Fig. 29 is a representation of the elongated and quasi-
columnar calicoblasts of Mussa distans, drawn under a high
magnification, ‘The section was stained by Heidenhain’s iron
hematoxylin method. Under a lower power (420 diameters)
the calicoblasts appear to form a layer of simple columnar
epithelium, but a higher power (1000 diameters) brings out the
characters shown in the figure. The cells are confluent, both at
their bases and at their free ends, especially the latter. The
alveolar structure of the cytoplasm is well shown, and besides
the minute alveoli there are a number of small vacuoles, each of
which contains a microsome. The presence of vacuoles con-
taining microsomata is characteristic of well-preserved calico-
blasts. A comparison with fig. 40 shows that there is no
essential difference between the calicoblasts of Mussa and
Caryophyllia, though the elements are much smaller in the
latter species. Fig. 29 does not show the presence of an
external limiting membrane, but there are indications of 1t in
fig. 27, which was drawn from the same slide. I have reason
to believe that there is always an external limiting membrane
to the calicoblasts in Madreporaria, and that when it 1s absent
‘n sections it has been destroyed in the course of decalcifica-
tion. As has already been stated, a similar membrane occurs
in Heliopora. The limiting membrane is interposed between
the ecalicoblasts and the corallum, and may be regarded as a
sheath of the latter. It is clear that the carbonate of lime
seereted by the calicoblasts must pass through the membrane,
just as it has to pass through the spicule sheath in Aleyo-
narians. Throughout my researches I have looked most

534 GILBERT C. BOURNE.

carefully for evidence of the formation of spicules or crystals
within cells; but no such evidence was to be found. Were
spicules or crystals formed within cells I could hardly have
failed to find them, either in Heliopora or the Madrepo-
raria in the numerous preparations which I searched with the
polariscope. I have found crystals in certain cases both in
Heliopora and in Caryophyllia, and have nearly been
deceived by them, but they occur in the endoderm and par-
ticularly in the zooxanthelle, and I have no doubt that they
are aleuron crystalloids, though I have not specially tested the
point.

Moreover I am convinced that the structure of the Madre-
porarian skeleton does not admit of a theory of spicular
formation. I can take no exception to Mrs. Gordon’s draw-
ings and descriptions of the numerous sections, and most of
the surface views of Madreporaria which she has given. They
are faithfully delineated, and constitute an important addition
to our knowledge of corals. But I join issue with her on the
subject of the ultimate skeletal unit. She has dismissed the
“crystalline spheroid” described by von Koch in almost con-
temptuous terms. “For what is this spheroid? Is it a
spicule? We can see and separate the spicule, then why not
the spheroid? It is noteworthy that the spheroid has been
presented from first to last by one or two transverse
sections, no longitudinal section has demonstrated it.”
Mrs. Gordon’s memory is at fault. The spheroid was repre-
sented very clearly by von Koch between the base and the
surface of attachment in an Astroides larva, and the section
was a longitudinal one. Quite recently De Lacaze-Duthiers,
in a beautifully illustrated memoir on the development of
Caryophyllia Smithii and Balanophyllia regia (24),
has completely rehabilitated the crystalline spheroid or
“‘sclerenchymatous nodule,” if that name be preferred. Speak-
ing of the larva of the former species, he says, “ Le premier
dépét de la charpente caleaire se produit sous la forme d’un
semis de globules trés refringents.” His illustration (loc.
cit., pl. ix, fig. 16) leaves no doubt on the subject; the first

CALCAREOUS SKELETON OF THE ANTHOZOA. 535

traces of the calcareous skeleton of Madreporaria are what
von Koch described, spheroids of calcium carbonate. I
shali show presently that similar spheroids may be observed
on the septal surface of the adult Caryophyllia. At the
same time the ‘‘calcareous scale” figured by Mrs. Gordon
for Galaxea exists, and calls for explanation. I have said
enough to show that the scales in question are certainly not
calcified desmocytes. I have found these scales in preparations
of the vesicular endotheea of Galaxea laperonsiana made
for the purpose of verifying Mrs. Gordon’s statements, but I
have found them best developed in preparations of Madrepora
subulata, M. rosacea, and M. hyacinthus, taken from
the collection of the late Mr. Geo. Brook. In tolerably thick
sections, longitudinal or transverse, of any of these species,
the calcareous scales are very evident at the sides of the septa,
cenenchymal trabecule, and costal spines included in the
section. The outer edges of the scales are commonly coloured
bright crimson in specimens stained in borax carmine (see fig.
41), but this effect is not produced by aniline dyes, and after
some study I became convinced that there is no staining of an
organic residue, but a deposit of carmine particles on the
edges of the scales. But the whole structure is so striking
that it seems to afford positive proof of the statement that
“the growth-lamellze of Madrepora from first to last are
composed of coalesced calcified calicoblasts, which once repre-
sented living ectoderm ” (Mrs. Gordon, loc, cit., p. 213). One’s
belief is shaken, however, by the study of sections, both of
hard and soft parts together and of decalcified specimens.
The layer of continuous vacuolated protoplasm shown in fig. 42
lends no support to the conception of successive shedding off
of calcified cells from a definite cell layer, nor does a side-by-
side comparison of as much of the calicoblasts as can be seen
in a ground-down section with the adjacent corallum. A
portion of such a section, magnified 420 times, is shown in
fig. 41. And here let me say that neither in this nor in any
other sections of several species of Madrepora could I find
any trace of a continuous layer of desmocytes. Here and

536 GILBERT C. BOURNE.

there was a desmocyte, such as is shown in fig. 42, conspicuous
among the calicoblasts for its dark coiour, but of tracts of
desmocytes no trace whatever.

In the section under consideration the soft tissues have
shrunk away from the corallum. At the edge of the latter the
striated scales (Mrs. Gordon’s calcified calicoblasts) are clearly
seen, and adherent to the mesogloea, and lying on the corallum
where the latter is in contact with the soft tissues, are calico-
blasts—irregularly-shaped rounded cells, and decidedly smaller
than the calcareous scales. That the calicoblasts appear in
the section as separate cells, and not as a continuous layer of
protoplasm, is probably due to shrinkage. In other parts of
the section, where the soft tissues have not shrunk away from
the corallum, the calicoblast layer appears continuous. The
rubbed-down sections are too thick for satisfactory examination
with an immersion lens, and one cannot speak with certainty,
but in decalcified sections from the same species the calicoblast
layer always has the characters represented in fig. 42.

The other point noticeable in the section is the fact that the
surface scales on the same level as the polished upper surface
of the section appear to be emergences of tracts of fibres
which diverge from a centre of calcification which is not
included in the drawing. Mrs. Gordon’s drawings of sections
of Madrepora are represented on a much smaller scale than
mine (loc. cit., figs. 60—63), but in her fig. 63 there is a
distinct indication of the same fact. Now if her theory were
true, viz. that each scale represented a calcified ectoderm cell,
it would follow as a necessary corollary that the calcified cells
were piled upon one another in perfectly regular series, cell
fitting accurately on the top of cell, and in such a manner as
to produce the regular diverging columns shown in my figure.
This is extremely improbable, as anybody well acquainted with
the arrangement and behaviour of cells will allow. Anda
comparison with the structure of the Spongodes spicule shows
that such an assumption is unnecessary for the explanation of
the structure. In the Spongodes spicule the ground substance
is made up of regularly disposed concentric lamellae, and each

CALCAREOUS SKELETON OF THE ANTHOZOA. 537

lamella consists of strands of crystalline fibres which have a
general direction parallel with the long axis of the spicule. I
cannot say exactly how this arrangement was brought about,
but it certainly is not due to the apposition of calcified cells,
for the whole spicule was formed inside a cell or ceenocyte, and
was covered in all. stages of its growth by a spicular sheath
of organic matter. In fact, the spicule was, from its early
origin, separated from the protoplasm which elaborated the
material necessary for its further growth by a layer of some
cuticular material.

The case in the Madreporaria, and alsoin Heliopora,
appears to be somewhat parallel. Fig. 19 shows that in
Heliopora a cuticular external limiting membrane extends from
the edges of the desmocyte over the adjoining calicoblasts.
Figs. 27, 33, and 43 show the same phenomenon in Madre-
poraria. And there are numerous indications in flat prepara-
tions that the calicoblast layer both in Heliopora and
Madreporaria is separated by an external limiting membrane
from the corallum. I have frequently noticed at the edges of
the septa of Caryophyllia that there was a space between the
edge of the corallum and the overlying soft tissues, and that
this space was occupied by a colloid substance in which minute
particles could be detected. I was for along time doubtful
whether this appearance was deceptive or not, but studies of the
growth of costal spines in sections through hard and soft parts
of the different species of Madrepora named above have settled
the question.

Fig. 43 represents a costal spine with its adherent tissues in
M. subulata. At the bottom of the figure is a dark “centre
of calcification,” from which lines, representing fibro-crystals,
diverge to the surfaces of the spine. At the tip of the spine
its contours become indefinite, and instead of the compact
crystalline arrangement we have separate diverging fibres, and
outside of these, minute particles which show a tendency to
arrange themselves in lines. At the base of the spine calico-
blasts may be seen lying close to the corallum. Towards the
apex of the spine they no longer lie close to the calcareous

538 GILBERT C. BOURNE.

deposit, but are seen to be separated from it by a very delicate
membrane, which is continued beyond the apex of the cal-
carcous spine as an unquestionable sheath. Outside of the
calicoblasts is the mesoglea, and outside of this again the
endoderm represented diagrammatically. Precisely the same
features are shown in all the costal spines of this and several
other sections, this particular spine being selected for repre-
sentation because the sheath or membrane between the calico-
blasts and the calcareous spine is very distinct. There can,
I think, be only one interpretation of the phenomena. The
spine grows by the addition of minute particles of carbo-
nate of lime, which crystallise out of an organic liquid matrix
secreted by the calicoblastic layer. These particles attach
themselves to the crystalline structure already present, and
become oriented conformably with it, so as to continue the
pre-existent crystalline figure in the same mode and in the
same direction. If anybody doubts the possibility of this, let
him study the formation and growth of crystals forming in a
concentrated solution of such a substance as potassium sulphate
to which a small amount of some colloid substance, such as
mucus or gelatine, has been added. A high power of the
microscope should be used. As the solution dries upon the
slide minute particles make their appearance: at first they do
not behave as crystals under polarised light, nor do the
particles in the apparently colloidal material lying between
the calicoblasts and the corallum of Madreporaria. After a
time the particles aggregate themselves at the growing point
of a crystal already formed : they increase in number, and then
_ slowly, imperceptibly, they orient themselves in the direction
of the axis of the crystal and become a part of it, producing
figures exactly like that shown in fig. 43. In the case of
the inorganic crystal, the particles, when they are oriented but
not yet attached to the crystal, show up with a pale lambent
light when viewed through crossed Nicols, the rest of the
crystal standing out brilliantly. Exactly the same phenomenon
is observed in the growing tip of a costal spine of Madrepora.

The conclusion is irresistible, though I only arrived at it

CALCAREOUS SKELETON OF THE ANTHOZOA. 539

after holding almost every other conceivable hypothesis on the
subject. The corallum of the Madreporaria and of Heliopora,
and the spicules of other Alcyonaria, are crystalline growths
formed by the deposition of needles of carbonate of lime in a
colloid matrix, 2nd obeying the ordinary laws of crystalline
growth in detail; but the general arrangement of the fasciculi
of crystals is dominated, in some manner of which we are
ignorant, by the living tissues which clothe the corallum.

We have seen that the characteristic structure of certain
Alcyonarian spicules is moulded upon an organic pattern of
fibres. I have looked for a similar organic pattern in Madre-
poraria, but without success. The corallum of Madrepores is
invariably penetrated by the mycelium of the parasitic fungus
which I have elsewhere described under the name given to it
by W. B. Carpenter, Achyla penetrans. I believe, how-
ever, that it does not really belong to the genus Achyla.
However, the presence of these fungus filaments have pre-
vented my making any trustworthy observations on the organic
matrix of the Madreporarian skeleton. I have slowly and
carefully decalcified sections of coralla on a slide, and have
invariably obtained a plexus of mycelial filaments, among
which I could detect fine threads whose arrangement I could
not determine owing to the presence of the fungus. I am of
the opinion that there is an organic pattern, probably con-
sisting of a fine network of threads analogous to those observed
in the spicules of Spongodes, but I cannot at present say
anything definite on the subject. Further, I suspect that the
dark “centres of calcification ”’ will be found to be the ex-
pression of a core of organic filaments, just as the central dark
line in the Alcyonarian spicule is the expression of the central
core of threads. In further support of the conclusions arrived
at as to the formation of the Madreporarian skeleton, I must
add that the “ scales”’? which are so conspicuous in Galaxea,
and in the genus Madrepora, as also in Mussa and some
other genera, are by no means a constant feature in Madre-
poraria. I have failed to find them in Caryophyllia,
Porites, Montipora, Lophohelia, and others. I have

vou. 41, part 4,—NEW SER. PP

540 GILBERT C. BOURNE.

examined Caryophyllia with special care. Fig. 31 is a
representation of an optical section of a septal granulation as
seen with the oil immersion. The apex of the granulation
was, of course, out of focus, and is not represented, but its
periphery is seen to be composed of a number of needle-like
crystals radiating outwards. The crystals are grouped together
in bunches, and all the crystals in a bunch diverge from an
ideal centre towards the free surface, not only horizontally,
but upwards and downwards in all directions where they can
extend. Fig. 32 is a not very satisfactory drawing of the
surface of avery thin palus of Car yophyllianear its extreme
upper edge. The surface is very irregular, and studded with
hemispheroidal projections, each of which is composed of
fibro-crystals, diverging in all directions where there is room
for their growth. This is a much more common appearance,
in my experience, than the flattish overlapping “ scales.”
Similar spheroids or hemispheroids are formed in many in-
organic solutions, and I opine that we find in these the “ crys-
talline spheroids” described by von Koch and derided by Mrs.
Gordon. It is worth noticing that the extreme irregularity of
the surface of the corallum of Caryophyllia corresponds
with the incoherent and irregular arrangement of the calico-
blasts which I have described for this species. The continued
growth and aggregation of such spheroids would give rise to
what mineralogists term a botryoidal formation, and the
Madreporarian skeleton may fairly be described as a complex
botryoidal growth, whose character has been influenced pro-
foundly by the living tissues which gave origin to it.

If my conclusions are correct the whole skeleton of a Madre-
porarian polyp has some sort of analogy toa single Alcyonarian
spicule. Both are composed of fibro-crystals having an infinitely
various, but for every species a definite and characteristic
arrangement. As the spicule is enclosed in a sheath, and
separate from the protoplasm of the cell or ccenocyte from
which it is developed, so is the Madreporarian corallum covered
with a membrane or corallum sheath, and separated from the
calicoblastic layer which gave rise to it. If there is no diffi-

CALCAREOUS SKELETON OF THE ANTHOZOA. 541

culty in admitting the fact that an Alcyonarian spicule can
grow by deposition of successive crystalline layers within its
sheath, there can be no difficulty in admitting the possibility
that the Madreporarian corallum grows in the same way, by
addition of new material elaborated by the calicoblasts and
passed through the membrane which lies between them and
the corallum. I do not wish to push the analogy further, but
only to emphasise the fact that an explanation which is satis-
factory in the case of such complex bodies as Alcyonarian
spicules, with their bewildering variety of form and their
complex arrangements of warts and spines, cannot be held
to be unsatisfactory in the case of Madreporarian coralla,
whose fundamental plan is not strikingly variable, and
whose detail is scarcely more complicated than that of the
spicules,

Of course there is no real explanation in either case. We
are as ignorant of the laws which govern the formation of these
organic crystalline growths as we are of the molecular laws
which determine why a given mineral solution shall crystallise
out according to a given system. But I enter my protest
against the discovery of a Deus ex machina in the form of
calcified cells.

I have been mainly occupied in the course of this paper in
criticising a particular statement made by Mrs. Gordon, and I
may appear to have been unappreciative of her excellent
memoir on the structure of Madreporarian corals. I must
conclude by saying that I do not conceive that her most
important conclusions are affected by the error which she has
made in the matter of the calicoblasts. The acceptance of my
views on the formation of the corallum would not detract in
the least from the importance and truthfulness of her observa-
tions on the different types of microscopical and macroscopical
structure in recent and fossil Madreporaria, nor would it make
them less useful as a basis for an amended classification of the
group. Saving this question of calcified ectoderm, and some
remarks on the anatomy of Fungia with which I shall have to
deal on another occasion, I have found her descriptions of fact

542 GILBERT 0. BOURNE.

to be remarkably accurate in the numerous cases in which I
have taken the trouble to verify them.

In conclusion, I must express my thanks firstly to Professor
Ray Lankester, in whose laboratory my investigations were
conducted ; next, to Professor De Lacaze-Duthiers for per-
mission to work in the Marine Laboratory at Roscoff, where I
was abundantly supplied with living specimens of Aleyonium,
Gorgonia, and Caryophyllia: to Mrs. Geo. Brook, to
whose kindness I am indebted for the valuable collection of
specimens and preparations of Madreporaria, without which I
could hardly have completed my work; to Professors Miers
and Sollas, who have spent much of their valuable time in
overcoming my profound ignorance on the subject of crystallo-
graphy ; and lastly, to my friend Mr. E. A. Minchin, whose
studies on sponge spicules, made in the same room in which
I have been working, have greatly influenced my work on the
Anthozoa,

List OF PAPERS REFERRED TO.

1. Bournn, G. C.—‘‘ On the Anatomy of the Madreporarian Coral Fungia,”
‘Quart. Journ. Mier. Sci.,’ vol. xxvii, n. s., p. 298.

2. Bourns, G. C.—‘‘On the Anatomy of Mussa and Euphyllia,” ‘ Quart.
Journ. Micr. Sci.,’ vol. xxviii, n. s., p. 25.

8. Bourng, G. C.—‘‘Onthe Post-embryonic Development of Fungia,” ‘Sci.
Trans. Royal Dublin Soc.,’ vol. v, ser. 2, 1893.

4. Bourns, G. C.—“*On the Structure and Affinities of Heliopora
coerulea,” ‘ Phil. Trans.,’ 1895, p. 455.

5. Epyer, V. von.— Ueber den feineren Bau der Skelettheile der Kalk-
schwimme nebst Bemerkungen tiber Kalkskelete tiberhaupt,” ‘Sitz.
der kais. Akad. der Wissensch., xev,’ 1887.

6. Fow.er, G. H.—*‘ The Anatomy of the Madreporaria,” i, ‘ Quart. Journ.
Mier. Sci.,’ vol. xxv.

7. Fowter, G. H.—* The Anatomy of the Madreporaria,”’ ii, ‘ Quart.
Journ. Mier. Scei.,’ vol. xxvii.

8. Fowier, G. H.—‘ The Anatomy of the Madreporaria,” iii, ‘ Quart.
Journ. Mier. Sci.,’ vol. xxviii.

9. Fowier, G. H.—“‘ The Anatomy of the Madreporaria,” iv, ‘Quart.
Journ. Mier. Sci.,’ vol. xxviii.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

CALCAREOUS SKELETON OF THE ANTHOZOA. 543

Fowter, G. H.—“The Anatomy of the Madreporaria,” v, ‘ Quart.
Journ. Mier. Sci.,’ vol. xxx, p. 405.

Herper, A. von.—* Die Gattung Cladocora,” ‘ Sitz. kais. Akad. Wiss.
Wien,’ |xxxiv.

Hwviper, A. voy.— Korallenstudien,” ‘ Arbeiten Zool. Inst. zu Graz.,’ 1,
1896, p. 153.

Hickson, S. J.—‘* The Anatomy of Aleyonium digitatum,” * Quart.
Journ. Mier. Sci.,’ vol. xxxvii, n. s., p. 343.

Kocu, G. von.— Das Skelet der Alcyonarien,” ‘ Morph. Jahrb.,’ iv,
1878, p. 447.

Kocu, G. voy.—‘ Bemerkungen iiber das Skelet der Aleyonarien,”
‘Morph. Jahrb.,’ v, 1879, p. 316.

Kocn, G. von.— Anatomie der Clavularia prolifera,” ‘ Morph. Jahrb.,’
vii, 1881, p. 467.

Kocn, G. von.—‘ Ueber die Entwicklung des Kalkskeletes von Aste-
roides calycularis ti. dessen morphologische Bedeutung,” ‘ Mitth.
Zool. Stat. Neapel,’ 1882, p. 284.

Kocu, G. von.—* Uber das Verhiiltniss der Skelet und Weichtheilen bei
den Madreporen,” ‘ Morph. Jahrb.,” xii, p. 154,

Kocn, G. von.—“ Die morphologische Bedeutung des Korallenskelets,”
‘Biol. Centralblatt,’ ii, p. 583.

Kocu, G. von.—** Das Skelet der Steinkorallen,” ‘Festschrift fir Gegen-
bauer,’ 1896.

Kouurxker, A. von.— Icones histologice,’ Leipzig, 1864.

Kouuxker, A. von.—* Die Pennatulide Umbellula und zwei neue Typen
der Aleyonarien,” ‘ Festschrift aur Feier des 25-jahrigen Bestehens
der phys.-med. Gesell. in Wiirzburg,’ 1874.

Kowatevsky, A., and Marion.—“ Documents pour histoire embryo-
génique des Aleyonaires,” ‘Ann, du Musée d’Hist. Nat. de Marseille,’ 1,
pt. 2, p. 20.

Lacaze-Duturers, H. pz. Faune du Golfe du Lion, Coralliaires,”’
‘Arch. Zool. expér. et générale,’ ser. 8, v, 1897.

Ocitvig, (Miss) M. M.—* Microscopic and Systematic Study of Madre-
porarian ‘Types of Corals,” ‘ Phil. Trans.,’ elxxxvii, 1896, p. 83.

Scrater, W. L.—* On a Madreporarian Coral of the Genus Stephano-
trochus,” ‘Proce. Zool. Soe.,’ 1886, p. 128.

Soutas, W. J.—‘ Physical Characters of Calcareous and Siliceous
Sponge Spicules,” ‘ Proc. Roy. Dublin Soe.,’ n. s., iv, 1888, p. 374.

544 GILBERT OC. BOURNE.

EXPLANATION OF PLATES 40—43.

Illustrating Mr, G, C. Bourne’s paper on “ The Structure and
Formation of the Caleareous Skeleton of the Anthozoa.”

PLATE 40,
Nias. l—5.—Aleyonium digitatum, Zeiss, ;45 hom. imm., c. oc. 8.

Vig. 1. A group of scleroblastic cells from the base of a polyp. sel. A
sinall selerite in course of formation enclosed in a vacuole in the
scleroblast. ov. Ovoid bodies. gv. Finely granular cells which pro-
bably form mesogleea.

Wig. 2. A seleroblast enclosing a larger sclerite; the cytoplasm is. still
abundant, and contains granules aud vacuoles,

Wig. 38. A larger sclerite enclosed in a scleroblast.

Vig. 4. ‘wo twin-sclerites enclosed in a twin cell with two nuclei.

Vig. 5. A characteristic spicule enclosed in a thin protoplasmic envelope
with two nuclei.

Via. 6.—A single seale-like spicule of Plumarella delicatissima,
Wright and Studer, Zeiss B, oc, 4.

lie, 7.—The same spicule viewed through the polariscope, showing the
dark cross due to the extinction of light by those component crystals whose
optic axes lie in four positions ab right angles to one another. Zeiss B, oc. 2.

lies. 8—12.—Spongodes, sp. ?

Vig. 8. A single spicule from the stem, showing the warty projections
and the dark axial core, ae.

Vig. 9. A transverse section of a similar spicule, showing aa, the axial
core; va, the radial cords, many of which are bifurcated; e.g, the
concentric lines of growth.

Wig, LO. A longitudinal section of a similar spicule, showing me., the
medullary portion ; ¢. p., the intermediate portion; and co. p., the
cortical portion: the other lettering as in Mig. 9.

Wig. LL. Part of a longitudinal section through a spicule which has been
crushed after grinding. The strands of fibro-crystals are clearly seen,
traversed by the radial cords, ra.

Hig. 12. Part of a spicule which has been carefully decalcified with weak
acetic acid, mg. Remains of mesoglea. sf. Spicular sheath. ov. e.
Core composed of interlacing organic threads. or. va. Radial organic
threads corresponding to the radial cords in Figs. 9 and 10.

CALCAREOUS SKELETON OF 'THE ANTHOZOA, 045

PLATE 41.

Fies. 13—25.—Heliopora cerulea, Pallas. ca. Calicoblast. ca. é.
Internal layer of calicoblasts. ca. e. External layer of calicoblasts. mg.
Mesoglea. ex. Endoderm. de. Desmocyte. ov. Ovoid bodies. 20.
Zooxanthelle.

Fig. 13. Portion of a section through the growing point of Heliopora,
showing the external and internal calicoblast layers. m. 2. Limiting
membrane formed outside the internal calicoblast layer. Zeiss, ;';
hom. imm., ¢. oc. 4.

Fig. 14. Ovoid bodies, showing nuclei and vacuoles.

Fig. 15. Karly stages in the formation of desmocytes. A striated body,
stained blue by picro-nigrosin, is contained within a much vacuolated
cell,

Fig. 16. Further stages in the development of desmocytes.

Fig. 17. A desmocyte is becoming attached to the mesoglea by a
process, p. m. /. Limiting membrane.

Fig. 18. Desmocyte in course of formation, attached by several bands to
the mesogloea.

Fig. 19. A fully formed desmocyte, showing the characteristic cupped
shape. m. /. External limiting membrane.

Fig. 20. A group of calicoblasts, some of which are undergoing dis-
integration.

Fig. 21. Calicoblasts stained with eosin, showing the eosinophilous
granules, gr. crys. f Groups of crystalline fibres adherent to a mesh-
work of organic filaments. Zeiss, ;4; hom. imm., c. oc. 4.

Fig. 22. Growing edge of a ceenenchymal spine, showing the manner in
which it is built up of numerous crystals. Zeiss D, oc. 4

Fig. 23. Surface view of part of a partition wall of a ceenenchymal
tubule from the growing edge of Heliopora. The component crystals
are arranged pell-mell in no definite order. Zeiss D, oe. 4.

Fig. 24. Isolated crystalline elements from the same preparation as
Vig. 23. Zeiss, j45 hom. imm.,, c. oc. 4.

Vig. 25. Portion of a transverse section taken about 2 mm. below the
growing edge of a frond of Heliopora. a, a. Centres of calcification,
from which erystalline fibres diverge to meet those of adjacent systems
at the sutures 6. 6. c¢. Coenenchymal tubes.

Fie. 26.—Surface view of the flattened endoderm cells forming part of the
investment of aseptum of Caryophyllia Smithii. #, Flagella, each arising
from a clear space closely contiguous to the nucleus. ‘The clear spaces are
in the centres of small asters, and appear from their position to be related to
the centrosome. The astral rays were stained as figured by safranin, but the
flagella are represented in red only for clearness’ sake.

046 GILBERT C. BOURNE.

PLATE 42,

Fre. 27.—Portion of a transverse section through Mussa distans.
Mes. Mesenteries. Sep. space occupied by a septum. Dissep. Space occu-
pied by a dissepiment. ca. Calicoblasts. de. Desmocytes. ez. HEndoderm.
m. l. Limiting membrane. Zeiss B, oc. 4.

Fie. 28.—Four desmocytes of Mussa distans, magnified 1000. x.
Nuclei with surrounding cytoplasm of the parent cells of the desmocytes.
mg. Mesogleea.

Fie. 29.—Calicoblasts of Mussa distans, magnified 1000. vac. Vacuoles
containing microsomata. mg. Mesoglea. ex. Kudoderm.

Fic. 30.—Surface view of a portion of the calicoblast layer from the lateral
surface of a septum of Caryophyllia Smithii, showing a rudimentary
desmocyte, dm. The vacuolated and pitted character of the protoplasm is
indicated. Zeiss, 4; hom. imm., c. oc. 4.

Fie. 31.—Optical section of a septal granulation of Caryophyllia, showing
that it is made up of a number of bunches of radiating fibro-crystals. Zeiss,
ji; hom. imm., c, oc. 4.

Fic. 32.—Surface view near the growing edge of a palus of Caryophyllia,
showing the hemispheroidal bunches of fibro-crystals of which it is com-
posed.

PLATE 483.

Fie. 33.—Part of a transverse section through the edge-zone of Kuphyllia.
ec. Eetoderm with two kinds of gland-cells and nematocysts. de. Desmocytes.
en. Endoderm. mes. Extra-thecal portion of a mesentery. mg. Mesogleea.
ca. Calicoblasts.

Fie. 34.—Part of a section of the same species, showing the fold of the
soft tissues over the inner and upper edge of a septum. ez. Endoderm.
mg. Mesoglea. ca. Calicoblast layer consisting of nuclei lying in a layer of
vacuolated protoplasm.

Fic. 35.—Surface view of a group of desmocytes of Caryophyllia
Smithii. Cor. Fragments of corallum adherent to the desmocytes. p. ¢.
Parent cells of desmocytes.

Fics. 836—39.—Stages in the development of desmocytes of Caryophyllia.
Lettering as in the previous figures. g/. c. in Fig. 39 is a gland-cell filled
with highly refringent globules. Such gland-cells are often very abundant in
the endoderm immediately overlying active calicoblasts.

Fic. 40.—Part of a transverse section of Caryophyllia taken at the point
where the tissues turn over the edge of the theca between two septa to form
the edge-zone. ex. Endoderm of the intra-theeal portion of the polyp.
ex’. Endoderm of the edge-zone. mg. Mesogloea. ca. Calicoblasts,

CALCAREOUS SKELETON OF THE ANTHOZOA. 547

Fie. 41.—Portion of a transverse section through the hard and soft
parts of Madrepora subulata, showing sc. The scale-like emergences of
bundles of fibro-crystals on the free surface of the corallum. 6. The bundles
of fibro-crystals as seen in section, lying parallel to one another and
emerging at the free surface to form the “scales.” (sp. is a costal spine.
ca. Calicoblasts. mg. Mesogloea. en. Endoderm.

Fie. 42.—Portion of a section through a radial ecenenchymal canal of the
same species, showing a desmocyte, dc. ; calicoblast, cv. ; and endoderm, ex.

Fic. 43,—A transverse section through a costal spine and the adjacent
soft tissues of Madrepora rosacea. Sp. Costal spive. Crys. Deposit of
erystalline fibres on its growing point. m. /. Limiting membrane. ca.
Calicoblasts. ex. Endoderm. mg. Mesoglea.

Veal

” f : - @ j
% . . te are i GV liek Kips a Tod, Ma i i
. ew 4 ; I f | i Ds ae yntens pi <4 Deh ae ty,
r) | : _ ae Mea pray Hee? ; Po tat Bh A
‘’ 4 Wake ae
i / i) ahha Me j
4 4
\ '
*
bh ’
‘
Pa
4

HAIRS OF MONOTREMES AND MARSUPIALS. 549

The Structure and Development of the Hairs of
Monotremes and Marsupials.

Part I.—Monotremes.
By

Baldwin Spencer, NIE.A.,
Professor of Biology in the University of Melbourne,

and

Georgina Sweet, M.Sc.,
University of Melbourne.

With Plates 44—46.

In his paper dealing with the structure of the bill and
hairs of Ornithorhynchus,' Mr. Poulton has described the
structure of the hair in an embryo of Ornithorhynchus
measuring 8°3 cm. long, in which the larger hairs had appeared
above the surface of the skin. Thanks to the kindness of
Professor J. T. Wilson, of Sydney, we have been able to study
much earlier stages as developed in an embryo measuring 40
mm. in length, the same embryo upon which he and Dr. Martin
have already worked when studying the structure of the bill,
To Mr. Dudley le Souef we are indebted for an embryo
measuring 77 cm., in which the hairs have not yet appeared
above the surface, and to Dr. Gregg Wilson for pieces of skin
of an embryo of the same age. In the case of Echidna we
have been able to study the structure of an embryo measuring
55 mm. in length in which the tips of the larger hairs, which
subsequently become modified into spines, have just appeared

1 «Quart. Journ, Micr. Sci.,’ vol. xxxvi, pt. 2, p. 143.

950 BALDWIN SPENCER AND GEHORGINA SWEET.

above the surface, and we are again indebted to Dr. Gregg
Wilson for pieces of skin of an embryo of the same age. Owing
to the fact that in different parts of the body the hairs are not all
developed to the same extent, we have been able to study various
stages—in the youngest one of which the follicle is only just
formed, while in the oldest the largest hairs are well developed.

We are indebted to Professor J. T. Wilson and Mr. J. P.
Hill for the opportunity of consulting literature which was
unavailable to us in Melbourne.

Whilst as the result of a considerably greater supply of
material, which is also probably in a better state of preservation
for histological work than was that available to Mr. Poulton,
we have been led to conclusions different from those to which
he came, especially in regard to the question of the develop-
ment of the hair in an open and not a closed tube—conclusions
to which, moreover, we feel sure that Mr. Poulton would have
come had he been able to study earlier stages—we desire to
express our indebtedness to the assistance which we have
received from his work, especially combined as it was with a
valuable résumé of previous work compiled by Dr. Benham,
certain of the memoirs dealt with by the latter being unavail-
able to us. We are at the present time engaged upon an
investigation of the development of hairs in Marsupials, but
have thought it better to divide our work into two parts, and
the following deals simply with the structure and development
of hairs in Monotremes.

To Professor G. B. Howes we are much indebted for his
kindness on this, as on other occasions, in revising the proof.

General Structure and Arrangement of the Hairs.

The presence of large and small hairs and their arrangement
has been dealt with by various authors, Leydig being the first
to point out that the small hairs are arranged in bundles with
a common follicular neck.

Leydig! described in Ornithorhynchus the presence of four

1 “Ueber die ausseren bedeckungen der Saugethiere,” ‘Arch. fir Anat,
Phys. und Wiss. Med.,’ 1859, pp. 677—747.

HAIRS OF MONOTREMES AND MARSUPIALS. 551

or five hairs in each bundle; Welcker gives as many as fifteen
to thirty; Poulton (p. 159) gives the number as varying
between seven and eleven, including both young and old hairs,
the greatest number of old hairs being ten, and, referring to
his figures, states that they “ prove that, at any rate in the
dorsal region, Leydig’s estimate of the number of hairs in a
bundle is too small, while that of Welcker is far too large.”

There is evidently a considerable amount of variation,
possibly between different individuals but more probably at
different times of the year, and this variation affects not only
the number of small hairs in each bundle but also the number
of bundles. On the dorsal area Poulton gives the latter as
constantly four. In the same part we have found the number
to be usually the same, but occasionally five and more rarely
even six bundles may be present. In his work dealing with
the arrangement of the hairs in various mammals, Meijere !
figures, on the back, two groups with five bundles each.

With regard to the number of hairs in each bundle, the
following table represents an average series in a fuily-grown
male examined by us; in the case of successional hairs the
latter are represented with the sign + in front of them.

Group 1. Group 2. Group 3. Group 4. Group 5.
Bundle— Bundle— Bundle— Bundle— Bundle—
1—16 1— 8 1—13 | 1—16 1—16
2—12 2—12 2—13 2—18 2— 7+6
3—14 38— 7+5 3—12 3—18 3—12
4—12 4—10+8 4—16 4—14 | 4-13
5—17+1 5—15 |

These show the number of small hairs in each bundle to
vary generally from twelve to eighteen, less than twelve being
very rarely met with.

Successional hairs are rarely met with,—there is, as Poulton
says, no difficulty in distinguishing them when they are
present,—and the few which are developing have not yet

reached the level of the common follicular opening of the
bundle.

1 «Morph. Jahrbuch,’ Bd. xxi, 1894, p. 312.

552 BALDWIN SPENCER AND GEORGINA SWEET.

In regard to the large hairs, Poulton (p. 158) says ‘‘The
protective large hairs are evidently subject to much wear and
tear, and succeed each other very rapidly ; the new successional
hair, which is to be met with in nearly every section, emerging
from the same follicular mouth, in front of and therefore over-
lapping the base of the old one.” In another part (p. 167)
Poulton says that “ the appearance of two hairs in one follicle
is spoken of as an occasional appearance in other mammals.
In the large hairs of Ornithorhynchus it is the invariable rule.”
A glance at the transverse sections figured by us (figs. 1, 2)
will show that no successional hair, such as is figured by
Poulton, is present; nor, though we have examined both
transverse and longitudinal sections of the skin of three adult
animals, have we been able to find any successional hairs. The
same author, as quoted, says that “the protective large hairs
are evidently subject to much wear and tear, and succeed each
other very rapidly.” From the method of life of Ornitho-
rhynchus we think that the reverse is the case, and that they
are not subject to much wear and tear; and further, that the
difference in regard to the successional hairs as described by
Poulton and ourselves is to be accounted for on the supposition
that there is not a constant but a periodic shedding and
replacement of hairs. Poulton’s sections are on this supposi-
tion taken from an animal in which the shedding and replace-
ment was taking place, while ours are from an animal in which
replacement had already taken place and in which the hairs
were in the maximum state of development.

With regard to the size of the hairs, that of the larger ones
(fig. 3) varies, of course, according to the part of the hair which
is cut through. In our sections the shaft—always round in
section—is cut through, the diameter varying from ‘0245 mm.
to ‘035 mm.; the smaller hairs vary from ‘0071 mm. to
‘(0105 mm. The latter measurement agrees closely with that
given by Meijere and Welcker, viz. ‘(008 mm.; the size of the
large hairs given by Welcker is ‘(048mm., and by Meijere
°045 mm.

In Echidna the hairs are arranged in groups as in Ornitho-

HAIRS OF MONOTREMES AND MARSUPIALS. 553

rhynchus, one (or very rarely two) large hairs usually
emerging from a common follicular opening together with a
relatively small number of small hairs (figs. 4,5). The exact
arrangement varies, however, in different parts of the body.
On the back, where the spines are strongly developed, each of
these issues from its own follicle and is not associated with
small hairs. Amongst these spines is, however, a strong
development of small distinctly wavy hairs arranged in groups
of from eight to ten in number. There may sometimes,
especially on the dorso-lateral aspects of the body, be a large
hair associated with these groups, but most often there is not.
In the mammary region also the large hair is apparently
wanting (fig. 6). The following table gives the number of
large and small hairs present in the bundles from different
parts of the body of a large adult female Echidna from
Tasmania.

Under Chin. Pre-axial Surface of Fore-
limb.

Bundle 1.—8 small, 1 large. Bundle 1.—2 small, 2 large.
Oe ceo gy Ly ay Weep tae ee gl are
oes 4s » 82 , 1 ,,

-o 4—O- 5, DL 5; » 4-2 . 1 ,,

sy OE sr ww lle 3s oF (Oasys

ee 6:6) ee Ae (ees ills
Centre of Mammary Area. Side of Body.

Bundle 1.—4 small, no large. Bundle 1.—6 small, 1 large.
Sy Cae eae ers Ge so Ol Oe
sialeSereOe sar fF whe. eras dO — Ol) sul osnees
cy Pn: co | re re
yy OSes San os 5 WS pp He
ee a pe Os sa aah Gs

In connection with the numbers now given it must be
remembered that (1) there is very great variation in the hairs
of various specimens from different localities such as Tasmania,
Victoria, Queensland, and Central Australia, and (2) it is very
likely indeed that both in different individuals and at different

554 BALDWIN SPENCER AND GEORGINA SWEBT.

times of the year in the case of the same individual the
numbers will vary.

The large hairs are easily distinguishable from the spines,
and this feature is most marked in the case of Central Australian
specimens, owing to the fact that whereas the spines are always
circular in section, the large hairs are flattened (figs. 7, 8),
though there is not the distinct division of the hair into haft,
shield, and tip as in the case of Ornithorhynchus.

The distinct wavy nature of the small hairs, especially of
those of the dorsal region, is a marked feature of Echidna.

On the ventral surface of the body the small hairs vary in
diameter from ‘0357 mm. to ‘054 mm.; the diameter of the
large hairs varies from ‘0612 mm. to "108mm. On the dorsal
surface the small hairs average 126 mm., while the large spines
vary from 3°5 mm. to 4°5 mm., those of the large size being
scattered about amongst a larger number of the smaller ones,
though they are more thickly developed over two areas, one on
the dorso-lateral surface on each side of the body about half-
way back from the head.

For further details of the adult hairs, reference may be made
to the descriptions of the figures.

Muscles of the Hairs.

In both Ornithorhynchus and Echidna a bundle of muscle
fibres is associated with each group of hairs. In the case of
Ornithorhynchus the fibres are, as usual, unstriated, but in the
case of Echidna the fibres are striated. The bundle, as
usual, passes from the superficial part of the corium obliquely
downwards, and is attached to the special modification of the
dermic layer which envelops the group of hair follicles; or, in
the case of the large hairs which develop into spines and are
not accompanied by small hairs, the muscle is attached near
the bottom of the follicle, there being no special development
of the outer root-sheath to form a swelling at this part.

From their disposition these striated fibres are evidently
the equivalents of those forming the arrector pili in other
mammals, and the only other instance known to us of the

HAIRS OF MONOTREMES AND MARSUPIALS. 55)

latter being formed of striated fibres is that of the muscles
moving vibrissz.!

Development of the Hairs.

So far as essential points are concerned, the development of
large and small hairs alike agrees in both Ornithorhynchus and
Echidna. As in all mammals, so in the Monotremata the
hair is developed in a solid downgrowth of the epi-
dermis. It is possible that owing to the remarkable develop-
ment of the inner root-sheath an indefinite central cavity may
be formed at a somewhat earlier stage than that in which the
follicle becomes tubular in other mammals, but a prolonged
investigation of a large series of sections comprising various
stages has led us to the conclusion that it is incorrect to say
that the hairs of either Ornithorhynchus or Echidna are
developed in tubes open to the- exterior, “in” to use Mr.
Poulton’s words, “a tubular and not a solid dowugrowth from
the exterior.” The downgrowth is at first a solid one
exactly as it is in other mammals, and it is only at
some considerable time after the hair has been
formed that there is developed any lumen opening to
the exterior. Exactly as in all other mammals so in the
Monotremata the developing hair has to push its way up
through the centre of a solid follicle.

For the sake of convenience we have taken eight stages, the
structure of the hair and its follicle in each one of these being
represented in the figures.

Stage 1 (fig. 9)—We have not, unfortunately, been able
to secure a young enough embryo to demonstrate the earliest
appearance of the follicular downgrowth. The section of the
earliest stage which we have so far been able to secure, and
which is figured, is, however, sufficiently early to indicate
without any doubt that, in Monotremes, the commencement of
the development of the hair and its follicle is fundamentally
identical with that of all other mammals. There is nothing

1 Owen, ‘Comp. Anat.,’ vol. iii, p. 621. See also Heneage Gibbes, ‘ Quart.
Journ. Mie, Sci.,’ vol. xxiv, p. 193.

vot, 41, PART 4,—NEW SERIES, QQ

556 BALDWIN SPENCER AND GEORGINA SWEET.

which gives any colour to the idea that the earliest trace of the
hair is formed on the surface and subsequently sinks. In fig.
9, the nuclei of the outer layer of the follicle, which are
continuous with those of the Malpighian layer, are seen to be
somewhat elongated, and the central part of the follicle is
occupied by a mass derived from, and continuous with, the
middle part of the epidermis, and in which the nuclei are
evidently undergoing rapid proliferation. The stratum corneum
extends continuously over the external surface, and there is
not the faintest trace of a tubular downgrowth ; the follicle at
this very early stage is, in fact, precisely similar to that of all
other mammals.

Stage 2.—The difference between this and the preceding
stage consists in a slightly greater elongation of the follicle
and in an aggregation of the dermic cells at the base of the
follicle affording the earliest indication of the formation of the
future dermic papilla. Not infrequently, as shown in fig. 10,
there is a trace of a blood-vessel running up towards the base
of the follicle. There is still no indication of anything like a
tube, the stratum corneum running continuously across above
the follicle. It will be noticed that in these early stages there
is no outline of cells to be distinguished except to a certain
extent in the stratum corneum; elsewhere, and until later
stages, there are simply series of nuclei, those of the Malpighian
layer being distinguished as usual by their more elongate form
and definite arrangement.

Stage 3 (figs. 11, 12).—Poulton was, we believe, the first
to draw attention to the fact that in regard to the arrange-
ment of the hairs in Ornithorhynchus there is a distinct
bilateral symmetry; this he also showed to obtain in respect
to the structure of the large hair. Whilst undoubtedly the
latter statement is true if the hair be examined at a relatively
late stage of development,—a stage such, for example, as
Poulton’s figures refer to,—we think, as will be shown later on,
that this bilateral symmetry is only of a secondary nature and
is not of any importance from a phylogenetic point of view.
There is, however, at this early stage an indication of an original

HAIRS OF MONOTREMES AND MARSUPIALS. 507

bilateral symmetry, which in its turn soon becomes lost and
has no influence upon the form of the hair; in fact, this
bilateral symmetry completely disappears before there is any
trace of the hair proper. To show the close agreement of
Ornithorhynchus and Echidna in this respect we have drawn
sections which pass longitudinally through the follicle of a
future large hair in each animal at this stage. The earliest
indication of any modification of the end of the follicle to give
rise to a bulb, takes the form of a flattening from above down-
wards of the end of the follicle, attended by a slight pitting in
of the plate-like structure thus formed. On this plate
(separated from it by a slight space, evidently an artefact in
the section) lies a mass of dermic cells, the early rudiment of
the dermic papilla. In both cases, and especially in that of
Ornithorhynchus, the nuclei of the plate are especially
elongated. The points a and @ indicate, as seen in section, the
future lowest parts or rim of the bulb, the point @ gradually
growing downwards ; while at the same time the dermic papilla
becomes more firmly established, the bulb becoming, as it
were, moulded upon the papilla. The follicle is still quite
solid, and there are indications that the nuclei in the central
part are beginning to arrange themselves so that their long
axes are parallel to the length of the follicle. This is espe-
_ cially seen in the case of the section of Platypus (fig. 11). It
will be seen that in each follicle we can distinguish four parts :
(1) that which lies nearest to the epidermis, and in connection
with which there can be seen in the section of Platypus a
small swelling, the earliest indication of the sebaceous gland ;
(2) a slightly swollen part which indicates the position in which
the hair when developed is most tightly enveloped by the wall
of the follicle, and where, as will be shown subsequently, there
is a special modification in the inner root-sheath ; (3) a some-
what narrower part, which, as the hair grows, increases greatly
in length in proportion to the other parts ; and (4) the rudiment
of the bulb.
Stage 4 (fig. 13).—The section figured is taken from an
Echidna, and shows the definite form of the papilla round which

558 BALDWIN SPENCER AND GEORGINA SWEET.

the bulb becomes moulded, though there is still an indication
of the rim on the upper side not having grown down to quite
the level of the lower side. The nuclei are beginning to have
a definite arrangement; those continuous with the stratum
Malpighii are set with their axes at right angles to the wall of
the follicle, except on the outer surface of the latter, where, as
in all later stages, they have become apparently subject to stress,
resulting in their lying with their long axes parallel to the
length of the bulb; on the inner surface of the latter, where
they lie upon the papilla, they are placed with their long axes
at right angles to the length of the latter. The central layer
of nuclei in the follicle are distinctly arranged with their long
axes parallel to the length of the follicle, while the outer
layers, on the contrary (as can be seen in transverse section),
show a tendency to elongation in the opposite direction. In
the bulb there is the earliest indication of the formation of
the hair itself. The nuclei of the outermost layer, that is those
lying on the dermic papilla, are continuous with a series which
lies in the central line of the hair rudiment, and around this
central series others are beginning to be arranged in somewhat
definite lines, giving the appearance of the hair rudiment just
beginning to grow up through the solid follicle. As in the one
which is figured, the follicle is often bent at a distinct angle to
the surface at the part where the swelling occurs, to which
reference has been made. It will be seen that this swelling
is particularly prominent in this follicle, and that considerable
growth relative to the other parts has taken place in the section
of the follicle above the bulb.

Stage 5 (fig. 14).—This is an important stage, as in it
the hair rudiment, though unmistakably discernible in the
fourth stage, is now clearly marked out. Perhaps the most
striking point at the first glance is the development of pig-
ment in the layer next to the papilla, which has still more
clearly begun, as it were, to grow upwards in the centre of the
follicle, the nuclei becoming elongate and also taking the stain
more deeply than those surrounding them in the follicle.
Running up within the bulb, more or less definite lines can be

HAIRS OF MONOTREMES AND MARSUPIALS. 509

seen separating the nuclei into series which converge towards
the point of the growing hair. These lines of nuclei become
more definitely outlined as development proceeds, and are
associated with certain definite parts of the hair and root-
sheaths, to which reference will be made later. The gland at
this stage is very clearly marked, and its nuclei are all small,
round, and take the stain deeply. The outline of cells can be
seen in parts of the gland (though not as yet in the hair
follicle); and though there is no opening to the exterior there
are traces towards the base of the gland of an internal cavity.
Though the follicle is quite solid there is in both the stratum
lucidum and stratum corneum an indication of the formation
ofalumen. The nuclei begin to dip down towards the central
line of the follicle, as indicated in the figure; this is most
strongly marked in the case of the stratum lucidum, the outer
layers of the stratum corneum still running continuously over
the spot where will subsequently be the opening of the lumen.
A transverse section across the bulb just above the level of the
tip of the papilla is shown in fig. 15. In the centre lie the
nuclei with pigment round them, continuous with those lining
the bulb on its inner surface. It will be seen that there is no
continuation upwards of the cells of the dermic papilla. This
part of the hair probably corresponds to both the medulla and
the cortex. Then follows a circle of six nuclei of considerable
size, which represent, we believe, the nuclei of the layer form-
ing the cuticle of the hair; outside these is an irregular circle
of smaller and more deeply staining nuclei, which, judging by
sections of later stages, represent those of the so-called cuticle
of the inner root-sheath ; outside these are one or more layers
of larger nuclei irregularly arranged and belonging to the
inner root-sheath, while the outer layer of nuclei represent
the outer root-sheath. The whole is enclosed in dermic tissue.

A transverse section at a level just below the origin of the
sebaceous gland is represented in fig. 16. In the centre lies
the follicle of the large hair, and at either side is the early
indication of the follicles in which the two first formed small
hairs will be developed. These, as shown in the next stage,

560 BALDWIN SPENCER AND GEORGINA SWERT.

are formed as outgrowths from the side of the large follicle,
and include, in the centre, a growth from the stratum lucidum
surrounded externally by a growth of the stratum Malpighii.
In the large follicle the central nuclei appear to be smaller
than those surrounding them, which latter are arranged in
roughly concentric circles, owing to the fact that they are cut
across their short axes. There is the earliest appearance in
the central part indicative of the formation of a network, which
becomes more definitely, in fact strongly developed in later
stages, as described by various observers in the case of dif-
ferent mammals. This section shows also the bilateral arrange-
ment of the large and small hairs. As yet only the rudiment
of one small hair is shown on either side, but from this will
subsequently be budded off (1) a second follicle giving two on
each side ; and from each of these, which remain proximally in
connection with one another and with the large follicle, are
budded off (2) the remaining follicles in which the small hairs
are developed, with the result that the large and small hairs
all have a common follicular opening.

Stage 6 (figs. 17—20).—The most important feature of
this stage is the strong development of the inner root-sheath,
which is, as described already by Poulton, a striking feature of
the developing hair in Ornithorhynchus. It is equally strongly
developed in Echidna. The section figured is taken from
Ornithorhynchus.

As compared with the last stage it will be seen that im-
portant changes have taken place in the follicle. The most
important one is that, except in the region of the bulb, the
inner part of the follicle now forms a network of corneous
material in which the now only faintly staining nuclei are
embedded. This network, which is in the case of Echidna and
Ornithorhynchus formed directly out of the central part of
the follicle itself, gives rise to the inner root-sheath. The
meshes of the network lie close together, and the strands
have at once, as regards the follicle, a longitudinal and a con-
centric arrangement—that is, as seen in longitudinal section
they appear to run irregularly along the length of the follicle,

HAIRS OF MONOTREMES AND MARSUPIALS. 561

while in transverse section they are seen to form more or less
roughly concentric bands, which in the lower part tightly
enclose the hair. If the corneous layer be traced downwards
towards the bulb it will be noticed that the corneous nature
gradually disappears, and that the network is really directly
continuous with the softer, undifferentiated, nucleated layers of
the inner part of the bulb, that is the part lying between the
outer root-sheath and the layer of columnar cells next to the
dermic papilla. On the other hand, at the opposite end of the
follicle the network is clearly continuous with the layer lying
immediately beneath the stratum corneum, which layers are in
their turn becoming gradually corneous. Owing to the way in
which the network closely envelops the hair which is growing
up through it, it is usually difficult at this stage to detect
exactly how far the hair has grown up. The central part of
the network, which lies in close contact with the hair, as well
as the tip of the latter, takes the stain very deeply, so that at
times this central part almost appears to be distinct from the
rest ; but anything like close examination at once shows that
it is only a special part of the general network which is most
deeply stained, but which is at the same time in direct con-
tinuation with the latter.

It is at this stage that the lumen of the follicle really makes
its appearance, though as yet it is ill-defined, except at the
surface, though even here it has the form of a somewhat in-
definite tubular cavity crossed by a meshwork of cornified cells,
the substance of which is gradually breaking down to form a
cavity. So far as the follicle is concerned, the structure can
be best understood from the study of a series of transverse
sections. In figs. 18 to 20 we have represented three such
sections taken at different levels. Fig. 18 is taken through
the bulb in such a way that the tip of the papilla is just cut
through. The nuclei of the layers, which are, when traced
upwards, found to be continuous with the medulla and cortex
of the growing hair, are surrounded and almost hidden from
view by the pigment, which is now strongly developed. Out-
side these layers is a clearly marked series of nuclei, the layer

562 BALDWIN SPENCER AND GEORGINA SWEET.

in which they lie being marked off by a more or less definite
line from those lying to the outside. This layer, which is a
more or less clearly marked feature in all stages from the
present onwards, represents the cuticle of the hair. Both in
this section and elsewhere along the course of the hair the
pigment passes continuously round the structure and is never
confined to the under side, as Poulton found it to be in his
preparations and as it is in the adult hair; nor after long
searching over many hundreds of sections have we been able
to find up to this stage any thickening of the cuticle on the
upper surface such as is figured by Poulton in his figs. 19
and 25, and such as again exists in the adult hair (fig. 3).
The cuticle is a clearly marked layer with large nuclei, and is
a striking feature in sections, both transverse and longitudinal,
from this stage onwards.

Immediately outside the cuticle layer lies a series of flat-
tened nuclei which stain darkly and correspond to those which,
in the previous stage, we described as representing the cuticle of
the inner root-sheath. Outside this layer lies a series of layers
arranged concentrically with regard to the central papilla and
with nuclei which follow the trend of the layers, and as a
geueral rule showing chromatin material in contact with the
outer membrane, the inner part of the nucleus being generally
devoid of stained material. These layers are, when traced
upwards, seen to be directly continuous with the corneous
inner root-sheath. There is not at this level, or indeed at this
stage of development, any distinction of the inner root-sheath
into an outer and an inner part corresponding respectively to
Henle’s and Huxley’s layers, though, as Poulton has already
clearly indicated in his descriptions, such a distinction can be
easily seen at a later stage and in a part of the inner root-
sheath close to the bulb. We are quite of Poulton’s opinion
when he says (p. 167) with regard to this differentiation that
“it does not, in Ornithorhynchus at least, imply any differen-
tiation of the sheath into layers, and when we consider the
immense development of the structure in this animal it seems
possible that the distinction can hardly be sustained through

HAIRS OF MONOTREMES AND MARSUPIALS. 563

mammals generally.” We may add that, in respect to this
point, Echidna agrees with Ornithorhynchus. The outermost
layers of nuclei, which are large and deeply stained, represent
the outer root-sheath, which, as usual at this level, is very thin
and at most two layers thick, the inner of the two layers being
represented probably by a very few small nuclei (fig. 18).

Fig. 19 represents a section cut across the follicle close
to the top of the hair, that is about halfway along the length
of the follicle. In the centre is seen the hair, circular in out-
line ; in fact there is as yet no flattening of the hair to be seen,
though when oblique sections are cut—and it is not always
easy to cut true transverse sections—the appearance of a slight
flattening is produced. The hair is enclosed in the meshwork
of the inner root-sheath, which here has the form of a corneous
network containing nuclei which stain in the way already
described. The cuticle of the hair is well developed, and no
nuclei can now be seen in it at this level; whilst the nuclei
which at a lower level represent the cuticle of the inner root-
sheath, cannot here be recognised. The outer root-sheath
contains apparently a single layer of deeply-stained nuclei, and
is well marked off from the inner root-sheath, owing to the fact
that it is not corneous. The meshwork of the inner root-sheath
is rendered distinct, even at this stage, and still more so at later
stages, by the way in which it stains with indigo or picric acid.

Fig. 20 represents a slightly oblique section across the
upper end of the follicle, cutting at one side through the level
at which one of the first formed small hairs is given off. In
the centre lies the corneous network, and outside this is the
outer root-sheath. 'T'wo points of importance may be noticed.
The first is that there is no sharp line of demarcation between
inner and outer root-sheath ; in fact, in the upper part of the
follicle the two always merge to a certain extent into one
another, while in the lower part they tend to become, as develop-
ment goes on, more and more strongly marked off from one
another. ‘The second is a fact of greater importance, viz. that
there is in this part of the follicle above the hair no distinct
lumen, much less any structure which could be described as an

564 BALDWIN SPENCER AND GEORGINA SWEET.

open tube. Ata higher level, that is right in the epidermis, there
is a more clearly outlined tubular space; but here, in the part
immediately above the developing hair, there is merely a more
or less loose corneous network up through which the hair
pushes its way. This feature in the development of the hair
is true of both Echidna and Ornithorhynchus, and can be
seen in sections taken from any part of the body. We have
seen it in sections taken from the following parts: the top of
the head, the shoulder, the middle of the back, the thigh,
under the tail, the chin. It is quite true that in the epidermis
itself, and leading down into the very uppermost part of the
follicle, there is a tubular space present, but this is only
formed when, and not until, the hair has developed to a very
considerable extent, usually for about three quarters of the
length of the follicle ; until it reaches the level of the very top
of the follicle the hair simply, as described, pushes its way
up through the corneous network which forms the inner root-
sheath, and only reaches the more open tubular part when its
tip lies a short way beneath the surface.

Stage 7 (fig. 21)—In this stage the hair itself can be
readily distinguished, with the pigment extending in an
unbroken way almost to the very tip. There is no distinction
to be drawn between medulla and cortical substance; the
cuticle towards the tip has the characteristic serrate outline.
The inner root-sheath in the lower part of the follicle through
which the hair has passed has become more compact, owing
doubtless in part to the pressure of the hair, while just
around and above the tip of the hair the network is clearly
seen, and here it takes the stain more deeply than elsewhere.
In the somewhat swollen part near to the upper end of the
follicle the network is very clearly seen; and within the
incompletely formed cavity is a certain amount of granular
material, produced apparently by the breaking down of certain
of the cells. Above this the tubular lumen, now for the first
time freely open, leads to the surface. In the lower part of
the follicle and in the bulb region we begin to see a more
clearly marked differentiation of layers than in the last stage,

HAIRS OF MONOTREMES AND MARSUPIALS. 565

and corresponding to those which will be more fully described
in the next stage. The nuclei of the outer root-sheath are
seen in parts to be proliferating, so as to give rise to the more
strongly developed and definite outer root-sheath of the more
highly developed hair.

Stage 8 (figs. 22—26).—The figures represent the
structure of the large hairs, the tips of which have just
appeared above the surface, and are drawn from sections taken
from the back of an Echidna measuring 5°5 em. in length.
They may be regarded as representing the structure of the hairs
when the various parts are well developed; and before, owing
to the great growth of the hair and extreme cornification and
subsequent absorption of the softer parts, the structure and
relationship of the different layers is more difficult to determine.

In fig. 22 we have represented a longitudinal section which,
apart from the special modifications which take place subse-
quently to form, on the one hand the flattened hairs of both
Ornithorhynchus and Echidna, and on the other the strongly
developed spines of Echidna, may be regarded as representing
the details of the typical structure of the hairs, large and small
alike, of both animals. It will be noticed, in the first place, that
the dermic papilla appears at this stage to lead up very distinctly
in the direction of the medulla, the nuclei of the cells at the
apex being elongated in the direction of the length of the
hair. Whether there be at this stage any direct continuation
of the dermic cells into the medulla of the hair, we are unable
to state with certainty ; and, despite the suggestive appearance
of the section figured, which is only one of very many in which
the same appearance is presented, we are strongly of opinion
that there is no such continuity. In early development the
medulla—which is very difficult to distinguish as a definite
structure at this stage, though it becomes more prominent in
the adult hair—is without any doubt whatever formed entirely
from the structure of the bulb, the dermic papilla taking no
part whatever in its formation. With the subsequent breaking
down of the central cells of the hair and the formation of an
Open space, an upward growth of the dermic. papilla may take

566 BALDWIN SPENCER AND GEORGINA SWEET.

place, but in early stages there is no indication of any dermic
structure to be seen in the medullary region immediately
above the papilla.

Poulton says (p. 153), in speaking of the hair of Ornitho-
rhynchus, “ From the tip of the papilla, at any rate in the larger
hairs, an axial rod of soft protoplasmic cells, deeply staining in
reagents, is continued. This, when dry and shrivelled, admits
the hair and forms the characteristic medulla ;” and again, “ the
great length of the papilla projecting through the bulb into the
lower part of the area is also very significant, suggesting a pre-
vious development like that of a scale or feather from the surface
of the epidermic covering of a papillary core traversing the
structures from base to apex.”

In the first place, we have been quite unable in either
animal to detect during development any such axial rod of
deeply-staining cells in the medulla, and certainly during the
early stages of development no such structure is present ; nor,
despite the suggestive appearance of the papilla, as seen in lon-
gitudinal section, can any upward prolongation of it be seenin
transverse sections of later stages, of which a typical one is
represented in fig. 26. Until a certain stage in development
has been reached there is no such upward prolongation of the
dermis in either Ornithorhynchus or Echidna, and we are
inclined to draw from this fact the conclusion that, if there
does take place, as perhaps there may, though we have failed
to convince ourselves of its existence, any such extension of the
dermic papilla into the medulla, this is to be regarded as a
secondary feature associated with the special modification of
the large hairs, and has no phylogenetic significance.

However, to return to the structure of the follicle and bulb.
By means of the method adopted by Norris and Shakespeare,
and used also by Mertsching, which consists in staining with
a mixture of Mayer’s hemalum, indigo and carmine, the
various layers become well differentiated, though we have not
been able to obtain the strong carmine stain indicated in
Mertsching’s beautiful figures, all the nuclei in our prepara-
tions being stained with hemalum. As a further differentia-

HAIRS OF MONOTREMES AND MARSUPIALS. 567

tion we have found it useful after over-staining to reduce with
picric acid, the result of which is to render still more clear the
variations in the amount of cornification undergone by the
various layers. The most corneous layer, that is the stratum
corneum, and the side of the inner root-sheath in contact with
the hair, stain a brilliant green which shades off into a yellow
tint as the cornification becomes less in the deeper (morpho-
logically) lying parts (figs. 24—26). In this way it is clearly
seen that no modification has taken place in the very base or
rim of the bulb, which is formed of an outer layer with elongate
nuclei and an inner part with rounded nuclei. In this part we
have been unable to obtain a clear demarcation of cell outlines
such as Mertsching figures,' though this may perhaps be due to
the fact that our specimens were preserved merely in strong
alcohol. As, however, we pass up from the base of the bulb into
the region of the cuticle and inner root-sheath, there is clearly
seen in parts an indication of cell outlines around the nuclei.
The layers which in life rest upon the dermic papilla are
directly continuous upwards with the medulla and cortex of
the hair, and are marked by a strong deposit of pigment.
Immediately outside these there is a most clearly marked layer
in which the nuclei are Jarger and more round than elsewhere.
Passing upwards this layer becomes resolved into a single
series of very distinctly marked cells set at an angle to the
surface of the cortex of the hair and developed equally all
round ; in the lower parts specks of chromatin can be seen in
the nuclei, then at a higher level the latter become flattened
and more darkly stained all over, and gradually above the level
of the top of the papilla the cell outlines and nuclei disappear
and the layer passes up into the strong cuticle with its well-
marked serrations. Just above the tip of the papilla there is a
definite contraction of the hair, followed by a slight expansion,
and then above this it tapers gradually away to the tip, which
has just protruded beyond the surface. Outside of and in
contact with the cuticle there can be detected three distinct
layers in the inner root-sheath ; next to the cuticle of the hair
1 ¢ Arch, f, Mikr, Anat.,’ Bd. xxxi, p. 32, Taf. 4 and 5.

568 BALDWIN SPENCER AND GEORGINA SWEET.

there lies a single layer of flattened, deeply-stained nuclei,
which, by the way in which they take the stain, can always be
distinguished from those lying to the outside. These nuclei
become more rounded as they are traced lower down in the
bulb, and in this part the layer in which they lie can sometimes,
as shown in the figure, be seen to be marked off by a more or
less clear line from those of the inner root-sheath. In the
region of the bulb the inner root-sheath is clearly distinguish-
able into an inner and an outer part, which, however, merge
into one another at the level at which the hair is constricted.
The outer layer has undergone cornification, while the inner
one stains more deeply with the hemalum; its nuclei are
clearly visible, and below it melts into the more or less undif-
ferentiated part which forms the rim of the bulb. Above the
level at which the hair is constricted, that is slightly above
that of the tip of the papilla, there is no differentiation of the
inner root-sheath into the equivalents of Huxley’s and Henle’s
layers. Outside the inner root-sheath, and in this part
sharply marked off from it (fig. 24) lies the outer root-sheath,
in which, in comparison with earlier stages, a considerable
proliferation of nuclei has taken place, the outer layer being
set with their long axes at right angles to that of the follicle.
A transverse section (fig. 26) shows clearly the relative size of
the various layers as they are seen when the hair at this stage
is cut across in its follicle. The whole hair is filled with pig-
ment, there is no clearly marked medulla, and the groundwork
of the whole structure is evidently, from the way in which the
stain is taken, of a corneous nature. The cuticle is sharply
outlined, and is closely invested by the thick inner root-sheath,
which has the form of a network, the nuclei having by this
stage completely disappeared, the part next to the cuticle
being more cornified than that on the outside. In somewhat
younger stages in which the hair is not so large (fig. 23) the
gradual cornification of the root-sheath can be well seen.
Here, next to the cuticle, the nuclei have disappeared, while
outside this they are evidently undergoing change prior to
complete degeneration.

HAIRS OF MONOTREMES AND MARSUPIALS. 569

For the purpose of showing the relationship of the inner root-
sheath we have represented in figs. 24 and 25, drawn under the
camera lucida, (1) two entire follicles as seen in a longitudinal
section of the skin of Echidna ; (2) a portion of one of these on
a larger scale. The exact relationship of the inner root-sheath
to other parts of hair and its method of formation are matters
of fundamental importance in connection with the considera-
tion of the relationship which may exist between hairs and
other structures.

The one point of fundamental importance is that the inner
root-sheath is a differentiation of the inner layers of the ori-
ginal epidermic follicle, and that as shown by various authors,
and most clearly perhaps by Mertsching (pl. iv, fig. 2) in a
section through the hair of a guinea-pig, the sheath is directly
continuous with the layers of the epidermis intermediate be-
tween the stratum corneum on the outside and the stratum
Malpighii on the inner side. This direct continuity is very
clearly brought out by adopting the method of staining already
indicated, and the appearance always presented in the case of
both large and small hairs of Ornithorhynchus and Echidna is
well seen in fig. 24. It will be seen that, in regard to this point,
our observations are at variance with those of Poulton, who
(p. 165) says “The inner root-sheath is always present in the
developing hair, and is a structure of great importance, throwing
much light upon the corresponding sheath as it is described
in other mammals. As in the latter, the inner root-sheath
surrounds that part of the hair which is enclosed in the follicle,
but growing less rapidly it does not extend to the neck through
which the hair protrudes ; hence we do not find it at all in
sections of the upper part of the follicle (figs. 17 and 18).”
On reference to the figures indicated by Mr. Poulton, we think
that there is in them traces of the inner root-sheath to be
seen; in fig. 17 the wavy lines surrounding the old hair, and
evidently also passing round the successional hair, are very
suggestive in this respect, as are also the wavy lines surround-
ing the old hair in fig. 18. In fact, judging by our own sections,
we cannot but think that the structures drawn, but not referred

570 BALDWIN SPENCER AND GEORGINA SWERT.

to, by Poulton are in reality associated with the inner root-
sheath. Our material is certainly in a very fair state of pre-
servation, and in all the stages examined in which the root-
sheath is developed it can be seen most clearly that the latter
extends throughout the whole length of the follicle, and at the
open end is directly continuous with the middle layers of the
epidermis. Possibly the difference in this respect in the
various accounts may be due to the fact that at one particular
part of the follicle there is a special modification of the sheath,
which can be seen by reference to fig. 25. This represents on
a large scale the part of the follicle which is slightly swollen
and lies near to the open end. On the upper side of the
follicle the swelling is more pronounced than on the lower
side, and the nuclei of the outer root-sheath, to the special de-
velopment of which the swelling is due, are arranged as shown
in the figure in a somewhat radiating manner. The most im-
portant feature, however, is concerned with a special modifica-
tion of the inner root-sheath ; the lumen of the follicle becomes
somewhat contracted in this part, and the hair is tightly
grasped by the sheath, which itself is somewhat more solid in
appearance than elsewhere, and is also less sharply marked off
from the outer root-sheath than at a lower level. In addition
to this, the lower part of the inner root-sheath as shown at #
(fig. 25) is slightly produced into a kind of collar arrange-
ment, so that on casual observation it might even be thought
that the inner root-sheath only extended as far as this point.
Anything like a close examination of sections in a good state
of preservation shows clearly that there is merely a local
differentiation in the sheath, which is again to be associated
with the fact that at this spot the hair is tightly grasped in the
follicle, and that we are, in reality, as can be seen from an in-
spection of the figures, dealing with a continuous structure.
The whole face of the inner root-sheath is clearly marked at
this stage by the downwardly directed serrations which repre-
sent the cuticle of the sheath and fit into the upwardly directed
ones on the surface of the cuticle. This thin serrated layer of
the inner root-sheath is morphologically continuous with the

HAIRS OF MONOTREMES AND MARSUPIALS. 571

outermost layer of the epidermis, into which, at the mouth of
the cuticle, it merges. For the greater part of the length of
the follicle this cuticle of the inner root-sheath, though its
serrations render it very easily distinguishable, shows, in
common with the rest of the sheath, no cellular structure;
indeed Poulton (p. 166) says, “I gained the impression that
it is not a distinct and definite layer, but merely the condensa-
tion as it were of the innermost part of the inner root-sheath
upon the exterior of the hair and the moulding of its surface by
contact with the cuticle of the latter.”’ If, however, the layer
be followed down to the bulb, then, where the cellular nature of
the sheath becomes evident, it is seen that the serrated cuticle
is directly continuous (fig. 22) with a special line of nuclei
which are distinct from those of the remainder of the sheath
and from those of the hair cuticle, next to which they lie, by
their very distinctly flattened appearance and dark staining.
Traced still further down into the bulb the flattened nuclei
become more rounded until they reach the lowest point at
which the nuclei of the hair cuticle can be distinguished as a
distinct layer, and at this point the two layers become con-
tinuous with one another (4, fig. 22).

General Considerations.

The figures and descriptions of Mertsching may be taken as
representing the relationship of the layers of the hair and
sheaths as most generally accepted. He shows the cuticle of
the inner root-sheath as running down into the bulb where it
turns back again in continuity with Huxley’s layer. The cuticle
of the hair is directly continuous with Henle’s layer. Of the
two layers into which the outer root-sheath resolves itself in
the lower part of the bulb, the outer one is continuous with the
medulla and the inner one becomes much expanded as it passes
upwards through the bulb and becomes continuous with the
cortical substance of the hair.

In Ornithorhynchus and Echidna, on the other hand, while
we have been unable to distinguish such complete and de-
finitely outlined cellular layers as Mertsching figures, the

voL. 41, PART 4,—NEW SERIES. RR

572 BALDWIN SPENCER AND GEORGINA SWEET.

arrangement is seen to be as follows:—The line of nuclei
representing the cuticle of the hair is directly continuous with
that of the inner root-sheath ; of this fact we feel satisfied,
after long examination of a very large number of well-pre-
served sections, and it may be pointed out that if, phyloge-
netically, the hair be regarded as a process from the surface
of the epidermis, which was developed subsequently in a tube,
and still later in a solid follicle, then this relationship is exactly
what would normally obtain ; for the cavity of the hair follicle
being regarded, ex hypothesi, as formed originally as a de-
pression of the surface at the base of which the hair arises, it
is perfectly natural for the cuticular layer which lines the de-
pression to be directly continuous on the one hand with that
on the hair which arises in the depression, and on the other
with the cuticle on the general surface of the body; indeed,
any other relationship seems to be difficult to understand.

In connection with this it may be noticed that the outer
root-sheath is in connection partly with the medulla and partly
with the cortex. During the earlier stages of development the
outer sheath is formed of a single layer of cells continuous
with the layer which forms the stratum Malpighii of the
epidermis. The layers which lie morphologically to the outside
are modified in the follicle to form the inner root-sheath, which
becomes strongly corneous, except in the region of the bulb,
where it, as well as the elements corresponding to the stratum
Malpighii, is in close relationship to the dermic papilla,—the
source of nutriment,—and here they retain their soft proto-
plasmic nature. The stratum lucidum is, during life, constantly
replenished, as the outer part of the epidermis is worn off, by
proliferation of deeper-lying elements, and at a certain stage
of development there takes place a similar proliferation of the
elements of the outer root-sheath, and hence it is only natural
that, in the region of the bulb, these newly formed elements
should give rise to a portion of the layer corresponding to the
one to which they would have belonged in the surface epidermis.

The continuity of the cuticle of the inner root-sheath and
that of the hair is a matter of great importance, as it implies

HAIRS OF MONOTREMES AND MARSUPIALS. 573

that between the hair and the fenestrated inner root-sheath
there lies a definite cuticular layer. In his discussion with regard
to the homology of the various parts of the hair Poulton has
suggested (p. 189) and has argued in favour of the theory that
“the hair represents the axial, its inner root-sheath the
appendicular part of a feather; and thus an_ intelligible
morphological significance is given to the mysterious inner
root-sheath—a true part of the hair itself, and with it a rising
from the bulb, but which, owing to the mode of development,
is buried deeply beneath the surface.”

A general résumé of the various theories held with regard
to the homologies and origin of mammalian hair has been
given by Benham and in the second part of this work, when
dealing with the development and structure of the hair in
Marsupials, we shall have more to say upon this point ; mean-
while the following will serve our present purpose. The two
most important views with regard to the development of the
various parts associated with the hair, so far as their origin
from the follicle is concerned, may be taken as those expressed
respectively by Gegenbaur and Klein. Gegenbaur! says
“the shaft of the hair is differentiated from the invaginated
epidermis by cornification of its cells, while other cellular
parts of the follicle form the root-sheaths.”’ Klein,? on the
other hand, states that ‘‘ henceforth the multiplication of the
cells at the bulb naturally leads to the new offspring being
pushed up in the axis of the hair rudiment towards the surface,
and becoming elongated constitute the elements of the hair
substance, its cuticle and inner root-sheath; the cells of the
primary solid cylinder represent the rudiment of the cells of
the outer root-sheath only.”

Schafer,’ in his figure of the longitudinal section of a hair,
very distinctly represents the inner root-sheath as directly
continuous with the outer more cornified layer of the
epidermis.

1 «Comparative Anatomy,’ English trans., p. 420.

2 «Atlas of Histology,’ p. 325.
3 * Hssentials of Histology,’ 1885, p. 110, fig. 133.

574 BALDWIN SPENCER AND GEORGINA SWEET.

Our observations upon the Monotremes point very clearly
to the fact that in the most primitive mammals known to us
the inner as well as the outer root-sheath is a direct trans-
formation of certain parts of the follicle, and that during
development the hair formed on and from the bulb forces its
way up through the network, into which the inner root-sheath
becomes transformed.

Giovannini,! in dealing with the successional growth of the
hair in man, describes the inner root-sheath as formed first
above the tip of the developing hair, which subsequently
pierces the sheath in its upward growth. His figures are quite
compatible with the idea that when the downgrowth from the
old bulb to give rise to the new one is formed the inner part
of the new short follicle thus formed gives rise to the inner
root-sheath, which then occupies, exactly as it does in
Monotremes, the central part of the tube through which the
growing hair has to push its way upwards.

The relationship of the various layers as indicated by
Mertsching and other workers on the one hand, and by our-
selves on the other, can be represented by the following
diagrams (pp. 575,576). For the sake of clearness we have in
each case represented the hair as if it were slightly separate
from the walls of the follicle.

In both diagrams the cuticle of the hair, and the layer with
which it is regarded as continuous, is indicated by the black
line, and if it be granted that the external surface of the hair
is to be regarded as morphologically equivalent in position to
the outside of the epidermic layer of the general body surface,
then it will hardly be denied that the diagram of the structure
as represented by ourselves, and as actually exists in Mono-
tremes, is, a priori, what might be expected to obtain. When
the hair has reached nearly to the summit of the follicle, and
the inner root-sheath is well cornified, there is, for the greater
part of its length, no distinct cuticle, the latter being repre-
sented merely by the clearly serrated edge of the sheath, but
near to the bulb this can be distinctly traced into connection

* Arch, f, mikr. Anat.,’ 1890, Bd. xxxvi, p. 528, pls. xxxv—xxxvili.

HAIRS OF MONOTREMES AND MARSUPIALS. 575

with a special and well-marked layer of nuclei continuous with
those in the layer, which traced upwards above the bulb is
seen to give rise to the hair cuticle. Further still, at the open
end of the follicle the cuticle of the root-sheath is directly

Fic. A.—Longitudinal section through the hair, showing the relationship
of the layers as indicated by Mertsching.
Fic. B.—Longitudinal section through the hair, showing the relationship of
the layers in Monotremes.
1. Cuticle of hair. 2. Cuticle of inner root-sheath. 3, Huxley’s layer.
4, Henle’s layer. 5. Outer root-sheath.

continuous with the outermost layer of the epidermis, and takes
the stain, in its course along the follicle, in precisely the same
way (figs. 24 and 25).

If now we represent diagrammatically the relationship of the
various parts in the developing hair of a Monotreme, this can
only be done as represented in the accompanying Fig. C, which,
though diagrammatic, represents what is actually the relation-
ship of the hair and inner root-sheath. If we suppose the hair

576 BALDWIN SPENCER AND GEORGINA SWEET.

to have been originally developed on the surface, then the
relationship of the parts can be represented in Fig. D.

The difference between ourselves and Mr. Poulton lies in the
fact that he regards the inner root-sheath as “a true part of
the hair itself, and with it arising from the bulb ;” whilst, with
the advantage of a larger series of stages than Mr. Poulton was
able to study, and with probably better preserved material, we
have been led to the conclusion that the inner root-sheath is
not a true part of the hair itself, and is not developed from the

Fic. C.—Longitudinal section through the hair of a Monotreme lying in its
follicle, showing the relationship of the layers.

bulb, a conclusion which our figures—all drawn under the
camera lucida—will, we think, serve to demonstrate. It ap-
pears to us that the relationship now described renders unten-
able the suggested homology between the inner root-sheath
and the appendicular parts of a feather.

We must, however, now return to a consideration of the
early stages of the development of the hair. Mr. Poulton
(p. 183) says, ‘It is, indeed, by no means improbable that the
first and earliest trace of the hair is formed at the surface in

HAIRS OF MONOTREMES AND MARSUPIALS. 577

Ornithorhynchus, and subsequently sinks with the deepening
tube. Material by which this suggestion could be tested is
unfortunately wanting. But the open tube is not by any
means the only, although it is the most important point about
the development of the hair of Ornithorhynchus. The great
length of the papilla projecting through the bulb into the
lower part of the hair is also very significant, suggesting a
previous development like that of a scale or feather from the
surface of the epidermic covering of a papillary core, traversing

Fic. D.—Longitudinal section through a hair which is supposed to have
developed on the surface and not in a follicle. The part corresponding
to the inner root-sheath is marked by wavy lines.

mh. Medulla of hair, ¢.4. Cuticle of hair. co.4. Cortex of hair.
e.i.r.s. Cuticle of inner root-sheath. 7..s. Inner root-sheath. 0.7.s.
Outer root-sheath. Mpg. Malpighian layer.

the structure from base to apex. Further confirmation is
afforded by the axial rod of soft protoplasmic cells forming the
medulla of hair; for a shortening papillary core, surrounded
by cells of the rete mucosum superficially undergoing corni-
fication, would tend to leave just such an indication of its
former presence.”

First of all as to the supposed open tube. We have already
pointed out that neither the large nor the small hair of either

578 BALDWIN SPENCER AND GEORGINA SWEET.

Ornithorhynchus or Echidna develops in what can be correctly
called an open tube. In the earlier stages—indeed, until the
hair is well formed and has reached a considerable distance
up the folliclke—development takes place in a solid follicle.
Possibly the latter may open to the surface at a slightly earlier
stage than it does in some other mammals, but the one im-
portant point in this connection is that the hair-follicle has the
form of a solid downgrowth, and that the early stages during
which the hair is laid down are absolutely indistinguishable
from those of other mammals. Figures which represent the
condition in one of the earliest stages of the follicle are pre-
cisely similar to those of corresponding stages in other mam-
mals, and, with the elongate nuclei at the base of the follicle,
correspond so closely to as to be in fact indistinguishable from
those of Marsupials, such as Macropus and Perameles. The
resemblance between the two is so complete, that we think
there can be very little doubt but that in Monotremes, just as
in Marsupials, in which our observations (to be published later)
to a large extent confirm those of Maurer, the very earliest in-
dication of the hair will be found to take the form of an elonga-
tion of the lowest epidermic cells. At all events, the figures now
given show (1) that the hair cannot be described as developing
in an open tube; and (2) that the earliest trace is not formed
on the surface, and then sinks with the deepening tube.

The next point which we desire to lay emphasis upon is that
in these lowest mammals, and indeed in all mammals, the hair
is a radially symmetrical structure ; by which we mean to imply
not that hairs may not possibly be derived from structures
which originally possessed a bilateral symmetry, but that,
from the earliest moment at which we get the first rudiment
of the hair itself laid down, the structure takes on a radial
symmetry, and that any bilateral symmetry, such as may be
found in the well-developed hairs of both Ornithorhynchus and
Echidna, is a secondary and not a primary feature.!

1 As we shall show later, the flattened bristles of such a Marsupial as

Perameles are, in their early development, radially symmetrical, the bilateral
symmetry, just as in Ornithorhynchus, being a secondary feature.

HAIRS OF MONOTREMES AND MARSUPIALS. 579

If we trace the development of the hair in Ornithorhynchus
we find that at an early stage, before there is any appearance
of a hair in the follicle, the base of the latter is somewhat
flattened into a plate-like structure, indicating possibly the
original development of the hair rudiment out of a bilaterally
symmetrical structure; but it is not until the radially sym-
metrical bulb with its dermic papilla has been formed that we
can see any trace of the hair itself, and the latter has at first,
and until it is well established, a radial symmetry with a per-
fectly circular outline in transverse section. Subsequently, and
as a secondary modification, there arises a flattening which
produces in the scale-like part of the hair a bilateral symmetry,
but this is in no sense a primitive feature of the hair. So far
as we have any evidence, every hair is primarily a radially
symmetrical structure,—that is, we never find the hair com-
mencing to grow until the bulb of the follicle from which it
grows has acquired a radial symmetry. In Echidna the radial
symmetry is emphasised in some of the large hairs, which
become modified into spines, and lost in others which become
flattened, the latter feature being especially emphasised in
Ornithorhynchus ; but both of these are very clearly secondary
developments, and cannot in any way be regarded as represen-
tative of the structure of the primitive mammalian hair.

In the same way the remarkably developed inner root-sheath
of the large hairs of Monotremes is not to be regarded as a
primitive feature; it is simply a secondary feature of no
phylogenetic importance, and is to be associated with the strong
development of all the various parts of the follicle. The
modification of the large hair requires, as it were, that all the
various parts of the walls of the follicle should be strengthened
and stiffened, and that at the same time the root of the hair
should be tightly held. These requirements are met by the
cornification of the inner root-sheath, that is of the walls of
the follicle, and also by the way in which it grasps the hair,
and has its surface strongly imbricated so as to oppose the
pulling out of the latter, which might otherwise be easily sepa.
rated from the soft structures of the bulb.

580 BALDWIN SPENCER AND GHORGINA SWEET.

We fail to see any relationship between the specially modi-
fied inner root-sheath of Monotremes and the appendicular
part of a feather. The sheath is formed perfectly indepen-
dently of the hair rudiment in the wall of the follicle, which,
with the development of the hair, becomes transformed into a
pit. A comparison of a transverse section of a developing
hair and feather will serve to show there is no real relationship
between the inner root-sheath of the one and the appendicular
part of the other.

Fic. E. Transverse section across a Monotreme hair lying in its follicle, at the
level of the top of the dermic papilla.

Fic. F.—Transverse section across a developing feather (modified from Newton
and Gadow), to show the relative positions of the developing main
shaft, after-shaft, and appendicular parts.

d.p. Dermic papilla. m.4. Medulla of hair. ¢o.. Cortex of hair.
ch. Cuticle of hair. c.c.7.s. Cuticle of inner root-sheath. «7.s.
Inner root-sheath. 0.7.s. Outer root-sheath. p.s. Primary shaft of
feather. 7. Rami. — sé.c. Layer continuous with stratum corneum
( = cuticle of inner root-sheath of hair). m. Malpighian cells of follicle.
a.s. After-shaft. d. Dermis.

We have already stated that the bilateral symmetry in the
large hair of Ornithorhynchus is not a primary, but a secon-
dary feature, and the same remark holds true in the case of
the large size of the bulb and papilla. The early development
of these is precisely similar, so far as relative size is concerned,
to that of other mammals, and it is only with the secondary
modification of the hair to form in the one case a flattened

HAIRS OF MONOTREMES AND MARSUPIALS, 581

structure, and in the other a spine, that the papilla begins to in-
crease in size, and to extend any distance up the shaft. Poulton
(p. 183) says, ‘The great length of the papilla projecting
through the bulb into the lower part of the hair is also very
significant, suggesting a previous development like that of a
scale or feather from the surface of the epidermic covering of
a papillary core traversing the structure from base to apex.”

Various authors, and especially Maurer, have insisted upon
the fact that the dermic papilla is always developed at a slightly
later time than the epidermic forecast. In the case of Marsu-
pials, such as Perameles, Macropus, Sminthopsis, and Dasyu-
roides, we can confirm the conclusions of Maurer. The earliest
indication of the hair forecast in these animals has the form
of a lengthening and definite arrangement of the elements of
the Malpighian layer, the dermic layer at first taking no part
whatever in the formation of the structure. In the typical
mammalian hair the dermic papilla is never of very large size,
and it is only in those cases in which we get special modifica-
tions to form spines, &c., that we find, not as a primary but as
a secondary feature, that the papilla increases in size and ex-
tends some distance up the shaft. The very fact that this
development is only found in what is clearly a secondary modi-
fication of the hair, which up to the period at which the modi-
cation begins to show itself develops in the ordinary way, is
sufficient to indicate the fact that in the case of the Mono-
tremes the special development in question has no phylogenetic
signification of any kind. All that it signifies is simply the
fact that, as might be expected to be the case, the larger the
hair (or its modification, a spine) becomes, the larger is the
dermic papilla.

It seems probable that the connection between feathers and
hairs, if any such thing really exists, is a very remote one, and
can only be traced back to some simple form of epidermic, and
perhaps scale-like structure, of the former existence of which
common ancestral structure we have in hairs a possible indi-
cation in the flat plate, which at a very early stage is developed
at the base of the follicle in Monotremes (figs. 11, 12, 12a).

582 BALDWIN SPENCER AND GEORGINA SWEET.

It must also be remembered that in the case of any comparison
between the two we must compare not the hair and the highly
developed quill or contour feather, but the former and the
primitive nestling feather, or “ Neossoptile,” with its short
calamus and terminal bunch of “ spikes.” In development
even this, the earliest form of feather, shows traces of bilateral
symmetry, while in the case of the hair we have a strongly
marked radial symmetry ; in the former the primitive symmetry
of a possibly simple ancestral scale is retained, while in the
latter it is lost before the rudiment of the hair is actually laid
down. It must also be remembered that we have no evidence
whatever as yet of any development of hair on the external
surface of the epidermis. From the lowest to the highest
mammals the development of the hair is practically identical,
the Monotremes even showing us nothing which can be re-
garded as primitive in the development of the actual hair. It
may, however, be worth while to draw attention to the fact that
in the Monotremes, and probably also in certain of the Marsu-
pials, the original follicle may give rise not only to the large
hair, but to small hairs which are developed in follicles budded
off from the large one, and that thus we may, perhaps, have
a homoplastic relationship between, on the one hand, the axial
part of a feather and the large hair, and, on the other hand,
between the appendicular part of a feather and the small hairs.
It appears to us that in the feather there is no representative
of the inner root-sheath in the hair which is to be regarded as
a special structure intimately associated with the growth in a
follicle, while in the hair there is no representative of the
feather sheath which is shed, and which is, it seems to us,
correctly compared by Maurer to the moulted cuticle of rep-
tiles.

We shall return subsequently, when dealing with the deve-
lopment of hairs in the Marsupialia, to the question of the re-
lationship of the hair to other epidermic structures ; meanwhile
our conclusions with regard to the development of hair in
Monotremes may be summarised as follows :

(1) The early development of the follicle is precisely similar

HAIRS OF MONOTREMES AND MARSUPIALS. 583

to that which takes place in other mammals, and is in the form
of a solid epidermic downgrowth.

(2) The dermic layer takes at first no share in the develop-
ment.

(3) The lower end of the follicle forms at first a flat
obliquely slanting plate, indicating possibly a primitive bilateral
symmetry in the structure from which the hair is originally
derived.

(4) The plate, by ingrowth of the dermic layer, is transformed
into a radially symmetrical bulb moulded on the dermic papilla,
this radially symmetrical bulb being formed before the deve-
lopment of the hair itself takes place.

(5) The hair is formed as an upgrowth from the bulb within
the solid follicle, up which it pushes its way just as in other
mammals, and it is not developed in a tube open to the
exterior.

(6) The inner root-sheath is developed as a modification of
the walls of the follicle, and subsequently becomes transformed
into a corneous network surrounding the growing hair. The
development of the inner root-sheath is fundamentally similar
in large and small hairs.

(7) There is no relationship between the inner root-sheath
of a hair and the appendicular part of a feather.

(8) The medulla of the hair is formed primarily as a solid
upgrowth of the cells which are continuous with those of the
stratum Malpighii of the epidermis.

(9) The cuticle of the inner root-sheath is directly con-
tinuous with the cuticle of the hair.

(10) There is no real distinction of the inner root-sheath into
Huxley’s and Henle’s layer.

(11) The larger size of the dermic bulb in the large hairs of
Ornithorhynchus, and in the spines of Echidna, is a secondary
feature of no phylogenetic importance.

(12) In all essential respects the development of the hairs
in Monotremes is precisely similar to that of other mammals.

584 BALDWIN SPENCER AND GEORGINA SWEET.

EXPLANATION OF PLATES 44—46,

Illustrating Mr. Baldwin Spencer’s and Miss Georgina Sweet’s
paper on “ The Structure and Development of the Hairs
of Monotremes and Marsupials.”

List oF REFERENCE LETTERS.

A. Point at which the cuticle of the hair is continuous with that of the
inner root-sheath. 6.v. Blood-vessel. c.4. Cuticle of hair. co.%. Cortex of
hair. d. Dermis. d’. Special modification of dermis in Echidna to form an
enclosure for the group of follicles. d.p. Dermic papilla. /.p. Plate-like
structure at the base of the solid follicle. 4%. Hair. 4,. Tip of developing
hair in follicle. 7.7. s. Inner root-sheath. 7.7.8, Equivalent of Henle’s,
and 7z.7.s,, of Huxley’s layer. 7.4. Large hair. m.p.g. Stratum Malpighii.
o.7.s. Outer root-sheath. s.g. Sebaceous gland. s.4. Small hair. — sé¢.e.
Stratum corneum. s¢./. Stratum lucidum.

The outlines of all the figures are drawn under the camera lucida.

Fie. 1.—Transverse section across a group of hairs from the back of an
adult Platypus, showing four groups of small hairs. ‘The left side lies at a
deeper level than the right side, where in one group all the root-sheaths have
coalesced to form a common follicle. In the other groups the sheaths of the
different hairs are distinct. There is no successional large hair. Zeiss,
apert. 0°95, oc. 1.

Fic. 2.—Transverse section across a group of hairs from the back of an
adult Platypus, showing six groups of small hairs. The root-sheaths of all the
hairs are distinct. Zeiss D, oc. 1.

Fic. 3.—Transverse section across the shield part of an adult large hair of
Platypus. The distinct flattening is seen, and also the dorsal thickening of
the cuticle, and the bilateral arrangement of the pigment in the ventral part
of the cortex, nearer to which side lies the medulla.

Fic. 4.—Transverse section across a group of hairs lying to the side of the
mammary area in an adult female Echidna. ‘he large hair lies anteriorly.
Bach hair has its distinct outer and inner root-sheath, and the follicles are bound
together into a group by a special deeply staining modification of the dermis.

Fie. 5.—Transverse section across a group of hairs lying on the ventral
surface just behind the bill of an adult Echidna. A large flattened hair and
six small hairs are present. The inner root-sheath is especially well de-
veloped. ‘There is no distinct medulla to be seen.

Fic. 6.—Transverse section across a group of hairs from the mammary area
of an adult female Echidna. No large hair is present. The section lies close
to the surface, and shows the root-sheaths coalescing. At a slightly higher

HAIRS OF MONOTREMES AND MARSUPIALS. 585

level, i.e. closer to the surface, they will coalesce completely to form a common
follicular opening through which the hairs emerge.

Fic. 7.—Transverse section across a large flattened hair of an adult Kehidna.
There is no dorsal thickening of the cuticle, or restriction of pigment to the
lower surface. The hair is from the dorso-lateral aspect.

Fic. 8.—Transverse section across a large flattened hair of an adult Echidna,
from the ventral surface just behind the bill. The anterior face is distinctly
concave. This form of hair is especially developed in specimens of Echidna
from Central Australia, and has not been noticed on specimens from Tasmania,
Victoria, or Queensland.

Fie. 9.—Longitudinal section through the follicle of an embryo Ornitho-
rhynchus on the chest region. The nuclei at the base of the follicle are slightly
elongate. ‘There is no trace of a dermic papilla. Zeiss, apert. 0°95, oc. 4.

Fie. 10.—Longitudinal section through a follicle at a slightly later stage
from the same embryo of Ornithorhynchus in the chest region. At this stage
there is the earliest appearance of a modification in the dermis (d. p.) to form
a dermic papilla, up towards which a small blood-vessel runs (4.v.). Zeiss,
apert. 0°95, oc. 2.

Fie. 11.—Longitudinal section through a follicle of the same embryo of
Ornithorhynchus from the back just in front of the tail. The slight swelling
indicating the future sebaceous gland (s.g.) is seen; below this the follicle is
slightly swollen, and at the base a flat plate-like structure (/p.) is formed,
which has been slightly pushed in by the developing dermic papilla. @and é
indicate the future lowest part or rim of the bulb. In the centre of the follicle
the nuclei are arranged with their long axes roughly parallel to that of the
follicle. Zeiss, apert. 0°95, oc. 1.

Fic. 11a.—Transverse section across the base of the follicle at the same
stage as Fig. 11, to show the flat plate with its elongate nuclei.

Fie. 12.—Longitudinal section through the same stage in an Echidna
embryo, to show the flat plate at the base of the follicle. There is no ap-
pearance as yet of the development of the sebaceous gland, the follicle being
probably at a slightly earlier stage of development than the one represented in
Fig. 11. Zeiss D, oc. 1.

Fic. 13.—Longitudinal section through the follicle of an embryo of Echidna,
showing the early formation of the bulb and the upgrowths of the dermic
papilla. At the external end of the follicle are two swellings (s. 4.7) which
may possibly indicate outgrowths to form the follicles of small hairs. The
nuclei within the follicle are arranged in lines following roughly the length of
the follicle. At the bulb end the nuclei are more definitely arranged, and
there is the earliest indication of the upgrowths to form the hair rudiment
itself, but as yet there is no pigment present and no cornification. Zeiss,
apert. 0°95, oc. 2.

586 BALDWIN SPENCER AND GEORGINA SWEET.

Fie. 14.—Longitudinal section through the follicle of an embryo Ornitho-
rhynchus. The rudiment of the hair can be clearly seen arising from the bulb,
faint but distinct lines being present, which converge towards the apex of the
hair rudiment and separate series of nuclei from one another. Pigment has
appeared, and stretches up towards the medulla. The follicle is solid, but at
the external end the nuclei are beginning to dip in towards the centre. The
gland is well developed, and reaches as far down as the level of the tip of the
dermic papilla, which is not shown in the drawing. The stratum corneum
is continuous over the top of the follicle. Zeiss apert. 0°95, oc. 1.

Fig. 15.—Transverse section just above the level of the tip of the dermic
papilla of a follicle at a slightly later stage than that represented in Fig. 14.
The layers of nuclei indicating the cuticle of the hair (c. 2.) and of the inner
root-sheath (¢.i.7.s.) can be seen. Outside these lie the inner root-sheath,
which is commencing to be corneous (¢.7.s.), and the outer root-sheath (0.7.s.).
Zeiss F, oc. 1.

Fie. 16.—Transverse section through the same follicle as the one repre-
sented in Fig. 15, at a higher level. In the centre lies the solid follicle in
which the large hair is being developed. Anteriorly is the gland, and on
either side a follicle, budded off from the central one in which the first formed
small hairs will be developed. Zeiss F, oc. 1.

Fic. 17.—Longitudinal section through the follicle of an Ornithorhynchus
from the dorsal surface of an embryo at a later stage than that represented in
Fig. 14. The hair has grown up, through the corneous network into which
the inner cells of the follicle have been modified, as far as the point 4,. The
cornified inner cells form the inner root-sheath, which tightly encloses the
growing hair and stains deeply. At the upper end an irregular lumen is de-
veloped leading to the exterior. The inner root-sheath is directly continuous
with the stratum lucidum of the epidermis. In the bulb region the layer of
nuclei continuous with the cuticle of the hair (¢.4.), and with the cuticle of
the inner root-sheath, can be distinguished (¢.7.7.s.). Only the proximal
part of the now well-developed sebaceous gland (s.g.) is indicated. Zeiss,
apert. 0°95, oc. 1.

Fics. 18—20.—Transverse sections across the follicle of the same embryo
as represented in Fig. 17.

Fig. 18.—Section at the level of the tip of the dermic papilla. The cells
filled with pigment, and representing the medulla and cortex, lie next
to the dermic papilla. The cuticle of the hair (¢. 4.) and of the inner
root-sheath (c.¢.7. s.) are clearly seen. Outside the latter lie the nucleated
layers of the inner root-sheath, and outside this the outer root-sheath
(0.7. 8.), Which is very thin in this part and consists of two layers of
cells, the nuclei of the outer layer being large and deeply stained, and
those of the inner layer very few in number and smaller. Zeiss F, oc. 4.

HAIRS OF MONOTREMES AND MARSUPIALS. 587

Fig. 19.—Section close to the tip of the developing hair. The follicle is
seen to be solid; the cornified network which forms the inner root-
sheath (¢.7.s.) tightly envelops the hair. Zeiss F, oc. 4.

Fig. 20.—Section close to the external end of the follicle, showing the
corneous and more open network of the inner root-sheath, the large
outer root-sheath, and an outgrowth from the main follicle to form the
follicle of a small hair. The main follicle is still solid, though the
central open network indicates the position of the future lumen. Zeiss
0°95, oc. 4.

Fic. 21.—Longitudinal section through the follicle of an embryo of Echidna,
showing the hair more highly developed. The inner root-sheath is more de-
finitely established ; the nuclei of the layers forming the cuticle of the hair
(c. A.) and of the inner root-sheath (c.7.7.s.) can be clearly seen in the lower
part of the follicle. The lumen is distinctly formed in the epidermis, and
below this the central part is occupied by disintegrating material. Imme-
diately above the hair there is still the definite network formed by the inner
root-sheath. Zeiss, apert. 0°95, oc. 4.

Fic. 22.—Longitudinal section through the lower part of the follicle of an
embryo of Echidna, in which the tip of the large hair has just appeared above
the surface. The outer root-sheath is well developed. The inner root-sheath
in the bulb region can, owing to the greater cornification of the outer part, be
distinguished into two layers, one (7.7.s.,) presumably the equivalent of
Henle’s, and the other of Huxley’s layer (i.17.s..). The cuticle of the hair
fits closely on to the surface of the inner root-sheath, the serrations being
clearly marked. Traced down towards the bulb the outermost layer of the
inner root-sheath is directly continuous with a series of nuclei (¢.7.7.s.) which
are flattened and stained deeply, and separated off in the lower part, where
they become more rounded, by a distinct line from the rest of the inner root-
sheath. At the point 4 this layer turns round and is continuous with the
layer which passes upwards into the well-developed cuticle of the hair (ce. 4.).
The cells of the latter are large and clearly outlined, the nuclei gradually
fading away as the layer is traced upwards into the strongly cornified part.
The medullary region has the appearance of opening up to admit possibly of a
secondary upward prolongation of the dermic papilla. Immediately above the
bulb is a constriction followed by a slight swelling of the hair. The cuticle is
equally developed on both sides. Zeiss C, oc. 4.

Fic. 23.—Transverse section across a group of hairs of an embryo Echidna.
The follicle is at a slightly earlier stage of development than that represented
in Fig. 22, and the section is at some little distance below the epidermis. In
the central follicle the large hair is seen surrounded by the inner root-sheath,
the outer layers of which are not yet completely cornified. Four follicles in
which small hairs will be formed are cut through; the one most to the right
hand is budding off a secondary follicle. Zeiss E, oc. 4.

VoL. 41, PART 4. —NEW SERIES. Ss

588 BALDWIN SPENCER AND GEORGINA SWEET.

Fries. 24—26.—Sections through the follicle and hair of an embryo of
Echidna of the same age as that represented in Fig. 22. They have been
stained with hemalum, carmine and indigo, and subsequently treated with
picric acid to show the cornification of the different structures and the con-
tinuity of the inner root-sheath with the stratum lucidum of the surface epi-
dermis.

Fig. 24.—Longitudinal section through two follicles; in the one on the
right side the hair has been torn away from the bulb. The inner root-
sheath is seen to run continuously along the whole length of the
follicle. At ifs lower end in the bulb region it can be divided into an
outer (é.7.s.,) and an inner part (¢.7.s..), the latter being less
cornified than the former. At 2 is a special collar arrangement where
the hair is most tightly grasped. The cuticle (c.7.7.s.) of the sheath
passes upwards, and is continuous with the stratum corneum.

Fig. 25.—More highly magnified part of the follicle shown in Fig. 24, to
show the special modification in the inner root-sheath to form a collar.
The face of the inner root-sheath on the side of the follicle is seen in
part.

Fig. 26.—Transverse section across the follicle. The strongly pigmented
medullary and cortical parts (m. and co. 4.) are seen, the medulla not

_ being distinct from the cortex. The network of the inner root-sheath
is seen more cornified on its face next to the hair than on that next to
the well-marked outer root-sheath.

TROPHOBLAST AND SEROSA, 589

Trophoblast and Serosa. A Contribution to the
Morphology of the Embryonic Membranes of
Insects.

By

Arthur Willey, D.Sc.Lond., Hon. M.A.Cantahb.,
Balfour Student of the University of Cambridge.

Tue substance of the following remarks formed the subject
of a paper! read by me at the last meeting of the British
Association in Bristol.

It is matter of common knowledge that, during the early
stages of development, the embryos of insects are protected
by two membranes,—an inner, the amnion, and an outer, the
serosa,—which resemble in principle the homonymous feetal
membranes of the higher vertebrates.

Both in the embryos of the Vertebrata Amniota and in those
of insects, the amnion and serosa are derivatives of the extra-
embryonic blastoderm (abstraction being made of the adven-
titious mesodermal elements which accompany the membranes
in vertebrates), and in both it may be stated in general terms
that the amnion is subsidiary to the serosa. In insects the
secondary character of the amnion, as compared with the
serosa, is particularly clearly marked, and it has been recog-
nised that the serosa, being a direct derivative of the blas-
toderm, is the older structure (Heymons, 4).

The peripheral, extra-embryonic, non-formative epiblast or
blastoderm of the mammalian blastodermic vesicle has been

1 Entitled “ Considerations bearing upon the Phylogeny of the Arthropod
Amnion.”

590 ARTHUR WILLEY.

defined as the trophoblast by Hubrecht (5), since its prin-
cipal function is to provide for the nutrition of the embryo.

More recently (6) Hubrecht has published a very remark-
able and what I venture to predict will become a classical
theory of the mammalian trophoblast, in which he seeks to
demonstrate a phyletie relationship between the latter and the
superficial ectoderm (‘‘embryonic epidermis,” Balfour; ‘‘ Deck-
schicht,” Goette) of the embryos of Amphibia.

Hubrecht points out that the vertebrate amnion is a de-
rivative of the trophoblast, which is a structure sui generis,

If the vertebrate amnion, in its capacity of derivative of
the trophoblast, has a profound phylogenetic significance, one
would be inclined to suppose that an analogous significance
would attach to the embryonic membranes of insects.

Hitherto the prevailing impression seems to have been that
the latter were of merely cenogenetic importance, and all
serious attempts to account for them morphologically have
been more or less coloured by this assumption.

The embryos of a species of Peripatus which I found in
New Britain last year (1897), of which I have recently pub-
lished a full description (14), seem to me to point the way out
of this somewhat barren and unsatisfactory position, and to
endow the embryonic membranes of insects with a singular
interest. . Because, if it could be shown that a common prin-
ciple governs the theories applied to the explanation of the
embryonic membranes of insects and of those of the higher
vertebrates, the one theory would constitute an important
complement of the other.

On this occasion I am only dealing with the embryonic
membranes of insects, because they are the best known, and
my own observations only bear upon these structures. There
can be little doubt that the embryonic membranes of scorpions
are capable of being explained in an analogous manner, only
the data are not suflicient in this case.

My own idea is that three theories are necessary for the ex-
planation of the embryonic membranes of scorpions, insects,
and vertebrates, and that a common principle underlies the

TROPHOBLAST AND SEROSA. 591

whole. For the vertebrates, Hubrecht has provided the requi-
site theory. The present paper is written for the purpose of
providing an analogous theory for the insects, while the
scorpions must be handed over to posterity.

My chief object is to demonstrate the applicability of the
conception of the trophoblast to invertebrate animals ; in fact,
to show that the serosa of the insect embryo can be traced
back to a primitive trophoblast.

I believe, in short, that the trophoblast, as it is pre-
served tous in the embryos of Peripatus nove-brit-
annie#, arose in adaptation to a viviparous habit
acquired by the terrestrial descendant of an aquatic
ancestor; and that it became transformed, whether
directly or by substitution, into the serosa, in corre-
lation with the secondary deposition of yolk-laden

eggs.

In the case of the vertebrates, Hubrecht starts with a pro-
tective layer (Deckschicht), which becomes transformed into a
nutritive layer (trophoblast). For the insects, 1 commence with
a nutritive layer, which becomes changed into a merely pro-
tective layer, the serosa. This example will suffice to show
that the two theories are quite distinct in their treatment of
the respective problems, only they have the principle in common
which is expressed in Hubrecht’s admirable conception of the
trophoblast.

Just as, throughout the series of the Amniota, the forma-
tion of the amnion by no means takes place according to the
stereotyped plan with which we are familiar in the chick, so in
insects there is a distinct gradation in the mode and
extent of development of the amnion. It attains its
completest development in the highest order of insects, namely
the Hymenoptera, while its most nascent condition is exhibited
in the most primitive insect in which an amnion has been
found to occur, namely the Thysanurid, Lepisma saccha-
rina, L. (Heymons, 4). The observations of Heymons on the
development of Lepisma are particularly noteworthy. Not

592 ARTHUR WILLEY.

only is the amniotic cavity of greater relative cubic capacity
in Lepisma than in any other Hexapod, but it never com-
pletely closes, remaining permanently open by a minute orifice,
the amniotic pore. Thus, in its most primitive condi-
tion, the amniotic cavity of insects is anopen sac.
This is in strong contrast with the vertebrate amniotic cavity,
as exemplified in the embryo of Erinaceus. Two quotations
from Hubrecht’s memoir will suffice to call to mind the point
of view adopted by him with regard to the vertebrate amnion.

He says (6, p. 24): “ Ich denke mir das Amnion in seiner
allerersten Entstehung gleich als geschlossene Blase;! ich
betrachte somit die Entwickelungsweise durch sich entgegen-
wachsende Faltenrander als ein bedeutend cenogenetisch modi-
ficirten Entwickelungsprozess.”

Again, on p. 25 he says: “ Meiner Ansicht nach muss man
nicht bei Sauropsiden, sondern bei den Séugern nach Andeu-
tungen der primitiveren, der mehr urspriinglichen Amnion-
entwickelung suchen.”

In Peripatus nove-britannie the egg is without yolk
and the embryo develops on the surface of a relatively enormous
blastodermic vesicle, whose wall consists of an internal layer of
endoderm and an external layer of ectoderm. The ectoderm
of this trophic vesicle does not consist of flattened and passive
cells, as does the serosa of insects, but it is a mucous epithelium,
the cells having vacuolar contents and serving for the absorp-
tion of nutrient fluids from the uterine wall and their trans-
ference to the trophic cavity, whence they are available for the
nutrition of the embryo. The trophic ectoderm is there-
fore the trophoblast of the embryos of this species of
Peripatus. The chorionic membrane always intervenes be-
tween the trophoblast and the uterine mucosa.

At first the embryonic tract lies at the posterior ventral
extremity of the trophic or blastodermic vesicle, and at this
stage the entire embryo bears a remarkable resemblance to an

1 The posterior amniotic canal of Chelonians (Mitsukuri) and Hatteria
(Dendy) requires special explanation.

TROPHOBLAST AND SEROSA, 593

insect egg with the embryonic tract lying upon the yolk (Figs.
1, a and B).

At a later stage the trophic vesicle develops a caudal exten-
sion, so {that the embryo attains a more central position,
although the cephalic portion of the vesicle is, as a rule, con-
siderably larger than the posterior portion (Fig. 2, A).

F B

Fic. 1.—a. Embryo of Peripatus nove-britannia, with embryonic
tract at posterior ventral extremity of trophic or blastodermic vesicle
(original).

B. Egg of Gryllus, with embryonic tract at the posterior ventral
surface of the vitellus. (After Heymons.)

The young embryo rests upon the trophic vesicle as on a
cushion, and there can be little doubt that in the fresh condi-
tion, when the cavity of the vessel is filled to repletion with its
nutrient fluid contents, the embryo lies in a sort of lap or bay
or depression, bounded by turgid lips, produced by the inflation
of the surrounding thin wall of the vesicle. I cannot state
this as a definite fact as I did not observe the living embryos.
Everything, however, points to such a state of things.

594 ARTHUR WILLEY.

As the embryo advances in development, the relative dimen-
sions of the trophic organ decrease pari passu with the
growth of the embryo. The posterior portion of the vesicle is
the first to disappear, being used up in the formation of the
dorsal body-wall of the embryo. When there is no longer any
caudal extension of the vesicle there may still be observed a

A

Fig. 2.—a. Embryo of Peripatus nove-britannis, showing posterior
extension of trophic vesicle and primary ventral flexure of embryonic
tract. (Original.)

B. Egg of Strongylosoma guerinii, Gerv. (a Diplopod), showing
primary ventral flexure of embryonic tract. (After Metschnikoff.)

large head-fold, which, in consequence of the cephalic flexure
subsequently undergone by the embryo, is reflected like a cap
over the ventral surface of the embryo (14, pl. ii, fig. 35).

TROPHOBLAST AND SEROSA. 595

This head-fold or cephalic portion of the trophic
vesicle is attached tothe embryointhenuchal region;
in other words, the point at which the final resorption of the
trophic organ of these embryos takes place, lies in the nuchal
region (cf. Fig. 3,4). This fact is of capital importance when
taken in comparison with the remarkable phenomena attendant
upon the involution of the serosa of insect embryos.

a B

Fic. 3.—a. Embryo of Peripatus nove-britan nia, in which the posterior
extension of trophic vesicle has almost entirely disappeared. The ante-
rior portion of the vesicle is present as a large lobe attached to the
head. The figure also illustrates the caudal flexure in addition to the
ventral flexure of the embryo. (After Willey, ‘ Zoological Results,’
Cambridge, 1898.)

B. Egg of Gryllus, at stage after the eversion of the embryo. am.
Amnion reflected over dorsum of embryo. ser. Serosa withdrawn to
anterior portion of vitellus and attached to head of embryo, like the
trophic vesicle in a. A slight caudal flexure is present, but no ventral
flexure. (After Heymons.)

After the amnion of the Insecta Pterygota is formed by
the fusion of tail-fold, lateral folds, and head-folds, the amniotic

596 ARTHUR WILLEY.

cavity is present as a closed sac, and the amnion itself has
completely separated from the rest of the blastoderm, which
has, ipso facto, become converted into the serosa.

At a later stage the amnion enters for a second time, i.e.
secondarily, into relations with the serosa, fusing with it near
the head end of the embryo. The fused area becomes perfo-
rated and the embryo undergoes a process of eversion (revolu-

Micr
'

SSD VY. St1

“. St 10

Fig. 4.—Hge of Forficula after eversion of embryo, with caudal flexure.
Micr. Micropyle. am. Amnion reflected over dorsum. sev. Dorsal organ
(remains of serosa). S¢,.—St,). Stigmata. S7/b/. Stink-gland. (After
Heymons.)

tion of Wheeler, Umrollung of Heymons and others), everting
itself through the opening thus produced.

The secondary fusion of the amnion with the serosa is the
initial stage in the involution or retrogressive development of
the serosa. By the eversion of the embryo, the amnion

TROPHOBLAST AND SEROSA. 597

becomes reflected over the surface of the yolk, while the serosa
shrinks away towards the anterior end of the egg (Fig. 3, B).
Finally the serosa shrivels up until it constitutes the dorsal
organ of the insect embryo, which in primitive in-
sects is situated in thenuchal region, e.g. Lepisma,
Gryllus, Forficula (cf. Fig. 4). The dorsal organ is subse-
quently withdrawn into the yolk, where it undergoes disinte-
gration and absorption.

In the more primitive insects, therefore, the point at which
the absorption of the serosa into the yolk takes place, les in
the nuchal region.

The resemblance between the trophic vesicle of the embryo
of Peripatus nove-britanniz shown in Fig. 3, a, with the
serosa of the embryo of Gryllus copied from Heymons in Fig.
3, B, is too striking to need further comment. If this were all,
I should almost regard the suggested homology between the
trophoblast of my Peripatus embryos and the serosa of insect
embryos asafait accompli. But it is not all.

The indusium of the Locustidz described by Wheeler (12),
and the dorsal organ (‘‘ micropyle,” “‘ micropylar organ”’) of
the Poduride have to be taken into consideration. They can
only be accounted for by invoking the aid of hypothesis ; but
then every explanation of any structure or phenomenon is
more or less hypothetical.

Since the embryos of Peripatus nove-britanniz have
enabled us for the first time, so far as the Invertebrata are
concerned, to seize upon the most precious conception of the
trophoblast, I think we are not only justified but absolutely
called upon to face any difficulties which may obscure the
application of this conception in all its phylogenetic ramifica-
tions.

To account for the dorsal organ of the Poduridz and the
indusium of the Locustide, we must call to our aid the prin-
ciple of substitution. Without doubt Kleinenberg has left us
a noble legacy in his principle of substitution, applicable as it
is both to ontogenetic and phylogenetic changes.

I assume that the transformation of the trophoblast into the

598 ARTHUR WILLEY.

serosa took place by substitution, and my meaning can best be
illustrated by an observation of my own with regard to the
endoderm in the embryos of P. nove-britanniz. The tro-
phic cavity of the embryo becomes the gastral cavity of the
adult, but the exceedingly thin layer of endoderm lining the
trophic cavity does not become the definitive endoderm by
simple continuous growth. On the contrary, the embryonic
endoderm undergoes histolytic changes, the cells losing their
mutual contiguity, rounding up and migrating as free tro-
phocytes into the trophic cavity. After this remarkable
histolysis, the endoderm reconstitutes itself, secretes a basal
membrane, and produces the high columnar epithelium of the
gut.

Thus, while the gastral endoderm of the adult is derived
from the trophic endoderm of the embryo, it is not exactly the
same as the latter since histolytic changes intervene.

The three principal phases in the development of the endo-
derm in the embryos of P. nove-britanniz may be sum-
marised as follows :

IL. i; slike
Trophic endoderm. Trophocytes. Gastral endoderm.

I regard this method of transformation of trophic endo-
derm into gastral endoderm as a special case of substitution,
and well adapted to elucidate the significance of the assump-
tion that the transformation of the trophoblast into the serosa
has taken place by substitution.

Many other still clearer cases of epithelial substitution could
be adduced. In insects, after the eversion of the embryo and
consequent reflection of the amnion over the yolk, the amnion
forms the provisional dorsal wall of the embryo. But it
becomes replaced later by true ectoderm which grows out from
the pleural edges of the embryo, while the amnion itself under-
goes disintegration and absorption (Wheeler, 11, 12; Hey-
mons, 2). The origin of the definitive hypodermis of holome-
tabolous insects from the imaginal discs is another case in
point.

TROPHOBLAST AND SEROSA. 599

The above are actual instances of ontogenetic substitution.
Instances of phylogenetic substitution can, in the nature of
things, only be hypothetical.

In the embryos of the Poduridz there is no amnion (Uljanin
[10], Wheeler [12] ), and therefore no serosa. The blastoderm,
however, is not entirely employed in forming the dorsal body
wall of the embryo, since there is a dorsal organ which is some-
times spoken of under the name micropylar organ (Wheeler)
and spherical organ (Uljanin). This dorsal organ is a local
thickening of the blastoderm lying in front of the embryonic
tract ; but at a later stage, when the true topographical rela-
tions are established, it is found to lie in the nuchal region (ef.
Lemoine’s figures [8] ). It is destined to be absorbed like the
dorsal organ of higher insects, which is the product of the
involution of the serosa.

Appearances are in favour of the Poduride never having
had an amnion, but the embryos present the remarkable pecu-
liarity that the blastoderm outgrows the vitellus to
such an extent that it is thrown into numerous complicated
folds at the surface of the egg. ‘“ La superficie de la couche
blastodermique,” says Uljanin (10, p. xvii), ‘ devient de plus
en plus inégale ; cette superficie, beaucoup plus agrandie A
cause des inégalités de la couche blastodermique, sécréte une
membrane cuticulaire, cuticule blastodermique, que l’on
voit a travers le chorion de l’ceuf fortement plissée et suivant
toutes les inégalités du blastoderme épaissi.” Lemoine (8)
figures eggs with such a folded blastoderm.

The fact that the blastoderm of Poduridz out-
grows the vitellus, involuntarily suggests an ata-
vistic repetition of a state in which the size of the
egg bore no relation whatever to the dimensions of
the blastodermic vesicle.

The indusium of Locustid embryos is a remarkable structure
discovered by Wheeler (12). It arises as a small circular
local thickening of the serosa immediately in front of the head
of the embryo, i.e. in a position corresponding to the nuchal

600 ARTHUR WILLEY.

region of the embryo at a later stage. According to Wheeler’s
perfectly clear account (12), the indusial thickening becomes
overgrown by the surrounding serosa, sinks below the surface,
becoming entirely separated from the serosa of which it isa
derivative, and gradually spreads over the entire egg until its
growing edges meet on the dorsal side of the vitellus and fuse
together. The indusium then forms a complete envelope round
the egg below the serosa, and it completely usurps the functions
of the latter. Why this substitution of an indusium for the
serosa should take place in Locustid embryos is a question not
easily answered. At present all that can be said is that it
happens so and there’s an end.

In the course of its development, Wheeler found that the

AV LUE

Fic. 5.—Median section of egg of Anurida maritima, showing dorsal
organ with vacuolar cell contents, as a thickening of the blastoderm.
(After Wheeler.)

indusium was subject to great time variations, growth varia-
tions, and even numerical variations. In two young embryos

TROPHOBLAST AND SBEROSA. 601

he found that each possessed two separate indusial thickenings
of the serosa or blastoderm.

In its first appearance as a local thickening of the serosa,
the indusium has essentially the same topographical relations
with regard to the nuchal region as the dorsal organ of the
Poduride.

Wheeler says (12, p. 56): ‘ Although much simpler in its
structure, I do not hesitate to homologise this ‘ micropylar ’
organ in Anurida and the Poduride in general with the
indusium of Xiphidium.”

Again, with reference to the indusium Wheeler says (p. 58) :
“That the organ is rudimental is shown by its tendency to
vary, especially during the earlier stages of its development ;
that it still performs some function is indicated by its some-
what complicated later development, and by its survival in but
very few forms [Locustidz] out of the vast group of Pterygo-
tous insects. This seeming paradox may be explained if we
suppose that the indusium was on the verge of disappearing,
being the last rudiment of some very ancient struc-
ture. Assucha rudiment it no longer fell under the influence
of natural selection, and for this reason began to vary consider-
ably like other rudimental organs. Some of these fortuitous
variations may have come to be advantageous to the embryo,
and were perhaps again seized upon by natural selection, the
nearly extinct organ being thus resuscitated and again forced
to take an active part in the processes of development.”

Eventually the indusium undergoes retrogressive develop-
ment like the serosa of other forms, and produces an indusial
dorsal organ, which, however, is not absorbed but shed
(Wheeler [12]).

Accepting Wheeler’s view of the homology between the
dorsal organ of Poduride and the indusium of Locustide, and
admitting that these structures are not the same as the serosal
dorsal organ of other insects, we are now in a position to
discuss the bearing of these various observations. A certain
amount of repetition is unavoidable if there is to be any

602 ARTHUR WILLEY.

approach to clearness. In the following considerations, facts
and hypotheses are given indifferently.

1. The trophoblast of the embryos of Peripatus nove-
britannie is a mucous membrane in virtue of its close
proximity to the uterine mucosa; the chorionic membrane,
however, always intervening between the trophoblast and the
mucosa. This is a special case of Driesch’s principle of the
dependence of function on position— Die Zelle ist Funktion
der Lage.”’

2. The eggs of P. nove-britanniz have retained the
primitive alecithality which must have characterised the eggs
of the aquatic ancestor of Peripatus, which presumably spawned
in the water. In connection with the neotropical species of
Peripatus, Kennel was also of the opinion that their lack of
yolk was a primary deficiency, and not a secondary loss.

If this postulate as to the primary character of the aleci-
thality of the eggs of my Peripatus be not conceded, my theory
would be thereby practically rendered nugatory.

3. The viviparity of Peripatus is to be regarded as a direct
result of the transition from an aquatic to a terrestrial life.
This was also Kennel’s opinion.

4. In adaptation to the requirements of the intra-uterine
nutrition of the embryo, various provisional embryonic trophic
organs were evolved, e.g. dorsal (ectodermal) hump of P.
capensis (Sedgwick), trophic vesicle of P. nove-britannie,
trophic sac of P. edwardsii, yolk of P. nove-zealandiz.
It is neither easy nor necessary to decide which of these came
first, or whether they were evolved more or less independently.
Future investigation of the development of other species, such
as P. tholloni, Bouvier, from the Gaboon’district, may throw
light on this matter.!

1 For facts indicating the relatively primitive character of P. nove-
britanniz the reader may be referred to my paper (14). According to the
descriptions of Kennel and Sclater, the trophic organ of the embryos of the neo-
tropical Peripatus has the form of a spheroidal vesicle, into which the embryo
is inverted and to the wall of which it is united by an at first solid stalk. The

vitelline membrane undergoes early resorption (Kennel), so that the wall of
the sac comes into direct contiguity with the uterine mucosa. This inversion

TROPHOBLAST AND SEROSA. 603

5. We may, however, be quite certain that yolk is not the
most primitive kind of trophic organ for the intra-uterine
nutrition of the embryo, for the simple reason that the accumu-
lation of yolk, such as occurs in the eggs of the Australian
and New Zealand species of Peripatus, is a step towards
secondary oviposition on terra firma, a condition which is
actually realised in the case of the Victorian species, P.
oviparus Dendy.

6. The lecithality and deposition of the eggs of insects are
both secondary.

7. With oviparity the trophoblast necessarily ceased to act
as an absorbent mucous membrane.

8. The trophoblast therefore became transformed into the
blastoderm and its derivative the serosa, by substitution, this
being apparently the prevailing method by which an epithelial
transformation or metamorphosis is effected.

9. The serosal dorsal organ of insects is the product of the
ontogenetic involution of the serosa.

10. The phylogenetic involution of the trophoblast has also
to be accounted for, it being understood that the serosa was
not directly derived from a primitive trophoblast, but by sub-
stitution. Although phyletically related, the trophoblast and
serosa are therefore not identical structures. Accordingly we
may well expect to find some vestiges of the true primitive
trophoblast cropping up here and there.

11. I see in the dorsal organ of the Poduridz and in the
indusium of the Locustide, vestiges of the true trophoblast,
both of these structures presenting, in the earlier stages of
their development, many of the characters of a mucous mem-
brane. ‘This interpretation of facts at least accounts for the
otherwise meaningless dorsal organ of the Poduride (Fig. 5).

12. The dorsal organs of Crustacean embryos and the primi-
tive cumulus of Arachnids must require special explanation,
of the embryo shows that this is a special case, and not one from which
phylogenetic conclusions may safely be drawn, An inverted development
must necessarily be secondary, as compared with a development which is not
inverted.

vol, 41, PART 4,—NEW SERIES. ne

604 ARTHUR WILLBY.

my theory assuming the diphyletic origin of the Tracheata
and the monophyletic origin of the Hexapoda.!

1 T am aware that the assumption as to the monophyletic origin of the
Hexapoda is a very grave one to make, and I only make it in the most general
way. It would take one much too far afield to attempt to justify and explain
such an assumption in detail. At least two authors, Pocock and Kingsley,
have made an onslaught against the group of the Myriapoda; and Pocock
(R. I. Pocock, “On the Classification of Tracheate Arthropoda,” ‘ Zool.
Anz.,’ xvi, 1893, p. 271) has gone so far as to suggest a scheme of classifica-
tion, in which the group of the Myriapoda finds no place. Taking the position
of the generative apertures as his criterion, he divides the Tracheata (apart
from Arachnida) into two large groups, the Progoneata (Diplopoda and
Pauropoda) and Opisthogoneata (Chilopoda, Symphyla, Hexapoda). The
group of the Symphyla (Scolopendrella) is one of the utmost importance in
phylogenetic speculations, and there seems to have been some misunderstanding
about it, which however was subsequently corrected (see ‘ Nature,’ vol. xlix,
p. 124). Grassi discovered in 1886 that the unpaired genital pore of
Scolopendrella lies on the fourth body-segment, and the same statement
is contained in an important memoir by Kenyon (F. C. Kenyon, “ The Mor-
phology and Classification of the Pauropoda, with Notes on the Morphology
of the Diplopoda,” ‘Tuft’s College Studies,’ No. iv, 1895, see p. 1386).
The Symphyla would seem to be related to the Chilopoda in an analogous
manner to that in which the Pauropoda are related to the Diplopoda. It
would, in fact, appear that the Symphyla are distinctly intermediate between
the Chilopoda and Diplopoda (Chilognatha), having Chilopod affinities in
respect of their ambulatory appendages, and Diplopod affinities in respect of
their manducatory appendages and generative organs (consult table at end of
Kenyon’s memoir). Even if the two principal sub-divisions of the Myriapoda
were sharply divergent, I think the loss of the name Myriapoda would out-
weigh the profit of a classification which omitted all mention of the name.
As a matter of fact, although well-marked, the Chilopoda and Diplopoda are
not irretrievably divergent.

Nevertheless, it seems not impossible that the Hexapoda have differential
or heterogeneous relations to the Myriapoda. The Insecta Apterygota are
divisible into two well-defined sub-orders, as shown by Stummer-Traunfels,
namely the Apterygota entognatha (Campodeide, Japygide, and Col-
lembola) and the Apterygota ectognatha (Machilide, Lepismide).
(Rudolf Ritter v. Stummer-'raunfels, “ Vergleichende Untersuchungen tber
die Mundwerkzeuge der Thysanuren und Collembolen,” ‘S. B. Ak. Wien,’
Bd. c, Abth. 1, April, 1891.)

The Apterygota entognatha (e. g. Campodea) show strong external
resemblance to Scolopendrella, and the embryonic development (Collembola)

TROPHOBLAST AND SEROSA. 605

Such are the considerations by which I seek to establish a
phyletic relationship between the trophoblast, as it is presented
to us inthe embryos of Peripatus nove-britanniz, and the
serosa of insect embryos, and for my part I am disposed to be
content with this demonstration. Nevertheless a few words
on the origin of the insect amnion may not be out of place.
In accounting for the serosa we have not, eo ipso, accounted
for the amnion, notwithstanding that they are both derivatives,
and inseparable derivatives, of the blastoderm. The amniotic
cavity requires special treatment.

Until Heymons discovered the amnion and serosa of
Lepisma, he supposed that the embryonic membranes were
to be regarded as a new acquisition of the Insecta Ptery-
gota, and that there was no basis upon which to frame any
hypothesis as to their phylogenetic history (2).

For statements of their own views and references to the
views of others, the reader would find it well worth while to
consult the well-known memoirs of Wheeler and Heymons.
Suffice it to say here that Heymons (3) has conclusively shown
that the superficial or ectoblastic type of embryonic tract
as exemplified in Orthoptera is more primitive than the
immersed or entoblastic type as exemplified in Libellulide
and Ephemeride.

So far as I have been able to form an opinion from the data
which are to hand, I think that the amniotic cavity of
iusect embryos was originally a product of invagina-
tion, and that this invagination was primitively de-
rived from and associated with a ventral flexure of
the embryo.

As already mentioned, Heymons has in fact found that the
amniotic cavity of Lepisma is formed by invagination, which is
brought about by the ventral flexure of the embryo, the orifice
strikingly calls to mind that of the Chilopoda, in that a dorsal flexure of the
embryo precedes the ventral flexure.

The Apterygota ectognatha, as exemplified in Lepisma, recall in their
embryonic development that of the Diplopoda in respect of the primary
ventral flexure of the embryo.

606 ARTHUR WILLEY.

of invagination narrowing down to a small pore which persists
as the amniotic pore. My views, however, differ from those of
Heymons, in that I regard the ventral flexure as primary,
while Heymons regards the dorsal flexure of young embryos of
Chilopoda and Poduridz as primary, and he thinks that this
dorsal flexure has been lost by the embryo of Lepisma.
Knowing what takes place in the embryos of P. nove-bri-
tannix, I am quite satisfied that the early ventral flexure
of the Chilognatha (Diplopoda) is primitive, and that the
provisional dorsal flexure of the embryos of Chilopoda and
Poduride is a transitory, cenogenetic, intercalated phase of
development.

The ventral flexure of the embryo of Lepisma is therefore to
be regarded as comparable with the primitive ventral flexure
which oceurs in Chilognatha and in Peripatus (see below).

Heymons says (4, p. 620) :-—“ Sucht man bei hoéheren
Insecten nach einem Anklang an die ausgesprochen ventrale
Kriimmung von Lepisma, so ist dieser wohl zweifellos in der
von mir Caudalkriimmung genannten Umbiegung des hinteren

Fig. 6.—Sagittal section through embryo of Lepisma saccharina, to
show ventral flexure, amniotic cavity, and amniotic pore. Mesoderm and
yolk omitted. @. Amnion. ¢. Amniotic cavity. 2. Head end of embryo.
p. Amniotie pore. s. Serosa, ¢. Tail end of embryo. (Simplified after
Heymons.)

Kérperendes vieler Insectenembryonen gegeben. Dieselbe
zeigt sich bei den Orthopteren, den Odonaten, Ephemeriden

TROPTIOBLAST AND SBROSA, 607

fast in allen bisher bekannt gewordenen FViillen, und sie ge-
langt selbst dann zum Ausdruck, wenn der Keimstreifen im
Ubrigen vollkommen dorsal gekriimmt ist.”

I cannot follow Heymons in the above comparison between
the ventral flexure of his Lepisma embryos and the caudal
flexure of other insect embryos.!

The embryos of P. nove-britanniw undergo three
absolutely distinet ventral flexures, namely, (1) the primitive
ventral flexure, (2) the caudal flexure, and (3) the cephalic
flexure ; the last-named flexure not being constant. It is very
important to note that there is no connection whatever
between the primitive ventral flexure and the caudal
flexure (see Fig. 3, a).

The primitive ventral flexure serves the immediate purpose
of releasing the primitive streak, i.e. the principal growing
point of the embryo, from the rest of the embryonic tract, so as
to enable it to continue its growth independently of the
trophic vesicle ; and it also serves the essential purpose of keep-
ing the growing embryo, during its early stages, in as small a
space as possible, in order to admit of the trophoblast coming
into contact with the uterine mucosa (the chorion intervening)
over as large an area as possible, The flexure of the embryo
is sufficiently accounted for on such physiological grounds as
these.

The occurrence of three ventral flexures in the
embryos of P, nove-britanniz cannot be too strongly
emphasised.

In the Myriapods the ventral flexure of the embryo is an
essential and characteristic feature of the development, and is
to be considered in connection with the absence of an amnion.
It takes place at a very early stage in the Chilognatha
(Fig. 2, 8), while in the Chilopoda it is preceded by a dorsal
flexure. The ventral flexure of the Myriapods (with special
reference to the Chilognatha) is quite obviously to be iden-

' It is, however, quite possible that, in certain cases, the caudal flexure
and the primary ventral flexure might coincide,

608 ARTHUR WILLEY.

tified with the primitive ventral flexure of the embryo of my
Peripatus (cf. Fig. 2, a and B), with the difference that in the
Myriapods the flexure goes much deeper and is much more
pronounced than in Peripatus (cf. Metschnikoff [9], and
Korschelt and Heider [7]).'

Thus in the Myriapods (as in Peripatus) the ventral flexure
of the embryo may be said to be a product of invagination, but
the orifice of invagination does not narrow down at all, remain-
ing freely open on all sides.

In Lepisma, as shown by the observations of Heymons, we
have a primary ventral flexure of the embryo, strictly com-
parable with the ventral flexure of the Myriapod embryo, and
the invagination which produces the ventral flexure is accom-
panied by a narrowing of the orifice of invagination, thus
giving rise to an amniotic cavity opening by an amniotic pore.

The preceding remarks may be resumed in tabular form as

follows :
P. NOVH-BRITANNIA. MyRIAPopA. LEPISMA,
Shallow ventral flexure.’ Deep ventral flexure. Amniotic cavity and

Amniotic pore.

Thus although the amnion itself first appears within the
group of the Hexapoda, it does not owe its origin to purely
mechanical causes, as has been so often supposed, but can be
traced back, through the link supplied by Lepisma, to the
primitive ventral flexure of ancestral forms.

I think I have now said enough to explain my theory of the
embryonic membranes of insects, and I give it for what it is
worth. I have endeavoured to trace the serosa back to a
primitive trophoblast, and the amniotic cavity back to a primi-
tive ventral flexure.

New Museums, CAMBRIDGE ;
September 22nd, 1898.

1 In the embryos of P. nove-britanniz the primary ventral flexure
does not involve the trophic vesicle; whereas in Myriapods it does involve

the vitellus.
2 That is to say, shallow in its primary relations to the trophic vesicle.

12.

13.

14.

TROPHOBLAST AND SEROSA. 609

LITERATURE,

. Herper, K.—‘ Die Embryonalentwickelung von Hydrophilus piceus,

L.,’ Jena (Gustav Fischer), 1889.

. Hermons, R.—‘ Die Embryonalentwickelung von Dermapteren und

Orthopteren,’ Jena (Gustav Fischer), 1895.

. Heymons, R.—‘‘ Grundziige der Entwickelung und des Kérperbanes von

Odonaten und Ephemeriden,” ‘ Abh. k. Akad. wiss.,’ Berlin, 1896,

. Hrymons, R.—‘“ Entwickelungsgeschichtliche Untersuchungen an Le-

pisma saccharina,’ L., ‘ Zeit. f. wiss. Zool.,’ Bd. lxii, p. 583, 1897.

. Husrecat, A. A. W.—“ Keimblatterbildung und Placentation des Igels,”

‘ Anat. Anz.,’ iii, p. 510, 1888.

. Huprecut, A. A. W.—‘‘ Die Phylogenese des Amnions und die Bedeut-

ung des Trophoblastes,” ‘ Ver. Kon. Akad. Wetenschappen,’ Amster-
dam, 1895.

. Korscuett, E., und Herper, K.—‘ Lehrbuch der vergleichenden Ent-

wickelungsgeschichte der wirbellosen Thiere,’ Jena, 1890-3.

. Lemorne, N.—“ Recherches sur le développement des Podurelles,” re-

printed from ‘ Association frangaise pour l’avancement des Sciences,
Congres de la Rochelle, 1882,’ Paris, Imprimerie Chaix, 1883.

. Metscunikorr, H.—* Embryologie der doppeltfiissigen Myriapoden

(Chilognatha),” ‘ Zeit. f. wiss. Zool.,’ Bd. xxiv, p. 253, 1874.

. Utsanin, W. N.—‘‘ Développement des Podurelles,”’ ‘Arch, de Zool.

exper.,’ t. v, p. xvil, 1876.

. WueELeR, W. M.—“The Embryology of Blatta germanica and

Doryphora decemlineata,” ‘Journ. Morph.,’ iii, p. 291, 1889.

Wueeter, W. M.—‘‘A Contribution to Insect Embryology,” ‘Journ.
Morph.,’ viii, p. 1, 1893.

Wit, L.—‘ Entwickelungsgeschichte der viviparen Aphiden,” ‘ Zool.
Jahrb.’ (¢ Abth. f. Anat. u. Ont.’), iii, p. 201, 1889.

Will thought that the immersed or entoblastic (Wheeler) type of
embryonic tract is the more primitive (opinion shared by Heider and
Wheeler), and ‘dass das Amnion als ein riickgebildeter oder meta-
morphosirter Theil des Keimstreifens aufzufassen ist.”

Witury, A.—‘‘The Anatomy and Development of Peripatus nove-
britannie,” ‘Zoological Results, &c.,’ part i, p. 1, 1898, Cambridge
University Press.

AGENDA.

Scumipt, P.— Beitrage zur Kenntniss der niederen Myriapoden,” ‘ Zeit. f.

wiss. Zool.,’ Bd. lix, 1895, p. 436.

Cuayroitn, A. M.—“ The Embryology and Odgenesis of Anurida mari-

tima (Guér.),” ‘Journ. Morph.,’ xiv, p. 219, 1898.

> every gee Eile ae bee EBV “Alls dlb tive he sdeondt

oa \ aCe Andee
ie nS ip tt) 20% abel Hi Pobanhdir Hie rs
oa rowhal tenlealé «ghar A
ful ay ots) SF) de SON ye vat Aral Kt ay ww yhithes yeah

be

tie an? ee ple A Oke ea! at Cane Jn Wea sk
bel ha, ruth 4h) Das A if at Ta) i) ‘ain nile ile Sieh rt;
: ie» b “Wee as met
sh aad he? Heh a YO NS an elite wR j

PAR UY, ta ee ht at, 1):

| ay. AME Al . ail Sofie y d

peel Od hi eerie | wre (ab ‘brad

eS , en “lf dike aes cena

; oe

i ee ieee by milan | Pier ee

en Hd o ue i sty WW ONE Ni rs

; oy wl it rhar* a pes

Sh. fhe ice eh
ir ines hee iki ue | ine
peer pe . a is Vee ak VWirsde :

ray he wi Sage sci ise

cet ne Me yi (ioe) ie oe ari :

oan. atAl a. en

Jon i CeO ee TA tae a 1% UVES os

A's. Oe
yi , ary ari: ‘5 | 4 ’ } d 4 Aa. Pig
a ve) : j Fr, ' iF ; Ai 2a eo

) j | iF ! A \ y

i ff - Lig Dla a at pups tay

pee at re ual th ro | jf ine

G ra hha tal Mh S00) Oh
‘ak giant gue tat anit ae daly ‘Sai i
: sf t ingtrawa é ie he oh Pokal
ja pill aS 2 Seay hn Ta
WL a <

Non een deeh tae | he tole ye Taek i ee "Ss
| | evi ae fe
ee ee overeat .. WtOniia
BAN: Ober tensty * reba cin ee Sets} atl

aye me ———— ee Bees Vile ve, ee

LBD EX TO VOL. Al,

NEW SERIES.

Anthozoa, structure and formation of
the calcareous skeleton of, by G. C.
Bourne, 499

Arenicola, habits and structure of, by
Gamble and Ashworth, 1

Ashworth and Gamble on the habits
and structure of Arenicola ma-
rina, l

Assheton, Richard, on the segmenta-
tion of the ovum of the sheep,
with observations on the hypothesis
of a hypoblastic origin of the Tro-
phoblast, 205

= on the develop-
ment of the pig during the first
ten days, 329

Bathanalia, anatomy of, by J. E. $
Moore, 181

Bourne, G. C., on the structure and
formation of the calcareous skeleton
of the Anthozoa, 499

Brown, A. W., on Tetracotyle
petromyzontis, a parasite of the
brain of Ammoceetes, 489

Chlorophylloid pigments in Inverte-
brata, by M. I. Newbigin, 391

Evans on two new species of Spon-
gilla from Lake Tanganyika, 471

Gamble and Ashworth on the habits
and structure of Arenicola ma-
rina, 1

Gastric glands of Mammalia, by R.
R. Hensley, 361

Glands, gastric, of Mammalia, by
Hensley, 361

Glycera, nephridia of, by Goodrich,
439

Goniada, nephridia of, by Goodrich,
439

Goodrich, Edwin S., on the nephridia
of Polycheta (Part IT), 439

53 on the reno-pericardial
canals of Patella, 323

Hair, structure and development of, in
Monotremes and Marsupials, by
Spencer and Sweet, 549

Harmer, Sydney F., on the develop-
ment of Tubulipora, 73

Heart-body and ceelomic fluid of cer-
tain Polycheta, by Lionel James
Picton, 263

Hensley on the structure of the Mam-
malian gastric glands, 361

Insects, embryonic membranes of, by
Willey, 589

Larval crustacean, by Lister, 433

612

Lister on a Stomatopod larva, 433

Marsupials, see Monotremes.

Miller, C. O., on the aseptic cultiva-
tion of Mycetozoa, 43

Molluses of the great African lakes,
by J. E. 8. Moore, 159

Monotremes and Marsupials, struc-
ture and development of the hair
of, by Spencer and Sweet, 549

Moore on the Molluscs of the great
African lakes, 159
», on the anatomy of Typhobia
and on the new genus Bathanalia,
181
», on the hypothesis that Lake
Tanganyika represents anold Juras-
sic sea, 303

Mycetozoa, aseptic cultivation of, by
Dr. Caspar O. Miller, 43

Nephridia of the Polycheeta (Part IT),
by Goodrich, 439

Newbigin, M. I., on chlorophylloid
pigments in Invertebrata, 391

Ovum of the sheep, segmentation of,
by Assheton, 205

Patella, reno-pericardial canals of, by
Edwin 8. Goodrich, 323

Petromyzon (Ammoceetes), parasite of
the brain of, by A. W. Brown, 489

Picton on the heart-body and eclomic
fluid of certain Polycheta, 263

Pig, development of, during first ten
days, by Assheton, 329

Pigments, chlorophylloid, in Inverte-
brata, by M. I. Newbigin, 391

Polycheeta, nephridia of (Part II), by
Goodrich, 439

- heart-body and ccelomic

fluid of, by Picton, 268

INDEX.

Segmentation of the ovum of the
sheep, by Assheton, 205

Serosa and trophoblast, by Willey,
589

Skeleton, calcareous, of Anthozoa, by
G. C. Bourne, 499

Spencer, Professor Baldwin, and
Georgina Sweet, on the structure
and development of the hairs of
Monotremes and Marsupials, 549

Spongilla, two new species of, from
Lake Tanganyika, by Evans, 471

Stomatopod larva, by Lister, 433

Sweet, see Spencer.

Tanganyika, an old Jurassic sea, by

J. E. 8. Moore, 303
ss two new species of Spon-

gilla from, by Evans, 471

Teeth of Gadide, by Tomes, 459

Tetracotyle petromyzontis, a
parasite of the brain of Ammoceetes,
by A. W. Brown, 489

Tomes on differences in the histologi-
cal structure of teeth occurring
within a single family—the Gadide,
459

Trophoblast and serosa, by Willey,
589

S hypoblastic origin of,

205

Tubulipora, development of, by Har-
mer, 73

Typhobia, anatomy of, by J. EK. S.
Moore, 181

Willey, Arthur, on trophoblast and
serosa, a contribution to the
morphology of the embryonic mem-
branes of insects, 589

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

b i ae hal Pree
i 7 rt 7 : f ,
.* : owes Pe a ee bat Sete, Wee a
i r - nag We a ad Pe,
~ ‘ S 5 | oe y
a Pe ; Bath hem
3 4 7 J f) iw
4 ad
‘ i ns =
! . ' .
f { >
;
\
x i}
q
Y ’
‘
.
Z 1
-
; a
7
' %
' i}
at
i
*
‘
i
i
1
ba ‘
F 1
i
' i if is
iad rt

f ful eet
2) OM ra ioe

Date Due

ERNST MAYR

INA

3 2044 110

i

RY

HNN

319 845

pe

Pac — ™ e+ ‘

a 7 * » , 2h. i
rae : 2 bri poey.

Me Senge 5 gta ee eee ae