I

oj

I

si

r =0

t
I

i 3-
_U
1 0~

\ m

! ^
a
a

r^

a
m
a

o

PHYCOLOGICAL MEMOIRS.

PHYCOLOGICAL MEMOIRS

BEING

Researches made in the Botanical Department

of the British Museum.

EDITED BY

GEORGE MURRAY, F.R.S.E., F.L.S.

WITH TWENTY LITHOGRAPHIC PLATES.

/., //., & in.

DULAU AND CO., SOHO SQUARE, LONDON.

1892-5.

6 *
TO

WILLIAM CARRUTHERS ESQ., F.R.S., F.L.S., F.G.S.,

KEEPER OF THE DEPARTMENT OF BOTANY,
BRITISH MUSEUM,

IN GRATITUDE AND FRIENDSHIP.

PREFACE.

IN every case but one I have verified every fact described
in these Memoirs for the first time. The exception is Mr.
Batters' Paper, which 1 did not presume to revise in this
fashion, since his subject is one of which he knows much more
than I do. I must cordially thank my fellow-workers for their
kindness, consideration, and patience.

GEORGE MURRAY.

CONTENTS.

PART I. APRIL, 1892.

PAGE.

I. ON SPLACHNIDIUM KUGOSUM GREV., THE TYPE OF A NEW ORDER
OF ALG^:. By Margaret O. Mitchell, former Bathurst Student,
Newnham College, Cambridge, and Frances G. Whitting, former
Student of Newnham College, Cambridge . . . . i

II. ON A FOSSIL ALGA BELONGING TO THE GENUS CAULERPA, FROM

THE OOLITE. By The Editor 1 1

III. ON THE STRUCTURE OF DICTYOSPHMRIA DECNE. By The Editor 16

IV. ON MALFORMATIONS OF ASCOPHYLLUM AND DESMARESTIA. By

Ethel Sard Barton . . . . . . . . .21

V. ON CONCHOCELIS, A NEW GENUS OF PERFORATING ALG^E. By E.

A. L. Batters, B.A., LL.B., F.L.S. 25

PART II. MAY, 1893.

VI. NOTES ON THE MORPHOLOGY OF THE FUCACE& : . . 29

COCCOPHORA LANGSDORFII GREV. By A. Lorrain Smith . . 30

SEIROCOCCUS AXILLARIS GREV. By A. Lot-rain Smith . . 32

XlPHOPHORA BlLLARDIERII MONT. By E. S. Barton . . 35

NOTHEIA ANOMALA BAIL, and HARV. By M. O. Mitchell . 36

SARCOPHYLUS POTATORUM KUTZ. By F. G. Whiting . . 38

VII. ON CHLOROCYSTIS SARCOPHYCI, \ NEW ENDOPHYTIC ALGA. By

F. G. Whitting ......... 41

VIII. ON HALICYSTIS AND VALONIA. By The Editor . . . -47

IX. ON THE STRUCTURE OF HYDROCLATHRUS BORY. By M. O. Mitchell 53

X. ON THE CRYPTOSTOMATA OF ADENOCYSTIS, ALARIA AND SAC-

CORHIZA. By The Editor 59

XI. A COMPARISON OF THE MARINE FLORAS OF THE WARM ATLANTIC,

INDIAN OCEAN, AND THE CAPE OF GOOD HOPE. By The Editor 65

(P

viii Contents.

PART III. APRIL, 1895.

PAGE.

XII. A NEW PART OF PACHYTHECA. By The Editor . . . .71

XIII. CALCAREOUS PEBBLES FORMED BY ALG/E. By The Editor. With

LIST OF DIATOMS. By E. Grove ...... 74

XIV. NOTES ON THE SORI OF MACROCYSTIS AND POSTELSIA. By A. L.

Smith and F. G. Whitting 83

XV. A COMPARISON OF THE ARCTIC AND ANTARCTIC MARINE FLORAS.

By The Editor and E. S. Barton 87

I.

ON SPLACHNIDIUM RUGOSUM, GREV., THE
TYPE OF A NEW ORDER OF ALGsE.

HISTORICAL. The earliest record of the subject of this research is the
description by Linnseus* of Ufoa rugosa, a plant collected at the Cape of
Good Hope by Koenig. His description was based on two specimens
preserved in his herbarium, one of them of mature stature, the other con-
siderably younger. Dawson Turner f describes and figures it under the
name of Fucus rugosus, and his illustration was drawn from the mature
specimen in the Linnean herbarium, of which it is a good representation.
He mentions that this plant had been collected in New Holland by Robert
Brown, who obtained it also at the Cape of Good Hope, as found on con-
sulting his herbarium in the British Museum. The next noteworthy record
is that by Suhr+, who, describing a specimen collected by Drege, assigned
it a place in the genus Dumontia, to which the dried specimens have a
certain outward resemblance. Greville next founded the genus Splack-
nidium for its reception, and it remains still the only representative of this
generic type. In the meantime the plant had been collected by Lesson ||
in New Zealand, and a short description and figures are given by Kiitzing
in Tabula Phycologica*^ Harvey gives details of its distribution, and
describes it in his Genera of South African Plants (p. 394), and in the
Flora of New Zealand (vol. ii., part ii., p. 215), and figures it on plate XIV.
of his Phycologia Australica. Lastly, we must note a paper by Mr. R. M.

* Mantissa, p. 311, and Syst. Nat. Ed. Gene/. //., p. 1391.

t Fuci, vol. iii., p. 118, tab. 185.

\ Beitrage zur Algenkunde in Flora, 1840, p. 275.

Synopsis, p. xxxvi.

|| Voyage d 'Astrolabe. Serf. Astrolab., p. 140.

IT Vol. x., p. 4.

A

2 Splacknidiwn rugosuin.

Laing,* in which the author gives a brief note of his observations on
Splachnidium rugosum.

Specimens collected by Harvey at the Cape and Australia have been
re-examined by us, as also those collected by Dr. Lyall in New Zealand,
and quoted by Harvey. In addition to these, we have examined specimens
from Seal Island, l Challenger Expedition ;' from the Cape, collected by
Mr. Boodle ; excellent spirit specimens from the same region by Mr.
Scott Elliott ; and from New Zealand by Dr. Berggren, all of them in the
British Museum.

The PLANT of Splachnidium rugosum consists of from five to six
branches or fronds, which spring from a small conical disc, the base of
which rests on the substratum to which the plant is attached. Each
branch has a well-marked main axis of cylindrical form, which gives off
at irregular intervals smaller branches, similar in form and structure. The
surface of the thallus is marked with numerous pits containing hairs in
their young state, and, at a later stage, reproductive organs.

THE MINUTE STRUCTURE OF THE THALLUS. Transverse sections of
a branch taken at any point between the apex and the base, but exclusive
of the pits just mentioned, show on the whole a uniform structure. The
external wall consists of rows of closely placed cells, on the inner side of
which lie strands of hyphal filaments surrounding a mass of mucilaginous
substance, which entirely fills the central part of the branch (plate II., fig. i).
Five layers of cells may be distinguished, the outer layers being clearly
defined from those that lie more internally by the shape of their cells. For
the sake of clearness in description, we will venture to call the two outer
layers epidermal, and the inner ones cortical layers. The cells of the
epidermal layers are small and approximately cubical in shape. They are
filled with a dense granular olive-brown protoplasm ; on staining with
carmine or picric aniline blue, a nucleus is apparent. The cells of the
cortical layers, when first formed, are polygonal ; at a later stage they
become ovate by unequal growth. The cell contents are olive-green when
young, changing to brown when maturity is reached.

The strands of filaments lie, some immediately below and in contact
with the innermost cortical layer, while the remainder, becoming more and
more separated from one another, lie in the periphery of the mucilage
with which the central part of the branch is filled ; they run in all direc-
tions, but take, on the whole, a longitudinal course from base to apex.
Each strand consists of one large central filament, surrounded by a varying
number of small ones. The former are made up of elongated thin-walled

f Observations on the Fucoidece of Banks 's Peninsula (Trans, and Proceedings of the
New Zealand Institute, vol. xviii., 1885).

Splachnidium rugosuni. 3

cells, which branch and anastomose irregularly (plate 1., fig. 2, a, b, c\ In
the young state there is active cell-division (plate II., fig. 3), and, as each
mature cell is greatly elongated, the granular protoplasm which at first fills
the whole cell becomes somewhat attenuated, and in some cases, though
this appearance is, perhaps, owing to the method of preservation, ruptures
altogether, so that a mass of protoplasm remains at each end of the cell
overlying the septum, and gradually tapering towards the centre of the
cell. The small filaments are produced from the large ones by lateral
branching ; the individual cells are very long, and the cell-cavity is small
(plate I., fig. 3, a, b, c, d).

The disc from which the branches spring is like the rest of the thallus
in colour, brown or olive-green, and it has no special organs of attachment.
A section taken in any direction through the lower part, i.e., the region
below the origin of the branches, shows a mass of closely woven un-
branched septate filaments (plate II., fig. 6), with thick walls and a small
amount of cell-contents. These are woven together so that interhyphal
spaces are absent ; where they approach the surface, their arrangement is
looser and their apices project freely ; these latter contain a dense granular
protoplasm. As in the Laminariacece and Fucacecz, the plant is covered
with a mucilaginous substance. In Splachnidium rugosum this is very
abundant, and not only covers the outside of the thallus, but is largely
developed within, and fills the branches. On staining some of the filaments
with haematoxylin, the stain is taken up by an otherwise invisible sub-
stance, which surrounds them, forming a thick coating on the outside of
the wall. The same appearance shows itself on staining a piece of the
thallus.

INCREASE IN LENGTH AND APICAL CELL. The chief increase in growth
of the plant takes place as the result of the activity of a zone of meristematic
tissue which surrounds the actual apical cell, and thus a contrast is offered
to the mode of growth which obtains in the Fucacece. The apical cell (plate
II., fig. 4), in Splachnidium rugosum differs entirely in appearance from the
surrounding meristem ; it has a pear-shaped body bounded by a thick
transparent mucilaginous wall, and a filiform basal process stretches
towards the interior of the thallus. This cell and the very active surround-
ing meristem are sunk in a slight depression ; the two outer layers of the
meristem by radial division form the epidermal cells, which continue to
divide radially until the part of the thallus in which they lie has reached
maturity. Some of the inner cells of the meristematic zone form the
adult cortical layers ; these are at first polygonal in shape, and when once
formed do not again divide, each increases in size and becomes oval ; by
subsequent separation of these cells the cortex keeps pace with the

4 Splachnidium rugosuvi.

growing epidermal layers. The cells of the meristematic zone lying
internally and immediately round the apical cell, divide tangentially and
give rise to large filaments ; the young cells undergo division and grow in
length, they branch and anastomose repeatedly, so that below the apex
there comes to be a filamentous network of cells filled with a dense olive
green protoplasm (plate IT., figs. 3 and 4) ; at this stage young lateral
filaments are as yet unformed.

The young cells of the outer epidermal layer, with the exception of those
lying directly round the apical cell, give rise to hairs. Each hair is produced
as a small protuberance which increases in length, and is finally cut off by
a wall ; cell divisions follow till a short filament is produced. The cells of
an adult hair are slightly elongated, sometimes the upper members of a
filament are spherical or bulge unilaterally, so that the whole hair takes on
a moniliform appearance, and the protoplasm which has shrunk away from
the wall lies in the centre as an irregular mass (plate II., fig. 4). All the
hairs bend towards the apical cell, and it is to be noted that in those which
are destined to become moniliform the bulging appears first on the side
away from the apex. These hairs are referred to by Mr. Laing as being
possible antheridia.*

The apical cell persists throughout the life of the plant, and not only
does it appear at the apex of the main branch, but cells of the same kind
are found at other points where growth is taking place, e.g., in the young
conceptacles and at the point of formation of lateral branches. It seems
as if its appearance were the herald of every local development.

When a new branch is about to be produced, one of the young
epidermal cells not far from the apex, and still in the region covered by
the apical hairs, develops a markedly mucilaginous wall, and takes on the
characters of an apical cell. The cells in its immediate neighbourhood
divide in the same way as they do at the primary apex, and cause the
thallus to grow out at this point (plate II., fig. 2) ; the filaments which
were originally below the thallus branch repeatedly and form a mass of
tissue, while the meiistem round the apical cell produces new filaments in
the same way as that already described. As the branch increases in size,
the basal filaments collect into strands which run at first radially, and then
longitudinally, below the cells of the newly formed cortex, and the new
branch shows precisely the same structure as the main axis. The apical
hairs remain, and are carried upwards by the growing branch.

THE DEVELOPMENT OF THE CONCEPTACLES. The first indication of
the formation of a conceptacle is seen in the alteration of one of the young
epidermal cells lying not far from the apical cell. By change in shape and

* Op. cit.

Splacknidium rugosiim. 5

character of its wall this becomes an exact copy of the apical cell. The
basal process grows between the cells of the thallus towards the centre of
the branch ; at its extremity it is club-shaped, and in some cases is united
to one of the cells of the large filaments, which forms a small outgrowth at
the point of contact (plate III., fig. i). It appears to us that this junction
may form a point of contrast with the vegetative apical cell described above,
and we may perhaps note a further although slight distinction in some indi-
cation of casting of the external layers of the mucilaginous wall in the latter
case fplate II., fig. 4), which we have never certainly observed in the concep-
tacles. This peculiarly modified cell of the conceptacle is homologous with
the initial cell of Prof. Bower,* but it undergoes no division and no further
development. The epidermal cells lying round it undergo division, and
the neighbouring cortical cells increase in size. These causes combine to
place the initial cell in a depression (plate II., fig. 7). Hairs arise from
the youngest epidermal cells, while others, which were formed earlier,
surround the mouth. At this stage, therefore, a young conceptacle is a
cylindrical depression in the thallus, with older hairs lining the mouth,
while young hairs are being produced round its sides, and the upper part of
the initial cell has come to stand out prominently from its base. There
is thus a marked contrast with the sexual conceptacle of Fucus serratus.
In Splachnidium rugosuni the hairs arise while the formation of the con-
ceptacle is in progress, and older hairs line the ostiole, which is never
closed, while in Fucus serratus a young conceptacle with a very small
ostiole is formed before the appearance of any hairs. In this plant also
the initial cell shrivels and finally almost disappears before the production
of any hairs ; in Splachnidium rugosuni the initial cell remains a most
prominent object in the conceptacle till it has reached maturity.

By repeated radial division of the lining cells the conceptacle gradually
enlarges, and the base comes to be of greater circumference than the
mouth ; the form of the conceptacle is, however, never flask-shaped, with
a small ostiole, as in many Fucacece. The hairs which are developed from
the lining cells are long, septate, and unbranched, and they increase in
length by successive divisions at the base. At this stage the conceptacles
present the appearance of the Fasergriibchen of many Fncacecz, the tufts of
hair extending a long distance beyond the ostioles (plate III., fig. 3). Later
reproductive organs are produced among the hairs in the form of long club-
shaped sacs filled with numerous spores (plate III., figs. 4 and 5). We shall
speak of these organs as sporangia. The number of sporangia in a single
conceptacle increases with its age, and in a mature conceptacle the hairs

* Formation of the Conceptacles in the Fucacccc. Q.J.M.S. iSSi.

6 Splachnidium rugosum.

are chiefly confined to the region surrounding the ostiole, only a few
remaining between the sporangia. Conceptacles which may be said to
have passed their maturity contain sporangia which have lost their
contents, only the clear hyaline walls remaining (plate III., fig. 5). The
sporangia are ruptured irregularly at the apex, and strongly resemble
those of the Laminariace& (cp. plate III., fig. 6, e, and fig. 8). They are
persistent, and in this offer a contrast to the empty oogonia of the
Fucacece, which disappear after a short time.

STRUCTURE AND DEVELOPMENT OF THE SPORANGIUM. A
sporangium arises as a papillose protuberance of one of the lining cells
of the conceptacle. This elongates and enlarges into a club-shaped body,
having its base sunk between the adjacent cells of the wall of the con-
ceptacle, and it becomes completely filled with a dark-coloured pro-
toplasm. When the sporangium has attained its full size, the protoplasm
becomes differentiated into spores, the differentiation proceeding simul-
taneously throughout the whole sporangium. The spores do not entirely
fill the cavity of the mature sporangium, the wall of which gives the
cellulose reaction when treated with iodine and sulphuric acid.

It should be noted that there is no division of the original papillose
protuberance into two cells, forming stalk-cell and sporangium respectively.
In all Fucacece, on the other hand, the oogonium is supported by a pedicel
cell, which is cut off from the protuberance in the very beginning ; only in
Pelvetia, so far as is known, is this cell somewhat obscure, owing to the
fact that it does not elongate to form a stalk as in other cases.

The sporangium is unilocular. Treatment with eau de javelle dissolves
the undifferentiated protoplasm or the spores within, but leaves no trace of
an intrasporangial network. Reinke* has shown that similar treatment of
the multilocular sporangia of some other Algae brings out a well-defined
intrasporangial network. That there is no inner membrane enclosing
the spores may also be demonstrated by the action of eau de javelle.

The spores when liberated by pressure from a sporangium of average
size are found to number from 500 to 600 ; they measure approximately
-2^0 mm. in diameter. It may be noted that this is also the average
diameter both of an antherozoid of Fucus serratn.s^ and of a zoospore of
the Laminar iacecz. It is calculated that the volume of an oosphere of
Fucus sermtus is 20,000-30,000 times as great as that of an antherozoid
of the same plant. These figures therefore express the ratio between the
volumes of the oosphere of Fucus serratus and of the spore of Splachnidium
rugosum.

* Reinke, Atlas deutscher ATeeresalgen, p. 49, tab. 32, fig. n.
t J. Thuret, Eludes Phycologiques.

Spiachnidium rugosum. 7

The reproductive organs here described are the only ones which have
been found in Spiachnidium rugosum ; and there are before us three
possible interpretations of these organs :

i. That they may be oogonia containing a very large number of small
oospheres.

2. That they may be large antheridia.

3. That they may be sporangia like those of the Laminariacecs containing
zoospores.

We will proceed to a consideration of each of those possibilities. In
connexion with the first we must note that Mr. R. M. Laing, in the paper
cited above, refers to these organs in the following words : ' The con-
ceptacle is surmounted by a ring of hairs, and in its interior contains a
number of unbranched hairs. The oogonia are obscurely pedicelled and
developed in the cells lining the wall of the cavity. Each oogonium gives
rise to a large number of oospheres, thus differing from all the other
Fucacecz that have hitherto been described. Each oosphere is very small
compared with the oospheres of any of the other Fucacea? A rough figure
(op. cit., plate X., fig. 7) accompanies this note, and it is apparent that the
author has anticipated the view that these bodies are oogonia with numerous
minute oospheres. Harvey* also regarded them as oogonia. Mr. Laing,
believing these bodies to be oogonia, and failing to find antheridia
developed as in typical Fucacece, suggests, as has been already said
that the antheridia are to be found in the peculiarly shaped hairs which
cover the apex of each branch ; that these are merely hairs has been
already shown.

That the sporangia of Spiachnidium rugosum, however, are not oogonia
homologous with those of the Fucacece appears obvious from the following
facts :

1. The enormous number of the spores (500-600), as compared with
the largest number of oospheres in an oogonium of the Fucacea, i.e.,
eight.

2. The small size of the spores, the bulk of an oosphere of the Fucacecs
being 20,000-30,000 times as great. Though deprecating size as an im-
portant characteristic, surely there is here a sufficiently wide margin to
entitle us to base an argument upon it.

3. The absence of a pedicel cell.

4. The absence of an inner membrane within the wall of the sporangium.

Secondly, that they are not antheridia homologous with those of the
Fucacecs appears plain from the following facts :

* Harvey. Fhycologia Austrcilica, vol. i.

8 Splacknidium rugosum.

1. The enormous number of the spores (500-600) compared with the
largest number of antherozoids contained in the antheridia of the Fucaccce,
i.e., 64.

2. The large size of the sporangium compared with that of an
antheridium.

3. The production of sporangia directly from the cells lining the
conceptacle, whereas antheridia are always borne on hairs within the
conceptacle.

In the face of these facts, it is impossible to regard the sporangia of
Splacknidium rugosum as homologous with either the oogonia or the
antheridia of the Fucacece. We are, therefore, led to the third possible
interpretation, i.e., that they are sporangia, containing zoospores homo-
logous with those of the Laminariacece. It is in those very points in which
the sporangia of Splachnidium differ so remarkably from the oogonia or the
antheridia of the Fucacece that we note their striking similarity to the
sporangia of the Laminariacece. The size of the sporangium, its unilocular
nature, its single wall, the number of the spores, the size of the spores,
and the persistence of the empty spore cases, all point definitely to the
conclusion that these sporangia are homologous with those of the Lami-
nariacece. A comparison of fig. 6 with fig. 8 (plate III.) will show the
general resemblance more clearly than it can be expressed in words.
Whether the spores produced in these sporangia germinate directly or
conjugate first, or, in the latter case, whether they show any sexual
differentiation, are questions which can only be settled by the examination
of fresh material and the cultivation of the spores.

While, then, the characteristics of Splachnidium rugosum are such
that it can be placed in no existing natural order of the PhaeopJiycece, its
nearest allies appear to be the Fucacece on the one hand and the Lami-
nariacece on the other. In its vegetative structure, i.e., in the nature of its
thallus and in the existence of conceptacles, Splachnidium bears a resem-
blance to the Fucacece ; it differs from them, however, in its mode of
growth, the former increasing by means of an apical meristem, while the
growth of the latter is due to divisions of an apical cell. In the presence
of an apical meristem Splachnidium approaches the Laminariacece, but at
the same time there is no plant in this order which has a cell corre-
sponding to the remarkable cell found at the apex of the main axis
and lateral branches of Splachnidium. In its reproduction it is allied
to the Laminariacece, and the production of sporangia within concep-
tacles might seem to point to a narrower limitation of the fertile sorus
of plants of this order, recalling the difference between perithecium and
apothecium in the fungi.

PKyc. Mem. Part I

BerjeaaJcHiohTeylth .

M.O.M,del.
SPL.AGHNIDIUM RUGOSUM

Hanha.ro imp

Phyc.Mem.Part.I .

Plate. II

'

e:- . ,.. laj..

C.JU. eel .

.1 RUGOSUIM Grev- .

Hanharfc

Pkyc .Mem . Part . 1

Plate. II I

B

M M ' W. del ,
SPL.AGHNID1UM rlUGOF.

.Mein.Parb.I.

Plate. IV.

Berjeau &L Higliley Jel.etTitk .

Hannarb imp .

CAULERPA CARKLJTHER51T G.Murr .

Splachnidium rugosum. 9

The sum of the characters of Splachnidium so expressly excludes it
from any existing natural order that there is no other course open to us
than to establish one for its reception under the name of Splachnidiacecz,
the main character of which shall be, reproduction by spores contained in
sporangia which are borne zcvV/^Vj conceptacles. It is possible that further
investigation of those genera of the Fucacece, about which little is at
present known, may lead to the discovery that such are really allied to
Splachnidium, and to their inclusion in the order. We may define this
order as follows, the subsidiary characters being necessarily provisional :
Splachnidiacece. Algae olivaceae, per fulcrum discoideum e fibris
radicalibus coalescentibus formatum, substrate affixse, frondibus ramosis,
externe e cellulis parenchymatibus interne e fills inter stratum mucosum
currentibus compositis. Sporse in sporangiis clavatis inclusae inter para-
nemata simplicia articulata in scaphidiis infra superficiem excavatis, per
totam frondem dispersis, collectae.

Finally we wish to record our thanks to Mr. George Murray for
pointing out this research to us, and for his help and criticism during
the progress of our work.

MARGARET O. MITCHELL
{Former Bathurst Student, Newnham College, Cambridge}.

FRANCES G. WHITTING

(Former Student of Newnham College, Cambridge).

EXPLANATION OF PLATES.

PLATE I.

Fig. i. Mature plant of Splachnidium rugosum. Natural size.
Fig. 2. a. b. c. Mode of junction of some of the large filaments ( x 235).
Fig. 3. a. b. c. d. Large filaments not far from the apex, giving rise to small
lateral branches ( x 235).

PLATE II.

Fig. i. Transverse section of Thallus (x235). a. Epidermal layers, b. Cortical
layers, c. Filament.

Fig. 2. Section of developing branch with its apical cell ( x 55).

Fig. 3. Young filaments from near the apex ( x 235).

Fig. 4. Section through the apex showing the apical cell with its basal process
and mucilaginous layers being shed externally, the apical hairs and the large fila-
ments in process of formation ( x 235).

B

io Splachnidium rugosum.

Fig. 5. Section across a branch showing the relative arrangement of tissues in
the thallus and position of conceptacles.

Fig. 6. Section of part of the disc showing interwoven filaments with free
apices ( x 235).

Fig. 7. Four stages in the development of a conceptacle (X235). a. The
youngest stage hardly to be distinguished from an epidermal cell. b. c. d. Older
stages ( x 235).

PLATE III.

Fig. i. Older conceptacles with young hairs in process of formation, a. Point
of contact of basal process with a large filament ( x 235).

Figs. 2 and 3. Further stages of Fig. i ( x 235).

Fig. 4. Conceptacle with developing sporangia. a. Initial cell. b. Young
sporangium.

Fig. 5. Mature conceptacle.

Fig. 6. a. b. c. d. Successive stages in the development of a sporangium ( x 235).
e. Empty sporangium ( x 235).

Fig. 7. Spores ( x 440).

Fig. 8. Sporangia of Sacchoriza bulbosa, after Thuret. Recherches sur les Zoospores
des Algues. Ann. Sci. Nat., ser. 3, torn, xiv., plate xxx., fig. 6 ( x 330).

II.

ON A FOSSIL ALGA BELONGING TO THE
GENUS CAULERPA FROM THE OOLITE.

ANY trustworthy evidence of the occurrence of Algae as fossil remains is
welcome, for the two reasons that the undoubted records of the preserva-
tion of such organisms are scanty, and that every one placed beyond doubt
enables us the better to deal with numerous cases hitherto uncertain, and
to judge of their claims to recognition. A large number of supposed
Algae have been described by Brongniart and other older writers under the
names of Chondrites, Confervites, Caulerpites, etc., and the validity of their
position has been disputed, as I believe, with success by botanists, who
have pointed out that such markings as are represented by these names
may well be the result of trails of animals and other casual impressions, or
the remains of other organisms, both plants and animals. Of the rights
of this discussion, which has been carried on mainly by Nathorst* and
Saporta,f I have no claim to judge other than that given by a working
knowledge of the forms of living Algae, but a recent examination of fossil
forms in the Geological Department of the British Museum has inclined
me to the point of view of Nathorst and those who with him reject the
validity of numerous recorded fossil Algae. No more eloquent testimony
could be sought against the views of Saporta than the plates of his own
memoir just cited. It is scarcely conceivable that a scientific inquirer
should not only accept such forms as many of those figured by Saporta,
but even use them in answer to a challenge.

* Memoire sur quelques traces d'animaux sans vertlbres, efc., et de leur portee palceonto-
logique, Kongl. Svensk. Vetenskaps-Akad. Handl., Bd. 18, No. 7, 1881, contains biblio-
graphy ; Nouvelles observations sur des traces animaux ; Ibid, Bd. 21, No. 14, 1886.

t A Propos dcs Algues fossiles par le Marquis de Saporta. Paris, G. Masson, 1882.

I I

12 A Fossil Alga.

If we take as trustworthy only such cases as exhibit microscopic
evidence of minute structure, we have before us a scanty record of forms.
The earliest of all is the Nematophycus of Carruthers* from the Devonian,
which exhibits its structure so plainly that there need be no hesitation in
accepting the position assigned it by its author in the Udote<z, a sub-order
of SipJionece. In the same beds there is found another organism,
PachytJieca, described by Sir Joseph Hooker, f and others; but its claims
to rank with the Algae need a more cautious examination. Mr. Barber +
has recently described it in two papers. In the former he obviously and
confessedly mistook the microscopic evidence, and in the second giving a
detailed and accurate description of its structure, while avoiding ' theoretical
considerations concerning the systematic relations of the plant,' he assigns
it a place near Cladophora, I have recently examined the beautiful slides
of Pachytheca prepared by Mr. John Storrie, who most kindly placed them
at my service, with some hope that it might prove to be an encrusting
algal growth surrounding some spherical Alga, such as a Valonia, and
pushing its rhizoids within the sphere; but this view, which at one time
appeared to me a highly probable explanation, was abandoned after
repeated examinations and comparisons with living forms. I confess to
feeling as little attached to this view as to Mr. Barber's determination of it
(near Cladophora), which it would be hard to get a phycologist to accept.

Its claims even to be reckoned an Alga may be regarded as still
doubtful. There is then a long break until we come to the Tertiary rocks,
in which besides the fossil Diatomacecs we have a series of Algae belonging
to the Dasycladece and other Siphonea described by Munier Chalmas, and
forms of Lithothamnion (Floridete), and several Characecz. These are those
which rest on the solid ground of structure, and the record is bare enough.

It would be going too far, however, to restrict our view absolutely to
fossil Algae exhibiting such structure, or even to those possessing a rind of
coal, which Nathorst considers indispensable evidence. With regard to the
latter view it has been pointed out that coal ' may entirely disappear in the
course of time from remains that are undoubtedly organic, if they are
deposited in a porous rock.' || There are many fossil plants the deter-
mination of which is unquestionably correct, affording no other characters
than those given by impressions of outward form, and I may cite as a

* Monthly Micr. Journ., 1872, vol. viii., p. 160 ; and 1873, vol. x., p. 208.
t Jotirn. Geol. Soc., 1853, p. 12 ; and Annals of Botany, vol. iii., No. x., 1889.
\ Annals of Botany, vol. iii., No. x.; also ibid., vol. v., No. xviii.

Comptes Rendus de FAcad. des Sc., vol. Ixxxv. (1877), pp. 814-817. See also Bull.
Soc. Geol. de France, 3 sdr., vol. vii., p. 66 1.

(I Graf zu Solms-Laubach. Fossil Botany, p. 47.

A Fossil Alga. 13

familiar example the case of many ferns in which, owing to the char-
acteristic outward forms on the one hand, and the excellence of the
impressions on the other, safe determinations of genera have been made.
An objection to the application of this method to the Algae might be
made at the outset, that it would only lead us back again along the old
path encumbered with the debris of Chondrites and the like. I venture to
think, however, while agreeing with Nathorst in the main, that he goes
too far in denying that this method may be employed with proper safe-
guards, such as a consideration of the nature of the bed as obtained from
other remains, the perfection of the remains themselves, and the degree
of their correspondence with the bodies of living Algse, especially if the
group in question be one of steadfast outward form. This latter character
is one unfortunately rarely to be met with among Algae. Their extremely
plastic bodies are readily moulded by their immediate environment, and
systematic literature is loaded with records of growth-forms erroneously
described as independent species.

Of all genera of Algae, one might almost say of plants, there is none
which varies through a wider range of outward forms than Cauhrpa ; but
the species themselves are, on the other hand, remarkably constant in their
characters. It is only here and there that we find one species run into
another, as the saying is. The variation is of a noteworthy character,
since the outward forms are astonishingly like those of the higher plants,
as their names cactoides, taxifolia, Abies marina, ericifolia, cupressoides,
tJiujoides, juniperoides, fontinaloides, Selago, Lycopodium^ liypnoides fre-
quently show, while their internal structure is one characteristic of the
genus in a remarkable degree. The whole body of the plant consists of
a single multinucleate cell, and the walls are sustained by an interwoven
system of trabeculae or cross beams and struts traversing the interior in
all directions, forming a great plexus, but at no point interrupting the
continuity of the cell. The Caulerpce form a group of Siphonea, the
same order to which belong those other Algae alluded to above, described
by Munier Chalmas, and Nematophyciis as well. Nature seems to have
shown in this genus the utmost possibilities of the siphoneous thallus in
an exhibition of the outward forms of the higher plants a prophecy,
as it were, of what was otherwise accomplished by the building up of
cellular and vascular tissues. Small wonder, then, if Caulerpa has been
chosen by palaeontologists as a convenient limbo for placing all imperfect
sorts and conditions of so-called fossil Algae. I have examined nearly
every species known to science of those at present existing, and have seen
a considerable number of them alive in West Indian waters, and with this
aid to judgment I may say that of all the described fossil Caulerpa and

14 A Fossil Alga.

Caitlerpites of which I have seen specimens and figures, there is not one
which might not with equal propriety be assigned a place elsewhere within
or without the vegetable kingdom.

With due consideration of all these facts, I yet propose to submit
reasons for believing that we have a veritable fossil Caulerpa represented
by specimens from the Kimmeridge clay. This fossil was discovered
by Mr. Damon, of Weymouth, and described * by him as an Equisetaceous
plant. He sent specimens to Mr. Carruthers, who very kindly placed
them in my hands for description, with the remark that they might
prove to be a true Caulerpa. I have examined other specimens of this
fossil in the Geological Department of the British Museum, and am
much indebted to the Director-General of the Geological Survey for the
permission to use and figure here (plate V., fig. I) an excellent specimen in
the Museum of Practical Geology, first pointed out to me by Mr. E. T.
Newton. These specimens represent a cast in the round (not a flattened
impression) of a verticillate organism. We are, therefore, on much safer
ground in describing it than we should be in dealing with a mere impres-
sion. In some specimens the hollow interior of the axis is now filled in
places with a solid substance crystallised from the minerals dissolved
in the water. The specimens come from a bed of marine origin, and
we may assume, without absolutely excluding other possibilities, that
the remains represent a sea-weed, if a plant at all. No zoologist has
been forthcoming who will claim it as an animal, and on the other hand
there are strong positive reasons for believing it to be an Alga. In the
first place, the stature, habit, and mode of branching favour this view,
and the correspondence with existing species of Caulerpa is so close,
that, were such a specimen clothed with cellulose walls and endued with
living contents to be placed before a phycologist, he would have little
hesitation in assigning it a place near Caulerpa cactoides, Ag. From this
species, in point of fact, the specimens in question differ in one single
respect of outward structure, viz., in the fossil we have verticillate ramenta
in place of the opposite pairs or, as it may be said, whorls reduced to two
members of the living species. In the lower whorls of the fossil (plate IV.,
fig. 2) there are thirteen ramenta in each whorl, and the number diminishes
towards the apex. In plate IV., fig. la, there is a whorl which must
be near the apex of a specimen exhibiting only six. The ramenta are
constricted at their insertion, and are club-shaped in form, as in C. cactoides.
Unfortunately we do not have the creeping surculus. The following may
be taken as a description of the species, with which I have much pleasure

* Supplement to the Geology of Wey mouth, Portland, and Coast of Dorset, 1888.
Plate xix., figs. 12, \2a.

A Fossil Alga. 15

in associating Mr. Carruthers' name: Caulerpa Carruthersii, n. sp.:
frondibus erectis, annulato-constrictis, simplicibus, ramentaceis, ramentis
verticillatis, clavato-obovatis, strictura conspicua a rachide sejunctis, quasi
articulatis ; surculus nondum inventus.

The species of Caulerpa at present existing are mostly tropical, a few
of them reaching as far north as the Mediterranean and as far south as
the Cape of Good Hope. We may say approximately that they are
confined within the fortieth* parallels of north and south latitude, but
occur more plentifully as we reach the equator from either hand. If we
may judge by the marine fauna of the Kimmeridge clay, it is such as would
favour the view of the climate being suitable to the occurrence of Caulerpa.

Though I do not venture to claim for this determination the cer-
tainty which would be yielded by tissues preserved in a state fit for
microscopic examination, yet I would point out that we are dealing
here with no mere flattened impression, but with a cast exhibiting in
the round all the details of external conformation in great perfection,
that these characters agree closely with living representatives, that the
deposit is a marine one, and the conditions of climate fitting. Taking
all this evidence pointing in one direction, I would therefore venture to
urge the recognition of my contention that we have here evidence of
the existence of an Alga in the secondary rocks of far greater weight than
has hitherto been brought forward. The fact that it, like the other fossil
Algae, certainly known to us as such (except the diatoms and Litho^
thamniori), is a siphoneous Alga may have a certain significance. Owing
to the plentiful lack of material, I do not choose to draw any inference
from such facts, though others may have a taste for the pastime.

GEORGE MURRAY.

EXPLANATION OF PLATES.
PLATE IV.

Fig. i. Caulerpa Carruthersii, G. Murr. Nat. size. a. Section near apex.
Fig. 2. a. b. c. Sections seen in cross fracture. Nat. size.
Fig. 3. Ideal figure of C. Carruthersii.

PLATE V.

Fig. i. Specimen of C. Carruthersii in Museum of Practical Geology, Jermyn
Street. Nat. size.

Fig. 2. a. Caulerpa cactoides Ag. var. gracilis. Nat. size. b. Cross section of
C. cactoides ( x 15 ). c. Longit. section ( x 15), after figures by the author in Linn. Soc.
Trans. ot. y vol. iii., pt. 4, plate LII.

* They come slightly to the north of 40 in the Mediterranean.

III.

ON THE STRUCTURE OF DICTYOSPHsERIA

DECNE.

THE genus Dictyosphceria, not to be mistaken for the widely different
DictyospJuzrinm Naegeli,* was founded by Decaisne,f who took Valonia
favulosa, Ag, for the type of it. The commonly accepted view of its
systematic position has remained very much the same as that indicated
by Decaisne, viz., near Valonia and Anadyomene, and I do not now pro-
pose to assign to it any other affinities, but rather to show how strongly
this original opinion is confirmed by an investigation of its minute
structure and the disclosure of an arrangement of its parts hitherto
undescribed.

To the typical species of the genus, Harvey + added another, D. sericea,
an exceedingly well-marked species, as will presently be shown, though
Harvey himself subsequently seems to have doubted it, from an imperfect
understanding of its structure. MM. Millardet and Montagne II added
another species, D. Enteromorpha, I have not seen a specimen of this
species, and it is very difficult to say from the illustration whether it is
to be regarded as a true Dictyospkceria or not. Finally, Zanardini 11
added a fourth species, which I have no hesitation in excluding from
the genus on the ground that it is plainly an imperfect Valonia.**

* Einzellige A I gen, p. 72.

t Class, des Algues, Ann. Set. Nat., 2 sdr., vol. xvii., 1842, p. 328.

Trans. R. Irish Acad., xxii., p. 565, and Fl. Tasin., ii., p. 339, tab. 196 A.

Phycologia Australica, vol. v., Synopsis, p. lix.

|| In Maillard, Notes sur File de la Reunion Algues, p. 4, plate xxv.

H Iconogr. Phycol. Adriat., p. 73, tab. xviii.

** Dr. Schmitz of Greifswald writes in confirmation of this view : ' D. valonioides
Zan. ist (wie ich durch eigene Untersuchung von Material aus Neapel feststellen
konnte) nichts anderes als ein unregelmassiges Regenerationsprodukt eines alten
halbzerstorten Exemplares von Valonia macrophysa?

16

The Structure of Dictyosphceria. \ 7

The best description known to me of D. favulosa is that furnished
by Harvey * :

'Fronds at first globose, like tubers, heaped together, hollow and empty or
filled with sea water, attached to the rock and to each other by a few short, rooting
processes; at length irregularly torn, and then forming expanded cartilaginous or
skinlike, coarsely reticulated membranes. The membrane is wholly composed of
a single layer of large globose, or by mutual compression hexagonal cells, which
closely adhere by their sides, leaving the convex ends of the cells free, and these
form the surface of the membrane, which when dry resembles a piece of fish skin,
or a miniature honeycomb. When the cells have been separated, each is seen to
be marked at the line of junction by a double row of circular discs. In full-grown
cells the primordial utricle is easily separable from the outer cell-wall, and contains
a green granular endochrome, from which by cell division four new cells are
formed, and thus the frond extends by repeated quadrisection of its component

cells I have seen hairlike processes issue from it internally, analogous

perhaps to the fibrous processes of the membrane of Caulerpa.' 1

Prof. J. G. Agardh f discusses the structure of Dictyosph&ria at
considerable length, with the general result that he does not carry us
much farther than Harvey. I should have said that KiitzingJ figures
D. favulosa, and attributes to the rows of circular discs (noted above)
the character of Keimzellen. There was, no doubt, considerable tempta-
tion to take this view, having regard to the affinity of Dictyospharia
with Valonia. Agardh deals at some length with these discs, pronouncing
against the Kiitzingian view, and, while leaving the matter doubtful, sug-
gesting, ' Forsan credere liceret eadem potissimum analoga esse organis,
quae in membrana aliarum Siphonearum (Espera, Penicillus] conspiciantur,
in quibus deposita calcarea magis normaliter coacervantur.' He also dis-
cusses the minute processes projecting into the interior of the cells, while
also leaving their nature in doubt. With regard to Harvey's idea, he says :
' Vix cum fibris introrsum prominulis Caulerparum eosdem comparare
auderem.'

D. favulosa occurs in all tropical seas. I have examined material from
Ceylon (Harvey and Ferguson), Mauritius (Pike), Red Sea (Hohenacker),
Philippines (Challenger Exp.), Barbadoes (Sir R. Rawson and Miss Watts),
St. Thomas (Challenger Exp.), Guadeloupe (Maze), and Grenada, collected
by myself. The specimens of MM. Maze and Schramm from Guadeloupe
called D. valonioides Zan. are D. favulosa. The material principally used
is a series of young specimens collected by Ferguson in Ceylon, and my

* Nereis Bar. Am., iii., p. 50, tab. xliv. B.

t Till. Alg. Sys/., viii. (Siphoneae), p. 113, tab. 2, figs. 1-3.

\ Tab. P/iyc., Bd. vii., p. 10, tab. xxv.

C

1 8 The Structure of Dictyosph&ria.

own specimens preserved in spirit from Grenada. The Ceylon material
corresponds with the descriptions of Harvey and Agardh in respect of the
hollow interior, while my material shows a solid mass of cells in the young
state, the interior cells being less firmly united than the peripheral cells,
but ultimately becoming hollow in the older specimens. The first observ-
ation was directed to discover the nature of the mysterious discs. These
proved to be tenacula emitted from and attached to the cells, and binding
the mass together. An inspection of figs. 2b, c, 3, will show that these tenacula
differ in no essential respect from those already described by Mr. Boodle
and myself in Stmvea* Spongocladia,-\ Microdictyon,% Boodlea,\ &c.
Where the attachment of the cells is intimate, as in the peripheral cells,
the tenacula are short, and in the internal cells, which are more loosely
compacted, the tenacula are produced at the ends of filaments (fig. 3)
in the familiar manner. The figures exhibit the mode of attachment so
well, that a detailed description is unnecessary. After reaching a certain
stage of development the tuberous bodies of D. favnlosa burst, and the
lobes appear to go on growing irregularly, giving the mature plant a very
different appearance from what it presented in its earlier stages.

With regard to the cell division into four, some of the Ferguson
specimens (fig. i c, upper part) show what I take to be an appearance of
the kind described by Harvey and Agardh, but I am not altogether con-
vinced that the explanation of the matter is so simple. On this point I
hesitate to put forward an alternative view for the present. The processes
projecting into the interior of the cells (fig. 4) are accurately described by
Agardh, and, like him, I fail entirely to apprehend their significance.

We have here, then, in Dictyosphcsria favnlosa one of the simplest forms
of valonioid organism, and one, moreover, of some interest from the wider
point of view of thallus-formation. Struvea, Boodlea, and Microdictyon
present to us examples of a branching reticulate thallus of this type, the
component members of which are held together by tenacula ; other
Siphonocladacece, like Spongocladia and Udotea, show us cases of strands
of filaments or fronds similarly connected. Valonia itself, the simplest of
these forms, branches, though unpossessed of tenacula for the junction of
its members. In Dictyosphcsria favnlosa we have simply an aggregate of
similar cells not forming a definite frond, but cohering in an unbranched
mass, this colony of units being held together solely by tenacula. It is
certainly an extraordinary form of tissue -formation, and represents, it
appears to me, the most reduced form of the siphoneous thallus, using the
term in the wide Agardhian sense.

* Annals of Botany, vol. ii., No. vii. t Ibid., vol. ii., No. vi.

% Journ. Linn. Soc. Bot., vol. xxv., p. 243. Ibid.

The Structure of Dictyosphceria. 19

Dr. Schmitz, of Greifswald, having heard of my examination of this
species, has been good enough to write me the following letter, giving a
brief account of his observations on D. favulosa. It will be seen that they
confirm the above description in certain particulars :

' Die jungeren Pflanzen von D. favulosa sind massive Zellkorper von unregel-
massig kugeliger Gestalt, oberseits mehr oder weniger abgeflacht, unterseits etwas
verdickt und mit schmaler Insertionsflache angewachsen. Diese massiven
Zellkorper sind grosszellig; die grossen Zellen aber sind angeordnet in unregel-
massige verzweigte Zellreihen, die von der Insertionsstelle aus aufwarts facherformig
auseinanderlaufen. Der ganze Zellkorper aber stellt ein congenital verwachsenes
Verzweigungssystem einer grosszelligen Cladophora (z.B. Cl. prolifera} oder einer
kleinzelligen Valonia (z.B. V. Cladophora Kiitz.) dar, ein Verzweigungssystem,
dessen Gliederzellen vielfach secundar querverkettet sind durch ganz kleine Hafter-
Zellchen (wie solche ja bei Siphonocladaceen vielfach vorkommen). Nachtraglich
erfolgt dann in dem starker verbreiterten Pflanzenkorper unterhalb der obersten
Zellenlage der Oberseite die Bildung eines mehr oder minder ausgedehnten
horizontalen Spalte, wodurch eine Deckschicht von clem Thallus -korper sich
abhebt. Dann reisst die abgehobene Deckschicht in der Mitte unregelmassig auf
und reisst von hier aus in wechselnder Weise ein, wahrend sie gleichzeitig unter
flachen-Wachstum sich ausdehnt und in verschiedenartigster Weise sich ausbreitet.
Eine " Viertheilung " der Zellen erfolgt bei diesem flachen-Wachstum der
aufgerissenen Deckschicht aber nicht. Die Lappen der aufgerissenen Deckschicht
konnen dann in mannigfaltigster Weise weitersprossen, haufig auch proliferirenden
Sprossungen den Ursprung geben. Fortpflanzungsorgane habe ich nicht
beobachtet.'

In Dictyospharia sericea, of which I have examined the material
collected by Harvey and Clifton in Australia, we have a more complicated
structure. In surface view the thallus presents (fig. 5) a remarkable
appearance, not indicated by Harvey either in his figure or description.
As figured by Harvey, the thallus is represented simply by a frond
consisting of hexagonal cells fitting into each other without intercellular
spaces. A minute examination of this view of the thallus discloses the
fact, however, that the borders of these cells are everywhere lined with
rows of smaller cells sometimes single, sometimes double rows running
along the tops of the hexagonal walls, and these are discovered to be
attached by tenacula (fig. 6) alternately on either side. The cross-
section (fig. 7) shows the thallus to be several cells thick, with a stratum
of large cells in the middle, and two or three above and below it. The
bordering cells with tenacula are right over the surfaces of junction, as
would be expected, and are here seen to be several in depth above and
below, penetrating wedge-fashion between the large cells of the middle
stratum. I have said that the thallus is several cells thick, but the view is

2Q 77/6' Structure of Dictyosphceria.

probably equally tenable that the cells immediately overlying and under-
lying the cells of the middle layer are merely spaces enclosed by mem-
branes separated from the wall of the middle layer by the growth of the
bordering cells, and now suspended, as it were, from one border to the
other. In favour of this latter view, I should state that I have never seen
any cell contents in these overlying and underlying spaces, while those are
plentiful, but disorganized, in the cells of the middle layer. Where the
section has run along a row of bordering cells, the appearance seen at a
(fig. 7) is presented. I account for the appearance at b by supposing
that it is an intrusion from the bordering cells, or more probably a cut-off
angle from an adjoining cell of the middle layer. Owing to the condition
of the material, which required very careful treatment to enable me to
obtain this much information, I have been unable to carry the
investigation of this species farther. It suffices to show us a form
differing very strikingly from the simple D. favulosa, and probably
confirming the view taken of the affinity of the genus with Anadyomene
a view much shaken by the examination of D. favulosa. I have to
thank Miss Mitchell for sketches of tenacula in the accompanying plate,
and for the interest she has taken in Dictyospkaeria.

GEORGE MURRAY.

EXPLANATION OF PLATE VI.

Fig. i. D. favulosa in various stages of development; a, c, d, e ( x 2) ; b ( x 8) ;

f, nat. size ; g, in section ( x 3) ; a and b represent a plant

consisting of a single cell ; in b, tenacula have already been

developed.
Fig. 2. ,, Cells joined by tenacula, diagrammatic view.

(a) alternate tenacula joining two cell-walls ; (b) tenacula

seen from below; (c) seen from above ( x 150).

Fig. 3. Tenacula from internal cells side view ( x 150).

Fig. 4. Internal projections from cell-wall ( x 150).

Fig. 5. D. sericea, surface view ( x 66).
Fig. 6. Bordering cells with tenacula ( x 375).

Fig. 7. ,, Section of thallus ( x 66).

Fig. 8. Nat. size.

IV.

ON MALFORMATIONS OF ASCOPHYLLUM
AND DESMARESTIA.

IN a short account of the galls in Rhodymenia palniata Grev., published
in the Journal of Botany for March, 1891, I referred to malformations
found in Ascophyllum nodosum which were shown me by Mr. George
Murray, having been collected by him at Stonehaven. During a visit
to Cumbrae last summer I found a number of Ascophyllum plants bearing
similar growths, and in one instance almost every branch of the thallus
was more or less affected. I collected specimens both from this plant
and from others near it, and preserved them for investigation in London
and for comparison with those I had already seen.

The malformations are almost invariably confined to the part of the
thallus immediately above or below the air-vesicles, and in many cases
the thallus is broken away across the middle of a vesicle, the remaining
half being transformed into a solid mass of diseased tissue. On the
younger vesicles the swellings are considerably smaller than on those
nearer the base of the thallus, probably from having been more recently
attacked. The thallus in the affected parts is much swollen and covered
with small, round swellings or lumps, forming a corrugated surface
(plate VII., figs. I and 2) ; and a transverse section through this part
shows that each lump is a more or less hollow space containing numerous
nematode worms (plate VII., fig. 3). In the younger stages of the disease
the central part of the tissue is not affected, and the cavities are not
in communication with one another ; but in the older stages the whole
of the tissue of the affected part becomes diseased, and the partitions
between the cavities are broken down, both cavity and passage being
bordered by a thick wall of disorganized tissue closely pressed together.
The cavities are flask-shaped and have a small narrow opening at the
top, by which the nematodes appear to make their way in or out. In the
youngest stages of malformations that I have found, a slight rupture of

21

22 Malformations of Ascophyllum and Desmarestia.

the epidermal layer and a forcing asunder of the tissue beneath is observed
(plate VII., fig. 4), caused probably by the entrance of the nematode from
without ; and the space thus made is filled with a granular substance,
in which I believe I have found parts of the animal cut through. It is
difficult, however, to find the nematode in the young galls, as the cross
sections of the cells of the thallus bear much superficial resemblance in
outline to similar sections of the animals. The effect of this forcible
entrance of the nematode appears to be that the thallus swells up all
round the spot inhabited by the animal, probably caused, like gall
structures in general, by the stimulation of some animal secretion. I
have in no case found any communication between the galls along the
interior of the thallus, and I am therefore forced to believe that the
nematode attacks each affected spot separately and from outside.

The material of Ascophyllum no do sum collected at Stonehaven shows
the same hollows filled with nematodes, and, as these growths appear
to be rather common, I imagine that Ascophyllum is a favourite resort
of this species of worm. I have in no instance found these malformations
inhabited by any other sort of animal, and only by nematodes in a fully
grown condition.

As regards the nematode itself, I have sent specimens of it to
Dr. de Man, of Middelburg, who has very kindly examined them, and
pronounces the species to be new to science. We have therefore called
it Tylenchus fudcolus, n. sp,, and a detailed description of it with plates
will shortly be published by Dr. de Man, to whom I here offer my best
thanks for the interest he has taken in this investigation.

I also wish to express my indebtedness to Professor King for the
trouble he has taken in sending me fresh material from time to time,
and for the interest he has shown in the work.

Through the kindness of Mr. Batters, I have been enabled to examine
malformations on Desmarestia aculeata, Lam., found by him on the island
of Cumbrae in September of last year. The malformations in this Alga are
small round bodies, sometimes growing singly on a young part of the
thallus, sometimes grouped together on the thick stem, forming a slightly
raised ridge about a quarter of an inch long. In other places a lump is
formed in the angle of the branches, in shape somewhat like the young
malformations on Ascophyllum Jtodosuw, described above ; and it is in
these lumps that there is the greatest likelihood of finding the animal
which causes this growth, viz., a copepod.

A vertical section through one of these swellings, transverse to the
thallus, shows that the hypertrophied tissue is composed of very closely
interwoven anastomosing hyphae, which somewhat resemble the ordinary

Malformations of Ascophy Hum and Desmarestia. 23

pseudo - cortex of Desmarestia, as seen in longitudinal section (fig. 5).
The walls are generally thinner, however, and the cells much smaller than
the normal tissue. The cells are very full of contents, and the dark yellow
mucilage, as well as the granular substance, which I found in Rhodymenia
palmata, are here in abundance. The Rhodymenia galls, it may be re-
membered, were also caused by a copepod, Harpacticus chelifer. In some
cases the tissue is torn, and an opening communicates with the outside,
looking as if some animal had forced its way through the cells into the
open. The space is always half rilled with the fine granular matter
mentioned above, and the cells bordering it are very small and thin-walled,
showing that some reparative process has been going on around the torn

^

cells. In cutting open some of the larger swellings, a quantity of oblong
brown cells pour out, sometimes separate, sometimes adhering together ;
and, in nearly every case where these brown cells are, there may be found
a specimen of the copepod which causes the outgrowths. I cannot in
any way account for the fact that, notwithstanding the large number of
outgrowths examined, I have found so few copepoda indeed, in the
whole investigation, I have only succeeded in teasing out one fully
developed specimen, the others being all rolled up in a more or less
immature condition.

In the case of Rhodymenia palmata, I took the view that the copepod
went through its various stages of development in the thallus of the Alga,
and on reaching maturity it forced its way out into the open. This view
was strongly opposed by an authority on Crustacea, and, until the eggs
of the animal are found in the youngest outgrowths, it is impossible
that this idea can be more than a theory ; but, if it is incorrect, I cannot
understand how the animal comes to be found in the outgrowths before
it is fully developed and while still in the nauplius state.

In Rhodymenia palmata, the small galls in which the tissue was quite
undisturbed were probably caused by irritation, which had spread over
the adjoining area round the invaded spot of the Alga, and in the case of
Desmarestia I account for the presence of the small outgrowths of un-
disturbed tissue near the nidus of the copepod in the same way. The
isolated swellings, however, on the young branches of the thallus must be
formed singly by a puncture or other invasion, as they are too far away
in many cases from the scene of disturbance to be accounted for on the
score of irritation. Their structure is the same as that of the larger
swellings (plate VII., fig. 6), and many of the cells contain darkly coloured
matter.

I have sent specimens of the copepoda, teased out from the mature
outgrowths, to Dr. G. S. Brady, who was so kind as to interest himself in

24 Malformations of Ascophyllum and Dcsmarestia.

my former investigation ; but, unfortunately, he has not been able to
decide the name of the species, through want of a sufficient number of
complete specimens, most of those I sent to him being immature. I hope,
therefore, when I shall have the opportunity of collecting fresh material,
to send more specimens to Dr. Brady, that the question as to the species
of copepod may be settled.

Finally, I must express my gratitude to him for the very kind help and
encouragement he always gives me.

ETHEL S. BARTON.

EXPLANATION OF PLATE VII.

Ascophyllum nodosum.

1. Malformation. Natural size.

2. ,, Trans, sect. ( x 4).

3. Mature ( x 66).

4. Young ( x 66).

Desmarestia aculeata Lam.

5. Vertical section of malformation (xi43).

6. small outgrowth ( x 37).

V,

ON CONCHOCELIS, A NEW GENUS OF
PERFORATING ALGsE.

IN their interesting account of the plants living in the calcareous shells
of molluscs, MM. Bornet and Flahault* mention eight genera of Algae
which may be met with in this strange situation. All of these genera
are members either of the Cyanopkycecs or Chloropkycecs, and are either
bluish or purplish green or grass-green in colour. It was therefore with
both surprise and pleasure that I detected among some perforating
Algae, extracted from shells which I had gathered near Cumbrae, in the
Clyde sea area, some filaments of a beautiful carmine-coloured Alga
belonging apparently to the Porphyracea. As Dr. Bornet had devoted
so much time and attention to the subject of the perforating Algae, I
sent the specimen to him, requesting him to let me know if, during
his researches, he had met with anything similar. He at once replied
that he had not previously seen anything resembling my plant, which
he believed to be a new Alga related to ErytJirotrickia. I propose in
the present paper to give a short account of this interesting addition
to our marine flora.

The shells from which this Alga was first obtained were dredged
from 6-8 fathoms water off the Tan buoy, between the islands of Great
and Little Cumbrae. The best specimens were got in the empty shells
of Mya tnmcata and Solen vagina (the common razor-shell). I have,
however, subsequently found the same plant, though in poor condition,
in many other shells left on the sandy beach by the receding tide.

To the naked eye the presence of the Alga is betrayed by a pink
stain, which cannot be removed by rubbing, on the inner surface of the
shells. If a flake, sufficiently thin to be semi-transparent, be broken from

* ' Sur quelques plants vivant dans le test calcaire des mollusques, par MM. Ed.
Bornet et Ch. Flahault.' Bulletin de la Societe botanique de France, vol. xxxvi.

25 D

26 Conckocelis.

a shell which contains Conchocelis, it is an easy matter, even with a pocket
lens, to see the delicate pink radiating filaments of the Alga. It is not
however, till the plant has been freed from the calcareous substance in
which it grows that its characters can be made out in a satisfactory way.
I have found that the best decalcifying agent is the ' liquide de Perenyi '
the formula for which is given by MM. Bornet and Flahault, which, though
it acts slowly, is much more satisfactory in its results than any other
solution I have tried. At first the Alga appears in the form of roundish
spots, which finally become confluent, their pink colour sharply dis-
tinguishing them from the spots formed by any of the other perforating
Algae which always, or at least very frequently, accompany them.

On removing the Alga from the shell by means of the ' liquide de
Perenyi/ one at once perceives that these spots are formed of a network
of interlaced branching filaments, the central portion of the patch being
so closely felted together that it is impossible to make out the ramifi-
cations of the filaments ; but at the edges the filaments are not so closely
interlaced, and it is easy to trace any particular thread for a considerable
portion of its length.

The thallus is formed of articulated branched filaments, which in the
young plants (figs, i and 2) radiate more or less from a central point.
Later, by the confluence of several groups, and the consequent interlacing
of their filaments, a more or less compact, continuous, horizontal network
(fig. 3), which finally covers a considerable portion of the shell, is formed
in the superficial layer of the shell. These filaments are of very various
widths ; the most slender that I have measured was only 1*5 /x in diameter,
while the most robust was 7-5 p, the usual width being about 4^. The
cells of which the horizontal filaments are composed are of very various
shapes straight or curved, clavate or triangular, or so irregular as to
defy description. Below the horizontal layer the filaments swell out here
and there into irregularly shaped, septate, simple, or slightly branched
inflations ( fig. 5 ), which are, of course, much more robust than the
ordinary filaments ; they are often 20-30 /u in diameter, though they
seldom reach more than from 70-1 io/* in length. The branching of the
ordinary filaments is very irregular ; sometimes for a considerable space
a filament is quite simple, and then suddenly, by the outgrowth of
lateral shoots, becomes densely branched. The lateral branches are
either opposite or alternate, and are either simple or more frequently
rebranched, their lateral shoots sometimes anastomosing. The cells of
these filaments and those of the inflations are always more or less con-
stricted at the joints.

The inflations (Fig. 5), which usually consist of from two to ten

Conchocehs. 2 7

cells, varying in length from 15-30^, are either quite simple or more or
less branched. In the centre of each cell of the inflations there is a
star-shaped chromatophore, the rays of which often reach to the edges
of the cell. I have been unable to detect with any degree of certainty
the manner in which the colouring matter is disposed in the ordinary
filaments, but it appears to be applied in a continuous layer to the
walls of the tubes, which are always, so far as I have observed, of a
uniform, or nearly uniform, carmine colour. The inflations often become
detached from the horizontal filaments, and are then capable of an
independent existence. I have frequently found these inflations of all
sizes and shapes, growing in company with Gomontia and other per-
forating Algae, but without any trace of the horizontal filaments of the
Conchocehs from which they have originated.

The plant appears to be reproduced by means of spores formed in
the cells of the inflations, one spore in each cell. I have seen globular
bodies which appear to be spores escape from the cells, and not unfre-
quently one sees them lying free among the filaments of other per-
forating Algse.

The shape and colour of the chromatophores, and the manner in which
the spores are formed, point to a strong relationship between Erythro-
trichia and the present genus ; but the branched frond and curious
inflations sharply separate the genera. In some respects, e.g., the infla-
tions and the branching, Conchocelis resembles Ostreobium, from which it
is at once distinguishable by its pink colour and articulated filaments.

Conchocelis nov. gen. Thallus minutus e fills ramosis articulatis hie illic
in utriculos septatos, forma irregulari dilatantibus, compositus. Propagatio
fit per sporas in cellulis utriculorum evolutas. Unica spora in singulis
cellulis. Genus ad Porphyraceas referendum.

Conchocelis rosea, n. sp., thallo immerso, roseo, maculas orbiculares
demum confluentes et ambitu indefinitas efficiente ; fills primariis in stra-
tum pannosum implicatis, ex cellulis cylindraceis, tortuosis aut forma
irregulari, 7 usque ad 75^ et ultra longis 1-5-7, saepius 4-5^ crassis
constantibus. Utriculis simplicibus aut ramosis usque ad iio/t longis ad
3O/z latis. Sporis globosis 13-15^ crassis.

Hab. in conchis vetustis saepe in consortio Gomontice ad oras Scotiae

prope Millport in insula Cumbrse.

EDW. A. BATTERS.

28 Conchocelis.

EXPLANATION OF PLATE VIII.

Prepared from specimens decalcified by Percnyt's solution.

Figs, i and 2. Young plants taken from a part of the shell very little attacked
by the Alga ( x 250).

Fig. 3. Portion of the horizontal network at a more advanced period ( x 250).

Fig. 4. Portion of the vertical filaments ( x 250).

Fig. 5. Various stages in the development of the inflations, showing the
chromatophores.

Fig. 6. Portion of an inflation with spores.

?hvc."Mem.Part .1

.V

Berjeau 8c"Higliley Jel.etTibh

CAULERPA CARRUTHERSII G.M:
.CACTOIDES Ag.v&r GRACILIS (2a. r

C

Hani

.

Plate VI

fl

IS

.

^~Y? > I ~>

Iff

Serj ean. & Hgnl cy del et litix - Hankart imp .

DICTYOSPHAERIA FAVULOSA .Z?ec (,1-4.) D. SERTCEA Ea,rv. (5-8)

Phyc Mem. Par I . I

ASCOPHYLLUM (1-4) DZSlMARr^S" . "6)

Pkya.Mem.Parb.I .

E.A.L.B.iel
COTTCHOCELTS ROSE A

ah

*

VI.

NOTES ON THE MORPHOLOGY OF THE

FUCACEsE.

THE object of these notes is to supply deficiencies in the account of
the Fncacete given by Oltmanns,* by a study of those types which he
was unable to examine or of which he had imperfect material. In his
grouping of the genera marks of uncertainty are attached to Coccophora,
Xip ' hop/io ra, Sarcopliycus, Myriodesma, Ecklonia, Scaberia, Hormosira,
and Splachnidinm. The first three of these genera are dealt with
below, together with Seirococcus and Notheia, which I thought worth
more minute examination. Myriodesma has been examined, with the
result that the position assigned to it by Oltmanns is confirmed, and
his query may be removed. Of Ecklonia and Scaberia satisfactory
material did not arrive in time for inclusion in this research, and an
account of them will appear later. Hornwsira had already been dealt
with by Mr. Mollet f (overlooked by Oltmanns) and Splachnidinm
was accounted for in the first part of these Memoirs. With regard to
this last genus, now excluded from the Fucacece, it is gratifying to
receive the following, confirmation in an essential point of the interpre-
tation of the structure, by Mr. R. M. Laing, whose previous work at
Splachnidium has been referred to. He writes to the authors from
Christchurch, New Zealand, on the I5th February of this year: 'I am
glad to say that I am able to bear out the conclusions you arrived at
as to the nature of the supposed oogonia, as several years ago I saw
zoospores in active motion escape from some that I had been examining,

* Bibliotheca Botanica, Heft 14, 1889.

t On the Structure of Hornwsira LabiHardicri (Trans. New Zealand Institute^ vol.
xiii., 1880).

29 E

30 Notes on the Morphology of the Fiicacece.

confirming your conclusions as to their real nature. I have written a
note on the subject for the Transactions of the New Zealand Institute' etc.

With regard to the other genera named, excepting the two reserved
for the present, the general result is that, with regard to Seirococcus
and CoccopJiora, Oltmanns' expectations are confirmed as to their oogonia
and the number of their oospores. XiphopJiora, on the other hand, yields
four oospores in each oogonium instead of one, as conjectured by
Oltmanns. It had better, therefore, be removed from his section
Loriformes and be placed in the Fucece with AscopJiyllitm, though its
receptacle bears a remarkable likeness to HimantJtalia, the type of
Loriformes. Notheia was not included in Oltmanns' review of genera.
The result of this examination would point to its true position in the
section Fucece as a very degenerate member. Hormosira is also in
the same section, judging by Mr. Mollet's observations (confirmed by
Miss Smith).

Sarcophycus, which has four oospores in each oogonium, is so plainly
the ally of Dnrvill<za that it may remain with it and Ecklonia pending
the further investigation of these genera. It may be noted that,
though the four oospores suggest affinity with the Fucea in this
respect, yet the different form of their segmentation described by Miss
Whitting, and the frequent situation of the oogonia on branching hairs
afford characters that may be looked for in its companion genera.

Oltmanns appears to me to attach undue weight, in estimating those
characters on which he rests his sections, to that afforded by the growth
of the plants, viz., the occurrence of three or four-sided apical cells.
My reason for depreciating the value of such a character is its proved
inconstancy, not only among Algae, but even in the much less plastic
forms of the Vascular Cryptogams.

Hearty thanks for material of Seirococcus, XipJiopIiora, Hormosira,
NotJieia, and SarcopJiycus are due to Mr. Bracebridge Wilson, of Geelong,
Australia. (ED.)

COCCOPHORA LANGSDORFII GREV.

DAWSON TURNER gives us the first account of this beautiful and
uncommon sea-weed in his Fitci, vol. Hi., p. 76, tab. 165.

It was brought to him from Japan, as a dried specimen, by Dr.
Langsdorff, the ' natural Philosopher ' attached to a Russian expedition
round the world.

Notes on tlic Morphology of the Fncacetf. 31

Dr. Langsdorff gave him at the same time a drawing of the plant in
its natural condition by Prof. Mertens, which Turner incorporated
along with the description of the new plant, and named it Fitcus
Langsdorfii in honour of its finder.

Tilesius, a companion of Dr. Langsdorff on this expedition, also
gathered and brought home a specimen which he gave to C. Agardh.
A preliminary description of it under the name of Fucns Tilesii, was
published by Agardh in 1812. (Alg. Dec. I.)

In his Species Algannn, p. 78 (1823), he described the plant more
fully and transferred it to the genus Cystoscira.

Finally, in 1830, Greville founded for it a new genus, Coccophora
(Synopsis, p. 34), retaining Turner's specific name Langsdorfii.

The specimen in the British Museum is from the Shuttleworth
Herbarium, dated 1830. The plant bears out the description given by
Turner, being of a dull black colour in the dry condition.

Coccophora Langsdorfii is one of the most highly organized of the
Fuel, recalling Sargassum with its leaves and berries. It has a central
axis or stem from which lateral members are given off, according to
Turner : ' alternate and separated by intervals of an inch, or standing
close together, and not uncommonly two or three rising together from the
same point, about half a foot long, and each swollen at its base into a
small bulb.' This latter statement I have been unable to verify.

These branches from the main stem (plate IX., fig. i) are spirally
clothed with short, sessile, entire, tapering leaves, which, towards the
apex, are replaced by a raceme of berry-like hollow receptacles, each
borne on a short stalk. This transformation of leaves into special organs
for fruit-bearing divides CoccopJwra from the Cystoseirce, and places it in
another group between Cystophora and Scaberia.

The vegetative tissue is very simple. There is a central cylinder of
long empty cells, with very small lumen, owing to the great thickness of
the cell-wall. Towards the outside the cells are shorter and broader with
thinner walls, thus forming a sort of cortex, and the epidermis is of
brick-shaped cells. This layer and the outer cortical cells are filled
with dark brown contents (plate IX., figs. 3 and 4).

The tissue of the leaves is continuous with that of the main axis ;
there is a middle strand of long thick-walled cells and wings of large
rectangular or polygonal cells, clothed with an epidermis similar to that
of the stem.

The hollow receptacles have an uniform tissue of comparatively thick-
walled cortical cells, which are richer in contents than those of the
vegetative tissue. The conceptacles are numerous and crowded, and arc

32 Notes on the Morphology of the Fncacctf.

separated from each other by a narrow strand of tissue (plate TX.,

fig. 2).

The plants examined by me are dioecious. The male conceptacles do
not differ from those of other Fuel ; the female conceptacles are full of
oogonia, with multicellular hairs interspersed among them.

The oogonium rises on a broad base from the wall of the conceptacle,
one division cutting off the large solitary oosphere from the pedicel which
is buried in the wall of the conceptacle, and probably therefore resembles
that of Pelvetia, where the pedicel cell is short, and does not elongate

to form a stalk.

A. LORRAIN SMITH.

EXPLANATION OF PLATE IX.

Fig. i. Coccophora Laiigsdnrfii. Natural size.

Fig. 2. Receptacle enlarged ( x 5).

Fig. 3. Longitudinal section of branch and base of leaf ( x 150).

Fig. 4. Transverse section of stem ( x 150).

Fig. 5. Male conceptacle ( x 150)

Fig. 6. Antheridia ( x 450).

Fig. 7. Female conceptacle ( x 150).

SEIROCOCCUS AXILLARIS GREV.

THIS Alga was found at Port Dalrymple by Robert Brown during
his expedition to New Holland at the beginning of the century. He
named it Fucus axillaris, and under that name placed it in the Herb.
Brit. Mus. In his MS. it is thus described : ' Fucus fronde plana ra-
mosissima ramis ramulisque alternis, integris, coriaceus aveneis siliquis
marginalibus confertis pedicellatis torulosis. Inter rejectimenta maris,
Port Dalrymple.'

He communicated specimens to Dawson Turner, who, in his
Fuel, vol. iii., p. 28, tab. 146, figured and described it under Brown's
name. Turner referred Mertens' Fiicus scorteus (Mertens MS. fide D. T.)
to this species as Var. /3; but, from an examination of the series in the

"A- on the Morphology of the Fucaccc?. 33

British Museum, I am satisfied that it does not deserve to be separated
from the species even as a variety.

Agardh, in his Systeuia, p. 291, places this Alga among the Cystoseirae,
and in his Species Algarum he repeats this determination, and names it
Cystoscira axillaris.

The genus Seirococcus was created by Greville for this plant, of which
it is the only known species. (Synopsis, p. 34.)

Montagne, in Dumont D'Urville's Voyage au Pole Slid (Botauiquc,
vol. i., p. 86, 1845), discussing the systematic position of Seirococcus
refuses to accept it as a separate genus, and insists on its resemblance to
ScytotJialia, under which he places it. The receptacular branch he allowed
was different, but did not think it of importance.

Klitzing records it as ScytotJialia axillaris (Species Algarmn, p. 59 2 )-
and Oltmanns* seems doubtfully to follow him.

I have adhered to Greville's classification for the following reasons.
There is an undoubted resemblance between the two plants in their
structure and mode of growth which makes them very closely allied, but
the receptacular branches differ from each other in a marked degree.
Harvey, who agrees with the Grevillean classification, adds another
slight distinction (PJiycologia Australica, plate IV.), 'ScytotJialia when
steeped in fresh water throws out, like Fucus, an immense quantity of
slimy gelatine ; but this is not the case with Seirococcus.' On examination
I find the cells of ScytotJialia have very much swollen mucilaginous walls.
The conceptacles of Seirococcus are borne on special branches, not on
metamorphosed organs like those of CoccopJiora and some other Fttci. It
is this rather striking peculiarity which separates it from the Cystoseirae,
the affinity with which Turner had already questioned, and places it near
the Sargassece. Oltmanns, in his arrangement of the Fucacece, places it
under that group, and the following details of its structure confirm his
judgment.

The plant has a flat stem from which side branches are given off
irregularly (plate x., fig. i.) These branches bear the flat sinuous fronds
which rise alternately at an acute angle. On the opposite edge from the
leaf insertion there is a narrow wing, which in turn broadens out to form a
frond. The fronds are simple and entire, flat and rather sinuous ; there is
no midrib, and they vary in width, tapering to a blunt tip, simple or bifid.

The plant has much the same structure throughout, in stem and frond.
There is a central portion in the stem of long cells with very much
thickened walls ; a cortex of larger looser cells, and an epidermis several

* Loc. cit., p. 70.

34 Notes on the Morphology of the Fucaccff.

layers deep of small brick-shaped cells very rich in contents. In the
frond the same structure is repeated (plate x., figs. 3 and 4), but the central
thick strand has been flattened out and stretches across the frond, giving
it a leathery consistence. The cortex and epidermis are not so strongly
developed.

The fruit-branches or receptacles are erect and torulose (plate X.,
figs, i and 2) ; they occur along the axils formed by the frond and the
wing of the stem. They vary in size and are crowded together, some
rising directly from the leaf margin ; others are borne on a flat or
round pedicel, and branch irregularly. Each swelling of the receptacular
branch encloses a diclinous conceptacle. The male conceptacles as a
rule are at the base of the fruit-branch, the female nearer the top,
and both occur on the same branch.

The antheridia in the male conceptacles are much branched, and are
accompanied by the usual branching paraphyses (plate X., figs. 5 and 6.)
The female conceptacles are crowded with oogonia. There is only one
oosphere cut off from the base of the oogonium (plate X., fig. 8),
and the pedicel cell, if present, is obscure. The paraphyses are large
with irregular cells rich in protoplasm and having a marked tendency
to branch (plate X., fig. 9). Harvey states that Seirococais occurs
only at Tasmania and on the south coast of Australia, and not further
west than Cape Northumberland. I find a record of it in the Herb.
Mus. Brit, as far north as Cape York in Torres Straits, well within
the tropics.

The specimen I have examined was collected last year at Geelong,
by Mr. Bracebridge Wilson.

A. LORRAIN SMITH.

EXPLANATION OF PLATE X.

Fig. i. Seirococais axillaris, part of plant. Natural size.

Fig. 2. Part of frond with receptacular branch. Slightly enlarged.

Fig. 3. Transverse section of frond ( x 150).

Fig. 4. Longitudinal section of frond ( x 150).

Fig. 5. Section through male conceptacle ( x 150).

Fig. 6. Antheridial branch ( x 450).

Fig. 7. Section through female conceptacle ( x 150).

Fig. 8. Oogonium (225).

Fig. 9. Paraphyses of female conceptacle ( x 225).

Notes on the Morphology of the Fucacea. 35

XIPHOPHORA BILLARDIERII MONT.

THE earliest description of Xiphophora Billaniierii is by Labillardiere
under the name of Fucus gladiatus (PL Nov. Ho//., ii., p. in, t. 256),
but neither here nor in the description of the plant in Turner's
Fnci (vol. iv., p. 102) is there any definite account of the structure.
The first detailed examination of the fruits was made by Montagne
and described by him in the Annales des Sciences naturelles, ser. 2,
vol. xviii., p. 200, where he founds the genus Xiphophora. In the
same place he dwells at some length on his theory that the antheridia
of the Fucacecs are possibly homologous with the tetraspores of the
Floridece, a theory which other botanists of the time seemed to favour.

In 1860, Harvey added to the genus XipJtopJiora the Fncus cJiondro-
pJiyllus of R. Brown, the type specimen of which, preserved in the
British Museum, bears on the label ' cf. Fitats gladiatus,' in R. Brown's
handwriting, showing that he recognised the relationship of the two
species. Since an authentic specimen of Fucits gladiatus named by
Labillardiere himself is also in the British Museum, I have been able
to compare the other specimens in the Herbarium with the types of both
species ; there are, however, many intermediate forms, showing Fncus
gladiatus with comparatively short receptacles, approaching more nearly
to the short, dichotomous branches of Fucus cJiondropJiyllus. Agardh in
his Spec. gen. et ord. Alg., vol. i., amalgamates the genera XiphopJiora,
PycnopJiycns, Osotliallia and Pelvetia, and forms of them all a new genus
Fucodium, merely reserving the former generic names for sections in
that genus. I have, however, followed the example of Oltmanns and
others, and retained Montague's generic name of Xiphophora.

This genus is confined to the southern hemisphere, where it is found
on the shores of Australia, Tasmania, New Zealand and the Auckland
Islands ; both species appear to be common.

I have examined material of Xiphophora Billardierii preserved in
the British Museum Herbarium from Lyall's Bay, New Zealand, collected
by Dr. Lyall ; from New Zealand, collected by the Antarctic Expedition ;
and a plant, preserved in spirit, sent from Geelong, by Mr. Bracebridge
Wilson.

Xiphophora Billardierii Mont, is, as its generic name implies, a plant
bearing long, sword-like branches, and these constitute the receptacles,
while the vegetative part of the thallus is shortly dichotomous. But, although
the generic name characterises the species which the genus was originally

36 Notes on the Morphology of the Fucacece.

formed to receive, the same cannot be said of X. chondrophyllus Harv.
Its main point of difference from X. Billardierii consists in the entire
absence of the long strap-shaped laciniae, all the branches of the thallus
ending in short dichotomous forks, which form the receptacles. It will
therefore be necessary to modify Montagne's generic description of the
receptacles if X. chondrophyllus is to be retained in the genus. Oltmanns
(Beilrdge zur Kenntniss der Fucaceen, p. 69) points out an external likeness
between XiphopJiora and Himanthalia, which, however, only extends
to both plants having the long receptacles ; and even these, as men-
tioned above, are often in XipJiopJwra very much reduced.

The thallus is composed of three layers of cells : the cortex, consisting
of one row of narrow radiating cells, the parenchyma below the cortex,
and a central strand of long, narrow cells with thick longitudinal and
very delicate transverse walls. The thick longitudinal walls are marked
at intervals by thin places resembling pits, as in the central strand of
Turbinaria ; but, the material at my command not being fresh, I have
been unable to detect continuity of protoplasm. Oltmanns (loc. cit., p. 70)
queries the number of oospores in an oogonium as i ; but, on examining a
large number of conceptacles, I find each oogonium contains four oospores
divided tetrahedrally as in Ascophyllum. So far as I can see, in dried
material the oogonia have no pedicel cell, but grow directly from the
cells at the base of the conceptacle. Both species of this genus are
monoecious. E. S. BARTON.

(For explanation of figures see p. 37.)

NOTHEIA ANOMOLA BAIL, and HARV.

EXCEPT a short account of it in the Transactions of the Neiv Zealand
Institute (vol. xviii., 1885), by Mr. R. M. Laing, I know of no adequate
description of Notheia anomola. The remarkable structure of the plant
led me to make a further examination of it. This so far has shown that
there are points of great interest waiting solution, but with the material
at my command I have been unable to make a detailed investigation of
the plant, and can only therefore give a preliminary note on the subject
at the present time.

The points which appear to me of interest, and worthy of further
investigation, arc :

Notes on the Morphology of the Fucacete. 37

1. The parasitic habit of the plant. Up to the present time but
few truly parasitic Algse have been recorded. Notheia anomola grows
vigorously on Hormosira and Fucodium.

A longitudinal section of the junction of the Notheia and Hormosira,
when stained with methylene blue, shows that the cells of the thallus
of Hormosira stain a deep blue, as also do the outer layers of cells
of Notheia, but that the central cells of Notheia, and a wedge-shaped mass
of tissue, which penetrates the thallus of the host, remain unstained. The
appearance of the cell-contents and the cell-walls in the neighbourhood of
the wedge is suggestive of tissue in a state of degeneration ; but it is im-
possible to say anything decisive on this point unless fresh tissues are
examined.

2. The production of the lateral branches. Each branch rises from
the base of a conceptacle, even though this be filled with hairs and
reproductive organs.

3. The presence of one kind of sexual reproductive organ. The
reproductive organs are developed in well-formed conceptacles. In the
latter, hairs are present at a very young stage ; they appear to grow
only from the base at first, and stand up parallel to one another and
perpendicularly to the base ; other hairs appear later, growing from the
sides, among which the reproductive organs are developed.

The reproductive organs have every appearance of oogonia, but no
antheridia have ever been recorded, and I have examined many plants
for them in vain. The accounts of the number of the oospheres have
differed in various records, but I have invariably found eight present,
generally arranged in the same manner, that is, one oosphere at either
end of the oogonium, the remaining six in the centre in two groups of
three oospheres each. MARGARET O. MITCHELL.

EXPLANATION OF PLATE XI.

Fig. i. Male conceptacle of Xiphophora Billardierii Mont, (x 66).

Fig. 2. Antheridia ( x 375).

Fig. 3. Female conceptacle ( x 66).

Fig. 4. Oogonium ( x 190).

Fig. 5. Notheia growing on Hormosira (a), and Fucodium (b}. Nat. size.

Fig. 6. Diagrammatic section of the thallus to show the origin of a lateral

branch from a conceptacle containing oogonia ( x 250).
Fig. 7. ,, Oogonium ( x 250).

Fig, 8. Section of the junction of Hormosira and Notheia : the cells with

darker contents represent those stained with methylene blue ;

the other cells are unstained ( x 250).

r

38 Notes on the Morphology of the Fticacece.

SARCOPHYCUS POTATORUM KUTZ.

THE earliest record of this plant is a description and figure of
Fucus potatorum in 1806 by Labillardiere,* who assigned to it the specific
name potatorum in consequence of having observed the natives of the woods
round Van Diemen's Land use portions of its great fronds, folded in the
form of a pouch, for the purpose of keeping fresh water. Dawson Turner-f-
alse describes it, and gives a figure of the whole plant drawn from a
specimen in the Herbarium of the Jardin des Plantes at Paris. The next
record is that of Lamouroux,t who, in 1813, mentions this plant under
the name Laminaria potatornin. The genus Sarcophycus was subsequently
founded for its reception by Kiitzing. He was, however, under a
misapprehension as to the nature of the oogonia, since he regarded these
as tetrasporangia, and, believing the plant to be one of the Floridece
placed it in the order Chatangiea. Its systematic position was,
however, clearly recognised by Hooker, for in his Flora Australica,
p. 456, after describing Durvillcea Harveyi, he mentions this plant under
the name Laurinaria potatorum, and observes that it probably belongs to
the genus Durvillcea. Lastly, Areschoug,|| in 1847, gives a description
of the plant under its name of Sarcophycus potatorum.

The genus founded, as has been noted, as the result of a misapprehension,
has been retained, for, although the genera Sarcophycus and Durvillcea
are very closely allied, they differ in the fact that, while the frond of
Sarcophycus is solid to the centre, that of Durvillcea is, beneath the
cortex, lacunar in structure, the lacunae being separated by strands of
anastomosing filaments. I am indebted to Mr. Bracebridge Wilson
of Geelong, Australia, for collecting the material on which I have
worked.

The mature plant of Sarcophycus potatorum consists of a disk, a stipe,
and a frond portion. From the flat and more or less circular disk there
rises a stipe, which is cylindrical below and flattened above. The stipe bears
a huge segmented frond, the segments of which are of tough, leatherlike
consistency, and reach an enormous length and considerable thickness.

A section through the frond shows that it is composed of a cortical

* PL Nov. Holt., ii., p. 112.

t Fuci, vol. iii., tab. 242.

J Ann. Mus. d'Htst. A T at., xx.

Phyc. Gen., p. 392.

|| Ofuersigt of Kongl. Vetensk.-Akad. Forhandl. Stockholm, 1847, p. 267.

Notes on the Morphology of the Fucacece. 39

and of a medullary portion. The cortical portion is made up of parallel
filaments running perpendicularly to the surface of the thallus. These
filaments are very closely packed and regularly septate, causing the cortical
portion to present the appearance of a compact mass of parallel rows of
cells. As the filaments pass down to the medullary portion, they cease to
run parallel to each other, and branch out irregularly in all directions,
anastomosing more freely, while the individual cells expand in diameter,
and the septa are further apart. Thus the medullary portion is made
up of a mass of branching and anastomosing filaments running in every
direction, forming a tissue which is looser than that of the cortical portion.
The conceptacles extend to the same depth as the cortical portion, their
bases being surrounded by filaments of the medullary portion. The
material at my disposal consisting only of mature parts of the fronds,
I have been unable to investigate either the mode of apical growth or
the mode of development of the conceptacles.

The male conceptacles are of the ordinary Fucaceous type, and contain
antheridia borne on branching filaments (plate XII., fig. ij.

The female conceptacles (plate XII., fig. 2) contain oogonia, the contents
of which divide into four oospheres in the following manner : the proto-
plasm of the immature oogonium divides into three transversely, and the
middle portion then divides again into two longitudinally. This mode of
division was first observed by Kiitzing, who, as has been already noted,
regarded the four resulting oospheres as tetraspores. A peculiarity about
the oogonia is ( that, while many of them are developed in the usual
manner from the lining cells of the wall of the conceptacle, others are
borne laterally on branching filaments (plate XII., fig. 3). These filaments
are longer, and their component cells are larger, than in the case of the
antheridia-bearing filaments. There are only a very few sterile filaments
on either the male or the female conceptacles, and these occur chiefly at
the mouth of the ostiole. FRANCES G. WHITTING.

(For explanation of figures on plate XII. see p. 46.)

VII.

ON CHLOROCYSTIS SARCOPHYCLA NEW

ENDOPHYTIC ALGA.

WHEN Mr. Bracebridge Wilson was collecting Sarcopliycus, he observed
the occurrence of raised circular patches, forming gall-like structures
on certain of the fronds. He collected and sent to the British Museum,
for the purpose of investigation, material on which such patches occurred,
saying that he found them to be caused by what he took to be an
unicellular Protococcaceous Alga. An examination of this material,
which was put into my hands, has proved that this is indeed the case,
and I will proceed to a description of these peculiar patches and the
endophyte producing them.

The patches (plate XII., figs. 4, 5, 6) vary in size from a quarter of
an inch to one inch in diameter, and are generally circular, but sometimes
rather oval in form. They are accompanied by a bulging of the frond
causing malformations, of which, while one surface is raised above, the
other is somewhat depressed below the surrounding portion of the
frond. The surface generally presents either a smooth or a corrugated,
rough, and decomposed appearance, sometimes only one, and sometimes
both surfaces being thus affected.

These malformations are caused by the growth of an unicellular
endophytic Alga in greater or less abundance between the filaments
of the cortical portion of the thallus.

The endophytic cells (plate XIL, fig. 10) are thin-walled, chloro-
phyllaceous, various in size, and irregular in shape. They appear to
vary in shape according to the pressure exerted upon them by the
surrounding portions of the thallus. The cell when young contains
homogeneous protoplasm (plate XIL, fig. loa), which at first, completely
filling the cell-cavity, often shrinks away later from the cell-wall.
Later, by the process of free-cell-formation, the protoplasmic contents
become differentiated into a very large number of small spores, which

4 2 Chtorocystis SarcopJiyci.

generally completely fill the cell-cavity (plate XIL, fig. n). I have
found the number of spores in many cases to exceed one hundred, the
number, however, varying according to the size of the cell containing
them. Even in the spirit material on which I have worked there often
remains a decided green colouration of the contents of a cell before spore-
formation ; on the other hand, there is always a marked diminution of
colour after spore-formation has occurred. Besides these two prevailing
conditions of the contents of an endophytic cell, there remains yet a
third condition, the meaning of which I shall have to discuss later. In
this the protoplasm is segmented into portions of considerably larger size
than the small spores which I have described (plate XII., fig. 10). The
segmentation appears to proceed simultaneously throughout the whole
protoplasm, and the resulting segments, though fairly regular in size, are
not very uniform in shape ; the green colouration is as well marked as it
is before such segmentation has occurred.

Of the endophytic and unicellular Algae hitherto described, the
species to which I believe this endophyte of SarcopJiycns to be the most
nearly allied is CJilorocJiytriuni incliiswu, described by Kjellman*, in the
fronds of the Floridean SarcopJiyllis arctica. The spore-containing cells
which I have described, though considerably smaller in size, yet bear a
most striking resemblance to those figured by Kjellman. Kjellman refers
his species to the genus Cklorockytrium, but it seems probable that
it, together with other marine forms, should be placed in the genus
CJilorocystis of Reinhard, and I venture to think that, so far as can be
judged without knowing the whole life-history, the species occurring in
SarcopJiycns belongs to the same genus, and I have therefore named it
CJilorocystis SarcopJiyci.

The genus CJilorocystis was founded by Reinhardf on ClilorocJiytrmm
CoJmii, described by Dr. Wright* as occurring in several species of
ScJiizonema, in PolysipJwnia urccolatn and other marine Algse.

The chief point of distinction between the two genera is that in CJiloro-
cJiytrinui spore-formation occurs by repeated division of the protoplasm
into two, the resulting portions being separated from each other by a
cell-membrane, while in CJilorocystis the spores arise by free-cell-formation.

The characteristics of the genus CJilorocystis are given as follows in
De Toni's Sylloge Algarum (vol. i., p. 637) : ' Cellulae sphaericae, tuber-
culo lateral! (ut in CJdorocJiytrio} instructae, chlorophora perforata varie

* Alga; of the Arctic Sea, p. 320, tab. 31.

t Proc. Neu> Russian Society of Naturalists, vol. ix., p. 214.

\ Trans, of Royal Irish Academy, vol. xxvi., p. 355.

A Nezv Endophytic Alga. 43

lacerata . . . pyrenoide unico instructa. Propagatio zoogonidiis e plasmatis
divisione simultanea ortis et sine copulatione e membrana cellulse
matricalis examinantibus.'

Chlorocystis Sarcophyci departs from the generic character in the absence
of any cellulose protuberance or neck-like portion. This, however, would
not appear to be a very important point, for Dr. Perceval Wright, to whom
I am indebted for very kindly examining some of my slides, informs me
that, when he has found Cli. CoJinii developed in the interior of a cell-
tissue, the cells are sometimes quite globular. (See also Kjellman, loc. cif.,

figs. 10, 17.)

There is difficulty in explaining the division of the protoplasm into the
larger segments I have described above. I thought at first that, in some
cases at least, the protoplasm might become thus segmented before the
final differentiation into spores. Since, however, I have never found any
of the segments in question differentiated into small spores, which always
arise, on the other hand, by differentiation of the original homogeneous
protoplasm, this interpretation is not tenable. A somewhat similar
segmentation of the protoplasm was observed by Cohn* in the cells of
Chlorochytrium Lemnce and was regarded by him as a stage in the
formation of the spores. This observation was, however, not confirmed
by Klebsf, who found the spores to be formed by repeated division of the
protoplasm into two.

Another possible interpretation of the segments is that they are larger
spores, in which case this would present a parallel with the small and large
spores of Chlorocystis Cohnii. Since, however, in the material I have ex-
amined I have found these segments, though fairly regular in size, to lack
uniformity of shape and in many cases the definitely rounded-off appearance
of the small spores, I should not be justified in offering this explanation with
any degree of certainty. The nature of these segments must therefore be left
open to be decided by the result of future investigation. An examination of
fresh material would probably offer an immediate solution of the difficulty.

There remains only to be described the mode in which the endophytic
cells produce the malformations, and the manner of their escape from the
tissue.

The malformations exhibiting considerable variations in external
appearance and minute structure, it will be convenient to describe them
at three arbitrarily chosen stages.

There are, in the first place, those which, while presenting the usual
bulging of the frond, have their surfaces smooth and not decomposed.

* Beilrdge, bd. i., heft 2. t Bot. Zeit., vol. xxxix.

44 Chlorocystis Sarcopkyci.

These, when cut across perpendicularly to the surface, show a very slight
amount of swelling in the cortical tissue of one or both surfaces, the whole
thickness of the tissue in the diseased portion being slightly greater
than that of the surrounding portions of the frond. A section of the
raised surface of such a malformation (plate XII., fig. 7) will show the
endophytic cells between the somewhat swollen filaments of the cortex,
thus diverting these from their perpendicular course. Although cells
containing spores may occur, yet the contents are for the most part of
homogeneous protoplasm.

Secondly, there are those of which one surface at least is rough and
disintegrated, and when cut across show increased swelling in the cortical
tissue, the whole thickness of the tissue being considerably greater than
that of the surrounding portion of the frond. A section (plate XII., fig. 8)
shows endophytic cells in the greater number of which spore-formation
has taken place among the cortical filaments which are in this case
swollen and becoming separated from each other. In parts the individual
cells have become rounded off from each other, forming a loose tissue
which at the surface is completely broken down into a discoloured mass
of disorganized cells. It is by this decomposition of the filaments sur-
rounding them that the endophytic cells are set free. As the decomposition
proceeds, the endophytic cells in deeper portions of the cortex are enabled
to escape ; in some cases this escape is facilitated by the exfoliation of
large portions of the diseased cortical tissue.

Thirdlv, there are those which when cut across show that the thickness

* '

of the thallus is considerably less than in the surrounding portions of the
frond, owing to the destruction of a portion of the cortical tissue. A
section (plate XII., fig. 9) shows the loose tissue described above broken
down at the surface into a disorganized mass without any remaining
endophytic cells. These, in fact, have all escaped from the tissue.

It appears, then, that the cells of Chlorocystis Sarcophyci are able, in
some manner (and perhaps especially during the process of spore-forma-
tion), to exert an influence upon the surrounding tissue, causing at first
swelling and loosening of the tissue, and finally complete disintegration of
the cells. In this point it differs from the species that have been hitherto
described, for these inflict no injury on the plant in which they live. The
three stages described above may, however, occur in one and the same
malformation, for the disintegration proceeds centrifugally, and in many
cases a gradation occurs from a peripheral portion, of which the cortex
is almost uninjured, to a central portion, in which it is completely dis-
integrated.

The disintegration of the tissue may proceed still further, for Mr.

A New Endophytic Alga. 45

Bracebridge Wilson describes these malformations as resulting in the
formation of circular holes. I had only one instance of a malformation
with a central hole, but I think that in this case the destruction of tissue
was certainly due to the action of the endophytic cells, for these were
extremely abundant in the very thin tissue surrounding the hole. It is
easy to suppose that, when once the denser cortical portions of both
surfaces had disappeared, the looser medullary portion would soon be
destroyed.

The endophyte may or may not occur among the filaments of the
depressed surface of the malformation, and when it is present the injury
to the tissue is not generally so far advanced as that of the raised surface.

The presence of the endophyte among the cortical filaments of both
surfaces of the thick mature frond would seem to imply that the infection
occurs when the frond is quite young and very thin. This perhaps would
also explain the bulging of the frond in the diseased portions, for an
infected surface becoming raised and swollen might, as it were, carry
up with it the whole tissue of a thin and delicate thallus. This, however,
is merely hypothesis for the mode of entry of the endophyte into the
plant, and the first stages of its life-history within the tissue are entirely
unknown to me. I have been unable to find any trace of the endophyte
in any younger stage than those I have described, probably because I have
had no young material to examine.

Briefly to sum up the results of this investigation, there exists among
the cortical filaments of the frond of SarcopJiycus potatorum an unicellular
Protococcaceous Alga, which, so far as I am able to judge from the life-
history known to me, belongs to the genus Chlorocystis, and I have named
it as follows :

Chlorocystis Sarcophyci n. sfi., cellulis globosis oblongis vel irregularibus,
10-40^ diam., in statu vegetativo viridibus, in matrice omnino inclusis,
collo destitute, zoogonidia emittentibus.

Hab. in SarcopJiyci frondibus ad oras Novae Hollandias prope Geelong.
coll. J. Bracebridge Wilson.

It differs from the species hitherto described, in that, while these inflict
no injury on the tissue of the plant in which they live, C. Sarcophyci pro-
duces conspicuous malformations, and, in consequence of the disintegration
which it in some way induces, its spores are enabled to escape from the
tissue. A study of fresh material might determine whether the endophyte
derives any nutritious benefit from the disintegration of the cells of
which it is the cause; whether, in fact, this is a case not merely of

'Raumparasitismus,' but of true parasitism.

FRANCES G. WHITTING.

G

46 Chlorocystis Sarcophyci.

EXPLANATION OF PLATE XII.

Fig. i. Sarcophycus potatorwn. Male conceptacle ( x 105).
Fig. 2. ,, Female conceptacle ( x 105).

Fig. 3. ,, Oogonia developed on branching filaments : a, b, c,

being stages in the formation of oospheres
( x 440).
Figs. 4, 5, 6. Chlorocystis Sarcophyci. Malformations on frond of Sarcophycus

fotatorum. Nat. size.
Fig. 7. ,, Section of malformations, first stage, showing

endophytic cells ( x 105).
Fig. 8. ,, Section showing the endophytic cells escap-

ing by the decomposition of the cortical

filaments ( x 105).
Fig. 9. ,, ., Section of the disorganized tissue after the

escape of the endophyte ( x 105).
Fig. 10. ,, Group of endophytic cells : a, contents

undivided ; ^, larger segments ; c, spores

Fig. ii. Group of spore-containing cells ( x 440).

VIII.
ON HALICYSTIS AND VALONIA,

WHILE dredging in the Clyde Sea area, last August, with Dr. Schmitz
of Greifswald, we were so fortunate as to find, both on the bar at the
entrance to Loch Goil and in the Kyles of Bute near Inchmarnock, an Alga
of a generic type new to British seas. It was growing in 8-11 fathoms
water, attached to shells and to Lithothamnion. Our discovery was the
Alga at present known as Halicyslis ovalis, Aresch. (= Valonia ovalis, Ag.
Spec. Alg., i. p. 431, and Gastridium ovale, Lyngbye, Tent. Hydrophyt.
Dan., p. 72, tab. i8b), and its nearest occurrence to our shores previously
recorded was Molde, Norway (Areschoug), and the Faroe Islands
(Lyngbye). Bornet has recorded it also from Biarritz (Alg. Schousb.,

P- 50).

H. ovalis has the appearance of a small round or oval bladder (plate
XIII., fig. i), of the colour of a green grape, attached to its substratum by a
very short, delicate, cylindrical stalk, terminating downwards in a minute
disk. The cavity of this relatively thick-walled stalk communicates
upwards directly with the interior of the bladder. There is no formation
of rhizoids such as are found in Valonia.

Dr. Schmitz and I have both independently examined the minute
structure of this plant, and our observations are in exact agreement. He
has been good enough to send me in his own words an account of his
examination, and I have translated it for convenience sake, as follows :
The somewhat thick membrane of the unicellular bladder appears out-
wardly quite smooth, and shows none of the striation to be seen easily for
the most part in Valonia cells. The stratification of this membrane is
exceedingly fine and barely recognisable, and has no appearance of
exfoliation (plate XIII., fig. 4). This membrane in the living plant is
coated with a thin layer of protoplasm, in which very numerous small
chlorophyll-grains and nuclei are embedded. The great lumen is filled

47

48 Halicystis and Valonia.

with cell-sap. The small, green chromatophores (plate xill., figs. 2 & 3)
lie almost everywhere in a single more or less dense layer in the proto-
plasmic coat. At the upper arched portion of the bladder this layer
becomes very dense, and in places so much so, that the granules no
longer lie flat, but become partly tilted over on each other. Below this
the press of granules is less, and on the sides where they lie flat, spaces
appear free from granules. The chlorophyll-grains themselves are small,
flat roundish or oval disks of somewhat varying size and rounded outline,
and I have never seen sharply angular or lobed forms. These disks are
wholly without pyrenoids, and I could not discover in the minutely
examined specimen of this Alga any amylum in the whole protoplasmic
layer, rich in chromatophores as it is.

Very numerous minute nuclei are distributed among the chromato-
phores, or lie on the inner surface of the chromatophore layer. They are
flattened disks, round or oval in outline, and contain each a minute
nucleolus. The nuclei are to be detected only here and there in specimens
hardened without staining ; but after this process they may be easily
recognised. Without staining, the minute glistening nucleoli may be
recognised in the spaces free from chlorophyll-grains. The nuclei are
scattered irregularly throughout the whole protoplasmic layer, now singly,
now in pairs, or in groups of several near each other. In the living
plant they probably wander in the protoplasmic layer, as for example
those of Codimn* do.'

Dr. Schmitz then proceeds to point out the characters which distin-
guish Halicystis from Valonia, to some species of which it bears, as will be
seen later, a very striking resemblance. The nuclei of Valonia, as he has
pointed out.t are much more compact and more evenly distributed at
fairly regular distances in the protoplasm. The chromatophores exhibit an
irregularity of shape, being roundish, but angular, some of them with
sharp angles, and of varying size ; moreover, some of them are provided
with pyrenoids, and these are of regular occurrence among the others,
which have no pyrenoids. He comes to the conclusion that the above
differences of characters in combination with the non-development of
rhizoids and a cell-membrane so little stratified as to exhibit no exfoliation
completely justify the separation of Halicystis from Valonia, as Areschoug
has done. Dr. Schmitz goes further, and insists on the removal of the

* Berthold, Zur Kenntniss der SipJwneen imd Bangiaceen (Mittheil. d. Zoolog. Station
A'capel, ii., p. 76).

+ Beobachtungen iiber die vielkernigen Zellen dcr Siphonocladiaceen. Halle, 1879.
\ Phyc. Scandinav., p. 446. Upsala, 1850.

Halicystis and Valonia. 49

genus from the Sipkonocladaceee, to which Valonia undoubtedly belongs,
to the SipJioneae, with which group it agrees in the characteristic chlorophyll
layer, and the arrangement of its nuclei. Although I am most strongly
inclined to agree with this determination of the position of Halicystis, I
confess to a preference for the more cautious attitude of awaiting the story
of its reproduction before committing myself to full agreement. I have
observed all the details described above by Dr. Schmitz, working quite
independently and simultaneously at the material we collected, but such
histological details appear to me to be by themselves insufficient support
for this step. Since I may fairly claim, however, to have given in several
papers unqualified support to Dr. Schmitz's admirable work in establishing
the group of Siphonodadacecs* I may take this opportunity of expressing
surprise at the small extent to which it has been adopted, and to my con-
viction that students who cling to the old system will, after due study,
discover the Siphonodadacece to be one of the most natural orders of the
green Algae. With regard to the present case I must be understood to
express not dissent, but merely hesitation. If we grant the point, the
question of its position among the Siphonecs becomes interesting, and on
this point Dr. Schmitz's remarks, which I translate, are of great value :

' The genus Halicystis stands in a somewhat isolated position among the
Siphonese, but its vegetative structure, which is all we know of it, recalls that of
Botrydiitm.\ This genus, apart from its occurrence in fresh water, is mainly
distinguished from Halicystis by its branching filamentous rhizoids. In its simple
bladder-like thallus Halicystis is further removed from such genera as Vancheria,
Derbesia and Bryopsis, which possess a more or less branched filamentous thallus.
A more exact determination of its position among the Siphoneae can only be made
when the whole life-history and reproduction of Halicysns has been ascertained.'

Since Reinke has recently* mentioned the discovery of H. ovalis in
Heligoland, it may be mentioned that Dr. Schmitz found it there so long
ago as September, 1881, while dredging in the Nordhafen, and he even
then came to the conclusion that it belonged to the SipJionecs rather than
the Siphonodadaccce, During the winter of 1879-80, Dr. Berthold called
his attention to another form obtained also by dredging in the Gulf of
Naples on stony ground, rich in corallines. On comparing the preparations

* Sitzber d. Naturf. Gesellsch. su Halle, Nov. 30, 1878.

t Botrydiitm possesses in the upper portion of its bladder-shaped thallus numerous
small disk-shaped chromatophores, free from pyrenoids, as Halicystis does. The asser-
tion of Wille (Engler Prantl. Pflanzenfamilien, i., p. 123), that in the Botrydiaceae
only a single much-lobed chromatophore is present, is not correct for Botrydium;
Codioluin, which Wille (loc. cit.} places in the Botrydiacece, does not belong to the
Siphonea: at all ; the unicellular thallus of C. gregarium possesses a single much-lobed
chromatophore with numerous pyrenoids and a single nucleus.

t Berichte der deutschen botan. Gesellsch. 1889, p. 369.

50 Halicystis and Valonia.

he then made of this form with H. ovalis, he has come to the conclusion
that the Neapolitan plant is another species apparently identical with the
form described by Zanardini* in the Adriatic as Valonia ovalis. This
form, though smaller, much resembles H. ovalis externally, but differs in
having a shorter and more blunt stalk, a more uneven outer surface, but
mainly in having larger chromatophores of different shape (plate XIII.,
fig. 5 ). These occur in great numbers, and are of a rather long spindle
shape, and are provided in the centre with a single clear pyrenoid, very
like the chromatophores of Bryopsis. Its nuclei, however, resemble gene-
rally those of H. ovalis. Dr. Schmitz proposes to retain this form under
Halicystis, under the name of H.parvula, and calls attention to the fact
that similar differences in the chromatophores of different species in the
same genus are to be found in Derbesia and Vaucheria. I may be excused
for pointing to such a circumstance in justification of my caution in
attaching what may be undue weight to evidence based on these characters.

The collection and examination of Halicystis ovalis recalled to me
certain observations I made in 1886 in Grenada on Valonia ventricosa, of
which I then obtained magnificent specimens. This remarkable plant,
consisting of a cell as large as a hen's egg, and of much the same shape,
varying to the shape of a pear (plate xill., fig. 6a) I found growing in
fairly shallow water (i to 2 fathoms) attached to Galaxaura lapidescens,
and in 5 fathoms water on the rhizoids of A vrainvillea longicaulis (plate
XIII., fig. 6b), the latter being much smaller specimens. It resembles
Halicystis in consisting of a single great, unbranched cell, and differs from
other species of Valonia (except V. Forbesii probably) in this respect. It
possesses no stalk like Halicystis, but agrees with Valonia in the pro-
duction of rhizoids terminating in tenacula (plate XIII., fig. 7). Its
cell-wall is stratified and exfoliates readily (plate XIII. , fig. 10) on cutting
it in section, and exhibits faint striation. The chromatophores possess
a pyrenoid (plate XIII., fig. 9). In its protoplasmic layer there are both
single starch-grains and masses of amylum, as in other species of Valonia.
In these respects it shows distinct differences from Halicystis.

The remarkable point, however, in these specimens (as also in others
collected in Bermuda by Mrs. Whelpdale) is the occurrence in them
of reproductive organs. Since such organs have been hitherto unknown
in Valonia (the so-called ' germ-cells/ figured by Nageli, Kiitzing, and
others, being concerned in branching, but possibly also in reproduction)
their occurrence deserves detailed description. The cells in question
(plate Xill., fig. 8a,&,c,d) have plainly arisen by free cell-formation within

* Saggio di classif. nat. d. Ficee., 1843, p. 59.

Halicystis and Valoma. 5 i

the great mother-cell, and in this process I have discovered various stages
in different specimens. It evidently proceeds slowly, and is not a
simultaneous act, since in some specimens comparatively few cells are
to be found, while a considerable amount of free protoplasm remains, and
in others the cells are very numerous, while mere traces of free protoplasm
are left over. They are, moreover, of various sizes, and, while generally
round, odd forms are to be met with (plate XIIL, fig. 8, b, c,d), dumb-bell
shaped in varying degree, and others suggesting a process of sprouting in
yeast fashion. Their membranes vary slightly in thickness. They occur,
not only within the great mother-cell cavity, but even in the rhizoids
and the small. marginal cells from which these spring.

I observed these bodies not only alive in Grenada, but have frequently
studied them since in the preserved material. Dr. Schmitz, to whom I
sent specimens in spirit, has been good enough to confirm my observations
in every detail, and to suggest that possibly they are abnormal repro-
ductive bodies produced by slight accident to the plant in being gathered,
as he has observed a similar case in SipJionodadus Wilbergi ( Vielkernige
Zellen der Siplionoclad., p. 33), or it may be by too strong illumination, as in
the formation of aplanospores in Botrydium. My own view of the matter
is simply that they are the normal reproductive organs of Valonia (a
possibility not excluded by any means by Dr. Schmitz also) for the
reason that I could find no support for the view that they are abnormal.
Not only do they occur in all the specimens I have examined, both from
Grenada (several localities) and from Bermuda, but I have seen them in
the condition described shortly after their removal from the water, before
accident, if it had occurred, could well have had time to operate in this
fashion. That they are not the result of strong illumination is plain from
the fact that they occur in specimens buried among the rhizoids of Avrain-
villea in coral-sand under five fathoms water. On the other hand,
their irregularity in size and shape favours the view of their abnormality,
but their variation in size may be accounted for by their gradual formation,
as mentioned above, and the exceptional forms in point of shape may
be explained by the slight abnormality of unequal division of the proto-
plasm in the process of their formation.

Attempts to cultivate them were unsuccessful under the circumstances
of my position in Grenada, though in one instance I observed the same
cells unchanged for several successive days. As the plant is not a rare
one in the West Indies, and its condition as described frequent, further
inquiry into the development of these reproductive bodies ought not to

present serious difficulty.

GEORGE MURRAY.

52 Halicystis and Vafama.

EXPLANATION OF PLATE XIII.

Fig. i. Halicystis ovalis. Natural size.

Fig. 2. ,, Chromatophores and nuclei. Drawn by Dr. Schmitz.

Fig. 3. ,, The same from a photograph ( x 1200).

Fig. 4. Section of membrane with layer of contents ( x 150).

Fig. 5. Halicystis parvula. Chromatophores and nuclei. Drawn by Dr. Schmitz.

Fig. 6. a and <, Valonia venlricosa. Natural size.

Fig. 7. ,, Rhizoids ( x 66).

Fig. 8. a, b, c, d. Reproductive bodies ( x 375).

Fig. 9. ,, ,, Chromatophores ( x 750).

Fig. 10. ,, Section of wall, showing exfoliation ( x 150).

Phyc.Mem.Parb IT .

Pla

Hfe

Berjeau 2c HigWeylibk.

A.L S. del.
COCCOPHORA LANGSDORFII

Thyc.Mern.Part IT

Plate X.

; \ v

t

"

-

.

Haithart

SEIROCOCCUS AXILLAR1S Grey .

\Iem.T?art II.

IPlate XI .

- \\ nv**' l v I

am

>&3DH5rP<
:5Q

>fe

/ \ I O5^ * ( lC^

<9

esu.&,Hig]hleyMi. E.S.B.8C EO-M. del . Hanhait imp.

' ! BILLARD1ER1I 3fc;t. (1-4). NOTHEIA ANOMALA Bail.* Hour r. (5-8) .

PKyc.Mem.Parb II

Plate XII

.

v

..

' -

,

: I
1

.ni

S T U|
,i**

F. &.W. del . Hankarb imp .

SARCOPHYCUS POTATORUM 2Gibz.(1-3) CHLOROCYSTIS SARCOPHYCI Wki,t.(4-.

. Mem .Part U

Plate

.
f

/

m

\ . 1 ' '

-

Vi \

',

\Sg

;

.

So

5o.

.

.

IL

f

'

fas

4

HAL I CYST IS OVAL IS ArescTi,. (J-4 ) .H.PARVULA
VALONIA VENTRICOSA JiAc. ('

HanKarb im

IX.

ON THE STRUCTURE OF HYDROCLATHRUS

BORY.

HYDROCLATHRUS SINUOSUS Zanard. and H. canccllatus Bory have by
recent authors been separated from the group of the Asperococcoidece in
which they were formerly placed, partly on account of the structure of
their thallus, and partly because they have not yet been shown to possess
any other than plurilocular sporangia, while all those plants now placed
in the group of the Asperococcoidecs have, with one exception,* well-
defined unilocular sporangia only.

I examined the plants because I have been unable to find any detailed
description of them, and also with a view to determining the method of
development of certain groups of hairs scattered abundantly over the sur-
face. These are, apparently, analogous to the so-called ' fasergriibchen 'f
or ' cryptostomata ' of the higher Fucacece, and it seemed possible that the
development of such simple forms might throw light on the origin of
structures which occur so generally throughout this group.

The plants examined were collected at Anguilla by Mr. W. R. Elliott,
and in the Gulf of Manaar by Mr. Thurston ; I have only, therefore, been
able to describe specimens preserved in alcohol.

* J. H. 'Rvfihscm., Journal of Botany. 1891

t Throughout this and the succeeding paper, 'cryptostomata' will be used in place
of 'fasergriibchen.' Though there are objections to the term, these do not outweigh the
advantage of using an adopted term of classical derivation in place of a German word. (ED.)

53 H

54 Structure of Hydroclathrus.

The form of Hydroclatlims sinuosus is that of a hollow sphere, with a
slightly corrugated surface, the largest specimens examined having a
diameter varying from two and a half to three inches. The plant adheres
to stones and to other Algae, but has no special organ of attachment. At
first it probably consists of a solid mass of cells, but this condition is only
transitory, since the youngest plants examined have always been found to
be hollow, while the cells of the layer which borders the hollow interior
generally present a ragged appearance, as if they had been torn apart from
each other.

A section of the thallus (plate XIV., fig. 2) shows that it is made up of
from five to six layers of cells ; those forming the outermost or epidermal
layer are small closely packed polygonal cells, each with a well-developed
nucleus and dense protoplasmic contents. Below them come one or two
layers of cells larger than those of the epidermis and with thinner walls.
The innermost cells of all resemble those above them, excepting that they
show a marked increase in size, and take on a more spherical form owing
to the absence of pressure. When they have reached their limit of growth
the walls become torn, either on account of the constant tension exerted
by the growing thallus or by degeneration of the substance of their cell-
walls.

The peripheral walls of the epidermal cells increase slightly in thick-
ness, thus forming a kind of cuticle which covers the whole plant, and is
ultimately pushed off as a result of the outward growth of the epidermal
cells below it during the formation of sporangia (plate XIV., fig. 3).

H. sinuosus has neither an apical cell nor any area of special growth,
and increase in size takes place by division of the epidermal cells.

While the plant is still quite young, there is no indication of the
formation either of cryptostomata or sporangia, and the epidermal layer
has everywhere an uniform appearance ; but, as the thallus increases, a
surface view of the epidermis (plate XV., fig. i) shows that localised changes
are taking place simultaneously; an isolated cell or several cells in a group
become separated off from the surrounding epidermis, each loses its
polygonal shape and becomes cylindrical. This change in appearance
denotes the first stage in the formation of a cryptostoma. In a radial
section of such a group each cell is seen to be divided by a transverse
wall, but there is no indication of such longitudinal division as occurs in
neighbouring epidermal cells. The lower of the two cells again divides
transversely, and this method of division continues till a long row of
cells has been formed, making in fact a hair. Simultaneously with the
formation of these hairs, the cells immediately surrounding them undergo
similar changes, and thus the cryptostoma enlarges radially. Meanwhile

Structure of Hydroclathrus. 5 5

the thallus continues its growth, so that the basal cells of the hairs which
were originally in the same plane as the epidermis have now come to lie
below it, and the whole structure is suggestive of a conceptacle.

H. sinuosus is reproduced by spores contained in plurilocular sporangia.
So soon as a cryptostoma is fully formed the epidermal cells immediately
surrounding it begin to elongate, pushing up the overlying cuticle, which is
finally ruptured (plate XIV., fig. 3), and peels off as the growth of the cells
spreads centrifugally. From the appearance of many sporangia at different
stages of development it may be seen that each epidermal cell becomes di-
vided, as in the case of the hairs, by a transverse wall ; the lower of the two
cells, which represents the basal cell, remains unchanged, while the upper
one divides transversely. Of the two cells thus formed the lower one divides
longitudinally into two, the protoplasm of each half going to form a spore,
while the upper half again divides transversely, and the same process of
division takes place, with the formation of two more spores. Growth con-
tinues until from twelve to sixteen spores have been produced (plate XIV.,
fig. 4). Of the sporangia forming a sorus, those immediately surrounding
the cryptostoma come to maturity first, and the spores are liberated; what
becomes of them afterwards it is impossible to determine unless growing
plants can be examined. After the liberation of the spores the sporangia 1
walls gradually disappear, and the basal cells originate a new growth.
They again elongate, and each is divided into two parts by a transverse
wall ; the lower half remains unchanged, and the upper one grows into a
long, club-shaped paraphysis (plate XIV., fig. 5), the contents of which, in
spirit material, are entirely concealed by a dark-brown colouring substance.
In a young paraphysis this is distributed evenly all over the cell ; in older
ones it appears to be collected into small globules, but how much of this
appearance is due to the method of preservation it is impossible to say.
While the basal cells nearest the cryptostomata are giving rise to
paraphyses, those farthest away are still producing sporangia, and, as the
growth of the sporangia spreads radially from the cryptostomata, so also
does that of the paraphyses, until finally all the sporangia have disappeared,
and there are scattered over the thallus in the place of the sori groups of
dark-coloured paraphyses, each having a central cryptostoma. The hairs
of the latter are shrunken, and the cells have lost their contents, but they,
together with the paraphyses, appear to persist throughout the remaining
life of the plant. It is of interest to note here slight differences in the
structure and appearance of the plants coming from different localities
(cp. plate XIV., figs. 3 and 6). Thus the form of the inner cells of the thallus
is round in the plants coming from Madras, and more elongated in those
from Anguilla. Again, in the plants coming from Anguilla the crypto-

56 Structure of Hydroclathrus.

stomata and paraphyses are sunk in slight depressions, whereas in the
plants collected at Madras the surface is uniformly even.

Hydrodathms cancdlatus. H. cancellatus, in outward appearance,
offers an entire contrast to H. sinuosits ; the thallus consists of branches
which anastomose to form a network with large or small intervening
meshes. The whole plant is usually more or less spherical or ovate, and it
is found sometimes growing together with and adhering to //. sinuosus or
by itself, and attached to some other Alga.

While one region of the thallus is fully formed and capable of producing
sporangia, another may be still in a state of growth. Some of the branches
are flattened, others are almost cylindrical. Along the inner side of the
latter, i.e., the side which is towards the interior of the whole net, the
continuity of the epidermal layer is destroyed, and the underlying tissue
appears to be ruptured almost as far as the central line of the branch
(plate XV., fig. 3) ; the epidermal layers on either side of the line of
division bend towards each other, and a section of such a branch is
reniform rather than cylindrical. On all the branches cryptostomata and
sporangia are present in various stages of development. Their growth,
as far as I have been able to observe it, follows closely that of H. sinnosns,
the material to which I had access being too young to show whether
vegetative growth takes place from the basal cell of the sporangia, giving
rise to paraphyses.

The point of interest about these plants is the very elementary method
of development of the cryptostomata ; and although no true conceptacle is
formed with special lining cells, such as occurs in the higher Fucacece, yet
the growth, both of the hairs and reproductive organs is initiated by the
alteration in form and by the division of one of the epidermal cells, which
might with truth be called an ' initial cell.'*

To leave this statement without a qualification would not, however, be
exact, since my observations do not exclude the possibility of the initiative
being taken by a small group of initial cells dividing simultaneously,
instead of a single one. Whether the decision of this point may ultimately
prove to have a bearing on the affinities of such organs with the
cryptostomata of the Fucacece is one that at present I am unable to
determine.

MARGARET O. MITCHELL.

* Bower, Quart. J our ti. iVicr. Sa., vol.

Structure of Hydroclathrus. 57

EXPLANATION OF PLATES.

9-

PLATE XIV.

Fig. i. Hydroclathrus sinuosus. Nat. size.

Fig. 2. Section of thallus with young cryptostoma( x 250)

a. Epidermal layer.
/; Inner cells of thallus.
c. Young cryptostoma.

Fig. 3. Mature cryptostoma with young sporangia ( x 250)

Fig. 4. ,, Part of sorus to show adult plurilocular sporangia

(x 250).
Fig. 5. ,, Section of part of sorus, paraphyses having taken

the place of sporangia ( x 250).

Fig. 6. ,, Section of thallus of specimen from Anguilla,

cryptostoma and paraphyses sunk in a
slight depression ( x 250).

PLATE XV.

Fig. i. H. sinuosus. Surface view of epidermis of young plant to show first

appearance of cryptostoma ( x 250).
Fig. 2. Hydroclathrns cancellatus. Natural size.
Fig. 3. ,, Section of thallus ( x 250).

Fig. 4. Cryptostoma ( x 250).

X.

ON THE CRYPTOSTOMATA OF ADENO-
CYSTIS, ALARIA AND SACCORHIZA.

THE bodies called ' fasergriibchen,' ' cryptostomata,' ' cryptes piliferes,"
' sterile,' ' neutral,' and ' vegetative conceptacles ' which occur somewhat
capriciously among the Phceophycea have been the subject of so much
speculation that any new attempt to explain their nature and significance
must be made on fresh evidence to have any claim on attention. While
submitting such evidence, I do not now propose to set forth any conclusive
argument, but to arrange old and new facts in such a fashion as to suggest
a way out of the difficulties.

The development of cryptostomata in the Fncacece has been lucidly
described, first, by Prof. Bower* in his paper on the development
of the conceptacle in Fucacetz. Valiante-f- and Oltmanns + have both
extended his observations ; and I find in a paper by Miss Barton
on Turbinaria an additional contribution to the subject. Prof. Bower
regards them as incomplete sexual conceptacles, a view which has
the advantage of definiteness, and one it would be hard to controvert ;
Oltmanns thinks that the fertile conceptacles are cryptostomata which
have in time come to bear organs of reproduction ; while Miss Barton puts

* Quart. Journ. After. Sci., xx.

t Fauna und Flora des Golfes v. NeapeL Leipzig, 1883.

\ Bibliotheca Botanica, heft 14.

Trans. Linn. Soc., vol. iii., part 5.

59

60 Adenocystis, Alaria and Saccorkiza.

forward a third view, which is just as likely to be true as the others, and
certainly is very ingenious. She says: ' My own view of the matter is that
the two forms of conceptacle are of equal antiquity, and were a later
development in the ancestors of the Fucacea than the reproductive organs ;
therefore I consider neither form a development of the other, and the fact
that one conceptacle contains reproductive organs, the other nothing but
paraphyses, is an interesting point, but does not bear on the phylogenetic
history of the conceptacles themselves.' I know so little about the
ancestors of the Fucacece, that I must be content with a respectful attitude
towards this statement.

As Prof. Bower has pointed out, the development of both conceptacles
and cryptostomata is the same. To quote his words, it ' is preceded by
the decay of one or more cells which occupy a central position with regard

to the changes which follow The cell or cells which decay are in

all cases members of a linear series.' These words appear to me to be the
true guide of those who investigate the development of such bodies.

When one comes, however, to compare the process as it takes place
in Splachnidium the process of making a conceptacular body for the
development, not of oogonia and antheridia, but of zoosporangia we
find a slight but noteworthy modification. The process is so nearly allied,
and the result achieved so similar, as to furnish material for direct com-
parison. Prof. Bower (loc. cit., p. 47) acknowledges the impossibility of
finding a parallel instance while speaking of the destruction of initial cells.
There is in Splachnidium not a parallel instance, but a closely comparable
one. The peculiarly modified initial cell of Splachnidium is homologous
with the initial cell of the Fitcacece but it persists. ' The epidermal cells '
(page 5, ante) ' lying around it undergo division, and the neighbouring
cortical cells increase in size. These causes combine to place the initial
cell in a depression (plate II., fig. 7). Hairs arise from the youngest
epidermal cells, while others which were formed earlier surround the
mouth.' In Fucus, it will be remembered that hairs are not produced
until the formation of a young conceptacle with an ostiole, while the
initial cell has decayed. The hairs in the conceptacle of Splachnidium
are long, septate, and unbranched, increasing in length by successive
divisions at the base, and giving the young conceptacle the appearance
of the cryptostomata of the Fucacea. The sporangia are developed later.
We have in fact in this type a sorus of sporangia and paraphyses com-
parable with those of Laminaria, rolled up and definitely limited in a
conceptacular body, produced, not exactly in the same manner as the
Fucaceous conceptacle, but after another manner directly comparable
with it.

AdenocystiS) Alaria and Saccorhiza. 61

The study of the structure and development of bodies of this character
among the Phceophycea is, therefore, one of interest, so far as such may
throw light on the origin of conccptacles. In connexion with the above
comparison of the Laminarian with the SplacJmidian sorus, it will be
remembered with interest that cryptostomata occur in the Laminariaceae.
Through the kindness of Mr. George Brebner, whom I asked to look out
for them at Cumbrae, I have been so fortunate as to obtain good examples
of their occurrence in Saccorhisa biilbosa and Alaria esculenta. Alaria has
been frequently described as possessing cryptostomata. Greville* definitely
refers to them as ' minute pores from which issue minute tufts of filaments'
and Prof. Bower (loc. cit., p. 36) refers to them also in passing. I have not
placed myself in antagonism to these authorities on this minute point,
without arming myself with certainty. On plate XVI., figs. 4, 5, and 6, there
will be seen a surface view of the Alaria cryptostomata, a transverse
section of the frond showing them in a very early stage, and another in an
advanced stage. It will be seen that so far no pits or conceptacles are
formed, and in a very old Alaria frond which I have examined there was
at most a depression one cell deep. The Alaria cryptostomata, if I may
call them so, are tufts of hairs with basal growth, the cell at the apex of
the hair being first cut off from the epidermal layer.

On plate XVI., fig. 7, I have illustrated the mature cryptostoma of
Saccorhiza bulbosa. So far as I am aware, the mature stage of these bodies
in Saccorhiza has not been figured before, and the reader who compares
this figure with that of the young conceptacle of SplacJinidimn (plate III.,
fig. 3), while it yet bears hairs only, cannot fail to be struck by the resem-
blance in all respects except the persistent initial cell of Splaclinidiuin .
We have here, then, a body which has every claim to rank as a cryptostoma.
Mr. Setchell,* in his admirable paper on the life-history of Saccorhiza
dennatodea has figured (loc. cit., plate II., fig. 22) an early stage in the
development of the cryptostomata of 3". dennatodea^ which bears the most
close resemblance possible to this early stage of Alaria. He says (p. 202):
' The first indication that a cryptostoma is forming is the appearance of a
shallow, saucer-shaped depression. This will be seen to be due to the fact
that the cells of the limiting layer in the region of this depression cease to
divide as actively in a tangential direction as they have been doing, or as
actively as their neighbours are doing. Furthermore, the cells of the outer
cortex immediately below this point do not increase in size either so rapidly
or to so great an extent as their neighbours do,' &c. This is so like what
happens in Splachnidium (without the persistent initial cell) as to be

t

* Alg. Brit., p. 25. t Proc. Amer. Acad. of Arts and Sciences, vol. xxvi.

62 Adenocystis, Alaria and Saccorhiza.

specially noteworthy. His description of the old cryptostomata, of which he
does not give a figure, as saucer or shallow bowl-shaped depressions, with
a large number of hairs projecting from the bottom, and with a rim pro-
jecting in over the edge of the depression, is sufficiently like my figure of
those of 5. bulbosa (plate XVI., fig. 7).

It appears to me to be a very probable explanation of the arrested
development of the cryptostomata in Alaria (if I may put it in that way)
that the frond of Alaria remains thin and easily lacerated so thin as to
give no room for the development of pits on either side and I am
confirmed in this view by Mr. Setchell's observation (loc. tit.} that ' the
cryptostomata are present on the one layered portion of the blade (of
Saccorhiza dermatodca] as clusters of hairs upon the flat surface.'

Kjellman,* in his paper on the forms of Adenocystis, has figured the
cryptostoma of A. Lessonii Hook, et Harv. Mr. R. M. Laingf has also
figured this body, and mentions and illustrates ' oval sacs containing
zoogonidia' in the cryptostoma of Adenocystis. These sacs do not appear
to belong to the species of Streblonema, which commonly invests the
cryptostomata of this plant, and their presence may or may not be
accidental. This plant has so many points of structure in common with
Splachnidium that an occurrence of such sporangia would possess a par-
ticular interest. On examining the original material of Adenocystis
collected during the Antarctic voyage, and also some New Zealand
specimens, kindly lent me by Mr. Harvey Gibson, I was not satisfied with
the accuracy of either Kjellman's or Laing's figures, and have therefore
given another rendering of this structure (plate XVI., fig. 2), from which it
will be seen that it bears a resemblance, both to the cryptostoma of its
ally, Saccorhiza, and to those of Hydroclathrus (plate XIV.), illustrated
by Miss Mitchell. Though I have examined a very young plant, I was
unable to secure sufficiently early stages to ascertain the development of
the cryptostoma. The result, however, is sufficiently remarkable. These
cryptostomata occur in the middle of sori of nniloatlar sporangia, and the
paraphyses and sporangia will be seen in the figure (plate XVI., fig. 2) on
the very border of the depression. The sporangia and paraphyses are truly
Laminarian, as would be expected. In the other Laininaricc examined,
the sorus occurs separately from the cryptostomata, but, in this form, in the
middle of the sporangia and paraphyses, as if in a nascent effort (or a
dying one, as the case may be), to form a conceptacle like that of Splac/t-
nidimn. It will be remembered that a young conceptacle of Splachnidium
bears hairs only at first and sporangia later on.

* Bihang till K. Svenska Vet. Akad. Handlingar, bd. xv., afd. Hi., No. I.
t Trans. New Zealand Institute, vol. xviii., p. 306, plate X., fig ?.

Adenocystis, Alarm and Saccorhiza. 63

In Hydroclathrus (plate Xiv.) the cryptostomata occur also among the
sporangia, which, in its case, are, of course, plurilocular.

In speaking of the conceptacular hairs of the Fucacea, Prof. Bower
(loc. cit., p. 47, foot-note), in quoting Reinke on the ' sprossfaden ' of the
Dictyotacece, which are the precursors of the reproductive cells (conf. SplacJi-
nidiiun), makes what I venture to think a prophetic remark of great
weight, ' How far these may be compared with the initial cell or hair of the
Fucacea it remains for closer observation to decide.' I anticipate with a
measure of confidence, so far justified by facts, that this remark will
eventually be borne out in many ways not then foreseen, otherwise than
dimly, by its author.

It has been pointed out (conf. Falkenberg in Schenk's Handbuch der
Botaniky vol. ii., p. 227), that in Scytosiphon single epidermal cells within a
sorus remain sterile and grow out in the form of club-shaped unicellular
paraphyses ; and in Asperococcus and other allied forms, similar cells
grow out in the form of long, branching filaments, of which the upper cells
are very long, while the basal ones remain short, and probably act as the
growing point of the hair. These hairs resemble those in the conceptacles
and cryptostomata of the Fucacea, SplacJmidium, SaccorJdza arid Adenocystis,
and those in the sorus of Cutleriacece and Dictyotacece. Interest is certainly
quickened when we add to these cases that of Hydrodatlirus, with its
cryptostomata situated among the sporangia.

A comparison of the Fucaceous conceptacle and cryptostoma, the
Splachnidian conceptacle with its persistent initial cell and the formation
of its hairs yielding place to sporangia, the development of the Adenocystis
cryptostoma in the heart of its sorus, the other Laminarian cryptostomata
(SaccorJiiza and Alaria] apart from the sorus, the cryptostoma of
Hydroclathrus among its plurilocular sporangia, and finally the cases of
the hairs in Asperococcus and the Cutleriacc(Z and Dictyotacecs a com-
parison of these cases, and of the evidence plainly furnished by them,
points very significantly to a possible origin of cryptostomata. I anticipate,
from further research into the development of these bodies, evidence that
may enable us to dispense with ' the ancestors of the Fncacecs' of which,
however, I would speak with respect. I am aware that the hairs of the
cryptostomata are regarded by many as adapted to absorptive and other
nutritive functions. This may or may not be ; there is no proof of the
matter, and probably more reasons to be cited against than in favour of
such an opinion but their function has nothing directly to do with the
points in question, which are purely morphological. Moreover, in seeking
for a solution of the difficulties surrounding the origin of these bodies, I
do not intend to make the mistake of looking at cryptostomata only.

GEORGE MURRAY.

64 Adenocystis, Alaria and Saccorhiza.

EXPLANATION OF PLATE XVI.

Fig. i. Adenocystis Lesson ii. Nat. size.

Fig. 2. Cryptostoma ( x 375).

Fig. 3. Alaria esculenta. Young state, bearing so-called cryptostomata. Nat. size.

Fig. 4. Surface view of so-called cryptostomata ( x 150).

Fig. 5. Transverse section of frond, showing very early stage

of same ( x 375).

Fig. 6. ,, Advanced stage of same ( x 375).

Fig. 7. Saccorhiza bulbosa. Mature cryptostoma ( x 375).

XL

A COMPARISON OF THE MARINE FLORAS

OF THE W 'ARM ATLANTIC, INDIAN
OCEAN, AND THE CAPE OF GOOD HOPE*

IN delimiting the above regions I have been guided by what may fairly
be taken to be their natural boundaries. The warm Atlantic is the tropical
Atlantic, with a slight northward extension, to include Florida, the
Bahamas, and Bermuda in the track of the Gulf Stream, and also Madeira
and the Canary Islands, washed by that branch of the same stream which
trends off backward to the south, the north equatorial current. I have not
included the Azores, since they are not sufficiently under this influence
and their marine flora, so far as we know it, appears to be more akin to
that of the north temperate Atlantic. On its southern boundary on the
African coast the Cape region is permitted to come slightly within the
tropics, so far as Walfisch Bay, on account of this coast being swept by a
cold current from the south, bringing with it up to this point at all events
such temperate forms as Laminaria, recently recorded from that place.
The Indian Ocean similarly is the tropical Indian Ocean, but including
the whole of the Red Sea, and extending to the south slightly outside the
tropics down the coast of Africa, and including the whole of Madagascar.
I am justified in this by the course of the warm Mozambique current. I
do not include on the east Sumatra, which appears to belong to another

* A paper on this subject was read by me at the British Association at Edinburgh,
1892, and a table appears in the Report. The Cape totals in that table were much smaller,
since the present table embodies the results obtained since then by Miss Barton, alluded
to in this paper.

65

66 Marine Floras of the Warm Atlantic,

region, though I have included a few forms from the Andaman Islands and
Mergui. The Cape of Good Hope region has already been indirectly
described, and, as has been said, extends for the reasons given slightly into
the tropics on the west coast, and recedes slightly from that boundary on
the east coast.

The accompanying table speaks for itself so far as the results of the
comparison go, but a few remarks on what I take to be the meaning of
the figures may be thought not too venturesome. If we look first at the
last line of the table where the aggregates are set forth, it will be seen that
the warm Atlantic has the largest recorded flora, viz., 859 species in 162
genera. I may explain that, out of this total, no less than 788 species in
150 genera occur in the West India region, and that the rest of the warm
Atlantic furnishes only 71 species in 12 genera not occurring in the West
Indies out of a much smaller total flora. Allowing for the undoubted
fact that a large number of West Indian species are bad species, there
still remains a large balance in its favour. It has been better examined
than any other part of the warm Atlantic, but still we may attribute
this preponderance mostly to the favourable natural conditions, princi-
pally the coral formation of large portions of its island shores. On
the coast of Africa there is not only no coral, but league after league
of muddy shore, making a marine desert so far as Algae are concerned.
The Indian Ocean comes next, with 514 species in 139 genera. It
possesses an enormous coast line, to a considerable extent favourable
to the growth of Algae (though including long desert stretches) ; but
the bulk of the records are from Ceylon, Mauritius, and the Red Sea,
while a very large proportion of the region is unexamined. As in

the West Indies, there is also here a considerable proportion of bad
species, principally Sargassa, from the Red Sea. From the Cape we
have 429 species in 141 genera. This remarkable total, from so short
a coast line, is obtained from Miss Barton's list in the Journal of
Botany, 1893. The flora previously recorded in books amounted only
t to 242 species in 99 genera, and this addition to its flora has resulted
from her examination of the British Museum Herbarium, and her naming
of the admirable collection made by Mr. Boodle, and also those made by
Mr. Scott Elliot and Mr. Tyson. The most noteworthy observation on
these aggregates is the proportion of species to genera. In the warm
Atlantic the genus averages well over 5 species ; in the Indian Ocean the
proportion is nearer 4 than 3 species to the genus ; while at the Cape it is
almost exactly 3. This is instructive when we remember, as I have
elsewhere pointed out,* that while the Arctic Algae average slightly more

* Trans. Biol. Soc. Liverpool, vol. v., p. 177.

Indian Ocean, and the Cape of Good Hope. 67

than 2 species only to the genus, the West Indies and Australia average
rather more than 5 and less than 5 respectively. I estimate that the
north temperate Atlantic yields an average of about 4^ species to the
genus, and the difference between this and 3 species per genus found
at the Cape is to be attributed primarily to the short coast line of the
Cape, and in a less degree to its Algae being less known. The calculation
of such averages and proportions appears to me to be justified only when
applied to the whole flora, and becomes more dangerous and apt to mislead
when applied to portions of it, since particular groups in all the floras have
been subjected to unequal treatment by collectors and describers, and we
may perhaps trust to these personal errors neutralising each other when
the complete totals are compared.

The warm Atlantic and Cape have 85 genera and 114 species in
common, while the Indian Ocean and Cape have 86 genera and 89 species
in common. That the number of genera in common should be so nearly
exactly similar is interesting, and to discover whether they are the same
genera in many cases it is only necessary to turn to the last table, where
the Algae common to all three regions are given to find that 72 genera are
are common to all three. Some years ago I hazarded the speculation
that, while the genera of the tropical Atlantic and those of the Indian
Ocean were largely the same, the species were, in a high proportion,
different.* We can now see that they have no less than 103 genera in
common out of a total of 139 occurring in the Indian Ocean and 162
in the warm Atlantic. They have certainly more species in common,
viz., 173, but these must be considered relatively to the two totals of
514 in the Indian Ocean and 859 in the warm Atlantic, when my expecta-
tion will appear to be fairly borne out. Nevertheless, I confess to having
anticipated an even greater diversity of species. That the absolute
number of genera occurring at the Cape should be by two greater than
those of the Indian Ocean completely puzzles me. I cannot fully
account for it on any theory. While the number of species in common
between any two of the floras is greater than the number of genera
(though in one case only three more), the number of species, as might be
expected, in common to all three vi/., 59 is less than the genera viz., 72.
Again I should have expected to find relatively fewer species in common.

When one comes to analyse these totals, the process must be carried
on in a more guarded fashion. One expects, as shown above, to find
fewer species to the genus at the Cape than in the tropical floras, but one
hardly expects to find that the genera of Floridece at the Cape are by five

* Catalogue of Marine Alga of the West Indian Region.

68 Marine Floras of the Warm Atlantic,

more numerous than in the warm Atlantic, and by 15 more than in the
Indian Ocean. There no less than 95 genera of P"loridece at the Cape, with
295 species, while the 90 of the warm Atlantic contain nearly 200 more
species ! Matters are much the same in the case of the Phceophycece, and
we have to come to the CJdorophycece to redress the balance in the case of
the warm Atlantic. They just fail to bring it level in the case of the
Indian Ocean. It has been remarked above that the genera which the
two tropical floras have in common with the Cape are almost identical in
number. The analysis shows that the figures are very steady, viz., 58 each
of FloridecB, 14 and 15 of PJuzophycea?, 11 each of Chlorophycece, and 2 each
of Protophycecz. The table shows the tropical character of such a group as
the Siphonece very markedly. There are 99 species in 23 genera in the warm
Atlantic, 72 species in 16 genera in the Indian Ocean, and only 20 species
in 7 genera at the Cape. It is interesting to observe that the whole of
the 16 genera of Siphonece in the Indian Ocean are represented in the
warm Atlantic. It has no peculiar generic type of its own in this tropical
group. While the genera of this tropical order are thus practically
identical, the species are in a very high proportion different. Only 29 are
possessed in common out of the two totals of 99 and 72. In the com-
parison of the two tropical floras there is the coincidence that the
genera and species of Siphone<z agree exactly in numbers, viz., 16 and 29,
with the total of all the PliceopJiycece a thing without significance, how-
ever. But I need not go farther into the details of the table.

The interest that is attached to the above comparison is mainly this.
We have here two tropical marine floras cut off from each other by a
permanent continental area, and communicating only via the Cape.
That these floras have been periodically mingled at the epochs of warmer
climate at the Cape seems a reasonable conclusion with regard to a group
of such antiquity as the Algae, and the proportions of species in common
and genera in common between the different regions, and among all
three may have a significance in this respect to students of distribution
(cf. the totals of Sipkone<z> a peculiarly tropical order). I have
elsewhere* commented on the fact that, ' while in the Arctic and
Australian regions the Ph(zophyec<z far outnumber the ChloropJiycece, in
the tropical West Indian flora the proportion is very markedly reversed,
and the green Algae outnumber the olive-brown. One is tempted to put
this down to the strong illumination of the tropical sea, but another
reason is to be found in the fact that a number of the Antilles richest
as regards Algae are subject to irruptions of fresh and brackish water

* Trans. Bid. Soc. Liverpool, vol. v., p. 178.

Indian Ocean, and the Cape of Good Hope. 69

from the Orinoco floods a condition that would operate in the same
direction.' We can now check this speculation by a comparison with
the figures for the Indian Ocean, mainly derived from such localities as
the Red Sea, Ceylon, Mauritius, &c., in no case affected by the question
of fresh-water floods. The figures for the Indian Ocean are very nearly
the same for both groups 24 genera and 117 species of PJiceopJiycece, and
26 genera and 121 species of Chlorophycea thus showing indirectly that
the irruptions of fresh water are, in all probability, potent in the case
of the West Indian Algae. One is much struck by the strength of
illumination of the bottom in a shallow coral sea, but the filtering action
by sea water of the rays of light, and the interception first of those
rays that are most efficient in the work of assimilation conditions
modifying the pigments of Algae are the same in all seas.* The
practically tideless character of the Antilles would also make for a
preponderance of green over olive-brown forms.

GEORGE MURRAY.

* Recent research on other pigments by Prof. Marshall Ward makes it appear to
me more probable that, in the case of the marine Algae, the pigments are rather shields
against the excess of blue rays than adaptations to heighten the succeptibility of
chlorophyll to the diminished supply of the others.

K

ALG^E.

TOTALS.

ALG.E COMMON TO

Warm
Atlantic.

Cape of
Good
Hope.

Indian
Ocean.

Warm

Atlantic
and Cape.

Indian
Ocean
and Cape.

Warm
Atlantic
and Indian
Ocean.

Warm
Atlantic,
Indian
Ocean,
and Cape.

Floridcce.

Gen.

Spec.

Gen.

Spec.

Gen.

Spec.

Gen.

Spec.

Gen.

Spec.

Gen.

Spec.

Gen.

Spec.

Ceramics;

6
9

10

I

I

-)

6

2

2

4

3
6

2

5

2

4
i

2
I
14

7

47
30
28
6

i

7

22

3

6

63

8

4i

2

17
22
12
2

25
I

9 6

43

1 1
6

7
i
i

i

7
i

i

2

7
6

3

2

5

2

4
i

2
I
15

9

33
1 1

42

4

i

2

18

4
i
6
13
17
5
4
13
8

5
i

15
i

43
48

7

6
i

2

5
i

2

6

2

6

2
o

2

3
I

2
I

18
6

21

16
IS

3

o

7

10
2

6

22

3

24
2
12
12

6
i
16

o

48
24

4
3
7
i
o

i

4
I
I

2
I

2

3
i

3

2

3
i

i
i

9

7

7
i
8

2

o

o

2
I

2
2
2

4
I

5
4

2
O

8
o

7
16

8

3
6
i

o
i

4
i

2
2
I
2
2
2
I

3
i

i
i

10

6

7

2

8

2
O

I
2
O

I

2
O
O

I

5

2

I

6
o

3

10

5

6
i

3
i

o

2
2
2

4
I

2
I
2

2
I
12

5

IO

4
6

3

o
3

2

I

7
i

10

i
3
9

2

8
o

12
12

5
3
6
i

o

i

2
I

2

I
I
2
I
2
I
2

I
I
O

8

5

5

i

4

2
O

o

I
o

o

2
O
O

I
3

2
I

5

i
6

Cryptonemiacese ...
Gigartineae

Spyridieae

Dumontiaceae

Areschougieae

Champieae

Rhodymeniaceae ...
Squamarieas

Hildenbrandtiaceas
Porphyraceae

Sphaerococcoideae . . .
Delesserieoe

Helminthocladiaceae
Chaetangieas

Gelidieas ..

Hypneaceae

Solierieae

Wrangelieae

Chondrieae

Lomentarieae

Rhodomeleas ....

Corallineas

Total

90

482

95

295

80

255

58

73

58

53 59

94

46

34

Phceophyceee.

Fucacere

4
7
o

i

2

6
i
i

3
o
i

38

45
o

15

4

9

i

3
9

i

7
4
i
i

o

4
i

o

i

5

i

21
21
I

6
6

o

o
4
9
i

4
7
o

2
I

5
i
i

2
O
I

59
23
o

12

5
7

i

2

7
o
i

4
4
o
i
i

2
O
O
I

I

7

9
o
o

i

2
O
O

2

3

4
o
i

i

3

i
o
i
o

i

8

5
o

i

->

3
i
o
i

o

3
4
o
i
i

3

i

2
O

I

10

9
o

2

I
i

o

3

o

3
4
o
i

i

o

I

I

4

5
o

i

i

2
O

I

o

Dictvotaceae

Splachnidiacea;

Ectocarpaceas

Sphacelariaceae

Chordariaceas

Punctariaceae

Arthrocladiaceae ...
Sporochnaceas

Laminarieae

Ralfsieas

Total

26

125

28

76

24

117

14

23

15

21

16

29

13

14

Chloropliycea.

Siphoneae

23

4
5

99
80

20

7
3
4

20

23
II

16
8

2

72
39

10

6
3

2

7

5
6

6
3

2

7
3

5

16

4

2

29
13

4

6
3

2

6

2

3

Conferveae

Ulveae

Total .

32

199

14

54

26

121

I I

18

I I

15

22

46

I I

1 1

Protophycece.

M

51

4

4

9

21

2

o

2

5

4

2

o

Aggregate .

162

859

141

429

139

5M

85

114

86

89

103

173

72

59

ThydVEemPart IT.

"Plate XIV.

'- " . ' . '

' Isy3sl eb lith .

Han.Tn.arb imp

HYDRO CLATHRUS SINUOSUS

Phyc.TyEem.Parb II .

Tlite

mm i : -

-;>-,. l.StaVi . A! ' -, * '

-

y$8v\'4r " v \

' ' $ ^ ' \ ^ ** -" NE^ y * = ^ ' (- ?'

^r -- iCkS^f^^S^^KfOT
' A ; %5^4|^Wi i)^/vM ft v

v ^| &yf)>? i?^Wm& .>>|CY?

\ v> &

'Berjea-u.&HigTaleyHfil.ebliblL. M.O.TVT.del .

HYD"ROCLATHRUS SITNJTJOSUS

Harvhart/ imp.
CANCEL'LATUS Sory,(2-4-)

PKyc.TVlem.Earfc II

Plibe XVI.

*iwiss
$$$$%

M!IV&U52

SESS

Berjeau.&,Higliley.deletlillv. Hanharb imp.

ADENOCYSTIS LESSONII Hook.et, ffoar.ft-2) ALARIA ESCULENTA GnY.(3-6)
SACCOKHIZA BULBOSA De la. Pyl . (7)

XII.

A NEW PART OF PACHYTHECA.

A PAPER by Sir Joseph Hooker, in the Annals of Botany (vol. iii., 1889),
gives the history of this remarkable and very puzzling fossil from the
Devonian rocks, while a study of its structure by Mr. Barber will be found
in two papers, one in the volume cited, the other in vol. v. A note on the
subject by Mr. Thiselton-Dyer appeared also in the latter volume. The
most ample discussion of the subject has not left us with any clear idea of
what PacJiytJieca may be, and this is certainly the fault of the inadequate
nature of the evidence it presents rather than any lack of ingenuity on the
part of those who have sought to interpret it. The discovery of an addition
to this evidence in the form of a new part of the fossil therefore raised my
hopes greatly of finding some clue to its true character. I am bound to
confess at the outset that, in my understanding of the matter, the discovery
of this new part only adds to the perplexities of the subject, and deepens
the gloom that already surrounds PacJiytJieca. It is, however, my duty to
make this new evidence known, and I do it in this place, not out of
a conviction that PacJiytJieca is an Alga, but because, if it be a plant at all,
it is most probably an Alga, and in any case there is no evidence to
justify us in placing it elsewhere.

Mr. John Storrie of Cardiff, who has done so much to further the study
of PacJiytJieca, sent to Mr. Carruthers the cup-like structure, with a PacJiy-
tJieca in it, which forms the new part about to be described. He says, in the
letter with which it was forwarded, that the specimen was quite loose in its
cavity, and that it ' was held in the cup like an acorn ; or a growth of some
arillus-like structure took place round it externally.' Mr. Storrie cut two
sections of the sphere, and he describes them in his letter as follows : ' The
section across the ends of the outer tubes I made from a piece that I thought
looked slightly different from the remainder of the outer skin, and, as I
think, had been connected with the cup at one time. Part of it still

I L

72 A New Part of Pachytkeca.

remains in the original state Please notice the shiny black external

coating still remaining on one part of the Pachytheca* (plate XVIII., fig. i).
Mr. Storrie thought that this shiny outer layer would repay cutting
through. I may say at once that the sections cut for me of the remainder
of the Packytheca-sphere have revealed nothing new whatever, not even
the section which passed through the outer coat just mentioned (plate
XVIIL, fig. 3). The sections (plate XVIII., figs. 2 and 3) have been useful
only in determining beyond question the fact that it is a true Pacliythcca
which rested within the cup.

The cup surrounded one half of the sphere ' like an acorn,' as Mr. Storrie
says, or as an egg-cup holds an egg. On the outside, the cup is clothed
with hairs (plate XVII., figs. I, 3 and 4) on the lower portion, and is other-
wise smooth. The broken edge (plate XVII., fig. 2) shows that the cup
itself is chambered radially and these chambers appear to spring from
an axial body the top of which may be seen in the bottom of the
cup (plate XVII., figs. 2 and 3). My first idea was that this projecting
axis would be found to correspond with the dimple found in many
specimens of PacJiytlieca, but this portion was not among the parts sent by
Mr. Storrie ; probably it was the part he used in making his two sections.
However likely such a view may be, I have no evidence of it. Unfortu-
nately the structure of the interior of the cup, as seen in vertical section, has
not been preserved, and a microscopical examination of it reveals only purely
mineral structure. We know, however, from examining its broken edge,
that the cup was radially chambered, like the sporangial rays of Acetabu-
laria, and in fact the cup itself (without the Pac/tyt/ieca-sphere) is not unlike
the remains of some organism like Acetabularia, and the resemblance is
heightened by the axial body in the centre. Suggestions of this kind are,
however, plainly of no use, since in none of the verticillate SipJioncce is
there anything to correspond with the sphere of Pachytkeca. There is, in
fact, no alga, nor any other plant known to me, which lends itself for
comparison with Pacliytlieca.

Since the Pachyt/icc a- sphere was loose in the cup, there must have been
a tissue, now perished, between the ends of the radiating filaments of the
sphere and the chambers of the cup. My only hope of gaining farther
light is in the discovery of specimens in which this tissue has been preserved,
and more of the cup and axial portion or that perhaps, in the mean
time, some zoologist may recognise it, and relieve botanical literature
of this standing puzzle.

I am much indebted to Mr. John Storrie, who found this unique
specimen and to Mr. Carruthers for the opportunity he kindly gave me of
making it known. GEORGE MURRAY.

A New Part of Pachytheca. 73

EXPLANATION OF PLATES.
PLATE XVII.

Fig. i. Side view of cup of Pachytheca ( x 14).
Fig. 2. The same from above ( x 14).
Fig. 3. The same in vertical section ( x 25).
Fig. 4. Hairs from the lower part of outside of cup ( x 61).
Fig. 5. a. The cup natural size, 5. b. The cup natural size, showing how the

rested in it.

PLATE XVIII.

Fig. i. Patfii't/ieca-sphere (from cup) showing portion of shiny outer layer

(x 14).

Fig. 2. Section through portion of same ( x 55).
Fig. 3. Section through another (marginal) portion, including shiny outer

layer ( x 86).

Fig. 4. Tangential section through another PatAytfoca-sphere ( x 125).
Fig 5. Portion of same ( x 375).
Fig. 6. Diagrammatic section through sphere and cup.

NOTE. Figs. 4 and 5 (plate xvm.) are from a section sent me by Mr. Storrie.
They were cut from a Pachytheca obtained from the same bed as the one described.
The section is tangential near the periphery, but a portion of the central core is
shown. Fig. 5 shows both central and peripheral tissues, and the arrangement of
the peripheral tubes shows inconsistencies with any known plant structure, a con-
dition which may or may not be due to their state of preservation. For example,
the tubes cut transversely sometimes appear between those cut obliquely, and
sometimes within them, even under the most careful focussing.

XIII.

CALCAREOUS PEBBLES FORMED BY

MR. E. GROVE has been good enough to give me several large calcareous
pebbles collected by Dr. J. W. Velie, Secretary of the Chicago Academy
of Sciences, who found them in a pond in Michigan. Mr. B. W. Thomas,
of Chicago, had sent them to Mr. Grove with a request for his opinion as to
their nature, and Mr. Grove forwarded them to me. I am indebted to Mr.
Thomas for a further supply. The pebbles were found resting free on a
sandy bottom under three to eight feet of clear water in a pond separated
from Lake Michigan by a sand bar. The specimens vary in size from
an inch to three inches and a half in diameter, are hollow in the interior
and show a stratified or concentric zoned structure. On decalcifying a
portion of one, I found that it was composed of a densely interwoven
mass of filaments evidently not all of the same nature, but the pre-
dominating kind was clearly a species of Sckizotknx, while mixed with
it there were other forms, notably filaments of Stigonema and Dichothrix.
I examined portions here and there from a number of the pebbles, and
in all cases found the strong stout sheaths and filaments of Schizothrix
composing by far the greater part of the decalcified mass. At the surface
the Schizothrix filaments had been clearly alive when the pebbles were
gathered, while nearer the centre older sheaths only were found. It was
apparent that the SchizotJirix died off internally while fresh crops were
produced on the surface adding to the growth of the pebble. On com-
paring this Schizothrix with the figures and description in M. Gomont's
admirable Monographic des Oscillarices, I found a difficulty in identifying
it, but none in placing it near certain other species. I accordingly con-
sulted him, and he had no hesitation in identifying Schizothrix fasciculata

74

Calcareous Pebbles formed by Algce. 75

Gomont in the mass I sent him. At the same time he says, ' The interior
of the calcareous mass is formed of entangled filaments ; they appear to
belong to a Schisotkrix t \xo!i which?' It was precisely these sheaths that
had puzzled me, and I had been inclined to regard them as those of an
unknown ScJiizotJirix. They do not agree very closely with the figures
given of 5. fasciculata, but M. Gomont finally says, ' I do not think
one can make anything very distinctly out of this specimen, except
S. fasciculata, which undoubtedly occurs in abundance.' I have figured
the specimen selected by M. Gomont and others on plate XIX., figs. 5a
and b, and have given an indication (fig. 4) of the density with which the
filaments are interwoven.

S. fascicidata is known from the countries of central Europe and forms
small, stony cushion-like calcareous masses. The filaments are always en-
tangled where they are incrusted, but on the surface, where they are more
or less free, there is very little entanglement and the filaments are almost
straight and parallel. The ordinary incrustation formed by this species is
said to be zoned in section, and its surface is generally mammillate. It
is constantly found growing in association with other Algae of the same
group, and there is a considerable colony of them in the present instance.
The most striking feature, however, in this respect is the large number
and variety of diatoms present in the mass. Mr. E. Grove has made an
exhaustive examination of them, and has most kindly furnished me with
the appended list of them, which extends to about 90 species and varieties.

A matter of particular interest is the massive occurrence of car-
bonate of lime in association with a kind of organism that is commonly
so incrusted only in hot springs, and to some extent it raises the question
of the influence of temperature on the deposition of carbonate of lime.
Murray and Irvine have made a series of most interesting experiments
leading to the conclusion stated by Murray (Challenger Reports
Summary of Results, p. 1456) as follows :

' The very slight development of carbonate of lime shells and other
carbonate of lime structures in the cold waters of the polar regions are
instructive when recalled in connexion with the massive coral reefs
constructed in the polar regions in Palaeozoic and even later geological
times. The waters of the ancient oceans must have had a temperature of
65 or 70 F. at the poles, for it has been shown that the deposition of
carbonate of lime is due to the secretion of carbonate of ammonia, one of
the effete or waste products of marine animals, which, decomposing the
soluble sulphate of lime in sea-water, produces insoluble carbonate of lime
to form shells, its precipitation taking place with great difficulty and very
slowly in cold water, but easily and very rapidly within the organism in

76 Calcareous Pebbles formed by Alga.

water of a high temperature ; hence in the cold polar and deep sea waters
of our day no massive carbonate of lime shells or other structures are
secreted by organisms.' Although plainly to be otherwise accounted for,
it is interesting to observe that the coralline plants of the sea are much
more massively developed in the warmer seas than in the colder ones. In
fresh waters it would be interesting to know how far the incrustation of
CJiaracece, for example, is affected by temperature. Certainly, in the
particular case under discussion, the massive occurrence of carbonate of
lime suggests a hot-spring Alga rather than one living at the temperature
of a lake in Michigan. The incrustation of carbonate of lime so often
found on the surface of plants is, no doubt, often due to the evaporation of
water holding the salt in solution, which has been excreted by the plant ;
but, as Vines (Lectures on the Physiology of Plants, p. 22) suggests, ' in the
cases of submerged plants it is possible that the calcium carbonate may
be deposited on the surface in consequence of the absorption of carbonic
acid from the water by the plant.' Since a relatively high temperature
would encourage this operation of dissociation, it is exceedingly probable
that we have in such a process an explanation of this distribution of
carbonate of lime deposits according to temperature. Gomont (loc. cil.}
says of OscillariecB that, when they live in waters strongly charged with
lime, the precipitation of carbonate of lime is caused by the decom-
position of carbonic acid, and that this occurs only where the Algai do not
find carbonic acid in sufficient quantity to meet the needs of vegetation.
When it is abundant, as in certain mineral waters, the salt is precipitated
only in very small quantities.

In the Annals of Botany (vol. v., p. 225), Mr. Thiselton-Dyer refers to
the occurrence of round calcareous pebbles in profusion at the bottom of
Lough Belvedere, near Mullingar. He collected and examined these
pebbles, and was much struck with their geological possibilities. ' An
accumulation of them might easily form a rock, and the interpretation of its
structure would not be easy.' The pebbles in this case (fig. 6) are not
nearly so large as those from Michigan, and Mr. Thiselton-Dyer describes
them as 'of all sizes up to that of a filbert.' He has most kindly placed at
my service a number of these pebbles, and in his letter calls my attention to
the resemblance of ' the material to pisolite [= Pea grit of the Inferior
Oolite, not the pisolite limestone of Eocene Age] though, of course, the
structure is very different.' I have seen a specimen of this rock from the
Cotteswolds, and it so exactly resembles the appearance that would be
presented by a heap of these pebbles that I am in no way surprised at Mr.
Thisclton-Dyer's insisting on the geological interest of these bodies. Those
pebbles I have cut open are solid, though the substance is less dense at the

Calcareous Pebbles formed by Algcs. 77

centre, and they would form a more compact rock than the Michigan
pebbles. Mr. Dyer, in a short examination of the algal contents, determined
them to be a Rivularia with Scytonema, and Mr. Archer had also noted
other Algae casually interwoven in the mass. I have examined it with
much care and have compared it with the Michigan specimens, and the
predominant Alga is the same in both cases viz., Schizothrix fasciculata.
As in the Michigan pebbles, there are a number of other Algae casually
present notably a species of Dichothrix ( = Scytoneina) as observed by Mr.
Thiselton-Dyer, Nostoc cells, diatoms, &c. I have found no trace of the
CladopJiora, nor of anything resembling one, mentioned by Mr. Barber in
Annals of Botany (vol. iii., p. 144), and Mr. Thiselton-Dyer (loc. cit^ also
has seen nothing to confirm the statement.

It is much to be wished that some of our Irish botanists would give
an account of the growth of these bodies. They are in shallow water in
both cases, and may be a product of the high temperature of summer ;
but speculation on this point is useless without further knowledge. Mr.
E. Grove has kindly examined the Lough Belvedere pebbles for diatoms,
and his lists, 'allowing for the differences between the diatomaceous flora
of North America and Great Britain, show much resemblance between
the two gatherings.'

GEORGE MURRAY.

EXPLANATION OF PLATE XIX.

Fig. i. Pebbles from Michigan natural size.

Fig. 2.

Fig. 3. in section

Fig. 4. Interwoven filaments of Schizothrix fasciculata ( x 86), from Michigan

pebbles.

Fig. 5. a, b. Filaments of pebbles (a x 232 ; a 1 x 375 ; b x 232 ; & x 375).

Fig. 6. Pebbles from Lough Belvedere, Mullingar.

78 Calcareous Pebbles formed by Alga.

DIATOMACEOUS REMAINS OBSERVED IN PREPARA-
TIONS OF CALCAREOUS ALG^E FROM A POND

IN MICHIGAN.

Achnanthes delicatula, Grun. Scarce.
microcephala, Grun.

minutissima, K.

var. cryptocephala, Grun.

Achnanthidium flexellum, Breb.
Amphipleura pellucida, K. Rare.
Amphora ovalis var. affinis, K. Rare.

veneta, K. Frequent.

Asterionella formosa, Hass. Rare.
Caloneis alpestris, Cl.

fasciata, Cl. Rare.

obtusa, Cl. Rare. ( = Nav. Hebes, Ralfs.)
Cocconeis Placentula, E.

var. lineata, Grun. Rare.

Cyclotella antiqua, W. S.

comta var. radiosa, Grun.
Kiitzingiana, Chauvin.
Meneghiniana, K.
Cymatopleura elliptica W. S. (A twisted form, perhaps the C.

Cochlea, Brun.
Solea, W. S.

Cymbella amphicephala, Naeg.
angustata, W. S.

aspera, E. (C. gastroides, K.) Rare.
var. minor, K. Rare.

Cistula, Hempr.
,, delicatula, K.
Ehrenbergii, K. Rare.
hybrida, Grun. Rare.
leptoceros, K.
microcephala, Grun.
parva, W. S.

Calcareous Pebbles formed by Alga. 79

Cymbella pusilla, Grun.

(Encyonema) Cesatii, Grun.

,, gracilis Rab. var. ? (n. sp. teste Cleve).

Triangulum, E.

turgida, Greg. Rare.

ventricosa, K.

var. obtusa, Grun.

Denticula elegans, K.
Diploneis oculata Bre"b. Rare.

ovalis, Hantzsch. Rare.
Epithemia Argus, E. Frequent.
gibba, E.

turgida var. granulata, Grun.

Veliei, Thomas. Cleve notes this as E. Cistula, E. var.

proboscidea, Grun., which Grunow considers the same as
E. proboscidea Sm. This form is, however, much larger,
and has much coarser strife than E. proboscidea, either in
Grunow's or Smith's figures.
Eunotia Arcus, E. Rare.
praerupta, Grun.

var. curta, Grun. Frequent.

Fragilaria mutabilis, Grun.
Harrisonii, Grun.

Gomphonema acuminatum, E.

var. elongata, Grun.

intricatum, K.

lanceolatum, E.

olivaceum, E.

subtile, E.

Mastogloia Grevillii, W. S. A short, broad form.

Smithii, Thwaites var. amphicephala, Grun.

var. lacustris, Grun. Frequent.

Melosira crenulata var. tenuis, Grun.
Navicula bacilliformis, Grun.
brachysira, Breb.
cincta, K.

var. Heufleri, Grun.

lanceolata, K. Rare.
radiosa, K. Rare.
var. acuta, K.

radiosa, var. tenella, Breb.

M

go Calcareous Pebbles formed by Alga.

Navicula Rcinhardtii, Grun. Rare.
vulpina, K.
Zellensis, Grun. Frequent.
Nitzschia amphibia, Grun.
angustata, Grun.

var. acuta, Grun.

Denticula, Grun.
sinuata, var. Tabellaria, Grun.
Pinnularia major, K.

microstauron, E.

Krookii, Grun. Abundant (teste Cleve).
oblonga, K.

viridis var. commutata, Grun.
Pleurosigma attenuatum, W. S. Rare.
Stauronis Phcenicenteron, E.

anceps, var. amphicephala, K. Rare.
Stephanodiscus Astraea var. spinulosa, Grun.
minutula, Grun.

Synedra Chasei, Walker.
delicatissima, W. S.
obtusa, W. S.
Tabellaria fenestrata, K.

E. GROVE.

DIATOMACEOUS REMAINS OBSERVED IN WASHINGS OF

CALCAREOUS ALG^E FROM LOUGH BELVIDERE,

MULLINGAR, IRELAND.

Achnanthes Biasolettiana, Grun.

Clevei, Grun. Rare (teste Cleve).

delicatula, Grun. Rare.

lanceolata, Grun. Scarce.

,, microcephala, Grun.

minutissima, K.

var. cryptocephala, Grun.

Achnanthidium flexellum, Breb.

Calcareous Pebbles formed by Algce. 81

Amphora ovalis, K.

,, var. affinis, Grun.

Pediculus, Grun.

salina, W. S. Rare (teste Cleve).
Caloneis alpestris, Cl.

bacillaris, Cl.

fasciata, Cl. Rare.

Silicula, Cl.
Cocconeis Pediculus, E. Rare.

Placentula, E.

var. lineata, Grun.

Cyclotella comta, K.

Kiitzingiana, Chauvin.

Meneghiniana, K. (teste Cleve).
Cymatopleura Solea. Rare.
Cymbella amphicephala, Naeg.

affinis, K.

aspera, E. Scarce.

Cistula, E.

cuspidata, K. Scarce.

cymbiformis, E.

Ehrenbergii, K. Rare (teste Cleve).

helvetica, K.

,, var. Balatonis, Grun.

lanceolata, E. Scarce.

lata, Grun.

microcephala, Grun.

aequalis, Sm.

pusilla, Grun.

laevis, Naeg. Rare.

naviculiformis, Auersvv. Rare (teste Cleve).

sinuata, Grun. Rare.

parva, Sm.

(Encyonema) turgida, Greg. Rare.

,, ventricosa, Cl.

var. ovata, Cl.

Denticula tenuis, K.
Diatoma lineare, Sm.

tenue, K.
Diploneis Boldtiana, Cl. Rare.

oculata, Breb.

82 Calcareous Pebbles formed by Algce.

Diploneis ovalis, Hantsch.
Epithemia Argus, K. Scarce.

gibba, K.
Eunotia Arcus, E.
Fragilaria Harrisonii, Grun.

capucina, Desm. Rare.
intermedia, Grun.
mutabilis, Sm.
Gomphonema acuminatum, E.

var. Brebissonii, CL Rare.

angustatum K. var. Sarcophagus, Cl. Scarce.

constrictum, E. Scarce.

geminatum, Ag. Rare.

gracile, E. Scarce.

intricatum, K.

var. dichotoma, Grun.

olivaceum, E.

subclavatum, Grun.

Mastogloia Grevillii, Sm. Scarce.

Smithii Thw. var. amphicephala, Grun. Scarce.

Navicula Bacillum, E.

cryptocephala, K.
cincta, E.

var. Heufleri, Cl. (teste Cleve).

exilis, K.
gracilis, K.

gregaria, Donk. (teste Cleve).
,, hungarica, Grun.
Menisculus, Schum. (teste Cleve).
Pupula, K.
radiosa, K.
var. tenella, Grun.

Reinhardtii, Grun.
rhynchocephala, K. Rare.
Tuscula, Grun. Scarce.
vulpina, K. Scarce.
Neidium productum, Cl. Rare.
Nitzschia amphibia, Grun.
,, angustata, Sm.

, Denticula, Grun. (Denticula obtusa, Sm.)
dissipata, Grun.

Calcareous Pebbles formed by Algtz* 83

Nitzschia Hnearis, Sm.

sinuata, Grun.
Pinnularia commutata, Grun.

rupestris, Hantzsch.

appendiculata, K.

subcapitata, Greg.

Rhaphoneis oregonica, E. (doubtful). Rare.
Pleurosigma attenuatum, Sm. Scarce.
Pleurostauron Smithii, Grun. One specimen seen.
Synedra delicatissima, Sm.

E. GROVE,

XIV.

NOTES ON THE SORI OF MACROCYSTIS

AND POSTELSIA.

THE occurrence and function of conceptacular bodies in so many
genera of olive-brown Algse is a subject that has received much
attention from phycologists. Without entering on a discussion of the
origin of these structures, it has seemed to us that the peculiar position
of the sori in the two genera, Macrocystis and Postelsia, throws
additional light on the function of conceptacles ; we have found the
sori in these plants confined to the furrows of the sporophyllous leaves,
a position of shelter such as is afforded elsewhere by conceptacles.

The Laminariacecs have been subdivided by Mr. Setchell into three
groups, of which the types are Laminaria, Lessonia, and Alaria, and
in that of Lessonia he has placed Postelsia and the gigantic Macrocystis,
The members of the Lessonia group are characterised by the complexity
of their fronds, due, in the case of Macrocystis, to unequal splitting
of the growing tissue, so that, while the main growth of the thallus
continues at the apex, and becomes a stem-like member, the parts that are
split off develop into long narrow leaves with large air-bladders at the
base. When quite young the fronds are thin and smooth, the older
members become thicker and stronger, and are longitudinally wrinkled
or furrowed. Along the margin spines occur, which develop after the
splitting has taken place.

Rosenthal published in Flora, 1890, a detailed account of the anatomy
and development of the vegetative organs of Macrocystis luxurians (== M.
pyrifera.} The leaves have an epidermis of one or sometimes two cells
in depth. These are small, compact, and full of contents. Within
this layer are the cortical cells, which become larger towards the centre
of the frond, and, finally, the central tissue, which in the older leaves

84

Notes on the Sori of Macrocystis and Postelsia. 85

is of long thick -walled cells. Large mucilaginous canals embedded
in the outer cortex occur all over the leaves.

The reproductive bodies are borne on the fertile leaves, which are
produced from the base of the plant on special branches. They are
much like the vegetative leaves in appearance, being long and narrow,
and bordered with spines ; but they are destitute of air-bladders, and
are more regularly and deeply furrowed than the floating leaves.

In the various descriptions hitherto published of the fruit of Macrocystis,
the sporangia have been described as occurring in irregular son', or in
patches on the lamina of the fertile leaf ; but this does not seem to
be an exact or adequate description. In the plant examined, M.
pyrifera, Ag., they are confined to the depressions on either side of the
furrowed leaf (plate XX., fig. i) ; they are not, however, invariably present,
and the furrows may be empty for considerable spaces. The tissue
of the fertile leaf resembles that of the vegetative leaf; the sporangia
which arise by repeated division of the epidermal cells are stout, club-
shaped bodies (plate XX., fig. 3), and are surrounded and over-topped
by the dark-brown paraphyses. These also arise from the epidermal
cells ; they are club-shaped, but more slender than the sporangia (plate
XX., fig. 3), and are full of dark-coloured contents.

The sporangia, as already stated, occur in the hollow part of the
furrow in longitudinal sori. The furrows are comparatively wide at
the base, and so closely pressed together that those adjacent almost
touch each other (plate XX., fig. i) ; the outlet is further closed by a
projection of the tissue of the leaf, which, from each side, bends over
the sorus (plate xx., fig. 2). The cells of the epidermis and outer
cortex forming the projection are elongated parallel to the surface of
the leaf, and, by narrowing the entrance of the furrow, afford additional
shelter and protection to the growing and ripening fruits. A section
across the fertile leaf has thus the appearance of a row of conceptacles,
and forcibly recalls in appearance the conceptacles of Splachnidium and
other brown Algae.

Postelsia palmceformis is, so far as is at present known, limited in

its habitat to the coast of California. It is of interest to note that the

Pacific Coast of North America, otherwise rich in Laininariacece, is

especially characterised by the number of peculiar genera which it

possesses, i.e., Dictyoneuron, Postelsia, Nereocystis, Cymathere, Pterygophora,

Egregia, and Eisenia, many of them monotypic and limited to California.

The only description of Postelsia palmceformis is that given by

Ruprecht (Nene oder nnvollstandig bekannte Pflanzen aus der nordlicJien

Theile des Stillen Oceans) from material collected on the shores of

86 Notes on the Sori of Macrocystis and Postelsia.

Bodega Bay, California. The plant which we have examined is in the
Herbarium of the British Museum, and was collected by Dr. Anderson
at Santa Cruz, California. The plant consists of a root-like member
bearing an unbranched, erect, hollow stem. The stem at its apex
divides into several portions, each of which bears directly, or after still
further divisions, a number of stalked fronds.

The plant described and figured by Ruprecht presents some points
of difference ; the stem bears the stalked fronds directly from its apex
without any further division, and the fronds of his plant have a toothed
margin, and are larger than the smaller untoothed fronds of the specimen
we have examined.

The fronds are longitudinally ribbed throughout their whole length,
the ribs being most marked in the central thicker portion of the thallus,
and diminishing in size towards the margins (plate XX., fig. 4.) The
epidermis of the frond consists of one or sometimes two layers of small
cells with coloured contents. Beneath this is the cortex, made up of
roundish thick-walled cells, of only a few layers in thickness in the
furrows, but forming the whole mass of the tissue of the ridges. The
medullary portion consists of filaments running in an irregularly longi-
tudinal direction forming a compact tissue.

The fronds are all sporophylls, the sporangia occurring in the
following manner. At the points where the ridge bends into the furrow
the epidermal cells grow out into paraphyses, and the whole furrow is
lined with a dense mass of sporangia and typical Laminarian paraphyses
(plate XX., figs. 5 & 6).

Ruprecht figures a transverse section of a frond showing the ridges
and furrows into which the thallus is folded ; he also describes and
figures the sporangia and paraphyses, but, in mentioning their position,
says only that they occur in patches near the apices of the mature
leaves. He does not appear to have noticed what has seemed a point
of interest to us, that, while throughout the whole frond the sporangia
occur regularly in the furrows, they never occur on the ridges of the
folds, a protection thus being afforded to them, by the natural structure
of the thallus, almost as perfect as that of a true conceptacle.

A. LORRAIN SMITH.
FRANCES G. WHITTING.

Notes on the Sori of Macrocystis and Postelsia. 87

EXPLANATION OF PLATE XX.

Fig. i. Macrocystis pyrifera. Trans, section of frond (x 10).

Fig. 2. Trans, section of a furrow of the fertile frond

(x 66).

Fig. 3. Sporangia and paraphyses ( x 450).

Fig. 4. Postelsia palmaformis, Trans, section of frond ( x 10).

Fig. 5. ,, Trans, section of a furrow of the frond ( x 50).

Fig. 6. Sporangia and paraphyses ( x 440).

N

XV.

A COMPARISON OF THE ARCTIC AND
ANTARCTIC MARINE FLORAS.

THE general similarity, and in many cases the identity, of species
of marine organisms in the Arctic and Antarctic regions have led to
a very interesting analysis of the facts, and an interpretation of them
by Dr. John Murray in the Summary of the Results of the Challenger
Expedition. His comparison is based mainly on the marine animals,
but he points out also, as confirming his argument, that the plankton
Algae are in many cases identical. ' Rhabdospheres are found only
in the warmer waters of the ocean, but the Coccospheres abound in the
northern and southern temperate zones ; in the Arctic and Antarctic
zones these calcareous unicellular Algae are replaced by species of
Protococci, which are identical in the two polar regions. Many pelagic
Diatoms of the Arctic and Antarctic are likewise identical.' His
theoretical explanation of this remarkable fact in the distribution of
organisms is of the highest interest. Speaking of carboniferous times,
he holds that 'the surface temperature of the sea could not well have
been less than about 70 F., and the same temperature and the same
marine fauna prevailed from equator to poles, the temperature not being

higher at the equator In early Mesozoic times cooling at the

poles and differentiation into zones of climate appear to have commenced,
and temperature conditions did not afterwards admit of coral reefs in
the polar area, but the colder, and hence denser, water that in
consequence descended to the great depths of the ocean carried with
it a large supply of oxygen, and life in the deep sea became possible
for the first time. There have been many speculations as to how a
nearly uniform temperature could have been brought about in sea-water
over the whole surface of the earth in early geological ages, as uell as

88

Comparison of the Arctic and Antarctic Marine Floras. 89

to how sufficient light could have been present at the poles to permit
of the luxuriant vegetation that once flourished in these regions. The
explanation that appears to me the most satisfactory is the one which
attributes these conditions to the very much greater size of the sun in
the early stages of the earth's history an idea first introduced into
geological speculations by Blandet {Bull. Soc. Geol. de France, ser. 2,
t. 25, p. 777, 1867-8), who likewise discussed the relations of Arctic
and Antarctic faunas together with the greater amount of aqueous
vapour in the atmosphere and the greater mass of the atmosphere.'
Dr. Murray holds that the pelagic Algae, Radiolaria, and Foraminifera
are probably the slightly modified descendants of a very ancient
universal pelagic fauna and flora. It is quite plain that the conditions
of environment of the pelagic flora are not such as to produce much
variation on any theory, and accordingly, though the mass of the
pelagic Algae enormously exceeds that of the littoral Algae, it consists,
so far as we know, of few species compared with the much-varied
littoral flora in fact, the discrepancy is so great as to hardly admit of
comparison at all. Similarly, zoologists have found that the mass of
individuals in the pelagic fauna probably greatly exceeds that of other
marine faunas, and that the species are few when compared with the
organisms of the shore and shallow water. Nearly all the fossil Algae
of the early Tertiary and the Secondary rocks, and the single form
known from the Devonian, are of a type found at the present time
within the Tropics, and with very few exceptions confined to the warmer
seas. They owe their preservation for the most part to their being
calcareous, which fits into the fact of their coming from warm seas
(see p. 75). Of course, we must assume that other Algae existed, and
their remains have been lost through their not being incrusted, &c. ;
but yet the argument has some weight.

If, then, we test Dr. Murray's theory by an examination and
comparison of the Arctic and Antarctic marine floras, we would begin
by expecting a stronger agreement between the plankton floras than
between the littoral floras of these regions from the known stable
character of the former, and the varied and variable character of the
latter. In the following comparison there is, therefore, less agreement
to be expected than from a comparison of the total plankton and
littoral floras.

In order to make the comparison we have taken Kjellman's Alga of
the Arctic Sea as representing the one part ; and for the Antarctic we have
compiled a list. The region includes Cape Horn and the Falkland
Islands, Marion, Kerguelen and Heard Islands, and Lord Auckland

9O Comparison of the Arctic and Antarctic Marine Floras.

and Campbell's Islands, from which we have records by Hooker and
Harvey (Antarctic Voyage], by Dickie, in the Challenger and Transit
of Venus (Kerguelen) Expeditions, and by Harlot in the Mission
Scientifique du Cap Horn. From these sources the list has been
compiled, and brought under the same system of classification as that
adopted by Kjellman ; and, of course, it has been tested by our
knowledge of the material.

It will be seen from the totals that the number of species in the two
floras is fairly even, viz., 259 in the Arctic and 269 in the Antarctic. The
Arctic species are in 1 1 1 genera, i.e., an average of 2\ species to the genus
exactly. The Antarctic species are in 98 genera, i.e., very nearly an average
of 2f species to the genus. The average proportion of species to genera
in other seas has been commented on before (p. 66). The genera
common to both oceans is 56, and the species 41. But no less than
42 of the species here recorded as Antarctic, and not Arctic, occur in
the north temperate zone. This probably means, in part, that we have
drawn our Antarctic line too near the south temperate ocean. Similarly,
on comparing the Arctic list with Miss Barton's Cape of Good Hope
list, we find another 9 species occurring in both, but not recorded from
the Antarctic. Adding these figures together, we get 92 species common
to northern and southern polar and partly temperate coasts. Of these,
however, 38 species occur in the intervening Tropics are cosmopolitan,
in short, and we must subtract them from the above total, yielding 54
species in common north and south of the tropical belt, and not occurring
within it. Out of the two totals of 259 and 269 there are thus 54
species in common which have not been found within the Tropics. If
we were to compare all the north temperate and south temperate
littoral seaweeds (instead of the polar forms mixed with certain
temperate forms that enter polar seas at either pole as we have done),
we should run a greater risk of counting species as extra-tropical,
which are probably really tropical though not recorded. Therefore it
appears to us to be a more safe proceeding to deal only with polar
and adjoining waters.

The tables disclose a number of interesting facts in distribution, the
principal one being this, that all the genera in the list of Fucacece and
LaminariacetZ) the largest seaweeds, are either Arctic or Antarctic none
of them are both. Some of these genera, e.g., Macrocystis, Laminaria,
are found, however, on both sides of the tropical belt, and other genera
not mentioned here, e.g., PycnopJiycus (Fucacecz} occur in both temperate
belts, but all such occurrences are exceptional among the great seaweeds.
Nothing, in fact, is more striking in the distribution of seaweeds than the

Comparison of the Arctic and Antarctic Marine Floras. 91

change from our northern Fucacece to the Sargasso, and other allied
genera of the tropical belt, and then to other Fucacea again in the
south temperate and Antarctic seas, these last resembling the northern
forms in general facies, but yet generically distinct in most cases.

When we compare the littoral Algae of the tropical Atlantic, viz., 859
species in 162 genera, with those of the tropical Indian Ocean, viz., 514
species in 139 genera, we get of these 173 species and 103 genera in common
(see Mem. XI.) a very considerable proportion. However, while these
oceans, or rather the tropical portions of them, may have been in
communication via the Cape by a change of climate sufficient to give
a tropical or sub-tropical character to the Cape, in comparatively recent
geological times, the heat barrier between the northern and southern
polar seas has always been there. The cold depths of the ocean do
not enter into our calculation in the case of Algae owing to the illu-
mination difficulty. These two polar marine floras have been separated
as long as there has been climate of any sort on the globe, and out of
their poor marine floras there are 54 species that occur north and
south of the tropical belt, and, so far as we know, not within it.
Whether this needs a new cosmical theory to account for it or not we
do not pretend to say, but it appears to support Dr. Murray's other
statistics, and is more significant in some respects than the agreement
of the plankton Algae, which have a more stable environment than the

littoral forms.

GEORGE MURRAY.

ETHEL S. BARTON.

92 Comparison of the Arctic and Antarctic Marine Floras.

Arctic No. of
Species.

Antarctic No
of Species.

Genera com-
mon to both

Species com-
mon to both.

Corallines :

Corallina

I

3

I

I

Lithothamnion

IO

2

I

I

Lithophyllum

2

6

I

I

Melobesia

3

5

I

Jania

o

i

o

o

Amphiroa

4

O

o

RJiodoviektz :

Rhodomela

2

5

I

o

Alsidium

O

i

O

o

Odonthalia

I

o

o

o

Polyzonia

O

i

o

o

Polysiphonia

II

18

I

2

Dasya

o

2

o

o

Pollexfenia

o

I

o

o

Bostrychia

o

4

o

Chondriecs :

Laurencia

o

i

o

o

Ptilonia

o

i

o

o

Delisea

o

i

o

o

Cladhymenia

i

o

o

Spongiocarpece :

Polyides

I

o

o

o

Wrangeliace& :

Spermothamnion

I

o

o

Chantransia

4

2

I

2

Solieriecz :

Catenella

o

I

Hypnece :

Hypnea

o

I

o

o

\f

Comparison of the Arctic and Antarctic Marine Floras. 93

Arctic No. of
Species.

Antarctic No.
of Species.

Genera com-
mon to both.

Species com-
mon to both.

Gelidietz :

Choreocolax

O

I

O

Gelidium ..

O

I

O

Ch&tangiece :

Chaetangium

2

O

Deles seriece :

Delesseria

7

IO

I

Nitophyllum

i

II

I

I

Sphcsrococcoidecs :

Gracilaria

2

o

Squamariece :

Hildenbrandtia

I

2

I

o

Peyssonnelia

I

I

I

o

Petrocelis

2

O

o

Cruoria

I

o

o

o

Haemescharia

I

o

o

RJiodymeniea :

Hydrolapathum

I

I

I

I

Rhodymenia

2

5

I

I

Epymenia

O

3

o

Plocamium

I

3

I

I

Acanthococcus

O

2

o

o

Rhodophyllis

I

2

I

o

Euthora

I

o

o

o

Champiecs :

Chylocladia

2

I

I

I

Dumontiecs :

Dumontia

I

2

I

I

Halosaccion

2

o

o

Sarcophyllis

2

o

o

FurcellariecB :

Furcellariea

I

o

o

Gigartinecz :

Cystoclonium

I

O

o

Chondrus

I

I

I

I

Callophyllis

I

7

I

o

Iridaea

O

7

o

o

Gigartina

I

/

6

I

Phyllophora

3

i

I

o

Ahnfeltia

i

2

I

I

94 Comparison of the Arctic and Antarctic Marine Floras.

Arctic No. of

Antarctic No.

Genera com-

Species com-

Species.

of Species.

mon to both.

mon to both.

Gigartinetz continued :

Gymnogongrus

O

3

O

O

Kallymenia

3

i

I

O

CryptonemiecB :

Halymenia

o

i

Grateloupia

O

2

o

CeramiecB :

Microcladia

I

O

o

Ceramium

4

7

I

2

Carpoblepharis

o

I

o

O

Rhodochorton

6

I

I

I

Callithamnion

9

12

I

O

Ballia

o

3

o

o

Griffithsia

j

4

o

o

Halurus

o

i

o

o

Ptilota

3

a

I

v^

o

Porphyracea :

J

j

Porphyra

2

5

I

I

Diploderma

2

o

o

Bangia

I

o

o

o

Erythrotrichia

I

i

I

1

IO4

179

29

20

Comparison of the Arctic and Antarctic Marine Floras. 95

PH^EOPHYCEsE.

Arctic No. of
Species.

Antarctic No.
of Species.

Genera com-
mon to both.

Species com-
mon to both.

Fucacea :

Himanthalia

I

O

O

O

Halidrys

I

O

O

Ozothallia

I

O

O

O

Fucus

IO

o

o

o

Pelvetia

I

o

O

o

Xiphophora

o

I

O

o

Marginaria

I

O

o

Scytothalia

1

O

o

Durvillaea

o

2

O

o

Tilopteridecs :

Scaphospora

I

O

O

o

Haplospora

I

O

o

Laminariacecs :

Alaria

7

o

o

o

/

Agarum

I

o

o

o

Phyllaria

2

o

o

Laminaria

II

o

o

o

Chorda

2

o

o

o

Adenocystis

O

I

o

o

Lessonia

O

4

o

o

Macrocystis

O

T^

I

o

o

Encceliea :

Stilophora

I

o

Asperococcus

2

2

I

2

Hydroclathrus

I

o

o

Ralfsiecz :

Ralfsia

2

o

o

o

Chordariecs :

Chordaria

I

I

I

I

Mesogkea

I

I

I

o

Stereocladon

2

o

o

96 Comparison of the Arctic and Antarctic Marine Floras.

Arctic No. o

Antarctic No.

Genera com-

Species com-

Species.

of Species.

mon to both.

mon to both.

Chordariea continued :

Cladothele

O

I

O

O

Castagnea

I

O

O

Eudesme

I

O

O

Myrioneinatce :

Leathesia

I

O

O

O

Elachista

2

o

O

Myrionema

I

O

O

L ithodermatecB :

Lithoderma

2

o

O

O

Scytosiphonecs :

Ilea

I

J

I

I

Scytosiphon

2

3

I

I

Punctariece :

Punctaria

I

I

I

O

DesinarestiecE :

Dichloria

I

I

I

I

Desmarestia

I

6

I

O

Dictyosiphon

4

i

I

o

Litosiphon

i

O

o

Phlc-eospora

3

O

o

Coilonema

2

o

o

o

Ciitleriacece :

Aglaozonia

I

o

o

o

Sphacelariece :

Cladostephus

I

2

I

I

Stypocaulon

I

O

o

o

Chretopteris

I

o

o

o

Sphacelaria

3

5

I

Ectocarpece :

Isthmoplea

i

o

o

Ectocarpus

9

6

I

2

Actinema

o

i

o

O

Streblonema

i

o

Pylaiella

3

o

o

o

Myriotrichia

i

o

o

Gloeothamnion

i

o

o

o

92

47

12

9

Comparison of the Arctic and Antarctic Marine Floras. 97

CHL OROPH FC^.

Arctic No. of

Antarctic No.

Genera com-

Species com-

Species.

of Species.

mon to both.

mon to both.

Ulvece :

Ulva

2

2

I

I

Enteromorpha

7

2

I

2

Monostroma

12

I

I

I

Prasiola

I

o

o

o

Diplonema

I

O

\J

O

Confervea; :

Chaetophora

2

Spongomorpha

3

I

I

I

Cladophora

5

15

I

I

Rhizoclonium

3

4

I

I

Chaetomorpha

5

3

I

I

Ulothrix

3

o

o

o

Siphonocladus

i

o

Urospora

i

i

I

I

Bulbocoleon

[

o

o

Characiece :

Chlorochytrium

i

I

I

I

Characium

i

o

o

o

Codiolum

3

o

o

Palmellece :

Chlorangium

i

o

o

Siphone<z :

Bryopsis

i

2

I

I

Derbesia

i

O

o

o

Codium

o

-3

o

o

J

54

36

IO

1 1

1

98 Comparison of the Arctic and Antarctic Marine Floras.

PROTOPHYCEsE.

Arctic No. of
Species.

Antarctic No.
of Species.

Genera com-
mon to both.

Species com-
mon to both.

Rivularia

2

I

I

Calothrix

3

2

I

O

Lyngbya
Oscillatoria

I
I

I
I

I
I

I

Spirulina
Glceocapsa
Dermocarpa

Grand Total

I
I
O

I
I

O
I
O

O
O
O

9

7

5

I

259

269

56

41

Phyc.Mem.Parb.ITI .

17.

BerjeauiX Hv^hiey del.et ktl\ .

PACHYTHECA

PLyc. Mem .Part. 111.

PI. 18.

S>

. /.

" rJf*,.

i -

8&2

&&*

, ^ ^- : .,x <'-'

J .'* jn, _ ,. f . -^y,

. ,

V 1 V

^m

'"^^iiJili^ilS^

,

;

.
: : . >-,, .

i
i

'?'!>*,

. - ; }

Berjeau-Sc Hicjhley del.etlith

PACHYTHECA ,

Hiyc.Tyfem.Parb.nl.

PL 19.

.

-

/: - -

:

'

im

U : : ^y

'

- : M:-~.,, '

K ,'

'&f&$2l&-&'

* y,^"D

J- "- .

Berjeau.2c.Hi(jKle)r Islet Jibk. Han. IT

CALCAREOUS PEBBLES -SCHIZOTHRIX "FASCICULATA

Pliyc.Mem.Parb.IIl

PI . 2C

^<N> J^K -2-sS^?r^O ^z3<?S^>

A.L.S.&P.G.W. del.
Berjeau^ HigHey libh .

MACROCYST1S AND POSTELSIA

HanKart

PHYCOLOGICAL MEMOIRS

BEING

RESEARCHES MADE IN THE BOTANICAL DEPARTMENT

OF THE BRITISH MUSEUM.

EDITED BY

GEORGE MURRAY, F.R.S.E., F.L.S.

PART I.

DULAU AND CO,,
SOHO SQUARE, LONDON.

APRIL, 1892.

The issue of the Phycological Memoirs is made for the -purpose of keeping
within the limits of one book the results of researches on Alg<e in the De-
partment of Botany, British Museum. 'The publication is not a Journal, and
is not meant to compete with the numerous excellent botanical magazines.
It is intended to issue Parts at about half-yearly intervals and in time to
complete a Volume containing about 50 plates. In addition to the promotion
of research, the authors of the Memoirs have the object of making better
known the treasures of the Museum, which they have received facilities to
investigate.

CONTENTS OF PART I.

I. ON SPLACHNIDIUM RUGOSUM, GREV., THE TYPE OF A
NEW ORDER OF ALG^E. By MARGARET O. MITCHELL, former
Bathurst Student, Newnham College, Cambridge, and FRANCES G.
WHITTING, former Student of Newnham College, Cambridge.

II. ON A FOSSIL ALGA BELONGING TO THE GENUS CAU-
LERPA, FROM THE OOLITE. By THE EDITOR.

III. ON THE STRUCTURE OF DIC1 YOSPH&RIA DECNE. By
THE EDITOR.

IV. ON MALFORMATIONS OF ASCOPHYLLUM AND DESMA-
RESTIA. By ETHEL SAREL BARTON.

V. ON CONCHOCELIS, A NEW GENUS OF PERFORATING ALG.E.
By E. A. L. BATTERS, B.A., LL.B., F.L.S.

PHYCOLOGICAL MEMOIRS

BEING

RESEARCHES MADE IN THE BOTANICAL DEPARTMENT

OF THE BRITISH MUSEUM.

EDITED BY

GEORGE MURRAY, F.R.S.E., F.L.S.

PART II.

DULAU AND CO.,

SOHO SQUARE, LONDON.

MAY, 1893.

1

CONTENTS OF PART I.

I. ON SPLACHNIDIUM RUGOSUM GREV., THE TYPE OF A
NEW ORDER OF ALG^E. By MARGARET O. MITCHELL, former
Bathurst Student, Newnham College, Cambridge, and FRANCES G.
WHITTING, former Student of Newnham College, Cambridge.

II. ON A FOSSIL ALGA BELONGING TO THE GENUS CAU-
LERPA, FROM THE OOLITE. By THE EDITOR.

III. ON THE STRUCTURE OF DICTYOSPH^ERIA DECNE. By
THE EDITOR.

IV. ON MALFORMATIONS OF ASCOPHYLLUM AND DESMA-
RESTIA. By ETHEL SAREL BARTON.

V. ON CONCHOCELIS, A NEW GENUS OF PERFORATING
By E. A. L. BATTERS, B.A., LL.B., F.L.S.

PRICE 7 s. 6d.

CONTENTS OF PART II.

VI. NOTES ON THE MORPHOLOGY OF THE FUCACE&:
COCCOPHORA LANGSDORFII GREV. By A. LORRAIN SMITH.

SEIROCOCCUS AXILLARIS GREV. By A. LORRAIN SMITH.
XIPHOPHORA BILLARDIERII MONT. By E. S. BARTON.
NOTHEIA ANOMALA BAIL, and HARV. By M. O. MITCHELL.
SARCOPHYCUS POTATORUM KUTZ. By F. G. WHITTING.

VII. ON CHLOROCYSTIS SARCOPHYCI, A NEW ENDOPHYTIC
ALGA. By F. G. WHITTING.

VIII. ON HALICYSTIS AND VALONIA. By THE EDITOR.

IX. ON THE STRUCTURE OF HYDROCLATHRUS BORY. By
M. O. MITCHELL.

X. ON THE CRYPTOSTOMATA OF ADENOCYSTIS, ALARIA
AND SACCORHIZA. By THE EDITOR.

XL A COMPARISON OF THE MARINE FLORAS OF THE WARM
ATLANTIC, INDIAN OCEAN, AND THE CAPE OF GOOD
HOPE. By THE EDITOR.

PRICE 73. 6d.

PHYCOLOGICAL MEMOIRS

BEING

RESEARCHES MADE IN THE BOTANICAL DEPARTMENT

OF THE BRITISH MUSEUM.

EDITED BY

GEORGE MURRAY, F.R.S.E., F.L.S.

DULAU AND CO.,
SOHO SQUARK, LONDON,

APRIL, 1895.

CONTENTS OF PART III.

XII. A NEW PART OF PACHYTHECA. By THE EDITOR.

XIII. CALCAREOUS PEBBLES FORMED BY ALG^E. By THE EDITOR.
With LIST OF DIATOMS. By E. GROVE.

XIV. NOTES ON THE SORI OF MACROCYSTIS AND POSTELSIA.

By A. L. SMITH and F- G. WHITTING.

XV. A COMPARISON OF THE ARCTIC AND ANTARCTIC MARINE
FLORAS. By THE EDITOR and E. S. BARTON.

PRICE ys. 6d.

CONTENTS OF PART I.

I. ON SPLACHNIDIUM RUGOSUM GREV., THE TYPE OF A
NEW ORDER OF ALG^E. By MARGARET O. MITCHELL, former
Bathurst Student, Newnham College, Cambridge, and FRANCES G.
WHITTING, former Student of Newnham College, Cambridge.

II. ON A FOSSIL ALGA BELONGING TO THE GENUS CAU-
LERPA, FROM THE OOLITE. By THE EDITOR.

III. ON THE STRUCTURE OF DICTYOSPH&RIA DECNE. By
THE EDITOR.

IV. ON MALFORMATIONS OF ASCOPHYLLUM AND DESMA-
RESTIA. By ETHEL SAREL BARTON.

V. ON CONCHOCELIS, A NEW GENUS OF PERFORATING ALG^E.
By E. A. L. BATTERS, B.A., LL.B., F.L.S.

PRICE 73. 6d.

CONTENTS OF PART II.

VI. NOTES ON THE MORPHOLOGY OF THE FUCACE&:
COCCOPHORA LANGSDORFII GREV. By A. LORRAIN SMITH.

SEIROCOCCUS AXILLARIS GREV. By A. LORRAIN SMITH.
XIPHOPHORA BILLARDIERII MONT. By E. S. BARTON.
NOTHEIA ANOMALA BAIL, and HARV. By M. O. MITCHELL.
SARCOPHYCUS POTATORUM KUTZ. By F. G. WHITTING.

VII. ON CHLOROCYSTIS SARCOPHYCI, A NEW ENDOPHYTIC
ALGA. By F. G. WHITTING.

VIII. ON HAUCYSTIS AND VALONIA. By THE EDITOR.

IX. ON THE STRUCTURE OF HYDROCLATHRUS BORY. By
M. O. MITCHELL.

X. ON THE CRYPTOSTOMATA OF ADENOCYSTIS, ALARIA
AND SACCORHIZA. By THE EDITOR.

XL A COMPARISON OF THE MARINE FLORAS OF THE WARM
ATLANTIC, INDIAN OCEAN, AND THE CAPE OF GOOD
HOPE. By THE EDITOR.

PRICE 73. ed.